PostGraduate Vascular Surgery

This page intentionally left blank
Postgraduate Vascular Surgery

The Candidate’s Guide to the FRCS
Postgraduate Vascular
Surgery
The Candidate’s Guide to the FRCS
Edited by
Vish Bhattacharya
Queen Elizabeth Hospital
Gerard Stansby
Freeman Hospital
ca mb rid g e un iv e r si t y pres s
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore,
São Paulo, Delhi, Dubai, Tokyo, Mexico City
Cambridge University Press
The Edinburgh Building, Cambridge CB2 8RU, UK
Published in the the United States of America by Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/9780521133524
© Cambridge University Press 2010
This publication is in copyright. Subject to statutory exception

and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2010
Printed in the United Kingdom at the University Press, Cambridge
A catalogue record for this publication is available from the British Library
Library of Congress Cataloguing in Publication data

Postgraduate vascular surgery : the candidate’s guide to the FRCS / [edited by] Vish Bhattacharya,
Gerard Stansby.
p. ; cm.
Includes bibliographical references and index.
ISBN 978-0-521-13352-4 (pbk. : alk. paper)
1. Blood-vessels–Surgery–Handbooks, manuals, etc. I. Bhattacharya, Vish. II. Stansby,
Gerard. III. Title.
[DNLM: 1. Vascular Surgical Procedures–Handbooks. 2. Vascular Diseases–surgery–
Handbooks. WG 39]
RD598.5.P67 2010
617.4′13–dc22 2010039933
ISBN 978-0-521-13352-4 Paperback
Cambridge University Press has no responsibility for the persistence or

accuracy of URLs for external or third-party internet websites referred to
in this publication, and does not guarantee that any content on such
websites is, or will remain, accurate or appropriate.
Every effort has been made in preparing this book to provide accurate and up-to-date
information which is in accord with accepted standards and practice at the time of publication.
Although case histories are drawn from actual cases, every effort has been made to disguise the
identities of the individuals involved. Nevertheless, the authors, editors and publishers can make no
warranties that the information contained herein is totally free from error, not least because clinical
standards are constantly changing through research and regulation. The authors, editors and
publishers therefore disclaim all liability for direct or consequential damages resulting from the use of
material contained in this book. Readers are strongly advised to pay careful attention
to information provided by the manufacturer of any drugs or equipment that they plan to use.
Contents
List of contributors page vii
Preface ix
Section 1. Final FRCS vascular 9 Femoral artery aneurysms 123

Robert Davies, Asif Mahmood and
clinicals Rajiv Vohra
Introduction to the examination 10 Carotid, subclavian and vertebral
and clinical cases 3 disease 129
Jonathan Smout and Asif Mahmood A. Ross Naylor
11 Diagnosis and management of
Section 2. Final FRCS vascular thoracic outlet syndrome 146
topics Hassan Badri and Vish
Bhattacharya
1 Vascular risk factors and their
management 37 12 Diagnosis and management of
Alasdair Wilson and Julie Brittenden hyperhidrosis 152
Hassan Badri and Vish
2 Management of acute limb Bhattacharya
ischaemia 49
Arun Balakrishnan and 13 Chronic mesenteric
David Lambert ischaemia 156
Mohamed Abdelhamid, Robert
3 Chronic lower limb ischaemia, Davies and Rajiv Vohra
critical ischaemia and the diabetic
foot 58 14 Acute ischaemic colitis 164
Chris Davies and Cliff Shearman Vish Bhattacharya and Gerard
Stansby
4 Endovascular and surgical options
for peripheral revascularisation 72 15 Vascular trauma 172
Colin Nice Robbie George and Paul Blair
5 Abdominal aortic aneurysms 86 16 Indications and management of

Vish Bhattacharya and Rob Williams lower limb amputation 183
Mark Kay and Colette Marshall
6 Thoracic, thoracoabdominal and
suprarenal aortic aneurysms 99 17 Leg swelling and
Jenny Richards and Rod Chalmers lymphoedema 195
Arun Balakrishnan and
7 Aortic dissection 108 Tim Lees
Jenn Richards and Rod Chalmers
18 Varicose veins and chronic venous
8 Popliteal artery aneurysms 115 insufficiency 204
Robert Davies, Asif Mahmood and Marcus Cleanthis and
Rajiv Vohra Tim Lees
vi Contents
19 Management of deep vein 24 Access surgery 288

thrombosis 220 David C. Mitchell and William D.
Dharmendra Garg and Vish Neary
Bhattacharya
25 Basic outline of solid organ
20 Infection in vascular surgery 229 transplantation 298
Mike Clarke Mathew Jacob, Jeremy French and
Derek Manas
21 Vascular malformations 242
George Hamilton and Andrew Platts
22 Vasospastic disorders and
Index 317
vasculitis 254
Mohammed Sharif, Jonathan Smout
and Gerard Stansby
23 Critical care considerations
and preoperative assessment
for general and vascular
surgery 272
Ian D. Nesbitt and David M. Cressey
Contributors
Mohamed Abdelhamid David M. Cressey

Department of Vascular Surgery, Department of Anaesthesia,
University Hospital Birmingham Freeman Hospital, High Heaton,
NHS Foundation Trust, Selly Oak Hospital, Newcastle upon Tyne, UK
Birmingham, UK
Chris Davies
Hassan Badri Department of Vascular Surgery,
Gateshead Health NHS Foundation Trust, Southampton General Hospital,
Queen Elizabeth Hospital, Sheriff Hill, Southampton, UK
Gateshead, UK
Robert Davies
Arun Balakrishnan University Department of Vascular
The Northern Vascular Centre, Surgery, Heart of England NHS
Freeman Hospital, High Heaton, Foundation Trust, Birmingham, UK
Newcastle upon Tyne, UK
Jeremy French
Vish Bhattacharya Transplantation and Hepatobiliary Surgery,
Gateshead Health NHS Foundation Trust, Freeman Hospital, High Heaton,
Queen Elizabeth Hospital, Newcastle upon Tyne, UK
Sheriff Hill, Gateshead, UK
Dharmendra Garg
Paul Blair Gateshead Health NHS Foundation Trust,
Department of Vascular Endovascular Queen Elizabeth Hospital,
Surgery, Royal Victoria Hospital, Belfast, UK Sheriff Hill, Gateshead, UK
Julie Brittenden Robbie George

University Department of Surgery, Department of Vascular Endovascular
Aberdeen Royal Infirmary, Aberdeen, Surgery, Royal Victoria Hospital, Belfast,
UK UK
Rod Chalmers George Hamilton

University Department of Surgery, University Department of Vascular
Royal Infirmary of Edinburgh, Surgery, Royal Free Hospital, London,
Edinburgh, UK UK
Mike Clarke Mathew Jacob

The Northern Vascular Centre, Transplantation and Hepatobiliary Surgery,
Freeman Hospital, High Heaton, Freeman Hospital, High Heaton,
Newcastle upon Tyne, UK Newcastle upon Tyne, UK
Marcus Cleanthis Mark Kay

The Northern Vascular Centre, Coventry and Warwickshire County
Freeman Hospital, High Heaton, Vascular Unit, University Hospitals
Newcastle upon Tyne, UK Coventry, UK
viii List of contributors
David Lambert Andrew Platts

The Northern Vascular Centre, Consultant Vascular Interventional
Freeman Hospital, High Heaton, Radiologist, Vascular Unit, Royal Free
Newcastle upon Tyne, UK Hospital, University College Medical
School, London, UK
Tim Lees
The Northern Vascular Centre, Jenny Richards
Freeman Hospital, High Heaton, University of Edinburgh, Royal Infirmary
Newcastle upon Tyne, UK of Edinburgh, Edinburgh, UK
Asif Mahmood Mohammed Sharif
Department of Surgery, Good Department of Vascular and
Hope Hospital, Sutton Coldfield, Endovascular Surgery, Belfast City
Birmingham, UK Hospital, Belfast, UK
Derek Manas Cliﬀ Shearman
Transplantation and Hepatobiliary Surgery, Department of Vascular Surgery,
Freeman Hospital, High Heaton, Southampton General Hospital,
Newcastle upon Tyne, UK Southampton, UK
Colette Marshall Jonathan Smout
Coventry and Warwickshire County The Northern Vascular Centre,
Vascular Unit, University Hospitals Freeman Hospital, High Heaton,
Coventry, UK Newcastle upon Tyne, UK
David C. Mitchell Gerard Stansby
Department of Vascular Surgery, The Northern Vascular Centre,
Southmead Hospital, Westbury-on-Trym, Freeman Hospital, High Heaton,
Bristol, UK Newcastle upon Tyne, UK
A. Ross Naylor Rajiv Vohra
Department of Surgery, Leicester Royal Department of Vascular Surgery,
Infirmary, Leicester, UK Selly Oak Hospital, Birmingham, UK
William D. Neary Rob Williams
Department of Vascular Surgery, Department of Radiology,
Southmead Hospital, Westbury-on-Trym, Freeman Hospital, High Heaton,
Bristol, UK Newcastle upon Tyne, UK
Ian D. Nesbitt Alasdair Wilson
Department of Anaesthesia, Department of Surgery, University of
Freeman Hospital, High Heaton, Aberdeen, Aberdeen Royal Infirmary,
Newcastle upon Tyne, UK Aberdeen, UK
Colin Nice
Gateshead Health NHS Foundation Trust,
Queen Elizabeth Hospital, Sheriff Hill,
Gateshead, UK
Preface
Examinations are always challenging although less so for the well prepared. The Intercollegiate
Exit Examination in Surgery, in particular, often causes considerable anguish for the candi-
dates. It is expensive to fail the examination, both financially and emotionally. It is a high
stake examination and comes at a critical stage in a young surgeon’s career when he or she is
nearing the end of training.
Over many years, and having discussed the examination with both successful and unsuc-
cessful candidates, we felt that a book was required that would provide a quick, easy read
for candidates already under pressure to cover a vast array of subjects. It would need to be
pitched at the right level and would complement both clinical experience and aid personal
revision.
This book has been divided into two sections. The clinical cases section has been written
by two recently successful candidates who have incorporated the views of several trainees to
make this section relevant to their needs.
The second section has been written by experts in their respective fields who have pro-
vided succinct chapters in a concise format. All the co-authors are surgical trainees, most of
whom have passed the examination or are preparing for it. Key points have been mentioned
at the beginning of the chapters to aid quick revision before the examinations. References
have been deliberately kept to a minimum.
We thank all the contributors for their hard work and timely submission of manuscripts.
We are especially grateful to Nick Dunton and Katie James of Cambridge University Press
for their continuous support and timely reminders.
Though we appreciate the fact that it takes more than a book to pass the examinations we
hope this book serves as a useful tool for candidates to achieve success. Good luck!
Section Final FRCS vascular clinicals
1
The current format of the Final FRCS clinical examination for vascular candidates consists
of both general surgery and vascular clinicals. These examinations have the same layout,
and are taken on the same day. Each clinical consists of a series of five short stations taken in
rapid succession. For each examination there are two examiners who will take turns in ques-
tioning and marking the candidate. The cases are either a patient encounter or interpreting
an investigation. With the final examination being directed at ascertaining competence to
become a consultant, the questions tend to relate to management issues rather than test-
ing your ability to perform a head-to-toe clinical assessment. However, as senior trainees
it is expected that you should know how to examine a patient in an orderly and effective
fashion. Failure to demonstrate this in the clinical encounters will ring alarm bells with the
examiners.
The short case format of the clinicals should not be viewed as a hurdle, but rather as an
opportunity for you to impart your fundamental knowledge on a broad range of topics. The
examination process is an efficient way of assessing a wide range of subjects in a limited
period of time, in a systematic manner. Candidates who have gone through the clinicals are
often left stunned by the number of topics that have been discussed in a blur of 30 minutes.
Candidates should therefore see the benefits of this system where a poor performance for
one case becomes a small part of the whole marking scheme. It is essential that candidates
who feel that they have done badly at one station do not dwell on their misfortune, but com-
pose themselves and get on with the rest of the assessment believing they can still pass.
It is important that as the candidate you listen carefully to the examiner’s questions. The
instructions will often be extremely focused, and initially seem quite a minor request. The
initial ‘starter’ question will then lead on to more complex issues. The questions are generally
not intended to catch you out, so do what you are asked to do. It is essential that you appear
comfortable dealing with patients. Although observed examination practice with colleagues
is very useful, it can also be helpful to get into a habit of examining patients in outpatient
clinics in the same systematic manner as you would use in the examination. Your actions will
then become effortless and automatic in the high adrenaline situation of the examination.
It is important to get over the ‘pass/fail’ information and common conditions before mov-
ing on to rarities. As the clinical examinations and vivas all depend on verbal interaction to
impart your knowledge, it is vitally important to practice viva questioning with colleagues.
This will help you formulate a structure to your answers in an orderly fashion, and you
should quickly notice an improvement in your performance.
When anticipating cases for your clinical examination, common conditions in mobile
patients will appear most frequently. If you have ever been in the situation of organising
patients for a clinical examination you realise that the mobile elderly and those with stable
2 Section 1: Final FRCS vascular clinicals
chronic disease are easiest to recruit. There will always be a small pool of rarities such as
Klippel–Trénaunay syndrome or carotid body tumours that are willing to turn up for exami-
nations. It would be extremely unusual for an acute life-, or limb-threatening problem to
turn up to an examination. When performing the clinicals try to smile when introducing
yourself to the patients and thank the patients following the encounter (the same applies to
thanking the examiners!). The examination day can actually become quite repetitive for the
patients and examiners alike. Try hard to make them feel that their participation is valued,
and demonstrate that you can put patients at their ease.
Last minute revision can be helpful as topics recently read have a habit of turning up in
examinations. Despite this, turning up to the examination sleep deprived will not help your
performance; hence a sensible balance must be met. Research has demonstrated that moder-
ate sleep deprivation produces impaired cognitive and motor performance similar to alcohol
intoxication. You would not expect to pass an examination whilst intoxicated!
Section 1 Final FRCS vascular clinicals
Introduction to the examination

and clinical cases
Jonathan Smout and Asif Mahmood
Popliteal aneurysm
The basics
Popliteal aneurysms (PAs) are the commonest peripheral aneurysm (Figure I.1).
Approximately half are bilateral and half are associated with an aortic aneurysm. Conversely,
5–10% of patients with an abdominal aortic aneurysm (AAA) have a PA. The majority of
PAs present with distal ischaemic complications in either the acute or chronic situation.
The prevalence of the PA is thought to be around 1% for those in their eighth decade. When
presenting acutely with distal limb ischaemia, limb loss occurs in up to 50% of cases. PAs
almost exclusively occur in males. When treatment is indicated PAs are generally treated by
surgical exclusion although endovascular management is a newer development in selected
cases. Occasionally patients with patent PAs and very diseased run-off may be managed long
term with anticoagulation to reduce the risk of aneurysm thrombosis.
The case
Popliteal aneurysms are usually easy to identify as an expansile, or prominent, pulsation in
the popliteal fossa. The artery is best palpated against the tibia in the midline of the popliteal
fossa, with the knee in the extended position (or with a few degrees of flexion). The artery can
also be palpated with the knee flexed to 130°; in this position the popliteal fascia loosens to
aid palpation. However, in doing so the manoeuvre deepens the artery from the skin surface.
When thrombosed, PAs may be more difficult to diagnose clinically. It is important to assess
the distal circulation for evidence of embolisation into the foot or calf vessels. Other poster-
ior knee swellings include a Baker’s cyst or a semimembranosus bursa. Remember a PA can
exist at any point along the course of the popliteal artery and include the lower SFA as well.
In contrast, Baker’s cyst originates below the level of the knee joint as it extends beneath the
gastrocnemius muscle. A Baker’s cyst will often be associated with symptoms and signs sug-
gestive of degenerative arthritis of the knee joint. When present, an enlarged semimembrano-
sus bursa will be located medially under the popliteal edge of the semimembranosus muscle.
Questions
How do PAs present?
In the acute situation PAs usually present with distal ischaemia as a consequence of acute
thrombosis or distal embolisation. In the chronic situation they present with intermittent
Postgraduate Vascular Surgery: The Candidate’s Guide to the FRCS, eds. Vish Bhattacharya and Gerard
Stansby. Published by Cambridge University Press. © Cambridge University Press 2010.
Figure I.1 Angiographic image of a popliteal

aneurysm.
claudication as a result of chronic embolisation to the tibial vessels. Asymptomatic PAs are
often identified when screening patients with known aortic aneurysms. In contrast to aortic
aneurysms, rupture of a PA is a relatively rare occurrence (<5%). Other rarer presentations
result from local pressure on surrounding nerves and/or popliteal vein, and they can be the
cause of a deep vein thrombosis (DVT).
Tell me about the anatomy of the popliteal artery?
The popliteal artery commences when the femoral artery passes through the adductor hiatus in
the thigh. The vessel terminates as it splits into the anterior tibial artery and tibioperoneal trunk
at the lower border of popliteus muscle. The popliteal artery gives off genicular branches at sev-
eral levels to form a large collateral network about the knee joint. The artery is the deepest major
structure in the popliteal fossa, and sits beneath the popliteal vein. The tibial nerve lies superficial
to the popliteal vein. This organisation is not in the classical vein/artery/nerve configuration.
How would you approach the proximal and distal popliteal artery?
Although the popliteal artery can be approached via posterior or lateral incisions, the
most common routes of proximal and distal access are via the medial approach. The supra-
geniculate artery is accessed via an incision in the distal third of thigh along the anterior
border of the sartorius muscle. This muscle is mobilised posterior and the artery is iden-
tified between the medial intramuscular septum anteriorly and semimembranosus mus-
cle posteriorly. The infrageniculate popliteal artery is exposed medially via a longitudinal
incision, 1 cm behind the posterior/medial border of the tibia. The long saphenous vein
(LSV) is usually located posterior to the incision, and care must me made not to damage
it! A tissue plane is bluntly created between the soleus and gastrocnemius muscles. The
tendons of sartorius, gracilis and semitendinosus often require division for more prox-
imal access. The popliteal vein must be mobilised as this sits in front of the artery from
the medial approach.
What are your indications for elective repair?

In contrast to AAAs, where there is clear consensus on indications for intervention in
terms of size, this is not the case for PAs. Most surgeons would treat PAs exceeding 2 cm in
Introduction 5
diameter although patient factors may influence the decision for intervention. These fac-
tors may include patient fitness, the anatomical configuration of the PA, evidence of distal
embolisation or the presence of critical ischaemia. Most surgeons would view distal embol-
isation as a strong indicator for treatment, irrespective of aneurysm size. The presence of
mural thrombus on Duplex scanning, and significant distortion of the aneurysm should be
viewed as concerning signs. Prevention of aneurysm thrombosis is critical as limb loss is
markedly worse in the acute setting than for elective surgery.
When is thrombolysis utilised?

It has been observed that catheter based thrombolysis is associated with higher risks of
ischaemic complications when used to manage acute PAs in comparison to treating an
acute graft occlusion. During the lysis process a large volume of thromboembolic mater-
ial is destabilised and inevitably embolised distally. Studies have demonstrated that in at
least 10% of patients the limb acutely deteriorates during the lysis process. The main role
of thrombolysis is ‘on table’ to clear thrombus from the run-off vessels during the process
of surgical revascularisation. Thrombolysis can sometimes be used where no distal target
vessels are seen for surgical bypass on initial angiography and the limb is only in the ‘mar-
ginally’ threatened category.
What major problem faces endovascular treatment of PAs?

It is without doubt that covered endovascular stents can effectively exclude PAs and provide
an adequate conduit to supply blood to the lower leg. The main concerns regarding pop-
liteal stent grafts relate to their long-term durability. With the constant flexing of the knee
joint, the physical stresses challenge the integrity and positioning of popliteal stent grafts.
Endovascular exclusion was first described in 1994, and most of the literature reports come
from institutional case series. Endovascular treatment was mainly performed in asymptom-
atic patients, and initial results were poor. With the development of newer, more flexible
devices these results have improved, with 5-year patency rates of >75% being published.
Carotid body tumour

The basics
Carotid body tumours (CBTs) are paragangliomas derived from the neural crest ectoderm.
Paragangliomas are a rare neoplasm that can be found in the abdomen, thorax, and head and
neck region. They are usually considered benign and complete surgical removal results in
cure. The rule of ‘5%’ is often quoted as 5% are bilateral, 5% familial, 5% systemically malig-
nant and 5% locally reoccur. In reality nearer 10% are familial and in these patients one-third
have bilateral tumours. In contrast to retroperitoneal paragangliomas, where the majority
are hormonally active, <5% of CBTs are hormonally activity. In the neck paragangliomas can
also arise from the vagus nerve (glomus vagale), and jugular bulb (glomus jugulare). There
are three distinct groups of patients: sporadic (majority); familial; and hyperplastic (associ-
ated with chronic hypoxia).
The case
The thought of a CBT in your final examination might overwhelm you with fear, but it
shouldn’t! With their management being a relatively specialised subject you will not be
expected to know a large amount about CBTs or to have treated one. Twenty minutes
reading will provide you with all the knowledge you need to impress the examiners. The
examination case will take the form of a neck mass, postoperative case and/or a computed
tomography (CT)/magenetic resonance (MR) scan to review. The mass will be palpated at
the level of (or above) the hyoid bone, along the anterior border of sternocleidomastoid.
The CBT is firm in consistency and hence often referred to as a ‘potato tumour’, the mass
is laterally mobile but vertically fixed. The tumour is itself not pulsatile although a trans-
mitted pulsation may be present, or a pulsation may be palpable from an overlying exter-
nal carotid artery. Differential diagnoses to consider are cervical lymphadenopathy (are
there nodes elsewhere?), branchial cyst, carotid artery aneurysm (expansile mass), carotid
artery tortuosity or other cervical paragangliomas. Due to the anatomical distortion and
intraoperative bleeding, cranial nerve injury is more common when treating CBTs than
during carotid endarterectomy (glossopharyngeal, vagus [including laryngeal branches],
hypoglossal, accessory).
Questions
How do CBTs usually present?
A CBT usually presents as a painless neck mass (>50%), and can also present with compres-
sion of local structures or pain. The most common nerves to be compressed are the glos-
sopharyngeal, vagus and hypoglossal nerves. CBTs rarely present with symptoms of cerebral
ischaemia.
What are the typical CT/MR findings?

Due to the location of the CBT they typically splay the carotid bifurcation on angiography
(arterial/CT or MR) – Figure I.2. If the tumour does not display this feature it is more likely
to be another type of paraganglioma. The tumour derives its blood supply from the external
carotid artery. The tumours are usually well defined, and when large can encase the carotid
vessels.
What is the preoperative assessment?

All patients should have had a Duplex scan and neurological examination as part of their ini-
tial investigations. Further investigations include laryngoscopy (vocal cord assessment), plus
selective catecholamine screening in patients experiencing hypertensive episodes or those with
other neuroendocrine tumours (and contralateral CBTs on imaging). Magnetic resonance
imaging (MRI) scanning is valuable for diagnostic purposes, and to identify the cranial limits
of the tumour. Angiography is useful with larger tumours to identify their blood supply.
What classification system is used for CBTs?

The Shamblin classification (I to III) is used to stratify CBTs. Shamblin I tumours are
small and easily dissected from the vessel wall, Shamblin II tumours are of medium size
and partially encircle the carotid vessels. Shamblin III tumours are large (>4 cm) and more
completely encircle the carotid vessels. The Shamblin class III tumours classically require
excision and a vascular reconstruction with an interposition graft.
What endovascular interventions can be helpful in managing large tumours?

CBTs have a rich blood supply, and can be associated with significant perioperative blood
loss. In addition, operative bleeding can make a safe dissection more difficult. The tumours
Introduction 7
Figure I.2 Intra-arterial angiography of carotid body

tumour. Note splaying of the carotid bifurcation and
high vascularity of the tumour.
derive their blood supply from external carotid artery branches. Preoperative tumour
embolisation, or covered stent placement over the external carotid artery feeding vessels,
have both been advocated to reduce bleeding for particularly large tumours. Both of these
options remain controversial as they both pose a small risk of cerebral embolisation.
Femoral anastomotic pseudoaneurysms

The basics
Femoral anastomotic pseudoaneurysms (FAPs) often occur as a consequence of previous
aortobifemoral bypass surgery (Figure I.3(a) and (b)). The relationship of the anastomosis to
a constantly moving hip joint may be a contributory factor in the degeneration of the arter-
ial wall at the site of the anastomosis. Compliance may also be an issue in the pathogenesis
of FAPs at the junction between the elastic artery and an inelastic prosthetic material. Their
incidence at 5 years is around 5–10%; studies with longer-term follow up naturally demon-
strate higher occurrence for FAPs.
Continued smoking and wound infection at the time of the original operation are
thought to be risk factors for FAP development. In the case of aortobifemoral surgery the
aneurysms are often bilateral.
The case
FAPs are an ideal examination case with their chronic nature and obvious clinical signs.
On clinical examination femoral anastomotic pseudoaneurysms are easy to palpate due to
Figure I.3 Femoral anas-

tomotic pseudoaneurysms.
(a) Visible on bilateral groin
inspection as late consequence
of aortobifemoral surgery. (b)
Angiogram of same patient,
demonstrating bilateral femoral
artery aneurysms. Note full
length bilateral superficial fem-
oral artery occlusions.
(a)
(b)
their superficial location. The leg and abdomen should be inspected for scars from the pre-
vious surgery. The main differential diagnoses are of other aneurysm of the femoral artery
including atherosclerotic, mycotic, traumatic (including iatrogenic) and aneurysms related
to connective tissue disorders.
Questions
How do FAPs present?
FAPs usually present with a visible or palpable pulsatile groin lump. Due to local pressure
effects the patient can experience pain from peripheral nerve irritation (mass effect). Patients
may also present less frequently with distal embolisation, rupture and aneurysm thrombosis.
Introduction 9
What factors predispose to FAP formation?

Factors that may predispose to FAP formation include infection (early and late), poor sur-
gical techniques (i.e. inadequate tissue bites, undue graft tension), concomitant endarter-
ectomy, and the continued processes of arterial degeneration from atherosclerosis (hence
promoted by continued smoking).
What are your indications for surgery?

Aneurysms >2–3 cm (debatable) in diameter should be considered for surgery. Patient fac-
tors and rate of growth may influence the size for intervention. The presence of distal embol-
isation should lower the threshold for repair.
What is your approach to surgical repair?

The approach to surgical repair will be influenced by evidence of an infective process in the
aneurysm or groin. For chronic FAPs where infection is not suspected the aneurysm can be
repaired by placement of an interposition graft between graft and normal native artery. The
aneurysm should be controlled proximally and distally and the aneurysm should be opened
throughout its length. The anastomosis should be tension free and good bites of healthy
arterial wall should be taken. Foggarty occlusion balloons may be useful for controlling
back-bleeding side branches, as scarring in the groin may have made the initial dissection
difficult. For cases where infection is evident the surgery should include local debridement
and removal of infected graft material. The revascularisation should be performed through
healthy tissue using an autologous conduit where possible. In all cases tissue should be sent
for culture. The laboratory should be made aware that slow growing bacteria such as staphy-
lococcus epidermidis may be implicated. Antibiotic therapy should be discussed with the
microbiologists and reflect the likely causative organisms in your local area.
Vascular access
The basics
Dialysis utilising arteriovenous fistulae has been practiced since the 1960s. The classical
Cimino-Brescia fistula connects the radial artery to the cephalic vein at the wrist level. For
use as a dialysis conduit, flow rates of at least 200 ml min-1 are required and there needs to
be a suitable length of vein for access. Preoperative Duplex scanning should be performed in
the presence of poor peripheral pulses, equivocal veins, failed fistula in the limb, a previous
subclavian catheter or signs of proximal venous obstruction.
The case
The autologous arteriovenous (AV) fistula is the preferred method of dialysis in patients
with long-term, end-stage renal failure. In the examination you may be presented with an
arm to examine in a patient with an AV fistula. On inspection the arm should examined
for dilated superficial veins and scars from a current (or previous) fistula. The arm should
be inspected for scarring consistent with current needle punctures for dialysis. It is also
important to examine the arm for evidence of distal ischaemia, particularly in the digits. On
palpation the fistula should have a palpable ‘thrill’, or if occluded a thrombosed vein may be
palpable. The distal pulses should be assessed. On auscultation a machinery murmur will be
audible if the flow is sufficient. Proximal to the fistula there should be a sufficient length of
vein for two needle dialysis.
Questions
What is the preferred location for a primary AV fistula?
The preferred location for an AV fistula is as distal as possible (artery and veins permitting)
in the non-dominant arm. Although many surgeons utilise the radial artery and cephalic
vein at the wrist for primary fistulae, some surgeons have demonstrated good results with
fistulae made with these vessels in the anatomical snuff box. Distal sites are utilised to allow
new fistulae to be created at more proximal locations in the case of fistula failure, plus a
lower risk of distal ischaemia. The non-dominant arm is used to allow the recipient to per-
form activities during the dialysis process. Use of the non-dominant arm also means that if
any complications occur as a consequence of the intervention they will have a lesser impact
on the patient’s function.
What findings on palpation can suggest that a fistula is at risk?

In a fistula functioning for dialysis a thrill will be easily palpable in the majority of cases. The
presence of a weak thrill suggests the presence of inflow disease or narrowing at the site of the
anastomosis. Pulsatility in the fistula suggests the presence of a stenosis or occlusion in the venous
run-off. If a thrill is unclear in the examination you should also listen with a stethoscope.
Why is cardiac failure problematic in patients with AV fistulae?

Most fistulae used for dialysis have a blood flow of 500–1000 ml min–1. In patients with more
proximal fistulae, such as those in the antecubital fossa, the flow rates may be even higher
than this. These high flow rates can be demanding on patients with existing cardiac failure.
High-output cardiac failure can be diagnosed by observing a fall in pulse rate on manual
occlusion of the fistula (Branham’s sign). Patients with poor cardiac output will also be at
risk of fistula occlusion or insufficient flow for effective dialysis from poor flow.
What are the complications of an AV fistula?

• Bleeding – bleeding can occur in the postoperative period from the anastomosis or
divided vessels. Early exploration is advocated if there is any concern that the pressure
effects from the haematoma could compromise the fistula.
• Thrombosis – early thrombosis can occur due to technical problems with the
anastomosis or underlying arterial or venous disease. Re-exploration has been
advocated to correct these technical issues unless there is evidence of non-correctible
arterial or venous problems at the time of the operation. This view has been challenged
by others with the observation that re-explored fistulae often re-thrombosis, hence, the
creation of a new fistula at a more proximal location may be a better option.
• Failure to mature – an autologous AV fistula requires time for remodelling and venous
dilatation prior to commencement of dialysis. This should be at least 4–6 weeks. In
fistulae that remain small and have a poor flow investigations should be performed to
look for an underlying reversible cause.
• Steal – steal occurs more commonly in proximal than distal fistulae. Treatment can
involve surgical narrowing of the fistula or ligation or ligation of the retrograde blood
Introduction 11
flow into the fistula from the distal limb. In some situations the fistula must be ligated.
More complex interventions such as the Distal Revascularisation-Interval Ligation
(DRIL) procedure have been successfully performed whereby the fistula is ligated at
the artery just distal to the anastomosis. A distal bypass is then performed from a more
proximal to a more distal arterial location to re-perfuse the distal arm. If there was
a case of steal with digital ischemia available to the organisers this would be a good
examination case.
• Infection – superficial sepsis usually respond to antibiotics. Severe infection may
present with a major haemorrhage requiring fistula ligation.
• Aneurysm formation – although aneurysmal dilatation of the fistula is common it does
not usually require surgical intervention.
• Ischaemic neuropathy – monomelic ischaemic neuropathy occurs with steal and
usually presents within hours of surgery. This condition will not correct unless the
fistula is reversed. If the condition is left untreated permanent neurological damage can
occur.
Lymphoedema
The basics
Lymphoedema occurs when the circulation of lymph from the peripheral tissues back to
the central system is impaired. The majority of cases are due to obliteration of the lympatics
(80%) although proximal obstruction and lymphatic valvular dysfunction are also causes.
Lymphoedema is generally classified into primary and secondary causes. Primary causes are
sub-classified by the age of onset into ‘congenital’ (<1 year), ‘praecox’ (<35 years) and ‘tarda’
(>35 years).
The case
Lymphoedema of the lower limbs is a regular vascular examination case (lymphoedema of
the arm following treatment for breast cancer treatment is also commonly seen in the vascu-
lar or general surgical examinations). The clinical findings are of a leg swelling that does not
easily pit. The skin and subcutaneous tissues become fibrosed and less compliant with time,
hence the ability to ‘pit’ the skin is lost. The skin can also become thickened and hyperkera-
totic in appearance. Typical features of lymphoedema include a tree-trunk appearance of the
lower leg (NB step for the trunk at the ankle), ‘buffalo’ hump on the dorsum of the foot and
squared-off toes (in cross section) (Figure I.4). ‘Stemmer’s’ sign occurs when it is no longer
possible to pinch the skin on the dorsum of the 2nd toe. Chylous vesicles may appear on the
shins. It is important to examine the patient for other causes of leg swelling and evidence of
secondary causes of lymphoedema (see below).
Questions
What are the secondary causes of lymphoedema?
Secondary lymphoedema mainly falls into three categories; infective, malignancy and
iatrogenic. The commonest cause worldwide is from the parasitic infestation filaria-
sis. In the UK the commonest cause for lymphoedema is malignancy; either due to the
Figure I. 4 Lymphoedema of
the foot.
disease process itself or as a consequence of its treatment. Recurrent minor infections can
cause lymphoedema by chronic lymphangitis and progressive damage to the lymphatic
system.
1. Infective – parasitic (filariasis), bacterial (Staphylococcus, Streptococcus, TB).
2. Malignancy – infiltration or compression of lymphatic vessels and nodes.
3. Iatrogenic – radiotherapy, surgery (either direct surgery on the lymph nodes,
via ‘collateral’ damage during vascular surgical procedures, or by major surgery
obliterating lymphatic routes).
How is lymphangioscintigraphy performed?

The colloid of a radioactive isotope (technetium) is injected bilaterally into the interdigital
spaces between the 2nd and 3rd toes. The proximal progression of the isotope is assessed
using a high-resolution collimator, which takes images at regular intervals. Bilateral ilioin-
guinal node visualisation with isotope should occur within 1 hour in the normal individual.
The images demonstrate the progression (or hold up) of lymphatic flow. Abnormalities that
may be seen include an interruption of lymphatic flow, collateral lymph vessels, dermal
backflow, delayed flow, delayed visualisation or non-visualisation of lymph nodes, a reduced
number of lymph nodes and dilated lymphatics.
What findings on MRI are suggestive of lymphoedema?

On MRI scanning there is generalised subcutaneous oedema with a honeycomb pattern. The
latter is a result of the subcutaneous fibrosis. MRI can also demonstrate anatomical detail of
lymphatics and is complementary to scintigraphy.
What surgical procedures are performed to manage lymphoedema?

Surgery is infrequently performed to manage lymphoedema. The mainstay of manage-
ment is with complex decongestive therapy. Surgery is confined to severe cases where
conventional measures have failed. Surgery is divided into debulking procedures and
lymphatic bypass surgery. The classic Homan’s procedure involves incisions along the
Introduction 13
affected portion of the limb. The lymphoedematous subcutaneous tissue is excised with
preservation of the skin flaps. The skin flaps are cut to size and closed. Several proce-
dures may be necessary to manage severe disease. The Charles’ procedure is a slightly
more aggressive procedure whereby the affected subcutaneous tissue is resected down to
muscle fascia and then covered with a skin graft (Charles never actually performed this
procedure on the leg!). Although the results from the operation can be good in selected
cases, there is significant morbidity including delayed wound healing, infection and
nerve injury.
Diabetic foot
The basics
In around half of diabetic patients who develop foot ulceration, ischaemia is the primary
cause of a significant contributing factor. The remaining cases will predominantly be neuro-
pathic in their aetiology. Limb loss is 15 times more frequent in diabetic than non-diabetic
patients, and it is commonly preceded by foot ulceration. The diabetic foot is prone to ulcer-
ation for a number of key factors, including neuropathy, impaired vascularity at micro- and
macro-vascular levels, deformity and immune effects. Diabetes is frequently complicated by
renal failure, and this accelerates the development of vascular disease. Diabetic retinopathy
also makes wound care all the more difficult.
The case
Patients with chronic diabetic foot disease are in plentiful supply and often more than will-
ing to attend hospital for an examination. No matter where the foot lesion is located the
same basic assessment of the foot must be performed. During the examination you must
assess the vascularity, look for evidence of neuropathy, assess foot deformity, look for signs
of infection and comment on the wound condition. Where bone can be probed at the base of
an ulcer the likelihood of osteomyelitis is greatly increased. It is probably easiest to perform
a general inspection of the foot and then assess each aspect in turn.
Questions
How does diabetic neuropathy influence foot disease?
To describe the changes it is best to split them up into motor, sensory and autonomic. The
motor changes are thought to predominantly affect the small muscles of the foot with pres-
ervation of the long flexors and extensors. This leads to clawing of the toes with prominence
of the plantar metatarsal heads. Patients are often unaware of their sensory neuropathy, and
its presence impedes their ability to avoid injuries and protect healing wounds. Autonomic
neuropathy has several effects; first, it reduces sweating and causes dryness and fissuring
of the skin. Second, it alters foot microcirculation, causing blood to shunt from the skin
circulation.
How is neuropathy tested?

The simplest and most practical method for testing for neuropathy utilises pressure percep-
tion with a 10 g nylon monofilament (Figure I.5). Buckling of the monofilament with pres-
sure indicates a skin pressure of 10 g. The skin is tested in a standardised pattern at several
Figure I.5 Testing neuropathy with a 10 g

monofilament.
points on the foot. Impaired sensation demonstrated by this method has been shown to
correlate with the risk of ulceration. The presence of neuropathy can also be assessed using
a biothesiometer (vibration sensation), or by nerve conduction studies.
What are the principles in managing acute diabetic foot disease?

• Vascular assessment and revascularisation where necessary.
• Treating infection – draining sepsis, antibiotics.
• Wound care – debridement, wound dressings, and ‘off loading’ of wound.
Following wound healing, prevention is of paramount importance.
What is ‘Charcot’s foot’?

Charcot’s neuroarthropathy is a complex condition whereby bony destruction and joint
deformity occurs. The changes result from neuropathy, and diabetes is the commonest cause
of neuropathy in the Western world. Presentation is often delayed due to the impaired sen-
sation of pain. In the early phase the foot is often warm and swollen, and these changes
are often mistaken for cellulitis. As the condition progresses the foot becomes structurally
deformed with collapse of the medial arch. There is a rocker bottom deformity to the foot
with bony prominences. The skin temperature is often several degrees higher in the affected
foot. Bony changes will usually be obvious on plain X-rays, although in the initial phase the
imaging may be normal. Once the diagnosis is suspected the foot must be immobilised until
the deformity has stabilised.
What imaging aids the diagnosis of osteomyelitis?

Osteomyelitis is often difficult to diagnose in the diabetic foot. The presence of neuropathy
and adjacent ulceration can complicate the situation. In around 70% of cases there will be
plain radiographic evidence of bony destruction. Serial plain radiographs in suspected cases
are valuable in looking for changes in the bony integrity; the serial radiographs should be
taken at 2-week intervals. Three phase bone scans and isotope white cell scans can also be
used in combination with plain radiographs to improve accuracy. In combination with
plain radiographs, these modalities are sensitive for osteomyelitis in over 90% of cases. MRI
Introduction 15
scanning is now becoming established as the investigation of choice in diagnosing osteo-

myelitis and deep infections in the diabetic foot. MRI findings of osteomyelitis include a
decreased bone marrow T1 signal, increased T2 signal, and post-gadolinium enhancement.
MRI can also identify associated abnormalities such as cellulitis, abscess formation, sinus
tracts, and cortical bone destruction.
Complications of carotid endarterectomy

The basics
The main complications of carotid endarterectomy (CEA) include stroke, nerve injury,
haemorrhage, cardiac ischaemia and death. Carotid patch infection is fortunately a rela-
tively infrequent complication (<1%), and you are unlikely to see one in the examination.
The case
In previous examinations candidates have been introduced to a patient with a visible scar
from a CEA. The candidates have been asked about the patient’s likely intervention, and then
questioned about various aspects of carotid surgery. Pre- or postoperative carotid patients
are a common examination case.
Questions
Which cranial nerves are at risk during carotid endarterectomy?
The cranial nerves most commonly injured during carotid surgery include the hypoglossal,
vagus and laryngeal nerves.
• The hypoglossal nerve is the most commonly injured major nerve during CEA. The
hypoglossal nerve crosses the internal carotid artery and external carotid artery near
the upper limit of internal carotid artery dissection. Due to its location it runs the risk
of division, diathermy or traction during surgery. Hypoglossal nerve injury classically
presents with tongue deviation towards the side of nerve injury. Injury during CEA
occurs in 5–10% of cases and is usually a transient phenomenon. It is more common if
the nerve has been mobilised.
• The vagus nerve is usually located posteriorly in the carotid sheath. This posterior
location makes it vulnerable to clamp injuries if the arterial clamp is carelessly placed
beyond the artery.
• The superior laryngeal nerve is susceptible to injury where it descends behind the
internal carotid artery and then passes posteriorly close to the superior thyroid
artery. Injury to the superior laryngeal nerve probably goes unnoticed most of the
time.
• The non-recurrent laryngeal nerve, when present, is at particular risk during carotid
endarterectomy. Non-recurrent nerves occur in 0.5–1% of people and are commonest
on the right-hand side. In its non-recurrent course, the nerve passes transversely from
under the carotid sheath and will be at right-angles to the normal position.
• The glossopharyngeal nerve is rarely damaged during a straightforward CEA. The
glossopharyngeal nerve is susceptible to injury when a more extensive cranial
dissection is performed. The nerve is deep too and courses in a similar direction to the
posterior belly of the digastric muscle.
• The spinal accessory nerve exits the skull in posterior direction just deep to the styloid
process. The nerve is susceptible to injury during ‘high’ carotid dissections or during
the retrojugular approach to the carotid artery. The spinal accessory nerve provides
motor innervation to the trapezius and sternocleidomastoid.
Would you use local or general anesthesia?

The general anaesthetic versus local anaesthetic for carotid surgery (GALA) trial showed no
advantage for either. It remains a choice between the patient and surgeon concerned. You
should have a view on what you should do in your practice and be prepared to justify it.
Does patching reduce the risk of perioperative stroke?

The Cochrane Stroke Review Group has published a meta-analysis comparing carotid patch-
ing to primary closure. The analysis demonstrated a reduction in the ipsilateral stroke rate
and restenosis with patching. Most surgeons use a Dacron patch although some may use
vein patches or material such as bovine pericardium. The best current practice is to use a
patch on virtually all patients.
What should be performed prior to bilateral carotid surgery?

The vocal cords should be checked for evidence of recurrent laryngeal nerve injury from the
first procedure. Bilateral nerve injury can cause significant difficulties with voice, swallowing
and obstruction of the upper airway – the neutral position of the cords is in the midline.
What is the risk of disabling stroke or death?

The risk of death or disabling stroke in both the European and North American carotid
endarterectomy trials (NASCET and ECST) was approximately 3%. The combined rate of
death and all strokes is virtually double this figure. Data from the European and North
American asymptomatic carotid endarterectomy trials (ACST and ACAS) 30-day any stroke
or death rates are near 2.5%, suggesting that the asymptomatic patients and/or the surgery
is lower risk. Vascular units should strive to audit their own results and present their own
complication rates when consenting patients.
What is the frequency of haematoma formation?

Data from randomised trials often identify much higher rates of complication than self-
reported case series. In the GALA trial haematoma rates were close to 10% and just over one
third of these cases required re-exploration for bleeding. Haematoma in the neck can result
in airway obstruction and should be considered a serious problem as per thyroid surgery.
The mechanism of airway obstruction is through a combination of laryngeal oedema plus
direct compression. Re-intubation should be considered sooner rather than later where air-
way compromise is suspected.
When would you use a carotid shunt?

Unless conclusive trial evidence becomes available controversy will reign over approaches to
carotid shunting. Policies include shunting all patients, selective shunting and not-shunting.
You should decide on your policy and be able to justify it in the examination. An appropri-
ate policy is to shunt all CEAs under general anaesthetic and selectively shunt those under
local anaesthesia.
Introduction 17
How is cerebral perfusion monitored during CEA?

Although sophisticated methods exist to measure cerebral perfusion such as xenon washout,
infra-red spectroscopy and electrophysiological studies, none are ideal for everyday carotid
practice. Commonly used methods include awake testing (local anaesthetic procedures),
measurement of stump pressures, and Transcranial Doppler (TCD). In cerebral monitor-
ing you want a device that is simple and cheap to use, is able to identify ipsilateral cerebral
blood flow, can identify embolic events and looks at neurological function. Unfortunately,
none of the methods available can satisfy all of the requirements fully. Awake testing may
seem the most ideal, but it will not necessarily warn you of embolisation; once cerebral
compromise is apparent inserting the shunt may be more difficult due to patient agitation.
TCD can assess reduction in cerebral blood flow (>50% reduction in middle cerebral artery
flow) and provide evidence of embolic events before and during shunt insertion; however,
it does not demonstrate whether the patient’s neurology has been affected. Stump pressure
measurement is cheap and readily available, stump pressure <50 mmHg are taken to indi-
cate the need for a shunt. Stump ‘pressures’ are taken as a proxy measurement of cerebral
blood ‘flow’, and will not identify changes during the shunting period, unless repeatedly
measured.
How would you manage a postoperative cerebrovascular event (CVE) following recent carotid
endarterectomy?
The management of a postoperative CVE will be contentious and dependent on the tim-
ing and facilities available as well as when the event occurs. Whatever investigations and
treatments are instituted, all patients should have their cerebral perfusion and oxygenation
optimised and this will usually require admission to an area such as a high dependency unit
(HDU) or an intensive therapy unit (ITU). Intubation and ventilation may be needed if
the conscious level is impaired or the patient is confused. The intention with these patients
is to identify those with carotid artery thrombosis, prevent further events (correct tech-
nical problems, antiplatelet therapy, mange hypertension and hypotension) and to identify
patients with haemorrhagic strokes.
When the CVE is apparent on waking, or occurs in the recovery room, the patient is
in the ideal location to return to theatre to correct carotid artery thrombosis or technical
problems. Ideally a Duplex scan should be performed immediately to see if the endarter-
ectomised vessel is patent or if there are technical problems with the endarterectomy. If
the scan is normal and the surgeon was ‘happy’ with the operation the benefits of re-oper-
ation will be questionable, and perhaps the best course of action is to exclude other causes
of embolisation and optimise antithrombotic therapy. These patients should have a cere-
bral CT scan as soon as possible to exclude a haemorrhage, particularly where enhanced
antithrombotic therapy is being considered. The scan should also include angiography,
this is particularly important if the aortic arch and intracranial vessels have not been pre-
viously imaged.
In patients where the blood pressure has been labile and there have been periods of sig-
nificant hypertension a haemorrhagic stroke should be more strongly considered. In these
patients a cerebral CT should be performed immediately to exclude a haemorrhage. It
should be remembered that cerebral haemorrhage occurs in the minority of patients, and
a case review has suggested that this represents about 10% of postoperative strokes. CVAs
that occur after the immediate postoperative period are less likely to derive the same benefit
from re-exploration, and should be urgently imaged with carotid Duplex (if available), CT
and CT angiography (CTA).
Buerger’s disease
The basics
Buerger’s disease (thromboangiitis obliterans) is an inflammatory arteriopathy that pre-
dominantly affects small- and medium-sized arteries. The changes are pathologically
distinct from atherosclerosis. The lower extremity is predominantly affected although
this is not exclusive. The disorder classically affects young male smokers; however, the
pattern of disease is changing. Leo Buerger first described the condition in pathologi-
cal specimens in 1908, and termed the disease thromboangiitis obliterans. In contrast
to the common clinical presentation of atherosclerosis with intermittent claudication,
patients with Buerger’s disease tend to present with rest pain and tissue loss. Near to half
of patients diagnosed with Buerger’s disease who continue to smoke end up requiring an
amputation of some sort. The exact pathogenic mechanism linking Buerger’s disease to
tobacco is unknown.
The case
Consider the diagnosis in any young patient with symptoms or signs of critical ischaemia.
Patients with Buerger’s disease are young and mobile enough to attend examinations, and
they have good clinical signs. In the clinical encounter the level of vascular disease should
be established and other alternative diagnosis considered. Do not make this diagnosis too
glibly but suspect it in young male smokers with significant ischemia and preserved proximal
pulses.
Questions
What are the pathological and radiological features of Buerger’s disease?
• The pathological features at a microscopic level involve an acute hypercellular
thrombosis causing arterial occlusion. There is a striking perivascular inflammation
that can also affects veins. Despite the intravascular thrombosis patients with Buerger’s
disease have not been demonstrated to be hypercoagulable.
• The radiological features are of relatively normal arteries to the knee level with abrupt
occlusions of the tibial vessels with ‘corkscrew’ collateral feeding of the distal vessels at
the ankle (Figure I.6).
How is the diagnosis made?

There are several diagnostic criteria described (Tel Aviv, Oregon) for Buerger’s disease. The
consistent factors in these schemes are:
• exclusion other causes;
• tobacco use;
• distal extremity disease (femoral pulses nearly always preserved);
• young age of onset (< 45);
• The presence of upper limb disease, phlebitis migrans or Raynaud’s phenomena, and
radiological signs strengthen the diagnosis.
Introduction 19
Figure I.6 Peripheral angiogram of tibial vessels

in Buerger’s disease. Arrow indicating corkscrew
collaterals.
What other forms of vasculitis are relevant to the vascular surgeon?

• Takayasu’s arteritis – this affects larger elastic arteries such as the aorta and its branches.
The majority of patients are female and present in their 2nd and 3rd decade. Two
phases exist of acute pre-pulseless and chronic pulseless.
• Giant cell arteritis (temporal arteritis) – this is a granulomatous vasculitis affecting
large- and medium-sized arteries, with the cranial vessels most commonly affected.
This can present with cranial ischaemia (visual loss, jaw ischaemia, headaches) or with
chronic constitutional symptoms.
• Polyarteritis nodosa – this is a necrotising vasculitis affecting small- and medium-sized
arteries. The disease is more common in males and usually presents in middle age.
Renal and gastrointestinal involvement is common.
• Drug induced – penicillins, sulphonamides, NSAIDs, cocaine, etc.
Connective tissue disorders should be considered in patients with digital ulceration, without
typical findings of atherosclerosis. Other rarer vascular disorders are also considered in the
popliteal entrapment scenario.
How is Buerger’s disease managed?

The absolute goal for managing Buerger’s disease is establishing permanent abstinence from
smoking. Aspirin should be prescribed for its antiplatelet effects, and analgesics for pain
control. Prostaglandin infusions may help with symptom control, although it is uncertain if
they alter the progression of tissue loss. Distal ischaemic lesions will often auto-amputate or
may require surgical amputation. Awaiting demarcation is helpful even if surgery is planned.
Antibiotics may be needed intermittently for any infective episodes. A lumbar chemical
sympathectomy is helpful in selected cases for chronic pain control.
Vascular malformations
The basics
Vascular anomalies fall into two main categories, haemangiomas and vascular malfor-
mations. Haemangiomas will be evident at birth, and have a distinct natural history
(proliferation, plateau and involution). Most will disappear during the first decade of
life. Previous terminology utilised the term ‘haemangioma’ for both lesions appearing
around birth and those appearing later in life. More recent terminology has defined
the self-involuting tumours as haemangiomas (hence the port wine stain is no longer
considered a haemangioma). Vascular malformations grow during childhood, and may
enlarge following hormonal change, trauma and sepsis. Most vascular malformations are
sporadic although some can be part of a specific syndrome, such as Klippel–Trénaunay.
Vascular malformations are derived from aberrations in vasculogenesis, and generally
do not undergo spontaneous involution. Although several classifications exist, they are
most simply considered in two main clinically relevant categories: fast- and slow-flowing
lesions.
• Fast-flow: arterial malformation, arteriovenous malformations and arteriovenous
fistulae.
• Slow-flow: venous malformations, lymphatic malformations, capillary malformations.
First-line investigations include Duplex and MRI scanning. Duplex scanning demonstrates
flow dynamics and morphology of the lesion. MRI scanning provides more detailed infor-
mation on structure and relationship to other soft tissues. MRI is also able to differentiate
between slow- and fast-flow malformations. Invasive investigations such as arteriography
and venography are still performed second line, and can provide valuable additional infor-
mation when planning treatment.
The case
Due to the diversity of lesions we will run through the more common groups of malforma-
tion for this scenario.
Capillary malformations
Capillary malformations are intra-dermal vascular anomalies. The malformations appear as
pink/red areas of discolouration, and can occur throughout the body. The lesions can cause
hypertrophy of the surrounding soft tissues. Capillary malformations may be spontaneous
or part of a syndrome such as Klippel–Trénaunay or Sturge-Weber. Imaging will often be
performed to look for associated abnormalities. Pulsed dye laser is an established treatment
for these lesions.
Venous malformations
Venous malformations are the most prevalent vascular malformation and tend to occur in
the head and neck. Because of their slow flow they often take considerable time to enlarge.
The lesions have previously been referred to as ‘cavernous haemangiomas’. On examination
they will be deep blue in colour and easily compressible. The lesions can calcify, and local
thrombosis can cause pain. Limb hypertrophy is seen with some extremity lesions. Treatment
is conservative with compression where possible. Sclerosant therapy is usually the first-line
Introduction 21
intervention, and repeat treatments are often needed. Surgical excision can be performed for
severe symptoms with or without preoperative sclerotherapy.
Arteriovenous malformations
Arteriovenous malformations are fast-flowing connections that bypass the capillary bed.
The lesions are usually apparent at birth, and enlarge in size as their blood flow increases.
The lesions are usually warm and pink/blue in colour. Arteriography is often required to
establish the anatomy of their arterial supply. Small lesions can be excised (including the
feeding vessel). Larger lesions often require a combination of embolisation and surgical exci-
sion. Ligation of the feeding vessel alone can cause collateralisation, making further treat-
ment more complex. When these lesions enlarge they can become destructive and lead to
cardiac failure.
Lymphatic malformations
Lymphatic malformations have historically included lymphangiomas and cystic hygromas.
These lesions are slow flow and usually occur in the cervical region. The majority will be
apparent within the first years of life. These lesions can be associated with both soft tissue
and skeletal overgrowth.
• Cystic hygroma: this consists of a collection of lymphatic sacs that have failed to
connect properly with the normal channels. The lesion are often found in the posterior
triangle at the base of the neck. They can rapidly fill up in response to an infection
or trauma, and can become very large. On palpation the lesion will be smooth and
transilluminate spectacularly.
• Lymphangioma circumscriptum: this is at the other spectrum where small localised
lymphatic vesicles are present that do not connect to the normal lymphatic system.
They are usually found around the shoulder, axilla, groin and buttocks. If these lesions
contain old blood they may turn brown in colour.
MRI scanning is useful in defining the anatomy of the larger lesions. The main sources of
symptoms for these lesions are infection and intra-lesional bleeding. Sclerotherapy has
been performed with a variety of agents with acceptable results. Surgery is reserved for
severe symptoms and recurrence is high. The aim of surgery is complete excision of the
lesion to minimise the risk of recurrence. Surgery can be complex and involve multiple
procedures. Lymphangioma circumscriptum requires wide local excision if treatment is
required.
Vascular malformations (specific syndromes):

Klippel–Trénaunay syndrome
The basics
Klippel–Trénaunay syndrome (KTS) was first described by the French physicians Maurice
Klippel and Paul Trénaunay in 1900. Most cases are sporadic although there have been
reports that the condition has occurred in an autosomal dominant pattern. KTS is a com-
plex and variable syndrome; hence treatment is planned on a case-by-case basis. The venous
component of the disease tends to be the most problematic. The physical signs will become
evident early in life and there is no gender predominance.
Figure 7 Port-wine’ staining and varicose veins in

Klippel–Trénaunay syndrome. There is a scar over the
hip from a leg shortening procedure.
The case
KTS is an excellent vascular case with its various manifestations and physical signs
(Figure I. 7). The clinic findings are discussed below.
Questions
What are the pathological components of KTS?
• Venous abnormalities (varicosities).
• Bony and soft tissues hypertrophy.
• Cutaneous angiomata (‘port wine’ stains).
Each component of the syndrome may be present to a varying degree. The limb hypertrophy
can be secondary to increased bony length and/or increased soft tissue girth. Abnormally
developed lymphatics have also been associated with this syndrome.
What is the distribution of varicosities in KTS?

Unlike the typical distribution of greater or lesser saphenous varicosities, the varicosities
seen in KTS tend to be located on the lateral side of the thigh and lower leg. The ‘port wine’
stains from cutaneous angiomata tend to be located in similar lateral distribution to the vari-
cosities. The ‘port wine’ stain has a distinct, linear border, and the lesion possesses neither a
proliferative nor a regressing phase.
What concerns would you have over performing superficial venous surgery in KTS?
In KTS the deep venous system can be abnormally developed. The superficial veins may pro-
vide an important role in venous function; hence, they should be left intact unless adequate
venous drainage has been demonstrated.
Introduction 23
What surgical options are available to treat KTS?

The surgical options for KTS have mainly included de-bulking procedures and venous sur-
gery. The de-bulking procedures are only really considered for severe cases where conserva-
tive options have failed. Venous surgery in KTS is associated with a higher rate of recurrence
than with conventional venous anatomy.
What non surgical options are available?

Compression therapy is an important treatment modality in the management of limb swelling.
Maintaining good compression therapy can reduce pain, swelling and inflammation. The poten-
tial side effects of compression therapy include displacement of fluid into other locations, and
the compression therapy itself may impede circulation through the limb. As with lymphoedema
management, massage therapy can provide an improvement in symptoms. Liquid or foam scle-
rotherapy offer a less invasive option to treat the varicosities than conventional surgery.
Vascular malformations (specific syndromes): Parkes-Weber

syndrome?
Parkes-Weber syndrome was first described by Frederick Parkes-Weber, an English derma-
tologist and physician to Queen Victoria. Parkes-Weber syndrome is similar in many of its
features to KTS; however, in KTS the cutaneous malformations are slow-flowing, whereas in
Parkes-Weber syndrome the vascular malformations are fast-flowing arteriovenous abnor-
malities, with associated limb hypertrophy (haemangiectatic hypertrophy). So, in essence,
Parkes-Weber syndrome is a fast-flow arteriovenous malformation in association with the
other abnormalities found in KTS such as cutaneous capillary malformation and skeletal or
soft tissue hypertrophy.
Vascular malformations (specific syndromes): Proteus syndrome

Proteus syndrome is an extremely rare congenital condition and unlikely to be seen in a
clinical examination. The condition is highly variable in appearance, and is named after
the Greek sea-god Proteus, who could change his shape at will. The condition consists of
cutaneous abnormalities (nevi), vascular abnormalities (capillary, venous, or combined),
plus skeletal and soft-tissue abnormalities such as hemihypertrophy. Proteus syndrome is
extremely disfiguring as a result of an overgrowth of skin, bones, muscles, fatty tissues, and
blood and lymphatic vessels. The changes often only occur over half of the body. The condi-
tion was publicised in the 1980s film The Elephant Man, about the life of Joseph Merrick, a
Proteus sufferer.
Vascular malformations (specific syndromes): popliteal

entrapment syndrome and cystic adventitial disease
The basics
Popliteal entrapment usually presents with calf claudication in young athletic individuals.
The symptoms may commence following a period of strenuous activity. Foot pulses will be
normal at rest unless a complication of the disease process has occurred. Sudden onset clau-
dication in a young person may indicate a vessel occlusion. Popliteal entrapment syndrome
can be classified into its anatomical and functional varieties. Long-term complications
include localised fibrosis and stenosis, aneurysmal dilatation, and embolisation from the
diseased segment of artery.
Making a diagnosis of this condition is particularly problematic for the functional group,
as at least 10% of the asymptomatic population demonstrate compression of the popliteal
artery with active dorsal or plantar flexion of the ankle joint. The diagnosis of popliteal
entrapment is usually made in younger individuals with symptoms of claudication and evi-
dence of vascular compromise on active ankle flexion. Primary atherosclerosis, other arter-
ial disorders, and alternative causes of leg pain must be excluded. Anatomical entrapment
should be corrected by removal of the entrapment mechanism (usually release of the medial
head of gastrocnemius) with or without resection of the involved segment of artery. This
prevents further damage to the artery and distal vasculature.
The case
The patient should have a lower limb vascular examination. Attention should be made
to assessing the quality of the ankle pulses in the plantar- and dorsi-flexed positions. The
patient should be inspected for venous skin changes, vasculitic lesions and examined for
sources of embolisation. The candidate should also indicate to the examiner that they would
perform a musculoskeletal examination of the lower limb.
Questions
What two important embryological factors are pertinent in the development of anatomical
popliteal entrapment?
Two important embryological factors that can be implicated in the development of popliteal
entrapment involve the differing embryonic origins of the mid portion and the distal pop-
liteal artery, and secondly, the migration of the medial head of gastrocnemius.
• In the definitive human anatomy the mid portion of the popliteal artery is a remnant
part of the primitive axial artery and definitive distal vessel (original axial vessel
lying beneath the popliteus muscle) from a more superficially placed vascular plexus.
This process of vascular development occurs between the 8th and 12th week of
embryology.
• The medial head of gastrocnemius migrates from a lateral location during embryonic
development. This process occurs around a similar time to the changes in the popliteal
artery.
What are the common configurations causing popliteal artery compression?

• The popliteal artery can be located medially to the medial head of the gastrocnemius.
This abnormal position can occur to a varying degree depending on the final location
of the gastrocnemius muscle.
• The popliteal artery may be located within the medial head of the gastrocnemius
muscle.
• The popliteal artery may exist deep to the popliteus muscle.
• In functional cases it is thought that muscle hypertrophy from exercise plus a
‘vulnerable’ location of the artery between the gastrocnemius heads predisposes to the
condition.
Introduction 25
What other structures can be involved in the compression syndrome?

Popliteal vein and the tibial nerves can also be compressed.
Do you know any classification schemes for popliteal entrapment?
A classification scheme for popliteal entrapment has been suggested by Levien and Veller [1].
Types I to III involve misplacement of the artery in relation to all, or part of, the medial head of
gastrocnemius. In Type IV the popliteal artery is developed main in an anomalous way beneath
the popliteus muscle. This layout is therefore unrelated to the positioning of the gastrocnemius
muscle. Type V was subsequently suggested as an addition to include cases where the popliteal
vein is involved. Functional entrapment, where there is compression of the vessel in stressed
positions without any apparent anatomic abnormality, is termed Type VI.
What other rare vascular disorders can causes arterial claudication in the younger patient?
• Fibromuscular dysplasia (FMD) – FMD is the commonest cause of ‘renal’ hypertension
in children. Most FMD affects the media of the arterial wall. The renal and carotid
vessels are most commonly affected, although the external iliac artery (EIA) is the most
commonly involved vessel in the lower limb vasculature. Disease of the EIA usually
presents with claudication, although it can be complicated by embolisation, aneurysm
or thrombosis. The classic appearance on arterial imaging is of a ‘string of beads’,
angioplasty has been successfully used to treat this disease.
• Persistent sciatic artery – the sciatic artery is the embryonic axial limb artery. In the
normal individual the majority of the vessel obliterates, apart from the the segments
becoming the internal iliac artery, part of the popliteal artery and the peroneal
artery. The condition may present with a pulsatile mass in the buttock. Aneurysmal
degeneration of the anomalous artery can occur due to trauma in the sciatic foramen.
The patent sciatic artery is associated with hypoplasia of the iliofemoral vessels. The
blood supply through these abnormal vessels may be inadequate during exercise, and
the diseased sciatic artery may acutely thrombose.
• Cystic adventitial disease (CAD) – CAD of the popliteal artery is thought to exist
due to inclusion, or extension, of mucin-secreting structures between the media and
adventitia of the popliteal artery. The condition usually presents with claudication
in the fourth and fifth decades. On clinical examination, flexing the knee joint may
cause the distal pulses to disappear. The typical appearance on angiography is of an
‘hourglass’ narrowing of the popliteal artery with normal distal vessels.
• Endofibrosis – arterial endofibrosis is a recently discovered condition that can affect
highly trained athletes, with cyclists predominantly at risk. Repetitive movement of the
hip joint and the cycling posture are thought to lead to chronic arterial injury, resulting
in progressive intimal thickening. The endofibrosis most often affects the external iliac
arteries. The presence of an arterial pressure drop is useful sign as peripheral pulses
and ankle brachial pressure index (ABPI) will usually be normal at rest. The condition
has been treated with resection and revascularisation using autologous vein. Prosthetic
materials should be avoided due to compliance issues.
• Premature atherosclerosis – lipid disorders and hyperhomocystinaemia should be
considered.
• Dissection – aortic dissection can be complicated by acute limb ischaemia.
Claudication may be a longer-term consequence of the event. Young patients with
acute dissection usually have significant hypertension or a collagen disorder such

as Marfan’s. Isolated spontaneous dissections have been described in the peripheral
vasculature.
• Embolisation – as with dissection, embolisation should present with a well-defined
acute event. Claudication may be an ongoing consequence if the presentation is delayed
or embolus is left untreated. Proximal sources of embolisation should be sought.
• Drug induced arteriopathies – cocaine, amphetamines, ergot, etc.
• Pseudoxanthoma elasticum (PXE) – PXE is a rare genetic disorder that produces
progressive calcification and fragmentation of elastic fibres in the skin, cardiovascular
system and retina. Extensive arteriosclerosis often occurs in the third or fourth decade
of life. The disease tends to spare the aorta, but involves lower limb arteries, producing
intermittent claudication. Patients with PXE often have coronary and valvular heart
diseases.
• Vasculitis.
Venous disease: varicose veins

The basics
Varicose veins are a common clinical condition and an extremely common examination
case. Although many patients may seek treatment for cosmetic reasons, a broad range of
symptoms such as heaviness, itching, aching, mild swelling, cramps can be attributed to
varicose veins. Symptoms tend to be worse towards the end of the day, and after prolonged
episodes of standing. In females the symptoms are often worse around the time of menstrua-
tion. Varicose veins frequently present as uncomplicated entities, but can also be associated
with changes of chronic venous insufficiency (CVI). The main patterns of venous incom-
petence involve the long (greater) and short (lesser) saphenous veins. With the advent of
Duplex scanning a Giacomini vein is often mentioned in reports. This is a thigh extension
from the short saphenous vein that joins with the long saphenous vein first described by
Giacomini.
The case
Varicose veins should initially be examined in the standing position.
• Inspection – comment on the extent of varicosities and their distribution in relation
to the superficial venous systems. The lower abdominal wall and perineal areas should
also be inspected for venous collaterals. Inspect for associated skin changes or areas of
ulceration suggestive of CVI. Look for a bluish tinge to the skin in the groin suggestive
of a saphena varix. Later in the examination you will want to examine the patient in
the supine position to ensure that the varicosities disappear on leg elevation (venous
occlusion, tricuspid disease).
• Palpation – the skin should be examined for oedema, thickening, and the veins
compressed to establish their patency. The saphenofemoral and saphenopopliteal
junctions should be palpated for the presence of a varix. The lower limb pulses should
be examined for co-existing arterial disease (+/- ABPI). Where abdominal wall venous
collaterals are seen Harvey’s test can be performed to determine the direction of the
flow of the veins. The test is performed by placing two fingers on a segment of vein
Introduction 27
several centimetres apart. By sliding one finger along the vein to empty it and then
releasing one finger (repeated in both directions) the direction of venous filling can be
determined.
• Percussion and auscultation – the ‘Tap’ test is best performed where a long segment of
LSV is palpable. Normally when a column of blood is present in the LSV, transmission
of a percussion wave should only occur in an antegrade direction. Where the valves
in the system are incompetent is a percussion wave can also travel in a retrograde
direction. Auscultation may be useful when looking for evidence of an arteriovenous
shunt (‘machinery’ murmur).
• Tourniquet test – it is worth understanding (and being able to perform) the tourniquet
test, although it is infrequently performed in modern clinical practice.
• Hand Held Doppler (HHD) examination for junction reflux. The HHD can be
used to identify reflux at the saphenofemoral junction, popliteal fossa and also in
the long saphenous vein itself, whilst the patient is in the standing position. When
listening for reflux, care must be taken to only apply light pressure to the area of
interest, as veins can easily be compressed, hence abolishing audible venous flow.
Significant reflux (bidirectional venous blood flow) is taken as >0.5 s in duration.
The saphenofemoral junction is located medial to the femoral artery, 2 cm below
the level of the pubic tubercle. Reflux can be augmented by compression of the calf
muscle or during a Valsalva manoeuvre. The saphenopopliteal junction is much
more variable in location, hence HHD examination is less reliable. Reflux in the
popliteal fossa is detected by identifying the arterial signal close to the midline,
and then moving the probe laterally. The calf muscle is again squeezed to augment
reflux. Popliteal fossa reflux will either relate to short saphenous, gastrocnemius or
popliteal venous incompetence. Examining the LSV at knee level with HHD is useful
where junction reflux is not present, but the varicosities appear distributed in the
LSV territory.
Following the peripheral examination you would then wish to examine the abdomen,
and rectum, vagina and testes where appropriate for evidence of malignancy or pelvic
mass. Prior to intervention, or where varicosities have newly developed, a pregnancy
test should be performed (pregnancy causes pelvic venous compression, and increased
progesterone levels cause smooth muscle relaxation and alterations in the structure of
collagen).
Venous disease: chronic venous insufficiency (CVI)

and ulceration
The basics
The prevalence of venous ulceration increases with age, with a rate of 20 per 1000 in subjects
over 80 years of age. Healing of venous ulcers is a major cost burden to society with the best
healing rates in specialist centres of 70% at 3 months. Venous ulcers are often recurrent and
usually have a history of venous disease. The skin at the site of the ulcer is often tender, painful
and inflamed prior to the development of the ulcer, and the ulcer formation can often be trig-
gered by minor trauma or scratching of the skin. Around half of patients with venous ulcers
have superficial venous incompetence without deep incompetence or obstruction. Chronic leg
Figure I.8 Image of lower limb venous skin

changes – lipodermatosclerosis and pigmentation. .
ulcers are attributable to venous disease in 60–80% of cases; in around 20% of these cases there
is underlying arterial insufficiency. Venous ulcers are also often associated with diabetes. Pure
venous ulcers are usually relatively painless; if pain is a significant feature you should consider
an inflammatory or arterial condition as a cause. Patients with chronic ulcers often develop
limited mobility at the ankle joint, this process further compromises function of the calf muscle
pump mechanism.
CVI is a complex condition whereby venous hypertension causes inflammation, thick-
ening and fibrin deposition within the skin and subcutaneous tissues. Clinical features of
CVI include swelling, ulceration, pain, skin changes (sclerosis, pigmentation, eczema) –
Figure I.8. Hypothesis for the formation of CVI skin changes include the white cell trap-
ping, perivascular fibrin cuff formation, tissue pressure effect, and macromolecule leakage
theories.
The case
Venous ulceration classically occurs in the gaiter area, usually in the medial location.
Ulceration occurring primarily in the foot is unlikely to have a venous cause. The venous
ulcer is usually shallow with gentle sloping edges. The surrounding skin will often have skin
changes of chronic venous insufficiency. Although the ulcer base may contain some slough,
once removed there should be pink epithelium where healing is taking place. There may be
areas of white fibrous tissue scarring in the ulcer from the healing process. Where there have
been previous venous ulcers there may be pale scars visible (‘atrophie blanche’). The absence
of visible varicosities does not exclude a venous cause to an ulcer.
Introduction 29
Questions
What are the secondary causes of varicose veins?
• Venous obstruction.
• Extrinsic compression – pelvic mass (including pregnancy), increased abdominal
pressure, and retroperitoneal fibrosis.
• Intrinsic obstruction – post deep vein thrombosis (DVT), May–Thurner syndrome
(DVT as a consequence of compression of the left common iliac vein by the overlying
right common iliac artery).
• Valve destruction – post phlebitic limb.
• High flow – arteriovenous fistula.
How does the calf muscle pump function?

The calf muscle pump is an integral part of the process for re-circulating blood from peripheral
veins back into the centre venous system. The soleal muscles contain large valveless venous
sinusoids with a total capacity of over 100 ml in the adult. At rest the sinusoids fill, and during
muscular activity blood is expelled from these veins. Valves in the deep and perforating veins
direct blood from the superficial system, and then through the deep system, in a retrograde
direction. In the standing subject at rest there is a standing column of blood that exerts a
hydrostatic pressure down to the foot level of around 90 mmHg (equivalent to its vertical
weight from the point of measurement to the right auricle of the heart). Following repeated
calf muscle contraction the hydrostatic venous pressure in the foot falls until it reaches a lower
plateau level of 20–30 mmHg (ambulatory venous pressure). If the patient stands still the
pressure returns back to a resting pressure of around 90 mmHg. Where deep venous obstruc-
tion or incompetence occurs (following a DVT) the pressure produced by the calf muscle
pump forces blood into the superficial system causing superficial venous hypertension.
What are the physiological mechanisms of venous hypertension?

• Superficial venous reflux.
• Deep venous reflux.
• Perforator incompetence.
• Deep venous obstruction.
• Calf pump failure.
Any of the above may exist in isolation or combination.
How does venous claudication differ from arterial claudication?

Exercise in the presence of venous insufficiency causes an increase in blood flow to the limb with a
compromise to normal venous return. The process can cause distension and a ‘bursting’ type dis-
comfort to the leg. Unlike the rapidly resolving pain of intermittent claudication (IC), relief in venous
claudication takes much longer to occur, and often requires sustained elevation to the limb.
How do you manage venous ulcers?

• Debridement – slough and debris should be regularly cleaned from the surface of an
ulcer. Soaking the limb in a bowl of tap water often helps local dressing removal and
the debridement of slough.
• Emollients – surrounding areas of dry skin should be kept moist with emollients.
(Topical steroids are occasionally used to treat surrounding venous eczema; if there are
persistent problems with eczema always consider an allergy to the dressings.)
• Dressing – ulcers can be dressed with non-adherent, hydrocolloid and foam dressings.
The benefits of each depend on the current state of the wound. An important factor in
dressing performance is that it does not stick to the wound, hence removing valuable
granulation tissue on dressing change.
• Graduated compression – the mainstay in managing venous ulcers is with graduated
compression therapy. It is useful to try to minimise pre-existing swelling of the limb
with rest and elevation prior to commencing therapy. Compression therapy should
produce maximum pressure at the ankle, gradually reducing towards the knee level.
Compression bandaging is multilayer, with a variety of layers for dressing support,
padding, elastic compression and an outer cover. The layers vary between compression
systems.
• Antibiotics – antibiotics are only prescribed for ulcers where there is evidence of active
infection (i.e. progressing cellulitis) rather than just colonisation. Topical antibiotics are
generally not given to treat infections, although metronidazole gel is sometimes used
for malodorous wounds.
What is the role of superficial venous surgery in managing venous ulcers?

The recent long term results of compression therapy alone versus compression plus sur-
gery in chronic venous ulceration (ESCHAR) trial has addressed the question of correc-
tion of superficial venous reflux in addition to standard compression therapy [2]. The study
assessed 500 legs managed in specialist nurse-led leg ulcer clinics (three centres). The inves-
tigators demonstrated that surgical correction of superficial venous reflux in addition to
compression bandaging did not improve ulcer healing. However, superficial venous surgery
did reduce ulcer recurrence at 4 years and also resulted in patients having a greater propor-
tion of ulcer-free time.
What factors influence sub-bandage pressure?

Sub-bandage pressure is proportional to TN/CW. Where T is the bandage tension, C is the
circumference of the limb, W is the width of the bandage and N is the number of layers.
From analysing the formula it is clear that a bandage applied with a constant ten-
sion will automatically produce a graduated fall in sub-bandage pressure from the narrow
ankle to the wider calf. The ankle, being the lowest diameter, will have the highest area
of pressure, providing the bandage tension remains the same throughout the dressing.
Hence there is an inverse relationship between leg circumference and pressure applied.
Bandages are generally applied with a 50% overlap. Care must be taken when applying
bandaging around bony prominences (malleoli and tibial crest) not to form areas of local-
ised pressure. Ankle sub-bandage pressures of 40 mmHg are often utilised in treating ven-
ous ulcers.
How is compression hosiery classified?

Compression hosiery has commonly been prescribed by the ‘Class’ system. This system
often leads to confusion due to the differing British standard (BS 6612:1985) and European
standard (SS-ENV 12718) categories. Whereas the British system is divided into three
Introduction 31
Table I.1 Classification of compression hosiery
Class British (mmHg) European (mmHg)

I 14–17 18–21
II 18–24 25–32
III 25–35 36–46
IV NA Over 59
classes the European has four classes according to the level of ankle compression. The
pressure levels in the European ‘classes’ are higher than those in the British systems (see
Table I.1). In clinical practice the best way to prescribe compression hosiery is by the level
of compression in mmHg, hence 18–24 mmHg compression hosiery should be requested
rather than simply ‘class II’.
Aortic aneurysm suitability for endovascular aneurysm

repair (EVAR)
The basics
EVAR is now established based upon the UK EVAR trials but techniques, suitability and
assessment methods are continually evolving. Assessment of suitability requires a com-
puted tomography (CT) angiogram with a minimum slice thickness of 3 mm. Axial images
produced by CT should be assessed to gain a general picture of the aneurysm, patency of
visceral arteries and anatomy of the thoracic aorta and arch. Multi-planar reconstructions
are then performed on CT work stations to allow measurements of diameters perpendicu-
lar to the axis of the aorta and length measurements. Diameters that must be measured
are the neck of the aneurysm, the distal landing zone (usually common iliac arteries) and
the aortic bifurcation. Important length measurements are length of aortic neck, distal
sealing zone, distance from lowest renal artery to aortic bifurcation and distance from
aortic bifurcation to intended landing zone (iliac bifurcation). Three-dimensional volume
rendering modes using automated techniques to plot the centre-line give the described
lengths.
The case
You are shown an abdominal CT scan with an AAA. These will be axial scans but may con-
tain reconstructed images.
Questions
What CT features of an AAA would you consider in assessing it for a conventional stent graft?
Adverse anatomical factors vary for different types of stent but in general these can be clas-
sified as shown below. The limitations for EVAR are becoming less. In the EVAR trials 50%
of AAAs were suitable for stenting whereas currently using a variety of different devices,
approximately 80% can be treated using EVAR. Although no single anatomical problem
should prevent one from performing EVAR, several anatomical constraints may prevent
sealing and fixation, thereby increasing the risk of endoleak and migration.
Graft introduction
Factors making insertion of the stent difficult include iliac artery tortuosity especially in
combination with circumferential calcification preventing straightening of the vessels.
Stenotic or occlusive disease of the iliac arteries may prevent access although minor degrees
can be overcome by angioplasty, use of ‘peel-back’ sheaths or even dilating within a sheath
for more severe cases. Ultimately a uni-iliac device may be used if one-sided access is impos-
sible with a femoro-femoral cross-over graft used to perfuse the limb with the diseased iliac
arteries.
Proximal sealing and fixation

A non-favourable neck may have angulation of more than 60° (although newer devices may
allow greater degrees of angulation), conical neck (more than 3 mm increase in diameter
distally for 1 cm length of neck for sealing), thrombus or calcification. The length of the
neck should be at least 10–15 mm for an adequate proximal seal. Stent grafts come in various
diameters and there will be a few AAAs that have a neck diameter that is too large (normally
32 mm diameter of neck, which requires 20% oversizing).
Distal sealing zone

Similar principles apply to the distal sealing zone as they do to the neck. Ideally this should
be close to the iliac bifurcation with a sealing length of 10–15 mm. With modern devices
iliac diameters as large as 25 mm can be treated. If the landing zone involves the external
iliac artery and coverage of the internal iliac artery, there may be a risk of buttock claudica-
tion or ischaemic colitis if done bilaterally. Coverage would in addition require embolisation
of the internal iliac artery to prevent a Type II endoleak. The internal iliac artery may be
preserved by using customised ‘iliac-branched grafts’.
Length from renal arteries to aortic bifurcation

The length of the aneurysm from the renal arteries has to be appropriate to allow the short
stump of the contralateral limb to fully open. Similarly any ‘waisting’ at the aortic bifurca-
tion may not allow this to be possible, hence the importance of measuring the diameter of
the aortic bifurcation.
How do you classify endoleaks and how are they managed?

Endoleaks are classified as four different types. Types I, III and IV endoleak require inter-
vention whereas most Type II leaks are benign and can be monitored conservatively unless
there is continued sac expansion. Table I.2 summarises classification, pathogenesis and
management.
Consent for an AAA repair

The basics
The discussion about consent will depend on whether the patient is to undergo open or
enodovascular repair and also the ability to obtain consent. The General Medical Council
(GMC) has published guidelines for good practice in obtaining consent.
Introduction 33
Table I.2 Classification of endoleaks
Type Pathogenesis Management

I (proximal) Poor seal or fixation at Balloon moulding (at first procedure)
neck Palmaz stent
Banding of aortic neck
I (distal) Poor seal or fixation at Balloon moulding (at first procedure)
distal landing zone Upsizing distal limb with extension
II Feeding IMA or iliolumbar Conservative treatment if AAA not expanding. If it
vessels expands, then CT or selective catheter angiography
to diagnose the cause. These may be embolised, IMA
may be ligated surgically (open or laparoscopic) or
translumbar injection of thrombotic material directly
into AAA
III Modular disconnection or Intraoperatively junctional zones can be ballooned.
tears in stent graft Postoperatively, bridging stent required
IV Porosity of stent graft or Intraoperative porosity or ‘sweating’ is benign.
undetected endoleak Endotension resulting in continued sac growth
requires relining with new stent graft or explantation
Notes: IMA, inferior mesenteric artery; AAA, abdominal aortic aneurysm; CT, computed tomography.
The case
There will usually be a patient who most likely is about to or has undergone the procedure.
Communication is the key. You must assess how much the patient knows about the proce-
dure and then go on to describe the procedure, its benefits and risks. The risks as stated on
the consent form should be those that are frequent or serious and in addition should take
into account the patient’s personal circumstances. These can be divided into general and
specific and furthermore early and late.
The benefits of surgery depend on the presentation. For asymptomatic AAAs, the reason
for intervention is to prevent rupture. In cases of distal embolisation the aim is limb salvage
and in acute cases of rupture or bleeding from fistulation, the aim is to save life. Rarely, large
aneurysms may cause obstruction of the duodenum.
The risks of open AAA repair

Early complications
1. These are related to aortic cross clamping and the systemic inflammatory response.
Although cardiac complications were the main causes of early mortality, more
modern series suggest that multi-organ failure (MOF) is an equal cause of death
in elective repair including both primary MOF and MOF secondary to visceral
ischaemia and pneumonia.
2. Isolated respiratory failure is the other prominent cause of mortality. With improvements
in optimization of cardiac function, patient selection, advancements in anaesthetic and
postoperative critical care, early mortality rates have been reduced to 5%.
3. Other complications to mention include distal embolisation leading to limb loss,

ischaemic colitis requiring a colostomy, renal failure and impotence. The frequency
of these complications will depend on the complexity of the aneurysm including
suprarenal clamping, thrombus adjacent to the renal artery origins and patency of
visceral and internal iliac arteries.
Late complications
1. Poor coverage of the graft by either the sac of the aneurysm or by an omental flap
may lead to graft-to enteric fistulae and graft infection.
2. False aneuryms particularly with anastamoses onto the femoral artery.
The risks of EVAR

Early complications
1. General complications are related to vascular surgical (groin seroma, haematoma
and false aneurysm) and radiological procedures involving contrast agents (direct
effects of contrast related to volume including renal failure and idiosynchratic
reactions).
2. Specific complications are classified as those relating to the stent and those relating
to the surgery. Stent related complications include endoleak and rupture. Coverage
of both internal iliac arteries in the presence of diseased visceral vessels may lead
to ischaemic colitis. Coverage of one internal iliac artery may lead to buttock
claudication. Inadvertent coverage of renal arteries may necessitate stenting the
renal artery to ‘open’ the ostium covered by the stent graft. Bilateral occlusion
normally necessitates open conversion with renal revascularisation. Balloon
moulding has a small risk of rupture in addition, which may be treated by a covered
stent graft or open conversion. The 30-day mortality for EVAR is 2%.
Late complications
1. Continued expansion and rupture (1% annual risk) can occur because of either
endoleak or endotension alone or in combination with stent migration.
2. Limbs of the stents may also occlude from kinking as a result of changes in
conformation of the stent due to sac shrinkage.
3. Although the risk of graft infections is lower than open repair, procedures utilising
uni-iliac stents, which need a femoro-femoral cross-over graft, may have a higher
risk.
References
1. Levieri LS, Veller MG. Popliteal enfragment synchrome: more common than previously
recognized. J Vcox Surg 1999; 30: 587–98.
2. Gohel et al. Long term results of compression therapy alone versus compression therapy alone
versus compression plus surgery m chronic venous ulceration (eschar): randamized controlled
trial. BMG 2007; 335: 83–87.
Section Final FRCS vascular topics
2
Section 2 Final FRCS vascular topics
Chapter
1
Vascular risk factors and their
management
Alasdair Wilson and Julie Brittenden
Key points
• Peripheral arterial disease (PAD) is an under-diagnosed and under-treated condition
• Patients with PAD have a cardiovascular risk profile equivalent to or worse than those
with coronary or cerebrovascular disease
• PAD patients with concomitant symptomatic cardiac or cerebrovascular disease,
diabetes or a low ankle pressure index are at even higher risk of sustaining a vascular
event
• Patients with PAD should receive the same risk factor management as patients with
other cardiovascular diseases
• Patient awareness of the need for cardiovascular secondary prevention therapy in PAD
is low
The need for cardiovascular risk factor management in patients

with peripheral arterial disease
PAD is a condition that is frequently under diagnosed and often the subject of suboptimal
care. The first line treatment for patients with PAD is cardiovascular risk factor management
with the aim of improving patient survival. This is because patients with PAD have a two- to
threefold increased risk of cardiovascular mortality compared to an age- and sex-matched
control population. The risk of a patient with PAD dying from a heart attack is believed to be
equivalent to those patients who have already survived their first myocardial infarction.
The global Reduction of Atherothrombosis for Continued Health (REACH) registry has
recently been established to determine atherothrombotic risk in more than 68 000 at-risk
patients [1]. To date it has shown that, compared to patients with coronary heart disease or
cerebrovascular disease, those with PAD had the highest rates of cardiovascular death, myo-
cardial infarction, stroke, or hospitalisation for atherothrombotic events at 1-year follow up
(Figure 1.1). It also showed that the number of events increased with the number of clinic-
ally involved vascular beds. Thus patients with PAD and symptomatic cardiac or cerebrovas-
cular disease have increased risk compared to those with PAD alone (Figure 1.1).
The risk of developing cardiovascular events in patients with PAD is also known to
increase with the severity of disease, such that patients with rest pain or tissue loss have
a worse prognosis than patients with intermittent claudication (Figure 1.2) [2]. However,
38 Section 2: Final FRCS vascular topics
Population (%) Figure 1.1 One-year cardiovas-

cular event rates in outpatients
30
with atherothrombosis.
Data for graph obtained
25
from: Steg PH, Bhatt D, Wilson
20
PWF, D’Iagostino R, Ohman
EM, Rother J et al. JAMA 2007;
15 297: 1197–206. Reproduced
with permission. PAD,
10 peripheral arterial disease; CHD,
coronary heart disease; CVD,
5 cerebrovascular disease.
0
PAD CHD CVD PAD+ PAD + PAD
CHD CVD +CHD+CVD
100 Figure 1.2 Kaplan–Meier sur-

vival curves based on mortality
Normal subjects from all causes in patients with
Survival (% of patients)
75 large vessel disease.

Criqui MH, Langer RD, Fronek
A, Feigelson HF, Klauber MR,
50 Asymptomatic PAD† McCann TJ et al. Mortality over
Symptomatic PAD† a period of 10 years in patients
with peripheral arterial disease.
25 Severe symptomatic PAD† N Engl J Med 1992; 326: 381–6.
Reproduced with permission.
PAD, peripheral arterial disease.
0
0 2 4 6 8 10 12
Year
patients who are asymptomatic but have PAD as defined by an ankle brachial pressure index
(ABPI) less than 0.9 have also been shown to have a reduced survival compared to a sex- and
age-matched control population. In one study, patients who had PAD involving the large
vessels were found to have a 6.6-fold (95% confidence interval 2.9–14.9) increased risk of
death from coronary heart disease at 10-year follow up compared to patients with no PAD.
Overall, less than one-quarter of patients with severe symptomatic large vessel PAD survived
10 years (Figure 1.2) [2].
The ABPI has also been shown to predict overall survival, independently of the meta-
bolic syndrome and other conventional cardiovascular risk factors. The hazard ratio for
mortality has been shown to increase consistently with decreasing ABPI for both males and
females (Figure 1.3)[3].
Why do we as vascular surgeons need to treat risk factors?

Despite the increased cardiovascular risk in patients with PAD, risk factor management in
these patients has been shown to be inadequate in both primary and secondary care set-
tings. In particular, when compared to patients with coronary heart disease (CHD), patients
with PAD (despite comparable risk) received less intensive treatment for lipid disorders and
hypertension and were prescribed antiplatelet therapy less frequently than were patients
with CHD.
Chapter 1: Vascular risk factors 39
8.0 Men
Wome
4.0
Hazard ratio (95%)
2.0
1.0
0.5
≤ 0.60 0.61–0.70 0.71–0.80 0.81–0.90 0.91–1.00 1.01–1.10 1.11–1.20 1.21–1.30 1.31–1.30 >1.40
(Reference)
Ankle brachial index
Figure 1.3 Hazard ratios for total mortality in men and women by ankle brachial pressure index (ABI) for all stud-
ies in the ABI collaboration. Ankle Brachial Index Collaboration. Ankle Brachial Index combined with Framingham
risk score to predict cardiovascular events and mortality. A meta-analysis. JAMA 2008; 300: 197–208. Reproduced
with permission.
In addition to being under-treated, PAD is often under-diagnosed and public awareness

of the condition has been shown to be poor throughout Europe and the USA. Patients are
also unaware of the need for secondary prevention therapy. Thus it is the responsibility of
the vascular surgeon to initiate risk factor management if this has not already been com-
menced in primary care and also to educate the patient on the need for this treatment.
Risk factors
The major risk factors for PAD are the same as for CHD. They can be considered as
either ‘modifiable’ or ‘non-modifiable’ and those that are modifiable may be treated by
‘lifestyle changes’ or drug treatments. Of all the modifiable risk factors smoking is the
most important but others include management of dyslipidaemia, hypertension, diabe-
tes, and the use of antiplatelet therapy. Evidence-based medicine has shown that reducing
cardiovascular risk in patients with symptomatic PAD improves survival. The rationale
and targets for treating these risks factors have been addressed in a number of national
and international guidelines. The Scottish Intercollegiate Guidelines Network (SIGN)
guidelines [4] on the management of peripheral arterial disease and the Transatlantic
Society (TASC) guidelines have both been recently updated [5]. All vascular surgeons
should check that their PAD patients have stopped smoking, check that they are on an
antiplatelet and a statin unless contraindicated, check that their hypertension is ade-
quately treated, exclude diabetes and give lifestyle advice. The management of these risk
factors is further summarised below.
Smoking
Smoking cessation is the most important component of secondary prevention in patients
with PAD. Cigarette smoking doubles the risk of a patient developing PAD and for those who
Favours Favours Figure 1.4 Pharmacotherapy

Pharmacotherapy Odds ratio (95% Crl) placebo treatment for smoking cessation: a
meta-analysis of randomized
Bupropion 2.12 (1.76–2.56)
controlled trials.
Nicotine gum 1.65 (1.37–2.01) Permission for use requested.
Eisenberg MJ, Filion KB, Belisle P
Nicotine inhaler 2.18 (1.38–3.45)
et al. Pharmacotherapies
Nicotine nasal spray 2.37 (1.57–3.60) for smoking cessation: a
meta-analysis of randomized
Nicotine patch 1.88 (1.60–2.22) controlled trials CMAJ 2008;
Nicotine tablet 2.06 (1.47–2.87) 179: 135–44. Reproduced with
permission.
Varenicline 2.55 (1.99–3.24)
0.2 1.0 7.0

Odds ration (95% Crl)
continue to smoke the chance of disease progression is also increased by twofold. Cessation
of smoking may not improving walking capacity in patients with PAD but can prevent the
onset of critical limb ischemia and reduces the risk of bypass graft occlusion by threefold.
However, most importantly, smoking cessation is associated with a reduction in all major
cardiovascular events. A recent Cochrane review of smoking cessation for the secondary
prevention of coronary heart disease has shown that smoking cessation is associated with a
36% reduction in all- cause mortality [6].
The vascular surgeon has an important role in promoting smoking cessation and should
‘strongly and repeatedly advise patients to stop smoking’ (Figure 1.4) [5, 7], and arrange
referral to a smoking cessation program. Encouraging patients to stop smoking through
smoking cessation support programs has been shown to double the smoking cessation rate.
Nicotine replacement therapy also increases the quit rate by approximately twofold. The
combination of these two methods has been associated with a 22% cessation rate at 5 years.
Antidepressants such as bupropion have also been found to be useful in achieving smoking
cessation and have a synergistic effect when used with nicotine replacement therapy. More
recently interest has focused on the role of the nicotine receptor partial agonist varenicline
in smoking cessation. This may reduce dependence by mimicking the actions of nicotine
on neuronal nicotinic receptors in the brain, thus maintaining some dopamine levels and
reducing withdrawal symptoms that are associated with reduced dopamine release. In add-
ition to this action, varenicline may also competitively inhibit binding of cigarette nicotine
to these receptors. A recent randomised controlled trial has shown that varenicline results
in four times greater odds of stopping smoking compared to a placebo and two times greater
odds than bupropion. Although effective, these drugs may have side effects (bupropion can
cause seizures, varenicline can cause depression and suicidal ideation). Pharmacotherapy
treatments are usually provided within a smoking cessation program that involves counsel-
ling and usually uses nicotine replacement as first-line treatment (Figure 1.5).
Smoking: current recommendations
• Patients with PAD who smoke should be advised to quit.
• Vascular surgeons should take the opportunity to advise all patients who smoke with
PAD to quit when they attend for a consultation.
• Patients with PAD who smoke should be referred to an intensive support service.
• Patients with PAD who are planning to stop smoking should be offered nicotine
replacement therapy (NRT) or varenicline or bupropion.
Intervention Effect size (OR) Figure 1.5 Effectiveness of

smoking cessation interventions.
Eisenberg MJ, Filion KB, Belisle P
Group behaviour therapy
et al. Effectiveness of smoking
cessation interventions among
Bupropion
adults: a systemic review of
reviews. Cancer Prevention 2008;
Physician advice 17: 535–44. Reproduced with
permission.
NRT
Individual counselling
Telephone counselling
Nursing interventions
Self-help interventions
–1 0 1 2 3
Dyslipidemia
Epidemiological, post mortem and angiographic studies have consistently shown a strong
positive correlation between plasma total cholesterol and the incidence of CHD. In Western
populations the increased risk of future cardiovascular events is observed with serum chol-
esterol levels that were considered to be in the normal range and increase progressively in a
linear manner with rising cholesterol concentrations. In patients with PAD, elevated levels
of total cholesterol, low density lipoprotein cholesterol, triglycerides and lipoprotein A are
independent risk factors for adverse vascular events. In contrast, increased levels of high-
density lipoprotein are protective.
Statins competitively inhibit the enzyme 3-hydroxyl-3-methylglutaryl coenzyme A
(HMG-CoA), which catalyzes the conversion of HMG-CoA to mevalonate, an early step
in the biosynthesis of cholesterol. This leads to a reduction in hepatocyte cholesterol con-
centration and increased expression of low-density lipoprotein (LDL) receptors, which are
involved in the clearance of LDLs and LDL precursors from the circulation. Statin therapy
to lower LDL cholesterol is recommended in all patients with PAD, even in those patients
whose cholesterol levels are within what is considered the normal range. This guidance is
based on the results of the Medical Research Council (MRC) Heart Protection study, which
showed that treatment with simvastatin 40 mg daily resulted in a 22% (95% confidence
interval: 15–29) relative risk reduction in the rates of myocardial infarction, stroke and of
revascularisation in patients with PAD who had a cholesterol level greater than 3.5 mmol l–1
[8]. There was also a significant reduction in all-cause mortality and in particular that due
to cardiac causes in patients with PAD allocated to simvastatin therapy compared to placebo
(Figure 1.6).
These benefits were observed irrespective of the baseline level of cholesterol and occurred
in patients with and without clinical disease in other arterial beds. In addition to starting
a patient with PAD on a statin it is also important to monitor the reduction in cholesterol
levels achieved. The reduction in cardiovascular risk achieved by statin therapy has been
shown to be proportional to the achieved reduction in LDL cholesterol reduction (Figure
1.7) [9]. Overall statins reduce the 5-year incidence of major coronary events, coronary
Major vascular event Simvastatin Placebo Event rate ratio Heterogeneity

& prior disease group -allocated -allocated (95% Cl) p-value
(10,269) (10,267)
Major coronary evebts

PAD 369 (10.9%) 465 (13.8%) p = 0.3
No PAD 529 (7.7%) 747 (10.8%)
0.73 (0.67–0.79)
Subtotal: coronary 898 (8.7%) 1212 (11.8%) p < 0.0001
Strokes
PAD 179 (5.3%) 242 (7.2%) p = 0.7
No PAD 265 (3.8%) 343 (5.0%)
0.75 (0.66–0.85)
Subtotal: stroke 444 (4.3%) 585 (5.7%) p < 0.0001
Revascularisations
PAD 466 (13.8%) 603 (17.9%) p = 0.7
No PAD 473 (6.9%) 602 (8.7%)
0.76 (0.70–0.83)
Subtotal: revascularisation 939 (9.1%) 1205 (11.7%) p < 0.0001
MAJOR VASCULAR EVENTS

PAD 895 (26.4%) 1101 (32.7%) p = 0.5
No PAD 1138 (16.5%) 1484 (21.5%)
ALL PATIENTS 2033 (19.8%) 2585 (25.2%) 0.76 (0.72–0.81)

p < 0.0001
0.4 0.6 0.8 1.0 1.2 1.4

Simvastatin better Placebo better
Figure 1.6 Heart Protection Study. Randomized trial of the effects of cholesterol-lowering with simvastatin on
peripheral vascular and other major vascular outcomes in 20 536 people with peripheral arterial disease and other
high-risk conditions. Heart Protection Study Collaborative Group. J Vasc Surg 2007; 45: 645–54. PAD, peripheral
arterial disease.
revascularisation and stroke by one-fifth per mmol l-1 reduction in LDL-cholesterol, irre-
spective of the patient’s baseline lipid profile. Furthermore, recent trials involving patients
with acute coronary syndromes and stable coronary disease have shown a greater reduc-
tion in cardiac events in patients receiving high-dose compared to conventional dose sta-
tin therapy [8] This is likely to be the case for patients with PAD and as such the TASC
guidelines recommend more aggressive lowering of cholesterol in those patients with PAD
most at risk such as those with concurrent disease in other vascular beds. This should also
be considered for those with diabetes or a low ABPI. In addition to statin therapy, diet-
ary measures are also recommended but when used alone have been shown to result in
only a 10% reduction in LDL-cholesterol and long-term compliance is known to be low.
The main limitations to using statins are side effects, principally muscle aches and rarely
rhabdomyolysis.
For blood sampling and other lipids, non-fasting samples are satisfactory for assessing
LDL-cholesterol levels. Triglycerides would need to be measured in fasting samples but are
not routinely directly targeted in patients with PAD. High-density lipid (HDL)-cholesterol
levels are known to be protective and a number of new pharmacological agents are currently
under investigation to increase HDL-cholesterol.
30 4S
Statin
Placebo
25
4S
20
Events (%)
LIPID
15
LIPID
CARE
CARE
10 HPS
HPS
TNT (10 mg of atorvastatin)
5 TNT (80 mg of atorvastatin)
0
0 70 90 110 130 150 170 190 210
LDL cholesterol (mg/dl)
Figure 1.7 Relationship between level of cholesterol reduction achieved with statin therapy and reduction in
events. La Rosa JC, Grundy SM, Waters DD. Intensive lipid lowering with atorvastatin in patients with stable coron-
ary disease. N Engl J Med 2005; 352: 1425–1435. Reproduced with permission.
Note: 100 mg dl–1=2.6 mmol.
Dyslipidaemia: current recommendations
• Patients with PAD with a total cholesterol >3.5 mmol l–1 should be commenced on
statin therapy (this will be the vast majority of patients and you do not need to wait for
lipid levels to come back before starting therapy).
• The aim is to lower the LDL cholesterol to less than 2.0 mmol l–1.
• Patients with PAD and known disease in other vascular beds should be considered
for more aggressive treatment with the aim of achieving LDL-cholesterol levels below
1.8 mmol l–1.
Statins and inflammatory response

This is still a controversial area but statins appear to do more than just lower cholesterol.
LDL-cholesterol is oxidized by free radicals and may cause direct oxidative damage. A reduc-
tion in the levels of LDL cholesterol by the use of statin therapy will result in reduced super-
oxide production. While many of the anti-inflammatory effects may be due to the reduction
in LDL-cholesterol, the reduction in vascular events achieved by statin therapy is greater and
occurs earlier than would be predicted from the lipid lowering effects alone. The non-lipid
lowering effects of statins have been attributed to their ability to inhibit the generation of
proteins called isoprenoids. Mevalonate is the precursor of these compounds and inhibited
by statins. The isoprenoids bind to a number of signalling proteins (Rho and Ras) on the cell
membrane, which are involved in the inflammatory response. Through these mechanisms
statins have widespread effects on the endothelium, coagulation pathways, and platelet func-
tion, all of which are implicated in the pathogenesis of acute ischaemic events and have been
shown to be activated in patients with PAD.
Vascular surgery has been shown to result in endothelial activation and a pro-thrombotic
state, as indeed has lower limb angioplasty. Statin therapy has been shown to decrease the
incidence of perioperative cardiac events in patients undergoing vascular surgery. Statin
therapy leads to plaque stabilisation and may also have beneficial effects on walking dis-
tances. One study has shown that patients with PAD on statin therapy have a lower annual
decline in lower extremity performance compared to those who are not on statins. A further
study has suggested that statins may increase walking distances in claudicants.
Hypertension
Elevated blood pressure is an independent risk factor for cardiovascular and cerebrovascu-
lar morbidity and mortality (SIGN) [4]. In the Framingham trial the age-adjusted risk ratio
for intermittent claudication in hypertensive patients compared to controls was increased
two and half to fourfold [3]. In the UK, treatment of hypertension in patients with PAD
should follow the recommendations of the British Hypertension Society Guidelines [10].
The current targets are a level of less than 140/90 mmHg or 130/80 mmHg for patients with
diabetes or chronic renal disease. It can be seen from Figure 1.8 that the recommended
first-line drug treatment varies depending on the age and race of the patient. ACE inhibi-
tors are first-line drugs for the treatment of hypertension in patients with PAD but should
be commenced with careful monitoring due to the possibility of co-existing renal artery
stenosis. The prevalence of this is difficult to ascertain but one study, involving a selective
group of patients with PAD who were undergoing angiography, found renal artery stenosis
Choosing drugs for patients newly diagnosed with hypertension
Abbreviations: Younger than 55 years or older

A = ACE inhibitor 55 years or black patients of any age
(consider angiotensin-II receptor
antagonist if ACE intolerant)
C = calcium-channel blocker A C or D Step 1
D = thiazide-type diuretic
A + C or A + D Step 2
Black patients are those of African or

Caribbean descent, and not mixed- A+C+D Step 3
race, Asian or Chinese patients
Add
further diuretic therapy
or
alpha-blocker Step 4
or
beta-blocker
Consider seeking
specialist advice
Figure 1.8 Current British Hypertensive Society and National Institute for Health and Clinical Excellence (NICE)
guidelines for treatment of newly diagnosed hypertension.
in a quarter of patients. Since the publication of the Heart Outcomes Prevention Evaluation
(HOPE) study there has been a move towards recommending the use of ACE inhibitors
in patients with PAD even in the absence of hypertension [11]. ACE inhibitors have been
shown to have various pleiotropic effects beyond their blood pressure lowering capacity.
While there appears to be a promising reduction in cardiovascular mortality, morbidity,
and stroke associated with ACE inhibitors, the TASC and SIGN guidelines do not recom-
mend their use in patients with PAD other than for their blood pressure lowering effects.
Only 30% of patients in the HOPE study were on statin therapy, and thus the pleiotropic
effects of ACE inhibitors, in addition to standard medical therapy in patients with PAD
who do not have hypertension, requires to be evaluated further. However, some recent
evidence suggests that the ACE inhibitor ramipril may improve pain-free and maximum
walking time in patients with PAD.
Beta-adrenergic blocking agents (β-blockers), in particular those with vasoconstrictor
properties, were previously not recommended in patients with PAD. However, a meta-anal-
ysis of 11 randomised controlled trials has shown that they do not worsen the symptoms of
claudication and can be used safely in patients with PAD. Furthermore, β-blockers may offer
additional cardio-protection to patients with symptomatic PAD and cardiovascular disease.
A number of studies have shown that perioperative use of β-blockers may reduce the post-
operative cardiovascular events in PAD patients undergoing major vascular surgery. Their
use is therefore recommended in the TASC guidelines. However, this is still a controversial
area and a more recent study (POISE study) has shown that they may actually increase mor-
tality [12]. Current advice would be to continue β-blockers if the patient is already on them
but not to use them first line for blood pressure control or general risk reduction.
Hypertension: current recommendations
• Patients with PAD and hypertension should be treated to reduce their blood pressure to
<140/90 mmHg.
• Patients with PAD, hypertension and either diabetes or renal impairment should be
treated to reduce their blood pressure to <130/80 mmHg.
• Beta-adrenergic blocking drugs are not contraindicated in patients with PAD.
Antiplatelet therapy
Antiplatelet therapy is recommended in all patients with PAD. The Antithrombotic Trialists’
Collaboration found that antiplatelet therapy (aspirin, ticlopidine or dipyridamole) was
associated with a 23% reduction in non-fatal myocardial infarction, non-fatal stroke and
vascular death in patients with PAD. Low dose aspirin (75–150 mg) is recommended as it
has been shown to be equally as effective as high doses and is associated with a lower rate of
gastrointestinal side effects [9].
Clopidogrel, a thienopyridine, has been shown to be of benefit in the treatment of
PAD. The use of clopidogrel versus aspirin in patients at high risk of ischaemic events trial
(CAPRIE) showed that clopidogrel reduced the relative risk of major vascular events by
8.7% ( 95% confidence interval 0.3–16.5%) compared to aspirin. In a subgroup analysis,
clopidogrel reduced the relative risk of major vascular events by 23.8% (95% confidence
interval 8.9–36.2%) compared to aspirin in patients with PAD. A subsequent economic
analysis by the National Institute of Health and Clinical Excellence (NICE) has shown that
the use of clopidogrel as first-line antiplatelet therapy is cost-effective for 2 years but not
beyond this time period [13]. It remains to be determined if clopidogrel is cost effective in
the ‘higher risk’ PAD groups. If used, clopidogrel should be prescribed as monotherapy. A
combination antiplatelet therapy of aspirin and clopidogrel has not been shown to be of
benefit to patients with PAD, and is not recommended due to the increased risk of bleeding
complications.
Platelet activation, despite the use of antiplatelet therapy, has been shown to be
increased in patients with PAD compared to healthy controls. Studies have shown
‘aspirin resistance’ occurs in 11% to 40% of patients with PAD. Similarly, a large vari-
ation in response to clopidogrel has also been shown to occur in patients with intermit-
tent claudication, with 10% of patients showing no reduction in platelet activation after
a loading dose.
Antiplatelet therapy: current guidelines

• Patients with PAD should be prescribed antiplatelet therapy.
• If aspirin is prescribed it should be used at low dose (75–150 mg).
• If clopidogrel is prescribed it should be used as monotherapy.
• There is currently no evidence for the use of dual antiplatelet therapy in PAD.
Diabetes
Diabetes and its poor control have long been recognised as a major risk factor for periph-
eral arterial disease. Diabetes increases the risk of PAD by two- to threefold. Approximately
20% of patients with PAD will have diabetes, but undiagnosed diabetes is common and may
occur in 12% or more of new patients referred to a vascular clinic. Thus patients attending
vascular clinics should be screened for the presence of possible diabetes. Tight diabetic con-
trol has been shown to reduce the risk of developing microvascular and macrovascular com-
plications. The UK diabetes prospective study showed that each 1% rise in haemoglobin A1c
(HbA1c) was associated with a 28% increased incidence of PAD and a 28% increased risk of
death [14]. Furthermore, each 1% reduction in HbA1cachieved by treatment was found to
correlate with a 14% reduction in myocardial infarction and a 43% decrease in amputation
or death from PAD (Figure 1.9).
The current National Diabetic Guidelines recommend a HbA1c of less than 7% [15].
Thus in vascular clinics the opportunity to measure the HbA1c level should be taken. The
current National Guidelines recommend that patients with a HbA1c level of greater than
6.5% should be started on a medical therapy after a trial of lifestyle measures [15]. All too
often patients have been maintained on ‘lifestyle measures’ alone despite persistently ele-
vated HbA1c levels. Furthermore, the presence of PAD and diabetes means that the patient
falls into a high-risk category and should have aggressive cardiovascular risk factor manage-
ment and appropriate foot care. In patients with type II diabetes and PAD, intensive blood
pressure control has been shown to significantly reduce the risk of cardiovascular events, as
has intensive treatment of dyslipidemia.
Diabetes: current recommendations
• Patients with diabetes and PAD should have aggressive control of blood glucose with
the aim of obtaining a HbA1c level of <7%.
• Patients with PAD should be screened for the presence of diabetes.
80 Figure 1.9 Diabetes, haemoglobin A1c and

Myocardial infarction macrovascular events. Stratton IM, Adler AI, Neil HAW,
Microvascular end points Mathews DR, Manley SE, Cull CA, et al. Association of
Adjusted incidence per 1000 person years (%)
macrovascular and microvascular complications of

type 2 diabetes (UKPDS:35): prospective observational
60 study. BMJ 2000; 321: 405–12. Reproduced with
permission.
40
20
0
5 6 7 8 9 10 11
Updated mean haemoglobin A1c concentration (%)
Lifestyle: weight reduction, diet and exercise

Obesity, in particular a BMI greater than 30 kg m–2 is associated with increased cardiovascular
risk (SIGN). Thus obese patients with PAD should be offered help with weight reduction in
terms of diet and exercise [4]. Patients with intermittent claudication who are physically active
have been shown to have improved survival compared to patients who are sedentary [16].
There is abundant evidence to show that exercise can improve walking distance in clau-
dicants if it is provided as a supervised exercise programme. However, there is little evidence
that simple advice to exercise more is effective on its own. Where they are available patients
with claudication should be referred to a local exercise programme. There are no studies
looking at long-term risk reduction benefits of exercise in PAD specifically, however, they
are proven to be effective in coronary disease and heart failure patients and prevent both
future events and need for hospitalisation.
Weight reduction: current recommendations

• Patients with PAD who are obese should be treated to reduce their weight.
Exercise: current recommendations
• Patients with PAD should be encouraged to exercise and when available referred to a
supervised exercise programme.
Homocysteine
This is currently a controversial risk factor for vascular disease. High plasma levels of the
non-essential sulphur-containing amino acid homocysteine appear to be an independent risk
factor for all types of atherosclerosis including PAD. Vitamins B6, B12 and folate are involved
in its metabolism. Hyperhomocysteinemia is common, present in up to 60% of patients with

PAD, and may be treated by folic acid and vitamin B6 supplements. However, currently there
is no evidence from randomised trials that treatment in patients with PAD will alter the nat-
ural history of the disease and indeed evidence from one study in patients with cardiovas-
cular disease suggests that treatment may have the potential of causing harm. There is still
debate about whether homocysteine itself, or cofactors such as folic acid, are the causative
agents. Thus the TASC guidelines currently recommend that patients with PAD should not
be given folate supplements routinely. Homocysteine should probably only be measured in
patients who develop PAD at a young age in the absence of traditional risk factors.
Homocysteine: current recommendations
• Patients with PAD should not routinely have homocysteine levels measured.
• Patients with PAD should not routinely be given folate supplements.
References review of reviews. Cancer Prevention 2008;

17: 535–44.
1. Steg PH, Bhatt D, Wilson PWF et al. 7. LaRosa JC, Grundy SM, Walters DD et al.
One-year cardiovascular event rates in Intensive lipid lowering with atorvastatin
outpatients with atherothrombosis. JAMA in patients with stable coronary heart
2007; 297: 1197–206. disease. N Engl J Med 2005; 352: 1425–35.
2. Criqui MH, Langer RD, Fronek A et al. 8. Antithrombotic Trialists’ Collaboration.
Mortality over a period of 10 years in Collaborative meta-analysis of randomised
patients with peripheral arterial disease. trials of antiplatelet therapy for the
N Engl J Med 1992; 326: 381–6. prevention of death, myocardial infarction
3. Ankle Brachial Index Collaboration. and stroke in high risk patients. BMJ 2002;
Ankle Brachial Index combined with 324: 71–86.
Framingham Risk Score to predict 9. Heart Protection Study Collaborative
cardiovascular events and mortality. JAMA Group. MRC/BHF Heart Protection Study
2008; 300: 197–208. of cholesterol lowering with Simvastatin I
4. Scottish Intercollegiate Guidelines 20, 536 high risk individuals: a randomised
Network (SIGN). Diagnosis & Management placebo controlled trial. Lancet 2002;
of Peripheral Arterial Disease 2006; 360: 7–22.
Guideline no. 89. 10. Stratton IM, Adler AI, Neil HAW
5. TASC II guidelines at http://www. et al. Association of macrovascular and
tasc-2-pad.org microvascular complications of type
6. Eisenberg MJ, Filion KB, Belisle P et al. 2 diabetes (UKPDS:35): prospective
Effectiveness of smoking cessation observational study. BMJ 2000;
interventions among adults: a systemic 321: 405–12.
Chapter
2
Management of acute limb
ischaemia
Arun Balakrishnan and David Lambert
Key points
• Acute limb ischaemia (ALI) is associated with significant mortality and morbidity
• Clinical assessment is paramount for planning management
• All cases of ALI should be assessed by a vascular specialist
• All cases should be started on intravenous heparin as soon as possible to prevention
extension of thrombus
• Surgery is preferred with severe ALI as time is of the essence
• Thrombolysis is associated with a lower mortality rate but higher failure rates
• Surgery is more durable but is associated with a higher mortality rate
• If compartment syndrome likely or suspected a fasciotomy is required
Definition
Acute limb ischaemia (ALI) can be defined as a sudden compromise of the blood supply to
a limb, threatening its viability. Symptoms are usually of less than 2 weeks in duration. The
lower limbs are more commonly affected than the upper limbs.
Background
Patients with ALI present depending on the severity of their symptoms. In patients with
acute arterial occlusions and no collaterals symptom onset is immediate and severe. This
scenario is seen in patients with embolic occlusions, trauma, thrombosed aneurysms and
occluded grafts. If the acute event occurs with a background of an artery or a graft narrow-
ing/occluding over a period of time then usually there are developed collaterals. In these
patients the symptoms are often not as severe.
After 3–6 hours of severe ischaemia muscle and nerve undergo irreversible changes. Ischaemia
of the limb for greater than 6 hours usually results in functional impairment or limb loss. Time is
therefore of the essence – the less the time interval between the event and treatment the better the
outcome. Acute limb ischaemia is a genuine surgical emergency with a high incidence of mortal-
ity and morbidity. These patients are best managed by a dedicated vascular service [1].
Aetiology
Acute limb ischaemia can be caused by occlusion of a native vessel or a graft. Arteries are
mainly occluded by thrombus or emboli (Table 2.1). Emboli tend to lodge in the bifurcation
Table 2.1 Causes of arterial occlusion
Embolic
• Mural thrombus following myocardial infarction
• The atrium in patients with atrial fibrillation [2]
• The atrium in patients with rheumatic heart disease
• Valvular vegetations in patients with endocarditis
• Atrial myxoma
• Aneurysms and atherosclerotic lesions proximal to the ischaemic limb
• Paradoxical emboli from the venous system in patients with atrial septal defects
Thrombotic
• Thrombosis of an artery due to atheroma
• Thrombosed aneurysm with peripheral embolisation
• Thrombosis of a reconstructed artery or bypass graft
• Arterial dissection
• External compression
• Popliteal entrapment
• Cystic adventitial disease
• Blunt trauma resulting in disruption of the intima
• Penetrating trauma resulting in division of the artery
• Compartment syndrome
• Low flow states in the limb
○ Hypotension
○ Low cardiac output
○ Vasoconstrictor drugs
○ Severe venous thrombosis
of an artery. The usual sites in the limbs are at the bifurcation of the common femoral artery,
iliac artery, popliteal artery, aorta and brachial artery. Almost all emboli are part of a throm-
bus although foreign body or tumour emboli can occur. Emboli lodging at the aortic bifur-
cation are termed saddle emboli.
Assessment of the acutely ischaemic limb

The limb is cool, pale with decreased sensation and muscle weakness. The clinical features of
an acutely ischaemic limb are often described as the ‘6 P’s’:
Pain Sudden onset, constant and severe

Pallor The limb is pale
Pulselessness Unilateral loss of pulses
Paralysis Inability to move limb
Paresthesia Altered sensation
Perishingly cold leg Cool or cold skin
Chapter 2: Acute limb ischaemia 51
Table 2.2 Classification of acute limb ischaemia
Arterial Venous
Capillary Muscle Sensory Doppler Doppler
Category Description return paralysis loss signals signals
I Viable Not Intact None None Audible Audible
immediately
threatened
IIa Threatened Salvageable Intact/slow None Partial Inaudible Audible
if promptly
treated
IIb Threatened Salvageable Slow/absent Partial Partial/ Inaudible Audible
if complete
immediately
treated
III Irreversible Primary Absent Complete Complete Inaudible Inaudible
amputation Staining Tense
compartment
Rutherford RB. Suggested standards for reports dealing with lower extremity ischemia. J Vasc Surg 1986;
4: 80–94. Reproduced with permission.
However, this is a rather simplistic approach best suited to medical students rather than a
vascular specialist. Patients rarely complain of any of these specifically, except pain. Most
commonly they describe the sudden onset of pain, inability to stand or walk on the leg and
then the onset of numbness in the foot.
An acutely ischaemic limb must be carefully assessed to determine the severity of
ischaemia. The main questions to answer are the following:
• Is the ischaemia reversible?
• Is the leg viable?
• Is the limb immediately threatened?
The severity of ischaemia influences management and decision making. A viable limb
has minor or no sensory or motor impairment. The presence of rest pain, decreased sensa-
tion and weak muscles indicate a threatened limb. Limbs with severe pain in the presence
of fixed mottling and tender muscles are irreversibly ischaemic. A viable limb allows time
for investigation to decide on appropriate intervention (Table 2.2). When the cause for limb
ischaemia is thrombosis in situ the symptoms/signs may be less pronounced. This is because
occlusion occurs in an artery/graft that has narrowed over a period of time, permitting col-
laterals to develop.
Management
The severity of ischaemia influences management, urgency and decision making (Figure 2.1).
When a diagnosis of severe ALI is made the patient should be adequately resuscitated. The
following measures in particular should be instituted:
1. Oxygen should be administered.
2. Intravenous (IV) access and fluids to achieve adequate hydration.
ACUTELY Figure 2.1 Flow chart showing

ISCHAEMIC LIMB initial decision-making pathway.
VIABLE LIMB THREATENED IRREVERSIBLY

LIMB ISCHAEMIC
IMAGING IMMEDIATE AMPUTATION

BEFORE INTERVENTION OR
INTERVENTION No Delay for PALLIATIVE
Imaging CARE
3. Heparin therapy: a bolus of 5000 units IV followed by the infusion of heparin at a

rate of 1000 units per hour. The APTR is monitored and heparin infusion adjusted
accordingly to maintain adequate anticoagulation.
4. Adequate analgesia.
5. Urine output should be monitored – usually a catheter is inserted.
Heparin is used to prevent propagation and extension of thrombus. It also reduces the
incidence of cardiovascular events and improves prognosis [3].
Investigation
All patients should have the following baseline tests, although treatment should not be
delayed for the results.
1. Full blood count.
2. Urea and electrolytes.
3. Blood glucose.
4. Clotting if already on anticoagulant therapy.
5. Electrocardiogram (ECG).
6. Group and save serum.
Vascular imaging is only indicated to guide treatment if the limb is viable. If the limb is
threatened then imaging will not usually change the planned procedure and will delay revas-
cularisation. Imaging will define arterial anatomy and help decide if surgery or endovascular
therapy is appropriate. If required on-table angiography may be preferred to avoid delays. If
the limb is not immediately threatened then standard imaging modalities including Duplex
and computed tomography angiography (CTA) or magnetic resonance angiography (MRA)
may be appropriate.
Treatment strategies for acute limb ischaemia

The various available treatment options are:
1. Surgical revascularisation.
2. Surgery in the form of amputation.
3. Endovascular including thrombolysis.
4. Palliative care.
If the limb is immediately threatened then immediate intervention is indicated.
Surgical revascularisation
Embolectomy
Femoral embolectomy can be performed under a local or general anesthetic depending upon
the fitness of the patient. If performed under a local anesthetic then an anaesthetist should be
present in theatre and the patient adequately monitored. The patient is appropriately prepared
and draped. In the absence of a femoral pulse both groins are exposed. The limbs should be
visible throughout the procedure and the feet should be placed in transparent bags. The femo-
ral vessels on the affected side should be accessed through a vertical incision centred on the
mid inguinal point. The common femoral, superficial femoral and profunda femoris arteries
should be dissected free and controlled with slings. A transverse arteriotomy should be fash-
ioned proximal to the bifurcation of the common femoral artery into the superficial femoral
and profunda femoris arteries in order to permit selective catheterisation of both arteries. The
embolus usually lodges in the femoral bifurcation, iliac bifurcation or the aortic bifurcation.
If there is an embolus at the femoral bifurcation this is removed using forceps or suction.
This manoeuvre will establish a good flow of blood down the common femoral artery if there
are no other proximal lesions. A Fogarty catheter is passed down the superficial femoral and
profunda arteries in an attempt to remove any thrombus that may have dropped down these
arteries. An intraoperative angiogram can be performed to check if the arteries have been
adequately cleared. If the embolus is proximal to the femoral artery then a Fogarty catheter will
need to be passed cranially. If flow cannot be established down one side then the patient should
have a femoral to femoral artery bypass if there is a normal pulse on the other side. If this is not
the case then the patient should have an ipsilateral axillary to femoral artery bypass.
Some patients with an ischaemic limb have a good femoral pulse with an embolus at the
popliteal bifurcation. If this diagnosis is suspected then the below-knee popliteal artery should
be exposed through an incision behind the medial border of the tibia. The below-knee pop-
liteal, anterior tibial, peroneal and the posterior tibial arteries should be dissected free and
controlled with sloops. An incision should be made proximal to the bifurcation of the popliteal
artery. Each of the above arteries can now be cleared by selectively catheterising them with a
Fogarty balloon catheter. An angiogram can be performed to check clearance of the vessels.
The arteriotomy is usually closed with a vein patch to avoid narrowing of the vessel.
If a saddle embolus is suspected at the aortic bifurcation then the femoral arteries are dis-
sected free and controlled in both groins. A Fogarty catheter is passed in a cranial direction
on both sides and the clot removed.
In the case of upper limb ischaemia the brachial artery and its bifurcation into the radial
and ulnar arteries should be accessed through a lazy S shaped incision in the ante cubital
fossa.The arteries are dissected free and controlled with sloops. A transverse arteriotomy is
made just proximal to the brachial bifurcation. Clot at the bifurcation can be removed with
forceps or by suction. A Fogarty catheter is then passed cranially and into the distal branches
to clear any possible embolic material. When good flow has been established the arteriotomy
is closed with prolene.
If an angiogram reveals the presence of residual clot in the vessels then intraoperative
thrombolysis can be used to dissolve the clot. An infusion of 100 000 units of streptokinase
or t-Pa 15 mg in 100 ml of normal saline can be used into the distal vessels over 30 min [4].
An angiogram is repeated to check adequacy of clearance.
Thrombosed arteries, thrombosed aneurysms and grafts can also present with threat-
ened limbs. In these scenarios the surgical approach is different. If there is an absent femoral
pulse and inflow cannot be established by embolectomy because of iliac atherosclerosis the
patient should have a femoral to femoral or axillary to femoral artery bypass.
The superficial femoral artery cannot be cleared sometimes due to atherosclerotic disease.
The patient should have an on table angiogram. Attempts should be made to clear residual
thrombus with intra arterial thrombolysis. If there is endovascularly treatable disease this
should be treated by an angioplasty. Lesions not amenable to angioplasty may require distal
bypass.
Bypass grafts may occlude in the immediate postoperative period or further down the
line. Grafts that thrombose in the immediate postoperative period should be explored as
this is usually due to technical reasons. Late graft occlusions are usually due to lesions
proximal to, in the graft, or distal to the graft due to vein valve site stenosis, neo intimal
hyperplasia or progression of disease. These are better first approached by thrombolysis.
These patients are sometimes primarily operated on due to the severity of their ischaemia.
Late graft occlusions are unlikely to be resurrected. They should have an on table angi-
ogram. If a distal vessel is identified then they should have a bypass procedure as long as
good inflow is established.
Thrombosed popliteal artery aneurysms require urgent intervention. An on table angi-
ogram will identify patent vessels if any are distal to the aneurysm. Embolic material in the
runoff should be cleared by thrombolysis. A bypass procedure should be performed and the
aneurysm excluded by ligation.
In the immediate postoperative period all patients should be anticoagulated with heparin.
They should be then anticoagulated with warfarin for a period of 3–6 months.
Surgery in the form of emergency amputation

A proportion of patients will present with irreversible ischaemia. They tend to have severe rest
pain, fixed staining and mottling, paralysis and profound sensory loss and usually undergo
above-knee amputation. No attempts should be made to revascularise these limbs as the
reperfusion injury is likely to prove fatal to the patient. About 10% of patients present with an
already nonviable limb and the 30-day amputation rate following ALI is 25–30% [5].
Endovascular treaments
Thrombolysis
Catheter directed thrombolysis is indicated in limbs that fall into categories I and IIa. [5]. As
the limb is not severely ischaemic, time is available to pursue thrombolysis.
Table 2.3 Contraindications to thrombolysis
• Active bleeding
• Cerebrovascular accident within 2 months
• Recent gastrointestinal (GI) bleeding
• Pregnancy
• Neurosurgical procedure within 2 months
• Vascular surgical procedure within 2 weeks
• Abdominal surgery within 2 weeks
• Bleeding disorder
• Extensive trauma
The advantages of thrombolysis are decreased risk of trauma to the endothelium,

issolution of clot in vessels that cannot be accessed by means of an embolectomy cath-
d
eter and low pressure reperfusion of the limb. They may reveal stenotic lesions that are
amenable to angioplasty. This will improve long-term patency rates if these lesions are
angioplastied.
Complications associated with thrombolysis include the following:
1. Myocardial infarction.
2. Cerebrovascular accidents.
3. Bleeding from puncture site.
4. Embolisation.
If persistent bleeding is an issue then the effect of the lytic agent can be reversed by
administering aprotinin. Fresh frozen plasma may be required and other clotting products
may need to be replaced.
The use of thrombolysis will depend upon availability, patient’s clinical condition and
local expertise. There is evidence from randomised clinical trials that at similar time inter-
vals the limb salvage rates with thrombolysis are comparable to surgery but with a lower
mortality rate [6–8]. Contraindications to thrombolysis are given in Table 2.3.
Other endovascular treatment options are percutaneous aspiration thrombectomy and
percutaneous mechanical thrombectomy.
• Aspiration thrombectomy involves the use of a large bore end hole catheter to aspirate
thrombus.
• Mechanical thrombectomy involves the use of devices that agitate, disperse and aspirate
thrombus.
The above procedures can be used along with thrombolysis to optimise results.
Palliative care
A proportion of patients will present with acute limb ischaemia while being very unwell. The
likely outcome for these patients is death. Attempts should not be made to intervene as this
will not change the outcome of their illness.
The aim of treatment should be to alleviate their symptoms. The local palliative care team
should be involved and the care of the dying pathway should be instituted.
Complications of acute limb ischaemia

Compartment syndrome
This is due to reperfusion of ischaemic tissue. Typically immediately after surgery the treated
limb is noted to be perfused but the patient complains of pain in the calf and inability to
dorsiflex the foot.
Reperfusion of an ischaemic muscle results in oedema of the muscle. Swelling is due to
failure of cellular membrane function and leaking of capillaries. As muscles are enclosed in a
bony fascial compartment this increases the volume and ultimately pressure within the com-
partment. As the pressure within the compartment increases, muscle perfusion decreases
resulting in further ischaemic injury. This in turn increases muscle edema. Thus a vicious
circle is established resulting in obstruction of veins, arteries, capillaries, nerve dysfunction
and infarction of the muscle. The anterior compartment is the most vulnerable.
If compartment syndrome is clinically suspected or with compartment pressures of more
than 20 mm Hg a fasciotomy is indicated. There are several techniques but usually a four
compartment fasciotomy should be performed.
Partially closed fasciotomy

The leg is cleaned and draped. The skin over the proximal part of the compartment is incised.
The deep fascia that is exposed through this incision is then incised. The remainder of the
deep fascia is then divided using scissors that are passed subcutaneously. This technique is
used in cases of chronic compartment syndrome. The anterior compartment is most com-
monly affected. In an acute situation this technique should not be used and an open, full
length fasciotomy should be performed.
Open fasciotomy
The leg is cleaned and draped. A four compartment fasciotomy is performed to decompress
the four compartments in the leg adequately. An anterolateral incision is made along the
fibula from below the knee down to the ankle. Through this incision the fascia overlying
the anterior compartment and peroneal compartment is incised. A posteriomedial incision
is made along the length of the leg. The underlying fascia is divided and the gastronemius
muscle in the superficial posterior compartment is exposed. The attachment of the soleus
muscle to the tibia is divided to decompress the deep posterior compartment. The fibula can
be excised through an anterolateral incision to achieve adequate decompression of all com-
partments. The lateral ends of all the fascial envelopes are attached to the fibula. Excision of
the fibula therefore decompresses all compartments.
The open wounds should be dressed with non adherent dressing material such as bac-
tigras and a loose bandage. The wounds are allowed to heal by secondary intention or split
skin grafts can be used.
Rhabdomyolysis
Muscle breakdown as a result of ischaemic injury can release myoglobin into the blood
stream. This can cause acute tubular necrosis leading to renal failure. Patients have dark
urine, elevated levels of serum creatine kinase and myoglobin in the urine. Treatment is
hydration, alkanising the urine and removing the source of myoglobin. Patients may require
haemofiltration or dialysis.
References for acute arterial occlusion of the legs.

Thrombolysis or Peripheral Arterial
1. Clason AE, Stonebridge PA, Duncan Surgery (TOPAS) Investigators. N Engl J
AJ et al. Acute ischaemia of the lower Med 1998; 338: 1105–11.
limb: the effect of centralising vascular 10. Williams B, Paulter NR, Brown MJ
surgical services on morbidity and et al. The BMS Guidelines Working
mortality. Br J Surg 1989; 76: 592–3. Party guidelines for management of
2. Earnshaw JJ. Demography and aetiology of hypertension: report of the Fourth
acute leg ischaemia. Semin Vasc Surg 2001; Working Party of the British Hypertension
14: 86–92. Society. J Hum Hypertens 2004; 18: 139–85.
3. Blasidell FW, Steele M, Allen RE. 11. Heart Outcomes Prevention Evaluation
Management of lower extremity Study Investigators. Effects of angiotensin-
arterial ischaemia due to embolism and converting enzyme inhibitor, ramipril, on
thrombosis. Surgery 1978; 84: 822–34. cardiovascular events in high risk patients.
4. Earnshaw JJ, Gaines PA, Beard JD. NEJM 2000; 342: 145–53.
Management of acute lower limb 12. POISE Study Group, Deveraux PJ,
ischaemia. In: Beard JD, Gaines PA, Yang H et al. Effects of extended-
eds. Vascular and Endovascular Surgery. release metoprolol succinate in patients
Elsevier, 2006; 169. undergoing non-cardiac surgery (POISE
5. Norgren L, Hiatt WR et al. Acute limb Trial): a randomised controlled trial.
ischaemia. In: TASC II Inter-Society Lancet 2008; 371(9627): 1839–97.
Consensus on Peripheral Arterial 13. Karnon J, Brennan A, Pandor A et al.
Disease. Eur J Vasc Endovasc Surg 2007; Modelling the long term cost effectiveness
33: Supplement 1. of clopidogrel for the secondary prevention
6. Critchley JA, Capewell S. Smoking of occlusive vascular events in the UK.
cessation for the secondary prevention of Curr Res Med Opin 2005; 21(1): 101–12.
coronary heart disease. Cochrane Database 14. Heart Protection Study Collaborative
of Systematic Reviews 2005; Issue 1. Art. Group. MRC/BHF Heart Protection Study
No.: CD003041. DOI: 10.1002/14651858. of cholesterol lowering with Simvastatin I
CD003641.pub2. 20, 536 high risk individuals: a randomised
7. Results of a prospective randomized placebo controlled trail. Lancet 2002;
trial evaluating surgery versus 360: 7–22.
thrombolysis for ischaemia of the lower 15. National Collaborating Centre for Chronic
extremity. The STILE trial. Ann Surg 1994; Conditions. Type 2 Diabetes: National
220: 251–66. Clinical Guideline for Management in
8. Ouriel K, Shortell C, Deweese J et al. A Primary and Secondary Care (Update).
comparison of thrombolytic therapy with London: Royal College of Physicians, 2009.
operative revascularisation in the initial 16. Gardner AW, Montgomery PS, Parker
treatment of acute peripheral arterial DE. Physical activity is a predictor of
ischaemia. J Vasc Surg 1994; 19: 1021–30. all cause mortality in patients with
9. Ouriel K, Veith F, Sasahara A. A intermittent claudication. J Vasc Surg 2008;
comparison of recombinant urokinase 47: 117–22.
with vascular surgery as initial treatment
Chapter
3
Chronic lower limb ischaemia,
critical ischaemia and the diabetic
foot
Chris Davies and Cliff Shearman
Key points
• P eripheral arterial disease (PAD) affects approximately 30% of the adult population
• PAD is a powerful marker of cardiovascular risk and the risk is related to the severity of
the PAD
• Risk factor management has been proven to reduce cardiovascular risk but many
patients do not receive adequate treatment
• The optimum treatment to improve walking in patients with intermittent claudication
is best medical treatment, supervised exercise and angioplasty
• Angioplasty and surgery appear to be equivalent in treating critical limb ischaemia
(CLI). However, angioplasty is cheaper due to reduced length of hospital stay
• Diabetic foot complications are a common cause of hospital admission and often
precede amputation. The majority of amputations in the UK are carried out in patients
with diabetes
• Patients with diabetes should be screened annually for neuropathy and PAD
• Multi-professional team approach to the management of diabetic foot complications
can reduce amputation rates
Epidemiology
Atherosclerotic arterial disease affecting the legs, peripheral arterial disease (PAD), is very
common. In many patients it is asymptomatic but commonly the first manifestation of PAD
is pain in the leg on walking, intermittent claudication. In some patients the blood supply to
the leg becomes further reduced to a level when pain is experienced at rest and ulceration
and gangrene occur (critical limb ischaemia, CLI). In the Edinburgh Artery study it was
found that 4.5% of men and women over 55 years of age had intermittent claudication but a
further 25% had evidence of asymptomatic disease. In one-third of the asymptomatic group
evidence of a major vessel occlusion was found. At 5-year follow up those that had subse-
quently developed claudication were found to have come from the asymptomatic group,
suggesting there may be a window of opportunity to prevent disease progression. Further
evidence of this high prevalence comes from the PAD Awareness Risk & Treatment: New
Resources for Survival (PARTNERS) study, which screened 6979 subjects over 70 years of
age or over 50–69 with a risk factor for vascular disease in 320 primary care practices in the
Chapter 3: Chronic lower limb ischaemia 59
USA. The study found 5.5% had symptomatic PAD and in total 29% of subjects had PAD.
Further, it is estimated that the prevalence of limb threatening or CLI is 220 per million of
the population [1].
A number of studies have identified that the majority of patients with PAD do not sig-
nificantly deteriorate from the point of view of their legs. In the Edinburgh Artery study the
annual risk of limb loss was less than 1–2%. However, 5–10% of subjects suffered a cardio-
vascular event per year, commonly a myocardial infarction or stroke. This is an extraordi-
narily high cardiovascular risk and patients with PAD are six times more likely to die from
cardiovascular disease over 10 years than non-PAD subjects.
The severity of PAD is a prognostic indicator of cardiovascular risk, those with the most
severe symptoms faring worse. In patients with CLI the cardiovascular death rate is even
worse and one in five will be dead within one year of diagnosis. Although less marked even
the asymptomatic group have an increased cardiovascular risk.
Over the past decade this observation has often led to the main focus of treatment shifting
to address cardiovascular risk in patients with PAD by attempting to modify their risk factors.
However, it is important to remember that many patients with intermittent claudication are
extremely handicapped by their symptoms and these may need addressing in their own right.
Although it might seem appealing to screen the adult population for PAD there is currently
no strong evidence that this approach would reduce cardiovascular risk or be cost effective.
Intermittent claudication
History
Tight, cramp like pain in the muscles of the calf, thigh or buttock is characteristic. Calf
claudication is the most common symptom, simply because the most common site for of
PAD is in the superficial femoral or popliteal arteries. More proximal disease is required for
buttock or thigh claudication. Occasionally buttock claudication can occur from isolated
internal iliac artery blocks – usually bilateral – but this is rare. Typically the pain of clau-
dication comes on only after walking for a pain free distance, is usually worse on hills, and
is relieved by resting in the standing position. Most other causes of hip or leg pain, such as
arthritis, may be exacerbated by walking but will also cause some pain at rest or in certain
positions or postures. If the pain only subsides when the patient sits down this suggests
the symptoms may be related to their back rather than arterial disease. Spinal canal clau-
dication typically produces numbness, weakness or heaviness in the leg rather than pain
localised to a specific muscle group. Spinal canal claudication, due to a narrow spinal canal,
may actually be easier going uphill as leaning forward opens the spinal canal up.
It is usual, when taking a history of claudication, to ask the patient to quantify their
walking distance – but this is notoriously inaccurate as it will vary with speed and terrain.
Additionally patients are often extremely poor at estimating distances. For research pur-
poses walking distances are usually assessed by walking on a treadmill at a set speed and
degree of slope. Usually the ‘initial claudication distance’, which is the distance at which the
pain first starts and the ‘absolute claudication distance’, which is the distance at which the
patient stops, are recorded.
Examination
A full cardiovascular examination is important to detect other manifestations of cardio-
vascular disease. Upper limb peripheral pulses should be palpated and the cardiac rhythm
checked. Measurement of blood pressure, cardiac auscultation and abdominal examination

for the presence of an abdominal aortic aneurysm should be performed. An abdominal aor-
tic aneurysm cannot be fully excluded by clinical examination and any suspicion should
prompt an ultrasound scan.
Examination of the peripheral circulation should include inspection for clinical signs
such as ulceration and gangrene. Other, less definitive signs of ischaemia such as cracked
skin, hair loss and nail damage should be recorded but are not very useful in their own right.
Capillary return is notoriously unreliable unless particularly prolonged, but it should prob-
ably also be assessed. In a patient with diabetes it is important to look for signs of abnor-
mal foot shape and callus formation. Skin colour and temperature should also be recorded.
The most important aspect of the peripheral examination is peripheral pulse palpation. The
femoral, popliteal, posterior tibial and anterior tibial/dorsalis pedis pulses should be identi-
fied. In a small number of normal subjects (10–15%) the dorsalis pedis may be absent and in
some individuals it is possible to palpate the anterior communicating branch of the peroneal
artery on the lateral side of the ankle.
Usually it is sufficient to judge whether a pulse is present or absent. Lower limb pulses
can be difficult to feel and trying to judge whether a pulse is strong or weak is rarely of
great use. In some patients a hard calcified vessel may be palpated, which is important to
identify as it may have implications for treatment. A rough rule is that in patients with calf
claudication the disease will be predominantly in the superficial femoral artery and so the
popliteal pulse will be absent. For thigh and buttock claudication the disease is predomi-
nantly in the aorto-iliac segment and the femoral pulse will be weak or absent. Although
this is not infallible, it is a useful concept to bear in mind when examining the peripheral
circulation.
Thigh and buttock claudication may be associated with male impotence (Leriche syn-
drome). It may therefore also be appropriate to enquire about the ability for a male to attain
an erection. A small number of patients will have isolated disease of their internal iliac arter-
ies causing their claudication and so will have palpable femoral pulses. This may also be
seen in patients who have had therapeutic occlusion of their internal iliac arteries to allow
endovascular repair of an aorto-iliac aneurysm.
Auscultation over the femoral artery may detect the presence of a bruit. We believe this
is of limited significance as it simply implies turbulent flow proximally. However, in patients
with a good history of claudication but palpable pulses a bruit may signify proximal disease,
which is only haemodynamically significant during exercise.
Ankle brachial pressure index (ABPI) measurement

Measurement of the ABPI is the third part of the diagnostic triad. The ankle pressure is
expressed as a ratio compared to the brachial pressure. An ABPI of 0.9 or less is usually
taken to be abnormal. In some patients, especially those with diabetes, the peripheral arter-
ies may be stiff or calcified, which can give an artificially high ankle pressure. This is usu-
ally obvious but if doubt remains further investigation may be required. In symptomatic
patients with an ABPI of <0.9 there is a 95% sensitivity of diagnosing PAD. To measure the
ABPI you need a hand held Doppler with an 8 MHz probe, sphygmomanometer and cuff.
Make sure the patient is lying flat, ideally for 10 min before starting. The test is invalid if the
patient’s legs are not at the same level as the heart. Secure the cuff around the arm as if tak-
ing a blood pressure reading in the normal manner. Apply gel over the brachial pulse and
apply the Doppler probe over the brachial pulse, at an angle of 60º to the skin. Repeat on
the other arm and use the higher of the two readings to calculate the ABPI. When measur-
ing pressures in the dorsalis pedis and posterior tibial arteries make sure the cuff is placed
on the lower calf.
Toe pressures/pole test

Patients with diabetes, renal failure and other conditions causing vascular calcification can
develop incompressible tibial arteries, causing falsely high systolic pressures when measur-
ing the ABPI. Measurement of toe pressures provides an accurate alternative measurement.
The toe pressure is normally about 30 mmHg less than the ankle pressure and an abnormal
toe brachial index (TBI) is defined as <0.70. Usually this is not done using a hand held
Doppler but by using a system of automatically inflating cuffs on the great toes (sometimes
other toes as well) and by recording the pulse using a plethysmographic technique or laser
Doppler.
To measure ABI using the pole test the leg is slowly elevated from the horizontal position
until the Doppler signal disappears. The height above the left ventricle is measured using a
calibrated pole (i.e. a ruler) to record the hydrostatic pressure in equivalents to mmHg. For
example 70 cm is equivalent to 50 mmHg.
Differential diagnosis
In the majority of patients the diagnosis of intermittent claudication due to PAD can
be confidently confirmed by the simple measures outlined above. Other conditions
that should be considered in the differential diagnosis are nerve root irritation due to
degenerative lumbar spine disease in which the pain often takes longer to wear off and
may only be relieved by sitting. True spinal canal stenosis is relatively uncommon and
again the pain is characteristically relieved by sitting or flexing the lumbar spine. Other
muscular skeletal conditions such as osteoarthritis are usually possible to identify due
to their associated symptoms of nocturnal pain, early morning stiffness and pain that
persists for hours after ceasing walking. Venous claudication is rare and usually the
patients will have a history of previous venous problems or signs of chronic venous
insufficiency.
There is a small group of patients who have a non-atherosclerotic cause of their vas-
cular intermittent claudication. They are often younger, have no other manifestations of
cardiovascular disease and their symptoms may only occur after walking some distance.
These patients should however be investigated further to exclude other rare causes of exer-
cise induced ischaemia, which may need specific treatment. These include cystic advential
disease, popliteal entrapment syndrome, congenital abnormalities such as persistent sciatic
arteries and fibromuscular dysplasia.
Investigations
In some patients the diagnosis of intermittent claudication may remain uncertain. They
may have other co-morbidities that make it difficult to determine the dominant cause of the
symptoms or they may have non-compressible vessels. In these patients treadmill exercise
testing by an experienced technician may be of value. Observation of the patient walking
and determination of the distance walked on a treadmill may be helpful in determining
the origin of the pain. However, relatively few patients will need this level of investigation.
Treadmill walking (the distance to the onset of pain, the claudication distance, and the maxi-
mum distance they can walk, absolute walking distance) is most commonly used when new
therapies are being evaluated.
If there still remains doubt as to the presence of PAD, such as in a patient with diabetes
or if intervention is being considered, then colour flow Duplex (CFD) ultrasound scan-
ning should be undertaken. CFD utilises the B mode ultrasound component to identify
the vessel and then interrogates them with Doppler ultrasound to determine the veloc-
ity of blood flow through them. Direction of flow is colour coded on the display, which
makes it easier and quicker to identify areas of disease due to turbulence or cessation of
flow. This will display the extent of the PAD and also, using the Doppler component, give
an objective evaluation of the haemodynamic severity of the disease based on the velocity
shift across the diseased segment. Duplex is ideal for the infra-inguinal vessels but can be
more difficult in the aorto-iliac segments. Usually an experienced ultrasonographer can get
enough information to determine the extent and severity of the disease. Bubble contrast
enhancing agents have been used but currently have not been found to be of great value
in the investigation of PAD. Duplex ultrasound will confirm the diagnosis of PAD but will
also usually give a strong indication of what interventions are likely to be applicable to an
individual, e.g. angioplasty or bypass surgery. In many patients with clear Duplex scans no
further imaging is required although some clinicians prefer more detailed imaging prior
to intervention.
Depending on the pattern of disease and nature of the planned intervention more detailed
information may be required. Digital subtraction angiography (DSA) is now rarely used for
diagnostic imaging. It is invasive and requires a significant volume of contrast medium,
Figure 3.1 Computed tomog-

raphy angiogram showing calf
vessels of patient with diabetes.
Calcification is clearly seen.
Figure 3.2 Magnetic resonance angiogram of

patient with right iliac artery occlusion.
which is nephrotoxic. Multi-slice computed tomography angiography (CTA) is the preferred

mode of imaging in many units (see Figure 3.1).
It is relatively non-invasive with very rapid acquisition times under 30 s. However an
intravenous contrast medium is required and contrast induced nephropathy can occur,
particularly in patients with impaired renal function. The radiation dose is also significant
being 80–200 chest X-ray equivalents and, although for single studies this is not an issue, in
patients having repeated scans this must be borne in mind. Image interpretation can pose
problems, particularly with calcification in the vessel wall. Magnetic resonance angiography
(MRA) is also widely used (see Figure 3.2).
Calcification does not cause the same problems for CTA but acquisition times are longer.
Recognition of nephrogenic systemic fibrosis, which may be induced by the contrast agent
gadolinium in patients with impaired renal function, has limited its role in these patients.
Both of these techniques are ideal to use as outpatient investigations.
Management
The two aims of treatment are to reduce the cardiovascular risk and improve the walking
ability of the patient.
Reducing cardiovascular risk

Smoking is the most important risk factor for the development of PAD and even passive
smoking increases cardiovascular risk. Not only is disease progression and amputation
more likely in smokers, but smoking increases the risk of hypertension and raised choles-
terol. Excess cardiovascular risk is halved within one year of cessation and is the same as
non-smokers within 5 years in those patients that successfully give up smoking. There is no
strong evidence for the benefits of smoking cessation to the limb but some observational
studies have suggested an improvement in walking distance and a reduction in amputa-

tion rates. Smoking cessation advice, when combined with nicotine replacement therapy,
improves quit rates to around 30%. Recent UK legislation prohibiting smoking in public
areas may also have an impact on the numbers of smokers.
The prevalence of diabetes is increasing across the world. Patients with type 2 diabetes
have 3–5 times more chance of developing PAD. Also the effects of other risk factors such
as hypertension and hypercholesterolaemia are amplified in patients with diabetes. Studies
looking at tight diabetic control in type 2 diabetes have shown a reduction in cardiovascu-
lar events but the benefits of this were outstripped by the complications of hypoglycaemia.
At present in patients with diabetes and PAD every attempt should be made to identify and
reduce any other risk factors and diabetic control should be optimised for the individual
patient. Patients with diabetes should also have regular checks of their feet for signs of neur-
opathy or PAD.
There is overwhelming evidence for the benefits of lowering cholesterol in patients with
PAD. In the Heart Protection Study, patients with PAD and a total cholesterol over 3.5 mmol l–1
who took simvastatin (a HMG-CoA reductase inhibitor) had a 17.6% reduction in cardio-
vascular events compared to those on placebo [2]. There was also a reduction in the subse-
quent need for both cardiac and non-cardiac revascularisation procedures. Based on these
results nearly all patients with PAD should be prescribed statin therapy. There is also emerg-
ing evidence that statins have a direct effect on atherosclerotic plaque, stabilising it and pos-
sibly causing plaque regression in high doses.
Up to 24% of the adult population are hypertensive and hypertension is associated with
a threefold increase risk of PAD as well as being strongly associated with stroke and myo-
cardial infarction. Treatment of hypertension will reduce stroke rates by 38% and cardio-
vascular deaths by 14%. In the Heart Outcomes Study, the angiotensin converting enzyme
inhibitor, ramapril, demonstrated an advantage in reducing cardiovascular events, even in
those patients whose blood pressure was not elevated [3]. However there are potential prob-
lems with the widespread use of ramapril in patients with PAD as many will have renal
artery disease. At present in those with PAD and hypertension ramapril should be consid-
ered as the first-line treatment but there is not enough evidence to suggest widespread use
in non-hypertensive patients.
The Antithrombotic Trialists’ Collaboration meta-analysis found that antiplatelet agents
(predominantly aspirin, a cyclo-oxygenase inhibitor) reduced the risk of cardiovascular
events by 23% in patients with PAD [3]. A dose of 75 mg was as effective as higher doses.
Approximately 20% of patients are unable to take aspirin largely due to gastrointestinal dis-
turbance and it is emerging that a similar proportion of patients have aspirin resistance. In
these patients normal doses of aspirin do not have the normal effect on patients. In these
patients clopidogrel should be used. Clopidogrel is a theopyridine derivative that blocks
ADP induced platelet activity. In the CAPRIE study, clopidogrel was shown to further
reduce cardiovascular events compared to aspirin (particularly in the PAD group), with
a relative risk reduction of 8.7%. However, as clopidogrel is more expensive the National
Institute for Health and Clinical Excellence (NICE) recommends that it should be used
only by patients who cannot take aspirin or by high-risk patients. Combination therapy of
aspirin and clopidogrel should be considered very carefully. In the Charisma study patients
on both drugs had a significantly greater risk of bleeding complications, which overall
exceeded any apparent benefit.
A number of other lifestyle changes should be advocated. Weight reduction and regular
exercise have proven cardiovascular benefit. They also have a positive effect on other risk
factors. Omega 3 fatty acids (fish oils) appear to have some beneficial effects but their clinical
role in PAD has not been established. Likewise antioxidants and other dietary additives have
not been demonstrated to be of benefit.
Despite the overwhelming evidence for the benefit of risk factor management recent
epidemiological studies have found that only around 30% of patients get adequate treatment
for these.
Improving walking distance

In many patients confirmation of the diagnosis of intermittent claudication, reassurance of
the natural history of the condition as regards the leg and risk factor management will be all
that is required. The decision to directly attempt to improve walking distance is something
that should be decided by the patient balanced by the impact of their symptoms on their
day-to-day life compared to the chance of success versus the risks of treatment. Patients
who have claudication are a heterogeneous group. Many will only be mildly troubled by
their symptoms or have other significant co-morbidity that reduces their mobility. Others,
however, may be severely restricted by their claudication, which can significantly alter their
lifestyle. It is the role of the clinician to help the patient decide on the best therapeutic option
for them based on the impact of their symptoms on their quality of life.
A number of vasoactive drugs have been promoted to increase walking distance. In the
UK two drugs are currently used, naftidrofuryl and cilostazol. The benefit of naftidrofuryl on
walking distance in claudication is variable and generally small. In a meta-analysis of eight
studies cilostazol has been demonstrated to improve walking by 50% with some improve-
ment in quality of life. Whether this is of benefit to individual patients or cost effective has
not been established. In the UK cilostazol is generally used in patients when other physical
interventions are not possible and the patients would benefit from some improvement in the
walking distance. However, if symptoms do not improve after 6–8 weeks the drug should be
discontinued.
The benefits of exercise have been realised for over a decade. Until recently no large
randomised studies of exercise have been conducted but a meta-analysis of 21 studies sug-
gested an improvement in walking distance of 124%. Supervised exercise with programmes
requiring three sessions of 30 min per week seemed to be the optimum, but even upper limb
exercises have been shown to improve walking. The main problem with this approach is
making it appealing to patients and generally less than 30% of them will either want to try
exercise or continue on a programme.
Two small randomised trials of angioplasty versus exercise showed the benefit of exer-
cise in terms of walking distance. The recently published Mimic Trial compared best medi-
cal treatment (BMT) and exercise to BMT, exercise and angioplasty [5]. In this study of
144 patients with stable claudication they found a significant advantage for angioplasty
above exercise and BMT alone, especially for patients with aorto-iliac disease. These results
strongly suggest a broad approach to the treatment of patients with claudication and com-
bining exercise with angioplasty, if indicated, to gain the maximum benefit.
Generally percutaneous angioplasty should be appealing for patients with intermittent
claudication. It is relatively non-invasive, often can be done as a day case and significant
complications occur in less than 3% and limb loss in around 0.3%. However re-stenosis
rates, particularly in more distal disease, are significant and can be as high as 40% for distal
superficial femoral artery disease. If patients with short stenoses or occlusion in the proximal
vessels only are treated then about 50–60% of patients will be suitable. Stent placement after
angioplasty may improve the initial technical success rates but has not been demonstrated
to improve clinical outcome. Until recently the only evidence of comparison of angioplasty
compared to exercise favoured exercise. This tended to restrict the use of angioplasty in the
UK. It is likely that the findings of the Mimic Trial will change this.
Surgical reconstruction (bypass or common femoral endarterectomy) has been dem-
onstrated to improve quality of life in patients with claudication. However, the risks of
morbidity and mortality are significant. There are still a few patients with simple lesions
who may be offered surgery. Usually for some reason they will not be suitable for endovas-
cular treatment or will have had multiple re-stenoses after angioplasty. However, the
expectation of a good result should be high and the patient should be fully aware of the
potential risks.
Critical limb ischaemia (CLI)

A large number of patients who present with limb threatening ischaemia have not had a
previous history of claudication. They are often less mobile so may not have precipitated
any symptoms but often other factors, such as infection, cause a sudden deterioration. For
this reason it has proved difficult to accurately define CLI. The Inter-Society Consensus for
the Management of Peripheral Arterial Disease (TASC II) suggests that a patient with per-
sistent pain, ulcers or gangrene, considered to be due to proven arterial disease, should be
considered to have CLI [1]. This definition is based on previous attempts to define the condi-
tion in the Trans-Atlantic Inter-Society Consensus (TASC) of 2000. The diagnosis is usually
confirmed by a low ankle blood pressure (under 50 mmHg) but this is not always the case
in some patient groups, e.g. for diabetes. The need to intervene in patients with CLI is obvi-
ously much greater as a significant proportion will deteriorate and face limb loss without
revascularisation.
Patients with CLI face an enormous cardiovascular risk and 50% will be dead within
1 year of diagnosis. It is important to recognise, and when possible correct, any associated
risk factors. These patients also tend to be older and have significant co-morbidities that
need to be optimised.
Options for revascularisation include angioplasty or surgery. Conventional trans-
luminal angioplasty has limited results for more distal disease but the development of low
profile catheters and stents for coronary artery work has improved outcomes. Sub-intimal
angioplasty seems to have greater initial success but the medium-term patency rates tend
to be variable. However, in many patients even if the original angioplasty site re-occludes,
ulcers have often healed and it seems the angioplasty worked long enough to achieve wound
healing.
Surgical bypass
Surgical bypass can either be inflow procedures to the femoral artery; anatomical bypasses,
such as aorto-bifemoral bypass; or extra-anatomical bypasses, such as cross femoral bypass
or axillo-femoral bypass. Distal bypass is technically demanding but if carried out in expe-
rienced units will achieve excellent results. A meta-analysis of infra-geniculate bypass in
high-risk patients found limb salvage rates of over 78% at 5 years. The optimum bypass
graft is autologous vein and if long saphenous vein is not available excellent results can be
obtained using arm vein, short saphenous vein or the deep veins. Although moderate results
can be obtained, there is little role now for the use of prosthetic grafts below the knee due to
the high risk of infection.
In patients selected for surgery great care needs to be taken to identify the optimum
run-off vessel and it is also important to carry out Duplex ultrasound scanning of he veins
to ascertain if they will be suitable and to minimise the dissection required to harvest them.
There is no clear advantage of in situ versus reversed vein techniques and it is really depend-
ent on the experience of the surgeon.
Up to 30% of vein bypass grafts will develop stenoses and if left many will go onto
occlude. It is not possible to detect these stenoses clinically and many patients are asympto-
matic. Many units therefore carry out regular Duplex ultrasound surveillance of the grafts
and correct any significant stenosis (greater than 50% or three times velocity shift across the
narrowed segment). However, a randomised controlled study has brought this practice into
question. Although the study found it was possible to detect stenoses, it found no differ-
ence in outcome between patients in surveillance and those simply followed up. As Duplex
surveillance is time consuming and costly it was found not to be cost effective. Many units,
however, continue to monitor these patients, arguing that there are other benefits, such as
checking on risk factor management and long-term outcomes.
There are then a number of options available to re-vascularise a critically ischaemic limb.
The TASC II document has classified lesion in the aorto-iliac and femoro-popliteal segment
and reviewed the evidence for treatment options [1]. However, a randomised controlled study,
the Bypass Versus Angioplasty in Severe Ischaemia of the Leg (BASIL) study, compared surgery
versus angioplasty for the treatment of CLI and severe limb ischaemia [6]. This study found no
difference in amputation free survival, which was 55% during the study period, between either
treatment. There was also no difference in mortality, 3% and 5%, respectively, for angioplasty
and surgery. However although there was an initial 20% failure rate from angioplasty, length of
stay in hospital was much shorter, making angioplasty more economical.
There are a number of important points that arise from this study. First only 30% admit-
ted to the units involved in the study were randomised, suggesting a large proportion were
not re-vascularised. Also 37% died during the study period, largely from cardiovascular dis-
ease, emphasising the high cardiovascular event rate in these patients. Few patients were
made worse by angioplasty, suggesting that in most of these patients who are elderly and frail
it is worth trying angioplasty initially.
One group of patients in whom surgery is usually the optimum treatment are those with
disease in the common femoral artery. Angioplasty here may put the profunda artery ori-
gin at risk and often these patients are treated by common femoral endarterectomy. There
still remains some doubt over which is the optimum technique to re-vascularise the distal
tissues in patients with diabetes and further studies are needed to determine this. The role
of surgical profundaplasty (usually an endarterectomy of the common femoral artery into
the profunda artery with patch closure) remains contentious. If no other options are avail-
able, or a distal procedure is not appealing due to poor quality distal vessels, it is worth
attempting.
Amputation
Amputation is the only option possible in some patients. Extensive infection may require
urgent intervention. If the patient is very sick then guillotine amputation of the infected tis-
sue with subsequent conversion to formal amputation may be the best option. It will achieve
Table 3.1 Eight clinical features to look for in the diabetic foot
• Neuropathy
• Ischaemia
• Deformity
• Callus
• Swelling
• Skin breakdown/ulceration
• Infection
• Necrosis
control of sepsis and avoid infection of a definitive amputation stump. The patient can be
returned to theatre when they are stable for conversion of the guillotine stump to the opti-
mal formal amputation.
Primary amputation may also be required in patients in whom there is extensive tis-
sue loss and no obvious method of re-vascularisation of the limb. Secondary amputation
is required when a revascularisation attempt has failed to achieve tissue healing or relief of
pain. Below-knee amputation has the advantages of greater mobility with a prosthesis due to
preservation of the knee joint. However the primary healing rate of below-knee amputations
is approximately 80% compared to 90% with above-knee amputation.
Selection of amputation level can be difficult. The presence of a good femoral pulse
together with a well developed profunda femoris artery is encouraging for a below-knee
amputation. The quality of the tissues in the leg, previous wounds and the presence of
sepsis and the general condition of the patient also make a major contribution to the
decision at what level to amputate. The use of measurements such as TcPO2 at the level
of amputation has been found to be helpful by some units but the general application
of these techniques has not been widespread. Often the decision has to be made based
on the clinical factors outlined above and the appearances of the tissues at the time of
surgery.
Diabetic vascular disease

It is estimated that by 2010 there will be 221 million people diagnosed with diabetes world-
wide [7] and in England the current prevalence is 3.75% [8]. In recent epidemiological stud-
ies it has been found that over 40% of patients presenting with PAD have diabetes and 20%
of all hospital admissions for complications of diabetes are due to foot problems. An esti-
mated 5% of patients with diabetes develop a foot ulcer every year and the lifetime risk of an
ulcer is 15% [9]. The ultimate consequence of foot disease without adequate management is
amputation; and the relative risk of amputation, associated with all types of diabetes, is 13
times compared to that of a non-diabetic [10]. It has been estimated that 85% of amputations
could have been avoided with adequate care.
There are two main reasons why patients with diabetes are so prone to foot complica-
tions, neuropathy and ischaemia. Screening patients with diabetes is essential to detect early
signs of neuropathy and PAD (Table 3.1). This can be simply done by the use of a 10 g mono-
filament test, peripheral pulse palpation and ankle blood pressure measurement. Patients
who have any evidence of neuropathy or PAD should be given help and advice to detect
problems early and seek help. There is encouraging evidence from some countries, such as
Finland, that a multi-disciplinary approach to diabetic foot complications can reduce major
amputation rates.
Neuropathy
1. Sensory neuropathy reduces the patient’s ability to appreciate damage to the foot such
as rubbing of ill-fitting footwear. Thus small lesions, such as from ill-fitting shoes, are
not appreciated at an early stage by the patient. Also lack of proprioception means that
when the patient is standing subtle movements in posture that would normally offload
pressure do not happen.
2. Motor neuropathy results in reduced power of the short flexor muscles of the foot. The
resulting in-balance between these and the long extensors results in an ‘intrinsic minus’
foot with clawing of the toes and prominence of the metatarsal heads. Abnormal weight
bearing on the metatarsal heads (combined with reduced sensory proprioception)
causes tissue damage and ulceration (Figure 3.3 and Table 3.1).
3. Finally autonomic neuropathy causes reduction in sweating and dry skin, which is
more prone to cracking and damage. The failure of the autonomic system also has
effects on the vasomotor tone in the skin circulation with abnormal arteriovenous
shunting. All of these factors make the foot prone to damage and ulceration.
Ischaemia
Patients with diabetes have a fourfold increased risk of developing PAD and the result-
ing ischaemia is another factor predisposing the foot to injury and poor healing. Patients
with diabetes and PAD have the greatest risk of suffering and amputation. There are also
Figure 3.3 Typical neuropathic foot ulcer. The ulcer

is surrounded by a ring of keratinised skin.
abnormalities of the microcirculation such as thickening of the basement membrane,

increased capillary permeability and increased platelet adherence, which reduce tissue
perfusion.
It has been estimated that of patients with foot complications approximately 10% will
be due to ischaemia, 45–60% will be due to neuropathy and 25–45% will be a combination
of the two (=neuro-ischaemia). However the final common presenting factor is often infec-
tion. The immunosuppressive effect of hyperglycaemia predisposes the patient with dam-
aged ischaemic tissue to infection, which can be extensive.
Management of the diabetic foot

The first-line treatment of a patient with a diabetic foot complication is to ensure that their
diabetes is controlled and this may require a sliding scale or correction of a keto acidosis.
Treatment of infection involves cultures and local wound swabs followed by intravenous
antibiotics. Units should have agreed regimens with their microbiology team, which will
cover the common causative agents such as staphylococcus aureus, streptococcus, pseu-
domonas and anaerobic bacteria. Superficial wound swabs may fail to identify the causative
organism and aspiration of tissue fluid using a 10 ml syringe and a green needle is advocated
by some groups to obtain more accurate deep tissue cultures.
The extent of the infection is determined by careful inspection of the foot and palpation
of the tissues. Tenderness or crepitus suggests extensive infection. Often infection of a digit
will spread to the associated web space and infect the tissue between the metatarsals. This
may only be apparent by a slight redness and tenderness over the dorsum of the foot. Plain
X-rays should be undertaken. They may reveal osteomyelitis or gas in the tissues. Surgical
drainage of any collection and removal of dead infected tissue should be undertaken within
24 hours of admission and further tissue sent for microbiological culture.
Once the situation has been stabilised by the above, full assessment of the neurological
and vascular status of the limb can be undertaken. The absence of pedal pulses together with
a low ankle blood pressure indicate a high-risk foot. Further investigation, starting with a
Duplex scan and probably including a CTA, should be obtained. Unless the foot shows signs
of healing revascularisation either by angioplasty or surgery should be performed.
Wounds on the foot should not be sutured but generally left to heal by secondary inten-
tion. Attempting to suture the skin may reduce skin edge blood flow and trap infected mate-
rial, leading to abscess formation. Care should be taken to avoid oedema of the limb and the
patients should rest with their leg elevated for the first few days. The use of negative pressure
wound therapy has transformed management of these patients. Although the precise way in
which it works is not clearly understood, negative pressure therapy will remove excess tissue
fluid, increase microcirculatory blood flow beneath the wound and reduce the wound size.
Patients can be mobilised early using total contact casts. A cast is placed on the lower limb
with a window to allow inspection of any wound. These need to be expertly fitted as if there
are abnormal areas of pressure the cast itself will cause damage.
Conclusion
PAD is a common problem and is strongly associated with cardiovascular risk. Despite
overwhelming evidence for the benefit of the treatment of cardiovascular risk factors many
patients remain undiagnosed and under-treated. Recent evidence has suggested that com-
bining best medical treatment, exercise and intervention has the best results for symptomatic
disease. Endovascular treatments seem to have the same result as surgical treatments in term
of limb salvage. The commonest cause of amputation is diabetes, the prevalence of which is
increasing. Evidence shows that early detection of foot problems in diabetes and a multi-
disciplinary approach to the problem can reduce amputation rates.
References supervised exercise, smoking cessation

advice and best medical therapy: results
1. Norgren L, Hiatt WR, Dormendy JA, from two randomised trials for stenotic
Nehler MR, Harris KA, Fowkes FGR femoro-popliteal and aortoiliac arterial
on behalf of the TASC II working disease. European J Vasc Endovasc 2008;
group. Inter-Society Consensus for the 36: 680–8.
Management of Peripheral arterial disease 6. Adam DJ, Beard JD, Cleveland T and
(TASC II). Norgren L, J Vasc Surg 2007; BASIL Trial Participants. Bypass Versus
45: Supplement. Angioplasty in Severe Ischaemia of the
2. Heart Protection Study Group Leg (BASIL): multicentre, randomised
Collaborators Group. MRC/BHF heart controlled trial. Lancet 2005; 366: 1925–34.
protection study of cholesterol lowering 7. Amos AF, McCarty DJ, Zimmet P. The
with simvastatin in 20,536 high-risk rising global burden of diabetes and its
individuals: randomised placebo complications: estimates and projections to
controlled trial. Lancet 2002; 360: 7–22. the year 2010. Diabet Med 1997; 14 Suppl
3. The Heart Outcomes Prevention Evaluation 5: S1–S85.
Study Investigators. Effects of angiotensin- 8. Scanlon P, Stratton I. Incidence and
converting enzyme inhibitor, ramapril, on prevalence of diabetes. NHS Evidence –
cardiovascular events in high-risk patients. Diabetes June 2008. http://www.library.
N Engl J Med 2000; 342: 145–53. nhs.uk/diabetes.
4. Antithrombotic Trialists’ Collaboration. 9. National Institute for Health and Clinical
Collaborative meta-analysis of randomised Excellence (NICE) Guideline, Type 2
trials of antiplatelet therapy for the diabetes: prevention and management
prevention of death, myocardial infarction, of foot problems. January 2004. Clinical
and stroke in high risk patients. BMJ 2002; guideline 10. http://www.nice.org.uk/
324: 71–86. CG010NICEguideline.
5. Greenhalgh RM, Belch JJ, Brown LC. The 10. International Diabetes Federation. Position
adjuvant benefit of angioplasty in patients Statement – The Diabetic Foot. http://www.
with mild to moderate intermittent idf.org/Position_statementsdiabetic_foot.
claudication (MIMIC) managed by
Chapter
4
Endovascular and surgical options
for peripheral revascularisation
Colin Nice
Key points
• R isk factor optimisation and best medical therapy are the standard of care for all patients
• Severe acute ischaemia is best managed with surgery, there is a role for thrombolysis in
less severe cases
• Thrombolysis requires intensive monitoring to identify and manage complications
• Surgical or endovascular revascularization is appropriate for patients with limiting
claudication or critical limb ischaemia
• Non-invasive imaging should be used for procedural planning
• Bypass grafts with autologous vein produce the best long-term patency rates
• Endovascular procedures have lower mortality and morbidity rates than the equivalent
surgery
• Stents and stent grafts improve endovascular results and are important for managing
complications
• Patient fitness, co-morbidity and preference are as important as lesion characteristics in
informing revascularization decisions
• Multi-disciplinary teams are best placed to manage individual patients in this rapidly
evolving field
Background
Many patients with peripheral arterial disease (PAD) do not require any revascularization
procedure. Identification and management of modifiable risk factors are effective in redu-
cing the excess risk of cardiovascular mortality and preventing acute limb ischaemia due to
disease progression. Also supervised exercise programmes can benefit those with intermit-
tent claudication, a Cochrane review of randomised trials in patients with stable intermit-
tent claudication suggested an improvement in walking distance of 150% with a regime of
three sessions per week of walking to near maximum pain [1].
However, surgical and endovascular revascularization procedures produce substan-
tial additional benefits when proficiently performed upon carefully selected and prepared
patients. Treatment strategies addressing the differing aetiologies of chronic peripheral
arterial disease and acute limb ischaemia determine the urgency, nature and effectiveness of
attempted revascularization.
Chapter 4: Peripheral revascularization 73
BASIL trial
The BASIL trial (Bypass Versus Angioplasty in Severe Ischaemia of the Leg) compared the
outcomes of 452 patients presenting to UK centres with severe limb ischaemia [2]. Patients
were randomized to either a surgery first (n = 228) or an angioplasty first (n = 224) strat-
egy. Only 60% of participants received antihypertensive treatment, 54% received antiplatelet
drugs and 34% were taking cholesterol lowering agents.
The primary endpoint was survival free from amputation for the trial limb. Trial dura-
tion was 5.5 years, with follow up ending when the endpoints of death or above-ankle ampu-
tation of the trial leg were reached.
At follow-up completion;
• 248 patients (55%) were alive without amputation of the trial leg;
• 38 (8%) were alive following amputation;
• 130 patients (29%) were dead without amputation.
A surgery first strategy was associated with a higher rate of early morbidity (57% versus
41%). Six-month amputation free survival was not significantly different between the treat-
ment groups and health related quality of life (HRQL) was similar. At 2 years surgery was
associated with a reduced risk of future amputation or death. A strategy of surgery first
increased the hospital costs by about one third.
These results indicate that there is scope to substantially improve the medical treatment
of risk factors. For patients with a short life expectancy an angioplasty first strategy reduces
early morbidity and treatment costs but for relatively fit patients, expected to live beyond
2 years, a surgery first strategy is more durable and carries a reduced re-intervention rate,
possibly outweighing the initial costs and increased short term morbidity.
Acute limb ischaemia

Acute limb ischaemia may be due to arterial occlusion arising from in situ thrombosis of
established atheromatous plaques or embolization from remote sites. The majority of emboli
are cardiogenic, resulting from atrial fibrillation or mural thrombus, complicating myocar-
dial infarction. Aortic and popliteal aneurysms are other important causes and these should
be sought with clinical examination and imaging.
Treatment decisions are often based on clinical assessment due to clinical urgency or
limited availability of other imaging modalities.
Anticoagulation
Immediately following the diagnosis of acute limb ischaemia, anticoagulation with intra-
venous heparin should be commenced to prevent thrombus propagation. Daily assessment
of activated partial thromboplastin time (APTT) ratio is required with a target ratio of 2
(range 1.5–2.5) considered optimal. This should be monitored at least daily and dose adjust-
ments made according to local protocols. Intravenous fluid rehydration may be needed prior
to and following endovascular revascularization to minimise the risk of contrast induced
nephropathy (CIN).
Analgesia
Acute limb ischaemia causes severe pain. Lying flat may exacerbate symptoms and many
patients are only able to tolerate a dependent limb position. Effective analgesia is needed to
relieve this and also to permit the patient to lie still enough to allow surgical or endovascular
procedures performed under local anaesthetic. Regional blocks of the femoral and sciatic
nerve are promising techniques that alleviate ischaemic pain without the need for epidural
or general anaesthetic [3].
Embolectomy
Surgical exposure and control of the common femoral artery can often be performed under
local anaesthetic. Removal of emboli with balloon catheters may effectively restore limb per-
fusion. Associated fasciotomy may be required.
The chances of technical success are greatest with larger calibre arteries and cardiogenic
emboli where the likelihood of an associated stenosis or occlusion is lower. Embolectomy
is effective for supra-inguinal occlusive emboli as absence of a femoral pulse hinders vascu-
lar access for catheter delivered thrombolysis. Selective catheterization and embolectomy
of below-knee vessels may prove impossible without fluoroscopic imaging and guidewire
compatible embolectomy catheters.
Completion angiographic appearances indicate technical success, identify associated
underlying lesions and determine the need for adjunctive procedures such as thrombolysis
or angioplasty.
Following successful embolectomy, anticoagulation decreases the risk of recurrent
embolization, particularly in those with atrial fibrillation. A target international normalised
ratio (INR) of 2.5 (range 2–3) is optimal. If femoral embolectomy is unsuccessful then either
on-table thrombolysis or surgical exploration of the distal popliteal artery or of the anterior
tibial artery and tibio-peroneal trunk may be required.
Thrombolysis
Pharmacological thrombus dissolution is used to treat embolic occlusions or resolve in situ
thrombosis and reveal the underlying stenotic lesions. In acute limb ischaemia thromboly-
sis delivered by an intra-arterial catheter within the thrombus is more effective and car-
ries a lower risk of complications than systemic thrombolysis. Three main agents have been
employed; they exhibit no major differences in efficacy:
• Recombinant tissue plasminogen activator (rTPA) is most widely used in the UK. This
acts by converting plasminogen into plasmin and promoting fibrin degradation.
• Streptokinase induces antibody production, preventing repeat administration. It is
ineffective in patients with prior streptococcal infections and is no longer widely used.
• Urokinase has similar efficacy to rTPA and has been favoured in the USA.
Concurrent administration of low dose heparin through the arterial sheath is used to
prevent pericatheter thrombus formation. Newer classes of thrombolytic agents acting
independently of the plasminogen system (alfimeprase and plasmin) offer the potential for
more rapid lysis coupled with a lower risk of serious bleeding and an improved safety pro-
file. Platelet GP2b/3a receptor antagonists, such as abciximab, produce a synergistic effect
with thrombolytics, resulting in more rapid lysis but this may be at the expense of increased
haemorrhagic complications. The role of these agents in PAD is still to be defined.
The efficacy of all thrombolytic agents diminishes rapidly and most units restrict usage
to within 4–6 weeks of the presumed thrombus formation (usually a major symptomatic
deterioration). Beyond this therapeutic window the low chance of successful lysis is out-
weighed by the relatively high complication rate.
In the acutely threatened ischaemic limb, emergency embolectomy is preferred to throm-

bolysis which can take up to 24 hours to substantially improve limb perfusion. The risks
and benefits of thrombolysis must be carefully considered for each patient. Situations where
thrombolysis is contraindicated include:
• active/recent bleeding;
• previous cerebral haemorrhage;
• surgery, major trauma or cardiopulmonary resuscitation within the preceding 2 weeks;
• proliferative diabetic retinopathy;
• recent ocular surgery or trauma;
• bleeding diathesis;
• pregnancy.
Prerequisites for successful thrombolysis are an informed, co-operative patient in whom
the risk of inadvertent arterial sheath or catheter removal is low (the resulting haemorrhage
may be very difficult to control) and careful supervision and monitoring, usually within a
high dependency environment. Effective analgesia during this period may require intrave-
nous opiates or specialist pain team input.
Several different thrombolysis regimes are described:
• Low dose regimes administer a continuous infusion typically for 24 hours or longer.
• High dose regimes involve an initial bolus followed by a high dose infusion with the
aim of achieving faster thrombolysis during core working hours.
Both strategies are effective and confer similar limb salvage and complication rates. Persistent
thrombotic occlusion resistant to embolectomy is often managed with intraoperative rTPA
thrombolysis. A typical dosage regime is 15 mg diluted to 100 ml and delivered by infusion
or repeated small bolus injections over 30 min.
Thrombolysis carries substantial risks. Thirty day mortality in a large UK based audit
was 12.4% [4], with the majority of deaths attributable to myocardial infarction or stroke.
Stroke occurs in approximately 3% of patients and may be haemorrhagic or thrombotic.
Major arterial haemorrhage, requiring cessation of thrombolysis affects about 10% of
patients. This is mostly from groin puncture sites but there is a significant risk of concealed
retroperitoneal and intra-abdominal haemorrhage. Haemodynamic monitoring, frequent
assessment of the puncture site, limb perfusion and neurological status are required to
promptly detect serious bleeding and to minimise the duration of thrombolysis. Titration
of thrombolytic dose to the results of coagulation studies is necessary to avoid excessive
depletion of fibrinogen.
There should be a high index of suspicion for concealed arterial bleeding and a low
threshold for performing CT examination to assess for cerebral or abdominal haemorrhage.
Minor bleeding from the puncture site is common (20–40%) and can be managed by local
compression and need not require thrombolytic cessation.
Distal embolization may produce a temporary deterioration in limb perfusion or be evi-
dent on check angiography. This usually resolves with continued thrombolysis or thrombus
aspiration.
Surgery versus thrombolysis

Randomized controlled trials demonstrate no clear superiority for surgery or thrombolysis
on 30-day limb salvage rates or mortality [5]. However, randomised trials include only those
suitable for either treatment. Registry data shows that surgery is performed three to five
times more frequently than thrombolysis.
Mechanical thrombectomy
A variety of adjunctive endovascular techniques involving mechanical thrombus aspiration
or thrombectomy aims to promote faster thrombus resolution. They are often combined
with thrombolysis. These add to the cost and complexity of treatment and are not currently
widely used.
Unsalvagable limbs
Even with current techniques a significant proportion of ischaemic limbs cannot be saved.
In these patients efforts must be directed towards pain management and resuscitation with
prompt amputation and rehabilitation. Futile attempts at revascularization pose a major risk
from rhabdomyolysis and hyperkalaemia.
Chronic limb ischaemia

In chronic arterial disease non-invasive imaging techniques including Duplex ultrasound,
magnetic resonance angiography (MRA) and computed tomography angiography (CTA)
have largely replaced catheter angiography for the majority of patients and allow accur-
ate assessment of disease distribution. The haemodynamic significance of stenoses dem-
onstrated by these techniques can be further evaluated with targeted Duplex ultrasound or
translesional pressure gradients obtained with catheter angiography. Typical thresholds for
treatment are a peak systolic velocity ratio (PSVR) of 2.5 for ultrasound and a 5–10 mmHg
pressure gradient without vasodilatation or 10–15 mmHg with vasodilator administration
measured at angiography.
Lesion prioritisation
When confronted with multiple lesions in a patient with chronic limb ischaemia, treatment of
significant proximal lesions is undertaken first. Symptomatic improvement following angio-
plasty or stenting of aorto-iliac lesions frequently exceeds that of femoro-popliteal lesions and
if a good response is achieved subsequent infra-inguinal revascularization may be unneces-
sary. Even when further surgical or endovascular procedures are required the optimization of
arterial inflow increases the likelihood of technical success. It may be unnecessary or undesir-
able, due to severe potential complications, to treat all identified lesions and a staged approach
with clinical review will optimise the risk/benefit balance for individual patients.
Surgical revascularization
Endarterectomy
Areas of bulky heavily calcified plaque respond poorly to angioplasty, which may fail to
fracture the plaque. Stenting is similarly ineffective as stents lack sufficient radial force to
overcome these lesions. For lesions in readily accessible surgical sites, such as the common
femoral artery, endarterectomy is an effective treatment.
Surgical bypass
Infra-inguinal bypasses may be constructed from autologous vein or prosthetic graft such
as woven Dacron or polytetrafluoroethylene (PTFE) There is a major advantage to using
vein grafts with 5-year patency rates of 74–76% versus 39–52% for PTFE [6, 7]. In addition
graft infection is reduced if autologous grafts are used. Most commonly the long saphe-
nous vein (LSV) is used either as a reversed vein graft or an in situ vein graft, following
valvotomy. There is no strong evidence that in situ grafts have better patency rates than
reversed grafts. Narrow calibre and varicose veins are unsuitable and many patients may
have already undergone LSV harvesting for prior lower limb or coronary revasculariza-
tion. In these circumstances short saphenous vein, cephalic vein or superficial femoral
vein are alternatives, with two more separate pieces being joined together if necessary
(spliced vein graft). Preoperative Duplex ultrasound assessment reliably demonstrates
suitable veins and avoids the unnecessary morbidity resulting from exploration of poor
quality veins.
Anastomotic techniques
Venous cuffs or patches performed at anastomoses between prosthetic grafts and small
(distal) arteries improve the patency of femoro-popliteal PTFE grafts performed below the
knee or more distally but do not produce any benefit in above-knee grafts [8].
Aorto-iliac bypass
Surgical bypass with a prosthetic graft from the distal aorta to either the iliac or common
femoral arteries is usually the preferred option for occlusive or diffuse stenotic disease. A
variety of trans-peritoneal or extra-peritoneal approaches may be employed. The patency of
aorto-bifemoral grafts is high (greater than 80% at 5 years).
Femoral-femoral crossover graft

Femoral-femoral crossover operations utilising prosthetic grafts are an alternative to
aorto-iliac or aorto-femoral bypass for unilateral iliac occlusive disease. They are also
employed following aorto-uni-iliac stent graft treatment of abdominal aortic aneu-
rysms. As an intra-abdominal incision is avoided this is a less invasive option better
suited to higher risk patients, although the majority will require either regional or gen-
eral anaesthesia.
Axillo-bifemoral grafts
Axillo-bifemoral prosthetic grafts (Figure 4.1) are used as a less invasive alternative to aorto-
iliac bypass procedures for occlusive disease or to treat aorto-iliac graft infection. Long-term
graft patency rates for extra-anatomic bypasses are lower than those for aorto-bifemoral
grafts and they are usually reserved for patients with critical limb ischaemia.
Femoro-popliteal bypass
Non-invasive imaging will confirm the adequacy of arterial inflow and determine whether
there is sufficient outflow to promote graft patency. The proximal anastomosis is usually
sited anteriorly on the common femoral artery. The distal anastomosis is in the popliteal
artery and may be above or below the knee. Autologous vein grafts are preferred as they
have the best long-term patency rates but prosthetic grafts may be necessary for patients
with insufficient suitable vein. Five-year patency rates for above-knee and below-knee
femoro-popliteal vein grafts are similar (62% and 68%, respectively). Prosthetic grafts
perform much worse below the knee with 5-year patency falling from 43% above knee to
27% below. Many units restrict below-knee prosthetic grafts to patients with critical limb
ischaemia.
Figure 4.1 Computed tomography (CT) scan of

right axillofemoral bypass.
Femoro-distal bypass
The quality of the outflow artery correlates with long-term graft patency rates. The choice
of distal arterial anastomotic site (tibial or peroneal artery) does not appear to affect out-
come and therefore the best quality artery with uninterrupted run off to the foot should be
selected. Five-year assisted patency rates for vein grafts are approximately 60% and less than
35% for prosthetic grafts.
Endovascular techniques
Angioplasty
Angioplasty techniques use pressure controlled balloon inflation to fracture arterial
plaque and remodel the artery. As the plaque is not removed this is most effective for short
focal stenoses without heavy calcification. Excessively calcified lesions are resistant to
angioplasty.
A recent randomised controlled trial demonstrated additional benefit of angioplasty
over a supervised exercise programme and best medical therapy [9]. At 24-months post
angioplasty absolute walking distance was improved by 78% for aorto-iliac lesions and 38%
for femoro-popliteal lesions.
Procedural outline
• atient symptoms, examination and pre-procedural imaging are reviewed.
P
• Arterial access site is planned (ultrasound guidance is a useful adjunct).
• Local anaesthetic infiltration.
• Arterial puncture and guidewire passage (Seldinger technique).
• Arterial sheath placed.
• Angiographic evaluation of lesion.
• Heparin anticoagulation.
• Guidewire passage across lesion (frequently requires a hydrophilic guidewire).
• Exchange for supportive guidewire.
• Angioplasty balloon placement and inflation.1
• Angiography to assess response.
• Repeat inflation or stenting if needed.
• Arterial puncture site compression/closure device.
Subintimal angioplasty differs from the usual intraluminal angioplasty because it is per-
formed in the wall of the artery to create a new non-diseased channel whereas traditional
angioplasty opens the narrowed lumen. This technique can be effective in recanalizing iliac,
superficial femoral artery (SFA) and long below-knee occlusions. It involves intentional
guidewire passage into the subintimal plane to create a channel that extends the length of
the occlusion. The guidewire is then redirected back into the vessel lumen and the entire
section angioplastied to create a new flow channel between the intima and media. Care must
be taken to avoid major collateral vessels as occlusion of these can precipitate acute limb
ischaemia. Heavily calcified vessels may prevent guidewire and catheter passage and it may
be impossible to re-enter the vessel lumen.
Although procedures performed on crural arteries in patients with critical limb ischae-
mia achieve lower technical success rates, compared with those performed above the knee,
this remains a valuable technique as many of this patient group are unfit for surgical revas-
cularization. Long-term patency rates are a less important measure for this group as patient
life expectancy is short and even a temporary increase in limb perfusion may be sufficient to
achieve relief of rest pain, ulcer healing and limb salvage.
Complications of angioplasty
Arterial puncture site haemorrhage
This is potentially life threatening. Risk factors include:
• vessel wall calcification (loss of elastic recoil);
• hypertension;
• number of arterial puncture attempts;
• size of arterial sheath;
• anticoagulation.
Most risk factors can be controlled with manual compression, although this may need to
be prolonged. Concealed retroperitoneal haemorrhage is a life-threatening emergency
Medicines and Healthcare Products Regulatory Agency (MHRA) alert notice from May 2002 mandates
1
use of a calibrated inflation device to minimise the risk of balloon rupture and embolization.
occurring within a few hours of the procedure. Pain, pallor, tachycardia, hypotension,
flank tenderness or visible haemorrhage are all useful indicators of serious bleeding. Early
detection prior to the onset of haemorrhagic shock is the aim and vigilant post-procedural
monitoring in an appropriate environment is mandatory. Any clinical suspicion of serious
bleeding should prompt emergency computed tomography (CT) examination and surgical
exploration and repair if active bleeding is confirmed.
Arterial rupture
Iliac artery rupture is a rare but potentially fatal procedural complication. Female patients
requiring angioplasty of heavily diseased, narrow calibre, external iliac arteries are most
at risk. Severe pain unrelieved by angioplasty balloon deflation indicates likely rupture.
Immediate angiography is required to confirm the site of rupture and then balloon inflation
to tamponade the site is performed.
Urgent cross matching of blood is undertaken and experienced vascular, surgical and
anaesthetic assistance urgently obtained. Once this team is assembled the tamponade bal-
loon may be deflated. Minor ruptures may resolve with balloon tamponade alone but con-
tinued extravasation requires immediate placement of a covered stent placed across the
rupture site. This usually resolves the problem but if not emergency surgery will be required.
Rupture of infra-inguinal arteries seldom causes any major clinical problems.
Dissection
Arterial dissection is a frequent event during angioplasty. Usually this is minor and non-
flow-limiting and no treatment is indicated. Flow restricting dissections may resolve with
prolonged (5–15 min) balloon inflation or placement of an uncovered stent.
Distal embolization
Embolization may be caused by plaque fracture and detachment or by thrombus. Completion
angiography of the distal vascular territory will detect this complication. If limb perfusion
is maintained then further intervention is unnecessary. In cases where the limb is compro-
mised then thrombus aspiration or thrombolysis are often effective but some patients will
still need surgery. Symptomatic plaque embolization is more likely to require surgery.
Contrast induced nephropathy (CIN)

Following iodinated contrast administration there is frequently a transient deterioration
in renal function which is maximal at 48 hours and has generally resolved by 5 days.
Patients with normal pre-procedural renal function are unlikely to suffer clinical problems
but for those with pre-existing renal impairment there may be an irreversible worsening of
renal function. Adequate hydration in the peri-procedural period is essential to minimise
this risk. Pre-treatment with n-acetylcysteine (NAC) may be beneficial. A meta-analysis
showed a moderate reduction in CIN in patients pre-treated with NAC. The use of iso-
osmolar contrast media lessens the degree of renal impairment in high risk patients with
diabetes. The minimum possible contrast volumes consistent with high quality imaging
should be used.
Carbon dioxide (CO2) may be used as an alternative contrast agent to guide angioplasty
in high risk patients. Infra-inguinal angioplasty performed solely with ultrasound guid-
ance is a viable alternative for centres lacking the necessary equipment or experience in CO2
angiography.
Stents
Stents are supportive frameworks that apply radial force to diseased arteries and promote
vessel remodelling. Most are made of stainless steel or nitinol, a nickel-titanium alloy with
the ability to regain its manufactured shape once deployed. Drug eluting stents, dissolvable
and biodegradable stents all offer attractive theoretical benefits but are not part of current
routine practice for peripheral arterial disease.
Stent deployment is by two basic mechanisms:
(1) Balloon expandable stents are securely mounted on a balloon angioplasty catheter by
the manufacturer and deployed by pressure regulated balloon inflation. Advantages
include their greater radial force and reliable accurate placement. Care must be taken
when introducing balloon mounted stents through arterial sheaths and tight stenoses
to prevent stent dislodgement.
(2) Self expanding stents are constrained by a low profile sheath and deploy when the
covering sheath is withdrawn. They apply less radial force than balloon expandable
stents and may require post-dilation. Some earlier versions shortened significantly
upon deployment which hindered precise placement. Current devices are much less
prone to this.
Continuous technological improvements mean that most stents can be delivered through
a sufficiently small sheath to permit percutaneous procedures.
Aortic angioplasty
Infrarenal aortic atheroma and minor stenoses are common and often clinically insignifi-
cant. The small minority of lesions that produce significant haemodynamic effects may be
treated with either aortic angioplasty or uncovered stent placement. Both methods result in
excellent long-term patency rates.
Iliac stenting
Iliac artery angioplasty has high technical success rates and produces 5-year patency rates
of approximately 70%.
Stent placement (Figures 4.2 and 4.3) either as a primary procedure or performed selec-
tively following suboptimal response to angioplasty significantly reduces the failure rate of
endovascular treatment. Although endovascular treatment for aorto-iliac occlusive disease
carries lower risks than surgical revascularization a recent large UK registry recorded a 2%
in-hospital mortality rate, with deaths mainly occurring in patients with critical limb ischae-
mia. Mortality for those with claudication was 0.2%.
SFA stenting
Previous attempts to stent SFA stenoses and occlusions failed. There was a very high inci-
dence of early stent fractures and the majority of these subsequently occluded.
Fundamental stent re-design has produced a generation of devices better suited to the
complex series of forces affecting the SFA. Recent trials suggest a much reduced early stent
fracture rate with substantially improved vessel patency.
Two-year follow-up data comparing primary SFA stenting with a strategy of SFA angio-
plasty and selective stent placement for residual stenosis or dissection showed clinical
Figure 4.2 Angiogram showing a left common iliac Figure 4.3 Left common iliac artery occlusion
artery occlusion. treated with balloon expandable stent placement. A
right-sided stent was also placed to preserve the right
common iliac origin (same patient as Figure 4.2).
improvement and lower re-intervention rates with primary stenting. The current role of SFA
stenting is as a ‘bail out’ procedure for significant residual stenosis or flow limiting dissec-
tion at the time of angioplasty. Longer-term data are needed before the substantial add-
itional costs generated by a primary stenting policy can be justified.
Below-knee stenting
Treatment of short focal lesions of the tibial arteries with either coronary artery stents or
dedicated below-knee stents is described. The majority of critically ischaemic limbs contain
diffuse stenoses and long occlusions involving more than one run off vessel and are therefore
anatomically unsuitable. There is currently very little evidence to support below-knee stent
placement.
Stent grafts
Stent grafts, also referred to as covered stents, consist of a fabric covering (Dacron or PTFE)
applied to a stent. The addition of the fabric covering increases the profile of the device and
stent grafts require larger calibre arterial access than uncovered stents of the same diameter.
Stent grafts allow endovascular treatment of arterial haemorrhage and exclusion of aneu-
rysms whilst maintaining perfusion of the distal vascular bed.
Several case series describing stent graft treatment of long SFA lesions (mainly occlu-
sions) suggest similar patency rates to prosthetic femoro-popliteal bypass procedures.
Great care must be taken to avoid covering the origin of profunda with the stent graft
proximally. The distal extent of any stenting procedure must also be very carefully
considered so that future surgical options are not compromised (i.e. conversion of an
above-knee femoro-popliteal bypass into a below-knee femoro-popliteal or femoro-
tibial bypass). Long-term antiplatelet therapy is recommended to maintain stent graft
patency.
Hybrid procedures
Hybrid procedures typically consist of a surgical bypass combined with either angi-
oplasty or stenting to optimise graft inflow or outflow. The order of these procedures
requires careful consideration and the endovascular component is usually performed
first to avoid compromising the inflow or outflow to a newly constructed graft by bal-
loon inflation or haemostatic manoeuvres. A dedicated endovascular theatre is the opti-
mal environment but satisfactory results can be obtained with a fluoroscopy compatible
theatre table and portable image intensifier with vascular processing software including
digital subtraction.
Radiation protection
Radiation protection legislation requires that all medical diagnostic and therapeutic proce-
dures are justified and that exposures are As Low As Reasonably Achievable (the ALARA
principle). Endovascular operators must be adequately trained in both the principles and
practicalities of radiation protection. There is an individual legal responsibility to comply
with local radiation protection policies and dosimetry monitoring arrangements.
Restenosis
Angioplasty or stent induced plaque fracture and vessel wall trauma induce a complex series
of responses frequently leading to intimal hyperplasia and vessel restenosis. The incidence of
restenosis varies between anatomical sites and small calibre vessels, long stenotic segments
or occlusions are particularly prone.
The use of stent grafts may prevent tissue ingrowth and reduce restenosis rates. A vari-
ety of response modifying techniques aim to reduce or prevent restenosis. External beam
radiotherapy significantly reduces restenosis and re-intervention rates in patients with a
suboptimal response to SFA angioplasty requiring stent placement. Endovascular brachy-
therapy and the use of radioactive stents are also under evaluation.
Graft/stent surveillance
Occlusion of surgical bypass grafts and arterial stents is frequently preceded by progressive
stenosis within or immediately adjacent to the graft or stent.
Retreatment at this stage is difficult and many units undertake graft and stent surveil-
lance programmes in which significant stenoses identified with Duplex ultrasound are
treated with the aim of avoiding graft or stent occlusion.
These programmes are resource intensive. Compliance rates with long-term follow up
are low and patients still present requiring treatment for lesions occurring outside the sur-
veillance zone (i.e. opposite limb).
The majority of re-interventions are required within the first year and it may be more
cost effective to limit surveillance to one year or to stratify follow up according to early sur-
veillance findings.
Upper limb revascularisation

Although less frequently affected than the lower limbs, revascularisation may be needed for
either acute or chronic ischaemia involving the arms.
Surgical embolectomy or catheter delivered thrombolysis are the main options for
embolic occlusions, with severity of ischaemia and local availability and expertise again the
determining factors.
Chronic ischaemia associated with limiting symptoms may be treated by surgical bypass
(carotid-subclavian or carotid-axillary) or by endovascular methods. Reduced upper limb
pulses or asymmetric blood pressure measurements in the absence of limiting symptoms do
not constitute an indication for revascularization.
Trauma from cervical or rudimentary first ribs in thoracic outlet syndrome (TOS)
may cause subclavian arterial stenoses or aneurysm formation and distal embolisation.
Revascularisation with autologous or prosthetic graft is combined with rib resection.
Adjunctive embolectomy or thrombolysis may be needed for embolic presentations.
Additional considerations for upper limb endovascular procedures include
• The need for alternative arterial access sites (usually brachial artery), with associated
risk of serious neurovascular complications (including median nerve compression) and
higher incidence of arterial spasm.
• Manipulations around the aortic arch carry a risk of stroke. This must be explained
to patients. Meticulous technique, avoidance of air bubbles in contrast media and
measures to avoid peri-catheter thrombus formation are mandatory.
• Awareness of the vertebral artery anatomy and careful planning to avoid inadvertent
embolisation or occlusion.
Evidence-based approach
The recently updated Transatlantic Inter-Society Consensus (TASC) Guidelines on the man-
agement of peripheral vascular disease are a collaboration involving international medical,
vascular surgical, interventional radiology and cardiology societies [10]. Their recommen-
dations are graded based upon the level of supporting evidence.
Lesions are stratified according to the results of surgical and endovascular revas
cularisation.
• Type A lesions give very good results with endovascular treatment and this is preferred.
• Type B lesions give good results with endovascular treatment and this is recommended
unless there are adjacent lesions requiring surgical treatment.
• Type C lesions – surgical revascularization produces superior long-term results and
endovascular methods should only be used in patients at high risk for surgery.
• Type D lesions respond poorly to current endovascular methods and surgical
revascularization is preferred for fit patients.
These guidelines emphasise that ‘patient’s comorbidities, fully informed patient prefer-
ence and the local operators long-term success rates must be considered when making treat-
ment recommendations for type B and type C lesions.’
Decision making
The evidence base for these interventions continues to evolve and individual manage-
ment decisions are best considered by a multi-disciplinary team experienced in best med-
ical management, anaesthesia and able to offer the full range of surgical and endovascular
techniques.
References saphenous vein in claudicant patients

with bilateral above knee femoropopliteal
1. Watson L, Ellis B, Leng GC. Exercise bypasses. Surgery 1999; 126: 594–602.
for intermittent claudication. Cochrane 7. Green RM et al. Prosthetic above-knee
Database of Systematic Reviews femoropopliteal bypass grafting: five year
2008, Issue 4. Art. No. CD000990. results of a randomized trial. J Vasc Surg
DOI: 10.1002/14651858.CD000990. 2000; 31: 417–25.
2. Adam DJ, Beard JD, Cleveland T 8. Stonebridge PA, Prescott RJ, Ruckley CV.
and BASIL Trial Participants. Bypass Randomized trial comparing infrainguinal
Versus Angioplasty in Severe Ischaemia polytetrafluoroethylene bypass grafting
of the Leg (BASIL): multicentre, with and without vein interposition
randomised, controlled trial. Lancet 2005; cuff at the distal anastamosis. The Joint
366: 1925–34. Vascular Research Group. J Vasc Surg 1997;
3. Marcus AJ, Lotzof, K, Kamath et al. A new 26: 543–50.
approach: regional nerve blockade for 9. Greenhalgh RM, Belch JJ, Brown LC et
angioplasty of the lower limb. Cardiovasc al. and Mimic Trial Participants. The
Intervent Radiol 2006; 29: 235–40. adjuvant benefit of angioplasty in patients
4. Earnshaw JJ, Whitma B, Foy C on behalf with mild to moderate intermittent
of the Thombolysis Study Group. National claudication (MIMIC) managed by
Audit of Thrombolysis for Acute Limb supervised exercise, smoking cessation
Ischaemia (NATALI) clinical factors advice and best medical therapy: results
associated with early outcome. J Vasc Surg from two randomised trials for stenotic
2004; 39: 1018–25. femoropopliteal and aortoiliac arterial
5. Berridge DC, Kessel DO, Robertson I. disease. Eur J Vasc Endovasc Surg 2008;
Surgery versus thrombolysis for initial 36: 680–8.
management of acute limb ischaemia. 10. Norgren T, Hiatt, L, Dormandy, WR.
Cochrane Database of Systematic Reviews Transatlantic Inter-Society Consensus for
2002, Issue 1. Art. No. CD002784. the Management of Peripheral Arterial
DOI: 10.1002/14651858.CD002784. Disease (TASC II). J Vasc Surg 2007;
6. AbuRahma AF et al. Prospective controlled 45: S5.
study of polytetrafluoroethylene versus
Chapter
5
Abdominal aortic aneurysms
Vish Bhattacharya and Rob Williams
Key points
• An aneurysm is an abnormal focal dilatation of a vessel with a greater than 50%
increase in its diameter
• Elastin degradation due to matrix metalloproteinases (mainly 2.9 and 12) in the aortic
media is the most significant pathology
• Smoking, male gender, hypertension and genetics are the main risk factors
• The annual risk of rupture between 5 and 5.9 cm is 5%
• Contrast enhanced computed tomography (CT) is the gold standard for measuring size
and morphology
• The UK Small Aneurysm Trial (UKSAT) helped reach a consensus about the minimal
size of treatment, which is 5.5 cm
• The UK Endovascular Aneurysm Repair (EVAR) 1 (in fit patients) trial showed that the
30-day mortality for EVAR was 1.7% versus 4.7% for open repair, although by 4 years
all-cause mortality was the same
• The EVAR 2 trial (in unfit patients) showed no difference in mortality between
conservative management and stenting
• Fenestrated grafts allow stenting of juxtarenal aneurysms
• Stenting for ruptured aortic aneurysms has shown promising results and a UK trial is
currently underway
Introduction
Abdominal aortic aneurysm (AAA) was first described around AD 100 by Roman physicians.
In 1923, Rudolf Matas successfully ligated the aorta of a patient with an AAA. The interven-
ing years have seen many advances and improvements in diagnosis and treatment, however
ruptured AAAs still cause around 6000 deaths in England and Wales per annum [1].
Definition
An aneurysm is defined as an abnormal, focal dilatation of a vessel, with greater than a 50%
increase in the diameter. The normal diameter for the abdominal aorta is up to 2 cm, there-
fore 3 cm is considered to be the minimum diameter of an AAA. All three vessel layers,
the intima, media and adventitia, are intact. Aneurysms can be fusiform i.e. a cylindrical
Chapter 5: Abdominal aortic aneurysms 87
dilatation of the whole vessel, or saccular, a focal bulge arising from the side of the ves-
sel. Abdominal aortic aneurysms involve the abdominal aorta and can extend into the iliac
arteries. Ninety per cent are infrarenal, the remaining 10% are para- or suprarenal.
Aetiology
Aneurysmal degeneration occurs most commonly in the elderly and is usually associated with
atherosclerosis. In simplistic terms aging results in weakening of the aortic wall and aneurys-
mal dilation results. According to the Laplace law, luminal dilation results in increased wall
tension and a cycle of progressive dilation and greater wall stress. The most marked patho-
logical changes occur in the media and intima. The media is responsible for the majority
of the tensile strength of the aortic wall. It is made of multiple layers of structural proteins,
the bulk of which are collagen and elastin. Over time elastin fibre fragmentation occurs and
the media degenerates. Collagen makes up about 25% of the wall of an atherosclerotic aorta
and only 6–18% in aneurysms. Matrix metalloproteinase 2, 9 and 12 are found in increased
amounts in the wall of aneurysms and cause degradation of the collagen and elastin matrix,
causing a decreased amount of both in the wall of the aortic aneurysm. The smooth muscle
cells in the media undergo a phenotypical change and are unable to maintain and protect the
matrix. Changes in the adventitial layer are less marked, but inflammatory infiltrates are often
present [2]. Inflammatory aneurysms will be mentioned later in this chapter.
Historically, aortic aneurysms were often referred to as ‘atherosclerotic’. However,
although atherosclerosis and aortic aneurysms share common risk factors and frequently
occur concomitantly, aortic aneurysms primarily are the result of degeneration, which leads
to a loss of aortic wall integrity. Subsequent enlargement and aneurysm formation predis-
poses the aorta to atherosclerosis and further degeneration of the aortic wall. It is likely that
a combination of genetic predisposition, aging and damage to the aortic wall from risk fac-
tors, such as hypertension and smoking, are all involved.
Aneurysms can also be caused by infection (also known as mycotic aneurysms); there is
normally a separate source of infection such as vertebral discitis rather than primary infection
of the aorta. The classical syphilitic aneurysm is now rare. Other causes include trauma, arter-
itis, connective tissue disorders such as Marfan’s, Ehlers–Danlos and Loeys–Dietz syndromes.
These are probably degenerative aneurysms with more rapid progression due to underlying
abnormalities. False aneurysms develop after dissection and lack of an intact intimal layer.
Various risk factors have been identified which can predispose an otherwise healthy indi-
vidual to develop an AAA. The most important of these is tobacco smoking, which causes
an eightfold increase in the risk of developing an aneurysm. There is an important genetic
predisposition, although the exact association(s) has not been described. The familial asso-
ciation is strongest amongst men. Male siblings have an absolute risk of 20–30%. The other
major risk factor is hypertension.
Most research is directed at degenerative aneurysms as these represent the largest and
most important group; however the treatment techniques described will normally apply to
aneurysms with other causes.
Inflammatory AAAs are defined by the presence of a thickened aneurysm wall, marked
peri-aneurysmal and retroperitoneal fibrosis and dense adhesions to adjacent organs. They
represent 3–10% of all AAAs and the triad of abdominal or back pain, weight loss and an
elevated erythrocyte sedimentation rate (ESR) in a patient with an AAA is highly suggest-
ive. The term periaortitis is used to describe the appearance of inflammatory tissue around
the aorta, which enhances with intravenous iodinated contrast media and can therefore be
picked up on CT scanning. The condition overlaps with, and is a variant of, retroperitoneal
fibrosis and the precise aetiology remains elusive.
Operative repair of a large inflammatory AAA can be problematical. The duodenum can
be stuck onto the neck of the AAA in a dense white sheet of inflammatory tissue, which will
need to be separated by sharp dissection into the sac. If there is associated ureteric involve-
ment and hydronephrosis prior ureteric stenting may be required and concomitant ureterol-
ysis can be performed. Traditionally open repair has been preferred to stenting, particularly
with renal involvement, as it has been thought that the inflammation in the retroperitoneum
is more likely to resolve. However, there have been no randomised studies.
Epidemiology
Andominal aortic aneurysms are four times more common in men than women. The mean
age for presentation is between 65 and 70 years, with an incidence of 2–6% in men over the
age of 60. The prevalence of aneurysms increases with age. Caucasian men are more likely to
be affected than other groups, with AAAs rare in Asian and African populations.
Natural history
Most AAAs behave in a relatively predictable manner until they rupture. Most gradually, but
exponentially, increase in size until either the aneurysm ruptures or the patient dies of other
causes. The annual rate of growth in diameter is generally around 10% of the sac diameter.
Higher rates of growth are frequently related to either infected or inflammatory aneurysms
and this is an indication for earlier and more urgent treatment.
The annual risk of rupture is directly related to, and increases with, the aneurysm size.
Below 5 cm the risk is <2%, between 5 and 5.9 cm the annual risk of rupture is 5%; this
increases to 6.6% for aneurysms from 6 to 7 cm and 19% for aneurysms over 7 cm [3]. Only
15% of patients with an AAA will rupture, the remainder will die of other causes.
Presentation
About 70 to 75% of aortic aneurysms are asymptomatic and are found either during routine
physical examination or during imaging investigations for another problem. Aneurysms are
commonly seen on abdominal ultrasound (US), computed tomography (CT) and magnetic
resonance imaging (MRI) studies.
Aneurysms tend to become palpable on abdominal examination when they reach 5 cm
or more in size. This figure varies greatly between patients depending on the size of the
patient and the overlying soft tissue. Slim patients can often palpate their own normal aorta
while obese patients can hide aneurysms much larger than 5 cm. Tumours adjacent to the
aorta, unusual tortuosity and lumbar lordosis can cause the aorta to be pulsatile although
only a true aneurysm will be expansile.
About 20% of patients present with a rupture. The classic triad of a rupture consists of
mid abdominal or diffuse abdominal pain radiating to the back, shock and a pulsatile mass.
The pain can be more severe in the back or flank and radiate into the groin, mimicking renal
colic. Although a catastrophic event, the rupture can be contained in the retro-peritoneum
with the contained haematoma tamponading the leaking aorta allowing time for assessment
and treatment. Free intra-peritoneal ruptures are normally rapidly fatal. It is believed that
only about 25% of patients survive a rupture due to high intraoperative and postoperative
mortality. However this may be an overestimate of survival rate as many elderly patients who
die in the community of undiagnosed ruptured AAAs do not have a post mortem and the
deaths are falsely attributed to cardiac causes.
Imaging
Plain radiographs
Plain radiographs are of limited use for diagnosing or investigating AAAs. Vascular calci-
fication can be seen in some instances allowing the aneurysm to be detected. Occasionally
vertebral body erosion is caused by the pressure effect of a gradually expanding aneurysm.
Plain radiographs are useful in the follow up of patients after endovascular repair of aneu-
rysms as they allow detection of graft migration or graft deterioration as long as a standard
projection is used.
Ultrasound
Many AAAs are found as incidental findings on abdominal US for other problems. With a few
exceptions detection is easy, although obesity and gas within the bowel lumen can obscure the
aorta. Abdominal ultrasound suffers from significant inter-observer errors so technique needs
to be meticulous and consistent within a unit. Due to inherent errors in tranverse measure-
ments on US most AAAs are measured in the antero-posterior (AP) plane. This rarely causes
problems with fusiform aneurysms, but can lead to measurement errors with saccular aneu-
rysms. Abdominal ultrasound is the mainstay of aneurysm surveillance for small aneurysms
as it is quick, relatively cheap and risk free for the patients. Surveillance should generally be
at 6–12-month intervals. Abdominal ultrasound can be used for follow-up surveillance of
patients after endovascular repair, although there is debate over the reliability. Generally the
aneurysm sac diameter is monitored and if concerns are raised then more advanced US tech-
niques using colour Doppler and contrast enhanced US can be used to look for endoleaks.
Ultrasound is unreliable for investigating ruptured aneurysms as the majority of patients who
survive long enough to receive emergency medical care have a retroperitoneal rupture. The
haematoma is often obscured by the bowel and other abdominal viscera.
Computerised tomography
Contrast enhanced CT has become the gold standard for diagnosing aortic aneurysms and
assessing the size and morphology. Dedicated arterial phase images acquired on a multi-slice
spiral scanner, with the slices reconstructed at thicknesses of 3 mm or less provide detailed
information that allows treatment to be tailored to the patient. This is particularly important
for planning endovascular treatment. Measurements are reproducible as long as the imaging
protocol remains the same and detailed assessment of the aneurysm volume is possible,
although unnecessary in the majority of cases. The incidence of non-aneurysm significant
clinical abnormalities is reported to be up to 20%. CT is also the gold standard investigation
for a ruptured aneurysm. Intravenous contrast is not required to visualise the haematoma
but is necessary if the patient is being considered for emergency endovascular repair. Some
clinicians are concerned that imaging introduces an unnecessary delay in treatment, which
may increase the risk of death. Certainly if the patient is unconscious, profoundly hypoten-
sive with strong clinical suspicion of rupture, then imaging in any form is not required.
CT is not without potential problems, the radiation dose is equivalent to around 5 years
of normal background radiation and this carries a small but appreciable risk of causing
malignancy. Repeated imaging increases this risk. The iodinated contrast media may cause
anaphylactoid reactions and can precipitate acute renal failure particularly in diabetics with
underlying renal disease. If contrast must be given then adequate pre-hydration is the most
important step to reduce the effects.
Magnetic resonance imaging

MRI can, like CT, provide detailed information about the aneurysm. Magnetic resonance
angiography protocols require intravenous (gadolinium based) contrast media to provide
high quality information and care needs to be taken with interpretation as some sequences
only display the vessel/aneurysm lumen and may underestimate the true size of the vessel.
The resolution is generally less than that of CT and MRI is not commonly used for treatment
planning unless there is a contraindication to contrast enhanced CT. MRI is susceptible to
artefacts from metals (particularly ferro-magnetic) and pacemakers; some metallic implants
are absolute contraindications to scanning. Gadolinium based contrast media should be
used with care in patients with impaired renal function due to the risk of nephrogenic sys-
temic fibrosis.
Intra-arterial angiography
Calibrated invasive angiography was the gold standard investigation used to plan endovas-
cular repair. It has been superseded by CT scanning due to both the quality of information
obtained and the inherent risks of the procedure. Intra-arterial angiography demonstrates
the lumen of the aneurysm rather than the true diameter.
Screening for AAA

A Cochrane review has identified four controlled trials involving 127 891 men and 9342
women who were randomly assigned to aortic aneurysm screening using ultrasound or no
screening [4]. Only one trial included women. Two of the trials were conducted in the UK,
one in Denmark and one in Australia. The results provide evidence of a benefit from screen-
ing in men with a strongly significant reduction in deaths from AAA. The odds ratio (OR)
for death was 0.60 (range 0.47 to 0.78, three trials) in men aged 65 to 83 years but was not
reduced for women. From one trial there was also a decreased incidence of ruptured aneu-
rysm in men but not women. Screening with ultrasound combined with elective treatment
reduces mortality by 42% in the age group 65–74.
The UK screening programme has started and is being extended nationally. Men will be
offered a single ultrasound scan at 65. Men over 65 will not be screened but will be offered
a scan if they request one. Patients with an AAA will be offered either surveillance or treat-
ment if the AAA is 5.5 cm or greater. It is envisaged that there will be about 70 screening
centres covering the UK population of 60.2 million with each centre covering 800 000 to
1.2 million.
Treatment
The aim of treatment for AAA is to prevent death from rupture of the aneurysm. Surgery
in a variety of forms has been the mainstay of treatment for many years. Early treatment
involved ligating the aneurysm. Subsequently aneurysms were wrapped in various mater-
ials to stimulate fibrosis with only limited success. Albert Einstein had his AAA treated
using this method in 1949 but subsequently died 5 years later when his aneurysm ruptured.
Surgery should be reserved for patients where the risk of rupture exceeds the risk of surgery,
for all others conservative management with aneurysm surveillance and best medical ther-
apy is indicated.
Aneurysm repair has a significant surgical mortality and small aneurysms have a
low risk of rupture: the risk of treatment can outweigh the benefits. Two landmark trials
(The UK Small Aneurysm Trial [UKSAT] [5] and the Aneurysm Detection and Mortality
[ADAM] [6]) have helped reach consensus about the minimum size of aneurysm at which
benefit of treatment outweighs the risks of surgical repair. The UKSAT randomized 1090
patients in 93 centres. About half received early surgery and the other half were under ultra-
sound surveillance between September 1991 and November 1995.
Overall the 30-day operative mortality was 5.5% in the early surgery group and 7.2% in
the surveillance group where surgery took place when it expanded to 5.5 or at 1 cm per year
or ruptured.
After 12 years mortality was 64% in the early surgery group versus 67% in the surveil-
lance group. Sixty percent of all deaths were due to cardiovascular disease. A comparison
with age- and sex-matched population showed a higher risk of death even after successful
open repair in small aneurysms. There is evidence that patients with peripheral vascular
disease including aortic aneurysms are under-treated with respect to cardiovascular risk
reduction, even though the presence of an AAA carries a similar risk as an established diag-
nosis of coronary heart disease.
Similarly the US ADAM Study showed a lower mortality of 2.7% in the elective surgery
group but also did not demonstrate a better survival in the early surgery group [6].
Elective aneurysm repair carries a significant risk of morbidity and mortality. The 30-day
mortality from elective AAA repair in the UK currently stands at 7%. This is significantly
higher than results reported by many other countries in Europe; however few countries have
similarly robust methods of data collection.
Patient selection is critical so preoperative assessment and work-up is important. All
patients should undergo standard assessment and risk scoring, including assessment of car-
diac, respiratory, renal, diabetes and peripheral vascular disease. Preoperative assessment
by a consultant vascular anaesthetist is recommended, normally with cardio-pulmonary
exercise testing. All patients should be discussed in a multi-disciplinary environment. CT
scanning should be used to assess suitability for endovascular repair if considered to be
appropriate for the patient.
Open aneurysm repair

In 1951, Dubost performed the first successful open surgical repair of an AAA with a hom-
ograft. Open repair has evolved over the last six decades but the basic principles remain the
same.
Blood should be cross-matched preoperatively, prophylactic antibiotics administered,
nasogastric (NG) tube and bladder catheters inserted, large calibre venous access estab-
lished. Both central venous pressure monitoring and invasive arterial monitoring are nor-
mally used. Normal body temperature is important for coagulation and metabolic function
so all intravenous fluids should be warmed and the patient covered with a forced-air warm-
ing blanket.
For most infrarenal aneurysms the aneurysm is approached via an anterior midline
abdominal approach. A retroperitoneal approach can be used, most commonly from the
left for juxtarenal or suprarenal repairs as it allows better access to the suprarenal aorta.
Shortly before proximal aortic control is obtained by clamping 5000 IU of heparin should be
administered. Where possible the clamp should be placed on the infrarenal aorta to main-
tain kidney perfusion; this requires the identification of the left renal vein. Five per cent of
individuals have a retro-aortic left renal vein, which can be vulnerable to iatrogenic trauma
if not appreciated. Clamping of the suprarenal aorta increases the risk of both renal ischae-
mia and renal artery embolus. Suprarenal control will normally require division of the left
renal vein. Supracoeliac control requires separation of the diaphragmatic crura and retrac-
tion of the left lobe of the liver. There is an increased risk of bowel ischaemia. The inferior
mesenteric artery is sacrificed, although it is often already occluded due to thrombus in the
aneurysm sac. Collateral supply to the distal large bowel from the internal iliac arteries may
be important, so these vessels should be preserved where possible.
When proximal and distal control of the aorta has been obtained the aneurysm sac is
opened and the aorta reconstructed using a graft. The grafts are normally made of Dacron
or polytetrafluoroethane (PTFE). They may be soaked or impregnated with antibiotics or
silver to reduce the risk of infection. The choice of graft configuration depends on the extent
of the aneurysm. The most common grafts are either straight tubes or aorto-iliac bifurcated
grafts.
Initial postoperative care normally requires intensive care facilities. Careful monitoring
of fluid balance is needed and the fluid requirements in the first 24 hours can be high. The
major risks in the early stages of recovery are of myocardial, visceral and limb ischaemia and
infarction and renal failure related to embolic complications and hypotension and pneumo-
nia. Longer-term complications include aortic pseudoaneurysms, wound and graft material
infection, incisional hernias and erectile dysfunction in men. Most patients are discharged
from hospital after a week.
Open repair has proved to be durable with a low rate of secondary interventions,
although when infection or pseudo-aneurysms occur the mortality rate of secondary inter-
ventions is high.
Endovascular aneurysm repair (EVAR)

Endovascular repair has developed in response to the significant mortality and morbidity
related to elective open repair. The rationale behind endovascular repair is the same as open
repair, i.e. to prevent death from aneurysm rupture by excluding the aneurysm sac from
the direct systemic circulation. Like open repair the endovascular technique involves using
tubular fabric grafts to exclude the aneurysm sac. The major difference is the route by which
the graft is inserted and secured.
In the late 1970s, Juan Parodi first proposed the concept of using an endovascular route
to insert fabric grafts reinforced with metal stents. After more than a decade of engineering
developments and animal experiments the first human implant was performed in 1990 on
a patient unfit for open repair. The patient was a close friend of the president of Argentina;
Parodi, who is Argentinean, was directly approached by the president and persuaded to
perform the operation. The operation was a success and the patient lived for 9 years before
dying from pancreatic carcinoma.
The early grafts were simple fabric tubes, reinforced with stents, which relied on radial
force to maintain their position within the aorta. These proved vulnerable to migration and
sometimes loss of the seal with the aortic wall. Over the next 29 years the grafts have evolved
significantly. Most grafts are now modular bifurcated tubes with some form of active fixation
at the proximal end to reduce the incidence of migration. The active fixation normally con-
sists of a bare metal stent that extends into the suprarenal aorta or barbs/hooks that embed
into the aortic wall. Some grafts use a combination of the two methods. The only other
mechanism to prevent distal migration involves resting the flow-divider of the stent graft
directly onto the aortic bifurcation. This requires a dedicated stent graft.
Preoperative work-up is generally identical to that required for open repair as rarely an
endovascular repair may need to be converted to an open repair during the procedure. All
patients considered for endovascular repair should have a planning CT scan to accurately
assess the aortic morphology. The scan needs to be performed with contrast and timed
for the arterial phase. The images need to be reconstructed at a maximum of 3 mm axial
slices or 1.5 mm for fenestrated/branched grafts. The scan should extend from the level
of the diaphragms to the groins. Around 25% of all pre-EVAR CT scans will have a sig-
nificant unsuspected abnormality so they should all be formally reported by a radiologist.
Planning for EVAR requires an understanding of how the images are acquired and how
to make accurate measurements from the study. Ideally three-dimensional reconstruc-
tions should be performed to allow true axial measurements (related to the vessel, not
the patient) and midline lumen measurements to be made. Accurate measurements of the
vessel diameters are absolutely critical to the success of the procedure. Most manufactur-
ers’ stent grafts require measurements from the adventitial layer to the adventitial layer
however some require intima–intima measurements. Accuracy is governed by both the
operator and the patient. There is a variation of diameter of around 1 to 2 mm, depending
on the phase of cardiac cycle when the scan was acquired. Length measurements do not
need to be as accurate due to significant and unpredictable variability in the conformation
of the grafts between patients. Fortunately the position of the stent graft can be adapted
during the procedure.
Aortic morphology is perhaps the biggest limiting factor for EVAR. Each of the com-
mercially available standard infrarenal stent grafts has different instructions for use (IFUs).
Most grafts require a portion of infrarenal aorta measuring at least 10–15 mm long with a
diameter of less than 30 mm to provide a proximal sealing zone. This is referred to as the
proximal neck. This portion of vessel should be angulated relative to the aneurysm by less
than 60°. The proximal neck should be relatively free of atheroma/thrombus. Both common
iliac arteries should have relatively normal portions for the distal sealing zone, although the
grafts can be extended into the external iliac arteries if required. The external iliac and com-
mon femoral arteries need to be large enough to allow the delivery system to pass through
into the aorta. Most grafts require diameters of 7–8 mm. Iliac tortuosity alone tends not
to cause major problems however extensive calcification (particularly in association with
tortuosity) can cause major difficulties. These problems can normally be overcome and in
extreme cases direct exposure of the iliac arteries can be useful. Strictly following the IFUs
of the currently available, commercially produced stent grafts allows approximately 40–50%
of all infrarenal AAAs to be treated with EVAR. Judicious use of more advanced techniques
and a variety of stent grafts has meant that some enthusiastic high volume UK centres suc-
cessfully treat 80–90% of AAAs using standard grafts and most of the remainder can be
offered fenestrated or branched stent grafts if appropriate, although the evidence to support
this approach is currently limited.
EVAR should generally take place in a theatre environment to reduce the infection risk to
the minimum possible. Fortunately infection of the graft is very rare but it is a catastrophic
complication. The quality of the fluoroscopic units used is critical to the success of the pro-
cedure. Ideally fixed high-specification angiographic fluoroscopy units should be used; how-
ever these are rarely found in a theatre environment and most centres use a mobile C-arm
with angiographic capabilities and a high heat capacity to allow for prolonged screening.
General purpose mobile fluoroscopic units are not suitable.
The procedure can take place under general, regional or local anaesthesia. The latter is
uncommon as the patient may experience significant discomfort due to lower limb ischaemia
while the delivery system occludes the femoral arteries. Most procedures start with surgical
exposure of both common femoral arteries. Guidewires are passed into the thoracic aorta.
An angiogram is performed to locate the renal arteries. The main body is deployed with the
first covered stent placed as close to the lowest renal artery as possible. If a bifurcated device
is used the contralateral limb is cannulated from the opposite groin and both stent graft
limbs are extended down to the level of the iliac bifurcations. The proximal and distal seals
and the junctions of the modular components can be balloon moulded to improve the seal.
A completion angiogram is performed to confirm the patency of the renal and iliac vessels
and to identify endoleaks.
Endoleaks
The term endoleak was coined to describe blood flow in the aneurysm sac despite the
presence of a stent graft. An endoleak does not mean that the aneurysm itself has leaked.
Endoleaks are classified by their source.
Type I
A type I endoleak is a failure of the seal between the stent graft and the artery wall. They
are divided into proximal – where there is a leak between the main body of the stent graft
and the infrarenal aortic neck and distal – where there is a leak between the stent graft
limb and the iliac artery. The blood leaking into the sac is at systolic arterial pressure and
results in a significant increase in the intra-sac pressure. The patient remains at signifi-
cant risk of aneurysm rupture (some clinicians argue that the risk is higher than if the
aneurysm had remained untreated) and the leaks must be treated at the earliest possible
opportunity.
Treatment is normally endovascular. The simplest treatment involves using a moulding
balloon to improve the contact between the stent graft and the arterial wall; this is com-
monly employed during the initial implantation procedure if a type I endoleak is seen on
angiography. If the stent graft is not appropriately positioned in relation to the renal or iliac
arteries it can be extended using additional pieces. If the stent graft is appropriately posi-
tioned but the leak persists then additional bare metal stents can be used to increase the
radial force and improve the apposition to the arterial wall. A few papers have reported
using coils or endovascular glue (either cyano-acrylate or ethylenevinylalcohol copolymer)
to seal the leak, although there is scepticism among many clinicians as to the efficacy of this
approach. Should these treatments fail the options involve surgery; either placing a band
around the outside of the aneurysm neck to provide support for the stent graft. This can be
done laparoscopically and does not involve opening the aneurysm sac. Otherwise the stent
graft can be explanted and the aneurysm repaired in a conventional open fashion. This car-
ries a significantly higher mortality than standard open repair and should not be undertaken
unless all other options have been considered.
Type II
The infrarenal aorta supplies the lumbar arteries and the inferior mesenteric artery. In
many AAAs these are occluded by atheroma; however they often remain patent. A type II
endoleak is filling of the aneurysm sac from these side branches with blood flowing in a
retrograde direction. They are often complex with blood flowing in through one branch and
out through another. They are independent of the graft used and around 20% of patients will
have a type II leak. They can be subtle and better quality imaging will discover more leaks.
The majority are benign as they do not pressurise the sac to the same degree as a type I leak
and around 60% will spontaneously thrombose by one year. A small proportion of all type II
leaks will cause persistent sac expansion and these require treatment. Aneurysm rupture due
to persistent type II endoleak is rare.
Treatment is normally endovascular. These endoleaks are analogous to arteriovenous
malformations (AVMs) with inflow vessels, a central nidus and outflow vessels. Treatment
should be aimed at occluding the cavity within the sac. This can be accessed via an endovas-
cular route using microcatheters (often via the superior mesenteric artery [SMA]-marginal
artery-inferior mesenteric artery [IMA]-aneurysm) or via a direct trans-lumbar route with
CT guidance. The cavity can be occluded with a variety of embolic materials, taking care to
avoid non-target embolisation of the outflow vessels. Simply occluding inflow or outflow
vessels will lead to a recurrence rate of over 50%.
Type III
These leaks are due to either fabric defects in the graft material or dislocation/separation of
the stent graft components. The former are normally due to manufacturing defects and are
fortunately very rare, they present at the time of implant. The latter tend to occur much later as
the aneurysm remodels and distorts the stent graft. The earlier generations of grafts were much
more vulnerable to this phenomenon than the current grafts. Whatever the cause the implica-
tions are similar to type I endoleaks and they must be treated at the earliest safe opportunity.
Treatment is either endovascular, by relining the original graft with new component(s),
or by surgical explantation.
Type IV
Immediately after implantation some graft material is porous, particularly while the patient is
anticoagulated with heparin. These leaks are visible as a contrast ‘blush’ in the aneurysm sac
on the angiogram. They are of no clinical consequence and generally resolve spontaneously.
Type V
Otherwise known as endotension, this is believed to be transmission of arterial pulse pres-
sure into the aneurysm sac without a true leak of blood or fluid. This concept is controversial
and many clinicians believe that endotension is in reality an endoleak of another type that is
subtle and difficult to identify.
Follow up post EVAR

Current management of endovascular repair demands long-term follow up of the patients.
This is due to the need for relatively frequent re-intervention (10–20%) to confirm and
maintain graft integrity. Initial practice was to use interval CT scans to monitor the aneu-
rysm sac size and to look for new endoleaks and plain X-rays to monitor the graft position
and integrity. Many centres are now using ultrasound to monitor the sac size and are only
using CT scanning if an increase in the aneurysm sac is recorded. Re-intervention is gener-
ally only indicated if the aneurysm sac size is increasing, the graft is migrating or if a type I
or a type III endoleak is identified. Fortunately the newer generations of stent grafts appear
to be much more robust than the previous models and the rates of graft migration and dis-
integration are significantly lower. The requirement for long-term follow up is one of the
major reasons for the increased costs associated with EVAR compared to open repair. As
ultrasound increasingly replaces CT and the grafts become more durable this cost difference
decreases.
EVAR versus open repair

Initially it was hoped that endovascular techniques would allow patients who were medically
unfit for open repair to be treated with less risk. Various trials have taken place to compare
the outcomes of EVAR with open repair and with conservative management for unruptured
aneurysms. The most important trials in the UK were the EVAR I and II trials.
EVAR I compared open repair with endovascular repair. It demonstrated that the 30-day
mortality for EVAR was less than half that of open repair (1.7% versus 4.7%) [7]. Patients
also recovered more quickly and were discharged from hospital much earlier. However
by 4 years all cause mortality was virtually identical (~28%) suggesting that endovascular
repair offers no long-term survival benefit. The EVAR group also had a significantly higher
re-intervention rate (20% versus 6%), however the majority of these re-interventions were
relatively benign.
EVAR II compared EVAR with conservative management for patients unfit for open
repair. It demonstrated no significant differences in all-cause mortality between the treat-
ment groups. By 4 years 64% of the patients had died [8]. The trial has been criticised for
several reasons. The definition of unfit for open repair was loose and varied between recruit-
ing centres. The crossover rate from conservative management to EVAR was high and these
patients generally did well but were still analysed under intention to treat rules, thereby
improving the outcome for the conservative management group. However, the conclusion
seems to be that these patients are generally at high risk of death from many causes, whether
or not the aneurysm is treated and intervention of any sort should only be offered after very
careful consideration and after the patient’s fitness has been optimised.
Open repair should still be considered the gold standard treatment for younger, fit
patients because of the increased rate of secondary intervention with EVAR and the require-
ment for long-term follow up.
Fenestrated and branched grafts

Conventional endovascular repair is often limited by the aneurysm morphology. The most
difficult problem to overcome is the length of the proximal neck.
Fenestrated grafts were developed to allow endovascular repair of AAAs with little or no
normal aorta below the level of the renal arteries [9]. At their simplest the grafts have holes
(fenestrations) that correspond with the positions of the renal arteries and/or the SMA to
allow the covered portion of the stent graft to be placed more proximally. These are rein-
forced with either small bare metal stents or covered stents, which extend into the renal
arteries. This helps to prevent migration and maintain alignment between the fenestrations
and the renal arteries. The grafts are custom made for each patient and require very accurate
planning.
Branched grafts come in two different forms. Iliac branch grafts are relatively simple
bifurcated grafts that extend the limb on a standard stent graft. One branch sits in the internal
iliac artery (IIA)and the other in the EIA. This allows treatment of aneurysms involving the
distal common iliac artery (CIA) while maintaining patency of the IIA. The other type of
branched graft is used to treat type intravenous aneurysms that involve the visceral segment
of the aorta. These grafts have multiple branches for the visceral vessels. These grafts are in
the early stages of development and are reserved for high volume, specialist canters.
Hybrid repairs
Type IV aneurysms can be treated by open repair, branched stent grafts or hybrid repairs.
Hybrid repair involves extra-anatomical bypass to the visceral vessels; generally from the
iliac arteries but occasionally from the ascending aorta. The origins of the visceral vessels are
ligated and the aneurysm then repaired with a stent graft that covers the visceral segment of
the aorta. The procedure can be performed in a single episode or staged. The complication
rate can be high, particularly as the consequences of occlusion of the bypass grafts can be
catastrophic.
Pharmacological management
No drug therapy has yet demonstrated a reduction in aneurysm size or mortality in vivo how-
ever various trails are ongoing. Doxycycline has been shown to down regulate the expression
of metalloproteinase 9 in the aneurysm wall in humans [10]. Statins reduce aneurysm for-
mation by their anti-inflammatory effect and this has been proven in human experiments.
Curcumin has been shown to suppress the development of experimental aneurysms and
reduce the expression of inflammatory cytokines, chemokines and proteinases that mediate
aneurysmal degeneration [11].
Treatment of ruptured AAAs

Despite the development of EVAR, open repair remains the normal treatment for ruptured
AAAs in most centres. The 30-day mortality for patients who get to the operating theatre
alive is between 40–50%. Deaths are usually due to the physiological insult of hypotension/
ischaemia to the heart, kidneys and other organs. Most patients require a prolonged period
of intensive care. Endovascular aneurysm repair for ruptured AAAs is complicated by the
need for planning and sizing of the stent graft. This inevitably takes time and delays surgery.
Some patients will not be suitable for EVAR and will have had an unnecessary wait. However
many clinicians feel that patients who are not stable enough to have a CT scan are highly
unlikely to survive surgery of any sort and therefore the delay is acceptable. To provide an
endovascular service requires provision of 24-hour CT scanning, 24-hour access to fluoros-
copy, a large stock of stent graft sizes and, most critically, trained staff (either interventional
radiologists or vascular surgeons) capable of performing the surgery. The most obvious
benefit of endovascular repair is the ability to repair the aneurysm without opening the
abdomen and de-tamponading the rupture. Control of the bleeding can normally be gained
very rapidly by inflating a balloon in the descending thoracic aorta at the expense of renal
ischaemia. The aneurysm can then be repaired in a relatively controlled manner. However,
the patient may still require a laparotomy to treat abdominal compartment syndrome. There
is little good evidence that EVAR for ruptured AAAs reduces the 30-day mortality. Some
registries and single centre series report a benefit but no randomised clinical trial has been
conducted to date [12]. However a UK trial is in the final stages of planning.
Conclusion
Abdominal aortic aneurysms still cause a significant number of deaths due to either rup-
ture or complications of treatment. However recent advances raise the possibility that these
deaths may soon be avoidable. Screening followed by endovascular repair when aneurysms
are relatively small and easy to treat will hopefully virtually eliminate ruptures and peri-
operative deaths. In the longer term, promising advances in pharmacological therapy may
prevent aneurysm growth and even allow healing and regression.
References 7. The EVAR Trial Participants. Endovascular

aneurysm repair and outcome in patients
1. Office of Population Census and unfit for open repair of abdominal aortic
Surveys. 2000 Mortality Statistics. aneurysm (EVAR 2): randomized control
London: Stationary Office; 2001. trial. Lancet 2005; 365: 2187–92.
2. Abdul-Hussien H, Soekhoe RG et al. 8. Verhoeven EL, Muhs BE et al. Fenestrated
Collagen degradation in the abdominal and branched stent-grafting after previous
aneurysm: a conspiracy of matrix surgery provides a good alternative to open
metalloproteinase and cysteine collagenases. redo surgery. Eur J Vasc Endovasc Surg
Am J Pathol 2007; 170: 809–17. 2007; 33: 84–90.
3. Szilagyi DE, Elliot JP, Smith RF. Clinical 9. Mosorin M, Juvonen J, Biancari F et al.
fate of the patient with asymptomatic Use of Doxycycline to decrease the
abdominal aortic aneurysm and unfit growth of abdominal aortic aneurysms: a
for surgical treatment. Arch Surg 1972; randomized double blind placebo
104: 600–6. controlled pilot study. J Vasc Surg 2001;
4. Cosford PA, Leng GC. Screening for 34: 606–10.
abdominal aortic aneurysm. Cochrane 10. Parodi FE, Mao D, Ennis TL, Pagano MB,
Database of Systematic Reviews Thompson RW. Oral administration of
2007, Issue 2. Art. no. CD002945. diferulolylmethane (curcumin) suppresses
DOI: 10.1002/14651858.CD002945. proinflammatory cytokines and destructive
5. UK Small Aneurysm Trial Participants. connective tissue remodelling in
Long term outcome of immediate repair experimental abdominal aortic aneurysms.
compared with surveillance for small Ann Vasc Surg 2006; 20: 360–8.
abdominal aortic aneurysm. N Engl J of 11. Hinchliffe RJ, Bruijstens L et al. A
Medicine 2002; 346: 1437–44. randomized trial of endovascular and
6. The EVAR Trial Participants. Endovascular open surgery for ruptured abdominal
aneurysm repair versus open repair in aortic aneurysm – results of a pilot study
patients with abdominal aortic aneurysm and lessons learned for future studies. Eur
(EVAR trial 1): randomized control trial. J Vasc Endovasc Surg 2006; 32: 506–13;
Lancet 2005; 365: 2179–86. discussion 514–15.
Chapter
6
Thoracic, thoracoabdominal and
suprarenal aortic aneurysms
Jenny Richards and Rod Chalmers
Key points
• A neurysm disease affects the thoracic aorta much less commonly than the infrarenal
aorta
• Most patients are asymptomatic but may have chest, back or abdominal pain
• Diagnosis is on computed tomography (CT) angiogram and the Crawford classification
is used to describe the extent of the aneurysm
• Intervention is recommended if the aneurysm is >6 cm (or 5 cm in the presence of a
connective tissue disorder or a family history of rupture)
• Open surgery remains the standard treatment for complex suprarenal, thoracic and
thoracoabdominal aneurysms
• Specific complications incude risks of paraplegia and renal and visceral ischaemia
• Endovascular repair is suitable for descending thoracic aneurysms
Epidemiology
The incidence of aneurysmal disease affecting the suprarenal aorta and the thoracic aorta
is difficult to estimate since it is a condition that is usually asymptomatic and rupture,
which is often rapidly fatal, is commonly misattributed to other causes such as myocar-
dial infarction or pulmonary embolism. The incidence is approximately 6 per 100 000
people per year. Males are affected about twice as commonly as females but interestingly
the gender difference is less than for infrarenal aortic aneurysms, where the ratio is up
to 7:1.
Natural history data are scarce, typically involving retrospective observation of patients
not fit for open repair. In a series of 1600 cases the rupture risk for a 6 cm aneurysm of
the thoracic aorta was 3.6% per year and for the composite endpoint of rupture/dissection/
death the annual risk was 14.1%.
Aetiology and pathophysiology

The causes of descending thoracic aortic aneurysm (DTAA) and thoracoabdominal aortic
aneurysm (TAAA) overlap significantly with those of infrarenal aortic aneurysm (described
in Chapter 5). While atherosclerotic disease is still the most common cause of thoracic aneu-
rysms other causes, including chronic dissection and connective tissue disease (e.g. Marfan’s
syndrome) as well as infection, trauma and arteritis, account for a higher percentage of cases
of thoracic than abdominal aneurysms. Aneurysmal tissue is characterised by deficiency of the
tunica media of the vessel wall. Neovascularisation, inflammation and proteolytic digestion of
the extracellular matrix are important processes, and biomechanical factors are also involved.
Presentation
The majority of DTAA and TAAA are asymptomatic. Those presenting to the vascular sur-
geon have either ruptured or have been detected as an incidental finding during an investi-
gation for another indication. Patients may complain of chest, back or abdominal pain and
if the aneurysm extends into the abdominal aorta may be detectable as an expansile mass on
examination. Distal emoblisation can occur and usually manifests as ‘trashing’ of the lower
limbs giving rise to the so called ‘blue toe syndrome’. Other symptoms such as hoarseness,
stridor, haemoptysis, haematemesis, nerve root pain and dysphagia aortica can result from
compression or erosion of surrounding structures. The small proportion of aneurysms that
are inflammatory can cause ureteric, bowel or venous obstruction, and occasionally a low-
grade temperature and weight loss. Patients with a ruptured aneurysm will have sudden
onset of severe chest, back or abdominal pain with syncope or profound cardiovascular col-
lapse, and mottling of the lower limbs in particular. A retroperitoneal haematoma may be
evident in a TAAA and superficial bruising is sometimes seen if the diagnosis is delayed.
Imaging and classification

A CT angiogram of the entire thoracic and abdominal aorta is required to define the extent
of the aneurysm and its relationship to the key aortic branches. CT is also the investigation
of choice for suspected rupture.
DTAAs commence beyond the left subclavian artery and terminate above the level of
the diaphragm. TAAAs affect a variable length of the thoracic and abdominal aorta and are
described according to the Crawford classification (Figure 6.1).
This practical classification system is useful for planning the surgical approach and for
predicting complications. For example a laparotomy is often appropriate for type IV TAAA
while an extensive thoracolaparotomy is required for aneurysms extending into the chest.
Regarding complications, aneurysms affecting the visceral aorta carry a higher risk of renal
and visceral ischaemia, and the risk of paraplegia is highest for the most extensive aneu-
rysms and lower for those starting below the level of T6.
Figure 6.1 Modified

Crawford classification of
thoracoabdominal aneurysms.
Safi HJ et al. Evolution of
risk for neurologic deficit
after descending and
thoracoabdominal aortic
repair. Ann Thorac Surg 2005;
80: 2173–9; discussion 2179.
Chapter 6: Aortic aneurysms 101
Management
Based on what is known of the natural history of TAAA and DTAA, intervention is gener-
ally considered in medically fit patients when the descending thoracic component reaches
6 cm or the ascending aorta reaches 5 cm. A lower threshold of 5 cm may also be considered
appropriate in patients with Marfan’s syndrome or a familial tendency to rupture. Open
surgical management of DTAA and TAAA is a major undertaking which, even with recent
advances in perioperative care and surgical technique, is associated with a significant mor-
bidity and mortality. There is an emerging and increasing role for the use of endovascular
techniques in certain patients, which can reduce the surgical insult. Comprehensive preop-
erative assessment is essential.
Management of DTAA and TAAA disease is complex, but the frequency of cases is rela-
tively low. Centralisation of services is essential and there is evidence that outcome is signifi-
cantly better when patients are managed by high volume surgeons in high volume centres
with dedicated vascular anaesthetists.
Non-operative management
Patients with an aneurysm below the diameter-threshold for intervention should remain
in a surveillance programme with CT scans at appropriate intervals. These patients should
receive standard cardiovascular best medical treatment including an antiplatelet agent and a
statin (regardless of cholesterol) unless contraindicated, with careful control of blood pres-
sure and blood sugar levels in diabetic patients. Patients who are not fit for surgical interven-
tion can generally be discharged from the surveillance programme and be treated with best
medical therapy in the community.
Preoperative assessment
The extent of the aneurysm and the relationship to key aortic branches is determined using
CT scanning, which may also reveal surgically important anomalies such as a retro-aortic
left renal vein, multiple renal arteries or a horseshoe kidney. From this information the
nature of the operation and the risk of important complications can be gauged. The other
side of the equation, how fit the patient is to withstand surgical intervention, is evaluated on
the basis of a number of investigations.
It is important to obtain a full medical history and perform a thorough physical exam-
ination. In the assessment of cardiac function, the information available from a standard
electrocardiogram (ECG) and echocardiogram is supplemented by some form of stress
testing, either an exercise tolerance test or a stress echocardiogram. While an exercise
bicycle or treadmill may be used for exercise testing, a treadmill is generally preferred
since most patients are unused to cycling and quadriceps fatigue may limit exercise cap-
acity. With continuous ECG monitoring the Bruce protocol involves up to 12 min of walk-
ing on a treadmill during which the speed and elevation are increased every 3 min, aiming
to achieve 85% of the patient’s maximum predicted heart rate. A positive test involves
the development of angina-related symptoms, a fall in blood pressure or ECG changes
including ventricular dysrhythmias or ST changes. Stress echocardiography may be pre-
ferred in patients with mobility difficulties and where motion affects data interpretation.
During this examination echocardiography is performed during infusion of a pharmaco-
logical agent such as dobutamine, which has positive inotropic and chronotropic effects.
The focus of this test is the detection of wall motion abnormalities and assessment of their
relationship to cardiac exertion, and is a sensitive way to detect stress-induced myocardial
ischaemia. Cardiac catheterisation and angiography is performed selectively following
non-invasive investigations. An assessment of pulmonary function is made using spirom-
etry (FEV1, FVC and FEV1/FVC ratio). In addition a functional assessment, such as ask-
ing a patient to walk up a flight of stairs, is often informative. Cardiopulmonary exercise
(CPX) testing is now available in a number of centres and gives detailed global informa-
tion on the patient’s cardiopulmonary performance during exercise. While exercising on a
static bicycle ergometer, a closed breathing circuit is used to measure inspired and expired
volumes and pO2 and pCO2 breath-by-breath. As the intensity of exercise increases the
ECG is observed for ischaemic changes and the anaerobic threshold and oxygen consump-
tion are calculated.
Renal function and hepatic function are assessed using standard blood laboratory meas-
ures. The patient’s current level of mobility and independence and their mental resilience,
although hard to assess objectively, are also important in their ability to rehabilitate follow-
ing major surgery.
The results of all these investigations should be considered by a surgeon and anaesthetist
with additional input from a radiologist in order that an appropriate management plan can
be recommended to the patient. The patient’s medications should be reviewed and consid-
eration given to whether they can be optimised in advance of any surgical or endovascu-
lar intervention. In particular, consideration is given to antiplatelet, β-blocker and statin
therapy.
Open surgical management

Type I, II, III and V thoracoabdominal aortic aneurysms
With the patient in the right lateral position an extensive thoracolaparotomy is made
through the left 6th intercostal space with partial division of the anterior muscular part
of the diaphragm, partial division of the left crus of the diaphragm and posterior division
of the 6th rib to facilitate access. The left lung is partially deflated using a double lumen
endotracheal tube and the aorta is exposed behind the left lung by division of the parietal
pleura allowing control of the proximal aorta to be secured. The dissection is continued
into the abdomen with medial visceral rotation of the left colon, left kidney and spleen
being performed through the left paracolic gutter to expose the entire length of the aorta
in the abdomen. The iliac arteries are dissected out and the left femoral artery is exposed
through a separate incision to accommodate the bypass circuit. Following full heparinisa-
tion left heart bypass is established with the afferent limb in the left pulmonary vein or left
atrium and the efferent limb in the femoral artery. For a type II TAAA the aorta is cross-
clamped distal to the left subclavian and in the mid-descending aorta (above T6). The aorta
is opened longitudinally between the clamps and transected proximal to the aneurysm. The
proximal end of a Dacron graft of appropriate diameter is anastomosed using a continu-
ous prolene suture, and Teflon pledgelets and cyanoacrylate glue can be used to reinforce
this suture line. The graft is clamped just beyond the anastomosis and the aortic clamp is
replaced in a supracoeliac position. The thoracic aorta is opened along its length and large
intercostals arteries, usually between T8 and L1, are selected for anastomosis to the graft
on a patch. Occasionally a jump graft is required if vessels are widely spaced. Remaining
small intercostal arteries are oversewn. The graft clamp is replaced beyond the intercostal
patch restoring in-line flow to the spinal artery for the remainder of the procedure. The
infrarenal aorta is then clamped and the renal, supracoeliac and superior mesenteric arter-
ies are anastomosed to the graft on a patch using a jump graft to incorporate the left renal
artery if necessary. Flow to the visceral vessels is restored while the distal anastomosis is
performed.
By using a serial clamping technique the ischaemic insult to the spinal cord and viscera
is minimised compared to clamping the entire aorta for the duration of the procedure. Prior
to closing, the patient is actively warmed since normothermia is important for achieving
haemostasis. The surgical techniques and adjuncts that can be used to reduce the complica-
tion rate are discussed below.
Descending thoracic aneurysms

An aneurysm of the descending thoracic aorta is repaired through a left-sided thoracotomy
without the need to open the abdomen. Otherwise the technique for open repair is similar to
that described for the thoracic component of an extensive TAAA. As discussed below DTAA
are readily amenable to endovascular repair, which has now largely superseded open repair
in this setting.
Aneurysms involving the aortic arch

Complete replacement of the ascending aorta and arch mandates hypothermic cardioplegic
circulatory arrest and cardiopulmonary bypass, and lies within the remit of the cardiotho-
racic surgeon. However, the vascular surgeon is frequently faced with an aneurysm of the
descending aorta, which also involves part of the aortic arch or which allows insufficient
space to clamp the aorta distal to the left subclavian artery. Endovascular stenting of the
arch would interrupt the blood supply to the brain if the proximal landing zone were to
occlude the left carotid artery and the left vertebral artery, which is a branch of the left
subclavian artery. In this situation a staged procedure should be performed involving a
carotid-carotid bypass and a carotid-subclavian bypass before the aneurysm is repaired
endovascularly.
Type IV thoracoabdominal and suprarenal aneurysms

The surgical approach for type IV and suprarenal aneurysms is similar and generally
involves a midline laparotomy or ‘roof-top’ incision, although some surgeons prefer a thora-
colaparotomy through the 8th, 9th or 10th interspace. The retroperitoneal space is accessed
though the left paracolic gutter and the viscera (descending colon, spleen and left kidney)
are mobilised and rotated medially to reach the aorta. The left crus of the diaphragm is
partially divided but the diaphragm itself is left intact. A Dacron graft is used to replace the
abdominal aorta. A bevel at the proximal end allows incorporation of the renal and visceral
vessels on a patch. The size of the patch should be minimised to reduce the chance of further
aneurysm formation so if the vessels are widely spaced any outlying vessels are anastomosed
using a jump graft. Cerebrospinal fluid (CSF) drainage and other adjuncts are not necessary
for the repair of type IV and suprarenal aneurysms.
Type IV and suprarenal aneurysms are less amenable to endovascular repair than infra-
renal and descending thoracic aneurysms due to the difficulty in preserving the visceral
vessels (see below). A hybrid approach can be used but does not offer improved outcomes in
most patients compared to standard open surgery. In a hybrid approach the coeliac, superior
mesenteric and renal arteries are revascularised in a retrograde fashion using a graft from
one of the iliac arteries prior to placement of an endovascular stent-graft to exclude the
aneurysm.
Endovascular management
Endovascular strategies offer a minimally invasive alternative to open surgical management
with the particular advantage of reducing cardiorespiratory complications. Descending
thoracic aneurysms are particularly amenable to a wholly endovascular approach since
there is no involvement of the arch or visceral vessels. Branched and fenestrated grafts open
up the possibility of treating aneurysms involving the arch and visceral vessels but these
techniques are still being evaluated and currently are only available in a small number of
centres with a particular interest in endovascular aneurysm treatment. Encouraging results
have been achieved in small case series but, while early technical success rates are high,
target vessels remain at risk from stent migration, fracture and kinking [1]. Such grafts are
individually made for elective patients but cannot be obtained in time for patients requir-
ing urgent intervention. Wholly endovascular techniques undoubtedly reduce the cardio-
pulmonary complications associated with open repair but the risk of paraplegia remains,
re-intervention may be necessary, long-term durability results are awaited and annual
post-procedure CT surveillance is required [1]. The stent graft industry is continuously
developing improved devices that make increasingly complex lesions accessible to endovas-
cular approaches. Hybrid procedures, combining endovascular exclusion of the aneurysm
with open surgical revascularisation of aortic branches covered by the stent, have also been
advocated but again accrual of outcome data is ongoing. Hybrid procedures have been used
with particular success to treat arch aneurysms when accompanied by an open debranch-
ing procedure of the arch.
Complications and adjuncts

The morbidity and mortality associated with repair of extensive aneurysms has been reduced
to acceptable levels over recent years with a better understanding of the pathophysiology
of complications leading to modifications to the perioperative strategy [2]. Despite this,
however, it is unsurprising that the complication rate, as a product of the magnitude of
the operation and the level of comorbidity present in the DTAA/TAAA patient population,
remains significant. Table 6.1 shows the incidence of the major complications associated
with thoracic and thoracoabdominal aneurysm repair reported in two large contemporary
case series.
Reducing the overall risk

Preoperative assessment is necessary to stratify the risk of complications on an individual
patient basis and to select patients suitable for intervention. Centralisation of services for
managing DTAA and TAAA is vital in order that experienced surgical, anaesthetic, critical
care and theatre teams can be assembled along with the wide range of facilities and equip-
ment that they require.
Intraoperatively systemic heparinisation is preventative against myocardial infarction,
minimises the risk of embolisation, preserves the microcirculation and reduces the risk of
thrombosis of key branches, particularly intercostal arteries, during clamping. Moderate
Table 6.1 Contemporary results of open repair of thoracoabdominal aneurysms
Conrad et al. (2007) [4] Coselli et al. (2007) [3]

Number of patients (dates) 445 (1987–2005) 2286 (since 1986)
Crawford extent I/II (%) 41.8 64.2
Ruptured (%) 11.4 6.1
Death (<30 day) (%) 8.3 6.6
Major paraplegia (%) 9.5 3.8
Renal failure (requiring dialysis) (%) 4.6 5.6
Cardiac complications (%) 14.7 7.9
Pulmonary complications (%) 49 32.1
permissive hypothermia (32–34 °C) reduces the metabolic demands of the viscera and the
spinal cord, thereby increasing their tolerance to ischaemia. Unlike repair of the ascend-
ing aorta and aortic arch, full hypothermic circulatory arrest is rarely required for DTAA
and TAAA repair. Left heart bypass is used for extensive repairs (Crawford extent I and
II) enabling retrograde aortic perfusion, facilitating cooling/re-warming and reducing
spinal cord and visceral ischaemia. Left heart bypass also allows patients with cardiac
disease to tolerate proximal aortic clamping more readily by offloading the left side of
the heart. Serial aortic clamping (as described above) involves moving the clamps distally
along the aorta as the operation progresses and reduces the ischaemic insult and subse-
quent reperfusion injuries. Additional strategies are directed at reducing the incidence of
specific complications.
Reducing cardiac and respiratory complications

The aetiology of cardiac and respiratory complications following DTAA and TAAA repair
is multifactorial involving, in particular, existing comorbidities, a large surgical incision
involving the thoracic cavity and fluid replacement to compensate for significant blood
loss. Unfortunately there is limited potential to modify these factors, which is the reason
that cardiopulmonary complications remain problematic in extensive aortic surgery and is
why the possibility of endovascular repair is so attractive. Nonetheless identification of risk
and optimisation of function preoperatively are important. Limitation of the incision to a
laparotomy for type IV repair avoids the need to enter the thorax. Chest drains are placed
routinely when the chest has been entered. Prompt extubation reduces ventilator-associated
pneumonia and respiratory muscle wasting, and early tracheostomy placement in selected
patients facilitates weaning.
Reducing the risk of paraplegia

The risk of paraplegia was much higher (in the region of 16%) before the introduction
of modern adjuncts, but remains higher for type II aneurysms (6.3–14.5%) than for less
extensive type IV aneurysms (2.9–3.8%) [3]. This is directly related to the length of aorta
that is replaced and the risk of paraplegia is also increased following previous aortic sur-
gery or dissection. The blood supply to the spinal cord is from multiple sources including
the vertebral arteries, anterior segmental medullary arteries (intercostal and lumbar) and
the internal iliac arteries, all of which contribute to the anterior spinal artery. The artery
of Adamkiewicz is the largest contributor to the anterior spinal artery and is generally
a left-sided intercostal artery but may sometimes arise from the right or from a lumbar
artery.
Reimplanation of the major segmental arteries, ideally including the artery of
Adamkiewicz, is thought to be important in reducing the risk of paraplegia. While some
surgeons advocate preoperative mapping of the spinal cord blood supply using MRI,
most would select one to three large pairs of intercostal vessels (usually between T6 and
T12) intraoperatively for reattachment. Serial clamping and left heart bypass reduce spi-
nal cord ischaemia during extensive repair but despite this some degree of spinal cord
ischaemia is inevitable. Placement of a CSF drain allows monitoring of CSF pressure
during the operation and postoperatively. CSF is actively drained during aortic cross
clamping and passively drained for 48 hours postoperatively allowing decompression of
the vertebral foramen in the event of spinal cord swelling. Since CSF drainage carries
the risks of infection and haematoma formation it is reserved for extensive aneurysm
repairs. Spinal cord perfusion is further optimised by maintaining the mean arterial
pressure >80 mmHg and avoiding hypotensive, vasodilating drugs where p ossible. Motor
evoked potentials can be used to detect spinal cord ischaemia during cross-clamping
and following reimplanation of vessels. Postoperatively, rather than being fully sedated,
even intubated patients are nursed in a rousable state to facilitate assessment of the
neurological function of the lower limbs and detection of signs of delayed-onset spinal
cord ischaemia.
Reducing the risk of renal and gastrointestinal tract ischaemia

Left heart bypass, moderate systemic hypothermia, heparinisation and serial clamping all
contribute to a reduced risk of visceral ischaemia. Additionally a trial of renal cooling com-
paring renal perfusion with normothermic blood and 4C crystalloid fluid found the latter to
be advantageous in reducing renal complications. Retrograde aortic perfusion on left heart
bypass maintains the blood supply to the abdominal viscera during surgery on the proximal
aorta. When the clamps are replaced on the infrarenal aorta, selective visceral perfusion can
be maintained using individual balloon perfusion catheters in the ostia of the visceral ves-
sels. Some surgeons use renal cooling and selective visceral perfusion routinely whilst others
use these measures selectively in patients perceived to have a higher than average risk of
complications.
Reducing complications of blood loss and coagulopathy

Significant blood loss occurs during the repair of extensive aneurysms and massive heter-
ologous transfusion is associated with a number of complications including coagulopathy,
hypothermia and transfusion reactions. The use of a cell saver and autologous transfusion
reduces heterologous transfusion requirements. Near-patient testing of coagulation, such
as thromboelastography, can be used to detect impending coagulopathy before it becomes
evident clinically. It also enables identification of the specific part of the coagulation cascade
that is deficient so that treatment can be targeted and timely, before bleeding complications
become problematic.
References 3. Coselli JS, Bozinovski J, LeMaire SA. Open

surgical repair of 2286 thoracoabdominal
1. Greenberg RK, Lytle B. Endovascular repair aortic aneurysms. Ann Thorac Surg 2007;
of thoracoabdominal aneurysms. Circulation 83: S862–S4; discussion S890–S2.
2008; 117: 2288–96. 4. Conrad MF, Crawford RS, Davison JK,
2. Coselli JS, LeMaire SA. Tips for successful Cambria RP. Thoracoabdominal
outcomes for descending thoracic and aneurysm repair: a 20-year perspective.
thoracoabdominal aortic aneurysm Ann Thorac Surg 2007; 83: S856–S61;
procedures. Semin Vasc Surg 2008; discussion S890–S2.
21: 13–20.
Chapter
7
Aortic dissection
Jenny Richards and Rod Chalmers
Key points
• A ortic dissection may be defined as acute (<2 weeks since symptoms) or chronic (>2
weeks since symptoms)
• Acute aortic dissection is associated with hypertension and pre-existing disease of the
aortic wall
• Patients usually present with severe, tearing chest and back pain
• Diagnosis is with transoesophageal echocardiogram and computed tomography (CT)
angiogram
• The Stanford or DeBakey classifications are used to describe the pattern of dissection
• Type A dissections require early surgical treatment. The in-hospital mortality is high
but the prognosis is very good in patients surviving to discharge from hospital
• Type B dissections are treated medically in the first instance. The in-hospital mortality
is much lower but complications may occur following discharge
• Chronic type B dissection may result in aortic dilatation and rupture so surveillance is
required
• Endovascular strategies have a role in some situations for acute type A dissection and
acute type B dissection, but the role in chronic type B dissection is unclear
Epidemiology
Aortic dissection occurs with an incidence of approximately 3 per 100 000 per year.
Accurate estimates are difficult due to the high out-of-hospital mortality for this condi-
tion and the low post-mortem rate in most countries. It affects males twice as commonly
as females and the risk increases with age. Interestingly, as with a number of other car-
diovascular conditions, it demonstrates circadian variation with a peak incidence early
in the morning and during the winter. This pattern correlates with times of peak blood
pressure.
Aetiology
Factors predisposing to aortic dissection involve pre-existing degenerative, genetic or inflam-
matory conditions that weaken the aorta and primary or secondary causes of hypertension.
Table 7.1 shows the factors that have an association with aortic dissection.
Chapter 7: Aortic dissection 109
Table 7.1 Factors associated with aortic dissection
• Increasing age
• Male sex
• Diabetes mellitus
• Pre-existing aortic disease e.g. aneurysmal disease, atherosclerosis, arteritis
• Connective tissue disorders e.g. Marfan’s syndrome, Ehlers–Danlos syndrome
• Congenital anomalies e.g. bicuspid valve, aortic hypoplasia
• Hypertension primary or secondary e.g. pregnancy, cocaine use, weightlifting
• Iatrogenic e.g. cardiac catheterisation and intervention, coronary artery bypass grafting, valve
replacement
• Trauma e.g. deceleration injury
• Previous aortic surgery
• Family history
Pathophysiology
A breach in the aortic intima allows blood to track within the media of the vessel wall, cre-
ating a true lumen and a false lumen. This entry tear usually lies in the ascending aorta,
aortic arch or just distal to the left subclavian artery. Dissection results from a combination
of the stress applied to the aortic wall due to high, fluctuating blood pressure and disease
processes, which damage and weaken the vessel wall. It is now recognised that the breach
in the intima may be caused by a penetrating atherosclerotic ulcer or alternatively an intra-
mural haematoma resulting from disruption of the vasa vasorum, which may subsequently
penetrate into the lumen. Although initially benign, intramural haematoma regresses in
only 10% of patients while 28–45% progress to aortic dissection, 20–45% rupture and more
proximal lesions are particularly prone to complication [1]. In a smaller number of patients
the intimal injury is traumatic or iatrogenic. Intimal disruption initiates the dissection and
proximal and/or distal propagation follows and is facilitated if the media has been weakened
by enzymatic (e.g. matrix metalloproteinases) degradation of the extracellular matrix. Blood
may travel within the false lumen along a variable length of the aorta and its branches. The
exit tear may re-enter the true lumen connecting it with the false lumen or it may rupture
outwards, with the latter resulting in rapid death. End-organ ischaemia occurs as branches
become thrombosed or perfusion may be dynamically influenced as the true lumen is vari-
ably compromised by changes in flow in the false lumen during the cardiac cycle and as
arterial pressure fluctuates.
Presentation
The classical presentation of acute aortic dissection is with sudden onset of extremely severe
chest pain, often described as ‘tearing’ in nature, located anteriorly radiating to the neck
or back and migrating as the dissection propagates. While pain is prominent in the major-
ity of patients, a proportion will complain of less severe pain or no discomfort at all and,
in fact, lack of pain is a predictor of poor outcome, perhaps because diagnosis is delayed.
Where chest pain is present the diagnosis is often confused with acute myocardial infarction
although, if a detailed history is taken, the sudden onset and nature of the symptoms may
be used to distinguish the true cause. The two may also occur together if the coronary arter-
ies are affected by the dissection. The integrity of the aortic valve may be disrupted causing
acute aortic regurgitation and, if the dissection communicates with the pericardium, tam-
ponade may result. Patients with a type A dissection are often hypotensive while hyperten-
sion is typical in type B dissection.
As the false lumen extends it may compromise some of the key branches of the aorta,
impairing end-organ perfusion and precipitating organ dysfunction or failure. Symptoms
include myocardial ischaemia (coronary arteries), cerebrovascular accident (carotid and ver-
tebral arteries), spinal cord ischaemia (intercostals and lumbar arteries), renal dysfunction
(renal arteries), mesenteric ischaemia (coeliac and mesenteric arteries) and limb ischaemia
(subclavian and iliac arteries), which often become evident in serial fashion as the dissection
extends. On examination pulses may be weak, asymmetrical or absent, and again may disap-
pear serially with progression of the dissection.
Diagnosis and classification

Sudden death may occur but in those surviving to present to the emergency department,
rapid diagnosis is important to plan treatment. Plain radiography and electrocardiogram
(ECG) are commonly performed in patients presenting with acute chest pain but, while
they may show some non-specific changes consistent with aortic dissection, neither is diag-
nostic. A plain radiograph of the chest may be normal in a third to a half of patients [2]
or reveal mediastinal widening, an abnormal contour of the aortic arch or a haemothorax.
Electrocardiogram is often normal but may show signs of myocardial ischaemia or left ven-
tricular strain/hypertrophy.
An echocardiogram and cross-sectional imaging (either magnetic resonance imaging
[MRI] or CT) are needed to confirm the diagnosis and the extent of the dissection proxim-
ally and distally. Whilst transthoracic echocardiography (TTE) has the advantage of being
rapid and non-invasive in unstable patients, transoesophageal echo (TOE) has a higher
sensitivity and specificity. CT or magnetic resonance angiography with contrast yield add-
itional information about organ perfusion and the relative flow in the true and false lumens
and, unlike echocardiography, both have the advantage of imaging the entire thoracic cav-
ity if an alternative diagnosis is possible. The performance of CT and MRI are comparable
in diagnosing and excluding aortic dissection. The advantage of CT is that it is more widely
available and the scanning time is short. The use of MRI for long-term surveillance to avoid
an accumulating ionising radiation dose is beneficial.
Acute dissection is defined as being within 2 weeks of the onset of symptoms and after this
time becomes chronic. The Stanford and Debakey classifications are useful for stratifying risk,
predicting complications and planning treatment. The Stanford classification distinguishes
dissections affecting the ascending aorta (type A) from those which affect only the descending
aorta (type B). The Debakey classification further subdivides dissections affecting the ascend-
ing aorta into those affecting the descending aorta as well (type I) and those only affecting the
ascending aorta (type II). Debakey type III is equivalent to Stanford type A (Figure 7.1).
If the ascending aorta is affected the patient may experience cardiac, cerebral and upper
limb complications since the coronary, carotid and subclavian arteries are at risk. Dissections
of the descending and abdominal aorta cause complications related to spinal cord, visceral
and limb ischaemia. Considering treatment, type A dissections generally require surgical
intervention while type B dissections can be managed medically in the majority of patients.
De Bakey I II III Figure 7.1 Stanford and

Stanford A A B Debakey classification of aortic
dissections.
Erbel R et al. Diagnosis and
management of aortic
dissection. Eur Heart J 2001;
22: 1642–81. Reproduced with
permission.
Management
The management of aortic dissection depends upon the extent and location of the dissection
and the stability and comorbidities of the patient [6]. Due to the nature of this condition
there is a paucity of high-level evidence to support practice but the International Registry of
Aortic Dissection (IRAD) has gathered valuable information about the trends and outcome
of recent practice [1, 4]. In addition the European Society of Cardiology taskforce produced
guidelines for the management of aortic dissection [2].
Type A
Type A dissections account for around two thirds of all cases of dissection [3]. Urgent
surgery is almost always required for type A dissections unless the patient is moribund or
has significant pre-existing co-morbidities. Untreated type A dissections have a mortality
in the region of 1–2% per hour for the first 24 hours, amounting to 50–75% by 2 weeks.
With medical management alone the mortality of type A dissections in the IRAD data-
base was 56% [1]. Surgical management involves sealing off the entry point to decom-
press the false lumen and reduce the risk of propagation of the dissection. Since the
entry tear usually lies in a diffusely diseased part of the aorta, this generally involves
resection of the aorta at the site of the entry tear and reconstruction using a short graft
(e.g. Dacron) through a median sternotomy approach with hypothermic cardioplegic
circulatory arrest and cardiopulmonary bypass. When the entry tear lies in the arch,
reconstruction or reimplanation of the innominate, carotid and subclavian vessels will be
necessary. If there is incompetence of the aortic valve this may have to be reconstructed
or, more commonly, replaced with a prosthesis. The ostia of the coronary arteries may
also require reconstruction or coronary artery bypass grafting. The operative mortality
for dissections of the ascending aorta is in the region of 16%, higher (31.4%) if the patient
is shocked before surgery [1].
The false lumen may extend for a variable distance into the descending aorta but due to
the morbidity associated with replacing long segments of the aorta, including the visceral
vessels, this is usually avoided in the first instance where possible. The operation on the
ascending aorta effectively converts this situation into a type B dissection, which can then
be managed medically as described below. While there is not yet an established role for
endovascular management of type A dissections, there is interest in using a hybrid approach
combining open surgery for the ascending aorta and arch with endovascular stenting of the
descending aorta. This approach may be useful for patients with malperfusion syndromes,
intractable pain, uncontrolled hypertension or continued propagation or expansion of the
false lumen.
An elephant trunk procedure is used when it is anticipated that the descending aorta
will need to be replaced subsequently. The ascending aorta and the arch are replaced with
a tube graft. The arch vessels are re-implanted into the graft on a single patch. Rather than
cutting the graft to length the distal end is left long and projects into the descending aorta
enabling a staged procedure to be performed. At a later date an open surgical approach to
the descending aorta can be used with the proximal anastomosis being made to the elephant
trunk, or the elephant trunk can be used as the proximal landing zone for an endovascular
stent graft. The advantage of this is that it avoids the need to graft directly onto diseased and
fragile aortic wall.
Type B
Type B – uncomplicated
The outcome for type B dissection is much more favourable than for type A dissection with
a 30-day survival rate of 89% [1]. Operative management is extremely hazardous and is
avoided if possible [5]. Aggressive medical management involving analgesia and control of
hypertension and heart rate is important. Patients should be admitted to a coronary care
or intensive care unit for close monitoring of blood pressure, end organ function and any
evidence of progression such as escalating or intractable pain or development of new symp-
toms. Intravenous β-blockers (e.g. esmolol, labetolol), unless contraindicated, are used in
the first instance to reduce systolic blood pressure to 100–120 mmHg. Rate-limiting calcium
channel blockers (e.g. diltiazem) are an alternative for patients intolerant of β-blockers, and
a nitrate infusion may be used if either of these agents is insufficient in isolation. In the
longer term, angiogenesis-converting enzyme (ACE) inhibitors may have the additional
advantage of promoting remodelling of the aortic wall but the evidence for this has not yet
been accumulated.
Type B – complicated

In the International Registry of Acute Aortic Dissection (IRAD) database, 17.2% of patients
with type B dissection required surgical intervention during the acute phase [6]. Indications
for surgery were rupture (23%), further propagation of the dissection (52%), uncontrolled
hypertension (15.5%), intractable pain (14.1%), malperfusion of the viscera (23.9%) or limbs
(15.5%) or a combination of these factors. The exact procedure is tailored to the anatomy
of the dissection for a given patient. Graft replacement of the descending aorta is the most
common open surgical procedure (69.3% of surgically managed acute type B dissections
in the IRAD series [9, 6]), which may be performed in conjunction with replacement of
the arch and fenestration, stenting or bypass for compromised vessels. The requirement for
surgical intervention was associated with high mortality (29.3%), particularly if required
during the first 2 days [6] and there has therefore been considerable interest in the possibil-
ity of employing endovascular techniques in the management of complicated acute type B
dissection. From the EUROSTAR/United Kingdom registry 131 patients with aortic dissec-
tion were treated with 89% primary technical success, a paraplegia rate of <1% and a mor-
tality rate of 8.4% at 30 days and 10% at one year [7]. In the IRAD database endovascular
treatment was associated with better short-term outcomes than open surgical treatment for
acute type B dissection [1]. Long-term follow-up data are awaited. In this context a stent
graft may be used to occlude the entry tear and bare metal stents hold the true lumen open
and compress the false lumen, which may improve dynamic malperfusion phenomena.
Compromised vessels may also be targeted directly with bare metal stents to restore flow.
In the longer term it is possible that endovascular occlusion of the false lumen will promote
thrombosis, thus encouraging remodelling of the dissected wall and reducing the risk of
late expansion and rupture. There is currently no definite evidence to support prophylactic
placement of bare metal stents for this purpose.
Type B – chronic

Accurate data on the outcome of patients with chronic dissection are sparse and 5-year
mortality is estimated at 50%, suggesting that medical management alone is suboptimal.
The European Society of Cardiology recommends regular surveillance at 1, 3, 6, 12 months
and annually thereafter to monitor for expansion of the affected aorta [2]. Thresholds for
intervention are recommended at 5 cm for the ascending aorta and 6 cm for the descend-
ing aorta [2]. The IRAD database showed partial thrombosis of the false lumen to be an
independent predictor of risk of death at 3 years [1]. Surgical management involves graft
replacement of the entire aorta as described for open thoracoabdominal aneurysm repair.
While early technical success can be achieved using endovascular repair, evidence to sup-
port or refute a role for endovascular treatment of chronic type B dissection is currently
limited but early results do not appear to be as encouraging as for acute dissection.
Complications
The main complications of type A dissection repair are cardiac, stroke and death. The com-
plications of intervention for type B aortic dissection are similar to those described for open
repair of extensive TAAA but patients with a history of dissection are at higher risk of para-
plegia, renal failure and death.
While acute type A dissection has a high in-hospital mortality (33%) among patients
who survive to discharge, the 1- and 3-year survival rates are 96% and 91%, respectively [1].
In contrast 90% of patients with acute type B dissection survive to discharge but amongst
those surviving to be discharged from hospital the 1- and 3-year survival rates are 78% and
76%, respectively [1].
References 3. Golledge J, Eagle KA. Acute

aortic dissection. Lancet 2008; 372:
1. Tsai TT, Trimarchi S, Nienaber CA. Acute 55–66.
aortic dissection: perspectives from the 4. Trimarchi S et al. Contemporary
International Registry of Acute Aortic results of surgery in acute type
Dissection (IRAD). Eur J Vasc Endovasc Surg A aortic dissection: the
2009; 37: 149–59. International Registry of Acute Aortic
2. Erbel R et al. Diagnosis and management Dissection experience.
of aortic dissection. Eur Heart J 2001; J Thorac CardiovascSurg 2005;
22: 1642–81. 129: 112–22.
5. Akin I, Kische S, Ince H, Nienaber Aortic Dissection (IRAD). Circulation 2006;

CA. Indication, timing and results 114: 357–64.
of endovascular treatment of type B 7. Leurs LJ et al. Endovascular treatment
dissection. Eur J Vasc Endovasc Surg 2009; of thoracic aortic diseases: combined
37: 289–96. experience from the EUROSTAR and
6. Trimarchi S et al. Role and results of surgery United Kingdom Thoracic Endograft
in acute type B aortic dissection: insights registries. J Vasc Surg 2004; 40: 670–9;
from the International Registry of Acute discussion 679–80.
Chapter
8
Popliteal artery aneurysms
Robert Davies, Asif Mahmood and Rajiv Vohra
Key points
• Popliteal artery aneurysm is the most common lower limb aneurysm
• Lower limb arterial aneurysms rarely occur in isolation
• The majority present either as an incidental finding or as lower limb ischaemia
• Aneurysm rupture is rare, but life and limb threatening when it occurs
• All symptomatic popliteal and femoral artery aneurysms should undergo repair
• Asymptomatic popliteal artery aneurysms >2 cm in maximum diameter should be
considered for elective repair
• Asymptomatic femoral artery aneurysms >2.5 cm in maximum diameter should be
• Adequate imaging of the proximal and distal vasculature is vital for successful repair
Introduction
Popliteal artery aneurysms are the second most common peripheral arterial aneurysm and
together with femoral artery aneurysms represent the vast majority of true lower limb arter-
ial aneurysms. They frequently occur in association with aortic aneurysms and are often
identified as an incidental finding during routine examination or radiological imaging.
However, they may present with lower limb thromboembolism or rupture and thus their
identification and subsequent management is vitally important to the patient’s quality of
life and life expectancy.
Pathophysiology
The majority of popliteal artery aneurysms (PAAs) develop secondary to a systemic athero-
sclerotic disease affecting the medial layer of the artery and are often associated with aneu-
rysms elsewhere. In early retrospective studies in which PAAs were detected clinically or
with the use of angiography rather than Duplex scanning, the incidence of PAAs in patients
with aortic aneurysms was between 3.2 and 3.9%. However, more recent prospective studies
have shown the incidence to be two to three times greater than this. Conversely, if a PAA
is detected, concomitant aortic and femoral aneurysms are present in up to 50% and 40%
of cases, respectively. It is also important to examine and image the contralateral popliteal
artery, as 50% are bilateral[1].
Table 8.1 Presentation of popliteal aneurysms
Presentation %
Limb ischaemia 55
Asymptomatic 37
Rupture 1.4
Local compression 6.5
Causal explanations for the preponderance of the popliteal artery to become aneurysmal
include fixation at the adductor hiatus and exposure to repetitive trauma with the use of
adductor magnus or extrinsic compression from knee flexion and extension. Uncontrolled
hypertension is associated with aneurysm expansion. Non-atherosclerotic causes of aneur-
ysmal change are rare and include inflammatory and connective-tissue disorders such as
Marfan’s syndrome and Behçet’s disease. Aneurysmal changes can also be seen after post-
stenotic blunt and penetrating trauma. Mycotic aneurysms in the popliteal fossa have also
been rarely reported.
Symptoms and signs

Most occur in men with a median age of 60 to 65 years at the time of diagnosis. Approximately
63% (Table 8.1) are symptomatic at presentation with asymptomatic aneurysms detected on
palpation or surveillance Duplex ultrasound scan (DUS) of the popliteal fossae in patients
with diagnosed aortic, iliac, femoral or contralateral PAAs.
A pulsatile mass on palpation of the popliteal fossa must be differentiated from other vas-
cular lumps such as pseduoaneurysm from trauma or anastomotic dehiscence after bypass
surgery. Venous aneurysm and non-vascular lumps with transmitted pulsation arising from
the soft tissues such as Baker’s cyst, bursae, tumours and lymph nodes are also included in
the differential diagnoses.
The most common complication of popliteal aneurysms is thrombosis, embolism or a
combination of both. This may result acutely with either intermittent claudication or as an
acutely ischaemic leg, threatening the limb unless revascularisation is rapidly undertaken.
Because of the need for aneurysm ligation during surgery and greater risk of embolism dur-
ing intra-arterial thrombolysis it is important to differentiate acute ischaemia with popliteal
aneurysm from ischaemia from other causes of thrombosis and embolism.
A more insidious presentation is that of gradual onset of claudication or chronic critical
limb ischaemia with rest pain and tissue loss. Silent thrombosis and embolism is also pos-
sible; clinically detected by impalpable pedal pulses and reduced ankle brachial pressure
index (ABPI), this is one of the indications for surgery. Large aneurysms may compress
neighbouring structures such as the popliteal vein and/or tibial nerve. The former may be
mistaken for other causes of lower limb oedema and venous hypertension while the latter
may be misdiagnosed as a primary neurological disorder. Rupture is the most unlikely pres-
entation of popliteal aneurysms but may threaten life as well as limb from the blood loss.
Investigation
Information is required regarding the extent of the aneurysm, the state of inflow and run-off
vessels due to coexistent disease. In the case of aneurysms that are producing compression
symptoms, information about the anatomy of the aneurysm, including diameter, extent and
Chapter 8: Popliteal artery aneurysms 117
(a) (b)
Figure 8.1 (a) Cross-sectional view of bilateral popliteal aneurysms. (b) Coronal view of bilateral popliteal
aneurysms.
effect on neighbouring structures is required. Duplex ultrasound scan provides excellent

information regarding inflow, outflow, diameter and presence of sac thrombus. Although
angiography will show the state of the run-off, it adds little to anatomical evaluation. Fusiform
aneurysms tend to involve the whole length of the popliteal artery and may have stenotic or
occluded segments in addition to the aneurysmal disease and may also affect the superficial
femoral artery. Sacular aneurysms, on the other hand, are more localised, affecting mainly
the mid-popliteal artery. In cases where there are compression symptoms, assessment of the
surrounding structures may be carried out with computed tomography (CT) angiography or
magnetic resonance imaging (MRI) (Figures 8.1(a) and (b)).
Management
When to repair
All patients with symptomatic aneurysms require intervention for the relief of symptoms
and prevention of limb loss. The indications for surgery in asymptomatic popliteal aneurysm
remain controversial with no randomised controlled trials available to establish clear indica-
tions for intervention. Enthusiasts of surgery argue that since up to 100% of asymptomatic
aneurysms will develop thromboembolic complications within 5 years without surgery, with
an associated amputation rate of <67%, intervention is always recommended [1, 2]. In one
prospective study, 18 of 58 aneurysms that had been treated conservatively developed com-
plications [1]. Better graft patency and limb salvage have been noted for asymptomatic rather
than symptomatic aneurysms. In fact, the results for reconstruction for popliteal aneurysm
are better than for occlusive disease [3]. This may be related to better run-off, use of shorter
length grafts or reduced thrombogenicity of blood in these patients compared with occlusive
disease. Some surgeons favour an expectant policy with the use of thrombolysis if required.
Schellack et al. showed that only 2 of 26 (8%) aneurysms developed thromboembolic com-
plications at a mean follow up of 37[4]. Bowyer et al. reported good results with intra-arterial
thrombolysis for asymptomatic aneurysm presenting with acute ischaemia [5].
Most of these studies have been retrospective with risk factors for complications being a
diameter greater than 2 cm, presence of aneurysm sac thrombus and signs of silent embol-
ism i.e. loss of pedal pulses or lowering of ABPI. Varga et al. attempted to identify indica-
tions for the management of asymptomatic aneurysm in a prospective multi-centre study
but the numbers with small aneurysms (<2 cm) were insufficient to establish risk of compli-
cations with conservative treatment in this group [1]. In another prospective study, Galland
et al. showed that those with a diameter of less than 3 cm were unlikely to thrombose [6].
Conversely, it has been shown that enlarging diameter of the sac does not increase the risk
of thromboembolism. Also, some small aneurysms (<2 cm diameter) may have dispropor-
tionate amounts of thrombus in the sac with a greater risk of embolism. Thus we advocate
elective surgical repair for all asymptomatic PAAs ≥2 cm in maximum diameter or those
PAAs ≥1.8 cm with high thrombus content in suitable patients.
Surgical approaches
The most widely utilised surgical technique for repair of PAA is proximal and distal aneu-
rysm ligation combined with long saphenous vein bypass through a medial approach. The
advantages of this approach are the ability to expose the crural vessels and harvest the long
saphenous vein from the same incision. However, the disadvantage is the potential for con-
tinued sac expansion post repair. In our series, 15% of PAA repairs treated by ligation and
bypass demonstrated sac perfusion at a median of 75 months after primary PAA repair. Forty
per cent of these patients presented with symptoms, including one rupture [7]. This compli-
cation of the ligation and bypass technique has been reported by others and is thought to be
secondary to patent collaterals, similar in endovascular terms to a type II collateral endoleak.
This imparts near systemic arterial pressure within the excluded sac, causing expansion with
the possibility of subsequent rupture or neurovascular compression.
Some authors advocate a posterior approach with interposition grafting in the presence
of compressive features or focal PAAs (Figure 8.2). This involves an ‘S’ shaped incision in the
popliteal fossa with dissection of the aneurysm from other popliteal structures followed by
inlay grafting
The advantage of this approach is the direct visualisation of the geniculate arteries allow-
ing them to be oversewn thereby preventing aneurysm sac reperfusion. Furthermore, the
sac may be excised, which is especially important in cases of neurovascular compression.
This technique is not generally suitable for PAAs extending proximal to the adductor hiatus
or where crural vessel exposure is required. Thus we advocate interposition grafting as the
operation of choice for focal, large PAAs that do not extend above the adductor hiatus.
Conduits
Although the autologous vein is the conduit of choice, the conduit for grafting depends
on the availability of a suitable vein, the presence of infection and the length of the graft
required. Hence, for mycotic aneurysms, prosthetic conduits should be avoided.
Inlay grafting requires a small length of graft and as such prosthetic conduits such as pol-
ytetrafluoroethane (PTFE) have been shown to produce adequate results, although a venous
conduit is recommended if the distal anastomosis is on the distal below-knee popliteal
artery.
For bypass grafts below the knee, a thorough search for a vein using a combination of
preoperative vein mapping and on-table exploration of the leg veins (saphenous trunks)
usually allows several segments of composite spliced vein to be anastomosed together to
Figure 8.2 (a) Posterior

approach to large popliteal
aneurysm. (b) Vein inlay
graft repair of large popliteal
aneurysm.
(a)
(b)
form a suitable conduit if a single length of vein is not available. Arm vein can be utilised if
the saphenous vein is not available. In our series the long-term results using this approach
are comparable to ligation and bypass with a trend for superior primary patency rates. The
second best option is the use of a composite-sequential conduit with above-knee prosthetic
conjoined by a three-panelled anastomosis to the above-knee popliteal artery and vein tra-
versing the knee. [7]. A prosthetic graft with the addition of a venous cuff or fistula remains
the least attractive option due to the relatively poor results in terms of graft patency, limb
salvage and risk of infection.
Management of thromboembolism
The acutely ischaemic leg from popliteal aneurysm remains a challenge because of the poor
limb salvage achieved. A multi-disciplinary approach involving a vascular surgeon and
interventional radiologist is required. Treatment options include intra-arterial thrombolysis
alone, thrombolysis to remove clot in the aneurysm sac and run-off followed by surgery or
intraoperative lysis with thromboembolectomy followed by reconstruction. The treatment
depends on the degree and duration of ischaemia and co-morbidities. In a chronically ischae-
mic leg, it is unlikely that lysis or embolectomy will clear a well packed clot in the run-off and
therefore a crural or pedal bypass remains the most realistic option. Interventional radio-
logical techniques include preoperative intra-arterial thrombolysis and aspiration embolec-
tomy. First described by Schwarz et al. in 1984, thrombolysis can be successfully used as the
primary intervention [8]. In a unit favouring conservative treatment of all asymptomatic
aneurysms, Bowyer et al. reported successful thrombolysis in six of nine patients with acute
thromboembolic complications [5]. Limb salvage was achieved in all nine cases, including
those with failed lysis requiring additional procedures such as lumbar sympathectomy and
angioplasty. The combination of preoperative lysis, followed by surgery, was shown to be
effective by Carpenter et al., who reported a limb salvage rate of 100% with combination
treatment in comparison to 57% for surgery alone [9]. However, details of run-off clearing
procedures were not given in this study. Preoperative intra-arterial thrombolysis should not
be considered in every case, especially in the presence of neurological signs, as it is associ-
ated with a poor outcome. Immediate surgery is required in this situation. Preoperative lysis
is also associated with a risk of distal embolism of 13% in popliteal aneurysms, compared
to 2% in patients with acute ischaemia due to occlusive disease. Not all cases are successful
and lysis may not be effective in dislodging well-packed clot or cholesterol emboli (‘trash’)
with clearance rates of only 58–66%.The theoretical risk of stroke as well as systemic bleed-
ing has been reduced when a low dose lytic agent has been used in the on-table technique.
Intraoperative thrombolysis may be delivered either trans-femorally, via the popliteal artery
or from the ankle routes with the latter two routes being favoured in cases of a threatened
limb with more rapid clearance of run-off being achieved with the catheter being closer
to the run-off. Isolated limb perfusion may also give good results and limit side effects by
reducing the dosage of lytic agent reaching the systemic circulation. Methods for improving
the removal of distal thrombus include popliteal trifurcation embolectomy and ankle level
micro-embolectomies. The embolectomy catheter may not pass easily from the popliteal
artery all the way down the calf vessels for mechanical reasons such as angulations or due
to the nature of occlusive disease. In these circumstances, embolectomy of the crural and
pedal vessels via arteriotomy at the ankle may successfully retrieve clot in the crural or pedal
vessels that is otherwise inaccessible. In this technique the anterior tibial, posterior tibial or
both arteries are exposed by small separate ankle incisions and opened transversely using
loupe magnification and a micro-knife. Proximal and distal embolectomy can be carried out
using a size 2 balloon catheter, directed proximally to the popliteal trunk and distally into
and around the pedal arch. Adjunctive thrombolysis may also be used in cases of incom-
plete embolectomy. In our series, of the 17 patients with acute ischemia, limb salvage was
achieved in 14 with an aggressive management with thromboembolectomy in 12 and bypass
grafting in 6. This was combined with thrombolysis in a few. Acutely ischemic limbs had a
cumulative 5-year secondary patency of 80% and a 30-day mortality of 11.8% [7].
The endovascular era

In recent years endovascular repair has been increasingly employed as a valid alternative
to open surgical repair. Reported benefits include significant reduction in operation time,
blood loss, perioperative morbidity and postoperative hospital stay compared with conven-
tional surgical repair [8, 9]. However, to date, only one randomised controlled trial (RCT)
and three case series with more than 10 PAAs are reported in the literature (Table 8.2).
Table 8.2 Summary of results of endovascular treatment of popliteal artery aneurysms (PAAs)
Secondary
Primary patency patency
Mean
No. of (months) 1 2 1 2
Author PAAs follow up year years year years Occlusions
Tielliu et al. [10] 57 24 80% 77% 90% 87% 12 (21%)
Mohan et al. [11] 30 24 80% 75% 89% 83% 6 (20%)
Antonello et al. [12] 15 46 87% 80% 100% 100% 1 (6.7%)
The experience of the Padua group who presented data on 30 PAAs randomised to sur-
gical (15 PAAs) or endovascular repair (15 PAAs) suggests comparable short-term patency
rates between surgical and endovascular repair. However the study was significantly com-
promised by its power limitation and short follow up. The largest cohort by Tielliu et al.
(57 endovascular repairs) reported inferior primary and secondary patency rates (80% and
90% at 1-year follow up, and 77% and 87% at 2-year follow up) compared to surgical repair
[10]. Multivarate analysis demonstrated treatment with clopidogrel in the postoperative
period was the only predictor of success. Endovascular PAA repair is associated with a risk
of stent occlusion of 6.7–44% at 2-year follow up, with 67–100% occurring within the first 6
months postoperative. Despite this the limb salvage rate is 100% with the majority of stent
occlusions being successfully treated by intra-arterial thrombolysis with residual claudica-
tion symptoms. Endovascular PAA repairs are also at risk of endoleak and stent migration.
Tielliu et al. reported two type I endoleaks secondary to late stent migration in short sealing
zones that were successfully repaired with graft extension. In the same series, early stent
migration caused disconnection of two overlapping stents, resulting in a type III endoleak
requiring a secondary intervention. Of particular interest is the lack of a type II endoleak
in Tielliu et al.’s series, which in open surgical repair is believed to account for the high rate
of sac reperfusion. This lack of type II endoleak may well represent diagnostic error as one
aneurysm sac subsequently showed evidence of enlargement during follow up. Mohan et al.
reported two type II endoleaks in their series [11].
Despite the inferior patency rates and risk of stent-related complications associated with
endovascular repair, proponents argue that this is offset by the benefits gained from a min-
imally-invasive procedure in a population with significant co-morbidity and reduced life
expectancy. In our series the 5- and 10-year patient survival rates were 81% and 64%, respect-
ively. For the majority of patients in this series, endovascular repair would not represent a
sufficiently durable procedure.
References disease and results of surgical treatment.

J Vasc Surg 1991; 13: 398–407.
1. Varga ZA, Locke-Edmonds JC, Baird 3. Shortell CK, DeWeese JA, Ouriel K, Green
RN. A multicenter study of popliteal RM. Popliteal artery aneurysms: a 25-year
aneurysms. Joint Vascular Research Group. experience. J Vasc Surg 1991; 14: 771–6.
J Vasc Surg 1994; 20: 171–7. 4. Schellack J, Smith RB, Perdue GD Jr.
2. Dawson L, van Bockel JH, Brand R, Non-operative management of selective
Terpstra JL. Popliteal artery aneurysms. popliteal aneurysms. Arch Surg 1987;
Long-term follow-up of aneurysmal 122: 372–5.
5. Bowyer RC, Cawthorn SJ, Walker WJ, management and outcome. J Vasc Surg
Giddings AEB. Conservative management 1994; 19: 65–73.
of asymptomatic popliteal aneurysm. Br J 10. Tielliu IF, Verhoeven EL, Zeebregts CJ,
Surg 1990; 77: 1132–5. Prins TR, Span MM, van den Dungen JJ.
6. Galland RB, Magee TR. Management Endovascular treatment of popliteal artery
of popliteal aneurysm. Br J Surg 2002; aneurysms: results of a prospective cohort
89: 1382–5. study. J Vasc Surg 2005; 41: 561–7.
7. Davies RS, Wall M, Rai S et al. Long-term 11. Mohan IV, Bray PJ, Harris JP et al.
results of surgical repair of popliteal artery Endovascular popliteal aneurysm
aneurysm Eur J Vasc Endovasc Surg 2007; repair: are the results comparable to open
34: 714–8. surgery? Eur J Vasc Endovasc Surg 2006;
8. Schwarz W, Berkowitz H, Taormina 32: 149–54.
V, Gatti J. The preoperative use of 12. Antonello M, Frigatti P, Battocchio P
intraarterialthrombolysis for a thrombosed et al. Open repair versus endovascular
popliteal artery aneurysm. J Cardiovasc treatment for asymptomatic popliteal
Surg (Torino) 1984; 25: 465–8. artery aneurysm: results of a prospective
9. Carpenter JP, Barker CF, Roberts B, randomised study. J Vasc Surg 2005;
Berkowitz HD, Lusk EJ, Perloff LJ. 42: 185–93.
Popliteal artery aneurysms: current
Chapter
9
Femoral artery aneurysms
Robert Davies, Asif Mahmood and Rajiv Vohra
Key points
• Lower limb arterial aneurysms rarely occur in isolation
• The majority present either as an incidental finding or as lower limb ischaemia
• Aneurysm rupture is rare, but life- and limb-threatening when it occurs
• All symptomatic femoral artery aneurysms should undergo repair
• Asymptomatic femoral artery aneurysms >2.5 cm in maximum diameter should be
True femoral aneurysms

Incidence
Femoral artery aneurysm (FAA) is the second most common peripheral artery aneurysm
after popliteal artery aneurysms. They most often occur in male patients over 65 years with
an age-adjusted incidence of combined femoral and popliteal artery aneurysms of 7.39 per
100 000 population in the USA; male-to-female ratio of 10:1 [1]. Patients with FAAs demon-
strate a high incidence of cardiovascular disease and associated risk factors, including hyper-
tension, smoking and hypercholesterolaemia. Diabetes mellitus appears to be protective for
femoral artery aneurysm. Although FAAs predominantly occur in association with athero-
sclerotic disease, there are reports available in the English literature suggesting an associa-
tion with vasculitides and connective tissue disorders.
FAAs rarely occur in isolation and are frequently associated with a contralateral aneur-
ysm or/and aneurysmal disease affecting the aorta or other peripheral arteries. In a cohort
of 100 patients with FAAs Graham et al. reported 72% were bilateral, 85% were associated
with aorto-iliac aneurysms and 44% were associated with popliteal artery aneurysms [2].
Alternatively, the incidence of FAAs in patients with abdominal aortic aneurysms is ≈5%.
Pathophysiology
The pathophysiological conditions predisposing to the development of FAAs are poorly
understood. It has been postulated that, in conjunction with weakening of the arterial wall by
atherosclerosis, turbulent flow plays an important part. Constriction at the level of inguinal
ligament may promote the development of poststenotic dilatation with resultant turbulent
flow and pressure fluctuations affecting the common femoral artery. This localised change in
flow dynamics causes the arterial segment to vibrate, clinically evident as a bruit, weakening
the arterial wall predisposing it to aneurysmal change [2, 3]. This would be consistent with
the clinical finding that femoral artery aneurysms classically affect the common femoral
artery (CFA) in isolation without extending proximal to the inguinal ligament.
Symptoms and signs

Of patients with FAAs 30–40% are asymptomatic at the time of diagnosis with an incidental
finding of a smooth, fusiform and pulsatile mass. However, the majority of patients with
FAAs present with localised symptomatology or lower limb claudication/rest pain. In 20%
of patients localised pain in the groin or anterior thigh is the sole symptom.
Slowly enlarging FAAs may cause chronic compression of the adjacent neurovascular
structures. Up to 10% of patients demonstrate changes attributable to lower limb chronic
venous hypertension secondary to chronic femoral vein compression. However, chronic
dysthesia or paraesthesia, resulting from femoral nerve compression, is uncommon unless
associated with spontaneous FAA rupture.
Femoral artery aneurysms may be complicated by acute or chronic thromboembolism or
rupture in as many as 50% of cases at the time of initial presentation. The incidence of FAA
thrombosis varies in the literature although the incidence of chronic and acute thrombosis
appears to be similar. Acute thrombosis complicates up to 16% of femoral artery aneurysms
at the time of presentation. Patients present with an acutely ischaemic lower limb (Rutherford
classification IIb or III), clinically mimicking a femoral artery embolus. Inflow into both the
superficial femoral artery and profunda femoris artery is disrupted, thereby dramatically
reducing any potential source of collateral flow. Chronic thrombosis may mimic peripheral
occlusive disease in presentation, particularly as a high number of patients have concomi-
tant atherosclerotic cardiovascular disease.
Symptomatic distal embolisation is less common with a reported incidence of 2–8%
[1, 2]. This may not reflect the true incidence as a large proportion of distal emboli are
clinically silent. The majority of reports cite femoral artery aneurysm rupture as occurring
less frequently than thrombosis or distal embolisation with an incidence of <15%. This may
partly reflect the ‘protective’ nature of the tough femoral sheath [2].
Classification
Femoral artery aneurysms can be classified into type I and II according to the involvement
of the superficial femoral artery (SFA) and profunda femoris artery (PFA). Type I involves
the CFA in isolation, type II involves the CFA and one or both of the SFA and PFA.
Investigation
Only 20% of femoral artery aneurysms are reliably identified on clinical examination alone.
Those measuring < 2 cm are particularly unlikely to be identified by routine physical exami-
nation. A high index of suspicion is required depending on the circumstances in which
the patient presents. All patients with abdominal aortic aneurysms or peripheral artery
aneurysms should undergo formalised imaging assessment of their lower limb vasculature.
Conversely, all patients with femoral artery aneurysms should undergo formalised assess-
ment of their abdominal aorta and popliteal arteries.
Modern Duplex ultrasonography (DUS) is an adequate screening tool in both emergency
and elective settings. High quality images can be obtained of the aneurysm sac size, thrombus
Chapter 9: Femoral artery aneurysms 125
content and morphology, while other vascular beds can be screened for aneurysmal or occlu-
sive disease.
Upon confirmation of the diagnosis a suitable imaging modality is required to accurately
quantify the extent of thromboembolic disease locally and distally as well as any concomi-
tant aneurysmal disease identified by DUS. This should be dictated by the patients’ clini-
cal picture. In those patients suitable for elective investigation percutaneous angiography
remains the gold standard. Computed tomography and magnetic resonance angiography
are providing an increasingly accurate crural vessel assessment and thus are an alternative
to percutaneous angiography, particularly in those in whom aorto-iliac disease is suspected.
In the patient with an acutely ischaemic limb, lengthy preoperative investigation should not
prevent a limb-saving operation from being performed expeditiously. In these situations the
use of on table angiography is the key.
Management
It is generally accepted that all medically fit patients with symptomatic aneurysms or aneu-
rysms complicated with thromboembolism or rupture should be offered primary surgical
repair. Aneurysms that are enlarging on serial imaging should also be considered for elective
surgical repair.
The treatment of asymptomatic aneurysms remains controversial. Conflicting evidence
regarding the safety of conservative management exists in the literature. Proponents of a con-
servative management strategy cite the lack of evidence between aneurysm size and the devel-
opment of ischaemic complications. Furthermore, the risk of developing limb-threatening
complications in an asymptomatic aneurysm with conservative treatment at 2-year follow up
is less than 5% [2]. Most vascular surgeons would consider a femoral aneurysm measuring
>2.5 cm, or enlarging on serial imaging, for elective repair.
Surgical approach
The majority of type I femoral aneurysms are best repaired with a straight interposition graft
anastomosed proximally at the external iliac-common femoral artery junction and distally to
the common femoral artery bifurcation. This is performed from within the aneurysm with
the redundant sac plicated over the graft. The conduit utilised is dependent on individual
surgeon preference, but unless there is a suspicion of infection, autologous conduit offers
no benefit over prosthetic. A variety of configurations have been utilised for type II femoral
aneurysms, often based around the concept of an interposition graft extending distally into
either the SFA or PFA with re-implantation of the remaining artery, with or without an inter-
position graft.
Often femoral aneurysms occur in conjunction with aorto-iliac and popliteal aneurys-
mal or occlusive disease. In these situations the type of repair needs to be tailored to the
anatomy of the presenting pathology. In particular the most pressing disease needs to be
dealt with first whilst simultaneously planning for any second- or third-stage procedures.
Outcome
Perioperative mortality rates following femoral artery aneurysm repair are low although rise
to 5% when undertaken concomitantly with other aneurysm repairs [2]. Five-year patency
rates in asymptomatic patients approaches 90%. Patients presenting with lower limb claudi-
cation or rest pain/gangrene have worse patency rates at 5 years [2].
Profunda femoris artery aneurysms

Isolated profunda femoris artery aneurysms (PFAas) are rare and account for < 3% of femo-
ral aneurysms. Three-quarters of patients have aneurysmal disease affecting one other vas-
cular bed, most often the popliteal artery [4]. Due to the difficulty in identifying isolated
profunda femoris aneurysms at clinical examination, spontaneous rupture rates are higher
than for other peripheral aneurysms, with up to 50% of patients presenting with rupture.
Diagnosis is predominantly by DUS and formal assessment of the aorta-iliac and popliteal
arteries is recommended to identify concomitant aneurysms.
There is no current evidence regarding the optimal size at which PFAas should be
repaired, however many surgeons recommend considering repair for those >2 cm in an oth-
erwise fit patient. Those medically unfit for repair should be evaluated regularly with DUS
to assess for aneurysm expansion.
The primary aim of treatment is to prevent rupture. However one series reports ipsi-
lateral SFA occlusion in 45% of cases, therefore maintaining PFA blood flow is important
[4]. Vessel reconstruction with an interposition graft or aneurysm ligation and autologous
bypass is recommended.
Femoral artery pseudoaneurysms

Pathophysiology
Femoral artery pseudoaneurysms (FApAs) develop as a result of high pressure blood flow
through an arterial wall defect that becomes constrained by the perivascular soft tissue or
surrounding haematoma. Unlike a true aneurysm, the wall consists of thin fibrous tissue and
not all three tunica layers.
Iatrogenic, postcatheterisation pseudoaneurysm formation is the commonest cause
(Table 9.1) with a reported incidence of 1–6% in prospective sonographic studies [5, 6].
Infection, particularly in the drug abuser population, and anastomotic leakage may lead to
FApA formation. At a median follow up of 8 years, 2.5% of all femoral anastamoses and 8%
of aortobifemoral bypasses are complicated by FApA formation [7, 8].
Symptoms and signs

FApAs may present as a pulsatile groin mass that may or may not be painful. Enlargement
may cause compression of the adjacent femoral nerve or vein with resultant paraesthesia/
dysthesia or symptoms/signs of venous hypertension, respectively. Mycotic aneurysms pre-
sent as a painful, erythematous groin mass with or without purulent/sanguinous discharge.
Investigation
Duplex ultrasonography is the initial investigation of choice with a reported sensitivity and
specificity of >90%. Computed tomography angiography may be utilised as an adjunct to
ultrasound scan (USS), particularly in the assessment of patients with anastomotic pseu-
doanuerysms, in whom complex reconstructive surgery may be required.
Treatment
Most small postcatheterisation pseudoaneurysms have an indolent clinical course with the
majority undergoing spontaneous thrombosis. Pseudoaneurysm size and anticoagulation
Chapter 9: Femoral artery aneurysms 127
Table 9.1 Factors predisposing to postcatheterisation pseudoaneurysm formation
• Anticoagulation
• Hypertension
• Haemodialysis
• Obesity
• Sheath size >7 French
• Cannulation of superior femoral/profunda femoris artery
• Simultaneous puncture of artery and vein
• Heavily calcified arteries
• Short vessel compression time postcatheterisation
Toursarkissian B, Allen BT, Petrinec D, Thompson RW, Rubin BG, Reilly JM, Anderson CB, Flye MW, Sicard GA.
Spontaneous closure of selected iatrogenic pseudoaneurysms and arteriovenous fistulae J Vasc Surg 1997;
25: 803–8.
status of the patient are the two most important predictors of spontaneous thrombosis.
Toursarkissan et al. reported 87% of FApAs <3 cm having undergone spontaneous throm-
bosis at a mean follow up of 23 days[9]. Thus it is acceptable to initially treat a small pseu-
doaneurysm (<3 cm) conservatively with serial DUS evaluation. However the economic
burden of such an approach can be prohibitive and many institutions advocate active treat-
ment of all FApAs, irrespective of size.
The advent of minimally invasive percutaneous treatment methods has limited the
indications for open surgical repair to those FApAs that are either rapidly expanding,
infected, causing neurovascular compression, causing skin necrosis or have undergone
failed percutaneous treatment. The presence or absence of infection dictates surgical tech-
nique. For mycotic pseudoaneurysms surgery is aimed at excising infected tissue whilst
endeavouring to preserve the distal circulation. This may require the use of autologous
arterial bypasses routed outside the field of infection. In cases of gross suppurative infec-
tion primary ligation of the femoral vessels may become necessary with a 30% risk of
subsequent amputation. Non-infected pseudoaneurysms are repaired primarily with
interrupted non-absorbable sutures or with a vein patch-angioplasty, depending on the
size of the arterial defect.
In recent years minimally invasive methods of treating FApAs, and particularly post-
catheterisation FApAs, have been trialled. Ultrasound guided compression was the first
to gain widespread acceptance with success rates between 63 and 100% [10]. Patients
receiving anticoagulation with pseudoaneurysm sizes >3 cm are predictors of the fail-
ure of ultrasound-guided compression. Complications including rupture, femoral vein
thrombosis and acute limb ischaemia have been reported in 2–4% of patients [10].
Ultrasound guided percutaneous thrombin injection into the pseudoanuerysm sac is
now in widespread usage with reported success rates of >90%. This may be performed
in combination with an intra-arterial balloon protection device to prevent spillage of
thrombin into the native vessels, particularly in pseudoaneurysms with side necks.
Complications including distal embolisation and anaphylaxis are reported in <4% of
cases.
References diagnostic and therapeutic angiography.

Radiology 1995; 195: 463–6.
1. Lawrence PF, Lorenzo-Rivero S, Lyon 6. Lumsden AB, Miller JM, Kosinski AS.
JL. The incidence of iliac, femoral, and A prospective evaluation of surgically
popliteal artery aneurysms in hospitalized treated groin complications following
patients. J Vasc Surg 1995; 22: 409–15. percutaneous cardiac procedures. Am Surg
2. Graham LM, Zelenock GB, Whitehouse 1994; 60: 132–7.
WM Jr et al. Clinical significance of 7. Marković DM, Davidović LB, Kostić DM
arteriosclerotic femoral artery aneurysms. et al. False anastomotic aneurysms.
Arch Surg 1980; 115: 502–7. Vascular 2007; 15: 141–8.
3. Gow BS, Legg MJ, Yu W, Kukongviriyapan 8. Biancari F , Ylönen K, Anttila V et al.
U, Lee LL. Does vibration cause Durability of open repair of infrarenal
poststenotic dilatation in vivo and abdominal aortic aneurysm: a 15-year
influence atherogenesis in cholesterol-fed follow-up study. J Vasc Surg 2002; 35: 87–93.
rabbits? J Biomech Eng 1992; 114: 20–5. 9. Toursarkissian B, Allen BT, Petrinec D etal.
4. Harbuzariu C, Duncan AA, Bower TC, Spontaneous closure of selected iatrogenic
Kalra M, Gloviczki P. Profunda femoris pseudoaneurysms and arteriovenous
artery aneurysms: association with fistulae J Vasc Surg 1997; 25: 803–8.
aneurysmal disease and limb ischemia. J 10. Morgan R, Belli AM. Current treatment
Vasc Surg 2008; 47: 31–4. methods for postcatheterization
5. Katzenschlager R, Ugurluoglu A, Ahmadi pseudoaneurysms. J Vasc Interv Radiol
A. Incidence of pseudoaneurysm after 2003; 14: 697–710.
Chapter
10
Carotid, subclavian and vertebral
disease
A. Ross Naylor
Key points
• The extracranial arteries are prone to involvement with a number of important
atherosclerotic and non-atherosclerotic conditions
• The provision of best medical therapy should not be delegated to the most junior
member of the team. It is an essential component of care
• Patients with symptomatic carotid disease benefit from very rapid intervention.
Investigative strategies and rapid access to the operating theatre should be geared
to ensuring all patients are treated within 2 weeks of suffering their index symptom
• Relatively few patients with asymptomatic carotid disease benefit from intervention
(especially females and patients aged >75 years). It is essential that trials identify high-
risk subgroups
• The role of several technical aspects of carotid surgery has been guided by large
randomised trials
• Patients with symptomatic vertebral stenoses may have a much worse prognosis than
was previously thought
Radiation arteritis
Demographics
Despite the widespread use of radiotherapy in the treatment of head and neck cancers, there
is still insufficient data regarding the incidence of symptomatic radiation arteritis.
Pathology
• Acute phase: fibrin deposition/endothelial swelling followed by intimal necrosis.
• Subacute phase: endothelial regeneration with destruction of the internal elastic
lamina. Inflammatory cell infiltration of the media and adventitia.
• Chronic phase: intima becomes thickened with a tendency towards accelerated
atherosclerosis. The media and adventitia become progressively fibrotic.
Clinical features
False aneurysm or vessel rupture (acute/subacute phase), stroke due to carotid/vertebral
thrombosis, or upper limb ischaemia following subclavian thrombosis (sub-acute phase). In
the chronic phase (where radiation arteritis co-exists with atherosclerosis) thromboembolic
stroke/transient ischaemic attack (TIA) or upper limb claudication (subclavian stenosis/
occlusion) can occur.
Investigation
First line is Duplex ultrasound. Computed tomography angiography (CTA) or magnetic
resonance imaging (MRI)/contrast enhanced magnetic resonance angiography (CEMRA)
should also be performed in order to evaluate the full extent of the arteritic process (prox-
imally and distally) and to evaluate the status of the circle of Willis and to exclude recurrent
malignant disease.
Management
Management decisions must take account of (1) the overall health status of the patient and (2)
whether there is ongoing malignant disease, acute/chronic nature of arteritic process, age of
patient, mode of presentation, presence of tracheostomy, likelihood of cranial nerve injury,
likelihood of compromising skin flaps, ongoing embolisation using transcranial Doppler
(TCD).
In general, asymptomatic stenoses should probably be treated conservatively, unless
there is a non-functioning circle of Willis. Symptomatic lesions should be considered for
intervention (autologous reconstruction/carotid stenting). Stenting is currently the pre-
ferred option (especially in patients with a tracheostomy), although surgery may be safer in
patients with crescendo symptoms and evidence of spontaneous embolisation on TCD.
Prognosis
Little quality data are available regarding the prognosis in untreated patients.
Fibromuscular dysplasia [1]

Demographics
Fibromuscular dysplasia (FMD) is found in 0.5% of patients undergoing angiography for
symptomatic cerebral vascular disease, while 10% of patients suffering a carotid dissection
will have FMD in the contralateral internal carotid artery (ICA). Fibromuscular dysplasia is
bilateral in up to 60% of patients, 25% of patients with carotid involvement will have renal
artery FMD. One quarter of FMD cases involve the vertebral arteries, while up to 50% of
patients with carotid/vertebral FMD will also have an intracranial aneurysm.
Pathology
Fibromuscular dysplasia is a non-inflammatory, non-atheromatous segmental disorder.
There are three main subtypes based upon involvement of media, intima or adventitia.
Intimal fibroplasia (<10% of FMD cases) causes band-like narrowing or long smooth
stenoses and carries a poorer prognosis. Medial dysplasia (90% of FMD cases) causes
a beading appearance and is subdivided into medial fibroplasia (75–85%), perimedial
fibroplasia (<10%) and medial hyperplasia (<5%). Adventitial fibroplasia is seen in <1%
of cases and is characterised by dense collagen formation.
Chapter 10: Carotid, subclavian and vertebral disease 131
Clinical features
There are a wide spectrum of features, ranging from being asymptomatic (the majority)
to subarachnoid haemorrhage, embolic TIA/stroke, neurovascular hypertension or features
suggestive of carotid dissection (see later).
Investigation
Duplex remains the first-line investigation, but it cannot image the distal ICA/vertebral
arteries. If there is any clinical or Duplex suspicion of FMD, CTA/CEMRA should be per-
formed (preferably including the renal vasculature). The intracranial circulation should also
be imaged to exclude an aneurysm.
Management
There is no Level I, Grade A evidence to guide practice. Asymptomatic patients should prob-
ably be treated conservatively and be subject to serial surveillance. Carotid/vertebral angi-
oplasty is now the first-line treatment for patients with symptomatic FMD, with stenting
being reserved for patients with either a poor technical result or secondary dissection.
Prognosis
Multi-focal intimal fibroplasia carries the worst prognosis; medial dysplasia carries a rea-
sonably good prognosis. It is important, however, to remember that more than one vascular
bed can be affected with FMD and do not ignore the possibility of there being an intracranial
aneurysm.
Carotid artery dissection

Demographics
Carotid dissection (CD) causes relatively few strokes, but traumatic CD complicates about
1% of all head injuries and about 25% of trauma patients with an unexplained focal neuro-
logical deficit will have a dissection.
Pathology
Carotid dissection occurs spontaneously (Marfan’s/FMD), it can follow iatrogenic injury
(cannulation, angioplasty) or follow trauma (forced lateral rotation and hyper-extension,
which causes the ICA to be crushed between the skull base and the transverse process of C2).
The plane of dissection usually starts 2–3 cm beyond the origin of the ICA, but then extends
over a variable distance towards the skull base. Type 1 lesions cause a minor intimal irregu-
larity or stenosis <50%. Type 2 lesions are either associated with false aneurysm formation
or a stenosis >50%. In type 3 lesions, the dissection can cause complete vessel occlusion
through compression of the true lumen by a thrombosed false lumen.
Clinical features
Ipsilateral headache and/or neck pain occur in 80% of patients. Ocular signs and symptoms
are present in about 60% (miosis, painful Horner’s syndrome, hemianopia, ischaemic optic
neuropathy) along with III, IV or VI nerve palsies. Stroke and TIA may also complicate a
dissection (thrombosis or distal embolisation from the false lumen).
Investigation
Awareness of the potential diagnosis is integral to instituting effective treatment. Duplex
is generally limited by its inability to image the upper reaches of the carotid and vertebral
arteries. Here CTA or CEMRA is the investigation of choice and is also useful for serial sur-
veillance of small false aneurysms.
Management
There is little Level I, Grade A evidence. Traditionally, the majority of patients with CD have
been managed conservatively with bed rest and anticoagulation (heparin followed by war-
farin). Evidence suggests that a significant proportion will recanalise and clinically improve.
A few symptomatic patients have undergone attempts at surgical revascularisation, but this
has now been replaced by endovascular technology (covered stents, etc.), which also lessens
the likelihood of cranial nerve injury.
Prognosis
Stroke following undiagnosed CD carries a high risk of morbidity and mortality. The key to
minimising stroke risk is early awareness and rapid institution of anticoagulation.
Giant cell arteritis (GCA) [2]

Demographics
This is the most common vasculitis in the Western world and seen predominantly in women
aged >50 years (mean age of presentation is 75). Over one-third of GCA patients will have
polymyalgia rheumatica at the time of diagnosis. Giant cell arteritis predominantly affects
large (thoracic aorta) and medium-sized vessels (extracranial carotid branches, subclavian,
axillary and vertebral).
Pathology
An infiltrate of lymphocytes and macrophages involving the entire vessel wall is typically seen.
The characteristic feature is fragmentation of the internal elastic lamnia. Giant cells are often
present.
Clinical features
Three syndromes may co-exist:
1. Systemic inflammatory syndrome: non-specific constitutional symptoms including
arthralgias, myalgias, anorexia, weight loss and night sweats.
2. Cranial arteritis: localised vasculitis of the carotid and vertebral arteries causing
headache/facial pain, scalp tenderness, jaw claudication, hoarseness and visual loss.
3. Large vessel vasculitis: symptoms secondary to stenotic/occlusive disease of the
subclavian and axillary arteries (arm claudication, Raynaud’s phenomenon).
Giant cell arteritis involvement of the thoracic aorta predisposes to aneurysm forma-
tion. Cerebrovascular events (stroke/TIA) occur in 3–4% of GCA patients and follow
inflammatory occlusion of the vertebral/carotid arteries. Episodes of transient visual loss

precede permanent visual loss in 50% of untreated patients within 7 days and are usually
due to inflammatory occlusion of the short posterior ciliary arteries, causing ischaemia of
the optic disc and choroids.
Investigations
Temporal artery biopsy should be performed as soon as possible (it will be negative in 50%
of patients with large vessel vasculitis). Duplex ultrasound will identify stenoses/occlusions,
but CEMRA or CTA is preferred in order to evaluate aortic arch and major branch vessels
(looking for bilateral patterns of disease, thoracic aneurysm) as well as excluding intrac-
ranial vasculitis. Laboratory investigations should include erythrocyte sedimentation rate
(ESR) (85% of GCA patients will have an ESR >50 mm h–1), C-reactive protein (CRP) (98%
sensitivity for active GCA), thrombocytosis (present in 48% of biopsy-positive patients) and
normocytic normochromic anaemia.
Management
High-dose steroid therapy (intravenous or oral), which is then gradually reduced over 6–12
months with titration against CRP and other inflammatory markers.
Prognosis
Visual loss persisting for >24 hours tends to be permanent. High dose steroid therapy is then
aimed at preventing visual loss in the other eye. Patients with thoracic aneurysm require
serial monitoring.
Takayasu arteritis (TA) [3]

Demographics
Pan-arteritis of unknown aetiology occurs, affecting the aorta and its main branches and
seen predominantly in young females, especially from the Orient. Overall incidence is about
2.6 cases per million per year.
Pathology
The inflammatory process starts with inflammatory infiltration (lymphocytes and occa-
sional giant cells) around the vasa vasorum, extending transmurally. If disease progression
is rapid, aneurysm formation becomes more likely. In the later stages of the condition, there
is progressive fibrosis, leading to occlusion (hence the term ‘pulseless’ disease). The common
carotid arteries are involved in 65–75% of cases, the subclavian arteries in 50–75%, while the
vertebral arteries are involved in 6–10%.
Clinical features
The American College of Rheumatology has determined that a diagnosis of TA requires at
least three of the following six criteria to be met: (1) age at onset <40 years; (2) claudication
of extremities; (3) decreased brachial artery pulse; (4) blood pressure difference >10 mmHg
between arms; (5) bruit over subclavian arteries or aorta; and (6) arteriogram abnormality.
Given the large number of arteries that can be subject to aneurysm formation, stenosis or
occlusion, the mode of presentation will vary considerably. In the early stage most complain
of constitutional symptoms (fatigue, malaise etc.). In the second stage, symptoms are related
to the increasing inflammatory reaction: (1) systemic inflammatory response (fatigue, fevers,
extremity pain, headache, rashes); (2) vascular insufficiency (claudication, arm numbness,
TIA/stroke/amaurosis fugax); or a combination of (1) and (2). In the final ‘burned out’ stage
of the disease, the inflammatory reaction is replaced with transmural fibrosis.
Investigations
Laboratory investigations (ESR, thrombocytosis, anaemia) show whether there is an under-
lying inflammatory reaction and also monitor the effect of treatment. Conventional angiog-
raphy has been replaced with CTA and CEMRA for evaluating the anatomical extent of the
inflammatory process. These imaging modalities also permit evaluation of aortic aneurysm
formation and measurement of arterial wall thickness and extent of oedema, which can also
be used for monitoring the effect of treatment.
Management
High-dose steroid therapy (intravenous then oral) is used, which is reduced over 6–12 months
with titration against CRP and other inflammatory markers. If it proves difficult to reduce the
steroid dose or adverse side effects occur, methotrexate, cyclophosphamide and azathioprine
are alternative immunosuppressive agents. Patients with renovascular hypertension require
aggressive treatment and patients should receive antiplatelet therapy unless contraindicated.
There is relatively limited experience with angioplasty or stenting in TA. Surgical revascu-
larisation of the carotid, subclavian and vertebral arteries is also rarely required. If revascu-
larisation becomes necessary, try to avoid this in the acute phase of the condition. It is also
important to perform an inflow from the aorta rather than the subclavian artery.
Prognosis
Provided medical treatment is initiated early, approximately 90% of patients will survive
5 years.
True and false aneurysms

Incidence
Aneurysms of the carotid, vertebral and subclavian arteries comprise <2% of all arterial
aneurysms.
Pathology
The prevalence of true or false aneurysms will reflect differing patient populations. South
African studies tend to describe aneurysms in predominantly young men with infection
(HIV, tuberculosis). In metropolitan areas, many will be secondary to trauma (gunshot,
knife), while in studies reporting outcomes from urban populations, the majority of true
aneurysms are classed as ‘atherosclerotic’. While the commonest cause of a false aneurysm is
prosthetic patch infection after carotid endarterectomy.
Clinical features
The commonest clinical feature is a pulsatile neck mass, followed by detection of a cer-
vical bruit. Aneurysm rupture is extremely rare. By contrast, TIA/stroke is a relatively
common presentation (presumably secondary to thromboembolism), as are cranial nerve
signs/symptoms due to direct compression. Patients with false aneurysms will present with
related symptoms and signs (e.g. history of recent trauma, evidence of prosthetic patch
infection).
Investigation
The majority presenting with pulsatile neck ‘masses’ will have coiling/ectasia of the com-
mon carotid or innominate arteries. Accordingly, the first-line investigation is Duplex ultra-
sound. Thereafter, investigations are directed towards determining the likely underlying
cause (FMD, trauma, etc.) as management strategies will vary. Second-line investigations
include CTA or CEMRA, which can rapidly image other arteries (e.g. renal arteries in sus-
pected FMD). They will also provide information regarding the feasibility of endovascular
treatment.
Management
Management will depend upon the underlying aetiology, urgency of symptoms and the
level/distal extent of the aneurysm. Operative strategies include: (1) proximal/distal liga-
tion; (2) open reconstruction (venous bypass, partial aneurysm excision, patch angioplasty,
prosthetic bypass, resection and end-to end-bypass); and (3) endovascular repair (stent
graft exclusions, carotid stenting augmented with coil exclusions, endovascular balloon
occlusion). In the carotid and vertebral circulations, ligation should only be considered
if reconstructive options have been excluded. Inflation of an endovascular balloon within
the artery under local anaesthesia may assist in determining whether ligation will be
tolerated.
Prognosis
Most patients with carotid (vertebral) aneurysms will become symptomatic with time.
Surgery, however, carries the risk of procedural stroke and cranial nerve injury in 5–7%
of patients. Management decisions must therefore balance the risks and benefits associ-
ated with intervention, which may mean adopting a more conservative strategy in selected
asymptomatic patients with small distal ICA aneurysms.
Carotid body tumour [4]

Demographics
The carotid body is a collection of chemoreceptor cells responsible for detecting changes
in blood oxygen/carbon dioxide levels and pH, and is located within the adventitia of the
carotid bifurcation. Carotid body tumours (CBTs) may be sporadic or familial. Familial
CBTs (10% of all CBTs) are more common in females and are more likely to be bilateral.
Carotid body tumours represent <1% of all neck tumours.
Pathology
Carotid body tumours are highly vascular tumours derived from the neural crest ectoderm
and are the commonest type of cervical paraganglionoma (glomus vagale, glomus jugulare,
glomus tympanicum). Most present in the fourth/fifth decades and cause characteristic
splaying of the bifurcation. By contrast, glomus vagale tumours (second commonest) cause
splaying of the ICA and ECA above a normal bifurcation. Approximately 5% of CBTs are
bilateral, 5% will be locally malignant and 5% systemically malignant.
Clinical features
Asymptomatic neck swelling is the most common feature. Larger lesions cause pain, cra-
nial nerve palsies (XII, IX, X) and rarely a Horner’s syndrome. Stroke and TIA are unusual,
while some will present with a neuroendocrine-mediated syndrome with flushing, dizziness,
arrhythmias and hypertension.
Investigations
Consider CBT in all patients with lateral neck swellings and definitely before any decision is
made to undertake an open biopsy. Duplex ultrasound will show the characteristic blush of
hypervascularity within a splayed carotid bifurcation. Cross-sectional imaging will provide
information regarding the upper and lower limits of the lesion, which is useful in plan-
ning resection strategies (i.e. do you need to plan for a high approach to the carotid artery).
Computed tomography and MRI are also useful in excluding bilateral lesions. Radionuclide
imaging and conventional angiography are not routinely necessary.
Management
Resection is the main treatment strategy and is typically described as being performed in
a subadventitial plane. A conservative approach is indicated in elderly patients with small
lesions. Occasionally, it may be preferable to resect the tumour and carotid bifurcation
(adherent tumour, carotid injury, suspicion of malignancy) and perform an interposition
bypass. Perioperative bleeding may be reduced by preoperative embolisation of ECA feed-
ing vessels or insertion of a covered stent within the first few centimetres of the ECA. This
strategy is probably only necessary in large lesions.
Prognosis
Resection carries a 1% mortality rate and a 2–3% risk of stroke. Cranial nerve injuries are
not uncommon, but tend to be transient. Provided a macroscopically complete excision has
been performed, the risk of recurrence is <5%.
Carotid occlusive disease

Demographics
Stroke (responsible for 12% of UK deaths) is defined as a focal (occasionally global) loss of
cerebral function, which lasts for >24 hours and has a vascular cause. A TIA carries a times-
cale of <24 hours. The incidence of first-ever stroke is 2.4/1000, but increases with age. The
annual incidence of TIA is 0.5/1000.
Pathology
Approximately 80% of all strokes are ischaemic (20% are haemorrhagic), while approximately
80% of ischaemic strokes affect the carotid territory. The main causes of carotid territory
ischaemic stroke include thromboembolism of the ICA and/or middle cerebral artery (50%),
small vessel occlusion of the deep penetrating end-arteries (25%), cardiogenic embolism
(15%), haematological disorders (myeloma, polycythaemia, thrombocytosis) in 5%, whilst 5%
have a miscellany of causes (tumour, arteritis, oral contraceptive, etc.). Risk factors for stroke
include hypertension, ischaemic heart disease, smoking, hyperlipidaemia, TIA, diabetes and
hyperfibrinogenaemia. The commonest single cause of ischaemic, carotid territory stroke is
thromboembolism from an atherosclerotic plaque at the origin of the ICA. The carotid bifur-
cation is prone to atherosclerosis, particularly on the outer aspect of the bulb. Some of these
plaques then undergo an acute change (plaque rupture, intraplaque haemorrhage), which
predisposes towards overlying thrombus formation and embolisation to the brain.
Clinical features
Asymptomatic disease
Of the population 10% will have an asymptomatic >50% ICA stenosis, but only 1% will have
a stenosis >70%. Asymptomatic stenoses are usually detected by auscultation of a bruit, with
the patient complaining of pulsatile tinnitus or by ultrasound. However, the term ‘asymp-
tomatic’ may be misleading as many patients do not consider a transient episode of hand
paraesthesia or weakness to be important and it could go unreported. Similarly, because
33% of our lives are spent sleeping, nocturnal TIAs will go unreported. Approximately 25%
of asymptomatic patients will have ischaemic brain injury on CT/MRI.
Symptomatic disease
‘Classical’ carotid territory symptoms include: hemisensory/motor signs; higher cortical
dysfunction (dysphasia, visuospatial neglect); and monocular blindness. There has previ-
ously been a tendency to ascribe a diagnosis of ‘non-hemispheric’ symptoms to patients with
blackouts, isolated diplopia, isolated vertigo, isolated dizziness, presyncope and syncope. In
practice, these should never be considered to be carotid (or vertebrobasilar in origin) unless
they coexist with more typical symptoms.
Investigations
Baseline investigations include biochemistry, lipids, glucose, full blood count, plasma vis-
cosity, chest X-ray and electrocardiogram (ECG). More specialised investigations (throm-
bophilia screening, autoantibodies, homocysteine levels, echocardiography and 24-hour
tapes) should be reserved for selected cases.
Routine catheter angiography (previously the gold standard) is not now indicated (radia-
tion, 1–2% stroke risk), having been replaced by non-invasive alternatives. Table 10.1 sum-
marises the sensitivity and specificity for Duplex, CTA (excluding multislice CT), MRA and
CEMRA from a recent systematic review [5]. Overall, CEMRA emerged as the best inves-
tigation, but it remains limited by accessibility and the potential for gadolinium-induced
nephrogenic systemic fibrosis. In practice, each imaging modality has an important role
because investigations in patients being worked up for carotid endarterectomy (CEA) are
different to those should the patient be considered for carotid artery stenting (CAS).
Table 10.1 Results of a meta-analysis of the accuracy of non-invasive imaging for all stenosis groups and
imaging modalities
Stenosis group (%) Imaging Sensitivity Specificity (%)

70–99 US 89 84
CTA 77 95
MRA 88 84
CEMRA 94 93
50–69 US 36 91
CTA 67 79
MRA 37 91
CEMRA 77 97
0–49, 100 US 83 84
CTA 81 91
MRA 81 88
CEMRA 96 96
US, ultrasound; CTA, computed tomography angiography; MRA, magnetic resonance angiography; CEMRA,
contrast enhanced magnetic resonance angiography.
Wardlaw JM, Chappell FM, Stevenson M et al. Accurate, practical and cost-effective assessment of carotid
stenosis in the UK. Health Technol Assess 2006; 10: no. 30. Department of Health Crown copyright material is
reproduced with the permission of the Controller of the HMSO and Queen’s Printers for Scotland. Document
available at: http://www.hta.ac.uk/fullmono/mon1030.pdf. Reproduced with permission.
Duplex remains the first-line investigation. In centres with internal validation, it is

acceptable for surgery to be performed on the basis of ultrasound provided the first scan has
been corroborated by a second (using a different technologist). In other hospitals, Duplex
findings are corroborated with either CTA or MRA, especially if there is any question of
inflow/outflow disease or excessive calcification. The most important advantage of CEMRA
and CTA is that it can simultaneously image the arch, great vessel origins, distal ICA and
intracranial circulation (a prerequisite for CAS). The main disadvantage of CEMRA and
CTA is that neither are as accessible as Duplex. Unless hospitals are prepared to provide
more ‘single visit’ facilities for CT and MRI, Duplex will remain the first-line investigation.
Management
Table 10.2 summarises what is currently considered to be ‘optimal medical therapy’ in
patients presenting with symptomatic and asymptomatic carotid disease [6]. In addition to
risk factor control and initiating medical therapy, selected patients will benefit from a more
invasive intervention (CEA or CAS).
Symptomatic carotid disease

Table 10.3 summarises the 5-year findings from the Carotid Endarterectomy Trialists
Collaboration (CETC) who combined data from the European Carotid Surgery Trial (ECST),
North American Symptomatic Carotid Endarterectomy Trial (NASCET) and Veterans
Adminstration (VA) trials (> 6000 patients), having remeasured the pre-randomisation
Table 10.2 European Stroke Initiative recommendations for what constitutes ‘best medical therapy’ in patients
with asymptomatic and symptomatic carotid disease [6]
Level of evidence
Treatment Asymptomatic Symptomatic
BP<140/90 mmHg or <130/80 mmHg I I
in diabetics
Glycaemic control to prevent other diabetic III III
complications
Statin therapy I I
Stop smoking II II
Avoid heavy consumption of alcohol I I
Regular physical activity II II
Low salt, low saturated fat, high fruit and II II
vegetable diet rich in fibre
If BMI elevated, reduce weight II II
HRT should not be used for stroke prevention I I
in women
Aspirin To prevent MI level IV I
Aspirin and dipyridamole Not recommended Level IV I
Clopidogrel Not recommended Level IV I
BMI, body mass index; HRT, hormone replacement therapy; MI, myocardial infarction.
Table 10.3 Effect of delay to surgery on overall benefit conferred by carotid endarterectomy
Strokes prevented
per 1000 CEAs at
Stenosis group Delay (weeks) ARR (%) NNT 5 years
50–69 <2 14.8 7 148
2–4 3.3 30 33
4–12 4.0 25 40
>12 –2.9 Nil Nil
70–99 <2 30.2 3 302
2–4 17.6 6 176
4–12 11.4 9 114
>12 8.9 11 89
ata recalculated from the Carotid Enarterectomy Trialists Collaboration (CETC) [8] and excludes patients
D
with ‘near occlusion’. ARR, absolute risk reduction in 5-year risk of ipsilateral stroke conferred by carotid
endarterectomy (CEA) over best medical therapy. NNT, number needed to treat to prevent 1 ipsilateral stroke
at 5 years. Strokes prevented is the number of ipsilateral strokes prevented at 5 years by performing 1000
CEAs.
angiograms using the NASCET method [7]. Carotid endarterectomy conferred no bene-
fit in symptomatic patients with 0–50% stenoses. A small, but significant benefit was seen
in patients with 50–69% stenoses, while maximum benefit was present in those with more
severe degrees of stenosis.
Secondary analyses from ECST, NASCET and the CETC have provided informa-
tion regarding which patients gain most (and least) benefit from intervention. Markers of
increased benefit include (1) males versus females, (2) increasing age, especially >75 years,
(3) hemispheric versus ocular symptoms (4) cortical versus lacunar stroke, (5) increasing
medical co-morbidity, (6) very recent symptoms, especially the first 2 weeks, (7) irregular
versus smooth plaques, (8) increasing degrees of stenosis (not subocclusion) (9) contralateral
occlusion, (10) tandem intracranial disease and (xi) a failure to recruit intracranial collater-
als. Patients with subocclusion (string sign) derived no apparent benefit from intervention.
One of the most striking predictors of benefit is speed to treatment. It has previously
been taught that the 7-day risk of stroke after TIA/minor stroke is 1–2% (2–4% at 30 days).
These data, along with a perception that procedural risks increase the quicker one inter-
venes, have made surgeons reluctant to intervene quickly. However, recent evidence sug-
gests that the 7-day risk of stroke may be as high as 8%. Table 10.4 summarises CETC data
regarding outcomes stratified for delays to surgery and show (unequivocally) that any delay
significantly reduces the long-term benefit accrued to the patient [8].
There has also been controversy about how long CEA should be delayed after suffering
a stroke (traditionally 6–8 weeks). Evidence now suggests that expedited surgery can be
undertaken safely in patients who meet the following criteria: rapid neurological recovery/
neurological plateau, no carotid occlusion, Rankin 0–2 in terms of disability, area of infarc-
tion less than one third of the middle cerebral territory, no intracranial haemorrhage and
patients should be lucid and able to give informed consent.
Carotid artery stenting has now emerged as an alternative to CEA in the management
of symptomatic carotid disease. The 2007 Cochrane Review of 12 published randomised
trials observed that while the prevention of cranial nerve injury significantly favoured
CAS, stenting was associated with a significantly higher 30-day risk of death/stroke. The
International Carotid Stenting Study (ICSS) was a multicentre randomised controlled trial
in 1713 patients. The results showed that stenting was associated with twice as many stokes
when compared to surgery (7.7% vs 4.1%, 65 vs 35, p = 0.002), most of these strokes in the
stenting group were, however, non-disabling stoke, death of periprocendural myocardial
infarction (MI) was higher in the stentin group (8.5% vs 5.2%, 72 vs 44, p = 0.006) [9].
The Carotid Revascularisation Endarterectomy Stenting Trial (CREST) results were recently
published. In this trial symptomatic and asymptomatic carotid artery 2502 stenosis patients
were randomised to stenting and surgery and were followed up for two and half years. There
was no significant difference in the rate of stroke, MI or death. However, during the peripro-
cedural period there was a higher risk of stroke with stenting and a higher risk of MI with
endarterectomy. Younger patients had slightly fewer events with stenting and older patients
fewer events with surgery [10].
Asymptomatic carotid disease

Table 10.3 summarises outcomes from Asymptomatic Carotid Artery Stenosis (ACAS) and
Asymptomatic Carotid Surgery Trial (ACST) who compared CEA with best medical ther-
apy in asymptomatic patients with 60–99% stenoses. The ‘headline’ conclusion was that
Table 10.4 Five-year prevention of ‘any’ stroke in the symptomatic and asymptomatic randomised trials comparing carotid endarterectomy (CEA) with medical therapy
5-year risk
30-day Strokes prevented
Trial Stenosis n CEA risk Surgery Medical ARR RRR NNT per 1000 CEAs
Symptomatic trials*
CETC <30 1 746 No data 18.36 15.71 –2.6 n/b n/b None at 5 years
30–49 1 429 6.7 22.80 25.45 +2.6 10 38 26 at 5 years
50–69 1 549 8.4 20.00 27.77 +7.8 28 13 78 at 5 years
70–99 1 095 6.2 17.13 32.71 +15.6 48 6 156 at 5 years
string 262 5.4 22.4 22.30 –0.1 n/v n/b None at 5 years
Asympotomatic trials**
ACAS 60–99 1 659 2.3 17.5 12.4 +5.1 29 20 51 at 5 years
ACST 60–99 3 120 2.8 11.8 6.4 +5.4 46 19 54 at 5 years
* Data derived from the Carotid Endarterectomy Trialists Collaboration (CETC) [7], which combined data from ECST, NASCET and the VA trial. All pre-randomisation
angiograms were remeasured using the NASCET method.
** ACAS data cited here are for the 5-year risk of ‘any’ stroke, in order to permit direct comparison with ACST.
n/b, no benefit conferred by CEA; ARR, absolute risk reduction; RRR, relative risk reduction; strokes prevented per 1000 CEAs, number of strokes prevented at 5 years. by
performing 1000 CEAs; NNT, number of operations to prevent one stroke at 5 years.
CEA conferred a significant reduction in the 5-year risk of stroke. More importantly, ACST
showed that CEA conferred a significant reduction in fatal/disabling stroke and that it was
not beneficial in patients aged >75 years. Neither ACAS nor ACST showed any relationship
between stenosis severity or bilateral severe disease and 5-year stroke risk. Note that the data
published in Table 10.3 are slightly different to those normally published as they specific-
ally refer to the 5-year risk of ‘any stroke’. The 5-year risks of ipsilateral stroke in ACAS were
11.0% (medical) and 5.9% (surgical).
The management of patients with asymptomatic carotid disease continues to arouse con-
troversy, largely because it is still not possible to identify who benefits most (and least) from
intervention. For example, neither trial showed that women gained significant benefit from
CEA (ACST only demonstrated benefit if the operative risk was excluded). Moreover, cost-
effectiveness analyses suggest that up to 94% of patients ultimately undergo an unnecessary
intervention, whilst incurring massive costs to health systems [11]. Finally, there is emerging
evidence that the natural history risk of stroke is diminishing with time, presumably due
to improvements in ‘optimal medical therapy’. It is imperative, therefore, that we identify
high risk cohorts in whom to target therapy. At present, it is not appropriate to offer CAS to
otherwise normal risk asymptomatic patients without participation in randomised trials.
Performance of CEA
Carotid endarterectomy has been subject to more scientific scrutiny than any other surgical
procedure, including a number of randomised trials aimed at evaluating different aspects of
the procedure. The principal findings will be summarised below.
• The Aspirin and Carotid Endarterectomy (ACE) trial showed that low-dose aspirin
(75–300 mg) conferred significant reductions in early and late death/stroke compared
with higher doses (650–1300 mg). This trial was performed in response to a NASCET
subgroup analysis suggesting the converse to be true. Low-dose aspirin is currently the
preferred choice and should not be stopped perioperatively.
• The GALA trial showed that there was no evidence that performing CEA under general
or locoregional anaesthesia influenced outcome, quality of life, hospital stay, intensive
therapy unit (ITU) stay or costs. Surgeons and anaesthetists may use either anaesthetic
technique according to their preference.
• A Cochrane meta-analysis of six randomised trials showed that routine patching
conferred a threefold reduction in the 30-day risk of death/stroke and thrombosis and
a similar reduction in late stroke/restenosis compared with routine primary closure.
No trial has compared selective with routine patching. The evidence favours a policy of
routine patching over routine primary closure and there is no evidence that patch type
(vein/prosthetic) influences outcome.
• A meta-analysis of five randomised trials showed that provided the arteriotomy
is patched, eversion endarterectomy does not confer any additional benefit over
traditional endarterectomy.
• A meta-analysis of two randomised trials showed that routine shunting conferred a
non-significant 25% reduction in the 30-day risk of death/stroke compared with no
shunting. These studies were, however, methodologically flawed and this remains an
unresolved issue. It is intuitively hard to defend a policy of ‘never shunting’. Unless
surgeons are prepared to perform CEA under locoregional anaesthesia, there is no safe
or reliable way of predicting who needs a shunt.
Subclavian occlusive disease

Pathology
The commonest cause of occlusive/stenotic disease of the subclavian artery is atheroscler-
osis, usually at its origin. Other important conditions include arteritis (GCA and TA), thor-
acic outlet compression, occasionally FMD and rarely aneurysms of the subclavian artery.
Clinical features
Acute obstruction (embolus/thrombosis) of the subclavian artery may cause acute ischae-
mia of the upper limb, as well as a posterior circulation stroke due to compromised flow in
the ipsilateral VA. The more common presentation is pain in the forearm with exercise (clau-
dication) or dizziness while using the ipsilateral limb. The latter condition (subclavian steal
syndrome) is due to a temporary reduction in flow in the posterior cranial circulation due to
reversed flow in the ipsilateral vertebral artery (VA) during arm exercise. A related condition
(coronary steal syndrome) is now being increasingly reported with the trend towards using
the internal mammary artery as a conduit for coronary bypass. In the presence of a proximal
subclavian stenosis/occlusion, there may be reversed flow in the internal mammary artery
graft during arm exercise, which can be sufficient to precipitate angina or breathlessness.
Investigations
Duplex remains the first-line investigation (it is very accessible and cheap), supplemented by
CTA or CEMRA. Patients suspected of having arteritis or FMD require additional investiga-
tions (see earlier).
Management
Risk factor modification and statin/antiplatelet therapy should be instituted. A significant
proportion of patients with subclavian occlusive disease can be managed conservatively,
especially if it is discovered as an incidental (asymptomatic finding). The decision to inter-
vene must be based on the mode of presentation and extent of disability (pain, employ-
ment, etc.) in the context of the potential risks (surgery and angioplasty carry a small but
significant risk of stroke). Compelling indications include severe arm pain with exercise
that compromises employment, vertebrobasilar symptoms at presentation, subclavian steal
syndrome and coronary steal syndrome. Less compelling indications include mild dizzi-
ness with head movement and minor forearm claudication (especially in the non-dominant
limb). In the past, surgical revascularisation was the cornerstone of management (carot-
id-subclavian bypass, transposition), but many centres now increasingly use angioplasty
+/- stenting as the first-line option. There is no Level I Grade A evidence to guide practice
in this situation. In general, surgery carries a slightly higher initial risk but probably offers
better long-term durability. Conversely, endovascular interventions are less invasive (and
less risky), but long-term patency rates may be slightly poorer.
Vertebral occlusive disease

Demographics
Approximately 20% of all ischaemic strokes are vertebrobasilar.
Pathology
The commonest aetiology is atherosclerosis, but dissection, arteritis and FMD must be
considered in patients with appropriate symptoms. In a recent population-based study of
CEMRA imaging in 151 patients presenting with vertebrobasilar symptoms, approximately
26% were found to have significant disease (>50% stenosis) in the vertebral or basilar arter-
ies [12]. Of those with demonstrable disease, 62% were located in the extracranial VA, 11%
were located in the intracranial VA, while 8% of lesions were found in the basilar artery.
Interestingly, in those patients presenting with vertebrobasilar symptoms and who had a
significant with extracranial VA stenosis, 69% had their stenoses either at the origin or near
to the origin of the VA, while 31% had lesions in the upper third of the VA.
Clinical features
‘Classical’ vertebrobasilar symptoms include; bilateral sensory/motor symptoms and signs,
hemisensory/motor symptoms and signs (seen in 10% of vertebrobasilar events), bilateral
visual loss (cortical blindness), dysarthria, nystagmus, and problems with gait and stance.
Patients with isolated non-hemispheric symptoms (without other more definite vertebro-
basilar symptoms) should not be considered to have suffered posterior circulation-based
symptoms. Similarly, it has become conventional to make a diagnosis of ‘positional’ verte-
brobasilar ischaemia in patients who develop symptoms (dizziness, vertigo) on lateral or
flexed head movements. Recent evidence suggests that the vast majority of these patients do
not have a vertebrobasilar cause for their symptoms. Such a diagnosis should only be made
after comprehensive investigation as most will have inner ear pathology.
Investigations
Duplex ultrasound remains the first-line investigation but it is limited by its ability to fully
evaluate the extracranial VA. Transcranial Duplex can image the intracranial VA and basilar
arteries, but it is usually preferable to undertake either CEMRA or CTA. This will provide
comprehensive information regarding the arch, subclavian/vertebral origins as well as the
intracranial vessels and circle of Willis. The other advantage of CT and MRI is that it will
provide valuable information regarding structural cranial abnormalities (tumour, infarc-
tion, A-V malformation). However, extracranial and transcranial Duplex is an excellent
method for seeing whether head movements cause any change in flow in the extracranial VA
or posterior cerebral artery.
Management
Most centres in the UK have relatively little experience of treating isolated vertebral and
basilar artery stenoses. In the past, the mainstay of management was medical therapy or
surgery (vertebral patch angioplasty, transposition to the carotid artery and proximal/dis-
tal bypass). Modern endovascular technological advances now offer an alternative means
of treatment (angioplasty, stenting), which is emerging as the new first-line intervention in
most centres. All patients will, of course, require optimisation of risk factors, and antiplate-
let/statin therapy.
Prognosis
It was previously thought that patients with vertebrobasilar symptoms faced a lower
risk of late stroke than those with carotid artery disease. However, emerging data from
opulation-based studies suggest that this assumption may be erroneous. The Oxford
p
group have recently shown that 46% of patients with a recently symptomatic VA stenosis
suffered either a recurrent TIA or stroke in the first 90 days after presentation [12]. This
compares with only 21% in similar patients with no VA stenosis.
References controlled trials of endarterectomy for

1. Olin JW. Recognizing and managing symptomatic carotid stenosis. Lancet
fibromuscular dysplasia. Cleveland Clin J 2003; 361: 107–16.
Med 2007; 74: 273–82. 8. Rothwell PM, Eliasziw M, Gutnikov SA,
2. Kawasaki A, Purvin V. Giant cell Warlow CP, Barnett HJM for the Carotid
arteritis: an updated review. Acta Ophth Endarterectomy Trialists Collaboration.
2009; 87: 13–32. Endarterectomy for symptomatic carotid
3. Johnston SL, Lock RJ, Gompels MM. stenosis in relation to clinical subgroups
Takayasu arteritis: a review. J Clin Pathol and timing of surgery. Lancet 2004;
2002; 55: 481–86. 363: 915–24.
4. Sajid MS, Hamilton G, Baker DM on 9. International Carotid Stenting Study
behalf of the Joint Vascular Research Investigators. Carotid artery stenting
Group. Eur J Vasc Endovasc Surg 2007; compared with endarterectomy in patients
34: 127–30. with symptomatic carotid stenosis. Lancet
5. Wardlaw JM, Chappell FM, Stevenson M 2010; 375(9719): 985–97.
et al. Accurate, practical and cost-effective 10. Brott TG, Hobson RW, Howard G et
assessment of carotid stenosis in the al. Stenting versus endarterectomy for
UK. Health Technol Assess 2006; 10: no. treatment of carotid artery stenosis. N Engl
30. Available at: http://www.hta.ac.uk/ J Med 2010; 363(1): 11–23.
fullmono/mon1030.pdf. 11. Naylor AR, Gaines PA, Rothwell PM.
6. The European Stroke Initiative Who benefits most from intervention for
Executive Committee and the EUSI asymptomatic carotid stenosis: patients
Writing Committee. European Stroke or professionals? Eur J Vasc Endovasc Surg
Initiative Recommendations for Stroke 2009; 37(6): 625–32.
Management – Update 2003. Cerebrovasc 12. Marquardt L, Kuker W, Chandratheva A,
Dis 2003; 16: 311–37. Geraghty O, Rothwell PM. Incidence and
7. Rothwell PM, Eliasziw M, Gutnikov SA prognosis of >50% symptomatic vertebral
et al. for the Carotid Endarterectomy or basilar artery stenosis: prospective
Trialists Collaboration. Analysis of population based study. Brain 2009;
pooled data from the randomised 132: 982–8.
Chapter
11
Diagnosis and management of
thoracic outlet syndrome
Hassan Badri and Vish Bhattacharya
Key points
• Th
oracic outlet syndrome (TOS) can be neurogenic, venous or arterial
• Neurogenic TOS is the commonest presentation, seen in 90% of cases
• Arterial presentation is very rare but may be more dramatic with digital
gangrene
• TOS is due to extrinsic compression from fibrous bands, cervical rib or first rib
• Physical examination may be helpful with the Roos test being positive in the
majority
• Plain X-rays, Duplex and magnetic resonance angiography (MRA) may be helpful but
the diagnosis is mainly clinical
• Electrophysiology testing is non-specific although median antebrachial nerve response
has recently shown to be useful
• Removal of fibrous bands, cervical ribs and the first rib may be needed along with
anterior scalenectomy
• Arterial reconstruction of the subclavian artery may be required
Introduction
TOS is one of the most controversial clinical entities in medicine. This is partly because
there is no definitive diagnostic test and debate continues as to whether the syndrome
even really exists in some of its forms! Its incidence has been estimated at 5:100 000 per
year in the UK although the true figure is still unknown [1].
The thoracic outlet is the region at the top of the rib cage between the base of the neck
and the axilla through which the brachial plexus and the subclavian vessels travel. The first
channel is the interscalene triangle, which is bordered by the scalenus anterior, scalenus
medius and the medial border of the first rib (Figure 11.1). This is followed by the costocla-
vicular space, bordered anteriorly by the middle part of the clavicle and posteriorly by the
first rib and the scapula. The last channel is the subcoracoid space below the coracoid proc-
ess deep to the pectoralis minor tendon.
Thoracic outlet syndrome refers to a variety of complex disorders in the upper extrem-
ity caused by damage to the brachial plexus, or the subclavian artery or vein, as they pass
through the thoracic outlet tunnels described above. Thoracic outlet syndrome is more
Chapter 11: Thoracic outlet syndrome 147
Figure 11.1 Anatomy of

the thoracic outlet showing
structures in the interscalene
triangle.
ius
ed
.M
Sc
r
erio
Ant
Sc.
lavian
Subc Ar t
ery
us
lex
hial p Clavicle
Brac
First
rib
Se rib
co
nd
common in women and in the age group between 30 and 50 years but can occur in all ages,
including in children.
Pathophysiology/causes
Anatomical factors such as cervical ribs, fibrous bands, repetitive injuries or whiplash caus-
ing scarring of the scalenus anterior muscle can lead to TOS. However, in many cases no
specific anatomical factor can be identified.
Cervical ribs are seen in 0.1% of adults and only 5–10% of these are symptomatic. Fifty
per cent of cervical ribs are bilateral (Figure 11.2). Fibrous bands, which traverse the thor-
acic outlet, are the most common congenital anomaly causing TOS and will not be seen on
plain X-ray and may not be seen with magnetic resonance imaging (MRI) either. Anomalous
muscular insertions and muscle hypertrophy have also been known to cause TOS. Malunion
and formation of prominent callus after clavicle fractures have also been reported to cause
TOS. Pancoast tumours, enlarged regional lymph nodes and developmental changes in the
cervical spine are rare causes of TOS [2].
Clinical presentation
Clinical features differ according to the compressed structure/s in the thoracic outlet tunnel.
Hence, there are four main clinical syndromes described:
1. neurogenic TOS (NTOS);
2. arterial TOS (ATOS);
3. venous TOS (VTOS);
4. non-specific (combined) TOS, which is a mixture of any combination of the above three.
Figure 11.2 Bilateral cervical ribs.
Neurogenic TOS (NTOS)

This is the commonest form of TOS (90%) and patients usually present with symptoms in
the ulnar nerve distribution (C8, T1 nerve roots). Pain, especially with activities with the
arms raised, is an early and very common presentation. The pain may radiate to the axilla,
shoulder, back of neck or down the arm. Paraesthesia is usually felt on the medial aspect of
the arm, forearm and hand. There may be tenderness over the scalene muscle and reproduc-
tion of symptoms on provocation tests as below. Often there are associated vasospastic-type
symptoms, such as coldness or Raynaud’s, which may lead to an erroneous suspicion that
there is arterial involvement.
Adson test
In this test the patient takes a deep breath and extends his neck and rotates the head towards
the side being examined. The test is positive if the radial pulse is abolished or patient’s symp-
toms are replicated. Another variant is described where the manoeuvre results in paraes-
thesiae in the hand. This test is reported to be 76% specific and 79% sensitive [3] although
diagnostic decisions are rarely based on this test alone.
Roos test (Elevated Arm Stress Test, EAST)

This is a more reliable test for screening for TOC. In this test the patient abducts both arms
at 90° from the body with the elbows bent and opens and closes the hands repeatedly for 3
min. It is positive if the patient’s symptoms are reproduced. This test has been found to be
84% sensitive and 30% specific [3].
Motor weakness of the intrinsic muscles, innervated by the ulnar and median nerve, is a
late finding. Fine hand movement and hand grip may become clumsy. Patients often com-
plain of occipital headaches. Muscle wasting is usually a late presentation and may be noted
in the thenar muscles and the abductor thumb muscles. Vasomotor disturbances include
bluish red discolouration and blanching of the hand. Intermittent attacks of cyanosis or
pallor of the hand usually accompanies emotional disturbances and exposure to cold.
Figure 11.3 Distal

embolization due to arterial
thoracic outlet syndrome.
Venous TOS (VTOS)

This is the second commonest presentation and patients may present acutely with cyanosis
and arm swelling. If established there may be prominent veins seen in the upper arm or
chest. Thrombosis of the subclavian vein develops at the site of compression as the vein
passes over the first rib, and it often occurs after strenuous activity to the upper limb and is
referred to as effort thrombosis or Paget–Schroetter syndrome. It is also found in younger
patients who are athletes, such as swimmers. It is thought that repeated extrinsic compres-
sion of the subclavian vein may cause fibrosis, stenosis and eventually thrombosis of the sub-
clavian vein. It may also be due to a sudden hyper-abduction injury causing intimal damage
and thrombosis of the subclavian vein.
Arterial TOS (ATOS)

This is the rarest presentation (1%). Constriction and repetitive trauma of the subclavian
artery can lead to stenosis, aneurysm formation or complete occlusion. Poststenotic dilata-
tion of the second part of the subclavian artery may lead to aneurysm formation, which
can present with thromboembolic complications, such as acute ischaemia of the arm or a
digit (Figure 11.3). Patients might present with attacks of pallor, pain and paraesthesia, or
display Raynaud’s phenomenon. The radial pulse may be absent and provocative tests may
cause the radial pulse to disappear although this is often found in normal subjects and is not
diagnostic.
Investigations
Cervical spine film and chest radiograph can rule out cervical ribs, lung lesions or clavicular
abnormalities. Magnetic resonance imaging and computed tomography (CT) scan of the
cervicothoracic area can be helpful in detecting the non-bony anomalies (fibrous bands,
muscular abnormality, tumours), and in excluding cervical spine lesions. Magnetic reson-
ance imaging findings in patients in a provocative position (the shoulder is abducted by
placing the hand behind the head) are also valuable [4].
Neurophysiologic testing is largely unhelpful and may show non-specific abnormalities
although it can be useful to exclude other disorders. Recently a new nerve test has been shown
to be abnormal in the majority of patients with neurogenic TOS. It is the determination of

the response to medial antebrachial cutaneous nerve stimulation. Electromyography (EMG)
may reveal abnormalities in the intrinsic muscles of the hand in late cases but is unreliable
in early diagnosis. Anterior scalene nerve blocks provide temporary relief but these patients
usually respond well to surgery.
Venous Duplex ultrasound, contrast venograms or magnetic resonance venography are
useful to rule out venous obstruction or arterial involvement and these can also be per-
formed with the shoulder in abduction.
Arterial Duplex ultrasound of the upper extremity is the first-line investigation in most
cases. Arteriography or MRAs help to confirm the diagnosis and to plan reconstruction.
Angiography can highlight subclavian stenosis, aneurysms or irregular filling defects con-
sistent with mural thrombus and allow for dynamic views with the arms in abduction and
adduction.
Treatment
Neurogenic TOS
Conservative measures to avoid repetitive overhead work, correct posture and strengthen the
shoulder elevating muscles, such as the trapezius, are useful and have proven to be effective
at reducing symptoms and improving function [5].
Selective botulinum chemodenervation of the scalene muscles can provide temporary
relief in NTOS [6, 7]. This can be done under electrophysiological and fluoroscopic guidance
and has proven to be more effective than injections with local anaesthetics and steroids.
Surgery is indicated in NTOS if the conservative treatment failed to improve symp-
toms, and in the case of severe symptoms that interfere with work or daily activities. This
includes anterior scalenectomy with or without cervical rib resection, removal of fibrous
bands and excision of the first rib. Various approaches to the first rib including transaxil-
lary, supraclavicular and infraclavicular approaches have been described. Several combi-
nations of surgery have been presented in large series and there is no consensus on the
optimal approach.
Venous TOS
Therapeutic protocols now include thrombolysis and correction of the anatomical abnor-
malities contributing to the thrombosis by surgery or endovascular means. First rib resec-
tion is recommended as soon as possible after thrombolysis [8]. Without decompression of
the thoracic outlet rethrombosis may occur and likewise without decompression there is
almost universal failure of venous stenting due to external compression to the stent causing
stent fracture or thrombosis.
Arterial TOS
Thrombectomy, followed by excision of the cervical or first rib is carried out. Arterial recon-
struction may include repairing a subclavian aneurysm, which forms as a result of the post-
stenotic dilatation or bypass grafting.
Symptomatic improvement after surgery is difficult to quantify and is largely subjective.
It has been shown to vary from 43% to 88% [9].
Results have usually been classed as excellent, good and fair, based on patients’ percep-
tions. Previous acute ischaemia, sensory or motor deficit and extended resection of the first
rib have been shown to be poor predictors of outcome [10].
References chemodenervation treatment of thoracic

outlet syndrome: comparison with
1. Thompson JF, Jannsen F. Thoracic outlet fluoroscopy and electromyography
syndromes. Br J Surg 1996; 83: 435–6. guidance. Pain Physician 2007; 10: 541–6.
2. Sanders RJ, Hammond SL, Rao NM. 7. Jordan SE, Ahn SS, Freischlag JA,
Diagnosis of thoracic outlet syndrome. Gelabert HA, Machleder HI. Selective
J Vasc Surg 2007; 46: 601–4. botulinum chemodenervation of
3. Rayan GM, Jensen C. Thoracic outlet the scalene muscles for treatment of
syndrome: provocative examination neurogenic thoracic outlet syndrome. Ann
manoeuvres in a typical population. Vasc Surg 2000; 14: 365–9.
J Shoulder Elbow Surg 1995; 4: 113–17. 8. Lee MC, Grassi CJ, Belkin M, Mannick JA,
4. Jordan SE, Machleder HI. Diagnosis Whittemore AD, Donaldson MC. Early
of thoracic outlet syndrome using operative intervention after thrombolytic
electrophysiologically guided anterior therapy for primary subclavian vein
scalene blocks. Ann Vasc Surg 1998; thrombosis: an effective treatment
12: 260–4. approach. J Vasc Surg 1998; 27: 1101–7.
5. Vanti C, Natalini L, Romeo A, Tosarelli 9. Bhattacharya V, Hansrani M, Wyatt MG,
D, Pillastrini P. Conservative treatment Lambert D, Jones NA. Outcome following
of thoracic outlet syndrome. A review surgery for thoracic outlet syndrome. Eur J
of the literature. Eura Medicophys 2007; Vasc Endovasc Surg 2003; 26: 170–5.
43: 55–70. 10. Degeorges R, Reynaud C, Becquemin JP.
6. Jordan SE, Ahn SS, Gelabert HA. Thoracic outlet syndrome surgery:
Combining ultrasonography and long-term functional results. Ann Vasc
electromyography for botulinum Surg 2004; 18: 558–65.
Chapter
12
Diagnosis and management
of hyperhidrosis
Hassan Badri and Vish Bhattacharya
Key points
• Postganglionic sympathetic C fibres supply the sweat glands
• Iontophoresis involves passing a small current into the skin using tap water
• Botox injections are useful for axillary, palmar or frontal hyperhidrosis but repeat
injections are required
• Thoracoscopic sympathectomy of T2 and T3 ganglion for palmar and T2 T3 and T4
ganglia for axillary hyperhidrosis is very effective
• Patients should be warned of side effects such as compensatory hyperhidrosis, Horner’s
syndrome, pneumothorax and haemothorax
• Local surgical treatments include curettage, skin excision or liposuction
Introduction
Hyperhidrosis is the production of excessive quantities of sweat, and is caused by hyper-
function of the exocrine sweat glands, which are controlled by the sympathetic nervous sys-
tem via postsynaptic cholinergic fibres.
Nerves from the hypothalamic preoptic sweat centre synapse in the intermedi-
olateral cell columns without crossing. The myelinated preganglionic fibres pass out
in the anterior roots to the sympathetic chain. Unmyelinated postganglionic C fibres
arising from the sympathetic ganglia join the peripheral nerves and pass out to the
sweat glands.
Sweating can be induced by thermal stimuli and emotional stress. Emotional sweating
can occur over the entire skin but is more prevalent in the palms, axillae and soles. This stops
during sleep when thermal sweating can continue.
A dysfunction of the central sympathetic nervous system, possibly of the hypothal-
amic nucleus or prefrontal areas is suspected to be the cause of hyperhidrosis.
Hyperhidrosis may be primary or secondary; localized or generalized. Secondary hyper-
hidrosis may be due to hyperthyroidism or phaeochromcytoma.
Primary hyperhidrosis is not uncommon, affecting between 0.6% and 1% of the general
population. The palms, soles and axillae are the most commonly affected sites. Patients are
usually in their second or third decade of life, with a positive family history noted in 30% to
50% of cases [1, 2].
Chapter 12: Hyperhidrosis 153
Symptoms
The only presentation is excessive sweating in a localized area of the patients’ body, which
restricts their private and professional lives. The condition itself might cause bromhidrosis,
dermal mycoses, and gram-negative infections of the feet or palmar and plantar warts.
Diagnosis
The diagnosis of this condition is usually made on clinical grounds. However, Minor
described a test in which a 2% iodine solution is applied to the affected area, followed by
starch powder once the solution has dried. The hyperhidrotic skin then develops a blue-
black colouration. Colorimetry is a similar procedure where colour changes, occurring on
specially coated paper placed in contact with the sweat, are analysed.
Management
Topical therapy
Aluminium chloride hexahydrate in absolute anhydrous ethyl alcohol is the most effective
antiperspirant for treating hyperhidrosis. The product is applied to the affected areas every
night until symptoms are controlled and then weekly or fortnightly to maintain control of
sweating. The drawbacks of topical treatment are that they are time consuming, and cannot
be applied to irritated, broken or recently shaven skin. They also can cause irritation, hyper-
sensitivity and staining. Boric acid, glutaraldehyde, formaldehyde, potassium permanganate
and tannic acid have shown less satisfactory results.
Systemic therapy
Anticholinergic agents block sweat production at the level of the neuroglandular junction by
competing with acetylcholine. Treatment with glycopyrollate, at a dose of 1–2 mg, has been
tried with some success. This does not cross the blood–brain barrier and does not have any
systemic side effects unlike atropine and probathaline bromide, which can cause dry mouth,
blurred vision, mydriasis, urinary retention and constipation.
Iontophoresis
Iontophoresis involves passing a direct electrical current of about 15 mA through the skin,
thereby reducing sweat excretion. The patients’ hands or feet are placed on a metal plate and
gauze in two trays containing tap water or anticholinergic agents such as glycopyrronium brom-
ide solution. The mechanism of action is poorly understood but it is thought to act by either
causing obstruction of sweat pores or impairment of the electrochemical gradient of sweat secre-
tion [3]. A repeated course of treatment is usually required with maintenance to prevent relapse.
Tap water has been shown to be more effective than saline. Adverse effects include initial aggra-
vation of symptoms, sensory disturbances and skin eruptions. Iontophoresis is contraindicated
in patients with pacemakers and in pregnancy. It is not as effective for axillary symptoms as for
palmo-plantar hyperhidrosis but it is simple, effective and not associated with rebound com-
pensatory hyperhidrosis [4]. The treated area may become dry, cracked or fissured.
Botulinum toxin
Botulinum toxin A is a neurotoxin produced by Clostridium botulinum, which is a gram
positive, spore-producing, anaerobic bacteria. It acts at the acetylcholine presynaptic
nerve endings at neuromuscular junctions and exocrine sweat glands. The toxin works
over an area of approximately 1.2 cm and injections should therefore be spaced up to
2.5 cm apart. Injections are made in the intracutaneous rather than subepidermal layer.
The effect however gradually wanes between 4–13 months and top-up injections may be
needed [5].
Common side effects include transitory pain, intrinsic muscle wasting of the hand in
patients undergoing palmar injections, haematoma and itching. Nerve blocks are the most
effective way of combating pain around injection sites and have been found to be more effec-
tive than topical solutions of local anaesthetic.
It is both effective and well tolerated in axillary, palmar and frontal hyperhidrosis.
Surgical treatment
Sympathectomy
Sympathectomy is used to eliminate the sympathetic innervation of the sweat glands and
consequently reduce the amount of sweating. Thoracoscopic sympathectomy has proved
to be the most effective and durable treatment for patients suffering from moderate to
severe hyperhidrosis. For palmar hyperhidrosis sympathectomy should be restricted to
T2, T3 ganglions. For axillary hyperhidrosis, the T2, T3, T4 ganglia are denervated. In a
large study with a 6-year follow up, Dumont et al. found satisfaction rates and improved
quality of life of 93% and 100% after sympathectomy for palmar hyperhidrosis and 67%
and 83% after sympathectomy for axillary hyperhidrosis [6]. A recent 10-year follow up
showed that satisfaction rate were lower (47%) after a mean follow up of 12 years after sur-
gery [7]. In the case of plantar hyperhidrosis, ablation of the L2, L3, L4 ganglia is required.
Clipping of the sympathetic chain has also been advocated in patients who have severe com-
pensatory hyperhidrosis as this would make the procedure reversible.
The commonest complication after sympathectomy is compensatory sweating with
reported rates of 33–87% [8]. This affects previously unaffected areas such as trunk, chest,
back, and lower limb. This can be avoided by limiting the extent of symapthectomy to T2
level only.
Other surgical complications included haemothorax (0.1%), pneumothorax (0.5%),
segmental atelectasis (0.35%), and mild wound infections (0.1%). The recurrence rates for
palmar and axillary hyperhidrosis 5 years after surgery have been reported to be 1.3% and
16.7%, respectively [9]. Transient Horner’s syndrome has been reported in up to 0.8% of
patients and permanently in 0.1% [10].
Local surgical treatments

Other surgical treatment options are local excision of sweat glands, curettage and liposuc-
tion. Excision requires identification using the starch-iodine test. While studies have shown
this to be effective it can lead to ugly scars, haematoma, wound infection and reduced shoul-
der abduction. Recurrence rates can be as high as 20–25%. Subcutaneous curettage involves
smaller incisions and a curette to remove all the subcutaneous fat within a demarcated
area. This reduces keloid and poor scar formation but has a high failure and relapse rate.
Liposuction enables the removal of sweat glands without comprising the overlying skin and
has better cosmesis than excision and less bleeding than curettage. This may require more
than one procedure and can be time consuming, with some authors proposing a trial of
inotophoresis prior to liposuction
Chapter 12: Hyperhidrosis 155
Chemical/thermo coagulation sympathectomy

Chemical sympathectomy is another method in which sympathetic ganglion blockade is
achieved by injecting ethanol or phenol using a closed percutaneous needle technique. In
addition to the risks of pneumothorax, intradural or intravascular injection, neuralgia and
compensatory sweating have been reported. This technique has been greatly facilitated by
use under CT-guidance.
Dorsal percutaneous stereotactic thermocoagulation sympathectomy is a newer alter-
native procedure. It involves the insertion of a thermocoagulation probe through the skin
of the back under local anaesthetic. Recurrence rates are low and retreatment can easily be
performed in an outpatient setting. However, complications such as Horner’s syndrome and
pneumothorax have been noted with this procedure as well.
References for hyperhidrosis. Ann Thorac Surg 2004;

78: 1801–7.
1. Moran KT, Brady MP. Surgical 7. Walles T, Somuncuoglu G, Steger
management of primary hyperhidrosis. V, Veit S, Friedel G. Long-term
Br J Surg 1991; 78: 279–83. efficiency of endoscopic thoracic
2. Mosek A, Korczyn A. Hyperhidrosis in sympathectomy: survey 10 years after
the palms and soles. In: Korczyn A, ed. surgery. Interact Cardiovasc Thorac Surg
Handbook of Autonomic System Dysfunction. 2009; 8(1): 54–7.
New York: Marcel Dekker, 1995; pp. 167–77. 8. Kwong KF, Hobbs JL, Cooper LB,
3. Hill AC, Baker GF, Jansen GT. Mechanism Burrows W, Gamliel Z, Krasna MJ.
of action of iontophoresis in the treatment Stratified analysis of clinical outcomes
of palmar hyperhidrosis. Cutis 1981; 28: in thoracoscopic sympathectotomy for
69–70, 72. hyperhidrosis. Ann Thorac Surg 2008;
4. Karakoc Y, Aydemir EH, Kalkan MT, 85: 390–4.
Unal G. Safe control of palmoplantar 9. Lin TS, Kuo SJ, Chou MC. Uniportal
hyperhidrosis with direct electrical endoscopic thoracic sympathectomy
current. Int J Dermatol 2002; 41: 602–5. for treatment of palmar and axillary
5. Karamfilov T, Konrad H, Karte K et al. hyperhidrosis: analysis of 2000 cases.
Lower relapse rate of botulinum toxin A Neurosurgery 2002; 51: 84–7.
therapy for axillary hyperhidrosis by dose 10. Kestenholz PB, Weder W. Thoracic
increase. Arch Dermatol 2000; 136: 487–90. sympathectomy Curr Probl Dermatol 2002;
6. Dumont P, Denoyer A, Robin P. Long-term 30: 64–76.
results of thoracoscopic sympathectomy
Chapter
13
Chronic mesenteric ischaemia
Mohamed Abdelhamid, Robert Davies and Rajiv Vohra
Key points
• C hronic mesenteric ischaemia (CMI) is a rare condition, accounting for less than 5% of
all intestinal ischaemic events
• More than 90% of cases are due to atherosclerotic occlusion or severe stenosis
• Classic symptoms include postprandial abdominal pain, sitophobia and weight loss
• At least two of the three main splanchnic arteries must be significantly compromised to
result in chronic mesenteric ischaemia
• Duplex ultrasonography is non-invasive and expedient but may miss up to 20% of
vascular lesions in the coeliac trunk
• Computed tomography angiography (CTA) and magnetic resonance angiography
(MRA) are equally excellent non-invasive modalities with highly accurate diagnosis
of vascular disease in the coeliac axis (CA) and superior mesenteric artery (SMA) and
replace conventional catheter angiography
• Conventional catheter angiography should be reserved for diagnosis of CMI only
when other modalities have been unhelpful or if intervention such as percutaneous
transluminal angioplasty (PTA) is planned
• Surgical vascular bypass is the traditional definitive therapy for CMI with an overall
5-year graft patency of 78%
• Endovascular therapy is optimal in short segment atherosclerotic lesions at the ostia of
the SMA and CA. Stenting and PTA in short-term follow up have a clinical benefit with
stent patency in more than 90% of cases
Background
CMI is an uncommon cause of abdominal pain. It accounts for 5% of all intestinal ischae-
mic events with acute ischaemia being much more common. Atherosclerotic occlusion, or
severe stenosis of the mesenteric arteries, is the most common aetiology and the incidence
of atherosclerotic lesions affecting the mesenteric arteries in a person >65 years is 17.5% [1].
Symptoms of CMI such as intestinal angina, weight loss and sitophobia (=fear of eating)
usually only occur when at least two of the three main splanchnic arteries are affected. This
is because the mesenteric arterial circulation is rich in collaterals. allowing for the gradual
stenosis/occlusion of one or two main arteries without any symptoms.
Chapter 13: Chronic mesenteric ischaemia 157
Table 13.1 Aetiology and differential diagnosis of chronic mesenteric ischaemia
Aetiology Differential diagnosis

• Atherosclerosis • Gall bladder disease
• Median arcuate ligament syndrome • Peptic ulcer
• Takayasu arteritis • Abdominal malignancy
• Dysplastic lesions • Chronic pancreatitis
• Thromboangiitis obliterans • Spastic colon
• Radiation induced vascular injury • Indigestion
Pathophysiology
The majority of those affected are elderly patients with generalized atherosclerosis.
Atherosclerotic occlusion or severe stenosis of the mesenteric arteries accounts for more
than 90% of causes of CMI [2]. Hyperlipidemia, diabetes and smoking contribute to the
occurrence of CMI. Thrombotic occlusion or stenosis usually occurs at the origin of the
artery adjacent to the ostium. Diffuse atherosclerosis of the whole vessel is seen in fewer
patients. The CA and SMA are more commonly affected than the inferior mesenteric artery
(IMA). In patients with peripheral arterial disease and renal artery stenosis a quarter of
the individuals examined had greater than 50% stenosis in either the CA or SMA but only
3.4% had significant occlusion of both arteries [3]. Other causes of CMI include constric-
tion of coeliac artery blood flow by diaphragmatic compression (median arcuate ligament
syndrome, which is more predominant in women), Takayasu arteritis, dysplastic lesions,
thromboangiitis obliterans and radiation-induced vascular injury (Table 13.1).
Relative ischaemia occurs after eating when there is an increased demand for flow within
the mesenteric circulation while the arteries are unable to dilate due to the fixed occlusive
lesions. This results in transient ischaemic pain, known as intestinal angina. Pain is dull
in nature and typically postprandial, 30 min after eating, and occurs in the periumbilical
region. It may last 1–4 hours and fades gradually. The patients develop fear from eating
resulting in reduction of the size of meals in order to avoid the pain, which eventually leads
to weight loss [2]. Chronic mesenteric ischaemia involving the coeliac artery may result
in disorders such as gastroparesis, gastric ulceration and gall bladder dyskinesia. Physical
examination is usually unremarkable except for abdominal pain that is out of proportion to
examination. Sometimes, an epigastric bruit may be audible.
Diagnosis
Diagnosis requires careful history-taking and exclusion of other illnesses such as malig-
nancy, chronic pancreatitis and gastric ulcer (Table 13.1). The traditional modality used to
diagnose CMI is mesenteric angiography. Other modalities include visceral Duplex ultra-
sound, CTA and MRA.
Visceral Duplex ultrasound (VDU) evaluation of the mesenteric arteries is non-invasive.
It has been used successfully to document occlusive disease in the proximal SMA and, to
a lesser extent, in the CA. The IMA is rarely imaged by transabdominal ultrasound due to
its anatomic location. Overall, VDU has a 90% accuracy in identifying significant proximal
SMA stenosis and 80% accuracy for coeliac trunk lesions [4]. Turbulence and velocity of
blood flow are the features of stenotic and occlusive lesions affecting the proximal portion
of the arteries. Peak systolic velocity of greater than 200 cm s–1 and an end-diastolic velocity
exceeding 55 cm s–1 have been shown to have high correlation with CA stenosis [2]. End-
diastolic velocity greater than 45 cm s–1 is specific to lesions of the proximal SMA, together
with peak systolic velocity greater than 275 cm s–1. However, limitations of Duplex include
effects of respiration, obesity, food ingestion, bowel gas, anatomic variations and operator. If
screening ultrasonography detects vascular stenosis or occlusion, further detailed imaging
is usually indicated.
Conventional interventional angiography is reserved until other more common disor-
ders of chronic abdominal pain have been excluded. Selective arterial catheterisation of the
branches of the CA or SMA is possible. It usually shows occlusion or near occlusion of the
CA and/or SMA near their origins from the aorta. The IMA is usually occluded due to dif-
fuse atherosclerosis and prominent collaterals are often present. Contraindications to arteri-
ography include hypotension or hypovolemia as these may cause vasoconstriction and make
the findings less accurate. Contrast-induced nephrotoxicity is another drawback. The over-
all rate of major complications from mesenteric angiography is 1.9–2.9% and may include
external iliac artery dissection or deep venous thrombosis.
CTA is used increasingly with high sensitivity and specificity to identify significant
splanchnic vascular stenosis. The availability of three-dimensional image reconstruction
can diagnose significant atherosclerotic lesions of all the three major mesenteric arteries
and many of their main branches. In addition to providing three-dimensional images, this
modality has faster scanning time. Thinner collimation of 0.5–1.0 mm thickness facilitates
better visualisation of small vessels and branches. This minimally invasive method is highly
comparable to conventional angiography but with less cost and morbidity. In suspected
intestinal angina, a negative CTA study of the mesenteric arteries makes the diagnosis of
CMI virtually unlikely. Recently, it has been shown that using multiple radiographic criteria,
CTA has a sensitivity of 96% and specificity of 94% for diagnosis of CMI [5].
In recent years MRA has become a valuable tool for diagnosing CMI, particularly since
its cost and image acquisition times have substantially decreased. MRA images provide
high-resolution mesenteric angiograms with sensitivity greater than 90% of SMA and CA
lesions, 81–88% of portal vein disease, and 25% of lesions affecting the IMA vessels [6].
Contrast-enhanced MRA has 100% sensitivity and 95% specificity for stenosis of the CA
and SMA when compared to conventional angiography. However, small peripheral arterial
branches are less well visualised. Unlike Duplex ultrasonography, the detection of proximal
CA and SMA stenosis by contrast-enhanced MRA is accurate with minimal interobserver
variability.
The choice between these investigations depends on the availability of technology,
allergy to contrast and renal function. It is therefore recommended that invasive angiog-
raphy should be used when IMA occlusion is suspected or when endovascular therapy of
stenotic or occlusive lesions is planned (Table 13.2)
Treatment
In most cases, the treatment of CMI is not considered urgent. However, the therapeutic goal
in patients with CMI is to revascularize the mesenteric arterial circulation to prevent the
Table 13.2 Diagnostic methods of chronic mesenteric ischaemia
Advantages Disadvantages
Duplex ultrasound Non-invasive, low cost, 90% and Operator dependent, not for IMA,
80% sensitivity for SMA and CA, limited by obesity, respiration,
respectively bowel gas and food ingestion
Angiography 100% sensitive for three arteries, Invasive, contrast allergy, renal
angioplasty possible impairment, complications
CTA Non-invasive, 96% sensitive for CA Not for IMA, contrast allergy,
and SMA, operator non-dependent, renal impairment, cost
MRA Non-invasive, 90% sensitive for CA Low sensitivity for IMA, cost,
and SMA, operator non-dependent claustrophobia
SMA, superior mesenteric artery; CA, coeliac axis; IMA, inferior mesenteric artery; CTA, computed tomography
angiography; MRA, magnetic resonance angiography.
development of bowel infarction. Mesenteric vascular stenosis or occlusion usually requires

open surgical repair or endovascular therapy.
Surgical repair
Surgical repair has been the standard treatment for CMI since the first successful repair
reported by Shaw in 1958. The surgical options include:
1. Transaortic endarterectomy. This is indicated for osteal lesions of patent CA and
SMA. This can be achieved by left medial visceral rotation to expose the aorta and
its mesenteric branches. Initial success of trap-door transaortic endarterectomy was
found to reach 93% with overall patency at 1 and 3 years of 85±10.0% and 77±11.7%,
respectively [6]. However, others reported high failure rates with recurrent thrombosis
and symptoms using this method. Transaortic endarterectomy is also beneficial in
patients with concomitant renal artery stenosis and CMI due to atherosclerosis. When
significant lesions are located at the origin of both the renal artery and splanchnic
arteries, transaortic endarterectomy can be effectively used to treat both conditions
with minimal mortality.
2. Surgical bypass is indicated for occlusive lesions located 1–2 cm from the origin of
the mesenteric arteries. Surgical bypass can be performed through either antegrade
or retrograde reconstruction using either autogenous or prosthetic grafts. Single or
multivessel reconstruction with outflow into the CA, SMA or rarely the IMA should be
achieved. Isolated IMA revascularization has been used for CMI in selected cases when
it is not possible to revascularize from either the CA or SMA.
Antegrade versus retrograde inflow reconstruction

In antegrade reconstruction, the arterial inflow arises from the thoracic or supracoeliac
aorta, so the bypass is placed in the direction of normal blood flow to reduce anastomotic
turbulence. This could be done through abdominal or thoracoabdominal incision. Apart
from a slightly higher 30-day mortality with the antegrade approach, there is no statistic-
ally significant difference between the antegrade and retrograde approaches concerning
Figure 13.1 Retrograde bypass. (Top) Superior

mensenteric (SMA) stenosis, (middle) proximal
anastomosis from the right limb of the aorto-bi-iliac
graft, (bottom) distal anastomosis to the distal SMA.
symptom free survival. In addition, an antegrade approach may lead to a higher incidence
of postoperative ileus [7]. In retrograde reconstruction (Figure 13.1), inflow arises from the
infrarenal aorta or the common iliac artery.
The drawbacks of retrograde revascularization include kinking of the graft and progres-
sion of atherosclerosis to the origin of the retrograde bypass graft from the infrarenal aorta
or from the common iliac artery. Kinking of the vein grafts occurred immediately after sur-
gery in the earlier series when short vein grafts were used. This problem has been avoided
with the use of prosthetic grafts, especially when a long loop is constructed.
The retrograde approach is useful in high-risk patients requiring shorter surgical time,
in patients with previous abdominal surgery and in those who have had a failed previous
antegrade bypass [8].
Autogenous versus prosthetic grafts

Both autogenous and prosthetic vascular reconstructions have been used with satisfactory
results. Both the long saphenous vein and and superficial femoral vein have been used with
equally good results although polytetrafluoroethane (PTFE) and Dacron have more long-
term durability. Surgical reconstruction has been associated with morbidity and mortality
(5–30%) in most series [8]. This could be expected in this group as many are elderly with
significant weight loss, malnutrition, and low albumin levels, which are all predictors of
increased morbidity and mortality after any major surgery. Symptomatic recurrence requir-
ing reintervention is required in about 10–40% of patients.
Percutaneous transluminal angioplasty and stenting (PTA/stent)

In the last two decades, endovascular therapy has become a more acceptable approach for
stenotic or occlusive lesions in the mesenteric arteries. This consists of percutaneous translu-
minal angioplasty and stenting (PTA/stent) of the mesenteric arteries. Interest in this method
arose from the potential for decreased morbidity and mortality in comparison to surgical
bypass. However, reported series on percutaneous transluminal angioplasty and stenting
(PTA/stent) have contained small numbers of patients. Short occlusive or stenotic lesions
are ideal for this modality (Figure 13.2 and Figure 13.3).
An endovascular approach could also be used for graft angioplasty in the case of recur-
rent symptoms after surgical repair (Figure 13.4).
Figure 13.2 (a) Tight stenosis

at the origin of coeliac artery,
(b) successful angioplasty and
stent deployment.
(a) (b)

at the origin of the superior
mesenteric artery (SMA),
(b) successful angioplasty and
stent insertion.
(a) (b)
Table 13.3 A comparison of surgical and endovascular repairs
Surgical repair
Endovascular
Endarterectomy Bypass repair
Suitable lesions Osteal lesions Lesions 1–2 cm from Stenotic or short
the origin occlusive lesions
Success rate Initial success 93% 90–100% success 80–100% success
Suitable arteries Suitable for CA, SMA Suitable for CA, SMA Suitable for renal
and associated renal and IMA stenosis artery, CA, SMA, IMA
artery stenosis and graft stenosis
Complications 30–60% 10–30%
Perioperative mortality 5–10% 0–5%
Primary patency 60–80% 60–90%
Recurrent symptoms 10–40% 10–50%
Re-intervention at 1 year 7–20% 8–20%
CA, coeliac axis; SMA, superior mesenteric artery; IMA inferior mesenteric artery.

in vein graft 18 months after
surgery, (b) successful graft
angioplasty.
(a) (b)
Endovascular mesenteric revascularization has been associated with high immediate

technical success of 91% ± 8% with immediate symptom relief in 79% ± 9%. Complications
occurred in 18% of cases with a periprocedural mortality rate of 4%. Subjective long-term
pain relief was found in 72% ± 10% with a radiographic patency of 70% ± 18% [9].
In a review of their practice during 14 years, the Mayo Clinic Group examined the
results of 229 patients undergoing surgery or PTA/stent for CMI. They found that open
repair was associated with higher morbidity and longer hospitalization. However, peripro-
cedural mortality was not statistically significant (2.5% surgery versus 3.6% PTA). They
reported that patients undergoing PTA/stent were five times more likely to develop resten-
osis, seven times more likely to have recurrent symptoms, and four times more likely to
undergo reintervention. They concluded that surgical repair was the preferred first-line
treatment in good-risk patients and that PTA/stent was reserved for those at increased risk
from surgery (Table 13.3) [10].
Conclusion
Although uncommon, CMI remains an important cause of abdominal pain that may lead to
serious consequences if misdiagnosed. The clinician has to have a high degree of suspicion
for its early detection. Diagnosis requires good history taking, physical examination and
diagnostic testing. Non-invasive investigations, CTA or MRA, are replacing the traditional
angiography as the standard investigation with high levels of accuracy and less side effects.
Whenever indicated, the minimally invasive endovascular approach should be attempted
and if there is restenosis or recurrence of symptoms, standard open surgical repair is the
definitive therapy.
References 6. Lau H, Chew DK, Whittemore AD et al.

Transaortic endarterectomy for primary
1. Hansen KJ, Wilson DB, Craven TE et al. mesenteric revascularization. Vasc
Mesenteric artery disease in the elderly. Endovasc Surg 2002; 36: 335–41.
J Vasc Surg 2004; 40: 45–52. 7. Kansal N, Logerfo F, Belfield A et al. A
2. Van Bockel JH, Geelkerken RH, Wasser comparison of antegrade and retrograde
MN. Chronic splanchnic ischaemia. Best mesenteric bypass. Ann Vasc Surg 2002;
Practice Res Clin Gastroenterol 2001; 16: 591–6.
15: 99–119. 8. Cho JS, Carr JA, Jacobsen G et al.
3. Zwolak RM. Can duplex ultrasound Long-term outcome after mesenteric artery
replace arteriography in screening for reconstruction: a 37-year experience.
mesenteric ischaemia? Semin Vasc Surg J Vasc Surg 2002; 35: 453–60.
1999; 12: 252–60. 9. Kasirajan K, O’Hara PJ, Gray BH et al.
4. Kirkpatrick ID, Kroeker MA, Greenberg Chronic mesenteric ischaemia: open
HM. Biphasic CT with mesenteric CT surgery versus percutaneous angioplasty
angiography in the valuation of acute and stenting. J Vasc Surg 2001; 33: 63–71.
mesenteric ischaemia: initial experience. 10. Oderich GS, Bower TC, Misra S et al. Open
Radiology 2003; 229: 91–98. versus endovascular revascularization
5. Laissy JP, Trillaud H, Douek P. MR for chronic mesenteric ischaemia: risk
angiography: noninvasive vascular imaging stratified outcomes. Society for Vascular
of the abdomen. Abdom Imaging 2002; Surgery Annual Meeting, Philadelphia, PA,
27: 488–506. 3 June 2006.
Chapter
14
Acute ischaemic colitis
Vish Bhattacharya and Gerard Stansby
Key points
• A cute ischaemic colitis can be due to occlusive or non-occlusive causes
• A high index of suspicion should be present in elderly patients presenting with sudden
abdominal pain and bloody diarrhoea
• Computed tomography (CT) scan may show a ‘halo sign’ and also rule out other
abnormalities
• D-Lactate is more specific for intestinal ischaemia than L-Lactate
• Angiography and lysis may be considered in early cases with no sign of peritonitis
• In case of bowel resection primary anastomosis is best avoided and a re-look
laparotomy recommended
• Postaneurysm repair colitis can be prevented by selective reimplantaion of the inferior
mesenteric artery (IMA) in high risk cases
Introduction
Ischaemic colitis is the result of an event that leads to a reduction in colonic blood flow suf-
ficient to cause ischaemia or infarction of the colonic wall but not sufficient to produce full
thickness infarction and perforation.
The term ischaemic colitis was first introduced by Marston et al. [1] in 1966. It is com-
monly due to acute thrombosis or embolism of the superior mesenteric artery (SMA) or
IMA, causing compromise of the colonic blood supply or due to hypotension causing hypo
perfusion and ischaemia.
The term is often used for cases where full thickness infarction is present acutely but this
usage is incorrect – not all ischaemic colons have ischaemic colitis, although the two may
coexist if the involvement is patchy.
Pathophysiology
The following causes predispose the colon to ischaemia more readily than the small bowel:
1. The colon differs from the small bowel in having no villi and therefore no
countercurrent mechanism.
2. The overall blood supplied per gram of tissue is lower in the colon compared to the
small bowel.
Chapter 14: Acute ischaemic colitis 165
3. The mucosa may be relatively hypoperfused in conditions of low cardiac output

because it is at the end of the microvascular arcades supplying the colonic wall.
4. There is decreased capacity for autoregulation, especially when the perfusion pressure
is lower than 50–60 mmHg [2].
5. In periods of physiological or surgical stress blood is shunted away from the splanchnic
circulation as part of the physiological responses to shock and vasoconstrictors.
The ischaemic colonic mucosa loses its barrier function rapidly allowing invasion by lumi-
nal bacteria and absorption of endotoxins. In less severe cases inflammatory cytokine release
may contribute to multiple organ failure with renal and respiratory impairment. The mucosa
may also slough in a patchy fashion and this can result in increased peristalsis and diarrhoea
mixed with blood. In the more severe forms of the condition, bacterial invasion leads to
portal pyaemia and death.
The gut mucosa is rich in the enzyme xanthine dehyrogenase, which results in the pro-
duction of reactive oxygen species and free radicals, such as superoxide. This causes oxi-
dative tissue damage, which is actually most severe when reperfusion follows a period of
ischaemia.
This reperfusion injury also results in the activation of polymorphonuclear leucocytes,
which then result in systemic events, which may potentially lead to tissue injury and sys-
temic inflammatory response syndrome (SIRS). In addition endothelin 1 (ET-1), a potent
vasoconstrictor, is released following ischaemia and reperfusion and may lead to splanchnic
vasoconstriction, thus further exacerbating the situation.
Aetiology
The following causes lead to ischaemic colitis:
1. Occlusive vascular disease of the major vessels due to embolism, thrombosis, trauma or
surgical ligation.
2. Low perfusion states with low cardiac output, e.g. in cardiogenic or septic shock.
3. Mesenteric venous occlusive disease, where increased venous resistance leads to
impaired microcirculatory arterial perfusion, e.g. in portal hypertension, mesenteric
venous thrombosis or sequestration of red cells in sickle cell disease.
4. Drug induced ischaemic colitis due to cocaine, catecholamines, oral contraceptives [3],
phenobarital and sumitriptan.
5. Hypercoagulability states caused by dehydration and physiological shunting, e.g. in
marathon runners [4].
6. Small vessel disease, e.g. in acute pancreatitis, polyarteritis nodosa, systemic lupus
erythematosus, Wegener’s granulomatosis.
Anatomical factors
The colon is normally supplied by the SMA and IMA, with some contributions from the
internal iliac arteries via the superior haemorrhoidal vessels. This anatomical arrangement
usually allows for free collateralisation between the coeliac, SMA and IMA territories, such
that occlusion of individual vessels does not necessarily result in ischaemia.
Important in the collateral pathways between SMA and IMA territories is the marginal
artery of Drumond, which runs along the splenic flexure and the Arc of Riolan forming a
collateral between the left colic artery and the SMA (meandering mesenteric artery). These
collaterals may not always be adequate, either because of congenital variations or previous
surgery. In acute conditions, for example after surgery or myocardial infarction, there may
be no time for collaterals to form.
The internal iliac arteries provide an important collateral supply to the IMA territory of
the rectum and descending colon via the middle and inferior haemorrhoidal vessels.
The watershed between the areas of the colon supplied by the SMA and IMA is usually
described to be between the proximal two thirds and distal third of the transverse colon,
based on the embryological division between the mid- and hind-gut. This anatomical
arrangement explains the vulnerability of the splenic flexure to ischaemia. However in the
arteriopath with an occluded inferior mesenteric artery the watershed point will be shifted
distally towards the rectum.
In up to 50% of people the marginal artery of the colon may be poorly developed and this
may result in right-sided ischaemic colitis. Right colonic involvement appears to be associated
with severe forms of ischaemic colitis and occurs frequently in patients with chronic cardiac dis-
ease, such as aortic stenosis and in patients with chronic renal failure requiring haemodialysis.
Clinical features
The typical presentation is of an elderly man with a history of vascular disease complaining
of abdominal pain and tenderness with bloody diarrhoea.
Mild degrees of ischaemic colitis may in fact be subclinical and never diagnosed.
Clinically evident ischaemic colitis appears to affect 1–2% of patients after aortic surgery but
it is found more frequently if colonoscopy or biopsy evidence is sought. In a study by Welch
et al. [5] where colonoscopy and biopsies were performed post-aortic surgery a surprisingly
high 30% were found to have features of ischaemic colitis on biopsy, although virtually all of
these were asymptomatic. If only the submucosa and mucosa are involved then the presen-
tation is more likely to be mild. If the muscularis is involved then symptoms are likely to be
worse and the colitis may fail to resolve or cause subsequent stricture formation.
Full thickness infarction of the muscularis can lead to perforation and peritonitis. This
situation should not be classified as true ischaemic colitis, but most series do include 10–20%
of such cases, presumably reflecting the fact that full thickness infarcts and true ischaemic
colitis can coexist in a patchy fashion.
This includes infective colitis, pseudomembranous colitis due to Clostridium difficile toxin,
inflammatory bowel disease, an acute diverticulitis and radiation colitis.
Perhaps the most common diagnostic dilemma is between C. difficile colitis and ischae-
mic colitis in the postoperative patient where C. difficile is probably more common.
C. difficile can be tested by detecting the toxin A and/or B in a fresh or frozen stool sam-
ple. Biopsy specimens may show diffuse pseudomembranes in C. difficile. Hyalization and
haemorrhage in the lamina propria along and atrophic micro crypts and a diffuse micro-
scopic distribution of pseudomembranes are more commonly seen in ischaemic colitis
[6]. Full-thickness mucosal necrosis is also significantly more common in ischaemia than
C. difficile.
Escherichia coli 0157 may cause a haemorrhagic colitis resembling ischaemic colitis. In
addition it may also be a cause of ischaemic colitis by causing thrombosis in the colonic ves-
sels as a secondary event.
Investigations
The diagnosis of ischaemic colitis requires a high index of clinical suspicion and confirm-
ation by further investigations.
Plain X-ray
Plain X-ray of the abdomen may show fluid levels in the colon, toxic colonic dilatation, intra-
mural gas, or free gas if a perforation has occurred. In severe cases intraportal air may be
seen, a finding that suggests an outcome with a high mortality. In the postoperative patient
it may help to exclude mechanical obstruction or indicate the need for laparotomy.
Stool culture
Stool samples should be sent for microscopy for ova cysts and parasites, culture and sensitiv-
ity and for analysis for C. difficile toxin.
Endoscopic examination
Sigmoidoscopy may reveal blood in the lumen but the mucosa may appear normal at this
point if the ‘watershed zone’ is higher up. If the mucosa appears macroscopically normal but
ischaemic colitis is suspected, mucosal biopsies should be taken.
Colonoscopy will determine the extent of ischaemia and may need to be undertaken in
an unprepared bowel if the patient is unwell. Examination will determine the extent of the
ischaemia. In mild disease the mucosa has a pale appearance with petechiae. In more severe
disease the mucosa may be blue or black with slough and ulceration. Colonoscopy can help
diagnose ischaemic colitis, but cannot separate transmural from mucosal ischaemia.
Barium enema
An instant enema may reveal thumb printing due to mucosal oedema. However, endoscopy
with biopsy is preferable as a means of investigation, if available.
Computed tomography (CT)/magnetic resonance imaging (MRI)

scanning
These may be normal in the early stages of ischaemic colitis or show non-specific thickening
or oedema with a ‘double halo’ or ‘target’ appearance. CT can be used to confirm the clinical
suspicion of ischaemic colitis and to diagnose complications. In ischaemic colitis, CT typically
demonstrates circumferential, symmetric wall thickening with fold enlargement. CT may also
demonstrate a thrombus in the mesenteric vessel, toxic megacolon or gas in the colonic wall.
Modern spiral CT may also allow assessment of patency of visceral vessels as may gadolinium-
enhanced magnetic resonance angiography (MRA).
Blood gases
In severe intestinal infarction the patient will develop a metabolic acidosis with a large base
excess and low pH. However, these metabolic changes are usually signs of advanced full
thickness infarction rather than ischaemic colitis.
L-lactate is produced by all cells as a product of glycolysis and is produced in excess
during conditions of hypoxia. Raised levels of L-lactate are therefore related to inadequate
tissue perfusion. However since L-lactate from the intestine is mostly removed by the liver
it is not usually helpful in the diagnosis of ischaemic colitis, although it may be raised in
colonic infarction. D-Lactate is produced by intestinal bacteria and may be more predictive
of colonic ischaemia after aortic surgery than L-lactate, although there has been no pro-
spective study.
Angiography/Duplex scanning
Angiography and Duplex scanning of mesenteric vessels are rarely helpful or diagnostic for
ischaemic colitis in the acute situation.
Management
Conservative
Conservative management is appropriate in most patients if they are clinically stable and
there are no signs of full thickness involvement or peritonitis. Patients should be given intra-
venous fluids and antibiotics and be carefully monitored. If there are conditions predispos-
ing to intestinal ischaemia, such as hypercoagulability, they should be treated.
Pharmacological
A number of pharmacological treatments have been shown experimentally to improve intes-
tinal blood flow, such as glucagon and prostanoid infusions, but are not widely used clin-
ically. Early colonoscopy should be carried out to confirm the diagnosis. The question of
angiography and other investigations remains controversial. In the case of a patient who
appears to be settling and who is systemically well, intervention is probably best avoided.
Surgical
In a patient who continues with significant bowel disturbance or systemic symptoms angi-
ography may be considered in order to assess the visceral circulation. If there is a significant
stenosis or occlusion of visceral arteries then thrombolysis followed by angioplasty may be
considered. Thrombolytic therapy should be undertaken within 8 hours of onset of symp-
toms and only if there is no sign of peritonitis or bowel necrosis. Angioplasty may be con-
sidered if there is any stenosis of the SMA. Heparin should be given concomitantly, provided
there is no bowel necrosis.
In cases with embolic acute myocardial infarction (AMI), where lysis is not indicated,
surgical embolectomy can be carried out. A transverse arteriotomy is made and the clot
removed. If the embolectomy fails, an aorto bi-iliac graft can be inserted.
Patients who have peritonitis or who deteriorate should have a laparotomy with resection
of the ischaemic segment and both ends of the bowel should be brought out as stomas. A pri-
mary anastomosis is probably always best avoided when the colon is involved. If attempted
then the anastomosis certainly should be covered by a defunctioning stoma. Re-look laparot-
omy the following day should also be considered. Although there is a high-associated mortal-
ity, in patients who survive it is often possible to reverse the stoma at a later date [7].
Mortality rates in patients requiring emergency surgery for peritonitis and resection of
ischaemic bowel may be as high as 50%. This is due to the associated comorbid medical con-
ditions including peripheral vascular disease, ischaemic heart disease and cerebrovascular
disease. If the original surgery has involved aortic grafting then there is also a high risk of
subsequent graft infection.
Colonic stricture may develop as a late complication of ischaemic colitis, especially

where the muscularis has been involved. Clinically these patients will typically present with
features of subacute obstruction. Bowel resection and anastamosis for these strictures may
lead to anastomotic dehiscence in cases where there has been an occlusive aetiology.
As postaortic ischaemic colitis is commonly seen this will be discussed in detail below.
Ischaemic colitis after aortic surgery

The single commonest cause of ischaemic colitis is postaortic surgery where a combination
of inferior mesenteric occlusion, postoperative hypoxia or hypotension and a vulnerable,
elderly patient are predisposing factors.
Incidence
The incidence of this condition is of the order of 1–2% in most series. In ruptured aortic
aneurysm repair, however, the incidence is much higher, up to 30%. Studies with colonos-
copy or sigmoidoscopy have, however, shown an even higher incidence: 7–35% of elective
cases and up to 60% of survivors of ruptured aneurysm repair.
Fanti et al. [8] reported a series of 105 patients who underwent rectosigmoidoscopy
within 72 hours of aortic surgery. Colonic ischaemia was found in 12 patients but 7 were
asymptomatic and all were managed conservatively.
Welch et al. studied a group of patients undergoing elective aortic surgery with pre- and
postoperative colonoscopy with biopsy in order to assess the true incidence of the condi-
tion [5]. They studied 28 patients each in two groups having aneurysm repair and reconstruc-
tion for occlusive aortoiliac disease, respectively. Postoperative colonoscopy and biopsy was
carried out at one week. All patients had normal appearances and biopsies before surgery.
Postoperatively, however, 30% had features of ischaemic colitis upon biopsy. Interestingly
there was no difference in the incidence of this finding between those with occlusive and
those with aneurysmal disease.
It has also been described following aortic stent grafting.
Risk factors
Renal disease, emergency surgery, age, type of hospital, aortobifemoral graft, operating time,
cross-clamping time and ligation of one or both internal iliac arteries are independent risk
factors for developing ischaemic colitis. Duration of hypotension, temperature less than
35°C, pH <7.3, fluid requirement of >5 l and packed red cells >6 units have been found to be
predictive of ischaemic colonic complications following ruptured aortic aneurysm repair.
Ischaemic colitis after aortic surgery may be due to interruption to colonic blood supply
due to division of the IMA, hypotension and hypoxia in the perioperative period and sten-
osis of other visceral vessels due to the underlying atherosclerotic disease process. Inferior
mesenteric artery ligation is probably the most important factor in the majority of cases.
Other possible factors include embolisation into the IMA territory during dissection of
the aneurysm, injury to mesenteric vessels by retractors and mesenteric compression by
haematoma.
Prevention
As postoperative ischaemic colitis occurs in an unpredictable fashion there are no clear
guidelines as to how surgeons may avoid its development. The normal IMA is not well
demonstrated at routine angiography and preoperative angiography may demonstrate only

the occasional at-risk patient with enlarged collaterals. Routine angiography can therefore
not be justified in all patients undergoing aortic surgery.
Various suggestions have been made to either reimplant the patent IMA routinely, or
selectively based on stump pressure, Doppler and intraoperative inspection.
Seeger et al. [9] reimplanted all patent IMAs routinely in 151 aortic reconstructions.
None of the patients developed colonic infarction compared with 2.7% of patients in the
previous series of 186 patients where reimplantation was done based on stump pressure,
Doppler and inspection.
Killen et al. [10] used IMA stump pressure of less than 50 mmHg as an indicator for
implantation of the IMA. However this did not improve the incidence of ischaemic colitis
in his series.
Routine reimplantation of IMAs may increase the risk of bleeding from the anastomosis,
is technically demanding and takes extra time. However this is recommended in patients
whose preoperative angiography has shown occlusion or stenosis of the SMA. This is also
recommended if neither of the internal iliacs is reconstructed and the colon appears ischae-
mic at the end of the operation.
In summary the following steps can prevent this complication after aneurysm repair:
1. At least one internal iliac artery should be revascularised whenever possible.
2. Blood pressure and oxygen levels should be maintained as any fall may be critical in
patients with an already compromised colonic mucosal perfusion.
3. Inspection of the colon after surgery is important, although confounding factors such
as hypovolemia, hypotension and the use of inotropes can make this unreliable.
4. Selective reimplantation of the IMA in high-risk patients as mentioned above.
Conclusion
Acute mesenteric ischemia is a serious condition associated with a 60–80% mortality.
The SMA is the main vessel usually involved in 85% of cases. CT angiography with three-
dimensional reconstruction is the diagnostic tool of choice. Conventional angiography is
only useful if this is associated with thrombolysis and or stenting.
The elderly are most commonly affected due to their higher incidence of underlying
systemic pathology, most notably atherosclerotic cardiovascular disease. A high index of
suspicion should be present, especially in the elderly arteriopath presenting with abdominal
pain. Early recognition and an aggressive therapeutic approach are essential if the usually
poor outcome is to be improved. Blood pressure support typically involves careful, but often
massive, fluid resuscitation and pharmacologic support. Thrombolysis and stenting have a
limited role and only in the early hours of diagnosis. A second look laparoscopy or laparot-
omy is indicated in the case of bowel resection.
References circulation. In: Shepherd AP, Granger DN,

1. Marston A, Pheils MT, Thomas ML, eds. Physiology of the Intestinal Circulation.
Morson BC. Ischaemic colitis. Gut 1966; New York: Raven Press, 1984.
7: 1–15. 3. Mann DE Jr, Kessel ER, Mullins DL,
2. Kvietys PR, Granger DN. Physiology, Lottenberg R. Ischemic colitis and acquired
pharmacology and pathology of the colonic resistance to activated protein C in a
woman using oral contraceptives. total colonic ischaemia. Dis Colon Rectum
Am J Gastroenterol 1998; 93: 1997; 40: 1448–54.
1960–2. 8. Fanti L, Masci E, Mariani A et al. Is
4. Lucas W, Schroy PC 3rd. Reversible endoscopy useful for early diagnosis of
ischaemic colitis in a high endurance ischaemic colitis after aortic surgery?
athlete. Am J Gastroenterol. 1998; Results of a prospective trial. Ital J
93: 2231–4. Gastroenterol Hepatol 1997; 29: 357–60.
5. Welch M, Baguneid MS, McMahon 9. Seeger JM, Coe DA, Kaelin LD, Flynn TC.
RF et al. Histological study of colonic Routine reimplantation of patent inferior
ischaemia after aortic surgery. Br J Surg mesenteric arteries limits colon infarction
1998; 85: 1095–8. after aortic reconstruction. J Vasc Surg
6. Dignan CR, Greenson JK. Can ischaemic 1992; 15: 635–41.
colitis be differentiated from C difficile 10. Killen DA, Reed WA, Gorton ME
colitis in biopsy specimens? Am J Surg et al. Is routine postaneurysmectomy
Pathol 1997; 21: 706–10. hemodynamic assessment of the inferior
7. Longo WE, Ward D, Vernava AM 3rd, mesenteric artery circulation helpful? Ann
Kaminski DL. Outcome of patients with Vasc Surg 1999; 13: 533–8.
Chapter
15
Vascular trauma
Robbie George and Paul Blair
Key points
• emember the whole patient, do not just focus on the vascular injury
R
• Time is of the essence, avoid delay
• Do not attempt to mobilise large veins, use local pressure
• Consider temporary intravascular shunts in complex limb injuries
• Consider damage limitation surgery in patients developing hypothermia or acidosis
Trauma is a leading cause of mortality in the first four decades of life. Vascular surgeons
are often involved in the management of a multiply-injured patient who may have limb-
and/or life-threatening vascular injuries. In addition the extended range of procedures
carried out by open and minimally invasive surgical and radiological techniques has cre-
ated its own unique set of vascular injuries. It is beyond the scope of this short chapter to
deal with specific vascular injuries in detail, however, general principles will be discussed
with specific details given in the more common sites of injury.
General considerations
Vascular trauma can occur as a result of a variety of mechanisms including penetrating,
blunt, crush, irradiation and a variety of iatrogenic injuries. The majority of penetrating
injuries in civilian life in the UK are caused by knives or low-velocity handguns. Penetrating
injuries in military and terrorist theatres are more often associated with high-velocity weap-
ons, bombs and missiles, the latter can cause extensive tissue damage due to a combination
of blast and shrapnel injuries.
Blunt vascular trauma is usually seen following road traffic accidents, falls, building col-
lapses, major disasters, etc.
Time is of the essence when dealing with vascular trauma and the need for early control
of haemorrhage and restoration of blood flow must be balanced with potential delay caused
by over investigation. Patients should be managed along the principles of the Advanced
Trauma Life Support (ATLS) system as is it important not to miss occult torso injuries. Rapid
exsanguinating local haemorrhage is best controlled with local pressure. The use of vascular
clamps should be avoided in the emergency department. A small but significant number of
trauma patients may benefit from immediate transfer to the operating room without further
investigation (see Table 15.1).
Chapter 15: Vascular trauma 173
Table 15.1 Indications for immediate transfer to the operating room
Cardiopulmonary resuscitation (CPR) in progress and penetrating torso injury

Systolic blood pressure <90 after 2 l of fluid
Major amputation of hip/shoulder
Major complex wounds and hypotension
Limb trauma
Pathophysiology
Distal ischaemia can result from a wide range of arterial injury, including complete tran-
section or laceration of vessels, contusions and haematomas, dissections, thromboses,
occlusion by haematoma from adjacent injury, false aneurysms and delayed presentation of
arteriovenous fistula.
Arterial injury affects distal flow and results in ischaemia and tissue hypoxia. Striated
muscle will likely undergo irreversible damage if warm ischaemia time exceeds 6–8 hours and
may result in multi-organ dysfunction as well. Limb ischaemia is associated with oedema;
tight fascial compartments can cause a further fall in perfusion pressure with worsening of
ischaemia resulting in compartment syndrome. Revascularisation of ischaemic limbs can
result in significant ischaemia reperfusion injury and rapidly destabilize an already unwell
patient as toxic metabolites are washed out. Although the upper limb will tolerate a pro-
longed period of warm ischaemia, >6 hours ischaemia is poorly tolerated in the lower limb.
Revascularisation of a non-viable limb is futile and hazardous for the patient.
Initial assessment and management

Clinical signs of limb vascular injuries can be classified into hard and soft signs. See
Table 15.2. The majority of patients with a single-level penetrating injury and one
hard sign should be transferred immediately to the operating theatre, without further
imaging, for immediate exploration. Exceptions to this rule include:
• proximal upper limb injuries with suspected subclavian or axillary artery involvement;
• extensive bone or soft tissue injury;
• close range shotgun injuries or multiple entry wounds;
• elderly patients with co-existing peripheral vascular disease (PVD).
Other issues to be considered:
• Is there any neurological deficit associated with the trauma?
• Is the limb salvageable? Various scores such as the Mangled Extremity Severity Score
(MESS) are in use. They are good predictors of likely limb salvage but not of likely limb
loss [1].
• Duration since injury, i.e. warm ischaemia time.
• Any features of compartment syndrome – is the arm/calf soft.
Investigations [2]
It can be difficult to assess distal limb perfusion in a hypothermic, hypovolaemic, multiply-
injured patient. Hand-held Doppler can be helpful but should be used with caution. Accuracy
Table 15.2 Clinical signs of vascular injury in a limb
Hard signs Soft signs

Absent pulses Pulses present but decreased (compare with
opposite limb and/or measure ankle brachial
pressure index)
Arterial thrill or bruit over or near the artery History of haemorrhage at the scene
Observed pulsatile bleeding Unexplained hypotension
Signs of distal ischaemia Peripheral nerve deficit
Haematoma (large or expanding)
is improved significantly when formal ankle brachial pressure index (ABPI) measurements
are performed. An ABPI of <0.9 should raise the suspicion of an upstream vascular compro-
mise. Stable patients with hard signs of vascular trauma, not falling into the urgent category
(Table 15.1), or those patients with single-level penetrating injury and hard sign of vascular
trauma, may be investigated in a more timely fashion.
Computed tomography (CT) is widely available and can provide superb contrast resolu-
tion with the additional visualisation of non-vascular structures, particularly useful in the
investigation of a trauma patient. Multi-slice technology has further expanded the image
quality of run-off vessels, although images can be sub-optimal. Catheter-directed angiog-
raphy remains the gold standard, particularly if therapeutic interventional procedures are
required. The development of endovascular surgery has improved the quality of digital sub-
traction angiography (DSA) available in the operating theatre.
The investigation of patients with soft signs of vascular trauma is more controversial.
The majority of publications, in the literature, concerning investigation of such patients,
are often produced from high volume North American or South African trauma centres
and are, therefore, not always applicable to the UK where the incidence of vascular trauma
is relatively low. While it is not appropriate to over-investigate patients, once the suspicion
of a vascular injury is raised it should be excluded by appropriate imaging. There is a high
incidence of occult arterial injury with posterior knee dislocations and investigations should
be utilised freely in this situation. Over-investigation of penetrating proximity injury, with
‘soft signs’, should be balanced with the risk of late sequelae of missed injuries such as false
aneurysms and arteriovenous fistulae.
General principles of arterial repair

Prior to exploring any vascular injury, proximal and distal control must be attempted, with
the use of additional axial incisions if required. Major bleeds from the femoral vessels in the
groin are best controlled by direct pressure until proximal control of the external iliac vessels
is obtained via an extraperitoneal approach in the iliac fossa. The contralateral limb should
be prepped in case great saphenous vein (GSV) harvest is required and the abdomen should
also be prepped in case iliac control is required. Axillary artery injuries may require proxi-
mal control of the subclavian artery in the supraclavicular space while the axillary artery
itself is approached through an incision just below the clavicle with division/separation of
pectoralis fibres. Axillary and subclavian artery injuries should be approached with care due
to the close association of the brachial plexus. Embolectomy catheters may be employed to
gain temporary proximal control.
Figure 15.1 Proximal and distal

compression using fingers and
swabs in venous injury.
In complex limb injury, or delayed management of simple arterial injury, the use of a tem-
porary intravascular shunt should be considered [3, 4]. This allows early restoration of tissue
perfusion prior to a planned arterial repair with appropriate soft tissue cover. It is important
to have a healthy vessel exposed proximally and distally, particularly in crush injury, and a
gentle distal embolectomy should be performed to remove distal thrombus prior to shunt
insertion. Temporary intraluminal shunts may also be employed in venous injury although
this is rarely required. Once the shunt is in place, associated damage to bone, nerve, muscles
and skin can be fully assessed and may require the presence of an orthopaedic and plastic
surgeon. If there is an associated fracture, bony stability via internal or external fixation can
be achieved with the shunt in place. This allows stability prior to definitive arterial and/or
venous repair. Intravenous heparin may be employed although is best avoided in unstable
hypovolaemic trauma patients.
Arterial repair can be in the form of lateral suture, patch angioplasty or some form of
arterial graft. Patch angioplasty can be difficult in a young patient, due to spasm, and if in
doubt a short segment GSV graft is the operation of choice. It is the authors’ opinion that
resection of a damaged vessel with end-to-end anastomosis is rarely possible and an inter-
position graft is preferable. In the majority of cases GSV should be harvested from the con-
tralateral limb in case of deep vein damage in the traumatised limb. The majority of venous
injuries require lateral suturing or occasionally ligation. Venous repair is rarely required in
the upper limb but may be required in the lower limb at popliteal level. Complex venous
repair employing panel or spiral grafts is best avoided in the majority of trauma patients.
Limb distal vessel injury

Isolated distal radial or ulnar artery injuries can be ligated if collateral circulation, as assessed
by Allen’s test, is adequate. However, they frequently need exploration because of the risk of
associated nerve injury. Isolated lower limb crural vessel injury rarely requires repair and
may be ligated in the majority of cases. If both anterior and posterior tibial vessels have
been transected in a young patient, attempts should be made to repair at least one of them,
preferably the posterior tibial artery. Achieving tissue cover for the vascular repair is critical
to success. Use of soft tissue flaps and involvement of the plastic surgeons is very helpful. In
extensive soft tissue loss extra-anatomic bypasses may be utilised.
Fasciotomies should be used liberally as compartmental hypertension from a combina-

tion of ischaemic damage, direct trauma, haematomas and third space losses associated with
trauma can all compromise blood flow.
Strong indications for fasciotomy include an ischaemia time greater than 6 hours,
major artery and vein injury, significant soft tissue injury or evidence of muscle damage,
proximal half of below-knee vascular tree injury or compartment pressures of greater than
40 mmHg.
Vascular injuries of the chest [5, 6]

Thoracic vascular injuries may occur in road traffic accidents associated with rapid decelera-
tion. They may also result from penetrating and gunshot wounds.
Diagnosis and assessment

Penetrating thoracic vascular injuries may be obvious while major vascular injury in the
setting of blunt trauma is typically one component of polytrauma and must always be sus-
pected. Following a major deceleration injury or gunshot wound a combination of severe
interscapular pain, decreased air entry and dullness in the chest, with reduced or diminished
pulses, is highly suggestive of free aortic rupture.
Hypotension and other major signs of significant blood loss may not be seen in a patient
with a contained rupture. Dissection or partial laceration of the aortic wall may be contained
by spasm and surrounding mediastinal tissues. This, however, is a very temporary situation
and it is imperative to detect and treat the injury as soon as possible. The clinical triad of
grossly widened mediastinum on chest X-ray, haemothorax and transient hemodynamic
instability is highly indicative of aortic injury. Aortic injuries can be associated with a 2–3%
incidence of paraplegia resulting from spinal cord ischaemia.
Investigations
Chest X-ray should be part of the adjuncts to primary survey in the management of all
trauma patients and the signs of aortic injury include a widened mediastinum, loss of aortic
knuckle, apical cap, depression of left main bronchus and deviation of the trachea/nasogas-
tric tube to the right.
If a vascular injury is suspected the next investigation should be contrast enhanced helical
CT. Transoesophageal echocardiography and arch aortography are other useful modalities.
The choice of investigation is dictated by the hemodynamic state, other injuries, local avail-
ability and expertise.
Management
A rapid rise in blood pressure in these patients may convert a contained rupture into a free
rupture, resulting in rapid exsanguinations and death. A regime of permissive hypotension,
where the aim is to maintain cerebral perfusion rather than a normal blood pressure, is
followed. In the conscious patient this is easy to implement, in unconscious or intubated
patients a systolic pressure of 70–80 mmHg is probably acceptable.
Injuries presenting with features of free bleeding need minimum investigations and
should quickly proceed to resuscitative surgery. In more contained situations the airway
and cervical spine should be secured and cardiac tamponade and tension pneumothorax
relieved prior to transfer to a trauma centre. No attempt should be made to drain a haemot-
horax unless there is respiratory compromise.
Operative management
In penetrating trauma, surgical intervention is required in the presence of cardiac arrest,
significant or sustained haemorrhage from the chest, mediastinal traversing injury, major
vascular injury, sucking chest wound, persistent massive air leak or diaphragmatic rupture.
In a minority of patients this might necessitate emergency room thoracotomy, however this
is a procedure best carried out in the operating theatre.
Contained thoracic aortic rupture due to blunt trauma is managed by blood pressure
control and treatment of other life threatening injuries with repair of the thoracic aorta on a
semi-urgent basis. β-blockers are the drug of choice in this situation.
A recent meta-analysis of retrospective cohort studies indicates that endovascular treat-
ment of descending thoracic aortic trauma is an alternative to open repair and is associated
with lower postoperative mortality and ischaemic spinal cord complication rates [7].
Vascular injuries of the neck [8, 9]

Carotid artery injury accounts for approximately 10% of all vascular trauma with penetrat-
ing trauma responsible for 90% of cases. The carotid artery is most commonly involved in
penetrating trauma with the vertebral artery more commonly damaged by blunt trauma.
Mortality remains around 10–30% and in survivors permanent neurological sequelae are
present in approximately 40%.
Diagnosis and assessment of penetrating neck vascular injuries

If a penetrating injury has breached the platysma, surgical exploration or arterial imaging is
essential. Besides active bleeding, a cervical bruit or thrill or a rapidly expanding haematoma
and an absent carotid pulse are diagnostic of carotid artery injury. It is essential to look for
evidence of neurological deficits, which may be the result of vessel occlusion (contralateral
deficits) or direct cranial nerve injuries (IX–XII). Associated head injuries, psychotropic
substance or systemic hypotension and hypothermia may make neurological assessment
difficult.
Investigations and management

Penetrating carotid injuries
Carotid artery injuries are classified into three zones and management is dependent on the
zone of injury.
Zone 1 – injuries from clavicle to cricoid cartilage.
Zone 2 – injuries from cricoid to angle of mandible.
Zone 3 – injuries above the angle of mandible.
Surgical access in zone 1 and 3 injuries is problematic and, therefore, imaging is essential.
The gold standard imaging modality is catheter angiography although CT angiography and
occasionally Duplex may be employed. Catheter-directed angiography has the additional

benefit of offering therapeutic endovascular interventions. Most patients, unless unstable,
will need further investigation to rule out concomitant aero-digestive tract injury.
Injuries with active bleeding, expanding cervical haematomas or airway compromise
should undergo immediate surgical exploration. In such patients early control of the airway
by intubation must be considered.
Management
Patients with minor angiography abnormalities, such as small pseudoaneurysms, small
intimal defects and non-obstructive downstream intimal plaques may be managed non-
operatively with close follow up.
Patients with carotid artery occlusion on angiography and a dense neurological deficit
due to brain infarct have a poor outcome, regardless of operative management.
Those with an occluded vessel but no neurological deficit should be managed with anti-
coagulation to prevent thrombus extension. High zone 3 injuries and those involving the
vertebral artery may be best managed by endovascular methods, such as the use of embolisa-
tion and covered stents. Most other injuries seen on angiography are likely to need surgical
intervention.
The carotid artery is approached by an incision along the anterior border of the sternomas-
toid with the chest being prepped for access if needed. If possible proximal and distal control
should be obtained prior to entering a haematoma although this may not always be possible.
Digital pressure may be used prior to vessel control, but care should be taken to avoid dam-
aging the vagus, hypoglossal and recurrent laryngeal nerves. Repair is undertaken by suture,
patch or interposition or transposition grafting as required. Occasionally a thrombosed
occluded vessel may be ligated to reduce the risk of embolic stroke. Thorough exploration of
the aerodigestive tract is essential to ensure that no other injuries are missed.
Blunt carotid artery injury in the neck

Blunt injury may cause dissection and thrombosis of the carotid or, more commonly, ver-
tebral arteries. These injuries can be extremely difficult to diagnose and may be missed in
up to two-thirds of patients. High-risk criteria for associated blunt cerebrovascular injuries
include severe hyperextension, rotation or flexion of the neck, a significant anterior triangle
haematoma or soft tissue injury, transient ischaemic attack (TIA) or stroke, cervical spine
fracture (vertebral artery), seat belt around the neck, bruit or thrill and a basilar fracture
involving the petrous bone (carotid artery). Occasionally patients develop transient neuro-
logical signs at the time of injury. There may be a classic history of a neurologically intact
road traffic victim later developing hemiparesis or other lateralising neurological signs.
Some patients may present with Horner’s syndrome.
Catheter angiography is the diagnostic gold standard, though there is also a role for
CT angiography and perhaps Duplex sonography. Imaging of the brain and spine is also
essential.
The keystones of management are to optimise haemodynamics and anticoagulation.
Hypotension must be avoided to prevent thromboses. Hypertension risks the development
of an intimal flap and dissection. Systemic anticoagulation is the preferred treatment for
most patients. Heparin or combinations of antiplatelets have both been utilised, with no clear
benefit for either. Endovascular placements of stents may have a role in the management of
pseudoaneurysms but their role in the management of dissections remains unclear. These
patients need to be followed up with MRI to detect late onset false aneurysm formation.
Vertebral artery injuries

The majority of vertebral artery injuries are silent due to the vertebral artery’s capacity for
collateral flow. However in 2–3% of patients the contralateral artery may be hypoplastic,
precipitating the development of neurological signs. Injuries occur rarely in the setting of
penetrating trauma and are more commonly associated with cervical spine injuries.
Catheter angiography is usually required for diagnosis and, where possible, endovascu-
lar interventions are the treatment of choice, as surgical access can be very difficult.
Abdominal vascular trauma

Patients with significant abdominal vascular trauma are usually haemodynamically unstable
with obvious signs of haemorrhage. A small group of patients, however, may present late
with relatively subtle signs. Abdominal vascular trauma can present as hypovolaemia, limb
ischaemia, anuria or obvious intraoperative haemorrhage. Most patients, unless extremely
unstable, will have a contrast enhanced CT of the abdomen and pelvis. Resuscitation and
diagnosis usually takes place simultaneously and unstable patients should proceed, without
delay, to the operating room as stated in Table 15.1.
Stable patients, who have sustained blunt trauma following CT evaluation, may have
attempted embolisation of arterial bleeding points i.e. those associated with pelvic fractures,
distal renal artery injury and lumbar vessels. It is important, however, to exclude non-
vascular injuries in these patients, e.g. occult bowel perforation.
A standard mid-line laparotomy is performed with the patient prepped for additional thora-
cotomy if required. Initial assessment requires four quadrant packing and careful evaluation
of injuries. If faced with uncontrollable haemorrhage, from multiple sites, control of the
aorta at the diaphragmatic hiatus may be a useful manoeuvre. Mobilisation of the left lobe
of liver is required and care must be taken to avoid damaging the oesophagus (more easily
identified if a nasogastric [NG] tube has been inserted). Finger dissection of the diaphrag-
matic crus in a vertical plane is required before the aorta can be visualised and clamped
in a vertical fashion. Formal vascular repair, in the abdomen, depends on the location of
the injury and also the general condition of the patient. In the extremely unstable patient,
principles of damage control surgery should be applied. The presence of metabolic acidosis,
hypothermia and coagulopathy may preclude formal vascular repair. In such patients, pack-
ing the abdomen and pelvis, followed by a period of resuscitation in intensive care, may be
entirely appropriate prior to second-look laparotomy.
Management of retroperitoneal haematomas

The decision to explore a retro-peritoneal haematoma depends on the mechanism of injury,
stability of the patient and the appearance of the haematoma. Exploration of a retro-perito-
neal haematoma can be challenging for an experienced vascular surgeon and it may be best
to simply pack the area and wait for additional help to arrive. Retro-peritoneal haematomas,
caused by penetrating trauma, require mandatory exploration, while a more selective policy
may be employed for those caused by blunt trauma. They are classified into three zones –
central, lateral and pelvic.
Central haematomas (zone 1) are always explored as the likely source of bleed is the great
vessels, pancreas or duodenum. For suprarenal haematomas proximal control of the aorta
is obtained at the diaphragmatic hiatus. The lesser sac is then opened to isolate the injury.
If that fails left or right visceral rotation is carried out to expose the aorta or inferior vena
cava (IVC), respectively. Infrarenal haematomas or those at the base of the mesentery are
approached after infrarenal aortic control. Special care must be taken to avoid injury to the
extremely delicate IVC, renal and iliac veins.
Lateral (zone 2) haematomas are likely of renal pedicle origin and should be explored
if expanding or pulsatile or if there is radiological evidence of serious injury. Renal vessels
should be controlled prior to opening the haematoma. Avulsed or lacerated renal arteries are
best treated by using a saphenous graft although outcomes are often poor.
Pelvic fracture associated injuries (zone 3) can be due to torn pelvic veins or injury to the
iliac vessels. In case of a pelvic fracture, fixation of the same is essential to reduce the pelvic
volume and therefore blood loss. If the patient is reasonably stable following pelvic fracture
fixation, investigation via catheter guided angiography may be particularly useful in achiev-
ing embolisation of arterial bleeders.
The retro-peritoneum often provides tamponade for zone 3 haematomas and if this is
entered catastrophic blood loss may occur. When faced with uncontrollable venous haemor-
rhage in this situation damage limitation surgery and packing should be considered. If the hae-
matoma is actively bleeding or expanding and pulsatile, especially with a missing pulse in the
groin, it should be explored. Some of these injuries can be managed by endovascular means.
Mesenteric vessel injuries [10]

The superior mesenteric artery (SMA) proximal to the trunk of the middle colic should be
repaired if at all possible or a saphenous graft taken off the aorta. Segmental branches to the
small bowel can be ligated and in between injuries should be repaired or ligated, dependent
on the patient’s condition.
• Coeliac artery injuries will tolerate ligation but should be repaired if possible.
• IVC injuries are best controlled by direct pressure and suture repair where possible. For
major uncontrollable haemorrhage IVC ligation is the best option and is well tolerated.
Retrohepatic IVC injuries have over an 80% mortality. Patients may need a median
sternotomy and atriocaval shunt or total occlusion of the portal system to achieve
control and repair.
• Portal vein injuries are also associated with extremely high mortality. Where possible
they should be repaired by direct suture or interposition grafting. Failing that ligation
would be the best option.
• Renal vein injuries on the right side need repair as the right kidney does not tolerate
ligation. On the left the vein can be ligated, providing drainage through collaterals to
the gonadal and adrenal vein are preserved.
• Iliac vein injuries are treated by packing if associated with pelvic fractures. Where
possible common or external iliac vein injuries should be repaired but ligation is a
feasible option as well.
Delayed presentation abdominal vascular injury

Patients may present with a pulsatile mass suggestive of a pseudoaneurysm. GI haemorrhage
may result from haemobilia secondary to liver injury or from false aneurysms eroding into
the rectum, colon and duodenum. Aortocaval fistula can present with lower limb oedema
and an arteriovenous bruit in the abdomen. They may present late with congestive heart fail-
ure and venous hypertension. The main investigation in this patient group is an angiogram
and investigations of the involved organ.
Iatrogenic injuries
A wide scope of iatrogenic vascular injuries may occur as surgical advances continue.
Injuries of arterial or venous catheterisation and

endovascular procedures
Arterial catheterisation and puncture may cause bleeding, stenosis or occlusion as a result
of laceration, thromboembolism, dissection or foreign body occlusion. If recognised at the
time of intervention numerous endovascular interventions in the form of angioplasty, cov-
ered stents etc. are available.
Major haemorrhage can occur especially if the back wall of the artery is perforated and
a concealed bleed into the retroperitoneum occurs. If suspected after a groin puncture, and
time allows, a quick CT scan will confirm the diagnosis. Most such injuries in the femoral
artery can be explored and controlled from the groin.
Thromboembolic occlusion or dissection will require surgical exploration.
Pseudoaneurysms formed as a result of arterial catheterisation should proceed to surgical
repair if there are any signs of active bleed or impending rupture. In stable patients Duplex
sonography can confirm the diagnosis. Pseudoaneurysms < 2 cm in maximum diameter will
likely respond to thrombin injection and compression.
Central venous line insertion can inadvertently puncture the subclavian artery or
carotid artery. If the catheter is withdrawn, especially after a dilator has been used, sig-
nificant bleeding can occur. A bleed from the carotid artery can rapidly cause airway com-
promise. If the carotid artery is punctured and dilated, it will need formal exploration and
repair and the dilator should be left in place until control has been achieved. Surgical access
to the subclavian artery can be especially difficult and endovascular solutions should be
considered.
Endovascular procedures with catheters and balloons can also cause vascular trauma in
the form of perforation and rupture of vessels distant from the site of puncture.
Immediate control can usually be achieved by balloon tamponade followed by surgical
exploration.
Specific injuries of open surgery

Lumbar disc surgery, especially at L4/L5 level, can cause vascular injury to the aorta, IVC or
iliac vessels. Significant intraoperative bleeding may necessitate an immediate laparotomy
and repair. Patients can also present late with arteriovenous fistulae, which are ordinarily
best dealt with by endovascular means.
Hip and knee surgery beside fracture fixation have all been associated with injury to
major lower limb vascular structures. Vascular injuries after hip/knee arthroplasty/replace-
ment are rare but have been known to result in amputations.
Varicose vein surgery has resulted in femoral vein injuries, ranging from laceration to
stripping of the vein. Redo groin surgery and attempts to pass a vein stripper from distal to
proximal are significant causative factors. Blind attempts at applying haemostats can make
a bad situation worse. Compression will control bleeding until help is available. Repair may
be by simple lateral suture, patch or interposition grafting.
Accidental injection of sclerosant into arteries can produce significant tissue damage. If
that is suspected and the needle is still in situ, heparinised saline and an α-blocker should be
injected through it.
Laparoscopic surgery and especially the creation of a pneumoperitoneum have been
reported to cause major vascular trauma, which will require rapid recognition and treat-
ment if a successful outcome is to be achieved.
Hepatic artery injury during cholecystectomy is often best dealt with by ligation, rather
than attempts at repair. Significant hepatic damage is unlikely to occur because of the dual
blood supply of the liver.
A variety of other vascular injuries have been reported in association with abdominal
surgery, hernia repairs and gynaecological surgery among others. The application of the
general principles and approaches to vessels should allow control of such injuries.
References
1. Ly TV, Travison TG, Castillo RC, Bosse 5. Neschis D, Scalea T, Flinn W, Griffith B.
MJ, MacKenzie EJ. LEAP Study Group. Blunt aortic injury current concepts. New
Ability of lower-extremity injury severity Engl J Med 2008; 359: 1708–16.
scores to predict functional outcome after 6. Arthurs ZM, Sohn VY, Starnes BW.
limb salvage. J Bone Joint Surg Am 2008; Vascular trauma: endovascular
90: 1738–43. management and techniques. Surg Clin N
2. Fishman EK, Horton KM, Johnson PT. Am 2007; 87: 1179–92.
Multidetector CT and three-dimensional 7. Xenos ES, Abedi NN, Davenport DL
CT angiography for suspected vascular et al. Meta-analysis of endovascular vs
trauma of the extremities. Radiographics open repair for traumatic descending
2008; 28: 653–65. thoracic aortic rupture. J Vasc Surg 2008;
3. Barros D’ Sa AA, Harkin DW, Blair 48: 1343–51.
PH, Hood JM, McIlrath E. The Belfast 8. Rathlev NK, Medzon R, Bracken ME.
approach to managing complex lower limb Evaluation and management of neck
vascular injuries. Eur J Vasc Endovasc Surg trauma. Emerg Med Clin N Am 2007; 25:
2006; 32: 246–56. 679–94.
4. Taller J, Kamdar JP, Greene JA et al. 9. Newton EJ. Acute complications of
Temporary vascular shunts as initial extremity trauma. Emerg Med Clin N Am
treatment of proximal extremity vascular 2007; 25: 751–61.
injuries during combat operations: the new 10. Asensio JA, Forno W, Roldán G et al.
standard of care at Echelon II facilities? Visceral vascular injuries. Surg Clin North
J Trauma 2008; 65: 595–603. Am 2002; 82: 1–20, xix.
Chapter
16
Indications and management of
lower limb amputation
Mark Kay and Colette Marshall
Key points
• Peripheral vascular disease is the leading cause of amputation in the Western world
• Diabetics are 8–12 times more likely to suffer amputation
• Mortality and subsequent loss of the contralateral limb following amputation is
high
• The level of amputation requires careful consideration of the rehabilitative potential of
the patient, level and pattern of vascular disease and likely healing
• Careful preoperative assessment and periopertive care using the multidiscliplinary
team is essential for successful outcomes
• Epidural analgesia provides the best perioperative analgesia but does not prevent
subsequent phantom pain
• The general principles of amputation surgery apply to all sites of amputation
• Rehabilitation of the patient should start immediately postoperatively
• Postamputation pain is the commonest postoperative complication, is multifactorial
and requires thorough assessment and possible onwards referral to a multidisciplinary
pain team for management
• Phantom limb pain is common and effective treatments remain elusive
History of amputation
Amputation, derived from the latin amputare, ‘to cut away’, is one of the oldest surgical oper-
ations. The first recorded amputation appears in the book of the Vedas, written in Sanskrit
in India, dated between 3500 and 1800 BC. It records that the leg of Queen Vishpla was
amputated in battle, and after healing, an iron leg was fitted to enable the Queen to return
to the battlefield [1, 2].
The early descriptions of amputation by Hippocrates and Celsus focused on amputa-
tion for the treatment of gangrene. Hippocrates, in the latter half of the fifth century BC,
recommended amputation for gangrene of the joint below the ‘boundaries of blackening’
as soon as it is ‘fairly dead and lost its sensibility.’ Celsus described the use of ligatures
to control bleeding, although the use of cautery was more commonly used. By AD 100,
Archigenes and Heliodorus described amputation for the management of ulcer, tumour,
injuries and deformity.
War served as the impetus for surgical developments in amputation. The mid-fourteenth
century saw the arrival of gunpowder into the wars of Europe, and with it came a new era of
injury that required control of major haemorrhage and limb fracture.
Amboise Paré, surgeon to the colonel general of the French infantry, is most famous for
re-introducing the use of the ligature to control bleeding. Paré first employed the ligature in
amputation of the leg at the siege of Danvier in 1552, and published his technique in 1564.
Paré is also known for his work on prosthetics, having designed an artificial hand and above-
knee prosthesis with a knee joint. The introduction of the tourniquet in 1674, by the French
barber surgeon Morell, further advanced surgical practice of haemorrhage control during
amputation.
During the Napoleonic wars, Dominique Larrey, Napoleon’s chief of surgery, advocated
the use of amputation ‘on-site’ rather than waiting for transfer to conventional hospital
settings. This practice was continued in the American civil war with immediate transfer to
adjacent field surgical sites, where some 50 000 amputations were carried out.
The introduction of ether anaesthetic by Morton and Warren in 1846 and the antiseptic
technique by Lord Lister in 1867 reduced mortality from amputation and allowed better
healing of wounds. Prior to antisepsis, mortality following limb amputation due to sepsis
was approximately 60%. However, as medical advances have developed, so too have weapons
of increasing destruction. The First World War saw somewhere between 300 000 and 500
000 amputations carried out.
Epidemiology of amputation
In the Western world, peripheral vascular disease is the dominant cause of lower limb ampu-
tation. Implicit to this is an increased prevalence of diabetes, currently over 1.7 million in
the UK, obesity, together with an ageing population. Amputation carries with it a significant
economic burden, and a high mortality rate, approaching 50% in diabetics at 2 years. A
1.6% annual incidence of major limb amputation is reported in patients with intermittent
claudication [3].
The incidence of amputation is reported to be 8–12 times higher in diabetics than non-
diabetics [4]. The mortality rate in both groups increases with age, the level of amputation
and is twice as high in men compared to women. A half of all diabetics who have a major
amputation will lose their contralateral leg within 5 years [5].
The UK National Amputee Statistical Data Base reported a total of 5000 new amputee
referrals in 2005–2006. Seventy-two per cent of these were due to dysvascularity, 9% trauma,
7% infection and 2% neoplasia (10% other or not specified). More than two-thirds of lower
limb amputees referred were male (70%).
In the USA an estimated 134 000 amputations occur annually, such that there are approx-
imately 1.7 million amputees in the USA. Dysvascularity accounts for 82% of amputations,
with the highest incidence among males.
Rates of trauma- and cancer-related amputations have both declined by approximately
half over the past 20 years. However, unlike the UK, the risk of traumatic amputations in
the USA has increased steadily with age, reaching its highest level among people age 85
or older.
There is evidence that focused management can reduce the incidence of amputation.
Regional UK studies have shown significant reduction of lower limb amputation rates in
diabetics of 70–80%, following the introduction of diabetic foot care monitoring [6].
Chapter 16: Lower limb amputation 185
These improvements are unlikely to be due to any single factor. Dedicated diabetic foot
care services with a multidisciplinary approach, advances in radiology such as the use of
subintimal angioplasty, and tighter control of risk factors with statins, antihypertensives and
antiplatelet agents are all important developments.
Aetiology of amputation
• Critical limb ischaemia
• Sepsis
• Trauma
• Neoplasm
• Congenital
Critical limb ischaemia warrants revascularisation by angioplasty or bypass, to relieve rest
pain or for the management of tissue loss/gangrene, in order to avoid major amputation.
Amputation may become necessary after unsuccessful attempts at revascularisation or it
may be considered as a primary option to control symptoms. Some patients choose primary
amputation over potentially complex and lengthy surgery, or it may be recommended if they
have significant comorbidities. The aims of amputation are to improve patient function and
quality of life.
Sepsis, with or without, gangrene is most commonly seen in the diabetic foot, but can
result from systemic infections such as meningitis. The diabetic foot is subject to a combin-
ation of neuropathy, ischaemia and biomechanical changes that result in abnormal weight
bearing. Pressure ulceration ensues and ischaemia impairs healing, allowing superimposed
infection. Minor infection may be treated with antibiotic treatment, but diabetic foot sepsis
can progress rapidly and have devastating consequences if not managed appropriately. The
diabetic patient with foot sepsis and palpable pedal pulses has a good prognosis with early
surgical intervention. Digital gangrene is treated by amputation of the digit and metatarsal
head with the wound left open. However the extent of underlying infection is often more
extensive than is apparent externally, necessitating a higher and more major level of amputa-
tion. This should be borne in mind when consenting such patients. In diabetics with absent
pulses attempts at revascularistion should be made at the earliest opportunity, ideally prior
to surgery but this may not be practical until after.
Trauma was a common reason for amputation during wartime but in the UK is now
more commonly performed following traumatic injury sustained in road traffic or machin-
ery accidents. Thermal injury by severe burns or extreme cold exposure may also necessitate
amputation.
Bone or soft tissue tumours (e.g. osteosarcoma, fibrosarcoma) may require limb amputa-
tion as their management. Finally, congenital deformities such as polydactyly may require
amputation to improve patient function.
Preoperative assessment of the patient includes consideration of the level of amputa-
tion. Decision on the level of amputation should include the likely rehabilitative potential
of the patient. Patients with peripheral vascular disease often have concomitant disease
that will limit ambulation due to the energy expenditure required for more proximal
Table 16.1 Key considerations in decision making about the level of major lower limb amputation
Energy
expenditure
Level of above normal Ambulation
amputation (%) rate (%) Advantages Disadvantages
Below-knee 10 >80 Best chance Flaps require well-
amputation of restoring perfused tissues, up to
long stump ambulation 10% conversion rate to
above-knee amputation.
Not suitable for bed-bound
patients who are unlikely
to ambulate due to stump
contracture and pressure
ulceration
Below-knee 40 80 May excise poorly If stump is too short a
amputation perfused tissues prosthesis cannot be
short stump successfully used
Through- 71.5 31 Useful if Unpredictable healing of
knee orthopaedic skin flaps
amputation metalware Poor cosmesis from
present in femur prosthesis with leg
Provides long appearing to dangle
stump for balance further than the normal
in a wheelchair side
Above-knee 63 38–50 (less Excellent healing Vascular or elderly patients
amputation in vascular rates unlikely to become
patients) ambulant, poor balance
due to short stump
amputations. A summary of the key considerations for major amputations are given in
Table 16.1.
Preoperative assessment should involve a multidisciplinary approach with input from
physiotherapists, occupational therapists, prosthetists, rehabilitation medicine special-
ists, psychologists, nursing staff, and the surgical and anaesthetic teams. As well as the
rehabilitative considerations a decision on the level of amputation should also include an
assessment of the likelihood of the flaps healing. This should take into account the degree
of tissue loss and ulceration, pattern and severity of vascular occlusion and the viability
of tissues in the area of the proposed flaps. Toe pressures measured by Doppler should be >40
mmHg for successful healing of toe amputations. Several adjunctive tests exist to aid deci-
sion making, e.g. laser Doppler flowmetry, transcutaneous oxygen measurement or isotopic
measurement of skin blood flow. However, the role of these tests is currently unclear and
usually decision making is based on clinician judgement.
A checklist of the other preoperative tests and assessments routinely required are given
in Table 16.2. Additional, more detailed tests and referral on to the appropriate subspecialty
may be required for individual patients based on clinical findings. Amputation patients are
Table 16.2 Preoperative checklist of tests and assessments required routinely prior to major lower limb
amputation
Test/assessment
required Reason Action required
Full blood count Patients often anaemic from chronic Tranfuse if haemoglobin less than
ulceration or other causes 10 g dl–1
Raised white cell count may indicate Preoperative antibiotic treatment
ongoing infection may be required if clinical supporting
evidence of infection
Urea and electrolytes Vascular patients often have Perioperative intravenous
associated renal disease and often rehydration whilst patient is fasted
are on medications that may disturb Treatment of individual disturbances
electrolyte balance
Involve renal team if necessary
Clotting screen Vascular patients may often be on Stop anticoagulants perioperatively to
a variety of anticoagulants or may bring INR < 1.5. Patients at high risk of
have disturbed clotting function cardiac thromboembolism may need
secondary to sepsis perioperative intravenous heparin or
treatment dose low molecular weight
heparin – consult local protocols
Correct clotting disorders in
discussion with a haematologist
Crossmatch (usually Transfusion frequently required Monitor haemoglobin
two units) perioperatively due to high blood postoperatively
losses. Concomitant disease
such as ischaemic heart disease
dictates maintaining relatively high
haemoglobin levels
Blood glucose Many amputation patients are Monitor blood glucose
diabetic, good perioperative perioperatively, prescribe insulin
diabetic control is associated with sliding scale or glucose/potassium/
better outcomes insulin infusion for insulin
dependent diabetics depending on
local protocols
Put diabetic patients first on the
operating list
Electrocardiogram Patients often have concomitant Referral to cardiology if necessary
ishaemic heart disease based on clinical findings
Chest X-ray Only if clinically indicated for Delay surgery if possible until active
suspected acute or new chest problems treated
problems
Thromboprophylaxis Major amputation puts patients Prescription of low molecular weight
risk assessment at high risk of thromboembolic heparin
disease Avoid TED stocking in patients with
known peripheral arterial disease.
Table 16.2 (cont.)
Test/assessment
required Reason Action required
MRSA screening/ Guides use of perioperative Routine prescription of intravenous
microbiological antibiotics broad spectrum antibiotics at
screen Allows isolation of MRSA-positive induction of anaesthesia. Modify
patients according to swab results
Anaesthetic To assess and plan analgesic Instigate preoperative epidural
assessment requirements analgesia if possible
To plan safe anaesthesia Book bed at appropriate level of
To plan level of critical care facilities care (levels 1, 2 or 3)
required
Careful history and To detect unexpected conditions Inform operating surgeon and
examination that may impact on surgery e.g. other team members
previous orthopaedic metalware or
vascular prostheses
Consent Some patients undergoing In-depth discussion with patient and
amputation may be unable to give family about procedure, its
consent risks and impact on quality of life
Life-changing procedure that Two doctor consent form may
requires careful counselling need to be signed if necessary
Surgery high risk with associated
high mortality and morbidity rates
INR, international normalized ratio; TED thromboembolic deterrent; MRSA, methicillin-resistant Staphylococcus
aureus.
often very sick, frail and elderly. Careful attention to detail with delivery of optimal peri-
operative care is essential for successful outcomes.
Principles of amputation surgery

The underlying general principles of amputation surgery apply to all surgical sites:
• Avoid undermining or devitalizing skin flaps.
• Use a tourniquet to control haemorrhage – a recent randomized trial has shown that
this reduces blood loss but does not compromise healing [7].
• Ligate vessels as they are encountered.
• Divide nerves cleanly and away from bone ends to avoid neuroma formation.
• Presence of muscle that does not bleed or contract in response to diathermy
stimulation indicates devitalisation and that a higher level should be selected for
amputation.
• Guillotine amputation of highly-infected tissue with later stage completion of
amputation is indicated for severe sepsis and may reduce revision rates.
• Avoid unnecessary tension on the flaps and unnecessary bulk in the stump when closing.
• Use a suction drain/s for major amputation.
• Avoid stump bandaging, which can cause skin breakdown.
Figure 16.1 Toe amputation: amputation of diabetic gangrenous toes.
Lower limb amputation

Toe amputation
An evaluation of the vascular supply must be performed prior to toe amputation. Fishmouth
or circular incisions are used to excise the affected toe. Amputation should not be per-
formed through a joint as the avascular cartilage that is exposed prevents healing. Therefore
toe amputations are usually performed through the proximal phalanx (Figure 16.1).
Ray amputation
This involves removal of a toe through the metatarsal bone. A tennis racquet-shaped inci-
sion is used to expose the distal part of the metatarsal bone and amputation is performed
through the neck. Dissection must be kept close to the bone to avoid damaging the blood
supply to neighbouring toes. Tendon remnants need to be excised as far proximally as pos-
sible. Ray amputation usually involves subsequent near normal ambulation but amputation
of the hallux or fifth toe can sometimes cause ulceration of the plantar skin due to tendon
imbalances and abnormal weight bearing.
Transmetatarsal amputation
This involves amputation of all the toes through the mid-metatarsal bones. A plantar flap
is fashioned to cover the end of the foot. Excellent ambulation results from a well-healed
amputation.
Mid-foot amputation
For more proximal foot disease amputation can be carried out proximal to the metatarsal
bones. Named amputations include the Chopart amputation – a disarticulation of the tal-
onavicular and cacaneocuboid joints or the Lisfranc amputation between the metarsal and
tarsal bones. Other unnamed amputations at this level can also be successful. Healing rates
at his level may be unpredictable and ambulation limited by the development of equinus
varus deformity due to sacrifice of tendons. Mid-foot amputation should only be considered
in patients without or with corrected ischaemia.
Ankle-level amputation
The Syme and Pirogoff amputations at ankle level are seldom performed in vascular
surgical practice. The Syme amputation occurs through the ankle joint with the distal
tibia and fibula cut in line with the joint. The Pirogoff modification conserves a piece of
posterior calcaneum, which is screwed onto the distal tibia. Problems associated with
these amputations include difficulty fitting a prosthesis, posterior migration of the heel
pad and sloughing of the skin. The failure rate in dysvascular patients is approximately
41%. Usually below-knee amputation is preferable and allows a more reliable stump to
be fashioned.
Below-knee amputation (BKA)

The most common technique for BKA is the long posterior flap technique first described
by Burgess and Romano in 1967. The preferred site for section of the tibia is 14 cm below
the knee joint or 10–12 cm below the tibial tuberosity. The absolute minimum required for
successful limb-fitting is a length of 7.5 cm below the joint line. The flaps are marked out
accurately using the rule of thirds (Figure 16.2). The tibia is divided 1 cm proximal to the
skin wound and bevelled smoothly. The fibula is divided 2 cm proximally. The posterior flap
is fashioned by excluding soleus from the wound to provide suitable coverage for the bone
end that is not too bulky. An alternative technique uses skew flaps. This technique is useful
when a long posterior flap is unsuitable due to tissue loss or non-viability extending onto the
proposed site of the posterior flap. A systematic review has demonstrated comparability of
the two techniques [8].
Through-knee amputation
Through-knee amputation may be useful when a BKA is contraindicated due to tissue
nonviability. The through-knee amputation provides a long lever and an end-bearing
stump. The patella may be preserved and wired onto the end of the femur (Gritti–Stokes
Figure 16.2 Marking the

amputation flaps of a below-
knee amputation using the rule
of thirds.
amputation) or may be sacrificed. Problems with through-knee amputation include

unpredictable healing of tissues and a stump that cosmetically is bulbous with unequal
levels of the knee when fitted with a prosthesis. However, when successful it can pro-
vide good ambulation with less energy demands than a more proximal amputation. For
the non-ambulant patient the long stump provides good leverage for transfer and good
balance for sitting.
Above-knee amputation (AKA)

The preferred site of AKA is at the mid-femoral level at least 15 cm above the tibial plateau.
The aim is to achieve a stump long enough to act as a lever arm for locomotion whilst allow-
ing adequate clearance at the knee for a jointed prosthesis. The shortest stump recommended
is 8 cm below the inferior pubic ramus. The flaps are fashioned using fishmouth incisions to
achieve equal anterior and posterior myoplastic flaps.
Hip disarticulation and hindquarter amputation

These may occasionally be required in cases of severe ischaemia extending proximally, usually due
to occlusion of aorto-iliac inflow. Mortality is high and successful ambulation exceedingly low.
Upper limb amputation

Upper limb amputation is rare in vascular surgical practice and may result from trauma,
failed or delayed revascularization or infection. Amputations for congenital causes or malig-
nancy may be undertaken by other specialties. The general principle considerations for
amputations apply to amputations of the upper limb.
Postoperative complications
Early complications
• Deep vein thrombosis
• Flap necrosis
• Wound infection
• Post-amputation pain
• Stump haematoma
• Flexion contractures
• Psychological problems
Late complications
• Excess bulbosity of the stump
• Bone erosion through the skin
• Neuroma formation
• Ischaemia
• Osteomyelitis
• Adherent scar tissue
• Ulceration
Table 16.3 Causes of post-amputation pain
Type of pain Definition Treatment strategies

Immediate Pain occurring in the immediate postoperative period due to Opiate analgesia
postoperative pain the trauma of surgery Epidural analgesia
Phantom limb sensation Any sensory phenomenon (except pain) felt in the amputated Reassurance
limb
Phantom pain Pain felt in the amputated limb or a portion of the absent limb Referral to a multidisciplinary pain team. Useful approaches
may include: simple analgesia, opiates, amitriptyline,
gabapentin or pregabalin, ketamine, biofeedback,
transcutaneous nerve stimulation, psychological techniques,
spinal cord stimulation, use of metallic stump liners and
application of hot or cold to stump. No treatment has been
conclusively proven in robust randomized trials to be effective
Residual limb pain
Internal pain:
• Neuroma formation Overgrowth of nerve-ending in stump – may cause Nerve injections, amitriptyline, gabapentin or pregabalin
spontaneous or mechanically induced pain Surgical exision/revision of nerve
• Bony overgrowth New bony formation may produce spurs that cause local pain Revision of stump
• Ischaemic pain Coldness, blueness and pain associated with elevation in the Revascularisation using inflow procedure, amputation at
stump +/- ulceration higher level
• Sympathetically Rare type of pain maintained by the sympathetic nervous Referral to multidisciplinary pain team
mediated pain system and similar to complex regional pain syndrome.
Associated with coldness, blueness, increased sweating and
allodynia (pain associated with a stimulus that is not usually
painful)
• Neuropathic pain Neuropathic pain usually associated with diabetes may affect Amitryptilline, gabapentin, pregabalin
a stump
External pain:
• Mechanical causes from Ill-fitting prosthesis Adjustment of prosthesis
ill-fitting prosthesis
Myofascial pain Pain occurring at the supporting joint usually from an ill- Physiotherapy, adjustment of prosthesis
syndrome fitting or heavy prosthesis
Post-amputation pain
Because post-amputation pain is so common and challenging to treat it deserves special
mention. Post-amputation pain occurs in about 80% of patients following amputation. There
are several different causes, which are categorized in Table 16.3. A careful history and exam-
ination is the key to identifying the cause of the pain and a treatment plan can then be tai-
lored to the underlying cause. Phantom limb pain occurs in about 70% of patients. Despite
the wide variety of treatments described none has been conclusively proven to be of benefit
in preventing or treating phantom limb pain. Pre-emptive epidural analgesia does not pre-
vent phantom pain in the long term but does provide superior perioperative analgesia [9].
For cases recalcitrant to treatment with first-line oral agents referral to a multidisciplinary
pain team is recommended.
Rehabilitation and prosthetics

Rehabilitation of the amputee should start as soon as possible postoperatively.
Physiotherapy in the early period is important to prevent flexion contractures and to
commence practice in transferring, sitting in bed, etc. Once the stump is healed, elasti-
cated graduated compression stump socks are used to shrink the stump to an acceptable
shape for fitting of a prosthesis. Early walking aids allow the patient to stand and start
walking early. The Pneumatic Post Amputation Mobility Aid uses a maximal pressure of
40 mmHg to attach a pneumatic aid to the stump to achieve early walking in the physio-
therapy gym.
Limb fitting is usually delayed until about 6 weeks postoperatively when oedema in the
stump has subsided and it has shrunk to an acceptable size. A prosthesis is a device that is
designed to replace, as much as possible, the function or appearance of a missing limb or
body part. Advances in prosthetic technology have developed very sophisticated artificial
limbs that use microchip technology to anticipate and respond to movements. These can be
useful for young athletic patients to maintain prior levels of activity. However, for elderly
dysvascular patients simpler and lighter modular components are recommended. Modern
materials such as silicone or thermoplastic materials are used for stump liners and sockets
to achieve a good fit with the stump [10].
References
1. Ellis H. The Cambridge Illustrated History 5. Nathan DM. Long-term complications
of Surgery. Cambridge: Cambridge of diabetes mellitus. N Engl J Med 1993;
University Press, 2009; 127–8. 328: 1676–85.
2. Stansbury LG, Branstetter JG, Lalliss SJ. 6. Canavan RJ, Unwin NC, Kelly WF,
Amputation in military surgery. Trauma Connolly VM. Diabetes- and nondiabetes-
2007; 63: 940–4. related lower extremity amputation
3. Dormandy JA, Murray GD. The fate of the incidence before and after the introduction
claudicant. Eur J Vasc Surg 1991; 5: 131–3. of better organized diabetes foot
4. Johannesson A, Larsson GU, Ramstrand N care: continuous longitudinal monitoring
et al. Incidences of lower limb amputation using a standard method. Diabetes Care
in the diabetic and nondiabetic general 2008; 31: 459–63.
population. Diabetes Care 2009; 7. Choksey PA, Chong PL, Smith C,
32: 275–80. Ireland M, Beard J. A randomized
controlled trial of the use of a tourniquet 9. Halbert J, Crotty M, Cameron ID. Evidence
to reduce blood loss during transtibial for the optimal management of acute and
amputation for peripheral arterial chronic phantom pain: a systematic review.
disease. Eur J Vasc Endovasc Surg 2006; Clin J Pain 2002; 18: 84–92.
31: 646–50. 10. Tang PCY, Ravji K, Key JJ, Mahler DB,
8. Tisi PV, Callam J. Type of incision for Blume PA, Sumpio B. Let them walk!
below knee amputation. Cochrane Current prosthesis options for leg and
Database of Systematic Reviews 2004, Issue foot amputees. J Am Coll Surg 2008;
1. Article no. CD003749. 206: 548–60.
Chapter
17
Leg swelling and lymphoedema
Arun Balakrishnan and Tim Lees
Key points
• The commonest cause of lymphoedema worldwide is filariasis
• In the Western world the commonest cause is malignancy and its treatment
• Oedema is initially pitting, but becomes non-pitting due to fibrosis of subcutaneous tissues
• Diagnosis is confirmed by isotope lymphangioscintigraphy
• Satisfactory treatment can usually be achieved by conservative measures that include
manual drainage, compression hosiery, complex decongestive therapy and prevention
of infection
There are various conditions that cause chronic lower limb swelling (Table 17.1). The two
most common are chronic venous insufficiency and lymphoedema. Lymphoedema is a
debilitating condition that has no cure. Several million people are affected worldwide.
Definition
Lymphoedema can be defined as the accumulation of fluid rich in protein in the skin and sub-
cutaneous tissues due to a defect in the lymphatic system resulting in swelling of the limb.
Classification
Lymphoedema can be primary or secondary.
Primary
Primary lymphoedema can be classified based on the age of onset (Table 17.2). Women are
more likely to be affected than men. The lower limbs are more frequently affected than the
upper limbs.
• Congenital lymphoedema is present at birth and can be autosomally inherited (Milroy’s
disease). Oedema is present at birth and two-thirds of affected patients have bilateral
lymphoedema. It accounts for 25% of all cases.
• Lymphoedema praecox presents up to the age of 35 years, usually during adolescence. It
is the commonest form of congenital lymphoedema. Most patients have unilateral limb
involvement.
• Lymphoedema tarda presents after the age of 35 years. This is the least common form
and accounts for about 10% of cases.
Table 17.1 Differential diagnosis of chronic leg swelling
Venous disease
• Primary varicose veins
• Primary deep venous incompetence
• Post-thrombotic syndrome
• Arteriovenous malformations
Lymphoedema
• Primary
• Secondary
General disease
• Lipoedema
• Congestive cardiac failure
• Pretibial myxoedema
• Nephrotic syndrome
• Hepatic failure
Tumours
• Pelvic tumours causing extrinsic compression
Drugs
Dependency
All forms of primary lymphoedema are likely to arise from an abnormality present at birth
that may manifest at birth or later in life. It is likely that these three groups represent differ-
ent parts of the same spectrum of disease, which has been attributed to aplasia, hypoplasia
or hyperplasia of the lymph vessels during development.
Secondary
Secondary lymphoedema occurs when the lymphatic vessels become occluded by an
acquired pathology. The lymphatic channels distal to the obstruction become dilated and
the valves secondarily incompetent. Worldwide the commonest cause is an infection caused
by the parasite Wuchereria bancrofti, resulting in filariasis.
In the Western world the commonest cause is neoplasia and its treatment, resulting in
damage or removal of lymph nodes, for example post-mastectomy lymphoedema of the
upper limb. This is particularly common in patients who undergo radiotherapy following
axillary lymph node removal.
The causes of secondary lymphoedema are listed in Table 17.2.
Pathophysiology
Interstitial fluid is composed of proteins, lipids and water. It is formed by the high hydro-
static pressure in the capillaries that forces fluid into the interstitial space. This results in an
increase in the oncoctic pressure, which attracts more water. Normally functioning lymphat-
ics return interstitial fluid to the intravascular space. High molecular weight proteins and
associated water pass through the lymphatics and eventually into the venous system.
Chapter 17: Leg swelling and lymphoedema 197
Table 17.2 Causes of lymphoedema
Primary
• Congenital (age <1 year)
• Familial (Milroy’s disease)
• Non-familial
• Praecox (age <35 years)
• Familial
• Non-familial
• Tarda (age >35 years)
Secondary
• Malignant disease
• Surgery
• Radical mastectomy
• Radical groin dissection
• Radiotherapy
• Infection
• Parasitic (filariasis)
• Pyogenic ( β-haemolytic streptococci, Staphylococcus aureus)
• Tuberculosis
• Arterial surgery
• Venous disease and venous surgery
If a disease process has caused obstruction of the lymphatics, the above transport system
is overwhelmed. There is a stagnation of protein-rich fluid in the interstitial space and high
concentrations of protein result in accumulation of more water. The lymphatic vessels dilate
and this results in secondary valvular incompetence.
Limbs may swell due to other local or systemic conditions. Chronic venous insuffi-
ciency results in a rise in venous pressure. This pressure is transmitted to the capillary bed,
resulting in structural changes in the endothelium, causing an increase in capillary per-
meability. This results in an increase in interstitial fluid volume, causing oedema. Systemic
conditions such as congestive cardiac failure, hypoproteinemia and nephrotic syndrome
result in a state of retention of salt and water, causing oedema.This may compound the
situation in the presence of diseased lymphatics or overwhelm a normally functioning
lymphatic system.
Initially the oedema is pitting. The accumulation of protein and fluid in the interstitial
space results in an inflammatory reaction. This, in time, leads to fibrosis of subcutaneous
tissues and the oedema becomes non-pitting.
Presentation
Patients initially present with peripheral oedema. Lymphoedema can be differentiated from
other causes of limb swelling by history and examination. A detailed history and examina-
tion of the patient may help to differentiate primary from secondary lymphoedema.
History
The initial presentation is swelling of the limb of varying degrees. As the swelling progresses
the patient may have difficulty fitting into clothes and footwear. As the oedema progresses
skin complications develop. Bacterial and fungal infections are common. Patients may have
discomfort related to the size of the limb, cosmetic concerns, difficulty mobilizing and
impairment of activities of daily living. Primary lymphoedema occurs predominantly in
females in their early teens. Patients with secondary lymphoedema will commonly have a
history of previous surgery, neoplastic disease or radiotherapy.
Examination
The patient may have unilateral or bilateral limb swelling. Initially the oedema is pitting but
with time the swelling becomes non-pitting due to fibrotic changes in the skin and subcuta-
neous tissues. The swelling is uniform and as it progresses the leg appears like a tree trunk.
The skin gradually thickens and becomes less elastic. The dorsum of the foot (Figure 17.1) is
usually involved, producing the characteristic ‘buffalo hump’ appearance. The skin becomes
thickened and has a typical peau d’orange appearance. There are thick deposits of keratin on
the epidermis.
Investigation
The diagnosis of lymphoedema can usually be made clinically. Investigation is needed when
the diagnosis is uncertain, to confirm diagnosis, when considering surgery, to plan treatment
and to exclude pelvic masses as the cause for lower limb lymphoedema. Oedema associated
with generalized disorders such as hypoproteinemia, congestive cardiac failure and nephrotic
syndrome are excluded or diagnosed by examination, biochemical analysis and urinalysis.
Duplex ultrasonography
This is useful to exclude chronic venous insufficiency. Chronic venous insufficiency is the
commonest differential diagnosis for lymphoedema. Deep venous thrombosis may cause
unilateral lower limb swelling.
Figure 17.1 Chronic

lymphoedema of the foot.
Lymphangioscintigraphy (isotope lymphangiography)

This is the most commonly performed investigation (Figure 17.2). It defines anatomy, evalu-
ates dynamics and determines the severity of obstruction [1]. A radiolabelled (usually tech-
netium) colloid is injected into the interdigital space and gamma-camera pictures are taken
at 5-min intervals to assess transit through the lymph channels. If the time taken for the
tracer to appear in the regional lymph node is greater than 60 min then this would sug-
gest delayed lymphatic transport. A negative scintigram effectively excludes the diagnosis of
lymphoedema. Primary and secondary lymphoedema are frequently associated with simi-
lar scintigraphic appearances, including delayed transit, the presence of collaterals, dermal
backflow, and reduced uptake in one or more groups of lymph nodes. It can be used to dis-
tinguish between a venous and lymphatic cause of limb swelling [2]. The test cannot differ-
entiate primary from secondary lymphoedema.
Computed tomography
The primary role of computed tomography (CT) is in the diagnosis of primary and secondary
malignancy as a cause for lymphoedema.The common findings on computed tomography
are skin thickening, thickening of subcutaneous fat and thickening of the perimuscular apo-
neurosis. It will also provide evidence of lymphoedema by the presence of a honeycomb
appearance of fluid in the subcutaneous tissues [3] and has been used to monitor the response
to compression therapy. Patients with a previous history of pelvic or abdominal malignancy
9/01 Figure 17.2

Lymphangioscintigram.
20 min 50 min 80 min
R
should be scanned for recurrent disease in order to diagnose enlarged lymph nodes or pelvic
masses that may be compressing the lymphatic channels.
Magnetic resonance (MR) imaging

The MR features of lymphoedema include circumferential oedema, increased volume of tis-
sues, honeycomb pattern of subcutaneous tissues and thickening of the dermis. Nodal archi-
tecture can be demonstrated. It can distinguish between lymphatic and venous swelling but
is not good at separating primary and secondary causes of lymphoedema.
Contrast lymphangiography
Lymphangiography used to be the gold standard test for evaluating lymphatic disorders. It is an
invasive test and can cause an inflammatory reaction in the lymphatics. This investigation is now
rarely used in the diagnosis of lymphoedema and has been largely replaced by scintigraphy.
Treatment
The aim of treatment is to reduce limb swelling, reduce the risk of infection and improve
function. Treatment can be surgical or non surgical. If management begins early in the dis-
ease process, before irreversible fibrotic changes occur, then conservative measures should
be successful. Once achieved the improvement must be maintained. Surgery is indicated
only in a small proportion of patients, and is palliative and not curative.
General measures
There is no cure for lymphoedema. The nature of the condition and its management should
be clearly explained to the patient. General measures are of benefit and should be followed
religiously throughout life to obtain maximum benefit. These include the following:
• skin care is essential to maintain healthy skin and reduce risk of infection;
• elevation of the limb at rest;
• regular exercise to encourage flow of lymph;
• weight reduction;
• high protein and low sodium diet;
• avoiding pressure, tight footwear and constrictive clothing;
• simple lymphatic drainage, which involves gentle massage of the affected area by
patients or their carers [4].
Manual lymphatic drainage

This is performed by specially trained therapists. Beginning with the proximal portion of
the affected limb, the limb is massaged over short segments in a distal to proximal manner.
Massage stimulates the flow of lymph via superficial lymphatics from an affected area to an
adjacent normal area.
Compression therapy
Compression can be achieved by multilayer bandaging during the intensive treatment phase.
Multi Layer Lymphoedema Bandaging (MLLB) involves the use of several layers of bandages
to achieve compression [4]. The pressure is more uniformly distributed. Compression stock-
ings need to exert a pressure of approximately 40–50 mmHg at the level of the ankle and are
generally used during the maintenance phase. The compression should be graduated from
distal to proximal. Support garments are essential to maintain limb size in the treatment of
lymphoedema.
Intermittent sequential pneumatic compression therapy

Intermittent pneumatic compression is a means of reducing the size of a limb in lym-
phoedema. They can be used at home or in an outpatient setting. They work best if used
before subcutaneous fibrosis sets in. The affected limb is placed in a sleeve or cuff that is
alternately inflated and deflated, creating a pressure gradient that moves fluid out of the
affected limb. These devices may have a single sleeve of uniform pressure or may consist
of several chambers that can be inflated in sequence. The direction of inflation should
be graduated from distal to proximal. Compression hosiery should be used between
treatments.
Thermal treatment
Hyperthermia of the leg is produced by immersing the limb in hot water or by microwave
heating. The reason behind its efficacy is not clear. It is presumed to mobilize fluid and soften
tissues.
Complex decongestive therapy

Complex decongestive physiotherapy generally involves an intensive treatment programme
over 4 weeks. The first phase (intensive therapy) involves skin care, exercise, multilayer
bandages and manual lymphatic drainage. Phase 2 (maintenance phase) aims to conserve
and optimize the results obtained in Phase 1. It involves the regular use of support gar-
ments, continued exercise, skin care and massage therapy. Reduction in limb volume can
be achieved and maintained. There is a reduction in the incidence of infection. Worldwide,
people have achieved good results but this process is resource intensive and compliance can
be poor.
Prevention of infection
Macrophages and lymphocytes are activated by inflammatory processes and carried through
the lymphatics to regional lymph nodes. Antigens are presented and an immune response is
mounted. Stagnation of lymph prevents this and increases the risk and severity of infection.
The common pathogens are β-haemolytic streptococci and staphylococcus aureus. With
each episode of infection there is further destruction of the lymph channels, making the
oedema worse. Well-fitting comfortable shoes prevent small cracks in the skin that may act
as a portal of entry. The affected limb should be washed daily with a mild soap and the feet
must be dry before putting on shoes. The patient must keep a very careful eye on the foot and
any early signs of infection must be treated aggressively with antibiotics. Recurrent infec-
tion can be managed by long-term, prophylactic, low-dose antibiotics such as amoxicillin,
flucloxacillin or a cephalosporin.
A Consensus Document on the Management of Cellulitis in Lymphoedema has been
published by the British Lymphology Society and the Lymphoedema Support Network [6].
This document makes recommendations about the use of antibiotics for cellulitis in patients
with lymphoedema, and advises when admission to hospital is indicated. Prompt treatment
is essential to avoid further damage to the affected part, which in turn may predispose to
repeated attacks.
Drugs
Drug therapy for lymphoedema is limited. Diuretics can be used in the early stages but
have no value in long-term treatment. Underlying filarial infection should be treated with
diethylcarbamazine. Benzopyrenes have been advocated by some. They induce phagocytosis
of proteins and proteolysis. The resulting fragments are more readily removed.
Surgical treatment
Surgery is indicated only in a very small proportion of patients as symptoms are usually
controlled adequately by conservative measures. Surgery is indicated if conservative meas-
ures have failed and there is severe disability, gross deformity or lymphorrhagia. These
can be divided into debulking operations and bypass procedures. Obliterative causes are
best treated by debulking procedures, whereas in lymphatic obstruction surgical bypass
is recommended.
Debulking operations
These procedures, which involve removal of variable amounts of the excess skin and sub-
cutaneous tissue from the affected limb, are indicated when there is gross oedema:
• Homan’s operation involves making an incision along the length of the limb. Anterior
and posterior skin flaps are fashioned and raised. Excess subcutaneous tissue is excised.
Tissue is removed down to the level of the deep fascia. The skin flaps are then fashioned
appropriately and closed primarily. The skin should be reasonably healthy to carry out
this procedure.
• Charles’ procedure involves excision of skin, subcutaneous tissues and deep fascia. The
resulting defect is closed with a split skin graft. The cosmetic results of this procedure
are poor and the procedure can be complicated by hyperkeratotic scars.
• Liposuction has been used to reduce the size of a lymphoedematous limb. It is
generally more effective in patients with minimal pitting oedema. Post-procedure,
patients are required to wear compression garments for life. The National Institute for
Health and Clinical Excellence (NICE) has issued guidance regarding treating chronic
lymphoedema by liposuction. NICE has said that if a doctor wants to use liposuction
to treat chronic lymphoedema, they should make sure that extra steps are taken
to explain the uncertainty about how well it works in the long term, as well as the
potential risks of the procedure. The patient should be told that they will need to wear
compression garments indefinitely after the procedure. This should happen before
the patient agrees (or does not agree) to the procedure. The patient should be given
this leaflet (Treating chronic lymphoedema by liposuction, Interventional procedure
guidance 251) and other written information as part of the discussion. There should
also be special arrangements for monitoring what happens to the patient after the
procedure [6].
Table 17.3 Bypass procedures for lymphoedema
• Skin and muscle flaps

• Omental bridges
• Enteromesenteric bridges
• Lymphatico-lymphatic anastomosis
• Lymphatico-venous anastomosis
Bypass procedures
Several procedures have been described (Table 17.3). Only a small number of patients show
long-term improvement.
Patients with proximal obstruction of lymphatics in the pelvis and patent distal lymphat-
ics can have a bypass procedure. Of note are lymphovenous anastomosis and lympholym-
phatic anastomosis.They are physiological methods for correcting lymphoedema. Multiple
lymphatics are anastomosed to subdermal venules using micro-surgical techniques to prod-
uce an improvement in signs and symptoms. Autologus lymphatic vessels harvested from
the contra lateral normal limb can be used to bypass obstruction by performing lymphol-
ymphatic anastomosis. For the procedures to succeed the patients should have patent distal
lymphatics and mild-to-moderate oedema. In patients with distal obliterative disease recon-
struction is not an option. The techniques of omental and enteromesenteric bridges have
been described to help improve lymphatic drainage.
Lipeodema
This condition is often diagnosed as lymphoedema. It is characterized by symmetrical
enlargement of the lower limbs, excluding the feet. The age of onset is early and the con-
dition almost exclusively affects women. There is no cure for the condition. Treatment is
multimodal and mainly involves the use of exercise and compression hosiery. Liposuction
has been used with some effect. Patients require adequate support.
References lymphoedema of the lower limb. AJR Am J

Roentgenol 1985; 144: 361–4.
1. Mortimer PS. Evaluation of lymphatic 4. Lymphoedema Support Network. What is
function: abnormal lymph drainage in lymphoedema? 2004. www.lymphoedema.
venous disease. Int Angiol 1995; 14: 32–5. org (accessed 13 June 2009).
2. Brautigam P, Vanscheidt W, Foldi E, 5. British Lymphology Society. Cellulitis
Krause T, Moser E. The importance of the Management Consensus 2007. www.thebls.
subfascial lymphatics in the diagnosis of com (accessed 13 June 2009).
lower limb edema: investigations with semi 6. National Institute for Health and
quantitative lymphoscintigraphy. Angiology Clinical Excellence. Treating chronic
1993; 44: 464–70. lymphoedema by liposuction. 2008. www.
3. Hadjis NS, Carr DH, Banks L, Pflug JJ. nice.org.uk/IPG251 (accessed 13 June
The role of CT in the diagnosis of primary 2009).
Chapter
18
Varicose veins and chronic venous
insufficiency
Marcus Cleanthis and Tim Lees
Key points
• T runcal varices are common with an age-adjusted prevalence of 40% in men and 32%
in women
• Varicosities may be primary or secondary due to previous deep vein thrombosis
(DVT), pelvic obstruction or deep venous reflux
• Patients present with discomfort, swelling, pain, pruritis, bleeding, thrombophlebitis
and skin changes
• Hand held Doppler has replaced the tourniquet test as a ‘bedside examination’ for
evaluating incompetence of valves
• Current treatments for varicose veins include conventional surgery, radiofrequency
ablation, laser ablation and injection sclerotherapy (liquid or foam)
• Surgery to the great saphenous vein involves high tie and stripping and has a
recurrence rate of 15%–20%
• Surgery to the small saphenous vein involves saphenopopliteal ligation, stripping and
phlebectomy
• Foam sclerotherapy and catheter ablation techniques using laser or radiofrequency can be
used under local anaesthesia, but less is known about their long-term outcome
• Chronic venous insufficiency affects between 7% and 9% of the adult population and is
due to venous reflux, obstruction or calf pump failure
• Clinical signs include swelling, venous eczema, pigmentation lipodermatosclerosis and
ulceration
• Multilayered compression is the gold standard used to heal venous ulceration
• The Effect of Surgery and Compression on Healing and Recurrence (ESCHAR) trial
showed a reduced ulcer recurrence rate with combined surgery and compression
• Venous reconstruction is rarely performed and usually reserved for severe disease,
resistant to other forms of treatment
Varicose veins
Epidemiology
Data from the Edinburgh Vein Study suggest a slightly greater prevalence of varicose veins
in males compared to females (4 : 3 .2) [1]. The age-adjusted prevalence of truncal varices
Chapter 18: Varicose veins and chronic venous insufficiency 205
was found to be 40% in men and 32% in women. Mild varices (hyphenweb and reticular
varices) were identified in 80% of the population aged 18–64 years.
Asymptomatic disease detected with duplex scanning and defined as significant venous
reflux greater than or equal to 0.5 s is estimated to occur in 35% of the population between
the ages of 18 and 64 years. The prevalence of varicose veins increases with age and affects
approximately 15% of the population aged between 25 and 34 years and 50–60% of the
population aged 55–64 years [1].
Aetiology
Primary varicose veins
Primary varicose veins occur in the absence of any known underlying cause. Risk factors for
developing primary varicose veins include age, parity, weight, posture and bowel habit.
• Pregnancy: it is not uncommon for female patients to relate the onset of their varicose
veins to pregnancy and childbirth. The question as to whether these individuals have
underlying predisposition to venous disease or whether the pregnancy causes it remains
unanswered. Venous tone is affected by changes in female sex hormones and increased
blood volume. Interestingly most varices appear in the first trimester when uterine size
is unlikely to cause obstruction to venous return. Multiparity is also a risk factor for
venous disease with a 20–30% increased risk associated with two or more pregnancies.
• Weight: increased weight is a risk factor for varicose veins and the evidence for this
is stronger in women. The association with elevated body mass index (BMI) has been
shown in numerous studies. However, although the risk of developing varicose veins
has been shown to increase with increasing BMI there is no defined BMI level at which
this risk is greatest.
• Posture: it has been suggested that prolonged standing may exacerbate varicose veins.
Prolonged sitting has also been linked in some studies. The mechanism may be related
to prolonged, increased hydrostatic pressure making the veins more susceptible to
other factors.
• Diet: the effect of diet is suggested by the geographical variation in venous disease.
Varicose veins are more common in Western societies with low fibre diets. Western
diets result in straining during defacation associated with raised intra-abdominal
pressures. This pressure may be applied hydrostatically to leg veins.
• Genetics: patients frequently describe the occurrence of varicose veins in other family
members suggesting an inherited component to the occurrence. Although this is
plausible, the evidence for genetic factors is limited. The prevalence of venous disease is
so high that it is of no surprise that other family members are afflicted.
Secondary varicose veins
Secondary varicose veins occur as a consequence of another condition (post deep venous
thrombosis [DVT], pelvic tumours, congenital malformations, deep venous reflux and out-
flow obstruction). In these situations the superficial veins act as a collateral venous return.
Clinical features (Table 18.1)

Asymptomatic varicose veins
The majority of patients with varicose veins are asymptomatic. Many seek medical atten-
tion for cosmetic reasons. This may have psychological implications altering a patient’s
Table 18.1 Clinical features of varicose veins
Symptoms Types of varices Chronic skin changes Acute complications

Pain Corona Eczema Haemorrhage
phlebectatica Lipodermatosclerosis
Swelling Heaviness Hyphenweb varices Pigmentation Superficial
thrombophlebitis
Pruritis Reticular varices Ulceration
Burning Truncal varices Atrophie blanche
Restlessness
Cramps
Figure 18.1 Typical features of thrombophlebitis are

displayed. This patient presented with inflammation
affecting varicosities arising from the long saphenous
system.
MIMS Cardiovascular Journal 2007; 2 No. 3.
confidence. Such concerns should be identified and discussed prior to any intervention to
ensure that the expectations of any treatment are realistic.
Some patients have concerns about the risk of future complications. Many have a fear
of developing ulceration and some are concerned about the risk of deep venous throm-
bosis. Other patients are concerned about the risk of developing a DVT during flying.
These patients should be offered reassurance as not all patients develop skin changes. If a
patient is likely to be travelling on a long-haul flight they should take the same precautions
as a person without varicose veins. Many of the available treatments have an underlying
risk of DVT and hence treatment of varicose veins should not be offered as a form of DVT
prophylaxis.
Figure 18.2 Clinical features of chronic venous

insufficiency include swelling, pigmentation, eczema,
lipodermatosclerosis (shown above) and ulceration.
MIMS Cardiovascular Journal 2007; 2 No. 3.
Symptomatic varicose veins

Symptoms can be variable and may include pain, swelling, heaviness, pruritis, burning, rest-
lessness and cramps. These symptoms may be exacerbated by ambulation and alleviated by
lower limb elevation. Consequently symptoms may be more troublesome during the day or
in the evening after prolonged standing or sitting and less problematic in the morning after
the legs have been elevated overnight.
Presentation with acute complications of varicose veins is not uncommon. The acute
complications are haemorrhage and thrombophlebitis (Figure 18.1).
Haemorrhage can be torrential, occasionally life threatening and frequently asso-
ciated with trauma (often trivial). Bleeding is initially managed with compression and
elevation. However after a single episode patients frequently seek definitive treatment
to avoid further haemorrhage. It can be very distressing and subsequent treatment is
indicated to prevent further episodes. Superficial thrombophlebitis is a consequence of
thrombosis and subsequent inflammation of the vein. The clinical presentation is that of
an inflamed swollen tender vein. The early phase is managed with analgesics and anti-
inflammatory agents. It is not uncommon for patients to receive a course of antibiotics in
the community, although the process is that of a sterile inflammation. Subsequent fibro-
sis can result in a thickened nodule with subsequent skin pigmentation. As recurrence
is common many surgeons would regard this as an indication for surgery after the acute
phase has been controlled. Chronic complications (pigmentation, eczema, lipoderma-
tosclerosis (Figure 18.2) and ulceration) are more commonly seen with chronic venous
insufficiency (CVI) although they can occur with isolated superficial venous reflux.
Clinical signs
Varicose veins show features of tortuosity, elongation and dilatation. They are described as
hyphenweb varices, reticular or truncal varicosties.
• Hyphenweb varices are also referred to as telengiectasia, thread veins, spider veins and
venous flare. They are intradermal dilated venules occurring in isolation or associated
with truncal or reticular varices.
• Reticular varices are subcuticular varices that do not belong to the main trunk or its
tributaries. They frequently appear as a bluish reticular pattern of veins beneath the
skin and can occur physiologically.
• Truncal varices arise from the long or short saphenous veins or their major tributaries
(first and second order tributaries).
• Chronic venous skin changes include eczema, lipodermatosclerosis and ulceration.
• Eczema results in dry, scaly, itchy skin. Scratching can result in bleeding, infection and
subsequent ulceration.
• Lipodermatosclerosis is the term given to the pigmented skin associated with
thickened/inflamed subcuticular and cutaneous tissue. The mechanism is unclear but
it is believed that elevated venous pressure facilitates extravasation of cells and fluid,
leading to inflammation.
• Pigmentation/skin staining is a consequence of haemosiderin deposits.
• Chronic venous ulceration is most commonly (but not exclusively) seen in the gaiter
area above the medial malleolus. It commonly occurs in an area of pre-existing
lipodermatosclerosis following trauma. Long-standing ulcers are at risk of malignant
conversion to squamous cell carcinoma (Marjolin’s ulcer) and this should be considered
in long-standing ulceration and those that fail to heal.
• Other skin changes seen include malleolar flare/corona phlebectatica (intradermal or
subdermal collection of dilated veins at the medial malleolus) and atrophie blanche
(scar tissue at a site of previous ulceration).
Investigations for venous disease

Hand held Doppler
Hand held Doppler has replaced the tourniquet test in assessing the source of incompetence.
It is now the initial choice of investigation at the bedside. The test is performed with the
patient standing. The probe is placed at the junction suspected of being incompetent and the
calf muscle squeezed. This produces a characteristic venous signal. On releasing the muscle
the presence of an audible signal suggests reflux. The great saphenous vein, saphenofemoral
junction and saphenopopliteal junction can all be assessed although false positives are more
commonly seen at the saphenopopliteal junction. Hence reflux at this site should always be
confirmed with a duplex scan.
Duplex
Duplex combines B-mode ultrasound and Doppler. This allows identification of individual
vessels and can generate information on the direction of flow in a vessel. Hence both ana-
tomical and functional information can be obtained. As well as defining the communica-
tion points between the deep and superficial venous systems (perforators, saphenofemoral
and saphenopopliteal junctions) the direction of blood flow through these junctions and
within the veins themselves can be assessed and hence venous reflux confirmed. Duplex is
non-invasive, quick and cheap. It is the first-line imaging technique for venous disease and
frequently the only imaging ever required.
There is controversy regarding the use of duplex scanning for all patients with varicose
veins. Some centres reserve duplex for recurrent disease, small saphenous disease, deep ven-
ous disease and equivocal hand held Doppler examination of the long saphenous system.
Surgery may be offered on positive hand held Doppler examination of primary long saphe-
nous varicose veins. There is evidence to support the use of duplex scanning for all patients
presenting with varicose veins. Reliance on clinical examination alone can result in inappro-
priate surgical procedures being performed [2, 3]. The increasing availability of affordable
portable scanners allows quick and convenient scanning in the outpatient environment,
reduces diagnostic error and decreases clinic follow-up requirements prior to surgery.
Venography
Prior to duplex scanning investigation of the venous system was an invasive technique
requiring direct injection of contrast into the venous system. Duplex scanning has almost
abolished the need for this investigation. Occasionally more detailed imaging is required
and venography can be performed using magnetic resonance venography (MRV) and com-
puted tomography venograpy (CTV). These techniques have almost abolished the need for
contrast venography described below.
Magnetic resonance venography (MRV)

MRV is useful for imaging the venous system, especially within the main body cavities (tho-
rax, abdomen and pelvis) where duplex imaging may be limited (especially in the obese). It
can be used to diagnose deep venous occlusions, stenosis, thrombosis and malformations.
Various techniques exist including non-contrast imaging, which utilises time of flight tech-
niques and contrast enhanced MRV. The contrast-enhanced techniques have the advantage
of being quicker and less susceptible to flow artefact in parallel vessels. It is also able to
differentiate between acute and chronic DVT and is less invasive than conventional venog-
raphy. The greatest limitation with MRV is its cost compared to other forms of venography
and duplex. Magnetic resonance venography has largely replaced conventional venography.
Computed tomography venograpy (CTV)

CTV, like MRV, may provide better imaging of the iliac and caval venous system than duplex
scanning. Extrinsic compression of the iliac veins or vena cava (inferior and superior) can
also be identified and pelvic/abdominal masses can be visualised. Computed tomography
venography has become the investigation of choice for pulmonary embolism.
Contrast venography
Varicography requires the injection of the contrast into a superficial varicosity. It was previ-
ously used to identify perforators and localise the saphenopopliteal junction. This is now per-
formed with duplex scanning although varicography may still have a role in defining ovarian
reflux (e.g. patients with vulval varices) and in the investigation of venous malformations.
Ascending venography requires cannulation of a pedal dorsal vein. Contrast is encouraged
into the deep venous system via occlusion of the superficial system using a tourniquet above
the ankle. This technique was traditionally used to define deep occlusions, stenosis, throm-
bosis and perforator vein incompetence. In addition iliac and inferior vena caval imaging
can be obtained. Deep venous incompetence cannot be diagnosed using ascending venog-
raphy. Prior to duplex the diagnosis of valvular incompetence required descending venog-
raphy. This requires injection of contrast directly into the common femoral vein. Contrast
venography is usually reserved for investigation of the deep venous system when duplex and
MRV are equivocal.
Functional calf measurements

Plethysmography
Overall lower limb venous function is influenced by valvular function, venous outflow and calf
muscle activity. All of these will affect venous filling, which subsequently influences calf volume.
The change in calf volume can be quantified using plethysmography. Photoplethysmography
utilises infrared light absorption to quantify the change in blood flow within cutaneous veins.
This is correlated with ambulatory venous pressure refilling time. Air plethysmography
involves enclosing the leg in a bag and measuring changes in the pressure within the bag fol-
lowing ankle exercise. From these measurements, changes in calf volume can be measured. A
similar principle is used in foot volumetry where patients stand in water filled boots. Changes
in foot volume are detected by measuring the volume of water expelled.
Ambulatory venous pressure (AVP)

This is an invasive technique requiring the cannulation of a dorsal foot vein and direct meas-
urement of the superficial venous pressure by connection to a pressure transducer, ampli-
fier and a computer. It remains the gold standard method for measuring venous pressure.
Patients are asked to tip-toe ten times and the pressure changes are recorded. Although
AVP can provide detailed information its use is now largely confined to research of venous
disease.
Treatment
Reassurance
Many patients require no treatment other than reassurance. Patients with asymptomatic
varicose veins who have no cosmetic concerns can be discharged after the benign nature
of the disease has been explained. Patients who are due to fly on long-haul flights should
be advised to wear anti-DVT flight socks and exercise their calves as for patients without
varicose veins. The risk of developing DVT with varicose veins is not significant enough to
justify intervention and patients should be reassured of this, especially as intervention itself
is associated with a risk of DVT.
Compression stockings
These can provide symptomatic relief in patients with varicose veins and should be offered
to those who do not want any form of invasive intervention. They have an important role in
the management of chronic venous insufficiency and reduce ulcer recurrence after healing
with compression dressings. They create a graduated pressure on the leg, aiding the action
of the venous calf pump and improving deep venous blood flow, which in turn reduces
reflux into the superficial system. Compression stockings are classified (and should be pre-
scribed) according to the pressure applied at the ankle. These stockings may be described as
class I to IV but this classification should be avoided as significant differences exist between
the British and European standard pressure classification (Table 18.2). The most commonly
Table 18.2 Classification of compression stockings
European standard British standard

Indication Support class pressure class pressure
Mild varices, deep vein thrombosis, Light I 18–21 mmHg I 14–17 mmHg
prophylaxis
Marked varices, oedema, chronic Medium II 23–32 mmHg II 18–24 mmHg
venous insufficiency
Chronic venous insufficiency, Strong III 34–46 mmHg III 25–35 mmHg
lymphoedema, prevention of
venous ulcers
Severe lymphoedema, chronic Heavy IV >49 mmHg
venous insufficiency
used stockings are those that produce a pressure ranging between 25 and 35 mmHg at the
ankle. Before prescribing compression stockings it is vital to exclude co-existing occlusive
arterial disease. They should be used with caution in diabetic patients. Stockings can result
in ulceration in these groups of patients and hence close monitoring is required.
Foam sclerotherapy
Foam sclerotherapy is an advancement on a technique used for many years to treat varicose
veins. The original use of sclerosing agents for varicosities became less popular after the
recurrence rate was shown to be significantly higher compared to surgery. However, the
efficacy of sclerosing agents was found to be improved by mixing with air and administering
the agent as foam. This facilitated the displacement of blood from the vein and improved
contact with the vein wall.
Today, foam sclerotherapy injection is achieved under ultrasound guidance. The most
common agents used are sodium tetradecyl sulphate (1–3%) and polidocanol (0.5–3%).
These agents are converted to foam by mixing with air using two syringes and a 3-way tap.
Using ultrasound the great saphenous or small saphenous vein is cannulated with a venflon
and the foam administered in 1 ml aliquots. Smaller tributaries are cannulated with butter-
fly needles and treated with lower concentrations of foam. While the foam is administered,
the leg is elevated and the patient encouraged to dorsiflex the ankle. This latter manoeuvre
encourages blood flow in the deep veins, reducing the risk of deep venous thrombosis. After
treatment, compression bandaging is applied for one or two weeks. The more common
complications include thrombophlebitis, haemosiderin skin staining and ulceration from
extravasation of sclerosant (Table 18.3). Less common complications include deep venous
thrombosis, transient visual disturbance and stroke (three cases reported worldwide). Foam
sclerotherapy can be offered as an outpatient treatment and is also suitable for recurrent
varicosities. It is the first-line treatment for patients unfit for general anaesthetic. Long-term
studies are required to compare recurrence rates of foam with conventional surgery.
Catheter ablation
Catheter ablation is a technique used to generate heat resulting in transmural injury to the
vein wall. There are two techniques in common use: radiofrequency ablation (RFA) and
endovenous laser treatment (EVLT). One trial comparing RFA with surgery has shown that
RFA is associated with less postoperative pain and faster recovery compared to surgery [4].
Early results have shown that both are effective in eliminating reflux. Endovenous laser
treatment has been shown to produce less postoperative bruising and swelling compared
to surgery [5]. One trial has shown RFA to be superior to EVLT in postoperative pain and
quality of life parameters [6].
Radiofrequency ablation (RFA)

Radiofrequency energy is used to provide the thermal insult required to destroy the vein. Using
a Seldinger technique the vein (e.g. great saphenous vein) is cannulated under ultrasound guid-
ance and the RFA catheter inserted and guided to a position just distal to the saphenofemoral
junction. Bipolar electrodes on the end of the catheter are exposed and an alternating electrical
current passes between the electrodes and the vein wall. The cells of the vein wall offer resist-
ance to the passage of the alternating current and as a result of this resistance the cells become
heated and the vein ablated. Temperatures of 120°C are generated. Complications include
nerve injury (resulting in paraesthesia) and skin burns (Table 18.3). The risk is reduced or
prevented by using tumescence anaesthesia (ultrasound guided injection of anaesthetic solu-
tion into the facial envelope surrounding the vein). The solution injected usually consists of
a vasoconstrictor such as adrenalin and local anaesthetic diluted in saline solution. This type
of anaesthesia protects the surrounding tissue from thermal insult, anaesthetises and vaso-
constricts the vein compressing it against the catheter/electrode within the lumen. The latter
effect drains blood from within the vein and facilitates contact of the catheter with the wall of
the vein lumen. This is aided further by manual compression. Successful long saphenous vein
ablation is seen in approximately 85% of patients at 2 years [7].
Endovenous laser treatment (EVLT)

The main difference between EVLT and RFA is with the use of a laser diode to generate the
thermal insult required for ablation of the vein. The technique of vein cannulation (ultra-
sound guidance, Seldinger technique) and tumescence anaesthesia is comparable to that
of RFA. The catheter contains a laser diode that generates a laser (usually of 810 nm wave-
length). This heats blood around the laser tip, generating steam, which heats the vein wall.
The risks are the same as RFA, although the temperatures generated are higher and the
results probably comparable. Both RFA and EVLT avoid the complications associated with
groin incisions and thigh haematomas, and have improved postoperative recovery with less
postoperative pain. Both techniques are dependent on causing cell death in the vein wall
whilst avoiding perforation and luminal thrombus. Although there are few randomised con-
trolled trials comparing recurrence after EVLT with surgery, published case series indicate
successful early ablation rates of 80–90% with EVLT [8, 9].
Surgery
Saphenofemoral junction ligation and great saphenous vein stripping
Surgical treatment of varicosities arising from an incompetent saphenofemoral junction and/
or its tributaries is one of the most common procedures performed by the vascular surgeon.
Flush ligation of the saphenofemoral junction, division of its tributaries and stripping of the
great saphenous vein (ideally to a level of approximately one hand breadth below the knee)
has become the gold standard surgical procedure. Stripping of the long saphenous vein to
this level reduces the risk of reoccurrence, disconnects the vein from the more proximal calf
perforator (Boyd’s perforator) and minimises the risk of damage to the saphenous nerve.
Saphenous neuralgia and chronic pain was a complication associated with full length long
saphenous vein stripping. The technique of full length stripping has now been abandoned
because of this complication.
Saphenopopliteal junction ligation

Varicosities associated with reflux of the saphenopopliteal junction are treated by ligation
of the saphenopopliteal junction. Preoperative diagnosis of saphenopopliteal reflux requires
duplex confirmation, and prior to surgery all patients should have the saphenopopliteal junc-
tion marked with the aid of duplex ultrasonography. This is because there is considerable
anatomical variation in the location of the saphenopopliteal junction location. Significant
variation exists in the techniques for small saphenous vein surgery. Approximately 15% of
UK surgeons strip the proximal third and there is evidence to suggest this reduces the risk of
reoccurrence [10, 11]. Stripping of the small saphenous vein is not carried out routinely by
all surgeons due to the risk of sural nerve damage. However, the sural nerve joins the vein in
the distal two thirds of the calf. Many surgeons will excise a segment of the small saphenous
vein, in the popliteal fossa.
Multiple stab avulsions (phlebectomies)

The cosmetic part of the procedure takes the form of phlebectomies performed through
tiny stab incisions using a special vein hook. These should be sufficiently small enough to be
closed with small steri-strip adhesive dressings.
Perforator surgery
Opinion remains divided regarding the relevance of perforator surgery. Open surgical liga-
tion of venous perforators may be complicated by poor wound healing and recurrence. A
number of perforators identified as being incompetent on duplex scan later become com-
petent after treatment. Some ulceration associated with perforator incompetence together
with incompetence in either the long or short saphenous system improves with treatment of
the long/short system only. Hence the majority of surgeons would not treat perforators in
primary varicose veins without skin changes or ulceration. Some would reserve perforator
surgery for resistant venous ulceration not responding to treatment of either the long or
short systems. As well as the technique of open ligation, perforator surgery may also be per-
formed using the minimally invasive technique of subfascial endoscopic perforator surgery
(SEPS). This involves endoscopic dissection of the sub-fascial plane, identification of the
perforator veins and subsequent division. The port site is positioned away from the diseased
skin. However, despite this advancement, there remains a lack of evidence to support the use
of perforator surgery and randomised controlled trials are needed.
Recurrent varicose veins

There exist a number of reasons for developing recurrence. After excluding persistent vari-
cosities (present but not treated at the time of original surgery) recurrent varicose veins
may develop for a number of reasons. Neovascularisation is the development of a new
connection between the deep and the superficial vein at a previously divided junction. The
mechanism is not fully understood but is believed to be either the consequence of new
vessel growth at a site of a previous junction or the dilatation of pre-exisiting tributaries.
Table 18.3 Complications of treatment
Catheter ablation
(RFA/EVLT) Surgery Foam sclerotherapy
Bruising Bleeding/bruising Hyperpigmentation
Thrombophlebitis Groin infection/abscess Visual disturbance
Skin burns Nerve injury Skin necrosis
Nerve injury Sensory loss Deep vein thrombosis
Sensory loss Motor loss Thrombophlebitis
Neuralgia Neuralgia Stroke
Perforation of deep veins Deep vein thrombosis Recurrence
Erythema Venous flare
Deep vein thrombosis Residual varicosities
Recurrence Arterial/venous injury
Recurrence
RFA, radiofrequency ablation; EVLT. Endovenous laser treatment.
This is the commonest cause of recurrence at the saphenofemoral junction. Other causes
of recurrence include the development of incompetent perforators and the development of
varicosities in a second saphenous system.
Approximately one in five patients treated for varicose veins are likely to develop recur-
rent varicosities. Surgery for recurrent varicose veins can be hazardous with a greater risk of
injuring the deep veins, wound complications such as bleeding, infections, lymphatic leak-
age and seromas. In addition there is a greater risk of nerve injury especially at the saphe-
nopopliteal juction. The need for re-exploration of the groin or saphenopopliteal junction
has diminished with the expansion of techniques available to treat varicose veins. Foam
sclerotherpy, RFA and EVLT can all be used and hence reduce the need for complicated
groin and popliteal re-explorations.
Complications of treatment (Table 18.3)

When consenting patients for surgery, potential complications need to be discussed. General
complications include bleeding, infection (more common with groin surgery), recurrence,
nerve injury (paraesthesia, chronic pain), DVT, residual varicosities, venous ‘flare’, haema-
toma, arterial and deep venous injury (common femoral vein, popliteal vein). Great saphe-
nous vein stripping is associated with significant bruising to the thigh and patients should
be warned of this. Saphenopopliteal ligation may be associated with injury to the common
peroneal nerve and the risk of foot drop should be mentioned. The nerve can be affected by
local anaethesia and any foot drop that persists beyond the duration of the local anaesthetic
should be referred for immediate investigation and treatment (nerve conduction studies,
exploration, nerve repair/grafting or tendon transfer). Other nerve injuries can occur as
a consequence of the phlebectomies. These can result in numbness or chronic neuralgia.
Nerves at risk include the common peroneal, saphenous, sural and tibial nerves. The lat-
ter two are especially at risk when avulsing behind the malleoli. Neuropraxia has also been
seen following application of compression dressings and staff should be advised to release
Table 18.4 Aetiology of chronic venous insufficiency
Aetiology Example
Superficial venous reflux Great saphenous/small saphenous varicosities
Deep venous insufficiency Valvular damage e.g. post DVT
Venous outflow obstruction Stenosis/occlusion e.g. post DVT, venous cannulation
Calf muscle pump failure Immobility, obesity, prolonged sitting
DVT, deep vein thrombosis.
dressings should the patient complain of significant pain and numbness in the postoperative
recovery room. Any concern about perfusion to the foot following the application of dress-
ings should be treated in the same way.
Chronic venous insufficiency

Epidemiology
As with varicose veins, the prevalence of chronic venous insufficiency (CVI) is marginally
greater in men (9% of the population aged between 18–64 years) compared to women (7% of
the female population aged between 18–64 years) [1]. The prevalence increases with age. It is
a major cause of ulceration and has significant economic implications both in lost working
days and cost to the NHS.
Aetiology
This condition results from impaired venous return and causes elevated ambulatory venous
pressure within the lower limbs. As a consequence of this elevated venous pressure, skin
changes occur in the form of eczema, pigentation, lipodermatosclerosis and ulceration. It
is associated with lower limb oedema, varicose veins and chronic pain. There are a number
of causes including venous reflux, venous obstruction and ‘pump’ failure of the calf muscle
(Table 18.4).
The combination of normal venous anatomy and calf muscle contraction during exercise
is responsible for reducing venous pressure in the lower leg and encouraging venous return.
These mechanisms are impaired by venous reflux (affecting either the superficial, deep or
perforating venous systems), venous occlusions (e.g. DVT) or abnormal calf pump action.
Numerous causes exist for impaired calf muscle function and these range from neurological
causes to reduced mobility associated with morbid obesity. The relevance of perforator reflux
remains a topic of debate. Perforator reflux frequently occurs in association with superficial
or deep reflux. Surgical treatment of the superficial system alone may be associated with a
reversal of reflux in the perforator system. Ambulatory pressures have been shown to nor-
malise after superficial venous surgery but not after perforator surgery. Studies of perforator
surgery (open and SEPS) for chronic venous ulceration have been associated with a high
incidence of ulcer recurrence. It is for this reason that primary surgical treatment should
be directed at the superficial system and correction of perforator reflux only considered in
resistant cases.
The mechanism by which raised ambulatory venous pressure generates skin changes
is not fully understood. Numerous mechanisms have been suggested including the fibrin
cuff hypothesis and white cell trapping hypothesis. The fibrin cuff hypothesis suggests that
elevated pressure associated with chronic venous insufficiency results in capillary damage
leading to the deposition of fibrinogen. This subsequently forms fibrin and is allowed to
accumulate due to an impaired fibrinolytic system. Subsequently oxygen transfer is impaired
and the resulting local ischaemia results in ulceration. The white cell trapping hypothesis
suggests that the elevated venous pressure slows and halts the passage of large white cells
through the capillaries. These become ‘plugged’ with white cells. These cells are activated
and release proteolytic enzymes and free radicals causing local tissue damage. A series of
events results in the release of factors, which favour increased vascular permeability facili-
tating the formation of the fibrin cuff. The latter and the trapped white cells cause local
ischaemia and facilitate the formation of ulcers.
Clinical features
Patients complain of a variety of symptoms including swelling, itching, ache and heaviness.
The symptoms tend to be exacerbated by prolonged standing and relieved by elevation.
Initially, oedema is pitting but with chronicity becomes non-pitting. The venous eczema
causes pruritis, which stimulates scratching and subsequent skin trauma. This can be the
precipitating event leading to ulceration. The most common site of ulceration is the distal
medial calf. Pigmentation results from haemosiderin deposition and when accompanied by
fibrosis results in lipodermatosclerosis. Varicose veins may be present and superficial reflux
should always be excluded. A previous history of deep venous thrombosis is not uncom-
mon and symptoms of venous claudication (pain and swelling exacerbated with exercise,
requiring rest and elevation for relief of symptoms) suggest significant venous occlusion
(ilio-femoral veins).
Management
Non operative treatment
Certain co-morbidities are known to be associated with the development of chronic venous
insufficiency and these should be treated. Obesity and immobility should be improved. Leg
elevation can reduce ankle venous pressure to 15 mmHg and should be encouraged. Bed
rest and elevation above the level of the heart may be required for patients with ulceration
resistant to other forms of treatment.
Most patients with chronic venous insufficiency are treated with either compression stock-
ings or dressings. These apply a graduated compression that decreases proximally. The com-
pression improves venous blood flow in the deep venous system. Multilayered compression is
the gold standard and used to heal venous ulceration. Stockings are used to prevent reoccur-
rence. Below knee compression stockings of 25–35 mmHg are sufficient for most patients.
Operative treatment
Superficial venous surgery
The role of superficial surgery (e.g. ligation of saphenopopliteal junction, ligation of
saphenofemoral junction and long saphenous vein stripping) in ulcer healing has been pre-
viously studied [12]. The ESCHAR trial was a randomised trial involving 500 leg ulcers asso-
ciated with isolated superficial incompetence and showed the 1-year reccurrence rate of leg
ulceration to be reduced with combined superficial venous surgery and elastic compression
(9% recurrence) compared to compression alone (38% recurrence) [12]. The initial healing
rate was not affected. Superficial surgery is recommended in patients with isolated superfi-
cial reflux and a previous history of ulceration.
Perforator surgery
The role of perforator surgery has already been discussed. There remains a debate regarding
its benefits as correction of superficial reflux is frequently associated with improvement in
perforator reflux. The use of minimally invasive perforator surgery (SEPS) is associated with
less morbidity but usually reserved for resistant ulceration.
Venous reconstruction
Venous reconstruction is usually reserved for patients with severe ulceration, venous claudi-
cation and limb swelling that has not responded to compression and superficial/perforator
surgery. The reconstruction may be either in the form of a venous bypass or, less com-
monly, valvular reconstruction. A variety of valvular reconstructive procedures have been
described and include valvuloplasty, valve transplant, valve transposition, vein wall plication
and external banding with a Dacron or polytetrafluoroethane (PTFE) cuff.
If the deep venous system is obstructed then bypass is the reconstructive procedure of
choice. The most common cause for deep venous obstruction is deep venous thrombosis. A
number of patients will undergo recanalisation and others will develop collaterals allowing
for some improvement in symptoms. Venous bypass is usually reserved for patients who
are severely symptomatic despite allowing sufficient time (at least 12 months) for collateral
development. Other causes of deep venous obstruction include malignant disease, retro-
peritoneal fibrosis and May–Thurner syndrome. Iliac vein obstruction can be treated with
a Palma operation. Using the contra-lateral long saphenous vein a bypass is completed by
anastomosing the distal end to the common femoral or profunda veins in the affected limb.
In the absence of a suitable long saphenous vein a prosthetic conduit may be considered.
Deep thigh vein occlusion can be treated with a sapheno-popliteal vein bypass (May–Husni
operation). Again the long saphenous vein is the conduit of choice. This procedure is very
rarely performed. Reconstructive surgery is contra-indicated in early DVT, arterial disease
and thrombophilias.
Endovascular treatment
The techniques of endovenous catheter ablation (RFA and EVLT) and foam sclerotherapy
described to treat the superficial system in varicose veins can also be used for treatment of
the superficial venous system in chronic venous insufficiency. These are an alternative to
surgery. Endo-luminal angioplasty and stenting can be used for treatment of the deep sys-
tem. Long-term patency results are not known and have not been compared directly with
surgery.
Management of leg ulcers

Venous ulceration accounts for approximately 80% of chronic ulceration. A significant num-
ber of ulcers are of mixed arterial and venous aetiology. Significant arterial disease should
be treated especially if compression is to be considered. Neoplastic disease can be a cause
of chronic ulceration and may occur in long standing ulcers (squamous cell carcinoma,
Marjolin’s ulcers). Biopsy needs to be considered.
In addition to management of the underlying venous disease, consideration should be

given to appropriate skin dressings, emollients and nutritional supplements (vitamins and
trace elements). Adjunctive surgical techniques that may facilitate wound healing should
be considered e.g. wound debridement and skin grafts. Various techniques to faciltate
debridement exist and range from topical applications to surgical debridement. Skin graft-
ing may be considered to accelerate the healing. The graft is usually a split skin graft or
pinch grafts.
Post-phlebitic syndrome
After a DVT a number of patients will develop chronic venous insufficiency. This is fre-
quently referred to as post-phlebitic or post-thrombotic syndrome. Mild disease is seen in
up to one-third of patients whilst severe disease may affect one in ten patients. The risk is
increased with more proximal venous thrombosis. The underlying mechanism is of elevated
ambulatory venous pressure. As a consequence of valvular destruction, outflow obstruction
and calf muscle dysfunction ambulatory venous pressure fails to undergo the normal physio-
logical reduction seen during walking. A normal pressure drop of 50% can be seen dur-
ing ambulation compared to standing. In chronic venous insufficiency these pressures may
actually rise and can result in venous claudication. The early management of DVT is aimed
at preventing propagation of the thrombosis by anticoagulation. Elevation and compression
stockings are encouraged to minimise swelling. If detected early, preferably within 14 days,
proximal/ileo-femoral DVTs may be treated with thrombolysis with the aim of restoring
venous patency and minimising the loss of venous function. Long-term treatment includes
compression stockings and all patients are encouraged to preserve calf muscle function with
regular exercise.
References comparison of the costs. J Vasc Surg 2002;

35: 958–65.
1. Evans CJ, Fowkes FG, Ruckley CV, Lee AJ. 5. de Medeiros CA, Luccas GC. Comparison
Prevalence of varicose veins and chronic of endovenous treatment with an 810 nm
venous insufficiency in men and women laser versus conventional stripping of
in the general population: Edinburgh Vein the great saphenous vein in patients with
Study. J Epidemiol Community Health 1999; primary varicose veins. Dermatol Surg
53: 149–53. 2005; 31: 1685–94.
2. London NJM. Duplex ultrasonograpy and 6. Almeida JI, Kaufman J, Göckeritz O et al.
varicose veins. Br J Surgery 2007; 94: 521–2. Radiofrequency endovenous ClosureFAST
3. Makris SA, Karkos CD, Awad S, London versus laser ablation for the treatment
NJM. An ‘all comers’ venous duplex scan of great saphenous reflux: a multicenter,
policy for patients with lower limb varicose single-blinded, randomized study
veins attending a one stop vascular (RECOVERY study). J Vasc Interv Radiol
clinic: is it justified? Eur J Vasc Endovasc 2009; 20: 752–9.
Surg 2002; 32: 718–24. 7. Suramonia S, Lees TA, Wyatt MG,
4. Rautio T, Ohinmaa A, Perälä J et al. Oates C. Radiofrequency ablation in the
Endovenous obliteration versus treatment of varicose veins. In: Wyatt
conventional stripping operation in the WA, ed. Endovascular Interventions.
treatment of primary varicose veins: a Shrewsbury: TMF Publishing Ltd, 2004;
randomized controlled trial with 271–6.
8. Disselhoff BC, der Kinderen DJ, Moll of Great Britain and Ireland. Eur J Vasc
FL. Is there recanalization of the great Endovasc Surg 2004; 28: 400–3.
saphenous vein 2 years after endovenous 11. O’Hare JL, Vandenbroeck CP, Whitman
laser treatment? J Endovasc Ther 2005; B et al., Joint Vascular Research Group.
12: 731–8. A prospective evaluation of the outcome
9. Sharif MA, Soong CV, Lau LL, Corvan after small saphenous varicose vein surgery
R, Lee B, Hannon RJ. Endovenous laser with one-year follow-up. J Vasc Surg 2008;
treatment for long saphenous vein 48: 669–73.
incompetence. Br J Surg 2006; 93: 831–5. 12. Gohel MS, Barwell JR, Taylor M et al. Long
10. Winterborn RJ, Campbell WB, Heather term results of compression therapy alone
BP, Earnshaw JJ. The management of short versus compression plus surgery in chronic
saphenous varicose veins: a survey of the venous ulceration (ESCHAR): randomised
members of the vascular surgical society controlled trial. BMJ 2007; 335: 83.
Chapter
19
Management of deep vein
thrombosis
Dharmendra Garg and Vish Bhattacharya
Key points
• Deep vein thrombosis (DVT) and pulmonary embolism (PE) are the leading causes of
preventable in-patient mortality following surgery
• Virchow’s triad (stasis, hypercoagulable state, vessel wall injury) forms the basis for
DVT formation
• Many DVTs are asymptomatic
• Heparin prevents propagation by its action on antithrombin III
• D-dimer level measurements are useful screening tests
• Heparin must be overlapped with warfarin because of the transient hypercoagulable
state induced by warfarin
• Outpatient treatment is carried out with low molecular weight heparin (LMWH) and
warfarin
• Newer anticoagulants such as rivoraxaban and dabigatran are now being used for
prophylaxis
• Inferior vena cava filters can be used when anticoagulation is contraindicated
Background
DVT and its sequela, PE, are the leading causes of preventable in-hospital mortality [1, 2].
In 1846, Virchow recognized the association between venous thrombosis in the legs and
PE. Heparin was only introduced to clinical practice in 1937. Over the last 25 years, consid-
erable progress has been made in understanding the pathophysiology, diagnosis, treatment
and prevention of venous thromboembolism (VTE). Many DVTs are asymptomatic and
almost half of all fatal cases of PE are associated with asymptomatic DVTs.
Pathophysiology
DVT is multifactorial with interaction between hereditary and acquired risk factors.
The Virchow triad (i.e. venous stasis, hypercoagulable state, vessel wall injury), con-
tinues to serve as the unifying concept in the pathogenesis of DVT. However the sig-
nificance of interplay between the elements of Virchow’s triad and environmental or
Chapter 19: Management of deep vein thrombosis 221
acquired risk factors is also important. The formation, propagation and dissolution of
venous thrombi represent a balance between thrombogenesis and the body’s protective
mechanisms, specifically the circulating inhibitors of coagulation and the fibrinolytic
system. Several risk factors in combination are needed for thrombosis to develop and
the risk is cumulative.
Inherent hypercoagulable states include thrombophilias, which can be primary or
secondary [3]. Whilst important they only account for <5% of VTE events and are not
routinely screened for. Congenital thrombophilias include conditions such as AT III (anti-
thrombin III) deficiency, protein C and protein S deficiency. Antiphospholipid syndrome is
the most common cause of acquired thrombophilia. It is characterized by the combination
of antiphospholipid antibodies, such as lupus anticoagulant or anticardiolipin antibodies.
This syndrome is usually secondary to cancer or an autoimmune condition such as systemic
lupus erythematosus (SLE) [3].
Clinical features
Traditional clinical features, such as swollen tender calf, venous swelling, are present in
less than one-half the patients. Homan’s sign, which is pain or dorsiflexion of foot, is rarely
present. Phlegmesia alba dolens (phlegmasia = inflammation, alba = white, dolens = pain-
ful) was originally used to describe massive iliofemoral venous thrombosis and associated
arterial spasm. The affected extremity is often pale with poor or even absent distal pulses.
The physical findings may suggest acute arterial occlusion, but the presence of swelling,
petechiae, and distended superficial veins point to a DVT.
In rare cases, the leg is cyanotic from extensive DVT where thrombosis extends into the
collateral circulation, resulting in venous congestion. This ischaemic form of venous occlu-
sion was originally described as phlegmasia cerulea dolens or painful blue inflammation.
The leg is usually markedly oedematous, painful and cyanotic. Petechiae are often present
and the viability of the limb may be threatened.
Calf vein thrombi are usually benign unless there is propagation into the proximal veins.
Thrombi in the proximal veins can result in pulmonary emboli in 40–50% of cases and iso-
lated calf vein DVTs cause pulmonary emboli in about 15% of cases.
Assessment of risks
Surgical patients are at increased risk of VTE if they meet one of the following criteria (NICE
guidelines January 2010) [1].
• Surgical procedure with a total anaesthetic and surgical time of more than 90 minutes
or 60 minutes if the surgery involves the pelvis or lower limb
• Acute surgical admission with inflammatory or intraabdominal condition
• Expected significant reduction in mobility
• One or more of the risk factors noted in Table 19.1.
The Wells clinical prediction guide incorporates risk factors, clinical signs, and the pres-
ence or absence of alternative diagnoses (Table 19.2). This clinical prediction guide
quantifies the pre-test probability of DVT [4]. The model enables physicians to reliably
stratify their patients into high-, moderate- or low-risk categories. Combining this with
the results of objective testing greatly simplifies the clinical work up of patients with
suspected DVT.
Table 19.1 Patient-related risk factors for venous thromboembolism (VTE) [1]
• Active cancer or cancer treatment

• Age over 60
• Critical care admission
• Dehydration
• Known thrombophilia
• Obesity with a BMI over 30 kg m-2
• One or more significant medical comorbidities (e.g. heart, metabolic,
endocrine or respiratory pathology)
• Personal history or first degree relative with a history of VTE
• Use of hormone replacement therapy
• Varicose veins with phlebitis
Table 19.2 Wells clinical score for deep vein thrombosis (DVT)
Clinical parameter score Score

Active cancer (treatment ongoing, or within 6 months or palliative) +1
Paralysis or recent plaster immobilization of the lower extremities +1
Recently bedridden for >3 days or major surgery <4 weeks +1
Localized tenderness along the distribution of the deep venous +1
system
Entire leg swelling +1
Calf swelling >3 cm compared to the asymptomatic leg +1
Pitting oedema (greater in the symptomatic leg) +1
Previous DVT documented +1
Collateral superficial veins (nonvaricose) +1
Alternative diagnosis (as likely or greater than that of DVT) –2
Total of above score
High probability ≥3
Moderate probability 1 or 2
Low probability ≤0
Diagnostic evaluation
Because of the inherent inaccuracy of clinical diagnosis, the history, physical examination
and assessment of risk factors should be used to determine who requires further objective
diagnostic testing.
Objective diagnostic testing for acute DVT has changed considerably over the past two
decades. Invasive tests such as venous angiography or fibrinogen uptake test have now been
replaced by duplex ultrasound. In the fibrinogen uptake test, iodine labelled fibrinogen is
injected intravenously and gets taken up by the developing thrombus. This test has a high
sensitivity and detects many small DVTs that would normally resolve spontaneously. In add-
ition this test is inaccurate in detecting the more serious clots in the pelvis and thigh. Duplex
is 100% sensitive in detecting proximal thrombus but is less so for calf vein thrombi.
Figure 19.1 Right popliteal

vein showing deep vein
thrombosis.
D-dimer
Negative result Positive result
Low or moderate High Serial compression

Pretest probability pretest probability ultrasonography
No further testing Serial compression Positive result Negative result

Withhold anticoagulants ultrasonography
Positive result Negative result Treat Follow for 3 months
Treat Follow for 3 months
Figure 19.2 Systematic approach to the diagnosis of deep vein thrombosis using clinical prediction models,
D-dimer assays, and ultrasound.
D-dimer testing is commonly indicated to rule out VTE in low risk patients. D-dimer is
a degradation product of cross-linked fibrin with a high sensitivity in patients with clotting
activity. It is measured using ELISA (enzyme linked immunosorbent assay) [5]. A negative
D-dimer ELISA test effectively rules out VTE and the need for unnecessary ultrasound scans
is therefore reduced. An algorithm combining the use of a clinical prediction score and a
D-dimer assay is shown in Figure 19.2.
Information and assessment

• All surgical patients should be assessed to identify their risk factors for VTE (Table
19.1), ideally before admission.
• They should be given verbal and written information, before surgery, about the risks of
DVT and the effectiveness of prophylaxis.
• They should be informed that the immobility associated with prolonged travel in the 4
weeks before or after surgery may increase the risk of DVT.
• Oestrogen containing oral contraceptive should be stopped 4 weeks before elective
surgery although there is no added risk with progesterone only pills, implants or
injections.
• Patients should be given verbal and written information on the signs and symptoms of
DVT and PE and the correct use of prophylaxis at home as part of their discharge plan
if extended prophylaxis is required [1].
Prophylaxis against DVT [1]

Several options for prophylaxis are available (Table 19.3).
• VTE prophylaxis for patients undergoing gastrointestinal, gynaecological or thoracic
surgery and who are assessed to be at increased risk of VTE should be started on
mechanical prophylaxis at admission with anti-embolism stockings, foot impulse
devices or intermittent pneumatic compression devices.
• The stocking compression profile should be equivalent to the Sigel profile and
approximately 18 mmHg at the ankle, 14 mmHg at the mid-calf and 8 mmHg at the
upper thigh. These figures are based on Sigel’s work in 1975 when he showed that
optimal compression for elastic stockings to be used by hospitalized patients confined to
bed should be 18 to 8 mmHg (ankle to mid thigh). At this compression, average femoral
vein blood flow velocity is increased significantly. Graduated compression produced a
greater femoral vein flow velocity than uniform compression of the lower limb.
• Pharmacological VTE prophylaxis for patients who have a low risk of major bleeding
should include fondaparinux sodium, low molecular weight heparin (LMWH) or
unfractionated heparin (UFH). This should be continued until the patient no longer
has significant reduced mobility, generally for 5–7 days.
• NICE guidelines advise that in elective hip and knee replacement surgery and hip
fractures, in addition to mechanical prophylaxis as above, VTE prophylaxis can be
carried out with newer anticoagulants e.g. dabigatran etexilate and rivaroxaban, or
fondaparinux sodium, LMWH or UFH.
• Dabigatran is started 1–4 hours after surgery, fondaparinux 6 hours after surgical
closure and LMWH 6–12 hours after surgery. Rivaroxaban and UFH are started 6–12
hours after surgery.
• Patients who are already on antiplatelet agents and who are at increased risk of VTE
should have pharmacological prophylaxis if the risk of VTE outweighs the risk of
bleeding. Conversely, if the risk of bleeding is higher then mechanical prophylaxis
should be used.
• If patients are on vitamin K antagonists and are within therapeutic range no
prophylaxis is required provided anticoagulant therapy is continued.
Treatment of DVT
Unfractionated heparin
The primary objectives of the treatment of DVT are to prevent PE, reduce morbidity and
prevent or minimize the risk of developing postphlebitic syndrome. Because of the risk of
Table 19.3 Methods of venous thromboembolism (VTE) prophylaxis
• General measures
Early mobilization
Leg exercises
Hydration
• Mechanical methods
Antiembolism stockings (thigh or knee length)
Foot impulse devices
Intermittent pneumatic compression devices (thigh or knee length)
• Pharmacological methods
Heparin
(i) unfractionated
(ii) low molecular weight
Direct thrombin inhibitors
(i) dabigatran
(ii) rivaroxaban
Synthetic pentasaccharides
(i) fondaparinux
Anticoagulants e.g. warfarin
Antiplatelet agents e.g. aspirin, clopidogrel (not usually used alone for VTE prophylaxis)
• Vena cava filters
proximal propagation with the potential risk of pulmonary embolism or post-thrombotic

syndrome most authors believe that calf DVTs should be treated.
Anticoagulation remains the mainstay of initial treatment for DVT. Regular unfraction-
ated heparin was the standard of care until the introduction of LMWH. Heparin prevents
extension of the thrombus and has been shown to significantly reduce, but not eliminate, the
incidence of fatal and nonfatal PE, as well as recurrent thrombosis [6]. The primary reason
for this is that heparin has no effect on preexisting nonadherent thrombus. It does not affect
the size of existing thrombus and has no intrinsic thrombolytic activity. Heparin is a hetero-
geneous mixture of polysaccharide fragments with varying molecular weights. The key unit
is a pentasaccharide sequence that is responsible for binding to antithrombin III (AT III)
to inhibit thrombin. Heparin also activates the plasma protein heparin cofactor II, which
inactivates thrombin. At higher concentrations it also binds to factor IXa, resulting in the
modulation of factor Xa generation.
The optimal regimen for the treatment of DVT is anticoagulation with heparin or
LMWH followed by full anticoagulation with oral warfarin for 3–6 months. Warfarin ther-
apy is overlapped with heparin for 4–5 days until the international normalized ratio (INR)
is therapeutically elevated to 2–3. Heparin must be overlapped with oral warfarin because
of the initial transient hypercoagulable state induced by warfarin. This effect is related to the
differential half-lives of protein C, protein S and the vitamin K-dependent clotting factors
II, VII, IX and X.
Long-term anticoagulation is definitely indicated for patients with recurrent ven-
ous thrombosis and/or persistent or irreversible risk factors. When IV UFH is initiated
for DVT, the goal is to achieve and maintain an elevated activated partial thromboplas-
tin time (aPTT) of at least 1.5 times control. After an initial bolus of 80 U kg-1, a constant
maintenance infusion of 18 U kg-1 is initiated. The aPTT is checked 6 hours after the bolus
and adjusted accordingly. The aPTT is checked every 6 hours until two successive aPTTs are
therapeutic. Thereafter, the aPTT, the haematocrit level and platelet count are monitored
every 24 hours.
Heparin-induced thrombocytopenia (HIT) is a serious but uncommon side effect. Type
I HIT occurs in 10% and Type II in 5% of patients on heparin. Type I is due to a mild
lowering of the platelet count due to platelet clumping and occurs in the first 24 hours
of therapy. Type II HIT results from the binding of heparin to platelets and subsequent
generation of immune complexes consisting of immunoglobulin G and heparin platelet
factor 4. This occurs five or more days after starting therapy and can cause thrombosis.
Unfortunately, the subset of patients who develop thrombosis is unpredictable. Heparin
can rarely cause hyperlipidaemia, hyperkalaemia, osteoporosis, skin necrosis and hyper-
sensitivity reactions.
Low molecular weight heparin

LMWH is prepared by selectively treating UFH to isolate the low-molecular-weight (<9000
Da) fragments. Its activity is measured in units of factor X inactivation, and monitoring of
the aPTT is not required. The dose is weight adjusted. LMWH is administered subcutane-
ously (SC), and its half-life permits single- or twice-daily dosing. Its use in the outpatient
treatment of DVT and PE has been evaluated in a number of studies.
Indirect factor Xa inhibitors

A recent further refinement of the heparin molecule is fondaparinux, a synthetic form of the
pentasaccharide sequence of heparin that binds to antithrombin. It acts as a selective potent,
indirect antithrombin-dependent inhibitor of factor Xa. It is administered once a day via
subcutaneous injection and does not require laboratory monitoring. Despite its advantages,
it is not recommended for patients with renal failure and has limited use due to its parenteral
route of administration and lack of an antidote.
Direct factor Xa inhibitors

Factor Xa inhibitors include rivaroxaban and apiaxaban. Rivaroxaban is well tolerated,
has high bioavailability and predictable pharmacokinetics and is orally available. It does
not require dose adjustments and laboratory monitoring. The RECORD trials (four multi
centre double blind trials compared the efficacy of rivaroxaban compared to enoxaparin
for the prevention of VTE following hip and knee replacement) showed significant reduc-
tion in primary efficacy end points (DVT, PE or death from all causes). Major bleeding
was noted to be slightly higher in the rivaroxaban group although this was not statistically
significant.
Direct thrombin inhibitors

Dabigatran and ximelagatran are a new class of drugs that act by binding directly to throm-
bin. There is no need for routine monitoring and drug interaction is rare. Clinical trials with
dabigatran in patients undergoing hip and knee replacement surgery showed significantly
less VTEs as compared to enoxaparin although bleeding risks were higher.
Other strategies
Inferior vena cava (IVC) filters [7]
IVC filters should be considered for surgical inpatients with recent (within 1 month) or exist-
ing DVT and in whom anticoagulation is contraindicated. They may be indicated in patients
requiring abdominal surgery when anticoagulation is contraindicated or in trauma patients.
Other indications include recurrent PE while being adequately anticoagulated or if there is
a complication with the anticoagulation therapy requiring its termination. Designs of filters
have improved since Greenfield first described his filter in 1973. Temporary and retrievable
filters are also available. Most of these can now be placed via 6–12 French sheaths using a
percutaneous route under fluoroscopic or ultrasound guidance. They are usually inserted via
the jugular vein route. Complications include thrombosis of the filter and migration of the
filter through the walls of the IVC.
Thrombolysis [8]
The indications for thrombolysis in acute DVT remain controversial and Level 1 evidence is
missing for its long-term benefit. Thrombolysis can be performed for acute (<10 days) ile-
ofemoral DVT when there is no contraindication to use of anticoagulants or recent bleeding
episodes, trauma or recent surgery. It is usually reserved for patients with major ileofemo-
ral thrombosis with significant acute swelling and symptoms. Techniques vary considerably
and are rapidly developing [9]. Venous catheterization can be performed via the internal
jugular, femoral or popliteal vein, depending on the site of thrombus and treatment aims.
Infusion of thrombolytic agents such as r-tPA is performed via a multiple side hole catheter
embedded into the thrombus. A Cochrane review has shown that complete clot lysis and
improved patency is more often seen with thrombolysis than conservative treatment with
anticoagulation [8]. Venous function is significantly improved and there is less post-throm-
botic syndrome. Usually a temporary caval filter is used in addition to prevent PE while lysis
is being performed. Acute venous thrombectomy is now rarely performed.
Mechanical devices for thrombus removal have been described but none are currently
approved. Some of them have a jet, suction or a brush tip to allow disruption and aspiration
of the clot. However all of these need IVC filters at the time of use to prevent PE.
Post-thrombotic syndrome (PTS)

This is a condition that appears up to several years after DVT, manifesting as chronic swell-
ing, pain, varicose veins, venous eczema, pigmentation and venous ulcers. It probably occurs
because of damage to deep venous valves, resulting in chronic venous hypertension. It affects
up to 40% of patients after DVT and is more common after extensive iliofemoral DVT. There
is evidence that long-term use of graduated compression stockings after DVT can reduce the
incidence of PTS.
Summary
It is essential to define the preoperative risk of DVT and use best practice advice to reduce the
risk in inpatients undergoing surgery. Treatment and care should take into account patients’
needs and preferences. UFH, LMWH, fondaparinux and newer agents such as rivaroxaban
and dabigatran are the main pharmacological agents currently used in prophylaxis and
treatment of DVT. In the past decade, tremendous advances have been made with respect
to the diagnosis and treatment of DVT. The use of invasive techniques to diagnose DVT has
been completely replaced with accurate, yet noninvasive, diagnostic modalities. The therapy
of DVT is undergoing exciting changes with the development of targeted antithrombotics
with greater therapeutic efficacy and safety, as well as considerably greater ease of use.
References 5. Bates SM, Kearon C, Crowther M et al. A

diagnostic strategy involving a quantitative
1. National Institute for Health latex D-dimer assay reliably excludes deep
and Clinical Excellence. Venous venous thrombosis. Ann Intern Med 2003;
thromboembolism: Reducing the risk 138: 787–94.
of venous thromboembolism (deep vein 6. Hirsh J, Warkentin TE, Shaughnessy
thrombosis and pulmonary embolism) in SG et al. Heparin and low-molecular-
patients admitted to hospital. NICE Clinical weight heparin: mechanisms of action,
Guideline 92. January 2010 http://www. pharmacokinetics, dosing, monitoring,
nice.org.uk efficacy, and safety. Chest 2001;
2. Geerts WH, Bergqvist D, Pineo GF, 119: 64S–94S.
Heit JA, Samama CM, Lassen MR, 7. Young T, Tang H, Aukes J, Hughes R.
Colwell CW; American College of Vena caval filters for the prevention of
Chest Physicians. Prevention of venous pulmonary embolism. Cochrane Database
thromboembolism: American College of of Systematic Reviews 2007, Issue 4. Art.
Chest Physicians Evidence-Based Clinical no. CD006212. DOI: 10.1002/14651858.
Practice Guidelines, 8th edn. Chest 2008; CD006212.pub3.
133(6 Suppl): 381S–453S. 8. Watson LI, Armon MP. Thrombolysis
3. Rosendaal FR. Venous thrombosis: a for acute deep vein thrombosis.
multicausal disease. Lancet 1999; Cochrane Database of Systematic Reviews
353: 1167–73. 2004, Issue 3. Art. no. CD002783.
4. Anand SS, Wells PS, Hunt D et al. Does this DOI: 10.1002/14651858.CD002783.pub2.
patient have deep vein thrombosis? JAMA 9. Comerota AJ, Gravett MH. Iliofemoral vein
1998; 279: 1094–9. thrombosis. J Vasc Surg 2007; 46: 1065–76.
Chapter
20
Infection in vascular surgery
Mike Clarke
Key points
• Infection of prosthetic vascular grafts is associated with a high mortality and morbidity
• Prosthetic grafts should be avoided if the risk of infection is high
• Early diagnosis requires a low index of suspicion
• The greatest chance of long-term success lies in complete removal of the infected
prosthesis and revascularisation with autologous material
• A groin abscess in an intravenous drug abuser should be considered to be an infected
false aneurysm of the femoral artery until positively excluded
Introduction
Managing the infective complications of arterial surgery represents one of the most complex
challenges facing the vascular surgeon. Medical management alone seldom produces a sat-
isfactory outcome but the removal of an infected prosthesis in a debilitated patient, possibly
in the face of life-threatening haemorrhage, is rarely straightforward. The problem of then
restoring distal perfusion may require innovative approaches whilst minimising the risk to
the patient’s life and reducing the likelihood of recurrent infection.
Epidemiology
Conventional surgical teaching is that ‘clean’ operative procedures should carry a post-
operative wound infection rate of less than 1% (Table 20.1). Data from the Health Protection
Agency (HPA) surveillance of surgical site infection rates however suggests that this is rarely
achieved and approaches the sort of rates generally seen with clean-contaminated or con-
taminated procedures [1].
Causative organisms
Forty-six per cent of organisms seen in early postoperative infections following vascular
procedures are staphylococcal, two-thirds of these being methicillin resistant Staphylococcus
aureus (MRSA). Coagulase negative staphylococci are an unusual cause of early infection but
may be more important in the development of late-presenting graft infections. Enterococcus
and Enterobacter species are responsible for around 30% of early postoperative infections
with the remainder consisting of Streptococci, Pseudomonas, anaerobes, fungi and other
Table 20.1 Incidence of infection following surgical procedures in England 1997–2005
Category No. of operations Infected (%)

Knee prosthesis 62 031 1.0
Total hip prosthesis 74 677 1.8
Coronary artery bypass graft 27 447 4.1
Vascular surgery 8 959 6.0
Large bowel surgery 14 296 9.2
Limb amputation 2 670 13.1
Table 20.2 Rates of Szilagyi grade III infection following arterial reconstruction by
nature of graft and anatomical location
Graft type Rate of grade III infection

Synthetic graft 1.9
Autogenous vein 0.4
Endarterectomies 0.2
Arterial allografts 0.4
Site of surgery
Aorto-iliac 0.7
Aorto-femoral 1.6
Femoro-popliteal 3.0
Femoro-femoral 0.9
Other (intrathoracic, carotid, visceral) 1.0
bacterial species. Although MRSA is the commonest infecting organism, most studies sug-
gest that outcomes are no worse with MRSA infection compared with other staphylococcal
infections.
The HPA data refer specifically to infection rates during the original hospital stay and
of these the vast majority relate to superficial wound infections. Nevertheless the data is
in keeping with some historical series that have suggested the rate of vascular prosthesis
implant infection may be as high as 7%.
Infection in following arterial reconstruction varies from simple superficial wound
infection to deep infection involving the arterial conduit itself. The classification system
proposed by Szilagyi in the 1970s remains relevant to vascular surgical practice today [2].
Grade I – dermis only.
Grade II – involvement of subcutaneous tissues.
Grade III – vascular graft involvement.
Although it is grade III infections that are of the greatest concern, it is recognised that
early grade I and II infections are associated with the later development of grade III infections
and so should always be taken seriously. Table 20.2 shows how rates of grade III infection
vary by the nature of the conduit and the site of arterial surgery. As can be seen, prosthetic
grafts are more at risk than autologous forms of reconstruction and risk is also increased by
surgery involving the groin.
Chapter 20: Infection in vascular surgery 231
Prevention
Although not always possible, there is currently a keen focus on reducing infection rates in
surgical patients. Some factors cannot be easily altered. Patient age and preoperative health
status (including the presence of diabetes) have both been shown to be associated with
higher instances of postoperative infection. Other risk factors include prolonged preopera-
tive hospital stay, the presence of non-healing ulcers or other tissue loss, admission from a
long-term care facility and excessive antibiotic usage.
Conduct of the operation

Surgical technique can go a long way to preventing infection following surgery. Simple
measures include preoperative washing of patients; reducing theatre traffic to a min-
imum; careful attention to the handing of tissues; avoidance of excessive retraction with
self-retaining retractors; preoperative marking of the long saphenous vein to avoid under-
mining skin flaps, careful skin closure without undue tension and with suture techniques
that avoid devascularising the skin edges. Other factors that may help reduce periopera-
tive surgical site infection include maintenance of normothermia during surgery and
careful glycaemic control in patients with diabetes. Approximately 5% of patients under-
going arterial surgery will be colonised with MRSA prior to surgery. Preoperative screen-
ing for MRSA allows attempts at eradication therapy prior to admission and patients
known to be colonised with MRSA should not be nursed alongside patients who are free
from MRSA.
Antibiotic prophylaxis
It is important to develop a local antibiotic prophylaxis policy in conjunction with your
microbiologists. Consideration should be given to observed patterns of infection in vascular
patients as well as sensitivity and resistance patterns amongst organisms encountered. As
well as routine prophylaxis, it is well worth the policy including an alternative regime for
patients with known sensitivities or renal/liver impairment that precludes them from receiv-
ing the standard regime as well as an option for second-line prophylaxis for those patients
requiring early re-exploration.
Diagnosis
Graft infections may present in a multitude of ways and it is important to maintain a high
index of suspicion. For superficial grafts, the diagnosis may be relatively straightforward
with, for example, an obviously infected false aneurysm within the groin or a discharg-
ing sinus overlying a graft. Any signs of bleeding should raise the possibility of anasto-
motic disruption and all too often a small herald bleed will be followed soon afterwards by
major haemorrhage. The diagnosis of infection related to intracavity grafts can be some-
what more complicated. Patients presenting with upper gastro-intestinal haemorrhage fol-
lowing previous aortic surgery should be considered to have an aorto-enteric fistula until
proven otherwise. In many cases however, the presentation is more subtle and may include
pyrexia of unknown origin, general malaise and weight loss, vertebral osteomyelitis and
hydronephrosis.
Laboratory investigations
Routine laboratory investigations can provide evidence in support of the diagnosis of graft
infection but most are non-specific. They should, in the first instance, include: full blood count;
erythrocyte sedimentation rate; C-reactive protein; urea and electrolytes; serum albumin; micro-
scopy; culture and sensitivity of any wound or sinus discharge; and peripheral blood cultures.
Imaging
Imaging plays an essential role in the evaluation of patients with suspected arterial graft infec-
tion. In particular it can potentially confirm the presence of infection or the development of
local complications, e.g. false aneurysm formation, hydronephrosis. In addition information
can be obtained regarding proximal and distal vessels for planning reconstruction.
Ultrasound
Often useful in situations where a graft is placed superficially, ultrasound can quickly and
easily detect the presence of peri-graft collections and the presence or absence of false aneu-
rysms. In the carotid territory, a distinct rippling of the surface of an infected patch has
been reported as being visible on ultrasound. Ultrasound guided aspiration of a peri-graft
collection may provide the opportunity for microbiological confirmation of infection. In the
absence of overt clinical signs of local infection one has to consider the possibility of intro-
ducing infection into an otherwise sterile field or precipitating other local complications and
such aspiration should only take place under the guidance of a vascular surgeon.
Computed tomography (CT)

For intracavity grafts, CT remains the most useful method of imaging currently in use. The
presence of a gas-filled fluid collection adjacent to the graft more than 3 months after implan-
tation can be considered pathognomonic of infection. Such collections may be present in the
first few weeks following implantation and at these times interpretation needs to be carefully
considered alongside the clinical features. In these circumstances and if time allows, serial
scans may be helpful in distinguishing normal postoperative change, which should progres-
sively resolve versus progressive postoperative infection. Other features that may suggest
infection include simple fluid collections, ‘streaking’ of surrounding fat planes or the pres-
ence of adjacent complications such as an anastomotic aneurysm.
In addition to confirming the diagnosis, CT angiography also offers the opportunity to
outline the inflow and run-off vessels, thus avoiding the need for supplementary vascular
imaging (Figure 20.1).
Angiography
Digital subtraction angiography provides little information on the presence or extent of
infection. It can be helpful in providing detailed images of the run-off vasculature but rarely
adds significantly more than CT.
Magnetic resonance (MR) scanning

Magnetic resonance offers many of the advantages seen with CT scanning including the
diagnosis of intracavity graft infection and imaging of adjacent vasculature. T2 weighted
Figure 20.1 Computed

tomography showing an
infected aortic stent graft 2
years after insertion. Note the
presence of gas within the
aortic sac.
images may be more sensitive in identifying peri-graft oedema prior to the development of
a frank collection. There is, however, little firm data to support its use in place of CT, which
is currently the more commonly utilised modality in UK practice.
Labelled white cell scans

Radionuclide-labelled white cell scintigraphy with technitium-99 and indium-111 have
both been used in the diagnosis of vascular graft infection. They may be a useful adjunct
to CT, particularly in confirming early graft infection although the results need to be
interpreted with caution as although sensitive, the technique suffers from relatively low
specificity.
Treatment strategies
In many instances, the ideal treatment strategy is excision of the infected prosthesis (if one
is present), debridement of adjacent involved arterial wall and in situ reconstruction with
autogenous vein. In the case of the infected lower limb prosthetic graft, the autologous
saphenous vein is the most obvious choice, although in many cases, the reason for implant-
ing a prosthetic graft in the first place has been the unsuitability of the saphenous vein.
Consideration should be given to the contralateral long saphenous vein (LSV), if present, as
well as short saphenous veins and upper limb veins.
In situ vein graft

Where the long saphenous vein is going to be of insufficient calibre, the superficial femo-
ral veins can be successfully utilised. This is particularly so in replacing aortic grafts. The
superficial femoral veins (SFVs) are frequently 10–12 mm in diameter and there are now a
number of series that have demonstrated success in replacing infected aortic prostheses. As
well as aortic replacement, the SFV is also a useful conduit when replacing infected femoro-
femoral cross over grafts. The SFVs can be harvested down to popliteal level and may even
be harvested when the LSV has already gone. At the proximal end, particular care should
be taken to ensure that the profunda vein remains intact. Some leg swelling may occur after
harvest of the SFV but this is rarely severe.
Excision and extra-anatomic bypass

In many instances, excision and extra-anatomic bypass would be considered the conven-
tional approach to managing prosthetic graft infection. It has the advantage of removing
the infected prosthesis whilst providing revascularisation without placing the new graft in
the infected bed. In most instances it is necessary to use a second prosthesis for the extra
anatomic bypass and there remains a relatively high rate of subsequent infection of the new
prosthetic conduit.
Infected aortic grafts

Infection of a prosthetic aortic graft is invariably a catastrophic complication following
aortic aneurysm repair. Presentation often occurs relatively late after the original surgery
and may represent haematogenous seeding of the prosthetic graft or low-grade contami-
nation from the time of implantation. Presentation varies. In some cases, patients present
with features of chronic sepsis including general malaise, pyrexia of unknown origin and
weight loss. Back pain and abdominal pain may be present but are not invariable and the
diagnosis is frequently delayed. More acutely, patients may present with signs of gastroin-
testinal (GI) bleeding either from an aorto-duodenal fistula (with or without false aneu-
rysm formation) or more commonly an erosion of the duodenum by the body of the graft.
In this instance, the arterial anasatamosis is not involved but bleeding is from the eroded
edges of the duodenum. In all patients with upper GI bleeding and a history of previous
aortic surgery, the diagnosis of aorto-duodenal fistula should be considered until proven
otherwise.
In patients presenting less acutely, time should be spent not only in investigation, but
also in attempting to optimise the patient’s status for surgery. Any surgery for an infected
aortic graft is going to be a major undertaking and any time spent improving the patient’s
general condition will ultimately pay dividends. Positive blood cultures may allow admin-
istration of appropriate antibiotics and attention should be given to the patient’s nutritional
status, which may well be depleted by the effects of chronic sepsis.
The surgical strategy will to a large extent be determined by the general condition of the
patient and the urgency for operative intervention (principally the presence of life-threat-
ening haemorrhage). The key elements of the surgical approach are the safe excision of the
infected graft and the provision of distal revascularisation.
Any revisional aortic surgery is demanding but particularly so in the presence of infec-
tion or when dealing with a proximal anastomotice false aneurysm. The ease by which
proximal control can be established is largely determined by the length of aorta between
the prosthetic graft and the origin of the renal arteries. In order to safely secure proximal
control, some surgeons prefer to expose the aorta at the level of the diaphragm prior to dis-
secting out the infrarenal aorta. The lesser sac is entered by dividing the lesser omentum
and the left lobe of the liver is retracted upwards to expose the crus of the diaphragm. In the
emergency situation this can be split bluntly with scissors but in the elective setting can be
progressively divided with diathermy. This exposes the most distal thoracic aorta above the
coeliac axis where a clamp can be placed without closing it. Attention can then be turned
to the infrarenal aorta, safe in the knowledge that if major haemorrhage is encountered,
proximal control can be easily be achieved. An alternative means of securing rapid proximal
control is to expose a more distal portion of the aortic graft and introduce an aortic occlu-
sion balloon through a small arteriotomy in the graft.
Having achieved proximal control, distal control is generally more straightforward. If it

is not possible to dissect out the iliac vessels, the graft can be divided and the iliac arteries
controlled with Pruitt occlusion catheters or Fogarty-type catheters fitted with three-way
taps.
Having excised the graft, the decision needs to be made about subsequent reconstruction.
In the emergency setting, where the patient has been operated on for life-threatening haem-
orrhage, the decision will be dictated by the condition of the patient at this stage. If the patient
is in extremis, the aortic stump should be over sewn and the patient returned to the intensive
care unit in an attempt to stabilise them. If they recover sufficiently in the next few hours they
may be returned to the operating theatre for lower limb revascularisation although it has to
be accepted that if they survive many will face bilateral lower limb amputation.
In dealing with the infrarenal aortic stump, sufficient tissue should be debrided to allow
safe ligation of the aorta and minimise the risk of the aorta subsequently giving way and
leading to stump blow-out. The aortic stump is most securely closed in two layers with a
non absorbable monofilament – a layer of interrupted mattress sutures and a second layer
of continuous suture.
Where revascularisation is deemed appropriate, two options are available – extra-
anatomic bypass or in situ replacement. The conventional approach is extra-anatomic
bypass. Controversy exists as to whether this is by axillo-bifemoral or bi-axillo-femoral
bypass. Furthermore, some authors advocate performance of the extra-anatomic bypass
prior to graft excision to avoid the otherwise potentially prolonged period of lower limb
ischaemia that occurs and the subsequent reperfusion effects. Others argue that pre-emptive
insertion of the extra-anatomic grafts risks bacteraemic seeding at the time of excision of
the infected aortic grafts and contributes to the 25% infection rate amongst extra-anatomic
grafts inserted in the treatment of aortic graft infection. The other major concern with this
approach is the risk of aortic stump blow-out. Historic series have reported blow rates of up
to 40% although more recent practice would indicate a rate closer to 10%.
In situ graft replacement avoids the issue of an aortic stump and is thus particularly
applicable when there is only a short infrarenal aortic segment although it does necessitate
placing a graft in an infected bed. A number of series have reported good outcomes utilising
prosthetic grafts treated to resist infection. The commonest way of doing this is to soak a
polyester graft for 20 min in a rifampicin solution (60 mg ml–1). The relative rarity of aortic
graft infection means that most evidence comes from single-centre retrospective case series.
One of the largest series of in situ prosthetic replacement grafts in 52 patients included a
mix of rifampicin-bonded grafts and grafts covered with omental patches. In this series the
perioperative mortality was 9% with 11% of patients developing recurrent infection of the
aortic graft [3].
More recently, attention has turned to using autologous material for in situ replacement.
Obviously, the LSV is of too narrow a calibre to be practicable but a number of series have
now reported good outcomes using SFV [4]. The vein can readily be harvested from the
subsartorial canal. Proximally it is divided immediately distal to its confluence with the pro-
funda vein where the common femoral vein is formed. Distally, it is easiest to divide it just
above the adductor hiatus – below this level there is an increasing number of geniculate
branches to deal with, although if it is essential to obtain the maximum length then it is quite
safe to harvest the entire above-knee popliteal vein in continuity. Postoperative leg swelling
is reported to be minimal, even in cases where the LSV has been previously removed. In an
adult, the femoral vein is generally 10–12 mm in diameter and can readily be fashioned to
Figure 20.2 In situ

reconstruction of a mycotic
aortic aneurysm with superficial
femoral vein.
allow anastomosis with the aorta. If necessary, both femoral veins can be harvested to allow
construction of a Y-graft and it can also prove useful in the replacement of an infected aortic
graft (Figure 20.2).
Another alternative to the autologous vein is fresh cryo-preserved human tissue.
This is available from a number of sites around the UK including Oxford, Liverpool and
Birmingham. The largest experience is with ascending aorta and aortic arch, which has been
extensively used in paediatric cardiac surgery but material is now available from more distal
aorta and superficial femoral artery. There is, as yet, little experience in the UK of using such
grafts for the replacement of infected peripheral vascular grafts but reports from elsewhere
suggest results comparable with autologous material [5].
Infected aortic endovascular grafts

Aortic stent grafts are not immune to infective complications [6]. A number of series have
reported incidences of around 1%, similar to that seen following open aortic surgery. The
mechanisms of presentation are also similar to those seen in infected open aortic grafts and
include aorto-enteric fistula. Furthermore, the same principles of treatment apply – removal
of the infected prosthesis, debridement of infected tissue and, where possible, distal revas-
cularisation. In most cases removal of the infected stent graft is relatively straightforward
as they do not become incorporated in the same manner as an open graft. Care must be
exercised however when removing grafts with suprarenal fixation. It is easy to damage the
suprarenal aorta whilst attempting to remove these devices and in some cases it is more judi-
cious to detach and leave in situ the uncovered bare metal portion of the graft, which extends
above the renal arteries.
Infected carotid patches

As with peripheral infection, infection related to a prosthetic carotid patch is generally a
straightforward diagnosis. Presentation may be in the form of clear signs of acute infection
with pain, swelling and erythema overlying the patch. In some cases the presentation is less
acute and manifests as a discharging sinus or the presence of a false aneurysm. Rupture of a
prosthetic patch is fortunately rare, as are recurrent neurological symptoms.
In most cases, first-line imaging is with Duplex scanning, to look for local features of
infection and to confirm whether the internal carotid artery remains patent – an occluded
internal carotid is obviously considerably easier to manage. Magnetic resonance or CT angi-

ography will give details about the distal carotid circulation. It can be very helpful to have
an indication of the likely consequences of occlusion of the internal carotid artery as this
is a very real risk following revisional surgery for infection. Information gathered at the
time of the original operation can be helpful, e.g. if performed under local anaesthetic, did
the patient tolerate the procedure without the need for a shunt? If this information is not
available, then angiography with a trial balloon occlusion of the internal carotid artery can
provide useful information.
As with infection of an arterial prosthesis elsewhere, the treatment strategy has to be
considered in light of the patient’s status, but successful eradication of infection generally
requires removal of the infected patch. The operation should only be undertaken by sur-
geons with considerable experience of carotid surgery. The patient should be warned of the
relatively high risk of cranial nerve injury. The most difficult part of the procedure is usually
obtaining adequate distal exposure to safely allow reconstruction. In patients with a low
bifurcation and a short patch this is not too difficult but if difficulty is anticipated, naso-tra-
cheal intubation allows the mandible to fully close and provides more room to manoeuvre
than conventional intubation. When further access is required, some surgeons advocate
approaching the distal internal carotid via an infratemporal fossa approach, usually in con-
junction with an ear, nose and throat (ENT) surgeon.
Having established control proximally and distally, and following heparinisation of the
patient, the patch can be opened and a suitable shunt inserted. In most cases we plan to
attempt reconstruction with autologous vein and this can be reversed and ‘pre-loaded’ onto
the distal limb of the shunt prior to insertion. In some circumstances and even with good
distal exposure, the accessible part of the internal carotid artery (ICA) proves too friable to
safely reconstruct. It is then necessary to ligate the artery and it is when faced with this situ-
ation that knowledge of the patients likely tolerance of internal carotid ligation can aid the
decision-making process as up to 50% of patients will suffer a stroke following acute ligation
of the ICA.
In those patients where reconstruction is feasible, the choice between utilising the vein
as a patch or interposition graft is largely determined by the amount of ICA wall left follow-
ing adequate debridement of the edges of the original anastomosis. In a large proportion of
cases, a patch is feasible and this is the simpler option. The vein that was loaded onto the
shunt can be opened longitudinally and following excision of any valve leaflets can be used
as a patch in the conventional manner.
If only a thin strip of back wall of the ICA remains, it is better to use the vein as an
interposition graft. The distal ICA is transacted and the distal anastomosis fashioned first.
Following completion of this anastomosis, the distal limb of the shunt is withdrawn into
the vein graft and the anastomosis tested for adequate haemostasis. It is important to ensure
this is satisfactory prior to undertaking the proximal anastomosis as access to the deep
aspect of the graft is subsequently very difficult. The proximal anastomosis can then be
completed, ideally incorporating the origin of the external carotid artery, although this is
not always practicable, in which case the orifice of the external carotid can be over sewn.
Infected peripheral grafts

The diagnosis of peripheral graft infection is generally more straightforward than is the
case with intracavity grafts. Patients generally have signs of localised sepsis including pain,
swelling and erythema over the graft/anastomosis. There may be a purulent discharge from
a wound sinus and any blood staining of the sinus should alert the surgeon to the possibility
of incipient major haemorrhage. Anastomotic false aneurysm development is not always
related to infection but even in the absence of other signs of sepsis, the possibility should
always be borne in mind.
In most cases of peripheral graft infection, the underlying graft is prosthetic and success-
ful treatment ultimately requires removal of the infected prosthesis. The anastomosis may
be quite friable or there may have been false aneurysm development. It is therefore generally
wise to ensure adequate proximal control prior to exposure of the anastomosis. With infra-
inguinal grafts in the groin, it is often possible to expose the inguinal ligament, the lower
margin of which can then be carefully dissected upwards to provide exposure of the distal
external iliac artery. If there is any doubt about the ability to safely gain control by this route
(e.g. with extensive scarring, in very obese patients or with haemorrhage requiring rapid
control) then an extraperitoneal approach to the external iliac artery should be followed.
Having achieved proximal and distal control the graft can then usually be detached from
its anastomoses and removed. In some cases, the graft infection has resulted in the graft
failing to become incorporated and removal from its track is relatively straightforward. It is
often the case, however, that the graft has remained incorporated along a large proportion of
its length. One useful technique is to utilise a Codman-style varicose vein stripper to remove
the graft, however one has to weigh up the risk of ongoing infection in a graft remnant
against that of damaging adjacent structures, e.g. the femoral vein in attempting to remove
a well incorporated graft.
Revascularisation
In any case of arterial graft infection, amongst the most feared complications is that of
recurrent infection with the possibility of life-threatening haemorrhage. Careful considera-
tion therefore needs to be given as to whether the risks inherent in further attempts at arte-
rial reconstruction are justified. In a proportion of cases, the limb in question will survive
without further revascularisation. Experience of patients with infected false femoral aneu-
rysms as a complication of drug misuse has shown that limb salvage can be maintained
even following acute ligation of the common femoral artery. Factors that may indicate the
possibility of achieving limb salvage without distal grafting include an already occluded
graft, patients whose grafts were originally inserted for claudication and the ability to pre-
serve flow down the profunda artery. Having taken down an infected common femoral
anastomosis, it is necessary to excise a reasonable portion of the adjacent arterial wall. The
atherosclerotic artery can harbour organisms and failure to adequately debride the artery
wall can result in late haemorrhage, even when closed primarily. If necessary, the artery
can be closed with a patch although this should ideally be autologous material. In many
cases it is possible to find a short segment of vein to do this or alternatively, if the super-
ficial femoral artery is occluded (which frequently it is) a segment can be excised, opened
longitudinally and fashioned into a patch by removal of the occluded intima. Failing this,
bovine pericardium may be used if no autologous material is available. Repair with further
prosthetic material is a very high risk strategy (even if soaked in antibiotics or silver nitrate)
and should be avoided.
It can be difficult to achieve sound wound closure of the groin, particularly when there
has been extensive scarring. Detachment of Sartorius from the anterior superior iliac spine
and mobilisation of its lateral border (to avoid devascularising it) allows the muscle to be
‘rolled over’ and tacked down to cover the common femoral artery.
Should distal revascularisation be deemed necessary then a number of options are avail-
able. Our own practice has changed over the years such that in nearly all cases prosthetic
grafts are only utilised where no autologous material is available. There are, however, still
some patients with an infected prosthetic infra-inguinal graft in whom the long saphenous
vein is still present and suitable to use as a conduit. This includes a number of patients in
whom the original operation record makes note of an ‘unsuitable’ vein! It is always worth
re-examining the vein with Duplex to assess its suitability. Obviously, the search for a suitable
vein should not stop at the ipsilateral long saphenous. The contralateral leg, short saphenous
and upper limb veins should all be considered if necessary.
Interventional radiology techniques have advanced considerably over the years.
Subintimal angioplasty, atherectomy catheters and remote endarterectomy have all increased
the scope of disease that can be treated but by wholly or partial endovascular means and
these can all be considered in cases where re-do bypass grafting with autologous vein is not
feasible.
In cases where sepsis is confined to the groin and with no other options available, consid-
eration may be given to extra-anatomic bypass. Iliopopliteal bypass via the obturator canal
and axillo-popliteal bypass have both been utilised in these circumstances and it is worth
remembering that both the above- and below-knee popliteal arteries can be approached by
a lateral approach.
Many of the same principles apply to other peripheral grafts as apply to infra-inguinal
grafts. Extra-anatomic grafts in particular have a relatively high incidence of graft infec-
tion. Although it is generally considered most appropriate to remove an infected graft,
there have been some reports of successfully managing grafts in situ. This should ideally
only be entertained when the infecting organism is of low virulence and the infection is
confined to the mid portion of the graft, not involving the anastomoses. Surgical manage-
ment includes adequate debridement of infected tissues, local application of antibacterial
agents and adequate coverage of the graft, which may necessitate a vascularised myocuta-
neous flap.
Infected femoral pseudoaneurysms

Infected primary false aneurysms of the common femoral artery are perhaps worth a spe-
cial mention. Invariably a consequence of intravenous drug abuse, they typically present
as an apparent acute abscess in the groin. This can be a real trap for the unwary and in all
such cases an underlying false aneurysm should be positively excluded prior to incision and
drainage. There is often a mixed growth of organisms although staphylococci and Gram
negative species are most common.
Direct exposure of the artery is potentially hazardous and made all the more difficult by
the indurated and thickened tissues. The judicious approach is to establish proximal control
via an extraperitoneal approach to the external iliac artery. On exposure there is often found
to be a large defect in the anterior wall of the common femoral artery, which precludes pri-
mary repair. The very real risk of ongoing sepsis and recurrent haemorrhage means that in
most cases the only prudent course of action is to ligate the common femoral artery without
reconstruction. Contrary to expectations, the majority of patients will tolerate this with the
affected limb surviving on a collateral circulation. In the unusual event of limb-threatening
Figure 20.3 Computed tomography showing

a saccular aneurysm of the distal thoracic aorta,
strongly suggestive of a mycotic aetiology. The
aneurysm was excised and repaired with an in situ
cryo-preserved aortic arch homograft.
critical ischaemia developing, consideration may be given to an extra-anatomic ilio-pop-

liteal bypass via the obturator canal although the safe option is a primary amputation.
Mycotic aneurysms
The term mycotic aneurysm is generally applied to any aneurysm with an infective aetiol-
ogy. Historically, the commonest cause was septic embolisation in patients with infective
endocarditis. Now a relatively rare cause, other aetiological mechanisms have come to
the fore. There are a number of pathogenetic mechanisms including septic embolisa-
tion (into a small branch vessel or into the large vasa vasorum in the aorta), seeding of
an established atherosclerotic plaque, secondary infection in an established aneurysm
and rarely contiguous spread from local septic focus. Ultimately the infective process
results in progressive destruction of the arterial wall with eventual aneurysm formation.
Once established, the natural history of mycotic aneurysms is to progressively enlarge
and eventually rupture.
As with graft infection, an appropriate index of suspicion is essential in making the diag-
nosis. In many patients, the ‘vascular’ presentation may occur some time after the initial
septic episode or indeed a particular septic episode may not be apparent. In these cases the
clue to the diagnosis often lies in the morphology of the aneurysm, which is frequently (but
not invariably) saccular rather than fusiform (Figure 20.3).
The commonest causative organism seen in modern vascular practice is salmonella spe-
cies although staphylococci, streptococci, haemophilus and pseudomonas are not infre-
quently cultured. Management is along similar lines to those for an infected prosthetic graft
in the same anatomical location. One possible exception is the use of endovascular stents to
treat mycotic aneurysms. This is a potentially attractive option in what is often a debilitated
patient; published results demonstrate acceptable early outcome results [7]. There is, how-
ever, obvious concern about the development of late graft infection and as yet there is little
long-term follow-up data.
References as a conduit for vascular reconstruction in

infected fields. Eur J Vasc Endovasc Surg
1. Health Protection Agency. Surveillance of 2003; 25: 424–31.
Surgical Site Infection in England: October 5. Brown KE, Heyer K, Rodriguez H et al.
1997 – September 2005. London: Health Arterial reconstruction with cryopreserved
Protection Agency, 2006. human allografts in the setting of
2. Szilagyi DE, Smith RF, Elliot JP et al. infection: a single-center experience with
Infection in arterial reconstruction midterm follow-up. J Vasc Surg 2009;
with synthetic grafts. Ann Surg 1972; 49: 660–6.
176: 321–32. 6. Sharif MA, Lee B, Lau LL et al. Prosthetic
3. Oderich GS, Bower TC, Cherry KJ et al. stent graft infection after endovascular
Evolution from axillofemoral to in situ abdominal aortic aneurysm repair. J Vasc
prosthetic reconstruction for the treatment Surg 2007; 46: 442–8.
of aortic graft infections at a single center. 7. Clough RE, Black SA, Lyons OT et al.
J Vasc Surg 2006; 43: 1166–74. Is endovascular repair of mycotic aortic
4. Gibbons CP, Ferguson CJ, Figelstone LJ aneurysms a durable treatment option? Eur J
et al. Experience with femoro-popliteal vein Vasc Endovasc Surg 2009; 37: 407–12.
Chapter
21
Vascular malformations
George Hamilton and Andrew Platts
Key points
• I n the neonate differentiate between haemangioma and vascular malformation
• Clinical assessment is of prime importance
– Is intervention or simple reassurance required?
– Realistically manage patient and parent expectations
• Investigate by ultrasound, Duplex and magnetic resonance imaging (short T1 inversion
recovery [STIR], T1 and T2 with selective use of magnetic resonance angiography and
magnetic resonance venography)
• Invasive angiography rarely indicated except for complex arteriovenous (AV)
malformations and planned embolisation
• Classify the malformation – extratruncular or truncular?
• The majority of congenital vascular malformations are venous – investigate for
disseminated intravascular coagulation (DIC) in diffuse extensive capillary/venous
malformations in the young child
• Malignant vascular malformations are rare
• Tailor treatment to the severity of the lesion – cosmesis, pain, function, asymmetry,
tissue damage etc.
• Foam sclerotherapy will be suitable and repeatable for the majority of venous
malformations
• Klippel–Trenaunay and Parkes-Weber syndromes will need multidisciplinary input –
vascular and orthopaedic surgeons, paediatrician and dermatologist
• Look for and ablate ectatic and aneurysmal lateral marginal veins because of risk of
thromboembolic disease
• Ethanol sclerotherapy and surgery for the minority usually extratruncular AV
malformations
• The best care for patients of all age groups with congenital vascular malformations
is provided by multidisciplinary groups (including plastic surgery) with a special
interest
Introduction
Vascular malformations, currently most commonly termed congenital vascular malforma-
tions (CVMs), have a wide spectrum of clinical presentation and behaviour, ranging from
Chapter 21: Vascular malformations 243
simple birthmarks to massive disfiguring lesions, which in a minority of patients can be

life threatening. Understanding the pathophysiology and anatomy of CVMs has been poor,
resulting in clinical confusion with regard to treatment strategies and resulting poor out-
comes. Since the 1990s, there has been a significant advance in understanding of the variety
of CVMs, the introduction of a clinically relevant and useful classification system and sig-
nificant improvement in their management.
Definition/classification of congenital vascular malformations

Neonatal/infantile haemangioma
Infantile and neonatal haemangioma usually appears after birth and rather than being a
malformation is a vascular tumour arising from endothelial cells. Infantile haemangioma
is characterised by early proliferation and growth can be rapid followed by spontaneous
involution between 5 and 10 years. Haemangioma is relatively common with an incidence
of 2–3% in newborn infants. The incidence of congenital vascular malformation is about 1%
and because of its presentation also during the neonatal period, it is important to distinguish
clinically between CVM and haemangioma. Congenital vascular malformation is character-
ised by malformed blood vessels while haemangioma is a localised vascular tumour behav-
ing as described above.
Congenital vascular malformations

Congenital vascular malformations can be classified according to their make up as either
arterial, venous, arteriovenous, lymphatic, capillary, and combined or mixed vascular mal-
formation. Venous malformations are the most common form of CVM with 15–20% being
mixed lesions, examples of which include the Klippel–Trenaunay syndrome and Parkes-
Weber syndrome [1].
Eponymous syndromes
Early in the last century, before angiography was available, several clinical syndromes were
described and named according to the clinician. Two eponymous syndromes, namely
Klippel–Trenaunay and Parkes-Weber, continue to be used in modern times. This is largely
because the syndromes describe the primary vascular malformation together with the sec-
ondary non-vascular pathologies.
Klippel–Trenaunay syndrome has venous, lymphatic and capillary components associ-
ated with soft tissue swelling, long bone growth, leg discrepancy and gigantism [2]. Parkes-
Weber syndrome has capillary, venous and lymphatic malformation but is also characterised
by arteriovenous shunting malformation (Table 21.1) [3].
The development of imaging modalities has led to a better understanding and classifica-
tion of congenital vascular malformations.
Hamburg classification
This classification was based on a consensus established at the International Workshop in
Hamburg in 1998 with subsequent modifications. The classification system is based on the
insights achieved in determining the underlying pathophysiology, anatomical distribution
both macroscopic and microscopic, the resulting underlying haemodynamic effects and
Table 21.1 Eponymous malformations
Klippel–Trenaunay syndrome
• Predominantly haemolymphatic (HLM)
• Venous, lymphatic and capillary malformation
• Vascular bone syndrome
Parkes-Weber syndrome
• Capillary, venous and lymphatic
• Arteriovenous shunting malformation (AVM)
Table 21.2 The Hamburg classification of congenital vascular malformations
A: Anatomical types
• Predominantly arterial defects
• Predominantly venous defects
• Predominantly arteriovenous (AV) shunting defects
• Predominantly lymphatic defects
• Combined vascular defects (Klippel–Trenaunay and Parkes-Weber syndromes)
B: Embryological types
• Extratruncular forms
Infiltrating, diffuse
Limited, localized
• Truncular forms
Aplasia or obstruction
Hypoplasia, aplasia, hyperplasia
Stenosis, membrane, congenital spur
• Dilation
Localized (aneurysm)
Diffuse (ectasia)
further differentiation relating to their time of development in the embryo. The classification
is summarized in Table 21.2 with the first component (A) being classification into one of five
types based on the vascular pathophysiology, namely arterial, venous, lymphatic, arterio-
venous and combined vascular malformations.
The second very important component of the classification (B) is based on the embryo-
logical stage at which the malformation develops and divides malformations into either
extra-truncular or truncular.
This part of the classification is particularly important in terms of management and
prognosis in that the clinical behaviour of the two forms is predicated at the embryological
stage at which the malformation develops and at which development arrest occurs.
Extra-truncular congenital vascular malformations

Extra-truncular cardiovascular malformations arise during the earlier phase of embryo-
logical development. At this stage, vascular cells are primarily angioblasts derived from
mesenchymal cells. These are similar to stem cells, which at any stage of life can be stim-
ulated to proliferate. These stimuli can be hormonal, such as menarche, adolescence and
pregnancy, or as a response to trauma or surgical intervention. Extra-truncular malforma-
tions most commonly are diffuse and infiltrating in nature, less commonly more localised
but typically having compressive effects on adjacent tissues and organs. A further clinically
important feature is their high propensity for recurrence if treatment is sub-optimal. In gen-
eral extra-truncular malformations have greater morbidity, worse prognosis and present a
greater therapeutic challenge.
Truncular congenital vascular malformations
These malformations develop later in the embryological vascular development. They are
therefore constituted of cells that have lost their primitive stem cell-like properties, in par-
ticular the potential to proliferate. These malformations may present as persistent foetal
remnants, such as the sciatic vein, or a defectively developed structure, often occlusive, with
a combination of stenotic, aneurysmal or web-like disease. The classification therefore fur-
ther subdivides truncular lesions in to obstructive or dilated. Truncular lesions are much
less likely to recur when treated.
Mulliken classification of cardiovascular malformations

This is a different system that characterises malformations as either slow flow or fast flow.
This further haemodynamic classification is clinically useful but the Hamburg classification
remains predominant, particularly with its ability to predict the clinical outcome for treat-
ment according to the malformation being either extra-truncular or truncular.
Most commonly presentation takes place in the peri-natal period. A congenital vascular
malformation having developed in the later stages of the foetus is an abnormal vascular
lesion present at birth. A haemangioma is most commonly not present at birth but develops
suddenly within the first 4 weeks of the neonatal period with rapid early growth, stimulation
and then regression between the ages of 5 and 10 years. The first step in the differential diag-
nosis is therefore to exclude the diagnosis of haemangioma (Table 21.3).
Classification of a congenital vascular malformation is the next step in the process of
differential diagnosis and as described above is made by applying the Hamburg classifica-
tion. The presence of an AV component predicts a more virulent and unpredictable clinical
course with higher morbidity. It is therefore also extremely important to investigate every
vascular malformation for the possibility of an arteriovenous component. Initially this may
not be clinically obvious but declares itself during follow up.
Investigation
After the initial steps of clinical assessment by history taking and examination, the first step
in the diagnostic process is by non-invasive assessment, primarily in the vascular laboratory.
Duplex ultrasonography based on colour Doppler imaging and spectral waveform analysis
primarily provides anatomical and haemodynamic information. This methodology can reli-
ably demonstrate arterial and venous components and define the anatomy and extent of the
lesion, in particular the presence of feeding arteries and draining veins. When there is deeper
involvement of anatomical structures, such as in the chest, abdomen or pelvis, or involvement
of long bones, Duplex sonography may be limited in its diagnostic potential (Table 21.4).
Table 21.3 Differential diagnosis
• Haemangioma occurs in 2–3% of neonates, rising to 10% at the end of the first year
– Vascular tumour
• Congenital vascular malformation occurs in 1%
– Congenital stable malformation that grows commensurately
• Venous malformations are the most common congenital vascular malformations (CVMs)
– 15–20% are mixed lesions (Klippel–Trenaunay and Parkes-Weber syndromes)
Table 21.4 Investigation
• Ultrasound
– Excellent screening tool, central to correct diagnosis
• Flow characteristics of the lesion can be defined
• Arterial, venous or combined
– High flow or low flow
– Confirms complications, such as thromboses
• Computed tomography (CT)
• Excellent anatomical detail with contrast for arteriovenous (AV) malformations
• Magnetic resonance imaging (MRI)
– Excellent anatomical detail of surrounding tissues
– Confirms venous nature of the lesion
– Excellent for picking up complications or occult features
Magnetic resonance imaging (MRI) is the second component of the initial diagnostic
process for all malformations. STIR, T1 and T2 weighted studies will provide excellent ana-
tomical detail of the extent of the cardiovascular malformation, differentiate between high
and low flow lesions and is particularly valuable in the diagnosis of venous malformations.
In addition, MR angiography and venography can be added to the above MRI investigation
for further anatomical detail. Computed tomography contrast scans are also valuable for
assessment, in particular of malformations involving bone, the thorax, abdomen and pelvis.
Three-dimensional reconstruction of CT angiograms is particularly useful in the diagnosis
of arteriovenous malformations (Figure 21.1).
Lympho-scintigraphy is the investigation of choice for assessment of lymphatic or
haemo-lymphatic malformations and MR lymph-angiography may also be used to provide
further anatomical detail of the extent of the lymphatic malformation.
Simple bone X-ray is indicated where there is limb asymmetry in the presence of vascu-
lar-bone syndrome, and MRI can confirm the presence of bony or articular involvement.
In cases of deep seated haemangioma or congenital vascular malformations, MR and CT
imaging are important. This will be needed to differentiate between various hamartomatous
pathologies and most importantly to exclude the possibility of sarcoma. Where there is still
doubt, CT-guided needle biopsy is of value in these cases.
Angiography and venography are important, more invasive investigations currently used
in planning interventional or surgical treatment of symptomatic vascular malformations.
Figure 21.1 Magnetic resonance imaging (MRI)

revealing the extent of a venous malformation with
diffuse involvement of the thigh muscles and the
right labium majorum.
Basic haematological investigations are of course important but should include coagula-
tion studies to detect the possible presence of a consumptive coagulopathy, which could com-
plicate extensive extra-truncular lesions or large marginal veins. Large marginal veins and
venous aneurysms have a high risk of venous thrombosis. Coagulopathy is found in children
with extensive haemangiomas and in these children anticoagulation might be required.
Diagnosis summary
Diagnosis in the vast majority of malformations is based on careful clinical assessment
and examination, Duplex sonography with STIR, T1 and T2 weighted MRI. In very young
children more complex investigations, which require sedation or general anaesthesia, can
be delayed where the malformation is clinically assessed as unlikely to be aggressive [4].
Earlier management principles focused on aggressive surgical excision of these lesions. The
poor outcomes in the absence of the clinical insight given by the Hamburg classification
resulted in a move to more conservative treatment. Currently, treatment outcomes are much
improved and are focused on a multidisciplinary approach. This will involve disciplines such
as paediatric dermatology, vascular surgery, plastic and reconstructive surgery, orthopaedic
surgery, head and neck surgery, ENT surgery. The outcome of multidisciplinary assessment
will be focused on dealing with the primary malformation and subsequently on dealing with
the secondary affects, particularly on the vascular and musculoskeletal systems.
Indications for treatment

Many malformations presenting in children will not need initial treatment but rather careful
monitoring of their development and or regression. This will require careful documentation
of clinical history, photographic and non-invasive assessment.
Table 21.5 Indications for intervention
Prevention of harm
– Heart failure
– Consumptive coagulopathy
– Ischaemia of limb, skin, etc.
– Compression of airway, etc.
– Bleeding
Treatment of symptoms
– Pain
– Unpleasant pulsation – tinnitus or bruit
Treatment of signs
– Unsightly vessels or swelling
– Limb or facial asymmetry
– Cosmesis
Absolute indications for intervention are in most cases clear and include bleeding from
arteriovenous or venous malformations, progression of high output heart failure in arterio-
venous malformations, persistent lymphatic leakage with infection or sepsis from lymphatic
or haemo-lymphatic malformations, and chronic venous insufficiency secondary to ven-
ous or haemo-lymphatic malformations. Vascular malformations situated in potentially
life-threatening regions such as next to the airway or significantly compromising limbs,
or vital functions such as seeing, hearing, eating, are also indications for intervention.
Relative indications for treatment include severe unremitting pain, non-healing ulceration,
lesions causing deep vein thrombosis, such as in the marginal vein with its associated risk
of thromboembolic disease, haemathrosis where the lesions involve the knee joint, and cos-
metically unacceptable lesions.
In venous malformations, the most common presentation, early aggressive treatment is
only indicated in situations as highlighted above but also where the vascular bone syndrome
results in limb gigantism. The treatment plan for diffuse extra-truncular lesions has to be
very carefully formulated because of its progressive nature and high potential for recurrence
(Table 21.5).
In treatment of all malformations the management of expectation of the patient and
parents is key. It must be made clear that multiple interventions may be required to control
and deal with the effects of the malformation and it is also important to stress that control
rather than cure is the likeliest outcome in the majority of patients [5].
Surgical treatment
The old principle of surgical ligation of a feeding or draining vessel rarely applies and
indeed ligation of a feeding artery is absolutely contraindicated since this will deny the pos-
sibility of endovascular treatment to future recurrences. Surgical treatment is frequently
combined with embolization in AV malformations, or may be required after sclerother-
apy in predominately venous malformations. The policy of preoperative and postoperative
Figure 21.2 The aneurysmal

marginal vein in the lateral
aspect of the right leg of this
young woman was removed
with preoperative marking
and intraoperative Duplex
localisation of the deep draining
veins for ligation. These venous
ectasias have significant
potential for thromboembolic
complications [6].
embolo-sclerotherapy allows treatment of complex and infiltrating congenital vascular

malformations that previously were beyond surgical therapy (Figure 21.2).
In general, the truncular forms will be best treated by surgery, particularly venous mal-
formations such as the marginal vein in the Klippel–Trenaunay syndrome and in venous
aneurysms [7]. Surgery to vascular malformations can be either haemodynamic or ablative.
Haemodynamic surgery is designed as the first step in dealing with the vascular malformation
where restoration of arterial blood flow around an infiltrating lesion, e.g. a femoro-popliteal
bypass graft, or a venous bypass to restore venous drainage, is required. After haemodynamic
surgery has been successfully achieved excisional or ablative surgery can be performed.
Complete ablative resection of CVMs has greater success when the malformation is super-
ficial to the deep fascia of any muscle and not invading into major muscles, bone or into the
thorax, abdomen or pelvis [8]. Secondary surgery to limb length discrepancy (epiphysiod-
esis) can then take place, usually between the ages of 3–7 years, by the orthopaedic service.
The more extensive excisional surgery will in many cases involve the reconstructive plas-
tic surgeons. Where there is gross deformity and marked compromise of function in a limb,
then primary amputation may be the best option, resulting in early optimisation of function
(Figure 21.3).
Endovascular therapy
Embolisation using coils or liquid embolic agents such as glues and Onyx™ are useful in
dealing with the arterial component of AV malformations. In extensive venous malforma-
tions coils and glues are not indicated because of the diffuse nature of the lesion and the
Figure 21.3 This extensive and progressive

extratruncular malformation flared up after trauma;
primary amputation of the index finger resulted in an
excellent symptomatic and functional outcome.
high propensity for these structures to be flushed out of the malformation. Although glue
is effective, it leaves a hard cast within the obliterated malformation that may be uncom-
fortable or unsightly. Coil and glue embolization will rarely be effective in the more diffuse
extra-truncular lesions. Sclerotherapy by injection of various agents is of value in many mal-
formations, both arteriovenous, venous, truncular and extra-truncular. These sclerosants
act by their toxic effects resulting in endothelial destruction, thrombosis and fibrosis. The
important factors to be considered are the strength of the sclerosant, the length of time that
the sclerosant is in contact with the vein wall and the area of contact between sclerosant and
endothelium. A mainstay of treatment with sclerosants is post-injection compression. There
are several sclerosants in use [9].
Ethanol
Absolute ethanol has been used for many decades. It is a very potent sclerosant so much so
that it can cause vessel necrosis and have local toxic effects. These include severe pain, swell-
ing with compartment syndrome, damage to surrounding structures, in particular nerves
and in more superficial lesions overlying skin necrosis. In addition ethanol has potent sys-
temic side effects including cardiac arrhythmias and pulmonary hypertension secondary to
pulmonary vasospasm. The reported incidence of complications including cardiovascular

collapse and death has been reported to be as high as 28%.
Ethanol is therefore a very powerful sclerosant but with potentially devastating side
effects and should only be used in more aggressive generally extra-truncular malformations
and after multidisciplinary assessment of the therapeutic options. Ethanol injection is par-
ticularly useful in dealing with diffuse extra-truncular lesions causing significant compli-
cations. The core or nidus of the lesion must be identified and targeted with the ethanol
injection to achieve a good result.
The ethanol is injected either directly under careful fluoroscopic control or by a catheter.
General anaesthesia is required and careful attention to postoperative monitoring and anal-
gesia is essential. The maximum dose of ethanol should not exceed 1 ml kg–1 of body weight.
Other sclerosing agents

Polidocanol, tetradecyl sulphate, sodium morrhuate and ethanolamine have been used in
the treatment of cardiovascular malformations. They are particularly valuable in dealing
with venous malformations and treatments can be repeated many times, in many cases with-
out general anaesthesia.
Technique
In arteriovenous malformations catheter delivery is most commonly used. In venous mal-
formations direct-injection sclerotherapy is used. Duplex ultrasonography is very useful in
imaging the malformation and very frequently multiple needles are used. Under fluoros-
copy, contrast is injected to define the extent of the malformation and also to assess its vol-
ume. The volume identified of sclerosant is injected – it is important to be able to monitor
the spread of the sclerosant into the malformation. This is not a problem with the foam
sclerosant but when a ‘neat’ sclerosant solution is being used, this should be mixed with
contrast. The aim is to compress the draining vein and then to inject additional sclerosant to
fill as much of the malformation as possible. With the use of multiple needles injections con-
tinue until the sclerosant begins to drain from the other needles (Figure 21.4). Compression
Figure 21.4 Using three separate needles the components of this venous malformation of the elbow were
sequentially filled with foam sclerosant to treat the entire malformation; immediate compression was applied.
is then applied. Using foam, intravenous sedation and pain relief may be all that is required.
With the injection of ‘neat’ sclerosants and certainly ethanol, general anaesthesia will be
required. More recently, with major venous malformations, a transvenous placement of an
occlusion balloon catheter into a large draining vein has been used, alternatively in more
superficial lesions tourniquets are useful for occluding draining veins. These manouevres
aid in maximal filling of the malformation with the sclerosant.
The results of injection sclerotherapy have been good with the modern focus being on
the use of foam sclerosant (this is prepared by mixing ‘neat’ sclerosant with either air or car-
bon dioxide at a ratio of 1 : 10). There are several outcome reports with success particularly
in reduction of the pain associated with venous malformations. The complication rate is low,
mainly with transient skin pigmentation and a very low incidence of skin necrosis. Multiple
episodes of injection sclerotherapy are required in the majority of patients with venous mal-
formations [10].
Summary
Over the last decade there have been significant advances and improvements in the manage-
ment of the difficult condition of congenital venous malformations. Central to this success is
the clinical classification of these disorders, in particular into truncular and extra-truncular
lesions. The multidisciplinary approach is of absolute importance with the vascular/endovas-
cular surgeon and interventional radiologist central to the team. Embolo-sclerotherapy has
proved to be a major advance in the management of these conditions, particularly venous
malformations. It has also revolutionised the approach to high flow and extensive malforma-
tions, facilitating surgical excision of life-threatening lesions or those associated with severe
complications. Advances in devices and agents such as coils, glues and sclerosant foam have
also reduced the morbidity of treatment and widened the indications. Ethanol remains the
most effective sclerosant but is toxic and has significant side effects. The current drive is
towards finding alternatives to ethanol.
Most patients referred to a multidiscipline unit will not require treatment but benefit
from the reassurance that their disease is unlikely to progress, that it is not a tumour and
that if symptoms do develop they have established contact with a unit that is able to deliver
all aspects of the care that they may need.
References 4. Lee BB, Laredo J, Lee SJ, Huh SH, Joe

JH, Neville R. Congenital vascular
1. Lee BB, Laredo J, Lee TS, Huh S, Neville malformations: general diagnostic
R. Terminology and classification of principles. Phlebology 2007; 22: 253–7.
congenital vascular malformations. 5. Lee BB, Laredo J, Kim YW, Neville R.
Phlebology 2007; 22: 249–52. Congenital vascular malformations: general
2. Gloviczki P, Driscoll DJ. Klippel– treatment principles. Phlebology 2007;
Trenaunay syndrome: current 22: 258–63.
management. Phlebology 2007; 22: 291–8. 6. Mattassi R, Vaghi M. Management of the
3. Comi AM. Update on Sturge-Weber marginal vein: current issues. Phlebology
syndrome: diagnosis, treatment, 2007; 22: 283–6.
quantitative measures, and controversies. 7. Loose DA. Surgical management of venous
Lymphat Res Biol 2007; 5: 257–64. malformations. Phlebology 2007; 22: 276–82.
8. Maftei N, Howard A, Brown LC, 9. Rosenblatt M. Endovascular management

Gunning MP, Standfield NJ. The surgical of venous malformations. Phlebology 2007;
management of 73 vascular malformations 22: 264–75.
and preoperative predictive factors of major 10. Bergan J, Cheng V. Foam sclerotherapy of
haemorrhage – a single centre experience. venous malformations. Phlebology 2007;
Eur J Vasc Endovasc Surg 2009; 38: 488–97. 22: 299–302.
Chapter
22
Vasospastic disorders and
vasculitis
Mohammed Sharif, Jonathan Smout and Gerard Stansby
Key points
• U nderstanding of the nomenclature used for the classification of vasospastic disorders
into primary and secondary Raynaud’s phenomena is essential
• The management of these disorders requires a multidisciplinary team approach
involving physicians, rheumatologists and vascular specialists
• Treatment of Raynaud’s includes general supportive measures, pharmacotherapy and
correction of underlying disorders
• Vasculitis is associated with a range of medical conditions and can present as digital
ischaemia
• A diagnosis of vasculitis is suggested by constitutional symptoms and confirmed by
raised inflammatory markers, autoantibodies and biopsy of the skin lesions
• Immunosuppressive therapy is the mainstay of treatment in vasculitic disorders
Introduction
Raynaud’s phenomenon refers to a clinical state characterised by episodic vasospasm, usually
involving the distal small arteries of the upper limb although sometimes toes and feet are also
affected. In addition, there are other vascular disorders characterized by inflammatory changes
in the arterial wall, known as ‘vasculitidies’, which can present with digital ischaemia.
Vasospasm (Raynaud’s phenomenon)

Maurice Raynaud first described this clinical picture in 1862. The classical presentation of
Raynaud’s phenomenon is characterised by a sequence of colour changes in the following
order:
• pallor, reflecting initial vasospasm;
• cyanosis as a result of deoxygenation of stagnant blood during maximum vasospasm;
• rubor, representing inflow of oxygenated blood and reactive hyperaemia as the
vasospasm subsides.
Episodes usually last for 30–60 min. However, some patients present with only cold hands
and do not exhibit the classical triple colour response although they demonstrate a similar
blood flow pattern to classical vasospasm. The factors triggering these changes and the parts
of the body affected are listed in Table 22.1.
Chapter 22: Vasospastic disorders and vasculitis 255
Table 22.1 Triggering factors and the parts of the body

affected by Raynaud’s phenomena
Factors provoking a vasospastic episode

• Cold exposure
• Emotional stress
• Vibration – hand arm vibration syndrome (HAVS)
• Industrial chemicals
• Tobacco smoke
• Trauma
• Drugs
Parts of the body affected by vasospasm
• Fingers (commonest)
• Toes
• Nose
• Ear lobes
• Tongue
• Nipples
Epidemiology
The overall incidence of Raynaud’s phenomena is 11.8% with women affected nine times
more often than men. In certain geographical areas that are prone to a cold and damp cli-
mate, 20–30% of the population is affected. A familial predisposition has also been reported,
which is more likely if the age at onset is less than 30 years.
Raynaud’s phenomenon has particularly been described as a complication of occupa-
tions involving vibrating tools such as pneumatic drills and vibrating saw chains, in which
context it is also referred to as ‘vibration white finger’. Over 50% of people using these
tools may ultimately show symptoms of the disease. The underlying mechanism appears
to be a high-frequency vibration, which exposes the small digital arteries in the fingers
to severe stress. This leads to inflammatory changes and fibrosis in the arterial wall. Since
1985, hand arm vibration syndrome (HAVS) has been listed as an industrial disease in the
UK and these patients might be eligible for compensation provided they fulfil disease spe-
cific criteria.
Chronic exposure to cold temperature at the industrial level has also been associated
with vasospastic disorder and up to 50% of workers dealing with alternate cold and hot water
exposure manifest some symptoms of vasospasm.
Nomenclature used in vasospastic disorders

• Raynaud’s phenomena (RP) – a blanket term used to describe all vasospastic disorders.
• Primary Raynaud’s disease (RD) – vasospastic symptoms manifested on their own
without any underlying systemic disease process.
• Secondary Raynaud’s syndrome (RS) – vasospasm associated with another known
disease entity.
Most people with mild symptoms would usually present to their general practitioner and
usually have primary RD. Those with more severe forms are more likely to have secondary
RS. It should be noted that sometimes the symptoms of vasospasm are present for many years
in advance of diagnosis of the associated secondary RS cause. The differentiation between
primary RD and secondary RS is helpful in planning the initial treatment but is sometimes
difficult, requiring a sound clinical knowledge of the diseases associated with secondary RS
as listed in Table 22.2.
Pathophysiology
Although the exact mechanism of RP is not completely understood, a number of key factors
are implicated in its aetiology and pathogenesis. These include:
1. neurogenic alterations;
2. haemodynamic changes;
3. inflammatory and immune regulation;
4. mechanical;
5. genetic predisposition.
For simplicity, we have described these different mechanisms separately. However, in actual
practice, there is likely to be a complex interaction between them [1].
Neurogenic alterations
Patients with RP have an increased sensitivity to cold and show increased vasospastic
tone in response to cold exposure and other triggering factors. This vascular neuro-
genic response can occur in the absence of an obstructive arterial lesion. The vaso-
spastic response is explained on the basis of enhanced sensitivity of both alpha and
beta adrenergic receptors in the peripheral sympathetic nervous system controlling the
arterial tone.
Haemodynamic changes
Normal vascular tone depends on interactions between the flowing stream of blood (cellular
and plasma components) and the endothelial lining of the vessel wall. Patients with RP show
changes in these microvascular components, which trigger an obstructive and vasospastic
response. These include the following:
• Changes in blood and blood components:
• Platelet activation and aggregation leads to formation of platelet plugs, causing
obstruction to the blood flow and release of thromboxane A2 resulting in
vasospasm.
• Activated leukocytes release free radicals causing further vasoconstriction.
• Red cells become stiff and obstruct the vessel lumen resulting in further obstruction
to microcirculation.
• Plasma viscosity is increased as a result of an increase in plasma proteins.
• Changes in endothelial function:
• Patients with RP show signs of endothelial dysfunction, such as activation of von
Willebrand factor (VWF), which promotes clotting and activates platelets.
• Tissue plasminogen activator (tPA) activity is reduced, resulting in less effective
fibrinolysis and a prothrombotic state in patients with RP.
Table 22.2 Disorders associated with secondary Raynaud’s syndrome
Connective tissue diseases (CTD) Drugs

• Systemic sclerosis • β-blockers
• Systemic lupus erythematosis • Cytotoxic drugs
• Henoch–Schönlein purpura • Anti-migraine drugs including ergotamine
• Scleroderma (CREST* syndrome) Industrial
• Sjögren’s syndrome • Hand arm vibration syndrome
• Rheumatoid arthritis • Vinyl chloride
• Mixed connective tissue disease • Frozen food workers
Obstructive Circulating globulins
• Atherosclerosis • Malignancy
• Buerger’s disease (thromboangitis • Multiple myeloma
obliterans)
• Microemboli • Cryoglobulinaemia
• Thoracic outlet syndrome (cervical rib/band) Miscellaneous
Myeloproliferative disorders • Chronic renal failure
• Leukaemia • Reflex sympathetic dystrophy
• Myeloid metaplasia • Hypothyroidism
• Polycythemia rubra vera
CREST = Calcinosis, Raynaud’s syndrome, Esophageal dysfunction, Sclerodactyly, Telangiectasia.
Inflammatory and immune regulation

Patients with an underlying connective tissue disorder manifest the most severe symptoms
of RS. It is a reflection of an intense systemic inflammatory and immune response involv-
ing the connective tissues in general including the connective tissue in the arterial wall.
The vessel wall in these cases shows a mixture of inflammatory cells including neutrophils,
lymphocytes and plasma cells.
Mechanical
Vibration white finger disease seen in patients exposed to high-frequency vibration tools is
associated with mechanical damage to vessel wall and secondary inflammatory changes.
Genetic predisposition
RP is reported in patients with Scleroderma and CREST (Calcinosis, Raynaud’s syndrome,
Esophageal dysfunction, Sclerodactyly, Telangiectasia) syndrome showing a familial association
with class II MHC antigens. In addition, primary RD has been seen in monozygotic twins.
Key points in the clinical assessment and diagnosis are:
• The clinical picture is episodic with each episode lasting from a few minutes to an hour.
• The RP is usually triggered by exposure to cold or emotional upset.
• Usually fingers and sometimes toes are affected. The distribution of involved digits
could be asymmetrical with one or more digits affected at one time although all digits
can be affected in one particular patient.
• Involvement of one limb only suggests a possible local cause such as a cervical rib.
• Patients either show a biphasic or triphasic response of colour changes or associated
symptoms.
• The first phase is observed after initial exposure to the triggering factor and is
associated with vasospasm characterized by clinical feature of pallor. Some patients
would experience cold hands and numbness at this stage as a result of reduced blood
flow to the digital arteries.
• The initial phase is followed by cyanosis and/or rubor because of reperfusion and is
sometimes associated with pain and paraesthesias.
• Other parts of the body can be affected as described earlier, including ear lobes, nose
and nipples.
• A cyanotic episode without preceding pallor is not usually because of RP.
Laboratory investigations
A number of laboratory tests, including blood flow measurement, blood tests and radio-
logical investigations, can be helpful during investigation of patients with RP.
• The diagnosis of RP is mainly clinical, based on a history of biphasic/triphasic
colour changes in response to cold exposure or emotional upset along with physical
examination if the presentation is during an acute episode.
• Laboratory investigations are mainly aimed at differentiating primary RD from
secondary RS by identifying any underlying systemic disease.
• Digital blood flow measurement is usually not required to make a diagnosis of RP
unless the clinical assessment is inconclusive. The methods used for assessing digital
blood flow include strain gauge plethysmography and computerized thermography, as
outlined in Table 22.3.
• Nail fold capilloroscopy has been established as a corner stone in the diagnosis of
connective tissue disease (CTD) in patients with RP.
• The relevant blood tests include the following:
• Full blood count, urea and electrolytes, C-reactive protein (CRP), erythrocyte
sedimentation rate (ESR) and urinalysis to detect renal dysfunction and anaemia of
chronic illness.
• An autoantibody screen, which should include rheumatoid factor and antinuclear
antibodies.
• Cryoglobulins and cold agglutinins are required only if there is a clinical suspicion
of a prothrombotic state.
• Thyroid function test if suspecting hypothyroidism as an underlying cause of RP.
• A bnormal nail fold capilloroscopy combined with an abnormal autoantibody test
predicts CTD in 90%.
• A baseline upper extremity arterial Duplex scan should be carried out to exclude an
underlying occlusive arterial disease.
• A plain chest radiograph can demonstrate basal pulmonary fibrosis seen in CTD and
cervical rib if present.
Table 22.3 Investigative techniques used for the assessment of blood flow in patients with RP
Investigation Technique Interpretation

Digital systolic blood Two techniques used >30 mmHg drop in systolic pressure on
pressure measurement for measuring blood cooling is considered significant
(measurements taken pressure, strain gauge Avoid testing for 2–3 hours after an
before and after cooling plethysmography and acute episode, as the fingers would still
hands at 15°C) photoplethysmography be in a state of reactive hyperaemia
Nail-fold capilloroscopy Microscopic examination Abnormal enlarged, dilated and
of the nail bed for abnormal tortuous blood vessels indicate
blood vessels. possibility of CTD
Beware that similar changes can occur
in diabetes and nail fold trauma
Computerized Skin temperature is assessed Caution is required in interpreting
thermography as a marker of blood flow the results as both the arterial as well
during different phases of RP. as venous flow can affect the skin
temperature
CTD, connective tissue disease.
Management
Patients with primary RD usually have mild or moderate symptoms and do not require spe-
cific drug treatment. However, they should be given general advice, reassurance and written
information such as that provided by the Raynaud’s and Scleroderma Association, UK. Any
causative drug, such as β-blockers, should be replaced with alternatives and any underlying
disease such as hypothyroidism should be treated. Good symptomatic control is possible
with these measures, despite the non-availability of a definite cure. Figure 22.1 outlines the
management algorithm in patients with RP.
General supportive measures

These are suitable for patients with RD showing mild to moderate symptoms without an
underlying disease process. The key points are:
• Reassurance, explanation and information leaflets.
• Stop smoking.
• Electric gloves and stockings.
• Special shoes with broad and padded fitting (Abel shoes). These are available from the
hospital appliance departments.
• Meticulous attention to areas of skin breakdown, ulcers and infection.
Pharmacological therapy
This should be reserved for more severe symptoms, which affect quality of life or work, as
outlined in Figure 22.1.
Calcium channel blockers

Nifedipine is the most effective and commonly used vasodilator, either by oral or sublingual
route for a quick action. Other drugs include diltiazem and amlodipine but they are less
effective and have fewer side effects than nifedipine because of less vasodilatory properties.
Initial assessment
• Clinical history and examination
• Blood tests
• CXR/Duplex imaging/thermography/capillaroscopy
Primary Raynaud’s Secondary Raynaud’s
General supportive measures

• Avoid cold exposure
• Avoid occupational factors
• Avoid other triggers
• Thermal gloves/boots
• Stop smoking
If not controlled consider:

• Nifedipine/diltiazem
• Treatment of
anatomical causes
(thoracic outlet/cervical ribs)
If not controlled consider:

• Prostaglandins or
prostacyclin
analogues
• Lumbar
sympathectomy (for
lower limb)
• Digital sympathectomy
• Amputation of digits (rare)
Figure 22.1 Algorithm for the management of Raynaud’s phenomenon.
Peripheral vasodilators
These include naftidrofuryl, inositol nicotinate, pentoxifylline and moxisylyte. Patients with
primary RD benefit more from these agents than those with secondary RS. A combination
of low dose nifedipine and one of the peripheral vasodilators may avoid the adverse effects
of both while achieving maximum symptomatic relief [2].
Prostaglandins (PGI2 and PGE1)

These are powerful vasodilatory agents with additional antiplatelet activity and cytoprotec-
tive effects. Iloprost is a synthetic analogue with a longer half-life and is commonly used as an
intravenous infusion over a period of 6 hours for 5 days during the course of one treatment.
The dose is titrated against the individual response and the maximum dose is adjusted at a rate
lower than the dose, causing headache, hypotension and flushing (maximum permissible daily
dose is 2 ng kg–1 min–1). A single course of treatment may provide relief for several months.
Sympathectomy
Cervical and lumbar sympathectomy using phenol injection can sometimes relieve symp-
toms of upper and lower extremity RP, respectively. However, it should be reserved only
as a last option for those with severe symptoms who do not respond to other methods of
treatment. The results of sympathectomy are not always predictable. Thoracoscopic cervical
sympathectomy, in particular, is less effective and is associated with a higher rate of relaps-
ing symptoms with only one-third achieving a long-term benefit [3]. In addition, there are
frequent side effects and for this reason it is not recommended as routine. Digital sympath-
ectomy can be helpful in alleviating the symptoms associated with chronic digital ischaemia
in patients with severe RP [4].
Vasculitidies
Vasculitis is characterised by inflammatory changes in the vessel wall, which can affect either
the arteries or veins and results in destruction of the normal architecture of the vessel wall.
The three commonest vasculitic disorders that vascular surgeons may be involved with are
Takayasu’s arteritis, Giant cell arteritis and Buerger’s disease. In this section we will consider
these common vasculitidies.
Takayasu’s arteritis
Takayasu’s arteritis (TA) produces segmental and patchy granulomatous inflammation of
the arterial wall. The condition was first described by Mikito Takayasu in 1908, and is of
unknown aetiology. The inflammatory process produces medial thickening and intimal
kinking. These changes result in stenoses, occlusion and aneurysmal dilatation. Takayasu’s
arteritis predominantly affects the aorta and its major branches.
Takyasu’s arteritis classically affects young women (90% women, and most aged between 15 and
30 years at onset). The condition usually presents with systemic symptoms and loss of normal
supra-aortic pulses; hence the term ‘Pulseless disease’. In simple terms the condition evolves
from an acute pre-pulseless phase to a chronic pulseless phase; although not all patients pro-
gress on to the chronic phase. The pre-pulseless phase can be divided into the early prodrome
followed by the occlusive phase. The subclavian vessels are most commonly affected followed
by the aorta and carotid vessels. Raynaud’s phenomenon is commonly associated with TA.
Diagnostic criteria
Various criteria have been designed to aid the early diagnosis of TA. These include the
American College of Rheumatology (ACR) criteria, the Ishikawa classification and more
Table 22.4 Modified Ishikawa classification for the diagnosis of Takayasu’s arteritis (TA) [5]. The presence of two
major or one major and two to four minor criteria suggests a high probability of TA.
Three major criteria:

1. Left mid-subclavian artery lesion
2. Right mid-subclavian artery lesion
3. Characteristic symptoms and signs of at least 1 month duration
Ten minor criteria:
1. High ESR (>20 mm h–1)
2. Carotid artery tenderness
3. Hypertension
4. Aortic regurgitation or annuloaortic ectasia
5. Pulmonary artery lesion
6. Left mid common carotid lesion
7. Distal brachiocephalic trunk lesion
8. Descending thoracic aorta lesion
9. Abdominal aorta lesion
10. Coronary artery lesion
ESR, erythrocyte sedimentation rate.
recently a modification of the Ishikawa classification (Table 22.4). The modified classifica-
tion had a sensitivity of 92.5% with a specificity of 95% for the diagnosis of TA in 106 patients
who had TA, as shown on angiograms compared to 20 control subjects [5].
Investigations
Clinical diagnosis of TA can pose diagnostic difficulty in view of its varied presentation.
Various non-invasive imaging studies can help not only in making an early diagnosis
but are also helpful in monitoring the progress of the disease and response to treat-
ment. Conventional angiography, on the other hand, can only provide luminal details.
Ultrasound can identify the early morphological changes in the vessel wall character-
ized by circumferential thickening. Oedema or enhancement of vessel wall on com-
puted tomography (CT) or magnetic resonance imaging (MRI) is considered a sign of
active inflammation. However, this may not lead to the characteristic morphological
changes of TA and, on the other hand, vessels showing no inflammatory changes can
later develop TA.
Sometimes it is difficult to differentiate between oedema of chronic inflammation and
chronic fibrosis in the vessel wall based on CT or MRI alone. Positron emission tomography
(PET) with labelled 18Fluorodeoxyglucose may provide better diagnostic imaging for evalu-
ation of active inflammation in the vessel wall [6].
Treatment
Takayasu’s arteritis is initially managed with high-dose steroids, with or without immune
modifying drugs such as azathioprine and methotrexate. A quarter of the patients do
not enter into remission. Where surgical intervention is being contemplated the patient
should ideally be in remission, although if this cannot be achieved surgery may be neces-
sary to prevent end organ damage. Bypass surgery should be performed to and from
disease free areas of the artery, with biopsies taken from anastomotic sites. Aneurysm
formation tends to occur at a later stage, and the indications for intervention will depend
on conventional factors such as aneurysm size, rate of growth and the presence of distal
embolisation.
Restenosis after intervention has been observed in upto one third of the patients with
TA. This can be minimised by delaying the intervention until the patient is in remission and
starting immunosuppressive treatment after the intervention [7].
Giant cell arteritis (temporal arteritis)

A surgeon’s most frequent encounter with giant cell arteritis (GCA) is during the request
for temporal artery biopsy. Giant cell arteritis is a systemic inflammatory vasculitis predom-
inantly affecting extracranial medium- and large-sized arteries. The inflammatory infiltrate
consists predominantly of mononuclear cells with giant cell formation. Although a variety
of causative factors have been implicated (genetic, immune, infective) the exact aetiology for
GCA remains unknown. The inflammation can be segmental in distribution and for this rea-
son the biopsy results should be interpreted with caution. The vertebral, superficial temporal
and ophthalmic arteries are the vessels most commonly involved. Less commonly the disease
affects larger vessels such as the aorta, subclavian arteries and branches of the abdominal
aorta [8].
The condition rarely presents before the fifth decade and affects women twice as often as
men. Involvement of the ophthalmic artery can result in ischaemic optic neuropathy, which
can result in sudden blindness and hence the condition is treated as a medical emergency.
Presenting symptoms often include following:
• fever;
• new onset headache often seen as the hallmark symptom (usually temporal or occipital
in location);
• scalp tenderness;
• jaw or tongue claudication;
• visual disturbances;
• tinnitus;
• peripheral neuropathy;
• aneurysm formation, if a large vessel is affected.
Investigations
The majority of patients with GCA have an elevated ESR. Traditionally temporal artery
biopsy has been performed to confirm the diagnosis, although a negative biopsy does not
exclude the disease. Ultrasound imaging of the temporal arteries is becoming more widely
available as a diagnostic tool as an alternative to biopsy. It shows a dark halo (oedema) sur-
rounding the diseased vessel – ‘halo sign’. Although not universally accepted, the combina-
tion of ultrasound evidence of inflammatory process plus a good clinical history is taken by
some clinicians to be diagnostic.
Management
Giant cell arteritis is managed with high-dose corticosteroid therapy to induce remission
and to reduce the inflammatory changes in the vessel wall. The presence of visual symptoms
warrants immediate treatment with corticosteroids, as prompt treatment reduces the rsik of
permanant visual impairment. Constitutional symptoms, vascular symptoms, and inflam-
matory markers (ESR) can guide immunosuppressant therapy. Therapeutic surgical inter-
vention is confined to managing consequences of the disease such as aneurysm formation.
Buerger’s disease
Buerger’s disease (thromboangiitis obliterans) is an inflammatory disease that predomi-
nantly affects medium- and small-sized arteries. Leo Buerger first described the condition
in 11 amputated limbs of patients of Jewish descent in 1908, and he termed the disease
thromboangiitis obliterans. The special features of the disease [9] include following:
• The pathological changes are distinct from atherosclerosis with minimal atheroma
formation and more cellular infiltrate along with a striking perivascular inflammatory
reaction resulting in occlusion of the vessel lumen.
• These inflammatory changes can be observed in both the arterial as well as the venous
system.
• Both the upper and lower extremities can be affected, although lower limb disease
occurs more frequently.
• Tobacco exposure is strongly associated with the initiation and progression of Buerger’s
disease, although the exact pathological mechanism is not known.
• Smoking cessation is the only way to stop the disease progression and nearly one-half
of those who continue to smoke end up with a major amputation.
The disease usually affects men at the age of <45 years. Unlike patients with atherosclerotic
disease, patients with Buerger’s usually present with rest pain and tissue loss, rather than
claudication. This reflects the involvement of more distal vessels. The prevalence of Buerger’s
disease is higher in patients of Middle or Far Eastern descent. The disease is historically rare
in women, although this pattern is changing as is reflected by the increasing proportion of
female smokers.
The tissue loss typically involves painful ulceration or necrosis of the digits of the hands
and feet. As the disease progresses it may involve more proximal vessels, although it would be
very unusual to lose proximal pulses from Buerger’s disease alone. Involvement of the small
vessels results in vasospastic symptoms. Patients may also present with sepsis or extensive
foot infections where local infections of necrotic tissue are neglected. The venous inflamma-
tory changes can cause a superficial thrombophlebitis.
Several diagnostic criteria have been suggested to aid correct diagnosis. These include
the following:
• Exclusion of other causes of limb ischaemia (hypercoagulability, emboli, diabetes, etc.).
• Tobacco use (previous or current).
• The presence of distal extremity arterial disease.
• Young age of onset (<45).
The presence of additional features such as upper limb disease, phlebitis migrans, Raynaud’s
phenomena, and radiological signs strengthen the diagnosis.
Investigations
The radiological appearance of Buerger’s disease comprises of relatively normal arteries to
the knee level, abrupt occlusions of the tibial vessels and ‘corkscrew’ collaterals feeding the
distal vessels at the ankle. It is often helpful to look at the images of the unaffected limb, as
these hallmark radiological features may already be present in the asymptomatic limb. It
should be remembered that the similar angiographic features could also be seen in diabetes
and other connective tissue disease.
There are no specific laboratory tests for Burger’s disease. The primary role of laboratory
tests is to exclude other causes of occlusive arterial disease. Likewise echocardiography is
useful to exclude a proximal embolic source.
Management
Medical
The absolute goal in managing Buerger’s disease is to stop smoking completely and per-
manently. Aspirin should be prescribed for its antiplatelet effects, and analgesics for pain
control. Prostaglandin infusions (iloprost) may help to control the symptoms. However, it
is uncertain if it could alter the progression of tissue loss. Patients should be educated about
foot care, prevention of injury and avoidance of cold exposure.
Surgical
Given the pattern of small- and medium-vessel occlusive disease, the options for surgical
revascularisation in Buerger’s are extremely limited. Any significant co-existing proximal ath-
erosclerotic arterial disease should be treated to improve the inflow. Attempting to intervene
with revascularisation is probably futile while the patient continues to smoke. Local amputa-
tion is often required to treat necrosis, non-healing ulcers or intractable pain. Distal ischae-
mic lesions will often auto-amputate, and it is usually helpful to await demarcation even if
surgery is planned. Antibiotics may be needed intermittently for any infective episodes.
Behçets’s disease
Behçet’s disease (BD) is a multisystem inflammatory condition that can affect both arter-
ies and veins. The disorder was first publicised by the Turkish dermatologist Hulusi Behçet
in 1937, who described a syndrome of aphthous ulceration, genital ulceration, and uveitis.
Behçet’s disease has historically been noted to occur more frequently along the old silk trad-
ing routes of the Middle East and in Central Asia; hence, the name ‘Silk Road’ disease. No
specific causative factor has been noted, although it is seen more commonly in individuals
with HLA-B51 gene.
Veins are affected more often than arteries. When large arteries are involved, inflamma-
tory changes affect the vasa vasorum, resulting in medial destruction and fibrosis. Subsequent
damage to the arterial wall can lead to aneurysm formation.
BD is a chronic inflammatory condition characterised by episodes of recurrence and remis-
sion. International diagnostic guidelines emphasise the importance of the following features
for the diagnosis:
• oral (aphthous) ulcers;
• genital ulcers;
• skin lesions;
• uveitis.
Venous involvement is manifested by thrombophlebitis of the superficial and deep veins.
These changes are not wholly confined to the peripheral vasculature and can also involve the
visceral and central veins. Arterial involvement is manifested by aneurysmal disease, which
necessitates intervention due to the risk of rupture. Occlusive disease can occur in BD and
tends to affect medium- and small-sized arteries.
Investigations
There is no specific diagnostic test for BD. The disease occurs more commonly in HLA-B51
individuals, and this gene is more prevalent in the Middle East. The ‘pathergy’ test can be
helpful in the diagnosis of BD but is not 100% specific. The test involves a needle prick to
the forearm. The presence of a papule (>2 mm in diameter) 1 to 2 days after the test consti-
tutes a positive result. Investigation for aneurysmal or venous disease would follow standard
investigative pathways.
Management
Steroid therapy is utilised to ease the symptoms and reduce the inflammatory changes of the
disease. Anti-tumour necrosis factor (TNF) therapy is beneficial in the management of uve-
itis, skin and mucosal symptoms. Aneurysm formation in BD necessitates surgical interven-
tion due to the risk of rupture. False aneurysms at the anastomotic sites are more common
in BD as compared to atherosclerotic disease. Similarly, inflammatory changes can occur at
the site of previous arterial surgery.
Connective tissue disorders

In this section, we will discuss some common connective tissue disorders affecting the vas-
cular system, including Marfan’s syndrome and Ehler–Danlos syndrome.
Marfan’s syndrome
Marfan’s syndrome is an autosomal dominant connective tissue disorder. One-quarter of the
affected population has a new genetic mutation. The underlying genetic defect arises from
mutations of the fibrillin-1 (FBN1) gene based on chromosome 15. Fibrillin is a complex
structural protein that serves as substrates for elastin in the aorta and other connective tis-
sues. Abnormalities of these microfibrils results in weakening of the aortic wall and cardiac
valves. In addition the normal fibrillin-1 protein has a role in vascular smooth muscle devel-
opment by binding to transforming growth factor beta (TGF-β). The role of angiotensin II
receptor blockers has been investigated in modifying the vascular changes induced by
TGF-β.
The major manifestations of Marfan’s syndrome include the following:
• mitral valve prolapse;
• aortic root dilatation;
• aortic aneurysm formation;
• aortic dissection;
• dural abnormalities;
• lens dislocation (ectopia lentis).
External features include tall stature, arachnodactyly, dolichostenomelia (limbs dispropor-
tionately long compared with trunk), joint hypermobility, scoliosis, pectus excavatum, high
arched palate and dental crowding.
Investigations
Genetic investigations have a limited role in the diagnosis of Marfan’s syndrome as not all
FBN1 mutations are associated with Marfan’s syndrome. In addition, genetic testing is not
available in all centres. The diagnosis is currently made on the basis of family history and
clinical features, with or without molecular testing if available. The ‘Ghent’ diagnostic cri-
teria consist of major and minor features based on the systems involved. To be diagnosed
with Marfan’s syndrome using the Ghent criteria requires two major criteria and one minor
criterion affecting different systems or if there is a family history one major criterion and
one minor criterion.
Major criteria include:
• Aneurysmal aorta.
• Dissection of aorta.
• Dislocation of the lens.
• Family history of the syndrome.
• At least four skeletal problems, such as:
• pectus carinatum or pectus excavatu;
• arm span greater than height;
• reduced upper to lower segment ratio – this is when the length of the torso (from
shoulders to legs) is shorter than the length of the legs;
• positive wrist sign – this is when the thumb and little finger overlap when you grasp
the other wrist;
• positive thumb sign – this is when you put your thumb on your hand and it extends
beyond the palm;
• curvature of the spine (scoliosis) with a curve greater than 20°;
• spondylolisthesis;
• flat feet (pes planus);
• protrusion acetabulae on hip X-ray;
• Dural ectasia of spinal cord – shown on CT or MRI.
Minor criteria include:
• myopia;
• unexplained stretch marks;
• loose joints;
• long, thin face;
• high, arched palate.
Patients with Marfan’s syndrome should be regularly screened for cardiac complica-
tions and aortic dilatation to allow prophylactic intervention against potentially fatal
complications.
Management
Open surgical repair of descending thoracic and abdominal aortic dilatation is consid-
ered a reliable treatment option in patients with Marfan’s syndrome and has resulted in
improved survival based on studies carried out over the last decade. It still remains the first
choice of treatment. Thoracic endovascular stenting for aneurysmal disease and dissection
in patients with Marfan’s syndrome has been increasingly used in recent years but the pub-
lished studies in this area are limited to small numbers. Further larger studies are required
with long-term follow up to show a conclusive benefit of thoracic endovascular stenting in
Marfan’s syndrome [10]. In particular there is a worry about the durability of endograft-
ing since the majority of these patients are young. It is also essential that the follow up
in these patients should include monitoring for new aneurysmal disease or anastomotic
aneurysms.
Loeys–Dietz syndrome
Loeys–Dietz (LD) syndrome is a recently discovered autosomal dominant disorder with
many of the clinical features of Marfan’s syndrome. The disorder originates from a defect in
the gene responsible for coding the transformation of growth factor β-receptors (TGF-β 1
or 2). The two genetic defects result in a similar syndrome and the condition was previously
considered to be a subtype of Marfan’s syndrome.
The main clinical features of LD syndrome include:
• arterial aneurysms;
• arterial tortuosity (often in the carotid vessels);
• hypertelorism (widely-spaced eyes);
• bifid or broad uvula.
This combination of major features is usually seen together in other connective tissue disor-
ders. As with Marfan’s syndrome, aneurysms occur at young age and frequently involve the
aortic root. LD syndrome has been classified into two subtypes, depending on the craniofa-
cial involvement. Many of the other features seen in LD are also seen in Marfan’s syndrome
as previously described.
Ehlers–Danlos
Ehlers–Danlos (ED) syndrome is a connective tissue disorder that results in impaired strength,
elasticity, and healing of the tissues. The condition is thought to have an incidence of 1 : 10 000,
and is usually inherited as an autosomal dominant disease. ED syndrome is caused by abnor-
malities of collagen synthesis. The clinical picture depends on the type of collagen defect and
the distribution of that type of collagen in the individual tissues. There are at least six pheno-
types with considerable overlap between them. This often makes the exact categorisation diffi-
cult in up to half of the patients. Type IV (‘vascular’ type) ED syndrome is of most relevance to
the vascular surgeon as it involves the abnormalities of type III collagen [11].
ED syndrome (‘vascular’ type) often presents with arterial or visceral rupture and the diag-
nosis is only made at post mortem. An arterial rupture is either spontaneous or follows
minor trauma, with or without the presence of an underlying aneurysm. Aneurysms often
involve multiple sites and arterial dissection is common. There is a high risk of bleeding
complications and aneurysm formation following surgical intervention in these patients in
view of the tissue friability.
Clinical findings include:
• thin translucent skin;
• easy bruising;
• hyperextensible skin;
• hypermobile joints;
• high and narrow palate;
• dental crowding;
• abnormal wound healing and scars;
• aneurysm formation;
• varicose veins;
• arteriovenous fistulae.
Investigations
The combination of the clinical features listed above in addition to a family history of the
disease or history of a family member unexpectedly dying at a young age should alert the
clinician to the possibility of ED syndrome. To confirm the diagnosis, collagen typing can
be performed on cultured skin fibroblasts. Investigations for vascular involvement should
be non-invasive as conventional diagnostic angiography carries an unacceptably high risk of
haemorrhagic complications.
Management
Treatment for ED is currently confined to the identification and treatment of complications.
Patients should also receive lifestyle advice to avoid trauma. As previously noted surgical or
endovascular intervention should be avoided unless absolutely necessary due to the friabil-
ity of the tissues and bleeding complications.
Investigations
The characteristic facial features and clinical findings will usually suggest the diagnosis. Genetic
tests provide more definitive confirmation of the disease. Patients with LD syndrome should be
placed under long-term CT or MR follow up to detect aneurysm formation at an early stage.
Management
Murine laboratory research has suggested that angiotensin II receptor blockade can reduce
the formation of aortic aneurysms in Marfan’s syndrome. This observation is particularly
relevant in LD where the abnormality in TGF-β activity is the fundamental defect. Therapy
with angiotensin II receptor blockers such as losartan reduces the potential complications
caused by increased TGF-β.
Pseudoxanthoma elasticum
Pseudoxanthoma elasticum (PXE) is a rare genetic condition with an incidence of 1 : 50 000.
The condition involves progressive calcification and fragmentation of elastic fibres contained
within the dermis of the skin, retina and cardiovascular system. More recent studies have
suggested that this may be a systemic metabolic condition. The cardiovascular and gastro-
intestinal complications are the major causes of morbidity and mortality. Vascular calcifica-
tion involves the intimal and medial layers of the vessel wall containing the elastic fibres. The
cardiovascular complications of PXE often occur later than the cutaneous and haemorrhagic
complications. The gastrointestinal bleeding complications result from the fragility of the
calcified submucosal vessels.
The condition is usually first manifested with cutaneous lesions on the lateral part of the
neck (often symmetrical). These lesions have the appearance of small yellow papules and
occur in a linear or reticular pattern. The lesions can coalesce and form plaques and take on
the appearance of ‘plucked chicken’ skin. The skin lesions also commonly occur in the folds
of major joints such as the axillae. As the skin disease progresses it often becomes lax, wrin-
kled and hangs in the form of thick folds. Other clinical features of PXE include:
• gastrointestinal haemorrhage;
• intermittent claudication;
• ischaemic heart disease;
• retinal haemorrhages;
• haematuria.
Investigations
Blood tests are carried out for anaemia, renal impairment and baseline lipid levels. Urine
should be tested for haematuria. Whenever possible, vascular imaging should follow a non-
invasive course because of the risk of haemorrhagic complications after conventional angi-
ography. The cardiac function should be assessed before vascular intervention in view of the
high risk of cardiac complications.
Management
Patents should be counselled regarding changes in lifestyle to reduce the risk of complica-
tions. There is no specific treatment at present to halt or reverse the disease progression. Best
medical therapy should be instituted to reduce the cardiovascular risk, although caution
should be exercised with antiplatelet therapy due to the haemorrhagic risk.
References 4. Kotsis SV, Chung KC. A systemic review

of the outcomes of digital sympathectomy
1. Bakst R, Merola JF, Franks AG Jr, Sanchez for treatment of chronic digital ischaemia.
M . Raynaud’s phenomenon: pathogenesis J Rheumatol 2003; 30: 1788–92.
and management. J Am Acad Dermatol 5. Sharma BK, Jain S, Suri S, Numano F.
2008; 59: 633–53. Diagnostic criteria for Takayasu’s arteritis.
2. Vinjar B, Stewart M. Oral vasodilators for Int J of Cardiol 1996, 54(Suppl): S141–7.
primary Raynaud’s phenomenon. Cochrane 6. Pipitone N, Versari A, Salvarani C. Role
Database Syst Rev 2008; 16: CD006687. of imaging studies in the diagnosis and
3. Thune TH, Ladegaard L, Licht PB. follow-up of large-vessel vasculitis: an
Thoracoscopic sympathectomy for Raynaud’s update. Rheumatology 2008; 47: 403–8.
phenomenon – a long term follow-up study. 7. Park MC, Lee SW, Park YB, Lee SK,
Eur J Vasc Endovasc Surg 2006; 32: 198–202. Choi D, Shim WH. Post-interventional
immunosuppressive treatment and disease. Atherosclerosis 2009; 206:

vascular restenosis in Takayasu’s arteritis. 328–34.
Rheumatolgy 2006; 45: 600–5. 10. Cooper DG, Walsh SR, Sadat U, Hayes
8. Salvarani C, Cantini F, Hunder GG. PD, Boyle JR. Treating the thoracic aorta
Polymyalgia rheumatica and giant-cell in Marfan syndrome: surgery or TEVAR?
arteritis. Lancet 2008; 372: 234–45. J Endovasc Ther 2009; 16: 60–70.
9. Małecki R, Zdrojowy K, Adamiec R. 11. Germain DP. Clinical and genetic features
Thromboangiitis obliterans in the of vascular Ehlers–Danlos syndrome. Ann
21st century – a new face of Vasc Surg 2002; 16: 391–7.
Chapter
23
Critical care considerations and
preoperative assessment for
general and vascular surgery
Ian D. Nesbitt and David M. Cressey
Key points
• Perioperative cardiac complications are the most serious risk to delineate and
pre-emptively manage
• Discussions between anaesthetist, surgeon and cardiologist are frequently required on a
case-by-case basis
• Critical care is an essential and rapidly developing support to many surgical procedures
Introduction
‘Good surgeons know how to operate, better surgeons know when to operate, and the best
surgeons know when not to operate.’ This aphorism reflects the intertwined nature of sur-
gery, anaesthesia and critical care. Poor patient selection or preparation for a particular sur-
gical procedure cannot be entirely compensated for by good anaesthesia or critical care. The
purposes of preoperative assessment include the identification and management of indi-
vidual patient risks as well as appropriate resource allocation.
Sixty per cent of patients undergoing major vascular surgery have significant coronary
artery disease (CAD). Similarly, CAD is common among patients having non-vascular pro-
cedures, so an understanding of the important principles of investigation and management
is important for all surgeons and anaesthetists. This section will therefore concentrate par-
ticularly on cardiovascular assessment, although other disease states are also considered.
General preoperative assessment
When considering an individual patient, the degree of CAD is often difficult to adequately
assess by history and examination alone (e.g. because of limitations in exercise capacity due
to claudication, general fatigue or the time limited nature of an emergency presentation).
However, a good history and examination can allow specific directed investigations to be
carried out. Examples include: echocardiography for patients with suspected aortic stenosis
and cardiopulmonary exercise testing for patients with poor functional reserve. Individual
history taking can be combined with population based information such as the Heart
Outcomes Preventation Evaluation (HOPE) study and Reduction of Atherothrombosis for
Chapter 23: Critical care and preoperative assessment 273
Table 23.1 Some risk scoring systems
Risk score Comments

Association of Anesthesiologists (ASA) General scoring system with significant variation in
scoring between different clinicians
Goldman Index Developed >30 years ago: superseded by later systems
Detsky Score Principally concerned with perioperative cardiac event
prediction
Lee Revised Cardiac Risk Index See Table 23.2 for additional details
V-POSSUM Specific vascular surgical score developed in the UK
Parsonnet Score Scoring system for cardiac surgery
Continued Health (REACH) registry (which show that a history of peripheral vascular dis-
ease strongly predicts adverse cardiovascular outcomes) to help direct other investigations
or management.
Risk assessment and scoring systems for surgery

Aside from a general preoperative assessment, more sophisticated investigations may have
additional benefits. It is important to assess the risk of a particular procedure, not only from
a patient’s viewpoint (adequate information is required for informed consent) but also from
a medical and organisational perspective. This includes appropriate targeting of resource
and therapy, e.g. preoperative investigation and physiological optimisation; efficient use of
critical care beds; consideration of conservative management rather than operative inter-
vention, and also comparisons of observed versus expected outcomes, which may be used
for wider comparative purposes (Table 23.1).
In all these scores, cardiac failure and recent myocardial infarction are the strongest
indicators of postoperative cardiac complications. Aortic stenosis is a strong independent
predictor of perioperative complications: a gradient of 25–50 mmHg increases operative risk
fivefold, and a gradient greater than 50 mmHg increases risk sevenfold.
The Revised Cardiac Risk Index (RCRI) performs best as a predictor in real life practice,
and is detailed in Table 23.2.
Ideally a scoring system allows accurate and individualised prediction of the outcome
from a particular operation for each patient, with low false positive and negative rates. No
scoring system yet allows this. Others use population based historical outcomes to prospect-
ively stratify patients. These can be useful to discuss general risks with patients, but are too
imprecise to be used to attempt individual predictions.
Attempts to improve the prognostic ability of preoperative investigations have yielded
mixed results. It is increasingly recognised that static testing, such as resting echocardiog-
raphy, may provide information on valvular anatomy or pulmonary artery pressures, but has
little prognostic value in vascular surgery. Dynamic testing such as dobutamine stress echo-
cardiography (DSE) adds value to preoperative investigations, but even this has limited prog-
nostic ability, and there is much current interest in cardiopulmonary exercise testing (CPX).
Cardiopulmonary exercise testing involves a combination of exercise electrocardi-
ography (ECG), usually on a bicycle, with concurrent measurement of exhaled CO2 and
oxygen consumption (Figure 23.1). These allow calculations of anaerobic threshold (AT)
Table 23.2 Revised Cardiac Risk Index (RCRI)
Criteria Details
Ischaemic heart disease Angina, myocardial infarction, previous PCI or CABG
Heart failure History of or examination compatible with left ventricular failure.
Paroxysmal nocturnal dyspnoea
Cerebrovascular disease Previous TIA or CVA
Insulin dependent diabetes
Chronic renal impairment Creatinine > 177 mmol l–1 (2 mg dl–1)
High risk surgical case Thoracic, abdominal or pelvic vascular operation
No RCRI criteria = low risk (0.4–1% risk of cardiac complications).
1–2 RCRI criteria = intermediate risk (2–7% risk).
3+ RCRI criteria = high risk (>9% risk).
PCI, percutaneous coronary intervention; CABG, coronary artery bypass grafting; TIA, transient ischaemic attack;
CVA, ceresro vascular accident.
Oxygen consumption CO2 production

25
O2 consumption/CO2 production
20
15
10
AT
5
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Elapsed time
Figure 23.1 Cardiopulmonary exercise (CPX) testing. During gentle (aerobic) exercise, O2 consumption and
CO2 production are similar, so the graphs have a similar slope. During strenuous (anaerobic) exercise, CO2
production is greater than O2 consumption, so the CO2 graph has a steeper slope. The transition between
aerobic and anaerobic metabolism is the anaerobic threshold (AT ), and reflects the integrated ability of the
cardiorespiratory and cellular mechanisms to manage an increased workload. Aside from training and gen-
eral conditioning, a variety of cardiac and respiratory disease states produce diagnostic patterns in CPX data.
and development of any myocardial ischaemia during exercise. An AT below 11ml min–1
kg–1 is considered an indicator of high-risk cases, although more complex evaluations
are increasingly being developed, based on nature of surgery, early versus late ischaemic
changes and absolute AT [1]. It is likely that prognostic cut-off values will vary between
population groups and surgical centres (e.g. the average AT for major surgery at the
authors’ institution is 9.5ml min–1 kg–1, with a mortality/morbidity rate below the UK
average).
These investigations can be used to aid decision making regarding intraoperative and
postoperative care. For example, a patient with a high AT having aortic reconstruction may
be suitable for a level 2 (HDU type) bed postoperatively, while a patient with a poor AT
having similar surgery may be much more likely to require a level 3 (ITU type) bed, with
consequent implications for hospital resource management.
Preoperative decision making requires more than simply assessing risk. If possible,
the perioperative risks should be reduced or eliminated. This may involve optimising the
patients’ condition. Typically, this includes achieving control of co-morbidities including
blood pressure, diabetes, dyslipidaemia and consideration of β-blockade (see below).
The actual investigative and management pathways will vary between institutions, but
ultimately, success depends on good communications and relationships between radiology,
anaesthesia, critical care and surgery. There is an increasing role for pre-assessment clin-
ics (PACs) as co-ordinators of investigations and optimisation, but to be successful, PACs
require strong leadership and clear strategies. Agreed inter-departmental policies are an
important part of this e.g. anticoagulation, major bleeding, cross matching, based partly
on national recommendations such as National Institute of Health and Clinical Excellence
(NICE) guidelines for preoperative investigation.
Co-ordination of complex cases, e.g. decisions about staging of carotid and coronary
surgery, proximal thoraco-abdominal aortic aneurysms requiring partial or complete car-
diopulmonary bypass, staged visceral re-vascularisation should be carried out in specialist
centres with an adequate volume of work to achieve good results. Increasingly unfit patients
are presenting for carotid, thoracic and abdominal endovascular procedures, and again
require expert multidisciplinary team assessment, if possible.
Specific management of other co-morbidities

Pulmonary disease
Obstructive pulmonary diseases (asthma, emphysema and bronchitis) should have long-term
drug therapy optimised. Typically, this is based on multi-modal anti-inflammatory drugs
and bronchodilators. Particular attention to pre-existing pulmonary disease is required if
surgery involves deflating a lung for surgical access.
Restrictive lung disease may be intrinsic, e.g. fibrosis, or more commonly extrinsic, e.g.
due to kyphosis or obesity. Treatment options for the specific restrictive lung disease may
be limited, but may influence the choice of regional versus local or general anaesthesia for a
particular surgical procedure.
As with cardiovascular diseases, close liaison with the anaesthetist and intensive care team
is required to optimally manage patients with significant lung disease. Pulmonary hyper-
tension (PHT) and cor pulmonale are important risk factors for postoperative death, and
patients with severe lung disease should have PHT actively excluded by echocardiography.
Epidural anaesthesia may reduce postoperative respiratory morbidity and reduce length of
ICU stay, but has implications for some drug therapies in the perioperative period (see below).
Obesity
Obesity is a multi-system disease with implications for both anaesthetic and surgical
approaches. Patients with obstructive sleep apnoea may require postoperative respiratory
support in critical care rather than immediate transfer to a ward environment, since respira-
tory complications are more common in the obese. Local services will dictate where these
patients are managed.
Diabetes
In addition to predisposing to macro- and microvascular disease, diabetes is a significant risk
factor for many postoperative complications, including renal failure, myocardial infarction
and death. The long-term benefits of good diabetic control are well established, but the optimal
details of short-term perioperative diabetic control are less clear. Tight (4.4–6.6 mmol l–1) gly-
caemic control may improve outcomes in selected (predominantly cardiac surgical) patients,
but less rigid control (perhaps below 10 mmol l–1) is more practical and possibly safer among
a general patient population. Traditionally, type 2 diabetics undergoing minor surgery are
placed early on an operating list, and simply omit any morning hypoglycaemic agents. Type 1
diabetics may either take their normal insulin or have a glucose-potassium-insulin (GKI) infu-
sion started, depending on the timing and nature of surgery and the local policies in use.
All diabetics having major surgery should have a GKI started preoperatively and contin-
ued until enteral feeding is established again.
Hypertension
Systemic hypertension is a risk factor in the long term for cardiovascular and cerebrovascu-
lar complications, but the evidence that perioperative outcomes are worse unless hyperten-
sion is severe (perhaps >180/100 mmHg) is weak. Local agreements about how an individual
patient with hypertension is managed will depend on multiple other factors.
Permanent pacemakers (PPMs) and implantable cardiac defibrillators (ICDs)

An increasing number of patients present for surgery with pacemakers or ICDs. Aside from
the traditional indications such as sick sinus syndrome and complete heart block, PPMs/
ICDs may now be indicated as treatment of choice for diseases as diverse as refractory atrial
fibrillation, vasovagal syncope, and post myocardial infarction ventricular dysfunction, as
well as in survivors of first presentation life-threatening arrhythmias.
It is important to know what type of device is in use prior to surgery. This allows plan-
ning both for deliberate actions and emergency management, should device interference
occur. Patients may be a source of useful information in this respect, and a 12 lead ECG is
mandatory, but recent device interrogation by a pacemaker clinic is most helpful.
Some PPMs and ICDs may require re-programming or deactivation prior to anaes-
thesia. Unipolar diathermy should be avoided if possible, as should prolonged periods of
continuous diathermy during surgery. This reduces the electromagnetic interference, which
may reset the PPM to a backup mode, or trigger inappropriate defibrillation by an ICD.
Postoperatively, the device may need re-programming, and should be rechecked before the
patient is discharged.
Percutaneous coronary interventions (PCIs), coronary stents and cardiac surgery

More than 90% of PCIs involve insertion of at least one stent, the majority of which are drug
eluting stents (DESs). Although DESs may have a short-term advantage over bare metal
stents (BMSs) regarding early stent thrombosis, they carry the disadvantage of requiring
prolonged (at least 9–12 months post procedure) dual antiplatelet therapy to prevent late
stent thrombsosis, which has a high mortality rate.
Percutaneous coronary intervention usually involves large doses of anticoagulant and
antiplatelet agents. Intravenous heparin has a relatively short (dose dependent) half-life,
but glycoprotein IIb/IIIa inhibitors such as abciximab or tirofiban may have activity up to
48 hours (and can inhibit the activity of any transfused platelets). This and the condition
precipitating PCI have obvious implications for the conduct of surgery and anaesthesia.
In the emergency situation, vascular emergencies may present around the time of a PCI,
e.g. retroperitoneal haemorrhage due to vessel injury during angiography. Additionally some
patients requiring vascular interventions will have had a relatively recent PCI for intercur-
rent active CAD.
Although no specific rules apply to management, since the risk–benefit will depend on
the nature and urgency of surgery, major elective surgery in patients with a DES in place may
be best deferred for at least 12 months. Patients having PCI with BMS or PCI alone should
have at least 6 weeks delay. Patients who have undergone coronary artery bypass grafting
should have non-cardiac surgery delayed for at least 30 days, if possible.
For emergency cases, or those where regional anaesthesia is considered essential, a
platelet transfusion may reduce major bleeding without substantially increasing the risk
of stent thrombosis. Recommended platelet target levels are between 50 and 80 000 µl–1
This is a rapidly evolving area of practice, and recommendations change frequently, so
focused discussion is essential when faced with such situations.
Specific management of drug therapy

The scientific understanding of atherosclerosis has changed over recent years, from a sim-
ple mechanical obstructive model to one of a variable and ongoing inflammatory patho-
logical process with central involvement of platelet activation and aggregation. This and the
observed hypercoagulable state following surgery have implications for both drug manage-
ment and surgical interventions.
Antiplatelet agents
Aspirin
This inhibits thromboxane A2 to reduce platelet aggregation, although up to 40% of patients
may be aspirin resistant. In general, the risk of excess perioperative haemorrhage is suf-
ficiently low to recommend that, for most surgery, aspirin should be continued if possible.
Prostate and intra-cranial operations may have a higher bleeding rate, so the use of aspirin
should be discussed with the anaesthetist involved.
Clopidogrel
This pro-drug, when activated, inhibits fibrinogen binding to platelet glycoprotein IIb/
IIIa receptors and reduces platelet aggregation. A small proportion of patients are clopi-
dogrel resistant, but generally, a loading dose takes several days to exert its full effect, and
the antiplatelet effect lasts for up to 7 days. During this time, any transfused platelets are
also affected, although often to only a small degree. Frequently, clopidogrel is used as either
combination therapy with aspirin, or as monotherapy for patients intolerant or resistant to
aspirin. A significant indication for antiplatelet therapy is to prevent coronary stent throm-
bosis – untreated, around 20% of stents thrombose, with a 20% mortality. Premature cessa-
tion of antiplatelet therapy is a major concern, although it is difficult to precisely quantify
the risk for most individual patients.
Generally, in cases where the risks of bleeding are small, and the risk of thrombosis high,
dual antiplatelet therapy should be continued. For patients with a high risk of bleeding,
discussion between anaesthetist, cardiologist and surgeon is essential. Although no firm
evidence exists, the emphasis is on continuing antiplatelet therapy whenever possible. This
poses a potential problem for many procedures where epidural or central neuraxial blockade
(CNB) is used.
Oral anticoagulants
Many vascular patients take warfarin. New oral anticoagulants, such as the direct thrombin
inhibitor dabigatran and the direct factor Xa inhibitor rivaroxaban, are now available, and
are likely to replace warfarin. These drugs may be easier and safer to manage from a patients’
perspective (predictable fixed dosing without intensive monitoring), but emergency reversal
may be more problematic.
Patients with mechanical heart valves should have an individual assessment of risk from
surgical haemorrhage against risk of cardiac thrombotic complications. This will depend on
the nature and location of the heart valve and the planned surgery. Traditional periopera-
tive management involves stopping warfarin 3 or 4 days before surgery, and using heparin
infusion until a few hours preoperatively, then re-starting the heparin postoperatively until
adequate oral intake allows rewarfarinisation.
Statins
There is some evidence from retrospective and case control studies (e.g. The Statins for Risk
Reduction in Surgery [StaRRS] study) that statins reduce mortality following major surgery,
possibly due in part to stabilisation of inflammatory atherosclerotic plaque. No prospective
trials have confirmed this observation, and any proposed mechanism of action is currently
putative. Nonetheless, patients with dyslipidaemias should have lipid-lowering therapy
started, irrespective of their need for surgery.
β-Blockers
The role of β-blockade in the perioperative period remains uncertain. Initial small trials showed
an all-cause reduction in mortality, but the more recent POISE study showed that myocardial
protection was more than balanced by all-cause mortality and stroke [2, 3]. Patients already
taking β-blockers should probably continue these in the perioperative period, but starting
β-blockers in the immediate preoperative phase for patients with uncomplicated coronary
artery disease should not be considered routine practice at present.
Angiotensin converting enzyme inhibitors (ACEIs) and angiotensin II antagonists

Angiotensin converting enzyme inhibitors are first-line therapy for heart failure.
Angiotensin II receptor (AT II) antagonists may be used as an alternative or additional
treatment. There is uncertainty regarding the optimal management of these drugs in the
perioperative period. Patients continuing these drugs may be more susceptible to hypoten-
sion during anaesthesia, and to renal impairment and possibly even postoperative death.
This is perhaps most marked in hypovolaemic patients. However, patients stopping ACEI
or ATII antagonists may be at risk from non fatal cardiac complications. If ACEIs are with-
held preoperatively, this should be for at least 12 hours. Angiotensin II receptor antagonists
should be withheld for at least 24 hours due to their longer half life.
Management of emergency or acute cases

Increasingly, carotid surgery should be carried out in an acute setting, and many vascu-
lar patients present as emergencies. Time to investigate, treat or stabilise may be limited,
and patient outcome then depends on the expertise and functioning of the whole team.
Outcomes are better when experienced teams, who have practiced and worked together
over prolonged periods, carry out these complex tasks, rather than random on-call teams
of varying abilities. This has implications for centralisation of services and specialisation of
staff, which are beyond the remit of this chapter.
Radiologic embolisation for control of bleeding is possible under local anaesthesia in
selected emergency cases to avoid the risks of general anaesthesia and open surgery. Again,
this has wider implications for service provision beyond the scope of this chapter.
Critical care considerations

Critical care issues cover a wide range of topics both in examinations and clinical prac-
tice relevant to many surgical specialties. Some topics (e.g. brain death, organ donation)
are covered elsewhere in this book. Others should be taken from source documents due
to their regular revision. Examples include cardiac arrest management, treatment of com-
mon arrhythmias and anaphylaxis. These topics are clearly and concisely set out on the
Resuscitation Council UK website (http://www.resus.org.uk, see the guidelines page for a
link to the algorithms for a concise revision aid). A precise knowledge of current algorithms
is vital not just for examination purposes but for everyday clinical practice. It is reasonable
to expect a senior surgical trainee to have a clear understanding of these areas of emergency
management.
Practical procedures often associated with critical care are central venous line insertion,
chest drain insertion and vascular access for haemodialysis. Indications for the procedure,
common complications and the details of surface anatomy and actual insertion technique
have all been asked in the exit examination and candidates should have a clear under-
standing of these. These areas are covered in the Care of the Critically Ill Surgical Patient
(CCrISP) course and the handbook for that course sets them out well.
Definitions of levels of critical care

Critical care can be considered as a spectrum from enhanced ward care, including sup-
port from Critical Care Outreach services, up to the advanced multiple organ support pro-
vided in an intensive care unit. In 2000, the Department of Health produced a document
‘Comprehensive Critical Care’ defining critical care into levels 0 to 3 (see Table 23.3).
In some hospitals each level of care is provided for in distinct clinical areas. This is the
traditional model of acute surgical ward (level 1), high dependency unit (HDU, level 2),
which may be directly supervised by the surgical directorates or managed as a step-down
unit from intensive care by the intensivists, and intensive care or intensive therapy units
(ICU or ITU level 3). Many hospitals are now moving to a system of integrated critical care
units (ICCUs) with all the enhanced level beds [4, 5] in one location.
Definitions of organ failure

Most critical care therapy supports failing organ systems to allow time for other treatments
and the patients own defence mechanisms to generate recovery. Multiple organ dysfunc-
tion syndrome (MODS) describes what is often a secondary injury to tissues caused by a
host inflammatory response to a primary insult. It is characterised by a progressive loss
of function in several different organ systems and its treatment is the core of ICU care.
Deterioration of any organ ranges from reduced to complete loss of function and there are
many definitions of exactly what constitutes organ failure. One set of definitions used in
several large trials of sepsis treatment is summarised in Table 23.4 along with definitions of
Table 23.3 Levels of critical care
Level 0 Patients whose needs can be met through normal ward care in an acute hospital
Level 1 Patients at risk of their condition deteriorating, or those recently relocated from higher
levels of care, whose needs can be met on an acute ward with additional advice and
support from the critical care team
Level 2 Patients requiring more detailed observation or intervention including support for a
single failing organ system or postoperative care and those ‘stepping down’ from
higher levels of care
Level 3 Patients requiring advanced respiratory support alone or basic respiratory support
together with support of at least two organ systems. This level includes all complex
patients requiring support for multi-organ failure
acute lung injury (ALI) and acute respiratory distress syndrome (ARDS). The Sepsis-related
Organ Failure Assessment (SOFA) score is another widely used measure of organ failure [5].
Calculated daily it can be used as a measure of severity of illness and response to treatment.
Monitoring and therapeutic measures for specific organ failure

All critical care measures must be built on a foundation of sound basic care. Treatment of
any organ failure needs to start with an assessment of airway, breathing and circulation (A,
B, C) as per CCrISP or Advances Trauma and Life Support (ATLS) guidelines and these
should always be followed with immediate treatment of any deficit detected. A full phys-
ical examination including measurement of pulse, blood pressure, respiratory rate, pulse-
oximetry and temperature is also mandatory. Critical care therapeutic options for specific
organ system failure are set out below.
Respiratory failure
Immediate measures to treat respiratory failure begin with optimal positioning of the
patient (e.g. sitting up in bed or in a chair to improve chest expansion). Oxygen should be
used in all acutely breathless patients. If there is a strong suspicion of chronic lung disease
likely to include carbon dioxide retention then arterial blood gas analysis (ABG) to monitor
this will be required soon after commencing oxygen. If sputum retention is suspected then
urgent physiotherapy may be required. When clinical signs suggest an active chest infection
then appropriate antibiotics should be started immediately, ideally after obtaining a sputum
specimen. Other treatable causes for breathlessness should be sought and rectified, such as
excessive pain or anxiety, pneumothorax, pleural effusion and cardiac arrhythmias.
For a patient who remains breathless or tachypnoeic, or whose peripheral oxygen sat-
urations remain depressed in spite of supplemental oxygen, ABG analysis is appropriate.
Arterial blood gas analysis results will provide information on pH, oxygen and carbon diox-
ide levels, base deficit, bicarbonate and often lactate. Respiratory failure can be differentiated
into types 1 and 2 on the basis of pO2 and pCO2 (Table 23.4).
Standard mixed concentration (MC) oxygen face masks can deliver up to a maximum of
around 60% oxygen regardless of how high a flow rate of oxygen is used due to entrainment
of air around the sides of the mask during rapid inspiration. The addition of a non-rebreathe
bag to the mask allows more oxygen to be drawn from the bag during inspiration and may
increase inspired oxygen concentration to approximately 80%. For CO2-retaining patients
Table 23.4 Indicators of organ dysfunction
Type I respiratory failure pO2 less than 8 kPa with normal or reduced pCO2
Type II respiratory failure pO2 less than 8 kPa with elevated pCO2 > 6.7 kPa; assistance with
ventilatory effort required
Acute lung injury (ALI) Acute onset
PaO2/FiO2 < 300 mmHG (despite normal pCO2 and regardless of
PEEP)
Bilateral diffuse infiltrates on CXR
No apparent cardiogenic cause (pulmonary capillary wedge
pressure < or = 18 mmHg or 2.4 kPa if measured, or no clinical
evidence of left atrial hypertension
Known triggering event or risk factor
Acute respiratory distress Same as ALI except oxygenation PaO2/FiO2 < 200 mmHg (despite
syndrome (ARDS) normal PaCO2 and regardless of PEEP)
CVS Systolic arterial pressure < 90 mmHg despite adequate fluid
resuscitation and/or vasopressor requirement
pH < 7.3 or base deficit >5 in association with lactate >1.5 × upper
limit of normal
Renal Urine output < 0.5 ml kg–1 for more than 1 hour despite adequate
fluid resuscitation
Haematological Platelet count < 80 000 mm–3 (or 50% fall in 3 days)
PEEP, positive end expiratory pressure; CXR, chest X-ray.
needing tightly controlled oxygen venturi masks will ensure a maximum inspired oxygen
concentration is delivered independent of oxygen flow rate and inspiratory effort.
If levels of oxygen greater than 80% are required then a tight-fitting mask is needed to
prevent entrainment. The use of continuous positive airway pressure (CPAP) may further
enhance oxygen delivery. CPAP (delivered via a nasal or face mask or hood device) increases
functional residual capacity (FRC) by preventing airway pressure falling to zero during
expiration. A rise in FRC reduces ventilation-perfusion mismatch thereby improving gas
exchange. Continuous positive airway pressure requires a cooperative patient with a patent
airway and intact airway reflexes. Above CPAP pressures of 20 cmH2O insufflation of the
stomach may occur with a risk of vomiting and aspiration. Oxygenation may be improved
by CPAP but CO2 removal may not be improved with its use. In some cases a reduced work
of breathing may permit improved respiration and affect pCO2 but this is not a predictable
response.
For patients with rising pCO2, or need for high oxygen concentrations in the presence
of impaired airway protection or impending exhaustion intubation and positive pressure,
ventilation may be indicated. A cuffed tube placed in the trachea will allow a degree of pro-
tection of the airway from aspiration and the application of airway pressure in excess of
20 cmH2O without risk of gastric insufflation. Intermittent positive pressure ventilation
(IPPV) can be delivered as a pressure controlled volume limited ventilation (i.e. bilevel
positive airway pressure [BIPAP]) or as a volume controlled pressure limited ventilation
(conventional intermittent mandatory ventilation [IMV], which may be synchronised with
patients own respiratory efforts [SIMV]). Currently accepted methods include the use of
high positive end expiratory pressure (PEEP) to enhance FRC and thus oxygenation in con-
junction with low tidal volumes (maximum 6 ml kg–1 body weight tidal volume) as per the
ARDSNET study to reduce volutrauma damage to the lungs in ALI [6].
Limiting tidal volumes may lead to rising pCO2 but in the absence of marked acidaemia
this is considered an acceptable side effect of this ventilatory strategy, described as permis-
sive hypercapnia.
Prolonged oral or nasal intubation may lead to long-term injury to vocal cords and to
ischaemic mucosal damage of the trachea leading to stenosis. Tracheostomy has long been
an established alternative. With the advent of percutaneous dilational techniques there has
been an increase in the use of tracheostomy and also a tendency to perform it earlier in a
patient’s ITU stay. Additional advantages include a reduced need for sedation, which in turn
may reduce vasopressor requirements and direct access to the trachea for suctioning in a
patient who is awake and able to cough, communicate and cooperate with physiotherapy.
This may speed the process of weaning from ventilation.
Cardiac failure
Standard assessment of the cardiovascular system begins with clinical examination and
non-invasive assessments including ECG and echocardiography, supplemented by invasive
methods including arterial and central venous access, and cardiac output estimation by
one of several methods. Therapy builds from optimisation of cardiac filling and correction
of rhythm abnormalities to inotropic and vasopressor regimes and perhaps cardiac-assist
devices such as intra-aortic balloon pumps. These therapies aim to ensure optimum oxygen
delivery to end organs. Global oxygen delivery is a product of the haemoglobin concen-
tration in the blood, the oxygen saturation of that haemoglobin and the cardiac output.
Dissolved haemoglobin makes up a very small percentage of oxygen content of the blood.
Therapeutic measures need to ensure each of these aspects is optimised. Less predictable
or amenable to treatment is tissue-level oxygen flux. Microcirculatory abnormalities and
impaired enzyme function may reduce oxygen delivery at cellular level. This is particularly
seen in severe sepsis.
The effectiveness of treatment of circulatory impairment at end-organ level can be
assessed by simple means including Glasgew Coma Scale (GCS) for adequacy of cerebral
perfusion or urine output and creatinine levels for renal perfusion. Assessment of adequacy
of global oxygen delivery can be estimated from lactate levels although local ischaemia or
liver failure may complicate this. Oxygen saturations of less than 70% on a central line ven-
ous sample may indicate inadequate oxygen delivery.
To maximise cardiac function, left ventricular filling, myocardial contractility, and
afterload (systemic vascular resistance [SVR]) should be optimised. Starling’s curve relates
increased stretch on myofibrils to contractility of those fibres. It can be extrapolated to pre-
dict the effect of increasing left ventricular filling on myocardial contractility. With increased
filling, myocardial contractility increases up to a certain point. Thereafter further increases in
volume lead to a decrease in contractility and a failing heart. In a normal heart central venous
pressure, which estimates right atrial pressure will be a reasonable measure of left atrial pres-
sure and therefore left ventricular filling. Where valves are damaged or pulmonary resistance
is increased this may not be true. Pulse contour analysis of arterial pressure traces using com-
plex algorithms are a surrogate measure of left ventricular filling, in particular showing what
effect fluid boluses have on stroke volume (SV). If a bolus causes less than a 10% rise in SV
then further filling may not benefit contractility and may even have adverse effects.
A variety of methods are available to estimate cardiac output itself. The gold standard
remains pulmonary artery flotation catheter methods using dilutional calculations but risks
associated with these, including death, mean their use is declining. Less invasive methods
include oesophageal Doppler (ODM), lithium dilution (LiDCO) and pulse contour cardiac
output analysis (PiCCO). Each method has its own problems and none necessarily gives an
exact measure of cardiac output. Most clinicians accept that the measure is an estimate and
are more interested in the dynamic effects of interventions on the reading to demonstrate
improvements or otherwise.
Once cardiac filling is optimised, contractility can be further assisted by ensuring a suit-
able electrolyte and pH balance in the myocardial tissues. Calcium, potassium, phosphate
and magnesium are all essential factors for muscle contraction and should be closely moni-
tored and optimised. Severe acidosis can have a detrimental effect on contractility and pH
should be normalised whenever possible (see renal failure below). Effective contraction is
also enhanced by sinus rhythm and any new onset arrhythmia should be corrected.
Thereafter contractility can be augmented by the use of positive inotropic agents. The
most commonly used are epinephrine (adrenaline) and dobutamine with strong agonist
actions on β1-adrenoreceptors. Although cardiac output influences global oxygen deliv-
ery, the systolic blood pressure determines localised perfusion of tissues such that a very
low systolic pressure is likely to be harmful. Drugs such as dopamine and dopexamine are
known to be positive inotropes but are often used in lower doses by clinicians who believe
they can improve specific regional blood flow (in particular renal and splanchnic) but the
evidence for this is not strong. Their other adverse side effects (tachyarrhythmia, dopamin-
ergic receptor stimulation) make these two drugs less attractive as pure inotropes.
Profound vasodilation in severe sepsis may reduce systolic pressure to harmful levels
even in the presence of a high cardiac output. Drugs acting as α-adrenoreceptor agonists
can produce vasoconstriction to improve overall tissue perfusion pressure and tissue oxy-
gen delivery. Norepinephrine (noradrenaline) and phenylephrine are first-line vasopres-
sor agents. Vasopressin and terlipressin are second-line agents, which may supplement the
actions of norepinephrine. In patients with severe sepsis and vasopressor resistant hypoten-
sion low-dose steroids (50 mg hydrocortisone IV qds) may reduce vasopressor requirements
but effects on outcome are unclear.
Renal failure
Classically causes of renal failure can be divided into pre-renal, renal and post-renal. Pre-
renal causes arise when an insufficient oxygen supply is available to the kidneys. Optimisation
of the cardiovascular system, as described above, is key to preventing or limiting pre-renal
injury. Clearly, ensuring patency of renal vessels is of primary importance. Thereafter opti-
mal filling, maintenance of cardiac output and adequate mean arterial pressure (with vaso-
pressors if needed) are the only proven therapies that reduce the degree and duration of
renal failure from this cause.
Treatable causes of renal failure usually involve removal or avoidance of nephrotoxic
agents. Non steroidal anti-inflammatory drugs (NSAIDs), aminoglycocides and iodinated
contrast media are commonly encountered in critical care but should be avoided when
feasible.
In the specific case of renal failure following rhabdomyolysis, hydration and a forced
alkaline diuresis may limit injury (myoglobin is precipitated in the collecting tubules at acid
pH). Sodium bicarbonate infusion may be needed to achieve this. Furosemide should be
used with caution, if at all.
As a general rule, diuretics increase urine volumes and are useful in patients with
volume overload but do not reduce the occurrence of or duration of renal failure. Their
use in patients with developing acute renal failure may worsen outcome by causing
hypovolaemia.
Mannitol has been used, particularly in the context of aortic surgery as a ‘reno-protective’
measure. It is an osmotic diuretic and may produce an increase in urine volume, but there is
minimal evidence to support any effect on onset or severity of renal impairment.
Dopamine and dopexamine have both been used at ‘renal’ doses to try to enhance the per-
fusion of the reno-splanchnic vascular systems. Again, the evidence for efficacy is limited.
Contrast induced nephropathy (CIN)

This usually occurs within 72 hours of exposure to iodinated contrast media. It is usually
transient, resolving in 7–10 days, but is occasionally permanent and is associated with an
increased morbidity and mortality. Contrast induced nephropathy is dose dependent and
use of high osmolar contrast media carries an increased risk. Using the lowest possible dose
of contrast media and adequate pre-procedure hydration will reduce the incidence of neph-
rotoxicity. N-acetyl-cysteine or sodium bicarbonate infusions are used in some centres to
reduce contrast injury but the evidence for this is weak.
Post-renal causes of renal failure consist of obstruction to outflow of urine, this may be
an occluded ureter post-surgery, prostatic hypertrophy or tumour or other forms of blad-
der outflow obstruction. In most patients with impending renal failure an ultrasound of the
renal tract is a mandatory part of investigation. Bladder catheterisation, both as a monitor-
ing and also a therapeutic measure, is usual in patients with renal impairment.
Established acute renal failure

Acute renal failure leads to raised levels of urea and creatinine, the potential for fluid over-
load and worsening acidosis and hyperkalaemia. This combination, if left untreated, will
result in arrhythmias and eventually death.
For treatment of severe hyperkalaemia calcium gluconate or calcium chloride 10 mmol
IV, should be given immediately. These act as membrane stabilisers to reducing the risk
of life-threatening ventricular fibrillation. Urgent reduction of serum potassium can be
achieved by pushing the potassium ions into the intracellular space using dextrose and
insulin (15 IU actrapid in 50 ml 50% dextrose over 15 min) and by correcting the acidosis.
Salbutamol either IV or nebulised can also effectively reduce serum potassium in the emer-
gency situation. Intravenous sodium bicarbonate infusion will increase serum pH as long as
the patient maintains an ability to increase ventilation to remove the extra carbon dioxide
this generates. Potassium is driven into the intracellular space by alkalosis. Chelating agents
such as calcium resonium (for adults 30 g given as a PR enema) will further help to remove
potassium from the serum.
Definitive treatment is still likely to be required. Intermittent haemodialysis (IHD) or
continuous veno-venous haemofiltration (CVVH) can provide this.
In a haemodynamically stable patient IHD over 3–4 hours via a twin lumen venous
access line (vascath) will clear acidosis, hyperkalaemia and uraemia rapidly. The rapid fluid
shifts involved with this make it difficult to achieve in the more unstable patient. Many units
prefer to use CVVH in this group.
In a CVVH circuit venous blood is pumped into a filter with a pore size suitable to allow
water and small molecules to pass through whilst preventing the passage of most proteins,
cells and platelets. In this ultrafiltrate, waste such as urea and creatinine along with a mixture
of electrolytes and water are separated out and diverted to a waste bag. The volume removed
is carefully measured and then replaced with a balanced solution of water and electrolytes.
Potassium is added separately as required. The replacement fluid is either lactate buffered
(which remains stable for long periods) or ‘lactate-free’ bicarbonate buffered (which requires
mixing immediately prior to use). Units treating patients with liver impairment tend to use
lactate free fluid as the liver is the main site of lactate metabolism. If a greater degree of clear-
ance is required there is an option to apply a counter-current flow of the balanced solution
through the filter with the aim of increasing the concentration gradient for solutes to exit
into the waste flow (haemodiafiltration).
Continuous veno-venous haemofiltration, as its name implies, is run as a 24 hours-a-day
process. By adjusting the amount of fluid replaced into the patient relative to the volume of
ultrafiltrate removed it is possible to remove water to achieve the desired daily fluid balance.
In unstable patients the fluid shifts and haemodynamic effects of CVVH are better tolerated.
There may also be less injury to the kidney during the filtration process due to the greater
haemodynamic stability than with IHD.
Nutrition
Postoperative patients and those with sepsis and systemic inflammatory response syn-
drome (SIRS) usually mount a highly catabolic response. Significant loss of muscle mass and
strength may prolong ICU and hospital stay, inadequate nutrition may also affect wound
healing. As such, nutrition is a vital part of their critical care therapy. The average catabolic
patient in ITU will need a daily calorie intake of around 1600 kcal day–1. Where possible, the
route of delivery should be via enteral feeding. Early enteral feeding even at low levels (10 ml
hour–1) has been shown to increase splanchnic blood flow and there is better maintenance
of gut mucosal anatomy.
In the unconscious patient this might be nasogastric (NG) or percutaneous endoscopic
gastrostomy (PEG) tube delivered. Many ICU patients develop gastroparesis for a variety of
reasons and post-pyloric feeding should be considered early in a patient with high gastric
aspirates not responding to prokinetics. With foresight, nasojejunal tubes can be sited dur-
ing a laparotomy using direct manipulation. Later insertion of post-pyloric tubes using blind
techniques or with endoscopy causes delay in feeding and is not without risk.
A range of feeds is available, each with its own benefits, and the choice should be tailored
to the individual. Osmolite is a standard feed with 1 kcal ml–1. Nepro is a low volume feed
with 2 kcal ml–1 with low potassium, sodium and phosphate loads suitable for those with
renal failure. Pulmocare and oxepa have a high fat : carbohydrate ratio so generate less CO2
on metabolism, which may be useful for those with severe respiratory failure. Impact has
relatively high protein content with added arginine, fish oils and omega 3. It may be bene-
ficial in immunocompromised patients with sepsis. Other feeds exist and the choice should
be made in consultation with a dietician.
Where it is not possible to feed via the gut then total parenteral nutrition (TPN) is used. A
dedicated central venous access port and exemplary aseptic technique is needed when using
Table 23.5 Definitions in critical illness
SIRS (systemic Requires two of:

inflammatory • Pyrexia > 38oC or hypothermia < 36oC
response syndrome) • Tachycardia > 90 bpm (in absence of β-blocker)
• Tachypnoea > 20 breaths per min or PaCO2 < 4.3 kPa (32 mmHg) or a
requirement for mechanical ventilation
• White cell count > 12 000 cells mm-3 or < 4000 cells mm-3
Sepsis SIRS + documented source of infection
Severe sepsis Sepsis plus altered organ perfusion or evidence of dysfunction in one or
more organs
Septic shock Refractory hypotension in addition to the above in the presence of systemic
infection
MODS (multiple Presence of altered organ function in acutely ill patients such that
organ dysfunction homeostasis cannot be maintained without intervention. Usually involves
syndrome) two or more organ systems
TPN. Line-related sepsis leading to septicaemia is a particular problem associated with TPN
as the solution provides ideal growth media for bacteria. The solution used for TPN should
be tailored to the individuals’ biochemical and nutritional needs although ‘off-the-shelf ’
preparations are available. Due to the high lipid content of TPN fluids, fatty infiltration of
the liver is relatively common and liver function tests should be measured regularly.
Severe sepsis
Definitions of SIRS, sepsis, severe sepsis and MODS are often a topic of discussion in the
examination situation and definitions are presented in Table 23.5. For a summary of the man-
agement of sepsis the European surviving sepsis campaign editorial sets out an extensive dis-
cussion of treatment options. A succinct summary can be derived from Tables 23.3–23.5 of
this paper [7]. Treatment revolves around source control to eradicate the infection site with
supportive therapy to maintain organ function during recovery. It is vital to liaise closely with
the microbiology department and send regular culture specimens to screen for infection,
identify pathogens and establish antibiotic sensitivities. Discussion of detailed antibiotic use
is beyond the scope of this chapter.
Specific therapy for severe sepsis may include activated protein C (aPC). The exact role of
this drug is unclear at present, since the initial encouraging results from a large international
trial have not been upheld in daily practice. A new trial is underway to clarify the issue.
Tight glycaemic control (blood glucose 4.5–6.0 mmol l–1) in patients with severe sepsis
was thought to be beneficial initially [8]. However a recently published Australian study
suggested very tight control may be deleterious, in part due to hypoglycaemia [9]. Current
thinking supports blood glucose control between 6–10 mmol l–1 in septic patients.
Despite all the currently available treatment options for severe sepsis with MODS the
mortality rate remains high (approximately 25% ITU mortality, 40% hospital mortality).
Finally it must always be remembered that the goal of any treatment is to allow the
patient to return home with an acceptable quality and quantity of life. If the chances of
achieving this become minimal then consideration of the appropriateness of continuing that
therapy is vital. This is certainly the case in ICU where many life-prolonging treatments are
available in the face of severe illness. National mortality figures for those admitted to ICU
show that for all-cause admissions 18% die in the ICU with overall hospital mortality being
27%. For elective and emergency surgery these figures are 2.8 versus 7% and 14.4 versus
24.4%, respectively (Intensive Care National Audit & Research Centre Case Mix Programme
(ICNARC CMP) data 2008). Where prognosis is poor a change to palliation should be seen
as an active decision in the course of ICU care. Appropriate use of palliative care schemes
such as the Liverpool Care Pathway may give dignity and ensure relief from unnecessary
suffering for patients at the end of their lives.
References organ dysfunction/failure. Intensive Care

Med 1996; 22: 707–10.
1. Cardiopulmonary Exercise Testing Website. 6. The Acute Respiratory Distress Syndrome
www.cpxtesting.com (accessed 2 April Network. Ventilation with lower tidal
2009). volumes as compared with traditional tidal
2. Deveraux PJ, Beattie WS, Choi, PT-L, et al. volumes for acute lung injury. N Engl J Med
How strong is the evidence for th use of 2000; 342: 1301–8.
perioperative ß-blockers in non-cardiac 7. Dellinger RP et al. Surviving Sepsis Campaign
surgery. BMJ 2005; 331(7512): 313–21. International Guidelines for Management of
3. POISE Study Group, Deveraux PJ, Yang H, Severe Sepsis and Septic Shock 2008. Intensive
et al. Effects of extending release metoprolol Care Med 2008; 34: 17–60.
succinate in patients undergoing non- 8. Van den Berghe G, Wouters P, Verwaest C,
cardiac surgery (POISE Trial): a radomized et al. Intensive insulin therapy in critically
controlled trial. Lancet 2008; 371(9627): ill patients. N Engl J Med 2001; 345:
1839–47. 1359–67.
4. Auerbach A, Goldman L. Assessing 9. The NICE SUGAR Study Investigators.
and reducing the cardiac risk of non cardiac Intensive versus conventional glucose control
surgery. Circulation 2006; 113: 1361–76. in critically ill patients. N Engl J Med 2009;
5. Vincent JL et al. The SOFA (Sepsis-related 360: 1283–97.
Organ Failure Assessment) score to describe
Chapter
24
Access surgery
David C. Mitchell and William D. Neary
Key points
• Planning for vascular access in renal failure needs to begin at least 6 months prior to
the predicted onset of dialysis
• Surgery should be aimed at the most distal veins first to preserve the more proximal ones
• Autologous arteriovenous (AV) fistula are the most durable form of access
• Most access procedures can be performed under local anaesthesia as day case surgery
• A good access programme should have an individual to coordinate investigations and
surgery
• Surveillance improves access graft function and longevity
Introduction
Vascular access is required in those patients where frequent repeated access to the circula-
tion is required. The vast majority need this for haemodialysis to treat renal failure. Other
examples are for plasmapharesis, injection of antibiotics (e.g. cystic fibrosis) or drugs (e.g. in
chemotherapy for neoplasia).
The focus of this chapter will be on the provision and maintenance of vascular access
for haemodialysis, but the principles of access placement and surveillance hold good for
patients with alternative requirements.
Diagnosis of need for access placement

At first sight this appears straightforward. Those patients with end-stage chronic kidney
disease will need dialysis and should have access placed. As AV fistula have the lowest mor-
bidity and failure rate, once established, this is regarded as the ‘ideal’ form of access. Many
fistula and all grafts will require surveillance and some may need interventions to keep them
functioning adequately. It serves little purpose to place a fistula some years in advance and
then spend a lot of resources maintaining it, never to see it used. Conversely patients who
present needing dialysis without established access have worse survival. For this reason
the National Service Framework in the UK identified that access needs to be placed some
months in advance of its anticipated need [1].
Studies show that using serum creatinine as a measure of severity of renal failure is not
very accurate. This is because renal function can vary significantly for the same creatinine
Chapter 24: Access surgery 289
depending on age, race and body mass. A better measure is the glomerular filtration rate
(GFR), but measurement of this is time consuming. Most centres use estimated GFR (eGFR),
which is calculated using one of several formulae from the combination of plasma creatinine,
age, sex and race. This method is sufficiently reproducible in adults to give reliable trends of
renal function over time.
There are opinions as to how best to decide the timing of access placement. Some recom-
mend the use of a single eGFR measurement as a trigger, with figures between 15 and 25
being those most commonly used. Studies in some centres, including our own, have shown
that eGFR is a good predictor of cardiac death, but a poor predictor of time to onset of dialy-
sis. The best predictor of the need for renal replacement therapy (RRT) (i.e. dialysis or trans-
plantation) is a combination of rate of creatinine fall and clinical decision by a nephrologist.
There are inaccuracies associated with each approach. As a result, some access will be placed
and never used. The key to an efficient service is to maximise those starting with established
access, whilst minimising unnecessary operations. Those units wanting to start a high pro-
portion of their patients using an established fistula will have a significant redundancy rate
built into their surgical programme to achieve this.
Organising an access service

Most Western medical services recognise that access surgery should ideally take place some
months prior to the anticipated date of first use. This allows for delays in providing operating
time and also for revisional surgery should the first procedure prove inadequate.
Superficially organisation of an access service appears straightforward. The patient is
seen, a diagnosis of impending need for dialysis is made and a referral made for surgery.
Unfortunately many patients will not have suitable visible veins for fistula placement and
will need preoperative assessment with a Duplex scan. Those with identified veins and pat-
ent arteries will need to be assessed to determine the likelihood of success.
Successful access surgery is dependent on both arterial (i.e. flexibility and size) and vein
quality (size, continuity, distensibility). Where the patient has had previous central lines, the
central veins will need checking for stenosis that may jeopardise the success of the procedure
by obstructing blood flow in the access.
Once all the information is available, an operating list space will need to be identified and
surgery will be performed. Following surgery, some assessment of the success of the proce-
dure is needed and the patient will need to be reviewed to ensure that the fistula is suitable
for dialysis. In those where the fistula is not able to be used within a few weeks, further surgi-
cal or radiological procedures will be required to make the fistula suitable for use.
The solution to meeting these challenges and delivering a timely service is best addressed
by having a defined pathway of care [2] and a multidisciplinary team able to respond to the
varying needs of the patient. A central figure in the service is the ‘access co-ordinator’ who
provides a link between patient, dialysis unit and the hospital team.
Timing of access placement

Patients need to come to dialysis with functioning access. This requires timely placement,
ideally about 6 months ahead of the anticipated date of first dialysis. The access co-ordi-
nator should be informed by the medical team of the proposed date for starting RRT. This
will trigger placement on the waiting list. If suitable veins are visible, adjacent to a palpable
arterial pulse, then no further preoperative planning is required. At this stage the patient
can be listed for surgery. Routine surgical clinical review serves to impose delays without
improving care and is best focused on difficult clinical problems. If no suitable veins and
arterial pulse can be identified in the clinic, then ultrasound scanning should be undertaken
to facilitate planning of surgery.
If a fistula is planned, placement should occur as soon as the patient is within 6 months
of the anticipated RRT date. If no suitable veins are identified, and placement of central
venous catheter or access graft is planned, then these can be inserted much closer to the time
of RRT commencement, as the time required before they can be used is much shorter.
From this, it is clear that planning needs to begin at least 6 months prior to the antici-
pated RRT date. This will ensure that the patient comes to RRT with appropriate access for
their needs.
Types of vascular access

There are three routes used for establishing vascular access. These are the arteriovenous fis-
tula (AVF), a prosthetic access graft or a central venous catheter (CVC). Each has their own
advantages and disadvantages (see Table 24.1).
Choosing which type of access to place requires experience and a good understanding of
the needs of the patient. There are some principles that govern the choice of access.
1. Fitness of the patient
Both AVF and grafts require a good blood flow both to maintain patency and to be
useable for dialysis. Patients with failing hearts, or those unable to increase their cardiac
output to maintain access flow, will not be suitable for these types of access. In this
situation, a decision has to be made to use a CVC or not to embark upon haemodialysis.
2. Adequacy of vessels
The state of both arteries and veins needs to be considered when planning vascular
access. The most commonly encountered difficulty is an absence of suitable veins
due to previous cannulation. Duplex scanning is a valuable resource for identifying
suitable veins. Most surgeons are reluctant to use veins below 2–2.5 mm in diameter
in adults, although smaller veins have been used successfully [3].
The population coming forward for access surgery is steadily ageing. Many will have
diabetes. Both diabetes and chronic kidney disease lead to vascular calcification. While
small amounts of vessel wall calcification may not prejudice the success of surgery,
heavily calcified (i.e. rigid) arteries are unable to increase flow to allow fistula maturation
or to maintain graft patency. Arteries need to be an adequate size and again 2–2.5 mm
diameter would be regarded as the lower limit for successful access placement.
3. Handedness
A 4-hour dialysis session is unpleasant enough without rendering patients unable
to use their dominant hand. For this reason it is customary to place access in the
non-dominant limb wherever possible. If no suitable vessels are identified, then it is
acceptable to use the dominant limb after discussion with the patient.
4. Central vein stenosis or occlusion
Patients with previous CVC placement, pacemakers or port-a-caths may have stenosis
of their central veins. In these patients, imaging of the central veins is advised.
Ultrasound may show normal respiratory phasicity suggesting adequate patency. If
there are doubts about the veins in the root of the neck, then arm venograms are a
good method of imaging the central veins.
Table 24.1 Advantages and disadvantages of vascular access procedures
AVF Graft CVC

Ease of Variable depending on Variable, often needs Usually straightforward
placement state of veins GA or loco-regional unless CV stenosis
Easy LA procedure for block
primary wrist and elbow
AVF, more complex for
transposition procedures
Advantages Usually LA day case Can bridge long Insertion under LA,
procedure (75%) distance between tunnelled if required for
Robust once established artery and vein more than 3 weeks
Resistant to thrombosis Usually easy to Not dependent on
and infection establish provided good cardiac output
vessels adequate Can be used
Revision rate about 15% pa
immediately for dialysis
Disadvantages May take months to mature Needs a couple of Vulnerable to sepsis
for use weeks to become Associated with higher
Difficult to establish, only incorporated prior to mortality than AVF
about 55% will be useable use (unless designed
Venous stenosis
1 year after formation for immediate use)
around catheter may
May need revising to Prone to thrombosis, preclude future access
promote maturation mostly due to venous placement in limb
intimal hyperplasia
More susceptible to
infection than AVF
Revision rates about
85% pa
AVF, arteriovenous fistula; CVC, central venous catheter; LA, local anaesthesia; GA, general anaesthesia.
5. Patient preference
Some patients may have a strong preference for certain types of access. Younger
patients may be particularly sensitive about the cosmetic appearance of a fistula and
may request brachio-cephalic AVF, thigh grafts or CVC. It is the role of the medical
team to make patients aware of the advantages and disadvantages of their choices and
then to support them in providing acceptable access.
Operative techniques
The choice of operation is based on the factors above. The ideal access is usually an AVF
in the distal part of the non-dominant limb. An AVF has the advantage of being robust,
thrombosis and infection resistant when compared to the alternatives. Arteriovenous fistula
requires the least maintenance of any type of access. The principle issue with AVFs is the dif-
ficulty faced in establishing them. This may take more than one operation. In our unit 55%
of AVFs are useable at one year after a single operation. This rises to 84% after two opera-
tions. Only a small number of patients require more than two operations. It is this need for
a second procedure in a substantial minority that dictates that patients planned for AVF for-
mation need to start on their surgical pathway some months prior to the onset of dialysis.
Patients may anticipate more than one operation to maintain access over the lifetime of
their renal failure. It is in their interests to minimise the trauma associated with surgery and
to plan the simplest procedure compatible with a good outcome. For this reason, local or
regional anaesthesia is preferred wherever possible. Many procedures can be carried out in
the day case facility, minimising hospital time. A survey of the workload in our unit (about
450 operations a year) identified that about 75% are done as day case procedures under local
anaesthesia.
Arteriovenous fistula (AVF)

These should be fashioned as end of vein to side of artery anastomoses. At the wrist or in the
anatomical snuffbox, some surgeons prefer end-to-end anastomoses. The results from the
two approaches seem to be broadly similar. At the wrist a long anastomosis is preferred. This
reduces narrowing from intimal hyperplasia and maximises flow in small vessels. Prolene
sutures are commonly used. There is a suggestion in the literature that the use of external
clips for the anastomosis may produce technically superior anastomoses [4]. The clips are
considerably more expensive than sutures, which may influence surgeon’s choice.
More proximal AVFs are fashioned using the end-to-side configuration, but more care
needs to be taken to keep the anastomosis at about 8 mm length. Larger anastomoses are prone
to develop very high flows, which may steal blood from the distal part of the limb. In some
cases high flows (usually of 2 l min–1 or more) can precipitate high output cardiac failure.
When the easily accessible superficial veins (cephalic and forearm basilic) are exhausted,
then attention may have to focus on the deeper veins (arm basilic and great saphenous).
These veins may form good quality AVFs but need transposing towards the skin to make
them accessible to dialysis needles. Most surgeons will undertake to transpose and form the
AVF at a single operation. Occasionally the vein may be a bit small and then it is accept-
able to perform a simple end-to-side anastomosis and await vein enlargement before sub-
jecting the patient to the bigger transposition operation. The results of either approach are
broadly equivalent, with greater success rates (at about 75% primary patency) from basilic
vein transposition.
Arteriovenous access grafts

The commonest material used is polytetrafluoroethane (PTFE), but many types of access
graft material can be used. Some grafts are designed with self sealing coatings to allow
immediate cannulation after placement. Grafts may be placed in any configuration between
a suitable artery and vein. Where they are placed over muscle (upper limb and thigh) it is
important to place them superficially to the deep fascia as peri-graft haematomas can cause
compartment syndrome if the graft is placed too deeply.
A graft of 6 mm diameter is usually adequate for dialysis. Smaller grafts may prove dif-
ficult to puncture accurately with dialysis needles. The exception is the stepped graft where
the arterial end is deliberately of lower calibre. This is to try and reduce turbulent flow, which
is thought to contribute to venous intimal hyperplasia and graft failure. The larger upper
part of the graft is the portion that is used for dialysis.
A number of variations have been employed to try to reduce the narrowing seen at the
venous end. Grafts with pre-dilated venous ends are sold commercially, but there is no good
evidence of superior outcome from their use. Vein cuffs may be used, as for lower limb arte-
rial bypass, but their efficacy remains unproven.
Where grafts are placed, they may need subsequent removal for sepsis. This can be
a very difficult operation, particularly around the artery in an infected groin. Complete
removal of all graft material is required to eliminate infection. It is the authors’ preference
to place a vein interposition cuff between artery and graft to facilitate subsequent removal.
The dissection of the infected graft goes down to the cuff:graft anastomosis. A clamp can
be placed across the cuff, which is then oversewn avoiding the need to formally dissect
the artery.
Central venous catheters (CVCs)

The main advantage of CVCs is that they can be placed and used immediately. They are the
mainstay of providing dialysis in acute kidney injury and in those with chronic kidney dis-
ease who present acutely before placement of permanent access.
The catheters should be placed in large central veins using a Seldinger technique under
ultrasound guidance. It is preferable to use the jugular veins. This avoids damage to the
subclavian veins from the trauma of placement, or subsequently due to scarring of the vein
around the catheter. Narrowing of the subclavian vein may jeopardise the placement of
access distally within the limb.
Central venous catheters are prone to infection, so if it is anticipated that the catheter
will remain in place for more than a week or two, a tunnelled catheter should be placed. The
catheter has a cuff, which is situated just within the tunnel and reduces migration of micro-
organisms along the catheter. Eradication of methicillin-resistant Staphylococcus aureus
(MRSA) carriage is an important part of reducing sepsis related to CVC use.
Care of access after placement

Non-maturation of AVF
A significant minority of AVF fail to mature. There is evidence that early identification and
correction of abnormalities significantly improves maturation rates [5]. Flow rates increase
significantly within minutes of AVF formation, and reach near maximal levels within 2–7
days. A Duplex scan at 1–2 weeks will detect those with low flow that are very unlikely to
mature to use. This allows early intervention. The commonest cause of non-maturation is
stenosis in the ‘swing’ segment of the fistula, where the vein is mobilised to swing across
to the artery. Revision or angioplasty are equally effective at correcting the abnormality,
although angioplasty may need to be repeated more often to obtain the same result [6].
Large tributaries may split the flow leaving no single vein large enough to use for dialysis.
Identification and ligation of such tributaries can improve maturation rates significantly.
Finally an increasingly common problem is with an AVF that is too deep to needle due
to patient obesity. The solution here is to mobilise the AVF up under the skin (disconnecting
and reconnecting the anastomosis if need be) to allow improved access.
Early failure of grafts

This should be a rare occurrence if patients have been carefully assessed and selected for graft
placement. The commonest cause is a technical surgical failure and failure within 24 hours
should stimulate the surgeon to re-explore the graft. Occasionally it will become apparent
that the artery or vein was insufficient for the procedure, but often a technical problem at
one or another anastomosis can be corrected to improve access function. Sometimes in
longer grafts and those with loop configurations, a kink in the tunnelling is responsible for
early failure.
Surveillance of access
Vascular access requires surveillance following placement to ensure that it functions cor-
rectly. This is of vital importance to patients as the access is their lifeline, without which
they are unable to dialyse. Each type of access has its own particular needs, but the common
theme of surveillance is to detect abnormalities before access failure and to resolve them
before the access becomes unuseable. There is debate about the value of surveillance with
some studies suggesting improved outcomes and others not [7, 8].
The problem of failure is most acute with grafts. This is usually due to stenosis at the
venous anastomosis and this is the first place to focus on if surgical re-exploration is
planned. Whether thrombectomy is performed with catheters or at open operation, it is
vitally important to treat any stenosis or the graft will probably clot again. One exception
to this is the overdialysis or patients, or volume depletion secondary to fluid loss (diarrhoea
and/or vomiting). In this situation the thrombosis is consequent on low flow states and sim-
ple thrombectomy plus re-adjustment of dialysis dose will suffice.
AVFs are less likely to fail, although there is some evidence that surveillance and inter-
vention may prolong the life of established AVFs [8]. They degrade rapidly after thrombosis.
Arteriovenous fistulae need to be declotted rapidly, within 48–72 hours (unlike grafts) if
they are to be salvaged after failure. Central venous catheters tend to clot but can often be
resuscitated with endovascular techniques. There is no formal surveillance programme for
CVCs and this may in part reflect the fact that many of the CVCs are placed temporarily as
a bridge to more permanent access.
Techniques for the surveillance of grafts and AVFs centre on clinical assessment, flow
estimation and assessment of dialysis efficacy. It is not the purpose of this chapter to provide
an in-depth critique of various techniques. The variety of methods speaks for the fact that
none is perfect. It is up to individual units to select a method that suits the needs of their
patients and then to understand both its advantages and shortcomings.
Clinical surveillance revolves around thrill detection in the access and recognition of
significant abnormalities. Thrill detection can be performed by the patient, and patients are
told to report loss of thrill immediately. This is an indication for immediate review and con-
firmation of thrombosis/critical stenosis at the earliest opportunity. Other factors that are
detected clinically are infection (rigours, redness and pain over access), aneurysm formation
and vascular steal syndrome. Some stenoses may be detected by noting a change in the char-
acter of the thrill/turgidity of the AVF along its length.
While clinical surveillance may detect significant abnormalities, the most common
clinical problem encountered with access is thrombosis. Ideally the abnormality leading to
thrombosis (usually a stenosis) should be detected before it is clinically problematic. Two
strands of investigation have emerged with this aim in mind. One centres on intra-dialytic
measurements of flow or recirculation within the access. The other relies on external Duplex
scanning to detect abnormalities at an early stage.
The advantage of intra-dialytic measurement is it avoids extra visits to hospital. Flow can
be measured in the fistula by ultrasound dilution techniques. These have been shown to be
both reproducible and sensitive. Significant deterioration in flow rates triggers investiga-
tion by ultrasound and interventions as appropriate. Similarly, rising rates of re-circulation
within the access (i.e. dialysing the blood within the access, rather than the patient) should
trigger investigation by ultrasound and then either angioplasty or surgery as appropriate.
Duplex ultrasound surveillance works in the same way as aneurysm or lower limb bypass
graft surveillance. The patient is scanned at regular intervals, looking for evidence of stenosis
or change in volume flow. There are no clear cut-off figures at which intervention becomes
essential. Most clinicians will be concerned with a fall in flow rate of 20% or more or an
absolute flow of less than 600 ml min–1. Duplex will often demonstrate the site of abnormal-
ity, allowing a directed choice of therapy. Once again such choices are best determined by
multidisciplinary teams.
Managing failing access

Failing AVF
Reducing flow or dialysis efficacy may be related to stenosis formation, often at the anas-
tomosis, sometimes at the site of fistula damage from needling. The treatment options lie
between angioplasty and surgery to revise any stenotic segments. The evidence suggests that
outcomes are broadly similar, but that angioplasty may need to be repeated more frequently.
Overall, costs of treatment are similar. For early failure due to anastomotic stenosis in wrist
fistula, surgical revision gives good results. However in brachio-cephalic fistula and mature
fistula with scarring from dialysis, mobilisation of vein may be more difficult. In this situa-
tion, angioplasty is probably to be preferred.
Some fistulae exhibit rapid aneurysmal enlargement. If this is throughout the fistula,
then ligation may be the preferred option. Often the aneurysm is associated with a down-
stream stenosis. Aneurysm growth may be arrested by dilating the stenosis.
Failing arteriovenous graft

Grafts usually work well initially unless there is some technical problem with their place-
ment. Subsequent reduction in flow is most commonly due to stenosis at the venous anasto-
mosis. Angioplasty is effective at prolonging graft survival and American studies have shown
that with repeated interventions the longevity of grafts can approach that of AVF.
When angioplasty is unsuccessful, or if the graft has been clotted for a few days and
cannot be lysed successfully, surgical revision can be effective. Once declotted, the graft
can be extended with a short extra segment, to an undamaged vein distal to the original
anastomosis.
Graft infection can be very difficult to eradicate. The simplest approach is graft removal,
but this often needs to be followed by temporary CVC placement risking catheter infection.
Localised infection at needling sites and small abscesses can be managed by antibiotics and
local graft excision with adjacent skip grafting. This is not always successful but may salvage
a particularly valuable access.
Failing central venous catheter

The commonest cause of failure is clotting or obstruction to flow by the development of a
sheath of fibrin around the catheter tip. Clotted catheters can be re-opened by a passage of
guidewires and small wire brushes. Fibrin sheaths can be difficult to remove. One technique
is to pass a Dormier basket from the groin, close it around the catheter and strip the fibrin
sheath away. An alternative is to change the CVC over a wire, which sometimes improves
flow.
Central venous catheters are the most susceptible form of access to infection. The best
management is to remove the CVC and treat the patients with antibiotics for a few days
until better. A new catheter can then be placed, ideally in a new location or at the same site
if there is no overt inflammation. Patients can develop severe complications from catheter
sepsis including infective endocarditis, arthritis and discitis, so early aggressive antibiotic
treatment of infection is important to minimise the risk of complications.
Vascular steal syndrome

This is a problem peculiar to fistulae and grafts. In the limb, a proportion of blood flow is
directed through the access. It is not uncommon to see reversed flow in the artery beyond
the anastomosis. This is not usually a problem at the wrist as the ulnar or interosseous arter-
ies provide adequate flow to the hand. At the elbow, reversed flow in the brachial artery may
be asymptomatic, providing collateral vessels at the elbow to maintain adequate flow to the
hand.
If the collateral flow is inadequate, the patient may complain of coolness, numbness
and tingling, or overt pain. If this occurs rapidly after access placement (i.e. on waking or
the anaesthetic block wearing off) this is a clinical emergency. Once neurological deficit is
present, it is difficult to reverse it, even with access ligation. This aggressive version of steal is
called ‘ischemic monomelic neuropathy” and usually leaves a permanent deficit.
True vascular steal usually presents within a few days, but can present later in AVFs if
they enlarge or if there is progressive arterial narrowing. Symptoms are not always present at
rest and may sometimes only be present after a period of dialysis.
Diagnosis can be clinical by arm elevation, demonstrating blanching with colour being
restored by temporary access occlusion. Digital : brachial pressure (DBI) measurements can
help and a DBI of less than 0.6 is strongly associated with steal syndrome.
The management of vascular steal is to restore flow to the distal part of the limb. This can
be achieved by access ligation, but this leaves the patient without their lifeline. Attempts to
place access in another limb risks re-creating the problem elsewhere. For this reason, other
attempts are often made to improve blood flow to the hand. This can be achieved by band-
ing the access and reducing flow, but banding is difficult to get exactly right and symptoms
may persist or the access may clot. Extending the access distally or proximally may increase
resistance and allow improved flow distally. The distal revascularisation and interval ligation
(DRIL) procedure ligates the artery just distal to the access. The artery is then bypassed from
at least 8 cm proximal to the access, creating two flow channels, one to the access and one
to the distal limb. Occasionally simple distal ligation may suffice, by abolishing the reversed
arterial flow beyond the access [9].
Conclusion
Access surgery requires not only good vascular specialist skills, but also requires good ser-
vice organisation. A well organised service will correctly identify those in need of access,
investigate them appropriately to plan the best procedure for the patients and will then
undertake surgery in a timely manner.
Once surgery is completed, it is necessary to provide quality assurance by both assessing

access adequacy and providing surveillance where needed. A well-organised team will need
to review scans and flow measurements regularly and decide on appropriate management of
those accesses at risk of failure. A service will need to be provided to deal with those accesses
that fail in a timely manner to ensure continuing provision of dialysis for the patient.
References in autogenous radiocephalic wrist access

for haemodialysis. Nephrol Dial Transplant
1. Renal NSF part 1. http://www.dh.gov.uk/ 2003; 18: 378–83.
en/Publicationsandstatistics/Publications/ 6. Tessitore N, Mansueto G, Lipari G et al.
PublicationsPolicyAndGuidance/ Endovascualr versus surgical preemptive
DH_4070359 (accessed 19 September 2010). repair of forearm arteriovenous fistula
2. Winearls C, Fluck R, Mitchell D, Gibbons juxta-anastomotic stenosis: analysis of data
C et al. The organisation and delivery collected prospectively from 1999 to 2004.
of the haemodialysis access service for Clin J Am Soc Nephrol 2006; 1: 448–54.
maintenance haemodialysis patients. 7. Robbin ML, Oser RF, Lee JY et al.
Joint Publication Vascular Society/ Randomized comparison of ultrasound
Renal Association British Society for surveillance and clinical monitoring on
Interventional Radiology, August 2006. arteriovenous graft outcomes. Kidney Int
http://www.vascularsociety.org.uk/Docs. 2006; 69: 730–5.
3. Wong V, Ward R, Taylor J et al. Factors 8. Tessitore N, Lipari G, Poli A et al. Can blood
associated with early failure of arteriovenous flow surveillance and pre-emptive repair of
fistulae for haemodialysis access. Eur J Vasc subclinical stenosis prolong the useful life
Endovasc Surg 1996; 12: 207–13. of arteriovenous fistulae? A randomized
4. Zeebregts CJ, Kirsch WM, van den Dungen controlled study. Nephrol Dial Transplant
JJ, Zhu YH, van Schilfgaarde R. Five years’ 2004; 19: 2325–33.
world experience with nonpenetrating clips 9. Bourquelot P, Guadric J, van Laere O.
for vascular anastomoses. Am J Surg 2004; Surgical techniques for steal treatment.
187: 751–60. In: Tordoir J, ed. Vascular Access. Turin: Publ
5. Tordoir JH, Rooyens P, Dammers R et al. Edizioni Minerva Medica, 2009.
Prospective evaluation of failure modes
Chapter
25
Basic outline of solid organ
transplantation
Mathew Jacob, Jeremy French and Derek Manas
Key points
• Solid organ transplantation is now commonplace and is the standard of care for
patients with end-stage organ failure
• Indications have changed over time and there are few absolute contraindications
• Equity of access to transplant waiting lists is paramount and selecting the correct
recipient and donor pair will optimise the outcome
• The surgical techniques for all organ transplantation are now well established and
standardized and as a result there has been a year-on-year improvement in 1-year
survival. Most recipeints die because of co-morbidity or poor organ function
• Live donor transplantation for both kidney and liver recipients has become an
extremely important source of donor organs
• Complications are general to surgical patients but indeed each organ has its own
specific risks
• Immunosuppression has advanced hugely over the past 10 years and, as a result, the
overall attrition rate due to acute rejection has reduced considerably
• The biggest problem facing transplantation today is the donor shortage. In 2008, the
organ donor taskforce set up by the Minister of Health has set out a plan to increase
donation by 50%
• Until this happens, transplantation will always have to deal with the ethical dilemmas
of allocation, utilization and fairness
Introduction
Solid organ (liver, pancreas and kidney) transplantation is an important treatment modality
for end-stage organ failure. Indeed if a vital organ such as the liver fails, transplantation is
the only management option currently available.
Organ transplantation increases life expectancy and quality of life (for the recipient and
their family), but is not without risk. Since the pioneering days of solid organ transplantion
(kidney 1950, liver 1963) there have been many advances, both surgical and medical, result-
ing in considerable reduction in overall risk.
While these advances clearly are beneficial in terms of graft and patient survival, these
successes have lowered the threshold for acceptance of patients onto transplant waiting lists
worldwide, thus significantly contributing to the observed increased demand.
Chapter 25: Solid organ transplantation 299
Table 25.1 Disease-specific indications for liver transplantation
Primary recipient disease

Cirrhosis Secondary sclerosing cholangitis
• Primary biliary cirrhosis Alpha-1-antitrypsin deficiency
• Secondary biliary cirrhosis Budd–Chiari syndrome
• Cryptogenic Wilson’s disease
• Alcoholic Biliary atresia
Non-alcoholic fatty liver disease Other congenital biliary abnormalities
Chronic active hepatitis (autoimmune) Acute/subacute fulminant hepatic failure (FHF)
Chronic viral hepatitis B Primary hepatocellular carcinoma in cirrhotic liver
Chronic viral hepatitis C Primary hepatic malignancy
Congenital hepatic fibrosis Inborn errors of metabolism not in chronic liver
failure (CLF) group
Primary sclerosing cholangitis
The fact that the demand for donor organs outstrips supply creates ethical and medical
considerations specific to transplantation such as: selection and de-selection criteria; wait-
ing list prioritisation; national organ sharing schemes; development of organ donation and
retrieval methods; and the concept of transplant benefit.
This chapter will explore some of these issues as well as giving an overview of the medical
and surgical technical aspects of transplantation.
Indications
Liver transplantation (LT)
Following the death of a young woman with liver failure, a colloquium was set up in the UK
in 1999 to establish guidelines for the selection of patients for LT. It was agreed that livers
donated for transplantation should be considered a national resource. Patients should be
considered for transplantation if they had an anticipated length of life (in the absence of
transplantation) of less than one year or an unacceptable quality of life. It was also agreed
that patients should be accepted for transplantation only if they had an estimated probabil-
ity of being alive 5 years after transplantation of at least 50% with a quality of life acceptable
to the patient. The British Society of Gastroenterology has published clinical guidelines on
the indications for referral and assessment in adult liver transplantation [1]. The common
indications for liver transplantation are shown in Table 25.1.
Special considerations
Alcohol-induced liver disease may be associated with significant damage to cardiovascular and
neurological systems, as well as the risk of patients reverting back to alcohol abuse, resulting in
them not complying with medication or follow-up schedules and thus damaging the new liver.
A multidisciplinary approach is required to select those patients who are likely to comply; all
potential recipients, once accepted onto the waiting list, have to enter into a written contract
with the transplanting centre not to return to alcohol consumption after transplantation.
Illegal drug use is not a contraindication to transplant if the patient will comply with the
required schedules. However, continued intravenous drug use is considered a contraindication.
Age in itself is not a contraindication, although the survival rate after transplant of the
over 65s is significantly worse than that of younger patients.
Self-inflicted conditions such as overdose of paracetamol would only be contraindicated
if there were good reasons to believe that the patient would, despite appropriate support,
return to their pre-morbid lifestyle that would lead to liver failure or result in a quality of life
unacceptable to the patient.
Co-morbid medical or psychiatric conditions are relevant if they affect the patient’s qual-
ity of life or prospect for survival post transplant. Patients in whom early graft damage from
recurrent disease can be anticipated such as recurrent hepatitis C virus (HCV) and hepatitis
B virus (HBV) infections should only be transplanted as part of an agreed protocol of treat-
ment. There are well-developed protocols now for prevention of recurrence. With the advent
of effective treatment, those co-infected with human immunodeficiency virus (HIV) may be
suitable candidates for transplantation.
Re-grafts will need special consideration dependent on the circumstances that gave rise
to the need for the re-graft. This is because the results after early re-graft are poor and of only
limited benefit.
Where potential liver allograft recipients have suffered from previous extra-hepatic
malignancy, the decision to proceed for liver transplantation depends on the probability of
malignancy recurring following liver transplantation. Some immunosuppressive agents may
encourage the growth of malignancy. In patients with primary hepatic malignancy (HCC),
there are agreed criteria that predict a high probability of tumour persistence after trans-
plantation: these include number and size of lesions. More than three liver tumours with a
maximum diameter of 5 cm indicates that HCC is likely to persist following liver transplant-
ation. However, these criteria are under regular review.
Currently in the UK, patients with cholangiocarcinoma are not appropriate candidates
for transplantation.
It has also been agreed that if the condition of patients awaiting a liver transplantation
deteriorates to the extent that the probability of a 5-year survival may fall below 50%, they
will be removed from the waiting list but only after full discussion with them. Such patients –
although in greatest need – are at greatest risk of not benefiting after transplantation.
Kidney transplantation
With the tremendous improvements in transplant management most patients with kidney
failure can be considered for transplantation. Diseases that may be indications for renal
transplantation are listed below:
Glomerulonephritis
1. Idiopathic and post-infectious crescentic
2. Membranous
3. Mesangiocapillary (Type I)
4. Mesangiocapillary (Type II) (dense-deposit disease)
5. IgA nephropathy
6. Antiglomerular basement membrane
7. Focal glomerulosclerosis
8. Henoch–Schönlein
Chronic pyelonephritis (reflux nephropathy)
Hereditary
1. Polycystic kidneys
2. Nephronophthisis (medullary cystic disease)
3. Nephritis (including Alport’s syndrome)
4. Tuberous sclerosis
Metabolic
1. Diabetes mellitus
2. Hyperoxaluria
3. Cystinosis
4. Fabry’s disease
5. Amyloid
6. Gout
7. Porphyria
Obstructive nephropathy
Toxic
1. Analgesic nephropathy
2. Opiate abuse
Multisystem diseases
1. Systemic lupus erythematosus
2. Vasculitis
3. Progressive systemic sclerosis
Haemolytic uraemic syndrome
Tumours
1. Wilms’ tumour
2. Renal cell carcinoma
3. Incidental carcinoma
4. Myeloma
Congenital
1. Hypoplasia
2. Horseshoe kidney
Irreversible acute renal failure
1. Cortical necrosis
2. Acute tubular necrosis
Trauma
All patients between the ages of 2 and 70, who require dialysis or expect to require dialysis
within the next 12 months, will be considered for transplantation. It is important to satis-
factorily resolve other co-morbidities to increase the safety of the transplant. Patients must
be evaluated early to allow them to consider their options for renal replacement therapy.
This is particularly valuable since living donor kidney transplantation can be considered and
timed appropriately to serve as renal replacement therapy, obviating the need for dialysis
and access surgery. Live donation has increased significantly over the past 5 years with the
advent of the laparoscopic donor operation. As a result, live donors now contribute up to
50% of the kidneys for transplantation in most large programmes in the UK.
Most patients are listed for a cadaver kidney when their creatinine clearance (Clcr),
calculated by the Cockcroft–Gault formula, is less than 30 ml min–1. The Cockcroft–Gault
formula for calculation of the Clcr is now considered to be superior to actual measured cre-
atinine clearance, as determined by 24-hour urine collection, due to inherent inaccuracies
and collection difficulties. The creatine clearance formula is as follows:
Clcr (ml min–1) = ((140 – age)(weight in kg))/(creatinine (mg dl–1) ×72)
For women, the result is multiplied by 0.85.
Although all causes for kidney failure can be considered for transplantation, some causes
of kidney failure, such as certain types of glomerulonephritis, may occasionally recur in the
new transplant. In most cases, transplantation is worthwhile since recurrence is usually very
slow to develop. These risks are discussed with patients on a case-by-case basis. Patients with
primary oxalosis require combined kidney-liver transplantation since without metabolic cor-
rection of oxalosis with liver transplantation, recurrent kidney disease would be very rapid.
Diseases that may recur in renal transplants are:
• diabetes mellitus;
• systemic lupus erythematosis;
• IgA nephropathy;
• focal segmental glomerulosclerosis;
• membranous glomerulonephritis;
• membranoproliferative glomerulonephritis;
• amyloidosis;
• cystinosis.
Contraindications for kidney transplantation

There are certain absolute contraindications to renal transplantation:
1. disseminated or untreated cancer;
2. severe psychiatric disease;
3. unresolvable psychosocial problems;
4. persistent substance abuse;
5. severe mental retardation;
6. un-reconstructable coronary artery disease or refractory congestive heart failure.
Relative contraindications:
• Treated malignancy. The cancer-free interval required will vary from 2 to 5 years,
depending on the stage and type of cancer.
• Substance abuse history. Patients must be involved in drug-free rehabilitation. This
includes negative random toxicology screens.
• Chronic liver disease. Patients with chronic hepatitis B or C or persistently abnormal
liver function testing must be seen by a hepatologist prior to consideration.
• Cardiac disease. All patients over the age of 55 or those with a history of diabetes,
hypertension, or tobacco abuse must have dobutamine stress echocardiography, or
exercise or pharmacologic stress cardiac scintigraphy. Any patient with a history of a
positive stress test or history of congestive heart failure must have cardiology evaluation
prior to consideration.
• Structural genito-urinary abnormality or recurrent urinary tract infection. Urologic

consultation is required prior to consideration.
• Past psychosocial abnormality. Social work or psychiatry evaluation, as appropriate.
• Aorto-iliac disease. Patients with abnormal femoral pulses or disabling claudication,
rest pain or gangrene will require evaluation by a vascular surgeon prior to
consideration. Patients with significant aorto-iliac occlusive disease may require
angioplasty or aorto-iliac grafting prior to transplantation.
Special consideration needs to be given to:
• morbid obesity;
• antibody status;
• re-transplantation;
• HIV-positive recipients.
Indications for simultaneous pancreas-kidney (SPK) transplantation patients with
insulin-dependent (type 1, juvenile diabetes) diabetes who have end-stage renal disease
(ESRD) and require dialysis or expect to require dialysis in the next 12 months may be
considered for SPK transplantation. Special care is taken to exclude recipients with type
2 diabetes. Candidates with a strong family history, or late age or gestational onset, have a
C-peptide level determined after glucose loading. Only those individuals with C-peptide
levels of 0.2 ng ml–1 are further considered for transplantation. In addition, evidence
of at least one type of progressive secondary diabetic complication including: diabetic
retinopathy, diabetic neuropathy, diabetic gastroparesis and accelerated atherosclerosis
should be present.
Indications for pancreas transplantation (solitary pancreas transplant, pancreas transplant

alone, pancreas after kidney)
Patients with insulin-dependent (type 1, juvenile diabetes) diabetes may be candidates
for pancreas transplantation if they have secondary diabetic complications that are pro-
gressive despite the best medical management. This includes patients for whom the
indication is brittle diabetes and hypoglycaemic unawareness with evidence of frequent
hypoglycaemic events, despite an attempt at optimal medical management. Patients
with brittle diabetes as the primary indication should have evidence of impairment of
employability, hypoglycaemic-induced accidents involving themselves or small children
in their care. Usually there is evidence of frequent emergency care for hypoglycaemia
or diabetic ketoacidosis. In some cases these patients will have received a prior kidney
transplant, usually from a living donor (living donor kidney transplant alone, LDKTA).
Absolute and relative contraindications for pancreas transplantation are similar to those for
kidney transplantation. Special consideration needs to be given to the cardiovascular system
because of the high incidence of asymptomatic coronary artery disease in this population.
All type 1 diabetic patients require dobutamine stress echocardiography.
Allocation and ethics of organs for transplantation

There is no doubt that there is a spectrum of ethical dilemmas within transplantation. These
range from more straightforward issues, such as the allocation of scarce resources, to com-
plex issues such as financial reward for organ donation and xenotransplantation.
The system of allocation of organs for transplantation in the UK varies with the individ-
ual organ. Factors common to different organs, which are important in allocation, are ABO
blood group compatibility (all organs) and the comparative sizes of donors and recipients
(liver, heart, heart/lung and lung). Other important considerations when allocating specific
organs include: tissue matching (histocompatibility) in kidney transplantation; the MELD
(Model for End-stage Liver Disease) score for liver transplantation; the quality of the donor
organ and how appropriately it matches a particular recipient, especially as it relates to mar-
ginal donors and recipients; and the concept of transplant benefit, currently being applied to
lung transplantation in the USA.
All patients who are waiting for transplants in the UK are registered on the National
Transplant Database held by UK Transplant – now part of an organisation linked with the
National Blood Service called National Blood and Transplant (NHSBT).
Currently UK Transplant run organ-specific national allocation schemes with an over-
arching principle of ensuring patients are treated equally. Donated organs are allocated in
a fair and unbiased way, based on the patient’s need and the importance of achieving the
closest possible match between donor and recipient. Kidneys are allocated according to a
national waiting list based on a weighted scoring system, which includes waiting time, time
on dialysis, sensitization levels and tissue match. Liver grafts are allocated to the centre,
which prioritizes locally based on the MELD/UKELD score (bilirubin, INR, creatinine and
serum sodium) equating to how ‘sick’ the potential recipient is. This has been validated in
the USA to predict survival up to 3 months post transplant. Currently no potential recipient
can be registered for a liver transplant unless they meet the minimum listing criteria, which
is a UKELD score of 49 [2]. A national allocation system for liver transplantation operates
for patients with acute liver failure deemed to have less than 72 hours to live. These patients
are categorized as ‘super-urgent’. Pancreas grafts are currently allocated to the retrieving
centre unless there is a potential recipient nationally who was previously sensitized but has
been shown to be suitable for a particular graft. These are patients usually those who have
transplanted previously and are awaiting a ‘window of opportunity’.
Some patients have a greater clinical need, resulting in others waiting longer. Donation
rates are greater in some ethnic groups, while in other ethnic groups the need for transplan-
tation is greater. Utilitarian principles therefore compete against duty-based ones.
Organ donation and transplantation are covered by the Human Tissue Act 2004 in
England, Wales and Northern Ireland and by the Human Tissue (Scotland) Act 2006.
Consent, or authorisation in Scotland, is the fundamental principle of both acts and is
required before organs can be removed from the deceased, stored and used. Consent is also
required from live patients offering organs, but is covered by common law.
The number of people needing organ transplants in the UK is far greater than the number
of donor organs available (Figure 25.1). In the financial year 2007–8 there were 2385 organs
transplanted from 809 deceased donors with a further 839 live donor transplants, but there
were 7655 patients on the active waiting list. This list grows at 8% per year, with approxi-
mately 1000 potential recipients dying each year while waiting or becoming too ill for a
transplant.
This means there has to be a system in place to ensure that patients are treated equally
and that donated organs are allocated in a fair and unbiased way, based on the patient’s
need and the importance of achieving the closest possible match between donor and recipi-
ent. The underlying ethical principles are straightforward in that organs should be allo-
cated irrespective of age, gender, race, religion or social standing. But the reality is more
8000
7655
7000 7219
6698
6000 6142
5604 5654 5673
5532
5000 5345 5354
Donors
Transplants
Number
4000 Transplant list
3000
2360 2428 2311 2388 2396 2385 2381
2247 2241 2196
2000
1000 738 777 773 745 777 770 751 764 793 809
0
1998-1999 1999-2000 2000-2001 2001-2002 2002-2003 2003-2004 2004-2005 2005-2006 2006-2007 2007-2008
Year
Figure 25.1 The number of deceased donors and transplants in the UK, 1 April 1998–31 March 2008, and patients
on the active transplant list.
complex because organs are a scarce resource and not every individual who needs an organ
will receive one.
Many of the ethical issues that require consideration in decisions to offer transplantation
to one patient in preference to another are shared in common, irrespective of the specific
organ or tissue to be transplanted. Factors that must be taken into account in organ alloca-
tion include:
1. Selection of the sickest patient: an offer of an organ to the patient most likely to die
without it might appear the most reasonable basis for organ allocation. However, this
method may also select the poorest outcome. This is not necessarily the ‘best use’ of a
limited resource.
2. Selection of the patient most likely to benefit based on medical or other criteria: if the
major emphasis in organ allocation is placed on guaranteeing that the greatest number
of transplanted organs are accepted and survive for the longest time, a preference
should be for the best possible tissue match in the patient with the best outlook.
Whilst this approach appears acceptable in isolation, it conflicts with a number of
other criteria. This would disadvantage any potential recipient with advanced disease
and result in an impaired chance of success. The best way of applying this criterion
is, having identified patients with similar priority for allocation on other grounds,
further choice might favour that case most likely to be successful.
3. Selection of the patient on the waiting list for the longest period: the length of a prior
waiting period appears fair. This criterion has identifiability and defensibility. Against
adopting this as the sole criteria is the fact that if a patient has survived for a long
period after meeting the requirements for entry to a waiting list this might indicate
that he or she was in better condition than others on that list. The question might then
be whether his or her ‘need’ was less.
4. All patients on the waiting list should have an equal chance of selection: it has the
advantage of being seen to be free of any favouritism. However, its application is
impractical. The method could only have a place in rare situations where several
possible recipients are judged to have equal priority on medical and other grounds
(this sometimes happens in the case of kidney transplantation).
5. Selection of patients on the basis of their importance for the well-being of others: is
it appropriate for a patient with a young family, dependent upon him or her for
support, or an individual with the capacity to make a unique contribution to his or
her community, to be accorded priority. This criterion raises questions about the
manner in which selection attributes are to be quantified and of who is entitled to do
so. During the early years of kidney transplantation programmes in the USA this was
considered and subsequently rejected.
6. Preference or not in selection to patients who have previously had one or more
transplants: patients who have already been transplanted but who have had the
misfortune of a failed graft, and so received no benefit, might be seen to have a claim
for priority for another try or conversely – ‘have had their chance’. Here the type of
organ graft is of importance. For example, recipients with failing liver or heart grafts
facing imminent death may achieve priority for this reason. Renal recipients may
return to dialysis following graft failure, but the basis of loss of the first graft may
persist (for example, high antibody titres) and remain transplantable at a later date.
7. Capacity of the patient to pay: equal access to medical care for all, irrespective of
capacity to pay, is a basic principle of the NHS.
8. What about a potential recipient’s lifestyle in selection for transplant allocation? There
are many who would argue that self-inflicted illnesses such as alcoholism and drug
abuse, necessitating liver transplantation due to alcoholic cirrhosis or HCV cirrhosis
should lessen a patient’s eligibility to be allocated a donor organ. The concern revolves
around the extent to which recurrence of alcoholism or HCV in the new graft may
compromise a successful outcome to transplantation. Any case for exclusion from
organ transplantation because of a self-inflicted illness involves non-medical ‘social’
judgements. Whenever the issue of possible exclusion of patients with a particular
lifestyle from access to any form of treatment that remains available to others is
considered, it is essential that attention be given to the development of processes that
will ensure adequate representation of the views and needs of marginalised groups
and this includes HIV co-infection in haemophiliac patients as well.
9. Can we exclude patients on the basis of anticipated lack of compliance: apart from
recidivism leading to a recurrence of the disease that required treatment originally,
grounds for exclusion might arise if there was considerable likelihood that a patient
would not be prepared to participate in essential post-transplant treatment, for
example the use of immunosuppressive agents. The reliability of any prediction
leading to exclusion creates a substantial ethical issue in most transplant units.
In Transplantation Ethics, Robert Veatch outlines an interesting way of assessing need [3].
He calls it the ‘over-a-lifetime perspective’. This approach takes into consideration a person’s
entire life when determining who is worst off. A 17-year-old and an 80-year-old both dying
of liver failure are equally badly off, but this perspective allows that the person who has had
63 more years of life is better off, so the 17 year old is neediest. Veatch writes, ‘from this
over-a-lifetime perspective, justice requires that we target organs for these younger persons
who are so poorly off that they will not make it to old age without being given special
priority – the younger the age of the person, the higher the claim.’ This furthers the goal of
utilising organs to their maximum potential. Presently, our system of allocation gives pri-
ority to those who are the sickest or most in need of a transplant. But sometimes those who
are the sickest and in the most immediate need will not receive the same benefit from the
transplant as someone whose medical condition is currently more stable. They may be so
sick that they have a higher chance of dying regardless of treatment. There is a moral obli-
gation, due to the scarcity of organs, to maximise the potential longevity of donated organs
and place them where they are most likely to do the most good (bring the most health) over
the longest period of time.
Donors and the donor procedure

The widening gap between organ demand and supply has resulted in the relaxation of the
criteria for organ donation as the clinician must weigh up the risk of transplanting the organ
against dying on the waiting list. There are however, contraindications, all of which are rela-
tive (Table 25.2). It is highly unlikely however that an individual with an active extra-cranial
malignancy would be considered as a donor.
It is important to note that organs used in transplantation can come from very different
clinical scenarios (Table 25.3).
Death confirmed by brainstem testing

The majority of donors in the UK come from heart-beating donors. Death is defined as irre-
versible loss of capacity for consciousness and the irresversible loss of the capacity to breath.
Before organ procurment can commence, death by brainstem testing needs to be performed
according to the UK code, which involves three steps.
1. Preconditions – comatose patient with irreversible injury, on a ventilator (>6–24
hours after last intervention aimed at reversing injury), with an identified underlying
cause for coma.
2. Exclusions – no drugs, alcohol, neuromuscular blocking agents or hypothermia.
3. Clinical testing – these tests (two sets of tests at least 2–3 hours apart performed by
two senior medical personnel) demonstrate absent brainstem relexes and total apnoea
(Table 25.4).
All the major religions of the UK support the principles of organ donation and transplan-
tation. However, within each religion there are different schools of thought, which mean that
views may differ. All the major religions accept that organ donation is an individual choice.
The donor operation

The principles are as follows:
• Preparation with antibiotics and neuromuscular blockade (to prevent spinal reflexes).
• Exposure is achieved through a midline laparotomy and sternotomy with
pericardotomy. The colon and small bowel are reflected superiorly to expose the
inferior mesenteric vein (IMV) and the aortic bifurcation. Hepatic vasculature is
dissected to their origin (alternative arterial vasculature is identified), portal dissection
and division of the bile duct is performed. The thoracic organs are then dissected.
Table 25.2 Organ donation (relative) contraindications
• Human immunodeficiency virus (HIV)/active hepatitis B infection

• Extra-cranial malignancy (current) or cranial
• Glioblastoma/medulloblastoma
• Severe systemic sepsis
• Disease of unknown aetiology
Table 25.3 Donor categories
Category Description
Non-heart beating donors (Maastricht classification)
I Brought in dead Uncontrolled
II Unsuccessful resuscitation Uncontrolled
III Awaiting cardiac arrest Controlled
IV Cardiac arrest after brain-stem Controlled
death
V Cardiac arrest in a hospital Uncontrolled
inpatient
Heart beating donors Usually patients in intensive care units having sustained irreversible
(brain stem dead) brain damage (e.g. intracranial haemorrhage, cerebrovascular accident
or head injury)
Live donors Increasingly common in kidney and liver donation
Table 25.4 Brainstem death tests
• Absent brainstem relexes

• No pupillary response to light
• Absent corneal reflexes
• Absent vestibulo-ocular reflex
• No motor response to adequate stimuli in cranial nerve distribution
• No gag reflex to bronchial stimulation by suction catheter
• Apnoea testing
• No attempt to breathe despite a PaC02 > 6.5 kPa. Hypoxia avoided by preoxygenation with 100%
oxygen for 10 minutes
• Perfusion cannula is placed in the aorta, the aorta is cross-clamped and 41 of cold
perfusate passed through the cannula. The inferior vena cava (IVC) is opened in the
chest and the abdomen filled with ice.
• Procurement of the liver is by done by dividing the arterial supply with an aortic patch,
dividing the IVC just above the renal veins, the superior vena cava (SVC) during the
cardiectomy and the portal vein preserving maximum length. Procurement of the
kidneys is done by dividing the ureters as long as possible (with adequate tissue around
them to preserve blood supply), preserving the renal arteries on an aortic patch and
the renal veins on an IVC patch. Important steps in procurement of the pancreas (in
combination with liver procurement) involve identification and division of the splenic
artery, preservation of the superior mesenteric artery and the length of the portal vein.
After removal of the liver the duodenum is stapled at the pylorus and 4th part, and
remaining attachments to the small bowel and transverse colon are divided. Following
pericardotomy and mobilisation of the great vessels, the SVC is ligated and divided, the
aorta cross-clamped and the heart perfused with a cardioplegic agent via an anterior
aortic puncture. After cardiac arrest the heart is emptied via incisions in the IVC and
left pulmonary vein. The heart is excised by dividing all the remaining attachments. The
procurement of the lungs follows as for the heart, except a perfusion cannula is placed
in the pulmonary artery, and the left heart is vented via the tip of the atrial appendage
as opposed to the pulmonary vein. Following manual venting the trachea is stapled and
divided above to keep the lungs inflated during transport.
• Completion of the procurement involves removing a portion of spleen and mesenteric
lymph nodes (tissue typing and cross-matching), iliac vessels (conduits) and removal of
blood and neat abdominal closure.
Operative technique
Liver transplantation
The operative technique can be classified based on the position of the graft in the receipient
(orthotopic if graft is placed in the usual position and heterotropic if placed elsewhere in
the body), whether all or part of the graft is transplanted (whole graft or partial graft trans-
plantation) and whether the native liver (or part of it) is retained (auxiliary graft). Partial
grafts can be a split liver, when the liver is divided for implantation into two recipients, or a
reduced-size graft, when only one part is retained to be transplanted.
Living donor transplantation and rarely domino transplantations are other techniques
used.
Standard liver transplantation technique

In this method recipient hepatectomy is performed en bloc with the retrohepatic cava.
Subsequently the whole graft is implanted by end-to-end anastomosis of the supra and infra
hepatic inferior vena cavae to the graft’s vena cava. Then the donor portal vein is anastmosed
end-to-end with the recipient’s portal vein, following which the graft is revascularised. The
arterial anastamosis is performed between the graft artery and the recipient hepatic artery. If
the recipient hepatic artery is not suitable then the graft artery can be anastamosed directly
to the recipient’s aorta or by using an arterial conduit. Lastly biliary anastamosis is per-
formed using a duct-to-duct technique or a Roux-en-Y biliary reconstruction. T-Tubes are
not routinely used.
The operative technique has evolved with time and more and more surgeons perform
liver transplantation using some, if not all, of the following modifications.
Veno-venous bypass
Cross clamping of the vena cava, which is a requirement for the classical method, often
results in haemodynamic instability and congestion of splanchnic circulation. Diverting the
blood from the portal vein and inferior vena cava to either the jugular or axillary vein on one
side (veno-venous bypass) overcomes this problem.
Preservation of vena cava

In this technique the recipient’s retrohepatic inferior vena cava is preserved during initial
hepatectomy by dissecting the liver off the vena cava. This maintains the blood flow in the
cava thus avoiding the need for systemic venous bypass. In addition a transient porto-caval
shunt may also be constructed to maintain splanchnic flow.
With the recipient cava intact the caval anastamosis can be performed with direct end-
to-end anastamosis of the donor cava to the unified stump of the recipient’s hepatic veins
or alternatively performing a lateral cavo-cavostomy on the anterior wall of the recipient
vena cava.
Kidney transplantation
The kidney transplant operation has been standardised over the last few decades. Unlike
liver transplantation, the transplanted kidney is placed in a heterotropic extraperitoneal
location, usually in the iliac fossa. A curvilinear incision in a lower quadrant of the abdomen
(Gibson’s incision) is made, with division of the muscles of the abdominal wall and dissec-
tion of the preperitoneal space to expose the iliac vessels and the bladder. The renal vein and
artery are anastamosed to the recipient iliac vein and artery, respectively. Then an uretero-
neocystostomy is created, with or without placement of a ureteric stent. If a stent is used it
is important to remove this in a few weeks to prevent complications of a non-removed stent
(e.g. haematuria, renal stones, infection).
The kidney may be placed on either side, depending on history of previous transplanta-
tion, surgeon preference and the side of the donor kidney.
Pancreas transplantation
The pancreas may be transplanted simultaneously with a kidney (SPK), sometimes following
a kidney transplant (PAK – pancreas after kidney) or as a pancreas alone (PTA).
The back table preparation of the pancreas is a crucial part of the procedure and this
can usually take 2 hours. Following careful ligation of all peri-pancreatic tissue to prevent
bleeding at re-perfusion, an iliac Y graft from the donor is anastamosed to the superior
mesenteric artery (SMA) and the splenic artery of the pancreas graft. The Y graft construc-
tion avoids the need for two separate arterial anastamoses betwen the donor and the recipi-
ent to vascularise the pancreas graft. Futher important preparation of the graft involves
merticulous attention to controlling the route of the small bowel mesentary, as well as pre-
paring the portal vein.
In SPK transplantation, the pancreas is implanted first due to the lower ischaemia toler-
ence of the pancreas. The pancreas is usually placed in an intraperitoneal position, although
extraperitoneal placement can also be done. The graft portal vein is commonly anastamosed
to the recipient lower inferior vena cava. Alternatively the venous drainage can be put into
the portal circulation (graft superior mesenteric vein), giving the theoretical benefit of
avoiding hyperinsulinaemia, which has been linked to atherogenesis. The Y graft is anasta-
mosed to the lower aorta or the common iliac artery (CIA). The management of exocrine
secretion is still a matter of considerable debate. The donor duodenum can be anastamosed
to a Roux-en-Y loop of recipient small bowel (enteric drainage) or alternatively this can be
anastamosed to the recipient urinary bladder (bladder drainage).
Postoperative complications
Organ transplantation is susceptible to all the recognised complications of any major surgi-
cal procedure (e.g. bleeding, infection, hernia). There are, however, issues that can arise that
are specific to organ transplanation. There are complications that are common to all trans-
plants and some specific to individual organs. The manifestation of each complication can
differ according to the organ involved. Complications are classified into early and late.
Liver
Early
1. Primary nonfunction – this can manifest as haemodynamic instability, hypoglycaemia,
elevated transaminases, coagulopathy, minimal bile output, encephalopathy, systemic
acidosis and renal failure. This is not compatible with life and most patients require
re-grafting (incidence: 1–3%). More commonly one sees a less dramatic version of this
scenario called initial poor function or delayed graft function. This is often related to
graft ischaemia and will improve.
2. Vascular thrombosis – arterial thrombosis can be early or late. This is more common
in the paediatric population and can manifest as rapid or slow deterioration of graft
function or as necrosis of bile ducts (incidence: adult liver transplant 2–4%; paediatric
liver transplant 8–10%).
3. Bile leak – occurs usually due to ischaemia of the donor duct or rarely an operative
technical problem (incidence: 20%).
4. Infection – this remains the most significant complication in liver transplantation and is
responsible for most of the early mortality from bacterial infections with resistant gram-
positive bacteria dominating in the first month. Multi-resistant bacterial and fungal
infections become a more prominent and life-threatening issue if infection persists.
5. Acute rejection – with the advent of the newer and more potent immunosuppressive
drugs this has become less of an issue in transplantation in general. It may present with
fever, abdominal pain and elevated liver enzymes. The diagnosis is confirmed by a liver
biopsy and most episodes are responsive to augmentation of immunosuppression with
high dose corticosteroids.
Late
1. Arterial stenosis – presentation is with slow deterioration of graft function.
2. Infection – late infections are usually due to opportunistic pathogens such as
cytomegalovirus, candida, aspergillosis, cryptococcus, legionella.
3. Chronic rejection – this is seen months or years after transplantation with poor
synthetic liver function and hyperbilirubinemia.
4. Biliary stricture – patients present with obstructive jaundice, usually due to an
ischaemic stricture.
5. Recurrent disease – recurrence of viral hepatitis is likely within a short time in infected
patients but this may be mild and in many cases will not result in graft loss.
6. Cancer – recurrence if the patient was transplanted for HCC (outcome for small
tumours: 75% 5-year survival). De Novo tumours such as skin cancers, lymphomas and
others may occur in up to 3% of transplant recipients. This is most often secondary to
immunosuppression.
7. Others – diabetes, hyperlipidaemia, hypertension and metabolic bone disease – these

are related to the immnosuppression agents.
Kidney
Early
1. Acute tubular necrosis (ATN) and delayed graft function – some degree of ATN occurs
in 5–30% of all heart-beating cadaveric donor transplantations. Delayed graft function
may be associated with a reduction in the 5-year graft survival by up to 10% in some
studies.
2. Primary non-function – the kidney never functions.
3. Arterial thrombosis – causes early postoperative oliguria or anuria. Immediate
re-exploration is the only chance for salvaging such a graft.
4. Venous thrombosis – can result from technical error or kinking or compression of the
renal vein.
5. Acute rejection – incidence varies but with newer immunosuppression and pre-emptive
treatment most centres report acute rejection rates of 10–20%. Diagnosis requires
biopsy and treatment usually involves steroid boluses.
6. Ureteral obstruction – could be due to blood clot in catheter, haematoma or oedema.
7. Urinary fistula – this occurs due to disruption of the ureteroneocystostomy or ureteral
necrosis. Fluid biochemistry showing urea content several folds higher than that of
serum is diagnostic.
8. Infection – 30–60% will suffer some type of infection during the first year. Conventional
bacterial infections occur during the first month. Infections can be confused with rejection.
Late
1. Renal artery stenosis – patients present with hypertension and diminished renal
function. This presentation can be confused with that of rejection. The aetiology of
renal artery stenosis is frequently technical. Most instances can be managed with
percutaneous transluminal angioplasty.
2. Ureteral obstruction – late presentation could be due to ureteral stenosis.
3. Lymphocele – manifests weeks or months postoperatively with swelling of the wound,
oedema of the scrotum or labia and lower extremity and urinary obstruction from
pressure on the collecting system or ureter. The treatment of choice is fenestration of
the cyst into the peritoneal cavity and external drainage should be avoided as this puts
the kidney at risk of infection.
4. Infection – the period between 30 and 180 days postoperative is the usual
time for opportunistic infections as this coincides with the period of maximal
immunosuppression. Viral infections are more important (e.g. cytomegalovirus).
Other pathogens include aspergillosis, blastomycosis, nocardiosis, toxoplasmosis,
cryptococcosis, candida and Pneumocystis carinii.
5. Hypergylcaemia – this is generally attributed to corticosteroid administration and
previously normoglycaemic patients may become diabetic.
6. Hyperparathyroidism – patients could suffer from tertiary hyperparathyroidism
with significant hypercalcaemia and elevated parathyroid hormone levels despite a
functioning graft. This is treated by total parathyroidectomy.
7. Cancers – in renal transplant recipients an incidence of 6% is reported for

de novo malignant neoplasms. This is related to the duration and degree of
immunosuppression rather than to any particular agent. More prevelant tumours are
skin cancers (squamous cell carcinomas), lymphomas, renal cancers, Kaposi’s sarcoma,
carcinoma of the vulva and uterine cervix.
8. Post-transplant lymphoproliferative disease (PTLD) – term used to cover the
spectrum of disease from benign hyperplasia to malignant lymphomas. Epstein–Barr
virus is implicated as the most important factor in PTLD.
9. Chronic allograft nephropathy – this complication is characterised by a progressive
decline in kidney function, which is not attributable to a specific cause. Chronic
changes to the kidney allograft are mediated by both immune and non-immune factors.
10. Recurrent disease – Glomerulonephritides (eg, mesangiocapillary glomerulonephritis
type 1, IgA nephropathy) are most likely to recur; however, loss of the kidney generally
occurs late, and, thus, these diseases are not contraindications to transplantation.
Similarly, patients with diabetes mellitus have poorer outcomes following
transplantation than do patients without diabetes; nearly all patients demonstrate
histological evidence of diabetic nephropathy within 4 years. Hence the treatment of
choice for diabetics with renal failure is combined kidney and pancreas transplantation.
Pancreas
Complications and issues specific to pancreas transplantation are discussed below.
Early
1. Vascular thrombosis – this is the most common non-immunological cause of graft loss.
2. Allograft pancreatitis – this occurs in 10–20% of all pancreas graft recipients. In its
most severe form it can result in graft necrosis and arterial thrombosis. This entity is
difficult to detect and can be confused with rejection and pancreatic fistula.
3. Pancreatic fistula – this is more common in enteric-drained than in bladder-drained grafts.
4. Rejection – hyperglycaemia is a late indicator of rejection as islet damage results
by the time such physiological evidence results. Rejection of a kidney and pancreas
transplanted simultaneously from the same donor often occurs at the same time. In such
patients careful monitoring of serum creatinine level is a sensitive indicator of rejection.
Late
1. Urological complications – haematuria, urethritis, recurrent urinary tract infections,
and bicarbonate loss are common in bladder-drained recipients. This can necessitate
enteric conversion if it does not respond to conservative treatments.
2. Autoimmune recurrence – the autoimmnune response to native islets can be
responsible for loss of transplanted pancreatic β-cells.
Graft rejection and immunosuppression

With the exception of identical twins the organ donor and recipient are genetically differ-
ent. Without medical manipulation a transplanted graft will be rejected within a number of
days. Understanding the rejection process is a prerequesite to understanding the principles
of immunospressive medication.
The rejection process

Organs are rejected as the recipient body recognises the graft as ‘foreign’. This recognition
is based on the fact that the graft major histocompatability complex (MHC) is different to
that of the host cell. The role of the MHC is to present antigens for other immunological
cells to screen. The antigens relevant to transplantation were first defined seriologically on
leucocytes and are therefore called the human leucocyte antigens (HLAs). Minor histocom-
patability systems (miHs) are of limited clinical importance in solid organ transplantation.
Major histocompatability class I proteins are found on most nucleated cells. They present
antigens and are recognised by the T-cell receptor (TCR) on T-cells bearing the CD8 protein.
CD8 T-cells cause lysis of the target (graft) cell. Major histocompatability class II proteins
are found on B lymphocytes and dendritic cells (antigen-presenting cells). They present anti-
gens and are recognised by the T-cell receptor on T-cells bearing the CD4 protein. CD4 cells
are referred to as ‘helper’ cells and have crucial specialised functions in the generation of the
immune response (and therefore graft rejection) such as cytotoxic T-cell generation, B-cell
maturation.
For immunological cells to be activated however, in addition to this MHC (with antigen)–
T-cell receptor (TCR) interaction, various co-stimulatory signals are needed. Many have
been detected (e.g. Il-2, the CD28-B7 and CD40ligand-CD40 family), which is important
when searching for immunosuppressive agents. It is noted that other cellular (natural killer
cells) and humoral (antibodies) mechanisms of rejecton exist.
Immunosuppressive agents
The main immunosuppressive groups and their mode of action are documented in
Table 25.5. In solid organ transplantation, many immunosuppressive protocols have been
used, but broadly speaking most protocols are based on the principles outlined below a
calcineurin inhibitor, an antimetabolite and a reducing dose of steroid. In selected cases
protein immunosuppressives are given. The calcineurin inhibitor is sometimes exchanged
for sirolimus at 3 month post-transplantation (Table 25.5).
Outcome
Outcomes have steadily improved due to better surgical techniques and more effective
immunosuppressive treatments and the development of transplant specialists and teams.
Transplants are now so successful in the UK that a year after surgery:
• 94% of kidneys in living donor transplants are still functioning well;
• 88% of kidneys from people who have died are still functioning well;
• 86% of liver transplants are still functioning well;
• 84% of heart transplants are still functioning well;
• 77% of lung transplants are still functioning well;
• 73% of heart/lung transplants are still functioning well.
Longer-term outcomes are similarly improving although most organs, with the exception of
liver grafts, suffer from ‘chronic fatigue’ otherwise known as ‘chronic rejection’. Most trans-
planted organs are still functioning at 5 and 10 years but the percentage attrition rate varies
from 10% at 5 years for livers through to 25% for lungs. A recent analysis based on 3673 adult
liver recipients for whom the 15-year patient survival rate was 58% (95% confidence interval
Table 25.5 Immunosuppressive agents
Side effects
Immunosuppressive (in addition to
agent group Examples Mode of action infection)
Drugs acting on Tacrolimus Forms a complex with Nephrotoxicity
immunophillins – Cyclosporin immunophillins and inhibits Neurotoxicity
calcineurin inhibitor calcineurin, which under normal
Diabetes
(CNI) circumstances induces the
transcription of interleukin-2. Marrow
It has been in use since
The drug also inhibits suppression
1983 and is one of the
most widely used group lymphokine production and
of immunosuppressive interleukin release, leading to
drugs a reduced function of effector
T-cells
Corticosteroids Prednisolone Multiple immunomodulatory Multiple including
effects e.g. reduces interleukin hyperglycaemia,
(IL) production and hence obesity,
activation of B- and T-cells, osteoporosis, skin
reduces immune cells protein fragility
transcription.
Antimetabolites Azothiaprine Inhibits synthesis pathway of Gastrointestinal
Mycophenolate B- and T-cells. (nausea, vomiting,
Mofetil diarrhoea,
ulcers, gastritis)
Bone marrow
suppression
mTOR inhibitor Sirolimus Inhibits response to IL-2 and Hyperlipidaemia
(mammalian target of hence blocks activation of B- Leukopenia
rapamycin) and T-cells.
Thrombocytopenia
Poor tissue healing
Protein Antithymocyte Contains cytotoxic/blocking
immunosuppressives globulin antibodies (anti-CD2, 3, 4, 8, 11a,
18, 25, 44, 52, HLA classes I and
II) to circulating lymphocytes
Anti-IL-2 (CD25) Blocks IL-2 and the resulting
receptor activation of B and T-cells.
Dacluzumab
Basiliximab
Anti-CD52 Causes apoptosis of circulating
Alemtuzumab lymphocytes, monocytes,
macrophages and natural killer
cells
54–62%) suggests that adult liver transplant recipients have an average life expectancy of
22 years. The average life expectancy of the equivalent UK adult population is 30 years, and
so on average 8 years of life are lost. Furthermore, female recipients lose fewer life-years than
male recipients, and younger recipients lose more life-years than older recipients.
References and Transplant. Selection of patients

for liver transplantation and allocation
1. Devlin J, O’Grady J. Indications for of donated livers in the UK. Gut 2008;
referral and assessment in adult liver 57: 252–7.
transplantation: a clinical guideline. British 3. Veatch RM. Transplantation Ethics.
Society of Gastroenterology. Gut 1999; 45 Washington, DC: Georgetown University
Suppl 6: VI1–VI22. Press 2000.
2. Neuberger J, Gimson A, Davies etal.
and Liver Advisory Group; UK Blood
Further reading 5. Talbot D, D’Alessandro A. Organ Donation

and Transplantation after Cardiac Death.
1. British Transplant Society. Standards for New York: Oxford University Press, 2009.
solid organ transplantation, 2003. http:// 6. Clavier P-A, Petrowsky H, OeOlveira ML,
www.bts.org.uk/Forms/standards%20 Graf R. Strategies for safer liver surgery and
document%20edition%202%20-%20final. partial liver transplantation. NEJM 2007;
pdf (accessed 1 March 2010). 356: 1545–59.
2. Forsythe JLR. Transplantation – A 7. Neuberger J, James O. Guidelines
Companion to Specialist Surgical Practice, for selection of patients for liver
3rd edn. Philadelphia, PA: Elsevier transplantation in the era of
Saunders, 2005. donor-organ shortage. Lancet 1999;
3. United network for organ sharing. 354: 206–14.
http://www.unos.org/ (accessed 1 March 8. Hirschfield GM, Gibbs P, Griffiths WJH.
2010). Adult liver transplantation: what non-
4. The Practice of Liver Transplantation. specialists need to know. BMJ 2009;
Oxford: W.B. Saunders, 1995. 338: b1670 1321–7.
Index
Note: Page numbers in italics – Tables
Page numbers in bold – Figures
AAA, see abdominal aortic above-knee amputation (AKA), ALI, see acute limb ischaemia
aneurysm 191 Ambulatory Venous Pressure
abdominal aortic aneurysm ABPI, see ankle : brachial (AVP), 210
(AAA), 86, 87 pressure index amputation, 183
ADAM study, 91 activated partial aetiology, 185
aneurysm repair, 91 thromboplastin time Chopart, 189
aneurysmal degeneration, 87 (aPTT), 225 epidemiology, 184
branched grafts, 97 acute limb ischaemia (ALI), level, 68
Cochrane review, 90 49, 73 lower limb, 191
CT, 32, 90 amputation, 54 pain, 193
curcumin, 97 aetiology, 49 pain cause, 192
doxycycline, 97 analgesia, 73 postoperative complications,
endoleak types, 94, 95 anticoagulation, 73 191
EVAR, 94, 95, 96, 97 and atherosclerotic disease, 54 preoperative assessment, 186
fenestrated grafts, 96 causes, 73 primary, 68
hybrid repairs, 97 classification, 51 prosthetics, 193
imaging, 90 compartment syndrome, 56 secondary, 68
inflammatory, 88 embolectomy, 74 surgical principles, 188
intra-arterial angiography, 90 endovascular treatments, 55 toe, 189
matrix metalloproteinase, 87 graft occlusion, 54 types, 186
mortality rate, 91 heparin, 52 upper limb, 191
MRI, 90 intra arterial thrombolysis, anaerobic threshold (AT), 274
open repair, 34, 92 54 anastomotic techniques, 77
operative repair, 88 investigations, 52 aneurysm
palpable, 88 irreversible, 51 aneurysmal degeneration, 87
patient selection, 91 mechanical thrombectomy, mycotic, 87, 241
periaortitis, 87 76 surgical management, 104,
plain radiographs, 89 palliative care, 55 see also pseudoaneurysm
prevalence, 88 revascularisation, 54 Aneurysm Detection and
risk factors, 87 rhabdomyolysis, 56 Management (ADAM), 91
rupture, 88 risk prone area, 49 aneurysmal tissue, 99
screening, 90 severity assessment, 51 angioplasty, 78
statins, 97 surgery vs thrombolysis, 75 aortic, 81
stent graft, 93 symptoms, 50 arterial dissection, 80
tobacco smoking, 87 threatened limb, 51, 54 arterial puncture site
treatment, 98 thrombolysis, 54, 75 haemorrhage, 79
UKSAT, 91 treatment, 51, 53 arterial rupture, 80
ultrasound imaging, 89 unsalvagable limbs, 76 complications, 80
vessel layers, 86 vascular imaging, 52 distal embolization, 80
abdominal vascular injuries, ADAM, see Aneurysm iliac artery, 81
179 Detection and procedure, 79
delayed presentation, 181 Management subintimal, 79
mesenteric vessel injuries, Adamkiewicz artery, 106 ankle brachial index (ABI), 37
180 Adson test, 148 ankle level amputation, 190
operative management, 179 adventitial fibroplasia, 130 ankle : brachial pressure index
retro-peritoneal haematoma, AKA, see above-knee (ABPI), 60
179 amputation measurement, 60
ABI, see ankle brachial index ALARA principle, 83 pole test, 61
318 Index
antiphospholipid syndrome, pulsatility in, 10 buttock claudication, 59

221–4 risk indicators, 10 bypass graft
anti-platelet agents arterio-venous malformations stenoses, 67
aspirin, 277 (AVMs), 21 veins, 66
clopidogrel, 277 aspiration thrombectomy, 55 Bypass versus Angioplasty in
anti-platelet therapy, 45 aspirin, 64, 277 Severe Ischaemia of the
clopidogrel vs. aspirin, 45 AT, see anaerobic threshold Leg (BASIL)
guidelines, 46 ATOS, see arterial TOS angioplasty first strategy, 73
aortic angioplasty, 81 auscultation, 60 surgery first strategy, 73
aortic dissection, 108 autonomic neuropathy, 69 survival rate, 73
acute, 110 AVF, see arteriovenous fistula
associated factors, 108 AVMs, see arterio-venous calcineurin inhibitor, 314
complications, 113 malformations CAD, see coronary artery
Debakey classification, 110, AVP, see Ambulatory Venous disease, cystic adventitial
111 Pressure disease
diagnosis, 110 axillo-bifemoral grafts, 77, 78 calcium channel blockers, 260
end-organ ischaemia, 109 calf claudication, 59
factors associated, 109 β-adrenergic blocking agents calf muscle pump, 29,
graft replacement, 112 Baker’s cyst, 3 see also chronic venous
intimal disruption, 109 BASIL, see Bypass versus insufficiency (CVI)
intramural haematoma, 109 Angioplasty in Severe cardiac failure, 283
luman creation, 108 Ischaemia of the Leg contractility, 283
management, 111 β-blockers, see β-adrenergic function maximization, 283
presentation, 110 blocking agents output estimation methods,
Stanford classification, 110, BD, see Bechet’s disease 283
111 Bechet’s disease (BD), 266 oxygen delivery, 282
symptoms, 110 investigation, 266 vasodialation, 283
tearing, 109 management, 267 Cardiac Risk Index, 273
type A, 112 presentation, 266 cardiopulmonary exercise
type B chronic, 113 below knee stenting, 82 testing (CPX), 102, 274
type B complicated, 113 below-knee amputation (BKA), cardiovascular events
type B uncomplicated, 112 190 aspirin, 64
aortic stent graft infection, β-Blockers, 278 clopidogrel, 64
236 bilateral carotid surgery life style modification, 64
aorto-iliac bypass, 77 preoperative assessment, 16, ramapril, 64
aPTT, see activated partial see also carotid body carotid aneurysms, 134
thromboplastin time. tumour investigations, 135
arterial Biothesiometer, 14 operative strategies, 135
claudication causes, 26 BKA, see below-knee prevalence, 134
dissection, 80 amputation prognosis, 135
puncture site haemorrhage, blue toe syndrome, 100 pulsatile neck mass, 135
79, 80 brainstem death tests, 307 carotid artery injury, 177
repair, 175 Brainstem death tests, 308 blunt injury, 179
rupture, 80 branched grafts, 97 classification, 177
vs. venous claudication, 29 bruit, 60 diagnosis, 177
arterial occlusion 50 Buerger’s disease, 18, 264 management, 178
arterial TOS (ATOS), 149 diagnostic criteria, 18, 265 carotid artery stenting (CAS),
treatment, 150, features, 265 140
see also neurogenic TOS, investigations, 265 carotid body tumour (CBT),
venous TOS management, 19, 266 5, 135
arteriography peripheral angiogram, 19 causes, 136
contraindications, 158 presentation, 265 characteristics, 6
arteriovenous fistula (AVF), 9 risk population, 18 cross-sectional imaging, 136
and cardiac failure, 10 symptoms, 18 CT/MR findings, 6
complications, 11 vasculitidies types, 19 diagnosis, 6
location, 10 ‘buffalo-hump’, 198 duplex ultrasound, 136
Index 319
endovascular interventions, 7 chronic limb ischaemia transaortic endarterectomy,

features, 136 ALARA principle, 83 159
intra-arterial angiography, 7 anastomosis, 77 chronic venous insufficiency
location, 6 anastomotic techniques, 77 (CVI), 28
perioperative bleeding angioplasty, 79, 81 aetiology, 215, 216
reduction, 136 angioplasty complications, clinical features, 207, 216
preop assessment, 6 80 epidemiology, 215
prognosis, 136 aorto-iliac bypass, 77 treatment, 217
resection treatment, 136 autologous grafts, 77 venous skin changes, 28
Shamblin classification, 6 axillo-bifemoral grafts, 77 chronic venous ulceration,
symptoms, 6 below knee stenting, 82 208
carotid dissection (CD), 131 CIN, 81 Cimino Brescia fistula, 9
causes, 131 considerations for CIN, see contrast induced
features, 131 endovascular nephropathy
investigation, 132 procedures, 84 CLI, see critical limb
management, 132 endarterectomy, 76 ischaemia66
prognosis, 132 endovascular theatre, 83 clopidogrel, 64, 277
carotid endarterectomy (CEA), femoral-femoral crossover CMI, see chronic mesenteric
138 graft, 77 ischaemia
benifits, 140 femoro-distal bypass, 78 Cockcroft-Gault formula, 302
carotid shunt, 16 femoro-popliteal bypass, 77 colorimetry, 153
cerebral perfusion, 17 hybrid procedures, 83 colour flow duplex (CFD), 62
complications, 15 iliac artery angioplasty, 81 compartment syndrome, 56
cranial nerves, 16 imaging techniques, 76 fasciotomy, 56,
criteria for surgery, 140 lesion prioritisation, 76 see also rhabdomyolysis
GALA trial, 16 LSV, 77 compression
patient management, 140 prosthetic grafts, 77 hosiery, 31
performance, 142 radiation protection, 83 pressure classes, 31
post-op CVE, 17 required treatments, 84 stockings, 211
risks, 16 restenosis, 83 therapy, 23
stroke reduction, 140 SFA stenting, 81 computerised tomography
treatment speed, 140 stent graft, 83 (CT)
carotid patching, 16 surgical bypass, 76 angiography (CTA), 158
CAS, see carotid artery stenting TASC guidelines, 84 contrast enhanced, 89
catheter ablation, 212 trauma, 84 risks, 89
EVLT, 212 upper limb revascularisation, congenital lymphoedema, 195
RFA, 212 84 congenital vascular
cavernous haemangiomas, 20 chronic lower limb swelling, malformations (CVMs),
CBT, see carotid body tumour 195 242
CD, see carotid dissection chronic mesenteric ischaemia classification, 243
CD4 cells, 314, see also major (CMI), 156 diagnosis, 247
histocompatability aetiology, 157 differential diagnosis, 245,
complex antegrade reconstruction, 246
CEA, see carotid 160 endovascular treatment, 251
endarterectomy clinical presentations, 157 eponymous syndromes, 243
cerebrospinal fluid (CSF), 106 diagnosis, 157, 158 extra-truncular, 244, 250
cerebrovascular event (CVE) diagnostic method Hamburg classification, 244
assessment, 17 evaluation, 159 indications for intervention,
treatment, 17 grafts, 160 248
cervical paraganglionoma, 136 pathophysiology, 157 investigation, 246
CFD, see colour flow duplex retrograde reconstruction, MRI, 247
Charcot’s foot, see Charcot’s 160 Mulliken classification, 245
neuroarthropathy surgical bypass, 159 sclerotherapy, 252
Charcot’s neuroarthropathy, 14 surgical vs. endovascular surgical treatment, 249
Charles’ procedure, 202 repair, 162 treatment indications, 248
Chopart amputation, 189 symptoms, 156 truncular, 245
320 Index
connective tissue disorders, 267 prevalence, 58 management, 101

Ehlers-Danlos (ED) surgical bypass, 67 pre-operative assessment,
syndrome, 270 trans-luminal angioplasty, 66 102
Loeys-Dietz (LD) syndrome, vein bypass graft, 67 renal and hepatic function,
269, 270 CSF, see cerebrospinal fluid 102
Marfan’s syndrome, 268 curcumin, 97 stress echocardiography, 101
Continuous Positive Airway CVE, see cerebrovascular event diabetes, 276
Pressure (CPAP), 281 CVI, see chronic venous HbA1C and macrovascular
Continuous Veno-Venous insufficiency events, 47
Haemofiltration (CVVH), CVMs, see congenital vascular recommendations, 46
285 malformations diabetic foot, 13, 69, 189
contrast induced nephropathy CVVH, see Continuous Veno- clinical assessment, 13
(CIN), 81, 284 Venous Haemofiltration clinical features, 68
balloon expandable stents, 81 cystic ddventitial disease cosequences, 185
carbon dioxide, 80 (CAD), 25 infection level, 70
N-acetylcysteine (NAC), 80 cystic hygroma, 21 limb loss, 13
stent, 81 management, 14, 70
coronary artery disease (CAD) dabigatran, 226 negative pressure wound
ACEI and ATII, 278 D-dimer testing, 223 therapy, 70
acute case management, 279 death, 307 diabetic neuropathy, 14
anti-platelet agents, 277 Debakey classification, 111 Biothesiometer, 14
β blockers, 278 dissections types, 110 testing, 14
co-morbidities, 277 deep vein thrombosis (DVT), diabetic vascular disease
drug therapy, 278 220–2, 221–4, 223–11 autonomic neuropathy, 69
oral anticoagulants, 278 clinical features, 221–5 diabetic foot management,
preoperative investigations, diagnostic evaluation, 223 70
272, 275 factor Xa inhibitors, 226 foot complications, 70
scoring systems for surgery, IVC filters, 227 ischaemia, 70
273, 274 LMWH treatment, 226–7 motor neuropathy, 69
statins, 278 patient assessment, 224 neuropathy, 69
surgery risk assessment, 273 prophylaxis, 224–6 prevalence, 68
coronary stents, 276 PTS, 227 sensory neuropathy, 69
costoclavicular space, 146 in right popliteal vein, distal emoblisation, 100
covered stents, see stent graft 223–10 distal endoleak, 94
CPAP, see Continuous Positive risk assessment, 221–5 distal ischaemia, 173
Airway Pressure risk factors, 222 arm inspection, 9
cranial arteritis, 132 thrombolysis, 227–10 popliteal aneurysms in, 3,
critical care, 279 UFH treatment, 224 see also limb ischaemia
cardiac failure, 283 Wells clinical score, 222 doxycycline, 97
levels, 279, 280 descending thoracic aortic Drumond marginal artery, 165
nutrition, 286 aneurysm (DTAA) DTAA, see descending thoracic
renal failure, 285 blue toe syndrome, 100 aortic aneurysm
respiratory failure, 282 branched and fenestrated duplex Doppler ultrasonography
sepsis, 287 grafts, 104 (DUS), 124
critical illness, 286 causes, 99 DUS, see duplex Doppler
critical limb ischaemia complication reduction, 104, ultrasonography
revascularization, 185 105, 106 DVT, see deep venous
critical limb ischaemia (CLI), complications and adjuncts, thrombosis.
58, 66 104 dyslipidaemia, 41
amputation, 68 CPX testing, 102 blood sampling, 42
BASIL study, 67 CT angiogram, 100 recommendations, 43
cardiovascular risk, 66 endovascular approach, 104 statin therapy, 42
diagnosis, 66 GI tract risk reduction, 106
Duplex ultrasound intraoperative heparin, 104 eczema, 208
surveillance, 67 involving aortic arch, 103 ED syndrome, see Ehlers-
infra-geniculate bypass, 66 left-sided thoracotomy, 103 Danlos (ED) syndrome
Index 321
effort thrombosis, see Paget- endovascular treatment, investigation, 126

Schroetter syndrome 5, 7, 161, 162 post-catheterisation, 126
Ehlers-Danlos (ED) syndrome, see also percutaneous symptoms, 126
269 transluminal angioplasty treatments, 127,
investigation, 270 and stenting see also profunda
presentation, 269 eponymous syndromes, 243, femoris artery
elephant trunk procedure, 244 aneurysms
112 EVAR, see endovascular femoral embolectomy
Elevated Arm Stress Test aneurysm repair at femoral aortic bifurcation,
(EAST), see Roos test excision, 154 53
embolectomy, 74 preparation, 53
brachial artery bifurcation, FAA, see femoral artery thrombosed popliteal artery
54 aneurysm aneurysm, 54
at femoral bifurcation, 53 Factor Xa inhibitors, 226 femoral-femoral crossover
femoral bypass, 54 FApA, see femoral artery graft, 77
at femoral popliteal pseudoaneurysms femoro-distal bypass, 78
bifurcation, 53 FAPs, see femoral anastomotic femoro-popliteal bypass, 77
post-op care, 54 pseudoaneurysm fenestrated grafts, 96
thrombotic occlusion fasciotomy, 56 fibrillin, 267
resistant, 75 indications, 55, 176 fibrin cuff hypothesis, 216
emboli see also compartment fibromuscular dysplasia
distal embolization, 75, 80 syndrome (FMD), 130, 131
saddle, 50 femoral anastomotic carotid/vertebral angioplasty,
supra-inguinal occlusive pseudoaneurysms (FAPs), 131
emboli, 74 7 causes, 130
symptomatic distal aortobifemoral surgery investigation, 131
embolisation, 124, consequence, 8 prognosis, 131
see also arterial bilateral femoral artery Fibromuscular dysplasia
occlusion, embolectomy, aneurysms, 8 (FMD), 25
femoral embolectomy, diagnosis, 8 FMD, see fibromuscular
graft occlusion factors inducing, 9 dysplasia
endarterectomy, 76 indicators for treatment, 9 foam sclerotherapy, 211
endofibrosis, 25 surgical repair, 9 fondaparinux, 226
endoleak, 32, 33, 94 symptoms, 7, 8 see also heparin
type I, 94 femoral artery aneurysm foot ulceration, see diabetic
type II, 95 (FAA), 123 foot
type III, 95 acute thrombosis, 124 FRCS examination, 1–2
type IV, 95 chronic thrombosis, 124 fusiform aneurysms, 117
type V, 95 classification, 124
end-organ ischaemia, 109 conservative management, gangrene, 185
endotension, see also endoleak 125 GCA, see giant cell arteritis
endovascular aneurysm repair DUS, 124 Ghent diagnostic criteria, 268
(EVAR), 31 interposition graft, 125 giant cell arteritis (GCA), 132,
aortic morphology, 93 investigation, 125 263
benefit, 97 occurrence, 123 cerebrovascular events, 132
branched grafts, 97 pathophysiological cranial arteritis, 132
CT scans, 93 conditions, 123 high dose steroid therapy,
fabric grafts, 92 perioperative mortality rates, 133
fenestrated grafts, 96 125 investigation, 133, 264
follow-ups, 95 symptomatic distal large vessel vasculitis, 132
hybrid repairs, 97 embolisation, 124 management, 264
infection risk reduction, 93 symptoms, 124 pathology, 132
proximal neck, 93 femoral artery presentation, 264
risks, 34 pseudoaneurysms (FApA), prognosis, 133
trial I, 96 126 systemic inflammatory
trial II, 96 DUS evaluation, 127 syndrome, 132,
322 Index
giant cell arteritis (GCA) (cont.) HHD, see Hand Held immunosuppressive agents,
see also temporal Doppler27 314, 315
arteritis hindquarter amputation, 191 malignant growth, 300,
glomus vagale tumours, 136 HIT, see heparin-induced see also graft rejection
graft thrombocytopenia. implantable cardiac
arm vein, 118 Homan’s operation, 202 defibrillators (ICDs),
autologous vein, 118 Homan’s sign, 221 276, see also permanent
composite-sequential, 119 homocysteine, 47 pacemakers
great saphenous vein, 174 hyperhomocysteinaemia, 48 infantile haemangioma, 243
inlay grafting, 118 recommendations, 48 infection
long saphenous vein, 77 hyperhidrosis, 152 femoral pseudoaneurysms,
graft infection aluminium chloride 240
aortic stent graft, 233 hexahydrate, 153 post operative, 230
clinical presentation, 231 botulinum toxin A inotropic agents, positive, 283
extra-anatomic bypass, 234 treatment, 153 INR, see international
grade III infection, 230 botulinum treatment side- normalized ratio.
infected aortic grafts, 236 effects, 154 intermittent claudication, 58,
infected carotid patches, 237 causes, 152 59
infected peripheral grafts, chemical sympathectomy, amputation, 184
238 155 angioplasty trials vs. exercise,
in-situ vein graft, 233 colorimetry, 153 66
investigations, 233 compensatory sweating, 154 claudication distances, 59
mycotic aneurysm, 241 diagnosis, 153 nerve root irritation, 61
revascularization, 239 glycopyrollate treatment, 153 non-atherosclerotic cause, 61
graft occlusion iontophoresis, 153 spinal canal stenosis, 61
treatment, 54, see also arterial primary, 152 surgical reconstruction, 66
occlusion, emboli secondary, 152 international normalized ratio
graft rejection, 313 surgery, 154 (INR), 225
mechanism, 314, sympathectomy, 154, 155 International Registry of Acute
see also organ symptoms, 152 Aortic Dissection (IRAD),
transplantation hyperkalaemia 112
effect on revascularisation, interscalene triangle, 146
haemangioma, 20 76 intestinal angina, 157
infantile, 243 treatment, 285, see also renal intimal disruption, 109
haematoma, 89 failure intimal fibroplasia, 130
haematoma effects, 16 hypertension intra-arterial angiography, 90
‘halo sign’, 264 ACE inhibitor, 44 iontophoresis, 153
Hamburg classification, 244 B-Blockers, 45 IRAD, see International
Hand Held Doppler (HHD), 27 recommendations, 45 Registry of Acute Aortic
‘helper’ cells, see CD4 cells treatment guidelines, 44 Dissection
heparin, 225 venous etiology, 29 ischaemia, 49
in ALI, 73 hyphenweb varices, 208 irreversibly ischaemic, 51
CD, 132 hypoglossal nerve, 15 signs, 60
in clinical practice, 220 thrombosis, 51,
intraoperative dose, 92, 104 iatrogenic vascular injuries, see also compartment
ischaemic colitis, 168 182 syndrome,
and mechanical heart valves, ICDs, see implantable cardiac rhabdomyolysis
278 defibrillators ischaemic colitis, 164
pericatheter thrombus, 74 iliac artery angioplasty, 81 aetiology, 165
in post-op care, 54 endovascular treatment, 81 anatomical factors, 166
and sclerosant, 182 left common iliac artery clinical features, 166
in vascular repair, 175, occlusion, 82 diagnosis, 166
see also thrombolysis. right sided stent, 82 etymology, 164
heparin-induced stent placement, 81 investigations, 168
thrombocytopenia (HIT), iliac vein injuries, 180 management, 168
226–7 Iloprost, 260 in mucosa, 165
Index 323
pathophysiology, 165 ankle level amputation, 190 chronic mesenteric

post aortic, 169 BKA, 190 ischaemia, 158
prevention, 170 decision making, 186 magnetic resonance imaging
risk factors, 169 hindquarter amputation, (MRI), 90
“Ischemic monomelic 191 major histocompatability
neuropathy”, 296 mid-foot amputation, 189 complex (MHC), 314
IVC injuries, 180 preoperative checklist, 187 mannitol, 284
ray amputation, 189 Marfan syndrome, 267
kidney transplantation, through-knee amputation, investigation, 268
see renal transplantation 190 management, 268
Klippel-Trenaunay syndrome, toe amputation, 189 presentation, 267
243 transmetatarsal amputation, mechanical thrombectomy,
Klippel-Trénaunay syndrome 189 55
(KTS), 21 lower limb swelling ALI, 76, see also thrombolysis
non surgical options, 23 differential diagnosis, 196 medial dysplasia, 130
pathological components, 22 DVT, 198 mesenteric vessel injuries,
surgical options, 23 LSV, see long saphenous vein 180
varicose veins, 22 LT, see liver transplantation MHC, see major
varicosities distribution, 22 lung disease, see pulmonary histocompatability
venous system, 22 diseases complex
KTS, see Klippel-Trénaunay lymphangioma mid-foot amputation, 189
syndrome circumscriptum, 21 Milroy’s disease, see congenital
lymphangioscintigraphy, 12, lymphoedema
large vessel vasculitis, 132 199 MODS, see Multiple Organ
laryngeal nerve, 15 lymphoedema, 11, 195 Dysfunction Syndrome
LD syndrome, see Loeys-Dietz bypass procedures, 203 motor neuropathy, 69
(LD) syndrome causes, 197 MRA, see magnetic resonance
limb distal vessel injury, 176 Charles’ procedure, 202 angiography
limb ischaemia, 173 classification, 11, 12, 196 MRI, see magnetic resonance
limb vascular injury compression therapy, 201 imaging
assessment, 173, 174 computed tomography, 199 Multiple Organ Dysfunction
investigations, 174 decongestive physiotherapy, Syndrome (MODS), 279
lipidema, 203 201 mycotic aneurysm, 236, 241
lipodermatosclerosis, 208 diagnosis, 198
liposuction, 154, 202 drugs, 202 National institute of clinical
liver disease, alcohol-induced, of foot, 12 excellence (NICE)
299 Homan’s operation, 202 neonatal haemangioma, 243
liver transplantation (LT), 300 infection prevention, 202 neovascularisation, 213
disease specific indications, investigations, 200 nerve root irritation, 61
299 of leg, 198 neurogenic TOS (NTOS),
operative technique, 310 liposuction, 202 148
postoperative complications, lymphangioscintigraphy, Adson test, 148
312 199 botulinum
L-lactate, 167 lymphatic drainage, 200 chemodenervation, 150
LMWH, see low molecular management, 200 paraesthesia, 148
weight heparin. MR features, 200 Roos test, 148
Loeys-Dietz (LD) syndrome, MRI findings, 12 surgical treatment, 150,
268 pathophysiology, 197 see also arterial TOS,
investigations, 270 secondary causes, 12 venous TOS
management, 270 surgical procedures, 13 neuropathic foot ulcer, 69
presentations, 269 symptoms, 11 neuropathy
long saphenous vein (LSV), 77 treatment, 200 autonomic, 69
low molecular weight heparin motor, 69
(LMWH), 224 magnetic resonance sensory, 69
lower limb amputation, 191 angiography (MRA), 63 NICE see National institute of
AKA, 191 chronic limb ischaemia, 76 clinical excellence
324 Index
nifedipine, 260 PCI, see Percutaneous phlegmasia cerulea dolens,

NTOS, see neurogenic TOS Coronary Interventions. 221–5
PE, see pulmonary embolism. phlegmesia alba dolens, 221–5
obesity, 275 Percutaneous Coronary plethysmography, 210
obstructive pulmonary Interventions (PCI), 277 Polyarteritis nodosa, 19
diseases, 275 percutaneous transluminal popliteal aneurysms (PAs), 3, 4
occlusive lesions, 158 angioplasty and stenting endovascular treatment, 5
open aneurysm repair, 92 (PTA/Stent), 161, 163 indications for treatment, 5
oral anticoagulants, 278 perforator surgery, 213 ischaemic complications, 3
organ allocation peripheral arterial disease need for thrombolysis, 5
ethical principles, 304, 307 (PAD), 37 symptoms, 4
factors considered, 306 ABI predictor, 38 popliteal artery, 4
system of allocation, ABPI measurement, 60 accessing, 4
304, see also organ ACE inhibitor, 44, 45 thrombosis, 54,
transplantation anti-platelet therapy, 45 see also popliteal
organ donation, 307 aspirin resistance, 46 aneurysms
contraindications, 308 auscultation, 60 popliteal artery aneurysms
death conformation, 307 B-Blockers, 45 (PAA), 115
donor categories, 308 cardiovascular events, 38 bilateral, 117
donor operation, 309 cardiovascular risk causes, 115
ethics, 307 management, 59, 60, 65 clopidogrel, 121
organ failure, 280 CFD ultrasound scanning, complication, 116
cardiac failure, 283 62 embolectomy, 120
indicators, 281 cholesterol reduction, 64 endovascular repair, 121
renal failure, 285 circulation examination, 60 endovascular treatment
respiratory failure, 282 claudication, 58, 59, 60 results, 121
therapeutic measures, 280 clopidogrel vs. aspirin, 45 fusiform aneurysms, 117
organ procurement, 309 CT angiogram, 62 graft repair, 119
organ transplantation, 305 CV risk, 59 intervention, 117
life expectancy, 315 diabetes, 46, 64 investigation, 116
postoperative complications, differential diagnosis, 61 ligation, 118
311 dyslipidaemia, 44 limb salvage, 120
solid, 298 hazard ratios, 42 posterior approach, 118,
osteomyelitis homocysteine, 48 119
diagnostic imaging, 15 hypertension, 45, 64 preoperative lysis, 120
and graft infection, 231 Kaplan Meier survival presentation, 116
x-ray, 70 curves, 40 pulsatile mass, 116
life style modification, 47 risk factors, 117, 121
PAA, see popliteal artery MRA, 63 rupture, 116
aneurysms platelet activation, 46 saccular aneurysms, 117
PAD, see peripheral arterial pole test, 61 symptoms, 116
disease prevalence, 58 thrombolysis, 120
Paget-Schroetter syndrome, pulse palpation, 60 treatment, 117, 119
149 risk factors, 37, 38 popliteal entrapment
pancreas transplantation, 303 simvastatin therapy, 42 syndrome, 23
operative technique, 310 smoking, 40, 64 classifications, 25
postoperative complications, spinal canal claudication, 59 clinical examination, 24
313, see also renal survival rate, 37 compression causes, 24
transplantation walking, 61, 66 embryological factors, 24
paragangliomas, 5 peripheral vasodilators, 260 portal vein injuries, 180
patient type, 5 permanent pacemakers (PPM), postoperative death
treatment, 5 276 risk factors, 275
paraplegia risk reduction, 106 permissive hypercapnia, 282 post operative infection, 230
Parkes-Weber syndrome, 23, PFAa, see profunda femoris antibiotic prophylaxis, 231
243 artery aneurysms causative organisms, 230
PAs, see popliteal aneurysms phlebectomies, 213 prevention, 231
Index 325
risk factors, 231, see also graft haemodynamic changes, 256 retro-peritoneal haematoma,

infection immune regulation, 258 179
postphlebitic syndrome, 218 investigations, 259 classification, 180
post-thrombotic syndrome management, 261 revascularisation, 239
(PTS), 227 mechanical damage, 258 critical limb ischaemia, 185
post-thrombotic syndrome neurogenic response, 256 femoral embolectomy, 53,
(PTS), 228, see also post- organs affected, 255 see also thrombectomy
phlebitic syndrome presentation, 254, 258 rhabdomyolysis, 56,
“potato tumour”, see carotid triggering factors, 255, see also compartment
body tumour (CBT) see also Takayasu syndrome
PPM, see permanent arteritis Roos test, 148
pacemakers RD, see primary Raynaud’s RP, see Raynaud’s phenomenon
primary Raynaud’s disease disease RRT, see renal replacement
(RD), 255 REACH, see Reduction of therapy
profunda femoris artery Atherothrombosis for RS, see secondary Raynaud’s
aneurysms (PFAa), 126 Continued Health syndrome
treatment aim, 126 recombinant tissue rTPA, see recombinant tissue
prophylaxis plasminogen activator plasminogen activator
antibiotic, 231 (rTPA), 74
prosthetics, 193 Reduction of Atherothrombosis saccular aneurysm, 117, 240
Proteus syndrome, 23 for Continued Health saphenofemoral junction
proximal endoleak, 94 (REACH), 37 ligation, 213
pseudoaneurysm renal failure, 285 saphenopopliteal junction
infected femoral, 240 access placement, 289 ligation, 213
mycotic, 127 acute, 284 saphenous vein stripping, 213
post-catheterisation, 126 causes, 283 SC, see subcutaneously
predisposing factors, 127 CIN, 284 sciatic artery aneurysm, 25
pseudoxanthoma elasticum management, 284 sclerosants, 250, 251
(PXE), 26, 270 vein injuries, 180 ethanol, 251
investigations, 271 renal replacement therapy sclerotherapy, 250
management, 271 (RRT), 289 technique, 252
presentation, 270 renal transplantation secondary Raynaud’s syndrome
PTA/Stent, see percutaneous contraindications, 303 (RS), 255
transluminal angioplasty indications for, 301 associated diorders, 257
and stenting live donation, 301 sensory neuropathy, 69
PTS, see post-thrombotic operative technique, 310 sepsis, 287, see also gangrene
syndrome. postoperative complications, SFA stenting, 81
pulmonary diseases, 275 313 ‘bail out’ procedure, 82
pulmonary embolism (PE), 220 recurring diseases, 302 Shamblin classification, 6
“Pulseless disease”, see Takayasu respiratory failure, 282 ‘Silk Road’ disease, see Bechet’s
arteritis arterial blood gas (ABG) disease
PXE, see pseudoxanthoma analysis, 280 sitophobia, 156
elasticum CPAP, 281 smoking cessation
oxygen face masks, 281 effectiveness, 41
radiation arteritis permissive hypercapnia, nicotine replacement
features, 129 282 therapy, 40
investigation, 130 positive pressure ventilation, pharmacotherapy, 44
pathology, 129 282 recommendations, 39, 40
treatment, 130 restenosis, 83 spinal canal claudication, 59
ray amputation, 189 external beam radiotherapy, spinal canal stenosis, 61
Raynaud’s phenomenon (RP), 83 Stanford classification, 111
254, 255 stent graft effect, 83 dissections types, 110
blood flow assessment restrictive pulmonary diseases, statin therapy
techniques, 259 275 AAA, 97
epidemiology, 255 reticular varices, 208 anti-inflammatory effects, 43
genetic predisposition, 258 retrograde reconstruction, 160 atherosclerotic plaque, 278
326 Index
statin therapy (cont.) inflammatory process, 133 stress echocardiography, 101

cholesterol reduction vs. investigation, 262 type I, 102
events, 43 lab investigations, 134 type II, 102
perioperative cardiac events, modified Ishikawa type IV, 104
44 classification, 263 thrombectomy, 227
stenotic lesions, 158 presentation, 262 thromboangiitis obliterans,
stent, 81 prognosis, 134 see Buerger’s disease
balloon expandable, 81 surgical revascularization, thrombolysis, 227
self-expanding, 81 134 advantages, 55
stent graft, 83 symptoms, 133 complications, 55
infected aortic, 233 systemic inflammatory contraindications, 55, 75
uses, 83 response, 134 distal embolization, 75
streptokinase, 74 treatment, 263 intra arterial, 54
stroke, 136 vascular insufficiency, 134 intra operative, 54
asymptomatic stenoses TASC, see Transatlantic low dose heparin
detection, 137 Inter-Society Consensus administration, 74
carotid bifurcation, 137 guidelines pharmacological thrombus
carotid territory symptoms, TBI, see toe-brachial index dissolution, 74
137 telengiectasia, see hyphenweb pre-requisites, 75
carotid territory, causes of, varices regimes, 75
137 temporal arteritis, 19, risks, 75
CEMRA and CTA, 138 see also giant cell arteritis rTPA, 74
duplex ultrasound, 138 thoracic outlet, 147 streptokinase, 74
investigations, 138 thoracic outlet syndrome thrombolytic agents, 74
management, 138 (TOS), 146 urokinase, 74
risk factors, 137 anatomical factors, 147 thrombophilia, 221–4
sub-bandage pressure, 30 arterial, 149 thrombophlebitis, 206
subclavian occlusive disease causes, 147 thrombosis, 51
cause, 142 channels, 146 acute, 124
clinical features, 143 clinical syndromes, 147 chronic, 124
indications, 143 investigations, 150 graft occlusion, 54
investigations, 143 neurogenic, 148 predictors, 127
management, 143 venous, 149 spontaneous, 127
subcoracoid space, 146 thoracic vascular injuries, 176 treatmement, 52,
subcutaneous curettage, 154 diagnosis, 176 see also arterial
sub-intimal angioplasty, 66, 79 investigations, 176 occlusion
suprarenal aneurysms, 104 management, 177 through-knee amputation,
sweating, 152 thoracoabdominal aortic 190
compensatory, 154, aneurysm (TAAA) TIA, see transient ischaemic
see also hyperhidrosis blue toe syndrome, 100 attack
sympathectomy, 154, 261 causes, 99 toe amputation, 189
chemical, 155 complication reduction, 105, toe pressure, 61
thermocoagulation, 155 106 toe-brachial index (TBI)
systemic inflammatory complications and adjuncts, TOS, see thoracic outlet
syndrome, 132 104 syndrome
CPX testing, 102 total parenteral nutrition
TA, see Takayasu arteritis Crawford classification, 100 (TPN), 286
TAAA, see thoracoabdominal CT angiogram, 100 TPN, see total parenteral
aortic aneurysm GI tract risk reduction, 106 nutrition
Takayasu arteritis (TA), 19, intraoperative heparin, 104 Transatlantic Inter-Society
133, 262 laparotomy, 100 Consensus (TASC)
diagnostic criteria, 262 management, 101 guidelines, 84
high dose steroid therapy, open repair results, 105 lesions types, 84
133 preoperative assessment, 102 patient’s comorbidities, 84
immuno-suppressive agents, renal and hepatic function, transient ischaemic attack
134 102 (TIA), 136
Index 327
transmetatarsal amputation, vascular anomalies, 20 slow-flow lesions, 20

189 vascular injury varicography, 209
treadmill walking, 61 criteria for emergency venous thromboembolism
truncal varices, 208 operation, 173 (VTE), 220
limb, 174 prophylaxis, 225–11
UFH, see unfractionated limb distal vessel injury, venous TOS (VTOS), 149
heparin. 176, see also abdominal Paget-Schroetter syndrome,
UK Small Aneurysm Trial vascular injuries, carotid 149
(UKSAT), 91 artery injury, iatrogenic rib resection, 150,
UKSAT, see UK Small vascular injuries, limb see also arterial TOS,
Aneurysm Trial. distal vessel injury, limb neurogenic TOS
unfractionated heparin (UFH), vascular injury, thoracic venous ulceration, 28, 218
224 vascular injuries, clinical asssessment, 28
DVT treatment, 224–6 vertebral artery injuries management, 30
upper limb amputation, 191 vascular malformations, 20 sub-bandage pressure, 30
urokinase, 74 arteriovenous malformations, vertebral artery injuries,
21 179
vagus nerve, 15 capillary malformation, 20 vertebral discitis, 87
varicose veins, 26 fast-flow, 20 vertebral occlusive disease,
aetiology, 205 investigations, 20 143
asymptomatic, 206 lymphatic malformations, investigations, 144
clinical examination, 27 21 pathology, 144
clinical features, 206, 208 slow-flow, 20 prognosis, 144
epidemiology, 204 syndromes, 23 symptoms, 144
investigations, 210 venous malformations, treatment methods, 144
recurrent, 214 20, see also congenital ‘vibration white finger’,
risk factors, 205 vascular malformations see Raynaud’s phenomenon
secondary causes, 29 vascular steal syndrome, Virchow triad, 220
symptomatic, 207 296 visceral duplex ultrasound
thrombophlebitis, 206 vascular trauma, 172 (VDU)
treatment, 213 vasculitidies, 19, 254, 262 chronic mesenteric
treatment complications, Buerger’s disease, 266 ischaemia, 157
214, 215 GCA, 264 limitations, 158
vascular access, 288 TA, 263 von Willebrand factor (VWF),
arterio-venous access grafts, vasospasm, see Raynaud’s 256
293 phenomenon VTE, see venous
arterio-venous fistula, 292 VDU, see visceral duplex thromboembolism.
AV grafts, 295 ultrasound VTOS, see venous TOS
AVF, 292, 294, 295 vena caval filters, 227–9 VWF, see von Willebrand
care after placement, 294 venography, 209 factor
central venous catheters, 293, contrast venography, 209
296 CTV, 209 walking distance, 65
failure management, 296 MRV, 209 exercise benefits, 65
operating techniques, 294 venous claudication, 61 vasoactive drugs, 65
placement diagnosis’ need, causes, 218 warfarin
289 symptoms, 216, DVT, 225
procedures, 291 see also intermittent warfarin, 278
RRT commencement, 290 claudication watershed region, 166
service organisation, 289 venous incompetence, 26 white cell trapping hypothesis,
stenosis, 294 venous malformations 216
surveillance, 295 AVM, 21
thrill detection, 294 cavernous haemangiomas, ximelagatran, 226
types, 291 20

PostGraduate Vascular Surgery

Uploaded by

Copyright:

Available Formats

PostGraduate Vascular Surgery

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

PostGraduate Vascular Surgery

Uploaded by

Copyright:

Available Formats

What is the book about?

What is the book about?

What topics are covered in the book?

What topics are covered in the book?

This page intentionally left blank

Postgraduate Vascular Surgery

© Cambridge University Press 2010

This publication is in copyright. Subject to statutory exception

First published 2010

Printed in the United Kingdom at the University Press, Cambridge

Library of Congress Cataloguing in Publication data

ISBN 978-0-521-13352-4 Paperback

Cambridge University Press has no responsibility for the persistence or

Section 1. Final FRCS vascular 9 Femoral artery aneurysms 123

5 Abdominal aortic aneurysms 86 16 Indications and management of

19 Management of deep vein 24 Access surgery 288

Mohamed Abdelhamid David M. Cressey

Julie Brittenden Robbie George

Rod Chalmers George Hamilton

Mike Clarke Mathew Jacob

Marcus Cleanthis Mark Kay

David Lambert Andrew Platts

Introduction to the examination

Figure I.1 Angiographic image of a popliteal

What are your indications for elective repair?

When is thrombolysis utilised?

What major problem faces endovascular treatment of PAs?

Carotid body tumour

What are the typical CT/MR findings?

What is the preoperative assessment?

What classification system is used for CBTs?

What endovascular interventions can be helpful in managing large tumours?

Figure I.2 Intra-arterial angiography of carotid body

Femoral anastomotic pseudoaneurysms

Figure I.3 Femoral anas-

What factors predispose to FAP formation?

What are your indications for surgery?

What is your approach to surgical repair?

What findings on palpation can suggest that a fistula is at risk?

Why is cardiac failure problematic in patients with AV fistulae?

What are the complications of an AV fistula?

How is lymphangioscintigraphy performed?

What findings on MRI are suggestive of lymphoedema?

What surgical procedures are performed to manage lymphoedema?

How is neuropathy tested?

Figure I.5 Testing neuropathy with a 10 g

What are the principles in managing acute diabetic foot disease?

What is ‘Charcot’s foot’?

What imaging aids the diagnosis of osteomyelitis?

scanning is now becoming established as the investigation of choice in diagnosing osteo-

Complications of carotid endarterectomy

Would you use local or general anesthesia?

Does patching reduce the risk of perioperative stroke?

What should be performed prior to bilateral carotid surgery?

What is the risk of disabling stroke or death?

What is the frequency of haematoma formation?

When would you use a carotid shunt?

How is cerebral perfusion monitored during CEA?

How is the diagnosis made?

Figure I.6 Peripheral angiogram of tibial vessels