Cardiac ICDs
Cardiac ICDs
Cardiac ICDs
doi: 10.1093/bjaed/mkw020
Advance Access Publication Date: 6 May 2016
Matrix reference
1A03, 2A03, 3G00
© The Author 2016. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved.
For Permissions, please email: journals.permissions@oup.com
388
Perioperative management of patients with CIEDs
Implantable cardioverter defibrillators device, the box responds to sensing of intrinsic electrical activity
in single or multiple chambers by either inhibiting or triggering
Implantable cardioverter defibrillators (ICDs) sense and analyse
pacing in one or more chambers.
myocardial electrical activity and are capable of pacing and shock
The Generic Pacemaker Code describes the PPM function; po-
therapy when necessary. Most ICDs are implanted (in accordance
sitions 1–3 refer to the chamber paced, the chamber sensed, and
with NICE guidance), for secondary prevention in patients who
the response to sensing, respectively (Table 1).4 For example, VVI
have survived a cardiac arrest or other significant haemodynamic
would denote a ventricular paced and sensed PPM with an inhi-
compromise after ventricular arrhythmias.3 ICD implantation is
biting response to sensing. The commonly encountered DDD
indicated in selected patients after myocardial infarction [usually
mode can have essentially four responses according to the de-
with scar-related ventricular tachycardia (VT) risk] or those with
tected intrinsic cardiac rhythm:
significant left ventricular dysfunction. Other patient groups at
high risk of sudden death due to ventricular arrhythmias include (i) Atrial sensing and ventricular sensing
those with familial cardiac conditions such as long QT syndrome,
Brugada syndrome, cardiomyopathies, and selected patients If both atrial and ventricular activity are detected within the
with congenital heart disease. There are specialized ICDs available appropriate pre-programmed time interval (which depends on
which permit CRT for patients with indications for biventricular the desired heart rate), then there will be inhibition of any pacing
PPMs at risk of arrhythmias. These devices, predominantly activity.
inserted for primary prevention, are referred to as CRT-Ds (ii) Atrial pacing and ventricular sensing
(‘D’ denotes defibrillator).
Atrial pacing leads to intrinsic ventricular activity within the pro-
grammed time interval which is sensed by the ventricular lead
Implantable loop recorders
and therefore inhibits pacing (e.g. sick sinus syndrome with a
These leadless cardiac monitoring devices function solely as a normal P–R interval).
diagnostic tool. They are indicated for patients with recurrent
(iii) Atrial sensing and ventricular pacing
unexplained syncope and/or palpitations and as a long-term
monitoring device for some patients with atrial fibrillation (AF). Sensed intrinsic atrial activity triggers ventricular pacing when
Implantable loop recorders (ILRs) have self-contained electrodes spontaneous ventricular depolarization is not sensed within a
capable of recording an ECG and can be automatically or patient programmed time interval (e.g. with complete AV block and nor-
triggered (via an external activator) when arrhythmias or symp- mal sino-atrial node function).
toms arise.
(iv) Atrial pacing and ventricular pacing
Physics and nomenclature of CIEDs In this scenario, there is pacing of both the atrium and ventricle if
intrinsic electrical activity is undetected within the specified pre-
Pacemakers set time interval (e.g. with complete heart block and inadequate
A PPM system comprises a pulse generator (which contains intrinsic atrial rate).
a battery and electronic circuitry) and one or more pacing The fourth position of the generic PPM code refers to rate-re-
leads. The pulse generator consists of a silicon chip and elec- sponsive pacing whereby the pacemaker can alter the paced
tronic sensing and output circuitry which analyses the cardiac heart rate in response to motion or sensed physiological condi-
rhythm and paces as programmed. Pacing configuration may tions. Most commonly, an accelerometer detects activity during
entail current delivery to a single cardiac chamber or to mul- exertion and increases the paced rate to optimize cardiac output.
tiple chambers. Unipolar leads have a single electrode at the Other sensing mechanisms may detect an increase in physio-
pacing tip which acts as the current emitting cathode; the logical parameters, including minute ventilation or myocardial
PPM box serving as the anode. With more commonly encoun- contractility, adjusting heart rate accordingly.
tered bipolar leads, the anode lies on the lead itself just prox-
imal to the cathode tip; reducing the susceptibility of the
Implantable cardioverter defibrillators
device to EMI.
The PPM box generates an electrical current between the Conventional ICDs are analogous to pacemaker devices but with
anode and cathode which is transmitted to the myocardium via a larger generator which houses electronic circuitry, batteries,
a lead to achieve a wave of depolarization. Successful pacing- and a capacitor. Defibrillation requires ∼750 V (to deliver an out-
induced depolarization is referred to as ‘electrical capture’. The put of 30–45 J); achieved by charging the capacitor and then dissi-
stimulation threshold is defined as the minimum electrical pating this energy. The device continuously senses and analyses
stimulus that consistently exceeds the excitation threshold native electrical activity (the R–R interval) to detect VT or ven-
of the myocardial cells to produce cardiac depolarization. De- tricular fibrillation. It incorporates anti-tachycardia pacing
pending on the number of leads and the programming of the (ATP) capabilities whereby a programmed burst of overdrive
Table 1 The Generic Pacemaker Code (adapted from Bernstein and colleagues4 with permission from John Wiley & Sons Ltd)
Position I: pacing chamber(s) Position II: sensing chamber(s) Position III: response(s) to sensing Position IV: programmability
pacing is used to attempt to terminate VT. In the case of VF, or if wall. The atrial lead is generally positioned in the right atrial
ATP fails, a capacitor within the ICD will charge and then release appendage and the ventricular lead in the right ventricular apex.
a shock which passes between the defibrillator coils. These coils Biventricular systems require the additional positioning of a lead
are either both located on the ventricular lead or alternatively the to stimulate the wall of the left ventricle; this is usually achieved
box acts as one coil with a single ventricular lead coil. Shock by passing a lead into the coronary sinus via the right atrium
delivery is painful and utilizes more battery life or capacity (Fig. 1). A retrospective study reported a 7.5% complication rate,
(measured in ampere-hours) than ATP. An ICD has the anti- most commonly from lead displacement (4.8%), pneumothorax
bradycardia capabilities of a pacemaker and is able to pace in (3.7%), or infection (1.5%).6 Recently developed leadless PPMs
the event of ‘post-shock’ bradyarrhythmias. have a small sensing and pacing device entirely contained within
the right ventricle with no leads (Fig. 2).
ICDs are implanted similarly to pacemakers with trans-
Sensing and the implications of EMI venous leads which sense and deliver shock therapy as re-
quired. A less invasive, subcutaneous ICD has been developed
Sensitivity of a PPM describes the minimum intrinsic atrial
(S-ICD), primarily for patients with difficult venous access or
or ventricular electrical activity measured in millivolts that is
complex cardiac anatomy. A lead is placed subcutaneously in
sensed by the device. This is set as a margin around the ampli-
the midline, connecting to a pulse generator sited in the left
tude of a sensed signal and depends on origin (atrial or ventricu-
axilla (Fig. 3). The device analyses and treats ventricular
lar) and lead type (unipolar or bipolar). If incorrectly set, the
arrhythmias by delivering an 80 J biphasic shock without the
device may ‘under-sense’ and fail to detect intrinsic atrial
need for endovascular lead placement and the potential
or ventricular activity. Consequently, ‘over-pacing’ can occur
associated complications. The anticipated battery longevity is
whereby the pulse generator fires, despite intrinsic activity with
shorter than conventional ICDs; pacing capabilities are limited
the potential risk of triggering malignant tachyarrhythmias. An
to post-shock pacing only.
alternative scenario of ‘over-sensing’ is when the device inappro-
priately registers myocardial activity when none exists, leading
to failure to pace in the context of no intrinsic electrical activity. Preoperative assessment of patients with CIEDs
The most frequent PPM interaction with EMI is over-sensing.5
EMI is particularly important to consider in a patient who is
History and examination
PPM-dependent since over-sensing may result in inappropriate The presence of a CIED indicates a high likelihood of coexisting
inhibition and significant haemodynamic compromise. In the significant cardiac disease and warrants conduct of a thorough
case of ICDs, EMI may be falsely interpreted as a shockable history and examination. Direct inquiry of any symptoms suggest-
rhythm and as such, there is a risk of intraoperative shock ing device malfunction should be made, including dizziness, syn-
delivery. Disabling the defibrillator function of a device before cope, or indicators of deteriorating cardiac function. Prescribed
exposure to EMI is therefore advised. anti-arrhythmic agents should be continued in the perioperative
period. Electrolyte abnormalities (including hypomagnesaemia),
acid–base disturbances, or blood gas abnormalities should be
CIED insertion corrected since they may influence the stimulation and/or
PPMs are usually implanted subcutaneously below the left clav- defibrillation thresholds.
icle with trans-venous access commonly via the left cephalic,
subclavian, or axillary veins. In paediatric patients, the PPM
may be implanted in other locations including the abdominal
Fig 1 Chest radiograph of a biventricular pacing system (CRT-P). PB, pacing box;
RA, right atrial lead; RV, right ventricular lead; LV, coronary sinus lead Fig 2 Chest radiograph demonstrating a leadless pacemaker system (arrowed)
stimulating the left ventricle. within the right ventricle.
ICD-specific analysis
Although advances in technology have limited the repeated need
to check the defibrillator function of ICDs, a patient may recently
have been sedated for a defibrillator safety margin test. This test
involves inducing VF under controlled conditions ( providing
there is no contraindication such as cavity thrombi) to ensure
that the device senses, detects, and terminates the arrhythmia
reliably and with adequate energy delivery.
Fig 4 ECG demonstrating sole atrial pacing—atrial pacing spikes (arrowed) are followed by atrial capture and normal width QRS complexes.
Fig 5 Sole ventricular pacing—pacing spikes (arrowed) are followed by broad QRS complexes.
Fig 6 A chest radiograph demonstrating a CRT-D device—note the widened shock Fig 7 Chest radiograph demonstrating right ventricular lead migration (arrowed)
coils (S) located on the right ventricular (RV) lead, in contrast to the right atrial (RA) outside of the heart.
lead and the narrow left ventricular (LV) pacing lead.
(i) Significant pacemaker dependency programmed off during surgery since mechanical ventilation
may stimulate excessive pacing rates. Sleep/rest mode is another
If the patient is highly pacemaker-dependent, and the procedure example of an advanced function whereby the programmed
involves potential EMI, a cardiac physiologist should be con- base rate is gradually reduced during a specified sleep phase.
sulted as temporary reprogramming of the device to an asyn- This should be deactivated if late surgery is planned.
chronous (non-sensing) mode (A00, V00, or D00 depending on
the set configuration) may be required. (iii) Defibrillator function
(ii) Advanced CIED functions
The defibrillator function should be deactivated immediately
Any device with a rate response function using minute ventilation before any surgery where EMI is deemed likely. If this is not
as the mechanism of regulating pacing should have this possible, the application of a magnet should be considered.
Fig 8 Example of a European Pacemaker Patient Identification Card for a patient with a CRT device programmed to VVI-R (‘D2, C8, E3’ indicates a patient with heart failure
and chronic AF who required a PPM due to AV-block caused by nodal ablation—see Supplementary Fig. S1).
Magnets medical grade magnet on the device for the duration of surgery.
This is the case with all commonly encountered ICDs, provided
A magnetically activated switch is incorporated into CIEDs to en-
that this magnet function has not been disabled. Some manufac-
able alteration of the pacing or defibrillator modes. Application of
turers (including Medtronic and Boston Scientific) have incorpo-
a magnet to a PPM varies according to the model and settings of
rated audible tones emitted from the ICD to indicate when a
the device. Usually, it delivers an asynchronous mode at a rate spe-
magnet has been applied and defibrillator functions have been
cific to the manufacturer; however, it may initiate a diagnostics
deactivated. Removal of the magnet at the end of surgery should
function and then revert to its programmed mode of pacing. It is
promptly reactivate the defibrillator function.7
important to appreciate that an asynchronous mode may be sub-
optimal in a patient with an underlying native rhythm and rarely
Implantable loop recorders
may induce malignant arrhythmias.5 The device should revert to
programmed baseline settings upon removal of the magnet.7 These devices do not deliver any treatment and they present no
It is possible to deactivate the defibrillator function of an ICD risk to the patient when EMI is encountered. Since noise created
in an emergency by placing and securing (with surgical tape) a by diathermy may fill up the memory banks of the device and
• Device type—pacemaker/ICD/CRTD
• Pacing percentages—sensing (S), pacing (P), and chamber (A and V) will be given for all four combinations, i.e. ASVS, ASVP, APVS, APVP. A high
percentage of pacing will indicate a higher degree of pacemaker dependency
• Pacing threshold—the report should confirm an adequate safety margin with the output on the lead ( pacing amplitude) programmed to at
least double the pacing threshold (in volts) to ensure capture
• Lead impedance—resistance to current flow (measured in ohms) and trends. Abnormal lead impedance may suggest lead crush or fracture
(excessively high impedance) or an insulation defect (excessively low impedance)
• Sensed P/R amplitude (mV)—confirmation of appropriate sensing parameters
• Battery life—presented as a graph, figure, or battery life in years
• ICD features—therapy since last interrogation will be displayed in the form of ATP or shock therapy. Some devices will indicate the total
number of therapies that the device has delivered
• ‘Current EGM (endocardial electrogram)’—will display the current signals from the atrial and ventricular channels allowing the underlying
rate and rhythm to be seen. Some devices will indicate any episodes of AF
• ‘Alerts’—most devices will provide a summary box that highlights any clinical or device functionality issues
overwrite existing data, it is prudent to contact the cardiac physi- Where central venous access in the upper body is deemed
ology department to allow them an opportunity to download any necessary, sites should be chosen well away from the site of
relevant clinical data before planned surgery. lead implantation. Caution should be taken to avoid passing
guide wires into the heart precipitating arrhythmias or potential-
ly (in recently implanted devices) lead dislodgement. Peripheral
Intraoperative considerations
nerve stimulators are considered safe, providing that they are
Anaesthesia should be tailored to the patient’s cardiac state, ex- distant from the device and that the stimulus is not in a vector
isting comorbidities, and the intended surgical intervention. parallel to that of the pacemaker current.
Hypoxia, hypercapnia, acidosis, and electrolyte abnormalities
(especially of potassium and magnesium) should be avoided Drug and fluid considerations
since they may precipitate arrhythmias and/or interfere with
pacemaker capture. If re-programming of the device is required, Succinylcholine should be used with caution since fasciculations
this may be done by a cardiac physiologist in the anaesthetic may cause over-sensing and result in pacing inhibition;8 ICDs
room immediately before induction. If an asynchronous mode have robust sensing algorithms and inappropriate shock delivery
is deemed necessary, reliable capture at an appropriate rate is unlikely. Antibiotic prophylaxis should be carefully considered
with haemodynamic stability should be confirmed. depending on the surgery. CIED infection, predominantly
staphylococcal, is difficult to diagnose and treat with a mortality
of up to 35%.9
Management of intraoperative arrhythmias
Many patients with CIEDs will have impaired cardiac function
There is a possibility of intraoperative arrhythmias, particularly and the potential for development of malignant arrhythmias.
in susceptible patients with ICDs when the defibrillator function Monitoring and anaesthetic technique should reflect the need
has been temporarily disabled. Equipment should be immedi- to ensure myocardial optimization in these patients. Whilst
ately available for external defibrillation and/or temporary pa- negative inotropic drugs should be avoided, positive inotropes
cing. External defibrillator/pacing pads should be attached to may precipitate (catecholamine sensitive) tachyarrhythmias in
patients with CIEDs before surgery, particularly when access susceptible patients.
to the chest wall might be difficult. Pads should be placed at Fluid balance is an important consideration since patients
least 10–15 cm away from the edge of the CIED to avoid the re- with a fixed ventricular rate will be unable to respond to hypovol-
mote possibility of damage to the device or the theoretical risk aemia with an increase in heart rate; this may compromise end
of damage to the myocardium as a consequence of excess cur- organ perfusion and oxygen delivery. Cardiac output monitoring
rent flow. Arrhythmias should be treated conventionally follow- is recommended for surgery with the potential for significant
ing standard ALS procedures and using the usual recommended haemodynamic compromise.
external defibrillation energy levels. Should external pacing be-
come necessary, capture can usually be achieved at currents of Reducing EMI
∼50–100 mA using external pads.
Bipolar electrodes and engineered shielding protection have
reduced the risk of EMI; however, sources of EMI around the pa-
Monitoring
tient should be avoided. If the device has been recently checked
Patients should receive standard recommended monitoring; an and the surgical site is remote, the likelihood of malfunction is
appropriate ECG lead should be displayed that demonstrates minimal. Although mobile phones are generally considered safe,
any pacing spikes. The ECG should be monitored throughout sur- care should be taken to avoid direct placement of any phone
gery with particular attention paid to the effects of diathermy. onto the CIED since they remain a source of EMI and could poten-
Note that the monitored heart rate may be inaccurate due to dou- tially activate the ‘magnet mode’ of the device. Medical equipment
ble-counting of the pacing spike and the QRS complex. Monitor- which incorporates wireless technology, for example, some infu-
ing must also include a plethysmographic pulse measurement sion pumps, monitoring devices, and ultrasound probes, should
and display. Invasive arterial pressure monitoring will provide also be distanced from a CIED since these devices may provide a
additional beat-to-beat evidence of mechanical capture and source of EMI. Diathermy should be avoided where possible; bipo-
may add valuable information in patients with cardiac failure. lar electrical diathermy is considered safer than monopolar.
Should monopolar diathermy be required, it should be used in conditions are produced by some manufacturers. Stipulations in-
short 1–2 s bursts with 10 s pauses and (where possible) the use clude specific MRI systems, combined use of generator and leads
of the cutting rather than coagulation current. The pathway designated MRI-compatible, the pacing system in place for longer
from the diathermy to the return electrode should not pass near than 6 weeks, and specific programming of the device outside the
the CIED and the current field should be at right angles to the pa- MRI safety zone.11 Programming will be set to an asynchronous or
cing leads. Diathermy cables should be kept well away from the non-pacing mode by a cardiac physiologist depending on the
site of the implant. device and the degree of dependency.12 Device design modifica-
tions include a reduction in ferromagnetic components, a sensor
Procedure-specific considerations designed to resist the magnetic field, and robust well-insulated
circuitry.
The likely risk of EMI should be considered with reference to the
site and type of procedure to be undertaken, and the diathermy (vi) Radiation therapy
required. Broadly speaking, over-sensing as a complication
This is usually considered safe, although direct radiation of the
of EMI is unlikely for procedures where the application of
CIED should be avoided and the device may need to be re-sited
diathermy and the return electrode are below the level of the
if it is located directly within the field of radiation.10 Device func-
umbilicus.5
tion should be verified frequently in the case of radiation-
(i) Radiofrequency ablation induced reversion to a backup safety mode (usually VVI, but
manufacturer-specific).7
Radiofrequency ablation involving the continuous application
(several minutes) of radiofrequency energy via an emitting elec- (vii) Extracorporeal shock wave lithotripsy
trode to cause local heating and tissue coagulation is used in
Shock waves are acoustic pressure waves created from electro-
many kinds of surgery. The risk of prolonged current exposure
magnetic, piezoelectric, electroconductive, or electrohydraulic
leading to EMI with this technique means that it is especially
sources. The wave form comprises a compressive and tensile
important to disable the defibrillation function of ICDs. Direct
phase which generates acoustic energy for stone disintegration
contact between the ablation catheter and any CIED should be
(lithotripsy).13 Generally, the risk of CIED dysfunction is low but
avoided and the radiofrequency current path should be as far
devices should be checked within 1 month of the procedure.7
away as possible from the CIED and lead.10
There have been case reports of pacing suppression and backup
(ii) Tissue expanders in breast surgery safety mode reversion. The lithotripter should be kept at least
6 inches away from the CIED and the beam should not be focused
Tissue expanders that use magnets to orientate a needle to
near the CIED.10 Lithotripsy pulses should be timed with the ECG,
allow fluid filling to occur should be avoided. Their close prox-
and rate modulation should be deactivated.
imity to a CIED may risk magnetic switch activation with con-
version to asynchronous pacing, or failure of an ICD to detect
a tachyarrhythmia. Postoperative considerations
(iii) Electroconvulsive therapy Patients with CIEDs are ideally managed in a high dependency
recovery environment with continuous monitoring and full re-
The brief (1–2 s) electrical stimulus with electroconvulsive ther-
suscitation equipment immediately available. The defibrillator
apy (ECT) may cause pacemaker inhibition, although this is un-
function of an ICD and any rate modulator pacing function
likely to be clinically significant. Subsequent seizure activity,
which has been suspended needs reactivating by a cardiac physi-
however, may cause more prolonged over-sensing. Reprogram-
ologist after surgery. The device should be checked at the earliest
ming to an asynchronous mode may be appropriate for selected
opportunity if a magnet is used to deactivate a CIED intraopera-
patients before ECT, although this may also risk arrhythmias
tively. Any adverse incident relating to a CIED in the perioperative
when intrinsic electrical activity exists. The defibrillator function
period should be addressed by a cardiac physiologist as soon as
of an ICD should be disabled since seizure activity may trigger an
possible.
inappropriate shock either due to EMI or if there is a marked re-
active sinus tachycardia that encroaches on the ICD tachycardia
trigger rate. ECT often occurs in isolated sites which adds com- Conclusion
plexity to planning. An increasing number of patients are presenting for surgery with
(iv) Transcutaneous electric nerve stimulation varying forms of cardiac implantable electronic devices. It is
important to consider the indications for insertion of the device,
Transcutaneous impulses may cause inappropriate sensing and the degree of pacemaker dependency, the nature of the procedure
subsequent inhibition of pacemaker function or inappropriate being performed, and the likelihood of EMI. Careful preoperative
shock administration by an ICD; transcutaneous electric nerve and intraoperative preparation can help reduce complications
stimulation is therefore considered contraindicated in patients and permit timely intervention if required.
with CIEDs.