Genitourinary Ultrasound

xi
ULTRASOUND
CLINICS
Ultrasound Clin 1 (2006) xixii
Preface
Genitourinary Ultrasound
Vikram S. Dogra, MD
Department of Imaging Sciences
University of Rochester Medical Center
601 Elmwood Avenue, Box 648
Rochester NY 14642, USA
E-mail address:
vikram_dogra@urmc.rochester.edu
Vikram S: Dogra, MD
Guest Editor
The rapid advances and refinements of modern

imaging systems have provided a wide array of
diagnostic tools for practicing radiologists. Ultrasound has enjoyed considerable technical improvement as a result of improved materials, electronics,
and computer processing, as has also been the case
with CT and MR imaging. These changes have
resulted in the availability of high-resolution realtime grayscale imaging, tissue harmonic evaluation,
and color and power Doppler analysis. As a consequence of this revival of ultrasound imaging, there
has been tremendous growth in the number of
ultrasound examinations performed. The increased
popularity of ultrasound imaging has therefore
encouraged an increased demand for knowledge
pertaining to medical ultrasound. There are few
ultrasound journals currently available that address
all the issues pertaining to sonography in its totality.
There has long been a need for a new medical
periodical devoted to medical ultrasound. This new
publication, the Ultrasound Clinics, is devoted to
furthering the knowledge of ultrasound imaging
involving every aspect of sonography. This quarterly
periodical will have issues dedicated to obstetrics

and gynecology, abdominal, vascular, emergency,
genitourinary disease, and cardiac imaging. It will
also cover topics pertaining to advances in ultrasound technology and new developments.
I am honored to be the Guest Editor of this
inaugural issue of the Ultrasound Clinics together
with Deborah Rubens, who is a very close friend
and colleague.
The topics in this issue have been chosen with
particular attention to genitourinary and vascular
sonography. The review articles have been written
by very experienced radiologists who have dedicated their lives to the field of ultrasonography.
Most of the articles have a section devoted to the
pertinent anatomy and sonographic technique
where relevant to aid those readers who are new
to the practice of ultrasonography. There is a liberal
use of line drawings to explain the ultrasound concepts for easy understanding and reading. Key concepts have been stressed. This issue provides both
imaging specialists and clinicians with up-to-date
information regarding what is new, exciting, and
1556-858X/06/$ see front matter 2005 Elsevier Inc. All rights reserved.
ultrasound.theclinics.com
doi:10.1016/j.cult.2005.10.001
xii
Preface
relevant in the practice of ultrasonography as it

pertains to genitourinary and sonography.
I wish to express my sincere thanks to Shweta
Bhatt, MD, for her assistance in helping me to complete this issue on schedule. I would also like to
thank Bonnie Hami, MA, for her editorial assistance and Joseph Molter for preparation of the illustrations. I am also extremely thankful to Barton
Dudlick at Elsevier for his administrative and editorial assistance.
ULTRASOUND
CLINICS
Ultrasound Clin 1 (2006) 113
Ultrasonographic Evaluation of
Renal Infections
Srinivas Vourganti, MDa, Piyush K. Agarwal, MDa,
Donald R. Bodner, MDa,*, Vikram S. Dogra, MDb
&
&
&
Ultrasound principles
Ultrasound technique
Renal infections
Acute pyelonephritis
Ultrasonographic features of acute
pyelonephritis
Acute focal and multifocal pyelonephritis
(acute lobar nephronia)
Ultrasonographic features of acute focal
and multifocal pyelonephritis
Renal abscess
Ultrasonographic features of renal
abscess
Emphysematous pyelonephritis
Ultrasonographic findings of
emphysematous pyelonephritis
Pyonephrosis
Ultrasonographic features of
pyonephrosis
Medical ultrasonography dates back to the 1930s

when it was adapted from technology used to test
the strength of metal hulls of ships and applied to
detect brain tumors [1]. Now with ultrasound
being performed outside of the radiology suite
and in emergency departments, patient clinics, hospital rooms, and doctors offices, it compromises
approximately 25% of all imaging studies performed worldwide [2]. Ultrasonography is noninvasive, rapid, readily available, portable, and
offers no exposure to contrast or radiation. Furthermore, it is easily interpretable by physicians
&
&
Xanthogranulomatous pyelonephritis
xanthogranulomatous pyelonephritis
Renal malakoplakia
Ultrasonographic features of malakoplakia
Hydatid disease of the kidney (renal
echinococcosis)
echinococcosis
Renal tuberculosis
tuberculosis
HIV-associated nephropathy
Ultrasonographic features of HIVassociated nephropathy
Summary
References
of several different disciplines and can result in

quick diagnosis and treatment of potentially lifethreatening conditions. Some of these conditions
include severe kidney infections. This article focuses on reviewing ultrasound characteristics of
various renal infections.
Ultrasound principles
Electric waveforms are applied to piezoelectric elements in the transducer causing them to vibrate
and emit sound waves. The frequency range of
Department of Urology, Case Western Reserve University School of Medicine and University Hospitals of
Cleveland, 11100 Euclid Avenue, Cleveland, OH 44022, USA
b
Department of Imaging Sciences, University of Rochester School of Medicine, 601 Elmwood Avenue, Box 648,
Rochester, NY 14642, USA
* Corresponding author.
E-mail address: Dbodner180@aol.com (D.R. Bodner).
doi:10.1016/j.cult.2005.09.002
Vourganti et al
the sound waves emitted is above the audible human range of 20 to 20,000 Hz (cycles per second).
The sound waves generated range in frequency
from 1 to 15 MHz (1,000,000 cycles per second)
and are directed by the transducer into the body
where they are either reflected, absorbed, or refracted based on the density of the different tissues
the waves pass through. Sound passes through soft
tissue at an average velocity of 1540 m/s. As the
sound passes through tissues of differing densities,
a portion of the sound waves is reflected back to
the transducer and converted into electrical signals
that are then amplified to produce an image. The
strength of the returning sound waves or echoes is
proportional to the difference in density between
the two tissues forming the interface through which
the sound waves are traveling [3]. If the sound
waves encounter a homogenous fluid medium,
such as the fluid in a renal cyst, they are transmitted
through without interruption. As a result, no echoes are reflected back to the transducer, which produces an anechoic image [4]. Sound waves that are
strongly reflected generate strong echoes and are
visualized as bright white lines, creating a hyperechoic image.
In imaging the kidney, the highest frequency that
produces adequate tissue penetration with a good
resolution is selected. Tissue penetration is inversely related to the frequency of the transducer.
Therefore, as the frequency increases, the depth
of tissue penetration decreases. Conversely, image
resolution is directly related to the frequency of the
transducer. Therefore, as the frequency increases,
the spatial resolution of the image increases [5].
To balance these two competing factors, a 3.5- or
5-MHz transducer is used to image the kidneys.
challenging as it is located more superiorly, lacks

an acoustic window such as the liver, and is covered
by overlying gas from the stomach and small
bowel. The left kidney can be localized by positioning the patient in the right lateral decubitus position and by placing the probe in the left posterior
axillary line or in the left costovertebral angle [6].
The renal examination should include long-axis
and transverse views of the upper poles, midportions, and lower poles. The cortex and renal pelvic
regions should then be assessed. A maximum measurement of renal length should be recorded for
both kidneys. Decubitus, prone, or upright positioning may provide better images of the kidney.
When possible, renal echogenicity should be compared with the adjacent liver and spleen. The kidneys and perirenal regions should be assessed for
abnormalities. Doppler may be used to differentiate vascular from nonvascular structures [7].
The normal kidney appears elliptical in longitudinal view [Fig. 1]. The right kidney varies in length
from 8 to 14 cm, whereas the left kidney measures
7 to 12.5 cm. The kidneys are generally within 2 cm
of each other in length and are 4 to 5 cm in width
[8]. The renal cortex is homogenous and hypoechoic to the liver or spleen. The renal sinus contains the peripelvic fat; lymphatic and renal vessels;
and the collecting system, and appears as a dense,
central echogenic complex. The medulla can sometimes be differentiated from the cortex by the presence of small and round hypoechoic structures
adjacent to the renal sinus [9].
Renal infections
Acute pyelonephritis
Ultrasound technique
Patients are imaged in the supine position and a
coupling medium (eg, gel) is applied to the transducer to reduce the interference that may be introduced by air between the transducer and the skin.
Generally a 3.5-MHz transducer is used, but a
5-MHz transducer can provide high-quality images
in children or thin, adult patients. A breath-hold
may be elicited by instructing the patient to hold
their breath at maximal inspiration. This action will
displace the kidneys inferiorly by approximately
2.5 cm and may provide a better view. The right
kidney can be found by placing the transducer
along the right lateral subcostal margin in the anterior axillary line during an inspiratory breath-hold.
If the kidney cannot be imaged because of overlying bowel gas, then the probe can be moved
laterally to the midaxillary line or the posterior
axillary line. Imaging the left kidney is often more
A patient who has acute pyelonephritis will classically appear with localized complaints of flank pain
and costovertebral angle tenderness accompanied
by generalized symptoms of fever, chills, nausea,
and vomiting. In addition, these findings may
be accompanied by further lower urinary tract
symptoms, including dysuria, increased urinary frequency, and voiding urgency [10]. Laboratory abnormalities indicative of the underlying infection
can be expected, including neutrophilic leukocytosis on the complete blood count and elevated
erythrocyte sedimentation rate and serum Creactive
protein levels. If the infection is severe, it may
interfere with renal function and cause an elevation of serum creatinine [11].
Evaluation of the urine will usually demonstrate
frank pyuria with urinalysis demonstrating the presence of leukocyte esterase and nitrites and microscopic findings of numerous leukocytes and bacteria
[12]. However, sterile urine can be seen despite acute
Renal Ultrasonography
Fig. 1. Normal kidney. Longitudinal (A) and transverse (B) gray-scale sonogram of the right kidney demonstrates
a hypoechoic renal cortex (asterisk) as compared with the liver, and a central hyperechoic renal sinus. L, liver;
S, renal sinus.
pyelonephritis, especially in the setting of obstruction of the infected kidney [11]. Urine cultures,
which should be collected before starting antibiotic
therapy, will almost exclusively demonstrate ascending infection from gram-negative bacteria. Eighty
percent of infections are caused by Escherichia coli.
The remainder of cases is mostly caused by other
gram-negative organisms, including Klebsiella, Proteus, Enterobacter, Pseudomonas, Serratia, and Citrobacter. With the exception of Enterococcus faecalis and
Staphylococcus epidermidis, gram-positive bacteria are
rarely the cause of acute pyelonephritis [10].
In addition to bacterial nephritis, fungal infections of the kidney are also possible. Infections
with fungi are more commonly present in the setting of diabetes, immunosuppression, urinary obstruction, or indwelling urinary catheters [12]. Most
commonly, Candida sp such as Candida albicans
and Candida tropicalis are the causative organisms.
In addition to the Candida, other fungi such as
Torulopsis glabrata, Aspergillus sp, Cryptococcus neoformans, Zygomycetes (ie, Rhizopus, Rhizomucor, Mucor, and Absidia sp), and Histoplasma capsulatum may
cause renal infections with less frequency. Clinically, these infections present similarly to bacterial
infections. Diagnosis can be accomplished by evaluation of the urine where fungus can be found
microscopically or through fungal cultures. These
infections can cause the formation of fungal balls,
otherwise called bezoars, in the renal pelvis and
collecting system, which can contribute to obstruction [13].
Ultrasonographic features of acute

pyelonephritis
Imaging is generally not necessary for the diagnosis and treatment of acute pyelonephritis. In un-
complicated cases, ultrasound imaging will usually

find a normal-appearing kidney [6]. However, in
20% of cases, generalized renal edema attributed to
inflammation and congestion is present, which can
be detected by ultrasound evaluation. This edema
is formally defined as an overall kidney length in
excess of 15 cm or, alternatively, an affected kidney
that is at least 1.5 cm longer than the unaffected
side [11]. Dilatation of the collecting system in the
absence of appreciable obstructive cause may also
be detected by ultrasound. A proposed mechanism
of this dilatation is that bacterial endotoxins may
inhibit normal ureteric peristaltic motion, resulting
in hydroureter and hydronephrosis [10]. Parallel lucent streaks in the renal pelvis and ureter,
which are most likely caused by mucosal edema,
may also be detected on ultrasound. This finding is
the equivalent of a striated nephrogram appearance
on CT. Additionally, the renal parenchyma may
be hypoechoic or attenuated. In cases of fungal
infection, collections of air may be seen in the
bladder or collecting system as the fungus may be
gas-forming. In addition, ultrasonography may
demonstrate evidence of fungal debris in the collecting system, such as a bezoar and consequent
obstruction [Fig. 2] [13].
In relation to other modalities of renal imaging
(Tc-99m dimercaptosuccinic acid [DMSA] scintigraphy, spiral CT, and MR), ultrasound has been
found to be less sensitive and specific in the diagnosis of acute pyelonephritis [14]. In the pediatric
population, where a missed diagnosis can mean
irreversible damage, Tc-99m DMSA scintigraphy is
still considered the gold standard of imaging [15].
Improvements in ultrasound techniques by way of
power Doppler ultrasonography have resulted in
better imaging than B-mode ultrasonography alone,
but do not improve on the accuracy of Tc-99m DMSA
Vourganti et al
who have uncomplicated pyelonephritis. In addition, focal pyelonephritis commonly progresses to

sepsis [10]. Treatment is similar to other complicated cases of acute pyelonephritis, with 7 days of
parenteral antibiotics followed by a 7-day course of
oral antibiotics [11].
Ultrasonographic features of acute focal and

multifocal pyelonephritis
Fig. 2. Fungal ball. Longitudinal gray-scale sonogram

of the right kidney in an immunocompromised patient
demonstrates an echogenic mass (arrow) within a dilated calyx confirmed to be a fungus ball. L, liver.
scintigraphy [16]. Some authors suggest that although

power Doppler cannot replace Tc-99m DMSA scintigraphy because of its lack of sensitivity, a positive
power Doppler ultrasound finding can obviate the
need for further imaging [17]. In addition, when
combined with concomitant laboratory findings
such as an elevated serum Creactive protein level,
sensitivity and specificity can be improved, and results
correlate with those of DMSA findings [18]. However,
suggestions that ultrasound can serve as a replacement
to other more sensitive modalities in the detection of
acute pyelonephritis remain controversial.
In imaging acute focal pyelonephritis, it is important to differentiate it from the more severe case of
a renal abscess that requires more aggressive management. On ultrasound, the classic description of
acute focal pyelonephritis is of a sonolucent mass
that is poorly marginated with occasional lowamplitude echoes that disrupt the corticomedullary
junction [Fig. 3] [6]. The absence of a distinct wall
is a defining feature that differentiates focal nephritis from the more serious renal abscess [19]. Farmer
and colleagues suggest that an ultrasonographic appearance of increased echogenicity, rather than sonolucent masses, may also be commonly seen in focal
nephritis [20]. The ultrasound evaluation should be
complemented with a CT evaluation, which is more
sensitive in detecting focal pyelonephritis [21]. CT
findings demonstrate a lobar distribution of inflammation that appears as a wedge-shaped area of
decreased contrast enhancement on delayed images.
In more severe disease, a hypodense mass lesion can
be seen [19]. The radiologic appearance of multifocal disease is identical to focal disease except that it
is seen in more than one lobe.
Acute focal and multifocal pyelonephritis

(acute lobar nephronia)
Renal abscess
Acute focal and multifocal pyelonephritis occur

when infection is confined to a single lobe or occurs in multiple lobes, respectively, of the kidney.
More common in patients who have diabetes and
those who are immunosuppressed, these infections
will present with clinical features similar to acute
pyelonephritis. However, the patient will generally
experience more severe symptoms than patients
Before the advent of antibiotics, most abscesses in

the kidney were caused by hematogenous spread
(usually of Staphylococcus sp) from distant sites.
These renal carbuncles would be associated with a
history (often remote) of a gram-positive infection
elsewhere in the body, such as a carbuncle of the
skin [10]. With antibiotic therapy now common,
renal carbuncles are now rare, and instances of
Fig. 3. Pyelonephritis. Transverse gray-scale (A) and color flow Doppler (B) sonography of the right kidney
demonstrate two wedge-shaped areas of decreased echogenicity (arrows) in the renal cortex with absence of
color flow, consistent with multifocal pyelonephritis.
renal abscess are now primarily caused by ascending infection with enteric, aerobic, gram-negative
bacilli, including Escherichia coli, Klebsiella sp, and
Proteus sp [19]. Patients are at an increased risk for
these abscesses if they have a complicated urinary
tract infection (with stasis or obstruction), are diabetic, or are pregnant [10]. Patients will present
with fever, chills, and pain in their back and abdomen. In addition, many will have symptoms characteristic of a urinary tract infection, such as
dysuria, frequency, urgency, and suprapubic pain.
Constitutional symptoms of malaise and weight
loss may also be seen [10]. Laboratory studies
will demonstrate a leukocytosis. In nearly all renal carbuncles, and up to 30% of gram-negative
abscesses, the abscess does not involve the collecting system and urine cultures will be negative [19].
In general, any positive urine culture will match the
blood culture in the setting of an ascending gramnegative abscess. In the event of a gram-positive
renal carbuncle, the urine culture and blood cultures may isolate different organisms from one
another [10].
The management of renal abscesses is generally

dictated by their size. Small abscesses (smaller than
3 cm) are treated conservatively with observation
and parenteral antibiotics. Similar-sized lesions in
patients who are immunocompromised may be
treated more aggressively with some form of abscess
drainage. Lesions between 3 and 5 cm are often
treated with percutaneous drainage. Any abscess
larger than 5 cm usually requires surgical drainage.
Ultrasonographic features of renal abscess

Ultrasound is particularly useful in the diagnosis of
renal abscess [19]. It usually shows an enlarged
kidney with distortion of the normal renal contour
[Fig. 4]. Acutely, the abscess will appear to have
indistinct margins with edema in the surrounding
renal parenchyma. However, after convalescence,
it will appear as a fluid-filled mass with a distinct
wall. This clear margin helps to distinguish this
entity from the less severe focal nephritis. Once
identified by ultrasound, CT scanning with contrast
enhancement can better characterize these lesions.
The abscess will be seen as a round or oval pa-
Fig. 4. Renal abscess. Longitudinal (A) and transverse (B) gray-scale ultrasound of the right kidney reveal presence
of a well-defined hypoechoic lesion (A) near the superior pole, with posterior through transmission (arrow).
Corresponding power Doppler image (C) demonstrates an increased peripheral vascularity. L, Liver.
Vourganti et al
renchymal mass with decreased levels of attenuation. A ring circumscribing the lesion will form
with contrast enhancement (ring sign) because
of the increased vascularity of the abscess wall [10].
In many instances, it will be difficult to definitively
distinguish a renal abscess from a renal tumor.
In these cases, radiologic-guided drainage with
analysis of fluid can be helpful in establishing
the diagnosis.
Emphysematous pyelonephritis
Emphysematous pyelonephritis is a complication
of acute pyelonephritis in which gas-forming organisms infect renal parenchyma. It is usually
caused by E coli (70% of cases), but Klebsiella pneumoniae and Proteus mirabilis can cause emphysematous pyelonephritis with less frequency [22]. A
necrotizing infection occurs in the renal parenchyma
and perirenal tissues in which tissue is used as a substrate with carbon dioxide gas released as a byproduct.
Clinically, this infection usually occurs in patients
who have diabetes in the setting of urinary tract obstruction. Women are affected more than men. There
have been no reported cases in children. Nearly all
patients will present with the following triad of symptoms: fever, vomiting, and flank pain. Pneumaturia
can be seen when the collecting system is involved.
However, focal physical findings are commonly absent [11].
Once discovered, prompt treatment is imperative.
Management should begin with supportive care,
management of diabetes, and relief of any underlying obstruction. If infection is discovered in one
kidney, the contralateral kidney should also be
thoroughly investigated, as bilateral involvement
is seen in up to 10% of cases. The classic management for emphysematous pyelonephritis is administration of broad-spectrum antibiotics along with
emergent nephrectomy [10]. Despite this aggressive
therapy, mortality is seen in 30% to 40% of cases.
Ultrasonographic findings of emphysematous

pyelonephritis
Emphysematous pyelonephritis is diagnosed by
demonstrating gas in the renal parenchyma with
or without extension into the perirenal tissue
[10]. Ultrasound examination will characteristically show an enlarged kidney containing highamplitude echoes within the renal parenchyma,
often with low-level posterior dirty acoustic shadowing known as reverberation artifacts [Fig. 5].
However, the depth of parenchymal involvement
may be underestimated during the ultrasound
examination. Consequently, multiple renal stones
may also manifest as echogenic foci without
Fig. 5. Emphysematous pyelonephritis. Longitudinal

gray-scale sonogram of the left kidney (small arrows)
demonstrates air within the renal parenchyma with
reverberation artifact (large arrows).
clean posterior shadowing [23,24]. The isolated

presence of gas within the collecting system can be
seen after many interventional procedures and
should not be confused with emphysematous pyelonephritis. In these cases, an evaluation using CT
is always warranted and is considered the ideal
study to visualize the extent and amount of gas.
In addition, CT can identify any local destruction to
perirenal tissues. Radiologic studies play an important role in evaluation of the effectiveness of therapy in emphysematous pyelonephritis. As carbon
dioxide is rapidly absorbed, any persistence of gas
after 10 days of appropriate treatment is indicative
of failed therapy [10].
Pyonephrosis
Pyonephrosis is a suppurative infection in the setting of hydronephrosis, which occurs as the result
of obstruction. The renal pelvis and calyces become
distended with pus [6]. Patients present with fevers,
chills, and flank pain. Because of the obstruction,
bacteriuria can be absent. It is imperative that this
obstruction is relieved through a nephrostomy or
ureteral stent. If untreated, pyonephrosis can cause
destruction of renal parenchyma and irreversible
loss of renal function [10].
Ultrasonographic features of pyonephrosis

Ultrasound findings are useful in early and accurate
diagnosis of pyonephrosis. On examination, persistent echoes are seen in a dilated collecting system
[Fig. 6]. This echogenicity is caused by debris in the
collecting system, and is therefore seen in dependent areas of the collecting system. Shifts in this
debris can sometimes be appreciated if the patient
is asked to change positions during the ultrasound
examination. In addition, air can be seen in these
infections. In this event, strong echoes with acous-
Fig. 6. Pyohydronephrosis. Longitudinal (A) and transverse (B) gray-scale sonogram of the right kidney demonstrate an enlarged hydronephrotic kidney with a fluidfluid level (arrows) in the dilated calyces secondary to pus
appearing as echogenic debris. L, liver.
tic shadowing can be seen behind the affected area

of the collecting system [11].
Xanthogranulomatous pyelonephritis
Xanthogranulomatous pyelonephritis (XGP) is a
rare inflammatory condition that is seen in the setting of long-term and recurrent obstruction from
nephrolithiasis accompanied by infection. It results
in the irreversible destruction of renal parenchyma.
This damage begins in the renal pelvis and calyces
and eventually extends into the renal parenchyma
and can occur in either a diffuse or segmental pattern [11]. Though the cause of XGP is unknown,
it is thought that the inflammatory process that
occurs in response to tissue damage by bacterial
infection (usually Proteus mirabilis or E coli) results
in the deposition of lipid-laden histiocytes at the
site of infection. These macrophages, or xanthoma
cells, along with other inflammatory cells result in
the formation of fibrous tissue. This granulomatous
process eventually replaces the adjacent normal
renal parenchyma and adjacent renal tissue [10].
Clinically, XGP is seen more commonly in
women than men. Incidence peaks during the
fifth to sixth decade of life. Patients who have
diabetes are predisposed to the formation of XGP.
Symptoms include those that suggest underlying
chronic infection in the setting of obstruction,
such as fever, flank pain, persistent bacteriuria, or
history of recurrent infected nephrolithiasis. XGP
results in the irregular enlargement of the kidney
and is often misdiagnosed as a tumor. Even by
pathologic examination, XGP can closely resemble
malignancy, such as renal cell carcinoma. Definitive diagnosis is often made only after surgical
removal, which allows thorough pathologic examination. Treatment of XGP involves surgical removal
of the entire inflammatory process. Limited disease

may be amenable to partial nephrectomy; however,
more widespread XGP requires total nephrectomy
and removal of the involved adjacent tissue [11].
Though classic management suggests that conservative intervention through simple incision and drainage commonly results in further complications,
some investigators suggest this course in cases of
limited focal disease [25].
xanthogranulomatous pyelonephritis
Definitive preoperative diagnosis is extremely difficult to establish in XGP. By ultrasound evaluation,
multiple hypoechoic round masses can be seen
in the affected kidney. These masses can demonstrate internal echoes and can be abscesses (with
increased sound through-transmission) or solid
granulomatous processes (with decreased sound
through-transmission) [11]. Global enlargement
with relative preservation of the renal contour can
be seen with diffuse disease. However, in focal or
segmental XGP a mass-like lesion may be appreciated. In addition, evidence of obstruction and
renal calculus is commonly seen (85%) [26]. In
general, CT evaluation is considered more informative than ultrasound in describing XGP. A large
reniform mass within the renal pelvis tightly surrounding a central calcification is seen on CT imaging [10]. Dilated calyces and abscesses that replace
normal renal parenchyma will appear as waterdensity masses. Calcifications and low attenuation
areas attributed to lipid-rich xanthogranulomatous
tissue may be seen within the masses [27]. If contrast is used, a blush is seen in the walls of these
masses because of their vascularity. This enhancement, which is limited to the mass wall only, will
Vourganti et al
help distinguish XGP from renal tumors and other

inflammatory processes that do enhance throughout [11].
Renal malakoplakia
Renal malakoplakia is a rare inflammatory disorder
associated with a chronic coliform gram-negative
urinary tract infection (usually E coli) resulting
in the deposition of soft, yellow-brown plaques
within the bladder and upper urinary tract. The
cause is thought to be abnormal macrophage function that causes incomplete intracellular bacterial
lysis. This lysis results in the deposition of histiocytes, called von Hansemann cells, that are filled
with these bacteria and bacterial fragments. The
bacteria form a nidus for calcium phosphate crystals, which form small basophilic bodies called
Michaelis-Gutmann bodies [10].
Clinically, malakoplakia of the urinary tract usually occurs in women. Most patients are older than
50 years. There is often an underlying condition
compromising the immune system, such as diabetes, immunosuppression, or the presence of a
chronic debilitating disease. Symptoms of a urinary
tract infection may be present, such as fever, irritative voiding symptoms, and flank pain. In addition,
a palpable mass may be appreciated [11]. If the
disease involves the bladder, symptoms of bladder
irritability and hematuria may be seen.
Ultrasonographic features of malakoplakia

Imaging findings of malakoplakia are nonspecific
and can often mimic other pathology, such as renal
tumors [28]. The most common ultrasonographic
feature of renal malakoplakia is diffuse enlargement
of the affected kidney [29]. Increased echogenicity
of the renal parenchyma can be seen because of
a confluence of the plaques [10]. In addition, hypoechoic lesions and distortion of parenchymal echoes may be appreciated [29].
Hydatid disease of the kidney (renal

echinococcosis)
Echinococcosis is a parasitic infection that is most
commonly seen in South Africa, the Mediterranean,
Eastern Europe, Australia, and New Zealand. It
is caused by the tapeworm Echinococcus granulosis. Although the adult form is zoonotic, mostly
found in the intestines of dogs, humans may serve
as an intermediate host of this parasite while it is
in the larval stage [11]. Infection more commonly
manifests in the liver and lungs, with only 4% of
echinococcosis involving the kidney, because the
larvae, which originally invade the body through
the gastrointestinal tract, must first escape sequestration in the liver and subsequently the lungs. Only
after these two defenses are surpassed are the larvae
able to gain widespread access to the systemic circulation, and correspondingly, the kidneys [30].
The offending lesion will most commonly form
as a solitary mass in the renal cortex. It is divided
into three distinct zones. The outermost adventitial
layer consists of host fibroblasts that may become
calcified. A middle laminated layer consists of hyaline that surrounds a third inner germinal layer.
The germinal layer is composed of nucleated epithelium and is where the echinococcal larvae reproduce. The larvae attach to the surrounding germinal
layer and form brood capsules. These brood capsules grow in size and will remain connected to the
germinal layer by a pedicle for nutrition. The core
of this hydatid cyst contains detached brood capsules (daughter cysts), free larvae, and fluid, a combination known as hydatid sand [10].
Clinically, most patients who have renal echinococcosis are asymptomatic, especially in the beginning stages of the disease process because the cyst
starts small and grows at a rate of only 1 cm annually. Because of their focal nature, small hydatid
cysts will rarely affect renal function. As the lesion
progresses, a mass effect will contribute to symptoms of dull flank pain, hematuria, and a palpable
mass on examination [10]. If the cyst ruptures, a
strong antigenic immune response ensues with possible urticaria and even anaphylaxis [30]. If a cyst
ruptures into the collecting system, the patient will
develop symptoms of hydatiduria, including renal colic and passage of urinary debris resembling
grape skins [11].
Treatment of echinococcal disease in the kidney
is primarily surgical. Medical therapy with antiparasitic agents, such as mebendazole, has been shown
to be largely unsuccessful. In removing a cyst, great
care must be taken to avoid its rupture. Any release
of cyst contents can contribute to anaphylaxis. In
addition, the release of the larvae can result in the
dissemination of the disease. In the event of rupture, or if resection of the entire cyst is not possible,
careful aspiration of the cyst is indicated. After the
contents of the cyst are removed, an infusion of
an antiparasitic agent (eg, 30% sodium chloride,
0.5% silver nitrate, 2% formalin, or 1% iodine) is
reinfused into the cyst [11].

echinococcosis
Ultrasonographic findings of echinococcosis demonstrate different findings based on the age, extent,
and complications of the hydatid cyst [30]. These
lesions can be classified by the Gharbi ultrasono-
Table 1: Gharbi ultrasonographic classification of hyatid cysts

Type
Pathology
I
II
Discrete univesicular mass

Univesicular mass with
detached membranes
Multivesicular mass
Heterogenous mass
22%
4%
Heterogenous mass with

calcifications
8%
III
IV
V
Frequency
54%
12%
graphic classification [Table 1] [30]. The Gharbi

classification assists in the characterization of
renal masses that are caused by hydatid disease.
Higher Gharbi type corresponds with further disease progression. Consequently, Gharbi type I cysts
are most commonly seen in children. Accordingly,
Gharbi types III through V are consistent with more
advanced disease and are seen almost exclusively in
adults. Most common are the Gharbi type III cysts,
which are multivesicular masses that can be de-
Ultrasonographic findings
Liquid-filled cyst with parietal echo backing
Liquid-filled cyst with ultrasonographic water
lily sign
Partitioned cyst with a spoke wheel appearance
Heterogeneous echo structure with mixed solid
and liquid components
Dense reflections with a posterior shade cone
caused by calcifications
tected on ultrasound as a partitioned cyst with a

spoke wheel appearance [Fig. 7A, B] [31]. Changes
in patient position can cause any hydatid sand that
is present to be disturbed and will result in the
shifting of bright echoes within the mass. This
finding has been described as the snowstorm sign
[11,32]. Less commonly seen are the univesicular
Gharbi type I and type II cysts, which demonstrate
less disease progression and are seen more commonly in young adults and children. Type I cysts
Fig. 7. Renal hydatid cyst. Gray-scale ultrasound (A) and contrast-enhanced CT scan (B) of the right kidney reveal a
well-defined cystic lesion (large arrow) with multiple internal septae (small arrows) suggestive of a hydatid cyst
with multiple daughter cysts. (Courtesy of SA Merchant, India.) (C) Gray-scale sonography of the right kidney on
a different patient demonstrates the floating membranes (arrowheads) of the hydatid cyst following rupture of
the cyst, referred to as the water lily sign. (Courtesy of Ercan Kocakoc, Turkey.)
10
Vourganti et al
are well-limited liquid cysts that can be differentiated from simple nonhydatid cysts by the presence of a parietal echo. Gharbi type II cysts
demonstrate a detached and floating membrane
that is pathognomonic for hydatid disease. This
detachment of the membranes inside the cyst has
been referred to as the ultrasound water lily sign
because of its resemblance to the radiographic
water lily sign seen in pulmonary cysts [Fig. 7C]
[33,34]. In contrast, the Gharbi type IV and V cysts
demonstrate more advanced disease and are correspondingly seen in older patients. Gharbi type IV
hydatid cysts will demonstrate heterogeneity of
echo structure with a combination of liquid and
solid cyst contents. Gharbi type V hydatid cysts are
calcified and will show dense reflections with a
posterior shade cone. The varying echogenic aspects
of these type IV and V lesions make diagnosis by
ultrasound more difficult [30]. In these cases, CT
studies can aid in characterization. On CT, the
presence of smaller round daughter cysts within
the mother cysts can help differentiate hydatid
lesions from other similar appearing pathology,

such as simple cysts, abscesses, and necrotic neoplasm [10].
Renal tuberculosis
Tuberculosis is an infection caused by Mycobacterium tuberculosis. Typically acquired by inhalation,
exposure initially results in a primary infection
with a silent bacillemia. This infection will result
in systemic dissemination of mycobacteria. Latent
foci may result in kidney lesions many years following primary infection, though only 5% of patients who have active tuberculosis will have
cavitary lesions in the urinary tract [11].
Clinically, this infection presents in younger patients, with 75% of those affected being younger
than 50 years. Renal tuberculosis should be considered in any patient who has a diagnosed history
of tuberculosis. Often patients will present asymptomatically, even in cases of advanced disease. If
disease involves the bladder, symptoms of urinary
Fig. 8. Renal tuberculosis. (A) Longitudinal gray-scale ultrasound of the right kidney demonstrates hypoechoic
areas (arrows) in the renal cortex suggestive of lobar caseation in this known case of tuberculosis. Longitudinal
gray-scale sonography (B, C) of the kidney in another patient who has renal tuberculosis demonstrates hypoechoic
areas of caseous necrosis (large arrows) with dense peripheral calcification (small arrows) with posterior acoustic
shadowing. (Panels B, C, Courtesy of SA Merchant, Mumbai, India.)
Fig. 9. HIV nephropathy. Longitudinal (A) and transverse (B) gray-scale sonograms of the right kidney in young
man who has no known history of medical disease reveals an enlarged, markedly echogenic kidney (bilateral;
left not shown) with loss of corticomedullary differentiation and obliteration of sinus fat suggestive of HIVnephropathy. Subsequently confirmed by histopathology. L, liver.
frequency may result. One quarter of patients will

present with findings of a unilateral poorly functioning kidney. Other suspicious findings include
chronic cystitis or epididymitis that is recalcitrant to
treatment; firm seminal vesicles on digital rectal
examination; or a chronic fistula tract that forms
at surgical sites. Diagnosis of urinary tract tuberculosis can be established through a urine culture
that demonstrates growth of M tuberculosis.
HIV-associated nephropathy

tuberculosis
Ultrasonographic features of HIV-associated

nephropathy
Early findings of urinary tract tuberculosis are best

characterized by intravenous urography. Initially,
cavities appear as small irregularities of the minor
calyces. These irregular changes are classically described as feathery and moth-eaten. As disease
progresses, it extends from the calyces into the
underlying renal parenchyma. Calcifications may
be appreciated in these areas of caseating necrosis.
In addition, tuberculosis involvement of the ureter
can result in ureteral strictures, which cause a urographic appearance of a rigid, irregular, pipe-stem
ureter [11]. Ultrasound findings in the diagnosis of
renal tuberculosis have traditionally been described
as limited. However, recent reports describe the role
of high-resolution ultrasonography in characterizing late and chronic changes in renal tuberculosis
[35]. Granulomatous mass lesions in the renal parenchyma can be seen as masses of mixed echogenicity, with or without necrotic areas of caseation
and calcifications [Fig. 8]. Mucosal thickening and
stenosis of the calyces is detectable by ultrasonography. In addition, findings of mucosal thickening
of the renal pelvis and ureter, ureteral stricture, and
hydronephrosis are seen. Finally, bladder changes
such as mucosal thickening and reduced capacity
are commonly detectable.
Sonography is a critical component in the evaluation of HIVAN. The major sonographic findings
include increased cortical echogenicity, decreased
corticomedullary definition, and decreased renal sinus fat [Fig. 9]. Renal size may be enlarged [36,37].
The increased cortical echogenicity is attributable to
prominent interstitial expansion by cellular infiltrate and markedly dilated tubules containing voluminous casts. Histologically, HIVAN demonstrates
tubular epithelial cell damage, glomerulosclerosis,
and tubulointerstitial scarring [38]. Most patients
who have HIVAN have proteinuria secondary to
tubular epithelial cell damage. In the presence of
marked increased cortical echogenicity in a young
patient who has known history of medical renal
disease, HIVAN must be considered.
Renal disease is a common complication in patients

who have HIV. This complication can result primarily from direct kidney infection with HIV or secondarily from adverse effects of the medications used
to treat HIV. HIV-associated nephropathy (HIVAN)
accounts for approximately 10% of new end stage
renal disease cases in the United States. Patients
who have HIVAN are not typically hypertensive.
Summary
The growing ubiquity, well-established safety, and
cost-effectiveness of ultrasound imaging have cemented its role in the diagnosis of renal infectious
diseases. It is imperative that all practitioners of
renal medicine understand the ultrasonographic
manifestations of these diseases, as early diagnosis
and treatment are the cornerstones of avoidance of
11
12
Vourganti et al
long-term morbidity and mortality. If the strengths

and limitations of ultrasonography are understood
properly, a practitioner will be able to achieve the
quickest and safest diagnosis with the minimal
amount of further invasive imaging. The advent
of new ultrasonographic techniques may allow it
to serve a more central role in the diagnosis and
characterization of renal infections.
[15]
[16]
[17]
References
[1] Newman PG, Rozycki GS. The history of ultrasound. Surg Clin North Am 1998;78(2):17995.
[2] Harvey CJ, Pilcher JM, Eckersley RJ, et al. Advances in ultrasound. Clin Radiol 2002;57(3):
15777.
[3] McAchran SE, Dogra VS, Resnick MI. Office based
ultrasound for urologists. Part I: ultrasound
physics, and of the kidney and bladder. AUA
Update 2004;23:22631.
[4] Spirnak JP, Resnick MI. Ultrasound. In: Gillenwater JY, Grayhack JT, Howards SS, et al, editors.
Adult & pediatric urology. 4th edition. Philadelphia: Lippincott, Williams & Williams; 2002.
p. 16593.
[5] Smith RS, Fry WR. Ultrasound instrumentation.
Surg Clin North Am 2004;84(4):95371.
[6] Noble VE, Brown DF. Renal ultrasound. Emerg
Med Clin North Am 2004;22(3):64159.
[7] Grant EG, Barr LL, Borgstede J, et al. AIUM standard for the performance of an ultrasound examination of the abdomen or retroperitoneum.
American Institute of Ultrasound in Medicine.
J Ultrasound Med 2002;21(10):11827.
[8] Brandt TD, Neiman HL, Dragowski MJ, et al.
Ultrasound assessment of normal renal dimensions. J Ultrasound Med 1982;1(2):4952.
[9] Horstman W, Watson L. Ultrasound of the genitourinary tract. In: Resnick MI, Older RA, editors.
Diagnosis of genitourinary disease. 2nd edition.
New York: Thieme; 1997. p. 79130.
[10] Schaeffer AJ. Infections of the urinary tract. In:
Walsh PC, Retik AB, Vaughn ED, et al, editors.
Campbells urology. 8th edition. Philadelphia:
Elsevier; 2002. p. 516602.
[11] Schaeffer AJ. Urinary tract infections. In: Gillenwater JY, Grayhack JT, Howards SS, et al, editors.
Adult & pediatric urology. 4th edition. Philadelphia: Lippincott, Williams & Williams; 2002.
p. 289351.
[12] Ramakrishnan K, Scheid DC. Diagnosis and
management of acute pyelonephritis in adults.
Am Fam Physician 2005;71(5):93342.
[13] Wise G. Fungal and actinomycotic infections
of the genitourinary system. In: Walsh PC, Retik
AB, Vaughn ED, et al, editors. Campbells urology. 8th edition. Philadelphia: Elsevier; 2002.
p. 797827.
[14] Majd M, Nussbaum Blask AR, Markle BM, et al.
Acute pyelonephritis: comparison of diagnosis
with 99mTc-DMSA, SPECT, spiral CT, MR imag-
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
ing, and power Doppler US in an experimental

pig model. Radiology 2001;218(1):1018.
Johansen TE. The role of imaging in urinary tract
infections. World J Urol 2004;22(5):3928.
Berro Y, Baratte B, Seryer D, et al. Comparison
between scintigraphy, B-mode, and power Doppler sonography in acute pyelonephritis in children. J Radiol 2000;81(5):5237.
Bykov S, Chervinsky L, Smolkin V, et al. Power
Doppler sonography versus Tc-99m DMSA scintigraphy for diagnosing acute pyelonephritis in
children: are these two methods comparable?
Clin Nucl Med 2003;28(3):198203.
Wang YT, Chiu NT, Chen MJ, et al. Correlation
of renal ultrasonographic findings with inflammatory volume from dimercaptosuccinic acid renal scans in children with acute pyelonephritis.
J Urol 2005;173(1):1904.
Dembry LM, Andriole VT. Renal and perirenal
abscesses. Infect Dis Clin North Am 1997;11(3):
66380.
Farmer KD, Gellett LR, Dubbins PA. The sonographic appearance of acute focal pyelonephritis 8 years experience. Clin Radiol 2002;57(6):
4837.
Cheng CH, Tsau YK, Hsu SY, et al. Effective
ultrasonographic predictor for the diagnosis of
acute lobar nephronia. Pediatr Infect Dis J 2004;
23(1):114.
Stone SC, Mallon WK, Childs JM, et al. Emphysematous pyelonephritis: clues to rapid diagnosis in the Emergency Department. J Emerg Med
2005;28(3):3159.
Narlawar RS, Raut AA, Nagar A, et al. Imaging
features and guided drainage in emphysematous
pyelonephritis: a study of 11 cases. Clin Radiol
2004;59(2):1927.
Best CD, Terris MK, Tacker JR, et al. linical and
radiological findings in patients with gas forming
renal abscess treated conservatively. J Urol 1999;
162(4):12736.
Bingol-Kologlu M, Ciftci AO, Senocak ME, et al.
Xanthogranulomatous pyelonephritis in children: diagnostic and therapeutic aspects. Eur J
Pediatr Surg 2002;12(1):428.
Tiu CM, Chou YH, Chiou HJ, et al. Sonographic
features of xanthogranulomatous pyelonephritis.
J Clin Ultrasound 2001;29(5):27985.
Kim JC. US and CT findings of xanthogranulomatous pyelonephritis. Clin Imaging 2001;25(2):
11821.
Evans NL, French J, Rose MB. Renal malacoplakia: an important consideration in the differential diagnosis of renal masses in the presence
of Escherichia coli infection. Br J Radiol 1998;
71(850):10835.
Venkatesh SK, Mehrotra N, Gujral RB. Sonographic findings in renal parenchymal malacoplakia. J Clin Ultrasound 2000;28(7):3537.
Zmerli S, Ayed M, Horchani A, et al. Hydatid cyst
of the kidney: diagnosis and treatment. World J
Surg 2001;25(1):6874.
[31] von Sinner WN. New diagnostic signs in hydatid

disease; radiography, ultrasound, CT and MRI
correlated to pathology. Eur J Radiol 1991;12(2):
1509.
[32] Marti-Bonmati L, Menor Serrano F. Complications of hepatic hydatid cysts: ultrasound, computed tomography, and magnetic resonance
diagnosis. Gastrointest Radiol 1990;15(2):11925.
[33] Beggs I. The radiology of hydatid disease. AJR Am
J Roentgenol 1985;145(3):63948.
[34] Moguillanski SJ, Gimenez CR, Villavicencio RL.
Radiologa de la hidatidosis abdominal. In:
Stoopen ME, Kimura K, Ros PR, editors. Radiologa e imagen diagnstica y terapetica: abdomen, Vol. 2. Philadelphia: Lippincott Williams &
Wilkins; 1999. p. 4772.
[35] Vijayaraghavan SB, Kandasamy SV, Arul M, et al.

Spectrum of high-resolution sonographic features of urinary tuberculosis. J Ultrasound Med
2004;23(5):58594.
[36] Di Fiori JL, Rodrigue D, Kaptein EM, et al. Diagnostic sonography of HIV-associated nephropathy: new observations and clinical correlation.
AJR Am J Roentgenol 1998;171(3):7136.
[37] Atta MG, Longenecker JC, Fine DM, et al.
Sonography as a predictor of human immunodeficiency virus-associated nephropathy.
[38] Hamper UM, Goldblum LE, Hutchins GM,
et al. Renal involvement in AIDS: sonographicpathologic correlation. AJR Am J Roentgenol 1988;
150(6):13215.
13
15
ULTRASOUND
CLINICS
Sonography of Benign Renal

Cystic Disease
Therese M. Weber,
&
&
&
&
&
&
&
&
MD
Simple cortical cysts

Complex renal cysts
Renal sinus cysts
Medullary cystic disease
Multiple renal cysts
Autosomal dominant polycystic kidney
disease
Autosomal recessive polycystic kidney
disease
Von Hippel-Lindau disease
When evaluating renal masses, differentiating

cysts from solid lesions is the primary role of ultrasound (US). US is also helpful and frequently
superior to CT, in demonstrating the complex internal architecture of cystic lesions in terms of internal fluid content, septations, tiny nodules, and
wall abnormalities, including associated soft tissue
masses. Renal cysts are common in the population
older than 50 years, occurring in at least 50% of
people [1]. Scanning technique is important to the
success of demonstrating renal masses with US. The
kidneys should be evaluated in multiple patient
positions, including supine, lateral decubitus, and
occasionally oblique or prone positions. The mass
should be scanned with an appropriate focal zone.
Simple renal cysts will frequently be better demonstrated with tissue harmonic imaging, which can
eliminate low-level internal echoes by reducing
background noise [2]. The Bosniak Classification
System of renal cysts, shown in Box 1, has become
an important tool used by radiologists and urologists to communicate the significance of renal cyst
imaging characteristics [3,4]. The primary goal of
the radiologist in evaluating cystic renal masses is
&
&
&
&
&
&
&
&
Tuberous sclerosis
Acquired cystic kidney disease associated
with dialysis
Multiloculated cystic renal masses
Multicystic dysplastic kidney
Multilocular cystic nephroma
Summary
Acknowledgments
References
the differentiation of nonsurgical from surgical lesions [5].
Simple cortical cysts

The sonographic criteria used to diagnose a simple
cyst include the following characteristics: internally
anechoic, posterior acoustic enhancement, and a
sharply defined, imperceptible, smooth far wall.
Simple cysts are usually round or ovoid in shape.
If all these sonographic criteria are met, further
evaluation or follow-up is not required. Maintaining rigid criteria is necessary to ensure the highest
possible accuracy with US [6]. The simple renal cyst
is a Bosniak category I cyst [Fig. 1] [3].
Complex renal cysts

Complex cysts do not meet the strict US criteria of a
simple renal cyst. Five to ten percent of all renal
cysts are not simple cysts, and 5% to 10% of renal
cysts with complex features prove to be tumors [7].
These complex renal cysts may contain complex
fluid, septations, calcification, perceptible defined
Department of Radiology, Wake Forest University School of Medicine, Medical Center Boulevard,
Winston-Salem, NC 27157-1088, USA
E-mail address: tweber@wfubmc.edu
doi:10.1016/j.cult.2005.09.005
16
Weber
Box 1: Bosniak classification of renal cystic

disease
Category I
Simple benign cyst by imaging criteria
Category II
Cystic lesions characterized by:
One or two thin ( 1 mm thick) septations or

thin, fine calcification in the wall or septa
Hyperdence, homogeneous benign cysts
Diameter of 3 cm or less
One quarter of its wall extending outside
the kidney so the wall can be assessed
No contrast enhancement
Category IIF
Cystic lesions that are minimally complicated
that need follow-up. Some suspicious features
in these lesions need follow-up to detect any
change in character.
Category III
True indeterminate cystic masses that need
surgical evaluation, although many prove to be
benign, characterized by:
Uniform wall thickening or nodularity

Thick or irregular peripheral calcification
Multilocular nature with multiple enhancing
septa
Hyperdense lesions that do not meet category II criteria
Category IV
Lesions with findings that are clearly malignant, including:
Nonuniform or enhancing thick wall

Enhancing or large nodules in the wall
Clearly solid components in the cystic lesions
Irregular margins
wall, or mural nodularity. Bosniak category II cysts

are minimally complicated cysts that may contain
septations or thin calcification, and include infected cysts and high-density cysts seen on noncontrast CT examination. High-density cysts are
defined as having a density greater than 20 Hounsfield units (HU) on noncontrast CT. US may confirm the cystic nature in 50% of these lesions,
especially if larger in size (1.53.0 cm). Equivocal
cases can be placed in category IIF, created for
minimally complicated cysts that require followup. Size criterion of 3 cm differentiates between II
and IIF lesions. These cysts need at least 6-month
follow-up and are most likely benign but somewhat suspicious [8]. Many of these cysts, depending
on the degree of abnormality, will require further

imaging with MR or CT.
Internal echoes within a renal cyst may be caused
by a complicating hemorrhage or infection. Infected cysts on US usually have a thickened wall
and may exhibit a debris-fluid or gas-fluid level.
Aspiration and drainage may be required for diagnosis and treatment. Hemorrhagic cysts can usually
be followed with serial US examination if there
is no evidence of malignancy at MR or CT evaluation. Septations within a cyst may occur following
hemorrhage, infection, or percutaneous aspiration
[Fig. 2]. Two adjacent cysts may share a common
wall and mimic a large septated cyst. There may be
a small amount of renal parenchyma between two
adjacent cysts, suggesting a thick septation. If septa
are paper thin (1 mm), smooth, and attached to
the cyst wall without focal thickening or nodularity,
a benign cyst can be diagnosed [3,8]. Fewer than
five septations should be present in a benign cyst. If
the cystic lesion shows septal irregularity, septal
thickness greater than 1 mm, thick calcifications,
or solid elements at the wall attachment, the lesion
must be presumed malignant and requires further
evaluation with MR or CT.
Calcifications in renal cysts may be fine and
linear, or amorphous and thick. The presence
of a small amount of calcium or thin, fine areas
of calcification in the wall or a septation, without an associated soft tissue nodule, likely represents a complicated cyst rather than malignancy.
Thick, irregular, amorphous calcification had been
thought to be more concerning for malignancy;
however, a recent report demonstrates that calcification in a cystic renal mass is not as important in
the diagnosis as the presence of enhancing softtissue elements [9]. Renal cysts with milk-ofcalcium will show layering, echogenic, dependent
material that may be mobile. These cysts are always
benign. At real-time US, bright echogenic foci with
ringdown may be seen on septa and in the cyst
walls; however, calcification may not be of sufficient density to be identified on CT examination.
Bosniak category III lesions are moderately complicated cysts that have more numerous or thickened septae; a thickened wall; thicker, irregular
calcification; or multiloculated features. Benign
and malignant tumors may be classified as Bosniak
category III lesions. Because most category III lesions will be malignant [10], all category III lesions
require surgical resection or biopsy. Benign lesions
in this category include the multilocular cystic
nephroma; complex septated and multiloculated
cysts; densely calcified cysts; a complex abscess; an
infected cyst; vascular malformation; xanthogranulomatous pyelonephritis (XGP); or echinococcosis.
Malignant lesions in this category include cystic
Sonography of Benign Renal Cystic Disease
Fig. 1. Simple cyst in the lower renal pole in longitudinal (A) and transverse (B) planes demonstrating characteristic
sonographic findings.
Wilms tumor in a child and cystic adenocarcinoma

in an adult. Bosniak category IV lesions are cystic or
necrotic carcinomas that may have solid elements,
irregular margins, and enhancement demonstrated
on CT, MRI, or US with contrast imaging. These
lesions are unequivocally malignant and require
treatment. Renal cystic lesions with perceptible,
defined, thickened wall; thick septae; mural nodularity; intracystic soft tissue mass; or abnormal calcifications on US are concerning for malignancy.
that include multiple, small, confluent cysts versus

a single, larger cyst that probably arises from the
adjacent parenchyma [11]. On US, these are simple
cysts located in the medullary or renal sinus area
of the kidney. Multiple renal sinus cysts mimic
hydronephrosis on US and noncontrasted CT.
If communication with the collecting system cannot be demonstrated with US to confirm hydronephrosis, additional studies such as intravenous
pyelogram, contrasted CT, or MR examination
may be needed.
Renal sinus cysts

Renal sinus cysts (RSCs) are common and have
been described as peripelvic cysts, parapelvic lymphatic cysts, parapelvic lymphangiectasia, and parapelvic cysts [11]. RSCs are likely lymphatic in origin
or develop from embryologic rests. These cysts do
not communicate with the collecting system [12].
Most RSCs are asymptomatic, but they may become
infected or bleed, and may cause hematuria, hypertension, or hydronephrosis [13]. There are two distinct patterns of cyst formation in the renal sinus
Medullary cystic disease

Medullary cystic disease is a group of similar diseases that occur as a result of progressive renal
tubular atrophy with secondary glomerular sclerosis and medullary cystic formation [14,15].
Medullary cystic disease is an important cause of
end-stage renal disease (ESRD) in children, accounting for 10% to 25% of the cases of ESRD.
Two distinct presentations occur. The childhood
form, which is most common, is inherited as an
Fig. 2. (A) Mildly complicated renal cyst with thin septation (white arrow). (B) Artifact (white arrowheads) from the
septation is frequently seen at real-time sonography.
17
18
Weber
autosomal recessive disorder. There is an association with extrarenal, ophthalmologic abnormalities. The adult form, inherited as an autosomal
dominant pattern, tends to present in early adulthood and is not associated with extrarenal abnormalities. US findings include kidneys that are small
to normal in size, are hyperechoic, and have small
(0.1 to 1.0 cm) cysts in the medulla and at the
corticomedullary junction [16,17]. Acquired cystic
kidney disease may resemble medullary cystic disease; however, cyst location in the cortex and a history of dialysis supports the diagnosis of acquired
cystic kidney disease.
Multiple renal cysts

With increased use of CT, US, and MR, patients
who are harboring multiple simple cysts or lesions
too small to characterize are being seen with increasing frequency. Multiple renal cysts can be seen
in polycystic kidney disease, Von Hippel-Lindau
disease (VHL), tuberous sclerosis (TS), acquired
cystic kidney disease associated with dialysis
(ACKDD), and mulitcystic dysplastic kidney.
Autosomal dominant polycystic kidney

disease
Autosomal dominant polycystic kidney disease
(ADPKD) is the third most common systemic hereditary condition and accounts for 10% to 15%
of all patients on dialysis [18]. Because the disease
is characterized by variable expression and occurs
with spontaneous mutation, up to 50% of patients
will have no family history of the disease [19].
Renal failure develops in 50% of patients and is
usually present by 60 years of age. Early signs of
the disease include hypertension and flank or back
pain with variable progression to ESRD. Nephrolithiasis occurs in 20% to 36% of patients because
of metabolic (lower glomerular filtration rate and

urine volume) and anatomic (associated with larger
and more numerous renal cysts) factors [20]. Extrarenal manifestations of ADPKD include hepatic,
pancreatic, ovarian, splenic, arachnoid, and other
cysts, and intracranial berry aneurysms with associated intracranial hemorrhage, abdominal aortic
aneurysm, cardiac valve abnormalities, and hernias. The vascular abnormalities and cyst development are related to a basement membrane defect
[18]. Eighty percent of people who have ADPKD
and ESRD will have colonic diverticulosis. At least
two different genes involving chromosomes 4
and 16 have been found to be associated with
ADPKD [21].
US, because of high sensitivity and low cost, has
become the primary method of diagnosing ADPKD
and following the cysts. US screening for ADPKD
typically begins between ages 10 and 15 years, but
has the problem of false negatives in about 14% of
patients younger than 30 years. Bear and colleagues
[22] developed criteria that are widely used to diagnose ADPKD. In adults who have a family history
of ADPKD, the presence of at least three cysts in
both kidneys, with at least one cyst in each kidney,
is a positive finding [23]. The cysts tend to involve
all portions of the kidney and are of variable size.
US typically reveals kidneys that are bilaterally
enlarged with compression of the central sinus
echo complex [Fig. 3]. When the kidneys are markedly enlarged with multiple complex cysts, detection of solid lesions may be difficult, and
correlation with MR may be necessary to evaluate for renal cell carcinoma (RCC). When a solid
mass is seen it can be confirmed with confidence
sonographically [Fig. 4]. Nephrolithiasis may be
difficult to demonstrate sonographically because
of distortion of the collecting system by numerous
large cysts. There is no increased risk for RCC in
patients who have ADPKD, except for increased risk
Fig. 3. Bilaterally enlarged kidneys (A, B) with multiple cysts of various sizes in adult polycystic kidney disease. The
kidneys measure 15.6 cm in length bilaterally.
Fig. 4. (A) Adult polycystic kidney disease with (B) solid renal masses (white arrows) in the right kidney confirmed
as papillary renal cell carcinoma following nephrectomy.
related to dialysis and the generally increased risk

for RCC in men.
Autosomal recessive polycystic kidney

disease
Autosomal recessive polycystic kidney disease
(ARPKD) is characterized by varying degrees of
renal failure and portal hypertension as a result
of dilatation of renal collecting tubules, dilatation
of biliary radicals, and periportal fibrosis that influences its presentation. The involved gene is located
on chromosome 6 [24]. In its severest form, renal
disease predominates and ARPKD manifests itself
immediately after birth with the early complication
of severe pulmonary failure. The diagnosis may be
made at fetal or neonatal US. Characteristically, the
kidneys are enlarged bilaterally, and oligohydramnios, Potters facies, and pulmonary complications
may be encountered. Children who present later
will have some element of renal impairment with
the complications of congenital hepatic fibrosis,
including portal hypertension, splenomegaly, and
bleeding varices [25]. With US, the kidneys maintain a reniform shape, but may be normal to bilaterally enlarged and echogenic [Fig. 5].
Von Hippel-Lindau disease

VHL disease is an uncommon condition characterized by multiple lesions, including hemangioblastomas in the central nervous system (CNS) and
retina; RCC; pheochromocytomas; pancreatic neuroendocrine tumors; epididymal cystadenomas; endolymphatic sac tumors; carcinoid tumors; and
multiple cysts of the kidney, pancreas, and epididymis. The condition is inherited in an autosomal
dominant pattern [26]. The VHL tumor suppressor
gene is located on chromosome 3 [27]. Inactivation
of the VHL suppressor gene gives rise to premalignant renal cysts. Additional genetic alterations are
probably required for conversion of these cysts to
RCC. Renal involvement in VHL includes multiple,
bilateral renal cysts and solid tumors. RCC, one of
the major complications of VHL disease, occurs in
up to 75% of patients by the age of 60 and occurs
25 years earlier than sporadic RCC [28]. In addition
to the development of RCC at a younger age, these
tumors tend to be bilateral and multicentric. Cysts,
as the most common renal manifestation of the
disease, are seen in 75% of patients. Although multiple, these cysts are generally fewer than in patients
who have ADPKD. In ADPKD, pancreatic cysts almost
Fig. 5. Newborn who has bilaterally enlarged, echogenic kidneys consistent with ARPKD. The right kidney
(A) measures 10 cm in length and the left kidney (B) measures 9.8 cm in length.
19
20
Weber
never occur in the absence of liver cysts, whereas in

VHL they are common. On US the cysts tend to be
cortical in location and range in size from 0.5 to
3.0 cm. RCC in VHL disease may show solid renal
masses without cystic components or predominantly cystic lesions with enhancing solid components that contain the RCC. US may be used for
screening these patients, especially those younger
than 18 years; however, CT and MR are more sensitive in detecting small enhancing solid masses [18].
Tuberous sclerosis
TS is a neurophakomatosis involving the skin and
CNS that most commonly occurs sporadically in
60% of patients. This condition is also inherited as
an autosomal dominant pattern and is characterized by hamartomatous growths in the CNS, eyes,
skin, heart, liver, kidney, and adrenal glands. The
major presenting symptom is seizures. CNS manifestations occur in up to 90% of patients. The
classic clinical triad is mental retardation, seizures,
and adenoma sebaceum. At least two genetic loci
have been associated with TS, involving chromosome 9 and 16 [18]. The leading causes of death in
TS are renal failure, cardiac tumors, increased intracranial pressure, and bleeding [29]. Renal manifestations occur in about half of patients who have
TS and include cysts, angiomyolipomas (AML), tumors, and perirenal lymphangiomas. There is an
increased incidence of clear cell carcinomas in patients who have TS, especially women. RCC occurs
bilaterally in 43% of affected TS patients, and
tumors occur at a younger age than sporadic RCC
[18,30,31]. The risk for malignancy is lower in TS
than in VHL disease [18]. On US, renal cysts are
seen in the cortex and medulla [Fig. 6]. The renal
cysts in TS appear at an earlier age than the cysts
seen in APKD. Renal cysts are not the primary
diagnostic feature of TS, however. Multiple renal
AML are a primary diagnostic feature of TS, occur
in about 15% of patients who have TS, and are
more common in women. The rate of hemorrhage
in TS-related AML is higher than in sporadic AML
[32]. If TS is a consideration, the diagnosis can
usually be confirmed on CNS imaging with subependymal hamartomas or giant cell astrocytomas.
Acquired cystic kidney disease associated

with dialysis
Acquired cystic kidney disease consists of renal cysts
found in individuals who have chronic renal failure
who may have a history of dialysis, but do not have
a history of hereditary cystic renal disease. Dunnill
and colleagues [33] first described this process in
1977. The most important factor in developing
ACKDD is the duration of ESRD or maintenance

dialysis [34]. ACKDD is seen in 8% to 13% of
patients who have ESRD, 10% to 20% of patients
who have a 1- to 3-year history of dialysis, 40% to
60% of patients who have history of 3 to 5 years of
dialysis, and greater than 90% of patients who have
a history of 5 to 10 years of dialysis. Cysts occur
with equal frequency in patients receiving peritoneal or hemodialysis. The cysts are of tubular origin
(arising from the proximal tubules), thought to be
caused by nondialyzable mitogenic or cytogenic
substances; tubular obstruction from fibrosis or
oxalate crystals; and ischemia. There is a histologic
continuum from cysts with single-layered epithelia
to multilayered (atypical) cysts to renal adenoma
and RCC. RCC occurs in a younger population in
ACKDD patients compared with sporadic RCC,
with a mean age at diagnosis of about 49 years as
compared with 62 years for patients who do not
have ESRD [35]. RCC in patients undergoing dialysis is seven times more common in men than
women, in contrast to the 2/1 ratio in patients
who do not have ESRD [36]. Criteria used to define
ACKDD include the presence of three to five cysts in
each kidney in a patient who has chronic renal
failure, which are not related to inherited renal
cystic disease. On US, multiple bilateral renal cysts
involving the cortex and medulla are seen in the
setting of medical renal disease [Fig. 7]. The cysts
are usually small, many measuring 0.5 cm or less.
Confirmation of the simple cystic nature may be
difficult because of the small size [37]. The cysts
may increase in size and number with time, and the
overall size of the kidneys may increase. This appearance may resemble ADPKD. Complications associated
with ACKDD include cyst infection or hemorrhage,
renal calculi, erythocytosis, and solid neoplasms.
Multiloculated cystic renal masses

A multiloculated appearance can be seen in neoplastic disease such as renal cell carcinoma, multilocular
cystic nephroma, Wilms tumor, and solid tumors
with central necrosis, which are Bosniak category III
lesions. Multiloculated renal masses can be seen in
renal cystic disease such as localized renal cystic
disease, septated cyst, and segmental multicystic dysplastic kidney. Inflammatory disease such as echinococcosis, segmental XGP, and abscess can appear as
a multiloculated cystic renal mass. An organizing
hematoma and vascular malformation can also
appear as a multiloculated cystic renal mass.
Multicystic dysplastic kidney

The pathogenesis of multicystic dysplastic kidney
(MCDK) is complete ureteral obstruction early in
Fig. 6. Ten-year-old who has bilateral enlarged kidneys with multiple simple renal cysts on US (A, B) and CT (C ).
The right kidney measures 11 cm in length and the left kidney measures 11.5 cm in length. MR of the brain
demonstrated multiple subependymal nodules protruding into the lateral ventricle on axial (D) and coronal
(E ) T2-weighted images.
fetal life before the eighth to tenth week, or incomplete ureteral obstruction occurring later in fetal life
between the 10th and 36th week. Contralateral
renal anomalies, most commonly uretero pelvic
junction (UPJ) obstruction, occur in 33% of patients.
Imaging characteristics vary depending on age of

the patient at diagnosis. Fetal US may show large
cysts with no intercommunication between cysts,
absence of identifiable cortical parenchyma, or central sinus structures [Fig. 8]. The MCDK may calcify or
21
22
Weber
Fig. 7. Longitudinal US images of the right (A) and left (B) kidneys and noncontrast CT (C ) demonstrate marked
increased cortical echogenicity and loss of cortical-medullary distinction in this patient who has ESRD and multiple
bilateral simple renal cysts consistent with acquired renal cystic disease.
may not be identifiable in a few patients who have

been followed [38].
Multilocular cystic nephroma

Multilocular cystic nephroma (MLCM) is a benign,
nonhereditary cystic neoplasm with a bimodal pre-
sentation that has greater incidence in males younger

than 2 years and in women during the fifth to
eighth decades. MLCN consists of multiple epithelially lined cysts that do not communicate. MLCN
is usually a benign neoplasm; however, metastases
have been reported [5,39]. US will show a mass
containing multiple cysts or internal septations
Fig. 8. Longitudinal (A) and transverse (B) images of the kidney show multiple large renal cysts with absence of
central sinus structures and no identifiable cortical parenchyma consistent with multicystic dysplastic kidney.
Fig. 9. Longitudinal US (A) and CT (B) images demonstrating a predominantly cystic mass with multiple internal
septations, proven at surgery to be a multilocular cystic nephroma. Note the better representation of the lesion's
internal architecture at US.
[Fig. 9]. It is not possible to conclusively distinguish MLCN from multiloculated RCC radiolographically, and these are usually surgical lesions.
Summary
US plays an important role in evaluation of the
kidney in cases of medical renal disease because
of lower cost, ready availability, lack of radiation,
and lack of need for iodinated contrast material.
The primary role of US in evaluating benign cystic
renal disease is the distinction of a simple cyst from
a solid mass, and in defining the characteristics of a
complex cyst.
Acknowledgments
The author would like to acknowledge Raymond B.
Dyer, MD, for his editorial assistance.
References
[1] Kissane JM. Congenital malformations. In: Hepinstall RH, editor. Pathology of the kidney.
Boston: Little, Brown; 1973. p. 69119.
[2] Schmidt T, Holh C, Haage P, et al. Diagnostic
accuracy of phase-inversion tissue harmonic
imaging versus fundamental B-mode sonography
in the evaluation of focal lesions of the kidney.
AJR Am J Roentgenol 2003;180:163947.
[3] Bosniak MA. The current radiological approach
to renal cysts. Radiology 1986;158:110.
[4] Leder RA. Radiological approach to renal cysts
and the Bosniak classification system. Curr Opin
Urol 1999;9(2):12933.
[5] Hartman DS, Choyke PL, Hartman MS. A
practical approach to the cystic renal mass.
Radiographics 2004;24:S10115.
[6] Bosniak MA. The small (< 3 cm) renal parenchymal tumor: detection, diagnosis, and controversies. Radiology 1991;179:30717.
[7] Zeman RK, Cronan JJ, Rosenfield AT, et al. Imaging approach to the suspected renal mass. Radiol
Clin North Am 1985;23(3):50329.
[8] Israel GM, Bosniak MA. Follow-up CT of
moderately complex cystic lesions of the kidney
(Bosniak category IIF). AJR Am J Roentgenol
2003;181:62733.
[9] Israel GM, Bosniak MA. Calcification in cystic
renal masses: is it important in diagnosis?
Radiology 2003;226:4752.
[10] Harisingani MG, Maher MM, Gervais DA, et al.
Incidence of malignancy in complex cystic renal
masses (Bosniak category III): should imagingguided biopsy precede surgery? AJR Am J
Roentgenol 2003;180:7558.
[11] Rha SE, Byun JY, Jung SE, et al. The renal sinus:
pathologic spectrum and multimodality imaging
approach. Radiographics 2004;24:S11731.
[12] Hidalgo H, Dunnick NR, Rosenburg ER, et al.
Parapelvic cysts: appearance on CT and sonography. AJR Am J Roentgenol 1982;138:66771.
[13] Chan JCM, Kodroff MB. Hypertension and
hematuria secondary to parapelvic cyst. Pediatrics 1980;65:8213.
[14] Gardner KD. Juvenile nephronophthiasis and
renal medullary cystic disease. In: Gardner KD,
editor. Cystic disease of the kidney. New York:
John Wiley & Sons; 1976. p. 17385.
[15] Wise SW, Hartman DS. Medullary cystic disease
of the kidney. In: Pollack HM, McClennan BL,
editors. Clinical urography: an atlas and textbook
of urologic imaging. 2nd edition. Philadelphia:
W.B. Saunders Company; 2000. p. 1398403.
[16] Resnick JS, Hartman DS. Medullary cystic disease
of the kidney. In: Polack HM, editor. Clinical
urology: an atlas and textbook of urologic
imaging. Philadelphia: W.B. Saunders Company;
1990. p. 117884.
[17] Rego JD, Laing FG, Jeffrey RB. Ultrasonic
diagnosis of medullary cystic disease. J Ultrasound Med 1983;2:4336.
[18] Choyke PL. Inherited cystic diseases of the kidney. Radiol Clin North Am 1996;34(5):92546.
[19] Dalgaard OZ. Bilateral polycystic disease of the
23
24
Weber
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
kidney: A follow-up of 284 patients and their

families. Acta Med Scand 1957;157(S328):1255.
Grampsas SA, Chandhoke PS, Fan J, et al.
Anatomic and metabolic risk factors for nephrolithiasis in patients with autosomal dominant
polycystic kidney disease. Am J Kidney Dis 2000;
36(1):537.
Fick GM, Gabow PA. Natural history of autosomal dominant polycystic kidney disease. Annu
Rev Med 1994;45:239.
Bear JC, McManamon P, Morgan J, et al. Age at
clinical onset and at ultrasound detection of
adult polycystic kidney disease. Data for genetic
counseling. Am J Med Genet 1984;18:4553.
Parfrey PS, Bear JC, Morgan J, et al. The diagnosis
and prognosis of autosomal dominant polycystic
kidney disease. N Engl J Med 1990;323:108590.
Dimitrakov JD, Dimitrakov DI. Autosomal recessive polycystic kidney disease. Clinical and
genetic profile. Folia Med (Plovdiv) 2003;45:57.
Harris PC, Rosetti S. Molecular genetics of
autosomal recessive polycystic kidney disease.
Mol Genet Metab 2004;81:7585.
Sano T, Horiguchi H. Von Hippel-Lindau disease. Microsc Res Tech 2003;60:15964.
Kaelin Jr WG. The von Hippel-Lindau tumor
suppressor gene and kidney cancer. Clin Cancer
Res 2004;10:6290S5S.
Richard S, David P, Marsot-Dupuch K, et al.
Central nervous system hemangioblastomas,
endolymphatic sac tumors, and von HippelLindau disease. Neurosurg Rev 2000;23:122.
Gomez MR. Tuberous sclerosis. New York: Raven
Press; 1988.
[30] Torres VE. Systemic manifestations of renal cystic

disease. In: Gardner KD, Bernstein J, editors. The
cystic kidney. Dordrecht (The Netherlands):
Kluwer Academic Publishers; 1990. p. 207.
[31] Zimmerhackl LB, Rehm M, Kaufmehl K, et al.
Renal involvement in tuberous sclerosis complex: a retrospective survey. Pediatr Nephrol
1994;8:4517.
[32] Hildebrandt F. Genetic renal disease in children.
Curr Opin Pediatr 1995;7:18291.
[33] Dunnill MS, Millard PR, Oliver D. Acquired
cystic kidney disease of the kidneys: a hazard of
long-term intermittent maintenance haemodialysis. J Clin Pathol 1977;30:86877.
[34] Levine E, Slusher SL, Grantham JJ, et al. Natural
history of acquired cystic kidney disease in
dialysis patients: a prospective longitudinal CT
study. AJR Am J Roentgenol 1991;156:5016.
[35] Port FK, Ragheb NE, Schwartz AG, et al.
Neoplasms in dialysis patients: a populationbases study. Am J Kidney Dis 1989;14:11923.
[36] Matson MA, Cohen EP. Acquired cystic kidney
disease: occurrence, prevalence, and renal cancers. Medicine 1990;69:21726.
[37] Allan PL. Ultrasonography of the native kidney
in dialysis and transplant patients. J Clin Ultrasound 1992;20:55767.
[38] Strife JF, Souza AS, Kirks DR, et al. Multicystic
Dysplastic kidney in children: US follow-up.
Radiology 1993;186:7858.
[39] Madewell JE, Goldman SM, Davis CJ, et al.
Multilocular cystic nephroma: a radiographicpathologic correlation of 58 patients. Radiology
1983;146:30921.
25
ULTRASOUND
CLINICS
Sonography in Benign and

Malignant Renal Masses
Raj Mohan Paspulati,
&
&
&
&
a,*
MD
, Shweta Bhatt,
Normal sonographic anatomy of the

kidney
Sonographic technique
Pseudolesions of kidney
Dromedary hump
Persistent fetal lobulation
Prominent column of Bertin
(hypertrophy)
Junctional parenchymal defect
Hypoechoic renal sinus
Inflammatory mass lesions
Benign renal tumors
Angiomyolipoma
Renal adenoma
Oncocytoma
Leiomyoma
Hemangioma
Juxtaglomerular tumor (reninoma)
Hemangiopericytoma
Ultrasonography is often the initial modality

for imaging of the kidneys, although contrastenhanced CT is the established imaging modality
for the diagnosis of renal tumors. Despite technical
limitations, a large percentage of renal tumors can
be characterized by ultrasonography. Cystic and
solid renal parenchymal mass lesions can be well
differentiated by ultrasonography. Technical advances in the gray-scale and color-flow Doppler
(CFD) ultrasound have improved the sensitivity
in detection of small renal tumors. Gray-scale and
CFD ultrasonography can demonstrate the vascular
&
&
&
&
&
&
MD
Renal cell carcinoma

Hereditary renal cell carcinoma
Clinical presentation of renal cell
carcinoma
Imaging strategies of renal cell carcinoma
Sonographic findings of renal cell
carcinomas
Malignant uroepithelial tumors of the
renal collecting system
Transitional carcinoma of renal pelvis
Squamous cell carcinoma and
adenocarcinoma
Renal metastases
Renal lymphoma
Summary
References
invasion in selected groups of patients who have

renal cell carcinoma (RCC). Contrast-enhanced
Doppler ultrasonography appears promising as a
cost-effective, noninvasive imaging technique in
the characterization and follow-up of indeterminate renal mass lesions. As nephron-sparing surgery
is being increasingly used in the management of
small RCC, intraoperative ultrasound (US) has become a useful tool in guiding the surgeon. This
article reviews the gray-scale and CFD features of
benign and malignant renal masses encountered
in radiology practice.
Department of Radiology, University Hospitals of Cleveland, Case Western Reserve University, 11100 Euclid
Avenue, Cleveland, OH 44106, USA
b
Department of Radiology, University of Rochester Medical Center, 601 Elmwood Avenue, Box 648,
Rochester, NY 14642, USA
E-mail address: paspulati@uhrad.com (R.M. Paspulati).
doi:10.1016/j.cult.2005.09.004
26
Paspulati & Bhatt
Normal sonographic anatomy of the kidney

The kidneys are bean-shaped retroperitoneal organs
with their medial aspects parallel to the lateral
margin of the adjacent psoas muscles. The normal
orientation of the kidneys is such that the upper
pole is medial and anterior to the lower pole. The
right kidney is 1 to 2 cm inferior in position as
compared with the left kidney because of the location of the liver superior to the right kidney. The
renal size varies with the age, sex, and body habitus. The measurement of renal volume is a more
effective way of assessing the renal size, though
measurement of renal length is more practical in
regular practice [1]. The normal adult kidney measures 10 to 12 cm in length, 4 to 5 cm in width, and
2.5 to 3 cm in thickness. A discrepancy of more
than 2 cm between the lengths of two kidneys is
considered significant and needs further evaluation. The liver and hepatic flexure of the colon
are situated anterior to the right kidney. The spleen
lies anterosuperior to the left kidney and the rest of
the left kidney is related anteriorly with the colon.
On ultrasonography of a normal kidney, there is

good differentiation of the renal capsule, cortex,
medulla, and central sinus complex [Fig. 1]. The
renal capsule is visible as an echogenic line because
of the interface between the echogenic perinephric
fat and renal cortex. The renal parenchyma is composed of outer cortex and inner medulla (pyramids). The renal cortex is echogenic as compared
with the medulla, but is iso- to hypoechoic as
compared with the normal hepatic or splenic parenchyma. The extension of renal cortex toward the
renal sinus between the renal pyramids forms the
columns of Bertin. The central sinus is composed
of fat, fibrous tissue, renal vessels, and lymphatic
vessels. It has highest echogenicity because of the
adipose tissue, and its size increases with the age of
the person.
Sonographic evaluation of the right kidney is ideally performed from an anterior oblique approach
Fig. 1. Normal kidney. Longitudinal (A) and transverse (B) gray-scale US of the normal right kidney (calipers show
the maximum longitudinal dimension of the kidney). (C ) The schematic representation of the sagittal section of
the kidney. L, liver.
Renal Mass Sonography
using liver as an acoustic window, whereas the left

kidney is scanned through a posterior oblique approach. The lower pole of the right kidney may
be imaged using a more posterior approach. The
upper pole of the left kidney is often best seen
through an intercostal approach using spleen as a
window. In addition to supine position, decubitus,
prone, or upright positions may provide better
images of the kidneys [2]. An appropriate transducer frequency ranging from 2.5 to 5 MHz should
be used, depending on the body habitus. Time gain
compensation and adjustment of other scanning
parameters will allow a uniform acoustic pattern
throughout the image [3]. Renal echogenicity
should be compared with the echogenicity of the
liver and spleen [2]. The renal parenchyma of
a normal adult kidney is hypoechoic to the liver
and spleen. The sonographic examination of the
kidneys should include long axis and transverse
views of the upper poles, midportions, and the
lower poles, with assessment of the cortex and
central sinus. Maximum measurement of renal
length should be recorded for both kidneys [2].
Kidneys and the perirenal regions should be assessed for abnormalities. CFD and Power Doppler
(PD) are used to differentiate vascular from nonvascular structures.
Pseudolesions of kidney
There are various developmental variants of the
kidney that need to be identified on sonography
to avoid misdiagnosis as renal neoplasm or other
renal pathology [Table 1].
Fig. 2. Dromedary hump. Longitudinal gray-scale sonogram of the left kidney demonstrates the dromedary
hump (arrow). SP, spleen.
of the renal surface to the adjacent spleen. It can be

easily differentiated from a renal mass because
of its similar echotexture to that of adjacent renal
parenchyma on gray-scale ultrasound. CFD and PD
will demonstrate similar perfusion to that of adjacent renal parenchyma.

Persistent fetal lobulation is another common renal
variant that can be mistaken for renal scarring, a
consequence of chronic infective process of the
kidneys. Persistent fetal lobulation can be differentiated from scarred kidneys by the location of the
renal surface indentations, which do not overlie the
medullary pyramids as in true renal scarring [4],
but overlie the space between the pyramids [Fig. 3].
The underlying medulla and the cortex are normal.
Prominent column of Bertin (hypertrophy)

Dromedary hump
Dromedary hump is a common renal variation
usually seen as a focal bulge on the lateral border
of the left kidney [Fig. 2]. It is a result of adaptation
Prominent column of Bertin is a prominent cortical

tissue that is present between the pyramids and
projects into the renal sinus [Fig. 4]. If not identified as a normal variant, it may be mistaken for an
Table 1: Renal pseudotumors

Pseudotumor
Congenital normal variants
Dromedary hump
Prominent column of Bertin
Junctional parenchymal
defect
Hypoechoic renal sinus fat
Inflammatory lesions
Focal bacterial nephritis
Renal abscess
Diagnostic imaging features

Focal bulge in the lateral contour of left kidney with echotexture
similar to renal parenchyma
Renal surface indentations overlying the space between the pyramids
Continuity with the normal cortex; echotexture and vascular
perfusion similar to the normal cortex
Characteristic location in the anterosuperior and posteroinferior
surface of the kidney and demonstration of continuity with the
central sinus
No distinct margin and normal vessels traversing the sinus
Diagnosis is based on the proper clinical context
27
28
Paspulati & Bhatt
Fig. 3. Persistent fetal lobulations. Longitudinal (A) gray-scale sonogram of the right kidney demonstrates
persistent fetal lobulation (arrow). L, liver. Schematic (B) appearance of persistent fetal lobulations (note fetal
lobulations may be single or multiple).
intrarenal tumor. Sonography can accurately identify it by depicting its continuity with the renal
cortex and a similar echo pattern as the renal parenchyma. CFD and PD imaging can further assist
by depicting a similar vascular pattern as that of
normal renal tissue [5,6]. Prominent columns of
Bertin are usually seen in the middle third of the

kidney and are more common on the left side [5].
Junctional parenchymal defect

Junctional parenchymal defect (JPD) is another
variant commonly mistaken for a cortical scar or
Fig. 4. Prominent column of Bertin. (A) Longitudinal gray-scale US of the left kidney demonstrates a prominent
column of Bertin (arrows) mimicking an isoechoic renal mass. MRI was performed to confirm ultrasound findings.
T1 flash fat-sat (B) and gadolinium-enhanced (C ) MRI images of the kidneys reveal a prominent column of Bertin
(arrows) seen in continuity with the renal cortex. (D) Schematic drawing of a prominent column of Bertin.
Fig. 5. Junctional parenchymal defect. (A) Longitudinal gray-scale US of the right kidney demonstrates a notch
in the lateral border (arrow). L, liver. (B) Contrast-enhanced CT of the kidneys in another patient demonstrates
the junctional parencymal defect (arrow).
a hyperechogenic renal tumor. JPD is a linear or

triangular hyperechoic structure in the anterosuperior or posteroinferior surface of the kidney [Fig. 5].
These are caused by normal extensions of the renal
sinus at the junction of the embryonic renunculi.
These are differentiated from pathologic lesions by
their characteristic location and demonstrate continuity with the central sinus by an echogenic line
called interrenicular septum [79].
Hypoechoic renal sinus

The echogenicity of the renal sinus may vary from
echogenic to anechoic. Hypoechoic renal sinus may
mimic a mass lesion [10]. Absence of a well-defined
margin and demonstration of normal vessels traversing the renal sinus by CFD will aid in differentiating a hypoechoic renal sinus from a mass
lesion [10].
Inflammatory mass lesions

Acute focal bacterial nephritis and renal abscess
may present as renal mass lesions indistinguishable
from a renal tumor by ultrasonography and contrastenhanced CT. The clinical presentation will aid in
differentiating these inflammatory pseudotumors
from RCC [1113].
Benign renal tumors

Angiomyolipoma
Angiomyolipoma (AML) is a hamartoma and has
variable amounts of mature adipose tissue, smooth
muscle, and thick-walled blood vessels. Eighty percent of the AMLs are sporadic in occurrence and
20% of them are associated with tuberous sclerosis
(TS). Presence of subependymal nodules and giant
cell astrocytoma are sine qua non of TS, but not
AML. On the contrary, 80% of the patients who
have TS develop AMLs [14,15]. Patients who have
TS develop AMLs at a much younger age, and these

tend to be multiple, bilateral, and larger than in
sporadic cases. AMLs in patients who have TS are
more likely to grow and become symptomatic
[15,16]. The presence of estrogen and progesterone
receptors in angiolipomas has been reported, and
such AMLs are more common in women and in TS.
These AMLs tend to grow during pregnancy and
present with hemorrhage [17,18]. Small AMLs are
asymptomatic and are incidental findings on imaging. AMLs smaller than 4 cm are symptomatic and
are at increased risk for spontaneous hemorrhage
[16,19]. Massive retroperitoneal hemorrhage from
AML, also known as Wunderlichs syndrome, has
been found in 10% of patients.
The characteristic sonographic appearance of
AML is a well-defined hyperechoic mass [Fig. 6].
This increased echogenicity is attributed to the fat
content, multiple interfaces, heterogeneous cellular
architecture, and multiple vessels within the tumor
[20,21]. However, there is significant overlap between the imaging features of AML and RCC. Small
RCCs can be hyperechoic and indistinguishable
from an AML on sonography. Acoustic shadowing,
hypoechoic rim, and intratumoral cystic changes
are some of the sonographic features found to be
helpful in differentiating an AML from RCC. Hypoechoic rim and intratumoral cystic changes are seen
only in RCC, whereas acoustic shadowing is observed with AML [Fig. 7] [2224]. PD of AML may
reveal focal intratumoral flow and a penetrating
flow pattern [25]. The demonstration of intratumoral fat on CT confirms the diagnosis of an
AML. The CT appearance of an AML also depends
on the relative proportion of smooth muscle and
vascular components of the tumor. Rarely, RCCs
can demonstrate fat attenuation caused by entrapment of the perinal or renal sinus fat, lipid necrosis,
or osseous metaplasia [26]. The characteristic intratumoral fat cannot be detected in 4.5% of AMLs,
29
30
Paspulati & Bhatt
Fig. 6. Angiomyolipoma. (A) Longitudinal gray-scale US of the right kidney demonstrates an echogenic mass
(arrow) with posterior acoustic shadowing (arrowheads). (B) Corresponding CT (excretory phase) confirms this
lesion to be an angiomyolipoma, seen as a fat attenuation lesion (arrow) with a household unit of -8.
and will have high attenuation on an unenhanced

CT scan. This finding has been attributed to minimal fat content or immature fat [25,27]. These
AMLs with low fat content demonstrate homogeneous and prolonged enhancement on a contrastenhanced scan, which distinguishes them from an
RCC [25,28]. These AMLs with minimal fat are
isoechoic with renal parenchyma on sonography
[25]. The demonstration of micro- or macroaneurysms at angiography is reported to be characteristic
of an AML [29].
The risk for spontaneous rupture and hemorrhage of an AML is related to the tumor size and
the size of microaneurysms. AMLs larger than 4 cm
and those with microaneurysms larger than 5 mm
are reported to be at increased risk for spontaneous
rupture [16,30,31]. Management options of AMLs
include observation, embolization, and partial or
total nephrectomy. Prophylactic transcatheter embolization of AMLs larger than 4 cm is reported
to prevent tumor growth and spontaneous rupture
[19,31,32]. Kothary and colleagues [33] have described a high recurrence rate of AMLs after embolization in patients who have TS, and recommend
long-term surveillance of these patients following embolization.
Renal adenoma
Renal cell adenoma is considered to be a benign
counterpart of RCC, though the true nature and
potential of this tumor is a subject of much debate.
The size criterion used by many pathologists in
distinguishing an adenoma from RCC is based on
the initial observation by Bell, that renal cortical
glandular tumors of smaller than 3 cm rarely metastasize [34,35]. There are no histopathologic, histochemical, immunologic, or imaging characteristics
that distinguish a benign adenoma from an RCC
[36]. Most pathologists consider these small renal
cortical tumors to be premalignant or potentially
malignant and believe that tumor size is not a valid
differentiating criterion [37]. The widespread use of
US and CT has resulted in the incidental detection
of these tumors.
Oncocytoma
Fig. 7. RCC. Longitudinal gray-scale sonogram of the

left kidney demonstrates a hyperechoic mass (M) arising from the lower pole with areas of intra tumoral
cystic changes (asterisk).
Renal oncocytoma is a benign tumor of renal tubular origin (renal tubular epithelium is also called
oncocyte). It has the male predominance and age
incidence similar to RCC. They are asymptomatic
and are discovered as incidental findings on imaging [38]. They are well-defined tumors of variable
size and can be as large as 20 cm [35]. The preoperative differentiation of oncocytomas from RCC
is invaluable, but is often difficult because of overlap of imaging features. The characteristic central
stellate scar on cross-sectional imaging and spokewheel pattern of enhancement on an angiogram are
infrequently seen in oncocytomas and can also be
seen in RCC [3941]. Oncocytomas can be hypoechoic, isoechoic, or hyperechoic to the renal parenchyma on sonography. MRI is reported to be
superior to CT and US in identifying the imaging
features of a small tumor [42]. The radiologic features, which are helpful in distinguishing an oncocytoma from RCC, include well-defined margins,
homogeneous enhancement without hemorrhage,
calcification or necrosis, presence of a central stellate scar, and spoke-wheel pattern of arterial enhancement. There are few reports of bilateral and
multicentric oncocytomas [43,44]. Renal oncocytoma and RCC can coexist in the same or contralateral kidney [45,46]. Dechet and colleagues [47]
have reported coexistent RCC in 10% of a total 138
cases of oncocytomas. Imaging-guided biopsy of
renal tumors is indicated whenever there is radiologic suspicion of an oncocytoma [4850].
Juxtaglomerular tumor (reninoma)
Leiomyoma
Hemangiopericytoma
Renal leiomyoma is a rare benign tumor of smooth

muscle origin. These tumors are either peripheral,
arising from the renal capsule, or central in parapelvic location. They are more common in women
between the second and fifth decades of life. Most
renal leiomyomas are asymptomatic, with incidental detection on routine diagnostic imaging. These
are well-defined tumors and are indistinguishable
from RCC by imaging. Renal leiomyomas have
variable appearance on imaging: from that of an
entirely solid, to a mixed solid/cystic, to an entirely
cystic lesion. Renal leiomyomas appear as welldefined hypoechoic solid mass lesions on ultrasonography. The peripheral lesions may extend
into the retroperitoneum and can resemble primary
retroperitoneal sarcomas. The central lesions will
have a mass effect over the collecting system and
renal vasculature. They are most often avascular or
hypovascular on angiogram [5153].
Hemangiopericytomas are rare renal tumors with a

malignant potential that arise from the pericytes.
Tumor-induced hypoglycemia is characteristic of
hemangiopericytoma and has been attributed to
the production of insulin-like growth factors by
the tumor. There are no distinguishing radiologic
features of hemangiopericytoma from RCC or other
mesenchymal tumors of the kidney [6365].
Hemangioma
Hemangiomas are uncommon benign tumors of
the kidney that can present with macroscopic hematuria. They are commonly located in the renal
pyramids and renal pelvis, and are classified into
capillary and cavernous hemangiomas. The vascular spaces are small in capillary hemangioma and
large in cavernous hemangiomas. They are predominantly smaller than 1 cm, but occasionally present
as large mass lesions [54]. Gray-scale US features
a nonspecific solid mass, and CT demonstrates a
well-defined low-density mass without significant
enhancement [55,56]. Larger lesions may cause displacement of the renal vessels and collecting system. Angiography may demonstrate a hypovascular
or hypervascular mass [57,58].
Juxtaglomerular tumors are benign, renin-producing

tumors of the kidney that arise from the afferent arterioles of the glomerulus. These tumors were first
described by Robertson and colleagues in 1967
[59]. They are twice as common in women as in
men. In a young patient who has hypertension, the
presence of a renal mass, elevated serum renin
levels, and hypokalemia should raise a suspicion
of reninoma. The tumor is either hypo- or hyperechoic on sonography and appears as a well-defined
hypodense solid mass on a contrast-enhanced CT.
Angiography demonstrates a hypovascular mass
with normal renal arteries. Renal vein sampling
demonstrates elevated renin levels in reninomas,
but renin is also elevated in renal artery stenosis.
Surgical resection of the tumor results in reversal of
hypertension and hypokalemia [6062].
Renal cell carcinoma

RCC is the most common primary malignancy of
the kidney. It accounts for 2% of all malignancies.
There has been a steady increase of 38% in the
incidence of RCC between 1974 and 1990 [66].
The survival rates have also improved from 52%
between 1974 and 1976 to 58% between 1983 and
1996 [66]. This trend has been attributed to the
improved imaging technique and early diagnosis.
Smith and colleagues [67] have reported that only
5.3% of the tumors between 1974 and 1977 were
3 cm or smaller as compared with 25.4% during
1982 to 1985. Of these small tumors in the later
group, 96.7% were incidentally discovered by ultrasonography and CT. Most RCCs that are amenable
for surgical cure by either partial nephrectomy or
nephron-sparing surgery are incidentally detected
by the increased use of cross-sectional imaging.
Ultrasonography, being the primary imaging modality of the kidneys, is useful for screening and
detection of small RCCs [68,69].
The RCCs are classified histologically into four
main types [Table 2]. These include clear cell carcinoma, papillary carcinoma, chromophobe carcinoma, and collecting duct carcinoma. The clear
cell carcinomas are the most common type, ac-
31
32
Paspulati & Bhatt
Table 2: Classification of renal cell carcinoma

Subtype
Incidence
Grade
Imaging features
Clear (conventional)
cell carcinoma
Papillary type
Type 1
Type 2
Chromophobe type
Collecting duct type
Medullary carcinoma
70%80%
Low-grade tumor
Poor enhancement
Low-grade tumor
Aggressive tumor
Aggressive tumor with poor prognosis

Aggressive tumor with poor prognosis
and common in sickle cell trait
Poor enhancement
Intense enhancement
10%15%
4%5%
<1%
<1%
counting for 70% of the RCCs. The papillary type is

the second most common type, accounting for
10% to 15% of the RCCs. The papillary type is
subclasssified into type 1 and type 2 tumors. The
type 2 papillary tumors are more aggressive than
type 1. Clear cell and papillary tumors arise from
the proximal tubular epithelium. The chromophobe carcinomas account for 5% of the RCCs
and arise from cells of distal tubule. The collecting duct carcinomas are the least common type,
arise from collecting duct epithelium, and are the
most aggressive of all RCCs. The medullary carcinoma is a subtype of collecting duct carcinoma that
is more common in patients who have sickle cell
trait. Imaging cannot differentiate the different histologic types of RCC. The incidence of RCC is
increased in acquired cystic disease of the kidney
(ACDK). Clear cell carcinoma is the most common
type of RCC associated with ACDK. The incidence
of papillary type of RCC in ACDK is also higher
than in the general population [70,71].
Hereditary renal cell carcinoma

RCCs are predominantly sporadic in occurrence
and only 4% of them are familial in nature. The
different types of hereditary RCCs are displayed in
Table 3. The hereditary RCCs are characterized by
autosomal dominant inheritance, presentation at a
young age (third to fifth decades), and multifocal
and bilateral tumors [72].
Clinical presentation of renal cell carcinoma

The classic clinical triad of hematuria, abdominal
pain, and abdominal mass is seen in less than 10%
of patients. About 20% to 40% present with paraneoplastic syndrome, which includes anemia, fever,
hypertension, hypercalcemia, and hepatic dysfunction [7375]. RCC can be associated with Stauffer
syndrome, which is characterized by nonmetastatic
intrahepatic cholestasis. This syndrome is a tumorinduced inflammatory response and is reversible
after resection of the tumor [7678]. About 2% of
Table 3: Hereditary renal cell carcinoma

Syndrome
Inheritance
Predominant renal
tumor
Other renal
lesions
Associated
abnormalities
Von Hippel-Lindau
AD
Clear cell carcinoma
Cysts
Hereditary papillary RCC

Hereditary leiomyoma RCC
AD
AD
Papillary type 1
Papillary type 2
None
None
Birt-Hogg-Dub
AD
Chromophobe
carcinoma
Other types
of RCC
Familial renal oncocytoma

Medullary carcinoma
Oncocytoma
Medullary carcinoma
None
None
Hemangioblastomas
Retinal angiomas
Pancreatic cysts
Neuroendocrine
tumors of pancreas
Phaeochromocytoma
None
Cutaneous and
uterine leiomyomas
Fibrofolliculomas
Lung cysts
Pneumothorax
Sickle cell trait
Abbreviation: AD, autosomal dominant.
the male patients present with left-sided varicocele

because of renal vein involvement [79].
lesions, hypovascular solid mass lesions, and AMLs

with minimal fat component [83].
Imaging strategies of renal cell carcinoma
Sonographic findings of renal cell carcinomas
The goal of imaging is detection, diagnosis, and

staging of RCC. Ultrasonography, CT, and MRI
have variable sensitivity in detecting and staging
RCC. Ultrasonography is less sensitive in detecting
small renal lesions, especially those that do not
deform the contour of the kidney. The sensitivity
of CT and ultrasonography for detection of lesions
3 cm and less is 94% and 79%, respectively [80].
CT and MRI have nearly 100% accuracy in the
diagnosis of RCC [81]. Ultrasonography is also less
accurate than CT and MRI in staging of RCC. The
accuracy of CT and MRI in staging of RCC ranges
from 67% to 96%. Catalano and colleagues [82]
have reported 96% sensitivity, 93% specificity, and
95% accuracy of multidetector CT (MDCT) in evaluating Robson stage I RCC. Robson and tumor,
nodes, and metastases (TNM) staging of RCC are
outlined in Table 4.
Despite these limitations, ultrasonography is still
the initial imaging modality for screening and characterization of renal mass lesions. Ultrasonography
is also useful in characterizing indeterminate renal
mass lesions detected by CT, such as atypical cystic
The sonographic spectrum of RCCs varies from hypoechoic to hyperechoic solid mass lesions [Fig. 8].
RCCs 3 cm and smaller are predominantly hyperechoic and must be differentiated from AMLs
[84,85]. The hyperechoic appearance is reported
to be caused by papillary, tubular, or microcystic architecture; minute calcification; intratumoral hemorrhage; cystic degeneration; or fibrosis [24]. The
presence of an anechoic rim caused by a pseudocapsule and intratumoral cystic changes can aid
in differentiation of hyperechoic RCC from AML
[24,86]. Several investigators have reported acoustic
shadowing as a useful sign of AML [22,23]. Small
isoechoic RCCs and those located at the poles can
be missed by ultrasonography [26]. The isoechoic
RCCs must be differentiated from pseudotumors,
which include prominent column of Bertin, dromedary hump, persistent fetal lobulation, and compensatory hypertrophy. Careful attention to the
morphology on gray-scale US will differentiate pseudotumors from a mass lesion. Power Doppler and
contrast-enhanced sonography are useful in differentiating pseudotumors from true renal mass
lesions by demonstrating similar vascularity of the
pseudotumors to that of adjacent normal renal cortex [87,88]. Power Doppler and contrast-enhanced
sonography will demonstrate the vascularity of a
renal mass, but cannot differentiate an RCC from
an AML [87,88].
Approximately 15% of the RCCs are cystic in
nature and may result from extensive necrosis of a
tumor, or represent a primary cystic renal carcinoma [89]. Histologically, the cystic RCCs are predominantly of clear cell type. RCCs with extensive
necrosis are more aggressive as compared with the
primary multilocular cystic RCCs [90,91]. Multilocular cystic RCC (MCRCC) is an uncommon subtype of RCC and constitutes about 3% of all RCCs.
MCRCCs have a benign clinical course and may
benefit from nephron-sparing surgery [92]. Crosssectional imaging with US and CT of MCRCC will
demonstrate well-defined, multilocular cystic mass
with thin septations. Dystrophic calcification and
mural nodules are less common and MCRCC
should be included in the differential diagnosis of
all multilocular cystic renal mass lesions in adults
[93]. Small MCRCCs of less than 3 cm are hyperechoic on US and can mimic solid mass lesions, but
show minimal enhancement on contrast-enhanced
CT or MRI [94]. Contrast-enhanced Doppler US
is reported to improve the diagnostic accuracy of
malignant cystic renal mass by demonstrating the
Table 4: Staging of renal cell carcinoma

Robson
stage
I
II
III-A
III-B
III-C
IV-A
IV-B
Tumor description
Tumor confined within
renal capsule
Tumor < 2.5 cm
Tumor > 2.5 cm
Tumor extension to
perinephric fat or
adrenal gland
Renal vein involvement
or infradiaphragmatic
IVC involvement
Supradiaphragmatic
IVC involvement
Regional lymph node
metastases
Venous involvement and
lymph node metastases
Invasion of adjacent
organs beyond the
,
Gerota s fascia
Distant metastases
TNM stage
T1
T2
T3a
T3b
T3c
N1N3
T4
Abbreviations: IVC, inferior vena cava; TNM, tumor,

nodes, metastases.
33
34
Paspulati & Bhatt
Fig. 8. RCC. Longitudinal gray-scale (A) and CFD (B) sonography of the right kidney demonstrates an iso- to
hypoechoic mass arising from the lower pole, which shows presence of vascularity consistent with a RCC.
vascularity of the intracystic septations and mural

nodules [95].
CT and MRI are the standard imaging methods
for staging of RCC. However, US is useful in detecting the venous invasion and for demonstrating
the cranial extent of the inferior vena cava (IVC)
thrombus. Overall accuracy, sensitivity, and specificity of CFD for detecting the tumor involvement
of renal vein and IVC is 93%, 81%, and 98%, respectively [96]. McGahan and colleagues [97] have
reported a 100% sensitivity in the detection of renal
vein involvement as compared with 89% sensitivity for IVC involvement by CFD sonography.
Hence, US may be used as a complementary imaging modality when CT findings are equivocal in the
assessment of venous extension of the tumor. The
tumor thrombus is seen as an echogenic intraluminal mass causing distension of the vein. CFD will
demonstrate flow around a bland thrombus and
vascularity within a tumor thrombus. Use of US
contrast agents is reported to improve the accuracy
not only in demonstrating the extent of the thrombus but also in differentiating a tumor from a
bland thrombus.
The prognosis of RCC will depend on the stage,
histologic type, and grade of the tumor. The 5-year
survival rates of TNM stages I, II, III, and IV are
reported to be 91%, 74%, 67%, and 32%, respec-
tively [98]. The presence of a sarcomatoid component is reported to have poor outcome [99].
Malignant uroepithelial tumors of the renal

collecting system
Malignant uroepithelial tumors of the renal pelvis
constitute about 5% of all the urinary tract neoplasms [100]. 90% of them are transitional cell
carcinomas (TCC), 5% to 10% are squamous cell
carcinomas, and less than 1% are adenocarcinomas
[101,102].
Transitional carcinoma of renal pelvis

TCCs of the renal pelvis have similar epidemiologic
features to those of bladder and ureter. The risk
factors include exposure to chemicals in petroleum,
rubber, and dye industries; analgesic abuse; and
chronic inflammations. TCC is one of the several
extracolonic manifestations of hereditary nonpolyposis colorectal cancer (HNPCC)/Lynch syndrome [103]. The mean age of presentation of
TCC is 68 years with a higher rate of incidence in
men than women. Painless hematuria is the characteristic clinical presentation of TCC [104]. Three
morphologic forms of TCC are described, including focal intraluminal mass, mural thickening with
narrowing of lumen, and an infiltrating mass in the
renal sinus [105107]. The excretory urogram has

been the primary imaging modality for the diagnosis of TCC and is being replaced by CT or MR
urogram [108,109]. These imaging modalities have
the advantage of evaluating the entire urinary tract,
which is crucial in the assessment of TCC.
Sonography demonstrates a poorly defined hypoor hyperechoic mass in the renal sinus with or
without pelvicaliectasis. The mass lesions are initially intraluminal and later invade the renal sinus
fat and renal parenchyma. Infiltrating tumors of the
renal parenchyma tend to preserve the reniform
shape of the kidney [106,110].
carcinoma and adenocarcinoma of the renal pelvis

[101,102,111]. The clinical presentation ranges from
painless hematuria to nonspecific flank pain caused
by hydronephrosis [101]. Squamous cell carcinomas are more aggressive than TCC and the tumor
manifests as an infiltrating mass involving the collecting system, renal sinus fat, and renal parenchyma [112]. It is often difficult to differentiate
squamous cell carcinoma of the renal pelvis from
xanthogranulomatous pyelonephritis by imaging
[113,114].
Squamous cell carcinoma and adenocarcinoma
The frequency of renal metastases is reported to

vary from 7% to 13% based on autopsy findings
[115,116]. More frequent use of cross-sectional
imaging has resulted in an increase in the detection
of renal metastases [117,118]. In patients who have
a known history of malignancy, renal metastases
are three times more common than primary renal
tumors and are usually asymptomatic [115,116].
The tumors that most commonly metastasize to
Squamous cell carcinoma is the second most common malignant uroepithelial tumor of the renal
collecting system [Fig. 9]. Chronic irritation of
the uroepithelium is the etiologic factor, which
leads to squamous or columnar metaplasia of the
transitional epithelium. Renal calculi with longstanding hydronephrosis and inflammation are
important predisposing factors for squamous cell
Renal metastases
Fig. 9. Squamous cell carcinoma. Longitudinal gray-scale (A) and CFD (B) US of the left kidney demonstrates an
enlarged kidney with areas of chunky calcification (arrows) with posterior acoustic shadowing (arrowheads).
There is increased vascularity in the mass with large areas of necrosis (asterisk). Corresponding contrast-enhanced
coronal CT (C ) confirms the presence of calcification (arrows) and necrosis (asterisk). This tumor was pathologically
confirmed to be a squamous cell carcinoma.
35
36
Paspulati & Bhatt
Fig. 10. Hyperechoic renal metastasis. Patient is a known case of esophageal carcinoma. Longitudinal gray-scale US
of the right (A) and the left (B) kidney demonstrate multiple hyperechoic mass lesions (arrows). Tranverse CFD
image (C ) of the right kidney reveals increased vascularity.
the kidney are carcinoma of the lung, breast, and

gastrointestinal tract, and melanoma [Fig. 10]
[115,119]. The most common manifestation is
bilateral, multiple renal mass lesions, though they
can present with unilateral and solitary lesions.
Renal metastases can be well-defined focal mass
lesions or infiltrating in nature [117,120].
The most common sonographic appearance is
hypoechoic, cortical mass lesions without throughtransmission [Fig. 11] [121,122]. CT has higher
sensitivity and accuracy than US in the detection
of renal metastases [121123]. In patients who have

a known extrarenal primary malignancy, tissue
sampling is necessary to differentiate metastases
from a synchronous primary RCC [124].
Renal lymphoma
Renal lymphoma is commonly secondary to hematogeneous dissemination or contiguous extension
from a retroperitoneal nodal disease. Primary lymphoma is rare as there is no lymphoid tissue in the
Fig. 11. Hypoechoic renal metastasis. Longitudinal gray-scale ultrasound of the right (A) and left (B) kidneys
demonstrate multiple hypoechoic masses (arrows) in the renal parenchyma consistent with metastasis. L, liver.
Fig. 12. Lymphoma. Longitudinal gray-scale US of the right (A) and the left (B) kidneys demonstrates bilaterally
enlarged kidneys (R, 15.6 cm; L, 14.8 cm). In addition, right kidney also demonstrates a focal mass (arrow) in
patient who has known nonHodgkins lymphoma.
kidney [125,126]. Though the reported incidence

of renal involvement on autopsy ranges from 30%
to 60%, the actual detection by imaging is only 3%
to 8% [127]. The kidney is most commonly
involved by the nonHodgkins B-cell type of lymphoma [128130]. There is a wide spectrum of renal
involvement of lymphoma. Unilateral or bilateral
multiple renal mass lesions are the most common
type of renal lymphoma. Bilateral renal involvement is reported to occur in 50% to 72% of lymphomas [Fig. 12]. Other manifestations include
solitary renal mass, diffuse infiltration of the renal
parenchyma, contiguous invasion from retroperitoneal disease, and isolated perinephric mass [127].
In the diffuse infiltrative form, there is proliferation of lymphoma within the interstitium of the
renal parenchyma, resulting in enlarged kidneys
with preservation of the reniform shape [131].
The renal mass lesions are homogenously hypoechoic on ultrasonography and are hypodense on
a contrast-enhanced CT [131133]. Spontaneous
hemorrhage, cystic changes, and calcification are
uncommon and are usually secondary to prior treatment [127].
Summary
CT is the gold standard for the detection and characterization of renal mass lesions and in staging of
RCC. Despite its limitations, ultrasonography is
often the first imaging modality of the kidneys and
plays an important role in the diagnosis of renal
tumors. Technical advances in the gray-scale ultrasonography have improved the detection of small
RCCs. CFD and contrast-enhanced Doppler ultrasonography are useful in characterization of renal
tumors and in the identification of pseudotumors.
As nephron-sparing surgery is now an established technique in the management of small
RCC, intraoperative US has a key role in guiding

the surgeon.
References
[1] Emamian SA, Nielsen MB, Pedersen JF, et al.
Kidney dimensions at sonography: correlation
with age, sex, and habitus in 665 adult volunteers. AJR Am J Roentgenol 1993;160:836.
[2] Grant EG, Barr LL, Borgstede J, et al. AIUM
standard for the performance of an ultrasound
examination of the abdomen or retroperitoneum. American Institute of Ultrasound in
Medicine. J Ultrasound Med 2002;21:11827.
[3] Hagen-Ansert SL, Levzow B. Kidneys and adrenal
glands. 3rd edition. St. Louis (MO): Mosby; 1993.
[4] Marchal G, Verbeken E, Oyen R, et al. Ultrasound of the normal kidney: a sonographic,
anatomic and histologic correlation. Ultrasound
Med Biol 1986;12:9991009.
[5] Lafortune M, Constantin A, Breton G, et al. Sonography of the hypertrophied column of Bertin.
[6] Ascenti G, Zimbaro G, Mazziotti S, et al. Contrast- enhanced power Doppler US in the diagnosis of renal pseudotumors. Eur Radiol 2001;
11:24969.
[7] Tsushima Y, Sato N, Ishizaka H, et al. US findings of junctional parenchymal defect of the
kidney. Nippon Igaku Hoshasen Gakkai Zasshi
1992;52:43642.
[8] Carter AR, Horgan JG, Jennings TA, et al. The
junctional parenchymal defect: a sonographic
variant of renal anatomy. Radiology 1985;154:
499502.
[9] Hoffer FA, Hanabergh AM, Teele RL. The interrenicular junction: a mimic of renal scarring
on normal pediatric sonograms. AJR Am J Roentgenol 1985;145:10758.
[10] Seong CK, Kim SH, Lee JS, et al. Hypoechoic
normal renal sinus and renal pelvis tumors:
sonographic differentiation. J Ultrasound Med
2002;21:9939 [quiz 10012].
37
38
Paspulati & Bhatt
[11] Schmidt H, Fischedick AR, Wiesmann W, et al.

Acute focal bacterial nephritis. Rofo 1986;145:
2459.
[12] Lee JK, McClennan BL, Melson GL, et al. Acute
focal bacterial nephritis: emphasis on gray scale
sonography and computed tomography. AJR
Am J Roentgenol 1980;135:8792.
[13] Soulen MC, Fishman EK, Goldman SM, et al.
Bacterial renal infection: role of CT. Radiology
1989;171:7037.
[14] Casper KA, Donnelly LF, Chen B, et al. Tuberous sclerosis complex: renal imaging findings. Radiology 2002;225:4516.
[15] Ewalt DH, Sheffield E, Sparagana SP, et al. Renal
lesion growth in children with tuberous sclerosis complex. J Urol 1991;160:1415.
[16] Steiner MS, Goldman SM, Fishman EK, et al.
The natural history of renal angiomyolipoma.
J Urol 1993;150:17826.
[17] LHostis H, Deminiere C, Ferriere JM, et al. Renal angiomyolipoma: a clinicopathologic, immunohistochemical, and follow-up study of
46 cases. Am J Surg Pathol 1999;23:101120.
[18] Hatakeyama S, Habuchi T, Ichimura Y, et al.
Rapidly growing renal angiomyolipoma during
pregnancy with tumor thrombus into the inferior vena cava: a case report. Nippon Hinyokika Gakkai Zasshi 2002;93:4851.
[19] Dickinson M, Ruckle H, Beaghler M, et al. Renal angiomyolipoma: optimal treatment based
on size and symptoms. Clin Nephrol 1998;49:
2816.
[20] Hartman DS, Goldman SM, Friedman AC, et al.
Angiomyolipoma: ultrasonic-pathologic correlation. Radiology 1981;139:4518.
[21] Scheible W, Ellenbogen PH, Leopold GR, et al.
Lipomatous tumors of the kidney and adrenal:
apparent echographic specificity. Radiology 1978;
129:1536.
[22] Siegel CL, Middleton WD, Teefey SA, et al.
Angiomyolipoma and renal cell carcinoma: US
differentiation. Radiology 1996;198:78993.
[23] Zebedin D, Kammerhuber F, Uggowitzer MM,
et al. Criteria for ultrasound differentiation of
small angiomyolipomas (< or = 3 cm) and renal
cell carcinomas. Rofo 1998;169:62732.
[24] Yamashita Y, Ueno S, Makita O, et al. Hyperechoic renal tumors: anechoic rim and intratumoral cysts in US differentiation of renal cell
carcinoma from angiomyolipoma. Radiology
1993;188:17982.
[25] Jinzaki M, Tanimoto A, Narimatsu Y, et al. Angiomyolipoma: imaging findings in lesions with
minimal fat. Radiology 1997;205:497502.
[26] Helenon O, Merran S, Paraf F, et al. Unusual fatcontaining tumors of the kidney: a diagnostic
dilemma. Radiographics 1997;17:12944.
[27] Sant GR, Heaney JA, Ucci Jr AA, et al. Computed
tomographic findings in renal angiomyolipoma: an histologic correlation. Urology 1984;
24:2936.
[28] Kim JK, Park SY, Shon JH, et al. Angiomyo-
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
lipoma with minimal fat: differentiation from

renal cell carcinoma at biphasic helical CT.
Radiology 2004;230:67784.
Silbiger ML, Peterson Jr CC. Renal angiomyolipoma: its distinctive angiographic characteristics. J Urol 1971;106:3635.
Zagoria RJ, Dyer RB, Assimos DG, et al. Spontaneous perinephric hemorrhage: imaging and
management. J Urol 1991;145:46871.
Yamakado K, Tanaka N, Nakagawa T, et al.
Renal angiomyolipoma: relationships between
tumor size, aneurysm formation, and rupture.
Radiology 2002;225:7882.
Mourikis D, Chatziioannou A, Antoniou A, et al.
Selective arterial embolization in the management of symptomatic renal angiomyolipomas.
Eur J Radiol 1999;32:1539.
Kothary N, Soulen MC, Clark TW, et al. Renal
angiomyolipoma: long-term results after arterial embolization. J Vasc Interv Radiol 2005;
16:4550.
Harrison RB, Dyer R. Benign space-occupying
conditions of the kidneys. Semin Roentgenol
1987;22:27583.
Davis Jr CJ. Pathology of renal neoplasms.
Semin Roentgenol 1987;22:23340.
Bennington JL. Proceedings: Cancer of the
kidneyetiology, epidemiology, and pathology.
Cancer 1973;32:101729.
Bosniak MA. The small (less than or equal to
3.0 cm) renal parenchymal tumor: detection,
diagnosis, and controversies. Radiology 1991;
179:30717.
Licht MR. Renal adenoma and oncocytoma.
Semin Urol Oncol 1995;13:2626.
Quinn MJ, Hartman DS, Friedman AC, et al.
Renal oncocytoma: new observations. Radiology 1984;153:4953.
Jasinski RW, Amendola MA, Glazer GM, et al.
Computed tomography of renal oncocytomas.
Comput Radiol 1985;9:30714.
Ambos MA, Bosniak MA, Valensi QJ, et al. Angiographic patterns in renal oncocytomas. Radiology 1978;129:61522.
De Carli P, Vidiri A, Lamanna L, et al. Renal
oncocytoma: image diagnostics and therapeutic
aspects. J Exp Clin Cancer Res 2000;19:28790.
Mead GO, Thomas Jr LR, Jackson JG. Renal
oncocytoma: report of a case with bilateral
multifocal oncocytomas. Clin Imaging 1990;14:
2314.
Zhang G, Monda L, Wasserman NF, et al. Bilateral renal oncocytoma: report of 2 cases and
literature review. J Urol 1985;133:846.
Kavoussi LR, Torrence RJ, Catalona WJ. Renal
oncocytoma with synchronous contralateral
renal cell carcinoma. J Urol 1985;134:11936.
Nishikawa K, Fujikawa S, Soga N, et al. Renal
oncocytoma with synchronous contralateral
renal cell carcinoma. Hinyokika Kiyo 2002;48:
8991.
Dechet CB, Bostwick DG, Blute ML, et al. Renal
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[55]
[56]
[57]
[58]
[59]
[60]
[61]
[62]
[63]
[64]
oncocytoma: multifocality, bilateralism, metachronous tumor development and coexistent

renal cell carcinoma. J Urol 1999;162:402.
Rodriguez CA, Buskop A, Johnson J, et al. Renal
oncocytoma: preoperative diagnosis by aspiration biopsy. Acta Cytol 1980;24:3559.
Nguyen GK, Amy RW, Tsang S. Fine needle
aspiration biopsy cytology of renal oncocytoma.
Acta Cytol 1985;29:336.
Alanen KA, Tyrkko JE, Nurmi MJ. Aspiration
biopsy cytology of renal oncocytoma. Acta Cytol
1985;29:85962.
Steiner M, Quinlan D, Goldman SM, et al. Leiomyoma of the kidney: presentation of 4 new
cases and the role of computerized tomography.
J Urol 1990;143:9948.
Kanno H, Senga Y, Kumagai H, et al. Two cases
of leiomyoma of the kidney. Hinyokika Kiyo
1992;38:18993.
Protzel C, Woenckhaus C, Zimmermann U, et al.
Leiomyoma of the kidney. Differential diagnostic aspects of renal cell carcinoma with increasing clinical Relevance. Urologe A 2001;40:
3847.
Yazaki T, Takahashi S, Ogawa Y, et al. Large
renal hemangioma necessitating nephrectomy.
Urology 1985;25:3024.
Fujii Y, Ajima J, Oka K, et al. Benign renal tumors detected among healthy adults by abdominal ultrasonography. Eur Urol 1995;27:
1247.
Stanley RJ, Cubillo E, Mancilla Jimenez R, et al.
Cavernous hemangioma of the kidney. Am J
Roentgenol Radium Ther Nucl Med 1975;125:
6827.
Gordon R, Rosenmann E, Barzilay B, et al.
Correlation of selective angiography and pathology in cavernous hemangioma of the kidney. J Urol 1976;115:6089.
Cubillo E, Hesker AE, Stanley RJ. Cavernous
hemangioma of the kidney: an angiographicpathologic correlation. J Can Assoc Radiol 1973;
24:2546.
Roswell RH. Renin-secreting tumors. J Okla State
Med Assoc 1990;83:579.
Dunnick NR, Hartman DS, Ford KK, et al. The
radiology of juxtaglomerular tumors. Radiology 1983;147:3216.
Haab F, Duclos JM, Guyenne T, et al. Renin
secreting tumors: diagnosis, conservative surgical approach and long-term results. J Urol 1995;
153:17814.
Niikura S, Komatsu K, Uchibayashi T, et al. Juxtaglomerular cell tumor of the kidney treated
with nephron-sparing surgery. Urol Int 2000;
65:1602.
Weiss JP, Pollack HM, McCormick JF, et al.
Renal hemangiopericytoma: surgical, radiological and pathological implications. J Urol 1984;
132:3379.
Matsuda S, Usui M, Sakurai H, et al. Insulin-like
growth factor II-producing intraabdominal
[65]
[66]
[67]
[68]
[69]
[70]
[71]
[72]
[73]
[74]
[75]
[76]
[77]
[78]
[79]
[80]
[81]
[82]
hemangiopericytoma associated with hypoglycemia. J Gastroenterol 2001;36:8515.

Chung J, Henry RR. Mechanisms of tumorinduced hypoglycemia with intraabdominal
hemangiopericytoma. J Clin Endocrinol Metab
1996;81:91925.
Motzer RJ, Bander NH, Nanus DM. Renal-cell
carcinoma. N Engl J Med 1996;335:86575.
Smith SJ, Bosniak MA, Megibow AJ, et al. Renal
cell carcinoma: earlier discovery and increased
detection. Radiology 1989;170:699703.
Filipas D, Spix C, Schulz-Lampel D, et al.
Screening for renal cell carcinoma using ultrasonography: a feasibility study. BJU Int 2003;
91(7):5959.
Tsuboi N, Horiuchi K, Kimura G, et al. Renal
masses detected by general health checkup.
Int J Urol 2000;7:4048.
Ishikawa I, Kovacs G. High incidence of papillary renal cell tumours in patients on chronic
haemodialysis. Histopathology 1993;22:1359.
Sasagawa I, Nakada T, Kubota Y, et al. Renal cell
carcinoma in dialysis patients. Urol Int 1994;
53:7981.
Choyke PL, Glenn GM, Walther MM, et al.
Hereditary renal cancers. Radiology 2003;226:
3346.
Steffens MG, de Mulder PH, Mulders PF. Paraneoplastic syndromes in three patients with renal cell carcinoma. Ned Tijdschr Geneeskd 2004;
148:48792.
Gold PJ, Fefer A, Thompson JA. Paraneoplastic
manifestations of renal cell carcinoma. Semin
Urol Oncol 1996;14:21622.
Kim HL, Belldegrun AS, Freitas DG, et al. Paraneoplastic signs and symptoms of renal cell
carcinoma: implications for prognosis. J Urol
2003;170:17426.
Gil H, de Wazieres B, Desmurs H, et al.
Stauffers syndrome disclosing kidney cancer:
another cause of inflammatory syndrome with
anicteric cholestasis. Rev Med Interne 1995;16:
7757.
Sarf I, el Mejjad A, Dakir M, et al. Stauffer
syndrome associated with a giant renal tumor.
Prog Urol 2003;13:2902.
Dourakis SP, Sinani C, Deutsch M, et al. Cholestatic jaundice as a paraneoplastic manifestation of renal cell carcinoma. Eur J Gastroenterol
Hepatol 1997;9:3114.
Ritchie AW, Chisholm GD. The natural history of renal carcinoma. Semin Oncol 1983;10:
390400.
Amendola MA, Bree RL, Pollack HM, et al.
Small renal cell carcinomas: resolving a diagnostic dilemma. Radiology 1988;166:63741.
Zagoria RJ, Wolfman NT, Karstaedt N, et al. CT
features of renal cell carcinoma with emphasis
on relation to tumor size. Invest Radiol 1990;
25:2616.
Catalano C, Fraioli F, Laghi A, et al. Highresolution multidetector CT in the preoperative
39
40
Paspulati & Bhatt
[83]
[84]
[85]
[86]
[87]
[88]
[89]
[90]
[91]
[92]
[93]
[94]
[95]
[96]
[97]
[98]
[99]
evaluation of patients with renal cell carcinoma.

Helenon O, Correas JM, Balleyguier C, et al.
Ultrasound of renal tumors. Eur Radiol 2001;
11:1890901.
Yamashita Y, Takahashi M, Watanabe O, et al.
Small renal cell carcinoma: pathologic and radiologic correlation. Radiology 1992;184:4938.
Forman HP, Middleton WD, Melson GL, et al.
Hyperechoic renal cell carcinomas: increase in
detection at US. Radiology 1993;188:4314.
Coleman BG, Arger PH, Mulhern Jr CB, et al.
Gray-scale sonographic spectrum of hypernephromas. Radiology 1980;137:75765.
Jinzaki M, Ohkuma K, Tanimoto A, et al. Small
solid renal lesions: usefulness of power Doppler
US. Radiology 1998;209:54350.
Ascenti G, Zimbaro G, Mazziotti S, et al. Usefulness of power Doppler and contrast- enhanced sonography in the differentiation of
hyperechoic renal masses. Abdom Imaging 2001;
26:65460.
Hartman DS, Davis Jr CJ, Johns T, et al. Cystic
renal cell carcinoma. Urology 1986;28:14553.
Brinker DA, Amin MB, de Peralta-Venturina M,
et al. Extensively necrotic cystic renal cell carcinoma: a clinicopathologic study with comparison to other cystic and necrotic renal cancers.
Am J Surg Pathol 2000;24:98895.
Murad T, Komaiko W, Oyasu R, et al. Multilocular cystic renal cell carcinoma. Am J Clin
Pathol 1991;95:6337.
Nassir A, Jollimore J, Gupta R, et al. Multilocular cystic renal cell carcinoma: a series of
12 cases and review of the literature. Urology
2002;60:4217.
Kim JC, Kim KH, Lee JW. CT and US findings of
multilocular cystic renal cell carcinoma. Korean
J Radiol 2000;1:1049.
Yamashita Y, Miyazaki T, Ishii A, et al. Multilocular cystic renal cell carcinoma presenting
as a solid mass: radiologic evaluation. Abdom
Imaging 1995;20:1648.
Kim AY, Kim SH, Kim YJ, et al. Contrastenhanced power Doppler sonography for the
differentiation of cystic renal lesions: preliminary study. J Ultrasound Med 1999;18:5818.
Habboub HK, Abu-Yousef MM, Williams RD,
et al. Accuracy of color Doppler sonography in
assessing venous thrombus extension in renal
cell carcinoma. AJR Am J Roentgenol 1997;168:
26771.
McGahan JP, Blake LC, deVere White R, et al.
Color flow sonographic mapping of intravascular extension of malignant renal tumors. J Ultrasound Med 1993;12:4039.
Tsui KH, Shvarts O, Smith RB, et al. Prognostic
indicators for renal cell carcinoma: a multivariate analysis of 643 patients using the revised
1997 TNM staging criteria. J Urol 2000;163:
10905.
Cheville JC, Lohse CM, Zincke H, et al. Sarco-
matoid renal cell carcinoma: an examination of

underlying histologic subtype and an analysis of
associations with patient outcome. Am J Surg
Pathol 2004;28:43541.
[100] Leder RA, Dunnick NR. Transitional cell carcinoma of the pelvicalices and ureter. AJR Am J
Roentgenol 1990;155:71322.
[101] Blacher EJ, Johnson DE, Abdul-Karim FW, et al.
Squamous cell carcinoma of renal Pelvis.
Urology 1985;25:1246.
[102] Stein A, Sova Y, Lurie M, et al. Adenocarcinoma
of the renal pelvis. Report of two cases, one
with simultaneous transitional cell carcinoma
of the bladder. Urol Int 1988;43:299301.
[103] Sijmons RH, Kiemeney LA, Witjes JA, et al.
Urinary tract cancer and hereditary nonpolyposis colorectal cancer: risks and screening options. J Urol 1998;160:46670.
[104] Nocks BN, Heney NM, Daly JJ, et al. Transitional cell carcinoma of renal pelvis. Urology
1982;19:4727.
[105] Yousem DM, Gatewood OM, Goldman SM,
et al. Synchronous and metachronous transitional cell carcinoma of the urinary tract: prevalence, incidence, and radiographic detection.
Radiology 1988;167:6138.
[106] Wong-You-Cheong JJ, Wagner BJ, Davis CJ. Transitional cell carcinoma of the urinary tract: radiologic-pathologic correlation. Radiographics
1998;18:12342.
[107] Baron RL, McLennan BL, Lee JKT, et al. Computed tomography of transitional cell carcinoma
of the renal pelvis and ureter. Radiology 1982;
144:12530.
[108] Kawashima A, Vrtiska TJ, LeRoy AJ, et al. CT
urography. Radiographics 2004;24:3554.
[109] Joffe SA, Servaes S, Okon S, et al. Multi-detector
row CT urography in the evaluation of hematuria. Radiographics 2003;23(6):144155.
[110] Subramanyam BR, Raghavendra BN, Madamba
MR. Renal transitional cell carcinoma: sonographic and pathologic correlation. J Clin Ultrasound 1982;10:20310.
[111] Kobayashi S, Ohmori M, Akaeda T, et al.
Primary adenocarcinoma of the renal pelvis.
Report of two cases and brief review of literature. Acta Pathol Jpn 1983;33:58997.
[112] Wimbish KJ, Sanders MM, Samuels BI, et al.
Squamous cell carcinoma of the renal pelvis:
case report emphasizing sonographic and CT
appearance. Urol Radiol 1983;5:2679.
[113] Kenney PJ. Imaging of chronic renal infections.
[114] Kim J. Ultrasonographic features of focal
xanthogranulomatous pyelonephritis. J Ultrasound Med 2004;23:40916.
[115] Choyke PL, White EM, Zeman RK, et al. Renal
metastases: clinicopathologic and radiologic correlation. Radiology 1987;162:35963.
[116] Bhatt GM, Bernardino ME, Graham Jr SD. CT
diagnosis of renal metastases. J Comput Assist
Tomogr 1983;7:10324.
[117] Mitnick JS, Bosniak MA, Rothberg M, et al.

Metastatic neoplasm to the kidney studied
by computed tomography and sonography.
J Comput Assist Tomogr 1985;9:439.
[118] Becker WE, Schellhammer PF. Renal metastases
from carcinoma of the lung. Br J Urol 1986;
58:4948.
[119] Volpe JP, Choyke PL. The radiologic evaluation
of renal metastases. Crit Rev Diagn Imaging
1990;30:21946.
[120] Hartman DS, Davidson AJ, Davis Jr CJ, et al.
Infiltrative renal lesions: CT- sonographicpathologic correlation. AJR Am J Roentgenol
1988;150:10614.
[121] Paivanalo M, Tikkakoski T, Merikanto J, et al.
Radiologic findings in renal metastases. Aktuelle Radiol 1993;3:3605.
[122] Dalla Palma L, Pozzi Mucelli RS, Zuiani C.
Ultrasonography and computerized tomography in the diagnosis of renal metastasis. Radiol
Med (Torino) 1991;82:95100.
[123] Honda H, Coffman CE, Berbaum KS, et al. CT
analysis of metastatic neoplasms of the kidney.
Comparison with primary renal cell carcinoma.
Acta Radiol 1992;33:3944.
[124] Pickhardt PJ, Lonergan GJ, Davis Jr CJ, et al.
From the archives of the AFIP. Infiltrative renal lesions: radiologic-pathologic correlation.
Armed Forces Institute of Pathology. Radiographics 2000;20:21543.
[125] Fernandez-Acenero MJ, Galindo M, Bengoechea
O, et al. Primary malignant lymphoma of the
[126]
[127]
[128]
[129]
[130]
[131]
[132]
[133]
kidney: case report and literature review. Gen

Diagn Pathol 1998;143:31720.
Porcaro AB, DAmico A, Novella G, et al. Primary lymphoma of the kidney. Report of a case
and update of the literature. Arch Ital Urol Androl 2002;74:447.
Heiken JP, Gold RP, Schnur MJ, et al. Computed tomography of renal lymphoma with
ultrasound correlation. J Comput Assist Tomogr
1983;7:24550.
Cohan RH, Dunnick NR, Leder RA, et al.
Computed tomography of renal lymphoma.
J Comput Assist Tomogr 1990;14:9338.
Ferry JA, Harris NL, Papanicolaou N, et al.
Lymphoma of the kidney. A report of 11 cases.
Am J Surg Pathol 1995;19:13444.
Richards MA, Mootoosamy I, Reznek RH, et al.
Renal involvement in patients with nonHodgkins lymphoma: clinical and pathological
features in 23 cases. Hematol Oncol 1990;8:
10510.
Hartman DS, Davidson AJ, Davis Jr CJ, et al.
Infiltrative renal lesions: CT-sonographicpathologic correlation. AJR Am J Roentgenol
1988;150:10614.
Horii SC, Bosniak MA, Megibow AJ, et al.
Correlation of CT and ultrasound in the evaluation of renal lymphoma. Urol Radiol 1983;5:
6976.
Sheeran SR, Sussman SK. Renal lymphoma:
spectrum of CT findings and potential mimics.
41
43
ULTRASOUND
CLINICS
Prostate Ultrasound: Past, Present,

and Future
Sarah E. McAchran,
&
&
&
&
MD,
Martin I. Resnick,
Ultrasonographically guided biopsy

Benign prostatic hyperplasia
Prostate volume
Prostatic abscess
The first application of ultrasound in medical

diagnosis occurred in 1942 when Karl Dussik [1]
placed transducers on patients craniums and recorded the through-transmission of the sound
beam in an attempt to localize a brain tumor. In
1949, the first successful pulse echo system, which
could record echoes from tissue interfaces, was
developed by Douglas Howry and W. Roderic
Bliss at the University of Colorado School of Medicine [2]. In 1951, Wild and Neal [3] of St. Martins
Hospital reported that A-mode display ultrasonography was capable of showing differences between normal and abnormal tissues. The following
year, Wild and Reid [4] published the first twodimensional echograms of biologic tissues (kidney
cortex and myoblastoma of the thigh). In 1963,
Takahashi and Ouchi [5] were the first investigators
to report the use of ultrasound to image the prostate by way of an abdominal approach, but unfortunately these A-mode display images were difficult
to interpret and not clinically useful. One year later,
these investigators successfully obtained tomographic pictures of the prostate with the use of a
transrectal probe equipped with a radial scanning
device [6]. Although resolution was improved, the
images also were of poor quality and of no clinical value.
Watanabe and associates [7], using a chair device,
are credited with obtaining the first clinically useful
&
&
&
MD*
Future
Summary
References
transrectal ultrasonographic images of the prostate

in 1967 [Fig. 1]. Remarkable improvement in ultrasonography of the prostate was obtained with the
use of a transrectal radial probe with a special
concave 3.5-MHz transducer covered with a waterfilled balloon [8]. A B-mode display was used and
reproducible transverse prostate images were visualized on a black and white screen. Transrectal ultrasound scanning was also evaluated by VonMicsky
[9], a gynecologist who developed an ultrasound
scanning device in conjunction with a sigmoidoscope to visualize pelvic structures.
The first reports of investigators in the United
States using transrectal ultrasound for evaluation
of prostate came in 1972, and the images shown
in the cathode ray tube were recorded with a Polaroid camera [Fig. 2] [10,11]. These initial reports
were the culmination of several experiments that
were conducted beginning in 1968 at the Bowman Gray School of Medicine under the direction of William H. Boyce [1214], who described
these early developments. Initially studies were
performed with skin-coupled B-mode transducers
through the perineum and other portals before
the final decision was made to use an intrarectal
balloon-coupled rotational array transducer. It was
mounted on a stable tripod, adjustable in height
to the cystoscopic table, and capable of fixation at
virtually any angle selected by the operator as the
Department of Urology, University Hospitals of Cleveland, 11100 Euclid Avenue, Cleveland, OH 44106-5046, USA
E-mail address: martin.resnick@case.edu (M.I. Resnick).
doi:10.1016/j.cult.2005.09.003
44
McAchran & Resnick
Fig. 1. Early chair device with a rectal probe for prostate imaging. Refinements in instrumentation simplified the examination, but all studies continued using
the chair device. (From Resnick MI. Ultrasonography of
the prostate and testes. J Ultrasound Med 2003;22:
86977; with permission.)
planned position of reference [Fig. 3]. The probe

had a ratchet and a scale, which permitted the
planned position to become zero on the scale.
The ratchet then moved the transducer a precise
distance (15 mm) along the axis of the probe,
permitting serial tomograms of the prostate and
related structures. These loaf-of-bread slices permitted a three-dimensional reconstruction of the
prostate gland [Fig. 4].
The first transducer was the 3.5-MHz generator
receiver mounted at 90 from the long axis of the
probe [Fig. 5]. A single 360 sweep at a precise rate
was made for each activation of an electronic
motor, providing highly reproducible static sonograms. A plastic sheet protected the balloon from
contact with the rotating element. The balloon covered the entire insertable probe and was filled with
fluid through a tube in the probe shaft.
In 1968 and 1969, work progressed on construction of prototype equipment and attempts to solve
the apparently simple problems of air bubbles;
detection of highly absorptive interrectal interfaces;
and establishment of reliable points of reference
(anatomic and ultrasonographic). More difficult
problems included those of focusing, energy control, and permanent records from a cathode ray
tube. Placing the patient in the knee-chest position ultimately eliminated the air bubble problem.
However, there was no electronic conversion ca-
Fig. 2. Early transrectal equipment with a cathode ray

tube and tripod device to position the probe. Early
studies used fluoroscopy to document probe position
and to correlate anatomic structures visualized with
ultrasonographic images. (From Resnick MI. Ultrasonography of the prostate and testes. J Ultrasound Med
2003;22:86977; with permission.)
pability, and it was essential to have a roadmap

orientation of the sonogram to cystoscopic, fluoroscopic, radiologic, and physical examination of the
patient. Several coupling agents and techniques
were investigated, but none was satisfactory. Virtually everything done was repeated in all reasonable variations, modes, and patient positions.
In late 1969 and early 1970, the equipment was
assembled in the Urology Department, and a proto-
Fig. 3. Patient in the lithotripsy position with the

probe in place. The probe is mounted on a tripod
with a planned position indicator. (From Resnick MI.
Ultrasonography of the prostate and testes. J Ultrasound Med 2003;22:86977; with permission.)
Prostate Ultrasound
Fig. 4. (A) Diagram showing successive cephalad and caudad movement of the probe for imaging the prostate
in the transverse plane. (B) Typical examination. Serial scans at 4-mm intervals beginning at the urinary bladder (upper left) and progressing to the apex of the prostate are shown. The patient is in the lithotomy position,
and the rectum is at the bottom of each scan. (From Resnick MI. Ultrasonography of the prostate and testes.
J Ultrasound Med 2003;22:86977; with permission.)
col was developed for standardized examinations.

Examinations were performed on patients who had
well-defined urologic disease who required fluoroscopic examinations, radiographic examinations, or
both, and cystoscopic evaluations. In addition to
Dr. William Boyce, Dr. William King, Dr. Mark
Wilkiemeyer, and James Willard conducted these
early investigations. Dr. James Martin, Dr. William
McKinney, and Dr. Fred Kremkau also lent their

radiologic and ultrasonic experience to these early
studies. In 1972, a scan converter and video screen
with a capacity of 8 to 10 shades of gray were obtained, with considerable improvement in image
quality [Fig. 6]. For the three ensuing years from
Fig. 5. Initial studies were performed with a 3.5-MHz

transducer mounted on a rotating probe. (From Resnick
MI. Ultrasonography of the prostate and testes. J Ultrasound Med 2003;22:86977; with permission.)
Fig. 6. First scan converter and video display, 1972. The

capacity was 8 to 10 shades of gray. (From Resnick MI.
Ultrasonography of the prostate and testes. J Ultrasound Med 2003;22:86977; with permission.)
45
46
McAchran & Resnick
Fig. 7. (A) Scan of prostate with benign hyperplasia using a 3.5-MHz transducer. (B) Scan of prostate with malignancy (arrow) using a 3.5-MHz transducer. (From Resnick MI. Ultrasonography of the prostate and testes.
J Ultrasound Med 2003;22:86977; with permission.)
1972 to 1975, many studies were performed to

assess urologic diseases of the pelvis. Data were
recoded using 35-mm filmstrips, and the first clinical research program began in 1972 studying patients who had localized carcinoma of the prostate.
Recognition of advanced stages of prostate cancer
was noted, but early stage disease could not be
detected. Furthermore, it was noted that the locations of histologic cancer in radical prostatectomy
specimens were often at variance with the sonographic impressions [Fig. 7].
In 1975, gray-scale equipment with more than 18
shades of gray was acquired, and studies ensued
addressing early diagnosis of prostate cancer; accurate staging of cancer of the bladder and prostate;
and monitoring the response of prostate cancer to
therapy. The National Prostate Cancer Project of
the National Cancer Institute funded much of this
work, with Dr. Martin Resnick as the principle
investigator. The studies recognized that higher-
frequency transducers were required, and with the

assistance of Dr. Ronald Heilman, higher-frequency
transducers of up to 7 MHz were used. Gray-scale
quality was further enhanced by increasing the
number of shades of gray produced by the scan
converter, and the use of higher-frequency transducers improved the resolution of ultrasonographic
imaging considerably [Figs. 8 and 9].
The introduction of electronic real-time imaging
in the early 1980s further enhanced the quality of
ultrasonographic images [Fig. 10] [15]. Shortly
thereafter, the clinical applications of transrectal
ultrasonography of the prostate began to expand,
as evidenced by the dramatic increase in publications on transrectal ultrasonography of the prostate
in the medical literature during the mid-1980s.
Most investigators concluded that the transrectal
techniques were preferable to the transabdominal,
transperineal, and transurethral techniques when
visualizing the prostate because of more consistent
and reproducible imaging. These alternative techniques often provided poor definition of the prostate margins depending on the approach and were
dependent on bladder distension to improve image
quality. Currently, transrectal techniques are used
almost exclusively. The future of prostate ultrasound will involve improvements in image processing; application of color and power Doppler
imaging; the development and use of ultrasonographic contrast agents; and the application of
three-dimensional imaging. All of these methods
promise to enhance the reliability and applicability
of this examination and are discussed later.
Ultrasonographically guided biopsy

Fig. 8. Transrectal probe with 3.5- and 5.0-MHz transducers to enhance imaging of the prostate. Only one
frequency was used at a time. (From Resnick MI. Ultrasonography of the prostate and testes. J Ultrasound
Med 2003;22:86977; with permission.)
The office-based performance of ultrasound-guided

prostate biopsies has become the primary modality
used to diagnose prostate adenocarcinoma in patients who have an increased prostate-specific antigen (PSA) or abnormal digital rectal examination.
Prostate Ultrasound
Fig. 9. (A) Scan of the prostate with benign hyperplasia using a 5.0-MHz transducer. (B) Scan of the prostate with
benign hyperplasia using a 3.5-MHz transducer. (From Resnick MI. Ultrasonography of the prostate and
testes. J Ultrasound Med 2003;22:86977; with permission.)
The first investigators to use transrectal ultrasonographic guidance performed a biopsy by way of the
transperineal route with the use of a radial scanner
equipped with a special puncture attachment containing holes aligned parallel to the transducer
[Fig. 11] [16]. Interestingly, Holm and colleagues
[17] were the first to report the use of this technique for the placement of radioactive seeds within
the prostate for treating localized cancer of the
prostate. This method of seed implantation continues today. Other investigators have used the longitudinal linear array transducer to perform the
biopsy by way of the perineal route [1820]. Popularized by Lee and associates [2124], more recently transrectal ultrasonographic guidance has
been used to perform transrectal biopsies and
has become the standard in todays practice. Transrectal techniques are preferred over transperineal approaches because the path of the needle
is shorter and deviation of the needle within the
prostate is less likely to occur, thus making positioning easier and more accurate. The procedure is
Fig. 10. Scan of the prostate, 3.5-MHz transducer with

improved gray-scale imaging. (From Resnick MI. Ultrasonography of the prostate and testes. J Ultrasound
Med 2003;22:86977; with permission.)
quicker because there is no skin preparation, and

local anesthesia, although not required, is being
used with increasing frequency and has considerably reduced pain associated with the procedure.
Although the risk for infection is greater with a
transrectal approach, the use of prophylactic antibiotics has resulted in an infection rate of approximately 1%. Improvements in biopsy needles
and the use of spring-loaded biopsy guns have
improved prostate tissue samples and have also
lessened patient discomfort.
The descriptions of prostate anatomy by Dr. John
McNeal [25,26] had a considerable impact on the
technique of prostate biopsy. McNeal described
the prostate as being composed of three glandular
zones (peripheral, transitional, and central) and
the periurethral glands. The peripheral zone accounts for approximately 75% of the glandular
tissue of the normal prostate and is the most frequent site of origin of carcinoma. With aging there
is appreciable growth of the transition zone, which
predominantly gives rise to benign prostatic hyperplasia. Initially McNeals concept of zonal anatomy
had little usefulness in the clinical sphere or in
Fig. 11. Biopsy technique as described by Holm and

colleagues. A similar technique is used for brachytherapy. (From Resnick MI. Ultrasonography of the
prostate and testes. J Ultrasound Med 2003;22:86977;
with permission.)
47
48
McAchran & Resnick
Fig. 12. (A) Normal prostate, sagittal view. C, central zone; P, peripheral zone; T, transitional zone; AF, anterior fibromuscular stroma. (B) Normal prostate, transverse view. PZ, peripheral zone; CZ/TZ, central zone/transitional zone.
diagnostic imaging; however, with the development of cross-sectional imaging techniques, the
zonal concept of anatomy became useful because
it not only permitted interpretation of sonographic
images of the internal structure of the prostate but
also assisted in guiding prostate biopsy [Fig. 12].
Cooner and colleagues [27] also brought attention
to the importance of zonal anatomy in the interpretation of ultrasonographic images and recognized
the potential value of prostate ultrasonography
in the detection and diagnosis of early cancer of
the prostate.
Exhaustive research has been aimed at determining the sonographic appearance of prostate cancer.
Early investigators demonstrated that tumors can
appear as hypoechoic areas in the peripheral zone
[Fig. 13]; therefore, it was hoped that transrectal
ultrasound would be able to locate malignant lesions for biopsy and serve as a noninvasive screening tool. Unfortunately, ultrasound proved to be
fairly disappointing in both regards [28]. The sonographic appearance of prostate cancer is varied, and
early stage lesions tend to be indistinct from normal prostate tissue. As PSA screening has evolved
and has led to the detection of early stage, lowvolume cancers, the strategy of first recognizing the
lesion on radiologic examination and then obtain-
Fig. 13. Prostate cancer appearing as a peripheral, hypoechoic lesion.
ing a biopsy of the lesion is not practical. Therefore,

the true use of transrectal ultrasound resides in its
ability to enable sampling of all relevant areas of
the prostate, including those that appear ultrasonically normal [29].
The normal prostate appears homogeneous with
a stipple gray echogenicity and a well-defined,
echogenic, continuous capsule. The zonal anatomy
can usually be identified. A distinct layer of echogenic fibrous tissue separates the transition zone
from the central and peripheral zones. In transverse
images the prostate appears symmetric with a semilunar shape. The periurethral tissue can be identified as hypoechoic and centrally located, whereas
the peripheral zone tissue has a fine, homogeneous
echo pattern. In longitudinal or sagittal images, the
relationship of the prostate to the bladder neck,
seminal vesicles, and normal prostatic urethra is
easily identified [Figs. 14 and 15]. In the midline
the urethra will appear as a curved structure within
the central portion of the gland. Oriented horizontally in the transverse plane, the seminal vesicles are
paired, symmetric, crescent-shaped structures that
appear less echogenic than the prostate and are
separated from the prostate base by hyperechoic
fatty tissue.
The practice of systematic, sextant biopsies was
described by Hodge and colleagues [30] in 1989
and remains the foundation of the prostate biopsy procedure today. The technical aspects of
transrectal ultrasound-guided prostate biopsy are
straightforward. Before the procedure, patients
should discontinue all anticoagulants. Recent studies have demonstrated that prebiopsy cleansing
enemas do not significantly reduce the incidence
of postprocedure infection and can be abandoned
[31]. Traditionally 2 to 3 days of prophylactic antibiotics have been prescribed but recent studies have
demonstrated that a single dose of a long-acting
fluoroquinolone 1 hour before the procedure is
sufficient antibiotic coverage and does not increase
the postprocedure infection rate of 1% [3234].
Prostate Ultrasound
Fig. 14. Prostatic ultrasound and biopsy, transverse. The sound beam is directed from the tip of the biplanar, or
end-fire probe. Manipulating the probe in the anteriorposterior direction will image the prostate in the
transverse section from the apex to the base, as shown. (From Torp-Pedersen ST, Lee F. Transrectal biopsy of
the prostate guided by transrectal ultrasound. Urol Clin N Am 1989;16(4):70312; with permission.)
With the patient in either the lateral decubitus

or dorsal lithotomy position, the well-lubricated
transrectal probe is guided into the rectum above
the anal verge and a thorough sonographic examination of the prostate is performed. Biplane, multiplane, and end-fire endorectal probes with a
frequency range from 6 to 8 MHz are used to image the prostate and adjacent structures, including
the seminal vesicles and urethra. Suspicious, hypoechoic areas should be recorded.
Periprostatic nerve blockade has recently been

studied as a localized and uncomplicated method
of decreasing the pain associated with prostate biopsy [3539]. Under ultrasound guidance a sterile
5-inch 22-gauge spinal needle is passed through
the biopsy needle port and positioned at the prostate base at the junction between the prostate and
seminal vesiclethe site of the neurovascular bundle. Before injecting 2.5 cm3 of 1% lidocaine, the
syringe is aspirated to ensure that intravascular in-
Fig. 15. Prostatic ultrasound and biopsy, longitudinal. Rotation of the probe 90 in either direction will reach the
longitudinal plane. Angling the probe from left to right in this plane will image the lateral borders of the
prostate. (From Torp-Pedersen ST, Lee F. Transrectal biopsy of the prostate guided by transrectal ultrasound. Urol
Clin N Am 1989;16(4):70312; with permission.)
49
50
McAchran & Resnick
Prostate volume
Fig. 16. The dotted line represents the needle guide,

and overlies the corresponding echogenic trail of
the needle.
jection will not occur, as this can cause seizures.

Periprostatic nerve blockade is performed bilaterally.
A sterile 18-gauge biopsy needle is placed into
a spring-loaded biopsy gun and advanced through
the needle guide attached to the ultrasound probe.
Using a marker line, available with most modern
ultrasound view monitors, any suspicious lesion
can be biopsied in either the sagittal or longitudinal plane [Fig. 16]. The rest of the prostate should
then be sampled in a systematic fashion, encompassing the full length of the peripheral zone laterally and anteriorly. There is considerable debate
over the number and location of core biopsies
necessary to detect prostate cancer, but the evidence
suggests that an extended biopsy technique, including 10 to 12 cores and more laterally placed biopsies, improves detection over the traditional sextant
technique [29,35,36].
Prostate ultrasound for the evaluation of benign

prostatic hyperplasia has minimal clinical utility,
except in providing an accurate volume assessment. Volume determination may help in planning the surgical management of the disease.
Measurement of prostate gland volume using ultrasound is accurate to within 5% of its true weight
[40]. It is far more accurate than estimation by
digital palpation. The length, width, and height of
the prostate are measured using the transverse and
longitudinal orientations. The volume of the prostate is then estimated using the formula for an
ellipsoid. Planimetry, which measures the surface
area of the prostate at regular intervals throughout the gland and is used to determine gland
volume, is the most accurate means of volume
measurement. This method accommodates individual variations in prostate shape. With its superior accuracy, planimetry is used when planning
brachytherapy treatment for prostate cancer.
Prostate volume is also used in the calculation
of PSA density, where serum PSA is interpreted
in relation to prostate size. PSA density may help
the clinician judge whether a PSA increase is
caused by BPH or malignancy. PSA density is calculated according to the formula: PSA density = PSA
(ng/cm3)/prostate volume (cm3). A PSA density
of 0.15 or greater is considered suspicious for cancer [41,42].
Prostatic abscess
Acute and chronic prostatitis are conditions that are
diagnosed by history, physical examination, and
urinalysis. The role of ultrasound is reserved for
situations in which acute prostatitis is complicated
by a prostatic abscess. Intravenous antibiotics may
penetrate the prostatic tissue poorly, and ultimately
ultrasound-guided drainage may be indicated.
Benign prostatic hyperplasia

Benign prostatic hyperplasia (BPH) is present in
50% of 50-year-old men, and the percentage increases with aging. Some men will remain asymptomatic, whereas others will develop symptoms of
bladder outlet obstruction. Ultrasonographically,
BPH is seen as the proliferation of transition zone
tissues, with compression of the adjacent peripheral zone [Fig. 17]. Occasionally, the peripheral
gland is so compressed that it is almost totally
displaced. The echotexture of the hyperplastic tissue is mixed, with regions of increased echogenicity
that may be foci of corpora amylacea. Corpora
amylacea lack the distal acoustic shadows that are
seen with true calcifications.
Fig. 17. Benign prostatic hyperplasia. The small arrow

indicates the hyperplastic transition zone.
Prostate Ultrasound
Sonographically, the prostatic abscess presents as a

complex collection of cystic and solid elements
within the parenchyma. The internal echoes represent cellular debris and necrotic material. It may
have a frankly fluid-looking appearance, or may be
more difficult to appreciate if it has more of the
characteristics of soft tissue. Where visualized, a
prostatic abscess may be drained by transrectally
guided transperineal or transrectal placement of a
wide-bore needle, or it may be resected transuretherally [43].
Future
The roles of three-dimensional, color, and power
Doppler ultrasound and the value of ultrasound
contrast materials for the evaluation of prostate
cancer are the subject of much current research
[44]. For tumors to grow, they require a blood
supply. This blood supply has been typified by
low, slow flow. Power Doppler enables these
small tumor vessels to be imaged by ultrasound
in real time [37]. Color Doppler analysis is performed during transrectal ultrasound with a probe
capable of color and pulsed Doppler. Increased
flow in the peripheral gland is considered abnormal and suggestive of malignancy [38]. However,
interpretation of this increased flow is not always
straightforward, and the use of color Doppler ultrasound to increase sensitivity and specificity of transrectal ultrasound remains to be proven [39,45].
Where color Doppler analysis may prove most
beneficial is in the planning of interstitial laser
treatment for BPH. With this treatment, a laser
fiber is placed into the prostate transurethrally
and allowed to heat the tissue to a predetermined
temperature for a predetermined length of time.
This heating causes the destruction of prostate
cells in an area surrounding the fiber, thereby wid-
ening the voiding channel and improving symptoms of urinary obstruction. The size and shape of
the area destroyed may strongly depend on the
vascular flow in that area. Therefore, using color
Doppler ultrasound to accurately quantify prostatic blood flow may help to select the optimal
thermal dose necessary to achieve the desired improvement in voiding parameters and symptoms
[39]. Furthermore, preoperative ultrasound with
color Doppler analysis may help to select patients
who would be optimal candidates for this type
of procedure.
Three recent studies have used intravenous ultrasound contrast in an attempt to enhance color
Doppler ultrasound of the prostate [4648]. Ultrasound contrast agents are injectable liquids that
yield circulating microbubbles measuring 1 to
5 m. These small air bubbles increase the echo
density of the blood, resulting in enhanced visualization of the vasculature and better spatial resolution of the ultrasound image [39]. They have
been used to characterize renal vascularity and
renal masses [49]. The study from Thomas Jefferson
University performed contrast injection and color
Doppler ultrasound in the office setting, demonstrating encouraging improvement in biopsy sensitivity from 38% to 65% without substantial loss
of specificity [48].
Similarly, three-dimensional ultrasound of the
prostate can be easily performed in the office setting [50]. A three-dimensional endorectal volume
transducer is substituted for the conventional twodimensional ultrasound probe and can be coupled
to a commercially available ultrasound unit. Automatic, planimetric volume scanning allows for
nearly immediate reconstruction and display of sectional anatomy in orthogonal and oblique planes.
It remains to be proven, but three-dimensional
ultrasound may be superior in depicting tumor
Fig. 18. Example of analysis of ultrasound image to obtain probability of malignancy. (A) Gray-scale ultrasound
image used for color coding. (B) Probability of malignancy. Red represents high probability and blue represents
low probability (From Aarnink RG, Beerlage HP, De La Rosette JJ, et al. Transrectal ultrasound of the prostate:
innovations and future applications. J Urol 1998;159:156879; with permission.)
51
52
McAchran & Resnick
presence and extraglandular extent of disease. Furthermore, it allows more accurate repeated measurement than does two-dimensional imaging. This
capability can be useful when accurate volume
assessments are required for dosimetry planning
for radiation therapy or for estimating prostatespecific antigen levels [51].
Although the above innovations deal with the
acquisition of ultrasound images, other areas of
research are focused on the interpretation of these
images. Computerized interpretation of ultrasound
images of the prostate may reveal information not
perceptible to the human eye. Abnormal structures
may have subtle changes in texture that can be
quantified on a structural, statistical, or spectral
basis. Using a neural network, these subtle changes
can be translated into color-coded images, where
different colors indicate different probabilities of
malignancy [Fig. 18] [52]. This method may help
to overcome the low predictive value of transrectal
ultrasound for the clinical diagnosis of prostate
cancer by better discriminating malignant from
benign. In one study from Memorial Sloan Kettering comparing prostate biopsy guided by traditional B-mode ultrasound with prostate biopsy
using spectrum analysis, the use of spectrum analysis significantly improved the sensitivity and specificity of prostate cancer detection [53].
It is no great leap to see how this would also
improve planning for brachytherapy for prostate
cancer. Because current prostatic imaging methods
cannot accurately identify cancerous regions within
the prostate, the whole prostate is irradiated, causing the inherent increase in the potential side
effects of radiation. Guidance for radiation therapy
using the described tissue characterization methods
would allow more effective targeting of cancerous
tissue and spare noncancerous tissue. Similarly, this
type of imaging could be applied to the evaluation
of men who have a rising PSA following definitive
treatment for prostate cancer. Spectrum analysis
based tissue characterization techniques have already shown promise for noninvasively monitoring
intraocular cancer treated with 60cobalt plaque
therapy [54].
Summary
Since its inception 60 years ago, there have been
profound advances in the field of prostatic ultrasound. Its major usefulness remains in the areas of
volume measurement and biopsy guidance; however, ongoing research into refinements of image
acquisition and processing promise to expand on
these applications. Equally brisk progress can be
expected in the next 60 years.
References
[1] Dussik K. Uber die Moglichkeit hochfrequente
mechanische Schwingungen als diagnostisches
Hilfsmittel zu verwenden [The use of high frequency ultrasound as a diagnostic tool]. Z Ges
Neurol Psychiatr 1942;174:153 [in German].
[2] Martin J. History of ultrasound. In: Resnick MI,
editor. Ultrasound in urology. Baltimore (MD):
Williams & Wilkins; 1984. p. 112.
[3] Wild JJ, Neal D. Use of high-frequency ultrasonic
waves for detecting changes of texture in living
tissues. Lancet 1951;1:6557.
[4] Wild JJ, Reid JM. Application of echo-ranging
techniques to the determination of structure of
biological tissues. Science 1952;115:22630.
[5] Takahashi H, Ouchi T. The ultrasonic diagnosis
in the field of biology. In: Japanese medicine and
ultrasonics: the first report, Vol 7. 1963.
[6] Takahashi H, Ouchi T. The ultrasonic diagnosis
in the field of urology. In: Proceedings of the
Fourth Meeting of the Japanese Society of Ultrasonics in Medicine 1964;2:35.
[7] Watanabe H, Kato H, Kato T, et al. Diagnostic
application of ultrasonography to the prostate.
Nippon Hinyokika Gakkai Zasshi 1968;59:2739.
[8] Watanabe H, Kaiho H, Tanaka M, et al. Diagnostic application of ultrasonotomography to the
prostate. Invest Urol 1971;8:54859.
[9] VonMicksy L. Gynecologic ultrasonography.
In: VonMicsky L, editor. Diagnostic ultrasound.
St. Louis (MO): CV Mosby Co.; 1974. p. 20741.
[10] Boyce W. Advances in urologic surgery. Med
World News 1972;13:2931.
[11] Boyce W. New test for prostate cancer. In: Health
bulletin: official publication of the North Carolina State Board of Health, Vol 87. :Raleigh (NC):
North Carolina State Board of Health; 1972. p. 6.
[12] Boyce W. History of prostatic ultrasonography. In:
Resnick M, editor. Prostatic ultrasonography.
Philadelphia: BC Decker Inc.; 1990. p. 115.
[13] King WW, Wilkiemeyer RM, Boyce WH, et al.
Current status of prostatic echography. JAMA
1973;226:4447.
[14] Boyce WH, McKinney WM, Resnick MI, et al.
Ultrasonography as an aid in the diagnosis and
management of surgical diseases of the pelvis:
special emphasis on the genitourinary system.
Ann Surg 1976;184:47789.
[15] Waterhouse RL, Resnick MI. The use of transrectal prostatic ultrasonography in the evaluation
of patients with prostatic carcinoma. J Urol
1989;141:2339.
[16] Holm HH, Gammelgaard J. Ultrasonically
guided precise needle placement in the prostate
and the seminal vesicles. J Urol 1981;125:3857.
[17] Holm HH, Juul N, Pedersen JF, et al. Transperineal 125iodine seed implantation in prostatic
cancer guided by transrectal ultrasonography.
J Urol 1983;130:2836.
[18] Rifkin MD, Kurtz AB, Goldberg BB. Sonographically guided transperineal prostatic biopsy:
Prostate Ultrasound
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
preliminary experience with a longitudinal lineararray transducer. AJR Am J Roentgenol 1983;

140:7457.
Fornage BD, Touche DH, Deglaire M, et al. Realtime ultrasound-guided prostatic biopsy using
a new transrectal linear-array probe. Radiology
1983;146:5478.
Abe M, Hashimoto T, Matsuda T, et al. Prostatic
biopsy guided by transrectal ultrasonography
using real-time linear scanner. Urology 1987;
29:5679.
Lee F, Littrup P, Torp-Pedersen S, et al. Transrectal US of prostate cancer with use of transrectal
guidance and an automatic biopsy system [abstract]. Radiology 1987;165:215.
Cooner WH, Mosley BR, Rutherford Jr CL, et al.
Clinical application of transrectal ultrasonography and prostate specific antigen in the search
for prostate cancer. J Urol 1988;139:75861.
Stamey T, Hodge K. Ultrasound visualization of
prostate anatomy and pathology. Monogr Urol
1988;9:5563.
Torp-Pedersen S, Lee F, Littrup PJ, et al. Transrectal biopsy of the prostate guided with
transrectal US: longitudinal and multiplanar
scanning. Radiology 1989;170:237.
McNeal JE. Normal and pathologic anatomy of
prostate. Urology 1981;17:116.
McNeal J. The prostate gland: morphology and
pathology. Monogr Urol 1983;4:333.
Cooner WH, Mosley BR, Rutherford Jr CL, et al.
Prostate cancer detection in a clinical urological
practice by ultrasonography, digital rectal examination and prostate specific antigen. J Urol 1990;
143:114652.
Shinohara K, Wheeler TM, Scardino PT. The
appearance of prostate cancer on transrectal ultrasonography: correlation of imaging and pathological examinations. J Urol 1989;142:7682.
Klein EA, Zippe CD. Transrectal ultrasound
guided prostate biopsydefining a new standard.
J Urol 2000;163:17980.
Hodge KK, McNeal JE, Terris MK, et al. Random
systematic versus directed ultrasound guided
transrectal core biopsies of the prostate. J Urol
1989;142:714.
Carey JM, Korman HJ. Transrectal ultrasound
guided biopsy of the prostate. Do enemas decrease clinically significant complications? J Urol
2001;166:825.
Terris MK, McNeal JE, Stamey TA. Ultrasonography
and biopsy of the prostate. In: Walsh PC, Retik AB,
Vaughan ED, et al, editors. Campbells urology.
8th edition. Philadelphia: WB Saunders; 2002.
p. 303854.
Aron M, Rajeev TP, Gupta NP. Antibiotic prophylaxis for transrectal needle biopsy of the
prostate: a randomized controlled study. BJU Int
2000;85:6825.
Griffith BC, Morey AF, Ali-Khan MM, et al. Single
dose levofloxacin prophylaxis for prostate biopsy
in patients at low risk. J Urol 2002;168:10213.
[35] Scherr DS, Eastham J, Ohori M, et al. Prostate

biopsy techniques and indications: when, where,
and how? Semin Urol Oncol 2002;20:1831.
[36] Eskew LA, Bare RL, McCullough DL. Systematic
5 region prostate biopsy is superior to sextant
method for diagnosing carcinoma of the prostate. J Urol 1997;157:199202.
[37] Cochlin DL, Dubbins PA, Goldberg BB, et al.
Urogenital ultrasound: a text atlas. Philadelphia:
JB Lippincott Co.; 1994.
[38] Rifkin MD, Sudakoff GS, Alexander AA. Prostate:
techniques, results, and potential applications
of color Doppler US scanning. Radiology 1993;
186:50913.
[39] Aarnink RG, Beerlage HP, De La Rosette JJ, et al.
Transrectal ultrasound of the prostate: innovations and future applications. J Urol 1998;
159:156879.
[40] Hastak SM, Gammelgaard J, Holm HH. Transrectal ultrasonic volume determination of the
prostatea preoperative and postoperative study.
J Urol 1982;127:11158.
[41] Horstman W, Watson L. Ultrasound of the genitourinary tract. In: Resnick MI, Older RA, editors.
Diagnosis of genitourinary disease. 2nd edition.
New York: Thieme; 1997.
[42] Benson MC, Whang IS, Olsson CA, et al. The use
of prostate specific antigen density to enhance
the predictive value of intermediate levels of
serum prostate specific antigen. J Urol 1992;
147:81721.
[43] Kinahan TJ, Goldenberg SL, Ajzen SA, et al.
Transurethral resection of prostatic abscess under
sonographic guidance. Urology 1991;37:4757.
[44] Goldman SM, Sandler CM. Genitourinary imaging: the past 40 years. Radiology 2000;215:
31324.
[45] Halpern EJ, Frauscher F, Strup SE, et al. Prostate:
high-frequency Doppler US imaging for cancer
detection. Radiology 2002;225:717.
[46] Roy C, Buy X, Lang H, et al. Contrast enhanced
color Doppler endorectal sonography of prostate: efficiency for detecting peripheral zone tumors and role for biopsy procedure. J Urol 2003;
170:6972.
[47] Frauscher F, Klauser A, Volgger H, et al. Comparison of contrast enhanced color Doppler
targeted biopsy with conventional systematic biopsy: impact on prostate cancer detection. J Urol
2002;167:164852.
[48] Halpern EJ, Rosenberg M, Gomella LG. Prostate
cancer: contrast-enhanced us for detection. Radiology 2001;219:21925.
[49] Robbin ML, Lockhart ME, Barr RG. Renal imaging with ultrasound contrast: current status.
Radiol Clin North Am 2003;41:96378.
[50] Hamper UM, Trapanotto V, DeJong MR, et al.
Three-dimensional US of the prostate: early experience. Radiology 1999;212:71923.
[51] Downey DB, Fenster A, Williams JC. Clinical
utility of three-dimensional US. Radiographics
2000;20:55971.
53
54
McAchran & Resnick
[52] Huynen AL, Giesen RJ, de la Rosette JJ, et al.

Analysis of ultrasonographic prostate images for
the detection of prostatic carcinoma: the automated urologic diagnostic expert system. Ultrasound Med Biol 1994;20:110.
[53] Balaji KC, Fair WR, Feleppa EJ, et al. Role
of advanced 2 and 3-dimensional ultrasound
for detecting prostate cancer. J Urol 2002;168:

24225.
[54] Lizzi L, Kalisz D, Coleman F, et al. Very-high
frequency ultrasonic imaging and spectral assays
of the eye. In: Lees S, Ferrara LA, editors. Acoustical imaging. New York: Plenum Press; 1998.
p. 10712.
55
ULTRASOUND
CLINICS
Sonographic Evaluation of
Testicular Torsion
Vikram S. Dogra, MD*, Shweta Bhatt, MD, Deborah J. Rubens, MD
&
&
&
&
&
Sonographic anatomy
Normal spectral Doppler
Clinical presentation
Testicular torsion
Extravaginal torsion
Intravaginal torsion
Gray-scale features
Color flow Doppler
Broadband high-frequency transducer sonography has become the gold standard for the evaluation of patients who have acute scrotal pain.
High-frequency transducer sonography can not
only help delineate the anatomic details of the
testes but also aids in evaluating testicular perfusion. Testicular perfusion can be studied with the
help of color or power Doppler sonography. Evaluation of testicular perfusion aids in the diagnosis
of testicular torsion. This article reviews the grayscale and color flow Doppler features of testicular
torsion, including the partial torsion and torsiondetorsion syndrome.
Sonographic anatomy
The postnatal human testis is intraperitoneal. The
adult human testis is also intraperitoneal but may
appear extraperitoneal. The apparent discrepancy
between the adult testis being intraperitoneal or
extraperitoneal is likely to result from differences
in the relative size of the tunica vaginalis between
infant boys and elderly men [1]. Testes are bilaterally symmetrical and housed within the scrotum.
&
&
&
&
Partial testicular torsion

Torsion-detorsion syndrome
Mimics of testicular torsion
Testicular appendigeal torsion
Summary
Acknowledgments
References
Testes are ovoid in shape with medium-level echoes

and measure 532 cm each. Tunica albuginea is
the fibrous covering of the testicle and is covered by
the tunica vaginalis. The tunica albuginea can be
seen with a high-frequency transducer as an echogenic line [Fig. 1]. Septa extend from the tunica
albuginea into the testicle, dividing the testis into
250 to 400 lobules. The posterior surface of the
tunica albuginea is reflected into the interior of the
testis, forming the incomplete septum known as
the mediastinum testis. Sonographically, the mediastinum testis is seen as an echogenic band running
in a cephalocaudal direction [Fig. 2]. Each lobule
consists of one to three seminiferous tubules supporting the Sertoli cells and the spermatocytes that
give rise to sperm. The seminiferous tubules open
through the tubuli recti into dilated spaces called
the rete testis within the mediastinum [Fig. 3]. The
rete testis drains into the epididymis through 10 to
15 efferent ductules. The epididymis, consisting
of a head, body, and tail, is located superior to
and is contiguous with the posterior aspect of the
testis. The tail of the epididymis continues as the
vas deferens.
University of Rochester School of Medicine and Dentistry, 601 Elmwood Ave, Rochester, NY 14642, USA
E-mail address: Vikram_Dogra@URMC.Rochester.Edu (V.S. Dogra).
doi:10.1016/j.cult.2005.09.006
56
Dogra et al
Fig. 3. Diagrammatic representation of the testis in

cross-section.
Fig. 1. Longitudinal sonogram of the testis reveals
homogeneous medium-level echoes. The arrow points
to the tunica albuginea.
Arterial blood supply to the testes is derived from

the testicular artery, a branch of the abdominal
aorta. The superior epididymal artery, a branch
of the testicular artery, predominantly supplies
the epididymis. The cremasteric artery, which is a
branch of the inferior epigastric artery, predominantly supplies the peritesticular tissue and anastomoses with the deferential artery. The deferential
artery, a branch of the superior vesicle artery, supplies the vas deferens [Fig. 4]. There are variable
anastomoses between the posterior epididymal,
deferential, and cremasteric arteries. A transmediastinal arterial branch of the testicular artery is
present in approximately half of normal testes; it
courses through the mediastinum to supply the
capsular arteries and is usually accompanied by a
large vein [Fig. 5] [2].
Scrotal US is performed with the patient in the
supine position and the scrotum supported by a
towel placed between the thighs. Optimal results
are obtained with a 10 to 14-MHz, high-frequency,
linear-array transducer. Scanning is performed with
Fig. 2. Longitudinal sonogram of the testis reveals an

echogenic band (arrow) consistent with mediastinum testis.
the transducer in direct contact with the skin; however, a stand-off pad can be used for evaluation of
superficial lesions if necessary.
The testes are examined in at least two planes: the
longitudinal and transverse axes. The size and echogenicity of each testis and of the epididymis are
compared with those on the opposite side. Scrotal
skin thickness is evaluated. Color Doppler and
pulsed Doppler parameters are optimized to display low-flow velocities and to demonstrate blood
flow in the testes and surrounding scrotal structures. Bilateral testicular spectral Doppler tracings
must be recorded. Power Doppler US may also be
used to demonstrate intratesticular blood flow in
patients who have an acute scrotum. In patients
being evaluated for an acute scrotum, the asymptomatic side should be scanned initially to set the
gray-scale and color Doppler gain settings to allow
comparison with the affected side, remembering
that testicular torsion can be a bilateral process in
2% of patients [3]. Transverse images with portions
of each testis on the same image should be acquired in gray-scale and color Doppler modes.
Additional techniques, such as use of the Valsalva
maneuver or upright positioning, can be used as
needed for venous evaluation.
Normal spectral Doppler

The proper interpretation of velocity waveforms
requires an understanding of the normal waveform
characteristics of a given vessel, and of the physiologic status of the circulation subtended by the
vessel. The normal spectral waveform of the testicular artery and artery supplying the epididymis
is a low-resistance, high-flow pattern [Fig. 6] [4],
whereas that of the cremasteric artery supplying the
scrotal wall has a high-resistance, low-flow pattern
[Fig. 7] [4].
Quantification of outflow resistance may be
achieved by determining the resistive index (RI).
In the testes of a normal healthy volunteer, the RI is
Sonography of Testicular Torsion
Fig. 4. Diagrammatic representation of the testicular blood supply.
rarely less than 0.5 [2]. It is the most widely used

parameter and is defined as the peak systolic velocity minus the end diastolic velocity, divided by
the peak systolic velocity.
Clinical presentation
In 1776, Hunter provided the first description of
testicular torsion [5]. The chances of torsion of the
testis or its appendage developing by the age of
25 years is about 1 in 160 [6]. Testicular torsion
can occur at any age; however, it is most frequent
in adolescent boys. Most testicular torsion occurs
in young patients, with 66% occurring between
12 and 18 years of age and peak incidence occurring at 14 years [7].

Patients who have acute testicular torsion present
after a sudden onset of pain followed by nausea,
vomiting, and a low-grade fever. Physical examination reveals a swollen, tender, and inflamed hemiscrotum. It is difficult to distinguish the testis from
the epididymis because of localized swelling. For
this reason, the condition is frequently misdiagnosed as epididymitis. The cremasteric reflex is
usually absent [8,9], and the pain cannot be relieved by elevating the scrotum [5]. In an adult in
a standing position, the normal testis hangs in a
near-vertical position, whereas the torsed testis will
tend to hang in a near-horizontal position [10].
Fig. 5. (A) Transverse oblique sonogram reveals a transmediastinal artery. (B) Demonstrates the arterial waveform
of the low-resistance pattern.
57
58
Dogra et al
Fig. 6. Longitudinal sonogram: (A) demonstrates the low-resistance, high-flow pattern of the intratesticular artery,
(B) demonstrates a similar pattern obtained from within the epididymis because anterior and posterior epididymal
arteries originate from the testicular artery.
Elevation and transverse location of the testis with

an anteriorly rotated epididymis associated with
loss of the ipsilateral cremasteric reflex strongly
suggests testicular torsion [11].
Testicular torsion
Extravaginal torsion
Two types of torsion have been described: extravaginal and intravaginal. Extravaginal testicular torsion occurs exclusively in newborns. Torsion occurs
outside the tunica vaginalis when the testes and
gubernacula are not fixed and are free to rotate
[12]. The affected neonate presents with swelling,
discoloration of the scrotum on the affected side,
and a firm, painless mass in the scrotum [13]. The
testis is typically infarcted and necrotic at birth. US
findings include an enlarged heterogeneous testis,
ipsilateral hydrocele, skin thickening, and no color

Doppler flow signal in the testis or spermatic cord
[Fig. 8] [14]. In children, power Doppler US is
more sensitive than color Doppler US for detection
of intratesticular blood flow. In one study [15],
power Doppler US demonstrated intratesticular
blood flow in 66 (97%) of 68 testes, whereas
color Doppler US demonstrated intratesticular
blood flow in 60 (88%) of the testes. Both techniques combined depicted blood flow in all 68
(100%) testes. Color Doppler US and scintigraphy
are comparable with regard to the diagnosis of
torsion in adolescent and adult populations [16].
Scintigraphy remains a reasonable alternative to
color flow Doppler for evaluation of acute scrotal
pain and should be used when color Doppler US
sensitivity for low-velocity, low-volume testicular
blood flow is inadequate and the diagnosis of torsion remains in question.
Intravaginal torsion
Fig. 7. Cremasteric artery at the periphery of the epididymis reveals a high-resistance, low-flow pattern of
spectral waveform.
Intravaginal torsion occurs within the tunica vaginalis and is the most frequent type of testicular
torsion seen in 80% of the population [17]. The
predisposing factors include a long mesorchium or
a bell-clapper deformity in which the tunica vaginalis completely encircles the epididymis, distal
spermatic cord, and testis rather than attaching to
the posterolateral aspect of the testis. This deformity leaves the testis free to swing and rotate within
the tunica vaginalis, much like a clapper inside a
bell [18]. The bell-clapper deformity can be diagnosed by US in the presence of moderate hydrocele
[Fig. 9]. In most cases, the bell-clapper deformity is
bilateral. A 12% prevalence of bell-clapper deformity was found in one autopsy series [19], thereby
suggesting that it is a more common deformity
than intravaginal testicular torsion. Testicular torsion, caused by long mesorchium, is associated
Fig. 8. (A) Sonogram of the testis in a newborn reveals the absence of blood flow within the testis (T) and the
presence of a small hydrocele (arrow). (B) Corresponding spectral Doppler tracing further demonstrates the
absence of blood flow (only noise is observed).
with cryptorchism [Fig. 10]. The most frequently

found anatomical relation between the testis and
the epididymis in patients who have testicular torsion is type I (epididymis united to the testis by its
head and tail) [17].
Gray-scale features
In testicular torsion, venous obstruction occurs
first, followed by obstruction of arterial flow, and
ultimately by testicular ischemia. The extent of testicular ischemia depends on the degree of torsion,
which ranges from 180 to 720 or greater. The
testicular salvage rate depends on the degree of
torsion and the duration of ischemia. A nearly
100% salvage rate exists within the first 6 hours
after the onset of symptoms, a 70% rate within
6 to 12 hours, and a 20% rate within 12 to
24 hours [20].
US findings vary with the duration and degree of
rotation of the spermatic cord. Gray-scale images
are nonspecific for testicular torsion [21] and often appear normal if the torsion has just occurred
[Fig. 11]. Testicular swelling and decreased echogenicity are the most commonly encountered findings 4 to 6 hours after the onset of torsion. At
24 hours after the onset, the testis has a heterogeneous echotexture secondary to vascular congestion, hemorrhage, and infarction. This condition
is referred to as late or missed torsion [Fig. 12].
The gray-scale features are summarized in Box 1.
An enlarged hypoechoic epididymal head may be
visible because the artery supplying the epididymis
is often involved in the torsion [22]. In a recent
prospective study [23], a spiral twisting of the spermatic cord at the external inguinal ring was seen
in 14 of 23 cases of torsion. The twisting induced
an abrupt change in the course, size, and shape of
the spermatic cord below the point of torsion and
appeared as a round or oval homogeneous extratesticular mass with or without blood flow that
could be traced cephalad to the normal spermatic
cord. In the setting of testicular torsion, normal
testicular echogenicity is a strong predictor of testicular viability [24]. Other indicators of testicular
torsion include the presence of scrotal wall thickening and reactive hydrocele.
Color flow Doppler
Fig. 9. Bell-Clapper Deformity in a 10-year-old male.

Sonogram reveals hydrocele (arrow) encircling the
distal third of the spermatic cord (asterisk). The child
underwent a bilateral orchiopexy. T, Testis.
Testicular perfusion can be evaluated by color Doppler, power Doppler, or spectral Doppler sonography.
Color Doppler sonography can reliably demonstrate
intratesticular flow [25]. Power Doppler sonography
uses the integrated power of the Doppler signal to
depict the presence of blood flow. Higher power
gains are more likely with power Doppler sonography than with standard color Doppler sonography, thereby resulting in increased sensitivity for
detecting blood flow. Power Doppler sonography
is valuable in scrotal sonography because of its
increased sensitivity to low-flow states and its independence from the Doppler angle correction [26].
Pulsed Doppler sonography is a useful method to
59
60
Dogra et al
Fig. 10. Testicular torsion in a cryptorchid testis. (A) Transverse gray-scale sonogram demonstrates the right testis
and empty left scrotal sac. (B) Transverse sonogram in the left inguinal region reveals a hypoechoic testis (within
calipers) with no color flow (C ).
Fig. 11. Color flow Doppler of the right testis in the

longitudinal plane reveals absent color flow with
normal gray-scale echotexture, suggestive of early
testicular torsion.
Fig. 12. Gray-scale sonogram of both testes in the

transverse plane demonstrates a normal gray-scale
appearance of the right testis. Left testis has heterogeneous echotexture with areas of increased echogenicity representing hemorrhage. This type of
appearance favors testicular torsion of more than at
least 6 hours.
Box 1: Gray-scale patterns of testicular torsion

Normal gray-scale appearance of testis; testis
may be enlarged
Hypoechoic pattern that may be total or
partial in the case of a partial infarct
Hyperechoic and heterogeneous echo
patterns; hemorrhagic infarct
Hypoechoic with small testis (chronic torsion
as end result)
identify flow in the testis using the timevelocity

spectrum to quantify blood flow [27]. The spectral
waveform of the intratesticular arteries characteristically has a low-resistance pattern [28] with a mean
RI of 0.62 (range, 0.480.75) [25]; however, this is
not true for testicular volumes less than 4 cm3 as are
often found in prepubertal boys when diastolic
arterial flow may not be detectable [29].
Because gray-scale US findings are often normal
in the early phases of torsion, the Doppler component of the examination is essential. The absence of
testicular flow on color and power Doppler US is
considered diagnostic of ischemia, provided that
the scanner is optimized for detection of slow
flow, is limited to the use of a small color-sampling
box, and is adjusted for the lowest repetition frequency and the lowest possible threshold setting
[30]. The threshold should be set just above the
level for detection of color noise.
The role of color Doppler and power Doppler US
in the diagnosis of acute testicular torsion is well
established [3133]. Using the absence of identifiable intratesticular flow as the only criterion for
detecting testicular torsion, color Doppler US was
86% sensitive, 100% specific, and 97% accurate in
the diagnosis of torsion and ischemia in painful
scrotum [Fig. 13] [16]. As it is sometimes possible
to record a small arterial signal in one part of the
Fig. 14. Testicular torsion. A 28-year-old man presented with sudden onset of scrotal pain of 5-hours
duration. Power Doppler examination revealed absent
blood flow in the right testis except for a tiny arterial
signal at the periphery. Please note the decreased
echogenicity of the right testis. Despite the presence
of minimal peripheral arterial flow, a diagnosis of
testicular torsion was advanced. The patient underwent surgical exploration and was found to have
bell-clapper deformity. Untwisting of the testis did
not result in return of normal blood flow to the testis.
Patient had to undergo orchiectomy.
testis [Fig. 14] [34,35] in the appropriate clinical

setting and age group, this should not dissuade one
from making the diagnosis of testicular torsion. The
presence of hypervascularity in paratesticular tissue is a characteristic finding of an infarcted testis
[Fig. 15] [36]. Color flow Doppler features of testicular torsion are summarized in Box 2.
Unilateral testicular torsion has bilateral effects
and is a form of ischemia-reperfusion injury. Treatment of torsion by detorsion alone does not prevent testicular damage and may result in infertility
[37]. If nonsalvagable, the necrotic testis is removed to decrease the risk for an autoimmune
reaction to the residual testis [38].
Partial testicular torsion
Fig. 13. Testicular torsion. Power Doppler reveals

absent flow in the left testis and a normal power
Doppler appearance in the right testis.
Torsion is not an all-or-nothing phenomenon but

may be complete, incomplete, or transient. Cases of
partial or transient torsion present a diagnostic
challenge. The ability of color and power Doppler
sonography to definitively diagnose partial testicular torsion remains uncertain. The role of spectral
Doppler US analysis is not well established with
regard to the diagnosis of partial torsion, but the
findings may be useful [39]. There are no studies
that validate the role of spectral Doppler US in
partial torsion; however, there are findings from
sporadic case reports [19,40] that suggest its usefulness. Asymmetry in resistive indices with decreased
diastolic flow or diastolic flow reversal may be seen
[Fig. 16]. The presence of color or power Doppler
61
62
Dogra et al
Fig. 15. Increased paratesticular blood flow. Color flow Doppler of the left testis demonstrates increased paratesticular blood flow (A) with absent blood flow within the left testis (B), consistent with testicular infarction.
signal in a patient who has the clinical manifestation of torsion does not exclude torsion [19,40].
Torsion-detorsion syndrome
Acute and intermittent sharp testicular pain and
scrotal swelling, interspersed with long intervals
without symptoms, are characteristic of torsiondetorsion. This type of presentation should arouse
the suspicion of torsion-detorsion syndrome [41].
Physical findings may include horizontal or very
mobile testes; an anteriorly located epididymis; or
bulkiness of the spermatic cord from partial twisting [41,42]. Horizontal testicular position, even in
the absence of pain at the time of physical examination, is a strong indication for exploration and
bilateral testicular fixation [43]. If scanned when
asymptomatic or immediately after detorsion, the
affected testis may demonstrate increased blood
flow [Fig. 17]. The testis may be enlarged, and
focal infarcts may or may not be present.
If the suspicion for torsion-detorsion is high, a
bell-clapper deformity can be expected on exploration and prophylactic orchiopexy should be performed bilaterally [44,45].
Box 2: Color flow Doppler patterns in

testicular torsion
Absent arterial and venous flow
Increased RI on affected side (diminished or
reversed diastolic flow)
Decreased flow velocity difficult to measure
because of small vessels/angle correction,
but may be subjectively inferred by the
relative difficulty in finding small, lowamplitude flow on the symptomatic side
Mimics of testicular torsion

Testicular infarction without torsion of the spermatic cord is a rare lesion, generally idiopathic
and spontaneous. Absence or decrease of color
flow within the testis is not always because of
testicular torsion and can result from vasculitis,
protein S, or antithrombin III deficiency [46].
Patients who have polyarteritis nodosa can present
with acute scrotum [47,48]. Other causes of decreased testicular perfusion include large hydroceles [Fig. 18] and extratesticular hematomas [49].
Idiopathic testicular infarcts also present with
acute pain, and testicular torsion is the main suspect [50]. These patients have a focal area of decreased echogenicity, usually anteriorly near the
upper pole. The conditions resulting in decreased
blood flow to the testis are listed in Box 3.
Testicular appendigeal torsion

Five testicular appendages are formed during the
development of the male genitourinary tract; these
are the remnants of the degenerating mesonephric
and paramesonephric ducts. The testicular and epididymal appendages, found at the upper pole of
the testis and at the head of the epididymis, respectively, are the most common [Fig. 19] [51].
Approximately 91% to 95% of twisted testicular
appendices involve the appendix testis and occur
most often in boys 7 to 14 years old. Patients who
have torsion of the appendix testis and appendix
epididymis present with acute scrotal pain, but
there are usually no other physical symptoms and
the cremasteric reflex can still be elicited. The classic
finding on physical examination is a small, firm,
palpable nodule on the superior aspect of the
Fig. 16. Partial testicular torsion. (A) Right testis spectral Doppler evaluation reveals decreased diastolic blood
flow. (B) In the same testis, the diastolic waveform can be seen below the baseline. This appearance is suggestive
of increased resistance to the inflow of arterial blood. The patient underwent surgical exploration and partial
torsion was confirmed. (From Dogra VS, Sessions A, Mevorach RA, et al. Reversal of diastolic plateau in partial
testicular torsion. J Clin Ultrasound 2001;29:1058; with permission.)
Fig. 17. Torsion-detorsion syndrome. The patient presented with a history of intermittent left testicular pain. He
was asymptomatic at the time of examination. (A) Color flow Doppler in the transverse plane demonstrates
increased blood flow to the left testis. Comparing spectral Doppler waveform of the left testis (B) with that of
the right testis (C ), it is evident that there is increased blood flow to the left testis.
63
64
Dogra et al
Fig. 18. Testicular torsion mimic. Patient who has right scrotal pain. (A) Gray-scale sonogram of the right testis
demonstrates a large hydrocele (H) pushing and causing pressure on the testis (arrow). (B) Color Doppler
evaluation of both testes side-by-side reveals decreased blood flow to the right testis. Blood flow returned to
normal after hydrocelectomy.
testis; it exhibits a bluish discoloration through

the overlying skin which is called the blue dot
sign [52].
US appearance of torsion of the appendages of the
testes is variable and may be isoechoic, hypoechoic,
or hyperechoic to the testis or epididymis [Fig. 20]
[53,54]. Reactive hydrocele and skin thickening
are common in these cases. An appendix testis size
of 5 mm or larger, spherical shape, and increased
periappendiceal blood flow are suggestive of a

torsed appendix testis [54]. The identification of a
testicular appendage larger than 5.6 mm is suggestive of torsion. Therefore, depending on a patients
clinical condition, such cases can be treated conservatively when an appendage larger than 5.6 mm is
identified [55]. The role of US examination in tor-
Box 3: Conditions that may result in decreased

blood flow in the testes
Poor technical parameters
Pediatric population (small testicular volume)
Hydrocele/hematoma
Vasculitis, such as polyarteritis nodosa
and lupus
Marked scrotal edema, resulting in poor
penetration of US beam
Epididymo-orchitis, resulting in testicular
infarction (a rare complication)
Protein S and antithrombin III deficiency
Fig. 19. Appendix testis. Sonogram demonstrates

appendix of the testis (arrow). The presence of fluid
facilitates this visualization.
Fig. 20. Testicular appendigeal torsion. A 10-year-old

man presented with testicular pain. Color flow Doppler demonstrates a hypoechoic mass with peripheral
hyperemia (arrow) separate from epididymis (E). This
hypoechoic mass resolved on follow-up. T, Testis.
sion of testicular appendages is to exclude testicular

torsion and acute epididymo-orchitis.
Summary
As US is considered as the first step in evaluation of
acute painful scrotum, it is important to optimize
the scan parameters. Testicular torsion is a urologic
emergency that can be accurately diagnosed by highfrequency transducer sonography in conjunction
with appropriate clinical history and physical examination. Absence of color flow Doppler does not
always signify testicular torsion as this could be
secondary to vasculitis. Similarly, the presence of
color flow does not exclude partial testicular torsion.
Acknowledgments
We would like to acknowledge our sincere thanks
to Bonnie Hami, MA, for her editorial assistance in
the preparation of this manuscript.
References
[1] Pham SB, Hong MK, Teague JA, et al. Is the testis
intraperitoneal? Pediatr Surg Int 2005;21(4):
2319.
[2] Dogra V, Bhatt S. Acute painful scrotum. Radiol
Clin North Am 2004;42:34963.
[3] Washowich TL. Synchronous bilateral testicular
torsion in an adult. J Ultrasound Med 2001;20:
9335.
[4] Dogra VS, Rubens DJ, Gottlieb RH, et al. Torsion
and beyond: new twists in spectral Doppler
evaluation of the scrotum. J Ultrasound Med
2004;23:107785.
[5] Noske HD, Kraus SW, Altinkilic BM, et al.
Historical milestones regarding torsion of the
scrotal organs. J Urol 1998;159:136.
[6] Williamson RC. Torsion of the testis and allied
conditions. Br J Surg 1976;63:46576.
[7] Haynes BE, Bessen HA, Haynes VE. The diagnosis
of testicular torsion. JAMA 1983;249:25227.
[8] Nelson CP, Williams JF, Bloom DA. The cremasteric reflex: a useful but imperfect sign in
testicular torsion. J Pediatr Surg 2003;38:12489.
[9] Rabinowitz R. The importance of the cremasteric
reflex in acute scrotal swelling in children. J Urol
1984;132:8990.
[10] Angell JC. Torsion of the testicle. A plea for
diagnosis. Lancet 1963;1:1921.
[11] Ciftci AO, Senocak ME, Tanyel FC, et al. Clinical
predictors for differential diagnosis of acute
scrotum. Eur J Pediatr Surg 2004;14:3338.
[12] Backhouse KM. Embryology of testicular descent
and maldescent. Urol Clin North Am 1982;9:
31525.
[13] Hawtrey CE. Assessment of acute scrotal symptoms and findings. A clinicians dilemma. Urol
Clin North Am 1998;25:71523 [x.].
[14] Brown SM, Casillas VJ, Montalvo BM, et al.
Intrauterine spermatic cord torsion in the newborn: sonographic and pathologic correlation.
Radiology 1990;177:7557.
[15] Barth RA, Shortliffe LD. Normal pediatric testis:
comparison of power Doppler and color Doppler US in the detection of blood flow. Radiology
1997;204:38993.
[16] Burks DD, Markey BJ, Burkhard TK, et al. Suspected testicular torsion and ischemia: evaluation with color Doppler sonography. Radiology
1990;175:81521.
[17] Favorito LA, Cavalcante AG, Costa WS. Anatomic
aspects of epididymis and tunica vaginalis in patients with testicular torsion. Int Braz J Urol 2004;
30:4204.
[18] Dogra V. Bell-clapper deformity. AJR Am J Roentgenol 2003;180:11767.
[19] Dogra VS, Sessions A, Mevorach RA, et al. Reversal of diastolic plateau in partial testicular torsion. J Clin Ultrasound 2001;29:1058.
[20] Patriquin HB, Yazbeck S, Trinh B, et al. Testicular torsion in infants and children: diagnosis with
Doppler sonography. Radiology 1993;188:7815.
[21] Horstman WG. Scrotal imaging. Urol Clin North
Am 1997;24:65371.
[22] Berman JM, Beidle TR, Kunberger LE, et al. Sonographic evaluation of acute intrascrotal pathology.
[23] Baud C, Veyrac C, Couture A, et al. Spiral twist of
the spermatic cord: a reliable sign of testicular
torsion. Pediatr Radiol 1998;28:9504.
[24] Middleton WD, Middleton MA, Dierks M, et al.
Sonographic prediction of viability in testicular
torsion: preliminary observations. J Ultrasound
Med 1997;16:237 [quiz 2930].
[25] Siegel MJ. The acute scrotum. Radiol Clin North
Am 1997;35:95976.
[26] Hamper UM, DeJong MR, Caskey CI, et al. Power
Doppler imaging: clinical experience and correlation with color Doppler US and other imaging
modalities. Radiographics 1997;17:499513.
[27] Scoutt LM, Zawin ML, Taylor KJ. Doppler US.
Part II. Clinical applications. Radiology 1990;
174:30919.
65
66
Dogra et al
[28] Middleton WD, Melson GL. Testicular ischemia:

color Doppler sonographic findings in five patients. AJR Am J Roentgenol 1989;152:12379.
[29] Paltiel HJ, Connolly LP, Atala A, et al. Acute
scrotal symptoms in boys with an indeterminate
clinical presentation: comparison of color Doppler sonography and scintigraphy. Radiology
1998;207:22331.
[30] Wilbert DM, Schaerfe CW, Stern WD, et al.
Evaluation of the acute scrotum by color-coded
Doppler ultrasonography. J Urol 1993;149:
14757.
[31] Lerner RM, Mevorach RA, Hulbert WC, et al.
Color Doppler US in the evaluation of acute
scrotal disease. Radiology 1990;176:3558.
[32] Horstman WG, Middleton WD, Melson GL, et al.
Color Doppler US of the scrotum. Radiographics
1991;11:94157 [discussion 958].
[33] Middleton WD, Siegel BA, Melson GL, et al.
Acute scrotal disorders: prospective comparison
of color Doppler US and testicular scintigraphy.
Radiology 1990;177:17781.
[34] Luker GD, Siegel MJ. Color Doppler sonography
of the scrotum in children. AJR Am J Roentgenol
1994;163:64955.
[35] Prando D. Torsion of the spermatic cord: sonographic diagnosis. Ultrasound Q 2002;18:4157.
[36] Luker GD, Siegel MJ. Scrotal US in pediatric
patients: comparison of power and standard
color Doppler US. Radiology 1996;198:3815.
[37] Adivarekar PK, Bhagwat SS, Raghavan V, et al.
Effect of Lomodex-MgSO(4) in the prevention of
reperfusion injury following unilateral testicular
torsion: an experimental study in rats. Pediatr
Surg Int 2005;21:18490.
[38] Donohue RE, Utley WL. Torsion of spermatic
cord. Urology 1978;11:336.
[39] Fitzgerald SW, Erickson S, DeWire DM, et al.
Color Doppler sonography in the evaluation of
the adult acute scrotum. J Ultrasound Med 1992;
11:5438.
[40] Sanelli PC, Burke BJ, Lee L. Color and spectral
doppler sonography of partial torsion of the
spermatic cord. AJR Am J Roentgenol 1999;172:
4951.
[41] Stillwell TJ, Kramer SA. Intermittent testicular
torsion. Pediatrics 1986;77:90811.
[42] Lopez Aramburu MA, Arroyo Munoz JL, Amon

Sesmero JH, et al. [Intermittent testicular torsion.] Arch Esp Urol 1996;49:14.
[43] Schulsinger D, Glassberg K, Strashun A. Intermittent torsion: association with horizontal lie of
the testicle. J Urol 1991;145:10535.
[44] Schneck F, Bellinger MF. Abnormalities of the
testes and scrotum and their surgical management. 8th edition. Philadelphia: Saunders; 2002.
[45] Kamaledeen S, Surana R. Intermittent testicular
pain: fix the testes. BJU Int 2003;91:4068.
[46] Ameur A, Zarzur J, Albouzidi A, et al. [Testicular
infarction without torsion in cryptorchism.] Prog
Urol 2003;13:3213.
[47] Suty JM, Hubert J, Duquenne M, et al. [Bilateral
testicular ischemia in vasculitis. Differential
diagnosis with torsion and the value of color
Doppler ultrasonography.] Prog Urol 1995;5:
5869.
[48] Leibovici D, Strauss S, Sharon A. [Acute, painful,
swollen testis in polyarteritis nodosa: a diagnostic
problem]. Harefuah 1999;136:9389.
[49] Jacob M, Barteczko K. Contribution to the origin
and development of the appendices of the testis
and epididymis in humans. Anat Embryol (Berl)
2005;209(4):287302.
[50] Fukuhara Y, Shiga Y, Omori Y, et al. [Idiopathic
testicular infarction: a case report.] Hinyokika
Kiyo 2005;51:12931.
[51] Sellars ME, Sidhu PS. Ultrasound appearances of
the testicular appendages: pictorial review. Eur
Radiol 2003;13:12735.
[52] Skoglund RW, McRoberts JW, Ragde H. Torsion
of testicular appendages: presentation of 43 new
cases and a collective review. J Urol 1970;104:
598600.
[53] Cohen HL, Shapiro MA, Haller JO, et al. Torsion
of the testicular appendage. Sonographic diagnosis. J Ultrasound Med 1992;11:813.
[54] Yang DM, Lim JW, Kim JE, et al. Torsed appendix
testis: gray scale and color Doppler sonographic
findings compared with normal appendix testis.
J Ultrasound Med 2005;24:8791.
[55] Baldisserotto M, de Souza JC, Pertence AP, et al.
Color Doppler sonography of normal and torsed
testicular appendages in children. AJR Am J
Roentgenol 2005;184:128792.
67
ULTRASOUND
CLINICS
Sonography of Pediatric Urinary

Tract Emergencies
Carlos J. Sivit,
&
&
&
&
MD
Ureteropelvic junction obstruction

Renal vein thrombosis
Urolithiasis
Infected urachal remnant
Urinary tract emergencies frequently result in

acute signs and symptoms in the pediatric age
group. The imaging evaluation of these conditions
plays an important role in their diagnosis and
management. The most compelling indication for
emergent imaging in children who have suspected
urinary tract emergencies is that nonemergent treatment may lead to irreversible loss of renal function.
Sonography is typically the initial examination of
choice in infants and children who have suspected
emergencies of the urinary tract. Sonography provides a rapid initial screening method to assess
the urinary tract without any pre-exam preparation
and without ionizing radiation. This article focuses
on the sonographic assessment of common urinary
tract emergencies in children, including the typical
clinical presentation and characteristic sonographic
findings of such conditions.
Ureteropelvic junction obstruction

Ureteropelvic junction (UPJ) obstruction is the
most common cause of urinary tract obstruction
in the pediatric age group. There is controversy regarding the underlying cause. Multiple underlying
factors have been proposed, including: (1) abnormal development of the proximal ureteral smooth
muscle; (2) aberrant vessels, or bands crossing the
upper ureter and renal pelvis; (3) delayed recana-
&
&
&
&
Hemorrhagic cystitis
Acute renal failure
Summary
References
lization of the fetal ureter; or (4) abnormal ureteral peristalsis.

The widespread use of prenatal sonography has
impacted how the diagnosis of UPJ obstruction is
made. The condition is now often diagnosed in
utero. However, it is still not unusual for the diagnosis to be made after birth. Infants and children who have a UPJ obstruction usually present
with a palpable abdominal mass representing the
enlarged kidney; flank or abdominal pain; hematuria; or symptoms related to urinary tract infection.
Sonography is typically the initial imaging examination of choice for the evaluation of suspected
UPJ obstruction. The sonographic findings associated with a UPJ obstruction include dilatation of
the intrarenal collecting system, including renal
pelvis and calyces [Fig. 1] [1]. When the degree of
dilatation is mild to moderate, it is typically easier
to identify the fluid-filled areas within the kidney as
the dilated collecting system. However, when the
dilatation is severe, multiple cystic lesions of uniform size are noted, representing the dilated calyces
[Fig. 2]. In these instances it is important to
demonstrate communication between the cysts to
identify the abnormality as dilated calices and differentiate the condition from a multicystic dysplastic kidney (MCDK) or renal dysplasia. The cysts in
MCDK or renal dysplasia vary in size and will not
communicate [Fig. 3]. Additionally, there is no
Department of Radiology, Rainbow Babies and Children's Hospital, 11100 Euclid Avenue, Cleveland, OH 441065056, USA
E-mail address: Sivit@uhrad.com
doi:10.1016/j.cult.2005.09.001
68
Sivit
Fig. 1. UPJ obstruction. Longitudinal sonogram through

the right kidney shows dilatation of the intrarenal collecting system (arrow). The right ureter was not dilated.
The findings are consistent with a UPJ obstruction.
recognizable renal parenchyma in an MCDK. The

ipsilateral ureter is typically not dilated in children
who have a UPJ obstruction, although ureteral
vesical junction (UVJ) may occasionally coexist
and result in a dilated ureter [2]. The obstruction
is bilateral in approximately one fourth of children.
Small amounts of fluid in the intrarenal collecting system resulting in a renal pelvis anteroposterior diameter of less than 10 mm in size may
occasionally be seen in infants and young children
in the absence of urinary tract obstruction [Fig. 4].
A renal pelvis diameter of less than 10 mm is
characterized as mild pyelectasis and generally has
no pathologic significance, particularly if the examination occurs after 48 hours of birth. The kidney
may occasionally appear normal at sonography in
neonates who have a UPJ obstruction, particularly
if the examination is performed in the first day of
life when urine production is low. Therefore, if the
diagnosis is suspected on prenatal sonography, it
is best to delay the postnatal sonogram until the
infant is at least 48 to 72 hours old.
There is an increased prevalence of UPJ obstruction in children who have a horseshoe kidney.
This condition is felt to be caused by the atypical
location of the proximal ureters in horseshoe kidney. They are positioned more anteriorly, resulting
in an increased risk for obstruction by crossing
blood vessels. There may also be coexistent urolithiasis associated with UPJ obstruction caused by
the associated urinary stasis [3]. A urinoma may
occasionally be noted because of urine leakage
secondary to a high-grade obstruction. Renal parenchymal thinning may also be noted because of
secondary renal scarring. This scarring usually
results following the development of acute pyelonephritis, which is prone to develop secondary to
urine stasis. UPJ obstruction may also be associated
with vesicoureteral reflux. Therefore, the imaging
evaluation of children who have a UPJ obstruction
should include a voiding cystourethrogram to evaluate for potential reflux. The diagnosis of UPJ
obstruction is confirmed with diuretic renal scintigraphy, which will show delayed renal isotope
excretion. Renal scintigraphy allows for assessment
of severity of obstruction and quantification of
residual renal function.
Renal vein thrombosis

Renal vein thrombosis most commonly presents
in the neonatal period. However, it may be seen
at any age. The clinical presentation includes a
palpable mass representing enlarged kidney, renal
insufficiency, hematuria, or hypertension. The
pathophysiology of renal vein thromobosis differs
between infants and older children. In infants the
condition is typically associated with hemoconcentration attributable to dehydration, sepsis, and
maternal diabetes mellitus. In this age group the
thrombus formation is felt to start in the arcuate
and interlobar venules and progress centrally to the
Fig. 2. Severe UPJ obstruction. Longitudinal sonogram through the right renal pelvis (A) and calyces (B) show
marked dilatation of the intrarenal collecting system. Note that on the image through the calyces, the dilated
calyces appear as large cysts. However, they are symmetrical in size and were noted to communicate by sonography. Also note the diminished renal parenchyma.
Pediatric Sonography
Fig. 3. Multicystic dysplastic kidney. Longitudinal (A) and transverse (B) views through the right kidney demonstrate multiple parenchymal cysts of varying size (arrowheads). They were not noted to communicate on real-time
sonography. Additionally, there is no normal renal parenchyma visualized.
renal hilum. Renal vein thrombosis may also be

seen in the neonatal age group in association with
indwelling umbilical venous catheters. These neonates typically develop thrombi in the inferior vena
cava with secondary extension into the renal vein.
In older children renal vein thrombosis is associated with renal tumor, glomerulonephritis, or
nephrotic syndrome and is felt to originate in the
main renal vein or inferior vena cava.
Doppler sonography is the examination of choice
for the evaluation of suspected renal vein thrombosis. The sonographic findings associated with
renal vein thrombosis vary depending on the extent
and duration of renal venous occlusion. Sonographic findings include the presence of filling
defects in the main renal vein or inferior vena
cava and absence or diminution of renal venous
flow surrounding the thrombus [4]. The acute
thrombus is usually echogenic. Occasionally, the
thrombus may be hypoechoic and difficult to recognize at sonography. The venous outflow obstruction caused by renal vein thrombosis results
in diminution of ipsilateral renal arterial flow. This
diminution in turn results in narrowing of the
systolic peak and reduction or reversal of diastolic
flow with an elevated resistive index in the ipsilat-
eral renal artery [Fig. 5] [5]. Enlargement of the

involved kidney with diffuse increase in parenchymal echogenicity or loss of corticomedullary differentiation is also typically seen in the acute period
[Fig. 6].
The treatment of renal vein thrombosis is usually
nonsurgical. It consists primarily of fluid and electrolyte management. Sonography is useful in the
serial evaluation of children who have renal vein
thrombosis to monitor progression of disease. The
outcome associated with renal vein thrombosis is
variable and dependent on extent and rapidity of
venous occlusion. If renal vein occlusion by thrombus is incomplete, or if the thrombus develops
slowly allowing time for development of collateral
venous drainage, then there is typically diminished
renal edema and increased renal arterial perfusion,
resulting in a more favorable outcome. If the renal
vein obstruction is complete, then renal ischemia
develops followed by parenchymal atrophy. In
severe cases, serial sonographic assessment 2 to
3 weeks following the initial event shows parenchymal echogenicity in the involved kidney gradually
becoming heterogenous and renal size diminishing.
Urolithiasis
Fig. 4. Pyelectasis and caliectasis. Longitudinal sonogram through the right kidney in a 2-week-old infant
shows a small amount of fluid (arrowhead) within the
intrarenal collecting system without dilatation.
Urolithiasis represents the formation of macroscopic calcification within the urinary collecting
system. It is not a disease but a complication of
many different diseases. It is a common condition,
with estimates that 12% of individuals will develop
a calculi sometime during their lives [6]. The calculi
are formed on the renal papillae by retention of
lithogenic particles either by obstruction or adherence to damaged renal epithelium. Urolithiasis is
a complication of many different disorders that
result in the supersaturation of urine by calculipromoting factors such as increased calcium or
oxalate excretion. It is also associated with conditions that result in urine stasis. The most common
69
70
Sivit
Fig. 5. Acute renal vein thrombosis. Doppler evaluation of the left (A) and right (B) main renal arteries in a child
who has left-sided renal vein thrombosis shows reversal of diastolic arterial flow in the left renal artery. Note the
normal diastolic flow in the right renal artery.
causes of urolithiasis in children are urinary tract

obstruction, infection, and bladder augmentation
in children who have a neurogenic bladder [7].
Most urinary tract calculi are composed of calcium
mixed with oxalate, phosphate, or a combination
of the two. Calculi may also be composed of struvite, cystine, and uric acid. A special type of calculus is the staghorn, which resembles a stags
antlers and is composed of calcium, magnesium,
ammonium, and phosphate. These calculi are typically large and have a laminated appearance.
Children who have urolithiasis commonly present with flank pain or hematuria. The flank
pain can be severe. It is referred to as renal colic
and is caused by associated ureteral obstruction.
Generalized symptoms, including fever and chills,
may also be noted if secondary urinary tract infection develops.
Sonography is commonly used in the initial assessment of suspected urolithiasis. Sonography is
highly accurate for the diagnosis of calculi within
the intrarenal collecting system and the bladder
[8,9]. However, it is insensitive in the assessment

of ureteral calculi [8,10]. The deep central location
of the ureters within the retroperitoneum makes
sonographic assessment difficult. Additionally, calculi in the ureters are often obscured by bowel gas.
Calculi typically appear as a highly echogenic
focus [Fig. 7]. Associated acoustic shadowing is
noted at sonography with calculi larger than
5 mm [Fig. 7]. However, smaller calculi may not
shadow. Dilatation of the urinary collecting system
proximal to the calculus may be noted if there is
associated urinary tract obstruction [Fig. 8]. The
identification of a ureteral jet at Doppler sonography is useful for excluding ureteral obstruction
secondary to urolithiasis [11]. The ability to detect
ureteral jets is improved when the child is well
hydrated. Sonography is also useful in the followup assessment of children who have calculi to
monitor the response to therapy.
Urolithiasis should be differentiated from nephrocalcinosis, which represents deposition of calcium
within the renal parenchyma and is classified
Fig. 6. Acute renal vein thrombosis. Longitudinal views through the left (A) and right (B) kidneys show asymmetric
enlargement of the involved left kidney with poor corticomedullary differentiation.
Fig. 7. Urolithiasis. Longitudinal (A) and transverse (B) views through the upper pole of the right kidney
demonstrate a renal calculi with acoustical shadowing (arrowhead).
according to the anatomic area involved [6,12]. Medullary nephrocalcinosis, the most common form, is
associated with various conditions, including distal
renal tubular acidosis, hypercalcemia, hypercalciuria, medullary sponge kidney, and renal papillary
necrosis [Fig. 9]. Cortical nephrocalcinosis is uncommon and may be secondary to acute cortical
necrosis, oxalosis, or chronic glomerulonephritis.
Unenhanced CT has been shown to have the
highest sensitivity in the diagnosis of urolithiasis.
In children who have acute flank pain in whom
the diagnosis is uncertain, CT can accurately determine the presence of renal, ureteral, or bladder calculi [1315]. The presence of renal collecting
system obstruction can also be accurately detected.
Additionally, CT also allows for the assessment of
extraurinary causes of acute flank pain.
Infected urachal remnant

The urachus is a remnant of an embryonic connection coursing in the midline between the umbilicus
and the dome of the bladder. It lies in the space of
Retzius behind the transversalis fascia and anterior
to the peritoneum. The urachal lumen is typically

obliterated during fetal life or early in the neonatal period and the remnant is reduced to a fibrous
cord known as the median umbilical ligament. Occasionally, there is failure of all or a portion of
the urachal lumen to be obliterated. Such abnormalities are characterized as urachal remnant abnormalities and include patent urachus, urachal
sinus, urachal diverticulum, and urachal cyst
[16,17]. Urachal remnants are usually asymptomatic. However, occasionally they may become
infected secondary to urine stasis, resulting in enlargement of the structure with abscess formation. The
patients typically present with acute abdominal
pain, fever, and leukocytosis.
At sonography, an infected urachal remnant appears as a complex midline mass between the
umbilicus and the dome of the bladder [Fig. 10]
[1618]. A tubular component can usually be identified, which along with the characteristic midline
location allows for differentiation of infected urachal remnants from other causes of intra-abdominal abscesses . The infected urachal remnant may
also demonstrate mass effect on the dome of the
bladder [Fig. 10].
Fig. 8. Urolithiasis with obstruction. (A) Longitudinal view through the mid-right ureter demonstrates a linear
calculus (arrow) with acoustical shadowing (arrowheads). Note the dilatated ureter proximal to the calculus
(asterisk). (B) Transverse view through the right renal pelvis shows dilatation.
71
72
Sivit
Fig. 9. Medullary nephrocalcinosis. Longitudinal (A) and transverse (B) views through the left kidney demonstrate
increased medullary echogenicity (arrowheads) indicative of medullary nephrocalcinosis.
CT may provide improved definition of the extent of abscess and enhanced delineation of the
anatomic relationship of the abscess to the bladder.
The treatment of an infected urachal remnant typically involves surgical drainage. These infections
usually do not respond to antibiotic therapy because of the diminished blood supply to the involved area.
Hemorrhagic cystitis
Hemorrhagic cystitis results from damage to the
bladder transitional epithelium, resulting in diffuse
bleeding. The cause may be multifactorial. The condition has been associated with drugs, infection,
and toxins. Chemotherapeutic agents, particularly
cyclophosphamide and busulfan, are the drugs
most commonly associated with the condition. It
has most frequently been reported following bone
marrow transplantation, particularly in younger
children [19]. It has been reported as early as
1 week and as late as 4 months following transplantation. Children who have hemorrhagic cystitis
typically present with gross hematuria, dysuria,
suprapubic pain, and urinary frequency. The extent

of hemorrhage may be severe and the condition
may be life-threatening. It can result in impaired
renal and bladder function. Treatment usually involves bladder irrigation, blood transfusions, and
maintenance of adequate hydration. Endoscopic
removal of blood clots is sometimes required, as
the clots may result in bladder outlet obstruction.
Sonography is usually the initial modality in the
evaluation of suspected hemorrhagic cystitis. Associated sonographic findings include focal [Fig. 11],
multifocal [Fig. 12], or circumferential bladder wall
thickening [Fig. 13] [1921]. The normal bladder
wall thickness should not exceed 3 mm for a distended bladder and 5 mm for a nondistended
bladder [22]. Bladder wall thickening associated
with hemorrhagic cystitis is most pronounced in
the hypoechoic muscularis layer. The echogenic
mucosal and submucosal layers are less involved.
The bladder wall thickening may be polypoid in
appearance and resemble a bladder wall mass.
One sonographic finding that is helpful in distinguishing hemorrhagic cystitis from a true bladder
wall mass is that bladder thickening with hemor-
Fig. 10. Infected urachal remnant. (A) Longitudinal midline sonogram through the bladder shows a complex
midline mass with mass effect on the superior aspect of the bladder (arrowheads). (B) Transverse midline
sonogram immediately above the dome of the bladder in same child demonstrates a rounded, solid, complex
mass (arrowheads). At surgery, an infected urachal remnant was noted. B, bladder.
Fig. 11. Hemorrhagic cystitis with focal bladder wall

thickening. Transverse sonogram through the bladder
in a child who has hemorrhagic cystitis shows focal
bladder wall thickening resembling a mass (arrowheads). B, bladder.
Fig. 13. Hemorrhagic cystitis with diffuse bladder wall

thickening. Transverse sonogram through the bladder
in a child who has hemorrhagic cystitis shows diffuse
bladder wall thickening (arrowhead). B, bladder.
lows for surveillance of bladder clots that can lead

to obstruction.
Acute renal failure
rhagic cystitis usually changes in contour and thickness with bladder wall filling, whereas a true bladder wall mass does not change. Echogenic clots
may be noted within the bladder lumen. The bladder may also have a reduced capacity. In severe
cases there may be complete contraction of the
bladder without a visualized lumen. There may be
associated hydronephrosis secondary to obstruction from bladder clots. In addition, the entire
urothelial surface is at risk and lesions of the
renal pelvis and ureter have been reported. Suburothelial thickening of the renal pelvis may be
noted at sonography.
Sonography is a useful modality for the serial
evaluation of children who have hemorrhagic cystitis in monitoring response to therapy. Serial followup by sonography typically demonstrates gradual
improvement in bladder wall thickening associated with clinical improvement [23]. It also al-
Acute renal failure (ARF) is defined as the sudden

loss of renal function with or without oliguria. It
typically results from an ischemichypoxic or toxic
event. ARF may result from prerenal, intrinsic, or
postrenal causes [24]. Prerenal causes result from
renal hypoperfusion usually secondary to volume
depletion. The kidneys are intrinsically normal in
prerenal failure and usually improve with restoration of renal perfusion. Intrinsic ARF results from
acute tubular necrosis secondary to prolongued
prerenal ARF or secondary to a primary renal insult.
Postrenal ARF develops secondary to acute bilateral obstructive uropathy. It is caused by an acute
elevation of intrarenal pressure resulting in reduction of renal blood flow.
Sonography is the examination of choice for the
initial evaluation of infants and children who have
ARF. The primary role of sonography is to detect
Fig. 12. Hemorrhagic cystitis with multifocal bladder

wall thickening. Longitudinal sonogram through the
bladder in a child who has hemorrhagic cystitis shows
multifocal areas of bladder wall thickening (arrowheads). B, bladder.
Fig. 14. Acute renal failure. Longitudinal sonogram

through the right kidney in a child who has ARF
demonstrates increased renal parenchymal echogenicity. Note that the renal parenchyma is hyperechoic
relative to the liver. L, liver.
73
74
Sivit
Fig. 15. Reversible acute renal failure. (A) Longitudinal sonogram through the right kidney in a child who has ARF
shows increased renal parenchymal echogenicity. (B) Follow-up sonogram in the same child 3 months after
restoration of normal renal function shows normal renal parenchymal echotexture.
obstructive uropathy resulting in postrenal ARF or

intrinsic causes such as renal vein thrombosis.
Renal sonography in children who have ARF typically demonstrates increased renal parenchymal
echogenicity [Fig. 14] [25]. The normal renal parenchymal echogenicity in children over the age of
4 months is usually hypoechoic relative to that of
liver or spleen. In most children who have ARF, the
renal parenchymal echogenicity will be equal to or
greater than liver or spleen. Correlation has been
shown between increased renal parenchymal echogenicity and elevated blood urea nitrogen and
serum creatinine values. Therefore, the sonographic
finding of increased parenchymal echogenicity is
significant in that it indicates the presence of renal
disease that is usually associated with decreased
renal function. However, increased renal parenchymal echogenicity is a nonspecific finding that may
be seen with various diffuse parenchymal renal
disorders that involve glomerular, tubular, interstitial, or vascular abnormalities [25,26]. This finding
may also be noted with renal involvement in leukemia or lymphoma. The entity is generally described as medical renal disease. Furthermore, in
children who have ARF, sonography cannot discriminate between reversible and irreversible renal
insufficiency. Improvement in renal parenchymal
echogenicity is often seen with improvement in
renal function [Fig. 15].
Summary
Sonography plays an important role in the evaluation of pediatric urinary tract emergencies. Sonography is often the initial imaging examination
of choice for evaluation of such conditions. The
precise characterization of urinary tract pathology
by sonography allows for prompt diagnosis and
treatment. Therefore, it is important for the practicing radiologist to have an understanding of
the typical sonographic findings associated with

such conditions.
References
[1] Brown T, Mandell J, Lebowitz RL. Neonatal hydronephrosis in the era of sonography. AJR Am J
Roentgenol 1987;148:95963.
[2] McGrath MA, Estroff J, Lebowitz RL. The coexistence of obstruction at the UPJ and UVJ. AJR
[3] Kraus SJ, Lebowitz RI, Royal SA. Renal calculi in
children. Pediatr Radiol 1999;28:62430.
[4] Helenon O, Rody FE, Correas J-M, et al. Color
Doppler US of renal vascular disease in native
kidneys. Radiographics 1995;15:83354.
[5] Laplante S, Patriquin HB, Robitaille P, et al. Renal vein thrombosis in children: evidence of early
flow recovery with Doppler US. Radiology 1993;
189:3742.
[6] Dyer RB, Chen MYM, Zagoria RJ. Abnormal calcifications in the urinary tract. Radiographics
1998;18:140524.
[7] Santos-Victoriano M, Brouhard BH, Cunningham III RJ. Renal stone disease in children. Clin
Pediatr (Phila) 1998;37:58399.
[8] Haddad M, Sharif HS, Shahed MS, et al. Renal
colic: Diagnosis and outcome. Radiology 1992;
184:838.
[9] Erwin BC, Carroll BA, Sommer FG. Renal colic:
the role of ultrasound in initial evaluation. Radiology 1984;152:14750.
[10] Bonner MP, Pollack HM. Urolithiasis in the
lower urinary tract. Semin Roentgenol 1982;17:
1408.
[11] Burge HJ, Middleton WD, McClennan RI, et al.
Ureteral jets in healthy subjects and in patients
with unilateral ureteral calculi: comparison with
color Doppler US. Radiology 1991;180:43742.
[12] Hernanz-Schulman M. Hyperechoic renal medullary pyramids in infants and children. Radiology 1991;181:911.
[13] Tamm EP, Silverman PM, Shuman WP. Evalua-
[14]
[15]
[16]
[17]
[18]
[19]
[20]
tion of the patient with flank pain and possible

ureteral calculus. Radiology 2003;228:31929.
Smith RC, Verga M, McCarthy S, et al. Diagnosis
of acute flank pain: value of unenhanced helical
CT. AJR Am J Roentgenol 1966;166:97101.
Sommer FG, Jeffrey RB, Rubin GD, et al. Detection of ureteral calculi in patients with
suspected renal colic: value of reformatted noncontrast helical CT. AJR Am J Roentgenol 1995;
165:50913.
Khati NJ, Enquist EG, Javitt MC. Imaging of the
umbilicus and periumbilical region. Radiographics 1998;18:41331.
DiSantis DJ, Siegel MJ, Katz ME. Simplified approach to umbilical remnant abnormalities. Radiographics 1991;11:5966.
Cacciorelli AA, Kass EJ, Yang SS. Urachal remnants: Sonographic demonstration in children.
Radiology 1990;174:4735.
Benya EC, Sivit CJ, Quinones RR. Abdominal
complications after bone marrow transplantation in children: sonographic and CT findings.
Yang CC, Hurd DD, Case LD, et al. Hemorrhagic
[21]
[22]
[23]
[24]
[25]
[26]
cystitis in bone marrow transplantation. Urology

1994;44:3228.
McCarville MB, Hoffer FA, Gingrich JR, et al.
Imaging findings of hemorrhagic cystitis in pediatric oncology patients. Pediatr Radiol 2000;30:
1318.
Jequier S, Rousseau O. Sonographic measurements of the normal bladder wall in children.
Cartoni C, Arcese W, Avvisati G, et al. Role of
ultrasonography in the diagnosis and followup
of hemorrhagic cystitis after bone marrow transplant. Bone Marrow Transplant 1993;12:4637.
Andreoli SP. Management of acute renal failure.
In: Barratt TM, Avni ED, Harmon DE, editors. Pediatric nephrology. 4th edition. Baltimore (MD):
Lippincott Williams & Wilkins; 1999. p. 11934.
Kraus RA. Increased renal parenchymal echogenicity: cases in pediatric patients. Radiographics
1990;10:100918.
Platt JF, Rubin JM, Bowerman RA, et al. The
inability to detect kidney disease on the basis of
echogenicity. AJR Am J Roentgenol 1988;151:
3179.
75
77
ULTRASOUND
CLINICS
Preface
Vascular Ultrasound
Deborah J. Rubens, MD
Department of Imaging Sciences
University of Rochester Medical Center
601 Elmwood Avenue
Rochester, NY 14542-8648, USA
E-mail address:
deborah_rubens@urmc.rochester.edu
Deborah J: Rubens, MD
Guest Editor
Rapid advancements in cross-sectional imaging

systems over the past two decades have provided
a wide array of diagnostic tools for the practicing
radiologist. Advances in computer technology with
faster processing and miniaturization have permitted tremendous anatomic imaging advances in
CT, MR imaging, and ultrasound.
One of the unique capabilities of ultrasound is its
elegant depiction of blood flow, both the real-time
anatomic display, as well as the physiologic information of blood flow velocity and flow direction.
In this inaugural issue of the Ultrasound Clinics, we
felt it was important to dedicate part of the issue
to this significant ultrasound feature. As such, we
have selected topics that emphasize the foundation
of vascular ultrasound, the technical aspects of
Doppler imaging, and the basic principles of carotid artery imaging. In addition, subjects include
advances in vascular Doppler imaging, including
imaging of nonvascular structures, advanced waveform analysis, Doppler applications in gynecologic
imaging, and new Doppler frontiers such as trans-
cranial Doppler imaging and Doppler-guided interventional arterial therapy. The hope is to familiarize readers with the essentials of Doppler
imaging in the modern ultrasound laboratory and
to excite them with the future possibilities of this
ever-changing and advancing technology.
My thanks to Dr. Vikram Dogra, editorial board
member for the Ultrasound Clinics, for his commitment to ultrasound and ultrasound education.
Thank you to Barton Dudlick for agreeing that
ultrasound deserved its own Clinics series and for
his editorial assistance, and thanks also to the
staff at Elsevier for their help. Thank you especially to Margaret Kowaluk and Holly Stiner for
graphics and manuscript preparation, respectively.
Lastly, and most importantly, thanks to all of
my collaborators and coauthors who contributed
such outstanding articles. Each and every one of
us is enthusiastic about the practice of vascular
ultrasound and the its exciting future. We hope
you will find our efforts valuable and stimulating
as well.
doi:10.1016/j.cult.2005.10.002
79
ULTRASOUND
CLINICS
Doppler Artifacts and Pitfalls

Deborah J. Rubens, MD*, Shweta Bhatt, MD,
Shannon Nedelka, MD, Jeanne Cullinan, MD
&
&
&
Understanding the technical challenge of

Doppler ultrasound or setting up your
equipment to get the best images and
spectral tracings
Choosing the correct transducer frequency
Doppler angle
Sample volume
Wall filters
Doppler Gain
Velocity scale
Doppler artifacts
Aliasing
Blooming artifact
Thirty years ago, use of Doppler ultrasound (US)

was limited to the vascular laboratory and was
mainly used to interrogate the carotid arteries.
Today, Doppler US has pervaded all of diagnostic
US imaging, is the mainstay of venous diagnosis,
and is used extensively throughout abdominal, pelvic, and obstetric imaging. In addition to continuous wave Doppler, pulsed wave (duplex) Doppler,
color Doppler, and power Doppler are now available. Motion is imaged in high-velocity settings
(ie, the aorta, carotid, and renal arteries) and in
low flow states (portal vein thrombosis, calf veins,
and so forth). Superficial structures (neck and arm
vessels, testicular and ovarian vessels) and deep structures (hepatic and renal arteries) are imaged. To
accomplish this range of diversity takes more than
one knob on a machine. The image obtained, the
particular organ being viewed, and the capabilities
of the machine itself are governed by the intrinsic
properties of US and Doppler. The physical properties of US give rise to several artifactssome occur
&
&
&
Directional ambiguity
Partial volume artifact
Pseudoflow
Flash artifact
Mirror-image artifact
Edge artifact
Twinkling artifact
Day-to day Doppler: too much flow versus
too little flow
Summary
References
in gray-scale and Doppler imaging and others are

specific to Doppler, especially color or power Doppler. Knowing an artifacts typical location and appearance helps avoid misinterpretation and can
actually be useful diagnostically [1]. Understanding
how to generate a Doppler signal enables the examiner to better avoid day-to-day scanning pitfalls,
which primarily fall into two clinical categories:
too little flow or too much flow. This article addresses the machine parameters first, then the artifacts, and concludes with the operational issues (or
pitfalls) as they apply to day-to-day scanning.
Understanding the technical challenge of

Doppler ultrasound or setting up your
equipment to get the best images and
spectral tracings
The Doppler effect measures a change in the reflected sound frequency generated by motion of
the source or the detector. The challenge lies in
Department of Radiology, University of Rochester Medical Center, 601 Elmwood Avenue, Rochester, NY
14642-8648, USA
E-mail address: Deborah_Rubens@urmc.rochester.edu (D.J. Rubens)
doi:10.1016/j.cult.2005.09.009
80
Rubens et al
the detection of the signal and the accurate display

of its direction and speed. Although the Doppler
effect is commonly used to measure flowing blood,
any tissue or fluid motion may generate a Doppler
signal. That signal is a shift or difference in frequency between the transmitted and the received
US pulse. The greatest difference or strongest signal
is achieved when the motion is parallel to the US
beam and no signal is generated when the motion
is perpendicular to it.
Choosing the correct transducer frequency

Of all the technical parameters that can be controlled, the choice of transducer frequency is paramount because the intensity of the scattered sound
varies in proportion to the fourth power of the
Doppler frequency [2]. Higher frequencies are,
therefore, much more sensitive to flow but sometimes cannot penetrate deep enough without attenuation; thus, for superficial structures such as
the testes, 7 to 10 MHz may be ideal, whereas for
deep abdominal structures, such as the hepatic
arteries or the portal vein, 3 MHz or lower may
be needed. Often the choice of Doppler transducer
frequency is empiric with a trial of different frequencies until the best compromise between penetration and signal strength is achieved [Fig. 1].
Doppler angle
Unlike in gray-scale US imaging whereby the best
image is obtained perpendicular to the US beam, in
Doppler US, the strongest signals (and best spectra)
result when the motion is parallel to the beam. A
Doppler angle of 90 does not display flow because
no component of the frequency shift is directed
Fig. 1. Pseudotesticular torsion. Four-day-old infant presents with left hydrocele and testicular torsion is suspected.
(A, B) Initial axial images of the symptomatic left (A) and asymptomatic right (B) sides at identical gain and scale
settings show symmetric spectral Doppler patterns equal above and below the baseline but do not have typical
vascular spectral Doppler waveform. This is noise. Note scanning frequency is 8.5 MHz and spectral Doppler
frequency is 5 MHz. (C, D) Axial images from repeat examination with appropriate high frequency transducer
shows normal symmetric arterial waveforms bilaterally. Note transducer frequency of 14 MHz and spectral
Doppler frequency of 7 MHz.
Fig. 2. Pseudothrombosis of main portal vein (MPV). (A) Color Doppler ultrasound (CDUS) of the MPV detects no
flow in confluence of MPV (arrows) suggesting thrombosis of MPV. Note reversed flow in splenic vein (arrowhead )
indicating portal hypertension and potentially slow-flow state in MPV. Wall filter set on medium, which may
exclude low velocity flow and MPV segment which lacks flow is parallel to transducer surface, and therefore at
90 to Doppler beam. (B) Portal venous phase of subsequent contrast enhanced CT on the same day reveals
completely patent MPV (arrow). (C ) Repeat CDUS examination with different machine following day shows
retrograde flow in MPV (arrows) and no apparent thrombus. Doppler angle has been improved (no longer 90)
and wall filter is corrected to low setting (20 Hz).
back toward the transducer [Fig. 2]. Any Doppler

angle other than zero requires angle correction to
adjust for the component of the signal not directed
parallel to the beam. The larger the Doppler angle,
the greater the correction is that needs to be done
and the greater chance for error; therefore, the
Doppler beam angle must always be kept as low
as possible. Ideally, it should be less than 60 and
always less than 70 because the errors associated
with the angle correction increase up to 20% to
30% with higher Doppler angles [Fig. 3] [3].
Sample volume
The sample volume is the three-dimensional space
from which the Doppler frequency shifts are measured. In color or power Doppler it is the color box,
and in pulsed wave Doppler it is the cursor one
places within the vessel. Although on the image the
sample looks like a flat box, it has a third dimension in and out of the plane of the image, which
may be much larger than anticipated (even 1 cm
or more in thickness, depending on frequency
and depth). Signals may be sampled and displayed
from unwanted areas of a vessel (ie, too close to the
vessel wall, giving more turbulence, and slower
velocities) or even from unwanted vessels (adjacent

arteries or veins). In a large vessel, blood flow is
not uniform across the vessel; it is generally slower
near the wall (as a result of friction and turbulence)
and faster in the center. Therefore, with spectral
Doppler, too wide a sample (which encompasses
the entire vessel lumen) includes the normal turbulence and slower velocities along the vessel margins, which result in spectral broadening (that may
be incorrectly interpreted as poststenotic turbulence) [3]. If the spectral sample is too small and
is not placed in the area of greatest flow, the resulting measured velocity is too low. If the sample
volume is too small and the vessel is mobile, a
discontinuous Doppler signal may result with loss
of the diastolic signal in each cycle. The ideal sample volume size for routine survey of a vessel is
about two thirds of the vessel width positioned in
the center of the vessel [3] excluding as much of the
unwanted clutter from near the vessel walls as
possible [2].
Wall filters
The Doppler frequency shift can be detected from
moving blood vessel walls and from the blood
81
82
Rubens et al
Fig. 3. Angle-corrected velocity. (A) Correct angle, as depicted by line through Doppler cursor, is parallel to center
of lumen and yields peak systolic velocity of 65 cm/s. (B) Angle is too low, at 38, which results in calculated velocity
of only 41 cm/s. (C ) Angle is only slightly off at 72 but velocity is now calculated to be 105 cm/s. Small changes in
angle greater than 60 result in much larger errors than small changes below 60.
itself. These wall echoes are large amplitude causing

a loud wall thump on the audio Doppler output
[4]. Fortunately, these signals are also low frequency. By using a threshold that cuts off these
low frequency noises, a cleaner high-velocity bloodflow signal is displayed; however, if the wall filter threshold is set too high, true blood flow also
is discarded from the display. Low velocity venous flow and the filter for venous Doppler should
be kept at the lowest practical level, usually 50 to
100 Hz or less [Fig. 2] [2].
Doppler Gain
This setting controls the amplitude of the color
display in color or power Doppler mode and the
spectral display in pulse Doppler mode. For spectral Doppler, the tracing should be continuous and
easy to visualize, without any low-level noise band
above and below the baseline. Excess spectral gain
in pulse wave Doppler produces noise that may be
mistaken for flow [Fig. 1]. In arterial spectra, excess

gain fills in the tracing as low velocity echoes and
mimics turbulent flow [5]. For color imaging, the
gain should be turned up until scattered isolated
color pixels can be seen overlying the gray-scale
background. Then the gain should be turned back
until they disappear.
If the color gain settings are too low, flow may be
present but not visualized. If the settings are too
high, color or power signals may overwrite grayscale clot. A machine setting related to gain for
color and power Doppler is the color-write priority.
The color-write priority determines whether a given
pixel is written as a gray-scale value or as color [3].
If the gray-scale signal is above some threshold
(eg, medium gray), the pixel remains gray, and if
the signal is below the threshold (ie, the pixel is dark
gray or black), the pixel is written as color. If the
gray-scale gain is too high or the color-write priority
too low, some color pixels may not be displayed.
Fig. 4. Portal vein pseudoclot. (A) Longitudinal CDUS image in cirrhotic patient with portal hypertension.
Velocity scale is set at 20 cm/s. Good flow in hepatic artery anteriorly (arrow) but none in adjacent portal
vein (arrowheads). (B) Scale is appropriately lowered to 7 cm/s and slower flow in portal vein (arrowheads) can
now be demonstrated.
83
84
Rubens et al
Fig. 5. Spectral aliasing. (A) Longitudinal spectral Doppler evaluation of common carotid artery (CCA) demonstrates
spectral aliasing as higher Doppler shift frequencies wrap around scale and peaks (arrows) are written coming
from opposite side of baseline. Note that peaks can actually cross baseline and overwrite existing spectral display.
(B) Dropping baseline (arrow) eliminates aliasing. (C ) Increasing scale (from 31 to 62 cm/s) eliminates aliasing.
Velocity scale
The velocity scale controls the range of frequencies
displayed and is critical in color and spectral Doppler imaging. If the scale is too high (similar to a
too-wide window in CT), the dynamic range is too
large and low velocity signals are missed simulating
an area of thrombosis [Fig. 4], particularly in low
flow vessels, such as the portal vein. If the velocity
scale is too low, the dynamic range is too small to
display the high-velocity signals accurately and
aliasing results (see later discussion).
Doppler artifacts
Doppler artifacts can be grouped into three broad
categories [1]: (1) artifacts caused by technical limitations, including aliasing, improper Doppler angle
with no flow, indeterminate Doppler angle, blooming, and partial volume artifact; (2) artifacts caused
by patient anatomy, including mirror image arti-
fact, flash artifact, and pseudoflow; and (3) artifacts caused by machine factors, including edge
artifact and twinkle artifact.
Aliasing
Aliasing is an inaccurate display of color or spectral
Doppler velocity and occurs when the velocity
range exceeds the scale available to display it. The
maximum velocity scale is limited by the number
of US pulses per second that can be transmitted and
received by the transducer (ie, the pulse repetition
frequency [PRF]). Accurate depiction of frequency
shifts requires a scale that is twice as large as the
maximum shift (known as the Nyquist limit) [4]. If
the scale is too small, large shifts exceed the available range and are displayed as multiples of small
shifts. Practically, the display wraps around the
scale and overwrites the existing data. For spectral
Doppler flow toward the transducer, the velocity
peak is cut off at the top of the scale and the
Fig. 6. Color Doppler aliasing. (A) Longitudinal CDUS image of CCA directed away from transducer should be red
with maximum central velocity displayed as bright yellow. Instead, color scale wraps around and colors are
displayed sequentially from red and yellow adjacent to wall to light blue and then dark blue in central lumen.
Velocity scale range is 12cm/s. (B) At proper scale range of 23 cm/s color display no longer aliases and flow
direction is depicted appropriately.
85
86
Rubens et al
missing portion is written from the lowest portion

of the scale back toward the top [Fig. 5]. The solutions to spectral aliasing are first to drop the baseline or increase the velocity scale (ie, the PRF) to
increase the available velocity range [Fig. 5]. If the
scale is still inadequate, decrease the Doppler frequency shift by using a lower insonating frequency
or by increasing the Doppler angle [4].
For color Doppler assume a scale ranging from
red (slow) to yellow (faster) toward the transducer
and dark blue (slow) to light blue (faster) away.
Aliasing within a vessel is displayed as adjacent
colors from red to yellow to light blue to dark
blue [Fig. 6]. Increasing the velocity scale [Fig. 6]
or decreasing the frequency can also be diminished by color Doppler aliasing. In areas of the
vessel where the flow actually reverses direction,
the color palette also goes from red to dark blue
but without the yellow and light blue components
in between. Instead there is the black line of no flow
dividing the areas in which the flow has changed
direction [Fig. 7]. Power Doppler has no aliasing
because it has no directional or velocity component.
Aliasing is disadvantageous in that high velocities
may not be accurately measured; however, in dayto-day scanning, color Doppler aliasing can be useful because it quickly localizes the highest velocity
region within a vessel for spectral sampling for carotid and other vascular studies [Fig. 8] [6]. Aliasing
rapidly identifies the abnormal area in assessment of
transjugular intrahepatic portal-systemic shunt TIPS
[Fig. 9] and displays the direction of high-velocity
jets for angle-corrected velocity determination. In
addition, color Doppler aliasing readily identifies
abnormal high-velocity vessels, which are often invisible on gray-scale. In particular, arteriovenous fis-
tulae, a common sequelae to renal or hepatic biopsy

[Fig. 10] [7] are often undetectable on gray-scale.
Blooming artifact
In common terms this is known as color bleed
because the color spreads out from within the vessel
and bleeds beyond the wall into adjacent areas.
Color bleed can occur because the color US image is
actually two images superimposed, the color and
the gray-scale; thus, depending on how the parameters are set, the color portion of the image can
extend beyond the true gray-scale vessel margin.
This extension usually occurs deep to the vessels
and, most commonly, is caused by abnormally
high gain settings [Fig. 11] [8]. The unwanted
result, however, is that the information within the
vessel (ie, partial thrombus) can be written over
and obscured. Color blooming artifact can also be
seen with US contrast agents and occurs soon after
the bolus injection, at the time the increase in signal
strength is the highest [9]. B-flow, an alternative
US-based blood flow detection method, does not
use Doppler and is acquired as part of the gray-scale
image; thus, the flow cannot overwrite the grayscale anatomy. This type of imaging may be useful
when color imaging is problematic [10].
Directional ambiguity
Directional ambiguity or indeterminate flow direction refers to a spectral Doppler tracing in which the
waveform is displayed with nearly equal amplitude
above and below the baseline in a mirror image
pattern. This pattern results when the interrogating
beam intercepts the vessel at a 90 angle [5] and is
most common in small vessels, especially those that
may be traveling in and out of the imaging plane
Fig. 7. Color Doppler aliasing and flow reversal. Longitudinal CDUS image of left CCA bifurcation demonstrates
focal aliasing centrally (arrow). True flow reversal (arrowheads) in ICA bulb is recognized by thin black line that
separates blue reversed flow near wall from adjacent red forward flow in central lumen. (From Zynda-Weiss A,
Carson NL. Carotid arterial and vertebral Doppler ultrasound. In: Dogra V, Rubens DJ, editors. Ultrasound secrets.
New York: Elsevier; 2004; with permission.)
Fig. 8. Color Doppler aliasing. (A) Longitudinal CDUS image of left common carotid artery bifurcation demonstrates focal aliasing within ICA (arrowheads) indicating high-velocity jet caused by stenosis. (B) Spectral Doppler
obtained at this region also demonstrates aliasing, even with maximized scale settings. Despite this, if peak
(arrow) is added to portion written above baseline, velocity can be calculated at 275 + 219 = 494 cm/s, which
indicates a severe stenosis. (From Zynda-Weiss A, Carson NL. Carotid arterial and vertebral Doppler ultrasound.
In: Dogra V, Rubens DJ, editors. Ultrasound secrets. New York: Elsevier; 2004; with permission.)
[Fig. 12]. In a study by Ratanakorn and colleagues

[11], this artifact adversely effected measured transcranial Doppler blood-flow velocities.
Directional ambiguity should not be confused
with true bidirectional flow. In the latter case,
blood actually flows in two directions, such as in
the neck of a pseudo-aneurysm [Fig. 13]. The clue
here is that the flow is first in one direction, then
in the opposite, all within a single cardiac cycle.
Another type of bidirectional flow occurs in the
setting of high resistance organ flow (eg, torsion,
venous thrombosis, or other causes of parenchymal
edema) and is represented as diastolic flow reversal
[Fig. 14] [12]. The difference between true bidirectional flow and an indeterminate direction spectral
tracing is that bidirectional flow is never simultaneously symmetric above and below the baseline.
The flow direction varies within the cardiac cycle.
True bidirectional flow is not an artifact. In the
visceral arteries it is always abnormal and must be
recognized to make the correct diagnosis.
Partial volume artifact

Partial volume artifact results from a slice thickness
that is not infinitely thin. Echoes and Doppler signals can be acquired from objects that may be partly
87
88
Rubens et al
Fig. 9. Color Doppler aliasing in transjugular intrahepatic portal-systemic shunt (TIPS). (A) Longitudinal CDUS
image of TIPS (arrows) demonstrates aliasing (arrowheads) in hepatic end of shunt suggesting focally elevated
velocity. (B) Corresponding Doppler spectrum confirms shunt stenosis with angle- corrected flow velocity measuring 256.7 cm/s (normal velocity is < 200 cm/s). (From Zynda-Weiss A, Carson NL. Carotid arterial and vertebral
Doppler ultrasound. In: Dogra V, Rubens DJ, editors. Ultrasound secrets. New York: Elsevier; 2004; with permission.)
within the slice and partly outside of it, similar to

slicing partly through a cherry in a piece of fruit
cake. If viewed from one side, the slice seems to
have a cherry. If viewed from the other side, no
cherry seems visible. Because the signals in the US
slice are summed together, the echoes produced are
attributed to structures in the assumed thin scan
plane [13]; thus, echoes can appear within anechoic
structures and Doppler signals are acquired in an
area in which no vessels are perceived on gray-scale
[3]. For example, on a longitudinal gray-scale image, echoes from gas in the duodenum may appear
within the gallbladder and mimic stones or polyps;
however, if you rotate the transducer and image
from the transverse plane, the gas is clearly adjacent
to the gallbladder and not within it. On color flow
imaging, an example of partial volume artifact is
visualization of a portion of the iliac artery within
the ovary giving the impression of abnormal cyst
wall flow. Spectral analysis of this vessel shows the
high resistance waveform typical of an iliac artery
[Fig. 15] and imaging from the 90 plane clearly
shows the vessel separate from the ovary. Partial
volume artifact may be produced by grating lobes
or side lobes, which generate information outside
the expected path of the main beam. These off-axis
lobes are located peripheral to the main beam axis
[Fig. 16] [14]. Side lobes occur close to the primary
beam whereas grating lobes can be far removed
from the central beam [15]. These off-axis lobes
can interrogate vessels that are separate from the
primary sample volume. The lobes may appear on
the spectral tracing as a flowing vessel where none is
expected or display bidirectional flow as a result
of interrogating the vessel from multiple angles
[Fig. 12]. These transducer related artifacts are

seen mainly with the high frequency, tightly curved,
convex, linear arrays used in endocavitary probes,
and depend on the crystal element size and the
spacing of the array elements [5].
Pseudoflow
Pseudoflow is defined as presence of flow of a fluid
other than blood [7]. Pseudoflow can mimic real
blood flow with color or power Doppler US, but no
true vessel containing the fluid exists [Fig. 17]. The
color or power Doppler signal appears as long as the
fluid motion continues. These artifacts may be misinterpreted as flow unless Doppler spectral analysis is
used. The spectral Doppler tracing does not exhibit a
normal arterial or venous waveform [1]. Spontaneous
examples of pseudoflow include ascites [Fig. 18],
amniotic fluid, and urine (bladder jets). Bladder jets
identify the ureteral orifice and are useful to exclude
complete obstruction or to denote asymmetric ureteral emptying in the case of partial obstruction
[Fig. 19] [16]. Bladder jets are not completely reliable,
however, because 30% of obstructed patients may
display normal jets [17]. Conversely, normal patients
in the 2nd and 3rd trimester of pregnancy may have
asymmetric or absent jets partly caused by uterine
pressure. These jets can mostly be restored by scanning in the decubitus position, however, because the
asymmetry may be physiologic and not necessarily
abnormal, using diminished jets to diagnose obstruction in pregnancy still remains problematic [18].
Flash artifact
Flash artifact is a sudden burst of random color
that fills the frame, obscuring the gray-scale image.
Fig. 10. Aliasing identifies an arteriovenous fistula (AVF). (A) Longitudinal CDUS image detects area of focal
aliasing (arrows) indicating high-velocity flow in renal hilum, and suggests arteriovenous fistula. (B) Doppler
spectrum demonstrates low resistance arterial waveform directed above baseline and high-velocity arterialized
venous waveform below baseline, diagnostic of AVF.
This artifact may be caused by object motion or

transducer motion [Figs. 20 and 21] [7]. Flash
artifact may occur anywhere but is most commonly
seen in the left lobe of the liver (as a result of
cardiac pulsation) and in hypoechoic areas, such
as cysts or fluid collections [Fig. 22] [5]. Flash
artifact can be used to denote the fluid nature of
solid-appearing material [Fig. 23] [1]. Power Doppler is more susceptible to flash artifact than color
flow Doppler because of the longer time required to
build the image (in general, more frames are averaged to create the image than with standard color
Doppler) [19].
Although generally disruptive, motion artifacts
can be extremely useful diagnostically. The so-called
perivascular artifact or color bruit is a tissue
motion artifact whereby the motion is generated
within an organ, rather than involving an entire
organ or image. This artifact appears as a random
color mosaic in the soft tissues (as opposed to a
single homogeneous color), occurs adjacent to vessels with turbulent flow, and is believed to be
caused by actual vascular tissue vibration [20].
This artifact is the imaging equivalent to an auditory bruit or palpable thrill; varies with the cardiac cycle; is most prominent in systole; is absent
or less prominent in diastole; is seen particularly
in association with anastomotic sites, stenotic arteries, or arteriovenous fistulae [Fig. 24]; and can
be extremely useful to detect their presence.
Mirror-image artifact
The mirror image artifact displays objects on both
sides of a strong reflector, though they are located
only on one side of it [21,22]. The reflector (eg, the
diaphragm, pleural surface, or aortic wall) directs
some of the echoes to a second reflector before
it returns them to the transducer, resulting in a
multipath reflection [Fig. 25] [14]. The machine
straightens out the multipath echoes assuming
89
90
Rubens et al
Fig. 11. Blooming artifact (A) Longitudinal CDUS image of right common femoral vein (CFV) shows blooming
artifact deep to vessel, displaying color (arrow) beyond vessel wall (arrowheads). Color is uniform and no clot
displayed. Initial examination was interpreted as normal. Scale is low at 0.11 and gain high at 50. (B) Axial CT just
above bifurcation shows intraluminal partial thrombus (arrow) in right CFV. CT was obtained same day as (A).
(C ) Following CT, directed CFV CDUS was performed. Increasing scale to 0.17 and decreasing gain to 38 shows
thrombus (arrow), which was initially missed. (D) Corresponding gray-scale image to C shows thrombus (cursors),
which is larger than in CDUS image, indicating some color pixels are still overwriting gray-scale, particularly in
darker portions of clot. (E) Axial CDUS displays thrombus centrally within CFV, identical to CT.
Fig. 12. Longitudinal CDUS image through infant testis shows arterial spectral Doppler waveform with
equal amplitude above and below baseline, yielding an indeterminate flow direction. This occurs most often in
small vessels.
that the echoes come from the initial transducer

beam and from a distance corresponding to the
actual time of flight, resulting in a display of the
echoes deeper in the image than they should be.
The resulting artifact shows up as the virtual object,
deep to the original image but identical to itthus
the term mirror.
Mirror images may be produced with gray-scale,
color, power, and spectral Doppler. With Doppler,
mirror image artifact commonly occurs adjacent to
the highly reflective lung in the supraclavicular
region [23]. Reflection off the pleura causes an apparent duplication of the subclavian artery or vein.
The sound is bounced back from the surface of
the lung to the moving blood cells and the resulting Doppler shift is reflected back to the surface
of the lung and then to the transducer. The extra
time taken causes the appearance of a second vessel
deep to the real subclavian vessel, referred to as
the mirror image artifact. The phantom vessel is
always projected deeper in the image [Fig. 26]
[5]. A carotid ghost is the term for a mirror image
of the common carotid artery. The carotid ghost is
always located deep to the common carotid artery
regardless of location and positioning of the transducer [23].
Edge artifact
Fig. 13. True bidirectional flow in a pseudoaneurysm. Doppler spectrum at neck of pseudoaneurysm
demonstrates true bidirectional flow with sequential
flow first into and then out of aneurysm in each cardiac cycle.
Edge artifact refers to the Doppler signal generated

at the margin of a strong, smooth, specular reflector, displayed on imaging as persistent color along
the rim of calcified structures, such as gallstones
[Fig. 27] or cortical bone, and may mimic vascularity unless the spectral tracing is obtained [24].
Edge artifact may be generated by any echogenic
surface, including manmade structures (eg, catheters and foley balloons [Fig. 28]). The diagnostic
feature is the Doppler spectrum, a straight-line pattern, equal above and below the baseline, and
representing noise, not flow. Edge artifact is seen
more commonly with power Doppler US than with
color Doppler US because of a larger dynamic
91
92
Rubens et al
Fig. 14. Diastolic flow reversal (A) Spectral tracing in left hepatic artery 2 days after liver transplantation shows
diastolic flow reversal (arrows) indicating high resistance to arterial flow. Resistive index (RI) is 1.0 (B) 1 day later
normal continuous forward diastolic flow has been re-established and RI is normal at 0.7.
range [25]. These artifacts are more frequent at low

PRF or velocity scale as a result of the increased
sensitivity of the system but may also be caused by
a low wall filter setting [26].
Twinkling artifact
In 1996, twinkling artifact was described by Rahmouni and colleagues [25] as color Doppler signals
that imitate motion or flow behind a stationary
strongly reflecting interface. The twinkling artifact
can be seen behind any granular (irregular or
rough) reflecting surface but is commonly caused
by renal calculi, bladder calcification, and cholesterol crystals in the gallbladder [Fig. 29]. The twin-
kling Doppler is a mosaic of rapidly changing

colors located deep to an echogenic reflector. With
power Doppler, the signal location is the same, but
the color is uniform [Fig. 30].
Twinkling artifact is believed to be caused by a
narrow band of intrinsic machine noise called
phase (or clock) jitter [27]. On a flat surface, system
noise generates a narrow band of Doppler shift as a
result of tiny clock errors. This tiny shift is usually
excluded by the wall filter and, therefore, is not
displayed as color. Rough surfaces increase the
delays in measuring signal and amplify the errors,
increasing the spectral bandwidth of this noise
above the level of the wall filter. The spectrum is
Fig. 15. Partial volume artifact in ovary. (A) Longitudinal CDUS image of left adnexa demonstrates vessel (arrow)
along margin of ovarian cystic mass (M), creating concern that mass is vascular. (B) Doppler spectrum of this vessel
reveals high resistance arterial waveform. (C ) Axial image shows vessel (arrows) is actually adjacent to ovary and
separate from it, not within cyst wall. (D) Doppler spectral waveform is identical to that in B, and is typical of
internal iliac artery.
93
94
Rubens et al
Fig. 15 (continued).
typical of noise, with multiple closely applied

spikes that are written equally above and below
the baseline [Fig. 30].
Detection of the twinkling signal depends on the
color-write priority and, in some instances, on grayscale gain. As color-write priority decreases, more
gray-scale is displayed, and the amount of twinkling artifact decreases behind the stone [27]. At
high color-write priorities, the gray-scale has less
effect. These settings vary from machine to machine
and the relationships between the settings also vary
Fig. 16. Sidelobes. Diagram displays origins of side

lobes and grating lobes and their relationship to
main beam. Echoes returning from either of these
additional lobe sources is displayed as though they
originated from main beam.
with manufacturer. To consistently obtain a twinkling artifact, a high color-write priority should be
selected and gray-scale gain kept to a minimum.
Why produce an artifact? Unlike many artifacts
that are problematic, the twinkling artifact can be
extremely useful. Similar to gray-scale shadowing,
twinkling artifact also may be useful to identify
stones. Small stones that may not generate a strong
echo or cast an acoustic shadow still can produce a
twinkling artifact, leading to their identification
[Fig. 31] [1]. Similar to an acoustic shadow, twinkling does not occur 100% of the time. In a study
of 32 patients by Lee and colleagues[28], only 86%
of urinary calculi demonstrated a twinkling artifact;
furthermore, the chemical composition of stones is
related to the production of the artifact. Chelfouh
and colleagues [29] reported that calcium oxalate
dihydrate and calcium phosphate calculi always
produced a twinkling artifact, whereas stones composed of calcium oxalate monohydrate and urate
lacked a twinkling artifact. Besides renal calculi, a
twinkling artifact may be seen behind material with
an irregularly reflective, granular surface, such as
iron filings, emery paper, ground chalk, wire mesh,
an aneurysm coil during transcranial Doppler
sonography [30], gall bladder adenomyomatosis
[31], or, recently, encrusted stents [32]. Although
the twinkling artifact cannot be generated 100% of
the time, it can be extremely useful in the detection
of renal calculi and some foreign bodies. The key to
the twinkling artifact is that the color produced
behind the calcification and the concomitant Doppler spectral tracing shows noise, not flow; thus, a
calcified carotid plaque with twinkling can be
Fig. 17. Pseudoflow caused by fluid in ligamentum teres. (A) Transverse CDUS color flow image of liver demonstrates simulated vessel (arrows) coursing along falciform ligament. (B) Longitudinal CDUS image of same
simulated vessel. (C ) Axial image at another time point shows flow in posteriorly located splenic vein (arrowhead)
but no flow around falciform. (D) Spectral Doppler tracing displays noise and no true flow. (From Campbell SC,
Cullinan JA, Rubens DJ. Slow flow or no flow? Color and power Doppler US pitfalls in the abdomen and pelvis.
Radiographics 2004;24:497506; with permission.)
differentiated from a potentially ulcerated plaque

with flow in the ulcer cavities [Fig. 32].
Day-to day Doppler: too much flow versus

too little flow
The artifacts described in the section Doppler Artifacts primarily relate to the generation of Doppler
signals by nonvascular structures or fluids. The key
to their recognition is (1) knowing that they can
occur, (2) knowing the common locations and
causes for their generation, and (3) identification
of the nonvascular Doppler spectrum they generate,
which clinches the diagnosis. In day-to-day clinical
practice, the more common problems are too much
flow, which may obscure thrombi, or too little

flow, giving the false diagnosis of thrombosis.
Too much flow usually can be recognized by
seeing color bleed [Fig. 11], or seeing aliasing in
a vessel that normally does not have it [Fig. 33].
This problem can be corrected by increasing the
scale or decreasing the gain. Another common imaging problem occurs when uninterrupted flow is
imaged from a segment of a vessel and flow is assumed the same throughout the rest of the lumen.
This problem usually occurrs in longitudinal vascular imaging whereby a partial thrombus or atheromatous plaque may not be imaged if it is not
centered in the imaging plane [Fig. 34]. The failsafe if inappropriate settings are not recognized is
95
96
Rubens et al
Fig. 18. Bladder jet. (A) Patient who presents with right renal colic. Transverse CDUS image through bladder shows
normal jet (arrow) from asymptomatic side. Note right ureteral calculus (arrowhead). (B) Transverse CDUS image
shows smaller right ureteral jet (arrow) indicating partial, but not complete, obstruction secondary to ureteral
calculus (arrowhead). (From Campbell SC, Cullinan JA, Rubens DJ. Slow flow or no flow? Color and power Doppler
US pitfalls in the abdomen and pelvis. Radiographics 2004;24:497506; with permission.)
always to image in two planes; thus, even if the colorwrite priority is too high or the imaging plane is not
centered and the thrombus is overwritten in the
long axis of the vessel, the clot can be recognized
in the short axis plane [Fig. 34].
The more common problem is too little flow,
which mimics thrombosis. First, Doppler angle
should be as small as possible. Obtaining signals
for flow at 90 to the probe is always difficult
[Fig. 2]. The scale should be set appropriately for
the vessel you being interrogated. Too high a scale
eliminates slow flow within the vessels [Fig. 35].
The frequency should be appropriate: low frequency for deep structures [Fig. 36] and high fre-
quency for superficial ones [Fig. 1]. The frequency

to demonstrate color flow Doppler is generally
lower than the frequency needed for gray-scale
imaging, so the Doppler frequency may need to
be decreased if it does not default to the correct
frequency. Frequency filters and other algorithms
designed to decrease color tissue noise can eliminate display of slow flowing blood if they are set
too high [Fig. 2]. In general, reducing the size of the
color box reduces the sample size and increases the
frame rate, leading to better overall sensitivity and
resolution of the color image. Lastly, the color-write
priority may be set automatically or manually. With
a high color-write priority the sensitivity for color
Fig. 19. Pseudoflow in ascites. (A) Longitudinal CDUS image in mid-abdomen in cirrhotic patient with ascites.
Two linear flowing streams exist. More caudal stream (with Doppler cursor) has spectral Doppler waveform, which
has random flow above and below baseline unrelated to any visceral vascular pattern. Motion in ascites occurs.
(B) Cranial stream is continuous, unidirectional, and monophasic (arrows), typical of portal vein, representing
patent umbilical collateral vessel. (From Campbell SC, Cullinan JA, Rubens DJ. Slow flow or no flow? Color and
power Doppler US pitfalls in the abdomen and pelvis. Radiographics 2004;24:497506; with permission.)
Fig. 20. Flash artifact: patient motion. (A) Longitudinal CDUS through the left lobe of liver with flash artifact (arrows)
produced by respiratory motion. (B) Longitudinal CDUS with no motion shows normal vascular flow with no artifact.
Fig. 21. Flash artifact: transducer motion. (A) Longitudinal CDUS of the left testis with flash artifact (arrows)
caused by transducer motion. (B) Without motion, normal testicular vessels are easily identified.
97
98
Rubens et al
Fig. 22. Flash artifact in amniotic fluid caused by motion of fetal head. CDUS image (left) of lower uterine segment
showing fetal head (H) and cervix (C). Flash of color (arrow) appears across internal os caused by application of
fundal pressure while scanning, and simulates vasa previa. On corresponding CDUS image with no fundal pressure
applied (right), no color is detected. (From Campbell SC, Cullinan JA, Rubens DJ. Slow flow or no flow? Color and
power Doppler US pitfalls in the abdomen and pelvis. Radiographics 2004;24:497506; with permission.)
Fig. 23. Flash artifact to identify fluid for aspiration. (A) Transverse CDUS image in left thigh of immunocompromised patient presenting with left leg pain, originally evaluated for venous thrombosis with negative examination. Imaging in area of tenderness showed hypoechoic mass (arrows) with some internal echoes. (B) With
compression applied to mass, anterior portion fills with color (arrows) indicating liquid, not solid mass. The
apparent mass was aspirated and was an abscess.
Fig. 24. Color Doppler bruit sampling in region of color Doppler bruit (mosaic of colors) shows typical spectrum of
arteriovenous fistula with high-velocity arterial flow (1 m/s) and even higher venous flow (below baseline).
Fig. 25. Multipath reflection diagram of US wave path required to produce mirror image artifact. Pulse begins at
transducer, is deflected by mirror (usually diaphragm or pleura) and hits target. Reflected echo returns to mirror
and then to transducer, requiring much longer transit time than if pulse had interacted with object directly; thus,
mirror image is displayed deeper in field of view.
99
100
Rubens et al
Fig. 26. Mirror-image artifact. (A) Anterior true vessel and (B) posterior mirror image of subclavian artery show
identical spectra. Mirror in case is pleura (arrows). (C ) Similar situation is noted with subclavian vein anteriorly
and (D) its mirror image posteriorly with pleura (arrows) between them. Mirror-image vein should not be
mistaken for collateral vessel.
Fig. 27. Edge artifact from gallstone. Power Doppler image demonstrates color signal along rim of gallstone
simulating a gallbladder mass. Spectral tracing is typical of noise, with nonvascular pattern displayed equally
above and below baseline. (From Campbell SC, Cullinan JA, Rubens DJ. Slow flow or no flow? Color and power
Doppler US pitfalls in the abdomen and pelvis. Radiographics 2004;24:497506; with permission.)
101
102
Rubens et al
Fig. 28. Edge artifact from foley catheter. (A) Transverse CDUS image through bladder shows spherical mass
centrally with marked Doppler signal around its margins (arrows). (B) Spectral Doppler confirms high amplitude
continuous noise, equal and symmetric above and below baseline.
Fig. 29. Twinkling artifact longitudinal CDUS in patient with cholesterol crystals in gallbladder. Crystals generate
twinkling artifact (arrowheads) posteriorly.
Fig. 30. Twinkling artifact. (A) Longitudinal image of bladder shows typical ureterovesical junction stone (arrow)
with posterior shadow. (B) Transverse CDUS image of bladder shows right ureteral calculus (arrow) and twinkling
artifact generated posteriorly (arrowheads). (C ) Power Doppler also generates signal (arrowheads) posterior to
stone. (D) Corresponding Doppler spectrum through twinkling color shows equal amplitude noise above and
below baseline. Same spectral tracing is generated whether color or power Doppler images twinkle. (From
Campbell Campbell SC, Cullinan JA, Rubens DJ. Slow flow or no flow? Color and power Doppler US pitfalls in the
abdomen and pelvis. Radiographics 2004;24:497506; with permission.)
103
104
Rubens et al
Fig. 31. Small renal stones with twinkling artifact. (A) Longitudinal US image through kidney shows minimal
hydronephrosis (arrow) but no stones. (B) CDUS image in same position shows marked twinkling artifacts (arrows)
at upper and lower poles, identifying stones, which do not cast an acoustic shadow.
Fig. 32. Twinkling artifact in carotid. (A) Twinkling artifact (arrowheads) occurs behind calcifications (arrows) in
atherosclerotic plaque, not to be mistaken for ulceration and disturbed flow. (B) Calcifications (arrows) are better
visualized on gray-scale image. (From Campbell SC, Cullinan JA, Rubens DJ. Slow flow or no flow? Color and power
Doppler US pitfalls in the abdomen and pelvis. Radiographics 2004;24:497506; with permission.)
Fig. 33. Portal vein clot obscured. (A) Initial transverse CDUS image through portal vein shows color filling lumen;
however, aliasing is occurring (arrow) and scale is too low at 23. (B) Repeat imaging with scale increased to 38 with
other factors remaining constant permits detection of clot (arrow), which is isoechoic to liver.
Fig. 34. Off axis imaging. (A) Longitudinal CDUS through common femoral vein (CFV) is normal. (B) With
slight repositioning of transducer, large partial thrombus is identified posteriorly (arrows). (C ) Transverse imaging
shows partial thrombus centered in vein. Imaging in sagittal plane angled either side of thrombus creates false
negative diagnosis.
Fig. 35. Partial thrombus in left portal vein? (A) Initial transverse CDUS image shows only partial filling of
left portal vein, simulating a thrombus (arrowheads) on right wall. Mean velocity scale is 17cm/s, too high for
slow-flowing portal vein. (B) Repeat transverse image with scale at 10 cm/s shows normal filling of vein and
no thrombus.
105
106
Rubens et al
Fig. 36. No flow in central TIPS. (A) Initial longitudinal spectral Doppler sampling through TIPS shows normal
inflow at portal vein end. (B) Hepatic vein portion also has normal flow. (C ) In mid TIPS between hepatic and
portal venous segments, there is no demonstrable flow. Spectral frequency is 3 MHz (arrow). (D) Normal flow in
mid TIPS is documented using frequency of 2 MHz (arrow).
107
108
Rubens et al
Box 1: Guidelines for an optimal color flow

Doppler examination
Adjust the gain and filter settings to obtain an optimal color signal and minimal
color noise.
Adjust the velocity scale (PRF) and baseline
according to the flow conditions. A low scale
is used for low flows and velocities; however,
it may produce aliasing. A high scale reduces
aliasing but is less sensitive for slow flows.
Obtain an optimal Doppler angle by adjusting the beam steering and probe position.
The angle should be 60 or less if velocity
measurements are to be made.
The color flow box should be kept as small as
possible to allow better frame rate for better resolution and sensitivity.
Adjust the pulsed Doppler sample volume
size appropriately (two thirds of the vessel
diameter) to obtain accurate velocities.
Avoid transducer motion.
flow can be prioritized so color shows up where no

vessel was visible on gray-scale imaging, and if the
color-write priority is set too low, flow may be
missed. Some practical guidelines to optimize a
color Doppler examination are listed in Box 1.
Summary
With increasing technologic advances in US, its
applications have continued to grow for the detection of pathology and physiology. To avoid misinterpretation of results, however, the Doppler US
practitioner must understand the factors that produce a Doppler signal, whether vascular, motion,
or artifact. Color or power Doppler artifacts can be
verified by their atypical spectral waveform. Some
artifacts, such as aliasing (for rapid detection of
stenoses or arteriovenous fistulae) and the twinkle
artifact (for identification of renal calculi and verification of other stones or crystals), are extremely
useful diagnostically. Careful attention to the technical parameters of frequency, gain, filter and scale
is required to correctly identify vascular patency or
thrombosis, especially in slow-flowing vessels.
References
[1] Campbell SC, Cullinan JA, Rubens DJ. Slow flow
or no flow? Color and power Doppler US pitfalls
in the abdomen and pelvis. Radiographics 2004;
24:497506.
[2] Merritt CRB. Physics of ultrasound. 3rd edition.
Philadelphia: Mosby; 2004.
[3] Zweibel WJ, Pellerito JS. Basic concepts of Doppler frequency spectrum analysis and ultrasound
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
blood flow imaging. Introduction to vascular

ultrasonography. Philadelphia: Elsevier Saunders; 2005.
Taylor KJ, Holland S. Doppler US. Part I. Basic
principles, instrumentation, and pitfalls. Radiology 1990;174:297307.
Pozniak MA, Zagzebski JA, Scanlan KA. Spectral
and color Doppler artifacts. Radiographics 1992;
12:3544.
Mitchell DG. Color Doppler imaging: principles, limitations, and artifacts. Radiology 1990;
177:110.
Campbell SC. Doppler Ultrasound artifacts. Philadelphia: Elsevier; 2004.
Nilsson A. Artifacts in sonography and Doppler.
Eur Radiol 2001;11:130815.
Forsberg F, Liu JB, Burns PN, et al. Artifacts in
ultrasonic contrast agent studies. J Ultrasound
Med 1994;13:35765.
Weskott HP. [B-flowa new method for detecting blood flow]. Ultraschall Med 2000;21(2):
5965.
Ratanakorn D, Kremkau FW, Myers LG, et al.
Mirror-image artifact can affect transcranial Doppler interpretation. J Neuroimaging 1998;8:1757.
Dogra VS, Sessions A, Mevorach RA, et al.
Reversal of diastolic plateau in partial testicular
torsion. J Clin Ultrasound 2001;29(2):1058.
Goldstein A, Madrazo BL. Slice-thickness artifacts
in gray-scale ultrasound. J Clin Ultrasound 1981;
9:36575.
Scanlan KA. Sonographic artifacts and their origins. AJR Am J Roentgenol 1991;156:126772.
Zagzebski JA. Doppler instrumentation. St.Louis
(MO): Mosby; 1989.
Burge HJ, Middleton WE, McClennan BL, et al.
Ureteral jets in healthy subjects and in patients
with unilateral ureteral calculi: comparison
with color Doppler US. Radiology 1991;180:
43742.
Sheafor DH, Hertzberg BS, Freed KS, et al. Nonenhanced Helical CT and US in the emergency
evaluation of patients with renal colic: prospective comparison. Radiology 2000;217(3):7927.
Karabulut N, Karabulut A. Colour Doppler evaluation of ureteral jets in normal second and third
trimester pregnancy:effect of patient position. Br
J Radiol 2002;75(892):3515.
Zagzebski JA. Physics and instrumentation in
Doppler and B mode ultrasonography. Introduction to vascular ultrasonography. Philadelphia:
W.B. Saunders; 2000.
Middleton WD, Erickson S, Melson GL. Perivascular color artifact: pathologic significance and
appearance on color Doppler US images. Radiology 1989;171:64752.
Kremkau FW, Taylor KJ. Artifacts in ultrasound
imaging. J Ultrasound Med 1986;5:22737.
Cakmakci H, Gulcu A, Zenger MN. Mirror-image
artifact mimicking epidural hematoma: usefulness of power Doppler sonography. J Clin Ultrasound 2003;31:4379.
[23] Middleton WD, Melson GL. The carotid ghost.

A color Doppler ultrasound duplication artifact.
[24] Cardinal E, Lafortune M, Burns P. Power Doppler US in synovitis: reality or artifact? Radiology
1996;200:8689.
[25] Rahmouni A, Bargoin R, Herment A, et al. Color
Doppler twinkling artifact in hyperechoic regions. Radiology 1996;199:26971.
[26] Rubaltelli L, Khadivi Y, Stramare R, et al. Power
Doppler signals produced by static structures: a
frequent cause of interpretation errors in the
study of slow flows. Radiol Med (Torino) 2000;
99:1614.
[27] Kamaya A, Tuthill T, Rubin JM. Twinkling artifact
on color Doppler sonography: dependence on
machine parameters and underlying cause. AJR
[28] Lee JY, Kim SH, Cho JY, et al. Color and power
Doppler twinkling artifacts from urinary stones:
clinical observations and phantom studies. AJR
[29] Chelfouh N, Grenier N, Higueret D, et al. Characterization of urinary calculi: in vitro study of
twinkling artifact revealed by color-flow sonography. AJR Am J Roentgenol 1998;171:105560.
[30] Khan HG, Gailloud P, Martin JB, et al. Twinkling
artifact on intracerebral color Doppler sonography. Am J Neuroradiol 1999;20:2467.
[31] Ghersin E, Soudack M, Gaitini D. Twinkling
artifact in gallbladder adenomyomatosis. J Ultrasound Med 2003;22:22931.
[32] Trillaud H, Pariente JL, Rabie A, et al. Detection of
encrusted indwelling ureteral stents using a twinkling artifact revealed on color Doppler sonography. AJR Am J Roentgenol 2001;176: 14468.
109
111
ULTRASOUND
CLINICS
Carotid Artery Ultrasound

Interpretation Using a Pattern
Recognition Approach
Michelle L. Robbin,
&
&
&
MD*,
Mark E. Lockhart,
General principles
Carotid ultrasound examination
Scanning technique
Waveform analysis
Normal carotid ultrasound waveform
pattern analysis
The carotid ultrasound examination is an excellent

screening tool to assess for hemodynamically significant carotid stenosis. The uncomplicated positive
carotid ultrasound is often the sole preoperative
study before angioplasty and stent placement or
endarterectomy. Although carotid ultrasound may
seem deceptively simple, there are many potential
pitfalls in the examination performance and interpretation. This article provides a conceptual framework of normal and abnormal carotid artery
physiology using waveform analysis. Once the hemodynamic contributions of stenoses and occlusions of
the various branches of the carotid artery are understood, common and uncommon presentations of
disease can be correctly diagnosed. Imaging protocols
and technique are discussed with an emphasis on
accurately depicting the pathology present.
General principles
Stroke is the third leading cause of death in the
Western world, after heart disease and cancer [1].
&
&
MD, MPH
Abnormal carotid ultrasound waveform

pattern analysis
Summary
References
The most common identifiable cause of stroke is

thromboembolic clots from the heart. Another
important identifiable cause of new stroke is atherosclerotic plaque causing stenosis of the carotid
bifurcation extending into the internal carotid artery [2]. Physical examination findings, such as a
neck bruit, the presence of extensive atherosclerotic
disease elsewhere, or clinical symptomatology suggestive of transient ischemic attack or stroke, are
important clues that suggest the possibility of hemodynamically significant carotid stenosis. Ultrasound is an inexpensive, highly accurate initial
screening study in the assessment of carotid artery
disease. When a high-quality ultrasound is performed by appropriately trained personnel, it is
generally unnecessary to obtain a corroborating preoperative angiogram or MRI unless there are discordant ultrasound findings.
The North American Symptomatic Carotid Endarterectomy Trial showed the benefit of surgery
in correcting hemodynamically significant carotid
artery stenoses. It demonstrated the strong benefit
of endarterectomy in appropriate symptomatic pa-
University of Alabama at Birmingham Hospital, 619 19th Street, South, Alabama 35249-6830, USA
E-mail address: mrobbin@uabmc.edu (M.L. Robbin).
doi:10.1016/j.cult.2005.09.011
112
Robbin & Lockhart
tients who had 70% to 99% internal carotid artery

stenosis [3,4]. The Asymptomatic Carotid Atherosclerosis Study evaluated the efficacy of endarterectomy in patients who were asymptomatic and
had carotid artery stenosis greater than 60% and
showed that these patients had a reduced risk for
subsequent stroke [5]. Before and after these impor-
tant studies, much effort has been directed at accurately detecting variable levels of stenosis with
ultrasound, using measurements such as peak systolic velocity and various spectral Doppler ratios
[6]. A multidisciplinary Society of Radiologists in
Ultrasound consensus conference met recently to
develop some general guidelines for carotid ultra-
Fig. 1. Normal carotid study. (A) Longitudinal spectral Doppler of ICA has normal low-resistance flow with higher
EDV (arrow) than the ECA or CCA. Prox, proximal. (B) Normal high-resistance spectral waveform is present in the
ECA. EDV (arrow) is lower than in the ICA or CCA. (C ) CCA spectrum shows low-resistance flow with no spectral
broadening. The EDV (arrow) is between that of the ICA and the ECA. (D) Spectral Doppler of vertebral artery
shows low-resistance waveform with moderate EDV (arrow).
sound for use in laboratories that do not have their

own internally validated criteria [7]. Although it is
optimal to perform an analysis of ones own laboratory ultrasound findings with respect to angiographic findings, it can be a difficult exercise.
Generally, few angiograms are performed to corroborate sonographic findings, and exact determination of the percent diameter stenosis at surgery is
not usually performed, making correlation difficult.
Using defined internal laboratory standards or
standards derived from a multidisciplinary consensus
Fig. 1 (continued ).
panel does not guarantee a high-quality ultrasound

examination. The examinations need to be performed by appropriately trained personnel, with
national credentials such as Registered Diagnostic
Medical Sonographer (RDMS), Registered Vascular
Technologist (RVT), or the American Registry of Radiologic Technologists, Vascular Sonography, R.T.
(R) (VS) (ARRT). Laboratory accreditation is essential
in demonstrating a commitment to excellence and
adherence to standards, as defined by the American
College of Radiology, American Institute of Ultra-
113
114
Robbin & Lockhart
sound in Medicine, or the Intersocietal Commission

for the Accreditation of Vascular Laboratories.
Carotid ultrasound examination

Scanning technique
General
The patient should be examined in the supine position with arms by the sides. The shoulders should
be lowered to increase access to the neck if needed.
It is often helpful to tell the patient to reach for the
feet if they have a short or thick neck. The head
should be turned slightly away from the side being
examined. In general, the patient is not scanned
with a pillow under the head. If adequate access to
the neck is still difficult, a pillow under the shoulders may help hyperextend the neck. Sonographer position is variable, with some sonographers
preferring to sit at the top of the patients head and
table, scanning facing downward from the patients head to his shoulders. Others prefer to scan
from the patients side, facing upward toward
the head.
A 5-MHz or higher linear transducer should be
used, with images optimized to examine the carotid
in the near field. A higher frequency linear transducer is encouraged to optimize gray-scale imaging
if technically possible. Particular attention should
be paid to placing the focal zone at the level of the
carotid so that the gray-scale images show the finest
detail possible. The transducer should be placed
slightly posterolaterally on the neck, using the sternocleidomastoid muscle as an acoustic window.
The transducer can be positioned more anteriorly,
along with changes in the patients neck position,
if necessary.
B-mode (gray-scale)
It is important to get an overall impression of
the amount of atherosclerotic disease in the carotid artery and the location of the branches before focusing on a particular arterial segment.
The transverse plane is used in this initial analysis,
scanning from the proximal common carotid
artery (CCA) through the bifurcation, and then
to the internal and external carotid arteries (ICA
and ECA, respectively). The probable positions
of the ICA and ECA can be determined and verified later with spectral Doppler. The ECA is usually a smaller artery oriented more medially and
anteriorly than the ICA, as it supplies the face.
The ICA tends to be slightly larger than the ECA,
with angulation posteriorly and laterally with respect to the ECA, although anatomic variability
can occur.
Next, the carotid artery is scanned longitudinally from the origin of the CCA through the bifur-
cation. The transducer is angled slightly anteriorly

toward the face to image the ECA. The ICA is found
using slight angulation of the transducer posteriorly
and laterally in most patients. A complete grayscale evaluation includes a general assessment of
soft and hard plaque and the documentation of
less-common findings, such as plaque ulceration
or dissection.
Color/power Doppler
Color Doppler can be useful in detecting areas of
flow abnormality that need to be further investigated with spectral Doppler to assess their significance. Optimizing color Doppler in an individual
vessel segment is important, as it can speed up the
task of finding stenoses. As is the case with spectral
Doppler, the angle of the color box should be 60
or less. Modern ultrasound scanners have preprogrammed scanner settings individualized for
different vessels, such as the carotid artery. The
sophisticated sonographer should feel comfortable
with the many aspects that are controllable in the
image, and change them as necessary to optimize
the information displayed.
One of the first things that should be addressed
when imaging a vessel is the color Doppler gain. It
should be increased until color speckles are seen
outside of the lumen, then turned down until the
color is displayed only within the vessel lumen. If
the color Doppler gain is set too high, color pixels
may be placed outside of the true lumen of the
artery, termed an artificial color bleed outside of
the vessel lumen. If this event occurs, plaque and
stenoses may be missed. If the color Doppler gain
is set too low, the amount of flow in a vessel can
be underestimated significantly and an occlusion
could be erroneously diagnosed. Power Doppler
should be used when occlusion or a trickle of
flow is suspected, as it is a more sensitive lowflow, angle-independent technique [8].
The color velocity scale is another parameter
that should be individually optimized for each
vessel segment. The color velocity scale should be
adjusted so that color is seen within the lumen
and the direction of flow in the vessel is clear. If
the velocity scale is set too low, aliasing will occur
and important information regarding areas of
flow disturbance and flow jets from stenoses
could potentially be overlooked. The velocity filter
controls the amount of flow seen within the vessel depending on its velocity characteristics, and
should be set as low as possible if an occlusion
is suspected.
The color Doppler image is based on the average
Doppler shift within the vessel, not the peak Doppler shift within the vessel [9]. Color Doppler will
give an estimate of the qualitative amount of the
stenosis. Spectral Doppler is a necessary component of the carotid ultrasound examination to

quantitate the degree of stenosis.
Spectral Doppler
After the gray-scale and color Doppler images have
been optimized, accurate measurement of the spectral Doppler signal is arguably the most important
portion of the examination. The angle of spectral

Doppler insonation should be 60 or less for reproducible and accurate velocity measurement. There
is some debate as to the correct placement of the
angle correction cursor. The normal default position should be parallel to the posterior carotid wall
segment. The exception to this rule occurs when
there is a color jet seen, usually extending distally
Fig. 2. High-resistance CCA. (A) Longitudinal spectral Doppler shows unilateral high-resistance flow (with no
diastolic flow) in the CCA, suggesting ICA stenosis versus occlusion. (B) No flow is identified in the mid-distal ICA
on color and spectral Doppler (arrow).
115
116
Robbin & Lockhart
or downstream from a stenosis. In this instance, the

angle should be placed parallel to the direction of
the jet and not the posterior carotid wall.
It is important to not only assess the angle, but
also the sample gate width when performing reproducible accurate spectral Doppler. A small sample gate width should be used, between 1.5 and
2.5 mm wide. Even in normal laminar flow, the
range of velocities present in the bloodstream
ranges from highest in the middle of the vessel to
lower near the vessel wall. The range gate is meant
to depict the representative flow in the vessel, not
all the spectral velocities that exist in the vessel.
Therefore, a small sample gate should be used at
the center of the vessel or in the center of a jet distal
to a stenosis. If too large a spectral gate is used, a
widened spectral Doppler tracing (spectral broadening) could lead to the erroneous diagnosis of
disturbed or turbulent flow.
Disturbed flow patterns indicate that the movement of blood is less orderly than in laminar flow,
and can be found in vessel kinks, curves, areas of
arterial branching, and areas of mild stenosis. Turbulent flow is defined as areas of flow reversal, and
usually occurs at areas of severe stenosis when a jet
of blood with high velocity encounters a normal or
increased lumen diameter beyond the stenosis [9].

Flow reversal normally occurs within the carotid
bulb, but is abnormal in any other location. Spectral broadening can also be erroneously depicted
on the spectral tracing when the Doppler gain is set
too high, causing a spurious filling in of the spectral
waveform. Likewise, if the Doppler gain is set too
low, not all of the clinically important velocities
may be displayed.
Because the accurate diagnosis of the degree of
carotid stenosis depends on velocity measurement,
the peak velocity needs to be properly displayed.
The pulse repetition frequency must be set high
enough to be able to accurately measure the arterial
signal, encompassing the entire signal within the
velocity scale chosen. The spectral waveform needs
be appropriately scaled so that the waveform is not
too small, but not so large as to have aliasing.
Imaging protocol
After all of the imaging parameters are optimized,
standard images should be obtained according to
laboratory and accreditation standards. Transverse
gray-scale images of at least the proximal and distal
CCA, bulb, bifurcation, and proximal ICA and ECA
should be obtained with detailed labeling. Longi-
Fig. 3. Low-resistance internalization of ECA flow. (A) Longitudinal spectral Doppler of the ECA shows lowresistance flow with prominent diastolic component (arrow). (B) Spectral Doppler of the ICA has brief highresistance flow (arrows), consistent with a high-grade stenosis or occlusion of the ICA.
Fig. 4. Low-resistance, low PSV flow distal to CCA stenosis. Longitudinal spectral Doppler of the ICA shows a typical
poststenotic parvus-tardus arterial waveform with delayed upstroke and low peak velocity of 13 cm/s. Dist, distal.
tudinal gray-scale images of at least the proximal

and distal CCA, bulb to CCA, proximal ECA extending into the CCA, and proximal ICA extending
into the CCA should be obtained. It is not necessary to record gray-scale images of the vertebral
artery without spectral Doppler, as only a limited
portion of the vertebral artery can be seen between
the vertebral segments.
After the gray-scale images are obtained, optimized color Doppler images with their accompanying spectral waveforms are recorded. Longitudinal
color Doppler and spectral waveforms are obtained
in the proximal and distal CCA; proximal ECA;
proximal, mid, and distal ICA; and the vertebral
artery. The peak systolic velocity (PSV) and end
diastolic velocity (EDV) are measured for each
segment. The location of the distal CCA PSV measurement is important, as CCA PSVs decrease
distally from the heart [10,11]. Accordingly, the
location at which the distal CCA waveform is measured within the CCA is standardized at 2 cm
proximal to the bulb. If any stenoses are found,
spectral waveforms should be obtained at and distal to the stenosis, with corresponding PSV and
EDV measurements.
stroke in systole with a moderate amount of diastolic flow. The entire waveform shows antegrade
flow throughout the cardiac cycle [Fig. 1A].
In contradistinction, the muscles of the face do
not require blood flow in diastole. Therefore, the
flow in the external carotid artery is high-resistance
flow. There is a tall and sharp upstroke in systole,
but the waveform quickly returns to near baseline,
with little flow during diastole [Fig. 1B]. A small
amount of diastolic flow in the ECA may normally
be seen.
The normal CCA waveform is a combination of
the distal high-resistance waveform pattern of the
ECA and the lower-resistance waveform pattern of
the ICA [Fig. 1C]. It is critical to begin with the
CCA waveform in the carotid ultrasound analysis
because the CCA is a combination of the downstream ECA and ICA hemodynamics.
Waveform analysis
Normal carotid ultrasound waveform pattern
analysis
The brain requires oxygen and nutrients throughout systole and diastole of the cardiac cycle. Thus,
the internal carotid artery has a low-resistance
waveform pattern. There is a tall and sharp up-
Fig. 5. CCA stenosis. The PSV ratio is used to characterize the severity of the stenosis in this region. PSVB is at
the CCA stenosis and PSVA is 2 cm proximal to the
stenosis. Prox, proximal.
117
118
Robbin & Lockhart
The last artery to be analyzed in the carotid

artery ultrasound is the vertebral artery. Like the
ICA, the vertebral artery has a sharp upstroke and
moderate low-resistance flow throughout all of
diastole [Fig. 1D].
Abnormal carotid ultrasound waveform

pattern analysis
After a high-quality accurate carotid ultrasound has
been performed using the techniques detailed earlier,
the next steps in interpretation are equally critical.

Following is a summary of the diagnostic possibilities, pearls, and pitfalls that should be considered
when interpreting a carotid ultrasound examination.
Common carotid artery
The first thing that should be assessed when interpreting a carotid ultrasound is the CCA waveform,
because the CCA waveform is a combination of the
ECA and ICA waveforms, and should have a mod-
Fig. 6. CCA stenosis. (A) Longitudinal spectral Doppler of proximal CCA has mildly elevated resistance. (B) Color
Doppler in the CCA demonstrates visible luminal narrowing with color aliasing (arrow). LT, left. (C ) Spectral
Doppler at the same location has high-velocity flow (PSV 627 cm/s) with severe turbulence. (D) Delayed systolic
acceleration (arrow) and low PSV of 43 cm/s (tardus-parvus waveform) are present in the ICA.
erate amount of diastolic flow. If this waveform is

abnormal, downstream or upstream stenoses or occlusions are likely. Thus, the more distal (or proximal) waveforms can be assessed with a high degree
of suspicion for abnormality. The usual waveform
abnormalities of the CCA encountered include an
abnormally high-resistance waveform, an abnormally low-resistance waveform, and either low or
high PSVs. These are considered individually later.
The classic cause of a high-resistance waveform
pattern in the CCA is a high-grade ICA stenosis or
occlusion, termed externalization of the CCA. If a
high-resistance waveform is seen in the CCA, the
diagnosis of severe ICA stenosis or occlusion is suggested, and the distal waveforms should be assessed
to determine whether they support the diagnosis
[Fig. 2]. However, there is an exception to this rule.
If the process is bilateral and the peak PSVs are low,
this waveform pattern could indicate aortic stenosis
or severe cardiac failure, discussed later.
A pitfall occurs when the ECA is providing collateral flow to adjacent arteries, typically occurring
with long-standing internal carotid artery occlusion. When collaterals are fed from the ECA, the
waveform becomes a lower-resistance waveform
with some diastolic flow. In this instance, there
may be an ICA occlusion present, but the CCA
waveform is not such a noticeably high-resistance
pattern because of the low-resistance contribution
of the collateralized ECA [Fig. 3]. When there is an
occlusion of one artery, and differentiation of ICA
versus ECA is not clear, the temporal tap can be a
helpful maneuver, as detailed later.
A normal CCA PSV should be in the range of

approximately 0.6 to 1.0 m/s [12]. If the CCA PSV
is lower than this range, the contralateral CCA
waveform should be examined. If the PSVs are
symmetric, the cause of the low PSV is likely secondary to a low cardiac output, such as that seen
with congestive heart failure. If the contralateral
PSV appears more normal, the CCA with the low
PSV is usually abnormal and needs to be investigated further.
Potential causes for a unilateral low PSV in the
CCA include a proximal or distal high-grade stenosis or occlusion. The first step in differentiating
between these two causes can be made by determining whether the waveform is high-resistance or
low-resistance. A high-grade stenosis can occur in
either proximal CCA. On the right side, the stenosis
or occlusion can also occur in the innominate artery (brachiocephalic artery), typically at its origin off the aortic arch. A high-grade proximal
CCA stenosis will be evidenced by a downstream
low-resistance waveform. Poststenotic turbulence
may be demonstrated if sampled close to the stenosis, and sometimes a tardus parvus waveform
can be seen [Fig. 4]. Occasionally the waveforms
look normal, and the only way a proximal CCA
stenosis can be suspected is by noting the asymmetry in the PSVs. A high-grade stenosis distal to the
point of sampling in the CCA, such as higher in the
neck or in the carotid siphon, will be reflected in
a high-resistance waveform.
If a focal stenosis in the CCA is seen, a ratio needs
to be calculated to determine the degree of stenosis.
119
120
Robbin & Lockhart
If the stenosis cannot be adequately sampled secondary to overlying calcification, the jet should be
assessed distal to the stenosis. It is important to
realize that the jet may dampen out and nearly
disappear within several vessel diameters. Thus,
considerable care needs to be taken in evaluating

the carotid visually with gray-scale and color Doppler. The ratio of the highest PSV at the CCA stenosis divided by the PSV 2 cm proximal to the
stenosis should be calculated as PSVcca at stenosis/
Fig. 7. Bisferious waveform pattern. (A) Longitudinal spectral Doppler of CCA has a double peak (arrows)
appearance of the systolic flow. (B,C ) Similar waveforms in the ipsilateral ICA and contralateral ICA confirm the
diagnosis of aortic regurgitation.
PSVcca prox [Fig. 5]. If the ratio is 2 or more and less

than 2.99, it is consistent with a stenosis of 50%
or more. If the ratio is 3 or more it is consistent
with a stenosis of 75% or more [Fig. 6]. This simple
ratio is also used if there are tandem (two) ipsilateral stenoses.
An unusual waveform not described in the cases
discussed earlier occurs in significant aortic regurgitation. This waveform is termed bisferious and has
a second sharp systolic peak at or above the velocity
of the first systolic peak. The peaks are separated by
a well-defined dicrotic notch. To make the diagnosis, this waveform needs to be found in at least
one vessel (CCA, ICA, or ECA) bilaterally [Fig. 7]
[13]. Another unusual finding is that of the CCA
occlusion with a patent ICA, bifurcation, and ECA.
Reversed flow is present in the ECA, fed by ECA
branches, to keep antegrade flow in the ICA
[Fig. 8]. Of course, the flow in the ICA is lowresistance with a low PSV flow, as it is fed by way
of collaterals [Fig. 9].
Internal carotid artery
It is important to be able to accurately differentiate
the ICA from the ECA. The ICA is typically located
more laterally and posteriorly than the ECA. A way
to remember this is to recognize that the ECA
supplies the muscles of the face, thus the ECA in
general will be located more anteriorly and medially in the carotid sheath. The ECA also tends to
be smaller in diameter in most patients. However,
Fig. 8. Occluded CCA. Reversed flow from the ECA

supplies the ICA and the brain.
this is not a constant relationship. Differentiation

between the ICA and ECA needs to be performed
based on the spectral waveform analysis, not on its
relative location.
Another way to differentiate the ICA from the ECA
is that the ICA has no branches in the neck. The
normal ICA waveform has a moderate amount of
diastolic flow, as the brain requires blood flow
throughout systole and diastole. The ICA is assessed
proximally, mid, and distally. Typically, most of the
atherosclerotic plaque, and therefore stenoses, will
occur from within the carotid bulb and extend into
the proximal ICA. However, an ICA stenosis may be
located in the mid or distal ICA, and even more distally than can be visualized with ultrasound, such as
at the base of the skull. If the stenosis is located
within the visible portions of the ICA, a luminal
narrowing may be seen on gray-scale imaging.
If there is substantial plaque overlying the stenosis, the stenosis itself may not be visible. However,
a stenosis can be inferred from a high-velocity jet
distal to an area of plaque with shadowing. If there
is substantial tortuosity of the artery, particularly
if the ICA dives posteriorly, the PSV can be somewhat elevated without a visible stenosis because of
a small angle theta in the Doppler equation. A tortuous ICA may have an artifactual elevation of
the PSV and should be correlated visually. A significant ICA stenosis that is distal to the visible ICA
will be evidenced by a high-resistance waveform.
As the degree of stenosis becomes hemodynamically significant, (50% to 60% diameter stenosis),
the PSV will begin to rise [Fig. 10]. The PSV will
continue to rise with increasing stenosis until the
degree of stenosis is so high that blood flow
through the stenotic area begins to drop. When
the available pressure is not sufficient to force the
same amount of blood flow through a tight stenosis, the amount of blood flow through the lesion
will decrease. In high-grade stenosis, the PSV will
therefore begin to fall [14].
Thus, a normal PSV can be either normal or
reflect an underlying high-grade stenosis. This potential pitfall occurs when the PSV is the only criterion used for stenosis quantification, and can be
avoided by also analyzing the ICAPSV/CCAPSV ratio.
In the case of high-grade stenosis, the ICA EDV will
also rise, as the blood flow downstream to the stenosis becomes more turbulent and low-resistance.
The ICA EDV should also be assessed to avoid
this pitfall.
The maximum PSV in the ICA (usually at the
proximal ICA) [Fig. 11] is compared with a table
that correlates the degree of stenosis with the PSV
[Table 1] [14]. The maximum PSV in the ICA is also
used to compute the ICAPSV/CCAPSV ratio. The
maximum PSV in the ICA is divided by the PSV
121
122
Robbin & Lockhart
Fig. 9. Reversal of ECA flow in CCA occlusion. (A) Longitudinal color Doppler of the CCA bifurcation shows the ICA
and ECA flow in opposite directions. (B) Spectral Doppler of ECA shows low-resistance reversed flow. (C ) Spectral
Doppler shows the ICA has delayed systolic acceleration and a PSV of 164 cm/s consistent with collateral flow.
(D) Occlusion versus trickle of flow is present in the proximal CCA.
in the CCA (measured at 2 cm proximal to the carotid bulb). This ratio is also used to assess the
degree of ICA stenosis (see Table 1). The ICAPSV/
CCAPSV ratio is an important double check of the
PSV measured in the ICA to determine whether
there is a high-grade stenosis as described earlier.

Additionally, if there is any concern regarding the
degree of ICA stenosis, the EDV of the ICA should
be assessed using the same table. This table is only
used for ICA stenosis, not ECA or CCA stenosis.
VELOCITY (cm/sec)
400
300
PEAK SYSTOLIC VELOCITY
200
100
PEAK DIASTOLIC
VELOCITY
0
0
10
20
30
40
50
60
70
80
90
100
ANGIOGRAPHY - DIAMETER STENOSIS (%)
VELOCITY RATIO
B 15
RATIO =
10
VEL. MAX. STENOSIS

VEL. CCA
5
SYSTOLIC
VELOCITY
RATIO
0
0
10
20
30
40
50
60
70
80
90
100
ANGIOGRAPHY - DIAMETER STENOSIS (%)

Fig. 10. (A) Elevation of PSV relative to degree of ICA stenosis. (B) ICA/CCA PSV ratio versus ICA stenosis. MAX,
maximum; VEL, velocity.
If there is a high-grade stenosis in one internal

carotid artery, there may be a compensatory elevation in the contralateral ICA PSV not caused by a
stenosis. More blood flow may be traveling through
the opposite ICA to compensate for a marked stenosis. This possibility should be considered in the
ultrasound report. Artificially elevated PSV contralateral to a high-grade stenosis is an important consideration, in that a patient may not have significant
bilateral stenoses and not require bilateral endarterectomies. It is possible that after treatment of a
unilateral high-grade ICA stenosis, the contralateral
PSVs may drop to indicate a stenosis of less than
60%. Therefore, endarterectomies are typically performed first on the side with the highest-grade stenosis, and the contralateral side is reassessed with
ultrasound before planning the next surgery [15].
Very high-grade near or complete ICA occlusion is not uncommonly seen. A characteristic audible thump occurs in the distal CCA with spectral
Doppler as the blood flow hits the occluded ICA

segment. A distinctive cranialcaudal piston-like
motion may be seen in the distal CCA if the ICA
occlusion is recent. In this instance, the ipsilateral
ECA may have a low-resistance waveform pattern
because it is providing an alternate pathway for
flow around the obstruction by way of collaterals.
Color and spectral Doppler settings should be optimized for slow flow to attempt to detect a trickle of
flow in the proximal ICA and more distally. If a
small patent residual lumen is present, the patient
is a candidate for endarterectomy. Patients who
have a sonographic diagnosis of ICA occlusion
should go to magnetic resonance angiography or
angiography to be certain that a trickle of flow is
not present. If the ICA is completely occluded, they
are not a candidate for endarterectomy.
The trickle of contrast seen traveling up the carotid on angiography in near occlusion does not
mean that the stenosis is severe all the way up the
123
124
Robbin & Lockhart
Fig. 11. ICA stenosis of 70% to near occlusion. (A) Longitudinal power Doppler of carotid bulb demonstrates only a
narrow stream of flow through a large region of soft plaque (P). (B) Narrow lumen is confirmed with spectral
Doppler PSV elevation of 255 cm/s. Prox, proximal.
ICA to the skull base [Fig. 12A]. Usually the ICA

stenosis is localized, extending from the bulb only
for a few centimeters [Fig. 12B]. The appearance
of a string sign, or narrow column of flow in the
distal ICA on the angiogram, is because the heavier
radiographic contrast does not fill the carotid
lumen distally. The narrow lumen seen on the
angiogram is not because there is a high-grade
stenosis throughout the entire ICA. Ultrasound
has limitations in its ability to differentiate occlusion [Fig. 13] from a trickle of flow.
The ICA should always be compared with the ECA
waveform. An uncommon pitfall occurs in the scenario of an occluded ICA. It is possible to overlook
an occluded ICA and think that the bifurcation
into the ECA and ICA occurs at the first visible
branch of the ECA. This situation can be diagnosed

by realizing that the ICA and ECA waveforms are
identical. Noticing that identical ICA and ECA waveforms are present should be a key toward making
this provisional diagnosis. The sonographer then
needs to re-examine that site for the diagnostic possibility of ICA occlusion. When this mistake occurs,
the sonographer will frequently remark that the
bifurcation looked unusual or that the examination was difficult. Occasionally this scenario will
be seen in a patient who has an extremely low
carotid bifurcation and no ICA occlusion [Fig. 14].
External carotid artery
A normal ECA waveform has a high-resistance
waveform pattern because the muscles of the face
Table 1: Doppler criteria for internal carotid artery diameter stenosis detection developed by the
Society of Radiologists in Ultrasound consensus conference
Normal
<50%
50%69%
70 to near occlusion
Near occlusion
Total occlusion
ICA PSV cm/s
Plaque/diameter
ICA/CCA ratio =
PSVICA/PSVCCA
ICA EDV
cm/s
<125
<125
125230
>230
High, low,
or undetectable
Undetectable
None
<50%
50%
50%
Visible
<2
<2
24
>4
Variable
<40
<40
40100
>100
Variable
Visible, no
detectable lumen
N/A
N/A
Abbreviation: N/A, not applicable.
do not need blood flow throughout diastole. In

general, little evaluation of the ECA distal to its
origin from the CCA needs to be performed. The
two major reasons the ECA needs to be evaluated
are to differentiate the ECA from the ICA and to
identify the source of an audible bruit other
than the ICA. The temporal tap maneuver is useful
for accurate differentiation when it is not immediately obvious which vessel is the ECA versus
the ICA.
The temporal artery is a branch of the ECA
located several finger breadths anterior to the ear
Fig. 12. Trickle of flow. Both of these plaque configurations correlate with the string sign on angiography.
(A) Severe concentric atherosclerotic plaque extends
all the way up the ICA into the distal ICA. (B) Marked
atherosclerotic plaque at the bulb extends for a short
distance into the ICA, causing severe narrowing of
the proximal ICA. Note the more distal ICA is relatively stenosis free.
on the patients cheek. Tapping in this location will

transmit pressure waveforms through the temporal
artery to the ECA [Fig. 15]. If there is no significant
stenosis of either the ECA or the ICA, the temporal
tap may be seen in both arteries. It is debatable
whether or not the temporal tap needs to be performed if no significant stenosis is seen in either the
ICA or ECA. However, most feel that it should be
included in every examination.
In the scenario of a significant stenosis in one
vessel off of the bifurcation, there are two choices
for the stenotic vesseleither the ICA or ECA. The
temporal tap can be useful in differentiating the
ICA from the ECA [Table 2]. In scenario 1, ECA
stenosis, if there is a significant stenosis in the
ECA, the temporal tap will not be transmitted to
the ICA from the ECA because of the significant
ECA stenosis. In this scenario, an alteration in the
waveform caused by the temporal tap will be appreciable in the ECA waveform and not in the ICA. In
the second scenario, ICA stenosis, if there is a significant stenosis in the ICA, the temporal tap waveforms will not be transmitted to the ICA from the
ECA because of the significant ICA stenosis. As in
the first scenario, the temporal tap will only be
appreciated in the ECA waveform. This maneuver
thus allows the definitive differentiation of the ECA
from the ICA once the hemodynamics behind the
technique are understood [16].
Vertebral artery
Evaluation of the vertebral artery is performed
bilaterally as a routine part of the carotid artery
ultrasound. Spectral waveforms and direction of
flow are assessed. Reversal of flow in the left vertebral artery is more common than the right and
indicates a proximal subclavian artery stenosis or
occlusion, usually at the origin from the aortic arch.
Flow reversal in the right vertebral artery indicates
either a proximal subclavian artery stenosis or
occlusion, or a significant brachiocephalic artery
stenosis or occlusion. If the flow reversal occurs
125
126
Robbin & Lockhart
Fig. 13. Occlusion versus trickle of flow. (A) Spectral Doppler of CCA shows high-resistance flow but no stenosis.
Prox, proximal. (B) Gray-scale image of the ICA and bulb shows severe plaque (arrows) with shadowing. (C ) Color
Doppler of the proximal ICA beyond the plaque shows no definite flow. A trickle of flow cannot be excluded
by conventional ultrasound.
in the right vertebral artery, a small footprint transducer should be used to attempt to visualize the
proximal right subclavian artery and a portion of
the brachiocephalic artery, which occasionally can
be well visualized. Lesser degrees of subclavian or
brachiocephalic stenosis result in transient systolic
flow reversal in the vertebral artery [17]. There are
progressive degrees of midsystolic deceleration patterns that ultimately completely reverse and correlate to the degree of proximal stenosis. Kliewer and
colleagues [17] developed an excellent grading
scheme for abnormal vertebral artery waveforms
that are not completely reversed, which is also covered in the subsequent article.
Fig. 14. Severe ICA stenosis and low carotid bifurcation with first ECA branch mistaken for carotid bulb.
(AC) Longitudinal spectral Doppler images that were initially labeled CCA, ECA, and ICA, and interpreted
as no stenosis, have identical waveforms, suggesting that all are part of the ECA system. LT, left. (D) Further
investigation demonstrates a very low bifurcation of the CCA and a focal ICA stenosis is confirmed on spectral
Doppler (PSV 219 cm/s). CCpk, common carotid peak systolic velocity.
127
128
Robbin & Lockhart
Fig. 15. Temporal tap for ECA identification. Repeated

jagged reverberations (arrow) are transmitted from
the temporal artery to the ECA and are displayed in
the spectral waveform.
Fig. 16. Abnormal vertebral artery high-resistance
flow. Spectral Doppler of a left vertebral artery in a
patient who has distal vertebral artery stenosis versus
occlusion shows abnormal high-resistance flow with
absence of any diastolic component.
A high-resistance waveform in the vertebral artery

is consistent with a distal stenosis, occlusion, or
a hypoplastic artery [Fig. 16]. Differentiation between these causes is important, as some centers are
performing vertebral artery angioplasty and stent
placement for significant vertebral artery stenosis.
A potential vertebral artery stenosis by ultrasound
should always be correlated with the patients clinical symptomatology. Dizziness and unsteady walking can be signs of vertebral basilar insufficiency or
impending stroke from vertebral artery stenosis. If
the vertebral artery cannot be detected, the differ-
Table 2: Temporal tap: internal and external

carotid artery findings
Normal
Sig. ICA stenosis
Sig. ECA stenosis
ICA
temporal
tap
ECA
temporal
tap
+/
+
+
+
+, present; , not present; +/, possibly present.
ential diagnosis is vertebral artery occlusion versus

a small or congenitally absent vertebral artery.
A low-resistance, low PSV waveform in the vertebral artery may indicate a more proximal stenosis
[18]. Although the left vertebral artery origin is
difficult to see with ultrasound because of overlying
lung, the right proximal vertebral artery origin off
of the subclavian artery can often be seen with a
small footprint transducer to identify an area of
stenosis [Fig. 17].
Summary
A high-quality ultrasound examination can be readily achieved using the techniques and principles
discussed above. Once the hemodynamics of the
carotid system are understood, the arterial waveforms encountered become understandable and
predictable. A pattern recognition approach to
waveform analysis makes accurate interpretation
129
130
Robbin & Lockhart
Fig. 17. Vertebral artery stenosis. (A) Longitudinal spectral Doppler of right vertebral artery demonstrates lowresistance parvus-tardus waveforms. RT, right. (B) Further investigation discovered a focal stenosis of the
vertebral artery origin with elevated PSV of 272 cm/s. RT, right; VA, vertebral artery.
more likely, with a greater chance of avoiding the

various pitfalls commonly encountered.
[8]
References
[1] Consensus Group. Consensus statement on the
management of patients with asymptomatic
atherosclerotic carotid bifurcation lesions. Int
Angiol 1995;14:517.
[2] Timsit SG. Early clinical differentiation of cerebral infarction from severe atherosclerotic stenosis and cardioembolism. Stroke 1992;23:48691.
[3] North American Symptomatic Carotid Endarterectomy Trial (NASCET) Steering Committee.
North American Symptomatic Carotid Endarterectomy Trial: methods, patient characteristics,
and progress. Stroke 1991;22:71120.
[4] North American Symptomatic Carotid Endarterectomy Trial Collaborators. Beneficial effect of
carotid endarterectomy in symptomatic patients
with high grade carotid stenosis. N Engl J Med
1991;325(7):44553.
[5] Executive Committee for the Asymptomatic Carotid Atherosclerosis Study. Endarterectomy for
asymptomatic carotid artery stenosis. JAMA 1995;
273(18):14218.
[6] Moneta GL, Edwards JM, Papanicolaou G, et al.
Screening for asymptomatic internal carotid
artery stenosis: duplex criteria for discriminating
60% to 99% stenosis. J Vasc Surg 1995;21(6):
98994.
[7] Grant EG, Benson CB, Moneta GL, et al. Carotid
artery stenosis: gray-scale and Doppler US
diagnosisSociety of Radiologists in Ultrasound
[9]
[10]
[11]
[12]
[13]
[14]
Consensus Conference. Radiology 2003;229(2):

3406.
Robbin ML. The utility of contrast in the extracranial carotid ultrasound examination. In:
Goldberg BB, Raichlen JS, Forsbert F, editors.
Ultrasound contrast agents: basic principles and
clinical applications. 2nd edition. London: Martin Dunitz Publishers; 2001. p. 23951.
Zwiebel WJ, Pellerito JS. Basic concepts of
Doppler frequency spectrum analysis and ultrasound blood flow imaging. In: Zwiebel WJ,
Pellerito JS, editors. Introduction to vascular
ultrasonography. 5th edition. Philadelphia: Elsevier Saunders; 2005. p. 6189.
Meyer JI, Khalil RM, Obuchowski NA, et al.
Common carotid artery: variability of Doppler
US velocity measurements. Radiology 1997;
204(2):33941.
Lee VS, Hertzberg BS, Workman MJ, et al.
Variability of Doppler US measurements along
the common carotid artery: effects on estimates
of internal carotid arterial stenosis in patients
with angiographically proved disease. Radiology
2000;214(2):38792.
Zwiebel WJ. Normal findings and technical aspects of carotid sonography. In: Zwiebel WJ,
Pellerito JS, editors. Introduction to vascular
ultrasonography. 5th edition. Philadelphia: Elsevier Saunders; 2005. p. 14354.
Kallman CE, Gosink BB, Gardner DJ. Carotid
duplex sonography: bisferious pulse contour in
patients with aortic valvular disease. AJR Am J
Roentgenol 1991;157(2):4037.
Bluth EI, Stavros AT, Marich KW, et al. Carotid
duplex sonography: a multicenter recommenda-
tion for standardized imaging and Doppler criteria. Radiographics 1988;8(3):487506.

[15] Abou-Zamzam Jr AM, Moneta GL, Edwards JM,
et al. Is a single preoperative duplex scan sufficient for planning bilateral carotid endarterectomy. J Vasc Surg 2000;31(2):2828.
[16] Budorick NE, Rojratanakiat W, OBoyle MK, et al.
Digital tapping of the superficial temporal ar-
tery: significance in carotid duplex sonography.

[17] Kliewer MA, Hertzberg BS, Kim DH, et al.
Vertebral artery Doppler waveform changes
indicating subclavian steal physiology. AJR Am
J Roentgenol 2000;174(3):8159.
[18] Sidhu PS. Ultrasound of the carotid and vertebral
arteries. Br Med Bull 2000;56(2):34666.
131
133
ULTRASOUND
CLINICS
Waveform Analysis of the

Carotid Arteries
Leslie M. Scoutta,*, Felix L. Lin,
&
&
&
Changes in systolic contour

Alteration in diastolic flow patterns
Abnormal waveforms effecting the entire
cardiac cycle
Stroke is the third leading cause of death in the

United States and a leading cause of severe longterm disability. It is estimated that 700,000 strokes
occur yearly in the United States with an associated
mortality of 162,672 and an estimated health care
cost of $56.8 billion [1]. It is believed that up to
20% to 30% of strokes are the result of disease
at the carotid bifurcation, with the mechanism of
injury most likely emboli from friable plaque dislodged by the high-velocity turbulent jet occurring
at the site of the atherosclerotic stenosis [2,3]. Several recently published prospective, multicenter,
randomized, double-blind trials demonstrate convincingly that carotid endarterectomy significantly
reduces the risk of stroke and death in comparison
to medical therapy in patients who have hemodynamically significant stenosis ( 60%70%) of the
internal carotid artery (ICA) [47]. Hence, when
performing a carotid ultrasound (US) examination,
pulse Doppler waveforms are used primarily to
measure peak systolic velocity (PSV) or PSV ratio
(PSVR = PSV in ICA PSV in distal common carotid
artery [CCA]) in order to quantitate the degree of
stenosis. More careful analysis of the shape and
MD
&
&
, Mark Kliewer,
MD
Summary
References
contour of the pulse Doppler waveform, however,

can provide physiologic information regarding
the distal circulation/vascular bed and the proximal
vessels, areas that cannot be interrogated or visualized directly on US examination. The purpose of
this review article is to discuss how changes in the
shape of the Doppler waveform tracing can provide
clues as to proximal or distal cardiovascular disease
and identify more unusual regional disease in the
carotid and vertebral vessels, including iatrogenic
conditions involving the carotid arteries.
Careful attention to technique of the US examination is required to reduce inherent error in
Doppler measurements of PSV and to ensure reproducibility of the Doppler examination. Pulse Doppler tracings should be obtained from longitudinal
images with the sample volume placed centrally
within the vessel or at the brightest spot in an
area of color aliasing. The sample volume should
be of adequate size to encompass the stenotic jet or
central parabolic flow but should not extend to the
vessel wall itself. The Doppler angle should be kept
between 45 and 60 and should be calculated in
relation to the direction of blood flow in the jet or
Diagnostic Radiology, Chief Section of Ultrasound, Yale University School of Medicine, 333 Cedar Street,
New Haven, CT 06520, USA
b
Diagnostic Radiology, Yale University School of Medicine, 333 Cedar Street, New Haven, CT 06520, USA
c
Diagnostic Radiology, University of Wisconsin Medical School, 600 Highland Avenue, Madison, WI
53792, USA
* Corresponding author. Yale University School of Medicine, Department of Diagnostic Radiology, 333 Cedar
Street, New Haven, CT 06520.
E-mail address: scoutt@biomed.med.yale.edu (L.M. Scoutt).
doi:10.1016/j.cult.2005.09.012
134
Scoutt et al
the vessel lumen rather than to the vessel wall. The

wall filter should be kept as low as possible and the
pulse repetition frequency or scale should be set to
maximize the size of the spectral training without
allowing aliasing.
Allowing for slight variability from patient to patient, the internal, external, common carotid, and
vertebral arteries demonstrate characteristic pulse
Doppler waveform patterns reflecting the nature

(oxygen consumption) of the vascular bed supplied. The ICA, which supplies the low-resistance
vascular bed of the brain, typically demonstrates
a low-resistance waveform pattern characterized
by continuous forward diastolic flow reflecting
the low impedence and high oxygen consumption
of the intracranial circulation. The systolic upstroke
Fig. 1. Normal Doppler waveforms of the carotid and vertebral arteries. (A) ICA: the systolic upstroke is sharp
with a thin spectral envelope. The systolic peak is blunted slightly and there is gradual tapering of velocity
from end systole through diastole with continuous forward diastolic flow. (B) ECA: the systolic upstroke is sharp
and the spectral envelope is thin. Note characteristic early diastolic notch (arrow) and reduced diastolic flow in
comparison to the ICA. (C ) ECA: this patient has no diastolic flow in the ECA. The amount of diastolic flow in
the ECA is variable but should be equal in the right and left ECA in a given patient. Transient reversal of flow
in early diastole can be a normal finding in the ECA. (D) CCA: note intermediate amount of diastolic flow
and sharp systolic upstroke. (E ) Vertebral artery: note sharp systolic upstroke and continuous forward diastolic
flow. Widening of the spectral envelope and fill in of the spectral window may be observed, most often the result
of poor visualization and, hence, poor angle of insonation.
Waveform Analysis of the Carotid Arteries
should be sharp and the spectral envelope thin,

although slightly thicker in diastole than in systole
[Fig. 1A]. The systolic peak may be slightly blunter
than the systolic peak of the external carotid artery
(ECA). The ECA, which supplies the higher impedance muscular bed of the face, neck, and scalp, has
a higher resistance waveform pattern, characterized
by an early diastolic notch and little to no diastolic
flow [see Fig. 1B,C]. There can be considerable
variability in the amount of diastolic flow in the
ECA from patient to patient. The amount of diastolic flow in the ECA, however, should be symmetric in the right and left ECA in a given patient.
The systolic upstroke of the ECA should be sharp
and the spectral envelope quite thin. The waveform

of the CCA is a combination of these two patterns,
characterized by an intermediate amount of continuous, forward diastolic flow, a sharp systolic upstroke, and thin spectral envelope [see Fig. 1D].
Flow below the baseline or filling in of the spectral window normally should not be seen in the
extracranial carotid arteries. Where the carotid
bulb widens, however, peripheral reversal of
flow or a helical flow pattern may be observed normally, reflecting a phenomenon, termed boundary layer separation. The vertebral artery usually
is visualized incompletely but should demonstrate
a sharp systolic upstroke and an intermediate
135
136
Scoutt et al
amount of continuous forward diastolic flow

[see Fig. 1E].
Changes in systolic contour

PSV and end-diastolic velocity increase at the site
of a carotid stenosis in the 50% to 96% diameter
reduction range. In addition, the waveform pattern
changes, demonstrating broadening of the spectral
envelope, fill-in of the spectral window, and turbulent flow (ie, flow above and below the baseline).
The rate of systolic upstroke remains brisk. Once
the degree of stenosis exceeds an approximately
96% diameter reduction, however, PSV begins to
fall although spectral broadening and turbulence
persists [8].
The pulsus tardus (delayed) and pulsus parvus
(diminished) waveform most commonly is observed distal to a severe proximal stenosis [9,10].
The tardus parvus waveform develops because of a
delay in systolic acceleration. Initially, loss of the
normal sharp systole upstroke with prolongation of
the acceleration time (length of time from the onset
of systole to the early peak systolic complex) is
observed. The slope of systolic upstroke becomes
more horizontal or flattened rather than sharply
vertical. When pronounced, the tardus parvus wave-
form is characterized by a round, widened systolic

peak of diminished absolute PSV [Fig. 2] [9]. The
tardus parvus waveform phenomenon becomes
more exaggerated the farther away from the stenosis that the vessel is sampled. Hence, if the stenosis
is in the CCA or at the carotid bifurcation, the
waveform in the distal ICA is more abnormal
than in the proximal ICA. Thus, if the carotid
bifurcation cannot be seen well because of shadowing from calcified plaque, for example, it is important to sample the distal ICA as high (cephelad) as
possible to look for this indirect, but important,
sign of a more proximal stenosis. If the stenosis is
unilateral, there is marked asymmetry in the systolic contour of the waveforms of the right and the
left ICAs. If the stenosis is central, such as with
aortic stenosis, the waveforms are affected bilaterally. Thus, the level of the proximal stenosis often
can be identified by the pattern of distribution
of the tardus parvus waveform phenomenon [11].
For example, a tardus parvus waveform is noted in
both CCAs, ICAs, ECAs, and both vertebral arteries in patients who have aortic stenosis [Fig. 3].
If the right inominate artery is stenosed, however,
the tardus parvus waveform is observed only in the
right CCA, ICA, ECA, and vertebral artery and the
waveforms in the left carotid and vertebral arteries
Fig. 2. Tardus parvus waveform. (A) Note delay (tardus) in systolic acceleration indicated by rounding and
widening of the systolic peak and diminished (parvus) PSV in the distal right CCA. PSV is diminished. Identification
of a tardus parvus waveform contour is indicative of a high-grade proximal stenosis in 91% of cases. The stenosis
may be found anywhere from the level of the aortic valve to the carotid bifurcation. This patient had a severe
right inominate stenosis. In comparison, note that the PSV is normal and that the systolic upstroke is sharp in the
left CCA (B). This patient also has reversed flow in the right vertebral artery (C ).
137
138
Scoutt et al
Fig. 3. Aortic stenosis. Tardus parvus waveforms are present in the right (A) and left (B) CCAs, right (C ) and left
(D) ICAs, and right vertebral artery (E) in this 57-year-old patient who has severe aortic stenosis. Note that the
tardus parvus waveform phenomenon is most pronounced in the ICA, which is more distal to the aortic valve than
the CCA.
have a normal, sharp systolic upstroke [11]. Reportedly, the degree of aortic or arterial stenosis must be
somewhat severe to cause a noticeable tardus parvus waveform [12].
Pulsus bisferiens, Latin for beat twice, is the
term used to describe a waveform characterized
by two systolic peaks with an interposed midsystolic retraction [Fig. 4]. Visualization of this waveform suggests the presence of aortic insufficiency
with or without concomitant aortic stenosis or hypertophic obstructive cardiomyopathy [13]. Coexistence of aortic stenosis and aortic regurgitation
magnifies the bisferiens effect. Occasionally, however, such a waveform may be seen in healthy,
athletic, young individuals or in older patients
who have no known aortic pathology, perhaps
reflecting an underlying nonspecific cardiomyopathy. The underlying mechanism of the bisferiens
effect is not well understood. One hypothesis is

that the first systolic peak represents an initial
high volume ejection of blood followed by an
abrupt midsystolic flow deceleration resulting from
aortic regurgitation and that the second systolic
peak reflects a tidal-wave phenomenon caused by
relaxation of the distended aorta. An alternative
explanation is that increased preload of the left ventricle creates a transient Venturi (suction) effect in
the aorta, which results in a midsystolic retraction
in the carotid artery waveform [14].
Occasionally, alternating peak systolic heights on
sequential beats are observed in patients who have
a regular normal sinus rhythm, termed pulsus alternans [Fig. 5]. A variety of clinical conditions, such
as intrinsic myocardial disease (ischemia, cardiomyopathies, or valvular heart disease), hypocalemia, or impairment of venous return (obstruction
139
140
Scoutt et al
Fig. 4. Pulsus bisferiens. Note prominent midsystolic retraction (arrow) in this 93-year-old man who has moderate
aortic regurgitation.
or compression of the inferior vena cava) are

reported to give rise to this characteristic, but rare,
spectral waveform. [15]
Patients who have subclavian steal phenomenon
can present with vertigo, headaches, transient
ischemic attacks, dizziness, syncope, and even visual disturbances or hearing loss, especially after
exercise of the upper extremity [16]. Although reversal of blood flow in the vertebral artery is diagnostic of a complete subclavian steal, the earliest
Fig. 5. Pulsus alternans. Note oscillating PSVs (arrows) in the right ECA in this 47-year-old man who has an
idiopathic dilated cardiomyopathy. The cardiac rhythm remains regular. (From the Rohren EM, Kliewer MA,
Carroll BA, et al. A spectrum of Doppler waveforms in the carotid and vertebral arteries. AJR Am J Roentgenol
sign of subclavian steal physiology is an abrupt,

sharp midsystolic deceleration of flow velocity
resulting in a midsystolic retraction after an extremely narrow initial systolic peak with a sharp
systolic upstroke [17,18]. The midsystolic retraction
can reach varying depths, remaining above or extending below the baseline, depending on the severity of the proximal subclavian stenosis. Kliewer
and colleagues [17] describe four progressively abnormal precursor subclavian steal vertebral artery waveforms categorized by increasing degrees
of decreased forward or reversed midsystolic flow.
In this series, angiography confirms moderately
good correlation of increasing degree of stenosis
(diameter reduction) with increasing waveform
grade [17]. All four waveforms are characterized
by a sharp but transient decrease in mid to late
systolic velocity, resulting in a midsystolic retraction or notch in the Doppler tracing and two systolic peaks and continuous forward diastolic flow.
The first systolic peak is narrow with a sharply
angulated point. The second peak is rounded and
broader with a variable absolute velocity and is
seen most clearly in type 1 and 2 waveform patterns. The waveform type, however, is defined by
the depth of the midsystolic retraction. In a type 1
waveform, the velocity at the nadir of the midsystolic notch remains greater than end-diastolic velocity. For a type 2 waveform, the velocity at the
trough of the midsystolic notch is equal to or
slightly less than end-diastolic velocity. In a type
3 waveform, the velocity at the nadir reaches the
baseline or just below with a rapid recovery of
forward flow in diastole. A second systolic peak
may not always be visible. In the type 4 waveform
pattern, midsystolic flow is frankly reversed, extending below the baseline [Fig. 6]. The type 2
vertebral artery waveform is described a bunny
waveform [Fig. 7]. Similar waveform patterns are
described in the right common carotid artery distal
to a right brachiocephalic artery stenosis [19]. With a
complete subclavian steal, blood flow is reversed
throughout systole, flowing towards the hand rather
than towards the head.
It is postulated that the high-velocity jet of blood
occurring just beyond the subclavian stenosis
courses across or underneath the origin of the vertebral artery during systole and produces a Venturi
effect, resulting in a transient pressure drop in the
subclavian artery causing an abrupt decline in flow
velocity and reduction of forward flow volume
during midsystole in the vertebral artery. The pressure drop is greatest during peak systole as flow
velocity across the stenosis is fastest at that point
in the cardiac cycle. Hence, the corresponding decrease in flow velocity in the vertebral artery also
occurs at peak systole, reflected in the prominent
midsystolic notch observed in the vertebral artery

Doppler tracing. Provocative maneuvers, which increase the oxygen requirement/consumption of
the distal arm and hand, such as exercise of the
hand or inflation of a blood pressure cuff above
systolic arterial pressure for 3 to 5 minutes or until
the hand is mildly symptomatic followed by rapid
release of the cuff and immediate re-evaluation of
the vertebral artery waveform, can convert a presteal waveform to a complete steal [Fig. 8] or can
increase the depth of the midsystolic retraction
substantially [Fig. 9] [16,20]. In the series reported
by Kliewer and colleagues [17], such provocative
maneuvers resulted in upgrading of the waveform
(ie, a greater decline in midsystolic velocity in 90%
of patients).
Alteration in diastolic flow patterns

The normal ECA has a high-resistance waveform
pattern characterized by little or no diastolic flow.
If the ipsilateral ICA is occluded, however, internalization of the ECA waveform may occur. In
this clinical setting, the ECA demonstrates a lowresistance waveform pattern with increased diastolic flow [Fig. 10] [21,22]. The increased diastolic
flow is believed to be the result of the development
of low-resistance collateral pathways between the
ECA and distal ICA via the ophthalmic bed or
between the two ECAs via a superficial vascular
network [22,23]. Occlusion of the CCA can cause
a similar waveform pattern although with reversed
flow in the ipsilateral ECA if the ICA is filled via
retrograde flow through the ECA [22]. In approximately 70% of proximal CCA occlusions, such
reconstituted flow in the ECA and ICA also demonstrates a blunted systolic peak with delayed systolic
acceleration similar to a tardus parvus waveform
[14]. Possibly, the collateral network may filter out
the high-frequency velocities [14]. If only one vessel is seen above the carotid bulb and if it demonstrates a large amount of diastolic flow, it may be
difficult to determine if this vessel represents
the ECA or the ICA. The best way to differentiate
between the ECA and ICA is to search for branch
vessels, which only should be seen arising from the
ECA. In the absence of branch vessels, a temporal
tapping maneuver can be performed. Tapping over
the temporal artery lateral to the orbit should produce small, regular, spike-like percussion waves
in the ECA [Fig. 11], best seen during diastole.
Although these percussion waves sometimes can
be transmitted around the bulb and, therefore,
seen in the ICA, the deflections typically are more
pronounced (ie sharper, better defined, and of
higher amplitude in the ECA) [24,25].
141
142
Scoutt et al
Reversed diastolic flow is an abnormal finding

in the carotid arteries. Aortic regurgitation, when
at least moderate to severe, often causes reversed
early diastolic flow in both CCAs in association
with elevation of PSV and a sharp systolic upstroke [Fig. 12]. This waveform pattern is the
Doppler correlate of the water hammer pulse
on physical examination. Depending on the severity of the aortic regurgitation, the reversal of flow
may be limited to early diastole with normalization of forward flow in end diastole, may be
associated with absent end-diastolic flow, or may
persist throughout diastole. The diastolic flow
reversal is bilateral and more pronounced the
closer to the aortic valve (ie, the more proximal)
the vessel is sampled. Hence, the diastolic flow
pattern normalizes in the distal CCA and ICA
[Fig. 13].
Fig. 7. Subclavian steal. (A) Note midsystolic retraction (arrow) in the left vertebral artery in this 54-year-old
woman who has an asymptomatic left subclavian stenosis. (B) Schematic comparing the contour of the spectral
tracing to the profile of a bunny. Hence, the derivation of the term bunny waveform. (From the Rohren EM,
Kliewer MA, Carroll BA, et al. A spectrum of Doppler waveforms in the carotid and vertebral arteries. AJR Am
J Roentgenol 2003:181;1695704; with permission.)
Loss of end-diastolic flow or reversal of diastolic

flow with a normal to low PSV in the CCA or ICA
suggests increased distal peripheral vascular resistance. This appearance is described as a knocking
waveform [26,27]. When unilateral, this waveform
is the result of a distal high-grade stenosis or occlusion, the result of either of atherosclerosis or carotid dissection [Fig. 14] [26,27]. Unlike in the case
of aortic regurgitation, the reversal of diastolic flow
becomes more pronounced the closer to the occlu-
Fig. 6. Progression of presubclavian steal waveform in the vertebral artery. Note synchronously recorded
EKG tracing below Doppler waveform tracing obtained in the midvertebral artery. (A) Type 1 waveform in a
63-year-old woman who has asymptomatic carotid bruit demonstrates a sharp midsystolic decrease in velocity
producing a notch (arrow) creating two systolic peaks. The first peak is narrow and rises to an acute angle.
The second has a rounded contour. Systolic velocity at the nadir of the notch is greater than end-diastolic velocity
Note separate early diastolic notch (arrow head). Flow is antegrade throughout the cardiac cycle. (B) Schematic
drawing of the type 1 waveform (A) correlating the waveform fluctuations with the cardiac cycle. (C ) Type 2
waveform obtained from a 53-year-old woman with coronary artery disease. Note that the midsystolic notch is
deeper with the velocity at the nadir of the notch slightly less than end-diastolic velocity. In the type 2 waveform,
the second systolic peak tends to be smaller and broader than the corresponding peak in the type 1 waveform.
(D) Schematic drawing of the type 2 waveform (C ) correlating the waveform contour with the cardiac cycle.
(E) Type 3 waveform. This 54-year-old man presented with angina. In the type 3 waveform, the nadir of the
midsystolic cleft reaches or minimally extends below the baseline. Rapid recovery of forward flow, however,
occurs before diastole. (F ) Schematic drawing of vertebral artery Doppler tracing [Fig. 6E] correlating the contour
of the type 3 waveform with the cardiac cycle. (G ) Type 4 waveform observed in an 81-year-old man who has
coronary artery disease. Note that the nadir of the midsystolic cleft falls well below baseline signifying greater
reversal of flow during most of systole. Forward flow, however, is restored in diastole. (H) Schematic drawing
of the type 4 waveform (G) correlating the waveform profile with the cardiac cycle. (From the Kliewer MA,
Hertzberg BS, Kim DH, et al. Vertebral artery Doppler waveform changes indicating subclavian steal physiology.
AJR Am J Roentgenol 2000;174;8159; with permission.)
143
144
Scoutt et al
Fig. 8. Subclavian steal: conversion of a presteal waveform to a complete steal following provocative maneuvers.
(A) Waveform of the left vertebral artery at rest in an 83-year-old woman who has a known left subclavian
stenosis demonstrates a prominent midsystolic reversal of flow (arrow). (B) After inflation of a blood pressure cuff
on the left arm and rapid deflation, there is conversion of the presteal waveform to a complete steal, with reversal
of direction of blood flow throughout the cardiac cycle. The blood pressure cuff maneuver induces reactive
hyperermia in the distal arm and increases blood flow across the subclavian stenosis, resulting in a complementary
pressure drop and change in direction of blood flow in the ipsilateral vertebral artery towards the now lowerpressure subclavian origin. (From the Rohren EM, Kliewer MA, Carroll BA, et al. A spectrum of Doppler waveforms in the carotid and vertebral arteries. AJR Am J Roentgenol 2003;181:1695704; with permission.)
sion/stenosis or cause of increased peripheral vascular resistance that the vessel is sampled. There is
a spectrum of waveform patterns progressing from
reversal of early diastolic flow to absence of diastolic flow, which reflects the severity of the distal
disease and the proximity to the obstructing lesion.
Ultimately, PSV drops just proximal to a vascular

occlusion, resulting in a blip of extremely low
velocity systolic flow with completely absent or
reversed early diastolic flow. It is possible to distinguish a carotid dissection from atherosclerotic
occlusive disease as a cause of such a waveform
Fig. 9. Vertebral artery waveform changes induced by provocative maneuver (reactive hyperemia maneuver).
(A) Doppler tracing of the vertebral artery obtained at rest in a 67-year-old woman status post stroke. Note
bunny profile characteristic of a type 2 waveform pattern. Although there is continuous antegrade flow
throughout the cardiac cycle, there is a prominent midsystolic notch. The velocity at the trough of the notch is
slightly below end-diastolic velocity. The first systolic peak is sharp and narrow; the midsystolic decrease in velocity
is abrupt and rapid; and the second systolic peak is broader and slightly rounded. Note synchronously obtained
EKG below Doppler tracing. (B) After inflation of a blood pressure cuff to greater than systolic arterial pressure on
the ipsilateral arm for 3 to 5 minutes and rapid deflation of the cuff, a repeat tracing from the vertebral artery
demonstrates conversion of the type 2 waveform to a type 4 waveform. Note transient reversal of flow (below
the baseline). (From Kliewer MA, Hertzberg BS, Kim DH, et al. Vertebral artery Doppler waveform changes
indicating subclavian steal physiology. AJR Am J Roentgenol 2000;174;8159; with permission).
by identification of the echogenic intraluminal dissection flap on gray-scale imaging or by the observation that in patients who have carotid
dissections, the vessel is narrowed and the waveform is abnormal over a relatively long segment
of the ICA and little atherosclerotic plaque is
observed. Increased intracranial pressure, diffuse
vasospasm, or bilateral severe atherosclerosis can
cause a similar pattern bilaterally in the carotid
arteries [Fig. 15].
Abnormal waveforms effecting the entire

cardiac cycle
Complex abnormal flow patterns affecting systole
and diastole also can be observed. Such diffusely
abnormal waveforms often are the result of iatrogenic or traumatic conditions, such as intra-aortic
balloon pumps, pseudoaneurysms (PSAs), arteriovenous fistulae (AVF), or carotid dissections.

In patients who have an intra-aortic balloon
pump, inflation of the balloon during systole
causes a second peak of forward flow. Deflation
of the balloon results in a transient decrease of
flow at end diastole [Fig. 16]. The intra-aortic balloon pump may not be inflated during every cardiac cycle (a 1:1 ratio). If it is set at a 1:2 ratio, it
inflates with every other systole and the second
systole peak is observed every second cardiac cycle
with an intervening normal systolic peak complex. Although placement of an intra-aortic balloon
pump results in an increase in total volume of
forward blood flow during systole, PSV typically
is decreased, which makes grading of carotid stenosis using only PSV criteria unreliable.
Carotid PSAs occur most commonly after inadvertent needle sticks to the carotid artery but also
145
146
Scoutt et al
Fig. 10. Internalization of the ECA waveformresulting from complete occlusion of the ipsilateral ICA. (A) Color
Doppler image demonstrates a single patent vessel above the carotid bifurcation. Is this the ECA or the ICA?
(B) Pulse Doppler waveform demonstrates a moderate to large amount of continuous forward diastolic flow
(ie a typical ICA waveform). Temporal tapping (arrow) confirms, however, that the patent vessel is the ECA and
the ICA is occluded.
may occur after other types of penetrating or blunt

trauma, carotid endarterectomy, carotid graft placement, or invasion of the carotid artery by a malignant mass. If the lumen of the neck of the PSA
is narrow, Doppler interrogation of the neck reveals a to-and-fro pattern characterized by flow
heading towards the PSA during systole and reversed flow heading towards the carotid artery
during diastole [Fig. 17]. If the neck is wide, this

to-and-fro flow pattern may not be observed and
the waveform is randomly irregular and turbulent. PSAs are recognized easily on gray-scale imaging as an outpouching arising from the carotid
artery, which fills in with color on color flow imaging. A variable amount of intraluminal thrombus
may be noted.
Fig. 11. Temporal tapping in the ECA. Note small regular deflections (TT). The frequency corresponds to the rate
of temporal tapping. The deflections are seen best during diastole.
AVF most often are congenital or traumatic in

origin and more rarely can occur secondary to malignant invasion. As in AVFs or arteriovenous malformations elsewhere, the feeding arterial vessel
demonstrates a low-resistance waveform pattern
characterized by increased PSV and increased continuous forward diastolic flow. The arterial waveform distal to the AVF, however, has a normal
waveform pattern. The draining vein should demonstrate a pulsatile arterialized waveform with a
normal venous flow pattern noted in the proximal
vein. A soft tissue color bruit also may be observed
as a result of tissue reverberation from the highvelocity flow [Fig. 18].
Dissections of the ICA occur most often in the
setting of trauma, especially seat belt injury or
repetitive trauma. The trauma can be minor but
still compress the ICA significantly on the spine.
Occasionally, dissections may occur spontaneously
and isolated to the carotid arteries in patients who
have predisposing conditions, such as Marfan syndrome, Ehlers-Danlos syndrome, fibromuscular
dysplasia, cystic medial necrosis, hypertension, or
drug abuse [28]. CCA dissections also may develop
from direct extension of an aortic dissection. Acute
onset of headache or Horners syndrome is the
most common presenting symptom. Although
rare, dissection of the ICA is the most common
cause of stroke in young patients [28]. Most ICA
dissections occur at the level of the carotid bifurcation. The echogenic intraluminal flap may be
visualized on longitudinal or transverse images

[Fig. 19A], and color flow imaging may reveal
flow in both lumens. The Doppler waveform pattern is highly variable, dependent on whether or
not the true or false lumen is sampled, the diameter
of the lumen, the length of the dissection, and
involvement of the proximal aorta. Often the waveform pattern is extremely bizarre in configuration:
low PSV velocity with a highly irregular waveform
contour with many spikes or fluttering with reversed or bidirectional of flow, such that it may
be difficult to distinguish systole from diastole
[Fig. 20] [2629]. Sometimes, however, a carotid
dissection results in an intramural hematoma, causing a long-segment tapering of the ICA without a
break in the intima and creation of a false lumen
[see Fig. 19B]. The residual lumen may be narrowed markedly, creating a string sign. Thrombosis of the false lumen creates a similar US
appearance on gray-scale or color flow imaging.
In such cases, a very pulsatile, sometimes dampened, waveform with little or no diastolic flow is
noted on pulse Doppler examination, particularly if
sampling is just proximal to a long-segment stenosis [27,28]. The waveform may be indistinguishable
from a stenosis except that typically it extends over
a much longer segment and often no plaque is
visualized. Trauma also is the leading cause of
vertebral artery dissections [27]. Vertebral artery
dissections tend to occur higher in the neck at
the level of the craniocerebral junction, however,
147
148
Scoutt et al
Fig. 12. Water hammer pulse. Note symmetrically reversed early and end-diastolic flow in the right (A) and left
(B) CCAs, indicating a widened pulse pressure. PSV is elevated and systolic upstroke is sharp with a rapid
deceleration in this patient who has severe aortic regurgitation.
Fig. 13. Aortic regurgitation. Reversed diastolic flow is noted in the right (A) and left (B) CCAs in this 63-year-old
patient who has moderate aortic regurgitation. Diastolic flow is normal in the more distal right (C ) and left
(D) ICAs, however. Note midsystolic retraction (arrow), the bisferiens phenomenon, also likely a result of the
aortic regurgitation.
149
150
Scoutt et al
Fig. 14. Knocking waveform. (A) Doppler tracing from the right CCA demonstrates a high-resistance waveform
pattern with no end-diastolic flow. This observation should raise suspicion for a distal occlusion or high-grade
stenosis. (B) PSV is diminished markedly with reversed/absent diastolic flow, a knocking waveform pattern in the
right ICA immediately proximal to the level of the high-grade ICA stenosis. (C ) Note near occlusion of the ICA with
distal string sign on this sagittal power Doppler image.
151
152
Scoutt et al
Fig. 15. A high-resistance waveform pattern is noted bilaterally in both CCAs (A,B) and right ICA (C ) in this patient
who has increased intracranial pressure resulting from a massive stroke. Note evidence of early herniation (arrow)
on CT scan of the brain (D).
153
154
Scoutt et al
Fig. 16. Intra-aortic balloon pump. Note second systolic peak of forward flow (arrow) resulting from inflation of
the intra-aortic balloon pump in this 58-year-old man who has severe ischemic cardiomyopathy. Deflation of the
balloon results in a transient reversal of flow (arrowhead) at the end of diastole.
Fig. 17. Carotid PSA. (A) Sagittal power Doppler image demonstrates a large PSA arising from the CCA just below
the bifurcation. (B) Transverse gray-scale image demonstrates a narrow neck (calipers) pseudoaneurysm (psa).
(C) Angiogram demonstrating carotid PSA. (D) Doppler waveform demonstrating to-and- fro flow with an irregular contour in the neck of the PSA.
155
156
Scoutt et al
Fig. 18. Carotid AVF. (A) Color Doppler image reveals a vertebrojugular AVF in a 19-year-old male status post
gunshot wound to the neck. (B) Waveform of the right vertebral artery demonstrates increased systolic and
diastolic velocities. Pulsatile, high-velocity flow is noted in the draining vein. (From Rohren EM, Kliewer MA,
Carroll BA, Hertzberg BS. A spectrum of doppler waveforms in the carotid and vertebral arteries. AJR Am J
Roentgenol 2003;181;1695704; with permission.)
Fig. 19. Carotid dissection. (A) Note thin echogenic intraluminal flap on sagittal gray-scale image in this patient
who has a carotid dissection. (B) This 22-year-old patient presented with a left neck hematoma and pain after
a motor vehicle accident. Color Doppler image demonstrates marked hypoechoic circumferential thickening
of the wall of the CCA consistent with an intramural hematoma/dissection. Note that the vessel wall is quite
narrowed over a long segment.
157
158
Scoutt et al
Fig. 20. Carotid dissection. (A) Gray-scale image reveals a thick echogenic flap in this 40-year-old patient who has a
carotid dissection. (B) Pulse Doppler image demonstrates a markedly irregular waveform.
and are, therefore, more difficult to diagnose by

US examination.
Summary
The normal Doppler waveform patterns of the carotid and vertebral arteries reflect the nature of
oxygen consumption of the vascular bed that they
supply. Careful attention to changes in the arterial
waveforms can provide important clues regarding
associated pathology, such as proximal cardiac
or great vessel pathology, distal intracranial disease, and nonatherosclerotic carotid disease. Careful analysis and judicious use of provocative
maneuvers also can help to distinguish accurately
between the ICA and ECA when only one vessel is
seen above the carotid bifurcation and to identify
a subclavian stenosis by converting a presteal to

a complete vertebral steal. In addition, certain iatrogenic conditions can be recognized by their
characteristic Doppler waveforms and gray-scale
features. Lastly, it is extremely important to recognize the effect of such conditions, especially proximal and distal disease, which cannot be visualized
on direct US examination, on the Doppler waveform to avoid pitfalls in grading carotid stenosis
using Doppler velocity criteria.
References
[1] American Heart Association. Heart disease and
stroke statistics2005 update. Dallas (TX): American Heart Association; 2005.
[2] Timsit SG, Sacco RL, Mohr JP, et al. Early clinical
differentiation of cerebral infarction from severe
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
atherosclerotic stenosis and cardioembolism.

Stroke 1992;23:48691.
Hademenos GJ, Massoud TF. Biophysical mechanisms or stroke. Stroke 1997;28:206777.
North American Symptomatic Carotid Endartectomy Trial Collaborators. Beneficial effect of carotid endarterectomy in symptomatic patients
with high grade stenosis. N Engl J Med 1991;325:
44553.
Barnett H, Taylor D, Eliasaw M, et al. Benefit of
carotid endarterectomy in patients with symptomatic moderate or severe stenosis. N Engl J Med
1998;339:141525.
Executive Committee for Asymptomatic Carotid
Atherosclerosis Study. Endarterectomy for asymptomatic carotid artery stenosis. JAMA 1995;273:
14218.
European Carotid Surgery Trialists Collaborative
Group. MRC European Carotid Surgery Trial: interim results for symptomatic patients with severe (7099%) or with mild (029%) carotid
stenosis. Lancet 1991;337:123543.
Spencer MP, Reid JM. Quantitation of carotid stenosis with continuous wave (CW) Doppler ultrasound. Stroke 1979;10:32630.
Kotval PS. Doppler waveform parvus and tardus:
a sign of proximal flow obstruction. J Ultrasound
Med 1989;8:43540.
Bude RO, Rubin JM, Platt JF, et al. Pulsus tardus:
its cause and potential limitations in detection or
arterial stenosis. Radiology 1994;190:77984.
Horrow MM, Stassi J. Sonography of the vertebral arteries: a window to disease of the proximal great vessels. AJR Am J Roentgenol 2001;
177:539.
OBoyle MK, Vibhakar N, Chung J, et al. Duplex
sonography of the carotid arteries in patients
with isolated aortic stenosis; imaging findings
and relation to severity of stenosis. AJR Am
J Roentgenol 1996;166:197202.
Kallman CE, Gosink BB, Gardner DJ. Carotid
duplex sonography: bisferious pulse contour in
patients with aortic valvular disease. AJR Am J
Roentgenol 1991;157:4037.
Rohren EM, Kliewer MA, Carroll BA, et al. A
spectrum of Doppler waveforms in the carotid
and vertebral arteries. AJR Am J Roentgenol 2003;
181:1695704.
Cohn KE, Sandler H, Hancock EW. Mechanisms
of pulsus alternans. Circulation 1967;36:37280.
[16] Gosselin G, Walker PM. Subclavian steal syndrome; existence, clinical features, diagnosis and
management. Semin Vasc Surg 1996;9:937.
[17] Kliewer MA, Hertzberg BS, Kim DH, et al. Vertebral artery Doppler waveform changes indicating
subclavian steal physiology. AJR Am J Roentgenol 2000;174:8159.
[18] Horrow MM, Stassi J. Sonography of the vertebral arteries: a window to disease of the proximal great vessels. AJR Am J Roentgenol 2001;
177:539.
[19] Grant EG, Elsaden S, Modrazo B, et al. Inominate
artery occlusive disease: sonographic findings.
AJR Am J Roentgenol, in press.
[20] Delaney CP, Couse NF, Mehigan D, et al. Investigation and management of subclavian steal syndrome. Br J Surg 1994;81:10935.
[21] AbuRahma AF, Pollack JA, Robinson PA, et al.
The reliability of color duplex ultrasound in diagnosing total carotid artery occlusion. Am J Surg
1997;174:1857.
[22] Verbeeck NY, Vazquez Rodriguez C. Patent internal and external carotid arteries beyond an
occluded common carotid artery: report a case
diagnosed by color Doppler. JBR-BTR 1999;82:
21921.
[23] Macchi C, Catini C. The anatomy and clinical
significance of the collateral circulation between
the internal and external carotid arteries through
the ophthalmic artery. Ital J Anat Embryol 1993;
98:239.
[24] Kliewer MA, Freed KS, Hertzberg BS, et al. Temporal artery tap: usefulness and limitations in
carotid sonography. Radiology 1996;201:4814.
[25] Budorick NE, Rojratanakiat W, OBoyel MK, et al.
Digital tapping of the superficial temporal artery: Significance ion carotid duplex sonography.
[26] Hennerici M, Neuerburg-Heusler D, editors. Vascular diagnosis with ultrasound. Stuttgart (Germany): Georg Thieme Verlag; 1998. p. 723.
[27] Hennerici M, Steinke W, Rautonberg W. Highresistance Doppler flow pattern in extracranial
carotid dissection. Arch Neurol 1989;46:6702.
[28] Khaw KT, Griffiths PD. Non-invasive imaging of
the cervical carotis and vertebral arteries. Imaging
2001;13:37690.
[29] Gardner DJ, Gosink BB, Kallman CE. Internal
carotid artery dissections: Duplex US imaging.
159
161
ULTRASOUND
CLINICS
Ultrasound of the
Intracranial Arteries
Susan L. Voci,
&
&
&
&
&
&
&
&
&
&
&
MD*,
Nancy Carson,
Transcranial Doppler and transcranial

color-coded duplex sonography
equipment
The transcranial Doppler examination
Transtemporal window
Suboccipital window
Transorbital window
Submandibular window
Documentation
TCD limitations
The transcranial color-coded duplex
sonography examination
Transcranial Doppler or transcranial
color-coded duplex sonography?
Clinical applications
Detection and monitoring of cerebral
vasospasm
In 1954, Leksell introduced echoencephalography to evaluate the midline and paramedian structures [1]. In 1982, transcranial Doppler sonography
was developed by Aaslid and colleagues [2]. Using
a 2 MHz Doppler device, they measured blood
mean flow velocities of the arteries of the circle
of Willis, which was a major advance to detect
and monitor noninvasively vasospasm related to
subarachnoid hemorrhage. In 1990, transcranial
color-coded duplex sonography (TCCS) provided
clinical-imaging capability in addition to transcranial Doppler (TCD). TCCS combines B-mode imaging with color flow Doppler, permitting direct
visualization of the basal cerebral arteries in addition to their flow direction so angle-corrected veloci-
&
&
&
&
&
&
&
&
&
&
MBA, RDMS, RVT
Detection of intracranial stenosis and

occlusion
Screening for stroke prevention in children
with sickle cell disease
Vasomotor reactivity testing
Detection of microembolic signals
Intraoperative monitoring
Brain death
Ultrasound contrast agents
New developments
Ultrasound enhanced thrombolysis
Summary
References
ties can be obtained. Ultrasound contrast agents and

other techniques, such as three-dimensional ultrasound, power Doppler, and harmonic imaging are
used with TCCS for more diagnostic accuracy [3].
This article discusses the technique for performing
TCD and TCCS and clinical applications and new
developments for intracranial vascular ultrasound.
Indications for TCD and TCCS include but are
not limited to (1) detection of intracranial stenosis and occlusion, (2) the course of vasospasm in
patients who develop subarachnoid hemorrhage,
(3) the detection of cerebral emboli, (4) prediction of stroke associated with sickle cell disease,
(5) evaluation of the vertebrobasilar system, (6) assessment of collateral pathways, (7) evaluation for
Department of Radiology, University of Rochester Medical Center, 601 Elmwood Avenue, Rochester,
NY 14642, USA
E-mail address: susan_voci@urmc.rochester.edu (S.L. Voci).
doi:10.1016/j.cult.2005.09.007
162
Voci & Carson
The transcranial Doppler examination
Fig. 1. Intracraninal arteries axial view at level of circle

of Willis: MCA, ACA, PCA, BA, and VA. (From Carson
NL, Voci SL. Transcranial Doppler. In: Dogra V, Rubens
DJ, editors. Ultrasound secrets. Philadelphia: Hanley &
Belfus; 2004; with permission.)
brain death, and (8) detection of possible emboli

from a patent foramen ovale. Examination protocols and the vessels interrogated vary based on the
indication for the examination.
The complete TCD examination obtains spectral

Doppler signals from the following arteries; middle
cerebral artery (MCA), anterior cerebral artery
(ACA), posterior cerebral artery (PCA), distal internal cerebral artery (dICA), basilar artery (BA), vertebral artery (VA), extracranial internal carotid
artery (eICA), and for some indications, the ophthalmic artery (OA) and carotid siphon [Fig. 1].
These arteries are known collectively as the basal
cerebral arteries of the circle of Willis [5]. To obtain
the Doppler information from the entire set of
vessels, the examiner must use four different windows: the transtemporal window, the transorbital
window, the suboccipital window, and the submandibular window [Fig. 2]. Vessel identification
during the TCD examination depends on several
factors: the depth of focus; window used (ie, submandibular, occipital, and so forth); the direction
of the reflected Doppler signal (toward or away
from the transducer); flow velocity (peak systolic
velocity [PSV], mean velocity); the traceability of
the vessel, and the direction of probe angulation
(ie, anterior, posterior, caudal) [5].
Transtemporal window
The transtemporal window location varies with
each patient and the ability to penetrate the temporal bone varies with age, sex, and ethnicity. The
Transcranial Doppler and transcranial

color-coded duplex sonography equipment
Transcranial imaging equipment must have an
excellent signal-to-noise ratio because most of the
examination is performed through the highly reflective skull. The machines specifically designed
for TCD have a lower bandwidth, which results in
a larger and less-defined sample volume [4]. The
TCD examination is performed with a 2-MHz, focused, bidirectional pulsed Doppler transducer that
displays only a spectral waveform. The TCCS examination uses a duplex ultrasound machine with a
low frequency (ie, 1.82.6 MHz) phased array sector transducer. The image consists of a gray-scale
picture with the vessels displayed as an overlay of
color (or power) Doppler. In addition to inherent
machine settings, such as adjustable Doppler gate
size and depth, the ultrasound beam must be able
to focus at a depth of 40 to 60 mm. The display
should be configured so the peak systolic, end
diastolic, and time-averaged velocities can be displayed in real-time [4].
Fig. 2. Acoustic windows for TCD. (From Carson NL,

Voci SL. Transcranial Doppler. In: Dogra V, Rubens DJ,
editors. Ultrasound secrets. Philadelphia: Hanley &
Belfus; 2004; with permission.)
Ultrasound of Intracranial Arteries
Fig. 3. Normal MCA. Using transtemporal window, depth is 56 mm, and mean velocity is 74 cm/s. Flow is toward transducer. (From Carson NL, Voci SL. Transcranial Doppler. In: Dogra V, Rubens DJ, editors. Ultrasound
secrets. Philadelphia: Hanley & Belfus; 2004; with permission.)
thinnest part of the bone tends to be above the

zygomatic arch. Rotating the transducer in a circular manner above the arch is helpful in finding the
best window. Once the window is located, small
adjustments in angulation are necessary to optimize the signal as the vessels are followed. The
vessels interrogated through the transtemporal
window include the MCA, the ACA, the PCA, and
the dICA. Several factors need to be checked if difficulty occurs in finding the vessels through this
window. First, the power should be set at maximum and the Doppler frequency chosen must be
the lowest available. Other factors and actions to
consider include: increasing the scale (or PRF),
opening the Doppler gate, increasing the gain,
and increasing the amplitude if possible.
The MCA can be found at a depth as shallow as
40 to 52 mm and can be followed in to a depth of
approximately 56 to 62 mm until it joins with the
ACA at the level of the dICA bifurcation. The M1
segment, which is the most distal segment closest
to the outer skull, is found at approximately 38 to
42 mm, depending on diameter of the head. Flow
in the MCA is toward the transducer, thus displayed
above the baseline [Fig. 3]. At the level of the

bifurcation of the dICA (5464 mm), MCA flow
is above the baseline and ACA flow is depicted
below the baseline signifying flow away from the
transducer [Fig. 4]. From the bifurcation at 54 to
64 mm, the ACA can be followed anteriorly. Flow
continues to be away from the transducer; it is
often necessary to angle the transducer slightly
superiorly or anteriorly to follow the ACA. The
ACA may be traced to depths of approximately
68 to 78 mm. The examiner may notice the MCA
signal disappearing from above the baseline as the
ACA is followed distally [Fig. 5].
Doppler signals are obtained from the dICA by
returning to the level of the bifurcation, 54 to
64 mm. Here, the transducer should be angled
slightly inferiorly; the signal from the dICA is toward the transducer and lower in amplitude than
that of the MCA or ACA. The dICA can be tortuous,
thus the Doppler signal may be displayed on both
sides of the baseline [Fig. 6].
The PCA is found by returning the sample
volume to the level of the bifurcation then angling
posterior. Flow in the PCA is displayed above the
Fig. 4. Normal bifurcation of ICA at depth of 64 mm. MCA is above baseline, and ACA is below baseline. Mean
velocity of MCA is 68 cm/s and flow is toward transducer. ACA has mean velocity of 53 cm/s, and flow is away
from transducer. (From Carson NL, Voci SL. Transcranial Doppler. In: Dogra V, Rubens DJ, editors. Ultrasound
163
164
Voci & Carson
Fig. 5. Normal ACA. Depth is 66 mm, mean velocity is 46 cm/s, and flow is away from transducer.
baseline. The PCA is distinguished from the MCA

(also above the baseline) by its lower velocity and
its posterior location [Fig. 7]. The PCA is followed
posteriorly 68 to 72 mm and the flow becomes
bidirectional. At this point, the PCA has been traced
to the top of the basilar artery (TOB), at its point of
bifurcation [Fig. 8]. The signal below the baseline
represents the contralateral PCA. While interrogating the intracranial vessels without color Doppler,
only the P-1 segment of the PCA is sampled be-
cause of the sharp angle of the P-2 segment as it

wraps around the midbrain.
The TCD examiner should keep in mind the normal depths and velocity relationships of the vessels.
The MCA has a higher mean velocity than the ACA,
and the MCA and ACA have higher mean velocities
than the PCA and dICA. The examiner may need to
go back and forth (shallow to deep, deep to shallow) several times to determine the identity of the
vessels before recording any waveforms.
Fig. 6. Distal ICA. Flow is toward transducer, depth is 58 mm, and mean velocity is 40 cm/s.
Fig. 7. Normal PCA. Transducer is angled slightly posterior, and depth is 68 mm. Flow is toward transducer and
mean velocity is 34 cm/s. (From Carson NL, Voci SL. Transcranial Doppler. In: Dogra V, Rubens DJ, editors.
Ultrasound secrets. Philadelphia: Hanley & Belfus; 2004; with permission.)
Occasionally, a patient has an adequate transtemporal window on one side and not the other. In this
case, often the contralateral MCA and ACA can be
interrogated by scanning from the side with the
optimum window. The examiner increases the
depth of the Doppler gate until the ipsilateral
MCA disappears and the midline has been crossed.
At this point, if the penetration is adequate, the
contralateral side begins to appear. The MCA is
depicted below the baseline, indicating flow away
from the transducer, and the ACA is above the
baseline, indicating flow toward the transducer.
Using calipers and measuring the diameter of the
head at the level of the zygomatic arches can be
helpful to determine where the midline is.
Suboccipital window
The intracranial VAs and the BA are evaluated using
the suboccipital approach. The patient is seated
with the head tipped forward slightly. The transducer is placed at the nape of the neck and the
foramen magnum is used as the window. If the patient cannot sit up, reclining in a sideways position
is sufficient as long as the spine is kept in a horizontal plane. The chin is still tipped forward.
The VA signals are obtained at a depth of about
60 to 85 mm. The transducer must be angled
slightly to the right or left of midline to follow
the vessels. The net direction of flow is away from
the transducer, but occasionally the signal may be
Fig. 8. PCAs at top of basilar artery (TOB). Spectral tracing above baseline represents ipsilateral PCA, tracing below
baseline is contra lateral PCA.
165
166
Voci & Carson
Fig. 9. Right VA from suboccipital window. Flow is away from transducer, depth is 66 mm, and mean velocity
is 20 cm/s.
bidirectional as the arteries course through the foramen of the transverse spinous processes [Fig. 9].
From this window, the VAs join to form the BA.
The BA is followed distally to a depth of 100 to
120 mm. The Doppler signal is below the baseline
with a higher mean velocity than that of the VAs
[Fig. 10]. If after several attempts, no signal is obtained from the suboccipital window the chin may
be tipped too far forward. Tip the head back and
angle the transducer so it is in the same plane as
the bridge of the nose.
Transorbital window
Doppler signals from the OA and the carotid siphon
are obtained using the transorbital approach. When
using the transorbital window, the ultrasound beam
travels through the eye. To minimize exposure to

the eye, the acoustic power must be reduced to a low
setting, 10% to 25% of maximum, and the examination time should be as short as possible. The
transducer is placed directly on the closed eyelid.
The depth of the sample volume should be approximately 40 to 50 mm with a slight medial angle.
Normal blood flow in the OA is directed toward
the transducer. The waveform may be highly pulsatile because it is supplying blood to the globe of the
eye [Fig. 11].
The carotid siphon is interrogated from this window at approximately 56 to 70 mm. The angle of
the transducer should be adjusted slightly as the
depth is increased. The signal from the carotid siphon changes from highly pulsatile in the OA to
low resistance in the carotid siphon. The direction
Fig. 10. Normal basilar artery. Suboccipital window at a depth of 94 mm, flow is directed away from transducer,
with mean velocity of 40 cm/s. (From Carson NL, Voci SL. Transcranial Doppler. In: Dogra V, Rubens DJ, editors.
Ultrasound secrets. Philadelphia: Hanley & Belfus; 2004; with permission.)
Fig. 11. OA normal. Right OA flow is towards transducer.
of the signal in the siphon depends on the location of the Doppler gate within the vessel. The carotid siphon is S-shaped and is divided into three
parts, the parasellar, the genu, and the supraclinoid.
The signal from the parasellar region is toward
the transducer, the signal is bidirectional in the
genu, and the signal is away from the transducer
in the supraclinoid section [Fig. 12]. All three sections may not be represented for each transorbital examination.
Submandibular window
The submandibular window is used to obtain a
signal from the extracranial portion of the internal
carotid artery. With the head tilted slightly to the
contralateral side, the transducer is placed at the
angle of the mandible aiming toward the head with
a slight medial angle. The depth of the sample

volume is at 50 to 64 mm and the direction of
flow is away from the transducer [Fig. 13].
Documentation
Several tracings are recorded at 2 to 4 mm intervals
as the vessels are followed. The velocity measurements taken from the spectral tracings include the
PSV, the end diastolic velocity, and the mean velocity. With the TCD method, all velocities are calculated by the equipment assuming a Doppler
angle of 0. In some laboratories, the pulsatility
index (PI), which provides information regarding
distal resistance, is recorded. In the setting of vasospasm, the ratio of the MCA to the eICA is
calculated. TCD studies show a decline in flow velocities with increasing age. Several published ta-
Fig. 12. Transorbital window. Transducer is placed on closed eyelid. From here, depth and angle are adjusted
to insonate OA and carotid siphon. Carotid siphon is divided into three parts: (1) parasellar portion, (2) genu, and
(3) supraclinoid portion.
167
168
Voci & Carson
Fig. 13. Extracranial ICA. From submandibular approach, depth is 60 mm. Flow is directed away from transducer at
mean velocity of 35 cm/s.
bles are available that can be used as normal criteria

[Table 1] [6,7].
TCD limitations
There are several limitations to the TCD approach.
The examination is highly operator dependent.
Vessel identification depends on the understanding
of normal anatomic relationships and knowledge
of possible variations from the norm. Approximately 50% of the population have an incomplete
circle of Willis [8]. Another limitation lies in the
actual spectral Doppler measurements. Doppler
frequency shift depends on the velocity of sound,
the velocity of flow, the insonating frequency, and
the angle between the ultrasound beam and the
vessel; thus, the calculated velocity and the true
velocity of blood flow differ as the angle of insonation increases [8]. The reported peak, diastolic, and
mean velocities may not be accurate with TCD

because the angle of the vessel is unknown. Another limitation lies in the transtemporal window
itself. At least 3% to 5% of patients lack an adequate window for the TCD examination. This percentage increases in older patient populations,
particularly among women [9]. Some patients simply cannot be examined.
The transcranial color-coded duplex

sonography examination
The TCCS method allows the examiner to visualize the vessels with color Doppler before obtaining
the spectral Doppler signals. Using the color Doppler not only makes vessel identification faster and
more accurate, it allows for angle correction of the
Doppler cursor and, thus, more accurate velocity
Table 1: Normal TCD velocities

Artery
MCA
ACA/MCA
ACA
dICA
PCA (P1)
Extracranial ICA
OA
VA
BA
Depth of
sample (mm)
3862
5464
6078
5664
6072
5064
4058
6085
80120
Mean velocity
(cm/sec)
Transducer window
Direction of flow
58
50
38
40
37
21
37
40
Transtemporal
Transtemporal
Transtemporal
Transtemporal
Transtemporal
Submandibular
Occipital
Suboccipital
Subocipital
Towards
Bidirectional
Away
12
12
10
10
9
5
10
10
Towards
Away
Away
Away
Away
Fig. 14. Transtemporal window. (A) Gray-scale reference point for circle of Willis is butterfly shaped hypoechoic
midbrain (arrows). (B) Color Doppler image of circle of Willis displays MCA toward transducer, ACA deep to MCA
and slightly anterior, and PCA directed posterior around midbrain. (C ) Power Doppler image shows more
complete view of circle of Willis than color Doppler with bilateral MCA, ACA, and PCA displayed.
169
170
Voci & Carson
readings. Direct visualization of the vessels also decreases training time for sonographers.
The same four windows are used to perform the
TCCS examination. The depths of the vessels with the
TCCS approach are similar to the depths used for
the TCD approach. The number of vessels interrogated
also depends on the indication for the examination.
To obtain the optimum window for the transtemporal approach, the probe is placed axially
along the orbitomeatal line and adjusted until the
hypoechoic, butterfly-shaped midbrain is seen at
the level of 6 to 8 cm [Fig. 14A] [4]. From this
point, the color Doppler or power Doppler is
turned on and the circle of Willis begins to appear
[Figs. 14B, C]. Next, the examiner can obtain anglecorrected spectral Doppler signals from the MCA,
ACA, PCA, and dICA. Angling the transducer
slightly may be necessary to see all of the vessels;
it is common to take the tracings from several
slightly different views. The transorbital TCCS approach is useful for vessel identification; however,
the gray-scale image of the actual brain parenchyma is technically limited.
After the vessels have been interrogated from the
transtemporal window, the examiner can proceed
to the occipital window. The landmarks are the
hypoechoic foramen magnum and the bright echo
of the clivus [Fig. 15A] [4]. With color or power
Doppler, the VAs and BA appear as a Y shape
[Fig. 15B].
Despite the clear advantages of vessel identification, and angle-corrected velocities, some limitations exist with the TCCS approach. The actual
shape and size of the phased array transducer
that is used may make it difficult to examine
patients who have small transtemporal windows.
The phased array tends to be larger than the transducer used in the TCD approach; thus, TCD may
Fig. 15. Occipital window. (A) Reference point for VAs and BAs is hypoechoic foramen magnum. (B) Power Doppler
shows normal VAs and BAs from occipital approach.
be better for those individuals. Another limitation

with TCCS lies in the physics of color Doppler.
Color Doppler, which is based on an average
Doppler frequency shift of moving blood, is limited by angle dependence and aliasing. Vessels
with low flow can be lost in background noise,
and vessels that are at large angles with respect
to the ultrasound beam may have weak or undetected signals. One solution to this is to use power
Doppler as suggested by Kenton and colleagues
[10] [Fig. 14C]. Their findings concluded power
Doppler particularly helped in the visualization
of the M2 portion of the MCA, the P2 portion of
the PCA, and the posterior communicating artery.
Another limitation is that in the presence of color
Doppler, the examiner may not pay close attention
to the audible Doppler signal and may miss the
segment of the vessel with the highest velocity. In
the TCD examination, the operator is trained to
listen for the signal with the highest pitch, which
correlates to the highest velocity [10].
Transcranial Doppler or transcranial

color-coded duplex sonography?
When it comes to the question of true anglecorrected velocity information, TCCS clearly provides more accurate data. Several investigators have
studied the differences between nonangle-corrected
and angle-corrected velocities of the various intracranial arteries, and whether just using color Doppler
(TCCS) without angle correction makes a difference. Neish and colleagues [11] studied a population of children with sickle cell disease and found
the mean velocity measurements in the MCA, dICA,
and ACA did not vary substantially between TCD
and TCCS techniques. Angle correction was not
used during the TCCS examination. Mean velocities
discovered in the dICA, the PCA, and the BA varied
substantially; these TCD velocities were higher than
the nonangle-corrected TCCS velocities for the
same vessels. The investigators explain that the discrepancy may be caused by the larger size of the
transducer limiting the windows for interrogation
of the dICA, PCA, and BA. When the examiner can
see the vessels, they may not be as tuned in to
listening for the highest pitch, which indicates the
highest velocity.
Another study by Schoning and colleagues
[12] compared TCD and TCCS data using an
adult population. In the MCA and ACA arteries,
the velocities were 10% to 15% higher using
TCCS than when TCD was used. They concluded
that the advantage of TCCS over TCD was more
qualitative than quantitative.
Using TCCS, Krejza and colleagues studied the
angle of insonation and its effect on flow velocity
measurements in patients who experience MCA

stenosis. Results indicated that the mean angle of
insonation for a stenotic MCA M1 segment is larger
than that of an unaffected normal artery. In patients who develop moderate or severe MCA stenosis, it was concluded that the angle of insonation
can be substantial, thus using nonangle-corrected
techniques can result in significantly underestimated peak and mean flow velocities [13].
Whether a department uses TCD or TCCS to
examine the intracranial arteries often depends on
the equipment available. The TCD examination
requires a separate dedicated piece of equipment,
an expense the laboratory may not be able to incur. The TCCS examination uses duplex equipment that already exists in the lab; however, the
examination requires a transducer specifically designed for TCD applications. The TCCS approach
allows for faster vessel identification but an experienced examiner can also perform the TCD examination in a timely fashion with a high level
of confidence. For reporting purposes, the department must take care in choosing its reference
data. The data must have been generated by the
same approach (eg, nonangled TCD data or anglecorrected TCCS data).
Clinical applications
Detection and monitoring of cerebral
vasospasm
One of the most serious complications of subarachnoid hemorrhage is cerebral vasospasm, which
is associated with significant morbidity and mortality. Cerebral vasospasm has been reported in up
to 40% of patients who present with subarachnoid
hemorrhage and typically occurs 4 to 14 days following the subarachnoid hemorrhage [14]. TCDdetected vasospasm can precede clinical vasospasm
by hours or days [4]. The velocity of blood flow is
inversely related to the lumen of the artery. With
vasospasm, the arterial lumen decreases, the blood
flow velocity increases, and the condition of the
vessels can be detected.
Transcranial Doppler has been shown to be a useful, noninvasive tool for the detection and evaluation of cerebral vasospasm. Flow velocity findings
in the MCA correlate with clinical grade, location
of blood clot, and the time course of angiographic
vasospasm [15]. For the MCA, mean flow velocities greater than 200 cm/s, sharp increase in flow
velocities, or a high Lindegaard ratio (mean velocity of the MCA/mean velocity of the ICA [vMCA/
vICA]) (6 .3) can reliably predict the presence
of clinically significant MCA vasospasm [15]. Similarly, a mean flow velocity less than 120 cm/s defines no clinically significant vasospasm [Table 2]
171
172
Voci & Carson
Table 2: Sonographic criteria for vasospasm of

the middle cerebral artery [6,1620]
Mean MCA velocity
(cm/s)
Vmca/
Vica
Degree of
stenosis
80
> 80120
> 120160
> 160
< 3.0
< 3.0
3.06.0
> 6.0
Normal
Mild
Moderate
Severe
Side-to-side discrepancy of the MCA greater than 14% is

abnormal; clinically significant vasospasm of the MCA
begins at 120 cm/s.
[1620]. A sharp increase in the flow velocity

(> 20cm/s/d) is associated with a poor prognosis
[Fig. 16] [4].
Hematocrit, blood pressure, PaO2 and PCO2,
collateral flow patterns, and intracranial pressure
may affect blood flow velocity measurements, and
therefore, should be taken into account when interpreting TCD results [15,20]. TCD velocity criteria
have been found to be most reliable for detecting
angiographic MCA and BA vasospasm [15]. Angiography is considered the gold standard for evaluation for cerebral vasospasm. Other modalities used
include CT angiography (CTA), and magnetic resonance angiography (MRA). Obvious advantages
of TCD over the other modalities are: it is noninvasive, can be performed at the bedside, can
be repeated as frequently as necessary, used for
continuous monitoring, does not require contrast
agents, and is less expensive. Disadvantages are:

it is operator dependant, an insufficient temporal
or foraminal window of insonation was found in
up to 20% of patients in one study [21], and vessel
anatomy is variable. TCCS is also useful for evaluation of vasospasm. Because the cerebral vessels
are directly visualized, angle-corrected velocities
can be obtained. Studies performed to evaluate normal angle-corrected flow velocities have shown
that angle-corrected velocities in adults tend to be
higher than velocities obtained with TCD. TCCS
may also directly visualize aneurysms, though at
this time TCCS is not the screening modality of
choice secondary to its low sensitivity and ready
availability of CTA or MRA [1].
Detection of intracranial stenosis and

occlusion
Intracranial atherosclerosis is the inciting factor in
up to 10% of patients presenting with TIA or stroke
[15,22]. Stenosis and occlusion of the ICA siphon,
M1 segment of the MCA, intracranial VA, proximal
BA, and P1 segment of the PCA can be reliably
detected by TCD [15]. Sensitivity, specificity, positive predictive value, and negative predictive value
of TCD are typically greater in the anterior circulation than in the vertebrobasilar circulation. This is
believed to be secondary to greater variation in the
vertebrobasilar anatomy and greater technical difficulties with the posterior circulation [15,3]. Fast,
Fig. 16. Vasospasm spectral Doppler tracing is from adult status postaneurysm clipping. Mean velocity of MCA is
elevated at 137 cm/s. (From Carson NL, Voci SL. Transcranial Doppler. In: Dogra V, Rubens DJ, editors. Ultrasound
noninvasive, portable, and cost effective diagnostic

tools for evaluating the acute stroke patient are
becoming more important because certain subgroups of patients may benefit from thrombolytic
and neuroprotective treatments [23]. TCD has
become extensively used in the evaluation of the
acute stroke patient.
Findings on TCD for a stenosis in a major basal
cerebral artery include increased flow velocity, spectral broadening of the waveform, and vibration of
the perivascular soft tissues [Fig. 17]. Normal mean
velocities for the basal cerebral arteries can be
found in Table 1. Felberg and colleagues [24], looking at the accuracy of different mean flow velocity
thresholds for determining the degree of stenosis,
found that using a mean velocity value of greater

than or equal to 100 cm/s had a sensitivity of
100%, specificity of 97.9%, a positive predictive
value of 88.8%, and a negative predictive value
of 94.9% in detecting greater than or equal to
50% stenosis of the MCA [24]. TCD criteria for
greater than 50% stenosis in intracranial arteries
have not yet been established. Rorick and colleagues [25] found the most important confounding factor in the evaluation of velocity criteria using
TCD was the presence of a greater than or equal to
75% stenosis of the eICA. Collateral flow may
result leading to false positive results or a false
negative result secondary to decrease in flow [25].
With TCCS, the angle-corrected PSV is used as
Fig. 17. Right MCA stenosis. (A) At the stenosis, velocities are elevated with PSV of 178 cm/s, and diastolic velocity
of 79 cm/s. (B) Distal to stenosis, PSV in MCA decreases to 64 cm/s. (C ) Ipsilateral ACA has increased PSV of 186 cm/s,
mean of 111 cm/s, and diastolic velocity of 77 cm/s, indicating increased flow through vessel. (D) Corresponding
angiogram shows MCA stenosis (arrow) and normal ACA. (From Carson NL, Voci SL. Transcranial Doppler. In:
Dogra V, Rubens DJ, editors. Ultrasound secrets. Philadelphia: Hanley & Belfus; 2004; with permission.)
173
174
Voci & Carson
the main criteria for intracranial stenosis. Table 3

(from work by Baumgartner and colleagues [22])
gives the upper limits of normal for PSV of the
intracranial arteries [4,22].
TCD can detect acute MCA occlusions with high
(> 90%) sensitivity, specificity, positive predictive
value, and negative predictive value as compared
with angiography [3,15]. TCD can also detect ICA
siphon, VA and BA occlusions with good (70%
90%) sensitivity and positive predictive value and
excellent specificity and negative predictive value
[15]. Cerebral arterial occlusion is diagnosed on
TCD by the following findings: lack of arterial
signal at expected depth, presence of signals in
vessels that communicate with the occluded vessel,
and findings corresponding to collateral flow, such
as elevated velocities in communicating arteries [4].
False positive TCD diagnoses of occlusion may
occur secondary to inadequate insonation window,
displacement of arteries secondary to mass lesions,

and variable anatomy of the circle of Willis [3,4].
TCCS definition of intracranial arterial occlusion
is based on absence of flow. With TCCS an inadequate window of insonation is not likely to lead to
a false positive. Poor Doppler angles or slow flow
can limit vessel visualization with TCCS. Diagnostic confidence for intracranial arterial occlusion
with TCCS is still close to 100% [4]. Ultrasound
contrast agents are being used with TCCS to improve diagnostic accuracy for intracranial arterial
stenoses and occlusions.

Stroke is the most severe cerebrovascular complication in sickle cell disease. An occlusive vasculopathy results from the adhesion of abnormal cells to
Table 3: Ultrasonic detection of 50% intracranial stenosis (n = 31) with angiography as standard
of reference
Ultrasound
ACA
MCA
PCA
BA
VA
PSV cutoff
(cm/s)
Sensitivity
(%)
Specificity
(%)
Positive
predictive
value (%)
155
220
145
140
120
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Angiography
Negative
predictive
value (%)
No.
Mean SD
degree (range)
100
100
91
100
100
4
11
10
3
3
60
67
63
67
69
8 (5271)
11 (5080)
7 (5072)
14 (5385)
14 (5584)
Abbreviation: SD, standard deviation.

From Baumgartner RW, Mattle HP, Schroth G. Assessment of >50% and <50% intracranial stenoses by transcranial colorcoded duplex sonography. Stroke 1999;30:8792; with permission.
the vascular endothelium and subendothelial matrix. The distal intracranial internal carotid arteries
and the proximal middle cerebral arteries are most
commonly involved. Eleven percent of children
with homozygous (HbSS) sickle cell disease experience a stroke before the age of 20 [2628]. The
Stroke Prevention in Sickle Cell Anemia Trial
(STOP) showed that children with mean flow velocities of 200cm/s or greater in the dICA or proximal MCA have a 10% risk for stroke per year. That
risk is 10 to 20 times that of age-matched children

who have sickle cell anemia and normal velocities [Fig. 18]. The STOP study also showed that
this risk could be reduced to less than 1% per
year with regular blood transfusions to decrease
the HbSS to less than 30% of total hemoglobin.
This study led to a recommendation from the National Heart Lung and Blood Institute of the National Institutes of Health that all children with
HbSS sickle cell anemia between the ages of 2
Fig. 18. Occlusive vasculopathy in sickle cell disease spectral Doppler tracing from a child with sickle cell disease
shows an elevated mean velocity of MCA at 224 cm/s.
175
176
Voci & Carson
Box 1: Interpretation of the transcranial

detection examination according to the Stroke
Prevention in Sickle Cell Anemia Trial protocol
1. Normal: <170 cm/s (mean velocity)
2. Conditional:
a. MCA, BIF, dICA 170cm/s but <200cm/s
(mean velocity)
b. PCA, TOB, BAS 170cm/s (mean velocity)
c. ACA 170cm/s (mean velocity)
3. Abnormal: 200 cm/s (mean velocity)
4. Inadequate (Inadequate studies are those
that cannot be interpreted. The examiner
is unable to identify a minimum of information, including the M1, MCA, and BIF
on temporal sides; however, if M1, MCA,
BIF, or dICA is greater than or equal to
200 cm/s, then the study is interpreted as
abnormal even if the other side is missing
or incomplete.)
and 16 should be screened for risk for stroke

[15,2628]. The STOP trial was performed with
TCD, not TCCS. The STOP protocol requires measurement of the bitemporal diameter with calipers
to aid in correct identification of the ICA bifurcation. The hematocrit is also evaluated before the
examination because there is an inverse relationship between hematocrit and cerebral blood flow
[Box 1] [29].
Because of the results of the STOP findings, an
increase in demand has occurred for TCD screening for children with HbSS. Many centers do not
have access to TCD equipment but do have access
to color Doppler imaging equipment; therefore,
studies have been performed to compare TCD
and TCCS findings in children who have sickle
cell disease. Neish and colleagues [11] found that
nonangle-corrected mean velocity measurements
in the MCA, distal internal carotid artery, and ACA
did not vary substantially between TCCS and TCD.
Jones and colleagues [28] found that nonanglecorrected mean velocity measurements were lower
with TCCS than TCD [27,28,30,31]. This potential
difference in mean velocity measurements should
be considered when screening children who develop HbSS for stroke risk with TCCS.

Vasomotor reactivity is the ability of the vessels to
undergo vasoconstriction or vasodilatation in response to various stimuli. Cerebral autoregulation
is the vascular response when the stimulus is a
change in perfusion pressure [32]. Transcranial
Doppler can demonstrate flow velocity changes
occurring over time, and therefore is an ideal test
for evaluation of autoregulation and vasomotor

reactivity. Vasomotor reactivity testing is performed
to evaluate the reserve mechanism of the cerebral
vasculature using various stimuli. Types of stimuli include CO2 inhalation, breath-holding, and
increased or decreased systemic arterial pressure.
With an intact vasomotor reserve a decrease in perfusion pressure can be offset by vasodilatation of
cortical arterioles to maintain adequate cortical
blood supply. In the situation where the resistance
vessels are already maximally dilated they cannot
respond to vasodilatory stimuli, and blood flow
does not increase. This can lead to ischemic brain
injury if perfusion pressure decreases further. Vasomotor reactivity testing with TCD has been used to
evaluate patients who present with symptomatic or
asymptomatic extracranial ICA stenosis or occlusion, cerebral small artery disease, head trauma,
and subarachnoid hemorrhage [4,15]. As per the
Report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of
Neurology, vasomotor reactivity testing was believed to be useful for detecting impaired cerebral
hemodynamics in patients who present with asymptomatic severe (greater than 70%) stenosis of the
eICA, symptomatic or asymptomatic eICA occlusion, and disease of the cerebral small arteries [15].

Transcranial Doppler can detect microembolic signals (MES) or high-intensity transient signals in
the intracranial arteries. Microemboli detection is
based on the backscatter from the emboli. The
backscatter of the ultrasound from flowing blood
is less than the backscatter from solid emboli
[15,33]. According to a consensus recommendation, an MES must have the following features:
(1) it must be transient, lasting less than 300 ms,
(2) the amplitude should be at least 3 dB greater
than background signal, (3) it should be unidirectional within the Doppler velocity spectrum, and
(4) the signal should be associated with a snap,
chirp, or moan [Fig. 19] [34]. In asymptomatic
patients, the identification of microembolic signals
may correlate to an active embolic source, possibly
identifying a subgroup of patients who are at high
risk for stroke. In symptomatic patients after an
acute event, identification of MES may identify
those at risk for recurrent stroke. This technique
may be useful in identifying the site of an embolizing lesion [33]. MES have been detected with
TCD in asymptomatic and symptomatic highgrade internal carotid stenosis, prosthetic valves,
myocardial infarction, atrial fibrillation, aortic arch
atheroma, fat embolization, and general cerebral
vascular disease. MES have also been detected during
Fig. 19. Typical MES detected with TCD are bright linear high amplitude echoes (arrowheads).
coronary catheterization, coronary angioplasty, cardio version, cerebral angiography, carotid endarterectomy, carotid angioplasty, and cardiopulmonary
bypass. TCD detection of MES has also been used to
monitor antithrombotic treatment in patients who
present with atherosclerotic cerebrovascular disease
[15]. The introduction of new hardware and software capabilities may improve the monitoring
procedure and help with discrimination from artifact [4,20].
A patent foramen ovale (PFO) is present in one
third of the population and up to 50% of patients
younger than 55 years of age presenting with ischemic stroke [35]. Contrast- enhanced transesophageal echocardiography (TEE) is the gold
standard for detection of paradoxical emboli by
way of a PFO. Data have shown a high correlation between contrast-enhanced TCD and contrastenhanced TEE in the detection of emboli from
PFO. The sensitivity and specificity of contrastenhanced TCD can be improved with the performance of a Valsalva maneuver. The sensitivity can
also be improved by increasing the volume of agitated saline used during the examination or by
using a contrast agent [15,35,36].
provide continuous monitoring. Most experience

with TCD monitoring in the operating room has
been during carotid endarterectomy where the ipsilateral MCA is monitored. Velocities and MES are
evaluated. A large decrease in velocities is considered an indication for blood pressure support,
shunt placement, and evaluation for shunt kinking
or thrombosis. MES occur most commonly during
dissection, shunting and unclamping, wound closure, and immediately postoperatively [15]. The
number of MES during the dissection phase and
wound closure is most predictive of postoperative
stroke [4]. Greater than 50 MES/h in the early postoperative period is predictive of the development of
ipsilateral focal cerebral ischemia [15]. The major
cause of stroke in the postoperative period is embolism from the operative site [15].
Cerebral infarction and encephalopathy occur in
up to 15% of patients undergoing coronary artery
bypass surgery. Neuropsychological deficits can occur in up to 70% of patients. The degree of neuropsychological deficit correlates with the number
of MES identified with intraoperative monitoring
with TCD [4,15]. TCD monitoring can document
changes in flow velocity and MES throughout all
phases of the surgery.
Intraoperative monitoring
Intraoperative monitoring with TCD has been used
during carotid endarterectomy and cardiac surgery
with cardiopulmonary bypass. Obvious advantages
of using TCD are that it is noninvasive and can
Brain death
The definition of brain death is the irreversible cessation of all functions of the entire brain. Clinical
177
178
Voci & Carson
criteria for brain death, such as a known cause of an

unresponsive comatose state, apnea, and absence
of brainstem reflexes, are usually adequate for the
diagnosis of brain death. Circumstances exist where
a confirmatory test is indicated. Patients treated
with sedatives, patients who present with facial
trauma, and ethical and legal considerations are
examples of circumstances requiring a confirmatory
test [37]. In addition, the increasing demand for
organ donors requires prompt identification of candidates before multiorgan failure [38]. Four vessel
angiography demonstrating non-filling of the cerebral arteries has been the gold standard. Angiography is invasive, expensive, and requires the patient
to be transported to the angiography suite. Radioisotope studies for cerebral blood flow have also
been usedthis is expensive, and uses radiopharmaceutical agents. TCD is noninvasive, portable,
and inexpensive. Specific intracranial arterial waveforms, which correspond to intracranial circulatory
arrest, have been described with TCD. These waveforms include oscillating flow, systolic spikes, and
no flow [Fig. 20]. To avoid false positive results,
right and MCAs and the BA must be evaluated.
Intracranial circulatory arrest and brain death may
not occur simultaneously. A lag can occur of almost
24 hours between the TCD documentation of intracranial circulatory arrest and complete loss of brain
function [38]. The sensitivity and specificity of TCD
in the confirmation of brain death is 91% to 100%,
and 100% respectively [37,38].
Ultrasound contrast agents

Ultrasound contrast agents in conjunction with
TCCS are currently used in clinical practice in
many countries, though are not yet approved by
the Food and Drug Administration for use in the
United States. As with TCD, a major limitation of
TCCS is insufficient insonation secondary to hyperostosis of the skull, obesity, and low flow velocities
[15,39]. Ultrasound contrast agents have been used
to offset this limitation. Ultrasound contrast agents
increase the Doppler signal intensity and increase
the signal-to -noise ratio for improved transcranial insonation [15]. In a study by Zunker and colleagues [39], looking at the frequency of use,
diagnostic benefit, and validity of results compared
with MRA, they found that ultrasound contrast
agents were indicated in 8.8% of their patients
who had undergone TCD and TCCS to reliably
evaluate the intracranial arteries. After ultrasound
contrast use, they had a diagnostic result in 75% of
cases with transtemporal and 81% during transforaminal insonation [39]. Artifacts associated with
ultrasound contrast agents include the blooming
effect (which occurs immediately after the bolus
arrives at the image location, and can obscure vessel anatomy) and spectral bubble noise that is
associated with a loud crackling noise, and an
increase in PSV by as much as 26 cm/s plus or
minus 10% [1,3,15]. Potential future applications
of ultrasound contrast agents include evaluation of
Fig. 20. Brain death characteristic to-and-fro flow is demonstrated in ICA. Flow is toward transducer in systole and
away from it in diastole, with net result of no forward flow. This same waveform pattern can be demonstrated in
intracranial arteries with TCD.
brain perfusion in stroke patients, and drug delivery, including delivery of thrombolytic agents [40].
New developments
Ultrasound enhanced thrombolysis
Experimental models have shown low frequency
ultrasound augments the activity of fibrinolytic
agents. Ultrasound enhanced thrombolysis occurs
by various mechanisms, including improved drug
transport, altering the fibrin structure, and increasing the binding of tissue plasminogen activator
(tPA) to fibrin. Studies have been performed with
ultrasound enhanced thrombolysis using KHz
frequencies, as these have better penetration, but
this resulted in increased risk for intracerebral
hemorrhage in stroke patients [40,41]. Experimental models found that 1- to 2.2-MHz TCD also
facilitated tPA infusion for clot dissolution, though
tPA-enhanced thrombolysis is slower with 1 MHz
than with 2 MHz frequencies [42]. At this time
2-MHz TCD is being used for diagnostic evaluation
and enhancement of thrombolysis with tPA infusion. The Combined Lysis of Thrombus in Brain
Ischemia Using Transcranial Ultrasound and Systemic t-PA (CLOTBUST) trial was a phase-2, multicenter, randomized clinical trial. This trial set out to
determine the differences in complete recanalization and early clinical recovery rates between
patients treated with tPA infusion and continuous
monitoring with 2-MHz TCD compared with those
treated with tPA infusion alone, and to determine
the safety of continuous 2-MHz TCD monitoring
of tPA infusion [43,44]. The trial concluded continuous monitoring with 2-MHz TCD did augment
tPA-induced arterial recanalization in patients who
experience acute ischemic stroke without an increase in bleeding. A trend toward an increase in
the rate of recovery from stroke exists, though this
trend was not found to be significant [41].
Summary
Since the introduction of TCD in 1982 by Aaslid
and colleagues [2], TCD and TCCS have become
essential tools in the diagnostic workup of cerebrovascular disorders. TCD and TCCS provide valuable
information as to anatomy, hemodynamic status,
and functional status, such as vasomotor reactivity
and autoregulation, [Box 2].
Major advantages of TCD and TCCS over other
cerebrovascular diagnostic modalities, such as angiography, CTA and MRA, are that they are noninvasive, portable, dynamic, and inexpensive. These
advantages promote the use of TCD and TCCS
in the operating room, emergency room, neurointensive care unit, and during thrombolysis. Excit-
Box 2: Main indications for transcranial

detection and transcranial color-coded duplex
sonography
Detection and monitoring of

cerebral vasospasm
Detection of intracranial stenosis
and occlusion
Intra-operative monitoring
Monitoring in the neurosurgical intensive
care unit
Evaluation for brain death
Detection of right to left shunts
Ultrasound assisted thrombolysis
(new development)
Evaluation of brain perfusion in the stroke
patient (new development)
ing new developments with TCD and TCCS include

ultrasound-enhanced thrombolysis and expanded
United States contrast applications in cerebral vascular diagnosis and monitoring of perfusion in the
stroke patient and contrast-assisted drug delivery.
References
[1] Baumgartner R. Transcranial color duplex sonography in cerebrovascular disease: a systematic
review. Cerebrovasc Dis 2003;16:413.
[2] Aaslid R, Markwalder TM, Norris N. Noninvasive
transcranial Doppler ultrasound recording of
flow velocity in basal cerebral arteries. Neurosurg
1982;57:769.
[3] Gahn G, von Kummer R. Ultrasound in acute
stroke: a review. Neuroradiology 2001;43:70211.
[4] Nabavi DG, Otis SM, Ringelstein EB. Ultrasound
assessment of the intracranial arteries. In: Zwiebel WJ, Pellerito JS, editors. Introduction to vascular ultrasonography. 5th edition. Philadelphia:
Elsevier Saunders; 2005. p. 22550.
[5] Otis SM, Ringelstein EB. Transcranial Doppler
sonography. In: Zwiebel WJ, editor. Introduction
to vascular ultrasonography. 4th edition. Pennsylvania: W.B. Saunders Co.; 2000. p. 177201.
[6] DeWitt LD, Rosengart A, Teal PA. Transcranial
Doppler ultrasonography: normal values. In:
Babikian VL, Wechsler LR, editors. Transcranial
Doppler ultrasound. St. Louis (Mo): Mosby; 1993.
p. 29.
[7] Fujioka KA, Douville CM. Anatomy and freehand
examination techniques. In: Newell DW, Aaslid
R, editors. Transcranial Doppler. New York:
Raven Press, Ltd; 1992. p. 931.
[8] Otis SM. Pitfalls in transcranial doppler diagnosis. In: Babikian VL, Wechsler LR, editors. Transcranial Doppler sonography. St Louis (MO):
Mosby-Year Book Inc.; 1993. p. 3950.
179
180
Voci & Carson
[9] Klotzsch C, Popescu O, Berlit P. A new 1-MHz

probe for transcranial Doppler sonography in patients with inadequate temporal bone windows.
Ultrasound Med Biol 1998;24(1):1013.
[10] Kenton AR, Martin PJ, Evans DH. Power Doppler: an advance over colour Doppler for transcranial imaging? Ultrasound Med Biol 1996;
22(3):3137.
[11] Neish AS, Blews DE, Simms CA, et al. Screening
for stroke in sickle cell anemia: comparison of
transcranial Doppler imaging and nonimaging
us techniques. Radiology 2002;222(3):70914.
[12] Schoning M, Buchholz R, Walter J. Comparative
study of color duplex sonography and transcranial Doppler sonography in adults. J Neurosurg
1993;78:77684.
[13] Krejza J, Mariak Z, Babikian V. Importance of
angle correction in the measurement of blood
flow velocity with transcranial Doppler sonography. Am J Neuroradiol 2001;22:17437.
[14] Suarez JI, Qureshi AI, Yahia AB, et al. Symptomatic vasospasm diagnosis after subarachnoid
hemorrhage: evaluation of transcranial Doppler
ultrasound and cerebral angiography as related
to compromised vascular distribution. Crit Care
Med 2002;30(6):134855.
[15] Sloan MA, Alexandrov AV, Tegeler CH, et al.
Assessment: transcranial Doppler ultrasonography. Neurology 2004;62:146881.
[16] Krejza J, Mariak Z, Lewko J. Standardization of
flow velocities with respect to age and sex improves the accuracy of transcranial color Doppler
sonography of the middle cerebral artery spasm.
[17] Proust F, Callonec F, Clavier E, et al. Usefulness
of transcranial color-coded sonography in the
diagnosis of cerebral vasospasm. Stroke 1999;30:
10918.
[18] Lindegaard KF. The role of transcranial Doppler in the management of patients with subarachnoid haemorrhage: a review. Acta Neurochir
(Wien) 1999;72:5971.
[19] Skirboll S, Newell DW. Noninvasive physiologic
evaluation of the aneurysm patient. Neurosurg
Clin N Am 1998;9(3):46383.
[20] Mascia L, Fedorko L, terBrugge K, et al. The
accuracy of transcranial Doppler to detect vasospasm in patients with aneurysmal subarachnoid hemorrhage. Intensive Care Med 2003;29:
108894.
[21] Droste DW, Kaps M, Vavabit DG, et al. Ultrasound contrast enhancing agents in neurosonology: principles, methods, future possibilities.
Acta Neurol Scand 2000;102:110.
[22] Baumgartner RW, Mattle HP, Schroth G. Assessment of 50% and < 50% intracranial stenoses
by transcranial color-coded duplex sonography.
Stroke 1999;30:8792.
[23] Gerriets T, Goertler M, Stolz E, et al. Feasibility
and validity of transcranial duplex sonography in
patients with acute stroke. J Neurol Neurosurg
Psychiatry 2002;73:1720.
[24] Felberg RA, Christour I, Demchuck AM, et al.

Screening for intracranial stenosis with transcranial Doppler: the accuracy of mean flow velocity
thresholds. J Neuroimaging 2002;12:914.
[25] Rorick MB, Nichols FT, Adams RJ. Transcranial Doppler correlation with angiography in detection of intracranial stenosis. Stroke 1994;25:
19314.
[26] Riebel T, Betzing Kebelmann C, Gutze R, et al.
Transcranial Doppler ultrasonography in neurologically asymptomatic children and young
adults with sickle cell disease. Eur Radiol 2003;
13:56370.
[27] Jones AM, Seibert JJ, Nichols FT, et al. Comparison of transcranial color Doppler imaging
(TCDI) and transcranial Doppler in children
with sickle cell anemia. Pediatr Radiol 2001;31:
4619.
[28] Neish AS, Blews DE, Simms CA, et al. Screening for stroke in sickle cell anemia: comparison of transcranial Doppler imaging and
nonimaging US techniques. Radiology 2002;222:
70914.
[29] TCD in children with sickle cell disease. An
intensive training course. Presented by School
of Medicine, Medical College of Georgia, the
Alumni Center. Augusta, January 2730, 1998.
[30] Bulas DI, Jones A, Seibert JJ, et al. Transcranial
Doppler (TCD) screening for stroke prevention in sickle cell anemia; pitfalls in technique
variation. Pediatr Radiol 2000;30:7338.
[31] Razumovsky AY. TCD methodology in children
with sickle-cell anemia. Pediatr Radiol 2002;32:
6901.
[32] Moppett IK, Mahanjan RP. Transcranial Doppler ultrasonography in anaesthesia and intensive
care. British Journal of Anaesthesia 2004;93(5):
71024.
[33] Ringelstein EB, Droste DW, Babikian VL, et al.
Consensus on microembolus detection by TCD.
Stroke 1998;29:2759.
[34] Babikian VL. Basic identification criteria of Doppler microembolic signals. Stroke 1995;26:1123.
[35] Uzuner N, Horner S, Pichler G, et al. Rightto-left shunt assessed by contrast transcranial
Doppler sonography. J Ultrasound Med 2004;23:
147582.
[36] Droste DW, Silling K, Stypmann J, et al. Contrast transcranial Doppler ultrasound in the detection of right-to-left shunts. Stroke 2000;31:
16405.
[37] Hadani M, Bruk B, Ram Z, et al. Application
of transcranial Doppler ultrasonography for the
diagnosis of brain death. Intensive Care Med
1999;25:8228.
[38] Dosemeci L, Dora B, Yilmaz M, et al. Utility of
transcranial Doppler ultrasonography for confirmatory diagnosis of brain death; two sides of
the coin. Transplantation 2004;77:715.
[39] Zunker P, Wilms H, Brossmann J, et al.
Echocontrast-enhanced transcranial ultrasound:
frequency of use, diagnostic benefit, and validity
of results compared with MRA. Stroke 2002;33:

26003.
[40] Droste D, Metz RJ. Clinical utility of echocontrast agents in neurosonology. Neurol Res 2004;
26:7549.
[41] Alexandrove AV, Wojner AW, Grotta JC, et al.
Ultrasound-enhanced thrombolysis for acute ischemic stroke. N Engl J Med 2004;351(21):2170.
[42] Alexandrove AV, Wojner AW, Grotta JC, et al.
Ultrasound-enhanced thrombolysis for acute
ischemic stroke: phase I findings of the CLOTBUST trial. J Neuroimaging 2004;14(21):1137.

[43] Alexandrove AV, Molina CA, Grotta JC. CLOTBUST: ultrasound-enhanced systemic thrombolysis for acute ischemic stroke. J Neuroimaging
2004;14(21):10812.
[44] Carson N, Voci SL. Transcranial Doppler ultrasound. In: Dogra V, Rubens DJ, editors. Ultrasound secrets. 1st edition. Philadelphia: Hanley &
Belfus; 2004. p. 394402.
181
183
ULTRASOUND
CLINICS
Ultrasound Diagnosis of Arterial

Injuries and the Role of Minimally
Invasive Techniques in
Their Management
Nael E.A. Saad, MB, BCh*, Wael E.A. Saad,
Deborah J. Rubens, MD, Patrick Fultz, MD
&
&
&
&
&
&
&
&
Natural history
Clinical features
Ultrasound imaging
Treatment
Surgery
Ultrasound-guided compression
Ultrasound-guided percutaneous thrombin
injection
Endoluminal management
Pseudoaneurysms (PsAs) arise from a localized

disruption of the arterial wall from a variety of
causes. Under high arterial pressure, blood dissects
into the tissues around the damaged artery forming
a perfused sac that communicates with the arterial
lumen [13]. The wall of the PsA is formed by
residual portions of the wall (the media or adventitia) or by the soft tissues surrounding the injured
vessel. Modern medical imaging modalities have
increased the diagnosis of PsAs [4] facilitating
early detection before onset of clinical symptoms.
Conventional arteriography, however, still remains
the gold standard [5] for diagnosis and anatomic
evaluation of PsAs. The other modalities, including
ultrasonography with Doppler evaluation, magnetic resonance arteriography (MRA), and helical
computed tomographic arteriography (CTA) offer
noninvasive imaging options with variable results
&
&
&
&
&
&
&
&
&
MB, Bch,
Arteriovenous fistulae
Natural history
Ultrasound imaging
Treatment
Surgery
Summary
References
[6,7]. As imaging guidance also improves, PsA

therapy has evolved from traditional surgical management toward a less invasive, percutaneous approach. The combination of early detection and
minimally invasive management has decreased dramatically the morbidity and mortality from PsAs.
Natural history
PsAs are the end result of multiple causes, including
inflammation, trauma, surgical procedures [1], and
other iatrogenic causes, such as percutaneous biopsy or drainage, may result in PsA formation in
any artery. The high sensitivity of modern imaging
modalities has made the diagnosis of PsAs more
common; however, there is also a real increase in
the incidence of PsAs [8] as a result of an increase in
the number of vascular surgical and arteriographic
Department of Radiology, University of Rochester Medical Center, 601 Elmwood Avenue, Box 648, Rochester,
NY 14642, USA
E-mail address: nael@mindless.com (N.E.A. Saad).
doi:10.1016/j.cult.2005.09.008
184
Saad et al
procedures performed. Surgery causes PsAs by direct injury to the vessel or introduction of infection.
Numerous procedures may induce PsAs, including
those considered minimally invasive such as diagnostic angiography whereby the incidence of PsA
formation is less than 1%. Therapeutic catheterization for angioplasty or stents often requires large
bore sheaths, periprocedural anticoagulation, and
antiplatelet therapy, which raise the incidence of
PsA formation to 3.2% to 7.7% [9,10]. PsAs are
also recognized complications of liver transplantation [11,12], cardiac transplantation [13], obstetric
procedures (eg, dilatation, curettage, and caesarian
sections [14]), and endovascular aortic stent grafts
[15]. Blunt and penetrating trauma may also cause
PsAs of multiple vessels, including the carotid [5],
extremity [6,16], splenic [17,18], and hepatic [19]
arteries. PsAs may also be caused by vascular infection or inflammation, including pancreatitis [20].
PsAs may undergo spontaneous thrombosis [7,21]
or may develop complications, such as local neurovascular compression, infection, or rupture [10].
Clinical features
Asymptomatic PsAs are detected incidentally during
radiologic investigation of other conditions or during surgery [22]. Symptomatic PsAs manifest with
local or systemic signs and symptoms. On physical
examination, a pulsatile mass, a palpable thrill, or
an audible bruit may be appreciated. Local effects of
a PsA are caused by mass effect on adjacent structures causing compromise of function. Ischemia of
the surrounding tissues (caused by vascular com-
promise) may lead to necrosis of the overlying skin

and subcutaneous tissue. Neurologic symptoms may
develop secondary to nerve compression/ischemia.
Compression of adjacent veins may cause edema or
deep venous thrombosis. Arterial thromboembolism
is also a potential complication. In addition, the PsA
may rupture leading to hemorrhage with potential
life threatening shock [7,10,21,23]. Rupture is the
most serious cause of morbidity from PsAs; for example, rupture of splenic artery PsAs has a mortality rate
approaching 100% [24]. PsAs can communicate with
and rupture into the bowel, or biliary system, and
the thoracic, peritoneal, pelvic, and retroperitoneal
spaces [25]. Hemorrhage from these ruptures may
present as a sentinel bleed, bleeding from a drain,
hematemesis, melena [23], splenic rupture, or subcapsular hepatic hematoma [26].
Ultrasound imaging
Gray-scale ultrasound (US) usually demonstrates a
predominantly anechoic mass adjacent to the artery
[Fig. 1A] [8,12,14]. This mass may be spherical or
irregular and may contain septations or debris
[Fig. 2A]. The size of the PsA sac, the number of
its compartments (lobes), its connection to the
artery, and the length and width of the neck of
the PsA can also be assessed [Fig. 1B] [9]. PsAs
may be simple (one lobe) or complex (two or
more lobes separated by a patent tract, which has
a diameter smaller than the minimal dimension of
the smallest lobe) [9]. In addition, a concentric
hematoma within the PsA may be seen occasionally
[Fig. 3AC]. Gray-scale US alone is not diagnostic
Fig. 1. (A) Gray-scale ultrasound with spectral waveform from the PsA neck depicting bidirectional flow within
neck (N). The donor artery (A) supplying the PsA can also be seen. (B) Color Doppler image of same PsA with easy
depiction of PsA neck (N) and the yin-yang (red-blue) Doppler appearance of the PsA sac. (From Saad N, Saad W,
Rubens DJ. Pseudoaneurysms and the role of minimally invasive techniques in their management. Radiographics
2005;25:S1739; with permission.)
Diagnosing and Managing Arterial Injuries
Fig. 2. (A) Gray-scale ultrasound image of right groin in patient who underwent catheterization demonstrates
irregular mass (cursors) with mixed echogenicity. (B) Color Doppler image of same mass demonstrating characteristic yin-yang (red-blue) flow in sac. Area lacking flow represents thrombus within sac. (C ) Spectral wave form
from PsA neck depicting bidirectional flow within neck.
185
186
Saad et al
Fig. 3. (A) Gray-scale transvaginal US image depicting PsA of left internal iliac artery (arrow head) and concentric
layers of mural thrombosis (arrow). (B) Transvaginal color Doppler image of same PsA demonstrating characteristic yin-yang (red-blue) flow in sac. (C ) Transvaginal color flow Doppler image with spectral wave form demonstrating to-and-fro flow (bidirectional flow) within sac. (From Saad N, Saad W, Rubens DJ. Pseudoaneurysms and
the role of minimally invasive techniques in their management. Radiographics 2005;25:S1739; with permission.)
because the findings above are seen in several conditions, including simple and complex cysts, hematomas, and abscesses.
Doppler US establishes the diagnosis of a PsA. The
blood flow within the mass is characterized by a
typical swirling motion called the yin-yang sign
whereby the blood flows into the sac and then out
of it, reversing colors as it reverses flow direction
[Figs. 1B, 2B, and 3B]. This pattern of flow may also
be seen in saccular aneurysms, and a diagnosis made
on this finding alone may prove to be inaccurate.
The hallmark of the diagnosis is the demonstration
of a communicating channel (neck) between the PsA
sac and the feeding artery and documentation of a
to-and-fro waveform on the spectral Doppler tracing
within the neck. The to component is caused by
blood entering the PsA in systole and the fro
component is caused by blood exiting the PsA during diastole [Figs. 1A and 3C] [26]. In addition,
placing the imaging findings in the clinical context
(history of the etiology of PsAs) allows the diagnosis
of a PsA versus a saccular, true aneurysm.
US is a valuable diagnostic tool for detection of

PsAsit is portable, readily available, inexpensive,
fast, uses no ionizing radiation or nephrotoxic contrast material, and is noninvasive [11,27,28]. US
has a sensitivity of 94% and specificity of 97% in
detecting postcatheterization PsAs [10]. Evaluation
of deep (visceral) arteries is limited, however, with
reports of low sensitivity [29]. US is operator and
patient dependent; thus, evaluation of vessels in the
setting of trauma with overlying fractures or adjacent hematomas can be limited [6].
Treatment
Symptomatic PsAs (intermittent or continuous bleeding) should be treated; however, the decision to treat
asymptomatic PsAs is controversial because of the
variable natural history of PsAs, particularly when
one considers the different outcomes based on anatomic locations of individual PsAs and the clinical
setting with patient comorbidities. PsAs may undergo
spontaneous thrombosis [20,30], and some investi-
gators have advocated observation for small, asymptomatic PsAs; however, no method exists to predict
which PsAs thrombose spontaneously [21,31].
The literature debate regarding spontaneous thrombosis of PsAs is mostly in regard to postcatheterization extremity PsAs versus post-traumatic visceral
(hepatic and splenic) PsAs [20,21,30,31]. In the
authors clinical practice, observation of small
asymptomatic PsAs is recommended in addition
to treatment only if they enlarge, do not resolve,
or become symptomatic; however, the risk for
spontaneous rupture of extra-organic visceral PsAs
is high, regardless of their size. The mortality rate of
these ruptures in the setting of morbid postsurgical
patients approaches 100% [23,24,26]; therefore,
some investigators believe that all PsAs in this setting should undergo definitive treatment regardless
of their size [24,27].
Traditionally, the treatment of PsAs has been
surgical repair [9,10,21,24], which is invasive and
associated with significant morbidity and mortality
[21,30]. Over the past few years, however, radiology has developed minimally invasive treatment
alternatives to the traditional surgical management [9,21,24]. These treatments include US-guided
compression, direct percutaneous management (including US-guided thrombin injection), and endoluminal management [3237].
The overall therapeutic options (including observation) of PsAs should be tailored to the site, rupture risk, comorbidities, and clinical setting (ie,
infection, relationship to graft or prosthesis, and
so forth) of the PsA in question. The following is an
overview of the various therapeutic approaches,
their relative merits, drawbacks and indications.
Surgery
Surgical management of PsAs in general is variable
and includes resection with bypass, arterial ligation, and partial or complete organ removal [38].
Intra-organic visceral PsAs can be managed surgically by partial or complete organ removal: example renal PsAs are treated with partial or complete
nephrectomy, hepatic PsAs with segmentectomy
and splenic PsAs with splenectomy. Surgical treatment of PsAs involving vital arteries aims at preservation of the arterial perfusion of the tissues
supplied by this artery. Resection of the PsA with
patch repair of the vital donor artery or arterial
ligation with surgical bypass may achieve this.
Expendable arteries may be managed by arterial
ligation alone. The definitions and examples of
vital versus expendable donor arteries are discussed
in the section Endoluminal Management.
Surgery is the traditional treatment of choice. The
risks and potential complications associated with
surgery are well-established and include anesthesia

risks, bleeding, wound infection, lymphocele formation, radiculopathy, prolonged recovery time,
perioperative myocardial infarction, and death
[21,30]. Surgical repair of an artery in the vicinity
of a prior surgical bed is technically difficult as a
result of fibrosis and adhesions, which may also
increase the risk for complications. As a result of
technologic advances in US-guided and endoluminal management of PsAs, an ongoing paradigm
shift has occurred to minimally invasive management of PsAs. Surgery, however, still plays a role in
certain situations.
An example of this paradigm shift and the particular role of surgery is in the management of
superficial arteries in the extremities. Surgical repair
of superficial extremity PsAs was the treatment of
choice until 1991 when US-guided compression
was introduced [10,21,39]. The technique of surgical repair is governed by the presence or absence of
concomitant infection. In noninfected PsAs, primary repair of the artery is performed. The segment
of the artery containing the PsA is resected and the
proximal and distal segments of the artery are anastomosed in an end-to-end fashion. If the edges
cannot be approximated an interposition or bypass
graft is used for the anastomosis. Autologous vein
grafts are preferred over synthetic grafts when possible. In the presence of infection, efforts are directed toward control of the infection with antibiotic
therapy and drainage of abscesses or debridement of
infected tissues. Surgical repair of the artery is then
performed preferably with vein bypass grafts rather
than synthetic grafts. The graft is positioned in an
extra-anatomic location far from the infected tissues
[10]. Despite the large and expanding role of imageguided therapy, surgical management still plays an
important role in those PsAs in which local mass
effect produces ischemia and neuropathy in an attempt to decompress the area rapidly. Surgery is still
indicated for infected PsAs and in cases that have
failed therapy by minimally invasive techniques [10].
Since first described by Fellmeth and colleagues [40]
in 1991, US-guided compression of PsAs rapidly replaced surgery to treat postcatheterization extremity
PsAs [10,21,40]. Compression is performed with
the US transducer itself, which permits direct and
continuous visualization of the vessels [34]. Pressure is applied at various PsA locations, depending
on lesion access. A typical protocol includes an
initial 10- to 20-minute compression of the PsA
neck [10,41]. If this protocol is not feasible, compression of the PsA itself is performed [10,31].
Compression should eliminate flow within the
187
188
Saad et al
PsA, but permit arterial perfusion to the extremity

[10]. Cycles are repeated up to 1 hour total. Anticoagulation therapy decreases success rate; therefore,
anticoagulants should be discontinued before the
procedure if possible [10].
The compression technique is also limited by
patient and operator discomfort caused by the extended time required to achieve thrombosis. Patient discomfort is significant, often requiring the
administration of analgesics [10,42]. Only fairly
superficial PsAs can be treated by this method,
mainly the common catheterization access sites,
including the femoral, axillary, and brachial arteries
[10]. Patient obesity or edema can hinder compression efficacy even at those sites. Other factors affecting success of this method include PsA size (small
ones do better than large), whether the PsA is
simple or complex (simple ones do better), age of
the PsA (fresh lesions do better), and the length
and width of the PsA neck (long, thin necks are
preferable) [10]. Rare complications include venous thrombosis, skin necrosis, PsA rupture [42],
local arterial thrombosis, and distal embolization
caused by occlusion of the feeding artery. [36]. The
initial failure rate is 15% to 38% and recurrence
rates are as high as 20% to 30% in anticoagulated
patients [10,21,42,43].
Ultrasound-guided percutaneous thrombin

injection
US-guided percutaneous thrombin injection has
replaced US-guided compression as the first-line therapy to treat postcatheterization PsAs [21,36,43].
Cope and Zeit [44] first reported fluoroscopically
guided percutaneous thrombin injection therapy in
1986 [34]. US-guided percutaneous thrombin injection was not embraced fully until Kang and
colleagues [45] presented their work in 2000 [34].
Fig. 4. (A) Doppler image of left common femoral artery PsA in patient who underwent cardiac catheterization.
(B) Doppler image of same PsA showing no flow after percutaneous injection of 1 mL of 1000 IU of thrombin
through 18-gauge needle. (From Saad N, Saad W, Rubens DJ. Pseudoaneurysms and the role of minimally invasive
techniques in their management. Radiographics 2005;25:S1739; with permission.)
The therapy is derived from a naturally occurring hemostatic event whereby thrombin converts
inactive fibrinogen to fibrin, leading to thrombus
formation [10]. Additional thrombin (beyond the
circulating thrombin in the patient) is prepared by
adding sterile normal saline to commercially available sterile thrombin powder [10]. Most protocols
use a 1000 IU/mL concentration [10,21,33,36,42];
however, some investigators advocate the use of a
lower concentration of 100 IU/mL [39].
Following complete sonographic evaluation of
the PsA, the site is prepared with sterile technique
and local anesthesia applied. Under real-time US
guidance the needle is positioned in the center of
the PsA sac and thrombin is injected at a constant
rate. Flow within the sac is monitored constantly
with color or power Doppler US. Thrombin is injected steadily until flow within the sac ceases,
usually within seconds [Figs. 4A, B] [10,21,42].
The volume of injected thrombin ranges from
0.51.0 mL [33,39].
Technical success rates using thrombin to treat postcatheterization PsAs are greater than 90% [10,21,
44], even in patients undergoing anticoagulation
or antiplatelet therapy [33,36,42,43,46,47], a major
improvement compared with success rates of USguided compression [33,39,41,43,47]. Unlike compression, this method is not limited to superficial
arteries; deep visceral artery PsAs may also be treated
with thrombin [33,35], especially when the donor
artery is inaccessible for endoluminal transcatheter
therapy and its occlusion is not a viable option [4].
Arteries above the inguinal ligament can also be

treated with thrombin [41], when US-guided compression is contraindicated because of the potential
risk for rupture [41].
Complication rates of thrombin therapy are can
be as low as 4% [46]. The major complication is
the generation of emboli caused by thrombin flowing beyond the clot into the arterial system rather
than remaining within the PsA sac [9,10,21,33,
35,36,39,42,46,47]; therefore, a lower concentration of thrombin (100 IU/mL) may be used [39].
Some investigators advocate balloon occlusion of
the PsA neck before thrombin injection, making
the procedure more invasive; however, the complications from this method outweigh the theoretic
potential benefit [10,35]. Postprocedural monitoring of peripheral pulses and the ankle-brachial index
allows early detection of peripheral arterial thromboembolism downstream from injection [33]. If
downstream thrombosis occurs, prompt intraarterial thrombolysis should be performed [39].
Other potential complications include venous thrombosis and allergic reactions, such as generalized urticaria and anaphylaxis [9,10,21,33,42,46,47]. In
fact, contraindications to the use of this method
include a history of allergic reaction to thrombin.
Skin testing before therapy has been proposed to
avoid potentially fatal reactions [33,43]. Other
contraindications include local infection and distal limb ischemia [36,42]. Although thrombin is
the preferred agent for percutaneous embolization
of PsAs, other materials, such as coils, may be used
Fig. 5. Characteristic features determining management of PsAs. (From Saad N, Saad W, Rubens DJ. Pseudoaneurysms and the role of minimally invasive techniques in their management. Radiographics 2005;25:S1739;
with permission.)
189
190
Saad et al
Fig. 6. Algorithm for management of PsAs. (From Saad N, Saad W, Rubens DJ. Pseudoaneurysms and the role of
minimally invasive techniques in their management. Radiographics 2005;25:S1739; with permission.)
independently or in conjunction with thrombin

[4,10].
These techniques serve to exclude a PsA from the
circulation by placing a stent across the PsA neck or
by occluding the PsA or the feeding artery. Selecting
the optimal method depends on the size of the

PsA neck and the expendability of the donor artery
[Figs. 5 and 6] [4,10]. Fig. 6 is a generalized schematic for noncatheterization-related extremity PsAs.
A PsA arising from an expendable donor artery
without collaterals is treated with embolization of
the afferent artery [Fig. 7A, B] [4]. Several visceral
arteries exist that have a well-established collat-
Fig. 7. (A) A 67-year-old man presented to emergency department after sustaining gunshot wound to right arm.
Digital subtraction angiogram shows PsA (arrow head) of the circumflex humeral artery. (B) Digital subtraction
angiogram showing obliterated PsA (arrow head). The PsA was excluded from donor artery by coil embolization.
(From Saad N, Saad W, Rubens DJ. Pseudoaneurysms and the role of minimally invasive techniques in their
management. Radiographics 2005;25:S1739; with permission.)
Fig. 8. (A) A 61-year-old woman with cholangiocarcinoma underwent Whipple procedure and multiple percutaneous transhepatic cholangiograms with biliary drainage catheter placement. Digital subtraction angiogram
demonstrates PsA (arrow head) off branch of right hepatic artery. (B) Digital subtraction angiogram showing
PsA with its afferent (arrow head) and efferent (arrow) hepatic artery segments. (C) Digital subtraction angiogram after excluding PsA by distal (arrow) and proximal (arrow head) embolization. (From Saad N, Saad W, Rubens
DJ. Pseudoaneurysms and the role of minimally invasive techniques in their management. Radiographics 2005;25:
S1739; with permission.)
eral supply, including the gastroduodenal, hepatic,

splenic, and other upper gastrointestinal arteries
[4,26]. In addition, some pelvic and distal extremity arteries have a collateral supply (eg, the internal
iliac artery and the profunda femoris artery respectively). When embolizing arteries with numerous
collaterals, embolize proximal and distal to the PsA
to exclude it completely from the circulation by
preventing back flow from collateral circulation
[Fig. 8AC] [4].
A PsA arising from a nonexpendable donor artery
must be excluded from the circulation while preserving the donor artery. A vital donor artery may be
embolized in certain emergent situations (eg, rupture
with active bleeding); however, this must be followed by restoration of blood flow distally by
means of surgical bypass [Fig. 9A, B]. Most of the
time, however, the PsA can be thrombosed by coils
delivered into the sac or by stent-graft placement. The
width of the PsA neck compared with the donor
artery diameter determines which method is optimal.
If the neck is narrow, the PsA may be embolized by
catheter-directed delivery of coils (the preferred em-
bolization material) into the sac itself [Fig. 10A, B]

[4,24,25]. Coils fall under two main categories:
(1) nondetachable coils and (2)detachable coils.
Nondetachable coils are available in a wide array
of diameters and lengths. They are linear in shape
and composed of stainless steel with the addition of polyester fibers to increase their thrombogenicity. Newer coils are made of titanium, which
makes them softer and allows for formation of
more complex shapes (helical) and for use of
fewer coils for embolization as they conform to
and fill the PsA sac. Because the newer coils are
softer, however, they may require initial placement
of stainless steel coils to act as scaffolding. Detachable coils are held to the pusher guidewire by a
mechanical connection that is released to deploy
the coil, facilitating more accurate deployment and
the possibility of readjusting the position of the coil
before its final deployment. A disadvantage of coils
as an embolization material is the potential for
recanalization of the embolized sac if the coils
are not packed tightly. This drawback has been
overcome largely by the use of the soft helical
191
192
Saad et al
Fig. 9. (A) Digital subtraction angiogram showing PsA of suprageniculate popliteal artery. (B) Digital subtraction
angiogram after successful embolization distal (arrow head) and proximal (arrow) to PsA. Patient subsequently
underwent surgical bypass of occluded popliteal artery. (From Saad N, Saad W, Rubens DJ. Pseudoaneurysms and
the role of minimally invasive techniques in their management. Radiographics 2005;25:S1739; with permission.)
coils, which may be packed tightly in the PsA sac

[48]. Other agents such as thrombin or N-butyl2cyanoacrylate (glue) may be used alone [25,37] or
in addition to coils [4].
If the neck is wide, the PsA may still be embolized
by catheter-directed delivery of embolization materials; however, remodeling is required to prevent
outflow of these materials into the donor artery.
The goal is to ensure adequate embolization of the
PsA sac while preventing distal embolization of the
donor artery. A stent cage can be used in which
the catheter tip is directed between the lattice before release of the coils [Fig. 11A, B], or the coils
can be trapped by inflating a temporary occlusion
balloon in the donor artery between coil deployments [10]. In cases concerning distal arterial embolization, detachable balloons may be used as the
embolic agent [9,22]. Another option if the PsA neck is

wide, is stent-graft (covered stent) placement across
the PsA neck to exclude it [Fig. 12AC] [10]. This
method is contraindicated in mycotic PsAs because
of potential stent-graft infection. Stent grafts require a larger diameter catheter and a stiffer delivery
system than coil embolization transcatheter delivery
systems. Accordingly, the arterial anatomy and the
caliber of the arteries leading to and at the PsA
site should be favorable (wide and straight arteries). Currently, stent grafts are suitable for larger
arteries only, because in small arteries they have a
higher risk for thrombosis. Visceral PsAs are usually
smaller and arise from small and tortuous donor
arteries, posing a particular challenge to stent grafts.
Endovascular techniques have a lower rate of complication for the treatment of visceral PsAs when
Fig. 10. (A) Digital subtraction angiogram showing PsA (arrow) off proper hepatic artery. (B) Digital subtraction angiogram after coil embolization of PsA itself (arrow) with preservation of flow in hepatic artery distally
(arrow head). (From Saad N, Saad W, Rubens DJ. Pseudoaneurysms and the role of minimally invasive techniques
in their management. Radiographics 2005;25:S1739; with permission.)
Fig. 11. (A) Digital subtraction angiogram of post-traumatic PsA of the common carotid artery (arrow). (B) Digital
subtraction angiogram following PsA exclusion using bare-stent (uncovered stent) and coil embolization (arrow)
of the PsA through interstices of stent. Stent acts as barrier and cages coils into PsA, excluding them from patent
vital carotid artery. (From Saad N, Saad W, Rubens DJ. Pseudoaneurysms and the role of minimally invasive
techniques in their management. Radiographics 2005;25:S1739; with permission.)
compared with surgical management [22,25]. Complications associated with endovascular techniques
include intraprocedural rupture of the PsA [24]. In
addition, recanalization of the embolized vessel
and reconstitution of arterial flow to the pseudoaneurysms (delayed failure of embolization) are
possible, though rare [24].
Arteriovenous fistulae
An arteriovenous fistula (AVF) is an abnormal communication between an artery and a vein, and may
be congenital or acquired [26,49]. Acquired AVFs
may occur spontaneously as a result of rupture of
arterial aneurysms into adjacent veins [50], in pa-
Fig. 12. (A) A 61-year-old man underwent prostatectomy and cystectomy with Indiana pouch creation for
treatment of bladder carcinoma. A Jackson-Pratt Drain placed in surgical bed at time of surgery subsequently
eroded into left external iliac artery. Contrast-enhanced CT was obtained showing PsA of left external iliac artery
(arrow head). (B) Digital subtraction angiogram showing PsA (arrow head) before stent-graft (covered stent)
placement. (C ) Digital subtraction angiogram showing successful exclusion of PsA using stent graft. (From Saad N,
Saad W, Rubens DJ. Pseudoaneurysms and the role of minimally invasive techniques in their management.
Radiographics 2005;25:S1739; with permission.)
193
194
Saad et al
tients who present with arteridities (Takayasus arteritis or HIV-related arteritis) and connective tissue
disorders (Ehlers-Danlos syndrome and Marfan syndrome), or as a result of infection or tumor [51,52].
Acquired AVFs, however, are more commonly a result of trauma (penetrating is more common than
blunt) or something iatrogenic, such as surgery,
catheterization, or percutaneous biopsies [5254].
Natural history
AVFs cause left-to-right shunting of blood resulting
in dilatation of the efferent veins [53]. The degree
of shunting and proximity of the AVF to the heart
correlates with the degree of cardiovascular derangement and clinical presentation [54]. Low-flow fistulas usually do not result in any cardiovascular
derangement, are asymptomatic (incidentally diagnosed), and can spontaneously thrombose [5456].
Moderate-flow AVFs cause venous dilatation but
usually do not result in flow-overload and cardiac
dilatation and often present years later [55]. Large
AVFs with high flow cause early left ventricular
hypertrophy and, eventually, cardiac failure [5356].
The left-to-right shunting of arterial blood may lead
to ischemia of the affected limb causing claudication
[51,56] or organ failure [54]. The increased venous
pressure may result in edema of the extremities
[49,51]. Hemorrhage is another clinical presentation
caused by rupture of the AVF [49,53,54]. Paradoxical
pulmonary embolism has also been reported [50].
On physical examination an audible bruit or palpable thrill (in superficial AVFs) may be present
[51,53].
Fig. 13. (A) Longitudinal US image of left superficial femoral artery in patient following motorcycle accident.
Spectral waveform demonstrates abnormal low resistive index, consistent with distal arteriovenous fistula. Normal
extremity waveform should demonstrate high resistive index. (B) Digital subtraction pelvic angiogram demonstrating early filling of left common femoral vein (CFV) in arterial phase from AVF between profunda femoris
artery (PFA) and CFV. Normal right common femoral vein is not seen in arterial phase imaging.
Ultrasound imaging
Gray-scale US often fails to demonstrate the connection between the artery and the vein, especially
if the fistula is small; therefore, the diagnosis may
be missed if color and spectral Doppler are not
used [49]. In peripheral AVFs, the communication
between the vessels is demonstrated as a colored jet
of blood by color or power Doppler [26], often
with aliasing on color Doppler caused by high
velocity flow. The affected vein is often larger
in caliber than the contralateral (unaffected) side
[Fig. 13A, B]. With a high-flow AVF, poor response
to a Valsalva maneuver in the involved vein occurs
[26]. A normal response to the Valsalva maneuver
suggests that the AVF is small with low-flow and
may resolve spontaneously [26]. Spectral Doppler
evaluates the flow patterns from a selected segment
of the AVF vessels [49,57]. Spectral waveforms obtained from the draining veins of the AVF demonstrate pulsatile periodic flow that varies with
the cardiac cycle, and often simulate an artery
[Fig. 14A, B]. The feeding artery shows increased
diastolic blood flow resulting from the high venous outflow with low resistance [26,49]. The low
resistance arterial and pulsatile venous waveforms
are the spectral diagnostic criteria for a systemic
AVF. In visceral AVFs (eg, arterioportal fistula),
the venous pattern includes reversed flow in particular portal vein radicals, and an arterialized
portal vein waveform [Fig. 15AC] [5860]. The
feeding hepatic artery exhibits increased diastolic
flow caused by reduced impedance, which is reflected in a reduced resistive index (peak systolic
velocity end diastolic velocity peak systolic velocity). A reduced resistive index less than 0.5 in
Fig. 14. (A) Axial contrast-enhanced CT image demonstrating increased attenuation in right renal hilum (arrow).
(B) Color flow Doppler image of right renal hilum demonstrating mosaic of colors indicates aliasing and high
velocity flow. Also note large volume of vessels which fill hilum. Spectral waveform shows pulsatile periodic
arterial waveform on both sides of baseline. Artery flows into kidney (above baseline) and high velocity vein
has an arterialized waveform directed below baseline as it flows out into renal hilum; diagnostic of AVF.
195
196
Saad et al
Fig. 15. (A) Gray-scale ultrasound with spectral waveform from right hepatic artery demonstrating low resistive
index of 0.48. (B) Gray-scale ultrasound with spectral waveform from right portal vein demonstrating abnormally
reversed flow direction, out of liver toward porta. (C ) Digital subtraction angiogram image from selective celiac
axis injection demonstrates early filling of portal venous system in hepatic arterial phase denoting an arterioportal
fistula. Right hepatic artery (RHA) and portal vein (PV).
the hepatic artery is usually abnormal and is caused

by ischemia or by arteriovenous shunting (tumor
or AVF).
Three-dimensional US improves localization and
orientation of the AVF relative to surrounding
structures with results comparable with MRA [52].
Intravascular US (IVUS) provides more accurate
delineation of AVFs; however, this is an invasive
modality and its use is reserved for guidance of
endoluminal therapy [61].
Treatment
Low-flow AVFs may be managed expectantly with
close monitoring by duplex US [49,54,56]. Symptomatic AVFs should undergo definitive treatment
[49,54]. As in PsAs, the traditional treatment of
AVFs has been surgical repair [51]; however, over
the past few years, minimally invasive therapy for
AVFs has also evolved to provide an alternate to
surgery [49,51]. The therapeutic options available
today include observation, surgery, US-guided compression, and endoluminal management [49,51].
The goal of treatment for AVFs is to obliterate the
connection between the artery and the vein [53].
The fate of the involved vessels depends on their
expendability. Efforts are made to preserve continuity of vital vessels, whereas expendable vessels
may be sacrificed [53].
Treatment of AVFs results in symptom relief, preservation of function, and, often, reversal of cardiomegaly [53,54].
Surgery
Surgery remains the traditional choice for therapy
and, because of multiple feeding arteries and draining veins, is more often required in AVF repair than
for PsA treatment. Surgical options include vascular
ligation of expendable vessels [54], ligation of the
AVF with preservation of vessel continuity or conduit bypass, and partial or total organ removal
[49,54].
US-guided compression may be used to treat superficial postcatheterization AVFs similar to postcatheterization PsAs [26,56]; however, success rates are
30% or lower [26]. The US probe directly compresses the fistulous track between the feeding
artery and the draining vein of the AVF [26]. Alternating compression cycles and concurrent continuous US monitoring are performed as they are for
PsA therapy.
The same basic principles of endoluminal therapy
for PsAs apply to AVFs. Expendable arteries may be
embolized selectively using coils or other embolic
agents [49,54,61]; however, these coils and agents
carry a substantial risk for inadvertent systemic or
paradoxical pulmonary embolization in large highflow AVFs [54,61]. In such cases, a covered stent
may be placed either in the artery or the vein [61].
If the stent is placed in the vein, the artery must be
embolized selectively to ensure fistula closure,
otherwise high pressure blood leaks around the
stent and back into the vein [58]. IVUS may be
helpful in the planning and deployment of stents
[61]. In general, the risks of percutaneous therapy
are higher and success rates lower in AVFs as compared with PsAs, often caused by multiple feeding
arteries or draining veins, making it more difficult
to isolate and eliminate all the abnormal vessels.
Summary
PsAs and AVFs are common vascular abnormalities.
They are diagnosed more frequently as a result of
increased percutaneous biopsies and procedures
that cause them, and noninvasive imaging with
CTA, US, and MRA, which often detect asymptomatic disease. Conventional angiography, the gold
standard for diagnosis, is an invasive procedure and
every effort should be made to use other noninvasive diagnostic modalities in the initial workup.
The potential complications of many but not all
PsAs and AVFs carry a high morbidity and mortality. Surgery has been the classic treatment of choice;
however, radiology has introduced alternative minimally invasive treatment techniques that carry a
lower morbidity and mortality. A complete workup
of the PsA or AVF and defining the location, surrounding structures, and vascular anatomy are
essential steps in the selection of the treatment technique. The therapeutic options (including observation) should be tailored to the site, rupture risk,
patient comorbidities, and symptoms of the PsA or
AVF in question.
References
[1] Schwartz LB, Clark ET, Gewertz BL. Anastomotic
and other pseudoaneurysms. In: Rutherford RB,
editor. Vascular surgery. 5th edition. Philadelphia: WB Saunders; 2000. p. 75263.
[2] Bromley PJ, Clark T, Weir IH, et al. Radiologic
diagnosis and management of uterine artery
pseudoaneurysm: case report. Can Assoc Radiol J
1997;48(2):11922.
[3] Zimon AE, Hwang JK, Principe DL, et al. Pseu-
197
198
Saad et al
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
doaneurysm of the uterine artery. Obstet Gynecol

1999;94(5 Part 2):82730.
Arata MA, Cope C. Principles used in the management of visceral aneurysms. Tech Vasc Interv
Radiol 2000;3:1249.
Munera F, Soto JA, Palacio D, et al. Diagnosis of
arterial injuries caused by penetrating trauma to
the neck: comparison of helical CT angiography
and conventional angiography. Radiology 2000;
216(2):35662.
Soto JA, Munera F, Morales C, et al. Focal arterial
injuries of the proximal extremities: helical CT
arteriography as the initial method of diagnosis.
Radiology 2001;218(1):18894.
Ahmed A, Samuels SL, Keeffe EB, et al. Delayed
fatal hemorrhage from pseudoaneurysm of the
hepatic artery after percutaneous liver biopsy. Am
J Gastroenterol 2001;96(1):2337.
Kresowik TF, Khoury MD, Miller BV, et al. A
prospective study of the incidence and natural
history of femoral vascular complications after
percutaneous transluminal coronary angioplasty.
J Vasc Surg 1991;13(2):32833.
Kruger K, Zahringer M, Sohngen FD, et al. Femoral pseudoaneurysms: management with percutaneous thrombin injections success rates
and effects on systemic coagulation. Radiology
2003;226(2):4528.
Morgan R, Belli A. Current treatment methods
for postcatheterization pseudoaneurysms. J Vasc
Interv Radiol 2003;14(6):697710.
Brancatelli G, Katyal S, Federle MP, et al. Threedimensional multislice helical computed tomography with the volume rendering technique in the
detection of vascular complications after liver transplantation. Transplantation 2002;73(2):23742.
Crossin JD, Muradali D, Wilson SR. US of liver
transplants: normal and abnormal. Radiographics
2003;23(5):1093114.
Knisley BL, Mastey LA, Collins J, et al. Imaging of
cardiac transplantation complications. Radiographics 1999;19(2):32139.
Kwon JH, Kim GS. Obstetric iatrogenic arterial
injuries of the uterus: diagnosis with US and
treatment with transcatheter arterial embolization. Radiographics 2002;22(1):3546.
Mita T, Arita T, Matsunaga N, et al. Complications of endovascular repair for thoracic and
abdominal aortic aneurysm: an imaging spectrum. Radiographics 2000;20(5):126378.
Busquets AR, Acosta JA, Colon E, et al. Helical
computed tomographic angiography for the
diagnosis of traumatic arterial injuries of the
extremities. J Trauma 2004;56(3):6258.
Sugg SL, Gerndt SJ, Hamilton BJ, et al. Pseudoaneurysms of the intraparenchymal splenic artery
after blunt abdominal trauma: a complication
of nonoperative therapy and its management.
J Trauma 1995;39(3):5935.
Shanmuganathan K, Mirvis SE, Boyd-Kranis R,
et al. Nonsurgical management of blunt splenic
injury: use of CT criteria to select patients for
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
splenic arteriography and potential endovascular

therapy. Radiology 2000;217(1):7582.
Mallek R, Mostbeck G, Gebauer A, et al. A
posttraumatic pseudoaneurysm of the hepatic
artery. Duplex sonographic diagnosis and follow
up in spontaneous thrombosis. Radiology 1990;
30(10):4848.
Soudack M, Epelman M, Gaitini D. Spontaneous
thrombosis of hepatic posttraumatic pseudoaneurysms. sonographic and computed tomographic features. Case report. J Ultrasound Med
2003;22(1):99103.
La Perna L, Olin JW, Goines D, et al. Ultrasoundguided thrombin injection for the treatment of
postcatheterization pseudoaneurysms. Circulation 2000;102(19):23915.
McDermott VG, Shlansky-Goldberg R, Cope C.
Endovascular management of splenic artery aneurysms and pseudoaneurysms. Cardiovasc Intervent Radiol 1994;17(4):17984.
Okuno A, Miyazaki M, Ito H, et al. Nonsurgical management of ruptured pseudoaneurysm in
patients with hepatobiliary pancreatic diseases.
Am J Gastroenterol 2001;96(4):106771.
Guillon R, Gracier JM, Abergel A, et al. Management of splenic artery aneurysms and false
aneurysms with endovascular treatment in 12 patients. Cardiovasc Intervent Radiol 2003;26(3):
25660.
Yamakado K, Nakatsuka A, Tanaka N, et al.
Transcatheter arterial embolization of ruptured
pseudoaneurysms with coils and n-butyl cyanoacrylate. J Vasc Interv Radiol 2000;11(1):6672.
Polak JF. The peripheral arteries. In: Rumack CM,
Wilson SR, Charboneau JW, editors. Diagnostic
ultrasound. 3rd edition. Philadelphia: Elsevier,
Mosby; 2005. p. 9931018.
Tessier DJ, Stone WM, Fowl RJ, et al. Clinical
features and management of splenic artery pseudoaneurysm: case series and cumulative review of
literature. J Vasc Surg 2003;38(5):96974.
Katz DS, Hon M. CT angiography of the lower
extremities and aortoiliac system with a multidetector row helical CT scanner: promise of new
opportunities fulfilled. Radiology 2001;221(1):
710.
Katyal S, Oliver III JH, Buck DG, et al. Detection
of vascular complications after liver transplantation:early experience in multislice CT angiography with volume rendering. AJR Am J
Roentgenol 2000;175(6):17359.
Ricci MA, Trevisani GT, Pilcher DB. Vascular
complications of cardiac catheterization. Am J
Surg 1994;167(4):3758.
Steinkamp HJ, Werk M, Felix R. Treatment of
postinterventional pseudoaneurysms by ultrasound-guided compression. Invest Radiol 2000;
35(3):18692.
Friedman SG, Pellerito JS, Scher L, et al.
Ultrasound-guided thrombin injection is the
treatment of choice for femoral pseudoaneurysms. Arch Surg 2002;137(4):4624.
[33] Kumins NH, Landau DS, Montalvo J, et al. Expanded indications for the treatment of postcatheterization femoral pseudoaneurysms with
ultrasound-guided compression. Am J Surg 1998;
176(2):1316.
[34] Patel JV, Weston MJ, Kessel DO, et al. Hepatic
artery pseudoaneurysm after liver transplantation: treatment with percutaneous thrombin
injection. Case report. Transplantation 2003;
75(10):17557.
[35] Lennox AF, Delis KT, Szendro G, et al. Duplexguided thrombin injection for iatrogenic femoral artery pseudoaneurysm is effective even in
anticoagulated patients. Br J Surg 2000;87(6):
796801.
[36] Badran MF, Gould DA, Sampson C, et al. Transluminal occlusion of a pseudoaneurysm arising
from a thoracic aortic graft patch using catheter
delivery of thrombin. J Vasc Interv Radiol 2003;
14(9, Part 1):12015.
[37] Thalhammer C, Kirchherr AS, Uhlich F, et al.
Postcatheterization pseudoaneurysm and arteriovenous fistulas: repair with percutaneous implantation of endovascular covered stents. Radiology
2000;214(1):12731.
[38] Bergert H, Hinterseher I, Kersting S, et al.
Management and outcome of hemorrhage due
to arterial pseudoaneurysms in pancreatitis. Surg
2005;137(3):3238.
[39] Sadiq S, Ibrahim W. Thromboembolism complicating thrombin injection of femoral artery
pseudoaneurysm: management with intraarterial
thrombolysis. Case report. J Vasc Interv Radiol
2001;12(5):6336.
[40] Fellmeth BD, Roberts AC, Bookstein JJ, et al.
Postangiographic femoral artery injuries: nonsurgical repair with US-guided compression.
Radiology 1991;178(3):6715.
[41] Paulson EK, Sheafor DH, Kliewer MA, et al.
Treatment of iatrogenic femoral arterial pseudoaneurysms: comparison of US-guided thrombin injection with compression repair. Radiology
2000;215(2):4038.
[42] Gale SS, Scissons RP, Jones L, et al. Femoral pseudoaneurysm thrombinjection. Am J Surg 2001;
181(4):37983.
[43] Brophy DP, Sheiman RG, Amatulle P, et al.
Iatrogenic femoral pseudoaneurysms: thrombin
injection after failed US-guided compression.
Radiology 2000;214(1):27882.
[44] Cope C, Zeit R. Coagulation of aneurysms by
direct percutaneous thrombin injection. AJR Am
J Roentgenol 1986;147(2):3837.
[45] Kang SS, Labropoulos N, Mansour MA, et al.
Expanded indications for ultrasound-guided
thrombin injection of pseudoaneurysms. J Vasc
Surg 2000;31(2):28998.
[46] Kurz DJ, Jungius K, Luscher TF. Delayed femoral
vein thrombosis after ultrasound-guided throm-
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[55]
[56]
[57]
[58]
[59]
[60]
[61]
bin injection of a postcatheterization pseudoaneurysm. Case report. J Vasc Interv Radiol 2003;
14(8):106770.
Sheldon PJ, Oglevie SB, Kaplan LA. Prolonged
generalized urticarial reaction after percutaneous
thrombin injection for treatment of a femoral
artery pseudoaneurysm. Case report. J Vasc Interv
Radiol 2000;11(6):75961.
Krysl J, Kumpe D. Embolization agents: a review.
Tech Vasc Interv Radiol 2000;3:15861.
Aziz N, Lenzi TA, Jeffrey Jr RB, et al. Postpartum
uterine arteriovenous fistula. Obstet Gynecol
2004;103(5, Part 2):10768.
Lay JP, Swainson CJ, Sukumar SA. Paradoxical
pulmonary embolism secondary to aortocaval
fistula: diagnosis by helical CT. Br J Radiol 1999;
72(857):5079.
Wang EA, Lee MH, Wang MC, et al. Iatrogenic
left iliac-caval fistula: imaging and endovascular
treatment. AJR Am J Roentgenol 2004;183(4):
10324.
Nair R, Chetty R, Woolgar J, et al. Spontaneous
arteriovenous fistula resulting from HIV arteritis.
J Vasc Surg 2001;33(1):1867.
Puppinck P, Chevalier J, Ducasse E, et al. Connection between a long-standing traumatic arteriovenous fistula and development of aneurysmal
disease. Ann Vasc Surg 2004;18(5):6047.
Chauvapun JP, Caty MG, Harris LM. Renal
arteriovenous aneurysm in a 4-year-old patient.
J Vasc Surg 2005;41(3):5358.
Hartung O, Garcia S, Alimi YS, et al. Extensive
arterial aneurysm developing after surgical closure of long-standing post-traumatic popliteal
arteriovenous fistula. J Vasc Surg 2004;39(4):
88992.
Ray C. Complications of percutaneous transluminal angioplasty. In: Darcy MD, Vedantham S,
Kaufman JA, editors. SCVIR syllabus: peripheral
vascular interventions. 2nd edition. Fairfax (VA):
The Society of Cardiovascular & Interventional
Radiology; 2001. p. 8190.
Rose SC, Nelson TR. Ultrasonographic modalities
to assess vascular anatomy and disease [review].
J Vasc Interv Radiol 2004;15(1, Part 1):2538.
Chavan A, Harms J, Picjlmayr R, et al. Transcatheter coil occlusion of an intrahepatic arterioportal fistula in a transplanted liver. Bildgebung
1993;60:2158.
Strodel E, Eckhauser FE, Lemmer JH, et al. Presentation and perioperative management of
arterioportal fistulas. Arch Surg 1987;122:56371.
Glockner JF, Forauer AR. Vascular or ischemic
complications after liver transplantation: pictorial essay. AJR Am J Roentgenol 1999;173:10559.
Sprouse II LR, Hamilton Jr IN. The endovascular
treatment of a renal arteriovenous fistula: placement of a covered stent. J Vasc Surg 2002;36(5):
10668.
199
201
ULTRASOUND
CLINICS
Doppler Imaging of the Uterus

and Adnexae
Shweta Bhatt,
&
&
&
&
&
&
&
MD,
Vikram S. Dogra,
Normal uterine and ovarian vascular

anatomy
Uterine pathologies
Leiomyoma
Color flow Doppler to assess leiomyomas
Vascular bridging sign
Vascular pedicle sign
Uterine artery embolization
Pre-embolization evaluation
Postembolization evaluation
Adenomyosis
Uterine artery embolization in
adenomyosis
Endometrial polyps
Endometrial carcinoma
Vascular abnormalities of the
endometrium
Arteriovenous malformation
Uterine artery aneurysm
Uterine artery pseudoaneurysm
Pelvic sonography constitutes a major proportion of the routine and the emergency ultrasound
(US) examinations performed. Application of color
Doppler US in addition to gray-scale imaging helps
in the characterization of uterine and adnexal
masses. Color flow Doppler imaging has a definitive role in the diagnosis and management of uterine leiomyomas (including arterial embolization),
ovarian torsion, placental abnormalities, vascular
malformations, and uterine artery aneurysms. This
article describes the normal CFD features of uter-
&
&
&
&
&
&
&
&
&
MD*
Gestational trophoblastic disease

Molar pregnancy
Persistent trophoblastic neoplasia
(invasive mole)
Abnormal placentation
Placenta accreta, increta, and percreta
Vasa previa
Retained products of conception
Ovarian pathologies
Torsion
Twisted vascular pedicle/whirlpool sign
Subtorsion of the ovary
Massive ovarian edema
Ectopic pregnancy
Ovarian vein thrombosis
Ovarian malignancy
Summary
References
ine and ovarian vessels and the usefulness of color

flow Doppler sonography in the evaluation of uterine and ovarian abnormalities.
Normal uterine and ovarian vascular

anatomy
The uterus derives its main arterial supply from the
uterine artery, a branch of the anterior division of
the internal iliac artery [Fig. 1]. Transvaginal (TV)
and transabdominal color Doppler sonography
Department of Radiology, University of Rochester School of Medicine and Dentistry, 601 Elmwood Avenue,
Box 648, Rochester, NY 14642, USA
E-mail address: Vikram_Dogra@URMC.Rochester.edu (V.S. Dogra).
doi:10.1016/j.cult.2005.09.010
202
Bhatt & Dogra
Fig. 1. Arterial blood supply of the uterus.
(CDS) can depict the origin of the uterine artery

that runs medially on the levator ani to a point
approximately 2 cm from the cervix where it crosses
over the ureter to become medial to it [1]. The
uterine artery ascends along the margin of the
uterus in the broad ligament up to the uterine cornua where it courses laterally to the hilum of the
ovary to join the ovarian artery. Branches arise from
the main uterine arteries and penetrate the outer
myometrium, forming a spoke wheel configuration
of arcuate vessels [2]. Right and left uterine artery
branches anastomose in the midline through the
arcuate vessels, located between the outer and middle layers of the myometrium. The radial branches
of the uterine artery course through the myometrium and form the basal and spiral arterioles
within the endometrium [2]. The uterine plexus
of veins accompanies the arcuate arteries and is
larger than the associated arterial channels [1].
Doppler waveforms of uterine artery flow typically
have a high-velocity, high-resistance pattern [3]
with an identifiable dicrotic notch [Fig. 2]. Persistence of the dicrotic notch beyond 23 weeks of
pregnancy is suggestive of intrauterine growth retar-
dation. Flow characteristics in the arcuate artery

may also vary during pregnancy with lower resistive
indices in placental than in nonplacental sites [4].
The ovary has a dual arterial supply through the
ovarian artery and the adnexal branches of the uterine artery [Fig. 3]. Ovarian arteries arise laterally
from the aorta inferior to the origin of the renal
arteries. They cross the external iliac arteries and
veins at the pelvic brim and course medially within
the suspensory ligament of the ovary and then
pass posteriorly in the meso-ovarium [1]. The right
ovarian (gonadal) vein drains into the inferior vena
cava (IVC), and the left ovarian vein drains into the
left renal vein. Doppler waveforms of the ovarian
arteries show physiologic variations within the menstrual cycle. The dormant ovary in the menstrual
cycle has a low-flow (high resistive index [RI]) pattern throughout the menstrual cycle, whereas the
dominant ovary demonstrates a high-flow (low RI)
pattern [Fig. 4]. This pattern coincides with the
luteinizing hormone surge that increases through
the periovulatory period and remains at this level
for 4 to 5 days after ovulation into the luteal phase
of the cycle. Ovarian blood flow then gradually
returns to a high-resistance pattern during the menstrual period [3,5,6].
On TV color and pulsed Doppler, the uterine artery
can be sampled along the lateral aspect of the
uterine body at the level of the corporocervical
junction. The arcuate branches can be sampled anywhere within the myometrium [4,7].
The ovarian artery should be sampled in the
infundibulopelvic ligament or from the intraovarian vessels for the best quantitative information [5].
Resistive and pulsatility indices are used to quantitate the blood flow in the ovary. RI is defined as
Fig. 2. Normal uterine arterial Doppler. (A) Longitudinal endovaginal gray-scale sonogram of uterus (UT) demonstrates normal arcuate vessels (arrows). (B) CFD image with spectral analysis of uterine artery reveals normal highvelocity, high-resistance waveform with dicrotic notch (arrow).
Doppler Imaging of the Uterus & Adnexae
Fig. 3. Normal arterial and venous ovarian Doppler. (A) Normal arterial and (B) venous Doppler waveforms of the
ovarian vessels.
(peak systolic velocity [PSV] end diastolic velocity

[EDV]) PSV and the pulsatility index (PI) is
defined as (PSV EDV) mean velocity.
Color Doppler parameters should be optimized
to detect flow, to avoid a mistaken diagnosis of
torsion or thrombosis. These parameters are summarized in Table 1. Power Doppler may be used for
better detection of flow. Pros and cons of the use of
power Doppler compared with color Doppler are
listed in Box 1.
Uterine pathologies
Leiomyoma
Uterine leiomyomas, also called myomas or fibroids, are the most common gynecologic disease
involving the uterus; they occur in 20% to 25% of
women over 30 years of age [8,9].
Leiomyomas are classified depending on their

location in the uterus. Subserosal leiomyomas are
the exophytic fibroids that protrude from the outer
surface of the uterus. Leiomyomas create a uterine
contour abnormality and may mimic an adnexal
mass. They may be pedunculated; if they are large
they are prone to torsion [10]. Intramural leiomyomas, the most common type, are located within
the myometrium. Submucosal leiomyomas are
subendometrial and may protrude into the endometrial cavity. Most fibroids are hypoechoic, wellcircumscribed masses, often with attenuation and
shadowing. As they become more complex, their
interfaces and, thus, their echogenicity increases.
When uterine leiomyomas increase in size, they
may outgrow their blood supply and undergo various types of degeneration, such as hyaline or myxoid degeneration, calcification, cystic degeneration,
and red degeneration [11]. The sonographic ap-
Fig. 4. Cyclical normal variation on ovarian arterial Doppler. Spectral Doppler waveform of (A) right ovary
(dormant ovary for this menstrual cycle), demonstrates low-flow (high-resistance, RI = 0.75) pattern compared
with high-flow (low-resistance, RI = 0.375) pattern seen in left ovary (B) (dominant ovary for this menstrual cycle).
203
204
Bhatt & Dogra
Table 1: Optimizing color flow Doppler image

Color box
Doppler gain
Color scale
(PRF)
Beam steering
Gate size
(sample
volume)
Wall filter
Focal zone
Keep color box small to

improve frame rate and
obtain better color resolution
Set just below noise level
Low PRF is more sensitive to
low-flow vessels but may lead
to aliasing
Adjust to obtain satisfactory
vessel angle (ie, not 90)
Set sample volume to
correct size, usually twothirds of the vessel lumen
High filter cuts out noise
and slower flow states.
Keep filter at 50100 Hz
Color flow image is optimized
at focal zone, so place
where vessels are expected
Abbreviation: PRF, pulse repetition frequency.
pearance of the fibroids may vary accordingly, demonstrating hypo- to anechoic areas as a result of
cystic degeneration or bright hyperechoic areas
with sharp acoustic shadowing in calcified leiomyomas [10].
Color flow Doppler to assess leiomyomas

Depiction of fibroid vascularity by CDS improves
the delineation of the size, location, and extent of
myometrial involvement [12]. Leiomyomas usually
demonstrate a peripheral rim of vascularity in the
pseudocapsule (covering almost three fourth circumference); this feature helps to identify isoechoic
intramural myomas and aids in the diagnosis of
subserosal myomas [Fig. 5] [2]. In addition, demonstration of a vascular bridging sign (VBS) or a
vascular pedicle between the uterus and the periuterine mass helps to discriminate a subserosal
leiomyoma from a true adnexal mass. Color flow
Doppler US also has a significant role in evaluating
patients for uterine artery embolization (UAE) preand postprocedure.
Box 1: Power Doppler versus color Doppler

1. No directional information is obtained with
power Doppler.
2. Power Doppler is more susceptible to motion
artifact than color Doppler.
3. Power Doppler has better sensitivity to low
flows than CFD.
4. Power Doppler has less temporal resolution
as a result of longer acquisition times (more
frame averages).
Fig. 5. Power Doppler image of uterus demonstrates

leiomyoma (M) arising from the uterus (UT), which
reveals peripheral draping pattern of vascularity typical of leiomyoma.
Vascular bridging sign

A large subserosal leiomyoma may mimic a solid
or heterogeneous ovarian mass, whereas the former
is benign, the latter may be malignant, and the
patient management is different. The presence of
multiple vessels between the uterus and the presumed adnexal mass is called the VBS. The VBS is
secondary to recruitment of multiple vessels feeding the exophytic uterine fibroid [Fig. 6] and confirms the origin of the vascular blood supply to
uterine fibroids from the uterine arteries, implying
that the tissue is uterine, not ovarian. Kim and
colleagues [13] first demonstrated the VBS in MRI
of leiomyomas whereby the bridging vessels appeared as tortuous channels with no signal connecting the uterus and the pelvic mass. In another
study comparing CDS and MRI for differentiating
subserosal myomas from extra-uterine tumors, the
VBS was present in 39 out of 41 subserosal myomas and in 3 out of 27 extra-uterine tumors when
using both modalities [14]. Color and power Doppler US alone have 100% sensitivity and 92% specificity for differentiating subserosal leiomyomas
from extra-uterine tumors [14].
Vascular pedicle sign

Color flow Doppler may also demonstrate a solitary vessel originating from the uterine artery that
supplies the subserosal leiomyoma, called the vascular pedicle sign [Fig. 7]. As with the VBS, demonstrating a common source of blood supply between
the mass and the uterus implies the origin of the
mass is from the uterus.
Uterine artery embolization

UAE of uterine fibroids is a minimally invasive alternative to surgical therapy (hysterectomy or myomectomy). The use of UAE in the treatment of
Fig. 6. VBS. (A) Longitudinal gray-scale sonogram of uterus demonstrates mass (M) adjacent to uterus (UT).
Corresponding color (B) and spectral Doppler (C ) demonstrate vascular bridge between mass and uterus, which
has identical spectral waveform as uterine artery with dicrotic notch (arrow), suggesting common blood supply.
This was confirmed to be exophytic leiomyoma.
uterine fibroids as an alternative to surgical treatment was first proposed by Ravina and colleagues
[15] in 1995. Originally, UAE was used before
myomectomy to reduce intraoperative bleeding
and also to treat post-partum hemorrhage [16]. As
a single therapeutic regimen for fibroids, UAE has
been reported to be a safe and successful method to
reduce fibroid-induced symptoms including bleeding, urinary frequency, and pelvic pain [1719]. Suc-
cessful pregnancies have been reported in patients

who have undergone UAE [19]; however, these pregnancies were shown to have an increased risk for
preterm delivery and malpresentation [20].
Pre-embolization evaluation
Imaging parameters used to predict the effectiveness of UAE, are baseline uterine volume, baseline
leiomyoma volume and location, and the number
Fig. 7. Vascular pedicle sign. (A) Transverse gray-scale sonogram of uterus demonstrates mass (M) adjacent to
uterus (UT). (B) Corresponding Doppler image demonstrates vascular pedicle sign (arrow).
205
206
Bhatt & Dogra
Fig. 8. UAE. (A) Power Doppler imaging of uterus before embolization shows fibroid (arrows) nearly isoechoic
with uterus (UT) and has marked peripheral vascularity (arrowheads). (B) Three months after UAE, leiomyoma
(arrows) has no vascularity and has cystic necrosis (arrowheads). (From Nicolson TA, Pelage JP, Ettles DF. Fibroid
calcification after uterine artery embolization: ultrasonographic appearance and pathology. J Vasc Interv Radiol
of leiomyomas. Spies [21] demonstrated that myomas with a larger baseline volume showed less
symptomatic improvement and less reduction in
size after UAE. The number of leiomyomas is not
considered a strong predictor of efficacy of UAE;
however, in Spies study, patients who presented
with two to five leiomyomas showed a better symptomatic response to therapy than patients who
presented with a single leiomyoma or more than
five leiomyomas [21]. The size of uterine leiomyomas remains controversial as a determining factor
in the outcome of UAE; however, a size greater than
8 cm is a predictor of UAE failure. Submucosal
leiomyomas have the best outcome after UAE
[21]. Pedunculated and subserosal fibroids are considered unsuitable for UAE [22].
Doppler examination of overall tumor vascularity predicts therapeutic outcome. Hypervascular
fibroids show more reduction in size than isovascular and hypovascular fibroids [12]. Doppler flow
measurements can be used to assess UAE outcome
also. McLucas and colleagues [23] reported that
patients presenting with fibroids that had a PSV
greater than 64 cm/s were more likely to have
UAE failure, emphasizing the use of Doppler flow
studies in screening for UAE.
Postembolization evaluation
MRI and color Doppler US have proven to be
excellent modalities for postprocedural follow-up
after UAE [24,25]. Follow-up sonographic examination is performed at 1, 3, 6, and 9 months, and
1 year after UAE [17]. Fibroid volume (size) and
vascularity have been used as parameters to evaluate the therapeutic efficacy of UAE [26]. CDS de-
picted that along with the reduction in the fibroid

volume, disappearance of the intrafibroid vessels
occurred without affecting uterine vascularity [27].
Shrinkage of the fibroids continues to take place
over a period of several months, peaking between
3 and 6 months after UAE, with measurable shrinkage sometimes noted for as long as 1 year [28,29].
Weintraub and colleagues [30] reported that postprocedure sonographic imaging showed decreased
uterine size and echogenicity. Concomitantly, color
Doppler imaging showed a marked decrease in
the blood flow to the leiomyoma [Fig. 8]. US
also may depict peripheral calcification in the fibroid after UAE, occurring as a result of aggregation
of the polyvinyl alcohol particles in the peripheral
tumoral arteries [31]. Gadolinium-enhanced MRI
is a useful diagnostic technique to assess uterine fibroids after UAE because it can evaluate the degree
of infarction in the treated tumors [32].
Adenomyosis
Adenomyosis is a benign condition of the uterus
characterized by the presence of endometrial glands
and stroma in the myometrium, identified on
imaging by the distortion of the endometrialmyometrial junctional zone and the asymmetry of
the uterus [33]. Diffuse adenomyosis appears on
US as uterine enlargement with a globular configuration, heterogeneous-appearing myometrium, and
subendometrial cysts [3436]. The RI in adenomyosis is usually less than 0.7, suggesting an abnormal flow (normal RI = > 0.7) [37]. Unlike myomas,
which have a peripheral blood supply, adenomyosis may be distinguished on imaging by its rain-
Fig. 9. Adenomyosis. Spectral Doppler evaluation of

adenomyosis reveals globular configuration of uterus
with increased central vascularity. RI in this patient
was 0.54.
drop appearance of blood flowing through the

center of the tissue [Fig. 9]. Surrounding tissue unaffected by adenomyosis remains mostly hypovascular [23]. The PI of the arteries within or around
the adenomyosis tissue is usually greater than 1.17,
whereas leiomyomas have a PI of less than or equal
to 1.17 [38]. The overall sensitivity and specificity
of TV US for the diagnosis of adenomyosis is considered to be 80% to 86% and 50% to 96%, respectively [39].
Uterine artery embolization in adenomyosis

UAE has been proposed as an alternative to surgery
for patients who present with menorrhagia as a
result of adenomyosis. The success rate of UAE
for controlling long-term symptoms of menorrhagia is higher (50%) in patients presenting with
coexisting myomas with adenomyosis but less
than in myomas alone (90%) [40]. Doppler measurement of PSV greater than 64cm/s is predictive of
UAE failure in adenomyosis and leiomyomas [23].
Significant improvement of clinical symptoms following UAE correlates well with gray-scale US and
CDS findings. Clinical improvement is associated
with decreased uterine size, decreased vascularity,
and reduced junctional zone thickness [41]. A decrease in the vascularity of normal myometrium
and the adenomyosis tissue is observed during
the initial, 1-week, postembolization period, with

restoration of normal myometrial vessels on the
3-month follow-up imaging [42]. Although the
short-term results (36 months) of embolization
seem to be good [43,44], the mid-term (2 years)
results are not as encouraging [45]. In patients who
present with pure adenomyosis, only 56% of patients demonstrated a complete resolution of adenomyosis at the 24-month follow-up after UAE as
reported by Pelage and colleagues [45]. Patients
who present with failed UAE in adenomyosis usually undergo hysterectomy [46].
Endometrial polyps
Endometrial polyps are intracavitatory lesions seen
in peri- and postmenopausal women who present
often with uterine bleeding. Polyps may be pedunculated, broad-based, or have a thin stalk. Twenty
percent of these polyps are multiple. The diagnosis of an endometrial polyp is considered when
hyperechoic endometrial thickening of greater than
5 mm is observed on US. Color Doppler or power
Doppler imaging enable easy detection of endometrial polyps by demonstrating a single feeding artery
known as the pedicle sign [Fig. 10] [47]. Observation of the feeding artery in a polyp may help to
distinguish endometrial polyp from a submucosal
fibroid [48]. Timmerman and colleagues [49] defined
the pedicle sign as the presence of an artery within
the central portion of the polyp. Using these criteria, the sensitivity of the pedicle artery sign in
diagnosing an endometrial polyp on endovaginal
sonography was 76%, the specificity was 95%,
and the negative predictive value was 94% [49].
In more recent study comparing transvaginal color
Doppler (TVCD) with sonohysterography (SHG), the
sensitivity and specificity for TVCD using the pedicle
artery sign for identifying endometrial polyps, was
95% and 80%, respectively. The specificity was
low because of two false negative cases. Overall,
TVCD has comparable efficacy to SHG in diagnosing
endometrial polyps [50].
Fig. 10. Pedicle artery sign in endometrial polyp. (A) Coronal section of uterus on endovaginal sonography,
demonstrates endometrial mass lesion (arrowhead) surrounded by fluid within endometrial cavity (arrow).
Corresponding CFD image (B) reveals solitary vessel (arrow) supplying polyp (arrowhead).
207
208
Bhatt & Dogra
Endometrial carcinoma
Endometrial carcinoma is the most common gynecologic malignancy in North America, occurring
most commonly in postmenopausal women and
manifesting as uterine bleeding, with increased incidence in women on replacement hormonal therapy.
Histologically, almost all endometrial carcinomas
are adenocarcinomas; typical endometrioid adenocarcinomas are the most common subtype of
endometrial carcinomas followed by adenocarcinomas with squamous elements. Other less common
subtypes include clear cell, serous, secretory, mucinous, and squamous carcinomas, but imaging characteristics cannot predict the histology of these
cancers [51].
Sonographically, a considerable overlap exists
between the morphologic characteristics of benign
and malignant endometrial neoplasms. A thickened endometrium (> 8 mm) in a postmenopausal
woman must be considered malignant until proven
otherwise. Malignancy is considered more likely
when the endometrial thickening or mass is of
heterogeneous echotexture with irregular borders
[51,52]. Numerous studies on the use of color
Doppler imaging in endometrial neoplasms have
yielded variable results.
CDS depicts vessels in benign and malignant endometrial lesions. Endometrial carcinomas show
an increased number of feeding vessels with moderate vascularity in the mass lesion [5355]. Irregular and randomly dispersed vessels with complex
branching in an endometrial lesion also favor malignancy [56]. Doppler indices may have a significant
role in suggesting malignancy. The mean RI in the

uterine artery in a normal postmenopausal woman,
is 0.93 plus or minus 0.09 [57]. Low-impedance
flow with a RI of less than 0.5 and a PI of less than
1, have been positively associated with the presence of malignancy [58]. Significantly lower RIs
are noticed in endometrial cancers of higher histologic grade and those with lymph node metastasis
[59]. An intratumoral RI of less than 0.4 has a high
incidence of pelvic nodal metastasis [60,61]. A combination of transvaginal sonography and CDS can
be useful in differentiating benign and malignant
endometrial lesions; however, endometrial biopsy
still remains the gold standard for diagnosing endometrial carcinoma.
Vascular abnormalities of the endometrium

Uterine vascular lesions are rare but may be lifethreatening. They usually manifest as unexplained
vaginal bleeding, occasionally massive in amount.
Color Doppler US with spectral analysis is helpful in diagnosis.
Arteriovenous malformation
Uterine arteriovenous malformations (AVMs) consist of a vascular plexus of arteries and veins without an intervening capillary network. They are
located in the myometrium and do not regress
spontaneously with time [52,62]. AVMs may be
congenital but are more commonly acquired following uterine trauma (eg, curettage, pelvic surgery,
Fig. 11. (A) Uterine AVM. CFD image of uterus, demonstrates focal area of mosaic red and blue pattern of color
signal in uterus. Corresponding spectral Doppler waveform (B) reveals a high velocity venous waveform suggestive
of an arterio-venous malformation.
and previous treatment for gestational trophoblastic disease [GTD]). Color Doppler US in the sonographic evaluation of any unexplained cystic mass
in the uterus significantly increases the detection
rate of uterine AVMs [63].
The gray-scale US morphology of uterine AVM is
nonspecific and may present as a hypoechoic or
heterogeneous area or as multiple contiguous distinct anechoic areas in the myometrium. Color
Doppler findings are more consistent and specific
and include intense multidirectional turbulent flow
typified by aliasing [Fig. 11] [52,64,65]. Spectral
analysis characteristically depicts a low-resistance,
high-velocity arterial flow with a low RI in the
range of 0.25 to 0.55 [62]. PSV recorded within
the AVM is also usually high in the range of 40
to 100 cm/s [62,65,66]. Three-dimensional power
Doppler sonography provides additional use in the
evaluation of uterine vascular AVMs by depicting a
clearer view of the orientation of the tortuous vessels comprising the AVM. Doppler US also is used
to monitor the AVMs for response or recurrence
after embolization [65].
Uterine artery aneurysm

Uterine artery aneurysms are rare. A true uterine
artery aneurysm is usually congenital and may
manifest as uterine bleeding from rupture during
pregnancy or in the puerperium [64]. On gray-scale
US, a uterine artery aneurysm is a pulsating anechoic structure in the myometrium. Doppler US
diagnoses the aneurysm with high sensitivity and
specificity by depicting the dilatation of the uterine
artery with its characteristic arterial flow pattern
[64]. UAE is an effective treatment modality for
uterine artery aneurysms and CDS can be used for
monitoring postembolization.
Fig. 12. Pseudoaneurysm of uterine artery. (A) Longitudinal section of uterus on endovaginal sonography reveals
retroverted uterus with absent endometrial lining (arrows). Corresponding color (B) and spectral (C ) Doppler
images demonstrate focal area of color signal (arrow) in endometrium (aliasing was observed on real-time
imaging) with to-and-fro spectral waveform suggestive of pseudoaneurysm resulting from thermal ablation.
209
210
Bhatt & Dogra
Uterine artery pseudoaneurysm

A pseudoaneurysm is an extraluminal collection
of blood with turbulent flow that communicates
with the parent vessel through a defect in the arterial wall [67]. A pseudo-aneurysm differs from a
true aneurysm because of the absence of the
three layers of a vessel wall [6870]. Uterine artery pseudoaneurysm is a rare vascular abnormality
but is seen more commonly than an aneurysm. It
usually arises as a complication after pelvic surgery
or trauma and can be diagnosed by arteriography
or, noninvasively, by CDS. CDS shows a bloodfilled cystic structure with varying colors caused
by the swirling movement of the arterial blood
in different directions [Fig. 12]. Spectral analysis
within the sac demonstrates turbulent multidirectional arterial flow. In the neck of the pseudoaneurysm, the blood flowing into the aneurysm
in systole and out in diastole creates a characteristic bidirectional flow spectrum [64,65]. Because of
the small size of the uterine artery, it may be diffi-
cult to visualize the neck of the pseudoaneurysm,

and in such instances, arteriography is more useful [68].
Gestational trophoblastic disease

Molar pregnancy
GTD is a spectrum of disorders that show abnormal trophoblastic proliferation with a progressive
malignant potential [71,72]. GTD includes molar
pregnancy (MP) (complete and partial), invasive
mole, choriocarcinoma, and placental site trophoblastic tumors (PSTT).
MP is the most common and benign form of
GTD. Depending on the severity of trophoblastic
proliferation and the presence or absence of fetal
tissue, it may be complete or partial MP [72]. Typically, gray-scale sonography reveals abnormal
hyperechoic placental tissue with cystic grapelike
cavities that fill the uterine cavity [73]. Proliferation
of the trophoblastic tissue with abnormal vascular
Fig. 13. Hydatiform mole. (A) Longitudinal gray-scale sonography of uterus demonstrates large hyperechoic mass
(arrows) with cystic spaces (arrowheads) filling uterine cavity. Corresponding power (B) and spectral (C ) Doppler
images reveal increased vascularity with high-velocity, low-resistance waveform pattern with RI of 0.35.
shunting (or abnormal vessels) results in a highvelocity, low-impedance flow on Doppler US in the
first and early second trimesters [Fig. 13] [71]. The
RI usually measures less than 0.4 compared with
0.66 in normal pregnancy [71].
Doppler findings confirm the diagnosis combined with the elevated human chorionic gonadotropin (hCG) level and the gray-scale US findings.
Persistent trophoblastic neoplasia

(invasive mole)
Persistent trophoblastic neoplasia (PTN) is a broad
term that includes invasive mole (the most common), choriocarcinoma, and PSTT; it is suggested
clinically by a persistent elevation of the hCG after
evacuation of MP. Sonographic and Doppler findings cannot differentiate between the three entities.
The most commonly described sonographic findings in PTN are focal echogenic myometrial nodules or a bulky enlarged uterus [7476]. Color
Doppler interrogation reveals marked hypervascularity of the invasive trophoblast with extensive
color aliasing within the myometrium. Spectral
tracings reveal low-resistance arterial flow within
the myometrium suggesting an invasion of the trophoblastic disease [7779]. Doppler indices in PTN
include a high PSV, usually greater than 50cm/s,
and a low RI, often less than 0.4 (normal myometrial blood flow PSV is less than 50 cm/s with an RI
of approximately 0.60.7) [72,77]. Doppler imaging may define the extent of invasion in an invasive
mole and also its subsequent response to chemotherapy [80]. Although the trophoblastic signals on
spectral Doppler are not unique to PTN, clinical
correlation with an elevated hCG level and with
gray-scale sonographic findings are virtually diagnostic of PTN.
Abnormal placentation
Placenta accreta, increta, and percreta
Placenta accreta is defined as a condition in which
the placental tissue extends from the endometrium
into the myometrium (minimal invasion). Placenta
increta refers to the extension of the placenta
deeper into the myometrium but confined to the
myometrium. Placenta percreta is the severest form
with extension of the placenta beyond the myometrium. Most of the published literature refers to
these abnormalities collectively as placenta accreta
[81,82]. Previous uterine surgery is a predisposing
factor for these abnormal placental attachments
[83]. The prenatal diagnosis of abnormal placentae
is critical to avoid postpartum hemorrhage.
Color flow Doppler of placenta accreta demonstrates several placental lacunae with a highly pul-
Fig. 14. Placenta accreta. Longitudinal CFD image of

placenta (P) demonstrates thinning of retroplacental
hypoechoic zone (arrowheads). CFD US demonstrates
enlarged abnormal vessels (arrows) invading adjacent
uterus (UT). (From Taipale P, Orden MR, Berg M, et al.
Prenatal diagnosis of placenta accreta and percreta
with ultrasonography, color Doppler, and magnetic
resonance imaging. Obstet Gynecol 2004;104:53740;
with permission.)
satile venous flow pattern that extends from the

placenta into the surrounding myometrium. Progressive thinning also occurs and disappearance of
the retroplacental hypoechoic zone representing
the decidua basalis [Fig. 14] [8487]. Placenta accreta is found commonly in association with placenta previa [86,88]. On postpartum US, cessation
of blood flow between the basal placenta and
the myometrium is the sonographic hallmark of
normal placental separation. Persistent blood flow
demonstrated by CDS is suggestive of placenta
accreta. [89]. Color Doppler has 100% sensitivity
and 94% specificity to detect placenta accreta [86].
Placenta increta appears on US as an echogenic
mass within the myometrium; isoechoic to the
placenta. Multiple large vessels exist within the
mass, which does not extend to the serosal margin
of the uterus [90]. Placenta percreta is diagnosed by
the interruption of the hyperechoic line between
the uterine serosa and the bladder and by the presence of multiple lacunae in the myometrium. Placent percreta has the worst prognosis. A patient
death caused by air embolism has been described
with placenta percreta extension into the bladder
[91]. Placenta percreta may more commonly result
in uterine rupture [92,93].
Vasa previa
Vasa previa is defined as umbilical vessels that
cross the internal os and are not supported by
the placenta or the umbilical cord [Fig. 15]. Fail-
211
212
Bhatt & Dogra
multiple pregnancies, and in vitro fertilization

[9698]. Documentation of the placental cord
insertion should be a part of the standard obstetric
examination to identify those at risk for vasa previa
[99]. Judicious use of CDS in these patients can help
establish the antenatal diagnosis of vasa previa and
reduce mortality by performing elective cesarean
section. Transperineal sonography with Doppler
evaluation may also identify vasa previa [100].
Retained products of conception
Fig. 15. Vasa previa. Sagittal CFD image of lower

uterine segment demonstrates posteriorly located
marginal placenta (P). Vessels (red and blue) (arrowheads) are seen communicating with accessory placental lobe (AP) across internal os (arrow points to cervix.
H = fetal head). (From Taipale P, Orden MR, Berg M,
et al. Prenatal diagnosis of placenta accreta and percreta with ultrasonography, color Doppler, and magnetic resonance imaging. Obstet Gynecol 2004;104:
53740; with permission.)
ure to diagnose vas previa antenatally results in

catastrophic hemorrhage following rupture of the
membranes with near 100% mortality of the fetus
[82]. Vasa previa is associated strongly with unusual placental findings, such as a bilobed placenta,
succenturiate lobe, placenta previa, or a velamentous cord insertion [94,95]. Other risk factors include low-lying placenta in the second trimester,
Retained products of conception (RPOC) is a wellknown complication of spontaneous or induced

abortion. The diagnosis usually is based on the
sonographic appearance of intrauterine echogenic
material. A high false-positive rate is reported from
clinical and sonographic diagnosis, leading to repeated curettage of the endometrial cavity, which
has immediate and long-term complications, such
as perforation, bleeding, infertility and adhesions
[101]. Although a focal echogenic mass in the endometrium on gray-scale US is the most useful finding
to predict the presence of RPOC [102,103], such a
mass may also represent a blood clot or residual
decidua. Doppler US can be helpful these situations. Depiction of flow in the echogenic mass
within the endometrium favors RPOC and excludes
a blood clot [104]. High-velocity, low-resistance
flow on color Doppler US is suspicious for RPOC
[Fig. 16] but does not exclude GTD. The two are
discriminated by clinical history of recent abortion
with negative or declining beta-hCG level. A PSV
Fig. 16. Retained products of conception. (A) Longitudinal gray-scale US of uterus demonstrates wide endometrium with heterogeneous-appearing contents (arrows). (B) Corresponding spectral Doppler evaluation
demonstrates increased vascularity in this heterogeneous area with low-resistance waveform pattern typical of
trophoblastic flow.
of 21 cm/s is used as the minimum threshold for

the diagnosis of residual trophoblastic tissue [105];
however, the lack of increased flow does not eliminate the possibility of RPOC. In a study by Alcazar
and colleagues, abundant endometrial or myometrial flow with an RI of less than 0.45 had a sensitivity of 93% and specificity of 96% for RPOC
[106]; however, low-impedance peritrophoblastic
flow also can be observed after abortion even
when products of conception are absent [107].
Box 2: Sonographic findings in ovarian torsion

Gray-scale US features
Midline enlarged ovary superior to fundus

of uterus
String of pearls sign
Coexistent mass within the torsed ovary
Free pelvic fluid
Extra-ovarian echogenic, target-appearing
mass = twisted vascular pedicle
Color flow Doppler features
Ovarian pathologies
Torsion
Ovarian torsion accounts for about 3% of the gynecologic emergencies presenting with acute, severe,
unilateral pain. Torsion most frequently involves
the ovary and the fallopian tube with the broad
ligament acting as the fulcrum, and is appropriately
termed adnexal torsion [108]. Ovarian torsion frequently is associated with an ipsilateral benign
ovarian mass but most commonly with a mature
cystic teratoma [109]. Sonographic features of ovarian torsion are not sensitive and cannot lead to
the diagnosis without clinical suspicion. Grayscale findings of a midline enlarged ovary plus or
minus a coexistent mass, peripheral ovarian follicles (the string of pearls sign) [110], pelvic free
fluid, and CFD findings of completely absent arterial and venous ovarian blood flow are specific for
ovarian torsion [Fig. 17]; however, arterial and
venous flow have been documented in the torsed
ovary. Evidence of flow within the ovary does not
exclude ovarian torsion. The presence of a dual
blood supply to the ovary (through the ovarian
and the adnexal branches of the uterine artery) is
responsible for the variable Doppler findings in
adnexal torsion. Technical errors with improper settings of the color flow parameters may result in a
false positive diagnosis of adnexal torsion [111].
CFD may identify the twisted vascular pedicle sign
or whirlpool sign [112].
Arterial and venous flow absent (diagnostic)

Arterial and venous flow present (does not
exclude the diagnosis of torsion)
Arterial flow present, venous flow absent
Arterial flow absent, venous flow present
Whirlpool sign in the twisted
vascular pedicle
Twisted vascular pedicle/whirlpool sign

A twisted vascular pedicle appears on sonography
as a round, echogenic, extraovarian mass with
multiple concentric, internal hypoechoic stripes,
wrapping around a central axis, giving a target appearance. On CDS, these intrapedicular hypoechoic
structures are identified as vessels, confirming the
twisted vascular pedicle and the resultant adnexal
torsion. The color Doppler whirlpool sign is similar to the whirlpool sign described with malrotation of the midgut with volvulus [113,114].
The presence of a whirlpool sign also indicates
the potential viability of the ovary preoperatively
[109,112,115]. The presence of flow should be
documented in the artery and the vein in the vascular pedicle. Documentation of intraovarian central venous flow is also essential to establish the
viability of the torsed ovary [116]. Intermittent
absence of blood flow in the vascular pedicle can
be visualized during episodes of pain brought on
by intermittent episodes of incomplete torsion
[115]. Flow returns when the pain is relieved. Use
Fig. 17. Ovarian torsion. (A) Contrast-enhanced axial CT scan through pelvis demonstrates enlarged ovary
(arrowhead) located in midline superior to fundus of uterus (arrow). CFD US (B) was performed because of
suspicion of ovarian torsion and revealed enlarged ovary with absent blood flow and with peripherally arranged
epithelial cysts (arrows), confirming diagnosis of ovarian torsion.
213
214
Bhatt & Dogra
of intravenous US contrast medium improves

demonstration of altered flow patterns in ischemic
adnexae secondary to torsion, allowing a more
confident diagnosis [117]. The sonographic findings in ovarian torsion are summarized in Box 2.
Subtorsion of the ovary

Subtorsion of the ovary is an early stage of torsion
consequent to incomplete vessel occlusion (torsion
of < 360o ) concomitant with a normal ovarian appearance on sonography [118]. Recognizing impending torsion of the ovary at this stage is
important so that appropriate detorsion methods
can be initiated to salvage the ovary before ischemia sets in. On sonography, a twisted vascular
pedicle cannot be identified in subtorsion. Unilateral, congested periovarian vessels in the presence
of a normal ovarian morphology and blood supply
in an appropriate clinical setting, indicate subtorsion of the ovary [119].
Massive ovarian edema

Massive ovarian edema is a rare condition characterized by massive enlargement of one or both
ovaries secondary to partial torsion of the ovary
occluding venous and lymphatic drainage but sparing the arteries. Chronic venous and lymphatic
obstruction results in marked stromal edema resulting in severe ovarian enlargement [52]. It is
usually unilateral and more commonly involves
the right ovary [120]. On gray-scale sonography,
massive ovarian edema appears as a heterogeneous
mass with cystic components [120]. Color and
pulsed Doppler demonstrate the presence of flow

in the pedicle of the ovary; therefore, Doppler
imaging cannot diagnose massive ovarian edema.
Ectopic pregnancy
Ectopic pregnancy (EP) is one of the most common
gynecologic emergencies, presenting as vaginal bleeding and abdominal pain. Sonographically, the presence of an extra-ovarian thick-walled adnexal mass
called the tubal ring is the most common feature of
an EP [121,122]; however, in a pregnant patient, an
exophytic corpus luteal cyst (CLC) may be mistaken
for an EP [123]. The sonographer or the radiologist
should identify an adnexal mass as a CLC or an
EP, particularly in the absence of a sonographically
visible intrauterine pregnancy so that appropriate
management can be planned. EP and CLC demonstrate abundant vascularity in the periphery of the
tubal ring [124] known as a ring of fire on color
flow imaging [Fig. 18]. A ring of fire is a nonspecific
sign and can also be seen around a mature ovarian
follicle or a hemorrhagic cyst. Various studies have
shown an overlap between the RIs of EPs and CLCs,
but, in general, EPs show lower RIs [125]. In his recent
study, Atri and colleagues[125] showed that low
(< 0.39) and high (> 0.7) RIs can represent EP and
were specific for an EP (100% specificity and positive
predictive value). A higher RI suggests the presence of
less active trophoblasts and, therefore, seems to be a
predictor of spontaneous resolution of the EP [126].
Color Doppler evaluation of the endometrium may
help discriminate between EP and CLC. EPs may also
show endometrial blood flow but the RI is usually
Fig. 18. Ring of fire. (A) CFD image of left adnexa in patient presenting with intrauterine pregnancy, demonstrates
exophytic CLC with ring of fire sign. Adjacent ovarian tissue is compressed (UT, uterus). (B) CFD image of
right adnexa demonstrates ring of fire sign in EP (live EP with cardiac pulsations visualized as red color
signal, arrowhead).
Fig. 19. Ovarian vein thrombosis. CFD image of right

ovarian vein (arrow), demonstrates echogenic thrombus (*) which partially occludes vessel. Some flow
(blue) exists adjacent to thrombus. This patient presented with right lower quadrant pain.
greater than 0.55 and the PSV is less than 15 cm/s

[127,128]. Demonstration of the presence of trophoblastic tissue in the endometrium, even in the absence of a visible double decidual sac, is suggestive of
an intrauterine pregnancy and excludes EP. Trophoblastic tissue is identified by the detection of a lowresistance flow in the endometrium with RI less than
0.55 and a PSV equal to or greater than 15cm/s [127].
Low resistance in the trophoblastic tissue is explained
by the absence of smooth muscles in the walls of
intervillous spaces.
Ovarian vein thrombosis

Ovarian vein thrombosis or thrombophlebitis, a
rare condition, is seen during the postpartum period; it results from venous stasis and the spread of
bacterial infection from endometritis. It commonly
involves the right ovarian vein (90% cases) [52]
and mimics acute appendicitis on clinical presentation [129]. CDS is the favored initial diagnostic
procedure, with CT being a supplementary tool
[129]. Color Doppler imaging demonstrates the
ovarian vein running cephalad and lateral to the
uterus, with hypoechoic to echogenic thrombus,
and absent or partially obstructed flow [Fig. 19].
The thrombus usually involves the junction of
the right ovarian vein with the IVC or extends
into the IVC [130]. Color Doppler imaging also
can be used to follow ovarian vein thrombosis
after anticoagulant therapy to demonstrate resolution of the thrombus.
Ovarian malignancy
Ovarian cancer is the fourth leading cause of cancer deaths among women in the United States and
comprises 25% of gynecologic malignancies in the
United States [52]. The clinician should identify
whether an adnexal mass is a benign or a malignant tumor so that appropriate treatment can be
planned. Ovarian masses less than 5cm in long
axis are generally benign, whereas those greater
than 10cm are more likely to be malignant. Imaging plays a major role in differentiating these
masses, and CDS serves as the main initial modality
for lesion characterization.
The two Doppler angle-independent indices, RI
and PI, predominantly have been used to differentiate malignant from benign adnexal masses. Malignant masses usually have a low PI (< 1.0) and RI
(< 0.4) [131,132], but an overlap of these values
has been noted [133]. In addition, lack of detectable flow by color Doppler US does not exclude
ovarian malignancy [134].
Fig. 20. Ovarian malignancy. (A) CFD image of right adnexa with spectral analysis, demonstrates complex ovarian
mass with blood flow in septae with low-resistance waveform pattern. (B) Power Doppler image in another
patient with ovarian malignancy demonstrates tumor vascularity in mural nodules.
215
216
Bhatt & Dogra
Box 3: Sonographic findings in ovarian

malignancy
> 10 cm size
Increased vascularization
PI < 1.0, RI < 0.4
End diastolic velocity distribution slope of
1.90 1.33
Shorter uptake and longer washout time of
US contrast
Malignant masses are characterized typically by

centrally located vessels that have low-impedance
flow without a diastolic notch [Fig. 20] [135]. The
decay constant of the Doppler waveform, which
characterizes its decrease from systole to diastole
as an exponential decay, has been recently presented as an additional measure of tumor malignancy [136]. In their study, Shaharabany and
colleagues designed a new parameter characterizing
the slope of the mean velocity spectrum at enddiastole (or end diastolic velocity distribution
slope [DVD_S]). This parameter provides much
better results than those obtained with previously
used variables in the detection of malignancy.
The mean DVD_S value for malignant tumors is
1.90 plus or minus 1.33 compared with 9.21 plus
or minus 5.34 for benign masses (area under reciever
operating characteristic curve = 0.983) and yields a
detection rate of approximately 94% [136]. CDS
and gray-scale sonography together with serum
cancer antigen [CA]-125 levels still have a better
sensitivity and specificity for diagnosing ovarian
malignancy than CDS alone [137140]; however,
in the future, vascular analysis may be more useful
because of the advent of US contrast media.
Increased vessel number is a hallmark of malignant
tumors. Contrast-enhanced sonography increases
the accuracy to discriminate benign from malignant
ovarian tumors by demonstrating a higher number
of vessels in malignant lesions [141]. The kinetic
properties of contrast agents, such as a shorter contrast uptake time and a longer washout time, also
have the potential to diagnose malignant ovarian
tumors successfully [142]. The degree, onset, and
duration of Doppler signal enhancement after US
contrast injection, are, therefore, supplemental to
CDS for the diagnosis of malignant ovarian tumors
[143]. The sonographic features of an ovarian
malignancy are summarized in Box 3.
Summary
CFD is performed routinely in all patients who
present with pelvic pain. CFD and power Doppler
sonography in combination with gray-scale imag-
ing and clinical history helps in characterization

and localization of uterine and ovarian masses. CFD
has a specific role in leiomyomas, in the diagnosis
and the management. Subserosal leiomyomas are
distinguished from adnexal masses by demonstrating a VBS or a vascular pedicle between the uterus
and the peri-uterine mass. UAE is a minimally
invasive therapy for leiomyomas. CFD plays an important role in pre-UAE evaluation to predict the
efficacy of the procedure and also in monitoring
the response post-UAE. Endometrial thickening can
be categorized as an endometrial polyp or malignancy with the help of CFD by identifying the pedicle artery sign in an endometrial polyp. Diagnosis
of MP, RPOC and placental abnormalities are aided
by CFD findings with clinical correlation. CFD findings may be variable in some cases of ovarian torsion
but clinical correlation may lead to the correct diagnosis. Other adnexal lesions, such as EP and ovarian
malignancy may be diagnosed with CFD in the
appropriate clinical setting. CFD of the uterus and
adnexae in combination with gray- scale findings
and clinical correlation leads to the correct diagnosis in most patients presenting with acute pelvic pain.
References
[1] Holt SC, Levi CS, Lyons EA. Normal anatomy of
the female pelvis. In: Callen PW, editor. Ultrasonography in obstetrics and gynecology.
3rd edition. Philadelphia: WB Saunders; 1994.
p. 54868.
[2] Fleischer AC. Color Doppler sonography of uterine disorders. Ultrasound Q 2003;19:17989.
[3] Taylor K, Burns PN, Wells PN. Ultrasound
Doppler flow studies of the ovarian and uterine
arteries. Br J Obstet Gynecol 1985;92:2406.
[4] Fleischer AC, Goldstein RB, Bruner JP, et al.
Doppler sonography in obstetrics and gynecology. In: Callen PW, editor. Ultrasonography in
obstetrics and gynecology. 3rd edition. Philadelphia: WB Saunders; 1994. p. 50323.
[5] Ritchie WGM. Ultrasound evaluation of normal and induced ovulation. In: Callen PW,
editor. Ultrasonography in obstetrics and gynecology. 3rd edition. Philadelphia: WB Saunders;
1994. p. 56985.
[6] Fleischer AC, Brader KR. Sonographic depiction
of ovarian vascularity and flow: current improvements and future applications. J Ultrasound Med 2001;20:24150.
[7] Pellizzari P, Esposito C, Siliotti F, et al. Colour
Doppler analysis of ovarian and uterine arteries
in women with hypoestrogenic amenorrhoea.
Hum Reprod 2002;17:320812.
[8] Schwartz SM. Epidemiology of uterine leiomyomata. Clin Obstet Gynecol 2001;44:31626.
[9] Tsuda H, Kawabata M, Nakamoto O, et al.
Clinical predictors in the natural history of
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
uterine leiomyoma: preliminary study. J Ultrasound Med 1998;17:1720.

Kurtz AB, Middleton WD. Pelvis and uterus. In:
Kurtz AB, Middleton WD, editors. Ultrasound:
the requisites. New York: Taylor & Francis Inc.;
2003. p. 53057.
Murase E, Siegelman ES, Outwater EK, et al.
Uterine leiomyomas: histopathologic features,
MR imaging findings, differential diagnosis, and
treatment. Radiographics 1999;19:117997.
Fleischer AC, Donnelly EF, Campbell MG, et al.
Three-dimensional color Doppler sonography
before and after fibroid embolization. J Ultrasound Med 2000;19:7015.
Kim JC, Kim SS, Park JY. Bridging vascular
sign in the MR diagnosis of exophytic uterine
leiomyoma. J Comput Assist Tomogr 2000;24:
5760.
Kim SH, Sim JS, Seong CK. Interface vessels on
color/power Doppler US and MRI: a clue to
differentiate subserosal uterine myomas from
extrauterine tumors. J Comput Assist Tomogr
2001;25:3642.
Ravina JH, Herbreteau D, Ciraru-Vigneron N,
et al. Arterial embolisation to treat uterine myomata. Lancet 1995;346:6712.
Heaston DK, Mineau DE, Brown BJ, et al.
Transcatheter arterial embolization for control
of persistent massive puerperal hemorrhage
after bilateral surgical hypogastric artery ligation. AJR Am J Roentgenol 1979;133:1524.
Ravina JH, Aymard A, Ciraru-Vigneron N, et al.
[Uterine fibroids embolization: results about
454 cases]. Gynecol Obstet Fertil 2003;31:597605.
Brunereau L, Herbreteau D, Gallas S, et al.
Uterine artery embolization in the primary treatment of uterine leiomyomas: technical features
and prospective follow-up with clinical and
sonographic examinations in 58 patients. AJR
Pron G, Mocarski E, Bennett J. Pregnancy after
uterine artery embolization for leiomyomata:
the Ontario multicenter trial. Obstet Gynecol
2005;105:6776.
Goldberg J, Pereira L, Berghella V, et al.
Pregnancy outcomes after treatment for fibromyomata: uterine artery embolization versus
laparoscopic myomectomy. Am J Obstet Gynecol 2004;191:1821.
Spies JB, Roth AR, Jha RC, et al. Leiomyomata
treated with uterine artery embolization: factors
associated with successful symptom and imaging outcome. Radiology 2002;222:4552.
Walker WJ, Pelage JP, Sutton C. Fibroid embolization. Clin Radiol 2002;57:32531.
McLucas B, Perrella R, Goodwin S, et al. Role of
uterine artery Doppler flow in fibroid embolization. J Ultrasound Med 2002;21:11320.
Pelage JP, Guaou NG, Jha RC, et al. Uterine
fibroid tumors: long-term MR imaging outcome after embolization. Radiology 2004;230:
8039.
[25] Pron G, Mocarski E, Bennett J, et al. Tolerance,

hospital stay, and recovery after uterine artery
embolization for fibroids: the Ontario Uterine
Fibroid Embolization Trial. J Vasc Interv Radiol
2003;14:124350.
[26] Muniz CJ, Fleischer AC, Donnelly EF, et al.
Three-dimensional color Doppler sonography
and uterine artery arteriography of fibroids:
assessment of changes in vascularity before and
after embolization. J Ultrasound Med 2002;21:
12933.
[27] Tranquart F, Brunereau L, Cottier JP, et al.
Prospective sonographic assessment of uterine
artery embolization for the treatment of fibroids. Ultrasound Obstet Gynecol 2002;19:
817.
[28] Lipman JC, Smith SJ, Spies JB, et al. IV. Uterine
fibroid embolization: follow-up. Tech Vasc
Interv Radiol 2002;5:4455.
[29] Walker WJ, Pelage JP. Uterine artery embolisation for symptomatic fibroids: clinical results
in 400 women with imaging follow up. BJOG
2002;109:126272.
[30] Weintraub JL, Romano WJ, Kirsch MJ, et al.
Uterine artery embolization: sonographic imaging findings. J Ultrasound Med 2002;21:
6337.
[31] Nicholson TA, Pelage JP, Ettles DF. Fibroid calcification after uterine artery embolization:
ultrasonographic appearance and pathology.
J Vasc Interv Radiol 2001;12:4436.
[32] Katsumori T, Nakajima K, Tokuhiro M. Gadolinium-enhanced MR imaging in the evaluation
of uterine fibroids treated with uterine artery
embolization. AJR Am J Roentgenol 2001;177:
3037.
[33] Azziz R. Adenomyosis: current perspectives. Obstet Gynecol Clin North Am 1989;16:22135.
[34] Bohlman ME, Ensor RE, Sanders RC. Sonographic findings in adenomyosis of the uterus.
[35] Reinhold C, Atri M, Mehio A, et al. Diffuse
uterine adenomyosis: morphologic criteria and
diagnostic accuracy of endovaginal sonography.
Radiology 1995;197:60914.
[36] Reinhold C, McCarthy S, Bret PM, et al. Diffuse
adenomyosis: comparison of endovaginal US
and MR imaging with histopathologic correlation. Radiology 1996;199:1518.
[37] Hata T, Hata K, Senoh D, et al. Doppler ultrasound assessment of tumor vascularity in gynecologic disorders. J Ultrasound Med 1989;8:
30914.
[38] Chiang CH, Chang MY, Hsu JJ, et al. Tumor
vascular pattern and blood flow impedance in
the differential diagnosis of leiomyoma and
adenomyosis by color Doppler sonography.
J Assist Reprod Genet 1999;16:26875.
[39] Reinhold C, Tafazoli F, Mehio A, et al. Uterine
adenomyosis: endovaginal US and MR imaging
features with histopathologic correlation. Radiographics 1999;19 Spec No:S14760.
217
218
Bhatt & Dogra
[40] McLucas B, Perrella R. Adenomyosis: MRI of the

uterus treated with uterine artery embolization.
[41] Siskin GP, Tublin ME, Stainken BF, et al. Uterine artery embolization for the treatment of
adenomyosis: clinical response and evaluation
with MR imaging. AJR Am J Roentgenol 2001;
177:297302.
[42] Liu P, Chen C, Liu L, et al. [Investigation of the
hemodynamic changes during uterine arterial
embolization in the treatment of adenomyosis].
Zhonghua Fu Chan Ke Za Zhi 2002;37:5368.
[43] Jha RC, Takahama J, Imaoka I, et al. Adenomyosis: MRI of the uterus treated with uterine
artery embolization. AJR Am J Roentgenol 2003;
181:8516.
[44] Kim MD, Won JW, Lee DY, et al. Uterine artery embolization for adenomyosis without fibroids. Clin Radiol 2004;59:5206.
[45] Pelage JP, Jacob D, Fazel A, et al. Midterm
results of uterine artery embolization for symptomatic adenomyosis: initial experience. Radiology 2005;234:94853.
[46] McLucas B, Adler L, Perrella R. Uterine fibroid embolization: nonsurgical treatment for symptomatic fibroids. J Am Coll Surg 2001;192:95105.
[47] Jakab A, Ovari L, Juhasz B, et al. Detection of
feeding artery improves the ultrasound diagnosis of endometrial polyps in asymptomatic
patients. Eur J Obstet Gynecol Reprod Biol 2005;
119:1037.
[48] Fleischer AC, Shappell HW. Color Doppler
sonohysterography of endometrial polyps and
submucosal fibroids. J Ultrasound Med 2003;
22:6014.
[49] Timmerman D, Verguts J, Konstantinovic ML,
et al. The pedicle artery sign based on sonography with color Doppler imaging can replace
second-stage tests in women with abnormal
vaginal bleeding. Ultrasound Obstet Gynecol
2003;22:16671.
[50] Alcazar JL, Galan MJ, Minguez JA, et al. Transvaginal color Doppler sonography versus sonohysterography in the diagnosis of endometrial
polyps. J Ultrasound Med 2004;23:7438.
[51] Reinhold C, Khalili I. Postmenopausal bleeding: value of imaging. Radiol Clin North Am 2002;
40:52762.
[52] Salem S, Wilson SR. Gynecologic ultrasound.
In: Rumack CM, Wilson SR, Charboneau JW,
editors. 3rd edition. Diagnostic ultrasound,
Vol 2. St. Louis (MO): Elsevier, Mosby; 2004.
p. 52787.
[53] Aleem F, Predanic M, Calame R, et al. Transvaginal color and pulsed Doppler sonography
of the endometrium: a possible role in reducing
the number of dilatation and curettage procedures. J Ultrasound Med 1995;14:13945.
[54] Carter J, Saltzman A, Hartenbach E, et al. Flow
characteristics in benign and malignant gynecologic tumors using transvaginal color flow
Doppler. Obstet Gynecol 1994;83:12530.
[55] Alcazar JL, Castillo G, Minguez JA, et al. Endometrial blood flow mapping using transvaginal power Doppler sonography in women
with postmenopausal bleeding and thickened
endometrium. Ultrasound Obstet Gynecol 2003;
21:5838.
[56] Kurjak A, Kupesic S. Three dimensional ultrasound and power doppler in assessment of
uterine and ovarian angiogenesis: a prospective
study. Croat Med J 1999;40:41320.
[57] Bonilla-Musoles F, Marti MC, Ballester MJ, et al.
Normal uterine arterial blood flow in postmenopausal women assessed by transvaginal
color Doppler ultrasonography. J Ultrasound Med
1995;14:4914.
[58] Alcazar JL, Galan MJ, Jurado M, et al. Intratumoral blood flow analysis in endometrial
carcinoma: correlation with tumor characteristics and risk for recurrence. Gynecol Oncol 2002;
84:25862.
[59] Lee CN, Cheng WF, Chen CA, et al. Angiogenesis of endometrial carcinomas assessed by
measurement of intratumoral blood flow, microvessel density, and vascular endothelial
growth factor levels. Obstet Gynecol 2000;96:
61521.
[60] Cheng WF, Chen CA, Lee CN, et al. Preoperative
ultrasound study in predicting lymph node
metastasis for endometrial cancer patients.
Gynecol Oncol 1998;71:4247.
[61] Cheng WF, Chen TM, Chen CA, et al. Clinical
application of intratumoral blood flow study
in patients with endometrial carcinoma. Cancer
1998;82:18816.
[62] Ghi T, Giunchi S, Rossi C, et al. Threedimensional power Doppler sonography in
the diagnosis of arteriovenous malformation
of the uterus. J Ultrasound Med 2005;24:
72731.
[63] Damani N, Wilson SR. Nongynecologic applications of transvaginal US. Radiographics 1999;
19 Spec No:S179200.
[64] Polat P, Suma S, Kantarcy M, et al. Color Doppler US in the evaluation of uterine vascular
abnormalities. Radiographics 2002;22:4753.
[65] Kwon JH, Kim GS. Obstetric iatrogenic arterial
injuries of the uterus: diagnosis with US and
treatment with transcatheter arterial embolization. Radiographics 2002;22:3546.
[66] Huang MW, Muradali D, Thurston WA, et al.
Uterine arteriovenous malformations: gray-scale
and Doppler US features with MR imaging
correlation. Radiology 1998;206:11523.
[67] Henrich W, Fuchs I, Luttkus A, et al. Pseudoaneurysm of the uterine artery after cesarean
delivery: sonographic diagnosis and treatment.
[68] Langer JE, Cope C. Ultrasonographic diagnosis
of uterine artery pseudoaneurysm after hysterectomy. J Ultrasound Med 1999;18:7114.
[69] Laubach M, Delahaye T, Van Tussenbroek F,
et al. Uterine artery pseudo-aneurysm: diagnosis
[70]
[71]
[72]
[73]
[74]
[75]
[76]
[77]
[78]
[79]
[80]
[81]
[82]
[83]
[84]
[85]
and therapy during pregnancy. J Perinat Med

2000;28:3215.
Lee WK, Roche CJ, Duddalwar VA, et al. Pseudoaneurysm of the uterine artery after abdominal hysterectomy: radiologic diagnosis and
management. Am J Obstet Gynecol 2001;185:
126972.
Zhou Q, Lei XY, Xie Q, et al. Sonographic and
Doppler imaging in the diagnosis and treatment of gestational trophoblastic disease:
a 12-year experience. J Ultrasound Med 2005;
24:1524.
Fraser-Hill F, Wilson SR. Gestational trophoblastic neoplasia. In: Rumack CM, Wilson SR,
Charboneau JW, editors. 3rd edition. Diagnostic ultrasound, Vol 2. St. Louis (MO): Elsevier,
Mosby; 2004. p. 589601.
Dogra V, Paspulati RM, Bhatt S. First trimester bleeding evaluation. Ultrasound Q 2005;21:
6985.
Fleischer AC, James Jr AE, Krause DA, et al.
Sonographic patterns in trophoblastic diseases.
Radiology 1978;126:21520.
Soper JT. Identification and management of
high-risk gestational trophoblastic disease. Semin
Oncol 1995;22:17284.
Mangili G, Spagnolo D, Valsecchi L, et al. Transvaginal ultrasonography in persistent trophoblastic tumor. Am J Obstet Gynecol 1993;169:
121823.
Taylor KJ, Schwartz PE, Kohorn EI. Gestational
trophoblastic neoplasia: diagnosis with Doppler
US. Radiology 1987;165:4458.
Desai RK, Desberg AL. Diagnosis of gestational
trophoblastic disease: value of endovaginal
color flow Doppler sonography. AJR Am J Roentgenol 1991;157:7878.
Carter J, Fowler J, Carlson J, et al. Transvaginal
color flow Doppler sonography in the assessment of gestational trophoblastic disease. J Ultrasound Med 1993;12:5959.
Paspulati RM, Bhatt S, Nour S. Sonographic evaluation of first-trimester bleeding. Radiol Clin
North Am 2004;42:297314.
Avva R, Shah HR, Angtuaco TL. US case of the
day. Placenta increta. Radiographics 1999;19:
108992.
Lazebnik N, Lazebnik RS. The role of ultrasound in pregnancy-related emergencies. Radiol
Clin North Am 2004;42:31527.
Taipale P, Orden MR, Berg M, et al. Prenatal
diagnosis of placenta accreta and percreta with
ultrasonography, color Doppler, and magnetic
resonance imaging. Obstet Gynecol 2004;104:
53740.
Hoffman-Tretin JC, Koenigsberg M, Rabin A,
et al. Placenta accreta. Additional sonographic
observations. J Ultrasound Med 1992;11:2934.
Rosemond RL, Kepple DM. Transvaginal color
Doppler sonography in the prenatal diagnosis
of placenta accreta. Obstet Gynecol 1992;80:
50810.
[86] Lerner JP, Deane S, Timor-Tritsch IE. Characterization of placenta accreta using transvaginal
sonography and color Doppler imaging. Ultrasound Obstet Gynecol 1995;5:198201.
[87] Levine D, Hulka CA, Ludmir J, et al. Placenta
accreta: evaluation with color Doppler US,
power Doppler US, and MR imaging. Radiology
1997;205:7736.
[88] Miller DA, Chollet JA, Goodwin TM. Clinical
risk factors for placenta previa-placenta accreta.
Am J Obstet Gynecol 1997;177:2104.
[89] Krapp M, Baschat AA, Hankeln M, et al. Gray
scale and color Doppler sonography in the third
stage of labor for early detection of failed placental separation. Ultrasound Obstet Gynecol
2000;15:13842.
[90] Finberg HJ, Williams JW. Placenta accreta:
prospective sonographic diagnosis in patients
with placenta previa and prior cesarean section.
[91] Megier P, Du Rouchet E, Desroches A, et al.
[Prenatal diagnosis of a placenta praevia percreta with color Doppler. A case report. Review
of the literature]. J Gynecol Obstet Biol Reprod
(Paris) 1994;23:3157.
[92] Passini Jr R, Knobel R, Barini R, et al. Placenta
percreta with silent rupture of the uterus. Sao
Paulo Med J 1996;114:12703.
[93] Mathieu E, Dufour P, Ernoult P, et al. [Uterine
rupture after twenty-two weeks of amenorrhea
due to placenta praevia percreta. A case report].
Rev Fr Gynecol Obstet 1995;90:22832.
[94] Lee W, Lee VL, Kirk JS, et al. Vasa previa:
prenatal diagnosis, natural evolution, and clinical outcome. Obstet Gynecol 2000;95:5726.
[95] Stafford IP, Neumann DE, Jarrell H. Abnormal
placental structure and vasa previa: confirmation of the relationship. J Ultrasound Med 2004;
23:15212.
[96] Oyelese KO, Turner M, Lees C, et al. Vasa previa:
an avoidable obstetric tragedy. Obstet Gynecol
Surv 1999;54:13845.
[97] Oyelese Y, Spong C, Fernandez MA, et al. Second trimester low-lying placenta and in-vitro
fertilization? Exclude vasa previa. J Matern Fetal
Med 2000;9:3702.
[98] Schachter M, Tovbin Y, Arieli S, et al. In vitro
fertilization is a risk factor for vasa previa. Fertil
Steril 2002;78:6423.
[99] Sepulveda W, Rojas I, Robert JA, et al. Prenatal
detection of velamentous insertion of the
umbilical cord: a prospective color Doppler ultrasound study. Ultrasound Obstet Gynecol 2003;
21:5649.
[100] Hertzberg BS, Kliewer MA. Vasa previa: prenatal
diagnosis by transperineal sonography with
Doppler evaluation. J Clin Ultrasound 1998;26:
4058.
[101] Sadan O, Golan A, Girtler O, et al. Role of
sonography in the diagnosis of retained products of conception. J Ultrasound Med 2004;
23:3714.
219
220
Bhatt & Dogra
[102] Achiron R, Goldenberg M, Lipitz S, et al. Transvaginal duplex Doppler ultrasonography in

bleeding patients suspected of having residual
trophoblastic tissue. Obstet Gynecol 1993;81:
50711.
[103] Cetin A, Cetin M. Diagnostic and therapeutic
decision-making with transvaginal sonography
for first trimester spontaneous abortion, clinically thought to be incomplete or complete.
Contraception 1998;57:3937.
[104] Keogan MT, Hertzberg BS, Kliewer MA. Low
resistance Doppler waveforms with retained
products of conception: potential for diagnostic
confusion with gestational trophoblastic disease. Eur J Radiol 1995;21:10911.
[105] Dillon EH, Feyock AL, Taylor KJ. Pseudogestational sacs: Doppler US differentiation from
normal or abnormal intrauterine pregnancies.
Radiology 1990;176:35964.
[106] Alcazar JL. Transvaginal ultrasonography combined with color velocity imaging and pulsed
Doppler to detect residual trophoblastic tissue.
Ultrasound Obstet Gynecol 1998;11:548.
[107] Dillon EH, Case CQ, Ramos IM, et al. Endovaginal US and Doppler findings after firsttrimester abortion. Radiology 1993;186:8791.
[108] Warner MA, Fleischer AC, Edell SL, et al. Uterine adnexal torsion: sonographic findings. Radiology 1985;154:7735.
[109] Lee EJ, Kwon HC, Joo HJ, et al. Diagnosis of
ovarian torsion with color Doppler sonography: depiction of twisted vascular pedicle.
[110] Graif M, Itzchak Y. Sonographic evaluation of
ovarian torsion in childhood and adolescence.
[111] Pellerito JS, Troiano RN, Quedens-Case C, et al.
Common pitfalls of endovaginal color Doppler
flow imaging. Radiographics 1995;15:3747.
[112] Vijayaraghavan SB. Sonographic whirlpool sign
in ovarian torsion. J Ultrasound Med 2004;23:
16439.
[113] Pracros JP, Sann L, Genin G, et al. Ultrasound
diagnosis of midgut volvulus: the whirlpool
sign. Pediatr Radiol 1992;22:1820.
[114] Shimanuki Y, Aihara T, Takano H, et al. Clockwise whirlpool sign at color Doppler US: an
objective and definite sign of midgut volvulus.
Radiology 1996;199:2614.
[115] Albayram F, Hamper UM. Ovarian and adnexal
torsion: spectrum of sonographic findings with
pathologic correlation. J Ultrasound Med 2001;
20:10839.
[116] Fleischer AC, Stein SM, Cullinan JA, et al. Color
Doppler sonography of adnexal torsion. J Ultrasound Med 1995;14:5238.
[117] Brown JM, Taylor KJ, Alderman JL, et al.
Contrast-enhanced ultrasonographic visualization of gonadal torsion. J Ultrasound Med 1997;
16:30916.
[118] Beyth Y, Bar-On E. Tuboovarian autoamputation and infertility. Fertil Steril 1984;42:9324.
[119] Shalev J, Mashiach R, Bar-Hava I, et al. Subtorsion of the ovary: sonographic features and
clinical management. J Ultrasound Med 2001;
20:84954.
[120] Chiang G, Levine D. Imaging of adnexal masses
in pregnancy. J Ultrasound Med 2004;23:80519.
[121] Cacciatore B, Stenman UH, Ylostalo P. Early
screening for ectopic pregnancy in high-risk
symptom-free women. Lancet 1994;343:5178.
[122] Braffman BH, Coleman BG, Ramchandani P,
et al. Emergency department screening for
ectopic pregnancy: a prospective US study. Radiology 1994;190:797802.
[123] Atri M, Leduc C, Gillett P, et al. Role of endovaginal sonography in the diagnosis and management of ectopic pregnancy [discussion 775].
Radiographics 1996;16:75574.
[124] Stein MW, Ricci ZJ, Novak L, et al. Sonographic
comparison of the tubal ring of ectopic pregnancy with the corpus luteum. J Ultrasound
Med 2004;23:5762.
[125] Atri M. Ectopic pregnancy versus corpus luteum
cyst revisited: best Doppler predictors. J Ultrasound Med 2003;22:11814.
[126] Atri M, Chow CM, Kintzen G, et al. Expectant
treatment of ectopic pregnancies: clinical and
sonographic predictors. AJR Am J Roentgenol
2001;176:1237.
[127] Parvey HR, Dubinsky TJ, Johnston DA, et al.
The chorionic rim and low-impedance intrauterine arterial flow in the diagnosis of early
intrauterine pregnancy: evaluation of efficacy.
[128] Wherry KL, Dubinsky TJ, Waitches GM, et al.
Low-resistance endometrial arterial flow in the
exclusion of ectopic pregnancy revisited. J Ultrasound Med 2001;20:33542.
[129] Prieto-Nieto MI, Perez-Robledo JP, RodriguezMontes JA, et al. Acute appendicitis-like symptoms as initial presentation of ovarian vein
thrombosis. Ann Vasc Surg 2004;18:4813.
[130] Grant TH, Schoettle BW, Buchsbaum MS. Postpartum ovarian vein thrombosis: diagnosis by
clot protrusion into the inferior vena cava at
sonography. AJR Am J Roentgenol 1993;160:
5512.
[131] Kurjak A, Zalud I, Alfirevic Z. Evaluation of adnexal masses with transvaginal color ultrasound.
[132] Weiner Z, Thaler I, Beck D, et al. Differentiating
malignant from benign ovarian tumors with
transvaginal color flow imaging. Obstet Gynecol 1992;79:15962.
[133] Hamper UM, Sheth S, Abbas FM, et al. Transvaginal color Doppler sonography of adnexal
masses: differences in blood flow impedance in
benign and malignant lesions. AJR Am J Roentgenol 1993;160:12258.
[134] Brown DL, Frates MC, Laing FC, et al. Ovarian
masses: can benign and malignant lesions be
differentiated with color and pulsed Doppler
US? Radiology 1994;190:3336.
[135] Fleischer AC, Rodgers WH, Kepple DM, et al.

Color Doppler sonography of ovarian masses:
a multiparameter analysis. J Ultrasound Med
1993;12:418.
[136] Shaharabany Y, Akselrod S, Tepper R. A sensitive
new indicator for diagnostics of ovarian malignancy, based on the Doppler velocity spectrum.
Ultrasound Med Biol 2004;30:295302.
[137] Fleischer AC, Cullinan JA, Kepple DM, et al.
Conventional and color Doppler transvaginal
sonography of pelvic masses: a comparison of
relative histologic specificities. J Ultrasound
Med 1993;12:70512.
[138] Kinkel K, Hricak H, Lu Y, et al. US characterization of ovarian masses: a meta-analysis. Radiology 2000;217:80311.
[139] Jain KA. Prospective evaluation of adnexal
masses with endovaginal gray-scale and duplex
[140]
[141]
[142]
[143]
and color Doppler US: correlation with pathologic findings. Radiology 1994;191:637.
Reles A, Wein U, Lichtenegger W. Transvaginal
color Doppler sonography and conventional
sonography in the preoperative assessment
of adnexal masses. J Clin Ultrasound 1997;25:
21725.
Marret H, Sauget S, Giraudeau B, et al. Contrastenhanced sonography helps in discrimination
of benign from malignant adnexal masses.
Orden MR, Gudmundsson S, Kirkinen P.
Contrast-enhanced sonography in the examination of benign and malignant adnexal masses.
Orden MR, Jurvelin JS, Kirkinen PP. Kinetics of
a US contrast agent in benign and malignant
adnexal tumors. Radiology 2003;226:40510.
221

Genitourinary Ultrasound

Uploaded by

Copyright:

Available Formats

Genitourinary Ultrasound

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

Genitourinary Ultrasound

Uploaded by

Copyright:

Available Formats

xi

The rapid advances and refinements of modern

periodical will have issues dedicated to obstetrics

relevant in the practice of ultrasonography as it

Medical ultrasonography dates back to the 1930s

of several different disciplines and can result in

challenging as it is located more superiorly, lacks

Ultrasonographic features of acute

complicated cases, ultrasound imaging will usually

who have uncomplicated pyelonephritis. In addition, focal pyelonephritis commonly progresses to

Ultrasonographic features of acute focal and

Fig. 2. Fungal ball. Longitudinal gray-scale sonogram

scintigraphy [16]. Some authors suggest that although

Acute focal and multifocal pyelonephritis

Acute focal and multifocal pyelonephritis occur

Before the advent of antibiotics, most abscesses in

The management of renal abscesses is generally

Ultrasonographic features of renal abscess

Ultrasonographic findings of emphysematous

Fig. 5. Emphysematous pyelonephritis. Longitudinal

clean posterior shadowing [23,24]. The isolated

Ultrasonographic features of pyonephrosis

tic shadowing can be seen behind the affected area

of the entire inflammatory process. Limited disease

help distinguish XGP from renal tumors and other

Ultrasonographic features of malakoplakia

Hydatid disease of the kidney (renal

Ultrasonographic features of renal

Table 1: Gharbi ultrasonographic classification of hyatid cysts

Discrete univesicular mass

Heterogenous mass with

graphic classification [Table 1] [30]. The Gharbi

tected on ultrasound as a partitioned cyst with a

lesions from other similar appearing pathology,

frequency may result. One quarter of patients will

Ultrasonographic features of renal

Ultrasonographic features of HIV-associated

Early findings of urinary tract tuberculosis are best

Renal disease is a common complication in patients

long-term morbidity and mortality. If the strengths

ing, and power Doppler US in an experimental

[31] von Sinner WN. New diagnostic signs in hydatid

[35] Vijayaraghavan SB, Kandasamy SV, Arul M, et al.

Sonography of Benign Renal

Simple cortical cysts

When evaluating renal masses, differentiating

the differentiation of nonsurgical from surgical lesions [5].

Simple cortical cysts

Complex renal cysts

Box 1: Bosniak classification of renal cystic

One or two thin ( 1 mm thick) septations or

Uniform wall thickening or nodularity

Nonuniform or enhancing thick wall

wall, or mural nodularity. Bosniak category II cysts

on the degree of abnormality, will require further

Sonography of Benign Renal Cystic Disease

Wilms tumor in a child and cystic adenocarcinoma

that include multiple, small, confluent cysts versus

Renal sinus cysts

Medullary cystic disease