Ultrasonographic Assessment of Cirrhosis and Portal Hypertension
1Radiology Department, Hippokration General Hospital, ?thens, Greece
2?cademic Department of Medicine, University of Athens Medical School, Hippokration General Hospital, Athens,
Greece
Abstract: Cirrhosis and portal hypertension are common diagnostic findings in the Western world. The aetiology is vari-
able, however alcohol abuse and hepatitis are the commonest causes. Ultrasound is usually the first diagnostic modality to
be used. It is easy to find, safe, radiation-free and cheap. Although computed tomography, magnetic resonance and biopsy
may be needed for diagnosis and follow up, ultrasound will always be used for initial assessment and in many cases will
solve the diagnostic problem. We review the findings on B-mode and Doppler, as well as contrast enhanced ultrasound,
according to a protocol which should be followed in order to evaluate a patient with cirrhosis and portal hypertension.
Keywords: Liver, cirrhosis, portal hypertension, ultrasound.
INTRODUCTION
Cirrhosis is defined by the World Health Organisation as
a diffuse progress characterised by fibrosis and conversion of
normal hepatic architecture into structurally abnormal nod-
ules [1]. Cell death, fibrosis and regeneration are thus com-
bined to create cirrhosis [2] which is classified as micro-
nodular (nodule diameter ranging between 0.1 and 1 cm,
alcohol consumption being the commonest cause) and
macronodular (nodules of variable size up to 5 cm with
chronic viral hepatitis as the commonest cause) [3].
A common complication of cirrhosis is portal hyperten-
sion. It is defined as a portal pressure increase over 6-10 mm
Hg or a pressure gradient between the hepatic veins and the
portal vein over 5 mm Hg due to increased resistance in por-
tal circulation and/or increased portal blood flow [4]. Causes
are divided into prehepatic, intrahepatic and suprahepatic.
Prehepatic aetiology is characterised by an obstruction of the
portal or superior mesenteric veins and includes aplasia, hy-
poplasia or thrombosis of the portal vein, infection, trauma
and malignancy. Intrahepatic aetiology includes cirrhosis,
congenital hepatic fibrosis, hepatitis, myeloproliferative dis-
orders and schistosomiasis. Suprahepatic portal hypertension
is caused by obstruction of blood flow away from the liver
(Budd-Chiari syndrome). Altogether, the commonest cause
of portal hypertension in the Western world is cirrhosis, ac-
counting for over 90% of cases [2].
CIRRHOSIS AND PORTAL HYPERTENSION UL-
TRASOUND FINDINGS
Ultrasound (US) is the first imaging examination used for
assessing the hepatic parenchyma and vasculature in patients
with cirrhosis and portal hypertension. An examination pro-
tocol should evaluate the following:
*Address correspondence to this author at the ?cademic Department of
Medicine, University of Athens Medical School, Hippokration General
Hospital, 124 Vas Sofias Ave, 115 25, Athens, Greece; Fax; ++30-210-
6993693; E-mail: spdour@med.uoa.gr
1. Portal vein diameter
2. Portal vein flow
3. Hepatic artery and veins flow
4. Cirrhotic liver morphology
5. Splenic size
6. Portosystemic collateral venous communica-
tions
7. Ascites
All these sonographic parameters will be described in
this paper.
1. Portal Vein Diameter
The normal caliber of the portal vein is up to 13 mm dur-
ing calm respiration. It increases up to 16 mm in deep inspi-
ration, as well as postprandially. A diameter increase of less
than 20% can diagnose cirrhosis with a sensitivity of 81%
and a specificity of 100% [5]. On the contrary, the portal
vein diameter decreases after exercise and in the erect posi-
tion [6]. Diameter assessment should be performed at the
confluence of the portal vein with the inferior vena cava in
the supine position. If these criteria are met, a diameter over
13 mm is indicative of portal hypertension with a specificity
of 100% and a sensitivity of 45-50% (Fig. 1). This poor sen-
sitivity may be due to the fact that very severe portal hyper-
tension is accompanied by reduced flow and a narrow portal
vein. Increase of sensitivity can be obtained if, besides the
portal vein, the diameters of the splenic and superior mesen-
teric veins during respiratory movements are also measured.
However, although portal vein diameter can reflect hy-
pertension (some authors have identified that a diameter over
17 mm is 100% predictive for large varices [4]), a normal
diameter does not exclude portal hypertension [5]. Thus,
although the portal vein caliber may initially increase [7],
when portosystemic shunts develop allowing pressure to
escape, diameter may decrease [8].
Ultrasonographic Assessment of Cirrhosis
Current Medical Imaging Reviews, 2009, Vol. 5, No. 1 63
An aneurysm of the portal vein is a rare complication of
portal hypertension, although it can also be congenital [9].
The most common location is at the splenic and superior
mesenteric veins junction or in the portal vein radicles. On
US, an anechoic mass connecting to the portal venous sys-
tem with turbulent venous flow can be noted.
2. Portal Vein Flow
Normally the portal vein flow is hepatopedal throughout
the cardiac cycle, with a median velocity of 15-18 cm/sec.
Flow normally varies with respiration and heart pulse. In
portal hypertension, velocity diminishes and the waveform is
dampened with decrease in amplitude of oscillations during
breathing. As portal pressure increases, flow may become
biphasic, towards and away from the liver during the cardiac
cycle. Finally it reverses, becoming monophasic and hepato-
fugal (Fig. 2). Intrahepatic arterial-portal venous shunting
and reversed flow in the splenic vein may also be observed.
However, once collateral circulation is set and portal hyper-
tension decreases, normal hepatopedal flow may recover.
When the portal vein is obstructed it is usually due to
thrombosis, neoplastic invasion or cirrhotic slow flow. Less
common causes are blood hypercoagulation, surgery and
peritoneal inflammation. Permanent obstruction occurs more
often than recanalisation. Decompression of the portal circu-
lation occurs when portosystemic collateral circulation is
formed. During the acute obstruction phase the portal vein
shows echogenic intraluminal thrombus, expansion of the
vein’s caliber, venous collateral formation and cavernous
transformation. In this last case, numerous wormlike vessels
are formed at the porta hepatis, representing periportal col-
lateral circulation (Fig. 3) [2]. This finding is seen in chronic
thrombosis, requiring up to 12 months to form, therefore
more often noted in benign disease [10]. If the obstruction is
malignant, then a pulsatile flow can be seen in the portal vein
with a specificity of 100% and a sensitivity of 62% [11]. On
Fig. (1). Increased caliber (2.75 cm) of the portal vein in a cirrhotic liver.
Fig. (2). Normal (left) and reversed (right) portal vein flow. In both cases cirrhosis with ascites is present.
64 Current Medical Imaging Reviews, 2009, Vol. 5, No. 1
Cokkinos and Dourakis
the contrary, a benign thrombotic obstruction shows no or
sporadic continuous flow. During the chronic phase, fibrosis
may result in an indiscreet, almost unidentifiable on US,
portal vein (Fig. 4).
3. HEPATIC ARTERY AND VEINS FLOW
3.1. Hepatic Veins
The hepatic vein normal waveform is triphasic, reflecting
pressure from the right atrium in a normally compliant liver
through the thin walls of the veins. Two antegrade diastolic
and systolic waves are followed by a smaller retrograde
wave which corresponds to the atrial “Kick” [2]. As fibrosis
evolves, the liver parenchyma stiffens, not allowing the he-
patic veins to reflect pressure alterations. This results into
decreased amplitude of phasic oscillations, reversed flow
loss and flattened waveform (Fig. 5) [12, 13]. As cirrhosis
progresses, the hepatic veins’ lumen narrows and velocity
increases with colour aliasing and turbulence. Therefore, in
cirrhotic patients, a triphasic pattern has been observed in
about half of cases and a biphasic waveform in the remaining
half. A completely flat waveform has been noted in up to 3%
of cirrhotic patients. Mean hepatic vein velocity is higher in
cirrhotics than non-cirrhotic controls. The poorer the grade
of cirrhosis, the higher is the hepatic vein mean velocity,
with very high values (? 20 cm/s) noted in patients with
moderate to massive ascites [14].
US contrast agents can be used to differentiate cirrhotic
from noncirrhotic patients by measuring hepatic vein arrival
and peak enhancement times. In normal patients, arrival of
contrast in a hepatic vein usually occurs after 30 s [15]. In
comparison, in cirrhosis there is an early arrival time of the
bolus injection (< 24 s) because of increased hepatic arterial
supply as well as intrahepatic and intrapulmonary shunts.
Therefore, the hepatic vein arrival and peak enhancement
Fig. (3). Portal vein thrombosis with porta hepatis cavernous transformation.
Fig. (4). Different cases of partial (A. Colour Doppler-B. Power Doppler) and complete (C. B-mode US) portal vein thrombosis. In case C,
chronicity results in an indiscreet portal vein. A stent is seen in the common bile duct.
Ultrasonographic Assessment of Cirrhosis
Current Medical Imaging Reviews, 2009, Vol. 5, No. 1 65
times [16], as well as transit time (time difference of contrast
enhancement between the hepatic vein and artery) [17] are
shorter in cirrhosis compared to noncirrhotic liver disease. A
hepatic vein arrival time cutoff of 17 s distinguishes cirrhotic
from noncirrhotic patients with high accuracy (100% sensi-
tivity, 93.3% specificity) and better sensitivity than B-mode
and Doppler US [16]. On another study, this cutoff time was
calculated at 21 s, while a hepatic artery to hepatic vein tran-
sit time of <12 s was 100% sensitive for cirrhosis [17].
3.2. Budd-Chiari Syndrome
Budd-Chiari syndrome includes the clinical and his-
tological changes caused by the acute obstruction of the he-
patic veins [18]. Changes may be partial, affecting some of
the veins or even a segment of a vein [4]. Patients usually
present with abdominal pain, hepatomegaly and ascites.
Laboratory results may show altered hepatic indices. US
reveals the thrombus inside the hepatic vein and possibly in
the inferior vena cava with centrolobular distension and
blood pooling in the sinusoidal capillaries (Fig. 6). US can
also give information on the nature (thrombotic or neoplas-
tic) of the portal vein obstruction. Hepatic US findings in
Budd-Chiari syndrome are direct and indirect [19].
• Direct findings include:
1. Echogenic material in the vessel lumen with no
flow. When partial thrombosis is present, the
patent part of the lumen shows increased veloc-
ity.
2. Two colour flow: normal in one branch and ab-
normal in the other.
3. Intrahepatic collateral circulation from the he-
patic towards the portal or splenic vein with a di-
ameter <1 cm.
4. Abnormal flow in the portal vein (reduced veloc-
ity, biphasic or reversed flow).
5. If thrombosis extends from the inferior vena cava
towards the hepatic veins, the latter may be par-
tially patent with reduced flow.
• Indirect findings consist of:
1. Enlargement and reduced echogenicity of the
liver in the acute phase.
2. Fibrosis, shrinking and increased echogenicity in
the chronic phase.
3. Increased size of the left and caudate lobes.
Fig. (5). Normal triphasic hepatic vein flow (left). Cirrhotic flattened hepatic vein waveform (right).
Fig. (6). A. Thrombosis of the inferior vena cava on B-mode US. Thrombosis at the confluence of the hepatic veins on B-mode (B) and Col-
our Doppler (C) US.
66 Current Medical Imaging Reviews, 2009, Vol. 5, No. 1
Cokkinos and Dourakis
Finally, extrahepatic findings include ascites, pleural ef-
fusion, gallbladder wall oedema, splenomegaly and collateral
circulation [20].
3.3. Hepatic Artery
The hepatic artery normally shows a typical splanchnic
waveform, with normal maximum velocity measuring 55.2 ±
12.0 cm/s. In cirrhotic patients, especially when portal vein
thrombosis may occur, this value can increase up to 64.4 ±
21.8 cm/s (Fig. 7) [14].
4. CIRRHOTIC LIVER MORPHOLOGY
4.1. Liver Size
Cirrhosis is a non typical damage to the liver tissue de-
fined by necrosis, fibrosis and nodular regeneration. During
this procedure, hepatic parenchyma volume is redistributed.
Thus, in early stages of cirrhosis the liver may be enlarged,
while in the chronic phase it is usually small with relative
enlargement of the caudate and left lobes compared to the
right [2]. Transverse diameters of the caudate and right lobes
should be measured in the portal vein bifurcation inside the
liver. A caudate/right lobe ratio of 0.65 or higher (Fig. 8) is
indicative of cirrhosis with high specificity (100%) but low
sensitivity (43-84%). Therefore, this ratio is useful only
when abnormal [21].
Segment IV size is also evaluated, decreasing in cirrho-
sis. Using as landmarks the umbilical fissure and the left
portal vein centrally, the gallbladder fissure laterally and the
middle hepatic vein posteriorly, a normal segment IV trans-
verse diameter is 43 ± 8 mm. In cirrhosis, this size dimin-
Fig. (7). Normal hepatic artery waveform (left). Increased hepatic artery velocity (109.9 cm/s) in a cirrhotic patient with portal vein thrombo-
sis (right).
Fig. (8). Measurement of caudate and right lobes transverse diameters in the portal vein bifurcation. In this case the ratio is normal.
Ultrasonographic Assessment of Cirrhosis
Current Medical Imaging Reviews, 2009, Vol. 5, No. 1 67
ishes to 28 ± 9 mm. However, this measurement shows no
relationship with severity or aetiology [22].
However, there are theories contemplating that there is
no need for routine size measurements and that visual as-
sessment is enough for cirrhosis characterisation1. Although
the caudate lobe is more markedly enlarged in patients with
alcohol abuse than with viral cirrhosis, this does not always
apply to individuals, due to major overlap [23]. Altogether,
one should always remember that the evaluation of morpho-
logical changes is only visual and requires some training and
attention to individual normal variants. A combination of 3
out of 4 parameters (Segment IV atrophy, left lobe hypertro-
phy, caudate lobe hypertrophy and right posterior atrophy) is
highly indicative of liver cirrhosis.
4.2. Liver Echotexture
Liver attenuation is correlated to the presence of fat and
not fibrosis [24]. Therefore, unless fatty infiltration also ex-
ists, a cirrhotic liver attenuation is similar to that of a normal
patient. Fat deposition may happen homogeneously, causing
generalised increase in liver echogenicity, or following dif-
ferent grades in different areas of the liver. The latter phe-
nomenon produces areas with different echogenicity: fatty
areas are more echogenic and areas with less fat appear less
echogenic. Thus, a “geographic liver” may appear on US
examination. Ultrasound is the first option for fatty liver di-
agnosis, but its accuracy depends on the operator and the
patient's features [25].
Three grades of fatty infiltration exist (Fig. 9):
• Mild: minimal diffuse increase in liver echo-
genicity.
• Moderate: moderate diffuse increase in liver
echogenicity with slightly impaired visualisation of
the intrahepatic vessels and the diaphragm.
• Severe: marked diffuse increase in liver echo-
genicity with poor penetration of the deeper part of
the liver and poor or non visualisation of the intra-
hepatic vessels and the diaphragm [2].
1: Menu Y, Le Kremlin-Bicêtre, France. Liver Cirrhosis and Portal Hyper-
tension. E3 820 Interactive Image Teaching Session, European Congress of
Radiology 2007. Abstract: European Radiology Supplements 2007; 17(1):
56.
In practice these grades apply to diffuse fatty infiltration,
with liver echogenicity increasing as steatosis progresses
from mild to severe. In the last grade, a focal lesion in the
deeper part of the liver may be hard to locate. Focal fatty
infiltration and sparing may appear similar to neoplasms.
They most commonly involve the periportal region of seg-
ment IV, the gallbladder fossa and the liver margins [2].
Coarse echotexture is also a common finding in cirrhosis.
However this is a subjective feature and may be confounded
by incorrect time gain compensation (TCG) and total gain
[2].
4.3. Liver Nodules
Nodular irregularity of the liver surface is a common sign
of cirrhosis due to the presence of regenerating nodules. The
latter represent regenerating hepatocytes surrounded by fi-
brotic septa. Micronodular cirrhosis can evolve into macro-
nodular, thus producing the nodular hepatic surface, a feature
which is more prominent when ascites is present (Fig. 10).
Smaller nodules are usually hypoechoic, while larger nod-
ules may become iso- or hyperechoic. Generally they have
similar architecture to the surrounding normal liver and are
therefore not easily spotted on US and even computed tomo-
graphy (CT). Stiff fibrous liver parenchyma causes pressure
on hepatic and portal vein branches, which appear in smaller
numbers than normally. Usually regenerating nodules are
more echogenic than annular fibrosis, while more vessels are
seen in fibrosis than in regeneration nodules. These differ-
ences are better imaged when a high frequency linear probe
is used [22, 26].
Regenerating nodules can turn into adenomatous hyper-
plastic (dysplastic) nodules, which have a diameter over 10
mm and are considered premalignant [27]. They consist of
well differentiated hepatocytes with portal venous blood
flow and atypical or malignant cells [2]. Dysplastic nodules
may eventually result into hepatocellular carcinoma (HCC).
The hepatocarcinogenetic process may evolve for years in a
stepwise fashion from premalignant to overt HCC [28].
Therefore early detection in patients at risk is very important.
A 6-month interval surveillance with alpha-fetoprotein
(AFP) concentration and US [29] is considered cost-
effective, picking up a single tumour smaller than 3 cm in
50-70% of the patients at risk. For tumours larger than 2 cm
tumors, arterial hypervascularisation of the node by contrast
Fig. (9). Three grades of fatty infiltration: A. mild, B. moderate, C. severe.
68 Current Medical Imaging Reviews, 2009, Vol. 5, No. 1
Cokkinos and Dourakis
US, triphasic spiral CT or magnetic resonance (MR) is diag-
nostic for HCC. However, diagnosis of a tumour smaller
than 2 cm in diameter is more difficult, due to (up to 50%)
false negative contrast enhanced examinations caused by
immature arterial vascularisation of small nodules [28].
When a hypoechoic nodule with a diameter up to 1 cm is
discovered, a follow-up is mandatory, as for this size find-
ings are usually atypical on US. A liver biopsy should be
proposed in the case of an enlarging tumour [30].
Altogether, differential diagnosis between HCC and re-
generating nodules can be difficult by US alone, while MR is
more sensitive than CT and US in detecting regenerating
nodules [31]. However, the addition of US contrast agents
can increase sensitivity up to 98% and specificity up to
92.7% for differentiating benign from malignant lesions
when all 3 phases of enhancement are studied [32]: Malig-
nant lesions show early contrast enhancement (being fed by
the hepatic artery), while in the portal and late sinusoidal
phases they present as defects. On the contrary, benign le-
sions enhance in the same way as the normal hepatic paren-
chyma (blood flow coming from the portal vein) and con-
tinue to enhance in the delayed phases [33]. Therefore, HCC
shows arterial enhancement with subsequent wash-out in the
portal and/or delayed phase, while confluent fibrosis en-
hancement increases in the delayed phase.
The European Association for the Study of the Liver
(EASL) Barcelona 2000 Conference Conclusions [29] sug-
gest that in case of a hypervascular nodule smaller than 1 cm
in a patient with cirrhosis, follow-up imaging should be per-
formed with the same imaging methods 3-4 months later in
order to assess enlargement, stability or disappearance. Dur-
ing that time, no surgery or biopsy are needed. According to
the same conclusions, for nodules larger than 2 cm imaging
may confidently establish the diagnosis without biopsy con-
firmation. When problematic, a follow-up with the same
imaging method should be performed. However, other re-
searchers [34, 35] advocate the need for biopsy in hepatocel-
lular carcinoma, suggesting that fine needle biopsy and cy-
tology are safe procedures for liver mass diagnosis of liver
masses, provided that safety measures should be taken to
minimise seeding.
5. SPLENIC SIZE
Splenomegaly is observed in most but not all patients
with cirrhosis, more often when complicated by portal hy-
pertension [36]. US shows a sensitivity of up to 95% and a
specificity of up to 98% in measuring the liver and spleen
[37]. Mild to moderate splenomegaly (craniocaudal diameter
of more than 13 cm) is a common finding of portal hyperten-
sion [2]. However, although there is no complete correlation
between this finding and the pressure in the portal vein, [38]
monitoring of the spleen diameter may allow a prognostic
stratification of cirrhotic patients [36].
6.
PORTOSYSTEMIC
VENOUS
COLLATERAL
COMMUNICATIONS
When portal resistance is higher than that of small com-
municating channels between the portal and systemic circu-
lation, portosystemic collaterals are formed. This causes a
subsequent decrease in the, initially dilated, caliber of the
portal vein [8]. Forming of collateral vessels is a definitive
finding of portal hypertension, although it may be seen in
other pathologic entities, such as venous obstructive disease.
US can reveal up to 65-90% of these vessels [33].
The following sites of portosystemic collateral veins can
be seen by US (Fig. 11):
Fig. (10). Nodular irregularity of the hepatic surface caused by regenerating nodules. Ascites is also present.
Ultrasonographic Assessment of Cirrhosis
Current Medical Imaging Reviews, 2009, Vol. 5, No. 1 69
6.1. Paraumbilical Vein
It is located in the falciform ligament and connects the
left portal vein with the systemic epigastric vein close to the
umbilicus. When patent, it is a specific sign of portal hyper-
tension and may protect cirrhotic patients from forming oe-
sophageal varices [39, 40].
6.2. Gastroesophageal Junction
The anastomosis is formed between the coronary and
short gastric veins (portal system) and the systemic oeso-
phageal veins. These varices can cause very severe haemor-
rhage. A coronary vein diameter over 0.7 cm is a sign of
serious portal hypertension [2].
6.3. Splenorenal-Gastrorenal Area
Tortuous veins arise close to the splenic and left renal
hili, representing collaterals between the splenic, coronary
and short gastric veins (portal system) and the left adrenal
and renal veins (systemic venous system).
6.4. Intestinal
The gastrointestinal portal venous system becomes
retroperitoneal in certain areas, allowing anastomoses be-
tween veins from the ascending and descending colon, liver,
pancreas and duodenum to communicate with the systemic
renal, phrenic and lumbar veins.
6.5. Haemorrhoidal
In the perianal region the superior rectal veins, connected
to the inferior mesenteric vein (portal system) form anasto-
moses with the systemic middle and inferior rectal veins [2].
6.6. Other Collaterals
Less prominent collateral veins are formed between the
liver and the abdominal wall, as well as in the wall of the
gallbladder.
7. ASCITES
Normally, about 50-75 ml of free fluid is present in the
peritoneum, acting as a lubricant [2]. An excess in this
amount results in ascites, which is classified into transudate
and exudate. Cirrhosis, peritoneal carcinomatosis, congestive
heart failure and tuberculosis are the causes of over 90% of
ascites. Pathophysiology of ascites in a cirrhotic patient in-
cludes portal hypertension, proteinaemia, increased hepatic
lymph production and renal sodium retention [41]. When the
patient is in the supine position, fluid accumulates first in the
paracolic gutters, Douglas and Morison’s pouches.
CONCLUSION
Assessment of patients with cirrhosis and portal hyper-
tension is an everyday imaging challenge. US is the first
modality to be used and in many cases will set the diagnosis,
provided that an examination protocol will be followed. This
includes studying of the portal vein diameter, portal vein
flow, hepatic artery and veins flow, liver morphology, size of
spleen, portosystemic collateral vessels and ascites. US con-
trast agents improve imaging of regenerating nodules and
differential diagnosis from dysplastic nodules and HCC.
REFERENCES
[1]
Anthony PP. The morphology of cirrhosis: Definition, nomencla-
ture and classification. Bull WHO 1977; 55: 521.
[2]
Wilson SR, Withers CE. In: Rumack CM, Wilson SR, Charboneau
JW, Johnson JAM. Diagnostic ultrasound, third edition. St. Louis,
Elsevier Mosby 2005; 77-145.
[3]
Millward-Sadler GH. In: MacSween RNM, Anthony PP, Scheuer
PJ. Pathology of the liver, second edition. New York, Churchill
Livingstone 1987; 342-63.
[4]
Dick R, Watkinson A, Olliff JFC, et al. In: Sutton D. Textbook of
radiology and imaging, seventh edition. New York, Churchill Liv-
ingstone 2003; 737-86.
[5]
Bolondi L, Mazziotti A, Arienti V, et al. Ultrasonographic study of
portal venous system in portal hypertension and after portosystemic
shunt operations. Surgery 1984; 95(3): 261-9.
[6]
Ohnishi K, Saito M, Nakayama T, et al. Portal venous hemody-
namics in chronic liver disease: Effects of posture change and exer-
cise. Radiology 1985; 155: 757-61.
[7]
Bolondi L, Gandolfi L, Arienti V, et al. Ultrasonography in the
diagnosis of portal hypertension: Diminished response of portal
vessels to respiration. Radiology 1982; 142: 167-72.
[8]
Lafortune M, Marleau D, Breton G, et al. Portal venous system
measurements in portal hypertension. Radiology 1984; 151: 27-30.
[9]
Vine HS, Sequira JC, Widrich WC, et al. Portal vein aneurysm.
AJR Am J Roentgenol 1979; 132: 557-60.
[10]
Aldrete JS, Slaughter RL, Han SY. Portal vein thrombosis resulting
in portal hypertension in adults. Am J Gastroenterol 1976; 65: 3-11.
[11]
Dodd GD, Memel OS, Baron RL, et al. Portal vein thrombosis in
patients with cirrhosis. Does sonographic detection of intrathrom-
bus flow allow differentiation of benign and malignant thrombus?
AJR Am J Roentgenol 1995; 165: 573-7.
Fig. (11). Portosystemic collaterals: Patent umbilical vein (A. B-mode, B. Colour Doppler) and splenorenal veins (C. Colour Doppler US).
70 Current Medical Imaging Reviews, 2009, Vol. 5, No. 1
Cokkinos and Dourakis
[12]
Bolondi L, Bassi S, Gaiani S, et al. Liver cirrhosis: Changes of
Doppler waveform of hepatic veins. Radiolgy 1991; 178: 513-6.
[13]
Colli A, Cocciolo M, Riva C, et al. Abnormalities of Doppler
waveform of the hepatic veins in patients with chronic liver dis-
ease: Correlation with histologic findings. AJR Am J Roentgenol
1994; 162: 833-837.
[14]
Sudhamshu KC, Matsutani S, Maruyama H, et al. Doppler study of
hepatic vein in cirrhotic patients: Correlation with liver dysfunction
and hepatic hemodynamics. World J Gastroenterol 2006; 12(36):
5853-8.
[15]
Albrecht T, Blomley MJ, Cosgrove DO, et al. Non-invasive diag-
nosis of hepatic cirrhosis by transit-time analysis of an ultrasound
contrast agent. Lancet 1999; 353: 1579-83.
[16]
Abbattista T, Ridolfi F, Ciabattoni E, et al. Diagnosis of liver cir-
rhosis by transit-time analysis at contrast-enhanced ultrasonogra-
phy. Radiol Med 2008; 113(6): 860-74.
[17]
Lim A, Cosgrove D. Functional studies. Eur Radiol Suppl 2004;
[Suppl 8]: P110-5.
[18]
Haskal ZJ, Carroll JW, Jacobs JE, et al. Sonography of transjugular
portosystemic shunts: Detection of elevated portosystemic gradi-
ents and loss of shunt function. J Vasc Interv Radiol 1997; 8: 549-
56.
[19]
Becker CD, Cooperberg PL. Sonography of the hepatic vascular
system. AJR Am J Roentgenol 1968; 150: 999-1005.
[20]
Zwiebel WJ. In: Zwiebel WJ. Introduction to vascular ultrasono-
graphy. 4th edition. Philadelphia, WB Saunders. 2000; 431- 54.
[21]
Giorgio A, Amoroso P, Lettiri G, et al. Cirrhosis: Value of caudate
to right lobe ratio in diagnosis with ultrasound. Radiology 1986;
161: 443-5.
[22]
Lafortune M, Matricardi L, Denys A, et al. Segment 4 (the quad-
rate lobe): a barometer of cirrhotic liver disease at US. Radiology
1998; 206(1): 157-60.
[23]
Okazaki H, Ito K, Fujita T, et al. Discrimination of alcoholic from
virus-induced cirrhosis on MR imaging. AJR Am J Roentgenol
2000; 175(6): 1677-81.
[24]
Taylor KJW, Riely CA, Hammers L, et al. Quantitative ultrasound
attenuation in normal liver and in patients with diffuse liver dis-
ease: Importance of fat. Radiology 1986; 160: 65-71.
[25]
Roldan-Valadez E, Favila R, Martínez-López M, et al. Imaging
techniques for assessing hepatic fat content in nonalcoholic fatty
liver disease. Ann Hepatol 2008; 7(3): 212-20.
[26]
Harbin WP, Robert NJ, Ferrucci JT Jr. Diagnosis of cirrhosis based
on regional changes in hepatic morphology: a radiological and
pathological analysis. Radiology 1980; 135(2): 273-83.
[27]
Theise ND. Macroregenerative (dysplastic) nodules and hepatocar-
cinogenesis: Theoretical and clinical considerations. Semin Liver
Dis 1995; 15: 360-71.
[28]
Colombo M. Natural history of hepatocellular carcinoma. Ann Ital
Chir 2008; 79(2): 91-7.
[29]
Bruix J, Sherman M, Llovet JM, et al. Clinical Management of
Hepatocellular Carcinoma. Conclusions of the Barcelona-2000
EASL Conference. J Hepatol 2001; 35: 421–30.
[30]
Zarski JP, Doffoel M, Filoche B, et al. [Hepatitis C, cirrhosis and
hepatocellular carcinoma].[Article in French]. Gastroenterol Clin
Biol 2008; 32(3 Pt 2): S117-20.
[31]
Murakami T, Nakamura H, Hori S, et al. Regenerating nodules in
hepatic cirrhosis. MR findings with pathologic correlation. AJR
Am J Roentgenol 1990; 155: 1227-31.
[32]
Nicolau C, Vilana R, Catalá V, et al. Importance of evaluating all
vascular phases on contrast-enhanced sonography in the differen-
tiation of benign from malignant focal liver lesions. AJR Am J
Roentgenol 2006; 186(1): 158-167.
[33]
Cosgrove D, Blomley M. Liver tumors: evaluation with contrast-
enhanced ultrasound. Abdom imaging 2004; 29(4): 446-54.
[34]
Llatjos M, Muns R, Tallada N. Need for biopsy in hepatocellular
carcinoma. J Hepatol 2002; 37(6): 874.
[35]
Bruix J, Sherman M. Diagnosis of Small HCC. Gastroenterology
2005; 129(4): 1364.
[36]
Berzigotti A, Zappoli P, Magalotti D, et al. Spleen enlargement on
follow-up evaluation: A noninvasive predictor of complications of
portal hypertension in cirrhosis. Clin Gastroenterol Hepatol 2008;
6(10): 1129-34.
[37]
Safak AA, Simsek E, Bahcebasi T. Sonographic assessment of the
normal limits and percentile curves of liver, spleen, and kidney di-
mensions in healthy school-aged children. J Ultrasound Med 2005;
24(10): 1359-64.
[38]
Koslin DB, Mulligan SA, Berland LL. Duplex assessment of the
splanchnic vasculature. Semin Ultrasound CT MR 1992; 13(1): 34-
9.
[39]
DiCandio G, Campatelli A, Mosca F, et al. Ultrasound detection of
unusual spontaneous portosystemic shunts associated with uncom-
plicated portal hypertension. J Ultrasound Med 1985; 4: 297-305.
[40]
Mostbeck GH, Wittich GR, Herold C, et al. Hemodynamic signifi-
cance of the paraumbilical vein in portal hypertension: Assessment
with duplex ultrasound. Radiology 1989; 170: 339-42.
[41]
Harrison's Principles of Internal Medicine Companion Handbook,
eleventh edition. New York, McGraw-Hill 1988; 72.
?
Received: October 18, 2008
Revised: November 30, 2008
Accepted: January 01, 2009
Không có nhận xét nào:
Đăng nhận xét