Updated: Feb 5, 2009
Portal hypertension (PH) represents an increase of the hydrostatic pressure within the portal vein or its tributaries. It is defined as an increase in the pressure gradient between the portal vein and hepatic veins or the inferior vena cava (IVC).
Hematemesis resulting from bleeding esophageal varices is the most common presentation in patients with PH, although some patients seek medical help because of decompensated liver disease. Findings from duplex ultrasonography (US) and color Doppler imaging (CDI), MRI, CT, and endoscopy may support the diagnosis of PH.
Related eMedicine topics:Portal hypertension (PH) is defined as an increase in pressure gradient between the portal vein and the hepatic veins or inferior vena cava (IVC).1 A pressure gradient of 12 mm Hg is regarded as clinically significant for PH. With such an increase in the pressure gradient, portosystemic collaterals develop. These bypass portal venous flow to the liver, diverting blood to the systemic veins.
PH may develop in a variety of clinical circumstances, but by far, most instances are related to cirrhosis.
The causes of PH may be divided into (1) posthepatic, (2) prehepatic, and (3) intrahepatic causes. The major causes of posthepatic causes of PH are right-sided heart failure, constrictive pericarditis, and Budd-Chiari syndrome (BCS). The prehepatic causes of PH include portal vein thrombosis (PVT) and portal compression or occlusion by biliary and pancreatic neoplasms and metastases. PH may be caused by an increase in flow secondary to arterioportal fistula, pancreatic arteriovenous malformations, and massive splenomegaly. The most common intrahepatic cause is cirrhosis. The common feature of all the causes is an increase in resistance to portal venous flow, although in a few cases, increased inflow into the portal venous system is present.
The basis of PH in patients with cirrhosis is an increase in resistance to portal venous flow at the level of the sinusoids as a result of the perisinusoidal deposition of collagen. This deposition results in narrowing and compression of the central veins caused by fibrosis. Pressure from regenerative nodules contributes to this compression. Arteriovenous anastomoses in a fibrous scar also contribute to the increase in portal venous pressure.
Cirrhosis has many causes. In patients with alcoholic liver disease, serious PH may develop in the absence of cirrhosis. In a minority of patients with acute alcoholic hepatitis, progressive obliteration of central veins occurs with resultant centrilobular fibrosis. These changes result in severe outflow obstruction to portal venous flow, which leads to the formation of ascites and esophageal varices. Alcoholism and malnutrition result in nodular or Laennec-type cirrhosis; bile duct obstruction results in biliary cirrhosis; schistosomiasis, prolonged congestive heart failure, and hepatitis cause postnecrotic cirrhosis. Whatever the cause of cirrhosis from a hemodynamic viewpoint, the common denominator is a progressive increase in resistance to portal venous blood flow that results in PH.
Chronic portal vein obstruction is an important cause of PH; it has a variety of causes. Cavernous transformation of the portal vein (ie, development of periportal collaterals) occurs in patients with long-standing PVT as a result of the development of multiple small vessels in and around the recanalizing or occluded main portal vein. A leash of fine or markedly enlarged serpiginous vessels is seen in place of the portal vein. Color Doppler and/or pulsed Doppler imaging shows blood flow in these periportal collaterals that form around the thrombosed portal vein or that replace the vein.
Causes of PVT include the following:
Splenic vein thrombosis may be caused by abdominal trauma, tumors, or pancreatitis. Pressure is increased in areas drained by the splenic vein, while pressure in the portal vein remains normal (left-sided PH). The diagnosis is suspected in patients with gastric and/or esophageal varices and normal liver biopsy results. Splenectomy is curative.
BCS is a manifestation of hepatic venous outflow obstruction. The hepatic outflow obstruction usually occurs at the level of the IVC; the hepatic vein; and, depending on the classification and the nomenclature, possibly at the venules. The causes of BCS are numerous. Two types exist: acute and chronic. An acute thrombosis of the main hepatic veins or the IVC causes the acute form. The chronic form is related to fibrosis of the intrahepatic veins, which is presumably related to inflammation. The classical presentation is with ascites, hepatomegaly, and abdominal pain.
Hepatic venous occlusion results in the elevation of sinusoidal pressure, which leads to a delay or reversal of portal venous blood inflow, ascites, and morphologic changes in the liver. The last changes are reflected in abnormal liver function results. Both the acute and chronic forms result in severe centrilobular congestion, hepatocellular necrosis, and atrophy. Two other entities with similar clinical characteristics are severe right-sided heart failure and veno-occlusive disease of the liver. Hepatic veno-occlusive disease is characterized by inflammation of the post-sinusoidal venules, which results in fibrosis and venous occlusion. BCS has been variously classified. Some authors distinguish between primary BCS, which is associated with IVC webs, and secondary BCS, which is ascribed to a myriad of causes, including tumor, thrombosis, and trauma.
Other authors classify the disease according to the location of obstruction as follows:
The causes of BCS are numerous. Examples include the following:
In most cases of BCS, the hepatic venous outflow is not completely eliminated, because a variety of accessory hepatic veins drain into the IVC above or below the site of obstruction. The most common of these accessory veins are the accessory inferior hepatic and caudate veins, which drain into the IVC inferior to the major hepatic veins. Vascular communications also exist via the azygos, intercostal, and paravertebral veins, which provide alternative pathways for hepatic venous drainage in BCS. Intrahepatic communication between the hepatic veins and portal veins also reverse flow in some portal venous branches, though flow in the main portal vein tends to remain centripetal.
As in BCS, some hepatic venous drainage is preserved, and the caudate lobe hypertrophies, sometimes massively; this may produce secondary IVC obstruction. Other parts of the liver with preserved venous drainage may also undergo hypertrophy.
Some venous drainage also occurs via the capsular veins, but this drainage is not sufficient to prevent peripheral atrophy in BCS. Parts of the liver with complete obstruction of its venous drainage tend to drain via the portal vein branches, depriving the involved parts of the liver of portal venous blood supply and the trophic effects of hormones. Hepatic hypertrophy and regeneration are always dependent on the trophic effect of portal blood supply. Thus, BCS is typically associated with peripheral atrophy of the liver and caudate and central hypertrophy. On cross-sectional imaging, the porta hepatis may be displaced anteriorly in BCS. Concomitant portal vein thrombosis (PVT) may be present in 9-20% of patients with BCS.
In idiopathic PH (Banti syndrome), no cirrhosis occurs, and the portal vein is patent. Liver biopsy results may be normal or may show fibrosis in the periportal areas and in the space of Disse; dilatation of sinusoids and intimal thickening with eccentric sclerosis of peripheral portal vein walls may be noted. The disease is progressive with increasing PH, and the liver eventually becomes small and fibrotic. A number of exogenous toxins such as copper salts, vinyl chloride, and arsenic may create a similar clinical picture.
Congenital hepatic fibrosis usually appears in childhood in association with autosomal recessive polycystic kidney disease, medullary sponge kidney, or Caroli disease. PH is a major consequence of this form of liver disease. Hepatic function is well maintained. Histologically, fibrous tissue is present within the hepatic parenchyma, with excess numbers of distorted terminal interlobular bile ducts and cysts that do not communicate with the bile ducts. In rare cases, hepatic sarcoidosis leads to hepatic fibrosis and PH.
When deprived of portal venous blood, the liver depends more on hepatic arterial blood. The liver undergoes atrophic changes and shrinks, and its capacity to regenerate is impaired. The changes are related to the lack of hepatotrophic hormones, largely insulin and glucagon, which are responsible for maintaining the normal structure and function of the liver. Vessels in the collateral venous circulation, particularly esophageal varices, are usually not prominent at autopsy because the veins collapse after death.
The spleen is enlarged and firm, with a thickened capsule. Malpighian bodies are inconspicuous. Histologically, sinusoids are dilated and lined by thickened epithelium. Histiocytes proliferate in the sinusoids with occasional erythrophagocytosis. Periarterial hemorrhages may progress to siderofibrotic nodules. The splenic artery and portal vein are enlarged and tortuous; they may be aneurysmal. The portal and splenic veins may show endothelial hemorrhages, mural thrombi, and intimal plaques, which may become calcified. These veins may be unsuitable for portosystemic shunt procedures.
Small, intrasplenic arterial aneurysms are seen in 50% of patients with cirrhosis. Hepatic changes depend on the cause of the PH. The degree of portal venous pressure is poorly correlated with the apparent degree of cirrhosis and fibrosis; a much better correlation is made with the degree of nodule formation.
The frequency of portal hypertension is related to the frequency of cirrhosis. Alcohol intake is the most common cause of liver disease in Western nations. Alcoholic cirrhosis is discovered in 1.6 – 9.9% of autopsies in the United States.
The exact worldwide incidence of cirrhosis is not known, but large regional variations occur, depending on the frequency of hepatitis B and hepatitis C.
Mortality and morbidity of portal hypertension (PH) are related to the underlying cause (eg, cirrhosis, portal vein thrombosis, splenic vein thrombosis, veno-occlusive disease). Hemorrhage caused by esophageal varices is a major complication of PH. Mortality rates in adults with cirrhosis vary; the rate ranges from 30 – 60% for each bleeding episode.
The most common cause of cirrhosis in North America is alcohol intake. In Africa, the Middle East, and the Far East, cirrhosis is virus related. Noncirrhotic idiopathic portal hypertension is more common in India and Japan.
Alcohol-related cirrhosis frequently affects males, although the incidence in females is increasing. The male-to-female ratio is 2:1. Primary biliary cirrhosis has a female preponderance (>90%).
The incidence of alcoholic cirrhosis peaks in patients 40 – 55 years of age; however, patients 20 – 30 years of age may also have advanced alcoholic liver disease. Primary biliary cirrhosis is found in patients as young as 23 years and as old as 72 years; however, most patients are 40 – 60 years of age.
The portal venous system includes all veins that carry blood from the abdominal part of the alimentary tract, spleen, pancreas, and gallbladder. The union of the superior mesenteric and splenic veins forms the portal vein posterior to the pancreatic head. The portal vein enters the liver at the porta hepatis and soon divides into left and right portal vein branches. When portal circulation is obstructed, a remarkable collateral circulation develops, redirecting portal venous blood into systemic veins.
Portosystemic collaterals are classified into 4 main groups, as follows:
Spontaneous splenorenal shunts are seen in 10 – 20% of patients with PH. Blood from gastroesophageal collaterals and retroperitoneal and venous systems of the abdomen ultimately reaches the superior vena cava via the azygos or hemiazygos system. A small volume of blood enters the IVC. Collaterals feeding into the pulmonary veins have been described.
The presence of portosystemic anastomosis usually implies PH, although occasionally, if the collateral circulation is extensive, portal pressure may decrease. Conversely, PH of short duration may exist without demonstrable collateral circulation.
The collateral pathways in extrahepatic venous obstruction depend on the site of obstruction. In the absence of liver disease, with splenic vein occlusion, portal-portal collateral pathways develop over gastric veins (splenic, short gastric, coronary portal veins) and omental veins (splenic, gastroepiploic or arch of Barkow, superior mesenteric portal veins). In superior mesenteric vein obstruction, collaterals develop via the pancreaticoduodenal or cystic veins. In portal vein occlusion, collaterals develop via peribiliary venous plexus, via veins in the hepatoduodenal ligament, and in the hepatic hilus.
Etiology of PH
The causes of portal hypertension are many; they may be subdivided into diseases causing prehepatic, hepatic, or posthepatic PH. The causes include alcoholic cirrhosis, schistosomiasis, hepatocellular carcinoma, Budd-Chiari syndrome (BCS), hepatoveno-occlusive disease, hepatitis and other chronic liver diseases, and the many causes of PVT (portal vein thrombosis).
Causes of PH include the following:
Clinical presentation
PH may be accompanied by the following 3 major complications:
Hepatic encephalopathy is devastating and usually results from gastrointestinal bleeding, which is life threatening. The development of hepatic encephalopathy in patients with portal hypertension is most often related to the size of the portosystemic anastomosis. Splenoportal shunts are usually large. Cirrhosis of the liver is the most common cause of PH. Stigmata of cirrhosis, including jaundice, spider nevi, caput medusae, and palmar erythema, may be associated with signs of PH.
Plain radiographs are not often obtained in cases of portal hypertension, but because most hospitalized patients undergo chest radiography, radiologists need to be aware of abnormalities that may be found in patients with PH. The appearance of calcification in the distribution of the portal vein on a plain abdominal radiograph may indicate PH. An upper GI tract barium series is often performed for the detection of esophageal varices (see Images 1-4).
Angiographic techniques such as splenosportography (SP), transhepatic portography, transumbilical catheterization, transjugular catheterization, wedge hepatic venography, and arterial portography are invasive. However, they are much more specific for the evaluation of PH hypertension; they are indicated when definitive surgery or radiologic intervention is contemplated.
The use of angiographic techniques is declining because noninvasive imaging techniques such as US, CT, computed tomographic angiography (CTA), and magnetic resonance angiography (MRA) are now available. These techniques are quickly improving, and this will lead to further decline in the use of angiographic methods.
Splenoportography and transumbilical catheterization are rarely performed. Arterial portography (indirect portography) and wedge hepatic venography with manometry is indicated before surgical portacaval shunt placement.
Carbon dioxide wedge hepatic venography is the most commonly used method for visualizing the portal vein before portal vein puncture for a transjugular intrahepatic portosystemic shunt (TIPS) procedure. TIPS is a radiology-guided creation of a shunt between the portal and hepatic veins in the liver by use of a percutaneous transjugular approach. Because of its proven safety and effectiveness, TIPS has largely replaced surgical decompressive shunt procedures.2,3,4
Plain radiographs are usually not indicated for patients with portal hypertension. Most plain radiographs are obtained for other reasons, and signs of PH are detected incidentally. Therefore, plain radiographs are of limited value.
Duplex US is a sensitive technique for the detection of PH in addition to other important features. When respiratory variation in the size of the portal, splenic, and superior mesenteric veins does not occur or when it is less than 20%, PH may be diagnosed with a sensitivity of 80% and a specificity of 100%.
In cases involving bleeding varices that are unresponsive to endoscopic sclerotherapy or when intractable ascites are present, a TIPS procedure is indicated. TIPS is performed after portal vein patency is documented at duplex US.
Budd-Chiari Syndrome
Cirrhosis
Congestive Heart Failure
Constrictive Pericarditis
Portal Vein Thrombosis
Veno-occlusive disease
Splenic vein thrombosis
Splenomegaly not resulting from liver disease
Arteriovenous fistula
Web lesion or thrombosis of the IVC
Idiopathic PH
Plain radiographic findings
Barium study findings of esophageal varices
Barium study findings of varices in other parts of the gastrointestinal tract
Gastric antral and duodenal varices are sometimes seen, usually in association with gastric fundal and esophageal varices.
Plain radiography is not sensitive in the diagnosis of portal hypertension. With good operator technique, barium examination may depict more than 90% of varices. The rate of detection with barium study has been reported to be 65 – 89%.
Other causes of calcification overlying the liver may mimic portal vein calcification. One example is hepatic arterial calcification.
Computed tomography
CT has been used to assess portal hypertension (see Images 10-11, 21-24).5,6,7,8,9
CT angiography
CTA is an exciting new application of helical CT. The speed of helical CT allows the maintenance of a higher concentration of intravenous contrast medium, particularly through the arterial enhancement phase, with the capability of 3-dimensional reconstruction. Both peripheral intravenous injections of contrast material and CT arterial portography have been used as a basis for CTA.
CTA has shown great promise in the evaluation of hepatic vessels before liver resection. It provides preoperative surgical information regarding the segmental location of liver tumors, the segmental venous anatomy, and significant arterial anomalies if present. The value of CTA for patients with PH remains unclear.
A diagnosis of portal hypertension frequently is made on the basis of the demonstration of signs of cirrhosis. Splenomegaly and ascites are demonstrated readily on CT; however, CT findings in cases of cirrhosis are highly variable. In some patients, the liver is of nonuniform attenuation; hypoattenuating components probably represent residual foci of infiltration.
Often, the attenuation of a cirrhotic liver is homogeneous and within the reference range. In the absence of obvious abnormalities of size and shape, the liver may appear entirely normal on CT scans. Nonenhanced CT is necessary when the cirrhotic liver is imaged for the identification of confluent fibrosis. Confluent fibrosis characteristically demonstrates low attenuation, which tends to become isoattenuating or minimally hypoattenuating on CT following intravenous contrast enhancement. Thus, confluent fibrosis is frequently missed if only contrast-enhanced CT is used.
Collateral veins are occasionally seen in the peritoneal cavity, retroperitoneum, abdominal wall, and mediastinum. CT lacks the dynamic capability of angiography in demonstrating the exact sites of portosystemic shunts and the feeding vessels.
In patients with cirrhosis, CT scans may appear entirely normal. Other causes of diffuse liver disease, such as splenomegaly and ascites, must be considered.
Magnetic resonance angiography
Magnetic resonance imaging
In patients with elevated creatinine levels who cannot undergo a CT scan with intravenous contrast enhancement, MRI with gadolinium enhancement may often be performed. These contrast-enhanced images are typically breath-hold fast spoiled gradient-echo sequences, which may be dynamically obtained in both the arterial and portal venous phases.10,11,12
Gadolinium-based contrast agents (gadopentetate dimeglumine [Magnevist], gadobenate dimeglumine [MultiHance], gadodiamide [Omniscan], gadoversetamide [OptiMARK], gadoteridol [ProHance]) have been linked to the development of nephrogenic systemic fibrosis (NSF) or nephrogenic fibrosing dermopathy (NFD). For more information, see the eMedicine topic Nephrogenic Fibrosing Dermopathy. The disease has occurred in patients with moderate to end-stage renal disease after being given a gadolinium-based contrast agent to enhance MRI or MRA scans.
NSF/NFD is a debilitating and sometimes fatal disease. Characteristics include red or dark patches on the skin; burning, itching, swelling, hardening, and tightening of the skin; yellow spots on the whites of the eyes; joint stiffness with trouble moving or straightening the arms, hands, legs, or feet; pain deep in the hip bones or ribs; and muscle weakness. For more information, see the FDA Public Health Advisory or Medscape.
MRI does not offer an advantage over CT in the diagnosis of cirrhosis; however, morphologic changes identified on US or CT are clearly depicted on MRI. One advantage of MRI over CT is its capacity to characterize regenerative nodules in a cirrhotic liver. MRA may offer a noninvasive, non–operator dependent evaluation of the portal venous system; however, the exact sensitivity and specificity still need to be determined.
Portal vein waveforms
The portal vein typically shows continuous forward flow with minor modulation caused by respiration and transmitted arterial pulsation. The rate of blood flow is greatest at the center of the vessel and is least at its margins; therefore, the average flow across a segment of the vessel is used to overcome the internal variation resulting from respiration and transmitted cardiac pulsation. The time-averaged velocity usually is on the order of 12–14 cm/s in adults.
Flow within the portal vein changes rapidly in response to eating. Blood flow to the small bowel begins to increase within 2 minutes of taking the first mouthful of food. The time-averaged blood flow velocity in the portal vein may increase to 25 cm/s after a meal, and with minor increase in vessel diameter, the flow volume may increase 4–5 times.
Most patients with mild hepatic parenchymal disease have normal portal venous blood flow. As hepatic disease becomes more severe, the first detectable flow abnormality is a reduction in the level of increase in flow seen after a meal. Then, the splenic and superior mesenteric veins may begin to distend, and the change in vessel caliber that is seen normally with respiration is lost. In severe hepatic parenchymal disease, portal venous blood flow is reduced, and a rough correlation is noted between the degree of reduction in portal flow velocity and the severity of hepatic parenchymal disease (providing that studies are performed in strictly fasting patients).
As portal venous flow is compromised further, forward flow may be seen only during systole, with reversed flow occurring during diastole. Eventually, flow within the portal vein may be reversed continuously, but the rate and direction of flow may vary from day to day, particularly in patients with acute exacerbation of chronic liver disease. Therefore, serial examinations provide a better picture than single scans. Clearly reduced portal flow velocity is associated with an increase in the risk of thrombosis.
Portal venous shunts
Percutaneous creation of a shunt between the portal and hepatic veins with TIPS is becoming increasingly common as a treatment for PH. Doppler US may be used to assess patency and show the direction of flow within the portal vein and shunt. If the walls of an artificial shunt prevent direct analysis of blood flow within it, an analysis of flow at the ends of the shunt usually allows adequate assessment of shunt function.
Surgically created shunts include portocaval (end or side of main portal vein to IVC), mesocaval, and splenorenal shunts (Warren shunt). Mesocaval and splenorenal shunts are used more frequently to maintain portal venous patency for transport than for other purposes. Color Doppler imaging is valuable for the evaluation of shunts.
US evaluation of PH
The diameter of the portal vein is measured with the patient in a supine position, in quiet respiration, and having fasted for a minimum of 4 hours. Measurements are made at the point at which the portal vein crosses the IVC. In an individual without portal hypertension (PH), the diameter of the portal vein is 13 mm or 16 mm during deep inspiration.
Under standard conditions, measurements greater than 13 mm indicate PH with a specificity of 100% but a low sensitivity of 45–50%. Sensitivity may be increased to 81% by measuring splenic vein and superior mesenteric vein diameters. An increase of 20–100% in diameter during deep inspiration is normal. An increase of less than 20% is associated with PH.
The differential diagnosis of a dilated portal vein includes PH splenomegaly (whatever the cause), acute portal vein thrombosis (PVT), and postprandial increase in portal vein diameter.
Portal flow direction and velocity
Usually, blood flow in the portal vein is hepatopetal (toward the liver) during the entire cardiac cycle. The mean velocity is 15–18 cm/s and varies with cardiac cycle. In PH, velocity fluctuations disappear, resulting in continuous flow. With a further increase in portal venous pressure, the blood flow direction becomes to-and-fro (biphasic), and finally, the direction is reversed (hepatofugal).
Differential diagnosis of hepatofugal portal venous flow includes PH, Budd-Chiari syndrome (BCS), side-to-side portocaval shunts, surgical or spontaneous splenorenal shunts with cirrhosis, and tricuspid regurgitation (tricuspid flow reversal). Differential diagnosis of portal venous flow reversal includes severe PH, tricuspid regurgitation, and congestive heart failure.
The differential diagnosis of reversal of hepatofugal-to-hepatopetal portal flow includes eating and the use of drugs that increase portal flow. Static flow without Doppler signal occurs occasionally.
Pulsatile portal vein flow
A pattern similar to that seen in patients with impaired right heart function occasionally is seen in patients with cirrhosis and/or PH. Patients with right-sided cardiac dysfunction with pulsatile portal venous flow invariably have abnormal liver function.
The differential diagnosis of pulsatile portal venous flow includes tricuspid regurgitation, aortic–right atrial fistula, fistula between the portal vein and hepatic vein, PH, and congestive heart failure. Rarely, it is a false-positive finding.
Decreased volume flow in the portal vein
In mild to moderate PH, the volume of flow in the portal vein is maintained. A reduction in volume flow occurs with advanced cirrhosis when intrahepatic obstruction to portal flow is severe, as indicated by hepatofugal flow and extensive portosystemic collaterals.
Congestive Index
PH may be recognized by use of the congestive index, in which the ratio of the portal vein (in units of square centimeter) is divided by the mean portal flow velocity (in units of centimeter per second). This ratio reflects the physiologic changes that occur in PH (ie, portal vein dilatation associated with diminished flow velocity). In individuals without PH, the ratio should not exceed 0.7.
Splenomegaly
The size of the spleen is not well correlated with the level of PH; however, if splenomegaly is absent, PH is unlikely. The spleen is best measured in the coronal plane. In the midaxillary line, a cephalocaudal measurement greater than 13 cm suggests enlargement.
Splenic interface sign
Linear reflective channels are observed in the splenic parenchyma in a variable number of patients with PH. Channels may be explained by dilatation of intrasplenic venous sinuses with increased collagen in the walls and by periarterial fibrosis. The pathologic changes are known to occur in PH. The splenic interface sign seldom is found in patients with splenomegaly that is unrelated to PH. The vascular nature of channels is readily confirmed by using CDI.
Ascites
Uncomplicated PH usually does not cause ascites. Usually, ascites occurs secondary to underlying liver diseases with liver cell failure.
Arterialization of hepatic blood supply
Hepatic arteries are enlarged and usually have aliased frequency shifts compared with those of normal hepatic arteries. The arteries also appear tortuous. As portal venous flow to the liver decreases, arterial flow increases. Increased arterial flow occurs with the development of large collaterals and hepatopetal flow.
The differential diagnosis of an enlarged hepatic artery includes an occluded or interrupted portal vein, a surgical portosystemic shunt, reversal of flow in the portal vein, parenteral feedings in newborns, hereditary hemorrhagic telangiectasia, cirrhosis or hepatic diseases associated with alcohol, vascular hepatic tumors, and primary hepatic artery dissection.
Portosystemic venous collaterals
The demonstration of portosystemic venous collaterals (PVCs) is diagnostic of PH. US findings of PVCs are as follows:
Banti syndrome
Banti syndrome, or noncirrhotic idiopathic PH, is a common cause of PH in India and Japan but is rare in the United States and Europe. The syndrome is characterized by signs of PH, but liver function test results tend to remain normal. Hepatic wedge pressure readings are usually normal or slightly elevated. Signs of hypersplenism are often present. US shows a normal-appearing liver; patent hepatic veins; and a patent portal vein, which may be associated with multiple portosystemic collaterals.
Portal venous thrombosis
PVT is being recognized with increasing frequency on US images. Reduced portal blood flow resulting from hepatic parenchymal disease and abdominal sepsis are the primary causes. Transient PVT is also being recognized with increasing frequency, in part because of the large increase in the use of US in evaluating patients with abdominal inflammation, such as appendicitis. Tumor within the portal vein may appear identical to thrombosis, but it is far less common. Tumor within the portal vein is most frequently related to a hepatocellular carcinoma, which gives rise to serpiginous filling defects in the portal venous luminal flow, but it usually persists around the tumor without complete occlusion.
Adults with acute PVT secondary to abdominal sepsis completely recover with vessel recanalization after successful treatment of underlying sepsis. In children, the portal vein may recanalize by developing multiple small collateral channels, which are seen as a partly echogenic band of small vessels running to the porta hepatis. These show reduced flow velocity of 2–7 cm/s. Nonvisualization of the portal vein is strongly suggestive of occlusion. Then, the portal vein may be seen as a band of high-level echoes at the porta hepatis.
Causes of PVT include idiopathic causes, malignancy (hepatocellular carcinoma, cholangiocarcinoma, pancreatic carcinoma, and stomach carcinoma), trauma (which may be iatrogenic, such as umbilical vein catheterization), and abdominal sepsis (pancreatitis, perinatal sepsis, omphalitis, appendicitis, diverticulitis, cirrhosis), especially in younger individuals.
On sonograms, clot exhibits variable echogenicity; it may be hypoechoic or anechoic if it has recently formed. Conversely, patent vessels may show increased intraluminal echogenicity because of artifact or erythrocyte rouleaux formation. In isolation, increased or decreased echogenicity in the lumen of the portal vein is not sufficient to make a diagnosis of PVT or to exclude PVT. In patients with PVT, the venous flow signal that is typically obtained from the lumen of the portal vein during pulsed or color-flow Doppler imaging is absent, and flow may be seen around a thrombus that partially occludes the vein. However, if flow is sluggish, the Doppler signal may not be detected. Color flow may be demonstrated in other small collaterals.
Incomplete occlusion (common in neoplastic invasion) or thrombolytic recanalization may occur. These cannot be differentiated by use of US. Hepatopetal flow may be demonstrated, and spontaneous shunts may be visualized (splenorenal).
US may demonstrate portosystemic collaterals, and the cause may be identified, such as hepatocellular carcinoma, metastases, cirrhosis, and pancreatic neoplasms. The incidence of PVT is reported to be low in patients with PH, but it is associated with sclerotherapy. The string sign, or thickening of the portal vein with narrowing of its lumen, which is interpreted as a portal phlebitis, is considered a precursor of PVT in patients with acute pancreatitis. The portal vein thrombus may become calcified.
Cavernous transformation of the portal vein
Cavernous transformation occurs in patients with long-standing PVT (19%) when numerous multiple collateral vessels develop around the occluded portal vein; these collateral vessels appear as vermiform tubular structures at the porta hepatis. Application of color-pulsed Doppler imaging shows blood flow in the periportal collaterals around the thrombosed portal vein. Cavernous transformation of the portal vein has been reported to appear as a subhepatic spongelike mass. This appearance also has been reported in patients with pancreatic hemangiosarcoma.
Bile duct varices, also called the pseudocholangiocarcinoma sign, are not infrequently observed during endoscopic retrograde cholangiopancreatography. In patients with PH, they result from cavernous transformation of the portal vein. Studies report that the sign may disappear after a TIPS procedure.
Assessment of TIPS
TIPS involves the percutaneous placement of a shunt via the jugular vein. TIPS is becoming popular as a definitive procedure for decompressing the portal venous system or as a prelude to liver transplantation.
Doppler US is a sensitive and relatively specific means of evaluating TIPS malfunction. US evaluation of the shunt is usually performed within 24 hours after shunt placement to establish baseline velocities within the portal vein, hepatic vein, and shunt. Follow-up studies are usually performed at 3-month intervals unless the clinical setting dictates a more emergent examination. The primary object of Doppler study of a TIPS is to document flow in the shunt and to demonstrate stenosis. The accuracy of Doppler US in shunt malfunction depends on several US parameters, which include the peak shunt velocity, distal shunt velocity, and antegrade flow in the left and right portal veins.
Flow velocities in the portal vein may double in comparison with the preoperative velocities in a successful TIPS placement. Direct observation of shunt thrombosis is possible with duplex or color Doppler US. Echo-enhanced color Doppler US may also be helpful in the assessment of TIPS.
Complications of TIPS that are detectable with US include the following:
The demonstration of portosystemic shunts is a specific sign of portal hypertension. Umbilical vein collateral flow is an important feature of PH because it has a specificity of 100%. In individuals without PH, the caliber of portal, splenic, and superior mesenteric veins shows significant variation during respiration. When respiratory variation does not occur or when it is slight (<20%), PH may be diagnosed with a sensitivity of 80% and a specificity of 100%. A portal vein diameter greater than 13 mm is specific for PH, but it is not a sensitive indicator of PH.
Angiographic evaluation of the portal venous system may be performed (see Images 5-6, 14-20, 25-26).
Splenoportography
Recently, because of the use of alternative imaging methods, the use of splenoportography (SP) has declined considerably after having played a primary role in the investigation of cirrhosis and PH for several years. However, restricted indications remain for the procedure. Even with the available methods of direct or indirect visualization of the portal vein, much of the information necessary to evaluate cirrhosis or PH may be obtained by use of SP (see Images 7-9, 19).
Splenic pulp measurement provides an accurate reflection of the portal venous pressure, both in hepatic and prehepatic causes of PH. An intravenous injection of radiographic contrast medium into the splenic pulp outlines the portosystemic collaterals (ie, splenic and portal veins); it demonstrates well the intrahepatic portal radicles and enables an assessment of the rapidity of washout of the vessels in a hepatopetal direction. With reversal of flow within the portal vein resulting from severe PH, the splenic and portal veins are not visualized; the contrast outflow tract is via gastroesophageal collaterals. With splenic thrombosis/PVT, not only is the site of the obstruction evident but also portal-portal collateral veins, which regularly develop, are demonstrated as well.
Contraindications to SP include coagulation defects, platelet count less than 50,000, ascites, an uncooperative patient, and splenic pathology (splenic mass lesions). To perform SP, the skin over the left side of the thorax and abdomen overlaying the spleen is prepared with an antiseptic after the spleen is localized. Historically, the spleen was localized by use of fluoroscopy and palpation and percussion, but currently, localization is more commonly performed by use of US. A skin bleb is raised near the lower pole of the spleen in mid respiration, and the soft tissues are infiltrated with local anesthetic down to the peritoneum.
The entry site of the needle usually is the 9th or 10th intercostal space in the midaxillary line. The needle is directed in a cephalic direction toward the splenic hilum. A nick is made at the site of the needle puncture, and a track is made into the soft tissues with a 12-gauge needle, which facilitates the placement of the SP needle. The SP needle is a 14-gauge, 6-inch-long polyethylene sheathed needle. With patient breath hold in mid respiration, the sheathed needle is introduced toward the direction of the splenic hilum, and the stilette is withdrawn, leaving the sheath in place. The patient is instructed to take shallow respirations during the examination.
If the spleen has been entered accurately, blood flows back via the sheath. Then, the sheath is connected to a spinal pressure manometer filled with saline, and the splenic pulp pressure is measured. The sheath is taped to the patient's side, and a test injection is made using 5 mL of contrast medium. When a sheath is placed correctly, contrast medium spreads into the splenic parenchyma, providing an uneven splenogram that rapidly drains into the splenic vein. With a subcapsular placement, the contrast agent is viewed as a homogeneous collection that spreads slightly as the agent is injected.
With subcapsular injection, the needle is withdrawn completely, and a fresh attempt is made to place the needle correctly. With the sheath correctly placed, a pump injection is performed using 40 mL of 75% contrast at a rate of 10 mL/s. The injection is usually painless; discomfort and shoulder pain may occur with subcapsular injection. After the procedure is complete, the needle tract is plugged with thromboembolic material, such as Gelfoam or Ivalon.
Complications of the procedure include hemorrhage, contrast agent extravasation, puncture of other organs, splenic hematomas, and capsular rupture. Capsular rupture may require emergency laparotomy and splenectomy. In addition to splenectomy, splenic artery embolization may be used to stop bleeding.
Indications for SP include the following:
In PH, the following findings may be noted with SP:
Carbon dioxide SP
Carbon dioxide has been employed as a contrast agent for SP; a 22- or 25-gauge needle is used (thereby exploiting the low viscosity of the gas). This technique is less traumatic than others, and it has been effective in visualizing the portal venous system. The hope is that the difference in the collateral venous filling patterns is significant, because carbon dioxide fills veins in the nondependent sites, whereas contrast material flows along the dependent sites. Contrast-enhanced SP has been infrequently used in the United States.
Transhepatic portography
With percutaneous transhepatic portography (PTP), the same technique is used as with percutaneous cholangiography. The portal vein is punctured directly and is replaced by a guidewire/catheter, and contrast material is injected into the splenic vein and/or the superior mesenteric vein, depending on the clinical setting. PTP is an easy and quick procedure. As a result of the straight course of the catheter through the liver substance, catheterization of several tributaries is possible, but the distance between the cutaneous entrance of the catheter and the liver makes catheter manipulation difficult. The few reported complications mostly result from nontarget organ puncture (gallbladder, pleura) and intra-abdominal hemorrhage. With US guidance, the incidence of nontarget organ puncture is expected to decrease.
Transcatheter obliteration of esophageal varices with various embolic agents is possible via transhepatic catheter placement, but this is not a procedure to be followed by suitable surgical procedure. Transcatheter obliteration should not be used as an elective procedure; even in an acute setting, hemostasis is best attempted with other methods.
Transumbilical catheterization
Transumbilical catheterization requires a surgical procedure in which a transverse incision is made 3 – 5 cm above the umbilicus, and the umbilical vein is catheterized. Complications are few, but the procedure is difficult and time consuming. With so many noninvasive procedures currently available, the procedure is seldom indicated.
Transjugular catheterization
The transjugular approach to the portal vein first was described by Rösch et al in 1969. Since then, the technique has improved; currently, it is used more for therapeutic applications, such as the establishment of a TIPS.
Wedged hepatic venography
When a catheter is placed in a small hepatic vein via the inferior vena cava (IVC) or jugular vein, pressures may be measured either with a transducer or with a saline manometer. In patients with sinusoidal or postsinusoidal portal venous obstruction, as occurs in cases of cirrhosis, pressure measured in this way accurately reflects the total portal pressure. The wedged hepatic venous pressure is the same as the splenic pulp pressure measured by use of SP. Normal portal pressure levels measured in this way are 40 – 150 mm of saline. Pressures above 150 mm of saline indicate PH.
Two components contribute to PH. The first is intrahepatic resistance to portal venous flow, and the second is transmitted pressure from the IVC. In addition, IVC pressures are measured, and the corrected sinusoidal pressure is derived by subtracting the IVC pressure from the wedged hepatic pressure. The corrected sinusoidal pressure is useful clinically because it reflects the true status of the liver disease responsible for development of portosystemic shunts and variceal bleeding. Corrected sinusoidal pressure readings of as high as 100 mm of saline are considered normal.
Although corrected sinusoidal pressure makes the most significant contribution to variceal bleeding in patients with cirrhosis, the total pressure usually contributes to bleeding from esophageal varices. Thus, a patient with both cirrhosis and congestive heart failure may bleed from esophageal varices because of a temporary increase in central venous pressure; conversely, esophageal bleeding may cease when the central venous pressure decreases. Typically, patients with cirrhosis do not bleed from varices with portal venous pressures of 200 – 250 mm of saline. Patients with cirrhosis who bleed at pressures lower than 200 mm of saline almost always bleed from an alternative source.
Arterial portography
Arterial portography (AP) is currently the preferred method of evaluating the portal venous system because it is less invasive and has a lower complication rate. AP involves the indirect opacification of the portal venous system with the injection of contrast material into the celiac axis (delayed images outline the splenic, gastric, and portal veins) or into the superior mesenteric artery (outlining superior mesenteric and portal veins).
The 3 major indications for AP include the following:
Preparation and contraindications for AP are identical to those of standard conventional angiography. Selective catheters are used to cannulate the appropriate artery. In a superior mesenteric artery injection, the tip of the catheter is placed such that it opacifies all the branches with contrast medium. Before delivery of the contrast agent, administration of a vasodilator (tolazoline, papaverine, nitroglycerin) improves opacification of the portal vein. Manual or digital subtraction imaging improves resolution. If digital subtraction is used, administration of an anticholinergic drug before filming reduces bowel movement. Left gastric artery injection consistently demonstrates esophageal varices.
AP findings in PH include the following:
Splenoportography (SP) is a fairly accurate method of outlining the portal venous system and the portosystemic communications for patients with portal hypertension; for this, SP remains the criterion standard. Diagnostic modalities such as US, CT, and MRI have reduced the diagnostic importance of arteriography. The major role of angiography is in mapping the vascular anatomy before surgery and in guiding the transcatheter treatment of liver tumors. In celiac-axis and superior angiography, the venous phase provides sufficient detail to render direct portography unnecessary in most patients.
Gastrointestinal bleeding from esophageal varices remains the most life-threatening condition resulting from portal hypertension. This is usually secondary to cirrhosis. The 30-day mortality rate approaches 30%, and rebleeding and death are common within 1 year. Surgery has failed to reduce the mortality rate. Endoscopic sclerotherapy has emerged as the most popular treatment. The goal is to arrest active bleeding by producing intravariceal thrombosis or extraluminal fibrosis around the varices. Various sclerosants are used.29
Percutaneous transhepatic coronary vein embolization is infrequently performed because of the high incidence of recurrent bleeding secondary to the development of new collaterals. Instead, most patients with variceal bleeding that is unresponsive to endoscopic sclerotherapy or banding undergo a TIPS procedure, during which the coronary vein may be embolized if gastroesophageal varices continue to fill after placement of a TIPS.
The technique used most often is percutaneous transhepatic portography (PTP) (see Angiography). The agent that provides the most effective long-term occlusion is large strips of compressed Gelfoam soaked in 3% sodium tetradecyl sulfate. The procedure may provide an alternative in arresting acute variceal bleeding that is not responsive to vasopressin or other medical methods. With transhepatic embolization, the patient's general status may improve enough to allow surgeons to perform elective shunt surgery.
Bosch J, Pizcueta P, Feu F, et al. Pathophysiology of portal hypertension. Gastroenterol Clin North Am. Mar 1992;21(1):1-14. [Medline].
Mergo PJ, Ros PR. Imaging of diffuse liver disease. Radiol Clin North Am. Mar 1998;36(2):365-75. [Medline].
Murakami T, Mochizuki K, Nakamura H. Imaging evaluation of the cirrhotic liver. Semin Liver Dis. May 2001;21(2):213-24. [Medline].
Parvey HR, Raval B, Sandler CM. Portal vein thrombosis: imaging findings. AJR Am J Roentgenol. Jan 1994;162(1):77-81. [Medline].
Henseler KP, Pozniak MA, Lee FT Jr, Winter TC 3rd. Three-dimensional CT angiography of spontaneous portosystemic shunts. Radiographics. May-Jun 2001;21(3):691-704. [Medline].
Mitchell DG, Nazarian LN. Hepatic vascular diseases: CT and MRI. Semin Ultrasound CT MR. Feb 1995;16(1):49-68. [Medline].
Winter TC 3rd, Nghiem HV, Schmiedl UP, Freeny PC. CT angiography of the visceral vessels. Semin Ultrasound CT MR. Aug 1996;17(4):339-51. [Medline].
Zhao LQ, He W, Chen G. Characteristics of paraesophageal varices: A study with 64-row multidetector computed tomograghy portal venography. World J Gastroenterol. Sep 14 2008;14(34):5331-5. [Medline].
Ulu EM, Kirbas I, Emiroglu FK, Cakir B, Harman A, Bakar C, et al. Multidetector CT findings of splenic artery aneurysm in children with chronic liver disease. Pediatr Radiol. Oct 2008;38(10):1095-8. [Medline].
Kim M, Mitchell DG, Ito K. Portosystemic collaterals of the upper abdomen: review of anatomy and demonstration on MR imaging. Abdom Imaging. Sep-Oct 2000;25(5):462-70. [Medline].
Nghiem HV, Winter TC, Schmiedl UP, Freeny PC. MR angiography of the portal venous system. Semin Ultrasound CT MR. Aug 1996;17(4):360-73. [Medline].
Spritzer CE. Vascular diseases and MR angiography of the liver. Magn Reson Imaging Clin N Am. May 1997;5(2):377-96. [Medline].
Bolognesi M, Sacerdoti D, Merkel C, et al. Noninvasive grading of the severity of portal hypertension in cirrhotic patients by echo-color-Doppler. Ultrasound Med Biol. Jul 2001;27(7):901-7. [Medline].
Colli A, Cocciolo M, Riva C, et al. Abnormalities of Doppler waveform of the hepatic veins in patients with chronic liver disease: correlation with histologic findings. AJR Am J Roentgenol. Apr 1994;162(4):833-7. [Medline].
Grant EG. Doppler imaging of the liver. Ultrasound Q. 1992;10:117-54.
Kedar RP, Merchant SA, Malde HH, Patel VH. Multiple reflective channels in the spleen: a sonographic sign of portal hypertension. Abdom Imaging. Sep-Oct 1994;19(5):453-8. [Medline].
Koslin DB, Mulligan SA, Berland LL. Duplex assessment of the portal venous system. Semin Ultrasound CT MR. Feb 1992;13(1):22-33. [Medline].
Millener P, Grant EG, Rose S, et al. Color Doppler imaging findings in patients with Budd-Chiari syndrome: correlation with venographic findings. AJR Am J Roentgenol. Aug 1993;161(2):307-12. [Medline].
Nelson RC, Sherbourne GM, Spencer HB, Chezmar JL. Splenic venous flow exceeding portal venous flow at Doppler sonography: relationship to portosystemic varices. AJR Am J Roentgenol. Sep 1993;161(3):563-7. [Medline].
Piscaglia F, Donati G, Serra C, et al. Value of splanchnic Doppler ultrasound in the diagnosis of portal hypertension. Ultrasound Med Biol. Jul 2001;27(7):893-9. [Medline].
Ralls PW. Color Doppler sonography of the hepatic artery and portal venous system. AJR Am J Roentgenol. Sep 1990;155(3):517-25. [Medline].
Ralls PW, Mack LA. Spectral and color Doppler sonography. Semin Ultrasound CT MR. Oct 1992;13(5):355-66. [Medline].
Restrepo-Schafer I, Wollenberg B, Riera-Knorrenschild J, Gorg C. Partially reversed intrasplenic venous blood flow detected by color Doppler sonography in two patients with hematologic diseases and splenomegaly. J Clin Ultrasound. Jun 2001;29(5):294-7. [Medline].
Tessler FN, Mitchell E, Siskin T. Sonography of the transjugular intrahepatic portosystemic shunt. Ultrasound Q. 1998;14(3):165-70.
Zwiebel WJ. Sonographic diagnosis of hepatic vascular disorders. Semin Ultrasound CT MR. Feb 1995;16(1):34-48. [Medline].
McKiernan PJ, Sharif K, Gupte GL. The role of endoscopic ultrasound for evaluating portal hypertension in children being assessed for intestinal transplantation. Transplantation. Nov 27 2008;86(10):1470-3. [Medline].
Pan JJ, Chen C, Geller B, Firpi R, Machicao VI, Caridi JG, et al. Is sonographic surveillance of polytetrafluoroethylene-covered transjugular intrahepatic portosystemic shunts (TIPS) necessary? A single centre experience comparing both types of stents. Clin Radiol. Oct 2008;63(10):1142-8. [Medline].
Morgan J, Sadler MA, Siegel S. US, CT, and MR imaging of hepatic masses in Alström syndrome: a case report. Clin Imaging. Sep-Oct 2008;32(5):393-5. [Medline].
Superina RA, Alonso EM. Medical and Surgical Management of Portal Hypertension in Children. Curr Treat Options Gastroenterol. Sep 2006;9(5):432-443. [Medline].
Habib SF, Muhammad R, Koulaouzidis A, Gasem J. Pulmonary embolism after sclerotherapy treatment of bleeding varices. Ann Hepatol. Jan-Mar 2008;7(1):91-3. [Medline].
Goyal AK, Pokharna DS, Sharma SK. Ultrasonic measurements of portal vasculature in diagnosis of portal hypertension. A controversial subject reviewed. J Ultrasound Med. Jan 1990;9(1):45-8. [Medline].
Schraufnagel DE, Kay JM. Structural and pathologic changes in the lung vasculature in chronic liver disease. Clin Chest Med. Mar 1996;17(1):1-15. [Medline].
portal hypertension, PH, portal HTN, portal venous pressure, hepatic venous pressure, portal venous flow, hematemesis, bleeding esophageal varices, portosystemic collateral vessels, portosystemic collaterals, cirrhosis, splenoportography
Ali Nawaz Khan, MBBS, FRCS, FRCP, FRCR, LRCP, Chairman of Medical Imaging, Professor of Radiology, NGHA, King Fahad National Guard Hospital, King Abdulaziz Medical City, Riyadh, Saudi Arabia
Ali Nawaz Khan, MBBS, FRCS, FRCP, FRCR, LRCP is a member of the following medical societies: American Institute of Ultrasound in Medicine, Radiological Society of North America, Royal College of Physicians, Royal College of Physicians and Surgeons of the United States, Royal College of Radiologists, and Royal College of Surgeons of England
Disclosure: Nothing to disclose.
Sumaira MacDonald, MBChB, PhD, MRCP, FRCR, Lecturer, Sheffield University Medical School; Endovascular Fellow, Sheffield Vascular Institute
Sumaira MacDonald, MBChB, PhD, MRCP, FRCR is a member of the following medical societies: British Medical Association, Royal College of Physicians, and Royal College of Radiologists
Disclosure: Nothing to disclose.
Murad Ali, MBBS, PhD, DTCD, Consulting Radiologist, Department of Radiology, Postgraduate Medical Institute, Lady Reading Hospital, Pakistan
Disclosure: Nothing to disclose.
David Sherlock, MBBS, FRCS, Consulting Staff, Department of Surgery, North Manchester General Hospital, Christie Hospital
Disclosure: Nothing to disclose.
Eric P Weinberg, MD, Associate Professor, Department of Radiology, University of Rochester Medical Center, Strong Memorial Hospital
Eric P Weinberg, MD is a member of the following medical societies: American College of Radiology, American Roentgen Ray Society, and Radiological Society of North America
Disclosure: Nothing to disclose.
Bernard D Coombs, MB, ChB, PhD, Consulting Staff, Department of Specialist Rehabilitation Services, Hutt Valley District Health Board, New Zealand
Disclosure: Nothing to disclose.
George Hartnell, MB, Professor of Radiology, Tufts University School of Medicine, Director of Cardiovascular and Interventional Radiology, Department of Radiology, Baystate Medical Center
George Hartnell, MB is a member of the following medical societies: American College of Cardiology, American College of Radiology, American Heart Association, Association of University Radiologists, British Institute of Radiology, British Medical Association, Massachusetts Medical Society, Radiological Society of North America, Royal College of Physicians, Royal College of Radiologists, and Society of Cardiovascular and Interventional Radiology
Disclosure: Nothing to disclose.
Robert M Krasny, MD, Consulting Staff, Department of Radiology, The Angeles Clinic and Research Institute
Robert M Krasny, MD is a member of the following medical societies: American Roentgen Ray Society and Radiological Society of North America
Disclosure: Nothing to disclose.
Kyung J Cho, MD, FACR, William Martel Professor of Radiology, Interventional Radiology Fellowship Director, University of Michigan Health System
Kyung J Cho, MD, FACR is a member of the following medical societies: American College of Radiology, American Heart Association, American Medical Association, American Roentgen Ray Society, Association of University Radiologists, and Radiological Society of North America
Disclosure: Nothing to disclose.
Guidelines and clinical studies:
The role of transjugular intrahepatic portosystemic shunt in the management of portal hypertension. American Association for the Study of Liver Diseases - Private Nonprofit Research Organization. 2005 Feb. 15 pages. NGC:004222
Peginterferon Alpha-2a Maintenance Therapy for Portal Hypertension in Patients With Hepatitis C
Use of Sanvar® With Endoscopic Treatment for the Control of Acute Variceal Bleeding
Efficacy and Safety of Simvastatin in the Treatment of Portal Hypertension
Efficacy and Safety of Cobiprostone in Patients With Portal Hypertension
Evaluation of Probiotics in the Treatment of Portal Hypertension
© 1994-
by Medscape.
All Rights Reserved
(http://www.medscape.com/public/copyright)