Digestive Diseases and Sciences

, Volume 54, Issue 6, pp 1178–1183

Hepatic Venous Pressure Gradient: Worth Another Look?


    • Department of MedicineBaylor College of Medicine

DOI: 10.1007/s10620-008-0491-8

Cite this article as:
Parikh, S. Dig Dis Sci (2009) 54: 1178. doi:10.1007/s10620-008-0491-8


Portal hypertension is one of the most important complications of chronic liver disease and accounts for significant morbidity and mortality. Measurement of the hepatic venous pressure gradient (HVPG) is a simple, invasive, and reproducible method of assessing portal venous pressure. Measurement of HVPG provides the clinician an estimate of the degree of intrahepatic portal flow resistance, guides therapy for variceal bleeding (primary and secondary prophylaxis), assesses feasibility of resection in patients with hepatocellular cancer, and predicts response to therapy of patients with chronic hepatitis C. Achieving hemodynamic targets of reducing the HVPG to <10 mmHg or a 20% reduction from baseline virtually eliminates complications related to portal hypertension from chronic liver disease. This review explores the role of HVPG measurement in the contemporary treatment of patients with cirrhosis and portal hypertension.


CirrhosisPortal hypertensionVariceal bleedingLiver function tests


Portal hypertension is characterized by an abnormal increase in portal venous pressure. Hepatic venous pressure gradient (HVPG) is defined as the difference in pressure between the portal vein and the hepatic vein (normally about 1–5 mmHg). When the HVPG increases above 10–12 mmHg, a number of life-threatening complications can occur. These include bleeding from esophageal varices, portal gastropathy, ascites, arterial hypoxemia, hepatorenal syndrome, and hepatic encephalopathy. The aim of this review was to explore the role of HVPG measurement in the contemporary treatment of patients with cirrhosis and portal hypertension (see Table 1).
Table 1

Utility of HVPG measurements in clinical practice

Uses of HVPG measurement

1. Diagnose portal hypertension (except due to pre-sinusoidal causes)

2. Monitor drug therapy in patients with esophageal varices

3. Predict prognosis of patients with alcoholic hepatitis

4. Assess disease progression and effects of therapy in patients with chronic hepatitis C

5. Assess feasibility of resection of hepatocellular cancer

6. Prognosticate patients with portal hypertension and cirrhosis


Direct measurement of portal venous pressure involves catheterization of the portal vein, which is a technically difficult procedure. Indirect evaluation of portal venous pressure by catheterization of the hepatic vein was introduced more than 50 years ago [1]. This procedure is done after a 6 h fast under conscious sedation with continuous electrocardiography and monitoring of vital signs. A balloon-tipped catheter is introduced from either the internal jugular vein or the femoral vein into the hepatic vein under fluoroscopic guidance. While the catheter traverses the inferior vena cava (IVC) into the hepatic vein, simultaneous measurements are made to record the IVC and free hepatic venous pressures. Normally, there is no significant difference between the two. The balloon is then inflated in the hepatic vein and the wedged hepatic venous pressure is recorded. These measurements are recorded continuously on a multi-channel recorder and are duplicated before the catheter is withdrawn.

Interpretation of Results

Portal hypertension can be divided into three major categories depending on its etiology: pre-sinusoidal, sinusoidal, and post-sinusoidal. In healthy individuals the wedged hepatic venous pressure is equal to sinusoidal pressure but is slightly less than portal venous pressure to allow for blood flow. However, in portal hypertension due to sinusoidal causes, there is damage to the hepatic sinusoids, and a continuous column of blood extends from the hepatic vein to the portal vein. This static column of blood equates the wedged hepatic venous pressure to the portal vein pressure [2]. The difference between the wedged and the free hepatic venous pressures represents the true HVPG. A pictorial depiction of this technique is illustrated in Fig. 1.
Fig. 1

A pictorial depiction of the measurement of HVPG. (a) In healthy individuals the wedged hepatic venous pressure is slightly lower than the portal venous pressure to allow for blood flow through the sinusoidal veins. (b) In patients with portal hypertension due to sinusoidal causes, the sinusoidal veins are damaged, and, hence, the wedged hepatic venous pressure equates to the portal pressure. IVC inferior vena cava, FHVP free hepatic venous pressure, WHVP wedged hepatic venous pressure, PVP portal venous pressure. Hepatic venous pressure gradient = WHVP − FHVP

Until a few years ago, the primary utility of HVPG was to diagnose different types of portal hypertension. Patients with pre-sinusoidal causes of portal hypertension have a normal wedged hepatic venous pressure, and portal venous pressure is falsely normal. This condition therefore requires direct cannulation of the portal vein to diagnose portal hypertension. However, in most cases of sinusoidal and post-sinusoidal portal hypertension, the wedged hepatic pressure is elevated and correlates well with the portal venous pressure (see Table 2). This is true for most chronic liver diseases, including cirrhosis from viral (hepatitis B and hepatitis C) and alcoholic causes [35]. An obvious advantage to this technique is that a transvenous biopsy of the liver can be performed simultaneously.
Table 2

Pressure measurements in different types of portal hypertension













PVP portal venous pressure, FHVP free hepatic venous pressure, WHVP wedged hepatic venous pressure, HVPG hepatic venous pressure gradient

Hepatic Venous Pressure Gradient to Monitor Drug Therapy

With significant advances in the technique of measurement of HVPG and its standardization, HVPG measurement is employed in major tertiary care centers to determine the effectiveness of pharmacologic therapy for portal hypertension. A number of studies have shown that reduction of the absolute HVPG to <12 mmHg or by 20% from its baseline value offers a significant benefit in reducing complications related to portal hypertension, chiefly variceal bleeding [615].

Primary Prophylaxis

Primary prophylaxis for variceal bleeding is defined as the initiation of pharmacologic therapy or performance of endoscopic variceal ligation in patients with esophageal varices due to portal hypertension who have never had an episode of bleeding. In routine clinical practice, treatment with a non-selective beta-blocker, such as propranolol, nadolol, or timolol, is initiated for the prophylaxis of a first episode of bleeding, and the dose is titrated up to decrease the resting heart rate by 25% from baseline value. Nitrates used in conjunction with beta-blockers do not reduce the rate of first hemorrhage and may even increase side effects [16]. Endoscopic variceal band ligation may be an alternative to non-selective beta-blockers if patients cannot tolerate them or in patients with large varices and Child–Pugh class C cirrhosis.

However, some patients who have seemingly responded to achieve the heart rate targets after the initiation of beta-blockers have a higher incidence of first bleeding. This is because lowering the heart rate does not necessarily correlate with a decrease in HVPG [17]. Measurement of HPVG can identify this subset of patients and allow for faster up-titration of medications to achieve the hemodynamic targets of 20% reduction in HVPG from baseline or absolute reduction to <12 mmHg.

However, there is a paucity of controlled clinical trials examining this question, and it is also not certain if it would be cost effective for baseline and repeat HVPG measurements to be performed in patients who have never experienced variceal bleeding.

Secondary Prophylaxis

Secondary prophylaxis for portal hypertension is employed for patients who have had at least one previous episode of variceal bleeding. These patients have a very high risk of rebleeding and death.

The pharmacologic therapy of choice for these patients is a combination of non-selective beta-blockers and nitrates. However, there is an increased incidence of side effects from the nitrates, so that most patients tolerate only beta-blockers. For patients who are naïve to treatment by beta-blockers or endoscopic therapy, either ligation or medical treatment is appropriate [18]. However, for patients who have already taken beta-blockers or have had prior ligation, the combination of pharmacologic and ligation techniques is superior to either alone [19].

The timing of HVPG measurement is critical in its ability to predict response to therapy. The lowest rate of rebleeding is observed in patients who are HVPG responders (defined as an absolute decrease to <12 mmHg or a reduction by 20% from baseline value). In patients with primary prophylaxis, who are otherwise stable, the timing of a repeat measurement does not matter as much, since it can safely be performed 8–12 weeks after initiation of therapy. However, patients with an episode of variceal bleeding have a 40% chance of rebleed in the first 3 months. The farther a repeat HVPG measurement is obtained from the index bleed, the more likely is a non-responder to rebleed and the less useful a repeat HVPG measurement will be [20]. It is difficult to interpret data in the literature, given the different times at which repeat measurements were done following the index bleed. This time varied from 19 days to 160 days, with a median of 90 days. Studies are ongoing to standardize repeat HVPG measurements within 14 days of the index bleed to be able to identify patients who are non-responders and who would benefit from an increased dose of beta-blockers, and/or with the addition of nitrates, to achieve specific hemodynamic targets. Alternatively, if those patients have not undergone variceal ligation, non-response to medical therapy by HVPG measurement would be an indication for ligation to reduce rebleeding risk.

Hepatic Venous Pressure Gradient in Patients with Alcoholic Liver Disease

Only one study has been published that assessed the relationship between HVPG and outcome of patients with alcoholic hepatitis. It concluded that patients with a high HVPG (>22 mmHg) and a high model for end-stage liver disease (MELD) score (>25) were independently associated with an increased risk of in-hospital death [21]. A liver biopsy and the Maddrey discriminant function, as assessed by the following formula: {4.6 × [prothrombin time-control time (s)]} + serum bilirubin (mg/dl) are the only validated tools available to clinicians to assess the severity of alcoholic hepatitis. MELD score has been shown to be a better prognosticator than the discriminant function in a few studies. More data are needed to clarify the role of HVPG in predicting mortality and assessing the need for therapy with corticosteroids and pentoxifylline in the treatment of patients with severe alcoholic hepatitis.

Hepatic Venous Pressure Gradient in Patients with Chronic Hepatitis C

A sustained virologic response, decrease in serum alanine aminotransferase levels, and improvement in fibrosis on liver biopsy, have all been used as surrogate endpoints in therapy for chronic hepatitis C. Since complications of portal hypertension related to cirrhosis from chronic hepatitis C account for the vast majority of deaths, improvements in these markers may not be the best end-points for therapy. HVPG, by its virtue of being able to demonstrate intrahepatic portal flow resistance, may allow determination of the degree of fibrosis and improvement in histology over and above that provided by liver histology alone. A number of studies have shown that serial HVPG measurements obtained with transjugular liver biopsy before and after therapy with peg-interferon and ribavarin may be the best predictors of response to therapy in patients treated for chronic hepatitis C. These studies have also demonstrated that treatment of chronic hepatitis C leads to improvements not only in fibrosis and virologic loads but also in portal hypertension [2224].

Measurement of HVPG is a valuable tool for follow-up in patients with recurrence of hepatitis C after orthotopic liver transplantation. Studies have shown that HVPG measurement correlates well with fibrosis and disease progression in patients infected with hepatitis C after transplantation [25, 26]. An HVPG of 6 mmHg or greater was accurate in identifying patients at risk for progression of disease 1 year after transplantation, as compared with liver biopsy alone [27].

Hepatic Venous Pressure Gradient in Patients with Hepatocellular Cancer

Hepatic resection is an important component of therapy for treatment of primary liver cancers, particularly hepatocellular carcinoma (HCC). Because of the high incidence of cirrhosis and portal hypertension in patients with HCC, such patients are at high risk of hepatic decompensation after resection of the tumor [28]. However, a number of subsequent reports have confirmed that patients with Child’s class A cirrhosis can safely undergo resection of HCC and that their long-term outcomes are similar to those of patients without portal hypertension [29]. The utility of HVPG in being able to predict the degree of portal hypertension and its impact on resection of HCC is currently being tested. HCC can falsely elevate HVPG measurement by arterio-venous shunting within the liver. HCC also leads to inaccuracies in the estimation of portal pressure, because it may cause both macroscopic and microscopic invasion of the portal vein [30]. It remains to be seen if clinical studies will be able to overcome these drawbacks so that patients with portal hypertension who can safely undergo resection of HCC will be able to be effectively identified.

Hepatic Venous Pressure Gradient Measurement Predicts Morbidity and Mortality

Traditionally, the Child–Pugh score has been used to predict death in patients with cirrhosis. Recently, however, the MELD score and the serum sodium model for end-stage liver disease (MELD-Na) score have been used more frequently as prognostic markers. HVPG measurement has not been shown to be superior to either the MELD score or serum sodium concentration for short term mortality. However, it may help identify a subset of patients with low MELD scores who have a higher mortality [31]. Another recently completed study has shown that HVPG, MELD score, and serum albumin concentration independently predict decompensation in patients with stable cirrhosis. It also showed that, for patients with an HVPG value of <10 mmHg, there is a <10% probability that they will develop clinical decompensation over a median follow-up period of 4 years [32].

Limitations of the Measurement of Hepatic Venous Pressure Gradient

Although the technique of HVPG measurement is relatively simple, safe, and reproducible, its biggest limiting factor is its invasive nature. It is important that the technique be performed properly and comply with the recommended standards to ensure that its results are reproducible. Certain minimal criteria for technical adequacy have been previously published [2, 33], and, for the benefit of both an individual patient and the general hepatology research community, it is important that they are routinely met for all patients undergoing HVPG measurement. Unfortunately, at this time, there are only a handful of centers of excellence that routinely perform this technique in the vast majority of patients with portal hypertension. The importance of paying meticulous attention to detail and achieving results that are consistent and comparable from center to center cannot be emphasized enough [2, 33].

Severe thrombocytopenia and coagulation factor deficiencies from liver dysfunction require transfusion of platelets and fresh frozen plasma, respectively, prior to the procedure. An allergy to radiocontrast medium is an absolute contraindication to HVPG measurement. Major complications reported during the procedure include cardiac arrhythmias, injury to the access vasculature (including the femoral, jugular, and antecubital veins), leakage of contrast agent, and arteriovenous fistulae formation.

Economic Analyses of Hepatic Venous Pressure Gradient Measurement

HVPG is generally measured by interventional radiologists or hepatologists trained to do this procedure while the patient is under conscious sedation. The average time it takes for the procedure to be completed is 60 min. If a transjugular liver biopsy or transjugular intrahepatic portosystemic shunt is performed simultaneously, it increases the time to about 75–90 min. The average Medicare reimbursement for measurement of HVPG is approximately $636 (technical and professional fees) [34], whereas the total costs (direct and indirect) for performance of the procedure may vary from $1,000 to $4,000 depending on the individual institution. A decision analysis study concluded that HVPG measurement to guide primary prophylaxis is an expensive strategy to reduce variceal bleeding and death, especially in patients with advanced decompensated cirrhosis [35]. However, another study reported that, for secondary prophylaxis of variceal bleeding, combination medical therapy guided by HVPG monitoring, compared with variceal band ligation, is more effective and only marginally expensive [36]. The cost effectiveness of measuring HVPG to guide prognosis and therapy in other complications of portal hypertension has not been reported.

Future Trends in the Use of Hepatic Venous Pressure Gradient to Assess Portal Hypertension

As in the use of intraprocedural nitric oxide inhalation to judge response in patients with pulmonary hypertension, trials are ongoing to determine if patients have a response to intravenous beta-blockers during measurement of HVPG. It is hoped that, in the future, one may consider variceal band ligation versus early nitrate therapy in this subset of patients who do not respond to beta-blocker administered during the procedure.

Newer non-invasive techniques to measure portal hypertension are being developed. Indirect markers of fibrosis, such as the Fibrotest [which combines alpha 2-macroglobulin, haptoglobin, apolipoprotein A1, total bilirubin, and gamma-glutamyl transferase (GGT)] and the Forns index (based on platelet count, GGT, age, and cholesterol levels) represent advances in determination of the degree of portal resistance. Transient elastography, a novel ultrasound-based technology, appears to be the most promising tool for assessment of liver stiffness, a surrogate for portal flow resistance and fibrosis in the liver [37]. Validation studies are underway to determine normal liver stiffness and changes in portal venous pressure relative to medication use and after therapy for chronic hepatitis C. Per-endoscopic measurement of variceal pressure may also act as an alternative to HVPG in measuring portal pressure, since this allows measurements to be taken during the endoscopic procedure itself [38] (see Table 3).
Table 3

Emerging alternatives to HVPG measurement to assess portal hypertension


1. Transient elastography

2. Per-endoscopic measurement of variceal pressure

3. Endoscopic ultrasonography

4. Duplex Doppler ultrasound of the liver


HVPG is an exciting and promising tool for measuring portal venous pressure. It is used to tailor therapy for variceal bleeding and predict the prognosis of patients who have not achieved the hemodynamic targets of 20% reduction of portal pressure from baseline value or an absolute decrease to <12 mmHg. HVPG measurement also aids in the assessment of response to therapy of patients with chronic hepatitis C and determines the feasibility of resection of hepatocellular cancers. A number of trials are underway to determine the role of HVPG measurement in other complications of portal hypertension, such as ascites, hepatopulmonary syndrome, and hepatic encephalopathy. Although this technique was first introduced in the USA over 20 years ago, it is now slowly gaining recognition, with more robust data becoming available from its ability to predict complications from chronic liver disease. It remains to be seen if HVPG measurement will be routinely applied as a “splanchnic sphygmomanometer” in all patients with portal hypertension or if non-invasive means of assessing portal hypertension will supercede this relatively invasive technique.


I would like to thank Ms. Yvette Pinales for her help in preparing this manuscript.

Copyright information

© Springer Science+Business Media, LLC 2008