The ability to non-surgically measure pressure within the portal venous system by wedging a catheter in the small hepatic veins was developed in the 1950s and has been shown to accurately reflect portal pressure in most patients with liver disease from a variety of causes.1, 2 However, wedged hepatic vein pressure (WHVP) measurement has remained a research tool and has not been applied to the care of patients with liver disease. A number of observations over the past 5 years are leading to a greater appreciation of how WHVP measurements can be used in the management of the patient with liver disease and the hepatology community may begin to use this technique in the same manner as they use liver biopsy.
WHVP measurements have been used to assess the efficacy of pharmacologic therapy for both primary and secondary prevention of variceal bleeding. A fall of the hepatic venous pressure gradient (WHVP— free hepatic vein pressure [HVPG]) by 20% or to less that 12 mm Hg following pharmacologic therapy is associated with much greater reduction in the risk of bleeding and improved survival as compared to those who lacked this hemodynamic response.3–6 Hemodynamic responders appear less likely to develop other complications of cirrhosis as well.7 Also, a high HVPG (>20 mm Hg) at the time of presentation of variceal bleeding is associated with a poor outcome as compared to patients with lower pressures.8 Despite the consistency of the reported series on the importance of measuring the HVPG in determining risk of variceal bleeding or rebleeding, numerous problems with the published studies exist, especially the absence of prospective data demonstrating that using the hemodynamic response to alter therapy is beneficial to the patients.3, 9 Thus, HVPG measurements remain little used in the care of patients with liver disease.
Two recent reports, one in this issue of HEPATOLOGY and another published recently in the New England Journal of Medicine may alter our thinking on the utility of HVPG measurements in the management of patients with liver disease.10, 11 Groszmann, Garcia-Tsao et al.11 reported on the efficacy of beta-blockers in preventing the development of varices in cirrhotics with portal hypertension but who lacked varices at the index endoscopy. No benefit of beta blocker therapy could be shown but the patients whose initial HVPG was greater or equal to 10 mm Hg, or whose HVPG increased by more than 10% were more likely to develop or to bleed from varices than were those whose pressure was less or failed to increase. Also a fall in HVPG of more than 10% was associated with a better outcome.11 As all patients had cirrhosis, it was the development of worsening portal hypertension and not the presence of cirrhosis per se that was predictive of outcome. In the study by Blasco and colleagues10 serial liver biopsies and HVPG measurements were obtained on patients with hepatitis C who had undergone liver transplantation allowing for a comparison of the accuracy of each approach in determining the rate of progression of hepatitis C induced liver disease. There was a good but not perfect relationship between the HV PG and severity of fibrosis. Several patients with stage 2 fibrosis or less on liver biopsy had portal hypertension suggesting that the liver biopsy had underestimated the severity of the fibrosis. Previous studies have shown that a single liver biopsy commonly underestimates the severity of fibrosis, especially if the biopsy is small, so the findings in the current study are not unexpected.12, 13 What was unexpected was the almost perfect correlation between development of an elevated HVPG and risk of future clinical decompensation.
HVPG measurements can be used to: (1) predict risk of rebleeding following an acute variceal bleed; (2) determine the efficacy of pharmacologic therapy in primary and secondary prophylaxis; (3) predict risk of the development of complications of portal hypertension; (4) determine which patients following liver transplantation for hepatitis C are in need of antiviral treatment. Given all these benefits of HVPG measurement, why is its use limited to academic centers? One argument is that the technique requires care and training in its performance in order to obtain reliable data.9, 14 What procedure in medicine doesn't require training and care in its performance? Liver biopsies are frequently too small or are misinterpreted by pathologists who have limited experience and yet that does not stop physicians from obtaining liver biopsies in the community. Training interventional radiologists how to perform HVPG correctly and with the needed precision would not be difficult as the criteria for a good HVPG are quite straightforward.14 A second argument is that measurement of HVPG is invasive. Liver biopsies are also invasive and have a higher risk of serious complications and death compared to measurement of the HVPG. A third concern is cost and cost-effectiveness.15, 16 In the study by Blasco et al.10 measurement of pressure was the same cost as a liver biopsy. The medicare reimbursement for a liver biopsy and HVPG measurement (technical plus professional fees) at my institution is $508 and $636 respectively. If one adds the cost for processing and interpreting the liver biopsy ($183) then the costs are virtually the same. Lastly, everyone is waiting for a non-invasive test for the measurement of portal pressure. The two most recent studies10, 11 show that pressure differences of as little as 1-2 mm Hg may have prognostic significance and it is very unlikely that a non-invasive alternative can be developed that will provide the same level of accuracy as the HVPG.
This writer feels that the hepatology community, rather than resisting HVPG measurement as a useful tool, should instead embrace the technique and design additional prospective studies that determine how changes in HVPG can be used to modify therapy in patients with liver disease. For example, if we can define which patients with hepatitis C post-liver transplant are in need of anti-viral therapy, then we can limit this expensive and toxic therapy to those who need it and not treat patients who have a good prognosis without therapy. Given the cost of antiviral therapy this approach may well be cost effective. We can also use the HVPG to judge the response to anti-fibrotic or anti-viral therapy in patients with hepatic fibrosis. Also, if pressure measurements can be shown prospectively to significantly improve the outcome with pharmacologic therapy in the prevention of variceal bleeding/rebleeding then physicians in the community will be motivated to use the HVPG in the clinical management of patients with cirrhosis. Lastly, the finding of a progressive rise in the HVPG over a period of 2-5 years in a patient with cirrhosis may be indicative of worsening liver disease and could be used as a criterion for referral to a liver transplant center.
The HVPG will not replace liver biopsy, as the latter defines the cause of liver disease; however, because HVPG samples a larger area of the liver the pressure measurement is more reflective of what is happening to the entire liver. HVPG also is a dynamic assessment of the progression of the liver disease whereas liver biopsy, once cirrhosis has developed, is static. Clearly, the time for using the WHVP more broadly has arrived and its utility in the management of patients with liver disease will only be limited by our lack of interest in using this old technique in new ways.