Gastroenterology
Volume 135, Issue 5 , Pages 1450-1451, November 2008

Portal Hypertension in Primary Biliary Cirrhosis: A Potentially Reversible Harbinger of Demise

  • Marlyn J. Mayo

      Affiliations

    • Corresponding Author InformationAddress requests for reprints to: Marlyn J. Mayo, MD, Associate Professor of Internal Medicine, Division of Digestive and Liver Diseases, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, Texas 75390-9151. fax: 214-648-8274

Division of Digestive and Liver Diseases, University of Texas Southwestern, Dallas, Texas

published online 09 October 2008.

Article Outline

 

See “Portal hypertension and primary biliary cirrhosis: Effect of long-term ursodeoxycholic acid treatment,” by Huet PM, Vincent C, Deslaurier J, et al, on page 1552.

In this issue of Gastroenterology, Huet et al1 report the results of serial direct portal and hepatic vein pressure measurements in a large cohort of patients with primary biliary cirrhosis (PBC) treated with ursodeoxycholic acid (UDCA). The portohepatic gradient (PHG) was determined biannually in 132 PBC subjects over a mean of 8.6 years. In addition, a subset of 30 subjects was initially randomized to UDCA or placebo for 2 years, allowing comparisons of PHG between those 2 groups. The sheer magnitude, duration, and invasiveness of the protocol make it an exceptional study that is unlikely to be repeated.

By using direct access to the portal vein, rather than wedged hepatic vein pressure, Huet et al1 encompass both sinusoidal and presinusoidal components of portal pressure, thought to be important in PBC. Indeed, much is said, but little is documented, about the presinusoidal component of portal hypertension in PBC. Shelia Sherlock and others have reported that varices and variceal bleeding may occur in PBC patients in the absence of cirrhosis,2, 3, 4 but the mechanism is not known. Some of these previous cases may have truly had cirrhosis, which was not appreciated on needle biopsy. It is now well recognized that histologic involvement of PBC is patchy and the degree of sampling error for staging is great.5 Indeed, in one of the frequently cited reports of varices in the absence of cirrhosis,3 16 of 18 patients with varices had “marked fibrosis,” although only 7 patients were labeled as having cirrhosis. A few cases, however, are well documented (eg, through necropsy) to have early stage disease and in these cases it is postulated that granulomas and dense portal infiltrates in the portal tracts lead to compression and/or thrombosis of portal venules, in the same way that schistosomiasis is also thought to cause portal hypertension. Nodular regenerative hyperplasia has also been linked to PBC as a potential cause of presinusoidal hypertension.6

To what degree does presinusoidal pressure contribute to portal hypertension in PBC? That answer is still not known precisely, but the results of this study indicate that it is minimal. In this series, 7 of 57 (12.3%) noncirrhotic patients had elevated PHG, a figure within the range of biopsy sampling error. It was, in fact, far more common in their series to have histologic cirrhosis without portal hypertension. In addition, purely presinusoidal hypertension is associated with favorable outcomes and survival because of relatively preserved liver function. Huet et al,1 however, have demonstrated through Kaplan–Meier analysis that elevated PHG in PBC is a harbinger of shortened survival. Thus, although medical providers caring for the individual with PBC must remember that occasional patients with PBC may have presinusoidal portal hypertension, in the vast majority of PBC patients, development of portal hypertension is an ominous sign. Portal hypertension, therefore, remains a valid clinical end point when studying outcomes of large populations of patients with PBC. Although significantly reduced survival was only detected in their study if the PHG was >12, the trend toward reduced survival in those with a PHG of 6–12 mmHg suggests that, with longer follow-up, this group might also be expected to have reduced survival.

Of all the potential clinical correlates studied, PHG correlated best with histologic stage. Mayo risk score and platelet count also correlated significantly with PHG, although not strongly. Exclusion of subjects with platelets <70,000, history of ascites, bleeding varices, or encephalopathy likely reduced the ability of this particular study to detect a strong correlation between PHG and platelet count or Child–Pugh score. Both the Mayo Risk score and thrombocytopenia have repeatedly been shown to predict the presence of varices in PBC patients with good accuracy.7, 8

Interestingly, UDCA seemed to have a protective effect on PHG. Subjects who were randomized to placebo experienced a mean increase in PHG from 7.0 to 9.7 mmHg over 2 years. After those 2 years, placebo subjects were crossed over to UDCA and PHG eventually returned to baseline. PHG in subjects randomized to UDCA did not change significantly over the 6-year period. The difference in PHG between the 2 groups at each time point, however, was not large and we do not know whether PHG would have continued to rise in the placebo group had it not been crossed over to UDCA. We also do not know if the difference between the UDCA and placebo groups was significant. Nevertheless, reductions in PHG, observed only in patients taking UDCA, provide for the first time evidence that portal hypertension may be partially reversible in PBC. A reversible component of portal hypertension in PBC is not yet explained, but could potentially involve relaxation of stellate cells around the sinusoids or decreased inflammation in the portal tracts. If a reversible component exists, it was not manifest in the group originally randomized to UDCA, because their PHG only fluctuated and did not improve overall. It is also important to note that of the 101 patients included in the long-term study, 20 patients on UDCA still experienced increases in PHG, almost half of whom developed new portal hypertension (PHG > 6 mmHg) while taking UDCA. A favorable effect of UDCA on portal hypertension (Table 1) was previously suggested by Lindor et al,9 who noted the delayed onset of esophageal varices in PBC patients treated with UDCA, but this was not confirmed in 2 other studies.10, 11 A recent update in 2008 of the Cochrane Database meta-analysis by Gong et al12 demonstrated no significant protective effect of UDCA treatment on the development of portal hypertensive complications (varices, variceal bleeding, or ascites) in the combined UDCA trials including hundreds of patients.

Table 1. Impact of UDCA Versus Placebo on Portal Hypertension in PBC
AuthorOutcomenSignificant Difference?
Huet et al1Portohepatic gradient (direct measurement)132Not directly compared
Lindor et al9, 15Development of varicesb114Yes
Combes et ala10Development of varices152No
Oka et al11Development of varices52No
Poupon et al16Variceal bleeding146No
Oka et al11Variceal bleeding52No
Pares et al13Variceal bleeding192No
Vuoristo et al14Variceal bleeding61No
Lindor et al9, 15Hepatic encephalopathy110No
Pares et al13Hepatic encephalopathy192No
Poupon et al16Ascites146No
Lindor et al9, 15Ascites110Trend
Oka et al11Ascites52No
Pares et al13Ascites192No

aThis trial also reported no significant difference in the combined outcome of variceal bleeding, development of ascites, or encephalopathy.

bOnly with prolonged follow-up. There was no significant difference in development of varices in the original 2-year trial.

The data in Huet et al's paper explain these discrepancies: portal pressures may only rise 2–3 mm Hg in placebo-treated patients over the duration of a 2-year study, probably not enough to be able to detect a difference between groups by screening for varices or performing examinations for ascites. Such small differences could only be reliably detected by PHG measurements.

Direct PHG measurement, however, is an invasive procedure and not easily justified in all patients. The paper by Huet et al establishes the important significance of portal pressure and its utility as a marker of survival, but noninvasive techniques of accurate PHG measurement in PBC need to be developed for more common use.

Back to Article Outline

References 

  1. Huet PM, Vincent C, Deslaurier J, et al. Portal hypertension and primary biliary cirrhosis: effect of long-term ursodeoxycholic acid treatment. Gastroenterology. 2008;153:1552–1560
  2. Zeeagen R, Stansfeld AG, Dawson AM, et al. Bleeding oesophageal varices as the presenting feature in primary biliary cirrhosis. Lancet. 1969;2:9–13
  3. Lebrec D, Sicot C, Degott C, et al. Portal hypertension and primary biliary cirrhosis. Digestion. 1976;14:220–226
  4. Kew MC, Varma RR, Dos Santos HA, et al. Portal hypertension in primary biliary cirrhosis. Gut. 1971;12:830–834
  5. Garrido MC, Hubscher SG. Accuracy of staging in primary biliary cirrhosis. J Clin Pathol. 1996;49:556–559
  6. Colina F, Pinedo F, Solis JA, et al. Nodular regenerative hyperplasia of the liver in early histological stages of primary biliary cirrhosis. Gastroenterology. 1992;102:1319–1324
  7. Levy K, Zein CO, Gomez J, et al. Prevalence and predictors of esophageal varices in patients with primary biliary cirrhosis. Clin Gastroenterol Hepatol. 2007;5:803–808
  8. Bressler B, Pinto R, El-Ashry D, Heathcote J. Which patients with primary biliary cirrhosis or primary sclerosing cholangitis should undergo endoscopic screening for oesophageal varices detection?. Gut. 2005;54:407–410
  9. Lindor KD, Jorgensen RA, Therneau TM, et al. Ursodeoxycholic acid delays the onset of esophageal varices in primary biliary cirrhosis. Mayo Clin Proc. 1997;72:1137–1140
  10. Combes B, Carithers RL, Maddrey WC, et al. A randomized double-blind placebo-controlled trial of ursodeoxycholic acid in primary biliary cirrhosis. Hepatology. 1995;22:759–766
  11. Oka H, Toda G, Ikeda Y, et al. A multi-center double blind controlled trial of ursodeoxycholic acid for primary biliary cirrhosis. Gastroenterologia Japonica. 1990;25:774–780
  12. Gong Y, Huang ZB, Christensen E, et al. Ursodeoxycholic acid for primary biliary cirrhosis. Cochrane Database of Systematic Reviews. 2008;1:CD000551;DOI: 10.1002/14651858.CD000551.pub2
  13. Pares A, Caballeria L, Rodes J, et al. Long term effects of ursodeoxycholic acid in primary biliary cirrhosis: results of a double-blind controlled multicentric trial. J Hepatol. 2000;32:561–566
  14. Vuoristo M, Farkkila M, Gylling H, et al. A placebo-controlled trial of primary biliary cirrhosis treatment with colchicine and ursodeoxycholic acid. Gastroenterology. 1995;108:1470–1478
  15. Lindor KD, Dickson Er, Baldus WP, et al. Ursodeoxycholic acid in the treatment of primary biliary cirrhosis. Gastroenterology. 1994;106:1284–1290
  16. Poupon RE, Balkau B, Eschwege E, et al. The UDCA-PBC Study Group A multi-center, controlled trial of ursodiol for the treatment of primary biliary cirrhosis. N Engl J Med. 1991;324:1548–1554

 The author discloses no conflicts.

PII: S0016-5085(08)01771-X

doi:10.1053/j.gastro.2008.09.041

Refers to article:

  • CME QuizEditorial Accompanies ArticleAdditional Online Content Available Portal Hypertension and Primary Biliary Cirrhosis: Effect of Long-Term Ursodeoxycholic Acid Treatment , 24 July 2008

    Pierre–Michel Huet, Catherine Vincent, Julie Deslaurier, Jean Coté, Shoichi Matsutami, Robert Boileau, Jacline Huet–van Kerckvoorde
    Gastroenterology November 2008 (Vol. 135, Issue 5, Pages 1552-1560)

Gastroenterology
Volume 135, Issue 5 , Pages 1450-1451, November 2008

Article Tools