Neil Osterweil

November 14, 2014

BOSTON — Prednisolone reduces the short-term risk for death in patients with severe alcoholic hepatitis, but pentoxifylline does not and should no longer be used in these patients, according to investigators from the STOPAH trial.

The work, also known as the Steroids or Pentoxifylline for Alcoholic Hepatitis trial and presented here at The Liver Meeting 2014, shows that prednisolone improves 28-day mortality, but has no impact beyond this time point.

"Pentoxifylline appears to have no benefit in this condition, and abstinence, not surprisingly, is the major determinant of survival," said lead investigator Michael Richard Thursz, MD, from Imperial College London in the United Kingdom.

At 1 year, however, there was no difference in death rates between the two drugs, he reported.

Alcoholic hepatitis is a distinct clinical syndrome seen in people who actively and chronically abuse alcohol, defined in this study as men who consume more than 80 g of alcohol day or women who consume more than 60 g daily. The syndrome is marked by jaundice, a serum bilirubin level above 80 µmol/L, and functional liver impairment.

Patients with severe alcoholic hepatitis have a short-term mortality rate greater than 30%.

Guidelines on the management of alcoholic liver disease from the European Association for the Study of the Liver (J Hepatol. 2012;57:399-420) and the American Association for the Study of Liver Diseases (Hepatology. 2010;51:307-328) mention the steroid prednisolone and the oral phosphodiesterase inhibitor pentoxifylline as treatment options for acute alcoholic hepatitis, "yet both of these have some controversy behind them," Dr Thursz said.

Controversial Options

A systematic review of 15 clinical studies showed no clear evidence to support the use of glucocorticosteroids to treat alcoholic hepatitis (Aliment Pharmacol Ther. 2008;27:1167-1178). In contrast, a meta-analysis from France showed significantly better 28-day survival in patients treated with steroids than in those treated with placebo (Gut. 2011;60:255-260).

The basis for the use of pentoxifylline comes from a single trial that showed a lower mortality rate in patients treated with pentoxifylline than in those treated with placebo (24.5% vs 46.1%) (Gastroenterology. 2000;119:1637-1648).

Abstinence, not surprisingly, is the major determinant of survival.

To determine whether either agent is clinically effective, Dr Thursz and colleagues conducted a randomized double-blind placebo controlled trial with a factorial 2 × 2 design.

All 1103 study participants had a clinical diagnosis of alcoholic hepatitis and were 18 years and older. About two-thirds of the cohort was male and about 95% was white. The mean daily consumption of alcohol was 142 to 157 g in women and 195 to 210 g in men. Mean time from admission for acute alcoholic hepatitis to treatment was about 6 days.

Patients were randomized to one of four treatment groups: prednisolone and pentoxifylline, prednisolone and placebo, pentoxifylline and placebo, or double placebo.

Prednisolone, but not pentoxifylline, was associated with a lower risk for 28-day mortality, the study's primary end point.

Table 1. STOPAH Mortality at 28 Days

Treatment Group Patients (n = 1103) Percent
Prednisolone and pentoxifylline 274 13.5
Prednisolone and placebo 277 14.3
Pentoxifylline and placebo 276 19.4
Double placebo 276 16.7


For all patients who received prednisolone, the mortality rate was 13.9%; for patients who received only pentoxifylline or placebo, the rate was 18.0%.

In contrast, the mortality rate was similar in patients who received pentoxifylline and those who did not (16.4% vs 15.5%).

In an unadjusted analysis, the odds ratio (OR) for 28-day mortality with prednisolone, compared with no prednisolone, was 0.72, which was of borderline significance (P = .056). When pentoxifylline was compared with no pentoxifylline, the OR was 1.07 (P = .686).

When disease severity and prognosis at baseline were accounted for, certain factors were shown to be predictors of 28-day mortality.

Table 2. Factors Significantly Associated With 28-Day Mortality

Factor Odds Ratio P Value
Prednisolone use 0.609 .015
Prothrombin time ratio 1.380 .002
Bilirubin level 1.002 .003
Age 1.050 <.001
White blood cell count 1.030 .037
Urea level 1.065 .037
Creatinine level 1.564 .028
Hepatic encephalopathy 3.073 <.001


Beyond 28 days, however, neither drug was significantly associated with a survival benefit.

Infections were about twice as frequent in the prednisolone group than in the no-prednisolone group (13.5% vs 7.9%; = .0026).

At 1 year, patients who did not cut down or who increased their alcohol consumption had a 3-fold risk for death, compared with patients who abstained (OR, 2.99; < .001).

Patients who reduced their drinking, but not below safety limits still had a more than a 2-fold risk for death at 1 year, compared with patients who abstained (OR, 2.28; P = .032), as did patients who reduced their drinking to below safety levels (OR, 2.17; = .031).

This study was well designed, said Evangelos Akriviadis, MD, from Αristotle University of Thessaloniki Medical School and the Hippokration General Hospital in Greece, who was involved in the 2000 study that showed lower mortality with pentoxifylline.

"In our study, the mortality rate in the control group was 45%," Dr Akriviadis said. "This was the experience with all previous studies in our center, so it is apparent that this may not be the same group of patients in terms of severity."

In response, Dr Thursz said, "If we look at the overall survival in your trial and we look at the steroid trials that go right back to 1971, the mortality rate in the earlier trials was undoubtedly a lot higher than we're seeing now."

"How do I explain that? I think we probably look after these patients a lot better," he explained. "In addition, we rarely see the old-fashioned 'skid row' alcoholic who's cachectic. Most of these patients, in fact, are well-nourished and overweight, so I think we're looking at a slightly different group of patients."

Dr Akriviadis asked how the investigators coped with the excess risk for infection with prednisolone in this severely ill population, and how they explain the difference in survival between the drug groups at 28 days but not at 1 year.

Dr Thursz said the reason for the convergence of the survival curves after 1 month might be related to more infections in patients treated with steroids, which occur at a higher rate even after they have stopped therapy.

The STOPAH trial was supported by the UK National Institute for Health Research. Dr Thursz and Dr Akriviadis have disclosed no relevant financial relationships.

The Liver Meeting 2014: American Association for the Study of Liver Diseases (AASLD). Abstract LB-1. Presented November 10, 2014.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.