Elsevier

Journal of Hepatology

Volume 48, Issue 3, March 2008, Pages 465-470
Journal of Hepatology

Early switch to pentoxifylline in patients with severe alcoholic hepatitis is inefficient in non-responders to corticosteroids

https://doi.org/10.1016/j.jhep.2007.10.010Get rights and content

Background/Aims

In severe alcoholic hepatitis (AH), 40% of patients will obtain no benefit from corticosteroids. Improvement in management of non-responders is warranted and only pentoxifylline can be considered an alternative. A two-step strategy was evaluated consisting of early withdrawal of corticosteroids and a switch to pentoxifylline for 28 additional days in non-responders identified using early change in bilirubin level.

Methods

One hundred and twenty-one patients with AH were treated prospectively with corticosteroids, and the two-step strategy was proposed to 29 non-responders treated according to a two-step strategy who were compared to 58 matched non-responders treated with corticosteroids only.

Results

Clinical and biological features of the two groups were similar. There was no survival improvement at 2 months in patients treated with the two-step strategy compared to controls: 35.5 ± 6.3% vs 31 ± 8.6%. After 21 days, biological evolution was similar for prothrombin time (−0.25 s vs +0.2 s), bilirubin (0.8 mg/dl vs 2.03 mg/dl) and creatinine (+0.16 mg/dl vs −0.7 mg/dl). In multivariate analysis, only age, evolution of bilirubin during the first week, creatinine and DF were associated with 2-month survival.

Conclusions

Non-responders to corticosteroids do not obtain any benefit from an early switch to pentoxifylline. Thus, the issue of management of non-responders remains unresolved.

Introduction

Several investigators have given priority to treatment of severe forms of alcoholic hepatitis (AH), as defined by a discriminant function of Maddrey (DF) ⩾32, so as to improve short-term survival [1], [2]. Among available therapeutic options, and although some clinicians are reluctant to use corticosteroids, compelling data have shown that this treatment improves short-term survival compared to placebo [1], [3], [4], [5], [6], [7], [8]. However, novel strategies or molecules are required in light of the fact that approximately 40% of patients treated with corticosteroids die within 6 months [9]. In order to progress in the management of patients treated with corticosteroids, early identification of patients who do not benefit from this treatment is important when choosing candidates for new therapeutic strategies. To achieve this objective, we proposed a simple criterion for early identification of so-called “non-responders to corticosteroids” [4]. This criterion, termed “early change in bilirubin levels (ECBL)”, is defined as a bilirubin level lower at 7 days than the bilirubin level on the first day of treatment. Seventy-three percent of patients registered ECBL at 7 days, and their 6-month survival was higher than that of patients without ECBL, 82.8 ± 3.3% vs 23 ± 5.8%.

Among available molecules that have already been tested, only pentoxifylline may be considered an alternative to corticosteroids, as shown in a double-blind randomized controlled trial comparing pentoxifylline to placebo [10]. Twenty-four percent of pentoxifylline-treated patients and 46.1% of control patients died during hospitalization, respectively. The survival benefit of pentoxifylline appears to be related to a significant reduction in development of the hepatorenal syndrome. Among patients who died, hepatorenal syndrome developed in 50% of pentoxifylline patients and 91.7% of placebo patients. The mechanisms involved in corticosteroid and pentoxifylline effects seem to differ under experimental conditions when compared to human levels. In vitro, pentoxifylline and steroids are anti-inflammatory molecules [11], [12], [13]. However, in human beings with severe AH, the main mechanism involved in the pentoxifylline protective effect does not seem to be related to its anti-inflammatory effect, but rather, to prevention of hepatorenal syndrome, without any decrease in pro-inflammatory cytokines or improvement in liver tests [10]. Conversely, several studies in human beings have shown that steroids ameliorate liver function, and have confirmed their inhibition effect in pro-inflammatory cytokines and polymorphonuclear neutrophil activation [14], [15].

Since pentoxifylline and corticosteroids seem to act via different pathways in humans, our aim was to test, in a prospective study, the efficacy of a two-step strategy consisting of a switch from corticosteroids to pentoxifylline in non-responders identified early on. This consisted of: (1) treating all patients having severe forms with corticosteroids in a first step, and identifying early on all non-responders to corticosteroids using ECBL; and (2) discontinuing corticosteroids in non-responders and switching to pentoxifylline.

Section snippets

Treatment criteria and treatment protocol

From December 2002 to January 2005, at two French hospitals (Hôpital Huriez, Lille, and Hôpital Germon et Gauthier, Béthune), we prospectively evaluated all patients admitted for a severe form of AH treated with corticosteroids. Indication of treatment with corticosteroids was based on the following criteria: (a) severity of AH defined by Maddrey function ⩾32; (b) recent onset of jaundice (less than 2 months); (c) history of long-standing alcoholism; (d) liver chemistry suggestive of severe AH;

Prospective cohort of patients treated with corticosteroids before ECBL evaluation

AH was biopsy-proven [17] in approximately 90% of cases using international criteria such as hepatocellular necrosis, Mallory bodies and infiltration of polymorphonuclear leukocytes. In the remaining cases, histological diagnosis was not available due to technical procedure failure or insufficient liver sample size.

All treated patients (n = 121) were classified according to ECBL at the end of the first week of treatment. Clinical and biological features at baseline and evolution of liver and

Discussion

Management of patients with severe AH would be improved by the development of a treatment strategy that integrated reversal of liver damage with prevention of extrahepatic injury such as hepatorenal syndrome. The present study confirms the absence of an early biological response as a strong predictor of poor outcome. For improving therapeutic strategies in such patients, clinicians can alternatively: (a) evaluate new strategies in order to increase the probability of biological response and

Cited by (0)

The authors who have taken part in the research of this paper declared that they do not have a relationship with the manufacturers of the drug involved either in the past or present and they did not receive funding from the manufacturers to carry out their research. They did not receive funding from any source to carry out this study.

View full text