Early liver biopsy, intraparenchymal cholestasis, and prognosis in patients with alcoholic steatohepatitis.

BACKGROUND: Alcoholic steatohepatitis (ASH) is a serious complication of alcoholic liver disease. The diagnosis of ASH requires the association of steatosis, evidence of hepatocellular injury with ballooning degeneration, and polynuclear neutrophil infiltration on liver biopsy. Whether these lesions, in addition to other histological features observed in liver tissue specimens, have prognostic significance is unclear.

METHODS: We studied 163 patients (age 55 yrs [35-78], male/female 102/61) with recent, heavy (> 80 gr/day) alcohol intake, histologically-proven ASH (97% with underlying cirrhosis, Maddrey's score 39 [13-200], no sepsis), who had a liver biopsy performed 3 days [0-10] after hospital admission for clinical decompensation. A semi-quantitative evaluation of steatosis, hepatocellular damage, neutrophilic infiltration, periportal ductular reaction, intraparenchymal cholestasis, and iron deposits was performed by two pathologists. All patients with a Maddrey's score ≥ 32 received steroids. The outcome at 3 months was determined. Statistical analysis was performed using the Wilcoxon and Fisher's exact tests, Kaplan-Meier method, and the Cox proportional hazard model.

RESULTS: 43 patients died after 31 days [5-85] following biopsy. The 3-month survival rate was 74%. Mean kappa value for histological assessment by the two pathologists was excellent (0.92). Univariate analysis identified age, the Maddrey's score, the Pugh's score, the MELD score and parenchymal cholestasis, but not other histological features, as factors associated with 3-month mortality. At multivariate analysis, age (p = 0.029, OR 2.83 [1.11-7.2], intraparenchymal cholestasis (p = 0.001, OR 3.9 [1.96-7.8], and the Maddrey's score (p = 0.027, OR 3.93 [1.17-13.23] were independent predictors of outcome. Intraparenchymal cholestasis was more frequent in non survivors compared to survivors (70% versus 25%, p < 0.001). Serum bilirubin was higher in patients with severe compared to those with no or mild intraparenchymal cholestasis (238 [27-636] versus 69 [22-640] umol/l, p < 0.001).

CONCLUSIONS: In this large cohort of patients with histologically documented ASH early after admission and no sepsis, liver biopsy identified marked intraparenchymal cholestasis as an independent predictor of poor short term outcome together with age and the Maddrey's score. It may be hypothesized that incorporation of this particular variable into existing disease severity scores for ASH would improve their performance.