Z Gastroenterol 2017; 55(05): e28-e56
DOI: 10.1055/s-0037-1603433
Hepatologie
Georg Thieme Verlag KG Stuttgart · New York

Limited value of clinical criteria for diagnosis of alcoholic steatohepatitis (ASH)

R Stauber
1   Department of Internal Medicine, Medical University of Graz, Graz, Austria
,
W Spindelböck
1   Department of Internal Medicine, Medical University of Graz, Graz, Austria
,
F Rainer
1   Department of Internal Medicine, Medical University of Graz, Graz, Austria
,
P Douschan
1   Department of Internal Medicine, Medical University of Graz, Graz, Austria
,
C Lackner
2   Institute of Pathology, Medical University of Graz, Graz, Austria
› Author Affiliations
Further Information

Publication History

Publication Date:
16 May 2017 (online)

Also available at
 

Background:

Current clinical practice guidelines recommend liver biopsy for diagnosis of alcoholic steatohepatitis (ASH) in patients with clinically suspected ASH and a Maddrey score of ≥32. A recent study suggested a high positive predictive value for diagnosing ASH based on clinical criteria including white blood count (WBC) ≥10.75 G/l and platelet count (PLT) ≥147.5 G/l (Hardy et al, 2013). The aim of the present study was to evaluate these criteria in an independent cohort of patients with biopsy-proven ASH.

Methods:

Consecutive patients with suspected severe alcoholic hepatitis on the basis of heavy drinking (alcohol consumption > 60 g/day) and Maddrey score ≥32 underwent percutaneous or transjugular liver biopsy. Patients with concomitant hepatocellular carcinoma or previous liver transplantation were excluded. Histological ASH was defined as the presence of both hepatocellular ballooning and lobular inflammation with any degree of steatosis. Patients were followed until death or liver transplantation and liver-related death was recorded.

Results:

Forty-eight patients underwent liver biopsy between 1995 and 2009 and were followed until December 2009 (median follow-up 4.1 years, range 0 – 15 years). Histological ASH was found in 29/48 patients (60%) while it was absent in 19/48 patients (40%). Five-year survival was 35% in ASH positive and 47% in ASH negative patients. In patients with WBC ≥10.75 G/l, ASH was found in 17/20 (85%) while in patients with PLT ≥147.5 G/l, ASH was found in 8/11 (73%). Among the 7 patients with both criteria, ASH was present in 6 (86%).

Conclusions:

Among patients with clinically suspected ASH and a Maddrey score ≥32, histological ASH was found in only 60%. When published WBC and PLT thresholds were applied as noninvasive predictors of histological ASH, specificity was increased but these criteria were applicable only in a minority of patients (7/48, 15%) and thus are of limited clinical value.