Differentiation of nonalcoholic from alcoholic steatohepatitis: are routine laboratory markers useful?

Background/Aims: Specific markers for differentiation of nonalcoholic (NASH) from alcoholic steatohepatitis (ASH) are lacking. We investigated the role of routine laboratory parameters in distinguishing NASH from ASH. Methods: Liver biopsies performed at our hospital over a 10-year period were reviewed, 95 patients with steatohepatitis identified and their data prior to biopsy reevaluated. The diagnosis NASH or ASH was assigned (other liver diseases excluded) on the basis of the biopsy and history of alcohol consumption (< 140 g/week). Logistic regression models were used for analysis. Results: NASH was diagnosed in 58 patients (61%; 30 f) and ASH in 37 (39%; 9 f). High-grade fibrosis (59% vs. 19%, P < 0.0001) and an AST/ALT ratio > 1 (54.1% vs 20.7%, P = 0.0008) were more common in ASH. The MCV was elevated in 53% of ASH patients and normal in all NASH patients (P < 0.0001). Multivariate analysis identified the MCV (P = 0.0013), the AST/ALT ratio (P = 0.011) and sex (P = 0.0029) as relevant regressors (aROC = 0.92). The AST/ALT ratio (P < 0.0001) and age (P = 0.00049) were independent predictors of high-grade fibrosis. Differences in MCV were more marked in high-grade fibrosis. Conclusions: Higher MCVs and AST/ALT ratios in ASH reflect the severity of underlying liver disease and do not differentiate NASH from ASH. Instead, these biomarkers might prove useful in guiding selection of patients for liver biopsy and in targeting therapy.