AST/ALT ratio > or = 1 is not diagnostic of cirrhosis in patients with chronic hepatitis C.

Abstract

Medical guidelines for interferon-alpha2a or -alpha2b (IFN-alpha) treatment of chronic hepatitis C virus (HCV) infection depend upon baseline liver histology. A better long-term response to IFN-alpha therapy correlates with less inflammation and absence of cirrhosis. It has been suggested that the presence of cirrhosis in patients with chronic hepatitis C virus infection may be predicted based on an AST/ALT ratio > or = 1. This study was designed to determine if the presence of cirrhosis can be predicted in patients with chronic HCV infection by such a ratio. Seventy-seven patients, including 23 cirrhotics, with chronic HCV infection were studied. Serum ALT, AST, and HCV-RNA levels and hepatic activity index (HAI), reflecting histologic inflammation in all liver biopsies, were assessed. AST/ALT ratios and mean ALT, AST, and HCV-RNA were determined for both cirrhotic and noncirrhotic patients. HAI was correlated with ALT, AST, and HCV-RNA levels, the latter determined by quantitative RT-PCR. The likelihood ratio (LR) and positive predictive value of an AST/ALT ratio > or = 1 for cirrhosis was 7.3 and only 77%, respectively. In cirrhotics vs noncirrhotics, there were no significant differences between mean serum ALT (149 +/- 28 vs 176 +/- 17 units/liter), AST (139 +/- 28 vs 102 +/- 8 units/liter), or HCV-RNA levels (589,160 +/- 147,053 vs 543,915 +/- 75,497 copies/ml), respectively. There was a significant, but clinically weak, correlation between serum ALT and HAI (r = 0.234), and none between HAI and either serum AST or HCV-RNA levels. Our results support the need for a liver biopsy prior to treatment of chronic HCV infection, since the AST/ALT ratio fails to predict accurately the presence of cirrhosis.