HEPATITIS-C INFECTION BY POLYMERASE CHAIN-REACTION IN ALCOHOLICS - FALSE-POSITIVE ELISA RESULTS AND THE INFLUENCE OF INFECTION ON A CLINICAL PROGNOSTIC SCORE

Antibody to hepatitis C as measured by the ELISA method is common in a lcoholics. The presence of antibody to C 100-3 has been associated wit h more advanced disease. However, few studies have investigated the cl inical significance of hepatitis C infection as defined by the presenc e of circulating viral RNA in alcoholics. We have prospectively examin ed 48 consecutive alcoholic patients for the presence of antibody to h epatitis C by an ELISA for antibody to the C100-3 antigen and by the r everse transcriptase polymerase chain reaction (PCR) using nested prim ers for the 5' nontranslated region of the viral RNA. Patients with li ver disease were scored for disease severity by the combined clinical and laboratory index (CCLI). Overall, 12 of 48 patients (25%) were ELI SA positive and eight of 48 (16%) were PCR positive. Among the 34 pati ents with liver disease, 10 (29%) were ELISA positive and six (18%) we re PCR positive. Alt PCR-positive patients were also ELISA positive. T here was no significant difference in the disease severity score (CCLI ) or the duration of clinical disease in PCR-positive versus PCR-negat ive patients with liver disease. However, PCR-positive patients were s ignificantly younger (43 +/- 6 vs. 55 +/- 10 yr, p = 0.001), indicatin g an earlier onset of severe disease in PCR-positive patients. There w ere no false-negative ELISA tests in either those with or those withou t liver disease. Among the 34 patients with liver disease, four of 10 patients with positive antibody were negative by PCR. Neither individu al immunoglobulin levels (IgG, IgM, IgA) nor total globulins were sign ificantly different between the ELISA-positive/PCR-negative patients a nd ELISA-positive/PCR-positive patients. When the entire group of 34 p atients with liver disease was considered, we could not detect a signi ficant correlation between ELISA absorbance and total globulins, and o nly a weak correlation between absorbance and immunoglobulin G(rho = 0 .49). These data show that the majority of alcoholic patients with liv er disease and positive antibody to hepatitis C also have demonstrable viremia by PCR, and may require further evaluation and treatment. Ele vated immunoglobulins in these patients do not correlate strongly with ELISA absorbance for anti-HCV. The presence of clinically advanced di sease at a significantly younger age in the PCR-positive group is cons istent with the concept of synergy between active viral infection and alcohol abuse in the development of liver disease in alcoholic patient s.