Recent studies have shown that hepatitis C virus antibodies are present in a large proportion of patients with autoimmune hepatitis type 2. We have studied 83 patients with liverkidney microsome antibodypositive type 1 hepatitis. Hepatitis C virus antibodies were sought in every case by second-generation tests (hepatitis C virus enzyme-linked immunosorbent assay and recombinant immunoblot assay). Hepatitis C virus RNA sequences were sought in 22 patients (12 with recombinant immunoblot assay-positive results and 10 with recombinant immunoblot assay-negative results) by means of polymerase chain reaction and by use of primers located in the 5' noncoding region. Sixty-four patients (77%) had positive results for hepatitis C virus antibodies in the enzyme-linked immunosorbent assay test, and 41 (49.3%) were confirmed by recombinant immunoblot assay. Hepatitis C virus RNA sequences were found in all the recombinant immunoblot assaypositive patients but in none of the 10 who were recombinant immunoblot assay-negative. The recombinant immunoblot assay-negative patients were younger than those who were positive (13 * 11 vs.
60 & 11 years) and had higher rglobulin levels and liverkidney microsome antibody-positive type 1 titers (61% had a titer of 1:1,000 or more, vs. only 17% of the recombinant immunoblot assay-positive patients). On the basis of these results, chronic hepatitis with liverfidney microsome antibody-positive type 1 can be divided into two subgroups: (a) true autoimmune hepatitis type 2 (mainly observed in young women), with high titers of liverkidney microsome antibodypositive type 1, and in which direct autoimmune mechanisms appear to be prominent; and (b) hepatitis C virus-associated hepatitis with liverkidney microsome antibody-positive type l (generally affecting older patients, especially men), with low titers of liverkidney microsome antibody-positive type 1, and in which the autoimmune process could be a come-