Recently, two distinct hepatitis C virus (HCV) serologic types have been identified on the basis of amino acid variations in the core region. The two serologic types can readily discriminate between genotypes I-II-V (serotype 1) and III-IV (serotype 2), according to the Okamoto classification. We compared HCV core serotyping with genotyping with sera from 363 anti-HCV-positive patients (309 HCV RNA positive by PCR) using a synthetic core peptide-based enzyme immunoassay and PCR amplification of core region sequences with type-specific primers, respectively. Serologic responses to HCV serotypes were successfully identified in 164 (45%) patients, of whom 153 were viremic. Eighty-nine patients had evidence of exposure to serotype 1: 8 of these were infected with genotype I, 50 were infected with genotype II, 2 were infected with genotype III, 7 were infected with genotype V, 13 had infections with mixed genotypes, 3 were infected with an indeterminate genotype, and 6 were nonviremic. Seventy-four patients had been exposed to serotype 2: 64 were infected with genotype III, 3 were infected with mixed genotypes, 2 were infected with an indeterminate genotype, and 5 were nonviremic. The serum of one patient, infected with genotype III, showed reactivity to both serotypes. Comparative evaluation of HCV core region serotyping and genotyping with sera from 294 viremic patients infected with a known HCV genotype showed a remarkable concordance between HCV core region genotyping and serotyping, with only 2 apparently discordant serum samples (both from patients with genotype III infection) of 148 (1.4%) successfully serotyped samples. Serotype 1 infection was more frequently observed in patients with overt chronic liver disease and accounted for all successfully serotyped samples from intravenous drug abusers. In contrast, serotype 2 was more prevalent in subjects with biochemically silent HCV infection (alanine aminotransferase, < 45 U/liter), in agreement with previous findings at the molecular level. HCV core serologic typing is a simple, inexpensive, and highly reproducible assay that can be applied to more than 50% of viremic HCV antibody carriers prior to the use of more sophisticated molecular typing techniques. Moreover, it may be helpful in tracking transmissions routes, particularly for incorrectly stored samples in which the RNA has degraded or for subjects who have cleared the virus and therefore have only antibodies remaining to testify to a remote infection. The lack of recognition of the core sequence from residues 67 to 81, which contains a minor B-cell epitope used to detect type-specific immunoreactivity, may explain the negative serologic findings for half of the patients.