Hepatitis C virus (HCV) is the main cause of parenteral non-A, non-B hepatitis and serum can be tested for the virus itself by reverse-transcription polymerase chain amplification. What of the level of this viraemia? To find out if quantitative study of HCV RNA might be useful clinically we took advantage of participation in trials of interferon-alpha in patients with chronic HCV infection and applied a new assay, branched DNA (bDNA) signal amplification. Paired serum and liver biopsy specimens from 47 patients with confirmed chronic HCV infection and evidence of HCV RNA in their serum were studied. The quantitative bDNA assay (detection limit 350,000 equivalents/mL [eq/mL]) was positive in 34 sera (sensitivity 72%). Patients who acquired HCV infection by blood transfusion had a higher viraemia (median 2,701,000 eq/mL, n = 29) than health workers and intravenous drug users (635,000 eq/mL, n = 13; p < 0.01). Patients with a sustained complete response to interferon-alpha therapy had lower pre-treatment viraemia levels (median at bDNA cut-off, n = 11) than complete responders who relapsed after the drug was stopped (1,613,000 eq/mL, n = 15; p < 0.01) and non-responders (3,066,000 eq/mL, n = 20; p < 0.01). High viraemia levels were not related to the histological diagnosis but were associated with lobular inflammation, lymphoid aggregates, and bile-duct lesions. These findings indicate that mode of acquisition is an important determinant of HCV viraemia and that patients with low HCV viraemia levels are more likely to respond to interferon in a sustained fashion.