BACKGROUND It is possible that antiretroviral treatment given early during primary infection with the human immunodeficiency virus (HIV) may reduce acute symptoms, help preserve immune function, and improve the long-term prognosis. METHODS To assess the effect of early antiviral treatment, we conducted a multicenter, double-blind, placebo-controlled trial in which 77 patients with primary HIV infection were randomly assigned to receive either zidovudine (250 mg twice daily; n = 39) or placebo (n = 38) for six months. RESULTS The mean time from the onset of symptoms until enrollment in the study was 25.1 days. Among the 43 patients who were still symptomatic at the time of enrollment, there was no appreciable difference in the mean (+/- SE) duration of the retroviral syndrome between the zidovudine group (15.0 +/- 4.1 days) and the placebo group (15.8 +/- 3.6 days). During a mean follow-up period of 15 months, minor opportunistic infections developed in eight patients: oral candidiasis in four, herpes zoster in two, and oral hairy leukoplakia in two. Disease progression was significantly less frequent in the zidovudine group (one opportunistic infection) than in the placebo group (seven opportunistic infections; P = 0.009 by the log-rank test). After adjustment for the base-line CD4 cell count, the patients treated with zidovudine had an average gain of 8.9 CD4 cells per cubic millimeter per month (95 percent confidence interval, -1.4 to 19.1) during the first six months of the study, whereas those receiving placebo had an average loss of 12.0 CD4 cells per cubic millimeter per month (95 percent confidence interval, 5.2 to 18.7), for a between-group difference of 20.9 CD4 cells per cubic millimeter per month (95 percent confidence interval, 8.5 to 33.2; P = 0.001). CONCLUSIONS Antiretroviral therapy administered during primary HIV infection may improve the subsequent clinical course and increase the CD4 cell count.