A multicenter, double-blind, placebo-controlled trial randomized 28 patients with primary (acute) human immunodeficiency virus (HIV)-1 infection (PHI) to receive zidovudine, 1000 mg daily, or placebo for 24 weeks. At week 48, compared with placebo patients, zidovudine-treated patients had significantly higher CD4 cell counts (zidovudine, 666 cells/mm3; placebo, 362; P = .004) and lower peripheral blood mononuclear cell (PBMC) culture titers (zidovudine, 0.58 log infectious units per million cells; placebo, 1.68; P = .02) but no difference in plasma RNA (zidovudine, 3.93 log copies/mL; placebo, 4.00; P = .83). Serious adverse events and minor clinical events were infrequent and comparable in both arms. There were two deaths: 1 patient died of sepsis and renal disease (zidovudine arm), and 1 patient died of sepsis and tension pneumothorax (placebo arm). Six months of high-dose zidovudine initiated during PHI results in higher CD4 cell counts and lower PBMC culture titers but no difference in plasma HIV-1 RNA. Further studies with more potent antiretroviral combination therapies are warranted.