Despite the use of vidarabine, herpes simplex virus (HSV) infection in neonates continues to be a disease of high morbidity and mortality. We undertook a controlled trial comparing vidarabine with acyclovir for the treatment of neonatal HSV infection. Babies less than one month of age with virologically confirmed HSV infection were randomly and blindly assigned to receive either intravenous vidarabine (30 mg per kilogram of body weight per day; n = 95) or acyclovir (30 mg per kilogram per day; n = 107) for 10 days. Actuarial rates of mortality and morbidity among the survivors after one year were compared overall and according to the extent of the disease at entry into the study (infection confined to the skin, eyes, or mouth; encephalitis; or disseminated disease). After adjustment for differences between groups in the extent of disease, there was no difference between vidarabine and acyclovir in either morbidity (P = 0.83) or mortality (P = 0.27). None of the 85 babies with disease confined to the skin, eyes, or mouth died. Of the 31 babies in this group who were treated with vidarabine and followed for a year, 88 percent (22 of 25) were judged to be developing normally after one year, as compared with 98 percent (45 of 46) of the 54 treated with acyclovir (95 percent confidence interval for the difference, –4 to 24). For the 71 babies with encephalitis, mortality was 14 percent with vidarabine (5 of 36) and with acyclovir (5 of 35); of the survivors, 43 percent (13 of 30) and 29 percent (8 of 28), respectively, were developing normally after one year (95 percent confidence interval for the difference, –11 to 39). For the 46 babies with disseminated disease, mortality was 50 percent (14 of 28) with vidarabine and 61 percent (11 of 18) with acyclovir (95 percent confidence interval for the difference, –20 to 40); of the survivors, 58 percent (7 of 12) and 60 percent (3 of 5), respectively, were judged to be developing normally after one year (95 percent confidence interval for the difference, –40 to 50). Both medications were without serious toxic effects. In this multicenter, randomized, blinded study there were no differences in outcome between vidarabine and acyclovir in the treatment of neonatal HSV infection. The study lacked statistical power to determine whether there were sizable differences within the subgroups of those with localized HSV, encephalitis, or disseminated disease. (N Engl J Med 1991; 324:444–9). © 1991, Massachusetts Medical Society. All rights reserved.