Objective. The objective of this investigation was to establish the safety of high-dose (HD) acyclovir for the treatment of neonatal herpes simplex virus (HSV) disease. In addition, an estimate of therapeutic efficacy was sought, both with respect to mortality and to morbidity. Virologic efficacy of HD acyclovir was also assessed. Participants. Infants who were ≤28 days old and whose disease was considered to be caused by HSV were enrolled in this study. Patients with central nervous system (CNS; N = 28) or disseminated (N = 41) HSV infection were offered participation in the trial. A small number of patients with HSV disease limited to the skin, eyes, or mouth (SEM; N = 10) or whose disease was clinically consistent with HSV but who did not have virologic confirmation of infection (N = 9) also were enrolled on a compassionate basis. Only patients with virologically confirmed HSV disease were included in efficacy analyses. All enrolled patients were included in safety analyses. Methods. The study was an open-label evaluation of intravenous acyclovir at dosages higher than the 30 mg/kg/d standard dosage approved by the US Food and Drug Administration. The first 16 patients enrolled received intermediate-dose (ID) acyclovir (45 mg/kg/d), and the next 72 patients received HD acyclovir (60 mg/kg/d). Acyclovir was administered in 3 divided daily doses for 21 days. Neonates were assessed prospectively throughout treatment and at scheduled follow-up visits for the first 4 years of life. Data were compared with those of a previous National Institute of Allergy and Infectious Diseases Collaborative Antiviral Study Group trial in which patients received standard-dose (SD) acyclovir for 10 days and in which identical methods (with the exception of acyclovir dosage and duration of therapy) were used. Results. Six (21%) of 29 HD acyclovir recipients whose HSV disease remained localized to the SEM or CNS experienced neutropenia. One of the 6 had an absolute neutrophil count <500/mm3, and 5 patients had an absolute neutrophil count (ANC) between 500/mm3 and 1000/mm3. In all 6 cases, the ANC recovered during continuation of acyclovir at the same dosage or after completion of acyclovir therapy, and there were no apparent adverse sequelae of the transient neutropenia. No other drug-related adverse events were reported among ID or HD recipients, and no other laboratory aberrations could be correlated specifically with antiviral therapy. The survival rate for the patients with disseminated HSV disease treated with HD acyclovir was significantly higher than for those in the previous study treated with SD acyclovir, with an odds ratio (OR) of 3.3 (95% confidence interval [CI]: 1.4-7.9). For patients with CNS disease, however, survival rates were similar for the HD and SD groups. To assess the effect of HD acyclovir on survival for the entire population with neonatal HSV disease, the Cox proportional hazards regression analysis was performed with stratification for disease category (CNS versus disseminated). In performing this analysis, differences in mortality for each disease category were weighted to allow statistical comparison of the treatment dosage groups (HD, ID, and SD). This analysis indicated that the survival rate for patients treated with HD acyclovir was statistically significantly higher than for patients treated with SD acyclovir (OR: 3.3; 95% CI: 1.5-7.3). Recipients of HD acyclovir had a borderline significant decrease in morbidity compared with SD recipients, after stratification for the extent of disease (SEM vs CNS vs disseminated) and controlling for the potential confounding factors of HSV type (HSV-1 vs. HSV-2), prematurity, and disease severity (seizures). Patients treated with HD acyclovir were 6.6 times (adjusted OR; 95% CI: 0.8-113.6) as likely to be developmentally normal at 12 months of age as patients treated with SD therapy. Conclusion. These data support the use of a 21-day course of HD (60 mg/kg/d) intravenous acyclovir to treat neonatal CNS and disseminated HSV disease. Throughout the course of HD acyclovir therapy, serial ANC determination should be made at least twice weekly. Decreasing the acyclovir dosage or administering granulocyte colony-stimulating factor should be considered if the ANC remains below 500/mm3 for a prolonged period.