Postherpetic neuralgia is the most common complication of herpes zoster (shingles) in the immunocompetent host. Its mechanism is incompletely understood, but one postulate is that continuous replication of varicella-zoster virus (VZV) in nerve tissues may be responsible for the pain. If this is so, antiviral treatment could be advantageous. To test this hypothesis, we performed a randomized, double-blind, placebo-controlled trial of intravenous acyclovir (10 mg/kg every 8 h [q8h]) for 14 days, followed by oral acyclovir (800 mg q6h) for 42 days in 10 subjects (median age, 71 years) who had experienced at least 6 months of severe pain (median duration of postherpetic neuralgia before enrollment, 3.2 years). Intensive and sparse pharmacokinetic sampling occurred during both dosing phases of the study. One- and two-compartment models were fitted to the oral and intravenous concentration-time data, respectively. The four men and four women assigned to acyclovir during either or both dosing phases tolerated it well. Pharmacokinetic results were similar to those previously reported in younger individuals. The mean oral clearance and elimination half-life following oral dosing were 1.47 liters/h/kg and 2.78 h, respectively. Total clearance and terminal half-life following intravenous administration were 0.16 liters/h/kg and 3.67 h, respectively. Only 1 of 10 participants reported definite improvement in the severity of postherpetic pain, and treatment had no effect on titers of humoral antibody to VZV. We concluded that 56 days of intravenous and oral acyclovir therapy were well tolerated but had little or no effect on the clinical course of postherpetic neuralgia.