In a randomized, placebo-controlled, double-blind trial of intravenous acyclovir in the treatment of varicella zoster virus (VZV) infections, 8 of 20 immunocompromised children with varicella received acyclovir (500 mg/m2/dose three times daily for 7 days). There was no significant difference in skin healing between the acyclovir and placebo groups although there was a significant reduction in the incidence of development of pulmonary involvement during acyclovir treatment. Nineteen out of 34 patients received vidarabine (10 mg/kg/day for 5 days). Vidarabine significantly shortened the duration of new vesicle formation. Both drugs significantly reduced the incidence of visceral varicella, the most serious complication of VZV infection. An open trial also concluded that early treatment of varicella in these patients is essential. Of the 94 patients with zoster infection, 52 received acyclovir (500 mg/m2/dose infused over one hour three times daily for 7 days). Acyclovir recipients healed more rapidly, had fewer days of pain and shorter duration of viral shedding compared with placebo patients. The most important finding was that acyclovir significantly protected against progression of zoster as defined by development or progression of cutaneous dissemination and development of visceral zoster. Vidarabine seemed to be equally effective in this respect. The likelihood of cutaneous dissemination is related to the nature of the underlying condition. The in vitro sensitivity of VZV isolates from patients with second episode VZV infection during the trial did not change appreciably which suggests that VZV does not become resistant to acyclovir during therapy.