In a double-blind study, 74 children were randomly treated with prophylactic methicillin at the time of ventriculo-peritoneal shunting. There were 7 infections in the placebo group and 2 in the methicillin-treated group. There was no statistically significant difference in infection or overall malfunction rate between the two groups. The risk of shunt malfunction in the 2nd to 6th month after surgery was significantly greater in the placebo group (7 of 26 patients) than in the methicillin-treated group (1 of 26 patients). Delayed shunt malfunction may be prevented by a short course of perioperative antibiotics.