Ondansetron was compared with placebo for nausea and vomiting prophylaxis after fentanyl/isoflurane/relaxant anesthesia and infratentorial craniotomy. Eight milligrams intravenous ondansetron or vehicle was administered at skin closure. Nausea, emesis, and antiemetic use were recorded at 0, 0.5, 1, 4, 8, 12, 24, and 48 hours. There were no significant intergroup differences for nausea incidence at any interval, but cumulatively the placebo group was 3.2 times more likely to develop nausea during the first 12 hours (P = .04). Nausea incidence was bimodal in both groups, peaking during the first 1 to 4 hours. A nadir occurred at 8 to 12 hours, but nausea increased during the next 36 hours. By 48 hours, approximately 40% of patients in both groups were still nauseated. Reduced vomiting frequency was seen with ondansetron at 4, 8, 12, and 24 hours (P < .05). Despite rescue antiemetics, emesis occurred in an irregular pattern with episodes still observed in 35% of placebo patients at 48 hours. For ondansetron, emesis was infrequent for the first 12 hours but then a persistent increase was observed (48 hours, 22%). The incidence of rescue antiemetic use was 65% for both groups. There was no effect of gender. Nausea and vomiting are frequent and protracted after infratentorial craniotomy. Administration of single-dose ondansetron (8 mg intravenously) at wound closure was partially effective in reducing acute nausea and vomiting but had little delayed benefit. Scheduled prophylactic administration of antiemetic therapy during the first 48 hours after infratentorial craniotomy should be evaluated for efficacy and safety.