It was previously reported that the Korean predictive model could be used to identify patients at high risk of postoperative nausea and vomiting (PONV). This study investigated whether PONV in the high-risk and very high-risk patients identified by the Korean predictive model could be prevented by multiple prophylactic antiemetics. A total of 2,456 patients were selected from our previous PONV study and assigned to the control group, and 374 new patients were recruited consecutively to the treatment group. Patients in each group were subdivided into two risk groups according to the Korean predictive model: high-risk group and very high-risk group. Patients in the treatment group received an antiemetic combination of dexamethasone 5 mg (minutes after induction) and ondansetron 4 mg (30 min before the end of surgery). The incidences of PONV were examined at two hours after the surgery in the postanesthetic care unit and, additionally, at 24 hr after the surgery in the ward, and were analyzed for any differences between the control and treatment groups. The overall incidence of PONV decreased significantly from 52.1% to 23.0% (p< or =0.001) after antiemetic prophylaxis. Specifically, the incidence decreased from 47.3% to 19.4% (p< or =0.001) in the high-risk group and from 61.3% to 28.3% (p< or =0.001) in the very high-risk group. Both groups showed a similar degree of relative risk reductions: 59.0% vs. 53.8% in the high-risk and very high-risk groups, respectively. The results of our study showed that the antiemetic prophylaxis with the combination of dexamethasone and ondansetron was effective in reducing the occurrence of PONV in both high-risk and very high-risk patients.