BACKGROUND: Few double-blind trials have examined the efficacy of a combination of a mood stabiliser and an atypical antipsychotic in acute mania. AIMS: To determine the efficacy of risperidone in combination with a mood stabiliser in acute mania. METHOD: Patients taking a mood stabiliser were randomised to 3 weeks' treatment with risperidone (n=75) or placebo (n=76). RESULTS: Young Mania Rating Scale (YMRS) scores improved rapidly with significantly greater reductions at week 1 in the risperidone group compared with the placebo group. At end-point YMRS scores decreased by 14.5 and 10.3 points in the risperidone and placebo groups, respectively. Significant improvements v. placebo (P<0.05) were noted in the risperidone group on several other clinically meaningful measures. Additionally, a post hoc analysis excluding carbamazepine-treated patients (plasma concentrations of risperidone active moiety were 40% lower in this group) revealed significantly greater reductions (P=0.047) in YMRS scores in the risperidone group than in the placebo group. Incidence of adverse events was similar in both groups. CONCLUSIONS: Risperidone is superior to placebo when used in combination with lithium or divalproex in acute mania.