In a 12-week randomized, placebo-controlled study evaluating the efficacy and safety of adjunctive asenapine, bipolar I disorder patients experiencing manic or mixed episodes despite pretreatment with lithium or valproate monotherapy were treated with flexible-dose, twice-daily asenapine 5 or 10 mg (n = 158) or placebo (n = 166). The primary efficacy end point was change from baseline Young Mania Rating Scale (YMRS) total score at week 3. Secondary outcomes included YMRS response and remission and Clinical Global Impression for Bipolar Disorder and Montgomery-Asberg Depression Rating Scale score changes. Patients completing the core study were eligible for a 40-week double-blind extension assessing safety and tolerability. Adjunctive asenapine significantly improved mania versus placebo at week 3 (primary end point) and weeks 2 to 12. The YMRS response rates were similar at week 3 but significantly better with asenapine at week 12. The YMRS remission rates and changes from baseline on Clinical Global Impression for Bipolar Disorder for mania and overall illness were significantly better with asenapine at weeks 3 and 12. No other statistically significant differences on secondary outcomes were observed. Only a small number of patients entered the extension, making firm statistical conclusions on efficacy difficult. Treatment-emergent adverse events reported by 5% or more of asenapine patients and at twice the incidence of placebo were sedation, somnolence, depression/depressive symptoms, oral hypoesthesia, and increased weight in the 12-week core study. Adjunctive asenapine to lithium or valproate was more effective than mood stabilizer monotherapy in the core study and was well tolerated for up to 52 weeks.