Background: Bipolar disorder is a recurrent, chronic illness that is one of the most notable causes of disability in persons 15 to 44 years of age. The traditional maintenance therapy, lithium, has a narrow therapeutic index and significant adverse effects, leading to problems in adherence to therapy. Many alternative therapies have been proposed, particularly anticonvulsant and second-generation antipsychotic agents. Many patients do not respond to monotherapy, leading to the widespread use of combination treatments. The Bipolar Affective Disorder: Lithium/Anticonvulsant Evaluation study compared combination therapy of lithium plus valproate (Depacon) with monotherapy using either drug alone to prevent relapse of bipolar disorder.

The Study: Patients with a clinical diagnosis of bipolar I disorder were enrolled at 41 sites in Europe and the United States if the combination of lithium plus valproate was considered reasonable, if there were no contraindications to any of the study medications, and if the patient was judged to be likely to adhere to the assigned therapy. During the four- to eight-week run-in period, participants received therapy with lithium and valproate titrated to a target dosage of 1,250 mg per day of valproate (or the highest dosage tolerated) and serum lithium levels of 0.4 to 1.0 mmol per L. After the run-in period, participants were randomly assigned to one of three treatment groups: lithium monotherapy, valproate monotherapy, or continued combination therapy. In patients assigned to monotherapy, the discontinued drug was withdrawn during a four-week period to reduce the risk of precipitating relapse. Participants were maintained on the assigned therapy for two years or until treatment was proven ineffective. Investigators and participants were aware of treatment allocation, and additional therapies could be continued during the study.

The primary outcome was time to intervention for a new mood episode or hospitalization. Additional outcome measures included global assessment of functioning, quality of life determined by a questionnaire, and adverse effects. Analysis was by intention to treat.

Results: A total of 110 participants were assigned to each of the three treatment groups and were comparable in all important variables. Approximately 22 participants were lost to follow-up in each group. The hazard for the primary outcome was significantly lower for patients in the combination therapy group than for those allocated to valproate monotherapy, but was similar to patients allocated to lithium monotherapy. Similarly, the hazard of the primary outcome was lower for participants treated with lithium than for those receiving valproate. This benefit was most apparent in prevention of manic episodes and was maintained for at least two years. These results persisted after adjustment for relevant factors. Serious adverse events affected five participants during the run-in period and 16 participants after randomization. No suicides occurred during the study. No obstetric complications or congenital abnormalities were noted in the five women who became pregnant during the study.

Conclusion: The authors conclude that lithium and valproate combination therapy is more likely than valproate monotherapy to prevent relapse of bipolar I disorder. Lithium monotherapy also seems more effective than valproate monotherapy. In this study, no significant differences were detected between combination therapy and lithium monotherapy. The study authors acknowledge several limitations and suggest that current guidelines recommending valproate as first-line long-term therapy might be improved by changing that recommendation to combination therapy or lithium monotherapy.