Bipolar pharmacotherapy and suicidal behavior. Part I: Lithium, divalproex and carbamazepine.

INTRODUCTION: The anti-suicidal benefit of lithium on suicidal behavior in bipolar disorder is well-established. Data are mixed on the effects of divalproex and carbamazepine.

METHODS: Retrospective chart review study of 405 veterans with bipolar disorder followed for a mean of 3 years, with month by month review of clinical progress notes, and systematic assessment of current pharmacotherapy and suicide completion, attempt or hospitalization for suicidality. Comparison of suicide event rates (events/100 patient years) between mood stabilizers and during-vs-after discontinuation of mood stabilizers, with linear regression analysis for influence of potential confounding variables, and robust bootstrap confirmation analysis.

RESULTS: No completed suicides occurred during or after discontinuation of monotherapy. Rates of non-lethal suicidal behavior were similar during lithium (2.49), divalproex (4.67) and carbamazepine (3.80) monotherapies. There was a sixteen fold greater, highly statistically significant non-lethal suicidal event rate after discontinuation compared with during mood stabilizer monotherapy (55.89 vs. 3.48 events/100 patient years; Chi2=13.95; df=1; p<0.0002). On compared with off treatment differences were similar for the three different agents.

LIMITATIONS: Treatments were uncontrolled in this naturalistic setting, and data were analyzed retrospectively.

CONCLUSIONS: Lithium and the anticonvulsants may show similar benefits in protecting bipolar patients from non-lethal suicidal behavior when careful analysis of clinical data is done to confirm medication adherence/non-adherence. Findings in this study were similar to those of a previous study that applied the same methodology in a private practice setting.