Comparison of therapeutic modalities for mania.

The results of sequential therapeutic trials in hospitalized manic patients conducted over a 16-year period are summarized, followed by an analysis of pooled data to assess relative efficacy. No clinically important baseline differences were found in the patients admitted to these studies despite the long time span. They were not overly "difficult to treat" or treatment resistant, and most were discharged to the community. Nearly all outcome measures showed statistically significant differences between groups after 8 weeks of treatment. The best responses occurred in the patients who received a mean series of nine electroconvulsive therapy (ECT) treatments with sparing use of neuroleptics followed by lithium maintenance. The next best outcome was observed with lithium combined with low doses of standard neuroleptics or risperidone. The combination of carbamazepine and lithium had significantly fewer neurological side effects than moderate doses of haloperidol with lithium, with equivalent therapeutic results. Monotherapy with either lithium or carbamazepine was less effective than the combination treatments. Minor differences in study design may contribute to the variance in outcome.