Recent reports have been increasingly critical of the effectiveness of lithium in the treatment of persons with bipolar disorder. Questions center around whether the drug has been “overvalued” in the past and whether long-term treatment with lithium has lost its effectiveness since the 1970s. Some of the perceived loss of efficacy may be associated with under-motivation, poor patient selection or non-compliance. In addition, the perception that lithium is dangerous or stigmatizing may also have limited its acceptance by physicians and patients. Baldessarini and Tondo reviewed the literature on lithium therapy published since 1970 to test the hypothesis that the drug has lost its clinical effectiveness. They also analyzed the experience of a large cohort of patients with bipolar disorder who were treated with lithium during the past three decades at a psychiatric research center.

The authors reviewed 11 controlled and 13 open long-term lithium treatment trials for bipolar or mixed major affective disorders that were conducted between 1970 and 1996. Patients with unipolar major depression were excluded. In addition to the review of studies, patients seen at a mood disorders center were also evaluated with several established psychiatric and diagnostic assessment methods.

Data from all 24 published trials were pooled, and no significant difference in response rates was evident between the controlled or open trials. The monthly recurrence rates of patients with manic depression who were not treated with lithium were similar throughout the 26-year period, averaging 18.7-fold higher than rates in patients who were treated with lithium. The monthly recurrence rates of patients treated with lithium did not rise during the study years but were higher from 1970 to 1981 compared with 1982 to 1996. A regression analysis of the 24 studies also revealed significant decreases in monthly recurrence rates of bipolar episodes during treatment over the 26-year time period.

The mood disorders clinic population included 360 patients with bipolar type I or type II disorder who had been in lithium maintenance treatment since 1970. Approximately 65 percent were women, and the mean age at entry into treatment was 37 ± 14 years. All patients were treated with lithium for a minimum of 12 months. The average number of recurrences of manic or depressive episodes was 0.81 ± 1.12 per year with therapy compared with 1.83 ± 2.14 episodes before therapy. The percentage of time that patients were reported to be “ill” was about 46 percent before treatment compared with 18 percent during lithium treatment. The frequency of new manic episodes or percentage of time ill during the three decades of the study did not differ significantly. The authors note that residual morbidity is expected during maintenance with lithium, and only 32.5 percent of the 360 patients enrolled in the study had no new episodes. However, 66 percent of patients showed a significant benefit to treatment with lithium, defined as at least a 50 percent or greater reduction in the percentage of time spent “ill.” Patients who were least responsive were those with psychotic features accompanying their bipolar disorder, patients who met criteria in the Diagnostic and Statistical Manual of Mental Disorders, 4th ed. (DSM-IV) for a substance abuse disorder, as well as patients who had rapid cycling.

The authors conclude from the review of published literature that recurrence rates have not risen among patients not undergoing lithium treatment over the past three decades. In addition, morbidity did not increase before and after lithium treatment in the large sample of patients in a stable clinical setting during the same era. The authors suggest that, based on their findings, abandoning the use of lithium maintenance therapy because of poor efficacy or toxicity is not justified. No other current medical therapy for bipolar disorder has such a history of long-term efficacy and has shown a substantial decrease in the risk of mortality.