Reduced morbidity after gradual discontinuation of lithium treatment for bipolar I and II disorders: a replication study

OBJECTIVE: The aim of this study was to verify reduction of early affective morbidity by gradual, rather than rapid, discontinuation of lithium treatment. METHOD: For 78 patients with bipolar disorders, lithium treatment was discontinued either rapidly (over 1-14 days) or gradually (over 15-30 days). The effects of the two schedules were compared by survival analysis of time to first recurrence. RESULTS: Median time to recurrence was 5.6 times as long for gradual discontinuation (14.0 months) as for rapid discontinuation (2.5 months). The ratios of the median survival times for gradual and rapid discontinuation were similar in I and II subtypes and in depression and mania (4-6:1). The polarities of the episodes at onset and at first recurrence after lithium discontinuation were 83.6% concordant. CONCLUSIONS: These results independently confirm a reduction in morbid risk from slow discontinuation of lithium treatment for bipolar disorders.