Risk for bipolar illness in patients initially hospitalized for unipolar depression.

OBJECTIVE: To assess the risk for subsequent development of mania or hypomania, the authors conducted a 15-year prospective follow-up study of a large, young cohort of patients originally hospitalized for unipolar major depression.
METHOD: Patients who were hospitalized for unipolar major depression (N=74; mean age=23.0 years, SD=3.8) were assessed prospectively as inpatients and then followed up five times over 15 years, at approximately 2, 5, 8, 11, and 15 years after discharge. Manic or hypomanic episodes, medications, and rehospitalizations were determined by standardized assessments at each follow-up. Polarity conversions were evaluated by survival analyses.
RESULTS: By the 15-year follow-up, 27% of the study group had developed one or more distinct periods of hypomania, while another 19% had at least one episode of full bipolar I mania. Depressed patients with psychosis at the index depressive episode were significantly more likely than nonpsychotic patients to demonstrate subsequent mania or hypomania at follow-up. Those with family histories of bipolar illness showed a nonsignificantly higher rate of switching to mania or hypomania. Spontaneous and antidepressant-associated manias did not differ in frequency. Fewer than one-half of the patients who showed an eventual bipolar course had received prescriptions for mood stabilizers in any follow-up year.
CONCLUSIONS: Young depressed inpatients with psychotic features may be at especially high risk for eventually developing mania. The probability for developing a bipolar spectrum disorder increases in linear fashion for patients at risk for polarity conversion during the first 10-15 years after an index depressive episode.