Reduced REM latency predicts response to tricyclic medication in depressed outpatients.

Forty-two outpatients with major depressive disorder entered a double-blind, randomized trial of either desipramine or amitriptyline for a minimum of 6 weeks. Pretreatment polysomnographic and clinical measures were used to predict response. Response was defined as a 17-item Hamilton Rating Scale for Depression score less than or equal to 9 at the end of treatment. There was a 61.1% response rate for patients treated with amitriptyline and a 66.7% response rate for patients treated with desipramine. Reduced REM latency (2-night mean less than or equal to 65.0 min) predicted a positive response to these tricyclic antidepressants. REM latency did not differentiate between desipramine or amitriptyline responders. More patients with reduced REM latency (80%) responded to treatment compared with patients with nonreduced REM latency (50%). The 80% response rate in reduced REM latency depressed patients confirms our previous findings in a mixed inpatient and outpatient sample. Contrary to our hypothesis, in this sample, endogenous depression was not associated with a good response to tricyclic medication.