Sleep and daytime psychomotor performance during acute and continuation treatment of major depressive disorder: double-blind randomised controlled trial of escitalopram vs paroxetine

Introduction. Disturbed sleep is a common depressive symptom which often persists despite otherwise successful pharmacological or psychological treatment, and is associated with increased risks of recurrence. As both worsened insomnia and daytime drowsiness are reported as treatment-emergent adverse effects with antidepressants, we wished to evaluate the effects of acute and continuation treatment on sleep and daytime psychomotor performance. Method. International, randomized, flexible-dose, parallel-group, comparator-controlled study involving 36 centres in 6 countries. Patients met DSM-IV criteria for current major depressive episode, with a baseline score of 22-40 on the Montgomery-Åsberg Depression Rating Scale (MADRS). After a single-blind one-week placebo run-in, patients were randomized to escitalopram (ESC) (10-20 mg/day) or paroxetine (PAR) (20-40 mg/day) for 8 weeks of acute treatment: those who were significantly improved could continue fixed-dose double-blind treatment for a further 19 weeks. Sleep and early morning behaviour were assessed by the Leeds Sleep Evaluation Questionnaire (LSEQ), psychomotor performance by Critical FIicker Fusion (CFF) and Choice Reaction Time, and cognitive function by the Cognitive Failures Questionnaire (CFQ). Assessments were undertaken at baseline and weeks 4, 8, 16 and 27. Results. 323 patients (ESC, 165; PAR, 158) started acute treatment and received at least one dose of study medication. There were no significant differences between treatment groups in reduction of depressive symptoms from baseline to week 8, but in severely depressed patients (baseline MADRS >30) escitalopram was superior (p<0.05; ANCOVA) to paroxetine at week 27. Sleep and daytime performance improved as depressive symptoms reduced. Two patients withdrew from the study due to the adverse effect of insomnia, and 1 due to somnolence. There were no significant differences between treatment groups in change from baseline on the LSEQ subscales for ‘getting to sleep’, ‘awakening from sleep’, or ‘behaviour following sleep’ although at week 4 ESC was superior (p<0.01; ANCOVA) to PAR on ‘quality of sleep’. There were no significant differences between treatment groups in the change from baseline in CFF, recognition or motor reaction times, or CFQ total score, at any of the scheduled assessments. Conclusion. In this study, only a small number of patients withdrew from antidepressant treatment due to adverse effects on sleep or daytime alertness. Repeated assessments of sleep and daytime performance indicated a steady improvement during acute and continuation treatment of major depressive disorder with both escitalopram and paroxetine. Source of funding. The overall randomised controlled trial was sponsored by Lundbeck Ltd