TABLE 8 

Summary of studies assessing the effects of amitryptiline (AMT) on sleep

StudyAgeDiagnosisDesign + Number of ParticipantsResultsAdverse EventsConclusion
Hartmann and Cravens (1973)21–35Healthy volunteers1125-night crossover study. AMT 50 mg/day vs. placebo vs. reserpine 0.50 mg/day vs. chlorpromazine 50 mg/day vs. chloral hydrate 500 mg/day vs. chlordiazepoxide 50 mg/day. N = 14.TST increased for AMT relative to placebo throughout trial. SWS increased and SOL decreased early during administration. REMS (“desynchronized sleep”) was reduced throughout the trial, most profoundly in the first few days of administration.Subjects themselves as feeling worse in the morning during AMT treatment. Reports of feeling sick increased for weeks 3 and 4 on and weeks 1 and 2 off medication. Reports of unusual psychologic feelings increased in week 4 on medication.AMT is probably effective at reducing symptoms of insomnia.
Mendlewicz et al. (1991)25–68Depressed inpatientsRetrospective study before treatment with AMT and at 5-7 wk follow up. AMT (165 ± 35 mg/day). N = 18.Marked suppression of REM sleep. Increase in stage 1 and stage 2 NREM sleep and no changes in SWS and latency to sleep onset.Not assessed.AMT induces a large REM sleep suppression in depressed inpatients.
Srisurapanont and Jarusuraisin (1998)18–65Opiate withdrawal insomnia6-day randomized, double-blind trial. AMT 25–100 mg/day vs. lorazepam 1–4 mg/day. N = 27.Ease of getting to sleep questionnaire scores on the SEQ were 177.4 ± 53.9 for AMT and 184.5 ± 85.4 for lorazepam, P = 0.80. Ease of awakening from sleep on the SEQ was 132.8 ± 47.9 for AMT and 167.6 ± 38.4 for lorazepam, P = 0.047.None of the patients reported any serious AEs during the study. Awakening from sleep after AMT was more difficult than after lorazepam.AMT is as effective as lorazepam at reducing symptoms of insomnia associated with opiate withdrawal syndrome, though it may be associated with a hangover effect.