Summary of studies assessing the effects of trazodone (TRZ) on sleep

StudyAgeDiagnosisDesign + Number of ParticipantsResultsAdverse EventsConclusion
Camargos et al. (2014)60+Alzheimer’s disease2-wk RCT. TRZ 50 mg/day vs. placebo. N = 30.Actigraphy: TRZ subjects slept 42.5 min more than placebo subjects and percent sleep increased 8.5%. No effect on cognition or WASO or number of awakeningsAEs transient and mild.Trazodone may be effective at reducing symptoms of insomnia in older adults with Alzheimer’s disease.
Cunningham et al. (1994)18+Major depression6-wk RCT of trazodone vs. venlafaxine vs. placebo. 96 responders continued in a 1-yr double-blind continuation phase. Average doses: 156–160 mg/day venlafaxine and 294–300 mg/day TRZ. N = 225.TRZ produced more improvement on the sleep disturbance factor of the HAM-D, but venlafaxine more effective in cognitive disturbance and retardation factors of the HAM-D.Significantly more patients discontinued TRZ due to dizziness than patients taking placebo or venlafaxine. Significantly more patients discontinued venlafaxine due to nausea.TRZ is effective at reducing symptoms of insomnia in patients with MDD.
Friedmann et al. (2008)16–65DSM-IV current alcohol dependence and sleep disturbance12-wk RCT. TRZ 50–150 mg/day vs. placebo. N = 173.TRZ group experienced less improvement in proportion of days abstinent during administration of medication, and an increase in the number of drinks per drinking day on cessation of study. TRZ improved sleep quality (PSQI mean change −3.02 [−3.38, −2.67]), but after cessation sleep quality equalized with placebo.TRZ group reported more dry mouth.Although TRZ reduced symptoms of insomnia, it may impede improvements in alcoholism during detoxification and lead to increased drinking when stopped.
Haffmans and Vos (1999)mean: 44Previous severe major depression; current insomnia secondary to brofaromine treatmentCrossover RCT. TRZ 50 mg/day as adjunct to brofaromine 150–250 mg/day. N = 7.TRZ did not improve SOL, total sleep time, or time awake. TRZ reduced the number of awakenings (P = 0.019) and increased stage IV sleep (P = 0.088).While taking TRZ, one patient reported constipation and nausea, one patient reported vertigo, dry mouth, and palpitations, and one patient had tingling feelings in chin and moderate heartburn.TRZ may not be effective at reducing symptoms of insomnia secondary to treatment with stimulating antidepressants, but may increase slow-wave sleep.
James and Mendelson (2004)Varied.Varied.Review.TRZ increases total sleep time in patients with MDD, though there are few data to support its use in non-depressed patients.There is a notable risk of priapism in 1 out of 6000 patients treated with TRZ. There is also a risk of orthostatic hypotension and the induction of cardiac arrhythmias in patients with preexisting cardiac disease.TRZ has more side effects than conventional hypnotics, making its risk-benefit ratio uncertain.
Karam-Hage and Brower (2003)see Gabapentin section
Kaynak et al. (2004)20–50Insomnia secondary to treatment w/ SSRI, DSM-IV depression2-wk crossover RCT w/ 1-wk washout. TRZ 100 mg/day + SSRIs vs. placebo + SSRIs. N = 12.TRZ significantly increased total sleep time, percentage of stage 3 and 4 sleep, SE, sleep continuity and decreased number of awakenings. At week 3, completing both TRZ and placebo treatments w/ washout, mean global PSQI scores in both groups reduced from 15 ± 2.5 to 5 ± 1.6. Total polysomnographic sleep time (minute) increased from 382.17 ± 58 to 428 ± 39 on last TRZ night, P < 0.05. SOL not significantly improved.TRZ period associated w/ one subject reporting mild acid indigestion and two reporting mild daytime sedation.TRZ may be effective at reducing symptoms of SSRI-induced insomnia.
Le Bon et al. (2003)18–65Alcohol dependence with physiologic dependence by DSM-IV; alcohol-induced sleep disorders, insomnia type (DSM-IV)4-wk RCT. TRZ 50 mg/day vs. placebo. N = 16.SE was increased in TRZ group. No benefit in placebo group. HAM-D and CGI were also better in TRZ group.Most frequent AEs in the TRZ group were hangovers and dizziness. Dose reduction from 200 to 150 mg/day reduced these AEs. In placebo group: AEs included headaches, hangover, and skin irritation.TRZ may be effective at reducing symptoms of insomnia.
Mendelson (2005)Varied.Varied.Review of 18 studies.Limited evidence of efficacy of TRZ, many trials were small and were conducted in depressed patients.High rate of discontinuation due to side effects including sedation, dizziness, and psychomotor impairment. This raises concerns about the use of TRZ in the elderly.TRZ risk-benefit ratio in insomnia remains uncertain.
Mouret et al. (1988)35–60depressed in-patients with a MADRS score >20Open-label study with TRZ (400–600 mg/day) for 5 wk. N = 10.TRZ decreased latency to sleep onset and intrasleep awakenings, increased TST and did not affect REM sleep time. An increase in REM sleep latency was found only at the end of the 5-wk trial.Nausea, dizziness.TRZ is effective at reducing symptoms of insomnia in depressed patients.
Nierenberg et al. (1994)mean:41.9 S.D.: 10Antidepressant-associated insomnia in patients treated with fluoxetine or bupropionCrossover RCT. TRZ 50 mg/day vs. placebo. Mean length of treatment: 6.5 days for TRZ vs. 4.6 days for placebo. N = 17.67% reported improvement with TRZ vs. 13% with placebo. Improvement with TRZ and not placebo on PSQI and Yale-New Haven scores.Daytime sedation occurred in one patient. Priapism did not occur, but one patient developed a prolonged erection, prompting a dose decrease.TRZ may be effective at reducing symptoms of insomnia.
Scharf and Sachais (1990)Depressed patients with sleep disturbances8-wk single-blind study with TRZ 150–400 mg/day. N = 6.After 5 wk, TRD improved sleep efficiency (from 80.6 ± 12.3% to 91. 9 ± 4.9%), increased TST (from 387.1 ± 59.2 to 441.3 ± 23.7 min), prolonged REM sleep latency but did not affect the amount of REM sleep. TRZ enhanced sleep-related penile tumescence.Not analyzed.TRZ affects REM sleep in young men and reduced awakenings and movement/arousals probably due to its sedating properties.
Stein et al. (2012)mean: 38.2 S.D.: 8.6Methadone maintenance treatment of opioid dependence6-mo RCT. TRZ 50–150 mg/day vs. placebo. N = 137.Placebo subjects reported significantly higher sleep quality ratings than TRZ subjects, P = 0.04. Polysomnography: total sleep time was not significantly improved in TRZ subjects, P = 0.18.Between baseline and 1 mo, TRZ group significantly more likely to report increased thirst or dry mouth, P = 0.001, and decreased appetite, P = 0.04.Trazodone is not effective at reducing symptoms of insomnia in patients on methadone maintenance treatment with sleep disturbance.
Walsh et al. (1998)21–65Primary insomnia by DSM-IIIR2-wk parallel-group comparison RCT. TRZ 50 mg/day vs. zolpidem 10 mg/day vs. placebo with 1-wk placebo lead-in. N = 306.During first week, both drugs produced significantly shorter self-reported SOL (SSL) and longer self-reported sleep duration (SSD) than placebo. SSL was significantly shorter with zolpidem than with TRZ. During Week 2, only the zolpidem group maintained a significantly shorter SOL than the placebo group, and SSD did not vary significantly among groups.Treatment-emergent AEs reported by 65.4% of placebo patients, 76.5% of zolpidem patients, and 75% of TRZ patients. Headache occurred, respectively, in 19%, 24%, and 30% of participants. Somnolence occurred, respectively, in 8%, 16%, and 23% of participants. Treatment generally well-tolerated.TRZ may be less effective than zolpidem at doses studied at reducing symptoms of insomnia.
Ware et al. (1994)20–34Healthy malesCrossover RCT. TRZ 100–200 mg/day vs. nefazodone 200–400 mg/day, vs. buspirone 10–20 mg/day, and vs. placebo. N = 12.TRZ and buspirone reduced REM sleep amount and increased REM sleep latency whereas nefazodone enhanced REM sleep without affecting REM sleep latency. TRZ The drugs did not affect NREM sleep. TRZ reduced the number of awakenings and the number of movement/arousals compared with placeboNot analyzed.TRZ suppress REM sleep and prolongs penile tumescence during sleep.
Yamadera et al. (1998)21–28Healthy malesSingle-blind study. TRZ 100 mg/day vs. imipramine 40 mg/day vs. placebo. N = 8.TRZ increased stage N3 and decreased stage N1-N2 vs. placebo, and unlikely imipramine, did not alter REM sleep.Not analyzed.TRZ increased NREM sleep without affecting REM sleep in young healthy males.