Summary of studies assessing the effects of gabapentin (GBP) on sleep

StudyAgeDiagnosisDesign + Number of ParticipantsResultsAdverse EventsConclusion
Brower et al. (2008)18+Alcohol dependence with comorbid insomnia6-wk RCT plus 2-wk placebo lead-in and follow up visit after 6 wk. GBP 1500 mg/day vs. placebo. N = 21.60% of GBP group relapsed to heavy drinking by 12 wk vs. 100% of the placebo group, P = 0.04. Treatment group did not predict changes in Sleep Problems Questionnaire (SPQ) score. There were no significant polysomnographic differences between groups.The most common side effects attributed to gabapentin vs. placebo, respectively, were somnolence (three subjects vs. one subject), headache (three subjects in each group), dizziness (two subjects vs. one subject), indigestion (two vs. four subjects), nerve or muscle pain (two subjects vs. none) .GBP is not effective at reducing symptoms of insomnia, though it delayed the onset to heavy drinking.
Foldvary-Schaefer et al. (2002)20–46Healthy participantsRandomized study. GBP titrated to 1800 mg/day vs. untreated group. N = 19.GBP-treated subjects had an increase in SWS compared with baseline. No difference in other polygraphic variables. GBP subjects had minor reductions in arousals, awakenings, and stage shifts was observed in treated subjects.One patient experienced dizziness at highest dose and was treated with 1500 mg/day.GBP may be effective at increasing SWS without affecting other aspects of sleep.
Karam-Hage and Brower (2003)mean: 44 S.D.: 14DSM-IV alcohol dependence with persistent insomnia4–6-wk open-label study. GBP 300–1800 mg/day vs. trazodone 25–300 mg/day. N = 55.Both groups showed significant improvement in sleep from baseline to follow up, P < 0.001 for each group as measured by the SPQ. Total change in SPQ scores between the groups: GBP group improved by 8.8 ± 4.0 while trazadone group improved 6.1 ± 3.4, P = 0.023.Both medications were well-tolerated.GBP is effective at reducing symptoms of insomnia in alcoholism and significantly more effective than trazodone.
Lo et al. (2010)mean: 43.2 S.D.: 15.4Primary insomnia4-wk open-label study. GBP 200–900 mg/day. N = 18.By polysomnography: GBP did not significantly improve SOL, which decreased from 17.58 to 14.58 min. GBP improved SE from 80.00% to 87.17%, P < 0.05. WASO was significantly reduced from 16.45% to 7.84%, P < 0.05. Sleep stage N3 increased from 10.47% to 17.68%, P < 0.005.Prolactin levels were significantly reduced after GBP treatment; whether the effect occurred as a result of GBP was uncertain.GBP is effective at reducing symptoms of insomnia and increases slow-wave sleep.
Robinson and Malow (2013)Pediatric: mean 7.2Refractory insomnia in children with neurodevelopmental or neuropsychiatric disordersOpen-label study. GBP 6–15 mg/kg per day. N = 23.Improved sleep was noted in 78% of children.AEs noted in six children, mostly agitation and worsened ability to sleep.GBP is effective at reducing symptoms of insomnia in children with neurodevelopmental and neuropsychiatric disorders.
Rosenberg et al. (2014)18 and olderOccasional disturbed sleepRCT using a 5-h phase advance insomnia model. GBP 250 mg vs. GBP 500 mg vs. placebo. N = 377.LPS was not significantly different among groups. Mean total sleep time was significantly greater for the gabapentin groups: 311.4 ± 8.4 min in placebo group, 356.5 ± 7.5 min in GBP 250 mg, and 378.8 min in GBP 500 mg group. Percent slow-wave sleep was significantly greater in GBP groups, 12.6%, 15.4%, and 17.0%, respectively.4.2% of participants reported at least one AE, with 1.6% of placebo participants, 4.0% of GBP 250 mg/day participants, and 7.2% GBP 500 mg/day. The most common AE was headache.GBP is effective at reducing symptoms of insomnia and increases slow-wave sleep.