International Meeting for Autism Research: Supplemental Melatonin Decreases Time to Fall Asleep IN Children with AUTISM

Supplemental Melatonin Decreases Time to Fall Asleep IN Children with AUTISM

Saturday, May 14, 2011: 2:30 PM
Elizabeth Ballroom GH (Manchester Grand Hyatt)
1:15 PM
B. A. Malow1, S. G. McGrew2, C. P. Burnette3, K. Adkins1, S. E. Goldman1, D. Wofford1, K. L. Surdyka1, D. Fawkes1 and L. Wang4, (1)Neurology/Sleep, Vanderbilt Medical Center, Nashville, TN, (2)Pediatrics, Monroe Carell Children's Hospital at Vanderbilt, Nashville, TN, (3)Department of Pediatrics, University of New Mexico, Albuquerque, NM, (4)Biostatistics, Vanderbilt University, Nashville, TN
Background:  Retrospective and small prospective studies have shown that supplemental melatonin promotes sleep in children with autism spectrum disorders (ASD). However, most studies have not used objective measures of sleep or have limited their samples to subsets of ASD (e.g., Asperger syndrome).

Objectives:  In preparation for a large multicenter randomized controlled trial, we carried out a pilot study of supplemental melatonin in children with ASD using actigraphy to measure changes in sleep parameters.

Methods:  Children were ages 3-10 years, with a clinical diagnosis of ASD based on DSM-IV-TR criteria.  Diagnosis was confirmed by the Autism Diagnostic Observation Schedule and by either clinical interview or the Autism Diagnostic Interview-Revised. All children had sleep delay, defined as inability to fall asleep within 30 minutes at least 3 nights a week. Medical causes of insomnia and primary sleep disorders were addressed prior to enrollment.  Children on medications which affect melatonin pharmacokinetics were excluded. Parents were instructed in optimizing sleep habits and bedtime routines prior to treatment. Objective measures of sleep were measured over 17 weeks with actigraphy (Phillips/Respironics/Mini Mitter) in conjunction with sleep diaries. All children received two weeks of an inert flavored liquid 30 minutes before bedtime, to acclimate them taking a liquid medication at bedtime. At week 3, children began 1 mg of supplemental melatonin (Natrol ®).  The child’s response to melatonin was reevaluated every three weeks based on mean sleep latency (SL), defined as time to fall asleep from lights out.  Dose was increased every three weeks (from 1 mg, to 3 mg, to 6 mg, to 9 mg) until the child reached a satisfactory response (SAT-R), defined as SL within 30 minutes on 5 or more nights in the week. Once a SAT-R was achieved, the child remained on that melatonin dose for the remainder of the 17 weeks.

Results: Twenty-one children were included, two girls and 19 boys, with a mean age of 6 years [standard deviation (SD) = 2.2]. A SAT-R was reached in five children at 1 mg, 14 children at 3 mg, and two children at 6 mg. Mean sleep latency improved with treatment (EOS) compared to baseline; Wilcoxon-signed rank test. Improvements were also noted in sleep efficiency and total sleep time. Fragmentation index, a measure of movement/restless sleep, did not improve. Actigraphy parameters did not change from the time a SAT-R was achieved as compared to the end of study. There were no adverse effects related to melatonin.

Actigraphy Parameter At Baseline At SAT-R At  EOS p-value
Mean (SD) Mean (SD) Mean (SD) EOS vs. baseline
Sleep Latency (min) 41.0 (33.9) 25.3 (26.4) 24.8 (21.6) <0.0001
Sleep Efficiency (sleep time/time in bed) 73.9 (9) 77.1 (9.2) 78.3 (7.5) 0.013
Total Sleep Time (min) 430.4 (73.1) 461.5 (77.2) 460.5 (66.5) 0.029
Fragmentation Index 38.7 (14.4) 37.4 (14.1) 34.4 (13.8) 0.88


In this pilot work, we documented, with actigraphy, that supplemental melatonin decreases sleep latency and improves other sleep parameters, with stability over several weeks.  Randomized clinical trials of supplemental melatonin appear warranted.

| More