Table 6 Pharmacological therapies for sleep problems in ASD

Posey et al. [91]

Subjects with neurodevelopmental disorder (PDDs), (N = 26, M = 21, F = 5, range = 3.8–23.5 years, mean age = 10.1 ± 4.8 years, 20 of them with ASD, 1 with Asperger’s disease, 1 with Rett’s disorder, 4 with PDDs not specified)

Treatment with mirtazapine (dose range = 7.5–45 mg daily, mean = 30.3 ± 12.6 mg daily)

Mirtazapine did not improve core symptoms of social or communication impairment. Adverse effects were minimal and included increased appetite, irritability, and transient sedation

Limitations: lack of a control group, disproportion between male and female number in group, small sample size

Thirumalai et al. [84]

ASD patients (N = 11, range 3–9 years, M = 9, F = 2, mean age = 5.09 years)

Polysomnography, EEG, EMG

REM sleep behavior disorder was identified in 5 of these 11 patients. Since REM sleep behavior disorder typically affects elderly males with neurodegenerative diseases, the identification of this phenomenon in autistic children could have profound implications for our understanding of the neurochemical and neurophysiologic bases of autism. Accurate diagnosis of REM sleep behavior disorder would enable specific treatment with clonazepam and help the family and the child

Limitations: small sample size, absence of a control group, disproportion between males and females

Ingrassia et al. [89]

Children (N = 6, 3 ADHD and 3 with mental retardation, range = 6–14 years, mean age = 11.2 years)

Clonidine administration (range dose = 50–100 mcg daily)

All children showed maintained improvements in their sleep pattern following the use of clonidine with only mild side-effects reported

Limitations: small sample size, lack of a control group, case series

Dosman et al. [94]

ASD children (N = 33, M = 27, F = 6, mean age = 6 years and 6 moths, range = 2 year 8 months–10 year 8 months)

Questionnaires (Sleep Disturbance Scale for Children, movements during sleep scale of Chervin and Hedger, Food records) made by parents after iron supplementation

High prevalence of restless sleep, which improved with oral iron supplementation, suggests that sleep disturbance may be related to iron deficiency in autism

Limitations: absence of a control group, disproportion between males and females

Rossignol et al. [36]

Initial studies review: 35 reviewed independently by two reviewers

Five of them were investigated through meta-analysis

Database used: PubMed, Google Scholar, CINAHL, EMBASE, Scopus, ERIC

Quality of studies was assessed through Downs and Black checklist

Six studies reported improvements in daytime behavior using melatonin; 18 studies on melatonin treatment in ASD reported improvements in sleep duration, sleep onset latency, night-time awakenings

From the meta-analysis: improvements in sleep duration but not in night-time awakenings

Strengths: the meta-analysis increases the statistical significance, funner plot didn’t indicate publication bias

Limitations: small sample size, protocol which measured changes in sleep parameters were variable

Buckley et al. [81]

ASD subjects (N = 5, range = 2.5–6.9 years) compared with within-lab controls

Polysomnography for REM sleep augmentation after donepezil administration

REM sleep as a percentage of Total Sleep Time was increased significantly and REM latency was decreased significantly after drug administration in all subjects

Limitations: open-label study without controls; very small sample size

Malow et al. [73]

ASD children (N = 24, range = 3–9 years, mean age = 5.9 years)

Melatonin supplementation, Actigraphy, Children’s Sleep Habits Questionnaire (CSHQ), Child Behavior Checklist (CBCL) scale

Supplemental melatonin improved sleep latency, as measured by actigraphy, in most children at 1 or 3 mg dosages. It was effective in week 1 of treatment, maintained effects over several months, was well tolerated and safe, and showed improvement in sleep, behavior, and parenting stress

Limitations: absence of a control group, small sample size

Mendez et al. [85]

ASD people (N = 3, range = 34–43 years, mean age = 39.33 years), controls (N = 3, range = 37–40 years, mean age = 38.66 years)

PET with receptor PET ligand [11C] Ro15-4513 was used to measure a1 and a5 subtypes of the GABA-A receptor levels

Lower [11C]Ro15-4513 binding was found throughout the brain of participants with ASD compared with controls. Planned region of interest analyses also revealed significant reductions in two limbic brain regions, namely the amygdala and nucleus accumbens bilaterally, thus suggesting a GABA-A a5 deficit in ASD

Limitations: very small number of participants

Strengths: controlled study, accurate technique of investigation (PET)

Maras et al. [74]

Initial participants: Children (N = 125, 96,8% of them ASD, 3,2% with Smith-Magenis syndrome, range = 2–17.5 years)

Final number of participants: N = 95, 51 of them received PedPRM, 44 placebo

Administration of 2, 5, or 10 mg PedPRM; Measures were: CSDI, PSQI, ESS, quality of life WHO-5 Well-Being Index

PedPRM, an easily swallowed formulation shown to be efficacious versus placebo, is an efficacious and safe option for long-term treatment (up to 52 weeks reported here) of children with ASD and NGD who suffer from insomnia and subsequently improves caregivers' quality of life

Limitations: open-label design of the study; lack of a control group made by healthy individuals; some individuals discontinuated treatment

Strengths: presence of a group receiving placebo

Ballester et al. [80]

ASD people (N = 23, M = 83%, mean age = 35 ± 12 years)

Administration of agomelatine or placebo

Agomelatine was effective and well tolerated for treating insomnia and circadian rhythm sleep problems present in adults with ASD and ID

Limitations: small sample size; absence of a control group

Strengths: placebo-controlled study

Gabis et al. [83]

ASD children (N = 60, range = 5–16 years, mean age = 9.5 ± 3.22 years)

AChE inhibitors and choline supplements in children and adolescents with ASD

Combined treatment of donepezil hydrochloride with choline supplement demonstrates a sustainable effect on receptive language skills in children with ASD for 6 months after treatment, with a more significant effect in those under the age of 10 years

Limitations: safety concerns limited the dose and the compounds used in the study; small sample size; two different language tests were used to assess global language skill; absence of a control group