Table 2.
Main significant findings, with p-value in brackets.
| First Author, Year, Country |
Macrostructural EEG Features | Microstructural EEG Features | Sleep Subjective Features | Other Findings |
|---|---|---|---|---|
| Elia, 2000, Italy [25] |
ASD vs. TD ↓ TIB (p < 0.01), TST (p < 0.02), SPT (p < 0.01) ASD vs. X-fragile ↓ SPT (p < 0.03), RL (p < 0.01), N1 (p < 0.05) |
|||
| Malow, 2006, USA [26] |
psASD vs. gsASD 1st night ↓ SE (p = 0.0091), REM% (p = 0.0226) ↑ SL (p < 0.0079), N3,N4 (p = 0.446) 2nd night ↓ TST (p= 0.3800) gsASD vs. TD 1st night ↓ TST (p= 0.5507) 2nd night ↓ TST (p= 0. 5483) |
|||
| Miano, 2007, Italy [27] |
ASD vs. TD ↓ TIB (p < 0.044), SPT (p < 0.007), RL (p < 0.02) |
ASD vs. TD ↓ CAP rate during N3,4 (p < 0.02) A1% (p < 0.0004) ↑ A2% (p < 0.006) A3% (p < 0.02) |
ASD vs. TD Sleep Questionnaire: ↑ Sleep less than 8 hours (p < 0.02) Latency to sleep > 30 min (p < 0.000001) Difficulty falling asleep at night (p < 0.002) Fluids or drugs to facilitate sleep (p < 0.00001) Hypnic jerks (p < 0.00001) Rhythmic movements while falling asleep (p < 0.00001) Poor sleep quality (p < 0.00001) More than two awakenings per night (p < 0.05) Waking up to drink or to eat at night (p < 0.015) Difficulty to fall asleep after awakenings (p < 0.00001) Parasomnias – bedwetting (p < 0.00001) Daytime somnolence (p < 0.03) Falling asleep at school (p < 0.02) ↓ Drinks stimulant beverages in the evening (p < 0.00001) |
|
| Bruni, 2007, Italy [28] |
No significant results | AS vs. TD ↑ A1% (η2 = 1.43; p < 0.009) ↓ A2% (η2 = −1.88; p < 0.003) AS vs. ASD ↑ CAP rate during N3,4 (η2 = 1.41; p < 0.02) A1% (η2 = 2.05; p < 0.001) |
AS Sleep Questionnaires: reluctant to go to bed (50%) need for light or TV in the bedroom (75%) difficulty getting to sleep at night (87%) falling asleep sweating (75%) nocturnal hyperkinesia(50%) feeling unrefreshed upon morning awakening (50%) difficulty in waking up in the morning (87%) daytime somnolence 87%) PDSS mean score 16.5 ± 3.4 |
AS Positive correlation between verbal IQ and: total CAP rate (r = 0.99) CAP rate in SWS (r = 0.95) global A1 index (r = 0.94) SWS A1 index (r = 0.76) Negative correlation between A2% and: FSIQ (r = −0.086) VIQ (r = −0.86) PIQ (r = −0.81) Positive correlation between CBCL total score and: cap rate (r = 0.76) A1 index (r = 0.88) Negative correlation between externalizing score and A3% (r= −0.81) |
| Goldman, 2009, USA [29] |
psASD vs. TD ↑ SL (p < 0.05) psASD vs. gsASD ↑ SL (p < 0.05) |
PCQ: poor sleepers rate among ASD: 64% psASD vs. gsASD CSHQ: ↑ sleep onset delay (p < 0.01), sleep duration (p < 0.01), night wakings and total (p < 0.01) psASD vs. TD ↑ for all dimensions except sleep disordered breathing |
||
| Ming, 2009, USA [30] |
ASD vs. TD ↓ REM% (p = 0.002) |
ASD Sleep Questionnaires: Parasomnias (60.8%), Disorder of Partial Arousal (55.6%) |
||
| Giannotti, 2010, Italy [31] |
NregASD vs. RegASD ↑ TST (p < 0.001), SE (p < 0.001) ↓ WASO (p < 0.001), SL (p < 0.001) RegASD vs. TD ↓ TST (p < 0.001), SE (p < 0.001), REM% (p < 0.01), N3,4 (p < 0.001) ↑ WASO (p < 0.001), SL (p < 0.001), RL (p < 0.01), N2 (p < 0.001) NRegASD vs. TD ↓ TST (p < 0.001), SE (p < 0.001) ↑ WASO (p < 0.001), SL (p < 0.001), RL (p < 0.01) |
ASD vs. TD ↓A1% (p < 0.001) ↑A2% (p < 0.01) A3% (p < 0.001) RegASD vs. TD ↓CAP rate during N1,2 (p < 0.01) |
NregASD vs. RegASD CSHQ: ↓ Bedtime, Bedtime resistance, Sleep onset delay, Sleep duration, Night-wakings (p < 0.001); Sleep latency (p < 0.05) ↑ Sleep length (p < 0.001) RegASD/NRegASD vs. TD ↓ Sleep length, (p < 0.001) ↑ Bedtime, Sleep latency, Bedtime resistance, Sleep onset delay, Sleep duration, Night-wakings (p < 0.001) |
|
| Buckley, 2010, USA [32] |
ASD vs. TD ↓TST (p = 0.004), REM% (p < 0.001) ↑RL (p = 0.016), N3,4 (p = 0.001) ASD vs. DD ↓TST (p = 0.001), REM% (p < 0.001) ↑RL (p = 0.012), N3,4 (p < 0.001) |
CSHQ: Median wake time: ASD 06.17 DD 06.45 TD 06.46 |
||
| Tessier, 2015, Canada [33] |
hfASD vs. TD Fp1 ↓Sleep Spindles duration (p < 0.05) Fp2 ↓Sleep Spindles density (p < 0.05) ↓Fast sigma EEG activity at C3, C4 (p < 0.05) |
Sleep diary: No sleep disturbances complained in the previous 14 days. |
TD negative correlation between VIQ and Fp2 spindle density for the last quarter of the night (r= −0.6, p < 0.04) positive correlation: between VIQ and C4 spindle duration for the total night (r = 0.72, p = 0.01) between PIQ and fast sigma activity in the end of the night at the C4 electrode (r = 0.59, p = 0.04) ASD negative correlation: between VIQ and C3 spindle density for the total night (r= −0.62, p = 0.02) between FSIQ and C3 spindle density for the total night (r= −0.55, p = 0.05) |
|
| Tessier, 2015 Canada [34] |
No significant results | Sleep diary: No sleep disturbances complained in the previous 14 days. |
ASD vs. TD ↑ neutral emotion reaction times on the delayed recognition task ((η2 = 0.16, p = 0.04) |
|
| Lambert, 2015 Canada [35] |
ASD vs. TD ↑SL (p = 0.02) ↓N3,4 (p = 0.026) |
ASD vs. TD Fp1 ↓K-complex (p = 0.006) Fp2 ↓Sleep Spindles density (p = 0.03), K-complex (p = 0.013) C3 ↓K-complex (p = 0.002) C4 ↓K-complex (p = 0.006) |
ASD vs. TD CSHQ: No significant results. Agendas: ↓Sleep onset latency (p < 0.05) Sleep quality (p < 0.02) |
ASD Negative correlation between N1% and FSIQ (r = −0.53, p = 0.009) and PIQ (r = −0.65, p = 0.001) Negative correlation between N3,4% and CBCL internalized behaviors (r = −0.41, p = 0.046). Positive correlation between SL reported in daily sleep agendas and in PSG in both groups (r = 0.75, p < 0.001). |
| Sahroni, 2015, Japan [36] |
ASD vs. TD ↑ absolute theta band power in T6 (p = 0.0379) ↑ absolute alpha band power in F7, Fz, F4, T3, Cz, C4, P3 (p < 0.03) ↑ relative delta band power in Fz, T6 (p = 0.0379) ↓ relative beta band power in T6 (p < 0.04) ↓ absolute and relative gamma band power in Fp1, T5, P3, T6, O1, O2 (p < 0.04) |
|||
| Maski, 2015, USA [37] |
ASD vs. TD ↑ TIB (p = 0.01), WASO (p = 0.02), SL (p = 0.01) ↓ SE (p < 0.001), REM% (p = 0.007) |
ASD vs. TD CSHQ: ↑ Bedtime resistance (p = 0.03), Sleep onset delay (p = 0.02), Sleep duration (p = 0.04), Sleep anxiety (p = 0.001), Daytime sleepiness (p < 0.02), Parasomnias (p = 0.02) |
ASD vs. TD No significant differences in benefiting from sleep in memory consolidation tasks |
|
| Lehoux, 2017, Canada [38] |
ASD vs. TD ↓N3,4 (p = 0.007) |
Sleep diary: No sleep disturbances complained in the previous 14 days. |
||
| Aathira, 2017, India [39] |
poor sleepers rate among ASD: 77.5% ASD vs. TD CSHQ: ↑ Daytime sleepiness (p < 0.001), Parasomnias (p < 0.001), Sleep anxiety (p = 0.002), Bedtime resistance (p < 0.001) |
psASD vs. gsASD ↑ higher CBCL mean score (p = 0.004), CBCL “withdrawn” score in the borderline or clinical range (p = 0.03) Not significant results about IQ and CARS |
||
| Vite, 2018, Messico [40] |
ASD vs. TD ↑ Mu rhythm peak in C3 (p = 0.003) ↓ Mu rhythm peak in C4 (p = 0.003) |
|||
| Farmer, 2018 USA [41] |
ASD vs. TD ↓Sleep Spindles density (p < 0.0001) ↓Sleep Spindles duration (p = 0.006) ASD vs. DD ↓Sleep Spindles density (p = 0.017) |
For the full sample, significant correlation between: spindle density and IQ (r = 0.26, p < 0.002) spindle density and Vineland subscales: socialization (r = 0.33, p = 0.0001) communication (r = 0.32, p = 0.0002) living skills (r = 0.25, p = 0.003) |
||
| Page, 2019, USA [42] |
ASD vs. TD ↓ theta band power in temporo-central regions (p < 0.05) ↑ beta band power in right temporo-occipital region (p < 0.05) ↑ slower sigma power over occipital and central regions (p < 0.05) ↓ higher frequency sigma power over frontal, central, and parietal regions (p < 0.05) |
ASD vs. TD Sleep diary: No significant differences in the naptime nor in the duration of wakefulness before the nap. |
No significant correlation between ADOS-2 score and NREM spectral power |
|
| Fletcher, 2019, UK [43] |
ASD vs. TD ↓ TST (p≤0.05), NREM (p < 0.05) |
ASD vs. TD ↓ sigma power(p ≤ 0.001) |
ASD vs. TD CSHQ: ↑Total sleep problems (p < 0.001) |
ASD vs. TD ↓performances in specific memory tasks with memory recalling after a month |
| Arazi, 2019, Israel [44] |
ASD vs. TD ↓ TIB (p = 0.02), TST (p = 0.03), REM% - second half of the night (p = 0.007) |
ASD vs. TD CSHQ: 50% of ASD children had scores that were above the mean score from previously published CSHQ scores from a large population of typically developing children in all domains, excluding sleep duration and sleep disordered breathing |
Negative correlation between SWA power and Bedtime resistance (r= −0.49, p = 0.01), Total sleep disturbances (r= −0.38, p = 0.05) and time to fall asleep (r = 0.42, p = 0.02). |
↓ Reduced value; ↑ Increased value; A1: subtype 1 of phase A of CAP; A2: subtype 2 of phase A of CAP; A3: subtype 3 of phase A of CAP; ADOS-2: Autism Diagnostic Observation Schedule-2nd Edition; AS: Asperger syndrome; ASD: autism spectrum disorder; CAP: cyclic alternating patterns; CARS: Childhood Autism Rating Scale; CBCL: Child Behavior Checklist; CSHQ: Children’s Sleep Habits Questionnaire; DD: developmental delay; FSIQ: Full Scale Intelligence Quotient; gsASD: good sleepers ASD; hfASD: high functioning ASD; IQ: Intelligence Quotient; N1: stage 1 of NREM; N2: stage 2 of NREM; N3: stage 3 of NREM; N4: stage 4 of NREM; NRegASD: not regressive ASD; NREM: non-rapid eye movement; PIQ: Performance Intelligence Quotient; psASD: poor sleepers ASD; PCQ: Parental Concerns Questionnaire; PSG: polysomnography; r: correlation coefficient; RegASD: regressive ASD; REM: rapid eye movement; RL: REM latency; SE: sleep efficiency; SL: sleep latency; SPT: sleep period time; SWA: slow wave activity; SWS: slow wave sleep; TD: typically developing children; TIB: time in bed; TST: total sleep time; VIQ: Verbal Intelligence Quotient; WASO: wakefulness after sleep onset; η2: Effect size.