Sleep in children with autism with and without autistic regression

The purpose of the present investigation was to characterize and compare traditional sleep architecture and non-rapid eye movement (NREM) sleep microstructure in a well-defined cohort of children with regressive and non-regressive autism, and in typically developing children (TD). We hypothesized that children with regressive autism would demonstrate a greater degree of sleep disruption either at a macrostructural or microstructural level and a more problematic sleep as reported by parents. Twenty-two children with non-regressive autism, 18 with regressive autism without comorbid pathologies and 12 with TD, aged 5-10years, underwent standard overnight multi-channel polysomnographic evaluation. Parents completed a structured questionnaire (Childrens' Sleep Habits Questionnaire-CSHQ). The initial hypothesis, that regressed children have more disrupted sleep, was supported by our findings that they scored significantly higher on CSHQ, particularly on bedtime resistance, sleep onset delay, sleep duration and night wakings CSHQ subdomains than non-regressed peers, and both scored more than typically developing controls. Regressive subjects had significantly less efficient sleep, less total sleep time, prolonged sleep latency, prolonged REM latency and more time awake after sleep onset than non-regressive children and the TD group. Regressive children showed lower cyclic alternating pattern (CAP) rates and A1 index in light sleep than non-regressive and TD children. Our findings suggest that, even though no particular differences in sleep architecture were found between the two groups of children with autism, those who experienced regression showed more sleep disorders and a disruption of sleep either from a macro- or from a microstructural viewpoint.