Developmental trajectories of sleep EEG in neurodevelopmental disorders: Does sex matter?

We aimed to examine the effect of sex on maturational changes in sleep neurophysiology indexed by polysomnography in children with neurodevelopmental disorders(NDD) compared to non-NDD individuals. We utilized whole-night polysomnography data from a large pediatric (2.5 to 17.5 years) sample available at the National Sleep Research Resource. Using ICD codes, we defined the NDD subsets for ASD(N=196), ADHD(N=525), intellectual disabilities(ID, N=167), and a no-NDD sample(N=1523). The open-source package Luna was used to derive multiple sleep features, including sleep spindles, slow oscillations(SO), and spectral power. Linear regression models were employed to investigate sex-by-age interaction. Out of 351 EEG features, 31% expressed nominally significant sex-by-age interaction in the non-NDD cohort. For example, during the N3 stage, the age-related increase in absolute sigma power was steeper in boys(p<0.001 across all channels), while girls displayed a steeper decline in beta power(p<0.003 across all channels). In the ADHD group, the majority of significant sex-by-age interactions were observed in power and were congruent in directions to the non-NDD group. In contrast, distinct developmental trajectories between boys and girls with ASD or ID samples were observed for sleep features that did not present significant sex-by-age effects in the non-NDD group and mostly included spindle and SO-spindle coupling metrics.Our findings suggest that multiple sleep EEG metrics follow distinctive developmental trajectories in boys and girls without NDDs. The NDD cohorts also exhibit sex-specific developmental patterns of sleep features often not present in the typical population which could potentially relate to disorder-relevant factors including symptom manifestation or disorder progression.