Abstract Introduction Changes in slow wave activity (SWA) are indicative of brain maturation, particularly in the transition to adolescence. However, whether abnormal maturational trajectories of SWA predict neurodevelopmental and psychopathological disorders has remained unclear given the lack of longitudinal studies across the lifespan. Methods A total of 674 participants from the Penn State Child Cohort, a randomly-selected population-based sample, underwent 9-hour in-lab polysomnography studies in childhood (median age 9y), adolescence (median age 16y), and young adulthood (median age 25y) with a median time-to-follow-up of 15y. Growth mixture models for clustering of longitudinal data series were used to identify latent age-related classes (trajectories) of log-SWA, while adjusting for childhood sex, age, race/ethnicity, diagnoses or medication use. Mixed-effects models were fitted to test the association of log-SWA trajectories with presence of attention-deficit/hyperactivity (ADHD), learning (LD), and internalizing (ID) disorders and suicidality from age 5 through 30, while adjusting for sex, race/ethnicity, insomnia symptoms, and body mass, apnea/hypopnea, and periodic limb movements indices. Results SWA showed four cubic maturational trajectories, that were significantly associated with the rate of neurodevelopmental and psychiatric disorders (p=.02). Compared to the reference SWA trajectory (n=357), the trajectory with a delayed peak-SWA in childhood (n=182) had a higher rate of LD in childhood (p.01). The trajectory with lowest SWA since childhood through young adulthood (n=53) showed a higher rate of ADHD since childhood (p=.002). The trajectory with highest SWA in childhood and largest cubic spline (n=82) showed a higher rate of ID and suicidality in young adulthood (p=.03). Conclusion Abnormal trajectories of the maturing sleeping brain, as indexed by SWA, are differentially associated with childhood-onset neurodevelopmental disorders and adult-onset internalizing psychopathology. Support (if any) R01MH136472
Fernandez-Mendoza et al. (Fri,) studied this question.