Sleep as a window for evaluating neurodevelopmental outcome: which impact on the brain of preterm infants? A prospective study on prematurity from a biological neuropsychiatry perspective

Background. Newborns and infants spend most of their time sleeping an immature sleep, which allows brain maturation a good neurodevelopment. Preterm birth is associated with abnormal brain development and alterations in later-in-life sleep patterns, carrying a high social burden, even when not accompanied by major neurological damages. Which impact prematurity itself can have on early sleep architecture, which influence it can have on well- known adverse outcomes, and which role brain lesions play in determining sleep patterns in preterm infants are still matter of debate. This exploratory pilot study aimed to describe the distribution of sleep states among very low birth weight (VLBW) infants, and to correlate it with neurobehavioral assessment at 35 weeks of post-menstrual age (PMA), and to observe if these persisted at term equivalent age (TEA) and at 6 months of corrected age (CA). Secondly, it aimed to assess if the presence of a major or minor brain lesion detected at MRI can affect sleep duration, distribution and quality. Methods. 10 VLBW were assessed at 34±2 weeks PMA with a 24-hours video- polysomnographic recording and received a neurobehavioral examination at the moment of the recording and at TEA (with Neonatal Behavior Assessment Scale; Hammersmith Neurological Neonatal Examination, and the neonatal visual battery). They were followed- up at 6 months CA with Griffiths’ Mental Development Scale III edition. Analysis of sleep stages distribution and spectra was conducted. Results. Total sleep time and total amount of transitional sleep (TS) significantly positively correlated with neurological, and neurobehavioral assessment at 34 weeks PMA, at TEA and with neurodevelopment at 6 months CA, while Sleep Onset Active Sleep (SOAS) had a negative association. Infants carrying severe-moderate brain lesions showed lower Total Sleep time (66.9% ± 7.39 vs 72.2% ± 3.52, p = 0.047) accompanied by a higher prevalence of SOAS (23.9% ± 10.2 vs 12.26% ± 5.5 p = 0.048), and showed a gradient for higher power of posterior slow activity (slow δ and δ) during SOAS from the posterior cerebral regions. Conclusions. Understanding sleep mechanism among preterm infants might provide future therapeutic/management strategies, which need to encompass sleep care. Further analyses with larger samples and more complex methods are claimed.