VANCOUVER—Among newborns at risk of neurologic dysfunction, measures of neonatal sleep help predict 18-month neurodevelopmental outcomes, according to research presented at the 45th Annual Meeting of the Child Neurology Society.
Studies suggest that abnormal sleep has neurocognitive consequences for older infants and children and that polysomnogram data are associated with brain function in newborns who require neonatal intensive care. “Although sleep is a highly sophisticated brain function, it is not typically included in the newborn clinical neurological assessment,” said Renée A. Shellhaas, MD, MS, Assistant Professor of Pediatrics and Communicable Diseases at the University of Michigan in Ann Arbor, and colleagues.
To evaluate how polysomnography measures may add to standard predictors of neurodevelopmental outcome for newborns who require intensive care and are at risk for neurologic dysfunction, Dr. Shellhaas and colleagues conducted a longitudinal study of 29 newborns. Patients had a gestational age of 35 weeks or more, were cared for in a neonatal intensive care unit, and were clinically determined to be at risk of seizures. Researchers excluded patients with congenital anomalies or syndromes known to affect neurodevelopmental outcome or predispose patients to sleep-disordered breathing. They also excluded patients who had severely abnormal EEG without sleep–wake cycling.
Once a newborn was medically stable, researchers conducted a 12-hour attended, bedside polysomnogram. Polysomnograms were scored by a polysomnography technologist and reviewed by a sleep-medicine physician. Researchers calculated the proportion of each sleep–wake stage, entropy of the sequence of sleep–wake state transitions, and power spectra of the EEG portion of the polysomnogram.
Researchers evaluated neurodevelopmental outcome at 18 months to 22 months using the third edition of the Bayley Scales of Infant Development (BSID). They assessed associations between polysomnogram results and neurodevelopmental outcomes using regression techniques that de-emphasized outliers. Patients’ mean gestational age was 39.6 weeks. Seventeen of the 29 patients were male. Mean birth weight was 3.42 kg, and median five-minute Apgar score was 8.
In univariate analysis, increased time in quiet sleep predicted lower 18-month cognitive, language, and motor BSID scores. Higher entropy of sleep–wake transitions predicted lower motor scores. Increased low-frequency EEG power during quiet sleep predicted higher motor and language BSID scores. Gestational age and illness severity were not predictive of BSID results. A more abnormal neonatal neurologic exam score (ie, Thompson score) predicted lower cognitive and motor BSID scores.
In analyses adjusted for Thompson score, higher EEG power during neonatal quiet sleep was associated with better 18-month motor and language scores. In addition, increased time in neonatal quiet sleep was associated with lower 18-month cognitive and motor scores. “Notably, Thompson score was not an independent predictor of outcome when the sleep data were included in the bivariate models,” Dr. Shellhaas and colleagues said.
“Our results suggest that inefficient neonatal quiet sleep—more time in quiet sleep and lower delta frequency power during that stage—predicts lower 18-month neurodevelopmental outcome scores,” the researchers concluded. “Importantly, these novel measures of brain functional integrity were robust predictors even after adjusting for the neonatal neurologic examination score.”
Shellhaas RA, Burns JW, Barks JD, Chervin RD. Quantitative sleep stage analyses as a window to neonatal neurologic function. Neurology. 2014;82(5):390-395.
Shellhaas RA, Burns JW, Wiggins SA, et al. Sleep-wake cycling and cerebral oxygen metabolism among critically ill neonates. J Child Neurol. 2014;29(4):530-533.