Abstract
BACKGROUND: The effect of supplemental oxygen on sleep has not been studied in preterm infants. METHODS: We studied 18 stable late-preterm infants with observed periodic breathing at a median gestational age of 36 weeks. Polysomnography was performed on room air and on 25% oxygen-enriched ambient air. RESULTS: Supplemental oxygen did not affect sleep stage distribution, sleep efficiency, the frequency of sleep stage transitions, the appearance of rapid-eye movement (REM) sleep periods, or the high number of spontaneous arousals. The percentage in periodic breathing out of total sleep time decreased from 10% (interquartile range [IQR] 5-9%) on room air to 1% (IQR 0-3%) (p < 0.001) on supplemental oxygen. Also, the number of central apneas decreased from 48 (IQR 32-68) to 23 (IRQ 15-32) per hour (p < 0.001), and the number of oxygen desaturations of a minimum 3% from 38 (IQR 29-74) to 10 (IQR 5-24) per hour (p < 0.001). On room air in non-REM sleep, the median end-tidal carbon dioxide values were systematically lower during periodic breathing at 5.1 (IQR 4.6-6.4) kPa than during stable breathing at 5.5 (4.9-5.9) kPa (p < 0.0001). CONCLUSIONS: In late-preterm infants, supplemental oxygen effectively reduces periodic breathing and the number of oxygen desaturations while having no significant effect on sleep. The results support the importance of carotid body over-reactivity on the genesis of periodic breathing in preterm infants.