Maternal Voice and Infant Sleep in the Neonatal Intensive Care Unit

Approximately 10% of US newborns require a NICU. We evaluated whether the NICU acoustic environment affects neonatal sleep and whether exposure to the mother's voice can modulate that impact. In a level IV NICU with single-infant rooms, 47 neonates underwent 12-hour polysomnography. Their mothers were recorded reading children's books. Continuous maternal voice playback was randomized to either the first or second 6 hours of the polysomnogram. Regression models were used to examine sleep-wake stages, entropy, EEG power, and the probability of awakening in response to ambient noise during and without voice playback. After epochs with elevated noise, the probability was higher with (versus without) maternal voice exposure of neonates staying asleep ( = .009). However, the 20 neonates born at ≥35 weeks' gestation, in contrast to those born at 33 to 34 weeks, showed an age-related increase in percent time awake ( = 0.52; < .001), a decrease in overall sleep ( = 0.52; < .001), a reduction in rapid eye movement sleep bouts per hour ( = 0.35; = .003), and an increase in sleep-wake entropy ( = 0.52; < .001) all confined solely to the 6 hours of maternal voice exposure. These associations remained significant ( = .02 to < .001) after adjustment for neurologic examination scores and ambient noise. Hospitalized newborns born at ≥35 weeks' gestation but not at 33 to 34 weeks' gestation show increasing wakefulness in response to their mother's voice. However, exposure to the mother's voice during sleep may also help protect newborns from awakening after bursts of loud hospital noise.