Title : Cortical response of extremely and very preterm newborns to recorded maternal or unknown woman’s voice

Background: Among the various neuroprotective interventions aimed at limiting the potential adverse effects related to preterm birth, maternal voice exposure, both recorded and live, have been proposed as a feasible care intervention, given its  potential positive effects  on preterm infant’s developmental outcomes. This study aimed to analyse, by multichannel near-infrared spectroscopy (NIRS), the activation of extremely and very preterm newborn’s cerebral cortex associated with the listening of registered maternal voice or that of an unknown woman. We hypothesized that the preterm newborn cortex reacts very early to maternal voice and that cortical activation is focused in the temporal region.

Methods: Four extremely (gestational age < 28 weeks) and eight very preterm newborns (gestational age < 32 weeks) were included and monitored, once a week, between the 32nd and 36th week postmenstrual age. By multichannel NIRS, the increase of oxygenated haemoglobin (HbO?), as cortical activation esteem, was assessed in association with listening to the recorded maternal or unknown woman’s voice, registered when reading Aesop’s tail “The country mouse and the city mouse”. The functional activation of the parietal and temporal cortex, bilaterally, was identified by paired t-test, comparing silent periods (baseline) to the listening of maternal voice or unknown woman voice. Furthermore, in significantly activated cortical sites, a linear regression analysis was performed, in which the increase in HbO2 constituted the dependent variable, while neonatal behavioural state, according to the Brazelton scale, represented the independent variable.

Results: During listening to maternal voice, no significant cortical activation was observed. During listening to the unknown woman voice, the left superior temporal cortex was significantly activated (p < 0.01). Furthermore, a positive effect of the newborn's behavioural state on the increase of HbO2 in the same cortical region was observed (p = 0.041): the more alert was the infant, the higher was his/her cortical response.

Conclusion: Contrary to our hypothesis, we did not observe any cortical activation while our preterm newborns were listening to maternal voice. However, the left superior temporal cortex showed to be activated, while infants listened to an unknown woman voice. Moreover, such activation was positively related to the neonatal behavioural state. These results may be attributed to the involvement of the neonatal attentional system, alerted by the unknown voice. However, as the familiar maternal voice did not significantly activate the infant cortex, it may be speculated that it does not alert the newborn's attentional system, promoting at the same time his/her psychophysical stability.

Dr. Stefano Bembich is a clinical psychologist and researcher. Currently, he works at the University Hospital of Udine (Italy), in the Mother-Child Department, as clinician. In collaboration with the Institute for Maternal and Child Health IRCCS “Burlo Garofolo” of Trieste (Italy), his research activity concerns the application of cerebral near-infrared spectroscopy, as a functional neuroimaging technique, in the field of neonatology. He published more than 20 papers on this topic.