Maternal Prenatal Distress, Neonatal Brain Volume, and Infant Sleep

• In Event: Prenatal Distress and Neurobehavior in Infancy and Early Childhood: Testing Putative Biological Mechanisms

Abstract

Maternal distress during pregnancy has been consistently linked to child neurodevelopment (e.g., Van den Bergh, 2018). Identifying intermediate phenotypes or endophenotypes associated with exposure to prenatal distress is an important research area with clear implications for prevention and intervention efforts. The current study tests whether there are associations between prenatal distress and infant sleep. Infant sleep represents a novel endophenotype, as it is both a marker of central nervous system integrity (Freudigman & Thoman, 1993) and a predictor of child cognitive and behavioral development (Bernier et al., 2010), yet to date sleep has not been examined in relation to prenatal distress. The field also lacks a clear understanding of the biological mechanisms by which maternal distress impacts child neurodevelopment. Brain structure is one putative mechanism, given previous evidence of structural brain differences in neonates of mothers with a history of psychiatric disorder (Knickmeyer et al., 2017). Thus, the current study tests whether alterations in neonatal brain volume mediate relationships between prenatal distress and sleep.

Pregnant women (N=115, 33% Black, MGA=27.3 weeks) reported their perceived stress (Cohen’s Perceived Stress Scale; Cohen, 1994) and symptoms of depression and anxiety (Brief Symptom Inventory-18; Derogatis, 2000). Given that these constructs were highly intercorrelated (Table 1), we created a cumulative prenatal distress (CPD) index. First, we dichotomized the three continuous measures to indicate high stress (≥ 75th percentile; 30%), clinically-elevated depression (T≥63; 10%), and clinically-elevated anxiety (T≥63; 5.2%). Then, we summed the three dichotomous indicators into a single measure.

MRI data were acquired from neonates (Mage=14.5 days) during natural sleep. White (WM) and gray matter (GM) volumes were segmented using the UNC NIRAL Multiseg processing pipeline. To account for confounding with intracranial volume (ICV), we used proportion of white (WMp=WM/ICV) and gray matter (GMp=GM/ICV) in analyses. When infants were 6 months old (Mage=6.67 months), mothers reported their infant’s sleep duration (daytime, nighttime, and 24-hour), fragmentation (night wakings), and consolidation (ratio of nighttime to 24-hour sleep) using three nights of sleep diaries.

Descriptive statistics are presented in Table 1. We tested associations between prenatal distress, neonatal brain volume, and infant sleep using path analysis. Separate models were run for each infant sleep measure and all models included relevant covariates (Figure 1). We observed significant direct effects between CPD and infant sleep, such that higher CPD predicted less nighttime (β = -.36, p < .001) and 24-hour sleep (β = -.30, p = .009). Follow-up models that replaced CPD with continuous measures of perceived stress and depression yielded similar findings (β = -.26 to -.46, p < .02). There was no evidence that neonatal brain volume mediated associations between prenatal distress and infant sleep.

Thus, we have evidence that prenatal distress impacts infant sleep duration, but not sleep quality. Further consideration of infant sleep in studies of prenatal distress may therefore be warranted. Global differences in neonatal brain structure did not mediate these associations. Future studies should examine white and gray matter microstructure as well as functional connectivity as mechanisms linking prenatal distress to infant neurodevelopment.

Authors

• Marie Camerota, Brown University
  Presenting Author

• Sarah J Short, University of Wisconsin - Madison
  Non-Presenting Author

• Rebecca Stephens, University of North Carolina at Chapel Hill
  Non-Presenting Author

• Cathi Propper, University of North Carolina at Chapel Hill
  Non-Presenting Author