Maternal prenatal stress and infant postnatal salivary cortisol levels: does maternal sensitivity moderate the link?

Background: Difficulties regulating biological, behavioural and emotional processes are fundamental to most childhood psychopathology. One possible mechanism for early dysregulation, supported by evidence from the animal literature, is the programming of the foetal hypothalamic-pituitary-adrenal (HPA) axis in utero. Maternal antenatal stress in animals is linked with impaired behavioural and emotional reactivity in offspring, along with alterations in HPA axis function. The effect of maternal rearing behaviours has been studied in rodents and high quality maternal behaviours in the postnatal period moderate the effect of prenatal stress on offspring outcomes. In humans, there is evidence that maternal stress in pregnancy predicts subsequent childhood behavioural difficulties, and emerging evidence that it may alter the function of the infants HPA axis, suggesting that the developing baby could be sensitive to maternal stress hormones in utero and caregiving behaviours in the postnatal period. Aims: The aims of the present study were three-fold. First, to examine the effect of prenatal stress, as indexed by maternal anxiety and depression symptoms on subsequent infant cortisol levels and reactivity to a social stressor at 6 months of age, taking into account potential confounding variables where appropriate. Second, to examine the potential moderating role of maternal sensitivity in the association between maternal prenatal stress infant cortisol levels. And third, to compare the effect of maternal prenatal stress measured at two gestational time points on infant cortisol outcome. Method: This was a prospective longitudinal study of a sub-sample of 91 mother-infant dyads, selected from a larger consecutively recruited sample of first time mothers for intensive study within the Wirral Child Health and Development Study based on their varying levels of intimate partner relationship dysfunction. Maternal prenatal stress was measured during the 2nd and 3rd trimesters of pregnancy, indexed by self-report of anxiety and depression symptoms. Infants’ salivary cortisol levels were measured before and after a social stressor paradigm, the still-face procedure, at 6 months of age and maternal sensitivity towards her infant was measured in an 8 minute playful interaction during the same laboratory visit. Results: No main effect of maternal prenatal stress in the 2nd or 3rd trimester of pregnancy on infant HPA axis function was found using multivariate regression analysis controlling for potential confounding variables. There was a significant association between maternal prenatal anxiety during the 2nd trimester of pregnancy in interaction with maternal sensitivity on infant baseline cortisol. Maternal sensitivity moderated the association between high maternal anxiety during the 2nd trimester and high infant baseline cortisol at 6 months of age. This association remained after controlling for potential confounding obstetric outcomes, maternal demographics and concurrent maternal mood. Conclusions: It appears that the function of infant’s HPA axis may be programmed by exposure to maternal stress during pregnancy but that this programming effect is subject to and moderated by the quality of maternal caregiving behaviour in the postnatal period. The importance of the impact of maternal caregiving behaviour on the relationship between maternal stress during pregnancy and infant outcomes is consistent with the animal and emerging human literature on the subject