The association between prenatal maternal anxiety and infant hypothalamic pituitary adrenal axis function

Background: Animal studies have amply demonstrated the negative
impact of maternal prenatal stress on offspring behavioural
adjustment and emotional reactivity. Experimental laboratory evidence
points to changes in the hypothalamic-pituitary-adrenal
(HPA) axis as one mechanism mediating this link. Although a
small number of studies suggest that analogous effects occur
in humans, the data is presently limited [1,2]. This study will
investigate the impact of maternal stress and anxiety in pregnancy
on infant behavioural and neuroendocrine responses to stress.
Aim: to prospectively examine the association between maternal
prenatal anxiety and infants’ HPA axis responses to a
laboratory stressor at 7 months postpartum. A novel contribution
of the study is the use of an interactive stress paradigm (the
“still-face” procedure) to measure both maternal sensitivity and
infant stress reactivity. The findings of this study are expected
to enhance the understanding of the mechanisms that underlie
individual differences in infants’ responses to stress.
Method: Maternal anxiety was assessed during the third
trimester of pregnancy (mean 32 weeks) using both self-report
Spielberger State and Trait Anxiety Questionnaire) and clinical
diagnostic interview (a modified version of the MINI). The
association between maternal anxiety and infant HPA axis was
assessed by measuring infant salivary cortisol upon arrival at the
laboratory, and at 15-, 25-, and 40-minutes following the still-face
procedure. The still-face procedure consists of two minutes play
between mother and infant; 2 minutes when mother reenters the
room but remains completely expressionless (the “still face”); and
then 2 minutes when she re-engages with the infant. We added an
extra component of 1 minute when mother was absent from the
room prior to the still-face segment. This laboratory procedure
has been demonstrated to be a stressfull activation paradigm for
the infant.
Results: 149 women were recruited but data could only be
analysed for 71 women and their 7 month-old infants, due to
a signficant attrition rate and missing data. Initial attrition was
related to maternal psychological risk, but data loss at follow up
was unrelated to maternal demographic variables or any infant
birth or clinical characteristics.
The sample was homogeneous in terms of maternal obstetric
history and outcomes (infant gestation, weight), socioeconomic
and marital status, and age. Repeated measures MANOVA yielded
significant interactions indicating that infants’ patterns of cortisol
response following the laboratory procedures differed as a function
of maternal prenatal anxiety, after adjusting for the effects of
maternal postnatal anxiety and depression symptoms. The results
were identical for maternal prenatal anxiety assessed using self
report, and diagnostic interview, Wilks’ Lambda = 0.91, F(2,65),
p = 0.03. The analyses remained significant after adjusting for the
effects of maternal postnatal anxiety and depression symptoms.
Infants of mothers who were less anxious during pregnancy
generally displayed a post-stress increase in cortisol, followed
by a decrease to below baseline at 40 minutes post-stress. In
contrast, the infants of prenatally anxious women typically showed
little change in their cortisol response from baseline through 40
minutes post-stress. Thus infants of prenatally anxious women
demonstrated what appears to be a blunted or diminished pattern
of HPA axis activity.
Conclusion: This study provides evidence that exposure to
prenatal anxiety in humans may be associated with subsequent
alterations in infants’ HPA axis responses to challenge. We shall
comment on how maternal sensitivity may also modulate infant
stress reactivity.