Abstract
BACKGROUND: Elevated maternal glucocorticoids during pregnancy may impact on fetal development in a sex-dependent way, leading to increased amygdala activation and increased risk for internalising disorders in females. Based on evidence implicating reduced amygdala activation in callous-unemotional (CU) traits, we predicted that elevated maternal cortisol in pregnancy would be associated with lower CU traits and elevated anxious-depressed symptoms, only in girls. METHODS: Participants were 225 members of a stratified subsample within an epidemiological longitudinal cohort (WCHADS). Salivary cortisol was measured over two days at 32 weeks gestation (on waking, 30-min post-waking and during the evening) and the log of the area under the curve (LogAUC) was calculated as an index of diurnal cortisol. Mothers reported on child CU traits and anxious-depressed symptoms at 2.5, 3.5 and 5.0 years of age. RESULTS: As predicted there was a sex of child by cortisol interaction (p < .001) whereby elevated maternal cortisol was associated with lower child CU traits, explaining 25% of the variance, in girls, but not in boys. This effect remained when controlling for relevant confounders and anxious-depressed symptoms. By contrast, elevated maternal cortisol did not predict higher anxious-depressed symptoms in girls. CONCLUSIONS: The study adds to growing evidence for sex-dependent effects of elevated maternal cortisol during pregnancy on early child psychopathology, consistent with mediation by elevated amygdala activation. The conditions under which, in girls, this is associated with heightened responsiveness to others' distress characteristic of low CU traits, or with increased affective symptoms, require further study.