Hair glucocorticoids during pregnancy in the context of trauma exposure and their predictive value for the development of childbirth-related posttraumatic stress disorder symptoms

Background: Childbirth-related posttraumatic stress disorder (CB-PTSD) is gaining attention as a mental disorder with negative sequela for mothers and their offspring. Maternal trauma history is a well-known vulnerability factor for CB-PTSD symptoms (CB-PTSS). Furthermore, alterations of the hypothalamus-pituitary-adrenal axis have been linked to both trauma exposure and PTSD development. Hence, we investigated whether trauma history was associated with long-term glucocorticoid (GC) levels during pregnancy and their predictive role for CB-PTSS. Further, we examined whether GCs act as a mediator in the relationship between trauma history and CB-PTSS and whether this was moderated by the subjective birth experience. Methods: 212 women participating in the prospective cohort study DREAMHAIR provided hair samples for quantification of long-term integrated cortisol and cortisone levels prior to their anticipated birth date accom-panied by measures of trauma history. CB-PTSS and subjective birth experience were assessed two months postpartum. Findings: Trauma history predicted elevated hair cortisol and hair cortisone during the third trimester of preg-nancy, however associations did not remain significant when Bonferroni correction due to multiple testing was applied. Trauma history also predicted higher CB-PTSS. Hair GC levels during pregnancy neither predicted CB-PTSS two months after birth nor mediated the relationship between trauma history and CB-PTSS. The subjective birth experience moderated the relationship of hair cortisol and cortisone with CB-PTSS.Conclusion: Our data suggest that a history of trauma contributes to a higher risk to develop CB-PTSS and elevated long-term GC levels during the third pregnancy trimester. Further, the predictive role of hair cortisol and cortisone levels for CB-PTSS may depend on subjective birth experience. This highlights the need to consider the latter in future investigations when examining the role of stress-related biomarkers in more severely affected samples.