Childhood maltreatment, pituitary volume and adolescent hypothalamic-pituitary-adrenal axis - Evidence for a maltreatment-related attenuation

Alterations of the development of the hypothalamic-pituitary-adrenal axis (HPAA) have been suggested to be related to experiences of early maltreatment. It has been postulated that early stress (i.e., maltreatment) leads to initial hyperactivation of the HPAA, which subsequently may progress to hypoactivation during the course of adolescence, however empirical studies on this hypothesis are rare. In the current study, we aimed to examine the longitudinal relationships between childhood maltreatment, early adolescent pituitary gland volume (PGV) and mid-adolescent cortisol output in an existing data set to explore the utility of PGV as a measure of HPAA function, and as an indirect test of the attenuation hypothesis.

The sample comprised 69 adolescents (30 females), subsampled from a larger longitudinal, community-based study on adolescent development. PGV, as an estimate of chronic childhood HPAA activity, was measured by magnetic resonance imaging during early adolescence (mean age 12.62 ± 0.45 years). Cortisol output was assessed via multiple salivary cortisol measures in mid-adolescence (mean age 15.52 ± 0.39 years). The cortisol awakening response (CAR) was calculated as a measure of HPAA functioning. Retrospective assessment of childhood maltreatment was performed using the Childhood Trauma Questionnaire (CTQ). Regression analyses were conducted to examine whether childhood maltreatment, PGV, and their interaction, predicted mid-adolescent CAR.

No main effect of PGV or maltreatment was found on adolescent CAR. PGV did however significantly interact with childhood maltreatment in predicting the CAR (t = -2.26; p = 0.024). Larger PGV positively predicted lower CAR in the context of relatively high childhood maltreatment (t = 2.032; p = 0.046), but showed no relationship in the context of relatively low maltreatment (t = 0.723; p = 0.472). Maltreatment also interacted with sex, such that (only) in females, higher levels of maltreatment predicted a lower CAR (t = -2.04, p = 0.042).

In the presence of childhood maltreatment, larger PGV was associated with lower CAR in adolescence, providing support for the application of PGV in studies of HPA axis function. Our finding is consistent with a maltreatment-related attenuation of HPAA functioning that may derive from a stress induced chronic hyperactivation during childhood. Prospective longitudinal studies are now required to further explicate these findings and relationships with psychopathology.