Bibian van der Voorn

193 GENERAL DISCUSSION AND SUMMARY puberty. In summary, these data suggest that both gender and genetic predisposition can result in differences in stress vulnerability. STRESS IN EARLY LIFE When a child is born very preterm, it is exposed to an excessive amount of stress during a critical window of development, while the body’s immaturity makes it difficult to respond adequately to these stressors. Indeed, in very preterm infants a high prevalence of relative adrenal insufficiency is seen in the first weeks of life, due to the immaturity of multiple levels along the HPA axis ( Chapter 1 ). In addition, the ongoing process of third trimester neurodevelopment makes the brain vulnerable for stressful events 3 . Animal studies suggest that adaptation to stressful events during this critical developmental window may lead to alterations in HPA axis set point through an altered balance between MR and GR expression in the brain 2 . Replication of such experiments in humans is impossible, but such data provide working hypotheses on the process of stress adaptation in very preterm newborns. Accordingly, we assessed several indicators of early life stress, including hair glucocorticoid levels, breast-milk glucocorticoids, and early life growth. In chapter 2 and 3 , we explored the reliability of a novel strategy for assessment of intra-uterine glucocorticoid regulation. Whereas in adult hair, cortisol and cortisone levels are shown to offer a valid retrospective view on glucocorticoid exposure 4,5 , we explored factors influencing neonatal hair glucocorticoids in utero and postpartum. In chapter 2 , we found a positive association between gestational age and neonatal hair glucocorticoid levels, which seemed to represent the positive feedback loop between placental CRH, and maternal and fetal cortisol. In chapter 3 , we found a negative association between maternal perinatal stress and neonatal hair glucocorticoid levels, which seemed to represent a downregulation of fetal HPA axis activity after overexposure to maternal glucocorticoids. Accordingly, these data suggest that newborn hair cortisol and cortisone levels could be an indicator of intra- uterine glucocorticoid regulation. When own mother’s milk is not available for very preterm newborns, pasteurized human donor milk is advised as the feeding of choice. Glucocorticoids in human milk might have the potential to provide beneficial effects on development of the preterm neonate. Therefore, we assessed variability in breast-milk glucocorticoid concentrations within and between mothers. In chapter 5 , we found that breast- milk cortisol and cortisone concentrations were lower in mothers who deliver