Bibian van der Voorn

194 CHAPTER 12 very preterm. However, it was also seen that breast-milk cortisol and cortisone concentrations varied widely between and within mothers, which suggested that time of collection was an important influencing factor. Subsequently, we tested the existence of diurnal rhythmicity and found that breast-milk glucocorticoids follow the diurnal rhythm of maternal HPA axis activity. As a result, whether concentrations in pretermmilk differ from that in termmilk necessitates confirmation in a future study with a timed study design. In chapter 6 , we tested whether pasteurization of human milk is harmful to breast- milk glucocorticoids and determined that processing of human milk by donor milk banks does not affect milk cortisol and cortisone levels. In chapter 8 , the influence of fetal growth restriction and impaired early life growth on HPA axis development was analyzed, as a proxy for stressful intra-uterine and early life conditions. Cross-sectionally, at corrected term age, the infants that were born SGA and did not show catch up growth seemed to have lower cortisol levels as compared to their AGA counterparts that had grown conform expectations. These findings might reflect a suppression of neonatal HPA axis activity due to experienced stress. LONG-TERM OUTCOMES Stress reactions are important defense mechanisms, offering an advantage in terms of morbidity and mortality. However, adaptive responses to early life challenges may become non-adaptive if they overshoot or are a mismatch to later life challenges 6 . In 1986, Professor Barker published unique, epidemiological data that suggested a relation between early life stress, infant mortality and later life cardiovascular disease 7 . Nowadays, robust data support this Developmental Origins of Health and Disease (DOHaD) paradigm. Subsequently, the ‘fetal cortisol hypothesis’ was postulated as an explanation for increased fetal HPA axis activity and risk of later life cardiovascular disease. Briefly, this hypothesis suggested that an impaired activity of the plancental barrier enzyme 11βHSD type 2 results in intra-uterine overexposure to maternal glucocorticoids, and thereby increased fetal HPA axis activity. We have suggested that such programming mechanisms might also occur postnatally, at least in preterm infants ( Chapter 1 ). Accordingly in chapter 7 , we found that genetic differences in glucocorticoid sensitivity can predispose very preterm survivors to long-term effects on cognitive and behavioral functioning after exposure to a single gift of synthetic glucocorticoids during a critical window of neurodevelopment.