regime controversial. Furthermore, it is not clear whether ACS is effective also in extremely preterm gestations. Objective: The overall objective with this thesis was to investigate the impact of repeat courses of ACS in exposed subjects, both on infant size at birth, and on longer term outcomes (study I-III). Another objective was to explore the association between timing of ACS administration and survival in extremely preterm infants (study IV). Methods: All studies in the thesis are cohort studies. In study I-III we used a cohort of about 100 children exposed to various courses of ACS in utero. We evaluated them regarding infant anthropometry at birth (study I), and risk factors for cardiovascular disease (study II) and neuropsychological function (study III) at follow up in adolescence/young adulthood. In study IV we evaluated a national population-based cohort of extremely preterm infants (EXPRESS – Extremely Preterm Infant in Sweden Study) regarding ACS administration-tobirth interval and survival. Results: We found a dose-dependent association between number of ACS-courses and restricted body size at birth (study I). There was no clear correlation between repeat courses of ACS in fetal life and cardiometabolic risk factors at 14-26 years of age (study II). In addition, there was no indication that repeat ACS exposure had an adverse impact on cognitive function or psychological health at follow-up in adolescents and young adults (study III). In study IV we found a significant reduction in mortality among extremely preterm infants after any ACS, with an optimal administration-to-birth interval of 1-7 days. Conclusions: Although exposure to repeat courses of ACS in utero were found to be related to gradually reduced body size at birth (indicating fetal growth restriction), it seems less likely from our findings that there are clinically important and long-standing adverse effects on cardiovascular and neuropsychological health. Another conclusion from this thesis is that ACS effectively reduces the mortality risk in extremely preterm infants and that timing of antenatal corticosteroids is important also in women delivering extremely preterm. PRETERM BIRTH 1.1.1 Definition and epidemiology The duration of a normal human pregnancy is 40 weeks or 280 days from the first day of last menstrual period.1 The currently accepted definition of preterm birth is birth before 37 completed weeks of pregnancy, very preterm before 32 weeks and extremely preterm before 28 weeks of pregnancy (figure 1).2 week 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 Extremely preterm 48 hours after birth). The incidence of septicemia in extremely preterm infants varies from 25-60%.48 16 Early neonatal infections are mainly contracted in utero or at birth and the most common microorganisms are group B streptococci. Late onset infections are mainly nosocomial and the most common microorganism is coagulase-negative staphylococci. Systemic infections are treated with antibiotics and antibiotic treatment is often given on wide indications before bacterial infection can be excluded. The use of ACS has been found to reduce the incidence of early systemic infections with an average of 40% (RR 0.60; 95% CI 0.41-0.88).12 1.2.4 Long-term outcomes of preterm birth 1.2.4.1 The Barker hypothesis In addition to the increased neonatal morbidity, survivors after preterm birth also face an increased risk of illness later in life. Barker et al discovered in the 1980´s that low birth weight as a surrogate marker of adverse fetal environment, was associated with increased risk of cardiovascular disease in adulthood.26 This has then been observed in numerous studies and it is now widely known that both low birth weight and prematurity are predisposing factors for developing pulmonary, cardiovascular and metabolic diseases as well as neurological and neurodevelopmental conditions later in life. Besides fetal malnutrition, excess exposure to glucocorticoids in utero could be a candidate mechanism underlying the associations between low birth weight and increased risk of cardiometabolic and other diseases in adulthood. 1.2.4.2 Cardiovascular and metabolic outcome after preterm birth Most studies reporting data in historical cohorts before the modernization of perinatal and neonatal medicine show no increased risk for coronary heart disease or hypertension among elderly people born preterm. 49,50 However, these results are biased because of the very selective survival after preterm birth during that time period. In the modern era of perinatal and neonatal medicine (after the 70´s) with a more universal survival, there are emerging numbers of studies showing an association between prematurity and increased risk of cardiovascular and metabolic diseases later in life. For example, preterm birth has been correlated to hypertension51-53, altered structure and function of the heart and the arterial tree54,55, diabetes56,57, increased levels of plasma low-density lipoproteins (LDL)53, overweight58 and stroke59,60 in adults. Since most women delivering preterm nowadays receive ACS, antenatal corticosteroid exposure could be one explanatory factor for these associations. In animals, ACS exposure has been associated with increased risk factors for cardiometabolic disease including enhanced fat deposition,61 impaired