Fetal programming and subsequent risks in adulthood : Are the associations confounded by genetic and/or environmental factors?

Sammanfattning: The fetal origins hypothesis proposes that fetal growth is inversely associated with subsequent adult disease risk. The objective of this thesis was to investigate whether familial (shared environmental and genetic) factors confound the association between birth weight and risks of hypertension, low intellectual performance, and type-2 diabetes. We performed three large population-based studies of singleton born boys conscripted for military service (Papers I-III) to investigate the association between birth characteristics and risk of high systolic blood pressure and low intellectual performance in early adulthood. We found that males born light for gestational age (<-2 standard deviation scores [SDs]) were at a 14 % increased risk of high systolic blood pressure and 22 % increased risk of low intellectual performance, after adjusting for social and maternal factors (Papers I and II). In sibling analyses we found that the association between birth weight for gestational age and risk of high systolic blood pressure was similar within and between full siblings. In contrast, the associations between birth weight, birth length, and head circumference and risks of low intellectual performance were weaker within siblings, compared with between siblings. Thus, whereas the association between birth weight for gestational age and risk of high systolic blood pressure is independent of familial factors, the associations between measurements of fetal growth and risk of low intellectual performance are at least partly confounded by familial factors. Furthermore, it appears that during early stages of gestation, birth length and head circumference is of greater concern for intellectual development than birth weight, whose importance may increase with gestation (Paper III). To investigate if the association between birth weight and risk of hypertension in adulthood is confounded by shared environmental or genetic factors, we performed a study on the association in a sample of middle aged and elderly Swedish like-sexed twins. We found that a 500-g decrease in birth weight was associated with a 42 % increased risk of hypertension in the whole cohort of twins. Co-twin control analyses showed that corresponding risks within dizygotic and monozygotic twin pairs were 34 % and 74 %, respectively. The results suggest that the inverse association between birth weight and hypertension is independent of shared familial environment and genetic factors. To assess whether there is a familial link between birth weight and type-2 diabetes, we studied the association between offspring birth weight for gestational age and parental risk of type-2 diabetes in the twin sample used in Paper IV. Decreasing offspring birth weight for gestational age (with 1 SDs) was associated with a 72 % increased risk of type-2 diabetes among fathers, and 57 % decreased risk among mothers, independent of measured social factors. Furthermore, we found that both the mother s and father s risk of type-2 diabetes associated with decreasing offspring birth weight was similar within and between twin pairs, although slightly smaller within pairs effects were found among fathers. Thus, the well established association between paternal type-2 diabetes and offspring birth weight seems to primarily be due to unique environmental factors experienced by each twin and its offspring.