Exposures in utero and chronic disease : an alternative methodological approach

Abstract

The aims of this thesis were to evaluate maternal and birth characteristics as risk factors for breast cancer and cardiovascular disease later in life, and to assess maternal serum concentrations of insulin-like growth factors in pregnancies complicated by preeclampsia.

Birth characteristics and risks of breast cancer were studied in a case-control study nested within the population-based Swedish Twin Register (Paper 1). Female same-sexed twin pairs discordant for breast cancer (i.e., only one twin of the pair has the disease) were identified by record linkage between the Swedish Twin Register and the Cancer Register. Birth characteristics of 96 discordant twin pairs and 86 individual (i.e., not related) twins were obtained from birth records. In comparisons between breast cancer cases and age-matched external twin controls an increased risk of breast cancer was found among twins with gestational age of more than 40 weeks, compared to twins with gestational age less than 33 weeks. In within-pair comparisons between breast cancer cases and co-twin controls mean birth weight and ponderal index were higher in cases compared to co-twin controls. The risk of breast cancer also increased with increasing birth weight. Our results suggest an effect of growth and/or hormonal factors in utero, reflected by birth characteristics, on risk of breast cancer later in life.

To investigate birth characteristics as risk factors for acute myocardial infarction (Paper 11), a similar study design was used as in Paper 1. Same-sexed twin pairs discordant for acute myocardial infarction were identified by record linkage between the Swedish Twin Register, the Cause of Death Register, and the Hospital Discharge Register. In all, birth characteristics of 132 discordant twin pairs and 118 individual twins were obtained from birth records. In comparisons between cases and sex and age-matched external (i.e., not related) twin controls, cases had significantly lower birth weight, birth length, and head circumference than external twin controls. In within-pair comparisons, no significant differences in birth characteristics were found between cases and co-twin controls. These results suggest that genetic and/or early environmental factors could influence the association between birth characteristics and risk of acute myocardial infarction in adulthood.

To assess the association between birth weight and angina pectoris, we conducted a study using self-reported data of 4594 same-sexed twins participating in a complete screening of the Twin Register, of whom 381 individuals reported angina pectoris (Paper 111). Low birth weight (<2.0 kg) was associated with increased risk of angina pectoris in the twin cohort, when data were adjusted for potential confounders the risk decreased and did not reach significance. Dizygotic (n=55) and monozygotic (n=37) twin pairs discordant for angina pectoris were analysed separately. Low birth weight was significantly associated with an adjusted increased risk of angina pectoris within dizygotic twin pairs, but not within monozygotic twin pairs. These results indicate that there are genetic factors associated with both fetal growth and risk of angina pectoris, yet the results of the within-pair comparisons are based on small numbers.

History of preeclamptic pregnancy reduces the risk of breast cancer both in the mother and the female offspring, and it has been suggested that lower levels of insulin-like growth factors (IGF) and higher levels of insulin-like growth factor binding proteins (IGFBP) play an important role. To evaluate the possible role of these factors in the aetiology of preeclampsia, we measured maternal serum concentrations of IGF-I, IGF-II, and IGFBP-3 in week 17 and 33 of gestation in 30 preeclamptic and 128 non-preeclamptic women (Paper IV). Results showed no significant differences in serum concentrations of IGF-I and IGFBP-3 neither in week 17 nor in week 33. Preeclamptic women had significantly higher serum levels of IGF-II in week 33, but there was no difference in week 17. We found no evidence for lower levels of IGF-I and IGF-II, or higher levels of IGFBP-3 in pregnancies complicated by preeclampsia before clinical signs of disease.