Spontaneous abortion : risk factors and measurement of exposures

Abstract: The overall objectives of this thesis were to increase the understanding of the etiology of spontaneous abortion and repeated spontaneous abortion, and to evaluate possibilities of exposure assessment during pregnancy. Data from a Swedish population-based case-control study of spontaneous abortion, performed in Uppsala County 1996-98, formed the base for the three first studies of the thesis. The final study population included 562 cases with a spontaneous abortion in gestational week 6-12 and 1,028 pregnant controls frequency-matched to the cases according to week of gestation. Exposure information was assessed through in-person interviews using structured questionnaires, and through plasma measurements of folate and cotinine levels. Intrauterine tissue was sent for chromosomal analyses among cases with incomplete spontaneous abortion. Conditional logistic regression models were used to calculate odds ratios to estimate the relative risk of spontaneous abortion. Women with folate levels <5 nmo/l were at 60% increased risk of spontaneous abortion as compared with women with folate levels 5.0-8.9 nmol/1. High maternal folate levels were associated with a non-significant trend towards a protective effect. Folate deficiency was predominantly associated with abnormal fetal karyotype, however we could not show a differential effect of folate between normal and abnormal fetal karyotype abortions. The use of a folic acid supplement, reported by <5% of all women, was not associated with a statistically significant increased risk of spontaneous abortion. The association between sociodemographic and anthropometric factors, obstetric history, and life style factors, with respect to risks of repeated spontaneous abortion, were explored among 108 cases with two or more consecutive first trimester spontaneous abortions and 583 controls with at least two pregnancies. High and low maternal age, and previous spontaneous abortions were found to be major risk factors. Smokers were at increased risk of repeated spontaneous abortion. Among nonsmoking women with high caffeine intake there was an increased risk, whereas there was no such association among smokers. Folate deficiency was not associated with increased risk of repeated spontaneous abortion. Measurement of plasma cotinine was used to assess the dose received from exposure to environmental tobacco smoke (ETS) and active smoking. The prevalence of ETS exposure (cotinine 0.1-15 ng/ml) was 24% among women with spontaneous abortion and 19% among women with a normal pregnancy in the first trimester. Nonsmoking pregnant women exposed to ETS faced a 70% increased risk of spontaneous abortion as compared with unexposed women. Active smoking was associated with a more than two-fold increased risk of spontaneous abortion. ETS exposure was associated with normal, abnormal, and unknown fetal karyotype abortions. In a prospective follow-up of 953 women from early pregnancy until delivery, maternal selfreported information on active smoking, smoking cessation during pregnancy, and ETS exposure during pregnancy, was validated using cotinine measurements as gold standard. Sensitivity, specificity, positive and negative predictive values, and likelihood ratios were calculated as measures of accuracy in gestational weeks 6-12 and 31-34. The validity of selfreported active smoking in early and late pregnancy was reasonably high. Among women reporting smoking cessation before the first interview and between the first and second interview, 13% and 25% misreported active smoking, respectively. Self-reported information on ETS exposure misclassified most women as unexposed throughout the pregnancy.