Abstract
BACKGROUND: The inverse association between prenatal smoking and preeclampsia is puzzling, given the increased risks of prematurity and low birthweight associated with both smoking and preeclampsia. We analyzed the Norwegian Mother and Child Birth Cohort (MoBa) to determine whether the associations varied by timing of prenatal smoking. METHODS: We conducted an analysis of 74,439 singleton pregnancies with completed second- and third- trimester questionnaires. Active and passive smoke exposure by trimester were determined by maternal self-report, and covered the period of preconception through approximately 30 weeks' gestation. Adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. RESULTS: Rates of active smoking declined dramatically during pregnancy: for trimester 1, 23%; trimester 2, 9%; and trimester 3, 8%. Active smoking in the third trimester was associated with reduced odds of preeclampsia and gestational hypertension, with the strongest association among continuous smokers (for preeclampsia, OR= 0.57 [95% CI = 0.46-0.70]). Women who quit smoking before the third trimester had approximately the same risk of preeclampsia and gestational hypertension as nonsmokers. There was some evidence of dose-response, with the heaviest smokers (more than eight cigarettes per day) having the lowest risks of preeclampsia (0.48 [0.32-0.73]) and gestational hypertension (0.51 [0.28-0.95]). There was little evidence of an association with passive smoking exposure. CONCLUSION: The association between smoking and preeclampsia varies substantially according to the timing and intensity of exposure. A better understanding of the biologic pathways that underlie these associations may provide important clues to the etiology of preeclampsia and the development of effective clinical interventions.