Prediabetes and perinatal mortality

OBJECTIVE - The association between gestational diabetes mellitus (GDM) and perinatal outcome is largely based on case series and retrospective studie s that found an increased risk of perinatal mortality and stillbirth as the onset of diabetes approached. Our objective was to assess the relationship between latency to diabetes and perinatal outcome of prediabetic pregnanci es in a contemporary population of women with adult-onset diabetes. RESEARCH DESIGN AND METHODS - A population of 403 diabetic women from two r ecruitment sites completed a pretested questionnaire. RESULTS - Details of 1,181 pregnancy outcomes were obtained. This comprised 1,024 live births, 22 stillbirths, and 8 early neonatal deaths. Crude anal ysis suggested a relationship between time to diabetes (latency) less than or equal to 20 years and both perinatal death and stillbirth: odds ratio (9 5% CI), 2.41 (1.17-4.95) and 2.15 (0.93-4.98). Generalized additive modelin g revealed a nonlinear relationship between the variables time to diabetes, and maternal age and perinatal outcome. Final logistic regression analysis was then performed for the outcomes perinatal death and stillbirth, with m aternal age as a second-degree polynomial, year of birth as a continuous va riable, and time to diabetes dichotomized less than or equal to 20 years to diagnosis and >20 years. This final analysis documented a significant asso ciation between time to diabetes less than or equal to 20 years and both pe rinatal death (4.06 [1.79-9.36]) and stillbirth (3.35 [1.25-9.05]). CONCLUSIONS - There appeared to be an increased risk of perinatal perinatal death and stillbirth in pregnancies occurring in the last 20 years before the diagnosis of diabetes.