Once gestational diabetes, always gestational diabetes? Maternal and neonatal outcomes of pregnancies with gestational diabetes preceding nongestational diabetes pregnancy: a retrospective cohort study

BACKGROUND: There is limited information about maternal and neonatal outcomes of pregnant individuals without gestational diabetes mellitus who were diagnosed with gestational diabetes mellitus in the preceding pregnancy. OBJECTIVE: This study aimed to investigate whether individuals previously diagnosed with gestational diabetes mellitus but not in subsequent pregnancy present similar diabetes-related complications despite the lack of diagnosis. STUDY DESIGN: A retrospective cohort design was employed, including individuals with at least 2 consecutive births at a tertiary medical center between July 2012 and December 2022. Participants were categorized into 3 groups: individuals with gestational diabetes mellitus in both pregnancies (group 1), individuals with gestational diabetes mellitus in the first pregnancy but not in the subsequent pregnancy (group 2), and individuals with no gestational diabetes mellitus in both pregnancies (group 3). The groups were compared for various diabetes-related complications. RESULTS: The study population was composed of 19,703 individuals. Group 2 showed higher rates of macrosomia (odds ratio, 1.40 [95% confidence interval, 1.01-1.92]; P=.03), large for gestational age (odds ratio, 1.60 [95% confidence interval, 1.20-2.00]; P<.01), and preeclampsia (odds ratio, 2.35 [95% confidence interval, 1.32-4.15]; P<.01) than group 3. The rates of large for gestational age, macrosomia, and preeclampsia were similar between groups 1 and 2. The composite maternal-neonatal adverse outcome was significantly elevated in groups 1 and 2 than in group 3 (odds ratios: 1.48 [95% confidence interval, 1.23-1.77; P<.01] and 1.91 [95% confidence interval, 1.58-2.3; P<.01], respectively). In a multivariate regression analysis, the adjusted odds ratios (accounting for a body mass index before the second birth, age at the second birth, and parity) for composite maternal-neonatal outcomes were 1.75 (95% confidence interval, 1.36-2.25; P<.01) in group 1 and 1.30 (95% confidence interval, 1.01-1.67; P=.03) in group 2 compared with group 3. The elective cesarean delivery rate was higher in groups 1 (22%) and 2 (13%) than in group 3 (8.7%). However, the rate was also significantly higher in group 1 than in group 2. The preterm birth rates were higher in group 1 (odds ratio, 1.70 [95% confidence interval, 1.20-2.30]; P<.01) but not in group 2 compared with group 3. CONCLUSION: Individuals with a history of gestational diabetes mellitus in a previous pregnancy but not in subsequent pregnancy are at increased risk of diabetes-related complications, including preeclampsia, macrosomia, and large for gestational age. Our findings suggest that these individuals may have underlying insulin resistance and other, still occurring, risk factors, and their absence of a gestational diabetes mellitus diagnosis does not eliminate the risk of adverse outcomes.