Hypothesis: Gestational diabetes mellitus (GDM) is a disorder of glucose metabolism during pregnancy characterized by pancreatic beta cell dysfunction and greater insulin resistance, but it is unclear whether dysfunction exists before pregnancy. The disposition index (DI) is a physiologic measure of beta cell compensation for insulin resistance strongly predictive of future diabetes. This prospective study evaluates whether a clinical approximation of DI before pregnancy is associated with risk of GDM.
Methods: This analysis included 696 women (45% black, 55% white) enrolled in the CARDIA Study, a U.S. multi-center prospective cohort of young adults aged 18-30 at baseline (1985-86) who gave birth at least once during 30 years of follow up, reported GDM status and had fasting glucose and insulin measured before one or more post-baseline births. DI was defined as HOMA-B divided by HOMA-IR using standard formulas. Multinomial logistic regression models estimated odds ratios (OR) and 95%CI for GDM among pre-pregnancy DI tertiles (low, moderate, high) and fully adjusted for time to birth, race, age, parity, BMI, lifestyle behaviors and family history of diabetes, and also stratified by pre-pregnancy BMI.
Results: 9% of women reported GDM (64/696) for 794 births. 55% of GDM and 30% of non-GDM were categorized as low DI. Low pre-pregnancy DI compared to moderate DI was associated with higher fully adjusted odds of GDM (OR=2.71, 95%CI:1.37-5.35) in the entire sample. In models stratified by pre-pregnancy BMI, low DI was associated with 4-fold higher odds of GDM among Overweight/Obese (OR=4.22, 95%CI: 1.35-13.91) and somewhat attenuated higher odds of GDM among Normal BMI (OR=1.94, 95%CI: 0.78–4.86); Table 1. Only family history of diabetes was strongly associated with GDM independent of DI.
Conclusions: Inadequate beta cell compensation is present before pregnancy and discriminates greatest risk of GDM among high BMI, and may identify higher risk among women of normal BMI.