Gestational Glucose Intolerance and Risk of Future Diabetes

Abstract

Objective: Pregnant individuals are universally screened for gestational diabetes mellitus (GDM). Gestational glucose intolerance (GGI) (an abnormal initial GDM screening test without a GDM diagnosis) is not a recognized diabetes risk factor. We tested for an association between GGI and diabetes after pregnancy.

Research design and methods: We conducted a retrospective cohort study of individuals followed for prenatal and primary care. We defined GGI as an abnormal screening glucose-loading test result at ≥24 weeks' gestation with an oral glucose tolerance test (OGTT) that did not meet GDM criteria. The primary outcome was incident diabetes. We used Cox proportional hazards models with time-varying exposures and covariates to compare incident diabetes risk in individuals with GGI and normal glucose tolerance.

Results: Among 16,836 individuals, there were 20,359 pregnancies with normal glucose tolerance, 2,943 with GGI, and 909 with GDM. Over a median of 8.4 years of follow-up, 428 individuals developed diabetes. Individuals with GGI had increased diabetes risk compared to those with normal glucose tolerance in pregnancy (adjusted hazard ratio [aHR] 2.01 [95% CI 1.54-2.62], P < 0.001). Diabetes risk increased with the number of abnormal OGTT values (zero, aHR 1.54 [1.09-2.16], P = 0.01; one, aHR 2.97 [2.07-4.27], P < 0.001; GDM, aHR 8.26 [6.49-10.51], P < 0.001 for each compared with normal glucose tolerance). The fraction of cases of diabetes 10 years after delivery attributable to GGI and GDM was 8.5% and 28.1%, respectively.

Conclusions: GGI confers an increased risk of future diabetes. Routinely available clinical data identify an unrecognized group who may benefit from enhanced diabetes screening and prevention.

Graphical abstract

Figure 1

Final analysis cohort and exposure groups based on glucose tolerance status in pregnancy. A: Flowchart describing exclusion criteria used to determine final analysis cohort from the Maternal Health Cohort. B: Total numbers included in our final analysis cohort are shown for individuals and pregnancies. Total numbers of pregnancies in each glucose tolerance exposure group used for analyses are shown. GDM screening was done according to the two-step screening test at ≥24 weeks’ gestation with a 1-h GLT of 50 g glucose with a glucose cutoff of 140 mg/dL used at MGH. If the GLT glucose result was ≥140 mg/dL, a confirmatory 3-hour OGTT of 100 g glucose was performed. GGI (zero [GGI-0] or one [GGI-1] abnormal OGTT value) and GDM (two to four abnormal OGTT values) were diagnosed according to Carpenter-Coustan criteria. GA, gestational age in weeks. *Where delivery for the first pregnancy in the data set did not occur at the study center, but at least one subsequent delivery did occur at the study center, individuals were still included. For these individuals (N = 1,847), follow-up began at the first delivery at the study center and subsequent consecutive pregnancies with deliveries at the study center were included.

Figure 2

Kaplan-Meier curves depicting time to diabetes diagnosis according to glucose tolerance status in pregnancy. Kaplan-Meier curves (solid lines) with 95% CIs (dashed lines) shown of unadjusted data with time since first observed delivery in years on the x-axis and proportion without diabetes on the y-axis. A: Normal glucose tolerance (NGT) (control group) is shown in blue, GGI in purple, and GDM in gray. B: NGT (control group) is shown in blue, and GDM in gray. GGI is divided into its components: zero abnormal OGTT values (GGI-0) in orange and one abnormal OGTT value (GGI-1) in red.