Antenatal oral glucose tolerance test in women with gestational diabetes mellitus: fasting plasma glucose is the best predictor of both large for-gestational-age newborns and postpartum glucose tolerance.

Abstract

Numerous studies have evaluated the predictors of large-for-gestational-age newborns (LGA) as well as abnormal glucose tolerance (AGT) after delivery in women with gestational diabetes mellitus, one at a time. The present study assesses the ability of glucose values on the antenatal oral glucose tolerance test (OGTT) to predict both outcomes. Retrospective analysis of a prospectively collected database. Gestational diabetes was diagnosed with a two-step approach. After delivery, reevaluation was performed using 2006 WHO definition of glucose tolerance. 1) singleton pregnancies; 2) available information on: offspring weight for age, antenatal and follow-up OGTT and additional potential predictors of LGA and AGT. The study included 1241 women. Data were analyzed by multivariate logistic regression and ROC analyses. Logistic regression models were used to calculate adjusted and unadjusted odds ratios and corresponding 95% CI for 1 mmol/L increase in each glucose value (FPG, 1 h, 2 h, 3 h) of the diagnostic 100-g OGTT, associated to the main outcomes LGA and postpartum AGT. Analyzing each glucose value on antenatal OGTT, fasting plasma glucose (FPG) showed the highest OR for both outcomes: LGA (1.537, 95% CI 1.237-1.910) and AGT (2.288, 95% CI 1.869-2.801). On the adjusted analysis the association was attenuated, but FPG continued to be the best predictor for both outcomes. In women with gestational diabetes mellitus, FPG on the diagnostic OGTT was the glycemic value that best predicted both LGA and after delivery AGT.