1409-P: Using Fasting Venous Plasma Glucose (FVPG) to Stratify Risk in Women with Gestational Diabetes (GDM)

Powe et al described variable “defects” (defined as parameter values <25^th centile for non GDMs) in OGTT derived insulin secretion (I_secr) and sensitivity (I_sens) estimates in GDM. They reported that pregnancy complications relate primarily to low I_sens. To explore applicability to routine OGTT data, an analogous but simplified analysis was undertaken, classifying IADPSG GDM women according to FVPG < or ≥ 75^th centile for non GDM women (cut-off 4.6 mmol/L - near the cohort mean (SD) of 4.5(0.4) mmol/L). Our multi-ethnic cohort included a diverse group of 6128 untreated women from 5 centers in the blinded epidemiologic HAPO study. Analyses compared outcome frequencies (LGA, excess neonatal (NN) adiposity, NN hypoglycemia, NN hyperinsulinemia, preterm birth, primary C section) with non GDM women and between GDM-FVPG groups using χ² tests and logistic regression analyses adjusted for maternal age, height, smoking, FH diabetes, BMI at OGTT visit and ethnicity (Table 1). The 25% of GDM women with FVPG < 4.6 mmol/L were at much lower risk of adverse outcomes. Although “mechanistic” classification considering I_secr and I_sens may offer insights for treatment, a simpler classification by FVPG performs better and appears simpler for routine use. A modified “two-step” approach for detection and classification of pregnancy hyperglycemia, using FVPG as the first step, deserves consideration.