Insulin Resistant Gestational Glucose Intolerance Is Associated With Adverse Perinatal Outcomes

Background: Women with gestational diabetes mellitus (GDM) and gestational glucose intolerance (GGI, abnormal initial GDM screening test) and their infants have an increased risk of adverse perinatal outcomes including large for gestational age birth weight (LGA), pregnancy-related hypertension, neonatal intensive care unit (NICU) admission, and cesarean delivery. We expanded a prior analysis defining physiologic subtypes of GGI categorized by insulin resistance, insulin deficiency, or mixed pathophysiology. We aimed to determine if GGI subtypes are at differential risk for adverse outcomes. Methods: We applied homeostasis model assessment (HOMA2) to fasting glucose and insulin levels at 16–20 weeks’ gestation to assess insulin resistance and deficiency, defined using the 50(th) percentile in 220 women with a normal glucose loading test (GLT) at 24–30 weeks’ gestation. We defined GGI as GLT 1-hr glucose ≥140 mg/dL (n=245) and normal glucose tolerance (NGT) as GLT 1-hr glucose <140 mg/dL (n=1538). We classified women with GGI into subtypes according to the presence of insulin resistance and/or deficiency. We compared odds of adverse outcomes in each subtype to odds in women with NGT using logistic regression with adjustment for age, race/ethnicity, marital status, and 1(st) trimester BMI, plus infant sex in LGA models. Results: Of women with GGI, 49.0% had the insulin resistant subtype (IR, n=120), 30.6% had the insulin deficient subtype (ID, n=75), 15.9% had mixed pathophysiology (MP, n=39), and 4.5% had no evidence of IR or ID (n=11). GLT results and GDM diagnosis were similar among GGI subtypes. We found increased odds of LGA (primary outcome) in women with IR compared to women with NGT (OR 1.97 [1.17–3.32], p=0.01) in an unadjusted model; this was attenuated in an adjusted model with BMI (adjusted OR 1.43 [0.82–2.49], p=0.21). There was a trend toward increased odds of LGA in women with ID (adjusted OR 1.87 [0.92–3.80], p=0.09) and no increased odds in women with MP (adjusted OR 1.33 [0.50–3.57], p=0.57) compared to NGT. The odds of pregnancy-related hypertension in the IR subtype were increased (adjusted OR 1.68 [1.02–2.77], p=0.04) compared to women with NGT; women with ID (adjusted OR 0.91 [0.44–1.88], p=0.79) or MP (adjusted OR 1.13 [0.48–2.67], p=0.78) did not have increased odds. Neither infants of women with IR nor ID had increased odds of NICU admission overall, yet among women with BMI <25, infants of those with IR had increased odds of NICU admission compared to those of women with NGT (adjusted OR 3.37 [1.04–10.96], p=0.02); odds of NICU admission were not increased in infants of women with ID and BMI <25 (adjusted OR 0.50 [0.07–3.83], p=0.50). There was no difference in cesarean delivery across subtypes. Conclusion: Insulin resistant GGI is a high-risk subtype for adverse perinatal outcomes. Using HOMA2 to delineate subtypes may provide opportunities for a personalized approach to GGI/GDM.