Accuracy of Third Trimester Ultrasound for Predicting Large-for-Gestational Age Newborn in Women with Gestational Diabetes Mellitus

Abstract

Objective: To determine the accuracy of ultrasonography for predicting a large-for-gestational-age (LGA) newborn in women with gestational diabetes mellitus (GDM).
Materials and Methods: Singleton pregnancy, diagnosed with GDM were recruited. They underwent ultrasonography at 32-36 weeks’ gestation for fetal biometry. Biparietal diameter (BPD), head circumference (HC), abdominal circumference (AC), and femur length (FL). Estimated fetal weight (EFW) was derived from these 4 parameters by Hadlock formula. Delivery of an LGA newborn in women with the ultrasound finding of LGA fetus was the primary outcome.
Results: Of 345 studied women, 107 (31%) had an LGA newborn. EFW of ≥ 90 th percentile at third trimester ultrasonography was found in 13 women, all of whom had an LGA newborn. It had a positive predictive value (PPV), specificity, sensitivity and negative predictive value (NPV) of 100%, 100%, 12.1% and 71.7% respectively to predict LGA at birth. Considering each fetal parameter individually, AC > 90 th percentile and HC > 90 th percentile had odds ratios (OR) with 95% confidence intervals of the newborn being LGA of 6.5 (3.3-12.8) and 2.0 (1.0-4.0) respectively while EFW > 85 th percentile had the highest OR of 9.3 (1.1-77.9). Lowering cutoff values of EFW to 80 th and 70 th percentile increased the sensitivity and NPV for prediction of LGA at birth while reducing the PPV and specificity slightly.
Conclusion: EFW derived from the third trimester ultrasonography in GDM had high PPV and specificity with low to moderate sensitivity and NPV to predict an LGA newborn in GDM.