Comparisons of the accuracy of fetal weight estimation between various ultrasound models and clinical assessments in term pregnancies complicated by diabetes mellitus

Abstract

© JOURNAL OF THE MEDICAL ASSOCIATION OF THAILAND | 2019. Objective: To compare the accuracy of fetal weight estimation on admission before delivery between subjective assessment, objective clinical assessment, and 2D-ultrasonography (2D-US) assessment in pregnancy with diabetes mellitus (DM). Materials and Methods: Pregnancy with DM admitted for delivery at Maharaj Nakorn Chiang Mai Hospital between March 24, 2017 and August 7, 2018, were prospectively recruited. Inclusion criteria were a term, singleton pregnancy with overt or gestational DM without other underlying diseases, no smoking, and no emergency condition for immediate delivery. Fetal weight estimation by subjective assessment (physical examination), objective assessment (a formula using multiple parameters including BMI, fundal height, maternal abdominal circumference, and parity), and 2D-US were performed on admission for delivery. Actual birth weight (BW) was measured within 30 minutes after birth. Accuracy in predicting actual BW by various methods was calculated and expressed as intra-class correlation coefficient (ICC), mean error, mean percentage error, and percentage of estimate within 10% of the actual BW. Results: Ninety-six women (gestational DM 92 and overt DM 4) were recruited. The 2D-US (Hadlock 4) was most accurate for estimation (ICC 0.896) and conversely objective assessment was least accurate (ICC 0.610). In predicting actual BW (±10%), 2D-US and subjective assessment were significantly superior to objective assessment (p=0.001 and p=0.017; respectively), whereas the accuracy of 2D-US and subjective assessment was comparable (76.4% versus 66.7%; p=0.151). Conclusion: In singleton pregnancy with DM, the fetal weight estimation with 2D-US (Hadlock 4) had more accuracy than subjective assessment (in terms of ICC), but no significant difference of accuracy between the two methods. The authors suggest using 2D-US as an adjunct in estimation of fetal weight among pregnancies with DM.