Increasing Maternal Body Mass Index and the Accuracy of Sonographic Estimation of Fetal Weight Near Delivery

• Published in: Journal of ultrasound in medicine Vol. 33; no. 12; pp. 2173 - 2179
• Main Authors: Kritzer, Sara, Magner, Kristin, Warshak, Carri R.
• Format: Journal Article
• Language: English
• Published: England American Institute of Ultrasound in Medicine 01.12.2014
• Subjects: Birth Weight - physiology; Body Mass Index; estimated fetal weight; Female; Fetal Weight - physiology; Humans; Image Interpretation, Computer-Assisted - methods; macrosomia; obesity; obstetric ultrasound; Ohio; Pregnancy - physiology; Pregnancy Trimester, Third - physiology; Reproducibility of Results; Retrospective Studies; Sensitivity and Specificity; sonography; Ultrasonography, Prenatal - methods; Ohio; estimated fetal weight; macrosomia; obstetric ultrasound; obesity; sonography
• ISSN: 0278-4297; 1550-9613
• DOI: 10.7863/ultra.33.12.2173

Objectives To evaluate whether an increasing body mass index (BMI) influences the accuracy of sonographic estimation of fetal weight. Methods We performed a retrospective cohort study of singleton deliveries over a 2‐year period in a single institution. Patients were included if they had a fetal weight estimation within 2 weeks of delivery. The Δ estimated fetal weight (EFW) was calculated by subtracting the sonographic EFW from the birth weight and compared among our study groups, which were based on the maternal BMI class. We also compared the absolute percentage error of estimation, rate of substantial error greater than 20%, rate of underestimation, and ability to predict fetal weight greater than 4000 g. Post hoc power analysis determined that our study group of 1200 patients allowed for an α of .05 and β of .90. Results We included 1177 women in our analysis. The median ΔEFW varied between study groups: 137, 202, 157, 200, and 189 g, respectively, in normal‐weight, overweight, and obese classes 1, 2, and 3 (P = .01). The median percentage error of estimation between study groups varied between 5.0% in normal‐weight women and 7.1% in class 2 obese women (P= .05). The rate of substantial error was similar between study groups and varied between 2.7% in class 1 obese women and 4.3% in normal‐weight and class 2 obese women. Linear regression analysis showed a weak association between maternal BMI and ΔEFW (R2 = 0.005; r = 0.069). Conclusions The absolute ΔEFW was lower in normal‐weight women; however, the percentage error of the EFW was similar between women of varying BMI classifications, as was the rate of substantial error and the rate of underestimation of the EFW.