Macrosomia – a new formula for optimized fetal weight estimation

Objectives: The aim of this study was to develop and to test a specific formula for estimating weight in the macrosomic fetus. Methods: This retrospective cross sectional study included 427 singleton fetuses with a birth weight of ≥ 4000g at the Perinatal Centre at Erlangen University Hospital. Ultrasound estimations of fetal weight were carried out within 1 week of delivery. Exclusion criteria were multiple pregnancy, intrauterine death, and major structural or chromosomal anomalies. Stepwise regression analysis with maternal and fetal biometric parameters was used to identify the best-fit formula for predicting fetal weight at birth. After a new formula had been developed in a formula-finding group (n=155), it was compared with commonly used weight equations (evaluation group, n=272). Results: The new formula — Estimated fetal weight (EFW)=–3906.596 + 0.001 * (FL)3 + 2742.512 * log(AC) + 335.169 * log(maternal weight) — proved to be superior to established equations, with the smallest mean error (–86±202g SD), the smallest mean percentage error (PE) (–1.8±4.5% SD), and the lowest mean absolute percentage error (APE) (3.6%; range: 0%–14.8%) when studied in the evaluation group. With the new formula, 72.4% of estimates fell within±5% of the actual weight at birth, 93.8% within±10%, and 100% within±15%. Conclusions: The new formula allows better weight estimation in the macrosomic fetus.

macrosomia - ultrasound - fetal biometry - fetal weight - regression formula