Grand Rounds

How Accurate are Estimated Fetal Weight Ultrasound Measurements between 23 and 25 Weeks?

BACKGROUND AND PURPOSE:

Haragan et al (American Journal of Perinatology, 2018) evaluated the accuracy of estimated fetal weight (EFW) to predict birth weight (BW) among infants born during the periviable period
- Authors also assessed the accuracy of EFW in SGA versus AGA neonates

Retrospective study
- Women who delivered between 23w0d and 25w7d
- EFW obtained within 7 days of delivery
EFW based on Hadlock-3 formula (HC, BPD, FL and AC)
Accuracy of EFW defined as
- EFW − BW)/BW × 100%
Median absolute percentage difference between EFW and BW
- |(EFW − BW)/BW × 100|
Excess error for EFW was defined as the absolute percentage difference between EFW and BW ≥20%

226 neonates were involved in the study
- Mean gestational age at birth: 24w1d
- 11% SGA; 10% BW <500g
- 74% spontaneous preterm
- Median BW 653 g
Median absolute percentage difference between EFW and BW of fetal weight estimates was 9.2%
EFW overestimated BW in 75.7% of the cohort
19% of cohort had excess error of ≥20%
- More likely in scans that overestimated vs underestimated EFW (22.2% vs 7.3%)
Among SGA infants
- EFW was a poorer predictor of BW among SGA infants
- Median absolute difference was greater among SGA infants (12.3%) vs AGA (8.6%), but did not achieve significance; p = 0.062
- Mean percentage difference was 16.2 ± 19.4% in SGA versus 6.9% ± 13.1% in AGA infants; p = 0.019
Infants <500 g
- Median absolute difference was significantly higher in infants <500g (15.2%) vs >500 g (8.6); p = 0.021
Mean percentage difference was greater among infants <500 g (18.2 ± 19.3%) versus 6.7 ± 13.0%; p = 0.003

EFW is a “reasonable proxy” for BW in periviable fetuses
There are important limitations, especially in SGA or <500g
- Ultrasound is likely to underestimate the true BW in these scenarios
Authors suggest that alternate ultrasound growth formulae should perhaps be developed, particularly infants predicted to be less than <500g

< Previous

Next >