226: the Accuracy of Targeted-Sga Fetal Weight Estimation – Results from the Multicentre PORTO Study

To determine if there is increased accuracy in switching to a targeted-SGA /LBW formula for fetal weight estimation in an SGA cohort. This is a secondary analysis of The Prospective Observational Trial to Optimize Paediatric Health. The PORTO Study recruited 1,200 ultrasound-dated singleton IUGR pregnancies, defined as EFW <10th centile (Hadlock IV, 1985), between 24+0 and 36+6 weeks gestation. All recruited pregnancies underwent serial sonographic assessment of fetal weight and multi-vessel Doppler studies until birth. Stillbirths and neonatal deaths were excluded from this analysis of fetal weight assessment, obtained within 1 week of delivery. Eight fetal weight formulae developed specifically for SGA fetuses or low birthweight (LBW) infants were compared, in addition to the Hadlock III & IV (1985) and the Intergrowth-21st (2017) formulae. The accuracy of the formulae was assessed using percentage error when compared to birthweight. A prediction model was also used to account for potential fetal growth during the interval to delivery. Complete fetal biometry within one week of delivery (median of 3 days) was available in 588 pregnancies recruited with an EFW < 10th centile. The mean birthweight at delivery was 2,311g (SD=647g) and the median gestational age at delivery was 37.7 weeks (range=26.4- 42.6 weeks). Boxplots of percentage error are presented in figure 1. The Hadlock III, IV and Sabbagha III formulae were similar in having almost no systematic bias (median percentage error < 1%). All three formulae had similar random errors (inter-quartiles ranges 10-13%) and had the highest proportion of estimated fetal weights within 10% of birthweight (77%). Percentage error was dependent upon both gestational age at scan (p<0.001) and time interval to delivery (p<0.001). Results using projection to delivery methods improved the accuracy of the other formulae but did not change the conclusions (Table 1). A retrospective analysis restricting to those with birthweight < 10th centile (n=442) found similar results. A sub-analysis in those with abnormal umbilical artery Doppler (n=303) also found similar results. Our study, which recruited pregnancies with suspected SGA (EFW < 10th centile using the Hadlock IV formula), suggests that there is no benefit to subsequently switching to formulae that is targeted for the SGA fetus or low birthweight infant.View Large Image Figure ViewerDownload Hi-res image Download (PPT)