Differences in prediction of adverse perinatal outcome in term pregnancies by choice of fetal growth reference: A...

Our objectives were to evaluate the association between fetal growth abnormalities and adverse perinatal outcomes in term pregnancies using four different fetal growth references: the recently published Swedish references by Lindström et al., the currently used Swedish references by Maršál et al., and the international standards by the WHO and INTERGROWTH-21st (IG21st). The study aimed to evaluate the performance of each reference and determine which reference most accurately identifies small for gestational age (SGA) infants at risk of perinatal mortality and morbidity.

This population-based cohort study included 1 126 059 singleton term births in Sweden from 2010 to 2020. Data were obtained from national registers, including the Swedish Medical Birth Register and the Swedish Neonatal Quality Register. Birthweight centiles were calculated using each growth reference. Adverse perinatal outcomes were categorized by severity and included stillbirth, neonatal death, and serious neonatal morbidity. Logistic regression models were used to assess predictive performance, and sensitivity and false positive rates (FPR) were calculated for SGA thresholds (<3rd and <10th centiles).

The distribution of birthweight centiles varied significantly across references. For SGA <3rd centile, the rate ranged from 9.6% for Lindström, 2.5% for Maršál, 1.9% for WHO, to 0.7% for IG21st. All references showed similar overall predictive performance (C-index ≈ 0.67) but with different discriminatory ability. The predicted risk of perinatal death increased at lower centiles for the Lindström reference than for the Maršál and WHO references, and at higher centiles for the IG21st reference. The Lindström reference identified the highest proportion of infants as SGA and had the highest sensitivity but also the highest FPR for detecting adverse outcomes. The IG21st reference classified the smallest proportion as SGA, resulting in the lowest sensitivity and FPR.

While all fetal growth references showed comparable predictive ability for adverse perinatal outcomes, they differed substantially in sensitivity and FPR. When the top priority is to identify as many at-risk fetuses as possible, Lindström et al.'s reference seems to be the best choice. However, when the top priority is a balanced sensitivity versus FPR, the WHO reference seems most suitable for clinical practice in this population of term births.