Comment on “Differences in Prediction of Adverse Perinatal Outcome in Term Pregnancies by Choice of Fetal Growth Reference: A Validation Study”

Abstract

Detecting abnormal growth of a fetus during pregnancy is important for clinicians to monitor and intervene when necessary. Multiple fetal growth references are used internationally to classify fetal size, including the World Health Organization (WHO) and INTERGROWTH-21st (IG21st) standards and the Swedish Maršál and Lindström references. The differences in expected fetal weight (EFW) for gestational age (GA) across these references can affect how well fetuses are identified and managed to prevent adverse perinatal outcomes. The aim of this study was to compare the four fetal weight references and assess which reference most accurately identifies infants at risk of perinatal mortality and morbidity. This was a population-based cohort study, using data from Swedish registries on antenatal, obstetric, neonatal care, and postnatal deaths. Included were 1,126,059 singleton, term births in Sweden from 2010 to 2020. Excluded were pregnancies with unknown GA or birthweight, GA <37+0 or >42+6 weeks, birthweight z-score >5 or <−5 SD from the mean for GA and sex, and structural malformations or chromosome aberrations. The 4 references included the Lindström reference, which covers the longest gestational span of 12 to 42 weeks (n=585 women in the study population); the Maršál reference, which begins at 25 weeks (n=86 women); and the WHO and IG21st growth, which end at 40 weeks (n=1362 and 4231, respectively). Small-for-gestational age (SGA) was defined as birthweight <3rd centile and <10th centile. Large-for-gestational age (LGA) was defined as EFW or birthweight >90th and >97th centile. Models to assess the predictive performance of the 4 references were designed to assess adverse perinatal outcomes, including perinatal death, serious neonatal morbidity, severe asphyxia, hypoxic ischemic encephalopathy, neonatal sepsis, neonatal resuscitation ≥10 minutes, admission to the neonatal ward, metabolic acidosis, Apgar score <6 at 10 minutes, or fetal distress requiring instrumental or cesarean delivery. Sensitivity and false-positive rates were calculated for SGA thresholds. The proportion of infants classified as SGA <3rd centile were 9.6% for the Lindström reference, 2.5% for Maršál, 1.9% for WHO, and 0.7% for IG21st. The proportion of infants classified >97th centile was 1.7% for Lindström, 4.1% for Maršál, 4.2% for WHO, and 14.4% for IG21st. Stillbirth and neonatal death occurred in 1.5 and 0.2 per 1000 births, respectively. When SGA was defined as <3rd centile, stillbirth sensitivity was 28.9% with the Lindström, 11.8% for Maršál, 11.3% for WHO, and 6.0% for IG21st, while false-positive rates ranged from 9.6% for Lindström to 0.7% for IG21st. A similar pattern was observed for detecting neonatal death and serious neonatal morbidity using SGA <3rd and <10th centiles, with the highest sensitivity and false-positive rate for Lindström and lowest for IG21st. In conclusion, all 4 references were comparable in predicting adverse perinatal outcomes. However, they differed in fetal size classification, sensitivity, and false-positive rates. The Lindström reference identified more at-risk infants but generated higher false-positive rates, while IG21st identified fewer at-risk infants and had lower false-positive rates. (Summarized from Lindström L, Ahlsson F, Axelsson Ove, et al. Differences in prediction of adverse perinatal outcome in term pregnancies by choice of fetal growth reference: A validation study. Acta Obstet Gynecol Scand. 2026; 105:466–478. doi: 10.1111/aogs.70136).