Performances of birthweight charts to predict adverse perinatal outcomes related to SGA in a cohort of nulliparas

Abstract

Background: Small-for-gestational-age neonates (SGA) are at increased risk of neonatal morbidity. Nulliparity represents a risk factor for SGA; birthweight charts may perform differently for the detection of SGA among nulliparas. This study aimed at describing the prevalence of SGA in nulliparas according to different birthweight charts and evaluating the diagnostic performance of these charts to maternal and perinatal outcomes. Methods: This is a secondary analysis of a Brazilian cohort of nulliparas named Preterm SAMBA study. Birthweight centiles were calculated using the Intergrowth-21st, WHO-Fetal Growth Charts, Birth in Brazil population chart and GROW-customised chart. The risks of outcomes among SGA neonates and their mothers in comparison to neonates with birthweights between the 40th-60th centiles were calculated, according to each chart. ROC curves were used to detect neonatal morbidity in neonates with birth weights below different cutoff centiles for each chart. Results: A sample of 997 nulliparas was assessed. The rate of SGA infants varied between 7.0–11.6%. All charts showed a significantly lower risk of caesarean sections in women delivering SGA neonates compared to those delivering adequate-for-gestational-age neonates (OR 0.55–0.64, p <.05). The charts had poor performance (AUC 0.492 – 0.522) for the detection of neonatal morbidity related to SGA born at term. Conclusion: The populational and customised birthweight charts detected different prevalence of small-for-gestational-age neonates and showed similar and poor performance to identify related neonatal adverse outcomes in this population.