Background: The optimal threshold of birthweight discordance (BWD) remains controversial. This study aimed to evaluate the associations between BWD at different thresholds and early neonatal outcomes and to assess their predictive accuracy.
Methods: This was a retrospective cohort study using a birthweight data with the chorionicity information of 2348 liveborn twin pairs at a gestational age of ≥26 weeks, from 2012 to 2018. The percentage of BWD was calculated by dividing the actual birthweight difference by the weight of the larger twin and multiplying by 100. Outcomes of interest included neonatal intensive care unit (NICU) admission, neonatal respiratory distress syndrome (NRDS), ventilator support and a composite outcome combining major morbidities and neonatal death. Logistic regression models were performed to estimate the association between neonatal outcomes and BWD with different thresholds (≥15.0%, ≥20.0%, ≥25% and ≥ 30%). Generalized estimated equation (GEE) models were used to address intertwin correlation. Restrictive cubic spline (RCS) models were established to draw the dose-response relationship between BWD and the odds ratios of outcomes. Clustered receiver operating characteristic (ROC) curve analyses were performed to assess the predictive accuracy.
Results: Of 2348 twin pairs, including 1946 dichorionic twin pairs and 402 monochorionic twin pairs, BWD was significantly associated with NICU admission, regardless of the thresholds used. The incidence of NRDS, ventilator support and the composite outcome were significantly higher when a threshold of ≥20% or greater was chosen. The dose-response relationship showed nonlinear growth in the risk of adverse neonatal outcomes with increasing BWD. ROC analyses showed a low significant AUROC of 0.569 (95% CI: 0.526-0.612) for predicting NICU admission but no significant AUROCs for predicting other outcomes. A BWD of ≥30% provided a moderate increase in the likelihood of NICU admission [positive likelihood ratio (LR+) = 5.77].
Conclusion: Although BWD is independently associated with adverse neonatal outcomes, it is not a single predictor for neonatal outcomes given the weak discriminative ability to predict neonatal outcomes. A cutoff of 30% is more practical for risk stratification among twin gestations.
Keywords: Birthweight discordance; Dose-response relationship; Neonatal outcome; Predictive accuracy; Twin pregnancies.
© 2022. The Author(s).