LAUSR

INTRODUCTION Very preterm delivery (22-32 weeks of gestation) remains a major cause of neonatal morbidity and mortality. The objective of this study was to validate a statistical model allowing to predict the risk of preterm delivery to use as a clinical decision-making tool for in utero transfer from a secondary to a tertiary care center. METHODS Retrospective observational study in a secondary care center (approximately 2500 births) in Paris, France. 137 women were admitted for threatened preterm delivery between 22 and 32 weeks. Women were retrospectively allocated to the following groups based on medical decision: "transfer group" (in utero transfer to a tertiary care unit) and "no transfer group" (no in utero transfer). The risk of preterm delivery within 48 h and before 32 weeks gestation was assessed for each group using a nomogram previously validated in a tertiary care center. The primary objective of the study was to determine the accuracy of the prediction model. RESULTS The discrimination and calibration of the nomogram were excellent (preterm delivery risk within 48 h, ROC AUC: 0.98, 95% CI: 0.95-1.00; probability of preterm delivery before 32 weeks gestation, ROC AUC: 0.94, 95% CI: 0.89-0.99). A threshold set at 0.16 helped minimize the risk of unnecessary in utero transfers with an excellent negative predictive value of 0.99. CONCLUSIONS We validated nomograms to predict the individual probability of preterm birth after admission in a secondary care center. Those nomograms could be helpful when making decisions regarding an in utero transfer to a tertiary care unit.