AIM: The aim of this study was to develop and validate an antepartum risk score based on maternal and obstetric characteristics to predict the requirement for neonatal intensive care unit (NICU) admission among late preterm infants. MATERIAL AND METHODS: A chart review was performed of 455 singleton late preterm deliveries at our institution between July 2010 and December 2011. Logistic regression analysis was used to develop a risk score, which was derived from β coefficients of the significant variables. A receiver-operator curve was plotted to determine the optimal cut-off score for predicting NICU admission. Validation of the score was tested in another cohort of 450 women who delivered a singleton late preterm infant between January 2012 and June 2013. RESULTS: A total of 98 infants (21.5%) in the development cohort were admitted to the NICU. The significant factors for NICU admission included: premature rupture of membranes, antepartum hemorrhage, medical disorders during pregnancy, prenatal estimation of fetal weight, gestational age at delivery, and mode of delivery. These six variables were integrated into a risk-scoring model, which ranged from -2 to 9 points. A cut-off score of ≥1 produced the maximum area under the receiver-operator curve of 0.764. At this cut-off point, the sensitivity was 79.6% and specificity was 73.1%. When the risk score was tested in the validation cohort, similar results were demonstrated. CONCLUSION: An antepartum risk score was developed to predict the requirement for NICU admission among late preterm infants and was validated in an independent cohort.
AIM: The aim of this study was to develop and validate an antepartum risk score based on maternal and obstetric characteristics to predict the requirement for neonatal intensive care unit (NICU) admission among late preterm infants. MATERIAL AND METHODS: A chart review was performed of 455 singleton late preterm deliveries at our institution between July 2010 and December 2011. Logistic regression analysis was used to develop a risk score, which was derived from β coefficients of the significant variables. A receiver-operator curve was plotted to determine the optimal cut-off score for predicting NICU admission. Validation of the score was tested in another cohort of 450 women who delivered a singleton late preterm infant between January 2012 and June 2013. RESULTS: A total of 98 infants (21.5%) in the development cohort were admitted to the NICU. The significant factors for NICU admission included: premature rupture of membranes, antepartum hemorrhage, medical disorders during pregnancy, prenatal estimation of fetal weight, gestational age at delivery, and mode of delivery. These six variables were integrated into a risk-scoring model, which ranged from -2 to 9 points. A cut-off score of ≥1 produced the maximum area under the receiver-operator curve of 0.764. At this cut-off point, the sensitivity was 79.6% and specificity was 73.1%. When the risk score was tested in the validation cohort, similar results were demonstrated. CONCLUSION: An antepartum risk score was developed to predict the requirement for NICU admission among late preterm infants and was validated in an independent cohort.
Authors: Ashley N Battarbee; Angelica V Glover; Catherine J Vladutiu; Cynthia Gyamfi-Bannerman; Sofia Aliaga; Tracy A Manuck; Kim A Boggess Journal: J Matern Fetal Neonatal Med Date: 2019-06-06