iPREFACE Score: External Validation and Bland–Altmann Analysis of Predictive Accuracy in Artificially Balanced Data From a Nested Case–Control Study [ID 1203]

INTRODUCTION: Our previously reported iPREFACE score, which predicts fetal acidemia by assessing fetal stress accumulation during labor based on cardiotocogram findings, raises concerns about its accuracy due to the rarity of fetal acidemia in real-world data. The objective of this study was to externally validate the iPREFACE score in artificially balanced data with more fetal acidemia cases. METHODS: This is a nested case–control study of full-term vaginal deliveries from January 2017 to July 2023. Eligible cases included all those with umbilical cord arterial blood (UmA) pH less than 7.2 during the study period, along with randomly selected cases with UmA pH greater than 7.2 in a 1:1 ratio. The predictive ability of the iPREFACE score for fetal acidemia was assessed using a receiver operating characteristic (ROC) curve. Bland–Altman analysis was performed to evaluate the difference between the predicted UmA pH, calculated using the previously reported quadratic curve of the iPREFACE score, and the measured values. RESULTS: The area under the curve for the ROC, with 95% CIs, was 0.91 (0.88–0.95) for UmA pH less than 7.2, 0.85 (0.77–0.93) for UmA pH less than 7.1, and 0.82 (0.66–0.99) for UmA pH less than 7.0. The mean difference between the predicted values from the quadratic curve and the measured values was −0.16. The upper and lower limits of agreement were 0.25 and −0.75, with approximately 95% of the data within this range. CONCLUSIONS/IMPLICATIONS: The iPREFACE score retains high predictive ability even in a cohort with a high number of fetal acidemia cases, raising expectations for its clinical application.