Validation of the discriminatory performance of the iPREFACE score for clinically significant fetal acidemia: an outcome-enriched nested case-control study.

BACKGROUND: Cardiotocography is widely used for monitoring fetal well-being during labor; however, its clinical utility remains limited due to a high false-positive rate, particularly in predicting adverse neonatal outcomes such as cerebral palsy. Despite its implementation, the incidence of cerebral palsy has not decreased, whereas cesarean section rates have increased. Severe fetal acidemia has been shown to be associated with neonatal brain injury. The iPREFACE score was developed to address this limitation by evaluating the accumulation of fetal stress due to uterine contractions based on the five-tier classification of fetal heart rate (FHR) patterns defined by the Japanese Society of Obstetrics and Gynecology. Previous studies have demonstrated that the iPREFACE score significantly correlates with umbilical artery (UmA) blood gas parameters, including pH, base excess (BE), and lactate, and has a high predictive ability for fetal acidemia.

OBJECTIVE: This study aimed to validate the discriminatory performance of the iPREFACE score for clinically significant fetal acidemia, defined as UmA pH<7.1, using an outcome-enriched artificially balanced cohort. Secondary analyses evaluated performance at UmA pH<7.2 and <7.0, and an exploratory analysis assessed UmA pH<7.3 to examine stepwise fetal deterioration during labor.

STUDY DESIGN: A nested case-control study was conducted involving women who had full-term vaginal deliveries at a single center between January 2017 and July 2023. To construct an outcome-enriched cohort, cases with UmA pH<7.2 were matched 1:1 with randomly selected controls with UmA pH≥7.2. The iPREFACE score was calculated by summing the levels of FHR waveform patterns in the 30 minutes before delivery. Receiver operating characteristic (ROC) curve analysis was performed to evaluate discriminatory performance at different thresholds of UmA pH (<7.1, <7.2, <7.0, and <7.3). Correlations between the iPREFACE score and UmA blood gas parameters (pH, pCO₂, pO₂, HCO₃⁻, BE, and lactate) were assessed using linear and quadratic regression models. Bland-Altman analysis was performed to assess agreement between predicted and measured UmA pH values.

RESULTS: Of the 272 patients, 15 (5.5%) had UmA pH<7.1 and 6 (2.2%) had UmA pH<7.0. The ROC-AUC for the iPREFACE score was 0.851 (95% confidence interval [CI]: 0.771-0.931) for the primary outcome of UmA pH<7.1. The AUC was 0.912 (95% CI: 0.879-0.945) for UmA pH<7.2 and 0.823 (95% CI: 0.656-0.989) for UmA pH<7.0. In the exploratory analysis, the AUC for UmA pH<7.3 was 0.833. Quadratic regression analysis revealed that the iPREFACE score had the highest coefficient of determination (R²) with pH (0.415), followed by BE (0.357), pCO₂ (0.230), lactate (0.165), and HCO₃⁻ (0.106), whereas pO₂ showed a notably lower correlation (R²=0.02). Bland-Altman analysis showed a mean difference of -0.16 between predicted and measured UmA pH values, with limits of agreement from -0.41 to 0.09.

CONCLUSION: In an outcome-enriched cohort, the iPREFACE score demonstrated discriminatory performance for clinically significant fetal acidemia (UmA pH<7.1) and significant correlations with key UmA blood gas parameters, particularly pH and BE. Consistent discriminatory performance across multiple pH thresholds supports further evaluation of the iPREFACE score for stepwise assessment of progressive fetal deterioration during labor.