The utility of fetal heart rate deceleration's descending slope in searching for a non-National Institute of Child Health and Human Development parameter for the detection of fetal acidosis.

OBJECTIVE: To identify new parameters predicting fetal acidemia. METHODS: A retrospective case-control study in a cohort of deliveries from a tertiary referral hospital-based cohort deliveries in Zaragoza, Spain between 2018 and 2021 was performed. To predict fetal acidemia, the NICHD categorizations and non-NICHD parameters were analyzed in the electronic fetal monitoring (EFM). Those included total reperfusion time, total deceleration area and the slope of the descending limb of the fetal heart rate of the last deceleration curve. The accuracy of the parameters was evaluated using the specificity for (80%, 85%, 90%, 95%) sensitivity and the area under the receiver operating characteristic curve (AUC). RESULTS: A total of 10 362 deliveries were reviewed, with 224 cases and 278 controls included in the study. The NICHD categorizations showed reasonable discriminatory ability (AUC = 0.727). The non-NICHD parameters measured during the 30-min fetal monitoring, total deceleration area (AUC = 0.807, 95% CI: 0.770, 0.845) and total reperfusion time (AUC = 0.750, 95% CI: 0.707, 0.792), exhibited higher discriminatory ability. The slope of the descending limb of the fetal heart rate of the last deceleration curve had the best AUC value (0.853, 95% CI: 0.816, 0.889). The combination of total deceleration area or total reperfusion time with the slope demonstrated high discriminatory ability (AUC = 0.908, 95% CI: 0.882, 0.933; specificities of 71.6% and 72.7% for a sensitivity of 90%). CONCLUSIONS: The slope of the descending limb of the fetal heart rate of the last deceleration curve is the strongest predictor of fetal acidosis, but its combination with the total reperfusion time shows better clinical utility.

Machine Learning Algorithm to Predict Acidemia Using Electronic Fetal Monitoring Recording Parameters.

BACKGROUND: Electronic fetal monitoring (EFM) is the universal method for the surveillance of fetal well-being in intrapartum. Our objective was to predict acidemia from fetal heart signal features using machine learning algorithms. METHODS: A case-control 1:2 study was carried out compromising 378 infants, born in the Miguel Servet University Hospital, Spain. Neonatal acidemia was defined as pH < 7.10. Using EFM recording logistic regression, random forest and neural networks models were built to predict acidemia. Validation of models was performed by means of discrimination, calibration, and clinical utility. RESULTS: Best performance was attained using a random forest model built with 100 trees. The discrimination ability was good, with an area under the Receiver Operating Characteristic curve (AUC) of 0.865. The calibration showed a slight overestimation of acidemia occurrence for probabilities above 0.4. The clinical utility showed that for 33% cutoff point, missing 5% of acidotic cases, 46% of unnecessary cesarean sections could be prevented. Logistic regression and neural networks showed similar discrimination ability but with worse calibration and clinical utility. CONCLUSIONS: The combination of the variables extracted from EFM recording provided a predictive model of acidemia that showed good accuracy and provides a practical tool to prevent unnecessary cesarean sections.