Time-Surrogate Variables Enhance the Association Between Cardiotocographic Features and Intrapartum Hypoxic-Ischemic Encephalopathy

BackgroundInterruptions to the flow of oxygenated blood to the fetal brain during labor can lead to hypoxic-ischemic encephalopathy (HIE). Timely preventive interventions for suspected hypoxemia are crucial to avoid progression to neurological injury. Prior research has used features from fetal heart rate (FHR) and uterine activity (UA) signals to predict adverse fetal outcomes. However, these systems focused only on the end of labor, when preventive measures are unlikely effective. Previously, we demonstrated that accounting for proximity to birth improved the association between FHR and UA features and the outcome group. Although proximity to birth cannot be known prospectively, other time-surrogate variables (TSVs) may be as useful. MethodsWe analyzed intrapartum data from 152,761 vaginal births, comprising 150,813 with healthy outcomes, 1,793 with perinatal acidosis, and 155 with confirmed HIE. Classical and novel FHR features were extracted across labor durations of up to 72 hours. This dataset represents the largest cohort of intrapartum FHR and UA signals to date, both in participant count and signal length. We evaluated four alternative TSVs to evaluate their ability to enhance the association of CTG features and the development of HIE. To do so, we applied information-theoretic methods to quantify their contribution to the association of fetal outcomes with FHR and UA features. FindingsOur results show that the cumulative contraction duration and the time from labor onset (TLO) were the most effective TSVs for strengthening the association FHR, UA, and fetal outcome. However, it is more practical to track TLO in clinical settings than a continuous contraction monitoring. ConclusionTLO is the most suitable TSV for prospective intrapartum CTG evaluation. Incorporating it may substantially enhance the performance of automated systems for early detection of intrapartum HIE. Author summaryOur study addressed a significant challenge in the development of classifiers to detect fetuses at high risk of developing hypoxic-ischemic encephalopathy (HIE) during labor. Most previous classifiers have concentrated on the end of labor, overlooking the dynamic evolution of FHR and UA patterns. This limits their clinical utility, as preventive interventions late in labor are unlikely to avert adverse fetal outcomes. To address this gap, we evaluated several TSVs that capture the progression of FHR and UA patterns and could support prospective, clinically usable intrapartum classifiers. Analyzing data from over 150,000 births, we found that time since labor onset was the most informative and practical TSV for enhancing the associations between FHR and UA features and the development of HIE during labor. Incorporating this variable would enable classifiers to learn the temporal dynamics of FHR and UA features, potentially improving early identification of fetuses at elevated risk for HIE and supporting more timely, effective interventions.