198: Application of fetal heart rate (FHR) algorithms to predict acidemia at birth


      To compare published intrapartum FHR monitoring algorithms for the prediction and early identification of fetal acidemia (arterial cord pH <7.1 at birth) and their effect on cesarean delivery (CD).

      Study Design

      Cases and controls were randomly selected from term, singleton deliveries between 2010 and 2014 with ≥2 hours of FHR monitoring and arterial cord blood gas results. Cases had pH <7.10 without a documented unpreventable cause of acidemia such as cord prolapse. We created rules for intervention from each of 3 published FHR interpretation systems (the 3 Tier (3T), 5 Tier (5T) and Clark systems)[references 1-3] that clearly defined when intervention was indicated, based on recommendations by the authors of each algorithm (figure 1). FHR tracings were reviewed in their entirety in 30-minute intervals by 10 NCC-certified MFM physicians or fellows who were blinded to the outcome. Sensitivity, specificity, false positive and false negative rates were calculated for each algorithm as well as the number of additional CDs that would be expected.


      265 tracings were reviewed. There were 41 cases (15.5%) with a pH <7.1 and 224 (84.5%) with a pH ≥7.1. Figure 2 shows the sensitivity, specificity, false positive and false negative rate for each algorithm. The algorithm recommended early intervention (≥30 minutes prior to actual delivery) in 4/10 cases identified by the 5T system and in 5/9 cases identified by the Clark algorithm. Application of the 5T algorithm would have resulted in 17 additional CDs (increase of 6.4%) while the Clark algorithm would have resulted in 24 additional CDs (increase of 9%).


      The 3T system does not effectively identify fetal acidemia. The 5T and Clark algorithms perform poorly in identifying fetal acidemia at birth, with detection rates of only 24% and 22%, respectively. Use of either system would result in 4 to 5 additional CD for every case of early detection. Improved methods of predicting fetal acidemia are needed.
      Figure thumbnail fx1
      Figure thumbnail fx2