Title : Differences in predicted survival of extremely preterm neonates: A descriptive abstract comparing survival and outcome probabilities based on estimated fetal weight and actual birth weight
Abstract:
Introduction: Periviable births, also referred to as “preterm births”, “micro-preemies”, and at the “threshold of viability”, are defined as neonates born at 22-25 weeks. In obstetrics, births over 26 weeks are generally accepted to be highly likely to survive with appropriate intervention. It is also agreed upon that the minimal gestational age for viability in most developed countries is approximately 22 weeks1. Births between 22 - 25 weeks and the survival of these periviable neonates is relatively unexplored and difficult to determine. In an effort to predict survival, the National Institute of Child Health and Human Development (NICHD) has created the Extremely Preterm Birth Outcomes Tool2. It uses factors such as gestational age, birth weight, sex, singleton, and the use of antenatal steroids to estimate the probability of survival and outcome of a neonate3. Because this tool is often used prior to the birth of the neonate, the estimated fetal weight determined by ultrasound is often used in this calculation instead of actual birth weight4. Although current formulas for determining EFW are generally accurate in predicting BW in term births, there is limited data on the accuracy of EFW to predict BW in periviable neonates. Multiple studies have concluded that there is a less than 10% mean percent difference between EFW and BW5,6. There has been little exploration of how the discrepancy between these two values can contribute to changes in survival probability. In review of previous literature, only one study was found to address this question. Results showed an absolute value of relative difference (AVRD) of predicted neonatal survival (PredNS) of greater than or equal to 10% in 58% of participants7. Since weight is a component of the calculator, we seek to explore if predicted survival using EFW and BW varies, and how these predictions compare to the actual outcomes of these neonates. This data can ultimately assist practitioners and families in making major decisions regarding life-sustaining measures for these neonates.
Methods: For this study, we will review approximately 100 maternal and respective neonate charts consisting of extremely preterm births (22-25 weeks) that occurred at Hutzel Women’s Hospital, Children’s Hospital of Michigan, and Sinai Grace Hospital located within the Detroit Medical Center from 2014-2024. Neonates must have a birth weight between 401 and 1000 grams per the NICHD extremely preterm birth outcome tool. Ultrasound-Estimated Fetal Weight (EFW), Birth Weight (BW), gestational age at time of birth, antenatal steroid use, neonate sex, and the survival of the neonate will be extracted from the patient charts. The NICHD Preterm Birth Outcomes Tool will be used to calculate probability using both EFW and BW, and the difference in probability will be calculated using PredNS BW - PredNS EFW. A negative value will indicate an overestimation in survival probability when using EFW, and a positive value indicates underestimation of survival probability when using EFW. The actual survival and outcomes of the neonates will also be calculated and compared to PredNS BW.
