Objective: To quantify the influence of maternal sociodemographic, medical, and pregnancy characteristics on decisions to offer or receive antepartum and neonatal interventions with deliveries occurring at 22-23 weeks of gestation.
Methods: This is a case-control study of U.S. live births at 22 0/7-23 6/7 weeks of gestation using National Center for Health Statistics vital statistics birth records from 2012 to 2016. We analyzed three outcomes in the treatment of periviable delivery: 1) maternal interventions (cesarean delivery, maternal hospital transfer or antenatal corticosteroid administration), 2) neonatal interventions (neonatal intensive care unit admission, surfactant administration, antibiotic administration, or assisted ventilation), and 3) combined interventions (at least one maternal and at least one neonatal intervention). Logistic regression estimated the influence of characteristics on interventions received.
Results: Of 19,844,580 U.S. live births from 2012 to 2016, 24,379 (0.12%) occurred at 22-23 weeks of gestation. Of these, 37.5% received maternal interventions, 51.7% received neonatal interventions, and 28.0% received combined interventions. Rates of births receiving at least one intervention were 38.9% and 78.3% for 22 and 23 weeks of gestation, respectively. Preeclampsia was the factor most positively associated with interventions. Other factors positively associated with interventions were increasing maternal age, Medicaid, low educational attainment, multiparity, twin gestation, and infertility treatment. Some factors had opposite influences on maternal compared with neonatal interventions. The presence of birth defects was positively associated with maternal interventions but negatively associated with neonatal interventions, whereas being of black race was negatively associated with maternal interventions but positively associated with neonatal interventions.
Conclusion: Maternal and neonatal interventions occur frequently at the threshold of viability, especially at 23 weeks of gestation where the occurrence of interventions exceeds 50%. This study identifies sociodemographic and medical factors associated with using interventions with periviable deliveries. These data elucidate observed practice patterns in the management of periviable births and may assist providers in the counseling of women at risk of periviable birth.