By Katherine Rivlin, MD, MSc
Assistant Professor, Obstetrics and Gynecology, The Ohio State University Wexner Medical Center, Columbus, OH
Dr. Rivlin reports no financial relationships relevant to this field of study.
SYNOPSIS: In this case control study using U.S. live birth records between 22 and 23 weeks of gestation, maternal intervention was positively associated with increasing maternal age, Medicaid use, preeclampsia, birth defects, twin gestation, multiparity, and infertility treatments while being negatively associated with non-Hispanic Black race. Positive associations for neonatal intervention included non-Hispanic Black race, preeclampsia, Medicaid use, infertility treatments, less than a high school education, increasing maternal age, and twin gestation, and negative associations included birth defects and small for gestational age pregnancies.
SOURCE: Hajdu SA, Rossi RM, DeFranco EA. Factors associated with maternal and neonatal interventions at the threshold of viability. Obstet Gynecol 2020;135:1398-1408.
Neonatal delivery during the periviable period between 22 and 23 weeks of gestation (specifically between 22 0/7 and 23 6/7 weeks of gestation) presents unique challenges for both clinicians and patients. Complex and ethically challenging decisions often must be made in rapidly evolving clinical situations that can be difficult to predict. Periviable birth has been associated with poor maternal outcomes and significant infant morbidity and mortality. Minimal data exist to support the efficacy of maternal and neonatal interventions, such as cesarean section, corticosteroid administration, and neonatal intensive care unit (NICU) admission. Intervention decisions may be influenced by maternal and fetal factors as well as available hospital resources. Provider and patient attitudes and biases likely also play a role in these decisions. In this study, Hajdu et al explored associations between maternal and pregnancy characteristics and periviable interventions to better understand the decision-making that occurs during this liminal time period.
This population-based, case control study of all live births in the United States from 2012 to 2016 used data from the birth certificates of neonatal deliveries occurring between 22 and 23 weeks of gestation. The study collected data on maternal sociodemographic, medical, and pregnancy characteristics that may influence patient and provider decision-making, and their associations with three primary outcomes: 1) maternal interventions (cesarean section, maternal hospital transfer, antenatal steroid administration); 2) neonatal interventions (NICU admission, surfactant administration, antibiotic administration, assisted ventilation); and 3) combined maternal-neonatal interventions (at least one maternal and at least one neonatal intervention). The case groups were births between 22 and 23 weeks of gestation that met criteria for one of these primary outcomes, and the control groups were births in the same gestational age category that did not. Chi-squared tests and multivariate logistic regression models estimated the influence of maternal and pregnancy characteristics on primary outcomes.
All live births were included in the study. Intrapartum stillbirths and births with missing data on the interventions analyzed were excluded. In multifetal gestations, only the first birth was included in maternal outcomes, but all live births were included in neonatal outcomes. Maternal characteristics were compared between births at 22 and 23 weeks of gestation with those at 24 and 42 weeks of gestation to identify the baseline characteristics of the periviable birth population. The study found that of the 19,844,580 U.S. live births between 2012 and 2016, 24,379 (0.12%) occurred between 22 and 23 weeks of gestation. When compared to births between 24 and 42 weeks, the frequency of non-Hispanic Black race births was higher between 22 and 23 weeks of gestation (P < 0.001). The frequency of Medicaid use, unmarried status, short interpregnancy interval, obesity, prior preterm birth, twin gestation, and no prenatal care also was higher in the 22- to 23-week group (no P value available). Among births between 22 and 23 weeks of gestation, 62% received at least one maternal or neonatal intervention, with interventions more likely in the 23rd week of gestation than in the 22nd week.
In multivariable analysis controlling for maternal characteristics and pregnancy characteristics, maternal intervention was more likely in cases of preeclampsia, birth defects, twin gestation, multiparity, Medicaid use, and increasing maternal age, and was less likely in non-Hispanic Black patients (all P < 0.001). Maternal characteristics included age, race/ethnicity, insurance type, educational attainment, and marital status, and pregnancy characteristics included parity, interpregnancy interval, body mass index (BMI), history of preterm birth, gestational weight gain, preeclampsia, fetal growth, birth defects, twin gestation, chronic hypertension, gestational and pre-gestational diabetes, infertility treatment, assisted reproduction, tobacco use, and no prenatal care. Neonatal intervention was more likely in cases of preeclampsia, non-Hispanic Black patients, twin gestation, Medicaid use, infertility treatments, less than a high school education, and increasing maternal age, and was less likely in cases of birth defects and small for gestational age (SGA) pregnancies (all P < 0.001). Combined maternal and neonatal interventions were more likely in cases of preeclampsia, infertility treatments, multiparity, and Medicaid use, and were less likely in cases of SGA pregnancies, birth defects, and non-Hispanic Black patients (all P < 0.001).
Clinical trends have shifted toward more aggressive periviable intervention in the years since the 2014 National Institute of Child Development joint workshop on the management, counseling, and treatment options for infants delivered during the periviable time period.1,2 Despite these shifts, there is little data to support the efficacy of such interventions. Recognizing that births on the threshold of viability are rare, Hajdu et al took advantage of a large study population (all live births in the Unites States over a five-year time period) to better explore clinical trends and to identify factors that may play a role in the decision to intervene.
Using vital statistics in research has its limitations. Birth certificate data may contain inaccuracies and missing information, and important research variables may not be included. For example, the use of magnesium sulfate — a highly relevant periviable intervention — is not reported on birth certificates. Additionally, birth certificate data are retrospective and observational, and, therefore, cannot establish causality. The authors noted that some associations were “diverse with limited commonality,” and that others were discordant. These findings may reflect the study’s large sample size (which can lead to modest effect sizes) but also may reflect the complexity of decision-making that occurs during this liminal time period. For example, such decisions may depend on factors that are hard to measure, such as available hospital resources, local hospital culture, and physician and patient values. Despite these limitations, I found this study to be highly relevant to the developing body of literature surrounding this clinical and ethical “gray zone” in obstetrics and gynecology. The study clearly demonstrates that clinical interventions are occurring in more than half of live births between 22 and 23 weeks of gestation, which may come as a surprise to some clinicians, depending on local hospital practices and definitions of viability.
Additionally, this study highlights the significant healthcare disparities that Black mothers face during their reproductive lives. Black woman are three times more likely to die in pregnancy than Caucasian and Hispanic women, and Black children are twice as likely to die in infancy as Caucasian children.3,4 As clearly demonstrated in this study, Black women have a higher preterm birth rate than Caucasian women. Also, this study identifies a higher rate of intervention among Black neonates compared to other races. These intervention trends have been reported elsewhere, with Black neonates more likely to undergo intubation than Caucasian neonates, perhaps reflecting not only underlying differences in patient preferences, but also improved survival among premature Black neonates compared to premature Caucasian neonates.5,6 This discordance between Black maternal and neonatal interventions deserves further exploration. Although factors contributing to racial disparities in healthcare are complex, stereotyping and implicit bias by healthcare providers may play an important role, particularly in situations where clinical guidance is limited. Providers should be aware of these disparities and adjust their practices to promote racial equity in women’s health.7
Decision-making around periviable interventions can be clinically complex and ethically and emotionally challenging for both patients and providers. Prediction models used to calculate chances of neonatal survival can be inaccurate, and little data exists to support interventions.8 This study does not provide guidance on when interventions should occur, but it does highlight the unpredictable nature of how and when interventions do occur. How should this study change practice? Observing the complex and, at times, discordant practice patterns occurring in the periviable period highlights the importance of reminding ourselves that although data supporting interventions are limited, best practices do exist and should guide how we counsel our patients.
In its Obstetrical Care Consensus on Periviable Birth, the American College of Obstetricians and Gynecologists outlines current best evidence on periviable interventions and outcomes and best practices in family counseling.8 Discussions between providers and families should seek to create an environment of mutual trust and understanding. Whenever possible, discussions should involve a multidisciplinary team, including an obstetrician, a maternal-fetal medicine specialist, a neonatologist, and any additional support that may benefit the family, such as psychological support, a social worker, a spiritual care provider, or a bioethicist. Clinicians should use a shared decision-making model with patients and their families, where the care team provides medical expertise and emotional support, but, ultimately, the patient’s values and preferences are honored.
- Raju TN, Mercer BM, Burchfield DJ, Joseph GF. Periviable birth: Executive summary of a joint workshop by the Eunice Kennedy Shriver National Institute of Child Health and Human Development, Society for Maternal-Fetal Medicine, American Academy of Pediatrics, and American College of Obstetricians and Gynecologists. Obstet Gynecol 2014;123:1083-1096.
- Rossi RM, Hall E, DeFranco EA. Contemporary trends in cesarean delivery utilization for live births between 22 0/7 and 23 6/7 weeks of gestation. Obstet Gynecol 2019;133:451-458.
- Creanga AA, Berg CJ, Syverson C, et al. Pregnancy-related mortality in the United States, 2006-2010. Obstet Gynecol 2015;125:5-12.
- Mathews TJ, MacDorman MF. Infant mortality statistics from the 2009 period linked birth/infant death data set. Natl Vital Stat Rep 2013;61:1-27.
- Tucker Edmonds B, Fager C, Srinivas S, Lorch S. Racial and ethnic differences in use of intubation for periviable neonates. Pediatrics 2011;127:e1120-1127.
- Loftin R, Chen A, Evans A, DeFranco E. Racial differences in gestational age-specific neonatal morbidity: Further evidence for different gestational lengths. Am J Obstet Gynecol 2012;206:259.e1-6.
- [No authors listed]. ACOG Committee Opinion No. 649: Racial and ethnic disparities in obstetrics and gynecology. Obstet Gynecol 2015;126:e130-134.
- American College of Obstetricians and Gynecologists, Society for Maternal-Fetal Medicine. Obstetric Care consensus No. 6: Periviable birth. Obstet Gynecol 2017;130:e187-e199.