How to manage premature births at gestational ages near the limit of viability has been debated for decades.1 The perceived boundary between futile and potentially beneficial intensive intervention has varied over time. For many years in the United States, the boundary has existed around 22 to 24 weeks’ gestation.2
In this issue of Pediatrics, LoRe and Groden and the Investigating Neonatal Decisions for Extremely Early Deliveries (INDEED) study group show how the gestational age at which intensive interventions are directed at neonatal survival has shifted during the past decade.3 In a cohort study of more than 2000 delivering patients who presented with a live fetus to teaching hospitals in 10 US states, the authors show that provision of antenatal steroids, consultation with a neonatologist before delivery, live birth, postnatal infant intervention, and infant survival to NICU discharge increased substantially for deliveries at 22 to 23 weeks’ gestation. There were no apparent changes in interventions or outcomes at 24 weeks. Importantly, changes did not occur consistently across all hospitals. For example, for deliveries at 22 weeks, between-hospital variation in antenatal corticosteroid provision, measured by the interquartile range, more than doubled during the study period.
The findings of the INDEED study group raise at least 4 important questions.
1. What Caused Changes in Intervention During the Past Decade?
A recent study of hospitals in 29 US states showed NICU admissions at 22 weeks’ gestation increased nearly 400% since 2008, with changes beginning around 2015.4 LoRe and Groden et al cite American Academy of Pediatrics (AAP) guidance on antenatal counseling 5 and a Neonatal Research Network study6 featured in news reports in 20157 as reasons for observed changes. There were also contemporary reports from hospitals in Japan (2013),8 Sweden (2015),9 and Germany (2016)10 showing surprisingly high rates of survival at 22 to 23 weeks.
No guideline in the United States has recommended intensive obstetric and neonatal care for deliveries at 22 weeks’ gestation. The American College of Obstetrics and Gynecology recommends to “consider” antenatal steroids11 and postnatal infant intervention,12 and the AAP recommends “decision-making regarding the delivery room management be individualized and family centered, taking into account known fetal and maternal conditions and risk factors as well as parental beliefs regarding the best interest of the child.”5 It remains unclear whether practice changes have been driven by changing parental beliefs, whether clinicians have offered or encouraged intervention more often, and whether the balance of risk and benefit also changed over time.
2. What Do Families Know About Between-Hospital Differences in Treatments and Outcomes?
Like LoRe and Groden et al, others have identified between-hospital variation in interventions and outcomes at 22 to 24 weeks.6,9 Deliveries at 22 to 24 weeks are typically unplanned and, as reported elsewhere by the INDEED group, approximately one-third occur within 12 hours of hospitalization.13 Despite this, when selecting an obstetrician or delivery hospital, are pregnant patients aware of whether they would have the option of obstetric or neonatal interventions at early gestational ages? Moreover, do patients know about differences in outcomes like live birth and infant survival among nearby hospitals, and does this affect their choices? Although in vitro fertilization success14 and cesarean delivery rates15 are publicly available, there is no public information on hospital-level practices or outcomes for delivery at 22 to 24 weeks.
3. Should All Hospitals Provide Intensive Interventions for Deliveries at 22 to 24 Weeks?
Considering the variation in volume, intervention, and outcomes observed by LoRe and Groden et al, is it possible that high-quality intensive obstetric and neonatal care at the earliest gestations could be better organized? Approximately 850 of 1400 NICUs in the United States are designated level III through IV.16 The AAP recommends that all level III through IV NICUs be equipped to care for infants born <32 weeks’ gestation and makes no distinction between the substantial differences in risk, resources, and specialized expertise for infants born at 31 weeks and 22 weeks.17 Moreover, risk-appropriate neonatal care does not always correspond with risk-appropriate maternal care.18 Should families be given the option to transfer to centers specializing in extremely early deliveries when possible? One population-based study found that, when centralization of extremely early deliveries was recommended, the proportion of live births at 22 to 23 weeks in designated centers increased from 67% to 87% and infant survival increased.19
4. What Evidence Supports Specific Interventions for Deliveries at 22 to 24 Weeks?
Most antenatal and delivery interventions included in the LoRe and Groden et al study have yet to be rigorously tested. A systematic review of all trials including extremely preterm infants published 2010 to 2019 showed only 1.4% of 32 000 infants enrolled in 200 clinical trials were 22 to 23 weeks.20 Because infants born at 22 to 24 weeks make up ∼1 in 500 live births4 (twice the incidence of trisomy 2121 ) and 1 in 5 infant deaths,22 scientific evidence to support specific obstetric and neonatal interventions is sorely needed.
The INDEED study group has highlighted vital questions requiring obstetric-neonatal collaboration. Who is affected by differences in intensive obstetric and neonatal interventions, how such differences impact patients, and how to equitably improve the care we provide families affected by birth at <25 weeks remain important areas for further study.
Drs Rysavy, Battarbee, and Gibson drafted the commentary and reviewed it critically for important intellectual content; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
COMPANION PAPER: A companion to this article can be found online at www.pediatrics.org/cgi/doi/10.1542/peds.2023-065521.
FUNDING: Funded by the National Institutes of Health (NIH), grant 5UG1HD040544 (Gibson) and grant 5UG1HD112093 (Rysavy).
CONFLICT OF INTEREST DISCLOSURES: The authors have indicated they have no potential conflicts of interest to disclose.
