Discordance in Antenatal Corticosteroid Use and Resuscitation Following Extremely Preterm Birth

Abstract

Objective: To describe discordance in antenatal corticosteroid use and resuscitation following extremely preterm birth and its relationship with infant survival and neurodevelopment.

Study design: A multicenter cohort study of 4858 infants 22-26 weeks of gestation born 2006-2011 at 24 US hospitals participating in the Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network, with follow-up through 2013. Survival and neurodevelopmental outcomes were available at 18-22 months of corrected age for 4576 (94.2%) infants. We described antenatal interventions, resuscitation, and infant outcomes. We modeled the effect on infant outcomes of each hospital increasing antenatal corticosteroid exposure for resuscitated infants born at 22-24 weeks of gestation to rates observed at 25-26 weeks of gestation.

Results: Discordant antenatal corticosteroid use and resuscitation, where one and not the other occurred, were more frequent for births at 22 and 23 but not 24 weeks (rate ratio [95% CI] at 22 weeks: 1.7 [1.3-2.2]; 23 weeks: 2.6 [2.2-3.2]; 24 weeks: 1.0 [0.8-1.2]) when compared with 25-26 weeks. Among infants resuscitated at 23 weeks, adjusting each hospital's rate of antenatal corticosteroid use to the average at 25-26 weeks (89.2%) was projected to increase infant survival by 7.1% (95% CI 5.4-8.8%) and survival without severe impairment by 6.4% (95% CI 4.7-8.1%). No significant change in outcomes was projected for infants resuscitated at 22 weeks, where few (n = 22) resuscitated infants received antenatal corticosteroids.

Conclusions: Infants born at 23 weeks were more frequently resuscitated without antenatal corticosteroids than other extremely preterm infants. When resuscitation is intended, consistent provision of antenatal corticosteroids may increase infant survival and survival without impairment.

Trial registration: ClinicalTrials.govNCT00063063 (Generic Database) and NCT00009633 (Follow-Up Study).

Copyright © 2019 Elsevier Inc. All rights reserved.

Figures

Figure 1.. Discordance in antenatal corticosteroid use and resuscitation by gestational age at birth

Bars represent average rates of discordance between antenatal corticosteroid use and resuscitation for births at each gestational age week, as described in the legend. Points and vertical lines represent rates and 95% confidence intervals for the combined discordance for births at each gestational age day. ANS=antenatal corticosteroids.

Figure 2.. Rates of obstetric interventions and resuscitation by gestational age and hospital of birth.

Point values represent the proportion of live births at the specified gestational age that received the specified intervention at each of the 24 hospitals in the study. Bars represent 95% CIs. The x-axis represents hospital rank order by mean active treatment rate (ordered from lowest to highest). A, Hospital rates of neonatal resuscitation; B, Hospital rates of antenatal corticosteroids exposure (at least 1 dose of antenatal corticosteroids, regardless of timing, prior to birth); C, Hospital rates of antepartum antibiotics (maternal receipt of antibiotics during the current hospitalization and prior to birth); D, Hospital rates of birth by cesarean delivery. Panel A has been modified from Rysavy et al.6

Figure 3.. Rates of obstetric interventions and resuscitation by gestational age in weeks and days.

Point values represent the proportion of live births at the specified gestational age in days that received the specified intervention. Vertical bars represent 95% CIs for these estimates. Horizontal dashed lines represent the proportion of births born at a given gestational age week with the specified intervention. The dotted lines on either side of the dashed lines represent the 95% CIs for these estimates. A, Rates of neonatal resuscitation; B, Rates of antenatal corticosteroids exposure (at least 1 dose of antenatal corticosteroids, regardless of timing, prior to birth); C, Rates of antepartum antibiotics (maternal receipt of antibiotics during the current hospitalization and prior to birth); D, Rates of birth by cesarean delivery. Panel A has been modified from Rysavy et al.6