Effect of antenatal corticosteroid administration-to-birth interval on maternal and newborn outcomes: a systematic review

Abstract

Background: Antenatal corticosteroids (ACS) are highly effective at improving outcomes for preterm newborns. Evidence suggests the benefits of ACS may vary with the time interval between administration-to-birth. However, the optimal ACS administration-to-birth interval is not yet known. In this systematic review, we synthesised available evidence on the relationship between ACS administration-to-birth interval and maternal and newborn outcomes.

Methods: This review was registered with PROSPERO (CRD42021253379). We searched Medline, Embase, CINAHL, Cochrane Library, Global Index Medicus on 11 Nov 2022 with no date or language restrictions. Randomised and non-randomised studies of pregnant women receiving ACS for preterm birth where maternal and newborn outcomes were reported for different administration-to-birth intervals were eligible. Eligibility screening, data extraction and risk of bias assessment were performed by two authors independently. Fetal and neonatal outcomes included perinatal and neonatal mortality, preterm birth-related morbidity outcomes and mean birthweight. Maternal outcomes included chorioamnionitis, maternal mortality, endometritis, and maternal intensive care unit admission.

Findings: Ten trials (4592 women; 5018 neonates), 45 cohort studies (at least 22,992 women; 30,974 neonates) and two case-control studies (355 women; 360 neonates) met the eligibility criteria. Across studies, 37 different time interval combinations were identified. There was considerable heterogeneity in included administration-to-birth intervals and populations. The odds of neonatal mortality, respiratory distress syndrome and intraventricular haemorrhage were associated with the ACS administration-to-birth interval. However, the interval associated with the greatest improvements in newborn outcomes was not consistent across studies. No reliable data were available for maternal outcomes, though odds of chorioamnionitis might be associated with longer intervals.

Intepretation: An optimal ACS administration-to-birth interval likely exists, however variations in study design limit identification of this interval from available evidence. Future research should consider advanced analysis techniques such as individual patient data meta-analysis to identify which ACS administration-to-birth intervals are most beneficial, and how these benefits can be optimised for women and newborns.

Funding: This study was conducted with funding support from the UNDP-UNFPA-UNICEF-WHO-World Bank Special Programme of Research, Development and Research Training in Human Reproduction (HRP), Department of Sexual and Reproductive Health and Research (SRH), a co-sponsored programme executed by the World Health Organization.

Keywords: Betamethasone; Dexamethasone; Evidence-synthesis; Glucocorticoids; Neonatal mortality.

Fig. 1

PRISMA flowchart ofincluded studies.

Fig. 2

Descriptive summary of reported neonatal mortality outcomes from randomised controlled trials. A visual representation summarising odds ratio (or relative risk—WHO 2020 trial) of neonatal mortality at various time intervals compared to “no antenatal corticosteroid” group (3 trials). Green data points indicate a significant reduction in the odds of neonatal mortality. Black data points indicate no effect of ACS on neonatal mortality.

Fig. 3

Descriptive summary of reported neonatal mortality outcomes from observational studies. A visual representation summarising odds ratio of neonatal mortality for different time intervals compared to “no antenatal corticosteroid” group (13 studies; 16 additional studies did not include a “no antenatal corticosteroid” group or had two few events). Green data points indicate a statistically significant decrease in odds ratio for neonatal mortality (i.e. upper bound of 95% CI was below 1). Red data points indicate a statistically significant increase in the odds of neonatal mortality (i.e. lower bound was above 1). Black data points indicate the odds ratio of neonatal mortality was not significantly different (i.e. 95% CI included 1).

Fig. 4

Descriptive summary of reported Respiratory Distress Syndrome outcomes from randomised controlled trials. A visual representation summarising odds ratio of respiratory distress syndrome for different time intervals compared to “no antenatal corticosteroid” group (5 trials; 4 additional trials did not include a “no antenatal corticosteroid” group or had two few events). Green data points indicate a statistically significant decrease in odds ratio for respiratory distress syndrome (i.e. upper bound of 95% CI was below 1). Black data points indicate the odds ratio of respiratory distress syndrome was not significantly different (i.e. 95% CI included 1).

Fig. 5

Descriptive summary of reported Respiratory Distress Syndrome outcomes from observational studies. A visual representation summarising odds ratio of respiratory distress syndrome for different time intervals compared to “no antenatal corticosteroid” group (7 studies; 23 additional studies did not include a “no antenatal corticosteroid” group or had too few events). Green data points indicate a statistically significant decrease in odds ratio for respiratory distress syndrome (i.e. upper bound of 95% CI was below 1). Black data points indicate the odds ratio of respiratory distress syndrome was not significantly different (i.e. 95% CI included 1).