Antenatal Corticosteroids and Neonatal Outcomes in Twins: A Systematic Review and Meta-analysis.

Abstract

To assess whether antenatal corticosteroid treatment is associated with improved neonatal outcomes in twins. We searched MEDLINE, PubMed, EMBASE, and the Cochrane Library, from inception through August 12, 2021. We did not search ClinicalTrials.gov because our inclusion criteria were restricted to nonrandomized studies. Records (n=7,802) were screened in Rayyan by two independent reviewers. We included all nonrandomized studies that compared antenatal corticosteroid treatment with no treatment in twins. Our outcomes of interest were neonatal mortality, respiratory distress syndrome (RDS), intraventricular hemorrhage, bronchopulmonary dysplasia, necrotizing enterocolitis, periventricular leukomalacia, and retinopathy of prematurity. We used the ROBINS-I tool (Risk Of Bias In Non-randomised Studies - of Interventions) to assess risk of bias. We performed random-effects meta-analyses of estimates from studies without critical risk of bias due to confounding, and reported summary adjusted odds ratios (aORs) and 95% CIs. Eighteen cohort studies (that reported on 33,152 neonates) met inclusion criteria. Sixteen studies restricted to preterm gestational ages, and 11 defined exposed neonates based on an optimal corticosteroid administration-to-birth interval. Limitations due to confounding and selection bias were common concerns for the risk-of-bias assessments (n=14 at critical or higher), and 11 studies did not account for clustering within twin pairs in their analyses. All included studies had at least moderate risk of bias. Meta-analysis showed that antenatal corticosteroid administration was associated with lower odds of neonatal mortality (aOR 0.59, 95% CI 0.43-0.80, I2 69%, five studies, 20,312 neonates) and RDS (aOR 0.70, 95% CI 0.57-0.86, I2 67%, seven studies, 20,628 neonates) in twins. Results were inconclusive for the other outcomes. Evidence from nonrandomized studies suggests antenatal corticosteroids are associated with lower incidence of neonatal mortality and RDS in twins. PROSPERO, CRD42020205302.