Steroid Use for Established Bronchopulmonary Dysplasia: A Systematic Review and Meta-Analysis.

Evidence on steroid treatment for established bronchopulmonary dysplasia (BPD) is sparse. To our knowledge, a systematic review has never been conducted on this topic. This meta-analysis aims to synthesize available evidence for the use of postnatal steroids to treat established BPD. MEDLINE, Embase, Cochrane databases, and gray literature sources were searched without time or language restrictions until October 2024. We included randomized and non-randomized trials (analyzed separately) that evaluated postnatal steroids started from 28 days of life in preterm infants diagnosed with BPD. Certainty of evidence was assessed using the GRADE approach. The search retrieved 9113 records, and 20 studies were included. Meta-analysis of the RCTs demonstrated that steroids significantly reduced oxygen requirement (daily mean difference of 1.6%, 95% CI 0.25-2.95), but the analysis did not identify significant differences in total duration of supplemental oxygen, length of stay, or mortality (moderate quality). From a safety perspective, steroids resulted in a transient increase in systemic blood pressure (mean difference of 6.8 mmHg, 95% CI 4.6-8.9) (moderate quality). Weight gain during treatment was lower in the systemic steroid group (-9.2 g/day, 95% CI -11.7 to -6.8) (moderate quality), although overall growth was reported as equal (2.4 g/day, 95% CI -0.3 to 6.3) (moderate quality). One retrospective study reported the incidence of steroid treatment among infants with established BPD (any definition) to be as high as 36%. Two single-arm studies reported a prolonged high-dose systemic steroid regimen as the routine treatment strategy for severe established BPD. Moderate quality of evidence suggests that steroid treatment cannot be recommended as standard of care for established BPD. However, corticosteroids are often used to this end. Large-scale RCTs designed to treat BPD are urgently needed. Furthermore, careful consideration for patient selection and compliance with GRADE methodology is essential.