Systemic steroids and bronchopulmonary dysplasia: a systematic review and meta-analysis.

IMPORTANCE

It is unclear if systemic steroids decrease the risk of Bronchopulmonary Dysplasia (BPD) while increasing the risk of neurodevelopmental impairment (NDI).

OBJECTIVE

Conduct a systematic review of randomized controlled trials of systemic steroids to evaluate the risk of BPD, mortality, and NDI in premature infants ≤30 weeks.

DATA SOURCES: MEDLINE, EBSCOhost, Web of Science, Cochrane Library, Embase, and CINAHL.

STUDY SELECTION: Randomized clinical trials of Dexamethasone (DEX) or Hydrocortisone (HC) to prevent BPD in premature infants ≤ 30 weeks.

DATA EXTRACTION AND SYNTHESIS: Data were extracted using the Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines. Random-effects meta-analyses and multivariable meta-regression were conducted.

MAIN OUTCOMES AND MEASURES: Primary outcomes were BPD, mortality, and NDI. Secondary outcomes were hypertension, hyperglycemia, sepsis, intestinal perforation, necrotizing enterocolitis (NEC), and retinopathy of prematurity (ROP). The a priori hypothesis was that steroids would reduce the risk of BPD without increasing NDI.

RESULTS

There were 6377 preterm infants in the 44 (32 DEX, 13 HC) selected studies. DEX significantly reduced the risk of BPD, RR = 0.66, (95% CI, 0.56-0.78). The most effective DEX regimen was medium cumulative dose (2 to 3 mg/kg), RR = 0.43 (95% CI, 0.29-0.65); day of initiation <8 days: RR = 0.68, (95% CI, 0.59-0.79); and treatment for ≥14 days: RR = 0.67 (95% CI, 0.55-0.80). HC did not significantly decrease the risk of BPD, RR = 0.98, (95% CI, 0.87-1.10). Neither DEX, (RR = 0.92, 95% CI, 0.78-1.09) nor HC (RR = 0.83, 95% CI, 0.68-1.01) decrease the risk of mortality. The risk of CP was not increased by either DEX (RR = 1.09, 95% CI, 0.55-2.17) or HC (RR = 1.18, 95% CI, 0.75-1.87). There were no significant differences between steroids and placebo for MDI/PDI scores. Multivariable meta-regression models showed that DEX significantly reduced the risk of BPD without increased risk of CP. DEX increased the risk of hypertension and hyperglycemia. Studies showed high heterogeneity, differing treatment regimen, missing data and different rates of follow-up.

CONCLUSION AND RELEVANCE: DEX, but not HC, significantly decreased the risk of BPD. Neither steroid showed an increased risk of NDI or mortality.