Corticosteroids for the prevention and treatment of bronchopulmonary dysplasia: an overview of systematic reviews.

BACKGROUND Bronchopulmonary dysplasia (BPD) remains an important complication of prematurity. Pulmonary inflammation plays a central role in the pathogenesis of BPD, explaining the rationale for investigating postnatal corticosteroids. Multiple systematic reviews (SRs) have summarised the evidence from numerous randomised controlled trials (RCTs) investigating different aspects of administrating postnatal corticosteroids. Besides beneficial effects on the outcome of death or BPD, potential short- and long-term harms have been reported. OBJECTIVES The primary objective of this overview was to summarise and appraise the evidence from SRs regarding the efficacy and safety of postnatal corticosteroids in preterm infants at risk of developing BPD. METHODS We searched the Cochrane Database of Systematic Reviews, MEDLINE, Embase, CINAHL, and Epistemonikos for SRs in April 2023. We included all SRs assessing any form of postnatal corticosteroid administration in preterm populations with the objective of ameliorating pulmonary disease. All regimens and comparisons were included. Two review authors independently checked the eligibility of the SRs comparing corticosteroids with placebo, and corticosteroids with different routes of administration and regimens. The included outcomes, considered key drivers in the decision to administer postnatal corticosteroids, were the composite outcome of death or BPD at 36 weeks' postmenstrual age (PMA), its individual components, long-term neurodevelopmental sequelae, sepsis, and gastrointestinal tract perforation. We independently assessed the methodological quality of the included SRs by using AMSTAR 2 (A Measurement Tool to Assess Systematic Reviews) and ROBIS (Risk Of Bias In Systematic reviews) tools. We assessed the certainty of the evidence using GRADE. We provided a narrative description of the characteristics, methodological quality, and results of the included SRs. MAIN RESULTS We included nine SRs (seven Cochrane, two non-Cochrane) containing 87 RCTs, 1 follow-up study, and 9419 preterm infants, investigating the effects of postnatal corticosteroids to prevent or treat BPD. The quality of the included SRs according to AMSTAR 2 varied from high to critically low. Risk of bias according to ROBIS was low. The certainty of the evidence according to GRADE ranged from very low to moderate. Early initiated systemic dexamethasone (< seven days after birth) likely has a beneficial effect on death or BPD at 36 weeks' PMA (risk ratio (RR) 0.88, 95% confidence interval (CI) 0.81 to 0.95; number needed to treat for an additional beneficial outcome (NNTB) 16, 95% CI 10 to 41; I2 = 39%; 17 studies; 2791 infants; moderate-certainty evidence) and on BPD at 36 weeks' PMA (RR 0.72, 95% CI 0.63 to 0.82; NNTB 13, 95% CI 9 to 21; I2 = 39%; 17 studies; 2791 infants; moderate-certainty evidence). Early initiated systemic hydrocortisone may also have a beneficial effect on death or BPD at 36 weeks' PMA (RR 0.90, 95% CI 0.82 to 0.99; NNTB 18, 95% CI 9 to 594; I2 = 43%; 9 studies; 1376 infants; low-certainty evidence). However, these benefits are likely accompanied by harmful effects like cerebral palsy or neurosensory disability (dexamethasone) or gastrointestinal perforation (both dexamethasone and hydrocortisone). Late initiated systemic dexamethasone (≥ seven days after birth) may have a beneficial effect on death or BPD at 36 weeks' PMA (RR 0.75, 95% CI 0.67 to 0.84; NNTB 5, 95% CI 4 to 9; I2 = 61%; 12 studies; 553 infants; low-certainty evidence), mostly contributed to by a beneficial effect on BPD at 36 weeks' PMA (RR 0.76, 95% CI 0.66 to 0.87; NNTB 6, 95% CI 4 to 13; I2 = 14%; 12 studies; 553 infants; low-certainty evidence). No harmful side effects were shown in the outcomes chosen as key drivers to the decision to start or withhold late systemic dexamethasone. No effects, either beneficial or harmful, were found in the subgroup meta-analyses of late hydrocortisone studies. Early initiated inhaled corticosteroids probably have a beneficial effect on death and BPD at 36 weeks' PMA (RR 0.86, 95% CI 0.75 to 0.99; NNTB 19, 95% CI not applicable; I2 = 0%; 6 studies; 1285 infants; moderate-certainty evidence), with no apparent adverse effects shown in the SRs. In contrast, late initiated inhaled corticosteroids do not appear to have any benefits or harms. Endotracheal instillation of corticosteroids (budesonide) with surfactant as a carrier likely has a beneficial effect on death or BPD at 36 weeks' PMA (RR 0.60, 95% CI 0.49 to 0.74; NNTB 4, 95% CI 3 to 6; I2 = 0%; 2 studies; 381 infants; moderate-certainty evidence) and on BPD at 36 weeks' PMA. No evidence of harmful effects was found. There was little evidence for effects of different starting doses or timing of systemic corticosteroids on death or BPD at 36 weeks' PMA, but potential adverse effects were observed for some comparisons. Lowering the dose might result in a more unfavourable balance of benefits and harms. Moderately early initiated systemic corticosteroids, compared with early systemic corticosteroids, may result in a higher incidence of BPD at 36 weeks' PMA. Pulse dosing instead of continuous dosing may have a negative effect on death and BPD at 36 weeks' PMA. We found no differences for the comparisons of inhaled versus systemic corticosteroids. AUTHORS' CONCLUSIONS This overview summarises the evidence of nine SRs investigating the effect of postnatal corticosteroids in preterm infants at risk for BPD. Late initiated (≥ seven days after birth) systemic administration of dexamethasone is considered an effective intervention to reduce the risk of BPD in infants with a high risk profile for BPD, based on a favourable balance between benefits and harms. Endotracheal instillation of corticosteroids (budesonide) with surfactant as a carrier is a promising intervention, based on the beneficial effect on desirable outcomes without (so far) negative side effects. Pending results of ongoing large, multicentre RCTs investigating both short- and long-term effects, endotracheal instillation of corticosteroids (budesonide) with surfactant as a carrier is not appropriate for clinical practice at present. Early initiated (< seven days after birth) systemic dexamethasone and hydrocortisone and late initiated (≥ seven days after birth) hydrocortisone are considered ineffective interventions, because of an unfavourable balance between benefits and harms. No conclusions are possible regarding early and late inhaled corticosteroids, as more research is needed.