Airway administration of corticosteroids for prevention of bronchopulmonary dysplasia in premature infants: a meta-analysis with trial sequential analysis

Abstract

Background: Uncertainly prevails with regard to the use of inhalation or instillation steroids to prevent bronchopulmonary dysplasia in preterm infants. The meta-analysis with sequential analysis was designed to evaluate the efficacy and safety of airway administration (inhalation or instillation) of corticosteroids for preventing bronchopulmonary dysplasia (BPD) in premature infants.

Methods: We searched MEDLINE, EMBASE, CINAHL, and Cochrane CENTRAL from their inceptions to February 2017. All published randomized controlled trials (RCTs) evaluating the effect of airway administration of corticosteroids (AACs) vs placebo or systemic corticosteroid in prematurity were included. All meta-analyses were performed using Review Manager 5.3.

Results: Twenty five RCTs retrieved (n = 3249) were eligible for further analysis. Meta-analysis and trial sequential analysis corrected the 95% confidence intervals estimated a lower risk of the primary outcome of BPD (relative risk 0.71, adjusted 95% confidence interval 0.57-0.87) and death or BPD (relative risk 0.81, adjusted 95% confidence interval 0.71-0.97) in AACs group than placebo and it is equivalent for preventing BPD than systemic corticosteroids. Moreover, AACs fail to increasing risk of death compared with placebo (relative risk 0.90, adjusted 95% confidence interval 0.40-2.03) or systemic corticosteroids (relative risk 0.81, 95% confidence interval 0.62-1.06).

Conclusions: Our findings suggests that AACs (especially instillation of budesonide using surfactant as a vehicle) are an effective and safe option for preventing BPD in preterm infants. Furthermore, the appropriate dose and duration, inhalation or instillation with surfactant as a vehicle and the long-term safety of airway administration of corticosteroids needs to be assessed in large trials.

Keywords: Bronchopulmonary dysplasia; Inhaled corticosteroids; Meta-analysis; Neurodevelopmental outcomes; Preterm infants.

Fig. 1

Flowchart of the study selection process. * Two article [48, 53] were follow-up study of two included trials [47, 52]

Fig. 2

Risk of bias graph

Fig. 3

Risk of bias summary for all included records

Fig. 4

Meta-analysis of primary outcome with the use of AACs or placebo, AACs: Airway administration of corticosteroids

Fig. 5

Trial sequential analysis of trials reporting primary outcome comparing pooled AACs and placebo. A diversity adjusted information size was calculated using ɑ = 0.05 (two sided), β = 0.20 (power 80%), D2 = 0%, an anticipated relative risk of 29.8%, −24.5%, 13.5%, respectively, and an event proportion of 28.2%, 11.4%, 46.3%, respectively in the control arm. The cumulative z curve was constructed using a random-effect model, and a penalized z curve was also constructed. The adjusted relative risk was 0.71, 0.90, 0.81, respectively, and the 95% confidence interval was corrected to 0.57–0.87, 0.40–2.03, and 0.71–0.97, respectively, AACs: Airway administration of corticosteroids

Fig. 6

Meta-analysis of primary outcome with the use of inhaled corticosteroids or systemic corticosteroids

Fig. 7

Trial sequential analysis of trials reporting primary outcome comparing pooled inhaled corticosteroids and systemic corticosteroids. A diversity adjusted information size was calculated using ɑ = 0.05 (two sided), β = 0.20 (power 80%), D2 = 0%, an anticipated relative risk of −10.5%, 18.4%, respectively, and an event proportion of 36.6%, 20.0%, respectively in the control arm. The cumulative z curve was constructed using a random-effect model, and a penalized z curve was also constructed. The adjusted relative risk was 1.08, 0.81, respectively, and the 95% confidence interval was corrected to 0.53–2.20, 0.43–1.53, respectively