Association of Inhaled Corticosteroids and Nebulized Furosemide with Reduced Hospitalizations and Oxygen Dependence in Preterm Infants with Bronchopulmonary Dysplasia: A Retrospective Cohort Study

Background Bronchopulmonary dysplasia (BPD) remains a leading cause of chronic lung disease in premature infants. Current pharmacotherapy often combines corticosteroids and diuretics, but robust comparative data on the efficacy of inhaled corticosteroids (ICS) and nebulized furosemide (NF) combinations are lacking. Objective This study evaluated associations of intravenous corticosteroids (IVCS), ICS, and NF, used alone or in combination, with hospitalization duration, oxygen dependency, and extubation outcomes in preterm infants with BPD. Methods In this retrospective cohort study, we analyzed data from 97 preterm infants with BPD (diagnosed by National Institutes of Health criteria) across two neonatal intensive care units in 2020. Participants were stratified into three treatment categories: monotherapy (IVCS, ICS, or NF alone; n = 9), dual therapy (IVCS+NF, n = 41; or ICS+NF, n = 27), and triple therapy (IVCS+ICS+NF, n = 20). Primary outcomes included duration of hospitalization and oxygen therapy, rates of early extubation (≤7 days), and reintubation. Nonparametric tests (Kruskal–Wallis, Mann–Whitney U ) and multivariable regression, adjusted for confounders, were employed. Effect sizes were calculated using rank-biserial correlation ( r rb ). Due to the nonrandomized design, the potential for indication bias is a major limitation; no adjustment for multiple comparisons was made, and findings should be interpreted as exploratory. Results The ICS+NF dual therapy group was associated with the shortest median hospitalization (1.53 months, interquartile range [IQR] 1.40-1.73; P = .001 vs. IVCS+NF; r rb = 0.82, 95% confidence interval [CI] 0.60-0.94) and the shortest duration of oxygen therapy dependency (1.22 months, IQR 1.10-1.63; P = .010; r rb = 0.71, 95% CI 0.45-0.88). These effect sizes are large but should be interpreted cautiously due to the observational design and small, imbalanced sample. Infants receiving ICS+NF also had significantly lower reintubation rates (37.0%) compared to those receiving IVCS+NF (56.1%, P = .037). Triple therapy showed no benefit ( P ≥ .41). No adverse events occurred. Conclusion In this retrospective cohort, ICS+NF was associated with reduced hospitalization and oxygen dependency, suggesting that this combination may be a promising therapeutic strategy. However, due to the observational design and potential for residual confounding, these findings require validation in prospective randomized controlled trials.