Factors associated with radiographic reversibility in paediatric Bronchiectasis

Abstract

Introduction: Bronchiectasis unrelated to cystic fibrosis (CF), is characterised by chronic respiratory symptoms with objective evidence of abnormal bronchial dilatation on chest multidetector-computed-tomography (MDCT) scans. Often considered irreversible, bronchiectasis is now known to be reversible in small studies. We aimed to identify factors associated with radiographic reversibility in paediatric bronchiectasis. Method: We identified children with bronchiectasis, who had a repeat MDCT. Exclusion: CF, surgical lobectomy, traction bronchiectasis, or lobar opacification precluded measurement of a broncho-arterial ratio (BAR) The diagnostic and most recent MDCT scan were scored by a single scorer using the modified Reiff-score (with paediatric correction). Inter-observer variability was assessed using a weighted kappa-coefficient (K) of the modified Reiff-scores from a random sample of 30 CTs scored by a paediatric respiratory-physician and radiologist, blinded to the clinical data. Demographic, clinical and laboratory data, were collected. Univariable and multivariable logistic regression models were used to determine factors associated with radiographic resolution. Results: 141 children (70 females, median age-at-diagnosis = 5.0 years (IQR 2.6-7.3)), median time between CTs = 4.3 years (IQR 2.9-5.4) were included. Weighted Kappa between scorer were 0.86 (DM and radiologist) and 0.83 (DM and paediatric respiratory physician) respectively. Bronchiectasis resolved in 40% (57/141), whilst in 60% (84/141) scores were static/improved but not resolved/worsened. On univariate logistic regression, lower age-at-diagnosis (OR = 0.82, 95% CI 0.73-0.92) was associated with improved, whereas number of lobes affected (OR = 0.79, 95% CI 0.62-0.99), highest BA ratio (OR = 0.36, 95% CI 0.20-0.64), Modified Reiff-score (OR = 0.75, 95% CI 0.6-0.9), need for intravenous antibiotics (OR = 0.27, 95% CI 0.11-0.63), presence of Pseudomonas aeruginosa (PsA) (OR = 0.04, 95% CI 0.01-0.3) predicted decreased likelihood of resolution. On multivariate regression, age at diagnosis (OR adj = 0.82, 95% CI 0.70-0.95) and PsA (OR adj = 0.10 95% CI 0.01-0.86) were significant factors associated with resolution. Conclusion: As cylindrical bronchiectasis is potentially reversible with prognosis significantly associated with early diagnosis, and worsened by PsA in respiratory samples, early diagnosis and microbiological surveillance for PsA might improve outcomes.