Varun Shah1, Bryan Husta2, Atul Mehta3, Soumya Ashok2, Oki Ishikawa1, Guillaume Stoffels4, Jennifer Hartzband1, Richard Lazzaro5, Byron Patton5, Viera Lakticova1, Suhail Raoof6. 1. Pulmonary and Critical Care Division, Lenox Hill Hospital, New York, NY. 2. Pulmonary Division, Memorial Sloan Kettering Cancer Center, New York, NY. 3. Pulmonary Division, Cleveland Clinic, Cleveland, OH. 4. Department of Biostatistics, Lenox Hill Hospital, New York, NY. 5. Department of Thoracic Surgery, Lenox Hill Hospital, New York, NY. 6. Pulmonary and Critical Care Division, Lenox Hill Hospital, New York, NY. Electronic address: suhailraoof@gmail.com.
Abstract
OBJECTIVE: The aim of this study was to assess any association between use of inhaled corticosteroids (ICS) and tracheobronchomalacia (TBM). METHODS: This study was a retrospective analysis of patients with asthma and COPD, with and without TBM. Patients were diagnosed with TBM on the basis of CT imaging, flexible bronchoscopy, or both. Patients were deemed to be on ICS if they had been receiving treatment for at least 3 months. Simple logistic regression models were used to assess the association between TBM status and each proposed factor. A multivariable logistic regression model was used to assess the association between TBM and steroid dose. RESULTS: A total of 463 patients with COPD (n = 153) and asthma (n = 310) were studied. In multivariate analysis, the odds of TBM were 3.5 times higher in patients on high-dose steroids compared with patients not on steroids (OR, 3.5; 95% CI, 1.4-8.5; P = .007). Age (P < .0001), presence of gastroesophageal reflux disease (P < .0001), use of long-acting muscarinic antagonists (P < .0001), and type of pulmonary disease (P = .002) were also associated with TBM. In patients using ICS, the odds of having TBM were 2.9 times greater in patients on high-dose inhaled steroids compared with those on low-dose inhaled steroids (OR, 2.9; 95% CI, 1.2-7.1; P = .02). Age (P = .003), presence of gastroesophageal reflux disease (P = .002), use of long-acting muscarinic antagonists (P = .004), type of ICS (P = .04), and number of months on ICS (P < .0001) were all associated with TBM. CONCLUSIONS: There was a significant association between ICS use in higher doses for a longer duration of time with TBM. Prospective randomized controlled trials are needed to show causality of this observed association.
OBJECTIVE: The aim of this study was to assess any association between use of inhaled corticosteroids (ICS) and tracheobronchomalacia (TBM). METHODS: This study was a retrospective analysis of patients with asthma and COPD, with and without TBM. Patients were diagnosed with TBM on the basis of CT imaging, flexible bronchoscopy, or both. Patients were deemed to be on ICS if they had been receiving treatment for at least 3 months. Simple logistic regression models were used to assess the association between TBM status and each proposed factor. A multivariable logistic regression model was used to assess the association between TBM and steroid dose. RESULTS: A total of 463 patients with COPD (n = 153) and asthma (n = 310) were studied. In multivariate analysis, the odds of TBM were 3.5 times higher in patients on high-dose steroids compared with patients not on steroids (OR, 3.5; 95% CI, 1.4-8.5; P = .007). Age (P < .0001), presence of gastroesophageal reflux disease (P < .0001), use of long-acting muscarinic antagonists (P < .0001), and type of pulmonary disease (P = .002) were also associated with TBM. In patients using ICS, the odds of having TBM were 2.9 times greater in patients on high-dose inhaled steroids compared with those on low-dose inhaled steroids (OR, 2.9; 95% CI, 1.2-7.1; P = .02). Age (P = .003), presence of gastroesophageal reflux disease (P = .002), use of long-acting muscarinic antagonists (P = .004), type of ICS (P = .04), and number of months on ICS (P < .0001) were all associated with TBM. CONCLUSIONS: There was a significant association between ICS use in higher doses for a longer duration of time with TBM. Prospective randomized controlled trials are needed to show causality of this observed association.