Sputum Eosinophils Predict Annual Remission of the Frequent Exacerbator Phenotype in COPD: A Prospective Cohort Study.
The frequent-exacerbator (FE) phenotype guides COPD management, yet its year-to-year stability and the role of airway eosinophilia in FE remission are uncertain.
In a single-center prospective cohort (n = 289), patients were classified as FE (≥ 2 moderate or ≥ 1 severe exacerbation in the prior year; n = 88) or non-FE (n = 201) and followed 12 months. FE patients were labeled FE-persistent or FE-remitted at 1 year. Clinical data, spirometry, CT, fractional exhaled nitric oxide (FeNO), blood counts, and induced-sputum cytology were collected. Logistic regression identified correlates of remission; ROC/AUC with bootstrap validation and calibration assessed models.
FE prevalence was 30.4%; 58.0% of FE patients remitted. Compared with non-FE, FE had higher SE% and more eosinophilic/mixed sputum phenotypes. Within FE, remitters had better lung function and greater airway eosinophilia. Each 1% SE% increase independently predicted remission (OR 1.21, 95% CI 1.03-1.47); SE% ≥ 3% tripled the odds (OR 3.76, 95% CI 1.29-11.91). Prior severe exacerbations and CT-defined chronic bronchitis predicted persistence (ORs 0.21-0.27). Models showed good discrimination (AUC 0.785-0.799; bootstrap-corrected 0.750-0.773) and calibration (Brier 0.179).
FE status is dynamic; over half remit within a year. Airway eosinophilia-especially SE% ≥ 3%-independently associates with FE remission, while prior severe exacerbations and CT-chronic bronchitis indicate persistence. Incorporating sputum cytology with history and imaging may enable earlier re-stratification and treatable-trait-guided COPD care.
In a single-center prospective cohort (n = 289), patients were classified as FE (≥ 2 moderate or ≥ 1 severe exacerbation in the prior year; n = 88) or non-FE (n = 201) and followed 12 months. FE patients were labeled FE-persistent or FE-remitted at 1 year. Clinical data, spirometry, CT, fractional exhaled nitric oxide (FeNO), blood counts, and induced-sputum cytology were collected. Logistic regression identified correlates of remission; ROC/AUC with bootstrap validation and calibration assessed models.
FE prevalence was 30.4%; 58.0% of FE patients remitted. Compared with non-FE, FE had higher SE% and more eosinophilic/mixed sputum phenotypes. Within FE, remitters had better lung function and greater airway eosinophilia. Each 1% SE% increase independently predicted remission (OR 1.21, 95% CI 1.03-1.47); SE% ≥ 3% tripled the odds (OR 3.76, 95% CI 1.29-11.91). Prior severe exacerbations and CT-defined chronic bronchitis predicted persistence (ORs 0.21-0.27). Models showed good discrimination (AUC 0.785-0.799; bootstrap-corrected 0.750-0.773) and calibration (Brier 0.179).
FE status is dynamic; over half remit within a year. Airway eosinophilia-especially SE% ≥ 3%-independently associates with FE remission, while prior severe exacerbations and CT-chronic bronchitis indicate persistence. Incorporating sputum cytology with history and imaging may enable earlier re-stratification and treatable-trait-guided COPD care.
Authors
Huang Huang, Hu Hu, Gong Gong, Wang Wang, Chen Chen, Yang Yang, Lan Lan, Ouyang Ouyang, Zhao Zhao, Zhao Zhao
View on Pubmed