Cluster analysis identifies characteristic phenotypes of asthma with accelerated lung function decline

Abstract

Objective: While the majority of individuals with asthma retain normal lung function over time, some exhibit accelerated lung function decline. Preservation of lung function is an important aspect of asthma management. Whether the asthma guidelines can prevent lung function decline remains controversial. This study was performed to determine the distribution of asthmatic subjects with greater lung function decline and to identify characteristic clinical features of such subjects treated in accordance with clinical guidelines by using hierarchical cluster analysis. Methods: Eighty-six asthmatic subjects without a history of smoking were assessed with respect to eight variables selected from clinical phenotypes by using step-wise multiple regression analysis. Hierarchical cluster analysis using Ward’s method generated a dendrogram for estimation of the number of clusters within the population and the differences between them. Results: Three distinct clusters were identified. Cluster 1 (n = 40) comprised women with late-onset asthma. Cluster 2 (n = 17) comprised subjects with early-onset asthma, atopy and long disease duration. Cluster 3 (n = 29) predominantly comprised older men who had late-onset asthma, a lower prevalence of exacerbation and a lower predicted % forced expiratory volume in 1 s (FEV1) at baseline. Subjects in cluster 3 showed a mean decline in FEV1 of 69 mL/year, which was the greatest lung function decline among the three clusters. Conclusion: We identified a subgroup of patients with accelerated lung function decline despite appropriate asthma treatment based on guidelines constructed by using subjective symptoms.