Asthma is a heterogeneous disease and has been classified into several phenotypes on the basis of allergic, inflammatory and clinical manifestations, such as atopic or non-atopic, eosinophilic or neutrophilic, and severe or non-severe subtypes. Cluster analysis is the task of grouping a set of objects in such a way that objects in the same group are more similar to each other than to those in other groups. Cluster analysis is applied to development of asthma sub-phenotypes and demonstrated the differences in clinical response to treatment, distinct phenotypes of severe asthma, clustering in extended populations including both asthma and chronic obstructive pulmonary disease, and clusters using additional parameters such as inflammatory biomarkers and obesity. In the present study, we investigated clusters reflecting the prognosis of asthma in terms of exacerbation over one or more years.
Methods
Clinical and demographic data on 1843 asthmatics registered in an asthma cohort in Korea were analyzed retrospectively. They were ethnic Koreans. Asthma was diagnosed by physicians on the basis of the Global Initiative for Asthma (GINA) guidelines. Among them, 632 subjects, who were regularly followed up for longer than 1 year, were included for the analysis after excluding current smokers and ex-smokers of 10 pack year or more. At the baseline visit, demographic information such as enrollment age, sex, BMI, asthma onset age, asthma duration, and smoking amount was collected. Uniform cluster analysis method was applied to each population using two-tiered approach, including hierarchical cluster analysis and K-mean cluster analysis.
Results
The subjects were classified into 4 major clusters using Ward’s method. Age, age of onset, duration of asthma, BMI, atopic status, FEV1%, and ΔFEV1 were robustly different among 4 clusters. Clinical characteristics of 4 clusters are as below: Cluster 1, early-onset with high frequency of atopy and relatively well preserved FEV1; Cluster 2, early-onset with long duration and moderately impaired FEV1; Cluster 3, middle age onset with short duration and severely impaired FEV1; Cluster 4, old age onset with high BMI, but well preserved FEV1. The annual average frequency of exacerbation was significantly different between the clusters (p=0.019), and markedly higher in the cluster 3 (0.46/year).
Conclusion
Exacerbation-prone asthma phenotype is found by using cluster analysis and clinical characteristics of the phenotype are middle-age onset, short duration, and severely impaired FEV1. Noticing this asthma phenotype can be useful in clinical practice to predict asthma exacerbation.