Background: In France, data regarding epidemiology and management of severe asthma are scarce. The objective of this study was to describe asthma phenotypes using a cluster analysis in severe asthmatics recruited in a real world setting.
Methods: The study design was prospective, observational and multicentric. The patients included were adults with severe asthma (GINA 4-5) followed-up in French Non Academic Hospital between May 2016 and June 2017. One hundred and seven physicians included 1502 patients. Both sociodemographic and clinical variables were collected. Hierarchical cluster analysis was performed by the Ward method followed by k-means cluster analysis on a population of 1424 patients.
Results: Five clusters were identified: cluster 1 (n = 690, 47%) called early onset allergic asthma (47.5% with asthma before 12 years), cluster 2 (n = 153, 10.5%): obese asthma (63.5% with BMI > 30 kg/m2), cluster 3 (n = 299, 20.4%): late-onset asthma with severe obstructive syndrome (89% without atopy), cluster 4 (n = 143, 9.8%): eosinophilic asthma (51.7% had more than 500 eosinophils/mm3), and cluster 5 (n = 139, 9.5%): aspirin sensitivity asthma (63% had severe asthma attacks).
Conclusions: In our population of adults with severe asthma followed by pulmonologists, five distinct phenotypes were identified and are quite different from those mentioned in previous studies.
Keywords: Cluster analysis; Phenotypes; Severe asthma.