2162 Identification of Phenotypes in Allergic Bronchopulmonary Aspergillosis Using Cluster Analysis

Thursday, 15 October 2015
Hall D1 Foyer (Floor 3) (Coex Convention Center)

Tsuyoshi Oguma, MD , Division of Pulmonary Medicine, Department of Medicine, Tokai University School of Medicine, Kanagawa, Japan

Jun Tanaka, MD , Division of Pulmonary Medicine, Department of Medicine, Tokai University School of Medicine, Kanagawa, Japan

Katsuyoshi Tomomatsu, MD , Division of Pulmonary Medicine, Department of Medicine, Tokai University School of Medicine, Kanagawa, Japan

Koichiro Asano, MD , Division of Pulmonary Medicine, Department of Medicine, Tokai University School of Medicine, Kanagawa, Japan

Background: Allergic bronchopulmonary aspergillosis (ABPA) is known as type I and III allergic disease for Aspergillus. Clinical presentation of ABPA ranges from asymptomatic mucus plugs to severe asthmatic symptoms or destructive and fibrotic lung disease. The heterogeneity may reflect not only the different clinical stages of the disease, but also different phenotypes. We herein attempt to identify phenotypes in ABPA using cluster analysis.  

Methods: We analyzed the data of 332 patients with possible ABPA from national-wide survey in Japan executed between 2013 and 2014. Definition of possible ABPA were, (1) positive skin test or specific IgE for Aspergillus, and (2) either a) positive precipitation or IgG antibody for Aspergillus or b) mucoid impaction or bronchiectasis in chest computed tomography. Non-hierarchical cluster analysis using k-means method was performed.

Results: Three clusters were identified. Cluster 1 (n=141) included the patients with later age at onset (mean ages, 68 years), female-dominance, and less frequent prevalence of asthma (76%). The patients in cluster 2 (n=95) were middle age at onset (55 years), female-dominant, and showed lower values of total serum IgE. Cluster 3 (n=96) was characterized with early-onset (37 years), male-dominance, and frequent recurrences (59%).

Conclusions: Three distinct clinical phenotypes were identified characterized by ages of onset, gender, asthma prevalence, total serum IgE levels, and the frequency of recurrences.