Chest Infections: Chest Infections: Bronchieactasis |

Unsupervised Learning Technique Identifies Bronchiectasis Phenotypes With Distinct Clinical Characteristics and Prognosis FREE TO VIEW

Wei-jie Guan, PhD; Mei Jiang, PhD; Yong-hua Gao, PhD; Hui-min Li, BA; Gang Xu, PhD; Jin-ping Zheng, MD; Rongchang Chen, MD; Nanshan Zhong, MD
Author and Funding Information

State Key Laboratory of Respiratory Disease, National Clinical Research Center of Respiratory Disease, Guangzhou, China

Copyright 2016, American College of Chest Physicians. All Rights Reserved.

Chest. 2016;149(4_S):A104. doi:10.1016/j.chest.2016.02.109
Text Size: A A A
Published online

SESSION TITLE: Chest Infections: Bronchieactasis

SESSION TYPE: Original Investigation Poster

PRESENTED ON: Saturday, April 16, 2016 at 11:45 AM - 12:45 PM

PURPOSE: Unsupervised learning technique has allowed researchers to identify different phenotypes of diseases with complex manifestations. We sought to identify major phenotypes of bronchiectasis and characterize their clinical manifestations and prognosis.

METHODS: We conducted hierarchical cluster analysis to identify major clusters that best distinguished clinical characteristics of bronchiectasis based on Guangzhou Bronchiectasis Study database. Demographics, lung function, sputum bacteriology, etiology, radiology, disease severity, quality-of-life, cough scale and sensitivity, exercise tolerance, healthcare utilization and exacerbation frequency were assessed using established methodologies and compared among different clusters.

RESULTS: Data of 148 adults with stable bronchiectasis were included in factor analysis. Four clusters were identified. Cluster 1 (n=69) consisted of the youngest patients with predominantly mild and idiopathic bronchiectasis who had minor healthcare resource utility. Patients in cluster 2 (n=22), in which post-infectious bronchiectasis predominated (77.3%), had the longest duration of symptom, comparatively greater disease severity, poorer lung function, airway Pseudomonas aeruginosa colonization and frequent healthcare resource utilization. Cluster 3 consisted of relatively older subjects with shorter duration of symptom onset and mostly idiopathic bronchiectasis (87.5%). These patients predominantly harbored severe bronchiectasis associated with poor lung function and airway Pseudomonas aeruginosa colonization. Cluster 4 (n=41) constituted the oldest subjects with modest duration of symptom onset and moderate disease severity. Clusters 2 and 3 tended to harbor greater risks of bronchiectasis exacerbations (P=0.06) compared with clusters 1 and 4.

CONCLUSIONS: Identification of distinct phenotypes will lead to greater insights into the characteristics and prognosis of bronchiectasis.

CLINICAL IMPLICATIONS: Phenotyping bronchiectasis has allowed physicians to systematically appraise the heterogeneity and homogeneity of patients, which might offer new insights into future management of the disease.

DISCLOSURE: The following authors have nothing to disclose: Wei-jie Guan, Mei Jiang, Yong-hua Gao, Hui-min Li, Gang Xu, Jin-ping Zheng, Rongchang Chen, Nanshan Zhong

No Product/Research Disclosure Information




Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
PubMed Articles
Cystic fibrosis: a clinical view. Cell Mol Life Sci Published online Oct 5, 2016;
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543