Clinical Predictors of High Blood Eosinophils in Chronic Obstructive Pulmonary Disease

Mei Yang,&ast; Ting Yang,&ast; Xiaoou Li,&ast; Diandian Li, Zenglin Liao, Yongchun Shen, Dan Xu, Lei Chen, Fuqiang Wen Department of Respiratory and Critical Care Medicine, West China Hospital, West China School of Medicine, Sichuan University, Chengdu, Sichuan, 610041, People’s Republic of China&ast;These authors contributed equally to this workCorrespondence: Lei Chen; Fuqiang WenDepartment of Respiratory and Critical Care Medicine, West China Hospital, West China School of Medicine, Sichuan University, Chengdu, Sichuan, 610041, People’s Republic of ChinaEmail lchens@126.com; wenfuqiang@scu.edu.cnPurpose: Elevated blood eosinophils have been implicated in chronic obstructive pulmonary disease (COPD) progression and exacerbation. We aim to investigate clinical predictors of high blood eosinophils in a Chinese COPD cohort.Patients and Methods: We conducted a retrospective cohort study in Sichuan province, a Southwest province with high prevalence of COPD in China. All patients in this cohort were extracted from the Chinese Pulmonary Health study, a large cross-sectional study on COPD epidemiology in China. Demographics, personal and family history, living condition, spirometry and blood eosinophil counts were obtained. Univariate and multiple linear regression analyses were performed to determine predictors of high blood eosinophils.Results: A total of 375 COPD patients were included in this cohort. The median absolute blood eosinophil count was 138.8 cells/μL, and the prevalence of COPD with high blood eosinophils was 66.7% and 14.7% when using the thresholds of 100 cells/μL and 300 cells/μL, respectively. Univariate analyses indicated that male gender, lower body mass index, high-density lipoprotein (HDL), lower family income, raising pets and biomass use were significantly associated with high blood eosinophils (p < 0.05). Multiple linear regression model further revealed male gender (unstandardized coefficient (B)=66.125, 95% confidence