Inflammatory Phenotypes and Associated Factors in Bronchial Asthma Patients: a Cross-sectional Survey

Abstract

Background The various inflammatory phenotypes of asthma, a common heterogeneous respiratory disease, are closely related to the pathogenesis, treatment, and prognosis of the disease. So screening the associated factors of inflammatory phenotypes will be helpful for the evaluation of patient condition and delivery of individualized diagnosis and treatment services. Objective To explore the distribution of inflammatory phenotypes and associated factors in bronchial asthma patients, providing a basis for the implementation of individualized diagnosis and treatment of the disease. Methods A cross-sectional study design was used. Clinical data of bronchial asthma outpatients and inpatients (n=184) were collected from the First Affiliated Hospital of Henan University of Chinese Medicine from November 2018 to December 2020. Inflammatory phenotypes in the patients were classified into four categories according to the type of inflammatory cells in the induced sputum: neutrophilic asthma (NA) , eosinophilic asthma (EA) , mixed granulocytic asthma (MA) , and paucigranulocytic asthma (PA) . Factors possibly associated with each of the inflammatory phenotypes were screened by three statistical methods (univariate analysis, multivariate Logistic regression analysis, and Spearman rank correlation analysis) , and were determined as the associated factors if they had significant associations with the phenotype by two of the aforementioned three methods. Results The prevalence of NA, EA, MA, and PA was 45.7% (84/184) , 20.7% (38/184) , 20.7% (38/184) , and 13.0% (24/184) , respectively. Univariate analysis showed that the prevalence of allergic rhinitis and fractioned exhaled nitric oxide (FeNO) level differed significantly between NA and non-NA patients (P<0.05) . And they also varied significantly between EA and non-EA patients (P<0.05) . There was significant difference in FeNO level between MA and non-MA patients (P<0.05) . There were significant differences in mean age, prevalence of previous respiratory disease and mean FeNO level between PA and non-PA patients (P<0.05) . Multivariate Logistic regression analysis showed that allergic rhinitis〔OR=0.417, 95%CI (0.205, 0.848) 〕 and FeNO〔OR=0.978, 95%CI (0.968, 0.989) 〕were associated with NA (P<0.05) ; FeNO〔OR=1.017, 95%CI (1.009, 1.025) 〕 was associated with EA (P<0.05) ; FeNO〔OR=1.007, 95%CI (1.000, 1.014) 〕was associated with MA (P<0.05) ; BMI〔OR=1.165, 95%CI (1.015, 1.337) 〕 and FeNO〔OR=0.981, 95%CI (0.965, 0.998) 〕were associated with PA (P<0.05) . Spearman rank correlation analysis indicated that NA prevalence decreased with increased allergic rhinitis prevalence and FeNO level (rs=-0.244, -0.361, P<0.05) ; EA prevalence increased with increased allergic rhinitis prevalence and FeNO level (rs=0.157, 0.341, P<0.05) ; MA prevalence increased with increased FeNO (rs=0.236, P<0.05) ; PA prevalence decreased with older age, prevalence of previous respiratory disease and increased FeNO (rs=-0.156, -0.163, -0.159, all P<0.05) . Based on the above analyses, allergic rhinitis and FeNO were associated factors for both EA and NA; FeNO was associated factors of MA; age, prevalence of previous respiratory disease and FeNO were associated factors of PA. Conclusion NA accounted for the largest percentage of the inflammatory phenotypes, while PA accounted for the least. FeNO was the associated factor for each inflammatory phenotype. It has specificity in recognizing EA and MA. FeNO combined with allergic rhinitis was associated with NA and EA. FeNO combined with age was associated with PA.

Keywords

• asthma|inflammation|phenotype|inflammatory phenotype|cross-sectional studies|root cause analysis