Pulmonary rehabilitation can alter the respiratory microbiome in patients with COPD

Abstract Background As one of the main causes ofchronic obstructive pulmonary disease (COPD), respiratory microbiome is of great significance to the disease progression. Meanwhile, pulmonary rehabilitation (PR), as an effective nonpharmacologicintervention for COPD, evaluation of the efficacy is limited to the improvement of symptoms and lung function, which is one-sided to a certain extent. Therefore, an individualized 2-week intensive PR programme was established in the present study for patients with COPD to investigate whether this short course of PR can have impact on respiratory microbiome in patients. Methods The patients with stable COPD who participated in this study underwent short-term, high-intensity and center-based PR programme, which included two major components: respiratory muscle training and vibration expectoration. Sputum samples were collected before and after the PR to detect alterations of respiratory microbiome by metagenomic sequencing. The main purpose was to observe changes of composition, function and resistance genes in respiratory microbiomebefore and after PR, and explore the relationship of respiratory microbiome with lung function and symptoms. Results After the 2-week intensive PR, the abundance of Firmicutes , Actinobacteria and Bacteroidetes decreased, while that of Proteobacteria increased at the phylum level; the abundance of Streptococcus and Prevotella decreased, while that of Neisseria increased at the genus level. Concomitantly, it was found that Firmicutes , Actinobacteria, Streptococcus and Prevotella were negatively correlated with lung function and positively correlated with CAT score, while Neisseria was positively correlated with lung function and negatively correlated with CAT score. Based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) database, the functional abundance of most respiratory microbiome decreased after 2-week PR. Simultaneously, the species function abundance was negatively correlated with the lung function of COPD. In addition, the resistance gene fluctuated with the attribution of the species, and the macrolide resistance genes of COPD were significantly reduced after 2-week PR. Conclusions The short course of intensive PR can alter the composition, function and antibiotic resistance genes of respiratory microbiome in patients with COPD, which plays a role by promoting the self-regulation ability of patients. This study provides a novel, gentle and non-pharmacologic intervention for the regulation of microbiome in the future.