Abstract

The homeostasis of pulmonary microbiota is crucial in maintaining human health and modulating disease progression. The stability of pulmonary microbial flora may be associated with diabetes, yet the specific alterations remain poorly characterized. This retrospective observational study aims to analyze the profiles in pulmonary microbiota between individuals with and without diabetes, using metagenomic next-generation sequencing (mNGS). A total of 632 patients were sequentially enrolled, including 77 patients with both pneumonia and diabetes, 46 patients without either pneumonia or diabetes, 499 patients with pneumonia but without diabetes, and 10 diabetic patients without pneumonia. Pathogens in bronchoalveolar lavage fluid (BALF) specimens were detected using mNGS (DNA). The lung microbiota of diabetic individuals significantly differs from that of non-diabetic individuals in the non-lower respiratory tract infection (non-LRTI) cohort. Parvimonas was more abundant in the diabetic group. Compared to non-diabetic patients with LRTI, those with diabetes and LRTI showed an increased relative abundance of Parvimonas, but decreased relative abundances of Prevotella and Malassezia. Our analysis revealed a negative correlation between Parvimonas and Malassezia, alongside a positive association of Parvimonas with the expression of antimicrobial resistance genes ICR-Mc and RbpA. This suggests a potential association between Parvimonas enrichment and microbial dysbiosis during infection, although the underlying host–microbe interactions require further validation. Interestingly, Parvimonas abundance showed no significant association with HbA1c levels. Our findings suggest that Parvimonas enrichment is associated with diabetes-related alterations in lower respiratory tract microbiota. Whether microbiota-associated alterations represent clinically actionable targets in diabetic patients with pulmonary infections remains to be determined in prospective and interventional studi