Abstract
Tuberculosis (TB) and non-tuberculous mycobacterial (NTM) infections, both caused by acid-fast bacteria, exhibit overlapping clinical presentations and radiological findings, making accurate and swift differentiation imperative for appropriate patient management. The ability to promptly distinguish between TB and NTM is pivotal, particularly in healthcare settings where TB patients require isolation. So, there is a pressing need to identify reliable biomarkers capable of delineating these conditions. In this study, we carefully analyzed a group of 108 patients, including 38 with TB, 29 with NTM, and 41 with other respiratory diseases. We used bacterial 16 S rRNA sequencing to examine the relative amounts of different microbial species in each patient group. We also used advanced methods like co-occurrence analysis, mutually exclusive analysis, and decision tree analysis to explore the complex patterns within the microbial communities. Our investigation revealed intriguing associations between microbial taxa and disease entities. In TB patients, prominent species were Enterococcus faecalis, Streptococcus mutans, and Snodgrassella alvi, while in NTM, prominent species were Cariobacterium hominis and Prevotella nigrescens. The absence of Mobiluncus curtisii indicated a higher probability of NTM, especially if Olsenella phocaeensis was also absent. This comprehensive analysis unveiled distinct microbial signatures that serve as discerning markers for discriminating TB, NTM infections, and other respiratory ailments. By elucidating microbial patterns unique to each condition, our findings offer valuable insights into the development of diagnostic strategies and therapeutic interventions tailored to specific respiratory infections.