When conventional microbiology goes silent, clinicians face a dangerous diagnostic gap: tuberculosis and nontuberculous mycobacterial lung disease look nearly identical on imaging, yet demand entirely different treatment regimens. A diagnostic tool that can simultaneously identify the culprit pathogen and characterize the surrounding microbial environment could meaningfully shorten the window of clinical uncertainty.
This retrospective cohort study analyzed bronchoalveolar lavage fluid from 74 patients with clinically confirmed pulmonary mycobacterial disease — 62 tuberculosis cases and 12 with NTM-pulmonary disease — using metagenomic next-generation sequencing (mNGS). Culture-based methods returned zero positives across all tested patients in routine clinical practice, exposing a critical failure point. Against that baseline, mNGS achieved 98.4% positivity in TB cases and 100% in NTM-PD cases. In a supplementary bias-adjusted analysis that excluded mNGS from the diagnostic classification process, sensitivity remained at 97.8% among 46 independently confirmed TB cases. Conventional comparators performed substantially worse: acid-fast bacilli smear reached only 8.82% positivity, while T-SPOT.TB and Xpert MTB/RIF achieved 71.43% and 69.23%, respectively, though these were applied to smaller subsets. Notably, mNGS also detected co-infecting non-mycobacterial pathogens in over 62% of patients, flagging polymicrobial disease that would otherwise go unrecognized.
This work sits within a rapidly growing body of evidence establishing mNGS as a high-sensitivity rescue tool for culture-negative infectious diseases, particularly in immunocompromised and difficult-to-diagnose populations. Several important constraints temper enthusiasm: the cohort is small at 74 patients, the design is retrospective, and specimen processing was limited to a single fluid type. The near-zero culture positivity rate is also unusually low, raising questions about patient selection. mNGS remains expensive and computationally demanding, limiting routine deployment. Still, the simultaneous microbiome profiling capability — revealing polymicrobial burden in nearly two-thirds of patients — represents a genuinely additive diagnostic dimension that standard tests cannot offer. This finding is confirmatory rather than paradigm-shifting, but meaningfully strengthens the evidentiary case for mNGS in challenging pulmonary mycobacterial workups.