Diagnostic value of metagenomic next-generation sequencing for bronchoalveolar lavage diagnostics in patients with lower respiratory tract infections

• mNGS identified pathogens in 92.7% of patients suspected of LRTIs, significantly outperforming CMT which only identified pathogens in 58.8% of cases. • The sensitivity, specificity, PPV, and NPV of mNGS were all significantly higher than those of CMT, with mNGS demonstrating a sensitivity of 99.7% compared to 57.1% for CMT. • The study detailed a diverse range of pathogens causing LRTIs, with significant differences observed in pathogen distribution between adults and pediatric patients, and varying severity associated with different types of infections. Current diagnostic methods of lower respiratory tract infections (LRTIs) often lack specificity, underscoring the necessity for advanced technologies such as metagenomic next-generation sequencing (mNGS). This retrospective study compared bronchoalveolar lavage fluid (BALF) analysis using mNGS and conventional microbiological tests (CMT) to evaluate their effectiveness in pathogen identification and alignment with clinical diagnoses. In this study involving 369 patients suspected of LTRIs, mNGS identified pathogens in 342 cases (92.7%), showing superior diagnostic performance compared to CMT (58.8%). The positive agreement and negative agreement rates of mNGS were 92.7% and 96.3%, respectively, both significantly higher than those of CMT (both p<0.001). The sensitivity, specificity, positive predictive value, and negative predictive value of mNGS were significantly higher than those of CMT, with values of 99.7% vs. 57.1%, 68.4% vs. 26.3%, 96.5% vs. 87.1%, and 96.3% vs. 6.3%, respectively (all p<0.001). Pathogen detection rates among the patients showed that 89.7% had evidence of LRTIs, with bacterial infections (20.1%), mycoplasma (13.6%), mycobacterium (4.3%), fungal (4.1%), viral (3.3%), and mixed infections (44.4%) being the most common. Furthermore, the study also differentiated the distribution of pathogens between adults and pediatric patients, and assessed the impact of pathogen types on severe outcomes using multivariate logistic regression, revealing that viral and fungal infections were more likely associated with severe symptoms, whereas mycoplasma infections typically presented with milder symptoms. BALF mNGS proves effective for rapid, comprehensive pathogen detection in LRTIs, warranting its early use for enhanced diagnosis and management, especially across different age groups.