LAUSR

Background Due to limitations of traditional microbiological methods and the presence of the oropharyngeal normal flora, there are still many pathogens that cause lower respiratory tract infections (LRTIs) cannot be detected. Metagenomic next-generation sequencing (mNGS) has the potential capacity to solve this problem. Methods This retrospective study successively reviewed 77 patients with LRTI and 29 patients without LRTI admitted to Tianjin Medical University General Hospital, China from August 2020 to June 2021. Pathogens in bronchoalveolar lavage fluid (BALF) specimens were detected adopting mNGS and traditional microbiological assays. The diagnostic performance of pathogens was compared between mNGS and BALF culture. The value of mNGS for aetiological and clinical impact investigation in LRTI was also evaluated. Results Among 77 patients with LRTI, 22.1%, 40.3%, and 65.0% of cases were detected as definite or probable pathogens by culture, all conventional microbiological tests, and mNGS, respectively. Using the final diagnosis as a gold standard, mNGS exhibited a sensitivity of 76.6% (95% confidence interval [CI], 65.6–85.5%), which was considerably superior to that of BALF culture (76.6% vs 18.2%; P < 0.01); specificity of 79.3% (95% CI, 60.3–92.0%), which was similar (79.3% vs 89.7%; P = 0.38); positive-predictive value of 90.8% (95% CI, 81.0–96.5%), and negative-predictive value of 56.1% (95% CI, 39.7–71.5%). According to our data, mNGS identified potential microorganisms in 66.7% (42/63) of culture-negative samples. Among 59 patients with pathogens identified by mNGS, conventional microbiological methods confirmed pathogenic infections in less than half (28/59) cases. Within the 77 patients, 34 (44.2%) patients received pathogen-directed therapy, 7 (9.1%) patients underwent antibiotic adjustment, and 3 (3.9%) patients stopped using antibiotics due to mNGS results. Conclusion mNGS exhibits high accuracy in diagnosing LRTI, and combine with traditional microbiological tests, causative pathogens can be detected in approximately 70.0% of cases, thus yields a positive effect on antibiotic application.