Background: We conducted this retrospective study to reveal the accuracy of metagenomic next-generation sequencing (mNGS) for diagnosing osteoarticular infections from fresh abscess specimens obtained from patients in an HIV-naive population. Methods: We retrospectively analyzed hospital records at three participating TB-specialized hospitals for patients admitted with suggestive diagnoses of osteoarticular tuberculosis between January 2018 and August 2019. Abscess specimens obtained from each patient were tested via pathogen culture, GeneXpert Mycobacterium tuberculosis (MTB)/rifampicin (RIF), and mNGS assay. Results: A total of 82 abscess samples were collected from patients with osteoarticular infections, including 53 cases with (64.6%) bacterial, 21 (25.6%) with mycobacterial, 7 (8.5%) with fungal, and 1 (1.2%) with actinomycetal organisms detected. Analysis of mNGS assay results identified potential pathogens in all cases, with M. tuberculosis complex (MTBC) most frequently isolated, followed by Staphylococcus aureus and Brucella melitensis. Conventional culture testing identified causative pathogens in only 48.4% of samples, a significantly lower rate than the mNGS pathogen identification rate (100%, p < 0.01). Culture-positive group specimens yielded significantly greater numbers of sequence reads than did culture-negative group specimens (p < 0.01). Of patients receiving surgical interventions and mNGS-guided treatment, 76 (92.7%) experienced favorable outcomes by the time of follow-up assessment at 3 months post-treatment. Notably, MTBC detection in two patients experiencing treatment failure suggests that they had mixed infections with MTBC and other pathogens. Conclusion: Results presented here demonstrate that mNGS has a greater pathogen detection rate in osteoarticular infections than conventional culture-based methods.
Keywords: diagnosis; infection; metagenomic; metagenomic next-generation sequencing; osteoarticular.
Copyright © 2020 Zhao, Tang, Liu, Zhou, Fan, Yan, Qin and Pang.