Diagnostic utility of metagenomic next-generation sequencing for tissue in patients with suspected infectious diseases

Metagenomic next-generation sequencing (mNGS) was suggested to potentially replace traditional microbiological methods because of its comprehensiveness. However, the diagnostic utility of mNGS for tissue hasn't been fully explored, especially for patient with HIV infection. HIV-positive and negative patients with suspected infectious diseases who performed tissue mNGS and conventional microbiological tests (CMTs) were retrospectively enrolled between October, 2020 and May 2024. The microbial spectrum of tissue mNGS and CMTs was analyzed, and the diagnostic accuracy and consistency of mNGS and CMTs for tissue were compared. The related factors of positive rate of mNGS was analyzed. Of 70 patients with suspected infectious diseases, 44 cases were confirmed with the infectious diseases. Among 44 patients with infectious diseases, aerobic bacteria (36.4%) was the most common detected pathogen, followed by mycobacterium tuberculosis (MTB, 18.2%), non-tuberculous mycobacteria (NTM, 13.6%) and fungus (11.4%). The sensitivity of tissue mNGS (72.7%, 95%CI 56.9%-84.5%) was significantly higher than that that in tissue CMTs (29.5%, 95%CI 17.2%-45.4%) (p<0.001), but the specificity was not statistically significant(P = 0.656). mNGS demonstrated higher detection rates than CMTs in the case with single microbial infections (70.0% vs. 30.0%; p<0.01). For the case with multiple microbial infections, the detection rates of mNGS and CMTs was 100.0% and 25.5% (p=1.000), respectively. Both positive mNGS and CMTs were observed in 22.7% patients with infectious diseases, and sole positive mNGS and sole positive CMTs were observed in 50.0% and 6.8% patients, respectively. There were no statistically differences in age, gender, HIV infection, PCT levels, neutrophil counts, CD4⁺ lymphocyte count and antibiotic exposure between mNGS positive and mNGS negative groups (P > 0.05). Tissue mNGS could provide a higher sensitivity, more robust and broader method for pathogen identification by comparison with CMTs. However, CMTs shouldn't be ignored since the low consistency between CMTs and mNGS.