Abstract / Description
BACKGROUND: Our initial goal was to assess the clinical efficacy of using plasma cell-free DNA (cfDNA) to perform metagenomic next-generation sequencing (mNGS) to detect suspected infections. METHODS: We retrospectively analyzed 425 patients who underwent plasma cfDNA mNGS. Of them, 84 patients had various systemic infections, 69 were ruled out of infectious diseases, and 272 had bloodstream infections. Conventional microbiological tests (CMTs) and plasma cfDNA mNGS were conducted concurrently in clinical practice. The sensitivity and specificity of the two techniques were examined based on the final diagnosis. RESULTS: The total positive rate (276/425) for the cfDNA mNGS test indicated the presence of microorganisms. The mean length of stay in the hospital for mNGS-positive patients was longer than for mNGS-negative patients (P = 0.0079). Compared with patients with negative mNGS tests, those with positive mNGS tests have significantly lower white blood cell counts in peripheral blood (P = 0.0085). Among all the disorders in our patients, bloodstream infection (272/425, 64.0%) accounted for the most significant percentage. The diagnostic sensitivity of mNGS is also higher than that of CMTs (72.8% vs. 32.9%). However, its diagnostic specificity is lower than CMT’s (75.4% vs. 85.5%). Bacteria were the most frequently identified potential pathogens by mNGS. Klebsiella pneumoniae (n = 46) was the most common pathogen. Candida albicans is the most common fungal infection (n = 18). Human cytomegalovirus (n = 45) is the most common viral infection. The detection rate of mNGS in empirically treated groups was significantly higher than in non-empirically treated groups(71.9% vs. 51.4%, p