Diagnostic performance of metagenomic sequencing in patients with suspected infection: a large-scale retrospective study

Autor: Ziyang Li, Li Tan, Jialiang Zhang, Qichen Long, Zhiyang Chen, Zhongyuan Xiang, Weimin Wu, Zhe Guo, Huifang Liu, Bingxue Hu, Bin Yang, Min Hu
Jazyk: angličtina
Rok vydání: 2024
Předmět:
Zdroj: Frontiers in Cellular and Infection Microbiology, Vol 14 (2024)
Druh dokumentu: article
ISSN: 2235-2988
27789446
DOI: 10.3389/fcimb.2024.1463081
Popis: BackgroundMetagenomic next-generation sequencing (mNGS) has been widely reported to identify pathogens in infectious diseases (IDs). In this work, we intended to investigate the diagnostic value and clinical acceptance of paired-samples mNGS as compared to the culture method.MethodsA total of 361 patients with suspected infection were retrospectively included. With reference to the clinical diagnosis, we compared the diagnostic performance and clinical acceptance in pathogen detection between mNGS and culture tests. Moreover, the pathogen concordance of paired blood and respiratory tract (RT) samples in mNGS assay was investigated.ResultsAmong 511 samples, 62.04% were shown to be pathogen positive by mNGS, and that for clinical diagnosis was 51.86% (265/511). When compared to culture assay (n = 428), mNGS had a significantly higher positivity rate (51.87% vs. 33.18%). With reference to the clinical diagnosis, the sensitivity of mNGS outperformed that of culture (89.08% vs. 56.72%). Importantly, mNGS exhibited a clinically accepted rate significantly superior to that of culture. In addition, the mNGS result from 53 paired blood and RT samples showed that most pairs were pathogen positive by both blood and RT, with pathogens largely being partially matched.ConclusionThrough this large-scale study, we further illustrated that mNGS had a clinically accepted rate and sensitivity superior to those of the traditional culture method in diagnosing infections. Moreover, blood and paired RT samples mostly shared partial-matched positive pathogens, especially for pathogens with abundant read numbers in RT, indicating that both blood and RT mNGS can aid the identification of pathogens for respiratory system infection.
Databáze: Directory of Open Access Journals
Externí odkaz: