Consistency between metagenomic next-generation sequencing versus traditional microbiological tests for infective disease: systemic review and meta-analysis

Pathogen identification is essential in sepsis and septic shock. Metagenomic next-generation sequencing (mNGS) is a novel pathogen detection method with several advantages over traditional tests. However, the consistency between mNGS and traditional pathogen tests requires further investigation. We aimed to assess the consistency between mNGS and traditional pathogen tests and to identify the factors influencing this consistency. This systematic review and meta-analysis involved a comprehensive search of mNGS and traditional pathogen tests in PubMed, Embase, Scopus, Web of Science, and the Cochrane Library. Data from included studies were extracted, and kappa consistency between mNGS and traditional tests was calculated. Study quality was evaluated using the QUADAS-2 tool. The search identified 415 studies, of which 27 were included in the analysis, involving 4112 individuals. Meta-analysis showed a pooled consistency of 0.319 ± 0.013 (p < 0.001), indicating a moderate relationship. In terms of sample type, cerebrospinal fluid showed the highest pooled kappa consistency at 0.500 ± 0.029 (p < 0.001). Immunocompromised patients had a lower pooled kappa consistency of 0.294 ± 0.014 (p < 0.001) compared to 0.321 ± 0.028 (p < 0.001) in immunocompetent patients. Positive percent agreement of mNGS was 83.63% over traditional microbiological test, and negative percent agreement was 54.59%. This review demonstrates a moderate relationship between mNGS and traditional pathogen tests, indicating a complex relationship between these two methods. Sterile samples show higher consistency than non-sterile samples. Immune function deficiency may reduce the consistency between mNGS and traditional tests. Further research is needed on the use of mNGS in sepsis and septic shock.