Diagnostic Insights Into Pathogen Spectrum and Mixed Microbial Detection in Critically Ill Patients With Pulmonary Infection Using Targeted Next-Generation Sequencing.
Pulmonary infections are common and potentially life-threatening in critically ill patients. Conventional microbiological tests (CMTs) often show limited sensitivity, particularly in the setting of prior antimicrobial exposure. Targeted next-generation sequencing (tNGS) has emerged as an alternative approach for broad pathogen detection; however, data describing pathogen spectrum, mixed infection patterns, and diagnostic performance in critically ill populations remain limited.
This retrospective study included 217 critically ill patients with suspected pulmonary infection who underwent tNGS testing of respiratory specimens. Baseline clinical characteristics were summarized. The pathogen spectrum and mixed detection patterns identified by tNGS were analyzed, including mixed microbial codetection. Diagnostic performance of tNGS was compared with that of CMTs using clinical diagnosis as the reference standard. Sensitivity, specificity, accuracy, agreement indices, and McNemar's test were applied.
tNGS detected at least one microorganism in 208 of 217 patients, yielding a significantly higher overall detection rate than CMTs. A broad spectrum of bacterial, viral, and fungal pathogens was identified, and mixed detections were common, frequently involving organisms from different pathogen categories. Co-occurrence network analysis highlighted recurrent patterns of mixed microbial detection in this cohort. Compared with CMTs, tNGS demonstrated substantially higher sensitivity and overall diagnostic accuracy, whereas conventional methods showed higher specificity.
tNGS provides a comprehensive overview of the pathogen spectrum and mixed detection patterns in critically ill patients with suspected pulmonary infection. Compared with CMTs, tNGS offers a markedly higher detection rate and sensitivity. However, careful clinical interpretation remains essential, particularly in the context of frequent mixed microbial detections, to ensure appropriate integration of sequencing results into clinical decision-making.
This retrospective study included 217 critically ill patients with suspected pulmonary infection who underwent tNGS testing of respiratory specimens. Baseline clinical characteristics were summarized. The pathogen spectrum and mixed detection patterns identified by tNGS were analyzed, including mixed microbial codetection. Diagnostic performance of tNGS was compared with that of CMTs using clinical diagnosis as the reference standard. Sensitivity, specificity, accuracy, agreement indices, and McNemar's test were applied.
tNGS detected at least one microorganism in 208 of 217 patients, yielding a significantly higher overall detection rate than CMTs. A broad spectrum of bacterial, viral, and fungal pathogens was identified, and mixed detections were common, frequently involving organisms from different pathogen categories. Co-occurrence network analysis highlighted recurrent patterns of mixed microbial detection in this cohort. Compared with CMTs, tNGS demonstrated substantially higher sensitivity and overall diagnostic accuracy, whereas conventional methods showed higher specificity.
tNGS provides a comprehensive overview of the pathogen spectrum and mixed detection patterns in critically ill patients with suspected pulmonary infection. Compared with CMTs, tNGS offers a markedly higher detection rate and sensitivity. However, careful clinical interpretation remains essential, particularly in the context of frequent mixed microbial detections, to ensure appropriate integration of sequencing results into clinical decision-making.
Authors
Zhang Zhang, Yang Yang, Zhao Zhao, Zhao Zhao, Zhang Zhang, Jiang Jiang, Liu Liu, Wu Wu, Du Du, Ma Ma, Chen Chen
View on Pubmed