<p>Targeted next-generation sequencing (tNGS) has emerged as a promising tool for pathogen detection in lower respiratory tract infections (LRTIs). Despite its potential, the diagnostic performance and clinical utility of tNGS in the intensive care unit (ICU) setting have not been rigorously defined. This study was designed to evaluate tNGS against conventional microbiological tests (CMT) in this specific patient population. A total of 404 BALF samples from 239 patients underwent parallel tNGS and CMT testing, with orthogonal validation (Sanger sequencing/qRT-PCR) for discrepant results. Using a composite microbiological standard, tNGS demonstrated superior accuracy (80.41% vs. 55.90%) and sensitivity (86.96% vs. 45.45%) but lower specificity (59.90% vs. 88.61%) compared to CMT (all <i>p</i> &lt; 0.05). For bacterial and viral detection, tNGS showed significantly higher accuracy and sensitivity, whereas fungal detection by tNGS was limited by panel design. In terms of concordance with clinical diagnosis, tNGS achieved higher agreement rates than CMT in patients with shock, invasive ventilation, or recent corticosteroid use (all <i>p</i> &lt; 0.05). Multivariable analysis identified shock as an independent predictor of tNGS–clinical concordance (OR = 3.37, <i>p</i> = 0.004). tNGS demonstrated enhanced sensitivity for bacterial and viral detection compared to CMT, albeit with reduced specificity, and fungal accuracy was panel-dependent. Shock independently predicted tNGS agreement with clinical diagnosis and discordance with CMT. From a diagnostic performance perspective, these findings support the preferential use of tNGS for pathogen identification in patients with shock. In other high-risk groups, integrating tNGS with CMT and clinical assessment remains essential to mitigate overdiagnosis.</p><p><b>Trial registration:</b> Chinese Clinical Trial Registry, ChiCTR2200060738. Registered 09 June 2022. </p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Diagnostic performance and clinical utility of targeted next-generation sequencing for diagnosing lower respiratory tract infections in intensive care unit patients

  • Bowen Ran,
  • Yijie Liu,
  • Xiaojing Wu,
  • Linna Huang,
  • Tianshu Zhai,
  • Ying Cai,
  • Yikun Song,
  • Qi Zhang,
  • Chao Zhang,
  • Jincheng Hao,
  • Longxian Jin,
  • Zhenyuan Huang,
  • Xiaoyan Lin,
  • Xuming Li,
  • Chen Wang,
  • Qingyuan Zhan

摘要

Targeted next-generation sequencing (tNGS) has emerged as a promising tool for pathogen detection in lower respiratory tract infections (LRTIs). Despite its potential, the diagnostic performance and clinical utility of tNGS in the intensive care unit (ICU) setting have not been rigorously defined. This study was designed to evaluate tNGS against conventional microbiological tests (CMT) in this specific patient population. A total of 404 BALF samples from 239 patients underwent parallel tNGS and CMT testing, with orthogonal validation (Sanger sequencing/qRT-PCR) for discrepant results. Using a composite microbiological standard, tNGS demonstrated superior accuracy (80.41% vs. 55.90%) and sensitivity (86.96% vs. 45.45%) but lower specificity (59.90% vs. 88.61%) compared to CMT (all p < 0.05). For bacterial and viral detection, tNGS showed significantly higher accuracy and sensitivity, whereas fungal detection by tNGS was limited by panel design. In terms of concordance with clinical diagnosis, tNGS achieved higher agreement rates than CMT in patients with shock, invasive ventilation, or recent corticosteroid use (all p < 0.05). Multivariable analysis identified shock as an independent predictor of tNGS–clinical concordance (OR = 3.37, p = 0.004). tNGS demonstrated enhanced sensitivity for bacterial and viral detection compared to CMT, albeit with reduced specificity, and fungal accuracy was panel-dependent. Shock independently predicted tNGS agreement with clinical diagnosis and discordance with CMT. From a diagnostic performance perspective, these findings support the preferential use of tNGS for pathogen identification in patients with shock. In other high-risk groups, integrating tNGS with CMT and clinical assessment remains essential to mitigate overdiagnosis.

Trial registration: Chinese Clinical Trial Registry, ChiCTR2200060738. Registered 09 June 2022.