Scholarlink

Purpose <p>Patients with suspected sepsis often receive broad-spectrum antibiotics before culture results are available. A rapid point-of-care test (POCT) that indicates Gram-negative (GN) versus Gram-positive (GP) infection could help tailor empiric therapy. We systematically compared available POCT for GN/GP differentiation and, for biomarkers, examined clinically usable thresholds.</p> Methods <p>PubMed, Embase, Web of Science and the Cochrane Library (January 2005 to August 2025) were searched for studies in adults with sepsis that evaluated rapid tests to distinguish GN from GP infection. We pooled host-response biomarkers, pathogen-directed assays, omics-based tests, and clinical-parameter approaches using a bivariate random-effects model to obtain pooled sensitivity, specificity, area under the curve (AUC) and Youden’s index. For biomarkers reported at multiple cut-offs, clinically relevant strata were prespecified and compared.</p> Results <p>Of 86 included studies, 72 were eligible for quantitative synthesis. Pathogen-directed rapid assays (PCR, MALDI-TOF MS) showed the highest and most consistent accuracy (pooled sensitivity and specificity &gt; 0.90; AUC 0.97–0.99; Youden 0.85–0.92). Among the biomarker studies, procalcitonin (PCT) showed clear threshold dependence: the 3–5&#xa0;ng/mL stratum provided the most balanced discrimination (sensitivity 0.84; specificity 0.83; AUC 0.90; Youden 0.67), whereas higher cut-offs did not yield further gains. Omics-based approaches showed variable accuracy, and clinical-parameter approaches alone achieved sensitivity and specificity &lt; 0.70.</p> Conclusion <p>In adults with sepsis, pathogen-directed rapid assays are the most reliable POCT for early GN/GP differentiation. When biomarker testing is available, a PCT range of 3–5&#xa0;ng/mL is a pragmatic working threshold that can support earlier tailoring of empiric antimicrobial therapy.</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

Comparative evaluation of multimodal point-of-care tests to differentiate gram-negative from gram-positive infections in critically ill adults: a diagnostic accuracy study

  • 期刊论文

  • 发表日期：

    2026年1月14日

查看全文
  • Weijia Huang,
  • Ting Yang,
  • Sinan Ma,
  • Jiatian Wang,
  • Huangxin Gong,
  • Haitao Wang,
  • Jingying Sun,
  • Na Wang,
  • Li Zhang,
  • Yan Wang

摘要

Purpose

Patients with suspected sepsis often receive broad-spectrum antibiotics before culture results are available. A rapid point-of-care test (POCT) that indicates Gram-negative (GN) versus Gram-positive (GP) infection could help tailor empiric therapy. We systematically compared available POCT for GN/GP differentiation and, for biomarkers, examined clinically usable thresholds.

Methods

PubMed, Embase, Web of Science and the Cochrane Library (January 2005 to August 2025) were searched for studies in adults with sepsis that evaluated rapid tests to distinguish GN from GP infection. We pooled host-response biomarkers, pathogen-directed assays, omics-based tests, and clinical-parameter approaches using a bivariate random-effects model to obtain pooled sensitivity, specificity, area under the curve (AUC) and Youden’s index. For biomarkers reported at multiple cut-offs, clinically relevant strata were prespecified and compared.

Results

Of 86 included studies, 72 were eligible for quantitative synthesis. Pathogen-directed rapid assays (PCR, MALDI-TOF MS) showed the highest and most consistent accuracy (pooled sensitivity and specificity > 0.90; AUC 0.97–0.99; Youden 0.85–0.92). Among the biomarker studies, procalcitonin (PCT) showed clear threshold dependence: the 3–5 ng/mL stratum provided the most balanced discrimination (sensitivity 0.84; specificity 0.83; AUC 0.90; Youden 0.67), whereas higher cut-offs did not yield further gains. Omics-based approaches showed variable accuracy, and clinical-parameter approaches alone achieved sensitivity and specificity < 0.70.

Conclusion

In adults with sepsis, pathogen-directed rapid assays are the most reliable POCT for early GN/GP differentiation. When biomarker testing is available, a PCT range of 3–5 ng/mL is a pragmatic working threshold that can support earlier tailoring of empiric antimicrobial therapy.