The diagnosis of infection in patients with suspected sepsis is frequently difficult to achieve with a reasonable degree of certainty. Currently, the diagnosis of infection still relies on a combination of systemic manifestations, manifestations of organ dysfunction, and microbiological documentation. In addition, the microbiologic confirmation of infection is obtained only after 2 to 3 days of empiric antibiotic therapy. These criteria are far from perfect being at least in part responsible for the overuse and misuse of antibiotics, in the community and in hospital, and probably the main drive for antibiotic resistance. Biomarkers have been studied and used in several clinical settings as surrogate markers of infection to improve their diagnostic accuracy as well as in the assessment of response to antibiotics and in antibiotic stewardship programs. The aim of this review is to provide a clear overview of the current evidence of usefulness of biomarkers in several clinical scenarios, namely, to diagnose infection to prescribe antibiotics, to exclude infection to withhold antibiotics, and to identify the causative pathogen to target antimicrobial treatment. In recent years, new evidence with “old” biomarkers, like C-reactive protein and procalcitonin, as well as new biomarkers and molecular tests, as breathomics or bacterial DNA identification by polymerase chain reaction, increased markedly in different areas adding useful information for clinical decision making at the bedside when adequately used. The recent evidence shows that the information given by biomarkers can support the suspicion of infection and pathogen identification but also, and not less important, can exclude its diagnosis. Although the ideal biomarker has not yet been found, there are various promising biomarkers that represent true evolutions in the diagnosis of infection in patients with suspected sepsis.

疑似脓毒症患者的感染诊断通常很难以合理的确定性实现。目前，感染的诊断仍然依赖于全身表现、器官功能障碍表现和微生物学文件的结合。此外，只有在经验性抗生素治疗 2 至 3 天后才能获得感染的微生物学确认。这些标准远非完美，至少部分原因是在社区和医院中过度使用和滥用抗生素，并且可能是抗生素耐药性的主要驱动力。生物标志物已被研究并在几种临床环境中用作感染的替代标志物，以提高其诊断准确性以及评估对抗生素的反应和抗生素管理计划。本综述的目的是清晰概述生物标志物在几种临床情况下有用的当前证据，即诊断感染以开抗生素，排除感染以停用抗生素，以及确定致病病原体以靶向抗菌治疗. 近年来，“旧”生物标志物的新证据，如 C 反应蛋白和降钙素原，以及新的生物标志物和分子测试，如呼吸组学或聚合酶链反应鉴定细菌 DNA，在不同领域显着增加，当充分使用时，为床边的临床决策增加了有用的信息。最近的证据表明，生物标志物提供的信息可以支持对感染和病原体鉴定的怀疑，但同样重要的是，它可以排除其诊断。尽管尚未找到理想的生物标志物，但有各种有希望的生物标志物代表了疑似脓毒症患者感染诊断的真正进展。