Department of Health and Society, Kristianstad University, Kristianstad, Sweden.
PreHospen - Centre for Prehospital Research, University of Borås, Borås, Sweden.
BMC Emerg Med. 2024 Oct 8;24(1):180. doi: 10.1186/s12873-024-01098-4.
Early recognition of sepsis by the EMS (Emergency Medical Services), along with communicating this concern to the emergency department, could improve patient prognosis and outcome. Knowledge is limited about the performance of sepsis identification screening tools in the EMS setting. Research is also limited on the effectiveness of prehospital use of blood tests for sepsis identification. Integrating blood analyses with screening tools could improve sepsis identification, leading to prompt interventions and improved patient outcomes.
The aim of the present study is firstly to evaluate the performance of various screening tools for sepsis identification in the EMS setting and secondly to assess the potential improvement in accuracy by incorporating blood analyses.
This is a retrospective observational cohort study. The data were collected from prehospital and hospital medical records in Region Halland. Data on demographics, vital signs, blood tests, treatment, and outcomes were collected from patients suspected by EMS personnel of having infection. The data were analysed using Student's t-test. Sensitivity, specificity, positive predictive value, negative predictive value and odds ratio were used to indicate accuracy and predictive value.
In total, 5,405 EMS missions concerning 3,225 unique patients were included. The incidence of sepsis was 9.8%. None of the eleven tools included had both high sensitivity and specificity for sepsis identification. White blood cell (WBC) count was the blood analysis with the highest sensitivity but the lowest specificity for identifying sepsis. Adding WBC, C-reactive protein (CRP) or lactate to the National Early Warning Score (NEWS) increased the specificity to > 80% but substantially lowered the sensitivity.
Identifying sepsis in EMS settings remains challenging, with existing screening tools offering limited accuracy. CRP, WBC, and lactate blood tests add minimal predictive value in distinguishing sepsis or determining non-conveyance eligibility.
通过急救医疗服务（EMS）早期识别脓毒症，并将这一担忧传达给急诊科，可能会改善患者的预后和结局。关于在 EMS 环境中使用脓毒症识别筛查工具的性能，人们的了解有限。关于在院前使用血液检测来识别脓毒症的有效性的研究也有限。将血液分析与筛查工具相结合可能会提高脓毒症的识别能力，从而及时进行干预并改善患者的结局。
本研究的目的首先是评估各种在 EMS 环境中用于脓毒症识别的筛查工具的性能，其次是评估通过整合血液分析来提高准确性的潜力。
这是一项回顾性观察性队列研究。数据来自哈兰地区的院前和医院病历。从被 EMS 人员怀疑患有感染的患者的病历中收集人口统计学、生命体征、血液检测、治疗和结局的数据。使用 Student's t 检验对数据进行分析。灵敏度、特异度、阳性预测值、阴性预测值和优势比用于表示准确性和预测价值。
共纳入了 5405 次涉及 3225 名患者的 EMS 任务。脓毒症的发生率为 9.8%。纳入的 11 种工具均不具备用于脓毒症识别的高灵敏度和特异度。白细胞（WBC）计数是血液分析中识别脓毒症的敏感性最高但特异性最低的指标。将 WBC、C 反应蛋白（CRP）或乳酸添加到国家早期预警评分（NEWS）中可以提高特异性>80%，但会显著降低灵敏度。
在 EMS 环境中识别脓毒症仍然具有挑战性，现有的筛查工具准确性有限。CRP、WBC 和乳酸血液检测在区分脓毒症或确定是否适合不转运方面几乎没有增加预测价值。
