Wallgren Ulrika M, Castrén Maaret, Svensson Alexandra E V, Kurland Lisa
aDepartment of Clinical Research and Education, Karolinska Institutet, Södersjukhuset bSection of Emergency Medicine, Södersjukhuset cFisksätra Vårdcentral, Saltsjöbaden, Sweden.
Eur J Emerg Med. 2014 Aug;21(4):260-5. doi: 10.1097/MEJ.0000000000000084.
Timely identification and treatment of sepsis is crucial for patient outcome. The aim of this study was to compare two previously unvalidated prehospital sepsis screening tools with clinical judgment by emergency medical services (EMS) personnel with respect to identification of septic patients.
We carried out a retrospective cross-sectional study of 353 adult patients, transported by the EMS, with a hospital discharge International Classification of Diseases code consistent with sepsis. We analyzed EMS records for the identification of sepsis according to two screening tools and clinical judgment by EMS providers. The Robson screening tool includes temperature, heart rate, respiratory rate, altered mental status, plasma glucose, and a history suggestive of a new infection. BAS 90-30-90 refers to the vital signs: oxygen saturation, respiratory rate, and systolic blood pressure. McNemar's two related samples test was used to compare the sensitivity of the two screening tools with the sensitivity of clinical judgment.
The Robson screening tool had a sensitivity of 75% (18 out of 24 patients for whom all parameters were documented, P<0.001, as compared with clinical judgment). BAS 90-30-90 had a sensitivity of 43% (76 out of 175 patients, P<0.001). EMS personnel documented suspected sepsis in 42 out of 353 (12%) patients with sepsis.
The Robson screening tool had a sensitivity superior to both BAS 90-30-90 and clinical judgment. This supports our hypothesis that the implementation of a screening tool could lead to increased prehospital identification of sepsis, which may enable a more timely treatment of these patients.
及时识别和治疗脓毒症对患者的预后至关重要。本研究的目的是比较两种先前未经验证的院前脓毒症筛查工具与急诊医疗服务（EMS）人员的临床判断在识别脓毒症患者方面的差异。
我们对353例成年患者进行了一项回顾性横断面研究，这些患者由EMS转运，出院时国际疾病分类代码与脓毒症相符。我们分析了EMS记录，以根据两种筛查工具和EMS提供者的临床判断来识别脓毒症。罗布森筛查工具包括体温、心率、呼吸频率、精神状态改变、血糖以及提示新感染的病史。BAS 90-30-90指的是生命体征：血氧饱和度、呼吸频率和收缩压。采用麦克尼马尔配对样本检验来比较两种筛查工具的敏感性与临床判断的敏感性。
罗布森筛查工具的敏感性为75%（在所有参数均有记录的24例患者中，有18例，与临床判断相比，P<0.001）。BAS 90-30-90的敏感性为43%（175例患者中有76例，P<0.001）。在353例（12%）脓毒症患者中，EMS人员记录了42例疑似脓毒症患者。
罗布森筛查工具的敏感性优于BAS 90-30-90和临床判断。这支持了我们的假设，即实施筛查工具可能会增加院前脓毒症的识别率，从而可能使这些患者得到更及时的治疗。
