Pajor Michael J, Munigala Satish, Reynolds Dan, Zeigler Julie, Gebru Danaye, Asaro Phillip V, Lawrence Steven J, Liang Stephen Y, Mudd Philip A
Department of Emergency Medicine Washington University School of Medicine Saint Louis Missouri USA.
Division of Infectious Diseases, Department of Medicine Washington University School of Medicine Saint Louis Missouri USA.
J Am Coll Emerg Physicians Open. 2023 Sep 20;4(5):e13045. doi: 10.1002/emp2.13045. eCollection 2023 Oct.
We sought to develop an evidence-based tool to risk stratify patients diagnosed with seasonal influenza in the emergency department (ED).
We performed a single-center retrospective cohort study of all adult patients diagnosed with influenza in a large tertiary care ED between 2008 and 2018. We evaluated demographics, triage vital signs, chest x-ray and laboratory results obtained in the ED. We used univariate and multivariate statistics to examine the composite primary outcome of death or need for intubation. We validated our findings in patients diagnosed between 2018 and 2020.
We collected data from 3128 subjects; 2196 in the derivation cohort and 932 in the validation cohort. Medical comorbidities, multifocal opacities or pleural effusion on chest radiography, older age, elevated respiratory rate, hypoxia, elevated blood urea nitrogen, blood glucose, blood lactate, and red blood cell distribution width were factors associated with intubation or death. We developed the Predicting Intubation in seasonal Influenza Patients diagnosed in the ED (PIIPED) risk-stratification tool from these factors. The PIIPED tool predicted intubation or death with an area under the receiver operating characteristic curve (AUC) of 0.899 in the derivation cohort and 0.895 in the validation cohort. A version of the tool including only factors available at ED triage, before laboratory or radiographic evaluation, exhibited AUC of 0.852 in the derivation cohort and 0.823 in the validation cohort.
Clinical findings during an ED visit predict severe outcomes in patients with seasonal influenza. The PIIPED risk stratification tool shows promise but requires prospective validation.
我们试图开发一种基于证据的工具，用于对急诊科诊断为季节性流感的患者进行风险分层。
我们对2008年至2018年期间在一家大型三级医疗急诊科诊断为流感的所有成年患者进行了单中心回顾性队列研究。我们评估了人口统计学、分诊生命体征、胸部X光和在急诊科获得的实验室结果。我们使用单变量和多变量统计方法来检查死亡或需要插管的综合主要结局。我们在2018年至2020年诊断的患者中验证了我们的发现。
我们收集了3128名受试者的数据；推导队列中有2196名，验证队列中有932名。医疗合并症、胸部X光上的多灶性opacity或胸腔积液、年龄较大、呼吸频率升高、缺氧、血尿素氮、血糖、血乳酸和红细胞分布宽度升高是与插管或死亡相关的因素。我们根据这些因素开发了急诊科诊断的季节性流感患者插管预测（PIIPED）风险分层工具。PIIPED工具在推导队列中预测插管或死亡的受试者工作特征曲线下面积（AUC）为0.899，在验证队列中为0.895。该工具的一个版本仅包括在实验室或影像学评估之前急诊科分诊时可用的因素，在推导队列中的AUC为0.852，在验证队列中的AUC为0.823。
急诊科就诊期间的临床发现可预测季节性流感患者的严重结局。PIIPED风险分层工具显示出前景，但需要前瞻性验证。
