Mackowiak Amy, Brenneman Ethan, Holland Thomas, Lee Hui-Jie, Jones Justin, Keil Elizabeth, Mando Jennifer, Theophanous Rebecca, Toler Rachel, Moehring Rebekah, Wrenn Rebekah
Department of Pharmacy, Duke University Hospital, Durham, North Carolina, USA.
Division of Infectious Diseases, Duke University School of Medicine, Durham, North Carolina, USA.
J Am Coll Emerg Physicians Open. 2025 Jan 10;6(1):100010. doi: 10.1016/j.acepjo.2024.100010. eCollection 2025 Feb.
Blood cultures obtained in the emergency department (ED) may become positive after discharge. Healthcare professionals must determine if these results represent true infection or a likely contaminant. An institutional algorithm was developed to assist with healthcare professional response to positive blood cultures for and coagulase-negative staphylococci (CoNS) in these situations.
We conducted a single system, multisite cohort study comparing before and after implementation of an ED decision-making algorithm from November 2022 to December 2023. Adults were included if they were discharged from the ED before blood cultures became positive for species. The primary outcome was the difference in rates of patients called back to the ED pre- and postalgorithm implementation. Secondary endpoints evaluated algorithm adherence and safety.
A total of 253 patients, 188 pre- and 65 postimplementation, were enrolled. There was a 7.3% reduction in patients called back to the ED after algorithm implementation (95% CI [-21.1 to 6.3], = .3). Algorithm adherence after implementation was 84.6% with a difference in actual and algorithm-based callback rates of 4.6%. After algorithm implementation, no patients deemed to have a contaminant experienced an infectious-related safety event.
This time-saving algorithm was well received by our ED professionals and served as a helpful tool in safely and effectively triaging patients who had positive blood cultures for species after discharge to determine who should be called back for further evaluation. There was a nonstatistically significant but clinically meaningful reduction in callback rates. Postimplementation algorithm adherence was high, and the majority of callback decisions were appropriate.
在急诊科采集的血培养标本在患者出院后可能呈阳性。医护人员必须确定这些结果代表的是真正的感染还是可能的污染物。为此开发了一种机构算法，以协助医护人员应对这些情况下凝固酶阴性葡萄球菌（CoNS）血培养阳性的情况。
我们进行了一项单系统、多中心队列研究，比较了2022年11月至2023年12月实施急诊科决策算法前后的情况。纳入标准为在血培养对特定菌种呈阳性之前从急诊科出院的成年人。主要结局是算法实施前后被召回急诊科的患者比例差异。次要终点评估算法的依从性和安全性。
共纳入253例患者，其中实施前188例，实施后65例。算法实施后被召回急诊科的患者减少了7.3%（95%CI[-21.1至6.3]，P = 0.3）。实施后算法依从性为84.6%，实际召回率与基于算法的召回率差异为4.6%。算法实施后，没有被认为是污染物感染的患者发生与感染相关的安全事件。
这种节省时间的算法受到了我们急诊科专业人员的好评，并成为一种有用的工具，用于安全有效地对出院后血培养对特定菌种呈阳性的患者进行分类，以确定哪些患者应被召回进行进一步评估。召回率有统计学上不显著但临床上有意义的降低。实施后算法依从性较高，大多数召回决策是合适的。
