Jones Jordan, Morrisette Taylor, Hamby Aaron, Hornback Krutika Mediwala, Burgoon Rachel
Department of Pharmacy Services, Medical University of South Carolina Health, Charleston, South Carolina, USA.
Department of Clinical Pharmacy & Outcomes Sciences, Medical University of South Carolina College of Pharmacy, Charleston, South Carolina, USA.
Pharmacotherapy. 2025 Feb;45(2):87-93. doi: 10.1002/phar.4646. Epub 2025 Jan 10.
Infections caused by extended-spectrum β-lactamase-producing Enterobacterales (ESBL-E) are increasing in the United States. Although many risk factor scoring tools exist, many are specific to bloodstream isolates and may not represent all patient populations. The purpose of this study was to create and validate an institution-specific scoring tool for select ESBL-E of non-urinary origin based on previously identified risk factors.
This retrospective, case-control analysis included inpatient adults at an academic medical center from July 2021 through August 2023 with a documented ESBL-E or non-ESBL-E infection of non-urinary origin. Patients with ESBL-E isolates were matched in a 1:1 ratio to non-ESBL-E isolates by organism and specimen type. Points for each risk factor were assigned by dividing their respective regression coefficient by half of the smallest regression coefficient and rounding to the nearest integer (prior ESBL-E within the past 12 months: 6 points, urinary catheter: 3 points, central venous catheter: 2 points, cirrhosis: 2 points). Sensitivities, specificities, positive predictive values (PPV), and negative predictive values (NPV) were calculated for each score, and discriminatory power was assessed via the receiver operating characteristic (ROC)-area under the curve (AUC).
Of the 1139 identified cultures, 140 patients met the criteria for inclusion into the ESBL-E case arm, thus 140 patients with non-ESBL-E cultures were matched as controls. Baseline characteristics were relatively similar between the groups. A score of 0 was associated with low risk of ESBL-E (PPV 0.31, NPV 0.36), whereas scores between 2 and 5 were considered moderate risk (PPV 0.56, NPV 0.55), and scores ≥6 were associated with high risk (PPV 0.91, NPV 0.56). The ROC curve AUC was 0.705.
The majority of ESBL-E risk factor scoring tools are specific to isolates causing bloodstream infections. This institution-specific scoring tool may be used to tailor empiric antimicrobial regimens and decrease unnecessary exposure to carbapenems in non-ESBL-E infections of non-urinary origin.
在美国，产超广谱β-内酰胺酶肠杆菌科细菌（ESBL-E）引起的感染正在增加。尽管存在许多风险因素评分工具，但许多工具特定于血流感染分离株，可能无法代表所有患者群体。本研究的目的是基于先前确定的风险因素，创建并验证一种针对非尿源性特定ESBL-E的机构特异性评分工具。
这项回顾性病例对照分析纳入了2021年7月至2023年8月在一家学术医疗中心住院的成年患者，这些患者有记录的非尿源性ESBL-E感染或非ESBL-E感染。ESBL-E分离株患者按生物体和标本类型与非ESBL-E分离株以1:1的比例匹配。通过将各自的回归系数除以最小回归系数的一半并四舍五入到最接近的整数来为每个风险因素分配分数（过去12个月内既往有ESBL-E：6分，导尿管：3分，中心静脉导管：2分，肝硬化：2分）。计算每个分数的敏感性、特异性、阳性预测值（PPV）和阴性预测值（NPV），并通过受试者工作特征（ROC）曲线下面积（AUC）评估鉴别能力。
在1139份鉴定出的培养物中，140名患者符合纳入ESBL-E病例组的标准，因此140名非ESBL-E培养物患者作为对照进行匹配。两组之间的基线特征相对相似。得分为0与ESBL-E低风险相关（PPV 0.31，NPV 0.36），而得分在2至5之间被认为是中度风险（PPV 0.56，NPV 0.55），得分≥6与高风险相关（PPV 0.91，NPV 0.56）。ROC曲线AUC为0.705。
大多数ESBL-E风险因素评分工具特定于引起血流感染的分离株。这种机构特异性评分工具可用于调整经验性抗菌治疗方案，并减少非尿源性非ESBL-E感染中碳青霉烯类药物的不必要暴露。
