Göcebe Deniz, Khatamzas Elham, Enk Alexander, Schäkel Knut
Department of Dermatology, University Hospital Heidelberg, Heidelberg, Germany.
Department of Infectious Diseases and Tropical Medicine, Center for Infectious Diseases, University Hospital Heidelberg, Heidelberg, Germany; German Center for Infection Research (Deutsches Zentrum für Infektionsforschung, DZIF), Partner site Heidelberg University Hospital, Heidelberg, Germany.
Ann Allergy Asthma Immunol. 2025 Jun 21. doi: 10.1016/j.anai.2025.06.019.
False antibiotic allergy labels lead to the unnecessary use of broad-spectrum agents, contributing to antimicrobial resistance and poorer clinical outcomes. Penicillin allergy delabeling initiatives have been supported by validated clinical decision rules such as PEN-FAST. However, there are currently no available standardized approaches for non-beta-lactam allergies, for which standardized and validated skin tests are also lacking.
First, to validate the safety and negative predictive value (NPV) of PEN-FAST for penicillin allergies. Second, to evaluate whether the performance of PEN-FAST could be replicated for non-beta-lactam antibiotic allergy labels.
This retrospective single-center comparative cohort study analyzed adult patients with penicillin or non-beta-lactam allergy labels referred to our allergy division at the Department of Dermatology, Heidelberg, Germany. Allergy assessment included skin tests and subsequent drug challenge. The performance of the previously published PEN-FAST criteria was evaluated for both cohorts.
A total of 102 penicillin allergy and 191 non-beta-lactam labels were analyzed. Confirmed allergies were more prevalent among non-beta-lactams (50.3% vs 37.3%). PEN-FAST exhibited high sensitivity for both penicillin (97.4%) and non-beta-lactams (94.8%), with an NPV of 95.8% and 82.8%, respectively. Misclassifications were limited to mild cutaneous reactions. Excluding clindamycin labels increased the NPV for non-beta-lactams to 95%.
Despite lower NPV for non-beta-lactam labels and varying performance across antibiotic classes, our results indicate that PEN-FAST may serve as a safe risk stratification tool for non-beta-lactam allergy labels. Prospective studies with direct drug challenges are required to assess its utility in the delabeling of non-beta-lactam allergies.
错误的抗生素过敏标签会导致不必要地使用广谱抗菌药物，从而导致抗菌药物耐药性增加和临床预后变差。青霉素过敏标签去除计划得到了如PEN-FAST等经过验证的临床决策规则的支持。然而，目前对于非β-内酰胺类过敏尚无可用的标准化方法，也缺乏标准化且经过验证的皮肤试验。
第一，验证PEN-FAST对于青霉素过敏的安全性和阴性预测值（NPV）。第二，评估PEN-FAST的性能是否可用于非β-内酰胺类抗生素过敏标签。
这项回顾性单中心比较队列研究分析了转诊至德国海德堡大学皮肤科过敏科的有青霉素或非β-内酰胺类过敏标签的成年患者。过敏评估包括皮肤试验及随后的药物激发试验。对两个队列均评估了先前发表的PEN-FAST标准的性能。
共分析了102个青霉素过敏标签和191个非β-内酰胺类过敏标签。确诊的过敏在非β-内酰胺类中更为普遍（50.3%对37.3%）。PEN-FAST对青霉素（97.4%）和非β-内酰胺类（94.8%）均表现出高敏感性，NPV分别为95.8%和82.8%。错误分类仅限于轻度皮肤反应。排除克林霉素标签后，非β-内酰胺类的NPV增至95%。
尽管非β-内酰胺类标签的NPV较低且不同抗生素类别之间性能存在差异，但我们的结果表明PEN-FAST可作为非β-内酰胺类过敏标签的一种安全的风险分层工具。需要进行直接药物激发试验的前瞻性研究来评估其在去除非β-内酰胺类过敏标签中的效用。
