Performance of PEN‐FAST and CEPH‐FAST for Cephalosporin Allergy Delabeling

Background Unverified antibiotic allergy labels lead to suboptimal antimicrobial prescribing and increased antibiotic resistance. Cephalosporins are among the most frequently administered antibiotics and cephalosporin allergy labels increasingly limit optimal therapy. Clinical decision rules that enable delabeling of low‐risk patients by non‐allergists are emerging as important tools. The PEN‐FAST score is a validated rule for identifying low‐risk penicillin allergy labels suitable for direct drug challenge. CEPH‐FAST is a recently proposed adaptation for cephalosporin allergy. We evaluated the diagnostic performance of PEN‐FAST and CEPH‐FAST for cephalosporin allergy. Methods In this single‐center retrospective cohort study, 100 adults with reported cephalosporin allergy labels underwent allergy assessment including skin testing and drug provocation testing at a tertiary allergy clinic in Heidelberg, Germany. Logistic regression was used to identify predictors of confirmed allergy. Diagnostic performance of PEN‐FAST and CEPH‐FAST scores was assessed using sensitivity, specificity, predictive values, and area under the receiver operating curve (AU‐ROC). Results Cephalosporin allergy was confirmed in 48 of 100 patients. PEN‐FAST categorized 20 patients at low risk, of whom one had a positive skin test (NPV 95.0%, sensitivity 97.9%, AU‐ROC 0.769). CEPH‐FAST classified 42 patients low‐risk, but 12 had confirmed allergies (NPV 71.4%, sensitivity 75.0%, AU‐ROC 0.667). Higher PEN‐FAST and CEPH‐FAST scores were significantly associated with confirmed allergy. Conclusion PEN‐FAST demonstrated high sensitivity and negative predictive value for identifying low‐risk cephalosporin allergy and may support safe delabeling strategies. In contrast, CEPH‐FAST showed reduced diagnostic performance in this cohort, highlighting the need for further validation before routine clinical implementation.