Development and validation of a cephalosporin allergy clinical decision rule

Abstract

Background: Like penicillin allergy labels, cephalosporin allergy labels go largely unverified and drive inappropriate antibiotic use. Clinical decision rules (CDR) have been validated to identify low-risk penicillin allergy labelled patients suitable for direct oral challenge (DOC); however, the generalisability to cephalosporin allergy remains uncertain.

Methods: Cephalosporin allergy tested cohorts from three hospitals in Australia were used for validation of a cephalosporin allergy CDR based on clinical variables utilised in the published penicillin allergy decision rule (PEN-FAST). Patients with a cephalosporin allergy label underwent allergy testing. North American tested cohorts were used for external validation.

Findings: From an Australian validation cohort of 228 patients and an external cohort of 167 patients, the four clinical features associated with a positive penicillin allergy from PEN-FAST showed similar associations to a positive cephalosporin test, with minor adjustments to scoring. Validation showed an AUROC of 0.921. A cut-off of less than three points for the newly directed CEPH-FAST was chosen to classify a low risk of cephalosporin allergy, for which six of 105 patients (5.7%) had positive allergy testing results.

Interpretation: Utilising the previously published and internationally validated tool PEN-FAST, we validated the same criteria with minor modifications for low-risk cephalosporin allergies. The results suggest that a CEPH-FAST score of less than three is associated with a high negative predictive value and could be used by clinicians and antimicrobial stewardship programmes to identify patients with low-risk cephalosporin allergies at the point of care, following local validation, who could proceed to DOC or use non-cross-reactive cephalosporins.

Keywords: Antimicrobial stewardship; Beta-lactam allergy; Cross-reactivity; Direct oral challenge; Drug allergy; Oral provocation.