Reviewed by James “Brad” Cutrell, MD, FIDSA; UT Southwestern Medical School

As antibiotic allergy delabeling has gained traction as a target for stewardship programs, there is a need for decision support tools to triage patients at low risk for true allergy, especially at centers lacking allergy expertise or a formal allergy skin testing program. This JAMA Internal Medicine study provides a simple point of care decision rule to identify “low risk” patients for penicillin (PCN) allergy delabeling without formal testing.

Trubiano et. al. conducted a prospective cohort study in 622 patients at two Australian hospitals to derive and internally validate a decision support rule to predict PCN allergy test results. Patients who self-reported a PCN allergy underwent testing via either skin prick, intradermal, or patch testing and/or oral challenge. The decision rule was externally validated in two retrospective cohorts with 945 total patients. The 4 clinical variables in the multivariable logistic regression associated with a positive PCN test and included in the final decision rule are the following (using PCN-FAST mnemonic): Five or fewer years ago (2 point), Anaphylaxis or angioedema (2 points), Severe cutaneous adverse reaction (2 points), and Treatment required for allergy episode (1 point). The final internal validation model had an AUC of 0.805. Using a cutoff value of less than 3 to define “low risk” patients, only 3.7% of these patients had a positive PCN allergy test, with a negative predictive value of 96.3% (95% CI; 94.1%-97.8%), with similar results in the external validation cohorts. These results suggest that a PCN-FAST score < 3 can safely identify “low risk” patients who can have their PCN allergy delabeled without formal allergy testing.

Reference:

Trubiano JA, et al. JAMA Intern Med. 2020; 180(5):745-752. doi:10.1001/jamainternmed.2020.0403
https://jamanetwork.com/journals/jamainternalmedicine/article-abstract/2762878