Evaluating penicillin allergy clinical decision-making tools to enhance penicillin allergy delabeling

Patient-reported penicillin allergy is associated with adverse outcomes.

We aimed to evaluate clinical decision-making tools used to risk-stratify reported penicillin allergies for delabeling.

Retrospective chart review of patients with reported penicillin allergies who underwent testing from October 9, 2020, to October 8, 2022 was completed at a tertiary referral health care system. Follow-up surveys were conducted to assess penicillin use after testing. The PEN-FAST clinical decision rule, 1-1-1 rule, Siew score, and Vanderbilt criteria were evaluated.

Of the 137 patients who underwent penicillin allergy evaluations, 58 completed follow-up surveys. In all, 48 patients (82.8%) had skin testing, and 46 patients had penicillin challenge following negative skin testing. Of the 10 patients (17.2%) who underwent direct challenge, none had reactions. A total of 4 patients (6.9%) tested positive; 2 had positive skin test results and 2 had negative skin test results but failed oral challenges. In predicting a positive penicillin test, the negative predictive values of the PEN-FAST (score ≤ 2), 1-1-1 rule (score ≤ 2), Siew score (score = 0), and Vanderbilt criteria with only low-risk features were 100% (95% CI = 92.0%-100.0%), 91.3% (95% CI = 72.0%-98.9%), 100.00% (95% CI = 79.4%-100.0%), and 96.6% (95% CI = (82.2%-99.9%), respectively. Of the 54 patients (93.1%) who had negative test results, 21 (38.9%) received subsequent courses of penicillin (to which 18 had no reaction and 3 had delayed rashes); none developed anaphylaxis.

Direct challenge was underutilized in penicillin allergy evaluations even though most patients tested had low-risk histories. The PEN-FAST, 1-1-1 rule, Siew score, and Vanderbilt criteria were all associated with high negative predictive values in our cohort, thus effectively identifying low-risk patients. We advocate the use of these tools to enhance penicillin allergy delabeling.