Towards an optimized approach for de-labeling penicillin allergy: comparison between PEN-FAST and PEN-FAST+ in an Italian cohort.

BACKGROUND

According to allergy work-up, fewer than 10% of adult and children who report a suspected penicillin allergy are actually allergic. The PEN-FAST score is currently validated for identifying patients at low-risk of penicillin allergy who benefit from a direct challenge test. The PEN-FAST+ has been developed with the aim of improving accuracy.

OBJECTIVES

The aim of this study was to compare the performance of the PEN-FAST score and PEN-FAST+ score, with the aim of identifying the best clinical decision score to apply in daily practice to simplify and implement the de-labeling processes of penicillins allergy.

METHODS

Phase I: Retrospective evaluation of the PEN-FAST score and PEN-FAST+ score in patients with proven penicillin allergy. Phase II: Comparison of diagnostic performances of PEN-FAST and PEN-FAST+ scores.

RESULTS

288 patients were studied for suspected beta-lactams allergy. 245 had a reported history of penicillin allergy. A total of 32/245 (13%) patients were confirmed to be allergic to penicillin according to EAACI/ENDA definition. The PEN-FAST score showed limited discrimination capacity (AUC = 0.639 and NPV 93.75%). For immediate allergic penicillin, PEN-FAST misclassified 31% of patients (5/16) while PEN-FAST+ nobody. Both PEN-FAST and PEN-FAST+, misclassified 30% of patients (3/10) with delayed hypersensitivity. For unknown timing hypersensitivity reactions, PEN-FAST misclassified 33%% of patients (2/6) while PEN-FAST+ 67% (4/6). The AUC (0.810 vs. 0.639, p < 0.001) and NVP (96.59% vs. 93.75%), was significantly higher for PEN-FAST+ than PEN-FAST.

CONCLUSION

Integrating PEN-FAST+ into diagnostic pathways could streamline patient selection for provocation tests, reduce false negatives, and optimize hospital resource use.