Development and validation of PEN-FAST clinical prediction tool for possible penicillin allergy

1. In patients with reported history of penicillin allergy, a clinical prediction tool based on timing, character, and severity of the reaction accurately predicted results of further confirmatory allergy testing.

2. Using a score cut-off of 3 points, the PEN-FAST prediction tool had a high negative predictive probability in ruling out true penicillin allergy in both the internal and external validation cohorts.

Evidence Rating Level: 2 (Good)       

Study Rundown: Patient-reported allergies to penicillin are common but often do not reflect true allergies when further investigated with skin-prick or other confirmatory testing. History of penicillin allergy may limit therapeutic options for common bacterial infections and is associated with worse clinical outcomes. Further evaluation with antibiotic allergy testing procedures take time and is costly. The current study sought to develop and validate a clinical prediction tool that could accurately predict results of antibiotic allergy testing and may aid clinicians in predicting patients with low risk of clinically significant penicillin allergy. The study found that a PEN-FAST score, utilizing the clinical severity, need for treatment, skin manifestations, and timing of the reported reaction, was able to predict patients who were low risk of having positive antibiotic allergy test results. These results were consistent in an external validation cohort data set.

The PEN-FAST score is a simple clinical prediction tool that may help clinicians and antimicrobial stewardship programs identify patients with low-risk penicillin allergy histories that may not need confirmatory allergy testing. The strengths of the study include the large number of patients with formal allergy assessment, and the use of geographically separate populations in Australia and the United States. The limitations of the design include the exclusion of non-penicillin beta-lactam reactions, and low numbers of patients with severe cutaneous adverse reactions included in the study.

Click to read the study in JAMA Internal Medicine

Relevant Reading: Designing Predictive Models for Beta-Lactam Allergy Using the Drug Allergy and Hypersensitivity Database

In-Depth [prospective cohort]: This study used data from population cohorts from Melbourne, Sydney, and Perth, Australia, and Nashville, Tennessee. Patients were included if they were 16 years or older, reported a history of penicillin allergy, and underwent subsequent skin, intradermal, and/or oral testing. Patients were excluded if they did not undergo oral penicillin challenge after negative skin/intradermal testing. Backward stepwise logistic regression was used to derive the model. The primary outcome for the model was positive result on penicillin allergy testing.

The model, PEN-FAST, included having the reaction within the last 5 years (2 points), angioedema/anaphylaxis (2 points), severe cutaneous adverse reaction (2 points), and treatment required for the reaction (1 point). For the internal validity cohort, the area under the receiver-operating curve was 0.805. A cutoff score of 3 points lead to a negative predictive value of 96.3% (95% CI, 94.1%-97.8%).

Image: PD

©2020 2 Minute Medicine, Inc. All rights reserved. No works may be reproduced without expressed written consent from 2 Minute Medicine, Inc. Inquire about licensing here. No article should be construed as medical advice and is not intended as such by the authors or by 2 Minute Medicine, Inc.