Customize your JAMA Network experience by selecting one or more topics from the list below.
Copaescu AM, Vogrin S, Shand G, Ben-Shoshan M, Trubiano JA. Validation of the PEN-FAST Score in a Pediatric Population. JAMA Netw Open. 2022;5(9):e2233703. doi:10.1001/jamanetworkopen.2022.33703
The pediatric prevalence of self-reported drug allergies is 10%,1 which carries significant health and economic implications.2 Following direct oral penicillin challenge, 94.6% of such labels are removed.3,4 However, despite published algorithms,5 there are no validated pediatric decision rules to guide clinician management. The aim of this cohort study was to examine the previously validated PEN-FAST adult score6 in children.
Using a Canadian prospective pediatric cohort from 3 centers,3 we examined the PEN-FAST score in 2028 children with 2031 penicillin allergy labels (eMethods in the Supplement). Data were collected from August 8, 2011, to March 3, 2021. This cohort study was approved by the McGill University Ethics Committee and the Research Ethics Board at the University of Manitoba, and written informed consent was collected. Sample characteristics are presented as median (IQR) and frequency (%). The PEN-FAST score and area under the curve (AUC) were calculated. Logistic regression with components of the score was performed. Sensitivity analysis with different time categories and removal of severe cutaneous adverse reaction (SCAR) was performed, and subgroup analysis for immediate and delayed reactions and various age groups were performed. All analyses were performed in Stata version 16.1 (StataCorp).
The median (IQR) age for the 2028 children in the cohort was 4.3 (2.1-8.0) years, with mostly male participants (1091 [53.7%]). Most reported reactions occurred in the past 5 years or at an unknown time (1661 [81.8%]), with amoxicillin suspected in 2022 reactions (99.6%) (Table 1). Anaphylaxis and angioedema were reported in 229 cases (11.3%). Treatment (or unknown) was administered for 1231 cases (60.6%). The AUC for the PEN-FAST score was calculated at 0.528, showing poor discrimination ability. Using the published adult PEN-FAST cutoff of 3 or greater, the AUC was 0.510 (95% CI, 0.47-0.56), and sensitivity and specificity were 57.0% (95% CI, 47.1%-66.5%) and 45.7% (95% CI, 43.5%-48.0%), respectively. The negative predictive value was 95.0% (95% CI, 93.4%-96.3%), considered poor in the context of a low prevalence positive challenge (5%). Furthermore, none of the individual variables were associated with a positive test.
Changing the coding for timing (<1 year) or removing the angioedema reported symptom did not improve the performance of the PEN-FAST tool in this pediatric population (Table 2). A subgroup analysis for the positive skin testing or challenges based on immediate vs delayed reaction or the time of the reported allergy showed similar results (Table 2). When the tool was used in children 13 years or older, the AUC was 0.622, indicating that despite variable adjustment, the tool is not useful (Table 2).
In this Canadian pediatric prospective multicenter cohort, the PEN-FAST tool did not help identify low-risk penicillin allergies. This previously validated tool in an adult population was not useful for risk stratification in children younger than 12 years. In teenagers (≥13 years), the predictive ability of the tool increased (higher AUC, specificity, and NPV but lower sensitivity), indicating that the tool could have some value in this population following further study. The extrapolation of the results is limited by the small number of teenagers included.
The PEN-FAST performed similarly for immediate and delayed reactions, and the timing of the reported reaction was not associated with the outcome of the allergy investigation. In this context, it is possible that the low validity of the PEN-FAST tool in children is explained by the increased prevalence of viral-induced reactions compared with true drug hypersensitivity. This analysis adds to the evidence that true drug allergies are rare among children and that they are often incorrectly labeled during a viral infection. Furthermore, the criteria included in the PEN-FAST score might not provide adequate information considering different index reactions in the pediatric population compared with the adult population. This study highlights that children are not little adults and clinical decision rules need to be derived and validated in the target population.
These findings suggest that the PEN-FAST drug allergy clinical decision rule should not be adapted to a pediatric population younger than 12 years at this time. New validated point-of-care clinical tools are required to identify low-risk penicillin allergies in a pediatric population, and validation of PEN-FAST in adolescents requires further examination in extended international cohorts.
Accepted for Publication: August 11, 2022.
Published: September 19, 2022. doi:10.1001/jamanetworkopen.2022.33703
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2022 Copaescu AM et al. JAMA Network Open.
Corresponding Author: Ana Maria Copaescu, MD, Department of Medicine, Division of Allergy and Clinical Immunology, McGill University Health Centre (MUHC), McGill University, 1650 Cedar Ave, Montreal, QC H3G 1A4, Canada (email@example.com).
Author Contributions: Drs Copaescu and Vogrin had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Ben-Shoshan and Trubiano are co–senior authors.
Concept and design: Copaescu, Ben-Shoshan, Trubiano.
Acquisition, analysis, or interpretation of data: Vogrin, Shand, Ben-Shoshan.
Drafting of the manuscript: Copaescu, Shand, Ben-Shoshan, Trubiano.
Critical revision of the manuscript for important intellectual content: Copaescu, Vogrin, Ben-Shoshan.
Statistical analysis: Vogrin, Ben-Shoshan.
Administrative, technical, or material support: Shand, Trubiano.
Supervision: Ben-Shoshan, Trubiano.
Conflict of Interest Disclosures: Dr Copaescu reported receiving grants from the University of Melbourne, the Anna Maria Solinas Laroche Career Award in Immunology, and Anita Garbarino Girard/Anna Maria Solinas/Dr Phil Gold Award of Distinction outside the submitted work. Dr Trubiano reported receiving grants from the Austin Medical Research Foundation and National Health and Medical Research Council. No other disclosures were reported.