Drug Allergy Claims in Children: From Self-reporting to Confirmed Diagnosis

Abstract

Rebelo Gomes E, Fonseca J, Araujo L, Demoly P. Clin Exp Allergy. 2008;38(1):191–198 PURPOSE OF THE STUDY. To assess the prevalence of self-reported adverse drug reactions and drug allergy in a pediatric population and confirm the diagnosis in children with suspected drug allergy. STUDY POPULATION. Patients (n = 1426) responded to an initial cross-sectional survey. A total of 60 of 67 patients with reported drug allergy were evaluated at an allergy clinic. METHODS. The first phase included a cross-sectional survey that assessed the life occurrence of adverse drug reactions and self-reported drug allergy in the outpatient clinic of a pediatric hospital. The second phase involved a diagnostic workup in children with parent-reported drug allergy, including detailed clinical history and in vitro and in vivo investigations. Specific immunoglobulin E (IgE) level determination for β-lactams, prick and intradermal skin testing for β-lactams, local anesthetics and sulfonamides, and patch tests (if a delayed reaction was reported) were performed. If all other investigations were inconclusive and a provocation test was not contraindicated, this test was performed. RESULTS. The prevalence of self-reported adverse drug reactions and drug allergy were 10.2% and 6.0%, respectively. The frequency of a medical diagnosis of drug allergy was 3.9%. The majority of the suspected allergic reactions were nonimmediate cutaneous events attributed to β-lactam antibiotics in younger children. Of 60 patients evaluated in the allergy clinic, 39 patients had a plausible clinical history, and additional investigation including a skin test, IgE-level measurement, and possible provocation tests were conducted. Drug allergy was diagnosed in 3 children on the basis of positive responses in skin (n = 1) and oral provocation (n = 2) tests. CONCLUSIONS. Although adverse drug reactions and suspected drug allergy are frequently reported in children, after a complete evaluation, only a few of these reactions can be attributed to immediate and nonimmediate drug allergy. Overall, 94% of the patients could tolerate the initially suspected drug. REVIEWER COMMENTS. This study underscores a serious problem: patients who experience or perceive a drug reaction are often classified as being truly allergic when this may not be the case. Such overdiagnosis and misdiagnosis may result in suboptimal medication choices. These results show that only 6% of the patients with initially suspected drug allergy were truly allergic. This study demonstrates the importance of a complete and detailed history, with consideration of additional testing including skin-prick tests, specific IgE-level determination, and provocation tests. It should be noted that for nonimmediate drug allergy, an oral provocation test may require prolonged treatment to observe for symptoms. Such provocation tests would not be undertaken for severe previous reactions (eg, toxic epidermal necrolysis).