Are you truly multidrug allergic?

Published online: June 22, 2019

For the past 30 years, the rate of patients reporting a drug allergy is growing alongside antibiotic use. With 15-20% of hospitalized patients being labeled as “drug allergic” in developed countries and about one third of them declaring multiple drug allergies, the consequences of ensuing drug avoidance and use of less effective alternatives are increasingly acknowledged. Where first-line antibiotics like penicillins are concerned, use of alternative antibiotics contributes to antimicrobial resistance and can have a negative impact on the healthcare quality and outcome. Furthermore, declaring multiple drug allergies can put physicians in a therapeutic dead-end. Thus, it is critical to distinguish true drug allergy, a potentially life-threatening reaction (proven by allergy tests), from intolerances that are non-specific, non-systemic and mostly non-severe reactions. Although intolerance or multiple intolerances should be taken into consideration for patient comfort, they should not be considered as drug allergies in the benefit-risk calculus. Therefore, isolated symptoms such as headache or gastrointestinal discomfort should not be listed in electronic health records as allergies.

In an original article in The Journal of Allergy and Clinical Immunology: In Practice, Landry et al selected all patients with potential multiple drug allergies in their 20-year database for review. From 1996 to 2018, 9,250 patients reporting 23,180 alleged allergies (approximately 2.4 each) were explored in their specialized allergy unit at the University Hospital of Montpellier (France). Tests were performed according to international recommendations at the time of the evaluation.

True multiple drug allergy was found in only 45 patients (0.48% of all tested patients and 2.5% of all patients with a confirmed drug allergy) accounting for 92 drug allergies, confirming the extreme rarity of this entity. Implicated drugs were mostly antibiotics (β-lactams, e.g., penicillins and cephalosporins; quinolones), non-steroidal anti-inflammatory drugs and iodinated radiocontrast media (iRCM), with both immediate and non-immediate reactions proven by skin-tests and/or drug provocation tests. β-Lactams and iRCM were the most frequent elicitors of positive skin test results. The most frequent manifestations in the 81 initial reactions were maculopapular exanthema (25.9%), anaphylaxis (20.9%), and angioedema (17.3%) or urticaria (16%). This series included 8 severe cutaneous adverse reactions (SCARs); 6 with drug reaction with eosinophilia and systemic symptoms, and 2 with acute generalized exanthematous pustulosis.  Most patients were allergic to 2 drugs, and rarely 3. But within this multiallergic group, a special phenotype is related to SCARs, in which the frequency of multiple drug allergy may reach 20-25% according to previous studies. No other patterns could be identified.

The fact that the authors only tested patients with clinical histories suggestive of allergy (and not intolerance symptoms or cross-reactivity) suggests that the proportion of multiple allergic patients is probably lower in the general hospitalized population and in the general population. Hence, developing de-labeling strategies should be a priority for every healthcare establishment and recent advances in algorithmic-based prescription tools are encouraging this attitude.   

The Journal of Allergy and Clinical Immunology: In Practice is an official journal of the AAAAI, focusing on practical information for the practicing clinician.

Full Article