Objective: Macrolide allergy is rarely reported, and there is limited knowledge of hypersensitivity reactions (HRs) in children. The negative predictive value (NPV) of drug provocation tests (DPTs) for macrolides is unresolved. We aim to evaluate the clinical features of macrolide allergy in children, and determine the NPV of macrolide DPTs.

Materials and Methods: Pediatric patients who were referred to our allergy department with a suspicion of macrolide allergy were evaluated by DPTs with or without prior skin tests between 2011 and 2020. Characteristics of the HRs and patients, the results of skin and DPTs were recorded. At least three months after evaluation of the patients with allergy work up, telephone interviews were performed. Patients were asked whether they had reused the suspected macrolide or not. Patients who reported HR during subsequent drug intake were invited for reevaluation.

Results: A total of 160 children (161 reactions) (55.6% male) with a suspicion of macrolide allergy were enrolled for the study, and all children had a mild index reaction. The median age was 48 (18-102) months, and the median time between the suspected allergic reaction and allergy work-up was 3 (2-8) months. The most frequently reported suspected agent was clarithromycin, in 151 patients (94.4%). Macrolide allergy was confirmed in 8 (5%) patients. Only one patient reported skin eruptions upon reuse despite a negative DPT and he was invited to be reevaluated. A second DPT was performed resulting in urticarial lesions. The NPV was found to be 97.4% for negative DPT with macrolides.

Conclusion: Confirmed macrolide allergy is rare in children, and DPTs are the gold standard to assess suspected macrolide allergy. The NPV of macrolide provocation tests seems to be high in children.