Résumé
BACKGROUND: The management of iodinated contrast medium (ICM) hypersensitivity has been a matter of debate. Skin testing to identify a subgroup of ICM allergic patients has been proposed, in addition to complete avoidance, provocation testing, or premedication. OBJECTIVE: The objective of this study was to assess the negative predictive value (NPV) of skin testing for ICM. METHODS: Patients with a hypersensitivity reaction to ICM who underwent skin testing during a 13.5-year period at a single center were evaluated for re-exposure to a negatively skin-tested ICM. Premedication, consisting of second-generation H1-antihistamines twice a day 48 hours before the examination, was advised only for patients with mast cell disorder or chronic urticaria who had negative skin tests. RESULTS: A total of 597 patients tested for 423 (70.9%) immediate, 118 (19.7%) nonimmediate, and 56 (9.4%) hypersensitivity reactions with undetermined chronology were included. Eighty (13.4%) patients were skin test positive. Re-exposure to ICM occurred in 233 (39.0%) patients and was tolerated in 16 of 17 (94.1%) with at least 1 positive skin test and 201 of 216 (93.1%) with all negative skin tests. Reaction intensity was similar in 4, milder in 10, unknown in 1, and worse in 1 patient although this reaction was deemed to be nonallergic in hindsight. Premedication was administered in 20.7% of patients and associated with more reactions (19.4% vs 5.7%, P~= .01). The overall NPV of skin testing for ICM was 93.1% (95% confidence interval [CI] 89.1% to 96.0%), and for immediate and nonimmediate hypersensitivity reactions 94.2% (95% CI 89.6% to 97.2%) and 86.1% (95% CI 72.1% to 94.7%), respectively. We cannot exclude some challenges occurred with a different than the initial culprit ICM, possibly overestimating the NPV. CONCLUSIONS: Skin testing for potential ICM hypersensitivity can identify safe alternative(s) for ICM re-exposure especially in patients with an immediate hypersensitivity reaction and/or skin test-proven ICM drug allergy. Reactions on re-exposure were infrequent, mostly milder, and occurred in some patients despite premedication.