Skin testing only with penicillin G in children with a history of penicillin allergy

Abstract Background The absence of commercially available penicilloyl-polylysine (PPL) for most of the last decade severely hampered the practice of penicillin allergy evaluation because skin testing without PPL is reported to have a poor negative predictive value (NPV). Objective To determine the safety and NPV of skin testing without PPL using only penicillin G followed by a 3-dose graded challenge to the incriminated penicillin in children with a history of penicillin allergy. Methods Patients evaluated for a history of penicillin allergy at the CHU Sainte-Justine Allergy Clinic between December 2006 and December 2009 were skin tested only with penicillin G and underwent a 3-dose graded challenge to the culprit penicillin if the skin test result was negative. Results Among 563 patients skin tested to penicillin G, 185 (33%) had a positive skin test result. These patients had a shorter interval between the initial reaction and skin testing compared with patients with a negative skin test result ( P  = .03). A total of 375 of 378 patients (99%) with a negative skin test result were challenged and 18 (4.8%) reacted, translating into a NPV of 95.2% (95% confidence interval [CI], 92.5%–97.1%). Three of 17 patients with a history of anaphylaxis and a negative skin test result reacted to challenge (NPV, 82.4%; 95% CI, 59.0–93.8%). All challenge reactions were mild and resolved promptly with treatment. Conclusion Among children with a history of penicillin allergy, skin testing only with penicillin G followed by a 3-dose graded challenge to the incriminated penicillin is safe and yields a good NPV. This approach could be useful when PPL is unavailable.