Penicillin hypersensitivity: value of clinical history and skin testing in daily practice.

Penicillin often is excluded as a treatment option based on patients' self-reported history of an adverse reaction to penicillin. The objective of this prospective study was to determine the likelihood of true penicillin allergy in patients with vague and convincing histories of penicillin allergy and to evaluate the diagnostic value added by appropriate skin testing. Six hundred thirty-eight patients with prior beta-lactam intake had a current indication for penicillin therapy and were referred for testing with the major (benzylpenicilloyl polylysine) and minor (minor determinant mixture) penicillin determinants from the inpatient and outpatient service of Athens University Dermatological hospital from January 2000 to December 2002. The prevalence of positive skin tests in the total group and in those patients with vague and convincing histories of penicillin allergy was determined. Positive skin tests were observed in 19/638 (3%) of the total group, 5 out of 542 (0.9%) patients without any history of penicillin allergy, 14 out of 96 (14.6%) patients with vague history (confidence interval [CI] 95% = 5.95-59.92), and 13 out of 18 (72.2%) patients with a convincing history of type I hypersensitivity reaction (chi2 = 286.3: odds ratio = 281.3: CI 95% = 62.19-1440.8). Patients with a vague history of penicillin allergy are 18 times more likely to have a positive penicillin skin test, and a convincing reaction history increases the likelihood by 281-fold compared with patients without a history of penicillin allergy. However, the fact that 5 of 18 (27.8%) patients with a convincing history were negative when skin tested points out that skin testing is helpful if the need for penicillin administration is compelling.