NONIMMEDIATE REACTIONS TO BETA-LACTAMS - PREVALENCE AND ROLE OF THE DIFFERENT PENICILLINS

In patients treated with penicillins, adverse cutaneous reactions can occur within minutes or may take several days to develop. IgE antibody-mediated reactions are well documented, but other mechanisms may also be involved. In particular, nonimmediate reactions have not been studied extensively, and the purpose of the present work was to establish the incidence of such reactions among a large group of patients and to study the penicillins involved. A total of 380 subjects with a history of a cutaneous reaction following administration of a penicillin antibiotic was included in the study. Skin tests and specific IgE measurements (RAST) were carried out using various penicillins and penicillin-related reagents, and patients were also challenged with various penicillins. In some patients with delayed skin test responses, skin biopsies were carried out. The tests confirmed that 74 subjects (19.4% of total investigated) had suffered a cutaneous reaction to a penicillin derivative, and 29 of these subjects (7.6% of total or 39% of confirmed) showed evidence of having suffered a nonimmediate reaction. The latter group were identified by giving a positive delayed direct challenge, and in 65% of the cases a delayed skin test response was detected. In most cases, these responses were to amino penicillins. Skin biopsies showed a lymphomonocytic cell infiltrate. Nonimmediate reactions to penicillins are a reproducible phenomenon, suggesting that a specific mechanism is responsible. By direct challenge, 93% of responders were positive to amino penicillins (10.3% ampicillin, 82.7% amoxicillin), indicating a major role for these penicillins in nonimmediate reactions. The high percentage (65%) of subjects in this group who showed delayed skin test responses, taken together with the biopsy results, suggests that a lymphocyte-mediated reaction occurred. However, the possibility that other mechanisms may also have been involved cannot be ruled out.