The observation of negative skin and challenge tests to beta-lactams in some patients with prior histories of reactions to these antibiotics led us to develop a protocol ("booster study") which systematically included the performance of skin and challenge tests to beta-lactams ten to thirty days after the study to confirm the negative results. From a total of 430 patients who came to our outpatient clinic because of a reaction to a beta-lactam or unknown antibiotic, 249 completed the study. Out of the patients who completed the second phase, or "booster study," the results were positive in the first phase, or conventional study in 42 patients, and negative in 207. The booster study was negative in 197 patients (95 percent) and positive in ten patients (5 percent). Skin tests were positive in 5 of them (penicilloyl-polylysine: 5; benzylpenicillin: 3; amoxicillin: 2; minor determinant mixture: 1), and 5 patients developed an allergic reaction after rechallenge (benzylpenicillin: 4; amoxicillin: 1). One patient with both negative skin tests and amoxicillin oral challenge in the booster study developed an immediate generalized urticaria with the oral intake of amoxicillin one month later at home; skin tests became positive at that moment. The negative results in the first phase of the study and the development of positive results in the second phase could be due to the existence of an immunological amnestic reaction or to sensitization after reexposure to beta-lactams in the diagnostic procedures.