BACKGROUND: Background: Patch tests are an efficient method to confirm the etiological diagnosis of allergic contact dermatitis. OBJECTIVES: 1) To determine the permanence of results between two tests performed with an interval of at least one year, in patients with allergic contact dermatitis; 2) To compare the positive results according to rates of intensity; 3) To evaluate the permanence of sensitization according to each substance that was tested. METHODS: Patients with previous diagnosis of allergic contact dermatitis, confirmed by patch tests carried out between the years 2005 and 2008, underwent new testing, using the same methodology, and data was compared. RESULTS: A total of 1470 results of both tests on 49 patients were analyzed. The negative results remained in the second test in a rate of 96%, and 4% became positive (+) without relevance to the clinical history. Moreover, moderately (++) and strongly (+++) positive results were also maintained in, respectively, 86% and 100%. Nevertheless, weakly (+) positive results became negative in 65%. By ignoring all weakly (+) positive tests, the calculation of Kappa Index of Agreement Statistics between the two tests showed a value of 0.88. CONCLUSION: Patch tests showed to be reliable for negative, moderately (++) positive and strongly (+++) positive results, by reproducing the same standard of individual response to allergens. However, for weakly (+) positive results, tests were not reliable.