Delayed readings of a standard screening patch test tray: Frequency of [ldquo ]lost,[rdquo ] [ldquo ]found,[rdquo ] and [ldquo ]persistent[rdquo ] reactions

Background: Several studies have shown that patch test results vary depending on the time at which the patch test sites are evaluated. Objective: This study aims to determine the frequency of [ldquo ]lost,[rdquo ] [ldquo ]found,[rdquo ] and [ldquo ]persistent[rdquo ] reactions to a standard screening tray by comparing initial and delayed readings of the same patch tests. Methods: We performed a retrospective review of patch test reactions for 587 patients tested with the standard screening tray at The Cleveland Clinic Foundation (Cleveland, OH) between January 1, 1988, and December 31, 1991, for which an initial day-3 and delayed day-4 to day-8 readings were available. Results: For the 587 patients, we identified 81 found (15%), 342 persistent (62%), and 127 lost (23%) reactions, for a total of 550 reactions (100%). The three most common found reactions were to neomycin sulfate (n = 19; 53%), p-phenylenediamine (n = 10; 29%), and thiomersal (n = 10; 18%). The three most common persistent reactions were to thiomersal (n = 41; 73%), nickel sulfate (n = 33; 73%), and quaternium 15 (n = 32; 74%). The three most common lost reactions were to balsam of Peru (n = 14; 39%), benzoyl peroxide (n = 12; 63%), and nickel sulfate (n = 9; 20%). Conclusion: Two patch test readings are necessary, followed by a final interpretation.