The influence of a history of a previous test on the prevalence and size of reactions to tuberculin.

Ten millimeters of more of induration to the Mantoux tuberculin test indicates present or past infection with Mycobacterium tuberculosis in the overwhelming majority of cases. Because a number of patients gave a history of a previous positive reaction that was not confirmed on retesting, we carefully evaluated, as part of a large tuberculin testing survey, 1,428 persons who claimed to have had a previous positive test. Of these, 606 (42.4 per cent) had a positive reaction on retesting. When compared to subjects with no history of a previous tuberculin test, this group had only a slightly higher percentage of large reactions (greater than 24 mm). The difference was not significant. This discrepancy between history and test results cannot be assumed to be due to technique or to anergy in the current testing because of appropriate controls. It probably reflects variability in the use of antigens, technique at the previous testing, or faulty recollection of the subject. Because presumption of tuberculosis infection may lead to further diagnostic or therapeutic decisions, it is recommended that all skin test histories be documented or confirmed by retest before consideration of further examinations or therapy. In those with a history of a previous reaction, very large reactions (greater than 24 mm) were only slightly more frequent (but not to a significant degree) than those without such a history; thus, a previous reaction history should not be sufficient to preclude retesting.