[Tuberculosis outbreak: significance of exposure time versus proximity to infection source].
BACKGROUND: The polymorphic length restriction fragment technique (PLRF) complements epidemiologic investigations. The aim of this study was to present an outbreak of tuberculosis in which the risk factors of infection and disease were studied. PATIENTS AND METHODS: A descriptive study was carried out in cases of tuberculosis. The isolated strains of Mycobacterium tuberculosis were typed by the PLRF technique. A study of the prevalence of infection was carried out among the 61 classmates of two classrooms (A and B) which the index case attended. An historical cohort study was thereafter performed among the cases with infection. The association of the dependent variable (infection or tuberculosis disease) with the remaining variables was determined by the odds ratio (OR). RESULTS: The incidence of disease was 9.8% (6/61). The strains isolated in 5 patients presented the same PLRF pattern. The prevalence study detected 28 infections (45.9%). Five cases (17.9%) were detected on the second tuberculin test. By multivariate analysis showed that the hours of exposure (1.8-3.2 hours, OR = 2.0; > 3.2 hours; OR = 10.2) were the risk factor for infection. The BCG vaccine, the intensity of the reaction to the tuberculin test and age could be associated with the risk of disease. CONCLUSIONS: The focus of the outbreak was confirmed by the PLRF technique. The importance of repeating the tuberculin test in whom the test was negative on the first test is of note. To evaluate the risk of infection the time of exposure is more important than the proximity to the index case.