Experiences with Mycobacterium leprae soluble antigens in a leprosy endemic population.

Rees and Convit antigens prepared from armadillo-derived Mycobacterium leprae were used for skin testing in two leprosy endemic villages to understand their use in the epidemiology of leprosy. In all, 2602 individuals comprising 202 patients with leprosy detected in a prevalence survey, 476 household contacts and 1924 persons residing in non-case households were tested with two antigens. There was a strong and positive correlation (r = 0.85) between reactions to the Rees and Convit antigens. The distribution of reactions was bimodal and considering reactions of 12 mm or more as 'positive', the positivity rate steeply increased with the increase in age. However, the distributions of reactions to these antigens in patients with leprosy, their household contacts and persons living in non-case households were very similar. These results indicate that Rees and Convit antigens are not useful in the identification of M. leprae infection or in the confirmation of leprosy diagnosis in a leprosy endemic population with a high prevalence of nonspecific sensitivity.