The Diagnosis of Latent Tuberculosis in HIV-Infected Persons

Abstract

TO THE EDITOR-We read with interest the article by The Antiretroviral Therapy Cohort Collaboration [ 1] on the incidence of tuberculosis among patients receiving antiretroviral therapy. The authors note that the effectiveness of the tuberculin skin test (TST) for detecting latent tuberculosis appears to be increased among patients receiving effective antiretroviral therapy. The TST uses purified protein derivative of tuberculin, a crude mixture of >200 antigens that cross-react with Mycobacterium bovis bacille Calmette-Guerin (BCG), and many environmental mycobacteria and, therefore, has a very low specificity. The 6 kDa early secretory antigenic target of M. tuberculosis (ESAT-6) and the 10 kDa culture filtrate protein (CFP-10) are peptides that are present only on M. tuberculosis and a few rare environmental mycobacteria (Mycobacterium kansasii [2], Mycobacterium marinum [2], Mycobacterium szulgai [2], and Mycobacterium leprae [3]). These peptides have been incorporated into 2 new commercially available diagnostic tests (T-SPOT. TB [Oxford Immunotec] and QuantiFERON-TB Gold [Cellestis]) that measure the release of IFN--y following stimulation of T lymphocytes with these peptides. These tests are not only more specific than the TST but also have the advantage that a nonspecific T cell stimulant (phytohaemagglutin) may be used as a positive control. This enables true-negative results to be distinguished from those that fail as a result of anergy or for technical reasons. It has recently been shown that the performance of the T-SPOT.TB test appears to be independent of the CD4+ T lymphocyte count [4]. Thus, tests incorporating ESAT6 and CFP-10 peptides, rather than the TST, should be better for the diagnosis of latent tuberculosis, regardless of the patient's CD4+ T lymphocyte count or use of antiretroviral therapy.