Description: Several studies conducted in developing countries shown a high mortality during the first year following the onset of antiretroviral therapy in late presenter. Some studies conducted with autopsy in Africa shown that an active TB is identified in 30 to 50% of the death patients, often not suspected before death. This highlights the weakness of the diagnosis tools available. In this context would the option of a systematic empirical TB treatment in severely immunosupressed patients (<100/mm3) significantly reduce early mortality (due to undiagnosed TB after ART initiation?
Goal: STATIS is a randomized controlled trial aiming to compare two experimental strategies to reduce the mortality and the occurrence of severe bacterial infection (incl. TB) at 6 months in severely immunodeficient adults infected with HIV (CD4 < 100/mm3): 1/ a strategy for intensive screening and repeated tuberculosis through workable tests during the day (Xpert MTB / RIF, LAM urinary, chest radiography); and, 2/ a strategy of systematic empirical anti -tuberculosis treatment initiated two weeks before the start of HAART.