Tuberculosis (TB) is a common and serious complication of human immunodeficiency virus-type 1 (HIV) infection in the developing world, especially in sub-Saharan Africa. Since the emergence of the HIV epidemic in Africa, the incidence rates of TB have risen dramatically, overwhelming national TB control programs across Africa. Over one-half of TB patients presenting to TB clinics in Africa are HIV-infected, often presenting in early stages of HIV infection. Recent WHO guidelines on the management of HIV-associated pulmonary TB recommend antiretroviral (ARV) therapy in patients with CD4+ T cells < 200 cells/muL but not for HIV-infected TB patients who present with high CD4+ cells. In Uganda, over half of HIV-infected patients with active TB present with CD4+ counts above 200 cells/?L. There are clinical and scientific reasons for treating these TB patients with ARV therapy even when they present in early stages of HIV infection. First, mortality in HIV-associated TB is high, even when patients respond to effective anti-tuberculous therapy. Second, excess mortality associated with TB is most evident when CD4+ counts are above 200 cell/muL. Third, in dually-infected patients, TB results in prolonged immune activation which may enhance viral replication and accelerate the decline of CD4+ cells. The proposed study will test whether a novel approach of punctuated ARV therapy given during treatment of active TB will slow progression of HIV disease in TB patients with CD4+ cells > 350 cells/muL. The study will also assess the possible risks (e.g., drug toxicities and resistance) and benefits (e.g., more rapid clearance of MTB and reduced TB relapse) of punctuated ARV therapy.