We evaluated the performance of T-SPOT.TB (a commercial interferon gamma release assay) and its accuracy for the diagnosis of active tuberculosis (TB) among HIV-infected subjects with advanced immunodeficiency. In a clinical prospective study, we assessed the performance of T-SPOT.TB for the diagnosis of active TB in HIV-infected patients with CD4 cell counts below 350 cells/mm(3) who were naive to antiretroviral and anti-TB therapies. Results were available from 147 patients, of whom 38 (25.9%) had active TB. The majority (85%) of the participants were male, with a median age of 40 years and a median CD4 cell count of 77 cells/mm(3). T-SPOT.TB yielded 15 (10.2%) indeterminate results. The indeterminate results were not associated with CD4 cell counts. However, younger patients were more likely to have an indeterminate result (OR = 0.91 per unit increase in age, p = 0.007). After excluding the indeterminate results, the sensitivity of T-SPOT.TB for the diagnosis of active TB was 37.1% and the specificity was 88.7%. The sensitivity of the T-SPOT.TB was independent of the CD4 cell count. However, its specificity was higher when used for patients with CD4 cell counts <100 cells/mm(3) when compared to patients with CD4 cell counts ≥100 cells/mm(3) (97.9% vs. 79.6%, p = 0.008). T-SPOT.TB could not be used as a routine tool to screen for active TB among HIV-infected patients with advanced immunodeficiency because of its poor performance and low sensitivity. However, it may be used as an adjunctive tool to diagnose active TB in this population due to its high specificity.