Treatment for tuberculosis-associated immune reconstitution inflammatory syndrome in 34 HIV-infected patients

Paradoxical tuberculosis immune reconstitution inflammatory syndrome (TB-IRIS) frequently follows the initiation of antiretroviral therapy (ART) in patients with tuberculosis (TB) and human immunodeficiency virus (HIV) co-infection. Treatment recommendations are nearly exclusively based on expert opinion.

To assess the clinical outcomes of patients treated using various strategies for TB-IRIS.

In a retrospective analysis of patients treated in Paris hospitals from 1996 to 2008, we describe TB-IRIS outcome, frequency of relapses and CD4 cell count changes after 12 months of ART for the following strategies: no treatment, interrupted ART and use of steroids.

Among 34 patients, TB-IRIS outcome was favourable in 10/10 with no treatment, 11/13 with ART interruption, 3/3 with ART interruption and simultaneous use of steroids and 8/8 with steroids alone. Relapses were observed in both the ART interruption (6/13, 46%) and steroids (4/8, 50%) groups, but were less frequent in the no-treatment group (1/10, 10%). Steroids were prescribed in 61% of the patients and had no significant side effects; steroid use was associated with a trend towards a lower median CD4 cell count at 12 months of ART compared to the others (230 vs. 322 cells/mm(3)), despite no baseline differences.

TB-IRIS outcome was favourable regardless of the therapeutic strategies employed. Although steroids were widely used and well-tolerated, an initial wait-and-see attitude in the case of non-severe IRIS remains an interesting strategy to be evaluated.