Impaired CD4 T-Cell Count Response to Combined Antiretroviral Therapy in Antiretroviral-Naive HIV-Infected Patients Presenting With Tuberculosis as AIDS-Defining Condition

Background. The impact of human immunodeficiency virus (HIV)‐associated tuberculosis on CD4 T-cell count response to combined antiretroviral therapy (cART) is poorly investigated. Methods. A collaborative analysis including HIV-infected patients prospectively enrolled in 4 Italian clinical cohorts was conducted. Patients were grouped according to Centers for Disease Control and Prevention stage at the start of cART as having tuberculosis, having AIDS but not tuberculosis (nontuberculosis AIDS), and not having AIDS (AIDS free). Time to CD4 T-cell count of at least 100, 200, and 300 cells/lL above pre-cART levels and to CD4 T-cell count of .500 cells/lL were major end points. Survival analysis with time-fixed and time-dependent covariates was used. Results. A total of 6528 patients were eligible; 125 patients (2%) had tuberculosis, 1062 (16%) had nontuberculosis AIDS, and 5341 (82%) were AIDS free. Patients with tuberculosis had a significantly reduced chance of CD4 T-cell count increase compared with AIDS-free patients as well as those with nontuberculosis AIDS, regardless of the primary outcome considered for a given value of confounders measured at baseline (eg, for .200 cells/lL above baseline; relative hazard, 0.71; P 5 .02), although it was no longer significant after further adjustment for current level of viral load suppression (relative hazard, 0.80; P 5 .11). There was a trend for reduced virological response in patients treated concomitantly for tuberculosis and HIV infection compared with those who were treated separately in time (P 5 .09). Conclusions. HIV-infected patients starting cART with a tuberculosis diagnosis showed an impaired immune recovery to cART compared with AIDS-free patients and those with nontuberculosis AIDS. It seems to be driven mainly by a delay in achieving viral suppression. Whether this may be due to interactions between antituberculosis drugs and antiretrovirals needs to be investigated.