Outcome of HIV-associated tuberculosis in the early- and late-HAART era

Abstract number: 1134_01_269

Palmieri F., Goletti D., DeMori P., Busi Rizzi E., Maddaluno R., DeMarco M., Licordari R., Rianda A., Girardi E.

Objectives:  

To assess outcome of pulmonary tuberculosis (TB) diagnosed in HIV-infected patients in the early and late time period of the widespread introduction in clinical practice of highly active antiretroviral therapy (HAART) in Italy (during 1996).

Methods:  

We reviewed clinical charts of 121 HIV-infected patients with culture-confirmed pulmonary TB hospitalized at National Institute for Infectious Diseases ‘‘L. Spallanzani’’, Rome from January 1996 to December 2002. 42 patients diagnosed in 1996–1998 (early-HAART patients) were compared to 79 patients diagnosed in 1999–2002 (late-HAART patients). Differences in categorical variables were analysed with the use of the 2 test or Fisher's exact test, as appropriate. The Kaplan–Meier method was used to estimate survival. In univariate analysis, survival differences between the two groups of patients were assessed using the log-rank test.

Results:  

At onset, patients diagnosed in 1999–2002 were more likely to be foreign born (39% vs 21%, p < 0.05), and less likely to be injection drug user (54% vs 74%, p < 0.05). Late-HAART patients had a higher CD4+ median count (155 vs 113/mm³). Moreover, 1999–2002 patients had more frequently started an HAART regimen at least 3 months before TB diagnosis (43% vs 21%, p < 0.05), and were less frequently on prior Nucleoside Reverse Transcriptase Inhibitors (NRTI) therapy before starting an HAART regimen (6% vs 48%, p < 0.001). By the end of the study period (December 31, 2003) 12 patients had died (15%) in 1999–02 period and 17 patients (40%) in 1996–98 period (p < 0.005). The survival proportion at 6, 12 and 24 months was 95%, 93% and 89% in patients diagnosed in 1999–2002, while it was 83%, 76% and 66% respectively in 1996–1998 patients (p = 0.016, log-rank). Compared to patients who started HAART when antiretroviral naive risk of death was significantly increased in patients who started HAART after NRTI therapy or did not take HAART at all (p < 0.001 for both comparison).

Conclusions:  

We found that patients diagnosed in the late-HAART era had a significantly longer survival. Our data suggest that the increasing use of HAART has continued to modify the survival of patients with HIV-associated TB, probably because of the increased proportion of patients starting HAART when antiretroviral naive in the late-HAART era.