In Thailand, people co-infected with HIV and tuberculosis who receive antiretroviral therapy are six times as likely to survive as those who do not, according to a study reported in the June 1st issue of the Journal of Acquired Immune Deficiency Syndromes. These results are drawn from one of the first large-scale studies to control for confounding factors which may have influenced estimates of survival benefit in this population.
In Thailand and Southeast Asia, co-infection with tuberculosis (TB) and HIV is widespread, and mortality rates from TB are extremely high (ranging up to 50%) among the co-infected. The World Health Organization currently recommends that HIV/TB co-infected people who are eligible for antiretroviral treatment (ART) should begin receiving it within two months of beginning anti-TB treatment. Several observational studies have demonstrated dramatically higher survival rates (during and after TB treatment) among ART-eligible co-infected patients who receive ART. (See, for instance, here.)
However, these have largely been single-site observational studies which may have contained biases (e.g., ART may have been avoided in patients at high risk of death); they have also not shown population-level data on the frequency and efficacy of ART use in HIV/TB co-infected patients.
The Thailand TB Active Surveillance Network is a project (begun in 2003) to enhance surveillance, monitoring, evaluation, and treatment of TB in Thailand. This study is a large, multi-site analysis from that project. The analysis measured the impact of ART on survival of HIV-infected TB patients in Thailand, while adjusting for factors associated with receiving ART.
Analysis was performed on data from HIV-positive patients registered for TB treatment across four provinces in Thailand between October 2004 and March 2006. Of the 3105 such patients, 1836 were excluded from analysis for the following reasons: 873 were undergoing TB retreatment (which has significantly different outcomes from initial treatment), 96 had a change in their initial TB diagnosis, 285 were already on ART, 151 transferred, and 431 did not have complete data. This left a total of 1269 patients in the analysis, with no previous history of TB treatment, and TB treatment outcomes of: cured, treatment completed, treatment failed, or died.
The median age of these 1269 patients was 35 years; 34% were female. Pulmonary TB was diagnosed in 54%, extrapulmonary in 35%, and both in 12%. Multidrug-resistant TB was diagnosed in 1%. Most patients had advanced HIV disease: of the 1012 for whom CD4 cell counts were available, the median was 54 cells/mm3 (range, 1 – 1169).
Who received ART?
ART was initiated after TB diagnosis in 626 (49%) of the 1269 patients. In multivariate analysis, ART use was more common in patients who received fluconazole after TB diagnosis (adjusted relative risk [aRR], 1.3; 95% confidence interval [CI], 1.1–1.4), in patients treated in an urban area (aRR, 1.2; 95% CI, 1.1–1.3), and in those with CD4 cell counts < 200 cells/mm3 (p<.05; aRR varied from 1.8 to 2.3 depending on CD4 cell count subrange).
ART use was less common in patients diagnosed with TB in a private clinic or hospital (aRR, 0.7; 95% CI, 0.5¬–1.0).
How did ART affect outcomes?
Of the 626 co-infected patients started on ART during TB treatment, in bivariate analysis, 68 (11%) died compared with 295 of the 643 (46%) not prescribed ART (relative risk [RR], 0.24, 95% CI, 0.19–0.30). Even in those with very low CD4 cell counts (<10 cells/mm3), 12 of 56 (21%) patients receiving ART died compared with 35 of 43 (81%) not receiving ART (RR, 0.26; 95% CI, 0.16–0.44).
In multivariate analysis, controlling for factors associated with death, ART use during TB treatment conferred a very strong survival advantage (HR, 0.18; 95% CI, 0.13–0.25). Results were very similar when limited to those with bacteriologically-confirmed or culture-confirmed TB diagnoses. Other factors independently associated with death were: Thai nationality, both pulmonary and extrapulmonary TB, multidrug-resistant TB, older age, and lower or unknown CD4 cell count.
A multivariate analysis was also done to control for factors affecting likelihood to receive ART, as described above. After controlling for propensity to receive ART, the hazard ratio for death remained substantially lower among patients treated with ART (HR, 0.17; 95% CI, 0.12–0.24).
The researchers concluded that, in this resource-limited, public-health setting in Thailand, "HIV-infected TB patients who received ART had approximately one sixth the risk of death of those not receiving ART. The survival benefit persisted even for those with a very low CD4 count." However, despite this "convincing evidence that most, if not all, HIV-infected TB patients … should receive ART during TB treatment", over half of the patients in this study did not receive it. These results "should reassure providers that the benefits of ART [in this scenario] outweigh the risks." Expanding the use of ART in HIV-infected TB patients will, in particular, require targeted efforts to increase ART use in the private sector and rural areas.
Reference:
Sanguanwongse N et al. Antiretroviral therapy for HIV-infected tuberculosis patients saves lives but needs to be used more frequently in Thailand. J Acquir Immune Defic Syndr 2008;48:181–189.