Antiretroviral therapy during tuberculosis treatment significantly reduces the death rate in HIV-infected Thai patients

Antiretroviral therapy (ART) in Thai tuberculosis (TB) patients receiving TB treatment is associated with a substantial reduction in deaths, according to the results of a study published in the July edition of Emerging Infectious Diseases. The paper recommends early access to HIV diagnosis and treatment for TB patients.

TB is a leading cause of death in HIV-infected patients throughout the world. In resource-poor countries, HIV diagnosis is often made after TB has been diagnosed and about 6-39% of HIV-infected TB patients die during TB treatment.

Thailand has co-endemic TB and HIV with almost 600,000 Thai patients being HIV-infected and more than 90,000 TB cases occurring annually. The proportion of Thais with both HIV and TB is 12-40%. Although ART access is widespread in Thailand physicians do not provide ART to HIV-infected TB patients.

ART improves survival in HIV-infected TB patients in resource-rich countries. However, the usefulness of ART in HIV-infected TB patients in routine public health programs in resource-poor countries is not known.

In order to address this issue, a team of Thai and US (CDC) investigators analyzed data from a prospective population-based surveillance study. The major goal was to assess the impact of ART on mortality during TB treatment in HIV-infected TB patients in rural Thailand.

The setting of the study is Ubon-ratchathani, a large rural province in northeastern Thailand with a population of 1.7 million people, a TB rate of 145/100,000 persons and a HIV prevalence of 0.6% in pregnant women attending antenatal clinics in 2004. Treatment for TB or HIV is offered by 25 health facilities on an outpatient basis.

The CDC started the Thailand TB Active Surveillance Program in 2003 in which surveillance, monitoring, evaluation, and treatment of TB in Ubon-ratchathani was carried out. For every patient with a diagnosis of TB, HIV counselling and testing was provided, standardised epidemiologic data were recorded, and sputum samples were collected for laboratory analysis.

The study population were all persons registered for TB regardless of their final diagnosis. The analyses were restricted to TB patients with a laboratory confirmation of HIV infection and who were registered for TB treatment during a one-year period. Patients received standardised TB regimens according to WHO guidelines. HIV-infected TB patients were referred for ART and co-trimoxazole treatment.

The following definitions of TB patients were employed: sputum smear-positive, pulmonary; sputum smear-negative, pulmonary; and extrapulmonary.

All deaths occurring during TB treatment were classified as TB deaths, irrespective of the cause. ART use was classified according to whether it was begun before or after TB treatment, or not taken during TB treatment.

Co-trimoxazole use was classified as either taken or not taken during TB treatment. Interruptions in ART or co-trimoxazole use were not considered. CD4 counts were stratified as < 50, 50-99, 100-199, and > 200 cells/mm³.

During twelve months, HIV was diagnosed in 329 (14%) of 2,342 patients registered for TB treatment. The median age of these 329 patients was 32 years (range 10 months to 68 years), 34% were females, and 93% had new TB cases.

Among these 329 patients, 36% were sputum smear-positive, 33% were sputum smear-negative, and 31% were extrapulmonary TB. CD4 counts were not available for 41% of the patients. The median CD4 counts in the remaining patients was 53 cells/mm³ (range 1-873); 93% had CD4 counts < 200 cells/mm³.

Sputum cultures were performed in 145 (64%) of 227 patients with pulmonary TB among whom 93 (78 %) had sputum smear-positive TB. Sixty five (70%) of the 93 patients with sputum smear-positive TB had cultures which grew Mycobacterium tuberculosis (MTB). Multiple drug resistant TB (MDR) was seen in 4 (6 %) of these isolates. Of the 52 sputum smear-negative patients, only 3 (6%) were culture positive and none was MDR.

Seventy five (23%) of the patients received ART before or during TB treatment while the remaining 254 (77%) never received ART. One hundred and eighty-seven (57%) of the 329 patients were cured or completed TB treatment while 90 (30 %) died during TB treatment.

Of patients with known outcomes, death during TB treatment occurred in 5 (7%) of 71 who received ART and 94 (43%) of 219 who did not. An increased likelihood of death was associated with having an unknown CD4 count. By contrast, receiving ART or co-trimoxazole was associated with reduced mortality. The protective effect of ART against death was confirmed by multivariate analysis.

The findings clearly demonstrate that ART use significantly reduces mortality during TB treatment in rural HIV infected TB Thai patients. The policy implication is that TB patients must have early access to HIV diagnosis and treatment. Educating providers about the usefulness of ART in HIV infected TB patients must be a priority, the authors say.