HIV treatment reduces risk of TB in South Africa more than previously thought

Antiretroviral therapy appears to reduce the risk of developing tuberculosis (TB) to a greater extent than was previously thought, according to a South African study published in the December 2^nd edition of AIDS. Investigators found that the incidence of new cases of tuberculosis fell to just one case per 100 patient years after five years of antiretroviral therapy. Risk factors for the development of tuberculosis during antiretroviral therapy included a low CD4 cell count before HIV treatment was initiated and poor immune restoration during HIV therapy.

Earlier studies suggest that the incidence of tuberculosis in HIV-positive individuals has fallen by between 70 – 90% since effective anti-HIV therapy became available. However, tuberculosis still occurs at significant levels in HIV-positive patients and models suggest that widespread use of effective anti-HIV drugs will have little impact on the overall level of tuberculosis amongst HIV-positive populations in low income countries. It is theorised that tuberculosis will occur with reduced frequency thanks to HIV treatment, but that the disease will still affect HIV-positive individuals over their extended life-span.

The World Health Organization (WHO) has developed a strategy for the integration of tuberculosis control and HIV programmes. The provision of HIV therapy will be one aspect of this combined programme and data are needed to further illuminate the impact of effective HIV treatment on tuberculosis control.

Doctors in Cape Town therefore studied the incidence of new cases of tuberculosis in individuals during their first five years of antiretroviral therapy. Data were also gathered on the risk factors for tuberculosis.

A total of 346 individuals who received potent anti-HIV therapy between 1996 and 2005 were included in the investigators’ analysis. The majority (55%) were men, the median age was 33 years, and just over half were assessed as having a low socioeconomic status. Before HIV therapy was initiated, the median CD4 cell count was 242 cells/mm³ and median viral load was 80,000 copies/ml. A total of 51% of individuals had symptoms of HIV disease or AIDS and 14% had a previous history of tuberculosis.

During a total of 1108 person years of follow-up, 27 new diagnoses of tuberculosis were made. All but five cases affected the lungs (extra-pulmonary tuberculosis is normally much more common in HIV-positive individuals) and three people with tuberculosis died.

The overall incidence of tuberculosis was 2.44 cases per 100 person years of follow-up. However, there was a significant decline in incidence from 3.35 cases per 100 person years in the first year to just 1.01 case per 100 person years in year five.

Development of tuberculosis was found to be significantly associated with the following baseline characteristics: age under 33 years (p = 0.01); a CD4 cell count below 100 cells/mm³ (p = 0.04); and symptomatic HIV disease or AIDS (p = 0.01). There was also a trend for a previous history of tuberculosis to be associated with a new tuberculosis diagnosis whilst taking anti-HIV drugs (p = 0.07).

The investigators then looked to see if there was any difference in the response to HIV therapy between patients who developed tuberculosis and those who did not. They found that although the patients who developed tuberculosis did have a response to anti-HIV therapy - with viral load falling to a median of 800 copies/ml from almost 100,000 copies/ml, and CD4 cell count increasing from a median of 112 cells/mm³ to 198 cells/mm³ - the median increase in CD4 cell count was much lower in patients with tuberculosis than those without (74 cells/mm³ vs 234 cells/mm³). “Tuberculosis”, write the investigators, “developed among patients whose immunological responses to HAART were suboptimal”, however, they also note that it was more likely to occur in individuals with very depressed immune systems before antiretroviral therapy was started and that "it is, therefore, unlikely that tuberculosis risk during long-term HAART will return to levels seen in individuals who do not have HIV infection."

“Long-term HAART confers a greater reduction in tuberculosis risk than previously reported and HAART may, therefore, contribute more to tuberculosis control in low-income countries than previously estimated”, comment the investigators.