Does Tuberculosis Increase HIV Load?

Abstract & Commentary

Synopsis: Poor prognosis for HIV-infected individuals after TB may be due to preexisting high HIV load, rather than to the TB event itself.

Source: Day JH, et al. Does Tuberculosis Increase HIV Load? J Infect Dis. 2004;190:1677-1684.

This is an observational cohort study that compared HIV patients who acquired TB or bacterial pneumonia with comparison of those 2 groups of patients without TB or bacterial pneumonia, respective ly. The site of the study was Free State, South Africa between 1999-2002. The regional hospital was the sole source of secondary care. Patients were prophlylaxed for PCP using cotrimoxazole, with CD4 cell counts that were < 200cells/uL. Antiretroviral therapy was not used in the clinic for these patients.

Pneumonia or TB was diagnosed either at hospital admission by a study team or at subsequent clinic visits. TB was diagnosed by culture or by response at 2 month to anti-TB therapy. The bacterial pneumonia group were diagnosed as having a chest film typical of bacterial pneumonia and had "culture of bacteria that typically cause pneumonia."

The control group was diagnosed with HIV without TB 3 months before study entry or 3 months after the final samples of the study had been drawn, and the same was true for bacterial pneumonia, except with a 1 month window before study entry and final blood draws.

The results compared 2 relatively small groups. The TB group had 30 TB patients, with 56 controls, and the pneumonia group had 14 patients and 35 controls. The TB group, but not the pneumonia group, showed a significant HIV load rise, however, both groups started with viral loads around log (10) = 4.5. The TB group increased from log (10) 4.73 to 5.02, and the pneumonia group from 4.47 to 4.51.

The most interesting finding was a CD4 cell decrease per year of 53.0 for the TB group, versus 6.7 for the control group. Amazingly, 778 other participants enrolled in the study for a minimum of 6 months, and actually had a CD4 decrease of only a 17.4/yr. On the other hand, the HIV load difference was not different in patients who were stratified by CD4 count above and below 200.

Comment by Joseph F. John, Jr., MD

The impression that TB, like some other opportunistic infections, accelerates the progression of HIV disease, is not supported by this study. From this study, it is more likely to conclude that higher viral loads, at the outset of TB disease, are more attributable to poor outcome. These data also show that HIV viral load is likely independent of CD4 count as a risk factor for TB. Day and colleagues further extrapolate that a difference, as seen in this study, though small, of 0.25 log (10), would amount to a difference of about 1 year in the time to progression of severe HIV disease or death. This extrapolation is based on pre-ART data, but is interesting, and stresses the fact that these patients have NO antiretroviral therapy on board, and thus, are susceptible to the full effect of unabated and uncontrolled viral loads.

The implication of this study, especially for the developing world and South Africa where there are high rates of HIV and TB, is that treatment of HIV should be paramount. By intervening in the natural history of HIV using ART, whatever role TB plays in hastening severe disease and death, can be minimized. The world waits as countries like South Africa, with some outside help, mobilize to make ART a reality for many of their HIV-infected population.

Joseph F. John, Jr., MD, Chief, Medical Subspecialty Services, Ralph H. Johnson Veterans Administration Medical Center; Professor of Medicine, Medical University of South Carolina, Charleston, SC, is Co-Editor of Infectious Disease Alert.