AEGiS-14IAC: Does tuberculosis increase HIV viral load?

14th International AIDS Conference

Barcelona, Spain - July 7-12, 2002

Does tuberculosis increase HIV viral load?

Int Conf AIDS 2002 Jul 7-12; 14:(abstract no. MoOrC1100)

Day JH, Grant AD, Fielding KL, Moloi VM, Morris L, Chaisson RE, Cock KM, Hayes RJ, Churchyard GJ
Aurum Health Research, Welkom, South Africa

BACKGROUND: Tuberculosis (TB) is the most important disease among HIV-infected persons world-wide. It is proposed that TB-associated immune activation may accelerate the progression of HIV disease. HIV viral load (VL) is elevated in individuals during episodes of TB and remains high at treatment completion. To determine whether TB causes this elevation in VL, we measured VL sequentially in a cohort of patients and compared final VL in those experiencing an episode of TB with controls who did not.

METHODS: 1400 consenting HIV-infected mineworkers had plasma samples stored 6-monthly, and at the time of any TB episode. VL was estimated on baseline (>3 m pre-TB) and final (>5 m post-TB) samples from individuals who developed TB, and from controls who did not experience TB or other major disease episode. Controls were matched to cases on baseline CD4 and duration of follow-up. A linear regression model was used to determine the ratio of mean final VL among cases vs. controls.

RESULTS: 17 patients with TB were compared to 29 controls; they were comparable in mean age (41.6 vs. 39.1 yrs), median baseline CD4 count (292 vs. 295/ é l) and median duration of follow-up (69 vs. 69 weeks). Baseline VL was significantly higher in individuals who subsequently experienced TB vs. controls (geometric mean [GM] VL 57,787 vs. 16, 639 copies/ml, P=0.03). Final VL was higher in cases than controls (GM VL 91,939 vs. 25,652 copies/ml, unadjusted ratio 3.6, P=0.02) but after adjustment for matched set, age, previous TB and baseline VL the difference became non-significant (adjusted ratio 1.5, 95% CI 0.7-2.6, P=0.3).

CONCLUSIONS: An episode of TB was not associated with a significant increase in final VL; the apparent difference in final VL between individuals with TB vs. controls seems attributable to the difference in baseline VL. This suggests that high VL may be a risk factor for rather than a consequence of tuberculosis in HIV-infected patients.

Keywords: AEGIS, Tuberculosis, CD4 Lymphocyte Count, HIV Infections, HIV, Anti-HIV Agents, HIV Seropositivity, Risk Factors, Antigens, CD4, Case-Control Studies, Human, immunology

020707
MoOrC1100