To The Editor:
We read with interest the article by Schiffer and Sterling1 concerning the optimum time to initiate antiretroviral therapy (ART) among patients with HIV-associated tuberculosis (TB). This is a complicated issue and remains a key challenge for clinicians treating such patients, especially in resource-limited settings, where most coinfected patients live. There are many competing factors that favor early or delayed initiation of ART. These include the risks of (1) HIV-associated morbidity and mortality attributable to delays in treatment versus risks of (2) drug cotoxicity, pharmacokinetic interactions, impaired treatment adherence, and immune reconstitution disease that may be increased by earlier treatment.2-4 Randomized controlled trials are greatly needed to provide definitive data. At a meeting of the STOP TB TB/HIV Working Group in Los Angeles in February 2007, however, it was revealed that of 6 large, randomized, controlled trials designed to address this question (reviewed by Colebunders et al5), none had well-established patient recruitment and the viability of several studies was in doubt. Thus, data are unlikely to be available for quite some time.
The analysis by Schiffer and Sterling1 is therefore welcome because it provides important insights into this question by drawing on existing data and clarifying this complicated issue. Standard decision analysis techniques were used to analyze mortality risk in a hypothetical cohort of 1000 HIV-infected adults with TB and CD4 cell counts <200 cells/μL. Data inputs were derived from existing studies in which patients had received standard 6-month rifamycin-based anti-TB treatment and triple-drug ART. The 2 overriding variables that defined the optimal timing were found to be the excess mortality risk resulting from delayed ART versus the excess mortality risk from immune reconstitution disease (which is associated with early ART). From our experience in South Africa, we agree with Schiffer and Sterling that the decision on timing of ART must primarily be mortality based and that these are indeed the 2 key variables. We strongly suspect, however, that the balance between these 2 competing factors is different when comparing treatment of patients in high-income and resource-limited settings.6
The community-based ART program in South Africa where we work now serves more than 3000 patients, and at entry to the program, approximately 25% of them have active TB.7 Despite concerns over pharmacokinetic drug interactions, impaired treatment compliance, and drug cotoxicity, we have found that immunologic and virologic responses are not compromised during overlapping efavirenz-based ART (using standard doses) and rifampicin-containing TB treatment; 93% of patients have viral load suppression <400 copies/mL at 48 weeks.7,8 Moreover, in this setting, where monitoring of serum hepatic transaminases is available, we have observed no deaths resulting from hepatotoxicity during concurrent treatment.9,10 Occasional deaths from TB-associated immune reconstitution disease have been observed, however.11 Thus, we agree with Schiffer and Sterling that this remains the most important variable potentially favoring delayed treatment.
With regard to the mortality risk associated with delays in ART, however, our experience in South Africa differs markedly from the input data used by Schiffer and Sterling1 in their analysis. They used data from the Tuberculosis Trials Consortium Study 23, which was based in North America.12 In this well-conducted study, the overall mortality risk among patients with HIV-associated TB was just 2% to 3% after 12 months and the mortality risk associated with delayed ART was thus small. In contrast, those with TB at entry to our ART program in South Africa have a 12-month mortality risk that exceeds 15% despite a similar median baseline CD4 cell count in the 2 settings.7 Many deaths occur just before starting ART, and we have previously shown that because of the extremely high pre-ART mortality rate, even short delays in ART initiation may be associated with a substantial mortality risk.9,13 More specifically, in a small observational study of patients with TB in this service in whom ART was started after a median delay of 42 days from TB diagnosis, most (71%) early deaths actually occurred while patients were waiting to start ART.10 Consistent with studies from elsewhere in Africa,14,15 this high early mortality risk had a strong independent association with advanced baseline immunodeficiency but not with TB disease activity or even TB diagnoses per se.6,10 Collectively, these data suggest that patients with HIV-associated TB in this setting may benefit from early initiation of ART.
The widely differing mortality risk between our program and the North American study is consistent with the well-established finding that early mortality rates within ART programs in resource-limited settings are several-fold greater than those in high-income countries, even after adjusting for baseline immunodeficiency.16 This fact has important implications for the decision analysis of Schiffer and Sterling1 and suggests that the optimum timing for ART initiation may be different for high- and low-income settings. High excess mortality risk associated with delays in ART initiation in resource-limited countries is likely to favor earlier ART initiation in this setting. This may be partially counterbalanced if the risk of death due to TB immune reconstitution disease is greater in settings with limited secondary health care11,17 or where more hepatotoxic ART regimens are used concurrently with rifampicin. Current data from our program in South Africa, however, show that the number of deaths that might be averted by earlier treatment far exceeds the number of deaths that might in any way be attributable to early treatment.6,10
Eventually, this question is likely to be answered by randomized controlled trials, but these studies are all being conducted in resource-limited settings.5 In view of the issues outlined above, the data subsequently derived from these studies may not be directly applicable to high-income settings. To provide further insight into this issue, we suggest that Schiffer and Sterling might consider doing comparative analyses of data from resource-limited settings and high-income settings.
Stephen D. Lawn, MD*†
Robin Wood, FCP (SA)*
*The Desmond Tutu HIV Centre Institute for Infectious Disease and Molecular Medicine Faculty of Health Sciences University of Cape Town Cape Town, South Africa
†Clinical Research Unit Department of Infectious and Tropical Diseases London School of Hygiene and Tropical Medicine London, United Kingdom
S. D. Lawn is funded by the Wellcome Trust, London, United Kingdom. R. Wood is funded in part by a US National Institutes of Health RO1 grant (A1058736-01A1) and CIPRA grant 1U19AI53217-01.
1. Schiffer JT, Sterling TR. Timing of antiretroviral therapy initiation in tuberculosis patients with AIDS: a decision analysis. J Acquir Immune Defic Syndr
2. Dean GL, Edwards SG, Ives NJ, et al. Treatment of tuberculosis in HIV-infected persons in the era of highly active antiretroviral therapy. AIDS
3. Burman WJ, Jones BE. Treatment of HIV-related tuberculosis in the era of effective antiretroviral therapy. Am J Respir Crit Care Med
4. de Jong BC, Israelski DM, Corbett EL, et al. Clinical management of tuberculosis in the context of HIV infection. Annu Rev Med
5. Colebunders R, John L, Huyst V, et al. Tuberculosis immune reconstitution inflammatory syndrome in countries with limited resources. Int J Tuberc Lung Dis
6. Lawn SD, Wood R. When should antiretroviral treatment be started in patients with tuberculosis in South Africa? S Afr Med J
7. Lawn SD, Myer L, Bekker LG, et al. Burden of tuberculosis in an antiretroviral treatment programme in sub-Saharan Africa: impact on treatment outcomes and implications for tuberculosis control. AIDS
8. Lawn SD, Wood R. Impact of antituberculosis treatment on virological response to highly active antiretroviral therapy: implications for resource-limited settings? J Infect Dis
9. Lawn SD, Myer L, Orrell C, et al. Early mortality among adults accessing a community-based antiretroviral service in South Africa: implications for programme design. AIDS
10. Lawn SD, Myer L, Bekker LG, et al. Early mortality in patients with HIV-associated tuberculosis in Africa: implications for time to initiation of treatment [abstract O-126]. Presented at: 13th Conference on Retroviruses and Opportunistic Infections (CROI); 2007; Los Angeles.
11. Lawn SD, Myer L, Bekker LG, et al. Tuberculosis-associated immune reconstitution disease: incidence, risk factors and impact in an antiretroviral treatment service in South Africa. AIDS
12. Burman W, Vernon A, Khan A, et al. Timing of antiretroviral therapy during treatment of HIV-related tuberculosis [abstract 846]. Presented at: 43rd Annual Meeting of the Infectious Diseases Society of America; 2005; San Francisco.
13. Lawn SD, Myer L, Harling G, et al. Determinants of mortality and nondeath losses from an antiretroviral treatment service in South Africa: implications for program evaluation. Clin Infect Dis
14. Zachariah R, Fitzgerald M, Massaquoi M, et al. Risk factors for high early mortality in patients on antiretroviral treatment in a rural district of Malawi. AIDS
15. Stringer JS, Zulu I, Levy J, et al. Rapid scale-up of antiretroviral therapy at primary care sites in Zambia: feasibility and early outcomes. JAMA
16. Braitstein P, Brinkhof MW, Dabis F, et al. Mortality of HIV-1-infected patients in the first year of antiretroviral therapy: comparison between low-income and high-income countries. Lancet
17. Lawn SD, Bekker LG, Miller RF. Immune reconstitution disease associated with mycobacterial infections in HIV-infected individuals receiving antiretrovirals. Lancet Infect Dis