Challenges in the Treatment of TB and HIV Co-Infection

March 16, 2012

By Santosh Vardhana, MD/PhD

Faculty Peer Reviewed

Ms. T is a 32- year-old woman with no past medical history who presents with a three month history of productive cough, shortness of breath, and a twenty pound weight loss. On review of systems, she also reports night sweats. On physical exam, she is cachectic. Pulmonary exam reveals dry bibasilar inspiratory crackles. Rapid HIV test is positive, and CD4 count returns at 46. Chest X-ray reveals bilateral increased interstitial markings at the lung bases as well as pleural thickening adjacent to the right lower lobe. Sputum for AFB is positive x2.

What are the concerns with co-initiation of HAART and anti-TB therapy, and what are the recommendations based on the most recent randomized controlled trials?

Cases of co-infection with Mycobacterium tuberculosis (TB) and the human immunodeficiency virus (HIV) have been prevalent since initial descriptions of an acquired immune deficiency syndrome in Zambia [1], Haiti [2], and New York City [3]. Indeed, despite vast amounts of progress in both anti-HIV and anti-TB therapy, co-infection of TB and HIV continues to be a common problem, with estimates ranging as high as 30% both in the United States and particularly in developing countries [4]. Beyond simple co-incidence, HIV seropositivity has been found to be a significant risk factor for reactivation of latent tuberculosis [5]. The molecular mechanisms underlying this link have been largely elucidated; CD4+ T cells activated by TB infection produce pro-inflammatory cytokines such as IFN [6], which in turn recruit and activate macrophages [7] that are ultimately required for formation of granulomas to wall off sites of infection [8]. Subsequent studies have found that anti-tuberculosis T cells are rapidly depleted following acute HIV infection [9]. In turn, pulmonary tuberculosis infection has been found to accelerate HIV viral replication within involved lung segments [10]. This may account for the high rate of immunological failure and poor CD4+ T cell count recovery observed in patients diagnosed with incident tuberculosis during anti-retroviral therapy (ART) [11]. Initiation of ART is critical to preventing treatment failure in TB and results in an 80% reduction in incident TB even in the absence of anti-TB therapy [12].

Thus, initiation of ART along with anti-TB therapy is critical both to elimination of TB and to prevention of HIV progression. However, a number of concerns are present with regards to co-initiation of anti-TB and anti-retroviral therapy. One concern with initiation of ART prior to complete resolution of tuberculosis infection is that of immune reconstitution inflammatory syndrome (IRIS). IRIS is thought to be the product of ART-induced rapid recovery of antigen-experienced CD4+ T cells, which initiate a massive, Th1-predominant inflammatory response to a disseminated yet previously clinically silent opportunistic infection (OI). This manifests as either a new, disseminated presentation of an OI or marked worsening of a previously diagnosed OI. IRIS was first described in 1998, with the identification of five HIV+ patients with CD4 counts less than 50 cells/microliter who developed the systemic inflammatory response syndrome (SIRS) shortly after treatment with the protease inhibitor indinavir; subsequent lymph node biopsies revealed a massive immune response to disseminated Mycobacterium avium complex (MAC) infection [13]. IRIS has been described in response to a number of other organisms, but occurs particularly in the setting of mycobacterial, fungal, and viral infections (all of which potently activate CD4+ T cells) [14]. (A spectacular review published in 2002 provides an in-depth characterization of early immune reconstitution as well as a comprehensive discussion of the various infections known to induce IRIS [15]). Some studies suggest that up to 45% of patients with TB/HIV co-infection may be at risk of developing IRIS following initiation of ART; risk factors for development of IRIS include with CD4 counts less than 100 and disseminated TB infection at baseline as well as rapid increases in absolute CD4 count following initiation of ART [16].

In 2006, a consortium of clinicians and researchers formed the International Network for the Study of HIV-associated IRIS (INSHI) and developed consensus case definitions for tuberculosis-associated IRIS. The consortium improved on the initial definition of IRIS, provided by French et al in 2004 [17] which required either atypical, exaggerated, or paradoxically worsening clinical presentation of an OI in the context of decreasing HIV viral load or increasing CD4 T cell count, by issuing consensus criteria for tuberculosis-associated IRIS in which they eliminated the need for objective measures of improvement in response to HAART (as studies had shown that IRIS can emerge before either viral load reduction or CD4 count increases are detectable) and definitively split TB-IRIS into two categories: paradoxical TB-associated IRIS, in which patients with known TB undergoing anti-TB therapy experience worsening of symptoms following HAART initiation, and unmasking TB-associated IRIS, in which HAART therapy results in new presentation of previously undetected (likely subclinical) TB infection. Paradoxical and unmasking TB-associated IRIS have distinct risk factors and prognoses. Paradoxical TB-IRIS is generally self-limited, with a median symptom duration of 2 months [18] and low risk of mortality. Unmasking TB-associated IRIS has been more difficult to describe, in part because of the difficulty in distinguishing it from new TB infection occurring during early HAART initiation (when the patient is likely still immunocompromised and susceptible to infection); however, some studies suggest that unmasking TB-associated IRIS may have a worse prognosis due to a heightened inflammatory response to a more diffuse, subclinical infectious process [19].

Thus, the potential of IRIS may serve as a relative contraindication to initiating antiretroviral therapy in a patient with known tuberculosis. The severity of this contraindication would depend largely on:

a) the prevalence of IRIS following initiation of ART

b) the mortality from IRIS as compared to delayed ART initiation, and

c) the balance between improvement in IRIS-related morbidity and AIDS-related morbidity from delaying ART initiation.

Reports estimating the prevalence of TB-IRIS in patients with undergoing new ART are variable, ranging from as low as 7.6% in one observational study conducted in India [20] to as high as 32% in the US [21]; rates of IRIS appear to be particularly high in patients who have initiated anti-TB therapy in the two months prior to starting ART (likely due to high circulating levels of antigen) [22]. A meta-analysis of 54 cohort studies containing over 13,000 patients starting ART, of whom nearly 1700 developed IRIS, reported an prevalence of 15.7% in patients with tuberculosis (as compared to nearly 20% of patients with cryptococcal meningitis and 38% of patients with CMV retinitis) [23]. The analysis also reported a cumulative mortality rate of 3.2% in patients with TB-IRIS; however, mortality in the various studies ranged from less than 1% to nearly 10%. Notably, the overall mortality rate of TB-IRIS was far lower than cryptococcal IRIS (over 20%). Also of note in this meta-analysis was the fact that development of IRIS was independent of socioeconomic status, suggesting that IRIS is not restricted to developing countries in which the prevalence of opportunistic infections is presumably higher.

Thus, existing evidence suggested IRIS is a common complication of combination antiretroviral and anti-TB therapy and raised the possibility that delaying ART until anti-TB therapy was either in the continuation (less intensive) phase or even completed might be beneficial by reducing the infectious burden and risk of IRIS. The first definitive insight with regards to this question was found in the preliminary results of the Starting Antiretroviral therapy at three Points In Tuberculosis (SAPIT) study, published in the New England Journal in early 2010 [24].

In this study, Abdool Karim and colleagues randomized 642 patients in South Africa with a diagnosis of both tuberculosis (sputum smear confirmed) and HIV (with a CD4 count of less than 500) to 3 groups: initiation of ART at the onset of TB treatment, following intensive phase of TB treatment (2 months), or following completion of treatment (6 months). This report was published following early termination of the sequential ART treatment arm after one year of follow-up due to a 54% increase in mortality as compared to the integrated arms (13% vs. 6%, p<0.001); however, the study has also reported an almost threefold increase in incidence of IRIS (12.4% vs. 3.8%, p < 0.001) in the integrated arm, demonstrating a clear adverse component to co-initiation of anti-TB and anti-retroviral therapy that might still benefit from a delay within the integrated arm.

The initial SAPIT study established that integrated ART and anti-TB therapy was superior to sequential anti-TB and ART; however, the question of whether temporarily delaying initiation of ART until completion of the intensive anti-TB treatment phase remained unanswered until publication of the 18-month follow-up of the patients in the integrated arm of the SAPIT study in October 2011 [25]. In these 429 patients that were randomized to anti-retroviral therapy either during the intensive or continuation phase of anti-TB therapy, early initiation of ART was associated with a twofold increase in IRIS (20% vs. 8%, p < 0.001) without any improvement in mortality (7% vs. 7%, NS). Notably, however, subgroup analysis demonstrated a 60% reduction in mortality (8% vs. 20%, p = 0.17) in patients with a CD4 count <50; this too was at the expense of an increased risk of IRIS (38% vs. 11%, p = 0.01). In patients with CD4 counts of 50-500, early ART was associated with a twofold increased risk of IRIS without an improvement in mortality. Thus, the composite SAPIT study results suggested immediate initiation of ART in patients with CD4 counts less than 50 but deferment of ART until completion of the intensive TB therapy phase in patients with CD4 counts of 50-500.

It is notable, however, that the inclusion criteria for patients in the SAPIT study included patients for whom ART is not necessarily indicated (CD4 counts >350) [26]; it remained possible that in patients in whom ART is definitively indicated (CD4<200), co-initiation of ART and anti-TB therapy might still be beneficial. That question was the subject of the Cambodian Early versus Late Introduction of Antiretrovirals (CAMELIA) in which patients with a CD4 count <200 were randomized to ART either two or eight weeks after initiation of anti-TB therapy. Preliminary reports from this study showed a 34% improvement in mortality in the early treatment arm, despite an almost fourfold elevation in incidence of IRIS [27]. At the completion of a two-year follow-up, this benefit was sustained; a 35% reduction in mortality was seen in the early ART group (17% vs. 26%, p = 0.002) despite a twofold increased risk of IRIS (33% vs. 14%, p < 0.001) [28]. However, despite the inclusion criteria of CD4 counts less than 200, the average CD4 count of patients participating in this study was 25. Therefore, while confirming the benefit of early ART initiation in patients with CD4 counts less than 50, this study did not resolve the question of when to initiate ART in patients with CD4 counts between 50 and 200. However, this answer may have come with the publication of the AIDS Clinical Trial Group Study A5221 [29]. In this study, 809 patients with known or suspected TB as well as HIV with CD4 counts <250 were randomized to initiation of ART either during the intensive or continuation phase of anti-TB therapy. Of note, the average CD4 count in this population was 77. At one year of follow-up, early initiation of ART resulted in a twofold increase in IRIS (11% vs. 5%, p = 0.002) without any improvement in mortality (6% vs. 7%, NS). Subgroup analysis of patients with CD4 counts less than 50 confirmed the results of prior studies, with a 42% reduction in the combined endpoint of mortality and AIDS-defining conditions (16% vs. 27%, p = 0.02) while no difference was seen in patients with CD4 counts of 50-250 (12% vs. 10%, NS). The results of all four studies are summarized in the table below (Table 1).

Thus, these studies indicate that the choice of when to initiate ART depends on the risk of progression to AIDS and death. Based on the SAPIT study, sequential therapy clearly results in higher mortality rates; however, synthesis of the SAPIT, CAMELIA, and A5221 studies suggest that initiation of ART during the intensive phase of anti-TB therapy should be restricted to patients with CD4<50; the remainder of patients should initiate ART during the continuation phase of anti-TB therapy.

Given the clear mortality benefit of integrated as compared to sequential ART/anti-TB therapy seen in SAPIT, integrated therapy is likely to become the standard of care in patients with new diagnoses of TB and HIV, and as demonstrated in all three RCTs, this carries with it a significantly elevated risk of IRIS. A continuing area of interest, therefore, is whether suppression of the hyperinflammatory response using steroids may help either prevent or alleviate IRIS in response to co-initiation of ART and anti-TB therapy, given the established data demonstrating a benefit for adjunctive steroid therapy in tuberculous pericarditis [30, 31] as well as data suggesting a possible, although controversial benefit of dexamethasone pretreatment for tuberculous meningitis [32, 33]. To date, no studies have investigated the use of prophylactic steroids at the time of initiation of ART and anti-TB combination therapy, but one RCT of prednisone given to patients who developed paradoxical TB-IRIS following initiation of ART showed a reduction in inpatient hospital days, improvement in quality of life, and improvement in chest radiographic findings with prednisone therapy at early, but not late timepoints; however, this was somewhat tempered by the increased risk of infection seen in the treatment group [34]. One other study noted an improvement in the hypercytokinemia seen in TB-IRIS patients, particularly of the pro-inflammatory cytokines IL-6 and TNF-, with prednisone therapy [35]. Further studies will be needed to assess for definitive clinical benefit with steroid therapy either at the time of initiation of ART or for resultant IRIS.


Current evidence suggests that development of IRIS is a common risk with co-initiation of anti-TB and anti-retroviral therapy. Evidence from RCTs suggests that the benefits of co-initiation of therapy outweigh the risks associated with delaying initiation of HAART until completion of anti-TB therapy, however, delaying initiation of ART until completion of the intensive phase of anti-TB treatment appears to lower the incidence of IRIS without affecting mortality in patients other than those with severely depressed CD4 counts. All patients receiving simultaneous treatment for TB and HIV should be closely monitored for development of TB-IRIS; if it develops, supportive care is appropriate, with some evidence for initiation of prednisone in patients with severe inflammatory manifestations.

Table 1. Comparison of recent RCTs evaluating co-initiation of ART and anti-TB therapy.

NEJM issue 362:697 365:1492 365:1482 365:1471
CD4 cutoff 500 500 250 200
TB status known known known/susp known
early tx (w) 0-8 0 2 2
late tx (w) 26 8 8-12 8
Age 34 34 34 35
CD4 147 150 77 25
f/u (y) 1 1.5 1 2
AIDS/ACM (early) N/A 8% 13% N/A
AIDS/ACM (late) N/A 9% 16% N/A
ACM (early) 6% 7% 6% 17%
ACM (late) 13% 7% 7% 26%
ACM HR 0.46 1.00 0.95 0.65
IRIS (early) 12% 20% 11% 33%
IRIS (late) 4% 8% 5% 14%
IRIS HR 3.3 2.4 2.2 2.4
CD4<50 AIDS/ACM (early) 11% 16%
AIDS/ACM (late) 29% 27%
AIDS/ACM HR 0.38 0.58
CD4>50 AIDS/ACM (early) 8% 12%
AIDS/ACM (late) 5% 10%
AIDS/ACM HR 1.6 1.1

Santosh Vardhana is a recent graduate from the MSTP program at NYU School of Medicine

Peer Reviewed by Melanie Maslow, Infectious Disease, Section Editor,  Clinical Correlations

Image courtesty of Wikimedia Commons


1. Melbye, M., et al., Evidence for heterosexual transmission and clinical manifestations of human immunodeficiency virus infection and related conditions in Lusaka, Zambia. Lancet, 1986. 2(8516): p. 1113-5.

2. Pitchenik, A.E., et al., Tuberculosis, atypical mycobacteriosis, and the acquired immunodeficiency syndrome among Haitian and non-Haitian patients in south Florida. Ann Intern Med, 1984. 101(5): p. 641-5.

3. Masur, H., et al., Opportunistic infection in previously healthy women. Initial manifestations of a community-acquired cellular immunodeficiency. Ann Intern Med, 1982. 97(4): p. 533-9.

4. Awoyemi, O.B., O.M. Ige, and B.O. Onadeko, Prevalence of active pulmonary tuberculosis in human immunodeficiency virus seropositive adult patients in University College Hospital, Ibadan, Nigeria. Afr J Med Med Sci, 2002. 31(4): p. 329-32.

5. Selwyn, P.A., et al., A prospective study of the risk of tuberculosis among intravenous drug users with human immunodeficiency virus infection. N Engl J Med, 1989. 320(9): p. 545-50.

6. Oftung, F., et al., Epitopes of the Mycobacterium tuberculosis 65-kilodalton protein antigen as recognized by human T cells. J Immunol, 1988. 141(8): p. 2749-54.

7. Orme, I.M., et al., Cytokine secretion by CD4 T lymphocytes acquired in response to Mycobacterium tuberculosis infection. J Immunol, 1993. 151(1): p. 518-25.

8. Hansch, H.C., et al., Mechanisms of granuloma formation in murine Mycobacterium avium infection: the contribution of CD4+ T cells. Int Immunol, 1996. 8(8): p. 1299-310.

9. Geldmacher, C., et al., Early depletion of Mycobacterium tuberculosis-specific T helper 1 cell responses after HIV-1 infection. J Infect Dis, 2008. 198(11): p. 1590-8.

10. Nakata, K., et al., Mycobacterium tuberculosis enhances human immunodeficiency virus-1 replication in the lung. Am J Respir Crit Care Med, 1997. 155(3): p. 996-1003.

11. Hermans, S.M., et al., Incident tuberculosis during antiretroviral therapy contributes to suboptimal immune reconstitution in a large urban HIV clinic in sub-Saharan Africa. PLoS One, 2010. 5(5): p. e10527.

12. Miranda, A., et al., Impact of antiretroviral therapy on the incidence of tuberculosis: the Brazilian experience, 1995-2001. PLoS One, 2007. 2(9): p. e826.

13. Race, E.M., et al., Focal mycobacterial lymphadenitis following initiation of protease-inhibitor therapy in patients with advanced HIV-1 disease. Lancet, 1998. 351(9098): p. 252-5.

14. Meintjes, G., et al., Tuberculosis-associated immune reconstitution inflammatory syndrome: case definitions for use in resource-limited settings. Lancet Infect Dis, 2008. 8(8): p. 516-23.

15. Shelburne, S.A., 3rd, et al., Immune reconstitution inflammatory syndrome: emergence of a unique syndrome during highly active antiretroviral therapy. Medicine (Baltimore), 2002. 81(3): p. 213-27.

16. Breton, G., et al., Determinants of immune reconstitution inflammatory syndrome in HIV type 1-infected patients with tuberculosis after initiation of antiretroviral therapy. Clin Infect Dis, 2004. 39(11): p. 1709-12.

17. French, M.A., P. Price, and S.F. Stone, Immune restoration disease after antiretroviral therapy. AIDS, 2004. 18(12): p. 1615-27.

18. Olalla, J., et al., Paradoxical responses in a cohort of HIV-1-infected patients with mycobacterial disease. Int J Tuberc Lung Dis, 2002. 6(1): p. 71-5.

19. Breen, R.A., et al., Does immune reconstitution syndrome promote active tuberculosis in patients receiving highly active antiretroviral therapy? AIDS, 2005. 19(11): p. 1201-6.

20. Kumarasamy, N., et al., Incidence of immune reconstitution syndrome in HIV/tuberculosis-coinfected patients after initiation of generic antiretroviral therapy in India. J Acquir Immune Defic Syndr, 2004. 37(5): p. 1574-6.

21. Shelburne, S.A., et al., Incidence and risk factors for immune reconstitution inflammatory syndrome during highly active antiretroviral therapy. AIDS, 2005. 19(4): p. 399-406.

22. Lawn, S.D., et al., Tuberculosis-associated immune reconstitution disease: incidence, risk factors and impact in an antiretroviral treatment service in South Africa. AIDS, 2007. 21(3): p. 335-41.

23. Muller, M., et al., Immune reconstitution inflammatory syndrome in patients starting antiretroviral therapy for HIV infection: a systematic review and meta-analysis. Lancet Infect Dis, 2010. 10(4): p. 251-61.

24. Abdool Karim, S.S., et al., Timing of initiation of antiretroviral drugs during tuberculosis therapy. N Engl J Med, 2010. 362(8): p. 697-706.

25. Abdool Karim, S.S., et al., Integration of antiretroviral therapy with tuberculosis treatment. N Engl J Med, 2011. 365(16): p. 1492-1501.

26. Writing committee for the CASCADE collaboration, Timing of HAART initiation and clinical outcomes in Human Immunodeficiency Virus type 1 seroconverters. Arch Intern Med, 2011. 171(17): p. 1560-1569.

27. Piggott, D.A. and P.C. Karakousis, Timing of antiretroviral therapy for HIV in the setting of TB treatment. Clin Dev Immunol, 2011. 2011: p. 103917.

28. Blanc, F., et al., Earlier versus later start of antiretroviral therapy in HIV-infected adults with tuberculosis. N Engl J Med, 2011. 365(16): p. 1471-81.

29. Havlir, D.V., et al., Timing of antiretroviral therapy for HIV-1 infection and tuberculosis. N Engl J Med, 2011. 365(16): p. 1482-91.

30. Hakim, J. G., et al., Double blind randomised placebo controlled trial of adjunctive prednisolone in the treatment of effusive tuberculous pericarditis in HIV seropositive patients. Heart, 2000. 84: p. 183-188.

31. Strang, J. I. G., et al., Controlled trial of prednisolone as adjuvant in treatment of tuberculous constrictive pericarditis in Transkei. Lancet, 1987. 2(8573): p. 1418-22.

32. Thwaites, G.E., et al., Dexamethasone for the treatment of tuberculous meningitis in adolescents and adults. N Engl J Med, 2004. 351(17): p. 1741-51.

33. Prasad, K., Singh, M.B., Corticosteroids for managing tuberculous meningitis. Cochrane Database Syst Rev, 2008. 1: CD002244.

34. Meintjes, G., et al., Randomized placebo-controlled trial of prednisone for paradoxical tuberculosis-associated immune reconstitution inflammatory syndrome. AIDS, 2010. 24(15): p. 2381-90.

35. Tadokera, R., et al., Hypercytokinaemia accompanies HIV-tuberculosis immune reconstitution inflammatory syndrome. Eur Respir J, 2010.

One comment on “Challenges in the Treatment of TB and HIV Co-Infection

Comments are closed.