University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Clinical deterioration during anti-tuberculosis treatment in Africa: Incidence, causes and risk factors
Global Health Sciences Literature Digest
Published September 29, 2010
Journal Article

Pepper DJ, Marais S, Wilkinson RJ, et al. Clinical deterioration during anti-tuberculosis treatment in Africa: Incidence, causes and risk factors. BMC Infect Dis. 2010 Mar 30;10:83.

In Context

Premature treatment of HIV infection in severely immunocompromised patients with concurrent tuberculosis (TB) can result in acute clinical deterioration through the immune reconstitution syndrome (IRIS).(1) On the other hand, delaying ART until the completion of TB therapy may lead to further progression of HIV and additional opportunistic infections. The World Health Organization (WHO) currently recommends that all patients with TB co-infection be treated with ART regardless of CD4 count and to begin ART within eight weeks of beginning TB treatment.(2) This study examined the incidence, causes and risk factors for clinical deterioration in HIV-infected patients being treated for TB in South Africa.


To determine the incidence, causes and risk factors for clinical deterioration in HIV-infected patients initiating TB therapy


The setting was a TB treatment center in Khayelitsha, South Africa. Participants were ≥18 years old, beginning treatment for active TB and eligible for ART (<200 CD4 cells/µL).


This was a cohort study. The main endpoint was clinical deterioration at 24 weeks after starting TB treatment. Clinical deterioration was defined as an AIDS-defining illness (other than TB), other non-AIDS-defining illnesses related to HIV, multidrug-resistant TB, deterioration due to poor adherence to TB treatment, paradoxical TB-IRIS, paradoxical reactions in HIV-uninfected patients and acute comorbid illnesses not related to HIV or TB (e.g., diabetic deep vein thrombosis). Tuberculosis treatment was isoniazid, rifampin, pyrazinamide and ethambutol for two months followed by isoniazid and rifampin for four months. First-line ART was stavudine, lamivudine and efavirenz. Typically, ART is begun two months after the start of TB treatment, but it can be begun as soon as two weeks after starting TB treatment for patients with <50 CD4 cells/µL. The study was conducted over a three-month period in 2008.


Of the 305 patients who began TB therapy during the study period, 292 had both HIV status and clinical records available; 209 (72%) were HIV-infected and 83 (28%) were uninfected. Patients infected with HIV were less likely to be lost to follow-up (22% vs. 35%, P=0.026) but more likely to die (8% vs. 1%, P=0.048). Overall, 117 (40%) experienced clinical deterioration; of these, 101 (48%) were HIV infected, and 16 (19%) were uninfected. Clinical deterioration related to TB occurred in 101 HIV-infected patients, including 22 episodes of IRIS, and in five uninfected patients; AIDS-defining illnesses occurred in 21 HIV-infected patients (11 esophageal candidiasis, five Pneumocystis jiroveci pneumonia, four cryptococcal meningitis); and non-AIDS defining illnesses in 25 HIV-infected patients (nine patients each superficial herpes infection, bacterial infection, and fungal infection). Other comorbid illnesses occurred in 59 HIV-infected patients and 11 uninfected illnesses. In multivariate analysis, HIV infection was found to be a significant risk factor for clinical deterioration (adjusted HR 2.0, 95% CI 1.1-3.6); in the HIV-infected stratum, low CD4 count at baseline was also found to be a significant predictor of deterioration. Of the 17 deaths in HIV-infected patients, 15 were in patients who had <200 CD4 cells at baseline with a median interval of 98 days from start of TB therapy to death. Three deaths were due to IRIS.


The investigators concluded that 40% of patients suffered clinical deterioration on TB therapy, especially those with HIV and low CD4 counts.

Study Quality

Using the Newcastle-Ottawa scale for cohort studies, this study was of moderate quality primarily because of the substantial and disproportionate loss-to-follow-up rate.

Programmatic Implications

This study was conducted before the new WHO guidelines were published and treated only patients with <200 CD4 cells/µL or with extrapulmonary tuberculosis. The expansion of treatment to all patients with pulmonary TB will likely be associated with less clinical deterioration because patients will not be as sick at the time ART is begun. Nonetheless, the implication of this study is that patients with HIV infection and lower CD4 counts when they begin TB treatment are at risk of serious clinical complications.


  1. Meintjes G, Lawn SD, Scano F, et al, International Network for the Study of HIV-associated IRIS: Tuberculosis-associated immune reconstitution inflammatory syndrome: case definitions for use in resource-limited settings. Lancet Infect Dis. 2008; 8:516-23.
  2. World Health Organization. Antiretroviral therapy for HIV infection in adults and adolescents. Recommendation for a public health approach. 2010 revision. Geneva, Switzerland: World Health Organization, 2010.