Starting Antiretroviral Therapy during Tuberculosis

HIV positive people with tuberculosis (TB) who start treatment for both infections at the same time have a lower risk of death than those who initiate antiretroviral therapy (ART) only after completing TB treatment, according to a study from South African published in the February 25, 2010 New England Journal of Medicine. While caution is warranted when combining HIV and TB drugs, rates of adverse events in this study were similar in the integrated and sequential therapy arms.

B is a major cause of death for people with HIV worldwide, especially in low-income countries. Mortality rates are high among HIV/TB coinfected patients who present for care with advanced disease, but the optimal timing for ART initiation in relation to tuberculosis therapy has been controversial.

People who start ART and experience large CD4 cell gains with untreated TB can develop immune reconstitution inflammatory syndrome (IRIS) as the recovering immune system starts to attacks existing pathogens in the body. Furthermore, some commonly used anti-TB drugs can interact with protease inhibitor and NNRTI antiretroviral agents, potentially causing worse side effects or compromising effectiveness.

In the present study, Salim Abdool Karim from the Centre for the AIDS Programme of Research in South Africa and colleagues compared simultaneous versus sequential treatment in 642 people with HIV and TB in Durban, South Africa.

In this controlled open-label trial, 429 adult participants were randomly assigned to receive integrated antiretroviral and anti-TB therapy, starting ART either within 4 weeks of TB treatment initiation (early integrated) or within 4 weeks after completing the intensive phase (late integrated); 213 patients were assigned to begin sequential ART within 4 weeks after completing TB treatment. Men and women were about equally represented, and the average age was 34 years. Participants all had a CD4 cell count < 500 cells/mm3, with a low median of about 150 cells/mm3.

All patients received standard TB therapy according to South African guidelines, which stipulate that a first episode of TB should betreated with a 2-month intensive combination regimen ofrifampin, isoniazid, ethambutol, and pyrazinamide, followed by a 4-month continuation regimen of isoniazid plus rifampin. Patients with prior tuberculosis received a 3-month intensive regimen (including streptomycin for the first 2 months), followed by a 5-month continuation phase.

Patients were offered directly observed therapy by clinic nurses, but some selected observers in their community, home, or workplace. Participants also received prophylaxis with trimethoprim-sulfamethoxazole. ART consisted of a once-daily regimen of didanosine (ddI, Videx), lamivudine (3TC, Epivir), and efavirenz (Sustiva or Stocrin).

In September 2008, based on interim data, a data and safety monitoring committee recommended that the sequential therapy arm should be halted ahead of schedule and those patients should start ART; the 2 integrated therapy groups continued. The recently published analysis includes data from the combined integrated therapy groups and the sequential therapy group up to that point.


  • Patients in the 2 integrated therapy groups had a significantly lower rate of death than those in the sequential therapy group (5.4 vs 12.1 deaths per 100 person-years).
  • This difference represented a relative risk reduction of 56% (hazard ratio 0.44 for the integrated therapy arm).
  • Mortality was lower in the integrated therapy groups compared with the sequential treatment group at all CD4 cell levels.
  • Among patients with CD4 counts < 200 cells/mm3 (indicative of AIDS), the mortality rate was 46% lower in the integrated therapygroups compared with the sequential therapy group.
  • Although the number of deaths among patients with 200-500 cells/mm3 was small, there was a trend toward lower mortality in the integrated therapy groups.
  • 12.4% of patients in the integrated therapy groups developed IRIS, compared with only 3.8% inthe sequential therapy group; however, there were no required ART changes or deaths due to IRIS.
  • Rates of serious (grade 3 or 4) adverse events not related to IRIS were similar in the integrated and sequential therapy arms (30 vs 32 per 100 person-years).

Based on these findings, the study authors concluded, "The initiation of antiretroviral therapy during tuberculosis therapy significantly improved survival and provides further impetus for the integration of tuberculosis and HIV services."

"The interval between the completion of tuberculosis therapy and the initiation of antiretroviral therapy is important; a considerable number of deaths in the sequential-therapy group occurred during this time," they elaborated in their discussion. "Once antiretroviral therapy was initiated,however, it was associated with similarly high levels of viral suppression in the two study groups."

Since even participants with relatively high CD4 counts of 200-500 cells/mm3 benefited from integrated therapy, the investigators suggested that treatment guidelines should recommend integrated therapy for everyone with a CD4 count < 500 cells/mm3. World Health Organization (WHO) antiretroviral guidelines in effect at the time of this study called for ART below 200 cells/mm3, but the threshold was recently raised to 350 cells/mm3, closer to guidelines for high-income countries.

Centre for the AIDS Programme of Research in South Africa (CAPRISA), University of KwaZulu-Natal, Durban, South Africa; Department of Epidemiology and International Center for AIDS Care and Treatment Programs (ICAP), Mailman School of Public Health, Columbia University, New York, NY; Department of Internal Medicine and Epidemiology, Yale University School of Medicine, New Haven, CT.



S Abdool Karim, K Naidoo, A Grobler, and others. Timing of Initiation of Antiretroviral Drugs during Tuberculosis Therapy. New England Journal of Medicine 362(8): 697-706 (Free full text). February 25, 2010.