Antiretroviral treatment in TB patients: patients with CD4 counts below 50 need to start ART quickly

Salim Abdool Karim and Diane Havlir at CROI 2011. Photo by Gus Cairns/
This article is more than 11 years old. Click here for more recent articles on this topic

Starting HIV treatment within two weeks of tuberculosis (TB) treatment is most necessary in patients with CD4 counts below 50 cells/mm3, while a slightly longer delay does not appear to be harmful for those with less advanced disease, two large international trials have shown.

One study also found that starting TB treatment around three months after starting antiretroviral treatment resulted in fewer cases of immune reconstitution inflammatory syndrome (IRIS) when compared to starting antiretroviral treatment within two weeks of TB treatment initiation.

The findings were reported on the opening day of the 18th Conference on Retroviruses and Opportunistic Infections in Boston.


immune reconstitution inflammatory syndrome (IRIS)

A collection of inflammatory disorders associated with paradoxical worsening (due to the ‘waking’ and improvement of the immune system) of pre-existing infectious processes following the initiation of antiretroviral therapy.



In a clinical trial, a clearly defined outcome which is used to evaluate whether a treatment is working or not. Trials usually have a single primary endpoint (e.g. having an undetectable viral load) as well as a few secondary endpoints, covering other aspects of treatment safety, tolerability and efficacy.

primary endpoint

The main result that is measured at the end of a clinical study to see if a given treatment worked (e.g., proportion of participants with viral suppression). The choice of primary endpoint is decided before the study begins.

AIDS defining condition

Any HIV-related illness included in the list of diagnostic criteria for AIDS, which in the presence of HIV infection result in an AIDS diagnosis. They include opportunistic infections and cancers that are life-threatening in a person with HIV.


Affecting the lungs.


The results confirm and extend the results of the CAMELIA trial, reported last year at the Eighteenth International AIDS Conference in Vienna, which showed that immediate antiretroviral treatment reduced the risk of death in TB patients by one-third when compared to waiting for eight weeks, in a population of Cambodian patients with a median CD4 count of 25 cells/mm3.

The high burden of TB in people with HIV infection means that decision-making about when to start antiretroviral treatment is a critical part of HIV care in many settings. Many clinicians have expressed concern about the risk of IRIS in patients with TB if antiretroviral treatment is started before TB is cured.

There have also been concerns about the use of antiretroviral drugs alongside the four-drug intensive phase of TB treatment, which usually lasts for the first eight weeks of TB treatment.

However in 2009 the World Health Organization issued new guidelines recommending that antiretroviral therapy should be started as soon as possible after starting TB treatment in co-infected patients.

The studies presented today shed further light on the question of how soon treatment needs to start, although the findings are likely to be of greater relevance in settings where CD4 counts are available to guide decision-making.

SAPIT study

The SAPIT (Starting Antiretroviral therapy at three Points In Tuberculosis therapy) trial was designed to test three strategies for antiretroviral initiation after starting TB treatment:

  • Wait until TB treatment is completed before starting ART (the 'deferred' arm)
  • Start ART within four weeks of completing the intensive phase of TB treatment (the 'later' arm)
  • Start ART within four weeks of starting TB treatment (the 'early' arm).

Three years ago the investigators of the SAPIT study reported in the New England Journal of Medicine that starting antiretroviral treatment soon after starting TB treatment significantly reduced the risk of death from all causes when compared to delaying the initiation of treatment until after the completion of TB treatment.

The SAPIT trial continued to follow patients randomised either to start treatment immediately or to wait until the completion of the intensive phase of TB treatment.

Professor Salim Abdool Karim of the Centre for AIDS Programme Research in South Africa (CAPRISA) presented final findings from the SAPIT study. The trial was conducted among people with HIV infection who had been diagnosed with smear-positive pulmonary TB in Durban, South Africa. The trial recruited individuals with CD4 counts below 500 cells/mm3.

The trial randomised 642 patients, of whom 214 were assigned to the 'early' arm and 215 to the 'later' arm. The remainder were assigned to the deferred group and are not included in this analysis. 72% completed follow-up in the early arm and 67% in the deferred arm.

The median CD4 count at baseline was 150 cells/mm3.

All participants in the study received a regimen of ddI, 3TC and efavirenz, designed to be dosed once daily at the same time as directly observed short-course TB treatment. Participants also received cotrimoxazole prophylaxis.

The primary endpoint of the study was the measurement of changes in all-cause mortality and new AIDS-defining illnesses.

The trial, showed that after two years of follow-up, there was no significant difference in the risk of AIDS or death between the early and later treatment groups, with an event rate of 6.9 and 7.8 per 100 person years of follow-up respectively in the two arms (p = 0.73).

Early treatment was associated with a 68% reduction in the risk of death or AIDS when compared to later treatment among those with CD4 counts below 50, and the incidence rate ratio was 0,32 (p = 0.06). The rate of IRIS was almost five times as high in the early treatment group (odds ratio 4.7, p = 0.01).

In patients with baseline CD4 counts above 50 cells/mm3 there was no significant benefit to early treatment, but a 2.2-fold higher risk of developing IRIS and a 6.8-fold higher risk of switching at least one drug in the antiretroviral regimen as a result of drug toxicity.

There was no significant difference between the two arms in rates of TB treatment completion or viral suppression.

Professor Karim said that the findings support further efforts to integrate HIV care into TB services in order to ensure rapid initiation of treatment for those with very low CD4 counts, but for those with higher CD4 counts, Professor Karim said the decision would remain a case-by-case judgment for the physician.

“It really depends on clinical judgment and the capability of the facilities to manage cases of IRIS, and the availability of second-line drugs for switching cases of toxicity,” said Professor Karim.

ACTG 5221 STRIDE study

A second international study, carried out by the US AIDS Clinical Trials Group at 26 sites in four continents, compared immediate and post-intensive phase initiation of ART in 906 individuals with confirmed or suspected pulmonary TB.

The study randomised patients to start treatment within four weeks of beginning TB treatment or within four weeks of beginning continuation phase TB treatment.

The study recruited 906 patients with CD4 counts below 250 cells/mm3, and stratified randomisation by CD4 count above and below 50 cells/mm3. All patients received standard TB treatment, cotrimoxazole prophylaxis and antiretroviral therapy consisting of tenofovir, FTC and efavirenz.

The median CD4 count in the study population was 77 cells/mm3.

The primary endpoint of the study was the proportion of patients who died or developed a new AIDS-defining illness during 48 weeks of follow-up after initiation of TB treatment.

There was no significant difference between the two study arms in primary endpoint outcomes across the study population as a whole (12.9% in the immediate arm vs 16.1% in the early arm, p = 0.45), but when investigators looked at primary endpoint outcomes in the participants with baseline CD4 counts below 50 cells/mm3, they found a significantly lower rate of death and new AIDS events in the immediate treatment group (15.5% vs 26.6%, p = 0.02).

There was no significant difference in deaths and new AIDS events between the two arms in patients with CD4 cell counts above 50 cells/mm3 (11.5% vs 10.3%, p = 0.67). Deaths were predominantly due to TB, cryptococcal disease, and pneumonia.

There was no significant difference in CD4 counts or viral load suppression at week 48 between the two groups.

A significant difference in the rate of IRIS was noted, with patients in the immediate treatment group 2.5 times more likely to develop IRIS (p = 0.001), although no deaths occurred in the study as a consequence of developing IRIS.

Professor Diane Havlir told a press conference that with results from three major trials all telling clinicians the same thing, it was time to speed up integration of TB services and HIV services in order to treat the patients with the most advanced HIV disease quickly. “Minutes matter” for these patients, she said.

Asked whether the findings of the STRIDE and SAPIT studies should lead to a change in guidelines, Professor Salim Abdool Karim said: “I personally think the guidelines were somewhat aggressive.”

However in practical terms the findings are unlikely to affect WHO recommendations, since the key finding of both studies – that antiretroviral treatment may be slightly less urgent in patients with CD4 counts above 50 – requires access to CD4 cell counting, which remains out of reach in many areas where the burden of HIV and TB co-infection is highest.

Abstracts and webcast

You can view the abstracts from this research on the official conference website:

Abstract 38:

Abstract 39LB:

You can also watch a webcast of the presentations made at this conference session, including the speakers Diane Havlir and Salim Abdool Karim.

Webcast from TB and Opportunistic Diseases.

We rely on donations to continue our work to help people with HIV, support us today at:


Havlir D et al. International randomised trial of immediate vs early ART in HIV+ patients treated for TB: ACTG 5221 STRIDE Study. 18th Conference on Retroviruses and Opportunistic Infections, Boston, abstract 38, 2011.

Karim SA et al. Optimal timing of ART during TB therapy: findings of the SAPiT trial. 18th Conference on Retroviruses and Opportunistic Infections, Boston, abstract 39LB, 2011.