Minimising the gap between the start of TB treatment and the
start of antiretroviral treatment had the biggest impact on death rates in
people with very low CD4 counts, but made much less difference in patients
with higher CD4 counts, observational cohort data from Rwanda show.
The findings, presented at the 41st Union World Conference on Lung Health in Berlin, also demonstrated a very similar reduction in the
risk of death in patients treated early to that observed in the CAMELIA study,
a randomised comparison of immediate versus delayed antiretroviral treatment in
TB patients conducted in Cambodia.
The question of when to start antiretroviral treatment in
people already receiving TB treatment has been the focus of considerable
attention among researchers due to concerns about drug interactions between
rifampicin and some antiretroviral drugs.
There has also been concern about the possible development
of immune reconstitution inflammatory syndrome (IRIS) in people who have not cleared
TB, and of a potential risk of death due to severe cases of IRIS.
For all these reasons there has been a tendency to delay the
start of antiretroviral treatment until at least two months after starting TB
treatment. Some physicians have preferred to delay antiretroviral treatment
until after the completion of TB treatment, particularly in those with less
advanced HIV disease and higher CD4 counts.
However, a large study in South Africa, the SAPIT trial,
reported in 2008 that delaying antiretroviral treatment until after the
completion of TB treatment was associated with a signficantly higher risk of death.
World Health Organization guidelines have subsequently been
updated to recommend that antiretroviral treatment should be commenced in all
HIV-positive patients receiving TB treatment, regardless of CD4 count.
What remains unclear is just how soon after starting TB
treatment it is necessary to start antiretroviral therapy in order to minimise
the risk of death.
The SAPIT study is continuing with a comparison of starting
antiretroviral treatment no more than two months after TB treatment, or starting
within two weeks of starting TB treatment.
A US AIDS Clinical Trials Group (ACTG) study, ACTG 5221, is
also comparing the effects of initiating antiretroviral therapy either two or
eight weeks after starting TB treatment.
A study conducted in Cambodia, the
CAMELIA trial, recently reported that among a group of patients with very
advanced HIV disease, starting antiretroviral treatment within two weeks of TB
treatment was associated with a significantly reduced risk of death.
Intriguingly, the reduction in risk became more pronounced over time, and was
not confined to the early period of TB treatment.
However, the results of the CAMELIA study may not be
applicable to all settings, or to patients with higher CD4 counts, so in
advance of results from further studies, researchers from Harvard
and Partners in Health used routinely collected observational data from Rwanda
to examine whether a similar pattern held true in their cohort.
In particular, they wanted to examine whether the results of
the CAMELIA study were generalisable to settings where it is not possible to
carry out microbiological confirmation of TB (microbiologically confirmed TB
was an entry criterion for the CAMELIA study).
Their study reviewed data from medical records of 308
patients with CD4 counts below 350 who had started TB treatment and
subsequently started HIV treatment.
The researchers evaluated the two-year survival rate based
on the delay between starting TB and HIV treatment regimens – 15, 30, 60 or 180
They controlled for baseline confounding factors (CD4 count,
age, gender, rural or urban care, inpatient or outpatient care), and for
time-variant confounding factors (current CD4 count and hospitalisation).
In this population, only 17% of patients had smear-positive
pulmonary TB. Twenty-five per cent had smear-negative pulmonary TB; the
remainder had extrapulmonary TB.
The median interval between starting TB treatment and
starting ART was 72 days. Starting antiretroviral treatment within 15 days of
TB treatment had a significant protective effect compared with any greater
delay only in those with baseline CD4 counts below 100.
When the investigators compared the survival probability
according to baseline CD4 count in the Rwanda cohort among those who started
ART no more than 15 days after beginning TB treatment, they found a very close
fit with the survival probabilities observed in the CAMELIA study among
patients with very low CD4 counts.
Starting ART by day 15 was associated with two-year survival
probabilities of 0.82 [0.76 to 0.89] and 0.86 [0.80 to 0.92] in persons with
baseline CD4 counts below 50 and 100 respectively in the Rwanda cohort.
In comparison, the two-year survival probability observed in
the CAMELIA trial among those who initiated treatment within 15 days was 0.82
(0.78 to 0.86). In that study the median CD4 count of participants was 25
Although the authors of the study are careful not to
overstate the significance of the findings, they point out that observational
databases such as the Rwanda
cohort have a valuable role to play in informing future practice, and in
providing information that may be lacking in clinical trials.
“Failure to draw on the experience of national treatment
programs may come at a formidable cost to clinicians, patients and their
families as they await results from randomized trials,” the researchers warned.