Chances of achieving an undetectable viral
load are equally poor for patients who inject heroin, cocaine, or a combination
of these two drugs, Canadian researchers report in Drug and Alcohol Dependence. Between 56 and 58% of people who
injected these drugs had an undetectable viral load one year after starting HIV
therapy, compared to 89% of individuals who were not active injecting drug
However, when considered as a time-dependent
variable, injection of these drugs was not associated with poorer virological
outcomes. The investigators believe that this is because injecting habits are
not constant, with people switching between drugs and between injecting and
not injecting. Factors associated with a greater chance of viral suppression
were baseline viral load, methadone therapy and, most strongly of all, high
levels of adherence to HIV therapy.
The investigators therefore stress the
importance of providing adherence support to people with a history of injecting
drug use who are starting HIV therapy.
It is well recognised that active injecting
drug use is associated with poorer responses to HIV therapy. However, the
impact of specific patterns of injecting on the chances of virological
suppression is poorly understood.
Therefore, investigators from Vancouver
designed a longitudinal study involving 267 people with a history of drug
use who started HIV therapy for the first time between 1996 and 2008.
These participants were interviewed at baseline
and then every six months. They were asked if they had injected heroin, cocaine,
or a combination of these drugs in the previous six months. Those who answered
yes were classified as active injecting drug users and their chances of
achieving viral suppression (below 500 copies/ml) was compared to individuals
who did not report recent injecting. Information was also gathered on the
participants’ demographic characteristics, baseline viral load and CD4 cell count, use
of methadone therapy, type of HIV therapy taken and adherence to this treatment.
The rate of viral suppression twelve months
after starting therapy was 56% for cocaine injectors, 58% for those who
injected heroin and 56% for participants who injected both drugs.
“The effects of various drugs (e.g. heroin
vs cocaine) on HIV-1 RNA viral suppression did not differ greatly when
baseline drug use was considered,” write the authors.
In contrast, 89% of patient who did not
report current injecting behaviour achieved virological suppression. The
difference in outcome between those who injected drugs and those who did not
was significant (p < 0.01).
However, when considered longitudinally as
a time-dependent variable, the only injecting pattern associated with a
significantly lower chance of viral suppression was combining cocaine and
heroin (HR = 0.67, 95% CI, 0.47-0.97; p < 0.05). This association ceased to
be significant after controlling for factors such as CD4 cell count, viral load
and the year therapy was started.
Further statistical analysis showed that
none of the drug use patterns were associated with suppression of viral load.
Nevertheless, the investigators found that
several factors were associated with a better chance of achieving viral
suppression. These included methadone use (AHR = 1.33, 95% CI, 1.01-1.76), therapy
with a protease inhibitor (AHR = 1.35, 95% CI, 1.03-1.77), the year treatment
was initiated (AHR = 1.10, 95% CI, 1.05-1.15) and, most strongly of all,
adherence to therapy of at least 95% (AHR = 4.00, 95% CI, 2.91-5.49).
“Suppression of HIV-1 RNA was most strongly
predicted by baseline clinical characteristics, use of methadone, and adherence
to ART,” emphasise the researchers.
They conclude: “Active injecting at the
time of ART initiation is associated with lower plasma HIV-1 RNA suppression
rates…when considered longitudinally as time-updated behaviors, there was
little association between patterns of drug injecting and plasma HIV-1 RNA suppression
rates.” The authors believe these findings show “adherence interventions should
be applied at the time of ART initiation for active drug injectors”.