The debate over whether the nucleoside reverse-transcriptase
inhibitor abacavir increases the risk for myocardial infarction (MI)
continues, with 4 presentations on the topic presented here at the
16th Conference on Retroviruses and Opportunistic Infections (CROI).

Data from the SMART and D:A:D studies presented last summer at the
International AIDS Conference in Mexico City suggested that the use
of abacavir heightened the risk for cardiovascular disease (CVD), a
contention that the drug's manufacturer, GlaxoSmithKline, disputed
with their own reanalysis of their clinical-trials data.

Here in Montreal, Jens Lundgren, MD, from the University of
Copenhagen, in Denmark, presented an additional analysis from D:A:D,
showing no statistically significant associations between recent or
cumulative use of tenofovir, zalcitabine, zidovudine, stavudine, or
lamivudine and risk for MI. Antiretroviral drugs associated with a
statistically significantly increased risk for MI were abacavir,
didanosine, indinavir, and lopinavir/ritonavir.

A separate caseˇVcontrol study from France (268 case patients, 865
control patients) found that the risk for MI was increased by
cumulative exposure to lopinavir and amprenavir or fosamprenavir.
More curiously, initiating abacavir also carried that increased risk,
but longer exposure to it did not.

But a third study, an analysis of the MACS and WIHS cohort data,
concluded that abacavir was not associated with elevated levels of
high sensitivity C-reactive protein, interleukin-6, or D-dimer levels.

Finally, Carl Grunfeld, MD, from the University of California, San
Francisco, reported on a cross-sectional study that compared scans
for carotid intima medial thickness in HIV-infected patients (n =
433) and control subjects (n = 5748). After adjustment for
demographic factors and CVD risk factors, he found that HIV infection
independently conferred a risk for greater intima medial thickness,
which is similar to other traditional CVD risk factors, such as
smoking.

"When you advise patients in your office, you can say, here is your
standard risk [for CVD], but because you have HIV, you are at higher
risk," he later told reporters. "It is the equivalent of being male,
smoking, or having diabetes. I believe that the effect of HIV is much
larger than the small signal seen with those drugs," such as abacavir.

Dr. Grunfeld acknowledged that 1 of the limitations in studying this
effect is that "97% of [HIV] patients have been treated, and 94%
[are] on HAART," so they can only compare with matched controls in
other CVD studies.

It was left to Peter Reiss, MD, from the University of Amsterdam, in
the Netherlands, to try to pull together these and other
presentations at CROI in a summary session on what it all means. He
said: "In a simplistic way, 4 studies basically give a signal of an
increased risk" of MI with abacavir, "and 3 do not."

He noted that the studies showing a signal generally were prospective
and more robust in size, and the study population also was 7 to 10
years older. "Age is a very important cardiovascular disease risk
factor. You could speculate that maybe it is just harder to pick up a
signal in that younger population."

Dr. Reiss said that inflammation really underlies all of the
processes of plaque formation and growth. He reviewed how untreated
HIV "is associated with upregulation of endothelial activation,
inflammation, and coagulation," markers that lead to CVD.

He said there is not that much evidence for an effect of
abacavir. "Once HIV is suppressed or the virus is replicating at low
levels, there is no consistent pattern emerging; there are still
question marks, but there is a hint that in that particular
circumstance, some of the inflammatory markers may be upregulated."

He acknowledged that if abacavir is proinflammatory, "against a
background of a very high inflammatory state" induced by HIV, it may
be difficult to discern such an additional signal. "By treating HIV
and downregulating the inflammatory state, maybe you have a better
opportunity for picking up a signal" from abacavir.

"The clinical data with this on/off phenomenon suggests a subacute
and reversible process rather than an aggressive progressive
pathogenic mechanism," Dr. Reiss said.

Clinically, Dr. Reiss urged a common-sense approach of identifying
and modifying underlying risk factors for CVD in individual
patients. "In those who are at high risk, and where there are clear
alternatives for abacavir, I think you just take the abacavir away.
But if there is no alternative, you continue on abacavir." For
someone who is at very low risk for CVD, abacavir only modestly
increases that risk; the absolute risk will be very low, he said.