CROI: More data on role of antiretrovirals, interventions, HCV, in cardiovascular disease risk

Although the biggest news regarding cardiovascular disease risk at last month's Thirteenth Conference on Retroviruses and Opportunistic Infections (CROI) came from the Data Collection on Adverse Events of Anti-HIV Drugs (D:A:D) study, reported here, several other important studies presented in Denver provided additional data on the role of antiretrovirals, interventions, and hepatitis C virus (HCV) in cardiovascular disease risk in both adults and children.

Could aggressive interventions be reducing the long-term risk of PI-induced cardiovascular disease?

Kaiser Permanente of Northern California is a private health insurance organisation looking after more than three million people in their own medical facilities, with more than 5000 HIV-positive patients. Utilising medical records they systematically identified rates of cardiovascular disease, including heart attack, or myocardial infarction, and compared the risk in HIV-positive males aged 35-64 with men aged 35-64 not diagnosed HIV-positive. (Klein)

Out of 5430 HIV-positive men, 140 experienced a cardiovascular disease event - of which 86 were heart attacks - over the nine-and-a-half years of observation. The other cardiovascular disease events consisted of atherosclerosis (hardening and narrowing of the arteries) which can lead to coronary artery disease and angina. Compared with the HIV-negative men of the same age, the HIV-positive men were found to have significantly higher rates of both cardiovascular disease (2.9 vs. 6.0 events per 1000 patient-years; p

When the rates of cardiovascular disease - and heart attacks, specifically - were compared between the HIV-positive men who had received protease inhibitors (PIs), for a mean of 4.5 years (resulting in 15,527 person-years of follow-up) with those had never received PIs (for a mean of 2.3 years, and 11,390 person-years of follow-up), the investigators found that there was a trend towards an increased risk with PI exposure. They found that the age-adjusted relative risk of a heart attack was increased by 16% for each year of PI exposure, roughly equivalent to a doubling of the risk every six years.

Although 16% is exactly the same increased relative heart attack risk for yearly PI exposure reported by the D:A:D study, in this case the investigators found that there was only a trend for this 16% increased risk (p=0.112) and that the risk was not necessarily cumulative. Rather, the relative risk for heart attacks on PIs appeared to have peaked at between four and six years of PI exposure, and was actually reduced after six years.

This may well have been due to the introduction of interventions that seek to modify traditional cardiovascular disease risks - like stopping smoking and using lipid- and blood pressure-reducing medication - as well as the availability of atazanavir (Reyataz), a newer PI that appears to have a neutral effect on lipids.

The investigators found that:

• between 2001 and 2005, the percentage of PI-treated patients on atazanavir rose from 6% to 35%
• between 1997 and 2005 the percentage of PI-treated patients on lipid lowering therapy increased from 1% to 27%
• and the percentage of current smokers fell from 21.1% in 2002-3 to 17.9% in 2004-5.

Together, these interventions appeared to offset the increased risk of both cumulative PI exposure and of ageing: the investigators found significantly improved mean total cholesterol, HDL and systolic blood pressure levels between 2000-1 and 2004-5 (p

Are traditional risk factors the most important driver of cardiovascular disease?

Interestingly, data from the 8000-strong HIV Outpatient Study (HOPS) cohort both supports and contrasts with the Kaiser Permanente study. Although they found that lipid-lowering medications reduced the risk of cardiovascular disease, they found that switching to atazanavir, or a non-nucleoside (NNRTI) had no effect on cardiovascular disease risk. (Lichtenstein)

HOPS is an ongoing multicentre study of HIV-positive individuals in seven US cities. Between 1993-2005, out of 8024 cohort members, there were 209 cardiovascular disease events, including 57 heart attacks, 44 strokes and 86 coronary artery disease events. Although incidence of heart attacks peaked between 2000-2 (at 3.5 events per 1000 patient-years) and was around 1.1 events per 1000 patient-years in 2005, incidence of strokes (2.8 events per 1000 patient-years) and coronary artery disease (11.5 events per 1000 patient-years) were highest in 2005.

When the investigators analysed the relative risk factors associated with cardiovascular disease events (restricted to the 1807 patients diagnosed with hyperlipidaemia, or high blood fats) they found that the risk of cardiovascular disease was associated only with traditional risk factors: age over 40 (Adjusted Odds Ratio, AOR, 3.31; p

Although the investigators found that use of lipid-lowering drugs reduced the risk of cardiovascular disease by two-thirds (Hazard Ratio 0.34; 95% CI, 0.14-0.85; p=0.021), no link was found with a specific antiretroviral agent. In addition, no reduction in risk was found with a switch to atazanavir (p=1.0) or to any NNRTI (p=0.27) from a PI. However, even the much larger D:A:D study does not include enough patients to allow detection of the role of individual protease inhibitors on the risk of cardiovascular disease.

A smaller study from Boston (Gerrior) which examined coronary calcification - a marker of subclinical atherosclerosis - in 129 HIV-positive individuals over three years, which found that those with calcification at baseline were more likely to have calcification after three years, also concluded that it was traditional risk factors that were the main driver of cardiovascular disease in their cohort.

Prematurely 'ageing' arteries in adults and children

A study from Frankfurt, Germany (Stephan) suggests that HIV-positive individuals have the equivalent artery health of HIV-negative individuals who are up to five years older, and that this means a higher risk of cardiovascular disease.

Investigators from Frankfurt's Goethe University compared artery wall, or intima-media, thickness in 292 HIV-positive individuals diagnosed HIV-positive a median 10.6 years ago, with 1168 individuals enrolled in a cardiovascular risk study who had not tested positive for HIV. Previous studies have shown that increased artery wall thickness is linked to an increased risk of heart attack.

They found that the HIV-positive individuals (95.5% of whom had taken, or were currently taking, antiretroviral therapy) were significantly more likely to have increased wall thickness at several artery locations. The investigators found no correlation between any individual drug or class of antiretrovirals.

However, the HIV-positive individuals were much more likely to smoke than the HIV-negative controls (27 vs. 13.4 pack-years; p

They concluded that HIV-positive individuals have artery health similar to HIV-negative people that are four or five years older, and that these 'ageing' arteries means an increase in cardiovascular risk of between 4-14% over five years.

Something similar appears to be happening in HIV-positive children on antiretroviral therapy, according to a small study from the United States. (McComsey)

Investigators measured intima-media thickness and other cardiovascular risk factors in 27 HIV-positive children with a median age of ten with 17 age-matched HIV-negative controls. The HIV-positive children had been on antiretroviral therapy for a median of 79 months, and on PIs for a median of 29 months.

Total and LDL cholesterol and triglycerides were significantly higher in the HIV-positive children (p=0.01 or greater for all three), and carotid artery wall thickness was also found to be greater in the HIV-positive children on the both the left (p=0.038) and right (p=0.08) sides.

The investigators concluded that these differences indicate that HIV-positive children on antiretroviral therapy may be at an increased risk of premature cardiovascular disease. They are following-up these children for 144 weeks, and will report their results at a future conference.

Does HCV genotype 3 reduce cardiovascular disease risk?

It has previously been reported by investigators from Spain and the United States that hepatitis C virus (HCV) coinfection reduces levels of certain blood fats.

At this year's CROI, researchers from Italy compared 415 HIV monoinfected and 307 HIV/HCV coinfected individuals on PI- or NNRTI-based antiretroviral therapy between January 2001 and November 2004. They, too, found that HIV/HCV coinfection resulted in lower levels of total cholesterol and triglycerides independently of antiretroviral therapy (Lapadula).

However, when they examined lipid levels by HCV genotype, they found that HCV genotype 3 (HCV-3) reduced lipid levels even more than the other HCV genotypes.

In comparison to HIV monoinfected individuals, HCV-3 was associated with a 45-fold lower risk of high (>200mg/dL) total cholesterol (p

Similarly, when the investigators compared triglyceride levels in HCV-3 coinfected individuals with HIV monoinfected individuals, HCV-3 was associated with a 39-fold lower risk of high (>180mg/dL) triglycerides (p=0.017). In contrast, the other HCV genotypes were associated with a 7-fold lower risk of high triglycerides, but this was not statistically significant (p=0.558).

The investigators concluded that hepatitis C virus (HCV) affected elevations in total cholesterol and triglycerides independently of antiretroviral therapy. Whilst HCV coinfection was associated with protection from high total cholesterol, and HCV coinfection per se was associated with a lower risk of high triglycerides on antiretroviral therapy, it is HCV-3 that is the driving factor here.

The exact reasons for HCV's protective effects - and HCV-3's in particular - against lipid elevations is not known. It is also not yet known whether these lower lipid levels translate into a lower cardiovascular disease risk over the long-term.