icon star paper   Articles  
Back grey_arrow_rt.gif
How bad is HAART for the HEART?
Peter Reiss
AIDS 2003; 17(17):2529-2531

From the Department of Infectious Diseases, Tropical Medicine and AIDS, and National AIDS Therapy Evaluation Center, Academic Medical Center, University of Amsterdam, the Netherlands.
This past week two published studies were reported on the risk of heart disease in HIV & associated with HAART. The DAD Study found an increased but low risk associated with HAART, you can read the article at this link:
The second study was reported by a French research group and found the risk of myocardial infarction increased with the duration of PI therapy. This editorial by Dr Reiss discusses his perspectives on these findings, the risks, potential limitations of the studies, and studies that report conflicting findings;
Current combination antiretroviral therapy (ART), particularly when containing HIV protease inhibitors (PI) has become associated with a range of metabolic abnormalities. These include dyslipidaemia, glucose intolerance associated with insulin resistance, as well as changes in the normal distribution of body fat resulting in loss of peripheral fat (lipoatrophy) and accumulation of 'central' fat. The latter is particularly, but not solely, reflected in an increase of intra-abdominal fat. Together these signs and symptoms have become known as the lipodystrophy syndrome. Very similar to what had been observed during the early course of treatment of HIV-infected patients, studies involving healthy HIV-uninfected volunteers have found that, at least for some of the licensed HIV PI, lipid and glucose metabolism may directly and rapidly be affected by these drugs, giving rise to elevated levels of both total and low density lipoprotein (LDL)-cholesterol and triglycerides, as well as to measurable degrees of insulin resistance and glucose intolerance. The molecular mechanisms underlying these changes have recently been partly characterized. The abnormal distribution of body fat, which in most cases only becomes noticable clinically during the second year of treatment, by itself may well provide a mechanism which reinforces and sustains the dyslipidaemia and insulin resistance initially triggered by PI-containing treatment. There is evidence that both PI and nucleoside reverse transcriptase inhibitors (NRTI) contribute to the pathogenesis of the changed distribution of body fat. PI may interfere with normal adipocyte differentiation, while it has been hypothesized that NRTI may be involved in inducing peripheral lipoatrophy possibly by virtue of depleting mitochondrial DNA and thereby adversely affecting mitochondrial function in specific adipose tissue depots. The mechanisms underlying the simultaneous loss of fat in certain areas and the gain of fat in others, remain incompletely understood.
Unfortunately these metabolic adverse effects associated with current ART are very common and have raised the concern that patients over the longer term may be at increased risk of developing premature coronary artery disease (CAD) as well as other clinical manifestations of atherosclerotic vascular disease. This is not surprising given that insulin resistance, intra-abdominal fat accumulation, and dyslipidaemia, in the absence of HIV infection, are all known to substantially increase the risk of CAD. Furthermore, these changes often are superimposed on additional pre-existing well-known risk factors for CAD such as cigarette smoking, which have been found to be commonly present in the average HIV-infected patient population receiving care in industrialized countries. Moreover, a greater intima-media thickness of carotid arteries, which is recognized to be a surrogate marker for the development of CAD, has been reported to be associated with the use of PI-containing ART, albeit not consistently in all studies performed to date.
The study by Mary-Krause et al. in this issue of AIDS involving almost 35 000 HIV-infected men providing almost 40,000 person-years of exposure to PI-containing ART, now indeed provides clinical evidence for an increased risk of myocardial infarction (MI) in relation to a more prolonged duration of treatment with PI. The findings, however, should be interpreted with caution in view of a number of potential biases. Although the authors have tried to correct for this, it cannot be definitively ruled out that the degree of risk attributed to PI exposure may have been biased by the lack of information concerning the influence of 'classic' cardiovascular risk factors such as smoking and hypertension. A lower rate of reporting MI during the early period of observation as a result of a lower degree of suspicion on the part of treating physicians for such events when PI were first introduced in 1996, may have resulted in an artificial overestimation of the rise in MI incidence over time.
In fact, another recently published retrospective study, similar in size to the one of Mary-Krause et al. and conducted among US veterans, failed to show any significant increase in cardio- and cerebrovascular events during the period between 1993 and 2001, despite the introduction and increasing use of potent combination ART including PI-containing ART. In contrast, initial analyses from the largest ongoing prospective study which was specifically designed to address the issue at hand, the 'Data collection on Adverse Events of Anti HIV drugs' study (DAD) providing over 36,000 years of follow up in more than 23 000 patients, after adjustment for demographic risk factors including age, showed a 26% relative increase in the rate of MI per year of exposure to combination antiretroviral therapy.
Potential explanations for these seemingly conflicting results may include differences in study design, in measures taken for the proper ascertainment of outcome measures including MI, as well as differences in the patient populations which were studied. For the time being, the overall absolute risk of premature CAD in HIV-1 infected patients treated with combination ART is likely to be only moderately increased. Clearly this does not outweigh the marked benefit which such treatment confers in terms of reducing HIV-1- associated morbidity and mortality. Nonetheless, in view of the biologically plausible relationship between ART-associated changes in lipid as well as glucose metabolism and atherogenesis, continued monitoring of the incidence of atherosclerotic vascular disease in patients who are increasingly exposed to these treatments remains warranted.
The continuation of studies such as DAD hopefully may also serve to establish algorithms which are valid to assess CAD risk specifically in patients with HIV-1 infection.
It is important to keep in mind that in the setting of HIV-1 infection, mechanisms other than ART-associated dyslipidaemia, insulin resistance and changes in body fat distribution, may be contributing to CAD risk. For instance, it is possible that PI may promote atherosclerosis independently of dyslipidaemia by increasing cholesterol accumulation in arterial wall macrophages, as was recently demonstrated in vitro and in vivo in a murine model. Conversely, combination ART, by suppressing HIV-1, may inhibit markers of vascular activation resulting from chronic virus infection. Therapy-induced dyslipidaemia may thereby to some degree be counterbalanced and the risk of developing atherosclerosis reduced. Finally, it is important to realize that the changes in plasma lipids and lipoproteins are not equal for regimens including different classes of drugs. For instance, non-nucleoside reverse transcriptase inhibitor (NNRTI)-, particularly nevirapine ,containing ART regimens have been shown to increase total and LDL-cholesterol to a lesser extent than PI, and to simultaneously markedly increase high density lipoprotein-cholesterol, potentially reducing the atherogenic potential of such regimens. Ultimately, the interplay between all of these potential mechanisms and factors taken together is what will determine the risk of an HIV-1-infected patient developing premature atherosclerosis and vascular disease.
Although much still remains to be investigated and learned, from a clinical perspective it seems prudent to at least consider individually managing patients with respect to their risk of developing CAD. This should include attempts to modify known cardiovascular risk factors such as cigarette smoking and hypertension, the promotion of diet and exercise, as well as the cautious use of lipid-lowering and anti-diabetic agents according to available guidelines.
In addition, where appropriate from the perspective of achieving and/or maintaining HIV-1 suppression, consideration should be given to avoiding the use of currently available PI, particularly in those with a high a priori risk of CAD. In such patients NNRTI-containing regimens may be the preferred option in view of their potentially more beneficial lipid and lipoprotein profile.
  icon paper stack View Older Articles   Back to Top   www.natap.org