Accelerated atherosclerosis an issue in the AIDS population

Issue: June 2009

ByMelvin Cheitlin, MD

When the first AIDS cases were recognized in the early 1980s, CVD proved to be a minor problem. Patients were dying from pulmonary disease, concomitant infection—frequently from organisms that had rarely before caused disease in man, Kaposi sarcoma and non-Hodgkin lymphoma. At autopsy, the finding of myocarditis, unsuspected during life, and the occasional case of cardiomyopathy and pulmonary hypertension were not significant causes of the tremendous mortality seen at that time. With the clinical development of AIDS, life expectancy was short and mortality high.

With the introduction of nucleoside reverse transcriptase inhibitors (NRTI), non-nucleoside reverse transcriptase inhibitors (NNRTI), protease inhibitors (PIs) and combined therapy termed highly active antiretroviral therapy (HAART), mortality decreased dramatically. With the proper adherence to drugs, HIV infection has become a chronic disease. Before PIs were introduced, annual mortality was 20%. Within a decade of effective therapy, the annual mortality has fallen to less than 2%.

Melvin D. Cheitlin, MD
Melvin D. Cheitlin

With patients having a longer survival, new problems have arisen that are possibly a direct effect of HAART. Abnormalities of fat distribution, including central fat accumulation and peripheral fat loss known as lipodystrophy, and dyslipidemia are seen in 70% of patients, diabetes in 8%, and insulin resistance. In addition, there is controversy concerning whether these metabolic changes characterized as the “metabolic syndrome” are the result of patients with HIV living longer or a direct effect of the HAART drugs. Finally, it is uncertain if these abnormalities have led to accelerated atherosclerosis in these patients.

HIV and CV risk factors

What is the evidence that patients with HIV have an increased incidence of coronary and vascular disease? Metabolic abnormalities have been associated with HAART drugs. Many PIs have been shown to increase cholesterol and triglyceride levels, and some PIs (ritonavir and indinavir, among others) are associated with insulin resistance. HAART markedly decreases the viral burden and subsequent endothelial damage due to the HIV organism. Some antiretroviral drugs may independently contribute to endothelial dysfunction and increase endothelial adhesion molecules.

Other CV risk factors are also increased in patients with HIV. The Multicenter AIDS Cohort Study (MACS) has followed more than 5,000 men with HIV for more than 19 years and has observed a significantly higher systolic BP in those patients taking HAART for longer than five years. The same study reported that over a four-year period the rate of developing diabetes was 4.7 cases/100 person-years in men with HIV who were taking HAART vs. 1.4 cases/100 person-years in HIV-seronegative men. HIV-positive men on HAART had more than 4.5 times the prevalence of diabetes compared with men who were HIV seronegative. The mechanism by which this occurs is possibly due to the down regulation of the glucose transporter isoform (GLUT4), which is the major transporter of glucose into muscle cells.

One of the earliest clues of an increased incidence of premature CHD and CVD associated with HIV infection came in autopsy studies of children with HIV who had these heart conditions. There are a number of studies investigating the increased incidence of CHD morbidity and mortality. Most of these studies are retrospective over a short period of time with conflicting results. From the French Hospital Database on HIV with a large number of patients with HIV followed for three years (88,029 person-years) there were 60 men with an acute MI including 49 patients on PIs. Patients on PIs for longer than 18 months had twice the risk compared with individuals with a shorter exposure. There was no increase in risk in those on NRTIs or NNRTIs. The problem with this study is that no adjustment for hypertension or smoking was made and there is no HIV-seronegative group. The study suggests that prolonged PI use increases the risk of MI.

The Kaiser Permanente of Northern California study evaluated more than 4,000 men with HIV and found no association between PI use and rates of hospitalization for CHD over a 5.5-year period of exposure. When the study was extended to 9.5 years, the CHD and MI rates were significantly higher in HIV-positive men than in HIV-seronegative men. There was no difference between those taking and not taking PIs.

The 1993 to 2001 U.S. Veterans Study is a retrospective study of more than 36,000 HIV-positive veterans for risk of CV or cerebrovascular events. There was no significant increase in risk, but the exposure to HAART was only 15 months, which is a relatively short time to develop vascular disease if HAART was responsible. A follow-up to this study of 41,213 Veterans Association patients between 1993 and 2003 similarly showed no difference for serious CVD events in those using HAART.

The Data Collection on Adverse Events of Antiretroviral Drugs Study Group prospectively studied the incidence of MI in more than 23,000 patients with HIV, including 24% women. The median age was relatively young (39 years), and the prevalence of CVD was low (1.5%). However, the CVD risk factors in this group were strikingly high, with 56% current or past smokers, 50% with dyslipidemia, 7% with hypertension and 2.8% with diabetes. The incidence of CVD increased independently of risk factors with increasing exposure to combined antiretroviral therapy. There was a 26% relative increase in the risk of MI/year of exposure during the first four to six years of use. The value of HAART in prolonging survival in this population must be balanced against this small risk of CVD.

Since thymidine analogues (zidovudine and stavudine) have been associated with dyslipidemia and insulin resistance, the cumulative recent (<6 months) vs. past (>6 months) exposure to these drugs was examined and no increase in risk of CVD was found in a 2007 study of more than 33,000 patients. However, recent use of abacavir and didanosine was associated with an increased risk of MI (RR 1.9 with abacavir and 1.4 with didanosine) not explained by classical CV risk factors. The increased risk for developing MI decreased after exposure to abacavir was stopped, suggesting that there may be a direct effect on the myocardium from the drug.

The prevalence of major CVD risk factors were evaluated in both men and women with HIV who participated in the Women’s Interagency HIV Study (WIHS) and the Multicenter AIDS Cohort Study (MACS). These risks were compared with those of HIV-seronegative control patients. The predicted 10-year risk based on the risk factors of developing CHD among the HIV-positive men was low (81%), moderate (2%) and high (17%). Among the HIV-seronegative men, the predicted 10-year risk was low (84%), moderate (5%) and high (11%). Among HIV-positive women, the predicted risk was low (86%), moderate (2%) and high (12%) and similar in the HIV-seronegative women. The risk of developing CHD was 43% lower in those not taking PIs compared with those patients taking PIs. There was a high incidence of modifiable classical risk factors, with 40% current smokers and >40% of men and >60% of women overweight or obese.

The SMART study, one of the largest treatment interruption studies, included 5,472 patients with HIV, and 79 of those patients had acute fatal and nonfatal MI. There was a 60% higher risk of MI in patients who stopped HAART when the CD4 count rose to >350/mm³ and started again when the CD4 count dropped to 250 mm³ compared with the group on continuous therapy. This suggests that suppression of HIV may be important in reducing proinflammatory cytokines.

These studies suggest there is probably an increase in the risk of developing coronary or vascular disease in patients with HIV compared with control patients who are HIV seronegative. It is unclear whether this increased risk is due to the HIV infection itself, the HAART drugs or both. There are ongoing studies attempting to assess the contribution of the HIV organism, the vascular toxicity of the HAART drugs and the metabolic abnormalities and high-risk CVD classical risk factor burden seen in patients with HIV. Until we have more information, the value of HAART in treating these patients is so valuable that it outweighs any measured increase in the risk of developing coronary or vascular disease. However, it is important to recognize the potential increased risk and to vigorously alter the modifiable risk factors, and to be alert to the possibility of premature coronary and vascular disease in patients with HIV.

Melvin Cheitlin, MD, is Emeritus Professor of Medicine, University of California, San Francisco. He is also a member of the CHD and Prevention section of the Cardiology Today Editorial Board.

For more information:

Acta Physiol Scand. 2005;183:75-88.
AIDS. 1998 May 7;12(7):F51-8.
AIDS. 2003;17:765-8.
AIDS. 2003;17:2479-86.
AIDS. 2005;19:953-60.
Am Heart J. 2005;150:933.
Arch Intern Med. 2005;165:1179-84.
Clin Infect Dis. 2007;45:1074-81.
Engl J Med. 1998;338:853-60.
J Acquir Immune Defic Syndr. 2002;30:471-7.
J Acquir Immune Defic Syndr. 2005;39:199-202.
J Acquir Immune Defic Syndr. 2005;40:12-9.
Klein, D. #737. Presented at: 13th Conference on Retroviruses and Opportunistic Infections, February 5-8, 2006; Denver.
Kuller, L. #139 Presented at: 15th Conference on Retroviruses and Opportunistic Infections; February, 3-6, 2008; Boston.
Lancet. 2006 Aug 5;368(9534):451-8.
Lancet. 2008;371(9622):1417-26.
Lancet Infect Dis. 2004;4:213-22.
N Engl J Med. 2003;348:702-10.
N Engl J Med. 2003;349:1993-2003.
N Engl J Med. 2005;352(1):48-62.
N Engl J Med. 2007;356:1723-35.
Pediatr Pathol. 1987;7:261-75.