A Review of the Cardiovascular Complications of HIV/HAART
- Hypertension in HIV-Infected Men and Women
- Hypertension in HIV-Infected Women
- Cardiovascular Risk of PI-Containing HAART
As morbidity and mortality rates due to HIV/AIDS have dramatically decreased over the past eight years, the focus of clinical management concerns has shifted to non-opportunistic diseases. A number of studies over the past few years, for example, have strongly suggested that there is an increased risk of cardiovascular disease in HIV-infected patients compared with age-matched controls.
Moreover, it appears that this risk is further increased for patients taking protease inhibitor (PI)-containing HAART, although there is controversy regarding this point. Several reports at this meeting continue to provide us with more information regarding the cardiovascular complications of HIV/AIDS in the third decade of the epidemic.
At last year's CROI meeting, results from the D:A:D cohort (The Data Collection on Adverse Events of Anti-HIV Drugs) -- which is a large, prospective, international, observational database of more than 23,000 HIV-positive patients -- demonstrated that the incidence of myocardial infarction increased by 26% per year of exposure to combination antiretroviral therapy (ART).1
At this year's CROI meeting, additional analyses of these patients were presented. The first study2 presented determined whether the 26% incidence rate pertained to cardiovascular events other than myocardial infarctions, and looked at whether the previous rate of myocardial infarction was similar to predictions based on the Framingham cohort. (The Framingham study is a landmark, ongoing, prospective study of the risk factors for cardiovascular disease in the general, HIV-negative population. It began in 1948.)
Two analyses were performed in this study. In the first analysis, the rate of the first cardiovascular event (myocardial infarction, invasive cardiovascular procedure, stroke or death from cardiovascular events other than a myocardial infarction) was found to be 5.5/1,000 person years in 35,151 person years of follow up: 199 cardiovascular events occurred, including 121 myocardial infarctions and 30 strokes.
The results confirmed that the longer the exposure to ART, the more a patient was at risk for a cardiovascular event, and that this rate was similar to using myocardial infarction as a single endpoint, which confirms the D:A:D results reported at CROI last year.
In the second part of this study, the Framingham model was used to determine the expected rate of myocardial infarction during the D:A:D follow up, and to ascertain whether the increased rate seen with increased exposure to ART was due to conventional cardiac risk factors.
Those patients not receiving ART had an observed rate of myocardial infarction of 3/1,000 person years compared to the predicted rate of 7.6/1,000 person years. In patients receiving ART, the number of myocardial infarctions was slightly higher than what is predicted by the Framingham equation, although they were still within the 95% confidence limits. Because the D:A:D cohort included patients from Europe, the U.S. and Australia, the Framingham equation was controlled for differences in cardiovascular events in different countries by using data from the World Health Organization.
The trends seen with the increased rate of myocardial infarction with a longer duration of exposure to ART were similar to those expected on the basis of a change in known cardiovascular risk factors. The authors concluded that the increase was due to ART-induced changes in conventional risk factors.
The results of this study, then, extend the previous findings suggesting that the longer someone is exposed to ART, the greater their risk for a cardiovascular event occurring. However, it must be kept in mind that the risk remains small compared with aged-matched controls based on the Framingham study. Whether ART causes the increased risk remains to be determined.
In the meantime, aggressive modification of known coronary risk factors needs to be a mainstay of our treatment strategies, treatment needs to be individualized and drugs that are known to raise lipids need to be avoided by patients with known coronary risks.
The second study3 presented from the D:A:D cohort examined the factors affecting blood pressure in HIV-infected patients, since very little data about this issue exists in the HIV-positive patient population. Here again, two different analyses were performed in order to determine both the incidence of hypertension and the factors that may affect both systolic blood pressure (SBP) and diastolic blood pressure (DBP) in patients in the D:A:D cohort.
During a median follow-up period of 1.5 years, more than 43,000 blood pressure measurements were obtained in over 16,000 patients. The median number of measurements for each patient was three. At baseline, more than half the patients were found to be normotensive (SBP <140 and/or DBP <90 mmHg).
During the evaluation period, 487 of the 8,341 normotensive patients developed hypertension with an incidence of 35.8/1,000 person years. This incidence is at the upper limit of normal in HIV-negative patients. Risk factors for the development of hypertension were as follows: male gender (hazard ratio, HR=1.69), higher body mass index or BMI (HR=2.20 for BMI >30 kg/m2 vs. <18), older age (HR=2.08 for age group 43-83 years versus 17-33), and a higher blood pressure at baseline (HR=2.40 for SBP 130-139 mmHg versus <120).
Further analysis attempted to elucidate those factors associated with an increase in systolic blood pressure compared to a reference group. Those factors that were associated with a higher predicted systolic blood pressure were the same as those mentioned above: male gender (+7 mmHg and +6 mmHg at baseline and 24 months versus female), higher BMI (+16.4 mmHg and +15.6 at baseline and 24 months for BMI >30 kg/m2 versus <18), and older age (+12.8 mmHg and 14.5 mmHg at baseline and 24 months for 60 years old versus 30).
In addition, the use of blood pressure-lowering medications at baseline was also associated with the increase (+8.3 mmHg at baseline and 24 months for those being treated). However, neither cumulative exposure to each class of antiretrovirals, nor the type of treatment at baseline, were associated with the development of hypertension. The authors concluded that hypertension in HIV-infected persons is associated with the same risk factors as those in the general population. In addition, the results do not support an association of antiretroviral medicine with an increase in blood pressure in HIV-infected persons.
Conversely, a study4 presented from the Women's Interagency HIV Study (WIHS) reported the opposite results. WIHS is an ongoing, prospective, cohort study being conducted at six U.S. sites and involving 2,057 HIV-positive women and 569 demographically similar HIV-negative women. Enrollment began between 1994-1995. As part of an attempt to determine the relationship, if any, between hypertension and lipodystrophy, this study determined the occurrence of hypertension in the large female cohort in a similar fashion to what has previously been reported for men in the large Multi-Center AIDS Cohort Study.5 Both univariate and multivariate analyses were performed; however, only the more important multivariate analysis will be discussed here.
The prevalence of hypertension at baseline was defined as having one of the following: diastolic blood pressure >90 mmHg or systolic blood pressure >140 mmHg or current use of anti-hypertensive medication or a self-reported history of hypertension. The incidence rate of hypertension was defined as any of the first three criteria present in at least two six-month intervals, one of which must have occurred after the introduction of HAART in April 1996. The results reported here demonstrated that neither the incidence nor the incidence rate of hypertension were different between the HIV-positive and HIV-negative cohorts.
The risk factors that were determined to be significantly associated with the development of hypertension were: increasing age, African-American race, less education, increasing duration of smoking and a higher BMI (>30 kg/m2). Being pregnant had a protective effect. In addition, the use of HAART and an increasing duration of time on HAART were independently associated with developing hypertension. Interestingly, zidovudine (ZDV, Retrovir) monotherapy was protective.
As a result of these conflicting results, we are again left wondering whether the use of ART does in fact increase the risk of developing hypertension. The discordant results presented here are undoubtedly the result of the different demographics that exist in large, observational cohorts.
Moreover, methods for assessing blood pressure can be highly variable depending upon who is taking the measurement, the number of times a reading is taken at each visit, as well as other factors that have been determined to influence blood pressure in the general population. It is this author's own observation that, indeed, there may be an increased rate of hypertension in both HIV-infected patients in general, as well as those on ART. Additional studies are required in order to elucidate whether there is truly an increased rate of hypertension.
Lastly, another study attempted to determine whether PI-containing HAART increases the risk of cardiovascular disease. HIV Insight6 is a prospective, observational cohort comprised of the 10 sites from the HIV Outpatient Study cohort and another nine from primary healthcare providers. There were 7,542 patients eligible for evaluation between Jan. 1, 1996 and June 30, 2003. The median duration of follow up for the PI-containing group was 3.5 years (mean 3.5 years; maximum 7.4 years) and 2 years (mean 2.5 years; maximum 7.4 years) for the non-PI group.
Median PI use was 1.7 years, with more than 95% of patients having been exposed to a PI for more than one month and 75% of patients for more than six months. Cardiovascular disease events were defined as any one of the following: acute myocardial infarction, angina pectoris, coronary artery disease, percutaneous transluminal coronary angioplasty, coronary artery bypass graft, cerebrovascular accident, transient ischemic attack and peripheral vascular disease.
During the study period, 127 events occurred: 112 in the PI group (incidence rate of 9.8/1,000 patient years of follow up (PYFU), and 15 in the non-PI group (incidence rate of 6.5/1,000 PYFU). In the subset of patients ages 35-65, there was also a higher incidence rate of 11.5 vs. 7.9/1,000 PYFU.
There were a number of significant differences between the demographics of the two groups: In the PI group there was a longer period of follow up; plus, overall, the group was older, more white, and also included more males, more patients with hyperlipidemia at baseline, fewer smokers and fewer African-Americans. These differences were, however, relatively small. By both univariate and multivariate analyses, the risk factors for the development of cardiovascular disease included older age, smoking, diabetes mellitus, hypertension, hyperlipidemia, pre-existing coronary artery disease and PI use for more than 60 days.
Sensitivity analyses performed in which all risk factors were controlled for, were also associated with more than 60 days of PI use in both the entire cohort, as well as the 35- to 65-year-old age group. The duration of PI use was further subdivided into time periods: PI use for 1-179 days, 180-364 days, and more than 365 days. Applying this in a multivariate analysis demonstrated that only those patients who had used PIs for more than 365 days, as well as the subset of patients ages 35-65, had an increased risk of cardiovascular disease.
The results of this study are in agreement with those of the D:A:D study and suggest that indeed there is an increased incidence of cardiovascular events the longer someone is exposed to ART. The nature and type of ART need to be "teased out," if possible, in order to more fully elucidate whether PI-containing or non-PI-containing ART may be the most causative. At the present time, we are still left with trying to do what is best for our patients by weighing the pros and cons of which ART to recommend for each individual patient. Moreover, as with the general population, aggressive interventions aimed at reducing known cardiovascular risk factors need to be an integral part of patient management in those deemed to be at risk.
<a class="aname" name="1" <="" a="">Friis-Møller N., Weber R., D'Arminio Monforte A. et al. Exposure to HAART Is Associated With an Increased Risk of Myocardial Infarction: The D:A:D Study. Presented at: 10th Conference on Retroviruses and Opportunistic Infections; February 10-14, 2003; Boston, Mass. Abstract 130.
<a class="aname" name="2" <="" a="">Law M. G., D'Arminio Monforte A., Friis-Møller N. et al. Cardio- and Cerebrovascular Events and Predicted Rates of Myocardial Infarction in the D:A:D Study. Presented at: 11th Conference on Retroviruses and Opportunistic Infections; February 8-11, 2004; San Francisco, Calif. Abstract 737.
<a class="aname" name="3" <="" a="">Thiébaut R., El-Sadr W., Chenuc G. et al. Predictors of Hypertension and Changes in Blood Pressure in HIV-Infected Patients in the D:A:D Study. Presented at: 11th Conference on Retroviruses and Opportunistic Infections; February 8-11, 2004; San Francisco, Calif. Abstract 75.
<a class="aname" name="4" <="" a="">Khalsa A., Karim R., Mack W. et al. Hypertension in HIV-Infected Women Related to HAART: Women's Interagency HIV Study. Presented at: 11th Conference on Retroviruses and Opportunistic Infections; February 8-11, 2004; San Francisco, Calif. Abstract 741.
<a class="aname" name="5" <="" a="">Riddler S. A., Smit E., Cole S. R. et al. Impact of HIV Infection and HAART on Serum Lipids in Men. JAMA. 2003 Jun 11;289(22):2978-82.
<a class="aname" name="6" <="" a="">Iloeje U., Yuan Y., L'Italien G. et al. Protease Inhibitor Exposure Time and Risk of Cardiovascular Disease in HIV-Infected Patients. Presented at: 11th Conference on Retroviruses and Opportunistic Infections; February 8-11, 2004; San Francisco, Calif. Abstract 736.