CARDIOVASCULAR DISEASE IN HIV

Abstract

People living with HIV are at increased risk for cardiovascular disease. This report summarizes our work demonstrating associations between HIV and multiple subclinical markers of cardiovascular disease risk. Some of this risk appears to be related to heightened immune activation and system inflammation that is present in people living with HIV even when there is an undetectable viral load.

Case Presentation

The patient is a 45-year-old woman living with HIV with a history of prior myocardial infarction; tobacco, cocaine, and heroin use; and medication noncompliance who presents with torsade de pointes on routine ambulatory electrographic (ECG) monitoring. Resting ECG revealed QTc 560 ms and a Q wave in lead V2. We will review data demonstrating potential mechanisms for the development of this potentially lethal arrhythmia.

In the general U.S. population, the proportion of deaths attributable to cardiovascular disease (CVD) has been declining over the last few decades. However, this same proportion has nearly doubled among people living with HIV (PLWH) (1). With the advent of combination antiretroviral therapy (ART), HIV has become a chronic disease, with PLWH living a near normal lifespan (2, 3). Multiple studies have demonstrated that PLWH have an increased risk for cardiovascular diseases (4, 5). This risk is likely related to (a) the presence of a heightened inflammatory response, and systemic inflammation, even when circulating HIV RNA is undetectable; (b) a high prevalence of traditional CVD risk factors, especially smoking; and (c) metabolic side effects of some ARTs (6, 7).

In order to understand more about the risk for coronary artery disease in PLWH, we studied the association between HIV and subclinical coronary atherosclerosis in the Multicenter AIDS cohort study (MACS) (8). The MACS is a longitudinal cohort study that includes gay and bisexual men living with HIV and at-risk uninfected men (9). MACS participants undergo semiannual research visits with standardized interviews, clinical evaluation, and laboratory tests. We performed coronary CT angiography (CTA) on 759 men between the ages of 40 and 70 without any prior history of coronary interventions or cardiac surgery and excluded men with chronic kidney disease or contrast allergy. We determined the prevalence and extent of coronary plaque in each coronary segment and the degree of luminal narrowing (stenosis). We also assessed the composition of the plaque as either non-calcified, mixed, or calcified in this cross-sectional analysis.

The mean age was 53.8 years, and 30.3% of the participants were Black men. The prevalence of having any coronary artery plaque was 76.3%. We found that men living with HIV (HIV+) had a 13% greater prevalence of having any coronary atherosclerotic plaque than uninfected men (HIV–), after adjusting for demographics and CVD risk factors [adjusted prevalence ratio 1.13 (95% CI 1.04– 1.23), p =0.004]. The prevalence of non-calcified plaque was 59.2%. Men living with HIV had a 25% greater prevalence of non-calcified plaque [adjusted prevalence ratio 1.25 (95% CI 1.10– 1.43), p =0.001]. Non-calcified plaque is thought to be potentially more likely to become unstable and lead to an acute coronary syndrome. Only 15.9% of men had a coronary stenosis >50%. Lower nadir CD4 T cell count and a longer duration of treatment with combination ART were each independently associated with a greater prevalence of coronary stenosis. A longer duration of ART is likely an indication of longstanding HIV, and lower nadir CD4 cell counts reflect a longer duration of uncontrolled viremia prior to initiation of ART. Although coronary CTA is not indicated as a screening test in asymptomatic individuals, this study demonstrated a link between HIV and coronary atherosclerosis, especially non-calcified plaque. The results suggest that PLWH should be treated aggressively with proven preventive strategies to try to limit development of coronary artery disease.

Even when HIV viral load is undetectable, HIV viral particles remain present in HIV reservoirs in the body. The presence of these viral reservoirs, co-infections, and microbial translocation in the gut may contribute to immune activation and systemic inflammation. Inflammation is an integral component of the atherosclerotic process (10, 11). Therefore, we evaluated the associations between monocyte activation markers and the presence of subclinical coronary atherosclerotic plaque, including sCD163, sCD14, and CCL2. We also evaluated associations with IL-6 and sTNRa2. Higher levels of inflammation markers were seen in HIV+ than HIV– men. These higher levels were also associated with the presence of atherosclerotic plaque, especially coronary stenosis >50%. These studies support the notion that immune activation and systemic inflammation contribute to the development of atherosclerosis in people living with HIV (12, 13).

More recently, we invited the men who completed coronary CT angiography back for a repeat coronary CT angiogram to evaluate the progression of coronary atherosclerosis. The median interval between the two CT scans was 4.5 years. We included 310 HIV+ men (31% of whom were viremic) and 234 HIV– men. We divided the PLWH into two categories: those with suppressed viral load and those who were viremic during the interscan interval. Suppressed viral load included <50 copies/ml during the interscan interval but allowed for one “blip” defined as <500 copies/ml. Men with an HIV viral load >500 copies/ml, or more than one “blip,” were considered viremic. We found that HIV+ men with a detectable HIV viral load had a 2.3 times greater risk for having progression of coronary artery stenosis during the interscan interval than HIV– men (adjusted RR 2.30; 95% CI, 1.32–4.00, p = 0.003), whereas there was no difference in coronary artery stenosis progression between HIV+ men with an undetectable HIV viral load and HIV– men. In addition, among men with a suppressed viral load, suboptimal adherence to cART was associated with greater progression of coronary artery stenosis (14). This study demonstrates the potential role of viremia in the progression of coronary artery disease and reinforces the importance of adherence to effective ART, which may lessen the risk for future coronary artery disease events.

PLWH have been reported to be at increased risk for sudden cardiac death (15, 16). In order to try to understand the mechanisms for this increased risk, we performed resting 12-lead electrocardiograms (ECG) and ambulatory ECG monitoring for 14 days in MACS participants. Prolongation of the electrocardiographic QT interval, reflecting ventricular repolarization, can increase risk for ventricular arrhythmias. We found that HIV+ men had a longer QT interval on a resting ECG, even after adjusting for CVD risk factors and QT prolonging drugs than HIV– men. Some of this association appeared to be attributable to greater levels of inflammatory biomarkers seen in HIV+ men. These results demonstrate the potential influence of systemic inflammation on ventricular repolarization as a potential mechanism contributing to increased risk for sudden cardiac death in PLWH (17, 18).

We also performed ambulatory ECG monitoring. We calculated the QT variability index in 1,123 men in the MACS (589 HIV+ and 534 HIV–) during four continuous days of monitoring. QT variability reflects variance in the QT interval relative to variance in the RR interval (heart rate). QT variability is a marker of increased ventricular repolarization lability, which may increase risk for ventricular arrhythmias. We found that HIV+ men had a greater QT variability index than HIV– men, even after adjusting for demographics, CVD risk factors, and QT prolonging medications. Men with a detectable HIV viral load had a greater QT variability index than HIV+ men with an undetectable viral load, and a greater extent of HIV viremia was associated with a greater QT variability index. In addition, greater levels of inflammation markers were associated with a greater QT variability index. This study describes an additional potential mechanism that might contribute to risk for ventricular arrhythmias in PLWH, also influenced by viremia and inflammation (19).

PLWH are also at increased risk for heart failure compared with the general population. This risk has been demonstrated for both heart failure with reduced ejection fraction (HFrEF) and heart failure with preserved ejection fraction (HFpEF) (20). In order to determine potential mechanisms leading to this increased risk, we performed cardiac MRI (CMR) with gadolinium enhancement in men and women living with HIV and uninfected people at risk for HIV from three ongoing prospective cohort studies, including MACS, Women’s Interagency HIV Study (WIHS), and AIDS linked to the Intravenous Experience (ALIVE) studies. We recruited 273 PLWH and 163 people living without HIV with a mean age of 54 years, ∼70% Black. When compared to people living without HIV, PLWH had a larger left atrial size and greater extracellular volume fraction (ECV), which is a measure of diffuse interstitial fibrosis. In addition, PLWH had a greater prevalence of subepicardial non-ischemic scar pattern. These associations all persisted after adjusting for demographic, CVD risk factors, and drug and alcohol use. Of note, the prevalence of any left ventricular systolic dysfunction was very low (<1%). We did not find any associations between HIV viral load or nadir CD4 cell count with any of our CMR outcomes. Greater left atrial volume, ECV, and subepicardial scar pattern in PLWH might represent early markers of an increased risk for future development of heart failure, especially HFpEF (21).

CONCLUSIONS

These studies demonstrate increased risk for subclinical coronary artery atherosclerosis; coronary artery stenosis progression on coronary CT angiography; prolongation of the cardiac repolarization (electrocardiographic QT interval); greater QT variability index from ambulatory ECG monitoring; and greater left atrial size, ECV, and subepicardial myocardial scar on cardiac MRI in PLWH compared to uninfected individuals. Although subclinical, these parameters are associated with increased risk for cardiovascular events. Heightened immune activation and inflammation, which is frequently seen in PLWH, appeared to contribute to many of these measures, as did uncontrolled HIV with viremia.

Getting back to our case presentation, we recognize multiple risks factors for torsade de pointe, including HIV, which is associated with prolonged ECG QT interval and QT variability index; medication non-adherence, likely leading to inflammation from immune activation related to untreated HIV; heroin use, which can prolong QT interval; and coronary artery disease.

The design of these studies had multiple strengths, such as the inclusion of very well-characterized participants from existing longitudinal cohort studies. The comparisons included individuals at risk for HIV who were not infected, and there were multiple adjustments for potential confounding factors.

Additional research is needed to inform CVD prevention guidelines and recommendations for preventing cardiovascular disease in PLWH. The Randomized Trial to Prevent Vascular Events in HIV (REPRIEVE) is testing statin use in PLWH who do not have preexisting coronary artery disease (CAD) (22), and ongoing studies aim to evaluate the effects of reducing inflammation pharmacologically (23–25). We know that current CVD risk scores used in the general population underestimate risk in PLWH (26). There is a need for a refined score that includes HIV-specific variables. Until further data are available, we use the American Heart Association/American College of Cardiology (AHA/ACC) guidelines for calculation of the Pooled Cohort Equation and designation of inflammatory conditions, such as HIV, as a risk-enhancing factor (27). We encourage all medical providers to treat PLWH aggressively with proven prevention guidelines.

DISCUSSION

Reiser, Chicago: I’m so glad I stayed for your talk; it’s fascinating. I suspect that long COVID and chronic HIV have a lot in common especially as it pertains to inflammation. I’m studying inflammation myself, in innate-immunity in particular. When I listened to your talk, I was reminded that it’s really important that we start breaking open inflammation and thinking about acute inflammation, C-reactive protein, and chronic inflammation, which may be driven by the innate-immune system in many ways. We need to see how adaptive immunity really intersects with long outcomes in cardiovascular and kidney disease and specifically measure key molecules and target them. There’s a lot of value in this research so I’m going to continue to follow your work. Thank you for the presentation.

Post, Baltimore: Thank you for those kind comments. Yes, there are a lot of corollaries to the COVID pandemic. In fact, one study showed that inflammation from COVID was associated with QT prolongation, which, of course, led to concern about using hydroxychloroquine. We already had shown that in HIV, but it didn’t get much attention. There are tremendous corollaries, and we are very interested in studying long COVID. Thank you for recognizing the corollaries there.

Mackowiak, Baltimore: Thank you for your presentation. Have you looked at the role of telomere length as a risk factor for coronary artery disease in these patients and how it interacts with other things?

Post, Baltimore: Thank you for the question. Telomere length, I think most of you know, is related to aging. Researchers at John Hopkins have some problems with how aging and accelerated aging are defined, so telomere length may be a way to objectively look at that. I haven’t looked at telomere length particularly in this population, but I think it has been measured and it might be a good thing for us to do so.

Patterson, Little Rock: Have you studied individuals who never had an immunologic response after having antiretroviral therapy for a long time? Some people never immunologically recover. I’m wondering if those individuals may be distinct from people who actually do have a robust immunologic response and how that might help us understand the inflammatory response in cardiovascular risk.

Post, Baltimore: Thanks for that question. I can’t say that we actually looked at that. We looked at the opposite. You probably know that some people suppress the virus without antiretroviral therapy. We were interested in whether those people who weren’t getting antiretroviral therapy because they had normal CD4-cell counts were at increased risk for cardiovascular disease. We weren’t able to demonstrate much of a difference, but we haven’t really differentiated people who are on antiretroviral therapy and are still viremic. In this observational study, it’s kind of hard to know whether they aren’t responding to therapy versus not being adherent to therapy. We usually just look at the viral load.

Patterson, Little Rock: So you have looked at the Women’s Interagency HIV Study (WIHS) cohort as well as the Multicenter AIDS cohort study (MACS)?

Post, Baltimore: The WIHS is the corollary to the MACS, so they have now been combined into the MACS/WIHS combined cohort study. Investigators in the WIHS are just starting to perform coronary CT angiography. I haven’t seen the results, but we have been collaborating with them for many years doing carotid ultrasound. We do most of our studies together looking at both men and women and find very similar results in men and women. The men are men who have sex with men (MSM), whereas the women are generally disadvantaged and often are using drugs or have multiple sexual partners. Lots of social determinants of health differ between the men and women, which makes it a little harder to look at true sex differences.

Wilson, Durham: I wanted to ask one other question regarding the women that you have studied. What data do we have for women, especially African American women, who are increasingly infected with HIV?

Post, Baltimore: Thank you for that question. Dr. Del Rio spoke about the HIV epidemic, which has now become an epidemic primarily in people of color and socially disadvantaged people in the United States. The epidemiology is very different for the reasons that I mentioned, because social determinants of health are such important predictors of adverse outcomes. Both the HIV effect and social determinants of health potentially lead to poor outcomes. Many medical people are trying to make sure that affected people have easy access to care and are treated with effective antiretroviral therapy.

Wilson, Durham: Are your studies including these younger African American -women?

Post, Baltimore: The WIHS study includes these women. I started with the MACS, but now our studies are combined. My future studies and the MRI studies will include both men and women.

Wilson, Durham: Okay, great. Thank you.

Post, Baltimore: Yes, thank you very much.