 |
|
|
| |
CROI 2013:
Cardiovascular Disease and Co-morbidities in HIV
|
| |
| |
David H. Shepp, MD
Associate Professor of Medicine, Hofstra North Shore-LIJ School of Medicine
North Shore University Hospital - Manhasset, NY
Combination antiretroviral therapy (ART) dramatically reduces AIDS-related illnesses and extends life expectancy in people with HIV infection. As a result, non-AIDS illnesses such as cardiovascular disease (CVD), non-AIDS cancer, and liver disease, now account for the majority of deaths. Smoking, obesity, diabetes, dyslipidemia and chronic inflammation all contribute to the pathogenesis of CVD in both HIV-infected and uninfected populations, but the contribution of each may vary between these two groups. The toxicities of ART also contribute to risk in HIV-infected individuals.
Many studies now suggest HIV infection confers an increased risk of CVD and certain other co-morbidities, even after accounting for traditional risk factors. This results in higher disease rates and possibly earlier onset. However, controversy exists as to whether or not HIV causes accelerated aging. New data from the recently concluded CROI helps to clarify the risk of CVD and other co-morbidities, their causes and treatment in HIV-infected patients.
Risk of CVD in HIV-positive vs. HIV-negative Individuals and Effects of Aging. Three studies reported on CVD rates and effects of aging in HIV-positive and HIV-negative populations. Althoff, et al [1] analyzed the Veterans Aging Cohort Study (VACS) database to determine incidence and age of onset for three important comorbid illnesses in HIV-positive and matched HIV-negative veterans. Compared to the HIV-negative group, HIV-infected individuals had an 81%, 43% and 84% increase in adjusted incidence of myocardial infarction (MI), end-stage renal disease (ESRD) and AIDS-related (not AIDS-defining) cancers, defined here as lung, liver, anal and oropharyngeal cancer and Hodgkins lymphoma. Incidence of other non-AIDS cancers was not increased. For MI and ESRD, age of diagnosis did not differ between the HIV-positive and -negative groups, but there was a slightly earlier mean age of diagnosis (7 months) for both types of cancer. The strengths of this study are its very large size (n >100,000), adjustment for most of the common risk factors and also for the very different age distributions in the two populations. The study provides strong support for the hypothesis that HIV-infection itself is a risk factor for these co-morbid illnesses, especially since the comparator group, HIV-negative veterans in care, is likely a high-risk population compared to the general population. However, it provides little evidence for the concept of premature aging.
Investigators from the DAD study group took an indirect approach to examining this problem, by modeling the increase in CVD risk from age 40 to 65 derived from the DAD CVD risk prediction equation, which includes HIV-related factors [2]. This was compared to the modeled risk conferred by 3 equations derived from the general population, including the Framingham risk equation and two others derived from Italian and Scottish patients. For the endpoint of coronary heart disease, which included MI, invasive procedures and sudden death, and the endpoint of cardiovascular disease events, which also included stroke, the risk conferred for the HIV equation was greater and appeared to increase somewhat more with age, although statistical comparisons were not possible and no precise quantifications were given. For the end-point of MI only, there were no differences, but comparison could be made only to the Framingham equation for this parameter. This study also supports increased risk of CVD in HIV and suggests there is a modest acceleration of aging, at least as measured by CVD events.
A third study used a computer model of HIV prevention cost effectiveness that includes data on co-morbidities and mortality to model rates of CVD in HIV and predict CVD risk at different ages compared to data from the general population and HIV-negative populations at high-risk for HIV [3]. This study projected higher rates of CVD between age 40-60 in the HIV-infected population but above age 60, rates were lower due to competing risk, meaning deaths due to other causes became more prominent in HIV-positives. However, this study built into their model the presumption that premature aging in HIV-positives shifted CVD risk by ten years. As a result, this study cannot be taken as new evidence supporting premature aging in HIV.
Pathogenesis and Risk Factors for Co-morbidities. Coronary heart disease in HIV-infected individuals may have more high-risk anatomic features and a different pathogenesis, as suggested in an abstract presented by Zanni et al [4]. In the general population, several angiographic features have been identified as correlates of increased risk of plaque rupture, leading to acute coronary syndrome, MI and sudden coronary death. Eccentric remodeling of the artery wall, low-attenuation plaque (LAP) and spotty microcalcification identify vulnerable plaque that is at increased risk for rupture. Using CT angiography, these investigators looked for vulnerable plaque features in 106 HIV-positive men, predominantly on ART, and 41 matched HIV-negative controls. None had known coronary heart disease. They found vulnerable plaque features (LAP and eccentric remodeling, but not microcalcification) were more common in HIV-positives. Vulnerable plaque with all 3 features were seen in 8% of HIV-positives and no controls. In univariate analysis, the presence of LAP correlated with both traditional CVD risk factors and inflammatory markers, but only plasma levels of soluble CD163, a marker of monocyte activation, remained independently associated with LAP in multivariate analysis. Similar data was described in women in a late-breaker poster [5]. This study suggests that CVD in HIV-positive individuals may be a more aggressive disease with more features that lead to serious events, and that systemic inflammation, particularly involving innate immunity and the monocyte/macrophage axis is central to the pathogenesis.
Less exotic data with more tangible clinical implications came from three posters describing trends in conventional risk factors for CVD and other co-morbidities. Many studies have shown smoking rates are substantially higher among HIV-infected compared to uninfected individuals in the US, but estimates of smoking prevalence have varied. CDC investigators analyzed data from the Medical Management Project in 2009 and found current smoking in 42% of HIV-infected individuals, a rate about twice that in the general population [6]. An additional 20% were former smokers. Smoking rates varied across demographic groups but were always substantially greater in HIV-infected individuals compare to the general population.
Using the same database, CDC investigators also reported on obesity [7]. Data for the general population came from the NHANES survey. Compared to the general population, the age adjusted prevalence of obesity (BMI >30 kg/m2) in HIV was lower overall (35.7% v. 22.8%), and in men (36% v.17%), but higher in women (36% v. 40%).
Weight gain is commonly seen after initiation of ART. Even though obesity is not more prevalent in HIV-infected men than in the general population, weight gain may carry more ominous health implications. Herrin et al. [8] analyzed the large VACS database to ascertain the effect of weight gain on risk of incident diabetes. At baseline they found only 10% of HIV-positives in the VACS database were obese compared to 36% in HIV-negatives. After one year, the HIV-positive group was more likely to experience a weight gain of >5 lbs. In multivariable analysis adjusted for baseline characteristics, every 5 lbs of weight gain conferred a 10% increase in risk of incident diabetes after 5 years of follow-up, compared with 6% in the HIV-negative population.
Cigarette smoking and obesity are strong risk factors for the co-morbidities that are now common in HIV-infected individuals, including CVD, diabetes, hypertension, and cancer. Obesity is prevalent in HIV, especially in women and the weight gain associated with ART appears to carry an amplified risk of diabetes. Because multiple other risk factors for these conditions are often present, strong efforts to achieve ideal body weight and smoking cessation are needed in management of HIV-infected individuals.
Heart Failure. HIV, hypertension, diabetes, alcohol abuse, and certain antiretroviral agents all are associated with increased risk of cardiac dysfunction. As patients with HIV-infection live longer with these conditions, heart failure (HF) may become increasingly common among cardiac co-morbidities. To examine the prevalence of HF among HIV-infected veterans, Freiberg et al.[9] used the VACS database to study HIV-positive and HIV-negative participants free of CVD at baseline. New onset of HF was recognized by ICD-9 codes and HF was categorized by echocardiogram results. After adjusting for multiple potential confounders, HIV-positive veterans were 60% more likely to develop HF than HIV-negatives. Both HF with preserved and reduced ejection fraction were more frequent. HF diagnosed before age 50 was 3 times as likely among HIV-infected individuals.
Treatment and prevention of CVD. Because CVD risk is increased in HIV, effective prevention methods are needed. Both aspirin and statins are effective for primary and secondary CVD prevention in the general population, but their effectiveness in HIV is not certain. Statins are an attractive option in HIV because, in addition to cholesterol-lowering properties, they have anti-inflammatory effects that may further reduce CVD and all cause mortality. A previously published cohort study has identified a significant survival benefit for HIV-infected statin users who were virologically controlled on ART. Two abstracts further examined the benefits of statin use in HIV. Using a national registry, Rasmussen et al. examined all-cause mortality among all HIV-positive individuals in Denmark, comparing statin users and non-statin users [10]. In a multivariable model adjusted for multiple HIV-related parameters know to affect survival, the presence of certain co-morbidities and censored for failure of viral suppression, these investigators found a trend toward reduced mortality that did not achieve statistical significance. Due to size, this study may have lacked statistical power to detect an effect. A much larger study from the VACS found, surprisingly, that statin use was significantly associated with a reduced risk of cancer, but not CVD or all-cause mortality [11]. An analysis limited to users of only the potent statins atorvastatin and rosuvastatin suggested greater reductions in cancer, CVD and mortality, but again achieved statistical significance only for cancer.
Statins are generally well tolerated, but typical side effects like myalgia and myopathy limit use in some patients. More recently, statin use has been associated with an increased risk of new onset diabetes, although this risk does not reverse the overall CVD benefit. Two cohort studies examining statin use as a risk factor for new onset of diabetes in HIV-infected populations reported opposite results. An Italian study found statin use was strongly associated with a decreased risk of diabetes [12], while a US analysis done through the HIV Out-patient study (HOPS) found statin use was associated with about a 10% increase in risk of diabetes per year of exposure [13]. These studies differ in the characteristics of their study populations, definitions used and statistical methods, and like all cohort study have the potential for a variety of confounding biases. Because of the conflicting results, the question of diabetes risk in HIV-infected statin users remain unanswered.
Little is know about aspirin use or its efficacy in HIV-infected individuals. Suchindran et al. investigated non-episodic aspirin use in a large database of HIV-positive and HIV-negative individuals receiving care from a large academic health system in Boston [14]. Aspirin was utilized less in HIV-positive men, but not women. The difference became greater among those with two or more CVD risk factors, where aspirin was used about twice as often in HIV-negative men and women. After adjustment for conventional risk factors, MI rates were significantly reduced among HIV-negative aspirin users, but there did not appear to be any beneficial effect among HIV-positive men or women at low or high risk for CVD. The study, like other cohort studies, is limited by possible confounding by unmeasured risk factors for CVD and also confounding by indication for aspirin use, but it is provocative and shows it cannot be assumed that inventions that work in the general population will be effective in HIV. CVD may behave differently in HIV-infected populations, due in part to different pathogenesis. Randomized trials of aspirin therapy in HIV, or greater inclusion of HIV-infected patients in CVD prevention trials in the general population are needed.
The improvement in survival after MI in the general population over the past few decades has been attributed to more widely available and more aggressive use of invasive procedures and secondary prevention measures. Only more recently has it been appreciated that HIV-infected individuals have an increased rate of MI, and with ART treatment, a life expectancy that warrants aggressive interventions for CVD. The DAD study group investigated whether improvement in cardiac care have been extended to the HIV population [15]. They examined their database for trends in utilization between 1999-2011. They found the number of MIs declined during the study period, as did associated short-term mortality, from 26 to 8%. They found use of invasive procedures before the diagnosis of MI was low, but was much higher and increased dramatically during the study period after MI. Primary prevention with statins, antihypertensive, ACE inhibitors and anti-platelet agents also was low before MI, increasing modestly over time. Use after MI was higher and showed more increase during the study period. Still, use of these agents for secondary prevention appeared to be low overall with fewer than half receiving each intervention in most time periods. However, there was no data from HIV-negative controls for comparison. In multivariable analysis of risk factors for short-term death after MI, most of the improvement in survival could be accounted for by the increase use of both invasive procedures and secondary prevention.
References
1. Althoff K, C Wyatt C, Gibert C, et al. HIV+ Adults Are at Greater Risk for Myocardial Infarction, Non-AIDS Cancer, and End-stage Renal Disease, but Events Occur at Similar Ages Compared to HIV- Adults. Abstract 59, 20th CROI, Atlanta, GA, March 3-6, 2013;
http://www.retroconference.org/2013b/Abstracts/47433.htm
2. Petoumenos K, El-Sadr W, d'Arminio Monforte A, et al. Increased Risk of Cardiovascular Disease with Age in Men: A Comparison of D:A:D with HIV- Cardiovascular Disease Risk Equations. Abstract 61, 20th CROI, Atlanta, GA, March 3-6, 2013;
http://www.retroconference.org/2013b/Abstracts/46310.htm
3. Losina E, Linas B, Hyle E, et al. Projecting 10-Year, 20-Year, and Lifetime Risks of Cardiovascular Disease in HIV+ Individuals in the US: Competing Risks and Premature Aging. Abstract 747, 20th CROI, Atlanta, GA, March 3-6, 2013;
http://www.retroconference.org/2013b/PDFs/747.pdf
http://www.retroconference.org/2013b/Abstracts/46310.htm
4. Zanni M, Lo J, Wai B, et al. Increased Coronary Atherosclerotic Plaque Vulnerability Features on Computed Tomography Angiography among HIV+ Subjects vs Matched HIV- Controls. Abstract 63, 20th CROI, Atlanta, GA, March 3-6, 2013;
http://www.retroconference.org/2013b/Abstracts/47634.htm
http://www.retroconference.org/2013b/Abstracts/46310.htm
5. Fitch K, Abbara S, T Burdo Y, et al. Non-Calcified Coronary Plaque and Macrophage Activation Markers Are Increased in HIV+ Women. Abstract 185LB, 20th CROI, Atlanta, GA, March 3-6, 2013;;
http://www.retroconference.org/2013b/Abstracts/48048.htm
http://www.retroconference.org/2013b/Abstracts/46310.htm
6. Mdodo R, Frazier E, Mattson C, et al. Cigarette Smoking among HIV+ Adults in Care: Medical Monitoring Project, US, 2009. Abstract 775, 20th CROI, Atlanta, GA, March 3-6, 2013;
http://www.retroconference.org/2013b/Abstracts/45702.htm
http://www.retroconference.org/2013b/Abstracts/46310.htm
7. Thompson-Paul A, Wei S, Mattson C, Skarbinski J. Prevalence of Obesity in a Nationally Representative Sample of HIV+ Adults Receiving Medical Care in the US: Medical Monitoring Project, 2009. Abstract 777, 20th CROI, Atlanta, GA, March 3-6, 2013;
http://www.retroconference.org/2013b/Abstracts/45605.htm
http://www.retroconference.org/2013b/Abstracts/46310.htm
8. Herrin M, Tate J, M Freiberg M, et al. Risk of Incident Diabetes Associated with Weight Gain after cART Initiation. Abstract 804, 20th CROI, Atlanta, GA, March 3-6, 2013;
http://www.retroconference.org/2013b/Abstracts/47714.htm
http://www.retroconference.org/2013b/Abstracts/46310.htm
9. Freiberg M, Chang C-C, Oursler KA, et al. The Risk of and Survival with Preserved vs Reduced Ejection Fraction Heart Failure by HIV Status. Abstract 750, 20th CROI, Atlanta, GA, March 3-6, 2013;
http://www.retroconference.org/2013b/Abstracts/47343.htm
http://www.retroconference.org/2013b/Abstracts/46310.htm
10. Rasmussen L, Kronborg G, Larsen C, et al. Statin Therapy and Mortality in HIV+ Individuals: A Danish Nationwide Population-based Cohort Study. Abstract 764, 20th CROI, Atlanta, GA, March 3-6, 2013;
http://www.retroconference.org/2013b/PDFs/764.pdf
http://www.retroconference.org/2013b/Abstracts/46310.htm
11. Drechsler H, Zhang S, Maalouf N, et al. Impact of Statin Exposure on Mortality and Non-AIDS Complications in HIV Patients on HAART. Abstract 765, 20th CROI, Atlanta, GA, March 3-6, 2013;
http://www.retroconference.org/2013b/PDFs/765.pdf
http://www.retroconference.org/2013b/Abstracts/46310.htm
12. Spagnuolo V, Galli L, Poli A, et al. Association between Statin Use and Type-2 Diabetes Mellitus Occurrence among HIV-1+ Patients Receiving ART. Abstract 766, 20th CROI, Atlanta, GA, March 3-6, 2013;
http://www.retroconference.org/2013b/Abstracts/47110.htm
http://www.retroconference.org/2013b/Abstracts/46310.htm
13. Lichtenstein K, Debes R, Wood K, et al. Statin Use Is Associated with Incident Diabetes Mellitus among Patients in the HIV Outpatient Study. Abstract 767, 20th CROI, Atlanta, GA, March 3-6, 2013;
http://www.retroconference.org/2013b/PDFs/767.pdf
http://www.retroconference.org/2013b/Abstracts/46310.htm
14. Suchindran S, Regan S, Meigs J, et al. Comparison of Aspirin Use and Incident Myocardial Infarction Rates in HIV+ and HIV- Patients in a Large US Healthcare System. Abstract 65, 20th CROI, Atlanta, GA, March 3-6, 2013;
http://www.retroconference.org/2013b/Abstracts/45699.htm
http://www.retroconference.org/2013b/Abstracts/46310.htm
15. Sabin C, Ryom L, Law M, et al. Improvements in Short-term Mortality following Myocardial Infarction: The Data Collection on Adverse events of Anti-HIV Drugs Study. Abstract 748, 20th CROI, Atlanta, GA, March 3-6, 2013;
http://www.retroconference.org/2013b/Abstracts/47305.htm
|
| |
|
|
|
|
|
