When and How to Screen for Cardiovascular Disease Risk in People With HIV

Table of Contents

• Introduction

• Does Framingham Reliably Frame CVD Risk With HIV?

• Which Patients Need Deeper Probing?

• References and Notes

Introduction

Guidelines for cardiovascular disease (CVD) screening in people with HIV are profuse yet elementary¹⁻⁴ (Table 1). And pediatric solons have devised a transparent algorithm for sizing up cardiovascular risk in children with HIV⁵ (Table 2). But as with any facet of HIV medicine, the deeper you dig, the bigger the hole you can find yourself in. As the first two articles and the interview with James Stein in this issue make clear, cardiovascular risk with HIV infection is fraught with confounders ranging from parental chromosomes to pernicious lifestyle choices. Not to mention antiretroviral picks.

For adults, the HIV Medicine Association (HIVMA) and collaborating groups,¹˒² the European AIDS Clinical Society (EACS),³ and the US Health Resources and Services Administration (HRSA)⁴ all recommend routine assessment of cardiovascular risk factors with a standard heart risk calculator when people enter care, start antiretroviral therapy (ART), switch antiretroviral regimens, and at regular intervals depending on calculated risk (Table 1). The EACS takes a more aggressive stance than the US groups, also calling for yearly electrocardiograms (ECGs) in men over 40 and women over 50.³ Lipids, blood pressure, and glucose also need regular checking (Table 1).

In 2011 an expert panel assembled by the National Heart, Lung, and Blood Institute (NHLBI) issued "Integrated guidelines for cardiovascular health and risk reduction in children and adolescents," which includes some HIV-specific screening advice (Table 2).⁵ The panel figures that HIV itself poses a moderate risk of accelerated atherosclerosis in children. Indeed, the Bogalusa Heart Study famously found that atherosclerotic changes can begin brewing in childhood.⁶ The NHLBI recommends assessing heart risk factors in children with HIV and -- in those with two or more risk factors -- taking steps to control weight, blood pressure, lipids, and glucose.

Does Framingham Reliably Frame CVD Risk With HIV?[](#ast)

Is the Framingham score an accurate cardiovascular risk predictor for people with HIV? Studies that address this question agree that the Framingham risk-reckoner does a decent job in HIV-positive people but may underestimate 10-year risk in some subgroups and overestimate risk in others -- depending on risk factors and geographic origin. And no wonder. As the first two articles in this issue of RITA! make clear, people with HIV tote an ample burden of classic heart risk factors. On top of that they have a chronic inflammatory infection and may take cardiotoxic drugs to treat it. So compared with the general population, HIV-positive people face a more ramified array of risk mediators that may sway the prediction one way or the other.

The Framingham tool figures 10-year risk of "hard coronary artery disease," meaning myocardial infarction or coronary death, in adults without coronary heart disease, diabetes, or intermittent claudication (leg or arm pain caused by inadequate blood flow).⁷ It does so by assigning points for six variables: age, total cholesterol, HDL cholesterol, systolic blood pressure, anti-hypertensive therapy, and smoking status for men or women (Figure 1). A 10-year risk score below 10% indicates low risk, 10% to 20% signals intermediate risk, and 21% or higher signifies high risk. These 10-increment cut points are arbitrary.

An HIV-specific cardiovascular risk calculator engineered by the Data Collection on Adverse Events of Anti-HIV Drugs (DAD) Study leaves out anti-hypertensive therapy but adds five others factors to the equation -- number of years taking indinavir or lopinavir; current treatment with indinavir, lopinavir, or abacavir; previous smoking; diabetes; and family history of cardiovascular disease (Figure 1).⁸ Unlike Framingham, the DAD tool predicts 5-year risk of coronary heart disease (in people without a heart disease history). The DAD team considers a 5-year CHD risk below 1% low, 1% to 5% moderate, 5% to 10% high, and above 10% very high. The Framingham and DAD calculators are online at the links in references 7 and 8.

DAD researchers stress the big differences between their study group and the Framingham cohort.⁸ The HIV-negative US-based Framingham population included 2590 men and 2983 women from 30 to 74 years old followed for 12 years from a baseline date of 1968 to 1975. The DAD cohort was bigger (16,765 men and 5860 women), younger (median 40 years), and mostly European, and follow-up was shorter (4.8 years starting in the year 2000) than in Framingham. Everyone in DAD had HIV infection and most were taking antiretrovirals. The DAD investigators cite previous research indicating that the Framingham equation overpredicts cardiovascular disease risk in European populations. They note that the limited number of cardiovascular endpoints in DAD women prevented them from developing sex-specific prediction models.

How can these differences in variables and study populations affect risk calculations? Say your patient is a 52-year-old male smoker with a total cholesterol of 237, an HDL of 42, a systolic blood pressure of 138, and naive to anti-hypertensives. Feed those numbers into the Framingham brain and you come away with a 10-year risk of 20%, right at the cusp of high risk. That 20% means 20 of 100 people with this risk will have a heart attack in the next 10 years. Not happy odds.

Now you turn to the DAD decoder and add that this same patient took lopinavir for 3 years and is still taking lopinavir with abacavir. He's a current smoker, does not have diabetes, but has a family history of heart disease. Click the CALCULATE button. In the next 5 years, this man runs a 14.1% risk of coronary heart disease. But if he stopped lopinavir and is not on abacavir, this 5-year risk shrinks to 7.1%. If he quit smoking, the 5-year risk dwindles to 4.2%. If you take smoking out of the Framingham calculation detailed above, the 10-year risk falls to 9%. Thus, depending on the patient subgroup involved, the DAD tool would predict a greater or lesser coronary heart disease risk than Framingham.

In the study in which DAD investigators developed their risk calculator, it proved more accurate than the Framingham equation in estimating cardiovascular disease risk in the overall (largely European) cohort and in certain subgroups.⁸ In the DAD cohort, the Framingham model underpredicted risk compared with the DAD model when forecasting myocardial infarction or coronary heart disease in women, former smokers, and people with diabetes. On the other hand, Framingham overpredicted risk in people who never smoked.

"Although pending external validation," DAD collaborators note, "our models are intended for clinical usage to inform doctor-patient discussions on CVD risks and interventions,"⁸ as well as for research applications.

Before DAD investigators devised their risk calculator, they analyzed the accuracy of the Framingham formula in predicting myocardial infarction.⁹ Among European, US, and Australian DAD participants taking antiretrovirals, Framingham underpredicted actual MI incidence (9 observed MIs versus 5.5 predicted). But among antiretroviral-naive people, Framingham overpredicted actual incidence (3 observed MIs versus 7.6 predicted).

In an interview in this issue, cardiologist James Stein counsels that HIV clinicians should consider the Framingham model "as a starting point for discussion," recognizing that it's not perfect in people with HIV or people who differ from the young to middle-aged white US population in the Framingham cohort.

Which Patients Need Deeper Probing?

When should you refer a patient with HIV for further cardiovascular workup? HIV/heart experts advise first figuring the pretest probability that a person has CHD.¹⁰ They suggest several tools for doing this (like the one in Table 3), cautioning that these formulas remain unvalidated in people with HIV.

In the Table 3 model, a score of 0 to 8 indicates low risk, 9 to 15 indicates intermediate risk, and 16 or higher signals high risk. People in the intermediate-risk stratum are the best candidates for a noninvasive stress test such as an exercise ECG, the HIV experts advise.¹⁰ People in the high-risk group often get false-negative results and thus are not great candidates for a noninvasive stress test. Instead, they should be referred for invasive coronary arteriography. People with a low-pretest probability of CHD tend to have false-positive test results, so they are not ideal candidates for noninvasive stress testing. Instead, they may be candidates for a stress test with nuclear perfusion imaging or wall motion imaging with echocardiography, but only if they have an intermediate global CHD risk or have a high-risk job, like flying airplanes.

This thoughtful review probes the ins and outs of noninvasive testing and cardiovascular markers, including high-sensitivity C-reactive protein (hsCRP), apolipoprotein (apo)B and apoA-1, carotid intimamedia thickness (cIMT), and coronary calcium scores.¹⁰ Although these markers see routine use in cohort studies and trials, their value in individual patients remains uncertain.

For the general population at least, hsCRP may have clinical value, according to a Centers for Disease Control and Prevention (CDC) panel -- but not for everyone who walks through the door.¹¹ Instead, these experts suggest that hsCRP in people with a 10% to 20% CVD risk over 10 years may pick out those who would benefit from medical intervention, for example, with anti-lipid agents, anti-platelets, or cardioprotective drugs. (See note 11 for details of this panel's advice on clinical use of hsCRP in the general population.) But because hsCRP reflects inflammation, and because HIV-positive people have so many potential inflammation inciters (including HIV itself), the HIV/heart panel says "the role of hsCRP in clinical [HIV] practice is less clear" and suggests the need for hsCRP studies that control for traditional risk factors.¹⁰ One study published after this panel wrote did control for traditional risk factors when measuring the impact of CRP and HIV -- independently and together -- on acute myocardial infarction risk.¹² In this analysis of 487 people with HIV and 69,870 HIV-negative people seen from January 1997 through December 2006, people with HIV and high CRP had 4-fold higher odds of an acute MI than HIV-negative people with normal CRP. (See Figure 13 in the first review article in this issue of RITA!)

In a 2010 issue of RITA!, HIV metabolics maven Steven Grinspoon, who headed this study in Boston's Partners HealthCare System, addressed the question of CRP use in the HIV clinic.¹³ "If you have a patient with a short duration of HIV and no other risk factors," he suggested, "measuring CRP early in the process is probably not useful. At the other end of the spectrum, in someone with severe dyslipidemia and diabetes, the additive value of CRP is probably irrelevant. But perhaps in borderline patients who have had chronic HIV for a long time and borderline dyslipidemia, much like the patients in this study,¹² measuring CRP may be useful because having HIV and a high CRP would raise the MI risk 4-fold. I couldn't say specifically whether the predictive value of CRP in such patients is entirely independent of other markers. Our study would suggest this is the case, but further studies need to be done."¹³

Should HIV-positive men over 40 and women over 50 have an annual ECG, as EACS guidelines recommend?³ A 4518-person SMART study analysis lends credence to this advice,¹⁴ but some authorities voice reservations about routine ECGs. One in two people in the SMART analysis had a minor ECG abnormality, and 1 in 13 had a major abnormality. During a median follow-up of 28.7 months, 155 people (3.4%) got diagnosed with cardiovascular disease. A statistical model adjusted for study arm, demographics, cardiovascular risk factors, and HIV variables figured that a major ECG abnormality almost doubled the risk of a new cardiovascular diagnosis (hazard ratio 1.83, 95% confidence interval 1.12 to 2.97, P < 0.015), but major and minor abnormalities combined did not. The SMART investigators believe these findings "suggest that the ECG could provide a convenient risk-screening tool in HIV-infected patients."¹⁴

But in an interview in this issue, HIV/heart expert James Stein explains that screening ECGs are not recommended in the United States because they lack sensitivity in identifying cardiovascular disease and because they may yield false-positive results. ECGs, Stein believes, should be reserved for people with heart disease symptoms, like shortness of breath and chest discomfort.

* Thanks to Nina Friis-Møller, MD, PhD, DMSc, Copenhagen HIV Programme, University of Copenhagen, for reviewing this section of this article.

References and Notes

1. Dubé MP, Stein JH, Aberg JA, et al. Guidelines for the evaluation and management of dyslipidemia in human immunodeficiency virus (HIV)-infected adults receiving antiretroviral therapy: recommendations of the HIV Medicine Association of the Infectious Disease Society of America and the Adult AIDS Clinical Trials Group. Clin Infect Dis. 2003;37: 613-627.

2. Aberg JA, Kaplan JE, Libman H, et al. Primary care guidelines for the management of persons infected with human immunodeficiency virus: 2009 update by the HIV Medicine Association of the Infectious Diseases Society of America. Clin Infect Dis. 2009;49:651-681.

3. Lundgren JD, Battegay M, Behrens G, et al. European AIDS Clinical Society (EACS) guidelines on the prevention and management of metabolic diseases in HIV. HIV Med. 2008;9:72-81.

4. US Department of Health and Human Services Health Resources and Services Administration HIV/AIDS Bureau. Guide for HIV/AIDS clinical care. January 2011.

5. Expert Panel on Integrated Guidelines for Cardiovascular Health and Risk Reduction in Children and Adolescents. Integrated guidelines for cardiovascular health and risk reduction in children and adolescents. The report of the expert panel. 2011.

6. Berenson GS, Srinivasan SR, Bao W, et al. Association between multiple cardiovascular risk factors and atherosclerosis in children and young adults. The Bogalusa Heart Study. N Engl J Med. 1998;338:1650-1656.

7. Framingham Heart Study. Hard coronary heart disease (10-year risk).

8. Friis-Møller N, Thiébaut R, Reiss P, et al; for the DAD study group: Predicting the risk of cardiovascular disease in HIV-infected patients: the Data Collection on Adverse Effects of Anti-HIV Drugs Study. Eur J Cardiovasc Prev Rehabil. 2010;17:491-501.

9. Law MG, Friis-Møller N, El-Sadr WM, et al. The use of the Framingham equation to predict myocardial infarctions in HIV-infected patients: comparison with observed events in the D:A:D Study. HIV Med. 2006;7:218-230.

10. Hsue PY, Squires K, Bolger AF, et al. Screening and assessment of coronary heart disease in HIV-infected patients. Circulation. 2008;118:e41-e47.

11. Pearson TA, Mensah GA, Alexander RW, et al. Markers of inflammation and cardiovascular disease: application to clinical and public health practice: a statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation. 2003;107:499-511. This panel makes the following points about clinical use of hsCRP in the general population:

12. Of current inflammatory markers identified, hsCRP has the analyte and assay characteristics most conducive to use in practice. (Class IIa, Level of Evidence B)

13. Measurement of hsCRP is an independent marker of risk and, in those judged at intermediate risk by global risk assessment (10 to 20% risk of CHD per 10 years), at the discretion of the physician, may help direct further evaluation and therapy in the primary prevention of CVD. The benefits of such therapy based on this strategy remain uncertain. (Class IIa, Level of Evidence B)

14. Patients with persistently unexplained, marked elevation of hsCRP (>10mg/L) after repeated testing should be evaluated for noncardiovascular etiologies. (Class IIa, Level of Evidence B)

15. In patients with stable coronary disease or acute coronary syndromes, hsCRP measurement may be useful as an independent marker of prognosis for recurrent events, including death, MI, and restenosis after PCI. The benefits of therapy based on this strategy remain uncertain. (Class IIa, Level of Evidence B)

16. Measurement of markers should be done twice (averaging results), optimally two weeks apart, fasting or nonfasting in metabolically stable patients. If hsCRP level is >10 mg/L, test should be repeated and patient examined for sources of infection or inflammation. (Class IIa, Level of Evidence B)

17. hsCRP levels, using standardized assays, categorize patients as follows:

1. Triant VA, Meigs JB, Grinspoon SK. Association of C-reactive protein and HIV infection with acute myocardial infarction. J Acquir Immune Defic Syndr. 2009;51:268-273.

2. Hidden heart disease and other non-AIDS risks in HIV patients. An interview with Steven Grinspoon, MD. Research Initiative, Treatment Action! 2010;15:5-12. http://centerforaids.org/pdfs/mar2010ritagm.pdf

3. Soliman EZ, Prineas RJ, Roediger MP, et al. Prevalence and prognostic significance of ECG abnormalities in HIV-infected patients: results from the Strategies for Management of Antiretroviral Therapy study. J Electrocardiol. 2011;44:779-785.