Preventing heart disease: Why our guidelines need to be modified

For more than 20 years, primary care physicians have been advised to follow National Heart, Lung, and Blood Institute–sponsored National Cholesterol Education Program guidelines recommending that all adults undergo cholesterol profile screening at least once every five years beginning at 20 years old. This effort has been relatively effective at educating the public, promoting healthy lifestyles and, when indicated, using drug therapy to reduce the risk for MI and stroke.

A major effect of the ATP III guidelines was the introduction of CHD risk equivalents such as peripheral artery disease, diabetes or those with >20% 10-year “hard” CHD risk based on Framingham risk score analysis. This innovation broadened the scope of patients who are eligible for more aggressive primary prevention interventions through lipid lowering.

Gabriel C. Brooks

Michael J. Blaha

Roger S. Blumenthal

However, the fact remains that half of all CV events in the United States occur among otherwise healthy adults who have average or below-average levels of LDL and total cholesterol. This does not mean that cholesterol screening is unimportant; lowering one’s cholesterol remains a crucial aspect of any comprehensive CV risk-reduction program. Rather, this implies that our guidelines need to move beyond traditional risk scoring, particularly when there is strong evidence that advanced subclinical atherosclerosis and inflammation as measured by high-sensitivity CRP identifies individuals at increased risk who many not otherwise be eligible for our highly effective interventions.

CRP levels

A recent systematic review of gene-level associations between variant CRP loci, CRP expression levels and development of CHD was published by Elliot et al. They concluded that a causal relationship of CRP with CHD was unlikely.

HsCRP is a simple, inexpensive blood test used to detect subclinical inflammation. This test has repeatedly been shown to predict increased risk for CVD, even at normal lipid levels. Measurement of hsCRP is currently recommended for adults at “intermediate risk” by the CDC and the American College of Cardiology.

Recent media reports suggesting limited ability of hsCRP to detect risk, however, may have confused clinicians and misinformed the public. We fear that this may cause a regression to the use of traditional risk factors alone to predict CVD risk instead of using assessments of subclinical inflammation or selective use of coronary calcium scoring to better classify the 30% to 50% of intermediate-risk patients in whom hsCRP assessment might directly affect treatment decisions.

The inclusion of hsCRP in global risk assessment has been shown to contribute to prediction of CV events at least as much as HDL and LDL. In considering the vasculature beyond the coronaries, inflammation as measured by hsCRP is implicated in carotid and cortical small vessel disease, as well as the peripheral vasculature.

It is not surprising that hsCRP levels are even more effective in predicting those who may go on to have ischemic strokes than our current measures of dyslipidemia. These two simple and inexpensive blood tests (lipid profile and hsCRP) are not in competition, but rather they are additive. Each test detects patient groups who otherwise might not be evaluated in a prevention-oriented primary care or preventive cardiology practice.

What should one do when a patient’s hsCRP is high? The answer is to counsel these patients to lose weight if they are overweight, to exercise at a brisk pace for 30 minutes or more most days of the week and to increase the consumption of fruits, vegetables and fiber. If the patient’s numbers do not improve satisfactorily during a defined period (eg, six months), treatment of dyslipidemia and/or inflammation via statin therapy and aspirin should be considered.

During the past decade, our understanding of the causes of heart disease has radically changed. No longer do we see CVD as simply a disorder of cholesterol accumulation in the arteries, but rather we understand CVD as having a fundamental inflammatory component. Therapies directly targeting inflammation may represent a key component in our armamentarium against CVD in the future.

JUPITER trial

The JUPITER trial may be prescient in this regard. We now know that treating elevated LDL or those with normal LDL but high hsCRP who are at least 60 years old can dramatically reduce CV event rates in men and women in nearly all ethnic groups.

In this trial of approximately 18,000 adults with normal LDL levels but who were nonetheless at increased CV risk because of an elevated hsCRP, statin therapy safely reduced MI, stroke and venous thromboembolic events by nearly 50% and also reduced all-cause mortality by 20%. This was true even among those who were already thin and who exercised regularly.

In the same trial, the rates of very expensive coronary artery revascularization procedure were also reduced by nearly half. Extrapolating the findings of JUPITER to the U.S. population at large using the National Health and Nutrition Examination Survey, approximately 6.5 million older adults would be candidates to initiate statin therapy.

As we enter an era of renewed scrutiny of health care utility and efficiency in health care spending, the development of preventive medicine tools, such as those that identify and treat vascular inflammation, will be invaluable in avoiding the expense of procedures required to ameliorate the end product of the athero-inflammatory process.

2010 prevention guidelines

These gains, however, cannot be realized until clinicians and their patients are made aware of the implications of the recent clinical trial data. Our upcoming ATP IV guidelines should reflect best clinical practices and best available evidence. In our view, it is already clear that people at intermediate risk, as defined by hard Framingham risk score of at least 10%, already qualify for aspirin statin therapy and with an optional LDL target of <100. We believe that adults with a hard Framingham risk score of 5% to 10% should be candidates for selective screening by hsCRP or coronary calcium scoring if they have multiple components of metabolic syndrome or a family history of premature CVD.

We found it intriguing that recently a few physicians from New England were quoted as saying that hsCRP is rarely, if ever, clinically useful for an individual patient. If certain Boston physicians wish to deny their patients innovations in prevention, we tongue-in-cheek suggest that they may wish to refer their patients across town to Brigham to be assessed for total CVD risk and heart age using the updated global Framingham risk score. Alternatively, they can refer them to Hopkins, where we can refine their CV risk prediction through a simple inexpensive blood test or cardiac CT scan. If they did come to Baltimore, we predict that their chances of finding an open seat for a Major League Baseball game at Camden Yards would be much higher than finding one at Fenway.

Roger S. Blumenthal, MD, is Professor of Medicine and Director of the Johns Hopkins Ciccarone Center for the Prevention of Heart Disease. He is Section Editor of the CHD and Prevention section of the Cardiology Today Editorial Board.

Gabriel C. Brooks, MD and Michael J. Blaha, MD, are from The Johns Hopkins Ciccarone Center for the Prevention of Heart Disease.

For more information:

• Brewer HB Jr. Clin Cardiol. 2003;4(suppl 3):III19-III22.
• Campbell PT. Med Sci Sports Exer. 2009;41:1533-1539.
• Cook NR. Ann Intern Med. 2006;145:21-29.
• Elliott P. JAMA. 2009;302:37-44.
• Everett BM. J Am Coll Cardiol. 2006;48:2235-2242.
• JAMA. 2001;285:2486-2497.
• Folsom AR. Arch Intern Med. 2006;166:1368-1373.
• Hamirani YS. Atherosclerosis. 2008;201:1-7.
• Lloyd-Jones D. Circulation. 2009;119:e21-e181.
• Michos ED. J Am Coll Cardiol. 2009;53:931-935.
• O’Malley PG. Am J Cardiol. 2000;85:945-948.
• Pearson TA. Circulation. 2003;107:499-511.
• Ridker PM. Circulation. 2004;109:2818-2822.
• Ridker PM. Circulation. 2008;118:2243-2251.
• Ridker PM. JAMA. 2007;297:611-619.
• Ridker PM. J Am Coll Cardiol. 2007;49:2129-2138.
• Ridker PM. N Engl J Med. 2001;344:1959-1965.
• Ridker PM. N Engl J Med. 2008;359:2195-2207.
• Rost NS. Stroke. 2001;32:2575-2579.
• Schillinger M. Circulation. 2005;111:2203-2209.
• van der Meer IM. Arch Intern Med. 2003;163:1323-1328.
• Wickramasinghe SR. Am J Cardiol. 2009;103:1174-1177.