Risk assessment and targeted risk factor modification are critical to the disease prevention paradigm in family medicine. Because coronary artery disease (CAD) represents the leading cause of death in the United States, prevention efforts have tremendous potential to reduce disease morbidity and mortality on a population scale.
In this issue of AFP, Mieres and colleagues1 discuss the potential role of noninvasive cardiac imaging technologies in estimating a patient's risk of a future myocardial infarction. In symptomatic patients, single-photon emission computed tomography myocardial perfusion scans and stress tests have a fairly high sensitivity and specificity and have proven to be useful in clinical decision making. In asymptomatic patients, however, the role of these tests is much more complicated. Currently, there is no evidence that electrocardiography, stress echocardiography, or cardiac computed tomography (CT) improves health outcomes in low-, intermediate-, or high-risk persons; application of such screening to the general, low-risk population may be harmful.2
The Framingham risk calculator is useful for stratifying patients into low-, intermediate-, and high-risk categories based on age, sex, lipid levels, presence of hypertension, and smoking status.
Unfortunately, a large group of asymptomatic patients falls into the intermediate-risk category (10 to 20 percent 10-year risk of CAD). It would be useful if these patients could be “bumped down” into the low-risk category to be reassured, or “bumped up” into the high-risk category to be prescribed more aggressive medical management (e.g., treatment to reduce a low-density lipoprotein or blood pressure goal, chemoprophylactic aspirin administration) or possibly have invasive interventions (e.g., coronary catheterization, bypass) if such interventions were to prove beneficial. For this to happen, there needs to be either a more refined understanding of the meaning of the intermediate-risk category or new methods of risk assessment, beyond what is already included in the Framingham calculator, that can more definitively sort persons into the “reassure” or “vigorous risk reduction” treatment categories.
For CAD risk stratification or a screening test to aid in the evaluation of asymptomatic, intermediate-risk patients, it must answer the following three questions. Does the test independently predict coronary events beyond Framingham risk scoring alone? Does a positive test occur often enough in those at highest risk of CAD to make mass application worth it (i.e., sufficient yield)? Do patients identified as high risk by the test stand to benefit from therapy? If the answer to any of these questions is uncertain, there is no way to know if applying such a test to large populations of asymptomatic patients would be beneficial. Even with adequate data to answer these questions, the test actually could be harmful by causing false-positive results, providing false reassurance, or by labeling persons with a disease when interventions are not effective.
Cardiac CT (or electron beam CT) is one of the most promising noninvasive technologies for screening asymptomatic, intermediate-risk patients. Coronary calcium scores are independent predictors of CAD (especially at values over 400), reliability is good, and population-based studies have shown that high scores are seen in a large proportion of intermediate-risk patients.3–6 However, it remains uncertain whether changes in the treatment of patients restratified as high risk based on coronary calcium scores improve clinical outcomes. For example, statins have not been shown to decrease mortality in patients screened with cardiac CT, and there is even conflicting evidence about whether they produce the intermediate outcome of reduction in coronary calcium scores.7,8
In October 2006, the American Heart Association (AHA) released a scientific statement encouraging Framingham risk assessment in asymptomatic persons and advising against the use of coronary calcium assessment in asymptomatic persons at low risk (less than 10 percent 10-year risk) and high risk (over 20 percent 10-year risk). It also states that in clinically selected intermediate-risk patients, it may be reasonable to measure the atherosclerotic burden using electron beam CT or multidetector CT to refine clinical risk prediction and to select patients for more aggressive target values for lipid-lowering therapies.9 Although some may interpret this to be a green light for cardiac CT in the intermediate-risk group, the recommendation is a class IIb recommendation, which means that there is conflicting evidence or a divergence of opinion about the usefulness or effectiveness of a procedure or treatment where usefulness or effectiveness is less well established by evidence or opinion. The AHA acknowledges in this statement that there are no outcome studies demonstrating that measurement of coronary calcium leads to improved outcomes.9
The U.S. Preventive Services Task Force is currently developing its recommendation statement for the use of nontraditional risk factors in intermediate-risk persons. It will include a review of coronary calcium assessment, as well as other risk factors (e.g., high-sensitivity C-reactive protein, ankle-brachial index, white blood cell count, fasting blood glucose, periodontal disease, carotid intimal thickness, homocysteine, lipoprotein [a]).
The bottom line is that the jury is still out on how noninvasive testing could or should be used to restratify asymptomatic patients at intermediate risk of CAD. Cardiac CT is expensive ($400 to $500 per scan) and thus far has not been proven to affect patient-oriented outcomes. Until further research is available, current evidence indicates that physicians should not order cardiac CT as a screening test in asymptomatic patients.