Several cardiologists have questioned whether assessment of coronary artery calcium (CAC) should be used to evaluate patients for statin therapy, as currently recommended in guidelines since the National Cholesterol Education Program (NCEP) Adult Treatment Panel III update was published in 2004. Serum biomarkers and imaging tests are used to target higher- and lower-risk patients in whom statins might produce a larger or smaller absolute event reduction.
CAC Is Guideline Recommended
Risk stratification with CAC is more robust than previously published data with C-reactive protein (CRP) or other biomarkers, and is a class IIa recommendation in the 2010 American College of Cardiology Foundation/American Heart Association (ACCF/AHA) Task Force Guideline for Assessment of Cardiovascular Risk in Asymptomatic Adults. These guidelines refer to a noncontrast, limited chest CT scan acquired with an approximate 3- to 5-second breath hold. The 2013 ACC/AHA Guideline on the Assessment of Cardiovascular Risk states that "assessing CAC is likely to be the most useful of the current approaches to improving risk assessment among individuals found to be at intermediate risk after formal risk assessment."
Disease Burden vs Risk Factor
Because CAC measurement allows for integration of the exposure to risk factors over a lifetime, CAC is best considered a measure of disease burden rather than a risk factor. This explains why it is such a powerful predictor of risk and is such a valuable negative predictor when the CAC score is 0. Influential cardiovascular epidemiologist Sir Geoffrey Rose (1926-1993) stated, "The best predictor of a life-threatening illness is the early manifestation of a life-threatening illness." Clearly this applies to CAC scanning, where finding atherosclerosis represents an opportune time to intervene with lifestyle changes, statins, aspirin, and potentially even angiotensin-converting enzyme (ACE) inhibitors. CAC testing has been validated in multiple studies, including Dallas Heart, Rotterdam, St Francis Heart, Multi-Ethnic Study of Atherosclerosis (MESA), and others. CAC has been shown to be the best predictor of future events, in the general population, the elderly, and in persons with diabetes. It provides more robust risk prediction than carotid intima media thickness, CRP, ankle-brachial index, and family history of premature heart disease.
CAC has been shown to better stratify those patients in MESA who would benefit from statins, aspirin, ACE inhibitors, and the polypill versus other biomarkers. All of these studies demonstrate that CAC is the most useful test we have for identifying those with early atherosclerosis, but more important, it may be an even more powerful determinant that patients don't need therapy (Power of Zero). The absence of atherosclerosis in a given patient would potentially obviate the need for anti-atherosclerotic therapies (such as aspirin, statin, and ACE inhibitors). A recent study demonstrated that a CAC score of 0 confers a 15-year warranty period against mortality in individuals at low to intermediate risk, unaffected by age or sex.
Is CAC Cost-Effective?
Many preventive physicians find direct measurement of the disease and integration of cumulative risk factor exposure very appealing. Imagine the cost savings if we were to apply CAC scoring to a study like the JUPITER trial, where statins prevented events but required a high number needed to treat (NNT). Blaha and colleagues observed a population in MESA that met inclusion criteria for JUPITER for a median follow-up of 5.8 years (IQR 5.7–5.9). Almost half of the patients in the MESA JUPITER population had CAC scores of 0, and in this group, rates of coronary heart disease (CHD) events were 0.8 per 1000 person-years (ie, < 0.1%). Nearly three quarters of all coronary events were in the 25% of participants with CAC scores higher than 100 (20.2 per 1000 person-years or 2% per year).
To prevent one CHD event, the predicted 5-year NNT with a statin was 549 for a CAC score of 0, 94 for scores of 1–100, and 24 for scores above 100. By comparison, high-sensitivity CRP was not associated with CHD or cardiovascular events after multivariable adjustment.
Thus, CAC demonstrated that 50% of the population who meet criteria for statins but have no measurable atherosclerosis would likely not benefit. Studies have shown that using CAC testing is more cost-effective[18,19] than the current widespread statin use that is advocated by the ACC/AHA pooled cohorts. Treating all is rarely cost-effective, and patients are less likely to want to stay on therapy if they do not perceive personal benefit. Visualizing coronary calcium is associated with improvements in adherence to statin therapy. Thus, not only can CAC identify who would benefit from anti-atherosclerotic therapies, but it also helps identify those who may not need treatment, and improves compliance when patients are then instructed to take these therapies.
Treating According to Calcium Score
What should you do differently when presented with different scores/findings? It is important to understand that a zero score portends a very low risk, and typically lifestyle changes alone will suffice; the NNT with statins to reduce a heart attack in individuals with zero calcium is > 500. Treating 500 persons to reduce one event is neither practical nor cost-effective. Side effects from statins, such as myalgias,[21,22] occur at a rate greater than 1 per 500, so the risk-benefit ratio of statins in a person with a zero score is either harmful or neutral and is not beneficial. Remember, CAC testing is not generally recommended in persons younger than 40 years. CAC scores of either > 75th percentile for age and gender or ≥ 300 (Agatston units) are deemed high risk per the 2013 cholesterol guidelines and warrant high-dose statins. In the St Francis Heart randomized study, atorvastatin reduced cardiovascular events by 42% in those with scores > 400, with an NNT to reduce one myocardial infarction or death of only 17.
On the basis of current guidelines from both NCEP and ACC/AHA:
CAC scores < 75th percentile and < 300 are to be treated with low- to moderate-dose statins.
CAC scores > 75th percentile or ≥ 300 are to be treated with high-dose statins.
CAC score of zero should be considered for lifestyle modification, unless a compelling indication for statin already exists.
As for aspirin, in the MESA study, individuals with CAC scores ≥ 100 had an estimated net benefit from aspirin regardless of their traditional risk status (NNT = 92; estimated 5-year number needed to harm [NNH] = 442 for a major bleed). Conversely, individuals with a score of zero were unlikely to benefit (5-year NNT = 2036 for individuals with a Framingham risk score [FRS] < 10% and 808 for FRS ≥10%; 5-year NNH = 442 for a major bleed). This was true for men and women and regardless of age. Similar data exist for ACE inhibitors and polypill therapies.
Treatment algorithms should be based on the results of CAC. You wouldn't put someone on chemotherapy if they didn't have cancer; similarly, we shouldn't use statins in people without atherosclerosis.