In Patients Without Cardiovascular Disease or Diabetes, Pooled Cohort Equations May Underestimate Risk of Cardiovascular Event if Other Risk Factors are Present
EBM Focus - Volume 10, Issue 38
Accurate prediction of cardiovascular disease risk is crucial for the success of preventative measures. The American College of Cardiology/American Heart Association (ACC/AHA) 2013 guidelines recommend using race and sex specific Pooled Cohort Equations (PCE) to determine the risk of atherosclerotic cardiovascular disease in patients aged 40-79 years (J Am Coll Cardiol 2014 Jul 1;63(25 Pt B):2935). High or moderate intensity statins are recommended in patients with a 10-year risk of atherosclerotic cardiovascular disease ≥ 7.5% on the PCE (J Am Coll Cardiol 2014 Jul 1;63(25 Pt B):2889). ACC/AHA guidelines also suggest considering other risk factor PCE risk < 7.5%, but this recommendation is limited by a lack of empirical evidence. A recent cohort study evaluated 5 additional risk factors for cardiovascular disease in 5,185 adults without baseline cardiovascular disease, statin use, or ankle brachial index ≥ 1.4 from the Multi-Ethnic Study of Atherosclerosis (MESA) to determine how statin eligibility by the PCE corresponded to observed cardiovascular events.
Atherosclerotic cardiovascular events, defined as a myocardial infarction, coronary heart disease death, or fatal/nonfatal stroke, occurred in 6.2% of adults during the median 10.2 year follow-up. The event rate was 13.8% in the 1,000 adults with ≥ 7.5% risk and 4.7% in 4,185 adults with < 7.5% PCE risk. The 5 additional risk factors studied were coronary artery calcium ≥ 300 Agatston units or ≥ 75th percentile for age, sex, and ethnicity; fatal or nonfatal myocardial infarction or stroke in first-degree relative (family history); high-sensitivity C-reactive protein (hs-CRP) ≥ 2 mg/dL; ankle brachial index < 0.9; and low-density lipoprotein (LDL) cholesterol ≥ 160 mg/dL (4.1 mmol/L), but < 190 mg/dL (4.9 mmol/L). These factors were assessed in the 3,882 adults without diabetes and with an initial PCE risk < 7.5% to reassess the degree of risk for purposes of statin initiation. Overall, 11.1% were reclassified using at least 1 additional factor to a ≥ 7.5% 10-year risk of atherosclerotic cardiovascular events and would be indicated for initiation of statin therapy by the 2013 ACC/AHA guideline. Adults reclassified using coronary artery calcium, family history, or hs-CRP had an observed 10-year cardiovascular event rates > 10%, and were therefore at a risk greater than the 7.5% threshold.
Three of the 5 additional risk factors examined identified patients originally classified as low risk who actually had a higher risk of cardiovascular events. The contributions of each risk factor were evaluated individually to determine the incremental value of each test, however the extent of overlap was not presented. The results of this study suggest that these equations may underestimate the risk of atherosclerotic cardiovascular events in some patients without diabetes who are classified as low risk. These results underscore the fact that the PCE is not perfect, but it is not to say the PCE does not have value for risk assessment purposes. Rather, the PCE should not be used blindly. The decision to use statin therapy requires individualized decision making including additional factors influencing risk assessment. Further validation in an independent population is needed to determine if these factors should be incorporated into future versions of the PCE.