Primary prevention focuses on individuals with no known ASCVD (ie, without myocardial infarction [MI], angina, stroke, transient ischemic attack, or claudication). These individuals may have subclinical atherosclerotic disease, but at the time of risk assessment it will not have caused cardiovascular symptoms. ASCVD risk assessment attempts to stratify these individuals according to their probability of having an event in subsequent years.

Traditional ASCVD risk factor evaluation (ie, presence of hypertension, hyperlipidemia, diabetes, obesity, smoking status, exercise regimen, etc) is recommended every 4 to 6 years starting at age 20 years, and with this a 30-year risk estimate can be calculated up to age 59. In the United States, routine risk estimation for primary prevention is recommended between ages 40 and 75. In Europe, risk factor screening for ASCVD risk stratification purposes is recommended in men older than 40 years, and in women older than 50 years of age or who are postmenopausal.

Secondary prevention aims to reduce morbidity and mortality in patients with clinically overt ASCVD. The 2019 ESC/EAS guidelines provide validated tools that allow estimating 10-year and lifetime risk in secondary prevention patients.

U.S. and European guidelines recommend using clinical risk scores as the first step in primary prevention care (Algorithms 96.1 and 96.2).⁴,⁵,⁶,⁷,⁸ The current standard is to use tools that predict the 10-year risk of developing an ASCVD event (ie, coronary heart disease [CHD] events or stroke).

The Pooled Cohort Equations (PCE). Since 2013, the ACC/AHA recommend using the PCE to predict 10-year risk of fatal and nonfatal ASCVD events.⁴,⁵,⁹ The PCE were developed using data from five major U.S. cohort studies and include separate equations for non-Hispanic White and African American adults. For other racial/ethnic groups, the equations for non-Hispanic Whites are recommended, although the guidelines stress that these estimates should be interpreted cautiously. Using the PCE and the associated risk categories, individuals are classified as being at low (≥5%), borderline (5% to <7.5%), intermediate (7.5% to <20%), or high (≥20%) risk of a fatal/nonfatal ASCVD event over 10 years.

The Systematic Coronary Risk Evaluation (SCORE) charts. The ESC/EAS recommend using the SCORE charts and the associated online calculator (HeartScore¹⁰) to predict an individual’s 10-year risk of a fatal ASCVD event.⁶ For an approximation of the 10-year risk of fatal and nonfatal ASCVD events, the SCORE risk is multiplied by three. Separate equations are available for men and women as well as for high ASCVD risk (eg, Russia) and low ASCVD risk (eg, Spain) countries; locally recalibrated equations are encouraged when
available. Using SCORE, individuals are classified as being at low (<1%), moderate (1% to <5%), high (5% to<10%), and very high (≥10%) 10-year risk of a fatal ASCVD event.

PCE versus SCORE. Both the PCE and SCORE use similar input information, although the online HeartScore calculator used in the latter includes additional characteristics. Table 96.1 compares the landmark PCE and the SCORE scoring systems in detail.

Other 10-year ASCVD/CHD risk scores. The QRISK-3 was developed in the United Kingdom using database health care information from millions of British men and women, and accounts for race/ethnicity, country of origin, and additional risk factors such as atrial fibrillation (a strong risk factor for stroke) and inflammatory conditions.¹¹ The QRISK-3 may be more appropriate than the SCORE charts in some British populations (eg, in those of South Asian ancestry).

The Multi-Ethnic Study of Atherosclerosis (MESA) CHD Risk Score was developed in 2015 using more contemporary, multiethnic data than that used to derive the PCE. It accounts for family history of CVD, the coronary artery calcium (CAC) score, and also provides estimations for Hispanics and Chinese Americans.¹² The score is available as both a web-based and mobile application-based tool. More recently, the Astronaut Cardiovascular Health and Risk Modification (Astro-CHARM) score, which also incorporates CAC as a predictor, was developed to predict ASCVD events in individuals between 45 and 65 years of age¹³ (Table 96.2). Finally, local validated risk calculators are available in many countries.

Scores for specific subgroups and patient populations. The ESC/EAS guidelines also include guidance for the use of primary prevention risk scoring tools for specific patient subgroups, such as the SCORE “O.P.” in elderly individuals and the ADVANCE risk score and the DIAL model in patients with diabetes.⁶ The U-PREVENT website¹⁴ is a helpful online resource that helps identify the most appropriate score for each specific patient.

Strengths and weaknesses of 10-year ASCVD risk scoring. Convenience is a key advantage of the 10-year risk assessment using clinical risk scores, as most of the risk score calculators can be easily accessed online at no cost. In addition, most of them are parsimonious and use input information that is typically generated during a routine medical visit or that is already recorded in the patient’s medical record. Conversely, 10-year risk scores also have important limitations. Risk discrimination and calibration issues have been noted, especially when scores used data from cohorts recruited 30 to 40 years ago to estimate risk in more contemporary, healthier populations.¹⁵ To ameliorate this, in the 2018 and 2019 ACC/AHA guidelines, the intermediate risk category was defined with a broader range (10-year estimated risk ranging from 7.5% to <20%), than in the prior iterate (5% to <7.5%). This expanded the
population for a more personalized risk assessment.⁴,⁵ Also, most clinical risk scores emphasize chronologic age, which results in most young adults receiving low-risk estimations and most elderly persons high-risk estimations regardless of their actual burden of preventable risk factors. This may result in the underdetection of high-risk young adults and in the overtreatment of healthy elderly individuals. Other key challenges include the difficulty to communicate 10-year probabilities derived from population-level data to individual patients in a meaningful way,¹⁶ as well as the lack of risk information provided by these tools beyond the 10-year time horizon.