Cardiovascular disease (CVD) is the leading cause of disease burden in the world, and prevalent cases have nearly doubled since the 1990s.1 A large proportion of CVD is caused by atherosclerosis.1,2 Ischemic heart disease encompasses diseases of the heart due to atherosclerotic coronary artery disease (CAD), while atherosclerotic CVD (ASCVD) is a general term for diseases of the cardiovascular system due to cholesterol plaque buildup. Primary prevention of ASCVD involves treating patients who are at risk of developing, or who have already developed, atherosclerotic disease before they experience a cardiovascular event. Risk stratification is undertaken to identify patients who would benefit from primary prevention, but commonly used methods for risk stratification are imperfect. Traditionally, cardiovascular risk stratification involves a review of the patient’s risk factors, characteristic symptoms (if any), a physical examination, a resting electrocardiogram, and bloodwork.3 The risk of ASCVD can then be estimated using a risk score. The 2021 Canadian Cardiovascular Society (CCS) guidelines4 recommend the use of the Framingham risk score (FRS),5 which aims to predict an end point that includes myocardial infarction (MI), angina, coronary death, stroke, claudication, or congestive heart failure.6 Of note, there are several versions of the FRS that estimate different end points (eg, hard CAD, which includes CAD but not angina5). A patient may have different FRS estimations depending on which end point is used and may falsely be placed in a lower-risk group if the full complement of outcomes is not considered.

Risk scores are not able to account for risk with complete accuracy, since traditional risk factors (smoking, hypertension, diabetes, dyslipidemia, age, and sex) account for only 65% to 85% of cardiovascular events.7,8 When abdominal obesity, psychosocial factors, diet, alcohol consumption, and physical activity are also accounted for, 90% to 94% of cardiovascular events can be explained,9 but these factors are difficult to quantify and are impractical for the purposes of risk scores. Furthermore, cardiovascular risk in women is often underestimated10,11: sex-specific differences in the rates of diagnosis and treatment of traditional risk factors, and sex-specific risk factors such as hypertensive disorders of pregnancy,12,13 are not accounted for by most risk scores. Biomarkers and imaging tests may help close the detection gap by accounting for the risk not explained by traditional risk factors.14

A 60-year-old woman consults her primary care provider after her brother dies from an MI at the age of 57. She is otherwise healthy, active, and asymptomatic. Her blood pressure is 125/82 mm Hg and physical examination findings are unremarkable. Electrocardiogram findings are normal. Bloodwork results are as follows: total cholesterol level of 4.72 mmol/L, high-density lipoprotein cholesterol (HDL-C) level of 1.57 mmol/L, low-density lipoprotein cholesterol (LDL-C) level of 2.89 mmol/L, non–HDL-C level of 3.15 mmol/L, and triglyceride level of 0.57 mmol/L. She is concerned about her risk of MI and wonders whether she can mitigate this risk.

Current guidelines were reviewed and an Ovid MEDLINE literature search was performed.

Novel biomarkers. Numerous biomarkers are associated with the incidence of cardiovascular events, but the usefulness of a biomarker is better measured by the additional information that is gained over traditional risk stratification. We will review several prominent biomarkers, including high-sensitivity C-reactive protein (hs-CRP) and lipoprotein(a) (Lp[a]), which are risk modifiers for ASCVD.3,4 The CCS guidelines endorse that the presence of risk modifiers (family history of premature ASCVD, hs-CRP level ≥2.0 mg/L, Lp[a] level ≥500 mg/L [≥50 mg/dL], or coronary artery calcium [CAC] >0) supports the use of statin therapy in those at intermediate risk (FRS 10% to 19.9%) who do not otherwise meet the recommendations for statin use.4

High-sensitivity troponin (hsTn): The development of hsTn assays, which can detect serum troponin at lower concentrations than older “contemporary” troponin assays, has allowed for accurate measurement of high-sensitivity cardiac troponin I and high-sensitivity cardiac troponin T (hs-cTnT) in seemingly healthy individuals.15,16 Baseline elevations in hsTn are independently predictive of all-cause mortality,17-20 cardiovascular mortality,18-22 and nonfatal MI18,19,20-22 in ambulatory patients without known ASCVD. In one large study, ambulatory patients with detectable hs-cTnT levels had a 10-year risk of ASCVD of 13.2%.23 Conversely, patients with hs-cTnT levels below the limit of detection had low rates of ASCVD.23 Studies have reported improvements in cardiovascular risk stratification with the addition of hsTn to the FRS20,24-27 and to the European Society of Cardiology SCORE (Systematic COronary Risk Evaluation) risk algorithm.28 Although more research is needed to investigate whether hsTn provides unique information that is not otherwise captured by an individual’s traditional risk factors, it remains useful as a risk marker and patients with detectable or elevated hsTn levels can be considered for earlier and more aggressive preventive interventions.29

C-reactive protein: Inflammatory cells play a pathogenic role in the formation of early “fatty streaks,” the progression to advanced atherosclerotic lesions, and the development of unstable plaques.30-33 One of the most ubiquitous markers of inflammation is CRP, an acute-phase reactant that is predominantly made by hepatocytes in response to inflammation, infection, malignancies, or tissue damage.34,35 Whether CRP is solely a marker for vascular disease or plays a pathogenic role in the development of atherosclerosis is an area of debate.36,37 Elevated baseline levels of hs-CRP are independently associated with increased all-cause mortality,38-40 cardiovascular death,39,40 and cardiovascular events.34,38,40,41 However, studies examining reclassification after the addition of CRP or hs-CRP to predictive models have shown mixed results.24,26,38,41-46 The largest analysis to date, which included 166,596 participants, found a very modest improvement in risk stratification with the use of CRP.45

Lipoprotein(a): Lipoprotein(a) has substantial atherogenic potential and is an LDL particle with apolipoprotein(a) (apo[a]) covalently bound to the apo B100 molecule.47,48 Circulating levels of Lp(a) are genetically determined, with little influence of lifestyle factors.48 There is a linear relationship between future cardiovascular risk and the concentration of Lp(a),47,49 and Mendelian randomization studies suggest a causative role of Lp(a) in the development of atherosclerosis.50,51 Research is ongoing to determine whether lowering Lp(a) will lead to reductions in ASCVD outcomes.52,53 The CCS guidelines consider Lp(a) a risk modifier and recommend Lp(a) testing once in a person’s lifetime with earlier and more intensive health behaviour modifications in those with Lp(a) levels 500 mg/L (50 mg/dL) or greater.4

Lipid measurements: Conventionally, total cholesterol, LDL-C, and HDL-C are used as biomarkers of dyslipidemia and included in many CVD risk calculators.5,28,54 However, newer measures to characterize lipid biology exist. Non–HDL-C and apoB have become more prevalent as alternatives to LDL-C measurement.55-59 Measurements of apo AI57 and LDL particle number and size56 are also used. Apolipoprotein AI and apoB are the principal protein components of HDL and non-HDL particles, respectively, and measuring apo AI or apoB is analogous to measuring the lipoprotein particle number.58-60 Because the number of atherogenic non-HDL particles is more strongly associated with ASCVD risk than is the cholesterol content of the particles,56 many researchers recommend the use of apoB over LDL-C or non–HDL-C.61 Canadian guidelines support the use of apoB as an alternative measurement to LDL-C or non–HDL-C.4

Low-density lipoprotein particle number, size, and density can also be measured directly by lipoprotein subfractionation techniques,62 but this is not practical for widespread, routine clinical use. Small dense LDL particles have emerged in population studies as independently associated with CVD risk.63-66 Small dense LDL particles are highly atherogenic owing to an increased propensity to oxidation, high endothelial permeability, and decreased clearance through hepatic LDL receptors.67

Despite the theoretical benefits of newer measures of a patient’s lipid profile, studies examining their use in risk stratification have been disappointing. The hazard ratios for CAD generated with apoB and apo AI are nearly identical to those generated with non–HDL-C and HDL-C, respectively,55 and the addition of apoB, apo AI, LDL, or Lp(a) to models that already included total cholesterol and HDL-C has not led to improvement in risk stratification.60,68

Imaging tests. Carotid intima-media thickness (C-IMT) and CAC scores are imaging tests commonly used in risk stratification. Abnormal C-IMT or CAC scores are correlated with risks of CAD69-71 and ASCVD events.72-75 Both tests have been shown to improve risk stratification73,75,76 but, when compared head to head, the CAC score performed better than C-IMT.75,77

Although C-IMT measurement is popular because of its affordability, availability, and lack of radiation exposure, an absence of standardization and inconsistent definitions of C-IMT have led to varied results in clinical trials.78-80 In a large meta-analysis, common C-IMT was found to improve risk prediction, but reclassification was small and unlikely to be of clinical importance.78

In contrast, although there are several different CAC scoring techniques, all are strongly correlated and have been shown to have excellent interobserver and intraobserver reproducibility.81 Coronary artery calcium measurements are associated with an average radiation dose of 0.89 mSv, which is approximately equal to 3.6 months of background radiation exposure.4 The addition of CAC scores improves risk stratification compared with the use of traditional risk factors alone, especially in middle-aged, intermediate-risk populations.77,82,83 Higher CAC scores are associated with increased risk, with a score of greater than 100 associated with a greater than 2% annual ASCVD risk and a score of greater than 300 associated with a 10-year risk of MI or cardiovascular death of 28%.4 Conversely, patients with a CAC score of 0 have a low risk of 10-year ASCVD,23,84 with a CAC of 0 being associated with the greatest downward shift in estimated risk compared with several other risk indicators.84 However, a CAC score of 0 does not imply a complete lack of risk, as a CAC score measures only late calcified plaque in the coronary arteries and may miss early, noncalcified plaque. Because of this, a patient’s age must be considered when ordering a CAC score. Ordering a CAC score for men older than 42.3 years and women older than 57.6 years who do not have other cardiovascular risk factors provides the greatest clinical usefulness.85 The CCS guidelines considers a CAC score greater than 0 to be a risk modifier and suggests consideration of CAC scoring to detect subclinical atherosclerosis in select individuals.4

When to use biomarkers and imaging tests. Biomarkers and imaging tests are most useful in patients at intermediate cardiovascular risk to help guide treatment recommendations. Since all patients, regardless of predicted cardiovascular risk, should be counseled to pursue health behaviour modifications including improved diet, increased exercise, and smoking cessation,4 the main treatment decision is whether the patient would benefit from statin therapy. A patient with a statin-indicated condition (Table 1)4 or whose FRS is 20% or greater has a strong recommendation to start statin therapy and, in general, no additional information from biomarkers is needed. In contrast, statin therapy is not recommended in most patients at very low (<5%) risk of ASCVD. For a patient with an FRS between 5% and 19.9%, treatment decisions are more nuanced. The CCS guidelines recommend statin therapy in patients at intermediate risk (FRS 10% to 19.9%) when cholesterol levels are elevated (LDL-C level ≥3.5 mmol/L, non–HDL-C level ≥4.2 mmol/L, or apoB level ≥1.05 g/L). In addition, statin therapy should be considered for patients at low risk (FRS 5% to 9.9%) with elevated cholesterol levels (using the same cutoffs), especially if risk modifiers are present. When cholesterol level is not elevated, evidence still favours the use of statins in intermediate-risk patients (FRS 10% to 19.9%) when risk modifiers are present and in men 50 years and older and women 60 years and older with 1 additional risk factor. A summary of these recommendations as they relate to statin therapy is shown in Figure 1.4

Statin-indicated conditions

Summary of statin therapy recommendations