Risk stratification in stable coronary artery disease
Abstract
Coronary artery disease is a very common chronic condition affecting large numbers of people across the world. Once detected the natural history of coronary disease is very variable between individuals. Hence, the assessment of symptoms and current burden of disease is a key determinant for the recommendation of drug treatments and other interventions, for example, revascularization, with the aim of improving the quality and quantity of a patient's life. We here review our current approach to assessing the risk of future cardiovascular events in patients presenting with stable coronary artery disease.
Answer questions and earn CME: https://wileyhealthlearning.com/Activity2/5222692/Activity.aspx
Introduction
Coronary heart disease affects more and more patients at younger ages across the globe and it is currently the major leading cause of death in Europe. Each year there are 2 million deaths due to cardiovascular disease in the European Union; the main cause being coronary artery disease, which causes 16% of deaths in men and 15% in women 1.
Stable angina is part of the spectrum of coronary artery disease (CAD), characterised by predictable chest discomfort due to myocardial ischemia, commonly precipitated by exertion, emotional stress, or cold. Stable coronary artery disease applies to patients known to have obstructive or nonobstructive CAD whose symptoms are stable on medical therapy as well as those who are diagnosed with CAD for the first time but judged to have stable angina from the length and stability of the clinical history. Within the continuum of the disease spectrum, there is large variation in the risk of later unstable coronary events such as unstable angina, myocardial infarction (MI), or cardiovascular death.
Identification of a patient's risk of these adverse cardiovascular outcomes is a key aspect of determining management strategy. The ultimate goal is identifying those patients at high risk of adverse outcomes who may benefit from revascularization procedures for more than symptom control. Similarly, the identification of a low-risk population as well as providing reassurance for patients may avoid unnecessary invasive tests and procedures, which potentially could be harmful,.
The 2013 ESC Guidelines on the management of stable coronary artery disease (SCAD) 2 define high event risk patients as those with an annual mortality >3%; this threshold is based on data from functional tests for ischemia and coronary anatomy-based outcomes beyond which prognostic benefit from revascularization has been suggested. Low-risk patients are those with an annual mortality <1%, with intermediate risk defined as ≥1% but ≤3%/year.
Risk stratification in individuals with coronary artery disease may be considered as a sequential, hierarchical approach:
- Clinical evaluation—patient demographics, risk factors, comorbidities, resting ECG, and potential use of biomarkers;
- Assessment of left ventricular function;
- Noninvasive functional tests for ischemia—exercise ECG, stress ECHO, myocardial perfusion scan, stress cardiac MRI;
- Assessment of coronary anatomy—invasive coronary angiography, CT coronary angiogram.
Risk stratification by clinical evaluation
The baseline clinical history and examination, along with 12-lead ECG will provide useful prognostic information. Age is a powerful (although nonmodifiable) determinant of survival. It has been clearly demonstrated that presence of cardiovascular risk factors such as diabetes, hypertension, current smoking, and hyperlipidemia 3 increases the likelihood of adverse outcomes in subjects with established CAD. For example, continued smoking can increase mortality by up to 70%. In addition, comorbidities such as chronic kidney disease, peripheral vascular disease, previous myocardial infarction, presence of signs and symptoms, or heart failure predict adverse cardiovascular outcomes. The pattern and frequency of anginal symptoms, particularly where leading to functional limitation, also predicts mortality 4.
The resting electrocardiogram (ECG) has a role in risk stratification, revealing signs of previous MI, bundle branch block, left ventricular hypertrophy, conduction defects, and arrhythmias. Stable angina patients with resting ECG abnormalities such as presence of Q waves, persistent ST-T changes, left ventricular hypertrophy, left bundle branch block, second- or third-degree heart block, or atrial fibrillation have worse outcome compared to angina patients with a normal resting ECG.
Biomarkers
Cardiac biomarkers are circulatory biochemical molecules released during myocardial stress or damage. The use of cardiac biomarkers in acute coronary syndromes is well established both in diagnosis and also providing valuable prognostic information. With the growing interest in the inflammatory concept of CAD initiation and progression, there have been studies in recent years using several novel biomarkers for risk stratification of SCAD patients.
Serum troponin levels are used mainly in the management of suspected unstable coronary artery disease patients. Very low level of high sensitivity troponin can also be detected in a number of patients with stable coronary artery disease and there are some studies that have found an association between detectable levels of high sensitivity troponin with medium term survival 5, 6.
Other biomarkers such as high sensitivity C-reactive protein (hs-CRP), growth differentiating factor (GDF-1), interleukin-6, interleukin-17, procalcitonin (PCT), and copeptin are found to have association with adverse outcome 7, 8. Hs-CRP has shown prognostic value in predicting adverse outcome in a number of studies, however, routine use is not recommended following systemic analysis of 83 studies, which raised uncertainty about its association 9.
More recently, a group of novel biomarkers were compared with traditional risk factors for predicting 5 year outcomes for cardiovascular events (stroke/MI/CV death) in subjects with stable CAD. The most potent predictors of adverse cardiovascular events were n-terminal protype brain natriuretic peptide, high sensitivity troponin T, and urinary albumin:creatinine ratio, with only current smoking from the traditional risk factor group contributing as much to the model of risk 10.
While at present current guidelines do not recommend their routine use in clinical practice, one can envisage the potential use of biomarkers in the future as a diagnostic and risk assessment tool not only in acute coronary syndrome but also as a combined risk assessment model incorporated with noninvasive tests in stable CAD patients.
Risk Stratification using Ventricular Function
Left ventricular systolic function is accepted as the strongest predictor of long-term survival in subjects with many forms of cardiac disease, including CAD. Historical registry data demonstrates that individuals with coronary disease and LV ejection fraction <50% have a significantly worse prognosis than those with preserved LV function, with a clear relationship between degree of LV systolic dysfunction and extent of risk. Thus in patients with stable CAD, the finding of left ventricular systolic dysfunction should considerably lower the threshold for further investigation. All patients with stable coronary disease should undergo a baseline clinical assessment, with an echocardiogram to quantify LV function a clear recommendation in ESC guidance.
While the default modality for assessing LV function is echocardiography, cardiac MR may add value to risk assessment, for example, by defining extent of scar which is predictor of sudden cardiac death 11.
Risk Stratification by Stress Tests
Despite a lack of randomized trial evidence demonstrating a better outcome for patients undergoing risk stratification by functional tests for ischemia (or stress testing), all major published guidance places stress testing at the core of risk stratification in patients with CAD. Many patients—particularly those at intermediate risk of CAD—will have undergone a stress test in order to make a diagnosis, However, even those with known CAD or a high pretest probability should be considered for stress testing for risk stratification purposes rather than diagnostic purposes.
There is a wide range of stress tests available, the choice of which depends on a combination of available resources, clinicians’ preference, and patient criteria. The relative benefits of these modalities are summarized in Table 1.
| Advantages | Disadvantages | |
|---|---|---|
| Exercise ECG |
Physiological Provides information about patient function Widely available Low cost Extensive evidence base |
Patient factors may limit applicability -inability to exercise - uninterpretable ECG (LBBB, LVH, paced rhythm) |
| Stress Echocardiography |
Wide availability Portability Low cost No radiation Provides quantitative assessment and localisation of ischaemia |
Image quality may be problematic Operator dependent Subjective assessment |
| MPI SPECT |
Generally available Extensive evidence base Provides quantitative assessment and localisation of ischaemia |
Radiation May not detect balanced 3 vessel disease |
| Stress CMR |
Excellent image quality No radiation Potential to assess ischaemia/ viability/ LV function in single scan |
Availability of hardware/ expertise Cost Contraindications (eg implanted devices, claustrophobia) Arrhythmias can cause problems Limited quantification of ischaemia |
| CT coronary angiography |
High sensitivity for detecting CAD High negative predictive value of normal scan Hardware generally available |
Radiation Assessment may be limited if extensive calcification or previous stents May overestimate intermediate lesions |
Exercise electrocardiography
The angina index = 0 point for no angina, one point for nonlimiting angina, two points for angina that limits exercise.
A Duke treadmill score ≥5 predicts low cardiovascular events (99% 4 year survival, annual mortality 0.25%). A score between 4 and -10 suggests intermediate risk and such a group may require further risk stratification with other noninvasive stress test such as myocardial perfusion scan, etc. A score of < −10 indicates high cardiovascular event rates (79% 4 years survival, 5% annual mortality) suggesting further investigation with coronary angiography with a view to revascularization is considered.
However, exercise stress tests may not be applicable in patients with limited exercise capacity due to orthopedic or other medical problems such as respiratory disease. An abnormal baseline ECG such as significant left ventricular hypertrophy, left bundle branch block, or paced rhythm makes interpretation of exercise ECG difficult.
Stress imaging techniques
The incorporation of an imaging technique into a stress protocol results in a number of theoretical advantages over the exercise ECG, including the ability to localize and quantify ischemic myocardium as well as provide information in the presence of resting ECG abnormalities that confound interpretation of dynamic changes. The two more established techniques of stress echocardiography and myocardial perfusion imaging using SPECT (single photon emission-computed tomography) provide similar levels of prognostic information and can be used with either exercise or pharmacological stress; the relatively newer technique of stress CMR (cardiac magnetic resonance imaging) requires pharmacological stress and has a smaller evidence base. The use of exercise as a stressor does reproduce the benefits of exercise ECG in demonstrating the functional capacity of the patient, in addition to also providing information from the ECG in response to a physiological stress. Consequently, ESC guidance suggests exercise as the preferred stressor, although accepts that the outcome data are comparable to pharmacological stress. Clearly, in patients who are unable to exercise adequately, pharmacological stress is needed.
Stress echocardiography
Stress echocardiography using either exercise, dobutamine or vasodilator stressor, is a low cost investigation, which is widely available in most hospital settings. It has been found to provide effective stratification of risk over and above that provided by an exercise ECG 14. Numerous studies have demonstrated that a negative stress ECHO (i.e., with no inducible wall motion abnormalities) carries a good prognosis with a low medium term risk of cardiac events (<0.5% annual mortality or myocardial infarction over 6 years) 15. This also true in subjects at higher baseline risk, for example, with diabetes 16. There is no gender difference in the negative predictive value 17. On the other hand, an abnormal scan is a predictor of increased mortality, with a greater extent of the induced abnormality being associated with a worse prognosis. A positive stress ECHO with regional wall motion abnormalities in more than three segments (in a 17 segment model) is known to have high cardiovascular adverse events (>3% annual mortality). Additional markers of increased risk during stress echocardiography include the involvement of multiple coronary territories, evidence of ischemia at submaximal stress and inducible ischemia in the context of resting regional wall motion abnormalities 14.
Single photon emission-computed tomography (SPECT)
Myocardial perfusion imaging using single photon emission-computed tomography (SPECT) is the imaging stress test with arguably the largest evidence base in terms of risk stratification of coronary disease, using either exercise or vasodilator stress (historically dipyridamole, more recently adenosine receptor agonists). Similar to stress echocardiography, a normal stress perfusion study is associated with low risk of cardiovascular death and MI (<1% annually) 18 indeed the results of the two modalities are considered to be comparable in terms of risk stratification 19. Increasing size of perfusion defect is associated with increasing rates of subsequent cardiovascular events. Historical observational data suggested that perfusion defects involving >10% of LV myocardium (determined by semiquantitatitve analysis incorporating extent and severity of defect) have an annual cardiovascular mortality of >2%, with overall mortality >3% per annum, with mortality rates significantly lower with smaller defects 20. Furthermore, in this cohort, cardiovascular mortality was seen to be lower in patients who underwent revascularization than those treated medically, if >10% myocardium was ischemic; if, however, <10% ischemic burden was demonstrated, then mortality was higher in patients undergoing revascularization. It is this observational data that has formed the cornerstone of using the threshold of ischemic burden >10% (or similar level with other modalities) as determining the need for invasive angiography and revascularization on prognostic grounds. However, it has to be remembered that this construct is based on observational, not randomized-controlled data, and perhaps more importantly is based on data from an era where the use of medical therapy proven to be of prognostic value (aspirin, statins, ACE-inhibitors, beta-blockers) was low. The COURAGE nuclear substudy 21, performed in the context of modern medical therapy, neither demonstrated a significant increase in events with increasing extent of ischemia nor demonstrated a benefit from revascularization, even in patients with a moderate-severe ischemic burden.
Consequently, it is not clear whether revascularization, driven by extent of inducible ischemia on a functional test, improves clinical outcomes in patients with stable CAD; this is a question the International Study of Comparative Health Effectiveness with Medical and Invasive Approaches (ISCHEMIA) trial, currently underway, is aiming to resolve.
Stress cardiac MRI
The use of cardiac MRI (CMR) has increased in recent years and is now more widely available, with an increasing evidence base for prognostication in stable coronary disease. Stress CMR can be performed with either dobutamine or vasodilator stress. Vasodilator first-pass contrast-enhanced perfusion has the benefit of being readily incorporated in a study providing viability assessment in addition to looking for inducible ischemia. The diagnostic accuracy of stress CMR has been demonstrated to be excellent, with the CE-MARC study 22 suggesting superiority to SPECT in a single-center head to head comparison using invasive coronary angiography as the gold standard. Furthermore, recently published 5 year follow-up data from the same trial has demonstrated that CMR is a stronger independent predictor of risk of adverse cardiovascular events than SPECT 23.
A recent meta-analysis 24 confirmed that patients with a normal stress CMR have an annual event rate (for cardiovascular mortality or myocardial infarction) of <1%, whereas the event rate increases to 5% per year when evidence of ischemia is seen. This study also confirmed that there was no significant difference between dobutamine and vasodilator stress in terms of prognostic accuracy. The detection of scar by late gadolinium enhancement in addition to perfusion abnormalities provides incremental information regarding cardiovascular risk 25.
Although conceptually the data regarding event risk increasing with increasing extent of ischemia seen in myocardial perfusion imaging and stress echocardiography should translate to CMR, to date that has not been demonstrated; the published data rely on binary distinctions between inducible ischemia or not, although with additional information relating to presence of scar, viability, and LV function.
Risk Stratification by Coronary Anatomy
Invasive coronary angiography
The ability of invasive coronary angiography to define the site, extent (length/severity), and nature of coronary arterial obstructions, along with the presence of collaterals and some information regarding flow, has led to this modality historically to be considered the “gold standard” for the diagnosis and assessment of coronary artery disease. While undoubtedly it remains key to determining the feasibility of revascularization procedures (either percutaneous or surgical), its role in the assessment an individual's risk of cardiovascular death or coronary events bears further examination.
The most widely used method of defining extent of coronary disease based on anatomy is that developed for the CASS registry 26—the presence of 1-vessel, 2-vessel, 3-vessel, or left main stem disease, with stenosis >70% luminal diameter (50% for left main stem disease). In the “medically treated” arm of the CASS registry, 12-year survival ranged from 71% for single vessel disease down to 40% with 3-vessel disease. Data from a large registry from the same era 27 suggest annual mortality of 1.4% with 1-vessel disease increasing to 8.2% with three vessel disease with >95% proximal LAD stenosis. The adverse impact of very severe stenoses (>95%) along with the involvement of the proximal LAD or left main stem is highlighted. However, while it is clear that the extent and severity of obstructive disease is a powerful predictor of mortality, as mentioned previously, in these studies there was very little use of secondary preventative therapies with now proven prognostic benefit, so there is likely to be a significant overestimate of risk compared to patients of the modern era of evidence-based medical therapy.
Furthermore, coronary angiography only provides anatomical information and thus is not reliable in determining the functional significance of a lesion. For revascularization to provide benefit, there is a prerequisite that it must address inducible ischemia. If that has not been demonstrated by a noninvasive functional test, then the use of fractional flow reserve (FFR) can be used to confirm the functional significance of lesions during invasive investigation to determine revacularization strategy. The FAME-2 trial suggested that patients with FFR-determined nonfunctionally significant coronary lesions had a low event rate when treated medically even in the presence of apparent angiographically significant disease 28. Thus FFR can provide useful prognostic information to supplement invasive coronary angiography in the context of stable coronary disease.
Computed tomography coronary angiography
Modern-computed tomography (CT) coronary angiography provides the opportunity to obtain similar levels of information about coronary anatomy as invasive angiography with a noninvasive approach. A number of large prospective trials have confirmed the potential for CTCA to risk stratify patients by not only the presence and severity of obstructive disease, but also the presence of nonobstructive coronary plaque. Subjects with neither coronary obstructive disease nor plaque have a very low risk of coronary events (<0.3% annual risk) 29. Mortality and coronary event rates increase as number of vessels with obstructive disease or total plaque burden increases 30, with patients with 3-vessel or left main disease having univariate hazard ratio for all cause mortality up to 15.52 29. It is notable that the annualized risk for the various levels of coronary disease appear lower than for the data from invasive coronary angiography. This may reflect a combination of the era in which the studies have been performed (i.e., effectiveness of secondary preventative therapy), the cohort of patients studied (with lower baseline risk in subjects undergoing CTCA) as well as the potential for CTCA to overestimate the degree of coronary obstruction.
Finally, CTCA may also provide the ability to identify nonobstructive high-risk coronary plaques to predict the risk of future coronary events. A combination of positive remodeling and low-density plaques predicted subsequent coronary events 31; however, whether this translates into a clinically useful tool remains to be seen.
Conclusion
Risk stratification is a dynamic process started with initial evaluation at the first clinical encounter to ongoing reevaluation of symptoms and risk assessment while conducting aggressive life-style modification and medical therapy. Optimal medical therapy can safely be continued deferring invasive revascularization in patients with minimal or acceptable symptoms and determined to be at low risk. Symptom relief and better quality of life are the main purpose of revascularization in low-risk patients. Higher risk patients may benefit from early revascularization in addition to optimal medical therapy to improve long-term prognosis; however, absolute risk, and thus the prognostic benefits of revascularization may be overestimated in data based on historical studies; further studies are currently underway aiming to clarify this.
Conflict of Interest
Dr. Aye and Dr. Graham have nothing to disclose.
