By Sarah Handzel, BSN, RN
As one of the most inexpensive and widely available diagnostic tests in the medical field, ECG has a primary role in the evaluation and determination of cardiac risk stratification in patients with coronary artery disease (CAD). Both resting and exercise ECGs can provide valuable insights to supplement information gathered using other diagnostic testing methods, such as cardiac imaging scans.
In many cases, clinicians avoid ECG (especially exercise ECG) as a first-line tool when they evaluate risk in patients with suspected or confirmed CAD. This is unfortunate, as CAD remains the leading cause of death for both men and women in the United States, according to the Centers for Disease Control and Prevention.1 By exploring new ways to use ECG, physicians may be able to estimate cardiac risk stratification more effectively and develop successful care plans.
The Value of Resting ECGs
While the importance of exercise ECG is well understood, resting ECG can also be a powerful prognostic tool. The power of having a baseline ECG is so that physicians can use it for comparison to detect new onsets of the abnormalities listed (arrhythmias, BBB, conduction defects, etc). They can help physicians identify signs of:
- Arrhythmias
- Bundle branch block (BBB)
- Conduction defects
- Left ventricular hypertrophy
- Myocardial infarction (MI)
In the clinical setting, a resting ECG should be the first test performed, according to Verywell Health.2 Stable angina patients with resting ECGs displaying abnormalities like Q waves, left ventricular hypertrophy, left BBB, second-degree or third-degree heart block, Afib, or persistent ST-T changes are more likely to have poorer outcomes than angina patients with normal resting ECGs. This is particularly important to note, as angina is one of the most common presenting symptoms of CAD.3
Insights from Exercise ECGs
Stress testing is at the core of risk stratification for patients with CAD. Though commonly used for diagnostic purposes, stress testing can also be used effectively for risk stratification purposes, even for patients with known CAD or a high pretest probability.4
Functional stress tests like exercise ECG have certain advantages over imaging stress tests. For instance, ECG provides real-time information about patient function, is supported by an extensive evidence base, and is widely available and inexpensive. However, these tests may be limited by a patient's ability to exercise, or the ECG may prove uninterpretable.
Still, exercise ECG can yield data that supplements other types of stress tests, such as imaging tests. As one study published in JACC: Cardiovascular Imaging notes, ST-segment depression is a powerful predictor of future cardiac events; the study found that the maximum ST-segment deviation was the strongest predictor of both cardiac death and a composite of cardiac death and nonfatal MI.5 The study also noted that ST-segment depression and exercise duration were the two most powerful prognostic markers of exercise testing.
Additional variables supplementing ST-segment depression during exercise can add to the prognostic function of exercise ECG and help to determine a patient's risk stratification for CAD. These variables include:
- Exercise capacity, measured in metabolic equivalents (METS)
- Lead aVR data
- QRS duration changes
- Rapid recovery of ST-segment changes
In many cases, lead aVR is ignored in ECG interpretation, but this lead may actually function as a "pseudo-intracavity" lead. As such, aVR may help to identify anterior wall transmural ischemia. The study in JACC: Cardiovascular Imaging found that a ≥1mm aVR elevation during exercise ECG was the strongest predictor of obstructive left main or ostial left anterior descending artery stenosis, with an accuracy of 80%.5 However, more research is necessary before this marker can be routinely applied in clinical practice.
To learn more about the power of the ECG in today's clinical landscape, browse our Diagnostic ECG Clinical Insights Center.
Novel Risk Stratification Method with Machine Learning
Machine learning is also emerging as a new prognostic tool that can be used in combination with ECG. In 2018, a review in Proceedings of Machine Learning Research detailed how researchers used a multiple-instance learning framework to interpret ECGs from over 5,000 patients admitted to a hospital with acute coronary syndrome.6 Risk stratification was determined by plotting ECG information and measuring both the area under the receiver operating curve and the odds ratio.
The study's ultimate goal was to identify patients at risk for cardiovascular death within 30, 60, 90, and 365 days of admission to the hospital. The machine learning model outperformed existing risk metrics, and future machine learning models that incorporate patient-specific features may prove even more accurate.
Determining risk stratification can be accomplished using various methods, but ECG is an especially valuable tool for gauging a patient's cardiac risk. As new risk measurement methods are discovered, more clinicians may realize that ECG is appropriate as a first-line test and that it may accurately predict risk without the need for additional, more invasive testing.
Resources
1. Heart disease facts. Centers for Disease Control and Prevention. Last reviewed May 15, 2023. https://www.cdc.gov/heartdisease/facts.htm. Accessed May 18, 2023.
2. Bailey A. What is a dangerous heart rate? Verywell Health. Updated April 8, 2023. https://www.verywellhealth.com/dangerous-heart-rate-5215509. Accessed May 18, 2023.
3. Coronary artery disease. Centers for Disease Control and Prevention. Last reviewed July 19, 2021. https://www.cdc.gov/heartdisease/coronary_ad.htm#:~:text=Angina%2C%20or%20chest%20pain%20and,the%20rest%20of%20your%20body. Accessed May 18, 2023.
4. Lak HM, Ranka S, Goyal A. Pharmacologic stress testing. StatPearls. 2022 August. https://www.ncbi.nlm.nih.gov/books/NBK555963/.
5. Beller GA, Bourque JM. Value of exercise ECG for risk stratification in suspected or known CAD in the era of advanced imaging technologies. JACC: Cardiovascular Imaging. 2015 Nov;8(11):1309–1321. https://www.jacc.org/doi/10.1016/j.jcmg.2015.09.006.
6. Shanmugam D, Blalock D, Guttag J. Multiple instance learning for ECG risk stratification. Proceedings of Machine Learning Research. 2018; 85:1–15. http://proceedings.mlr.press/v106/shanmugam19a/shanmugam19a.pdf.
Sarah Handzel, BSN, RN has been writing professionally since 2016 after spending over nine years in clinical practice in various specialties.
The opinions, beliefs and viewpoints expressed in this article are solely those of the author and do not necessarily reflect the opinions, beliefs and viewpoints of GE HealthCare. The author is a paid consultant for GE HealthCare and was compensated for creation of this article.
