By Michael Crawford, MD, Editor
SYNOPSIS: A post-hoc analysis of the SCOT-HEART trial demonstrated exercise ECG is predictive of future coronary heart disease events and mortality. However, coronary CT angiography is more accurate for the detection of coronary artery disease and is more strongly associated with future coronary events.
SOURCE: Singh T, Bing R, Dweck MR, et al. Exercise electrocardiography and computed tomography coronary angiography for patients with suspected stable angina pectoris: A post hoc analysis of the randomized SCOT-HEART trial. JAMA Cardiol 2020; Jun 3. doi: 10.1001/jamacardio.2020.1567. [Online ahead of print].
Some guidelines recommend exercise ECG as the initial test for intermediate-risk patients despite its low sensitivity. Coronary CT angiography (CCTA) is becoming a popular recommendation for this purpose because of its excellent sensitivity.
In the Scottish Computed Tomography of the heart (SCOT-HEART) trial, Singh et al performed a post-hoc analysis in the 79% of participants who had an exercise ECG test to assess the diagnostic, therapeutic, and prognostic benefits of exercise ECG in a contemporary clinical practice and compared it to CCTA. The study population consisted of 3,283 patients who underwent exercise ECG in both the standard care and standard care plus CCTA arms. The attending clinician categorized the exercise ECGs as abnormal, inconclusive, or normal, and were not adjudicated further. Obstructive coronary artery disease (CAD) was defined as a stenosis > 70% of the cross-sectional area or > 50% in the left main coronary artery. Non-obstructive was 10-70%, and normal was < 10% obstruction. The primary endpoint was death from coronary heart disease or a non-fatal myocardial infarction (MI) at five years.
Exercise ECG was normal in 67%, abnormal in 16%, and inconclusive in 17%. In the CCTA arm, among those with a normal exercise ECG, 56% had obstructive (15%) or non-obstructive (41%) CAD by CCTA. Among the 768 patients who underwent invasive coronary angiography, abnormal results on exercise ECG had a 39% sensitivity, a specificity of 91%, a positive predictive value of 58%, and a negative predictive value of 82% for detecting obstructive CAD.
An abnormal exercise ECG was associated with a 14-fold increase in coronary revascularization at one year and a three-fold increase in coronary heart disease death or MI at five years (P < 0.001). Compared to exercise ECG alone, CCTA was more strongly associated with five-year coronary death or MI (hazard ratio, 11; 95% confidence interval, 2-49; P = 0.002). The authors concluded that although an abnormal exercise ECG is associated with future coronary heart disease death, MI, or revascularization, CCTA is more accurate for the detection of CAD and is more strongly associated with future coronary events.
This study confirms the poor sensitivity (39%) and high specificity (82%) of exercise ECG for detecting obstructive CAD. CCTA was much better, with a sensitivity of 97% and a specificity of 86%. Thus, if detecting obstructive CAD is the only goal, there is no contest. However, there are other values to exercise ECG testing, such as the reproducibility, severity, and consequence of exercise-induced symptoms. ECG testing also assesses functional capacity and has been shown to carry prognostic value in many prior studies. The Singh et al study confirms the prognostic value of exercise testing in a contemporary setting, although CCTA performed better in this regard. CCTA detects non-obstructive CAD and can inform risk reduction therapies that an inconclusive or normal exercise ECG test does not.
An exercise ECG may be of value in the initial evaluation of patients with chest pain or other symptoms suggestive of CAD. If abnormal, the test would be highly predictive of obstructive CAD. However, if an exercise ECG is inconclusive, further testing may be required. Additional assessments could include an imaging stress test, which would be expected to be more sensitive in the search for obstructive CAD, but feature a similar specificity as an exercise ECG.
Perhaps CCTA should be the next step for inconclusive exercise ECG patients. The Singh et al study supports that conclusion because many of their patients underwent both tests. What is was not considered here is the decision to include imaging in the initial exercise test. Despite the added cost, exercise imaging might be the best first test in those with stable symptoms. Although CCTA probably is more sensitive when trying to detect obstructive and non-obstructive CAD, it does not provide the other information an exercise ECG does.
There were limitations to this study. It really was an observational study because not all patients underwent both tests, and the application of exercise ECG was not randomized. This was a pragmatic study, with no core lab to adjudicate the test results. But the biggest criticism is CCTA patients received more risk reduction therapy than those who only underwent exercise ECG testing because non-obstructive CAD was detected. The real conclusion of this study is that exercise ECG testing carries therapeutic and prognostic significance, but CCTA offers the added benefit of informing preventive care. Also, exercise ECG is a relatively static technology, whereas CCTA is evolving, with promise for plaque characterization data, microcirculatory assessment, and other measures of potential significance.