Refined interpretation of exercise ECG testing: Opportunities for a comeback in the era of expanding advanced cardiac imaging technologies?

Ever since the initial description of the Master step test in 19421 exercise electrocardiography (ECG) is the most widely used non-invasive method for the diagnosis of angina pectoris and the risk stratification of patients with known coronary artery disease (CAD). Its diagnostic accuracy in detecting significantly obstructive CAD based on the development of ‘ischaemic’ horizontal or down-sloping ST-segment depressions during exercise is, however, rather mediocre, with a sensitivity of 68% and a specificity of 77%.2 When tested in populations without referral bias from coronary angiography the sensitivity dropped to 50% whereas the specificity increased to 90%.3 However, the diagnostic and prognostic value of exercise ECG testing can be enhanced by incorporating in the assessment additional clinical data that exercise testing provides beyond the detection of ST-segment depression: exercise capacity, chronotropic and blood pressure response, heart rate recovery, calculation of the Duke treadmill score, T-wave alternans.4–7 The diagnostic accuracy of exercise ECG can further be improved by analysis of the relation between ST-segment depression and heart rate changes during exercise and recovery.8 Heart rate adjustment of ST-segment depression during exercise performed by calculating the ST-segment/heart rate (ST/HR) slope or the ST/HR index can improve the sensitivity with preservation of specificity, mainly from improved classification of patients with equivocal test responses related to up-sloping ST-segments.9 Analysis of the ST/HR loop during exercise and early recovery for the presence of hysteresis provides an alternative method for determining exercise ECG test positivity and negativity. ST-segment depression caused by myocardial ischaemia resolves in patients more slowly during the early recovery phase than in normal individuals. This leads to the presence of a positive ST/HR hysteresis in ischaemic patients as the ST-depression during the first few minutes is greater than it was at the same heart rate during exercise, whereas normal individuals show a negative ST/HR hysteresis (Figure 1). Analysis of post-exercise ST/HR hysteresis has been shown to provide higher diagnostic and prognostic test accuracy than that found for the standard ST-segment depression criteria and other ST/HR indices.10,11 Assessment of ST-segment/heart rate (ST/HR) hysteresis from the ST/HR diagram of a single electrocardiography (ECG) lead in a normal patient and a patient with exercise-induced myocardial ischaemia. ST/HR data pairs are plotted from the start of exercise, at the end of each minute of exercise, at peak exercise and after each of the first 3 minutes of recovery. ST/HR hysteresis is obtained by dividing the area included between the recovery and exercise ST-segment depression values by HR changes in the recovery phase. Redrawn from Zimarino et al.12 Exercise ECG has a much lower specificity in patients with left ventricular hypertrophy in whom false positive tests are frequently observed.13–15 The present study by Zimarino et al. shows an enhanced diagnostic accuracy of ST/HR hysteresis in patients with hypertension-related left ventricular hypertrophy and confirms the very low specificity of the classic maximal ST-segment depression criterion that is clinically largely unreliable in detecting significant CAD in those patients.16 The authors call for the adoption of the ST/HR hysteresis analysis in the interpretation of exercise testing in patients with left ventricular hypertrophy in order to limit the need for additional functional imaging tests in triaging patients towards invasive coronary arteriography. Based on the rather average diagnostic accuracy of exercise ECG testing in a general not by referral biased population where specificity is above 90%, the National Institute for Health and Care Excellence (NICE) UK recommended that ECG testing should not be used for the diagnosis (or exclusion) of stable angina in patients without known CAD.17 This recommendation has been followed in part by the latest European Society of Cardiology guideline on the management of stable CAD,18 which recommends the use of functional imaging over exercise ECG testing in patients with suspected stable angina pectoris, except in patients with low-to-intermediate pre-test probability (15–65% estimated by the CAD consortium prediction rule) in whom exercise ECG testing still remains sufficient as a first-line test as the number of false positive tests in this subset of patients remains acceptable. In intermediate-to-high probability patients (65–85%) the use of exercise ECG testing alone is expected to result in a high number of false negative test results and therefore functional imaging is recommended. Modern cardiology has a wide portfolio of clinical investigation methods available to aid the diagnosis and risk stratification of patients with stable CAD. The moderate accuracy of exercise ECG, new guidelines favouring functional imaging, financial drivers and fear of a claim if a diagnosis is missed may stimulate cardiologists to prescribe stress ischaemia imaging as a first-line test in an increasing number of patients. In the USA, 75% of patients with suspected CAD underwent stress nuclear myocardial perfusion imaging or echocardiography, whereas exercise ECG testing alone was performed in 25% of the patients.19 In a large registry in the USA 73% of the myocardial perfusion imaging studies ordered for the evaluation of suspected or known CAD were completely normal20 suggesting a dramatic overuse of myocardial perfusion imaging. A shift to a higher use (and possibly overuse) of modern imaging techniques may also occur in Europe as guidelines start to recommend more strongly their use: in a recent draft update of the 2010 clinical guideline on stable angina NICE now recommends to use non-invasive coronary arteriography by multislice computed tomography as a first-line test followed by functional imaging in patients with typical and atypical angina pain symptoms, and continues to ban the use of exercise ECG testing to diagnose or exclude stable angina for people without known CAD.21 Exercise ECG is the cheapest of our diagnostic tests and therefore optimisation of its use in the diagnostic algorithm of patients with suspected angina pectoris may result in important cost savings. The great majority of the patients with suspected stable angina pectoris referred for stress testing are a low-risk group with a low rate of downstream procedures (8.8% cardiac catheterisations and 2.7% revascularisations within 60 days) and with a very low one-year rate of major cardiac events (myocardial infarction rate: 0.29%; death rate: 0.16% and stroke rate: 0.10%).19 Routine anatomical and/or functional imaging of all these patients would constitute a dramatic overuse of expensive highbrow diagnostic testing with a meager outcome benefit, and represents an avoidable burden of cost on the healthcare system. For a great majority of the patients a maximal exercise ECG test is the preferred and only needed first-line test. When a patient shows no inducible ischaemic repolarisation changes during exercise ECG testing at or above the maximal target heart rate with a normal or higher exercise tolerance then the probability of having significant obstructive CAD and for having an adverse event during follow-up is very low. Further anatomical or functional imaging will therefore only be needed in a low number of patients with an equivocal test or positive results. Such a gatekeeper function of exercise ECG testing for more advanced and expensive imaging procedures can result in important cost savings, as was demonstrated in the WOMEN trial: in low-risk women who were able to exercise, a diagnostic strategy that used exercise ECG testing alone provided significant diagnostic cost savings (248%) and was associated with two-year post-test clinical outcomes as good as observed after exercise nuclear myocardial perfusion imaging.22 Further refinement of the interpretation of the exercise ECG test by incorporating supplementary exercise test variables and scores6,23 and by performing a ST/HR hysteresis analysis of the ST-segment changes12 may improve the diagnostic accuracy and provide incremental prognostic information. Although ST/HR hysteresis analysis may enhance considerably the diagnostic accuracy of exercise ECG testing in patients with left ventricular hypertrophy, it must be admitted that the specificity remains suboptimal (75%) and still may lead to a substantial number of false positive test results. Furthermore, patients with left ventricular hypertrophy constitute a high-risk subgroup with an almost three times increased risk of cardiovascular morbidity and mortality.24–26 Therefore, refined analysis of exercise ECG should be combined with exercise echocardiography in order to increase the diagnostic accuracy and the prognostic predictive power in this subset of patients. In patients with known or clinically suspected left ventricular hypertrophy, exercise echocardiography has been shown to be more accurate than exercise ECG testing for the detection of CAD27 and to provide incremental prognostic information in the prediction of death and other major adverse cardiac events.24,25 Moreover, the anginal symptoms in patients with left ventricular hypertrophy may in part be related to coronary microvascular dysfunction and are not always easy to discern from dyspnoea caused by an associated left ventricular diastolic dysfunction. Transthoracic Doppler may allow the assessment of coronary flow velocity reserve at rest28 and to monitor changes of pulmonary pressure changes and diastolic left ventricular function parameters29 during exercise that may increase significantly in patients with an underlying concealed left ventricular diastolic dysfunction. In summary, refined interpretation of exercise ECG by including other exercise variables and performing ST/HR hysteresis analysis may improve the diagnostic accuracy in low-to-intermediate probability patients with suspected stable angina pectoris. However, in intermediate-to-high probability patients including those with left ventricular hypertrophy exercise ECG needs to be bolstered by stress echocardiography.30 The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. The author(s) received no financial support for the research, authorship, and/or publication of this article.