Printer Friendly

Understanding ST depression in the stress-test ECG.


Objective: The electrocardiogram (ECG) obtained during stress testing often shows a typical pattern of primary ST depression. A similar pattern can occur in unstable angina. Current textbooks consider ST depression as a direct result of partial occlusion of a coronary artery. However, animal models could not reproduce this phenomenon. An alternative explanation for ST depression specific to stress testing involves global subendocardial ischemia. In this study, we evaluated both explanations with a realistic mathematical model of the human heart.

Methods: The ECG was simulated with an anisotropic reaction-diffusion model of the human heart and an inhomogeneous boundary-element model of the human torso.

Results: Limited subendocardial ischemic zones caused small ST depression in ECG leads not overlying the ischemic region. An ischemic zone of 50% transmural extent covering the entire left ventricular subendocardium caused an ST-depression pattern similar to that observed during stress test.

Conclusion: In contrast to regional subendocardial ischemia, global subendocardial ischemia can explain ST depression in our model.

Key words: ischemia, ST deviation, non-ST-elevation myocardial infarction, computer model, stress test


The ECG obtained during stress testing often shows a typical pattern of ST depression. A similar pattern can occur spontaneously in patients with unstable angina (1). The current textbook explanation of ST depression involves regional subendocardial ischemia as a direct result of partial occlusion of one or more coronary arteries. This would imply that ST-segment changes depend on the affected artery or arteries (2). However, in contrast to ST elevation, ST-depression patterns appear to be independent of the affected arteries (1, 3). Moreover, animal models could not reproduce ST depression at a resting heart rate (4-6). Recent theoretical work has shown that the classical relation between regional subendocardial ischemia and epicardial ST depression relies on an isotropic mathematical model of the myocardium (7, 8). Isotropic models were previously used because of practical limitations. In a more realistic anisotropic computer model of the human heart, ST depression could only be obtained with subendocardial ischemic zones that covered more than half of the left ventricle (8). In this study, we investigated whether such a realistic model can reproduce the pattern of ST depression that is typical for the stress test.


The ECG was simulated with a reaction-diffusion model of the human heart incorporating anisotropic myocardium with transmurally rotating fiber orientation at 0.25-mm resolution, and an inhomogeneous boundary-element torso model. Details of this model have been published previously (9). Ionic currents in the propagation model were computed with the Ten Tusscher-Noble-Noble-Panfilov (TNNP) model of the human ventricular myocyte (10). Ischemia was represented by setting the extracellular potassium concentration to 10 mM (normal value 5 mM).

Our model is based on the bi-domain model of the myocardium. It represents the myocytes and gap junctions as a continuum called the "intracellular domain," and the interstitium and microvasculature as another continuum, called the "extracellular domain." Both domains have anisotropic conductivity. The model accounts for both these anisotropies when it computes propagating action potentials, but it cannot deal with extracellular anisotropy (Re) when computing the ECG. To compensate, we used a reduced intracellular anisotropy (Ri) for ECG computation. Thus, the ratio of intracellular to extracellular anisotropy (R) was realistic. This ratio is much more important for ST-segment changes than the individual anisotropy ratios of the two domains (8). Normal values are Ri=10 and Re=2.5, so R=Ri /Re=4 (11). Bound to using Re=1, we set Ri=4 for ECG computation.


An ischemic zone of 50% transmural extent covering the entire left ventricular subendocardium caused an ST-depression pattern similar to that observed during stress test. This was verified for intracellular to extracellular anisotropy ratios R=1 and R=4 (Fig. 1). This increase in anisotropy ratio slightly increased T-wave amplitude in the precordial leads, but did not significantly affect the ST segment. Compared to a simulated normal ECG, the QRS complex was only slightly affected: R peak amplitudes were reduced by 20% in leads II, III and AVF.


In contrast to a global subendocardial ischemia, regional subendocardial ischemia induced ST-segment changes that depended strongly on the anisotropy ratio of the myocardium. Figure 2 shows ST changes due to an ischemic zone with 5 cm diameter and 50% transmural extent in the lateral wall of the left ventricle. The ECGs were simulated with isotropic tissue (R=1) and anisotropic tissue (R=4). Lead V6, which overlies the ischemic zone, is shown in panel A. Panel b shows the ST deviation in all precordial leads. ST depression in the leads overlying the ischemic zone was only obtained with isotropic tissue. With a more realistic anisotropy ratio of 4 the maximum ST depression shifted to lead V3.


Despite the smaller affected muscle mass, regional ischemia led to more prominent QRS changes than global subendocardial ischemia. Especially the end of the QRS complex was affected. As a result of regional ischemia, the QRS complex became narrower in lead III.


We have shown that regional subendocardial ischemia leads to an ST-deviation pattern that strongly depends on the anisotropy of the tissue. When a realistic anisotropy ratio was used, ST depression was not centered over the ischemic region. Global subendocardial ischemia led to considerable ST depression in all standard leads, with little dependence on anisotropy.

Classically, primary ST depression has been considered to be a direct result of subendocardial ischemia due to partial occlusion of one or more coronary arteries. Previous modeling studies have shown that if regional subendocardial ischemia is present, it can be localized on the body surface (2). Our simulations of regional subendocardial ischemia confirm this result. If the myocardium was isotropic (R=1), the ST depression was maximal in the leads overlying the ischemic region. With a more realistic anisotropy ratio R=4 the pattern still depended on the location of the ischemia, but the maximum ST depression occurred adjacent to the ischemic region.

Thus, modeling studies suggest that if partial occlusion leads to regional subendocardial ischemia, the occlusion can be localized by the ECG. However, in contrast to ST elevation, primary ST depression during stress testing and in unstable angina usually occurs in a typical pattern that does not depend on the affected artery or arteries (1, 3). Moreover, partial occlusion of a coronary artery in animal models did not lead to ST depression (4-6). We hypothesized therefore that the "stress-test ECG" is not caused by regional ischemia, and is not a direct result of partial occlusion. During stress testing, reduced diastolic coronary filling time and elevated left ventricular end-diastolic pressure can reduce perfusion in the subendocardium, leading to a global subendocardial ischemia (12). Our present results suggest that global subendocardial ischemia can explain the stress-test ECG. This ST depression would not depend on the territory of the affected artery.


Computational resources for this work were provided by the Reseau quebecois de calcul de haute performance (RQCHP). M. Potse was supported by a postdoctoral award from the Groupe de recherche en sciences et technologie biomedicale (GRSTB), Ecole Polytechnique and Universite de Montreal; and by the Research Center of Sacre-Coeur Hospital, Montreal, Quebec, Canada.


(1.) Nasmith JB, Pharand C, Dube B, Matteau S, LeBlanc AR, Nadeau R. Localization of maximal ST segment displacement in various ischemic settings by orthogonal ECG: implications for lead selection and the mechanism of ST shift. Can J Cardiol 2001; 17: 57-62.

(2.) MacLachlan MC, Sundnes J, Lines GT. Simulation of ST segment changes during subendocardial ischemia using a realistic 3-D cardiac geometry. IEEE Trans Biomed Eng 2005; 52: 799-807.

(3.) Mark DB, Hlatky MA, Lee KL, Harrell FE, Jr, Califf RM, Pryor DB. Localizing coronary artery obstructions with the exercise treadmill test. Ann Intern Med 1987; 106: 53-5.

(4.) Li D, Li CY, Yong AC, Kilpatrick D. Source of electrocardiographic ST changes in subendocardial ischemia. Circ Res 1998; 82: 957-70.

(5.) Kilpatrick D, Johnston PR, Li DS. Mechanisms of ST change in partial thickness ischemia. J Electrocardiol 2003; 36 Suppl: 7-12.

(6.) de Chantal M, Diodati JG, Nasmith JB, Amyot R, LeBlanc AR, Schampaert E, et al. Progressive epicardial coronary blood flow reduction fails to produce ST-segment depression at normal heart rates. Am J Physiol Heart Circ Physiol 2006; 291: H2889-96.

(7.) Hopenfeld B, Stinstra JG, MacLeod RS. Mechanism for ST depression associated with contiguous subendocardial ischemia. J Cardiovasc Electrophysiol 2004; 15:1200-6.

(8.) Potse M, Coronel R, Falcao S, LeBlanc AR, Vinet A. The effect of lesion size and tissue remodeling on ST deviation in partial-thickness ischemia. Heart Rhythm 2007; 4: 200-6.

(9.) Potse M, Dube B, Richer J, Vinet A, Gulrajani RM. A comparison of monodomain and bidomain reaction-diffusion models for action potential propagation in the human heart. IEEE Trans Biomed Eng 2006; 53: 2425-35.

(10.) ten Tusscher KHWJ, Noble D, Noble PJ, Panfilov AV. A model for human ventricular tissue. Am J Physiol Heart Circ Physiol 2004; 286: H1573-89.

(11.) Roth BJ. Electrical conductivity values used with the bidomain model of cardiac tissue. IEEE Trans Biomed Eng 1997; 44: 326-8.

(12.) Ellestad MH, Selvester RHS, Mishkin FS, James FW. Stress Testing; Principles and Practice. 5th ed. New York; Oxford University Press: 2003.

Mark Potse (1,2), Alain Vinet (1,2), A.-Robert LeBlanc (1,2), Jean G. Diodati (1,3), Reginald Nadeau (1,3)

(1) Research Center, Hopital du Sacre-Coeur, Montreal,

(2) Department of Physiology, Institute of Biomedical Engineering, Universite de Montreal, Montreal

(3) Department of Medicine, Universite de Montreal, Montreal, Quebec, Canada

Address for Correspondence: Mark Potse, PhD, Centre de recherche, Hopital du Sacre-Coeur, 5400 Boulevard Gouin Ouest, Montreal, Quebec H4J 1C5 Canada Phone: +1 514 338-2222 #2519 Fax: +1 514 338-2694 E-mail:
COPYRIGHT 2007 Galenos Yayincilik
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2007 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Original Investigation; electrocardiogram
Author:Potse, Mark; Vinet, Alain; LeBlanc, A.-Robert; Diodati, Jean G.; Nadeau, Reginald
Publication:The Anatolian Journal of Cardiology (Anadolu Kardiyoloji Dergisi)
Article Type:Clinical report
Geographic Code:1CANA
Date:Jul 1, 2007
Previous Article:The Pierre Rijlant lecture 2007: the future of electrocardiography.
Next Article:Reliability-based rearrangement of ECG automated interpretation chain.

Related Articles
Update on ACC/ESC criteria for acute ST-elevation myocardial infarction.
Understanding ST depression in the stress-test ECG.
The relation between QRS amplitude and left ventricular mass in patients with hypertension identified at screening.
Comparative assessment of ECG dynamics in myocardial infarction according to reperfusion therapy approach (primary and facilitated coronary...
Pseudonormalization: clinical, electrocardiographic, echocardiographic, and angiographic characteristics.
Evaluation of the QRS-T angle using the high-resolution 64-lead electrocardiography.
The Pierre Rijlant lecture 2007: the future of electrocardiography.
Reliability-based rearrangement of ECG automated interpretation chain.
Clinical presentation and ECG changes--how good is it in diagnosing troponin positive acute coronary syndrome.
Prevalence of ST-elevation in right precordial leads in patients presenting with acute coronary syndrome without ST-elevation in standard 12-lead...

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters