(Circulation. 1996;93:205-207.)
© 1996 American Heart Association, Inc.
Articles |
The Search for a Better Exercise Test
A Self-Fulfilling Prophecy?
From the Cardiology Branch, National Heart, Lung, and Blood Institute (R.O.C.), National Institutes of Health, Bethesda, Md, and the Department of Internal Medicine, Henry Ford Hospital (M.L.), Detroit, Mich.
Correspondence to Richard O. Cannon, III, MD, National Institutes of Health, Bldg 10, Room 7B15, 10 Center Dr, MSC 1650, Bethesda, MD 20892-1650, or Michael Lesch, MD, Henry Ford Hospital, 2799 W Grand Blvd, Detroit, MI 48202-2608.
Key Words: Editorials • coronary disease • exercise • electrocardiography • ischemia
 |
Introduction |
|---|
 Top Introduction References |
|---|
Since the demonstration that ST-segment depression is commonly found on the ECGs of coronary artery disease (CAD) patients during stress, graded exercise testing with ECG monitoring has been used to identify patients likely (and unlikely) to have CAD with flow-limiting stenoses that cause myocardial ischemia. Unfortunately, the clinical value of the exercise ECG as a diagnostic test for CAD has been limited by the imperfect association between ST-segment depression (even if >1.0 mm with horizontal or downsloping configuration) and angiographically significant coronary artery lesions, and by use of testing in populations at various degrees of risk for disease.1 2 3
Investigators have derived variations of ECG analysis in an attempt to improve the sensitivity of exercise testing for diagnosing CAD, such as heart rate adjustments of ST-segment depression (ST/heart rate slope, ST/heart rate index) during exercise.4 5 However, these analyses are not widely used, in part because of continued dispute regarding the advantages of these data manipulations over conventional ST-segment analysis alone.6 7 Frustrated by limitations in the diagnostic utility of the exercise ECG, many cardiologists initially perform nuclear or echocardiographic stress (exercise or pharmacological) testing to diagnose CAD in patients with unknown coronary anatomy or go straight to coronary angiography without prior noninvasive stress testing.8
In this issue of Circulation, Barthélémy and coworkers9 report a novel approach to analyzing the response to exercise stress, which they believe enhances the clinical usefulness of exercise testing in diagnosing CAD. In addition to recording the ECG and vital signs during exercise, the authors measured plasma lactate levels within minutes after termination of exercise. Lactic acid is released by exercising skeletal muscles as the oxygen requirements of working muscle cells exceed adequate oxygen delivery to the cells (the so-called oxygen debt).
Appropriate oxygen delivery is determined by the oxygen-carrying
capacity of blood and adequate skeletal muscle perfusion, which in turn
is determined by the cardiac output response to exercise and the
magnitude of vasodilation in working skeletal muscles. At low
workloads, plasma lactate levels are low because of adequate or
near-adequate muscle oxygenation. However, as the intensity of exercise
increases and muscles work harder, lactate release increases
substantially and blood lactate levels rise, with potential development
of a metabolic acidosis if the buffering capacity of bicarbonate in the
blood is exceeded. This point of rapid increase in plasma lactate
levels during strenuous exercise has been called the "anaerobic
threshold," indicative of a shift from aerobic to anaerobic
metabolism in working muscles.10 Beyond this threshold,
continued exercise results in rapid increases in plasma lactate levels
until exercise is terminated. Wasserman and coworkers10
have estimated that the anaerobic threshold in normal individuals is
measurable at 
55% of their maximum exercise capacity. However,
there is variability in the onset of this threshold because of the
level of fitness and training (which affects both the cardiac output
response to exercise and skeletal muscle perfusion during exercise) and
the oxygen-carrying capacity of blood.
The patient cohort of Barthélémy and
coworkers9 consisted of 236 symptomatic men with normal
ECGs at rest referred for suspicion of CAD. Although an assessment of
resting left ventricular function was not reported, patients had no
evidence of prior myocardial infarction, cardiomyopathy, or valvular
heart disease. Exercise was performed using a bicycle ergometer, with
the beginning load and load increases during exercise individualized
such that the total exercise would last approximately 10 minutes. The
exercise test was discontinued with notation of horizontal ST-segment
depression 
1.0 mm (or upsloping ST-segment depression 1.5 mm),
angina pectoris, or exhaustion. All patients in this series underwent
catheterization, including 146 with ST-segment criteria for a
"positive" test; the remaining 90 patients with a
"negative" ECG response underwent catheterization because of
specific referral instructions at the participating institutions. The
authors reported that the plasma lactate levels of the 153 patients
with CAD (70% luminal diameter narrowing of at least one major
coronary artery) were significantly lower than those of the 83 patients
without CAD (7.68±0.42 versus 10.56±0.51 mmol/L,
P<.0001).
Construction of receiver operator curves, which describe the relationship between sensitivity and specificity of a test result in determining the presence or absence of disease, showed highly significant areas under these curves (W values) for plasma lactate levels as well as conventional exercise test parameters: maximum load, exercise duration, percent maximum predicted heart rate, maximum ST-segment depression, age, and heart rate–systolic blood pressure product (in descending order of W values). By multiple logistic regression analysis, only the plasma lactate level and presence of angina during the test appeared to be independently and significantly associated with the presence of angiographically significant CAD. On the basis of plasma lactate levels, the authors then constructed a decision tree, with input of exercise hemodynamic and ST-segment response variables along with plasma lactate levels between the highest and lowest values. By use of this decision tree in their study population, the authors reported a negative predictive value of 96% and a positive predictive value of 97% for their testing in the determination of angiographically significant stenoses involving at least one major coronary artery.
The use of plasma lactate levels as a determinant of CAD has intuitive appeal, because patients with myocardial ischemia during exercise stress would likely not exercise as far or to as high a workload as healthy subjects, either because of angina pectoris and/or dyspnea due to myocardial ischemia or, as in this study, because the physician stops the test upon ECG evidence of ischemia. However, the success of plasma lactate levels as an important discriminant in detecting CAD in this study required considerable cardiologist input, not only in stopping the test at the earliest ECG evidence of ischemia but also in determining the ergometer loads at which to initiate exercise and the load increments for escalation during bicycle exercise. Very likely, the cardiologist chose lower initiating workloads and lower workload increments during exercise for patients who provided a history of angina-like pain or effort dyspnea likely to be a consequence of myocardial ischemia and chose higher workloads for individuals without any history of effort intolerance or individuals with atypical chest pain syndromes. Because the magnitude of lactate release depends in large part on the exercise load and thus skeletal muscle work, one might have been able to roughly predict the ultimate lactate levels on the basis of the clinical determinants that led the cardiologist to choose a particular exercise regimen, with further refinement in this prediction by observing ischemic-appearing ST-segment changes or angina during the test. Thus, in a population at risk for CAD referred for cardiac evaluation (such as the middle-aged and older men participating in this study), the clinical history may be as good a discriminant for the presence or absence of hemodynamically significant CAD as the exercise ECG and, very likely, as the plasma lactate level obtained soon after exercise. Indeed, in this study, angina during exercise testing was a more powerful predictor of angiographically significant CAD than the plasma lactate level, indicative of the high pretest likelihood of disease in their symptomatic study population.
Further, plasma lactate levels as determinants of CAD may not be useful in standardized exercise testing in which relatively large increments in workload changes are used during the course of the exercise test (such as the standard Bruce protocol), when patients develop left ventricular dysfunction during exercise, or when exercise testing is allowed to continue beyond 1.0-mm horizontal ST depression in an attempt to obtain prognostic information. Indeed, it is this last point that diminishes the usefulness of plasma lactate levels during exercise testing: The magnitude of plasma lactate level elevation bore no relation to the angiographic extent of CAD in the authors' study. Treadmill exercise testing, as is true for nuclear and echocardiographic stress testing, is used not only to diagnose CAD but also more commonly to obtain information regarding the likely extent and severity of CAD, findings with important prognostic implications. Further, in a population of patients at low risk for CAD, termination of exercise at an early stage because of a "false-positive" ST-segment response might be associated with a relatively low plasma lactate level, thus diminishing the specificity of plasma lactate level as a determinant of angiographically significant disease.
Although conventional exercise testing with analysis of ST-segment
changes may be less commonly used by cardiologists for the purpose of
diagnosing CAD, it is being more commonly used by internists and other
"gatekeepers" to determine the need for cardiology
referral.11 Cardiologists and noncardiologists alike need
to know which variables permit optimal (70%) sensitivity and
specificity in the performance of exercise testing for the purpose of
diagnosing CAD; attention to skin preparation and electrode placement
for 12-lead ECG recording, exclusion of patients with abnormal rest
ECGs (including left ventricular hypertrophy), performance of study off
medications (especially digoxin), avoidance of preexercise
hyperventilation, adequate exercise performance, and acceptance of
horizontal or downsloping ST-segment depression >1.0 mm during or
after exercise or ST-segment elevation in the absence of baseline
pathological Q waves as evidence for ischemia.12 In a
patient population at intermediate to high risk for CAD, the positive
and negative predictive values of exercise test results with attention
to these considerations should be sufficient to justify continued use
of this relatively inexpensive test modality, recognizing that test
results provide only estimates of the probability of significant CAD.
Equivocal or nondiagnostic studies should be repeated with nuclear or
echocardiographic imaging during exercise or pharmacological stress.
The necessity of adding plasma lactate measurement should await proof
of its independent diagnostic value in large, unselected patient
populations undergoing diagnostic testing for CAD.
|
Footnotes |
|---|
The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association.
|
References |
|---|
|
TopIntroductionReferences |
|---|
-
Goldschlager N, Selzer A, Cohn K. Treadmill stress tests
as indicators of the presence and severity of coronary artery disease.
Ann Intern Med. 1976;85:277-286.
-
Epstein SE. Value and limitations of the electrocardiographic
response to exercise in the assessment of patients with coronary
disease. Am J Cardiol. 1978;42:667-674.[Medline]
[Order article via Infotrieve]
-
Patterson RE, Horowitz SF. Importance of epidemiology and
biostatistics in deciding clinical strategies for using diagnostic
tests: a simplified approach using examples from coronary artery
disease.J Am Coll Cardiol. 1989;13:1653-1665.[Abstract]
-
Elamin MS, Mary DASG, Smith DR, Linden RJ. Prediction of
severity of coronary artery disease using slope of submaximal ST
segment/heart rate relationship. Cardiovasc Res. 1980;14:681-691.[Medline]
[Order article via Infotrieve]
-
Kligfield P, Ameisen O, Okin PM. Relation of the exercise
ST/HR slope to simple heart rate adjustment of ST segment
depression. J Electrocardiol. 1987;20(suppl):135-140.
-
Thwaites BC, Quyyumi AA, Raphael MJ, Cancepa-Anson R, Fox KM.
Comparison of the ST/heart slope with the modified Bruce exercise test
in the detection of coronary artery disease. Am J Cardiol. 1986;57:554-556.[Medline]
[Order article via Infotrieve]
-
Lachterman B, Lehmann KG, Detrano R, Neutel J, Froelicher VF.
Comparison of ST segment/heart rate index to standard ST criteria for
analysis of exercise electrocardiogram. Circulation. 1990;82:44-50.
[Abstract/Free Full Text]
-
Topol EJ, Ellis SG, Cosgrove DM, Bates ER, Muller DWM, Schork
NJ, Schork MA, Loop FD. Analysis of coronary angiography practice in
the United States with an insurance-claims data base.
Circulation. 1993;87:1489-1497.
[Abstract/Free Full Text]
-
Barthélémy J-C, Roche F, Gaspoz J-M, Geyssant A,
Minini P, Antoniadis A, Page E, Wolf J-E, Wilner C, Isaaz K, Cavallaro
C, Lacour J-R. Maximal plasmatic lactate level acts as a major
discriminant variable in exercise testing for coronary artery disease
detection in men. Circulation. 1996;93:246-252.
[Abstract/Free Full Text]
-
Wasserman K, Beaver WL, Whipp BJ. Gas exchange theory and the
lactic acidosis (anaerobic) threshold. Circulation.
1990;81(suppl II):II-14-II-30.
-
Wigton RS, Nicolas JA, Blank LL. Procedural skills of the
general internist: a survey of 2500 physicians. Ann Intern
Med. 1990;111:1023-1034.
-
Gianrossi R, Detrano R, Mulvihill D, Lehmann K, Dubach P,
Colombo A, McArthur D, Froelicher V. Exercise-induced ST depression in
the diagnosis of coronary artery disease: a meta-analysis.
Circulation. 1989;80:87-98.
[Abstract/Free Full Text]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||