Maximal Treadmill Stress Testingfor Cardiovascular Evaluation

By MYRVIN H. ELLESTAD, M.D., WILLIAM ALLEN, M.D.,

MAURICE C. K. WAN, M.D., AND GEORGE L. KEMP, M.D.

SUMMARYExperience with a maximal treadmill stress testing procedure which is relatively safe,

simple, and reproducible is reported. It has been used in normal persons and cardiacpatients with ages ranging from 7 to 83 years. There have been no deaths in our totalexperience of 4,028 maximal capacity stress tests. Maximal capacity is predictedby the patient's peak predicted pulse rate. Sixty-three per cent of those with ischemicS-T segments did not experience pain of any type.

Additional Indexing Words:Maximal capacity testing Ischemic S-T changesMonitoring Cardiac diagnosis

ALMOST 40 years ago Master and Oppen-heimer' introduced an exercise test for

the detection of coronary insufficiency whichmany still consider to be the standard. Thesignificance of exercise induced S-T segmentdepression as indicative of coronary insuffi-ciency is accepted by most cardiologists.2 4Master's single and double two-step test isaccepted as the standard submaximal stresstest, but there has been no uniform acceptanceof a protocol for a maximal stress test.

It is the purpose of this paper to describe amethod which has evolved in our laboratoryand propose it as a standard maximal stresstest. The procedure is simple to perform andrequires a limited amount of special equip-ment. Experience with it has been extensiveenough to establish norms of performance andto demonstrate its safety.

MethodsThe first 1,000 patients referred to the Division

of Clinical Physiology, of Memorial Hospital ofLong Beach, for maximal treadmill stress testing

From the Division of Clinical Physiology, Memori-al Hospital of Long Beach, Long Beach, California.Work was supported in part by the Long Beach

Heart Association and the Memorial Hospital Re-search Foundation.

Circulation, Volume XXXIX, April 1969

Peak predicted pulse rates

were chosen for detailed analysis. Informationas to their previous health was only known for284 executives previously examined and thoughtto be normal by history and resting ECG. Manyof the remainder were sent for evaluation ofknown or suspected angina, and many were sentfor screening prior to embarking on a physicalfitness program. Careful questioning as to symp-toms, medication, and previous cardiac diseasewas done to rule out unstable coronary insuffi-ciency and congestive failure. In the group were205 females and 795 males with ages varyingbetween 7 and 83 years.No special attempt was made to standardize

the time of day or the relationship of the lastmeal.The patients are prepared by applying gel

(Lectrocardiographic Gel) to the Telectrodeelectrode and affixing these self-adherent electrodesto the upper part of the manubrium sterni andthe standard left chest V5 position (CM-5).5The cable attachments are then snapped to theelectrodes and the cable is connected to a direct-writing Sanborn electrocardiograph. The electro-cardiographic complexes are monitored continuallywith an oscilloscope. An aneroid sphygmoma-nometer is placed on the right arm for measure-ments of blood pressure. A cardiotachometer givesa constant read-out of the heart rate. Oxygen,emergency drugs, and a DC defibrillator are avail-able in the room.

Resting electrocardiograms are taken while thepatient is sitting and also while standing,before and after hyperventilation, and are used

517

by guest on April 17, 2018

http://circ.ahajournals.org/D

ownloaded from

ELLESTAD ET AL.

Table 1Ages and Mlaximal Pulse Rates (1IIPR)*

MPR Age(yr)

Age(yr)

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

383940

200

199

198

197

196

195194

193

192

191190

190

189

188

187

186

186

185

184

183

182

4112

43

44

45

46

4i

48

49

50

5152

53

54

55

56

57

58

59

60

X1I1I1

181

180

180

180

179

177

177177

176

175

174

173

172

171

171

170

170

169

168

168

Age(yr)

6162

6.364

65

66

67

68

69707172737475a

MIPR

167

167166

165164163162161161160

160

160160160

160

* These figures are based upon regression figures ofRobinson.8a Age groups from 10 to 20 years have beencompiled by Astrand and associates, 6-8 and range ofMPR from 210 to 197, respectively.

as a base line for changes occurring during andafter exercise. Blood pressure is also taken in thesitting and standing positions. The patient thensteps onto a treadmill which has a fixed inclineof 10% and walks for 3 min at 1.7 mph, 2 minat 3 mph, 2 min at 4 mph and finally, 3 min at5 mph. Blood pressure and electrocardiogramare recorded at 1-min intervals during exerciseand for a period of 8 min following exercise.The exercise is terminated if the patient be-

comes exhausted, if the blood pressure fallssignificantly, if there is progressive S-T segmentdepression or pain, or if multiple prematureventricular contractions or ventricular tachycardiaoccurs. If none of these occur, the patient isurged to continue until he or she reaches atleast 95% of the predicted maximal pulse rate. Acompilation of maximal pulse rates (MPR)based upon age is listed from studies by Astrandand Norris and their associates6-8 (table 1).

Ischemic S-T change in the electrocardiogramoccurring during or in the 8-min period of moni-toring after maximal exercise stress testing isdefined as a 2-mm depression below the iso-electric line lasting for 0.08 sec from the J-point(fig. 1). When the depressed S-T segment is nothorizontal but slopes upward, a point 0.08 sec

after the J-point is measured, and if it is 2 mmbelow a line drawn through the P-Q junction,the tracing is read as positive for ischemia. De-pressions of lesser magnitude are read as equiv-ocal for ischemia if the S-T segment is concave,but are considered as a normal finding if the S-Tsegment is convex. T-wave changes by themselvesare not considered in the evaluation of ischemicresponse to exercise if there are no coexistingabnormalities in the S-T segment. Precipitationof frequent premature atrial or ventricular con-tractions or an increase in these if present at restas well as runs of ventricular tachycardia duringor after exercise are considered equivocal findingsnot necessarily indicative of myocardial ischema.

Results

SafetyNo deaths occurred during the testing in

the 1,000 cases presented herein or from anyof the subsequent 3,028 treadmill tests per-formed in our laboratory. Ventricular asystoleand ventricular fibrillation were not seen. Tran-sient ventricular tachycardia, lasting less than20 sec and reverting spontaneously, occurrednine times. Only one patient required anytherapy for ventricular tachycardia. He be-came slightly hypotensive and diaphoreticand was converted to normal sinus rhythm byDC countershock. Disturbances in A-V con-duction were precipitated in only two patients.No patient fell from the apparatus althoughphysical support was frequently needed at theend of walking to assist the patient in sittingdown.

Transient vasovagal reactions, that is, mildhypotension and bradycardia in the early re-covery period, occurred in less than 1%; of thegroup. Hypotension during or after exerciseoccurred infrequently and was rarely a prob-lem. Tvo patients had myocardial infarctionstemporally related to the test, but both sur-vived.The safety of maximal stress testing has

been demonstrated not only by this study butalso by Bruce and others.9-11 It must be em-phasized that safety requires continuous ob-servation by an experienced physician. Thetest can then be terminated in time to avoidtrouble, or the patient can be encouraged tocontinue long enough for us to obtain definitivedata. A careful history, and if indicated, a

Circulaijon, Volume XXXIX, April 1969

S18

by guest on April 17, 2018

http://circ.ahajournals.org/D

ownloaded from

MAXIMAL TREADMILL STRESS TESTING

S-T Segment Change With Exercise

A B C

Figure 1(A) This QRS-T complex demonstrates a normal configuration noted during or after maximalstress. f-point depression is associated with a convex J-X curve. (B) The J-point depression isassociated with a slow return of the S-T segment to the base line. By measurement, a point0.08 sec after the J-point is projected upward to bisect the S-T segment. If this intersection is2 mm or more below the iso-electric line, the tracing is considered positive for ischemia. TheJ-X curve is concave. (C) This complex demonstrates the typical ischemic S-T segment de-pression seen after maximal stress. The S-T segment is depressed 2 mm or more below theiso-electric line and there is usually straightening of the S-T segment.

REPEATABILITY OF MAXIMAL EFFORT

& ISCHEMIC S-T CHANGES

Duration of ISCHEMICMAXIMAL S-TEFFORT CHANGES

- IDENTICAL WITHIN ONE-i ,N9s3.-; T ME :5: : : MI NUTE

Figure 2

resting ECG should be repeated to preventexercising a patient with recent infarction orunstable angina.

Reproducibility

To evaluate the reproducibility of a pa-

tient's response to maximal treadmill stresstesting, 25 males, 40 to 68 years old, had repeatstress tests within 1 to 90 days. Twenty-twoof the patients had angina pectoris or a his-tory of myocardial infarction, or both, andthree did not. Fifteen patients (60%) per-

formed for an identical time, eight (32%)walked 1 min less or 1 min more than they hadon their previous treadmill test, and only two(8%) had a difference in duration of effortgreater than 1 min on repeat study. Thus, in92% the duration of exercise was within 1 minon the repeat study (fig. 2). The hypothesis

Of a total of 25 males who were retested by maximaltreadmill exercise within a 90-day period, 92% per-formed for a similar length of time (within 1 min)Circulation, Volume XXXIX, April 1969

and 95% developed S-T segment abnormalities at asimilar time interval (within 1 min).

100 -

90 -

8o

( 70

60 -

50 -

40 -

30 -

20 -

I

519

by guest on April 17, 2018

http://circ.ahajournals.org/D

ownloaded from

ELLESTAD ET AL.

of more than chance reproducibility of per-formance was tested using the standard nor-mal deviate, Z. The frequency categories weresplit into performance within 1 min and great-er than 1 min, and were significant at lessthan the 0.01 level.Twenty-two patients, all of whom had a pos-

itive history, exhibited ischemic S-T changes.Fifteen patients (68%) developed ischemicS-T changes at the identical time interval, six(27%) within 1 min of the previous time, andone (5%) had his S-T changes at a time inter-val greater than 1 min. The onset of ischemicS-T changes was separated by only 1 min orless in 95% of the patients. By splitting thefrequency categories into less than 1 min andgreater than 1 min, a Z test indicated signifi-cant reproducibility (less than 0.01 level ofsignificance).

Incidence of Chest PainOf 284 apparently normal executives referred

for treadmill stress testing as part of a routinephysical examination, 30 (11%) developedischemic S-T changes, and 10 (3.5%) hadequivocal S-T changes during or after exercise.All of the executives were males, aged 30 to59 years, and in none was heart disease pre-viously suspected. It was surprising that in noinstance was chest pain associated with ischem-ic abnormalities. This executive group will bethe subject of a subsequent report.A detailed analysis of the remaining 716

patients is not presented, since they were re-ferred for various reasons, some of which werenot apparent at the time of the test. However,the relationship between positive tests andchest pain in the total group is of interest.Only 88 (37%) of the 236 patients with posi-tive tests had chest pain. Sixty-three per centhad ischemic changes without pain. The fe-males had a 16% higher incidence of chestpain than the males (50% and 34%, respectively)with Z significant at less than 0.01 level. Theyounger males (31 to 40 years of age) standout as the group which had the lowest per-centage of chest pain (13%). Of the 88 pa-tients who developed chest pain as well asischemic S-T patterns, 61 (69%) experiencedpain within the first 5 min.

Discussion

When considering an exercise stress test,several objectives should be kept in mind:(1) It should be safe, (2) should require alimited amount of special equipment, (3)should not be too time consuming, (4) shouldbe adaptable enough in design so that it doesnot overstress some and understress othercardiac patients, (5) should use a familiarform of exercise, and (6) results should bereproducible.9, 12-14

The reproducible association between theexercise work load and the onset of S-T seg-ment depression has been documented byAreskag,12 Burkart and their associates,13 andothers.9' 1 The product of systolic blood pres-sure multiplied by the pulse rate has beenfound by others to be even more predictive ofthe end point in patients with coronary in-sufficiency.16 It has been our impression thatthe more severe the disease, the more re-producible the test.

Since the cardiac output and oxygen con-sumption increase in nearly a linear relation-ship with the pulse, the peak pulse responseallows us to estimate the maximal cardiacoutput.6-8 Many patients with coronary dis-ease do not reach their predicted maximalpulse rate; however, in those with normaltests only 6% failed to reach a pulse rate of atleast 80% of predicted value.The advantage of our procedure over the

Bruce procedure is based mainly on its sim-plicity and the fact that it takes less time toperform. The 10% incline is kept constant inour test in contrast to their changing inclines.However, if one wishes to study the well-trained athlete, the Bruce test has the advan-tage of requiring much higher levels of energyexpenditure toward the end of the test.The timing of the work periods has been

arbitrarily set for convenience. The initial 3-min period functions as a warm-up and allowsthe patient to become acquainted with theuphill grade. When the speed reaches 4 mph,some subjects must jog and almost all mustdo so at 5 mph. It has been suggested that in-creasing the grade would be a more satis-factory method of increasing the work load,

Circulation, Volume XXXIX, April 1969

520

by guest on April 17, 2018

http://circ.ahajournals.org/D

ownloaded from

MAXIMAL TREADMILL STRESS TESTING

but we have found that the stretch in thesoleus muscles is very uncomfortable for un-

trained individuals at the higher inclines.More study on the range of oxygen con-

sumption at each speed is underway, but some

preliminary data collected with the Webbmetabolic rate monitor are available. At a

speed of 1.7 mph the oxygen consumption isequivalent to 4 to 5 times the basal consump-tion of oxygen, at 3 mph to 6 to 7 times basal,at 4 mph to 9 to 10 times basal, and 5 mph to15 times basal. Vo2/kg/min for 10 middle-aged males averaged 4.2 at rest, 5 at 1.7 mph,7 at 3 mph, 9 at 4 mph and 16 at 5 mph.17These data are consistent with that presentedby Gordon,18 Rowell and associates19 and Fordand Hellerstein20 and would suggest that 3mph at a 10% grade is roughly equivalent toMaster's single-step test and 4 mph is equiva-lent to Master's two-step test. We do not be-lieve that it is practical to strive for a steadystate at each work load.The controversy over ECG criteria for diag-

nosis of ischemia by exercise has yet to besettled. We have used more stringent criteriafor ischemia than those used by Master andRosenfeld21 in order to avoid false positives as

reported by Mattingly.14 Friedberg and asso-

ciates22 found no false positives only when theS-T segment was depressed 2 mm or more.

However, we believe that the criteria listedby us may be established eventually as toostrict. Robb and Marks,23 using the standarddouble Master's test, have shown that even

minimal S-T depression is associated with a

higher mortality risk. This may not be as validwith maximal stress testing, however, becausemany young, seemingly healthy males andfemales have minimal S-T depression with thistechnic. Sensitivity might also be increasedby using more than one ECG lead.'2

It should be stressed that little is knownabout the specificity and prognostic signifi-cance of ischemic S-T changes during andafter maximal exercise. A recent paper byMost and co-workers24 supports the view, how-ever, that relative ischemia is the most likelycause.

The infrequent occurrence of pain associatedCirculation, Volume XXXIX, Aprl 1969

with S-T depression should be consideredwhen the absence of angina is used to ruleout the likelihood of coronary insufficiency.Exercise is often prescribed to the point ofpain on the grounds that it is a reliable indi-cator of myocardial ischemia. Our experienceindicates that more often than not ischemia,often associated with ventricular irritability,will develop unannounced by pain or any

other symptom easily recognizable by thepatient.The fact that the incidence of pain is higher

in patients manifesting S-T segment depressionearly in the test suggests that it may be partlyrelated to the degree of ischemia. In thosewith S-T depression at or near peak pulserate it was invariably absent. On the otherhand, typical coronary pain in the absence ofS-T depression is rare, and we have not seen

it more than a few times. While we were ableto discover 11% positive ischemic responses

to the treadmill stress test in a group of execu-

tives without clinical heart disease, in no case

were these ischemic changes associated withpain in the chest, even though each man was

exercised to his maximal capacity.Using maximal stress testing in the younger

age groups seems particularly important. Sixtyper cent of our males, aged 31 to 40 years withpositive tests, would have been missed bysubmaximal testing.We have used treadmill stress testing to

discover subclinical coronary artery diseaseand to clarify the etiology of chest pain, toevaluate the results of cardiac surgery, and toassess medical management of coronary dis-ease. It may also be helpful in developing ex-

ercise prescriptions for patients with coronary

disease.*

References

1. MASTER, A. M., AND OPPENHEIMER, E. T.: Simpleexercise tolerance test for circulatory deficiency,with standard tables for normal individuals.Amer J Med Sci 177: 223, 1929.

2. MASTER, A. M., AND ROSENFELD, I.: Can the

*Note: Pulse response graphs for untrained normalindividuals, male and female, age 20 to 70 years, areavailable on request from M. H. Ellestad.

521

by guest on April 17, 2018

http://circ.ahajournals.org/D

ownloaded from

ELLESTAD ET AL.

amount of S-T segment depression after the"two-step" test be correlated with the severityof ischemic heart disease? Amer J Cardiol15: 139, 1965.

3. BELLET, S., AND MULLER, 0. F.: Electrocardio-gram during exercise: Its value in the diagno-sis of angina pectoris. Circulation 32: 477,1965.

4. ARBARQUEZ, R. F., JR., KINTANAR, Q. L., VALDEZ,E. V., AND DAYRTrr, C.: Evaluation of somecriteria for the dynamic and post exerciseelectrocardiogram in diagnosing coronary in-sufficiency. Amer J Cardiol 13: 310, 1964.

5. BLACKBURN, H., TAYLOR, H. L., VASQUEZ, C. L.,AND PUCHNER, T. C.: Electrocardiogram dur-ing exercise. Circulationi 34: 1034, 1966.

6. SHOCK, N. W., ANDRES, R., LANDOWNE, M.,NORMS, A. H., SIMONSON, E., AND SWARTZ,F. C.: Aging of the cardiovascular system.Nat Conf Cardiovasc Dis 2: 558, 1964.

7. ASTRAND, I., ASTRAND, P. O., CHRISTENSEN, E. H.,AND HEDMAN, R.: Circulatory and respiratoryadaptation to severe muscular work. ActaPhysiol Scand 50: 254, 1960.

8. ASTRAND, L., ASTRAND, P. O., AND RODAHL, K.:Maximal heart rate during work in oldermen. J Appl Physiol 14: 562, 1959.

8a. ROBINSON, SID: Experimental studies of physicalfitness. Arbeitsphysiologie 10: 251, 1938.

9. BRUCE, R. A., BLACKMAN, J. R., JONES, J. W.,AND STRAIT, G.: Exercise testing in adultnormal subjects and cardiac patients. Pediat-rics 32: 742, 1963.

10. HASKELL, W. L., AND Fox, S. M.: Possibleplace of stress testing to discover, and physicalactivity to prevent, coronary heart disease.Southern Med J 59: 642, 1966.

11. SANDLER, G.: Comparison of radiocardiographyand conventional electrocardiography in theexercise tolerance test. Brit Heart J 29:719, 1967.

12. ARESKAG, N. H., BJORK, L., BjoRK, V. O.,HALLEN, A., AND STROM, G.: Physical work

capacity: ECG reaction to work test andcoronary angiogram in coronary artery disease.Acta Med Scand 472: 9, 1967.

13. BURKART, F., BAROLD, S., AND SOWTON, E.:Hemodynamic effects of repeated exercise.Amer J Cardiol 20: 509, 1967.

14. MATTINGLY, T. W.: Post exercise electrocardio-gram. Amer J Cardiol 9: 395, 1962.

15. HALLEN, A.: Angina pectoris, a clinical studywith special reference to surgical treatment.Acta Chir Scand (suppl 323): 43, 1964.

16. ROBINSON, B. F.: Relation of heart rate andsystolic blood pressure to the onset of painin angina pectoris. Circulation 35: 1073, 1967.

17. KEhip, G. L.: Rapid analysis of oxygen consump-tion during treadmill exercise. A.C.C. course,Instrumental Acquisition of Cardiological Data,Memorial Hospital of Long Beach, August 1 to3, 1968.

18. GORDON, E. E.: Use of energy costs in regulatingphysical activity in chronic disease. CaliforniaPublic Health Service, Berkeley, Califomia,Nov. 1957.

19. ROWELL, L. B., TAYLOR, H. L., SIMONSON, E.,AND CARLSON, W. S.: Physiologic fallacy ofadjusting for body weight in performance ofthe Master two-step test. Amer Heart J 70:461, 1965.

20. FoRD, A. B., AND HELLERSTEIN, H. K.: Energycost of the Master two-step test. JAMA 164:1868, 1957.

21. MASTER, A. M., AND ROSENFELD, I.: Two-steptest: Current status after twenty-five years.Mod Conc Cardiovasc Dis 36: 19, 1967.

22. FRIEDBERG, C. K., JAFFE, H. L., PoRDY, L., ANDCHESKY, K.: Two-step exercise program. Cir-culation 26: 1254, 1962.

23. ROBB, G. P., AND MARKs, H. H.: Latent coronaryartery disease: Determination of its presenceand severity by the exercise electrocardiogram.Amer J Cardiol 13: 603, 1964.

24. MOST, A. S., HORNSTEN, T. R., HOFER, V., ANDBRUCE, R. A.: Exercise S-T changes in healthymen. Arch Intern Med 121: 225, 1968.

Circulation, Volume XXXIX, April 1969

5u,2d2

by guest on April 17, 2018

http://circ.ahajournals.org/D

ownloaded from

GEORGE L. KEMPMYRVIN H. ELLESTAD, WILLIAM ALLEN, MAURICE C. K. WAN and

Maximal Treadmill Stress Testing for Cardiovascular Evaluation

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1969 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.39.4.517

1969;39:517-522Circulation. 

http://circ.ahajournals.org/content/39/4/517located on the World Wide Web at:

The online version of this article, along with updated information and services, is

  http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. and Rights Question and Answer

Permissionsthe Web page under Services. Further information about this process is available in thewhich permission is being requested is located, click Request Permissions in the middle column ofClearance Center, not the Editorial Office. Once the online version of the published article for

can be obtained via RightsLink, a service of the CopyrightCirculationoriginally published in Requests for permissions to reproduce figures, tables, or portions of articlesPermissions:

by guest on April 17, 2018

http://circ.ahajournals.org/D

ownloaded from