five-year follow-up of maximal treadmill stress test in asymptomatic...

Five-year Follow-up of Maximal Treadmill StressTest in Asymptomatic Men and Women

WILLIAM H. ALLEN, M.D., WILBERT S. ARONOW, M.D.,PHILLIP GOODMAN, M.D., AND PERRI STINSON, PH.D.

SUMMARY A five-year follow-up of 888 asymptomatic men and women without known coronary heart dis-ease (CHD) who had a maximal treadmill stress test (MTST) revealed a CHD incidence of 1.1% per year. Inwomen, exercise duration of 3 minutes or less by the Ellestad protocol correlated with subsequent developmentof CHD (p < 0.001), although abnormal ST-segment and R-wave responses did not. In men 40 years of age oryounger, the MTST did not correlate with subsequent CHD. In men older than 40 years, ischemic ST response(p < 0.01), an increase or no change in R wave (p < 0.01), and an exercise duration of 5 minutes or less (p <0.001) all correlated with subsequent development of CHD. Five of five men (100%) who had all three criteriadeveloped CHD within 5 years. When men older than 40 years who had all three criteria either present or ab-sent were considered, specificity was 100%. The sensitivity, specificity, predictive value of an abnormal test,and risk ratio for developing CHD within 5 years for the various MTST criteria alone and in combination aretabulated.

MAXIMAL AND NEAR-MAXIMAL stress test-ing have prognostic value in asymptomatic subjects.Studies have revealed that asymptomatic subjects withan abnormal maximal or near-maximal treadmillstress test have a higher probability of developingangina pectoris, myocardial infarction or suddendeath than asymptomatic subjects with a normalstress test.1'9We previously reported the initial results of max-

imal stress testing in 1077 asymptomatic adultswithout known heart disease."0 The present report is a5-year follow-up of this population, correlating thedevelopment of coronary heart disease (CHD) withage, sex, duration of exercise, R-wave response andelectrocardiographic response. The study is unique inthe follow-up of stress testing for women and in theprognostic evaluation of R-wave response and exerciseduration in an asymptomatic population withoutknown disease.

Materials and MethodsBetween August 1, 1973 and May 15, 1974, 1077

consecutive asymptomatic adults performed a max-imal treadmill stress test (MTST). None were onmedications that would affect the electrocardio-graphic response to exercise. None had left ventricularhypertrophy, bundle branch block or pathologic Qwaves on their resting ECG. None had clinicalevidence of pulmonary disease or vascular disease byhistory or by physical examination. No subject was in-cluded who developed a morbid arrhythmia, a conduc-

From the Medical Service, Cardiology Section, MemorialHospital, Veterans Administration Medical Center, CaliforniaState University, and the Long Beach Heart Association, LongBeach, and the University of California, Irvine, California.

Supported by the Memorial and Children's Medical CenterFoundation, Long Beach, California.

Address for correspondence: Wilbert S. Aronow, M.D., Chief,Cardiovascular Section, Veterans Administration Medical Center,Long Beach, California 90822.

Received October 9, 1979; revision accepted February 21, 1980.Circulation 62, No. 3, 1980.

tion abnormality, chest pain or any symptom sugges-tive of angina pectoris during or after the MTST.

All subjects performed the MTST at MemorialHospital of Long Beach using the Ellestad protocol."1Subjects were advised not to eat for 12 hours and notto smoke for 2 hours before the test.

Leads CM6 and V1 and a bipolar vertical lead wererecorded simultaneously with the subject at rest, sit-ting, standing, before and after hyperventilation, everyminute during exercise, at the end of exercise, andevery minute after exercise for at least 6 minutes. Theelectrocardiographic leads were calibrated before ex-ercise. Lead CM5 was continuously monitored by os-cilloscope. The subjects exercised until they reached100% of their predicted maximal heart rate,'0overwhelming fatigue, or marked dyspnea with aplateau of the heart rate.

Flat ST-segment depression of 1 mm (0.1 mV) orgreater and downsloping of the ST segment were con-sidered a positive MTST response if the ST segmentwas isoelectric at rest, after standing and afterhyperventilation. Resting ST segments were normal inmost of the subjects. Subjects with major ST-segmentchanges at rest were excluded. If there were minorchanges in the ST segment before exercise, an ad-ditional 1.5 mm of depression at 80 msec from the Jpoint was required to indicate a positive MTST. J-point depression with an upsloping ST segment of 1.5mm or greater depression at 80 msec after the J pointwas considered an equivocal MTST response.The R wave was measured from the isoelectric

baseline or PQ junction to the R-wave zenith for anaverage of six beats during the control period and im-mediately after exercise. An increase or no change inthe R wave immediately after exercise compared withcontrol was defined as an abnormal response. Adecrease in R-wave amplitude was defined as a normalresponse.

Between August 1, 1978 and May 15, 1979, follow-up was obtained on 888 of the original 1077 subjects(82.5%). Subjects and/or their physicians were con-tacted by letter and/or telephone to determinewhether the subjects had developed angina pectoris,

522

by guest on June 28, 2018http://circ.ahajournals.org/

Dow

nloaded from

http://circ.ahajournals.org/

MAXIMAL TREADMILL STRESS TEST FOLLOW-UP/Allen et al.

TABLE 1. Relation ofST Patterns After Maximal Treadmill Stress Test with Development of Coronary Heart Disease Within 5 Years

EquivocalPositive Upsloping with ST

Flat Flat Flat depression from JMTST depression depression depression point ofinterpretation > 2.0 1.5-1.9 1.0-1.4 1.5-1.9 1-1.4ST pattern Downsloping mm mm mm Total 2 mm mm mm Total Negative Z

CHD 11 2 1 1 15 1 0 2 3 30 48(15.7%) (11.1%) (11.1%) (14.3%) (14.4%) (9.1%) (0%) (10%) (6.1%) (4.1%) (5.4%)

No CHD 59 16 8 6 89 10 18 18 46 705 840(84.3%) (88.9%) (88.9%) (85.7%) (85.6%) (90.9%) (100%) (90%) (93.9%) (95.9%) (94.6%)

Abbreviations: MTST = maximal treadmill stress test; CHD = coronarv heart disease.

myocardial infarction or sudden cardiac death.Angina pectoris was diagnosed if classic angina pec-toris was present according to the subject's physician.Follow-up data were obtained for 105 of 113 subjectswith initially positive MTSTs (92.9%). Follow-up datawere obtained in 311 of 368 women (84.5%), com-pared with 577 of 709 men (81.4%).

Statistical Analysis

Data analysis involved processing a maximum of 20information bits for each of the original 1077 sub-jects. This and subsequent analyses were accom-plished using the California State University's centraltime share system as well as the CDC 3300 system atCalifornia State University, Long Beach. Computerprograms from the Statistical Package for the SocialSciences, as well as the UCLA Biomedical ComputerPrograms, which included crosstabs and contingency-table analyses, were used to analyze the data.

ResultsOf the 888 adults, 48 (5.4%) developed CHD

(angina pectoris, myocardial infarction or sudden car-diac death) over the 5-year follow-up, for an incidenceof 1.1% per year.

Table 1 indicates the percentage of subjects who didor did not develop CHD within 5 years related to thevarious types of ST-segment patterns and the degreesof ST-segment depression. The types of positive ST-segment patterns and the degrees of ST-segmentdepression shown in table 1 were equally sensitive inpredicting subsequent development of CHD. Upslop-ing patterns of ST depression were not predictive, andin subsequent tables, they will be included with thetreadmill responses interpreted as negative.

Table 2 shows the percentages of men and womenwho did or did not develop CHD within 5 years relatedto a positive or negative MTST as defined by ST-segment criteria. For women, a positive MTST by STcriteria did not correlate with subsequent developmentof CHD. For men, a positive MTST using ST criteriadid correlate with subsequent development of CHD(chi-square 15.22, p < 0.001).

Table 3 indicates the percentage of women who didor did not develop CHD within 5 years related to R-wave criteria and related to an exercise duration of 3

TABLE 2. Correlation of Positive vs Negative MaximalTreadmill Stress Test Using ST Criteria with Developmentof Coronary Heart Disease Within 5 Years in Women and inMen

Women MenPositive Negative Positive NegativeMTST MTST MTST MTST

CHD 1* 10 14t 23

(6.3%) (3.4%) (15.7%) (4.7%)NoCHD 15 285 75 465

(93.7%) (96.6%) (84.3%) (95.3%)

P = NS.tP < 0.001.Abbreviations: MTST = maximal treadmill stress test;

CHD = coronary heart disease.

minutes or less vs an exercise duration greater than 3minutes. The R-wave response did not correlate withsubsequent CHD in women. An exercise duration of 3minutes or less correlated with subsequent CHD inwomen (chi-square 31.99, p < 0.001).

ST-segment analysis was performed twice after thesubjects completed the MTST. However, analysis ofthe R-wave and exercise duration of the MTST wasperformed at follow-up. The MTST in 20 subjects, in-

TABLE 3. Correlation of Positive vs Negative MaximalTreadmill Stress Test Using R-wave and Exercise DurationCriteria with Development of Coronary Heart Disease Within5 Years in Women

R wave Exercise durationGreater

No change 3 minutes than 3or increase Decrease or less minutes

CHD 2* 8 4t 6

(5.7%) (3.0%) (30.8%) (2.1%)NoCHD 33 259 9 283

(94.3%) (97.0%) (69.2%) (97.9%)*p = NS.tp < 0.001.Abbreviation: CHD = coronary heart disease.

523


Dow

nloaded from


VOL 62, No 3, SEPTEMBER 1980

TABLE 4. Correlation of Positive vs Negative Maximal Treadmill Stress Test Using ST, R-wave and ExerciseDuration Criteria with Development of Coronary Heart Disease Within 5 Years in Men 40 Years or Younger

ST segment R wave Exercise durationNo change or 5 minutes Greater then

Positive Negative increase Decrease or less 5 minutes

CHD 0* 2 0* 2 0* 2(0%) (0.9%) (0%) (1.0%) (0%) (0.9%)

NoCHD 8 211 23 190 4 209

(100%) (99.1%) (100%) (99.0%) (100%) (99.1C%)

*p= NS.Abbreviation: CHD coronary heart disease.

cluding nine women and 11 men, could not be found.Of the nine women, one developed subsequent CHD.Six men were 40 years old or less, and five men wereolder than age 40 years; none developed subsequentCHD.

Table 4 shows the percentage of men 40 years oryounger who did or did not develop CHD within 5years related to positive vs negative ST-segmentcriteria, related to an increase or no change in R wavevs a decrease in R wave, and related to an exerciseduration of 5 minutes or less vs an exercise durationgreater than 5 minutes. In men 40 years or younger,there was no correlation between development ofCHD and ST criteria, an increase or no change in theR wave, and exercise duration of 5 minutes or less.

Table 5 indicates the percentage of men over age 40years who did or did not develop CHD within 5 yearsrelated to positive vs negative ST-segment criteria,related to an increase or no change in R wave vs adecrease in R wave, and related to an exercise dura-tion of 5 minutes or less vs an exercise duration ofgreater than 5 minutes. For men over age 40 years,significant correlations were present between subse-quent development of CHD and ST criteria (chi-square 7.20, p < 0.01), an increase or no change in theR wave (chi-square 10.62, p < 0.01), and exerciseduration of 5 minutes or less (chi-square 27.97, p <0.001).When combining either of two MTST criteria in

men over 40 years, sensitivity is enhanced (table 6). As

sensitivity is increased, specificity is decreased whenevaluating either/or combinations. When combiningtwo criteria, with both criteria being present or absent,sensitivity is decreased and specificity is increased(table 7).Any one of the three MTST criteria being present

had no greater sensitivity (64.7%, table 8) in predictingsubsequent development of CHD than either STcriteria or no change or an increase in R wave (64.7%,table 6). When comparing the presence vs the absenceof all three criteria, the specficity is 100% (table 9).Table 10 summarizes the sensitivity, specificity,predictive value of an abnormal test and risk ratio forvarious MTST criteria.

DiscussionUpsloping ST depression in a population including

subjects with known CHD has been shown to havepredictive value for future CHD events.12 However, inthe present study of asymptomatic subjects, upslopingST depression induced by a MTST did not havepredictive value for subsequent CHD. Further, in-creasing degrees of ischemic ST-segment depressioninduced by exercise did not correlate with a further in-crease in the subsequent development of CHD in thisasymptomatic group.

In women, ST-segment depression and the R-waveresponse to exercise did not correlate with the subse-quent development of CHD. Exercise duration of 3

TABLE 5. Correlation of Positive vs Negative Maximal Treadmill Stress Test Using ST, R-wave and ExerciseDuration Criteria with Development of Coronary Heart Disease Within 5 Years in Men Older Than 40 Years

ST segment R wave Exercise durationNo change or 5 minutes Greater than

Positive Negative increase Decrease or less 5 minutes

CHD 14* 20 16* 18 9+ 25

(17.3%) (7.2%) (18.6%) (6.8%) (42.9%) (7.6%)NoCHD 67 254 69 247 12 304

(82.7%) (92.8%) (81.4%) (93.2%) (57.1%) (92.4%)*p < 0.01.tP < 0.001.Abbreviation: CHD = coronary heart disease.

CIRCULATION524


Dow

nloaded from



TABLE 6. Correlation of Combining Either of Two Positive Criteria (ST, R-wave and Exercise Duration) vsAbsence of Criteria with Development of Coronary Heart Disease Within 5 Years in Men Older Than 40 Years

Duration < 5ST positive minutes or Duration > 5

ST positive ST negative or increase or ST negative increase or no minutes andor duration and duration no change in and decrease change in R decrease in R< 5 minutes > 5 minutes R wave in R wave wave wave

CHD 18* 16 22t 12 16t 18(18.9%) (6.3%) (16.2%) (5.6%) (16.8%) (7.1%)

NoCHD 77 239 114 202 79 237

(81.1%) (93.7%) (83.8%) (94.4%) (83.2%) (92.9%)

Sensitivity 52.9% 64.7% 47.1%Specificity 75.6% 63.9% 75.0%

Predictive value 18.9% 16.2% 16.8%

Risk ratio 3.0 2.9 2.4

*p < 0.001.tP < 0.01.Abbreviation: CHD = coronary heart disease.

TABLE 7. Correlation of Combining Two Positive Criteria (ST, R-wave and Exercise Duration) vs Absenceof Criteria with Development of Coronary Heart Disease Within 5 Years in Men Older Than 40 Years

Duration < 5ST positive minutes and Duration > 5

ST positive ST negative and increase ST negative increase or no minutes andand duration and duration or no change and decrease change in R decrease in R< 5 minutes > 5 minutes in R wave in R wave wave wave

CHD 5* 16 8* 12 9* 18

(71.4%) (6.3%) (26.7%) (5.6%) (81.8%) (7.1%)

NoCHD 2 239 22 202 2 237

(28.6%) (93.7%) (73.3%) (94.4%) (18.2%) (92.9%)

Sensitivity 23.8% 40% 33.3%Specificity 99.2% 86.3% 99.2%Predictive value 71.4% 26.7% 81.8%Risk ratio 11.3 4.8 11.5

*p < 0.001.Abbreviation: CHD = coronary heart disease.

TABLE 8. Correlation of Combining Any of Three PositiveCriteria (ST, R-wave and Exercise Duration) vs Absence ofCriteria with Development of Coronary Heart Disease Within5 Years in Men Older Than 40 Years

ST positive or ST negative andincrease or no change decrease in R wavein R wave or duration and duration > 5

< 5 minutes minutes

CHD 22* 12

(15.3%) (5.8%)

No CHD 122 194

(84.7%) (94.2%)

Sensitivity 64.7%; specificity 61.4%; predictive valueof a positive test 15.3%; risk ratio 2.6.

*p < 0.01.Abbreviation: CHD = coronary heart disease.

TABLE 9. Correlation of Combining A 11 Three PositiveCriteria (ST, R-wave and Exercise Duration) vs Absence ofCriteria with Development of Coronary Heart Disease Within5 Years in Men Older Than 40 Years

ST positive and ST negative andincrease or no change decrease in R wave,

in R wave and and duration > 5duration < 5 minutes minutes

CHD 5* 12

(100%) (5.8%)

No CHD 0 194

(0%) (94.2%)Sensitivity 29.4%; specificity 100%; predictive value

of a positive test 100%; risk ratio 17.2.*p < 0.001.Abbreviation: CHD = coronary heart disease.

525


Dow

nloaded from


VOL 62, No 3, SEPTEMBER 1980

TABLE 10. Significant Correlates of Maximal Treadmill Stress Test with Development of Coronary Heart DiseaseWithin 5 Years in Asymptomatic Men and Women

Predictive valueof an abnormal

Sensitivity (%) Specificity (%) test (%) Risk ratio

WomenDuration < 3 minutes 40.0 96.9 30.8 14.7

Men > 40 years oldST* 41.2 79.1 17.3 2.4Rt 47.1 78.2 18.8 2.8

Exercise durationt 26.5 96.2 42.9 5.6

ST* or exercise durationt 52.9 75.6 18.9 3.0

ST* or Rt 64.7 63.9 16.2 2.9

Rt or exercise durationt 47.1 75.0 16.8 2.4

Any of ST*, Rt, or exercisedurationt 64.7 61.4 15.3 2.6

ST* and Rt 40.0 86.3 26.7 4.8

ST* and exercise durationt 23.8 99.2 71.4 11.3

Rt and exercise durationt 33.3 99.2 81.8 11.5All of ST*, Rt and exercise

durationt 29.4 100.0 100.0 17.2

*ST segment downsloping or > 1 mm flat ST-segment depression.tR-wave increase or no change with exercise.tExercise duration of 5 minutes or less.

Sensitivity (() true positives X 100true positives + false negatives

Specificity (' true true negatives X 100true negatives + false positives

Predictive value of an true positivesabnormal test (%) = true positives + false positives X 100

Risk ratio = predictive value of an abnormal test (%) X 100[false negatives/(false negatives + true negatives)]

True positives = those with an abnormal test who developed coronary heart disease (CHD); false negatives- those with a normal test who developed CHD; true negatives = those with a normal test who did notdevelop CHD; false positives = those with an abnormal test who did not develop CHD.

minutes or less was manifested in four of 10 womenwho subsequently developed CHD, and was the onlysignificant predictor of subsequent CHD in women.However, only four of 13 (30.8%) women with thislimited exercise tolerance developed CHD. In view ofthese findings, MTSTs in asymptomatic women seemto be of limited value.Only two of 221 men age 40 years or younger

developed CHD, and none of these had ST abnormali-ties, an abnormal R-wave response, or an exerciseduration of 5 minutes or less. Therefore, in asymp-tomatic men 40 years or younger, MTSTs have verylimited value.

In asymptomatic men older than 40 years, subse-quent CHD within 5 years correlates well with an ab-normal ST-segment response, an increase or no

change in R wave, and an exercise duration of 5minutes or less. Bonoris, Greenberg, and others haveshown the predictive value of the R-wave response as

it relates to the concurrent presence of CHD in a pop-ulation that has known CHD.1'- 1 It has also beenshown that by using ST segment and/or R-wavecriteria together, sensitivity is enhanced, withspecificity and predictive value remaining high, as itrelates to the concurrent presence of known CHD.16 Inour study, sensitivity of ST criteria alone in men olderthan 40 years (41.2%) or of R-wave criteria alone(47.1%) is enhanced when interpreting a MTST aspositive when either criterion is present (64.7%). Add-ing exercise duration 5 minutes or less as a thirdcriterion for a positive MTST did not change sen-sitivity.The association between the predictive value of a

positive test and subsequent CHD within 5 years inmen over age 40 years was highest when positivity wasinterpreted as exercise duration of 5 minutes or less(42.9%), exercise duration of 5 minutes or less and anincrease or no change in R wave (81.8%), and when all

CIRCULATION526


Dow

nloaded from



criteria (exercise duration, R wave and ST segment)were present (100%). Further, when men over age 40years who had all three criteria either present or ab-sent were considered, specificity was 100%.The exercise duration data in our study relate to the

Ellestad protocol. These data could be applied toother treadmill protocols if one considered ap-proximate oxygen consumption rather than exerciseduration.

In patients with known CHD, limited exercisecapability has been highly predictive of future CHDevents17-19 and extensive coronary atherosclerosis. 19, 20When limited exercise tolerance is present in asymp-tomatic men older than 40 years, even in the absenceof ischemic ST-segment change or abnormal R-waveresponse, the possibility of underlying CHD should beconsidered. In women, exercise duration of 3 minutesor less was the only significant predictor of subsequentCHD in our study.

References1. Ellestad MH, Wan MKC: Predictive implications of stress test-

ing. Follow-up of 2700 subjects after maximum treadmill stresstesting. Circulation 51: 363, 1975

2. Bruce RA, McDonough JR: Stress testing in screening for car-diovascular disease. Bull NY Acad Med 45: 1288, 1969

3. Kattus AA, Jorgensen CR, Worden RE, Alvaro AB: ST seg-ment depression with near maximal exercise in detection ofpreclinical coronary heart disease. Circulation 44: 585, 1971

4. Aronow WS: Thirty-month follow-up of maximal treadmillstress test and double Master's test in normal subjects. Circula-tion 47: 287, 1973

5. Froelicher VF Jr, Yanowitz FG, Thompson AJ, Lancaster MC:The correlation of coronary angiography and the electro-cardiographic response to maximal treadmill testing in 76asymptomatic men. Circulation 48: 597, 1973

6. Stuart RJ Jr, Ellestad MH: Upsloping S-T segments in stresstesting. A 6-year follow up: correlation in 248 angiograms.(abstr) Circulation 50 (suppl III): II1-8, 1974

7. Aronow WS: Postexercise evaluation of ischemic heart disease

by electrocardiography, phonocardiography, apexcardiog-raphy, and systolic time intervals. In Noninvasive Methodsin Cardiology, edited by Zoneraich S. Springfield, Illinois,Charles C Thomas, 1974, p 244

8. Aronow WS, Cassidy J: Five year follow-up of double Master'stest, maximal treadmill stress test and resting and postexerciseapexcardiogram in asymptomatic persons. Circulation 52: 616,1975

9. Froelicher VF, Thomas MM, Pillow C, Lancaster MC:Epidemiologic study of asymptomatic men screened by max-imal treadmill testing for latent coronary artery disease. Am JCardiol 34: 770, 1974

10. Allen WH, Aronow WS, DeCristofaro D: Treadmill exercisetesting in mass screening for coronary risk factors. Cathet Car-diovasc Diagn 2: 39, 1976

11. Ellestad MH, Allen WH, Wan MKC, Kemp GL: Maximaltreadmill stress testing for cardiovascular evaluation. Circula-tion 39: 517, 1969

12. Stuart RJ, Ellestad MH: Upsloping S-T segments in exercisestress testing. Am J Cardiol 37: 19, 1976

13. Bonoris P, Greenberg P, Castellanet M, Ellestad MH: Predic-tive value of R-wave amplitude changes in treadmill stresstesting. (abstr) Circulation 56 (suppl III): III-197, 1977

14. Bonoris PE, Greenberg PS, Castellanet MJ, Ellestad MH:Significance of changes in R wave amplitude during treadmilltesting: angiographic correlation. Am J Cardiol 41: 846, 1978

15. Bonoris PE, Greenberg PS, Christison GW, Castellanet MJ,Ellestad MH: Evaluation of R wave amplitude changes versusST segment depression in stress testing. Circulation 57: 904,1978

16. Greenberg PS, Friscia DA, Ellestad MH: Predictive accuracyof Q-X/Q-T ratio, Q-T, interval, ST depression and R waveamplitude during stress testing. Am J Cardiol 44: 18, 1979

17. Bruce RA, DeRouen T, Peterson DR, Irving JB, Chinn N,Blake B, Hofer V: Noninvasive predictors of sudden cardiacdeath in men with coronary disease. Predictive value of max-imal stress testing. Am J Cardiol 39: 833, 1977

18. Hammermeister KE, DeRouen TA, Dodge HT: Variablespredictive of survival in patients with coronary disease. Circula-tion 59: 421, 1979

19. McNeer JF, Margolis JR, Lee KL, Kisslo JA, Peter RH, KongY, Behar VS, Wallace AG, McCants CB, Rosati RA: The roleof the exercise test in the evaluation of patients for ischemicheart disease. Circulation 57: 64, 1978

20. Goldschlager N, Selzer A, Cohn K: Treadmill stress tests as in-dicators of presence and severity of coronary artery disease.Ann Intern Med 85: 277, 1976

527


Dow

nloaded from


W H Allen, W S Aronow, P Goodman and P StinsonFive-year follow-up of maximal treadmill stress test in asymptomatic men and women.

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1980 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.62.3.522

1980;62:522-527Circulation.

http://circ.ahajournals.org/content/62/3/522the World Wide Web at:

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

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. Permissions and Rights Question and Answer information about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. FurtherEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculationpublished in Requests for permissions to reproduce figures, tables, or portions of articles originallyPermissions:


Dow

nloaded from

http://circ.ahajournals.org/content/62/3/522

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

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


five-year follow-up of maximal treadmill stress test in asymptomatic...

Documents