heart journal, electrocardiographic abnormalities during

British Heart_Journal, 197I, 33, 87I-877.

Electrocardiographic abnormalities duringexercise 3 weeks to I8 months afteranterior myocardial infarction

J.-H. Atterhog, L.-G. Ekelund, and L. KaijserFrom the Department of Clinical Physiology, Karolinska Sjukhuset, Stockholm, Sweden

The electrocardiographic abnormalities produced by exercise at varying time intervals after ananterior myocardial infarction were studied in 12 patients. Three weeks after the infarction,exercise induced an obvious rise in the ST segment in leads over the infarcted area in IO patients.In the course of the ensuing months this reaction to exercise gradually disappeared, and in somesubjects exercise then produced ST segment depression. The ST rise produced by exercise inthe early phase after an infarction is interpreted as a sign of anoxia in the infarcted zone, and therate at which this pattern resolves may be ofprognostic value. These changes in the ST-T waveproduced by exercise were not found in left praecordial leads distant from the site of infarction.The T wave amplitude wasfound to be increased during exercise at all times after infarction, butthis change is nonspecific.

The effect of exercise on the electrocardio-gram is used extensively as a method of diag-nosis of the presence of ischaemic heartdisease. However, little attention has beenpaid to the effects of exercise on the electro-cardiogram during the period after acute in-farction. The purpose of the present study wastherefore to examine the electrocardiographicpattern and sequential effects of exercise dur-ing the time course in a group of patients atintervals from 3 weeks to I8 months after theonset of acute myocardial infarction. Further-more, it was considered to be of interest to seeif the changes might be of prognostic value.

PatientsSince I968 an exercise test performed 3 weeksafter the onset of symptoms of acute myocardialinfarction has been a routine part of the manage-ment of all patients admitted to KarolinskaSjukhuset with this diagnosis, unless cardiacfailure, arrhythmias, or other contraindicationsare present at this time. Tests were repeated atvarying intervals up to I8 months in I50 patientswho were willing to attend the hospital for thispurpose. All patients with anterior myocardialinfarction who had never received digitalis andwho had had two or more tests performed duringthe period were examined. Patients with anteriorinfarction were chosen because this area is wellrepresented in the conventional leads duringexercise (Scandinavian Committee on ECG

Received 14 January 1971.

Classification, I967). Twelve patients with a meanage of 58 years fulfilled these criteria. They hadall given a typical history of acute myocardialinfarction. Ten patients had had abnormal Qwaves (>004 sec) accompanied by ST-T waverise in leads CR2-4 and raised levels ofserum enzymes (CPK, GOT, or LDH). Twopatients had had ST-T wave rises and raisedlevels of serum enzymes without the developmentof abnormal Q waves. The clinical data are sum-marized in the Table.

MethodsIn addition to the test at 3 weeks, 9 of the 12 hadtests between 2 and 4 months, 9 between 5 and 6months, and 6 a year or more after infarction.Exercise was performed in the sitting positionon an electrodynamically braked cycle ergometer(Holmgren and Mattsson, 1954) with stepwiseincreasing loads (Sj6strand, 1947; Wahlund,1948). The first load was I00-300 kpm/min, de-pending on the amount of work the patient wasexpected to tolerate, and the load was increasedby I00-300 kpm/min every sixth minute. On thefirst occasion after the infarction, the exercise wascontinued until the heart rate had reached I20 to130 beats a minute. In no subject was permanentdamage to the myocardium induced by the worktest, as evidenced by electrocardiograms over thenext few days. In later exercise tests the heartrate was allowed to increase to about I5o beats aminute unless frequent ectopic beats or anginaoccurred.The electrocardiogram was recorded with an

ink-jet electrocardiograph (Elema, Mingograf 8I)

on April 30, 2022 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.33.6.871 on 1 Novem

ber 1971. Dow

nloaded from

http://heart.bmj.com/

872 Atterhog, Ekelund, and Kaijser

TABLE Clinical data of 12 patients with anterior myocardial infarction examined repeatedlywith exercise electrocardiogram

Case Sex Age Duration of Other diseases Type of Max. GOT ComplicationsNo. (yr) coronary disease QRS values*

previous to infarction complex (units)

I M 62 - Q > o0o4 sec <100 -2 M 54 Angina pectoris Moderate hyper- ,, I00-200

5 yr tension 5 yr3 M 6i Slight angina ,, > 200

pectoris 2 wk4 M 66 Lateral infarc- ,, ioo-2oo Atrial fibrillation on admission, fre-

tion 3 wk quent ventricular ectopic beatsearlier first days in hospital

5 M 58 ,, Gout ? ,, > 2006 F 46 Moderate hyper- No Q wave I00-200 -

cholesterolaemia7 M 6i Angina pectoris ,, <100 Heart rate <5o beats/min on

i mth admission, frequent ventricularectopic beats first days in hospital

8 M 5-7 Q>oo4sec < I009 M 49 Abnormal glucose ,, I00-200 -

tolerance testI0 F 69 Angina pectoris Moderate ,, > 200 -

3 dy hypertensionII M 54 Angina pectoris Abnormal glucose ,, < I00

4 yr tolerance testI2 M 58 - Hypoplastic ,, I00-200 -

anaemia ofmoderate degree

*Normal value < 25 units.

at a paper speed of 50 mm/sec and a calibrationof i mV equal to I0 mm amplitude. Six chest-head leads (CH leads) were used (Sj6strand, I95I;Holmgren and Strandell, I96I). The abnor-malities were examined in the lead showing thedeepest Q wave corresponding approximately tothe centre of the infarction (lead CH2 in 7 sub-jects, CH4 in 5 subjects) and also in CH7, repre-senting the non-infarcted lateral part of the leftventricle. Alterations of the ST segment from theisoelectric line were measured 20 msec after theS-ST segment junction. The T wave was mea-sured at its highest positive value and, when di-phasic or negative, also at its most negative point.The average of 3 consecutive heart cycles wasused.

ResultsWork performance In the first exercisetest, when the work load was limited to give aheart rate of 120-I30 beats/min the averagefinal load was 370 kpm/min (range iio-600),with an average final heart rate of I24 beats/min (range 95-I42, the highest rate dependingon an obvious increase ofthe heart rate betweentwo loads). Between 5 and 6 months after theinfarction the average final load was 6ookpm/min (range 400-800) and the average

final heart rate 130 beats/min (range II5-144).Those patients examined 5 to 6 months afterthe infarction had increased their workingcapacity at a heart rate of I30 (W130) from 400

kpm/min (range 50-625) at 3 weeks to 590kpm/min (range 475-925).

Angina pectoris Of the I2 patients, 3stopped exercising because of angina pectorisat the 3-week and subsequent tests. The re-maimnng 9 were able to reach the heart ratelimit at the 3-week test, but 2 of these devel-

FIG. i Relation between heart rate (beatslmin) and the ST level in the 12 subjects duringexercise 3 weeks after acute myocardial infarc-tion. ST level measured in a lead correspon-ding to the centre of the infarcted area.Numbers refer to the case numbers of thepatients.

ST level, mm6 12

5

4 39

3'2 12 110

2

-1

-2-

I I I I I I i .

60 80 100 120 140 160 HR



j.com/

Br H


ber 1971. Dow

nloaded from


Electrocardiographic abnormalities during exercise 873

FIG. 2 Relation between heart rate duringexercise and T wave amplitude (measured inthe same lead as ini Fig. I). Patients withdiphasic T waves are represented by two linesindicating the most positive and the mostnegative part of the T wave (x= positive,o = negative).

oped angina during a later test when achievinghigher heart rates.

Electrocardiographic abnormalities 3weeks after infarction The ST segmentin the anterior wall leads (CH2-4) had anaverage level of + I.3 mm (range - i to + 3mm). During exercise there was a continuousincrease in the ST segment rise withrising heart rate in I0 subjects (average + i 6mm, range + o05 to + 4 mm) (Fig. I). Of theremaining 2 subjects, one had isoelectric STsegment at rest and no Q wave. The T waveamplitude increased in all subjects. In thosesubjects with a negative T wave component,this decreased and in most cases became posi-tive (Fig. 2). At higher heart rates the ST-Twave configuration resembled that during anacute infarction in all but 2 patients. Therewere no differences between patients with andwithout chest pain during exercise. No chan-ges in the QRS configuration occurred duringthe test. Two examples of the electrocardio-graphic abnormalities are shown in Fig. 3and 4.

In lead CH7, 9 subjects showed slight STsegment depression of the type considered tobe normal in relation to the increased heartrate (Sj6strand, I950). Three subjects devel-oped pronounced ST-T changes consideredtypical of coronary insufficiency as describedby, among others, Kaijser (i966) and Areskoget al. (i967) (Fig. 5). Angina pectoris devel-oped in one patient with abnormal ST-Tchanges in CH7 and in 2 of the patients with-out such changes.

Electrocardiographic abnormalities 2 to4 months after infarction At rest, 2 ofthe 9 patients tested had an ST level of o to+o5 mm, and 7 had an ST level of +I to+ i 5 mm in leads CH2-4. During exercise 2patients had a considerable further increasein ST level with increasing heart rate ( + 3 to

FIG. 3 Case 12, transmural infarction.Leads: CH4R-CH7 resting electrocardiogramon the left. Upper part: on exercise 3 weeksafter infarction. Lower part: on exercise 6weeks after infarction.

T amplitude, mm98 12

7/6 - 9

54-3 -

1 1C0 z3

-2 - 2o So-3 e ..

-4 -

-5-6-7

~~~12o 4

-8-7-9-10 6

50 60 70 80 90 160 110 120 130 l40

REST

:V 4'

V1n:.. ... ....:...... .§.,> ..

_:X,

... #. ......................

FIR 74: : .... . . ..... : .: . .:. . ... ..... .. . . ..... s .. .. ... .... ....... . ............. ..... ... . ....... ..... .. . ,; ... ... ..

2,...... V ....... .. .. .... V. .T: f.5(. ...

*.... :..: .... .. :. f- ..... . | . ...... L ...... .. f ;..HR 77

EXERCISE_S \V

HR 110 HR 140

HR 92 HR133 HR 15O

150 HR

.e on April 30, 2022 by guest. P


j.com/

Br H


ber 1971. Dow

nloaded from



FIG. 4 Case 7, subendocardial infarction.Leads: CH4R-CH7 resting electrocardiogramon the left. Upper part: on exercise 3 weeksafter infarction. Lower part: on exercise 9weeks after infarction.

+ 4 mm), 3 had a slight increase (+ o05 to+I-5 mm), and in 2 the ST segment re-mained at the same level. Two patients hada I to I.5 mm depression of the ST level.Two examples of the electrocardiographicabnormalities are shown in Fig. 3 and 4.

Electrocardiographic abnormalities 5 to6 months after infarction At rest, 4 ofthe 9 patients tested had an ST segment levelof o to + I mm, and 5 had an ST segmentlevel of + I 5 to + 2 mm in leads CH2-4.During exercise, 3 patients had a further STsegment increase of +o 5 to +3 mm; in 2the ST segment remained unchanged; and in4 subjects there was a depression of o 5 to 2mm (Fig. 6). The T wave amplitude at restbecame more positive compared to the 3-weektest, and increased during exercise as it haddone during the 2 to 4 month test (Fig. 7).

Electrocardiographic abnormalities 12to i8 months after infarction At rest, 3of the 6 patients studied had an ST segmentlevel of +o5 to + I mm and 3 an ST seg-ment level of +I5 mm in leads CH2-4.During exercise 3 patients showed an increaseof the ST level (+i to +2-5 mm) and 3 adepression of the ST segment (i to 2 mm).

Time course of ST-T wave changes Thetime course of the change in ST segment re-sponse to exercise is summarized in Fig. 8.The greatest decrease in the exercise-inducedST rise appears to occur during the first2 months. In no patient in whom an exercisetest was performed in the second month orlater after the infarction was there ST-T seg-ment deviation of the configuration seen in anacute infarction.The T wave amplitude during exercise at

the same heart rate but at different times afterthe infarction is summarized in Fig. 9. Theamplitude became more positive during thefirst 3 months and then appeared to remainat this level.During the early months, the QRS con-

FIG. 5 Relation between heart rate and STand T level in lead CH7 during exercise 3weeks after infarction. Recordings madesimultaneously with those presented in Fig. i.

REST

83;

Qr

.. ~

i;A~

HRt

EXERCISE

s.~~~~~~~~r^AAs 100%I

V

HR 84 HRIO HR125

* ,.

HR7S~~ HS9 'HSII2

ST level, mm3-2-1 - l0- W 1

-1-2 0

-2--3 2

60 80 100 120 140 160 HR

T amplitude, mm65 s4 75 7493 1

2 24 0

1 2. a 10S 6

0-12

-1-2

60 80 100 120 140 160 HR

Sm.



j.com/

Br H


ber 1971. Dow

nloaded from



ST5,4-3-2-1 -

0 -

-1--2-

level, mm T amplitt11109 -

I I I I I I I a I I- -I

60 80 100 120 140 HR

FIG. 6 Relation between heart rate and STlevel during exercise 5 to 6 months after theinfarction. ST level measured in a leadcorresponding to the centre of the infarction.

figuration changed in 3 patients with anappearance of a small R wave.

8-7 -

6-5 .

4-3-2-1*0*

-.1

ujde, mm 3

GM ~~~~~~~~6

3o

50 60 70 80 90 100 110 120 130 140 150 HR

FIG. 7 Relation between heart rate and Twave amplitude during exercise 5 to 6 monthsafter the infarction (measured in the same leadas in Fig. 6).

DiscussionST segment Reversible ST increases re-sembling those recorded in an acute infarctionhave been described earlier in connexion withattacks of angina pectoris (Wilson and John-ston, I94I; Randles and Fradkin, 1948;Myers and Talmers, I955; Prinzmetal et al.,I959). Bellet et al. (I962) found that insome patients with angina pectoris ST risecould be induced by exercise. In postinfarc-tion patients, Soderholm et al. (I962) occa-sionally found ST rise during exercise.Ekelund and Strandell (I960) called atten-tion to the occurrence of ST rise duringexercise when studying the electrocardio-graphic reaction 6 months after infarction.However, the electrocardiographic reactionduring exercise in the first few weeks after aninfarction and the time course of the change inexercise electrocardiographic pattern in thepostinfarction period have not been describedpreviously.The underlying cause of the ST rise

seen in the early weeks after an acute infarc-tion is not known. The probable explanationis that the partially ischaemic zone round theinfarction becomes anoxic during exerciseand, in consequence, ST rise develops,similar to that seen in an acute infarction.In later tests exercise no longer produces theappearance of an acute infarction. During thisperiod the hypoxic cells have either recovereddue to improved collateral blood flow or havebeen replaced by fibrous tissue. Thus a com-plete demarcation between impaired and nor-

FIG. 8 Change in ST level with exercise atdifferent times after myocardial infarction(measured in a lead corresponding to the centreof the infarction). ST level measured in eachindividual at the same heart rate ( ± 5 beats)in each exercise test. Mean heart rate 117beats/min (range 90o-40). (One patient isexcluded because of obvious variation in heartrate during different exercise tests.)

ST level, mm6- 125

5-

4- 3441

2-

11 ~ 11-1

-2 6

0 1 i ~3 4 5 6 7 8 12 46 8 months



j.com/

Br H


ber 1971. Dow

nloaded from



mal tissue has developed and then the patho-physiological basis for the conspicuous STrise produced by exercise no longer exists.Formation of granulation tissue is normallycomplete throughout the infarcted area 3 to 4weeks after the signs of infarction and changesto a less cellular and less vascular granulationtissue during the next 2 to 3 months. Afterabout 3 months a fully fibrous scar has de-veloped (Mallory, White, and Salcedo-Salgar,I939). Thus the time course for the change inthe electrocardiogram responses to exerciseseems to parallel the morphological changesof healing. The changing pattern of the STreaction during exercise may thus be of prog-nostic value as a possible indication of thecondition of the injured myocardium.

T wave During the months after infarctionthe T wave inversion in the resting electro-cardiogram may partly or totally resolve(Bohning and Katz, I935; Langendorf andPick, 1938). The increased positive amplitudeof a negative or low T wave over the infarctedarea induced by exercise during the earlyweeks after an infarction and later is not speci-fic but is seen after a large variety of injuriesto the myocardium (Fritzlar, I940; Klemola,I942; Bengtsson, I957; Rihl, Huttmann, andSpiegl, 1935; Stokes, I946; Lepeschkin, I95I;Thoren, I964).

CH7 The electrocardiographic changes seenin the lateral lead CH7 during exercise 3weeks after the infarction (Fig. 5) indicatethat an electrocardiogram in the neighbour-hood of the infarction is not necessarily dis-turbed by the changes occurring in the in-farcted area. The ST-T wave changes duringexercise may be normal or show the patterntypical of coronary insufficiency depending onthe blood supply in that region in relation tothe work load. The changes in CH7 cannotjust be reflecting what is occurring in the in-farcted area as they remain in later exercisetests when the anterior electrocardiographicabnormalities have another pattern. Conse-quently, a correct electrocardiographic diagno-sis of coronary insufficiency from the exercisetest may still be obtainable.

ReferencesAreskog, N.-H., Bjork, L., Bjork, V. O., Hallen, A.,

and Str6m, G. (I967). Physical work capacity, Ecgreaction to work test and coronary angiogram incoronary artery disease. Acta Medica Scandinavica,SuppI. 472, p. 9.

Bellet, S., Eliakim, M., Deliyiannis, S., and LaVan, D.(I962). Radioelectrocardiography during exercisein patients with angina pectoris. Circulation, 25, 5.

Bengtsson, E. (I957). Electrocardiographic studies inpatients with abnormalities in serial examinations

T amplitude, mm10

9-

8- 1\7-

6 S32

44- S

3- S

2- 7

II6

00

2 -

0 1 2 3 4 5 6 7

FIG. 9 Change in T wave amplitude withexercise at different times aftermyocardialinfarc-tion,from the same leads, loads, and exercisetests as in Fig. 8.

with standard leads during acute infectious diseases.II. Chest and extremity leads during and afterexercise. Acta Medica Scandinavica, 159, 411.

Bohning, A., and Katz, L. N. (I935). The four-leadelectrocardiogram in coronary sclerosis. A studyof a series of consecutive patients. American3Journalof the Medical Sciences, I89, 833.

Ekelund, L.-G., and Strandell, T. (I960). EKG-reaktionen under och efter arbete hos man medtidigare genomgangen hjartinfarkt. Report given atthe meeting of the Swedish Society for ClinicalPhysiology.

Fritzlar, G. (1940). Das Verhalten der Nachschwan-kung im Belastungselektrokardiogramm beischweren Myokardschaden und Kammerextra-systolen. Zeitschrift fur Kreislaufforschung, 32, 409.

Holmgren, A., and Mattsson, K. H. (I954). A new

ergometer with constant work load at varyingpedalling rate. Scandinavian3Journal of Clinical andLaboratory Investigation, 6, I37.

Holmgren, A., and Strandell, T. (I96I). On the use ofchest-head leads for recording of electrocardio-gram during exercise. Acta Medica Scandinavica,I69, 57.

Kaijser, L. (i966). EKG-forandringar vid koronar-insufficiens som funktion av arbetsintensitet ochduration. Lakartidningen, 63, 3340.

Klemola, E. (I942). Untersuchungen uber das Belast-ungselektrokardiogramm bei Myokardschadigun-gen nach akuten Infektionskrankheiten. ActaSocietatis Medicorum Fennicae "Duodecim ", 32, 59.

Langendorf, R., and Pick, A. (I938). Zur Diagnose desMyokardinfarktes mit Hilfe von Brustwandableit-ungen. Acta Medica Scandinavica, 96, 8o.

Lepeschkin, E. (i95i). Modern Electrocardiography.Vol. I: The P-Q-R-S-T-U complex. Williams andWilkins, Baltimore.

8 10 12 14 16 18 months



j.com/

Br H


ber 1971. Dow

nloaded from



Mallory, G. K., White, P. D., and Salcedo-Salgar, J.(I939). The speed of healing of myocardial infarc-tion. A study of the pathologic anatomy in seventy-two cases. American HeartJournal, i8, 647.

Myers, G. B., and Talmers, F. N. (i955). The electro-cardiographic diagnosis of acute myocardial ische-mia. Annals of Internal Medicine, 43, 36I.

Prinzmetal, M., Kennamer, R., Merliss, R., Wade, T.,and Bor, N. (I959). Angina pectoris. I. A variantform of angina pectoris. American Journal ofMedicine, 27, 375.

Randles, F. S., and Fradkin, N. F. (I948). Electro-cardiographic alterations resembling those pro-duced by myocardial infarction observed during aspontaneous attack of angina pectoris. Annals ofInternal Medicine, 28, 671.

Rihl, J., Huttmann, A., and Spiegl, E. (1935). Uberdas Arbeitselektrokardiogramm. Zeitschrift furKreislaufforschung, 27, 659.

Scandinavian Committee on ECG Classification (I967).The 'Minnesota Code' for ECG classification.Adaptation to CR leads and modification of thecode for ECGs recorded during and after exercise.Acta Medica Scandinavica, Suppl. 48I.

Sj6strand, T. (I947). Changes in the respiratory organsof workmen at an ore smelting works. Acta MedicaScandinavica, Suppl. I96, p. 687.

Sjostrand, T. (1950). The relationship between theheart frequency and the S-T level of the electro-cardiogram. Acta Medica Scandinavica, 138, 20I.

Sj6strand, T. (I95i). The electrocardiographic workand hypoxemia tests. Scandinavian Journal ofClinical and Laboratory Investigation, 3, I.

Soderholm, B. Thulesius, O., Heyman, F., Malm-crona, R., and Bj6rntorp, P. (I962). Myocardialinfarction in the younger age groups. III. Follow-up observations with special reference to exercisetolerance tests. Acta Medica Scandinavica, 172, 585.

Stokes, W. (1946). The effect of nitrite and exerciseon the inverted T wave. British Heart Journal, 8,62.

Thoren, C. (I964). Cardiomyopathy in Friedreich'sataxia. With studies of cardiovascular and respira-tory function. Acta Paediatrica, 53, SuppI. I53.

Wahlund, H. (I948). Determination of the physicalworking capacity. A physiological and clinical studywith special reference to standardization of cardio-pulmonary functional tests. Acta Medica Scandin-avica, Suppl. 2I5.

Wilson, F. N., and Johnston, F. D. (I94i). The occur-rence in angina pectoris of electrocardiographicchanges similar in magnitude and in kind to thoseproduced by myocardial infarction. American HeartJournal, 22, 64.



j.com/

Br H


ber 1971. Dow

nloaded from


heart journal, electrocardiographic abnormalities during

Documents