Br HeartJ1 1984; 51: 77-83

Electrocardiographic changes after myocardialinfarction as indicators of deranged regional leftventricular wall motionA serialM mode echocardiographic mapping study

KAJ LINDVALL, NINA REHNQVISTFrom the Department ofMedicine, Division of Cardiology, Karolinska Institute, Danderyd Hospital, Danderyd,Sweden

summARY Electrocardiographic and echocardiographic findings were compared in 44 patients with afirst transmural infarction. Each patient was investigated on days 1, 2, 10, and 360. The electrocar-diogram was classified according to QRS and ST segment changes. Local left ventricular functionwas determined from mean systolic wall velocity measurements by an M mode echocardiographicmapping technique in I0 of 16 segments suitable also for electrocardiographic evaluation. Meansystolic wall velocity was corrected for differences in anterior and inferior wall motion.Wall motion was normal in segments without QRS or ST changes throughout the study. All

segments with QRS or ST changes showed significantly lower corrected systolic wall velocity valuesdunrng the acute stage. Segments with ST depression, alone or in combination with a minor Q wave,had corrected mean systolic wall velocity values similar to those of normal segments after one year.Segments with major Q waves and all segments with ST elevation showed reduced corrected meansystolic wall velocity values throughout the study. Segments with ST elevation, irrespective of Qwaves, showed the most severely reduced wall motion with significanty lower corrected mean valuesthan segments with minor or major Q waves without ST elevation on days 10 and 360.Thus when electrocardiograms are used for defining local left ventricular function, consideration

must be given to the phase of illness, .QRS morphology, and presence of ST segment elevation.

Transmural myocardial infarction is characterised bychanges in the QRS complex, which are preceded bytransient ST segment elevation indicating ischaemia.'Persistent ST elevation, on the other hand, may indi-cate left ventricular dyskinesia.3 By contrast, suben-docardial necrosis is accompanied by non-specific STsegment changes and generally not by QRSchanges.14 Knowledge of wall motion disturbances inthese cases is incomplete.

Myocardial necrosis causes wall motion distur-bances which can be visualised with left ventricularcineangiography- 6 and echocardiography.7-9 Duringan ischaemic attack left ventricular dysfunction simi-lar to that found during myocardial infarction occurstransiently at the time of pain and changes on theelectrocardiogram. 1 0-12

Previous studies on echocardiographic and elec-trocardiographic findings in acute myocardial infarc-

Accepted for publication 26 July 1983

tion are mostly limited to single comparisons in thelater stages of the illness. In our study repeatedechocardiographic analyses in different stages of thedisease were performed to consider the relation 4et-ween certain QRS and ST segment changes and s3pg-mental wall motion dysfunction. We have also inves-tigated whether these electrocardiographic changes,indicative of either transmural or subendocardialinfarction, carry different information about regionalwall motion in relation to time.

Patients and methods

Consecutive patients with electrocardiographic signssuggesting a transmural myocardial infarction onadmission to our coronary care unit were investigatedon the day of admission (day 1) and on days 2, 10, and360. The following features excluded patients frominclusion in the study: a delay longer than 48 h; ahistory or electrocardiographic signs of previous

77

on October 20, 2021 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.51.1.77 on 1 January 1984. Dow

nloaded from

78

Table 1 Pertinent data firm the study group

Day of investigation

1 2 10 360

No of patients* 44 42 39 27

Reasons for withdrawalComplete heart block - 1 1 1

Left bundle branch block - - 1 1

Reinfarction - - - 2Died - 1 3 11Not available for investigation - - - 2

TreatmentDigitalis 9 11 13 11Diuretics 31 28 22 14Beta blocking agents 6 6 6 7None of above drugs 8 9 11 7

*33 men, 11 women; mean age 62± 11 years.

myocardial infarction; valvular heart disease; leftbundle branch block; complete heart block; orpacemaker treatment. Patients who requiredpacemaker treatment, who developed left bundlebranch block, or had a reinfarction were excludedfrom further analyses. This left 47 patients for inves-tigation, but in 3 of these unacceptable echocardiog-rams from more than six of 16 left ventricular seg-ments were obtained and they were thereforeexcluded. Thirty three men and 11 women with amean age of 62 (range 31-82) years were thereforestudied, and Table 1 gives the data including reasonsfor withdrawal and their treatment.

ECHOCARDIOGRAPHYEchocardiograms were obtained with an Organon-Technica echocardiovisor and recorded on a Honey-well 1856 fibreoptic line scan recorder. Two 2*25MHz collimated transducers (D= 13 mm) focused oneither 7-5 or 4 cm were used.The M mode investigation was guided by a preced-

ing cross sectional survey of left ventricular anatomy.The M mode transducer was positioned on theanterior chest wall in electrocardiographic positionsV2, V4, and V5 as well as one subxiphoidal and twopositions on the high anterolateral chest wall. Withthis multiple positioning 16 separate left ventricularsegments could be studied (Fig. 1).9 11-13 Segmentalmean systolic wall velocity from at least -two beats wascalculated as shown in Fig. 2.To obtain comparable values of systolic wall veloc-

ity from anterior and inferior segments a correctionwas made. Data from previously investigated normalsubjects, and from other reports, show that valuesfrom inferior wall segments are higher than thosefrom the anterior wall.91415 This difference mayreflect cardiac pendulum motion rather than true dif-ferences in contractile behaviour.16 The different cor-rection values for each segment were obtained from

Lindvall, Rehnqvist

Fig. 1 Illustration of the four recording planes by schematicsections ofthe left ventrick and the segment model including eightbasal (la to 8a) and 8 mid-ventclar(lb to 8b) segments. Wallmotion from segments marked with thick lines were usedforcompanson with ekltrocardiography. B, basal; M,mid-venricular; A, apex; Ant, antrior; Lat, lateral; Post,posterior; Inf, inferior; Sept, septal.

our previous investigation of normal subjects.9 Meanvalues of systolic wall velocity from anatomicallyopposite segments were measured and the differencefor each pair halved. By adding the halved difference,the correction value, to the anterior segment and sub-tracting it from the opposite inferior segment the dif-ferent segments show comparable corrected values.The numerical values of the obtained differences havebeen used in the present study. The corrected valuesgiven are the measured mean values plus or minus thecorrection value for each segment. Table 2 shows thedata from normal subjects with measured values, cor-rection values, and corrected values.

^'N

t. I

F

on October 20, 2021 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.51.1.77 on 1 January 1984. Dow

nloaded from

Electrocardiographic changes after myocardial infarction

_ ....

.. _'....

s,s, "sVmean5

LV

% e. 1,...-E_.....f Xp vm..eanpW:S

ELECTROCARDIOGRAPHYConventional 12 lead electrocardiograms wererecorded on Siemens-Elema Mingograph immediatelybefore each echocardiographic examiation. Accord-ing to a previous study based on postmortem findingsthe conventional electrocardiogram allows for infor-mation on 10 of the 16 segments; namely, la, 2b, 3b,4a, 4b, 5a, 5b, 6b, 8a, and 8b'7 (Fig. 1). Reciprocalelectrocardiographic signs of posterior or septalinfarctions were not evaluated. The QRS and STchanges were graded by two independent observerswithout knowing the echocardiographic data. Weconsidered ST segment depression to be present if thedownward displacement exceeded 0. 1 mV and eleva-tion to be present if the upward displacementexceeded 0*2 mV. Major Q waves were considered tobe present if the Q wave was more than 0.2 mV deepand was larger than the R wave or QS complexes.

Minor Q waves were considered to be present if the Qwave was more than 0-2 mV deep and 0x04 s wide or if

Fig. 2 M mode recordingfrom a

parastemal recording VositioniUustrating measurementsofmean systolic waU velocity (Vmeanmmls) in the septal (S) and theposterior wall (PW) segments la andSa. RV, right ventricle; LV, leftventricle.

the R wave amplitude was reduced by 0*2 mV or lessthan 50%/o of the normal value for age (Table 3) or ifthe peak to peak amplitude

was less than 0.5 mV (leads aVL, aVF, I, II, and III),Q wave being less than R wave.

The procedure used was based on studies byAskenazi et al.I8 and the Minnesota code formyocardial infarction.'920 Each recording positionwas marked with ink in order to secure constantrecording sites during hospital admission.

STATISTICSConventional statistical methods were used. Resultsare expressed as mean ±+1 SD. Student's t test wasused to evaluate differences between groups, and the5% level was used for determining significance.

Table 2 Left ventricular systolic mean wall velocity (Vmean) without and with correction (Vmeand in healthy controls (n=37) ofsimilar age and sex to the study group

Antrior segmet8a 8b la lb 2a 2b 3a 3b

Vmean (mm/s) 16 18 17 21 19 19 17 17Correction +12 +115 +12 +8-5 +13.5 +105 +145 +11Vmeanc 28 28-5 29 29 5 32-5 29-5 31-5 28

Poserior segments

4a 4b Sa Sb 6a 6b 7a 7b

Vmean (mm/s) 40 39 41 38 46 40 46 39Correction -12 -11-5 -12 -8.5 -13.5 -10-5 -145 -11Vmeanc 28 28.5 29 29 5 32 5 29-5 31-5 28

79

..,-ftil, 1-1/1 '.

10

on October 20, 2021 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.51.1.77 on 1 January 1984. Dow

nloaded from

80

Table 3 Normal R wave amplitudes according to age

Age (years)

40-49 50-59 60

V2 3.4 40 7-0V4 17.0 12-9 209V5 13.4 12.5 15-4

Results

Echocardiograms were adequate at the first emina-tion in 89% of segments in the 44 patients. The cor-responding figures for the exmination 24 h later andon days 10 and 360 were 90% in 42, 83% in 39, and87% in 27 respectively. The reproducibility of twoblind observations from the same recordings and bet-ween two recordings was analysed in 15 patients.Significant correlations -(r=0.90-0.95) were obtainedfor all positions, with SEE between 4.90 parasternallyand 8-1 in V5. The highest deviation between twomeasurements was in the range 10-44 mm/s.

ELECTROCARDIOGRAPHICALLY NORMALSEGMENTSSegments without QRS and ST changes showed cor-rected mean systolic wall velocities within the normalrange at the first examination. Similar values were alsorecorded on days 2, 10, and 360 (Table 4, Fig. 3 and4).

ELECTROCARDIOGRAPHICALLY ABNORMALSEGMENTSSegments without ST elevation v normal (Table 4; Fig.3)-Segments with isolated ST depression hadsignificantly lower (p<0-001) corrected mean systolic

Lindvall, Rehnqvist

30-

,_

EE

8

8 20-

10"

o..............b Normal

b

.. ST depression* Minor Q waveb changes

:C.. ..a

a . .

a

a .. Major Q wve

..- acd changes

a ..--acd

ac

1 2 10 360Days after acute myocardial infarction

Fig. 3 Corrected mean systolic waU velocity in segmentsgrouped according to electrocardiography into normal, STdepression, minor or major Q wave changes. a=sigmficantlylower than electrocardiographicalUy normal segments on the sameday; b =significantly higher than on day 1; c =significant4y lower

than segments with isolatedSTdepression; d =significantly lowerthan segments with minor Q wave charges.

wall velocities on days 1 and 2. This was followed byan improvement with significantly (p<0.01) highervalues after day 1. Segments with minor Q waves hadreduced (p<0-01) corrected mean systolic wall vel-ocities on days 1, 2, and 10. Contractile performanceimproved with higher values (p<0.01) on day 360compared with the first and second measurements.The lowest corrected mean systolic wall velocity

values were seen in segments with major Q waves,

Table 4 Corrected mean systolic wall velocity (mmis) in electrocardiographically classified segments

Days after acut myocardial infauois1 2 10 360

Normal 28-4± 16*4 28.9±17.7 30-6±+17-8 27-0+15-7(155) (138) (124) (135)

ST depression 12-4±+17-4 18-2+15-0 26.4+1 '.8 24.1+15-0(27) (49) (54, (17)

ST elevation 15.0±13.2 12-1+17-6 10.5±13.9 -

(49) (25) (6)Minor Q wave 13-9_12-1 13-9±21.4 20-0±13-1 23.2± 17.8

(14) (19) (29) (33)Minor Q wave +ST elevation 9-(4_110 6-4±15.1 6-1±17.1 7-9+16.4

(41) (40) (26) (12)Major Q wave 5.7t 9.5 3-5_141 7-0_12-9 11.4±11.8

(7) (6) (14) (15)Major Q wave +ST elevation 8-0+ 11-3 3-1+12-8 5-9+ 13.4 2.6± 14-6

(46) (61) (51) (25)All with ST elevation 10-8+11-8 6.3±15.0 7.1t14*7 4.3±15.2

(136) (126) (83) (37)All without ST elevation 11-2±11-2 11-4+19-1 15.7±13-0 19.5±15-8

(21) (25) (43) (48)

Values given as mean ± 1 SD.Figures in parentheses = No of evaluated segments.

m

on October 20, 2021 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.51.1.77 on 1 January 1984. Dow

nloaded from

Electocardiographic changes after myocardial infarction

which remained significantly (p<0.01) depressedthroughout the year. Segments with major Q wavesalso differed (p<0*01) from segments with isolated STdepression on days 2, 10, and 360 and from segmentswith minor Q waves on days 10 and 360. Values forsegments with major Q waves did not changesignificantly over the study year.The possible role of digitalis as a cause of the ST

depression was investigated in the nine patients takingthis drug (144 segments). These segments showedhigher corrected mean systolic wall velocities(15.1±20-3 mm/s) on day 1 than the remainder withST depression (10.0± 19*9 mmns) although this differ-ence was not significant. This slight difference waseven smaller on days 2, 10, and 360.

Segments with ST elevation v normal (Table 4 Fig. 4)Segments with isolated ST elevation showed lower(p<0.001) corrected mean systolic wall velocities ondays 1, 2, and 10. ST elevation without QRS changeswas not seen on day 360.

Segments with ST elevation and minor Q waveswere associated with slightly lower corrected meansystolic wall velocities throughout the study comparedwith normal segments and with segments with iso-lated ST elevation on day 1 (p<0.05). This minordifference was no longer evident on day 2 and similarvalues were seen subsequently.The greatest reduction in corrected mean systolic

wall velocities was seen in segments with ST elevationwith major Q waves which also showed lower values

30-

EE

a 20-

In

10C.Yw

(A

...............

o Normal

a *--...aST elevationac

....-a Minor Q wave changesac---. - a and ST elevation*--- .............

* Major wave changesa and ST elevation

1 2 10 360Days after acute myocardial infarction

Fig. 4 Corrected mean systolic wall velocity in segmentsgrouped according to ekltrocardiography into normal, STekvation, minor or major Q wave changes with ST elevation.a=signficantly lower than normal segments on correspondingday; c=significantly lower than segments with isolated STelevation.

81

E 20-

8

._L

6 10''DiU)

C.-

* Q wave changes

S ...............

0

I0........---- o

bD ST elevationbc

1 2 10 360Days after acute myocardial infarction

Fig. 5 Corrected mean systolic wall velocity in segments with Qwaves without ST elevation and in all segments with STelevation regardless of QRS. b =significantly differentfrom day1; c=significantly lower than segments without ST elevation.

than those with isolated ST elevation on days 1 and 2(p<0*001) and p<001 respectively). There was nosignificant difference, however, when these werecompared with segments with ST elevation withminor Q waves, and there was no significant changeover the year.

Segments with Q waves without ST elevation v all seg-ments with ST ekvation (Table 4, Fig. 5)Corrected mean systolic wall velocities on day 1 werereduced by the same amount in both groups. Valuesfor segments with Q waves-without ST elevation werehigher on subsequent days with significantly (p<0-05)higher values on day 360 compared with days 1 and 2.In contrast, all segments with ST elevation showedlower values on day 2, 10 (p<0-05), and 360 (p<001)when compared with day 1. Of all segments with STevaluation 12% showed dyskinesia on admission, and24, 22, and 390/o on days 2, 10, and 360 respectively.This difference in wall motion between segments withQ waves without ST elevation and those with ST ele-vation was significant (p<0.01) from day 10 with afurther increase in difference at the one year followup.

Discussion

The role of electrocardiography in diagnosing, localis-ing, and sizing myocardial infarction has been evalu-ated in both animal studies and postmortem examina-tions.19-21 The diagnostic value of a pathological Qwave as an indicator of left ventricular dysfunction isuncertain: some workers have found strong correla-tions between pathological Q waves and segmentalasynergy,5 22 23 while others have not.20 24 We there-fore considered that a study of wall motion properties

on October 20, 2021 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.51.1.77 on 1 January 1984. Dow

nloaded from

82

after a myocardial infarction measured by serialechocardiography with different levels of derangedQRS and ST segments was indicated.Any comparison between echocardiography, which

gives information about local wall motion, and elec-trocardiography, which reflects electrical activity,must be made with caution. They not only measure

different qualities, but also information is obtainedfrom different sample volumes. The electrocardiog-ram reflects a much larger portion of the ventricularwall than the narrow segment covered by the echocar-diographic beam. In the present study the threeanterior chest wall positions were the same forechocardiography and electrocardiography. Butmatching of echocardiographic and electrocardiog-raphic information is more difficult for the inferiorwall. We have based our comparisons on a previousechocardiographic study where systolic wall velocityand QRS changes were related to postmortemfindings in patients with myocardial infarction: thisstudy showed that conventional electrocardiographyreflected 10 out of the 16 left ventricular segmentswith sufficient sensitivity and specificity.'3 We there-fore limited this investigation to these 10 segments,which include three of four anteroseptal segments,four of six anterolateral segments, and three of fourinferoposterior segments. The middle third of theseptum (two segments) and the basal anterolateralwall, as well as one posterior and one mid-ventricularanteroseptal segment were not evaluated.When making M mode measurements recording

errors must be considered. In order to minimise spa-tial problems internal landmarks are used-that is,posterior and anterior edges of the right ventricle,papillary muscles, and leaflets, defined for eachview.'3 25 Certain limitations inherent in the echocar-diographic method remain and require comment.When comparing wall motion in the anterior andinferior left ventricular walls, using a fixed externalprobe location, we, like others, have found highermean systolic wall velocity values in inferoposteriorwall segments (39-46 mm/s) compared with anterior(16-21 mm/s) wall segments in healthy controls.9 It isunlikely that the normal left ventricle displays aninhomogeneous contraction pattern of this mag-nitude,'4 and a more likely explanation is cardiac rota-tional and pendulum motions, which cannot be sepa-rated from ventricular contraction motion by M modeechocardiography. Rotational motion is fortunatelyrelatively small. Instead, most of the differences seenin anterior and inferior wall motion are explained bythe anterior pendulum motion of the left ventricleduring systolic contraction.'5 16 Corrections for thesemovements of the heart are often made in wall motionstudies with cross sectional echocardiography,2'26but this has not previously been done in M mode

Lindvall, Rehnqvistrecordings. We have attempted to correct for cardiacmotion by using data from a healthy population ofsimilar age and sex.9 The corrected mean systolic wallvelocity obtained by this method is 28-32 mm/s for allsegments.

Different QRS scoring systems have been used toestimate left ventricular damage in myocardial infarc-tion.172728 Strong correlations between electrocar-diographic scores and overall ventricular functionas well as enzymatically estimated infarction sizehave been found by several groups.'72930 We haveused a simplified scoring system with three QRS clas-ses: normal, minor, and major Q waves subgroupedaccording to changes in the ST segment.The non-specific electrocardiographic finding of ST

depression is often seen in subendocardial infarction.'Nevertheless, it was a good indicator of left ventricu-lar dysfunction in the acute stage of the myocardialinfarction and was almost as good as Q waves or STelevation. During the later stages, however, the rela-tion between ST depression and wall motion distur-bances weakened. Several other mechanisms, includ-ing drugs such as digitalis and beta blockers, also pro-duce ST segment changes.3132 Beta blocking drugswere used infrequently but digitalis, which was pre-scribed to an increasing percentage, did not appear tobe important. The overall effect of drugs on the STsegment, however, was impossible to evaluate.

In contrast to ST depression, ST segment elevationis influenced less by parameters not associated withthe ischaemic process. Acute ischaemia causing STelevation leads to a reduction in wall motion,3334 andour findings from the acute phase confirm this. Wallmotion in segments with ST elevation even withoutthe appearance of Q waves was a third of normal.Segments with persistent ST elevation have beenstudied extensively and found to show poor ventricu-lar function.23 This is also supported by our findingsas segments with ST elevation also showed the poorestwall motion, while segments with only Q wavesshowed less pronounced derangement with time.Indeed, ST elevation on admission suggested dys-kinesia in 12% of the segments while the same elec-trocardiographic finding during late convalescencewas an even stronger (39% of segments) indicator ofdyskinesia.

Pathological Q waves in myocardial infarction areusually attributed to transmural necrosis. Ourfindings show that large Q waves indicate poor seg-mental function compatible with a concept of trans-mural necrosis. In contrast, minor Q waves recordedduring the first two days corresponded to poor con-tractile function. This electrocardiographic sign,however, does not indicate reduced left ventricularwall motion at later stages of the disease; contrary wallmotion is similar to that of segments without QRS

on October 20, 2021 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.51.1.77 on 1 January 1984. Dow

nloaded from

Electrocardiographic changes after myocardial infarction

changes.

This study was supported by grants from OlofNyhlin's Foundation and the Foundation ofSerafimerlasarettet.

References

1 Guyton R, McClenathan JH, Newman GE, Michaeis LL.Significance of subendocardial ST segment elevation caused bycoronary stenosis in the dog. Am Y Cardiol 1977; 40: 373-80.

2 Gorlin R, Klein MD, Sullivan JM. Prospective correlative studyof ventricular aneurysm. AmJ Med 1967; 42: 512-31.

3 Miller RR, Amsterdam EA, Bogren H, Massumi RA, Zetis R,Mason DT. Electrocardiographic and cineangiographic correla-tions in assessment of the location, nature and extent of abnormalleft ventricular segmental contraction in coronary artery disease.Circukation 1974; 49: 447-54.

4 Cook RW, Edwards JE, Pruitt RD. Electrocardiographic changesin acute subendocardial infarction 1. Large subendocardial andlarge nontransmul infarcts. Circulation 1958; 18: 603-12.

5 Arkin BM, Hueter DC, Ryan JT. Predictive value of electrocar-diographic patterns in localizng left ventricular asynergy in coro-nary artery disease. Am HeartJ 1979; 97: 453-59.

6 Hecht HS, Taylor R, Wong M, Shah PM. Comparative evalua-tion of segmental asynergy in remote myocardial infarction byradionuclide angiography, two-dimensional echocardiography,and contrast ventriculography. Am HeartJ 1981; 101: 740-9.

7 Jacobs JJ, Feigenbaum H, Corya BC, Phillips JF. Detection ofleft ventricular asynergy by echocardiography. Circulaion 1973;48: 263-71.

8 Heikkita J, Nieminen M. Echoventriculographic detection, local-ization, and quantification of left ventricular asynergy in acutemyocardial infarction. A correlative echo- and electrocardio-graphic study. Br HeartJ 1975; 37: 46-59.

9 Lindvall K. M-mode echocardiographic mapping in differentia-tion of normal from dysfunctioning left ventricular myocardium.Acta Med Scand 1981; 209: 149-60.

10 Carlens P. Effort angina. Studies on left ventricular pump at restand during exercise and the influence .of nitroglycerin, pro-pranolol, verapamil and coronary by pass surgery. Stockholm,1979. Academic thesis.

11 Heikkila J, Nieminen M. Rapid monitoring of regional myocar-dial ischaemia with echocardiography and ST segment shifts inman. Modification of "infarct size" and hemodynamics bydopamine and betablockade. Acta Med Scand 1979; 623 (suppl):71-95.

12 Nixon JV, Brown CN, Smitherman TC. Identification of trans-ient and persistent segmental wall motion abnormalities inpatients with unstable angina by two-dimensional echocardiogra-phy. Circulation 1982; 65: 1497-1503.

13 Lindvall K, Erhardt L, Sjo5gren A. Echo- and electrocardio-graphic findings in relation to autopsy in myocardial infarction.Clin Cardiol 1982; 5: 51-61.

14 Sniderman AD, Marpole D, Fallen E. Regional contraction pat-terns in the normal and ischemic left ventricle in man. Am JCardiol 1973; 31: 484-9.

15 Nieminen M. Normal left echoventriculography. Ann Clin Res1975; 7: 1-16.

16 Ingels N Jr, Daughters GT II, Stinson EB, Alderman EL. Evalu-ation of methods for quantitating left ventricular segmental wallmotion in man using myocardial markers as a standard. Circula-tion 1980; 61: 966-72.

83

17 Rose G, Blackburn H. Cardiovascular survey methods. WHOMonogr Ser 1968; 56:1.

18 Askenazi J, Maroko P, Lesch M, Braunwald E. Usefulness of STsegment elevations as predictors of electroeardiographic signs ofnecrosis in patients with acute myocardial infarction. Br Heart J1977; 39: 764-70.

19 Savage R, Wagner GS, Ideker RE, Podolsky SA, Hackel DB.Correlation of postmortem anatomic findings with electrocardio-graphic changes in patients with myocardial infarction. Retros-pective study of patients with typical anterior and posteriorinfarcts. Circulation 1977; 55: 279-85.

20 Horan LG, Flowers NC, Johnson JC. Significance of the diagnos-tic Q wave of myocardial infarction. Circulation 1971; 43: 428-36.

21 Heger JJ, Weyman AE, Wann L, Dillon JC, Feigenbaum H.Cross-sectional echocardiography in acute myocardial infarction:detection and localization of regional left ventricular asynergy.Circulation 1979; 60: 531-8.

22 Bodenheimer MM, Banka VS, Helfant RH. Q waves and ven-tricular asynergy: predictive value and hemodynamic significanceof anatomic localization. Am J Cardiol 1975; 35: 615-8.

23 Williams RA, Cohn PF, Vokonas P, Young E, Herman MV,Gorlin R. Electrocardiographic, arteriographic and ventriculo-graphic correlations in transmural myocardial infarction. Am JCardiol 1973; 31: 595-9.

24 Cheng TO. Incidence of ventricular aneurysm in coronary arterydisease. Am J Med 1971; 50: 340-55.

25 Lindvall K. Echocardiographic studies following myocardialinfarction. Sweden: Danderyd Hospital, 1983. Academic thesis.

26 Parisi AF, Moynihan PF, Felland ED, Feldman CL. Quantita-tive detection of regional left ventricular contraction abnor-malties by two-dimensional echocardiography. II. Accuracy incoronary artery disease. Circulation 1981; 63: 761-7.

27 Wagner GS, Freye CJ, Palmeri ST, et al. Evaluation of a QRSscoring system for estimating myocardial infarct size. I.Specificity and observer agreement. Circulation 1982; 65: 342-7.

28 Ideker R, Wagner GS, Ruth WW, et al. Evaluation of a QRSscoring system for estimating myocardial infarct size. II. Correla-tion with quantitative anatomic findings for anterior infarcts. AmJ Cardiol 1982; 49: 1604-14.

29 Palmeri ST, Harrison DG, Cobb FR, et al. A QRS scoring systemfor assessing left ventricular function after myocardial infarction.N EnglJ7 Med 1982; 306: 4-9.

30 Awan NA, Miller RR, Vera Z, Janzen DA, Amsterdam EA,Mason DT. Noninvasive assessment of cardiac function and ven-tricular dyssynergy by precordial Q wave mapping in anteriormyocardial infarction. Circulation 1977; 55: 833-41.

31 Nordstrom-Ohrberg G. Effect of digitalis glucosides on elec-trocardiogram and exercise test in healthy subjects. Acta MedScand 1969; 420 (suppl): 1-75.

32 Olsson G, Rehnqvist N, Lundman T, Melcher A. Metoprololtreatment after acute myocardial infarction. Acta Med Scand1981; 210: 59-65.

33 Nieminen M, Heikkila J. Echoventriculography in acute myocar-dial infarction. III. Clinical correlations and implication of thenoninfarcted myocardium. Am J Cardiol 1976; 38: 1-8.

34 Kerber RE, Marcus ML. Evaluation of regional myocardial func-tion in ischemic heart disease by echocardiography. Prog Car-diovasc Dis 1978; 20: 441-50.

Requests for reprints to Dr Kaj Lindvall, Department ofMedicine, Danderyd Hospital, S-182 88 Danderyd, Sweden.

on October 20, 2021 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.51.1.77 on 1 January 1984. Dow

nloaded from