JACC Vol. 10, No. IJuly 1987:17-24

17

Prospective Analysis of Electrocardiographic Variables as Markers forExtent and Location of Acute Wall Motion Abnormalities ObservedDuring Coronary Angioplasty in Human Subjects

MARC COHEN, MD, FACC, STEVEN J. SCHARPF, SA, K. PETER RENTROP, MD, FACC

New York, New York

To assess the usefulness of different electrocardiographicvariables as markers for the presence, extent and lo­cation of new wall motion abnormalities seen after sud­den controlled coronary occlusion, 23 patients with iso­lated left anterior descending (n = 12), or right (n =11) coronary artery disease and a normal baseline leftventriculogram were prospectively studied during trans­luminal coronary angioplasty. A simultaneous 12 leadelectrocardiogram was recorded before passing the bal­loon catheter and again at 30 seconds into the fourthinflation cycle. Using a second arterial catheter, a leftventriculogram was obtained at 40 seconds into the fourthinflation cycle. The extent of wall motion abnormalitieswas described as the percent of left ventricular perimetershowing hypocontractility.

During balloon inflation, 19 of the 23 patients devel­oped new hypocontractility ranging from 3 to 40%. STsegment elevation in lead V2 was the most sensitivemarkerfor anterior wall hypocontractility and ST segment el-

Clinical investigations reviewing the electrocardiographic(ECG) changes seen during the early phase of acute coronarythrombosis or coronary spasm indicate that several ECGpatterns may be observed. The most common change ob­served is ST segment elevation (1,2). Some investigators(3) observed that T wave peaking occurred before ST seg­ment elevation, whereas others (4-6) observed that acuteincreases in R wave amplitude occurred along with ST seg­ment elevation. A few investigators (7,8) have even ob­served transient Q waves within minutes of acute myocardialischemia.

Patients with coronary artery disease undergoing elective

From the Division of Cardiology, Department of Medicine, MountSinai School of Medicine of the City University of New York and MountSinai Hospital, New York, New York.

Manuscript received September 19, 1986; revised manuscript receivedDecember 28, 1986, accepted January 9, 1987.

Address for reprints: Marc Cohen, MD, Division of Cardiology, MountSinai Medical Center, One Gustave L. Levy Place, New York, New York10029.

© 1987 by the American College of Cardiology

evation in lead 111 was the most sensitive marker forinferior wall hypocontractility. Highly significant cor­relations were observed between the extent of the hy­pocontractile perimeter and 1) the sum of ST segmentelevation in all 12 leads; 2) the magnitude ofST segmentelevation in either lead V2 or lead 111; and 3) the totalnumber of leads with ST elevation 2:0.5 mV. No signif­icant changes were seen in the sum ofR wave amplitudes,but a significant prolongation of the QT interval wasseen during ischemia.

In conclusion, acute ST segment elevation parallelsthe development of new asynergy during transluminalcoronary angioplasty. The close association between STsegment elevation and the extent of asynergy may beuseful in predicting the impact of sudden coronary dis­section and occlusion on ventricular function duringcoronary angioplasty.

(J Am Coil Cardiol1987;10:17-24)

percutaneous trans luminal coronary angioplasty provide amodel of sudden cessation of anterograde blood flow. Usingthis model, one can prospectively assess the changes seenon the 12 lead ECG that occur within the first minutes ofcoronary occlusion. Furthermore, by using a second arterialcatheter for contrast left ventriculography during angioplastyballoon occlusion, correlations can be measured betweendifferent ECG variables and the size of the acute wall motionabnormality induced by coronary occlusion. In this fashion,we prospectively studied 23 patients to assess the usefulnessof different ECG indexes as markers for the presence, extentand location of new wall motion abnormalities seen duringcontrolled coronary occlusion by an inflated angioplastyballoon.

MethodsStudy patients. Patients were prospectively selected from

142 consecutive patients undergoing elective transluminalcoronary angioplasty. All patients had: I) single vessel cora-

0735-1097/87/$3.50

18 COHEN ET AL.ELECTROCARDIOGRAM DURING ANGIOPLASTY

JACC Vol. 10, No. IJuly 1987:17-24

nary disease of either the left anterior descending or a dom­inant right coronary artery and 2) normal global and seg­mental left ventricular function. Patients with the followingcharacteristics were excluded: 1) history of hypertension,(diastolic pressure >90 mm Hg), additional noncoronarycardiac disease, chronic obstructive lung disease, anemiaor bleeding diathesis; 2) peripheral vascular disease com­promising arterial access or renal insufficiency and 3) base­line ECG showing pathologic Q waves, left ventricular hy­pertrophy, bundle branch block or ST segment deviation of>0.5 mY. Twenty-six patients met these criteria and gaveinformed consent for this study, which was approved by theMount Sinai Hospital institutional review board. Three pa­tients were excluded from this analysis because ECGs ob­tained during the transluminal coronary angioplasty wereinadequate for analysis. The baseline clinical and angio­graphic characteristics of the remaining 23 patients are de­scribed in Table I. Patients with left circumflex artery an­gioplasty were excluded because of the lack of a wellestablished method for quantifying the extent of posteriorwall motion abnormalities.

Cardiac catheterization and angioplasty protocol. Theprinciples of our coronary angioplasty premedications,hemodynamic monitoring, guiding and balloon dilationcatheters and guide wires have been described in previouspublications (9,10). 111 addition to the routine angioplastycatheter system, a 5F pigtail catheter (Cook) was advancedthrough the femoral artery for left ventriculography.

Electrocardiography. Standard 12 lead ECGs were re­corded using a Hewlett-Packard 4700A Pagewriter modelwhich acquires into memory all 12 leads within 10 seconds.Precordial leads were monitored using X-ray-transparentelectrodes. Recordings were made during held end-expi­ration, standardization was 1.0 mV = 1.0 em and paperspeed was 25 mm/s. ST segment elevation, ST segmentdepression, R wave amplitude and the corrected QT interval(QTc) were measured on all ECG tracings. New hyperacuteT waves or sudden inversion of T waves as compared withthe baseline ECG tracing were noted. ST segment elevationabove the baseline (as defined by two consecutive PR seg­ments) was measured to the closest 0.5 mm 0.02 second

Table 1. Preangioplasty Clinical and AngiographicCharacteristics of 23 Patients

Age (yr)Sex (male/female)Duration of angina (mo)Left ventricular ejection fraction (%)Percent stenosis of lesion dilatedVessel dilated

Left anterior descending arteryProximal/distal

Right coronary arteryProximal/distal

57 ± 8

18/55.2 ± 7.9

65.5 ± 5.488 ± 11

127/5II7/4

after the J point (II). The magnitude of ST segment ele­vation in each lead showing elevation was measured andsummed over the 12 leads. The total number of leads show­ing ST segment elevation was also calculated. ST segmentdepression 2:0.5 mm below the baseline 0.08 second afterthe J point in any lead was considered significant.

R wave amplitude was measured to the closest 0.5 mmas the vertical distance from the preceding PR interval tothe peak of the R wave. The sum of R wave amplitudes inall 12 leads was calculated. The sum of R wave amplitudesin only those leads with ST segment elevation during isch­emia was also noted.

The QT interval was measured to the closest 0.02 secondin each of the 12 leads and, using Bazett's formula (12),was expressed as the QTc interval. The longest QTc intervalin each lead was noted and, in addition, the sum of the QTcintervals in all 12 leads was calculated for each electrocar­diogram. Heart rate was calculated from a Honeywell Elec­tronics for Medicine recording console.

All EeG measurements were made manually by one ofus (S.J.S.), who was completely unaware of the clinicaland ventriculographic impact of balloon inflation duringtransluminal coronary angioplasty. In five randomly chosenpatients, two consecutive baseline ECGs were recorded be­fore angioplasty to assess the intrinsic variability of our ECGmeasurements. The difference between the two measure­ments of the sum of ST segment elevation, the sum of Rwave amplitudes, the QTc interval and the sum of the 12QTc intervals for each of the five patients was analyzed andthe standard deviation of the mean value of the differenceswas calculated; intrinsic variability was described as ± 2SD. The sum of ST segment elevation varied by ± 0.5 mm,the sum of R wave amplitude varied by ± 7.6 mm, the QTcinterval varied by ±0.04 second and the sum of the 12 QTcintervals varied by ±0.6 second.

Arteriography and left ventriculography. All lesionsdilated in the right coronary artery were situated proximalto the posterior descending artery. All lesions dilated in theleft anterior descending artery were in the proximal or mid­segments of this vessel.

The Diasonics/Fischer digital subtraction imaging systemwas used for creation and analysis of the left ventricularimages obtained from injection of 40 cc of ReI1ografin-38%(meglumine diatrizoate) contrast directly into the left ven­tricle in the right anterior oblique view. Left ventricularejection fraction was calculated using the area-length methodof Dodge et al. (13). The extent of segmental wall motionabnormalities induced by angioplasty balloon occlusion wasquantified using the radial axis model ofIngels et al. (14, 15).In this model, radii are spaced 5° apart around the centroidand radial shortening <2 SD below the mean value for anormal population (14,15) defined a "hypocontractile seg­ment." The "percent hypocontractile perimeter" was de­fined as the length of the hypocontractile segment divided

JACC Vol. 10. No. IJuly 1987:17-24

COHEN ET AL.ELECTROCARDIOGRAM DURING ANGIOPLASTY

19

Pre TCA% Hyp=O

B. During TCA% Hyp = 40

A. Pre TCA

...• ,........ ...- .. ·.·0 _. . ........ Ut!4'J """""~ il l: : .. : . : j • • , . : • • :':J~.;"; : "d,i:: • • "

B. During TCA, ::s ST Elevat ion =10

@E' nllIIIFl:~j",,:::::--: :::T::':-f:n:: :::i :"fLHPIm-l+1·· ' tl, _,.. .._ , ,1 I I II!

..:::n::-~l-.=:1..: ~: rrr: . . L:-:·:.: :: : : • •• : : : I : : : : : I." '! 'l j 1

.... t . 1r.:;-..1:,... .. aVRr•.. •.... v' ·r.. ...... , V. I , I I", : : :.: rr: rrrrr: :~.: : I : : I . : : :~ : : : : : : : . : : : r: : : . , : . I : ~! : I . II•• t- . t l . ,..... ~ t •. o!.•• t •• • • ••• • •• ••~ ••f f ...n ..

~j ~W.$~r~'~ jil~i{nf1Fm~n~ ~ ~ ~ ~ ~ ~~ ~ j~~ ~ ~ iTm 1 ~ j.. . . . .. II ?-L· ~Fj l. t·lavL ... , .. .. , V2 ••• ••••••• v~ ·r 1... . . ..-. - . . ·· ·f-.- ......-· ·· ··· ·· ·· .. ~ ....... -.. ,.\~4ili.;.l.;.;~.ilj ,~ ,. , , ~ ..

~m~~:Wt~?d!:r\~~ ~m~Gi; . ~ ~ ;.;~ : :7ft:r;:r :::+~ III ili..· :::.;~~!..;+:. : : : : : : .~3. ::::: : : : : . ~.6. : : . : : . : .:-~!'H-'+ ~:.~~~:::~f:8.~;i ~~~I~~ : :

Figure I. Patient 20. A, From left toright, the pretransluminal coronaryangioplasty (Pre TCA) electrocardio­gram (ECG) and left ventriculogram.respectively. B, From left to right, theECG and left ventriculogram duringballoon inflation (During TCA), Incomparison with the preangioplastyECG, the recording during balloon in­flation reveals a sum of ST elevation= 10 mm. In comparison with thenormal preangioplasty left ventricu­logram (%Hyp = 0), the ventriculo­gram during balloon inflation revealshypocontractility invol ving 40% of theleft ventricular perimeter (%Hyp40).

by the total ventricular perimeter excluding the aortic andmitral valves. The percent hypocontractile perimeter wasused as an index of the extent of myocardial ischemia.Segments displaying isolated or combined hypokinesia, ak­inesia or dyskinesia were all considered "hypocontractile"segments.

Study protocol. /) Baseline pretransluminal coronaryangioplasty. A 12 lead ECG was obtained followed by aleft ventriculogram in the 200 right anterior oblique viewusing the 5F pigtail catheter. Routine transluminal coronaryangioplasty proceeded and several minutes after the thirdballoon inflation, the "study" cycle began.

Figure 2. Patient 16. A, From left toright, the preangioplasty electrocar­diogram (ECG) and left ventriculo­gram, respectively. B, From left toright, the ECG and left ventriculo­gram during balloon inflation. In com­parison with the preangioplasty ECG,the recording during balloon inflationreveals a sum of ST elevation = 16 B. During TCA, ::s ST Elevation = 16mm. In comparison with the normal , : : : : :1. : : : ' : . ' aVR , VIpreangioplasty left ventriculogram(%Hyp = 0), the ventriculogram dur­ing balloon inflation reveals hypocon­tractility involving 30% of the leftventricular perimeter (%Hyp = 30).Abbreviations as in Figure I.

B, During TCA% HYP=30

A. Pre TeA% Hyp =0

20 COHEN ET AL.ELECTROCARDIOGRAM DURING ANGIOPLASTY

JACC Vol. 10, No. 1July 1987:17- 24

Table 3. Changes in R Wave Amplitude and QT Intervals

A. R Wave Amplitude During Baseline (Ro) Cycle and Inflation Cycle(R.) in Patients With New Hypocontractility

B. QTc Interval During Baseline (QTco) and Inflation Cycle (QTc.) inPatients With New Hypocontractility

*p < 0.05 compared with baseline cycle using the paired t test. LAD= left anterior descending artery; RCA = right coronary artery.

RCA Occlusion (n = 9)LAD Occlusion (n = 10)

Sum of the QTc in all 12 leads (ms)I. Sumof Q'Tc., = 4.5 ± O. I3 I. Sum ofQ'Tc., = 4.70 ± 0.162. SumofQTc. = 4.80 ± 0.33* 2. Sumof QTc. = 4.80 ± 0.1 9*

QTc in lead with longest QTc (ms)I . QTeo = 0.44 ± 0.03 1. Q'Tc, = 0.46 ± 0.032. QTC4 = 0.48 ± 0.04* 2. QTc. = 0.47 ± 0.03

Sum of Rwaves in all 12 leads (mm)I. Sum of Ro = 79.2 ± 27.9 I. SumofRo = 76.6 ± 14.02. Sumof R. = 75.0 ± 28.2 2. Sum of R. = 75. 1 ± 15.6

SumofR waves only in leads with ST elevation (mm)I . Sumof Ro = 34.9 ± 24.5 I . SumofRo = 20.2 ± 15.32. SumofR. = 30.0 ± 23.6 2. Sumof R. = 18.9 ± 13.0

abnormalities developed acute ST segment elevation > 0.5mm compared with their baseline ECG. In contrast, only13 of these 19patients manifested ST segment depression.In two patients with a midright coronary artery lesion anda hypocontractile segment of only 3 and 5%, respectively,the only ST or T segment change observed was isolated STsegment depression in the inferior leads. In one other patientwith a midright coronary lesion and a hypocontractile seg­ment of only 5%, no ECG changes were observed. Fourpatients had no new wall motion abnormalities; none ofthese patients had any ST or T wave changes. None of the23 patients had significant T wave changes in the absenceof ST segment shifts. Comparing ST segment and T wavechanges, acute ST segment elevation was the most sensitive

2) Study cycle during transluminal coronary angioplasty,At 30 seconds into the fourth balloon inflation, a simulta­neous 12 lead ECG was repeated. At 40 seconds, a leftventriculogram was acquired. Examples of the ECGs andleft ventriculograms obtained before and during balloon in­fl ation in two patients are shown in Figures I and 2.

Statistical analysis. The study patients were classifiedinto two groups: Group I, 12 patients with left anteriordescending artery dilation and Group II , II patients withright coronary dilation. Each group was analyzed separately.All continuous variables are presented as the mean ± SD.The examination of the differences between the baselinecycle and the angioplasty inflationcycle involved the pairedt test. The relation between the ECG variables I) sum ofST segment elevation and 2) magnitude of ST segmentelevation in lead V2 or lead III and the percent hypocon­tractile perimeter was best fitted by applying a second orderpolynomial equation and a correlation coefficient was cal­culated. The relation between the ECG variable and thenumber of leads with ST segment elevation with the percenthypocontractile perimeter was assessed by applying non­parametric statistics and calculating the Spearman rank cor­relation coefficient. Probability (p) values 2::0 .05 were con­sidered not significant.

ResultsECG detection of the presence and location of new

wall motion abnormalities. Sudden coronary occlusion bythe angioplasty balloon resulted in significant myocardialischemia in the majority of patients. Left ventriculographyduring balloon inflation revealed that 19 of the 23 patientsexhibited new areas of hypocontractility ranging from 3 to40% of the left ventricular perimeter. The patients experi­encing changes in ST segments or T waves are listed inTable 2 according to left versus right coronary dilation.Sixteen of the 19patients who developed new wall motion

Table 2. Number of Patients With ST Segment or T Wave Shifts in Each Lead

Lead

Artery Occluded aVL VI V2 Vj V. V,

ST Elevation

LAD (n = 12) 2 5 6 10 9 7 3RCA(n = II ) 0 0 0 0 0 0 0

ST Depression

LAD(n = 12) 0 0 0 0 0 0 IRCA(n = II ) 2 4 0 4 5 2 2

T Wave Changes

LAD (n = 12) 1 4 I 2 2 2 2RCA(n = II ) 2 2 0 I 0 1

LAD = left anterior descending coronary artery; RCA = right coronary artery.

3o

II

o6

51

33

III

o6

52

53

aVF

o5

52

33

JACC Vol. 10, No. IJuly 1987:17-24

COHEN ET AL.ELECTROCARDIOGRAM DURING ANGIOPLASTY

21

ECG indicator of new asynergy (sensitivity = 85%, spec­ificity = 100%, predictive value = 100%).

Among the 19 patients with new asynergy. 10had anteriorwall and 7 had inferior wall asynergy, All of the 10patientswith new anterior wall asynergy experienced ST segmentelevation in lead V2 :::::0.5 mm (Table 2). No other lead wasmore sensitive in localizing new anterior wall asynergy. Onepatient had a greater magnitude of ST segment elevation inlead V3 than in lead V2, but both leads demonstrated STsegment elevation. Among the seven patients with new in­ferior wall motion abnormalities, five exhibited ST segmentelevation in leads II and III :::::0.5 mm; lead aVF was lesssensitive.

Changes in R wave amplitude and QT interval betweenthe baseline cycle and the inflation cycle were analyzed inthose patients with new asynergy (Table 3). No significantdifference was observed between the sum of R wave am­plitudes on the inflation cycle ECG compared with the base­line ECG. When this analysis was carried out summing theR waves only in those leads in which there was ST segmentelevation there was still no significant change in the sum ofR wave amplitude during coronary occlusion. In contrast,both groups of patients experienced a significant prolon­gation in the sum of the QTc interval during ischemia (Table3). When one examined only the lead with the longest QTcinterval, before and during balloon occlusion, only the pa­tients undergoing left anterior descending dilation demon­strated a significant prolongation in the longest QTc intervalduring ischemia.

ECG correlation with extent of new wall motion ab­normalities (Fig. 3 to 5). Three ECG variables utilizingST segment elevation were analyzed to assess the correlationbetween the ECG and the extent of the new hypocontractilesegment. Using a second order polynomial equation to de­scribe the relation between the sum of ST segment elevationand percent hypocontractility, a highly significant correla­tion was observed for left coronary as well as for rightcoronary artery occlusion (r = 0.96 and 0.89, respectively)(Fig. 3). Because the prior analysis revealed that lead V2

and lead III were the most sensitive individual leads foranterior and inferior asynergy, respectively, we analyzedthe relation between the magnitude of ST segment elevation

in only one lead (lead V2 or lead 1II) and percent hypocon­tractile segment. A highly significant correlation was ob­served for leads V2 (r = 0.92) and 1II (r = 0.90) (Fig. 4).Last, there was a significant positive correlation betweenthe total number of leads with ST segment elevation andthe percent hypocontractile perimeter (Spearman rank r =

0.77 for the left anterior descending artery and r = 0.92for the right coronary artery) (Fig. 5).

Discussion

Our study constitutes the first prospective analysis of therelation between the ECG and an angiographic estimate ofthe extent of the ischemic segment assessed within minutesof sudden coronary occlusion by balloon inflation in a largegroup of human subjects with fixed, severe atheroscleroticheart disease. Using the angioplasty model described, weassessed the ECG variables of ST segment elevation, STsegment depression, R wave amplitude and corrected QTinterval and examined the acute changes in these variablesrelative to the extent of ischemic wall motion abnormalities.Our observations suggest that in humans there is a highlysignificant relation between ST segment elevation and thesize of the ischemic hypocontractile zone.

Previous studies comparing 81 segment changes withinfarct size. Using epicardial electrodes in a dog model ofsudden ischemia, Maroko et al. (16) suggested that the sumof ST segment elevation and the number of leads showingST elevation correlated well with pathologic estimates ofinfarct size. In 1975, Muller et al. (17), using a dog model,showed that the sum of ST segment elevation obtained fromprecordial leads also correlated well with infarct size, Usinga pig model of sudden ischemia, Holland and Brooks (18,19)showed that infarct size correlated with ST segment ele­vation in precordial but not in epicardial measurements of

Figure 3. The relation between the sum of ST segment elevationduring coronary occlusion (abscissa) and the percent hypocon­tractile perimeter (%Hyp) (ordinate). Left panel, Twelve patientswith left anterior descending coronary (LCA) balloon dilation;right panel, 11 patients with right coronary artery (RCA) balloondilation.

40 LCA . RCA

~30c,>-I

~ 20

10 .. yo-12 +8,I(x)-4(x 2) V· yo 3 + 4,2(x)-'2(x2)

ro096:SEE049; n ol2 r 0 0,89; SEEo 5 5; n el t

°0 •2 4 6 8 10 12 14 16 0 2 14 16

L OF 5T ELEVATION (rnrn)

22 COHEN ET AL.ELECTROCARDIOGRAM DURING ANGIOPLASTY

JACC Vol. 10. No. IJuly 1987:17-24

4 5

y< -4 + 291,) - 49(, 2)« 0.92; 5[ E<6 9,0 <12

Figure 4. The relation between themagnitude of ST segment elevation ineither leadV2 for leftanteriordescend­ingartery(LCA) balloon dilations (leftpanel) or in lead \II for right coronaryartery (RCA) balloon dilations (rightpanel) during coronary occlusion (ab­scissa) and the percent hypocontractileperimeter (%Hyp) (ordinate).6

y - 2.9 + 11 ( . ) - I II,2)r - 90 ; 5[ E' 5.4; 0 - 1I

34 5LEAD ill

2

RCA

2

40 LCA

30o,>­:I:~20

MAGNITUDE OF ST ELEVAT ION ( mm)

monitoring just one lead gave a good reflection of the extentof the ischemic zone . This finding is in keeping with theobservations of Capone and Most (24), who showed a goodcorrelation between the magnitude of ST segment elevationin a single precordial lead and the sum of ST seg mentelevation . In several of our patients , howe ver , minimal ECGchanges were observed despite total balloon occlusion .Therefore, other factors, such as site of balloon occlu sion ,duration of occlu sion or good collateral flow , may acco untfor the extent of ECG changes .

Previous studies on R wave amplitude and QT inter­vals during ischemia. Earlier investigators (4-6) suggestedthat giant R waves were a common early sign of acuteischem ia. Using atrial pacing to induce ischem ia, David etal. (25) observed an increase in R wave amplitude in leadV5 in 15 of 17 patients. Barnhill et al. (26) assessed thechanges in the QRS complex using digital computer analysisin five patients with spontaneous variant angina and ob­served significant chan ges in the absolute height of the QRScomplex during ischemia . In contrast. Myers et al. (27)measured spatial R wave amplitude during exercise testingand observed no relation between changes in R wave heightand evidence of ischemia based on thallium-201 imaging .Recently, Feldman et al. (28), using precordial and intra­coronary unipolar ECGs durin g angioplasty, failed to showany acute changes in R wave height durin g balloon occlu­sion.

In the present study, no significant differences were ob­served between the R wave ampl itude s during ischemiacompared with baseline. In an attempt to minimize anydilutional effect of random changes in R wave height, anadditional analysis was performed of R wave changes onlyin those leads showing ischemic ST segment elevation . Thistoo , revealed no significant difference s. Several individualpatient s did manife st a large change in R wave height.However , analysis of two repeated baseline measurementsrevealed a large intrin sic variability in R wave ampl itude .Therefore , given the size of our patient group, the changesobserved did not reach statistical significance .

Mandel et al. (29) observed a significant shortening inthe refractory period within the ischemi c zone early after

1 1 I I I I I J I I I

- LeA · . : f- RCA -

~.

f- · .·

f- ·.· ·. ,<077 f- . · '00.92

n: 12 n " ll

I I I10

a,>­:I:oe 20

30

40

oo I 2 3 4 56 7 0 I 23 4 567NUMBER OF LEAD S WITH ST ELEVATION

ST deflections . In addition, they outlined several nonspatialfactors that could have a major impact on the extent of STsegment elevation . Work in humans, however, has yieldedconflicting results . Whereas two groups of investigators(20 ,21) observed poor correlation between the sum of STsegment elevation and infarct size estimated from creatinekinase washout curves, two other groups (22,23) observedgood correlation between these variables .

Our observations indicate that both the sum of ST seg­ment elev ation and the number of leads with ST segmentelevation observed within the first minute of ischemia cor­relate well with angiographic estimates of the size of theischemic hypocontractile zone . This good correlation ap­plies to patients with either left or right coronary occlusion .The curves that best describe this relation have a steep earlyrise followed by a plateau. This would indicate that abovea certain size of hypocontractile zone other factor s probablyplay a role in determining the ultimate extent of the STsegment elevation (18 ,19). A greater number of patientswould be needed, however, to further validate the corre­lations we observed. Because full 12 lead ECGs may notbe readily available during routine coronary angioplasty, weanalyzed the relation between the magnitude of ST elevationin the most sensitive lead for either anterior or infer iorasynergy and extent of hypocontractility and found that

Figure 5. The relation between the total numberof leads with STsegment elevation during coronary occlusion (abscissa) and thepercent hypocontractile perimeter (%Hyp) (ordinate). Abbrevia­tions as in Figure 4.

JACC Vol. 10. No.Joly 1987: 17-24

COHEN ET AL.ELECTROCARDIOGRAM DURING ANGIOPLASTY

23

coronary occlusion in a dog model, suggesting that ischemiamay affect the QT interval. In a prospective study in hu­mans, Cinca et al. (30) measured the QT interval within 6hours of acute transmural infarction. They observed a rel­ative shortening in the QT interval in the lead overlying theischemic zone in the early phase, followed by a prolongationafter 12 hours. Other investigators (31,32) suggested thatQT interval prolongation was relatively common in the earlyphase of acute myocardial infarction. Furthermore, data fromAhnve et al. (33) indicate that QT prolongation after myo­cardial infarction may be associated with a higher postin­farction mortality rate.

Our study indicates that within the first minute of coro­nary occlusion there is a small but significant increase inthe QTc interval assessed over all 12 leads for both left andright coronary occlusion. Because previous investigationsused only one particular lead to assess changes in the QTcinterval, we also analyzed the lead with the longest QTinterval. Significant prolongation occurred in patients withleft occlusion but not with right coronary occlusion. Theimplication of these QT changes in anyone lead, however,should be viewed with caution because of the relatively largeintrinsic variability observed in our analysis of two repeatedbaseline ECGs.

Clinical implications. The ECG is a very sensitive andsimple tool for assessing sudden onset of myocardial isch­emia during coronary angioplasty. Of the ECG variablesanalyzed, ST segment elevation best parallels the devel­opment of new asynergy during trans luminal coronary an­gioplasty. The close association between ST segment ele­vation and extent of asynergy may be useful in predictingthe impact of sudden coronary dissection and occlusion onventricular function during coronary angioplasty. Thisknowledge may help the physician better assess the needfor emergency revascularization after such a complication.

We thank Paul Schweitzer, MD, FACC and John E. Madias. MD. FACC.of the Division of Cardiology. for advice during this investigation.

ReferencesI. Pardee HEB. An electrocardiographic sign of coronary artery obstruc­

tion. Arch Intern Med 1920;26:244-50.

2. Prinzmetal M, Kennamer R, Merliss R, Wada T. Bor N. Anginapectoris. I. A variant form of angina pectoris: preliminary report. AmJ Med 1959;27:375-88.

3. Dressler W, Roesler H. High T waves in the earliest stage of myo­cardial infarction. Am Heart J 1947;34:627-35.

4. Prinzmetal M, Toyoshima H, Ekmekci A, Nagaya T. Angina pectoris.VI. The nature of ST segment elevation and other ECG changes inacute severe myocardial ischemia. Clin Sci 1962;23:489-514.

5. MacAlpin RN, Kattus AA, Alvara AB. Angina pectoris at rest withpreservation of exercise capacity: Prinzmetal's variant angina. Cir­culation 1973;47:946-58.

6. Madias JE, Krikelis EN. Transient giant R waves in the early phase

of acute myocardial infarction: association with ventricular fibrillation.Clin Cardiol 1981;4:339-49.

7. Meller J. Conde CA, Donoso E. Dack S. Transient Q waves in Prinz­metal's angina. Am J Cardiol 1975;35:691-5.

8. Roesler H, Dressler W. Transient electrocardiographic changes iden­tical with those of acute myocardial infarction accompanying attacksof angina pectoris. Am Heart J 1954;47:520-6.

9. Cohen M. Rentrop KP. Limitation of myocardial ischemia by collateralcirculation during sudden controlled coronary artery occlusion in hu­man subjects. A prospective study. Circulation 1986;74:469-76.

10. Rentrop KP. Cohen M, Blanke H, Phillips RA. Changes in collateralchannel filling immediately after controlled coronary artery occlusionby an angioplasty balloon in human subjects. J Am Coli Cardiol1985:5:587-92.

II. Rude RE. Poole K. Muller JE. et al. and the Milis Study Group.Electrocardiographic and clinical criteria for recognition of acute myo­cardial infarction based on analysis of 3,697 patients. Am J Cardiol1983:52:936-42.

12. Bazett He. An analysis of the time relationship of the electrocardio­gram. Heart 1920;7:353-70.

13. Dodge HT. Sandler H, Baxley WA, Hawley RR. Usefulness andlimitations of radiographic methods for determining left ventricularvolumes. Am J Cardiol 1966;18:10-20.

14. Ingels NB Jr , Mead CW, Daughters GT II, Stinson EB, AldermanEL. A new method for assessment of left ventricular wall motion.Comput Cardiol IEEE 1978;57-61.

15. Ingels NB Jr. Daughters GT II. Stinson EB, Alderman EL. Evaluationof methods for quantitating left ventricular segmental wall motion inman using myocardial markers as a standard. Circulation 1980;61:966-72.

16. Maroko PR, Kjekshus JK. Sobel BE, et al. Factors influencing infarctsize following experimental coronary artery occlusions. Circulation1971;43:67-82.

17. Muller JE, Maroko PR, Braunwald E. Evaluation of precordial electro­cardiographic mapping as a means of assessing changes in myocardialischemic injury. Circulation 1975;52:16-27.

18. Holland RP. Brooks H. Precordial and epicardial surface potentialsduring myocardial ischemia in the pig: a theoretical and experimentalanalysis of the TQ and ST segments. Circ Res 1975;37:471-80.

19. Holland RP, Brooks H. Spatial and nonspatial influences on the TQ­ST segment deflection of ischemia: theoretical and experimental anal­ysis in the pig. J Clin Invest 1977;60:197-214.

20. Thompson PL. Katavatis V. Acute myocardial infarction: evaluationof precordial ST segment mapping. Br Heart J 1976;38:1020-4.

21. Norris RM. Barrat-Boyes C. Heng MK, Singh BN. Failure of STsegment elevation to predict severity of acute myocardial infarction.Br Heart J 1976;38:85-92.

22. Henning H. Hardarson T. Francis G. O'Rourke RA, Ryan W, RossJ Jr. Approach to the estimation of myocardial infarct size by analysisof precordial ST-segment and R wave maps. Am J Cardiol 1978;41:1-8.

23. Inoue M. Hori M, Fukunami M, et al. Evaluation of praecordial STsegment mapping as an index of infarct size in patients with acutemyocardial infarction. Br Heart J 1979;42:726-33.

24. Capone RJ, Most AS. The single precordial lead for ST segmentmonitoring: comparison with the multiple lead map. Am Heart J 1979;97:753-8.

25. David D, Kitchen JG III, Michelson EL, Naito M, Sawin HS, ChenCC. R-wave amplitude responses to rapid atrial pacing: a marker formyocardial ischemia. Am Heart J 1984;107:53-61.

26. Barnhill JE, Wikswo JP Jr. Dawson AK, et al. The QRS complexduring transient myocardial ischemia: studies in patients with variantangina pectoris and in a canine preparation. Circulation 1985;71 :901-11.

27. Myers 1, Ahnve S, Froelicher V, Sullivan M. Spatial R wave amplitude

24 COHEN ET AL.ELECTROCARDIOGRAM DURING ANGIOPLASTY

changes during exercise: relation with left ventricular ischemia andfunction. 1 Am Coli Cardiol 1985;6:603-8.

28. Feldman T, Chua KG, Childers RW. R wave of the surface andintracoronary electrogram during acute coronary artery occlusion. Am1 Cardiol 1986;58:885-90.

29. Mandel Wl, Burgess Ml, Neville 1 lr, Abildskov lA. Analysis of Twave abnormalities associated with myocardial infarction using a the­oretical model. Circulation 1968;38:178-88.

30. Cinca 1, Figueras 1, Tenorio L, et al. Time course and rate dependenceof Q-T interval changes during noncomplicated acute transmural myo­cardial infarction in human beings. Am 1 Cardiol 1982;48:1023-8.

JACe Vol. 10, No. IJuly 1987:17-24

31. Dorghazi RM, Childers R. Time-related changes in the Q-T intervalin acute myocardial infarction: possible relation to local hypocalcemia.Am 1 Cardiol 1978;41:684-8.

32. Taylor Gl, Crampton RS, Gibson RS, Stebbins PT, Waldman MTG,Beller GA. Prolonged QT interval at onset of acute myocardial in­farction in predicting early phase ventricular tachycardia. Am Heart1 1981;102:16-24.

33. Ahnve S, Gilpin E, Madsen EB, Froelicher Y, Henning H, Ross 1If. Prognostic importance of QTc interval at discharge after acutemyocardial infarction: a multicenter study of 865 patients. Am Heart1 1984;108:395-400.