Qualitative and quantitative analysis of characteristic body surface potential map features in anterior and inferior myocardial infarction

Qualitativeand QuantitativeAnalysisof Characteristic BodySurfacePotential Map Featuresin Anterior and Inferior MyocardialInfarction FRED KORNREICH, MD, PhD, TERRENCE J. MONTAGUE, MD, MICHAIL KAVADIAS, BSc, JORIS SEGERS, MSc, PENlTl M. RAUTAHARJU, MILAN B. HORACEK, PhD, and BRUNO TACCARDI, MD, PhD

Body surface potential maps were recorded from 120 electrode sites in 236 normal subjects and 258 patients with initial evidence of either anterior myocardial infarction (Ml) or inferior Ml to identify characteristic map patterns in both groups. After time normalization, averaged map distributions were displayed at 18 equal time intervals during both QRS and ST-T waveforms from the normal, anterior Ml and inferior Ml groups. At each time instant, the 120-point averaged normal map was subtracted in turn from the corresponding anterior and inferior Ml maps; the resulting differences at each electrode site were divided by the pooled standard deviation and the obtained values (discriminant indexes), plotted as contour lines with 1 standard deviation increments, producing discriminant maps for each bigroup comparison. The most consistent discriminant patterns in 114 patients with anterior Ml were observed in early QRS in the upper left anterior chest where abnormal negative voltages reflected loss of electric potentials while reciprocal changes were noticed in the lower back; by mid-QRS, both distributions had moved jointly and vertically, the former in the lower torso on the midsternal line, the latter in the upper back. In 144 patients with inferior Ml, abnormal positive distributions were observed in early QRS in the upper back, followed later by excessive negative voltages in the inferior right anterior chest; at mid-QRS, both distributions had migrated horizontally, the former proceeding toward the upper

I

t is widely recognized today that body surface potential maps provide more information on the electrical activity of the heart than can be obtained from the standard l&lead electrocardiogram or the vectorcardiogram. 1-3Numerous studies have demonstrated the diagnostic superiority of surface maps over conventional electrocardiography in a large variety of clinical conditions.4-8 Moreover, sequential instant-by-instant maps of cardiac electrical events as projected on the body surface allow direct association of surface potentials with major wavefronts within the heart and have advanced the electrophysiologic bases of diagnostic electrocardiographic criteria.gJO 1230

MD, PhD,

anterior torso, the latter to the lower left dorsal area. Abnormal negative voltages were seen in the precordial region during ST in the inferior Ml group, moving toward the lower left flank where they stayed throughout T; in the anterior Ml group abnormal negativities appeared in the precordial area at the beginning of T and remained there until the end of repolarization. lntragroup variability was investigated by producing scattergrams of extrema and of abnormal peak discriminant indexes (2 2 standard deviations) derived from individual patients within each population. The presence of electrocardiographic subgroups was suggested for both classes of infarction: anterior Ml with or without apical involvement in the anterior group and inferior-posterior Ml with or without right ventricular involvement and with or without apical extension in the inferior group. Thus, both types of infarction share a temporally common but spatially discriminating portion of the early QRS. Repolarization patterns in both infarction groups were also spatially discriminant from normal subjects, but temporally heterogeneous one from the other. Retrospective classification of patients based on the presence of one or more patterns typical in time and in location of their respective groups yielded 96% and 93% of correct assignment to the anterior Ml and the inferior Ml classes, respectively. The specificity was 94% for normal control subjects. (Am J Cardiol 1987;60:1230-1238) Despite these advantages, widespread use of maps as a diagnostic tool has not occurred. Several factors contribute to this situation, namely, the cost and bulk of the required equipment, the difficulty of applying between 100 and 200 electrodes to the thoracic surface and the collection of extremely large amounts of data to be processed, displayed and analyzed. Acceptance by the clinical community has been further hampered by the nontraditional data display and the lack of an abundance of accumulated clinicopathologic correlations. Only extensive lead sets provide the display format and the spatial and temporal information that allows

December

1, 1987

the most meaningful physiologic correlation between surface potential features and underlying electrical events, particularly for abnormalities of short duration or only present between very close adjacent sites. In addition, significant changes in low-magnitude potential distributions, which may not result in clearly visible changes in scalar waveforms, can be examined using body surface maps.11-13 The determination of map criteria for differentiating various cardiac states requires a large data base consisting of both normal mapsI and a variety of abnormal map morphologies. The present study aims to contribute to this catalog. By using 236 normal subjects and 258 patients with documented first Q-wave myocardial infarction (MI), qualitative and quantitative interpretation criteria are presented in a format that can be easily viewed and interpreted by the cardiologist.

Methods Study population: We retrospectively studied 590 subjects, 236 of whom were normal and 354 were patients with first MI. The normal group consisted of 71 (30%) women and 165 (70%) men with no evidence of heart disease by history, physical examination, 12-lead electrocardiogram and, when available, echocardiogram. They were 20 to 55 years old (mean 36): 52% of normal subjects were in the decade 30 to 39 years, 26% in the decade 20 to 29 years and 22% were age 40 to 55 years. All patients in the MI group had a typical history of first prolonged, ischemic-type cardiac pain and characteristic changes in enzyme levels. In many patients the diagnosis was further substantiated by coronary angiography and ventriculography, echocardiography or nuclear imaging. Patients were excluded if they had electrocardiographic evidence of complete left (9 patients) or right (14 patients] bundle branch block or major nonspecific intraventricular conduction delay (QRS 1120 ms, 10 patients]. Other criteria for exclusion were: electrocardiographic signs of combined anterior and inferior Q-wave infarction, present at the time of admission (with either or both locations representing the acute infarct, 17 patients] or occurring between initial acute event and map recording (11 paFrom the Unit for Cardiovascular Research and Engineering, the Free University of Brussels, Brussels, Belgium; the Departments of Physiology and Biophysics, and Medicine, Dalhousie University, and the Victoria General Hospital, Halifax, Nova Scotia, Canada; and the Istituto di Fisiologia Generale, University of Parma, Parma, Italy. This study was supported by the National Fund of Scientific Research (FGWO 3.00%.86), the Free University of Brussels, the Belgian Ministry of Education (FCFO], the Nora Eccles Treadwell Foundation, the Richard A. and Nora Eccles Harrison Fund for Cardiovascular Research, the Medical Research Council of Canada (PG-301, the Nova Scotia Heart Foundation and the Italian National Research Council, Special Project on Bioengineering. Manuscript received May 28,1987; revised manuscript received and accepted August 17,1987. Address for reprints: Fred Kornreich, MD, PhD, Unit for Cardiovascular Research and Engineering, Free University Brussels (VUB), Laarbeeklaan 103, B-1090 Brussels, Belgium.

THE AMERICAN

JOURNAL

OF CARDIOLOGY

Volume 60

1231

tients); clinical infarction without typical QRS changes (29 patients); and presence of Wolff-Parkinson-White syndrome (6 patients]. Of the remaining 258 patients, 114 were classified as having anterior MI if they had Q waves 230 ms in duration in leads I, aVL or V1 to V6 or initial R waves of SO.2mV in leads V1 and Vz. Patients were classified as having (posterior] inferior MI if abnormal Q waves were present in leads II or aVF, or if R waves in VI and VZ were exceedingly tall (R to S ratio 1 or more] or broad (at least 40 ms). Subjects in the infarct population were 34 to 73 years old (mean 55): 50% of patients were in the 40 to N-year-old group, 42% were age 60 to 73 years and only 8% were in the decade 30 to 39 years; 62 (24%) patients were women and 196 (76%) were men. Body surface mapping: Electrocardiographic signals were recorded simultaneously on digital tape from 117 torso and 3 limb electrode sites in each subj ect and digitized at 500 samples/s with Wilson’s central terminal as reference potential. Tracing quality was monitored visually during the recording; later, the stored data were processed by performing selective averaging and again carefully edited. Details of the procedure have been reported previously.15 All measurements were performed with the PR segment as baseline and common time instants for QRS onset, QRS offset and ST-T offset were derived from superimposed Frank X, Y and Z leads. We then time normalized separately the QRS waveform and the ST-T waveform and represented them by 70 to 180 points, respectively. Patients with MI were studied a mean of 10 months after the acute event (range 1 day to 24 months); the interval between the event and recording of body surface potential maps was l month in 188 (73%). Processing and map display: Three types of display are presented. Isopotential maps: In this representation, sequential individual and group mean maps during the QRST complex are drawn as isocontour lines, connecting points of equal voltage at successive time instants Group mean maps were computed separately for the normal group and each MI group by averaging the potentials at each instant, at each electrode site from all subjects in each group. The depicted sequential maps were obtained by sampling time-normalized QRS and ST-T waveforms at equal intervals; this resulted in 18 instants for QRS and 18 instants for ST-T. The first 9 instants for QRS in the normal groups are shown in Figure 1, second column. Discriminant maps: For each time instant, difference maps were computed by subtracting at each electrode site the normal group mean voltage from the anterior MI and inferior MI group mean voltages; sequential discriminant maps for each 2 group comparison were obtained by further dividin,g each resulting difference by the corresponding composite standard deviation computed from the population under consideration. The values thus achieved, referred to as “discriminant (map] indexes,” were strictly proportional to t test statistics and provided, in contrast to difference maps, information on the capability for

1232

SURFACE

MAPS IN ANTERIOR

AND INFERIOR MYOCARDIAL

INFARCTION

each measurement at each electrode site and at each instant, to separate each class of MI from the normal group. (Discriminant maps differ from departure maps4 in that the values represented are standardized voltages instead of absolute voltages; these values have a statistical significance, useful for subsequent discriminant analysis and their magnitude is independe.nt of the actual voltages.) Discriminant maps were also represented as contour lines, connecting discriminant indexes of equal value and sign at each instant. Identical values with opposite signs have the same discriminant capability; negative and positive discriminant indexes are to be interpreted as either loss or excess of electrical forces with respect to average normal voltages. Increments are expressed in standard deviation as units. Figure 1 [columns 4 and 6) shows

sequential discriminant maps for the anterior MI and inferior MI groups, respectively, during the first half of activation. Individual maps are computed in a similar fashion, subtracting from the patient’s voltage data, normal mean group voltages and dividing the resulting differences by the standard deviations of the normal population at corresponding sites and instants. Scattermaps: In order to evaluate the intragroup dispersion range of map features within each class at each time instant and developing quantitative criteria for possible subgrouping of patients within each class, the sites where individual minimum and maximum voltages and peak discriminant indexes (2 or more standard deviations) are observed were placed on the torso. Examples of such instantaneous scattergrams of voltage extrema and discriminant scattermaps are discussed later.

Results AM,

ah-AM1

I M

AN-It.1

Normal group: The second column of Figures 1 through 4 depicts sequential group mean surface potential distributions in the normal population. The map features are similar to those previously published from normal subjects. l6 At the onset of activation, a midsternal maximum appears together with a low to mid-dorsal minimum; these patterns represent left to right septal activation. l7 Later, the maximum moves downward and leftward with positive potentials progressively extending to the inferior border around the thorax while the minimum migrates toward the upper back and the right shoulder. At this time, excitation is spreading through the walls of both ventricles. Soon after right ventricular breakthrough (at approximately 7/18 QRS], the anterior minimum develops rapidly and the entire anterior potential distribution becomes ORS

FIGURE 1. Group mean body surface potential maps of normal subjects (N) and patients with either anterior myocardial infarction (AMI) or inferior myocardial infarction (IMI) and discriminant maps for the separation of normal subjects from, respectively, anterior Ml pat/ents (AN-AMI) and inferior Ml patients (AN-IMI). Surface maps and corresponding bi-group discriminant maps are depicted for the first 9 of 18 equidistant instants calculated on QRS. The left half of each map represents the anterior torso and the right half, the back. The small circles correspond to the V, through Vs standard electrode positions. The top of the map is at the level of the sternal notch and the bottom at the umbilical level. Contour lines in surface maps are drawn for the following voltages f 20, 40, 80, 180, 320, 840, 1,280 and 2,580 @. Contour lines in the discriminant maps are drawn with increments of 1 U (the magnitude of the discriminant index equals 1 when mean voltages of corresponding pairs of measurements are 1 composite standard deviation apart). Solid lines indicate positive values and doffed lines, negative values. Time markers are represented on reference electrocardiographic waveforms by vertical bars (first column).

18 A-- q

N

AMI

AN-AM1

mQ, /

FIGURE 2. The second half of QRS is depicted

(see Fig. 1).

December

negative; this prominent anterior sink indicates that the activation front has reached the epicardial surface of a large part of both ventricles. Meanwhile, the maximum moves to the lower left flank and then dorsally, representing further activation of the lateral and posterobasal walls of the left ventricle. During the last quarter of QRS (14/18 QRS), a new maximum appears in the upper sternal region while the receding dorsal maximum is still present. At the end of depolarization (17/18 QRS), a new positive pole appears in the left precordial area and merges rapidly with the upper sternal maximum; this new maximum, representing early recovery potentials, expands over the inferior tqrso both anteriorly and posteriorly and persists throughout the ST-T inteyval. A minimum located in the upper dorsal region extends toward the mid-dorsal and upper right thoracic areas and recedes at the end of the T wave. Anterior myocardial infarction: The third column of Figures 1 through 4 represents the sequence of group mean potential distributions from 114 patienfs in the anterior MI group. Early negative voltages appear in the upper left precordial area and increase rapidly in amplitude and perimeter during the first part of QRS. During the s&cond part of QRS, the minimum decreases and proceeds slightly downward to the xiphoid appendix; at the same time, the maximum follows a trajectory from the inferior torso to the left flank and the low to mid back. At the end of activation (17/18 QRS), a new maximum appears at the site of early negativity. As recovery progresses, a dorsal minimum of low amplitude proceeds from the upper back to the upper anterior torso; near the peak of the T wave, the smooth distribution patterns start deteriorating, with

ST1

u

AM

3hU-AMI

FIGURE 3. The first half of ST-T is depicted

IMI

(see Fig. 1).

AN-NM

1, 1987

THE AMERICAN

JOURNAL

OF CARDIOLOGY

Volume 60

1233

the minimum migrating toward the left precordium, encroaching on the positive potential distribution and forcing it downward. The fourth column of Figures 1 through 4 depicts corresponding discriminant maps as they vary in time. These maps stress the significant differences existing at each instant between normal maps and maps from patients with anterior MI. Significant loss of electrical forces in the anterior torso occurs early (21'18QRS] in depolarization when actual mean voltages barely exceed f 20 pV. Peak negative discriminant indexes are at first located in the upper left precordial region, which presumably overlies the infarcted area, before moving near mid-QRS toward the lower anterior torso, under the xiphoid appendix. Meanwhile, Ithe peak positive indexes at first observed in the lower back proceed to the upper dorsal region as the negative discriminant indexes migrate downward. Peak negative indexes are observed between 131'18ST-T and 16/ 18 ST-T and coincide in time and space with the dip in positive voltages caused by the downward migration of the anterior minimum; here, too, reciprocal changes are oriented along a front to back axis. Inferior myocardial infarction: The fifth column of Figures 1 through 4 shows the time course of map distributions from 144 patients in the inferior MI group. In early activation (Z/18 and 3/18 QRS), the map features are similar to those of the normal population. Later, contrary to the normal subjects, the minimum located in the lower left flank and back migrates further downward with negative voltages progressively spreading over most of the back and the right half of the anterior torso. At the same time, the positive potentials, instead of extending toward the inferior border around the torso, are being enclosed on both sides by STT

N

FIGURE 4. The second half of ST-T is depicted

(see Fig. 1).

1234

SURFACE

MAPS IN ANTERIOR

AND INFERIOR MYOCARDIAL

INFARCTION

of inferior electrical forces not projected on the body surface or, more likely, unexplored by our electrode grid. During the first half of QRS, a symmetrical displacement of negative and positive indexes is noted: while the positive indexes moves horizontally from upper back to upper sternum, negative indexes proceed inferiorly toward the lower back, in contrast to the anterior MI group where opposite indexes moved jointly in a vertical direction. At the end of QRS, significant negative indexes are observed in the precordial area; they persist during ST, together with significant positive indexes in the mid-dorsal region. Peak indexes are reached immediately after the peak of the T wave: negative indexes dominate in the lower left flank and positive indexes in the upper thorax both anteriorly and posteriorly. Variability range in anterior and inferior myocardial infarction: Figure 5 depicts scattermaps and corresponding discriminant scattermaps at selected instants for the anterior MI group. From 5/18 to lo/18 A M / ExTKEMA A M INDEXES QRS, minima cluster very tightly in the midthoracic region while maxima spread from lower left chest to the back; this distribution seems to indicate the presence of a very homogeneous group of patients. Corresponding scattergrams of peak discriminant indexes, fhowever, show that while at 5/18 QRS, 90% of all QRS 5 anterior MI patients have significant loss of electrical forces concentrating in the upper left precordial area, at 7/18 QRS, this cluster breaks down: of 88% of patients with significant anterior negativity, only 40% remain grouped in the left precordium while 60% have moved over and under the xiphoid appendix; at B/18 QRS, these proportions are 5% and 95%, respectively. At lo/18 QRS, 75% of anterior MI patients still exhibit significant electrical losses; most of them are located in the inferior midchest. At 15/18 ST-T, 2 distributions of minima can be seen, one in the upper sternum, the other in the precordial area. The corresponding discriminant scattermap shows that 85% of the patients have significant negative discriminant inJ i ,/ *J , dexes, most of which cluster in the precordial area at the approximate site of loss of electrical forces in early depolarization. These findings suggest that at least 2 types of patients are present in the anterior MI population: a larger group with significant deficit of electrical forces extending in a base to apex direction and lasting until past mid-QRS, and a smaller group with abnormalities of shorter duration covering a smaller surface. Figure 6 summarizes the time course of peak positive and negative discriminant indexes in the anterior MI group. Figure 7 similarly displays scattermaps and disFIGURE 5. Scattergrams of voltage extrema (left column) and corcriminant scattermaps for,the inferior MI group. Early responding peak diicriminant indexes (right column) determined in depolarization (4/18 QRS), clusters of extrema can from the anterior Ml population (AMI) at,selected instants. Circles in be observed with maxima concentrating in the midthe extrema scattermaps represent minima and crosses, maxima, sternal region and minima in the lower left anterior observed in individual patients. When only 1 minimum or 1 maxiand posterior thorax. At 6/18 QRS, the minima scatter mum was observed at a givenelectrode site, it is not reported on the in the lower back and in the right anterior torso: by map. Circles and crosses in discriminant scatter maps represent mid-QRS, both the maxima and the minima cluster peak negative and positive discriminant, indexes, respectively, proagain: the former in the midsternal region, the latter in vided their value was equal to or exceeded 2 U (deviation from the left precordium. Corresponding discriminant scatnormal values by 2 or more standard deviations); again sites with tergrams indicate that most of the 39% of the patients only 1 positive or 1 negative peak index are not displayed.

negative potentials forming 2 parallel vertical separation lines (7/18 to 9/18 QRS]. During the second half of QRS, the negative voltages cover the entire anterior chest while the maximum proceeds toward the lower dorsal region. During recovery, a maximum appears in the midsternal area and remains there throughout repolarization with little inferior expansion to the right and to the left.Negative voltages of low amplitude appear in the back and in the upper right chest at the end of the ST interval, then migrate inferiorly and persist during the T wave. The sixth column of Figures 1 through 4 displays time-varying discriminant maps for the inferior MI group at corresponding instants. Significant indexes are first observed at 3/18 QRS in the upper dorsal region and indicate excess of electrical forces. This can be explained as a reciprocal effect due to the loss

A

r-‘-

A

F--

r--i?--

December

1,1987

THE AMERICAN

JOURNAL

OF CARDIOLOGY

1235

Volume 60

100%

a0

FIGURE 6. Number of anterior MI patients (expressed in percent) with significant (22 U) positive discriminant indexes (so/id /ine) and negative indexes (doffed /he) as they vary in time; 18 QRS and 18 ST-T equidistant instants are depicted.

60

40

STT 1

2

I

4

I

6

/

8

with early loss of electrical forces are grouped in the inferior right corner of the thorax; positive discriminant indexes from 59% of the patients are located in the upper dorsal area. At 6/18 QRS, the number of patients with abnormal negative indexes increases to i’l%, two-thirds of which cluster in the right anterior corner and one-third scatters over the mid- and left inferior torso. At the same time, significant reciprocal positive indexes are distributed in the upper back and the left subclavicular region in equal amounts. Just before mid-QRS, the amount of negative discriminant indexes in the lower right anterior torso decreases; a dominant cluster is noticed in the lower left flank (twothirds of the population) and a smaller one in the substernal region, Positive indexes from 81% of the patients regroup in the upper mid- and left thorax. At lO/ 18 QRS, 49% of the patients still exhibit significant negative discriminant indexes, half of which are located in the back. At peak T, maxima cluster in the precordial area and minima scatter in the inferior half of the torso; positive discriminant indexes from 81% of the patients are located mainly in the left subclavicular region while negative discriminant indexes from 83% of the patients are grouped primarily in the lower left flank. A smaller cluster is seen in the lower right anterior chest. The clusters and time course of discriminant indexes described during the first half of depolarization seem to point to at least 2 subgroups of patients within the inferior MI population: a first group showing early loss of electrical forces in the inferior right anterior torso and, somewhat later, another one appearing in the lower left flank, including two-thirds of the patients with abnormal negative discriminant indexes and coexisting with the first subgroup, which in the meantime became much smaller. Both distributions persist until after mid-QRS. The single reciprocal dorsal cluster of positive indexes seen at mid-QRS

I

10

I

12

/

14

I

16

I

18

I

2

i M I -

QRS

4

OR’S

6

QRS

IO

,

4

!

6

,

8

I

10

EXTREMA

t I

I

12

/

14

I M I -

/

16

I

18

INDEXES

0 ... ‘0

%

0 0

FIGURE 7. Scattergrams of voltage extrema and corresponding peak discriminant indexes from the inferior Ml (IMI) group at selected instants (see Fig. 5).

1236

SURFACE MAPS IN ANTERIOR

AND INFERIOR MYOCARDIAL

INFARCTION

and peak T suggests the existence of a considerable amount of overlap between both subgroups. Figure 8 shows the variation in time of peak negative and positive discriminant indexes in the inferior MI group.

Discussion Limitations of the study: Of 354 patients with nonelectrocardiographic evidence of MI, 258 were assigned to an anterior (114) or an inferior (144) group on the basis of commonly accepted QRS criteria measured on the 12-lead electrocardiogram at the time of the acute event. Very clear-cut groups emerge from this procedure; however, because the present study aims at analyzing typical map features in 2 broad classes of MI, this is not necessarily a disadvantage. Despite evidence of imperfect correlation between anterior-inferior electrocardiographic designation and underlying myocardial involvement, these terms were found acceptable as long as they indicate general location of necrosis.l*Jg Excluded from this study were patients with clinical and enzymatic evidence of MI, many of whom had ST-T changes and abnormal wall motion on nuclear imaging, which limits the conclusions of this study to typical Q-wave infarctions. In order to correlate the present findings to non-Q-wave MI, further investigation is required. Another important limitation is the discrepancy in age distribution between the normal control subjects and the MI population; 78% of normal subjects are 20 to 39 years old, but 92% of the MI patients are more than 40 years of age. Although this difference certainly accounts for part of the difference in voltage amplitudes, it does not account for the very large differences observed between the normal group and both MI populations [up to 5 standard deviations at peak QRS and ST-T). However, it is likely that measurements resulting from age-

matched groups would allow more accurate quantification of MI abnormalities. Interpretation and quantification of map patterns: It is generally recognized that myocardial necrosis is accompanied by loss of electrical forces previously generated by the intact tissue. When the loss is complete, negative voltages are recorded in electrodes facing the infarcted area; the most widely accepted theory for explaining this negative potential distribution is that of Wilson et alzOwho suggested that infarcted and electrically inert myocardium acts as a window for overlying electrodes that record negative cavity potentials. If viable cells are still present in the injured zone, reduced voltages will be recorded. In both cases, the new balance of electrical forces will produce reciprocal increases in potential in leads on the opposite side of the infarcted area. Postulating a direct physical relation between the surface potential distribution and the underlying cardiac electrical events, map patterns are expected to show a minimum of the anterior chest in anterior MI and in the inferior torso in inferior MI. Reciprocal maxima would then be seen in midback and upper torso regions, respectively. This simplistic view does not take into account the complexity of the conducting medium, the variability of heart-electrode geometry and the effects of abnormal activation sequences on surface potential distributions. Despite these limitations, mean group map features in both MI classes exhibited the aforementioned patterns in early depolarization. Although qualitative differences between normal and abnormal distributions are often visible, a large amount of overlap exists; by subtracting from successive MI patterns the range of potential distributions generated by the normal activation sequence, abnormal loss or excess of electrical forces is quantitatively determined and displayed. Repolarization abnormalities are indicated in a similar fashion; interestingly, their location in late recovery

I M I 100%

-

FIGURE 8. Time-varying number of inferior MI patients (expressed in percent) with significant (22 U) positive (solid Me) and negative (doffed/he) discriminant indexes at 18 QRS and 18 ST-T equidistant instants.

60

40

20

ORS 2

4

6

8

10

12

14

16

18

2

4

6

8

10

12

14

16

1.3

December

1, 1987

THE AMERICAN

JOURNAL

OF CARDIOLOGY

Volume 50

4237

Several features described in the present investigacorresponds in both groups to the site of early depolarization, This property should also contribute to the tion have been reported in other related studies: the presence of “sinks” facing infarcted region in early localization and the quantification of MI. Location and extent of anterior and inferior myo- QRS and in late T as well as opposite, reciprocal cardial infarction: Scattergrams provide information “sources” was described by Taccardi et alz3 The same on interindividual variability within each MI popula- group also reported late depolarization patterns that tion; identification of separate clusters of subjects, they attributed to absence of depolarization of fibers varying in location and time, could therefore lead to surrounding the necrotic tissue,3 whereas others assubdivisions within each class. Scattergrams of extre- cribed these patterns to delayed activation in periMinima distributed around the lowma are not suited for that purpose; for example, early necrotic areas.24,25 er half of the chest and migrating further down and to minima in both MI groups cluster remarkably well, the right anterior hemithorax have also been deconsidering the variety of electrocardiographic types-based on variable location of diagnostic Q scribed in inferior MI by Vincent at ak5 Flowers et al4 waves-accepted for inclusion in each class. This ho- and DeAmbroggi et al. 26Reciprocal positives were ofmogeneity is misleading because only minima are dis- ten observed in the upper dorsal and left subclavicular played, not the corresponding distributions of negative region by mid-QRS in inferior MI patients as well as voltages: 2 infarctions can have coinciding sites of early anterior maximan Using autopsy material in 22 minimum voltage and still exhibit quite different dis- patients with anterior MI, Matsushita et alz7concluded tributions of negative potentials. Information concern- that medium size anterior MIS demonstrated Q waves in the left anterior chest, whereas large anterior MIS ing location, extent and duration of electrical deficit can be derived from sequential discriminant scatter- involving the apex had Q waves extending to the lower anterior chest; these findings lend support to our hymaps; further quantification is provided by individual pothesis suggesting the presence of 2 subgroups in the discriminant maps. Analysis of sequential discriminant scattergrams in each class together with observa- anterior MI population. Montague et a1,28using isointion of individual discriminant maps has suggested the tegral contour maps and radionuclide evidence of abpossible presence of 2 types of infarctions in the ante- normal myocardium in a group of inferior MI patients, were able to define criteria for the separation of parior MI group [anterior and anterior with apical involvement] and at least 2 types in the inferior MI group tients with and without right ventricular involvement, (inferoposterior with and without right ventricular in- identifying greater areas of early QRS negativity and ST positivity in the right anterior-inferior chest. Revolvement, both with or without apical involvement). Our findings also underscore the frequent overlap be- ports on surface map repolarization patterns in MI are tween anterior and inferior MI, mainly because of scarce; the occurrence of abnormal positivities during apical involvement; this overlap results in part from ST and abnormal negativities at the end of T, both the particular blood supply to apical segments.l~~21~z2overlying the infarcted area, has been described by Another factor is the variability in heart-thorax geom- Taccardi et alz3 and Mirviszg etry, resulting in the projection of anterior apical asFuture developments will probably combine surpects in the inferior chest and vice versa; frequent face maps with epicardial maps, both measured and circumferential apical involvement further confuses computed, and with the various new techniques for electrocardiographic typingv21 cardiac imaging, yielding more refined and accurate Diagnostic criteria: Using only 1 measurement for spatial and quantitative electrical and mechanical, as the anterior MI group (negative index at 5/18 QRS in well as perfusion and biochemical, correlations. With the upper left precordial area) and 1 measurement for the use of data reduction methods and advanced statisthe inferior MI group (negative index at 14/18 ST-T in tical procedures, body surface potential mapping conthe lower left flank) 90% and 83% of the patients in the tributions to our understanding of pathophysiologic ischemic heart syndromes may be enhanced. respective groups could be correctly identified. Adding a second measurement in each group, namely, negAcknowledgment: The authors wish to thank Gaye ative index at 15/18 ST-T in the precordial region for anterior MI patients and positive index at lo/18 QRS A. Strong for expert secretarial assistance. in the upper sternal area for inferior MI patients, 93% and 8870, respectively, of the patients were recog- References nized. Using any of the 5 patterns depicted in Figures 5 1. Abildskov IA. Burness Ml. Urie PM. Lux RL. Wvatt RF. The unidentified dontent of the klectrocordiogram. ‘Cir”c Res 1977;40:3-7. ’ and 7 for inclusion of individual patients in 1 of the 2 2.information Kornreich F, Rautaharju PM. The missing waveform and diagnostic inforclasses, 96% of anterior MI patients and 93% of inferimotion in the standard 12.lead electrocardiogram. l EJectro&rdiol I&I; or MI patients were correctly assigned. Applying the x4:341-350. Taccardi B. De Ambroeai L. Viaanotti C. Characteristic features of surface same rules to the normal population, 6% of the sub- 3.potential maps during QR; and
1238

SURFACE

MAPS IN ANTERIOR

AND INFERIOR MYOCARDIAL

INFARCTION

ventricular hypertrophy. Circulation 1968:38:917-932. 7. Holt JH, Barnard ACL, Lynn MS. A study of the human heart as a multiple dipole electrical source. II. Diagnosis and quantification of left ventricular hypertrophy. Circulation 1969;46:69i’-710. 8. Stilli D, Musso E, Macchi E, Taccardi B, Rolli A, Aurier E, Favaro L, Botti G. Diagnostic value of body surface maps in left bundle-branch block. Adv Cardiof 1981;28:36-40. 9. Spach MS, Barr RC, Blumenschein SD, Boineau JP. Clinical implications of isopotential surface maps. Ann Intern Med 1968:69:919-928. 10. Boineau JP, Spach MS. The relationship between the electrocardiogram and the electrical activity of the heart. J Electrocardiol 1968;1:117-124. 11. Miller WT, Spach MS, Warren RB. Total body surface mapping during exercise: Q&?-T-wave changes in normal young adults. Circulation 1980; 62:632-645. 12. Spach MS, Barr RC, Warren RB, Benson DW, Walston A, Edwards SB. Isopotential body surface mapping in subjects of ail ages: emphasis on Iowlevel potentials with analysis of the method. Circulation 1979;59:805-821. 13. Taccardi B. Body surface distribution of equipotential lines during atria1 depolarization and ventricular repolarization. Circ Res 1966;19:865-878. 14. Green LS, Lux RL, Haws CW, Williams RR, Hunt SC, Burgess MJ. Effects of age, sex, and body habitus on QRS and ST-T potential maps of liO0 normal subjects. Circulation 1985;71:244-253. 15. Montague TJ, Smith ER, Cameron DA, Rautaharju PM, Klassen GA, Flemington CS, Horacek BM. Isointegral analysis of body surface maps: surface distribution and temporal variability in normal subjects. Circulation 1981; 63:1166-1172. 16. Taccardi B. Distribution of heart potentials on the thoracic surface of normal human subjects. Circ Res 1963;12:341-352. 17. Durrer D, van Dam RT, Freud GE, Janse MJ, Meijler FL, Arzbaecher RC. Total excitation of the isolated human heart. Circulation 1970;41:499-512. 18. Roberts WC, Gardin JM. Location of myocardial infarcts: a confusion of terms and definitions. Am J Cardiol 1978;42:868-872. 19. Williams RA, Cohn PF, Vokonas PS, Young E, Herman MV, Gorlin R. Electrocardiographic, arteriographic and ventriculographic correlations in

transmural myocardial infarction. Am ] Cardiol 1973;31:595-699. 20. Wilson FN, Johnston FD, Rosenbaum FF, Etlanger H, Kossman CE, Hecht H, Cotrim N, Menezez de Oliviera R, Scarsi R, Barker PS. The precordial electrocardiogram. Am Heart J 1944;27:19-85. 21. Sullivan W, Vlodaver Z, Tuna N, Long L, Edwards JE. Correlation of electrocardiographic and pathologic findings in healed myocardial infarction. Am J Cardiol 1978;42:724-732. 22. Savage RM, Wagner GS, Ideker RE, Podolsky SA, Hackel DB. Correlation of postmortem anatomic findings with electrocardiographic changes in patients with myocardial infarction. Retrospective study of patients with typical anterior and posterior infarcts. Circulation 1977;55:279-285. 23. Taccardi B, Musso E, De Ambroggi L. Potential fields of normal and ischemic hearts during rest, ventricular excitation and recovery. In: Hoffman I, ed. Vectorcardiographic 2. Amsterdam: North-Holland, 1971:142-145. 24. Flowers NC, Horan LG, Johnson JC. Anterior infarctional changes occurring during mid and late activation detectable by surface mapping technique. Circulation 1976;54:906-913. 25. Durrer D, Van Lier AAW, Biiller J. Epicardial and intramural excitation in chronic myocardial infarction. Am Heart J 1964;68:765-776. 26. De Ambroggi L, Bertoni T, Rabbia C, Landolina M. Body surface potential maps in old inferior myocardial infarction. Assessment of diagnostic criteria. J Electrocardiol 1986;19:225-234. 27. Matsushita S, Yamamoto Y, Murayama M, Kuramoto K, Murakami M. Correlation between body surface mapping and autopsy proved myocardial infarction. In: Yamada & Harumi K, Musha T, eds. Advances in Body Surface Potential Mapping. Nagoya, Japan: University of Nagoya Press, 1983: 213216. 28. Montague TJ, Smith ER, Spencer CA, Johnstone DE, Lalonde LD, Bessoundo RM, Gardner MJ, Anderson RN, Horacek BM. Body surface electrocardiographic mapping in inferior myocardial infarction: manifestations of feft and right ventricular involvement. Circulation 1983;67:665-673. 29. Mirvis DM. Body surface distributions of repolarization forces during acute myocardial infarction. I. Isopotential and isoarea mapping. Circulation 1980;62:878-887.