Temporal evolution of body surface map p
J. ELECTROCARDIOLOGY 17 (4), 1984, 319-328
Original Communications
Temporal Evolution of Body Surface Map Patterns
Following Acute Inferior Myocardial Infarction
BY TERRENCE J . MONTAGUE, M.D., ELDON R. SMITH, M.D.*, DAVID E. JOHNSTONE, M.D.,
C. ANNE SPENCER, B.Sc., LUCILLE D. LALONDE, M.D., RICARDO M. BESSOUDO, M . D . ,
MARTIN J. GARDNER, M.D., ROBERT N. ANDERSON, M . D . AND B. MILAN HORACEK, P H . D .
WITH THE TECHNICAL ASSISTANCE OF SttARON A. BLACK, R.N.
SUMMARY
We studied the evolution of body-surface potential map (BSPM) patterns in 32 patients following first acute inferior myocardial infarction. Initial BSPMs were obtained
at a mean of 79 hours post-infarction; follow-up BSPMs, a mean of eight months postinfarction. Temporal area-of-difference maps, constructed by subtracting initial from
follow-up group-mean BSPMs, revealed reciprocal changes over the superior and inferior
torso for both Q-zone and ST-segment time-integral distributions. The temporal changes
in Q-zone patterns were small but definite: over the inferior torso there was a relative
gain in Q-zone values and, over the superior torso, a relative decrease. In contrast, there
were marked spatial and quantitative changes of ST-segment distributions during the
follow-up period. Over the superior torso, particularly anteriorly, there was a gain in STsegment values; over the inferior torso, a decrease. With the small temporal changes in
Q-zone time-integral distributions, individual Q-zone maps continued to reflect a pattern
of inferior myocardial infarction at follow-up. In contrast, the marked temporal changes
in ST-segment time-integral distributions resulted in individual map patterns at followup that were nearly indistinguishable from normal ST-segment maps.
The relatively small changes in depolarization time-integral patterns during the early
post-infarction period suggest that the Q-zone patterns of the acute phase of myocardial
infarction reflect near-irreversible or completed myocardial damage. The marked normalization of repolarization time-integral patterns during the recovery phase suggests,
however, that there are also considerable areas of myocardium-at-risk during the early
phase of the infarction process which stabilize with time
myocardial ischemic injury 2"7. A potential problem of ST-segment mapping in this setting, however, is that complications of acute infarction
such as pericarditis or electrolyte imbalance m a y
induce ST-segment changes, thus diluting the
predictive value of such changes for exclusively
ischemic injury. Another problem, at least in the
assessment of infarct size, is that ST-segment
changes are thought to reflect primarily myocardial ischemia, rather than necrosis. Consequently,
other ECG parameters, such as QRS maps, have
more recently been utilized to assess infarct size
and degree of myocardial dysfunction following
acute myocardial infarction s'11.
In a previous study, we found that depolarization (Q-zone) and repolarization (ST-segment)
time-integral B S P M s provided valuable data in
patients with acute inferior myocardial infarction, reflecting the inferior site of myocardial inj u r y and discriminating between patient groups
with either exclusive left ventricular infarction or
The E C G has a well-defined role in the
diagnosis of acute myocardial infarction 1. The
ECG has also been utilized to assess the extent of
myocardial damage during infarction 2"11. Earlier
studies have focused on sequential changes in STsegment patterns during the very early phase of
From the Departments of Medicine and Physiology and
Biophysics, Dalhousie University and the Victoria General
Hospital, Halifax, Nova Scotia, Canada.
Supported by grants from the Nova Scotia Heart Foundation
and the Medical Research Council of Canada (PG-18).
*Current Address: Head, Cardiology Division, University of
Calgary, Calgary, Alberta, T2N 1N4.
The costs of publication of this article were defrayed in part
by the payment of page charges. This article must therefore
be hereby marked "advertisement" in accordance with 18
U.S.C. w1734 solely to indicate this fact.
~Reprint requests to: T.J. Montague, M.D., F.R.C.P.IC),
iDepartment of Medicine (Cardiology), Room 3054, Amibulatory Care Centre, Victoria General Hospital, Halifax,
N.S., Canada, B3tt 2Y9.
319
320
MONTAGUE El" AL
additional right ventricular involvement 12. T h e purpose of this s t u d y was to use similar B S P M techn i q u e s to m e a s u r e the t e m p o r a l c h a n g e s of
Q-zone and S T - s e g m e n t p a t t e r n s d u r i n g t h e posti n f a r c t i o n p e r i o d in an e f f o r t to assess their
c a p a c i t y for r e t u r n t o w a r d n o r m a l i t y .
MATERIALS AND METHODS
Patients.
Of 36 patients initially studied with BSPM during
the acute phase of a first inferior myocardial infarction TM, 32 were followed up with repeat BSPMs. These
32 patients form the present study group. There were
24 males and 8 females with an age range of 46-76
years. Of the four original patients not having followup BSPM, one died during the follow-up period of noncardiac causes; two developed serious systemic disease
FRRNKS X,YIZ
.
Fig. 1. The top panel illustrates superimposed Frank
X, Y and Z leads of one study subject; the vertical lines
indicate QRS-onset, QRS-offset and ST-T offset,
respectively. The bottom panel illustrates the signalaveraged recording from a right anterior-inferior torso
lead from the same subject (lead 23; 10 cm inferior to
the V1 torso position 14. The horizontal arrows indicate
the various ECG time-integrals; Q-zone and STsegment integrals are indicated by the diagonal lines
and stippling respectively. See Methods for details.
and were unable to return; and one refused follow-up
study. At the time of initial BSPM, all of the study
subjects had compatible histories, cardiac-enzyme and
12-lead electrocardiographic abnormalities diagnostic
of acute inferior infarction, as well as 2 + or greater (according to the method of Berman et al.13) focal, 99m
technetium pyrophosphate scintigraphic activity involving the inferior-posterior left ventricular wall on
myocardial scans performed within 72 hours of admission 12. The initial BSPMs were recorded at a mean of
79 hours (range 19 to 213 hours) after onset of symptoms. The time of follow-up BSPMs ranged from 5 to
15 months and averaged 8 months.
Normal Control Subjects.
A control group of 42 clinically normal subjects, 25
males and 17 females, aged 21 to 66 years, was studied.
All of the normal subjects had normal physical and
echocardiographic examinations and 12-lead ECG tracings. BSPMs were recorded on two separate occasions,
with separate application of electrodes, in each normal
subject. The average time between the initial and
follow-up BSPM studies averaged three months
(range, one week to seven months).
Protocol.
All BSPM recordings were obtained in the Coronary
Care Unit of the Victoria General Hospital with the
subjects in the resting supine position and during tidal
volume respiration 1~. At the time of initial and followup recording, all study subjects were in sinus rhythm
and none had bundle branch block. Individual maps, including Q-zone, QRS, ST-T, ST-segment and QRST,
were plotted for all subjects at the time of initial and
follow-up study. Group-mean Q-zone and ST-segment
time-integral maps were also constructed for the two
groups of patients at both study times. The technique
of plotting group-mean maps has been previously
reported 14. Briefly, for each time integral assessed,
isointegral BSPMs were constructed from the groupmean values at each lead site.
To further assess spatial and quantitative changes in
time-integral distributions with time, area-of-difference
maps were constructed for normal and patient groups
by subtracting initial group-mean Q-zone and STsegment maps from the respective group-mean Q-zone
and ST-segment maps recorded at follow-up. As previously described, the areas-of-difference identified by
the subtraction or difference maps, Served as areas of
interest in each individual subject 12. Within the respective areas-of-difference, the average value of Q-zone
and ST-segment integrals were quantitated for each individual at the time of initial and follow-up examinations 12. Temporal variability in time-integral torso
values was also globally assessed by utilizing a root
mean square index of intermap variability according to
the previously described formulal4'ls:
v=
~zN1 (XT2 -XT1)2
~-
J. ELECTROCARDIOLOGY 17 (4), 1984
BSPM PATTERNS IN MI
Where N is the number of lead sites, X represents the
respective time-integral values at a specific lead site,
and T1 and T2 refer to initial and follow-up map acquisition times.
BSPM Recording.
The system of BSPM recording utilized in our
laboratory has been previously described 14. In summary, the system utilizes a Varian V72 computer and
obtains signals simultaneously from 117 torso and 3
limb leads, using Wilson's central terminal as
reference, at a sampling frequency of 500 samples/
second/channel and with a sampling duration of 15 continuous seconds. The system has a resolution of 10
t~volts in the dynamic range of _+ 5 mvolts.
BSPM Processing.
Our system of ECG signal processing has been
described it. The program allows selective averaging of
P-QRS-T complexes at each lead site with deletion of
artifacts and ectopic beats and correction of baseline
drift. Measurements of specific ECG time integrals of
the averaged torso complexes are obtained using time
instants determined from edited Frank X, Y and Z
leads (Fig. 1).
Several time integrals may be evaluated. They are
graphically illustrated in Fig. 1. Briefly, the QRS time
integral is the algebraic sum of all potentials from the
time instant of QRS onset to QRS offset multiplied by
the sampling interval -- this represents the net area
under the curve with PR interval as baseline. The ST-T
time integral is similarly calculated from the time instant of QRS offset to T offset; the QRST time integral
represents the algebraic sum of QRS plus ST-T time integrals. The Q-zone time integral is calculated from the
QRS onset to the mid-point of the QRS complex and
the ST-segment from the first 3/8 of the ST-T segment.
All time integrals are expressed as microvolt.seconds
(gV. sec). In this study of patients with myocardial infarction, the analysis focused on Q-zone and St-segment
time integrals.
BSPM Display.
The isointegral map display format utilized for all
time-integral surface distributions is illustrated in
Fig. 2. In individual and group-mean maps, the solid
lines represent positive values; the interrupted lines,
negative values. In temporal subtraction or area-ofdifference maps, the solid lines represent areas in
which the integral values were greater at follow-up
than initial values; interrupted lines designate areas in
which integral values were less at follow-up t h a n
values initially. In all maps, the contour lines extend in
a logarithmic progression from the zone of near-zero
integral values. Each decade line, as well as the maximum and minimum, are numerically identified.
Statistical Analysis.
All grouped data are expressed as mean _+ 1 S.D.
For both normal and patient groups, the null
hypotheses that there were no differences in average
Q-zone and ST-segment time-integral values within
temporal areas-of-difference were tested by t-tests for
J. ELECTROCARDIOLOGY 17 (4), 1984
r'"
.!
321
]
;
9
..
-10
/t
i
f
Fig. 2. Isointegral map display format. The rectangle
represents the torso with the left half indicating the
front, and the right half, the back. The locations of all
torso electrode sites are indicated by the small dots.
The standard ECG V1 to V6 lead sites are indicated by
the small squares. The isocontour lines connect points
of equal time-integral value. This Q-zone map of a
study subject with acute inferior myocardial infarction
indicates a distribution of negative Q-zone values over
the inferior torso 12.
paired data. The null hypotheses that there were no differences in Q-zone and ST-segment root mean square
temporal variability between normal and patient
groups were also tested by t-tests for unpaired data.
RESULTS
The initial and follow-up group-mean Q-zone
isointegral maps, as well as the mean Q-zone areaof-difference maps, of the control and p a tie n t
groups are illustrated in Fig. 3. B o t h initial and
follow-up maps of the pat i ent group reflect the
p a t t e r n of inferior myocardial infarction with abnormal distributions of negative Q-zone integrals
over the inferior torso and increase of positive
distributions over the superior precordium 12.
A l t hough t he surface distributions of the Q-zone
integrals were similar at the two mapping times,
t here Was, however, some small spatial diminution of the negative inferior torso Q-zone distributions and positive superior distributions. This is
confirmed b y the Q-zone subtraction map {Fig. 3).
The group-mean Q-zone. maps of the control
group showed near-identical initial and follow-up
p a t t e r n s and the relative lack of spatial change
with time was supported by the small areas-ofdifference of the subtraction map (Fig. 3}.
The mean ST-segment isointegral maps of the
control and pat i ent groups are shown in Fig. 4.
The initial pat i ent group map shows an abnormal
d i s t r i b u t i o n of positive S T - s e g m e n t integral
values over the inferior torso and a negative
322
MONTAGUE ET AL
Q-ZONE
INFERIOR INFARCTION
NORMAL
T1
....::" /
"",.
~:
- 13
."=:7""',
'-'~
T2
T2
minus
T1
9
...~
""...... ..
NEGATIVE AREA
15
T1
T2
",
-1
~
POSITIVE AREA
T1
T2
/f
NEGATIVE AREA
34]
T1
POSITIVEAREA
T2
T1
T2
,/,
10
5
t~
0
~-5
iii'
'i~l,
~ "-'18-e
[~
r
(u
-10
0.
-15-
:
~..j.~
_25.]
e
=
-6-12-
p< 0.001
p< 0.05
p< 0.02
p< 0.005
Fig. 3. Comparison of mean initial (T1), follow-up (T2) and difference (T2 - T1) Q-zone maps from 42 normal subjects
and 32 patients with inferior myocardial infarction. Individual average Q-zone values within the temporal areas-ofdifference at initial and follow-up study are illustrated below the respective difference maps. Normal subjects had
small spatial changes with time. In contrast, inferior infarction patients had spatially greater temporal change with
an increase of Q-zone integral values inferiorly and a decrease anteriorly and superiorly. Quantitatively, both groups
had similar degrees of change within their respective areas-of-difference.
J. ELECTROCARDIOLOGY17 (4), 1984
BSPM PATTERNSIN M]
323
ST
INFERIOR INFARCTION
NORMAL
;',
".-....7~..:
....":
"-.
~
.......... "-.:--.-:~ -..._::--- ...... .-
~
T1
...................
-~
. ..........................
~~ :::::::::::::::::::::::::::::::::::::::::
'
....
--.-,o........ )( . . /
iiiiiiiiiiiiiiiiiioiiiiii'
T2
T2
minus
T1
~..1.:
~,f:_!~?..!!!!!:~;,i
... ~;;.........-,,..............
50-
NEGATIVE AREA
NEGATIVE AREA
T1
T2
15"
T1
T2
POSITIVEAREA
T1
T2
10"
4030-
Q,3
~
0
(fr
,rTr
nil
')~);~)
-(~
K
,
2010
0
--15
p = N.S.
p< 0.001
p< 0.001
Fig. 4. Comparison of mean initial (T1), follow-up (T2) and difference (T2 - T1) ST-segment maps of 42 normal subjeers and 32 patients with acute inferior myocardial infarction. Individual-average ST-segment integral values
within the temporal areas-of-difference are illustrated below the respective difference maps. Normal subjects had no
significant spatial or quantitative change in ST-segment distributions with time. In contrast, there were marked
spatial and quantitative changes with near-normalization of initially abnormal ST-segment distributions with time
in the inferior infarction group.
J, ELECTROCARDIOLOGY17 (4), 1984
324
MONTAGUE ET AL
distribution over the superior torso, compatible
with acute inferior myocardial infarction 12. At
follow-up, however, the patient group had a nearnormal distribution of ST-segment integrals
(Fig. 4). In comparison to the small spatial
changes observed in Q-zone distributions over the
average eight month follow-up interval (Fig: 3),
these spatial changes in ST-segment distributions were marked. Comparison of the ST
distributions of the normal group reveals no apparent change between initial and follow-up patterns and almost no area-of-difference (Fig. 4).
Quantitative comparisons of individual subjects' average integral values within the Q-zone
and ST-segment areas~f-difference are illustrated
by the data columns below the respective groupmean subtraction maps (Figs. 3, 4). The normal
group d e m o n s t r a t e d significant changes in
average integral values within the small Q-zone
areas-of-difference but no change within the very
small ST area-of-difference. In contrast, the infarction p a t i e n t s d e m o n s t r a t e d statistically
significant changes in all Q-zone and ST areas-ofdifference. The relative degree of quantitative
change within the patient groups was, however,
greater for ST-segment values.
The root mean square intermap variability of
Q-zone and ST time integrals for both groups is
compared in Fig. 5. The temporal change, as
assessed by this global measure of the change in
torso distribution, was significantly greater in in-
INDEX OF VARIABILITY
ST
Q - ZONE
INFERIOR
NORMAL INFARCTION
( n = 42) ( n = 32)
111.
81
18-
NORMAL
1n=42)
9
9
16-
9
10 ~
9149
9
6
,~
t,
9
0
6"
4
(co
'i|
((t
(((0
~
(r
r
(0
o-
INFERIOR
INFARCTION
(n=32)
p< 0.001
4-
9
2-
';
c((C(o
(((((($
Ill
9
9
oJ~)
1- 9
({((((((~ 9
(((r(( 9
p< 0.001
Fig. 5. Comparison of Q-zone and ST-segment root,
mean square temporal variability from initial iT1) to
follow-up (T2) maps in both study groups.
farction patients for both Q-zone and ST-segment
values.
Individual Q-zone and ST-segment is 9
maps of two representative patients and a normal
control subject are compared in Figs. 6 and 7,
respectively. One (patient A, Fig. 6) had relatively little spatial or quantitative regression of
Q-zone infarction distributions. Another {patient
B, Fig. 6) showed considerable change, with gain
in Q-zone values inferiorly and regression or
reduction of values superiorly. Thus, as might be
expected, t h e r e is c o n s i d e r a b l e i n d i v i d u a l
variability of post-infarction Q-zone evolutionary
changes. In a similar fashion, the ST-segment
patterns of the same patients also demonstrate
individual variability in the evolution of this
repolarization integral (Fig. 7). Patient A, who
had relatively little change of Q-zone distributions (Fig. 6), had near-normalization of ST patterns (Fig. 7). Patient B, on the other hand, who
did demonstrate some temporal normalization of
initially abnormal Q-zone patterns (Fig. 6), had
initially abnormal ST distributions which evolved
to an equally abnormal distribution at follow-up
(Fig. 7).
DISCUSSION
The results of this study demonstrate that during the post-infarction period in patients with inferior myocardial infarction there was a relative
temporal gain in Q-zone integral values over the
inferior torso and diminution of values over the
superior precordium. The data also showed a temporal gain in ST-segment integral values over the
precordium and a concomitant reduction of STsegment values over the inferior torso. Thus, similar to the reciprocal torso distributions of Q-zone
and ST-segment during acute infarction le, there
was a reciprocity of the torso distributions of the
post-infarction evolutionary changes of these
depolarization and repolarization time integrals.
Although there was a similar directional
change in post-infarction Q-zone and ST-segment
patterns in that both integrals showed regression
from initially a b n o r m a l d i s t r i b u t i o n s , t h e
magnitude of change of the respective integrals
was not similar. Rather, the changes in the
repolarization (ST-segment) time integral patterns were spatially and quantitatively much
greater than those of the depolarization time integral (Q-zone). At the time of follow-up examination, the ST-segment maps had returned to a
near-normal distribution while Q-zone maps continued to show a pattern of inferior infarction.
J. ELECTROCARDIOLOGY 17 (4), 1984
BSPM
PATTERNS
IN M I
325
regional wall motion abnormalities with time
may be slight. Nonetheless, some measurable
diminution of abnormal Q-zone distributions was
apparent during the post-infarction period. It remains to be determined whether the speed or extent of changes in the Q-zone (and ST-segment)
integral patterns during the early post-infarction
period parallels, or is predictive of, the changes in
segmental or global myocardial wall function or
late clinical course. Likewise, the exact temporal
pattern of the BSPM pattern changes, and the influence of the size and site of the original myocardial infarction on these changes, are as yet
unknown. Further studies focusing on the recovery processes following myocardial infarction are
required to better define the natural history of
this part of the infarction process and to explore
the nature of the apparent protective effect(s) provided by interventions, such as beta-blockade,
during this phase ]6.
One explanation for the marked changes in STsegment patterns is that a large part of the initially ischemic myocardial tissue improved its
blood supply during the post-infarction period
with consequent partial normalization of repolarization activity. If this theory is correct, then its
corollary is probably also true: that is, because of
the relatively small changes in Q-zone distributions with time, the initially abnormal Q-zone patterns probably reflected largely irreversible
myocardial damage. In this study, correlations of
Q-zone and ST-segment pattern changes with
changes in myocardial wall motion abnormalities
and ventricular function were not attempted.
However, a recent study has confirmed that the
presence of abnormal Q-waves is highly correlated with the presence of underlying areas of
myocardial dysfunctionlL Thus, the small change
in abnormal Q-zone distributions observed in this
study suggests that the degree of improvement in
Q-ZONE
NORMAL
PATIENT A
,.--..
i., .............
-" ..
.. \
,.:........
:::'.'-::::.'-"-:.-,
, _...;--~~.
T1
".--i.."
X
"-'.'.
:,
,~--,
"- ",
...-;;.."
.."
Z
i
.-
.....
l
,.":;"
.." .-'"
"" - . . . .
.....
.. .......
.........
:"---t,
--.;.::.. ,; ..~!.7
~ , ~ 7 ~ "
_._7,_~
-15
"-
"" iii!-l
',, ",.~
li
.."
~;-"
PATIENT B
!
V'-~L:-. - .......... - - - ' - .
-
.
..-,:-
,,~-~-.~..~
/i
/t/---.~'~.
:"
..'..: .."
..... ...
I
-
.........
:._~ ....
.
."
.:
..
t
"/.- .......
,
"
i
/
-
7"7]
"
!:w. -.-8C.;,
"2'
":.........
-:"""
-,. ,..,- .......
,,'A,
.
: .;- ......
.'~c-~.:.~
-",
.--
7
~:
I
-.
'.,
T2
t ::
"-,
L/"- ..... 7:7
T2
minus
T1
R
!"/-a
"-,
.; :"
;"
7?,,'
........ J);)
,," :- ,.'
....
-7
/---,.
-i
..:"- '.!. _~:"",..
:
**
"-..,"
., ....
Fig. 6. Comparison of initial (T1), follow-up(T2) and difference (T2 - T1) Q-zone maps of a normal subject and two
representative inferior infarction patients. Patient A had very little change in inferior torso distributions; patient B,
in contrast, had marked gain in Q-zonevalues inferiorly. There was no obvious spatial change in Q-zonedistributions
in the normal subject, and the quantitative differences identified by the difference maps were small and irregularly
distributed.
J. E L E C T R O C A R D I O L O G Y
17 (4), 1 9 8 4
326
M O N T A G U E ET AL
minimum value), and a small area-of-difference,
group size may have to be relatively large to
avoid a n alpha error. Individual subtraction
maps, with each subject as his own control, may,
therefore, be more appropriate for quantitation of
temporal change, particularly for correlation with
changes in other parameters of myocardial function. Until more experience is gained in the
utilization of such techniques, we feel that caution in interpretation of results is necessary.
Specifically, the size and torso location of surface
areas-of-difference and their spatial pattern must
all be carefully assessed in terms of biological
meaning and appropriateness -- it is not enough
to simply apply statistical tests to integral values
within group-mean areas-of-difference.
In summary, BSPM techniques allow location
and quantitation of changes in depolarization and
repolarization time-integral patterns during the
period following acute myocardial infarction and
The area-of-difference or subtraction BSPM
methodology, which was used to assess spatial
and quantitative temporal ECG changes in this
study, is a very sensitive indicator of temporal
electrocardiographic differences. The spatial
areas-of-difference identified by subtraction of
group-mean maps provided, in fact, the major
areas of interest for study of temporal change.
Quantitation of integral values within the groupmean areas-of-difference for each individual
allowed quantitative comparisons within groups
over the time of this study. Since t h e area of interest is determined from the mean difference,
this method may not, however, be the ideal
method to reflect individual differences that
deviate widely from the mean. In addition, why
normal subjects should display temporal areas-ofdifference for Q-zone distributions is uncertain. It
may be that when dealing with a time integral
with a small data range {maximum minus
ST
".,
"l~"
:" !
"'.......J.+J
..... :::;.
"..:
PATIENT B
PATIENT A
NORMAL
..*"
,'"
"L
:-',o,-.....,-"'--.,
-il
. . ...........
..
""
,,,,..,-:
.:..- ....-.!
i
~"----.'..........
........ :::~".. ':,
T1
/"
=======================
..""~:i ""-:~.......... '
"'k ":" .. :;.,
"".i!i............(i;
I~
"
.::::::
.......
- . . . . -__.._ . . . . . . . . . . .
T2
..., ):::::::::.
T2
minus
T1
:
!ii
Fig. 7. Comparison of initial (T1), follow-up {T2) and difference (T2 - T1) ST-segment maps in the same study subjects as illustrated in Fig. 6. Patient A had near-normalization of ST-segment distributions with time; patient B, in
contrast, although having marked change in his ST-segment distributions with time, continued to have an abnormal
torso distribution at T2. No temporal changes in spatial ST-segment distributions of the normal subject were evident although the difference map showed small and irregularly distributed quantitative changes.
J. ELECTROCARDIOLOGY 17 (4), 1984
BSPM PATTERNS IN MI
offer promise as one means of assessing recovery
from myocardial infarction. Additional, m u l t i p l e
p a r a m e t e r q u a n t i t a t i v e and correlative s t u d i e s
are required to b e t t e r delineate the n a t u r a l
history of this i m p o r t a n t phase following acute
myocardial infarction and injury.
9.
Acknowledgement: We are grateful to Mary Jayne
Tilley and Gaye Strong for expert secretarial
assistance.
I0.
REFERENCES
1. ALPERT, J S AND BRAUNWALD,E: Pathological and
Clinical Manifestations of Acute Myocardial Infarction. In: Heart Disease, E BRAUNWALD,Ed.
Saunders, Philadelphia, 1980, pp 231-246, 1333
2. MAROKO,P R, Lmnv, P, COVELL,J W, SOBEL,B E,
Ross, J AND BRAUNWALD, E: Precordial STsegment elevation mapping: An atraumatic
method for assessing alterations in the extent of
myocardial ischemic injury. Am J Cardiol 29:223,
1972
3. FLAIIERTY,J T, REID, P R, KELLY,D T, TAYLOR,D
R , WEISFELDT, M L AND PITT, B: Intravenous
nitroglycerin in acute myocardial infarction. Circulation 51:132, 1975
4. MULLER,J E, MAROKO,P R AND BRAUNWALD,E:
Evaluation of precordial electrocardiographic mapping as a means of assessing changes in myocardial ischemic injury. Circulation 52:16, 1975
5. MARORO,P R, DAVIDSON,D M, LIBBY, P, HAGAN,
A D AND BRADNWALD,E: Effects of hyaluronidase
administration on myocardial ischemic injury in
acute infarction: A preliminary study in 24 patients. Ann Intern Med 82:516, 1975
6. GOLD, H K, LEINBACH, R C AND MAROKO, P R:
Propranolol-induced reduction of signs of ischemic
injury during acute myocardial infarction. A m J
Cardiol 38:689, 1976
7. MADIAS, J E, MADIAS, N E AND HOOD, W B:
Precordial ST-segment mapping II. Effects of oxygen inhalation on ischemic injury in patients with
acute myocardial infarction. Circulation 53:411,
1976
8. HILLIS, L D, ASKENAZI, J, BRAUNWALD, E, RADVANY, P, MULLER, J E, FISHREIN, M C AND
MAROKO, P R: Use of changes in the epicardial
Q R S complex to assess interventions which
J. ELECTROCARDIOLOGY 17 (4), 1984
11.
12.
13.
14.
15.
327
modify the extent of myocardial necrosis following
coronary artery occlusion. Circulation 54:591, 1976
MAROKO,P R, HILLIS,L D, MULLER,J E, TAVAZZI,
L, HEYNDRICKX, G R, RAY, M, CIIIARIELLO, M,
DISTANTE, A, ASKENZAI, J, SALERNO, J, CARPENTIER, J, RESHETNAYA, N I, RADVANY, P, LIBBY, P,
RAABE, D S, CHAZOV, E I, BOBBA, P AND BRAUNWALD, E: Favorable effects of hyaluronidase on
electrocardiograhicevidence of necrosis in patients
with acute myocardial infarction. N Engl J M e d
296:898, 1977
HENNING, H, HARDARSON, T, FRANCIS, G,
O'RoURKE, R A, RYAN, W ANn ROSS, J: Approach
to the estimation of myocardial infarct size by
analysis of ST-segment and R wave maps. A m J
Cardiol 41:1, 1978
PALMERI, S T, HARRISON, D G, CoaB, F R, MORRIS,
K G, HARRELL,F E, IDEKER,R E, SELVESTER, R H
AND WAGNER, G S: A QRS scoring system for
assessing left ventricular function after myocardial infarction. N Engl J Med 306:4, 1982
MONTAGDE, T J, SMITI{, E R, SPENCER, C A,
JOtlNSTONE, D E, LALONDE,L D, BESSOUDO, R M,
GARDNER, M J, ANDERSON R N AND HORACEK, B
M: Body surface electrocardiograhic mapping in
inferior myocardial infarction: Manifestations of
left and right ventricularinvolvement. Circulation
67:665, 1983
BERMAN, D S, AMSTERDAM, E A, HINES, H H,
SALEL, A F, BAILEY, G J, DENARDO, G L AND
MASON,D T: A new approach to the interpretation
of technetium 99m pyrophosphate scintigraphy in
the detection of acute myocardial infarction.
Clinical assessment of diagnostic accuracy. Am J
Cardiol 39:341, 1977
MONTAGUE, T J, SMITII, E R, CAMERON, D A,
RAUTAIIARJU, P M, KLASSEN, G A, FLEMINGTON, C
S AND HORACEK, B M: Isointegralanalysis of body
surface maps: Surface distribution and temporal
variabilityin normal subjects:Circulation63:1166,
1981
SUTIIERLAND, D J, McPIIERSON, D D, SPENCER, C
A, ARMSTRONG, C S, HORACEK, B M AND MONTAGUE, T J: Effects of posture and respiration on
the body surface electrocardiogram. A m J Cardiol
52:595, 1983
16. The Norwegian Multicenter Study Group: Timololinduced reduction in mortality and reinfarctionin
patients surviving acute myocardial infarction.N
Eng J M e d 304:801, 1981
Original Communications
Temporal Evolution of Body Surface Map Patterns
Following Acute Inferior Myocardial Infarction
BY TERRENCE J . MONTAGUE, M.D., ELDON R. SMITH, M.D.*, DAVID E. JOHNSTONE, M.D.,
C. ANNE SPENCER, B.Sc., LUCILLE D. LALONDE, M.D., RICARDO M. BESSOUDO, M . D . ,
MARTIN J. GARDNER, M.D., ROBERT N. ANDERSON, M . D . AND B. MILAN HORACEK, P H . D .
WITH THE TECHNICAL ASSISTANCE OF SttARON A. BLACK, R.N.
SUMMARY
We studied the evolution of body-surface potential map (BSPM) patterns in 32 patients following first acute inferior myocardial infarction. Initial BSPMs were obtained
at a mean of 79 hours post-infarction; follow-up BSPMs, a mean of eight months postinfarction. Temporal area-of-difference maps, constructed by subtracting initial from
follow-up group-mean BSPMs, revealed reciprocal changes over the superior and inferior
torso for both Q-zone and ST-segment time-integral distributions. The temporal changes
in Q-zone patterns were small but definite: over the inferior torso there was a relative
gain in Q-zone values and, over the superior torso, a relative decrease. In contrast, there
were marked spatial and quantitative changes of ST-segment distributions during the
follow-up period. Over the superior torso, particularly anteriorly, there was a gain in STsegment values; over the inferior torso, a decrease. With the small temporal changes in
Q-zone time-integral distributions, individual Q-zone maps continued to reflect a pattern
of inferior myocardial infarction at follow-up. In contrast, the marked temporal changes
in ST-segment time-integral distributions resulted in individual map patterns at followup that were nearly indistinguishable from normal ST-segment maps.
The relatively small changes in depolarization time-integral patterns during the early
post-infarction period suggest that the Q-zone patterns of the acute phase of myocardial
infarction reflect near-irreversible or completed myocardial damage. The marked normalization of repolarization time-integral patterns during the recovery phase suggests,
however, that there are also considerable areas of myocardium-at-risk during the early
phase of the infarction process which stabilize with time
myocardial ischemic injury 2"7. A potential problem of ST-segment mapping in this setting, however, is that complications of acute infarction
such as pericarditis or electrolyte imbalance m a y
induce ST-segment changes, thus diluting the
predictive value of such changes for exclusively
ischemic injury. Another problem, at least in the
assessment of infarct size, is that ST-segment
changes are thought to reflect primarily myocardial ischemia, rather than necrosis. Consequently,
other ECG parameters, such as QRS maps, have
more recently been utilized to assess infarct size
and degree of myocardial dysfunction following
acute myocardial infarction s'11.
In a previous study, we found that depolarization (Q-zone) and repolarization (ST-segment)
time-integral B S P M s provided valuable data in
patients with acute inferior myocardial infarction, reflecting the inferior site of myocardial inj u r y and discriminating between patient groups
with either exclusive left ventricular infarction or
The E C G has a well-defined role in the
diagnosis of acute myocardial infarction 1. The
ECG has also been utilized to assess the extent of
myocardial damage during infarction 2"11. Earlier
studies have focused on sequential changes in STsegment patterns during the very early phase of
From the Departments of Medicine and Physiology and
Biophysics, Dalhousie University and the Victoria General
Hospital, Halifax, Nova Scotia, Canada.
Supported by grants from the Nova Scotia Heart Foundation
and the Medical Research Council of Canada (PG-18).
*Current Address: Head, Cardiology Division, University of
Calgary, Calgary, Alberta, T2N 1N4.
The costs of publication of this article were defrayed in part
by the payment of page charges. This article must therefore
be hereby marked "advertisement" in accordance with 18
U.S.C. w1734 solely to indicate this fact.
~Reprint requests to: T.J. Montague, M.D., F.R.C.P.IC),
iDepartment of Medicine (Cardiology), Room 3054, Amibulatory Care Centre, Victoria General Hospital, Halifax,
N.S., Canada, B3tt 2Y9.
319
320
MONTAGUE El" AL
additional right ventricular involvement 12. T h e purpose of this s t u d y was to use similar B S P M techn i q u e s to m e a s u r e the t e m p o r a l c h a n g e s of
Q-zone and S T - s e g m e n t p a t t e r n s d u r i n g t h e posti n f a r c t i o n p e r i o d in an e f f o r t to assess their
c a p a c i t y for r e t u r n t o w a r d n o r m a l i t y .
MATERIALS AND METHODS
Patients.
Of 36 patients initially studied with BSPM during
the acute phase of a first inferior myocardial infarction TM, 32 were followed up with repeat BSPMs. These
32 patients form the present study group. There were
24 males and 8 females with an age range of 46-76
years. Of the four original patients not having followup BSPM, one died during the follow-up period of noncardiac causes; two developed serious systemic disease
FRRNKS X,YIZ
.
Fig. 1. The top panel illustrates superimposed Frank
X, Y and Z leads of one study subject; the vertical lines
indicate QRS-onset, QRS-offset and ST-T offset,
respectively. The bottom panel illustrates the signalaveraged recording from a right anterior-inferior torso
lead from the same subject (lead 23; 10 cm inferior to
the V1 torso position 14. The horizontal arrows indicate
the various ECG time-integrals; Q-zone and STsegment integrals are indicated by the diagonal lines
and stippling respectively. See Methods for details.
and were unable to return; and one refused follow-up
study. At the time of initial BSPM, all of the study
subjects had compatible histories, cardiac-enzyme and
12-lead electrocardiographic abnormalities diagnostic
of acute inferior infarction, as well as 2 + or greater (according to the method of Berman et al.13) focal, 99m
technetium pyrophosphate scintigraphic activity involving the inferior-posterior left ventricular wall on
myocardial scans performed within 72 hours of admission 12. The initial BSPMs were recorded at a mean of
79 hours (range 19 to 213 hours) after onset of symptoms. The time of follow-up BSPMs ranged from 5 to
15 months and averaged 8 months.
Normal Control Subjects.
A control group of 42 clinically normal subjects, 25
males and 17 females, aged 21 to 66 years, was studied.
All of the normal subjects had normal physical and
echocardiographic examinations and 12-lead ECG tracings. BSPMs were recorded on two separate occasions,
with separate application of electrodes, in each normal
subject. The average time between the initial and
follow-up BSPM studies averaged three months
(range, one week to seven months).
Protocol.
All BSPM recordings were obtained in the Coronary
Care Unit of the Victoria General Hospital with the
subjects in the resting supine position and during tidal
volume respiration 1~. At the time of initial and followup recording, all study subjects were in sinus rhythm
and none had bundle branch block. Individual maps, including Q-zone, QRS, ST-T, ST-segment and QRST,
were plotted for all subjects at the time of initial and
follow-up study. Group-mean Q-zone and ST-segment
time-integral maps were also constructed for the two
groups of patients at both study times. The technique
of plotting group-mean maps has been previously
reported 14. Briefly, for each time integral assessed,
isointegral BSPMs were constructed from the groupmean values at each lead site.
To further assess spatial and quantitative changes in
time-integral distributions with time, area-of-difference
maps were constructed for normal and patient groups
by subtracting initial group-mean Q-zone and STsegment maps from the respective group-mean Q-zone
and ST-segment maps recorded at follow-up. As previously described, the areas-of-difference identified by
the subtraction or difference maps, Served as areas of
interest in each individual subject 12. Within the respective areas-of-difference, the average value of Q-zone
and ST-segment integrals were quantitated for each individual at the time of initial and follow-up examinations 12. Temporal variability in time-integral torso
values was also globally assessed by utilizing a root
mean square index of intermap variability according to
the previously described formulal4'ls:
v=
~zN1 (XT2 -XT1)2
~-
J. ELECTROCARDIOLOGY 17 (4), 1984
BSPM PATTERNS IN MI
Where N is the number of lead sites, X represents the
respective time-integral values at a specific lead site,
and T1 and T2 refer to initial and follow-up map acquisition times.
BSPM Recording.
The system of BSPM recording utilized in our
laboratory has been previously described 14. In summary, the system utilizes a Varian V72 computer and
obtains signals simultaneously from 117 torso and 3
limb leads, using Wilson's central terminal as
reference, at a sampling frequency of 500 samples/
second/channel and with a sampling duration of 15 continuous seconds. The system has a resolution of 10
t~volts in the dynamic range of _+ 5 mvolts.
BSPM Processing.
Our system of ECG signal processing has been
described it. The program allows selective averaging of
P-QRS-T complexes at each lead site with deletion of
artifacts and ectopic beats and correction of baseline
drift. Measurements of specific ECG time integrals of
the averaged torso complexes are obtained using time
instants determined from edited Frank X, Y and Z
leads (Fig. 1).
Several time integrals may be evaluated. They are
graphically illustrated in Fig. 1. Briefly, the QRS time
integral is the algebraic sum of all potentials from the
time instant of QRS onset to QRS offset multiplied by
the sampling interval -- this represents the net area
under the curve with PR interval as baseline. The ST-T
time integral is similarly calculated from the time instant of QRS offset to T offset; the QRST time integral
represents the algebraic sum of QRS plus ST-T time integrals. The Q-zone time integral is calculated from the
QRS onset to the mid-point of the QRS complex and
the ST-segment from the first 3/8 of the ST-T segment.
All time integrals are expressed as microvolt.seconds
(gV. sec). In this study of patients with myocardial infarction, the analysis focused on Q-zone and St-segment
time integrals.
BSPM Display.
The isointegral map display format utilized for all
time-integral surface distributions is illustrated in
Fig. 2. In individual and group-mean maps, the solid
lines represent positive values; the interrupted lines,
negative values. In temporal subtraction or area-ofdifference maps, the solid lines represent areas in
which the integral values were greater at follow-up
than initial values; interrupted lines designate areas in
which integral values were less at follow-up t h a n
values initially. In all maps, the contour lines extend in
a logarithmic progression from the zone of near-zero
integral values. Each decade line, as well as the maximum and minimum, are numerically identified.
Statistical Analysis.
All grouped data are expressed as mean _+ 1 S.D.
For both normal and patient groups, the null
hypotheses that there were no differences in average
Q-zone and ST-segment time-integral values within
temporal areas-of-difference were tested by t-tests for
J. ELECTROCARDIOLOGY 17 (4), 1984
r'"
.!
321
]
;
9
..
-10
/t
i
f
Fig. 2. Isointegral map display format. The rectangle
represents the torso with the left half indicating the
front, and the right half, the back. The locations of all
torso electrode sites are indicated by the small dots.
The standard ECG V1 to V6 lead sites are indicated by
the small squares. The isocontour lines connect points
of equal time-integral value. This Q-zone map of a
study subject with acute inferior myocardial infarction
indicates a distribution of negative Q-zone values over
the inferior torso 12.
paired data. The null hypotheses that there were no differences in Q-zone and ST-segment root mean square
temporal variability between normal and patient
groups were also tested by t-tests for unpaired data.
RESULTS
The initial and follow-up group-mean Q-zone
isointegral maps, as well as the mean Q-zone areaof-difference maps, of the control and p a tie n t
groups are illustrated in Fig. 3. B o t h initial and
follow-up maps of the pat i ent group reflect the
p a t t e r n of inferior myocardial infarction with abnormal distributions of negative Q-zone integrals
over the inferior torso and increase of positive
distributions over the superior precordium 12.
A l t hough t he surface distributions of the Q-zone
integrals were similar at the two mapping times,
t here Was, however, some small spatial diminution of the negative inferior torso Q-zone distributions and positive superior distributions. This is
confirmed b y the Q-zone subtraction map {Fig. 3).
The group-mean Q-zone. maps of the control
group showed near-identical initial and follow-up
p a t t e r n s and the relative lack of spatial change
with time was supported by the small areas-ofdifference of the subtraction map (Fig. 3}.
The mean ST-segment isointegral maps of the
control and pat i ent groups are shown in Fig. 4.
The initial pat i ent group map shows an abnormal
d i s t r i b u t i o n of positive S T - s e g m e n t integral
values over the inferior torso and a negative
322
MONTAGUE ET AL
Q-ZONE
INFERIOR INFARCTION
NORMAL
T1
....::" /
"",.
~:
- 13
."=:7""',
'-'~
T2
T2
minus
T1
9
...~
""...... ..
NEGATIVE AREA
15
T1
T2
",
-1
~
POSITIVE AREA
T1
T2
/f
NEGATIVE AREA
34]
T1
POSITIVEAREA
T2
T1
T2
,/,
10
5
t~
0
~-5
iii'
'i~l,
~ "-'18-e
[~
r
(u
-10
0.
-15-
:
~..j.~
_25.]
e
=
-6-12-
p< 0.001
p< 0.05
p< 0.02
p< 0.005
Fig. 3. Comparison of mean initial (T1), follow-up (T2) and difference (T2 - T1) Q-zone maps from 42 normal subjects
and 32 patients with inferior myocardial infarction. Individual average Q-zone values within the temporal areas-ofdifference at initial and follow-up study are illustrated below the respective difference maps. Normal subjects had
small spatial changes with time. In contrast, inferior infarction patients had spatially greater temporal change with
an increase of Q-zone integral values inferiorly and a decrease anteriorly and superiorly. Quantitatively, both groups
had similar degrees of change within their respective areas-of-difference.
J. ELECTROCARDIOLOGY17 (4), 1984
BSPM PATTERNSIN M]
323
ST
INFERIOR INFARCTION
NORMAL
;',
".-....7~..:
....":
"-.
~
.......... "-.:--.-:~ -..._::--- ...... .-
~
T1
...................
-~
. ..........................
~~ :::::::::::::::::::::::::::::::::::::::::
'
....
--.-,o........ )( . . /
iiiiiiiiiiiiiiiiiioiiiiii'
T2
T2
minus
T1
~..1.:
~,f:_!~?..!!!!!:~;,i
... ~;;.........-,,..............
50-
NEGATIVE AREA
NEGATIVE AREA
T1
T2
15"
T1
T2
POSITIVEAREA
T1
T2
10"
4030-
Q,3
~
0
(fr
,rTr
nil
')~);~)
-(~
K
,
2010
0
--15
p = N.S.
p< 0.001
p< 0.001
Fig. 4. Comparison of mean initial (T1), follow-up (T2) and difference (T2 - T1) ST-segment maps of 42 normal subjeers and 32 patients with acute inferior myocardial infarction. Individual-average ST-segment integral values
within the temporal areas-of-difference are illustrated below the respective difference maps. Normal subjects had no
significant spatial or quantitative change in ST-segment distributions with time. In contrast, there were marked
spatial and quantitative changes with near-normalization of initially abnormal ST-segment distributions with time
in the inferior infarction group.
J, ELECTROCARDIOLOGY17 (4), 1984
324
MONTAGUE ET AL
distribution over the superior torso, compatible
with acute inferior myocardial infarction 12. At
follow-up, however, the patient group had a nearnormal distribution of ST-segment integrals
(Fig. 4). In comparison to the small spatial
changes observed in Q-zone distributions over the
average eight month follow-up interval (Fig: 3),
these spatial changes in ST-segment distributions were marked. Comparison of the ST
distributions of the normal group reveals no apparent change between initial and follow-up patterns and almost no area-of-difference (Fig. 4).
Quantitative comparisons of individual subjects' average integral values within the Q-zone
and ST-segment areas~f-difference are illustrated
by the data columns below the respective groupmean subtraction maps (Figs. 3, 4). The normal
group d e m o n s t r a t e d significant changes in
average integral values within the small Q-zone
areas-of-difference but no change within the very
small ST area-of-difference. In contrast, the infarction p a t i e n t s d e m o n s t r a t e d statistically
significant changes in all Q-zone and ST areas-ofdifference. The relative degree of quantitative
change within the patient groups was, however,
greater for ST-segment values.
The root mean square intermap variability of
Q-zone and ST time integrals for both groups is
compared in Fig. 5. The temporal change, as
assessed by this global measure of the change in
torso distribution, was significantly greater in in-
INDEX OF VARIABILITY
ST
Q - ZONE
INFERIOR
NORMAL INFARCTION
( n = 42) ( n = 32)
111.
81
18-
NORMAL
1n=42)
9
9
16-
9
10 ~
9149
9
6
,~
t,
9
0
6"
4
(co
'i|
((t
(((0
~
(r
r
(0
o-
INFERIOR
INFARCTION
(n=32)
p< 0.001
4-
9
2-
';
c((C(o
(((((($
Ill
9
9
oJ~)
1- 9
({((((((~ 9
(((r(( 9
p< 0.001
Fig. 5. Comparison of Q-zone and ST-segment root,
mean square temporal variability from initial iT1) to
follow-up (T2) maps in both study groups.
farction patients for both Q-zone and ST-segment
values.
Individual Q-zone and ST-segment is 9
maps of two representative patients and a normal
control subject are compared in Figs. 6 and 7,
respectively. One (patient A, Fig. 6) had relatively little spatial or quantitative regression of
Q-zone infarction distributions. Another {patient
B, Fig. 6) showed considerable change, with gain
in Q-zone values inferiorly and regression or
reduction of values superiorly. Thus, as might be
expected, t h e r e is c o n s i d e r a b l e i n d i v i d u a l
variability of post-infarction Q-zone evolutionary
changes. In a similar fashion, the ST-segment
patterns of the same patients also demonstrate
individual variability in the evolution of this
repolarization integral (Fig. 7). Patient A, who
had relatively little change of Q-zone distributions (Fig. 6), had near-normalization of ST patterns (Fig. 7). Patient B, on the other hand, who
did demonstrate some temporal normalization of
initially abnormal Q-zone patterns (Fig. 6), had
initially abnormal ST distributions which evolved
to an equally abnormal distribution at follow-up
(Fig. 7).
DISCUSSION
The results of this study demonstrate that during the post-infarction period in patients with inferior myocardial infarction there was a relative
temporal gain in Q-zone integral values over the
inferior torso and diminution of values over the
superior precordium. The data also showed a temporal gain in ST-segment integral values over the
precordium and a concomitant reduction of STsegment values over the inferior torso. Thus, similar to the reciprocal torso distributions of Q-zone
and ST-segment during acute infarction le, there
was a reciprocity of the torso distributions of the
post-infarction evolutionary changes of these
depolarization and repolarization time integrals.
Although there was a similar directional
change in post-infarction Q-zone and ST-segment
patterns in that both integrals showed regression
from initially a b n o r m a l d i s t r i b u t i o n s , t h e
magnitude of change of the respective integrals
was not similar. Rather, the changes in the
repolarization (ST-segment) time integral patterns were spatially and quantitatively much
greater than those of the depolarization time integral (Q-zone). At the time of follow-up examination, the ST-segment maps had returned to a
near-normal distribution while Q-zone maps continued to show a pattern of inferior infarction.
J. ELECTROCARDIOLOGY 17 (4), 1984
BSPM
PATTERNS
IN M I
325
regional wall motion abnormalities with time
may be slight. Nonetheless, some measurable
diminution of abnormal Q-zone distributions was
apparent during the post-infarction period. It remains to be determined whether the speed or extent of changes in the Q-zone (and ST-segment)
integral patterns during the early post-infarction
period parallels, or is predictive of, the changes in
segmental or global myocardial wall function or
late clinical course. Likewise, the exact temporal
pattern of the BSPM pattern changes, and the influence of the size and site of the original myocardial infarction on these changes, are as yet
unknown. Further studies focusing on the recovery processes following myocardial infarction are
required to better define the natural history of
this part of the infarction process and to explore
the nature of the apparent protective effect(s) provided by interventions, such as beta-blockade,
during this phase ]6.
One explanation for the marked changes in STsegment patterns is that a large part of the initially ischemic myocardial tissue improved its
blood supply during the post-infarction period
with consequent partial normalization of repolarization activity. If this theory is correct, then its
corollary is probably also true: that is, because of
the relatively small changes in Q-zone distributions with time, the initially abnormal Q-zone patterns probably reflected largely irreversible
myocardial damage. In this study, correlations of
Q-zone and ST-segment pattern changes with
changes in myocardial wall motion abnormalities
and ventricular function were not attempted.
However, a recent study has confirmed that the
presence of abnormal Q-waves is highly correlated with the presence of underlying areas of
myocardial dysfunctionlL Thus, the small change
in abnormal Q-zone distributions observed in this
study suggests that the degree of improvement in
Q-ZONE
NORMAL
PATIENT A
,.--..
i., .............
-" ..
.. \
,.:........
:::'.'-::::.'-"-:.-,
, _...;--~~.
T1
".--i.."
X
"-'.'.
:,
,~--,
"- ",
...-;;.."
.."
Z
i
.-
.....
l
,.":;"
.." .-'"
"" - . . . .
.....
.. .......
.........
:"---t,
--.;.::.. ,; ..~!.7
~ , ~ 7 ~ "
_._7,_~
-15
"-
"" iii!-l
',, ",.~
li
.."
~;-"
PATIENT B
!
V'-~L:-. - .......... - - - ' - .
-
.
..-,:-
,,~-~-.~..~
/i
/t/---.~'~.
:"
..'..: .."
..... ...
I
-
.........
:._~ ....
.
."
.:
..
t
"/.- .......
,
"
i
/
-
7"7]
"
!:w. -.-8C.;,
"2'
":.........
-:"""
-,. ,..,- .......
,,'A,
.
: .;- ......
.'~c-~.:.~
-",
.--
7
~:
I
-.
'.,
T2
t ::
"-,
L/"- ..... 7:7
T2
minus
T1
R
!"/-a
"-,
.; :"
;"
7?,,'
........ J);)
,," :- ,.'
....
-7
/---,.
-i
..:"- '.!. _~:"",..
:
**
"-..,"
., ....
Fig. 6. Comparison of initial (T1), follow-up(T2) and difference (T2 - T1) Q-zone maps of a normal subject and two
representative inferior infarction patients. Patient A had very little change in inferior torso distributions; patient B,
in contrast, had marked gain in Q-zonevalues inferiorly. There was no obvious spatial change in Q-zonedistributions
in the normal subject, and the quantitative differences identified by the difference maps were small and irregularly
distributed.
J. E L E C T R O C A R D I O L O G Y
17 (4), 1 9 8 4
326
M O N T A G U E ET AL
minimum value), and a small area-of-difference,
group size may have to be relatively large to
avoid a n alpha error. Individual subtraction
maps, with each subject as his own control, may,
therefore, be more appropriate for quantitation of
temporal change, particularly for correlation with
changes in other parameters of myocardial function. Until more experience is gained in the
utilization of such techniques, we feel that caution in interpretation of results is necessary.
Specifically, the size and torso location of surface
areas-of-difference and their spatial pattern must
all be carefully assessed in terms of biological
meaning and appropriateness -- it is not enough
to simply apply statistical tests to integral values
within group-mean areas-of-difference.
In summary, BSPM techniques allow location
and quantitation of changes in depolarization and
repolarization time-integral patterns during the
period following acute myocardial infarction and
The area-of-difference or subtraction BSPM
methodology, which was used to assess spatial
and quantitative temporal ECG changes in this
study, is a very sensitive indicator of temporal
electrocardiographic differences. The spatial
areas-of-difference identified by subtraction of
group-mean maps provided, in fact, the major
areas of interest for study of temporal change.
Quantitation of integral values within the groupmean areas-of-difference for each individual
allowed quantitative comparisons within groups
over the time of this study. Since t h e area of interest is determined from the mean difference,
this method may not, however, be the ideal
method to reflect individual differences that
deviate widely from the mean. In addition, why
normal subjects should display temporal areas-ofdifference for Q-zone distributions is uncertain. It
may be that when dealing with a time integral
with a small data range {maximum minus
ST
".,
"l~"
:" !
"'.......J.+J
..... :::;.
"..:
PATIENT B
PATIENT A
NORMAL
..*"
,'"
"L
:-',o,-.....,-"'--.,
-il
. . ...........
..
""
,,,,..,-:
.:..- ....-.!
i
~"----.'..........
........ :::~".. ':,
T1
/"
=======================
..""~:i ""-:~.......... '
"'k ":" .. :;.,
"".i!i............(i;
I~
"
.::::::
.......
- . . . . -__.._ . . . . . . . . . . .
T2
..., ):::::::::.
T2
minus
T1
:
!ii
Fig. 7. Comparison of initial (T1), follow-up {T2) and difference (T2 - T1) ST-segment maps in the same study subjects as illustrated in Fig. 6. Patient A had near-normalization of ST-segment distributions with time; patient B, in
contrast, although having marked change in his ST-segment distributions with time, continued to have an abnormal
torso distribution at T2. No temporal changes in spatial ST-segment distributions of the normal subject were evident although the difference map showed small and irregularly distributed quantitative changes.
J. ELECTROCARDIOLOGY 17 (4), 1984
BSPM PATTERNS IN MI
offer promise as one means of assessing recovery
from myocardial infarction. Additional, m u l t i p l e
p a r a m e t e r q u a n t i t a t i v e and correlative s t u d i e s
are required to b e t t e r delineate the n a t u r a l
history of this i m p o r t a n t phase following acute
myocardial infarction and injury.
9.
Acknowledgement: We are grateful to Mary Jayne
Tilley and Gaye Strong for expert secretarial
assistance.
I0.
REFERENCES
1. ALPERT, J S AND BRAUNWALD,E: Pathological and
Clinical Manifestations of Acute Myocardial Infarction. In: Heart Disease, E BRAUNWALD,Ed.
Saunders, Philadelphia, 1980, pp 231-246, 1333
2. MAROKO,P R, Lmnv, P, COVELL,J W, SOBEL,B E,
Ross, J AND BRAUNWALD, E: Precordial STsegment elevation mapping: An atraumatic
method for assessing alterations in the extent of
myocardial ischemic injury. Am J Cardiol 29:223,
1972
3. FLAIIERTY,J T, REID, P R, KELLY,D T, TAYLOR,D
R , WEISFELDT, M L AND PITT, B: Intravenous
nitroglycerin in acute myocardial infarction. Circulation 51:132, 1975
4. MULLER,J E, MAROKO,P R AND BRAUNWALD,E:
Evaluation of precordial electrocardiographic mapping as a means of assessing changes in myocardial ischemic injury. Circulation 52:16, 1975
5. MARORO,P R, DAVIDSON,D M, LIBBY, P, HAGAN,
A D AND BRADNWALD,E: Effects of hyaluronidase
administration on myocardial ischemic injury in
acute infarction: A preliminary study in 24 patients. Ann Intern Med 82:516, 1975
6. GOLD, H K, LEINBACH, R C AND MAROKO, P R:
Propranolol-induced reduction of signs of ischemic
injury during acute myocardial infarction. A m J
Cardiol 38:689, 1976
7. MADIAS, J E, MADIAS, N E AND HOOD, W B:
Precordial ST-segment mapping II. Effects of oxygen inhalation on ischemic injury in patients with
acute myocardial infarction. Circulation 53:411,
1976
8. HILLIS, L D, ASKENAZI, J, BRAUNWALD, E, RADVANY, P, MULLER, J E, FISHREIN, M C AND
MAROKO, P R: Use of changes in the epicardial
Q R S complex to assess interventions which
J. ELECTROCARDIOLOGY 17 (4), 1984
11.
12.
13.
14.
15.
327
modify the extent of myocardial necrosis following
coronary artery occlusion. Circulation 54:591, 1976
MAROKO,P R, HILLIS,L D, MULLER,J E, TAVAZZI,
L, HEYNDRICKX, G R, RAY, M, CIIIARIELLO, M,
DISTANTE, A, ASKENZAI, J, SALERNO, J, CARPENTIER, J, RESHETNAYA, N I, RADVANY, P, LIBBY, P,
RAABE, D S, CHAZOV, E I, BOBBA, P AND BRAUNWALD, E: Favorable effects of hyaluronidase on
electrocardiograhicevidence of necrosis in patients
with acute myocardial infarction. N Engl J M e d
296:898, 1977
HENNING, H, HARDARSON, T, FRANCIS, G,
O'RoURKE, R A, RYAN, W ANn ROSS, J: Approach
to the estimation of myocardial infarct size by
analysis of ST-segment and R wave maps. A m J
Cardiol 41:1, 1978
PALMERI, S T, HARRISON, D G, CoaB, F R, MORRIS,
K G, HARRELL,F E, IDEKER,R E, SELVESTER, R H
AND WAGNER, G S: A QRS scoring system for
assessing left ventricular function after myocardial infarction. N Engl J Med 306:4, 1982
MONTAGDE, T J, SMITI{, E R, SPENCER, C A,
JOtlNSTONE, D E, LALONDE,L D, BESSOUDO, R M,
GARDNER, M J, ANDERSON R N AND HORACEK, B
M: Body surface electrocardiograhic mapping in
inferior myocardial infarction: Manifestations of
left and right ventricularinvolvement. Circulation
67:665, 1983
BERMAN, D S, AMSTERDAM, E A, HINES, H H,
SALEL, A F, BAILEY, G J, DENARDO, G L AND
MASON,D T: A new approach to the interpretation
of technetium 99m pyrophosphate scintigraphy in
the detection of acute myocardial infarction.
Clinical assessment of diagnostic accuracy. Am J
Cardiol 39:341, 1977
MONTAGUE, T J, SMITII, E R, CAMERON, D A,
RAUTAIIARJU, P M, KLASSEN, G A, FLEMINGTON, C
S AND HORACEK, B M: Isointegralanalysis of body
surface maps: Surface distribution and temporal
variabilityin normal subjects:Circulation63:1166,
1981
SUTIIERLAND, D J, McPIIERSON, D D, SPENCER, C
A, ARMSTRONG, C S, HORACEK, B M AND MONTAGUE, T J: Effects of posture and respiration on
the body surface electrocardiogram. A m J Cardiol
52:595, 1983
16. The Norwegian Multicenter Study Group: Timololinduced reduction in mortality and reinfarctionin
patients surviving acute myocardial infarction.N
Eng J M e d 304:801, 1981