Natural History of S-T Segment Elevation After

Acute Myocardial Infarction

ROGER M. MILLS, Jr., MD Clinical, electrocardiographic and cineventricuiographic data in two

ELIOT YOUNG, MD patient groups were analyzed to define the natural history of S-T seg-
RICHARD GORLIN, MD, FACC’ ment elevation after myocardial infarction. in sixteen of 22 patients (73
MICHAEL LESCH, MD, FACC+
percent) with acute inferior myocardial infarction, S-T segment eieva-
Boston, Massachusetts tion was present on hospital admission, persisting in 1 (5 percent) by
the 2nd week. S-T segment elevation was present on admission in 18
of 23 patients (78 percent) with acute anterior myocardial infarction
and persisted In 13 after 1 week and in 9 of 14 (64 percent) during a
follow-up period of 1 to 6 months. S-T segment elevation lasting more
than 2 weeks after myocardial infarction did not resolve. Compared
with patients with inferior myocardial infarction or anterior infarction
without persistent S-T segment elevation, patients with anterior infarc-
tion and persistent S-T segment elevation had a higher level of mean
maximal serum creatine phosphokinase (CPK), more severe left ven-
tricular decompensation and a greater frequency of death in the early
follow-up period. in a separate series of 95 patients with cineangio-
graphically documented coronary artery disease, 40 of 65 patients (62
percent) with advanced anterior and apical asynergy had persistent
S-T segment elevation. By contrast, only 1 of 30 (3 percent) with coro-
nary disease and normal ventricuiograms had persistent S-T segment
elevation.
We concluded that (1) the natural history of S-T segment elevation
after myocardial infarction is resolution within 2 weeks in 95 percent of
inferior but in only 40 percent of anterior infarctions; (2) S-T segment
elevation persisting more than 2 weeks after myocardial infarction
does not resolve; (3) persistent S-T segment elevation is associated
with clinically more severe myocardial infarctton; and (4) in patients
with coronary artery disease, persistent S-T segment elevatlon after
myocardiai infarction is a specific but insensitive index of advanced
asynergy.

An empirical association between persistent electrocardiographic S-T

From the Cardiovascular Division, Depart-
ment of Medicine, Harvard Medical School and
the Peter Bent Brigham Hospital, Boston, Mass.
This study was supported by U. S. Public Health
Service Grants TO1 HL-05679 and 2 PO1 HL-
11306. Manuscript accepted October 9, 1974.
‘Present address: Department of Medicine,
Mount Sinai School of Medicine of the City Uni-
versity of New York, 5th Ave. and 100th St.,
New York, N. Y. 10029.
T Established Investigator of the American
Heart Association.
Address for reprints: Michael Lesch, MD,
Peter Bent Brigham Hospital, 721 Huntington
Ave., Boston, Mass. 02115.
segment elevation after myocardial infarction and presence of ven-
tricular aneurysm has been recognized for more than 25 years.1-4 The
introduction of coronary care unit monitoring of serial electrocardio-
graphic records and of left cineventriculography as a diagnostic pro-
cedure for evaluating contraction disorders has engendered problems
of definition regarding the relation between “persistent” S-T seg-
ment elevation and “ventricular aneurysm.” These include (1)defin-
ing the natural history of S-T segment changes after myocardial in-
farction, particularly in light of the current widespread interest in S-
T segment mapping to assess the effect of therapeutic interventions
on infarct size; (2) defining that point in time when the chronic S-T
elevation of “ventricular aneurysm” may be differentiated from acute
S-T elevation of injury; (3) defining the clinical characteristics of
myocardial infarction that result in persistent elevation of the S-T

May 1975 The American Journal of CARDIOLOGY Volume 35 669

S-T ELEVATION AFTER MYOCARDIAL INFARCTION-MILLS ET AL

INFERIOR MYOCARDIAL INFARCTION ANlERlOR MYOCARDIAL INFARCTION

0 ,o,al t,me 0, ,oiiox-up
mpI Per,od 01S-l lle”af,an

1234561 2 3 I-6
1234561234 5 1-b
DClp Weeks+Months
1 Days : Weekr---_jMonths
Time Post MI TIME POST MI
FIGURE 1. Duration of S-T elevation for the study population (left, inferior infarction; right, anterior infarction) is plotted as a function of time with
each bar representing the data from an individual patient. Hatched areas correspond to the period of S-T segment elevation.

segment; and (4) defining the sensitivity and specific-
ity of S-T elevation as an index of cineangiographic
abnormalities of wall motion, as opposed to anatomic
aneurysm at autopsy.
In this report we have attempted to establish ap-
propriate definitions based on retrospective evalua-
tion of clinical, electrocardiographic and cineangio-
graphic data.
Methods
To investigate the natural history of S-T segment eleva-
tion after myocardial infarction, we used the record of ad-
missions to the coronary care unit of the Peter Bent Brig-
ham Hospital from January 1971 through December 1972
and reviewed 45 consecutive cases that met the following
criteria: documentation of infarction by characteristic
changes in both serial serum enzyme (creatine phosphoki-
nase (CPK] and glutamic oxaloacetic transaminase
[SGOT]) determinations and electrocardiograms (Q wave
or chronic T wave changes of nontransmural infarction),
absence of intraventricular conduction defects, survival at
least 3 weeks after admission and absence of severe noncar-
disc disease such as chronic renal failure or neoplasm. S-T
segment changes as such were not used as a criterion for in-
clusion in the study, nor was there a correction for the in-
terval from onset of symptoms to time of admission. As a
result, some patients with definite infarction had no re-
corded period of S-T segment elevation. Since the natural
history of this subset was unknown, these patients were in-
cluded in the study. No patient had clinical evidence of re-
current infarction during the period of S-T segment analy-
sis. Follow-up information was obtained from the patient’s
hospital record and the attending physician or by tele-
phone contact with the patient or his family.
Myocardial infarction was classified as anterior if the
primary QRS changes and S-T elevation were present in
leads Vz to Vd and inferior if the changes were in leads II,
III or aVF. S-T elevation was defined as greater than 1 mm
elevation of the mid-portion of the S-T segment (regardless
of contour) above the base line defined by successive T-P
segments in any of leads Vz to V4 with anterior infarction
or similar elevation in any of leads II, III or aVF with infe-
rior infarction. The lateral leads Vg to Vs were excluded in
classification because of their variable involvement in both
inferior and anterior infarction. Standard practice in the
coronary care unit includes marking the position of chest
leads to obtain a series of comparable tracings. In the fol-
low-up period, electrocardiograms were assumed not to
have changed in any time period spanned by identical trac-
ings; similarly, the electrocardiogram was presumed to be
unchanged until a change was documented.
To assess whether persistence versus resolution of S-T
elevation represented (1) differential resolution of initially
similar elevations, or (2) similar resolution of initially dif-
ferent elevations, the resolution rates of S-T elevation in
inferior infarction and anterior infarction with and without
persistent S-T elevation were compared using an S-T index
computed as:
S-T index =
Z (sum of S-T elevation in mm in all involved leads)
Number of patients with S-T elevation
The term X3-T for a given person has been used by Mar-
oko et a1.5 and Kantkiewski et a1.6 The expression given
here represents the arithmetic average of S-T elevation for
a group of patients.
Left cineventriculograms of high quality were obtained
and analyzed using standard techniques previously de-
scribed.7 To evaluate the sensitiuity of S-T elevation as an

610 May 1975 The American Journal of CARDIOLOGY Volume 35

 S-T ELEVATION AFTER MYOCARDIAL INFARCTION-MILLS ET. AL.

TABLE I TABLE II

Persistent S-T Elevation and Location of Myocardial Infarction S-T Elevation at 2 Weeks and 1 to 6 Months After Infarction
in 30 Patients
S-T Elevation
S-T Segment Elevation
On Admission At 1 to 6 Months
Site of Infarction (no.) (no.) At 1 to 6 Months

At 2 Weeks Resolved Present

Inferior 16 of 22 1 of 16
Anterior 18 of 23 9 of 14
Resolved 20 0
Difference Not significant P <O.Ol
Present 0 10

index of advanced left ventricular asynergy by cineven-
triculography, the immediate precatheterization electro-
cardiograms of 65 patients with abnormalities of left ven-
tricular motion defined as severe hypokinesis, akinesis or
dyskinesis’ of the anteroapical left ventricular wall due to
coronary artery disease were examined for S-T elevation
using the criteria previously outlined. To evaluate the spec-
ificity of the association between S-T elevation and asyner-
gy, the electrocardiograms of 30 patients with angiographi-
tally proved coronary artery disease and normal left cine-
ventriculograms were reviewed for S-T elevation.
Results
Natural history of S-T elevation: Sixteen of 22
patients (73 percent) with inferior infarction had S-T
elevation on admission to the hospital. By the 2nd
week of hospitalization, S-T elevation persisted in
only one (5 percent); this was unchanged at 6
months. Eighteen of 23 patients (78 percent) with an-
terior infarction had on admission S-T elevation that
persisted in 13 of 23 after 1 week and in 9 of 14 (64
percent) during a follow-up period of 1 to 6 months.
There was no significant difference in the prevalence
of S-T elevation on admission in patients with inferi-
or or anterior infarction, but the difference at 1 to 6
months of follow-up was highly significant (P <O.Ol).
S-T segment elevation persisting more than 2 weeks
after acute infarction did not return to base-line lev-
els (Fig. 1, Tables I and II).
The plot of S-T index versus time demonstrated
that the initial resolution rate of S-T elevation (the
slope of the line obtained by the method of least
squares) was similar in all groups. However, the pa-
tients with anterior infarction and persistent S-T ele-
vation had an initial elevation index more than dou-
ble that of the groups whose S-T segment returned to
base-line levels (Fig. 2).
Clinical characteristics of patients with per-
sistent S-T elevation: The 45 patients with acute
myocardial infarction were classified in three groups:
(1) inferior infarction, (2) anterior infarction without
persistent S-T elevation, and (3) anterior infarction
with persistent S-T elevation (2 2 weeks). These
groups were compared with regard to mean maximal
serum CPK, clinical evidence of left ventricular de-
compensation and mortality in a short-term follow-
up period (25 to 27 months). Mean maximal serum
CPK was essentially equal for patients with either in-
ferior infarction or anterior infarction without persis-

16 X= AMI with Persistent S-T Elevation

t
0; AMI without Persistent S-T Elevation
t
l = IMI

2-
FIGURE 2. Index of S-T elevation (B-T) in patients
with inferior infarction (IMI) and anterior infarction (AMI)
with or without S-T elevation (STE) at 2 weeks, plotted 0
as a function of time for the 1st week of hospitaliza- 1 2 3 4 5 6-7
tion. TIME jdayri

May 1975 The American Journal of CARDIOLOGY Volume 35 611

S-T ELEVATION AFTER MYOCARDIAL INFARCTION-MILLS ET AL.

TABLE III disease and advanced degrees of left ventricular ante-

Clinical Characteristics of Myocardial Infarctions Resulting rior and apical asynergy, 40 (62 percent) had persis-
in Persistent S-T Elevation tent S-T elevation, as defined earlier. The adminis-
tration of digitalis was not related to the presence or
Anterior Myocardial absence of S-T elevation. Of the 30 patients with cor-
infarction onary disease and normal left ventriculograms, only 1
Inferior S-T Elevation S-T Elevation
(3 percent) had persistent S-T elevation (Table IV).
Myocardial Resolved Present
Infarction at 2 Weeks at 2 Weeks Discussion
Subjects (no.) 10 13 Our data demonstrate that S-T elevation per-
Male (no., %) :: (64%) 8 (80%) 9 (69%) sisting 2 or more weeks after acute myocardial infarc-
Mean maximal serum 387 249 527 tion does not resolve in a follow-up period of up to 6
CPK (lU/ml) months and cannot be associated clinically with re-
Left ventricular decom- peated or ongoing ischemic injury. Thus, the electro-
pensation (no., %)
cardiographic diagnosis of persistent S-T segment el-
Mild 6 (27%) 100%) 5 (39%) evation consistent with ventricular aneurysm forma-
Severe 0 0 6 (46%)
Average duration of
tion may be made as early as 2 weeks after acute
follow-up (mo) 26 25 27 myocardial infarction.
Status at follow-up (no.) A review of major textbooks of general cardi-
Dead 1 0 3 ologys10 yields no precise definition of “persistent”
Alive 17 8 9 S-T elevation. Schamroth’l requires a delay of 6
Not known 4 2 1 months after myocardial infarction until “aneurysm”
may be diagnosed on the basis of the electrocardio-
gram. The origin of the 6 month dictum is unclear.
TABLE IV
Parkinson and Bedford,i2 describing the evolution of
S-T Elevation and Cineventriculographic Findings electrocardiographic changes after acute myocardial
infarction in the three standard leads, observed that
S-T Elevation S-T Elevation
S-T segment elevation usually resolved within 2
Present Absent
weeks. Neither Steven3 nor Ford and Levine4 had ac-
Group no. % no. % cess to serial electrocardiograms from coronary care
units in their studies associating persistent S-T ele-
65 patients with advanced 40 62 25 38 vation with anatomic evidence of ventricular aneu-
anteroapical asynergy
rysm, and neither of their reports attempted a pre-
With digitalis 31 59 21 41
cise definition of “persistent” S-T elevation. Steven3
Without digitalis 9 69 4 31
1
included cases in which autopsy was performed 1, 2,
30 patients with coronary 3 29 97
artery disease and normal
5, 6 and 8 weeks after acute infarction. He stated
cineventriculograms with regard to his case 8, “While 6 days may be too
short a time, usually the S-T changes due to a current
of injury will have disappeared in this time.”
tent S-T elevation. The mean maximal CPK eleva- Prognostic significance of persistent S-T ele-
tion in the group with anterior infarction and persis- vation: Patients whose electrocardiograms showed
tent S-T elevation was twice that recorded for other persistent S-T elevation had clinically more severe
groups (Table III). This difference was not statisti- myocardial infarction, as judged by greater CPK ele-
cally significant because of the wide variations within vations,13 a greater incidence of left ventricular de-
each group. compensation and an increased mortality due to con-
Mild left ventricular decompensation, defined as gestive heart failure in the early follow-up period.
rales and diastolic filling sounds that responded With regard to the striking difference in the “S-T
promptly to digitalis or diuretic therapy, or both, was index” between patients with and without persistent
seen in all three groups. Severe left ventricular fail- S-T elevation, our findings are similar to those of
ure with pulmonary edema was limited to 6 of 13 pa- Nielsen,14 who demonstrated that marked elevation
tients (46 percent) with anterior infarction and per- of the S-T segment immediately after acute myocar-
sistent S-T elevation (Table III). dial infarction predicted a poorer prognosis. Maroko
Only four deaths were recorded in the 38 patients et a1.5 found S-T segment elevation to be an index of
whose follow-up data were available. Of these, three the extent of myocardial infarction. Their laboratory
occurred in the group with anterior infarction and findings lend further support to the thesis that the
persistent S-T elevation; two of the three deaths were patients with greater elevations of both serum CPK
due to refractory congestive heart failure (Table III). and electrocardiographic S-T segments have SUS-
Sensitivity and specificity of S-T elevation as tained more extensive myocardial infarction. As stat-
an index of disordered left ventricular contrac- ed earlier, the arbitrary calculation of an “S-T index”
tion: In the series of 65 patients with coronary artery for the three groups of patients (inferior infarction,

612 May 1975 The American Journal of CARDIOLOGY Volume 35

 B-T ELEVATION AFTER MYOCARDIAL INFARCTION-MILLS ET AL.

anterior infarction with and without S-T elevation) TABLE V

was performed in an attempt to differentiate between Reported Cases of Persistent S-T Elevation and
two possible modes of development of persistent S-T Postinfarction Ventricular Aneurysm
elevation: (1) differential resolution of initially simi-
lar S-T elevations; or (2) similar resolution of initially Patients with
different S-T elevations. The observation that persis- Persistent S-T
tent S-T elevation reflects an increase in the magni- Diagnosis of
Elevation
tude of initial S-T elevation rather than a decrease in Reference Date Aneurysm no. %
the initial rate of resolution remains extremely inter-
esting and unexplained. Dubnow et al.ls 1965 At autopsy 30 of 41 73
Throughout the study, the magnitude of S-T ele- Gorlin et al.16 1967 By cineventric- 12 of 24* 50
vation in limb and chest leads was compared despite ulography
accepted geometric differences in the lead systems. Mourdjinis et al.17 1968 At autopsy (13); 25 of 39 64
This procedure seemed justified on clinical grounds clinically (26)
Total 67 of 104 64
since these systems are the only ones available to the
practicing physician. It is impossible to assess what
* Resting electrocardiogram.
differences the lead systems themselves may have
contributed to the difference in prevalence of persis-
tent S-T elevation in inferior and anterior infarction.
One might argue that if semidirect leads could be re- tion suggests that the development of persistent S-T
corded from the diaphragmatic surface of the heart elevation after myocardial infarction is quite specific
after inferior infarction, persistent S-T elevation for an advanced degree of asynergy.
might be much more prevalent. However, the clear Incidence of ventricular aneurysm: The ques-
association of persistent S-T elevation with other evi- tion of the incidence of ventricular aneurysm after
dence of clinically more extensive myocardial infarc- infarction remains unanswered. In fact, one might
tion, and the hemodynamically benign course of most hypothesize that no ventricle that has sustained an
patients with inferior infarction or anterior infarction infarction moves entirely normally. Therefore, the in-
without persistent S-T elevation, favors our empiri- cidence of wall motion abnormality in any given se-
cal approach. ries of patients studied after infarction will reflect (1)
Persistent S-T elevation and left ventricular the sensitivity of the criteria for diagnosis of infarc-
asynergy: Persistent S-T elevation, although quite tion, (2) the method of analysis of left ventricular
specific, was only moderately sensitive as an indica- motion, and (3) the relative number of patients with
tor of advanced cineventriculographic abnormalities anterior versus inferior infarction in the series.
of left ventricular wall motion. The prevalence of Corollaries to this hypothesis are the concepts that
persistent S-T elevation in various series of patients ventricular aneurysm is not a discrete entity but
with ventricular aneurysm is affected by the tech- rather an advanced stage in the spectrum of left ven-
nique and criteria-used for diagnosis of ventricular tricular motion abnormalities due to coronary artery
aneurysm and by the number of leads in the electro- disease and that persistent S-T elevation when pres-
cardiographic records. Data from three recent stud- ent is a clinically useful indicator of this stage.
ies15-17 using postmortem, clinical and cineangio- Definitions: With regard to the problems of defi-
graphic techniques of diagnosis with standard 12 lead nition alluded to, the following statements are ad-
electrocardiograms are summarized in Table V. As vanced:
one might expect, the incidence of persistent S-T ele- 1. The natural history of S-T elevation after acute
vation was highest in the autopsy series, in which myocardial infarction is resolution within 2 weeks in
definite external protrusion of scar (fibrotic) tissue 95 percent of inferior infarctions but in only 40 per-
was required for diagnosis,15 and lowest in the angio- cent of anterior infarctions.
graphic series, in which highly variable proportions of 2. S-T elevation persisting more than 2 weeks
scar and muscle were demonstrated in the areas of after acute myocardial infarction does not resolve in
ventricular asynergy.16 However, the overall figure of the follow-up period and may therefore be considered
64 percent of patients with ventricular aneurysm chronic.
showing persistent S-T elevation computed by com- 3. Myocardial infarctions resulting in persistent
bining the data of these three studies (Table V) is re- S-T elevation were predominantly anterior and were
markably close to the 62 percent incidence in our se- characterized by greater S-T elevation, higher levels
ries. of mean maximal serum CPK, more severe left ven-
We are not aware of previously reported data de- tricular decompensation and higher short-term mor-
scribing the prevalence of S-T elevation in patients tality rates than those of infarctions in which acute
with coronary disease and normal left ventriculo- S-T elevations returned to base-line levels.
grams. Although not a proof of specificity in rigorous 4. In patients with coronary artery disease, persis-
terms, the finding that only 1 of 30 patients (3 per- tent S-T elevation is a specific, but relatively insensi-
cent) with angiographically documented coronary tive, index of advanced degrees of left ventricular
disease and normal ventriculograms had S-T eleva- asynergy.

May 1975 The American Journal of CARDIOLOGY Volume 35 613

S-T ELEVATION AFTER MYOCARDIAL INFARCTION-MILLS ET AL.

References
1. Scherf D, Boyd TJ: Cardiac aneurysm. Med Clin N Amer 26: Schlant RC, et al., ed). New York, McGraw-Hill, 1974, p 297-
919-927, 1942 313
2. Rosenberg B, Messlnger WJ: The electrocardiogram in ventric- 10. Hirsch JI: The electrocardiogram in ischemlc heart disease. In,
ular aneurysm. Am Heart J 37:267-277, 1949 Cardiac and Vascular Diseases (Conn HL Jr, Horwitz 0, ed).
3. Steven RA: Electrocardiographic findings in cardiac aneurysm. Philadelphia, Lea & Febiger, 1971, p 1064
Ann Intern Med 34:747-756, 1951 11. Schamroth L: An Introduction to Electrocardiography, 4th edi-
4. Ford RV, Levine HD: The electrocardiographic clue to ventricu- tion. London, Blackwell Scientific Publications, 197 1, p 29
lar aneurysm. Ann Intern Med 34:996-l 0 16, 195 1 12. Parklnson J, Bedford DE: Successive changes in the electro-
5. Maroko PR, Libby P, Covell JW, et ai: Precordial S-T segment cardiogram after cardiac infarction. Heart 14:195-212, 1927-
mapping: an atraumatic method for assesslng alterations in the 2a
extent of myocardlal lschemic injury. Am J Cardiol 29:223-230, 13. Sobel BE, Bresnahan OF, Shell WE, et al: Estimation of infarct
1972 size in man and its relation to prognosis. Circulation 46:640-
6. Krotklewskl A, Gajewska-Llpka J, Szelemetko J, et al J: Multl- 648.1972
lead electrocardiogram In relation to serum enzyme In acute 14. Nielsen BL: ST-segment elevation in acute myocardlal Infarc-
myocardlal Infarctlon. Br Heart J 353991-996, 1973 tion. Circulation 48:33a-345, 1973
7. Herman MV, Helnle RA, Klein MD, et al: Localized dlsorders In 15. Dubnow MH, Burchell HB, Tltus JL: Postinfarction ventricular
myocardlal contractlon. N Engl J Med 277:222-232, 1967 aneurysm. Am Heart J 70:753-760, 1965
6. Frledberg CK: Acute coronary occlusion and myocardlal Infarc- 16. Gorlln R, Klein MD, Sullivan JM: Prospective correlative study
tion. Cllnical features. In, Diseases of the Heart, 3rd edltlon. Phll- of ventricular aneurysm. Am J Med 42:5 12-531, 1967
adelphla, WB Saunders, 1966, p 656 17. Mourdjlnls A, Olsen E, Raphael MJ, et al: Cllnclal diagnosis and
9. Ester EH Jr: Electrocardiography and vectorcardiography. In, prognosis of ventricular aneurysm. Br Heart J 30:497-513,
The Heart, Arteries, and Veins, 3rd edltlon (Hurst JW. Logue RB, 1968

614 May 1975 The American Journal of CARDIOLOGY Volume 35