Full text of DOI:10.1002/clc.4960260205

Clin. Cardiol. 26, 67–70 (2003)
Predischarge Exercise Electrocardiogram and Stress Echocardiography Can
Predict Long-Term Clinically Driven Revascularization Following Acute
Myocardial Infarction
ALESSANDRO DESIDERI, M.D., FESC, FACC, RICCARDO BIGI, M.D., FESC, LAURO CORTIGIANI, M.D., RICCARDO RAMBALDI, M.D., FESC,
DANIELA SABBADIN, R.N., GIANPIERO CURTI, R.N., LEOPOLDO CELEGON, M.D.
Cardiovascular Research Foundation, S.Giacomo Hospital, Castelfranco Veneto, Italy
Summary
Background: Predischarge stress testing provides subopti-
variate analysis selected peak WMSI, positive exercise ECG,
and low workload positive exercise ECG as independent pre-
dictors of late revascularizations.
mal prediction of spontaneous hard events following uncom-
plicated acute myocardial infarction (AMI).
Hypothesis: This study was aimed at assessing whether soft
cardiac ischemic events requiring late revascularization could
be predicted more accurately.
Conclusions: Predischarge stress testing identifies the long-
term occurrence of soft ischemic events driving late revascu-
larization after uncomplicated AMI.
Methods: In all, 428 patients undergoing exercise electro-
cardiography (ECG) and stress echocardiography (SE, 345
dobutamine and 83 dypiridamole) within 15 days of uncom-
plicated AMI were followed up for 425 (range 20–2220)
days. Soft ischemic events (effort angina > class II [Canadian
Cardiovascular Society Classification] and unstable angina)
driving late (> 6 months) revascularization were regarded
as endpoints.
Results: A total of 58 events (29 effort and 29 unstable angi-
na with subsequent 47 coronary artery bypass grafts and 11
percutaneous transluminal coronary angioplasties) occurred:
26 in patients with positive exercise ECG and 34 in patients
with positive SE. Univariate predictors of revascularizations
were positive exercise ECG (p = 0.0001), peak wall motion
score index (WMSI) (p = 0.0009), low workload (p = 0.0018),
rest WMSI (p = 0.02) and positive SE (p = 0.02). Cox multi-
Key words: prognosis, acute myocardial infarction, stress test
Introduction
Patients undergoing revascularization dictated by coronary
symptoms are generally excluded from subsequent follow-up
in prognostic studies to avoid post-test referral bias in primary
revascularization. However, subsequent clinical events driving
late revascularization could be underestimated. The aim of our
study was to assess the ability of predischarge stress testing to
predict clinically driven late revascularization procedures after
acute myocardial infarction (AMI).
Materials and Methods
Patient Population
The initial study population consisted of 511 consecutive
survivors from uncomplicated AMI undergoing predischarge
stress testing. All of them underwent maximum symptom-
limited exercise electrocardiogram (ECG) and stress echocar-
diography (SE) within 15 days of hospital admission on differ-
ent days and in random order, after a 48-h pharmacologic
washout. Eighty-three patients underwent immediate revascu-
larization as a result of stress testing and were excluded from
analysis. The remaining 428 patients represented the study
population and were prospectively followed up for revascular-
ization procedures that were undertaken for documented clin-
ical reasons.
Address for reprints:
Alessandro Desideri, M.D.
Cardiovascular Research Foundation
S.Giacomo Hospital
Castelfranco Veneto 31033, Italy
e-mail: aldesi@tin.it
Received: July 20, 2001
Accepted with revision: February 26, 2002

68
Clin. Cardiol. Vol. 26, February 2003
Exercise Electrocardiogram
Statistical Analysis
Exercise ECG was performed by means of an electromag-
netically braked, upright bicycle ergometer with 25 W incre-
mental loading every 2 min. Low workload was defined as an
achieved load < 100 W. A positive test was defined by the ap-
pearance of >1 mm horizontal or downsloping ST-segment
depression at 0.08 s after the J point, with or without chest
pain. Test endpoints were achievement of maximal age-relat-
ed heart rate, limiting symptoms, typical chest pain, ST-seg-
ment depression > 3 mm, decrease in blood pressure, or ap-
pearance of complex arrhythmias.
Continuous variables were expressed as mean ± 1 standard
deviation and were compared using the t-test. The 95% confi-
dence interval (CI) was reported where appropriate. Propor-
tions were compared by means of chi-square statistics. Cox
multivariate stepwise analysis (SAS statistical software, SAS
Institute, Cary, N.C., USA) was used to assess the outcome
predictors. Clinical variables (age, gender, pretest probability
calculated according to post-AMI Gruppo Italiano per lo
Studio della Streptochinasi nell’ Infarto miocardico (GISSI)
risk chart,4 exercise ECG (positive exercise ECG, peak dou-
ble product, workload at positivity), and echocardiographic
variables (positive stress echo, rest and peak WMSI, rest-
stress WMSI variation [(∆WMSI) representing the net contri-
bution of ischemia to peak WMSI] were tested. Cut-off value
for discriminating risk was determined by receiver-operator
characteristics (ROC) curve analysis. A p level of <0.05 was
set as statistically significant.
Stress Echocardiography
Stress echocardiography was performed using Hewlett
Packard Sonos 2500 or Sonos 5500 (Philips Medical Systems/
Agilent Technologies, Andover, Mass., USA) or Acuson XP-
10 (Acuson Corp. [Siemens], Mountain View, Calif. USA)
imaging systems. Either dipyridamole (n = 83) or dobutamine
(n = 345) were used as pharmacologic stressors. Dipyridamole
was given intravenously according to the high-dose protocol
(up to 0.84 mg/kg over 10 min).1 Dobutamine was infused ac-
cording to a protocol based on 3-min stages of 5, 10, 20, 30,
and 40 mcg/kg/min plus additional atropine (0.25 up to 1 mg)
if required for achieving the 85% of age-predicted maximum
heart rate. The selection of either dypiridamole or dobutamine
was the choice of the cardiologist performing the examination.
Test positivity was defined as the appearance of a transient
ventricular dyssynergy, which was absent or minor at baseline,
and was regarded among stop criteria. The test was also dis-
continued in case of peak drug dose, achievement of conven-
tional nonechocardiographic “ischemic” endpoints (severe
chest pain and/or ST-segment shift > 0.2 mV), intolerable
symptoms, and limiting side effects such as hypertension (sys-
tolic blood pressure > 220 mmHg, diastolic blood pressure
> 120 mmHg), hypotension (30 mmHg decrease in blood
pressure), and complex arrhythmias. The left ventricular wall
was divided into 16 segments according to the recommenda-
tions of the American Society of Echocardiography,2 and mo-
tion was scored as follows: 1 = normal, 2 = hypokinetic, 3 =
akinetic, 4 = dyskinetic. A rest and peak wall motion score in-
dex (WMSI) was derived by the summation of the individual
segment score divided by the number of segments interpreted.
Resting akinesia becoming diskinesia during stress was not
taken into account.3
Results
The characteristics of the study population are reported in
Table I; those of revascularized and nonrevascularized patients
are reported in Table II. No major complication occurred with
exercise ECG, but one case of sustained and three cases of
nonsustained ventricular tachycardia occurred during dobu-
tamine SE. Both tests were performed at a mean of 9 ± 4 days
after AMI.
Exercise Electrocardiogram
Maximal achieved workload and rate–pressure product
were 98 ± 30 W and 24116 ± 6057, respectively. Of the sub-
jects studied, 144 were not able to exercise up to 100 W. Ex-
ercise ECG was positive in 122 (28.5%) and negative in 306
(71.5%) patients.
Stress Echocardiography
The average dose of dipyridamole was 55 ± 5 mg and that
of dobutamine was 35 ± 2 µg/kg/min. Mean rest and peak
WMSI were 1.43 ± 0.29 and 1.51 ± 0.30, respectively. Peak
TABLE I Clinical characteristics of the study population
Follow-Up
Age (years)
58 ± 9
For the purpose of the present analysis, events were de-
fined as any late (>6 months) clinically driven revasculariza-
tion by means of either coronary artery bypass graft (CABG)
or percutaneous transluminal coronary angioplasty (PTCA).
Patients were referred for coronary angiography at the discre-
tion of the staff cardiologist on the basis of clinical symptoms
(effort angina CCS > 2 and unstable angina). The decision to
perform revascularization procedures was made by the cardi-
ologist blinded to the study aim.
Male gender
Infarct location (ECG)
Anterior
Inferior
Non-Q
429 (84)
158 (31)
255 (50)
98 (19)
47 ± 5
Ejection fraction (%)
Pretest likelihood of cardiac events
5.3 ± 2.4
Datapresentedarenumber(%)ofpatientsforcategoricalvariablesand
mean (± standard deviation) for continuous variables, respectively.

A. Desideri et al.: Stress testing and late revascularizations
69
TABLE II Clinical and stress testing characteristics of revascularized
and nonrevascularized patients
while positive PV was 25% for exercise ECG and 21% for SE.
Log-rank test showed a significantly lower (p < 0.001) event-
free survival in patients with positive exercise ECG or SE than
in those with negative tests. Univariate predictors of late revas-
cularizations were positive exercise ECG (p = 0.0001), peak
WMSI (p = 0.0009), low workload (p = 0.0018), rest WMSI (p
= 0.02), and positive SE (p = 0.02). Cox multivariate analysis
selected peak WMSI, (relative risk [RR] 3.8, 95% CI 1.8–8.1,
p < 0.001), positive exercise ECG (RR 2.6, 95% CI 1.5–4.4,
p < 0.001), and low workload-positive exercise ECG (RR 1.8,
95% CI 1–3, p < 0.05) as independent predictors of late revas-
cularization. Receiver-operator characteristics analysis identi-
fied 1.43 as the best cut-off for peak WMSI to discriminate be-
tween patients at high and low risk for late revascularization.
Non
Revascularized revascularized
p
(n = 58)
(n = 370)
Value
Age (years)
Females (%)
56 ± 9
6 (10)
58 ± 10
65 (18)
NS
NS
Non-Q AMI (%)
Thrombolysis (%)
Pretest likelihood (%)
Positive exercise ECG (%) 26 (45)
Workload (W) 84 ± 30
6 (10)
33 (57)
5.1 ± 2.4
63 (17)
NS
NS
NS
< 0.05
< 0.0001
200 (54)
5.4 ± 2.3
96 (26)
100 ± 30
Rate–pressure product 23,000 ± 6,153 24,243 ± 5,996 NS
Positive SE (%)
Rest WMSI
Peak WMSI
∆WMSI
34 (59)
1.5 ± 0.3
1.6 ± 0.3
0.1 ± 0.1
583 ± 455
151 (41)
1.4 ± 0.3
1.5 ± 0.3
0.1 ± 0.1
725 ± 578
< 0.05
< 0.05
< 0.05
NS
Discussion
Follow-up (days)
NS
The main result of our study is that predischarge risk strati-
fication by means of noninvasive stress testing allows identi-
fication of patients at higher risk for late (> 6 months) revas-
cularization procedures following uncomplicated AMI. Con-
ventional stratification strategies for patients recovering from
AMI are presently aimed at identifying those at risk for
“hard” cardiovascular events, that is, death or reinfarction. On
the other hand, in the thrombolytic era, these patients are at
overall low risk, and indeed the real impact of risk stratifica-
tion with noninvasive stress testing is a matter of debate.5 At
the same time, economic issues and need for cost contain-
ment give potential importance to “soft events,” for example,
unstable angina, that do not predicate length of life but may
enhance social burden. It therefore seems important to verify
whether noninvasive evaluation can early identify patients at
higher risk of late revascularization procedures.
Abbreviations: AMI = acute myocardial infarction, ECG = electro-
cardiogram, SE = stress echocardiography, WMSI = wall motion
score index, ∆WMSI = wall motion score index variation, NS = not
significant.
rest WMSI variation was 0.08 ± 0.12. Stress echocardiogra-
phy was positive in 185 (43%) and negative in 243 (57%) pa-
tients. Positivity was found in 147 of 346 (42%) patients un-
dergoing dobutamine and in 38 of 83 (45%) patients under-
going dypiridamole SE. Agreement between exercise ECG
and SE was poor (kappa value 0.20, 95% CI 0.10–0.30).
Outcome Prediction
We and others1, 6–14 previously demonstrated the ability of
stress echocardiography to identify risk in patients recovering
from AMI, but to the best of our knowledge this is the only
study that specifically addresses the issue of late clinically driv-
en revascularizations. The extent of ischemia at peak stress
proved to be the strongest independent predictor of late inter-
ventions in the present study. This confirms and expands previ-
ous observations that WMSI at peak represents a major deter-
minant of prognosis after AMI.1 Therefore, future prospective
studies focusing on interventions (maximal medical vs. direct
surgical) with cost–benefit analysis taking into account impact
on late revascularization seem warranted.
Median follow-up time was 425 days (range 20–2,220, 1st
quartile 265 and 3rd quartile 1,105 days). During follow-up,
58 events (29 effort and 29 unstable angina with subsequent 47
CABG and 11 PTCA) occurred. The results of stress testing
according to presence/absence of revascularization are shown
in Figure 1. Stress echocardiography was more sensitive (59
vs. 45%) and less specific (60 vs. 76%) than exercise ECG.
Both tests had high (89%) negative predictive value (PV),
74
59
59
Study Limitations
55
45
41
41
Limitations of our study are represented by pooled results
of dipyridamole and dobutamine stress echocardiography.
However, although the two drugs can induce ischemia by dif-
ferent physiologic mechanisms, they have very similar diag-
nostic and prognostic accuracy.15–17 Moreover, we do not pro-
vide information about medical treatment in patients with and
without events. It seems conceivable though that patients with
positive tests might be treated even more aggressively, which
may strengthen the outcome of the study.
26
ECG pos
SE pos
ECG neg SE neg
FIG. 1 Results (% of patients) of stress testing according to revas-
cularization procedures. ꢀ = No-revasc (n = 370), ꢀ = revasc (n =
58). ECG = exercise ECG, neg = negative, pos = positive, revasc =
revascularized, SE = stress echocardiography.

70
Clin. Cardiol. Vol. 26, February 2003
7. Desideri A, Bigi R, Suzzi GL, Coletta C, Gregori D, Valente G, Fioretti P:
Medical regimen of the two groups was not included in data
collection, and the importance of this confounding factor
could not be evaluated in the present study.
Stress echocardiography and exercise electrocardiography for risk stratifi-
cation after non-Q wave uncomplicated myocardial infarction. Am J Cardiol
1999;84:739–741
8. Desideri A, Bigi R, Terlizzi R, Cortigiani L, Suzzi GL, Ginocchio G,
Celegon L, Fioretti P: Noninvasive risk stratification in women with un-
complicated acute myocardial infarction. Am J Cardiol 2000;86:333–336
9. Bigi R, Galati A, Desideri A, Coletta C, Bax JJ, Occhi G, Gregari D, Fioretti
P: Incremental prognostic value of stress echocardiography as adjunctive
test to exercise electrocardiography following uncomplicated myocardial
infarction. Heart 2001;85:417–423
10. Rambaldi R, Bigi R, Desideri A, Curti G, Occhi G: Prognostic usefulness of
dobutamine-induced ST-segment elevation and T-wave normalization after
uncomplicated acute myocardial infarction. Am J Cardiol 2000,1;86:
786–789
11. Bigi R, Desideri A, Bax JJ, Fiorentini C, Galati A, Fioretti P: Prognostic in-
teraction between viability and residual myocardial ischemia by dobutamine
stress echocardiography in patients with acute myocardial infarction and
mildly impaired left ventricular function. Am J Cardiol 2001;87:283–288
12. Carlos M, Smart S, Wynsen J, Sagar K: Dobutamine stress echocardiogra-
phy for risk stratification after myocardial infarction. Circulation 1997;95:
1402–1406
Conclusions
Our study shows that predischarge testing can predict long-
term need for revascularization procedures. In particular,
echocardiographic indices of inducible ischemia are the best
predictors of late (> 6 months) PTCA and CABG. Cost–bene-
fit analysis of aggressive secondary prevention measures in
these patients seems advisable.
References
13. Greco CA, Salustri A, Seccareccia F, Ciavatti M, Biferali F, Valtorta C,
Guzzardi G, Falcone M, Palmara A: Prognostic value of dobutamine stress
echocardiography early after uncomplicated myocardial infarction: A com-
parison with exercise electrocardiography. J Am Coll Cardiol 1997;29:
261–267
14. Sicari R, Picano E, Landi P, Pingitore A, Bigi R, Coletta C, Heyman J,
Casazza F, Previtali M, Mathias W Jr, Dodi C, Minardi G, Lowenstein J,
Garvfallidis X, Cortigiani L, Morales M, Raciti M: Prognostic value of
dobutamine-atropine stress echocardiography early after acute myocardial
infarction. J Am Coll Cardiol 1997;29:254–260
15. Minardi G, Di Segni M, Manzara C, Pulignano G, Chiantera A, De Santis F,
Armiento G, Vajola F, Giovannini E: Diagnostic and prognostic value of
dipyridamole and dobutamine stress echocardiography in patients with Q-
wave acute myocardial infarction. Am J Cardiol 1997;80:847–851
16. Schroeder K, Wieckhorst A, Voller H: Comparison of the prognostic value
of dipyridamole- and dobutamine-stress echocardiography in patients with
known or suspected coronary artery disease. Am J Cardiol 1997;79:
1516–1517
17. Pingitore A, Picano A, Varga A, Gigli G, Cortigiani L, Previtali M, Minardi
G, Colosso M, Lowenstein J, Mathias W Jr, Landi P: Prognostic value of
pharmacological stress echocardiography in patients with known or sus-
pected coronary artery disease: A prospective, large scale, multicenter,
head-to-head comparison between dipyridamole and dobutamine test. J Am
Coll Cardiol 1999;34:1769–1777
1. Picano E, Landi P, Bolognese L, Chiaranda G, Chiarella F, Seveso G,
Sclavo MG, Gandolfo N, Previtali M, Orlandini A, for the EPIC Study
Group: Prognostic value of dipyridamole echocardiography early after un-
complicated myocardial infarction: A large-scale, multicenter trial. Am J
Med 1993;6:608–618
2. American Society of Echocardiography Committee on Standards, Subcom-
mittee on Quantitation of Two-Dimensional Echocardiograms: Recommen-
dations for quantitation of the left ventricle by two-dimensional echocardio-
graphy. J Am Soc Echocardiogr 1989;2:358–367
3. Arnese M, Fioretti PM, Cornel J, Postma-Tioa J, Reiis A, Roelandt J:
Akinesis becoming dyskinesis during high-dose dobutamine stress echocar-
diography: A marker of myocardial ischemia or a mechanical phenomenon?
Am J Cardiol 1994;73:896–899
4. Marchioli R, Bomba E, Di Pasquale A, Geraci E, Giannuzzi P, Marfisi R,
Pede S, Schweiger C, Tavazzi L, Tognoni G, Valagussa F: A chart of post-
acute myocardial infarct risk: Results in the first 18 months of the GISSI fol-
low-up study. G Ital Cardiol 1998;28:416–433
5. Desideri A: Role, timing and cost-effectiveness of non-invasive strategies af-
ter uncomplicated myocardial infarction. G Ital Cardiol 1999;29:1549–1552
6. Bigi R, Galati A, Curti G, Coletta C, Barlera S, Partesana N, Bordi L, Ceci
V, Occhi G, Fiorentini C: Prognostic value of residual ischaemia assessed by
exercise electrocardiography and dobutamine stress echocardiography in
low-risk patients following acute myocardial infarction. Eur Heart J
1997;18:1873–1881