Download Improvement of left ventricular contractile function by

THERAPY AND PREVENTION
EXERCISE TRAINING
Improvement of left ventricular contractile function
by exercise training in patients with coronary artery
disease*
ALI A. EHSANI, M.D., DANIEL R. BIELLO, M.D., JOAN SCHULTZ, M.S., BURTON E. SOBEL, M.D.,
AND JOHN 0. HOLLOSZY, M.D.
Downloaded from http://circ.ahajournals.org/ by guest on June 11, 2017
ABSTRACT To determine whether prolonged, intense exercise training can improve left ventricular
function in patients with coronary artery disease, we studied 25 patients, 52 ± 2 years old (mean +
SE), who completed a 12 month program of endurance exercise training and 14 additional patients with
comparable maximal exercise capacities and ejection fractions who did not exercise. The training
program consisted of endurance exercise of progressively increasing intensity, frequency, and duration. During the last 3 months the patients were running an average of 18 miles/week, or doing an
equivalent amount of exercise on a cycle ergometer. Maximal attainable V02 increased 37% (p <
.001). Of the 10 patients with effort angina, five became asymptomatic, three experienced less angina,
and two were unchanged after training. Ejection fraction was determined by equilibrium radionuclide
ventriculography. At rest, ejection fraction was 53 + 3% before and 54 ± 3% after training (p NS).
Ejection fraction did not change during maximal supine exercise before training (52 ± 3%), but after
training it increased to 58 + 3% (p < .01). During maximal exercise, systolic blood pressure and the
rate-pressure product were higher after training. The systolic blood pressure-end-systolic volume
relationship was shifted upward and to the left, with an increase in maximal systolic blood pressure (p
< .001) and a smaller end-systolic volume (p < .05), providing evidence for an improvement in
contractile state after training. In patients who did not participate in training neither this relationship nor
the ejection fraction response to exercise was changed after 12 months. Exercise-induced regional wall
motion disorders worsened in the training group. Our finding that prolonged, intense exercise training
can bring about an improvement in left ventricular contractile function essentially independent of
cardiac loading conditions in some patients with coronary artery disease provides evidence for a
reduction in the severity of myocardial ischemia despite an increase in the myocardial 02 requirement.
Circulation 74, No. 2, 350-358, 1986.
=
EXERCISE TRAINING increases maximal exercise
capacity, endurance, and the minimal work rate required to induce myocardial ischemia in patients with
coronary artery disease.'-4 These effects have been attributed to adaptations in skeletal muscle and the autonomic nervous system2' 7 that result in smaller increases in heart rate and systolic blood pressure, and
From the Section of Applied Physiology and the Cardiovascular
Division, Department of Medicine, the Division of Nuclear Medicine,
Department of Radiology, the Irene Walter Johnson Institute of Rehabilitation, and the Mallinckrodt Institute of Radiology, Washington
University School of Medicine.
Supported by NHLBI grants HL22215 and HL17646-SCOR in Ischemic Heart Disease.
Address for correspondence: Ali A. Ehsani, M.D., Department of
Medicine, Washington University School of Medicine, 4566 Scott
Ave., St. Louis, MO 63110.
Received Aug. 12, 1985; revision accepted April 17, 1986.
*AIl editorial decisions for this article, including selection of reviewers and the final disposition, were made by a guest editor. This procedure applies to all manuscripts with authors from the Washington University School of Medicine.
350
therefore in a reduced myocardial 0° requirement, at
any given submaximal exercise intensity.2 6 The results of previous studies have suggested that exercise
training does not improve myocardial blood supply
and left ventricular contractile function at the same
myocardial 02 requirement5' 9-12 in patients with coronary artery disease. However, in experimental animals
training has been shown to reduce myocardial ischemia and improve myocardial blood supply and left
ventricular contractile function.'3-19
One explanation for this difference may be that an
insufficient training stimulus was used in previous
clinical studies. Our recent experience with the effects
of prolonged, high-intensity training is consistent with
this view.2>23 We have found that, in addition to peripheral adaptations, long-term exercise training of
progressively increasing duration, frequency, and intensity can elicit further adaptations suggestive of improvement in myocardial ischemia and left ventricular
CIRCULATION
THERAPY AND PREVENTION-EXERCISE TRAINING
change in ejection fraction from rest to exercise of 5% or less
and/or development of discrete systolic regional wall motion
abnormalities with exercise. Coronary artery disease was documented by unequivocal prior myocardial infarction and/or effort
angina with angiographically proven fixed coronary artery stenosis. Of the 25 study patients, 22 had sustained a prior myocardial infarction, and 10 had chronic stable effort angina. Four
had undergone coronary revascularization, one with recurrence
of angina and two with myocardial infarction after surgery but
before enrollment in the study. The interval between the major
coronary event (myocardial infarction or coronary revascularization) and enrollment in the study was at least 3 months and
averaged 11 + 3 months. Two patients had left bundle branch
block (LBBB); LBBB was persistent in one patient (No. 11,
table 1) and induced by exercise (rate dependent) in the other
(No. 12, table 1). Two patients had a left ventricular aneurysm
(Nos. 10 and 24, table 1). Of the 14 patients who had coronary
angiography, four had triple-vessel disease, six double-vessel
disease, and four single-vessel disease. Seventeen patients were
taking ,3-adrenergic-blocking agents, 10 long-acting nitrates,
four calcium antagonists, and two patients were taking digoxin.
function, as reflected by less ST segment depression at
the same rate-pressure product20 and a higher stroke
volume at a comparable heart rate and peripheral
vascular resistance.21 The purpose of the present study
was to test the hypothesis that prolonged and intense
endurance exercise can improve left ventricular
contractile function in patients with coronary artery
disease.
Methods
Patients. Twenty-five patients with coronary artery disease
who had an abnormal left ventricular exercise response before
training completed 12 months of endurance exercise training.
This group consisted of 24 men and one woman with an average
age of 52 + 2 years (mean + SE). All of the patients provided
written consent, and the study protocol was approved by the
Human Studies Committee of Washington University. An abnormal left ventricular exercise response was defined as a
Downloaded from http://circ.ahajournals.org/ by guest on June 11, 2017
TABLE 1
Effects of exercise training on left ventricular function
EF
Patient
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Mean
+SE
Controls
Mean
±SE
Rest
Age/
Peak
gender
Initial
Final
Initial
42/M
46/M
51lM
54/M
55/M
43/M
62/M
39/M
50/M
63/M
64/M
69/M
56/M
46/M
64/M
60/M
551M
55/M
57/M
59/M
34/F
38/M
56/M
61/M
31/M
52
49
54
52
45
63
53
21
61
43
43
29
73
55
53
50
44
57
64
60
61
64
60
71
31
66
53
59
46
48
58
65
58
20
64
43
45
30
73
55
58
43
44
64
73
56
60
60
70
68
32
66
54
±2
+3
+3
34
51
51
50
65
75
21
63
47
41
29
73
53
56
48
42
47
60
49
60
63
55
69
28
67
52
3
(%o)
Change
exercise
Final
68
59
50
60
70
82
22
65
45
40
19
76
56
59
50
51
61
73
59
60
72
67
72
35
75
58A,B
+3
48.4
52
54
52
53
±2
±3
±3
+3
+3
Peak exercise
(RPP x 103)
Initial
Final
Initial
Final
-15
-3
-1
5
2
23
0
2
4
-2
0
0
-2
3
-2
-2
-10
-4
- 11
-1
-1
-5
-2
-3
9
13
2
2
5
24
2
1
2
-5
-11
3
1
1
7
7
-3
0
3
0
12
-3
4
3
9
4c
±1
25.95
17.71
21.80
27.40
21.31
19.89
27.20
15.45
16.63
21.81
17.25
18.13
15.32
13.10
16.15
21.54
24.42
24.57
26.25
16.07
16.75
24.25
26.67
25.94
24.10
21.03
24.01
22.88
27.40
31.02
22.43
33.18
27.47
20.70
19.04
27.93
18.79
20.46
19.52
16.15
24.87
26.40
25.31
23.96
27.74
22.89
18.88
30.89
26.16
23.22
25.30
1
-1
+1
-0.4
±1.0
1.4
+2
EF = ejection fraction; RPP = rate-pressure product.
Ap < .01, compared with EF at rest; Bp < .001, compared with initial peak exercise EF; cp < .005,
initial vs final.
Vol. 74, No. 2, August 1986
+0.88
24.36
+1.10
initial vs
24.22D
+0.84
23.94
±1.16
final; Dp < .001,
351
EHSANI et al.
Downloaded from http://circ.ahajournals.org/ by guest on June 11, 2017
The medications and their dosages were constant throughout the
study except in one patient in whom the dose of propranolol was
reduced. In this patient, propranolol was adjusted to the initial
dosage for 10 days before final evaluation. The interval between
the last dose of medication and the exercise test was also similar
for each patient and averaged 13.8+ 2 hr initially and 13.2+ 2
hr (p= NS) at the end of the study.
Sixteen of the 25 patients, 53+ 2 years old, had exerciseinduced myocardial ischemia evidenced by effort angina, 0.1
mV or greater horizontal or downsloping ST segment depression, and/or discrete severe regional left ventricular contraction
abnormalities during exercise. Eight patients, 49 ± 4 years old,
all with a previous myocardial infarction, did not exhibit exercise-induced ischemia. One patient could not be classified because of LBBB and extensive myocardial scar (No. 1 1, table 1).
An additional group of patients who did not exercise was used
to evaluate the reproducibility of the measurements and the
likelihood of spontaneous improvement in left ventricular function over a 12 month period. It comprised 13 men and one
woman (average age 48+ 2 years) with documented coronary
artery disease who were eligible for enrollment in our exercise
program and similar to the training group in terms of age,
maximal exercise capacity, and ejection fraction (table 1).
These patients lived too far away or had schedule conflicts that
prevented them from participating in the exercise program.
Eleven of these patients had sustained a prior myocardial infarction. Five had chronic stable effort angina. The interval between
the major coronary event (myocardial infarction or coronary
revascularization) and initial testing was at least 3 months and
averaged 13.6 ± 4 months.
Treadmill exercise test and maximal 02 uptake capacity
(VO2max). Maximal exercise testing was performed on a motor-driven treadmill according to the Bruce protocol24 with a
repeat treadmill test 1 week later for measurement ofVO2max
by the following protocol: After 5 min of warm-up exercise that
consisted of walking at 0 grade at a speed of 1.7 or 2.5 mph,
patients began to exercise at the speed equivalent to the next to
the last stage attained with the prior Bruce protocol with the
grade set at either 5% or 10%, depending on the patients' exercise capacity. From this point on the speed and grade were
increased alternately every 2 min. The patients breathed through
a Daniels valve, and expired gases were collected in neoprene
meteorologic balloons at 60 or 30 sec consecutive intervals.
Oxygen and CO2 were analyzed with a mass spectrometer (Perkin-Elmer MA 11 00). Expired volumes were measured with a
Tissot spirometer. In 15 patients (14 without and one with
angina) it was possible to obtain true V02max, defined as attainment of the leveling-off criterion, and/or a respiratory exchange
ratio of 1. 15 or greater, signifying marked hyperventilation that
usually occurs with VO2max.25 In the remaining 10 patients
(nine with angina and one with LBBB), peak or symptomlimited 02 uptake (peak V02) rather than VO2max was obtained.
Left ventricular function at rest and with exercise. Left
ventricular performance was assessed by electrocardiographically gated cardiac blood pool imaging with erythrocytes labeled in vivo by intravenous injection of 7.7 mg of stannous
pyrophosphate followed in 20 min by injection of 25 mCi of
99mTc. Images were obtained with a standard-field-of-view scintillation camera (Siemens LEM) equipped with a 0.64 cm thick
Nal crystal and with a low-energy, medium-resolution, parallelhole collimator. Images were obtained with the patients supine
and with the scintillation camera positioned in the left anterior
oblique (LAO) projection providing optimal separation of the
ventricles (35 degrees). Caudal angulation (15 degrees) was
used to maximally separate the left atrial and left ventricular
images. Data were collected in the frame mode (32 frames per
352
RR interval) in a 64x 64 pixel matrix and processed off-line
with a VAX 11/750 minicomputer equipped with a Lexidata
display unit. The average number of background-subtracted
counts per frame in the left ventricular region of interest at enddiastole at rest was 11,288 and that at peak supine bike exercise
was 5377. Ejection fraction (EF) was calculated as: EF= (EDC
- ESC) 1 00/EDC, where EDC and ESC are the left ventricular end-diastolic and end-systolic counts, respectively, corrected for background activity. With this method, reproducibility is
high, and left ventricular ejection fraction correlates well with
results of contrast left ventriculography.26
The left ventricular end-diastolic volume (LVEDV) was calV= 8
culated by the standard geometric area-length
A2/3 1, where V is volume, A is the area, and1 is the long axis
of the left ventricle. Spatial calibration factors for the X and Y
axes of the digital images were obtained with the use of a
phantom as described by Esser et a The area and the long axis
of the left ventricle were determined with an end-diastolic region of interest as previously described.29 The left ventricular
end-systolic volume (LVESV) and stroke volume were derived
and LVEDV. The scintigraphic method
ejection
used for volume measurements has been validated in our laboratory. Results correlate closely with those of contrast ventriculography (r = .97).
After images had been obtained in patients at rest, each performed a graded supine cycle exercise test with an electronically
braked bicycle ergometer (Engineering Dynamics Corp.) that
maintains a constant work rate over a wide range of pedaling
frequencies. The pedaling rate was between 65 and 70 rpm.
Work rates were increased by 25 W every 3 min until severe
fatigue or angina developed. Images were obtained during the
last 2 min of each stage of the exercise in the same modified
LAO projection as that used for rest images. Heart rate was
recorded every minute. Blood pressure was measured with a
mercury sphygmomanometer at the second and third minutes of
each stage of the exercise test. Peak heart rate and systolic blood
pressure during supine cycle ergometer exercise are reported as
the average of the values measured at the second and third
minutes of the last stage of the exercise test.
The following parameters were used to evaluate left ventricular contractile function:
changes in ejection fraction as a
function of systolic blood pressure (used as an estimate of afterload), (2) the systolic blood pressure-end-systolic volume relationship, and (3) the relationship between left ventricular stroke
work and end-diastolic volume (Frank-Starling mechanism).
Regional left ventricular contraction abnormalities were evaluated subjectively by three investigators blinded to the patients'
category (training or control) and clinical status. Differences in
opinion were resolved by consensus. Left ventricular ejection
fraction and end-diastolic volume were measured by only one
investigator blinded to the patients' status. The intraobserver
variabilities in the determinations
ventricular ejection
fraction and end-diastolic volume at rest were 0. 1 + 0. 7% (r =
.95) and 6 ± 3 ml (r = .88), and those at peak exercise were 1.7
± 0.9% (r .92) and 7 3 ml (r = .92), respectively, in 10
randomly selected subjects.
Plasma lipids. Blood samples were obtained after subjects
had fasted for 14 hr. Plasma cholesterol, triglyceride, and highdensity lipoprotein (HDL) cholesterol levels were assayed as
The low-density lipoprotein (LDL) chopreviously
lesterol level was calculated as previously reported.30 No specific diet recommendations were made. However, patients were
encouraged to adhere to diets recommended by their personal
physicians before referral to the exercise program.
Exercise program. The 12 month long exercise program
used in this study has previously been described in detail.20
Briefly, patients were expected to exercise three times a week
x
method27:
wr
.28
fraction
from
(I)
of left
=
+
described.30
CIRCULATION
THERAPY AND PREVENTION-EXERCISE TRAINING
for the first 3 months and five times per week thereafter. The
duration of exercise sessions was 40 to 45 min for the first 3
months and was increased progressively to 50 to 60 min of
exercise exclusive of the warm-up and cool-down periods over
the next 3 months. The training intensity ranged from 60% to
70% of the maximal attainable V02 for the first 3 months and
was then gradually increased to 70% to 90% of maximal attainable V02 over a 6 month period. The intensity of training was
assessed by measurement of heart rate and the relationship between heart rate and V02 and was verified periodically by collection of expired air and measurement of V02 during exercise.
All exercise sessions were supervised by a physician.
Statistical analysis. Student's t test for paired observations
was used for comparison of the data before and after training.
Chi-square analysis was performed when appropriate. Because
LVEDV and LVESV conformed to log normal distributions,
they were logarithmically transformed for the purposes of statistical analysis. However, the actual values are presented in the
text, figures, and tables. Data are expressed as mean ± SE.
Downloaded from http://circ.ahajournals.org/ by guest on June 11, 2017
Results
Baseline characteristics of patients. There were no significant differences with respect to age (52 ± 2 vs 48
± 2 years), VO2max (23 ± 0.6 vs 23 ± 1 ml/kg/min),
or left ventricular ejection fraction (rest: 53 ± 3 vs 52
+ 3; exercise: 52 ± 3 vs 52 ± 3) between the exercising patients and those who did not exercise.
Exercise training. Exercise capacity and endurance
improved markedly in response to the 12 months of
training. For 10 patients the primary mode of training
was running; during the last 3 months of the program
they were running between 3.6 ± 0.3 to 5.2 ± 0.3
miles continuously per session, averaging 18.1 ± 1.6
miles/week. Thirteen patients were running 7.7 ± 1.3
miles/week in addition to performing exercise on a
cycle ergometer for 20 to 30 min per day. One patient
exercised predominantly on a cycle ergometer 60 min/
day, 5 days/week during the last 3 months of the program. The peak training intensity in the last 3 months
of the program was 89.4 ± 1.3% of maximal attainable V02 as documented by measurement of VO2 during exercise sessions, and the attendance rate was 4.2
± 0.1 sessions/week (minimum of 3 and maximum
of 5).
Maximal attainable V02 and symptoms. Maximal attainable V02 was increased by 37% (p < .001), from
23 ± 1 to 31.5 ± 1 ml/kg/min (1.85 ± 0.04 to 2.36
± 0.08 1/min, p < .001) for the entire group. True
VO2max rose by 39%, from 23 ± 1 to 32 ± 1
ml/kg/min (p < .001) in the 15 patients in whom it was
attainable. Peak Vo2 increased from 24 ± 1 to 31 ± 2
ml/kg/min (p < .001) in the remaining patients who
did not attain true VO2max. The exercise time during
maximal treadmill exercise (Bruce protocol) was increased by 41% (367 ± 25 vs 517 ± 25 sec, p <
.001). Maximal work rate during supine cycle exercise
Vol. 74, No. 2, August 1986
increased from 97 ± 4 to 122 ± 3 W (p < .001). In
control subjects peak supine work rate was 97 ± 5 W
initially and 93 ± 6 W a year later (p = NS).
Five of the 10 patients (Nos. 2, 5, 6, 8, and 15; table
1) who had effort angina became entirely asymptomatic, even during maximal treadmill exercise. In three of
the remaining five patients (Nos. 13, 19, and 20; table
1), angina was considerably less in frequency and severity. Angina was unchanged in the other two patients
(Nos. 14 and 18, table 1). There were no major complications attributable to exercise testing or training.
Neither symptoms nor maximal exercise capacity
changed in the 14 patients who did not participate in
exercise training.
Heart rate, blood pressure, and rate-pressure product.
Heart rate at rest decreased significantly from 64 ± 2
to 56 ± 2 beats/min (p < .001) with training. Systolic
and diastolic pressure at rest did not change. Submaximal heart rate, systolic and diastolic blood pressure,
and rate-pressure product at the same work intensity
(stage I of the Bruce protocol) were lower after training
(figure 1).
The peak heart rate attained during supine exercise
was 125 ± 4 before and 130 ± 3 beats/min after
training (p < .01), averaging 87% and 86% of the
maximal heart rate during treadmill exercise initially
and 1 year later, respectively. Systolic blood pressure
(figure 2) and rate-pressure product (table 1) during
supine cycle ergometer exercise were higher after 12
months of training. The rate-pressure product during
maximal treadmill and supine cycle ergometer exercise was similar.
In the 14 patients who did not participate in exercise
training, heart rate, blood pressure, and rate-pressure
product during submaximal and maximal treadmill or
supine cycle ergometer exercise did not change significantly over a 12 month interval (table 1).
ST segment changes. In the seven patients of the training group who were not on digoxin and for whom
volumetric data and good quality electrocardiographic
HEART RATE
150
BLOOD PRESSURE
HRXSBP x103
Systolic
IniialIO
Final M
130 -
191
161
m110o
=l
90 70-
D.astolic
_ MC
13
10
m
.ri.
FIGURE 1. Peripheral adaptations to intense exercise training characterized by lower heart rate (*p < .001), systolic (tp < .005) and
diastolic blood pressure (**p < .01), and rate-pressure product (*p <
.01) at a given absolute work rate. Data are mean -+ SE for 25 patients.
353
EHSANI et al.
B
A
Initiol 0
Final o
210 r
L
I 190
I
E
X
E
170
Left ventricular contractile function
190 _
-
X
E E~ ~ ~ ~ -170 17
~ ~ ~~~0
LCD,
10
O5 0~
-6
0
+6
0
+20
A ESV (ml)
Downloaded from http://circ.ahajournals.org/ by guest on June 11, 2017
FIGURE 2. Effect of training on systolic blood pressure-end systolic
volume (SBP-ESV) relationship. A, In the training group, ESV decreased from rest to exercise (AESV) after (0) but not before (0) training
(*p < .05). SBP was significantly higher (tp < .001) after training,
shifting the pressure-volume relationship upward and to the left, consistent with an increase in contractile state. Data are mean ± SE for 25
patients. B, In patients who did not train, the SBP-ESV relationship did
not change after 12 months. Data are mean + SE for 14 untrained
patients.
recordings were available, the magnitude of ST segment depression was significantly less (0.16 ± 0.02
mV before and 0.09 ± 0.03 mV after, p < .005) at an
equivalent rate-pressure product (19.95 X 103 + 1. 56
X 103 before and 19.89 x 103 + 1.44 x 103 after)
during supine cycle ergometer exercise after training
(figure 3). End-diastolic volume (126 + 12 ml before
and 155 ± 13 after, p < .05) and ejection fraction (54
± 3% vs 62 ± 3%, p < .005) were both higher during
exercise that elicited the same rate-pressure product
after training (figure 3). ST segment depression did not
change in the 14 patients who did not undergo exercise
training.
Left ventricular volumes. Volumetric data were available in 23 of the patients. LVEDV at rest was slightly
but significantly increased from 148 + 10 to 159 + 10
ml (p < .025) after 12 months of training. At peak
supine exercise, LVEDV was also significantly higher
after training (148 ± 19 ml before vs 163 + 10 ml
after, p < .005). However, the changes in LVEDV
from rest to exercise were negligible both in the trained
and untrained states. In the 14 patients who did not
exercise, LVEDV at rest (150 ± 14 vs 155 ± 13 ml)
and at peak supine exercise (162 + 18 vs 159 ± 15
ml) did not change over 12 months.
LVESV at rest was 75 ± 9 ml before and 78 ± 9 ml
(p = NS) after training. Peak exercise values for
LVESV were the same in the trained and untrained
states (75 + 9 vs 75 ± 11 ml). However, the directional changes in LVESV from rest to exercise were
different, showing a decrease in LVESV after training
(p < .05; figure 2). In the 14 patients who did not
exercise, LVESV did not change significantly at rest
354
(74 ± 10 vs 72 ± 9 ml) or with peak supine exercise
(82 + 13 vs 77 ± 11 ml).
Ejection fraction. Exercise training did not affect left
ventricular ejection fraction at rest (table 1). After 12
months of training, left ventricular ejection fraction at
the work rate equivalent to the peak work rate attained
before training (97 + 4 W) was significantly higher
(51 + 3% before vs 57 ± 3% after, n = 22, p <
.005). Left ventricular ejection fraction during maximal supine exercise was also significantly higher,
averaging 58 + 3% after compared with 52 ± 3%
before training (p < .001, n = 25; figure 3, table 1),
despite a significantly higher rate-pressure product (table 1) and systolic blood pressure, a crude index of
afterload (figure 4).
70
4
601_
*
Initial
Final
*
0
LL
UJ
J
50
41I
I
E 1501p
C
LU
WE
>
11(
Id
-
7
~
~~~_
Dr
-0.22
Lt
LULLJ
-()
ip A
0 16
.1
I--
20
24
HR X SBP x 103
FIGURE 3. Effect of training on ST segment depression, LVEDV,
and left ventricular ejection fraction (LVEF) at the same rate-pressure
product. The ST segment depression was less (*p < .005), and LVEDV
and LVEF were higher (tp < .05 and *p < .005, respectively) after (0)
than before (0) training (n = 7). Data are mean -+- SE. HR = heart rate.
CIRCULATION
THERAPY AND PREVENTION-EXERCISE TRAINING
64[A
rest
exercise
601-
t
B rest
exercise
Initial @--*
Final 0--0
,'0
r
56-
.
UL_
52-
48t
Systolic blood pressure-end-systolic volume relationship.
LVESV decreased with exercise after but not before
training (p < .05; figure 2). The systolic blood pressure attained during maximal exercise was significantly higher after training (171 + 5 mm Hg before
vs 189 + 4 after, p < .001), shifting the pressurevolume relationship up and to the left (figure 2).
In the patients who did not participate in training, no
significant changes in LVESV, systolic blood pressure, or the pressure-volume relationship occurred
(figure 2).
Left ventricular stroke work-end-diastolic volume rela-
110
110
140 170
SYSTOLIC BLOOD PRESSURE (mmHg)
140
170
200
200
Downloaded from http://circ.ahajournals.org/ by guest on June 11, 2017
FIGURE 4. Effect of training on left ventricular ejection fraction
(LVEF). A, In the training group, LVEF at rest was not significantly
changed after training. During maximal exercise, LVEF increased significantly (*p < .01) above the resting level after (0) but not before (0)
training. During maximal exercise, LVEF was significantly higher (tp
< .001) after training despite the attainment of a higher systolic blood
pressure (p < .001). B, In the nonexercising patients, LVEF did not
change with exercise initially (0) or 12 months later (0). Systolic blood
pressure values were also similar. Data are mean + SE for 25 trained
(A) and 14 untrained patients (B).
Before training, ejection fraction decreased from
rest to exercise in 15 patients, remained unchanged in
three, and increased modestly (-<5%) in seven. In contrast, ejection fraction decreased in only four patients,
was unchanged in two, but increased in the remaining
19 patients after training (p < .005). Left ventricular
exercise reserve (change in ejection fraction from rest
to exercise) was - 1.0 + 1% before and 4.0 + 1%
after training (p < .005). One patient (No. 6, table 1)
had a large increase in ejection fraction in response to
exercise both before and after training; he was included
in the study because he developed a regional wall
motion abnormality in response to exercise before
training.
In the patients who did not participate in the training
program, neither left ventricular ejection fraction at
rest nor at peak exercise changed significantly over 12
months (figure 4, table 1). It decreased with exercise in
seven of the 14 patients initially and six of the 14
patients 12 months later. Left ventricular exercise re2% a
1% initially and -1.4
serve was -0.4
year later (p = NS).
Ejection fraction at rest was virtually identical in the
patients who did not exercise and the training group
initially and after 12 months. However, the training
group showed a significantly higher peak exercise
ejection fraction than that of the sedentary patients 1
3% vs 53 + 3%, p < .001; table 1).
year later (58
Vol. 74, No. 2, August 1986
tionship (Frank-Starling mechanism). Left ventricular
stroke work was higher at rest and during maximal
exercise after (106 ± 6 g-m at rest and 159 ± 1 g-m
with exercise) than before training (93 ± 5 g-m at rest
and 125 ± 9 g-m with exercise, both p < .01). Furthermore, the change in left ventricular stroke work
increased from 31 ± 7 g-m before to 50 ± 8 g-m after
training (p < .005). LVEDV was significantly higher
after training at rest and during maximal exercise.
However, LVEDV did not increase from rest to exercise either in the trained or untrained subjects (0.1 ± 3
vs 4 ± 5 ml, p = NS). Therefore, the increase in left
ventricular stroke work from rest to maximal exercise
could not be attributed to an increase in preload.
In the 14 subjects who did not exercise, no significant changes in LVEDV or in left ventricular stroke
work from rest to exercise occurred over 12 months.
Changes in regional left ventricular contraction abnor-
malities. Twenty-two of the 25 patients exhibited regional contraction abnormalities at rest before training. In four patients it was not possible to assess
regional wall motion abnormalities during exercise because of the extensive contraction abnormalities present at rest. Exercise-induced contraction abnormalities
were clearly detectable in 10 patients. Of these, eight
showed improvement in exercise-induced wall motion
disorders after training. In one patient the regional wall
motion disorder became worse, and in another it remained unchanged. The improvements in regional
wall motion abnormalities occurred despite a higher
rate-pressure product (20.65 x 103 + 1.4 x 103 VS
25.1 x 103 + 1.4 x 103, p < .025), a larger LVEDV
(126 ± 17 vs 135 ± 14 ml, p < .05), and a higher
ejection fraction (55 ± 6% vs 61 ± 6%, p < .025)
attained during maximal exercise.
Among the 14 patients who did not undergo training, an exercise-induced regional wall motion disorder
was detected in five initially and in seven patients 12
months later. Of the five patients who had exerciseinduced regional contraction disorders on the initial
355
EHSANI et al.
radionuclide ventriculogram, four showed no change
or deterioration of regional wall motion abnormalities,
and one showed improvement in regional contraction
abnormalities. The lack of improvement or deterioration of regional wall motion disorders was evident
despite similar levels of rate-pressure product (23.55
x 103 + 1.4 x 103vs23.56 x 103 + 1.8 X 103,p =
NS), LVEDV, and ejection fraction at peak exercise
initially and a year later. The changes in regional wall
motion disorders over a 12 month interval were significantly different between the training and nonexercising control groups (X2 = 8.305, p < .005).
Adaptive responses to training in patients with and with-
out detectable myocardial ischemia. Maximal attainable
V02 increased by 35% and 36% after 12 months of
training in patients with and without evidence of myoDownloaded from http://circ.ahajournals.org/ by guest on June 11, 2017
cardial ischemia, respectively. Ejection fraction did
not change significantly in either subgroup with exercise before training (figure 5). After training, ejection
fraction increased significantly during exercise in both
ISCHEMIA
A
rest
exercise
U-
_ 200
*t
62
56.
E
50 z
c-
180
a
i
u)
110 140 170 200
160
O7C
L
-10
SBP (mmHg)
0
a LVESV
(ml)
Initial*--
NO ISCHEMIA
B
Final
D
rest
L56,1,
_62[
200
CD
'~~~~~tt
J
exercise
m-
[E
u-56
0
140 170 200
SBP (mmHg)
110
0-0
T**
180[
'E4- 1640
L
+10
0
A
LVESV
(ml)
FIGURE 5. Adaptive responses to training in patients with and without
detectable exercise-induced myocardial ischemia. A and C, Patients
with exercise-induced ischemia exhibited a large increase in left ventricular ejection fraction (LVEF) at peak exercise after training (*p < .01
rest vs exercise; t p < .005 before vs after training at peak exercise),
despite a higher peak exercise systolic blood pressure (SBP) (p < .005).
The SBP-end-systolic volume (ESV) relationship was shifted upward
and to the left with a decrease in LVESV (**p < .01) and a larger peak
SBP (tp < .005). B and D, Similar changes in LVEF response were
noted in patients without apparent provocable ischemia (*p < .01 rest vs
exercise; tp < .05 before vs after training at peak exercise). However,
the SBP-ESV relationship showed an increase in SBP (**p < .01)
without a change in ESV. Data are mean + SE.
356
subgroups (figure 5). Furthermore, peak exercise ejection fraction was significantly higher in the trained
compared with the untrained states in both subgroups
(figure 5). The higher ejection fraction was attained
despite a significantly higher systolic blood pressure
at peak exercise after training in both subgroups
(figure 5).
In the subgroup with exercise-induced myocardial
ischemia, the systolic blood pressure-LVESV relationship was shifted upward and to the left, with a
smaller LVESV and higher systolic blood pressure,
after 12 months of training. In patients with no apparent exercise-induced myocardial ischemia, the systolic
blood pressure-end-systolic volume relationship was
shifted only upward, with no significant change in
LVESV but a significantly higher systolic blood pressure after training (figure 5). In the former subgroup,
rate-pressure product at peak exercise increased from
19.87 X 103 + 4.26 X 103to23.93 x 103 + 4.41 x
103 (p < .001) after training. However, rate-pressure
product did not change significantly at peak exercise
(23.8 x 103 ± 1.19 x 103 vs 25.50 x 103 + 1.20 x
103) in patients with no apparent myocardial ischemia.
In both subgroups the changes in LVEDV from rest to
exercise were insignificant.
Efect of training on plasma lipids. Training resulted in
a significant weight loss from 80.8 ± 2 to 76.2 ± 2 kg
(p < .001). No significant changes in weight occurred
in the group that did not exercise.
Endurance exercise training had no significant
effect on plasma total cholesterol (214 ± 10 vs 205 ±
9 mg/dl) or plasma triglycerides (174 ± 26 vs 142 ±
11 mg/dl). However, the HDL cholesterol level increased by 13% (39 ± 2 vs 44 ± 2 mg/dl, p < .005),
improving the atherogenic index (total cholesterol to
HDL cholesterol ratio) from 5.7 ± 0.3 to 4.7 ± 0.3 (p
< .001). The LDL cholesterol level was 135 ± 8
mg/dl before and 126 ± 7 mg/dl after training (p =
NS). This increase in the level of HDL cholesterol
without a change in total cholesterol is not surprising
because exercise was the only experimental intervention used.
Discussion
Our results provide evidence that endurance exercise training of progressively increasing intensity can
improve left ventricular contractile function in some
patients with coronary artery disease. This improvement appears to reflect a reduction in the severity of
myocardial ischemia.
A rise in ejection fraction, as seen in healthy subjects during exercise, generally reflects an enhanced
CIRCULATION
THERAPY AND PREVENTION-EXERCISE TRAINING
Downloaded from http://circ.ahajournals.org/ by guest on June 11, 2017
contractile state,3' reduced afterload,34' 5 and/or a
large increase in preload.35 36 Thus, a higher maximal
exercise ejection fraction after training could be the
result of either improved left ventricular contractile
function secondary to a reduction in myocardial ischemia or of favorable changes in cardiac loading conditions. However, it is unlikely that an increase in preload contributed significantly to the higher exercise
ejection fraction in our patients because the changes
from rest to exercise in end-diastolic volume were negligible. Furthermore, a higher end-diastolic volume
would be expected to raise the myocardial 02 requirement37 and potentiate myocardial ischemia, which
could in turn further impair left ventricular function.
Although left ventricular wall stress cannot be measured reliably during exercise with the currently available noninvasive techniques in patients with coronary
artery disease, it is unlikely that afterload was lower
after training because peak exercise systolic blood
pressure and LVEDV were significantly higher and
end-systolic volume did not change. The upward and
leftward shift in the systolic blood pressure-end-systolic volume relationship also suggests improvement
in left ventricular contractile function,3840 even though
peak systolic blood pressure may not reflect end-systolic pressure.
Improvement in left ventricular function in response
to 12 months of training was evident both in patients
with clear-cut evidence of exercise-induced myocardial ischemia and in those with no apparent provocable
myocardial ischemia. However, the extent of the improvement in left ventricular function in response to
training appeared to be more impressive in the patients
with provocable ischemia than in those without it; this
is evidenced by the difference in the systolic blood
pressure-end-systolic volume relationship in the two
subgroups, probably because the subjects in the latter
group had larger myocardial scar than those in the
former group. However, the presence of small areas of
myocardial ischemia cannot be excluded in the subgroup in which ischemia was not detectable with our
methodology. It is therefore likely that the primary
mechanism for improvement in left ventricular function in the majority of our patients was a reduction in
myocardial ischemia.
Spontaneous improvement in left ventricular contractile function over a 12 month interval is unlikely to
account for our findings. Williams et al.4' did not observe any improvement in left ventricular function in a
large number of patients who underwent exercise training of moderate intensity. Furthermore, the patients in
the present study who did not exercise did not show
Vol. 74, No. 2, August 1986
improved left ventricular function 12 months later.
Interventions designed to increase coronary blood
flow, such as coronary artery bypass graft surgery,
improve global and regional left ventricular function
during maximal exercise by decreasing myocardial ischemia,42'43 making it possible to attain a higher myocardial V02. The myocardial 02 requirement is influenced by heart rate, contractile state, and left
ventricular wall tension.34 Left ventricular contractile
function and regional wall motion abnormalities improved in our patients in response to training, despite
attainment during maximal exercise of a higher heart
rate. It is unlikely that left ventricular wall tension was
lower after training because systolic blood pressure
and end-diastolic volume were higher and end-systolic
volume was unchanged at peak exercise. Furthermore,
we have previously reported that patients who have
adapted to the training program used in this study attain a higher concentration of plasma norepinephrine
during maximal exercise,23 which, per se, raises myocardial 02 demand.31 Thus, the improvement in left
ventricular contractile function in our patients was not
likely due to a lower myocardial oxygen requirement,
but rather to an improvement in myocardial oxygenation. The present results extend our previous findings
that intense exercise training can result in volume overload left ventricular hypertrophy in patients with coronary artery disease similar to that seen in healthy subjects.22 This physiologic hypertrophy is characterized
by proportional increases in left ventricular radius and
wall thickness, as reported previously.22
Several previous studies did not show improvement
in left ventricular function9' 10" 41 or myocardial ischemia5' 11 12 in response to exercise training in patients
with coronary artery disease. These negative results
most likely reflect an insufficient training stimulus
rather than differences in the patient populations. The
clinical status of our patients on entry to the study, in
terms of exercise capacity and/or ejection fraction at
rest, was similar to that of the patients in most other
studies.4 5, 9-12, 41, 44 The major obvious difference between our study and those of others is the nature of the
training stimulus. The training intensity used in our
study was high enough to induce a 37% increase in
measured V02max. This was accomplished by progressively increasing the intensity, duration, and frequency of the exercise over the 12 month period rather
than keeping the patients on a maintenance exercise
regimen after 3 months of training.
The results of this study show that in addition to
inducing adaptations that result in a lower heart rate
and systolic blood pressure at the same submaximal
357
EHSANI et al.
work rate, prolonged, high-intensity endurance exercise training can improve left ventricular systolic function during maximal exercise independent of cardiac
loading conditions in some patients with coronary artery disease who can exercise regularly and intensely.
This improvement is likely due to improved oxygenation of some of the underperfused regions of the
myocardium.
References
Downloaded from http://circ.ahajournals.org/ by guest on June 11, 2017
1. Mitchell JH: Exercise training in the treatment of coronary heart
disease. Adv Intern Med 20: 249, 1975
2. Clausen JP: Circulatory adjustment to dynamic exercise and effect
of exercise training in normal subects and patients with ischemic
heart disease. Prog Cardiovasc Dis 18: 459, 1976
3. Varnauskas E, Bergman M, Houk P, Bjorntorp P: Hemodynamic
effect of physical training in coronary patients. Lancet 2: 8, 1966
4. Detry JMR, Rousseau M, Vanderbroucke G, Kusumi F, Brasseur
LA, Bruce RA: Increased arteriovenous oxygen difference after
physical training in coronary heart disease. Circulation 44: 109,
1971
5. Detry JMR, Bruce RA: Effects of physical training on exertional
ST-segment depression in coronary heart disease. Circulation 44:
390, 1971
6. Clausen JP, Larsen OA, Trap-Jensen JT: Physical training in the
management of coronary artery disease. Circulation 40: 143, 1969
7. Galbo H: Hormonal and metabolic adaptations to exercise. New
York, 1983, Thieme-Stratton, pp 2-25
8. Ferguson RJ, Cote P, Gautheir P, Bourasa MG: Changes in exercise coronary sinus blood flow with training in patients with angina
pectoris. Circulation 58: 41, 1978
9. Letac B, Cribier A, Desplanches JF: A study of left ventricular
function in coronary patients before and after physical training.
Circulation 56: 375, 1977
10. Cobb FR, Williams RS, McEwan P, Jones RH, Coleman RE,
Wallace AG: Effects of exercise training on ventricular function in
patients with recent myocardial infarction. Circulation 66: 100,
1982
11. Sim DN, Neill WA: Investigation of the physiological basis for
increased exercise threshold for angina pectoris after physical conditioning. J Clin Invest 54: 763, 1974
12. Nolewajka AJ, Kostuk WL, Rechnitzer PA, Cuningham DA: Exercise and human collaterization: an angiographic and scintigraphic
assessment. Circulation 60: 114, 1979
13. Scheuer JS, Tipton CM: Cardiovascular adaptations to physical
training. Annu Rev Physiol 39: 221, 1977
14. Bersohn M, Scheuer JS: Effect of ischemia on the performance of
hearts from physically trained rats. Am J Physiol 234: H215, 1978
15. Eckstein RW: Effect of exercise and coronary artery narrowing on
coronary collateral circulation. Circ Res 5: 230, 1957
16. Heaton WH, Marr KC, Capurro NL, Goldstein RE, Epstein SE:
Beneficial effect of physical training on blood flow to myocardial
perfused by chronic collaterals in the exercising dog. Circulation
56: 575, 1978
17. McElroy CL, Giesen SA, Fishbein MC: Exercise induced reduction in myocardial infarct size after coronary occlusion in rat.
Circulation 57: 598, 1978
18. Scheel KW, Ingram LA, Wilson JL: Effects of exercise on coronary collateral vasculature of beagles with and without coronary
occlusion. Circ Res 48: 523, 1981
19. Bloor CM, White FC, Sanders TM: Effects of exercise on collateral
development in myocardial ischemia in pigs. J Appl Physiol 56:
656, 1984
20. Ehsani AA, Heath GW, Hagberg JM, Sobel BE, Holloszy JO:
Effects of 12 months of intense exercise training on ischemic ST
segment depression in patients with coronary artery disease. Circulation 64: 1116, 1981
21. Hagberg JM, Ehsani AA, Holloszy JO: Effect of 12 months of
358
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
intense exercise training on stroke volume in patients with coronary
artery disease. Circulation 67: 1194. 1983
Ehsani AA, Martin WH III, Heath CW, Coyle EF: Cardiac effects
of prolonged and intense exercise training in patients with coronary
artery disease. Am J Cardiol 50: 246, 1982
Ehsani AA, Heath GW, Martin WH III, Hagberg JM, Holloszy JO:
Effects of intense exercise training on plasma catecholamines in
coronary patients. J Appl Physiol 57: 154, 1984
Bruce RA: Exercise testing of patients with coronary heart disease.
Ann Clin Res 3: 323, 1971
Issekutz B, Birkhead NC, RodahlK: Use of respiratory quotients in
assessment of aerobic work capacity. J Appl Physiol 17: 47, 1962
Biello DR, Sampathkumaran KS, Geltman ED, Briston WA, Scott
DJ, Grbac RT: Determination of left ventricular ejection fraction:
A new method that requires minimal operator training. J Nucl Med
Techn 9: 77, 1981
Dodge HT, Sandler H, Ballew DW, Lord JD Jr: The use of biplane
angiocardiography for the measurement of left ventricular volume
in man. Am Heart J 60: 762, 1960
Esser PD, Seldin DW, Nichols AB, Anderson PO: Spatial calibration of digital scintigraphic images. Radiology 144: 901, 1982
Ehsani AA, Biello DR, Seals DR, Austin MB, Schultz J: The effect
of left ventricular systolic function on maximal aerobic exercise
capacity in asymptomatic patients with coronary artery disease.
Circulation 70: 552, 1984
Heath GW, Ehsani AA, Hagberg JM, Hinderliter JM, Goldberg
AP: Exercise training improves lipoprotein lipid profiles in patients
with coronary artery disease. Am Heart J 105: 889, 1983
Poliner LR, Dehmer GJ, Lewis SE, Parkey RW, Blomqvist GG,
Willerson JT: Left ventricular performance in healthy subjects: a
comparison of the responses to exercise in the upright and supine
positions. Circulation 62: 528, 1980
Sonnenblick EH, Braunwald E, Williams JF Jr, Glick G: Effects of
exercise on myocardial force-velocity relations on intact unanesthetized man: Relative roles of changes in heart rate, sympathetic
activity, and ventricular dimensions. J Clin Invest 44: 2051, 1985
Okada RD, Boucher CA, Strauss HW, Pohost GM: Exercise radionuclide imaging approaches to coronary artery disease. Am J
Cardiol 46: 1188, 1980
Ross J Jr: Afterload mismatch and preload reserve. A conceptual
framework for analysis of ventricular function. Prog Cardiovasc
Dis 18: 255, 1976
Mitchell JH, Wildenthal K, Mullins CB: Geometrical studies of the
left ventricle utilizing biplane cinefluoroscopy. Fed Proc 28: 1334,
1969
Nixon JR, Murray RG, Leonard PD, Mitchell JH, Blomqvist CG:
Effect of large variations in preload on left ventricular performance
characteristics in normal subjects. Circulation 65: 698, 1982
Braunwald E: Control of myocardial oxygen consumption: Physiological and clinical considerations. Am J Cardiol 27: 416, 1971
Carabello B, Spann J: Usefulness and limitations of the end-systolic index in evaluating cardiac function. Circulation 69: 1058, 1984
Grossman W, Braunwald E, Mann T, McLaurin LP, Green LH:
Contractile state of the left ventricle in man as evaluated from end
systolic pressure volume relations. Circulation 56: 845, 1977
Mahler F, Covell JW, Ross J Jr: Systolic pressure-diameter relations in the normal conscious dog. Cardiovasc Res 9: 477, 1975
Williams RS, McKinnis RA, Cobb FR, Higginbotham MB, Wallace AG, Coleman RE, CaliffRM: Effects of physical conditioning
on left ventricular ejection fraction in patients with coronary artery
disease. Circulation 70: 69, 1984
Kent KM, Borer JS, Green MV, Bacharach SL, McIntosh CL,
Conke DM, Epstein SE: Effects of coronary-artery bypass on global and regional left ventricular function during exercise. N Engl J
Med 298: 1434, 1978
Bussman WD, Mayer V, Kober G, Kaltenbach M: Ventricular
function at rest during leg raising, and physical exercise before and
after aortocoronary bypass surgery. Am J Cardiol 43: 488, 1979
Froelicher V, Jensen D, Genter F, Sullivan M, McKirnan MD,
Witztum K, Scharf J, Strong ML, Ashburm W: A randomized trial
of exercise training in patients with coronary heart disease. JAMA
252: 1291, 1984
CIRCULATION
Improvement of left ventricular contractile function by exercise training in patients with
coronary artery disease.
A A Ehsani, D R Biello, J Schultz, B E Sobel and J O Holloszy
Downloaded from http://circ.ahajournals.org/ by guest on June 11, 2017
Circulation. 1986;74:350-358
doi: 10.1161/01.CIR.74.2.350
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1986 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circ.ahajournals.org/content/74/2/350
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally
published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further information
about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/