Content uploaded by Di Yang
Author content
All content in this area was uploaded by Di Yang on Jun 15, 2016
Content may be subject to copyright.
This article was downloaded by: [112.2.36.22]
On: 07 May 2015, At: 18:36
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK
Disability and Rehabilitation
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/idre20
The effects of different initiation time of exercise
training on left ventricular remodeling and
cardiopulmonary rehabilitation in patients with left
ventricular dysfunction after myocardial infarction
You-Ming Zhangab, Yao Luab, Yun Tangc, Di Yanga, Heng-Fang Wua, Zhi-Ping Bianb, Jin-Dan
Xua, Chun-Rong Gub, Lian-Sheng Wanga & Xiang-Jian Chenb
a Department of Cardiology and
b Research Institute of Cardiovascular Disease, First Affiliated Hospital of Nanjing Medical
University, Nanjing, China, and
c Department of Neonatology, Huai’an First People’s Hospital, Nanjing Medical University,
Huai’an, Jiangsu, China
Published online: 07 May 2015.
To cite this article: You-Ming Zhang, Yao Lu, Yun Tang, Di Yang, Heng-Fang Wu, Zhi-Ping Bian, Jin-Dan Xu, Chun-Rong Gu,
Lian-Sheng Wang & Xiang-Jian Chen (2015): The effects of different initiation time of exercise training on left ventricular
remodeling and cardiopulmonary rehabilitation in patients with left ventricular dysfunction after myocardial infarction,
Disability and Rehabilitation
To link to this article: http://dx.doi.org/10.3109/09638288.2015.1036174
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
http://informahealthcare.com/dre
ISSN 0963-8288 print/ISSN 1464-5165 online
Disabil Rehabil, Early Online: 1–9
!2015 Informa UK Ltd. DOI: 10.3109/09638288.2015.1036174
REVIEW PAPER
The effects of different initiation time of exercise training on left
ventricular remodeling and cardiopulmonary rehabilitation in patients
with left ventricular dysfunction after myocardial infarction
You-Ming Zhang
1,2
, Yao Lu
1,2
, Yun Tang
3
, Di Yang
1
, Heng-Fang Wu
1
, Zhi-Ping Bian
2
, Jin-Dan Xu
1
, Chun-Rong Gu
2
,
Lian-Sheng Wang
1
, and Xiang-Jian Chen
2
1
Department of Cardiology and
2
Research Institute of Cardiovascular Disease, First Affiliated Hospital of Nanjing Medical University, Nanjing, China,
and
3
Department of Neonatology, Huai’an First People’s Hospital, Nanjing Medical University, Huai’an, Jiangsu, China
Abstract
Purpose: The purpose of this study was to determine whether different initiation of exercise
training (ET) produces different effect sizes for left ventricular (LV) remodeling and
cardiopulmonary rehabilitation in patients with LV dysfunction after myocardial infarction
(MI). Method: Trials evaluating ET outcomes identified by searches in OVID MEDLINE, EMBASE,
PubMed and WEB OF SCIENCE were used. Meta-analysis was conducted with the use of the
software STATA 11.0. The results were expressed as the standardized mean difference (SMD),
with corresponding 95% CI and pvalue. Results: The largest changes in LV remodeling and
cardiopulmonary capacity rehabilitation were obtained when programs began the acute phase
after MI. With the healing of MI, the beneficial effects of ET on LV ejection fraction (LVEF), LV
end-systolic diameter (LVDs) and peak VO
2
were gradually weakened even worse. The
incidence of major adverse cardiac events was not significantly increased in acute phase post-
MI. Sensitivity analyses show that ET still had significant effect in reducing LVDs and increasing
peak VO
2
, while ET no longer had statistical effect in increasing LVEF but showed favorable
trends when the same research institution’s works were excluded. Conclusions: ET has favorable
effects on LV remodeling and cardiopulmonary rehabilitation in LV dysfunction post-MI
patients. The greatest benefits are obtained when ET starts at the acute phase following MI.
äImplications for Rehabilitation
Early exercise training is safe and feasible in acute and healing phase after myocardial
infarction.
Early exercise training could attenuate LV remodeling and improve cardiopulmonary capacity
in patients with myocardial infarction after hospital discharge (around one week post-MI).
Exercise training has favorable effects on LV remodeling and cardiopulmonary capacity
rehabilitation. Exercise training should be treated to have the same roles with drugs in
secondary prevention of myocardial infarction.
Keywords
Cardiopulmonary capacity, exercise, left
ventricular remodeling, meta-analysis,
post-myocardial infarction
History
Received 1 November 2014
Revised 24 March 2015
Accepted 27 March 2015
Published online 17 April 2015
Introduction
Exercise training (ET) in patients with acute myocardial infarction
(AMI) has several beneficial effects on morbidity and mortality,
functional capacity, cardiac symptoms and quality of life [1].
Jugdutt et al. [2] early reported that ET might be associated with
left ventricular (LV) remodeling in patients with myocardial
infarction (MI). Post-infarction LV remodeling is a complex
process characterized by continuing long after infarction healing,
LV size and shape, leading to further LV dysfunction and
heart failure, representing an important predictor of long-term
mortality [3]. However, it is not clear how ET affects LV
remodeling.
Some previous trials [2,4] have reported that ET on patients
with recent MI may deteriorate myocardial remolding while other
studies [5–7] have shown that ET may attenuate the progression of
LV remodeling in patients with recent AMI. In addition, other
studies [8–10] identified that post-MI ET may not alter LV
remodeling. In these studies, however, initiation of following MI
ET was different and LV ejection fraction (LVEF) was 32–50%.
Confirming these inconsistencies is necessary because the
findings could be used to enhance the benefits of exercise after
Address for correspondence: Xiang-Jian Chen, Research Institute of
Cardiovascular Disease, First Affiliated Hospital of Nanjing Medical
University, 300 Guangzhou Road, Nanjing 210029, China. Tel: +86 25
6813 6721. Fax: +86 25 8373 8572. E-mail: chenxiangjian@njmu.edu.cn
Downloaded by [112.2.36.22] at 18:36 07 May 2015
MI. Peak oxygen uptake (VO
2
) is considered as the most
accurate method for assessing cardiopulmonary capacity [9,11].
Therefore, we designed a systematic review and meta-analysis
to determine whether the different initiation of ET produces
different effect sizes for LV remodeling, cardiopulmonary
rehabilitation and adverse events in patients with LV dysfunction
after MI.
Methods
Data sources
A systematic literature search up to April 2014 was performed
in OVID MEDLINE (1950–2014), EMBASE (1988–2014),
PubMed (1966–2014) and WEB OF SCIENCE (1950–2014)
to identify relevant studies. We used search terms ‘‘myocardial
infarction’’, ‘‘acute myocardial infarction’’, ‘‘post-myocardial
infarction’’, ‘‘heart attack’’, ‘‘ventricular remodeling’’, ‘‘left
ventricular function’’, ‘‘ejection fraction’’, ‘‘left ventricular end-
systolic diameter’’, combined with ‘‘exercise’’, ‘‘exercise train-
ing’’ and ‘‘exercise therapy’’, in the full-text option. Two authors
conducted all searches independently.
Outcomes measures
Some studies [12,13] have shown that LVEF and LV end-systolic
diameter (LVDs) were important predictors of LV remodeling.
Therefore, the following data were collected: country, age, gender,
MI location, drug treatment, LVEF, LVDs, peak VO
2
, training
frequency, intensity, duration per session, length of the program,
withdraw and adverse events.
Study selection
Two reviewers independently read the titles/abstracts of all
published sources to identify studies reporting the effect of ET
on LVEF and/or LVDs in recent post-MI patients. Studies were
excluded if they were not randomized, evaluated single modality
interventions except exercise, did not have usual care group, the
mean LVEF of the subjects was more than 50% or less than 30%,
they tested interventions delivered to inpatients, or the interven-
tions were not provided by health professionals, and non-English
and non-human studies.
Data extraction and quality assessment
Two reviewers independently extracted relevant data from the
studies. The third author checked the data and resolved any
discrepancies by consensus. We assessed study quality using the
slightly modified Physiotherapy Evidence Database (PEDRO)
scale [14]. Due to subject blind is unlikely in exercise science
studies, the maximum possible PEDRO score was 9 [15].
Data synthesis and analysis
In our review, all statistical analyses were conducted with
software STATA 11.0 (Stata Corporation, College Station, TX).
Continuous data were expressed with mean ± SD. We assessed
treatment effects based on pooled data from individual studies.
The statistical data included the number of participants in the ET
and usual care control groups, the mean change from baseline and
the SD of the mean change [16]. In the majority of studies, only
the baseline and post-ET means and SDs were reported. We tried
to contact corresponding author of related studies to obtain
missing data necessary for meta-analysis, but rarely got reply of
the author. Accordingly, missing SDs for change scores were
calculated using baseline and post-ET means and SDs, with a
conservative correlation coefficient of 0.5 [17,18]. In this
situations, sensitivity analyses were carried out with different
values of the correlation coefficient (r¼0.25 and 0.80), to
determine whether the overall result of the analysis is robust to the
use of imputed correlation coefficients [19]. In meta-analysis,
statistical heterogeneity was assessed by the chi-square test
(p50.05) and I
2
. Given I
2
550%, p40.05, the fixed effects model
would be adopted. Otherwise, we applied the random effects
model. The results were expressed as the standardized mean
difference (SMD), with corresponding 95% CI and pvalue.
According to Jacob Cohen, effect size of 0.2 is considered as a
small effect, those of 0.5 as a moderate effect and from 0.8 is a
large effect [20]. Subgroup analyses were performed to try to
explore statistical heterogeneity between studies. The Begg’
funnel plot [21] and Egger test [22] were used to examine the
publication bias. A significant correlation (p50.05) suggests the
presence of publication bias.
Results
Study selection
The details of literature search are shown in Figure 1. Among the
2117 articles initially identified through the electronic database
searches, 41 were fully retrieved for more detailed evaluation, 19
of them were excluded, due to incomplete EF data (n¼5), EF
more than 50% or less than 30% (n¼9), ineligible study design
(n¼2), initiation of ET was unreported (n¼3).
Studies included in the systematic review
Supplementary Table 1 [2,5–10,23–36,50] presents summary data
from the 22 unique randomized controlled trials (RCTs) that were
eligible for our systematic review. The selected studies included
a total of 1137 participants, who were characterized by
younger age (mean age: 58 years) and predominantly male
gender (1042/1122). Studies incorporated ET at 60–85% heart
rate reserve or 70–90% peak heat rate or 55–85% peak VO
2
for
20–60 min per session. The intensity exercise was categorized as
vigorous, according to a recent position statement by Exercise and
Sports Science Australia [37].
Assessment of study quality
Owing to the traditionally used tools JADAD [38] are relatively
insensitive in assessing study quality of exercise interventions, we
chose the PEDRO scale which designed by physical therapists
is probably the tool most closely related to the needs of
Figure 1. Flow diagram of studies through the different phases of the
review.
2Y.-M. Zhang et al. Disabil Rehabil, Early Online: 1–9
Downloaded by [112.2.36.22] at 18:36 07 May 2015
physical training scientists [15]. For detailed quality assessments
see Table 1. No study described allocation concealed. The median
PEDRO score was 7 for all studies.
Total estimate
The effects of ET on LVEF and LVDs
The result of meta-analysis showed that, compared with the
control group, ET had a small, significant effect in increasing
LVEF for post-MI [SMD ¼0.29, 95% CI (0.09, 0.49), p¼0.005]
and moderate, significant effect in reducing LVDs for post-MI
[SMD ¼0.53, 95% CI (0.89, 0.16), p¼0.005]. Measures of
heterogeneity (the I
2
statistic) were 62.7%, 83.9% and p50.001,
respectively.
Therefore, there was significant heterogeneity in the initial
meta-analysis of EF and LVDs. To explore the heterogeneity, we
performed a subgroup analysis according to the initiation of ET
post-MI: as evolving (56 h), acute (6 h–7 days), healing (7–28
days) and healed (29 days and beyond) [39].
ET and LVEF at three different phases post-MI
There were seven articles in which the time post-MI to initiation of
ET was considered as acute period. The result of meta-analysis
showed that, compared with the control group, ET had a moderate,
significant effect in increasing LVEF for post-MI at acute phase
[SMD ¼0.60, 95% CI (0.27, 0.93), p50.001] (Figure 2A).
There were eight articles in which the time post-MI to
initiation of ET was considered as healing period. The result of
our meta-analysis showed that, compared with the control group,
ET had a small, significant effect in increasing LVEF for post-MI
at healing phase [SMD ¼0.36, 95% CI (0.14, 0.58), p¼0.02]
(Figure 2B).
There were eight articles in which the time post-MI to
initiation of ET was considered as healed period. The result
of meta-analysis showed that, compared with the control group,
ET had no statistical effect in increasing LVEF for post-MI at
healed phase [SMD ¼0.11, 95% CI (0.32, 0.10), p¼0.325]
(Figure 2C).
ET and LVDs at three different phases post-MI
There were six articles in which the time post-MI to initiation of
ET was considered as acute period. The result of meta-analysis
showed that, compared with the control group, ET had a great,
significant effect in reducing LVDs for post-MI at acute phase
[SMD ¼1.03, 95% CI (1.83, 0.22), p¼0.012] (Figure 3A).
There were five articles in which the time post-MI to initiation
of ET was considered as the healing period. The result of meta-
analysis showed that, compared with the control group, ET had a
small, significant effect in reducing LVDs for post-MI at healing
phase [SMD ¼0.24, 95% CI (0.47, 0.007), p¼0.043]
(Figure 3B).
There were two articles in which the time post-MI to initiation
of ET was considered as healed period. The result of meta-
analysis showed that, compared with the control group, ET
had no statistical effect in reducing LVDs for post-MI at
healed phase [SMD ¼0.13, 95% CI (0.46, 0.21), p¼0.469]
(Figure 3C).
ET and peak VO
2
at three different phases post-MI
There were seven articles in which the time post-MI to initiation
of ET was considered as acute period. The result of meta-analysis
showed that, compared with the control group, ET had a great,
significant effect in increasing peak VO
2
for post-MI at acute
phase [SMD ¼1.52, 95% CI (0.88, 2.16), p50.001] (Figure 4A).
There were five articles in which the time post-MI to initiation
of ET was considered as the healing period. The result of meta-
analysis showed that, compared with the control group, ET had a
great, significant effect in increasing peak VO
2
for post-MI at
healing phase [SMD ¼1.14, 95% CI (0.15, 2.21), p¼0.024]
(Figure 4B).
There were three articles in which the time post-MI to
initiation of ET was considered as healed period. The result of
meta-analysis showed that, compared with the control group, ET
had a moderate, significant effect in increasing peak VO
2
for post-
MI at healed phase [SMD ¼0.62, 95% CI (0.18, 1.06), p¼0.006]
(Figure 4C).
Table 1. Assessment of study quality.
Study/year A B C D E F G H I Overall pedro
Giallauria et al. (2013) Y YNYYYNYY 7
Giallauria et al. (2012) Y YNYYYNYY 7
Giallauria et al. (2011) Y YNYYYNYY 7
Chung et al. (2010) Y YNYYYNYY 7
Brehm et al. (2009) Y YNYYYNYY 7
Giallauria et al. (2009) Y YNYYYNYY 7
Zheng et al. (2008) Y YNYYYNYY 7
Lee et al. (2008) Y YNYYYNYY 7
Giallauria et al. (2008) Y YNYYYNYY 7
Giallauria et al. (2006) Y YNYYYNYY 7
Giallauria et al. (2006) Y YNYYYNYY 7
Mimura et al. (2005) Y YNYYYNYY 7
Yu et al. (2004) Y YNYYYNYY 7
Kubo et al. (2004) Y YNYYYYYY 8
Koizumi et al. (2003) Y YNYYYNYY 7
Heldal et al. (2000) Y YNYYYNYY 7
Dubach et al. (1997) Y YNYYYNYY 7
Giannuzzi et al. (1997) Y YNYYYNYY 7
Giannuzzi et al. (1993) Y YNYYYNYY 7
Jette et al. (1991) Y YNYYYNYY 7
Jugdutt et al. (1988) Y YNYYYNYY 7
Grodzinski et al. (1987) Y YNYYYNYY 7
A: eligibility criteria specified; B: random allocation of participants; C: allocation concealed; D: groups similar at baseline; E: assessors blinded; F:
outcome measures assessed in 85% of participants; G: intention to treat analysis; H: reporting of between group statistical comparison; I: point
measures and measures of variability reported; N: no; Y: yes.
DOI: 10.3109/09638288.2015.1036174 Effects of exercise on cardiac rehabilitation 3
Downloaded by [112.2.36.22] at 18:36 07 May 2015
NOTE: Weights are from random effects analysis
Overall (I−squared = 64.7%, p = 0.009)
Giallauria (2013)
Chung (2010)
ID
Giallauria (2008)
Giallauria, F (2012)
Giallauria, F (2011)
Zheng, H (2008)
Giallauria (2009)
Study
(A)
0.60 (0.27, 0.93)
0.67 (0.07, 1.26)
0.05 (−0.37, 0.47)
SMD (95% CI )
0.54 (0.03, 1.05)
0.11 (−0.44, 0.67)
1.05 (0.56, 1.53)
0.70 (0.18, 1.22)
1.16 (0.61, 1.70)
100.00
12.80
16.28
Weight
14.43
13.57
14.98
14.22
13.71
%
0−1.7 1.7
Overall (I−squared = 43.8%, p = 0.087)
ID
Kubo (2004)
Koizumi (2003)
Giallauria (2006)
Study
Giallauria (2006)
Giannuzzi (1997)
Mimura, J (2005)
Brehm, M (2009)
Yu, C. M (2004)
0.36 (0.14, 0.58)
SMD (95% CI )
−0.22 (−0.81, 0.38)
0.60 (−0.15, 1.35)
0.48 (−0.12, 1.08)
0.04 (−0.58, 0.66)
0.90 (0.43, 1.37)
−0.15 (−0.86, 0.57)
0.34 (−0.36, 1.03)
0.50 (−0.24, 1.24)
100.00
Weight
13.84
8.82
13.64
%
12.76
22.24
9.55
10.22
8.94
−1.37 0 1.37
Overall (I−squared = 0.0%, p = 0.632)
Dubach (1997)
Jugdutt, B (1988)
ID
Heldal (2000)
Giannuzzi (1993)
Grodzinski (1987)
Study
Lee, B. C (2008)
Jette (1991)
−0.11 (−0.32, 0.10)
−0.10 (−0.89, 0.68)
−0.31 (−0.99, 0.36)
SMD (95% CI)
0.12 (−0.53, 0.77)
0.07 (−0.33, 0.47)
−0.30 (−0.70, 0.10)
0.16 (−0.47, 0.79)
−0.58 (−1.46, 0.29)
100.00
7.20
9.63
Weight
10.66
27.39
28.13
%
11.22
5.78
−1.46 0 1.46
(B)
(C)
Figure 2. Forest plot of effects of ET on LVEF in patients with LV dysfunction after AMI including subgroup analysis.
4Y.-M. Zhang et al. Disabil Rehabil, Early Online: 1–9
Downloaded by [112.2.36.22] at 18:36 07 May 2015
NOTE: Weights are from random effects analysis
Overall (I−squared = 92.2%,
p = 0.000)
Giallauria (2008)
Giallauria, F (2012)
Chung (2010)
ID
Giallauria, F (2011)
Giallauria (2013)
Study
(A)
Giallauria (2009)
−1.03 (−1.83, −0.22)
−1.66 (−2.24, −1.08)
−0.10 (−0.66, 0.45)
−0.13 (−0.56, 0.29)
SMD (95% CI)
−2.70 (−3.33, −2.07)
−0.32 (−0.90, 0.27)
−1.31 (−1.87, −0.75)
100.00
16.54
16.68
17.27
Weight
16.31
16.54
%
16.66
0−3.33 3.33
Overall (I−squared = 0.0%, p = 0.945)
Mimura, J (2005)
Koizumi (2003)
ID
Giallauria (2006)
Yu, C. M (2004)
Giannuzzi (1997)
Study
Giallauria (2006)
Kubo (2004)
−0.24 (−0.47, −0.01)
0.05 (−0.66, 0.77)
−0.24 (−0.97, 0.50)
SMD (95% CI)
−0.27 (−0.86, 0.33)
−0.40 (−1.14, 0.34)
−0.41 (−0.86, 0.04)
−0.07 (−0.69, 0.55)
−0.16 (−0.76, 0.43)
100.00
10.36
9.93
Weight
15.06
9.79
26.04
%
13.81
15.02
0−1.14 1.14
Overall (I−squared = 0.0%,
p = 0.642)
ID
Study
Lee, B. C (2008)
Giannuzzi (1993)
−0.13 (−0.46, 0.21)
SMD (95% CI)
0.00 (−0.63, 0.63)
−0.18 (−0.58, 0.23)
100.00
Weight
%
29.19
70.81
0−.628 .628
(B)
(C)
Figure 3. Forest plot of effects of ET on LVDs in patients with LV dysfunction after AMI including subgroup analysis.
DOI: 10.3109/09638288.2015.1036174 Effects of exercise on cardiac rehabilitation 5
Downloaded by [112.2.36.22] at 18:36 07 May 2015
NOTE: Weights are from random effects analysis
Overall (I−squared = 88.4%, p = 0.000)
Giallauria, F (2011)
Zheng, H (2008)
ID
Giallauria, F (2012)
Chung (2010)
Giallauria (2008)
Study
(A)
Giallauria (2013)
Giallauria (2009)
1.52 (0.88, 2.16)
2.42 (1.82, 3.02)
1.47 (0.90, 2.05)
SMD (95% CI )
0.79 (0.21, 1.36)
0.59 (0.16, 1.02)
2.31 (1.66, 2.97)
0.70 (0.10, 1.30)
2.47 (1.79, 3.15)
100.00
14.23
14.38
Weight
14.36
15.15
13.90
%
14.23
13.75
0−3.15 3.15
NOTE: Weights are from random effects analysis
Overall (I−squared = 89.4%, p = 0.000)
Giallauria (2006)
Kubo (2004)
Study
Giallauria (2006)
Brehm, M (2009)
Mimura, J (2005)
ID
1.14 (0.15, 2.12)
2.66 (1.80, 3.53)
0.69 (0.08, 1.30)
2.12 (1.38, 2.87)
−0.21 (−0.90, 0.48)
0.51 (−0.22, 1.24)
SMD (95% CI)
100.00
19.17
20.67
%
19.91
20.23
20.01
Weight
−3.53 0 3.53
Overall (I−squared = 0.0%, p = 0.383)
Dubach (1997)
Study
Jette (1991)
Lee, B. C (2008)
ID
0.62 (0.18, 1.06)
0.81 (−0.01, 1.63)
0.10 (−0.76, 0.95)
0.80 (0.14, 1.45)
SMD (95% CI)
100.00
28.71
%
26.22
45.07
Weight
−1.63 0 1.63
(B)
(C)
Figure 4. Forest plot of effects of ET on peak VO
2
in patients with LV dysfunction after AMI including subgroup analysis.
6Y.-M. Zhang et al. Disabil Rehabil, Early Online: 1–9
Downloaded by [112.2.36.22] at 18:36 07 May 2015
Adverse events
All studies reported withdrawal and adverse events (deaths,
hospitalization and cardiac events) (Supplementary Table 2). In
the acute phase studies, there were three cardiac events (one
patient experienced re-infarction and two patients underwent
repeated revascularization) in the ET group and four cardiac events
(two patients experienced re-infarction and two patients underwent
repeated revascularization) in the control group. In the healing
phase studies, two patients were hospitalized from the study (one
for congestive heart failure in the ET group and one for unstable
angina requiring coronary artery bypass graft surgery in the
control), and one patient in the control group died from refractory
heart failure. In the healed phase studies, six of 11 patients (one in
the training group and five in the control group) were withdrawn
from the study because of unstable angina requiring coronary
artery bypass graft surgery, and two patients died (one from sudden
death in the ET group and one from cancer in the control group).
No adverse events took place during any of the training
sessions in the ET group patients. The reasons of subjects
withdrawal have nothing to do with ET. The incidence of major
adverse cardiac events was not significantly increased in the acute
phase post-MI.
Sensitivity analysis
The findings may be influenced by seven trials which came from
the same research institution of Italy. Therefore, analyses were
repeated excluding these trials. After this removal, ET still had
small significant effect in reducing LVDs [SMD ¼0.21, 95% CI
(0.39, 0.03), p¼0.025] and moderate significant effect in
increasing peak VO
2
[SMD ¼0.64, 95% CI (0.33, 0.94),
p50.001] for post-MI. ET no longer had statistical effect in
increasing LVEF [SMD ¼0.17, 95% CI (0.04, 0.38), p¼0.121]
but still showed favorable trends when these studies were
excluded. Statistical analyses depended on extrapolated data.
Therefore, we performed sensitivity analyses that changed the
correlation coefficient, which did not result in any significant
variation in overall results.
Publication bias
ET and LVEF
Publication bias was detected by drawing Begg’s funnel plot
[Z¼0.96 (continuity corrected), Pr4jZj¼0.338 (continuity cor-
rected)] and diagram (Egger test; t¼0.37, 95% CI (4.102,
2.872), p¼0.717) in the meta-analysis (Supplementary Figure
1A). There was no indication of publication bias.
ET and LVDs
Publication bias was detected by drawing Begg’s funnel plot
[Z¼0.79 (continuity corrected), Pr4jZj¼0.428 (continuity cor-
rected)] and diagram (Egger test; t¼0.66, 95% CI (9.986,
5.325), p¼0.522) in the meta-analysis (Supplementary Figure
1B). There was no indication of publication bias.
ET and peak VO
2
Publication bias was detected by drawing Begg’s funnel plot
[Z¼0.20 (continuity corrected), Pr4jZj¼0.843 (continuity cor-
rected)] and diagram (Egger test; t¼0.72, 95% CI (5.618,
11.227), p¼0.485) in the meta-analysis (Supplementary Figure
1C). There was no indication of publication bias.
Discussion
Overall, the present study provides a review of RCTs detecting the
effects of ET on LV remodeling after post-MI. The effect sizes of
changes were dependent on time of instigation of the exercise
intervention. The largest changes in LV remodeling and cardio-
pulmonary capacity rehabilitation were obtained when programs
began the acute phase after MI. With the healing of MI, the
beneficial effects of ET on LVEF, LVDs and peak VO
2
were
gradually weakened even worse.
The pathology change after MI
During the acute phase after MI, necrotic tissue is absorbed, and
infracted region becomes thin and elongated. Afterwards, con-
nective tissue cells enter the myocyte compartment and connect
disrupted myocyte fibers, providing resistance to further stretch-
ing [40]. Finally, fibroblast proliferation, collagen deposition and
scar formation were founded at the infracted area. Ventricular
expansion occurs in the acute and healing phases after MI before
extensive fibroblast proliferation and collagen deposition. During
this phase, the necrotic area is particularly susceptible to
distorting forces. In the healed phase, the scar is relatively non-
expandable and is much more resistant to further deformation
[40]. Therefore, ET started early after AMI may not be safe and
may cause deterioration in cardiac functioning. However, the
present systematic review has demonstrated that ET starting acute
and healing phases after MI significantly improves LV remodel-
ing, and major adverse cardiac events do not significantly increase
in clinically stable post-MI patients.
The putative mechanisms of ET on LV remodeling
Regarding the mechanism underlying the effect of ET on LV
remodeling, there is currently no specific analysis. The following
findings are biologically plausible. First, the LV wall stress
induced by ET may have only a slight impact on the development
of LV remodeling [41]. Previous study shows that LV remodeling
developed in patients with low LVEF is independent of ET [35].
Second, the appropriate exercise intensity may not aggravate LV
remodeling. Previous experimental studies showed that high
exercise intensity aggravated LV remodeling, while moderate
exercise intensity did not deteriorate LV remodeling in post-MI
rats [42,43]. Third, the other potential mechanisms are that
exercise may have even a remodeling attenuating effect. Melo
et al. reported that ET may attenuate LV remodeling by inducing
up-regulation of the expression of miRNA-29, which was related
to a significant decrease in gene expression of collagen [44].
Brehm et al. demonstrated that ET may attenuate LV remodeling
by increasing endothelial progenitor cells (EPCs) that may
participate in angiogenesis of the damaged tissue area after
AMI [23]. ET increases parasympathetic tone and restores arterial
baroreflex sensitivity (BRS) [45] and sympathetic nerve activity,
which improves LVEF and heart rate variability [46]. Post-ET
decreases plasma pro-NT-BNP levels, reduces LVDs and improve
in early LV diastolic filling [27] and increases shear stress that
stimulates endothelial NO release, which improves vasodilation
and vascular endothelial function [47]. Giallauria et al. reported
that ET decreases plasma high-mobility group box-1 (HMGB1)
levels, which were associated with improvements of cardiopul-
monary and autonomic function, and with favorable LV remodel-
ing [6]. Therefore, in order to more accurately elucidate the effect
of ET on LV remodeling process in patients with LV dysfunction
after MI, further study is still necessary.
Clinical implications
Recent guidelines on physical activity also do not provide any
recommendations on the initiation of ET program after AMI.
Most rehabilitation post-MI programs start at least four to six
weeks after hospital discharge [48]. In this review, early ET could
DOI: 10.3109/09638288.2015.1036174 Effects of exercise on cardiac rehabilitation 7
Downloaded by [112.2.36.22] at 18:36 07 May 2015
attenuate LV remodeling and improve cardiopulmonary capacity
in patients with MI after hospital discharge (around one week
post-MI). Meanwhile, the incidence of major adverse cardiac
events was not significantly raised in the acute and healing phases
compared with the healed phase. Consistent with previous
research results, exercise stress testing is safe and feasible in the
majority of patients with 3 days after MI [49]. It is a pity, in
secondary prevention after MI, ET is not as attention as drug
therapy. The results of this review are likely to provide existing
research evidence and could help to enhance the rehabilitation
process after MI.
Study limitations
Several potential limitations of our study deserve mention. First,
patients included in those trials were characterized by younger
age and predominantly male gender. Therefore, our findings may
not be extrapolated to elderly patients and females. Second, all
studies incorporated intensity exercise were categorized as
vigorous intensity exercise. Whether the high intensity, moderate
intensity and low intensity exercise and sedentary lifestyles
produces different effect sizes for change in LVEF or LVDs
compared with the vigorous intensity exercise in patients with LV
dysfunction (LVEF 30–50%) after AMI. Third, the heterogeneity
of these studies may come from other factors, such as nationality,
gender, location of MI. Therefore, future studies are necessary to
assess the effects of these factors on LV remodeling after MI.
Conclusions
ET has favorable effects on LV remodeling and cardiopulmonary
capacity rehabilitation in LV dysfunction post-MI patients. The
greatest benefits are obtained when ET starts at the acute phase
following MI.
Declaration of interest
No commercial party having a direct financial interest in the
results of the research supporting this article has or will confer a
benefit on the authors or on any organization with which the
authors are associated. This work has been funded by the National
Natural Science Foundation of China (Nos. 81170220 and
81100156), Jiangsu Province Health Foundation (RC2011075)
and Priority Academic Program Development of Jiangsu Higher
Education Institutions.
References
1. Taylor RS, Brown A, Ebrahim S, et al. Exercise-based rehabilitation
for patients with coronary heart disease: systematic review and
meta-analysis of randomized controlled trials. Am J Med 2004;116:
682–92.
2. Jugdutt BI, Michorowski BL, Kappagoda CT. Exercise training after
anterior Q wave myocardial infarction: importance of regional left
ventricular function and topography. J Am Coll Cardiol 1988;12:
362–72.
3. Cohn JN, Ferrari R, Sharpe N. Cardiac remodeling – concepts and
clinical implications: a consensus paper from an international forum
on cardiac remodeling. Behalf of an International Forum on Cardiac
Remodeling. J Am Coll Cardiol 2000;35:569–82.
4. Kubo N, Ohmura N, Nakada I, et al. Exercise at ventilatory
threshold aggravates left ventricular remodeling in patients with
extensive anterior acute myocardial infarction. Am Heart J 2004;
147:113–20.
5. Giallauria F, Acampa W, Ricci F, et al. Exercise training early after
acute myocardial infarction reduces stress-induced hypoperfusion
and improves left ventricular function. Eur J Nucl Med Mol Imaging
2013;40:315–24.
6. Giallauria F, Cirillo P, D’Agostino M, et al. Effects of exercise
training on high-mobility group box-1 levels after acute myocardial
infarction. J Card Fail 2011;17:108–14.
7. Giannuzzi P, Temporelli PL, Corra U, et al. Attenuation of
unfavorable remodeling by exercise training in postinfarction
patients with left ventricular dysfunction: results of the Exercise in
Left Ventricular Dysfunction (ELVD) trial. Circulation 1997;96:
1790–7.
8. Giallauria F, Acampa W, Ricci F, et al. Effects of exercise training
started within 2 weeks after acute myocardial infarction on
myocardial perfusion and left ventricular function: a gated SPECT
imaging study. Eur J Prev Cardiol 2012;19:1410–19.
9. Chung CC, Huang WC, Chiou KR, et al. Ratio of early mitral inflow
peak velocity to flow propagation velocity predicts training effects
of cardiac rehabilitation in patients after acute myocardial infarction.
J Rehabil Med 2010;42:232–8.
10. Giallauria F, Lucci R, De Lorenzo A, et al. Favourable effects of
exercise training on N-terminal pro-brain natriuretic peptide plasma
levels in elderly patients after acute myocardial infarction. Age
Ageing 2006;35:601–7.
11. Bassett DJ, Howley ET. Maximal oxygen uptake: ‘‘classical’’ versus
‘‘contemporary’’ viewpoints. Med Sci Sports Exerc 1997;29:
591–603.
12. Matsumura T, Ohtaki E, Tanaka K, et al. Echocardiographic
prediction of left ventricular dysfunction after mitral valve repair
for mitral regurgitation as an indicator to decide the optimal timing
of repair. J Am Coll Cardiol 2003;42:458–63.
13. Enriquez-Sarano M, Tajik AJ, Schaff HV, et al. Echocardiographic
prediction of left ventricular function after correction of mitral
regurgitation: results and clinical implications. J Am Coll Cardiol
1994;24:1536–43.
14. Maher CG, Sherrington C, Herbert RD, et al. Reliability of the
PEDro scale for rating quality of randomized controlled trials. Phys
Ther 2003;83:713–21.
15. Ismail H, McFarlane JR, Nojoumian AH, et al. Clinical outcomes
and cardiovascular responses to different exercise training intensities
in patients with heart failure: a systematic review and meta-analysis.
JACC Heart Fail 2013;1:514–22.
16. Kelly ME, Loughrey D, Lawlor BA, et al. The impact of exercise on
the cognitive functioning of healthy older adults: a systematic
review and meta-analysis. Ageing Res Rev 2014;16C:12–31.
17. Bacon AP, Carter RE, Ogle EA, Joyner MJ. VO2max trainability
and high intensity interval training in humans: a meta-analysis.
PLoS One 2013;8:e73182.
18. Follmann D, Elliott P, Suh I, Cutler J. Variance imputation for
overviews of clinical trials with continuous response. J Clin
Epidemiol 1992;45:769–73.
19. Filson CP, Hollingsworth JM, Clemens JQ, Wei JT. The efficacy and
safety of combined therapy with alpha-blockers and anticholinergics
for men with benign prostatic hyperplasia: a meta-analysis. J Urol
2013;190:2153–60.
20. Du S, Dong J, Zhang H, et al. Taichi exercise for self-rated sleep
quality in older people: a systematic review and meta-analysis. Int J
Nurs Stud 2015;52:368–79.
21. Begg CB, Mazumdar M. Operating characteristics of a rank
correlation test for publication bias. Biometrics 1994;50:1088–101.
22. Egger M, Davey SG, Schneider M, Minder C. Bias in meta-analysis
detected by a simple, graphical test. BMJ 1997;315:629–34.
23. Brehm M, Picard F, Ebner P, et al. Effects of exercise training on
mobilization and functional activity of blood-derived progenitor
cells in patients with acute myocardial infarction. Eur J Med Res
2009;14:393–405.
24. Giallauria F, Galizia G, Lucci R, et al. Favourable effects of
exercise-based Cardiac Rehabilitation after acute myocardial infarc-
tion on left atrial remodeling. Int J Cardiol 2009;136:300–6.
25. Zheng H, Luo M, Shen Y, et al. Effects of 6 months exercise training
on ventricular remodelling and autonomic tone in patients with acute
myocardial infarction and percutaneous coronary intervention. J
Rehabil Med 2008;40:776–9.
26. Lee BC, Chen SY, Hsu HC, et al. Effect of cardiac rehabilitation on
myocardial perfusion reserve in postinfarction patients. Am J
Cardiol 2008;101:1395–402.
27. Giallauria F, Cirillo P, Lucci R, et al. Left ventricular remodelling in
patients with moderate systolic dysfunction after myocardial
infarction: favourable effects of exercise training and predictive
8Y.-M. Zhang et al. Disabil Rehabil, Early Online: 1–9
Downloaded by [112.2.36.22] at 18:36 07 May 2015
role of N-terminal pro-brain natriuretic peptide. Eur J Cardiovasc
Prev Rehabil 2008;15:113–18.
28. Giallauria F, De Lorenzo A, Pilerci F, et al. Reduction of N
terminal-pro-brain (B-type) natriuretic peptide levels with exercise-
based cardiac rehabilitation in patients with left ventricular
dysfunction after myocardial infarction. Eur J Cardiovasc Prev
Rehabil 2006;13:625–32.
29. Mimura J, Yuasa F, Yuyama R, et al. The effect of residential
exercise training on baroreflex control of heart rate and sympathetic
nerve activity in patients with acute myocardial infarction. Chest
2005;127:1108–15.
30. Yu CM, Li LS, Lam MF, et al. Effect of a cardiac rehabilitation
program on left ventricular diastolic function and its relationship to
exercise capacity in patients with coronary heart disease: experience
from a randomized, controlled study. Am Heart J 2004;147:e24.
31. Kubo N, Ohmura N, Nakada I, et al. Exercise at ventilatory
threshold aggravates left ventricular remodeling in patients with
extensive anterior acute myocardial infarction. Am Heart J 2004;
147:113–20.
32. Koizumi T, Miyazaki A, Komiyama N, et al. Improvement of left
ventricular dysfunction during exercise by walking in patients with
successful percutaneous coronary intervention for acute myocardial
infarction. Circ J 2003;67:233–7.
33. Heldal M, Rootwelt K, Sire S, et al. Short-term physical training
reduces left ventricular dilatation during exercise soon after
myocardial infarction. Scand Cardiovasc J 2000;34:254–60.
34. Dubach P, Myers J, Dziekan G, et al. Effect of exercise training on
myocardial remodeling in patients with reduced left ventricular
function after myocardial infarction: application of magnetic
resonance imaging. Circulation 1997;95:2060–7.
35. Giannuzzi P, Tavazzi L, Temporelli PL, et al. Long-term physical
training and left ventricular remodeling after anterior myocardial
infarction: results of the Exercise in Anterior Myocardial Infarction
(EAMI) trial. EAMI Study Group. J Am Coll Cardiol 1993;22:
1821–9.
36. Jette M, Heller R, Landry F, Blumchen G. Randomized 4-week
exercise program in patients with impaired left ventricular function.
Circulation 1991;84:1561–7.
37. Norton K, Norton L, Sadgrove D. Position statement on physical
activity and exercise intensity terminology. J Sci Med Sport 2010;
13:496–502.
38. Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of
reports of randomized clinical trials: is blinding necessary? Control
Clin Trials 1996;17:1–12.
39. Thygesen K, Alpert JS, White HD, et al. Universal definition of
myocardial infarction. Circulation 2007;116:2634–53.
40. Pfeffer MA, Braunwald E. Ventricular remodeling after myocardial
infarction. Experimental observations and clinical implications.
Circulation 1990;81:1161–72.
41. Otsuka Y, Takaki H, Okano Y, et al. Exercise training without
ventricular remodeling in patients with moderate to severe left
ventricular dysfunction early after acute myocardial infarction. Int J
Cardiol 2003;87:237–44.
42. Gaudron P, Hu K, Schamberger R, et al. Effect of endurance training
early or late after coronary artery occlusion on left ventricular
remodeling, hemodynamics, and survival in rats with chronic
transmural myocardial infarction. Circulation 1994;89:402–12.
43. Alhaddad IA, Hakim I, Siddiqi F, et al. Early exercise after
experimental myocardial infarction: effect on left ventricular
remodeling. Coron Artery Dis 1998; 9:319–27.
44. Melo SF, Fernandes T, Barauna VG, et al. Expression of
MicroRNA-29 and collagen in cardiac muscle after swimming
training in myocardial-infarcted rats. Cell Physiol Biochem 2014;33:
657–69.
45. Martinez DG, Nicolau JC, Lage RL, et al. Effects of long-term
exercise training on autonomic control in myocardial infarction
patients. Hypertension 2011;58:1049–56.
46. Malfatto G, Facchini M, Bragato R, et al. Short and long term effects
of exercise training on the tonic autonomic modulation of heart rate
variability after myocardial infarction. Eur Heart J 1996;17:532–8.
47. Vona M, Rossi A, Capodaglio P, et al. Impact of physical training
and detraining on endothelium-dependent vasodilation in patients
with recent acute myocardial infarction. Am Heart J 2004;147:
1039–46.
48. Clark AM, Hartling L, Vandermeer B, McAlister FA. Meta-analysis:
secondary prevention programs for patients with coronary artery
disease. Ann Intern Med 2005;143:659–72.
49. Senaratne MP, Smith G, Gulamhusein SS. Feasibility and safety of
early exercise testing using the Bruce protocol after acute myocar-
dial infarction. J Am Coll Cardiol 2000;35:1212–20.
50. Grodzinski E, Jette M, Blumchen G, Borer JS. Effects of a four-week
training program on left ventricular function as assessed by
radionuclide ventriculography. J Cardiopulmonary Rehabil 1987;
7:517–24.
Supplementary available online
Supplementary Table S1 and S2 and Supplementary Figure S1
DOI: 10.3109/09638288.2015.1036174 Effects of exercise on cardiac rehabilitation 9
Downloaded by [112.2.36.22] at 18:36 07 May 2015
