Effects of High-Intensity Interval Training on Aerobic Capacity in Cardiac Patients: A Systematic Review with Meta-Analysis

Abstract

Purpose . The aim of this study was to compare the effects of high-intensity interval training (INTERVAL) and moderate-intensity continuous training (CONTINUOUS) on aerobic capacity in cardiac patients. Methods . A meta-analysis identified by searching the PubMed, Cochrane Library, EMBASE, and Web of Science databases from inception through December 2016 compared the effects of INTERVAL and CONTINUOUS among cardiac patients. Results . Twenty-one studies involving 736 participants with cardiac diseases were included. Compared with CONTINUOUS, INTERVAL was associated with greater improvement in peak VO 2 (mean difference 1.76 mL/kg/min, 95% confidence interval 1.06 to 2.46 mL/kg/min, p<0.001 ) and VO 2 at AT (mean difference 0.90 mL/kg/min, 95% confidence interval 0.0 to 1.72 mL/kg/min, p=0.03 ). No significant difference between the INTERVAL and CONTINUOUS groups was observed in terms of peak heart rate, peak minute ventilation, VE/VCO 2 slope and respiratory exchange ratio, body mass, systolic or diastolic blood pressure, triglyceride or low- or high-density lipoprotein cholesterol level, flow-mediated dilation, or left ventricular ejection fraction. Conclusions . This study showed that INTERVAL improves aerobic capacity more effectively than does CONTINUOUS in cardiac patients. Further studies with larger samples are needed to confirm our observations.

Full text

Research Article
Effects of High-Intensity Interval Training on Aerobic Capacity
in Cardiac Patients: A Systematic Review with Meta-Analysis
Bin Xie,1Xianfeng Yan,1Xiangna Cai,2and Jilin Li1
1Department of Cardiology, First Aliated Hospital of Shantou University Medical College, Shantou, Guangdong 515041, China
2Department of Plastic Surgery, First Aliated Hospital of Shantou University Medical College, Shantou, Guangdong 515041, China
Correspondence should be addressed to Xiangna Cai; caixiangna@.com and Jilin Li; lijilin@.com
Received 22 October 2016; Revised 7 January 2017; Accepted 19 January 2017; Published 12 March 2017
Academic Editor: Nikolaos G. Koulouris
Copyright ©  Bin Xie et al. is is an open access article distributed under the Creative Commons Attribution License, which
permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Purpose. e aim of this study was to compare the eects of high-intensity interval training (INTERVAL) and moderate-intensity
continuous training (CONTINUOUS) on aerobic capacity in cardiac patients. Methods. A meta-analysis identied by searching
the PubMed, Cochrane Library, EMBASE, and Web of Science databases from inception through December  compared the
eects of INTERVAL and CONTINUOUS among cardiac patients. Results. Twenty-one studies involving  participants with
cardiac diseases were included. C ompared with CONTINUOUS,INTERVAL was associated with greater improvement in peak VO2
(mean dierence .mL/kg/min, % condence interval . to . mL/kg/min, 𝑝 < 0.001)andVO
2at AT (mean dierence
. mL/kg/min, % condence interval . to . mL/kg/min, 𝑝 = 0.03). No signicant dierence between the INTERVAL
and CONTINUOUS groups was observed in terms of peak heart rate, peak minute ventilation, VE/VCO2slope and respiratory
exchange ratio, body mass, systolic or diastolic blood pressure, triglyceride or low- or high-density lipoprotein cholesterol level,
ow-mediated dilation, or le ventricular ejection fraction. Conclusions. is study showed that INTERVAL improves aerobic
capacity more eectively than does CONTINUOUS in cardiac patients. Further studies with larger samples are needed to conrm
our observations.
1. Introduction
Cardiovascular diseases (CVDs) remain the greatest cause
of death worldwide. In , more than  million people
died due to CVDs, of whom . million died of heart
attacks []. Interventions are urgently needed to address this
worrying trend. CVDs are largely preventable, and cardiac
rehabilitation is increasingly recognized as an important
component of the continuum of care for patients with
coronary artery disease (CAD) and chronic heart failure
(CHF). It is included in Class  recommendations of the
American Heart Association and the American College of
Cardiology for the treatment of these patients [, ].
According to the World Health Organization, insu-
cient physical activity is the fourth leading risk factor for
mortality, with % of deaths worldwide attributed to this
factor []. Exercise training is essential for cardiac patients.
It has an important role in improving endothelial function,
which in turn enhances blood ow by causing vasodilatation
and improving vasomotor function. Exercise training also
contributes to the improvement of many other functions,
such as the achievement of good glycemic control and
insulin sensitivity, leading to weight loss; the improvement
of blood pressure; and the correction of deranged lipid
proles [, ]. Proper exercise training is a cost-eective
and well-established primary intervention that delays the
onset of health burdens associated with various chronic
diseases in many cases. e appropriate amount, frequency,
and mode of exercise, however, remain unknown. Moreover,
the optimum “dose” of exercise to obtain maximum cardiac
benets remains unclear.
Aerobic capacity has been found to be the single best
parameter of cardiac function and all-cause death among
knowncasesofCVDs[].Itismeasureddirectlyaspeak
VO2.eimprovementofthepeakVO
2can improve aer-
obic capacity and promote cardiac rehabilitation. Moreover,
Hindawi
BioMed Research International
Volume 2017, Article ID 5420840, 16 pages
https://doi.org/10.1155/2017/5420840

BioMed Research International
reduction of the most common traditional risk factors for
CVDs (e.g., hypertension, hyperlipidemia, and obesity) can
decrease the occurrence of cardiovascular events. Research
suggests that CAD and CHF are associated with impaired
endothelial dysfunction, which is evaluated by ow-mediated
dilation (FMD) and can be improved through physical
exercise []. us, the identication of more eective exercise
programs is needed to improve cardiovascular benets in
cardiac patients.
Moderate-intensity continuous training (CONTINU-
OUS), a traditional exercise prescription, usually involves
walkingorcyclingfor–min[toreach–%peak
oxygen uptake (peak VO2)] []. However, recent evidence
from patients with CHF [] and CAD [] suggests that
high-intensity interval training (INTERVAL) may be a better
modality for the improvement of aerobic capacity. Although
INTERVAL has no standard denition, it refers to repeated
sessions of brief intermittent exercise, oen performed with
maximal eort or intensity (i.e., to achieve ≥% peak VO2)
[]. is intensity can be achieved by a single eort lasting
a few seconds to several minutes, or with multiple eorts
separated by a few minutes of rest or low-intensity exercise.
INTERVAL has been shown to have signicant benets,
including improved aerobic capacity, endothelial function,
and other cardiac functions, in patients with CAD and CHF
[, ].
Although several reviews and meta-analyses of INTER-
VAL for CAD and CHF were published [–], no consen-
sus has been reached about whether INTERVAL produces
superior physical, clinical, and functional benets compared
to CONTINUOUS. We are also unware of any systematic
reviews that have assessed the eect of INTERVAL among
cardiac patients.
issystematicreviewwasconductedtoassesswhether
INTERVAL produces larger eect sizes for change in aero-
bic capacity [peak VO2, oxygen consumption at anaerobic
threshold (VO2at AT), VE/VCO2slope, respiratory exchange
ratio (RER), peak minute ventilation (peak VE), peak heart
rate (PHR)], and physiological and clinical parameters com-
pared with CONTINUOUS among patients with known
cardiac disease (including CAD and CHF). e hypothesis
ofourstudywasthatINTERVALwillhaveagreatereect
on aerobic capacity given the superior improvement in
mitochondrial function and cardiac contractility.
2. Methods
We conducted this study according to the methods of the
Cochrane Handbook for Systematic Reviews of Interventions
[].
2.1. Search Strategy. The PubMed, Cochrane Library, EMBASE,
and Web of Science electronic databases were searched to
identify relevant clinical trials published between the earliest
available date and December  using the key words “heart
failure,” “coronary artery disease,” “high intensity interval
training,” “interval exercise,” and “high-intensity interval
exercise.” e reference lists of retrieved articles were also
searched to identify other appropriate studies.
2.2. Inclusion and Exclusion Criteria. Only full-text English-
language reports of clinical trials were considered for inclu-
sion. In addition, we considered only studies that com-
pared outcomes between an intervention group performing
INTERVAL and a control group performing CONTINUOUS,
with rhythmic aerobic exercise programs lasting at least 
weeks. Eligible studies also reported on at least one cardiores-
piratory exercise training outcome measure in patients with
cardiac disease. Reviews, cases reports, editorial comment,
communications, and reports without sucient data were
excluded in our meta-analysis.
2.3. Study Selection. Figure  illustrates the ow of study
selection. Two reviewers independently screened article titles
and abstracts, excluding irrelevant studies. Full texts were
then reviewed, and any study not fullling the inclusion
criteria was excluded. Dierences in the assessment of study
eligibility were resolved by discussion.
2.4. Data Extraction and Management. One reviewer col-
lected the data and the second reviewer rechecked it. Col-
lected data included authors’ names, year of publication,
country in which the study was conducted, duration of the
trial period, participant characteristics, intervention descrip-
tion, and outcomes assessed [peak VO2,VE/VCO
2slope,
RER, peak VE,PHR,VO
2at AT, body mass, blood pressure,
blood lipid parameters, FMD ndings, and le ventricular
ejection fraction (LVEF)]. Disagreements regarding the data
collectedwereresolvedbyconsensus.
2.5. Quality Assessment. e Cochrane collaboration’s tool
for assessing risk of bias was used for assessing the quality
of randomized controlled trials (RCTs) and Physiotherapy
Evidence Database (PEDro) scale nonrandomized controlled
studies, respectively [, ].
2.6. Statistical Analysis. e Cochrane Collaboration so-
ware (RevMan .; Cochrane Collaboration, Oxford, UK)
was used for meta-analyses. We calculated eect sizes by
subtracting preintervention from postintervention values.
When only baseline and postintervention standard devia-
tions (SDs) were reported, the following formula was used to
obtain the missing change value []: SDchange =√[(SDpre)2
+(SD
post)2−×corr(pre, post) ×SDpre ×SDpost], where
corr is the correlation coecient calculated for each outcome
using the formula of Conraads et al. []: corr = (SDpre2+
SDpost2−SDchange 2)/(2×SDpre ×SDpost). e heterogeneity
of included trials was assessed using the 𝐼2statistic and the
chi-squared test for heterogeneity. We used a xed-eects
model for studies showing signicant homogeneity (𝐼2<
%) and a random-eects model for other studies. Results
were considered signicant when 𝑝<.. To determine
the inuence of individual studies on the results obtained, we
conducted a sensitivity analysis with one-by-one removal of
studies. Publication bias was investigated using funnel plots
and Egger’s regression model.

BioMed Research International 
Identication
ScreeningEligibilityIncluded
Records identied through
database searching (n = 1712)
Additional records identied
through other sources (n=0)
Records aer duplicates removed
(n = 1194)
Records screened by titles and abstracts
(n = 1194)
Full-text articles assessed for eligibility
(n=63)
Studies included in qualitative synthesis
(aer combining two articles on the same
study) (n=21)
Studies included in quantitative synthesis
(meta-analysis) (n=21)
Records excluded
(n = 1131)
Full-text articles excluded
(n=40)
F : Flow chart of the study selection procedure.
3. Results
3.1. Characteristics of Identied Studies. e database search
yielded  titles. Aer the removal of duplicate records and
the screening of abstracts and titles to assess relevance, 
studies were selected for full-text review. Aer the exclusion
of  articles which did not comply with the inclusion
criteria, the nal sample consisted of  articles [, , ,
, –] that reported on  studies. e characteristics
of included studies are summarized in Table . All included
studies were the randomized controlled trials. e  studies
involvedatotalofpatients(%male,%female)with
cardiac disease (eleven studies examined patients with CAD
and ten studies examined those with CHF). Four studies
were conducted in Norway, three were conducted in Brazil,
two each were conducted in the United States, Greece, and
Canada, and one each was conducted in the Republic of
Korea, Belgium, Netherlands, France, Taiwan, Italy, Spain,
and United Kingdom. e duration of training programs
ranged from  to  weeks, and the frequency of exercise
training ranged from  to  days/week.
3.2. Risk of Bias. Figureshowstheriskofbiasoftheselected
studies. Six (.%) studies described the methods used to
generate and conceal allocation sequences. Participants were
not blinded in any study. Outcome assessors were blinded
to treatment allocation in sixteen (.%) studies. Seventeen
(.%) studies had incomplete descriptions of outcomes,
andeleven(.%)studieshadlowrisksofselectivereporting
bias.
3.3. Eects of Interventions on the Cardiorespiratory
Measurements
3.3.1. Peak VO2.e authors of  studies [, , , , –]
involving  patients reported on peak oxygen uptake fol-
lowing INTERVAL and CONTINUOUS. Peak VO2improved
by . mL/kg/min [% condence interval (CI) . to
. mL/kg/min] among patients in the INTERVAL groups,
which was greater than observed in the CONTINUOUS
groups, based on a random-eects model (overall 𝑍=.,
𝑝<.). However, this outcome showed signicant heter-
ogeneity (𝐼2=%,𝑝<.; Figure ).
ere was signicant heterogeneity in the study out-
comes. erefore, subgroup analysis was performed based
on the patient’s mean age and disease types. INTERVAL led
to signicantly greater improvements in peak VO2than did
CONTINUOUS in patients aged < years [mean dierence
(MD) . mL/kg/min, % CI . to . mL/kg/min, p<
., 𝐼2= %], those aged – years (MD . mL/kg/min,
% CI . to . mL/kg/min, p=.,𝐼2=%),andthose
aged > years (included in only one study []; Figure ).
From disease types subgroup analyses, INTERVAL also led
to signicantly greater improvements in peak VO2than did
CONTINUOUS in patients with CAD (MD . mL/kg/min,
% CI . to . mL/kg/min, 𝑝 < 0.001,𝐼2=%)
and those with CHF (MD . mL/kg/min, % CI . to
. mL/kg/min, 𝑝 = 0.004,𝐼2=%;Figure).
Sensitivity analysis did not change the statistical signi-
cance of the overall results. Exclusion of the study conducted

BioMed Research International
T : Characteristics of included studies.
Study Country Disease Patient number Mean age, year Mode Exercise program Exercise duration
(weeks)
INTERVAL CONTINUOUS INTERVAL CONTINUOUS INTERVAL CONTINUOUS
Angadi et al.,
 US CHF    . TM
d/wk,× mins @
%–%ofpeakHR,
 min recovery
 d/wk,  mins @ % of
peak HR 
Benda et al.,
 Netherlands CHF     Cycling
d/wk,.× min @
%, of maximal
workload,  periods
 d/wk,  min @
%–% of maximal
workload

Cardozo et al.,
 Brazil CAD     TM
 d/wk, mins @ % of
peak HR
 min recovery @ %
peak HR
 d/wk,  mins @
%–%ofpeakHR 
Conraads et
al.,  Belgium CAD    . Cycling
d/wk,× mins @
–% of peak HR,
 min recovery.  min
total
 d/wk, mins @ –%
of peak HR.  min total 
Currie et al.,
 Canada CAD     Cycling
d/wk,× min @
%–% PPO,  min
recovery@% PPO
 d/wk, –  min @
%–% PPO 
Dimopoulos
et al.,  Greece CHF   . . Cycling
 d/wk,  s@% of
WRp,  s rest,  min
total
 d/wk,  mins @ % of
WRp  ( sessions)
Freyssin et al.,
 France CHF     Cycling
d/wk,× s @ %
( wks) +  % ( wks) of
maximum power,  min
recovery @ rest
 d/wk, mins @ HRVT 
Fu et al.,  Taiwan CHF   . . Cycling
d/wk,× mins @ %
of peak VO2
 min recovery @ %
peak VO2
 d/wk,  mins @ % of
peak VO2
Iellamo et al.,
 Italy CHF   . . TM
– d/wk, – × mins @
%–% of HHR
 min recovery @
%–% of HHR
–d/wk, –mins@
%–% of HHR 
Jaureguizar et
al.,  Spain CAD     TM (–) × s @ the rst
(second) steep ramp test – mins @ VT 
Keteyian et al.,
 US CAD     TM
d/wk,× mins @
%–% of HRR,  min
recovery @ %–% of
HRR
 d/wk,  mins @
%–% of HRR 
Kim et al.,

Republic of
Korea CAD    . TM
walking
d/wk,× mins @
%–% of HRR,  min
recovery@%–%of
HRR.  min total
 d/wk,  mins @
%–% of HRR.  min
total


BioMed Research International 
T  : C on t i nued.
Study Country Disease Patient number Mean age, year Mode Exercise program Exercise duration
(weeks)
INTERVAL CONTINUOUS INTERVAL CONTINUOUS INTERVAL CONTINUOUS
Koufaki et al.,
 UK CHF   . . Cycling
d/wk,× min @ %
of PPO, min recovery @
%–% of PPO
d/wk,–months: ×
(–) mins; - months:
 mins; @ %–% of
peak VO2

Madssen et al.,
 Norway CAD   . . TM
d/wk,× mins @
%–% of peak HR,
 min recovery @ % of
peak HR
 d/wk,  mins @ % of
peak HR 
Moholdt et al.,
 Norway CAD   .  TM
d/wk,× mins @ %
of peak HR,  min
recovery @ % peak HR,
 min total
 d/wk,  mins @ % of
peak HR 
Rocco et al.,
2012∗
Prado et al.,
2016∗
Brazil CAD   . . TM
d/wk,× mins @ RCP,
× min @ VAT.  min
total
 d/wk,  mins @ VAT 
Roditis et al.,
 Greece CHF     Cycling d/wk,× s @ %
of WRp,  min total
 d/wk,  mins @ % of
WRp  ( sessions)
Rognmo et al.,
2004∗
Amundsen et
al., 2008∗
Norway CAD   . . TM
walking
d/wk,× mins @
%–% of peak HR,
 min recovery.  min
total
 d/wk,  mins @
%–% of peak VO2
Ulbrich et al.,
 Brazil CHF   .  TM
 d/wk, (–) × mins @
%ofpeakHR,min
recovery @ % peak HR.
 min total
 d/wk,  mins @ % of
peak HR .  min total 
Warburton et
al.,  Canada CAD     TM
 d/wk,  mins @ % of
VO2reserve,  min
recovery.  min total
 d/wk,  mins @ % of
VO2reser ve 
Wislø et al.,
 Norway CHF   . . TM
 d/wk,  mins @
%–%ofpeakHR,
 min recovery @
%–%peakHR.
 min total
 d/wk,  mins @
%–%ofpeakHR 
INTERVAL, high-intensity interval training; CONTINUOUS, moderate-intensity continuous training; CHF, chronic heart failure; CAD, coronary artery disease; TM, treadmill; HR, heart rate; PPO, peak power
output; HRR, heart rate reserve. WRp, %peak work rate. VAT, ventilatory anaerobic threshold. VO2,oxygenuptake.∗means two articles on the same study.

BioMed Research International
Random sequence generation (selection bias)
Allocation concealment (selection bias)
Blinding of participants and personnel (performance bias)
Blinding of outcome assessment (detection bias)
Incomplete outcome data (attrition bias)
Selective reporting (reporting bias)
Other biases
Low risk of bias
High risk of bias
Unclear risk of bias
25%50%75%100%0%
F : Quality assessment of RCTs using Cochrane collaboration’s tool for assessing risk of bias.
Heterogeneity: 𝜏2= 1.41;𝜒2= 50.09,df=20(p = 0.0002); I2=60%
Test for overall eect: Z=4.92(p < 0.00001)Favours
[CONTINUOUS]
Favours
[INTERVAL]
420−2−4
1.76 [1.06, 2.46]100.0%376362Total (95% CI)
4.10 [3.24,4.96]
8
0.6
1.991.156
1.15 [−3.06,5.36]
7
2.69
4.0175.01
Warburton et al. 2005 5.16
0.95 [−1.11,3.01]5.1%
10
2.41
1.84122.512.79
Ulbrich et al. 2016
3.30 [−1.85,8.45]1.5%
9
3.13
2.786.826
Rognmo et al. 2004
−0.10 [−2.18,1.98]5.1%
10
2.52
1.3112.321.2
Roditis et al. 2007
1.00 [−0.83,2.83]5.6%
20
3.19
3.4172.484.4
Rocco et al. 2012
1.00 [−0.55,2.55]6.3%
31
3.08
2.3282.993.3
Moholdt et al. 2009
1.30 [−3.03,5.63]2.0%
21
7.77
2155.53.3
Madssen et al. 2014
1.10 [−2.20,4.40]3.0%
9
4.15
1.382.722.4
Koufaki et al. 2014
3.99 [0.62,7.36]
14 2.9%
4.78
2.47144.316.46
Kim et al. 2015
5.6%
13
1.7
1.7153.13.6
Keteyian et al. 2014
0.15 [−2.25,2.55]4.4%
8
2.38
4.0982.524.24
Iellamo et al. 2013
3.50 [1.18,5.82]4.6%
13
3.56
0.1142.443.6
Fu et al. 2013
2.70 [1.13,4.27]6.3%
14
2.32
0.2121.752.9
Freyssin et al. 2012
0.30 [−1.72,2.32]5.2%
14
2.12
0.9102.721.2
Dimopoulos et al. 2006
1.10 [−1.43,3.63]4.2%
10
3.36
3.6112.444.7
Currie et al. 2013
0.70 [−0.40,1.80]7.5%
89
3.64
4.4853.755.1
Conraads et al. 2015
3.70 [1.92,5.48]5.7%
24
3.39
0.1232.833.8
Cardozo et al. 2015
1.20 [−0.75,3.15]5.3%
10
2.07
0.1102.381.3
Benda et al. 2015
4.0%
2.2
−0.192.941.8
Angadi et al. 2015
2.00 [0.07,3.93]5.4%
36
3.6
2.5364.74.5
Jaureguizar et al. 2016
INTERVAL CONTINUOUS
Study or subgroup Mean dierence
IV, random, 95% CI
Mean dierence
IV, random, 95% CI
Weight
Tot a lSDMean Total
SD
Mean
Wislø et al. 2007
[−0.71,4.51]1.90
[0.08,3.72]1.90
6
2.1%
8.0%
F:Meta-analysisofeectsofINTERVALonpeakVO
2.
by Wislø et al. [], which provided inferior evidence for
theeectofINTERVALonpeakVO
2,signicantlyimproved
homogeneity.
3.3.2. VO2at AT. e authors of fourteen studies [, ,
,–,,,,]involvingpatientsreported
on VO2at AT following INTERVAL and CONTINUOUS.
VO2at AT improved by .mL/kg/min [% CI . to
. mL/kg/min] among patients in the INTERVAL groups,
which was greater than observed in the CONTINUOUS
groups, based on a random-eects model (overall Z=
., p= .). However, this outcome showed signicant
heterogeneity (𝐼2=%,p<.; Figure ).
3.3.3. Peak Heart Rate. e authors of seventeen studies
[, , , –, –, , , –] involving  patients

BioMed Research International 
CONTINUOUS
Study or subgroup Weight
Tot a lSDMean
INTERVAL
Tot a lSDMean
Mean dierence
IV, random, 95% CI
Mean dierence
3.70 [1.92,5.48]5.7%
243.390.1232.833.8
Cardozo et al. 2015
0.70 [−0.40,1.80]7.5%
893.644.4853.755.1
Conraads et al. 2015
6.3%
142.320.2121.752.9
Freyssin et al. 2012
2.00 [0.07,3.93]5.4%
363.62.5364.74.5
Jaureguizar et al. 2016
1.90 [0.08,3.72]5.6%
131.71.7153.13.6
Keteyian et al. 2014
3.99 [0.62,7.36]2.9%
144.782.47146.46
Kim et al. 2015
3.0%
94.151.382.722.4
Koufaki et al. 2014
1.30 [−3.03,5.63]2.0%
217.772155.53.3
Madssen et al. 2014
1.00 [−0.83,2.83]5.6%
203.193.4172.484.4
Rocco et al. 2012
0.95 [−1.11,3.01]5.1%
102.411.84122.512.79
Ulbrich et al. 2016
1.15 [−3.06,5.36]2.1%
72.694.0175.015.16
Warburton et al. 2005
1.80 [1.10, 2.50]51.3%257244Subtotal (95% CI)
4.0%
2.2−0.192.941.8
Angadi et al. 2015
5.3%
102.070.1102.381.3
Benda et al. 2015
1.10 [−1.43,3.63]4.2%
103.363.6112.444.7
Currie et al. 2013
0.30 [−1.72,2.32]5.2%
142.120.9102.721.2
Dimopoulos et al. 2006
3.50 [1.18,5.82]4.6%
133.560.1142.443.6
Fu et al. 2013
0.15 [−2.25,2.55]4.4%
8
6
2.384.0982.524.24
Iellamo et al.2013
1.00 [−0.55,2.55]6.3%
313.082.3282.993.3
Moholdt et al. 2009
−0.10 [−2.18,1.98]5.1%
102.521.3112.321.2
Roditis et al. 2007
3.30 [−1.85,8.45]1.5%
93.132.786.826
Rognmo et al. 2004
1.10 [0.36, 1.83]40.6%111109Subtotal (95% CI)
4.10 [3.24,4.96]8.0%
0.61.99
9
1.156
Wislø et al. 2007
4.10 [3.24, 4.96]8.0%
8
8Subtotal (95% CI)
1.76 [1.06, 2.46]100.0%376362Total (95% CI)
[1.13,4.27]
2.70
−2.20,
1.10 [ 4.40]
4.31
1.90 [−0.71, 4.51]
[−0.75,3.15]
1.20
Heterogeneity: 𝜏2= 0.29;𝜒2= 12.78,df=10(p = 0.24); I2=22%
Test for overall eect: Z = 5.06 (p < 0.00001 )
Heterogeneity: 𝜏2= 0.00;𝜒2= 7.69,df=8(p = 0.46); I2=0%
Test for overall eect: Z = 2.93 (p = 0.003)
Heterogeneity: not applicable
Test for overall eect: Z=9.36 (p<0.00001)
Test for overall eect: Z=4.92 (p<0.00001)
Test for subgroup dierences: =28.68, df =2(p<0.00001), I2=93.0%
<60 years
1.3.1 Age
1.3.2 Age 60–75 years
Heterogeneity: 𝜏2=1.41; 𝜒2=
𝜒2
50.09, df =20(p=0.0002); I2=60%
Favours [CONTINUOUS] Favours [INTERVAL]
420−2−4
1.3.3 Age>75 years
IV, random, 95% CI
F : Meta-analysis of the eects of INTERVAL on peak VO2according to age.
reported on peak heart rate following INTERVAL and CON-
TINUOUS. A random-eects model revealed no signicant
dierence between groups (MD . bpm, % CI −. to
. bpm, 𝑝 = 0.55).
3.3.4. Peak Minute Ventilation. e authors of ve studies
[, , –] involving  patients reported on peak
VE following INTERVAL and CONTINUOUS. A random-
eects model revealed no signicant dierence between
groups (MD . l/min, % CI −. to . l/min, 𝑝 = 0.19).
3.3.5. VE/VCO2Slope. e authors of nine studies [–, ,
, , , ] involving  patients reported on VE/VCO2
slope following INTERVAL and CONTINUOUS. A xed-
eects model revealed no signicant dierence between
groups (MD ., % CI −. to ., 𝑝 = 0.46).
3.3.6. Respiratory Exchange Ratio. e authors of fourteen
studies[,,,,,–,]involvingpatients
reported on RER following INTERVAL and CONTINUOUS.
A random-eects model revealed no signicant dierence
between groups (MD ., % CI −. to ., 𝑝 = 0.25).
3.4. Eects of Interventions on Physiological and
Clinical Parameters
3.4.1. Body Mass. e authors of eight studies [, , ,
,,,,]involvingpatientsreporteddecreased
body mass following INTERVAL and CONTINUOUS. A
xed-eects model revealed no signicant dierence between
groups (MD . kg , % CI −. to . kg, 𝑝 = 0.31).
3.4.2. Blood Pressure. e authors of eight studies [, , ,
,,,,]involvingpatientsreportedonsystolic
blood pressure (SBP) and diastolic blood pressure (DBP) fol-
lowing INTERVAL and CONTINUOUS. A random-eects
model revealed no signicant dierence between groups

BioMed Research International
51.0%
5.7%
7.5%
4.2%
5.4%
5.6%
2.9%
2.0%
6.3%
5.6%
1.5%
2.1%
49.0%
100.0%
8.0%
5.1%
5.1%
3.0%
4.4%
4.6%
6.3%
5.2%
5.3%
4.0%
INTERVAL CONTINUOUS
Study or subgroup IV, random, 95% CI
103 102
243.390.1232.833.8
893.644.4853.755.1
103.363.6112.444.7
363.62.5364.74.5
131.71.7153.13.6
144.782.47144.316.46
217.772155.53.3
313.082.3282.993.3
203.193.4172.484.4
93.132.786.826
72.694.0175.015.16
274259
376362
80.61.991.156
102.411.84122.512.79
102.521.3112.321.2
94.151.382.722.4
82.384.0982.524.24
133.560.1142.443.6
142.320.2121.752.9
142.120.9102.721.2
102.070.1102.381.3
62.2−0.192.941.8
Weight
Tot a lSDMeanTot a lSDMean
Mean dierence Mean dierence
1.2.1 CAD
1.2.2 CHF
Subtotal (95% CI) 1.70 [0.53, 2.86]
3.70 [1.92,5.48]Cardozo et al. 2015
0.70 [−0.40,1.80]Conraads et al. 2015
1.10 [−1.43,3.63]
Currie et al. 2013
2.00 [0.07,3.93]Jaureguizar et al. 2016
1.90 [0.08,3.72]Keteyian et al. 2014
3.99 [0.62,7.36]Kim et al. 2015
1.30 [−3.03,5.63]Madssen et al. 2014
1.00 [−0.55,2.55]Moholdt et al. 2009
1.00 [−0.83,2.83]
Rocco et al. 2012
Rognmo et al. 2004
1.15 [−3.06,5.36]Warburton et al. 2005
1.62 [0.94, 2.30]Subtotal (95% CI)
1.76 [1.06, 2.46]Total (95% CI)
4.10 [3.24,4.96]Wislø et al. 2007
0.95 [−1.11,3.01]Ulbrich et al. 2016
−0.10 [−2.18,1.98]Roditis et al. 2007
1.10 [−2.20,4.40]Koufaki et al. 2014
0.15 [−2.25,2.55]Iellamo et al. 2013
3.50 [1.18,5.82]
Fu et al. 2013
2.70 [1.13,4.27]Freyssin et al. 2012
0.30 [−1.72,2.32]Dimopoulos et al. 2006
Benda et al. 2015
1.90 [−0.71,4.51]Angadi et al. 2015
[−1.85,8.45]
3.30
[−0.75,3.15]
1.20
Favours [CONTINUOUS] Favours [INTERVAL]
420−2−4
Heterogeneity: 𝜏2= 0.19;𝜒2= 11.69,df=10(p = 0.31); I2=14%
Test for overall eect: Z=4.69(p < 0.00001)
Heterogeneity: 𝜏2=2.38; 𝜒2=33.13, df =9(p=0.0001); I2=73%
Test for overall eect: Z = 2.85 (p = 0.004)
Heterogeneity: 𝜏2=1.41; 𝜒2=50.09, df =20(p=0.0002); I2=60%
Test for overall eect: Z=4.92(p<0.00001)
Test for subgroup dierences: 𝜒2=0.01, df =1(p = 0.91), I2=0%
IV, random, 95% CI
F : Meta-analysis of the eects of INTERVAL on peak VO2according to disease types.
Weight
Tot a lSDMeanTot a lSDMean
Study or subgroup Mean dierence Mean dierence
CONTINUOUS
INTERVAL
0.30 [−1.46,2.06]6.7%
1.30.692.170.9
Angadi et al. 2015
1.80 [0.51,3.09]7.8%
241.98−0.4232.51.4
Cardozo et al. 2015
1.20 [−0.34,2.74]7.2%
101.882.1111.713.3
Currie et al. 2013
1.50 [0.25,2.75]7.9%
141.90.2121.331.7
Freyssin et al. 2012
7.6%
141.611.1101.780.6
Dimopoulos et al. 2006
0.90 [0.08, 1.72]100.0%192190Total (95% CI)
1.30 [0.81,1.79]9.3%
80.492.190.543.4
Wislø et al. 2007
5.00 [1.00,9.00]2.9%
71.71275.127
Warburton et al. 2005
0.10 [−1.65,1.85]6.7%
101.950.9112.131
Roditis et al. 2007
8.3%
201.772.5171.571.7
Rocco et al. 2012
4.6%
93.262.782.541.5
Koufaki et al. 2014
2.30 [0.45,4.15]6.5%
132.20.7152.83
Keteyian et al. 2014
0.70 [−0.50,1.90]8.0%
362.11.83632.5
Jaureguizar et al. 2016
−1.31 [−2.63,0.01]7.7%
81.313.6581.382.34
Iellamo et al. 2013
3.60 [2.75,4.45]8.8%
131.02−0.6141.223
Fu et al. 2013
6
Heterogeneity: 𝜏2= 1.82;𝜒2=75.90, df =13(p<0.00001); I2=83%
Test for overall eect: Z = 2.14 (p = 0.03)Favours [CONTINUOUS] Favours [INTERVAL]
420−2−4
−0.50 [−1.89, 0.89]
−0.80 [−1.88, 0.28]
−1.20 [−3.96, 1.56]
IV, random, 95% CIIV, random, 95% CI
F : Meta-analysis of eects of INTERVAL on VO2at AT.

BioMed Research International 
5
4
3
2
1
0
SE (MD)
402−2−4
MD
F : Funnel plot of publication bias.
(SBP: MD −. mmHg, % CI −. to . mmHg, 𝑝=
0.97;DBP:MD−. mmHg, % CI −. to . mmHg,
𝑝 = 0.60).
3.4.3. Blood Lipids. Data on high-density lipoprotein choles-
terol (HDL-C), low-density lipoprotein cholesterol (LDL-
C), and triglyceride (TG) levels following INTERVAL and
CONTINUOUS were reported in six studies [, , ,
, , ] involving  patients. A random-eects model
showed no signicant dierence between groups (TG: stan-
dardized mean dierence (SMD) −., % CI −. to
.; LDL-C: SMD −., % CI −. to .; HDL-C: SMD
., % CI −. to .). e result of cholesterol was
assessed in four studies [, , , ] involving  patients.
A random-eects model revealed no signicant dierence
between groups (SMD ., % CI −. to .).
3.4.4. Flow-Mediated Dilation. e authors of six studies [,
,,,,]involvingpatientsreportedonFMDfol-
lowing INTERVAL and CONTINUOUS. A random-eects
model showed no signicant dierence between groups (MD
.%, % CI −.% to .%, 𝑝 = 0.09).
3.4.5. Le Ventricular Ejection Fraction. e authors of eight
studies [, , –, , , ] involving  patients reported
increased LVEF following INTERVAL and CONTINUOUS.
A random-eects model showed no signicant dierence
between groups (MD .%, % CI −.% to .%, 𝑝=
0.12).
3.5. Publication Bias. Egger’s regression analysis excluded
relevant publication bias for peak VO2(p=.),andthe
funnel plot of these data was symmetrical.
4. Discussion
To our knowledge, most previous systematic reviews on this
topic have focused on patients with specic diseases, such
as CAD and CHF. One previous review [] has examined
whether INTERVAL is more eective than CONTINUOUS
for improving peak VO2andLVEFinCHFpatients.However,
thisreviewfocusedonlyonCHFandonlysevenarticles
were included in the review. is systematic review examined
the ecacy of INTERVAL as a part of cardiac rehabilita-
tion in patients with cardiac disease (including CHF and
CAD). Twenty-one studies involving  cardiac patients
were included in the review. e main ndings were that
INTERVAL appears to be at least as eective as and in some
cases more eective than CONTINUOUS, for the improve-
ment of aerobic capacity, although we found evidence of
heterogeneity among studies. Heterogeneous results for this
outcome in the study conducted by Wislø et al. [] were due
mainly to the inclusion of elderly patients.
4.1. Rationale and Potential Working Mechanisms of INTER-
VAL. Duetorepeatedalternationofhigh-andlow-intensity
exercise, INTERVAL’s stimulation of the body uctuates. e
rationale is to accumulate more time in high-intensity zones
compared to a continuous exercise where exhaustion would
occur more prematurely and therefore to produce a stronger
stimulusforcardiovascularandmuscularadaptations[,
]. e mechanisms involved in the superiority of INTER-
VAL to CONTINUOUS have not been clearly elucidated. e
potential mechanisms for the greater improvement in aerobic
capacity achieved by INTERVAL include increased activation
of peroxisome-proliferator activated receptor 𝛾coactivator
(PGC-𝛼), which improves mitochondrial function [, ,
], and increased maximal rate of Ca2+ reuptake into
the sarcoplasmic reticulum, which reduces skeletal muscle
fatigue [, ]. e increase in PGC-𝛼to be strongly
correlated with the improved VO2peak (𝑟 = 0.72,𝑝 < 0.01)
was found by Wislø et al. [], supporting the inuence of
mitochondrial function on exercise capacity.
INTERVAL has been demonstrated to activate p
mitogen-activated protein kinase and 󸀠-adenosine mono-
phosphate-activated protein kinase. Both of these exercise-
responsive signaling kinases are implicated in direct phos-
phorylation and activation of PGC-𝛼.Increasednuclear
abundance of PGC-𝛼following INTERVAL may coacti-
vate transcription factors to increase mitochondrial gene
transcription, ultimately resulting in accumulation of more
mitochondrial proteins to drive mitochondrial biogenesis
[]. Mitochondrial biogenesis is essential to maintain the
structural integrity of skeletal muscle. Mitochondrial func-
tion is associated with aerobic physical tness and plays
an important pathophysiological role in cardiac patients.
Consequently, the major benets of INTERVAL interven-
tions include enhanced peripheral blood circulation [],
as well as increased skeletal muscle and functional capacity
[–]. e improvement of peak VO2, a strong, inde-
pendent predictor of all-cause and cardiovascular-specic
mortality [, ], through INTERVAL is thus of clinical
signicance.
e magnitude of dierence in the eects of INTERVAL
and CONTINUOUS in terms of VE/VCO2slope, RER, peak
VE, PHR, body mass, blood pressure, blood lipids, FMD,
and LVEF was small in the present analysis, which may be
related to the examination of short-term outcomes in the
included studies. us, more research is necessary to provide
information on the long-term eects of INTERVAL.

 BioMed Research International
+
++
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
++
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
??
?
??
?
?
?
??
?
?
?
Random sequence generation (selection bias)
Allocation concealment (selection bias)
Blinding of participants and personnel (performance bias)
Blinding of outcome assessment (detection bias)
Incomplete outcome data (attrition bias)
Selective reporting (reporting bias)
Other biases
Wislø et al. 2007
Warburton et al. 2005
Ulbrich et al. 2016
Rognmo et al. 2004
Roditis et al. 2007
Rocco et al. 2012
Moholdt et al. 2009
Madssen et al. 2014
Koufaki et al. 2014
Kim et al. 2015
Keteyian et al. 2014
Jaureguizar et al. 2016
Iellamo et al. 2013
Fu et al. 2013
Freyssin et al. 2012
Dimopoulos et al. 2006
Currie et al. 2013
Conraads et al. 2015
Cardozo et al. 2015
Benda et al. 2015
Angadi et al. 2015
F : Risk of bias summary.
0 12.55 9 6 16.09 6 3.1%
8.82 10 3 13.07 10 5.4%
2 13.89 23 1 10.51 24 7.2%
11 11.41 85 9 12.06 89 9.8%
6 8.82 10 3 13.67 10 5.2%
2 9.81 10 10 11.48 14 6.1%
6 12.15 14 3 23.48 13 3.4%
7.05 8 4.9 8 8.0%
10.6 11.7 36 1
1
11.64 36 8.4%
6 9.62 15 13 5.8%
13.1 9.59 14
14.27
0.1 12.3 14 6.4%
4 16.7 15 3 23.99 21 3.7%
2 14.59 17 11.28 18 6.1%
9.54 11 10.5 11.48 10 5.8%
3 11.48 8 0 7.7 9 5.6%
10.36 7 4 8.15 7 5.4%
11.95 9 11.4 8 4.7%
310 100.0%
20010
Angadi et al. 2015
Benda et al. 2015
Cardozo et al. 2015
Conraads et al. 2015
Currie et al. 2013
Dimopoulos et al. 2006
Fu et al. 2013
Iellamo et al. 2013
Jaureguizar et al. 2016
Keteyian et al. 2014
Kim et al. 2015
Madssen et al. 2014
Prado et al. 2016
Roditis et al. 2007
Rognmo et al. 2004
Warburton et al. 2005
Wislø et al. 2007
−6.00 [−21.26,9.26]
−4.00 [−13.77,5.77]
1.00 [−6.06,8.06]
2.00 [−1.49,5.49]
3.00 [−7.08,13.08]
−8.00 [−16.55,0.55]
−2.00 [−7.95,3.95]
9.60 [4.21,14.99]
9.00 [−0.16,18.16]
13.00 [4.83,21.17]
1.00 [−12.29,14.29]
4.00 [−4.68,12.68]
−10.80 [−19.88,−1.72]
3.00 [−6.41,12.41]
0.00 [−11.11,11.11]
0.97 [−2.19, 4.12]
−9.00 [−18.76,0.76]
−1
−1
−0.3
−5
−2
−3
−2
−2
Mean SD IV, random, 95% CI
INTERVAL CONTINUOUS Mean dierence Mean dierence
Study or subgroup Weight
Tot a lSDMeanTo t a l
Heterogeneity: 𝜏2=23.27; 𝜒2=39.25, df =16(p=0.001); I2=59%
Test for overall eect: Z = 0.60 (p = 0.55)
Total (95% CI) 301
3.00 [−11.26,17.26]
−20 −10
Favours [CONTINUOUS] Favours [INTERVAL]
IV, random, 95% CI
F:Meta-analysisofeectsofINTERVALonPHR.
Mean SD
INTERVAL CONTINUOUS Mean dierence
Study or subgroup
−1.70 [−8.42,5.02]25.7%
149.466.6107.334.9
Dimopoulos et al. 2006
28.1%
137.05−1.1148.848.5
Fu et al. 2013
20.8%
1011.226.4117.897
Roditis et al. 2007
7.70 [−3.86,19.26]14.0%
910.26.48
7
13.6314.1
Rognmo et al. 2004
0.00 [−13.21,13.21]11.5%
714.231010.7510
Warburton et al. 2005
3.46 [−1.75, 8.67]100.0%5350Total (95% CI)
Weight
Tot a lSDMeanTot a l
Mean dierence
20010
−20 −10
Favours
[CONTINUOUS]
Favours
[INTERVAL]
Heterogeneity: 𝜏2=15.75; 𝜒2=7.46, df =4(p=0.11); I2=46%
Test for overall eect: Z = 1.30 (p = 0.19)
0.60 [−7.77,8.97]
9.60 [3.59, 15.61]
IV, random, 95% CIIV, random, 95% CI
F : Meta-analysis of eects of INTERVAL on peak VE.

BioMed Research International 
0.20 [−4.11,4.51]2.2%
61.690.296.270.4
Angadi et al. 2015
0.20 [−3.21,3.61]3.5%
103.520.5104.230.7
Benda et al. 2015
0.80 [−0.59,2.19]20.9%
242.56−1.1232.29−0.3
Cardozo et al. 2015
5.9%
143.380.5103.08−0.8
Dimopoulos et al. 2006
−1.25 [−3.29,0.79]9.7%
82.44−0.7581.65−2
Iellamo et al. 2013
−0.10 [−1.93,1.73]12.0%
132.830151.96−0.1
Keteyian et al. 2014
1.10 [0.00,2.20]33.5%
182.17−3.5170.93−2.4
Prado et al. 2016
−0.50 [−3.34,2.34]5.0%
103.910.9112.50.4
Roditis et al. 2007
−1.00 [−3.34,1.34]7.4%
72.2172.260
Warburton et al. 2005
INTERVAL CONTINUOUS
Study or subgroup Weight
Tot a lSDMeanTot a lSDMean
Mean dierence Mean dierence
Heterogeneity: 𝜒2=7.86, df =8(p=0.45); I2=0%
Test for overall eect: Z = 0.74 (p = 0.46)Favours
[INTERVAL]
Favours
[CONTINUOUS]
420−2−4
0.24 [−0.40, 0.87]
100.0%110110
Total (95% CI)
1.30 [−3.90, 1.30]
−
IV, xed, 95% CI IV, xed, 95% CI
F : Meta-analysis of eects of INTERVAL on VE/VCO2slope.
INTERVAL CONTINUOUS
SD
Mean dierence
IV, random, 95% CI
Mean dierence
IV, random, 95% CI
0.07 9 0.05 6 4.6%
0.06 23 0.06 24 8.6%
0.07 85 0.06 89 12.0%
0.07 9
0.04 8
0.04
0.02
0.04 9 5.5%
0.04 8 7.6%
0.1 36 0.03 0.1 36 6.4%
0.06 15 0.05 13 7.4%
0.08 14 0.08 0.09 14 4.3%
0.06 8 0.08 9 4.0%
0.05 15 0 0.06 21 8.3%
0.04 28 0 0.04 31 11.8%
0.46 17 0.48 18 0.2%
0.04 8 0.02 0.02 9 9.4%
0.03 9 0.03 8 9.9%
284 295 100.0%
Study or subgroup
Angadi et al. 2015
Cardozo et al. 2015
Conraads et al. 2015
Currie et al. 2013
Iellamo et al. 2013
Jaureguizar et al. 2016
Keteyian et al. 2014
Kim et al. 2015
Koufaki et al. 2014
Madssen et al. 2014
Moholdt et al. 2009
Prado et al. 2016
Rognmo et al. 2004
Wislø et al. 2007
Total (95% CI)
Mean Mean
0.07
0.02
0.02
−0.02
−0.01
0.04
0.03
0.01
0
0
0.01
−0.04
0.03
0.03
0.08 [0.02,0.14]
0.03 [−0.00,0.06]
0.01 [−0.01,0.03]
−0.06 [−0.11,−0.01]
−0.03 [−0.07,0.01]
0.01 [−0.04,0.06]
0.04 [−0.00,0.08]
−0.07 [−0.13,−0.01]
0.02 [−0.05,0.09]
0.00 [−0.04,0.04]
0.01 [−0.01,0.03]
−0.02 [−0.33,0.29]
0.01 [−0.02,0.04]
0.04 [0.01,0.07]
0.01 [−0.01, 0.02 ]
−0.01
−0.02
−0.02
−0.01
−0.01
0.01
−0.01
Weight
Tot a lSD Total
0.10.050−0.05
−0.1
Favours [CONTINUOUS] Favours [INTERVAL]
Heterogeneity: 𝜏2= 0.00;𝜒2=30.04, df =13(p=0.005); I2=57%
Test for overall eect: Z = 1.15 (p = 0.25)
F : Meta-analysis of eects of INTERVAL on RER.
Angadi et al. 2015
Conraads et al. 2015
Currie et al. 2013
Fu et al. 2013
Moholdt et al. 2009
Rocco et al. 2012
Rognmo et al. 2004
Warburton et al. 2005
−2.40 [−11.93,7.13]
1.00 [−1.36,3.36]
0.60 [−1.16,2.36]
0.00 [−2.31,2.31]
1.00 [−3.12,5.12]
0.40 [−4.69,5.49]
0.70 [−3.42,4.82]
1.00 [−5.71,7.71]
−1.7
0.5
−0.6
−0.2
0.2
0.3
−0.1
−3
10.94
8.12
2.2
3.05
9.22
7.44
4.34
3.96
9
85
10
14
28
17
8
7
0.7
−0.5
−1.2
−0.2
−0.8
−0.1
−0.8
−4
7.87
7.74
1.8
3.06
6.55
8.35
4.3
8.15
6
89
10
13
31
20
9
7
1.3%
20.4%
36.6%
21.4%
6.7%
4.4%
6.7%
2.5%
0.55 [−0.52, 1.62]100.0%185178
Total (95% CI)
Mean SD IV, xed, 95% CI
INTERVAL CONTINUOUS Mean dierence Mean dierence
Study or subgroup Weight
Tot a lSDMeanTo tal
Heterogeneity: 𝜒2=0.80, df =7(p=1.00); I2=0%
Test for overall eect: Z = 1.01 (p = 0.31)420−2
−4
Favours
[INTERVAL]
Favours
[CONTINUOUS]
IV, dixed, 95% CI
F : Meta-analysis of eects of INTERVAL on body mass.

 BioMed Research International
INTERVAL CONTINUOUS Mean dierence Mean dierence
Study or subgroup
2.2.1 SBP
Angadi et al. 2015 9 6
Conraads et al. 2015 8.09 85 89
Currie et al. 2013 11 10
Fu et al. 2013 14 13
Jaureguizar et al. 2016 17.7 36 36
Keteyian et al. 2014 15 13
Rognmo et al. 2004 8 9
Ulbrich et al. 2016 12 10
Subtotal (95% CI)
2.2.2 DBP
Angadi et al. 2015 9 6
Conraads et al. 2015 85 89
Currie et al. 2013 11 10
Fu et al. 2013 14 7.93 13
Jaureguizar et al. 2016 11 36 11.3 36
Keteyian et al. 2014 15 0
0
6.6 13
Rognmo et al. 2004 8 0 7.8 9
Ulbrich et al. 2016 12 10
Mean SD Weight
8.0%
19.6%
12.3%
17.2%
12.7%
11.6%
7.5%
11.1%
100.0%
13.1%
18.1%
12.3%
9.7%
12.1%
12.9%
9.8%
12.0%
100.0%
8.00 [−5.07,21.07]
6.00 [3.40,8.60]
3.00 [−5.63,11.63]
0.00 [−4.72,4.72]
2.50 [−5.82,10.82]
−11.00 [−20.26,−1.74]
−15.80,11.80]
−2.00 [
−10.50 [−20.20,−0.80]
−0.09 [−4.82, 4.65]
−4.00 [−8.57,0.57]
2.60 [1.03,4.17]
5.00 [0.00,10.00]
−1.00 [−7.62,5.62]
−1.40 [−6.55,3.75]
−7.00 [−11.69,−2.31]
1.00 [−5.59,7.59]
−2.40 [−7.62,2.82]
−0.79 [−3.75, 2.16]
7.5−8.1
−1 16.62
2 14.41
18.3−5.6
−2 6.12
−6 10.36
9.4−6
−12 14.71
−5 4.94
−1
−7 5.66
0.4
−
−
3.7 5.79
−4 4.35
−7.4 7.47
1 6.02
−7 5.97
−1.8
−2 9.59
−2 6.02
−1.1 4.75
−8 4.53
−18.6 15.05
−3 12.3
−9 9.77
−3.1
−2 6.39
−3 9.75
8.7−4
IV, random, 95% CI IV, random, 95% CITot a l Mean SD Total
Heterogeneity: 𝜏2=28.06; 𝜒2=23.95,df=7(p=0.001); I2=71%
Test for overall eect: Z = 0.04 (p = 0.97)
Heterogeneity: 𝜏2=11.88; 𝜒2= 25.02,df=7(p=0.0008); I2=72%
Test for overall eect: Z = 0.53 (p = 0.60)
186190
Subtotal (95% CI) 186190
20010
−20 −10
Favours
[INTERVAL]
Favours
[CONTINUOUS]
F : Meta-analysis of eects of INTERVAL on blood pressure.
A meta-analysis focused mainly on patients with CHF
by Haykowsky et al. [] showed INTERVAL is more
eective than CONTINUOUS for improving peak VO2
(MD . mL/kg/min, % CI . to . mL/kg/min)
but not the LVEF. Another systematic analysis by Smart
et al. [] that analyzed  CHF patients revealed that
INTERVAL determined a signicant increase in peak VO2
(MD . mL/kg/min, % CI . to . mL/kg/min) and
VE/VCO2slope (MD −., % CI −. to −.). A more
recent meta-analysis including CAD patients by Pattyn et al.
[] reported higher increase in VO peak with INTERVAL
(MD . mL/kg/min, % CI . to .mL/kg/min) but
VE/VCO2slope, VO at AT, and body mass. From our
analysis, INTERVAL had similar eect results in improving
peak VO2in above meta-analysis. In addition, our systematic
review included more evaluative indicators, such as PHR,
peak VE,VE/VCO
2slope, RER, VO2at AT, blood pressure,
bloodlipids,FMD,andLVEF,thandidthepreviousmeta-
analysis.
Pooled estimates showed signicant heterogeneity among
studies included in this review. Important clinical and
methodological dierences may have aected the results
obtained in the intervention and control groups. Some
of these dierences were in inclusion criteria and among
participants, who were in dierent countries and of dierent
ages.
4.2. Study Limitations. Our systematic review has some
limitations.Fewtrialsincludedinthestudyprovidedclear
descriptions of the randomization and allocation of partic-
ipants to treatments. Many of the studies failed to describe
the blinding of assessors to treatment allocation, which raises
the possibility of performance bias. In addition, although
we examined publication bias because we searched only
four electronic databases, we did not search for unpublished
trials.Moreover,thereviewincludedonlyRCTspublished
in English. Consequently, our results may have been aected
by publication bias. Several meta-analyses were aected by
statistical heterogeneity, possibly due to dierences in study
methodologies and data collection techniques (e.g., wide
ranges of variability in age, sex, and follow-up duration),
which may have aected our ndings. Finally, most of
thestudieshadsmallsamples,andnolarge-scaleclinical
RCT was included, which likely aected the objectivity and
reliability of this meta-analysis and systematic review.
4.3. Conclusion. e current analysis indicated that INTER-
VAL can provide more benets than CONTINUOUS in
terms of improving peak VO2and VO2at AT in patients
with cardiac disease. INTERVAL programs, which increase
exercise capacity compared with traditional exercise, are
thus preferable. Dierences in the eects of INTERVAL and
CONTINUOUS in terms of PHR, peak VE,VE/VCO
2slope,
RER, body mass, blood pressure, blood lipids, FMD, and

BioMed Research International 
0.02 [−0.27,0.32]
0.02 [−0.27,0.32]
0.04 [−0.72,0.79]
−0.19 [−1.17,0.80]
−0.44 [−1.19,0.32]
0.00 [−0.66,0.66]
−0.10 [−0.62,0.41]
−0.05 [−0.26, 0.17]
0.06 [−0.24,0.36]
−3.43 [−4.67,−2.18]
0.69 [−0.32,1.71]
−0.74 [−1.50,0.03]
−0.16 [−0.82,0.51]
−0.67 [−1.20,−0.15]
−0.60 [−1.30, 0.11]
−0.06 [−0.36,0.24]
0.00 [−0.75,0.75]
0.55 [−0.46,1.55]
0.00 [−0.74,0.74]
−0.24 [−0.90,0.43]
0.45 [−0.06,0.97]
0.05 [−0.17, 0.26 ]
13 8.0%
8 4.5%
14 8.3%
31 17.0%
51.7%
89
10.3%
21
100.0%176
21 25.7%
15.7%
22.5%
36.1%
13 8.0%
8 4.7%
14 8.1%
21 10.4%
100.0%131
51.5%
89
17.4%
31
100.0%176
1.29 [0.18, 2.40]
0.17 0.41 89
13
8
89
13
8
14
21
31
176
20.0%
12.6%
14.5%
16.7%
17.6%
18.6%
100.0%
−10
4.87
−0.1
−0.02
−3
−7.87
0.05
0.1
−12
4.12
−0.52
0.09
−0.09
−0.2
−0.46
0.08
−1
1.75
0
0
0.11
−0.75 [−1.53,0.04]
−0.13 [−0.79,0.54]
0.01 [−0.55, 0.57]
18.39
12.47
0.98
0.54
25.58
51.52
0.32
1.02
0.38
0.29
14.22
18.21
0.17
0.59
0.6
0.15
4.79
8.51
0.05
0.4
0.22
85
14
8
15
122
85
14
8
14
15
28
164
85
14
8
14
15
28
164
85
14
8
14
15
28
164
0.16
2
−9.5
0
−0.03
−4
1.25
0.22
0.1
−0.05
0.07
29.38
−6.88
−0.34
−0.1
−0.01
0.09
−1
−3.25
0
0.1
0.02
0.44
11.74
8.1
0.6
0.27
29.38
40.62
0.43
1.03
0.38
0.37
8.15
10.88
0.29
0.65
0.71
0.17
4.62
8.78
0.05
0.42
0.17
2.3.1 TC
Conraads et al. 2015
Fu et al. 2013
Iellamo et al. 2013
Madssen et al. 2014
Subtotal (95% CI)
2.3.2 TG
Cardozo et al. 2015
Fu et al. 2013
Kim et al. 2015
Madssen et al. 2014
Moholdt et al. 2009
Subtotal (95% CI)
2.3.3 LDL-C
Conraads et al. 2015
Fu et al. 2013
Kim et al. 2015
Madssen et al. 2014
Moholdt et al. 2009
Subtotal (95% CI)
Subtotal (95% CI)
2.3.4 HDL-C
Conraads et al. 2015
Fu et al. 2013
Kim et al. 2015
Madssen et al. 2014
Moholdt et al. 2009
INTERVAL CONTINUOUS
Study or subgroup Std. mean dierence
Weight
Tot a lSDMean TotalSDMean IV, random, 95% CI
Heterogeneity: 𝜏2=0.00; 𝜒2=1.43, df =5(p=0.92); I2=0%
Test for overall eect: Z = 0.43 (p = 0.67)
Heterogeneity: 𝜏2=0.63; 𝜒2=36.59, df =5(p0.00001);I2=86%
Test for overall eect: Z = 1.65 (p = 0.10)
Heterogeneity: 𝜏2= 0.00;𝜒2= 4.58,df=5(p=0.47); I2=0%
Test for overall eect: Z = 0.42 (p = 0.68)
Test for subgroup dierences: 𝜒2=2.98, df =3(p = 0.39), I2=0%
Heterogeneity: 𝜏2=0.20; 𝜒2= 8.87,df=3(p=0.03); I2=66%
Test for overall eect: Z = 0.04 (p = 0.97)
Favours
[INTERVAL]
Favours
[CONTINUOUS]
420−2
−4
Std. mean dierence
Iellamo et al. 2013
Iellamo et al. 2013
Iellamo et al. 2013
<
IV, random, 95% CI
F : Meta-analysis of eects of INTERVAL on blood lipid.
Angadi et al. 2015
Benda et al. 2015
Conraads et al. 2015
Currie et al. 2013
Madssen et al. 2014
Wislø et al. 2007
Total (95% CI) 130 139 100.0% 1.47 [−0.20, 3.14]
4.80 [2.47,7.13]
−0.20 [−1.55,1.15]
0.14 [−0.38,0.66]
0.00 [−1.71,1.71]
−0.30 [−9.55,8.95]
3.57 [2.28,4.86]
Heterogeneity: 𝜏2= 3.16;𝜒2= 37.76,df=5(p < 0.00001); I2=87%
Test for overall eect: Z = 1.72 (p = 0.09)
15.9%
6
2.23
20.0%
101.63
22.5%
841.69
18.5%
102
2.9%
2118.56
20.2%
81.46
−4.792.290.1
−0.4101.45−0.6
1.07761.661.21
1.5111.991.5
3.2159.382.9
4.6991.228.26
Weight
Tot a lSDMeanTo t a lSDMean
Study or subgroup Mean dierence Mean dierenceCONTINUOUS
INTERVAL
Favours [CONTINUOUS]
50 10−5
−10
Favours [INTERVAL]
IV, random, 95% CI IV, random, 95% CI
F : Meta-analysis of eects of INTERVAL on FMD.

 BioMed Research International
Heterogeneity: 𝜏2= 10.86;𝜒2=26.30,df=7(p=0.0004); I2=73%
Test for overall eect: Z = 1.57 (p = 0.12)Favours
[CONTINUOUS]
5010−5
−10
Favours
[INTERVAL]
2.33 9 5.21 6 1 12.0%
3.26 9 2.72 6 14.5%
4 4.92 10 0 2.72 10 13.7%
10.3 7.23 14 4.5 11.55 13 7.7%
0.87 3.02 8 0.6 3.79 8 14.0%
0 5.46 19 2 4.9 19 14.1%
4.5 4.88 12 2.9 6.39 10 11.4%
10 5.4 9 0.7 3.09 8 12.6%
INTERVAL CONTINUOUS
Study or subgroup Weight
Tot a lSDMeanTo t a lSDMe an
Mean dierence Mean dierence
Wislø et al. 2007
−3.00 [−7.44,1.44]
3.00 [−0.05,6.05]
4.00 [0.52,7.48]
5.80 [−1.53,13.13]
0.27 [−3.09,3.63]
−2.00 [−5.30,1.30]
1.60 [−3.23,6.43]
9.30 [5.17,13.43]
Total (95% CI) 90 80 100.0% 2.18 [−0.54, 4.90]
−2
−2 −5
Amundsen et al. 2008
Angadi et al. 2015
Benda et al. 2015
Fu et al. 2013
Iellamo et al. 2013
Moholdt et al. 2009
Ulbrich et al. 2016
IV, random, 95% CIIV, random, 95% CI
F : Meta-analysis of eects of INTERVAL on LVEF.
LVEF were small and may not be clinically meaningful. e
resultsofthisanalysisshouldbeinterpretedwithcautiondue
to the small sample. Accordingly, more high-quality, large-
sample, multicenter, long-term randomized interventional
studies are needed to assess the eects of INTERVAL in
cardiac patients.
Appendix
See Figures –.
Competing Interests
e authors declare that there are no competing interests
regarding the publication of this paper.
References
[] S. Mendis, P. Puska, and B. Norrving, Eds., Global Atlas on
Cardiovascular Disease Prevention and Control,WorldHealth
Organization, Geneva, Switzerland, .
[] E.M.Antman,D.T.Anbe,P.W.Armstrongetal.,“ACC/AHA
guidelines for the management of patients with ST-elevation
myocardial infarction—executive summary. A report of the
American College of Cardiology/American Heart Association
Task Force on Practice Guidelines (Writing Committee to revise
the  guidelines for the management of patients with acute
myocardial infarction),” Journal of the American College of
Cardiology,vol.,no.,pp.–,.
[] S. A. Hunt, W. T. Abraham, M. H. Chin et al., “ACC/AHA
 Guideline Update for the Diagnosis and Management of
Chronic Heart Failure in the Adult: a report of the American
College of Cardiology/American Heart Association Task Force
on Practice Guidelines (Writing Committee to Update the
 Guidelines for the Evaluation and Management of Heart
Failure): developed in collaboration with the American College
of Chest Physicians and the International Society for Heart and
Lung Transplantation: endorsed by the Heart Rhythm Society,”
Circulation,vol.,no.,pp.e–e,.
[]G.A.Kelley,K.S.Kelley,andZ.VuTran,“Aerobicexercise,
lipids and lipoproteins in overweight and obese adults: a meta-
analysis of randomized controlled trials,” International Journal
of Obesity,vol.,no.,pp.–,.
[] V. A. Corneliss en and R. H. Fagard, “Eect of resistance training
on resting blood pressure: a meta-analysis of randomized
controlled trials,” Journal of Hypertension,vol.,no.,pp.–
, .
[] S. J. Keteyian, C. A. Brawner, P. D. Savage et al., “Peak aerobic
capacity predicts prognosis in patients with coronary heart
disease,” American Heart Journal,vol.,no.,pp.–,
.
[]H.Brunner,J.R.Cockcro,J.Deaneldetal.,“Endothelial
function and dysfunction. Part II: association with cardiovascu-
lar risk factors and diseases. A statement by the Working Group
on Endothelins and Endothelial Factorsof the European Society
of Hypertension,” Journal of Hypertension,vol.,no.,pp.
–, .
[] G. F. Fletcher, P. A. Ades, P. Kligeld et al., “Exercise standards
for testing and training: a scientic statement from the Ameri-
can heart association,” Circulation,vol.,no.,pp.–,
.
[] A.Z.Ulbrich,V.G.Angarten,A.SchmittNettoetal.,“Compar-
ative eects of high intensity interval training versus moderate
intensity continuous training on quality of life in patients with
heart failure: study protocol for a randomized controlled trial,”
Clinical Trials and Regulatory Science in Cardiology,vol.,pp.
–, .
[] V. M. Conraads, N. Pattyn, C. De Maeyer et al., “Aerobic inter-
val training and continuous training equally improve aerobic
exercise capacity in patients with coronary artery disease: the
SAINTEX-CAD study,” International Journal of Cardiology,vol.
,pp.–,.
[] M. J. Gibala and S. L. McGee, “Metabolic adaptations to short-
term high-intensity interval training: a little pain for a lot of
gain?” Exercise and Sport Sciences Reviews,vol.,no.,pp.
–, .
[]U.Wislø,A.Støylen,J.P.Loennechenetal.,“Superiorcar-
diovascular eect of aerobic interval training versus moderate
continuous training in heart failure patients: a randomized
study,” Circulation,vol.,no.,pp.–,.

BioMed Research International 
[] K. D. Currie, J. B. Dubberley, R. S. McKelvie, and M. J. Macdon-
ald, “Low-volume, high-intensity interval training in patients
with CAD,” Medicine and Science in Sports and Exercise,vol.
, no. , pp. –, .
[] S. Ito, T. Mizoguchi, and T. Saeki, “Review of high-intensity
interval training in cardiac rehabilitation,” Internal Medicine,
vol.,no.,pp.–,.
[] F. Giallauria, N. A. Smart, A. Cittadini, and C. Vigorito,
“Exercise training modalities in chronic heart failure: does
high intensity aerobic interval training make the dierence?”
Monaldi Archives for Chest Disease,vol.,no.-,.
[] A. D. Elliott, K. Rajopadhyaya, D. J. Bentley, J. F. Beltrame, and
E. C. Aromataris, “Interval training versus continuous exercise
in patients with coronary artery disease: a meta-analysis,” Heart
Lung and Circulation,vol.,no.,pp.–,.
[] J. P. T. Higgins and S. Green, Cochrane Handbook for Systematic
Reviews of Interventions Version 5.1.0, e Cochrane Collabora-
tion, , http://handbook.cochrane.org/.
[] “eGeorgeInstituteforGlobalHealthandtheUniversityof
Sydney. Physiotherapy Evidence Database,” , http://www
.pedro.org.au/english/downloads/pedro-scale/.
[] B. H. Amundsen, Ø. Rognmo, G. Hatlen-Rebhan, and S. A.
Slørdahl, “High-intensity aerobic exercise improves diastolic
function in coronary artery disease,” Scandinavian Cardiovas-
cular Journal,vol.,no.,pp.–,.
[] S. S. Angadi, F. Mookadam, C. D. Lee, W. J. Tucker, M. J.
Haykowsky, and G. A. Gaesser, “High-intensity interval training
vs. moderate-intensity continuous exercise training in heart
failure with preserved ejection fraction: a pilot study,” Journal
of Applied Physiology,vol.,no.,pp.–,.
[] N. M. Benda, J. P. Seeger, G. G. Stevens et al., “Eects of
high-intensity interval training versus continuous training on
physical tness, cardiovascular function and quality of life in
heart failure patients,” PLOS ONE,vol.,no.,.
[] G. G. Cardozo, R. B. Oliveira, and P. T. V. Farinatti, “Eects of
high intensity interval versus moderate continuous training on
markers of ventilatory and cardiac eciency in coronary heart
disease patients,” e Scientic World Journal,vol.,Article
ID ,  pages, .
[]S.Dimopoulos,M.Anastasiou-Nana,D.Sakellariouetal.,
“Eectsofexerciserehabilitationprogramonheartraterecov-
ery in patients with chronic heart failure,” European Journal of
Cardiovascular Prevention & Rehabilitation,vol.,no.,pp.–
, .
[] C. Freyssin, C. Verkindt, F. Prieur, P. Benaich, S. Maunier, and
P. Blanc, “Cardiac rehabilitation in chronic heart failure: eect
of an -week, high-intensity interval training versus continuous
training,” Archives of Physical Medicine and Rehabilitation,vol.
,no.,pp.–,.
[] T.-C. Fu, C.-H. Wang, P.-S. Lin et al., “Aerobic interval training
improves oxygen uptake eciency by enhancing cerebral and
muscular hemodynamics in patients with heart failure,” Inter-
national Journal of Cardiology,vol.,no.,pp.–,.
[] F. Iellamo, V. Manzi, G. Caminiti et al., “Matched dose interval
and continuous exercise training induce similar cardiorespira-
tory and metabolic adaptations in patients with heart failure,”
International Journal of Cardiology,vol.,no.,pp.–
, .
[] K. V. Jaureguizar, D. Vicente-Campos, L. R. Bautista et al.,
“Eect of high-intensity interval versus continuous exercise
training on functional capacity and quality of life in patients
with coronary artery disease: a randomized clinical trial,”
JournalofCardiopulmonaryRehabilitationandPrevention,vol.
,no.,pp.–,.
[] S. J. Keteyian, B. A. Hibner, K. Bronsteen et al., “Greater im-
provement in cardiorespiratory tness using higher-intensity
interval training in the standard cardiac rehabilitation setting,”
JournalofCardiopulmonaryRehabilitationandPrevention,vol.
, no. , pp. –, .
[] C. Kim, H. E. Choi, and M. H. Lim, “Eect of high interval train-
ing in acute myocardial infarction patients with drug-eluting
stent,” AmericanJournalofPhysicalMedicineandRehabilitation,
vol. , no. , pp. –, .
[]P.Koufaki,T.H.Mercer,K.P.George,andJ.Nolan,“Low-
volume high-intensity interval training vs continuous aerobic
cycling in patients with chronic heart failure: a pragmatic
randomised clinical trial of feasibility and eectiveness,” Journal
of Rehabilitation Medicine, vol. , no. , pp. –, .
[] E. Madssen, T. Moholdt, V. Videm, U. Wislø, K. Hegbom, and
R. Wiseth, “Coronary atheroma regression and plaque charac-
teristics assessed by grayscale and radiofrequency intravascular
ultrasound aer aerobic exercise,” American Journal of Cardiol-
ogy,vol.,no.,pp.–,.
[ ] T. T. Mo h o l d t , B . H . A mundsen, L. A. Ru s t a d e t a l . , “Aero b i c
interval training versus continuous moderate exercise aer
coronary artery bypass surgery: a randomized study of cardio-
vascular eects and quality of life,” American Heart Journal,vol.
, no. , pp. –, .
[] D. Prado, E. Rocco, A. Silva et al., “Eects of continuous vs
interval exercise training on oxygen uptake eciency slope
in patients with coronary artery disease,” Brazilian Journal of
Medical and Biological Research,vol.,no.,ArticleIDe,
.
[]E.A.Rocco,D.M.L.Prado,A.G.Silvaetal.,“Eectof
continuous and interval exercise training on the PETCO
response during a graded exercise test in patients with coronary
artery disease,” Clinics,vol.,no.,pp.–,.
[] P. Roditis, S. Dimopoulos, D. Sakellariou et al., “e eects of
exercise training on the kinetics of oxygen uptake in patients
with chronic heart failure,” European Journal of Cardiovascular
Prevention & Rehabilitation,vol.,no.,pp.–,.
[] Ø.Rognmo,E.Hetland,J.Helgerud,J.Ho,andS.A.Slørdahl,
“High intensity aerobic interval exercise is superior to moderate
intensity exercise for increasing aerobic capacity in patients
with coronary artery disease,” European Journal of Cardiovas-
cular Prevention and Rehabilitation,vol.,no.,pp.–,
.
[] D. E. R. Warburton, D. C. McKenzie, M. J. Haykowsky et al.,
“Eectiveness of high-intensity interval training for the reha-
bilitation of patients with coronary artery disease,” American
Journal of Cardiology,vol.,no.,pp.–,.
[] M. J. Haykowsky, M. P. Timmons, C. Kruger, M. McNeely, D.
A. Taylor, and A. M. Clark, “Meta-analysis of aerobic interval
training on exercise capacity and systolic function in patients
with heart failure and reduced ejection fractions,” American
Journal of Cardiology, vol. , no. , pp. –, .
[] H. A. Wenger and G. J. Bell, “e interactions of intensity,
frequency and duration of exercise training in altering car-
diorespiratory tness,” Sports Medicine,vol.,no.,pp.–
, .
[] U.Wislø,Ø.Ellingsen,andO.J.Kemi,“High-intensityinterval
training to maximize cardiac benets of exercise training?”

 BioMed Research International
Exercise and Sport Sciences Reviews,vol.,no.,pp.–,
.
[] D. Hoshino, Y. Yoshida, Y. Kitaoka, H. Hatta, and A. Bonen,
“High-intensity interval training increases intrinsic rates of
mitochondrial fatty acid oxidation in rat red and white skeletal
muscle,” Applied Physiology, Nutrition, and Metabolism,vol.,
no. , pp. –, .
[]A.E.Tjønna,S.J.Lee,Ø.Rognmoetal.,“Aerobicinterval
training versus continuous moderate exercise as a treatment for
the metabolic syndrome: A Pilot Study,” Circulation,vol.,no.
, pp. –, .
[] M. J. Gibala, J. P. Little, M. J. Macdonald, and J. A. Hawley,
“Physiological adaptations to low-volume, high-intensity inter-
val training in health and disease,” Journal of Physiology,vol.
, no. , pp. –, .
[] A. S. Leon, B. A. Franklin, F. Costa et al., “Cardiac rehabil-
itation and secondary prevention of coronary heart disease:
an American Heart Association scientic statement from the
Council on Clinical Cardiology (Subcommittee on Exercise,
Cardiac Rehabilitation, and Prevention) and the Council on
Nutrition, Physical Activity, and Metabolism (Subcommittee
on Physical Activity), in collaboration with the American
association of Cardiovascular and Pulmonary Rehabilitation,”
Circulation, vol. , no. , pp. –, .
[] P. A. Ades, N. M. Green, and C. E. Coello, “Eects of exercise
and cardiac rehabilitation on cardiovascular outcomes,” Cardi-
ology Clinics,vol.,no.,pp.–,.
[] C.J.Lavie,R.J.omas,R.W.Squires,T.G.Allison,andR.V.
Milani, “Exercise training and cardiac rehabilitation in primary
and secondary prevention of coronary heart disease,” Mayo
Clinic Proceedings,vol.,no.,pp.–,.
[] F.Roveda,H.R.Middlekau,M.U.P.B.Rondonetal.,“e
eects of exercise training on sympathetic neural activation in
advanced heart failure: a randomized controlled trial,” Journal
of the American College of Cardiolog y,vol.,no.,pp.–,
.
[] T. Guiraud, A. Nigam, V. Gremeaux, P. Meyer, M. Juneau, and
L. Bosquet, “High-intensity interval training in cardiac rehabil-
itation,” Sports Medicine,vol.,no.,pp.–,.
[] N. Oldridge, “Exercise-based cardiac rehabilitation in patients
with coronary heart disease: meta-analysis outcomes revisited,”
Future Cardiology,vol.,no.,pp.–,.
[] N. A. Smart, G. Dieberg, and F. Giallauria, “Intermittent versus
continuous exercise training in chronic heart failure: a meta-
analysis,” International Journal of Cardiology,vol.,no.,pp.
–, .
[] N. Pattyn, E. Coeckelberghs, R. Buys, V. A. Cornelissen, and
L. Vanhees, “Aerobic interval training vs. moderate continuous
training in coronary artery disease patients: a systematic review
and meta-analysis,” Sports Medicine,vol.,no.,pp.–,
.