See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/282038947

High Intensity Interval versus Moderate Intensity Continuous Training in

Patients with Coronary Artery Disease: A Meta-analysis of Physiological and
Clinical Parameters

Article  in  Heart, Lung and Circulation · September 2015

DOI: 10.1016/j.hlc.2015.06.828

CITATIONS READS

42 406

4 authors, including:

Kevin Liou Sze-Yuan Ooi

28 PUBLICATIONS   130 CITATIONS   
Prince of Wales Hospital and Community Health Services
60 PUBLICATIONS   161 CITATIONS   
SEE PROFILE
SEE PROFILE

All content following this page was uploaded by Kevin Liou on 13 November 2017.

The user has requested enhancement of the downloaded file.

Heart, Lung and Circulation (2016) 25, 166–174 ORIGINAL ARTICLE
1443-9506/04/$36.00
http://dx.doi.org/10.1016/j.hlc.2015.06.828

High Intensity Interval versus Moderate

Intensity Continuous Training in Patients
with Coronary Artery Disease:
A Meta-analysis of Physiological and
Clinical Parameters
Kevin Liou, MPH, FRACP a,c*, Suyen Ho, FANZCA b,
Jennifer Fildes, RN, CNC d, Sze-Yuan Ooi, MD, FRACP a,c
a
Eastern Heart Clinic, Prince of Wales Hospital, Sydney, NSW, Australia
b
Department of Anaesthetics, Royal Prince Alfred Hospital and St George Hospital, Sydney, NSW, Australia
c
Prince of Wales Clinical School, University of New South Wales, Sydney, NSW, Australia
d
Cardiac Rehabilitation Unit, Prince of Wales Hospital, Sydney, NSW, Australia

Received 5 April 2015; received in revised form 14 June 2015; accepted 16 June 2015; online published-ahead-of-print 22 July 2015

Introduction Exercise-based cardiac rehabilitation for patients with coronary artery disease (CAD) significantly improves
their outcome, although the optimal mode of exercise training remains undetermined. Previous analyses
have been constrained by small sample sizes and a limited focus on clinical parameters. Further, results
from previous studies have been contradicted by a recently published large RCT.
Method We performed a meta-analysis of published randomised controlled trials to compare high intensity interval
training (HIIT) and moderate intensity continuous training (MCT) in their ability to improve patients’
aerobic exercise capacity (VO2peak) and various cardiovascular risk factors. We included patients with
established coronary artery disease without or without impaired ejection fraction.
Results Ten studies with 472 patients were included for analyses (218 HIIT, 254 MCT). Overall, HIIT was associated
with a more pronounced incremental gain in participants’ mean VO2peak when compared with MCT
(+1.78 mL/kg/min, 95% CI: 0.45-3.11). Moderate intensity continuous training however was associated
with a more marked decline in patients’ mean resting heart rate (-1.8/min, 95% CI: 0.71-2.89) and body
weight (-0.48 kg, 95% CI: 0.15-0.81). No significant differences were noted in the level of glucose, triglyceride
and HDL at the end of exercise program between the two groups.
Conclusion High intensity interval training improves the mean VO2peak in patients with CAD more than MCT, although
MCT was associated with a more pronounced numerical decline in patients’ resting heart rate and body
weight. The underlying mechanisms and clinical relevance of these results are uncertain, and remain a
potential focus for future studies.
Keywords Interval training  Continuous training  Coronary artery disease  VO2peak

*Corresponding author at: Eastern Heart Clinic, Prince of Wales Hospital, Barker Street, Randwick, NSW 2031, Australia. Tel.: +61 2 93820700;
fax: +61 2 93820799, Email: Kevin.Liou@SESIAHS.HEALTH.NSW.GOV.AU
Crown Copyright © 2015 Published by Elsevier Inc on behalf of Australian and New Zealand Society of Cardiac and Thoracic Surgeons (ANZSCTS) and the Cardiac Society of Australia
and New Zealand (CSANZ). All rights reserved.
HIIT v.s. MCT in Patients with CAD 167

studies, specifically in relation to the patient characteristics

Introduction and the type, intensity and duration of continuous training
Exercise based cardiac rehabilitation for patients with coro- also limits the interpretation and clinical application of these
nary artery disease significantly improves their outcome [1]. results. Further, most studies have focussed predominantly
This in part may be attributed to the improvement in on changes in patients’ physiological characteristics, and not
patients’ aerobic exercise capacity [2,3]. The optimal exercise their clinical parameters and risk factor profile. Lastly, no
regime has yet to be determined however, and many have meta-analyses to date have included the largest multi-cen-
attempted to compare the efficacy of high-intensity interval tred randomised controlled trial (RCT) [6] recently pub-
training (HIIT) against moderate intensity continuous train- lished, which produced a contradictory result to previous
ing (MCT) in their ability to improve patients’ aerobic exer- analyses. Herein we present the results of our meta-analyses
cise capacity as measured by peak oxygen consumption which include the latest RCT rendering it the largest analysis
(VO2peak). Most recently, Elliott et al. have shown HIIT to to date. We also focussed on both physiological as well as
be superior to MCT in improving the VO2peak of patients with clinical parameters to improve the clinical relevance of our
coronary artery disease [4]. This echoed the findings of a analyses.
previous meta-analysis in this patient cohort [5], as well as
meta-analyses undertaken in other patient groups [22].
While these studies have been instrumental in our under- Methods
standing of exercise physiology particularly in the cardiac A systematic literature search was performed by the primary
population, the number of subjects in each individual study author in May 2015 using Ovid Medline and Embase with no
has been relatively small. The heterogeneity among the date restriction. A combined search strategy was employed

Figure 1 PRISMA flow diagram depicting the study selection process.

168 K. Liou et al.

Table 1 Inclusion and exclusion criteria.

Inclusion Criteria Exclusion Criteria

1. Randomised controlled trials where the physiological and clinical outcomes of 1. Studies where incomplete data were reported
interest were directly compared between subjects undergoing HIIT and MCT, 2. Studies containing patients not representative
respectively of the cohort of interest
2. The duration of HIIT and MCT must be at least 4 weeks or more 3. Conference abstracts
3. Studies containing patients with documented coronary artery disease with or 4. Studies published in language other than
without left ventricular impairment English
4. Studies must contain raw data for retrieving directly or permitting indirect
derivation of outcomes of interest as well as the associated 95% confidential
intervals

using the following search terms: (‘‘interval training’’ OR which allowed reporting of their pooled results. Only param-
‘‘interval exercise’’ OR ‘‘high intensity training’’ OR ‘‘high eters reported by at least four studies were analysed. Anal-
intensity exercise’’) AND (‘‘continuous training’’ OR ‘‘contin- yses were performed with Review Manager Ver 5.3.
uous exercise’’ OR ‘‘moderate intensity training’’ OR ‘‘mod-
erate intensity exercise’’) AND (‘‘coronary artery disease’’ OR
‘‘ischaemic heart disease’’ OR ‘‘ischemic heart disease’’ OR
‘‘coronary heart disease’’ OR ‘‘myocardial infarction’’). Fur-
Results
ther, the reference lists of relevant studies were reviewed for From 322 citations generated, 10 studies [6–15] with 472
additional studies addressing this subject. We followed the patients were eventually included for analyses (218 HIIT,
Preferred Reporting Items for Systematic reviews and Meta- 254 MCT). The study characteristics are listed in Table 2.
Analyses (PRISMA) guidelines where possible in performing The duration of exercise program ranged from 4-16 weeks
our systematic review (Figure and Table 1). (median: 12). There was no significant publication bias (Fig-
Heterogeneity among studies was examined with ure 2; p: 0.16). The quality of the included studies was
Cochran’s Q and I2 statistic. Publication bias is assessed reasonable (Figure 3). Overall, HIIT was associated with a
based on study distribution on the funnel plot, and Egger’s more pronounced incremental rise in VO2peak when com-
regression. Due to the anticipated heterogeneous nature of pared with MCT (+1.78 mL/kg/min, 95% CI: 0.45-3.11,
the included studies, we have decided a priori to analyse the p=0.009; Figure 4A, 4B and 4C), although significant hetero-
data with the random effect model. Sensitivity analysis was geneity was observed among the studies included (I2: 93%,
also performed to identify the most homogenous studies, p<0.01) (Figure 4). When sensitivity analysis was performed,

Figure 2 Funnel plot for assessment of publication bias.

 HIIT v.s. MCT in Patients with CAD
Table 2 Study characteristics.
Study Inclusion Criteria Exclusion Criteria Patient Number Patient Age (Years) Exercise program Exercise Attrition Rate
Duration
HIIT MCT HIIT MCT HIIT MCT HIIT MCT
(weeks)

Conraads 1) Angiographically documented NR 85 89 57 59.9 3 times a week 3 times a week 12 15% 11%
et al. (2015) CAD or previous AMI 90-95% 70-75% of peak HR
2) LVEF > 40% of peak HR on on bicycle
3) On optimal medical treatment bicycle
4) Stable with regard to symptoms
and medication for at least 4 weeks
5) Included between 4 and
12 weeks following AMI, PCI or
CABG
Keteyian 1) Men and women between the NR 15 13 60 58 3 times a week 3 times a week 10 29% 28%
et al. (2014) ages of 18 and 75 years 80-90% of peak HR 60-80% of peak HR
participating in phase 2 CR on treadmill on bicycle
programs
2) Sinus rhythm
3) LVEF > 40%
4) > 3 weeks following myocardial
infarction or percutaneous coronary
intervention
5) > 4 weeks after coronary artery
bypass surgery
6) Successful completion of a run-in
period (ie, attending  4 of the first
6 scheduled standard CR sessions
without complications using MCT)
7) Free of a comorbidity that might
limit treadmill exercise
Iellamo 1) Post-infarction heart failure 1) Unstable angina or recent acute 8 8 62.2 62.6 2 days a week for 2 days a week for 12 10% 10%
et al. (2013) 2) LVEF < 40% myocardial infarction (less than the first 3 weeks, the first 3 weeks,
3) Symptomatic heart failure with 6 months) 3 days a week for 3 days a week for
functional NYHA class II or III 2) Pacemaker the second 3 weeks, the second 3 weeks,
4) Clinical stability without hospital 3) Uncontrolled hypertension 4 days a week for 4 days a week for
admission for HF in the previous 4) History of severe kidney diseases, the third 3 weeks, the third 3 weeks,
3 months lung disease, severe lower extremities 5 days a week for 5 days a week for
5) Sinus rhythm vascular or other diseases which the last 3 weeks. the last 3 weeks.
6) On optimal medical treatment could prevent a symptom limited 45-60% of peak HR 75-80% of peak HR
exercise test on treadmill on treadmill
5) Coexisting valvular disease
6) Insulin-dependent diabetes

169
 170
Table 2. (continued)
Study Inclusion Criteria Exclusion Criteria Patient Number Patient Age (Years) Exercise program Exercise Attrition Rate
Duration
HIIT MCT HIIT MCT HIIT MCT HIIT MCT
(weeks)
Currie Patients with recent CAD event that 1) Smoking within 3 months 11 10 62 68 2 days a week 2 days a week 12 27% 27%
et al. (2013) was defined as the patient having 2) Non-cardiac surgical procedure 80-104% peak 51-65% peak power
at least one of the following: within 2 months power output on output on bicycle
1) Angiographically documented 3) Myocardial infarction or coronary bicycle
stenosis > 50% in at least one majorartery bypass graft within 2 months
coronary artery 4) Percutaneous coronary intervention
2) Myocardial infarction within 1 month
3) Percutaneous coronary 5) New York Heart Association class
intervention II-IV symptoms of heart failure
6) Documented valve stenosis
7) Documented severe chronic
obstructive pulmonary disease
8) Symptomatic peripheral arterial
disease
9) Unstable angina
10) Uncontrolled hypertension
11) Uncontrolled atrial arrhythmia or
ventricular dysrhythmia
12) Any musculoskeletal abnormality
that would limit exercise participation
Rocco Stable coronary artery disease 1) Unstable angina pectoris 17 20 56.5 62.5 3 sessions per week 3 sessions per week 12 0% 0%
et al. (2012) diagnosed by coronary angiography 2) Complex ventricular arrhythmias Moderate Intensity Treadmill
3) Pulmonary congestion Treadmill
4) Orthopaedic or neurological
limitations to exercise
Moholdt Patients who had suffered a MI two 1) LVEF <30% 30 59 56.7 57.7 2 sessions at the 2 sessions at the 12 14% 18%
et al. (2011) to 12 weeks ago in three Norwegian 2) Contraindications to vigorous hospital, 1 session hospital, 1 session
hospitals were candidates for physical activity at home per week at home per week
inclusion in the study 3) Pulmonary disease clearly limiting 85-95% peak HR Group/casual
exercise capacity Various activity exercise
4) Drug abuse. Various activity
Moholdt Coronary artery bypass grafting 1) Heart failure 28 31 60.2 62 5 days a week 5 days a week 4 12% 14%
et al. (2009) patients referred to a residential 2) Inability to exercise 90% peak HR 70% peak HR
rehabilitation centre 4 to 16 weeks 3) Drug abuse. Treadmill Treadmill
postoperatively were included
Wisloff Post-infarction heart failure MI within 12 months 9 8 76.5 74.4 2 sessions at the 2 sessions at the 12 0% 11%
et al. (2007) hospital, 1 session hospital, 1 session
at home per week at home per week
90-95% peak HR 70-75% peak HR

K. Liou et al.
Treadmill Treadmill
 HIIT v.s. MCT in Patients with CAD
Warburton 1) Men with CAD who had NR 7 7 55 57 2 days a week 2 days a week 65% 16 0% 0%
et al. (2005) undergone (6 months previously) 90% VO2 reserve VO2 reserve
bypass surgery or angioplasty Additional 3 day a Additional 3 days a
2) Negative stress test week @ week @
65% VO2 reserve 65% VO2 reserve
Various activity Various activity
Rognmo 1) Angiographically documented 1) Left main coronary artery disease 8 9 62.9 61.2 3 times a week 3 times a week 10 27% 10%
et al. (2004) CAD in at least one major months 80-90% of peak HR 50-60% of peak HR
epicardial vessel. 2) Unstable angina pectoris on treadmill on treadmill
2) Previous MI 3) Claudication
3) Previous CABG 4) MI within the last 3 months
4) Previous PCI 5) Positive stress 5) CABG or PCI performed within
test the last 12 months
6) Complex ventricular arrhythmia
7) LVEF < 40%
8) Orthopaedic or neurological
limitations to exercise
9) Regular exercise for the past
3 months

Abbreviations: CABG: cardiac artery bypass graft surgery; CAD: coronary artery disease; CR: cardiac rehabilitation; HIIT: high intensity interval training; HR: heart rate; LVEF: left ventricular ejection fraction;
MCT: moderate intensity continuous training; MI: myocardial infarction; NR: not reported; PCI: percutaneous coronary intervention.
Figure 3 Risk of bias summary.

171
172 K. Liou et al.

Figure 4 A) Changes in VO2peak by Mode of Exercise Training; B) Changes in Body Mass by Mode of Exercise Training; C)
Changes in Resting Heart Rate by Mode of Exercise Training. The size of the square reflects the weighting of each of the
included study.

a significant improvement in study homogeneity (ie reduced

variance) was observed when Wisloff (2007) [13] is excluded
Discussion
from the analysis (Figure 5A & 5B, I2: 0%, p: 0.62). While the Our results echoed the findings by Elliott et al. in demonstrat-
result remains statistically significant, its numerical value ing a more pronounced, albeit numerically small, improve-
appears more modest when compared to the overall analysis. ment in patients’ mean VO2peak with HIIT. There is, however,
MCT was associated with a more marked decrease in no significant benefit seen in patients’ metabolic profile while
resting heart rate (-1.8/min, 95% CI: 0.71-2.89; Figure 1B) paradoxically, patients tended to lose more weight and have a
and body weight (-0.48 kg, 95% CI: 0.15-0.81; Figure 1C). Both greater reduction in resting heart rate with MCT.
results were highly consistent among included studies As the rehabilitation process and follow-up duration for all
(I2=0%, P=NS). While there is a strong numerical trend in these studies were short, it is likely that some of the metabolic
favour of HIIT, no statistically significant differences were and clinical consequences of HIIT may be borne out over time
noted in the level of glucose (Figure S1), HDL (Figure S2) and with a sustained effort and commitment. It has also been
triglycerides (Figure S3) at the end of exercise program shown that exercise volume and duration correlate with fat
between the two groups. loss [16], potentially accounting for the increased weight loss
HIIT v.s. MCT in Patients with CAD 173

Figure 5 A) Sensitivity analysis demonstrating consistency in the incremental gain in VO2peak with each study removed; B)
Changes in VO2peak by Mode of Exercise Training with Wisloff (2007) excluded.

observed in the MCT cohort, although it is also possible that clinical outcome in the long term. Specifically, a cost-effective-
patients’ muscle mass increased with HIIT thus leading to a ness study has not been performed. Similarly, while the attri-
paradoxical gain in body mass. Resting bradycardia has been tion rate is similar in each study arm (Table 2), patients’
associated with improved cardiovascular outcome [17], adherence to each strategy after program cessation has not
although the mechanism underlying training induced brady- been thoroughly evaluated. Given the significant health and
cardia is not clear. Some believe it is related to enhanced economic considerations associated with the cardiac rehabili-
parasympathetic regulation [18], although the nature of its tation programs in Australia [20] and worldwide, it is impor-
interactions with the type, volume and intensity of the training tant to address these issues with either a nationwide registry or
programs has yet to be defined. The magnitude of difference in large, well designed randomised controlled trials with ade-
terms of resting heart rate between HIIT and MCT is small and quate mechanisms for long-term follow-up. For the time being,
its significance in patients’ overall prognosis currently is particularly in the absence of a safety signal [21], it would seem
uncertain. reasonable to tailor the exercise programs to individuals’ cir-
Overall, the clinical relevance of these findings remains cumstances and institutional practice and protocols.
unclear as no studies were powered to examine the effect of
the mode of training on hard clinical endpoints such as major
adverse cardiovascular events or death. The differences in
Conclusion
analysed outcomes were also small and may not be clinically HIIT improves the mean VO2peak in patients with CAD more
meaningful. Significant separation in clinical endpoints may than MCT, although MCT was associated with a more pro-
become apparent over time, particularly as HIIT has been nounced numerical decline in patients’ resting heart rate and
shown to improve patients’ vascular function and drivers of body weight. The magnitude in the treatment effects how-
atherosclerosis [19]. However, none of these studies were ever is small, and the clinical significance of our findings
designed to examine the feasibility of a maintenance exercise remains uncertain. Larger and longer term studies are
program and its effect on patients’ overall physiological and required to address the deficiencies in the current evidence
174 K. Liou et al.

base, as the associated health related and economic implica- response during a graded exercise test in patients with coronary artery
disease. Clinics 2012;67:623–7.
tions can be significant. [11] Moholdt T, Aamot IL, Granoien I, Gjerde L, Myklebust G, Walderhaug L,
et al. Aerobic interval training increases peak oxygen uptake more than
usual care exercise training in myocardial infarction patients: a random-
Appendix A. Supplementary data ized controlled study. Clinical Rehabilitation 2012;26:33–44.
[12] Moholdt TT, Amundsen BH, Rustad LA, Wahba A, Lovo KT, Gullikstad
Supplementary data associated with this article can be found, LR, et al. Aerobic interval training versus continuous moderate exercise
after coronary artery bypass surgery: a randomized study of cardiovas-
in the online version, at http://dx.doi.org/10.1016/j.hlc.
cular effects and quality of life. American Heart Journal 2009;158:1031–7.
2015.06.828. [13] Wisloff U, Stoylen A, Loennechen JP, Bruvold M, Rognmo O, Haram PM,
et al. Superior cardiovascular effect of aerobic interval training versus
moderate continuous training in heart failure patients: a randomized
References study. Circulation 2007;115:3086–94.
[14] Warburton DE, McKenzie DC, Haykowsky MJ, Taylor A, Shoemaker P,
[1] Heran BS, Chen JM, Ebrahim S, Moxham T, Oldridge N, Rees K, et al. Ignaszewski AP, et al. Effectiveness of high-intensity interval training for
Exercise-based cardiac rehabilitation for coronary heart disease. The the rehabilitation of patients with coronary artery disease. The American
Cochrane database of systematic reviews 2011. CD001800. Journal of Cardiology 2005;95:1080–4.
[2] Kodama S, Saito K, Tanaka S, Maki M, Yachi Y, Asumi M. Cardiorespiratory [15] Rognmo O, Hetland E, Helgerud J, Hoff J, Slordahl S. High intensity aerobic
fitness as a quantitative predictor of all-cause mortality and cardiovascular interval exercise is superior to moderate intensity exercise for increasing
events in healthy men and women: A meta-analysis. JAMA 2009;301:2024–35. aerobic capacity in patients with coronary artery disease. European Journal
[3]] Keteyian SJ, Brawner CA, Savage PD, Ehrman JK, Schairer J, Divine G, of Cardiovascular Prevention & Rehabilitation 2004;11:216–22.
et al. Peak aerobic capacity predicts prognosis in patients with coronary [16] Vanhees L, Geladas N, Hansen D, Kouidi E, Niebauer J, Reiner Z, et al.
heart disease. American Heart Journal 156: 292–300. Importance of characteristics and modalities of physical activity and
[4] Elliott AD, Rajopadhyaya K, Bentley DJ, Beltrame JF, Aromataris EC. exercise in the management of cardiovascular health in individuals with
Interval training versus continuous exercise in patients with coronary artery cardiovascular risk factors: recommendations from the EACPR (Part II).
disease: a meta-analysis. Heart, Lung and Circulation 2015;24:149–57. European Journal of Preventive Cardiology 2011.
[5] Pattyn N, Coeckelberghs E, Buys R, Cornelissen VA, Vanhees L. Aerobic [17] Fox K, Borer JS, Camm AJ, Danchin N, Ferrari R, Lopez Sendon JL, et al.
interval training vs. moderate continuous training in coronary artery Resting heart rate in cardiovascular disease. Journal of the American
disease patients: a systematic review and meta-analysis. Sports Medicine College of Cardiology 2007;50:823–30.
2014;44:687–700. [18] Billman GE, Cagnoli KL, Csepe T, Li N, Wright P, Mohler PJ, et al.
[6] Conraads VM, Pattyn N, De Maeyer C, Beckers PJ, Coeckelberghs E, Exercise training induced bradycardia: evidence for enhanced parasym-
Cornelissen VA, et al. Aerobic interval training and continuous training pathetic regulation without changes in ‘‘intrinsic’’ sinoatrial node func-
equally improve aerobic exercise capacity in patients with coronary tion. J Appl Physiol 1985;118(11 (Jun 1)):1344–55. 2015.
artery disease: the SAINTEX-CAD study. International Journal of Car- [19] Ramos JS, Dalleck LC, Tjonna AE, Beetham KS, Coombes JS. The Impact
diology 2015;179:203–10. of High-Intensity Interval Training Versus Moderate-Intensity Continu-
[7] Keteyian SJ, Hibner BA, Bronsteen K, Kerrigan D, Aldred HA, Reasons LM, ous Training on Vascular Function: a Systematic Review and Meta-
et al. Greater improvement in cardiorespiratory fitness using higher-inten- Analysis. Sports Medicine 2015.
sity interval training in the standard cardiac rehabilitation setting. Journal [20] Briffa T, Eckermann S, Griffiths A, Keech A, Harris P, Heath M, et al.
of Cardiopulmonary Rehabilitation and Prevention 2014;34:98–105. Cost-effectiveness of rehabilitation after an acute coronary event: a ran-
[8] Iellamo F, Manzi V, Caminiti G, Vitale C, Castagna C, Massaro M, et al. domised controlled trial. Med J Aust 2005;183:450–5.
Matched dose interval and continuous exercise training induce similar [21] Rognmo O, Moholdt T, Bakken H, Hole T, Molstad P, Myhr NE, et al.
cardiorespiratory and metabolic adaptations in patients with heart fail- Cardiovascular risk of high- versus moderate-intensity aerobic exercise
ure. International Journal of Cardiology 2013;167:2561–5. in coronary heart disease patients. Circulation 2012;126:1436–40.
[9] Currie KD, Dubberley JB, McKelvie RS, MacDonald MJ. Low-volume, [22] Haykowsky M, Timmon M, Kruger C, McNeely M, Taylor D, Clark A.
high-intensity interval training in patients with CAD. Medicine and Meta-analysis of aerobic interval training on exercise capacity and sys-
Science in Sports and Exercise 2013;45:1436–42. tolic function in patients with heart failure and reduced ejection fractions.
[10] Rocco EA, Prado DM, Silva AG, Lazzari JM, Bortz PC, Rocco DF, et al. Am J Cardiol 2013;111:1466–9.
Effect of continuous and interval exercise training on the PETCO2

View publication stats