Show Summary Details
Page of

Evidence for cardiac rehabilitation in the modern era 

Evidence for cardiac rehabilitation in the modern era
Evidence for cardiac rehabilitation in the modern era

Constantinos H. Davos

and Bernhard Rauch

Page of

PRINTED FROM OXFORD MEDICINE ONLINE ( © Oxford University Press, 2021. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

date: 05 March 2021


Management of cardiovascular disease (CVD) has rapidly improved during recent decades, and is still changing with the introduction of novel medication and advanced invasive procedures and devices. Notwithstanding these developments, cardiac rehabilitation (CR) is still a cornerstone of secondary prevention. Its effectiveness in improving the physical condition of chronic coronary syndrome (CCS) patients is beyond doubt, but its effectiveness on extending life expectancy is still a matter of debate. This chapter provides insights into the latest evidence (mainly presented in a recent meta-analysis of randomized controlled trials (RCTs) or controlled cohort studies) on the role of CR on morbidity and mortality in patients after an acute coronary event.


CR is considered a fundamental strategy in the prevention of secondary CVD. It has received a class IA recommendation in European and international guidelines for improving outcome in patients after an acute coronary event or revascularization procedure. As many of the RCTs supporting this recommendation have been of small size with limited follow-up periods, the effect of CR on morbidity and mortality has mainly been evaluated by meta-analyses.

The evidence

The first meta-analyses by Oldridge et al. [1] and O’Connor et al. [2] were published more than 30 years ago, and included 10–22 RCTs with more than 4300 participants. These meta-analyses showed that exercise-based CR may lead to a 20–25% reduction in all-cause and CVD mortality compared with standard care methods. Subsequently, the effect of CR on clinical outcome was evaluated in a series of Cochrane systematic reviews. Cochrane publications are established as a highly reliable tool for assessment of scientific evidence with respect to the effectiveness of clinical interventions because of the extended systematic literature searches undertaken, the rigorous study selection and evaluation, and their improved statistical methodology. Therefore Cochrane meta-analyses often serve as the basis for clinical recommendations and guidelines.

The first Cochrane meta-analysis on the clinical effect of CR by Jolliffe et al. [3] was published in 2001, and subsequently updated by Taylor et al. (2004) [4] and Heran et al. (2011) [5]. The results of these Cochrane meta-analyses did not significantly change during this 10-year period, and showed that exercise-based CR may reduce all-cause mortality by 13–27% and CV mortality by 26–36%. However, despite their professionalism, accuracy, and completeness, these meta-analyses have been criticized for including RCTs of doubtful size and quality, in which women, the elderly, and high-risk populations were poorly represented. It has also been argued that the introduction of statins, angiotensin-converting enzyme (ACE) inhibitors, and dual anti-platelet therapy, as well as modern invasive techniques and devices, has changed the clinical course of coronary artery disease (CAD) in recent years, leading to significantly lower mortality after acute CAD events [6]. However, lower baseline mortality may significantly influence the efficacy of traditional therapeutic tools as represented by CR after acute coronary syndromes (ACS). Therefore, extrapolation of data on CR effectiveness obtained before the implementation of modern therapeutic options may lead to considerable bias. As a consequence, the impression has arisen that the actual benefit of exercise-based CR may be largely overestimated.

Addressing these doubts, the most recent Cochrane review was published in 2016 by Anderson et al. [7] (Table 1.1). A significant reduction in CV mortality (10.4–7.6%) and hospitalization (30.7–26.1%) compared with controls was demonstrated in the exercise-based CR group. However, total mortality or the risk of fatal or non-fatal myocardial infarction (MI), coronary artery bypass grafting (CABG) procedures, or percutaneous coronary intervention (PCI) did not differ between the study groups. With respect to the case mix of the study populations, CR was particularly effective in reducing the CV mortality of post-MI patients but re-hospitalizations and PCI rates were unchanged. This Cochrane review also distinguished between the subgroups of RCTs published before and after 1995, which showed an interesting result: CV mortality was significantly reduced by CR in both subgroups. However, a significant reduction in total mortality could only be shown in studies published before 1995.

Table 1.1 Most recent meta-analyses on the effects of cardiac rehabilitation

Anderson et al, 2016 [7]

Rauch et al. 2016 [14]

van Halewijn et al. 2017 [12]

Sumner et al. 2017 [15]

Santiago de Araújo Pio et al. 2017 [11]

Abell et al. 2017 [17]

Powell et al. 2018 [18]

No. of trials








Literature search

Until July, 2014

1995 onwards


2000 onwards

Until Nov. 2015

Until Jan. 2016

2000 onwards

No of participants

14 486

219 702



15 133

13 423



56 (median)






59.5 (mean)


<15% F

57–90% M

16–30% F

71–90% M

80.3% M

83% M

78.4% M


MI, CABG or PCI, angina, CHD angiography defined


MI, CABG or PCI, angina, CHD angiography defined

AMI (medically managed or revascularized)

Post-MI, HF, various cardiac diagnoses


MI, CABG, PCI, angina, CHD defined by angiography

Study design





RCT, nRS, pOS, rOS (reporting CR dose)



Minimum follow-up

6 months

6 months

6 months

3 months

6 months

3 months

6 months


12 months (median)

40 months (mean)

24 months (median)

3–24 months

2 years (mean)

3 years (median)

24 weeks – 10 years

24.7 months (mean)


EBCR (supervised/ unsupervised ExTr alone or with psychosocial and/or educational interventions)

Supervised multi-component CR

Start <3 months after discharge

ExTr ≥2 times/week plus at least one of: information, motivational techniques, education, psychological support & interventions, social and vocational support

CV prevention and CR (ExTr and/or lifestyle based programme with at least one face-to-face session between healthcare provider and patient)

Supervised/ unsupervised, structured multi-component CR with ExTr and/or structured physical activity plus at least one of information provision, education, health behaviour change, psychological support or intervention, social support.

Comprehensive CR

CR dose subgroups

Low: 4-11 sessions

Medium: 12–35 sessions

High: ≥36 sessions

EBCR with structured ExTr (supervised or unsupervised), with or without lifestyle modification and counseling

Supervised or unsupervised ExTr alone or as part of a comprehensive CR programme (educational/ psychosocial components)



Centre-based CR: inpatient/outpatient/mixed/teleCR


Outpatient (≥4 sessions of structured ExTr plus at least patient education)

Supervised (hospital-based/medical-centre-based) and/or unsupervised (home-based/community-based)

Any (home-based/community-based/outpatient centre based)



Standard medical care and psychosocial and/or educational interventions, not structured ExTr

No CR participation

Usual care

No CR participation

Usual care (no CR dose)

Usual care

Standard medical care (optimal medical therapy, education, advice on diet and exercise, psychosocial support, no formal ExTr)


Total mortality

No effect


ACS: (pCCS: HR 0.37;

rCCS: HR 0.64)

CABG:(rCCS: HR 0.62)

Mixed:(rCCS: HR 0.52)

Reduced only in comprehensive CR programmes managing ≥6 risk factors

(RR 0.63)


(unadjusted OR 0.25; adjusted OR 0.47)

Low dose: No effect

Medium dose: reduced (RR 0.58)

High dose: reduced (RR 0.56)


RR 0.90, up to 10 years

No effect up to 19 years

No effect

CV mortality


(RR 0.74)


(RR 0.42)


(unadjusted OR 0.21; adjusted OR: 0.43)

No effect


(RR 0.74 up to 10y, RR 0.87 up to 19y)

No effect

Recurrent MI

No effect (fatal and/or non-fatal)


(RR 0.70)


(unadjusted OR: 0.31)

No effect (non-fatal MI)


(RR 0.80 up to 10 years)

Hospital admissions

Reduced (RR 0.82)

No effect

No effect

No effect

Borderline reduction

Cerebrovascular events


(RR 0.40)


No effect

No effect

Reduced in high dose

(RR 0.60)

No effect


No effect

No effect

Reduced in high dose

(RR 0.65)

No effect

* For clarity 95% confidence intervals are not given.

If not otherwise noted all results arepresented as HR, OR, RR represent statistically significance.

ACS, acute coronary syndrome; (A)MI, (acute) myocardial infarction; CABG: coronary artery bypass grafting; CHD, coronary heart disease; CV, cardiovascular; EBCR, exercise-based cardiac rehabilitation; ExTr, exercise training; F, female; HF, heart failure; HR, hazards ratio; M, male; nRS, non-randomized studies; (n)STEMI, (non)ST elevation myocardial infarction; pCCS, prospective control cohort studies; PCI, percutaneous coronary interventions; pOS, prospective observational studies; rCCS, randomized control cohort studies; RCT, randomized control trials; rOS, randomized observational studies; RR, relative risk; UA, unstable angina.

The Rehabilitation After Myocardial Infarction Trial (RAMIT) attempted to address the lack of large prospective RCTs in this field, but failed to show a significant difference in all-cause mortality between MI patients who were and were not referred to CR during 2 years of follow-up [8]. CR delivered in this study was comprehensive, including exercise training (ET) as the baseline intervention supplemented by health education and advice regarding secondary prevention. Unfortunately, RAMIT also had some serious limitations. The sample size was not large enough to reach a sufficient statistical power. Only 23% of the number of patients originally anticipated were enrolled in each group, and more than 20% of patients dropped out of the CR programme. Apart from this, the neutral result could be explained by the heterogeneity of programmes and participating centres, resulting in variable doses of CR. The physical exercise dose during CR, the exercise intensity, and the number of CR sessions including exercise, information, education, and psychosocial support have been shown to be essential for CR success. Thus, RAMIT missed the opportunity of providing an undisputed answer to the effects of CR on patient mortality.

Importantly, the effectiveness of CR provided in everyday clinical practice is determined by the content, duration, intensity, and volume of exercise. This is in line with the observation that the physical fitness of patients in CR strongly correlates with mortality [9,10]. Accordingly, the Cochrane analysis convincingly showed that CR is particularly effective in reducing CV mortality or the re-infarction rate if the exercise volume delivered during CR exceeds a minimum [7]. Moreover, exercise intensity during CR should achieve the upper third of maximum oxygen consumption (VO2max) or an equivalent measure [9,11], and all individual CV risks should consequently be addressed and treated [10,12].

Therefore, a minimum level of supervised and individually adjusted exercise volume is a basic requirement for CR to be successful in terms of improvement of clinical outcome. However, the content and features of CR often vary between and within different countries [13], and there are no internationally accepted minimum standards for evaluating the quality of CR delivery. The Cardiac Rehabilitation Outcomes Study (CROS) has been designed to take these inadequacies into account [14].

CROS was the first meta-analysis to include not only RCTs but also prospective and retrospective controlled cohort studies enrolling patients after ACS or CABG, or mixed populations with CAD. Studies were included in the subsequent meta-analysis only if the patients were participating in supervised comprehensive multidisciplinary CR starting no longer than three months after the index event and consisting of at least two weekly sessions of structured and supervised physical exercise. This should be supplemented by at least one of the following components: information, motivational techniques, education, psychological support and intervention, and social and vocational support. CROS investigated the effectiveness of CR on total mortality in the modern era of cardiology by only including studies published in 1995 or later, thereby reflecting the introduction of statin treatment and routine acute revascularization for treating ACS.

Apart from the RAMIT trial, no other RCT satisfied the CROS inclusion criteria, indicating that RCTs included in previous meta-analyses were mainly ‘exercise-only CR studies’ or were performed in the era before modern CAD treatment. CROS was the first meta-analysis to show a significant reduction in total mortality for CR participants after an ACS or CABG and in mixed CAD populations beyond the beneficial effects of modern medication (i.e. statins) and therapeutic procedures (i.e. acute coronary revascularization). For the first time, mortality reduction was also related to some minimal standards of CR delivery, which must be structured and multicomponent, including not only physical exercise but also educational sessions and psychosocial interventions, and must start early after the acute event (Table 1.1).

In line with the results of CROS were the results of the meta-analysis by Sumner et al. [15] which included eight observational studies published after the year 2000, enrolling acute MI patients. All-cause and cardiac mortality were reduced in the CR arm. As in CROS, no effect on re-hospitalization was found. The results of a later study by van Halewijn et al. [12], which included 18 RCTs with CAD patients published between 2010 and 2015, were able to elucidate these differences in morbidity and mortality. Based on the total study population, CV mortality, MIs and cerebrovascular events were reduced, but all-cause mortality was not. However, a significant reduction of all-cause mortality was observed under the condition that comprehensive CR programmes managed six or more CV risk factors (RFs) during the intervention, thus demonstrating the augmented effectiveness of the comprehensive CR by consequently addressing and treating all individual CV RFs. This finding was in line with the results of a previous meta-analysis by Lawler et al. [16] who showed that, in post-MI patients, only comprehensive CR could reduce total and cardiac mortality compared with exercise-based CR alone. In this study, CR was only associated with a significant reduction of cardiac mortality if its duration exceeded three months, again highlighting the role of CR dose. Further, a recent meta-analysis by Santiago de Araújo Pio et al. [11] demonstrated that total mortality reduction was only feasible in CVD patients who participated in medium- and high-dose CR sessions (Table 1.1) [11]. Another CR component, the level of adherence to the exercise intervention, was the only covariate affecting total mortality (28% relative risk reduction) and CV mortality (28% relative risk reduction) in a meta-regression analysis by Abell et al. [17] which included RCTs performed until 2017 enrolling participants with diverse CAD diagnoses (Table 1.1). Total mortality was significantly reduced in the exercised-based CR group within a follow-up period of 10 years, but this effect was not sustained during an extended follow-up period of 19 years. However, the significant reduction in CV mortality observed in this study was maintained over the whole follow-up period of 19 years.

The most recent meta-analysis by Powell et al. [18] focused on a re-evaluation of the Cochrane analysis by restricting the study inclusion period to the year 2000 onwards and including RCTs with zero events during follow-up (Table 1.1). Risk ratios (RRs) or risk differences (RDs) were calculated. No significant differences between the study groups could be detected under these conditions. However, the serious methodological limitations of this meta-analysis hamper a sound interpretation of the results. Thirteen studies (out of 22) investigated a mixed population, including stable CAD patients, and differentiation of patients with acute CV events initiating CR delivery (e.g. ACS or CABG) was not possible. Only one of the six studies including ACS patients satisfied the minimal inclusion criteria of CROS [14]. The preconditions for a successful CR, as discussed previously, were not sufficiently taken into account, and internationally accepted requirements for the biometric evaluation of zero-event studies were not followed [19,20]. Finally, the most recent update of CROS (CROS II) reconfirmed the effectiveness of CR participation after ACS or CABG in actual clinical practice by reducing total mortality in patients under evidence-based medical treatment [21].


The most important reason for practising evidence-based medicine is to improve the quality of healthcare by identifying clinically effective practices, while rejecting those that are ineffective. Although meta-analyses are not infallible and may include studies with high heterogeneity in design, quality, and quantity of CR delivered, they contain important information on CR effectiveness. They provide evidence that exercise-based CR may reduce CV mortality and all-cause mortality after ACS or CABG in the modern era of revascularization and medication. In particular, reduction of mortality could be achieved if CR starts early after the acute coronary event and is structured, multicomponent, and delivered by an organized team of qualified health professionals. The duration of CR, the exercise dose, and the adherence to CR are important parameters in achieving these favourable effects (Table 1.2). However, further evidence is needed regarding endpoints such as re-hospitalization, non-fatal MI, and revascularization.

Table 1.2 Minimal requirements for successful cardiac rehabilitation derived from recent meta-analyses


Supervised multicomponent cardiac rehabilitation


Within 3 months after discharge


Any (inpatient/outpatient/community-based/home-based/mixed/tele-rehabilitation)

Exercise components

  • Frequency: ≥ 2 times/week

  • Duration: > 3 months

≥ 36 sessions

  • Intensity: upper third of VO2max

≥1000 units (no. of exercise weeks × average no. of sessions/week × average duration of session in minutes)

Other components

  • At least once a week: information, motivational techniques, education, psychological support and interventions, social and vocational support

  • Management of six or more of the following risk factors: smoking cessation, physical exercise training, counselling for exercise/activity, diet, blood pressure, cholesterol, glucose levels, checking medication, stress management


1. Oldridge NB, Guyatt GH, Fischer ME, Rimm AA. Cardiac rehabilitation after myocardial infarction. Combined experience of randomized clinical trials. JAMA 1988; 260: 945–50.Find this resource:

2. O’Connor GT, Buring JE, Yusuf S, et al. An overview of randomized trials of rehabilitation with exercise after myocardial infarction. Circulation 1989; 80: 234–44.Find this resource:

3. Jolliffe JA, Rees K, Taylor RS, et al. Exercise-based rehabilitation for coronary heart disease. Cochrane Database Syst Rev 2000; (4): CD001800.Find this resource:

4. 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.Find this resource:

5. Heran BS, Chen JM, Ebrahim S, et al. Exercise-based cardiac rehabilitation for coronary heart disease. Cochrane Database Syst Rev 2011; (7): CD001800.Find this resource:

6. Puymirat E, Simon T, Cayla G, et al. Acute myocardial infarction: changes in patient characteristics, management, and 6-month outcomes over a period of 20 years in the FAST-MI program (French Registry of Acute ST-Elevation or Non-ST-Elevation Myocardial Infarction) 1995 to 2015. Circulation 2017; 136: 1908–19.Find this resource:

7. Anderson L, Thompson DR, Oldridge N, et al. Exercise-based cardiac rehabilitation for coronary heart disease. Cochrane Database Syst Rev 2016;(1): CD001800.Find this resource:

8. West RR, Jones DA, Henderson AH. Rehabilitation After Myocardial Infarction Trial (RAMIT): multi-centre randomized controlled trial of comprehensive cardiac rehabilitation in patients following myocardial infarction. Heart 2012; 98: 637–44.Find this resource:

9. Kavanagh T, Mertens DJ, Hamm LF, et al. Peak oxygen intake and cardiac mortality in women referred for cardiac rehabilitation. J Am Coll Cardiol 2003; 42: 2139–43.Find this resource:

10. Martin BJ, Arena R, Haykowsky M, et al. Cardiovascular fitness and mortality after contemporary cardiac rehabilitation. Mayo Clin Proc 2013; 88: 455–63.Find this resource:

11. Santiago de Araújo Pio C, Marzolini S, Pakosh M, Grace SL. Effect of cardiac rehabilitation dose on mortality and morbidity: a systematic review and meta-regression analysis. Mayo Clin Proc 2017; 92: 1644–59.Find this resource:

12. van Halewijn G, Deckers J, Tay HY, et al. Lessons from contemporary trials of cardiovascular prevention and rehabilitation: a systematic review and meta-analysis. Int J Cardiol 2017; 232: 294–303.Find this resource:

13. Benzer W, Rauch B, Schmid JP, et al. Exercise-based cardiac rehabilitation in twelve European countries results of the European Cardiac Rehabilitation Registry. Int J Cardiol 2017; 228: 58–67.Find this resource:

14. Rauch B, Davos CH, Doherty P, et al. The prognostic effect of cardiac rehabilitation in the era of acute revascularisation and statin therapy: a systematic review and meta-analysis of randomized and non-randomized studies—the Cardiac Rehabilitation Outcome Study (CROS). Eur J Prev Cardiol 2016; 23: 1914–39.Find this resource:

15. Sumner J, Harrison A, Doherty P. The effectiveness of modern cardiac rehabilitation: a systematic review of recent observational studies in non-attenders versus attenders. PLoS One 2017; 12: e0177658.Find this resource:

16. Lawler PR, Filion KB, Eisenberg MJ. Efficacy of exercise-based cardiac rehabilitation post-myocardial infarction: a systematic review and meta-analysis of randomized controlled trials. Am Heart J 2011; 162: 571–84.e2.Find this resource:

17. Abell B, Glasziou P, Hoffmann T. The contribution of individual exercise training components to clinical outcomes in randomised controlled trials of cardiac rehabilitation: a systematic review and meta-regression. Sports Med Open 2017; 3: 19.Find this resource:

18. Powell R, McGregor G, Ennis S, et al. Is exercise-based cardiac rehabilitation effective? A systematic review and meta-analysis to re-examine the evidence. BMJ Open 2018; 8(3): e019656.Find this resource:

19. Bradburn MJ, Deeks JJ, Berlin JA, Russell Localio A. Much ado about nothing: a comparison of the performance of meta-analytical methods with rare events. Stat Med 2007; 26: 53–77.Find this resource:

20. Lane PW. Meta-analysis of incidence of rare events. Stat Methods Med Res 2013; 22: 117–32.Find this resource:

21. Salzwedel A, Jensen K, Rauch B, et al. Effectiveness of comprehensive cardiac rehabilitation in coronary artery disease patients treated according to contemporary evidence based medicine. Update of the Cardiac Rehabilitation Outcome Study (CROS-II). Eur J Prev Cardiol 2020; 23 February:2047487320905719.Find this resource:

Further reading

Ambrosetti M, Abreu A, Corrà U, et al. Secondary prevention through comprehensive cardiac rehabilitation: from knowledge to implementation. 2020 update. A position paper from the Secondary Prevention and Rehabilitation Section of the European Association of Preventive Cardiology. Eur J Prev Cardiol 2020; 30 March: 2047487320913379.Find this resource:

Piepoli MF, Corrà U, Abreu A, et al. Challenges in secondary prevention of cardiovascular diseases: a review of the current practice. Int J Cardiol 2015; 180: 114–19.Find this resource:

Piepoli MF, Hoes AW, Agewall S, et al. European Guidelines on cardiovascular disease prevention in clinical practice: The Sixth Joint Task Force of the European Society of Cardiology and Other Societies on Cardiovascular Disease Prevention in Clinical Practice. Eur Heart J 2016; 37: 2315–81.Find this resource: