Skip to main content
The Cochrane Database of Systematic Reviews logoLink to The Cochrane Database of Systematic Reviews
. 2021 May 7;2021(5):CD010876. doi: 10.1002/14651858.CD010876.pub3

Exercise‐based cardiac rehabilitation for adults after heart valve surgery

Lizette N Abraham 1, Kirstine L Sibilitz 2, Selina K Berg 2, Lars H Tang 3,4, Signe S Risom 5,6,7, Jane Lindschou 8, Rod S Taylor 9,, Britt Borregaard 10, Ann-Dorthe Zwisler 11,12
Editor: Cochrane Heart Group
PMCID: PMC8105032  PMID: 33962483

Abstract

Background

The impact of exercise‐based cardiac rehabilitation (CR) following heart valve surgery is uncertain. We conducted an update of this systematic review and a meta‐analysis to assess randomised controlled trial evidence for the use of exercise‐based CR following heart valve surgery.

Objectives

To assess the benefits and harms of exercise‐based CR compared with no exercise training in adults following heart valve surgery or repair, including both percutaneous and surgical procedures. We considered CR programmes consisting of exercise training with or without another intervention (such as an intervention with a psycho‐educational component).

Search methods

We searched the Cochrane Central Register of Clinical Trials (CENTRAL), in the Cochrane Library; MEDLINE (Ovid); Embase (Ovid); the Cumulative Index to Nursing and Allied Health Literature (CINAHL; EBSCO); PsycINFO (Ovid); Latin American Caribbean Health Sciences Literature (LILACS; Bireme); and Conference Proceedings Citation Index‐Science (CPCI‐S) on the Web of Science (Clarivate Analytics) on 10 January 2020. We searched for ongoing trials from ClinicalTrials.gov, Clinical‐trials.com, and the World Health Organization International Clinical Trials Registry Platform on 15 May 2020.

Selection criteria

We included randomised controlled trials that compared exercise‐based CR interventions with no exercise training. Trial participants comprised adults aged 18 years or older who had undergone heart valve surgery for heart valve disease (from any cause) and had received heart valve replacement or heart valve repair. Both percutaneous and surgical procedures were included.

Data collection and analysis

Two review authors independently extracted data. We assessed the risk of systematic errors (‘bias’) by evaluating risk domains using the 'Risk of bias' (RoB2) tool. We assessed clinical and statistical heterogeneity. We performed meta‐analyses using both fixed‐effect and random‐effects models. We used the GRADE approach to assess the quality of evidence for primary outcomes (all‐cause mortality, all‐cause hospitalisation, and health‐related quality of life).

Main results

We included six trials with a total of 364 participants who have had open or percutaneous heart valve surgery. For this updated review, we identified four additional trials (216 participants). One trial had an overall low risk of bias, and we classified the remaining five trials as having some concerns.

Follow‐up ranged across included trials from 3 to 24 months. Based on data at longest follow‐up, a total of nine participants died: 4 CR versus 5 control (relative risk (RR) 0.83, 95% confidence interval (CI) 0.26 to 2.68; 2 trials, 131 participants; GRADE quality of evidence very low). No trials reported on cardiovascular mortality. One trial reported one cardiac‐related hospitalisation in the CR group and none in the control group (RR 2.72, 95% CI 0.11 to 65.56; 1 trial, 122 participants; GRADE quality of evidence very low). We are uncertain about health‐related quality of life at completion of the intervention in CR compared to control (Short Form (SF)‐12/36 mental component: mean difference (MD) 1.28, 95% CI ‐1.60 to 4.16; 2 trials, 150 participants; GRADE quality of evidence very low; and SF‐12/36 physical component: MD 2.99, 95% CI ‐5.24 to 11.21; 2 trials, 150 participants; GRADE quality of evidence very low), or at longest follow‐up (SF‐12/36 mental component: MD ‐1.45, 95% CI ‐4.70 to 1.80; 2 trials, 139 participants; GRADE quality of evidence very low; and SF‐12/36 physical component: MD ‐0.87, 95% CI ‐3.57 to 1.83; 2 trials, 139 participants; GRADE quality of evidence very low). 

Authors' conclusions

Due to lack of evidence and the very low quality of available evidence, this updated review is uncertain about the impact of exercise‐CR in this population in terms of mortality, hospitalisation, and health‐related quality of life. High‐quality (low risk of bias) evidence on the impact of CR is needed to inform clinical guidelines and routine practice.

Keywords: Adult; Female; Humans; Male; Middle Aged; Aortic Valve; Aortic Valve/surgery; Cardiac Rehabilitation; Cardiac Rehabilitation/methods; Exercise; Exercise Tolerance; Heart Valve Prosthesis Implantation; Heart Valve Prosthesis Implantation/mortality; Heart Valve Prosthesis Implantation/rehabilitation; Mitral Valve; Mitral Valve/surgery; Physical Conditioning, Human; Physical Conditioning, Human/methods; Randomized Controlled Trials as Topic; Resistance Training; Return to Work; Time Factors

Plain language summary

Exercise‐based cardiac rehabilitation for adults after heart valve surgery

Background

Cardiac rehabilitation (CR) that includes exercise training has been recommended as treatment for people after heart valve surgery. However, the strength of this evidence is uncertain. This updated review aimed to assess the benefits and harms of exercise‐based CR for adults who have undergone heart valve surgery or repair. All types of heart valve surgery were included.

Trial characteristics

We searched for studies examining the effects of exercise‐based CR compared with no exercise ('control') after heart valve surgery for adults (18 years or older) with heart valve disease (from any cause). The evidence is current to 10 January 2020.

Key results

We found six trials with a total of 364 participants. In this update, we added four new trials (216 participants) to those included in the previously published review. We are uncertain about the effects of exercise‐based CR compared to control on the outcomes of all‐cause mortality, health‐related quality of life, and all‐cause hospitalisation. 

Quality of the evidence

Results from this Review should be interpreted with caution because of some concerns about risk of bias (potential for systematic error) in five out of six trials. Only one trial had low risk of bias. Additional high‐quality randomised controlled trials are needed to fully assess the effects of exercise‐based CR interventions.

Summary of findings

Summary of findings 1. Exercise compared to no exercise for adults after heart valve surgery.

Exercise compared to no exercise for adults after heart valve surgery
Patient or population: adults after heart valve surgery
Setting: hospital‐ and home‐based
Intervention: exercise
Comparison: no exercise
Outcomes Anticipated absolute effects* (95% CI) Relative effect
(95% CI) №. of participants
(studies) Certainty of the evidence
(GRADE) Comments
Risk with no exercise Risk with exercise
All‐cause mortality
Follow‐up range: 3 to 24 months
Study population RR 0.83
(0.26 to 2.68) 131
(2 RCTs) ⊕⊝⊝⊝  VERY LOWa,b,c   
79 per 1000 66 per 1000
(21 to 213)
Cardiovascular mortality No study reported this outcome        
All‐cause hospitalisation
Follow‐up: 6 months
Study population RR 2.72
(0.11 to 65.56) 122
(1 RCT) ⊕⊝⊝⊝ VERY LOWb,c,d  There were 0 events in the control group
0 per 1000 0 per 1000
(0 to 0)
HRQoL (SF‐12/36 mental component) at end of intervention
Follow‐up range: 2 to 3 months
Mean HRQoL range (mental component) at end of intervention was 51.3 to 53.9 MD 1.28 higher
(1.60 lower to 4.16 higher) 150
(2 RCTs) ⊕⊝⊝⊝ VERY LOWb,c,d   
HRQoL (SF‐12/36 physical component) at end of intervention
Follow‐up range: 2 to 3 months
Mean HRQoL range (physical component) at end of intervention was 38 to 51 MD 2.99 higher
(5.24 lower to 11.21 higher) 150
(2 RCTs) ⊕⊝⊝⊝
VERY LOWb,c,d,e  
HRQoL (SF‐12/36 mental component) at maximum follow‐up
Follow‐up range: 3 to 24 months
Mean HRQoL range (mental component) at maximum follow‐up was 54.9 to 55.1 MD 1.45 lower
(4.70 lower to 1.80 higher) 139
(2 RCTs) ⊕⊝⊝⊝ VERY LOWb,c,d  
HRQoL (SF‐12/36 physical component) at maximum follow‐up
Follow‐up range: 3 to 24 months
Mean HRQoL range (physical component) at maximum follow‐up was 36.9 to 52.2 MD 0.87 lower
(3.57 lower to 1.83 higher) 139
(2 RCTs) ⊕⊝⊝⊝ VERY LOWb,c,d  
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: confidence interval; HRQoL: health‐related quality of life; MD: mean difference; RCT: randomised controlled trial; RR: risk ratio; SMD: standardised mean difference.
GRADE Working Group grades of evidence.High certainty: we are very confident that the true effect lies close to that of the estimate of the effect.
Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.
Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect.
Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

aAt least one trial has some concerns for overall risk of bias. Downgraded by one level for risk of bias.

bSmall sample size/number of events and optimal information size (OIS) criterion not reached, or OIS criterion reached but 95% CI includes RR/MD/SMD of 1/0. Downgraded by one level for inconsistency.

cConfidence interval includes possible benefit or harm (i.e. effect crosses RR of 0). Downgraded by one level for imprecision. 

dAll trials providing data for this outcome have an overall risk of bias judged as 'high'. Downgraded by one level for risk of bias.

eSubstantial I2 (between 50% and 90%). Downgraded by one level for imprecision.

Background

Description of the condition

Heart valve disease accounts for one‐third of all heart disease and is increasing in prevalence due to an ageing population, population growth, and advances in treatment methods. Heart valve disease is mostly degenerative in nature (Nkomo 2006), and it is highly prevalent in developing countries due to rheumatic heart disease (Iung 2003; Nkomo 2006Sibilitz 2015aSupino 2006; Yagdir 2020).

Heart valve disease can be left‐sided (aortic and mitral valve diseases), right‐sided (tricuspid and pulmonary valves), or, in rare cases, a combination of both. The cause may be congenital, degenerative, or calcific, and physiological consequences may include valve insufficiency, valve stenosis, or both (Baumgartner 2017Nkomo 2006). Heart valve disease is often asymptomatic at first. When it becomes symptomatic, the clinical presentation includes dyspnoea (difficulty breathing), fatigue, fluid retention, and decreased physical capacity. Symptomatic heart valve disease is associated with increased risks of mortality and morbidity, and it negatively impacts health‐related quality of life (HRQoL) and physical capacity (Baumgartner 2017Ben‐Dor 2010; Frank 1973). Medical follow‐up of valve disease includes regular clinical and echocardiographic follow‐up (Baumgartner 2017; Vahanian 2012), as well as assessment of treatment indications. The treatment of choice when serious symptoms and/or haemodynamic changes occur is valve surgery with valve repair or replacement (Baumgartner 2017Nishimura 2014; Vahanian 2012).

The changing disease pattern and expected increase in healthcare burden of patients after heart valve surgery require a well‐established after‐care programme to support the patient in managing postsurgical problems. These problems include physical and psychological issues and the challenge of returning to work. The large number of acute hospitalisations after valve surgery highlights the importance of follow‐up (Sibilitz 2015a). One trial to date has shown that individualised follow‐up programmes after surgery can reduce the risk of hospital admission (Borregaard 2019). Transcatheter aortic valve replacement (TAVR) is increasingly used for treatment of people with aortic stenosis and low surgical risk, impacting recovery following surgery. Data from the NOTION 3, PARTNER‐3, and Evolut Low Risk trials show that TAVR is at least non‐inferior and may be superior to surgery (Kolte 2019Mack 2019Popma 2019). However, the shorter stay in hospital at the time of TAVR (typically 1 to 3 days) has increased the demand for patient‐centred follow‐up and careful planning of rehabilitation. This is reflected in the latest (2017) European Society of Cardiology/European Association for Cardio‐Thoracic Surgery (ESC/EACTS) Guidelines, in which TAVR is recommended for patients older than 75 years of age (Baumgartner 2017); these guidelines were updated in 2020, and it is expected that results from PARTNER‐3 and Evolut Low Risk trials have been integrated (Ambrosetti 2020).

Physical inactivity is a problem for heart valve surgery patients,  who may experience presurgical dyspnoea and physical incapacity, immobilisation during hospitalisation, and potential postsurgical complications and restrictions due to healing of the sternum. Open heart surgery is a stressful life event (Karlsson 2010), and HRQoL is likely to be negatively affected (Hansen 2009), along with mental health; patients may require support for depressive symptoms and anxiety (Fredericks 2012). Although such problems may also occur following percutaneous procedures, recent studies suggest that after TAVR, patients have much better HRQoL within two weeks of the procedure (Lauck 2020). A Cochrane Review showed that participants who had undergone surgery for a coronary artery bypass graft might benefit from psychological interventions; however, risk of bias of included trials was considered to be high (Whalley 2011). Little is known about the effects of psychological interventions for patients after heart valve surgery.

In summary, risks of mortality and morbidity leading to hospital re‐admission are increased after heart valve surgery, resulting in high potential healthcare costs. In addition, patients are likely to experience physical, mental, or social recovery problems that negatively impact their HRQoL and physical capacity. Therefore, careful postsurgical recovery programmes are needed. One key solution may be exercise‐based cardiac rehabilitation (CR) (Baumgartner 2017Butchart 2005).

Description of the intervention

CR is defined as "the coordinated sum of activities required to influence favourably the underlying cause of cardiovascular disease, as well as to provide the best possible physical, mental and social conditions, so that the participants may, by their own efforts, preserve or resume optimal functioning in their community and through improved health behaviour, slow or reverse progression of disease" (BACPR 2012). Although a central component of rehabilitation programmes is exercise training, it is recognised that CR programmes should be 'comprehensive' and combined with other interventions, particularly those with psycho‐educational components (Ambrosetti 2020Piepoli 2010).

Current European guidelines recommend that rehabilitation following heart valve surgery should include exercise training, anticoagulant therapy, and medical and echocardiographic follow‐up. However, these guidelines do not explicitly state that psycho‐educational interventions should be part of the rehabilitation programme (Baumgartner 2017Butchart 2005). In contrast, American guidelines do not currently include any recommendations or information about CR after heart valve surgery (Balady 2007; Nishimura 2014).

A meta‐analysis published in 2017 and including six trials showed that participation in exercise training after TAVR can increase exercise capacity within the first year after the procedure (Ribeiro 2017). This is supported by a systematic review and meta‐analysis published in 2019 reporting that exercise‐based CR improves exercise capacity of post‐transcatheter aortic valve replacement (TAVR) and post‐surgical aortic valve replacement (SAVR) patients in the short term (Anayo 2019). This review concludes that further evidence is needed to assess the clinical effects and cost‐effectiveness of exercise‐based CR in people with valve disease. A reported cohort trial showed that CR is associated with decreased one‐year cumulative hospitalisation and mortality risk after valve surgery (Patel 2019).

The European Society of Cardiology recommends that physical activity for patients with cardiovascular disease should comprise 150 minutes per week, while others recommend three to four hours per week (Piepoli 2010). Further, recommendations state that low‐risk patients should perform 30 minutes of aerobic exercise daily to achieve a weekly expenditure of 1000 kcal, whereas the amount of physical activity should be individually prescribed for high‐risk patients (Gianuzzi 2003). Exercise training should be performed three times weekly for 12 weeks, through a local hospital or a community‐based facility (Piepoli 2010). Exercise should consist of submaximal endurance training, the intensity of which is increased over time, and the programme should be expanded to include weight/resistance training. Psychological and educational interventions should offer individual and/or small group education and counselling on adjustment to heart disease, stress management, and health‐related lifestyle changes (Gianuzzi 2003).

How the intervention might work

CR interventions following heart valve surgery can positively affect physical recovery, reduce blood pressure, reduce disease severity, and improve left ventricular ejection fraction (Gohlke‐Bärwolf 1992; Landry 1984; Newell 1980; Pardaens 2014; Sibilitz 2016; Sire 1987). Exercise training may confer direct benefits for the heart and the coronary vasculature involving myocardial oxygen demand, endothelial function, autonomic tone, coagulation and clotting factors, inflammatory markers, and development of coronary collateral vessels (Clausen 1976; Hambrecht 2000).

We might anticipate effects of exercise‐based CR after heart valve surgery similar to those seen in other cardiac populations that typically receive CR (i.e. post myocardial infarction and revascularisation and heart failure). Two Cochrane Reviews have shown that exercise‐based CR has several positive effects in these latter populations (Anderson 2016Long 2019), including reductions in hospitalisation and improvements in HRQoL. Furthermore, heart function changes due to valve dysfunction such as reduced cardiac output, stroke volume, and left ventricular ejection fraction may positively respond to exercise training. Exercise‐based CR following heart valve surgery might also be expected to reduce the symptom burden, improve symptom and disease management, and decrease rates of anxiety and depression, as has been shown for patients with atrial fibrillation (Smart 2018).

Possible harmful effects of exercise‐based CR after heart valve surgery include increased risk of surgery‐related adverse events (e.g. arrhythmias, arterial embolism, death), as well as adverse events associated with valve disease (e.g. any arrhythmias, heart failure, death). A prospective study of patients post cardiac surgery reported a rate of adverse events (defined as chest pain with typical electrocardiographic modifications, severe ventricular arrhythmias, syncope, cardiopulmonary arrest, or a clinical condition necessitating cardiopulmonary resuscitation, immediate transfer to a coronary care unit or cardiac surgery, and/or use of intravenous drugs) of only 1 per 49,565 patient‐hours of exercise training (Pavy 2006).

Why it is important to do this review

This systematic review is an update of a previous review that was undertaken to assess the benefits and harms of exercise‐based CR in adults who have undergone heart valve surgery or repair (Sibilitz 2016 SR). Since the time of first publication of this review, two non‐Cochrane systematic reviews and meta‐analyses on this topic have been published (Anayo 2019; Ribeiro 2017). 

Objectives

To assess the benefits and harms of exercise‐based CR compared with no exercise training in adults following heart valve surgery or repair, including both percutaneous and surgical procedures. We considered CR programmes consisting of exercise training with or without another intervention (such as an intervention with a psycho‐educational component).

Methods

Criteria for considering studies for this review

Types of studies

Randomised controlled trials (RCTs) (including individual participant/cluster allocation or cross‐over design) irrespective of language of publication, publication year, publication type, and publication status were eligible for inclusion in the review.

Types of participants

We included adults aged 18 years or older of both sexes and of any ethnicity who had undergone heart valve surgery for any cause of heart valve disease (i.e. aortic valve disease, mitral valve disease, tricuspid or pulmonary valve disease, or a combination) and had received heart valve replacement or heart valve repair (surgery to the valve and related anatomical areas without valve replacement, e.g. mitraclips, mitral ring, chordae rupture treatment). We included both percutaneous and surgical procedures.

Types of interventions

Exercise‐based CR interventions with or without a psycho‐educational intervention. Exercise‐based CR interventions include supervised and unsupervised programmes conducted in an inpatient, outpatient, community, or home‐based setting, including any kind of exercise training. The intervention must have included an exercise training component focused on increasing exercise capacity, and it may have included a psycho‐educational intervention that focused on improving mental health and the patient's self‐management skills. Patients could engage in an exercise intervention before or after discharge from the hospital for heart valve surgery (Kiel 2011). However, for inclusion in this review, the intervention must have included a postsurgical element. We applied no restriction in length, intensity, or content of the exercise training intervention.

Control interventions

We sought any of the following control interventions as long as they did not include a physical exercise element. 

  • Treatment as usual (e.g. standard medical care, such as drug and anticoagulant therapy; medical follow‐up with echocardiography).

  • No intervention.

  • Any other type of CR programme.

Co‐interventions

We included trials with co‐interventions to CR, as long as these were delivered equally to participants in the intervention and control groups. Co‐interventions could include drug, surgical (percutaneous versus transthoracic surgery), or dietary interventions.

Types of outcome measures

Reporting one or more of the outcomes listed here for the trial is not an inclusion criterion for this review. When a published report did not appear to report one of these outcomes, we accessed the trial protocol and contacted the trial authors to ascertain whether outcomes were measured but not reported. Relevant trials that measured these outcomes but did not report the data at all, or did not provide data in a usable format, were included in the review as part of the narrative. We did not use hierarchy to choose between multiple measures of the same outcome but instead sought to report all outcome results.

Outcomes are assessed at two time points: (1) at completion of the intervention (as defined by trialists); and (2) at longest available follow‐up. There was no minimum length of follow‐up for trials that were eligible for inclusion in the review.

Primary outcomes

We sought the following primary outcomes.

  • All‐cause mortality.

  • Cardiovascular mortality.

  • All‐cause hospitalisation.

  • Health‐related quality of life assessed by generic or disease‐specific validated instruments (e.g. Short Form‐36, EuroQoL Group Quality of Life Questionnaire based on 5 dimensions (EQ‐5D) ‐ generic measures, HeartQoL ‐ heart disease‐specific measure).

Secondary outcomes

We sought the following secondary outcomes.

  • Exercise capacity: any measure of exercise capacity including direct measurement of oxygen uptake (VO2 peak/VO2 max) or indirect measures such as exercise time, walking distance (e.g. 6‐minute walk text), etc.

  • Serious adverse events: defined as any untoward medical occurrences that are life‐threatening, result in death, or are persistent or lead to significant disability; or any medical events that have jeopardised the patient or required intervention to prevent them, or any hospitalisation or prolongation of existing hospitalisation (ICH‐GCP 1997).

  • Return to work.

  • Costs and cost‐effectiveness.

Search methods for identification of studies

Electronic searches

We searched the following electronic databases from their inception to 10 January 2020 (unless otherwise stated).

  • Cochrane Central Register of Controlled Trials (CENTRAL; 2020, Issue 1 of 12), in the Cochrane Library.

  • Database of Abstracts of Reviews of Effectiveness (DARE; 2015, Issue 1 of 4), in the Cochrane Library (last issue available, so not updated for this latest version).

  • MEDLINE and Epub Ahead of Print, In‐Process & Other Non‐Indexed Citations and Daily (Ovid) (1946 to 9 January 2020).

  • Embase Classic and Embase (Ovid) (1947 to 9 January 2020).

  • Cumulative Index to Nursing and Allied Health Literature (CINAHL) plus Full Text (EBSCO) (1937 to 10 January 2020).

  • PsycINFO (Ovid) (1806 to January week 1 2020).

  • Latin American Caribbean Health Sciences Literature (LILACS; Bireme), in English (1982 to 10 January 2020).

  • Conference Proceedings Citation Index‐S (CPCI‐S) on Web of Science (Clarivate Analytics) (1990 to 10 January 2020).

Searches for the previous review were run on 23 March 2015, and were updated and re‐run on 10 January 2020. Some additional search terms were added for each database in the latest search (Appendix 1). The RCT filter used for MEDLINE was the Cochrane sensitivity‐maximising RCT filter, and for Embase, terms as recommended in the Cochrane Handbook for Systematic Reviews of Interventions were applied (Lefebvre 2011). RCT filters used for the other databases, except CENTRAL, were adaptations of the Cochrane RCT filter.

We applied no language restrictions. Trials written in languages that the review authors did not understand were translated professionally.

We checked the status of studies identified as ongoing (7 February 2021) to determine their current publication status. None of the 10 ongoing studies were found to have been published. 

Searching other resources

We also searched the following clinical trials registers for ongoing trials on 15 May 2020.

We searched these other sources using the search terms 'heart valve surgery', 'heart valve replacement', 'exercise', and 'cardiac rehabilitation'. Several of the co‐authors are experts in the field with knowledge of current unpublished trials. We searched the reference lists of previous systematic reviews and trials included in this review .

Data collection and analysis

Selection of studies

Two review authors (LA and KLS) independently assessed all titles and abstracts for inclusion, excluding trials that did not meet the inclusion criteria. We retrieved full publications of all potentially relevant trials, stored them electronically, and translated them when required. We resolved disagreements by discussion between the two review authors (LA and KLS), or, when necessary, by consultation with a third review author (RST). We detailed excluded trials and reasons for their exclusion in the Characteristics of excluded studies table.

Data extraction and management

Two review authors (LA and KLS) independently extracted data from the included trials using a standardised data extraction form. This form was used in the previous version of this review and has been adapted from previous Cochrane cardiac rehabilitation reviews (e.g. Anderson 2016). When not reported in the text or tables, we extracted outcome data from graphs. A third review author (RST) checked all numerical calculations and data extractions. We resolved any discrepancies by consensus. One of the included trials was available only in Chinese. Data extraction for this paper was undertaken by one of the review authors (KLS) in the presence of a translator (native Chinese speaker). Data for the Chinese article were double‐checked against the English abstract (LA and KLS).

We extracted the following data.

  • General information: publication status, title, authors' names, source, country, contact address, language of publication, year of publication, duplicate publication, financial conditions.

  • Trial characteristics: design, duration.

  • Intervention: type of exercise training, type of rehabilitation programme (comprehensive CR or only exercise training), setting (e.g. in‐patient, out‐patient, community, home setting, a combination), time after hospitalisation, nature of the control group.

  • Participants: sampling method (e.g. convenience, random), inclusion and exclusion criteria, numbers of participants in intervention and control groups, participant demographics such as sex and age, baseline characteristics including type of valve affected and classification of heart valve disease, number of participants lost to follow‐up.

  • Outcomes: data sought for primary and secondary outcomes as defined earlier.

  • Risk of bias: see Assessment of risk of bias in included studies below.

One review author (LA) transferred data into Review Manager 5.4 (RevMan 2020), and another review author (KLS) double‐checked that data were entered correctly by checking trial characteristics for accuracy.

Assessment of risk of bias in included studies

For this review, the effect of interest is the effect of assignment to the intervention. Two review authors (LA and KS) independently assessed risk of bias using the Cochrane 'Risk of bias in randomised trials' tool (RoB2) for all primary outcomes (when data were provided) (i.e. at latest follow‐up for all‐cause mortality and all‐cause hospitalisation, at the end of the intervention, and at latest follow‐up for both exercise capacity and HRQoL outcomes) (Higgins 2019a; Sterne 2019). Secondary outcomes were not assessed for risk of bias. As all review authors but one (LA) were involved with one of the included trials (Sibilitz 2016), an independent RoB2 experienced review author Michele Hilton Boon (MHB) independently assessed all of the primary outcomes for this trial. Differences between RoB2 assessments were discussed between MHB and LA  (for details, see https://www.gla.ac.uk/media/Media_775195_smxx.xlsm).

We resolved all disagreements through discussion or by consultation with a third review author (RST).

We assessed risk of bias using the following Cochrane RoB2 criteria (Higgins 2019a; Sterne 2019).

  • Bias arising from the randomisation process.

  • Bias due to deviations from intended interventions.

  • Bias due to missing outcome data.

  • Bias in measurement of the outcome.

  • Bias in selection of the reported result.

For each domain, a series of signalling questions (with the answers yes, probably yes, no information, probably no, and no) will determine the risk of bias (low risk, some concerns, or high risk). We included text alongside the judgements to provide supporting information for our decisions (see 'Risk of bias in included trials'). We decided the risk of bias for an outcome (e.g. HRQoL) by noting its performance in each domain; if one domain was judged as 'some concerns' or 'high risk', this judgement was taken for the whole outcome. To manage the assessment of bias and to implement RoB2, we used the RoB2 Excel tool (available on the riskofbiasinfo.org website). The RoB2 tool was accessed from 18 to 20 May 2020.

Measures of treatment effect

We processed data in accordance with the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2019c). We expressed dichotomous data as risk ratios (RRs) with 95% confidence intervals (CIs). For continuous variables, we compared net changes (i.e. exercise‐based CR minus control) to detect differences. For each trial, we sought the mean change (and the standard deviation (SD)) in outcomes between baseline and follow‐up for both exercise and control groups. When not available, we used the absolute mean (and SD) outcome at follow‐up for both groups. We expressed results as mean differences (MDs), except when trials used different scales or measurements, in which case we used standardised mean differences (SMDs) (Thompson 2002). We interpreted SMD as 0.2, 0.5, and 0.8, representing 'small', 'medium', and 'large' effect sizes, respectively (Higgins 2019b).

Unit of analysis issues

If any cluster‐randomised controlled trials had been included, we planned to contact the trial authors to obtain an estimate of the intra‐cluster correlation when appropriate adjustments for the correlation between participants within clusters had not been made, or otherwise to impute it using estimates from the other included trials, or from similar external trials. Similarly, if we had included data from cross‐over trials, we would have included both periods of any cross‐over trials identified, assuming that (1) there had been a washout period considered long enough to reduce carry‐over, (2) no irreversible events such as mortality had occurred, and (3) appropriate statistical approaches had been used.

Dealing with missing data

As we did not obtain missing data by contacting triallists, we sought to undertake sensitivity analysis to explore the effect of this missingness. For dichotomous outcomes, we performed analyses using the intention‐to‐treat method (Higgins 2019c), which includes all participants according to their original random group allocation, irrespective of compliance or follow‐up. For primary analyses, we assumed that participants lost to follow‐up were alive and had no serious adverse events. For continuous outcomes, we performed available participant analysis and included data only on those for whom results are known (Higgins 2019c). It was possible to obtain SDs directly from the articles or by calculation (Furukawa 2006). When trials reported outcomes with medians and interquartile ranges, we calculated the means and the standard deviations by using the quantile method for estimating means and standard deviations. To calculate means and standard deviations, we divided the sum of the median, the first quartile range, and the third quartile range by three, and we subtracted the first quartile from the third quartile, then divided by 1.35, respectively (Higgins 2019ca; Chapter 6.5.2.5). When trials reported maximal oxygen consumption (VO2 max) in metabolic equivalent of tasks (METS), we converted this to mL/kg/min by multiplying by 3.5. We sought to undertake two sensitivity analyses for binary primary outcomes to examine the impact of losses to follow‐up.

Assessment of heterogeneity

We explored clinical heterogeneity by comparing population, intervention, and control groups across included trials. We observed statistical heterogeneity in the trials by visually inspecting forest plots, by using a standard Chi2 value with a significance cut‐off level of P = 0.10, and by using the I2 statistic. We interpreted an I2 estimate greater than or equal to 50% with a significant value for Chi2 as evidence of 'substantial' statistical heterogeneity (Higgins 2019c).

Small‐trial (publication) bias

We planned to construct funnel plots and to undertake Egger tests for each outcome when we identified 10 or more trials, to establish the potential influence of small‐trial effects and potential publication bias (Sterne 2011; Wood 2008). However, due to the limited number of included trials (six), this was not possible.

Assessment of reporting biases

See Assessment of risk of bias in included studies and small‐trial (publication) bias. There was no language bias, as relevant trials published in other languages were sought and translated.

Data synthesis

We performed data synthesis according to recommendations provided in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2019c), using Review Manager 5.4 (RevMan 2020). We implemented RoB2 in RevMan Web, available at revman.cochrane.org. The primary analysis will include all eligible studies, irrespective of their risk of bias status.

We pooled data from each trial using a fixed‐effect model, except when we identified substantial statistical heterogeneity (I² statistic > 50%), in which case we applied a random‐effects model, which provided a more conservative statistical comparison of differences between intervention and control, because a confidence interval around a random‐effects estimate is wider than a confidence interval around a fixed‐effect estimate.

Subgroup analysis and investigation of heterogeneity

We planned to analyse primary outcomes using stratified meta‐analysis, according to the following subgroups.

  • Trials at overall low risk of bias compared to trials at overall high risk of bias based on RoB2; for trials categorised as being at overall low risk of bias, we would perform subgroup analysis on trials at overall lower risk of bias compared to trials at overall higher risk of bias.

  • Trials including women only versus trials including men only.

  • Trials including younger participants (< 60 years old) only versus trials including older participants (≥ 60 years old) only.

  • Trials with an exercise intervention only compared to trials with an exercise intervention plus any other co‐intervention, such as a psycho‐educational intervention.

However, due to the small number of included trials and a limited quantity of data, it was not possible to perform these subgroup analyses.

Sensitivity analysis

For primary outcomes, we planned to perform the following sensitivity analyses.

Binary outcomes

Best/worst‐case scenario: for this analysis, we would assume that all participants lost to follow‐up in the intervention group have survived, and have had no serious adverse events; and that all those with missing outcomes in the control group have not survived, and have had serious adverse events.

Worst/best‐case scenario: for this analysis, we would assume that all participants lost to follow‐up in the intervention group have not survived, and have had serious adverse events; and that all those with missing outcomes in the control group have survived, and have had no serious adverse events.

Continuous data

Assumptions for lost data: when assumptions had been made for lost data (Dealing with missing data), we compared the findings from our assumptions with data only from those participants who completed the trials.

Summary of findings and assessment of the certainty of the evidence

One review author (LA) independently employed the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach to interpret study results (Schünemann 2013). We used the five GRADE considerations (overall risk of bias, consistency of effect, imprecision, indirectness, and publication bias) to assess the quality of a body of evidence as it relates to trials that contributed data to meta‐analyses and narrative summaries for pre‐specified outcomes. We (LA, KLS, RST) resolved any discrepancies in judgement through discussion. One review author (LA) used GRADEpro GDT software to import data from Review Manager to create a 'Summary of findings’ table that included the following pre‐specified outcomes: all‐cause mortality; cardiovascular mortality; all‐cause hospital hospitalisations; and health‐related quality of life (GradePro SoftwareSchünemann 2013). .

Results

Description of studies

The updated search results can be seen in Table 2; the trial selection process is shown in the PRISMA flow chart in Figure 1.

1. Updated search results.

Database searched Date searched February 2019 October 2019 January 2020 Total number of results
CENTRAL (January 2020; Issue 1 of 12), in the Cochrane Library 10/01/2020 211 132 28 371
MEDLINE and Epub Ahead of Print, In‐Process & Other Non‐Indexed Citations and Daily (Ovid, 1946 to 9 January 2020) 10/01/2020 242 46 27 315
Embase Classic + Embase (Ovid, 1947 to 9 January 2019) 10/01/2020 121 17 12 150
CINAHL Plus with Full Text (EBSCO, 1937 to 10 January 2020) 10/01/2020 160 31 16 207
PsycINFO (Ovid, 1806 to January week 1 2020) 10/01/2020 27 8 3 38
LILACS (Bireme, 1982 to 10 January 2020) in English 10/01/2020 38 6 1 45
Conference Proceedings Citation Index ‐ Science (CPCI‐S) on Web of Science (Clarivate Analytics, 1990 to 10 January 2020) 10/01/2020 10 0 2 12
DARE (2015, Issue 1 of 4), in the Cochrane Library No longer updated since March 2015 0 0 0 0
Total 809 240 89 1138
After de‐duplication 606 216 76 898

Exercise‐based cardiac rehabilitation for adults after heart valve surgery: updated search results

1.

1

Updated study flow diagram.

Results of the search

Through updated searches, we retrieved a total of 904 titles after de‐duplication, of which 865 did not fulfil the inclusion criteria and were excluded. At full paper review stage, we excluded 16 records. One was not randomised, one was an editorial, one was a protocol for an included trial, one was a conference abstract for an included trial, two had an inappropriate population, and 10 had an inappropriate intervention.

Five records are awaiting classification, as we contacted the triallists about details of their trials but received no response, and the detail we had was insufficient to warrant inclusion in this review (Characteristics of studies awaiting classification).

We identified 10 ongoing trials from results of the electronic searches, as well as from our search of other resources. Details of these ongoing trials can be found in the section on Characteristics of ongoing studies (ACTIVE AFTER TAVR 2017; Exercise Training After TAVI; Feng 2019; HBCR‐TAVR 2019; Post Cardiac Valvular Surgery Rehabilitation (PORT); PREPARE TAVR Pilot Study; REHAB‐TAVR 2017; The PACO TrialValve‐ex 2009Wang 2019). They will be assessed during future updates of this review.

Four new trials (six publications: two from a recent systematic review ‐ Anayo 2019) met the inclusion criteria and were therefore included in this review update. In total, this review included six trials ‐ two from the previous version of this review.

Included studies

See Characteristics of included studies and Characteristics of excluded studies.

Population

The six included trials randomised a total of 364 participants who had undergone heart valve replacement or repair. Four trials included participants after aortic valve replacement only (Nilsson 2019; Pressler 2016; Rogers 2018Sire 1987), one trial involved mitral valve replacement only (Lin 2004), and one trial included all heart valves (Sibilitz 2016). Some trials included participants undergoing several valve procedures at a time (e.g. two valve procedures) (Lin 2004; Sire 1987), but all trials excluded participants with other heart co‐morbidities, or with other co‐morbidities complicating physical activity. All trials had published abstracts in English, and all but the Lin 2004 trial (Chinese) were published in full in English. Five trials were single‐centre studies. Pressler 2016 was conducted at three different centres. None of the trials were reported to be industry‐sponsored.

Trial participants were predominantly male in four trials (57% ‐ Lin 2004, 75% ‐ Nilsson 2019, 76% ‐ Sibilitz 2016, and 72% ‐ Sire 1987); in the other two trials, the proportion of males was equal to the proportion of females (50% ‐ Pressler 2016, or slightly lower (44%) ‐ Rogers 2018). Mean participant age across trials varied from 31 years in Lin 2004 to 82 years in Rogers 2018. Although ethnicity of participants was not reported, five trials took place in Europe, and one in China. The longest reported trial follow‐up time ranged from 3 months in Lin 2004 to 24 months in Pressler 2016.

Interventions

Included exercise‐based interventions consisted of combined aerobic and resistance training that began one day to three months post surgery (Lin 2004Pressler 2016). Lin 2004 also included a psychological intervention and an exercise training element, both of which were undertaken before surgery. In three trials, the intervention was provided in a combined hospital‐ and home‐based setting (Lin 2004; Sibilitz 2016; Sire 1987), and in the other three trials, the intervention was given entirely in a hospital setting (Nilsson 2019; Pressler 2016; Rogers 2018). The dose and intensity of prescribed exercise training varied from 20 to 60 minutes per session across two to three sessions per week, except for one trial that recommended up to four hours daily (Sire 1987). The total duration of exercise programmes varied between trials from approximately one month in Sire 1987 to over three months in Lin 2004Nilsson 2019, and Sibilitz 2016. In Rogers 2018, the dose, frequency, length, and intensity of exercise were individualised based on information gained from participants' functional capacity tests and discussion around their specific goals. Further details of the trials included in this review are shown under Characteristics of included studies.

Comparison
All trials compared interventions to no exercise and usual care.

Excluded studies

We excluded 16 trials and have presented reasons for their exclusion in the section Characteristics of excluded studies. The most common reason for trial exclusion was the type of intervention used, as it was not appropriate for this review.

Risk of bias in included studies

We performed risk of bias assessment using the RoB2 tool for all primary outcomes (when data were provided) and summarised results of this assessment in the results‐level RoB2 tables (Higgins 2019c). Although some trials failed to give sufficient detail to enable a clear assessment of the potential risk of bias for outcomes measured (Lin 2004; Sire 1987), most trials provided sufficient information to allow for potential risk of bias assessment (for details, see https://www.gla.ac.uk/media/Media_775195_smxx.xlsm).

For all‐cause mortality outcomes, we assumed an overall risk of bias with some concern, as one of the two trials was at overall high risk of bias and the other was at low risk of bias. However, no trials intended to measure mortality as a primary or secondary outcome. Only Sibilitz 2016 reported all‐cause hospitalisations and was judged at high risk of bias, with short‐term follow‐up and few patients/events. We judged HRQoL physical and mental component outcomes to be at high risk of bias due to the small numbers of patients and the high level of missing outcome data at follow‐up. 

Given the nature of exercise‐based CR interventions and controls, it is not possible to blind participants or people delivering the intervention. Nevertheless, blinding of outcome assessors can reduce risk of bias in measurement of outcomes that involve clinician assessment (exercise capacity) or participant self‐reported outcomes (HRQoL, return to work). Three trials did not report any information on assessment of outcomes (Lin 2004Nilsson 2019Sire 1987). 

Effects of interventions

See: Table 1

Primary outcomes

All‐cause mortality

Nine deaths were reported by two trials (Lin 2004; Pressler 2016). We found lack of evidence of a difference between exercise‐CR and control (risk ratio (RR) 0.83, 95% confidence interval (CI) 0.26 to 2.68; 2 trials, 131 participants; I= 49%; GRADE quality of evidence very low; Analysis 1.1). In Lin 2004, two participants in the exercise‐based CR group died (2/55; 3.6%) (1 sudden death, 1 brain stem death) versus none in the control group (0/49; 0%). Pressler 2016 reported seven deaths: two in the exercise‐based CR arm (2/13; 15.4%) (1 intracranial bleeding, 1 unknown cause) versus five in the control arm (5/14; 35.7%) (3 pneumonia, 2 unknown cause). Sensitivity analyses (best/worst‐case scenario: RR 0.44, 95% CI 0.15 to 1.32; worst/best‐case scenario: RR 2.15, 95% CI 0.16 to 28.78) confirmed the lack of evidence of differences in all‐cause mortality between exercise‐based CR and control. 

1.1. Analysis.

1.1

Comparison 1: Exercise versus no exercise, Outcome 1: All‐cause mortality at longest follow‐up

For all‐cause mortality, the overall risk of bias for Pressler 2016 was 'low' and that for Lin 2004 was 'high' (see Analysis 1.1). Lin 2004 had some concerns with the randomisation process and deviations from intended interventions and was at high risk of bias for missing outcome data. Pressler 2016 led to a low risk of bias judgement for this outcome. Therefore caution should be applied when all‐cause mortality results are interpreted.

Cardiovascular mortality

Cardiovascular mortality was not reported.

All‐cause hospitalisations

Only one trial reported all‐cause hospitalisations at six months' follow‐up (Sibilitz 2016). This trial reported a cardiac‐related hospitalisation in the exercise‐CR group as one of the serious adverse events. No hospitalisations were reported in the control group (RR 2.72, 95% CI 0.11 to 65.56; fixed‐effect model; 1 trial, 122 participants; I2 = NA; GRADE quality of evidence very low; Analysis 1.4). We judged the trial as having overall high risk of bias, with both missing outcome data and measurement of outcomes judged at high risk of bias (see Analysis 1.4). Caution should therefore be applied when these results are interpreted.

1.4. Analysis.

1.4

Comparison 1: Exercise versus no exercise, Outcome 4: All‐cause hospitalisation at longest follow‐up

Health‐related quality of life

Pressler 2016 and Sibilitz 2016 reported HRQoL in a total of 139 participants using the 12‐Item and 36‐Item Short‐Form Health Survey questionnaires (SF‐12 and SF‐36), respectively. These questionnaires were subdivided into mental component and physical component sub‐scores, assessed at baseline, at completion of the intervention, and at longest follow‐up. At completion of the intervention (ranging from two to three months), there was no difference between exercise‐based CR and control groups in these sub‐scores (mental component: mean difference (MD) 1.28, 95% CI ‐1.60 to 4.16; fixed‐effect model; 2 trials, 150 participants; I2 = 0%; GRADE quality of evidence very low; Analysis 1.5; physical component: MD 2.99, 95% CI ‐5.24 to 11.21; random‐effects model; 2 trials, 150 participants; I2 = 79%; GRADE quality of evidence very low; Analysis 1.6). At longest follow‐up (six months in Sibilitz 2016 and 24 months in Pressler 2018), there was also no difference in sub‐scores  (mental component: MD ‐1.45, 95% CI ‐4.70 to 1.80; fixed‐effect model; 2 trials, 139 participants; I2 = 0%; GRADE quality of evidence very low; Analysis 1.7; physical component: MD ‐0.87, 95% CI ‐3.57 to 1.83; fixed‐effect model; 2 trials, 139 participants; I2 = 0%; GRADE quality of evidence very low; Analysis 1.8).

1.5. Analysis.

1.5

Comparison 1: Exercise versus no exercise, Outcome 5: HRQoL (mental component) at end of intervention

1.6. Analysis.

1.6

Comparison 1: Exercise versus no exercise, Outcome 6: HRQoL (physical component) at end of intervention

1.7. Analysis.

1.7

Comparison 1: Exercise versus no exercise, Outcome 7: HRQoL (mental component) at maximum follow‐up

1.8. Analysis.

1.8

Comparison 1: Exercise versus no exercise, Outcome 8: HRQoL (physical component) at maximum follow‐up

The overall risk of bias for both mental component and physical component sub‐scores at completion of the intervention and at maximum follow‐up was 'high' for both Sibilitz 2016 and Pressler 2016. Both trials also had some concerns for missing outcome data. Caution should therefore be applied when this outcome is interpreted; GRADE quality of evidence was very low.

Secondary outcomes

Exercise capacity

All six trials reported exercise capacity in 321 participants assessed as VO2 peak/max (Lin 2004; Nilsson 2019; Sire 1987), as six‐minute walk test (6MWT) (Rogers 2018), or as both (Pressler 2016; Sibilitz 2016). All trials reporting VO2 max were converted to mL/kg/min, except Sire 1987, which could not be recalculated from reported kilojoules. Due to these differences in reporting, exercise capacity is presented in three ways: (1) direct measures of VO2 max data in mL/kg/min across four trials, (2) maximal measures (contained all peak exercise capacity data as standardised mean difference (SMD) across five trials), and (3) submaximal data based on 6MWT from three trials.

At completion of the intervention, and compared to control, exercise‐based CR resulted in a moderate increase in exercise capacity for maximal measures (SMD 0.38, 95% CI 0.15 to 0.61; fixed‐effect model; 5 trials, 194 participants; I2 = 0%; Analysis 1.11) and direct measures of VO2 max (MD 2.38 mL/kg/min, 95% CI 0.36 to 4.40; 4 trials, 250 participants; I2 = 0%; fixed‐effect model; Analysis 1.9) but not for submaximal 6MWT (MD ‐3.89 metres, 95% CI ‐58.72 to 50.95; 3 trials, 167 participants; I2 = 85%; random‐effects model; Analysis 1.13).

1.11. Analysis.

1.11

Comparison 1: Exercise versus no exercise, Outcome 11: Exercise capacity (maximum measures) at end of Intervention

1.9. Analysis.

1.9

Comparison 1: Exercise versus no exercise, Outcome 9: Exercise capacity (direct: VO2 max) at end of intervention

1.13. Analysis.

1.13

Comparison 1: Exercise versus no exercise, Outcome 13: Exercise capacity (indirect/submaximal: 6MWT) at end of Intervention

At longest follow‐up, moderate benefit in favour of exercise was still seen for maximal measures (SMD 0.37, 95% CI 0.13 to 0.61; 5 trials, 284 participants; I2 = 0%; fixed‐effect model; Analysis 1.12) but not for direct measures of VO2 max (MD 1.53 mL/kg/min, 95% CI ‐0.44 to 3.50; 4 trials, 240 participants; I2 = 0%; fixed‐effect model; Analysis 1.10) nor of 6MWT (MD ‐25.48 meters, 95% CI ‐103.04 to 52.08; 3 trials, 158 participants; I2 = 84%; random‐effects model; Analysis 1.14).

1.12. Analysis.

1.12

Comparison 1: Exercise versus no exercise, Outcome 12: Exercise capacity (maximum measures) at longest follow‐up

1.10. Analysis.

1.10

Comparison 1: Exercise versus no exercise, Outcome 10: Exercise capacity (direct: VO2 max) at longest follow‐up

1.14. Analysis.

1.14

Comparison 1: Exercise versus no exercise, Outcome 14: Exercise capacity (indirect/submaximal: 6MWT) at longest follow‐up

Serious adverse events

A total of 23 serious adverse events (exercise‐based CR 12/164 (7.3%) versus control 11/162 (6.8%)) were reported across four trials (Lin 2004; Pressler 2016; Sibilitz 2016; Sire 1987), with no differences between groups (RR 1.07, 95% CI 0.50 to 2.27; 4 trials, 326 participants; I2 = 0%; fixed‐effect model; Analysis 1.15Table 3).

1.15. Analysis.

1.15

Comparison 1: Exercise versus no exercise, Outcome 15: Serious adverse events

2. Description of severe adverse events.
  Lin 2004 Sire 1987 Pressler 2018 Sibilitz 2016 Total events
No exercise group 3 patients:
1 pericardial effusion
1 paravalvular leakage
1 endocarditis
2 patients:
2 non‐fatal thromboembolism
5 patients:
5 died before 24 months'
follow‐up
1 patient:
Not reported
11
Exercise group 4 patients:
2 heart arrhythmias
1 sudden death
1 brain stem death
2 patients:
1 haematoma in abdominal muscle
1 angina pectoris
4 patients:
(but not due to exercise)
1 fell due to icy conditions leading
to severe cerebral trauma
1 lethal cerebral haemorrhage due to
oral anticoagulant
2 died before 24 months' follow‐up
2 patients:
(but not due to exercise)
1 postsurgical cardiac tamponade
1 heart failure‐related re‐admission
12
Return to work

Only one trial reported return to work in a total of 44 participants (Sire 1987). At 12 months' follow‐up, there was no difference in the proportion of participants who had returned to work in the exercise‐based CR group (4/21; 19%) compared to the control group (8/23; 35%) (RR 1.24, 95% CI 0.86 to 1.79; Analysis 1.16).

1.16. Analysis.

1.16

Comparison 1: Exercise versus no exercise, Outcome 16: Return to work

Costs and cost‐effectiveness

Only Sibilitz 2016 reported economic data, with cost data collected in the trial from the time of surgery to six months' follow‐up and assessed from a societal perspective (Hansen 2017). Although there was no difference between exercise‐CR and control in HRQoL or costs (see Table 4) driven by a trend towards cost savings with CR, trial authors reported a probability ≥ 75% that CR was cost‐effective (Hansen 2017).

3. Mean total societal cost.
Exercise group Control group Group difference (95% CI) Statistical Significance
14,185 Euros/patient 17,448 Euros/patient ‐1609 Euros/patient (‐6162 to 2942) NS

Table showing mean total societal cost between exercise‐CR and control groups from Sibilitz 2016. NS: not statistically significant. 

Subgroup analyses

Due to the small number of included trials and a limited quantity of data, it was not possible to perform any of the planned subgroup analyses.

Discussion

Summary of main results

We identified six randomised trials including a total of 364 people following open or percutaneous valve surgery who received exercise‐based cardiac rehabilitation (CR) or the no exercise control. Two trials reported a total of nine deaths, one trial reported one hospitalisation, and evidence of the impact on health‐related quality of life (HRQoL) was of very low certainty. Exercise‐based CR programmes in these trials were consistently based on aerobic exercise and were in accord with the European Society of Cardiology recommendation for physical activity for secondary prevention (Ambrosetti 2020; Corra 2010). In summary, although potentially beneficial in terms of short‐term exercise capacity, data remain inadequate for definitive assessment of the impact of exercise‐based CR on the key patient‐related primary outcomes of mortality, hospitalisations, and HRQoL.

Overall completeness and applicability of evidence

Several issues need to be addressed when implications of the findings of this review are interpreted for daily clinical practice. First and foremost, the generalisability of the findings of this review is limited by the small quantity of data identified. Furthermore, almost all included trials recruited highly selected trial populations consisting of younger participants with low to moderate risk and few women, except for Sibilitz 2016, which included a broad representation of participants. Throughout the last decade, novel valve repair techniques have evolved, including less invasive techniques such as percutaneous valve procedures, with resultant changes in the treatment and participant pathway following valve repair or replacement; without sternotomy, exercise‐based CR programmes can start earlier and patients are older with more co‐morbidities. Included trials provide few data on postsurgical complications, such as hospitalisation, atrial fibrillation, pericardial exudate, and impact on overall HRQoL. These considerations are important when postsurgery management is planned, especially after open heart surgery, and when suitable patients are selected for a rehabilitation programme after valve surgery. In summary, the applicability of the evidence in this review to current practice is limited, and the generalisability of results should be interpreted with caution.

Quality of the evidence

We judged all primary outcomes to have 'very low' quality of evidence based on GRADE analysis. The quality of evidence for total mortality was 'very low' and was downgraded for inconsistency and small sample size/numbers of events. The quality of evidence for hospitalisation admission was 'very low' and was downgraded for risk of bias, inconsistency, and small sample size/numbers of events. The quality of evidence for HRQoL was judged to be 'very low', with downgrading due to small sample size/numbers of events, inconsistency, and lack of patient blinding (with the HRQoL physical component score at completion of the intervention also having high statistical heterogeneity).

Potential biases in the review process

We conducted this updated review according to recommendations provided in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2019c). We followed our peer‐reviewed published protocol (Sibilitz 2013b), with its predefined participants, interventions, comparisons, and outcomes, to avoid biases during review preparation. We performed a comprehensive literature search to identify published and unpublished trials, abided by our prespecified inclusion and exclusion criteria, and conducted the meta‐analysis using available data or based it on intention‐to‐treat when possible. However, the bias of omission of full copies of papers that may have included important data due to no response from study authors is difficult to assess.

The included trials were relatively small and had short‐term follow‐up and small numbers of reported events (mortality, hospitalisations, and serious adverse events). With the exception of Sibilitz 2016, none of the included trials sought to formally collect mortality or serious adverse events as outcomes, and we were able to capture these outcomes from studies based only on their reporting of losses to follow‐up and dropouts. Translation of Lin 2004, which was published in Chinese, may have resulted in reporting bias.

Agreements and disagreements with other studies or reviews

Since the time this Cochrane Review was first published, two other non‐Cochrane systematic reviews and meta‐analyses have been published (Anayo 2019; Ribeiro 2017). The review by Ribeiro and colleagues (5 uncontrolled before‐and‐after studies, 862  patients) showed that the six‐minute walk distance test (6MWT) significantly improved with exercise‐based CR compared to control (standardised mean difference (SMD) 0.69, 95% confidence interval (CI) 0.47 to 0.91). Similarly, the Anayo et al review (3 randomised controlled trials (RCTs) and 3 non‐RCTs, 255 participants) showed improvement in 6MWT favouring exercise‐based CR (mean difference (MD) 22.90 metres, 95% CI −31.64 to 77.43). Although the present review found no clear evidence of improvement in 6MWT with exercise‐based CR, our finding of improvement in short‐term exercise capacity with CR is consistent with the findings of both of these previous reviews. In accord with this review, Anayo et al found no difference between exercise‐based CR and control in 12‐Item/36‐Item Short‐Form Health Survey questionnaire (SF‐12/36) HRQoL scores (mental component: MD −0.44, 95% CI −3.43 to 2.56; physical component: MD 2.81, 95% CI −5.82 to 11.44).

Authors' conclusions

Implications for practice.

Current European Society of Cardiology guidelines recommend exercise‐based CR following heart valve surgery. However, this updated systematic review of randomised controlled trial evidence shows that a more cautious recommendation is needed. In particular, the impact of exercise‐based CR after heart valve surgery on mortality, serious adverse events, HRQoL, return to work, and costs remains unclear. Additionally, its impact on postsurgical adverse events needs to be further investigated, and this information used to inform targeting of exercise‐based CR to the most relevant heart valve patients. Nevertheless, our review supports the potential use of exercise‐based CR to improve short‐term exercise capacity following heart valve surgery.

The trials included in this review have investigated CR interventions based on exercise training. It is widely accepted that contemporary CR should be 'comprehensive' and should incorporate risk factor education/counselling and psychosocial interventions (Anderson 2014; Corra 2010). For use post valve surgery, CR interventions may also need to include breathing and coughing exercises and vocational evaluation advice. Moreover, due to the risk of complications and of hospitalisations, a CR programme for heart valve surgery patients also needs to address medical issues and medical stabilisation, along with anticoagulation treatment, and needs to provide thorough information about endocarditis prophylaxis. An important question for future updates on CR is whether patients could benefit from alternative modalities to centre‐based CR, including home‐based programmes.

Implications for research.

To date, research evidence for CR has focused mainly on trials showing the benefits of CR in ischaemic heart disease (post myocardial infarction and revascularisation) and heart failure. This updated systematic review shows that further randomised controlled trial evidence at low risk of bias is needed to definitively assess the impact of exercise‐based CR on patients following valve surgery. Information is especially needed on the outcomes that matter most to patients, clinicians, and policymakers (i.e. mortality, hospitalisations, HRQoL, return to work, and costs and cost‐effectiveness).

We identified 10 ongoing (information from clinicaltrials.gov and World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP)) randomised controlled trials, most of which are still recruiting. These trials seek to include a total of 2435 participants (with sample sizes ranging from 30 to 800 participants/trial) and report that they are collecting a range of outcomes that include mortality, exercise capacity, HRQoL, hospitalisations, and adverse events.

Critique of this new evidence should include the following considerations.  

  • Trial quality including consideration of sample size calculation based on participant‐relevant outcomes that may include composite events (such as mortality and hospitalisation) and health‐related quality of life  and conduct/reporting in accordance with the Consolidated Standards of Reporting Trials (CONSORT) guidelines for non‐pharmacological interventions (Boutron 2008).  

  • CR interventions that address the specific needs and preferences of heart valve patients with focus on maximising uptake, such as home‐based programmes (especially given the global impact of the COVID‐19 pandemic on healthcare systems).

  • Routine reporting of fidelity to CR prescription delivery and patient adherence.  

  • Generalisability of trial populations to practice (i.e. inclusion of women, patients with baseline phenotypes including different types of valve lesions, open versus percutaneous and replacement versus repair valve surgery, inclusion of older participants).

  • Long‐term follow‐up (≥ 12 months) to fully assess the clinical and cost‐effectiveness implications of CR.

What's new

Date Event Description
9 July 2020 New search has been performed Four new studies added in updated review with evidence current to January 2020
6 July 2020 New citation required but conclusions have not changed We conducted an update of the previous systematic review and meta‐analysis to assess randomised clinical trial evidence for the use of exercise‐based CR following heart valve surgery

History

Protocol first published: Issue 12, 2013
Review first published: Issue 3, 2016

Risk of bias

Risk of bias for analysis 1.1 All‐cause mortality at longest follow‐up.

Study Bias
Randomisation process Deviations from intended interventions Missing outcome data Measurement of the outcome Selection of the reported results Overall
Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement
Lin 2004 Some concerns The review authors noted that the only information reported about randomisation is a statement that the patients were randomly divided into experimental or control groups. There is no reference to concealment of allocation sequence. The review authors judged there to be no imbalance between the two study groups according to baseline characteristics. Some concerns Given the nature of an exercise‐based CR intervention,  the review authors judged participants were aware of their intervention and it might not be possible to blind them, their carers, or the people delivering this intervention. The triallists did not provide any information to help us ascertain whether people may or may not have deviated from the intended intervention. The triallists did not state if any analysis was carried out to estimate the effect of assignment to the intervention. The review authors judged that there was no reason to believe that participants were excluded from the analysis or assessed in the wrong intervention group. High risk of bias The triallists reported that only participants who did not drop out and were not lost to follow‐up were included in analysis. The triallists state that they used available case analysis. Even though the outcome was mortality, which is registered centrally, because the trial only included participants who were not lost to follow‐up, this might have affected the number of deaths recorded in this trial. The triallists did not provide enough information to allow us to assess further whether or not the missingness depended on its true value. Low risk of bias Mortality was reported from a central database; because of the nature of the outcome and due to it being reported centrally, measurement or ascertainment could not have differed between the two groups. The triallists did not provide any information with regard to blinding of the outcome assessors/assessment. The review authors judged that knowledge of the assigned intervention was unlikely to influence all‐cause mortality as this was an observer‐reported outcome that did not involve judgement. Low risk of bias Although there was no protocol published for this study and the triallists made no mention of whether there was a prespecified analysis plan. Furthermore, all‐cause mortality was centrally recorded and the review authors judged that there can not have been bias in selection of this outcome. High risk of bias At least one risk of bias domain judged as 'High'.
Pressler 2016 Low risk of bias The triallists stated that “patients were randomly allocated 1:1 according to a computer‐generated allocation program code”. Although there was no mention of allocation sequence concealment, the fact that a computer‐generated program code was used, leads the review authors to judge that this would prevent the enrolling investigator and the participant from having any knowledge of the forthcoming allocation. The triallists state no differences for TG and UC with respect to age, proportion of females, the Society of Thoracic Surgeons mortality risk score, NYHA class, pre‐existing conditions, or medication (see Table I). Low risk of bias Given the nature of an exercise‐based cardiac rehabilitation intervention, participants were aware of their intervention and it might not be possible to blind them, their carers, or the people delivering this intervention. Although the triallists did not specifically state if there were any deviations from the intended intervention, the flow of participants through the study (Figure 1 of 2016 paper and Figure 1 of 2018 paper) showed that no participant changed groups during the study and all changes to the intervention were either consistent with the intervention (i.e. dropout due to unrelated causes) or consistent with what could occur outside the trial context (i.e. dropout due to unwillingness to continue participation).
Triallists tested their data for their distribution and the primary end point and key secondary outcomes were normally distributed; thus it was decided to present data as mean (standard deviation). Baseline differences between groups were analysed using the Student t test for independent samples for continuous variables and the Fisher exact test for categorical variables. Between‐group differences in changes of the primary and secondary end points were analysed using the Student t test for independent samples. For relevant effect sizes (mean differences between study groups), 95% CIs are presented. Because of the low number of dropouts that were all unrelated to the intervention and the primary intention of the study to evaluate efficacy and safety of exercise, a per‐protocol analysis was performed by the triallists. To account for observed differences in baseline parameters between the study groups, linear regression models using the change in the measure of interest as a dependent variable and the study group as well as the baseline value of the measure of interest as independent variable were fitted to the data by the triallists.
Low risk of bias Triallists presented outcome data for all participants. Low risk of bias Mortality was reported from a central database; because of the nature of the outcome and due to it being reported centrally, the review authors judged that measurement or ascertainment could not have differed between the two groups. The triallists stated that "All endpoints were performed and supervised by experienced staff blinded to original group assignments". Low risk of bias The trial was registered on clinicaltrials.gov before start of the study (NCT01935297) with researchers’ prespecified intentions available in sufficient detail. As all‐cause mortality was centrally recorded, the the review authors that there cannot have been selection bias of this outcome. The triallists provided results for all time points at which this outcome was assessed, with all extra analysis intended as exploratory analysis to provide additional information on the group‐specific impact of the intervention. The triallists reported all results. Low risk of bias All risk of bias domains have judgement of low risk of bias.

Risk of bias for analysis 1.4 All‐cause hospitalisation at longest follow‐up.

Study Bias
Randomisation process Deviations from intended interventions Missing outcome data Measurement of the outcome Selection of the reported results Overall
Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement
Sibilitz 2016 Low risk of bias The triallists stated that “Participants were allocated 1:1 to intervention or control using computer‐generated allocation sequence with varying block sizes of 8, 6 and 4, concealed to the investigators by central telephone correspondence with the Copenhagen Trial Unit.”
“There was no evidence of baseline imbalances.” The review authors judged that table 1 shows a good balance in patient characteristics between groups.
Low risk of bias Given the nature of an exercise‐based CR intervention, participants were aware of their intervention and the review authors judged it not possible to blind them, their carers, or the people delivering this intervention. Although the triallists did not expressly mention if there were any deviations from the intended intervention, their CONSORT flow diagram of patients (Figure 1 in article) shows that all participants were analysed in their initial intervention group.
The triallists state the following: analysis was based on intention to treat using a mixed model with repeated measures (MMRM) and adjusted for the stratification variable of LVEF, ensuring that missing data will not create bias as long as the values are missing at random. The MMRM was used for continuous outcomes (both primary and secondary outcomes). This model assumes normally (Gaussian) distributed residuals. However, the residuals were not normally distributed for the primary outcome, and so we transformed them. We used log‐transformation, and when log‐transformation did not result in normally distributed residuals, we used Box–Cox transformation. In the MMRM we assumed correlation within the individual patient, but no correlation between patients. The fixed effects for the primary outcome were randomisation group, time, interaction between random and time and LVEF.
Some concerns Considering that the appropriate study population for an analysis of the intention‐to‐treat effect is all of the randomised participants, in this study less than 95% of the participants in both the exercise group and the control group provided data for this outcome. The triallists stated that the dropouts were due to complications after surgery and withdrawal of consent;  9 people dropped out of the study before completing 1 month of exercise training or usual care. The review authors judged that there is no evidence to show that the results were not biased by missing outcome data, as the only sensitivity analysis carried out by the triallists was for VO2. The review authors judged that there was a minimal difference in the proportions of missing outcome data between the two groups. The triallists stated that the dropouts were due to complications after surgery and withdrawal of consent; there was no suggestion that the intervention played a part in withdrawals. Triallists reported reasons for missing outcome data were similar in both groups. The circumstances of the trial did not make it likely that missingness in the outcome depended on its true value, as more participants dropped out of the control group than the exercise group. Low risk of bias Because of its central reporting, the review authors judged that measurement or ascertainment of this outcome could not have differed between the two groups. “The trial applied central, stratified randomisation securing against selection bias, and blinded outcome assessment and blinded statistical analyses, reducing detection and interpretation bias”. Low risk of bias The trial was registered on clinicaltrials.gov before start of the study (NCT01558765) with researchers’ prespecified intentions available in sufficient detail. The triallists reported results for all time points at which this outcome was assessed (6 months) within the scope of this article. There was no evidence of multiple eligible analyses for this outcome. Some concerns At least one risk of bias domain judged as 'Some concerns'.

Risk of bias for analysis 1.5 HRQoL (mental component) at end of intervention.

Study Bias
Randomisation process Deviations from intended interventions Missing outcome data Measurement of the outcome Selection of the reported results Overall
Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement
Pressler 2016 Low risk of bias The triallists stated that “patients were randomly allocated 1:1 according to a computer‐generated allocation program code”. Although there was no mention of allocation sequence concealment, the fact that a computer‐generated program code was used, leads the review authors to judge that this would prevent the enrolling investigator and the participant from having any knowledge of the forthcoming allocation. The triallists state no differences for TG and UC with respect to age, proportion of females, the Society of Thoracic Surgeons mortality risk score, NYHA class, pre‐existing conditions, or medication (see Table I). Low risk of bias The nature of an exercise‐based CR intervention means that participants are aware of their intervention and it might not be possible to blind them, their carers, or the people delivering this intervention. Although the triallists did not state specifically if there were any deviations from the intended intervention, the flow of participants through the study (Figure 1 of 2016 paper) showed that no participant changed groups during the study and all changes to the intervention were either consistent with the intervention (i.e. dropout due to unrelated causes) or consistent with what could occur outside the trial context (i.e. dropout due to unwillingness to continue participating).
Triallists reported that they tested their data distribution and the primary end point and key secondary outcomes were normally distributed; thus it was decided to present data as mean (SD). Baseline differences between groups were analysed by triallists using the Student t test for independent samples for continuous variables and the Fisher exact test for categorical variables. Between‐group differences in changes of the primary and secondary end points were analysed using the Student t test for independent samples. For relevant effect sizes (mean differences between study groups), 95% CIs are presented. Because of the low number of dropouts that were all unrelated to the intervention and the primary intention of the study to evaluate efficacy and safety of exercise, a per‐protocol analysis was performed by triallists. To account for observed differences in baseline parameters between the study groups, linear regression models using the change in the measure of interest as dependent variable and the study group as well as the baseline value of the measure of interest as independent variable were fit to the data by triallists.
Low risk of bias The appropriate study population for an analysis of the intention‐to‐treat effect is all of the randomised participants; in this trial, the review authors noted that less than 95% of the participants in both the exercise group and the control group provided data for this outcome, as there were three dropouts overall. Because of the low number of dropouts that were all unrelated to the intervention and the primary intention of the study to evaluate efficacy and safety of exercise, the triallists performed a per‐protocol analysis. High risk of bias Health related qualit of life (HRQoL) was measured using SF36 and KCCQ, both of which are validated questionnaires. The triallists reported that measurements were made in the same way for both intervention and control.The person reporting this outcome was the participant; participants were aware of the intervention they received. Collation and analysis of the results for this outcome were most likely carried out by study personnel who were blinded to the intervention received by the study participant.
The assessment by the participants could have been affected by knowledge of the intervention as this is a subjective outcome. Exercise is generally believed to be a beneficial intervention in this patient group and it is possible that participants might have considered their HRQoL would be better if they had the intervention. Low risk of bias The trial was registered on clinicaltrials.gov before the start of the study (NCT01935297), with the researchers’ prespecified intentions available in sufficient detail. The review authors judged that there was no reason to believe that the data which produced this result were not analysed in accordance with the prespecified analysis plan. The triallists reported all results. The review authors judged there was therefore no reason to believe that this result might have been selected from multiple eligible analyses of the data. High risk of bias At least one risk of bias domain has a 'High' risk of bias judgement.
Sibilitz 2016 Low risk of bias The triallists stated that “Participants were allocated 1:1 to intervention or control using computer‐generated allocation sequence with varying block sizes of 8, 6 and 4, concealed to the investigators by central telephone correspondence with the Copenhagen Trial Unit.”
“There was no evidence of baseline imbalances.” The review authors judged that table 1 shows a good balance in patient characteristics between groups.
Low risk of bias The nature of an exercise‐based CR intervention means that participants are aware of their intervention and it might not be possible to blind them, their carers, or the people delivering this intervention. Although the triallists did not mention expressly if there were any deviations from the intended intervention, their CONSORT flow diagram of patients (Figure 1 in article) shows that all participants were analysed in their initial intervention group.
The triallists reported: the analysis was based on intention to treat using a mixed model with repeated measures (MMRM) and adjusted for the stratification variable of LVEF, ensuring that missing data will not create bias as long as the values are missing at random. The MMRM was used for continuous outcomes (both primary and secondary outcomes). This model assumes normally (Gaussian) distributed residuals. However, the residuals were not normally distributed for the primary outcome, and therefore we transformed the data. We used log‐transformation, and when log‐transformation did not result in normally distributed residuals, we used Box–Cox transformation. In the MMRM we assumed correlation within individual patients, but no correlation between patients. 
Some concerns The appropriate study population for an analysis of the intention‐to‐treat effect is all of the randomised participants; in this trial less than 95% of the participants in both the exercise group and the control group provided data for this outcome. The triallists stated that the dropouts were due to complications after surgery and withdrawal of consent;  9 people dropped out of the study before completing 1 month of exercise training or usual care. There was no evidence to show that the results were not biased by missing outcome data. There was a minimal difference in the proportions of missing outcome data between the two groups. The triallists stated that the dropouts were due to complications after surgery and withdrawal of consent, no suggestion was made that the intervention played a part in this withdrawal. The reported reasons for missing outcome data were similar in both groups.
The circumstances of the trial do not make it likely that missingness in the outcome depends on its true value as more participants dropped out of the control group than from the exercise group.
High risk of bias HRQoL was measured using SF36, a validated questionnaire. The triallists reported that measurements were made in the same way for both intervention and control. The person reporting this outcome was the participant and participants were aware of the intervention they received. Collation and analysis of the results was most likely carried out by study personnel who were blinded. “The trial applied central, stratified randomisation securing against selection bias, and blinded outcome assessment and blinded statistical analyses, reducing detection and interpretation bias”. The review authors judged that the assessment by the participants could have been affected by knowledge of the intervention as it was subjective. Exercise is generally believed to be a beneficial intervention in this patient group, and it is possible that participants might have considered their HRQoL would be better if they had the intervention. Low risk of bias The trial was registered on clinicaltrials.gov before it started (NCT01558765), with researchers’ pre‐specified intentions available in sufficient detail. The review authors therefore judged that there was no reason to believe that the results would not have been analysed in accordance with the prespecified analysis plan. The triallists reported results for all time points at which this outcome was assessed within the scope of this article. The review authors judged that there was no evidence of multiple eligible analyses for this outcome. High risk of bias At least one risk of bias domain had a judgement of 'High' risk of bias.

Risk of bias for analysis 1.6 HRQoL (physical component) at end of intervention.

Study Bias
Randomisation process Deviations from intended interventions Missing outcome data Measurement of the outcome Selection of the reported results Overall
Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement
Pressler 2016 Low risk of bias The triallists stated that “patients were randomly allocated 1:1 according to a computer‐generated allocation program code”. Although there was no mention of allocation sequence concealment, the fact that a computer‐generated program code was used, leads the review authors to judge that this would prevent the enrolling investigator and the participant from having any knowledge of the forthcoming allocation. The triallists state no differences for TG and UC with respect to age, proportion of females, the Society of Thoracic Surgeons mortality risk score, NYHA class, pre‐existing conditions, or medication (see Table I). Low risk of bias The nature of an exercise‐based CR intervention means that participants are aware of their intervention and it might not be possible to blind them, their carers, or the people delivering this intervention. Although the triallists did not state specifically if there were any deviations from the intended intervention, the flow of participants through the study (Fig 1 of 2016 paper) shows that no participant changed groups during the study and all changes to the intervention were either consistent with the intervention (i.e. dropout due to unrelated causes) or consistent with what could occur outside the trial context (i.e. dropout due to unwillingness to continue participation).
Triallists reported that data were tested for their distribution and the primary end point and key secondary outcomes were normally distributed; thus it was decided to present data as mean (SD). Baseline differences between groups were analysed using the Student t test for independent samples for continuous variables and the Fisher exact test for categorical variables. Between‐group differences in changes of the primary and secondary end points were analysed using the Student t test for independent samples. For relevant effect sizes (mean differences between study groups), 95% CIs are presented. Because of the low number of dropouts that were all unrelated to the intervention and the primary intention of the study to evaluate efficacy and safety of exercise, a per‐protocol analysis was performed. To account for observed differences in baseline parameters between the study groups, linear regression models using the change in the measure of interest as dependent variable and the study group as well as the baseline value of the measure of interest as independent variable were fitted to the data.
Low risk of bias The appropriate study population for an analysis of the intention‐to‐treat effect is all of the randomised participants; in this trial less than 95% of the participants in both the exercise group and the control group provided data for this outcome, as there were three dropouts overall. Because of the low number of dropouts that were all unrelated to the intervention and the primary intention of the study to evaluate efficacy and safety of exercise, the triallists performed a per‐protocol analysis. High risk of bias HRQoL was measured using SF36 and KCCQ, both of which are validated questionnaires. The review authors noted that measurements were made in the same way for both intervention and control groups. The review authors noted that the person reporting this outcome was the participant; participants were aware of the intervention they received. Collation and analysis of the results for this outcome were most likely carried out by study personnel who were blinded to the intervention received by study participants. The review authors judged that the assessment by the participants could have been affected by knowledge of the intervention as this is a subjective outcome. Exercise is generally believed to be a beneficial intervention in this patient group and it is possible that participants might have considered their HRQoL would be better if they had the intervention. Low risk of bias The trial was registered on clinicaltrials.gov before start of the study (NCT01935297), with the researchers’ prespecified intentions available in sufficient detail. The review authors judged that there was no reason to believe that the data which produced this result were not analysed in accordance with the prespecified analysis plan. The triallists reported results for all time points at which this outcome was assessed with all extra analysis intended as exploratory analysis to provide additional information on the group‐specific impact of the intervention. The triallists reported all results. The review authors therefore have no reason to believe that this result might have been selected from multiple eligible analyses of the data. High risk of bias At least one risk of bias domain was rated as 'High'.
Sibilitz 2016 Low risk of bias The triallists stated that “Participants were allocated 1:1 to intervention or control using computer‐generated allocation sequence with varying block sizes of 8, 6 and 4, concealed to the investigators by central telephone correspondence with the Copenhagen Trial Unit.”
“There was no evidence of baseline imbalances.” The review authors judged that table 1 shows a good balance in patient characteristics between groups.
Low risk of bias The nature of an exercise‐based CR intervention means that participants are aware of their intervention and it might not be possible to blind them, their carers, or the people delivering this intervention. Although the triallists did not mention expressly mention if there were any deviations from the intended intervention, their CONSORT flow diagram of patients (Figure 1 in article) shows that all participants were analysed in their initial intervention group.
The triallists reported: the analysis was based on intention to treat using a mixed model with repeated measures (MMRM) and adjusted for the stratification variable of LVEF, ensuring that missing data would not create bias as long as the values were missing at random. The MMRM was used for continuous outcomes (both primary and secondary outcomes). This model assumes normally (Gaussian) distributed residuals. However, the residuals were not normally distributed for the primary outcome, so we transformed the data. We used log‐transformation, and when log‐transformation did not result in normally distributed residuals, we used Box–Cox transformation. In the MMRM we assumed correlation within the individual patient, but no correlation between patients. 
Some concerns The appropriate study population for an analysis of the intention‐to‐treat effect is all of the randomised participants; in this trial less than 95% of the participants in both the exercise group and the control group provided data for this outcome. The triallists stated that the dropouts were due to complications after surgery and withdrawal of consent;  9 people dropped out of the study before completing one month of exercise training or usual care. The review authors judged that there was no evidence to show that the results were not biased by missing outcome data. There is a minimal difference in the proportions of missing outcome data between the two groups. Reported reasons for missing outcome data were similar in both groups. The circumstances of the trial did not make it likely that missingness in the outcome depends on its true value as more participants dropped out of the control group than from the exercise group. High risk of bias HRQoL was measured using SF36, a validated questionnaire. The review authors judged that measurements were made in the same way for both intervention and control. The review authors noted that the person reporting this outcome was the participant and participants were aware of the intervention they received. Collation and analysis of the results was most likely carried out by study personnel who were blinded.
“The trial applied central, stratified randomisation securing against selection bias, and blinded outcome assessment and blinded statistical analyses, reducing detection and interpretation bias”. The review authors judged that the assessment by the participants could have been affected by knowledge of the intervention as it is subjective. Exercise is generally believed to be a beneficial intervention in this patient group, and it is possible that participants might have considered their HRQoL would be better if they had the intervention.
Low risk of bias Trial was registered on clinicaltrials.gov before it started (NCT01558765), with researchers’ pre‐specified intentions available in sufficient detail. The review authors judged that there was therefore no reason to believe that the results would not have been analysed in accordance with the pre‐specified analysis plan. The triallists provided results for all time points at which this outcome was assessed within the scope of this article. The review authors judged that there was no evidence of multiple eligible analyses for this outcome. High risk of bias At least one risk of bias domain was judged as 'High'.

Risk of bias for analysis 1.7 HRQoL (mental component) at maximum follow‐up.

Study Bias
Randomisation process Deviations from intended interventions Missing outcome data Measurement of the outcome Selection of the reported results Overall
Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement
Pressler 2016 Low risk of bias The triallists stated that “patients were randomly allocated 1:1 according to a computer‐generated allocation program code”. Although there was no mention of allocation sequence concealment, the fact that a computer‐generated program code was used, leads the review authors to judge that this would prevent the enrolling investigator and the participant from having any knowledge of the forthcoming allocation. The triallists state no differences for TG and UC with respect to age, proportion of females, the Society of Thoracic Surgeons mortality risk score, NYHA class, pre‐existing conditions, or medication (see Table I). Low risk of bias The nature of an exercise‐based CR intervention means that participants are aware of their intervention and it might not be possible to blind them, their carers, or the people delivering this intervention. Although the triallists did not state specifically if there were any deviations from the intended intervention, the flow of participants through the study (Figure 1 of 2016 paper) shows that no participant changed groups during the study and all changes to the intervention were either consistent with the intervention (i.e. drop out due to unrelated causes), or consistent with what could occur outside the trial context (i.e. dropout due to unwillingness to continue participation).
Triallists reported: data were tested for their distribution and the primary end point and key secondary outcomes were normally distributed; thus it was decided to present data as mean (SD). Baseline differences between groups were analysed using the Student t test for independent samples for continuous variables and the Fisher exact test for categorical variables. Between‐group differences in changes of the primary and secondary end points were analysed using the Student t test for independent samples. For relevant effect sizes (mean differences between study groups), 95% CIs are presented. Because of the low number of dropouts that were all unrelated to the intervention and the primary intention of the study to evaluate efficacy and safety of exercise, a per‐protocol analysis was performed. To account for observed differences in baseline parameters between the study groups, linear regression models using the change in the measure of interest as dependent variable and the study group as well as the baseline value of the measure of interest as independent variable were fitted to the data.
Some concerns The appropriate study population for an analysis of the intention‐to‐treat effect is all randomised participants; in this trial less than 95% of the participants in both the exercise group and the control group provided data for this outcome. The review authors judged that there was no evidence to suggest that the result was not biased by missing outcome data.
The review authors judged that although reported reasons for missing outcome data were similar between both groups, there was a minimal difference in the proportions of missing outcome data between the two groups overall with data for 5 participants missing from the exercise group and data for 8 participants missing from the control group. Most of the missing data were missing because of death or other health conditions that were not necessarily related to the intervention. The circumstances of the trial did not make it likely that missingness in the outcome depends on its true value as more participants dropped out of the control group than from the exercise group.
High risk of bias HRQoL was measured using SF36 and KCCQ, both of which are validated questionnaires. The review authors noted that measurements were made in the same way for both intervention and control. The person reporting this outcome was the participant; participants were aware of the intervention they received. Collation and analysis of the results was most likely carried out by study personnel who were blinded to the intervention received by study participants. The review authors judged that the assessment by the participants could have been affected by knowledge of the intervention as it is subjective. Exercise is generally believed to be a beneficial intervention in this patient group and it is possible that participants might have considered their HRQoL would be better if they had the intervention. Low risk of bias The trial was registered on clinicaltrials.gov before it started (NCT01935297), with researchers’ prespecified intentions available in sufficient detail. There was no reason to believe that the data which produced this result were not analysed in accordance with the prespecified analysis plan. The triallists reported results for all time points at which this outcome was assessed with all extra analysis intended as exploratory analysis to provide additional information on the group‐specific impact of the intervention. The triallists reported all results. There was no reason to believe that this result might have been selected from multiple eligible analyses of the data. High risk of bias At least one risk of bias domain with judgement of 'High'.
Sibilitz 2016 Low risk of bias The triallists stated that “Participants were allocated 1:1 to intervention or control using computer‐generated allocation sequence with varying block sizes of 8, 6 and 4, concealed to the investigators by central telephone correspondence with the Copenhagen Trial Unit.”
“There was no evidence of baseline imbalances.” The review authors judged that table 1 shows a good balance in patient characteristics between groups.
Low risk of bias The nature of an exercise‐based CR intervention means that participants are aware of their intervention and it might not be possible to blind them, their carers, or the people delivering this intervention. Although the triallists did not mention expressly if there were any deviations from the intended intervention, their CONSORT flow diagram of patients (Figure 1 in article) shows that all participants were analysed in their initial intervention group.
The triallists reported: the analysis was based on intention to treat using a mixed model with repeated measures (MMRM) and adjusted for the stratification variable of LVEF, ensuring that missing data will not create bias as long as the values are missing at random. The MMRM was used for continuous outcomes (both primary and secondary outcomes). This model assumes normally (Gaussian) distributed residuals. However, the residuals were not normally distributed for the primary outcome, and therefore transformed. We used log‐transformation, and when log‐transformation did not result in normally distributed residuals, we used Box–Cox transformation. In the MMRM we assumed correlation within the individual patient, but no correlation between patients.
Some concerns The appropriate study population for an analysis of the intention‐to‐treat effect is all of the randomised participants; in this trial less than 95% of the participants in both the exercise group and the control group provided data for this outcome. The triallists stated that the dropouts were due to complications after surgery and withdrawal of consent;  9 people dropped out of the study before completing 1 month of exercise training or usual care, no suggestion was made that the intervention played a part in these withdrawals. The review authors judged there was no evidence to show that the results was not biased by missing outcome data. The review authors noted that there was a minimal difference in the proportions of missing outcome data between the two groups. Reported reasons for missing outcome data were similar between both groups. The circumstances of the trial did not make it likely that missingness in the outcome would depend on its true value as more participants dropped out of the control group than from the exercise group. High risk of bias HRQoL was measured using SF36, a validated questionnaire. The review authors noted measurements were made in the same way for both intervention and control. The person reporting this outcome was the participant and participants are aware of the intervention they received. Collation and analysis of the results was most likely carried out by blinded study personnel .
“The trial applied central, stratified randomisation securing against selection bias, and blinded outcome assessment and blinded statistical analyses, reducing detection and interpretation bias”. 
The review authors judged assessment by the participants could have been affected by knowledge of the intervention, as it is subjective. Exercise is generally believed to be a beneficial intervention in this patient group and it is possible that participants might have considered their HRQoL would be better if they had the intervention.
Low risk of bias Trial was registered on clinicaltrials.gov before it started (NCT01558765), with researchers’ prespecified intentions available in sufficient detail. The review authors judged that there is therefore no reason to believe that the results would not have been analysed in accordance with the prespecified analysis plan. The triallists reported results for all time points at which this outcome was assessed within the scope of this article. The review authors noted that there was no evidence of multiple eligible analyses for this outcome. High risk of bias At least one risk of bias domain with judgement of 'High'.

Risk of bias for analysis 1.8 HRQoL (physical component) at maximum follow‐up.

Study Bias
Randomisation process Deviations from intended interventions Missing outcome data Measurement of the outcome Selection of the reported results Overall
Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement Authors' judgement Support for judgement
Pressler 2016 Low risk of bias The triallists stated that “patients were randomly allocated 1:1 according to a computer‐generated allocation program code”. Although there was no mention of allocation sequence concealment, the fact that a computer‐generated program code was used, leads the review authors to judge that this would prevent the enrolling investigator and the participant from having any knowledge of the forthcoming allocation. The triallists state no differences for TG and UC with respect to age, proportion of females, the Society of Thoracic Surgeons mortality risk score, NYHA class, pre‐existing conditions, or medication (see Table I). Low risk of bias The nature of an exercise‐based CR intervention means that participants are aware of their intervention and it might not be possible to blind them, their carers, or the people delivering this intervention. Although the triallists did not state specifically if there were any deviations from the intended intervention, the flow of participants through the study (Figure 1 of 2016 paper) shows that no participant changed groups during the study and all changes to the intervention were either consistent with the intervention (i.e. dropout due to unrelated causes) or consistent with what could occur outside the trial context (i.e. drop out due to unwillingness to continue participation).
Triallists tested data for their distribution and the primary end point and key secondary outcomes were normally distributed; thus it was decided to present data as mean (SD). Baseline differences between groups were analysed using the Student t test for independent samples for continuous variables and the Fisher exact test for categorical variables. Between‐group differences in changes of the primary and secondary end points were analysed using the Student t test for independent samples. For relevant effect sizes (mean differences between study groups), 95% CIs are presented. Because of the low number of dropouts that were all unrelated to the intervention and the primary intention of the study to evaluate efficacy and safety of exercise, a per‐protocol analysis was performed. To account for observed differences in baseline parameters between the study groups, linear regression models using the change in the measure of interest as dependent variable and the study group as well as the baseline value of the measure of interest as independent variable were fit to the data.
Some concerns The appropriate study population for an analysis of the intention‐to‐treat effect is all randomised participants; in this trial less than 95% of the participants in both the exercise group and the control group provided data for this outcome. There was no evidence to suggest that the result was not biased by missing outcome data. Although reported reasons for missing outcome data were similar between both groups, there was a minimal difference in the proportions of missing outcome data between the two groups overall with data for 5 participants missing from the exercise group and data for 8 participants missing from the control group. Most of the missing data were missing because of death or other health conditions that were not necessarily related to the intervention. The circumstances of the trial did not make it likely that missingness in the outcome depends on its true value as more participants dropped out of the control group than from the exercise group. High risk of bias HRQoL was measured using SF36 and KCCQ, both of which are validated questionnaires. Measurements were made in the same way for both intervention and control. The person reporting this outcome was the participant and participants are aware of the intervention they received. Collation and analysis of the results was most likely carried out by study personnel who were blinded to the intervention received by the study participant. The assessment by the participants could have been affected by knowledge of the intervention, as it is subjective. Exercise is generally believed to be a beneficial intervention in this patient group and it is possible that participants might have considered their HRQoL would be better if they had the intervention. Low risk of bias The trial was registered on clinicaltrials.gov before it started (NCT01935297), with researchers’ prespecified intentions available in sufficient detail. There was no reason to believe that the data which produced this result were not analysed in accordance with the prespecified analysis plan. The triallists provided results for all time points at which this outcome was assessed with all extra analysis intended as exploratory analysis to provide additional information on the group‐specific impact of the intervention. The triallists reported all results. There is therefore no reason to believe that this result might have been selected from multiple eligible analyses of the data. High risk of bias At least one risk of bias domain with a judgement of 'High'.
Sibilitz 2016 Low risk of bias The triallists stated that “Participants were allocated 1:1 to intervention or control using computer‐generated allocation sequence with varying block sizes of 8, 6 and 4, concealed to the investigators by central telephone correspondence with the Copenhagen Trial Unit.”
“There was no evidence of baseline imbalances.” The review authors judged that table 1 shows a good balance in patient characteristics between groups.
Low risk of bias The nature of an exercise‐based CR intervention means that participants are aware of their intervention and it might not be possible to blind them, their carers, or the people delivering this intervention. Although the triallists did not mention expressly if there were any deviations from the intended intervention, their CONSORT flow diagram of patients (Figure 1 in article) shows that all participants were analysed in their initial intervention group.
The triallists reported: analysis was based on intention to treat using a mixed model with repeated measures (MMRM) and adjusted for the stratification variable of LVEF, ensuring that missing data would not create bias as long as the values are missing at random. The MMRM was used for continuous outcomes (both primary and secondary outcomes). This model assumes normally (Gaussian) distributed residuals. However, the residuals were not normally distributed for the primary outcome, and therefore transformed. We used log‐transformation, and when log‐transformation did not result in normally distributed residuals, we used Box–Cox transformation. In the MMRM we assumed correlation within the individual patient, but no correlation between patients. 
Some concerns The appropriate study population for an analysis of the intention‐to‐treat effect is all of the randomised participants; in this trial less than 95% of the participants in both the exercise group and the control group provided data for this outcome. The triallists stated that the dropouts were due to complications after surgery and withdrawal of consent; 9 participants dropped out of the study before completing 1 month of exercise training or usual care. There was no suggestion that the intervention played a part in withdrawals. The review authors judged that there was no evidence to show that the results were not biased by missing outcome data. There is a minimal difference in the proportions of missing outcome data between the two groups. The review authors judged that the reported reasons for missing outcome data are similar between both groups. The review authors judged that the circumstances of the trial do not make it likely that missingness in the outcome depends on its true value as more participants dropped out of the control group than from the exercise group. High risk of bias HRQoL was measured using SF36, a validated questionnaire. The review authors noted that measurements were made in the same way for both intervention and control. The person reporting this outcome was the participant and participants are aware of the intervention they received. Collation and analysis of the results was most likely carried out by blinded study personnel. “The trial applied central, stratified randomisation securing against selection bias, and blinded outcome assessment and blinded statistical analyses, reducing detection and interpretation bias”. The review authors judged that the assessment by the participants could have been affected by knowledge of the intervention, as it is subjective. Exercise is generally believed to be a beneficial intervention in this patient group, and it is possible that participants might have considered their HRQoL would be better if they had the intervention. Low risk of bias Trial was registered on clinicaltrials.gov before it started (NCT01558765), with researchers’ prespecified intentions available in sufficient detail. The review authors judged that there is therefore no reason to believe that the results would not have been analysed in accordance with the prespecified analysis plan.
The triallists reported results for all time points at which this outcome was assessed within the scope of this article. The review authors judge that there was no evidence of multiple eligible analyses for this outcome. 
High risk of bias At least one risk of bias domain with a judgement of 'High'.

Acknowledgements

We thank all the authors who provided important contributions to drafting of this review. We are indebted to Swenyu Hu, Danish Institute for Trial Abroad, Copenhagen, Denmark, and Henry Lishi Li, London School of Hygiene and Tropical Medicine, London, UK, for their excellent translation services. Further, we thank Dr. Lindsey Anderson for alignment of this review with the other Cochrane Reviews in the cardiac rehabilitation portfolio.

We thank Christian Gluud, Lars Kober, and Christian Hassager, who were also co‐authors of the previous version of this review; Michele Hilton Boon, who provided an independent RoB2 assessment for primary outcomes of the Sibilitz 2016 trial; and Dr. Wilby Williamson from the University of Oxford, who peer‐reviewed the manuscript.

The Background and Methods sections of this review are based on a standard template provided by the Cochrane Heart Group.

Appendices

Appendix 1. Search strategies for review update

CENTRAL

#1 MeSH descriptor: [Exercise] explode all trees

#2 MeSH descriptor: [Exercise Therapy] explode all trees

#3 MeSH descriptor: [Exercise Tolerance] this term only

#4 MeSH descriptor: [Sports] explode all trees

#5 MeSH descriptor: [Physical Exertion] this term only

#6 exercis*

#7 sport*

#8 MeSH descriptor: [Physical Fitness] this term only

#9 MeSH descriptor: [Physical Education and Training] explode all trees

#10 (fitness or fitter or fit)

#11 muscle* near/3 (train* or activ*)

#12 train* near/5 (strength* or aerobic* or exercise*)

#13 (aerobic or resistance) near/3 (train* or activ*)

#14 physical* near/5 (fit* or train* or therap* or activ* or strength or endur* or exert* or capacit*)

#15 (exercise* or fitness) near/3 (treat* or interven* or program* or train* or physical or activ*)

#16 MeSH descriptor: [Rehabilitation] this term only

#17 MeSH descriptor: [Rehabilitation Centers] this term only

#18 rehabilitat*

#19 MeSH descriptor: [Dance Therapy] this term only

#20 kinesiotherap*

#21 danc*

#22 ("lifestyle" or life‐style) near/5 activ*

#23 ("lifestyle" or life‐style) near/5 physical*

#24 walk*

#25 run*

#26 jog*

#27 #1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or #14 or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 or #23 or #24 or #25 or #26

#28 MeSH descriptor: [Heart Valve Diseases] explode all trees

#29 valve near/2 (disease* or stenos* or insufficien*)

#30 MeSH descriptor: [Heart Valve Prosthesis Implantation] this term only

#31 MeSH descriptor: [Heart Valve Prosthesis] this term only

#32 (valve near/2 (surg* or replace* or repair* or prosthe* or implant* or procedure*))

#33 MitraClip

#34 MeSH descriptor: [Transcatheter Aortic Valve Replacement] this term only

#35 TAVI

#36 #28 or #29 or #30 or #31 or #32 or #33 or #34 or #35

#37 #27 and #36

MEDLINE Ovid

1 exp Exercise/

2 exp Exercise Therapy/

3 Exercise Tolerance/

4 exp Sports/

5 Physical Exertion/

6 exercis*.tw.

7 sport*.tw.

8 Physical Fitness/

9 exp "Physical Education and Training"/

10 (fitness or fitter or fit).tw.

11 (muscle* adj3 (train* or activ*)).tw.

12 (train* adj5 (strength* or aerobic* or exercise*)).tw.

13 ((aerobic or resistance) adj3 (train* or activ*)).tw.

14 (physical* adj5 (fit* or train* or therap* or activ* or strength or endur* or exert* or capacit*)).tw.

15 ((exercise* or fitness) adj3 (treat* or interven* or program* or train* or physical or activ*)).tw.

16 Rehabilitation/

17 Rehabilitation Centers/

18 rehabilitat*.tw.

19 Dance Therapy/

20 kinesiotherap*.tw.

21 danc*.tw.

22 (("lifestyle" or life‐style) adj5 activ$).tw.

23 (("lifestyle" or life‐style) adj5 physical$).tw.

24 walk*.tw.

25 run*.tw.

26 jog*.tw.

27 or/1‐26

28 exp Heart Valve Diseases/

29 (valve adj2 (disease* or stenos* or insufficien*)).tw.

30 Heart Valve Prosthesis Implantation/

31 Heart Valve Prosthesis/

32 (valve adj2 (surg* or replace* or repair* or prosthe* or implant* or procedure*)).tw.

33 MitraClip.tw.

34 Transcatheter Aortic Valve Replacement/

35 TAVI.tw.

36 or/28‐35

37 27 and 36

38 randomized controlled trial.pt.

39 controlled clinical trial.pt.

40 randomized.ab.

41 placebo.ab.

42 drug therapy.fs.

43 randomly.ab.

44 trial.ab.

45 groups.ab.

46 38 or 39 or 40 or 41 or 42 or 43 or 44 or 45

47 exp animals/ not humans.sh.

48 46 not 47

49 37 and 48

Embase Ovid

1 exp exercise/

2 exp kinesiotherapy/

3 exercise tolerance/

4 exp sport/

5 exercis*.tw.

6 sport*.tw.

7 fitness/

8 physical education/

9 (fitness or fitter or fit).tw.

10 (muscle* adj3 (train* or activ*)).tw.

11 (train* adj5 (strength* or aerobic* or exercise*)).tw.

12 ((aerobic or resistance) adj3 (train* or activ*)).tw.

13 (physical* adj5 (fit* or train* or therap* or activ* or strength or endur* or exert* or capacit*)).tw.

14 ((exercise* or fitness) adj3 (treat* or interven* or program* or train* or physical or activ*)).tw.

15 Rehabilitation/

16 rehabilitation center/

17 rehabilitat*.tw.

18 dance therapy/

19 kinesiotherap*.tw.

20 danc*.tw.

21 (("lifestyle" or life‐style) adj5 activ$).tw.

22 (("lifestyle" or life‐style) adj5 physical$).tw.

23 walk*.tw.

24 run*.tw.

25 jog*.tw.

26 or/1‐25

27 exp valvular heart disease/

28 (valve adj2 (disease* or stenos* or insufficien*)).tw.

29 exp heart valve replacement/

30 exp heart valve prosthesis/

31 (valve adj2 (surg* or replace* or repair* or prosthe* or implant* or procedure*)).tw.

32 MitraClip.tw.

33 transcatheter aortic valve implantation/

34 TAVI.tw.

35 or/27‐34

36 26 and 35

37 random$.tw.

38 factorial$.tw.

39 crossover$.tw.

40 cross over$.tw.

41 cross‐over$.tw.

42 placebo$.tw.

43 (doubl$ adj blind$).tw.

44 (singl$ adj blind$).tw.

45 assign$.tw.

46 allocat$.tw.

47 volunteer$.tw.

48 crossover procedure/

49 double blind procedure/

50 randomized controlled trial/

51 single blind procedure/

52 37 or 38 or 39 or 40 or 41 or 42 or 43 or 44 or 45 or 46 or 47 or 48 or 49 or 50 or 51

53 (animal/ or nonhuman/) not human/

54 52 not 53

55 36 and 54

56 limit 55 to embase

CINAHL

S50 S31 AND S49

S49 S32 OR S33 OR S34 OR S35 OR S36 OR S37 OR S38 OR S39 OR S40 OR S41 OR S42 OR S43 OR S44 OR S45 OR S46 OR S47 OR S48

S48 TX cross‐over*

S47 TX crossover*

S46 TX volunteer*

S45 (MH "Crossover Design")

S44 TX allocat*

S43 TX control*

S42 TX assign*

S41 TX placebo*

S40 (MH "Placebos")

S39 TX random*

S38 TX (doubl* N1 mask*)

S37 TX (singl* N1 mask*)

S36 TX (doubl* N1 blind*)

S35 TX (singl* N1 blind*)

S34 TX (clinic* N1 trial?)

S33 PT clinical trial

S32 (MH "Clinical Trials+")

S31 S22 AND S30

S30 S23 OR S24 OR S25 OR S26 OR S27 OR S28 OR S29

S29 TAVI

S28 transcatheter aortic valve replacement*

S27 MitraClip

S26 valve N2 (surg* or replace* or repair* or prosthe* or implant* or procedure*)

S25 (MH "Heart Valve Prosthesis")

S24 valve N2 (disease* or stenos* or insufficien*)

S23 (MH "Heart Valve Diseases+")

S22 S1 OR S2 OR S3 OR S4 OR S5 OR S6 OR S7 OR S8 OR S9 OR S10 OR S11 OR S12 OR S13 OR S14 OR S15 OR S16 OR S17 OR S18 OR S19 OR S20 OR S21

S21 walk* or run* or jog*

S20 (("lifestyle" or life‐style) N5 physical*)

S19 (("lifestyle" or life‐style) N5 activ*)

S18 kinesiotherap* or danc*

S17 (MH "Dance Therapy")

S16 rehabilitat*

S15 (MH "Rehabilitation Centers")

S14 (MH "Rehabilitation")

S13 (exercise* or fitness) N3 (treat* or interven* or program* or train* or physical or activ*)

S12 physical* N5 (fit* or train* or therap* or activ* or strength or endur* or exert* or capacit*)

S11 (aerobic or resistance) N3 (train* or activ*)

S10 train* N5 (strength* or aerobic* or exercise*)

S9 muscle* N3 (train* or activ*)

S8 fitness or fitter or fit

S7 (MH "Physical Education and Training+")

S6 (MH "Physical Fitness")

S5 exercis* or sport*

S4 (MH "Sports+")

S3 (MH "Exercise Tolerance+")

S2 (MH "Therapeutic Exercise+")

S1 (MH "Exercise+")

PsycINFO

1 exp Exercise/

2 exp Sports/

3 exercis*.tw.

4 sport*.tw.

5 physical fitness/

6 physical education/

7 (fitness or fitter or fit).tw.

8 (muscle* adj3 (train* or activ*)).tw.

9 (train* adj5 (strength* or aerobic* or exercise*)).tw.

10 ((aerobic or resistance) adj3 (train* or activ*)).tw.

11 (physical* adj5 (fit* or train* or therap* or activ* or strength or endur* or exert* or capacit*)).tw.

12 ((exercise* or fitness) adj3 (treat* or interven* or program* or train* or physical or activ*)).tw.

13 rehabilitation/

14 rehabilitation centers/

15 rehabilitat*.tw.

16 dance therapy/

17 kinesiotherap*.tw.

18 danc*.tw.

19 (("lifestyle" or life‐style) adj5 activ*).tw.

20 (("lifestyle" or life‐style) adj5 physical*).tw.

21 walk*.tw.

22 run*.tw.

23 jog*.tw.

24 or/1‐23

25 (valve adj2 (disease* or stenos* or insufficien*)).tw.

26 prostheses/

27 (valve adj2 (surg* or replace* or repair* or prosthe* or implant* or procedure*)).tw.

28 MitraClip.tw.

29 TAVI.tw.

30 25 or 26 or 27 or 28 or 29

31 24 and 30

32 random$.tw.

33 factorial$.tw.

34 crossover$.tw.

35 cross‐over$.tw.

36 placebo$.tw.

37 (doubl$ adj blind$).tw.

38 (singl$ adj blind$).tw.

39 assign$.tw.

40 allocat$.tw.

41 volunteer$.tw.

42 control*.tw.

43 "2000".md.

44 or/32‐43

45 31 and 44

LILACS

(exercis$ or sport$ or fit$ or train$ or activ$ or aerobic$ or rehabilit$ or walk$ or jog$ or run$) [Words] and ("heart valve$" or "heart prosthe$") [Words]

CPCI‐S

# 18 #17 AND #16

# 17 TS=(random* or blind* or allocat* or assign* or trial* or placebo* or crossover* or cross‐over*)

# 16 #15 AND #10

# 15 #14 OR #13 OR #12 OR #11

# 14 TS=TAVI

# 13 TS=MitraClip

# 12 TS=(valve and (surg* or replace* or repair* or prosthe* or implant* or procedure*))

# 11 TS=(valve and (disease* or stenos* or insufficien*))

# 10 #9 OR #8 OR #7 OR #6 OR #5 OR #4 OR #3 OR #2 OR #1

# 9 TS=(("lifestyle" or life‐style) and physical*)

# 8 TS=(("lifestyle" or life‐style) and activ*)

# 7 TS=(rehabilitat* or danc* or kinesiotherap* or walk* or run* or jog*)

# 6 TS=((exercise* or fitness) and (treat* or interven* or program* or train* or physical or activ*))

# 5 TS=(physical* and (fit* or train* or therap* or activ* or strength or endur* or exert* or capacit*))

# 4 TS=((aerobic or resistance) and (train* or activ*))

# 3 TS=(train* and (strength* or aerobic* or exercise*))

# 2 TS=(muscle* and (train* or active*))

# 1 TS=(exercis* or sport* or fitness or fitter or fit)

Appendix 2. Previous search strategies

Cochrane Library

#1MeSH descriptor: [Exercise] explode all trees
#2MeSH descriptor: [Exercise Therapy] explode all trees
#3MeSH descriptor: [Exercise Tolerance] this term only
#4MeSH descriptor: [Sports] explode all trees
#5MeSH descriptor: [Physical Exertion] this term only
#6exercis*
#7sport*
#8MeSH descriptor: [Physical Fitness] this term only
#9MeSH descriptor: [Physical Education and Training] explode all trees
#10(fitness or fitter or fit)
#11muscle* near/3 (train* or activ*)
#12train* near/5 (strength* or aerobic* or exercise*)
#13(aerobic or resistance) near/3 (train* or activ*)
#14physical* near/5 (fit* or train* or therap* or activ* or strength or endur* or exert* or capacit*)
#15(exercise* or fitness) near/3 (treat* or interven* or program* or train* or physical or activ*)
#16MeSH descriptor: [Rehabilitation] this term only
#17MeSH descriptor: [Rehabilitation Centers] this term only
#18rehabilitat*
#19MeSH descriptor: [Dance Therapy] this term only
#20kinesiotherap*
#21danc*
#22("lifestyle" or life‐style) near/5 activ*
#23("lifestyle" or life‐style) near/5 physical*
#24walk*
#25run*
#26jog*
#27#1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or #14 or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 or #23 or #24 or #25 or #26
#28MeSH descriptor: [Heart Valve Diseases] explode all trees
#29valve near/2 (disease* or stenos* or insufficien*)
#30MeSH descriptor: [Heart Valve Prosthesis Implantation] this term only
#31MeSH descriptor: [Heart Valve Prosthesis] this term only
#32valve near/2 (surg* or replace* or repair* or prosthe*)
#33#28 or #29 or #30 or #31 or #32
#34#27 and #33

MEDLINE

1 exp Exercise/
2 exp Exercise Therapy/
3 Exercise Tolerance/
4 exp Sports/
5 Physical Exertion/
6 exercis*.tw.
7 sport*.tw.
8 Physical Fitness/
9 exp "Physical Education and Training"/
10 (fitness or fitter or fit).tw.
11 (muscle* adj3 (train* or activ*)).tw.
12 (train* adj5 (strength* or aerobic* or exercise*)).tw.
13 ((aerobic or resistance) adj3 (train* or activ*)).tw.
14 (physical* adj5 (fit* or train* or therap* or activ* or strength or endur* or exert* or capacit*)).tw.
15 ((exercise* or fitness) adj3 (treat* or interven* or program* or train* or physical or activ*)).tw.
16 Rehabilitation/
17 Rehabilitation Centers/
18 rehabilitat*.tw.
19 Dance Therapy/
20 kinesiotherap*.tw.
21 danc*.tw.
22 (("lifestyle" or life‐style) adj5 activ$).tw.
23 (("lifestyle" or life‐style) adj5 physical$).tw.
24 walk*.tw.
25 run*.tw.
26 jog*.tw.
27 or/1‐26
28 exp Heart Valve Diseases/
29 (valve adj2 (disease* or stenos* or insufficien*)).tw.
30 Heart Valve Prosthesis Implantation/
31 Heart Valve Prosthesis/
32 (valve adj2 (surg* or replace* or repair* or prosthe*)).tw.
33 or/28‐32
34 27 and 33
35 randomized controlled trial.pt.
36 controlled clinical trial.pt.
37 randomized.ab.
38 placebo.ab.
39 drug therapy.fs.
40 randomly.ab.
41 trial.ab.
42 groups.ab.
43 35 or 36 or 37 or 38 or 39 or 40 or 41 or 42
44 exp animals/ not humans.sh.
45 43 not 44
46 34 and 45

Embase

1. exp exercise/
2. exp kinesiotherapy/
3. exercise tolerance/
4. exp sport/
5. exercis*.tw.
6. sport*.tw.
7. fitness/
8. fitness/
9. physical education/
10. (fitness or fitter or fit).tw.
11. (muscle* adj3 (train* or activ*)).tw.
12. (train* adj5 (strength* or aerobic* or exercise*)).tw.
13. ((aerobic or resistance) adj3 (train* or activ*)).tw.
14. (physical* adj5 (fit* or train* or therap* or activ* or strength or endur* or exert* or capacit*)).tw.
15. ((exercise* or fitness) adj3 (treat* or interven* or program* or train* or physical or activ*)).tw.
16. rehabilitation/
17. rehabilitation center/
18. rehabilitat*.tw.
19. dance therapy/
20. kinesiotherap*.tw.
21. danc*.tw.
22. (("lifestyle" or life‐style) adj5 activ$).tw.
23. (("lifestyle" or life‐style) adj5 physical$).tw.
24. walk*.tw.
25. run*.tw.
26. jog*.tw.
27. or/1‐26
28. exp valvular heart disease/
29. (valve adj2 (disease* or stenos* or insufficien*)).tw.
30. exp heart valve replacement/
31. exp heart valve prosthesis/
32. (valve adj2 (surg* or replace* or repair* or prosthe*)).tw.
33. or/28‐32
34. 27 and 33
35. random$.tw.
36. factorial$.tw.
37. crossover$.tw.
38. cross over$.tw.
39. cross‐over$.tw.
40. placebo$.tw.
41. (doubl$ adj blind$).tw.
42. (singl$ adj blind$).tw.
43. assign$.tw.
44. allocat$.tw.
45. volunteer$.tw.
46. crossover procedure/
47. double blind procedure/
48. randomized controlled trial/
49. single blind procedure/
50. 35 or 36 or 37 or 38 or 39 or 40 or 41 or 42 or 43 or 44 or 45 or 46 or 47 or 48 or 49
51. (animal/ or nonhuman/) not human/
52. 50 not 51
53. 34 and 52
54. limit 53 to embase

CINAHL

S47 S28 AND S46
S46 S29 or S30 or S31 or S32 or S33 or S34 or S35 or S36 or S37 or S38 or S39 or S40 or S41 or S42 or S43 or S44 or S45
S45 TX cross‐over*
S44 TX crossover*
S43 TX volunteer*
S42 (MH "Crossover Design")
S41 TX allocat*
S40 TX control*
S39 TX assign*
S38 TX placebo*
S37 (MH "Placebos")
S36 TX random*
S35 TX (doubl* N1 mask*)
S34 TX (singl* N1 mask*)
S33 TX (doubl* N1 blind*)
S32 TX (singl* N1 blind*)
S31 TX (clinic* N1 trial?)
S30 PT clinical trial
S29 (MH "Clinical Trials+")
S28 S22 AND S27
S27 S23 OR S24 OR S25 OR S26
S26 valve N2 (surg* or replace* or repair* or prosthe*)
S25 (MH "Heart Valve Prosthesis")
S24 valve N2 (disease* or stenos* or insufficien*)
S23 (MH "Heart Valve Diseases+")
S22 S1 OR S2 OR S3 OR S4 OR S5 OR S6 OR S7 OR S8 OR S9 OR S10 OR S11 OR S12 OR S13 OR S14 OR S15 OR S16 OR S17 OR S18 OR S19 OR S20 OR S21
S21 walk* or run* or jog*
S20 (("lifestyle" or life‐style) N5 physical*)
S19 (("lifestyle" or life‐style) N5 activ*)
S18 kinesiotherap* or danc*
S17 (MH "Dance Therapy")
S16 rehabilitat*
S15 (MH "Rehabilitation Centers")
S14 (MH "Rehabilitation")
S13 (exercise* or fitness) N3 (treat* or interven* or program* or train* or physical or activ*)
S12 physical* N5 (fit* or train* or therap* or activ* or strength or endur* or exert* or capacit*)
S11 (aerobic or resistance) N3 (train* or activ*)
S10 train* N5 (strength* or aerobic* or exercise*)
S9 muscle* N3 (train* or activ*)
S8 fitness or fitter or fit
S7 (MH "Physical Education and Training+")
S6 (MH "Physical Fitness")
S5 exercis* or sport*
S4 (MH "Sports+")
S3 (MH "Exercise Tolerance+")
S2 (MH "Therapeutic Exercise+")
S1 (MH "Exercise+")

PsycINFO

1. exp Exercise/
2. exp Sports/
3. exercis*.tw.
4. sport*.tw.
5. physical fitness/
6. physical education/
7. (fitness or fitter or fit).tw.
8. (muscle* adj3 (train* or activ*)).tw.
9. (train* adj5 (strength* or aerobic* or exercise*)).tw.
10. ((aerobic or resistance) adj3 (train* or activ*)).tw.
11. (physical* adj5 (fit* or train* or therap* or activ* or strength or endur* or exert* or capacit*)).tw.
12. ((exercise* or fitness) adj3 (treat* or interven* or program* or train* or physical or activ*)).tw.
13. rehabilitation/
14. rehabilitation centers/
15. rehabilitat*.tw.
16. dance therapy/
17. kinesiotherap*.tw.
18. danc*.tw.
19. (("lifestyle" or life‐style) adj5 activ*).tw.
20. (("lifestyle" or life‐style) adj5 physical*).tw.
21. walk*.tw.
22. run*.tw.
23. jog*.tw.
24. or/1‐23
25. (valve adj2 (disease* or stenos* or insufficien*)).tw.
26. prostheses/
27. (valve adj2 (surg* or replace* or repair* or prosthe*)).tw.
28. or/25‐27
29. 24 and 28
30. random$.tw.
31. factorial$.tw.
32. crossover$.tw.
33. cross‐over$.tw.
34. placebo$.tw.
35. (doubl$ adj blind$).tw.
36. (singl$ adj blind$).tw.
37. assign$.tw.
38. allocat$.tw.
39. volunteer$.tw.
40. control*.tw.
41. "2000".md.
42. or/30‐41
43. 29 and 42

LILACS

(exercis$ or sport$ or fit$ or train$ or activ$ or aerobic$ or rehabilit$ or walk$ or jog$ or run$) [Words] and ("heart valve$" or "heart prosthe$") [Words]

CPCI‐S

# 16 #15 AND #14
# 15 TS=(random* or blind* or allocat* or assign* or trial* or placebo* or crossover* or cross‐over*)
# 14 #13 AND #10
# 13 #12 OR #11
# 12 TS=(valve and (surg* or replace* or repair* or prosthe*))
# 11 TS=(valve and (disease* or stenos* or insufficien*))
# 10 #9 OR #8 OR #7 OR #6 OR #5 OR #4 OR #3 OR #2 OR #1
# 9 TS=(("lifestyle" or life‐style) and physical*)
# 8 TS=(("lifestyle" or life‐style) and activ*)
# 7 TS=(rehabilitat* or danc* or kinesiotherap* or walk* or run* or jog*)
# 6 TS=((exercise* or fitness) and (treat* or interven* or program* or train* or physical or activ*))
# 5 TS=(physical* and (fit* or train* or therap* or activ* or strength or endur* or exert* or capacit*))
# 4 TS=((aerobic or resistance) and (train* or activ*))
# 3 TS=(train* and (strength* or aerobic* or exercise*))
# 2 TS=(muscle* and (train* or active*))
# 1 TS=(exercis* or sport* or fitness or fitter or fit)

Data and analyses

Comparison 1. Exercise versus no exercise.

Outcome or subgroup title No. of studies No. of participants Statistical method Effect size
1.1 All‐cause mortality at longest follow‐up 2 131 Risk Ratio (M‐H, Fixed, 95% CI) 0.83 [0.26, 2.68]
1.2 All‐cause mortality: best/worst‐case scenario 2 131 Risk Ratio (M‐H, Fixed, 95% CI) 0.44 [0.15, 1.32]
1.3 All‐cause mortality: worst/best‐case scenario 2 131 Risk Ratio (M‐H, Random, 95% CI) 2.15 [0.16, 28.78]
1.4 All‐cause hospitalisation at longest follow‐up 1   Risk Ratio (M‐H, Fixed, 95% CI) Totals not selected
1.5 HRQoL (mental component) at end of intervention 2 150 Mean Difference (IV, Fixed, 95% CI) 1.28 [‐1.60, 4.16]
1.6 HRQoL (physical component) at end of intervention 2 150 Mean Difference (IV, Random, 95% CI) 2.99 [‐5.24, 11.21]
1.7 HRQoL (mental component) at maximum follow‐up 2 139 Mean Difference (IV, Fixed, 95% CI) ‐1.45 [‐4.70, 1.80]
1.8 HRQoL (physical component) at maximum follow‐up 2 139 Mean Difference (IV, Fixed, 95% CI) ‐0.87 [‐3.57, 1.83]
1.9 Exercise capacity (direct: VO2 max) at end of intervention 4 250 Mean Difference (IV, Fixed, 95% CI) 2.38 [0.36, 4.40]
1.10 Exercise capacity (direct: VO2 max) at longest follow‐up 4 240 Mean Difference (IV, Fixed, 95% CI) 1.53 [‐0.44, 3.50]
1.11 Exercise capacity (maximum measures) at end of Intervention 5 294 Std. Mean Difference (IV, Fixed, 95% CI) 0.38 [0.15, 0.61]
1.12 Exercise capacity (maximum measures) at longest follow‐up 5 284 Std. Mean Difference (IV, Fixed, 95% CI) 0.37 [0.13, 0.61]
1.13 Exercise capacity (indirect/submaximal: 6MWT) at end of Intervention 3 167 Mean Difference (IV, Random, 95% CI) ‐3.89 [‐58.72, 50.95]
1.14 Exercise capacity (indirect/submaximal: 6MWT) at longest follow‐up 3 157 Mean Difference (IV, Random, 95% CI) ‐25.48 [‐103.04, 52.08]
1.15 Serious adverse events 4 326 Risk Ratio (M‐H, Fixed, 95% CI) 1.07 [0.50, 2.27]
1.16 Return to work 1   Risk Ratio (M‐H, Fixed, 95% CI) Totals not selected

1.2. Analysis.

1.2

Comparison 1: Exercise versus no exercise, Outcome 2: All‐cause mortality: best/worst‐case scenario

1.3. Analysis.

1.3

Comparison 1: Exercise versus no exercise, Outcome 3: All‐cause mortality: worst/best‐case scenario

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Lin 2004.

Study characteristics
Methods Study design: parallel‐group randomised controlled trial
No of centres: 1
Country: China
Dates patients recruited: NR
When randomised: NR
Maximum follow‐up (from baseline): 3 months
Participants Inclusion criteria: 20 to 45 years of age who have undergone single or double heart valve replacement
Exclusion criteria: comorbidities including pathological changes associated with coronary arteries, reoperations for valve replacement surgeries (patients who have undergone valve replacement before), severe pathological changes associated with other organs
N Randomised: total: 104; intervention: 55; comparator: 49
Number of participants lost to follow‐up: 7
Number of dropouts: 3 (2 due to irregular heart rhythm, 1 for delayed pericardial tamponade)
Number with complications: 4 (rehabilitation group: 1 sudden death, 1 brain stem disease; control group: 1 paravalvular leakage, 1 endocarditis)
 
Diagnosis (% of pts):
e.g.
Aetiology: the kind of valve disease is not specified; we assume that all kinds of valve diseases are included
Kind of surgery: mechanical valve replacement of any kind
NYHA: NR
LVEF: NR
Case mix: NR
Age (mean ± SD): total: NR; intervention: 32.8 ± 12.1;comparator: 29.8 ± 9.4
Percentage male: total: 56.73%; intervention: 56.36%;comparator: 57.14%
 
Ethnicity: NR
Interventions Intervention (exercise‐based CR)
Description
Type of rehabilitation programme: combined physical exercise, breathing exercises, and psychological intervention
Setting: hospital‐based and home‐based. At hospital and at home before and after surgery
Time after hospitalisation: the day after surgery, and continuing until 3 months after surgery
Total duration: starting the week before surgery with breathing exercises and psychological intervention, and the day after surgery with physical exercise
  • Psychological intervention


Conducted before surgery, to prevent anxiety and mental pressure before surgery. Introduction to the surgery in detail, and information about safety of the surgery
  • Breathing and coughing exercises


Conducted before and after surgery
Frequency and duration: 2 times a day 1 week before surgery and after surgery
Before surgery
Breathing exercises: lie down or sit up, pillow under knees, relax muscles in stomach, breathe in through the nose so stomach puffs up, breathe out through the nose. 10 to 12 times per minute. Patients monitor themselves
Coughing exercises: after deep breath, use chest and stomach power to cough as much as possible, 2 times daily, 20 times each session, the week before surgery. Breathing machine (Sherwood Voldyne) controls frequency. The patient can look over the results during exercises. Exercises are to be performed both sitting up and half lying down
After surgery
Day 1: stomach breathing exercise, coughing exercise to get rid of mucus, half lying down, relaxing whole body
Day 2: both breathing and coughing exercises
  • Physical exercise


Conducted after surgery. Includes limb stretch/joint exercises and aerobic exercises
Frequency: limb stretch/joint exercises: patients were advised to do this whenever they felt like it at home; aerobic exercise 2 to 3 times per week
Duration: 3 to 5 minutes limb stretch/joint exercises and 20 to 30 minutes aerobic exercise/session
Purpose: the purpose of the training is to increase endurance and increase pulmonary and cardiac capacity
At hospital (after surgery)
Day 2: joint exercises with passive arms and switch exercises
Day 3: joint exercises including both arm and leg exercises
Day 4: going out of the hospital, sitting, standing, getting out of bed, walking exercises. Aerobic exercises
At home (after discharge)
Resistance training: stretch arms and legs 3 to 5 minutes equivalent to 5 to 7 metabolic equivalents (METs) each session. Patients were encouraged to do the exercises whenever possible. The purpose of the exercises was to increase joint mobility, warm up the body, and relieve chest pressure
Aerobic exercise: consisted of walking slowly uphill, using treadmill or exercise bike at home. Goal of 5 to 7 METs per session
Intensity: not reported
Modality: not relevant
Both groups: follow regular principles and normal procedure for surgery. During surgery, the same equipment is used for all patients. After surgery, all patients receive the same quantities of analgesics, antibiotics, and anticoagulants
Comparator
Description: usual care by the hospital's heart doctor
Co‐interventions: NR
Outcomes Outcomes (scale measured in)
Postoperative incidence of pulmonary complications after surgery: measured once in all patients in % of control group and rehabilitation group, respectively, during the 3‐month period
Duration of hospitalisation for surgery: days of hospitalisation calculated once after all patients have been discharged after surgery. The number of days between groups was compared
Body activity energy level: measured at baseline and after 3 months in METs spent, using low strenuous physical exercises to test pulmonary and cardiac capacity
Besides outcome measurement, the purpose of the test was to determine for which patients the exercise could include potential risk and thus tailor the exercise plan in the most appropriate way
Other outcomes measured
Notes Follow‐up: 3 months from procedure
First author involved in patient selection, not in randomisation. Study authors emphasise that cardiac rehabilitation including physical exercise should be tailored and concrete, based on different patients' needs, and adjusted if necessary

Nilsson 2019.

Study characteristics
Methods Study design: RCT
No of centres: 1
Country: Sweden
Dates patients recruited: August 2011 and December 2014
When randomised: after surgery
Maximum follow‐up (from baseline): 1 year
Participants Inclusion criteria: all adult patients undergoing AVR due to AS
Exclusion criteria: any other concomitant cardiac disease, symptomatic lung disease, or mental or physical disability possibly limiting participation in the study
N Randomised: total: 12;intervention: 6;comparator: 6
 
Diagnosis (% of pts):
e.g.
Aetiology: (total): HR at rest TG (50 to 93), UC (48 to 91); SBP at rest TG (110 to 145), UC (110 to 170); DBP at rest (mmHg) TG (60 to 90), UC (70 to 95)
NYHA: NR
LVEF: NR
Case mix: NR
Age (mean ± SD): total: 62.5 (39 to 75); intervention: 58.5 (39 to 75);comparator: 65.5 (60 to 71)
Percentage male: total: 75%; intervention: 83.33%;comparator: 66.67%
 
Ethnicity: NR
Interventions Intervention (exercise‐based CR)
Description: the exercise training protocol was designed according to the most recent European position paper concerning exercise training in cardiac patients in addition to feasibility over a large span of age and fitness. Heart rate, workload (Watts), and perceived exertion (Borg RPE scale) were recorded every 5 minutes, and the workload was adjusted to preserve HR within the given interval according to the protocol
Time of start after event: 5 to 6 weeks postoperatively
Components: aerobic exercise
Detail of exercise: patients allocated to EX performed heart rate‐guided supervised exercise training on a bicycle ergometer
Modality: bicycle ergometer
Dose of exercise (calculated as overall no. of weeks of training multiplied by mean number of sessions per week multiplied by mean duration of sessions in minutes): 12 x 3 x 20 vigorous aerobic activity ± 12 x 5 x 30 light to moderate physical activity
Length of session: not clearly stated but about 45 to 60 minutes
Frequency/no. of sessions: 3 sessions per week
Intensity: workload was adjusted to preserve HR within the given interval according to the protocol
Resistance training included? NR
Total duration: 12 weeks
Setting: hospital
Supervision: yes, heart rate‐guided supervised
Intermittent nurse or exercise specialist support? NR
Co‐interventions: NR
Comparator
Description: patients in CON received the same general physical activity recommendations as those in EX at discharge and were contacted on 3 occasions during the 12 weeks to encourage them to follow these recommendations and to give them the opportunity to ask any questions connected to recovery and physical activity
Co‐interventions: NR
Outcomes Outcomes (scale measured in): peak VO2 measured during maximal exercise test on a cycle ergometer using cardiopulmonary exercise testing with oxygen uptake
Other outcomes measured
Effect on submaximal cardiopulmonary variables including oxygen uptake kinetics (tau), oxygen uptake efficiency slope (OUES), and ventilatory efficiency (VE/VCO2 slope)
Notes Follow‐up: baseline (i.e. 5 to 6 weeks postoperatively), at the end of the 12‐week intervention (i.e. 3 months from baseline), and 1 year hereafter
Study was supported by the Medical Research Council of Southeast Sweden (FORSS) and ALF Grants, Region Östergötland
Study authors have no conflicts of interest

Pressler 2016.

Study characteristics
Methods Study design: randomised controlled pilot trial
No of centres: 3
Country: Germany
Dates patients recruited: October 2012 to April 2014
When randomised: 83 ± 34 days (range 42 to 132) after intervention
Maximum follow‐up (from baseline): 24 ± 6 months
Participants Inclusion criteria: TAVI within previous 6 months, physically able and clinically stable to perform regular exercise as judged by study investigators, optimal medical treatment for cardiac and concomitant diseases, written informed consent. Only patients living within a reasonable distance from the exercise centre were contacted and were consecutively included in the screening process
Exclusion criteria: patients' decision to undergo TAVI despite receiving a recommendation for SAVR by the heart team (to avoid inclusion of atypical, low‐risk TAVI patients), physical disabilities making regular exercise impossible, unstable cardiac conditions (e.g. decompensated heart failure, New York Heart Association (NYHA) Class IV, severe rhythm disorders), uncontrolled hypertension or diabetes, severe obstructive pulmonary disease (forced expiratory volume in 1 second b50%). Patients were not included in cases of echocardiographic signs of prosthesis dysfunction according to the Valve Academic Research Consortium (valve orifice area of b1.2 cm2 plus mean transaortic pressure gradient ≥ 20 mmHg, or velocity ≥ 3 m/s, at least moderate paravalvular regurgitation, signs of ischaemia, severe arrhythmias, or haemodynamic deterioration during the initial exercise test)
N Randomised: total: 30;intervention: 13;comparator: 14
 
Diagnosis (% of pts)
e.g.
Aetiology: (total): aortic regurgitation (TG = 53%, UC = 73%), coronary artery disease (TG = 69%, UC = 71%), previous myocardial infarction (TG = 15%, UC = 35%), coronary artery bypass graft (TG = 23%, UC = 14%), atrial fibrillation (TG = 54%, UC = 36%), pacemaker/ICD (TG = 15%, UC = 21%), previous cerebrovascular event (TG = 8%, UC = 21%)
NYHA: TG: Class I: 1 (8), Class II: 10 (77), Class III: 2 (15); UC: Class I: 4 (29), Class II: 6 (42), Class III: 4 (29)
LVEF: TG: 58 ± 8%; UC: 57 ± 10%
Case mix: NR
Age (mean ± SD): total: 81 ± 6; intervention: 81 ± 7;comparator: 81 ± 5
Percentage male: 15/30 (50%):intervention: 47% (N = 7/15);comparator: 53% (8/15)
 
Ethnicity: NR
Interventions Intervention (exercise‐based CR)
Description: the training group received combined endurance and resistance exercise starting with 2 exercise sessions during the first week, followed by 3 sessions per week during Weeks 2 to 8. Resistance training started in Week 2 and was conducted subsequent to the endurance exercise portion in 2 of the 3 weekly workouts
Time of start after event: 81 days ± 27 days post TAVI in the exercise group;
84 days ± 41 days post TAVI in the usual care group
Components: exercise
Detail of exercise: exercise consisted of endurance training on cycle ergometers at moderate intensities, starting with 20 minutes and gradually increasing to 45 minutes by Week 8. Resistance training occurred after endurance training twice weekly from Week 2
Modality: cycle ergometer
Dose of exercise: (calculated as overall no. of weeks of training multiplied by mean number of sessions per week multiplied by mean duration of sessions in minutes): NR
Length of session: 20 to 45 minutes/session
Frequency/no. of sessions: Week 1: 2/week; Weeks 2 to 8: 2 to 3/week
Intensity : 45% to 70% VO2 peak
Resistance training included: yes + muscular endurance (bench press, rowing, shoulder press, pull‐down, leg press) 1 to 3 sets at 50% to 60% 1 RM
Total duration: 8 weeks
Setting: hospital
Supervision: Supervised
Intermittent nurse or exercise specialist support? NR
Co‐interventions: NR
Comparator
Description: usual care. Not receiving structured exercise
Co‐interventions: both groups received usual medical care
Outcomes Outcomes (scale measured in): exercise tolerance assessed by cardiopulmonary testing (VO2 peak), exercise capacity (6‐minute walk distance), HRQoL (KCCQ and SF‐12), mortality, all‐cause or cardiovascular
Other outcomes measured
Muscular strength with 1 repetition maximum testing, prosthetic aortic valve function with echocardiography
Notes Follow‐up: baseline, 8 weeks after baseline visit, 24 ± 6 months after baseline
This study received grant support from the German Heart Foundation/German Foundation of Heart Research (Frankfurt, Germany; F/14/12). Author BL received financial support from the German Cardiac Society (Düsseldorf, Germany) via the Otto‐Hess‐Research‐Grant
Conflict of interest: none declared
There were 3 dropouts: 2 from the training group that were unrelated to the intervention (1 had an accident, 1 had a lethal cerebral haemorrhage) and 1 from the usual care group who was not willing to continue in the study

Rogers 2018.

Study characteristics
Methods Study design: pilot RCT
No of centres: 1
Country: UK
Dates patients recruited: June 2016 to March 2017
When randomised: 4 weeks after TAVI
Maximum follow‐up (from baseline): 6 months post randomisation
Participants Inclusion criteria: severe symptomatic aortic stenosis accepted for TAVI in our institutional Multidisciplinary Team Meeting, age ≥ 75 years, able to give written informed consent, in the
Investigator’s opinion able to comply with all study requirements
Exclusion criteria: CR deemed inappropriate due to comorbidity or frailty, life expectancy < 1 year due to comorbidity, previous AVR or TAVI, predominant aortic regurgitation
N Randomised: total: 27;intervention: 14;comparator: 13
 
Diagnosis (% of pts)
e.g.
Aetiology: (total): previous MI, n (%), UC 2 (14.3), TG 3 (23.1); history of pulmonary disease, n (%), UC 4 (28.6), TG 3 (23.1); preoperative arrhythmia, n (%), UC 7 (50.0), TG 8 (61.5); previous cardiac surgery, n (%), UC 3 (21.4), TG 4 (30.8); previous PCI, n (%), UC 5 (35.7), TG 6 (46.2)
NYHA: NR
LVEF: ≥ 50% UC 12 (85.7%), TG 9 (69.2%); 30% to 49% UC 2 (14.3%), TG 3 (23.1%); < 30% UC 0, TG 1 (7.7%)
Case mix: NR
Age (mean ± SD): total: 82.04 ± 4.8; intervention: 82.92 ± 6.0;comparator: 81.21 ± 3.6
Percentage male: total: 44.4%; intervention: 46.2%;comparator: 42.9%
 
Ethnicity: NR
Interventions Intervention (exercise‐based CR)
Description: patients randomised to the intervention group underwent a comprehensive biopsychosocial assessment with a member of the exercise team, initiated 1 month post procedure and comprising once‐weekly sessions for 60 to 90 minutes for 6 sessions. An individualised programme was prescribed for each patient based on information gained from his/her functional capacity test and discussion around his/her specific goals
Time of start after event: 1 month post procedure
Components: exercise
Details of exercise: comprehensive biopsychosocial assessment comprising once‐weekly sessions for 60 to 90 minutes for 6 sessions. An individualised programme was then prescribed for each patient based on information gained from his/her functional capacity test and discussion around his/her specific goals. After each exercise session, each individual’s prescription was reviewed and was altered appropriately for the subsequent session. The intensity of the exercise was progressively increased based on self‐reported BORG intensity. Patients were offered further sessions if able to attend, in line with our institutional programme and British Association for Cardiovascular Prevention and Rehabilitation (BACPR) recommendations
Modality: exercise prescription consisted of graduated cardiovascular training and resistance training (both upper body and lower body) using cardiovascular exercise machines (treadmill and bike) as well as functional exercise such as ‘sit to stand’
Dose of exercise: (calculated as overall no. of weeks of training multiplied by mean number of sessions per week multiplied by mean number of sessions per week multiplied by mean duration of sessions in minutes): individualised
Length of session: individualised (avg ± SD: 7.5 ± 4.25) (77% completed 6 sessions; 3 participants completed 15, 13, and 12 sessions, respectively)
Frequency/no. of sessions: individualised
Intensity: individualised
Resistance training included? yes, + cardiovascular training
Total duration: individualised
Setting: hospital
Supervision: supervised
Intermittent nurse or exercise specialist support? NR
Co‐interventions: both control and intervention groups received routine medical care, which included an outpatient clinic follow‐up appointment, appropriate drug therapy, and concomitant medical management of co‐morbidities according to local practice
Comparator
Description: patients randomised to the control group received SOC according to our institutional protocols
Co‐interventions: both control and intervention groups received routine medical care, which included an outpatient clinic follow‐up appointment, appropriate drug therapy, and concomitant medical management of co‐morbidities according to local practice
Outcomes Outcomes (scale measured in): exercise capacity measured by 6‐minute walk test (6MWT), Nottingham Activities of Daily Living (ADL; scale of 0 for least activity to 22 for most activity), FRIED Frailty score (0 = not frail, 1 to 2 = pre‐frail, 3 = frail), Edmonton Frailty Score (9 domains, scale of 0 for non‐frail to 17 for severely frail), and Hospital Anxiety and Depression Scores (HADS, 0 to 7 normal, 8 to 10 borderline, 11 to 21 abnormal) score
Other outcomes measured
Thirty‐eight separate post‐TAVI patients completed the KCCQ with mean clinical summary score in a substudy
Notes Follow‐up: baseline (pre‐randomisation), 3 months and 6 months post randomisation
The RECOVER‐TAVI trial was funded through a pump priming grant from the Royal Brompton & Harefield NHS Foundation Trust Biomedical Research Unit
Conflicts of Interest: MD has received research grants, consultancy and proctorship fees from Astra Zeneca, Eli Lilly, Abbott Vascular, Daiichi Sankyo, Daiichi Sankyo, Lilly Alliance, Abbott Vascular, Sanofi, Medtronic, Boston Scientific, Edwards Lifesciences. NM has received honoraria, consultancy and proctorship fees from Abbott Vascular, Medtronic, and Edwards Lifesciences. MS has received research grants, consultancy and proctorship fees from Medtronic, Edwards Lifesciences, St Jude (now Abbott Vascular), and Boston Scientific. RST is the lead for the ongoing portfolio of Cochrane Reviews of cardiac rehabilitation. RST is a named scientific advisor for ongoing National Institutes of Health and Care Excellence (NICE) updated clinical guidelines for management of heart failure (CG108). HP is a member of the British Association for Cardiovascular Prevention and Rehabilitation (BACPR) and the Association of Chartered Physiotherapists in Cardiac Rehabilitation (ACPICR). HP chaired the referenced ACPICR Working Group for the national standards document
Thirteen control group patients completed the study assessment. Ten in the 13 intervention group completed the CR and assessment; 3 were too unwell to do so; and all patients were followed up

Sibilitz 2016.

Study characteristics
Methods Study design: randomised controlled trial
No of centres: 1
Country: Denmark
Dates patients recruited: 17 February 2012 and 7 May 2014
When randomised: after baseline outcome assessment
Maximum follow‐up (from baseline): 24 months (but data for 12 and 24 months recorded elsewhere)
Participants Inclusion criteria: elective right‐sided or left‐sided heart valve surgery, age ≥ 18 years, able to speak and understand Danish, ability to provide informed written consent
Exclusion criteria: known ischaemic heart disease before surgery, current recruitment to other rehabilitation trials or participating in trials precluding patients from participating, expected to not cooperate according to trial instructions, diseases in the musculoskeletal system, comorbidity complicating physical activity, competitive sports, and pregnancy and/or breastfeeding
N Randomised: total: 147;intervention: 72;comparator: 75
 
Diagnosis (% of pts)
e.g.
Aetiology: (total): atrial fibrillation 21% (intervention), 85% (control); symptoms before surgery are self‐reported and include dyspnoea, angina pectoris, palpitations, and decreased physical activity level – 92% (intervention), 92% (control)
NYHA: intervention NYHA Class I to II: 74%, Class III to IV: 26%; control NYHA Class I to II: 69%, Class III to IV: 31%
LVEF: intervention 55 ± 9.6 (89%); control 54 ± 10.2 (85%) ADD
Case mix: cardiac rehab group – aortic valve surgery 46 (64%), mitral valve surgery 27 (38%), pulmonal and tricuspid valve surgery 1 (1.4%)
Control group – aortic valve surgery 45 (60%), mitral valve surgery 26 (35%), pulmonal and tricuspid valve surgery 2 (3%)
Age (mean ± SD): total: 62; intervention: 62.0 ± 11.5;comparator: 61.0 ± 9.9
Percentage male: total: 76% (112/147); intervention: 82% (59/82);comparator: 71% (53/75)
 
Ethnicity: NR
Interventions Intervention (exercise‐based CR)
Description: exercise comprising 3 weekly exercise sessions for 12 weeks
Time of start after event: 1 month after surgery
Components: exercise
Detail of exercise: the programme consisted of graduated cardiovascular training (based on intensity on the Borg Scale, with progressively increasing intensity during the 12 weeks) and strength exercises (lower body exercises)
Modality: exercise training combining aerobic and resistance training
Dose of exercise: (calculated as overall no. of weeks of training multiplied by mean number of sessions per week multiplied by mean duration of sessions in minutes): NR
Length of session: 40 minutes/session (including 10‐minute warm‐up/10‐minute cool‐down)
Frequency/no. of sessions: 3 sessions/week
Intensity : 13 to 17 on Borg Scale
Resistance training included? yes, strength training for lower body (60% to 70% 1 RM)
Total duration: 12 weeks
Setting: home and hospital or local study protocol‐certified supervised facility
Supervision: hospital supervised, home unsupervised (had contact with a physiotherapist when indicated)
Intermittent nurse or exercise specialist support? NR
Co‐interventions: monthly psychoeducational consultations
Comparator
Description: all patients were provided early mobilisation immediately following surgery as part of usual care. Participants were not allowed to participate in a physical exercise programme
Co‐interventions: none
Outcomes Outcomes (scale measured in): exercise capacity (measured by VO2 peak) and self‐reported mental health (measured by Short Form‐36), 6MWT
Other outcomes measured
Notes Follow‐up: baseline; then 1, 4, and 6 months after randomisation
The Danish Strategic Research Foundation (10‐092790); the Heart Centre Research Council, Rigshospitalet; Familien Hede Nielsen Foundation (2013‐1226); National Institutes of Public Health, University of Southern Denmark; Region Zealand Health Research Foundation, Denmark (12‐000095/jun2014). Funders had no influence on trial design, execution of the trial, nor interpretation of data
Conflicts of interest: none declared
Due to pitfalls (such as calibration errors, flow errors, and mask leakage), 16 tests were estimated, with no overrepresentation in either randomisation group, using the following estimation equation: VO2 = 10.8 × (Watt max/weight) + 3.5. Estimation was validated on all measurements and was compared with non‐estimated values; the equation generally underestimated the VO2 peak value
Two serious adverse events were reported in the intervention group versus 1 in the control group at 6 months. Serious adverse events in the intervention group were evaluated as not caused by the intervention (1 with postsurgical cardiac tamponade and 1 with heart failure‐related re‐admission). Eleven of 72 (15.3%) in the intervention group versus 3 of 75 (4.0%) in the control group had self‐reported non‐serious adverse events (P = 0.02). These events were caused primarily by musculoskeletal problems and were related to exercise training in general
7 patients dropped out of the intervention group, and 11 dropped out of the control group due to complications after surgery and withdrawal of consent

Sire 1987.

Study characteristics
Methods Study design: prospective randomised study
No of centres: 1 trial centre but 2 patients received training at local hospital
Country: Norway
Dates patients recruited: NR
When randomised: 2 months after operation
Maximum follow‐up (from baseline): 12 months
Participants Inclusion criteria: had isolated aortic valve replacement and could tolerate and perform a physical training programme
Exclusion criteria: signs and symptoms of other heart disease, over 60 years of age, disease in the locomotor system, obvious mental ailments or social disturbances (e.g. alcoholics). Male patients with heart volumes exceeding 750 mL m‐2 BSA and females with hearts larger than 650 mL m‐2 BSA were also excluded
N Randomised: total: 44;intervention: 21;comparator: 23
 
Diagnosis (% of pts)
e.g.
Aetiology: (total): 27.3% due to aortic stenosis (n = 12), 31.8% due to aortic insufficiency (n = 14), 40.9% due to combined aortic stenosis and insufficiency (n = 18)
NYHA: NR
LVEF: NR
Case mix
Age (mean ± SD): total: NR; intervention: 45.5 ± 11.7;comparator: 45.5 ± 12.2
Percentage male: total: male 36, female 8; intervention: male 18, female 3; comparator: male 18, female 3
 
Ethnicity: NR
Interventions Intervention (exercise‐based CR)
Description: exercise was divided into 2 phases: centre‐based training (consisting of several types of exercise + 30‐minute cooling down period at the end), and home‐based training (consisting of a few simple daily exercises)
Time of start after event: 2 months after surgery
Components: exercise
Detail of exercise: started with 15‐minute bicycle warm‐up session, then short programme of 30 minutes (with 20 different arm and leg exercises of 1 to 2 minutes each). Calisthenics of alternative heavy (e.g. jogging, jumping) or light (e.g. rocking sit‐ups, arm flinging at slow speeds) exercises were then carried out for 1 hour, followed by playing volleyball for 30 minutes and a 1‐hour break. Selected exercises from the above were then repeated, before the session concluded with a 30‐minute cooling down period
Modality: bicycle ergometer + aerobics + calisthenics
Dose of exercise: (calculated as overall no. of weeks of training multiplied by mean number of sessions per week multiplied by mean duration of sessions in minutes): NR (centre) + NR (home)
Length of session: 3 to 4 hours
Frequency/no. of sessions: daily
Intensity: individualised to patient (upper pulse limit during training was adjusted to 85% to 90% of maximal heart rate obtained at initial exercise test)
Resistance training included: yes, isometric arm and leg exercises
Total duration: 4 weeks
Setting: home/hospital/Internet delivery or combination: hospital + home
Supervision: supervised/unsupervised/not reported: centre‐based supervised, home‐based not supervised
Intermittent nurse or exercise specialist support? NR
Co‐interventions: NR
Comparator
Description: patients were not encouraged to start any systematic training (no patients started this). Patients reported moderate daily physical activity at each control visit
Co‐interventions: NR
Outcomes Outcomes (scale measured in): return to work, exercise capacity (cumulated work, i.e. work performed + workload)
Other outcomes measured
Physical work capacity
Notes Follow‐up at 2, 6, and 12 months
  • In training group, 3 patients did not perform the exercise test at the end of the training period (i.e. at 3 months after surgery) for non‐medical reasons, and 1 patient did not attend the 12‐month control

  • In the control group, 2 patients were unable to participate 7.5 and 8 months following surgery due to a non‐fatal thromboembolic episode, and 1 patient did not come to the 12‐month review for non‐medical reasons

  • Only 15 male participants from the training group and 16 male participants from the control group were included in the exercise capacity assessments, as females could not reach the highest comparable workload (100W)

Characteristics of excluded studies [ordered by study ID]

Study Reason for exclusion
Amat Santos 2012 Patient population not appropriate. Conference paper
Bakhshayesh 2018 Population
Batra 2012 Not a randomised trial
Brosseau 1995 Patient population not appropriate
Cargnin 2019 Inappropriate intervention
Chambers 2005 Letter to the Editor; not a randomised trial
Chan 2012 Not a randomised trial (systematic review of effectiveness of qigong in cardiac rehabilitation)
CTRI 2017 Inappropriate Intervention
de Charmoy 2000 Intervention not appropriate (chest physiotherapy)
Deepa 2018 Not an RCT
Dull 1983 Patient population not appropriate
Editorial 2018 Editorial to paper that compares CR referral and outcomes in TAVR vs SAVR patients
Fang 2002 Inappropriate intervention (rehabilitation guidance at 24 hours after surgery and QoL measure) and unclear patient population (including both patients with rheumatic heart disease and patients after valve replacement)
Ferreira 2009 Intervention not appropriate (inspiratory breathing exercises)
Fontes Cerqueira 2018 Inappropriate intervention
Gaita 1999 Patient population not appropriate (randomisation method and study population unclear)
Ghalamghash 2008 Not a randomised trial
Gortner 1988 Intervention not appropriate (nursing intervention, no physical exercise)
Green 2013 Not a randomised trial
Grunewald 1971 Not a randomised trial
Ha 2011 Not a randomised trial. Not possible to obtain full paper
Hokanson 2011 Letter to the Editor; not a randomised trial
Hui 2006 Patient population not appropriate
Jairath 1995 Not a randomised trial (non‐randomised cluster trial)
Johnson 1996 Intervention not appropriate (physical intervention in control group)
Kardis 2007 Not a randomised trial (a randomised case control study)
Kassirskii 1983 Not a randomised trial (an observational study)
Kassirskii 1991 Not a randomised trial
Kodric 2013 Patient population not appropriate (patients after all kinds of major cardiac surgery)
Kübler 1984 Patient population not appropriate
Liao 2004 Intervention not eligible (no physical intervention, only psychological and behavioural interventions)
Lim 1998 Patient population not appropriate
Martsinkiavichus 1980 Not a randomised trial
McDermott 2019 Inappropriate intervention
Nagashio 2003 Patient population not appropriate
Nehyba 2009 Not a randomised trial (a non‐randomised cluster trial); patient population including patients with coronary artery bypass surgery
Newell 1980 Not a randomised trial (a non‐randomised cluster trial)
Patel 2019 Investigators were looking into the rate of CR enrolment in the studied population
Peng 2018 Inappropriate population
Petrunina 1980 Not a randomised trial
Prasciene 2019 Inappropriate intervention
Pressler 2015 Conference abstract for included study
RBR‐8swgc3 2017 Inappropriate intervention
Rizwan 2012 Not a randomised trial
Rogers2018 Conference abstract for included study
Roseler 1997 Not a randomised trial and inappropriate patient population
Rosenfeldt 2011 Patient population not appropriate (patients with valve surgery and coronary artery bypass graft surgery)
Royse 2015 Inappropriate intervention
Song 2019 Non‐RCT
Stoickov 2018 Outcomes
Sumide 2009 Not a randomised trial
Tang 2019 The only RCT of interest in this study is the one that has been updated
Therrien 2003 Patient population not appropriate (repaired tetralogy of Fallot)
Ueshima 2004 Not a randomised trial
Viana 2018 Not an RCT
Weber 2019 Inappropriate intervention
Widimsky 2009 Patient population not appropriate (patients with acute myocardial infarction)
Yan 2016 Not an RCT
Yau 2018 Inappropriate intervention

CR: cardiac rehabilitation.
QoL: quality of life.
RCT: randomised controlled trial.
SAVR: surgical aortic valve replacement.
TAVR: transcatheter aortic valve replacement.

Characteristics of ongoing studies [ordered by study ID]

ACTIVE AFTER TAVR 2017.

Study name A pragmatiC sTrategy to Promote actIVity and Enhance Quality of Life AFTER Transcatheter Aortic Valve Replacement (ACTIVE AFTER TAVR): a pilot study
Methods Parallel‐assignment RCT
Participants Subjects who have been treated commercially with TAVR with a SAPIEN 3 valve and are being discharged to home
Interventions Active comparator: no resistance exercise and no activity goal arm; blinded use of Fitbit with no daily activity goal and no resistance exercises
Experimental: resistance exercise and activity goal arm; unblinded use of Fitbit with daily activity goal (steps per day) and resistance exercises
Outcomes Primary outcome measures
  • Average daily steps [Time Frame: randomization to 6 weeks, average daily steps over the intervention period]

  • Short physical performance battery score [Time Frame: 6‐week value, adjusted for baseline value, combination of gait speed, balance test, and chair‐to‐stand test at end of intervention]

  • Quality of life as measured with KCCQ Overall Summary Score [Time Frame: 6‐week value, adjusted for baseline value, KCCQ overall summary score]


Secondary outcome measures
  • 5‐meter gait time at end of intervention period [Time Frame: randomisation to 6 weeks, 5‐meter gait time at end of intervention period, adjusted for baseline]

  • Chair sit‐to‐stand test [Time Frame: 6‐week value, adjusted for baseline value, time to complete 5 chair stands]

  • Balance test score at end of intervention period [Time Frame: randomisation to 6 weeks, balance test score at end of intervention period, adjusted for baseline]

  • 6‐minute walk [Time Frame: 6‐week value, adjusted for baseline value, 6‐minute walk distance at end of intervention period]

  • Handgrip [Time Frame: 6‐week value, adjusted for baseline value, handgrip strength'

  • Average number of hours per day with 250 or more steps [Time Frame: randomisation to 6 weeks, average number of hours per day with 250 or more steps over intervention period]

  • Average global physical health as assessed by PROMIS Global Health 10 Short Form [Time Frame: randomisation to 6 weeks, average global physical health as assessed by PROMIS Global Health 10 Short Form over intervention period]

  • Average global mental health as assessed by PROMIS Global Health 10 Short Form [Time Frame: randomisation to 6 weeks, average global mental health as assessed by PROMIS Global Health 10 Short Form over intervention period]

  • Physical function as assessed by NIH PROMIS computerised adaptive test [Time Frame: randomisation to 6 weeks, physical function as assessed by NIH PROMIS computerised adaptive test, adjusted for baseline]

  • Depression as assessed by NIH PROMIS computerised adaptive test [Time Frame: randomisation to 6 weeks, depression as assessed by NIH PROMIS computerised adaptive test, adjusted for baseline]

  • Fatigue as assessed by NIH PROMIS computerised adaptive test [Time Frame: randomisation to 6 weeks, fatigue as assessed by NIH PROMIS computerised adaptive test, adjusted for baseline]

  • Dyspnoea as assessed by NIH PROMIS computerised adaptive test [Time Frame: randomisation to 6 weeks, dyspnoea as assessed by NIH PROMIS computerised adaptive test, adjusted for baseline]

  • Daily active minutes (total) [Time Frame: randomisation to 6 weeks, average daily active minutes (total)

  • Daily active minutes of moderate to high intensity [Time Frame: randomisation to 6 weeks, average daily minutes of moderate to high intensity]

  • Sedentary minutes [Time Frame: randomisation to 6 weeks, average daily sedentary minutes]

  • Daily steps [Time Frame: 6 weeks post baseline to end of study, average daily steps]

  • Daily active minutes (total) [Time Frame: 6 weeks post baseline to end of study, average daily active minutes (total)]

  • Daily active minutes of moderate to high intensity [Time Frame: 6 weeks post baseline to end of study, average daily active minutes of moderate to high intensity]

  • Daily sedentary minutes [Time Frame: 6 weeks post baseline to end of study, average daily sedentary minutes]

  • KCCQ Overall Summary Score [Time Frame: 6 weeks post baseline to end of study, KCCQ overall summary score, adjusted for baseline]

  • Global physical health [Time Frame: 6 weeks post baseline to end of study, global physical health as assessed by PROMIS Global Health 10 Short Form, adjusted for baseline]

  • Global mental health [Time Frame: 6 weeks post baseline to end of study, global mental health as assessed by PROMIS Global Health 10 Short Form, adjusted for baseline]

Starting date 7 November 2017
Contact information Brian Lindman, Associate Professor, Vanderbilt University Medical Center
Notes Estimated enrolment: 85 participants. Estimated study completion date: August 2020
Location: Massachusetts General Hospital, Dartmouth‐Hitchcock Medical Center, Atlantic Health ‐ Morristown Medical Center, Vanderbilt University Medical Center, University of Utah

Exercise Training After TAVI.

Study name Exercise training after transcatheter aortic valve implantation
Methods Parallel‐assignment RCT
Participants Patients after transcatheter aortic valve replacement (TAVI)
Interventions Continuous exercise training 2 times per week for a period of 12 weeks
Patients will undergo moderate continuous exercise training at 75% of VO2 max
Outcomes Primary
  • Change in maximal oxygen uptake during exercise [Time Frame: 3 months, mL/kg/min]


Secondary
  • Change in flow‐mediated dilatation (FMD) of the brachial artery [Time Frame: 3 months, % flow‐mediated dilatation and arterial stiffness]

  • Change in arterial stiffness coefficient [Time Frame: 3 months, coefficient]

  • Change in value of blood N terminal‐proBNP [Time Frame: 3 months, ng/L]

  • Change in value of blood D‐dimer [Time Frame: 3 months, microg/L]

  • Change in value from questionnaire‐obtained quality of life [Time Frame: 3 months, points]

  • Change in ECG waves [Time Frame: 3 months, estimated with digital high‐resolution ECG]

  • Change in result of the 6‐minute walking test [Time Frame: 3 months, metres]

  • Change in heart rate variability [Time Frame: 3 months, estimated with digital high‐resolution ECG]


Other outcome measures
  • Change in heart rate recovery [Time Frame: 3 months, beats/min]

Starting date 18 June 2019
Contact information luka.vitez@gmail.comborut.jug@kclj.si
Notes Estimated enrolment: 40 participants. Estimated study completion date: December 2020
Location: UMC Ljubljana Slovenia

Feng 2019.

Study name The effects of stage I cardiac rehabilitation on cardiopulmonary function in patients undergoing open heart surgery: a randomized controlled study.
Methods Randomised parallel controlled trial
Participants Adults after open heart surgery
Interventions General exercise rehabilitation group: general exercise rehabilitation 
Intensive exercise rehabilitation group: intensive exercise rehabilitation
Outcomes Primary
  • PVO2


Secondary
  • Peak cardiac output

  • Resting cardiac output

  • Cardiac NYHA grading

  • Echocardiography

Starting date 1 July 2017
Contact information 29611290@qq.com; liubomiao424@sina.cn
Notes Estimated enrolment: general exercise rehabilitation group: 60; intensive exercise rehabilitation group: 60
Estimated study finish date: 31 March 2020
Location: Fuwai Hospital; Chinese Academy of Medical Sciences, Beijing, China

HBCR‐TAVR 2019.

Study name Impact of home‐based cardiac rehabilitation on outcomes after TAVR (HBCR‐TAVR)
Methods Parallel‐assignment RCT
Participants Chinese patients after transcatheter aortic valve replacement (TAVR)
Interventions Placebo comparator: control group: routine care
Experimental: intervention group: home‐based cardiac rehabilitation
Outcomes Primary outcome measures
  • 6‐minute walk test [Time Frame: 6 weeks, total distance walked in meters during 6 minutes]


Secondary outcome measures
  • Number of participants to die [Time Frame: 6 weeks, 12 months, number of participants who die during the study due to cardiovascular or non‐cardiovascular causes]

  • Number of participants re‐hospitalised [Time Frame: 6 weeks, 12 months, number of participants re‐hospitalised during the study]

  • Number of participants completing home‐based cardiac rehabilitation [Time Frame: 6 weeks, number of participants completing home‐based cardiac rehabilitation]

  • Cardiac function [Time Frame: 12 months, ejection fraction estimated by echocardiography]

  • Aortic valve function [Time Frame: 12 months, aortic valve function estimated by echocardiography]

  • Number of participants injured [Time Frame: 6 weeks, number of participants injured or dying during the course of home‐based cardiac rehabilitation]

  • Time spent performing activities [Time Frame: 6 weeks, 12 months, number of minutes in a typical week that participants spent performing activities]

  • 6‐minute walk test [Time Frame: 12 months, total distance walked in meters during 6 minutes]

Starting date 9 May 2020
Contact information Xiaoya Wang, 15715702712
wxyonce@zju.edu.cn
Notes Estimated enrolment: 300 participants. Estimated study completion date: 31 December 2023
Locations: Second Affiliated Hospital of Zhejiang University, School of Medicine, Hangzhou, Zhejiang, China, 310000

Post Cardiac Valvular Surgery Rehabilitation (PORT.

Study name Post Cardiac Valvular Surgery Rehabilitation (PORT)
Methods Parallel‐assignment RCT
Participants Chinese patients after heart valve surgery
Interventions No intervention: conventional treatment group: this arm will receive usual care
Cardiac rehabilitation: cardiac rehabilitation consists of exercise rehabilitation, psychological counselling, and dietary guidance
Rehabilitation starts preoperatively with education and exercise management. After screening with cardiopulmonary exercise test, the participant will receive daily preoperative exercise rehabilitation till surgery. This lasts for 20 minutes per day, starting with a 40% to 60% anaerobic threshold and gradually advancing to 80%. Each patient was motivated to adhere to the basic protocol, but individual adjustments were allowed in case of slower progress. Physical exercise starts 1 month postoperatively after the first cardiopulmonary exercise testing and comprises the following 3 elements: individual planning of physical exercise, a specially trained physiotherapist conduction, and integrating of detailed information concerning medical treatment and diet. The exercise diary and the heart rate monitor recordings are essential for monitoring during the whole intervention
Outcomes Primary outcomes
  • Composite endpoint of in‐hospital all‐cause death, pulmonary complications, and ratio of postoperative hospitalisation longer than 7 days [Time Frame: through hospitalisation (up to 2 months), composite of in‐hospital all‐cause death and pulmonary complications, such as pulmonary infection, postoperative hospitalisation days]

  • Postoperative duration of hospitalisation [Time Frame: through hospitalisation (up to 2 months), length of hospital stay]


Secondary outcomes
  • Incidence of all‐cause death in 3 months [Time Frame: 3 months, incidence of all‐cause death at 3‐month follow‐up]

  • Incidence of pulmonary complications in 3 months [Time Frame: 3 months, incidence of pulmonary complications, such as pulmonary infection at 3‐month follow‐up]

  • Individualised Short Form‐36 (SF‐36) living quality scores in 3 months [Time Frame: 3 months, scores from self‐administered SF‐36 living quality questionnaire are measured. Higher mean scores reflect better outcomes]

  • VO2 peak in 3 months [Time Frame: 3 months, peak oxygen consumption at cardiopulmonary exercise test is measured through a metabolic cart during a graded exercise test on a treadmill at 3 months' follow‐up]

  • Length of ICU treatment [Time Frame: through hospitalisation (up to 2 months), total length of treatment at intensive care unit]

  • Total length of in‐hospital stays [Time Frame: through hospitalisation (up to 2 months), total length of in‐hospital stays]

  • Length of bed rest [Time Frame: through hospitalisation (up to 2 months), length of bed rest] Description: postoperative duration of bed rest until off‐bed activity supervised by rehabilitation therapists

  • Total postoperative cost of medical expenses [Time Frame: through hospitalisation (up to 2 months), total postoperative cost of medical expenses]

  • Incidence of treatment‐emergent adverse events [Emerging Arrhythmia or/and Muscle Injury or/and Acute Heart Failure] [Time Frame: through hospitalisation (up to 2 months), evaluation of treatment‐emergent adverse events during hospitalisation: Emerging Arrhythmia or/and Muscle Injury or/and Acute Heart Failure]

Starting date 1 January 2018
Contact information Jiyan Chen, MD; 02083827812; chenjiyandr@126.com
Notes Estimated enrolment: 800 participants. Estimated study completion date: 30 December 2021
Locations: Guangdong General Hospital, China

PREPARE TAVR Pilot Study.

Study name Physiological reconditioning program administered remotely in patients undergoing transcatheter aortic valve replacement pilot study
Methods Parallel‐assignment RCT
Participants Frail adults undergoing transcatheter aortic valve replacement (TAVR) procedures
Interventions Patients assigned to intervention arm will be provided a personalised, tailored, and graduated exercise programme to improve physical strength and conditioning
Outcomes Primary
  • Quality of life (QoL) [Time Frame: 1 year]


Quality of life as assessed by the Kansas City Cardiomyopathy Questionnaire (KCCQ). KCCQE is a 23‐item self‐administered questionnaire developed to independently measure patients' perceptions of their health status, which includes heart failure symptoms, impact on physical and social function, and how their heart failure impacts their quality of life (QoL) within a 2‐week recall period. KCCQ responses are provided along a rating scale continuum with equal spacing from worst to best
Secondary
  • LOS [Time Frame: index hospitalisation, length of stay post TAVR]

  • MACE [Time Frame: 1 year, composite of mortality and repeat hospitalisation]

Starting date 1 February 2019
Contact information Syed Ishba; syedi@smh.ca
Notes Estimated enrolment: 160 participants. Estimated study finish date: 31 March 2021
Location: St. Michael's Hospital, Toronto

REHAB‐TAVR 2017.

Study name Home‐based exercise program for recovery after transcatheter aortic valve replacement: a pilot study
Methods Parallel‐assignment RCT
Participants Older adults after TAVR
Interventions Experimental: exercise and cognitive‐behavioural intervention. A physical therapist will make home visits, beginning within 1 week of discharge, to deliver an individualised exercise programme and cognitive‐behavioural interventions
Experimental: exercise alone. A physical therapist will make home visits, beginning within 1 week of discharge, to deliver an individualised exercise programme, without cognitive‐behavioural interventions
Active comparator: attention control education programme. Participants will receive telephone‐based education sessions from a study health professional
Outcomes Primary outcome measure
  • Change in Late‐Life Function and Disability Instrument (LLFDI) score [Time Frame: at baseline and at Week 8, LLFDI is a validated patient‐reported outcome questionnaire that measures both functional limitations (inability to perform physical tasks) and disability (inability to perform major life tasks and social roles) (range 0 to 100)]


Secondary outcome measures
  • Change in Short Physical Performance Battery (SPPB) summary score [Time Frame: at baseline and at Week 8, summary score is calculated based on chair stands, walking speed, and standing balance (range 0 to 12)]

  • Change in 2‐minute walk distance (meters) [Time Frame: at baseline and at Week 8, 2‐minute walk distance measures endurance]

  • Change in dominant handgrip strength (kg) [Time Frame: at baseline and at Week 8, dominant handgrip strength measures upper extremity strength]

  • Number of participants who experienced adverse events [Time Frame: at Week 8]


Other outcome measures
  • Change in Mini‐Mental State Examination (MMSE) score [Time Frame: at baseline and at Week 8, MMSE is an instrument that assesses general cognitive function]

  • Change in New York Heart Association (NYHA) functional class [Time Frame: at baseline and at Week 8, NYHA assesses the extent of physical activity limitation due to heart failure]

  • Change in Self‐Efficacy Scale for Exercise (SEE) [Time Frame: at baseline and at Week 8, SEE Scale measures self‐efficacy about exercise (range 0 to 90)]

  • Change in Outcome Expectation Scale for Exercise (OEE) [Time Frame: at baseline and at Week 8, OEE Scale measures outcome expectation about exercise (range 1 to 5)]

  • Adherence to home‐based exercise programme [Time Frame: at Week 8, proportion of days with completed daily task during entire study period will be measured]

Starting date 1 August 2017
Contact information Dae Hyun Kim, Associate Physician, Brigham and Women's Hospital
Notes Estimated enrolment: 60 participants. Estimated study completion date: 31 May 2020
Location: United States, Massachusetts

The PACO Trial.

Study name Personalized intervention to increase physical activity and reduce sedentary behaviour in rehabilitation after cardiac operations (the PACO trial)
Methods Parallel‐assignment RCT
Participants Coronary artery disease, aortic valve stenosis, and mitral valve insufficiency patients preparing for elective coronary artery bypass grafting (CABG), aortic valve replacement (AVR), or mitral valve repair (MVR)
Interventions The group of aortic valve stenosis patients receiving the PACO intervention for AVR/MVR patients besides the standard postoperative rehabilitation of Kuopio and Turku University Hospitals after aortic valve replacement. The PACO intervention includes activity guidance (i.e. goals to improve daily steps and physical activity levels, while reducing prolonged sitting) provided to patients with the novel combination of ExSed application, MoveSense accelerometer, and cloud system. In addition, exercise guidance (short video files) and regular mobile phone contacts from physiotherapist will be included in the intervention
Outcomes Primary outcome
  • Improvement in mean daily number of steps [Time Frame: improvement between baseline (during last preoperative month) and first 3 (and 12) months after discharge]


Improvement in mean daily number of steps after 3 months from discharge. In addition, follow‐up will be continued until 12 months after discharge. Baseline values of mean daily number of steps will be determined in a 7‐day accelerometer measurement conducted for patients before elective cardiac operation. Mean daily number of steps after the first 3 and 12 months of postoperative rehabilitation at home will be also determined in 7‐day (24‐hour) accelerometer measurements. Raw accelerometer data will be analysed with mean amplitude deviation and angle for posture estimation algorithms to recognise daily steps for the 7 days for which average will be calculated for each study patient
Secondary outcomes
  • Change in mean daily accumulated total time of light PA and MVPA [Time Frame: change between baseline (during last preoperative month) and first 3 months after discharge, postoperative change in patient's mean daily accumulated total time of light and moderate to vigorous physical activity]

  • Change in mean daily total time of sedentary behaviour (SB) [Time Frame: change between baseline (during last preoperative month) and first 3 months after discharge, postoperative change in patient's mean daily total time of SB]

  • Change in maximal oxygen consumption [Time Frame: change between first and third months after discharge, evolvement of patient's maximal oxygen consumption (VO2 peak) will be determined in 6‐minute walking test, conducted for patients twice (after 1 and 3 months) postoperatively. Only some of the randomised patients coming from city areas of Kuopio and Turku will be included for measurements of maximal oxygen consumption

  • Improvement in self‐perceived quality of life (QoL) assessed with SAQ‐7 questionnaire [Time Frame: improvement between baseline (during last preoperative month) and first 3 months after discharge, improvement in patient's postoperative quality of life after 3 months of rehabilitation; quality of life will be determined with Seattle Angina Questionnaire 7 (SAQ‐7)]

  • Improvement in self‐perceived quality of life (QoL) assessed with SF‐36 questionnaire [Time Frame: change between baseline (during last preoperative month) and first 3 months after discharge, improvement in patient's postoperative quality of life after 3 months of rehabilitation; quality of life will be determined with SF‐36 questionnaire]

  • Improvement in self‐perceived quality of life (QoL) assessed with 15 D questionnaire [Time Frame: improvement between baseline (during last preoperative month) and first 3 months after discharge, improvement in patient's postoperative quality of life after 3 months of rehabilitation; quality of life will be determined with 15 D questionnaire]

  • Improvement in self‐perceived quality of life (QoL) assessed with PHQ‐2 questionnaire [Time Frame: improvement between baseline (during last preoperative month) and first 3 months after discharge, improvement in patient's postoperative quality of life after 3 months of rehabilitation; quality of life will be determined with PHQ‐2 questionnaire]

  • Improvement in self‐perceived quality of life (QoL) assessed with Rose Dyspnoea Index [Time Frame: improvement between baseline (during last preoperative month) and first 3 months after discharge, improvement in patient's postoperative quality of life after 3 months of rehabilitation; quality of life will be determined with Rose Dyspnoea Index]

  • Incidence of major cardiovascular events [Time Frame: first 12 postoperative months, major cardiovascular events include all‐cause mortality, any re‐hospitalisations due to CVD, repeat coronary re‐vascularisation, non‐operational myocardial infarction, and stroke. The incidence of major cardiovascular events will be monitored from patient records at the hospitals and from HILMO database during the first 12 postoperative months. In addition, patients will be asked about cardiovascular events during research telephone contact (after 12 months of rehabilitation)]

  • Change in accelerometer‐derived portion of deep sleep [Time Frame: change between baseline (during last preoperative month) and first 3 months after discharge, change in patient's deep sleep portion after cardiac operations. Deep sleep will be recognised with accelerometer attached to patient's wrist during sleep. Accelerometer will be used during 7 days]

  • Change in heart rate variability [Time Frame: change between baseline (during last preoperative month) and first 3 months after discharge, change in heart rate variability]

Starting date 6 April 2018
Contact information villevas@uef.fi ; jari.halonen@kuh.fi
Notes Specific operation groups (CABG, AVR, and MVR) will be analysed separately
Estimated enrolment: 540 participants. Estimated study completion date: 1 March 2028
Location: Kuopio University Hospital, Kuopio, Finland, 70029

Valve‐ex 2009.

Study name Physical activity in patients after aortic valve replacement (Valve‐ex) [influence of regular physical activity on exercise capacity, cardiac remodeling and endothelial function in patients after aortic valve replacement]
Methods Parallel‐assignment RCT
Participants Patients after aortic valve replacement due to severe stenosis
Interventions Active comparator: training group: physical activity
B controls: no intervention
Outcomes Maximum oxygen uptake
Starting date 12 March 2009
Contact information Technische Universität München
Notes Estimated enrolment: 30 participants. Estimated study completion date: not reported
Location: Department of Prevention and Sports Medicine, Technische Universität München, München, Bavaria, Germany, 80802

Wang 2019.

Study name A study of the impact of home‐based cardiac rehabilitation on outcomes after transcatheter aortic valve replacement (TAVR)
Methods Parallel‐assignment RCT
Participants Adult Chinese patients after transcatheter aortic valve replacement (TAVR)
Interventions Placebo comparator; control group: routine care; experimental group/interventional group: home‐based cardiac rehabilitation
Outcomes Primary outcomes
  • 6‐minute walk test [Time Frame: 6 weeks, total distance walked in meters during 6 minutes]


Secondary outcomes
  • Number of participants who will die [Time Frame: 6 weeks, 12 months, number of participants who die during the study due to cardiovascular or non‐cardiovascular causes]

  • Number of participants re‐hospitalised [Time Frame: 6 weeks, 12 months, number of participants re‐hospitalised during the study]

  • Number of participants completing home‐based cardiac rehabilitation [Time Frame: 6 weeks, number of participants completing home‐based cardiac rehabilitation]

  • Cardiac function [Time Frame: 12 months, ejection fraction estimated by echocardiography]

  • Aortic valve function [Time Frame: 12 months, aortic valve function estimated by echocardiography]

  • Number of participants injured [Time Frame: 6 weeks, number of participants injured or who die during the course of home‐based cardiac rehabilitation]

  • Time spent performing activities [Time Frame: 6 weeks, 12 months, number of minutes in a typical week that participants spent performing activities]

  • 6‐minute walk test [Time Frame: 12 months, total distance walked in meters during 6 minutes]

Starting date 1 January 2020
Contact information wxyonce@zju.edu.cn
Notes Estimated enrolment: 300 participants. Estimated study finish date: 31 December 2023
Location: Second Affiliated Hospital, School of Medicine, Zhejiang University

AVR: aortic valve replacement.
CABG: coronary artery bypass graft.
ECG: electrocardiogram.
KCCQ: Kansas City Cardiomyopathy Questionnaire.
LOS: length of stay.
MACE: major adverse cardiovascular event.
MVR: mitral valve replacement.
NYHA: New York Heart Association.
PVO2: mixed venous oxygen tension.
RCT: randomised controlled trial.
TAVR: transaortic valve replacement.
VO2: maximal oxygen consumption.

Differences between protocol and review

This updated review included the RoB2 assessment, which was not included in the last review, nor in the protocol. The protocol has now been updated to account for RoB2 and MECIR guidance. The RoB2 assessment for all primary outcomes and secondary outcomes of exercise capacity has been included. 

Given their importance to policymakers, this update added the following secondary outcomes to the review: (1) return to work, (2) costs, and (3) cost‐effectiveness.

We deleted the outcomes of NYHA classification and LVEF as we considered them to be population characteristics rather than outcomes of interventions, and we therefore did not believe it was important to include them.

Contributions of authors

KLS and ADZ initiated and raised funding for the initial review. KLS drafted the initial review. LA, KLS, and RST contributed to updating of the text. LA and KLS carried out trial selection, data extraction, and RoB2 analysis, with RST confirming all data extractions and resolving any disagreements. LA carried out meta‐analysis with supervision from RST. All review authors have revised and contributed to drafting of the review, and all have approved the final version of the review for publication.

Sources of support

Internal sources

  • No sources of support supplied

External sources

  • NIHR, UK

    This project was supported by the NIHR via Cochrane Infrastructure funding to the Heart Group. The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the Systematic Reviews Programme, NIHR, NHS, or the Department of Health and Social Care

Declarations of interest

Lizette Abraham declares no conflict of interest.

Kirstine L Sibilitz, Selina K Berg, Lars H Tang, Signe S Risom, Britt Borregaard, Jane Lindschou, Rod Taylor, and Ann‐Dorthe Zwisler are involved in conducting previous and/or current randomised clinical trials (including the included trial of Sibilitz et al) investigating the effects of cardiac rehabilitation for different cardiac populations. None of these trials were or are industry sponsored, but studies were sponsored by private and public funding. None of the founders had any involvement in analyses, collection of data, or interpretation of trial results.

Kirstine L Sibilitz and Ann‐Dorthe Zwisler are currently co‐authoring other Cochrane Reviews of cardiac rehabilitation.

Rod S Taylor is an author on previous Cochrane Reviews on cardiac rehabilitation and is the Chief Investigator for ongoing trials (REACH‐HFpEF, SCOT:REACH‐HF, DK:REACH‐HF) assessing the clinical effectiveness and cost‐effectiveness of home‐based self‐directed exercise‐based cardiac rehabilitation interventions for patients with heart failure and their carers.

Ann‐Dorthe Zwisler declares financial support for expert testimony as part of her employment as professor.

New search for studies and content updated (no change to conclusions)

References

References to studies included in this review

Lin 2004 {published data only}

  1. Lin CY, He Z, Chen J, Yang B, Gu JX. Efficacy analysis of rehabilitation therapy on patients with heart valve replacement. Chinese Journal of Clinical Rehabilitation 2004;8:426-7. [Google Scholar]

Nilsson 2019 {published data only}

  1. Nilsson H, Nylander E, Borg S, Tamas E, Hedman K. Cardiopulmonary exercise testing for evaluation of a randomized exercise training intervention following aortic valve replacement. Clinical Physiology and Functional Imaging 2019;39(1):103–10. [DOI] [PMC free article] [PubMed] [Google Scholar]

Pressler 2016 {published data only}

  1. NCT01935297. Exercise training after transcatheter aortic valve implantation (SPORT:TAVI). clinicaltrials.gov/ct2/show/NCT01935297 (first posted 5 September 2013).
  2. Pressler A, Forschner L, Hummel J, Haller B, Christle JW, Halle M. Long-term effect of exercise training in patients after transcatheter aortic valve implantation: follow-up of the SPORT:TAVI randomised pilot study. European Journal of Preventative Cardiology 2018;25(8):794-801. [DOI] [PubMed] [Google Scholar]
  3. Pressler A, Christle JW, Lechner B, Grabs V, Haller B, Hettich I, et al. Exercise training improves exercise capacity and quality of life after transcatheter aortic valve implantation: a randomized pilot trial. American Heart Journal 2016;182:44-53. [DOI] [PubMed] [Google Scholar]

Rogers 2018 {published data only}

  1. Rogers P, Al-Aidrous S, Banya W, Haley SR, Mittal T, Kabir T, et al. Cardiac rehabilitation to improve health-related quality of life following trans-catheter aortic valve implantation: a randomised controlled feasibility study RECOVER–TAVI Pilot, ORCA 4, for the Optimal Restoration of Cardiac Activity Group. Pilot and Feasibility Studies 2018;13(4):185. [DOI] [PMC free article] [PubMed] [Google Scholar]

Sibilitz 2016 {published data only}

  1. Hansen TB, Zwisler AD, Berg SK, Sibilitz KL, Thygesen LC, Kjellberg J, et al. Cost-utility analysis of cardiac rehabilitation after conventional heart valve surgery versus usual care. European Journal of Preventive Cardiology 2017;24:698-707. [DOI: 10.1177/2047487317689908] [DOI] [PubMed] [Google Scholar]
  2. NCT01558765. CopenHeartVR - integrated rehabilitation of patients after heart valve surgery (CopenHeartVR). clinicaltrials.gov/ct2/show/NCT01558765 (first posted 20 March 2012).
  3. Sibilitz KL, Berg SK, Rasmussen TB, Risom SS, Thygesen LC, Tang L, et al. Cardiac rehabilitation increases physical capacity but not mental health after heart valve surgery: a randomised clinical trial. Heart 2016;102:1995–2003. [DOI] [PubMed] [Google Scholar]

Sire 1987 {published data only}

  1. Sire S. Physical training and occupational rehabilitation after aortic valve replacement. European Heart Journal 1987;8:1215-20. [DOI] [PubMed] [Google Scholar]

References to studies excluded from this review

Amat Santos 2012 {unpublished data only}

  1. Amat Santos IJ, Duro IC, Hernandez Luis C, Alvarado M, De La Fuente L, Amado MG, et al. Benefits of cardiac rehabilitation programmes in high-risk patients with symptomatic chronic heart failure and low left ventricular ejection fraction. In: European Heart Journal. Vol. 33. 2012:1056.

Bakhshayesh 2018 {published data only}

  1. IRCT20180730040643N1. The effect of IT interventions in cardiac rehabilitation patients [Home based cardiac rehabilitation after coronary angioplasty and heart surgery using booklet and mobile application: assessing the impact of interventions on health related quality of life]. en.irct.ir/trial/33791 (first received 3 November 2018).

Batra 2012 {published and unpublished data}

  1. Batra AS, McElhinney DB, Wang W, Zakheim R, Garofano RP, Daniels C, et al. Cardiopulmonary exercise function among patients undergoing transcatheter pulmonary valve implantation in the US Melody valve investigational trial. American Heart Journal 2012;163:280-7. [DOI] [PubMed] [Google Scholar]

Brosseau 1995 {published and unpublished data}

  1. Brosseau R, Juneau M, Sirard A, Savard A, Marchand C, Boudreau MH, et al. Safety and feasibility of a self monitored, home-based phase II exercise program for high risk patients after cardiac surgery [La securite d'un entrainement a domicile en phase II apres une chirurgie cardiaque pour les patients a haut risque]. Canadian Journal of Cardiology 1995;11:675-85. [PubMed] [Google Scholar]

Cargnin 2019 {published data only}

  1. Cargnin C, Karsten M,  Guaragna JCVDC, Dal Lago P. Inspiratory muscle training after heart valve replacement surgery improves inspiratory muscle strength, lung function, and functional capacity: a randomized controlled trial. Journal of Cardiopulmonary Rehabilitation and Prevention 2019;39(5):E1-E7. [DOI] [PubMed] [Google Scholar]

Chambers 2005 {published and unpublished data}

  1. Chambers J, Rimington H, Rajani R, Hodson F, Blauth C. Reply [2]. Journal of Heart Valve Disease 2005. [PubMed]

Chan 2012 {published and unpublished data}

  1. Chan CL, Wang CW, Ho RT, Ho AH, Ziea ET, Taam Wong VC, et al. A systematic review of the effectiveness of qigong exercise in cardiac rehabilitation. American Journal of Chinese Medicine 2012;40:255-67. [DOI] [PubMed] [Google Scholar]

CTRI 2017 {published data only}

  1. CTRI/2017/09/009707. Effect of physiotherapy in open heart surgery patients [Maximum inspiratory pressure, functional status and quality of life of the patients after open heart surgery following physiotherapy]. www.who.int/trialsearch/Trial2.aspx?TrialID=CTRI/2017/09/009707 (first received 2017).

de Charmoy 2000 {published and unpublished data}

  1. Charmoy SB, Eales CJ. The role of prophylactic chest physiotherapy after cardiac valvular surgery: is there one? South African Journal of Physiotherapy 2000;56:24-8. [Google Scholar]

Deepa 2018 {published data only}

  1. CTRI/2018/02/011705. To compare the utility and efficacy of the routine exercise program and bicycle ergometer program among heart valve surgery patients [Role of bicycle ergometer training in phase one among heart valve surgery patients]. www.who.int/trialsearch/Trial2.aspx?TrialID=CTRI/2018/02/011705 (first received 2018).

Dull 1983 {published and unpublished data}

  1. Dull JL, William JL. Are maximal inspiratory breathing exercises or incentive spirometry better than early mobilization after cardiopulmonary bypass? Physical Therapy 1983;63:655-9. [DOI] [PubMed] [Google Scholar]

Editorial 2018 {published data only}

  1. Francesco G, Carlo V. Editorial to 'Comparison of phase 2 cardiac rehabilitation outcomes between patients after transcatheter versus surgical aortic valve replacement'. European Journal of Preventive Cardiology 2018;25(15):1575‐6. [DOI] [PubMed] [Google Scholar]

Fang 2002 {published and unpublished data}

  1. Fang W, Chen YT, Liang JJ. Effects of early rehabilitative guidance on quality of life after cardiac valve replacement. Chinese Journal of Clinical Rehabilitation 2002;6:1106-7. [Google Scholar]

Ferreira 2009 {published and unpublished data}

  1. Ferreira PE, Rodrigues AJ,  Evora PR. Effects of an inspiratory muscle rehabilitation program in the postoperative period of cardiac surgery. Arquivos Brasileiros de Cardiologia 2009;92:275-82. [DOI] [PubMed] [Google Scholar]

Fontes Cerqueira 2018 {published data only}

  1. Fontes Cerqueira TC, De Cerqueira Neto ML,  De Assis Pereira Cacau L,  Oliveira GU,  Da Silva WM,  Carvalho VO,  et al. Ambulation capacity and functional outcome in patients undergoing neuromuscular electrical stimulation after cardiac valve surgery: a randomised clinical trial. Medicine (United States) 2018;97(46):e13012. [DOI] [PMC free article] [PubMed] [Google Scholar]

Gaita 1999 {published and unpublished data}

  1. Gaita D, Branea I, Dragulescu S, Gaita A, Socoteanu I, Gaspar M, et al. Benefit of Exercise Training in Patients With Valve Prostheses and Chronic Heart Failure. Bologna: Medimond S R L, 1999. [Google Scholar]

Ghalamghash 2008 {published and unpublished data}

  1. Ghalamghash R, Goosheh B, Emrani A,  Keyhani MR. Effect of cardiac rehabilitation program on functional capacity following valvular heart surgery. Journal of Medical Council of Islamic Republic of Iran 2008;26:292-3. [Google Scholar]

Gortner 1988 {published and unpublished data}

  1. Gortner SR, Gilliss CL, Shinn JA, Sparacino PA, Rankin S, Leavitt M, et al. Improving recovery following cardiac surgery: a randomized clinical trial. Journal of Advanced Nursing 1988;13:649-61. [DOI] [PubMed] [Google Scholar]

Green 2013 {published and unpublished data}

  1. Green P, Kirtane A, Genereux P, McAndrew T, Hueter I, Alu M, et al. The impact of six-minute walk test performance on outcomes after transcatheter aortic valve replacement: insights from the PARTNER trial. Journal of the American College of Cardiology 2013;62:E1971. [DOI] [PMC free article] [PubMed] [Google Scholar]

Grunewald 1971 {published and unpublished data}

  1. Grunewald B. Prevention and rehabilitation of rheumatic heart valve diseases. Zeitschrift fur Allgemeinmedizin 1971;47:705-9. [PubMed] [Google Scholar]

Ha 2011 {published and unpublished data}

  1. Ha YK, Jung YY. Development and application of an early exercise program for open heart surgery patients. Journal of Korean Critical Care Nursing 2011;4:1. [Google Scholar]

Hokanson 2011 {published and unpublished data}

  1. Hokanson JS,  Ralphe JC. Exercise restrictions after balloon aortic valvuloplasty (Letter to the Editor). Journal of the American College of Cardiology (JACC) 2011;57:2034. [DOI] [PubMed] [Google Scholar]

Hui 2006 {published and unpublished data}

  1. Hui PN, Wan M, Chan WK,  Yung PM. An evaluation of two behavioral rehabilitation programs, qigong versus progressive relaxation, in improving the quality of life in cardiac patients. Journal of Alternative & Complementary Medicine 2006;12:373-8. [DOI] [PubMed] [Google Scholar]

Jairath 1995 {published and unpublished data}

  1. Jairath N, Salerno T, Chapman J, Dornan J,  Weisel R. The effect of moderate exercise training on oxygen uptake post-aortic/mitral valve surgery. Journal of Cardiopulmonary Rehabilitation 1995;15:424-30. [DOI] [PubMed] [Google Scholar]

Johnson 1996 {published and unpublished data}

  1. Johnson D, Kelm C, Thomson D, Burbridge B, Mayers I. The effect of physical therapy on respiratory complications following cardiac valve surgery. Chest 1996;109:638-44. [DOI] [PubMed] [Google Scholar]

Kardis 2007 {published and unpublished data}

  1. Kardis P, Sherman M, Barnett SD. Association of age and quality of life following phase II cardiac rehabilitation. Journal of Nursing Care Quality 2007;22:255-9. [DOI] [PubMed] [Google Scholar]

Kassirskii 1983 {published and unpublished data}

  1. Kassirskii GI, Semenovskii ML, Petrunina LV,  Fomina NG. Evaluation of patient rehabilitation effectiveness after aortic valve prosthesis based on spiroergometric study data. Kardiologiia 1983;23:82-6. [PubMed] [Google Scholar]

Kassirskii 1991 {published and unpublished data}

  1. Kassirskii GI, Grosheva TV. Physical rehabilitation using bicycle ergometry training of patients after aortic valve replacement. Terapevticheskii Arkhiv 1991;63:59-62. [PubMed] [Google Scholar]

Kodric 2013 {published and unpublished data}

  1. Kodric M, Trevisan R, Torregiani C, Cifaldi R, Longo C, Cantarutti F, et al. Inspiratory muscle training for diaphragm dysfunction after cardiac surgery. Journal of Thoracic & Cardiovascular Surgery 2013;145:819-23. [DOI] [PubMed] [Google Scholar]

Kübler 1984 {published data only (unpublished sought but not used)}

  1. Kübler W,  Opherk D. Angina pectoris with normal coronary arteries. Acta Medica Scandinavica. Supplementum 1984;694:55-7. [DOI] [PubMed] [Google Scholar]

Liao 2004 {published and unpublished data}

  1. Liao B, Liu H, Chen L, Guo SM, Liu AM, et al. Effect of psychological and behavioral interventions on the negative emotion of patients with heart valve replacements during perioperative period. Chinese Journal of Clinical Rehabilitation 2004;8:7360-1. [Google Scholar]

Lim 1998 {published and unpublished data}

  1. Lim HY, Lee CW, Park SW, Kim JJ, Song JK, Hong MK, et al. Effects of percutaneous balloon mitral valvuloplasty and exercise training on the kinetics of recovery oxygen consumption after exercise in patients with mitral stenosis. European Heart Journal 1998;19:1865-71. [DOI] [PubMed] [Google Scholar]

Martsinkiavichus 1980 {published and unpublished data}

  1. Martsinkiavichus A, Paliushchinskaia N, Iozaitis M, Zhukauskene I, Norvilene A, et al. Physical activity and the work capacity of patients with artificial heart valves before and after rehabilitation. Kardiologiia 1980;20:28-31. [PubMed] [Google Scholar]

McDermott 2019 {published data only}

  1. McDermott C, McCormack CM, McDermott L,  O'Shea O,  Kelly SM, McCarren A, et al. Comparison of selected health indices in Irish and Belgian participants commencing a home-based, technology enabled cardiac rehabilitation program. European Journal of Preventive Cardiology 2019;26:S40-S41. [Google Scholar]

Nagashio 2003 {published and unpublished data}

  1. Nagashio M, Suzuki Y, Onishi Y, Tomatsu A, Shimada Y, Okamura S, et al. Post-operative cardiac rehabilitation programs for early returning to daily life activities: self-measurement of heart rate and blood pressure during staying at home trial. Journal of Cardiology 2003;41:91-5. [PubMed] [Google Scholar]

Nehyba 2009 {published data only}

  1. Nehyba S, Chaloupka V, Soucek R, Chaloupkova S, Vysoky R, Stetka F, et al. The programme of managed ambulatory rehabilitation for patients after heart valve defect surgery. Vnitrni Lekarstvi 2009;55:1118-25. [PubMed] [Google Scholar]

Newell 1980 {published and unpublished data}

  1. Newell JP, Kappagoda CT, Stoker JB, Deverall PB, Watson DA, Linden RJ. Physical training after heart valve replacement. British Heart Journal 1980;44:638-49. [DOI] [PMC free article] [PubMed] [Google Scholar]

Patel 2019 {published data only}

  1. Patel D,  Duncan M,  Shah A,  Lindman B, Greevy R, Savage P, et al. Association of cardiac rehabilitation with decreased hospitalization and mortality risk after cardiac valve surgery. JAMA Cardiology 2019;4(12):1250-9. [DOI] [PMC free article] [PubMed] [Google Scholar]

Peng 2018 {published data only}

  1. Peng X, Su Y, Hu Z, Sun X, Li X,  Dolansky MA, et al. Home-based telehealth exercise training program in Chinese patients with heart failure: a randomized controlled trial. Medicine (United States) 2018;97(35):e12069. [DOI] [PMC free article] [PubMed] [Google Scholar]

Petrunina 1980 {published and unpublished data}

  1. Petrunina LV. Patient physical rehabilitation after mitral or aortic valve prosthesis. Kardiologiia 1980;20:50-3. [PubMed] [Google Scholar]

Prasciene 2019 {published data only}

  1. Prasciene E, Icking N, Kisieliute E,  Kubilius R, Bjarnason-Wehrens B, et al. Feasibility and short-term effectiveness of an additional resistance and balance training in cardiac rehabilitation for older patients after valve surgery: results of a pilot study. European Journal of Preventive Cardiology 2019;26:S39. [Google Scholar]

Pressler 2015 {published data only}

  1. Pressler A,  Lechner B,  Christle J,  Bleiziffer S,  Mehilli J,  Jochheim D,  et al. Safety, applicability and outcome of regular exercise training after transcatheter aortic valve implantation: a randomized pilot trial. European Journal of Preventive Cardiology 2015;22 Suppl 1:S78. [Google Scholar]

RBR‐8swgc3 2017 {published data only}

  1. RBR‐8swgc3. Inspiratory muscle training in patients after cardiac surgery [Inspiratory muscle training in patients undergoing cardiac surgery]. http://www.who.int/trialsearch/Trial2.aspx?TrialID=RBR-8swgc3 (first received 2019).

Rizwan 2012 {published and unpublished data}

  1. Rizwan A, Jalwan J, Mukherjee S. To compare the immediate effect of deep breathing exercise and incentive spirometry on ABG after mitral valve replacement surgery. Indian Journal of Physiotherapy & Occupational Therapy 2012;6:12-7. [Google Scholar]

Rogers2018 {published data only}

  1. Rogers P, Banya W, Kabir T, Panoulas V, Probert H, Prendergast C, et al. Does cardiac rehabilitation improve functional, independence, frailty and emotional outcomes following trans catheter aortic valve replacement? European Heart Journal 2018;39 Suppl 1:1121. [Google Scholar]

Roseler 1997 {published and unpublished data}

  1. Roseler S, Schwartz FW,  Karoff M. Evaluation of the effectiveness of an ambulatory after-care rehabilitation program. Gesundheitswesen 1997;59:236-41. [PubMed] [Google Scholar]

Rosenfeldt 2011 {published and unpublished data}

  1. Rosenfeldt F, Braun L, Spitzer O, Bradley S, Shepherd J, Bailey M, et al. Physical conditioning and mental stress reduction - a randomised trial in patients undergoing cardiac surgery. BMC Complementary & Alternative Medicine 2011;11:20. [DOI] [PMC free article] [PubMed] [Google Scholar]

Royse 2015 {published data only}

  1. ACTRN12615000968572. Sternal management accelerated recovery trial (S.M.A.R.T.): the efficacy of modified sternal precautions on improving physical function in patients following cardiac surgery via a midline sternal incision [A randomized controlled trial of the efficacy of modified sternal precautions versus standard care on improving physical function following cardiac surgery via a median sternotomy]. www.who.int/trialsearch/Trial2.aspx?TrialID=ACTRN12615000968572 (first received 2019).

Song 2019 {published data only}

  1. Song F,  Zhan H, Liang Y,  He X, Guo L, et al. Cardiac rehabilitation improved oxygen uptake measured by cardiopulmonary exercise test in patients after aortic valve surgery. Reviews in Cardiovascular Medicine 2019;20(1):47-52. [DOI] [PubMed] [Google Scholar]

Stoickov 2018 {published data only}

  1. Stoickov V,  Deljanin Ilic M,  Stoickov M,  Marinkovic D,  Simonovic D,  Golubovic M, et al. Impact of short-term exercise training on QT dispersion and double product in patients after surgical aortic valve replacement. European Heart Journal 2018;39:85-6. [Google Scholar]

Sumide 2009 {published and unpublished data}

  1. Sumide T, Shimada K, Ohmura H, Onishi T, Kawakami K, Masaki Y, et al. Relationship between exercise tolerance and muscle strength following cardiac rehabilitation: comparison of patients after cardiac surgery and patients with myocardial infarction. Journal of Cardiology 2009;54:273-81. [DOI] [PubMed] [Google Scholar]

Tang 2019 {published data only}

  1. Tang LH, Zwisler AD,  Doherty P,  Oldridge N,  Berg SK, Christensen J. Changes in physical performance and their association with health-related quality of life in a mixed nonischemic cardiac population that participates in rehabilitation. Journal of Cardiopulmonary Rehabilitation and Prevention 2020;40(2):102-7. [DOI] [PubMed] [Google Scholar]

Therrien 2003 {published and unpublished data}

  1. Therrien J, Fredriksen PM, Walker M, Granton J, Reid GJ, Webb G. A pilot study of exercise training in adult patients with repaired tetralogy of Fallot. Canadian Journal of Cardiology 2003;19:685-9. [PubMed] [Google Scholar]

Ueshima 2004 {published data only (unpublished sought but not used)}

  1. Ueshima K, Kamata J, Kobayashi N, Saito M, Sato S, Kawazoe K, et al. Effects of exercise training after open heart surgery on quality of life and exercise tolerance in patients with mitral regurgitation or aortic regurgitation. Japanese Heart Journal 2004;45(5):789-97. [DOI] [PubMed] [Google Scholar]

Viana 2018 {published data only}

  1. RBR‐8ysb3n. Evaluation of the physiotherapy with cycle ergometer in the funcional recovery after cardiac surgery [Cycle ergometer and conventional physiotherapy in patients submitted the cardiac surgery]. www.who.int/trialsearch/Trial2.aspx?TrialID=RBR-8ysb3n (first received 2019).

Weber 2019 {published data only}

  1. DRKS00017239. Use of pre- and intensified post-procedural physiotherapy in patients with symptomatic aortic stenosis undergoing transcatheter aortic valve replacement - the 4P TAVR study. www.who.int/trialsearch/Trial2.aspx?TrialID=DRKS00017239 (first received 2019). [DOI] [PMC free article] [PubMed]

Widimsky 2009 {published and unpublished data}

  1. Widimsky P, Suryapranata H. "PRAGUE" randomized clinical trials: pride of Czech cardiology. Cor et Vasa 2009;51(10):679-83. [Google Scholar]

Yan 2016 {published data only}

  1. Yan XY. Early rehabilitation mode in transcatheter domestic aortic valve implantation of perioperative implementation and effects on patient life ability. Journal of the Hong Kong College of Cardiology 2016;24:A25. [Google Scholar]

Yau 2018 {published data only}

  1. ChiCTR1800016098. PREQUEL (PREhabilitation for improving QUality of recovery after ELective cardiac surgery) study: a randomised controlled trial. www.who.int/trialsearch/Trial2.aspx?TrialID=ChiCTR1800016098 (first received 2019). [DOI] [PMC free article] [PubMed]

References to ongoing studies

ACTIVE AFTER TAVR 2017 {published data only}

  1. NCT03270124. A pragmatiC sTrategy to Promote actIVity and Enhance Quality of Life AFTER Transcatheter Aortic Valve Replacement (ACTIVE AFTER TAVR): a pilot study. https://clinicaltrials.gov/ct2/show/NCT03270124 (first received 1 September 2017).

Exercise Training After TAVI {unpublished data only}

  1. NCT03966417. Exercise training after transcatheter aortic valve implantation (FitTAVI). clinicaltrials.gov/show/NCT03966417 (first received 29 May 2019).

Feng 2019 {published data only}

  1. ChiCTR‐IPR‐17011445. The effects of I stage cardiac rehabilitation on cardiopulmonary function in patients undergoing open heart surgery: a randomized controlled study. www.who.int/trialsearch/Trial2.aspx?TrialID=ChiCTR-IPR-17011445 (first received 2019).

HBCR‐TAVR 2019 {published data only}

  1. NCT04166682. Impact of home-based cardiac rehabilitation on outcomes after TAVR (HBCR-TAVR). https://clinicaltrials.gov/ct2/show/NCT04166682 (first received 18 November 2019).

Post Cardiac Valvular Surgery Rehabilitation (PORT {published data only}

  1. NCT03709511. Post cardiac valvular surgery rehabilitation (PORT). https://clinicaltrials.gov/ct2/show/NCT03709511 (first received 17 October 2018).

PREPARE TAVR Pilot Study {unpublished data only}

  1. NCT01504737. Rehabilitation in the form of exercise training in aortic stenosis patients (RASP). http://clinicaltrials.gov/show/NCT01504737 (first received 5 January 2012).

REHAB‐TAVR 2017 {published data only}

  1. NCT02805309. Home-based exercise program for recovery after transcatheter aortic valve replacement: a pilot study. https://clinicaltrials.gov/ct2/show/NCT02805309 (first received 20 June 2016).

The PACO Trial {published data only}

  1. Vasankari V, Jari Husu PY, Henri Tokola K,  Jaana Sievanen H, Vesa Airaksinen J, Tommi Hartikainen J. Personalised eHealth intervention to increase physical activity and reduce sedentary behaviour in rehabilitation after cardiac operations: study protocol for the PACO randomised controlled trial. BMJ Open Sport & Exercise Medicine 2019;5(1):e000539. [DOI] [PMC free article] [PubMed] [Google Scholar]

Valve‐ex 2009 {published data only}

  1. NCT00860678. Physical activity in patients after aortic valve replacement (Valve-ex). clinicaltrials.gov/ct2/show/NCT00860678 (first received 12 March 2009).

Wang 2019 {published data only}

  1. NCT04166682. Impact of home-based cardiac rehabilitation on outcomes after TAVR (HBCR-TAVR) [A study of the impact of home-based cardiac rehabilitation on outcomes after transcatheter aortic valve replacement (TAVR)]. clinicaltrials.gov/ct2/show/NCT04166682 (first received 18 November 2019).

Additional references

Ambrosetti 2020

  1. Ambrosetti M, Abreu A, Corra U, Davos CH, Hansen D, Frederix I, et al. Secondary prevention through comprehensive cardiovascular rehabilitation: from knowledge to implementation. 2020 update. A position paper from the Secondary Prevention and Rehabilitation Section of the European Association of Preventive Cardiology. European Journal of Preventive Cardiology 2020;0(0):1-42 [Epub ahead of print]. [DOI: 10.1177/2047487320913379] [PMID: PMID: 32223332] [DOI] [PubMed] [Google Scholar]

Anayo 2019

  1. Anayo L, Rogers P, Long L, Dalby M, Taylor R. Exercise-based cardiac rehabilitation for patients following open surgical aortic valve replacement and transcatheter aortic valve implant: a systematic review and meta-analysis. Open Heart 2019;6(1):e000922. [DOI: 10.1136/openhrt-2018-000922] [PMID: PMCID: PMC6519423 PMID: 31168371] [DOI] [PMC free article] [PubMed] [Google Scholar]

Anderson 2014

  1. Anderson LJ, Taylor RS. Cardiac rehabilitation for people with heart disease: an overview of Cochrane systematic reviews. International Journal of Cardiology 2014;177(2):348-61. [DOI: ] [DOI] [PubMed] [Google Scholar]

Anderson 2016

  1. Anderson L, Thompson DR, Oldridge N, Zwisler AD, Rees K, Martin N, et al. Exercise‐based cardiac rehabilitation for coronary heart disease. Cochrane Database of Systematic Reviews 2016, Issue 1. Art. No: CD001800. [DOI: 10.1002/14651858.CD001800.pub3] [DOI] [PMC free article] [PubMed] [Google Scholar]

BACPR 2012

  1. The British Association for Cardiovascular Prevention and Rehabilitation. The BACPR Standards and Core Components for Cardiovascular Disease Prevention and Rehabilitation 2012 (2nd edition). The British Cardiovascular Society 2012.

Balady 2007

  1. Balady GJ, Williams MA, Ades PA, Bittner V, Comoss P, Foody JM, et al. Core components of cardiac rehabilitation/secondary prevention programs: 2007 update. A scientific statement from the American Heart Association Exercise, Cardiac Rehabilitation, and Prevention Committee, the Council on Clinical Cardiology; the Councils on Cardiovascular Nursing, Epidemiology and Prevention, and Nutrition, Physical Activity, and Metabolism; and the American Association of Cardiovascular and Pulmonary Rehabilitation. Circulation 2007;115:2675-82. [DOI] [PubMed] [Google Scholar]

Baumgartner 2017

  1. Baumgartner H, Falk V, Bax JJ, De Bonis M, Hamm C, Holm PJ, et al. 2017 ESC/EACTS guidelines for the management of valvular heart disease. European Heart Journal 2017;38(36):2739-91. [DOI] [PubMed] [Google Scholar]

Ben‐Dor 2010

  1. Ben-Dor I, Pichard AD, Gonzalez MA, Weissman G, Li Y, Goldstein SA, et al. Correlates and causes of death in patients with severe symptomatic aortic stenosis who are not eligible to participate in a clinical trial of transcatheter aortic valve implantation. Circulation 2010;122(11):37-42. [DOI: 10.1161/CIRCULATIONAHA.109.926873] [DOI] [PubMed] [Google Scholar]

Borregaard 2019

  1. Borregaard B, Dahl JS, Riber LPS, Ekholm O, Sibilitz KL, Weiss M, et al. Effect of early, individualised and intensified follow-up after open heart valve surgery on unplanned cardiac hospital readmissions and all-cause mortality. International Journal of Cardiology 2019;15(289):30-6 [Epub 2019 Apr 11]. [DOI: 10.1016/j.ijcard.2019.02.056] [DOI] [PubMed] [Google Scholar]

Boutron 2008

  1. Boutron I, Moher D, Altman DG, Schultz KF, Ravaud P, CONSORT Group. Extending the CONSORT statement to randomized trials of nonpharmacologic treatment: explanation and elaboration. Annals of Internal Medicine 2008;148(4):295-309. [DOI] [PubMed] [Google Scholar]

Butchart 2005

  1. Butchart EG, Gohlke-Bärwolf C, Antunes MJ, Tornos P, Caterina RD, Cormier B, et al. Recommendations for the management of patients after heart valve surgery. European Heart Journal 2005;26:2463–71. [DOI] [PubMed] [Google Scholar]

Clausen 1976

  1. Clausen JP, Trap-Jensen J. Heart rate and arterial blood pressure during exercise in patients with angina pectoris: effects of exercise training and of nitroglycerin. Circulation 1976;53:436-42. [DOI] [PubMed] [Google Scholar]

Corra 2010

  1. European Association of Cardiovascular Prevention and Rehabilitation Committee for Science Guidelines, EACPR, Corrà U, Piepoli MF, Carré F, Heuschmann P, Hoffmann U, et al. Secondary prevention through cardiac rehabilitation: physical activity counselling and exercise training: key components of the position paper from the Cardiac Rehabilitation Section of the European Association of Cardiovascular Prevention and Rehabilitation. European Heart Journal 2014;31(16):1967-74. [PMID: ] [DOI] [PubMed] [Google Scholar]

Frank 1973

  1. Frank S, Johnson A, Ross J. Natural history of valvular aortic stenosis. British Heart Journal 1973;35:41-6. [DOI] [PMC free article] [PubMed] [Google Scholar]

Fredericks 2012

  1. Fredericks S, Lapum J, Lo J. Anxiety, depression, and self-management: a systematic review. Clinical Nursing Research 2012;21(4):411-30. [DOI: 10.1177/1054773812436681] [DOI] [PubMed] [Google Scholar]

Furukawa 2006

  1. Furukawa TA, Barbuib C, Ciprianib A, Brambillac P, Watanabe N. Imputing missing standard deviations in meta-analyses can provide accurate results. Journal of Clinical Epidemiology 2006;59:7–10. [DOI: 10.1016/j.jclinepi.2005.06.006] [DOI] [PubMed] [Google Scholar]

Gianuzzi 2003

  1. Giannuzzi P, Mezzani A, Saner H, Björnstad H, Fioretti P, Mendes M, et al. Physical activity for primary and secondary prevention. Position paper of the Working Group on Cardiac Rehabilitation and Exercise Physiology of the European Society of Cardiology. European Journal of Cardiovascular Prevention & Rehabilitation 2003;10:319–27. [DOI: 10.1097/01.hjr.0000086303.28200.50] [DOI] [PubMed] [Google Scholar]

Gohlke‐Bärwolf 1992

  1. Gohlke-Bärwolf C, Gohlke H, Samek L, Peters K, Betz P, Eschenbruch E, et al. Exercise tolerance and working capacity after valve replacement. Journal of Heart Valve Disease 1992;1(2):189-95. [PMID: ] [PubMed] [Google Scholar]

GradePro Software [Computer program]

  1. GRADEpro GDT: GRADEpro Guideline Development Tool [Software]. GRADEpro, 2015 (developed by Evidence Prime, Inc.).

Hambrecht 2000

  1. Hambrecht R, Gielen S, Linke A, Fiehn E, Yu J, Walther C, et al. Effects of exercise training on left ventricular function and peripheral resistance in patients with chronic heart failure: a randomised trial. JAMA 2000;283(23):3095–101. [DOI] [PubMed] [Google Scholar]

Hansen 2009

  1. Hansen L, Winkel S, Kuhr J, Bader R, Bleese N, Riess F-C. Factors influencing survival and postoperative quality of life after mitral valve reconstruction. European Journal of Cardio-Thoracic Surgery 2010;37:635-44. [DOI: 10.1016/j.ejcts.2009.07.044] [DOI] [PubMed] [Google Scholar]

Hansen 2017

  1. Hansen TB, Zwisler AD, Berg SK, et al. Cost-utility analysis of cardiac rehabilitation after conventional heart valve surgery versus usual care. European Journal of Preventive Cardiology 2017;24:698–707. [DOI: 10.1177/2047487317689908] [DOI] [PubMed] [Google Scholar]

Higgins 2019a

  1. Higgins JP, Savović J, Page MJ, Elbers RG, Sterne JA. Chapter 8. Assessing risk of bias in a randomized trial. In: Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA (editors). Cochrane Handbook for Systematic Reviews of Interventions version 6.0 (updated July 2019). Cochrane 2019. Available from www.training.cochrane.org/handbook.

Higgins 2019b

  1. Higgins JP, Li T, Deeks JJ (editors). Chapter 6. Choosing effect measures and computing estimates of effect. In: Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA (editors). Cochrane Handbook for Systematic Reviews of Interventions version 6.0 (updated July 2019). Cochrane 2019. Available from www.training.cochrane.org/handbook.

Higgins 2019c

  1. Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA (editors). Cochrane Handbook for Systematic Reviews of Interventions version 6.0 (updated July 2019). Cochrane 2019. Version 6. Available from www.training.cochrane.org/handbook.

Iung 2003

  1. Iung B, Baron G, Butchart EG, Delahaye F, Gohlke-Bärwolf C, Levang OW, et al. A prospective survey of patients with valvular heart disease in Europe: the Euro Heart Survey on Valvular Heart Disease. European Heart Journal 2003;24:1231-43. [DOI: 10.1016/S0195-668X(03)00201-X] [DOI] [PubMed] [Google Scholar]

Karlsson 2010

  1. Karlsson AK, Mattson B, Johansson M, Lidell E. Well-being in patients and relatives after open-heart surgery from the perspective of health care professionals. Journal of Clinical Nursing 2010;19:840-6. [DOI: 10.1111/j.1365-2702.2009.03017.x] [DOI] [PubMed] [Google Scholar]

Kiel 2011

  1. Kiel MK. Cardiac rehabilitation after heart valve surgery. Journal of Injury, Function and Rehabilitation 2011;3(10):962-7. [DOI: 10.1016/j.pmrj.2011.06.007] [DOI] [PubMed] [Google Scholar]

Kolte 2019

  1. Kolte D, Vlahakes GJ, Palacios IF, Sakhuja R, Passeri JJ, Inglessis I, et al. Transcatheter versus surgical aortic valve replacement in low-risk patients. Journal of the American College of Cardiology 2019;74:1532-40. [DOI] [PubMed] [Google Scholar]

Landry 1984

  1. Landry F, Habel C, Desaulniers D, Dagenais GR, Moisan A, Cote L. Vigorous physical training after aortic valve replacement: analysis of 10 patients. American Journal of Cardiology 1984;53(4):562-6. [PMID: ] [DOI] [PubMed] [Google Scholar]

Lauck 2020

  1. Lauck SB, Arnold SV, Borregaard B, Sathananthan J, Humphries KH,  Baron SJ, et al,  3M TAVR Investigators. Very early changes in quality of life after transcatheter aortic valve replacement: results from the 3M TAVR trial. Cardiovascular Revascularization Medicine 2020;S1553-8389(20):30349-3. [DOI: 10.1016/j.carrev.2020.05.044] [DOI] [PubMed] [Google Scholar]

Lefebvre 2011

  1. Lefebvre C, Manheimer E, Glanville J. Chapter 6. Searching for studies. In: Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011). The Cochrane Collaboration, 2011. Available from www.cochrane-handbook.org.

Long 2019

  1. Long L, Mordi IR, Bridges C, Sagar VA, Davies EJ, Coats AJS, et al. Exercise‐based cardiac rehabilitation for adults with heart failure. Cochrane Database of Systematic Reviews 2019, Issue 1. Art. No: CD003331. [DOI: 10.1002/14651858.CD003331.pub5] [DOI] [PMC free article] [PubMed] [Google Scholar]

Mack 2019

  1. Mack MJ, Leon MB, Thourani VH, Makkar R, Kodali SK, Russo M, et al. Transcatheter aortic-valve replacement with a balloon-expandable valve in low-risk patients. New England Journal of Medicine 2019;380:1695-705. [DOI] [PubMed] [Google Scholar]

Nishimura 2014

  1. Nishimura RA, Otto CM, Bonow RO, Carabello BA, Erwin III JP, Guyton RA, et al. 2014 AHA/ACC Guideline for the Management of Patients With Valvular Heart Disease: Executive Summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation 2014;129:2440-92. [DOI: 10.1161/CIR.0000000000000029] [DOI] [PubMed] [Google Scholar]

Nkomo 2006

  1. Nkomo VT, Gardin JM, Skelton TN, Gottdiener JS, Scott CG, Enriquez-Sarano M. Burden of valvular heart diseases: a population-based study. Lancet 2006;368(9540):1005-11. [DOI] [PubMed] [Google Scholar]

Pardaens 2014

  1. Pardaens S, Moerman V, Willems AM, Calders P, Bartunek J, Vanderheyden M, et al. Impact of the preoperative risk and the type of surgery on exercise capacity and training after valvular surgery. American Journal of Cardiology 2014;113(8):1383-9. [DOI: 10.1016/j.amjcard.2014.01.413] [DOI] [PubMed] [Google Scholar]

Pavy 2006

  1. Pavy B, Iliou MC, Meurin P, Tabet JY, Corone S. Safety of exercise training for cardiac patients: results of the French registry of complications during cardiac rehabilitation. Archives of Internal Medicine 2006;166(21):2329-34. [PMID: ] [DOI] [PubMed] [Google Scholar]

Piepoli 2010

  1. Piepoli MF, Corra U, Benzer W, Bjarnason-Wehrens B, Dendale P, Gaita D, et al. Secondary prevention through cardiac rehabilitation: from knowledge to implementation. A position paper from the Cardiac Rehabilitation Section of the European Association of Cardiovascular Prevention and Rehabilitation. European Journal of Cardiovascular Prevention & Rehabilitation 2010;17(1):1-17. [DOI: 10.1097/HJR.0b013e3283313592] [DOI] [PubMed]

Popma 2019

  1. Popma JJ, Deeb GM, Yakubov SJ, Mumtaz M, Gada H, O'Hair D, et al. Transcatheter aortic-valve replacement with a self-expanding valve in low-risk patients. New England Journal of Medicine 2019;380:1706-15. [DOI] [PubMed] [Google Scholar]

Pressler 2018

  1. Pressler A, Forschner L, Hummel J, et al. Long-term effect of exercise training in patients after transcatheter aortic valve implantation: follow-up of the SPORT:TAVI randomised pilot study. European Journal of Preventative Cardiology 2018;25(8):794-801. [DOI] [PubMed] [Google Scholar]

RevMan 2020 [Computer program]

  1. Review Manager (RevMan). Version 5.4. The Cochrane Collaboration, May 2020.

Ribeiro 2017

  1. Ribeiro GS, Melo RD, Deresz LF, Dal Lago P, Pontes MR, Karsten M. Cardiac rehabilitation programme after transcatheter aortic valve implantation versus surgical aortic valve replacement: systematic review and meta-analysis. European Journal of Preventive Cardiology 2017;24:688-97. [DOI] [PubMed] [Google Scholar]

Schünemann 2013

  1. Schünemann H, Brożek J, Guyatt G, Oxman A, editors. GRADE handbook for grading quality of evidence and strength of recommendations. Updated October 2013. The GRADE Working Group, 2013. Available from guidelinedevelopment.org/handbook.

Sibilitz 2015

  1. Sibilitz K, Berg S, Hansen T, et al. Update to the trial protocol, including statistical analysis plan for a randomized clinical trial comparing comprehensive cardiac rehabilitation after heart valve surgery with control: the CopenHeart VR trial. Trials 2015;16:38. [DOI] [PMC free article] [PubMed] [Google Scholar]

Sibilitz 2015a

  1. Sibilitz KL, Berg SK, Thygesen LC, Hansen TB,  Køber L, Hassager C, et al. High readmission rate after heart valve surgery: a nationwide cohort study. International Journal of Cardiology 2015;189:96-104. [DOI: 10.1016/j.ijcard.2015.04.078] [DOI] [PubMed] [Google Scholar]

Sibilitz 2016 SR

  1. Sibilitz KL, Berg SK, Tang LH, Risom SS, Gluud C, Lindschou J, et al. Exercise‐based cardiac rehabilitation for adults after heart valve surgery. Cochrane Database of Systematic Reviews 2016;Issue 3:Art. No.: CD010876. [DOI: 10.1002/14651858.CD010876.pub2] [DOI] [PubMed] [Google Scholar]

Smart 2018

  1. Smart NA, King N, Lambert JD, et al. Exercise based cardiac rehabilitation improves exercise capacity and health-related quality of life in participants with atrial fibrillation: a systematic review and meta-analysis of randomised and non-randomised trials. Open Heart 2018;5:e000880. [DOI: 10.1136/openhrt-2018-000880] [DOI] [PMC free article] [PubMed] [Google Scholar]

Sterne 2011

  1. Sterne JAC, Egger M, Moher D. Chapter 10. Addressing reporting biases. In: Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011). The Cochrane Collaboration, 2011. Available from www.cochrane-handbook.org.

Sterne 2019

  1. Sterne JAC, Savović J, Page MJ, Elbers RG, Blencowe NS, Boutron I, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ 2019;366:l4898. [DOI: ] [DOI] [PubMed] [Google Scholar]

Supino 2006

  1. Supino PG, Borer JS, Preibisz J, Bornstein A. The epidemiology of valvular heart disease: a growing public health problem. Heart Failure Clinics 2006;2(4):379-93. [DOI] [PubMed] [Google Scholar]

Thompson 2002

  1. Thompson SG, Higgens JP. How should meta-regression analyses be undertaken and interpreted? Statistics in Medicine 2002;21(11):1559-73. [PMID: ] [DOI] [PubMed] [Google Scholar]

Vahanian 2012

  1. Vahanian A, Alfieri O, Andreotti F, Antunes MJ, Baron-Esquivias G, Baumgartner H, et al. Guidelines on the management of valvular heart disease: the Task Force on the Management of Valvular Heart Disease of the European Society of Cardiology (ESC) and the Eurpoean Association for Cardio-Thoracic Surgery (EACTS). European Heart Journal 2012;33(19):2451-96. [DOI: 10.1093/eurheartj/ehs109] [DOI] [PubMed] [Google Scholar]

Whalley 2011

  1. Whalley B, Rees K, Davies P, Bennett P, Ebrahim S, Liu Z, et al. Psychological interventions for coronary heart disease. Cochrane Database of Systematic Reviews 2011, Issue 8. Art. No: CD002902. [DOI: 10.1002/14651858.CD002902.pub3] [DOI] [PubMed] [Google Scholar]

Wood 2008

  1. Wood L, Egger M, Gluud LL, Schulz KF, Jüni P, Altman DG, et al. Empirical evidence of bias in treatment effect estimates in controlled trials with different interventions and outcomes: meta-epidemiological study. BMJ 2008;336(7644):601-5. [DOI: ] [DOI] [PMC free article] [PubMed] [Google Scholar]

Yagdir 2020

  1. Yadgir S, Johnson CO, Aboyans V, et al. Global, regional, and national burden of calcific aortic valve and degenerative mitral valve diseases, 1990-2017. Circulation 2020;121:1670–80. [DOI: 10.1161/CIRCULATIONAHA.119.043391] [PMID: PMID: 32223336] [DOI] [PubMed] [Google Scholar]

References to other published versions of this review

Sibilitz 2013b

  1. Sibilitz KL, Berg SK, Tang LH, Risom SS, Gluud C, Lindschou J, et al. Exercise-based cardiac rehabilitation for adults after heart valve surgery. Cochrane Database of Systematic Reviews 2013, Issue 12. Art. No: CD010876. [DOI: 10.1002/14651858.CD010876] [DOI] [PubMed] [Google Scholar]

Articles from The Cochrane Database of Systematic Reviews are provided here courtesy of Wiley

RESOURCES