Effects of Supervised Cardiac Rehabilitation Programmes on Quality of Life among Myocardial Infarction Patients: A Systematic Review and Meta-Analysis

Coronary heart disease is the leading cause of death and disability worldwide. Traditionally, cardiac rehabilitation programmes are offered after cardiac events to aid recovery, improve quality of life, and reduce adverse events. The objective of this review was to assess the health-related quality of life, after a supervised cardiac rehabilitation programme, of patients who suffered a myocardial infarction. A systematic review was carried out in the CINAHL, Cochrane, LILACS, Medline, Scopus, and SciELO databases, according to the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines. Randomised controlled trials were selected. Meta-analyses were performed for the Short Form Health Survey SF-36, Myocardial Infarction Dimensional Assessment Scale (MIDAS), MacNew Heart Disease-Health-Related Quality of Life (HRQL) questionnaire, and European Quality of Life-Visual Analogue Scale (EuroQol-VAS) with the software Cochrane RevMan Web. Ten articles were found covering a total of 3577 patients. In the meta-analysis, the effect size of the cardiac rehabilitation programme was statistically significant in the intervention group for physical activity, emotional reaction, and dependency dimensions of the MIDAS questionnaire. For the control group, the score improved for SF-36 physical functioning, and body pain dimensions. The mean difference between the control and intervention group was not significant for the remaining dimensions, and neither for the MacNew Heart Disease-HRQL and EuroQol-VAS questionnaires. Supervised cardiac rehabilitation programmes were effective in improving health-related quality of life, however, there was a potential variability in the interventions; therefore, the results should be interpreted with caution. This study supports the importance of providing care and evaluating interventions via the supervision of trained health professionals, and further randomised clinical trials are needed to analyse the positive changes in mental and physical health outcomes.


Introduction
Coronary heart disease is the leading cause of death worldwide [1] and its main manifestation is myocardial infarction (MI). This heart disease causes 1.8 million deaths per year, corresponding to 27% of all deaths in Europe [2], and its prevalence is estimated to increase by 18% from 2013 to 2030 [3]. exercise programme, record of level of physical activity, telephone follow-up, or individual counselling). The comparison was addressed to usual care programmes (defined as standard care based on pharmacologic treatment or other non-supervised rehabilitation programmes and may include health education related to diet, education support, or nonstructured exercise). The outcome was the measurement of health-related quality of life through validated instruments. Therefore, the research question was: Does a supervised exercise-based cardiac rehabilitation programme influence the health-related quality of life of patients after MI?

Eligibility Criteria and Study Selection
The included studies were: (1) randomised clinical trials, (2) acute myocardial infarction patients, (3) adult samples, (4) hospital or outpatient interventions, (5) health-related quality of life measurements during or after a cardiac rehabilitation programme (baseline data collection before intervention and the follow-up during or after a cardiac rehabilitation programme), (6) rehabilitation programme based on controlled and supervised physical activity, (7) studies published in the last 10 years, (8) not restricted by publication language.
The exclusion criteria were: (1) pilot study or protocols; (2) assessed the health-related quality of life with different interventions, (3) cardiac rehabilitation interventions that did not include physical activity, (4) paediatric patients.
In the selection process, the first two authors independently reviewed the title and abstract of the articles found. Finally, the full text was read. A third author was consulted in case of disagreement.

Data Extraction
The data were recorded by two authors using a data coding manual. A third author verified the data in case of disagreement. The following variables were obtained for each of the articles: (1) author, year and country; (2) design; (3) aims; (4) sample; (5) type of intervention; (6) duration; (7) measuring tool; and (8) main results.
The intraclass correlation coefficient was calculated to assess the reliability of the data coding by the researchers: it was 0.98 (minimum = 0.96; maximum = 1). Cohen's Kappa coefficient of the categorical variables was 0.97 (minimum = 0.95; maximum = 1).

Quality Assessment and Risk of Bias
The quality and risk of bias of each study were assessed by two authors who collected the data in a table, which were subsequently verified by two other authors. Quality was checked in accordance with the recommendations of the Oxford Center for Evidence-Based Medicine (OCEBM) [28]. The risk of bias of each study was analysed using the Cochrane Collaboration Risk of Bias tool [29].

Data Analysis
Cochrane RevMan Web software was used to carry out the meta-analysis. A total of 18 meta-analyses were carried out, 8 based on the dimensions of the Short Form Health Survey SF-36 (SF-36), 4 based on the dimensions of the MacNew Heart Disease-Health-Related Quality of Life (HRQL) questionnaire, 5 based on the dimensions of the Myocardial Infarction Dimensional Assessment Scale (MIDAS), and 1 based on the European Quality of Life-Visual Analogue Scale (EuroQol-VAS). Heterogeneity was analysed using I 2 value. Publication bias was assessed with Egger linear regression and sensitivity analysis was performed. Due to the low sample sizes of some of the studies included in the metaanalyses, a random-effects analysis was performed. The questions included in RevMan Web were used for bias analysis. The effect size used was the post-intervention mean and standard deviation provided by the included studies.

Results
After conducting the search 218 articles were found. One-hundred and eighty-six articles were eliminated after reading the title and abstract and removing duplicates. Finally, after reading the full text, the final sample was n = 10. The search and selection process is described in Figure 1. I 2 value. Publication bias was assessed with Egger linear regression and sensitivity analysis was performed. Due to the low sample sizes of some of the studies included in the meta-analyses, a random-effects analysis was performed. The questions included in RevMan Web were used for bias analysis. The effect size used was the post-intervention mean and standard deviation provided by the included studies.

Results
After conducting the search 218 articles were found. One-hundred and eighty-six articles were eliminated after reading the title and abstract and removing duplicates. Finally, after reading the full text, the final sample was n = 10. The search and selection process is described in Figure 1.
The health-related quality of life was measured with the questionnaires SF-36 (n = 4), MIDAS (n = 2), MacNew Heart Disease-HRQL (n = 4), EuroQol-VAS (n = 3), European Quality of Life-5 Dimensions (EuroQol-5D) (n = 2), and others such as the Self-Rated Health General and Health Questionnaire (n = 1). In all studies, the intervention was based on supervised cardiac rehabilitation training, with a duration that ranged from 1 month [30] to 36 months [31]. The exercise included individualised or group programmes, and

Meta-Analysis of the Effect Size of Cardiac Rehabilitation Program on Quality of Life
Studies that provided sufficient statistical information (n = 7) were included in the meta-analysis. There were four studies that calculated the effect size in the SF-36 dimensions, and two studies for the MIDAS dimensions, MacNew Heart Disease-HRQL, and EuroQol-VAS.
In the meta-analyses based on the SF-36 questionnaire (n = 4), the size of the intervention group was n = 1049 patients, while in the control group it was n = 1056. Postintervention means differences were statistically significant for physical functioning and body pain dimensions. In these two cases, the difference in post-intervention means was in favour of the control group. The meta-analysis of the studies using the MIDAS questionnaire had a sample of n = 113 in the control group and n = 110 in the intervention group. In this questionnaire, the difference was statistically significant for physical activity, emotional reaction, and dependency dimensions in favour of the intervention group. Finally, the differences in means were not significant of the MacNew Heart Disease-HRQL dimensions or to EuroQol-VAS. The effect sizes of each questionnaire are shown in Table 2. Forest plots and the risk of bias are shown in Figure S1-S4 in the Supplementary Materials.

Discussion
The purpose of this systematic review and meta-analysis was to assess health-related quality of life after a supervised cardiac rehabilitation programme in patients post-MI. The rehabilitation programmes analysed included a supervised exercise programme with a record of the quantity of physical activity, telephone follow-up, or individual counselling. In the intervention group, the results from the MIDAS questionnaire showed an improvement after supervised cardiac rehabilitation in physical activity, emotional reaction, and dependency dimensions, compared to the control group. Other studies found similar results with significant improvements in the intervention group in physical dimension although there was no significant change in mental and emotional dimensions [40]. Normally, patients in the cardiac rehabilitation programmes exercise more frequently and for longer periods and have more information about the benefits of exercise on risk factors, this fact significantly improved health-related quality of life [41,42].
The results from this study indicated improvements in the control group in the dimensions of SF-36 physical functioning and body pain. Other authors found improvements in the control group in all dimensions, except for the role of emotional body pain and vitality [22,43].
In this meta-analysis, we found no improvements after the intervention in healthrelated quality of life in any other dimension or measurement tool for the intervention group. As corroborated by another meta-analyses, after analysing supervised and nonsupervised, exercise-based cardiac rehabilitation, no statistically significant difference, between groups were found for MacNew Heart Disease-HRQL questionnaire [22]. Additionally, other studies corroborated these facts by not finding significant differences between groups [44,45] or by only finding improvements in physical functioning and general health [43,46] or body pain [44] of the SF-36 dimensions.
It seems that the quantity of physical activity performed is closely linked to the healthrelated quality of life in physical and emotional terms [47]. Therefore, the greater the frequency and duration of the physical activity programmes, the higher the results in the score of each dimension of the SF-36 and MIDAS questionnaires, thus leading to an improvement in health-related quality of life [48,49]. In addition, previous research showed that the early initiation of low-level exercise before discharge from hospital was safe to perform in patients after MI, leading to a significant improvement in exercise tolerance [50][51][52]. Therefore, early exercise led by trained health professionals could positively increase the motivation, which could be translated into increased adherence and tolerance in order to improve health status [51]. Sustained physical activity could also be a key to the quality of life, as well as determining whether the dose and high levels of intensity in the exercise development would be even more beneficial [42,53].
Furthermore, patients who experience MI are more likely to have negative emotional effects that lead to a deterioration of health-related quality of life [54], and thus leaving the treatment and preventing healthy habits. Anxiety and depression are commonly experienced after MI and could persist for months or even years. This fact could also affect access and adherence to rehabilitation programmes; therefore, the early implementation of cardiac rehabilitation programmes could be disrupted [55].
The preventive effects of physical activity, including properly prescribed strength training, are safe and effective in patients with cardiovascular disease [56,57]. Physical training after a cardiac event is essential for improving patient outcomes, as reflected in the recommendation of the American Heart Association [58]. However, this vulnerable population often only receives secondary prevention strategies based on health education, and exercise-based interventions are provided without supervision by health professionals with specific training in this area [59]. Furthermore, few studies include health-related quality of life as an outcome measure when evaluating the effects of cardiac rehabilitation [45,60]. In this study, we found little evidence about the type of intervention, duration of effects over time, or setting (home or centre-based exercise interventions) associated with a true improvement. Providing interventions based on educational support, follow-up and counseling, and supervision by trained health professionals is strongly supported, in order to to improve functional status and health-related quality of life. Developing more randomised clinical trials in different settings, timing, intensity, the type of exercise, and quantity of physical activity could provide evidence for the positive effects on mental and physical health.

Limitations and Further Research
The present study had several limitations. First, the population included in many of the studies was very small. On the other hand, the interventions were relatively short in time. Furthermore, few studies analysed the adherence to the intervention, hence a compromised control programme could yield different results.
There is also a potential variability due to the types of settings, characteristics of the intervention, follow-up time, and modality (individualised programmes or by groups). Therefore, although the research aim was to analyse the effect of supervised cardiac rehabilitation on the improvement of the health-related quality of life after myocardial infarction, the heterogeneity of approaches adopted may influence the study findings.
Supervised cardiac rehabilitation programmes are effective for improving healthrelated quality of life. Health policymakers should improve cardiac rehabilitation programmes, promoting supervision by health professionals, with specific training in this area to generate better public health outcomes [61].
Providing more individualized perspectives offers opportunities to measure the health benefits of interventions in terms of survival and quality of life [22]; thus, more clinical trials with larger sample sizes and longer follow-up are needed. In addition, it would be useful to conduct in-depth studies on the adherence to programmes with motivational interventions, such as gamification [62] or coaching interventions [63].

Conclusions
In the meta-analysis, the effect size of the cardiac rehabilitation programme was statistically significant in the intervention group for physical activity, emotional reaction, and dependency dimensions of the MIDAS questionnaire. For the control group, the score improved in the dimensions for SF-36 physical functioning and body pain. The mean differences between the control and intervention groups were not significant for the remaining dimensions, for MacNew Heart Disease-HRQL, or for EuroQol-VAS questionnaires. Despite finding improvements after cardiac rehabilitation programmes, few studies analyse the effect of a programme supervised by health professionals with the improvement of health-related quality of life as the main objective. More clinical trials with larger sample sizes and longer follow-ups are needed, as well as interventions that support adherence and participation in these programmes.

Conflicts of Interest:
The authors declare no conflict of interest.