Rehabilitation effects of circuit resistance training in coronary heart disease patients: A systematic review and meta‐analysis

Abstract Background and Hypothesis The rehabilitation effect of circuit resistance training in coronary heart disease (CHD) patients remains unclear. We perform this review to examine the rehabilitation effect of circuit resistance training in CHD patients and to provide a basis for the formulation of reasonable individual exercise prescriptions for CHD patients. Methods Randomized controlled trials (RCTs) were searched on PubMed, Web of Science, The Cochrane Library, Embase, Clinical Trials, and CNKI. About 1232 studies were identified. Nine RCTs were finally used for the present meta‐analysis to determine the rehabilitation effect of circuit resistance training in CHD patients, compared to aerobic training. Individuals enrolled for the studies were at a mean age of 60.5 years old and were all CHD patients. Following the PRISMA guidelines, we extracted basic information about the study and patient characteristics, as well as measurements (e.g., the peak oxygen uptake, the body mass index [BMI], the body fat percentage, the systolic blood pressure, the total cholesterol, and triglycerides). Subsequently, this meta‐analysis determined the overall effect by using standardized mean difference (SMD) and 95% confidence interval (CI). Results Compared with aerobic training, circuit resistance training significantly decrease the BMI and the body fat percentage. Conclusions As suggested from the present meta‐analysis of RCTs, circuit resistance training is effective in improving the BMI and the body fat percentage in CHD patients and may help delay the progression of CHD. CRT has the advantage of lower load in most cases with a similar effect.


| INTRODUCTION
Coronary heart disease (CHD) refers to one of the most common cardiovascular system diseases, which has imposed a huge social, medical, and economic burden. 1 Obesity and lack of exercise were believed as independent factors affecting the development of CHD. 2 Regular physical activity and systematic exercise are vital components of most cardiovascular disease treatments, which are associated with the all-cause mortality of the cardiovascular disease. Accordingly, some guidelines 3  Cardiorespiratory Fitness (CRF) refers to an effective and independent predictor of cardiovascular mortality and the risk of all-cause mortality. It is recommended as a clinical indicator, in accordance with a statement issued by the AHA. 4 Exercise training is capable of significantly improving the CRF of patients with CHD, improving the quality of life and long-term prognosis of patients, and greatly reducing the mortality of CHD. 5 Over the past few years, some scholars have proposed circuit resistance training, inconsistent with conventional simple resistance training, which is progressive resistance training based on aerobic training. Compared with conventional resistance training, it is characterized by lighter weight, more repetitions, and shorter training intervals.
Exercise test assessment has been the most extensively used functional test, which can provide a relatively scientific and reasonable basis for formulating exercise prescription. 6 And anaerobic threshold level acts as a vital reference index for the formulation of exercise prescriptions for CHD patients.
Resistance exercise can help to increase myocardial perfusion and improves myocardial ischemia while improving muscle mass and exercise ability. 7 As indicated from existing studies, aerobic training combined with resistance exercise and aerobic training can significantly improve the cardiopulmonary response in patients with CHD. 7 Hansen et al.'s study 8 reported that aerobic combined with resistance exercise exerts a better effect on exercise therapy for patients with CHD than aerobic training independently. However, controversies remain about the evidence of whether circular resistance training can improve cardiopulmonary endurance, weight loss, and lipid reduction in CHD patients.
This systematic review aimed to examine the exercise training effect of circuit resistance training in CHD patients and lay a basis for the formulation of reasonable individual exercise prescriptions for CHD patients.

| Article type
This study is a systematic review and meta-analysis.

| Inclusion and exclusion criteria
Inclusion criteria included the following: stable angina pectoris; old myocardial infarction; percutaneous transluminal coronary angioplasty; and coronary artery bypass grafting. The training period should be no less than 7 days. Studies should report the rehabilitation effect of circuit resistance training in CHD patients compared to aerobic training. The outcome measures should include at least one of these: the peak oxygen uptake, the body mass index (BMI), the body fat percentage, the systolic blood pressure, the total cholesterol, and triglycerides (TGs).
Exclusion criteria consisted of: the presence of uncontrollable arrhythmias; unstable angina pectoris, and uncontrolled high blood pressure; congestive heart failure, exercise-induced angina; exercise-induced hypotension; and subjects with absolute exercise restriction.
Review articles, case-reports were not used. Studies with incomplete follow-up work were excluded.

| Data extraction
All eligible studies were identified by two authors, and the differences were resolved by reaching a consensus with a third researcher.
Several data were extracted: (e.g., the first author article, the country of publication, the year, the number of patients, the average age, the duration of exercise intervention, the peak oxygen uptake, the BMI, the body fat percentage, TG, the total cholesterol, and the systolic blood pressure).

| Statistical analysis
RevMan V. 5.2 and STATA 15.0 were used for statistical analysis.
Combined SMD and its 95% confidence interval [CI] were measured. I 2 test was calculated as a measure of heterogeneity, for which we believe I 2 values of 25%, 50%, and 75% indicate a low, moderate, or high heterogeneity. 10 We took the heterogeneity into account during the evaluation of the statistical effect. And the quality of the studies was assessed independently by two of us (Chunchun Wu and Yaoguo Wang). The random-effects model was adopted to combine effect size when I 2 > 50%; otherwise, we used a fixed-effects model. The Begg test was performed to assess publication bias. Finally, a two-tailed p < .05 was considered statistically significant.

| Searching results
After eliminating duplicates, our search initially identified a total of 1232 references. Next, 176 articles were deemed eligible after a review of titles and abstracts. Then, for the reasons listed in Figure 1, 167 studies were further removed. And we have found some incomplete studies by searching the registry center. Unfortunately, we were unable to contact the relevant units or authors for relevant data. Finally, nine randomized controlled trials (RCTs) focused on the rehabilitation effect of circuit resistance training were used in the meta-analysis. 7,[11][12][13][14][15][16][17][18] And those in the intervention groups were trained by circuit resistance training, while those in the control groups were trained by aerobic training. We outlined the retrieval strategy in Figure 1.

| Study characteristics
Nine RCTs were included in the meta-analysis.
Basic information from publications, mean age, exercise program, main outcome indicators, duration of exercise, and exercise frequency is listed in Table 1. And to show the effect of different training methods on the outcomes more intuitively, we define a new variable: the time ratio, which means the product of the training period and the training duration. And all the absolute values were divided by the time ratio to get the outcomes adjusted by training intensity. The adjusted outcomes were summarized in Table 2.

| Risk of bias
Risk assessment of bias is shown in Figures 2 and 3, and All studies had a low-risk bias in randomized, selective reporting. Most studies were well done for assignment hiding, participant blindness, evaluator blindness, and data integrity. It also indicated that there might be other deviation risks in the study. Heterogeneity test of the body fat percentage showed no statistical significance (Χ 2 = 2.97, p = .40; I 2 = 0%). Accordingly, the fixed effect model was selected for further analysis and treatment.
Systolic blood pressure heterogeneity test difference was statistically significant (Χ 2 = 11.84 p = .003; I 2 = 83%). Thus, random effect model was selected for the further analysis and processing.
Heterogeneity test of the total cholesterol showed no statistical significance (Χ 2 = 1.06 p = .79; I 2 = 0%). Therefore, the fixed effect model was selected for the further analysis and treatment. Result display, no significant difference was identified in the total cholesterol between the experimental and the control groups (SMD = −0.26, 95% CI = −0.53 to 0.00, z = 1.95, p = .05; Figure 4).
No significant difference was identified in TG heterogeneity test (Χ 2 = 1.06 p = .79; I 2 = 0%). Therefore, the fixed effect model was selected for further analysis and treatment. Result display, no significant difference was identified in TG between the experimental and the control groups (SMD = −0.14, 95% CI = −0.41 to 0.12, z = 1.08, p = .28; Figure 4).

| Publication bias
Supporting Information: Figure 1 indicates that VO 2 Peak may have a publication bias risk. However, the Begg test found no evidence of publication bias (p =.652).
Supporting Information: Figure 2 shows the possibility of publication bias in the BMI. However, the Begg test found no evidence of publication bias (p = .09).
Supporting Information: Figure 3  Supporting Information: Figure 6 shows that there is no possibility of publication bias risk for triglycerides. The Begg test found no evidence of publication bias (p = .497).

| DISCUSSION
To the best of the authors' knowledge, this has been the first randomized controlled meta-analysis to date to investigate the rehabilitation effects of circuit resistance training in CHD patients.
As demonstrated from the results of this meta-analysis, compared with aerobic training, circuit resistance training has insignificantly impacted the peak oxygen uptake, the systolic blood pressure, total blood cholesterol, and triglyceride in CHD patients, whereas it can significantly decrease the BMI and the body fat percentage.
The meta-analysis of RCT shows its own advantages and the conclusions drawn are more reliable and accurate since all the data originate from RCTs. As indicated by Begg tests, no significant evidence of publication bias was found in our study. According to the results, the VO 2 Peak, the BMI, the SBP group I 2 , respectively (I 2 = 75%, I 2 = 81%, I 2 = 83%) were more than 50%, and the p values is currently considered the optimal index to assess cardiopulmonary function and exercise tolerance of patients. 20 It was reported that circuit resistance training could significantly increase the VO 2 peak in CHD patients compared with aerobic training. 7,14,17 Other studies have shown that circuit resistance training does not improve the VO 2 peak in patients with CHD. 16 Overall, CRT did not differ significantly from conventional resistance training in terms of patient benefit. However, CRT has the advantage of less load per session and lower training intensity.
This makes CRT more beneficial for CHD patients, especially those with severe CHD or too weak physical conditions. Furthermore, CRT is obviously better than traditional resistance training in improving body fat in patients with CHD, which also makes CRT a better application prospect.
In this systematic review, circuit resistance training significantly reduced the BMI and the body fat content compared with aerobic training. Thus, it can effectively delay the progression of CHD and reduce mortality.

| Local
In this systematic review, some studies have high inter-heterogeneity and small sample size, so large-sample and high-quality RCT studies should be further conducted. The population and nations included in the respective study are different, and there may be regional bias.
Differences exist in the frequency, intensity, duration, and type of exercise training adopted in various studies, thereby probably causing differences in the final measurement index results. The literature retrieval method has some limitations, and there may be omissions.
This meta-analysis includes only studies published in English, which can cause potential publication bias. A small number of included studies are insufficiently rigorous for the assignment hiding, the participant blindness, as well as the evaluator blindness.

| CONCLUSIONS
In brief, although circuit resistance training fails to significantly impacts peak VO 2 as compared with aerobic training independently, it helps reduce the body fat percentage and the BMI and more | 829