Introduction

Kidney transplantation is currently the most desired treatment option for patients suffering from end-stage renal disease (ESRD). Compared with other treatments, kidney transplantation has some advantages in improving survival and quality of life [1, 2]. Despite this, patients generally experience multiorgan dysfunction following the procedure, after allograft transplantation. In addition, patients after kidney transplantation often require lifelong immunosuppressants to prevent graft rejection. These immunosuppressive drugs often result in adverse events such as muscle weakness, osteoporosis, and cardiovascular disease [3, 4]. Therefore, postoperative management of kidney transplantation plays a crucial role after renal transplantation.

In recent years, with the promotion of the enhanced recovery after surgery (ERAS) concept, the effect of appropriate exercise intervention for postoperative patients has gradually begun to receive attention. Exercise is recognized as an effective non-pharmacological intervention that is generally categorized as aerobic, anaerobic, and flexibility exercises. The health benefits of exercise have been demonstrated in healthy people and people with chronic diseases [5]. Related studies have also shown that exercise interventions are effective in patients with solid organ transplants, including heart, kidney, lung, and liver transplants [6]. Although exercise interventions are considered beneficial, routine exercise intervention programs for renal transplant recipients are not used as part of standard clinical care. Besides, the evidence on the impact of exercise intervention on kidney transplant recipients is limited. The few available studies have only focused on the effects of exercise intervention on several aspects of exercise tolerance, cardiorespiratory fitness, and quality of life in kidney transplant recipients [7,8,9].

This systematic review and meta-analysis aimed to compare the changes of the exercise intervention group and control group, to fully assess the effects of an exercise intervention on kidney transplant recipients.

Methods

Search strategy

We searched the PubMed, Embase, and Cochrane Library databases using kidney transplantation, exercise training, and randomized controlled trials as keywords. Depending on PICOS (populations, interventions, comparators, outcomes, and study designs) strategy, four authors independently conducted the searches. Table 1 shows the search strategy. This meta-analysis has been registered on PROSPERO with the number CRD42022357574. PRISMA 2020 checklist is shown in the supplementary material.

Table 1 Search strategy according to populations, interventions, comparators, outcomes, and study designs (PICOS)
Full size table

Inclusion criteria

The RCTs included in this study were required to fulfill all of the following inclusion criteria: (1) the study analyzed the effect of exercise intervention for kidney transplant recipients was analyzed; (2) the study contained valued data that could be analyzed and related outcome index; (3) full-text content was accessible; (4) the study was an RCT. The population inclusion criteria for RCTs were more stringent than other prospective and retrospective studies.

Quality assessment

Studies were categorized according to the Cochrane Risk of Bias Tool for Randomized Trials [10], version 2 (RoB2), recommended by the Cochrane Handbook for the Systematic Evaluation of Interventions [11], version 6.2. According to RoB2, we categorized studies into three levels: low risk of bias, moderate risk of bias, and high risk of bias.

Data extraction

From each included RCT, the following information was extracted: (I) the name of the first author; (II) the time of publication and the type of design; (III) the sample size of each group; (IV) the methods of exercise intervention; (V) the time of intervention; (VI) the outcomes of study: anthropometric characteristics, body composition, renal function, cardiorespiratory function, blood parameters, serum cytokine levels, physical performance and quality of life.

Statistical and meta-analysis

Data were analyzed using Review Manager software (RevMan, version 5.3.0, Cochrane Collaboration) [12]. This study adopted mean difference (MD) for assessing continuous data and adopted odds ratios (ORs) with 95% confidence intervals (CIs) for assessing dichotomous data. We considered studies with p values > 0.05 as homogeneous and conducted the analysis using a fixed-effects model. Conversely, we employed a random-effects for analyzing heterogeneous studies. The present study checked for inconsistency through I2 statistics. The value of p < 0.05 was deemed to be statistically significant.

Results

Characteristics of eligible studies

195 articles met the above inclusion criteria and were retrieved from the database. After screening the titles and abstracts, 163 articles were excluded. We reviewed the remaining articles. Among them, 12 studies were removed due to they were not RCTs. Then, 4 studies were eliminated because they missed key information. In the end, 16 RCTs [7,8,9, 13,14,15,16,17,18,19,20,21,22,23,24,25] were included in the final analyses. The flowchart of the selection process is shown in Fig. 1. The details of the included studies are given in Table 2.

Fig. 1
figure 1

Flowchart of the study selection process. RCT, randomized controlled trials

Full size image
Table 2 Study and patient characteristics
Full size table

Quality of eligible studies

The studies included in our meta-analysis were all RCTs. All studies performed a sample size calculation. Eleven of these RCTs were graded A for quality. Only one study reported an intention-to-treat analysis [13]. No patients were lost during follow-up in the ten studies [7, 8, 13, 15,16,17, 19, 21, 22, 24]. The quality of included studies is shown in Table 3.

Table 3 Quality assessment of individual study
Full size table

Efficacy

We studied the effects of exercise intervention on measurement parameters in kidney transplant patients. Patients in the control group received standard treatment for the same length of time.

Anthropometric characteristics

Body mass index (BMI)

Seven RCTs involving 393 patients compared the differences between the two groups after the intervention in terms of BMI (Supplementary Fig. 1A). Because of p > 0.05, we conducted a fixed-effects model for the study. The results showed no statistical difference in BMI between the two groups after the intervention treatment (MD: 0.12, 95% CI − 0.72 to 0.96, Chi2 = 3.40, p = 0.78).

Waist circumference

Three RCTs reported the changes between the two groups of patients after the intervention in terms of waist circumference (Supplementary Fig. 1B). Since p > 0.05, a fixed-effects model was used to analyze group differences. The model indicated that the MD was 1.50, the 95% CI was − 3.74 to 6.74, the I2 was 25%, and the Chi2 value was 2.67 (p = 0.58). We suggested that the exercise intervention and control groups were similar in terms of the waist circumference of patients.

Hip circumference

Two RCTs analyzed the changes in the hip circumference of 32 patients after the intervention (Supplementary Fig. 1C). A fixed‐effects model was used to assess changes between the two groups, which showed an MD of − 1.19 (95% CI − 6.10–3.72, p = 0.63). There was no significant difference between the two groups concerning hip circumference.

Body composition

Bone mineral density (BMD)

Because of p > 0.05, we employed a fixed-effects model to compare the BMD between the exercise intervention and control groups from two RCTs (Supplementary Fig. 2A). The pooled estimate of MD was − 0.02, 95% CI was − 0.07 to 0.03, I2 was 45%, and Chi2 was 1.80 (p = 0.45). The results showed that the exercise intervention and control groups were similar regarding BMD.

Lean body mass (LBM)

Three RCTs analyzed the differences between the two groups after the intervention in terms of LBM (Supplementary Fig. 2B). Due to p > 0.05, we conducted a fixed-effects model for the study. The results showed no statistical difference in LBM between the two groups after the intervention treatment (MD: 1.21, 95% CI − 2.35 to 4.78, Chi2 = 5.45, p = 0.50).

Renal function

Creatinine

Six RCTs involving 317 patients reported the changes between the two groups of patients after the intervention in terms of creatinine (Fig. 2A). Since p = 0.05, a fixed-effects model was used to analyze group differences. The model revealed that the MD was − 0.29, the 95% CI was − 0.46 to − 0.11, the I2 was 54%, and the Chi2 value was 10.90 (p = 0.001). We concluded that creatinine was greatly improved in the exercise intervention group than in the control group.

Fig. 2
figure 2

Forest plots showing changes in A creatinine; B urea; C estimated glomerular filtration rate (eGFR)

Full size image
Urea

Two RCTs analyzed the changes in the area of 28 patients after the intervention (15 in the exercise intervention group and 13 in the control group) (Fig. 2B). We performed a fixed-effects model to analyze differences between groups, due to p > 0.05. The model revealed that the MD was − 21.57, the 95% CI was − 35.84 to − 7.29, the I2 was 0%, and the Chi2 value was 0.62 (p = 0.003). Significant improvements in urea were found in the exercise intervention group.

Estimated glomerular filtration rate (eGFR)

Three RCTs were included in our study. A random-effects model showed that there was no difference between the exercise intervention group and the control group in eGFR (MD: 16.16, 95% CI − 3.98 to 36.29, Chi2 = 9.33, p = 0.12, Fig. 2C).

Cardiorespiratory function

Systolic blood pressure

Because of p < 0.05, we employed a random-effects model to compare the systolic blood pressure between the exercise intervention group and control group from six RCTs (Supplementary Fig. 3A). The pooled estimate of MD was − 1.53, 95% CI was − 4.70 to 1.64, I2 was 0%, and Chi2 was 3.34 (p = 0.34). The results showed that the exercise intervention and control groups were similar regarding systolic blood pressure.

Diastolic blood pressure

Six RCTs involving 258 patients analyzed the differences between the two groups after the intervention in terms of diastolic blood pressure (Supplementary Fig. 3B). Due to p > 0.05, a fixed-effects model was utilized for analyzing data. The results showed no statistical difference in diastolic blood pressure between the two groups after the intervention treatment (MD: − 0.03, 95% CI − 2.25 to 2.19, Chi2 = 4.52, p = 0.98).

Heart rate

Because of p < 0.05, we compared the heart rate between the exercise intervention and control groups from four RCTs by a random-effects model (Supplementary Fig. 3C). The pooled estimate of MD was − 2.32, 95% CI was − 9.26 to 4.62, I2 was 79%, and Chi2 was 14.12 (p = 0.51). The results showed that the exercise intervention and control groups were similar in heart rate.

Peak oxygen uptake (VO2peak)

Six RCTs involving 251 patients (136 patients in the exercise intervention group, and 115 patients in the control group) reported the changes between the two patients after the intervention regarding VO2peak (Supplementary Fig. 3D). Since p > 0.05, a fixed-effects model was used to analyze group differences. The model revealed that the MD was 3.20, the 95% CI was 1.97–4.43, the I2 was 6%, and the Chi2 value was 5.34 (p < 0.00001). We concluded that the exercise intervention group recorded a statistically significant improvement regarding the VO2peak.

Blood parameters

Total cholesterol

Five RCTs analyzed the changes in total cholesterol of 335 patients after the intervention (Supplementary Fig. 4A). A fixed‐effects model was utilized to evaluate differences between the two groups, which showed an MD of − 0.06 (95% CI − 0.33 to 0.21, p = 0.65). These results reflect no significant effect on kidney transplant recipients of total cholesterol with exercise intervention.

High-density lipoprotein-cholesterol (HDL-C)

Four RCTs reported differences in HDL-C of 229 patients after the intervention (Supplementary Fig. 4B). A fixed-effects model was used to conduct the analysis, due to p > 0.05. The model revealed that the MD was 0.21, the 95% CI was 0.04–0.37, the I2 was 0%, and the Chi2 value was 0.62 (p = 0.01), confirming greater improvements in HDL-C in the exercise intervention group.

Hemoglobin

Because of p > 0.05, we analyzed the hemoglobin between the exercise intervention and control groups from three RCTs using a fixed-effects model (Supplementary Fig. 4C). The pooled estimate of MD was 0.23, 95% CI was − 0.10 to 0.56, I2 was 0%, and Chi2 was 0.88 (p = 0.18). The exercise intervention group had a similar hemoglobin compared to the control group.

Serum cytokine levels

Tumor necrosis factor-α (TNF-α)

Two RCTs were included in our study. A random-effects model suggested that there was no difference between the exercise intervention group and the control group in TNF-α (MD: − 1.44, 95% CI − 4.39 to 1.51, Chi2 = 8.17, p = 0.34, Supplementary Fig. 5A).

Interleukin-6 (IL-6)

Two RCTs analyzed levels of IL-6 (Supplementary Fig. 5B). Pooled results from a fixed-effects model suggested that the exercise intervention group did not differ significantly from that of the control group regarding levels of IL-6 (MD = − 0.70, 95% CI − 1.56 to 0.17, p = 0.11).

Physical performance

60-s sit-to-stand test (60-STS)

Two RCTs recorded the differences in 60-STS of 62 patients after the intervention (Fig. 3A). We employed a fixed-effects model to compare differences between groups, due to p > 0.05. The model revealed that the MD was 14.47, the 95% CI was 8.89–20.04, the I2 was 52%, and the Chi2 value was 2.07 (p < 0.00001). The exercise-trained patients had obvious improvement in the 60-STS.

Fig. 3
figure 3

Forest plots showing changes in A 60-s sit-to-stand test (60-STS); B 6-min walk distance (6-MWD); C 6-min walk test (6-MWT); D handgrip strength

Full size image
6-Minute walk distance (6-MWD)

Three RCTs analyzed the changes in 6-MWD of 201 patients after the intervention (Fig. 3B). A random‐effects model was utilized to evaluate differences between the two groups, which showed an MD of 91.87 (95% CI 38.34–145.39, p = 0.0008). The results suggested a statistically significant improvement in the exercise intervention group regarding the 6-MWD.

6-Minute walk test (6-MWT)

Two RCTs analyzed the changes in 6-MWT of 169 patients after the intervention (Fig. 3C). A fixed‐effects model was utilized to evaluate differences between the two groups, which showed an MD of 44.08 (95% CI 20.30–67.87, p = 0.0003). The results suggested that the exercise intervention group was significantly superior to the control group in 6-MWT.

Handgrip strength

Because of p < 0.05, we used a random-effects model to analyze the handgrip strength between the exercise intervention and control groups from two RCTs (Fig. 3D). The pooled estimate of MD was 3.66, 95% CI was − 5.80 to 13.11, I2 was 75%, and Chi2 was 4.02 (p = 0.45). These results reflect no statistical difference in handgrip strength between the two groups after the intervention treatment.

Quality of life short form-36 questionnaire (SF-36)

Physical function score

Four RCTs reported the changes between the two patients after the intervention regarding physical function score (Fig. 4A). Since p < 0.05, a random-effects model was employed to analyze group differences. The model revealed that the MD was 6.93, the 95% CI was − 4.76 to 18.61, the I2 was 81%, and the Chi2 value was 15.54 (p = 0.25). The results suggested that the exercise intervention and control groups were similar in physical function scores.

Fig. 4
figure 4

Forest plots showing changes in A physical function score; B social function score; C role-physical score; D mental composite score; E general health score; F body pain score; G vitality score; H role-emotional score

Full size image
Social function score

Four RCTs analyzed the changes in social function scores of 170 patients after the intervention (Fig. 4B). We performed a random-effects model to analyze differences between groups. The model revealed that the MD was 9.22, the 95% CI was − 5.35 to 23.80, the I2 was 80%, and the Chi2 value was 15.28 (p = 0.21). There was no significant difference between the two groups about the social function score.

Role-physical score

Four RCTs analyzed differences in role-physical scores (Fig. 4C). Pooled results from a fixed-effects model suggested that the exercise intervention group had no significant effect on the role-physical score (MD = 4.94, 95% CI − 4.76 to 14.63, p = 0.32).

Mental composite score

Our study included three RCTs (Fig. 4D). A fixed-effects model suggested that the MD was − 0.17, the 95% CI was − 4.80 to 4.46, the I2 was 0%, and the Chi2 value was 0.83 (p = 0.94), thus indicating that there was no significant difference between the two groups about the mental composite score.

General health score

Three RCTs analyzed the changes in the general health score of 139 patients after the intervention (Fig. 4E). We performed a random-effects model to analyze differences between groups. The model revealed that the MD was 7.90, the 95% CI was − 10.66 to 26.47, the I2 was 87%, and the Chi2 value was 14.87 (p = 0.40). There was no significant difference between the two groups about the general health score.

Body pain score

Three RCTs analyzed differences in body pain scores (Fig. 4F). Pooled results from a random-effects model suggested that the exercise intervention group had no significant effect on the body pain score (MD = 17.00, 95% CI − 9.13 to 43.14, p = 0.20).

Vitality score

Our study included three RCTs (Fig. 4G). A random-effects model suggested that the MD was 16.45, the 95% CI was − 7.82 to 40.71, the I2 was 93%, and the Chi2 value was 30.10 (p = 0.18), thus indicating that there was no significant difference between the two groups about the vitality score.

Role-emotional score

Because of p > 0.05, we analyzed the role-emotional score between the exercise intervention and control groups from three RCTs using a fixed-effects model (Fig. 4H). The pooled estimate of MD was − 1.21, 95% CI was − 11.69 to 9.27, I2 was 0%, and Chi2 was 0.40 (p = 0.82). The exercise intervention group had a similar role-emotional score compared to the control group.

Discussion

Many patients with ESRD eventually ultimately require kidney transplantation to stay alive. Although advances have been made in surgical procedures, many difficult clinical issues remain in the management of patients during post-transplant. The incidence of postoperative cardiovascular disease (CVDS) is 4–6 times higher in kidney transplant recipients than in the general population [26, 27]. It is currently the leading cause of death in kidney transplant recipients [28]. Relevant studies have reported that kidney transplant recipients are also at increased risk of dyslipidemia, possibly related to using immunosuppressive drugs such as cyclosporine, glucocorticoids, and sirolimus [29]. In addition, patients after kidney transplantation often present with significant motor dysfunction [30]. Therefore, post-transplant management is important for patient recovery, and adjuvant treatment strategies may have important prognostic potential.

Exercise intervention therapy is a rehabilitation method that focuses on functional exercise. Over recent years, exercise training has generated interest as an adjunctive treatment strategy for surgical procedures. Many studies pointed out that perioperative exercise intervention in patients with gastrointestinal tumors can reduce the risk of complications [31] and shorten postoperative hospital stays [32]. Cavalheri et al. [33] suggested training can improve exercise capacity and quadriceps strength in patients after lung cancer surgery. Besides, many studies also analyzed the effects of exercise intervention programs on patients after solid organ transplantation. Raphael et al. [34] found that exercise intervention can significantly improve heart transplantation patients' peak heart rate and aerobic capacity. Langer et al. [35] concluded that exercise training improves functional recovery in postoperative uncomplicated lung transplant patients and that postoperative exercise intervention programs should be strongly encouraged in elderly lung transplant recipients. Stefan [36] demonstrated that exercise training is safe for liver transplant recipients, improves physical functional aspects, and may benefit cardiopulmonary and muscle health. Emily et al. [37] proved that an exercise training intervention can improve exercise capacity and quality of life in patients after lung transplantation.

VO2peak is an important measure to evaluate cardiorespiratory fitness [17]. It is commonly used to analyze various patients' aerobic work capacity [38, 39] and physical exercise effects [40]. In addition, some research confirmed excellent test–retest reliability for VO2peak [41, 42]. Our study considered VO2peak as an index to evaluate cardiopulmonary function in patients after kidney transplantation.

The SF-36 is a commonly used tool to evaluate subjective health-related quality of life [43]. It is now widely available for evaluating the quality of life in various diseases. The SF-36 comprises eight items, each reflecting different health aspects: physical function, role-physical, body pain, general health, vitality, social function, role-emotional, and mental health. In the present study, we used the SF-36 scores to analyze the patients’ quality of life.

Our study included 16 RCTs containing 827 patients. We analyzed the effects of exercise intervention in kidney transplant recipients from eight dimensions, including anthropometric characteristics, body composition, renal function, cardiorespiratory function, blood parameters, serum cytokine levels, physical performance, and SF-36 scores. Analysis of the results revealed that exercise intervention had some positive effects on improving renal function. Specifically, patients in the exercise intervention group showed significantly improved creatinine and urea than the control group. In terms of cardiorespiratory function, patients following the exercise intervention program had significant superiority in improving VO2 peak. Moreover, exercise intervention induced improvements in HDL-C, 60-STS, 6-MWD, and 6-MWT. And no difference was found between groups in terms of anthropometric characteristics, body composition, serum cytokine levels, and SF-36 scores. These findings laid a theoretical foundation for introducing exercise intervention in kidney transplant recipients.

In contrast to previously published studies [44, 45], we found that exercise intervention had shown advantages in improving renal function and dyslipidemia in kidney transplant recipients. These findings may be more clinically relevant. Additionally, this study analyzed the efficacy of exercise intervention in patients with kidney transplants from multiple dimensions and involved more evaluation indicators, offering more comprehensive results. Finally, our study, with more RCTs and a larger sample size, included, may provide more reliable results.

There are some shortcomings in the present study. First, the intervention duration and exercise programs of each RCT included were not unified, which may also result in bias of results. Second, given that many studies in recent years have reported on the effects of exercise intervention on renal transplant recipients, our study was limited by its lack of novelty. Therefore, we will continue the research topic and focus on the latest RCTs to address this.

Conclusions

Our meta-analysis concluded that appropriate exercise intervention can improve renal function, cardiopulmonary function, dyslipidemia, physical performance, and quality of life in renal transplant recipients. The patients should be encouraged to participate in an exercise training intervention after kidney transplantation. These findings will assist clinicians in developing and applying exercise rehabilitation programs specifically designed for kidney transplant recipients as part of standard medical care. In addition, our study will help bridge the gap in knowledge about the importance of exercise intervention programs in kidney transplant patients.