Introduction

Chronic kidney disease (CKD) is currently recognized as a global public health concern, with a substantial amount of end-stage renal disease patient in need for renal replacement therapy or kidney transplantation [1,2,3]. Apart from increased morbidity and mortality, the CKD causes decreased quality of life, malnutrition, impaired cognitive function, deteriorated sleep and increased depression rate among CKD patients [4, 5].

Patients with CKD/ESRD are often inactive. This has both a physiologic and a psychological basis. With routine dialysis interventions performed three times per week for 4–5 h per session, physical activities of dialysis patients are significantly limited, resulting in functional disability and inactivity [6, 7]. Furthermore, the patients have protein energy wasting and also muscle atrophy which also limits their physical activity [8]. As these patients are also depressive, most of them have limited energy and a decreased will for daily activities [9]. Each of these conditions (decreased quality of life, depression, decreased physical activity, muscle atrophy) per se is also related to increased morbidity and mortality [10,11,12]. Given the fact that low physical activity is frequent and related to morbidity and mortality, the Kidney Disease Outcomes Quality Initiative Clinical Practice Guidelines recommends to routinely counsel dialysis patients on increasing their physical activity levels [13, 14]. It was suggested that exercise has also a direct effect on glomerular filtration rate (GFR) and effective renal plasma flow (ERPF) [15].

Various studies have proven exercise to have impact on physical activity, quality of life, depression and cardiovascular function [16,17,18,19,20]. These data stimulate researchers to perform studies regarding the effects of aerobic and/or anaerobic exercise in CKD and ESRD patients. Bearing these issues in mind, in this review we tried to summarize the studies regarding the impact of exercise in CKD and ESRD patients including hemodialysis and peritoneal dialysis patients.

Methods

Data sources and literature search

A literature search was performed using electronic databases MEDLINE, Ovid/MEDLINE (1988–2017), PubMed/MEDLINE, Embase and ISI Web/Web of Science for published studies from January 1988 to April 2017. We searched for relevant studies using the keywords “chronic kidney disease,” “not on dialysis,” “predialysis,” “uremia,” “hemodialysis,” “peritoneal dialysis,” “end stage renal disease,” “exercise,” “quality of life,” “physical activity,” “depression,” “muscle atrophy,” “cardiovascular events” and “all-cause mortality” limiting the search. Only research articles involving humans and published in English were included. Neither unpublished data nor abstracts were included (Fig. 1).

Fig. 1
figure 1

Flow diagram of the study selection process

Full size image

Study selection

Eligibility criteria for inclusion in this review were: randomized or observational design, patients with CKD and ESRD (both hemodialysis and peritoneal dialysis), exercise test performed, reports of quality of life, “physical activity,” “depression,” “muscle atrophy,” “cardiovascular events” and “all-cause mortality.”

The quality of the studies was assessed by the Newcastle–Ottawa scale [21]. This scale used selection of the study groups, the comparability of the groups and the assessment of outcome. Stars were given for each quality item to serve as a quick visual assessment, and the highest quality studies are awarded up to nine stars (Table 1).

Table 1 Newcastle–Ottawa quality assessment of the included trials
Full size table

Outcome measures

We assessed the association between any exercise program (endurance, resistance or combination) with “physical activity,” “depression,” “muscle atrophy,” “cardiovascular events” and “all-cause mortality.”

Results

There are currently 38 studies, 22 observational [6, 22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43] and 16 randomized [16, 20, 29, 44,45,46,47,48,49,50,51,52,53,54], investigating the relationship between exercise and the change in body composition, arterial stiffness, blood pressure (BP), physical function, muscle performance, cardiopulmonary exercise performance and/or quality of life. Studies involving ESRD patients and CKD patients are summarized in Tables 2 and 3, respectively. The majority of these studies were performed in HD [6, 22,23,24,25,26,27,28,29,30,31,32,33, 44, 47, 48, 52, 53], while 6 were done in CKD not on dialysis [16, 20, 34, 35, 50, 51], 2 in both HD and PD [45, 46] and 1 in transplanted patients [49]. The exercise period varied between 4 weeks [22] and 5 years [28]. The types of training involved in these studies were heterogeneous: only aerobic—cycle ergometer or stationary bike [6, 22, 23, 25, 29,30,31, 44, 47, 49, 52], walking [16, 45, 46] or a combination of different aerobic methods [20, 26, 51]; only anaerobic [24, 34, 47, 52, 53]; both aerobic and anaerobic [27,28,29,30, 32, 48,49,50]. There was also a study that used a virtual reality exercise program [33].

Table 2 Characteristics of the included trials that involve end-stage renal disease patients
Full size table
Table 3 Characteristics of the included studies that involve chronic kidney disease patients
Full size table

Exercise and body composition

Four studies evaluated this relationship [6, 16, 33, 53]. Three studies used bioimpedance to assess body composition [6, 33, 53], while the last one used dual-energy X-ray absorptiometry [16]. While some of these studies showed that there was no improvement in muscle mass [6, 16], body fat mass [6], fat percentage [6, 16], arm muscle mass [33] or body fat rate [33], some showed a significant increase in skeletal muscle mass [33, 53], leg muscle mass [33] or decrease in body fat rate [53].

Exercise and arterial stiffness

Three studies examined this association [20, 49, 50]. Mustata et al. [20] showed that in patients with CKD stages 3 and 4, with different aerobic exercises for 12 months the augmentation index is improved in the exercise group. Similarly, in CKD non-dialysis patients, Greenwood et al. observed a reduction in the pulse wave velocity, both in 6 and 12 months, by using a combined aerobic and anaerobic training as compared to a usual-care group [50]. Later, the same group also showed that in transplanted patients, either an aerobic or anaerobic type of exercise for 12 weeks significantly reduced the pulse wave velocity as compared with a control group [49].

Exercise and blood pressure

This relationship was evaluated in 5 studies [16, 23, 25, 32, 50]. Henrique et al. [23] showed that a 12-week program of 30 min on cycle ergometer in the first 2 h of each hemodialysis session significantly reduces average systolic, diastolic and mean BP (assessed with ambulatory BP monitoring) despite maintaining the same doses of antihypertensive drugs and dry weight of patients. Musavian et al. [25] revealed that passive intradialytic pedaling exercise (and not active pedaling) was associated with an improvement in diastolic BP as compared with the control group. There was no benefit of passive or active pedaling in regard to systolic BP. Aoike et al. [16] also showed that walking for 30 min for 12 weeks is associated with an improvement in both systolic and diastolic BP control. However, the remaining studies did not find any benefit for BP control by using a combined aerobic and anaerobic training [32, 50].

Exercise and physical function

This relationship was evaluated in 18 studies [6, 16, 22,23,24, 26,27,28,29,30, 32, 34, 45,46,47, 49, 51,52,53]. Numerous tests were used to assess physical function, the more frequent being: the six- [6, 16, 23, 24, 26,27,28,29, 34, 45,46,47, 51] or 10-min [22] walk test; sit-to-stand test [27,28,29,30, 45, 47, 48, 50, 51, 53]; timed up and go test [16, 22, 28, 34]; and stair climb time [22, 32, 52].

In regard to the walk test, an improvement in the distance walked in the specific time interval (6 or 10 min) was observed in the majority of the studies [6, 16, 23, 26, 27, 29, 34, 45, 46, 51]. Anding et al. [28] observed a beneficial effect only at 12 months, but not at 6 months of follow-up. There were 2 studies that did not find a statistically significant improvement of this test with exercise [24, 47]. Similar findings were also observed for the sit-to-stand test. Exercise was associated with a significant increase in the number of repetitions or the time to perform a number of repetitions in the majority [16, 27,28,29, 45, 48, 50, 53], but not in all the studies [30, 47, 53]. Greenwood et al. observed an improvement in this test only in the anaerobic group and not also in the aerobic one, as compared with controls [49].

All the studies that assessed physical function using the timed up and go test showed a significant beneficial effect of training [16, 22, 28, 34]. Two studies found a significant decrease in time to climb [22, 52], while in the study by Molested et al. [32], this time did not change after 12 training.

Exercise and muscle performance

Ten studies evaluated the relationship between exercise and muscle performance [6, 22, 24, 27, 33,34,35, 47, 50, 53]. Lower body muscle performance was assessed by various tests including leg-press strength [22, 24, 33, 53], one repetition maximum [34, 35, 47], leg extension power [22, 27, 50], maximum peak torque of the knee [6]. Handgrip was used as a surrogate for upper body muscle performance [24, 27, 33, 53].

Leg-press strength is improved following exercise in all the studies [22, 24, 33, 53]. The results for the other tests are more heterogeneous. Thompson et al. [47] did not observe a significant increase in one repetition maximum after training, while the other two studies have found an improvement in this parameter [34, 35]. Similarly, Esteve et al. [27] showed an increase in leg extension power, Greenwood et al. [49] found a similar increase, but only in those that were performing anaerobic, and not aerobic, exercise, while Storer et al. [22] did not find any beneficial effect of training on this test. There was no improvement in knee peak torque after training [6]. Handgrip was improved in only one study [27].

Exercise and cardiopulmonary exercise performance

Cardiopulmonary exercise performance was assessed by measuring maximum oxygen uptake [6, 16, 20, 22, 23, 32, 48,49,50]. Only five of these studies found an improvement in maximum oxygen uptake following training [20, 22, 32, 49, 50].

Exercise and quality of life

Sixteen studies assessed the effect of training on quality of life [6, 20, 24,25,26,27,28,29,30, 32, 45, 47, 48, 51,52,53] and/or depression [48, 51]. The most frequently used questionnaire was the short form 36 (SF-36) [6, 20, 24,25,26, 29, 30, 32, 47, 51, 52], which evaluates the changes in eight subscales of quality of life: physical functioning, role physical, bodily pain, general health, vitality, social functioning, role emotional and mental health. Based on these subscales, physical component summary (PCS) and mental component summary (MCS) domain scores could also be calculated. Other questionnaires used were: the Kidney Disease Quality of Life Short Form (KDQOL-SF) [45, 52], the EuroQol-5D (EQ-5D) [20, 27], Short-Form General Health Survey (RAND-36) [48].

The effect of exercise on quality of life differs depending on the study chosen. Among studies that reported on every subscale of SF-36, three studies showed no statistically significant change in any of the domains [6, 20, 26]. While some studies suggested a significant improvement in the PCS [29, 32, 47, 51] and/or MCS [51, 53] with training, others did not [20, 24, 26, 47]. Painter et al. [29] observed an increase in the PCS, but not in the MCS, only in those patients with low baseline PCS values.

Using the KDQOL-SF questionnaire, Manfredini et al. showed that the global score on average changed favorably in the exercise condition compared with the control arm, but the difference largely failed to achieve any statistical significance. When compared with changes in the control arm, only two items—both in the kidney disease component (cognitive function and quality of social interaction)—achieved formal statistical significance [45]. In the other study that used the same questionnaire, the anaerobic training group reported a better quality of life in terms of social support, patient satisfaction and general health, while the aerobic training group described a general well-being related to domains referring to physical functioning, pain, symptoms, sleeping, sexual function and energy/fatigue [52].

There was no improvement in the quality of life when the EQ-5D questionnaire was used [20, 27]. There was an improvement for the RAND-36 components of vitality, general health perception and health change [48].

The effect on depression was assessed separately using the geriatric depression scale [24], the beck depression inventory [27], the self-rating depression scale [48] or the hospital anxiety and depression scale [51]. With the exception of one study [48], all studies mentioned a significant improvement in depression assessments [24, 27, 51].

Discussion

Exercise has favorable effects on various parameters including physical function and muscle function/atrophy. It is known that physical inactivity is associated with increased mortality in CKD patients [55, 56].

It was shown that physical activity is increased in CKD and ESRD patients after exercise programs [18, 22, 23, 26, 29, 31, 34, 45, 57, 58]. Physical work capacity is generally decreased in HD patients due to myopathies, neuropathies, peripheral vascular pathology or anemia [59]. As these pathologies are associated with uremic toxins, it was hypothesized that increased toxin clearance with intradialytic exercise would minimize their effect on various physiologic systems, thereby enhancing cardiovascular and skeletal muscle performance [26]. With regard to muscle wasting, it was shown that both high-intensity and low-intensity training programs improved the muscle strength in HD patients [60,61,62,63,64]. Storer et al. demonstrated that after 8.6 ± 2.3 weeks of endurance exercise performed three times every week significantly increases leg-press strength and fatigability. Similarly, a trend for improved leg extension power has been observed [22]. The finding is somehow contradictive since endurance training has little effect on the development of muscle strength or power in healthy individuals [65]. However, individuals in chronic deconditioned states such as CKD/ESRD are expected to exhibit the greatest initial gain in muscle function with training in conjunction with a large adaptation potential [22]. Thus, endurance exercise training, even at lower intensities, may provide adequate resistance to improve muscle function in CKD patients. Cho et al. demonstrated that a virtual reality exercise program (VREP) for 40 min, 3 times a week for 8 weeks, improved leg muscle mass in HD patients. Back strength (kg), leg strength (kg), balance (second) have all been improved after VREP [33].

Pomidori et al. investigated the effect of two 10-min walking sessions every other day at an intensity below a speed specific to the patient considering the patient’s lung function and respiratory muscle strength, evaluated by spirometry and maximal inspiratory pressure, respectively. Minimal dose of structured exercise maintained a stable respiratory muscle function, in contrast to the control group (self-care without exercise program) where it worsened [46].

Exercise training also improves cardiorespiratory function and cardiovascular outcomes. One of the most commonly studied parameters in this respect is the oxygen consumption at peak exercise (VO2 peak). As a metric that provides an index of exercise capacity, VO2 peak is indicative of the cardiovascular system’s ability to take up, distribute and use oxygen at maximal exercise [38]. Low peak VO2 in CKD patients can be due to a variety of conditions including anemia, electrolyte imbalance, hyperparathyroidism and respiratory problems [40, 66]. Several previous studies have demonstrated VO2 peak increase after endurance exercise in HD and CKD patients [22, 67, 68]. However, other studies did not show any improvement in VO2 peak [23] which may be attributed to prescribing low-intensity exercise [69] or short-duration aerobic training [70].

A very recent study has proposed a lower VO2 in CKD patients due to endothelial dysfunction. Downey et al. demonstrated that lower FMD values were associated with an upwardly concave systolic BP during exercise in CKD and with poorer exercise capacity measured as VO2 peak. The authors suggested that exercise intolerance in CKD patients may be due to decreased nitric oxide (NO) bioavailability and endothelial dysfunction causing impaired vasodilation during exercise [71]. On the contrary, Habedank et al. noticed no correlation between endothelial (dys)function and peak VO2. The authors argue that the lack of association may relate to the physical condition of the cohort investigated which may be rather preserved high peak VO2 levels (about 24 mL/min/kg) [39].

Arterial stiffness is an important marker of cardiovascular health and is predictive of outcome in HD, CKD and renal transplant patients [72, 73]. Greenwood et al. in a single-blind, randomized, controlled, parallel trial randomly assigned 60 kidney transplant patients to aerobic training (n = 20), resistance training (n = 20) or usual care (n = 20). Aerobic training and resistance training were delivered 3 days per week for a 12-week period. The usual-care group received standard care. Pulse wave velocity, peak VO2, sit-to-stand 60, isometric quadriceps force and inflammatory biomarkers were assessed at 0 and 12 weeks. After 12 weeks pulse wave velocity decreased significantly in both aerobic training and resistance training groups compared with the control group [49]. As opposed to usual care, both the aerobic training and resistance training interventions demonstrated a significant improvement in average peak VO2, associated with pulse wave velocity. Toussaint et al. [74] in the HD population suggest a significant decrease in pulse wave velocity as a result of a 12-week intradialytic aerobic exercise training program. Another study by Mihaescu et al. investigated the effect of resistance training in HD patients. The group reported a decline of 1 m/s in pulse wave velocity in the exercise training group, associated with lower systolic BP The control group demonstrated an increase in pulse wave velocity of 1.3 m/s [75]. The beneficial effect of exercise program has also been demonstrated in other studies relating to endothelial function [20, 50, 76].

In continuous ambulatory peritoneal dialysis (CAPD) patients peak VO2 has been demonstrated to be low. Ulubay et al. investigated factors that influence peak VO2 in renal transplant candidates receiving CAPD therapy. Cardiopulmonary exercise tests were performed on a cycle ergometer at the same time of the day for all patients, and exercise duration, maximum work rate and peak VO2 level were analyzed. Peak VO2 was found to be correlated with serum phosphate levels and no other parameter [40]. The same group also argued that peak VO2 did not change when the peritoneal cavity was filled with solution (full status) and again when the cavity had been drained (empty status) [41].

Exercise may also have impact on BP. Henrique et al. evaluated 14 HD patients, before and after 12 weeks of aerobic exercise, performed during hemodialysis sessions. There was a significant reduction in both systolic and diastolic BP, from 151 ± 18.4 to 143 ± 14.7 mmHg and from 94 ± 10.5 to 91 ± 9.6 mmHg, respectively. Similarly, average arterial BP declined from 114 ± 13.0 to 109 ± 11.4 mmHg. This beneficial effect of exercise with regard to BP reduction was also confirmed with some [77,78,79] but not in other studies [32].

The timing of the exercise is an important parameter in HD patients. Exercise elicits immediate cardiovascular responses, with respect to which it is important to consider whether exercise training is performed “on dialysis” (i.e., simultaneously exercising and dialyzing) or “off dialysis” (i.e., exercising and dialyzing at separate times). Simultaneous fluid removal during exercise may limit the exercise tolerance. Moore et al. showed that fluid removal at a rate of 1356 mL/h during dialysis had no significant cardiovascular effects during the first 2 h. However, at 3 h, cardiac output, stroke volume and mean arterial pressure were found to be decreased, limiting exercise tolerance. The hemodynamic instability at 3 h appears to have been due to an inappropriate decrease in heart rate, as in vasodepressor syncope. In this study, however the patient number is low which precludes to investigate independent factors such as autonomic dysfunction related to these findings [37]. On the other hand, Dungey et al. [54] demonstrated that although BP falls during exercise in HD patients, the cause is not cardiac injury as no change was observed in concentrations of cTnI, myoglobin or CK-MB. It is also important to notice that no serious adverse event has been reported after around 28,000 h of intradialytic exercise [80]. One may consider intradialytic exercise safe, given careful selection of the patients.

Cardiovascular outcomes

Exercise has also been associated with undesired outcomes such as cardiovascular events. Manfredini et al. [45] demonstrated that among patients who completed the regular exercise trial, patients in the active group had poorer hospitalization-free survival than the control group. In another study, Scrutinio et al. investigated the correlation of renal function with peak VO2 consumption in heart failure patients. In total, 2.938 systolic heart failure patients underwent clinical, laboratory, echocardiography and cardiopulmonary exercise testing. The patients were then stratified according to estimated GFR. Mean follow-up was 3.7 years during which the primary outcome was a composite of cardiovascular death and urgent heart transplantation. On multivariable regression, eGFR was predictor of peak VO2 (P < 0.0001). After adjusting for significant covariates, low peak VO2 has been found to be associated with primary outcome. The strength of this association increased as eGFR decreased [81] (Table 4).

Table 4 Evaluation of cardiovascular events and death on the trials
Full size table

Quality of life, cognitive function and depression

As suggested above, regular exercise has favorable actions on quality of life, cognitive function and depression. The beneficial effect of exercise program with respect to these parameters can be explained by a number of factors including increased muscular strength [24], increased social interaction [45], release of neurotransmitters (e.g., endorphins). The direct effects of exercise on emotional and behavioral aspects range from substitution of negative thoughts and low self-esteem to decreased anxiety and improved attitude toward self. In parallel, group exercise has been shown to promote socialization with participating in a fun, organized activity during HD sessions [18, 27, 37, 82].

Mechanisms

After all these evidences, this section briefly explains the underlying mechanisms of regular exercise on these positive findings.

Resistance training highly increases the metabolism of protein synthesis, leading to increased cross-sectional volume of muscle fibers. Exercise may increase the levels of growth factors such as insulin-like growth factor-I receptor and decrease the inhibitors of muscle hypertrophy in HD patients [83]. Thus, these factors may be responsible for the beneficial effects on physical function and muscle atrophy.

During intradialytic exercise enhanced blood flow and increased capillary surface area result in a greater flux of urea and other toxins from the tissue to the vascular compartment, hence an increased removal at the dialyzer. This may alleviate symptoms of uremia [25, 26].

Another important contributor may be the presence of heart failure which is commonly seen in CKD and ESRD patients with either reduced or preserved EF [84]. CKD patients have sympathetic hyperactivity [85] which is proportional to decreased kidney function [86]. Petersson et al. demonstrated increased renal norepinephrine spillover to underlie the pathophysiological mechanism in heart failure, in parallel with and independent of cardiac sympathetic drive [87]. Despite chronic sympathetic activation, decreased responsiveness of the failing heart to catecholamines may potentially limit exercise capacity [81]. Sympathetic drive may contribute to skeletal myopathy, further decreasing exercise capacity [88, 89]. It was also demonstrated that CKD and ESRD patients have exaggerated increases in BP during handgrip exercise due, in part, to over-activation of the sympathetic nervous system [90, 91]. Last but not least, reduced nitric oxide bioavailability may also contribute to exaggerated autonomic responses during exercise with the evidence of NO-mediated inhibition of sympathetic nervous system activation under normal conditions [92, 93].

Exercise has also direct actions on myocardial function in CKD. Luiz et al. evaluated the effects of long-term aerobic swimming exercise with overload on renal and cardiac function in rats undergone 5/6 nephrectomy (5/6Nx). Eight Wistar rats were divided into 4 groups: Control (C), Control + Exercise (E), sedentary 5/6Nx (NxS) and 5/6Nx + Exercise (NxE). The rats were subjected to swimming exercise sessions with overload for 30 min 5 days per week for 5 weeks. Exercise reduced proteinuria, diminished the decline of eGFR and attenuated sclerosis index at the glomerulus. The NxS group had higher LV posterior wall in diastole and systole compared with the E group. The developed isometric tension in Lmax of the heart papillary muscle was lower in the NxS group compared with the C, E and NxE groups. Sedentary animals with nephrectomy were observed to have disrupted in myocardial contractility [94].

Chronic inflammatory state, which is present in CKD patients, can be another potential factor leading to decreased exercise capacity in these patients. Systemic inflammation may induce proteolysis in skeletal muscle leading to muscle atrophy [95, 96]. Physical exercise on the other hand has direct anti-inflammatory actions [97]. It was shown that intradialytic exercise increased the levels of IL-10, a potent anti-inflammatory cytokine. Similarly, tendency for decreased tumor necrosis alpha levels was observed [43]. Thirty minutes of aerobic exercise in pre-dialysis individuals induced a significant elevation of IL-6 and IL-10, with little effect in the TNF receptors (sTNF-RI and sTNF-RII) in post-exercise period. After 1 h from the exercise session, IL-6 and IL-10 remained in high concentrations and an increase in sTNF-RII was found. These findings support the idea of interaction between immune system and exercise [98].

Controversies and future perspectives

One of the most important issues is the valid judgment regarding the effectiveness of exercise programming. To judge the effectiveness of exercise training in CKD and HD patients, most of the available literature relies on the changes in oxygen uptake (VO2) at peak incremental work rate (IWR) exercise testing as the main laboratory-based criterion. As previously mentioned, with regard to VO2, there are contrasting findings, which were previously attributed to prescribing low-intensity [69] or short-duration aerobic training [70]. Other factors are thought to be involved as well. It is possible that training may improve several submaximal responses (e.g., work and ventilatory efficiencies, cardiovascular stress), which are not necessarily translated into higher maximal aerobic capacity measured by VO2 or 6-min walking test [44]. Thus, simpler but more sensitive tests should be developed to better evaluate the effectiveness of the exercise programs.

Secondly, although various exercise programs have been implemented, how exercise training should be performed is still a matter of debate (intradialysis vs. off dialysis, aerobic vs. anaerobic, endurance resistance or in combination in dialysis center and at home, duration, intensity, etc.) [45]. Although it was observed that exercise during dialysis is more efficient and has less dropout rates [57], other trials show the beneficial effect of home dialysis programs in CKD patients. Manfredini et al. [45] successfully implemented home exercise programs and subsequently argued improved physical function, cognitive function and social interaction.

Another conflicting issue is the type of exercise. Performed studies generally use one type of exercise program—either endurance or resistance exercise. However, a recent study showed that resistance and endurance exercises can be applied in combination even in elderly patients. Anding et al. investigated the effect of a structured physical exercise program (combination of resistance and endurance exercise) in 46 patients with HD. Combination strategy has been proven to improve muscle strength, physical activity (sit-to-stand test and 6-min walk test) physical functioning, role of physical limitations, role of emotional limitations and mental health subscales of SF-36 [28]. Thompson et al. also investigated the effect of combination of cycling and resistance exercise program on quality of life. In this factorial (2 × 2) pilot trial, 31 HD patients were randomized to cycling, resistance, cycling and resistance or an attention control groups. Participants completed the Kidney Disease Quality of Life Short Form 36 (KDQOL-SF 36), and the 6MWT was used as a measure of aerobic capacity. No significant differences between baseline and 12 weeks were found in the PCS or MCS components of the SF-36 or physical performance tests among subgroups [47]. Similarly, the existing literature does not conclude whether passive or active exercise is more important [25].

A single exercise program may not be proper for every patient as some could not be done safely nor satisfactorily by all participants due to risk of muscle injuries and adverse cardiovascular events causing a high number of dropouts [27]. Hence, it needs to be determined whether the exercise program is suitable for the patient. When necessary, the program should be adapted according to individual characteristics together with an exercise counselor providing motivational support to stimulate patients to stay more active [48]. Even the most effective exercise programs would not keep the interest of patients if composed of repetitive routines only. Therefore, careful planning of the content of the exercise programs with the motivation and continued participation of the patients in mind is essential [33, 91]. Innovative ideas such as virtual reality exercise programs have been successfully implemented to pass these barriers [33].

One should also consider making meta-analysis instead of writing a narrative review. However, as previous studies were very heterogeneous regarding the type of exercise performed (type of training and duration) and the type of assessment, we proposed that a narrative review would be better suited to describe the findings. This also brings the issue that while writing the review we could not be certain to suggest firm conclusions. Since this review involves both observational and a few interventional studies, higher quality and strength is needed to make such suggestions. So the mode of presentations both in text and in tables is arranged accordingly and not so stringent.

Conclusion

Exercise training in CKD and ESRD patients has various beneficial effects on various domains such as physical function, muscle atrophy, depression and quality of life. The mechanisms behind these beneficial effects of exercise are not fully elucidated. However, mechanisms such as increased growth factors, increased secretion of endorphins, decreased sympathetic overactivity and decreased inflammation all have been suggested. Since the area is still evolving, there are conflicting issues and unknowns. More research is needed to fully elucidate the beneficial effects of exercise and to investigate whether exercise has also impact on hard outcomes such as mortality.