Intradialytic Exercise: Effects on Arterial Stiffness and Gait Speed in Patients Undergoing Hemodialysis

Introduction

According to the most recent update on chronic kidney disease epidemiology in 2022, it affects more than 800 million individuals globally, or more than 10% of the global population [1]. For the last decade, Taiwan reported the highest incidence and prevalence rates of end-stage renal disease, resulting in a heavy healthcare burden [2–4]. Patients on chronic hemodialysis (CHD) experience worse quality of life and higher comorbidity and mortality rates than the general population [3]. Importantly, arterial stiffness (AS), a surrogate for endothelial dysfunction, has been identified as a predictor of cardiovascular mortality in patients with CHD [5].

Carotid–femoral pulse wave velocity (cfPWV) is recognized as the leading indicator of arterial stiffness and is a highly predictive tool for assessing cardiovascular diseases in patients with HD. Moreover, our prior research established that cfPWV is an independent predictor of all-cause and cardiovascular mortality in these patients [6–9].

The benefit of aerobic exercises in reducing AS and lowering cardiovascular mortality has been demonstrated in the general population [10–13]. The findings of studies investigating the effect of in-hospital/clinic (intradialytic) and out-of-hospital/clinic (interdialytic) exercise on AS in patients with CHD are controversial due to variations in the design of the exercise programs [14–21]. In fact, achieving sufficient intensity and exercise compliance are the main challenges in promoting exercise in patients with CHD. Several obstacles in implementing intradialytic exercise, such as fear of injury due to aerobic activity and lack of exercise prescription and motivation, were identified [22]. The accumulating evidence suggests that instructions and supervision provided by medical staff can ensure effective and safe intradialytic exercise [15,23]. The design of this exercise program incorporated an intensity level acceptable to participants, aiming to enhance their willingness to complete the intervention and ensure greater compatibility with real-world practice.

Intradialytic cycling exercise is a form of aerobic exercise, and its potential clinical benefits have been investigated in patients with CHD [16,23]. Given the controversial evidence regarding the effects of intradialytic exercise on vascular and physical health in patients with CHD, this study evaluated the outcomes of low-to-moderate-intensity intradialytic cycling exercise on cfPWV, gait speed, and a sit-to-stand test in patients with CHD.

Material and Methods

ETHICS STATEMENT:

The study (Clinicaltrials.gov identifier: NCT04098848) was approved by the Research Ethics Committee of Hualien Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation (approval no: IRB106-139-A). The study procedures followed the principles of the Declaration of Helsinki, and written informed consent was obtained from all participants.

PARTICIPANTS AND STUDY DESIGN:

The study participants were recruited among patients aged between 20 and 75 years who were undergoing hemodialysis 3 times a week, with 4 hours per session, for more than 3 months in Hualien Tzu Chi Hospital between March 1, 2018 and February 28, 2019. We excluded patients with lower-limb amputation, hemodynamic instability, history of deep vein thrombosis, inability to perform cycling exercise, active infections, unstable angina, unstable cardiac arrhythmia, acute myocardial infarction, severe heart failure, and stroke that occurred within 3 months before enrollment. All eligible participants provided written informed consent before study entry. The study participants in the control group received regular care and the exercise group received regular care plus intradialytic exercise. These 2 groups were matched according to age and sex at a 1: 2 ratio. Randomization was performed using a computer-generated randomization code. Due to the characteristics of the intervention, it was not possible to blind participants and clinical staff.

INTRADIALYTIC EXERCISE:

During the second hour of routine hemodialysis, participants in the exercise group performed a 5-minute warm-up exercise lying down before stepping onto the cycling ergometry (KM-300 Athlete Auto-mini Bike, Taiwan) for a 20-minute cycling session in supine position. The cycling speed gradually increased and maintained an average speed of 20–40 rpm, with an estimated power output of 50–70 watts. The exercise protocol was set at a low-to-moderate intensity, defined as activities eliciting 3 to 5.9 metabolic equivalents, based on values from the Compendium of Physical Activities. Subsequently, the participants engaged in a 5-minute cool-down period, during which they cycled slowly. The total time for each exercise session was 30 minutes. The patients performed 3 sessions per week during the study period. The control group received regular care without intradialytic exercise.

SAFETY MONITORING:

Blood pressure, heart rate, respiration rate, and blood oxygen concentration were monitored before, 10 minutes after, and at the end of the exercise session. The exercise was stopped at any point during the session if the participant experienced intolerance or requested to stop, and all events were recorded.

MEASUREMENTS OF PRIMARY ENDPOINT AND OTHER CARDIOVASCULAR PARAMETERS:

The primary endpoint of this study was arterial function. The participants rested supine for at least 10 minutes before any measurement. cfPWV was determined from the time taken for the arterial pulse to propagate from the carotid to the femoral artery. cfPWV was calculated using the elapsed time and the distance of the point of maximum upslope between the 2 recording points. An electrocardiogram was simultaneously performed as a timing reference for the R-wave signal. The quality and consistency of the measurements were confirmed using the integral software that calculated the mean time interval between the R and pulse waves during each heartbeat within 10 consecutive cardiac cycles. Aortic augmentation index (AIx) was measured using an applanation tonometer with the SphygmoCor software (AtCor Medical, Sydney, Australia) according to the following formula [13] aortic Aix = (aortic systolic pressure – inflection pressure) / pulse pressure [22]. All these parameters were measured before hemodialysis on the test day. Systolic and diastolic blood pressures and pulse rate were measured 3 times via the brachial artery of the arm without the arteriovenous shunt using an automatic oscillometer placed on the upper arm following a resting period of 10 minutes after the exercise session.

MEASUREMENTS OF BLOOD CHEMISTRY:

Fasting blood samples were collected to measure blood urea nitrogen and serum levels of creatinine, albumin, uric acid, calcium, phosphorus, total cholesterol, triglycerides, fasting glucose, and hemoglobin. All measurements were performed using a COBAS Integra 800 autoanalyzer (Roche Diagnostics, Basel, Switzerland). Additionally, dialysis efficiency (Kt/V) and normalized protein catabolic rate (nPCR) were calculated.

MEASUREMENTS OF BODY COMPOSITION AND PHYSICAL PERFORMANCE:

The secondary outcomes of this study were 1) changes in body composition, including thigh and calf circumferences, fat mass, and lean mass, and 2) changes in physical performance, including gait speed and the 5-times sit-to-stand test. These tests were conducted before hemodialysis sessions and supervised by the same skilled staff. Gait speed was measured with the participant walking a 6-meter distance, and the result of the 5-times sit-to-stand test was measured by timing how quickly the participant could stand up from a sitting position 5 consecutive times. The thigh and calf circumferences were measured 20 cm distal to the iliac crest and 10 cm distal to the patella, respectively. Bioelectrical impedance measurements of fat and lean mass were conducted using a tetrapolar whole-body technique with a single-frequency (50 kHz) analyzer (Biodynamic-450; Biodynamics Corporation, Seattle, WA, USA) [24].

DATA COLLECTION:

The baseline characteristics of the study patients were collected. Data on these parameters were collected at baseline, 6 months, and 12 months in both study groups. In the exercise group, the total number of intradialytic exercise sessions was assessed at the 6^th and 12^th-month timepoints. Participants who completed fewer than two-thirds of the required sessions were excluded from the study.

SAMPLE SIZE CALCULATION:

The sample size estimation aimed to detect significant differences in arterial function, body composition, and physical performance resulting from intradialytic cycling. The power calculation was based on previous studies, indicating that enrolling 144 patients (matched by age and sex in a 1: 2 ratio for the control and exercise groups) would provide 80% power to achieve statistical significance, assuming a type I error rate of 0.05. However, due to limitations in patient willingness, a total of 114 patients were ultimately recruited for this clinical study [19,20].

STATISTICAL ANALYSIS:

Categorical data are presented as frequency and percentages and were compared using the chi-squared test. The Kolmogorov-Smirnov test was used to check the distribution of the continuous variables. Normally distributed data are expressed as means±standard deviation and were compared using Student’s t test, while non-normally distributed data are expressed as medians with interquartile ranges and were compared using the Mann-Whitney U test. Between-group differences in primary and secondary endpoints at specific timepoints (baseline, 6 and 12 months) were analyzed using the independent-sample t test, and within-group comparisons were performed using the paired t test. Generalized estimating equation (GEE) analysis was used with groups (control and exercise), timepoints (baseline and 6 and 12 months), and the interactions between the groups and the timepoints, to test the effects of groups, study duration, and the group–time interaction on cfPWV and gait speed. The correlation structure generated the smallest Akaike’s information criterion and the quasi-information criterion. Since the history of hypertension and the duration of dialysis were significantly different between the exercise and control groups at baseline, these 2 covariates, the baseline cfPWV, and the baseline gait speed were adjusted. SPSS version 22.0 (SPSS, Chicago, IL, USA) was utilized for all data analyses. A p value of 0.05 was considered statistically significant.

Results

ENROLLMENT OF PATIENTS:

A total of 256 patients were recruited, and 114 patients who agreed to participate in the study were randomized to the control (n=39) and exercise (n=75) at a 1: 2 ratio. In the overall cohort, 89 and 82 patients completed the study evaluations after 6 and 12 months of hemodialysis sessions, respectively. During the study period, 7 and 8 patients in the exercise and control groups refused baseline evaluations. Additionally, 5 patients in the exercise group were excluded because they failed to meet the requirements of the exercise program. The causes of death were infection (n=2 per group), CAD (n=1 per group), and cancer (n=1 in the exercise group), which were not related to the study intervention (Figure 1). The proportion of patients who completed the study and evaluations in the exercise group (76%) was comparable to that of the control group (64%).

PATIENT CHARACTERISTICS:

The demographic and baseline characteristics, except for the duration of dialysis and history of hypertension, were comparable between the exercise and control groups; the dialysis duration was longer and the number of patients with a history of hypertension was significantly higher in the exercise group than in the control group (Table 1). No significant complications, such as chest discomfort or dyspnea during exercise, were detected throughout the exercise program.

PARAMETERS OF VASCULAR FUNCTION:

Table 2 shows the changes in the parameters of vascular function, including cfPWV, aortic AIx, systolic and diastolic blood pressures, and pulse rate over time in both groups. Between-group comparisons revealed no significant differences in any of the vascular parameters at baseline. cfPWV, which exhibited a decreasing trend over the 12-month study period in the exercise group, increased during the same period in the control group (Figure 2). After adjusting for factors that were significantly associated with AS by multivariate linear regression analysis (Table 3), the decrease in cfPWV was significantly more pronounced in the exercise group than in the control group (−4.31 n/s, 95% CI, −5.71 to −2.90; p<0.001). Improvement in cfPWV was only observed with 12 months of intervention in the exercise group, as shown in Figure 2. Further evaluation with the GEE analysis to determine the effect of exercise and duration of exercise on cfPWV revealed significant group and study duration effects and group–time interactions at 6 and 12 months after the study initiation. The group–time effects of exercise on cfPWV after 6 and 12 months after intervention were β=−4.11 (p<0.001) and β=−4.98 (p<0.001), respectively. The group effect of exercise on cfPWV was β=1.35 (p=0.023) and the time effects of exercise on cfPWV 6 and 12 months after intervention were β=2.70 (p<0.001) and β=2.80 (p<0.001), respectively (Table 4). Conversely, aortic AIx, systolic and diastolic blood pressures, and pulse rate were not significantly different between the groups at the end of the study period (Table 2).

BODY COMPOSITION AND PHYSICAL FUNCTION:

The comparisons of the secondary study outcomes on physical performance between the 2 groups are summarized in Table 5. Briefly, gait speed was significantly improved over the course of 12 months in the exercise group. After adjustment by multivariate linear regression analysis (Table 6), gait speed was significantly faster in the exercise group than in the control group (0.17, 95% CI 0.03–0.30; p=0.019). However, the difference in the group–time interaction for gait speed observed between the 2 groups was not present with further evaluation using GEE analysis (p=0.134 and 0.435 at 6 and 12 months, respectively) (Table 7). This finding indicated that exercise training significantly affected gait speed compared to the control group after 12 months of training; however, gait speed did not improve significantly over time in our study (Figure 3). The 5-times sit-to-stand test results and body composition parameters were not significantly different between the 2 groups at all timepoints.

Discussion

In the present study, we found that intradialytic cycling exercise more than 6 months significantly reduced AS, as measured by cfPWV, a noninvasive method to determine central AS. The observed benefit of cycling exercise was even more pronounced in patients who performed the exercise for 12 months compared to the control group including patients who did not exercise. In addition, the physical performance of the patients, such as gait speed, increased after participation in the cycling exercise program used in the study. However, the body composition did not significantly differ between the groups at the end of the 12-month exercise period. To our knowledge, this is the first study with the longest observation period to demonstrate the beneficial effects of cycling exercise on AS in patients undergoing CHD, even at low-to-moderate intensity.

Studies in the general population reveal that aerobic exercise reduces the degree of AS and decreases the risk of cardiovascular disease and mortality through vasodilatation and improved vascular perfusion [10–12,19,25]. In patients with CHD, endothelial damage due to older age, anemia, uremic toxin accumulation, mineral abnormalities, oxidative stress, inflammation, increased advanced glycation end-products, and reduced nitric oxide levels are responsible for the rapid deterioration in vascular function and the development of AS [26–28]. Furthermore, in patients with CHD, sedentary lifestyle with reduced physical activity impairs glucose control and enhances insulin resistance [29]. Exercise can reverse these risks and lessen endothelial damage [30]. Several studies in patients with CHD have found that exercise can induce endothelial nitric oxide production, which causes coronary artery vasodilation and improves perfusion and cardiac output [18,31–33]. Endurance exercise can also induce an adaptive response, leading to increased vascular density and vasodilatory capacity with further improvement of perfusion [25].

A meta-analysis evaluating the impact of intradialytic exercise on cfPWV has revealed discrepancies and heterogeneity among the studies [19]. Two of the evaluated 4 studies were randomized trials that demonstrated the benefits of exercise on arteries in patients with hemodialysis despite the limited number of patients and shorter duration of exercise [20,34]. In the randomized crossover study with a total of 18 participants, conducted by Toussaint et al [20], the observation period for the cycling exercise was approximately 30 minutes per hemodialysis session, spanning only 3 months. Conversely, the prospective observational study by Mihaescu et al [34] enlisted 16 patients with CHD who underwent intradialytic exercise for up to 40 minutes per session, with an observational period of only 3 months. Another randomized controlled trial including 32 patients with hemodialysis found a significant reduction in cfPWV in the intradialytic exercise group, with a relatively short exercise period of 4 months [18]. Importantly, our study included a larger number of patients with CHD and a longer observation period of intradialytic exercise, providing evidence on the beneficial effects of cycling exercise on AS. In contrast, the remaining 2 studies included in the meta-analysis failed to demonstrate the beneficial effect of intradialytic exercise. Specifically, Chan et al [21] adopted intradialytic progressive resistance training rather than aerobic cycling exercise whereas Koh et al [14] evaluated lower-intensity exercise. The disparate findings among the studies might be due to the limited number of patients and the differences in the intensity and duration of exercise.

Specifically, findings regarding the effects of intradialytic cycling exercise on cardiovascular outcomes in this patient population are inconsistent. A recent meta-analysis [35] investigated the effects of intradialytic cycling exercise on arterial resistance. Two of the evaluated studies [36,37] reported significantly beneficial effects on PWV, while the other 2 studies [14,20] reported a null effect. Also, 2 studies [14,37] reported significantly beneficial effects on AIx, while one study [20] reported no effect. The results from the meta-analysis of these studies demonstrated a statistically significant mean difference in PWV with 13–26 weeks of intradialytic cycling as compared with controls. By contrast, the results from the meta-analysis of these studies showed no statistically significant difference in AIx with intradialytic cycling as compared with controls. The observations of the present study are generally consistent with the findings derived from that meta-analysis [35].

Our assessment of the other parameters of vascular function revealed no significant differences in pulse rate, systolic and diastolic blood pressures, and aortic AIx between the 2 groups. In contrast, we observed a trend of lower pulse rate in the exercise group at the end of study. Consistently, Carter et al. [38] observed a reduction in heart rate following exercise and proposed that such adaptations after exercise were a result of decreased sympathetic response, which in turn suppressed cardiac output [39–41]. However, cfPWV represents AS, and systolic blood pressure and aortic AIx are more directly influenced by heart rate and ventricular contractility [42–45], which might explain the observed discrepancy in the impact of exercise on the evaluated markers. Therefore, the significant reduction in cfPWV, but not blood pressure, pulse rate, or aortic AIx, observed in the exercise group suggests that intradialytic exercise significantly improved AS in patients with CHD in the current study [46,47].

Our analyses show improved muscle function reflected in gait speed, with no differences in body composition, after intradialytic exercise in patients with CHD, consistent with previous studies [17,48–51]. Gait speed is a well-known method to evaluate skeletal muscle function in patients with hemodialysis [48]. A change of 0.1 m/sec in gait speed is considered a significant change that is sufficient to improve daily independent activities [49,52,53]. Previous studies have shown that muscle strength is more critical than muscle mass regarding outcomes in patients with hemodialysis [48,54]. In patients with hemodialysis, type II fibers are atrophic and accompanied by mitochondrial degeneration and loose capillary network whereas exercise training improves type II fibers and capillaries [55]. In addition, consistent and long-term exercise can increase intracellular calcium concentrations and ATP turnover alongside reactive oxygen species production, promoting mitochondrial adaptation [56,57].

Previous studies have identified several barriers that prevent the implementation of exercise in patients with hemodialysis. These include fear of complications from aerobic activity, lack of an exercise prescription, and deficiency in motivation despite receiving education and encouragement from the healthcare team. Our intradialytic cycling program was designed to overcome these barriers through close supervision during hemodialysis, education on the proper exercise program, and support from other patients and healthcare team members. Nonetheless, we acknowledge the limitations of our study, including the tolerance of higher intensity of the exercise and the limited size of the cohort. However, the current study included the largest number of participants and had the longest observation period among all similar studies conducted to date. In real-life clinical practice, improving exercise strength in patients with several comorbidities is challenging and safety concerns should remain a priority.

Conclusions

Intradialytic cycling is a beneficial and safe exercise modality that integrates regular physical activity into hemodialysis sessions. This clinical study demonstrated significant improvements in arterial stiffness following intradialytic cycling, with more pronounced benefits observed over extended exercise observation periods, even at low-to-moderate intensity. Additionally, physical performance, exemplified by improved gait speed, was enhanced in the exercise group. These findings demonstrate the efficacy of intradialytic cycling as a practical and effective intervention for enhancing vascular health and physical performance in patients receiving hemodialysis.