Background
Despite continuous progress in dialysis technology and disease management, mortality rate remains high in patients on hemodialysis. Physically inactive could contribute to mortality excess of hemodialysis patient, as hemodialysis patients are inactive compared to individuals with normal kidney function [
1‐
4] and approximately 60% of hemodialysis patients were reported to exercise less than once a week [
5]. However, routine care for hemodialysis patients does not include interventions to address this sedentary behavior.
A previous systematic review indicated a significant association between lower baseline level of physical activity and elevated risks of chronic diseases, such as cancer, diabetes, and vascular disease [
6]. In addition to baseline physical activity, an inverse association was observed between decrease in physical activity over time and the risks of adverse events in subjects with impaired glucose tolerance and chronic heart failure [
7,
8]. Hemodialysis patients showed an association between lower initial levels of physical activity and elevated mortality risk [
9‐
13]. As the levels of physical activity in hemodialysis patients are affected by not only time constraints during dialysis treatment but also by aging, physical function decline, exacerbation of dialysis-related symptoms or comorbidities, depression, and other social factors, these patients would show further reductions in physical activity over time. Therefore, the effects of yearly changes in physical activity on prognosis in hemodialysis patients should be evaluated.
To our knowledge, there have been no previously reports regarding the influence of yearly changes in physical activity on prognosis in hemodialysis patients, and it is unclear whether changes in physical activity over time are associated with mortality risk. The present study was performed to examine whether changes in physical activity affected all-cause mortality in hemodialysis patients.
Discussion
The present study was performed to examine all-cause mortality rate in a cohort of 192 hemodialysis patients. Forty (20.8%) patients died during the observation period lasting up to 7 years, with cardiovascular disease as the leading cause of death. Almost one quarter of the patients showed a decline in physical activity over time, which was significantly related to elevated mortality risk independent of patient characteristics and baseline physical activity. To our knowledge, this is the first report of an association between change in physical activity and mortality in hemodialysis patients. Our findings suggest that it is important to prevent a decline in physical activity over time in hemodialysis patients to improve their prognosis.
Only a few studies have examined changes in physical activity evaluated with an accelerometer or pedometer and mortality. Yates et al. reported that a decrease and an increase of approximately 25% in steps per day from baseline were noted in 25.2 and 24.9% of people with impaired glucose tolerance, respectively, at 1 year follow-up [
7]. In the present study, 24.5% and 26.6% of hemodialysis patients experienced a decrease and an increase of 30%, respectively, in physical activity level from baseline. Our findings are consistent with those of the previous study targeting people with chronic illness. In addition, 90 patients dropped out during baseline and follow-up surveys in the present study. If an assessment of yearly changes in physical activity levels of these patients was possible, the rate of patients who experienced a > 30% decrease in physical activity might have been higher.
There are several possible explanations for the association between decline in physical activity over time and poor prognosis in hemodialysis patients observed in the present study. First, arteriosclerotic disease is the main cause of death in patients on hemodialysis, as also indicated in this study. Previous reports indicated that engaging in physical activity or low-intensity walking exercise improved risk factors for arteriosclerotic disease in hemodialysis patients, i.e., hypertension, arterial stiffness, plasma triglyceride and cholesterol levels, cardiac autonomic system dysfunction, and reduced maximal oxygen consumption [
20‐
28]. Therefore, reductions in level of physical activity over time in such patients could lead to elevated risk of arteriosclerosis. The results of the present study suggested that changes in physical activity may be related to risk factors for cardiovascular disease. However, the associations could have been confounded by lifestyle-related factors that were either not measured or were measured inaccurately. In addition, the observational design of this study did not allow us to establish a causal link between change in physical activity and all-cause mortality rate. Second, patients whose physical activity at 12 months could not be determined due to severe condition were excluded from the analyses in the present study. However, we may have failed to exclude some patients with slight deterioration of overall condition that did not reach a level requiring medical treatment. Regardless of these possibilities, this study indicated that it is important to evaluate changes in physical activity among hemodialysis patients to predict their prognosis and to allow better disease management.
Tudor-Locke et al. reported that 5000 steps per day at baseline are a “borderline for mortality” in older adults [
18]. In this study, participants indicates superior survival in patients with > = 5000 steps per day. Previous studies indicated an association between baseline physical activity and survival in hemodialysis patients [
10,
12,
13]. Therefore, the results of the present observational study extend previous research by suggesting that baseline physical activity and changes in physical activity are both important and independent determinants of all-cause mortality in hemodialysis patients.
Byberg et al. examined the effects of changes in physical activity on prognosis in a large community dwelling population-based cohort [
29], and reported a higher mortality risk in people with a decline in physical activity over time from a high to low or moderate level, even if their baseline levels of physical activity was high. In addition, Yates et al. reported that each increase of 2000 steps per day from baseline to 12 months was significantly associated with an additional 8% difference in the cardiovascular event rate in people with impaired glucose tolerance [
7]. These previous reports indicating relations between decline in physical activity over time and poor prognosis in myocardial infarction patients, diabetes, and community dwelling populations were similar to the results of the present study [
17,
29‐
31]. As a more representative measure of physical activity, especially in hemodialysis patients, not only baseline physical activity but also change in physical activity should be evaluated. In addition, we measured physical activity with an accelerometer rather than by interview or questionnaire; determination of physical activity by interview or questionnaire has been used frequently due to its ease, but use of an accelerometer during routine daily activities is recommended in hemodialysis patients [
32], as previous studies determined physical activity of hemodialysis patients using accelerometers or pedometers [
1,
3]. The present study indicated the association between changes in physical activity measured objectively using an accelerometer and mortality in hemodialysis patients.
Despite evidence regarding the health benefits of physical activity and exercise, most hemodialysis patients are not sufficiently physically active. There are several means of increasing physical activity, including the use of a pedometer that reports the change in number of steps walked per day. Adults were reported to show an increase in physical activity by 26.9% over baseline over 18 weeks of pedometer use [
33]. Early studies also indicated that use of a pedometer motivated hemodialysis patients to increase their level of physical activity [
34,
35]. The level of physical activity can also be increased by exercise training. Frailty is a primary pathway to disability, defined as a pathological condition that results in a constellation of signs and symptoms and is characterized by high susceptibility to adverse health outcomes, impending decline in physical function, and high risk of death [
36], which is common in hemodialysis patients [
37‐
41]. Frailty in such patients is a major factor preventing the adoption of an active lifestyle. However, there is a great deal of evidence that exercise training is beneficial for older adults at high risk of frailty [
42]. Hence, it is possible that the levels of physical activity can be increased by ameliorating frailty. In fact, Chen et al. reported increases in physical activity among hemodialysis patients that participated in a low-intensity intradialytic exercise program accompanied by improvement in physical performance [
43]. Therefore, most hemodialysis patients should change from a sedentary to a non-sedentary lifestyle to improve their long-term prognosis.
This study had a number of limitations. First, in this study, the patients in
becoming less active group were more likely older, and had a greater prevalence of cerebrovascular accident/TIA and atherosclerotic heart disease, and had higher physical activity level at baseline than the others. Although we should perform subgroup analyses according to age, the prevalence of these comorbidities and baseline physical activity level in order to adjust the effects of differences in baseline characteristics between 3 groups, we abandon the analyses because of the small sample size. Instead of this, we analyzed the association between change in physical activity and mortality risks using Cox proportional hazards regression models adjusting for the differences in baseline characteristics of the study participants. And, this is one of only a few studies reported to date involving examination of objectively measured physical activity using an accelerometer or pedometer in hemodialysis patients. Further large-scale observational studies are needed. Second, we excluded patients that required assistance with walking and did not complete the evaluation of physical activity at 12 months because of adverse events or lower adherence. Therefore, the comorbidities in the participants were mild, which should be taken into consideration when generalizing our results to patients with more severe limitations. Third, we measured participant physical activity over the course of 4-day monitoring. Although previous studies recommended the use of data from at least 3- to 4-day monitoring to reliably predict total physical activity behavior over long term periods [
44,
45], our survey period was short. In addition, the lack of data that could interfere with the measurement of true physical activity (e.g., season, social factors) might have introduced bias in our assessment. Forth, one of the five participants was classified as “becoming less active” (24.5%). We could not provide an explanation as to why physical activity levels decreased in many patients after 12 months, as we did not collect information relating to factors that could limit physical activities in hemodialysis patients, such as higher age, comorbidities, physical function, body pain, exacerbation of dialysis-related symptoms, depression symptoms, adherence to exercise, and cohabiting family. Accordingly, an observational study will be necessary to further assess the causes of sedentary lifestyle among hemodialysis patients, including factors associated with changes in physical activity over time. Finally, although we showed that mortality risk was higher in patients with a decline in physical activity over time compared with the other groups, the mechanisms underlying these observations remain to be elucidated.
Acknowledgements
We thank all of the investigators and contributors to our study.