Effect of Intradialytic Versus Home-Based Aerobic Exercise Training ...

Effect of Intradialytic Versus Home-Based Aerobic Exercise Training on Physical Function and Vascular Parameters in Hemodialysis Patients: A Randomized Pilot Study Kirsten P. Koh, BHM(Hons),1 Robert G. Fassett, PhD,2,3,4,5 James E. Sharman, PhD,6 Jeff S. Coombes, PhD,2,3 and Andrew D. Williams, PhD1 Background: Hemodialysis patients show reduced physical function and greater risk of increased arterial stiffness because of hypertension, metabolic disturbances, and vascular calcification. Exercise interventions potentially could improve their vascular risk profile. Study Design: Randomized controlled pilot clinical study comparing the effects of 6 months of supervised intradialytic exercise training versus home-based exercise training or usual care on physical function and arterial stiffness in hemodialysis patients. Setting & Participants: 70 hemodialysis patients from 3 renal units. Intervention: Intradialytic-exercise patients trained 3 times/wk for 6 months on a cycle ergometer and home-based-exercise patients followed a walking program to achieve the same weekly physical activity. Usual-care patients received no specific intervention. Outcomes & Measurements: Primary outcome measures were distance traveled during a 6-minute walk test and aortic pulse wave velocity. Secondary outcome measures included augmentation index (augmentation pressure as a percentage of central pulse pressure), peripheral (brachial) and central blood pressures (measured noninvasively using radial tonometry), physical activity, and self-reported physical functioning. Measurements were made at baseline and 6 months. Results: At 6 months, there were no significant differences between changes in 6-minute walk test distance (intradialytic exercise, ⫹14%; home-based exercise, ⫹11%; usual care, ⫹5%), pulse wave velocity (intradialytic exercise, ⫺4%; home-based exercise, ⫺2%; usual care, ⫹5%), or any secondary outcome measure. Limitations: Lack of medication data limited the analysis of vascular parameters in this study. Conclusion: There were no differences between intradialytic or home-based exercise training and usual care for either physical function or vascular parameters. Am J Kidney Dis 55:88-99. © 2009 by the National Kidney Foundation, Inc. INDEX WORDS: Hemodialysis; exercise prescription; physical function; arterial health.

P

hysical function and activity are low in hemodialysis (HD) patients,1 and physical function predicts survival.2 HD exercise training studies reported positive effects on physical function, with improved outcomes.3-5 Systemic vascular dysfunction is common in HD patients and may manifest as impaired peripheral vascular function,6 decreased large-artery distensibility, and increased stiffness.7,8 Vascular parameters improve after 4 weeks of exercise training in

healthy sedentary older men and patients with coronary artery disease.9-12 Mechanisms underlying increased arterial stiffness in HD patients are not well understood. However, adverse arterial remodeling occurs, with large-artery dilation and thickening and calcification of arterial walls.13 This increases aortic pulse wave velocity (PWV), a marker of regional large-artery stiffness, and central augmentation index (AIx), a composite marker of

From the 1School of Human Life Sciences, University of Tasmania, Launceston, Tasmania; 2School of Human Movement Studies, The University of Queensland, St. Lucia, Queensland; 3Renal Research Tasmania, Launceston General Hospital, Charles St. Launceston, Tasmania; 4Renal Medicine, Royal Brisbane and Women’s Hospital; 5School of Medicine, University of Queensland, Brisbane, Queensland; and 6Menzies Research Institute, University of Tasmania, Hobart, Tasmania, Australia. Received April 21, 2009. Accepted in revised form September 14, 2009. Originally published online as doi:10.1053/j. ajkd.2009.09.025 on November 23, 2009.

Trial registration: www.anzctr.org.au; study number: ACTRN12608000247370. Address correspondence to Andrew D. Williams, PhD, School of Human Life Sciences, University of Tasmania, Locked Bag 1320, Launceston, Tasmania, 7250 Australia. E-mail: [email protected] © 2009 by the National Kidney Foundation, Inc. 0272-6386/09/5501-0014$36.00/0 doi:10.1053/j.ajkd.2009.09.025

88

American Journal of Kidney Diseases, Vol 55, No 1 (January), 2010: pp 88-99

Exercise Training in End-Stage Renal Disease

systemic arterial stiffness and left ventricular systolic loading.13 These measures are independent predictors of cardiovascular morbidity and mortality,14 particularly in HD patients.15-17 Two studies have investigated the effect of exercise training on vascular parameters in HD patients.18,19 In an uncontrolled study of 11 patients, Mustata et al18 found significant improvements in AIx after 3 months of supervised exercise training outside of dialysis. Aortic PWV was significantly decreased after 3 months of intradialytic exercise using a crossover study design in 19 HD patients.19 However, supervised and intradialytic exercise training is resource intensive, which may preclude use in clinical practice. This study compared the effects of 6 months of supervised intradialytic or unsupervised home-based exercise training and usual care on physical function, arterial stiffness, and self-reported health. We hypothesized that home-based and intradialytic exercise training would improve exercise capacity and arterial stiffness.

METHODS Participants The study was a multicenter randomized controlled trial with participants recruited from 3 Tasmanian renal units (population, 500,000). The study protocol has been published.20 Sample-size calculation indicated that 17 patients per study group would be required to detect a 10% improvement in 6-minute walk distance (␤ ⫽ 90%; ␣ ⫽ 0.05), the measure expected to be the most variable.20 Because of our assumption of variability in 6-minute walk distance not being met and the lack of availability of additional patients (all HD patients in Tasmania meeting inclusion criteria were approached), these data are presented as a pilot study. After enrollment, participants were randomly assigned by an individual not associated with the trial using unrestricted computer-generated random numbers. It was not possible to blind participants or researchers to group assignment. Patients aged ⬎ 18 years on stable adequate dialysis therapy with urea reduction ratio ⬎ 70% for ⬎ 3 months were eligible for inclusion. Patients with unstable angina, those with lower-limb amputation, or those who met or exceeded the exercise recommendation of 120 minutes of moderate intensity physical activity per week were ineligible.21 Eligible HD patients (n ⫽ 113) were approached to participate. Written informed consent was obtained. Ethical approval was gained from the Tasmanian Human Research Ethics Committee.

89 months. Participants exercised at a rating of 12-13 on the Borg Rating of Perceived Exertion (RPE) scale, which ranges from 6-20 points,22 with intensity monitored using heart rate. When training heart rates were low, at Borg RPE of 12-13, the supervisor increased the resistance to elicit a greater cardiorespiratory response. Patients were permitted to rest or request training at lower intensity. If exercise blood pressure was ⬎ 200/110 mm Hg, participants were instructed to temporarily cease exercise and were monitored until blood pressure stabilized (⬍180/110 mm Hg). Patients using medications affecting cardiac sinus rhythm were trained according to reported Borg RPE. Although participants were encouraged to progress exercise duration according to individual capabilities, a guideline directed participants to complete at least 15 minutes per exercise session in the first 2 weeks and progress to 30 minutes per exercise session by week 12 and 45 minutes by week 24. Power output (watts) and duration (minutes) of each exercise session were recorded to estimate individual energy expenditure per session. Home-based-exercise participants were asked to perform thrice-weekly unsupervised walking for 6 months at Borg RPE of 12-13. To ensure treatment similar to the intradialyticexercise group, home-based-exercise participants were requested to start and progress their walking program according to individual capabilities, and investigators encouraged participants to start at 15 min/session and progress to 45 minutes by week 24. Participants were telephoned fortnightly to provide encouragement and allow feedback. The duration of each walking session was monitored. Homebased-exercise participants were encouraged to increase intensity by walking faster or on an incline. Adherence data captured using training diaries were completed by supervisors at every intradialytic exercise training session, whereas home-based-exercise participants completed diaries individually. The number of sessions performed in 6 months was calculated as a percentage of total possible sessions. Usual-care participants were requested to maintain usual lifestyles with regular reminders. Ethical approval required investigators to provide brochures about exercise benefits to all patients regardless of their decision to participate or to which group they were allocated.

Outcome Measures Participants underwent tests for the primary (6-minute walk distance and aortic PWV) and secondary (AIx, peripheral and central blood pressures, Timed Up and Go [TUG], grip strength, physical activity, and self-reported health) outcome measures at baseline and 6 months. Details have been described previously.20 The same researcher recorded outcome measures for participants at their respective HD centers. Participants were instructed to refrain from physical activity for 24 hours and food and caffeine intake for 3 hours before each session.

Interventions

Physical Function

Intradialytic-exercise participants trained on cycle ergometers (Rehab Trainer 881E; Monark, www.monarkexercise. se) within the first 2 hours of dialysis 3 times/wk for 6

Physical function was assessed using the 6-minute walk distance, TUG, and grip strength tests, as described previously.20 The 6-minute walk distance involved participants

90 walking as far as possible in 6 minutes along a 25-m stretch of a quiet hospital corridor. TUG measured the time taken to stand from a seated position, walk 3 m, turn, walk back to a chair, and return to the seated position as quickly as possible. Grip strength of the dominant arm was measured using a handgrip dynamometer (Smedleys, TTM, www.stoeltingco. com). Patients stood comfortably with the dynamometer held above the head with a nearly straight elbow and squeezed on the lever as hard as possible while bringing the hand down to their side. The distance between the handle and the lever was set at 60 mm for men and 55 mm for women. Participants were given 2 attempts, and the highest force was recorded.

Arterial Stiffness To ensure that diurnal variations in PWV were minimized, vascular measures were obtained at the same time of day in the hour preceding dialysis. Aortic PWV was the primary outcome measure for arterial stiffness. The method involved electrocardiogram-gated sequential applanation tonometry (SPT-301 Mikro-Tip; Millar Instruments, www.millarinstruments.com) of the common carotid and femoral arteries (SphygmoCor 7.01; AtCor, www.atcormedical.com).23 This technique is highly reproducible in patients with renal disease.24 Peripheral PWV was assessed from waveforms acquired at the carotid and radial arteries. Central blood pressures and AIx were estimated using radial applanation tonometry performed on the nonfistula arm using customized software (SphygmorCor 7.01) with a validated25 and reproducible24 generalized transfer function. In patients with a fistula in both arms, the brachial artery (medial to biceps) of the dominant arm was used.20 Physical activity and perceived physical functioning were measured as previously described.26 Participants randomly assigned to exercise were requested to include any training performed in the previous week in their response, which was used to calculate weekly physical activity in metabolic equivalents ⫻ min/wk. The questionnaire has been shown to be reliable and valid.27,28 Perceived physical functioning was evaluated using the Medical Outcomes Short-Form 36-Item Health Survey (SF36).29 Widely used previously in HD populations,30,31 it includes 8 independent scales and assesses physical and mental dimensions of health.

Statistical Analyses Baseline categorical data were compared using logistic regression, and baseline continuous data, using 1-way analysis of variance. Comparison of descriptive variables for completers and noncompleters was performed using t tests. Parametric longitudinal data were compared using generalized estimating equation models corrected for repeated measures. Results are expressed as mean ⫾ standard deviation, and the mean difference (with 95% confidence intervals [CIs]) in change from baseline to end of treatment in the intradialytic-exercise and home-based-exercise groups was compared with the usual-care group. Weekly physical activity (metabolic equivalents ⫻ min/wk) was analyzed as Box-Cox–transformed values in the generalized estimating

Koh et al equation models because of skewed distribution of residual values (using Cameron and Trivedi’s information matrix test) when untransformed values were used. P values were adjusted for multiple comparisons when appropriate using the Holm test. All statistical analyses were performed using Stata 10 (StataCorp; www.stata.com).

RESULTS All eligible patients were recruited between February 2006 and June 2007. The final patient completed the intervention in January 2008. Of 113 patients approached, 70 consented and were randomly assigned to receive either intradialytic exercise (n ⫽ 27), home-based exercise (n ⫽ 21), or usual care (n ⫽ 22). Patient flow through the study, including reasons for withdrawal, is shown in Fig 1. Forty-six HD participants (29 men; mean age, 51.9 ⫾ 12.8 years; mean body mass index, 27.4 ⫾ 8.0 kg/m2; mean time on HD therapy, 30.5 ⫾ 26.6 months) completed the study (Table 1). No significant differences were found in any descriptive variable (Table 1) or medication use (Table 2) between groups at baseline. There were no significant differences in baseline descriptive variables between participants who completed the study and those who withdrew. Exercise training adherence between the intradialytic-exercise (75% ⫾ 19%) and home-basedexercise (71% ⫾ 13%) groups was similar (P ⫽ 0.9). Intradialytic-exercise participants reported vascular access stability issues, tiredness, and dialysis-related hypotension as the main reasons for not training, whereas home-based-exercise participants reported fatigue, postdialysis hypotension, heart palpitations, and work commitments as reasons for missing walking sessions. No intervention-related serious adverse events were reported. There were no significant differences in 6-minute walk distance (P ⫽ 0.6) between groups at baseline (Table 3). Compared with usual care, there were no significant changes in 6-minute walk distance (intradialytic exercise, ⫹14%, P ⫽ 0.2; home-based exercise, ⫹11%, P ⫽ 0.3; observed power, 0.41). Our data indicated that the assumptions used in our sample-size calculation20 were not met and that 40 participants would be needed per group to achieve statistical significance for a clinically significant 10% improvement in 6-minute walk distance (␤ ⫽ 80%; ␣ ⫽ 0.05).


91

Figure 1. Flow diagram shows study progress from assessment of eligibility to data analysis. Abbreviations: HB, home based; ID, intradialytic; UC, usual care.

No differences were observed between groups for TUG (P ⫽ 0.7) or grip strength (P ⫽ 0.3) at baseline. Compared with usual care, no treatment effect was observed for either TUG or grip strength during the 6-month period (Table 3). No differences in weekly physical activity (metabolic equivalents ⫻ min/wk; P ⫽ 0.9) were observed between groups at baseline. Selfreported physical activity increased in the intradialytic-exercise group (P ⫽ 0.03), but not in the home-based-exercise group (P ⫽ 0.3; Table 3). SF-36 Physical Function scores were not different between groups at baseline (P ⫽ 0.9; Table 4). Compared with usual care, physical function decreased significantly in the intradialytic-exercise group (⫺25%; P ⫽ 0.01) and was unchanged in the home-based-exercise group (⫹5%; P ⫽ 0.6; Table 4). There were no significant changes in any other SF-36 scores. There were no baseline differences between groups in central (systolic, P ⫽ 0.5; diastolic,

P ⫽ 0.6) or peripheral (systolic, P ⫽ 0.9; diastolic, P ⫽ 0.6) blood pressures, AIx (P ⫽ 0.5), AIx at heart rate of 75 beats/min (P ⫽ 0.7), or aortic (P ⫽ 0.7) or peripheral (P ⫽ 0.6) PWV (Table 5). Compared with usual care, no treatment effect was observed for either aortic (intradialytic exercise, ⫺0.8 m/s, P ⫽ 0.4; home-based exercise, ⫺0.7 m/s, P ⫽ 0.3; observed power, 0.42) or peripheral (intradialytic exercise, ⫺0.9 m/s, P ⫽ 0.2; home-based exercise, ⫺0.6 m/s, P ⫽ 0.3; observed power, 0.23) PWV or any measure of central or peripheral blood pressure after the intervention (Table 5). Data obtained in this study indicated that 36 participants would be needed per group to attain statistical significance for a clinically significant improvement of ⫺1.0 m/s in aortic PWV15 (␤ ⫽ 80%; ␣ ⫽ 0.05). When data were separated and analyzed on the basis of hypertension status at baseline, no treatment effects were observed in any measure of arterial health for those without high blood pres-

92

Koh et al Table 1. Descriptive Data for Participants at Baseline Exercise

Age (y)

Usual Care

Intradialytic

Home-Based

Total

51.3 ⫾ 14.4

52.3 ⫾ 10.9

52.1 ⫾ 13.6

51.9 ⫾ 12.8

8/8

10/5

11/4

29/17

Men/women

167 ⫾ 8

167 ⫾ 8

169 ⫾ 11

168 ⫾ 9

80.8 ⫾ 25.2

75.9 ⫾ 13.9

80.7 ⫾ 19.1

79.1 ⫾ 19.8

Body mass index (kg/m )

28.6 ⫾ 7.3

27.6 ⫾ 7.2

27.9 ⫾ 4.9

28.1 ⫾ 6.5

Systolic blood pressure at rest (mm Hg)

145 ⫾ 18

148 ⫾ 22

143 ⫾ 32

145 ⫾ 24

Diastolic blood pressure at rest (mm Hg)

80 ⫾ 9

82 ⫾ 10

78 ⫾ 16

80 ⫾ 12

11 2 9 2 3 0 2 1 0 0

10 5 5 1 3 0 0 1 1 1

8 3 5 2 7 2 3 1 0 1

29 10 19 5 13 2 5 3 1 2

25.8 ⫾ 22.2

32.1 ⫾ 26.7

37.0 ⫾ 31.1

32.0 ⫾ 27.8

Height (cm) Weight (kg) 2

Comorbidities (%) Hypertension Essential hypertension Isolated systolic hypertension Diabetes Other cardiovascular disease Prior myocardial infarction Cardiac procedure Angina Mitral regurgitation Other arterial disease Length of end-stage renal disease (mo)

Note: Data presented as mean ⫾ standard deviation. All P values are nonsignificant.

sure or those with either essential or isolated systolic hypertension.

DISCUSSION Four major findings emerged from this study. First, neither form of training significantly improved 6-minute walk distance compared with usual care. Second, physical activity increased in the intradialytic-exercise group, but not in the home-based-exercise group. Third, self-reported Table 2. Participants’ Prescribed Medication at Baseline Exercise Usual Care Intradialytic Home-Based

No. of patients Antihypertensives Diuretic Hypolipidemic agents Anticoagulants Bone and calcium agents Hematopoietic agents

16 6 4 3 4 10 12

15 8 2 3 2 14 14

Note: All P values are nonsignificant.

15 6 3 5 3 9 10

physical function decreased significantly with intradialytic-exercise training. Finally, there were no improvements in vascular parameters. 6-Minute walk distance improved by similar amounts after both the intradialytic-exercise and home-based-exercise protocols, although this was not statistically significant. However, given that low exercise capacity measured using peak oxygen consumption has been identified as a powerful predictor of mortality in patients with endstage renal disease (ESRD),2 the observed changes in 6-minute walk distance may be clinically significant. A larger study may be required to achieve significance. Previously,4 peak oxygen consumption has been reported to increase after 1 year of thrice-weekly intradialytic exercise training on stationary bicycles compared with a smaller, yet still significant, improvement after an outpatient program. In this previous study, the outpatient exercise program was supervised and included a combination of calisthenics, stepping, and flexibility, making it resource intensive and limiting the comparisons between it and


93

Table 3. Effects of Home-Based or Intradialytic Exercise Training and Usual Care on Physical Function and Weekly Physical Activity Difference in Change vs Usual Care No. of Patients

Baseline

End

Mean (95% confidence interval)a

Pb

0 (reference) ⫹42.3 (⫺6.5 to 91.0) ⫹27.6 (⫺21.2 to 76.3)

— 0.2 0.3

0 (reference) ⫺0.3 (⫺1.39 to 0.77) ⫹0.3 (⫺0.84 to 1.36)

— 0.9 0.6

0 (reference) ⫺2 (⫺5.6 to 2.0) ⫺2 (⫺4.3 to 3.4)

— 0.7 0.8

6-Minute Walk (m) Usual care Intradialytic Home based

16 14 14

431 ⫾ 160 463 ⫾ 127 444 ⫾ 127

Usual care Intradialytic Home based

14 15 14

6.3 ⫾ 2.5 5.8 ⫾ 1.5 5.7 ⫾ 2.0


14 15 14

28 ⫾ 13 34 ⫾ 10 36 ⫾ 15

452 ⫾ 144 526 ⫾ 97 493 ⫾ 143 Timed Up and Go (s) 6.1 ⫾ 1.5 5.3 ⫾ 1.5 5.8 ⫾ 2.1 Grip Strength (kg) 31 ⫾ 12 35 ⫾ 11 37 ⫾ 14

Weekly Physical Activity (metabolic equivalents ⴛ min/wk)c Usual care Intradialytic Home based

15 15 15

692 ⫾ 771 528 ⫾ 795 848 ⫾ 1,470

943 ⫾ 1,701 1,920 ⫾ 3,273 1,712 ⫾ 3,868

0 (reference) ⫹6.42 (1.47 to 28.08) ⫹1.84 (0.69 to 5.48)

— 0.03 0.3

Note: Baseline and end values given as mean ⫾ standard deviation. The mean difference in change in exercise levels between baseline and end of treatment of the inpatient and home care groups was estimated using generalized estimating equation models, corrected for repeated measures. b P values were corrected for multiple comparisons using the Holm method. c Weekly physical activity was expressed as untransformed values for mean and standard deviation, and as a Box-Cox transformation in the generalized estimating equation models due to skewed distribution of residual values when untransformed values were used. a

the home-based exercise training used in our study. In contrast, true comparisons of supervised versus unsupervised training have reported similar between-group improvements in exercise capacity in coronary artery bypass graft patients,32 whereas a recent review33 indicates that supervised exercise results in greater improvements in walking distance than unsupervised exercise in patients with peripheral arterial disease. A potential reason for the nonsignificant improvements observed in 6-minute walk distance may involve the volume or intensity of activity undertaken. Participants were requested to exercise at an intensity of 12-13 on the 6-20–point Borg RPE scale.22 Whereas intradialytic-exercise participants were provided with feedback for exercise heart rate and blood pressure to ensure that appropriate training intensities were maintained, home-based-exercise participants trained with no supervision. Because Borg RPE correlates with exercise heart rate through a

range of relative exercise intensities,22 we believed that exercise intensity prescribed on the basis of the Borg RPE would be suitable for all patients. It is possible that fatigue and lack of motivation resulted in increased ratings of Borg RPE in both training groups. Although this would have been reflected in blunted training heart rates and blood pressures, which were not observed in the intradialytic-exercise group, a similar effect in the home-based-exercise group would have gone unobserved. This theory may be supported partially by the changes in self-reported physical activity. Although the increase in self-reported physical activity in the intradialytic-exercise group during the intervention period was significant, there was no statistically significant change in the home-based-exercise group (Table 3), indicating that the volume or intensity of training using home-based exercise may not have met the requested levels or that home-based-exercise participants reduced their involvement in nonprescribed physical activity during the intervention

94

Koh et al Table 4. Effects of Home-Based or Intradialytic-Exercise Training and Usual Care on Self-Reported Health Difference in Change vs Usual Care No. of Patients

Baseline


15 15 15

63 ⫾ 34 68 ⫾ 24 66 ⫾ 23


15 15 15

60 ⫾ 58 38 ⫾ 43 48 ⫾ 42

End


Pb

0 (reference) ⫺17 (⫺27.8 to ⫺5.5) ⫹3 (⫺7.8 to 14.5)

0.01 0.6

0 (reference) ⫹5 (⫺27.1 to 37.1) ⫹7 (⫺25.5 to 38.8)

0.8 0.9

0 (reference) ⫺11 (⫺34.6 to 12.5) ⫹19 (⫺4.7 to 42.3)

0.4 0.2

0 (reference) 0 (⫺14.3 to 14.0) ⫹7 (⫺7.6 to 20.8)

0.9 0.7

0 (reference) ⫹2 (⫺10.0 to 13.4) ⫹7 (⫺4.4 to 19.0)

0.8 0.4

0 (reference) ⫹1 (⫺16.6 to 18.3) 0 (⫺17.6 to 17.3)

0.9 0.9

0 (reference) ⫹4 (⫺22.4 to 31.1) ⫹24 (⫺2.4 to 51.1)

0.8 0.1

0 (reference) ⫺8 (⫺17.2 to 1.2) ⫹1 (⫺8.4 to 10.0)

0.2 0.9

0 (reference) ⫺7 (⫺19.4 to 5.8) ⫹6 (⫺6.8 to 18.5)

0.6 0.4

0 (reference) 0 (⫺9.4 to 9.2) ⫹8 (⫺1.4 to 17.2)

0.9 0.2

Physical Function 70 ⫾ 26 58 ⫾ 23 77 ⫾ 24 Role–Physical 48 ⫾ 44 31 ⫾ 42 43 ⫾ 38 Bodily Pain Usual care Intradialytic Home based

15 15 15

59 ⫾ 49 71 ⫾ 27 50 ⫾ 29


15 15 15

51 ⫾ 32 39 ⫾ 20 38 ⫾ 25


15 15 15

56 ⫾ 25 51 ⫾ 26 50 ⫾ 22


15 15 15

82 ⫾ 27 74 ⫾ 25 79 ⫾ 23

57 ⫾ 31 58 ⫾ 27 67 ⫾ 33 General Health 48 ⫾ 27 36 ⫾ 21 42 ⫾ 26 Vitality 52 ⫾ 23 49 ⫾ 21 53 ⫾ 26

Social Function 73 ⫾ 30 67 ⫾ 23 70 ⫾ 32

Role–Emotional Usual care Intradialytic Home based

15 15 15

69 ⫾ 43 60 ⫾ 48 60 ⫾ 46


15 15 15

73 ⫾ 25 77 ⫾ 19 71 ⫾ 19


15 15 15

55 ⫾ 29 53 ⫾ 22 50 ⫾ 23


15 15 15

66 ⫾ 26 60 ⫾ 22 59 ⫾ 22

69 ⫾ 41 64 ⫾ 46 84 ⫾ 35 Mental Health 77 ⫾ 16 73 ⫾ 16 76 ⫾ 16

Physical Component Score 55 ⫾ 25 47 ⫾ 20 56 ⫾ 25

Mental Component Score 64 ⫾ 25 58 ⫾ 20 65 ⫾ 22

Note: Baseline and end values given as mean ⫾ standard deviation. The mean difference in change in exercise levels between baseline and end of treatment of the inpatient and home care groups was estimated using generalized estimating equation models, corrected for repeated measures. b P values were corrected for multiple comparisons using the Holm method. a


period. Although this provides a potential explanation for the lack of increase in the home-basedexercise group, it also must be acknowledged that the lack of change may be caused by the mean increase in physical activity observed in the usual-care group, as well as the large variability in data. No significant changes were observed in TUG or grip strength in any group after the intervention. Patients with ESRD previously have been reported to have low functional mobility (median TUG time, 12.21 seconds; 25th-75th percentiles, 8.7-13.6).34 In comparison, baseline TUG times observed in the 3 groups in the present study (Table 3) were no different from those previously reported for healthy similarly aged (50-59 years) individuals.35 The lack of improvement in TUG therefore could be caused by limited room for improvement in our “high-functioning” group in this measure. However, it is possible that the lack of improvement in exercising patients also may have occurred because there was no functional strength training component in the exercise intervention. The lack of improvement in grip strength in either treatment group likely is caused by the failure to include upper-limb exercises in either training protocol. In contrast to changes observed in exercise capacity and physical activity, intradialytic exercise training resulted in significant decreases in self-reported physical functioning (Table 4). This decrease in the intradialytic-exercise group is difficult to explain. It is possible that the intradialytic nature of the exercise caused intradialyticexercise participants to associate the exercise with their ongoing treatment and therefore, although they experienced improvements in exercise capacity and physical activity, quality of life, including self-reported physical function, was influenced negatively by exercise being associated with the dialysis treatment. Compared with usual care, there were no significant changes in PWV or AIx with either intervention, indicating that aerobic exercise training at the intensity, duration, and frequency prescribed may not benefit aortic stiffness or factors that decrease AIx (such as improved peripheral vasodilation) in patients with ESRD. These findings contrast with those of previous reports examining the effect of exercise training on arterial stiffness in patients with ESRD.18,19 Mustata et

95

al18 reported that AIx significantly improved in 11 HD patients who attended 3 months of a twice-weekly outpatient aerobic exercise program. The decrease in AIx paralleled that of brachial pulse pressure, suggesting that largeartery compliance may have been improved in response to training, although this is only speculation because regional or local arterial measures were not acquired.18 In contrast, we found no significant changes in AIx or pulse pressure in either treatment group during the intervention period. The reason for the contrasting findings is uncertain, but may relate to differences in initial blood pressures. Nineteen of our 46 patients had isolated systolic hypertension, and endurance training does not affect arterial compliance in older participants with this condition.36,37 When patients with isolated systolic hypertension were removed from this analysis, there were no significant changes in AIx in the intradialytic-exercise (⫹4%; 95% CI, ⫺7.7 to 16.1; P ⫽ 0.9) or home-based-exercise group (⫹2%; 95% CI, ⫺9.8 to 13.3; P ⫽ 0.8). Although this may be caused by the smaller sample (n ⫽ 27) remaining in the analysis, this is unlikely because of the mean increase in AIx in both treatment groups and the range of within-group changes. Patients enrolled in the study of Mustata et al18 may not have been hypertensive, although it is difficult to be certain because they reported only pulse pressure and did not measure PWV, thereby precluding a comparison to our primary outcome measure of arterial stiffness. In a recent crossover study, Toussaint et al19 compared 3 months of intradialytic-exercise training with 3 months of usual care. Aortic PWV was found to decrease significantly. A method difference compared with our study was the exercise intensity prescribed. Our study guided patients to train at 12-13 on the Borg RPE22 scale. In contrast, patients of Toussaint et al19 determined their own levels of exertion and were not formally supervised. Although absolute intensity was not reported by Toussaint et al,19 average work performed per exercise session was ⬃70 kcal. This compares with ⬃35 kcal of work performed per session at the end of our protocol. Consequently, participants in the study by Toussaint et al19 worked at a higher intensity or longer duration than those in our study, which may explain the disparate results.

96

Koh et al Table 5. Effects of Home-Based or Intradialytic-Exercise Training and Usual Care on Cardiovascular Parameters Difference in Change vs Usual Care No. of Patients

Baseline


16 15 15

74 ⫾ 10 70 ⫾ 13 73 ⫾ 9


16 15 15

End


Pb

Heart Rate (beats/min) 75 ⫾ 12 69 ⫾ 11 71 ⫾ 10

0 (reference) ⫺2 (⫺9.1 to 5.7) ⫺3 (⫺10.2 to 4.5)

0.7 0.9

Peripheral Systolic Blood Pressure (mm Hg) 145 ⫾ 18 148 ⫾ 22 144 ⫾ 32

136 ⫾ 29 139 ⫾ 22 142 ⫾ 29

0 (reference) 0 (⫺17.1 to 17.7) 8 (⫺9.7 to 25.1)

0.9 0.8

Peripheral Diastolic Blood Pressure (mm Hg) Usual care Intradialytic Home based

16 15 15

80 ⫾ 9 83 ⫾ 10 78 ⫾ 16


16 15 15

65 ⫾ 18 65 ⫾ 20 66 ⫾ 20


16 15 15

129 ⫾ 18 137 ⫾ 23 129 ⫾ 31


16 15 15

75 ⫾ 15 77 ⫾ 10 79 ⫾ 16

0 (reference) ⫺1 (⫺9.6 to 8.6) 6 (⫺2.8 to 15.3)

0.9 0.4

0 (reference) 1 (⫺9.9 to 11.7) 2 (⫺9.3 to 12.3)

0.9 0.9

0 (reference) ⫺1 (⫺17.7 to 16.4) 7 (⫺10.1 to 24.0)

0.9 0.9

0 (reference) ⫺1 (⫺10.6 to 8.1) 6 (⫺3.7 to 15.0)

0.8 0.5

0 (reference) 0 (⫺10.3 to 9.3) 1 (⫺9.2 to 10.4)

0.9 0.9

0 (reference) ⫺1 (⫺13.9 to 11.6) 6 (⫺7.2 to 18.2)

0.9 0.8

0 (reference) 0 (⫺23.1 to 22.3) ⫺4 (⫺26.5 to 19.0)

0.9 0.9

0 (reference) 2 (⫺12.2 to 16.6) 8 (⫺6.5 to 21.9)

0.8 0.6

0 (reference) ⫺0.01 (⫺0.14 to 0.12) 0 (⫺0.13 to 0.14)

0.9 0.9

Peripheral Pulse Pressure (mm Hg) 61 ⫾ 19 62 ⫾ 20 63 ⫾ 18

Central Systolic Blood Pressure (mm Hg) 122 ⫾ 27 129 ⫾ 22 129 ⫾ 30

Central Diastolic Blood Pressure (mm Hg) 81 ⫾ 10 85 ⫾ 11 80 ⫾ 16

77 ⫾ 16 78 ⫾ 11 81 ⫾ 17

Central Pulse Pressure (mm Hg) Usual care Intradialytic Home based

16 15 15

48 ⫾ 14 52 ⫾ 21 49 ⫾ 19


16 15 15

102 ⫾ 13 106 ⫾ 14 101 ⫾ 21


16 15 15

322 ⫾ 21 333 ⫾ 31 324 ⫾ 19


15 14 15

138 ⫾ 13 139 ⫾ 8 141 ⫾ 16

46 ⫾ 15 50 ⫾ 20 49 ⫾ 19

Mean Arterial Pressure (mm Hg) 96 ⫾ 21 100 ⫾ 13 101 ⫾ 21 Ejection Duration (ms) 324 ⫾ 38 334 ⫾ 32 322 ⫾ 41 Time to Reflection (ms) 140 ⫾ 17 143 ⫾ 15 150 ⫾ 26

Pulse Pressure Amplification Usual care Intradialytic Home based

16 15 15

1.37 ⫾ 0.21 1.31 ⫾ 0.21 1.38 ⫾ 0.23

1.34 ⫾ 0.23 1.27 ⫾ 0.16 1.35 ⫾ 0.21 (Continued)


97

Table 5 (Cont’d). Effects of Home-Based or Intradialytic-Exercise Training and Usual Care on Cardiovascular Parameters Difference in Change vs Usual Care No. of Patients

Baseline


16 15 15

36 ⫾ 10 37 ⫾ 12 37 ⫾ 11


16 15 15

11.3 ⫾ 9.9 15.5 ⫾ 10.6 12.2 ⫾ 11.6

End


Pb

P1 Height (mm Hg) 34 ⫾ 10 36 ⫾ 13 36 ⫾ 10

0 (reference) 1 (⫺5.0 to 7.5) 2 (⫺4.7 to 7.8)

0.7 0.9

0 (reference) ⫺1.3 (⫺6.75 to 4.07) ⫺0.8 (⫺6.22 to 4.60)

0.9 0.8

0 (reference) ⫺1 (⫺10.2 to 8.1) ⫺3 (⫺11.8 to 6.5)

0.8 0.9

Augmentation (mm Hg) 12.2 ⫾ 8.7 15.0 ⫾ 9.8 12.3 ⫾ 11.9

Augmentation Index (%) 22 ⫾ 17 27 ⫾ 12 21 ⫾ 19


16 15 15


16 15 15

22 ⫾ 17 24 ⫾ 11 20 ⫾ 18


15 13 14

8.7 ⫾ 2.5 9.1 ⫾ 2.8 9.7 ⫾ 3.2


16 15 15

8.0 ⫾ 2.2 8.6 ⫾ 1.2 8.3 ⫾ 1.8

24 ⫾ 17 28 ⫾ 11 19 ⫾ 17

Augmentation Index at Heart Rate of 75 beats/min (%) 24 ⫾ 17 25 ⫾ 10 19 ⫾ 17

0 (reference) ⫺1 (⫺9.5 to 6.6) ⫺4 (⫺11.8 to 4.3)

0.7 0.7

0 (reference) ⫺0.8 (⫺2.11 to 0.48) ⫺0.7 (⫺1.92 to 0.62)

0.4 0.3

0 (reference) ⫺0.9 (⫺2.02 to 0.18) ⫺0.6 (⫺1.68 to 0.52)

0.2 0.3

Pulse Wave Velocity Aortic (m/s) 9.2 ⫾ 3.5 8.8 ⫾ 2.9 9.5 ⫾ 3.4

Pulse Wave Velocity Peripheral (m/s) 8.7 ⫾ 1.8 8.3 ⫾ 1.1 8.4 ⫾ 1.7

Note: Baseline and end values given as mean ⫾ standard deviation. P1 height denotes central pressure at systolic peak minus diastolic pressure. a The mean difference in change in exercise levels between baseline and end of treatment of the inpatient and home care groups was estimated using generalized estimating equation models, corrected for repeated measures. b P values were corrected for multiple comparisons using the Holm method.

The initial degree of arterial stiffness may have contributed to the lack of change in vascular parameters after either exercise intervention. Adult patients with ESRD have aortic PWV that tends to be higher than in healthy age-, sex-, and blood pressure–matched controls.17 Although the patient group in the previous study19 to examine the effect of exercise training on PWV in patients with ESRD had baseline values similar to those previously reported in patients with ESRD,17 baseline values of participants in our study (8.27 ⫾ 1.74 m/s) were closer to those observed in healthy controls.17 It therefore is tempting to speculate that the lack of improvement in vascular parameters in our study was caused by a smaller margin for improvement. However, this is unlikely be-

cause studies of healthy participants with aortic PWV similar to those reported in our study have reported decreases after as few as 8 weeks of training at moderate intensities.38,39 Whereas arterial stiffness in healthy populations has been associated with age and activity levels,11 arterial stiffness in patients with ESRD is accelerated by metabolic imbalances, insulin resistance, and hypervolemia.16 Several limitations of this study include the failure of our sample-size calculation assumptions to be met. The assumed variability derived from other authors30 was lower than the variability observed in our population. Retrospective power analyses of our data indicate that this resulted in 41% and 42% power for the primary

98

Koh et al

outcome measures. A possible explanation for the greater variability involves mean improvements in 6-minute walk distance in the usualcare group, which may have been caused by their 36% increase in physical activity during the trial period. This may have been influenced by the ethics committee’s requirement that all participants receive material highlighting the benefits of exercise. Other limitations affecting our ability to record improvements in 6-minute walk distance include learning effects from repeated trials, the possibility that the 6-minute walk distance test may not be a good indicator of changes in physical function in patients with ESRD,40 and the high average activity levels of patients entering the study. Although we gave participants a practice attempt before each testing session, it is possible that improvements in walk distance in all groups were caused in part by better pacing strategies. Another limitation was that although baseline medications were no different between groups, changes in doses were not recorded, potentially confounding observed results. The proportion of diabetic patients with ESRD was underrepresented in this study despite all dialysis patients being offered entry. Diabetic patients were often excluded because of unstable angina and/or lower-limb amputation. The results indicate that both intradialyticexercise and home-based-exercise training are likely to improve physical function, measured using the 6-minute walk distance. However, this did not translate to improvements in patients’ self-reported physical function or measurements of arterial health. Further research with a larger sample and longer duration is needed to determine the best approach to exercise prescription for HD patients.

ACKNOWLEDGEMENTS The authors wish to acknowledge the assistance of Dr Matthew Jose, Ms Marianne Smith and Ms Lisa Anderson for their assistance with patient recruitment, and Drs Iain Robertson and Kiran Ahuja for their assistance with statistical analyses. Support: This project was funded by the Clifford Craig Medical Research Trust. Financial Disclosure: None.

REFERENCES 1. Johansen KL, Chertow GM, Ng AV, et al. Physical activity levels in patients on haemodialysis and healthy sedentary controls. Kidney Int. 2000;57(6):2564-2570.

2. Sietsema KE, Amato A, Adler SG, Brass EP. Exercise capacity as a predictor of survival among ambulatory patients with end-stage renal disease. Kidney Int. 2004;65 (2):719-724. 3. Storer TW, Casaburi R, Sawelson S, Kopple JD. Endurance exercise training during haemodialysis improves strength, power, fatigability and physical performance in maintenance haemodialysis patients. Nephrol Dial Transplant. 2005;20(7):1429-1437. 4. Kouidi E, Grekas D, Deligiannis A, Tourkantonis A. Outcomes of long-term exercise training in dialysis patients: comparison of two training programs. Clin Nephrol. 2004; 61(suppl 1):S31-38. 5. Van Vilsteren MCBA, De Greef MHG, Huisman RM. The effects of a low-to-moderate intensity pre-conditioning exercise programme linked with exercise counselling for sedentary haemodialysis patients in The Netherlands: results of a randomised clinical trial. Nephrol Dial Transplant. 2005;20(1):141-146. 6. Tetzner F, Scholze A, Wittstock A, Zidek W, Tepel M. Impaired vascular reactivity in patients with chronic kidney disease. Am J Nephrol. 2008;28(2):218-223. 7. Briet M, Bozec E, Laurent S, et al. Arterial stiffness and enlargement in mild-to-moderate chronic kidney disease. Kidney Int. 2006;69(2):350-357. 8. Wang MC, Tsai WC, Chen JY, Huang JJ. Stepwise increase in arterial stiffness corresponding with the stages of chronic kidney disease. Am J Kidney Dis. 2005;45(3):494501. 9. Hambrecht R, Adams V, Erbs S, et al. Regular physical activity improves endothelial function in patients with coronary artery disease by increasing phosphorylation of endothelial nitric oxide synthase. Circulation. 2003;107(25):31523158. 10. Tanaka H, Dinenno FA, Monahan KD, Clevenger CM, De Souza CA, Seals DR. Aging, habitual exercise, and dynamic arterial compliance. Circulation. 2000;102(11): 1270-1275. 11. Vaitkevicius PV, Fleg JL, Engel JH, et al. Effects of age and aerobic capacity on arterial stiffness in healthy adults. Circulation. 1993;88(4 pt 1):1456-1462. 12. Edwards DG, Schofield RS, Magyari PM, Nichols WW, Braith RW. Effect of exercise training on central aortic pressure wave reflection in coronary artery disease. Am J Hypertens. 2004;17(6):540-543. 13. London GM, Marchais SJ, Guerin AP, Metivier F, Adda H. Arterial structure and function in end-stage renal disease. Nephrol Dial Transplant. 2002;17(10):1713-1724. 14. Laurent S, Cockcroft J, Van Bortel L, et al. Expert consensus document on arterial stiffness: methodological issues and clinical applications. Eur Heart J. 2006;27(21): 2588-2605. 15. Blacher J, Guerin AP, Pannier B, Marchais SJ, Safar ME, London GM. Impact of aortic stiffness on survival in end-stage renal disease. Circulation. 1999;99(18):24342439. 16. Barenbrock M, Spieker C, Laske V, et al. Effect of long-term hemodialysis on arterial compliance in end-stage renal failure. Nephron. 1993;65(2):249-253.

Exercise Training in End-Stage Renal Disease 17. London GM, Guerin AP, Marchais SJ, et al. Cardiac and arterial interactions in end-stage renal disease. Kidney Int. 1996;50(2):600-608. 18. Mustata S, Chan C, Lai V, Miller JA. Impact of an exercise program on arterial stiffness and insulin resistance in hemodialysis patients. J Am Soc Nephrol. 2004;15(10): 2713-2718. 19. Toussaint ND, Polkinghorne KR, Kerr PG. Impact of intradialytic exercise on arterial compliance and B-type natriuretic peptide levels in hemodialysis patients. Hemodial Int. 2008;12(2):254-263. 20. Koh KP, Fassett RG, Sharman JE, Coombes JS, Williams AD. Intradialytic versus home based exercise training in hemodialysis patients: a randomised controlled trial. BMC Nephrol. 2009;10(1):2. 21. US Department of Health and Human Services. Physical Activity and Health: A Report of the Surgeon General. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion; 1996. 22. Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports Exerc. 1982;14(5):377-381. 23. Wilkinson IB, Fuchs SA, Jansen IM, et al. Reproducibility of pulse wave velocity and augmentation index measured by pulse wave analysis. J Hypertens. 1998;16(12 pt 2):2079-2084. 24. Frimodt-Moller M, Nielsen AH, Kamper AL, Strandgaard S. Reproducibility of pulse-wave analysis and pulse-wave velocity determination in chronic kidney disease. Nephrol Dial Transplant. 2008;23(2):594-600. 25. Chen C-H, Nevo E, Fetics BE, et al. Estimation of central aortic pressure waveform by mathematical transformation of radial tonometry pressure: validation of generalised transfer function. Circulation. 1997;95(7):1827-1836. 26. Fassett RG, Robertson IK, Geraghty DP, Ball MJ, Burton NW, Coombes JS. Physical activity levels in chronic kidney disease patients entering the LORD trial. Med Sci Sports Exerc. 2009;41(5):985-991. 27. Timperio A, Salmon J, Crawford D. Validity and reliability of a physical activity recall instrument among overweight and non-overweight men and women. J Sci Med Sport. 2003;6(4):477-491. 28. Brown WJ, Trost SG, Bauman A, Mummery K, Owen N. Test-retest reliability of four physical activity measures used in population surveys. J Sci Med Sport. 2004;7(2):205-215. 29. Ware Jr JE, Snow KK, Kosinski M, Gandek B. SF-36 Health Survey Manual and Interpretation Guide. Boston,

99 MA: The Health Institute, New England Medical Center; 1997. 30. Painter P, Carlson L, Carey S, Paul SM, Myll S. Physical functioning and health-related quality of life changes with exercise training in hemodialysis patients. Am J Kidney Dis. 2000;35(3):482-492. 31. Knight EL, Ofsthun N, Teng M, Lazarus JM, Curhan GC. The association between mental health, physical function, and hemodialysis mortality. Kidney Int. 2003;63(5): 1843-1851. 32. Kodis J, Smith KM, Arthur HM, Daniels C, Suskin N, McKelvie RS. Changes in exercise capacity and lipids after clinic versus home-based aerobic training in coronary artery bypass graft surgery patients. J Cardiopulm Rehabil. 2001; 21(1):31-36. 33. Bendermacher BL, Willigendael EM, Teijink JA, Prins MH. Supervised exercise therapy versus non-supervised exercise therapy for intermittent claudication. Cochrane Database Syst Rev. 2006(2):CD005263. 34. Jamal SA, Leiter RE, Jassal V, Hamilton CJ, Bauer DC. Impaired muscle strength is associated with fractures in hemodialysis patients. Osteoporos Int. 2006;17(9): 1390-1397. 35. Isles RC, Choy NLL, Steer M, Nitz JC. Normal values of balance tests in women aged 20-80. J Am Geriat Soc. 2004;52(8):1367-1372. 36. Ferrier KE, Waddell TK, Gatzka CD, Cameron JD, Dart AM, Kingwell BA. Aerobic exercise training does not modify large-artery compliance in isolated systolic hypertension. Hypertension. 2001;38(2):222-226. 37. Westhoff TH, Franke N, Schmidt S, et al. Too old to benefit from sports? The cardiovascular effects of exercise training in elderly subjects treated for isolated systolic hypertension. Kidney Blood Press Res. 2007;30(4):240-247. 38. Hayashi K, Sugawara J, Komine H, Maeda S, Yokoi T. Effects of aerobic exercise training on the stiffness of central and peripheral arteries in middle-aged sedentary men. Jpn J Physiol. 2005;55(4):235-239. 39. Kakiyama T, Sugawara J, Murakami H, Maeda S, Kuno S, Matsuda M. Effects of short-term endurance training on aortic distensibility in young males. Med Sci Sports Exerc. 2005;37(2):267-271. 40. Canadian Erythropoietin Study Group. Association between recombinant human erythropoietin and quality of life and exercise capacity of patients receiving haemodialysis. Canadian Erythropoietin Study Group. BMJ. 1990; 300(6724):573-578.