The association between intradialytic exercise and hospital usage among hemodialysis patients

Abstract

Hemodialysis (HD) patients have high hospitalization rates. Benefits of intradialytic exercise have been proven in numerous studies yet exercise programs are still rarely used in the treatment of end-stage kidney disease (ESKD). Our objective was to determine if there was an association between a 6-month intradialytic bicycling program and hospitalization rates and length of stay (LOS) in ESKD patients. This was a retrospective cohort study that took place 6 months prior to and 6 months during an intradialtyic exercise program at an outpatient HD unit in Calgary, Alberta, Canada. Participants comprised 102 patients who had commenced HD <6 months (incident) or >6 months (prevalent) prior to starting exercise. The intervention comprised a 6-month intradialytic bicycling program. Main outcome measures were hospitalization rate, cause of hospitalization, and LOS. Patients were predominantly male (67.6%) aged 65.6 ± 13.5 years and median HD vintage 1 year (range: 0-12). Comorbidities included diabetes mellitus (50%) and cardiac disease (38.2%). The hospitalization incidence rate ratio (IRR) was 0.48 (0.23-0.98; P = 0.04) in incident and 0.89 (0.56-1.42; P = 0.64) in prevalent patients. The LOS decreased from 7.8 (95% confidence interval (CI): 7.3-8.4) to 3.1 (95% CI: 2.8-3.4) days and LOS IRR was 0.39 (0.35-0.45; P < 0.001). The main predictors of hospitalization were lower albumin levels (P = 0.007) and lack of intradialytic exercise program participation (P < 0.001). In conclusion, 6 months of intradialytic exercise was associated with decreased LOS in both incident and prevalent HD patients.

Full text

ARTICLE
The association between intradialytic exercise and hospital
usage among hemodialysis patients
Kristen Parker, Xin Zhang, Adriane Lewin, and Jennifer M. MacRae
Abstract: Hemodialysis (HD) patients have high hospitalization rates. Benefits of intradialytic exercise have been proven in numerous
studies yet exercise programs are still rarely used in the treatment of end-stage kidney disease (ESKD). Our objective was to determine
if there was an association between a 6-month intradialytic bicycling program and hospitalization rates and length of stay (LOS) in
ESKD patients. This was a retrospective cohort study that took place 6 months prior to and 6 months during an intradialtyic exercise
program at an outpatient HD unit in Calgary, Alberta, Canada. Participants comprised 102 patients who had commenced HD <6 months
(incident) or >6 months (prevalent) prior to starting exercise. The intervention comprised a 6-month intradialytic bicycling program.
Main outcome measures were hospitalization rate, cause of hospitalization, and LOS. Patients were predominantly male (67.6%) aged
65.6 ± 13.5 years and median HD vintage 1 year (range: 0–12). Comorbidities included diabetes mellitus (50%) and cardiac disease (38.2%).
The hospitalization incidence rate ratio (IRR) was 0.48 (0.23–0.98; P= 0.04) in incident and 0.89 (0.56–1.42; P= 0.64) in prevalent
patients. The LOS decreased from 7.8 (95% confidence interval (CI): 7.3–8.4) to 3.1 (95% CI: 2.8 –3.4) days and LOS IRR was 0.39 (0.35–0.45;
P< 0.001). The main predictors of hospitalization were lower albumin levels (P= 0.007) and lack of intradialytic exercise program
participation (P< 0.001). In conclusion, 6 months of intradialytic exercise was associated with decreased LOS in both incident and
prevalent HD patients.
Key words: hospitalization, length of stay, end-stage kidney disease.
Résumé : Les patients en hémodialyse (« HD ») présentent un taux élevé d’hospitalisation. Même si les bienfaits de l’exercice
intradialyse sont démontrés dans de nombreuses études, il y a peu de programmes d’exercices utilisés lors du traitement de la maladie
rénale au stade terminal (« ESKD »). Le but de cette étude est de vérifier l’association entre un programme d’exercices (6 mois) sur
bicyclette intradialyse et le taux d’hospitalisation et la durée du séjour (« LOS ») chez des patients EKSD. Il s’agit d’une étude
rétrospective de cohorte débutant 6 mois avant le programme d’exercices intradialyse et se poursuivant durant 6 mois de ce même
programme a
`la clinique externe HD de Calgary en Alberta au Canada. Les participants sont 102 patients aux prises avec HD <6 mois
(incident) ou >6 mois (prévalent) après le début des exercices. L’intervention consiste en un programme d’exercices sur bicyclette
intradialyse d’une durée de 6 mois. Les variables dépendantes sont le taux d’hospitalisation, les causes de l’hospitalisation et la durée
du séjour. Les patients sont surtout des hommes (67,6 %) âgés de 65,6 ± 13,5 ans et présentant une médiane de 1 an (écart : 0–12) aux
prises avec HD. Les comorbidités sont le diabète sucré (50 %) et la cardiopathie (38,2 %). Le ratio du taux d’incidence (IRR)
d’hospitalisation est de 0,48 (0,23, 0,98); P= 0,04 chez les patients incidents et de 0,89 (0,56, 1,42) P= 0,64 chez les patients prévalents.
La LOS diminue de 7,8 jours (intervalle de confiance (IC) a
`95 % 7,3–8,4) a
`3,1 jours (IC-95 % 2,8–3,4) et l’IRR de la LOS est de
0,39 (0,35–0,45); P< 0,001. Les principaux prédicteurs de l’hospitalisation sont le taux d’albumine (P= 0,007) et le manque de
participation au programme d’exercices intradialyse (P< 0,001). En conclusion, le programme d’exercices intradialyse est associé
a
`une diminution de la LOS chez les patients HD incidents et prévalents. [Traduit par la Rédaction]
Mots-clés : hospitalisation, durée du séjour, maladie rénale au stade terminal.
Introduction
The number of individuals diagnosed with end-stage kidney
disease (ESKD) has doubled in the last decade and most new pa-
tients start hemodialysis (HD) as their initial therapy (United States
Renal Data System (USRDS) Annual Report 2011;Canadian Organ
Replacement Register (CORR) Report 2013). Patients with chronic
kidney disease (CKD) have a greater risk of hospitalization than
the general population or others with chronic conditions (Schneider
et al. 2009;Strijack et al. 2009). As an individual with CKD pro-
gresses through the stages of kidney disease, the risk for hospital-
ization increases, making HD patients the most likely to experience
this outcome (Daratha et al. 2012). In addition, the length of stay
(LOS) associated with each hospital admission is longer for these
individuals (Schneider et al. 2009). Annually, HD patients experi-
ence an average of 2 hospital admissions with a combined mean
total of 11.9 days in hospital (USRDS Annual Report 2011). Many of
these admissions in the HD population are attributed to vascular
access complications, heart failure, and diabetes (Daratha et al. 2012).
Furthermore, the physical decline that occurs during lengthy hospi-
tal stays can also increase the risk of readmission (Chan et al. 2009).
The sedentary nature and poor physical functioning of the HD
population has been described previously (Johansen 2007;Kaysen
et al. 2011) and is a significant risk factor for mortality (O’Hare et al.
2003). Exercise is highly beneficial for all stages of CKD (Greenwood
et al. 2012;Heiwe and Jacobson 2011) with improvements in car-
diac function (Painter et al. 2002;Parsons and King-VanVlack
Received 8 September 2014. Accepted 8 December 2014.
K. Parker. Southern Alberta Renal Program, South Calgary Hemodialysis, 31 Sunpark Plaza SE, Calgary, AB T2X 3W5, Canada.
X. Zhang and A. Lewin. Department of Community Health Sciences, Faculty of Medicine, University of Calgary, Calgary, AB T2N 2T9, Canada.
J.M. MacRae. Division of Nephrology and Department of Cardiac Sciences, Cumming School of Medicine, University of Calgary, Calgary, AB T2N 2T9,
Canada.
Corresponding author: Jennifer MacRae (e-mail: jennifer.macrae@albertahealthservices.ca).
371
Appl. Physiol. Nutr. Metab. 40: 371–378 (2015) dx.doi.org/10.1139/apnm-2014-0326 Published at www.nrcresearchpress.com/apnm on 15 December 2014.
Appl. Physiol. Nutr. Metab. Downloaded from www.nrcresearchpress.com by University of Alberta on 08/19/15
For personal use only.

2009;Wilund et al. 2010), heart rate variability (Smart and Titus
2011) blood pressure (Miller et al. 2002), quality of life (Kolewaski
et al. 2005;Van Vilsteren et al. 2005), strength (Headley et al 2002;
Smart and Titus 2011;Storer et al. 2005), depression (Smart et al.
2013a), and self-reported physical function (Johansen et al. 2006;
Painter et al. 2000). Moreover, intradialytic exercise studies have
shown benefits of exercise on small solute clearance such as urea
(Kong et al. 1999;Parsons et al. 2006) and phosphate (Farese et al.
2008;Vaithilingam et al. 2004).
While there is evidence showing the benefits of intradialytic
exercise on various health and quality of life outcomes, it is not
known whether there is an association between intradialytic ex-
ercise and hospital admission rates or LOS. Therefore, this study
sought to address the impact of participation in a 6-month intra-
dialytic bicycling program on the rate of hospital admission and
LOS amongst a cohort of incident and prevalent HD patients.
Materials and methods
We performed a retrospective observational study in a cohort of
incident and prevalent HD patients. The procedures followed
were in accordance with the Declaration of Helsinki and its revi-
sions. The study protocol was approved by the Conjoint Health
Ethics Research Board of the University of Calgary and all partic-
ipants gave informed consent.
Study population and recruitment
Participants were recruited from 2 dialysis units in Calgary,
Alberta, Canada, between 1 August 2008 and 14 June 2012. These
2 dialysis units had an established intradialytic bicycling program
as a part of routine care for HD patients. Potential participants
were assessed for exercise readiness by the staff kinesiologist us-
ing a questionnaire that was developed in conjunction with the
Medical Director (Appendix A). Patients were excluded from exer-
cise if they had experienced a myocardial infarction within the
previous 6 months, had unstable angina, decompensated conges-
tive heart failure, physical limitations that would affect usage of
the bike, or other concerns (vascular access issues, poor blood
sugar control, and hemodynamic instability; Fig. 1). Resolution of
the aforementioned conditions to a more stable state could result
in participation with a doctor’s clearance. No stress tests were
required in any of the 102 patients prior to commencing exercise.
Exercise program description
A kinesiologist or nurse placed the Monark Rehab Trainer 881E
(Monark Exercise AB, Vansbro, Sweden) at the base of the dialysis
chair (Champion Medical Chair, Gobal Medical Products, Ont.,
Canada) during the first 2 h of an HD treatment. The study partic-
ipants were encouraged to exercise during each of their thrice
weekly dialysis treatments if they felt well enough to do so. All par-
ticipants started with a 5- to 10-min trial and were given an orienta-
tion on safety, gradual progression, and proper warm-up/cool-down
procedures. Participants self-selected an exercise intensity that fit
their perception of “moderate” to “somewhat hard” on the Borg
scale (Borg 1970). Participants self-progressed their exercise duration
at the rate of 2%–5% per week with the goal of achieving at least
30 min of exercise during each dialysis run, and were encouraged to
increase the duration beyond 30 min if they felt able to do so. Pre-,
mid-, and postexercise values (blood pressure, heart rate, and oxygen
saturation) were monitored by the kinesiologist or nurses.
Data sources and outcome measures
Medical records from the Southern Alberta Renal Program (SARP)
database were reviewed to determine cause of CKD and hospital
admission data. The SARP database collects all CKD patients’ medical
information, hospital stays, interventions, medications, bloodwork
values, and multidisciplinary progress notes (Manns et al. 2001). The
cause of CKD was classified as diabetes, hypertension/ischemia, au-
toimmune/glomerulonephritis, cystic, or other/unknown. Causes of
hospital admission were categorized as fluid overload, infection, car-
diac, metabolic, gastrointestinal, respiratory, and “other”. Examples
of “other” causes are injury, cancer, syncope, and failure to thrive.
Reasons for not completing the 6-month exercise program were due
to medical instability, hospitalizations, lack of motivation, unavail-
able staff, moving or transferring to other units, transplantation,
peritoneal dialysis transfer, or bereavement. Participants who had
commenced dialysis less than 6 months prior to starting the exercise
program were considered incident patients, and participants who
had been on dialysis for 6 months or longer prior to starting the
exercise program were considered prevalent patients.
Data on participant hospital admissions, LOS, hemoglobin, albu-
min, creatinine, urea, urea reduction ratio, hemoglobin A1c, and
body mass index (BMI) were extracted retrospectively from the SARP
database. Monthly bloodwork was tracked 6 months prior to exer-
cise initiation up until 6 months after exercise initiation. Exercise log
sheets from the participants’ medical charts were used to calculate
the number of sessions each participant completed in the first
6 months of exercise.
Study outcomes
The primary outcome was the hospitalization rate 6 months
prior to the exercise program and during the 6 months of the
exercise program. A hospital admission was defined as any un-
planned visit to the emergency room resulting in at least 1 over-
night stay. Scheduled surgeries and planned clinic appointments
with doctors or specialists were not included as hospital admis-
sions. Secondary outcomes such as causes of hospital admissions
and LOS were captured in the SARP database and confirmed for
accuracy by cross-referencing with hospital discharge reports.
Statistical analysis
Our analysis was “intention-to-treat” and therefore included all
102 participants regardless of the number of exercise sessions
Fig. 1. Patient flow during the study.
372 Appl. Physiol. Nutr. Metab. Vol. 40, 2015
Published by NRC Research Press
Appl. Physiol. Nutr. Metab. Downloaded from www.nrcresearchpress.com by University of Alberta on 08/19/15
For personal use only.

performed during the 6-month exercise program. Study partici-
pants’ pre- and postexercise measurements are presented as means
with standard deviation for continuous variables (age, weight,
BMI, laboratory values, number of exercise session completed) or
as number and percent for categorical variables (sex, etiology of
ESKD, comorbidities, and reasons for hospitalization). Hospital-
ization rates and mean length of stay pre- and postexercise were
calculated using Poisson regression. Predictors of hospitalization
were calculated using a zero-inflated negative binomial model
adjusted for baseline patient characteristics and accounting for
clustering within patient. Results were calculated for the full
group and then stratified by incident/prevalent dialysis status. All
statistical analyses were performed using Stata version 11 (Stata-
Corp LP, College Station, Texas, USA).
Results
Of 276 patients with HD screened at 2 dialysis units, 114 consented
to participate and 102 were eligible and started the 6-month exer-
cise program (Fig. 1). All patients were followed up for the duration
of the 6-month exercise program. Participants were predominantly
male (67.6%), with a mean age of 65.6 ± 13.5 years with a median
HD vintage of 1 year (range: 0–12). Fifty percent (n= 51) had diabe-
tes, 66.7% (n= 68) had hypertension, and 38.2% (n= 39) had coro-
nary artery disease, heart failure, or arrhythmias (Table 1). At
baseline, 74.5% of participants (n= 76) were prevalent and 25.5%
(n= 26) were incident to HD. Incident patients had been on dialysis
for a median of 3 months (range: 0–5) versus 33 months (6–145),
were younger (mean age 59 ± 14 vs 68 ± 13 years; P= 0.002), and had
a lower urea reduction ratio (URR) than the prevalent patients
(incident: mean 73% ± 7%; prevalent: mean 77% ± 7%; P= 0.03).
Incident patients had less cardiovascular comorbidity than prev-
alent patients (11.5% vs 47.4%; P= 0.001).
Delivered dose of exercise
Overall, study participants completed a mean of 4.5 exercise
sessions per month (range: 0.7–11.8). Incident patients had a higher
mean number of workouts per month than prevalent patients
(4.9; range 1.5–11.8 vs 4.3; range 0.7–9.5; P= 0.37); however, a
smaller proportion of incident patients completed all 6 months of
exercise (65.3% (n= 17) vs 79% (n= 60); P= 0.16). Of the 102 participants
who started the program, 77 (75.5%) completed all 6 months of exer-
cise, participating in a mean of 5.2 sessions per month. Overall, 102
patients completed a total of 2102 exercise sessions in 6 months. The
2 most common reasons for failing to complete 6 months of exercise
were medical instability (12.9%; n= 13) and hospital admissions
(5.9%; n= 6). The most common causes of medical instability were
vascular access problems, hypotension/hypertension, nausea, anemia/
fatigue, or poor blood sugar control. Only a few participants had
issues with motivation (n= 2), unavailable staff (n= 2), transfers to
peritoneal dialysis (n= 1), and transplantation (n= 1) that attrib-
uted to their inability to complete 6 months of exercise. No deaths
were recorded during the study period.
Primary outcome
The hospitalization rate for all study participants was 0.60 (95%
confidence interval (CI): 0.47–0.77) during the 6 months prior to
the exercise program (Table 2). This decreased to 0.44 (95% CI:
0.33–0.59) during the 6-month exercise program but the differ-
ence was not significant (incidence rate ratio (IRR): 0.74 (95% CI:
0.50–1.08); P= 0.12). Incident patients experienced a greater reduc-
tion in hospitalization rates during the 6-month exercise program
(IRR: 0.48, 95% CI: 0.23–0.98; P= 0.04) than prevalent patients (IRR:
0.89, 95% CI: 0.56–1.42; P= 0.64).
Secondary outcomes
The secondary outcome, mean LOS, fell from 7.8 days (95% CI:
7.3–8.4) to 3.1 days (95% CI: 2.8–3.4) during the 6-month exercise
program (IRR: 0.39, 95% CI: 0.35– 0.45; P< 0.001). This held true for
both the incident (IRR: 0.16, 95% CI: 0.11–0.22; P< 0.001) and prev-
alent patient groups (IRR: 0.48, 95% CI: 0.42–0.55; P< 0.001).
The most common reason for hospitalization during the 6-month
pre-exercise period was infection among prevalent patients
Table 1. Baseline characteristics of study participants.
Prevalent cases;
n= 76, 74.5%
Incident cases;
n= 26, 25.5%
Full cohort;
N= 102
Age, mean (±SD) 68.0 (12.6) 58.8 (14.0) 65.6 (13.5)
Female (%) 25 (32.9) 8 (30.8) 32.4
Etiology end-stage kidney disease (%)
Diabetes mellitus 34 (44.7) 11 (42.3) 45 (44.1)
Hypertension 17 (22.3) 3 (11.5) 20 (19.6)
Autoimmune or glomerulonephritis 5 (6.6) 5 (19.2) 10 (9.8)
Cystic disease 2 (2.6) 2 (7.7) 4 (3.9)
Other 18 (23.7) 5 (19.2) 23 (22.6)
Years on HD, median (range) 2 (0−12) 0 1 (0−12)
Comorbidity (%)
Diabetes mellitus 38 (50.0) 13 (50.0) 51 (50.0)
Hypertension 48 (63.2) 20 (76.9) 68 (66.7)
Coronary artery disease and/or heart
failure and/or arrhythmia, combined
36 (47.4) 3 (11.5) 39 (38.2)
Cerebrovascular accident 16 (21.1) 3 (11.5) 19 (18.6)
Peripheral vascular disease 7 (9.2) 1 (3.9) 8 (7.8)
Other, combined with cancer 31 (40.8) 8 (30.8) 39 (38.2)
Body mass index, mean (±SD) 26.4 (6.3) 26.4 (6.7) 26.4 (6.4)
Weight, mean (±SD) 74.2 (17.4) 77.2 (24.1) 75.0 19.3
Laboratory values at baseline, mean (±SD)
Hemoglobin 110 (12) 113 (14) 111.1 (12.6)
Albumin 34 (3.9) 33 (4.4) 33.8 (4.0)
Creatinine 641 (198) 568 (201) 622 (200)
Urea 18.6 (4.7) 18.6 (5.8) 18.6 (5.0)
Urea reduction ratio 76.6 (6.5) 73.3 (7.1) 75.8 (6.8)
Hemoglobin AIc; n= 51, limited to
those with Diabetes mellitus
7.01 (1.38) 7.25 (1.57) 7.07 (1.42)
Parker et al. 373
Published by NRC Research Press
Appl. Physiol. Nutr. Metab. Downloaded from www.nrcresearchpress.com by University of Alberta on 08/19/15
For personal use only.

(42.1% hospitalizations), and fluid overload among incident patients
(39.1% hospitalizations). During the 6-month exercise program, the
most common reason for hospitalization in prevalent patients was
cardiac-related (26.5% hospitalizations) while infections and respira-
tory issues were the causes for admissions among incident patients
(27.3% hospitalizations) (Table 3).
In the full cohort there were statistically significant increases
in URR (75.7% ± 6.8% vs. 76.8% ± 5.7% after 6 months of exercise;
P= 0.02) and albumin (33.8 ± 4.0 g/L vs. 34.9 ± 3.9 g/L after 6 months
of exercise; P= 0.001). We also observed similar changes when
examining only incident patients URR (73.3% ± 7.1% vs. 75.5% ±
6.2% after 6 months of exercise, P= 0.02) and albumin values
(33.3 ± 4.4 vs. 35.7 ± 3.7 after 6 months of exercise, P< 0.001).
Incident patients also had increased creatinine (6 months pre-exer-
cise: 568 ± 201 ␮mol/L vs. 630 ± 182 ␮mol/L after 6 months of exercise,
P= 0.03). No changes were evident in urea, hemoglobin A1c, or BMI
values (Table 4). A zero-inflated negative binomial model adjusted for
baseline patient characteristics found that lower albumin levels and
Table 2. Number of exercise sessions completed throughout the study.
Prevalent cases;
n= 76, 74.5%
Incident cases;
n= 26, 25.5%
All participants;
N= 102
Median number of exercise
sessions completed (range)
26 (4−57) 29 (9−71) 27 (4−71)
Median number of exercise
sessions per mo (range)
4.3 (0.7−9.5) 4.9 (1.5−11.8) 4.5 (0.7−11.8)
No. patients who completed
6 mo of exercise (%)
60 (79.0) 17 (65.3) 77 (75.5)
Table 3. Rate of hospitalization and length of stay at baseline and post 6-month exercise program by
incident/prevalent status.
Prevalent cases;
n= 76, 74.5%
Incident cases;
n= 26, 25.5%
All participants;
N= 102
Hospitalizations
Hospitalization rate at baseline (95% CI) 0.50 (0.36−0.69) 0.88 (0.59−1.33) 0.60 (0.47−0.77)
Hospitalization rate at 6 mo (95% CI) 0.45 (0.32−0.63) 0.42 (0.23−0.76) 0.44 (0.33−0.59)
Incidence rate ratio 0.89 (0.56−1.42) 0.48 (0.23−0.98) 0.74 (0.50−1.08)
Hospitalization rate post- vs pre-exercise P= 0.64 P= 0.04 P= 0.12
LOS
Mean LOS at baseline (95% CI) 7.7 (7.1−8.4) 8.2 (7.1−9.3) 7.8 (7.3−8.4)
Mean LOS at 6 mo (95% CI) 3.7 (3.3−4.2) 1.3 (0.9−1.8) 3.1 (2.8−3.4)
Incidence rate ratio 0.48 (0.42−0.55) 0.16 (0.11−0.22) 0.39 (0.35−0.45)
LOS post- vs pre-exercise P< 0.001 P< 0.001 P< 0.001
Note: CI, confidence interval; LOS, length of stay.
Table 4. Anthropometric and lab measurements pre− and post−6-month exercise program by
incident/prevalent status.
(a) Patient variables.
Prevalent, n= 76 Incident, n=26
Pre-exercise Postexercise Pre-exercise Postexercise
Weight, mean (±SD) 74.2 (17.4) 74.2 (17.2) 77.2 (24.1) 77.3 (24.1)
Body mass index, mean (±SD) 26.4 (6.3) 26.4 (6.3) 26.4 (6.7) 26.4 (6.8)
Laboratory results, mean (±SD)
Hemoglobin 110.4 (12.0) 111.8 (10.8) 113.1 (14.3) 113.0 (14.2)
Albumin 34.0 (3.9) 34.6 (4.0) 33.3 (4.4) 35.7 (3.7)
Creatinine 641 (198) 646 (215) 568 (201) 630 (182)
Urea 18.6 (4.7) 18.5 (5.2) 18.6 (5.8) 19.8 (5.8)
Urea reduction ratio 76.6 (6.5) 77.2 (5.5) 73.3 (7.1) 75.5 (6.2)
Hemoglobin AIc 7.01 (1.38) 7.14 (1.23) 7.25 (1.57) 7.25 (1.61)
(b) Reason for hospitalization.
Prevalent patients Incident patients
Reason for hospitalization, n(%)
Pre-exercise,
n=38
Postexercise,
n=34
Pre-exercise,
n=23
Postexercise,
n=11
Infection 16 (42.1) 6 (17.6) 7 (30.4) 3 (27.3)
Coronary artery disease 1 (2.6) 9 (26.5) 0 0
Fluid overload 0 1 (2.9) 9 (39.1) 0
Metabolic 4 (10.5) 2 (5.9) 3 (13.0) 1 (9.1)
Gastrointestinal 2 (5.3) 4 (11.8) 0 1 (9.1)
Respiratory 2 (5.3) 1 (2.9) 0 3 (27.3)
Other 13 (34.2) 11 (32.4) 4 (17.4) 3 (27.3)
Note: The reason for hospitalization is depicted above. In some cases there was more than 1 main reason for
hospitalization, which is why the hospitalization reasons are different from the patient numbers.
374 Appl. Physiol. Nutr. Metab. Vol. 40, 2015
Published by NRC Research Press
Appl. Physiol. Nutr. Metab. Downloaded from www.nrcresearchpress.com by University of Alberta on 08/19/15
For personal use only.

the lack of participation in intradialytic exercise were predictors of
hospitalization (P= 0.007 and P= 0.001, respectively).
Discussion
To our knowledge, this study is the first to report on the associa-
tion between intradialytic exercise program and hospital utilization.
Consistent with previous reports, albumin (Lacson et al. 2007) and
physical inactivity (Belardinelli et al. 2001) were predictors of hospi-
talization. Hospitalization rates declined in incident patients and the
length of hospital stay declined in both incident and prevalent pa-
tients during a 6-month intradialytic bicycling program. Incident
patients had a significant decrease in hospital admissions and both
incident and prevalent patients had a reduction in LOS.
Our study findings are similar to those on other chronic disease
populations. Exercise programs for those with cardiac (Belardinelli
et al. 2001;Davidson et al. 2010;Plüss et al. 2011), pulmonary
(Griffiths et al. 2000;Hui and Hewitt 2003), or other chronic con-
ditions (Courtney et al. 2009) have shown a decrease in either
hospital admissions, length of stay, or both. Cardiac patients who
underwent 1 year of “extended” rehabilitation had an average
stay in hospital of 6 days versus a 10-day stay in those who had
3 months of rehabilitation (Plüss et al. 2011), while pulmonary
patients who participated in a 6-week exercise program had LOS
shortened by a mean of 3 days per admission (Griffiths et al. 2000).
Similar results have been seen in older patients with a variety of
comorbidities. Courtney (Courtney et al. 2009) targeted 128 older
patients who were admitted to hospital and then introduced to a
simple, in-centre exercise program, with a home visit and phone
follow-up after discharge. Fewer patients in the intervention
group were re-admitted (22%) versus those in the control group
(47%). In addition, the exercise group had significantly fewer visits
to the doctor and commented on an improved quality of life.
Although we found a reduction in hospital visits during a 6-month
period of intradialytic exercise training amongst incident dialysis
patients, other factors may have played a role. In fact, Smart (Smart
and Titus 2011) showed that early nephrology referral is associated
with a reduction in hospitalization, possibly because of better pa-
tient preparation for dialysis. Mix (Mix et al. 2003) investigated the
hospital admission rates of over 100 000 patients with CKD for
2 years prior to dialysis start and found a substantial increase in
hospital admissions 3 months before and after commencing dialysis
because of vascular access issues, coronary artery disease, and infec-
tions (Mix et al. 2003). Similarly, another report suggested that the
cause of increased hospitalizations in ESKD is related to initiating HD
itself (Kassam et al. 2011). This may explain the elevated hospital rate
and LOS observed in our incident patients during the pre-exercise
period. The most common cause of hospitalization in the 6 months
before incident patients started exercise was due to fluid overload, a
condition treated by starting HD. In the postexercise period, infec-
tions and respiratory complications were equal contributors to hos-
pitalizations in this group. It is possible that the reductions in
hospital utilization seen in the 6 months after exercise started may
have been due to HD stability rather than a true benefit of the exer-
cise; however, the decline in hospital LOS in the 76 prevalent patients
could suggest otherwise. It is also possible that confounding by indi-
cation is present, whereby people who are less likely to exercise are
also more likely to be hospitalized. This could explain why our re-
gression model found “lack of participation in intradialytic exercise”
to be a predictor of hospitalization.
The reduction in LOS in prevalent patients is of great interest
considering those with a longer HD vintage are potentially subjected
to more complications from long-standing comorbidities and the
progression of disease. Although the prevalent patients decreased
their total hospital LOS, they saw an increase in cardiac-related ad-
missions after exercise commenced. As 47% of prevalent patients had
cardiac comorbidities at baseline, this is not an unexpected finding.
CKD patients have been known to have a high prevalence of left
ventricle dysfunction, hypertension, and coronary artery disease
(Weiner et al. 2004). For this reason, pre-screening and a thorough
medical history were carefully assessed to ensure safety of exercise. It
is important to note that none of the participants in this study re-
quired emergency care immediately following an exercise session. It
is likely that prevalent patients have been subjected to lengthy pro-
gression of their cardiac condition and these issues may have per-
sisted whether they were involved in an exercise program or not.
Unfortunately, with the absence of a control group, we have no way
of determining what may have occurred in the prevalent patients
with cardiovascular histories. There is a need for future research
studies to address this limitation.
Another factor to consider is that the exercise dose (4.5 sessions/
month) is below the recommended dose for CKD patients (Smart
et al. 2013b) or the minimal requirement to elicit an improvement
in cardiovascular function (Parsons and King-VanVlack 2009;American
College of Sports Medicine Position 1994). Since HD patients are
known to have significantly poor physical function, the idea that
“something is better than nothing” may be the case. Further work
regarding optimal exercise dose is warranted in this area.
It is important to note that many of our incident patients started
the exercise program during what is considered to be a “high risk”
time of dialysis. Collins (Collins et al. 2009) examined USRD reports
and commented on the peak in all-cause and cause-specific mortality
between the first and fourth month after commencing dialysis. Our
incident patients were offered exercise as early as 1 month after HD
initiation if they demonstrated hemodynamic stability during their
dialysis treatment. No adverse events were reported. Early exercise
intervention has been proven and a number of studies have demon-
strated a safe start as soon as 1 week after hospital discharge from a
cardiovascular event (Haykowsky et al. 2011), 6 days after a stroke
(Stoller et al. 2012), or within 48 h of hospital admission (Tang et al.
2009).
Limitations
This retrospective observational study showed a decreased hos-
pital admission rate in patients after exercise commenced; how-
ever, these findings should be interpreted with caution. Given
that this is an observational study, our findings cannot be inter-
preted as causal but rather as hypothesis generating. We acknowl-
edge that a major limitation is the lack of a control group and the
lack of baseline fitness testing. Furthermore, our results are lim-
ited by the accuracy of the data extracted from the renal program
database. However, all hospitalization dates and causes were cross
referenced with the hospital discharge summaries and all data cap-
ture was done by a single trained individual to ensure consistency.
Another potential limitation was the lack of a “set intensity” as we
encouraged our patients to choose a pace that fit their perceived
definition of “moderate to somewhat hard”, thus making it difficult
to draw any conclusions about an ideal intensity or dose–response
relationship. Finally, with the absence of a control group, we have no
way of determining longer term hospital outcomes with incident
patients or what may have occurred in the prevalent patients with
cardiovascular comorbidities in the absence of an exercise program.
Future studies must address this limitation.
Conclusions
This observational cohort study found an association between
intradialytic exercise and decreased hospital admissions and mean
LOS. Well-designed, prospective studies on exercise programs and
health-care utilization amongst patients with ESKD are needed.
Conflict of interest statement
The authors have no conflicts of interest to declare.
Acknowledgement
The study was funded by the Division of Nephrology, University
of Calgary, Calgary, Alberta, Canada.
Parker et al. 375
Published by NRC Research Press
Appl. Physiol. Nutr. Metab. Downloaded from www.nrcresearchpress.com by University of Alberta on 08/19/15
For personal use only.

References
American College of Sports Medicine Position. 1994. Exercise for patients with
coronary artery disease. Med. Sci. Sports Exerc. 26(3): i–v.
Belardinelli, R., Paolini, I., Cianci, G., Piva, R., Georgiou, D., and Purcaro, A. 2001.
Exercise training intervention after coronary angioplasty: the ETICA trial.
J. Am. Coll. Cardiol. 37(7): 1891–1900. doi:10.1016/S0735-1097(01)01236-0.
PMID:11401128.
Borg, G. 1970. Perceived exertion as an indicator of somatic stress. Scand. J.
Rehabil. Med. 2(2): 92–98. PMID:5523831.
Chan, K.E., Lazarus, J.M., Wingard, R.L., and Hakim, R.M. 2009. Association
between repeat hospitalization and early intervention in dialysis patients
following hospital discharge. Kidney Int. 76(3): 331–341. doi:10.1038/ki.2009.
199. PMID:19516243.
Collins, A.J., Foley, R.N., Gilbertson, D.T., and Chen, S. 2009. The state of chronic
kidney disease, ESRD, and morbidity and mortality in the first year of dialy-
sis. Clin. J. Am. Soc. Nephrol. 4: S5–S11. doi:10.2215/CJN.05980809. PMID:
19996006.
CORR Report. 2013. Treatment of end-stage organ failure in Canada, 2002 to 2011.
Canadian Institute for Health Information, Ottawa, Ont., Canada.
Courtney, M., Edwards, H., Chang, A., Parker, A., Finlayson, K., and Hamilton, K.
2009. Fewer emergency readmissions and better quality of life for older
adults at risk of hospital readmission: a randomized controlled trial to de-
termine the effectiveness of a 24-week exercise and telephone follow-up
program. J. Am. Geriatr. Soc. 57: 395–402. doi:10.1111/j.1532-5415.2009.02138.x.
PMID:19245413.
Daratha, K.B., Short, R.A., Corbett, C.F., Ring, M.E., Alicic, R., Choka, R., et al.
2012. Risks of subsequent hospitalization and death in patients with kidney
disease. Clin. J. Am. Soc. Nephrol. 7: 409– 416. doi:10.2215/CJN.05070511. PMID:
22266573.
Davidson, P.M., Cockburn, J., Newton, P.J., Webster, J.K., Betihavas, V., Howes, L.,
et al. 2010. Can a heart failure-specific cardiac rehabilitation program de-
crease hospitalizations and improve outcomes in high-risk patients? Eur. J.
Cardiovasc. Prev. Rehabil. 17: 393–402. doi:10.1097/HJR.0b013e328334ea56.
PMID:20498608.
Farese, S., Budmiger, R., Aregger, F., Bergmann, I., Frey, F.J., and Uehlinger, D.E.
2008. Effect of transcutaneous electrical muscle stimulation and passive cy-
cling movements on blood pressure and removal of urea and phosphate
during hemodialysis. Am. J. Kidney Dis. 52(4): 745–752. doi:10.1053/j.ajkd.
2008.03.017. PMID:18487001.
Greenwood, S.A., Lindup, H., Taylor, K., Koufaki, P., Rush, R., Macdougall, I.C.,
et al. 2012. Evaluation of a pragmatic exercise rehabilitation programme in
chronic kidney disease. Nephrol. Dial. Transplant. 27(Suppl. 3): iii126–iii134.
doi:10.1093/ndt/gfs272.
Griffiths, T.L., Burr, M.L., Campbell, I.A., Lewis-Jenkins, V., Mullins, J., Shiels, K.,
et al. 2000. Results at 1 year of outpatient multidisciplinary pulmonary reha-
bilitation: a randomised controlled trial. Lancet, 355: 362–368. doi:10.1016/
S0140-6736(99)07042-7. PMID:10665556.
Haykowsky, M., Scott, J., Esch, B., Schopflocher, D., Myers, J., Paterson, I., et al.
2011. A meta-analysis of the effects of exercise training on left ventricular
remodeling following myocardial infarction: start early and go longer for
greatest exercise benefits on remodeling. Trials, 12: 92. doi:10.1186/1745-6215-
12-92. PMID:21463531.
Headley, S., Germain, M., Mailloux, P., Mulhern, J., Ashworth, B., Burris, J., et al.
2002. Resistance training improves strength and functional measures in pa-
tients with end-stage renal disease. Am. J. Kidney Dis. 40: 355–364. doi:10.
1053/ajkd.2002.34520. PMID:12148109.
Heiwe, S., and Jacobson, S.H. 2011. Exercise training for adults with chronic
kidney disease. Cochrane Database Syst. Rev. 5(10): CD003236. doi:10.1002/
14651858.CD003236.pub2. PMID:21975737.
Hui, K.P., and Hewitt, A.B. 2003. A simple pulmonary rehabilitation program
improves health outcomes and reduces hospital utilization in patients with
COPD. Chest, 124: 94–97. doi:10.1378/chest.124.1.94. PMID:12853508.
Johansen, K.L. 2007. Exercise in the end-stage renal disease population. J. Am.
Soc. Nephrol. 18: 1845–1854. doi:10.1681/ASN.2007010009. PMID:17442789.
Johansen, K.L., Painter, P., Sakkas, G.K., Gordon, P., Doyle, J., and Shubert, T.
2006. Effects of resistance exercise training and nandrolone decanoate on
body composition and muscle function among patients who receive hemo-
dialysis: A randomized, controlled trial. J. Am. Soc. Nephrol. 17: 2307–2314.
doi:10.1681/ASN.2006010034. PMID:16825332.
Kassam, H., Sun, Y., Adeniyi, M., Agaba, E.I., Martinez, M., Servilla, K.S., et al.
2011. Hospitalizations before and after initiation of chronic hemodialysis.
Hemodialysis Int. 15: 341–349. doi:10.1111/j.1542-4758.2011.00551.x.
Kaysen, G.A., Larive, B., Painter, P., Craig, A., Lindsay, R.M., Rocco, M.V., et al.
2011. Baseline physical performance, health, and functioning of participants
in the Frequent Hemodialysis Network (FHN) Trial. Am. J. Kidney Dis. 57(1):
101–112. doi:10.1053/j.ajkd.2010.08.021. PMID:21184919.
Kolewaski, C.D., Mullally, M.C., Parsons, T.L., Paterson, M.L., Toffelmire, E.B.,
and King-VanVlack, C.E. 2005. Quality of life and exercise rehabilitation in
end stage renal disease. CANNT J. 15(4): 22–29. PMID:16491995.
Kong, C., Tattersall, J., Greenwood, R., and Farrington, K. 1999. The effect of
exercise during haemodialysis on solute removal. Nephrol. Dial. Transplant.
14: 2927–2931. doi:10.1093/ndt/14.12.2927.
Lacson, E., Ikizler, T.A., Lazarus, J.M., Teng, M., and Hakim, R.M. 2007. Potential
impact of nutritional intervention on end-stage renal disease hospitaliza-
tion, death, and treatment costs. J. Renal Nutr. 17(6): 363–371. doi:10.1053/j.
jrn.2007.08.009.
Manns, B.J., Mortis, G.P., Taub, K.J., McLaughlin, K., Donaldson, C., and
Ghali, W.A. 2001. The Southern Alberta Renal Program database: a prototype
for patient management and research initiatives. Clin. Invest. Med. 24(4):
164–170. PMID:11558850.
Miller, B., Cress, C., Johnson, M., Nichols, D., and Schnitzler, M. 2002. Exercise
during hemodialysis decreases the use of antihypertensive medications. Am.
J. Kidney Dis. 39(4): 828–833. doi:10.1053/ajkd.2002.32004. PMID:11920350.
Mix, T.-C., St. Peter, W.L., Ebben, J., Xue, J., Pereira, B.J.G., Kausz, A.T., et al. 2003.
Hospitalization during advancing chronic kidney disease. Am. J. Kidney Dis.
42(5): 972–981. doi:10.1016/j.ajkd.2003.06.001. PMID:14582041.
O’Hare, A.M., Tawney, K., Bacchetti, P., and Johansen, K. 2003. Decreased sur-
vival among sedentary patients undergoing dialysis: results from the dialysis
morbidity and mortality study wave 2. Am. J. Kidney Dis. 41: 447–454. doi:10.
1053/ajkd.2003.50055. PMID:12552509.
Painter, P., Carlson, L., Carey, S., Paul, S.M., and Myll, J. 2000. Low-functioning
hemodialysis patients improve with exercise training. Am. J. Kidney Dis.
36(3): 600–608. doi:10.1053/ajkd.2000.16200. PMID:10977793.
Painter, P., Moore, G., Carlson, L., Paul, S., Myll, J., Phillips, W., et al. 2002. Effects
of exercise training plus normalization of hematocrit on exercise capacity
and health-related quality of life. Am. J. Kidney Dis. 39: 257–265. doi:10.1053/
ajkd.2002.30544. PMID:11840365.
Parsons, T.L., and King-VanVlack, C.E. 2009. Exercise and end-stage kidney disease:
functional exercise capacity and cardiovascular outcomes. Adv. Chronic Kid-
ney Dis. 16(6): 459–481. doi:10.1053/j.ackd.2009.08.009. PMID:19801136.
Parsons, T., Toffelmire, E., and King-VanVlack, C. 2006. Exercise training during
hemodialysis improves dialysis efficacy and physical performance. Arch. Phys. Med.
Rehabil. 87: 680–687. doi:10.1016/j.apmr.2005.12.044. PMID:16635631.
Plüss, C.E., Billing, E., Held, C., Henriksson, P., Kiessling, A., Karlsson, M.R., et al.
2011. Long-term effects of an expanded cardiac rehabilitation programme
after myocardial infarction or coronary artery bypass surgery: a five-year
follow-up of a randomized controlled study. Clin. Rehabil. 25(1): 79–87. doi:
10.1177/0269215510376006. PMID:20702510.
Schneider, K.M., O’Donnell, B.E., and Dean, D. 2009. Prevalence of multiple
chronic conditions in the United States’ Medicare population. Health Qual.
Life Outcomes, 7: 82. doi:10.1186/1477-7525-7-82. PMID:19737412.
Smart, N.A., and Titus, T.T. 2011. Outcomes of early versus late nephrology
referral in chronic kidney disease: a systematic review. Am. J. Med. 124(11):
1073–1080. doi:10.1016/j.amjmed.2011.04.026. PMID:22017785.
Smart, N.A., McFarlane, J., and Cornelissen, V. 2013a. The effect of exercise
therapy on physical function, biochemistry and dialysis adequacy in haemo-
dialysis patients: a systematic review and meta-analysis. Open J. Nephrol. 3:
25–36. doi:10.4236/ojneph.2013.31005.
Smart, N.A., Williams, A.D., Levinger, I., Selig, S., Howden, E., Coombes, J.S., et al.
2013b. Exercise & Sport Science Australia (ESSA) position statement on exer-
cise and chronic kidney disease. J. Sci. Med. Sport. 16(5): 406– 411. doi:10.1016/
j.jsams.2013.01.005. PMID:23434075.
Stoller, O., de Bruin, E., Knols, R., and Hunt, K. 2012. Effects of cardiovascular
exercise early after stroke: systematic review and meta-analysis. BMC Neurol.
12: 45. doi:10.1186/1471-2377-12-45. PMID:22727172.
Storer, T.W., Casaburi, R., Sawelson, S., and Kopple, J.D. 2005. Endurance exer-
cise training during haemodialysis improves strength, power, fatigability
and physical performance in maintenance haemodialysis patients. Nephrol.
Dial. Transplant. 20: 1429–1437. doi:10.1093/ndt/gfh784. PMID:15840667.
Strijack, B., Mojica, J., Sood, M., Komenda, P., Bueti, J., Reslerova, M., et al. 2009.
Outcomes of chronic dialysis patients admitted to the intensive care unit.
J. Am. Soc. Nephrol. 20: 2441–2447. doi:10.1681/ASN.2009040366. PMID:19729437.
Tang, A., Sibley, K., Thomas, S., Bayley, M., Richardson, D., McIlroy, W., et al.
2009. Effects of an aerobic exercise program on aerobic capacity, spatiotem-
poral gait parameters, and functional capacity in subacute stroke. Neurore-
habil. Neural Repair, 23: 398–406. PMID:19088223.
USRDS. 2011. Annual data report: atlas of chronic kidney disease and end-stage
renal disease in the United States. National Institutes of Health, National
Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, Md., USA.
Vaithilingam, I., Polkinghorne, K.R., Atkins, R.C., and Kerr, P.G. 2004. Time and
exercise improve phosphate removal in hemodialysis patients. Am. J. Kidney
Dis. 43(1): 85–89. doi:10.1053/j.ajkd.2003.09.016. PMID:14712431.
van Vilsteren, M., de Greef, M., and Huisman, R. 2005. The effects of a low-to-
moderate intensity pre-conditioning exercise programme linked with exer-
cise counselling for sedentary haemodialysis patients in The Netherlands:
results of a randomized clinical trial. Nephrol. Dial. Transplant. 20: 141–146.
doi:10.1093/ndt/gfh560. PMID:15522901.
Weiner, D.E., Tighiouart, H., Amin, M.G., Stark, P.C., MacLeod, B., Griffith, J.L.,
et al. 2004. Chronic kidney disease as a risk factor for cardiovascular disease
and all-cause mortality: a pooled analysis of community-based studies. J. Am.
Soc. Nephrol. 15: 1307–1315. doi:10.1097/01.ASN.0000123691.46138.E2. PMID:
15100371.
Wilund, K.R., Tomayko, E.J., Wu, P., Chung, H.R., Vallurupalli, S.,
Lakshminarayanan, B., et al. 2010. Intradialytic exercise training reduces
oxidative stress and epicardial fat: a pilot study. Nephrol. Dial. Transplant.
25: 2695–2701. doi:10.1093/ndt/gfq106. PMID:20190243.
376 Appl. Physiol. Nutr. Metab. Vol. 40, 2015
Published by NRC Research Press
Appl. Physiol. Nutr. Metab. Downloaded from www.nrcresearchpress.com by University of Alberta on 08/19/15
For personal use only.

Appendix A
Fig. A1. Exercise Needs Assessment. ADL, activities of daily living; AV, atrioventricular; BMI, body mass index; BP, blood pressure; CAD, coronary
artery disease; CHF, congestive heart failure; DM, diabetes mellitus; K, potassium; HbA1c, hemoglobin A1c; Hgb, hemoglobin; MI, myocardial
infarction; PTH, parathyroid hormone; SOB, shortness of breath.
Parker et al. 377
Published by NRC Research Press
Appl. Physiol. Nutr. Metab. Downloaded from www.nrcresearchpress.com by University of Alberta on 08/19/15
For personal use only.

Fig. A1 (concluded).
378 Appl. Physiol. Nutr. Metab. Vol. 40, 2015
Published by NRC Research Press
Appl. Physiol. Nutr. Metab. Downloaded from www.nrcresearchpress.com by University of Alberta on 08/19/15
For personal use only.