Clin Exp Nephrol

DOI 10.1007/s10157-017-1429-2

ORIGINAL ARTICLE

Home-based versus center-based aerobic exercise

on cardiopulmonary performance, physical function, quality
of life and quality of sleep of overweight patients with chronic
kidney disease
Danilo Takashi Aoike1 • Flavia Baria2 • Maria Ayako Kamimura1,2 •

Adriano Ammirati1 • Lilian Cuppari1,2

Received: 17 November 2016 / Accepted: 4 June 2017

Ó Japanese Society of Nephrology 2017

Abstract observed between the exercise groups, and no changes in

Background The association between chronic kidney dis- any of the parameters investigated were found in the con-
ease (CKD) and obesity can decrease the patients’ car- trol group.
diopulmonary capacity, physical functioning and quality of Conclusion Home-based aerobic training was as effective
life. The search for effective and practical alternative as center-based training in improving the physical and
methods of exercise to engage patients in training programs functional capabilities, quality of life and sleep in over-
is of great importance. Therefore, we aimed to compare the weight NDD-CKD patients.
effects of home-based versus center-based aerobic exercise
on the cardiopulmonary and functional capacities, quality Keywords Aerobic exercise
Chronic kidney disease

of life and quality of sleep of overweight non-dialysis- Home-based exercise
Obesity
Physical function

dependent patients with CKD (NDD-CKD). Quality of sleep
Methods Forty sedentary overweight patients CKD stages
3 and 4 were randomly assigned to an exercise group
[home-based group (n = 12) or center-based exercise Introduction
group (n = 13)] or to a control group (n = 15) that did not
perform any exercise. Cardiopulmonary exercise test, Physical inactivity, a common condition observed in
functional capacity tests, quality of life, quality of sleep patients with CKD, is associated with low cardiopul-
and clinical parameters were assessed at baseline, 12 and monary and functional capacities [1, 2] and a higher risk
24 weeks. for morbidity and mortality [3, 4]. The association of
Results The VO2peak and all cardiopulmonary parameters CKD and obesity may decrease the physical capacity and,
evaluated were similarly improved (p \ 0.05) after 12 and consequently, the quality of life of these patients. Regular
24 weeks in both exercise groups. The functional capacity physical exercise plays a major role in the prevention,
tests improved during the follow-up in the home-based improvement and rehabilitation of various chronic dis-
group (p \ 0.05) and reached values similar to those eases. In the CKD population, a growing number of
obtained in the center-based group. The benefits achieved publications have, in recent years, clearly demonstrated
in both exercise groups were also reflected in improvement the benefits of exercise on several aspects of the disease
of quality of life and sleep (p \ 0.05). No differences were [5, 6]. However, although it is recommended by the
current guidelines for all stages of the disease, the
incorporation of physical exercise into standard CKD
& Lilian Cuppari treatment is a still a challenge [6, 7]. Barriers such as the
[email protected]
lack of knowledge of health professionals and patients,
1
Division of Nephrology, Federal University of São Paulo, the fear of adverse events, a lack of appropriate facilities
São Paulo, SP, Brazil and equipment to perform the exercise and the lack of
2
Nutrition Program, Federal University of São Paulo, studies showing real benefits in affordable and convenient
São Paulo, Brazil training programs have made the implementation of

123
 Clin Exp Nephrol

exercise programs even more difficult for these patients. and those with arrhythmia or myocardial ischemia detected
In other chronic diseases, home-based exercise has shown by cardiovascular stress testing were not included. No
to be effective in improving cardiopulmonary and func- patients had clinical history of other heart disease,
tional capacity of patients [8, 9]. However, as far as we including valve heart disease, peripheral arterial disease or
know there is no study that compared the impact of coronary artery disease. The patients were prescribed anti-
home-based exercise with that of center-based on physical hypertensive, diuretics, statins, sodium bicarbonate and
capacity and quality of life in NDD-CKD patients. Hence, oral hypoglycemic medications. The patients were not
in the present study, we hypothesized that a home-based engaged in any exercise program and the activity of daily
aerobic exercise program would provide similar benefits living included only personal care characterizing a seden-
as a center-based program for a group of overweight tary life style.
NDD-CKD patients. As shown in Fig. 1, 50 patients agreed to participate in
the study. Before randomization, five patients were exclu-
ded: three were due to an eGFR below 15 mL/min, and two
Subjects and methods withdrew their consent. Therefore, 45 patients were ran-
domized. Five patients were lost during the follow-up due
Subjects to social and financial issues. Thus, the present study was
completed with a total of 40 patients. The majority of the
Sedentary patients with CKD stages 3 and 4 were recruited patients were men (67.5%), with a mean age of
from the outpatient clinic of the Federal University of São 55.8 ± 8.3 years. The most frequent etiologies of CKD
Paulo—Oswaldo Ramos Foundation (São Paulo, SP, Bra- were hypertensive nephropathy (37.5%), diabetic
zil). The inclusion criteria were body mass index (BMI) nephropathy (12.5%) and glomerulonephritis (17.5%).
[25 kg/m2, age between 18 and 70 years, systolic and Overall, 13 patients had diabetes (35%). The body mass
diastolic blood pressure \180 and \100 mmHg, respec- index (BMI) was 31.2 ± 4.4 kg/m2, and in 47.5% of
tively, serum hemoglobin [11 g/dL, glycated hemoglobin patients the BMI was indicative of obesity (BMI C30 kg/
(HbA1c) \8%, and absence of chronic obstructive pul- m2). The estimated glomerular filtration rate (eGFR) was
monary disease, congestive heart failure or active coronary 26.9 ± 11.7 mL/min/1.73 m2, and 60% of the patients
disease. Patients who used beta-blockers or erythropoietin were in stage 4 of CKD.

Fig. 1 Participant flow chart

123
Clin Exp Nephrol

The Human Investigation Review Committee of Federal American College of Sports Medicine [13]. All training
University of São Paulo approved the study, and informed sessions were preceded by stretching of the large muscle
consent was obtained from each subject. The trial is reg- groups and warm-up (5 min) and ended with cooling down
istered at Clinical Trials.gov (NCT02379533). and stretching (5 min) in both exercise groups. The training
was performed for 30 min with increments of 10 min in
Study design and protocol duration every 4 weeks until week eight.

This was a 24-week, randomized controlled interventional Cardiopulmonary exercise test

trial. The patients were assigned to exercise or control
groups after a blocked randomization procedure using a The cardiopulmonary exercise test was performed on a
random block of 6 participants. The patients assigned to the treadmill and was used to determine each patient’s VO2peak.
exercise group were allowed to choose their preferred The ventilatory variables were measured using a gas ana-
mode of exercise between home- and center-based. All of lyzer (Quark PFT Cosmed 4, Rome, Italy) and were col-
the patients were instructed to follow a renal-specific diet lected using the breath-by-breath method. Before each test,
containing approximately 30 kcal/kg/day and 0.6–0.8 g/ the analyzer was calibrated with reference gases. The test
kg/day of protein. Assessments were performed at baseline, began with a fixed inclination of 1%. The initial velocity
12 and 24 weeks. The patients included in the exercise was 2 km/h during the first 3 min and increased by incre-
groups were submitted to a moderate aerobic exercise ments of 0.5 km/h every minute until the patient reached
program. To correctly perform the exercise, the patients physical exhaustion.
assigned to the home-based group initially underwent three The ventilatory threshold (VT) was determined in the
supervised exercise sessions, and they were then monitored stage preceding the first occurrence of an exponential
weekly by telephone calls and monthly by individual visits increase in ventilation, in the ventilatory equivalent for
to assess their adherence and progress and to provide oxygen (VE/VO2) and in the expired fraction of oxygen.
support. Each subject in this group received a heart rate The respiratory compensation point (RCP) is the maximum
monitor and a manual with detailed instructions on how to intensity at which there is a balance between production
perform the exercise. After each session, they recorded and removal of serum lactate, characterizing an intense
their average heart rate and perceived exertion using the workout, and was determined as the stage preceding the
Borg scale [10]. The records were collected, the instruc- second occurrence of the exponential increase in ventila-
tions were reinforced, and any adjustments to the exercise tion and in the ventilatory equivalent for carbon dioxide
workload were made during the monthly visits. The home- (VE/VCO2) and a decrease in end-tidal carbon dioxide. The
based training consisted of walking at locations near the highest VO2 obtained during the last stage reached was
patient’s home, such as a backyard, park or street, three considered the VO2peak. The data were analyzed according
times per week on alternate days. The patients from the to 20-s averages.
center-based group participated in the aerobic exercise The VO2peak was also estimated using the equation
program performed at an exercise center, using a treadmill, proposed by Bruce et al. The equation accounts for sex,
three times per week on alternate days under the supervi- age, weight and whether the individual is active or
sion of an exercise physiologist. The patients assigned to sedentary [14].
the control group were provided with usual care and were
advised not to engage in any exercise training program. Functional capacity tests

Aerobic training prescription Functional capacity was assessed by a variety of objective

measures. These included a 6-min walk test (maximal
The exercise training intensity was prescribed according to distance walked along an internal corridor during 6 min; to
each patient’s ventilatory threshold (VT) obtained during evaluate the aerobic power), 2-min step test (maximal
the cardiopulmonary exercise test. The VT is characterized number of steps achieved in stationary walking during
by the highest intensity of physical exertion that was fully 2 min; to assess the aerobic power), sit-to-stand test
maintained by aerobic energy pathways, was considered a (maximal sit-to-stand cycles achieved in 30 s; to assess the
marker of exercise consistent with mild-to-moderate muscular endurance of the legs), arm curl test (maximal
intensity and was usually between 40 and 60% of the number of arm curl cycles in 30 s; to assess the muscular
maximum VO2 [11, 12]. The control of intensity was endurance of the arms), sit-and-reach test (maximal dis-
maintained by the heart rate value obtained at VT, using a tance achieved in the Wells bench; to assess general flex-
Polar FS-1 heart rate monitor. The training program was ibility) and timed up-and-go test (shorter time to rise from a
conducted in accordance with the recommendation of the chair, walk 3 m and sit back; to assess the functional

123
 Clin Exp Nephrol

mobility). To minimize the effect of learning, the patients Data are presented as the mean and standard deviation,
completed a pre-test before the assessment. All of the tests median and 95% confidence interval or proportions on the
were applied in accordance with the methods described by basis of the distribution of the variables. Skewed variables
Rikli [15]. were log-transformed. The Chi-square test was used to
compare proportions. A General Linear Model test for
Quality of life assessment repeated measures was used to compare variables. Corre-
lation coefficients were calculated according to Pearson
The Short-Form Health Survey (SF-36) questionnaire was analysis. All tests were 2-sided, and statistical significance
administered to assess the participants’ quality of life. The for all analyses was established at p values \0.05. The
scores for each domain range from 0 to 100%. Higher SPSSÒ software version 18.0 for Windows (SPSS Inc.,
scores indicated a better quality of life [16]. Chicago, IL, USA) was used for the analysis.

Sleep assessment
Results
The Pittsburgh Sleep Quality Index (PSQI) was used to
assess the quality of sleep and sleep disturbances. The Forty patients completed the follow-up of 24 weeks. As
PSQI assesses the quality of sleep for the last month prior seen in Table 1, at baseline, no differences were observed
to the interview. The questionnaire has 19 questions among the groups regarding sex, age, BMI, eGFR, blood
comprising seven components: quality of sleep, latency, pressure, presence of diabetes, or etiology of CKD. BMI
duration, efficiency, nocturnal sleep disturbances, use of and eGFR did not change during the follow-up. The sys-
sleep medication, and daytime sleepiness. Each component tolic and diastolic blood pressures were significantly
receives a score from zero to three; thus, the final score reduced in both exercise groups after 12 and 24 weeks of
ranges between zero and 21. The higher the score is, the training (Table 1). No change was observed in the blood
worse the quality of sleep is: scores higher than five are pressure in the control group. A significant group-by-time
indicative of sleep disturbances [17]. interaction was observed for systolic and diastolic blood
All questionnaires were individually administered by the pressure in both exercise groups at 24 weeks. No changes
same observer in a quiet room with the patient rested. in type or dose of anti-hypertensive medication were nec-
essary during the follow-up. At baseline no differences
Laboratory data were observed between groups in the glycated hemoglobin,
urinary sodium and urinary protein. At 24 weeks urinary
Blood samples were drawn after an overnight fast of at sodium increased in the home-based group and did not
least 12 h. Serum creatinine was determined. The change in center-based or control group. No changes were
glomerular filtration rate (eGFR) was estimated using the observed in glycated hemoglobin and urinary protein dur-
CKD-EPI creatinine equation [18]. ing the follow-up. A significant group-by-time interaction
was observed for glycated hemoglobin in both exercised
Blood pressure assessment groups at 24 weeks (Table 1).
Table 2 shows that the groups were similar in their
Blood pressure was assessed during the week of evaluation cardiopulmonary parameters and functional capacity test
with the patient in a sitting position after 10 min of rest. results at baseline. At 12 weeks, except for VO2 at VT,
The average of three readings performed on alternate days heart rate at VT and at RCP, Borg at VT and at RCP, all of
was used to determine the systolic and diastolic blood the other cardiopulmonary parameters increased in both
pressures. exercise groups and were maintained or increased further
after 24 weeks (Fig. 2a, b). At baseline, the measured
Statistical analysis VO2peak was approximately 20% lower than the estimated
VO2 in all groups. After 12 and 24 weeks of follow-up, the
The sample size was calculated considering a mean patients in both exercise groups, but not those in the control
increase of 11% in the VO2peak within the group [19]. On group, achieved the level of estimated VO2 for sedentary
the basis of repeated measures analysis, a total of 14 healthy individuals according to sex, age and body weight.
patients for each group was calculated to ensure a power Regarding the functional capacity test results, the timed
of 80%, a dropout of 15% and p value of \0.05. The up-and-go test, 2-min step test and arm curl test improved
Gpower software version 3.1.2. (Franz Faul, University significantly in both exercise groups at 12 weeks and after
of Kiel, Germany) was used to determine the sample 24 weeks compared with the baseline results. The sit-and-
size. reach test improved only in the home-based group, and the

123
 Clin Exp Nephrol

Table 1 Baseline demographic and clinical characteristics of the patients according to groups
Variables Control (n = 15) Center-based (n = 13) Home-based (n = 12) p
Baseline 12 weeks 24 weeks Baseline 12 weeks 24 weeks Baseline 12 weeks 24 weeks

Sex (M%) 10 (66.6) 9 (69.2) 8 (66.6) 0.987

Age (years) 54.3 ± 8.7 56.3 ± 7.9 56.0 ± 8.3 0.883
DM (%) 5 (33.3) 4 (30.7) 4 (33.3) 0.988
2 a
eGFR (mL/min/1.73 m ) 25.3 ± 13.5 23.9 ± 12.2 24.1 ± 13.0 26.9 ± 9.9 29.5 ± 11.8 27.8 ± 11.0 28.5 ± 11.8 32.3 ± 14.7 30.9 ± 15.0
BMI (kg/m2) 30.7 ± 4.1 30.8 ± 4.1 31.5 ± 4.8 31.8 ± 4.5 31.4 ± 4.7 31.2 ± 5.2 31.1 ± 4.6 30.9 ± 4.1 30.4 ± 3.4
SBP (mmHg) 130.6 ± 11.1 126.8 ± 6.7 130.4 ± 7.2 127.6 ± 10.3 115.9 ± 6.6a 114.9 ± 8.1a,c 134.3 ± 12.6 117.9 ± 7.6a 120.3 ± 6.3a,c
a a,c a
DBP (mmHg) 81.9 ± 7.3 81.0 ± 3.6 83.8 ± 6.7 80.8 ± 7.1 74.6 ± 6.3 72.8 ± 6.1 83.2 ± 89.0 75.4 ± 4.5 76.9 ± 3.6a,c
Etiology of CKD 0.143
Hypertensive nephropathy 5 4 6
Diabetic nephropathy 2 3 0
Glomerulonephritis 4 1 2
Undetermined/other 4 5 4
HbA1c (%) 5.8 ± 0.8 6.0 ± 1.0 6.1 ± 1.0 6.2 ± 0.7 6.0 ± 0.5 5.9 ± 0.6c 6.3 ± 1.1 6.1 ± 1.0 6.0 ± 0.8c
Urinary sodium (mEq/24 h) 207.0 ± 110 202.0 ± 61 215.3 ± 83 205.5 ± 81 232.0 ± 116 236.5 ± 94 175.3 ± 45 215.0 ± 76 226.0 ± 84a
Urinary protein (g/24 h) 1.6 ± 1.6 1.5 ± 1.4 1.4 ± 1.4 1.1 ± 2.7 1.1 ± 2.9 1.4 ± 3.9 1.5 ± 1.4 2.7 ± 3.1 2.0 ± 1.7
Data are presented as the mean ± SD, or frequency (%)
DM diabetes mellitus, eGFR estimated glomerular filtration rate, BMI body mass index, SBP systolic blood pressure, DBP diastolic blood pressure
a
p \ 0.05 versus baseline
b
p \ 0.05 center-based or home-based versus control group
c
p \ 0.05 interaction group 9 time, control group versus center-based or home-based

123
123
Table 2 Cardiopulmonary parameters and functional capacity tests, baseline, 12 and 24 weeks of follow-up
Variable Control (n = 15) Center-based (n = 13) Home-based (n = 12)
Cardiopulmonary parameters Baseline 12 weeks 24 weeks Baseline 12 weeks 24 weeks Baseline 12 weeks 24 weeks

VO2peak (mL/kg/min) 23.6 ± 8.2 24.2 ± 7.1 22.9 ± 7.1 22.8 ± 4.3 26.1 ± 5.5a 28.2 ± 7.9a,c 23.4 ± 6.6 25.6 ± 5.9a 25.7 ± 5.8a,c
a,c a,c a,c
Maximal ventilation (L/min) 80.1 ± 29.4 76.7 ± 23.3 77.3 ± 27.2 81.1 ± 19.4 93.3 ± 16.8 94.7 ± 27.4 76.0 ± 25.5 91.1 ± 28.3 92.5 ± 27.1a,c
Speed at VO2peak (km/h) 7.5 ± 1.5 7.3 ± 1.7 7.2 ± 1.6 7.0 ± 1.3 8.4 ± 1.4a,c 9.0 ± 1.7a,b,c 7.0 ± 1.3 7.8 ± 1.4a,c 8.6 ± 1.8a,b,c
a,c
VO2 at VT (mL/kg/min) 15.6 ± 4.7 15.1 ± 4.3 14.8 ± 4.1 15.3 ± 3.0 15.9 ± 3.7 17.4 ± 4.2 14.7 ± 2.7 16.7 ± 3.8 17.4 ± 3.9a,c
b a,c a,b,c a,c
Speed at VT (km/h) 4.7 ± 1.0 4.4 ± 0.9 4.4 ± 0.9 4.4 ± 0.9 5.6 ± 1.0 5.8 ± 1.2 4.8 ± 0.7 5.6 ± 1.0 5.8 ± 1.3a,b,c
Heart rate at VT (bpm) 114.7 ± 13.2 109.1 ± 13.8 108.1 ± 15.2 114.9 ± 11.1 109.0 ± 11.3 115.6 ± 10.1 111.8 ± 8.3 118.0 ± 12.6 123.6 ± 15.8a,c
Borg at VT 11.8 ± 2.9 11.3 ± 3.7 12.7 ± 3.2 10.7 ± 3.0 9.6 ± 1.5 9.5 ± 2.2 10.75 ± 2.5 11.08 ± 2.1 12.0 ± 2.6
VO2 at RCP (mL/kg/min) 18.9 ± 5.9 19.0 ± 5.6 17.3 ± 4.9 17.6 ± 3.2 21.0 ± 4.0a 23.5 ± 6.7a,c 18.4 ± 4.4 21.5 ± 5.3a 21.9 ± 5.6a,c
a,c a,b,c a,c
Speed at RCP (km/h) 6.2 ± 1.0 5.8 ± 1.0 5.6 ± 1.0 5.5 ± 1.1 6.8 ± 1.1 7.5 ± 1.7 5.8 ± 0.9 6.8 ± 1.0 7.4 ± 1.5a,b,c
a,c a
Heart rate at RCP (bpm) 133.0 ± 14.3 128.5 ± 15.3 124.3 ± 18.7 127.2 ± 14.8 132.2 ± 16.5 143.7 ± 183.3 127.8 ± 15.8 139.8 ± 15.1 149.2 ± 18.2a,c
Borg at RCP 15.73 ± 3.4 14.7 ± 3.6 15.4 ± 3.1 13.3 ± 3.3 12.9 ± 3.1 14.5 ± 3.2 13.3 ± 3.0 15.0 ± 2.3 16.2 ± 3.6a
d d d d a,c a,c d a,c
VO2peak/estimated VO2 (%) 81.6 ± 17.7 83.3 ± 9.9 79.4 ± 14.1 84.3 ± 21.6 96.6 ± 26.5 101.1 ± 22.9 83.2 ± 12.2 91.9 ± 15.2 92.6 ± 12.4a,c
Functional capacity tests
6-min walk (m) 546.5 ± 74.9 561.2 ± 91.2 536.8 ± 80.9 528.8 ± 98.4 600.1 ± 96.4a,c 634.0 ± 97.9a,b,c 526.5 ± 81.9 579.9 ± 81.4a,c 632.3 ± 95.5a,b,c
a a,c a,c a,c
Timed up and go (s) 6.0 ± 8.1 6.4 ± 1.1 6.5 ± 1.2 6.7 ± 1.3 5.6 ± 1.4 5.6 ± 1.3 6.5 ± 1.8 5.8 ± 1.4 5.5 ± 1.6a,c
a,c a,b,c a,c
2-min step (step) 189.5 ± 33.8 179.9 ± 36.3 171.0 ± 35.9 189.1 ± 24.2 239.3 ± 25.6 240.5 ± 38.9 176.2 ± 32.6 218.8 ± 31.4 233.8 ± 28.3a,b,c
a,b,c a,b,c a,b,c
Sit and stand (repetition) 18.0 ± 4.3 18.3 ± 4.8 17.7 ± 4.4 17.5 ± 3.6 23.8 ± 5.2 26.1 ± 5.7 16.6 ± 4.2 23.8 ± 6.9 25.4 ± 6.5a,b,c
b a,b a,b
Sit and reach (cm) 13.5 ± 11.8 14.3 ± 11.8 14.8 ± 12.0 19.4 ± 6.3 22.8 ± 9.6 22.9 ± 11.1 21.9 ± 5.9 24.5 ± 5.9 27.5 ± 4.8a,b
Arm curl (repetition) 18.6 ± 3.0 18.1 ± 3.1 17.5 ± 3.4 19.08 ± 3.3 23.2 ± 3.6a,b,c 24.85 ± 3.3a,b,c 17.3 ± 2.5 22.8 ± 4.7a,b,c 24.4 ± 4.8a,b,c
Data are presented as the mean ± SD
VT ventilatory threshold, RCP respiratory compensation point
a
p \ 0.05 versus baseline
b
p \ 0.05 center-based or home-based versus control group
c
p \ 0.05 interaction group 9 time, control group versus center-based or home-based
d
p \ 0.05 VO2peak versus estimated VO2
Clin Exp Nephrol
Clin Exp Nephrol

Fig. 2 Changes in speed at VO2peak (a), speed at respiratory compensation point (b), 6-min walk test (c) and sit-and-stand test (d). Data are
presented as the means and 95% CI. ap \ 0.05—versus control group, bp \ 0.05—12 versus 24 weeks

stand-and-reach test improved only in the center group at based and center-based groups. All of the parameters were
12 weeks, but both tests were increased after 24 weeks in unchanged in the control group.
both exercise groups compared to baseline (Table 2). The Figure 3 shows the results obtained in the assessments
6-min walk test and the sit-and-stand test improved sig- of quality of life (SF-36) and quality of sleep (PSQI) in the
nificantly after 12 weeks and were further increased after study groups. At baseline, the SF-36 score was higher in
24 weeks in both exercise groups (Fig. 2c, d). In the both exercise groups compared with the control group. No
exercise groups, a significant group-by-time interaction differences were observed in the PSQI score among the
was observed for all of the cardiopulmonary parameters groups at baseline. After 12 weeks of training, both exer-
and for the majority of functional capacity tests (Table 2). cise groups showed an improvement in the SF-36 total
The changes in the cardiopulmonary parameters and in the score, and this improvement was maintained after
functional capacity were not different between the home- 24 weeks (Fig. 3a). Among the domains of SF-36, physical

123
 Clin Exp Nephrol

Discussion

This study aimed to compare the impact of home-based

versus center-based aerobic training in overweight NDD-
CKD patients. We showed that this alternative method of
training promoted similar benefits to those found in center-
based training with respect to the cardiorespiratory fitness,
functional capacity, blood pressure, quality of life and
quality of sleep of the patients.
As far as we know, no study has compared these two
approaches of exercise in NDD-CKD patients. When
investigated separately in studies with CKD patients, both
methods have been shown to be efficient in promoting
benefits in the cardiorespiratory fitness and these benefits
were determined mostly by improvement in the VO2peak
[19–22]. VO2peak is considered the gold standard and is the
most widely used parameter for evaluating cardiorespira-
tory capacity. In CKD, VO2peak has also shown to be an
independent predictor of mortality [3, 23]. In the present
study, the VO2peak obtained from our patients was, on
average, 20% below the estimated value of healthy
sedentary individuals [14], thus indicating a high degree of
sedentary lifestyle, which has been observed in several
studies of CKD patients [24, 25]. In addition to the reduced
physical activity, various disorders resulting from
decreased kidney function, such as anemia, metabolic
acidosis, chronic inflammation, and muscle and bone
metabolism disturbances, have been implicated in such low
physical performance [26, 27]. The combination of
sedentary lifestyle and CKD-related disturbances may
result in less efficient energy production by aerobic
Fig. 3 SF-36 total score (a) and PSQI score (b). Data are presented
as the means and 95% CI. ap \ 0.05 versus baseline, bp \ 0.05 mechanisms, thus facilitating the occurrence of premature
center-based or home-based versus control group, cp \ 0.05 interac- muscle fatigue, and in a lower tolerance for both exercise
tion group 9 time, control group versus center-based or home-based and everyday life activities [28, 29]. Despite the deficit in
the cardiorespiratory fitness observed, the results achieved
functioning, bodily pain, general health perception and in the current study indicated that the patients were able to
vitality improved in the center-based group. In the home- respond adequately to the exercise because virtually all of
based group, except for social functioning, all other the cardiopulmonary parameters measured improved in
domains were improved. No changes were found in the both exercise groups. It is worth mentioning that the mean
control group (data not shown). Improvement in the PSQI VO2peak achieved after 24 weeks of training reached the
score was observed only after 24 weeks (Fig. 3b). Among value predicted for healthy sedentary subjects in both
the components assessed, sleep latency and daytime exercise groups. It should also be noted that the greater
sleepiness were the components that improved in both magnitude of improvement in the VO2peak occurred in the
exercise groups. Both components contributed significantly first 12 weeks of training, with maintenance and no further
for the improvement in final score of PSQI (data not improvement after 24 weeks. This result probably occurred
shown). The quality of life and quality of sleep scores due to the difficulty in increasing the patients’ walking
remained unchanged in the control group. At 24 weeks, a speed; otherwise, it would have been necessary for them to
significant group-by-time interaction was observed for the start running, which they were unable to do. In the center-
SF-36 score and PSQI score in both exercise groups. based group, it was possible to compensate for this diffi-
No cardiovascular, muscular or metabolic adverse events culty by increasing the inclination of the treadmill, which
were observed during the follow-up of the study, and no did not require the patient to increase the rhythm of their
changes in the doses or types of medication were necessary. gait or to run. This alternative could not be employed for

123
Clin Exp Nephrol

patients in the home-based group; nevertheless, after fol- undergoing dialysis [34–36]. The limited existing data for
low-up, no differences were observed between the groups. NDD-CKD patients have shown that these patients have
The improvements in the VO2peak were accompanied by better quality of life scores than do patients on dialysis,
an increase in the speed achieved at VT and at RCP, thus possibly due to the formers’ better clinical condition and
indicating a greater metabolic efficiency at the same less impaired physical capacity [21, 33, 37]. The scores
workload. Such exercise markers have been poorly inves- obtained in the present study indicated the low impact of
tigated in CKD patients. Short-term studies with NDD- the disease on the quality of life of the patients. Despite
CKD patients who underwent aerobic training reported baseline differences in the perception of quality of life
increases of approximately 15 and 10% in oxygen con- between the exercise groups and the control group, the
sumption at VT and RCP, respectively [19, 30]. In the groups were not different in their clinical and physical
present study, after 24 weeks of training, the oxygen con- conditions. After 24 weeks of follow-up, we observed that
sumption at VT increased 18 and 14%, and at RCP, it the control group maintained the same score as baseline,
increased 19 and 33% in the home- and center-based but that the exercise groups showed significant improve-
groups, respectively. More importantly, the workload at ment in their total score.
those exercise intensities increased 21 and 32% at VT and In addition to the quality of life, we also subjectively
28 and 36% at RCP in the home-based and center-based evaluated the patients’ sleep quality. Insufficient hours of
groups, respectively. These results can be interpreted as an sleep (\6 h per night), insomnia, restless legs syndrome,
improvement in the metabolic efficiency to sustain a sleep apnea and excessive nighttime awakenings are all
greater workload that may provide a better physical toler- indicators of low quality of sleep [38]. These sleep disorders
ance for daily activities. Most important was the finding are frequently reported by patients with CKD and can
that the speed achieved at VT after the period of training become worse as the disease progresses, possibly due to the
was very similar to that achieved at RCP before training. deleterious effect of uremic toxins in the central nervous
This means that the baseline workload, which could only system [38–40]. Sleep disorders may contribute to the onset
be achieved with a large contribution of anaerobic meta- or exacerbation of important risk factors for progression of
bolism (RCP), was reached almost exclusively by aerobic CKD, such as hypertension, type 2 diabetes and obesity
metabolism (VT) after 24 weeks of training. [41, 42]. Studies in the general population have shown that
The increase in the distance achieved in the 6-min walk moderate aerobic exercise improves the architecture of slow
test and in the number of steps in the 2-min walk test in wave sleep and increases the total sleep time with reduced
both trained groups after 12 and 24 weeks also supported latency sleep onset and reductions in the frequency and
the finding of an improvement in the cardiorespiratory duration of nocturnal awakenings [43, 44]. The assessment
capacity. The average increase of 20% in the 6-min walk of quality of sleep in the present study was performed using
test reached the reference value of sedentary healthy the Pittsburgh Sleep Quality Index questionnaire, which is a
individuals with normal renal function [31, 32]. validated tool that is widely used in research about sleep
In addition to the aerobic capacity, other capabilities quality and shows good agreement with polysomnography,
were also tested in the current study. This provided a which is the gold standard method of assessing sleep. Scores
broader overview of not only the level of physical fitness of greater than five are indicative of poor-quality sleep. When
the patients, but also the benefits obtained from improved we considered this threshold score, the majority of our
aerobic capacity. Indeed, a significant improvement in the patients (65%) were classified as having a poor sleep quality
anaerobic characteristic tests, such as the sit-and-stand test at baseline. This finding is in accordance with other studies
(50% in both exercise groups), arm curl test (40% in home- with CKD patients that evaluated sleep quality using the
based and 30% in center-based group) and timed up-and-go PSQI [45–47]. As far as we know, no study has evaluated
test (15% in both exercise groups) were observed in the the impact of exercise on sleep quality in CKD patients. Our
exercise patients. These findings suggested that the aerobic data indicated that both aerobic training methods promoted
training may have resulted in benefits in other capabilities, improvements in the patient́s quality of sleep, with a
possibly by promoting a better interaction between the decrease in the score of approximately 46% in the home-
nervous and muscular systems. This possibility, however, and center-based exercise groups, whereas no change
needs to be investigated further. occurred in the control group. Although additional studies in
Quality of life is an important marker of how a disease is this area are still necessary in CKD patients, these results
affecting the patient’s life, and it has been used to evaluate suggest that the benefits achieved with physical exercise are
the effectiveness of an intervention [33]. A number of not limited to physical capacity.
studies have shown the important negative impact of CKD Studies investigating the effectiveness of home-based
on quality of life, especially in the physical aspects of exercise in comparison with center-based training in terms
patients’ lives and, more particularly, in those patients of improvements in physical capacity in patients with

123
 Clin Exp Nephrol

chronic diseases are scarce. In a group of patients with addition, we investigated a number of physical capacity
chronic heart failure submitted to an exercise program, both markers other than VO2peak to compare the two modalities
methods were equally effective in improving the physical of exercise. Furthermore, our results demonstrated that
condition of patients [48]. To our knowledge, there is only aerobic exercise provided physical and clinical improve-
one study with CKD patients that compared home-based and ment in overweight patients, even without the benefit of
center-based training. Patients undergoing hemodialysis body weight reduction.
were submitted to center-based training on interdialytic In conclusion, the home-based aerobic training was as
days, center-based training during hemodialysis or home- effective as center-based exercise in improving the physi-
based training on the interdialytic days. After the follow-up cal and functional capabilities of overweight CKD patients
period of 24 weeks, all groups showed significant and resulted in improvements in their clinical condition,
improvements in their aerobic capacity, with a slightly better quality of life and quality of sleep. Therefore, this aerobic
result for the center-based interdialytic day group. However, exercise method may constitute a safe and practical method
because the training protocols were quite different among to promote the benefits of exercise for overweight patients
the groups, it was not possible to conclude that the home- with NDD-CKD.
based group was inferior to the center-based group in terms
of improvements in physical capacity [49]. In the present Acknowledgements This study was supported by São Paulo
Research Foundation (FAPESP) (2009/14786-0), Coordination for the
study, both groups were exposed to the same training pro- Improvement of Higher Education Personnel (CAPES), Oswaldo
tocol, including the intensity, duration, progression and Ramos Foundation, Psychobiology and Exercise Study Centre
number of sessions, so the comparison is less biased. (CEPE) and the Research Incentive Fund Association (AFIP).
From a clinical point of view, an important benefit that
Compliance with ethical standards
was found after 12 weeks and maintained after 24 weeks of
training was the reduction in systolic and diastolic blood Conflict of interest The authors have no conflicts of interest to
pressures in both of the groups that participated in physical disclose.
exercise. It is important to highlight that the decrease in
blood pressure occurred even in the absence of changes in Ethical statement All procedures performed were in accordance
with the ethical standards of the Human Investigation Review Com-
anti-hypertensive medications, decrease of sodium intake mittee of Federal University of São Paulo that approved the study.
(estimated by 24 h urinary sodium) or body weight. The
decrease in blood pressure with mild-to-moderate aerobic Informed consent Informed consent was obtained from all individ-
training has already been well established in hypertensive ual participants included in the study.
subjects [50, 51] and in patients with NDD-CKD
[19, 52, 53]. The effect of aerobic exercise on blood
pressure seems to be mediated by a reduction in vascular
References
resistance due to improvements in endothelial function,
reduced activity of the sympathetic nervous system and 1. Hsieh RL, Lee WC, Huang HY, Chang CH. Quality of life and its
vascular remodeling [50]. correlates in ambulatory hemodialysis patients. J Nephrol.
We acknowledge some limitations of the present study. 2007;20(6):731–8.
2. Kouidi EJ. Central and peripheral adaptations to physical training
The relatively small sample size of clinically stable and in patients with end-stage renal disease. Sports Med.
motivated overweight NDD-CKD patients may limit the 2001;31(9):651–65.
generalizability of our findings to other patients with CKD. 3. Sietsema KE, Amato A, Adler SG, Brass EP. Exercise capacity as
Despite the small sample the changes were of such a degree a predictor of survival among ambulatory patients with end-stage
renal disease. Kidney Int. 2004;65(2):719–24. doi:10.1111/j.
that the statistical power observed for the majority of
1523-1755.2004.00411.x.
parameters was greater than 80% indicating that sample 4. Knight EL, Ofsthun N, Teng M, Lazarus JM, Curhan GC. The
size was sufficient to reproducibility of the observed association between mental health, physical function, and
results. Furthermore, the fact that the patients were allowed hemodialysis mortality. Kidney Int. 2003;63(5):1843–51. doi:10.
1046/j.1523-1755.2003.00931.x.
to choose between the exercise modes (center- or home- 5. Segura-Orti E, Johansen KL. Exercise in end-stage renal disease.
based) might have resulted in a selection bias. However, it Semin Dial. 2010;23(4):422–30. doi:10.1111/j.1525-139X.2010.
has been suggested that the possibility of choosing between 00766.x.
modes of exercise can be advantageous in terms of 6. Gould DW, Graham-Brown MP, Watson EL, Viana JL, Smith
AC. Physiological benefits of exercise in pre-dialysis chronic
adherence to the protocol [54]. In addition, the baseline kidney disease. Nephrology (Carlton). 2014;19(9):519–27.
characteristics of the exercise groups were similar. doi:10.1111/nep.12285.
The strengths of our study should also be mentioned. 7. KDIGO. KDIGO 2012 clinical practice guidelines for the eval-
We used a cardiopulmonary exercise test, which is the gold uation and management of CHronic kidney disease. Kidney Int
Suppl. 2013;3(1):1–150. doi:10.1038/kisup.2012.73.
standard for measuring cardiorespiratory fitness. In

123
Clin Exp Nephrol

8. Manfredini F, Malagoni AM, Mascoli F, Mandini S, Taddia MC, 25. Avesani CM, Trolonge S, Fau-Deleaval P, Deleaval P, Fau-Baria
Basaglia N, et al. Training rather than walking: the test in -train F, Baria F, Fau-Mafra D, Mafra D, Fau-Faxen-Irving G, Faxen-
out program for home-based rehabilitation in peripheral arteri- Irving G, Fau-Chauveau P, et al. Physical activity and energy
opathy. Circ J. 2008;72(6):946–52. expenditure in haemodialysis patients: an international survey.
9. du Moulin M, Taube K, Wegscheider K, Behnke M, van den Bussche Nephrol Dial Transplant. 2012;27:1460–2385. doi:10.1093/ndt/
H. Home-based exercise training as maintenance after outpatient gfr692.
pulmonary rehabilitation. Respiration. 2009;77(2):139–45. doi:10. 26. Johansen KL. Exercise in the end-stage renal disease population.
1159/000150315. J Am Soc Nephrol. 2007;18(6):1845–54. doi:10.1681/asn.
10. Borg E, Kaijser L. A comparison between three rating scales for 2007010009.
perceived exertion and two different work tests. Scand J Med Sci 27. Clyne N, Jogestrand T, Lins LE, Pehrsson SK, Ekelund LG.
Sports. 2006;16(1):57–69. doi:10.1111/j.1600-0838.2005.00448.x. Factors limiting physical working capacity in predialytic uraemic
11. Zhang JG, Ohta T, Ishikawa-Takata K, Tabata I, Miyashita M. patients. Acta Med Scand. 1987;222(2):183–90.
Effects of daily activity recorded by pedometer on peak oxygen 28. Dipp T. Força muscular respiratória e capacidade funcional na
consumption (VO2peak), ventilatory threshold and leg extension insuficiência renal terminal. Rev Bras Med Esporte.
power in 30- to 69-year-old Japanese without exercise habit. Eur 2010;16(4):246–9.
J Appl Physiol. 2003;90(1–2):109–13. doi:10.1007/s00421-003- 29. Cury JL, Brunetto AF, Aydos RD. Efeitos negativos da insu-
0860-0. ficiência renal crônica sobre a função pulmonar e a capacidade
12. Bhambhani Y, Singh M. Ventilatory thresholds during a graded funcional. Rev Bras Fisioter. 2010;14(2):91–8.
exercise test. Respiration. 1985;47(2):120–8. 30. Toyama K, Sugiyama S, Oka H, Sumida H, Ogawa H. Exercise
13. American College of Sports Medicine Position Stand. The rec- therapy correlates with improving renal function through modi-
ommended quantity and quality of exercise for developing and fying lipid metabolism in patients with cardiovascular disease
maintaining cardiorespiratory and muscular fitness, and flexibility and chronic kidney disease. J Cardiol. 2010;56(2):142–6. doi:10.
in healthy adults. Med Sci Sports Exerc. 1998;30(6):975–91. 1016/j.jjcc.2010.06.007.
14. Bruce RA, Kusumi F, Hosmer D. Maximal oxygen intake and 31. Chetta A, Zanini A, Pisi G, Aiello M, Tzani P, Neri M, et al.
nomographic assessment of functional aerobic impairment in Reference values for the 6-min walk test in healthy subjects
cardiovascular disease. Am Heart J. 1973;85(4):546–62. 20–50 years old. Respir Med. 2006;100(9):1573–8. doi:10.1016/
15. Rikli RE, Jones C. Development and validation of functional j.rmed.2006.01.001.
fitness test for community-residing older adults. J Aging Phys 32. Camarri B, Eastwood PR, Cecins NM, Thompson PJ, Jenkins S.
Act. 1999;6:127–59. Six minute walk distance in healthy subjects aged 55–75 years.
16. McHorney CA, Ware JE Jr, Raczek AE. The MOS 36-Item Short- Respir Med. 2006;100(4):658–65. doi:10.1016/j.rmed.2005.08.
Form Health Survey (SF-36): II. Psychometric and clinical tests 003.
of validity in measuring physical and mental health constructs. 33. Perlman RL, Finkelstein FO, Liu L, Roys E, Kiser M, Eisele G,
Med Care. 1993;31(3):247–63. et al. Quality of life in chronic kidney disease (CKD): a cross-
17. Buysse DJ, Reynolds CF, 3rd, Monk TH, Berman SR, Kupfer DJ. sectional analysis in the Renal Research Institute-CKD study. Am
The Pittsburgh Sleep Quality Index: a new instrument for psy- J Kidney Dis. 2005;45(4):658–66.
chiatric practice and research. Psychiatry Res. 1989;28(2): 34. Walters BA, Hays RD, Spritzer KL, Fridman M, Carter WB.
193–213. Health-related quality of life, depressive symptoms, anemia, and
18. Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF 3rd, malnutrition at hemodialysis initiation. Am J Kidney Dis.
Feldman HI, et al. A new equation to estimate glomerular fil- 2002;40(6):1185–94. doi:10.1053/ajkd.2002.36879.
tration rate. Ann Intern Med. 2009;150(9):604–12. 35. Nonoyama ML, Brooks D, Ponikvar A, Jassal SV, Kontos P,
19. Boyce ML, Robergs RA, Avasthi PS, Roldan C, Foster A, Devins GM, Heck C, Laprade J, Naglie G. Exercise program to
Montner P et al. Exercise training by individuals with predialysis enhance physical performance and quality of life of older
renal failure: cardiorespiratory endurance, hypertension, and hemodialysis patients: a feasibility study. Int Urol Nephrol.
renal function. Am J Kidney Dis. 1997;30(2):180–92. doi:10. 2010;42(4):1125–30. doi:10.1007/s11255-010-9718-7.
1016/S0272-6386(97)90051-2. 36. Rebollo-Rubio A, Morales-Asencio JM, Pons-Raventos ME,
20. Eidemak I, Haaber AB, Feldt-Rasmussen B, Kanstrup IL, Mansilla-Francisco JJ. Review of studies on health related quality
Strandgaard S. Exercise training and the progression of chronic of life in patients with advanced chronic kidney disease in Spain.
renal failure. Nephron. 1997;75(1):36–40. Nefrologia. 2015;35(1):92–109. doi:10.3265/Nefrologia.pre2014.
21. Mustata S, Groeneveld S, Davidson W, Ford G, Kiland K, Manns Jul.12133.
B. Effects of exercise training on physical impairment, arterial 37. Headley S, Germain M, Mailloux P, Mulhern J, Ashworth B,
stiffness and health-related quality of life in patients with chronic Burris J et al. Resistance training improves strength and func-
kidney disease: a pilot study. Int Urol Nephrol. tional measures in patients with end-stage renal disease. Am J
2011;43(4):1133–41. doi:10.1007/s11255-010-9823-7. Kidney Dis. 2002;40(2):355–64. doi:10.1053/ajkd.2002.34520.
22. Aoike DT, Baria F, Rocha ML, Kamimura MA, Mello MT, Tufik 38. Aoike DT, Baria F, Kamimura MA, Ammirati A, de Mello MT,
S, et al. Impact of training at ventilatory threshold on car- Cuppari L. Impact of home-based aerobic exercise on the physical
diopulmonary and functional capacity in overweight patients with capacity of overweight patients with chronic kidney disease. Int Urol
chronic kidney disease. J Bras Nefrol. 2012;34(2):139–47. Nephrol. 2015;47(2):359–67. doi:10.1007/s11255-014-0894-8.
23. Roshanravan B, Robinson-Cohen C, Patel KV, Ayers E, Littman 39. Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The
AJ, de Boer IH, et al. Association between physical performance occurrence of sleep-disordered breathing among middle-aged
and all-cause mortality in CKD. J Am Soc Nephrol. adults. N Engl J Med. 1993;328(17):1230–5. doi:10.1056/
2013;24(5):822–30. doi:10.1681/asn.2012070702. nejm199304293281704.
24. Sietsema KE, Hiatt WR, Esler A, Adler S, Amato A, Brass EP. 40. Hirotsu C, Tufik S, Bergamaschi CT, Tenorio NM, Araujo P,
Clinical and demographic predictors of exercise capacity in end- Andersen ML. Sleep pattern in an experimental model of chronic
stage renal disease. Am J Kidney Dis. 2002;39(1):76–85. doi:10. kidney disease. Am J Physiol Renal Physiol.
1053/ajkd.2002.29884. 2010;299(6):F1379–88. doi:10.1152/ajprenal.00118.2010.

123
 Clin Exp Nephrol

41. Sakkas GK, Gourgoulianis KI, Karatzaferi C, Liakopoulos V, Database Syst Rev. 2015;8:CD007130. doi:10.1002/14651858.
Maridaki MD, Pastaka C, et al. Haemodialysis patients with sleep CD007130.pub3.
apnoea syndrome experience increased central adiposity and 49. Konstantinidou E, Koukouvou G, Kouidi E, Deligiannis A,
altered muscular composition and functionality. Nephrol Dial Tourkantonis A. Exercise training in patients with end-stage renal
Transpl. 2008;23(1):336–44. doi:10.1093/ndt/gfm559. disease on hemodialysis: comparison of three rehabilitation pro-
42. Turek NF, Ricardo AC, Lash JP. Sleep disturbances as nontra- grams. J Rehabil Med. 2002;34(1):40–5.
ditional risk factors for development and progression of CKD: 50. Pescatello LS, Franklin BA, Fagard R, Farquhar WB, Kelley GA,
review of the evidence. Am J Kidney Dis. 2012;60(5):823–33. Ray CA. American College of Sports Medicine position stand.
doi:10.1053/j.ajkd.2012.04.027. Exercise and hypertension. Med Sci Sports Exerc.
43. Iftikhar IH, Kline CE, Youngstedt SD. Effects of exercise train- 2004;36(3):533–53. doi:10.1249/01.MSS.0000115224.88514.3A.
ing on sleep apnea: a meta-analysis. Lung. 2014;192(1):175–84. 51. Nybo L, Sundstrup E, Jakobsen MD, Mohr M, Hornstrup T,
doi:10.1007/s00408-013-9511-3. Simonsen L, et al. High-intensity training versus traditional
44. Chennaoui M, Arnal PJ, Sauvet F, Leger D. Sleep and exercise: a exercise interventions for promoting health. Med Sci Sports
reciprocal issue? Sleep Med Rev. 2014;. doi:10.1016/j.smrv. Exerc. 2010;42(10):1951–8. doi:10.1249/MSS.0b013e3181d
2014.06.008. 99203.
45. Afsar B. Relationship between total testosterone, cognitive 52. Pechter U, Maaroos J, Mesikepp S, Veraksits A, Ots M. Regular
function, depressive behavior, and sleep quality in chronic kidney low-intensity aquatic exercise improves cardio-respiratory func-
disease patients not on dialysis. Clin Exp Nephrol. tional capacity and reduces proteinuria in chronic renal failure
2013;17(1):59–65. doi:10.1007/s10157-012-0652-0. patients. Nephrol Dial Transplant. 2003;18(3):624–5.
46. De Santo RM, Bilancio G, Santoro D, Vecchi ML, Perna A, De 53. Pechter U, Ots M, Mesikepp S, Zilmer K, Kullissaar T, Viha-
Santo NG, et al. A longitudinal study of sleep disorders in early- lemm T, et al. Beneficial effects of water-based exercise in
stage chronic kidney disease. J Ren Nutr. 2010;20(5 patients with chronic kidney disease. Int J Rehabil Res.
Suppl):S59–63. doi:10.1053/j.jrn.2010.06.003. 2003;26(2):153–6. doi:10.1097/01.mrr.0000070755.63544.5a.
47. Cohen SD, Patel SS, Khetpal P, Peterson RA, Kimmel PL. Pain, 54. Dalal HM, Evans PH, Campbell JL, Taylor RS, Watt A, Read
sleep disturbance, and quality of life in patients with chronic KL, et al. Home-based versus hospital-based rehabilitation after
kidney disease. Clin J Am Soc Nephrol. 2007;2(5):919–25. myocardial infarction: a randomized trial with preference arms—
doi:10.2215/cjn.00820207. Cornwall Heart Attack Rehabilitation Management Study
48. Taylor RS, Dalal H, Jolly K, Zawada A, Dean SG, Cowie A, et al. (CHARMS). Int J Cardiol. 2007;119(2):202–11. doi:10.1016/j.
Home-based versus centre-based cardiac rehabilitation. Cochrane ijcard.2006.11.018.

123