Original Investigation Effect of Moderate Aerobic Exercise Training on Endothelial Function and Arterial Stiffness in CKD Stages 3-4: A Randomized Controlled Trial Amaryllis H. Van Craenenbroeck, MD,1,2,3 Emeline M. Van Craenenbroeck, MD, PhD,2,4 Katrijn Van Ackeren, MSc,2 Christiaan J. Vrints, MD, PhD,2,4 Viviane M. Conraads, MD, PhD,2,4,y Gert A. Verpooten, MD, PhD,3 Evangelia Kouidi, MD, PhD,5 and Marie M. Couttenye, MD, PhD1 Background: Evidence of a beneficial effect of exercise training on mediators of vascular disease is accumulating in chronic kidney disease (CKD), but its effect on vascular function in vivo still has to be established. The present study was designed to investigate whether a formal aerobic exercise training program improves peripheral endothelial function in patients with CKD stages 3 to 4. Study Design: Randomized controlled trial with a parallel-group design. Setting & Participants: 48 patients with CKD stages 3 to 4 without established cardiovascular disease were randomly assigned to either an exercise training program or usual care. 40 patients completed the study (exercise training, 19; usual care, 21). Intervention: The 3-month home-based aerobic training program consisted of 4 daily cycling sessions of 10 minutes each at a target heart rate, calculated as 90% of the heart rate achieved at the anaerobic threshold. Patients in the usual-care group were given standard therapy. Outcomes: The primary outcome was peripheral endothelial function. Secondary outcomes were aerobic capacity, arterial stiffness, numbers of endothelial (EPCs) and osteogenic progenitor cells (OPCs), migratory function of circulatory angiogenic cells, and health-related quality of life. Measurements: Endothelial function was assessed with flow-mediated dilation of the brachial artery, aerobic capacity by peak oxygen uptake (Vo2peak), arterial stiffness by carotid-femoral pulse wave velocity, numbers of EPCs and OPCs by flow cytometry, circulatory angiogenic cell function by an in vitro migratory assay, and quality of life by the Kidney Disease Quality of Life2Short Form questionnaire. Results: Exercise training significantly improved Vo2peak and quality of life, but not in vivo vascular function (flow-mediated dilation and carotid-femoral pulse wave velocity) or cellular markers for vascular function (EPC and OPC count and circulatory angiogenic cell migratory function). Limitations: Short duration and intermittent nature of the exercise intervention. Conclusions: In patients with CKD stages 3 to 4 without overt cardiovascular disease, 3 months of aerobic exercise training improved Vo2peak and quality of life, without altering endothelial function or arterial stiffness. Am J Kidney Dis. -(-):---. ª 2015 by the National Kidney Foundation, Inc. INDEX WORDS: Chronic kidney disease (CKD); exercise training; aerobic exercise; intermittent training; randomized controlled trial; endothelial dysfunction; aerobic capacity; arterial stiffness; quality of life; cardiovascular disease prevention.

C

ardiovascular disease remains the main cause of death in patients with chronic kidney disease (CKD), which cannot be explained by traditional risk factors alone. Vascular dysfunction, characterized by endothelial dysfunction and arterial stiffness, is an important nontraditional cardiovascular risk factor, occurring early in the course of kidney failure and predicting future cardiovascular events.1 Already in

mild CKD, endothelial dysfunction coincides with defective endothelial repair mechanisms: low levels of bone marrow–derived endothelial progenitor cells (EPCs) and reduced migratory function of circulating angiogenic cells.2-5 More recently, increased numbers of circulating progenitor cells with an osteogenic potential (OPCs) have been related to vascular calcification and arterial stiffness.6,7

From the 1Department of Nephrology, Antwerp University Hospital; 2Laboratory for Molecular and Cellular Cardiology and 3 Laboratory of Experimental Medicine and Pediatrics, University of Antwerp; 4Department of Cardiology, Antwerp University Hospital, Edegem, Belgium; and 5Laboratory of Sports Medicine, Aristotle University of Thessaloniki, Thessaloniki, Greece. y Deceased. Received October 29, 2014. Accepted in revised form March 5, 2015.

Trial registration: www.ClinicalTrials.gov; study number: NCT02209402. Address correspondence to Amaryllis H. Van Craenenbroeck, MD, Department of Nephrology, Antwerp University Hospital, Wilrijkstraat 10, 2650 Edegem, Belgium. E-mail: amaryllis. [email protected]  2015 by the National Kidney Foundation, Inc. 0272-6386 http://dx.doi.org/10.1053/j.ajkd.2015.03.015

Am J Kidney Dis. 2015;-(-):---

1

Van Craenenbroeck et al

Because endothelial dysfunction is found to play a key role in the pathogenesis of cardiovascular complications, including arterial stiffness,8 it is an attractive treatment target in trying to better the high cardiovascular morbidity and mortality in CKD.9 Among possible therapies for endothelial dysfunction, exercise training is an effective nonpharmacologic approach. In patients with established cardiovascular disease, exercise training leads to improvement and even correction of endothelial dysfunction, mainly through shear stress–induced upregulation of endothelial nitric oxide (NO) synthase (eNOS),10 as well as anti-inflammatory and antioxidative effects.11 In addition, ameliorated endothelial repair, through increased number and function of EPCs and circulatory angiogenic cells, contributes to restoration of endothelial health following exercise training.12-14 In patients with CKD, regular exercise lowers blood pressure and improves muscular strength, peak oxygen uptake (Vo2peak), physical performance, and quality of life.15-17 Evidence of a beneficial effect of exercise training on mediators of vascular disease (eg, inflammation and oxidative stress) is accumulating, but whether this translates into an improvement of vascular function in vivo and eventually could prevent cardiovascular disease in CKD still has to be established.18 Moreover, it is unclear whether vascular adaptations or other still undefined mechanisms account for the observed increase in Vo2peak following exercise training. Vo2peak relates to both oxygen delivery (cardiac output, vascular function, and erythrocyte count) and oxygen-utilizing factors (skeletal muscle). In the present study, we hypothesized that a functional adaptation to exercise is partially mediated by an improvement of vascular function and hence a beneficial effect on the working muscles through a decrease in peripheral vascular resistance and redistribution of blood flow. This study was designed to evaluate the effect of a 3-month home-based intermittent aerobic exercise training program on endothelial function in patients with CKD stages 3 to 4 without established cardiovascular disease in comparison to usual care. Secondary end points were aerobic capacity and arterial stiffness, as well as numbers of circulating EPCs and OPCs, circulatory angiogenic cell migratory function, and health-related quality of life.

METHODS Participants and Study Design Patients with CKD stages 3 to 4 according to NKF-KDOQI (National Kidney Foundation–Kidney Disease Outcomes Quality Initiative) guidelines (estimated glomerular filtration rate, 15-59 mL/min/1.73 m2, using the 4-variable MDRD [Modification of Diet in Renal Disease] Study equation) and without established cardiovascular disease were recruited from the outpatient clinic of the Antwerp University Hospital from April 2012 through July 2

2014. Established cardiovascular disease was defined as a positive history of any of the following: coronary artery disease, peripheral vascular disease, or cerebrovascular disease. Other exclusion criteria were pregnancy, age younger than 18 years, treatment with immunosuppressive or oral anticoagulation therapy, and active malignant disease. Prior to formal inclusion and randomization, eligible patients underwent electrocardiography, echocardiography, and exercise testing to exclude ischemic and/or structural heart disease (see Fig 1 for trial flow chart). Patients were randomly assigned in a 1:1 ratio by sealed opaque envelopes to exercise training or usual care. At baseline and after 3 months, patients underwent cardiopulmonary exercise testing, vascular function assessments, and blood sampling. Patients in the exercise-training group were called in within 1 week after completion of the last training session. The study complies with the Declaration of Helsinki, was approved by the local ethics committee, and written informed consent was obtained.

Exercise Training Patients in the exercise-training group underwent a 3-month home-based intermittent aerobic training program at moderate intensity, based on the recent KDIGO (Kidney Disease: Improving Global Outcomes) CKD guideline and cardiovascular prevention guidelines.19 The program consisted of 4 daily cycling sessions of 10 minutes each at a target heart rate calculated as 90% of the heart rate achieved at the anaerobic threshold on baseline testing.20 Magnetically braked home exercise bikes (DKN Mag 410B) and heart rate transmitters (Polar FT7) were provided to participating patients for the study period by the study center. Figure 2 illustrates the scheduled study visits and contacts for evaluation of training course and adherence. In the first 2 weeks of the study period, at least 3 training sessions were supervised in hospital by an experienced medical doctor (A.H.V.C.). For the following 2 weeks, a supervised training session was organized once a week. Adherence was monitored monthly by heart rate data obtained during training and detailed training logs. Requirements for adherence were as follows: 70 or more training days with at least the predefined 40 minutes per day. If more than 5 consecutive days of training were missed, the study period was prolonged by 1 week. Before final assessment, patients had to exercise for 2 consecutive weeks. Patients who did not adhere to these requirements were excluded. Patients in the usual-care group were given standard therapy, without specific instructions about physical activity. In both groups, medical therapy was unchanged during the study period.

Cardiopulmonary Exercise Test A symptom-limited cardiopulmonary exercise test with gas analysis was performed on a bicycle ergometer (CARDIOVIT CS200; Schiller AG). A ramp protocol with incremental steps of 10 or 20 W/min was used. A 12-lead electrocardiogram was recorded continuously and blood pressure was measured at baseline and every 2 minutes. Vo2peak was defined as mean oxygen uptake during the final 30 seconds of exercise and also expressed as a percentage of predicted value.21 Anaerobic threshold was determined using the V-slope method by a single, blinded, experienced exercise physiologist (E.M.V.C.). Circulatory power, a surrogate of cardiac power at peak exercise, was calculated as Vo2peak 3 peak systolic blood pressure. Peak oxygen pulse (milliliters per beat) was estimated from the carbon dioxide rebreathing method as a surrogate marker of stroke volume. Online analysis of VE/Vo2 and VE/VCO2 (ventilatory equivalents for oxygen and carbon dioxide, respectively) curves increased the likelihood that patients exercised to an exhaustion level of effort, confirmed by a respiratory exchange ratio . 1.10 for all participants. A single independent technician, blinded to the group allocation, performed all cardiopulmonary exercise testing. Am J Kidney Dis. 2015;-(-):---

Exercise Training in CKD

Vascular Assessments: In Vivo Patients were called in after an overnight fast and had refrained from exercise, food, and caffeine for at least 8 hours before assessment. Analyses were performed in a temperature-controlled room (21 C-24 C) by a single trained independent operator (K.V.A.).

Endothelial Function Endothelial function was assessed by flow-mediated dilation (FMD) of the brachial artery (ProSound 6; Hitachi-Aloka Medical Ltd) according to the International Brachial Artery Reactivity Task Force guidelines.22 Briefly, after recording of the baseline diameter for at least 1 minute of stable distension waveforms, a blood pressure cuff was inflated to 200 mm Hg or 50 mm Hg suprasystolic on the forearm for 5 minutes. During the final 30 seconds of occlusion, diameter was recorded meticulously in order to calculate low flow-mediated constriction as percentage constriction from baseline diameter. After cuff release, the longitudinal image of the artery was recorded continuously until 4 minutes after cuff deflation. A midartery pulsed Doppler signal was obtained immediately after cuff release and no later than 15 seconds after cuff deflation to assess hyperemic velocity. FMD was expressed as percent dilation from baseline to maximal postocclusion diameter. Endothelial-independent vasodilation was measured after sublingual administration of glyceryl trinitrate (GTN-MD). Off-line analyses were done with FMD-I software (FLOMEDI, version 2009) by a single trained investigator (K.V.A.) blinded to allocated treatment. Baseline, minimal, and maximal diameters were manually or automatically detected using edge detection software by the single trained operator.

Arterial Stiffness Carotid-femoral pulse wave velocity (PWV) was measured in triplicate using SphygmoCor (CVMS CPV model; AtCor

Medical).23 The system uses a single high-fidelity applanation tonometer (Millar Instruments) to obtain a proximal (ie, carotid) and distal (ie, femoral) pulse recorded consecutively and calculates carotid-femoral PWV from the transit time between the 2 arterial sites as PWV 5 distance (meters)/transit time (seconds). To determine the distance, a scaling factor of 0.8 was applied to the direct carotid-femoral distance determined by tape measure in order to prevent overestimation.24 Augmentation index was derived from carotid and radial pulse wave analysis. Central blood pressure was estimated noninvasively from the radial artery waveform using the Sphygmocor transfer function. All measurements were done in triplicate and were repeated when they did not meet the quality control guidelines as defined by the manufacturer.

Cellular Markers for Vascular Function Prior to in vivo vascular assessment, venous blood was collected from an antecubital vein. Peripheral-blood mononuclear cells were isolated from 70 mL of anticoagulated blood (acid citrate dextrose) by density gradient centrifugation. Next, peripheral-blood mononuclear cells were processed for immunostaining and cell culture.

Quantification of EPCs and OPCs Circulating EPCs and OPCs were quantified by multiparametric flow cytometry using the following antibodies: CD34phycoerythrin-Cy7, CD45-allophycocyanin-H7 (both are fluorescently labeled mouse anti-human antibodies from BD Pharmingen), KDR-phycoerythrin (a fluorescently labeled mouse anti-human antibody to vascular endothelial growth factor receptor 2 from R&D Systems), goat anti-human osteocalcin (OCN; from Santa Cruz Biotechnology), and a fluorescein isothiocyanate2 F(ab)2 donkey antigoat secondary antibody (Jackson ImmunoResearch). EPCs were defined as CD45-dim, CD34-positive, and

Figure 1. Trial flowchart. Of 778 patients who were screened, 104 patients were considered eligible. Fifty-six patients refused participation and 48 patients were randomly assigned to exercise training (25) or usual care (23). In the exercise training group, 19 patients completed the study, compared with 21 patients in the usual-care group. Abbreviation: CV, cardiovascular. Am J Kidney Dis. 2015;-(-):---

3

Van Craenenbroeck et al KDR-positive cells25; OPCs, as CD451CD341OCN1 cells.26 CD451CD341 cell and EPC and OPC count were expressed as number per 106 CD451 cells with low forward scatter and side scatter (see Item S1, available as online supplementary material).

Migratory Activity of Circulating Angiogenic Cells Peripheral-blood mononuclear cells were cultured in fibronectin-coated wells using Endothelial Growth Medium-2-MV (Lonza) as previously described.14 After 7 days, migration capacity of circulatory angiogenic cells toward vascular endothelial growth factor (50 ng/mL; R&D Systems) and stromal cell–derived factor (SDF-1a, 100 ng/mL; R&D Systems) was evaluated using a modified Boyden chamber (see Item S1).

Health-Related Quality of Life The Kidney Disease Quality of Life2Short Form (KDQoL-SF) consists of 24 questions evaluating both kidney disease–targeted items and global health status. The questionnaire was administered at baseline and after 3 months.

Statistical Analysis Because to our knowledge no prior exercise intervention studies are available in CKD, sample size calculation was based on data from patients with chronic heart failure27 and revealed that 17 patients per group would be sufficient to detect a 40% increase in FMD (chosen for its clinical relevance) at an a error level of 0.05 and b level of 0.20.14,28 Continuous data were expressed as mean 6 standard deviation. Skewed distributed data (1-sample Kolmogorov-Smirnov test) were logarithmically transformed. Baseline comparisons were performed using independent-sample t test or c2 test. Different trends over time between groups (interaction) were assessed by 2-way repeated-measures analysis of variance with group identity as first factor and within-subject factor (baseline vs 3-month follow-up) as second factor. All statistical tests were 2 sided, and P , 0.05 was considered statistically significant. Analyses were performed using SPSS, version 22 (IBM).

RESULTS Patient Characteristics and Adherence to Training A total of 48 patients were included and 40 patients (exercise training, 19; usual care, 21) completed the study. Eight patients withdrew during the course of the study for various reasons, as set out in Fig 1. Patient characteristics are summarized in Table 1. Groups were comparable in demographic and clinical characteristics, including vascular function and exercise capacity. Adherence to the prescribed training program was excellent, as was evident from the read-outs of heart rate transmitters and the training logs. There were no exclusions due to nonadherence. On average, patients completed 3.6 6 0.4 of 4 sessions per day and 26.7 6 2.6 of the prescribed 28 sessions per week. Attained heart rate during training was 101% 6 1.6% of target heart rate (mean, 124 6 14 beats/min). At baseline, no statistically significant difference existed between the basal heart rate of patients with or without b-blocker use (heart rate, 69 6 7 and 78 6 11 beats/min, respectively; P . 0.05). Mean target heart rate did not differ between the groups of patients with or without b-blocker use (119 6 14 vs 129 6 12 beats/min, respectively; P . 0.05). Training-Induced Effects Effects on Exercise Capacity At follow-up, Vo2peak, percent predicted Vo2peak, Vo2 at anaerobic threshold, maximal workload, circulatory power, exercise duration, and peak oxygen

Figure 2. Overview of training sessions. In the first 2 weeks of the study period, at least 3 training sessions were supervised in the hospital by an experienced medical doctor. For the following 2 weeks, a supervised training session was organized once a week. Motivational telephone calls and readouts of the heart rate transmitters were organized monthly. *Read-out heart rate transmitter. # Telephone call. Abbreviation: ET, exercise training. 4

Am J Kidney Dis. 2015;-(-):---

Exercise Training in CKD Table 1. Characteristics of Patients Who Completed the Study Exercise Training (n 5 19)

Usual Care (n 5 21)

P

Age (y) Male sex BMI (kg/m2) SBP (mm Hg) DBP (mm Hg) Hemoglobin (g/dL) Cholesterol Total (mg/dL) LDL (mg/dL) HDL (mg/dL)

51.5 6 11.8 54.7 6 14.1 11 (58) 11 (52) 28.3 6 6.2 28.3 6 5.8 129.0 6 17.5 122.9 6 16.2 79.5 6 8.0 81.2 6 9.9 13.0 6 1.4 13.4 6 1.1

eGFR (mL/min/1.73 m2) PCR category ,30 mg/g 30-300 mg/g .300 mg/g

37.5 6 13.23 39.6 6 12.9 0.6 0.9 1 (5) 1 (10) 12 (63) 12 (57) 6 (32) 7 (33)

171 6 20 101 6 19 47 6 8

Kidney disease cause Reflux nephropathy Solitary kidney Medication-induced nephropathy Diabetic nephropathy Nephroangiosclerosis IgA nephropathy ADPKD FSGS MCKD1 Lithiasis

0.1 0.6 0.07

0.6

CV risk factors Familial history of CV disease History of smoking Diabetes mellitus Medication Diuretics RAAS inhibitors b-Blockers Calcium channel blockers Erythropoietin Acetylsalicylic acid Statin Vitamin D supplementation

187 6 38 105 6 28 54 6 15

0.4 0.5 0.9 0.3 0.5 0.9

3 (14) 2 (11) 0 (0)

2 (10) 6 (28) 1 (5)

2 4 2 5 1 0 0

1 4 3 2 0 1 1

(11) (21) (10) (27) (6) (0) (0)

(5) (18) (14) (10) (0) (5) (5)

8 (42)

8 (38)

0.5

9 (47) 2 (11)

7 (33) 2 (9)

0.3 0.7

4 15 9 11 3 3 12 6

(21) (79) (47) (58) (16) (16) (63) (31)

4 17 7 7 2 4 10 7

(19) (80) (33) (33) (10) (14) (48) (33)

0.9 0.9 0.4 0.1 0.6 0.9 0.3 0.9

Note: Values for categorical variables are given as number (percentage); values for continuous variables, as mean 6 standard deviation. Conversion factor for cholesterol in mg/dL to mmol/ L, 30.02586. Abbreviations: ADPKD, autosomal dominant polycystic kidney disease; BMI, body mass index; CV, cardiovascular; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; FSGS, focal segmental glomerulosclerosis; HDL, high-density lipoprotein; IgA, immunoglobulin A; LDL, low-density lipoprotein; MCKD1, medullary cystic kidney disease type 1; PCR, protein-creatinine ratio; RAAS, renin-angiotensin-aldosterone system; SBP, systolic blood pressure.

pulse all improved significantly in the exercisetraining group but were unchanged in the usual-care group (P for interaction , 0.001; Table 2; Fig 3A). Am J Kidney Dis. 2015;-(-):---

Heart rate at rest, peak heart rate, and peak blood pressure were unchanged in both groups. Effects on Vascular Function: In Vivo and Cellular Markers As was evident from the P for interaction and 95% confidence intervals of the mean difference between the 2 groups, exercise training did not result in a significant improvement in peripheral endothelialdependent vasodilation, measured by FMD, as compared to usual care (P for interaction 5 0.9; Table 2; Fig 3B). Similarly, no significant differences were observed between groups for change in low flow-mediated constriction, endothelial-independent vasodilation (GTN-MD), or absolute diameters. Likewise, arterial stiffness, as evaluated by carotid-femoral PWV and augmentation index, did not improve significantly in the exercise-training group in comparison to the usual-care group (Table 2; Fig 3C). There was no significant difference between groups over time for central blood pressure (Table 2). Regarding cellular markers of vascular function, numbers of circulating EPCs and OPCs did not change significantly in the exercise-training group compared to the usual-care group. Moreover, there was no change in circulatory angiogenic cell migratory function in the 2 groups (Table 2). However, as was evident from the 95% confidence interval of the mean difference, variability of these parameters was high. Effects on Other Cardiovascular Risk Factors Despite nominal improvements in the exercisetraining group, the evolution of body mass index (P , 0.05 in exercise training) and high-sensitivity Creactive protein levels was not different between groups. Although no significant change was noted in total cholesterol levels in both groups separately, there was a significant difference between both groups (P for interaction 5 0.04) with a nominal improvement in the usual-care group. Low- and highdensity lipoprotein cholesterol levels, as well as kidney function, were unchanged in both groups (Table 3). Effects on Health-Related Quality of Life KDQoL-SF questionnaires were completed for 33 patients (exercise training, 16; usual care, 17). Following exercise training, health-related quality of life significantly improved in the fields of cognitive function, sleep quality, and energy levels in comparison to usual care (P , 0.05; Table 4).

DISCUSSION To our knowledge, the present study is the first evaluation of the effect of a 3-month home-based 5

6

Table 2. Cardiopulmonary Exercise Testing and Vascular Parameters at Baseline and 3-Month Follow-up ET (n 5 19) Baseline

UC (n 5 21) Final

Baseline

Final

Vo2peak (mL/kg/min) 26.5 6 5.4 32.3 6 6.8 24.8 6 6.5 24.0 6 6.5 % of predicted Vo2peak 88 6 19 107 6 19 87 6 26 85 6 26 Vo2 at AT (mL/kg/min) 21.9 6 4.2 29.0 6 6.5 22.4 6 5.2 22.5 6 5.5 85 6 13 91 6 5 94 6 7 91 6 6 Vo2 as % of Vo2peak Maximal workload (W) 164 6 41 201 6 53 160 6 48 151 6 45 Circulatory power (mm Hg $ mL 5,193 6 1,377 6,350 6 2,013 4,747 6 1,428 4,154 6 1,668 Vo2/[kg/min]) Exercise duration (s) 433 6 119 538 6 153 398 6 116 385 6 120 Peak oxygen pulse (mL/beat) 14.2 6 4.0 16.7 6 3.9 13.3 6 3.3 13.4 6 3.2 Heart rate at rest (beats/min) 74 6 10 73 6 9 78 6 15 77 6 15 Peak heart rate (beats/min) 156 6 20 156 6 19 149 6 28 150 6 32 Peak SBP (mm Hg) 194 6 20 193 6 32 189 6 23 180 6 24 Peak DBP (mm Hg) 81 6 17 77 6 16 83 6 14 82 6 14 t1/2 Vo2peak (s) 178 6 33 172 6 29 206 6 49 198 6 44 Peak RER 1.38 6 0.15 1.32 6 0.05 1.37 6 0.11 1.35 6 0.08 FMD (%) 4.0 6 1.9 4.6 6 3.0 5.2 6 3.4 5.3 6 3.1 L-FMC (%) 20.34 6 2.18 20.05 6 2.07 20.93 6 1.64 20.80 6 2.81 GTN-MD (%) 16.4 6 6.8 17.8 6 7.3 18.6 6 8.7 20.0 6 11.6 Diameter Baseline (mm) 3.78 6 0.72 3.74 6 0.68 3.68 6 0.63 3.62 6 0.51 Maximum (mm) 3.95 6 0.73 3.92 6 0.68 3.87 6 0.63 3.81 6 0.53 Minimum (mm) 3.67 6 0.71 3.63 6 0.70 3.55 6 0.62 3.55 6 0.49 9.1 6 1.9 8.9 6 2.2 14.2 6 18.1 20.0 6 18.7 26.1 6 13.6 25.5 6 13.3 135 6 21 132 6 16 82 6 12 79 6 7 2,335 6 1,224 2,021 6 816 49 6 56 45 6 81 228 6 178 214 6 187 42.6 6 17.2 41.1 6 16.4

8.8 6 1.6 8.3 6 1.3 19.8 6 20.2 18.6 6 18.3 28.6 6 9.5 28.1 6 10.5 140 6 21 131 6 16 80 6 9 78 6 8 2,577 6 1,574 2,396 6 1,371 35 6 32 37 6 29 204 6 176 214 6 179 40.78 6 22.6 42.3 6 18.1

5.82 18.48 7.11 5.9 37.37 1,157 105.26 2.47 21.32 0.58 21.05 23.84 26.32 20.07 0.41 0.41 0.86

(4.22 to 7.42) (14.16 to 22.81) (4.18 to 10.05) (21 to 12.9) (26.73 to 48.01) (730 to 1,583) (71.32 to 139.21) (1.65 to 3.28) (25.02 to 2.38) (23.41 to 4.57) (211.18 to 9.07) (210.74 to 3.08) (223.56 to 10.93) (20.13 to 0.0001) (20.95 to 1.76) (20.84 to 1.66) (22.65 to 4.37)

20.04 (22.28 to 1.60) 20.04 (20.19 to 0.11) 20.01 (20.14 to 0.11) 20.13 5.82 20.44 23.58 23.41 2313 24.34 257.83 21.71

(20.59 to 0.34) (0.28 to 11.37) (22.84 to 1.95) (211.91 to 4.73) (29.72 to 2.90) (2674 to 47) (244.49 to 35.81) (2191 to 76) (213.28 to 9.86)

UC Mean Diff Post 2 Pre (95% CI)

20.79 22.46 0.06 22.5 29.16 2592 212.78 0.18 20.68 1.37 29.05 21.50 27.22 20.02 0.08 0.22 0.66

(22.25 to 0.66) (27.45 to 2.53) (21.28 to 1.40) (24.7 to 20.23) (223.03 to 4.69) (21,294 to 109)

6.61 20.94 7.05 8.4 46.53 1,749

(4.52 to 8.71) (14.59 to 27.28) (3.68 to 10.42) (0.7 to 16.4) (29.80 to 63.27) (956 to 2,542)

(226.51 to 0.95) 118.04 (81.95 to 154.13) (20.67 to 1.03) 2.29 (1.15 to 3.43) (26.42 to 5.05) 20.63 (27.22 to 5.96) (29.67 to 12.41) 20.79 (212.12 to 10.54) (219.67 to 1.57) 8 (26.17 to 22.17) (28.79 to 5.79) 22.34 (212.02 to 7.33) (236.90 to 22.46) 0.91 (231.72 to 33.54) (20.07 to 0.04) 20.04 (20.13 to 0.04) (21.88 to 2.05) 0.32 (21.88 to 2.53) (20.71 to 1.14) 0.19 (21.24 to 1.63) (23.42 to 4.74) 0.19 (24.93 to 5.33)

20.06 (20.90 to 0.80) 20.06 (20.22 to 0.10) 20.03 (20.19 to 0.12) 20.87 21.23 20.68 28.88 21.88 2181 2.44 8.97 1.31

Mean Diff ET 2 UC (95% CI)

(21.58 to 20.16) (26.21 to 3.73) (22.65 to 1.28) (213.52 to 24.24) (24.99 to 2.23) (2468 to 105) (215.91 to 20.79) (280.56 to 98.50) (211.94 to 14.56)

0.02 (20.18 to 0.24) 0.02 (20.19 to 0.24) 0.02 (20.17 to 0.21) 0.74 7.05 0.24 5.29 21.53 2132 26.78 266.81 23.02

(20.09 to 1.58) (20.09 to 14.21) (22.73 to 3.21) (23.86 to 14.44) (28.24 to 5.24) (2573 to 309) (247.29 to 33.73) (2219.84 to 86.22) (220.48 to 14.43)

P for Interactiona

,0.001 ,0.001 ,0.001 0.04 ,0.001 ,0.001 ,0.001 0.05 0.8 0.8 0.3 0.6 0.9 0.3 0.9b 0.8 0.9 0.8 0.8 0.9 0.1b 0.05 0.8 0.2 0.6 0.4b 0.7 0.4 0.7

Note: Unless otherwise indicated, values are given as mean 6 standard deviation. Abbreviations: AIx, augmentation index; AT, anaerobic threshold; CF-PWV, carotid-femoral pulse wave velocity; CI, confidence interval; DBP, diastolic blood pressure; Diff, difference; EPC, endothelial progenitor cell; ET, exercise training; FMD, flow-mediated dilation; GTN-MD, glyceryl trinitrate-mediated dilation; L-FMC, low flow2mediated constriction; OPC, osteogenic progenitor cell; RER, respiratory exchange ratio; SBP, systolic blood pressure; t1/2, half-life peak oxygen uptake; UC, usual care; Vo2peak, peak oxygen uptake, a P value for difference over time between the 2 groups (interaction). b P value is shown following logarithmic transformation of this skewed variable. c By flow cytometry; events are CD451 cells with low forward scatter and side scatter.

Van Craenenbroeck et al

Am J Kidney Dis. 2015;-(-):---

CF-PWV (m/s) AIx carotid artery AIx radial artery Central SBP (mm Hg) Central DBP (mm Hg) CD451CD341cells/106 eventsc EPC/106 eventsc OPC/106 eventsc Circulating angiogenic cell migratory capacity (%)

ET Mean Diff Post 2 Pre (95% CI)

Exercise Training in CKD

A

EXERCISE

USUAL CARE

VO2peak (ml/min/kg)

40

30

20

B

EXERCISE

l na Fi

B as el in e

l na Fi

B as el in e

10

USUAL CARE

14

Flow-mediated dilation (%)

12 10 8 6 4 2

C

EXERCISE

l na Fi

B as el in e

l na Fi

B as el in e

0

USUAL CARE

20

PWV (m/s)

15

10

l na Fi

B as el in e

l na Fi

B as el in e

5

Figure 3. Predefined end points at baseline and follow-up. (A) Peak oxygen consumption (Vo2peak), (B) brachial artery flow-mediated dilation, and (C) carotid-femoral pulse wave velocity (PWV) in patients with moderate to severe chronic kidney disease. *P , 0.001 between groups. The thick dotted line represents the group mean.

Am J Kidney Dis. 2015;-(-):---

intermittent moderate aerobic exercise training program on peripheral endothelial function in patients with CKD stages 3 to 4. Contrary to our hypothesis, exercise training did not improve endothelial function, arterial stiffness, or endothelial repair mechanisms despite a 27% increase in Vo2peak. This indicates that changes in vascular function are likely not primarily responsible for the benefits on Vo2peak induced by exercise training. There is consistent evidence that regular physical exercise curbs cardiovascular risk and partially corrects endothelial dysfunction in prehypertensive patients,29 patients with coronary artery disease,30,31 and patients with chronic heart failure,27,32 coinciding with a gain in Vo2peak.30,32 This suggests that vasodilatory capacity substantially contributes to the functional adaptations following training, at least in these patient groups. However, in patients with CKD stages 3 to 4, a total of 3 months of daily aerobic exercise training did not reverse the observed endothelial dysfunction. Of note, even after exclusion of overt cardiovascular disease, FMD in the present CKD population appeared to be severely impaired, which is in line with other independent studies in moderate to severe CKD, in which mean FMD was reported to be 6.12% to 6.77%.33,34 In addition to FMD, which is regarded as the chief surrogate for vascular endothelial health, we studied low flow2 mediated constriction for evaluating endothelial function. Whereas FMD is mediated by a shear stress2induced increase in NO, the ability to constrict in low-flow states provides information on vascular tone at rest,35,36 and both tests are complementary in assessment of endothelial function. Evidence of an increase in low flow2mediated constriction following exercise has been described in healthy individuals37 and patients with coronary artery disease.36 However, in our study, exercise training did not influence low flow2mediated constriction. Likewise, whereas our study patients were characterized by increased arterial stiffness,24 we did not demonstrate a significant effect of exercise training on arterial stiffness, which is in line with the findings of Headley et al.38 In that study, 4-month aerobic exercise training had no effect on carotid-femoral PWV in patients with CKD stage 3. Reversibility of arterial stiffness is still a matter of debate, with a more pronounced effect with sustained training and a clear association with substantial blood pressure reduction, at least in (pre)hypertensive patients.39 Whereas effects on the vasculature were negligible, the improvement in aerobic capacity following training was impressive, with normalization of exercise capacity (107% 6 19% predicted Vo2peak). This strongly supports findings from previous randomized

7

8

Note: Unless otherwise indicated, values are given as mean 6 standard deviation. Conversion factor for cholesterol in mg/dL to mmol/L, 30.02586. Abbreviations: BMI, body mass index; CI, confidence interval; Diff, difference; ET, exercise training; eGFR, estimated glomerular filtration rate; HDL, high-density lipoprotein; hs-CRP, high-sensitivity C-reactive protein; LDL, low-density lipoprotein; UC, usual care. a P value for the difference over time between the 2 groups (interaction).

0.5 0.6 1.07 (21.90 to 4.04) 20.04 (20.22 to 0.14) 20.33 (22.23 to 1.56) 20.03 (20.17 to 0.11) 0.74 (21.73 to 3.20) 20.07 (20.19 to 0.05) 39.2 6 15.2 0.31 6 0.28 38.6 6 14.2 0.23 6 0.25 37.5 6 13.3 0.30 6 0.35 eGFR (mL/min/1.73 m2) hs-CRP (mg/dL)

39.6 6 12.9 0.34 6 0.32

0.04 0.07 0.2 13.48 (0.59 to 26.35) 8.84 (20.86 to 18.54) 3.79 (21.49 to 9.08) 213.48 (223.88 to 23.06) 29.95 (218.29 to 21.61) 20.29 (24.36 to 3.78) 0 (27.94 to 7.94) 21.11 (25.79 to 3.57) 3.5 (20.18 to 7.18) 174 6 33 95 6 28 51 6 15 171 6 27 99 6 21 50 6 13 171 6 20 101 6 19 47 6 8

187 6 38 105 6 28 54 6 14

0.08 20.77 (21.63 to 0.85) 0.21 (20.49 to 0.91) 20.57 (21.09 to 20.04) 28.7 6 5.6 27.7 6 5.7 28.3 6 6.2

BMI (kg/m2) Cholesterol Total (mg/dL) LDL (mg/dL) HDL (mg/dL)

28.3 6 5.8

UC Mean Diff Post 2 Pre (95% CI) ET Mean Diff Post 2 Pre (95% CI) Final Baseline Final Baseline

UC (n 5 21) ET (n 5 19)

Table 3. Cardiovascular Risk Factors at Baseline and 3-Month Follow-up

Mean Diff ET 2 UC (95% CI)

P for Interactiona

Van Craenenbroeck et al

controlled trials.15 Despite the home-based design, training adherence was excellent, resulting in an even larger increase in Vo2peak (effect size, 5.8 mL/kg/min) than that reported by Headley et al38 (2.3 mL/kg/min) or Mustata et al40 (3.6 mL/kg/min). The absent increase in circulating numbers of EPCs or the migratory function of circulatory angiogenic cells confirms the in vivo observation of this study and provides a possible mechanistic explanation. The recruitment of EPCs and increase in function of circulatory angiogenic cells after exercise have been linked to improved endothelial function in other populations with preexisting endothelial dysfunction.27,41,42 Here, 2 processes are important: release from the bone marrow, mainly driven by NO,43 and, once in the circulation, appropriate recruitment to the site of endothelial injury, mainly driven by the SDF-1a/CXCR4 axis.44 As such, a positive regeneration process is initiated, with a progressively stronger defense against noxious stimuli on the endothelium, increased NO bioavailability and subsequently enhanced EPC recruitment. Our findings fit the hypothesis that an absent increase in NO bioavailability could be further responsible for the lack in vascular repair. Systemic inflammation, decreased NO bioavailability, and deficient endothelial repair are characteristics of CKD and contribute to cardiovascular disease. In patients with cardiovascular disease, exercise training exerts its benefits on endothelial function mainly through these mechanisms.10,11,31,41 However, our study did not confirm a beneficial effect of exercise training on endothelial function in CKD. With an absolute mean difference between exercise training and usual care in FMD of only 0.32% (95% confidence interval, 21.88 to 2.53), it is unlikely that the study missed a clinically relevant response in FMD. The observed power to find a similar increase in FMD as in the original study (in patients with chronic heart failure) was 90%.28 A 40% increase in FMD was chosen for sample size calculation because of its clinical relevance. FMD is a well-known prognostic factor for cardiovascular outcome in CKD.45 According to Yilmaz et al,45 each (absolute) decrease in FMD of 1% results in an increase in cardiovascular risk of 41% in the CKD population. From the same study, we learn that a (relative) increase in FMD of 40% in patients with CKD stage 4 would result in a cardiovascular risk equaling that of patients with CKD stage 1, which is considered be highly clinically significant. Recently, Beck et al29 reported a 63% increase in percent dilation after endurance training in prehypertensive individuals. A first explanation for this observed discrepancy between patients with CKD and cardiovascular disease could be that NO bioavailability is so deeply reduced in CKD that the training-induced shear stress Am J Kidney Dis. 2015;-(-):---

Exercise Training in CKD

Am J Kidney Dis. 2015;-(-):---

Table 4. Results From the Health-Related Quality-of-Life Assessment at Baseline and After 3 Months ET (n 5 16)

UC (n 5 17) Final

ET Mean Diff Post 2 Pre (95% CI)

UC Mean Diff Post 2 Pre (95% CI)

Mean Diff ET 2 UC (95% CI)

P for Interactiona

Baseline

Final

Baseline

Kidney disease2targeted scale Symptoms/problems Effects of kidney disease Burden of kidney disease Work status Cognitive function Quality of social interaction Sleep Social support Overall health rating

83.09 6 13.17 90.57 6 10.77 80.86 6 12.18 75.00 6 31.62 87.5 6 13.96 82.08 6 14.13 71.72 6 19.25 91.66 6 12.17 68.13 6 14.24

88.35 6 13.40 91.60 6 8.74 87.89 6 11.74 87.5 6 22.36 92.92 6 9.57 83.75 6 13.33 77.97 6 13.29 89.58 6 14.75 76.25 6 15.00

73.11 6 15.45 83.88 6 12.87 75.00 6 27.86 69.44 6 15.81 78.15 6 15.81 82.22 6 14.64 61.48 6 23.51 87.04 6 16.72 67.22 6 19.34

75.67 6 15.15 2.14 (21.02 to 5.29) 80.72 6 14.56 0 (25.12 to 5.12) 75.73 6 25.66 8.59 (2.65 to 14.53) 61.76 6 37.62 3.33 (23.82 to 10.48) 78.03 6 13.69 3.92 (20.93 to 8.77) 73.33 6 18.10 20.39 (29.73 to 8.95) 62.65 6 21.86 5.83 (0.45 to 11.21) 81.37 6 20.31 0 (29.23 to 9.23) 70.58 6 18.86 6.87 (22.39 to 16.15)

0.30 23.97 20.42 23.33 23.11 28.00 22.77 24.44 3.33

(23.62 to 4.22) (210.07 to 2.13) (29.24 to 8.40) (210.48 to 3.82) (27.49 to 1.27) (211.47 to 24.52) (26.68 to 1.22) (214.59 to 5.71) (21.65 to 8.31)

1.83 3.97 9.01 6.67 7.03 7.61 8.61 4.44 3.54

(22.93 to 6.61) (23.62 to 11.55) (21.04 to 19.06) (22.99 to 16.32) (0.69 to 13.37) (22.48 to 17.69) (2.26 to 14.96) (28.66 to 17.54) (26.74 to 13.83)

0.4 0.3 0.08 0.2 0.03 0.1 0.01 0.5 0.5

36-Item Health Survey Scales Role–Physical Pain General Health Perceptions Emotional Well-Being Role–Emotional Social Function Energy/Fatigue

85.71 6 28.94 85.89 6 22.26 63.57 6 19.06 81.14 6 13.98 90.48 6 27.51 91.96 6 14.38 68.93 6 15.95

96.43 6 13.36 90.00 6 14.07 69.28 6 15.17 85.71 6 6.96 100 6 0.00 93.75 6 10.69 77.14 6 11.04

75.00 6 36.51 71.87 6 26.38 50.31 6 19.45 65.00 6 24.72 81.25 6 29.74 71.87 6 26.81 51.25 6 22.47

70.83 6 39.65 9.09 (22.23 69.79 6 23.09 1.66 (25.22 50.83 6 19.98 3.88 (23.95 61.00 6 20.68 2.22 (25.65 80.55 6 38.81 3.03 (23.72 66.66 6 21.54 21.47 (29.93 49.16 6 23.92 4.72 (20.90

25.55 24.68 4.37 24.75 27.40 23.43 23.91

(224.23 to 13.12) (213.59 to 4.21) (20.07 to 8.82) (27.88 to 21.62) (238.20 to 23.38) (28.46 to 1.59) (212.58 to 4.77)

14.64 6.35 20.48 6.97 10.43 2.43 8.16

(24.71 to 34.0) (24.32 to 17.03) (29.47 to 8.49) (21.58 to 15.52) (215.75 to 36.6) (29.19 to 14.07) (0.83 to 15.48)

0.1 0.2 0.9 0.1 0.4 0.7 0.03

Note: Unless otherwise indicated, values are given as mean 6 standard deviation. Abbreviations: CI, confidence interval; Diff, difference; ET, exercise training; UC, usual care. a P value for the difference over time between the 2 groups (interaction).

to to to to to to to

20.41) 8.55) 11.73) 10.09) 9.78) 6.99) 10.34)

9

Van Craenenbroeck et al

is insufficient. Myriad factors are involved in the reduction of NO bioavailability in patients with CKD.18,46 First, eNOS activity is reduced due to the presence of endogenous inhibitors such as asymmetric dimethylarginine (ADMA).47 Next, a net NO deficiency occurs when there is oxidative stress because of scavenging of NO by oxygen radicals, as well as switching of eNOS into a superoxide generator (eNOS uncoupling).48 Although evidence exists that exercise training is effective on both levels (increasing eNOS activity through the increase in shear stress as well as reducing oxidative stress) in patients with established cardiovascular disease, formal confirmation of this molecular adaptation is still lacking in CKD. In CKD, the effects of endogenous inhibitors could be too strong to overcome or the increased NO could be scavenged too quickly. The observation that exercise training had no effect on circulating NO metabolites in CKD stage 3 by Headley et al38 supports this explanation. Alternatively, although the increase in NO bioactivity erodes within weeks when training ends, studies have shown that if exercise continues, the short-term functional adaptation is followed by NO-dependent structural changes, which results in arterial remodeling and structural normalization of shear.49 However, no changes were detected in arterial diameter, making this a very unlikely explanation. In the present randomized controlled trial, a significant benefit on certain aspects of quality of life could be demonstrated in the exercise group in comparison to usual care—cognitive function, sleep quality, and energy levels, in particular. Self-reported sleep quality is an important factor for global well-being and even predicts mortality in hemodialysis patients50 and patients with CKD stages 3 to 5.51 A strength of our study is that patients with clinical atherosclerotic disease were excluded (primary prevention). Preexisting cardiovascular disease could have contaminated outcome assessment and/or exercise training performance. This study also has some limitations. Although we showed a relevant increase in Vo2peak, a longer training period, higher exercise intensity, or a continuous instead of intermittent training program could have induced important clinical changes in vascular function. In view of a favorable outcome of intermittent exercise training in terms of aerobic capacity and quality of life in healthy individuals52 and patients with other chronic diseases,53,54 we chose to implement an intermittent training regimen. The effect of other training modalities should be studied in future trials. Moreover, patients included in our study sample agreed to engage in a demanding exercise training scheme, which could limit generalizability of our findings to other patient populations 10

in CKD. The possibility that exercise training improves vascular function in a broader, perhaps more representative, CKD patient population cannot be excluded. In conclusion, our study supports prior reports that exercise training improves aerobic capacity and quality of life in patients with moderate to severe CKD without established cardiovascular disease. These functional adaptations could not be explained by an improvement in vascular function or in vascular repair, suggesting that other mechanisms are involved at early stages (eg, adaptations of the skeletal muscle and central hemodynamic effects). Care must be taken when interpreting our data in a general CKD population, which is typically characterized by the presence of cardiovascular disease. Future studies should be designed to elucidate underlying mechanisms of functional adaptation to exercise, as well as define the best intensity, modality, and volume of exercise training in this vulnerable population.

ACKNOWLEDGEMENTS The authors thank Bert Ectors for performing cardiopulmonary exercise tests, Paul Beckers for advice regarding the exercise training schemes, Bruno Verdonck for valuable technical assistance, and Ingrid Aelbrecht for help with eligibility screening. Support: Dr A.H. Van Craenenbroeck is supported by a research grant from the University of Antwerp. Dr E.M. Van Craenenbroeck is supported by the Research Foundation Flanders (FWO). Financial Disclosure: The authors declare that they have no other relevant financial interests. Contributions: research idea and study design: AHVC, EMVC, CJV, GAV; data acquisition: AHVC, KVA; data analysis/interpretation: AHVC, EMVC, GAV, EK, MMC; statistical analysis: AHVC, GAV; supervision or mentorship: GAV, CJV, MMC. VMC died before this manuscript was submitted; AHVC affirms that she contributed to the research idea and study design and had a prominent role as supervisor/mentor and vouches for her coauthorship status; all other authors approved the final author list. Except as noted, each author contributed important intellectual content during manuscript drafting or revision and accepts accountability for the overall work by ensuring that questions pertaining to the accuracy or integrity of any portion of the work are appropriately investigated and resolved. AHVC takes responsibility that this study has been reported honestly, accurately and transparently; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and registered) have been explained.

SUPPLEMENTARY MATERIAL Item S1: Supplemental methods. Note: The supplementary material accompanying this article (http://dx.doi.org/10.1053/j.ajkd.2015.03.015) is available at www.ajkd.org

REFERENCES 1. Gansevoort RT, Correa-Rotter R, Hemmelgarn BR, et al. Chronic kidney disease and cardiovascular risk: epidemiology, mechanisms, and prevention. Lancet. 2013;382:339-352. 2. Krenning G, Dankers PY, Drouven JW, et al. Endothelial progenitor cell dysfunction in patients with progressive Am J Kidney Dis. 2015;-(-):---

Exercise Training in CKD chronic kidney disease. Am J Physiol Renal Physiol. 2009;296: F1314-F1322. 3. Lorenzen J, David S, Bahlmann FH, et al. Endothelial progenitor cells and cardiovascular events in patients with chronic kidney disease—a prospective follow-up study. PLoS One. 2010;5:e11477. 4. Kuczmarski JM, Darocki MD, DuPont JJ, et al. Effect of moderate-to-severe chronic kidney disease on flow-mediated dilation and progenitor cells. Exp Biol Med (Maywood). 2011;236:1085-1092. 5. Jie KE, Zaikova MA, Bergevoet MW, et al. Progenitor cells and vascular function are impaired in patients with chronic kidney disease. Nephrol Dial Transplant. 2010;25:1875-1882. 6. Moe SM, Chen NX. Mechanisms of vascular calcification in chronic kidney disease. J Am Soc Nephrol. 2008;19:213-216. 7. Pirro M, Schillaci G, Mannarino MR, et al. Circulating immature osteoprogenitor cells and arterial stiffening in postmenopausal osteoporosis. Nutr Metab Cardiovasc Dis. 2011;21: 636-642. 8. Bellien J, Favre J, Iacob M, et al. Arterial stiffness is regulated by nitric oxide and endothelium-derived hyperpolarizing factor during changes in blood flow in humans. Hypertension. 2010;55:674-680. 9. Hirata Y, Sugiyama S, Yamamoto E, et al. Endothelial function and cardiovascular events in chronic kidney disease. Int J Cardiol. 2014;173:481-486. 10. Hambrecht R, Adams V, Erbs S, et al. Regular physical activity improves endothelial function in patients with coronary artery disease by increasing phosphorylation of endothelial nitric oxide synthase. Circulation. 2003;107:3152-3158. 11. Conraads VM, Beckers P, Bosmans J, et al. Combined endurance/resistance training reduces plasma TNF-alpha receptor levels in patients with chronic heart failure and coronary artery disease. Eur Heart J. 2002;23:1854-1860. 12. Asahara T, Murohara T, Sullivan A, et al. Isolation of putative progenitor endothelial cells for angiogenesis. Science. 1997;275:964-967. 13. Hur J, Yoon CH, Kim HS, et al. Characterization of two types of endothelial progenitor cells and their different contributions to neovasculogenesis. Arterioscler Thromb Vasc Biol. 2004;24:288-293. 14. Van Craenenbroeck EM, Beckers PJ, Possemiers NM, et al. Exercise acutely reverses dysfunction of circulating angiogenic cells in chronic heart failure. Eur Heart J. 2010;31:1924-1934. 15. Heiwe S, Jacobson SH. Exercise training in adults with CKD: a systematic review and meta-analysis. Am J Kidney Dis. 2014;64(3):383-393. 16. Johansen KL, Painter P. Exercise in individuals with CKD. Am J Kidney Dis. 2012;59:126-134. 17. Heiwe S, Jacobson SH. Exercise training for adults with chronic kidney disease. Cochrane Database Syst Rev. 2011;10: CD003236. 18. Van Craenenbroeck AH, Van Craenenbroeck EM, Kouidi E, Vrints CJ, Couttenye MM, Conraads VM. Vascular effects of exercise training in CKD: current evidence and pathophysiological mechanisms. Clin J Am Soc Nephrol. 2014;9: 1305-1318. 19. Eckel RH, Jakicic JM, Ard JD, et al. 2013 AHA/ACC guideline on lifestyle management to reduce cardiovascular risk: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation. 2013;129(25)(suppl 2):S76-S99.

Am J Kidney Dis. 2015;-(-):---

20. Beckers PJ, Denollet J, Possemiers NM, Wuyts FL, Vrints CJ, Conraads VM. Combined endurance-resistance training vs. endurance training in patients with chronic heart failure: a prospective randomized study. Eur Heart J. 2008;29: 1858-1866. 21. Hansen JE, Sue DY, Wasserman K. Predicted values for clinical exercise testing. Am Rev Respir Dis. 1984;129(2 Pt 2): S49-S55. 22. Corretti MC, Anderson TJ, Benjamin EJ, et al. Guidelines for the ultrasound assessment of endothelial-dependent flowmediated vasodilation of the brachial artery: a report of the International Brachial Artery Reactivity Task Force. J Am Coll Cardiol. 2002;39:257-265. 23. Laurent S, Cockcroft J, Van Bortel L, et al. Expert consensus document on arterial stiffness: methodological issues and clinical applications. Eur Heart J. 2006;27:2588-2605. 24. Reference Values for Arterial Stiffness’ Collaboration. Determinants of pulse wave velocity in healthy people and in the presence of cardiovascular risk factors: ‘establishing normal and reference values’. Eur Heart J. 2010;31:2338-2350. 25. Schmidt-Lucke C, Fichtlscherer S, Aicher A, et al. Quantification of circulating endothelial progenitor cells using the modified ISHAGE protocol. PLoS One. 2010;5:e13790. 26. Eghbali-Fatourechi GZ, Modder UI, Charatcharoenwitthaya N, et al. Characterization of circulating osteoblast lineage cells in humans. Bone. 2007;40:1370-1377. 27. Van Craenenbroeck EM, Hoymans VY, Beckers PJ, et al. Exercise training improves function of circulating angiogenic cells in patients with chronic heart failure. Basic Res Cardiol. 2010;105: 665-676. 28. Diggle P, Liang K, Zeger SL. Analysis of Longitudinal Data. Oxford, United Kingdom: Clarendon Press; 1994. 29. Beck DT, Casey DP, Martin JS, Emerson BD, Braith RW. Exercise training improves endothelial function in young prehypertensives. Exp Biol Med (Maywood). 2013;238:433-441. 30. Luk TH, Dai YL, Siu CW, et al. Effect of exercise training on vascular endothelial function in patients with stable coronary artery disease: a randomized controlled trial. Eur J Prev Cardiol. 2012;19:830-839. 31. Hambrecht R, Wolf A, Gielen S, et al. Effect of exercise on coronary endothelial function in patients with coronary artery disease. N Engl J Med. 2000;342:454-460. 32. Hambrecht R, Fiehn E, Weigl C, et al. Regular physical exercise corrects endothelial dysfunction and improves exercise capacity in patients with chronic heart failure. Circulation. 1998;98:2709-2715. 33. Thambyrajah J, Landray MJ, McGlynn FJ, Jones HJ, Wheeler DC, Townend JN. Abnormalities of endothelial function in patients with predialysis renal failure. Heart. 2000;83:205-209. 34. Caglar K, Yilmaz MI, Saglam M, et al. Serum fetuin-a concentration and endothelial dysfunction in chronic kidney disease. Nephron Clin Pract. 2008;108:c233-c240. 35. Gori T, Dragoni S, Lisi M, et al. Conduit artery constriction mediated by low flow a novel noninvasive method for the assessment of vascular function. J Am Coll Cardiol. 2008;51: 1953-1958. 36. Dawson EA, Alkarmi A, Thijssen DH, et al. Low-flow mediated constriction is endothelium-dependent: effects of exercise training after radial artery catheterization. Circ Cardiovasc Interv. 2012;5:713-719. 37. Rakobowchuk M, Harris E, Taylor A, et al. Heavy and moderate interval exercise training alters low-flow-mediated

11

Van Craenenbroeck et al constriction but does not increase circulating progenitor cells in healthy humans. Exp Physiol. 2012;97:375-385. 38. Headley S, Germain M, Wood R, et al. Short-term aerobic exercise and vascular function in CKD stage 3: a randomized controlled trial. Am J Kidney Dis. 2014;64(2):222-229. 39. Montero D, Roche E, Martinez-Rodriguez A. The impact of aerobic exercise training on arterial stiffness in pre- and hypertensive subjects: a systematic review and meta-analysis. Int J Cardiol. 2014;173:361-368. 40. Mustata S, Groeneveld S, Davidson W, Ford G, Kiland K, Manns B. Effects of exercise training on physical impairment, arterial stiffness and health-related quality of life in patients with chronic kidney disease: a pilot study. Int Urol Nephrol. 2011;43: 1133-1141. 41. Gielen S, Schuler G, Adams V. Cardiovascular effects of exercise training: molecular mechanisms. Circulation. 2010;122: 1221-1238. 42. Volaklis KA, Tokmakidis SP, Halle M. Acute and chronic effects of exercise on circulating endothelial progenitor cells in healthy and diseased patients. Clin Res Cardiol. 2013;102:249-257. 43. Petit I, Jin D, Rafii S. The SDF-1-CXCR4 signaling pathway: a molecular hub modulating neo-angiogenesis. Trends Immunol. 2007;28:299-307. 44. Tachibana K, Hirota S, Iizasa H, et al. The chemokine receptor CXCR4 is essential for vascularization of the gastrointestinal tract. Nature. 1998;393:591-594. 45. Yilmaz MI, Stenvinkel P, Sonmez A, et al. Vascular health, systemic inflammation and progressive reduction in kidney function; clinical determinants and impact on cardiovascular outcomes. Nephrol Dial Transplant. 2011;26:3537-3543.

12

46. Baylis C. Nitric oxide deficiency in chronic kidney disease. Am J Physiol Renal Physiol. 2008;294:F1-F9. 47. Schwedhelm E, Boger RH. The role of asymmetric and symmetric dimethylarginines in renal disease. Nat Rev Nephrol. 2011;7:275-285. 48. Balligand JL, Feron O, Dessy C. eNOS activation by physical forces: from short-term regulation of contraction to chronic remodeling of cardiovascular tissues. Physiol Rev. 2009;89:481-534. 49. Green DJ, Maiorana A, O’Driscoll G, Taylor R. Effect of exercise training on endothelium-derived nitric oxide function in humans. J Physiol. 2004;561(1):1-25. 50. Elder SJ, Pisoni RL, Akizawa T, et al. Sleep quality predicts quality of life and mortality risk in haemodialysis patients: results from the Dialysis Outcomes and Practice Patterns Study (DOPPS). Nephrol Dial Transplant. 2008;23:998-1004. 51. Kumar B, Tilea A, Gillespie BW, et al. Significance of selfreported sleep quality (SQ) in chronic kidney disease (CKD): the Renal Research Institute (RRI)-CKD study. Clin Nephrol. 2010;73:104-114. 52. Quinn TJ, Klooster JR, Kenefick RW. Two short, daily activity bouts vs. one long bout: are health and fitness improvements similar over twelve and twenty-four weeks? J Strength Cond Res. 2006;20:130-135. 53. Gui Y. Intermittent exercises reduce the hypertension syndromes and improve the quality of life [published online ahead of print January 22, 2015]. J Sports Med Phys Fitness; 2015. 54. Smart NA, Steele M. A comparison of 16 weeks of continuous vs intermittent exercise training in chronic heart failure patients. Congest Heart Fail. 2012;18:205-211.

Am J Kidney Dis. 2015;-(-):---