Effect of a 3-Year Lifestyle Intervention in Patients with Chronic Kidney Disease: A Randomized Clinical Trial

Significance Statement

Modifiable risk factors, such as low physical activity, contribute to cardiovascular mortality in patients with CKD. In a randomized trial of 160 patients with CKD, a nurse-led, multidisciplinary lifestyle intervention that included exercise training significantly increased cardiorespiratory fitness (peak O₂ uptake) in patients with CKD at 12 months compared with usual care; peak O₂ uptake declined to near baseline levels thereafter but remained elevated compared with usual care after 3 years of follow-up. The intervention also markedly increased exercise capacity and physical function, and prevented adverse anthropometric changes that were observed in the usual care group. Importantly, benefits of the intervention were sustained over 3 years. This study provides evidence of the effectiveness of a multidisciplinary team–supported lifestyle intervention in enhancing the health of patients with CKD.

Visual Abstract

Abstract

Background

Supervised lifestyle interventions have the potential to significantly improve physical activity and fitness in patients with CKD.

Methods

To assess the efficacy of a lifestyle intervention in patients with CKD to improve cardiorespiratory fitness and exercise capacity over 36 months, we conducted a randomized clinical trial, enrolling 160 patients with stage 3–4 CKD, with 81 randomized to usual care and 79 to a 3-year lifestyle intervention. The lifestyle intervention comprised care from a multidisciplinary team, including a nephrologist, nurse practitioner, exercise physiologist, dietitian, diabetes educator, psychologist, and social worker. The exercise training component consisted of an 8-week individualized and supervised gym-based exercise intervention followed by 34 months of a predominantly home-based program. Self-reported physical activity (metabolic equivalent of tasks [METs] minutes per week), cardiorespiratory fitness (peak O₂ consumption [VO_2peak]), exercise capacity (maximum METs and 6-minute walk distance) and neuromuscular fitness (grip strength and get-up-and-go test time) were evaluated at 12, 24, and 36 months.

Results

The intervention increased the percentage of patients meeting physical activity guideline targets of 500 MET min/wk from 29% at baseline to 63% at 3 years. At 12 months, both VO_2peak and METs increased significantly in the lifestyle intervention group by 9.7% and 30%, respectively, without change in the usual care group. Thereafter, VO_2peak declined to near baseline levels, whereas METs remained elevated in the lifestyle intervention group at 24 and 36 months. After 3 years, the intervention had increased the 6-minute walk distance and blunted declines in the get-up-and-go test time.

Conclusions

A 3-year lifestyle intervention doubled the percentage of CKD patients meeting physical activity guidelines, improved exercise capacity, and ameliorated losses in neuromuscular and cardiorespiratory fitness.

CKD is associated with a marked increased risk of cardiovascular morbidity and mortality. The increase in cardiovascular burden is in part due to the high number of traditional and nontraditional cardiovascular risk factors.¹ Cardiovascular risk factors are also associated with accelerated loss of kidney function in patients with CKD.² Physical activity levels are a potentially modifiable risk factor, with low levels and poor physical function associated with increased mortality and adverse clinical events including declining kidney function.³ Incorporating exercise training into a multidisciplinary lifestyle intervention is one strategy that has emerged as an effective tool to reduce cardiovascular risk. Indeed, a recent systematic review demonstrated that exercise improves physical function and cardiorespiratory fitness, but these improvements were not associated with improved kidney function or survival.⁴ Further, only one small study followed patients beyond 12 months.⁵ Thus, the implications of a long-term lifestyle intervention have not been previously investigated.

Current exercise recommendations for patients with CKD suggest patients should perform a combination of aerobic and resistance exercise training,⁶ with several research groups demonstrating this approach to be effective in increasing exercise capacity,^7–9 cardiorespiratory fitness,^10,11 and muscular strength.⁷ Indeed, we have previously demonstrated in this cohort that a 12-month lifestyle intervention was effective in improving cardiorespiratory fitness, exercise capacity, and physical activity levels, as well as reducing cardiovascular risk factors and improving left ventricular function.^12,13 Because patients with CKD have complex health requirements and often see multiple specialist teams there is often a lack of coordinated care between teams. To improve the care of CKD patients, multidisciplinary clinics involving nurse practitioners, dietitians, social workers, and pharmacists have shown promise.¹⁴ However, these clinics have not routinely included input from exercise physiologists or a focus on adopting healthy lifestyle practices such as exercise. We have previously shown that a lifestyle intervention that included exercise training improved cardiorespiratory fitness, body composition, and metabolic function in patients with type 2 diabetes.¹⁵ Thus, the aim of this study was to assess the efficacy of a lifestyle intervention in patients with CKD to improve cardiorespiratory fitness and exercise capacity over 36 months. It was hypothesized that the lifestyle intervention would elicit a significant increase in cardiorespiratory fitness and exercise capacity compared with a usual care control group.

Methods

Participants

The LANDMARK III study (Longitudinal Assessment of Numerous Discrete Modifications of Atherosclerotic Risk Kidney disease) was an open-label, single-center randomized controlled trial. The study was conducted at the Princess Alexandra Hospital in Brisbane, Australia and compared the effects of a 3-year multidisciplinary lifestyle intervention in patients with stage 3–4 CKD (Modification of Diet in Renal Disease [MDRD] eGFR 25–60 ml/min per 1.73 m²). Recruitment, intervention, and follow-up were conducted from 2008 to 2014. Patients were eligible for inclusion if they were aged 18–75 years of age, and had at least one of the following cardiovascular risk factors: BP or lipids not at target; overweight (body mass index [BMI] >25 kg/m²); or poor diabetic control (hemoglobin A1c >7%). Exclusion criteria were the following: intervention for, or, symptomatic coronary artery disease (within 3 months); current heart failure (New York Heart Association classes III and IV) or significant valvular heart disease; pregnant or planning to become pregnant; and life expectancy or anticipated time to dialysis or organ transplant <6 months. The study protocol was approved by the Princess Alexandra Human Research Ethics Committee (HREC 2007/190) and was registered at www.anzctr.org.au (Registration Number ANZCTR12608000337370). At the time of publication, the prespecified primary outcome as defined in the clinical trial registration had not been analyzed. All patients gave written, informed consent to participate in this study. The study complied with the Declaration of Helsinki.

Randomization

Patients were randomized to either the lifestyle intervention or usual care, in a ratio of 1:1 by the clinical trial coordinator using a random assignment computer program. A person external to the study also witnessed the randomization of all participants. There were three levels of stratification: kidney function (eGFR high [>44 ml/min per 1.73 m²] or low [≤44 ml/min per 1.73 m²]), sex (male or female), and diabetes status (yes or no); and randomization was carried out using a permuted block size of 6.

Intervention

Nurse-Led Multidisciplinary Clinic

To address the complex and multifactorial needs of patients, a nurse-led multidisciplinary model of care focused on translating current evidence-based recommendations into clinical practice. The multidisciplinary team consisted of nurse practitioner, accredited exercise physiologist, accredited practicing dietitian, diabetes educator, psychologist, and nephrologist. Social workers were also involved in care management as required. The nurse practitioner coordinated the care for the patients, including the cardiovascular risk factor modification, and led the team. Informed by the Caring for Australians and New Zealanders with Kidney Impairment (CARI) guidelines,¹⁶ the targeted risk factors were BP control, improved blood glucose control, use of antiproteinuria agents (e.g., angiotensin-converting enzyme inhibitors and angiotensin receptor antagonists), cholesterol lowering, healthy lifestyle with weight loss and cardiorespiratory fitness, anemia management, and calcium phosphate balance. The multidisciplinary team met quarterly to review patients.

Lifestyle Intervention

Patients allocated to the lifestyle intervention received a 4-week behavior and lifestyle modification program, followed by ongoing individual intervention with a dietitian. The program was conducted in groups of up to five patients, facilitated by a clinical psychologist, and a dietitian. The program focused on sustainable diet and behavior changes to facilitate weight management with the following weekly topics: Week 1—goal setting, guide to healthy eating, self-monitoring; Week 2—Mediterranean-style diet (education on cholesterol, fats, sugars, and sodium) and developing a healthy meal plan; Week 3—motivating change; and Week 4—sustaining change, including label reading and recipe modification. Patients were provided with a workbook including information on the discussed topics, self-monitoring exercises, homework, and evaluation. Following the 4-week program, patients received individual reviews and counseling by a dietitian every 3 months in person or via telephone, for the remainder of the trial. This dietetics therapy complied with the Evidence-Based Practice Guidelines for Nutritional Management of Chronic Kidney Disease for patients with an eGFR between 25 and 60 ml/min per 1.73 m².¹⁷ In addition to the dietetic and psychology support, patients were also reviewed by a diabetes educator, psychologist, and social worker as required.

Subsequent Visits

Patients allocated to the lifestyle intervention continued to attend the multidisciplinary team for the duration of the study, for further adjustment and monitoring of diabetic and antihypertensive medication. In particular, education about the adjustment of insulin and diuretic therapies during exercise was frequently required. In addition, support and motivation to persist with lifestyle change were emphasized.

Exercise Program

The exercise intervention was delivered in two phases: phase one, 8 weeks of center-based supervised exercise; and phase two, home-based exercise with center-based refresher visits as required. The program aimed to have patients complete 150 min/wk of moderate-intensity aerobic and resistance exercise. Participants were educated on the benefits of vigorous exercise and encouraged to also perform this type of exercise if possible. Exercise intensity was monitored within the session using the Borg scale of relative perceived exertion. Participants were encouraged to exercise at a moderate intensity (relative perceived exertion of 13–14) during aerobic and resistance-based exercises. Each supervised session included a combination of aerobic and resistance training and was individualized (based on comorbidities, motivation, and interests) to allow the best adherence to the program. , The goal of the program was to encourage the participants and provide them with the tools to faciliatate exercise at home (Swiss ball and TheraBand) and education to be as physically active as possible. To maximize exercise adherence and provide ongoing support, telephone follow-up calls were completed approximately every week for the first 3 months, every fortnight from months 6–8, and once a month thereafter, or as needed. Regular gym refresher sessions were also offered to participants to utilize as needed. The telephone follow-up calls and gym refresher sessions were an opportunity to increase motivation to exercise by modifying the exercise prescription as appropriate and encouraging accountability. Furthermore, they allowed an opportunity to address any relevant disruptions or barriers to exercise and provide suggestions to increase adherence. The clinical judgment of the accredited exercise physiologist dictated the appropriate exercise prescription for each individual.

Usual Care Group

Patients randomized to the usual care group received standard nephrological care according to clinical guidelines. Medications were prescribed by physicians as needed and patients were referred to other health professionals or specialists as required.

Experimental Protocol

Participants completed all tests described below, at baseline (before randomization) and 12, 24, and 36 months. Tests are categorized using the domains of fitness model.¹⁸

Cardiopulmonary Exercise Test

An incremental exercise test was performed on a treadmill until volitional fatigue or until signs or symptoms of myocardial ischemia. The Duke Activity Statement Index was performed before exercise to determine the suitable exercise protocol (e.g., Bruce, Balke, or Naughton¹⁹). Expired gases were analyzed by a metabolic cart with a mixing chamber (Vmax29c; SensorMedics, Yorba Linda, CA). Peak oxygen uptake (VO_2peak) was defined as the highest 20 second mean oxygen uptake from the final minute of exercise. Exercise capacity was determined from the treadmill test as metabolic equivalent of tasks (METs) by cardiac assessment system for exercise testing software (CASE V6.51; GE Healthcare, Waukesha, WI) based on the speed and grade of the treadmill at test termination. In addition, the 6-minute walk test was used to measure the total distance walked in 6 minutes over a 20-m course. Patients were tested in an identical order at each visit (i.e., maximal exercise test followed by 6-minute walk test) and testing was conducted during the morning.

Neuromuscular Fitness

To assess changes in neuromuscular strength, a handgrip dynamometer (Jamar 5030 J1;Patterson Medical Ltd., Warrenville, IL) was used to measure grip strength. Neuromuscular power was assessed using the get-up-and-go test.²⁰

Physical Activity

The self-reported Active Australia questionnaire^21,22 was used to assess average weekly physical activity levels from the preceding 6 months. The research exercise physiologist asked each question to the participant and recorded their answer. This allowed the exercise physiologist to check that the participant understood each question.

Clinical Measurements

At each visit, all participants underwent a series of standardized clinical assessments including resting BP by an automated machine (BPM-300; VSM MedTech, Vancouver, Canada) and anthropometric measures (height, weight, and waist and hip circumference). A fasting blood sample was collected for biochemical analysis for the measurement of creatinine,²³ C-reactive protein, glucose, hemoglobin A1c, lipid profile, and hemoglobin. Kidney function was determined as eGFR using the MDRD formula based on the isotope dilution mass spectrometry standardized creatinine assay (MDRD₇₅).²⁴

Safety

All adverse events, as defined by the National Health and Medical Research Council National Statement,²⁵ were recorded. The relevance of each event to the intervention was determined by the study nephrologist.

Study End Points

We report the effect of the lifestyle intervention in patients with CKD as part of the LANDMARK III trial on cardiorespiratory fitness (VO_2peak). The trial was powered to detect a significant change in the primary outcome, carotid intima-media thickness, at 3 years. At the time of publication, the primary outcome has not been analyzed. Additional outcomes reported here include exercise capacity (peak METs, 6-minute walk distance,), neuromuscular fitness (get-up-and-go time, grip strength), physical activity levels, and cardiovascular disease risk factors.

Sample Size and Power Calculation

The study was powered (90%) to detect a between-group difference in cardiorespiratory fitness (VO_2peak) of 20% at 2 years. Assuming baseline VO_2peak of 18±6 ml/kg per minute, 59 patients in each group were required to detect this difference.

Statistical Analyses

Continuous variables were checked for normal distribution using the Kolmogorov–Smirnov test and are presented as mean±SD and mean (95% confidence intervals), with percentages used to describe frequencies for categorical variables. Continuous variables measured over the 3 years were analyzed by generalized linear mixed-effects modeling (specifying a normal distribution of the outcome variables, and an identity link function) adjusted for age and sex (where age was modeled as a continuous linear variable and no age×sex interaction was included) and presented in tables as mean (95% confidence intervals) at each time point (baseline, 12, 24, and 36 months). First-order autoregressive repeated covariance structure was used, based on a smaller Akaike’s information criterion. Multinomial logistic regression was used to assess the group×time interaction on categorical variables (meeting physical activity guidelines). The analysis was performed on available data from participants who had completed baseline exercise testing (n=160). A P value of <0.05 was considered statistically significant. Statistical analyses were performed using IBM SPSS version 25.

Results

Participant Characteristics

One hundred and eighty-seven participants were screened between March 2008 and March 2013, with 160 randomized patients included in this analysis (Figure 1). Three participants were excluded from the analysis of the effect of the intervention on cardiorespiratory fitness. This was because they did not complete baseline exercise testing owing to preexisting severe musculoskeletal issues or morbid obesity. The study was terminated before 29 patients could complete the final follow-up assessment due to funding limitations. Table 1 provides the baseline demographic and clinical characteristics for each group. The participants had multiple comorbidities, with hypertension the most common cardiovascular risk factor.

Figure 1.

Consort diagram. *Study ceased due to funding limitation. CPET, cardiopulmonary exercise test.

Table 1.

Participant baseline characteristics

Variable	Usual Care, n=81	Lifestyle Intervention, n=79
Age, yr	60.4±10.2	59.5±9.9
Sex (male), n (%)	43 (53.0)	48 (60.7)
eGFR, ml/min per 1.73 m2	41.0±10.7	41.2±10.2
BMI, kg/m2	33.8±6.8	33.1±6.0
Systolic BP, mm Hg	140±21	136±21
Diastolic BP, mm Hg	83±13	81±13
Patient history, n (%)
Diabetes	37 (45.7)	35 (44.3)
Hyperlipidemia	59 (72.8)	50 (63.3)
Myocardial infarction	14 (17.3)	10 (12.7)
Heart failure	2 (2.5)	4 (5.0)
Peripheral vascular disease	12 (14.8)	18 (22.7)
Hypertension	78 (96.3)	74 (93.7)
Medications, n (%)
ACEi	45 (55.5)	38 (48.1)
ARB	38 (46.9)	49 (62.0)
Beta-blocker	34 (42.0)	25 (31.6)
Calcium channel blocker	43 (53.1)	33 (41.8)
Thiazide	20 (24.7)	16 (20.2)
Statin	53 (65.4)	51 (64.6)
Insulin	22 (27.2)	17 (21.5)
Cause of kidney disease, n (%)
Diabetes	18 (22.2)	14 (17.7)
Glomerulonephritis	3 (3.7)	7 (8.9)
Polycystic kidney disease	4 (4.9)	4 (5.1)
Renovascular disease	6 (7.4)	6 (7.6)
Other/unknown	46 (56.8)	44 (55.7)

Continuous variables are reported as mean±SD and categorical as n (%). Diabetes includes type 1 diabetes mellitus and type 2 insulin- and non-insulin-dependent diabetes mellitus. Glomerulonephritis includes IgA and FSGS. ACEi, angiotensin converting enzyme inhibitor; ARB, angiotensin receptor blocker.

Physical Activity Levels

The physical activity data are summarized in Table 2 and Supplemental Table 1. Both groups reported very low levels of physical activity at baseline with 29% of the lifestyle intervention and 40% of the usual care groups meeting guidelines. The 3-year intervention increased the percentage of participants in the lifestyle intervention group who met the physical activity guidelines (increased from 29% to 63% at 3 years). In comparison, the physical activity levels declined over time in the usual care group with 32% of participants meeting recommended levels (group × time interaction, P<0.001).

Table 2.

Effect of a 3-year lifestyle intervention on physical activity, cardiorespiratory fitness, exercise capacity, and neuromuscular fitness compared with usual care in patients with CKD

	Usual Care				Lifestyle Intervention
Variable	Baseline	Year 1	Year 2	Year 3	Baseline	Year 1	Year 2	Year 3	P Value
Meeting PA guidelines	29 (40)	22 (34)	13 (27)	12 (32)	22 (29)	39 (61)a	30 (60)b	24 (63)b	<0.001
VO2peak, L/min	2.1 (2.0 to 2.2)	2.1 (1.9 to 2.2)	2.0 (1.9 to 2.1)	1.9 (1.7 to 2.0)a	2.0 (1.9 to 2.1)	2.2 (2.0 to 2.3)b	2.0 (1.9 to 2.1)	1.9 (1.7 to 2.0)c	0.031
Respiratory exchange ratio	1.05 (1.03 to 1.08)	1.03 (0.99 to 1.06)	1.06 (1.02 to 1.10)	1.04 (1.00 to 1.09)	1.04 (1.01 to 1.06)	1.07 (1.04 to 1.10)	1.08 (1.04 to 1.11)	1.07 (1.03 to 1.11)	0.14
Peak heart rate, bpm	145 (140 to 150)	140 (136 to 146)	142 (136 to 147)	140 (134 to 147)	146 (141 to 150)	150 (145 to 156)	146 (140 to 151)	144 (138 to 150)	0.07
6-minute walk, m	468 (446 to 489)	476 (453 to 499)	468 (443 to 493)	467 (442 to 493)	457 (435 to 479)	499 (476 to 522)a	498 (473 to 520)a	481 (456 to 506)c	0.009
Grip strength, kg	32 (30 to 34)	31 (29 to 33)	30 (28 to 32)	29 (27 to 31)	32 (30 to 34)	31 (29 to 33)	29 (27 to 31)	29 (27 to 30)	0.45
Get-up-and-go test time, s	5.4 (5.1 to 5.8)	5.9 (5.6 to 6.3)a	6.0 (5.6 to 6.4)b	5.7 (5.3 to 6.2)	5.7 (5.4 to 6.1)	5.7 (5.3 to 6.1)	5.6 (5.2 to 6.0)	5.6 (5.2 to 6.1)	0.006

Variables are reported as mean (95% confidence interval) for continuous variables and n (%) for categorical variables. Linear mixed-effects model adjusted for age and sex (continuous variables) or logistic regression (categorical variables) identifying the main effect of group×time. Superscript letters indicate within-group significant difference from baseline. PA, physical activity.

^aP<0.001.

^bP<0.01.

^cP<0.05.

Cardiorespiratory Fitness and Exercise Capacity

There was a significant group × time interaction for both absolute and relative VO_2peak and METs (Figure 2, Table 2). The intervention significantly increased relative VO_2peak by 9.7% at year 1 (+2.13 ml/kg per minute) before declining to below baseline levels at year 3. In contrast, in the usual care group, VO_2peak declined gradually over the 3 years, with an overall 14.7% decline at year 3 (−3.28 ml/kg per minute, P<0.001). The VO_2peak of the lifestyle intervention group was 10.7% higher than the usual care group at the final follow-up visit. There was a marked increase in peak METs (31.7%) at year 1 in the lifestyle intervention group (+2.08 METs, Figure 2B). This remained elevated at each of the follow-up visits. The usual care group maintained relatively stable METs throughout the follow-up period. Consistent with the effects on peak exercise capacity, the intervention increased the 6-minute walk distance by 5% (Table 2).

Figure 2.

Changes in cardiorespiratory fitness and exercise capacity over 3 years in the usual care and lifestyle intervention groups. (A) Cardiorespiratory fitness (VO_2peak) and (B) exercise capacity (METs). METs were determined from the speed and grade of the treadmill at test termination. BL, baseline.

Neuromuscular Fitness

The timed get-up-and-go test performance deteriorated in the usual care group but remained unchanged in the lifestyle intervention group (Table 2). There were no statistically significant changes in the grip strength in either group.

Cardiovascular Risk Factors

There were significant group×time interaction effects for measures of anthropometry, with reductions in weight, BMI, and waist and hip circumference observed in the lifestyle intervention group (Table 3). In contrast, the usual care group had adverse changes to anthropometry including increases in weight, BMI, and waist and hip circumferences. There was no significant effect of the intervention on systolic or diastolic BP, or other traditional risk factors. Kidney function declined similarly over the 3 years and there were no significant changes in inflammation measured by C-reactive protein.

Table 3.

Effect of the 3-year lifestyle intervention on cardiovascular risk factors and markers of kidney function

	Usual Care				Lifestyle Intervention				P Value
Variable	Baseline	Year 1	Year 2	Year 3	Baseline	Year 1	Year 2	Year 3
Weight, kg	95 (91 to 99)	97 (92 to 101)a	98 (94 to 103)b	98 (94 to 103)b	93 (88 to 97)	91 (86 to 95)a	91 (87 to 96)	92 (88 to 96)	0.001
BMI, kg/m2	34 (32 to 35)	34 (33 to 36)a	35 (33 to 36)b	35 (33 to 36)c	33 (32 to 35)	32 (31 to 34)a	33 (31 to 34)	33 (31 to 34)	0.001
Waist, cm	110 (107 to 113)	111 (107 to 114)	113 (109 to 116)a	113 (110 to 117)c	109 (105 to 112)	106 (103 to 110)a	107 (104 to 111)	109 (105 to 113)	0.026
Hip, cm	114 (111 to 117)	115 (113 to 118)	117 (114 to 119)c	117 (114 to 120)b	114 (111 to 117)	113 (110 to 116)	114 (111 to 117)	115 (112 to 118)	0.03
Waist:hip ratio	0.97 (0.95 to 0.98)	0.96 (0.95 to 0.98)	0.97 (0.95 to 0.99)	0.97 (0.95 to 0.99)	0.95 (0.93 to 0.97)	0.94 (0.93 to 0.96)	0.94 (0.93 to 0.96)	0.94 (0.93 to 0.97)	0.61
Systolic BP, mmHg	140 (136 to 144)	137 (132 to 142)	139 (134 to 144)	136 (130 to 142)	137 (132 to 141)	133 (128 to 138)	136 (131 to 141)	133 (127 to 140)	0.99
Diastolic BP, mmHg	82 (80 to 85)	81 (78 to 84)	82 (79 to 85)	79 (76 to 83)	80 (77 to 83)	79 (76 to 82)	79 (76 to 82)	78 (75 to 82)	0.67
eGFR, ml/min per 1.73 m2	39.7 (37.2 to 42.2)	39.5 (37.0 to 42.1)	37.4 (34.8 to 35.0)	37.7 (35.0 to 40.5)	39.8 (37.2 to 42.3)	39.6 (37.0 to 42.2)	37.7 (35.0 to 40.3)	37.1 (34.3 to 39.9)	0.96
Creatinine, mmol/L	145 (135 to 156)	150 (139 to 161)	160 (148 to 172)	167 (155 to 180)	144 (131 to 155)	149 (137 to 160)	162 (150 to 173)	169 (156 to 182)	0.96
C-reactive protein, mg/L	5.9 (3.2 to 8.5)	9.2 (6.2 to 12.2)	7.0 (3.5 to 10.4)	6.3 (1.8 to 10.9)	6.8 (4.2 to 9.5)	8.4 (5.5 to 11.3)	6.6 (3.2 to 9.9)	5.6 (1.3 to 9.9)	0.89
Glucose, mmol/L	6.8 (6.1 to 7.4)	6.8 (6.1 to 7.4)	7.0 (6.3 to 7.7)	6.5 (5.6 to 7.3)	7.0 (6.3 to 7.6)	6.4 (5.7 to 7.0)	6.7 (6.0 to 7.4)	6.6 (5.8 to 7.4)	0.47
Hemoglobin A1C, %	6.6 (6.2 to 6.9)	6.8 (6.4 to 7.1)	6.9 (6.5 to 7.3)	6.9 (6.5 to 7.3)	6.6 (6.3 to 7.0)	6.7 (6.3 to 7.0)	6.9 (6.6 to 7.3)	6.9 (6.6 to 7.3)	0.72
Total cholesterol, mmol/L	4.4 (4.2 to 4.7)	4.3 (4.1 to 4.6)	4.3 (4.0 to 4.6)	4.3 (3.9 to 4.6)	4.4 (4.1 to 4.6)	4.2 (3.9 to 4.4)	4.2 (4.0 to 4.5)	4.0 (3.7 to 4.3)	0.85
Triglyceride, mmol/L	2.1 (1.7 to 2.4)	2.0 (1.6 to 2.3)	2.2 (1.8 to 2.6)	2.0 (1.6 to 2.5)	1.6 (1.3 to 1.9)	1.6 (1.3 to 1.9)	1.6 (1.3 to 2.0)	1.6 (1.2 to 2.1)	0.91
HDL, mmol/L	1.1(1.0 to 1.2)	1.2(1.1 to 1.3)	1.2(1.1 to 1.3)	1.2(1.1 to 1.3)	1.3(1.2 to 1.3)	1.3(1.2 to 1.4)	1.2(1.1 to 1.3)	1.3(1.2 to 1.4)	0.53
LDL, mmol/L	2.4 (2.2 to 2.6)	2.4 (2.2 to 2.6)	2.1 (1.9 to 2.4)	2.2 (1.9 to 2.5)	2.5 (2.3 to 2.7)	2.3 (2.0 to 2.5)	2.3 (2.1 to 2.5)	2.1 (1.8 to 2.3)	0.09
Hemoglobin, g/dl	13.2 (12.9 to 13.6)	13.1 (12.8 to 13.5)	13.2 (12.6 to 13.4)	13.0 (12.5 to 13.4)	13.2 (12.8 to 13.5)	13.1 (12.7 to 13.4)	12.9 (12.6 to 13.3)	12.9 (12.5 to 13.3)	0.99

P values indicate group×time interaction by linear mixed-effects model adjusted for age and sex. Superscript letters indicate within-group significant difference from baseline. Variables are reported as mean (95% confidence intervals).

^aP<0.05.

^bP<0.001.

^cP<0.01.

There was one adverse event directly related to exercise training during the study. An intervention participant had a syncopal episode during a supervised gym session and was admitted to the hospital, but was released the same day with no further follow-up required.

Discussion

We report the first randomized controlled trial to investigate the effects of a 3-year lifestyle intervention on cardiorespiratory fitness, physical activity levels, and cardiovascular health in patients with CKD. The main finding is that the intervention doubled the fraction of patients meeting physical activity guidelines. This resulted in a significant increase in cardiorespiratory fitness at year 1 and attenuated the declines in fitness that were observed in the usual care group over the remaining 2 years of follow-up. The intervention also improved exercise capacity (METs), physical performance (6-minute walk distance and get-up-and-go test time), and prevented adverse changes in anthropometry when compared with the deterioration in these measures that were observed in the usual care group. These effects were sustained for the duration of the study.

The sustained beneficial effects of the lifestyle intervention program were likely due to the regular, complementary support of the multidisciplinary team who through the intervention sought to improve the management of multiple risk factors. Indeed, other exercise programs with sustained follow-up and good adherence to the intervention have also demonstrated beneficial effects in patients with impaired kidney function.^26,27 Over the 3-year study period, patients were provided with continuity in care from the team which resulted in the lifestyle intervention group demonstrating increased physical activity levels, attenuation of weight gain, and preservation of physical performance. Although exercise interventions alone have been shown to favorably affect these outcomes in patients with CKD,^26,28 this study is the first, to our knowledge, to combine an exercise program with other allied health services, embedded in a nurse-led model of care. Indeed, lack of personnel in renal departments with exercise and physical activity knowledge has been identified as a common barrier to implementing exercise programs as part of kidney multidisciplinary teams.²⁹ Our findings demonstrate the benefits of a multidisciplinary team that includes an exercise specialist and provide evidence of the potential clinical effect.

Increasing physical activity through regular exercise is recognized as a cornerstone for cardiovascular disease prevention.³⁰ We observed a 9.7% increase in VO_2peak after 1 year of intervention compared with the usual care group. Although VO_2peak declined thereafter, the between-group difference remained. In addition to improved cardiorespiratory fitness, METs (calculated from time-on-test) also increased by approximately 31% (approximately 2.1 METs). The improvement in aerobic fitness measures was clinically significant given that a 1-MET increase in exercise capacity is associated with a 13% and 15% lower risk of all-cause mortality and cardiovascular disease, respectively.³¹ Reduced exercise capacity is an independent risk factor for cardiovascular disease outcomes, given that improvements in exercise capacity even without changes in traditional risk factors such as BP and lipids are still likely to favorably affect prognosis.³² Prior exercise studies of varying duration and follow-up periods in patients across all stages of CKD have demonstrated increases in VO_2peak.^33–35 The results of this study, combined with the reported beneficial effects on all-cause mortality and cardiovascular morbidity from other recent long-term evaluations of exercise rehabilitation programs for patients with CKD,³⁶ support the inclusion of exercise programs as standard of care.

VO_2peak declined in the usual care group, at a rate greater than the expected effect of aging (approximately 5% per year instead of normal age-related losses of 1% per year).³⁷ This increased rate of decline in fitness was also observed by Headley et al. VO_2peak declined by approximately 10% over 48 weeks of follow-up in a group of CKD patients.¹¹ Further, the average cardiorespiratory fitness in the usual care group approached levels consistent with “functional disability” at 3 years (e.g., VO_2peak ≤18.0 ml/kg per minute).³⁸ Thus, the lifestyle intervention prevented marked reductions in aerobic fitness that were observed in the usual care group, which may have important implications for maintaining functional independence.

The significant increase in peak METs at year 3 in the lifestyle intervention group, without an increase in VO_2peak, suggests a disconnect between exercise capacity and cardiorespiratory fitness in these patients. Cardiorespiratory fitness, quantified as VO₂ uptake, is an integrative measure of the cardiovascular, respiratory, circulatory, and skeletal muscular systems to deliver and utilize oxygen, whereas exercise capacity is the functional performance of these systems, and is additionally affected by factors such as movement efficiency, pacing, and psychologic motivation. It is well known that there is large intra-individual variability in the changes in cardiorespiratory fitness in response to a standardized exercise training program.³⁹ It has been suggested that nonresponders to changes in VO_2peak are not necessarily nonresponders in other measurements of training response.^40,41 There may be other mechanisms occurring through the pathogenesis of CKD which are dampening the physiologic adaptations to exercise training, but not exercise capacity. Indeed, a 12-month analysis of the current cohort demonstrated no change in autonomic function between the lifestyle intervention and usual care groups.⁴² This lack of improvement in the autonomic nervous function could at least partly explain the discrepancy between exercise capacity and physiologic improvements in oxygen uptake.

Although there is good evidence to support nurse-led care models to improve cardiovascular risk factor modification in patients with CKD,^43,44 this study is the first to include a structured exercise program as an integral component of the multidisciplinary approach. Interestingly, there was no significant benefit of the program on cardiovascular risk factors including BP, lipids, or glycemic control over the 3 years. We speculate that this is due to the high use of cardioprotective medications at baseline, therefore stalling further optimization of therapies. However, it is important to note that both groups retained an above normal systolic BP in the range of 130–140 mm Hg throughout the study. Given the complex interaction between hypertension and the progression of kidney disease, further investigation of adjunctive therapies is warranted.

A recent meta-analysis found that exercise increased cardiorespiratory fitness and HDL levels in patients with CKD, but did not alter BP or other lipids.⁴⁵ This lack of improvement in BP, may be related to vascular dysfunction as it is suggested that patients with CKD may have irreversible vascular changes that are unresponsive to exercise training. Indeed, short-term exercise studies in patients with CKD have failed to observe changes in central arterial stiffness,^46,47 which may explain the lack of change in BP in this study. Future research should investigate the dose-response of exercise intensity (i.e., increase in arterial shear stress) on vascular remodeling in patients with CKD. Additionally, intervening in earlier stages of CKD before vascular changes become irreversible might be more effective and requires investigation.

The lack of additional cardiovascular risk factor improvement in the lifestyle intervention group may reflect the increased focus on managing cardiovascular risk by nephrology teams to slow the progression of kidney disease. Additionally, prior studies in CKD patients that have examined the effect of exercise on cardiovascular risk have been largely inconclusive, due to small sample sizes and short follow-up.⁴⁸ However, the benefits of exercise for cardiovascular health likely extend beyond changes to traditional risk factors. For example, exercise has been found to improve ventricular,⁴⁹ hemostatic,⁵⁰ and immune function.⁵¹ Further, the intervention attenuated increases in body weight observed in the usual care group. Thus, the potential cardiovascular benefits of exercise are multifactorial, and although not all cardiovascular risk factors were improved in this study, other important indicators of cardiovascular health were enhanced.

Kidney function remained relatively stable in both groups, highlighting the efficacy of the care provided by the nephrology team in preserving kidney function. The evidence on the effects of exercise interventions on eGFR is mixed in patients with reduced kidney function. A recent meta-analysis has shown an improvement in eGFR through a reduction in serum creatinine after an exercise intervention compared with a usual care group.⁵² However, another meta-analysis published in the same year (2020) with twice as many participants showed no significant effect of exercise training on eGFR compared with usual care, despite significant improvements in VO_2peak.⁴ The latter study included studies with longer interventions compared with the previously mentioned meta-analysis. As such, it may be that previous studies have been of insufficient length to accurately report on the long-term effects of exercise on eGFR. Indeed, our 3-year intervention demonstrated that despite significant improvements in important cardiovascular disease risk factors, such as exercise capacity, an exercise intervention does not change eGFR decline compared with a usual care group. This was also found in the meta-analysis by Zhang et al.,⁵³ who showed a small improvement in eGFR with short-term interventions, with no difference in eGFR found with 3- to 6-month or 6- to 12-month interventions.

Limitations

Due to the trial ending prematurely, we may have underestimated the true effects of the intervention. Further, due to the longitudinal nature of this study, the causes of missing data are not consistent at each visit (Figure 1), and for some patients, the reason that they did not complete the exercise testing may be related to their health status. To minimize the biases introduced by either omitting patients with incomplete data or using multiple imputation methods, we included all individuals who completed exercise testing at their baseline testing visit. However, this assumes that the dropout before the termination of the trial was nonrandom, which is not able to be delineated. A further possible contributor to the missing data is that healthier patients could be more likely to complete the follow-up testing, introducing a survivorship bias. It is also important to acknowledge that due to the longitudinal nature of this study there were unexpected changes to study personnel which has consequently affected the analysis and reporting of the primary outcome as specified in the clinical trials registration.

As with all multicomponent interventions, we are unable to differentiate the contribution of individual components of the intervention to the change in outcomes. However, the integration of the multiple disciplines was likely complementary to the overall success of the intervention and is representative of multidisciplinary models of care. A cost-benefit analysis of multidisciplinary care should be investigated in future studies. The Active Australia questionnaire, used to measure physical activity levels, is based on patients’ recall of their average physical activity levels per week in the previous 6 months. Thus, there is potential for issues with patient recall and this approach may have limited the accuracy of the reporting. Further, the precise dose of activity performed across the 3 years was not documented for each participant. Future studies should seek to quantify the dose of exercise performed.

The changes in body mass between the two groups may have slightly influenced estimations of eGFR. However, our group did analyze the 12-month data from both lifestyle intervention and usual care groups using both creatinine-based eGFR measures (dependent on muscle mass) and cystatin C–based eGFR measures (independent of muscle mass) and found similar estimates before and after 12 months of both intervention and usual care.²³ As such, any influence of changes in body mass are likely to have little influence on eGFR estimates.

Conclusions

The 3-year lifestyle intervention doubled the fraction of CKD patients achieving physical activity guidelines. Furthermore, the lifestyle intervention improved exercise capacity and ameliorated cardiorespiratory fitness losses usually seen in the CKD population. These findings have important implications for increasing the access to exercise programs for patients with CKD to ameliorate the cardiovascular risk in these individuals.

Disclosures

B. Douglas reportsbeing a member of the Renal Society of Australasia (RSA) and Australian College of Nurse Practitioners; is a member of the editorial board of the RSA Journal and reports honoraria with Fresenius Medical Care. E.J. Howden is supported by a National Heart Foundation Australia Future Leader Fellowship (102536). N. Isbel reports consultancy agreements with Alexion Pharmaceuticals; reports honoraria with Alexion; reports scientific advisor or membership with Alexion; is on the speakers bureau with Alexion; and reports being a member of the Australian and New Zealand Society of Nephrology (ANZSN) and the Transplantation Society of Australia and New Zealand. R. Krishnasamy reports personal fees from Amgen and Baxter Healthcare; reports grant support from Baxter, outside the submitted work; reports being a recipient of Queensland Advancing Clinical Research Fellowship; reports honoraria with Shire Australia; reports scientific advisor or membership with ANZSN (President Elect); and reports being a member of the International Society of Nephrology and CARI. T. Stanton reports scientific advisor or membership with Heart, Lung & Circulation. All remaining authors have nothing to disclose.

Funding

This research was supported by the National Health and Medical Research Council–funded Centre for Clinical Research Excellence Vascular and Metabolic Health (grant GNT0455832).

Data Sharing Statement

Data sharing may be available upon request.

Supplemental Material

This article contains the following supplemental material online at http://jasn.asnjournals.org/lookup/suppl/doi:10.1681/ASN.2021050668/-/DCSupplemental.

Supplemental Table 1. Self-reported weekly minutes of physical activity.