Long-Term Physical Exercise for Preventing CKD in Older Adults

Visual Abstract

Abstract

Key Points

• Physical exercise mitigates several kidney-related pathophysiological pathways, yet its effect on kidney function remains underexplored.

• This randomized clinical trial found that high-intensity interval training significantly reduced the risk of rapid eGFR decline in older adults.

• Physical exercise can preserve kidney function, benefiting public health and individual patient outcomes similar to other cardiovascular diseases.

Background

CKD is a growing global public health issue. Physical exercise mitigates several kidney-related pathophysiological pathways, yet its effect on kidney function remains underexplored. We investigated the dose-response effect of physical exercise on kidney function preservation in older adults.

Methods

We conducted a post hoc analysis of the Generation 100 Study, a 5-year open-label, randomized, parallel-arm clinical trial conducted from 2012 to 2018 in Trondheim, Norway. All inhabitants aged 70–77 years were invited (n=6966). Main exclusion criteria were dementia; severe, uncontrolled cardiovascular disease or hypertension; or conditions precluding exercise. The control group (n=385) received information on national physical activity recommendations. Intervention groups underwent supervised moderate-intensity continuous training (70% of peak heart rate) for 50 minutes (n=380) twice weekly for 5 years or high-intensity interval training (90% of peak heart rate) for 4 minutes ×4 (n=391) twice weekly for 5 years. The primary outcome was cystatin C–based rapid eGFR decline (>5 ml/min per 1.73 m² per year).

Results

A total of 1156 participants were randomized, with median (interquartile range) age 72 (3) years and eGFR 95 (20) ml/min per 1.73 m². Oxygen uptake increased by 1.8, 2.3, and 3.3 ml/kg per minutes in the control, moderate-intensity, and high-intensity groups, respectively. Rapid eGFR decline occurred in 117 (30%), 108 (28%), and 92 (23%) participants, respectively. Compared with controls, relative risk (RR) was 0.93 (95% confidence interval [CI], 0.75 to 1.16) for the moderate-intensity and 0.75 (95% CI, 0.59 to 0.95) for high-intensity groups, demonstrating a significant dose-response relationship (P for trend 0.02). In observational analyses, participants with decreasing moderate-to-vigorous intensity activity (>−20 min/wk) had RR 1.30 (95% CI, 0.93 to 1.83) for rapid eGFR decline, while those with increasing activity (>+20 min/wk) had RR 0.73 (95% CI, 0.53 to 0.99) compared with stable activity after adjusting for baseline age, sex, peak oxygen uptake, and eGFR.

Conclusions

High-intensity interval training significantly reduced the risk of rapid eGFR decline in older adults.

Clinical Trial registry name and registration number:

ClinicalTrials.gov: NCT01666340.

Introduction

CKD is a growing global public health issue with enormous effects on individuals and societies because of its high prevalence, associated treatment costs, morbidity, and mortality.¹ Diabetes and hypertension are major risk factors of initiation of CKD and progression to kidney failure,^2,3 and associated pathophysiological mechanisms include endothelial dysfunction, ischemia, and inflammation leading to microvascular rarefaction and fibrosis.⁴ Physical exercise is known to mitigate several of these pathways, preventing vascular diseases in the heart, brain, and musculoskeletal systems.⁵ However, the effects of physical exercise on kidney function have not been adequately studied.

For patients with CKD, international guidelines recommend engaging in moderate-intensity continuous training (MICT) for at least 150 minutes per week, but the supporting evidence is very weak (recommendation level 1D) and guidelines have strongly advocated for additional evidence.⁶ Several small randomized clinical trials (RCTs) have shown that exercise induced significant improvements in cardiovascular fitness and muscle strength in nondialysis patients with CKD, but a meta-analysis found no risk reduction for mortality or CKD progression.⁷ Recently, a randomized trial demonstrated that MICT can preserve kidney function in inactive older adults with moderately decreased eGFR at inclusion.⁸ Whether the intensity or type of exercise affects CKD progression remains unknown, and more knowledge is especially needed among the older population where 25%–30% experience a rapid kidney function decline.^9,10

When compared with moderate-intensity training, high-intensity interval training (HIIT) has demonstrated promising results in improving cardiorespiratory capacity, endothelial function, and other critical aspects of cardiovascular health,¹¹ but its potential as a strategy to prevent rapid eGFR loss has not been evaluated. Optimally, physical exercise should preserve kidney function in patients with early and more advanced CKD, thereby contributing to both population-based and high-risk prevention strategies. Such combined management strategies are widely recommended for noncommunicable diseases,^12,13 and even small improvements can have significant public health effect given the large number of individuals treated, leading to a shift of the entire distribution of risk.¹⁴

The Generation 100 randomized clinical trial (RCT) studied the effect of cardiovascular exercise over 5 years in older adults.¹⁵ The primary aim was to determine the effect of MICT or HIIT on overall mortality. The primary trial results have been reported elsewhere and showed lower all-cause mortality in those randomized to HIIT compared with MICT or controls.¹⁶ Secondary aims included assessing the effects of exercise on morbidity and cardiovascular risk factors, including kidney function, a key component of the cardiovascular–kidney–metabolic syndrome and integral to cardiovascular health.^15,17 In this post hoc analysis, we studied the effect of randomization to the different intensities of physical exercise on kidney function decline in older, community-based adults with normal or mildly decreased eGFR. We hypothesized that randomization to supervised HIIT would more effectively reduce the risk of rapid eGFR decline compared with general information on exercise recommendations or supervised MICT.

Methods

Design

The Generation 100 study was a 5-year, open-label, randomized, parallel-arm trial investigating health-related effects of three different intensities of systematic, long-term physical exercise training in older adults (ClinicalTrials.gov: NCT01666340).^15,16 Participants were assigned at random to one of three groups: (1) following national guidelines for physical activity (control group),^18,19 (2) supervised MICT, or (3) supervised HIIT. A computerized, web-based block randomization service from The Unit for Applied Clinical Research at the Norwegian University of Science and Technology was used for randomizing participants 2:1:1 to control, MICT, and HIIT, stratified by sex and marital status. Starting in August 2012, all inhabitants (6966) aged 70–77 years in Trondheim, Norway, were invited to participate.

All individuals who signed the consent form were invited for an interview, blood sampling, and general health assessment including strength and cardiopulmonary fitness measurements at baseline and years 1, 3, and 5. We excluded only participants with severe uncontrolled hypertension (>220/110 mm Hg), symptomatic valvular heart disease, hypertrophic cardiomyopathy, unstable angina pectoris, heart failure, severe arrhythmia, dementia, or disabilities that precluded exercise. All test personnel were blinded to the intervention. For this study, kidney function testing was conducted in the HIIT and MICT groups, as well as in a 50% random sample of the control group, resulting in three equal-sized groups of approximately 400 participants each.

Intervention

Everybody, including the control group, was advised to follow national guidelines recommending 5 days with 30 minutes of moderate-to-vigorous intensity exercise per week.^18,19 For the MICT group, two of these sessions were extended and supervised, and for the HIIT group, two sessions were supervised doing HIIT instead of MICT. The two MICT sessions were supervised by exercise physiologists and consisted of 50 minutes of continuous exercise at approximately 70% of their peak heart rate, with a level of 11–14 on the Borg Rating of Perceived Exertion scale. This scale assesses exercise intensity and the participant's endurance, where 6 represents no exertion and 20 represents maximum exertion.²⁰ The HIIT group attended two supervised sessions per week, which included a 10-minute warm-up followed by four 4-minute intervals at around 90% of their peak heart rate, with a perceived exertion level of 15 or higher on the Borg scale.²¹ The MICT and HIIT exercise sessions were isocaloric throughout the intervention period, in an effort to ensure that only exercise intensity varied between the groups.²² Correspondingly, there was no significant difference in body mass index (BMI) between the groups during the study.²³ The HIIT group did more running and cycling and generally had a higher intensity in all types of training, which also included brisk walking, swimming, cross-country skiing, and other types of endurance and resistance training.²⁴ Every 6 weeks, the HIIT and MICT groups met separately for standardized, supervised spinning sessions (ergometer cycling) with heart rate monitors to ensure they achieved their recommended exercise intensities.

Clinical Outcomes

We used the cystatin C–based CKD Epidemiology Collaboration equation (2012) to estimate GFR.²⁵ Muscle mass variations may introduce substantial bias to creatinine-based GFR estimates, whereas cystatin C depends much less on muscle mass.^26,27 Furthermore, results on the basis of cystatin C are less modified by other risk factors such as age, sex, BMI, smoking, and physical exercise,^27,28 and cystatin C more often correctly displays linear risk associations among elderly patients with mild CKD.^27,29

Our primary outcome measure was rapid kidney function decline, which is defined internationally as a decrease >5 ml/min per 1.73 m² per year by the Kidney Disease Improving Global Outcomes organization.³⁰ In addition, we used >30% eGFR decline and annual eGFR slope as secondary outcomes.

Measurements

Baseline characteristics included self-reported sociodemographic factors, including age, sex, marital status, and education level. Medical history was used to establish exclusion criteria, and participants reported on general health (very good, good, fair, or poor), smoking habits (never smoked, former smoker, or current smoker), and prevalent diabetes and cardiovascular diseases (angina pectoris, myocardial infarction, or stroke). In addition, physical examinations were conducted to measure BMI, heart rate, and systolic and diastolic BPs. Hypertension was defined as a BP above 140/90 or using antihypertensive medication.

Participants' achieved exercise levels were categorized as low, medium, and high intensity on the basis of perceived Borg scores (6–10, 11–14, and 15–20, respectively). Peak oxygen uptake (VO_{2 peak}) was measured during all study visits through an incremental exercise test on a treadmill using a Cortex MetaMax II gas analyzer (Leipzig, Germany). The protocol gradually increased the workload by 1 km/h or 2% inclination every 1–2 minutes until exhaustion. Participants with previously diagnosed heart disease were closely monitored by a physician (BP and electrocardiography). Physical activity levels were monitored for 1 week before all visits using the triaxial Actigraph GT3X+ accelerometer (Actigraph, FL) and categorized using Troiano cutoffs.³¹

Cystatin C was analyzed in a single batch using the Gentian Cystatin C Immunoassay, a particle-enhanced turbidimetric immunoassay method, on an ABX Pentra 400 instrument (Horiba, Japan). The cystatin C method demonstrated an intrarun coefficient of variation of 1.8%.

Statistical Analyses

We conducted intention-to-treat analyses to determine the proportion of participants experiencing study outcomes over the 5-year intervention period. We also analyzed outcomes by the participants' achieved level of exercise intensity, as per protocol analyses. To calculate relative risk (RR) estimates and use all available data, we used generalized linear models (GLMs) with a log link function. In accordance with current RCT recommendations,^32,33 adjustments were made for stratification variables used during randomization (sex and marital status). The dose-effect relationship between training and risk of rapid kidney function decline was tested by contrasting the marginal linear predictions after the GLM analysis. The effects of increasing physical activity levels on risk of rapid eGFR decline and eGFR slope were quantified with GLM and linear regression analyses, respectively. On the basis of our prior research findings, we estimated a baseline probability of 0.30 for rapid kidney function decline in the control group, anticipated a RR reduction to 0.80 in the MICT group, and hypothesized a further reduction to 0.70 in the HIIT group. With 400 participants per group, we calculated 84% power to detect significant differences using a two-sided Cochran–Armitage trend test at an α level of 0.05.

Results

Study Population Characteristics

After a 10-month enrollment period, 1177 participants (22% of all adults aged 70–77 years in Trondheim, Norway) were included and randomly assigned to three levels of physical exercise (Figure 1). In total, 1156 participants had GFR estimates from two or more follow-up visits and were included in this analysis: 385 in the control group, 380 in the MICT group, and 391 in the HIIT group. At baseline, the median (interquartile range) age was 72 (3) years and eGFR was 95 (20) ml/min per 1.73 m² (Table 1). Half were male (50%), 5% had diabetes mellitus, and 12% had prevalent cardiovascular disease. No significant differences in these or other baseline characteristics were observed across the intervention groups.

Figure 1.

Flow chart describing inclusion of participants, randomization, and follow-up. ex., participant excluded because of medical concerns; w/d, participant withdrawn.

Table 1.

Baseline characteristics of participants by randomization group

Characteristics	Controls (n=385)	MICT (n=380)	HIIT (n=391)
Age, yra	72 (3)	72 (3)	72 (3)
Male, No. (%)	187 (49)	183 (48)	203 (52)
Living alone, No. (%)	94 (25)	107 (28)	89 (23)
Lower education, No. (%)	158 (43)	157 (43)	169 (45)
Reduced general health, No. (%)	50 (14)	48 (13)	41 (11)
Current smoking, No. (%)	25 (7)	36 (9)	29 (7)
Diabetes, No. (%)	16 (4)	18 (5)	23 (6)
Cardiovascular disease, No. (%)	44 (11)	54 (14)	36 (9)
BMI, kg/m2	25.9 (3.3)	26.0 (3.7)	26.1 (3.7)
Systolic BP, mm Hg	135 (18)	134 (18)	135 (18)
Diastolic BP, mm Hg	75 (10)	75 (9)	75 (10)
Heart rate, bpm	65 (11)	65 (11)	65 (11)
VO2 peak, ml/min per kg	28.8 (6.5)	28.6 (6.6)	28.9 (6.4)
MVPA, min/d	27 (21)	29 (23)	27 (18)
Hgb, g/dl	14.4 (1.2)	14.3 (1.2)	14.5 (1.1)
HbA1c (%)	5.6 (0.4)	5.7 (0.5)	5.7 (0.4)
Cholesterol, mg/dl	221 (44)	216 (44)	216 (42)
HDL-cholesterol, mg/dl	69 (20)	67 (19)	67 (20)
eGFR, ml/min per 1.73 m2a	96 (19)	94 (19)	95 (21)
eGFR ≥90 ml/min per 1.73 m2	255 (66)	232 (61)	242 (62)
eGFR ≥75–89 ml/min per 1.73 m2	73 (19)	91 (24)	87 (22)
eGFR ≥60–74 ml/min per 1.73 m2	39 (10)	38 (10)	45 (12)
eGFR <60 ml/min per 1.73 m2	18 (5)	19 (5)	17 (4)

Continuous data are given as mean (SD) and categorical data as number (%). Lower education is <12 years. Reduced general health is bad or fair general health (versus good or very good). Moderate or vigorous physical activity was measured with an accelerometer over 1 week. GFR was estimated with the serum cystatin C 2012 CKD Epidemiology Collaboration formula. BMI, body mass index; bpm, beats per minute; Hgb, hemoglobin; HIIT, high-intensity interval training; MICT, moderate-intensity continuous training; MVPA, moderate-to-vigorous intensity physical activity; VO_{2 peak}, peak oxygen uptake.

^aAge and eGFR are reported as median (interquartile range).

Exposure to Physical Activity

The study interventions led to significant changes in overall weekly physical activity over time and between groups (Supplemental Table 1): There was no significant change in reported physical activity intensity in the control group over time. By contrast, after 1 year of intervention, more of the activity in the MICT group was of medium intensity (increased from 62% to 80%, P < 0.001), and substantially more of the activity in the HIIT group was performed as high-intensity exercise (increased from 26% to 59%, P < 0.001). The absolute increases in VO_{2 peak} from baseline to year 1 for the control, MICT, and HIIT groups were 1.8, 2.3, and 3.3 ml/kg per minutes, respectively (P ≤ 0.001 for all). From year 1 to year 5, there was a decrease in physical activity and VO_{2 peak} in all groups, but the HIIT group maintained ≥1 ml/kg per minute higher oxygen uptake compared with the other groups at all times (Supplemental Figure 1).

Primary Outcome

At study completion in June 2018, a total of 317 participants (27%) had experienced rapid decline in eGFR, defined by >5 ml/min per 1.73 m² per year, after a mean follow-up of 3.9 years (Figure 2). There were 117 participants among those randomized to the control group, 108 in the MICT group, and 92 in the HIIT group, corresponding to 5-year risks of 30%, 28%, and 23%, respectively. The risk of rapid eGFR decline in the MICT group was not statistically different from the control group (RR, 0.93; 95% confidence interval [CI], 0.75 to 1.16), whereas participants randomized to the HIIT group had significantly lower risk than the control group (RR, 0.75; 95% CI, 0.59 to 0.95). The magnitude of risk reductions was consistent with a dose-response relationship (P for trend 0.02).

Figure 2.

Absolute and RR (95% CIs) for rapid eGFR decline per intervention group. CI, confidence interval; HIIT, high-intensity interval training; MICT, moderate-intensity continuous training; RR, relative risk.

Subgroup analyses were performed to evaluate potential differences on the risk of rapid kidney function decline across key subgroups (Figure 3). The effect of randomization to the MICT or HIIT group on rapid kidney function decline was similar irrespective of sex, prevalence of diabetes, hypertension, obesity, smoking status, baseline physical activity, hemoglobin VO_{2 peak}, or eGFR (P interaction all ≥0.27).

Figure 3.

Subgroup analysis to evaluate potential effect modification (interaction) on the risk of rapid kidney function decline. (A) Risk among participants randomized to MICT compared with the control group, and (B) participants randomized to HIIT compared with the control group. Continuous variables were dichotomized by their median baseline levels (MVPA 24 min/d, Hgb 14.4 g/dl, VO_{2 peak} 28.1 ml/kg per minutes, and eGFR 95 ml/min per 1.73 m²). DM, diabetes mellitus; Hgb, hemoglobin; MVPA, moderate-to-vigorous intensity physical activity; VO_{2 peak}, peak oxygen uptake.

Because adherence to randomized training type and dose was only moderate (Supplemental Table 1), we also report observational associations of achieved activity level and intensity with kidney outcomes (i.e., per protocol). Participants increasing their moderate-to-vigorous intensity physical activity with more than 20 minutes per week had a RR of 0.73 (95% CI, 0.53 to 0.99) for rapid eGFR decline compared with those with a stable activity (−20 to +20 min/wk) after adjusting for age, sex, VO_{2 peak}, and eGFR at baseline (Table 2). Participants decreasing their physical activity more than 20 minutes per week had a 30% higher risk of rapid eGFR decline (RR, 1.30; 95% CI, 0.93 to 1.83) when adjusted for the same covariates.

Table 2.

Risk of primary outcome (rapid eGFR decline) by observed change in moderate-to-vigorous physical activity per week

Physical Activity	Participants	Total	5-yr Risk	Risk Ratio	Adjusted RRa
MVPA (Δ min per week)
Decreasing (<-20)	20	42	47.6% (32.5 to 62.7)	1.39 (0.99 to 1.96)	1.30 (0.93 to 1.83)
Stable (−20 to 20)	156	457	34.1% (29.8 to 38.5)	1.00 (Ref)	1.00 (Ref)
Increasing (>20)	36	136	26.5% (19.1 to 33.9)	0.78 (0.57 to 1.05)	0.73 (0.53 to 0.99)

Risk ratios are based on a generalized linear model. Numbers in parenthesis are 95% confidence intervals. Rapid decline was defined as eGFR decline >5 ml/min per 1.73 m² per year. Physical activity was measured as minutes per week with moderate or vigorous activity using an Actigraph accelerometer for 1 week, and data are change from baseline to year 1. MVPA, moderate-to-vigorous intensity physical activity; RR, relative risk.

^aAdjusted for age, sex, peak oxygen uptake, and eGFR at baseline.

Secondary Outcomes

Results were similar when we evaluated >30% reduction in eGFR and eGFR slopes (Table 3). Participants randomized to the HIIT group had RR 0.73 (95% CI, 0.57 to 0.94) for a 30% reduction compared with controls. Although there were no statistically significant differences in mean slopes of eGFR change across randomized treatment groups, the point estimates were directionally consistent with the binary outcomes. Observational analysis showed that participants who increased their moderate-to-vigorous intensity physical activity had a 0.8 ml/min slower decrease in eGFR per year (95% CI, 0.04 to 1.57). Similarly, increasing physical activity was associated with RR 0.71 (95% CI, 0.50 to 1.01) for 30% eGFR reduction. None of the participants progressed to kidney failure or initiated dialysis during trial follow-up.

Table 3.

Risk of secondary outcomes by randomization group and by observed change in physical activity

Analysis by Randomization Group				Analysis by Observed Change in Physical Activity
Exposure	30% eGFR Reduction		eGFR Slope	Exposure		30% eGFR Reduction		eGFR Slope
	Cases/Total	RR (95% CI)	β (95% CI)			Cases/Total	RR (95% CI)	β (95% CI)
Control	108/385	Reference=1	Reference=0	Decrease		26/42	1.28 (0.85 to 1.92)	−0.61 (−1.86 to 0.64)
MICT	96/380	0.90 (0.71 to 1.13)	0.22 (−0.12 to 0.56)	Stable		135/457	Reference=1	Reference=0
HIIT	83/391	0.73 (0.57 to 0.94)	0.29 (−0.06 to 0.64)	Increase		29/136	0.71 (0.50 to 1.01)	0.80 (0.04 to 1.57)

Analyses by randomization group were adjusted for sex and marital status. Analyses by observed change were adjusted for age, sex, peak oxygen uptake, and eGFR at baseline. Decrease, more than 20 minutes' decrease of moderate-to-vigorous activity per week; increase, more than 20 minutes' increase of moderate-to-vigorous intensity activity per week compared with baseline. CI, confidence interval; HIIT, high-intensity interval training; MICT, moderate-intensity continuous training; RR, relative risk.

Safety

As previously reported,¹⁶ no cardiovascular disease–related events occurred during the supervised exercise sessions, but two MICT participants and one HIIT participant experienced fracture injuries while exercising on a slippery surface during outdoor training.

Discussion

In this study of community-dwelling older Norwegians, we found that randomization to supervised HIIT resulted in a 25% reduction in the risk of rapid kidney function decline over 5 years compared with active controls. The analysis revealed a dose-response relationship, with HIIT showing more pronounced benefits than MICT and the control group. Analyses of secondary outcomes (>30% eGFR decline and annual eGFR slope) by randomization groups and by achieved exercise level supported our main findings.

Most people with CKD do not adhere to lifestyle recommendations such as maintaining a healthy weight, following a nutritious diet, and engaging in regular physical activity.³⁴ Observational studies have consistently shown a strong association between higher levels of physical activity and slower declines in kidney function,^35–39 but inferring cause-and-effect relationships from these studies is difficult. Randomized clinical trials address many of these issues, but past kidney-related exercise trials have had limitations, including small sample sizes (fewer than 100 participants); selective populations; reliance on serum creatinine, which is significantly influenced by muscle mass; and short durations, typically over months.⁷ Furthermore, a meta-analysis of all relevant studies up to 2020 found no significant kidney function benefits.⁷ Recently, however, we examined the effect of physical exercise in a RCT among 1199 sedentary, high-risk older Americans with a mean eGFR at inclusion of 54 ml/min per 1.73 m². Supervised moderate-intensity training over 2 years significantly slowed the rate of kidney function decline in this trial, with a mean difference of 1 ml/min per 1.73 m² and an odds ratio for rapid eGFR decline of 0.79.⁸

In the current Generation 100 trial, we tested for the first time the effect of a substantially more intensive exercise program on long-term eGFR decline in a general population of older individuals not selected to be at high risk. We found that randomization to supervised HIIT was significantly more effective than simply advising adherence to traditional MICT-focused guidelines. Likewise, our observational results showed that higher achieved doses of moderate-to-vigorous intensity exercise were associated with lower risk of rapid eGFR (>5 ml/min per year) and with a more favorable eGFR slope. For cardiovascular diseases, HIIT has been shown to be superior to MICT in several meta-analyses,^40,41 but to our knowledge, there have been no prior evaluations of how HIIT relates to changes in kidney function. The mechanisms underlying the beneficial effects of HIIT on kidney health are not fully understood but likely extend beyond weight reduction and improved cardiovascular fitness per se. Experimental studies have shown that HIIT more effectively reduces inflammation and renin-angiotensin system activation compared with MICT.^42–44 Furthermore, direct beneficial effects of contracting muscles include improved maintenance of glomerular and peritubular capillaries, improved mitochondrial biogenesis and respiration, and reduced endothelial dysfunction and inflammation,^45–47 all of which are important pathophysiological mechanisms in CKD.^48,49

Physical exercise offers significant potential for widespread public health prevention efforts and for disease treatment because of its low cost, minimal side effects, and numerous benefits for various organ systems, now including the kidney. Despite these advantages, implementation has been slow and limited. However, a recent review found that HIIT protocols were well tolerated and feasible even for older adults with a wide range of diseases and low physical capacity at baseline.⁵⁰ The general HIIT principle is that each person performs short bouts of exercise with an intensity close to the maximum of what the individual person can achieve in their current state of fitness. This study demonstrates that high-intensity physical activity is feasible as a long-term intervention, with adherence rates comparable with those in studies involving less-intensive activities among older adults.⁵¹

Kidney function decline is recognized as one of the top ten modifiable risk factors globally,⁵² and the current CKD framework emphasizes prevention and progression control at all disease stages.⁶ Our previous study on MICT demonstrated significant preservation of kidney function among sedentary high-risk elders with a mean eGFR of 54 ml/min per 1.73 m²,⁸ and now, we extend this to older adults with fewer comorbidities and higher baseline eGFR. In general, relative treatment effects are similar across the spectrum of risks and disease stages,⁵³ and although absolute effects can be smaller in population-based studies, important preventions can be achieved because of the high number of participants involved. Furthermore, many patients do not display the typical high-risk profile. Population-based treatment strategies should therefore be at least as effective as high-risk strategies because of their broader reach and ability to shift the overall distribution of risk within the population.

Clearly, not all elderly in a general population will benefit directly from the kidney function–preserving effects of exercise. However, substantial proportions of our cohort had eGFR below 90 ml/min and/or rapid kidney function decline, indicating that many would have developed CKD with clinical effects during their remaining lifetime. In addition to kidney function preservation demonstrated in this study, physical exercise has numerous other beneficial effects on cardiovascular, respiratory, and muscle bone dimensions; inflammation and nutritional parameters; and health-related quality of life in patients with CKD.^54–56 Kidney disease in particular is an asymptomatic disease until there is irreversible organ damage, and thus, efforts related to prevention and intervention at early stages are paramount to curbing the epidemic of CKD.⁶ Our data support a broader implementation of physical training in older community-living persons, showing positive benefits for kidney health and diseases at various stages, a dose-response relationship similar to that seen in other cardiovascular diseases,^57,58 and an excellent safety profile with no cardiovascular events during supervised exercise sessions.¹⁶

This study has important limitations. Although we had few medical exclusions, and randomization counteracts selection bias, participants were healthier and may have been more motivated to perform and adhere to physical exercise than those electing not to participate. This could lead to higher adherence rates but potentially also fewer participants reaching our study outcome and reducing statistical power. There was also substantial overlap in training intensity between the groups, which also would bias toward the null hypothesis. Evaluation of achieved exercise in supplemental observational analyses supports the same conclusions with stronger effect sizes, consistent with the hypothesized effect of this crossover limitation. Generalization of our results to other age groups, ethnicities, and populations with other disease stages may currently be limited because our population was predominantly White with moderate-to-high cardiovascular fitness and normal or only mildly decreased GFR at randomization.⁵⁹ However, the results relating MICT to controls in persons with lower eGFR were qualitatively similar in prior studies. Key strengths to this study include its RCT design, number of participants, high participation rates and adherence, and the duration of 5 years. Kidney function was estimated using serum cystatin C measurements, reducing the potential of bias induced by muscle mass changes in response to the physical exercise intervention.^27–29 Although cystatin C can be influenced by some inflammatory and endocrine diseases,⁶⁰ this is probably less relevant because these conditions are less frequent and have smaller effects on GFR estimates than the non-GFR determinants of creatinine.

This 5-year RCT involving community-dwelling older Norwegians demonstrated a dose-response relationship between physical activity and rapid kidney function decline. The results were similar across a wide variety of subgroups, and the achieved physical exercise level was associated with the magnitude of eGFR changes in companion analyses. Preservation of kidney function should be recognized as one of the numerous health benefits of high-intensity exercise training.

Supplementary Material

jasn-36-1352-s001.pdf ^{(1.4MB, pdf)}

jasn-36-1352-s002.pdf ^{(179.3KB, pdf)}

Disclosures

Disclosure forms, as provided by each author, are available with the online version of the article at http://links.lww.com/JSN/F66.

Funding

S.I. Hallan: Norges Forskningsråd, Norges Teknisk-Naturvitenskapelige Universitet, and Helse Midt-Norge.

Author Contributions

Conceptualization: Stein I. Hallan, Knut A. Langlo, Dorthe Stensvold, Ulrik Wisløff.

Data curation: Stein I. Hallan, Dorthe Stensvold.

Formal analysis: Stein I. Hallan, Knut A. Langlo, Marius A. Øvrehus.

Funding acquisition: Knut A. Langlo, Dorthe Stensvold, Ulrik Wisløff.

Investigation: Nils P. Aspvik, Knut A. Langlo, Dorthe Stensvold, Ulrik Wisløff.

Methodology: Nils P. Aspvik, Dorthe Stensvold, Ulrik Wisløff.

Project administration: Dorthe Stensvold, Ulrik Wisløff.

Resources: Knut A. Langlo, Dorthe Stensvold, Ulrik Wisløff.

Supervision: Stein I. Hallan, Joachim H. Ix, Knut A. Langlo, Michael G. Shlipak.

Writing – original draft: Stein I. Hallan, Knut A. Langlo.

Writing – review & editing: Nils P. Aspvik, Joachim H. Ix, Marius A. Øvrehus, O. Alison Potok, Solfrid Romundstad, Michael G. Shlipak, Dorthe Stensvold, Ulrik Wisløff.

Data Sharing Statement

Partial restrictions to the data and/or materials apply. We are not permitted to share individual data from the current trial, but we are open to collaborative research with researchers worldwide, who can have access to analyzed data from our university.

Supplemental Material

This article contains the following supplemental material online at http://links.lww.com/JSN/F65.

Supplemental Methods

Supplemental Table 1. Exposure to physical activity by randomization group on the basis of self-reported physical activity intensity level.

Supplemental Figure 1. Training exposure measured as changes in cardiorespiratory fitness (mean VO_{2 peak} and SEM) for the 5-year period.