Abstract
Introduction
Mildly reduced renal function and elevated urine protein levels are each prospectively associated with hip fracture risk in older adults. Here we determine whether these markers are associated with reduced appendicular muscle performance.
Methods
We prospectively examined the associations of urine albumin:creatinine ratio (ACR) and reduced estimated glomerular filtration rate (eGFR) with longitudinal changes in grip strength and gait speed >2 years in 2317 older community-dwelling men and women (median age 77 years). The median ACR was 9.8 [interquartile range (IQR) 5.40–21.50] mg/g creatinine and the median eGFR was 71.6 (IQR 59.1–83.56) mL/min/1.73 m2. Models were adjusted for demographic factors, clinical history and biochemical measures in four candidate pathways: diabetes, oxidative stress, inflammation and fibrosis.
Results
In demographic- and covariate-adjusted models, a 2-fold higher baseline urine ACR was associated with longitudinal changes of −0.17 kg [95% confidence interval (CI) −0.29 to −0.06) in grip strength and −1.10 cm/s (95% CI −1.67 to −0.53) gait speed per year. Corresponding estimates for a 10 mL/min/1.73 m2 lower baseline eGFR were −0.13 kg (95% CI −0.23 to −0.04) and −0.89 cm/s (95% CI −1.37 to −0.40), respectively. The associations of a 2-fold higher baseline ACR and a 10 mL/min/1.73 m2 lower baseline eGFR using cystatin C with grip strength and gait speed were equivalent to ∼1.2–1.9 additional years of age. Adjustment for covariates in candidate pathways did not attenuate these estimates.
Conclusions
In older adults, higher ACR and lower eGFR are potential risk factors for a decline of physical performance >2 years.
Keywords: albuminuria, appendicular muscle mass, gait speed, grip strength, physical performance
INTRODUCTION
Epidemiological studies demonstrate that end-stage renal disease and advanced chronic kidney disease (CKD) are associated with frailty [1, 2], a constellation of symptoms and signs that includes involuntary weight loss, exhaustion, low physical activity and performance deficits in muscle strength and gait speed [3]. Frailty is associated with a poor prognosis [4]. This phenotype predicts incident falls, worsening mobility or activities of daily living, disability, hospitalization and death [3]. Moreover, the rates of decline in gait speed and grip strength influence the risk of disability and mortality [5, 6]. Lesser degrees of renal disease [estimated glomerular filtration rate (eGFR) <60 and <45 mL/min/1.73 m2, respectively] are also associated with graded declines in physical performance and with prefrailty, a precursor of frailty [7–9].
From these data, it may be extrapolated that the association of renal disease with physical decline may occur early in the process of renal decline, when eGFR is ≥60 mL/min/1.73 m2. Little is known about the progression of objectively measured performance deficits at this stage of renal disease and whether urinary protein excretion and impaired renal filtration have differential effects on physical performance. This issue is of special importance among the elderly in whom the age-related decline in filtration rate accelerates and in whom the prevalence rates of albuminuria (∼35% by ages 70–75 years) and CKD (eGFR <60 mL/min/1.73 m2; 17% at ages 60–79 years) are high [10, 11].
We have previously shown that albuminuria—adjusted for eGFR, falling and frailty—is associated with reduced hip and total body bone mineral density and with increased hip fracture risk [12]. We have also shown that mildly reduced eGFR, adjusted for a history of falling, physical activity and gait speed are associated with hip fracture risk [13]. Since appendicular bone and muscle share common arterial supplies, risk factors for impairment (e.g. advancing age, oxidative stress, inflammation, insulin resistance and fat infiltration) and common function (ambulation), it follows that early markers of renal disease may also be associated with low peripheral muscle quality.
Although the causes of physical dysfunction are manifold, a few appear to be relevant to renal disease. In particular, diabetes [14], inflammation [15] and oxidative stress [16] have been implicated as mediators of aging and muscle dysfunction in kidney disease. For example, carboxymethyl-lysine (CML) (a marker of oxidative stress) predicts hip fracture, coronary heart disease, stroke and disability risks [17–19]. Moreover, with increasing recognition of fibrosis as a common final pathway in CKD [20], it too appears to be a plausible candidate mechanism to link early renal impairment with physical dysfunction.
To our knowledge, no prospective study has evaluated the comparable associations of both eGFR and albuminuria with risk of physical function decline in a population-based cohort of elders. To elucidate the independent associations of mild impairments in eGFR and albuminuria with prospective muscle strength and gait speed, and the roles of candidate pathways in these associations, we prospectively studied participants from the Cardiovascular Health Study (CHS), an intensively phenotyped cohort of older Americans.
MATERIALS AND METHODS
CHS is an ongoing longitudinal study of adults ≥65 years of age in four US communities drawn from Medicare lists [21]. In 1989–90, 5201 participants were recruited, followed by an additional 687 African Americans in 1992–93. All participants gave informed consent upon study entry. Institutional review board approval was received at all clinical sites. From 1989–90 to 1998–99, participants were seen in clinic annually and had annual telephone contact between clinic visits.
Of the 5888 original and African American participants, 3826 (∼65%) were seen in 1996–97, the baseline year for this study. At that time, 3051 had grip strength testing, 3144 had gait speed measurement, 2828 had urine testing and 3115 had cystatin measured. Two years later, in 1998–99, during a follow-up examination, 2888 had grip strength and 2889 had gait speed re-measured. A total of 2317 participants had complete baseline and follow-up muscle function assessment and baseline urine and renal testing (See Figure, Appendix).
Kidney function testing
In 1996–97, the baseline year for the current analysis, participants underwent spot urine testing in the fasting state for albumin and creatinine levels. Urine albumin was measured soon after collection on the Array 360 CE Protein Analyzer (Beckman Instruments, Fullerton, CA, USA) and urine creatinine on the Ektachem 700 Analyzer (Eastman Kodak, Rochester, NY, USA). Urinary albumin excretion was estimated as the urinary albumin:creatinine ratio (ACR) in milligrams of albumin per gram of creatinine. This method, based on a spot urine, yields results comparable to those from a 24-h urine collection [22]. At the same visit, participants underwent fasting phlebotomy. Cystatin levels were measured using an immunonephelometric assay [23]. We have reported that cystatin C is a stronger predictor of the risk of death and cardiovascular events in elderly persons than is creatinine [23]. We estimated the eGFR using cystatin C (eGFRcys) using the Chronic Kidney Disease Epidemiology Collaboration equations for cystatin.
Tests of muscle performance
Two tests of muscle function were performed in 1996–97 [24]: grip strength and gait speed. For grip strength, forearm muscle strength was measured by a hand-held Jamar A dynamometer, which recorded the force in kilograms of three maximal attempts with the subject’s dominant and nondominant hands. The best of three attempts in the dominant hand was used. For gait speed, the time (to 0.1 s) required for a participant to walk a 15-foot course at his or her usual pace after starting from a standstill. Values were converted into centimeter per second analytically.
These two tests were repeated at a follow-up visit 2 years later (1998–99) {median follow-up 2.0 years [interquartile range (IQR) 1.95–2.03]}.
Covariates
Information was collected from the baseline year on demographics (sex, age, race), smoking status, medical history, seated blood pressure and technician-measured height, weight and waist circumference. Coronary heart disease was centrally adjudicated, as previously described [25]. Diabetes and impaired fasting glucose were defined using American Diabetes Association cut points and the use of hypoglycemic medication.
To examine candidate pathways, we used three measurements taken from the 1996–97 visit and performed at the CHS central laboratory [26]. High-sensitivity C-reactive protein (CRP) levels were measured with an assay developed in the CHS central laboratory [27]. Plasma levels of CML (an advanced glycation end product) were used as a measure of oxidative stress [17]. Procollagen type III N-terminal pro-peptide (PIIINP) was evaluated as a measure of systemic fibrosis [28].
Statistical methods
Baseline characteristics of the study cohort are presented by the presence or absence of albuminuria (≥30 mg albumin/g creatinine). To evaluate whether the ACR was prospectively associated with muscle strength and function during follow-up, we examined all participants with evaluable urine and blood samples and completed serial grip strength and gait speed tests. Because of a long right skew, ACR was log2 transformed and winsorized at the 95th percentile (i.e. values above this threshold were set to the 95th percentile). The associations of log2 ACR and eGFRcys with grip strength and gait speed change were examined through incrementally adjusted linear regression models. In all cases, the dependent variable was the performance measure in 1998–99, with adjustment for baseline values as recommended [29]. These models included Model 1 adjusted for age, gender and race; Model 2 additionally adjusted for height, waist circumference, current smoking, current use of alcohol, glycemic status, systolic blood pressure, prevalent cardiovascular disease, current hypertension medications and CRP; and Model 3 additionally adjusted for markers of two potential mediators, CML and PIIINP levels. We tested for linearity in the relationships of log2 ACR and eGFRcys with both outcomes using generalized additive models, with no meaningful departures from linearity.
Several sensitivity analyses were performed to gain further insights into the relationship of renal impairment and functional decline. (i) To explore changes in outcomes of ‘clinical importance’, we further examined the effect of ACR and eGFRcys on end points deemed clinically significant but conservative from prior studies of aging—a decline of at least 0.5 kg grip strength [30] or a decline of ≥30 cm/s [31] gait speed. Logistic regression models were adjusted as above. (ii) To put our findings into clinical context, we estimated the number of years of aging equivalent to the changes in the β coefficient for grip strength and gait speed. (iii) To test whether our findings were relevant to advanced older age or not, we categorized our analyses by the median age of the cohort (77 years). (iv) Finally, we examined whether the use of angiotensin-converting enzyme inhibitors (ACEis) or angiotensin II receptor blockers (ARBs) (which can lower ACR levels) modified the associations of ACR and CKD with functional decline.
Analyses were done using R (R Foundation for Statistical Computing, Vienna, Austria) [32].
RESULTS
A summary of baseline variables appears in Supplementary data, Table S1. The median eGFR was 71.6 (IQR 59.51–83.56) mL/min/1.73 m2 and the median ACR was 9.8 (IQR 5.4–21.5) mg/g creatinine.
Baseline characteristics of participants characterized by the presence or absence of ALB are shown in Table 1. Participants with ALB were older and more likely male and to have hypertension, diabetes and to smoke, but less heavily. They had higher CRP, CML and PIIINP levels.
Table 1.
Baseline characteristics of the CHS cohort from 1996 to 1997 categorized by albuminuria (≥30 mg/g creatinine) status
| Characteristics | Total | No albuminuria | Albuminuria | P-value |
|---|---|---|---|---|
| Sample, n | 2317 | 1903 | 414 | |
| Male, n (%) | 950 (41.0) | 751 (39.4) | 199 (48.1) | <0.01 |
| Age (years), n | 77.6 ± 4.4 | 77.4 ± 4.4 | 78.4 ± 4.9 | <0.01 |
| Black race, n (%) | 362 (15.6) | 299 (15.7) | 63 (15.2) | 0.86 |
| Education ≥12 grades, n (%) | 1141 (49.4) | 945 (49.8) | 196 (47.6) | 0.44 |
| Height (m) | 1.64 ± 0.09 | 1.64 ± 0.09 | 1.64 ± 0.1 | 0.8 |
| Waist (cm) | 97.3 ± 12.74 | 97.39 ± 12.66 | 96.88 ± 13.14 | 0.47 |
| BMI (kg/m2) | 27.01 ± 4.52 | 27.06 ± 4.51 | 26.78 ± 4.55 | 0.26 |
| BMI <18, n | 30 (1.3) | 23 (1.1) | 9 (2.2) | |
| BMI 18–24.9, n | 766 (33.2) | 617 (32.6) | 149 (36.1) | |
| BMI 25–29.9, n | 1016 (44) | 856 (45.2) | 160 (38.7) | |
| BMI >30, n | 496 (21.5) | 401 (21.2) | 95 (23) | |
| Smoking status, n (%) | n = 2280 | n = 1873 | n = 407 | 0.05 |
| Current | 171 (7.5) | 135 (7.2) | 36 (8.8) | |
| Former | 991 (43.5) | 798 (42.6) | 193 (47.4) | |
| Never | 1118 (49) | 940 (50.2) | 178 (43.7) | |
| Alcohol, n (%) | n = 2309 | n = 189 | n = 413 | 0.78 |
| None | 1291 (55.9) | 1054 (55.6) | 237 (57.4) | |
| 1–7 drinks/week | 771 (33.4) | 639 (33.7) | 132 (32.0) | |
| >7 drinks/week | 247 (10.7) | 203 (10.7) | 44 (10.7) | |
| Glucose status, n (%) | n = 2308 | n = 189 | n = 412 | <0.01 |
| Normal | 1800 (78) | 1542 (81.3) | 258 (62.6) | |
| Impaired fasting glucose | 163 (7.1) | 125 (6.6) | 38 (9.2) | |
| Diabetes | 345 (14.9) | 229 (12.1) | 116 (28.2) | |
| Hypertension, n (%) | 1465 (63.4) | 1132 (59.6) | 333 (80.6) | <0.01 |
| Hypertension medications | 1349 (58.2) | 1042 (54.8) | 307 (74.2) | <0.01 |
| Prevalent CHD, n (%) | 532 (23) | 394 (20.7) | 138 (33.3) | <0.01 |
| eGFR mean (mL/min/1.73 m2) | 71.65 ± 19.0 | 73.54 ± 18.2 | 62.9 ± 20.0 | <0.01 |
| eGFR <60 mL/min/1.73 m2, n (%) | 586 (25.3) | 408 (21.4) | 178 (43) | <0.01 |
| CRP >3 mg/L, n (%) | 982 (42.7) | 782 (41.4) | 200 (48.5) | <0.01 |
| CML (ng/mL) | 620.6 ± 212.7 | 611.8 ± 196.6 | 661.1 ± 271.5 | <0.01 |
| PIIINP (ng/mL) | 4.8 ± 1.7 | 4.7 ± 1.6 | 5.2 ± 2.3 | <0.01 |
| Daily physical activity (kcal), median (IQR) | 787.5 (270–1695) | 795 (280–1704) | 682.5 (245–1665) | 0.55 |
| Baseline gait speed (cm/s) | 900 ± 200 | 900 ± 200 | 900 ± 300 | <0.01 |
| Baseline grip (kg) | 28.0 ± 9.8 | 28.0 ± 9.9 | 28.0 ± 9.6 | 0.95 |
Values presented as mean ± standard deviation unless stated otherwise. All participants had complete baseline (1996–97) and follow-up (1998–99) muscle functional assessment and baseline urine and renal testing (1996–97).
Primary analyses of grip strength
The mean difference in grip strength during the 2-year follow-up was −2.00 kg [95% confidence interval (CI) −2.17 to −1.83; P < 0.001] and the median change was −2.00 kg (IQR −4.67–0.67). Linear regression models of grip strength and gait speed change using ACR (per 2-fold higher level) and eGFRcys (per 10 mL/min/1.73 m2/year lower level) are shown in Table 2.
Table 2.
Change in grip strength in kilograms and change in gait speed in centimeters/second at follow-up
| Measure of muscle function | β | SE (95% CI) | P-value |
|---|---|---|---|
| Grip | |||
| Model 1 | |||
| Log2 ACR | −0.23 | 0.05 (−0.33 to −0.12) | <0.001 |
| eGFRcys10 | −0.10 | 0.05 (−0.20 to −0.01) | 0.03 |
| Model 2 | |||
| Log2 ACR | −0.17 | 0.06 (−0.29 to −0.06) | 0.003 |
| eGFRcys10 | −0.13 | 0.05 (−0.23 to −0.04) | 0.007 |
| Model 3 | |||
| Log2 ACR | −0.16 | 0.06 (−0.28 to −0.05) | 0.006 |
| eGFRcys10 | −0.12 | 0.05 (−0.22 to −0.02) | 0.02 |
| Gait | |||
| Model 1 | |||
| Log2 ACR | −1.39 | 0.27 (−1.91 to −0.87) | <0.001 |
| eGFRcys10 | −1.15 | 0.23 (−1.60 to −0.69) | <0.001 |
| Model 2 | |||
| Log2 ACR | −1.10 | 0.29 (−1.67 to −0.53) | <0.001 |
| eGFRcys10 | −0.89 | 0.25 (−1.37 to −0.40) | <0.001 |
| Model 3 | |||
| Log2 ACR | −1.06 | 0.30 (−1.64 to −0.48) | <0.001 |
| eGFRcys10 | −0.74 | 0.26 (−1.24 to −0.23) | 0.005 |
Each model shows two coefficients: per doubling of ACR (log2 ACR) and 10 mL/min/1.73 m2 lower eGFR (eGFRcys10). Models (M) are adjusted as follows: M1: grip or gait speed at baseline, age, sex, race, clinic; M2: M1 + height, waist circumference, current smoking, current alcohol, log(CRP), glucose status (normal, impaired fasting glucose, diabetes mellitus), systolic blood pressure, prevalent CVD, hypertension medications; M3: M2 + CML and PIIINP levels.
After adjustment for age, sex and race, a 2-fold higher ACR was associated with a 0.23 kg decline in grip strength, while a 10 mL/min/1.73 m2/year lower eGFRcys was associated with a 0.1 kg decrease per year. Further adjustments (Model 2), including glycemic status and CRP level, modestly attenuated the association of ACR with grip strength, but results remained statistically significant. These adjustments did not impact the association of lower eGFRcys on grip strength. Further adjustment for CML and PIIINP had no appreciable effect on either estimate. CML (per ng/mL) was associated with a −0.001 kg (95% CI −0.002–0.0; P = 0.051) change in grip strength (Model 3); per standard deviation of CML (225 ng/mL) grip strength was reduced 0.179 kg (−0.359–0.001; P = 0.051) and gait speed by 0.934 cm/s (−1.834 to −0.035; P = 0.042).
The regression coefficient for 1 year of aging was −0.16 kg of grip strength. Hence a doubling of baseline ACR was equivalent to −0.23/−0.16 = ∼1.4 years of aging, while a 10 mL/1.73 m2/min/year lower eGFR was equivalent to 1.2 years of aging (−0.12/−0.10).
When analyses were repeated using the dichotomous variable ALB (i.e. ≥30 mg albumin/g creatinine), similar associations were observed (Supplementary data, Table S2).
Primary analyses of gait speed
The mean and standard deviation change of gait speed during 2-year follow-up was −6.1 cm/s (95% CI −7.0 to −5.1; P < 0.001) and the median decline was 1.0 cm/s (95% CI −22.6–0.3). A 2-fold higher baseline ACR was associated with a significant decline in gait speed of 1.39 cm/s (equivalent to a decline of ∼420 cm over 5 min) after adjustment for age, sex and race (Table 2). A 10 mL/min/1.73 m2 lower baseline eGFRcys was associated with a 1.15 cm/s (equivalent to a decline of ∼340 cm over 5 min) slower gait speed. Further adjustments (Model 2), including glucose status and CRP level, attenuated these associations, but they remained statistically significant. Adjustment for CML and PIIINP had no effect on the models. CML (per ng/mL) was associated with a −0.004 cm/s (95% CI −0.008–0.0; P = 0.042) change in gait speed (Model 3).
The regression coefficient for a 10 mL/min/1.73 m2/year lower eGFRcys was −0.74 cm/s. Hence a doubling of baseline ACR (−1.39) was equivalent to ∼1.9 extra years of age; a 10-unit lower baseline eGFRcys (−1.15/−0.75) was equivalent to ∼1.5 extra years of age (1.15/0.75).
When analyses were repeated using ALB, similar associations were observed (Supplementary data, Table S2).
Sensitivity analyses
The Model 2–adjusted odds ratios for clinically significant declines were 1.09 (95% CI 1.02–1.7; P = 0.02) for a doubling in ACR and 1.06 (95% CI 1.01–1.13; P = 0.04) for a 10 mL/min/1.73 m2/year decrease in eGFRcys.
When analyses were categorized by age (above or below the cohort median) (Supplementary data, Table S3), the associations with both outcomes tended to be of similar strength for ACR. Although they tended to be weaker for eGFRcys in participants above the median age, tests of interaction with age were not significant (P > 0.2).
Finally, among participants using an ACEi/ARB, gait speed was not significantly decreased with a doubling of ACR [Model 2: log2 ACR 0.2 cm/s (95% CI −0.93–1.33; P = 0.73)]. In contrast, participants not taking these medications demonstrated a significant decline in gait speed [Model 2: −1.52 cm/s (95% CI −2.21 to −0.94; P < 0.001)] (Pinteraction = 0.01). We observed no significant interactions between grip strength decline or the impact of a 10 mL/min/1.73 m2 lower eGFR in participants treated or not treated with an ACEi/ARB.
DISCUSSION
In this study of older adults with mostly mild renal dysfunction, higher baseline ACR and lower baseline eGFRcys were each prospectively associated with statistically significant lower grip strength and walking speed 2 years later. These relationships were independent of each other and were not appreciably altered by concomitant derangements in glycemia, oxidative stress, fibrosis or inflammation. The changes are apt to have meaningful consequences, particularly because we examined the effects of modest changes in both excess urine protein and eGFR. The associations of a higher baseline ACR with grip and gait change were approximately equivalent to an additional 1.4 and 1.9 years of age; the corresponding associations of 10 mL/min/1.73 m2/year lower baseline eGFRcys approximated 1.2 and 1.5 additional years of age. When we formally examined the likelihood of previously defined clinical thresholds in grip strength or gait speed, both measures of kidney function were associated with risk, highlighting their potential role in loss of function among elders. Finally, our exploratory results suggest (but do not prove) that the use of an ACEi and ARB may attenuate the impact of ACR on gait speed decline.
It may be argued that a 2-year follow-up period is too short for meaningful conclusions regarding the prospective associations of early kidney dysfunction with physical performance. With a median age of ∼77 years in the cohort, 2 years of follow-up constitutes a significant percentage of remaining life expectancy. Longer-term findings are likely to be seriously influenced by the competing risks of attrition and death.
Three prior cross-sectional studies are consistent with our prospective findings. In the I-Lan Longitudinal Aging Study from Taiwan [33] (mean age 63 ± 9 years), the prevalence of prefrailty/frailty and the components of weakness and slowness of gait increased across quartiles of ACR <30 mg/g creatinine. There were no differences across quartiles of eGFR. In the Systolic Blood Pressure Intervention Trial hypertension cohort [31] (mean age 79.8 years), mean gait speed was significantly higher in participants with an ACR (milligram/gram creatinine) <30 versus those ≥30–299 and ≥300. Finally, a prior CHS study [34] reported that a high burden of multiple microvascular disorders (brain white matter, retinal vascular disease and albuminuria) was more strongly associated with functional decline than eGFR levels.
How ALB is mechanistically related to lower muscle function is uncertain. Our current findings were present at a stage of renal disease that is usually not related to metabolic complications (e.g. acidosis, hyperphosphatemia) that can affect muscle function [35]. Likewise, while ALB is related to inflammation factors and oxidative stress [36], we adjusted our findings for these processes and our results were largely unaffected even though we have previously reported that higher CML levels in CHS—our marker of oxidative stress—are associated with incident disability and prevalent frailty [19]. It is therefore likely that alternate mechanisms are present. Several explanations are possible. First, we have shown that ALB is related to cognitive impairment [37]. Prospective studies show that slow gait speed and grip strength are associated with cognitive decline [38]. Second, ALB is related to impairments in other microvascular beds (e.g. skin and heart [39, 40]) and to generalized endothelial dysfunction [41], suggesting that ALB is a marker of a systemic disorder that could affect the microcirculation of muscle. Third, there is a bidirectional association of ALB with sarcopenia and muscle weakness. Kidney disease leads to decreased muscle mass [42] but sarcopenia can also predate the development of ALB [43].
Strengths of this study include a large number of participants in whom kidney function and physical performance were measured systematically and in multiple complementary ways. We included many covariates, some of which are plausible intermediates in the causal pathway between renal and muscle disease yet have rarely been measured in cohort studies. We used cystatin C levels to estimate GFR, which is more accurate than creatinine for assessing renal function in older adults [44]. Our study has limitations. First, urine testing was conducted only once; urine albumin levels vary, which may have biased estimates to the null because of potential misclassification of the ACR predictor. However, a single urine albumin specimen is a reliable means of predicting the geometric mean of multiple measures of ALB [45]. Second, the present cohort was exclusively elderly. Therefore we cannot extrapolate our results to younger people, although declines in physical performance are less common in such populations. Third, the markers of inflammation, oxidative stress and fibrosis that we adjusted for were not comprehensive markers of their associated processes, so these processes may still play a role in the association of renal disease and functional decline. Finally, life expectancy has improved over the past 20 years since the time of urine testing done in this study, such that participants reaching extreme old age now may receive different medical care than did participants in 1999; whether this would mitigate the adverse effect of albuminuria or reduced kidney function is unknown.
In conclusion, early impairments of ACR and eGFR are each independently and prospectively associated with a decline in muscle strength and function in older adults. Adjustment for hyperglycemia, fibrosis, inflammation and oxidative stress do not materially attenuate these results. The current findings suggest an early onset of aging-related functional decline in peripheral muscle in association with processes associated with early renal disease. These findings also suggest that small changes in ACR and eGFR may serve as biomarkers for the frailty syndrome and for the multiple processes that lead to biological aging. Further studies are needed to explore this hypothesis.
FUNDING
This research was supported by contracts HHSN268201200036C, HHSN268200800007C, N01HC55222, N01HC85079, N01HC85080, N01HC85081, N01HC85082, N01HC85083, N01HC85086 and grant U01HL080295 from the National Heart, Lung and Blood Institute (NHLBI), with additional contributions from the National Institute of Neurological Disorders and Stroke. Additional support was provided by the National Institute on Aging (R01AG023629). A full list of principal CHS investigators and institutions can be found at CHS-NHLBI.org. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The funding institutions did not play a role in the creation of this publication.
AUTHORS’ CONTRIBUTIONS
J.I.B. provided the research idea. P.B., J.I.B. and K.J.M. provided the study design. J.A.R., J.A.C., J.H.I. and K.J.M. were responsible for data acquisition. P.B., J.I.B., H.A.F., J.A.R., J.A.C., J.H.I. and K.J.M. were responsible for data analysis and interpretation. P.B. provided statistical analysis. Each author contributed important intellectual content during manuscript writing and accepts accountability for the overall work.
CONFLICT OF INTEREST STATEMENT
None declared.
Supplementary Material
Appendix
REFERENCES
- 1. Clark DA, Khan U, Kiberd BA. et al. Frailty in end-stage renal disease: comparing patient, caregiver, and clinician perspectives. BMC Nephrol 2017; 18: 148. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Lee S-Y, Yang DH, Hwang E. et al. The prevalence, association, and clinical outcomes of frailty in maintenance dialysis patients. J Ren Nutr 2017; 27: 106–112 [DOI] [PubMed] [Google Scholar]
- 3. Fried LP, Tangen CM, Walston J. et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56: M146–M156 [DOI] [PubMed] [Google Scholar]
- 4. Mitnitski AB, Rutenberg AD, Farrell S. et al. Aging, frailty and complex networks. Biogerontology 2017; 18: 433–446 [DOI] [PubMed] [Google Scholar]
- 5. Hirsch CH1, Buzková P, Robbins JA. et al. Predicting late-life disability and death by the rate of decline in physical performance measures. Age Ageing 2012; 41: 155–161 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Diehr PH, Thielke SM, Newman AB. et al. Decline in health for older adults: five-year change in 13 key measures of standardized health. J Gerontol A Biol Sci Med Sci 2013; 68: 1059–1067 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Reese PP, Cappola AR, Shults J. et al. Physical performance and frailty in chronic kidney disease. Am J Nephrol 2013; 38: 307–315 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. König M, Gollasch M, Spira D. et al. Mild-to-moderate chronic kidney disease and geriatric outcomes: analysis of cross-sectional data from the Berlin Aging Study II. Gerontology 2018; 64: 118–126 [DOI] [PubMed] [Google Scholar]
- 9. Roshanravan B, Patel KV, Robinson-Cohen C. et al. Creatinine clearance, walking speed, and muscle atrophy: a cohort study. Am J Kidney Dis 2015; 65: 737–747 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Rowe JW, Andres R, Tobin JD. et al. The effect of age on creatinine clearance in men: a cross-sectional and longitudinal study. J Gerontol 1976; 31: 155–163 [DOI] [PubMed] [Google Scholar]
- 11. Garg AX, Kiberd BA, Clark WF. et al. Albuminuria and renal insufficiency prevalence guides population screening: results from the NHANES III. Kidney Int 2002; 61: 2165–2175 [DOI] [PubMed] [Google Scholar]
- 12. Barzilay JI, Bůžková P, Chen Z. et al. Albuminuria is associated with hip fracture risk in older adults. The Cardiovascular Health Study. Osteoporos Int 2013; 24: 2993–3000 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Fried LF, Biggs ML, Shlipak MG. et al. Association of kidney function with incident hip fracture in older adults. J Am Soc Nephrol 2007; 18: 282–286 [DOI] [PubMed] [Google Scholar]
- 14. Bianchi L, Volpato S.. Muscle dysfunction in type 2 diabetes: a major threat to patient’s mobility and independence. Acta Diabetol 2016; 53: 879–889 [DOI] [PubMed] [Google Scholar]
- 15. Kooman JP, Dekker MJ, Usvyat LA. et al. Inflammation and premature aging in advanced chronic kidney disease. Am J Physiol Renal Physiol 2017; 313: F938–F950 [DOI] [PubMed] [Google Scholar]
- 16. Leitner LM, Wilson RJ, Yan Z. et al. Reactive oxygen species/nitric oxide mediated inter-organ communication in skeletal muscle wasting diseases. Antioxid Redox Signal 2017; 26: 700–717 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Barzilay JI, Bůžková P, Zieman SJ. et al. Circulating levels of carboxy-methyl-lysine (CML) are associated with hip fracture risk: the Cardiovascular Health Study. J Bone Miner Res 2014; 29: 1061–1066 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Kizer JR, Benkeser D, Arnold AM. et al. Advanced glycation/glycoxidation endproduct carboxymethyl-lysine and incidence of coronary heart disease and stroke in older adults. Atherosclerosis 2014; 235: 116–121 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Whitson HE, Arnold AM, Yee LM. et al. Serum carboxymethyl-lysine, disability, and frailty in older persons: the Cardiovascular Health Study. J Gerontol A Biol Sci Med Sci 2014; 69; 710–716 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Prunotto M, Ghiggeri G, Bruschi M. et al. Renal fibrosis and proteomics: current knowledge and still key open questions for proteomic investigation. J Proteomics 2011; 74: 1855–1870 [DOI] [PubMed] [Google Scholar]
- 21. Fried LP, Borhani NO, Enright P. et al. The Cardiovascular Health Study: design and rationale. Ann Epidemiol 1991; 1: 263–276 [DOI] [PubMed] [Google Scholar]
- 22. Busby DE, Bakris GL.. Comparison of commonly used assays for the detection of microalbuminuria. J Clin Hypertens 2004; 6: 8–12 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Shlipak MG, Sarnak MJ, Katz R. et al. Cystatin C and the risk of death and cardiovascular events among elderly persons. N Engl J Med 2005; 352: 2049–2060 [DOI] [PubMed] [Google Scholar]
- 24. Hirsch CH, Fried LP, Harris T. et al. Correlates of performance-based measures of muscle function in the elderly: the Cardiovascular Health Study. J Gerontol A Biol Sci Med Sci 1997; 52: M192–M200 [DOI] [PubMed] [Google Scholar]
- 25. Psaty BM, Kuller LH, Bild D. et al. Methods of assessing prevalent cardiovascular disease in the Cardiovascular Health Study. Ann Epidemiol 1995; 5: 270–277 [DOI] [PubMed] [Google Scholar]
- 26. Cushman M, Cornell ES, Howard PR. et al. Laboratory methods and quality assurance in the Cardiovascular Health Study. Clin Chem 1995; 41: 264–270 [PubMed] [Google Scholar]
- 27. Macy EM, Hayes TE, Tracy RP.. Variability in the measurement of C-reactive protein in healthy subjects: implications for reference intervals and epidemiological applications. Clin Chem 1997; 43: 52–58 [PubMed] [Google Scholar]
- 28. Barzilay JI, Bůžková P, Kizer JR. et al. Fibrosis markers, hip fracture risk, and bone density in older adults. Osteoporos Int 2016; 27: 815–820 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Vickers AJ, Altman DG.. Statistics notes: analysing controlled trials with baseline and follow up measurements. BMJ 2001; 323: 1123–1124 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Granic A, Davies K, Jagger C. et al. Grip strength decline and its determinants in the very old: longitudinal findings from the Newcastle 85+ study. PLoS One 2016; 11: e0163183. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Wolfgram DF, Garcia K, Evans G. et al. Association of albuminuria and estimated glomerular filtration rate with functional performance measures in older adults with chronic kidney disease. Am J Nephrol 2017; 45: 172–179 [DOI] [PubMed] [Google Scholar]
- 32.R Development Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing, 2009 [Google Scholar]
- 33. Chang CC, Hsu CY, Chang TY. et al. Association between low-grade albuminuria and frailty among community-dwelling middle-aged and older people: a cross-sectional analysis from I-Lan Longitudinal Aging Study. Sci Rep 2016; 6: 39434. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Kim DH, Grodstein F, Newman AB. et al. Microvascular and macrovascular abnormalities and cognitive and physical function in older adults: Cardiovascular Health Study. J Am Geriatr Soc 2015; 63: 1886–1893 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Inker LA, Coresh J, Levey AS. et al. Estimated GFR, albuminuria, and complications of chronic kidney disease. J Am Soc Nephrol 2011; 22: 2322–2331 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Miranda-Díaz AG, Pazarín-Villaseñor L, Yanowsky-Escatell FG. et al. Oxidative stress in diabetic nephropathy with early chronic kidney disease. J Diabetes Res 2016; 2016: 7047238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Barzilay JI, Fitzpatrick AL, Luchsinger J. et al. Albuminuria and dementia in the elderly: a community study. Am J Kidney Dis 2008; 52: 216–226 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Hooghiemstra AM, Ramakers IHGB, Sistermans N. et al. Gait speed and grip strength reflect cognitive impairment and are modestly related to incident cognitive decline in memory clinic patients with subjective cognitive decline and mild cognitive impairment: findings from the 4C study. J Gerontol A Biol Sci Med Sci 2017; 72: 846–854 [DOI] [PubMed] [Google Scholar]
- 39. Martens RJH, Henry RMA, Houben AJHM. et al. Capillary rarefaction associates with albuminuria: the Maastricht Study. J Am Soc Nephrol 2016; 27: 3748–3757 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Imamura S, Hirata K, Orii M, et al. Relation of albuminuria to coronary microvascular function in patients with chronic kidney disease. Am J Cardiol 2014; 113: 779–785 [DOI] [PubMed] [Google Scholar]
- 41. Schmiedel O, Schroeter ML, Harvey JN.. Microalbuminuria in type 2 diabetes indicates impaired microvascular vasomotion and perfusion. Am J Physiol Heart Circ Physiol 2007; 293: H3424–H3431 [DOI] [PubMed] [Google Scholar]
- 42. Bouchi R, Fukuda T, Takeuchi T. et al. Sarcopenia is associated with incident albuminuria in patients with type 2 diabetes: a retrospective observational study. J Diabetes Investig 2017; 8: 783–787 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Chung HS, Hwang SY, Choi JH. et al. Effects of low muscle mass on albuminuria and chronic kidney disease in patients with type 2 diabetes: the Korean Sarcopenic Obesity Study (KSOS). J Gerontol A Biol Sci Med Sci 2018; 73: 386–392 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Peralta CA, Katz R, Sarnak MJ. et al. Cystatin C identifies chronic kidney disease patients at higher risk for complications. J Am Soc Nephrol 2011; 22: 147–155 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Pugliese G, Solini A, Fondelli C. et al. Reproducibility of albuminuria in type 2 diabetic subjects. Findings from the Renal Insufficiency and Cardiovascular Events (RIACE) Study. Nephrol Dial Transplant 2011; 26: 3950–3954 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.