Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/ou_press/kidney-function-gait-pattern-and-fall-in-the-general-population-a-lOz0RQDfwl
- Publisher
- Oxford University Press
- Copyright
- © The Author(s) 2018. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.
- ISSN
- 0931-0509
- eISSN
- 1460-2385
- DOI
- 10.1093/ndt/gfy043
- Publisher site
- See Article on Publisher Site
Abstract
ABSTRACT Background Gait disturbance is proposed as a mechanism for higher risk of fall in kidney disease patients. We investigated the association of kidney function with gait pattern in the general population and tested whether the association between impaired kidney function and fall is more pronounced in subjects with lower gait function. Methods We included 1430 participants (mean age: 60 years) from the Rotterdam Study. Kidney function was assessed using estimated glomerular filtration rate (eGFR) and albumin-to-creatinine ratio (ACR). We assessed global gait, gait velocity and seven independent gait domains: Rhythm, Phases, Variability, Pace, Tandem, Turning and Base of Support. Regression models adjusted for cardiometabolic and neurological factors were used. We evaluated whether participants with impaired kidney function and impaired gait fell more in the previous year. Results The study population had a median (interquartile range) ACR of 3.6 (2.5–6.2) mg/g and mean ± SD eGFR of 87.6 ± 15 mL/min/1.73 m2. Higher ACR and lower eGFR were associated with lower global gait score [per doubling of ACR: −0.10, 95% confidence interval (CI): −0.14 to −0.06, and per SD eGFR:−0.09, 95% CI: −0.14 to −0.03] and slower gait speed (ACR: −1.44 cm/s, CI: −2.12 to −0.76; eGFR: −1.55 cm/s, CI: −2.43 to −0.67). Worse kidney function was associated with lower scores in Variability domain. The association between impaired kidney function and history of fall was present only in participants with lower gait scores [odds ratio (95% CI): ACR: 1.34 (1.09–1.65); eGFR: 1.58 (1.07–2.33)]. Conclusions We observed a graded association between lower kidney function and impaired gait suggesting that individuals with decreased kidney function, even at an early stage, need to be evaluated for gait abnormalities and might benefit from fall prevention programmes. albumin-to-creatinine ratio, fall, gait, gait speed, glomerular filtration rate INTRODUCTION Decline in physical functioning is a common finding in patients with chronic kidney disease (CKD) [1]. Deterioration in activity of daily living begins already in early stages of CKD and progressively worsens, leading to a lower quality of life and ultimately shorter survival in these patients [2, 3]. One of the key contributors in performing activities of daily living is a proper walking pattern or gait [4, 5]. Previous studies showed that patients with CKD more frequently experience gait abnormalities, which could lead to higher risk of fall in these patients [6]. Nevertheless, so far it has not been investigated whether there is a graded association between decline in kidney function and gait disturbances and whether the link between impaired kidney function and fall is more pronounced in subjects with lower gait function. In addition, gait is a complex process that includes various domains representing different abilities such as balance, physical strength and cognition. Evaluation of each domain can potentially provide unique information on the pathophysiology and level of the damage. No study has yet evaluated whether kidney function is differentially related to various domains of gait. In this population-based study, we studied the independent associations of kidney function with gait pattern and various gait domains. In addition, we investigated whether the association between kidney function and prevalence of fall is stronger in subjects with lower gait scores. MATERIALS AND METHODS Study population The study was performed within the framework of the Rotterdam Study. The design of the Rotterdam Study has been described previously by Hofman et al. [7]. For this study, we used the third cohort of the Rotterdam Study including participants 45 years and older living in Ommoord, a district of Rotterdam, The Netherlands. Between March 2009 and March 2011, 1643 individuals from the third cohort were invited to undergo gait assessment. In total, gait measurements were performed in 1509 participants (exclusions were due to physical inability, refusal and technical problems). Given the tight link between gait and neurological disorders [8], we further excluded 24 participants from the analyses due to the history of stroke (symptomatic cerebral ischaemia) (n = 21), dementia (n = 1) and symptoms of Parkinsonism (n = 2). This resulted in 1485 participants, of whom 1430 had serum cystatin C data, and 1400 had urine albumin and creatinine measurements. The Rotterdam Study has been approved by the Medical Ethics Committee according to the Population Study Act Rotterdam Study, executed by the Ministry of Health, Welfare and Sports of the Netherlands. A written informed consent was obtained from all participants [7]. Kidney function Serum creatinine is influenced by muscle mass, and muscle mass is an important component of gait. Therefore, we estimated glomerular filtration rate (eGFR) based on cystatin C measurement, which is believed to be independent of muscle mass [9]. Serum cystatin C was measured with a particle-enhanced immunonephelometric assay. eGFR was calculated for cystatin C based on the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) formula [10]. Participants collected the first morning urine before arriving at the research centre. Urine albumin and creatinine were determined by a turbidimetric method and measured by a Hitachi MODULAR P analyser (Roche/Hitachi Diagnostics, Mannheim, Germany). Albumin-to-creatinine ratio (ACR) (mg/g) was estimated by dividing urine albumin by creatinine. Since ACR was not normally distributed, we used base 2 log-transformed values to obtain values per doubling of ACR. We added 1 to the non-transformed values to account for those who did not have albuminuria [11]. Apart from continuous measures of kidney function, we made three kidney function categories. We made the categories on the basis of two criteria: eGFR >60 mL/min/1.73 m2 and ACR <30 mg/g. The first category includes participants who met both criteria, while the second category included participants who met only one criterion. Participants who met none of the criteria were classified as the third category. Gait and fall assessment Details of the gait assessment have been published elsewhere [5]. In brief, a 5.79-m long electronic walkway with pressure sensors (GAITRite Platinum; CIR systems, Sparta, NJ, USA: 4.88 m active area; 120 Hz sampling rate) was used to assess gait. Standardized gait protocol consists of three walking conditions: normal walk, turning and tandem walk. In ‘normal walk’, participants were asked to walk eight times at their usual pace across the walkway. We considered the first walk as a practice walk and did not use it for gait parameter calculations. In ‘turning’, participants walked across the walkway, turned halfway and returned to their starting position. In ‘tandem walk’, participants were asked to walk heel-to-toe over a visible line on the walkway (Figure 1). Using principal components analysis, we summarized mean gait parameters of both legs into seven gait domains: (i) Rhythm, reflecting cadence and single support time (time when the body mass is carried by a single limb); (ii) Variability, reflecting variability in step length and time; (iii) Phases, reflecting double support time (time when both feet are in ground contact) and single support time as a percentage of the total stride time; (iv) Pace, reflecting step length and velocity; (v) Tandem, reflecting errors in tandem walking; (vi) Turning, reflecting the number of steps and time needed to turn; and (vii) Base of support, reflecting stride width and stride width variability (Figure 1 and Supplementary data, Table S1) [5]. If necessary, gait domains were inverted so that lower values of the gait represent a worse gait. Subsequently, global gait was calculated as the z-standardized average of the seven gait domains. Since gait speed is the most commonly assessed gait parameter and has been shown to be a strong predictor of survival [12], we additionally used gait speed from ‘normal walking’ (cm/s) as an additional measure of global gait. Fall was assessed using an interview questionnaire asking whether participants who fell in the last 12 months had serious consequences such as breaking a bone [13]. FIGURE 1 View largeDownload slide Three walking conditions and different gait domains. Normal walk: the gait domains (in the parentheses) are defined through the highly correlating gait parameters. Turn: turning time was measured as the time between the last contact of Foot 1 to first contact of Foot 7. Number of feet—2 was calculated as the turn step count. Tandem walk: tandem walk was defined as the error in tandem walk; in this example, the side step would be considered as the error. FIGURE 1 View largeDownload slide Three walking conditions and different gait domains. Normal walk: the gait domains (in the parentheses) are defined through the highly correlating gait parameters. Turn: turning time was measured as the time between the last contact of Foot 1 to first contact of Foot 7. Number of feet—2 was calculated as the turn step count. Tandem walk: tandem walk was defined as the error in tandem walk; in this example, the side step would be considered as the error. Covariates Cardiovascular risk factors Blood pressure was measured twice on the right arm with a random zero sphygmomanometer, after the participant had been seated for at least 5 min. The mean of both blood pressure values was used in the analyses. Information on antihypertensive medication use was based on home interview. Antihypertensive medications include diuretics, β blockers, angiotensin-converting enzyme inhibitors and calcium channel blockers. Information related to smoking (past/current/never) was based on interviews using questionnaires. Serum total and high-density lipoprotein cholesterol levels were measured using an automated enzymatic method. History of coronary heart disease was considered as a history of myocardial infarction or coronary revascularization procedures. Diabetes mellitus was ascertained using general practitioners’ records (including laboratory glucose measurements), hospital discharge letters and serum glucose measurements from the Rotterdam Study visits, which take place roughly every 4 years. Inflammation and metabolic factors C-reactive protein was measured in non-fasting serum samples kept frozen at 20°C by use of Rate Near Infrared Particle Immunoassay (Immage Immunochemistry System; Beckman Coulter, Brea, CA, USA). Haemoglobin values were measured using the COULTER AC•T diff2 haematology analyser. Serum vitamin D concentrations measurements were performed with an electrochemiluminescence immunoassay (COBAS, Roche Diagnostics GmbH, Germany). Serum levels of calcium and phosphate were calculated using Roche/Hitachi cobas c analyser. Locomotor factors Instrumental activity of daily living (IADL) was assessed based on the IADL scale by Lawton and Brody and scaled between 0 and 24 [4]. Radiographic assessment of hips and knees was performed on participants. Hip and knee osteoarthritis were scored using the Kellgren and Lawrence (K&L) grading system. Radiographic osteoarthritis was defined as a K&L score ≥2 [14]. Neurological factors Brain magnetic resonance imaging (MRI) scanning was performed on a 1.5 Tesla MRI scanner (GE Signa Excite). Scan protocol and sequence details are described extensively elsewhere [15]. In brief, an automated segmentation approach, based on the intensities of the T1-weighted, proton density-weighted and the fluid-attenuated inversion recovery (FLAIR) scans and a conventional k-nearest-neighbour classifier, which was extended with post-processing white matter lesion segmentation, was used to segment scans into grey matter, white matter, white matter lesion, cerebrospinal fluid and background tissue. All segmentation results were visually inspected and, if needed, manually corrected. Supratentorial intracranial volume was estimated by summing grey and white matter, and cerebrospinal fluid (CSF) volumes. Cortical infarcts were rated on structural sequences, and in case of involvement of cortical grey matter, they were classified as cortical infarcts. Lacunes were defined as focal hyperintensities (size ≥3 and <15 mm) with the signal intensity of cerebrospinal fluid (CSF) on all sequences, and when located supratentorially with a hyperintense rim on FLAIR sequence. The presence of cerebral microbleeds was rated on a three-dimensional T2*-weighted gradient-recalled echo MRI scan by one of five trained research physicians, blinded to the clinical data. Apolipoprotein E (APOE) ε4 allele carriership was assessed on coded genomic Deoxyribonucleic acid (DNA) samples. We performed multiple imputations for missing data in the covariates using a Markov chain Monte Carlo method, and used the imputed data for all the analyses. Statistical analysis We performed analysis of covariance where adjusted mean values of standardized z-scores of global gait were compared across different categories of kidney function based on both eGFR and ACR values. Mean values were adjusted for age, age squared, sex, height and weight. We used linear regression models to investigate the associations of kidney function markers with the outcomes including global gait, gait speed and different gait domains. Betas and 95% confidence interval (CI) were estimated per SD increase for eGFR and per doubling of ACR. We adjusted all analyses for age, age squared (given the quadratic association between age and gait [16]), height, weight and sex. In additional models, we adjusted the analyses for cardiovascular risk factors (diabetes mellitus, systolic and diastolic blood pressure, smoking and antihypertensive medication), inflammation and metabolic factors (C-reactive protein, vitamin D, calcium, phosphate and haemoglobin levels), locomotor factors (presence of either hip or knee osteoarthritis and IADL) and neurological factors [intracranial volume, total brain volume, microbleeds, MRI-defined infarcts, white matter lesions, APOE4 carriers and mini-mental state examination (MMSE)]. To improve interpretability, in an additional analysis, we log-transformed global gait scores (a constant was added to all values to avoid missing for negative values) and reported the estimates as percentage increase in global gait scores. To explore whether any of the above-mentioned factors could modify the association between kidney function and global gait, we included an interaction term separately in each model. The interaction term was a product of kidney function markers and the above-mentioned factors. Nested models were compared using F-tests. In an additional analysis, we further adjusted the associations for use of nervous system medications [N02–N07 Anatomical Therapeutic Chemical (ATC) Classification codes]. For Tandem domain, analyses were adjusted for the step length and step count in the tandem walk. To study whether the association of kidney function measures with global gait and gait speed is influenced by individuals with comorbidities such as coronary heart disease, diabetes mellitus and kidney dysfunction, we repeated the analyses excluding participants with history of coronary heart disease, diabetes mellitus and eGFR <60 mL/min/1.73 m2 (n = 222). To explore whether the link between kidney function and falling is through worse gait, we used logistic regression models including kidney function as an independent variable and history of fall (in the past 12 months) as an outcome in participants with high and low global gait scores. We divided the population to low and high global gait based on the median of the global gait scores. Interaction was assessed by adding an interaction term in the regression model. The interaction term was the product of the kidney function measures and global gait scores. In the stratified analysis, we investigated the association of ACR and eGFR with history of fall in two groups of participants with low and high scores of global gait. To explore whether this association is independent of gait speed, we repeated the analyses adjusting for gait speed. All analyses were carried out using SPSS 20.0.2 for windows or R version 3.1.2. RESULTS Characteristics of the participants are shown in Table 1 and Supplementary data, Table S2. The study population had a mean age of 59.7 ± 5 years and 57% were women. There were 45 participants with eGFR <60 mL/min/1.73 m2 and 73 individuals with ACR >30 mg/g. Figure 2 shows mean values of global gait and gait speed in categories of kidney function. We observed a trend (P-value <0.05) when plotting the adjusted mean values of global gait and gait speed across different categories of kidney function based on both ACR and eGFR, indicating that participants with worse kidney function have worse global gait and slower walking speed (Figure 2). Table 1 Population characteristics Characteristics n= 1430 Age, years 59.7 (5.1) Women 816 (57.1) Cardiovascular risk factors Systolic blood pressure, mmHg 131.0 (18.4) Diastolic blood pressure, mmHg 82.2 (10.7) Antihypertensive medication 288 (20.1) Smoking Current 376 (26.3) Former 608 (42.5) Total cholesterol, mmol/L 5.6 (1.0) HDL cholesterol, mmol/L 1.4 (0.4) Diabetes mellitus 121 (8.5) Coronary heart disease 39 (2.6) Weight, kg 80.2 (15.3) Height, cm 170.5 (9.4) Inflammation and metabolic factors Haemoglobin, mmol/L 8.8 (0.7) C-reactive protein, mg/L 1.2 (0.6–2.5) Serum vitamin D, nmol/L 61.3 (27.3) Serum calcium, mmol/L 2.5 (0.1) Serum phosphate, mmol/L 1.1 (0.2) Motor factors IADL, points 1.4 (2.2) Osteoarthritis (knee/hip) 143 (10.0) Neurological factors APOE4 carriers 401 (30.2) White matter lesion volume, mL 1.9 (1.3–3.2) MMSE, score 29 (27–29) Microbleeds 149 (10.4) Lacunar infarcts 45 (3.1) Cortical infarcts 17 (1.2) Intracranial volume, mL 1127.7 (121.5) Total brain volume, mL 958.4 (102.0) Kidney function measures ACR, mg/g 3.6 (2.3–6.2) eGFR, mL/min/1.73 m2 87.6 (14.8) Gait parameters Global gait 0.11 (0.3) Gait speed, cm/s 126.1 (16.2) Rhythm −0.05 (0.9) Phases 0.16 (0.9) Pace 0.15 (0.9) Variability 0.23 (0.8) Turning 0.05 (0.9) Tandem 0.19 (0.7) Base of support 0.07 (0.9) Characteristics n= 1430 Age, years 59.7 (5.1) Women 816 (57.1) Cardiovascular risk factors Systolic blood pressure, mmHg 131.0 (18.4) Diastolic blood pressure, mmHg 82.2 (10.7) Antihypertensive medication 288 (20.1) Smoking Current 376 (26.3) Former 608 (42.5) Total cholesterol, mmol/L 5.6 (1.0) HDL cholesterol, mmol/L 1.4 (0.4) Diabetes mellitus 121 (8.5) Coronary heart disease 39 (2.6) Weight, kg 80.2 (15.3) Height, cm 170.5 (9.4) Inflammation and metabolic factors Haemoglobin, mmol/L 8.8 (0.7) C-reactive protein, mg/L 1.2 (0.6–2.5) Serum vitamin D, nmol/L 61.3 (27.3) Serum calcium, mmol/L 2.5 (0.1) Serum phosphate, mmol/L 1.1 (0.2) Motor factors IADL, points 1.4 (2.2) Osteoarthritis (knee/hip) 143 (10.0) Neurological factors APOE4 carriers 401 (30.2) White matter lesion volume, mL 1.9 (1.3–3.2) MMSE, score 29 (27–29) Microbleeds 149 (10.4) Lacunar infarcts 45 (3.1) Cortical infarcts 17 (1.2) Intracranial volume, mL 1127.7 (121.5) Total brain volume, mL 958.4 (102.0) Kidney function measures ACR, mg/g 3.6 (2.3–6.2) eGFR, mL/min/1.73 m2 87.6 (14.8) Gait parameters Global gait 0.11 (0.3) Gait speed, cm/s 126.1 (16.2) Rhythm −0.05 (0.9) Phases 0.16 (0.9) Pace 0.15 (0.9) Variability 0.23 (0.8) Turning 0.05 (0.9) Tandem 0.19 (0.7) Base of support 0.07 (0.9) Categorical variables are presented as numbers (percentages), continuous variables as means (SDs) and white matter lesions and ACR are presented as medians (interquartile ranges). The following variables had missing data: blood pressure (n = 4), smoking (n = 1), coronary heart disease (n = 39), antihypertensive medication (n = 13), HDL (n = 3), total cholesterol (n= 2), weight and height (n = 1), phosphate (n = 1), calcium (n = 1), vitamin D (n = 7), haemoglobin (n = 1), IADL (n = 369), MMSE (n = 7), white matter lesion (n = 3), intracranial and total brain volume (n = 3), microbleeds (n = 17), APOE4 (n = 104) and osteoarthritis (n = 200). Conversion factors: total cholesterol and HDL cholesterol: multiply by 38.61 to obtain mg/dL. Haemoglobin: multiply by 1.61 to obtain g/dL. Serum vitamin D: multiply by 0.4 to obtain ng/mL. Calcium: multiply by 4.01 to obtain mg/dL. Serum phosphate: multiply by 3.1 to obtain mg/dL. HDL, high-density lipoprotein cholesterol. Table 1 Population characteristics Characteristics n= 1430 Age, years 59.7 (5.1) Women 816 (57.1) Cardiovascular risk factors Systolic blood pressure, mmHg 131.0 (18.4) Diastolic blood pressure, mmHg 82.2 (10.7) Antihypertensive medication 288 (20.1) Smoking Current 376 (26.3) Former 608 (42.5) Total cholesterol, mmol/L 5.6 (1.0) HDL cholesterol, mmol/L 1.4 (0.4) Diabetes mellitus 121 (8.5) Coronary heart disease 39 (2.6) Weight, kg 80.2 (15.3) Height, cm 170.5 (9.4) Inflammation and metabolic factors Haemoglobin, mmol/L 8.8 (0.7) C-reactive protein, mg/L 1.2 (0.6–2.5) Serum vitamin D, nmol/L 61.3 (27.3) Serum calcium, mmol/L 2.5 (0.1) Serum phosphate, mmol/L 1.1 (0.2) Motor factors IADL, points 1.4 (2.2) Osteoarthritis (knee/hip) 143 (10.0) Neurological factors APOE4 carriers 401 (30.2) White matter lesion volume, mL 1.9 (1.3–3.2) MMSE, score 29 (27–29) Microbleeds 149 (10.4) Lacunar infarcts 45 (3.1) Cortical infarcts 17 (1.2) Intracranial volume, mL 1127.7 (121.5) Total brain volume, mL 958.4 (102.0) Kidney function measures ACR, mg/g 3.6 (2.3–6.2) eGFR, mL/min/1.73 m2 87.6 (14.8) Gait parameters Global gait 0.11 (0.3) Gait speed, cm/s 126.1 (16.2) Rhythm −0.05 (0.9) Phases 0.16 (0.9) Pace 0.15 (0.9) Variability 0.23 (0.8) Turning 0.05 (0.9) Tandem 0.19 (0.7) Base of support 0.07 (0.9) Characteristics n= 1430 Age, years 59.7 (5.1) Women 816 (57.1) Cardiovascular risk factors Systolic blood pressure, mmHg 131.0 (18.4) Diastolic blood pressure, mmHg 82.2 (10.7) Antihypertensive medication 288 (20.1) Smoking Current 376 (26.3) Former 608 (42.5) Total cholesterol, mmol/L 5.6 (1.0) HDL cholesterol, mmol/L 1.4 (0.4) Diabetes mellitus 121 (8.5) Coronary heart disease 39 (2.6) Weight, kg 80.2 (15.3) Height, cm 170.5 (9.4) Inflammation and metabolic factors Haemoglobin, mmol/L 8.8 (0.7) C-reactive protein, mg/L 1.2 (0.6–2.5) Serum vitamin D, nmol/L 61.3 (27.3) Serum calcium, mmol/L 2.5 (0.1) Serum phosphate, mmol/L 1.1 (0.2) Motor factors IADL, points 1.4 (2.2) Osteoarthritis (knee/hip) 143 (10.0) Neurological factors APOE4 carriers 401 (30.2) White matter lesion volume, mL 1.9 (1.3–3.2) MMSE, score 29 (27–29) Microbleeds 149 (10.4) Lacunar infarcts 45 (3.1) Cortical infarcts 17 (1.2) Intracranial volume, mL 1127.7 (121.5) Total brain volume, mL 958.4 (102.0) Kidney function measures ACR, mg/g 3.6 (2.3–6.2) eGFR, mL/min/1.73 m2 87.6 (14.8) Gait parameters Global gait 0.11 (0.3) Gait speed, cm/s 126.1 (16.2) Rhythm −0.05 (0.9) Phases 0.16 (0.9) Pace 0.15 (0.9) Variability 0.23 (0.8) Turning 0.05 (0.9) Tandem 0.19 (0.7) Base of support 0.07 (0.9) Categorical variables are presented as numbers (percentages), continuous variables as means (SDs) and white matter lesions and ACR are presented as medians (interquartile ranges). The following variables had missing data: blood pressure (n = 4), smoking (n = 1), coronary heart disease (n = 39), antihypertensive medication (n = 13), HDL (n = 3), total cholesterol (n= 2), weight and height (n = 1), phosphate (n = 1), calcium (n = 1), vitamin D (n = 7), haemoglobin (n = 1), IADL (n = 369), MMSE (n = 7), white matter lesion (n = 3), intracranial and total brain volume (n = 3), microbleeds (n = 17), APOE4 (n = 104) and osteoarthritis (n = 200). Conversion factors: total cholesterol and HDL cholesterol: multiply by 38.61 to obtain mg/dL. Haemoglobin: multiply by 1.61 to obtain g/dL. Serum vitamin D: multiply by 0.4 to obtain ng/mL. Calcium: multiply by 4.01 to obtain mg/dL. Serum phosphate: multiply by 3.1 to obtain mg/dL. HDL, high-density lipoprotein cholesterol. FIGURE 2 View largeDownload slide Adjusted mean and standard error (SE) of standardized global gait and gait speed values across categories of kidney function. Mean values were adjusted for age, age squared, sex, height and weight. First: eGFR >60 mL/min/1.73 m2 and ACR <30 mg/g. Second: eGFR >60 mL/min/1.73 m2 and ACR >30 mg/g or eGFR <60 mL/min/1.73 m2 and ACR <30 mg/g. Third: eGFR <60 mL/min/1.73 m2 and ACR >30 mg/g. FIGURE 2 View largeDownload slide Adjusted mean and standard error (SE) of standardized global gait and gait speed values across categories of kidney function. Mean values were adjusted for age, age squared, sex, height and weight. First: eGFR >60 mL/min/1.73 m2 and ACR <30 mg/g. Second: eGFR >60 mL/min/1.73 m2 and ACR >30 mg/g or eGFR <60 mL/min/1.73 m2 and ACR <30 mg/g. Third: eGFR <60 mL/min/1.73 m2 and ACR >30 mg/g. Kidney function, global gait and gait speed The association of kidney function markers with global gait and gait speed is presented in Figure 3. In the basic model (adjusted for age, age squared, sex, height and weight), higher ACR was associated with worse global gait [difference in standardized z-scores of gait per doubling of ACR]: −0.10; 95% CI: −0.14 to −0.06. Similarly, each SD for lower eGFR was associated with 0.09 lower values of standardized z-scores of global gait (95% CI: −0.14 to −0.03). Higher ACR was associated with slower gait speed (−1.44 cm/s; 95% CI: −2.12 to −0.76). Each SD lower eGFR was associated with slower gait speed (−1.55 cm/s; 95% CI: −2.43 to −0.67). Separate adjustments for motor, metabolic and neurological factors did not essentially change the associations. Associations attenuated after adjusting for cardiovascular risk factors. When adjusting for all factors in a single model, the effect estimates attenuated (from 2% to 1% change in global gait score per SD eGFR increase or doubling of ACR) but remained statistically significant (Supplementary data, Table S3 and Figure 3). Adjusting the associations for use of central nervous system medications did not change the associations (data not shown). Excluding participants with history of coronary heart disease, diabetes mellitus and eGFR <60 mL/min/1.73 m2 did not change our findings (Supplementary data, Table S4). No significant interactions were observed between kidney function markers and motor, inflammatory, metabolic, neurological and cardiovascular risk factors in relation to global gait or gait speed (data not shown). FIGURE 3 View largeDownload slide Association between kidney function and global gait and gait speed adjusting for various potential factors. Base model: adjusted for age, age squared, sex, height and weight. Cardiovascular risk factors: adjusted for age, age squared, sex, height, weight, diabetes mellitus, systolic and diastolic blood pressure, smoking and antihypertensive medication. Inflammation and metabolic factors: adjusted for age, age squared, sex, height, weight, C-reactive protein, vitamin D, calcium, phosphate and haemoglobin. Motor factor: adjusted for age, age squared, sex, height, weight, presence of either hip or knee osteoarthritis and IADL. Neurological factors: adjusted for age, age squared, sex, height, weight, intracranial volume, total brain volume, microbleeds, lacunar infarct, cortical infarct, white matter lesion, APOE4 carriers and MMSE. FIGURE 3 View largeDownload slide Association between kidney function and global gait and gait speed adjusting for various potential factors. Base model: adjusted for age, age squared, sex, height and weight. Cardiovascular risk factors: adjusted for age, age squared, sex, height, weight, diabetes mellitus, systolic and diastolic blood pressure, smoking and antihypertensive medication. Inflammation and metabolic factors: adjusted for age, age squared, sex, height, weight, C-reactive protein, vitamin D, calcium, phosphate and haemoglobin. Motor factor: adjusted for age, age squared, sex, height, weight, presence of either hip or knee osteoarthritis and IADL. Neurological factors: adjusted for age, age squared, sex, height, weight, intracranial volume, total brain volume, microbleeds, lacunar infarct, cortical infarct, white matter lesion, APOE4 carriers and MMSE. Kidney function and gait domains Figure 4 shows the association of kidney function markers with various gait domains. From seven gait domains, both higher ACR and lower eGFR were associated with lower scores of Variability (P-values <0.05) indicating that people with impaired kidney function have exaggerated gait variability. Furthermore, higher ACR was associated with slower scores in Pace domain and lower scores in Phase domain (all P-values <0.05). FIGURE 4 View largeDownload slide The association of kidney function parameters with different gait domains. Grey bars with stars indicate statistically significant results. Differences (betas), and 95% CI are calculated using the per SD increase in eGFR and subject-specific standardized z-scores for various gait domains. Differences for ACR represent measures of gait components per doubling of the ACR. All analyses were adjusted for age, age squared, sex, height, weight and all the gait domains. For Tandem, analyses were adjusted additionally for step count and step size. FIGURE 4 View largeDownload slide The association of kidney function parameters with different gait domains. Grey bars with stars indicate statistically significant results. Differences (betas), and 95% CI are calculated using the per SD increase in eGFR and subject-specific standardized z-scores for various gait domains. Differences for ACR represent measures of gait components per doubling of the ACR. All analyses were adjusted for age, age squared, sex, height, weight and all the gait domains. For Tandem, analyses were adjusted additionally for step count and step size. Kidney function, gait and risk of fall In the total population, each doubling of ACR was associated with 1.24 higher odds of falling in the past year. Similarly, each SD (14.8 mL/min/1.73 m2) was associated with 1.30 higher odds of falling in the past year. When dividing the population to low and high global gait, we observed that in subjects with low global gait scores, higher ACR was associated with higher prevalence of falls, whereas such an association was not found in subjects with better global gait scores [odds ratio (OR): 1.34; 95% CI: 1.09–1.65 versus OR: 1.15; 95% CI: 0.86–1.53].This difference was statistically significant (P for interaction = 0.035). Similarly, lower eGFR was associated with higher prevalence of fall only in participants with lower global gait scores (OR: 1.58; 95% CI: 1.07–2.33 versus OR: 1.09; 95% CI: 0.74–1.60). This difference was statistically significant (P for interaction = 0.011) (Table 2). Adjusting for cardiovascular risk factors, motor, metabolic and neurological factors as well as gait speed did not change the results (Table 2, model 2 and Supplementary data, Table S5). This finding suggests that the association between kidney function and fall is more prominent in subjects with low global gait scores. Table 2 Association of kidney function markers with prevalence of fall in categories of global gait Prevalence of fall OR (95% CI)a P-value P for interaction Fall cases/n Total population Fall cases/n Low global gait (lowest 50%) Fall cases/n High global gait (highest 50%) ACR 62/1400 31/701 31/699 Model 1 1.24 (1.06, 1.46) 0.009 1.34 (1.09, 1.65) 0.005 1.15 (0.86, 1.53) 0.357 0.035 Model 2 1.23 (1.03, 1.48) 0.026 1.45 (1.14, 1.85) 0.003 1.08 (0.78, 1.49) 0.625 0.046 eGFR 64/1430 31/712 33/717 Model 1 1.30 (1.00, 1.69) 0.054 1.58 (1.07 2.33) 0.021 1.09 (0.74, 1.60) 0.652 0.011 Model 2 1.24 (0.93, 1.65) 0.142 1.81 (1.12, 2.93) 0.015 1.04 (0.69, 1.58) 0.845 0.016 Prevalence of fall OR (95% CI)a P-value P for interaction Fall cases/n Total population Fall cases/n Low global gait (lowest 50%) Fall cases/n High global gait (highest 50%) ACR 62/1400 31/701 31/699 Model 1 1.24 (1.06, 1.46) 0.009 1.34 (1.09, 1.65) 0.005 1.15 (0.86, 1.53) 0.357 0.035 Model 2 1.23 (1.03, 1.48) 0.026 1.45 (1.14, 1.85) 0.003 1.08 (0.78, 1.49) 0.625 0.046 eGFR 64/1430 31/712 33/717 Model 1 1.30 (1.00, 1.69) 0.054 1.58 (1.07 2.33) 0.021 1.09 (0.74, 1.60) 0.652 0.011 Model 2 1.24 (0.93, 1.65) 0.142 1.81 (1.12, 2.93) 0.015 1.04 (0.69, 1.58) 0.845 0.016 Global gait is categorized based on median value (0.14). Model 1: adjusted for age, age squared, sex, height and weight. Model 2: additionally adjusted for diabetes mellitus, systolic and diastolic blood pressure, smoking, antihypertensive medication, C-reactive protein, vitamin D, calcium, phosphate, haemoglobin, presence of either hip or knee osteoarthritis, IADL, intracranial volume, total brain volume, microbleeds, lacunar infarct, cortical infarct, white matter lesion, APOE4 carriers and MMSE. a ORs are presented per each negative SD of eGFR and per doubling of ACR. Table 2 Association of kidney function markers with prevalence of fall in categories of global gait Prevalence of fall OR (95% CI)a P-value P for interaction Fall cases/n Total population Fall cases/n Low global gait (lowest 50%) Fall cases/n High global gait (highest 50%) ACR 62/1400 31/701 31/699 Model 1 1.24 (1.06, 1.46) 0.009 1.34 (1.09, 1.65) 0.005 1.15 (0.86, 1.53) 0.357 0.035 Model 2 1.23 (1.03, 1.48) 0.026 1.45 (1.14, 1.85) 0.003 1.08 (0.78, 1.49) 0.625 0.046 eGFR 64/1430 31/712 33/717 Model 1 1.30 (1.00, 1.69) 0.054 1.58 (1.07 2.33) 0.021 1.09 (0.74, 1.60) 0.652 0.011 Model 2 1.24 (0.93, 1.65) 0.142 1.81 (1.12, 2.93) 0.015 1.04 (0.69, 1.58) 0.845 0.016 Prevalence of fall OR (95% CI)a P-value P for interaction Fall cases/n Total population Fall cases/n Low global gait (lowest 50%) Fall cases/n High global gait (highest 50%) ACR 62/1400 31/701 31/699 Model 1 1.24 (1.06, 1.46) 0.009 1.34 (1.09, 1.65) 0.005 1.15 (0.86, 1.53) 0.357 0.035 Model 2 1.23 (1.03, 1.48) 0.026 1.45 (1.14, 1.85) 0.003 1.08 (0.78, 1.49) 0.625 0.046 eGFR 64/1430 31/712 33/717 Model 1 1.30 (1.00, 1.69) 0.054 1.58 (1.07 2.33) 0.021 1.09 (0.74, 1.60) 0.652 0.011 Model 2 1.24 (0.93, 1.65) 0.142 1.81 (1.12, 2.93) 0.015 1.04 (0.69, 1.58) 0.845 0.016 Global gait is categorized based on median value (0.14). Model 1: adjusted for age, age squared, sex, height and weight. Model 2: additionally adjusted for diabetes mellitus, systolic and diastolic blood pressure, smoking, antihypertensive medication, C-reactive protein, vitamin D, calcium, phosphate, haemoglobin, presence of either hip or knee osteoarthritis, IADL, intracranial volume, total brain volume, microbleeds, lacunar infarct, cortical infarct, white matter lesion, APOE4 carriers and MMSE. a ORs are presented per each negative SD of eGFR and per doubling of ACR. DISCUSSION In this population-based study, we observed that worse kidney function is independently associated with worse performance in global gait function and slower gait speed. From all the gait domains, both kidney function measures were associated with lower scores of Variability domain. Furthermore, in participants with lower scores of global gait, impaired kidney function was associated with more reports of falls in the past year. The link between worse kidney function and decline in physical ability has been shown previously [2, 3, 17, 18]. Investigators from the Health, Aging and Body Composition (Health ABC) Study found that individuals with eGFR <60 mL/min/1.73 m2 are more likely to have slower gait speed, reduction in lower extremity performance and lower strength [2]. Consistently, in the Cardiovascular Health Study, the investigators showed that a higher level of serum creatinine is associated with lower scores in activity of daily living [18]. The Framingham Offspring Study including participants with mean age of 68 years found that patients with CKD (as defined by cystatin C) experience a steeper decline in gait speed [17]. In this study, we showed that impaired kidney function is associated with higher prevalence of falls only in participants with low scores of gait. We also found that impaired kidney function was most prominently related to worse Variability domain, which previously has been shown to be specifically related to higher risk of falls [19]. Given the high occurrence of falls in patients with kidney disease [6], gait abnormalities could be considered as potential predisposing factors for higher risk of fall in these patients. Most gait domains, Variability in particular, are difficult, if not impossible, to assess visually; hence, a thorough computerized gait assessment is potentially a useful tool to identify people most in need of fall prevention programmes. Different factors can play a role in the association between kidney function and gait. Impairment in kidney function is known to be related to osteoarthritis [20], inflammation and metabolic factors [21, 22]. In addition, patients with kidney disease are at higher risk of cerebrovascular and neurodegenerative disorders [11], which has been reported to be tightly linked to gait [8]. Finally, cardiovascular risk factors such as smoking, high blood pressure and diabetes are also related to both impairment in kidney function and gait deterioration. Adjusting for these factors separately and including them in a model as an interacting factor did not fully explain the associations between kidney function and gait [4, 23]. From all the factors, adjusting for cardiovascular factors attenuated the association between kidney function and gait the most, suggesting that the relation could be partly explained by cardiovascular risk factors. Another explanation for the association between impaired kidney function and gait could be the direct impact of kidney function on gait. Impairments in kidney function, even in early stages, is associated with the accumulation of neurotoxins that have been shown to cause axonal loss with secondary or predominant demyelination, β2-microglobulin deposition in joints and connective tissue, and increased levels of inflammatory cytokines, which can all result in muscle mass reduction, muscle strength deterioration and loss of balance [24, 25]. Strengths of our study include the relatively large sample size and availability of extensive information on various kidney function parameters and different neurological and cardio-metabolic risk factors, which enabled us to control for several potential factors that can influence the association between kidney function and gait. Limitations of our study should be acknowledged. First, gait was assessed in a research setting and it is possible that people with severe gait problems are under-represented in a population-based study. Nonetheless, our findings indicate that lower kidney function is associated with more subtle gait abnormalities before full-blown gait abnormalities become evident. Secondly, we have assessed the link between kidney function, gait and fall in a cross-sectional setting, which makes it difficult to address the temporal relationship of the association. It is possible that a fall results in gait problems and subsequent declines in health and kidney function. Finally, cystatin C levels and spot albuminuria may vary within persons over time and therefore single measurement of these markers may not reflect true values. In conclusion, we show that worse kidney function is associated with slower gait speed and worse gait pattern. From all the gait domains, both low eGFR and high ACR were associated with lower scores in Variability domain. Our findings suggest that kidney function, even in early stages, is associated with impairments in gait, which may increase the risk of falling and lead to reduced quality of life and disability. ACKNOWLEDGEMENTS The Rotterdam Study investigators are grateful to the participants and staff from the Rotterdam Study, the participating general practitioners and the pharmacists. FUNDING The Rotterdam Study is supported by the Erasmus Medical Center and Erasmus University Rotterdam; the Netherlands Organization for Scientific Research; the Netherlands Organization for Health Research and Development (ZonMw); the Research Institute for Diseases in the Elderly; the Ministry of Education, Culture and Science; the Ministry of Health, Welfare and Sports; the European Commission; and the Municipality of Rotterdam. J.N.v.d.G. was supported by the Prinses Beatrix Fonds. Funding for gait assessment was obtained from the International Parkinson Fonds. O.H.F. works in ErasmusAGE, a centre for ageing research across the life course funded by Nestlé Nutrition (Nestec Ltd), Metagenics Inc. and AXA. The sponsors had no involvement in any of the following: the study design; the collection, analysis and interpretation of data; the writing of the manuscript; and the decision to submit the manuscript for publication. CONFLICT OF INTEREST STATEMENT None declared. REFERENCES 1 Wilhelm-Leen ER , Hall YN , Tamura MK et al. Frailty and chronic kidney disease: the Third National Health and Nutrition Evaluation Survey . Am J Med 2009 ; 122 : 664 – 671.e2 Google Scholar Crossref Search ADS PubMed 2 Odden MC , Chertow GM , Fried LF et al. Cystatin C and measures of physical function in elderly adults: the Health, Aging, and Body Composition (HABC) Study . Am J Epidemiol 2006 ; 164 : 1180 – 1189 Google Scholar Crossref Search ADS PubMed 3 Bowling CB , Sawyer P , Campbell RC et al. Impact of chronic kidney disease on activities of daily living in community-dwelling older adults . J Gerontol A Biol Sci Med Sci 2011 ; 66 : 689 – 694 Google Scholar Crossref Search ADS PubMed 4 Verlinden VJ , van der Geest JN , Heeringa J et al. Gait shows a sex-specific pattern of associations with daily functioning in a community-dwelling population of older people . Gait Posture 2015 ; 41 : 119 – 124 Google Scholar Crossref Search ADS PubMed 5 Verlinden VJ , van der Geest JN , Hoogendam YY et al. Gait patterns in a community-dwelling population aged 50 years and older . Gait Posture 2013 ; 37 : 500 – 505 Google Scholar Crossref Search ADS PubMed 6 Lopez-Soto PJ , De Giorgi A , Senno E et al. Renal disease and accidental falls: a review of published evidence . BMC Nephrol 2015 ; 16 : 176 Google Scholar Crossref Search ADS PubMed 7 Hofman A , Brusselle GG , Darwish Murad S et al. The Rotterdam Study: 2016 objectives and design update . Eur J Epidemiol 2015 ; 30 : 661 – 708 Google Scholar Crossref Search ADS PubMed 8 Verlinden VJ , van der Geest JN , Hofman A et al. Cognition and gait show a distinct pattern of association in the general population . Alzheimers Dement 2014 ; 10 : 328 – 335 Google Scholar Crossref Search ADS PubMed 9 Baxmann AC , Ahmed MS , Marques NC et al. Influence of muscle mass and physical activity on serum and urinary creatinine and serum cystatin C . Clin J Am Soc Nephrol 2008 ; 3 : 348 – 354 Google Scholar Crossref Search ADS PubMed 10 Inker LA , Schmid CH , Tighiouart H et al. Estimating glomerular filtration rate from serum creatinine and cystatin C . N Engl J Med 2012 ; 367 : 20 – 29 Google Scholar Crossref Search ADS PubMed 11 Sedaghat S , Cremers LG , de Groot M et al. Kidney function and microstructural integrity of brain white matter . Neurology 2015 ; 85 : 154 – 161 Google Scholar Crossref Search ADS PubMed 12 Studenski S , Perera S , Patel K et al. Gait speed and survival in older adults . JAMA 2011 ; 305 : 50 – 58 Google Scholar Crossref Search ADS PubMed 13 Lahousse L , Verlinden VJ , van der Geest JN et al. Gait patterns in COPD: the Rotterdam Study . Eur Respir J 2015 ; 46 : 88 – 95 Google Scholar Crossref Search ADS PubMed 14 Hoeven TA , Kavousi M , Clockaerts S et al. Association of atherosclerosis with presence and progression of osteoarthritis: the Rotterdam Study . Ann Rheum Dis 2013 ; 72 : 646 – 651 Google Scholar Crossref Search ADS PubMed 15 Ikram MA , van der Lugt A , Niessen WJ et al. The Rotterdam Scan Study: design update 2016 and main findings . Eur J Epidemiol 2015 ; 30 : 1299 – 1315 Google Scholar Crossref Search ADS PubMed 16 Terrier P , Reynard F. Effect of age on the variability and stability of gait: a cross-sectional treadmill study in healthy individuals between 20 and 69 years of age . Gait Posture 2015 ; 41 : 170 – 174 Google Scholar Crossref Search ADS PubMed 17 Liu CK , Lyass A , Massaro JM et al. Chronic kidney disease defined by cystatin C predicts mobility disability and changes in gait speed: the Framingham Offspring Study . J Gerontol A Biol Sci Med Sci 2014 ; 69 : 301 – 307 Google Scholar Crossref Search ADS PubMed 18 Shlipak MG , Stehman-Breen C , Fried LF et al. The presence of frailty in elderly persons with chronic renal insufficiency . Am J Kidney Dis 2004 ; 43 : 861 – 867 Google Scholar Crossref Search ADS PubMed 19 Verghese J , Holtzer R , Lipton RB et al. Quantitative gait markers and incident fall risk in older adults . J Gerontol A Biol Sci Med Sci 2009 ; 64 : 896 – 901 Google Scholar Crossref Search ADS PubMed 20 Cohen SD , Patel SS , Khetpal P et al. Pain, sleep disturbance, and quality of life in patients with chronic kidney disease . Clin J Am Soc Nephrol 2007 ; 2 : 919 – 925 Google Scholar Crossref Search ADS PubMed 21 Muir SW , Montero-Odasso M. Effect of vitamin D supplementation on muscle strength, gait and balance in older adults: a systematic review and meta-analysis . J Am Geriatr Soc 2011 ; 59 : 2291 – 2300 Google Scholar Crossref Search ADS PubMed 22 Padilla J , Krasnoff J , Da Silva M et al. Physical functioning in patients with chronic kidney disease . J Nephrol 2008 ; 21 : 550 – 559 Google Scholar PubMed 23 Verlinden VJ , Maksimovic A , Mirza SS et al. The associations of alcohol, coffee and tobacco consumption with gait in a community-dwelling population . Eur J Clin Nutr 2016 ; 70 : 116 – 122 Google Scholar Crossref Search ADS PubMed 24 Yokoi H , Yanagita M. Decrease of muscle volume in chronic kidney disease: the role of mitochondria in skeletal muscle . Kidney Int 2014 ; 85 : 1258 – 1260 Google Scholar Crossref Search ADS PubMed 25 Galassi G , Ferrari S , Cobelli M et al. Neuromuscular complications of kidney diseases . Nephrol Dial Transplant 1998 ; 13 (Suppl 7) : 41 – 47 Google Scholar Crossref Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
Journal
Nephrology Dialysis Transplantation – Oxford University Press
Published: Dec 1, 2018