Abstract
Background/Objectives:
Neuroimaging indicators of reduced brain health in the form of lower gray matter volume, lower fractional anisotropy, and higher white matter hyperintensity volume, have been related to global mobility measures such as gait speed in older adults. The purpose was to identify associations between brain regions and specific mobility functions in order to provide a greater understanding of the contribution of the central nervous system to independent living.
Design:
Cross Sectional Study
Setting:
Research lab
Participants:
Seventy community-ambulating healthy older adults (mean age 76 ± 5 years)
Measurements:
Participants performed the following tests: gait speed, Five Times Sit to Stand (5xSTS), Four Square Step Test (FSST), and Dynamic Gait Index (DGI). Structural MRI of each participant’s brain was collected. Measures of regional gray matter volume (GMV), tract-specific white matter hyperintensity volume (WMHV) and fractional anisotropy (FA) were extracted. Correlational analyses between the mobility measures and neuroimaging measures were conducted using whole brain and regional and tract-specific measures. This was followed by linear regression models relating the mobility measures to regions or tracts identified in the correlation analysis, and adjusting for age, sex and body mass index.
Results:
Significant associations were found between higher GMV in multiple regions, primarily the parietal and temporal lobes, and better performance in gait speed, DGI and FSST. After adjusting for personal factors, greater parahippocampus GMV was independently associated with greater gait speed. Greater inferior parietal lobe, supramarginal gyrus and superior temporal gyrus GMV were associated with gait function. Greater postcentral gyrus, parahippocampus and superior temporal gyrus, GMV were associated with faster FSST performance. The WMHV and FA were not significantly correlated with the mobility measures.
Conclusions:
Grey matter regions associated with higher performance in mobility measures serving gait function and multidirectional stepping were those structures related to vestibular sensation, spatial navigation, and somatosensation.
Keywords: Aging, neurology, balance
INTRODUCTION
According to the most recent data from the US Centers for Disease Control, the fall rate in adults over the age of 65 has increased by 3% per year over the span of 2007-2016 (i.e. there were 11,000 more fall-related deaths in 2016 compared with 2007).1 Impaired mobility has been associated with increased risk for falls and shown to be predictive of increased risk for mortality and loss of independence in community-dwelling older adults.2,3 Mobility impairments are complex with many contributing factors. It is crucial to identify the intrinsic and extrinsic factors that lead to impaired mobility so that future treatments can be tailored to an individual’s unique characteristics.
Investigations of the relationship between central nervous system changes and impaired mobility have been undertaken in the past two decades. Decreased gray matter volumes have been associated with poorer scores on mobility outcomes.4–6 Conversely, increased levels of whole brain gray matter have been shown to have a protective effect against falls risk.7 It has also been shown that white matter lesions and progression of white matter hyperintensities are associated with impaired balance, slower gait speed, and reduction in mobility.8,9 Reduced white matter microstructural integrity, such as fractional anisotropy, has also been associated with poor mobility performance.10–12
In addition to measures of whole brain gray and white matter, it is important to investigate the contribution of specific brain regions to mobility. For example, decreased gray matter volumes in regions of the brain associated with motor control (cerebellum, basal ganglia, prefrontal cortex) appear to be related to poorer scores on balance measures and slower gait speed independent of white matter integrity.13 Furthermore, in order to develop more targeted treatment strategies, investigation of the association between regional neuroimaging measures and mobility should move beyond global measures of mobility such as gait speed, to specific mobility functions such as functional strength, gait stability, and multi-directional stepping. These functions allow people to move safely in the variable and complex environments encountered in community living.
The purpose of this study was to explore the relationships between brain structural neuroimaging measures and impairments in functional mobility in community-dwelling older adults. The hypothesis was that decreased performance on functional mobility measures will be associated with decreased integrity of selected mobility-related regions and tracts. Specifically, we focus on gray matter volume, white matter hyperintensity volume and fractional anisotropy as the main measures of interest.
METHODS
Participants
Seventy (30 male, 40 female) community-ambulating healthy older adults aged 70 and older participated in the study. Ten of the participants were African-American and 60 were Caucasian. All subjects were recruited from the Claude D. Pepper Center Research Registry. The inclusion criteria were: age 70 y and older, ability to stand unassisted for at least 5 minutes and walk unassisted for 3 minutes. The exclusion criteria were: Primary neurological, cardiopulmonary or orthopaedic disorder that would affect balance or limit mobility (e.g. Parkinson’s Disease, mild brain injury, multiple sclerosis, uncontrolled hypertension, uncontrolled diabetes, chronic obstructive pulmonary disease, significant vision loss, peripheral vestibulopathy, severe arthritis), psychoactive medication use, abnormal central neurological exam, corrected binocular visual acuity worse than 20/40 and contrast sensitivity worse than 3%, unilateral or bilateral vestibular hypofunction, clinically significant peripheral neuropathy, cognitive impairment greater than 1.5 standard deviations below age-adjusted mean on Repeated Battery of Neuropsychological Status testing, or clinically significant depression. The number of and indications of medications was recorded by self-report. The Duke Comorbidity Index was utilized to estimate the medical comorbidities, which may impair functional status and contribute to disability.14
Study Design
This study was part of a larger study to investigate step initiation performance in healthy older adults.15,16 After providing informed consent and passing screening tests, eligible participants returned for a second visit for neuroimaging and a third visit to perform the functional mobility tests. The functional mobility tests were selected to capture a more comprehensive assessment of each individual’s current functional status, and included: gait speed over 6 m, Five Times Sit to Stand (5xSTS), the Four-Square Step Test (FSST), the Dynamic Gait Index (DGI). All outcome measures were scored by a licensed physical therapist who was blinded to the results of the MRI scans.
Functional mobility tests
Gait speed was assessed by taking the average of time to complete the 6 meter walk test at a self-selected speed. Gait speed has been shown to represent a composite of disability and morbidity in community-dwelling older adults.17,18
The Five Times Sit to Stand Test was measured due to its strong ability to capture lower extremity strength and functioning. Participants were instructed to stand up from a standard height chair and sit down five times consecutively, as quickly as possible.19 Individuals that take greater than 15 seconds to complete the task should be assessed for recurrent falls risk.20
The DGI is an outcome measure that assesses multiple aspects of dynamic gait and balance. The DGI is marked by a 20-foot straight path on level surface with each component being scored on a 4- point scale based on degree of impairment. The components assessed include steady state ambulation, walking with changing speeds, walking forward with horizontal and vertical head turns, walking while stepping over and around objects, pivoting while walking, and stair climbing. A total of 24 points are available to be scored. Scores less than 19 on the DGI have been correlated with increased risk for falls community- dwelling older adults.21
The FSST is a valid and reliable tool that assesses dynamic standing balance, motor planning and identifies individuals at increased risk for falls.22 Individuals are timed while they complete a multidirectional stepping task over canes on the floor. A cutoff time of 15 seconds or great to complete the task has been indicated as a cutoff score for increased risk of falls in older adults.22
Neuroimaging
MR imaging was performed using a Siemens 3 Tesla TIM TRIO scanner with an 8-channel head coil. The following axial series oriented to the plane connecting the anterior and posterior commissures was acquired:1) T1-weighted, 2) fast spin-echo T2-weighted, 3) fast spin-echo proton density-weighted, 4) fast fluid-attenuated inversion recovery (FLAIR), 5) magnetization-prepared rapid-acquisition gradient-echo (MPRAGE) and 5) diffusion-tensor imaging (DTI). Section thickness was 3 mm with no gap. Automated methods were used for quantification of the whole brain grey matter volume (GMV) and white matter hyperintensity volume (WMHV) using the Automated Labeling Pathway (ALP).23 The GMV was segmented into 47 regions that were averaged across hemispheres. Of these 47 regions, 18 were selected because of their relationship to balance or lower extremity function (refer to Table 3 in Results). An automated WMH extraction method provided a WMHV estimate for each of 11 white matter tracts identified in the Johns Hopkins Atlas (refer to Table 4 in Results).24 GMV and WMHV were represented as a percentage of total brain volume. The whole brain and white matter tract fractional anisotropy (FA) was calculated from the DTI image. A tract-based spatial statistic method (TBSS) was used to map the FA values onto the main white matter tracts.25 The mean FA value was calculated for the ‘normal’ appearing white matter in the tract, excluding the voxels classified as WMH on the FLAIR image. Eleven white matter tracts were selected for analysis of the WMHV and FA.
Statistical analysis
Demographics, clinical characteristics, and performance on the functional mobility tests and quantification of whole brain GMV, WMHV and FA were summarized using descriptive statistics (mean, standard deviation, and range). Deviations of normality were noted for the FSST, DGI, 5xSTS, and WMHV. Bivariate associations between population characteristics and the mobility tests were tested using unadjusted correlations or chi square test for continuous or binary variables, respectively. Associations between the mobility tests and whole brain neuroimaging measures were first tested with unadjusted correlations, using both Pearson and Spearman correlation coefficients. Despite the normality deviations, the strength of the associations was similar, and the parametric tests were used for reporting. Based on the significant correlations that were observed between the functional mobility tests and GMV, we proceeded to investigate relationships between the functional mobility measures and regional GMV values through multivariable models (no significant associations with WMHV and tract-specific FA were found). False discovery rate (FDR) method was used to correct for multiple comparisons and select neuroimaging measures for multivariable regression models.26A series of 4 models were performed, where the first model was unadjusted, model 2 was adjusted for age, model 3 was adjusted for age and sex, and model 4 was adjusted for age, sex, and body mass index (BMI). In sensitivity analyses, models were further adjusted for other population characteristics that were significantly associated with mobility measures, such as number of medications taken. The significance level for all analyses was set at 0.05.
RESULTS
The clinical characteristics, performance on the functional mobility tests and neuroimaging measures are detailed in Table 1. The mean age was 76 (SD 5) years and the body mass index was 26.7 (3.4) kg/m2. The median number of prescription medications taken was two; a majority of the participants were taking anti-hypertensive medication and a majority were taking calcium supplements. The median number of comorbidities was 3, with the most common being cataracts (61%), arthritis (59%), history of fracture (44%), and cancer (33%). Even though mean values of the mobility measures would be considered to be normal, we observed a wide range of values indicating a heterogeneous sample of healthy older adults. Among the functional mobility tests, the inter-correlations ranged between 0.26 and 0.70, indicating that the functional mobility tests generally represent different constructs. Likewise, the intercorrelations among the whole brain neuroimaging measures ranged from 0.01 to 0.46.
Table 1:
Mean (SD) of the clinical characteristics, mobility measures, and whole brain neuroimaging measures, in N=70, (40 females, 10 African-American)
| Measure | Mean (SD) [Range] |
|---|---|
| Age | 76 (5) [70 – 94] |
| Body mass index (kg/m2) | 26.7 (3.4) [20.2 – 37.2] |
| Number of medications | 3 (3) [0 – 13] Median 2 |
| Number of comorbidities | 3 (2) [0 – 7] Median 3 |
| Gait Speed (m/s) | 1.24 (0.20) [ 0.71 – 1.63] |
| Five Times Sit to Stand (s) | 11.2 (4.4) [3.9 – 41.0] |
| Dynamic Gait Index (maximum 24) | 22 (1) [18 – 24] |
| Four Square Step Test (s) | 9.9 (2.1) [6.8 – 18.8] |
| Grey Matter Volume (% total brain volume) | 32 (2) [25 – 37] |
| White Matter Hyperintensities Volume (% total brain volume) | 0.14 (0.17) [0 – 0.93] |
| Fractional Anisotropy | 0.37 (0.01) [0.33 – 0.40] |
The correlations between the mobility measures and whole brain neuroimaging measures are displayed in Table 2. Higher GMV was associated with better mobility performance, reflected by faster gait speed, higher DGI, and faster time to complete the FSST. There were no significant relationships between the volume of white matter hyperintensities, FA and clinical mobility measures, after correcting for the false discovery rate; these modalities were not examined in further models.
Table 2:
Pearson correlation coefficients between functional mobility measures and whole brain neuroimaging measures.
| Measure | Gait Speed | Five Times Sit to Stand | Dynamic Gait Index | Four Square Step Test |
|---|---|---|---|---|
| Grey Matter Volume | 0.39* | −0.13 | 0.41* | −0.36* |
| White Matter Hyperintensities | −0.03 | 0.02 | −0.19 | 0.11 |
| Fractional Anisotropy | 0.08 | −0.13 | 0.24 | −0.18 |
False Discovery Rate method used to correct for multiple comparisons in each column, with adjusted p < 0.05 (Benjamini and Hochberg, 1995)
Because we observed a significant association between whole brain GMV and mobility test performance, we further investigated the relationship between mobility measures and 18 regions that have been related to mobility function (Table 3). After applying FDR, greater gait speed was correlated with greater regional GMV, most significantly in the parahippocampus and superior temporal gyrus regions. Greater DGI performance was also associated with higher GMV, especially the superior temporal gyrus, supramarginal gyrus and inferior parietal region. Additional significant correlations included hippocampus and parahippocampus. Faster performance on the FSST was related to increased GMV in the parahippocampus and superior temporal gyrus. Slightly weaker correlations were also found in all of the examined regions of the parietal lobe, and paracentral lobule.
Table 3:
Pearson correlation coefficients between functional mobility measures and regional grey matter volume measures.
| Regional grey matter volume | Gait Speed | Dynamic Gait Index | Four Square Step Test |
|---|---|---|---|
| Frontal |
|||
| Precentral gyrus | 0.12 | 0.25 | −0.20 |
| Middle frontal gyrus | 0.13 | 0.27 | −0.01 |
| Supplementary motor area | 0.17 | 0.22 | −0.15 |
| Paracentral lobule | 0.20 | 0.15 | −0.32* |
| Parietal |
|||
| Postcentral gyrus | 0.25 | 0.28 | −0.40* |
| Superior parietal lobe | 0.28 | 0.22 | −0.29* |
| Inferior parietal lobe | 0.29 | 0.44* | −0.33* |
| Supramarginal gyrus | 0.18 | 0.47* | −0.29* |
| Precuneus | 0.30 | 0.24 | −0.30* |
| Temporal |
|||
| Hippocampus | 0.22 | 0.33* | −0.26 |
| Parahippocampus | 0.42* | 0.39* | −0.44* |
| Superior temporal gyrus | 0.36* | 0.52* | −0.44* |
| Cerebellum | 0.17 | 0.11 | −0.12 |
| Vermis | 0.12 | 0.15 | −0.15 |
| Basal Ganglia |
|||
| Caudate | 0.26 | 0.11 | −0.26 |
| Pallidum | 0.12 | 0.12 | −0.003 |
| Putamen | 0.31 | 0.15 | −0.21 |
| Thalamus | 0.002 | 0.06 | −0.11 |
False Discovery Rate method used to correct for multiple comparisons in each column, with adjusted p < 0.05 (Benjamini and Hochberg, 1995)
Finally, we examined the independent associations between the neuroimaging measures and clinical mobility performance, after accounting for age, sex, and body mass index (Table 4). Greater parahippocampal volume was associated with both faster gait speed and FSST after adjusting for age, sex, and BMI. Postcentral gyrus, superior temporal gyrus GMV were associated with faster FSST performance. Greater superior temporal gyrus, supramarginal gyrus, and inferior parietal lobe GMV were related to higher DGI scores. In sensitivity analyses further adjusted for number of medications, results were substantially unchanged.
Table 4:
Standardized Coefficient Beta (standard error) values from multiple linear regression between functional mobility measures and regional grey matter volume (GMV) measures (Model 1), after adjusting for age (Model 2), age and sex (Model 3), and age, sex, and body mass index (Model 4).
| Model 1 | Model 2 | Model 3 | Model 4 | |
|---|---|---|---|---|
| Gait Speed |
||||
| Parahippocampus | 0.42 (0.11) | 0.32 (0.12) | 0.28 (0.12) | 0.23 (0.11) |
| Superior Temporal Gyrus | 0.36 (0.11) | 0.24 (0.13) | 0.26 (0.13) | 0.16 (0.12) |
| Dynamic Gait Index |
||||
| Parahippocampus | 0.39 (0.11) | 0.21 (0.11) | 0.17 (0.11) | 0.12 (0.11) |
| Superior Temporal Gyrus | 0.52 (0.10) | 0.35 (0.12) | 0.37 (0.11) | 0.29 (0.11) |
| Inferior Parietal lobe | 0.44 (0.11) | 0.28 (0.11) | 0.31 (0.11) | 0.25 (0.10) |
| Supramarginal gyrus | 0.47 (0.11) | 0.30 (0.11) | 0.35 (0.11) | 0.28 (0.11) |
| Hippocampus | 0.33 (0.11) | 0.17 (0.11) | 0.16 (0.11) | 0.20 (0.10) |
| Four Square Step Test |
||||
| Parahippocampus | −0.44 (0.11) | −0.40 (0.12) | −0.39 (0.12) | −0.32 (0.11) |
| Superior Temporal gyrus | −0.44 (0.11) | −0.41 (0.13) | −0.42 (0.13) | −0.30 (0.12) |
| Inferior Parietal lobe | −0.33 (0.12) | −0.26 (0.12) | −0.27 (0.13) | −0.19 (0.11) |
| Supramarginal gyrus | −0.29 (0.12) | −0.20 (0.13) | −0.24 (0.13) | −0.13 (0.12) |
| Paracentral lobule | −0.32 (0.12) | −0.25 (0.13) | −0.25 (0.13) | −0.22 (0.11) |
| Postcentral gyrus | −0.40 (0.11) | −0.34 (0.13) | −0.38 (0.13) | −0.30 (0.12) |
| Superior Parietal lobe | −0.29 (0.12) | −0.22 (0.13) | −0.26 (0.13) | −0.19 (0.12) |
| Precuneus | −0.30 (0.12) | −0.23 (0.12) | −0.23 (0.12) | −0.20 (0.11) |
Bold values indicate significant t-scores (p < 0.05)
DISCUSSION
Neuroimaging measures of a healthier brain, defined as higher whole brain gray matter volume, were related to better functional mobility represented by faster gait speed and FSST performance, as well as higher DGI scores. These findings are consistent with the predominant findings in literature.4–13 A novel finding of this research was the observation of significant associations between GMV, varying gait tasks (DGI) and multidirectional stepping (FSST). Greater whole brain GMV was related to greater performance in the DGI, which assesses functional gait activities such as walking with head turns, stepping over obstacles, and stair climbing, and faster completion of the FSST, which involves planning of multi-directional stepping as quickly as possible over canes. The lack of a significant correlation between the WMHV and FA and the mobility measures was surprising given the established findings in literature.8,9 However, it is possible that the relatively small sample that was included had a relatively narrow distribution of WMHV and FA, which precluded finding significant relationships. It is also possible that these mobility tasks rely less on the integrity of the neural pathways compared with the amount of grey matter in these specific regions.
In adjusted multivariate models, regional GMV were significantly associated with several of the mobility measures. In particular, higher GMV in the parahippocampus was related to faster gait speed and FSST performance. The parahippocampus has previously been related to gait activities in several functional neuroimaging studies that compared imagined walking to rest and its function has been attributed to processing of vestibular, somatosensory, and visual inputs during walking.27–29 The association of the superior temporal gyrus with DGI and FSST, combined with the strong association between inferior parietal lobe and supramarginal gyri with DGI, is consistent with the established role of these regions as part of the human vestibular cortex.30–33 Thus, it is meaningful that performance of dynamic gait tasks requiring intact vestibular function was related to volume in specific regions synonymous with this function. Additionally, larger postcentral gyrus GMV contributed independently to faster FSST times. Given its role as the primary somatosensory cortex, the relationship of the postcentral gyrus to FSST performance could reflect an important function of sensory feedback during performance of the multidirectional stepping task. Although several studies have demonstrated increased activation of the postcentral gyrus during imaging and real walking performance,28,34,35 others have reported decreased activation.29
Several weaknesses were noted with this study. The small sample size (n=70) may have limited the power to detect additional significant associations between neuroimaging measures and scores on functional balance outcomes. Additionally, the subjects were taken from a convenience sample (research registry) with baseline data indicating this population was healthier than the general older adult population; approximately 20% had falls in the past year which is less than the approximate amount of 29% for non-institutionalized older adults36. A more diverse selection would be beneficial to explore relationships with greater generalizability. This study is limited by the cross-sectional design. Future studies should analyze data in a longitudinal fashion to track changes in individuals over time to determine a relationship between gray matter volume changes and changes in functional mobility scores.
CONCLUSION:
Gray matter volumes of distinct regions were significantly correlated with scores on functional mobility measures of walking and multidirectional stepping. Regions associated with mobility measures are considered to be related to vestibular processing, motor planning, and somatosensory processing. It is important to consider changes in these central nervous system areas that may affect changes in mobility in older adults. Future research should track neuroimaging markers and scores on balance and mobility outcome measures longitudinally to measure these relationships over time.
ACKNOWLEDGMENTS
Sponsor’s Role: This research was supported by funding from the National Institutes of Health (R01 AG031118, P30 DC005205, T32 AG021885), including the Pittsburgh Claude D. Pepper Older Americans Independence Center (P30 AG024827), and the Eye and Ear Foundation. The sponsors did not contribute to the design, methods, subject recruitment, data collections, analysis and preparation of paper.
Abstract presented at the International Society of Posture and Gait Research World Congress June 2017. Supported by NIH grant T32 AG021885
Footnotes
Conflict of Interest: The authors have no conflicts.
REFERENCES
- 1.Burns E, Kakara R. Deaths from falls among persons aged≥ 65 years—united states, 2007–2016. Morbidity and Mortality Weekly Report. 2018;67(18):509. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Hirvensalo M, Rantanen T, Heikkinen E. Mobility difficulties and physical activity as predictors of mortality and loss of independence in the community-living older population. J Am Geriatr Soc. 2000;48(5):493–498. [DOI] [PubMed] [Google Scholar]
- 3.Tinetti ME, Kumar C. The patient who falls:“It’s always a trade-off”. JAMA. 2010;303(3):258–266. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Holtzer R, Epstein N, Mahoney JR, Izzetoglu M, Blumen HM. Neuroimaging of mobility in aging: A targeted review. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences. 2014;69(11):1375–1388. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Nadkarni NK, Nunley KA, Aizenstein H, et al. Association between cerebellar gray matter volumes, gait speed, and information-processing ability in older adults enrolled in the health abc study. Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences. 2013;69(8):996–1003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Rosano C, Bennett DA, Newman AB, et al. Patterns of focal gray matter atrophy are associated with bradykinesia and gait disturbances in older adults. Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences. 2012;67(9):957–962. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Boisgontier MP, Cheval B, van Ruitenbeek P, et al. Whole-brain grey matter density predicts balance stability irrespective of age and protects older adults from falling. Gait Posture. 2016;45:143–150. [DOI] [PubMed] [Google Scholar]
- 8.Zheng JJ, Delbaere K, Close JC, Sachdev PS, Lord SR. Impact of white matter lesions on physical functioning and fall risk in older people a systematic review. Stroke. 2011;42(7):2086–2090. [DOI] [PubMed] [Google Scholar]
- 9.Baloh RW, Ying SH, Jacobson KM. A longitudinal study of gait and balance dysfunction in normal older people. Arch Neurol. 2003;60(6):835–839. [DOI] [PubMed] [Google Scholar]
- 10.Moscufo N, Wakefield DB, Meier DS, et al. Longitudinal microstructural changes of cerebral white matter and their association with mobility performance in older persons. PLoS One. 2018;13(3):e0194051. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Cavallari M, Moscufo N, Skudlarski P, et al. Mobility impairment is associated with reduced microstructural integrity of the inferior and superior cerebellar peduncles in elderly with no clinical signs of cerebellar dysfunction. NeuroImage: Clinical. 2013;2:332–340. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Tian Q, Glynn NW, Erickson KI, et al. Objective measures of physical activity, white matter integrity and cognitive status in adults over age 80. Behav Brain Res. 2015;284:51–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Rosano C, Aizenstein HJ, Studenski S, Newman AB. A regions-of-interest volumetric analysis of mobility limitations in community-dwelling older adults. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences. 2007;62(9):1048–1055. [DOI] [PubMed] [Google Scholar]
- 14.Rigler SK, Studenski S, Wallace D, Reker DM, Duncan PW. Co-morbidity adjustment for functional outcomes in community-dwelling older adults. Clin Rehabil. 2002;16(4):420–428. [DOI] [PubMed] [Google Scholar]
- 15.Sparto PJ, Jennings JR, Furman JM, Redfern MS. Lateral step initiation behavior in older adults. Gait Posture. 2014;39(2):799–803. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Sparto PJ, Fuhrman SI, Redfern MS, et al. Postural adjustment errors reveal deficits in inhibition during lateral step initiation in older adults. J Neurophysiol. 2012;109(2):415–428. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Perera S, Patel KV, Rosano C, et al. Gait speed predicts incident disability: A pooled analysis. Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences. 2015;71(1):63–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Studenski S, Perera S, Patel K, et al. Gait speed and survival in older adults. JAMA. 2011;305(1):50–58. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Guralnik JM, Simonsick EM, Ferrucci L, et al. A short physical performance battery assessing lower extremity function: Association with self-reported disability and prediction of mortality and nursing home admission. J Gerontol. 1994;49(2):M85–M94. [DOI] [PubMed] [Google Scholar]
- 20.Buatois S, Perret-Guillaume C, Gueguen R, et al. A simple clinical scale to stratify risk of recurrent falls in community-dwelling adults aged 65 years and older. Phys Ther. 2010;90(4):550–560. [DOI] [PubMed] [Google Scholar]
- 21.Shumway-Cook A, Baldwin M, Polissar NL, Gruber W. Predicting the probability for falls in community-dwelling older adults. Phys Ther. 1997;77(8):812–819. [DOI] [PubMed] [Google Scholar]
- 22.Dite W, Temple VA. A clinical test of stepping and change of direction to identify multiple falling older adults. Arch Phys Med Rehabil. 2002;83(11):1566–1571. [DOI] [PubMed] [Google Scholar]
- 23.Wu M, Rosano C, Butters M, et al. A fully automated method for quantifying and localizing white matter hyperintensities on mr images. Psychiatry Res. 2006;148(2-3):133–142. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Wakana S, Jiang H, Nagae-Poetscher LM, van Zijl PC, Mori S. Fiber tract-based atlas of human white matter anatomy. Radiology. 2004;230(1):77–87. [DOI] [PubMed] [Google Scholar]
- 25.Smith SM, Jenkinson M, Johansen-Berg H, et al. Tract-based spatial statistics: Voxelwise analysis of multi-subject diffusion data. Neuroimage. 2006;31(4):1487–1505. [DOI] [PubMed] [Google Scholar]
- 26.Benjamini Y, Hochberg Y. Controlling the false discovery rate: A practical and powerful approach to multiple testing. Journal of the Royal statistical society: series B (Methodological). 1995;57(1):289–300. [Google Scholar]
- 27.Hamacher D, Herold F, Wiegel P, Hamacher D, Schega L. Brain activity during walking: A systematic review. Neurosci Biobehav Rev. 2015;57:310–327. [DOI] [PubMed] [Google Scholar]
- 28.La Fougere C, Zwergal A, Rominger A, et al. Real versus imagined locomotion: A [18f]-fdg pet-fmri comparison. Neuroimage. 2010;50(4):1589–1598. [DOI] [PubMed] [Google Scholar]
- 29.Deutschländer A, Stephan T, Hüfner K, et al. Imagined locomotion in the blind: An fmri study. Neuroimage. 2009;45(1):122–128. [DOI] [PubMed] [Google Scholar]
- 30.Zwergal A, Linn J, Xiong G, Brandt T, Strupp M, Jahn K. Aging of human supraspinal locomotor and postural control in fmri. Neurobiol Aging. 2012;33(6):1073–1084. [DOI] [PubMed] [Google Scholar]
- 31.Fasold O, von Brevern M, Kuhberg M, et al. Human vestibular cortex as identified with caloric stimulation in functional magnetic resonance imaging. Neuroimage. 2002;17(3):1384–1393. [DOI] [PubMed] [Google Scholar]
- 32.Dieterich M, Bense S, Lutz S, et al. Dominance for vestibular cortical function in the non-dominant hemisphere. Cereb Cortex. 2003;13(9):994–1007. [DOI] [PubMed] [Google Scholar]
- 33.Lopez C, Blanke O. The thalamocortical vestibular system in animals and humans. Brain Res Rev. 2011;67(1-2):119–146. [DOI] [PubMed] [Google Scholar]
- 34.Sacco K, Cauda F, Cerliani L, Mate D, Duca S, Geminiani G. Motor imagery of walking following training in locomotor attention. The effect of ‘the tango lesson’. Neuroimage. 2006;32(3):1441–1449. [DOI] [PubMed] [Google Scholar]
- 35.Van Der Meulen M, Allali G, Rieger SW, Assal F, Vuilleumier P. The influence of individual motor imagery ability on cerebral recruitment during gait imagery. Human brain mapping. 2014;35(2):455–470. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Bergen G. Falls and fall injuries among adults aged≥ 65 years—united states, 2014. MMWR Morbidity and mortality weekly report. 2016;65. [DOI] [PubMed] [Google Scholar]