[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address 50.16.52.237. Please contact the publisher to request reinstatement.
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Download PDF
Figure 1. Gait-related white matter (fractional anisotropy) alterations in all subjects and those with risk of falls. A, Relationship between the entire range of Tinetti scale scores (with or without risk of falls) and fractional anisotropy. Significant clusters are seen in the frontostriatal and temporal part of the superior longitudinal pathway. B, Relationship between Tinetti scale scores and fractional anisotropy in subjects with risk of falls. Clusters showing correlation between Mini-Mental State Examination (MMSE) and Tinetti scale scores (blue to light blue) are seen in the medial frontal and parietal peripheral pathways (I); MMSE-independent clusters (red to yellow) are seen in the genu and splenium of corpus callosum, posterior cingulum, prefrontal and orbitofrontal pathways (II), and longitudinal pathways that connect frontal-parietal-temporal circuits (III). Green indicates white matter skeleton.

Figure 1. Gait-related white matter (fractional anisotropy) alterations in all subjects and those with risk of falls. A, Relationship between the entire range of Tinetti scale scores (with or without risk of falls) and fractional anisotropy. Significant clusters are seen in the frontostriatal and temporal part of the superior longitudinal pathway. B, Relationship between Tinetti scale scores and fractional anisotropy in subjects with risk of falls. Clusters showing correlation between Mini-Mental State Examination (MMSE) and Tinetti scale scores (blue to light blue) are seen in the medial frontal and parietal peripheral pathways (I); MMSE-independent clusters (red to yellow) are seen in the genu and splenium of corpus callosum, posterior cingulum, prefrontal and orbitofrontal pathways (II), and longitudinal pathways that connect frontal-parietal-temporal circuits (III). Green indicates white matter skeleton.

Figure 2. Gait-related white matter (fractional anisotropy) alterations in subjects with risk of falls and no cognitive impairment (NCI). Significant albeit smaller abnormal fractional anisotropy clusters are seen mostly in the areas shown to be independent of Mini-Mental State Examination scores in Figure 1.

Figure 2. Gait-related white matter (fractional anisotropy) alterations in subjects with risk of falls and no cognitive impairment (NCI). Significant albeit smaller abnormal fractional anisotropy clusters are seen mostly in the areas shown to be independent of Mini-Mental State Examination scores in Figure 1.

Table. Demographic, Clinical, and Imaging Characteristics of Subjects
Table. Demographic, Clinical, and Imaging Characteristics of Subjects
1.
Tinetti ME, Williams CS. Falls, injuries due to falls, and the risk of admission to a nursing home.  N Engl J Med. 1997;337(18):1279-1284PubMedArticle
2.
Gates S, Smith LA, Fisher JD, Lamb SE. Systematic review of accuracy of screening instruments for predicting fall risk among independently living older adults.  J Rehabil Res Dev. 2008;45(8):1105-1116PubMedArticle
3.
Tinetti ME. Performance-oriented assessment of mobility problems in elderly patients.  J Am Geriatr Soc. 1986;34(2):119-126PubMed
4.
Tinetti ME. Clinical practice: preventing falls in elderly persons.  N Engl J Med. 2003;348(1):42-49PubMedArticle
5.
Fields RD. White matter in learning, cognition and psychiatric disorders.  Trends Neurosci. 2008;31(7):361-370PubMedArticle
6.
Scherder E, Eggermont L, Visscher C, Scheltens P, Swaab D. Understanding higher level gait disturbances in mild dementia in order to improve rehabilitation: “last in-first out.”  Neurosci Biobehav Rev. 2011;35(3):699-714PubMedArticle
7.
Snijders AH, van de Warrenburg BP, Giladi N, Bloem BR. Neurological gait disorders in elderly people: clinical approach and classification.  Lancet Neurol. 2007;6(1):63-74PubMedArticle
8.
Bhadelia RA, Price LL, Tedesco KL,  et al.  Diffusion tensor imaging, white matter lesions, the corpus callosum, and gait in the elderly.  Stroke. 2009;40(12):3816-3820PubMedArticle
9.
de Laat KF, Tuladhar AM, van Norden AG, Norris DG, Zwiers MP, de Leeuw FE. Loss of white matter integrity is associated with gait disorders in cerebral small vessel disease.  Brain. 2011;134(pt 1):73-83PubMedArticle
10.
Srikanth V, Phan TG, Chen J, Beare R, Stapleton JM, Reutens DC. The location of white matter lesions and gait: a voxel-based study.  Ann Neurol. 2010;67(2):265-269PubMedArticle
11.
Scott TM, Peter I, Tucker KL,  et al.  The Nutrition, Aging, and Memory in Elders (NAME) study: design and methods for a study of micronutrients and cognitive function in a homebound elderly population.  Int J Geriatr Psychiatry. 2006;21(6):519-528PubMedArticle
12.
Smith SM, Jenkinson M, Johansen-Berg H,  et al.  Tract-based spatial statistics: voxelwise analysis of multi-subject diffusion data.  Neuroimage. 2006;31(4):1487-1505PubMedArticle
13.
DeCarli C, Maisog J, Murphy DG, Teichberg D, Rapoport SI, Horwitz B. Method for quantification of brain, ventricular, and subarachnoid CSF volumes from MR images.  J Comput Assist Tomogr. 1992;16(2):274-284PubMedArticle
14.
DeCarli C, Murphy DG, Tranh M,  et al.  The effect of white matter hyperintensity volume on brain structure, cognitive performance, and cerebral metabolism of glucose in 51 healthy adults.  Neurology. 1995;45(11):2077-2084PubMedArticle
15.
Tucker KL, Falcon LM, Bianchi LA, Cacho E, Bermudez OI. Self-reported prevalence and health correlates of functional limitation among Massachusetts elderly Puerto Ricans, Dominicans, and non-Hispanic white neighborhood comparison group.  J Gerontol A Biol Sci Med Sci. 2000;55(2):M90-M97PubMedArticle
16.
Hamilton M. A rating scale for depression.  J Neurol Neurosurg Psychiatry. 1960;23:56-62PubMedArticle
17.
Morris JC. The Clinical Dementia Rating (CDR): current version and scoring rules.  Neurology. 1993;43(11):2412-2414PubMedArticle
18.
Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”: a practical method for grading the cognitive state of patients for the clinician.  J Psychiatr Res. 1975;12(3):189-198PubMedArticle
19.
Perneczky R, Wagenpfeil S, Komossa K, Grimmer T, Diehl J, Kurz A. Mapping scores onto stages: Mini-Mental State Examination and Clinical Dementia Rating.  Am J Geriatr Psychiatry. 2006;14(2):139-144PubMedArticle
20.
McKhann GM, Knopman DS, Chertkow H,  et al.  The diagnosis of dementia due to Alzheimer's disease: recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease.  Alzheimers Dement. 2011;7(3):263-269PubMedArticle
21.
Román GC, Tatemichi TK, Erkinjuntti T,  et al.  Vascular dementia: diagnostic criteria for research studies: report of the NINDS-AIREN International Workshop.  Neurology. 1993;43(2):250-260PubMedArticle
22.
Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E. Mild cognitive impairment: clinical characterization and outcome.  Arch Neurol. 1999;56(3):303-308PubMedArticle
23.
Sun X, Bhadelia R, Liebson E,  et al.  The relationship between plasma amyloid-β peptides and the medial temporal lobe in the homebound elderly.  Int J Geriatr Psychiatry. 2011;26(6):593-601PubMedArticle
24.
Shumway-Cook A, Woollacott MH. Motor Control: Theory and Practical Applications. 2nd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2001
25.
Society of General Internal Medicine.  Tinetti Assessment Scale: description. http://www.sgim.org/userfiles/file/handout16TinettiAssessmentTool1.pdf. Accessed August 23, 2011
26.
Nichols TE, Holmes AP. Nonparametric permutation tests for functional neuroimaging: a primer with examples.  Hum Brain Mapp. 2002;15(1):1-25PubMedArticle
27.
Moscufo N, Guttmann CR, Meier D,  et al.  Brain regional lesion burden and impaired mobility in the elderly.  Neurobiol Aging. 2011;32(4):646-654PubMedArticle
28.
Buracchio T, Dodge HH, Howieson D, Wasserman D, Kaye J. The trajectory of gait speed preceding mild cognitive impairment.  Arch Neurol. 2010;67(8):980-986PubMedArticle
29.
Dubost V, Kressig RW, Gonthier R,  et al.  Relationships between dual-task related changes in stride velocity and stride time variability in healthy older adults.  Hum Mov Sci. 2006;25(3):372-382PubMedArticle
30.
Sahyoun C, Floyer-Lea A, Johansen-Berg H, Matthews PM. Towards an understanding of gait control: brain activation during the anticipation, preparation and execution of foot movements.  Neuroimage. 2004;21(2):568-575PubMedArticle
31.
Verghese J, Wang C, Lipton RB, Holtzer R, Xue X. Quantitative gait dysfunction and risk of cognitive decline and dementia.  J Neurol Neurosurg Psychiatry. 2007;78(9):929-935PubMedArticle
32.
Köpke S, Meyer G. The Tinetti test: Babylon in geriatric assessment.  Z Gerontol Geriatr. 2006;39(4):288-291PubMedArticle
33.
Kegelmeyer DA, Kloos AD, Thomas KM, Kostyk SK. Reliability and validity of the Tinetti Mobility Test for individuals with Parkinson disease.  Phys Ther. 2007;87(10):1369-1378PubMedArticle
34.
Lin MR, Hwang HF, Hu MH, Wu HD, Wang YW, Huang FC. Psychometric comparisons of the timed up and go, one-leg stand, functional reach, and Tinetti balance measures in community-dwelling older people.  J Am Geriatr Soc. 2004;52(8):1343-1348PubMedArticle
35.
Raîche M, Hébert R, Prince F, Corriveau H. Screening older adults at risk of falling with the Tinetti balance scale.  Lancet. 2000;356(9234):1001-1002PubMedArticle
36.
Kloos AD, Kegelmeyer DA, Young GS, Kostyk SK. Fall risk assessment using the Tinetti mobility test in individuals with Huntington's disease.  Mov Disord. 2010;25(16):2838-2844PubMedArticle
37.
Blahak C, Baezner H, Pantoni L,  et al; LADIS Study Group.  Deep frontal and periventricular age related white matter changes but not basal ganglia and infratentorial hyperintensities are associated with falls: cross sectional results from the LADIS study.  J Neurol Neurosurg Psychiatry. 2009;80(6):608-613PubMedArticle
38.
Taguchi N, Higaki Y, Inoue S, Kimura H, Tanaka K. Effects of a 12-month multicomponent exercise program on physical performance, daily physical activity, and quality of life in very elderly people with minor disabilities: an intervention study.  J Epidemiol. 2010;20(1):21-29PubMedArticle
Original Contribution
June 2012

Clinical Prediction of Fall Risk and White Matter AbnormalitiesA Diffusion Tensor Imaging Study

Author Affiliations

Author Affiliations: Departments of Anatomy and Neurobiology (Drs Koo and Bergethon and Mr Hussain) and Psychiatry (Dr Qiu), Boston University School of Medicine, Tufts Medical Center and School of Medicine (Drs Bergethon, Qiu, and Scott) and Gerald J. and Dorothy R. Friedman School of Nutrition Science and Policy (Drs Scott and Rosenberg), Tufts University, and Departments of Neurology (Dr Caplan) and Radiology (Dr Bhadelia), Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, Massachusetts.

Arch Neurol. 2012;69(6):733-738. doi:10.1001/archneurol.2011.2272
Abstract

Background The Tinetti scale is a simple clinical tool designed to predict risk of falling by focusing on gait and stance impairment in elderly persons. Gait impairment is also associated with white matter (WM) abnormalities.

Objective To test the hypothesis that elderly subjects at risk for falling, as determined by the Tinetti scale, have specific patterns of WM abnormalities on diffusion tensor imaging.

Design, Setting, and Patients Community-based cohort of 125 homebound elderly individuals.

Main Outcome Measures Diffusion tensor imaging scans were analyzed using tract-based spatial statistics analysis to determine the location of WM abnormalities in subjects with Tinetti scale scores of 25 or higher (without risk of falls) and lower than 25 (with risk of falls). Multivariate linear least squares correlation analysis was performed to determine the association between Tinetti scale scores and local fractional anisotropy values on each skeletal voxel controlling for possible confounders.

Results In subjects with risk of falls (Tinetti scale score <25), clusters of abnormal WM were seen in the medial frontal and parietal subcortical pathways, genu and splenium of corpus callosum, posterior cingulum, prefrontal and orbitofrontal pathways, and longitudinal pathways that connect frontal-parietal-temporal lobes. Among these abnormalities, those in medial frontal and parietal subcortical pathways correlated with Mini-Mental State Examination scores, while the other locations were unrelated to these scores.

Conclusions Elderly individuals at risk for falls as determined by the Tinetti scale have WM abnormalities in specific locations on diffusion tensor imaging, some of which correlate with cognitive function scores.

Gait impairment is common in elderly persons and is associated with increased risk of falling. Falls often cause trauma and fractures that can lead to hospitalization, loss of independence, and death. Nearly 30% of individuals older than 65 years have 1 or more falls each year, and this number increases to 50% by age 80 years.1 Given the importance of fall assessment in elderly persons, several tools that evaluate gait impairment as related to risk of falling have been designed for clinical practices.2 These tools can help detect, identify, and then guide interventions to prevent injury in patients at high risk for falls. The Tinetti scale is an exemplar of a simple clinical tool designed to evaluate gait function that can be used to assess the risk of falling.3 It is usually used in conjunction with other clinical information such as a history of sedative and other drug use, orthostatic blood pressure, vestibular integrity, and assessment of cognitive and behavioral functions.1,4

Gait impairment can be a result of peripheral nerve, spinal cord, and/or brain dysfunction. White matter (WM) tracts are of particular interest because they communicate between key cortical and subcortical regions that relate to stance and gait.57 Natural gait is dependent on integration of these functions. Several previous imaging studies investigated gait-related WM abnormalities.810 However, correlations between Tinetti scale scores and WM structure as delineated by brain imaging have not been made. Predicting specific patterns of WM abnormalities in subjects identified to be at risk for falling is important not only for understanding the pathobiological basis of these clinical observations but also for increasing the value and applied use of clinical tools such as the Tinetti scale.

We investigated the hypothesis that elderly subjects at risk for falling as determined by the Tinetti scale would have specific patterns of WM abnormalities on diffusion tensor imaging (DTI). We further hypothesized that certain patterns of DTI abnormalities in these subjects are related to cognitive function and that some are not.

METHODS
SUBJECTS

The study sample consisted of 173 participants of the Nutrition, Aging, and Memory in Elders study who had magnetic resonance imaging (MRI) evaluations with DTI. The study designs, including the inclusion and exclusion criteria and clinical assessment, have been described in detail previously.8,11 Briefly, a subset of 366 subjects from a total of 1246 subjects recruited from Boston's 3 Aging Services Access Points had detailed psychiatric, neurological, and MRI examinations in addition to the routine nutritional, neuropsychological, medical-historical, and blood chemistry evaluations. The Aging Services Access Points provide services and support for independent living to elderly individuals who are at least 60 years old and have low income, diminishment in activities of daily living, and needs in a critical area such as food or personal care. The Tufts–New England Medical Center Institutional Review Board approved the study, and all participants signed informed consent. The original DTI sample consisted of 187 subjects, of whom 13 were excluded owing to missing Tinetti scale scores, bookkeeping errors, and, in 1 case, an amputated leg (limiting assessment by the Tinetti scale). Additionally, 42 subjects were excluded owing to insufficient MRI fields of view for covering the whole cortical area, a requirement for the normalization process for tract-based spatial statistics (TBSS) analysis.12 Six additional subjects were excluded based on Mini-Mental State Examination (MMSE) scores as explained later, leaving a final sample of 125 participants.

MRI SCANNING AND IMAGE PROCESSING

All subjects were imaged on a 1.5-T scanner (Siemens Symphony). The DTI was performed using a single-shot, spin-echo, echo-planar sequence and 6 independent directions. Imaging data were preprocessed to extract noise and noncortical regions. Fractional anisotropy (FA) values were then calculated and put into the TBSS pipeline,12 where we created a study-specific template and FA skeleton indicating major fiber pathways. All normalized FA data were then projected onto this skeleton. After this projection, statistical processing was done on each overlaid voxel. We determined WM hyperintensity volume and intracranial volumes by using the histogram analysis method.13,14

HEALTH, NEUROPSYCHOLOGICAL, AND NEUROLOGICAL ASSESSMENTS

Extensive demographic and laboratory data were collected for each Nutrition, Aging, and Memory in Elders study subject.11 Participants responded to questions documenting the presence of abnormalities from a list of chronic conditions and health events. Physical functioning was examined using modified activities of daily living.15 A board-certified psychiatrist evaluated the subjects and recorded the Hamilton Rating Scale for Depression, Clinical Dementia Rating Scale, and MMSE scores (0-30).1618 Six of the 131 subjects with MMSE scores below 21 were excluded from the analysis because of the likelihood of association with moderate to severe dementia.19 A board-certified neurologist performed neurological evaluations and determined whether the subject had findings of symptomatic stroke and a peripheral neuropathic or spinal cord disorder.

A consensus diagnosis of psychiatric and neurological disorders was made at a meeting attended by the study psychiatrist, neuropsychologist, neurologist, and neuroradiologist.11 Each subject was assigned to any (or none) of the diagnoses of Alzheimer disease (possible or probable), mild cognitive impairment, vascular dementia, and depression according to defined criteria.11,2023 Subjects were considered to have no cognitive impairment if they were not demented and had scored no more than 1 SD below the mean of age- and education-defined strata on MMSE scores and no more than 1.5 SD below the mean of age- and education-defined strata on neuropsychological tests.23

GAIT ASSESSMENT AND CATEGORIZATION OF TINETTI SCALE SCORES

Each participant's gait and balance were assessed during neurological examination including a standardized Tinetti scale.3 The Tinetti balance and gait evaluation is designed to assess both balance and gait using a simple bedside clinical evaluation scale that has a maximum total score of 28 (16 points for best balance; 12 points for best gait). The balance examination begins with the subject seated in a hard, armless chair, and evaluation of each of 3 broad performance outcomes are made: (1) sitting and rising (0-5 points); (2) standing and turning (0-9 points); and (3) sitting down (0-2 points). The gait examination is conducted after the subject is standing and is asked to walk a distance down a hallway first at a normal pace and then returning at a rapid but safe pace. Gait is evaluated in the following categories: (1) initiation (0-1 point); (2) step morphology (length, height, symmetry, continuity; 0-6 points); and (3) gait trajectory (path, truncal sway or flexion, walking stance; 0-5 points). For the purpose of analysis, subjects with Tinetti scale scores of 25 or higher were considered to be without risk of falls and those with scores lower than 25 were considered to be with risk of falls.24,25

STATISTICAL ANALYSIS

All of the demographic and clinical variables were compared between the group without risk of falls and the group with risk of falls. For the TBSS analysis, we used multivariate nonparametric correlation to determine the association between Tinetti scale scores and local FA values on each skeletal voxel controlling for possible confounders such as age, sex, arthritis, neuropathy, stroke, and WM hyperintensity volume.8 Parkinsonism or any variety of Parkinson disease was present in only 2 subjects and therefore was not used as a confounding variable for analysis. First, the entire range of Tinetti scale scores was used as an independent variable to assess the relationship with FA. Second, relationships between Tinetti scale scores and FA were assessed individually in subjects without and with risk of falls. Finally, the relationship between Tinetti scale scores and FA was assessed in subjects without and with risk of falls who were found to have no cognitive impairment. All TBSS voxelwise statistical analyses were based on a permutation-based inference method for nonparametric statistical thresholding, which corrects for multiple comparisons by using the null distribution of the maximum (across the image) voxelwise test statistics.26 This approach allows inference on statistical maps in the case of unknown null distribution. Here, 5000 iterations were used to calculate statistical inferences (corrected P < .05) for all statistical analyses.

RESULTS

The demographic, clinical, and imaging characteristics of the subjects are shown in the Table. Of the 125 subjects in the study, 78 had Tinetti scale scores of 25 or higher and were considered without risk of falls; 47 had Tinetti scale scores lower than 25 and were considered with risk of falls. The 2 groups of subjects had no significant differences except for MMSE scores. Although the percentage of those with stroke was higher in those with risk of falls compared with those without risk of falls, the difference did not reach statistical significance (P = .07). The MMSE scores were lower in those with risk of falls compared with those without risk of falls (P = .004). The percentage of subjects with mild cognitive impairment was slightly higher in those without risk of falls and the percentage of those with Alzheimer disease was slightly higher in those with risk of falls, but neither reached statistical significance.

Significant abnormal FA clusters (representing loss of WM integrity) were seen in the genu of corpus callosum as well as the frontostriatal and temporal part of the superior longitudinal pathway when the entire study population, including the entire range of Tinetti scale scores (without or with risk of falls), was evaluated (Figure 1A).

When subjects were grouped into those without and with risk of falls (high and low Tinetti scale scores), no abnormal WM FA clusters were seen in those without risk of falls (Tinetti scale score ≥25). However, when subjects with risk of falls (Tinetti scale score <25) were evaluated, significant abnormal FA clusters were seen in the medial frontal and parietal peripheral pathways, genu and splenium of corpus callosum, posterior cingulum, prefrontal and orbitofrontal pathways, and the longitudinal pathways that connect frontal-parietal-temporal lobes (Figure 1B). Furthermore, clusters in the medial frontal and parietal subcortical pathways (Figure 1B) were correlated with MMSE scores and were not seen after adjusting for MMSE scores in the regression analysis. However, the other FA clusters (Figure 1B) were independent of MMSE scores and were seen after MMSE adjustments were made in the regression analysis.

Finally, when the subjects with no cognitive impairment were evaluated separately, only those with risk of falls (Tinetti scale score <25) showed significant albeit smaller abnormal FA clusters (Figure 2). These occurred mostly in the areas shown to be independent of MMSE scores in Figure 1.

COMMENT

The results of this DTI study in a community-based sample of elderly subjects add to previous clinical and neuroimaging observations.69,2730 The study shows that subjects who were clinically determined to be at risk for falls by the Tinetti scale had WM abnormalities in specific locations, which were not seen in those who were determined not to be at risk for falls by the same scale. In those with risk of falls, we observed a pattern of WM abnormalities in the medial frontal and parietal peripheral pathways as well as in the genu and splenium of corpus callosum, posterior cingulum, prefrontal and orbitofrontal pathways, and longitudinal pathways that connect frontal-parietal-temporal circuits. We also showed that the WM abnormalities in the medial frontal and parietal subcortical pathways were dependent on the MMSE scores, with a higher probability of finding these lesions in subjects with low MMSE scores. Locations of the other WM abnormalities were unrelated to the MMSE scores. Finally, we also observed that individuals with risk of falls who had no cognitive impairment also had WM abnormalities mostly in the areas shown to be independent of MMSE scores. However, the abnormalities in this group were smaller than those observed in all subjects with risk of falls, likely owing to a smaller number of individuals included for analysis.

Gait assessment has been used in the elderly population for detection of neurodegenerative disorders.31 A recent longitudinal study showed that gait speed decline can be a useful predictor of mild cognitive impairment up to 10 years before the initial diagnosis is secured.28 Most obviously and importantly, gait assessment can be used to evaluate an individual's risk of falling. Gait assessment can be performed by clinical tools such as the Tinetti scale or by more extensive analysis of gait velocity, stride length, width, and cadence. While the latter approach provides a more quantitative assessment of gait, tools like the Tinetti scale have the advantage of providing quick and simple bedside assessment, albeit at a risk of subjective interpretations. The Tinetti scale has been used independently or as a component of more elaborate clinical fall assessment instruments in which other clinical variables are also considered.

Wide variations are found in the literature in interpretation of the Tinetti scale owing to different scoring systems and cutoff values used to determine its validity and reliability.32 Nevertheless, the Tinetti scale, which provides information regarding gait and stance of an individual, is found to be an effective tool for assessment of fall risk.24,3335 The gait and balance subscales of the Tinetti scale have been used separately or in a combined maximum score of 28, with a higher score indicating better mobility function. Using the Tinetti scale, an individual's risk of falling is considered to be moderate when the score is between 19 and 24 and high when the score is lower than 19.24,25,33 In this study, we used a cutoff score of less than 25 for assessing any fall risk in absence of actual fall data. Using similar cutoff values, previous studies have shown the sensitivity of the Tinetti scale to be about 70% and the specificity to be between 53% and 60% in predicting a fall within 1 year.35,36

Abnormalities of the WM have been described in subjects with gait impairment using both conventional MRI and DTI. In a previous study, using a region-of-interest analysis approach to DTI, we found a significant relationship between Tinetti scale scores and FA in the genu of corpus callosum.8 This relationship was independent of various other confounders including MMSE scores. In a recent large DTI study, de Laat et al9 evaluated the relationship between gait function and FA. These investigators found widespread WM abnormalities in almost all regions of commissural, association, and projection fibers related to various gait parameters. In our study, we have further explored the relationship between the Tinetti scale score and gait using a TBSS approach to DTI. Quantitative evaluation of whole-brain WM was possible in this study by using this approach. Furthermore, TBSS analysis ensures minimum bias from including non-WM voxels during the smoothing process and reduces structural variability among the subjects.

The significant relationships between gait, MMSE scores, and WM alterations in frontal connections that we describe here further argue for an anatomical basis for previous suggestions, which considered the bilateral frontal network as a core functional subsystem for both cognitive and gait functioning. In addition, we also observed parietal peripheral abnormalities on DTI that were dependent on MMSE scores. The idea that there is involvement of higher-level neural function in gait is supported by dual-task paradigms showing interference with gait function by loading with a cognitive task29 and was also shown in a recent functional neuroimaging study.30 Blahak et al37 reported an association between the incidence of falls and deep frontal WM hyperintensities on conventional T2-weighted MRI. Our findings confirm their observations and are further evidence that these focal abnormalities are related to cognitive function.

We also observed several WM abnormalities in specific locations that were not correlated with a subject's MMSE scores. The MMSE-independent WM alterations in the genu of corpus callosum observed in this study confirm previous neuroimaging observations of their important role in gait disturbances.8,9 We observed abnormal WM in the splenium of corpus callosum, a finding described previously.27 From these results, the genu and splenium of corpus callosum might be regarded as functioning as a connector hub that provides coordination and interaction between the neural networks within the hemispheres facilitating normal gait. In addition, several other MMSE-independent WM alterations were seen in the association tracts connecting anterior to posterior cortical regions and cortico-pontine-cerebellar pathways. These abnormalities are located within the areas of cortico-cortical and cortico-subcortical connections believed to be responsible for maintenance of normal gait.6,7

The rationale for dividing study subjects into 2 main categories of those without and with fall risk based on the Tinetti scale score was to identify the specific WM abnormalities that were clinically relevant. While MRI can be used to evaluate a pathoanatomical basis for gait disorders, using it widely for the routine clinical assessment of any patient with gait impairment would be neither practical nor fiscally prudent. The Tinetti scale can be used as a screening tool to select patients for MRI. Our results have shown that patients with a Tinetti scale score of 25 or higher will not show significant WM abnormalities, and MRI is not likely to add relevant information. However, those with Tinetti scale scores lower than 25 have multiple WM abnormalities, and these subjects might benefit from the additional data provided in MRI using tractography.

Our findings suggest that judicious use of MRI can help in improving the sensitivity and specificity of easily accessible and inexpensive clinical evaluation methods of gait and balance such as the Tinetti scale to identify patients at risk for falls. However, before our results from a group of homebound elderly individuals (requiring some assistance for independent living) are generalized, we suggest that our approach might be used as a basis for future studies by longitudinally following the subjects with gait impairment with information on the actual history of falls. Such a study using DTI measures with more directions than the 6 used in this study should enable more accurate fiber tracking for localization of WM abnormalities in relation to specific tracts. By demonstrating brain abnormalities in specific locations on DTI, such studies may help in developing an evidence-based clinical matrix for fall risk consisting of Tinetti scale and cognitive function scores along with other clinical risk factors for falls. Furthermore, this future study might also address the question of whether the use of expensive MRI technology can benefit in guiding specific prevention and rehabilitation strategies for people with gait impairment who are at risk for falls by demonstrating WM abnormalities in specific tracts.6,38

Back to top
Article Information

Correspondence: Rafeeque A. Bhadelia, MD, Department of Radiology, Beth Israel Deaconess Medical Center, WCB90, 330 Brookline Ave, Boston, MA 02115 (rbhadeli@bidmc.harvard.edu).

Accepted for Publication: September 30, 2011.

Published Online: February 13, 2012. doi:10.1001/archneurol.2011.2272

Author Contributions:Study concept and design: Koo, Bergethon, and Bhadelia. Acquisition of data: Qiu, Scott, Hussain, Rosenberg, and Bhadelia. Analysis and interpretation of data: Koo, Bergethon, Qiu, Scott, Hussain, Rosenberg, Caplan, and Bhadelia. Drafting of the manuscript: Koo, Bergethon, Qiu, Hussain, and Bhadelia. Critical revision of the manuscript for important intellectual content: Koo, Bergethon, Scott, Rosenberg, Caplan, and Bhadelia. Statistical analysis: Koo and Scott. Obtained funding: Bergethon, Qiu, and Rosenberg. Administrative, technical, and material support: Bergethon, Qiu, Scott, Hussain, Rosenberg, and Bhadelia. Study supervision: Bergethon, Caplan, and Bhadelia.

Financial Disclosure: None reported.

Funding/Support: This work was supported by grant AG21790-01 from the National Institute on Aging.

REFERENCES
1.
Tinetti ME, Williams CS. Falls, injuries due to falls, and the risk of admission to a nursing home.  N Engl J Med. 1997;337(18):1279-1284PubMedArticle
2.
Gates S, Smith LA, Fisher JD, Lamb SE. Systematic review of accuracy of screening instruments for predicting fall risk among independently living older adults.  J Rehabil Res Dev. 2008;45(8):1105-1116PubMedArticle
3.
Tinetti ME. Performance-oriented assessment of mobility problems in elderly patients.  J Am Geriatr Soc. 1986;34(2):119-126PubMed
4.
Tinetti ME. Clinical practice: preventing falls in elderly persons.  N Engl J Med. 2003;348(1):42-49PubMedArticle
5.
Fields RD. White matter in learning, cognition and psychiatric disorders.  Trends Neurosci. 2008;31(7):361-370PubMedArticle
6.
Scherder E, Eggermont L, Visscher C, Scheltens P, Swaab D. Understanding higher level gait disturbances in mild dementia in order to improve rehabilitation: “last in-first out.”  Neurosci Biobehav Rev. 2011;35(3):699-714PubMedArticle
7.
Snijders AH, van de Warrenburg BP, Giladi N, Bloem BR. Neurological gait disorders in elderly people: clinical approach and classification.  Lancet Neurol. 2007;6(1):63-74PubMedArticle
8.
Bhadelia RA, Price LL, Tedesco KL,  et al.  Diffusion tensor imaging, white matter lesions, the corpus callosum, and gait in the elderly.  Stroke. 2009;40(12):3816-3820PubMedArticle
9.
de Laat KF, Tuladhar AM, van Norden AG, Norris DG, Zwiers MP, de Leeuw FE. Loss of white matter integrity is associated with gait disorders in cerebral small vessel disease.  Brain. 2011;134(pt 1):73-83PubMedArticle
10.
Srikanth V, Phan TG, Chen J, Beare R, Stapleton JM, Reutens DC. The location of white matter lesions and gait: a voxel-based study.  Ann Neurol. 2010;67(2):265-269PubMedArticle
11.
Scott TM, Peter I, Tucker KL,  et al.  The Nutrition, Aging, and Memory in Elders (NAME) study: design and methods for a study of micronutrients and cognitive function in a homebound elderly population.  Int J Geriatr Psychiatry. 2006;21(6):519-528PubMedArticle
12.
Smith SM, Jenkinson M, Johansen-Berg H,  et al.  Tract-based spatial statistics: voxelwise analysis of multi-subject diffusion data.  Neuroimage. 2006;31(4):1487-1505PubMedArticle
13.
DeCarli C, Maisog J, Murphy DG, Teichberg D, Rapoport SI, Horwitz B. Method for quantification of brain, ventricular, and subarachnoid CSF volumes from MR images.  J Comput Assist Tomogr. 1992;16(2):274-284PubMedArticle
14.
DeCarli C, Murphy DG, Tranh M,  et al.  The effect of white matter hyperintensity volume on brain structure, cognitive performance, and cerebral metabolism of glucose in 51 healthy adults.  Neurology. 1995;45(11):2077-2084PubMedArticle
15.
Tucker KL, Falcon LM, Bianchi LA, Cacho E, Bermudez OI. Self-reported prevalence and health correlates of functional limitation among Massachusetts elderly Puerto Ricans, Dominicans, and non-Hispanic white neighborhood comparison group.  J Gerontol A Biol Sci Med Sci. 2000;55(2):M90-M97PubMedArticle
16.
Hamilton M. A rating scale for depression.  J Neurol Neurosurg Psychiatry. 1960;23:56-62PubMedArticle
17.
Morris JC. The Clinical Dementia Rating (CDR): current version and scoring rules.  Neurology. 1993;43(11):2412-2414PubMedArticle
18.
Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”: a practical method for grading the cognitive state of patients for the clinician.  J Psychiatr Res. 1975;12(3):189-198PubMedArticle
19.
Perneczky R, Wagenpfeil S, Komossa K, Grimmer T, Diehl J, Kurz A. Mapping scores onto stages: Mini-Mental State Examination and Clinical Dementia Rating.  Am J Geriatr Psychiatry. 2006;14(2):139-144PubMedArticle
20.
McKhann GM, Knopman DS, Chertkow H,  et al.  The diagnosis of dementia due to Alzheimer's disease: recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease.  Alzheimers Dement. 2011;7(3):263-269PubMedArticle
21.
Román GC, Tatemichi TK, Erkinjuntti T,  et al.  Vascular dementia: diagnostic criteria for research studies: report of the NINDS-AIREN International Workshop.  Neurology. 1993;43(2):250-260PubMedArticle
22.
Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E. Mild cognitive impairment: clinical characterization and outcome.  Arch Neurol. 1999;56(3):303-308PubMedArticle
23.
Sun X, Bhadelia R, Liebson E,  et al.  The relationship between plasma amyloid-β peptides and the medial temporal lobe in the homebound elderly.  Int J Geriatr Psychiatry. 2011;26(6):593-601PubMedArticle
24.
Shumway-Cook A, Woollacott MH. Motor Control: Theory and Practical Applications. 2nd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2001
25.
Society of General Internal Medicine.  Tinetti Assessment Scale: description. http://www.sgim.org/userfiles/file/handout16TinettiAssessmentTool1.pdf. Accessed August 23, 2011
26.
Nichols TE, Holmes AP. Nonparametric permutation tests for functional neuroimaging: a primer with examples.  Hum Brain Mapp. 2002;15(1):1-25PubMedArticle
27.
Moscufo N, Guttmann CR, Meier D,  et al.  Brain regional lesion burden and impaired mobility in the elderly.  Neurobiol Aging. 2011;32(4):646-654PubMedArticle
28.
Buracchio T, Dodge HH, Howieson D, Wasserman D, Kaye J. The trajectory of gait speed preceding mild cognitive impairment.  Arch Neurol. 2010;67(8):980-986PubMedArticle
29.
Dubost V, Kressig RW, Gonthier R,  et al.  Relationships between dual-task related changes in stride velocity and stride time variability in healthy older adults.  Hum Mov Sci. 2006;25(3):372-382PubMedArticle
30.
Sahyoun C, Floyer-Lea A, Johansen-Berg H, Matthews PM. Towards an understanding of gait control: brain activation during the anticipation, preparation and execution of foot movements.  Neuroimage. 2004;21(2):568-575PubMedArticle
31.
Verghese J, Wang C, Lipton RB, Holtzer R, Xue X. Quantitative gait dysfunction and risk of cognitive decline and dementia.  J Neurol Neurosurg Psychiatry. 2007;78(9):929-935PubMedArticle
32.
Köpke S, Meyer G. The Tinetti test: Babylon in geriatric assessment.  Z Gerontol Geriatr. 2006;39(4):288-291PubMedArticle
33.
Kegelmeyer DA, Kloos AD, Thomas KM, Kostyk SK. Reliability and validity of the Tinetti Mobility Test for individuals with Parkinson disease.  Phys Ther. 2007;87(10):1369-1378PubMedArticle
34.
Lin MR, Hwang HF, Hu MH, Wu HD, Wang YW, Huang FC. Psychometric comparisons of the timed up and go, one-leg stand, functional reach, and Tinetti balance measures in community-dwelling older people.  J Am Geriatr Soc. 2004;52(8):1343-1348PubMedArticle
35.
Raîche M, Hébert R, Prince F, Corriveau H. Screening older adults at risk of falling with the Tinetti balance scale.  Lancet. 2000;356(9234):1001-1002PubMedArticle
36.
Kloos AD, Kegelmeyer DA, Young GS, Kostyk SK. Fall risk assessment using the Tinetti mobility test in individuals with Huntington's disease.  Mov Disord. 2010;25(16):2838-2844PubMedArticle
37.
Blahak C, Baezner H, Pantoni L,  et al; LADIS Study Group.  Deep frontal and periventricular age related white matter changes but not basal ganglia and infratentorial hyperintensities are associated with falls: cross sectional results from the LADIS study.  J Neurol Neurosurg Psychiatry. 2009;80(6):608-613PubMedArticle
38.
Taguchi N, Higaki Y, Inoue S, Kimura H, Tanaka K. Effects of a 12-month multicomponent exercise program on physical performance, daily physical activity, and quality of life in very elderly people with minor disabilities: an intervention study.  J Epidemiol. 2010;20(1):21-29PubMedArticle
×