AAIC Press Release
New Research Indicates Gait Changes Could Signal Increased Risk for Cognitive Impairment
VANCOUVER, July 15, 2012 – Gait disturbances – such as a slowing of walking pace or a more variable stride – could indicate a decline in cognitive function, according to new research studies reported today at the Alzheimer's Association's International Conference® 2012 (AAIC® 2012).
"With an aging baby boomer generation advancing into greater risk for Alzheimer's and dementia, it is important for physicians to be aware of the associations between gait and mental function. These studies suggest that observing and measuring gait changes could be a valuable tool for signaling the need for further cognitive evaluation," said William Thies, PhD, Alzheimer's Association® Chief Medical and Scientific Officer.
"For busy doctors who have limited time with their patients, monitoring deterioration and other changes in a person's gait is ideal because it doesn't require any expensive technology or take a lot of time to assess. It is relatively simple and straightforward," Thies added.
Gait Analysis Shows that Stride Speed and Variability May Track with Cognitive Impairment
Difficulties with walking are not inevitable consequences of aging. They are, however, common and relevant problems among older adults. Research shows that people with walking difficulties not only have an increased risk of falling, but may also have an increased risk developing memory disorders and dementia.
Stephanie A. Bridenbaugh, MD, of the Basel Mobility Center in Basel, Switzerland, and colleagues used quantitative gait analysis to explore this issue. The study followed 1,153 participants (average age=77) including outpatients from the Basel Memory Clinic and Basel Mobility Center, plus cognitively healthy participants in a Basel cohort study, from 2007 to 2011.
Participants were divided into groups based on their cognitive diagnoses: cognitively healthy, mild cognitive impairment (MCI) or Alzheimer's dementia. Those with Alzheimer's dementia were subdivided into mild, moderate or severe. Gait was measured using a 10-meter-long electronic walkway with almost 30,000 integrated pressure sensors. All participants performed one "normal" walk and two different "dual tasks" – normal walking while simultaneously counting backwards out loud or while simultaneously naming animals.
The scientists found that gait became slower and more variable as cognitive decline progressed. For all groups, walking speeds were slower during dual tasking than during normal walking alone. "Those with Alzheimer's dementia walked slower than those with MCI, who in turn walked slower than those who were cognitively healthy," said Bridenbaugh.
"Mobility impairments are often associated with dementia, and some gait changes may even appear before cognitive decline can be detected by traditional testing methods. Gait analysis can simply, quickly and objectively measure walking. When problems emerge, this may provide early detection of fall risk and the earliest stages of cognitive impairment in older adults," Bridenbaugh added. "A gait analysis will not replace a comprehensive neuropsychological assessment to diagnose a patient's cognitive status. Gait analysis, however, may prove to be an important tool to aid diagnosis, and record treatment effects or disease progression."
Specific Aspects of Gait may be Associated with Specific Cognitive Abilities and Functions
With aging and in people with Alzheimer's disease, various brain functions deteriorate. Most research has focused on cognition. Recent evidence suggests that gait is also affected by aging and Alzheimer's, yet the exact relationship remains unclear.
Mohammad Ikram, MD, PhD, and colleagues at Erasmus MC, Rotterdam, the Netherlands investigated the relationship between cognition and gait in community-dwelling elderly. The researchers studied 1,232 individuals age 49 and older from The Rotterdam Study (Note: data included here is updated since the original abstract submission to AAIC 2012). Standardized neuropsychological tests were used to measure information processing speed, memory, fine motor speed, and executive function. Gait was assessed using an electronic walkway.
Each participant performed a normal walk, a tandem walk (where the heel of your front foot is placed directly touching the toes of your back foot), and a turn. Gait variables were grouped into seven independent factors:
- Rhythm (reflecting stride time and cadence)
- Pace (reflecting stride length and velocity)
- Phases (reflecting the amount of time spent on one or both feet)
- Variability (reflecting the variation in gait within persons)
- Base of Support (reflecting step width and stride width)
- Tandem (the amount of errors in a tandem walk)
- Turn (the amount of time and steps needed to turn around)
Interesting patterns emerged in the data analysis; the researchers found that certain cognitive domains were only associated with certain aspects of gait.
- Information processing speed was associated with the Rhythm aspect of gait.
- Executive function was associated with Pace and Variability.
- Fine motor speed was associated with Tandem.
- Memory was not associated with any aspect of gait.
"Our results suggest that cognition and gait are tightly linked according to a specified pattern, in which certain cognitive domains only associate with corresponding aspects of gait," Ikram said.
Reduced Gait Velocity, Cadence, and Stride Length may be Associated with Cognitive Decline
Some previous studies have reported that gait abnormalities may be associated with cognitive impairment and dementing illnesses. However, it is unclear which gait components may be associated with a future cognitive decline.
Rodolfo Savica, MD, MSc, and colleagues at the Mayo Clinic Study of Aging (MCSA) measured the stride length, cadence and velocity of more than 1,341 study participants through a computerized gait instrument (GAITRite) at two or more visits roughly 15 months apart. The visits also included neurological and neuropsychological evaluations covering four domains: memory, executive functioning, language, and visuospatial ability. Participants were either cognitively normal (1,172), or diagnosed with MCI (158) or dementia (11).
The researchers found that study participants with lower cadence, velocity and amplitude of the stride length experienced significantly larger declines in global cognition, memory and executive function.
"We observed an association between reduced gait velocity, cadence and stride length, and both global and domain-specific cognitive decline in our population," said Savica. "These results support a possible role of gait changes as an early predictor of cognitive impairment."
Continuous In-Home Monitoring may be a More Accurate Measure of Gait than Single Tests
Traditionally, walking speed has been collected at a single, intermittent time point, such as during a yearly physical exam.
"Advanced technology now allows us to measure walking speed in one's own home, derived from hundreds of walking episodes, and using information collected continuously by motion sensors," said Lisa Silbert, MD, MCR, of Oregon Health & Science University, Portland. "This potentially provides a better measure that links real-world walking abilities and brain health."
Silbert and colleagues worked with 19 dementia-free volunteers (mean MMSE 28.7) enrolled in the Intelligent Systems for Assessment of Aging Changes (ISAAC) study. All participants underwent brain MRI to measure the volume of the total brain and various brain sections. Gait speed was determined in two ways: (1) at the time of MRI, by assessing the time to walk nine meters, and (2) by using an in-home assessment system that continuously collected data over a one month period using motion activity sensors.
The researchers found that:
- Study participants walked faster when measured once in person than when walking in their home under conditions of continuous assessment.
- Slower walking speed determined with continuous in-home assessment technology was associated with smaller total brain size, while single walking speed measures were not.
- Slower in-home walking speed was more highly associated with smaller volumes of the hippocampus (a section of the brain important for memory) than walking speed obtained during a single time point.
"Walking speed taken at a single time point may over-estimate walking abilities in the elderly. Our data suggests that continuous in-home monitoring may provide a more accurate reflection of walking speed and may be more sensitive at detecting motor changes associated with future cognitive decline," Silbert said.
Gait Changes Correlate with Dementia Symptoms in an "Old-Old" Population
The Kurihara Project, conducted by Kenichi Meguro and colleagues at the Tohoku University Graduate School of Medicine, Sendai, Japan, examined the relationship between gait and cognition in 525 community dwelling persons age 75 and older in Kurihara and Osaki, Japan.
Researchers gathered participants' demographics, medical history, general medical and neurological examination results, MRI results, and neuropsychological exams including the Mini-Mental State Examination (MMSE) and the Clinical Dementia Rating (CDR). Participants walked six meters at their fastest pace. Gate measures included gait pattern, velocity and stride length.
The researchers found that 385 study participants had a normal gait pattern, 65 had "neurological gait," and 73 had abnormal gait due to bone and joint disease (such as osteoarthritis). On the CDR scale: 175 participants were classified CDR 0, 287 as CDR 0.5, 44 as CDR 1, 20 as CDR 2, and 2 as CDR 3. (CDR 0 is considered normal, CDR .5 = very mild dementia, with dementia severity increasing to CDR 3 = severe dementia.) They also found that MRI-measured atrophy of the entorhinal cortex – a section of the brain that functions as a hub in a widespread network for memory and navigation – was significantly correlated with gait velocity.
"Our research found that gait velocity was significantly decreased as the severity of dementia symptoms increased," said Meguro. "Gait should no longer be considered a simple, automatic, motor activity that is independent of cognition. They are linked."
The Alzheimer's Association International Conference® (AAIC) is the world's largest conference of its kind, bringing together researchers from around the world to report and discuss groundbreaking research and information on the cause, diagnosis, treatment and prevention of Alzheimer's disease and related disorders. As a part of the Alzheimer's Association's research program, AAIC serves as a catalyst for generating new knowledge about dementia and fostering a vital, collegial research community.
About the Alzheimer's Association®
The Alzheimer's Association is the world's leading voluntary health organization in Alzheimer's care, support and research. Our mission is to eliminate Alzheimer's through the advancement of research, to provide and enhance care and support for all affected, and to reduce the risk of dementia through the promotion of brain health. Our vision is a world without Alzheimer's. Visit www.alz.org or call 800-272-3900.
- Stephanie Bridenbaugh, et al. How does gait change as cognitive decline progresses in the elderly?
- Mohammad Ikram, et al. Cognition and gait reveal distinct patterns of association in an aging population. (Funders: Erasmus MC University Medical Center Rotterdam, Prinses Beatrix Fonds)
- Rodolfo Savica, et al. Slow gait predicts cognitive decline: a population based cohort study. (Funders: National Institutes of Health, Robert H. and Clarice Smith and Abigail Van Buren Alzheimer's Disease Research Program of the Mayo Foundation)
- Lisa Silbert, et al. In Home Continuous Monitoring of Gait Speed: a sensitive method for detecting motor slowing associated with smaller brain volumes and dementia risk. (Funders: National Institutes of Health, Department of Veterans Affairs, Max Millis Fund for Neurological Research, Storms Family Fund at the Oregon Community Foundation, T&J Meyer Family Foundation, Intel Corporation).
- Kenichi Meguro, et al. Clinical gait assessment in the old-old population in a community: The Kurihara Project.
P1-073 Sunday, July 15 / Poster session
How does gait change as cognitive decline progresses in the elderly?
Stephanie Bridenbaugh1, Andreas U. Monsch2, Reto W. Kressig1
1Acute Geriatric Dept., University Hospital Basel, Basel, Switzerland; 2Memory Clinic, Basel, Switzerland
Presenting author e-mail: email@example.com
Background: Mobility impairments are often associated with dementia. Some gait changes may even appear before cognitive decline can be detected by neuropsychological assessment. Gait analysis may be able to aid diagnosis of cognitive impairment. However, it is not yet well characterized how gait changes as cognitive decline progresses.
Methods: The 1,153 older adults (mean age 78.05±5.92 years, age range 65.02-97.43 years, 54.81% females) in this retrospective study were outpatients from our Memory Clinic or participants in the Project BASEL (Basel Study on the Elderly) between 2007 and 2011. Participants had healthy cognition (HC, n=357), mild cognitive impairment (MCI, n=309) or mild (n=219), moderate (n=230) or severe (n=38) Alzheimer's dementia (AD). Spatio-temporal gait parameters were quantified with the GAITRite® electronic walkway at the Basel Mobility Center. Gait was tested during normal walking (self-selected speed) as a single task and during a working memory and a semantic memory dual task. Cognitive diagnoses were based on neuropsychological test battery results.
Results: Gait speed slowed as cognitive impairment progressed. Compared to single task normal walking, gait speed reduction was greater for the semantic memory than for the working memory dual task in those with HC, MCI and mild AD. Gait cycle time variability, a marker of gait regularity, stayed relatively stable during the normal walking single task for those with HC, MCI and mild AD, and yet increased during dual-tasking with the variability highest during the semantic memory dual task. This change was greater in those with MCI and mild AD than in those with HC. Age-stratified results showed that these changes in velocity and variability were not simply age-related.
Conclusions: Cognitive impairment is characterized not only by memory loss, but also by functional impairment. Using quantitative gait analysis with dual task paradigms, we show that gait becomes slower and more variable as cognitive decline progresses. Particularly, gait cycle time variability increases during cognitive dual tasking compared to normal walking in those with MCI and mild AD. Early detection of these mobility impairments may be used as a tool to aid diagnosis of those in the earliest stages of cognitive impairment.
O1-08-02 Sunday, July 15 / Oral session, 3:30-5 pm
Cognition and gait reveal distinct patterns of association in an aging population.
Mohammad Ikram, Vincentius Verlinden, Albert Hofman, Jos van der Geest
Erasmus MC, Rotterdam, Netherlands
Presenting author e-mail: firstname.lastname@example.org
Background: With aging and in Alzheimer's disease, various brain functions deteriorate. Most research has focused on cognition, which includes processing speed, memory and executive function. Recent evidence suggests that gait is also affected by aging and AD. Gait, too, consists of various aspects, such as Time (e.g. swing time), Space (e.g. stride length), and Variability (consistency of walking). It is unclear if and how cognitive impairment associates with gait deterioration. We investigated the relation between cognition and gait in community-dwelling elderly.
Methods: From the Rotterdam Study, 1,465 elderly (49 years and older) underwent cognitive testing and assessment of gait. Cognition was tested using a validated test-battery, measuring processing speed, memory, and executive function. Gait was assessed using an electronic walkway. Each person performed 6 regular walks, which yielded a total of 21 variables. PCA was used to group these variables together into three factors: Time (e.g. stride length and velocity), Space (e.g. stride length and width), and Variability (variance of the variables across 6 walks). We used age/sex-adjusted linear regression to relate cognition with gait. Additionally, in multivariate models we adjusted cognitive domains for each other.
Results: In univariate analyses we found that all cognitive domains were associated with various aspects of gait. However, in multi-variate analyses interesting patterns emerged: certain cognitive domains were only association with certain aspects of gait, whilst the association with other aspects of gait strongly attenuated. Processing speed was associated with Time-aspect of gait (difference in Z-score of Time per SD decrease in cognition -0.22 (-0.34;-0.10), but not Space or Variability. Executive function was associated with Variability (-0.17 (-0.29;-0.04)) and Space (-0.15 (-0.26;-0.03)), but not Time. Memory was not associated with any gait-aspect. The magnitude of effect of these associations correspond to a 10-year deterioration in gait.
Conclusions: Cognition and gait are tightly linked with processing speed associating with Time, and executive function associated with Space and Variability. These data provide novel insights into the full clinical spectrum of brain pathology. Future studies should investigate the role of gait in aging, MCI, and Alzheimer's disease.
P2-169 Monday, July 16 / Poster session
Slow gait predicts cognitive decline: a population based cohort study
Michelle Mielke, Rodolfo Savica, Dina Drubach, Teresa Christianson, Rosebud Roberts, David Knopman, Vernon Pankratz, Boeve Bradley, Walter Rocca, Ronald Petersen
Mayo Clinic, Rochester, Minnesota, United States
Presenting author e-mail: email@example.com
Background: Gait may be susceptible to age-related changes that depend on the individual characteristics and on the concurrent presence of other diseases. Some studies have reported that gait abnormalities may be associated with cognitive impairment and dementing illnesses, supporting the role of an association between gait and cognitive impairment. However, it is unclear which gait components have been associated with a future cognitive decline.
Methods: Participants included 1,341 subjects who were enrolled in the Mayo Clinic Study of Aging (MCSA) and were examined with a computerized gait instrument, GAITRite, at two or more visits. Visits, including neurological and neuropsychological evaluations, were conducted approximately 15 months apart. The primary gait assessments were three spatio-temporal parameters: stride length, cadence, and velocity. Outcomes included the Short Test of Mental Status (STMS) and global and domain-specific Z-scores. Domain-specific Z-scores were calculated using nine neuropsychological tests covering four domains: memory, attention/executive functioning, language, and visuospatial ability. Linear mixed effects models were used to assess differences in the annualized rate of change in cognitive domain scores associated with differences in the gait measures at baseline while adjusting for age, sex, education, APOE genotype, hypertension, diabetes and dyslipidemia.
Results: Participants with lower cadence, velocity, and amplitude of the stride length experienced significantly larger declines in global cognition, memory, and executive function. Lower gait measures were also significantly associated with a greater decline on the STMS. The average decrease in the STMS associated with a 10 steps per minute lower cadence was 0.07 points per year (95% CI: 0.02 - 0.12, p=0.01). The average decrease in the STMS associated with a 1 meter per second slower velocity was 0.5 points per year (95% CI: 0.2 - 0.8, p=0.002). Excluding persons with stroke, parkinsonism, vascular dementia, head injury, or normal pressure hydrocephalus from the analyses did not affect the results.
Conclusions: Our study is consistent with previous reports. In particular, we observed an association between reduced gait velocity, cadence, and stride length and global- and domain-specific cognitive decline in a large population based cohort
P2-022 Monday, July 16 / Poster session
In Home Continuous Monitoring of Gait Speed: a sensitive method for detecting motor slowing associated with smaller brain volumes and dementia risk.
Lisa Silbert, Hiroko Dodge, David Lhana, Louie Perkins, Tamara Hayes, Nora Matteck, Daniel Austin, Brittany Stone, Jeffrey Kaye
Oregon Health & Science University, Portland, Oregon, United States
Presenting author e-mail: firstname.lastname@example.org
Background: Slower walking is common with advanced age, and is associated with loss of independence and increased dementia risk. Single gait speed (SGS) measures are typically obtained only at an annual visit. Continuous in-home monitoring of gait speed (CIHM-GS) derived from hundreds of walking episodes per person may provide a better measure linking real-world motor abilities and brain integrity.
Methods: 19 dementia-free volunteers (mean MMSE 28.7) enrolled in the Intelligent Systems for Assessing Aging Change study underwent 3T brain MRI. FreeSurfer was used to determine brain, ventricular, and hippocampal volumes. Gait speed (GS) was determined: 1) at the time of the MRI, assessing the number of seconds to walk 9 meters with a stopwatch, and 2) using an in-home assessment system that continuously collected data using passive infra-red motion-activity sensors averaged over a one month period aligned with the volunteer's MRI acquisition. Regression analyses examining MRI volumetric measures adjusted for ICV were performed with the outcome of GS.
Results: Mean SGS was 81.7 cm/sec, while CIHM-GS was 76.4 cm/sec (correlation: R2 = 0.27, p = 0.02). Slower CIHM-GS was associated with decreased total brain (R2 = 0.29, p = 0.02), increased ventricular (R2 = 0.28, p = 0.02), and decreased hippocampal (R2 = 0.44, p = 0.002) volumes. SGS measures were not associated with either brain or ventricular volumes, but were associated with hippocampal size (R2 = 0.30, p = 0.02). CIHM-GS was associated with brain (p = 0.02) and CSF (p = 0.02) in two separate step-wise regressions with age, and gender as covariates. In a step-wise regression with hippocampal volume as an outcome, and age, gender, and both SGS, and CIHM-GS as covariates, only CIHM-GS remained related to hippocampal volume (0.002)
Conclusions: SGS measures may over-estimate walking abilities in the elderly. Compared with SGS, CIHM-GS measures were associated with smaller global and regional brain volumes, indicators of increased dementia risk. CIHM may provide a more accurate reflection of GS in dementia-free elderly gauging biologically relevant changes in motor performance associated with cognitive decline.
O1-08-05 Sunday, July 15 / Oral session, 3:30-5 pm
Clinical gait assessment in the old-old population in a community: The Kurihara Project
Naofumi Tanaka, Hiroyasu Ishikawa, Kei Nakamura, Masayuki Satoh, Satoshi Yamaguchi, Kenichi Meguro
Tohoku University Graduate School of Medicine, Sendai, Japan
Presenting author e-mail: email@example.com
Background: Several studies have underscored that gait is no longer considered a simple automatic motor activity that is independent of cognition. Several epidemiological studies focused on a negative relationship between gait dysfunction and the risk of dementia in older people. We investigated the relationship between clinical gait and cognition in a community-based sample of people aged 75 and older.
Methods: Five hundred and twenty five subjects participated among the 1,374 members of the population aged 75 years and older in Kurihara and Osaki, Japan and 164 underwent volumetric magnetic resonance imaging. Our database comprised participant demographics, medical history and current symptoms, general medical examination, neurological examination, MRIs, laboratory tests, electrocardiograms, various neuropsychological tests including the Mini-Mental State Examination (MMSE), and Clinical Dementia Rating (CDR). Gait function was assessed using clinical gait measures including gait pattern, gait velocity, and stride length. Gait patterns were classified as normal, neurological gaits, and abnormal due to bone and joint disease such as osteoarthritis and spodylosis. We asked participants to walk 6 m at the fastest pace to measure the gait velocity and stride length.
Results: We found 385 participants with normal gait pattern, 65 with neurological gait, and 73 with abnormal gait due to bone and joint disease. As for the CDR scale, 175 participants were classified as CDR 0, 287 as CDR 0.5, 44 as CDR 1, 20 as CDR 2, and 2 as CDR3. The degree of atrophy of the entorhinal cortex was significantly correlated with gait velocity.
Conclusions: Gait velocity was significantly decreased as CDR rating increased. A comprehensive approach of motor as well as cognitive assessment is needed for community residents.
Media line: 312.335.4078