Increasing Gait Automaticity in Older Adults by Exploiting Locomotor Adaptation
- Conditions
- Gait AutomaticityLocomotor AdaptabilityCommunity Mobility of Older Adults
- Interventions
- Other: Split-belt walkingOther: Multiple transitions between split-belt and tied-belt walking
- Registration Number
- NCT04934956
- Lead Sponsor
- University of Pittsburgh
- Brief Summary
The investigators will test the following: 1) the extent of locomotor adaptation improvement in individuals aged 65 years and older; 2) the association between initial walking automaticity (i.e. less PFC activity while walking with a cognitive load) and prefrontal-subcortical function (measured via neuropsychological testing); and 3) whether improvements in locomotor adaptability result in improvements in the Functional Gait Assessment (FGA), a clinically relevant indicator of dynamic balance and mobility in older adults. To answer these questions, the investigators will combine innovative techniques from multiple laboratories at the University of Pittsburgh. Automatic motor control (Dr. Rosso's expertise) will be assessed by wireless functional near-infrared spectroscopy (fNIRS) of the PFC during challenged walking conditions (walking on an uneven surface and walking while reciting every other letter of the alphabet). fNIRS allows for real-time assessment of cortical activity while a participant is upright and moving by way of light-based measurements of changes in oxygenated and deoxygenated hemoglobin. Locomotor adaptation (Dr. Torres-Oviedo's expertise) will be evaluated with a split-belt walking protocol (i.e., legs moving at different speeds) that the investigators and others have used to robustly quantify motor adaptation capacity in older individuals and have shown to be reliant on cerebellar and basal ganglia function. The investigators will focus on two important aspects of locomotor adaptation that the investigators have quantified before: (Aim 1) rate at which individuals adapt to the new (split) walking environment and (Aim 2) capacity to transition between distinct walking patterns (i.e., the split-belt and the overground walking patterns), defined as motor switching. Adaptation rate and motor switching are quantified using step length asymmetry, which is the difference between a step length taken with one leg vs. the other. The investigators will focus on this gait parameter because it robustly characterizes gait adaptation evoked by split-belt walking protocols. Finally, the investigators will quantify participant's cognitive function (Dr. Weinstein's expertise) through neuropsychological battery sensitive to prefrontal-subcortical function. The investigators will mainly focus on evaluating 1) learning capacity reliant on cerebellar structures and 2) assessing executive function heavily reliant on PFC and, to a lesser extent, the basal ganglia.
- Detailed Description
Restrictions in community mobility, the ability to move outside of one's home, are common in older ages and contribute to disability, institutionalization, and poor quality of life. Successful community mobility requires rapid integration of information from both external (e.g. surface quality, distances) and internal (e.g. fatigue, pain) to the individual. Under normal conditions, integration of these inputs occurs in subcortical-frontal (e.g., basal ganglia and cerebellum to primary motor cortex) networks and favors automatic motor control with few demands on the attention- related networks that primarily reside in the prefrontal cortex (PFC). As gait automaticity diminishes in older adults6, activation of the PFC during walking tasks increases. Lack of gait automaticity can interfere with community mobility, as the PFC is no longer free to process other information for navigating community environments. Another potential contributor to reduced community mobility is diminished locomotor adaptation. Specifically, older adults are slower at adjusting their movements while interacting with a new environment and have more difficulty switching motor patterns when transitioning across distinct walking conditions. This difficulty in switching motor patterns is related to cognitive switching ability, which is reliant upon similar subcortical-frontal processes that underlie motor control. While locomotor adaptation is reduced in normal aging, data from our lab indicates that older adults maintain plasticity and can improve locomotor adaptation. Our central hypothesis is that the ability to improve locomotor adaptation is greater in those with higher gait automaticity and greater integrity of the prefrontal-subcortical connections.
The extent of gait automaticity can be tested by increasing the cognitive load during walking (e.g., completing a cognitive task while walking) and measuring the related PFC response. Small changes in PFC activity and motor performance in response to the imposed cognitive load indicate intact gait automaticity. Conversely, a large change in PFC activity to maintain motor performance with addition of a cognitive load indicates diminished gait automaticity. Locomotor adaptability can be measured by manipulating walking context on a split-belt treadmill where the legs are moving at different speeds. Adaptation rate to the split-belt environment can be measured as well as the ability to switch motor patterns from the split-belt to overground walking. Promising data from our labs (n=8) indicate that older participants improve locomotor adaptation after experiencing multiple transitions between the split condition (belts' speed ratio 2:1) and regular walking (belts' speed ratio 1:1). However, neither the underlying mechanisms nor the clinical relevance of such improvements are known.
The investigators will test the following: 1) the extent of locomotor adaptation improvement in individuals aged 65 years and older; 2) the association between initial walking automaticity (i.e. less PFC activity while walking with a cognitive load) and prefrontal-subcortical function (measured via neuropsychological testing); and 3) whether improvements in locomotor adaptability result in improvements in the Functional Gait Assessment (FGA), a clinically relevant indicator of dynamic balance and mobility in older adults. To answer these questions, the investigators will combine innovative techniques from multiple laboratories at the University of Pittsburgh. Automatic motor control (Dr. Rosso's expertise) will be assessed by wireless functional near-infrared spectroscopy (fNIRS) of the PFC during challenged walking conditions (walking on an uneven surface and walking while reciting every other letter of the alphabet). fNIRS allows for real-time assessment of cortical activity while a participant is upright and moving by way of light-based measurements of changes in oxygenated and deoxygenated hemoglobin. Locomotor adaptation (Dr. Torres-Oviedo's expertise) will be evaluated with a split-belt walking protocol (i.e., legs moving at different speeds) that the investigators and others have used to robustly quantify motor adaptation capacity in older individuals and have shown to be reliant on cerebellar and basal ganglia function. The investigators will focus on two important aspects of locomotor adaptation that the investigators have quantified before: (Aim 1) rate at which individuals adapt to the new (split) walking environment and (Aim 2) capacity to transition between distinct walking patterns (i.e., the split-belt and the overground walking patterns), defined as motor switching. Adaptation rate and motor switching are quantified using step length asymmetry, which is the difference between a step length taken with one leg vs. the other. The investigators will focus on this gait parameter because it robustly characterizes gait adaptation evoked by split-belt walking protocols. Finally, the investigators will quantify participant's cognitive function (Dr. Weinstein's expertise) through neuropsychological battery sensitive to prefrontal-subcortical function. The investigators will mainly focus on evaluating 1) learning capacity reliant on cerebellar structures and 2) assessing executive function heavily reliant on PFC and, to a lesser extent, the basal ganglia.
With this data, the investigators will be able to address the following Aims:
Aim 1. Determine the association between improved locomotor adaptation rate and 1) individuals' gait automaticity and 2) cognitive function. Hypothesis: changes in adaptation rate will be predicted by initial walking automaticity and cerebellar-mediated learning capacity. This is predicated on the evidence that motor adjustments during split-belt walking depend on basal ganglia and cerebellar function.
Aim 2. Determine the association between improved locomotor switching and individuals' gait automaticity and cognitive function. Hypothesis: initial walking automaticity and executive control will predict improvements in locomotor switching. This is predicated on the evidence that motor switching is directly associated with basal ganglia-dependent cognitive tasks such as set-shifting.
Aim 3. Determine the extent to which improved locomotor adaptability could improve mobility. Hypothesis: changes in locomotor adaptability will not be exclusive to the laboratory context but will generalize to other locomotor tasks that require adaptability, as measured in the Functional Gait Assessment.
These results will provide strong preliminary data for a future study to explore these associations in a larger sample with more comprehensive measures of mobility contributors, neuroimaging for integrity of key brain regions, and objective measures of community mobility. These results will identify novel contributors to loss of community mobility in older adults and could identify novel therapeutic targets for interventions that improve gait adaptation to prevent falls and enhance independence.
Recruitment & Eligibility
- Status
- RECRUITING
- Sex
- All
- Target Recruitment
- 44
- 19 years old or older.
- Body Mass Index of 35 or less. Muscle activities will be recorded for distinct muscles in the legs and fatty tissue could interfere with these measurements.
- Able to walk without a hand held device
- Able to walk for 5 minutes at their self-paced speed
- Any past or present history of neurological disorders, heart or respiratory disease, brain injury, seizures, spinal cord surgery, or strokes.
- Pregnancy.
- Unable to follow two part commands;
- Uncorrected vision or severe visual impairment with visual acuity < 20/70 with best correction;
- Cognitive impairments defined as modified mini-mental score <84;
- orthopedic or pain conditions (lower extremity pain, back pain, calf pain);
- refuse to walk on a treadmill;
- hospitalized 6 months prior to the study for acute illness or surgery, other than minor surgical procedures;
- lower extremity orthopedic surgery within 1 year;
- uncontrolled hypertension (> 190/110 mmHg);
- diagnosed dementia;
- dyspnea at rest or during daily leaving activities;
- use supplemental oxygen, resting heart rate> 100 or <40 beats per minute;
- fixed or fused hip, knee, or ankle joints;
- progressive movement disorder such as Multiple Sclerosis (MS), Amyotrophic Lateral Sclerosis (ALS), or Parkinson's disease
Study & Design
- Study Type
- INTERVENTIONAL
- Study Design
- SINGLE_GROUP
- Arm && Interventions
Group Intervention Description Intervention: Split-belt walking; Multiple transitions between split-belt and tied-belt walking Split-belt walking Split-belt walking will be used in all experiments and consists of a time period during which the legs move at different speeds (0.5 m/s vs. 1 m/s). The investigators select those speeds since the investigators have observed in our preliminary data and published study (Sombric et al. 2017) that older individuals adapted at these speeds exhibit large deficits at motor switching when transitioning to overground walking. This large reference signal will facilitate the detection of a change in motor switching (Aim 2) following the Intervention. This second intervention consists of multiple short adaptation blocks (i.e., 6 blocks of 200 strides each) interleaved with short de-adaptation blocks (i.e., 5 blocks of 200 strides of tied-belt walking each). It was designed based on several studies showing improvements in adaptation rate in young adults with a similar protocol (Malone et al. 2011; Day et al. 2018; Leech et al. 2018). Intervention: Split-belt walking; Multiple transitions between split-belt and tied-belt walking Multiple transitions between split-belt and tied-belt walking Split-belt walking will be used in all experiments and consists of a time period during which the legs move at different speeds (0.5 m/s vs. 1 m/s). The investigators select those speeds since the investigators have observed in our preliminary data and published study (Sombric et al. 2017) that older individuals adapted at these speeds exhibit large deficits at motor switching when transitioning to overground walking. This large reference signal will facilitate the detection of a change in motor switching (Aim 2) following the Intervention. This second intervention consists of multiple short adaptation blocks (i.e., 6 blocks of 200 strides each) interleaved with short de-adaptation blocks (i.e., 5 blocks of 200 strides of tied-belt walking each). It was designed based on several studies showing improvements in adaptation rate in young adults with a similar protocol (Malone et al. 2011; Day et al. 2018; Leech et al. 2018).
- Primary Outcome Measures
Name Time Method Change in Adaptation rate 1 week pre-Intervention, during Intervention and 1 week post-Intervention Adaptation is measured by the mean of the first 32 strides of in adaptation. This metric approximates the rate of learning. Higher adaptation rate means better learning rate. The adaptation rate will be measured during the visits 1 week pre-intervention, during intervention and 1 week post-intervention. The changes through the process, most importantly before and after intervention, will be calculated. Larger the change, better the learning capacity.
Change in Aftereffect 1 week pre-Intervention, during Intervention and 1 week post-Intervention Aftereffect is the difference in error between the last 40 strides of baseline walking and the first 5 strides of walking over ground after adaptation. Higher aftereffect value represents higher transfer of the learning. The aftereffects will be measured during the visits 1 week pre-intervention, during intervention and 1 week post-intervention. The changes through the process, most importantly before and after intervention, will be calculated. Larger the change, better the cognitive switching ability.
Change in FGA 2 weeks pre-Intervention and 1 week post-Intervention Change in Functional gait assessment (FGA) score post-Intervention relative to pre-Intervention. The FGA consists of 10 items: gait on level surface, change in gait speed, gait with horizontal and vertical head turns, gait with 180° pivot turn, stepping over obstacles, gait with narrow base of support, gait with eyes closed, backwards gait and stairs. Scoring of each of these activities is done on a 4-point ordinal scale ranging from 0-3, with 0 indicating severe impairment (cannot perform without assistance, severe gait deviations or imbalance, increased time to perform task), 1 indicating moderate impairment, 2 indicating mild impairment, and 3 indicating normal ambulation (no gait or balance impairment, completion of task in a timely manner). All items are summed to calculate a total score (max. 30).
Executive function 2 weeks pre-Intervention Subtests of Delis-Kaplan Executive Function System (D-KEFS) will be used: (1) Color-Word Interference Task that measures the ability to inhibit a dominant and automatic verbal response (inhibition) and the ability to switch between inhibiting and executing an automatic verbal response (inhibition/switching); (2) Trail Making Test that measures flexibility of thinking and set-shifting ability on a visual-motor sequencing task (Condition 4), and (3) Verbal Fluency (for letters and categories) that measures letter fluency, category fluency, and category switching. The performance of each of test is measured by seconds to completion, except for verbal fluency, which is determined by total number of correct responses and switches between categories. Raw scores are normed using the D-KEFS normative data structure (mean = 10, SD = 3). All tests are co-normed to allow averaging to create a single, composite executive function measure. Higher score means better executive function.
Subcortical/basal ganglia function 2 weeks pre-Intervention he Action verbal fluency test will be used to measure subcortical/basal ganglia cognitive function. This task requires the participant to rapidly generate as many verbs (i.e., "things that people do") as possible within 1 min. The score is the number of correct words within 1 minute, excluding rule-breaks and intrusions (i.e., non-verbs), from 0 to the max number of correct words participant can generate. Higher score means better basal ganglia cognitive function.
- Secondary Outcome Measures
Name Time Method Change in prefrontal cortex activity 2 weeks pre-Intervention and 1 week post-Intervention Gait automaticity will be measured by dual-task related changes in oxygenated (O2Hb) and deoxygenated hemoglobin (HHb) at the prefrontal cortex as detected by functional near-infrared spectroscopy (fNIRS). A wearable, wireless, continuous wave NIRS system with a probe that includes eight bilateral measurements will be used for functional central nervous system measurements during dual task walking. Changes in O2Hb and HHb concentrations from the baseline (quiet standing) to each of the task conditions will be calculated using methods similar to functional magnetic resonance imaging. A coefficient from canonical general linear model will be used to estimate the combined hemodynamic response of O2Hb and HHb changes. This coefficient will be the quantification of changes in brain activity of each task compared to quiet standing and will be averaged for the repeated (4-6) trials of each condition. The changes in hemodynamic response before and after intervention will be calculated.
Attention, language, immediate memory, delayed memory, and visuospatial function measures 2 weeks pre-Intervention Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) is an integrated, co-normed tool to measure cognitive performance across measures of attention, language, memory (immediate and delayed), and visuospatial ability. Each test is scored based on number correct in accordance with the test manual. Individual test scores are normed, then combined into predetermined index scores. Attention index is comprised of the digit span and coding subtests; language is comprised of the picture naming and semantic fluency subtests; immediate memory is comprised of the list and story learning subtests; delayed memory is comprised of the list recall, list recognition, story recall, and figure recall subtests; visuospatial construction is comprised of the figure copy and line orientation subtests. These individual index scores can be combined to generate an overall index standard. The score range is 40-154 (mean = 100 and SD = 15). Higher score means better cognitive performance.
Premorbid estimated verbal ability 2 weeks pre-Intervention The Wechsler Test of Adult Reading (WTAR) estimates premorbid verbal ability/crystallized intellectual function. This is a measure that shows relative stability even in the face of cerebral trauma or insult, or neurodegenerative decline. As such, this provides an estimate of a participant's projected level of function for which to compare other measures of cognitive ability. Raw scores are converted to an estimated IQ/verbal ability score. The score ranges from 50 to 126 (mean = 100, SD = 15). Higher score means better verbal ability.
Switching Ability 2 weeks pre-Intervention The computer-based task inspired by Wisconsin Card Sorting Test will be used to quantify cognitive perseveration. There are four electronic cards displayed with a specific count of colored shapes on a monitor. Participants are showed a reference card and they must find the matching rule (count, shape, or color) by switching their matching strategy based on the feedback on each trial (e.g., if they match the card to the reference by color and they get "incorrect", in the next trial they can try to match by shape). Subjects perform a total a 128 matching trials, with 5 s to respond. Rule changes after 3-5 consecutive correct matches. The score is measured by perseveration error, computed as the total number of matches that were made based on a previous matching rule. The score range is 0-128. Higher score means poorer switching.
Trial Locations
- Locations (1)
Sensorimotor Learning Laboratory, Schenley Place Suite 110
🇺🇸Pittsburgh, Pennsylvania, United States