Parkinson's disease (PD) is characterized by severe motor symptoms, including upper limb dysfunction, that is only partially alleviated by medication. PD is also a motor learning disease due to the degradation of the striatum, involved in the consolidation of motor memory. We showed earlier that motor practice improves writing deficits and that there is long term potential when it is applied in a focused manner. However, retention difficulties were also apparent. What is currently unclear, is which learning method leads to optimal retention in PD and how it is expressed in underlying neural network changes. In healthy controls, retention is improved by incorporating dual task (DT) conditions or by loading cognition during learning. Our own work showed that DT training also led to better retention than single task (ST) learning, at least in a subgroup of PD. Using a combination of behavioral assessment, functional magnetic resonance imaging and upper limb task training, this project aims to understand how to boost the robustness of practice in PD. Throughout, we will contrast ST with DT learning. As complex practice can now easily be delivered via novel technology, this study will set out future avenues for rehabilitation targeted at specific neural circuitry.
For this study, 40 healthy elderly and 40 patients with Parkinson's disease (PD) will be included. Sample size was calculated combining results of a pilot study using the Swipe-Slide Pattern (SSP) task (Nackaerts et al. 2020, Behav Brain Res) and literature on complex task training (Lin et al. 2012, Neuroimage; Sidaway et al. 2016, J Mot Behav). The pilot study showed that movement time at retention, after ST training, was on average 2.824s ± 1.015 for sequence swiping. Based on the literature, we assumed a 13.5% difference between DT and ST training at retention. Importantly, we hypothesized a similar benefit from DT-training in PD patients and HC. Using a β = 0.20 and α = 0.05 sample size was calculated for a repeated measured ANOVA analysis with a within-between factor interaction design (within: pre vs post vs retention; between: ST vs DT). Total sample size was estimated at 62 participants, divided in 31 in ST-training and 31 in DT-training. Taking into account a 30% dropout (either from the study or due to data loss), this resulted in 40 subjects in ST-training and 40 in DT-training. PD patients and HC will be divided equally across training types, resulting in 40 PD patients and 40 HC to be randomized to a ST or DT training arm. Patients will be tested during the OFF phase of the medication cycle, i.e. approximately 12 h after last medication intake. Participants will first undergo an inclusion session at home. During this session, they will undergo an extensive behavioral test battery, assessing cognitive and motor skills. Patients will start this session in the OFF phase of the medication phase and will therefore be asked to postpone their morning medication. First, motor skills will be assessed by means of the Movement Disorders Society Unified Parkinson's Disease Rating Scale (MDS-UPDRS) and Clinch token transfer test (C3T). Cognitive assessment includes the Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA) and Trail Making Test (TMT). Additionally, tablet skills will be tested extensively using a newly developed test battery, including performance of tapping between two spots, swiping in a single direction and swiping in multiple directions in random order, in single and dual task conditions. Additionally, a mobile phone task will be performed in which participants have to type in a pre-defined phone number on a smartphone. This part of the assessment will take approximately 60 min. Afterwards patients will be able to take their regular medication and several questionnaires will be filled out. These include the New Freezing Of Gait Questionnaire (NFOG-Q), the non-gait freezing questionnaire, the Dexterity questionnaire (DextQ-24), the Mobile Device Proficiency Questionnaire (MDPQ-16), questions regarding smartphone and tablet use and remaining parts of the MDS-UPDRS. Furthermore, daily levodopa doses will be recorded. Healthy controls will undergo a similar protocol, though disease-specific assessments and questionnaires will not be performed (i.e. MDS-UPDRS, NFOG-Q, non-gait freezing questionnaire and medication intake). Following inclusion, participants will be invited to the radiology dept. UZ Leuven. They will train the SSP-task on an MRI-compatible tablet, while their hemodynamic responses are measured using fMRI. Training will include two runs of 7 min 50 s, either as ST or DT depending on randomization. The SSP-task is based on the finger movements that have to be made to unlock smartphones or tablets or the trajectory that can be used to quickly form words using a keyboard on a smartphone. During this test, participants will have to make different pre-defined patterns. To reduce cognitive load, the pattern will be visible in one of the upper corners of the screen. During the task, participants will be able to see the lines they are drawing. Every pattern will begin in one of nine circles and will consist of equally long movements. Participants will be asked to move the hand without fully lifting the finger stylus from the screen to maintain the online trace. In addition, participants will be instructed to return to a fixed starting point when the pattern is complete. The secondary task consists of counting the number of red or green lights that are illuminated in the peripheral view during task performance. An MRI-compatible version of the touch-sensitive tablet will be used. Participants will see the trace of their pen on the tablet by means of a built-in mirror on top of the head coil. In addition to the task-based fMRI, a high-resolution T1-weighted anatomical scan and diffusion weighted imaging will be performed. Before and after training, ST and DT performance on the SSP-task will be assessed outside the scanner, using a different pattern to avoid learning. Participants will also perform the mobile phone task and tapping between two spots test. Additionally, participants will fill out the Hospital Anxiety and Depression Scale (HADS) and Pittsburgh Sleep Quality Index (PSQI). From day 2 till 5, participants will continue practice of the SSP-task at home. For this, participants will be asked to perform the SSP-task on a tablet each morning, for patients this will be just before taking their regular medication. These training sessions will be limited to 10 minutes and contain the pattern that was learned on day 1, as well as two new patterns to allow variation. The patterns will be offered in a random order, as research has shown that random practice can improve retention and transfer in both healthy elderly adults and patients with PD. On day 5, the researcher will go to the participants' home and perform an immediate retention test, involving the single- and dual task version of the SSP-task. On days 6 and 7 participants will not practice to allow for a retention period. On day 8, all participants will have a (delayed) retention/transfer scan, consisting of two runs in either ST or DT mode: (i) a run containing the learned pattern; and (ii) a run including a new pattern to assess transfer. Again, ST and DT performance on the SSP-task will be assessed outside the scanner, using a different pattern to avoid learning. Participants will also perform the mobile phone task and tapping between two spots test. To minimize head movements during the scan itself, a vacuum fixation pillow to accommodate these difficulties will be used.
Study Type
INTERVENTIONAL
Allocation
RANDOMIZED
Purpose
TREATMENT
Masking
NONE
Enrollment
80
Participants will practice the Swipe Slide Pattern (SSP) task in combination with a secondary task, i.e. counting red or green lights. On day 1, participants perform 2 runs of the SSP-task (only one pattern), each lasting approx. 10 min, within an MR scanner. During each run, 9 trials of 30s are performed, alternated with rest periods of 14s. Instructions are provided before each trial (i.e. view of the pattern in combination with instruction to count red or green lights, 4s), as well as an answer option (i.e. participants have to indicate how many red or green lights they counted by circling a number from zero to nine, 6s). For the following 4 days, participants will continue practice of the SSP-task at home. During the at-home-sessions, participants will perform 9 trials of 12 patterns each, alternated with rest periods of 14s. Instruction and answers are also included. During at home practice, the same pattern as during scans is included, in combination with 2 new patterns.
Participants will practice the Swipe Slide Pattern (SSP) task without the presence of a secondary task. On day 1, participants perform two runs of the SSP-task (only one pattern), each lasting approx. 10 min, within an MR scanner. During each run, nine trials of 30s are performed, alternated with rest periods of 14s. In addition, instructions are provided before each trial (i.e. view of the pattern, 4s), as well as an answer option (i.e. participants have circle a random number from zero to nine, 6s). For the following four days (day 2-5), participants will continue practice of the SSP-task at home. During the at-home-sessions, participants will perform nine trials of 12 patterns each, alternated with rest periods of 14s. Instruction are also included. During at home practice, the same pattern as during scans is included, in combination with two new patterns.
Department of Rehabilitation Sciences KU Leuven
Leuven, Belgium
Change in movement time (s) of trained pattern
Using the behavioral data gathered during task-based fMRI, the learning index and retention index, as described in Nackaerts et al. 2020, will be determined and compared between training types (ST vs DT) and groups (PD vs HC).
Time frame: 7 days
Change in dual task effect
Using the behavioral data gathered before and immediately after task-based fMRI, as well as at immediate and delayed retention, dual task interference will be calculated and compared between training types (ST vs DT), groups (PD vs HC) and time points.
Time frame: 7 days
Change in brain activity during performance of trained pattern
The BOLD activity pattern will be determined and compared between the 3 training phases (i.e. early learning, late learning and retention), between training types (ST vs DT) and groups (PD vs HC).
Time frame: 7 days
Change in brain connectivity during performance of trained pattern
The BOLD activity pattern will be determined and connectivity measures will be extracted. Neural network changes will be compared between the 3 training phases (i.e. early learning, late learning and retention), between training types (ST vs DT) and groups (PD vs HC).
Time frame: 7 days
Diffusion weighted imaging as a predictor
Anatomical connectivity at baseline will be calculated and investigated as a predictive factor for learning capacity.
Time frame: 7 days
Change in Euclidean distance of trained pattern
Using the behavioral data gathered during task-based fMRI, the learning index and retention index, as described in Nackaerts et al. 2020, will be determined and compared between training types (ST vs DT) and groups (PD vs HC).
Time frame: 7 days
Change in movement time (s) of the untrained pattern
Using the behavioral data gathered during task-based fMRI, the transfer index, as described in Nackaerts et al. 2020, will be determined and compared between training types (ST vs DT) and groups (PD vs HC).
Time frame: 7 days
Change in Euclidean distance of the untrained pattern
Using the behavioral data gathered during task-based fMRI, the transfer index, as described in Nackaerts et al. 2020, will be determined and compared between training types (ST vs DT) and groups (PD vs HC).
Time frame: 7 days
Change in brain activity during performance of untrained pattern
The BOLD activity pattern will be determined and compared between the training phases, training types (ST vs DT) and groups (PD vs HC).
Time frame: 7 days
Change in brain connectivity during performance of untrained pattern
The BOLD activity pattern will be determined and connectivity measures will be extracted. Neural network changes will be compared between the training phases, between training types (ST vs DT) and groups (PD vs HC).
Time frame: 7 days
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.