Study design overview
This study is designed as a randomized, single-blind, single-center trial and will be conducted at the local rehabilitation hospital. This clinical trial was approved by the Ethics Committee of Ningbo College of Health Sciences (LLSC2023017) and registered in the China Clinical Trial Registry (ChiCTR2300075048). The design of this interventional study follows the Standard Protocol Items: Recommended for Interventional Trials (SPIRIT) 2013 statement [19]. Participants will be randomly allocated into two intervention groups and one waitlist control group for 12 weeks. An informed consent will be signed prior to joining the study. Assessments will be conducted at baseline, after 12 weeks of intervention, and after a 12-week follow-up and will include gait parameters, changes in the HbO2 concentration in the PFC, and other motor function parameters. The schedule of observations is illustrated in Table 1.
Participants
Recruitment strategy
We will enroll participants from the Neurology Department of the local rehabilitation hospital and local communities by means of printed leaflets, online PD patient groups, and popular science lectures by the hospital. To minimize potential expectation bias and confirm eligibility, a research assistant will contact those referred by a clinician and follow up with interested participants by phone. Participants will be informed that they will be randomly allocated to two experimental groups (treadmill training and Baduanjin) and a waitlist control group. Figure 1 illustrates the recruitment strategy, intervention procedure, and time points of the three assessments.
Inclusion criteria
(1) Age 60–75 years old; (2) diagnosed with idiopathic PD according to the UK Brain Bank criteria[20]; (3) Hoehn and Yahr (H&Y) stage I–III; (4) Mini-Mental State Examination score ≥24; and (5) ability to walk at least 5 min independently (walking aids are permitted).
Exclusion criteria
(1) Unstable medical conditions, including cardiopulmonary disease, cancer, or orthopedic problems affecting performance; (2) severe cognitive, visual, or auditory impairment; (3) any co-morbidity of the motor system that may restrict the gait; and (4) other ongoing forms of intensive training.
Sample size
In the absence of preliminary data on a meaningful change in prefrontal activation for DT walking, the effect size will be calculated by the gait speed. A clinically important difference in the gait speed among people with PD under the on-medication condition ranged from 0.05 m/s to 0.22 m/s by a distribution-based analysis [21]. We set a statistical power of 0.8 and two-tailed α level of 0.05 in repeated-measures multivariate analysis of variance (MANOVA). Based on the data of a previous study, we assumed that there would be a difference of 15% between the Treadmill training group and Baduanjin group in favor of DT gait[22]. Considering a loss to follow-up, with an estimated drop-out rate of 5%, we will recruit 48 participants per group (a total of 144 patients).
Randomization, allocation, and blinding
Eligible participants will be randomized to one of the groups with an allocation of 1:1:1 ratio through an independent statistician-implemented, computerized block randomization. The treating physiotherapist will be notified of the results of the group allocation by e-mail to ensure concealed allocation. All study assessors who collect outcome measures will be blinded to the study hypotheses and group assignments. In addition, participants will be instructed not to reveal their training regime at any time to prevent unbinding. Both therapists and participants will be informed that both training methods may effectively improve DT gait performance to control for expectancy effects.
Study interventions
Participants in the two intervention groups will complete an exercise program during the “on” medication phase that will be supervised by a physiotherapist. The program will consist of a 5-min warm-up, 30-min treadmill training or Baduanjin, and 10-min cool-down. The training period will encompass 12 consecutive weeks with a total of 36 sessions, including three weekly sessions lasing 45 min each. The intensity will be monitored by measuring the heart rate (Polar Team2; Polar Electro Finland) and perceived exertion (PRE). The calculation of the heart rate will adopt 60%–70% of the HRmax, determined by the formula HRmax = 208 − (0.7 × age) [23] or reaching a fatigue limit at the RPE of 12 [24]. Attendance will be recorded for each training session. An assistant will inquire about the reason of absence by telephone for those participants with low attendance, and encourage them to persist in exercising. Moreover, participants will write an exercise log throughout the study, the content will involve in the self-report training effects, training duration per session and fall events.
a) Treadmill training
Participants will perform gait training on a treadmill (GaitTrainer System 2-Biodex Medical Systems, NY, USA). A safety jacket will be attached to an overhead suspension system, without providing weight support.
At the first training session, participants will be familiarized with the equipment and will receive a gait speed measurement over-ground by 10 m walking. With the progression of training, treadmill speed will be elevated, and the duration of intervals of training will be reduced. During the first 2 weeks of the program, the treadmill speed will be set at 80% of the usual gait speed. The 30-min training sessions will be conducted in three sessions with 5-min breaks at the intervals. During the third and fourth weeks, the treadmill speed will be increased to 90% of the usual speed, and the training session will be completed in two 15-min sessions with a 5-min break between the sessions. During the latter 8 weeks, the treadmill speed will be increased to its usual speed and the training will be continuous for 30 min. Participants will conduct 5-min active stretching exercises to warm up. Ten min of relaxation exercise will be performed on the ground at the end of treadmill training (S1 Table).
The participants will be encouraged to maintain a postural alignment and use the frontal support of the treadmill to minimize physical exertion[25]. Cognitive components of the DT will be integrated into the gait training after 4 weeks of training, including 1) walking while phoning; 2) walking while performing a backward digit span; and 3) walking while carrying a bag filled with food.
b) Baduanjin exercise
Participants will be instructed to perform Baduanjin, whose characteristics include symmetrical physical postures and dynamic movements, postural control, meditation, and rhythmic breathing regulation [26]. It consists of eight different movements, and each movement is repeated six times (S2 Table). During the initial 2–4 weeks, participants will learn to adjust their breathing and coordinate it with the prescribed movements. The whole set of Baduanjin Qigong movements usually takes 12–15 min to complete at a regular pace[18]. During the later 5–12 weeks, a concurrent cognitive task will be added to the exercise. The purpose is to strengthen continuous attentional processing demands while performing another task under varying conditions. Participants will practice Baduanjin while performing the following tasks: 1) stating the form of each movement; 2) counting a 4-digit number backwards; and 3) reciting Chinese poems.
c) The waitlist control group will continue to maintain routine care and lifestyle and will not perform any form of intensive training. A research assistant will follow up their health status twice a month by phone or WeChat. They will participate in either the treadmill training or Baduanjin exercise after the 12-week follow-up assessment.
Outcome measures
All assessments will be performed in the practical, self-reported on-medication status, at about the same time of day. Participants will be required to walk under four conditions on a 4.6 m electronic walkway, while wearing a portable fNIRS device. The walking conditions will be: 1) walking at a usual speed; 2) walking while conducting a backward digital span; 3) walking while performing a serial-3 subtraction; and 4) walking while negotiating two obstacles positioned on the walkway at specific locations. The obstacles will be set up at a half knee heightÍ60 cm and width Í3 cm depth [27].
To avoid a learning effect, the order of each task will be changed randomly. Participants will walk three consecutive loops on the walkway to obtain valid trials and to minimize within-individual variability, and the interval of each task will include 1-min rest [28]. All experimental data will be collected by the same trained assessor to ensure reproducibility.
Primary Outcomes measures
a) Change of prefrontal activity
Prefrontal activity will be measured by monitoring the change in HbO2 concentration of the anterior PFC using a multi-channel, continuous wave, fNIRS instrument (NIRScout; NIRx Medical Technologies LLC; Minneapolis, MN, USA) during task performance. The space between the emitter and detector will be maintained at 3 cm. HbO2 concentration is sensitive to walking-related changes in cortical activity and has been used as a primary outcome to measure prefrontal activity [29].
The concentration of HbO2 will be exported to MATLAB (MathWorks; Natick, MA, United States) for further data analysis. A bandpass filter with a cutoff frequency of 0.01 Hz will be used to remove low-frequency noise, such as head movements, and a low-pass filter with a cutoff frequency of 0.15 Hz will be utilized to reduce high-frequency noise and cardiovascular artifacts [30]. The concentration changes of HbO2 in the targeted PFC will be calculated from the changes in detected light intensity according to the modified Beer-Lambert Law [31]. For each experimental condition, the average concentration of HbO2 during task performance and 10 s baseline (i.e., the value during the 10 s quiet standing prior to the task) will be used to determine relative changes in HbO2 concentration. We will analyze the average of the left and right PFC HbO2 concentrations [4, 5].
b) Gait parameters
We will select the gait speed (m/s), stride length (m), double-phase time (%), stride variability (%), and step width (cm). The gait parameters will be measured using an electronic walkway with pressure sensors embedded in a 4.6 m carpet. Data processing and storing will be performed using the PKMAS software (PKMAS walkway, ProtoKinetics, Havertown, PA, USA) connected to a personal computer. Prior to performing each task, the participants will be required to stand quietly for 20 s and refrain from talking or moving their heads. The instruction “start” will be given after these 20 s.
Secondary outcome measures
a) Disease severity will be determined using the Movement Disorder Society-Unified Parkinson’s Disease Rating Scale (MDS-UPDRS).
The MDS-UPDRS is structured in four parts, namely, Non-motor Experiences of Daily Living (I); Motor Experiences of Daily Living (II); Motor Examination (III); and Motor Complications (IV). Each item is scored on a 5-point scale (0–4 points), with a minimum score of 0 indicating normal and a maximum score of 4 indicating a severe impairment in performing activities. Thus, higher scores reflect worse functioning. The H&Y stage in the MDS-UPDRS includes 5 stages, with the early, middle, and late stages of PD corresponding to stage I-II, stage III-IV, and stage V, respectively [32].
b) The balance function will be assessed by the Mini-Balance Evaluation Systems Test (Mini-BESTest). The scale consists of 14 items from four different balance control systems, including anticipatory postural adjustments, reactive postural control, sensory orientation, and dynamic gait. Each item is assessed from 0 to 2 points, and the total score ranges from 0 to 28, with a higher score representing better balance function [33].
c) The Time-Up and Go Test (TUG) is widely used in clinical research to test balance, mobility, and fall risk. It is quick and easy to conduct and requires a participant to stand up from a chair, walk 3 m, turn 180°, walk back to the chair, and sit down. The outcome is the total duration, which is usually recorded with a stopwatch [34].
d) Health related quality of life (HRQoL) will be used to reflect the progression of PD and will be assessed using the 39-item Parkinson’s Disease Questionnaire (PDQ-39). It has eight domains: mobility, activities of daily living, emotional well-being, stigma, social support, cognition, communication, and bodily discomfort. A summary index of the eight domain scores ranges from 0 to 100, with higher scores indicating more frequent self-perceived difficulties in HRQoL [35].
e) Demographics. The participants’ demographic characteristics include age, gender, duration of disease, weight, height, educational level, health status, medication dosage, exercise habits, and the numbers of falls within 6 months.
Statistical analysis
Statistical analysis will be carried out using IBM SPSS Statistics 24.0. Group comparisons on baseline demographic descriptors and primary and secondary outcome measures will be performed utilizing analysis of variance (ANOVA) for continuous variables and the chi-square independent test for categorical variables. Primary and secondary intervention effects for between/within-group differences will be analyzed by mixed repeated-measures ANOVA. The H&Y stage and MMSE score will be included as covariates since these variables will probably differ significantly among the groups. Bonferroni’s post hoc test will be used to compare results when the main effects are significant. Paired t-test will be adopted to test within-group changes from baseline to post 12-week follow-up.
HbO2 differences among the groups and tasks will be measured using a linear regression, including the interactions between groups and tasks. Pearson correlation coefficients will be adopted to analyze the association between gait performance and PFC activation with tasks in the three groups [7]. We will employ repeated measures ANOVA to analyze differences among conditions with regards to HbO2 levels and gait parameters. Post-hoc analysis will be used to assess for differences between the tasks in both gait and HbO2. Data normality will be tested utilizing the Kolmogorov–Smirnov test. If the data are not normally distributed, the Mann-Whitney U test will be applied to analyze between-group differences under different tasks following training (baseline minus post-intervention), and the Wilcoxon signed-ranks test will be utilized to analyze within-group differences. Significant differences will be set at P values less than 0.05. Missing data and lost to follow-up data will be analyzed based on the intention-to-treat (ITT) principle, and all data obtained from the participants will be included in the analysis.