Effects of Biofeedback Footwear in Parkinson's Disease: Assessment of Functional Motor Abilities and Locomotion

Parkinson's Disease (PD) is a chronic and progressive neurodegenerative condition, clinically characterized by the presence of bradykinesia, resting tremor, rigidity, and postural instability. Individuals affected by this condition exhibit several gait abnormalities, characterized by reduced speed, short and shuffling steps, and increased step variability.

In recent decades, efforts have been made towards the development of sustainable and implementable strategies in the daily lives of these individuals, aimed at improving posture and gait performances.

In this context, the role of enhanced plantar biofeedback is under investigation. For example, plantar sensory stimulation seems to improve postural adjustments and gait. Moreover, it has been observed that the barefoot contact of the foot with the ground, as opposed to wearing a shoe, is capable of providing greater feedback, thus improving posture and locomotion. Improved locomotion leads to increased safety while performing daily living activities, a reduction in the risk of falls, and therefore an increase in quality of life. Studies in the literature are primarily focused on stimulating insoles and footwear with integration of vibratory, auditory, and/or visual biofeedback.

The applicability of electronic footwear is currently limited. A possible sustainable solution could be the use of footwear that minimizes the separation of the foot from the external environment, while effectively protecting it from possible injuries. This type of footwear conforms to the shape of the foot, potentially allowing for greater plantar feedback when compared to traditional shoes. Gait using inertial sensors represents an easily implementable method in research and clinical practice, with application in numerous neurological syndromes. A device worn at the lumbar level can measure spatiotemporal parameters of gait. Moreover, it is possible to calculate trunk-derived accelerometric indexes. In particular, the Harmonic Ratio (HR) is a representative index of the fluidity and rhythm of physiological gait. It is therefore a reliable parameter for evaluating possible gait alterations in neurological patients.

The primary aim of the study is to verify whether the use of footwear providing increased plantar feedback improves HR in PD patients.

As a secondary aim, changes in other trunk-derived gait indexes, static postural control and functional capabilities will be evaluated. Moreover, middle-term changes (after four weeks of plantar feedback shoes wearing) in these parameteres, as well as the occurence of falls and freezing of gait (FOG), will be assessed.

The study is divided into 2 work-packages (WP). WP1 is an interventional open-label study, in which short-term changes induced by footwear providing increased plantar feedback will be observed.

At the initial assessment (T0), clinical-demographic data will be collected, and patients will undergo a comprehensive neurological examination. Disease-specific scales (UPDRS III part 3, NFOG-q, LEDD, DASS-21), will be administered.

Patients will undergo inertial gait analysis, static barefoot posturography, and functional performance tests (TUG, 2MWT, 5-STST, 10-mFW) under three conditions in a randomized order:

• Barefoot condition (BC): without wearing any shoes
• Neutral condition (NC): wearing neutral gym shoes
• Increased plantar feedback condition (PFC): wearing plantar shoes

WP2 is a prospective, interventional randomized controlled study, in which changes induced by the adoption of plantar shoes for four weeks will be evaluated. After the baseline evaluations (T0), patients will be randomized and assigned to one of the following groups:

• Neutral condition (NC): wearing neutral gym shoes
• Greater plantar feedback condition (PFC): wearing plantar shoes

In both groups, participants will be provided with appropriately sized shoes and will be asked to adopt the assigned shoes as their usual footwear for everyday activities over a period of four weeks. Patients will be required to keep a diary to indicate any issues with the use of the shoes and daily usage time, as well as keep record of falls and FOG episodes.

Evaluations scheme:

Gait analysis procedure:

To acquire gait data, an inertial sensor (BTS G-Walk, BTS, Milan, Italy) will be positioned at the level of the fifth lumbar vertebra (L5) using an ergonomic belt. The inertial sensor communicates with a laptop via Bluetooth for data recording and offline analysis. The sensor incorporates a triaxial accelerometer (16 bits/ axis), triaxial magnetometer (13 bits), and triaxial gyroscope (16 bits/axis). The sampling frequency is 100 Hz, recording linear and angular trunk accelerations in the anteroposterior (AP), mediolateral (ML), and vertical (V) directions.

The "Walk+" protocol of the G-STUDIO software (G-STUDIO, BTS, Milan, Italy) will be used to detect trunk acceleration, phases of the right and left step cycles, and spatiotemporal parameters of pelvic kinematics. The harmonic ratio will be calculated as trunk acceleration data along three directions (vertical, medio-lateral and anteroposterior), decomposing the components of the signal into its harmonics, as the ratio between the sum of the first 10 even and the first 10 odd harmonic multiples of the fundamental frequencies.

The other Trunk Inertial Indexes calculated will be the largest Lyapunov exponent (LLE), coefficient of variation (CV), log dimensionless jerk score (LDLJ), the recurrence quantification analysis (RQA), as described in a previous work by Castiglia et al.

Before the experimental session, participants will be instructed to walk on the ground along a predetermined path to become familiar with the procedure. They will be instructed to walk at a pace consistent with their usual gait, along a corridor approximately 3 meters wide and 30 meters long, in the absence of external factors that may interfere with step cadence and rhythm. Five consecutive trials will be performed. The trials will be stopped if the patient reports discomfort during the procedure, fatigue, or pain.

Posturographic evaluation:

For posturographic evaluation, computerized stabilometry will be performed, recording the patient in an orthostatic position, under quiet conditions and without disturbances with eyes open and eyes closed, using force platforms available at our Institute (BTS P-6000, BTS, Milan, Italy). Patients will be asked to position themselves on the posturographic platforms. Once the position is achieved and it is ensured that the patient is comfortable, the technician will record 3 trials of 10 seconds with eyes open and 3 trials of 10 seconds with eyes closed. To reduce the learning effect, the trials will be recorded in a randomized manner. The trials will be stopped if the patient reports discomfort during the procedure, fatigue, or pain.

Administration of Functional Capacity Scales:

• Timed-Up and Go (TUG) Test: The TUG is a simple test for the assessment of overall motor functionality. Participants are asked to rise from a chair, walk a linear distance of 3 meters, turn 180 degrees, and return to a seated position in the shortest time possible.
• 2-minute Walk Test (2MWT): The 2MWT evaluates autonomy and aerobic walking capacity. In the 2MWT, the subject is instructed to cover the greatest distance by walking along a flat 30-meter path within a span of 2 minutes.
• 5 Sit to Stand Test (5-STST): The 5-STST is used to assess the functionality and strength of the lower limbs. The subject is asked to stand up and sit down from a chair five times in the shortest time possible.
• 10-meter Free Walk (10-mFW): In the 10-meter test, the subject is asked to walk along a flat 10-meter path at a perceived comfortable (usual) pace.

Statistical Plan A preliminary normality analysis will be conducted to determine whether to use parametric or non-parametric methods, using graphical representation and the Shapiro-Wilk test. Numeric variables will be described in terms of mean and standard deviation (or median and quartiles if appropriate), while categorical variables will be presented as raw count and percentage.

Within WP1, the three groups will be compared using ANOVA with Bonferroni post-hoc tests or Friedman tests for dependent samples.

Correlations with quantitative parameters will be assessed using the Pearson correlation coefficient (or Spearman if necessary), both within groups and globally.

Within WP2, changes in HR will be evaluated with repeated measures ANOVA (or a similar non-parametric test) with two factors: TIME (2 levels, T0 vs. T1, within groups) and GROUP (2 levels, plantar shoe vs. neutral shoe, between groups).