Vibrotactile Foot Device for Freezing of Gait in Parkinson's Disease: A Pilot Study

Abstract

Background

Vibrotactile stimulation has been studied in its efficacy of reducing freezing of gait (FOG) in patients with Parkinson's disease (PD). However, the results are still controversial. We evaluated the efficacy of a newly developed vibrotactile foot device on freezing severity and gait measures in PD patients with FOG.

Objective

To evaluate the efficacy of vibrotactile foot device on PD patients with FOG.

Methods

Thirty‐three PD patients with FOG were examined during their “off” medication state. The efficacy of the vibrotactile foot device was evaluated using a gait protocol comprising walking trials with vibrotactile stimulation “off” and “on.” Walking trials were videotaped for the offline rating by two movement disorder specialists. The Opal inertial sensor unit (128 Hz; Mobility Lab; APDM Inc., Portland, OR, USA) was used for quantitative gait analysis.

Results

The results demonstrated 33.1% reduction in number of FOG episodes (P < 0.001) and 32.6% reduction of freezing episodes (P < 0.001). Quantitative gait analysis showed a significant increase in step length (P = 0.033). A moderate negative correlation was observed between the change of percent time frozen and age (r = −0.415, P = 0.016). 73% of participants reported minimal to substantial improvement in walking with this vibrating stimulation delivered by the vibrotactile foot device.

Conclusions

The vibrotactile foot device is an efficient device that could significantly reduce freezing severity and provide gait regulation to patients with PD experiencing frequent freezing. It could potentially be used in the home environment for improving the quality of life.

Freezing of gait (FOG) is a common, disabling symptom of patients with late‐stage Parkinson's disease (PD), defined as sudden and brief episodes of inability to produce effective stepping. ¹ FOG can lead to a notable increase in morbidity and mortality by increasing risks of falls as the disease advances. ² Even with optimal medical treatment and deep brain stimulation (DBS) therapy, many patients with PD are incapacitated by FOG and gait disorders. ¹

External cueing, as a non‐pharmacological and non‐invasive treatment, is an important complementary approach in addressing gait disorders in PD. ³ It involves utilizing temporal or spatial stimuli to regulate the initiation and continuation of gait. ⁴ While the precise neural mechanisms remain unclear, it is often recommended for PD patients presenting FOG. The compensation strategy is considered to shift patients from habitual motor control to goal‐directed control by utilizing external cues to focus their attention on walking. ⁵ Both visual and auditory cueing have been extensively studied and shown to be effective in reducing FOG. ⁶ , ⁷ , ⁸ However, practical challenges exist with these modalities, such as requiring dual task functioning which is affected by PD, and impeding communication and drawing attention from other individuals. ⁴ Vibrotactile stimulation, on the other hand, appears promising in overcoming these limitations associated with visual and auditory cues. It works in the form of vibration or electrical stimulation and is considered to act similarly to visual or auditory cues or through sensory feedback processes. ⁹

Several studies have reported that vibrotactile stimulation led to a significant reduction in the percentage of time spent frozen and the number of freezing episodes. ¹⁰ , ¹¹ , ¹² However, most of these studies had relatively small sample sizes. In 2018, a wearable VibroGait device that provided vibrating stimulation on the wrist demonstrated immediate improvement in turning among 35 PD patients with FOG using both open‐loop (constant rhythmical stimulus) and closed‐loop (intermittent stimulus) modes. ¹³ More recently, a study compared a vibrating sock to an auditory cueing device in 31 PD patients with FOG and reported negative results in terms of both the percentage of time frozen and gait parameters. ¹⁴

Therefore, there is a lack of convenient vibrotactile devices ready for daily use supported by reliable clinical trial data. We have developed a vibrotactile foot device with the aim of addressing this gap. This is a single‐center pilot study, designed to assess the effect of the vibrotactile foot device on FOG in patients with PD.

Methods

Participants

Thirty‐three PD patients with FOG were recruited from the Movement Disorders Clinic of the Xuanwu Hospital of the Capital Medical University between May 2022 and August 2023. Idiopathic PD was diagnosed according to the criteria of the Movement Disorder Society. ¹⁵ All patients had a recent history of disabling FOG, defined as (i) presence of FOG frequently in the last month (score of 1 “Yes” on question 1: “Have you experienced FOG in the past month?” and score of 3 “Very often, more than once a day” on question 2: “How often do you experience FOG?” of the New Freezing of Gait Questionnaire [NFOGQ]) ¹⁶ ; (ii) response to a standardized gait task that contained specific elements known to provoke FOG, including gait initiation, a narrow passage, dual tasking, and rapid 360° axial turns in both directions; (iii) able to walk unaided at least 20 m. Exclusion criteria were: (i) other diseases interfering ambulation, such as stroke; (ii) severe foot sensory impairments disabling the individual to perceive vibratory stimulation; (iii) history of deep brain stimulation surgery. We screened participants for sensory impairments during the recruitment process in two steps. First, participants were asked if they had any foot sensory impairments or any medical history that could cause this. Second, we had the participants wear the foot device and stand for a while. If they could perceive vibrations at both the front and the heel of the foot device, they were included in the study. The sample size was based on comparable studie. ¹⁴

The study was approved by the Institutional Review Board of Xuanwu Hospital of Capital Medical University and registered at the clinicaltrial.gov database (NCT06147713). Written informed consent was obtained from all participants prior to the study.

Study Design

All participants were assessed in their “off” medication state, following a 12‐hour withdrawal or postponing the next dose of dopaminergic medication (if they could not tolerate a 12‐hour withdrawal). The efficacy of the vibrotactile foot device was evaluated using a gait protocol comprising three trials ¹⁷ : (1) straight walking: participants walked back and forth over a 10‐m corridor; (2) stop & start trial: participants walked through a 10‐m corridor with a passage set at 5‐m (created by two chairs placed 50 cm apart), stopping and restarting walking at the 3‐m and 7‐m points; (3) full turn trial: participants walked through a 10‐m corridor with a passage set at 5‐m (created by two chairs placed 50 cm apart), making two 360° full turns at the 3‐m and 7‐m points in two directions. The three trials were randomized. All participants completed all trials with vibrotactile stimulation turned “off” and “on.” The participants wore the foot device throughout the entire gait tasks during both the stimulation off and on conditions. Before the trial, participants had 5 minutes to become familiar with the vibrotactile foot device. Throughout the walking trials, participants were videotaped, allowing for offline rating of freezing severity.

Device Description

The vibrotactile foot device (Dailai Technology Inc., Beijing) (see Fig. 1A,B) consists of four flat small vibration disks positioned at both sides of the heel, the head of the 1st metatarsal, and the head of the 5th metatarsal to deliver vibrotactile stimulation. Two pressure sensors at the shoe insole trigger the vibration when the participant's foot touches the floor. The foot device provides vibration in the range of 5–250 Hz, and was set at 200 Hz during the pilot study based on findings from previous studies. ¹³ , ¹⁴ The foot devices are paired with an android application (Dailai Technology Inc., Beijing) using a Pad (Xiaomi Pad 5, Xiaomi Inc, Beijing) that enables control of vibratory stimulation and videotaping during the study (see Fig. 1C).

Figure 1.

(A) Schematic overview of the vibrotictile foot device. Four small vibration disks (black dots) on each side provide stimulation, triggered by two pressure sensors (red dots) on the insole. A switch on the tongue controls “off” and “on.” (B) Photograph of the device. (C) A paired Android application.

Clinical Assessments

The clinical evaluation included the assessment of Movement Disorder Society‐Unified Parkinson's Disease Rating Scale (MDS‐UPDRS III) motor scores, ¹⁸ Hoehn and Yahr (H–Y) stage, New Freezing of Gait Questionnaire (NFOG‐Q), ¹⁶ Mini‐Mental State Examination (MMSE), ¹⁹ Montreal Cognitive Assessment (MoCA), ²⁰ Frontal Assessment Battery (FAB), ²¹ Hamilton Anxiety Rating Scale (HAM‐A), ²² and the Hamilton Depression Rating Scale (HAM‐D). ²³ Participant satisfaction was rated using The Clinical Global Impressions Scale (CGI‐I). ²⁴

Freezing and Gait Assessments

Two movement disorder specialists from the Movement Disorder Department at Xuanwu Hospital, were recruited to independently evaluate the offline videos for Freezing of Gait (FOG). The movement disorder specialists were blinded to whether the vibrotactile stimulation was on or off. Video data were annotated offline in ELAN (Version 6.2). ²⁵ Each rater received 198 videos in a random order, with 6 trials for each participant. Then, a FOG tool was used to build the consensus between the two raters. ²⁶ A tolerance of 2 seconds and the correction parameter “include” were set up to determine the consensus. ¹⁴

For quantitative gait analysis, straight walking trial was conducted using the Opal inertial sensor unit (128 Hz; Mobility Lab; APDM Inc., Portland, OR, USA), which was worn on the bilateral wrists, ankles, and the trunk (sternal and lumbar region). Automatically extracted gait parameters include cadence, double support, stride length, stride velocity, stride time, turn duration, and turn velocity.

Statistical Analysis

Normality and homogeneity were examined for all variables. Paired sample t test or Wilcoxon signed‐rank test were used to assess efficacy between vibrotactile stimulation off and on. The independent‐samples t test or Mann–Whitney U test were performed to compare demographic and clinical characteristics between participants with and without response to the foot vibrotactile device, based on the mean value of percent time frozen. A Bonferroni correction was applied to adjust for multiple comparisons across measures of freezing and gait (adjusted P value 0.006). Spearman's correlation coefficient was used to analyze the relationship between outcome measurements and clinical characteristics. All statistical analyses were conducted using SPSS, version 28 (IBM Corp., New York, USA). P < 0.05 was considered statistically significant; correlations were interpreted based on |r| values as follows: 0.8–1.0, very strong; 0.6–0.8, strong; 0.4–0.6, moderate; 0.2–0.4, weak; and 0.0–0.2 very weak or no correlation.

Results

Demographic Characteristics

Forty‐two patients were enrolled in this study of which six did not exhibit objective freezing episodes during the recruitment process when assessed using provoking tasks or during the walking trials despite reporting frequent episodes of freezing in their daily lives. Additionally, three patients were excluded due to difficulties perceiving vibrations while wearing the vibrotactile foot device. Despite screening for foot sensory impairment, these individuals still experienced difficulties in vibration sensation. The demographic and clinical characteristics of the remaining 33 participants are presented in Table 1.

TABLE 1.

Baseline demographics and clinical information for all participants

Characteristics	Mean ± SD or n (%)
Age, y	67.5 ± 8.8
Men	18 (54.5%)
Disease duration, y	7.5 ± 4.4
MDS‐UPDRSIII	36.0 ± 15.7
LEDD (mg)	881.7 ± 315.3
Hoehn & Yahr	2.5 ± 0.6
NFOGQ (0–28)	23.1 ± 4.2
MMSE (0–30)	26.3 ± 3.1
MoCA (0–30)	21.7 ± 5.3
FAB (0–18)	14.7 ± 2.9
HAMA (0–56)	6.8 ± 7.0
HAMD (0–54)	4.1 ± 5.6

Abbreviations: NFOGQ, New freezing of gait questionnaire; MDS‐UPDRS III, Movement Disorder Society–Unified Parkinson's Disease Rating Scale part III; LEDD, Levodopa Equivalent Daily Dose; FAB, frontal assessment battery; MMSE, Mini‐mental state examination; HAMA, Hamilton Anxiety Rating Scale; HAMD, Hamilton Depression Rating Scale.

Outcome Measures

Freezing Measurement

All participants experienced freezing episodes in at least one gait trial. One patient did not experience a freezing episode during stimulation off but did experience one during stimulation on and was therefore included in the final analysis. Three out of 33 participants were not able to complete the full turn trial, and the remaining trials were used for analysis. Vibrotactile stimulation was associated with a notable reduction in the number of FOG episodes (stimulation‐off: 4.5 ± 4.4; stimulation‐on: 3.0 ± 3.1, P < 0.001). With stimulation off, 32 out of 33 experienced a total of 148 FOG episodes, whereas 99 FOG episodes occurred in 29 patients when stimulation on, reflecting a 33.1% reduction with vibrotactile stimulation. Additionally, vibrotactile stimulation showed a significant decrease in the percentage of time spent frozen (stimulation‐off: 27.9% ± 24.2%; stimulation‐on: 18.8% ± 19.4%, P < 0.001), representing a 32.6% improvement. These differences did survive after adjusting for multiple comparisons (P < 0.006). The results are shown in Table 2 and Figure 2. The results for each walking trial are shown in Supplemental Table S2. A sample video showing greatly enhanced walking performance using the vibrotactile foot device, as shown in Video 1.

TABLE 2.

Freezing and gait measure changes with vibrotactile stimulation

Variables	Stim‐OFF	Stim‐ON	P
Freezing measures
Freeze episodes a	4.5 ± 4.4	3.0 ± 3.1	<0.001**
Percent time freeze a	27.9 ± 24.2	18.8 ± 19.4	<0.001**
Gait measures
Cadence (steps/min)	114.5 ± 12.5	116.0 ± 12.9	0.518
Double support (%) a	23.3 ± 11.1	27.5 ± 13.7	0.078
Stride time (s)	1.1 ± 0.1	1.1 ± 0.1	0.787
Stride velocity (m/s)	0.7 ± 0.2	0.7 ± 0.2	0.119
Stride length (m)	0.7 ± 0.2	0.8 ± 0.2	0.033*
Turn duration (s)	3.0 ± 0.6	2.8 ± 0.5	0.236
Turn velocity (degrees/s)	118.8 ± 34.7	126.1 ± 38.6	0.068

Note: Data presented as mean ± SD.

^* P < 0.05;

^** P < 0.001;

^a Analyzed by wilcoxon test.

Figure 2.

(A) The mean freeze episodes with vibrotactile stimulation “off” and “on.” (B) The mean percent time freeze with vibrotactile stimulation “off” and “on.” **P < 0.01. StimOFF, stimulation off. StimON, stimulation on. Error bars represent the standard error of the mean.

Video 1.

This video demonstrates a patient with Parkinson's disease exhibiting freezing of gait. The first part shows walking without vibrotactile stimulation. Freezing episodes worsen during turning, initiating, straight walking, and in narrow passages. The second part illustrates improvement with the vibrotactile foot device turned on.

Gait Measurement

Comparing with the stimulation off condition, the vibrotactile stimulation exhibited a significant increase in stride length (m) (stimulation‐off: 0.7 ± 0.2; stimulation‐on: 0.8 ± 0.2, P = 0.033). This difference did not survive after adjusting for multiple comparisons (P < 0.006). No significant differences were observed in any other variables. The results are shown in Table 2.

Patient Satisfaction

Participant satisfaction was assessed using the Clinical Global Impressions Scale (CGI‐I). Out of 33 patients, nine reported much improvement (one reported very much improvement), 15 reported minimal improvement, and nine reported no change. None of the patients reported a worsening of their walking condition.

Comparisons of Patients with or without Response to Stimulation

The participants were divided into two groups based on their response to the stimulation, which was determined by the mean value of change of percent time frozen. The aim was to explore the characteristics of participants who respond well to vibrotactile stimulation. Patients who responded to the stimulation were of younger age (P = 0.137) and with slightly higher FAB scores (P = 0.624) compared to those who did not, but these differences did not reach statistical significance. No other clinical characteristics differed between the two groups. The results are presented in Table 3.

TABLE 3.

Comparison in demographic and clinical characteristics between participants with or without response to vibrotactile stimulation

Characteristics	FOG‐r	FOG‐nr	P
N	10	23
Age, y a	61.6 ± 12.2	68.3 ± 5.1	0.137
Disease duration, y	6.8 ± 5.0	7.1 ± 3.5	0.501
MDS‐UPDRSIII a	36.1 ± 19.5	33.4 ± 13.6	0.638
LEDD (mg)	855.4 ± 229.1	754.3 ± 381.6	0.853
Hoehn & Yahr a	2.2 ± 0.4	2.5 ± 0.6	0.208
NFOGQ (0–28) a	23.7 ± 4.0	22.8 ± 4.3	0.325
MMSE (0‐30) a	26.0 ± 3.4	26.5 ± 3.2	0.540
MoCA (0‐30) a	21.0 ± 6.6	22.4 ± 4.7	0.432
FAB (0‐18) a	15.3 ± 3.0	14.5 ± 3.0	0.624
HAMA (0–56) a	6.7 ± 8.5	7.0 ± 7.0	0.890
HAMD (0‐54) a	4.9 ± 7.5	4.1 ± 5.1	0.858

Note: Data presented as mean ± SD. FOG‐r, PD patients with freezing of gait who response to vibrotactile stimulation; FOG‐nr, PD patients with freezing of gait who do not response to vibrotactile stimulation.

Abbreviations: y, years; NFOGQ, New freezing of gait questionnaire; MDS‐UPDRS III, Movement Disorder Society–Unified Parkinson's Disease Rating Scale part III; LEDD, Levodopa Equivalent Daily Dose; FAB, frontal assessment battery; MMSE, Mini‐mental state examination; HAMA, Hamilton Anxiety Rating Scale; HAMD, Hamilton Depression Rating Scale.

^a Analyzed by Mann‐Whitney U test.

Correlations between Outcome Measures and Clinical Characteristics of Patients

A moderate negative correlation was observed between the change of percent time frozen and age (r = −0.415, P = 0.016). No significant correlations were found in other items. The results are presented in Supplemental Table S1.

Discussion

This study evaluates the efficacy of a newly developed vibrotactile foot device providing a closed‐loop vibration stimulation for PD patients with FOG in their “off” medication state. The results demonstrated a moderate and significant reduction in both number of FOG episodes (33.1% reduction) and percent time freezing (32.6% reduction). Quantitative gait measurement showed a significant improvement of stride length, with an average increase of approximately 0.1 m. These objective improvements were consistent with the positive subjective experiences reported by the patients, with 73% reporting varying degrees of improvement.

Vibrotactile cueing devices have been studied in patients with PD who experience FOG, and most have been reported to be effective. ¹⁰ , ¹¹ , ¹³ , ¹⁴ , ²⁷ However, the results are still controversial due to the small sample sizes. For instance, a study using a shoe insole delivering 275 Hz vibration through a small, attached vibrator, triggered by an inertial measurement unit (IMU) and a pressure sensor to detect freeze episodes was tested in a single PD patient with FOG, resulting in a 34% reduction in walking time. ¹¹ In another study, 20 PD patients with FOG were recruited to evaluate a rhythmic vibration metronome provided on the pelvis. ²⁷ However, this metronome showed no significant improvement of FOG compared with a laser cane, U‐step, and an auditory cue provided by the same metronome. Additionally, a rhythmic vibrotactile belt designed to provide 250 Hz vibration on patients’ ankles via two vibrating motors yielded significant reductions in freezing episode duration. ¹⁰ It correlated with Tinetti scale evaluations of clinical gait and balance. ¹⁰ Furthermore, 25 PD patients with FOG were tested on VibroGait device—delivering vibration in the 200–300 Hz range to the wrist by two modes: rhythmic vibration (open‐loop) and the other while in the stance phase of gait (closed‐loop). Both modes reported a significant reduction in the percent time freezing during turning. ¹³ However, a recent study, using vibrating socks, delivering 183 Hz vibration through a motor disc below the arch of the feet with two modes similar to VibroGait, have been examined in 31 PD patients with FOG and compared with an auditory cueing device showed no significant improvement. ¹⁴ In our study, the vibrotactile foot device was similar to the closed‐loop mode of VibroGait and vibrating socks, providing a 200 Hz vibration during the stance phase of walking. We assessed its effectiveness with a relatively larger sample size compared to previous studies, involving 33 PD patients with FOG, all of whom experienced at least one freezing episode during the walking trials. The foot device in our study contained larger number of vibrating disks compared to previous studies and positioned at four different locations near the bony areas of each foot, which are more sensitive to perceiving vibration. This is particularly relevant for patients with PD, who are known to have underlying proprioceptive deficits. ²⁸ The increased number of vibrating disks aims to enhance the patient's ability to sense vibration. The vibrotactile foot device used in our study significantly reduced freezing episodes and the percent time freezing, in line with the satisfaction of the participants. 73% of the participants reported minimal to substantial improvement in walking with this pattern of vibrating stimulation.

External cueing devices have been previously shown to provide rhythmic and continuous stimuli that may interfere with patients’ normal walking patterns and lead to reliance on cue in patients with freezing of gait. ²⁹ , ³⁰ Such reliance could impede the potential for long‐term learning effects. ³¹ However, recent technological advancements have explored the “on‐demand” delivery of cues, providing stimuli only when specific gait alterations occur. ⁴ This approach offers feedback on performance and may reduce the development of reliance on cue. ³² In the current study, the closed‐loop cueing modality of the vibrotactile foot device controlled by two pressure sensors placed on the shoe insole, offers a “phase‐dependent” cueing system during the stance phase of gait. Further studies are needed to examine the efficacy of using the vibrotactile foot device in daily life.

There is increasing evidence of FOG not being a sole motor symptom but a complex phenomenon caused by interaction between motor, cognitive, and affective factors. ³³ It is well established that cognition is closely related to FOG. Studies have shown that PD patients with freezing demonstrate poorer cognitive function across various domains, including executive function/attention, language, memory, and visuospatial abilities. ³⁴ In our study, attention‐demanding tasks were incorporated into walking trials, such as 360° turning or navigating through a passage. Patients who positively responded to vibrotactile stimulation, although not reaching statistical significance, demonstrated slightly higher cognitive performance as assessed by the Frontal Assessment Battery (FAB) scale. In addition, individuals with younger ages were observed to have more benefit from this vibrotactile stimulation. This suggests that PD patients with greater cognitive reserves may be more likely to benefit from the vibrotactile cueing device. This finding aligns with previous suggestions that cognitive impairment may hinder the use of attention‐demanding compensatory strategies, such as external cueing, particularly in individuals with FOG, may struggle to implement these methods. ³²

Limitation

Limitations of this study should be acknowledged. First, the current study only examined the immediate effect of the vibrotactile foot device on PD patients with FOG during their “off” medication stage, limiting in a laboratory environment. During the recruitment process, 6 out of 42 patients did not exhibit objective freezing episodes, despite reporting frequent episodes of freezing in their daily lives. Future studies should consider home‐based research to assess this device's efficacy in a real life scenario. Second, the vibrotactile foot device cannot be used in individuals with sensory impairments, particularly considering vibration sensation. In this study, three patients were unable to feel the vibrations while wearing the foot device, ruling out the possibilities of using this device. Furthermore, this was a single‐center pilot study without a double‐blind and control group design. Future studies should incorporate a more rigorous protocol to address this limitation. In this study, the vibrotactile stimulation was performed in a fixed order, with stimulation off first, followed by stimulation on. This fixed order design could potentially impact the results. Future studies should improve the design by randomizing the order of stimulation on and off.

Conclusion

In conclusion, the vibrotactile shoe device is an efficient device that provides immediate relief to PD patients experiencing frequent freezing, and it may become a practical device for facilitating movement in daily life. Future studies will need to examine the effectiveness and usability of the vibrotactile shoe device in a home‐based setting, to assess whether it has the potential to provide a learning effect through long‐term training.

Author Roles

(1) Research Project: A. Conception, B. Organization, C. Execution; (2) Statistical Analysis: A. Design, B. Execution, C. Review and Critique; (3) Manuscript: A. Writing of the First Draft, B. Review and Critique

S.C.: 1A, 1B, 1C, 2A, 2B, 3A

J.M.: 1C, 3B

H.S.: 1C, 3B

H.Z.: 1C, 3B

Y.L.: 1C, 3B

W.M.: 1C, 3B

E.X.: 1C, 3B

S.M.: 1C, 3B

J.C.: 3B

Z.R.: 3B

H.W.: 1C

K.W.: 1A, 1B, 2C, 3B

P.C.: 1A, 1B, 1C, 2C, 3B

Disclosures

Ethical Compliance Statement: This study was approved by the Institutional Review Board of Xuanwu Hospital of Capital Medical University and registered at the clinicaltrial.gov (NCT06147713). Written informed consent was obtained from all participants prior to the study. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this work is consistent with those guidelines.

Funding Sources and Conflict of Interest: The authors declare that there are no conflicts of interest relevant to this work. K.W. and P.C. serve as consultant for Dailai Technology Inc.

Financial Disclosures for the previous 12 months: This work was supported by the National Key R&D Program of China [No. 2021YFC2501200], Talent Program by Xuanwu Hospital Capital Medical University [No. YC20220119], National Natural Science Foundation of China [No. 82301423].

Supporting information

Table S1. Correlations between the change in outcome measures and demographic and clinical characteristics of all participants.

Table S2. Freezing measure changes with vibrotictile stimulation.