Comparing the effect of off-loading knee brace with and without local muscle vibration on clinical and biomechanical outcomes in patients with medial knee osteoarthritis: A randomized clinical trial

Abstract

Background

Medial knee osteoarthritis (KOA) is a prevalent degenerative joint disease causing pain and functional impairment. Off-loading knee braces reduce pain but may decrease muscle activity, leading to weakness. Integrating local muscle vibration (LMV) into off-loading braces may enhance muscle activation and clinical outcomes.

Objective

To design a portable LMV system synchronized with gait phases and compare its efficacy to a conventional off-loading brace in patients with medial KOA.

Methods

In this randomized clinical trial, 16 patients with medial KOA were assigned to either an LMV-equipped off-loading brace group or a conventional brace group for 4 weeks. Clinical outcomes (Western Ontario and McMaster Universities Osteoarthritis Index [WOMAC], Visual Analog Scale [VAS]) and biomechanical parameters (knee adduction moment [KAM1, KAM2], impulse, range of motion [ROM], cadence, stride length) were assessed pre- and post-intervention using validated questionnaires and a motion analysis system.

Results

The LMV-equipped brace group demonstrated a significantly greater reduction in KAM impulse (−24.79% vs. −7.68%, p=0.050) and improved knee ROM (p=0.048) compared to the conventional brace group. Significant improvements in WOMAC (p=0.001) and VAS (p=0.011) scores were observed in the LMV group, indicating enhanced functional status and pain relief.

Conclusion

The LMV-equipped off-loading brace provides superior biomechanical (KAM impulse, ROM) and clinical outcomes compared to conventional braces, offering a promising intervention for medial KOA.

Introduction

Knee osteoarthritis (KOA) is a prevalent chronic degenerative disease characterized by the progressive deterioration of articular cartilage, leading to inflammation and pain in the knee joint. It is estimated that approximately 364.6 million individuals are affected by KOA globally, representing around 4.9% of the world population. This condition is particularly common among women and individuals over the age of 55, with aging being a significant risk factor for its development.^1,2 The underlying pathophysiology involves not only cartilage breakdown but also synovial inflammation, subchondral bone remodeling, and osteophyte formation, which collectively contribute to the chronic nature of the disease and its resistance to complete reversal through conservative measures.

The clinical manifestations of KOA include pain that is often exacerbated by activity, instability, crepitus (a grating sensation), tenderness, stiffness, and restricted range of motion. These symptoms can severely impair daily activities and overall quality of life.^1,2 Notably, KOA is associated with significant weakness in the quadriceps muscle, with strength deficits reported to range between 20% and 45% compared to healthy individuals.^3–5 This weakness contributes to impaired dynamic knee stability and a decline in physical function, further complicating the management of the condition.^4–6 The quadriceps’ role extends beyond simple movement; it serves as a primary stabilizer, influencing joint congruence and load distribution during weight-bearing activities, thereby amplifying the impact of weakness on disease progression.

Recent longitudinal studies suggest that maintaining stronger quadriceps at the onset of KOA may mitigate the progression of knee pain, patellofemoral cartilage loss, and joint space narrowing. ⁷ The quadriceps play a crucial protective role by absorbing shock during ambulation. They work eccentrically to cushion the knee and decelerate the leg before foot strike, which helps to reduce impulsive loading on the joint.^8,9 Research has demonstrated a clear link between weaker quadriceps and an increased rate of loading at the knee, which can exacerbate the degenerative processes associated with KOA.^10,11 This relationship underscores the importance of targeted interventions that address muscle strength as a modifiable factor in slowing disease advancement.

A significant contributor to quadriceps weakness in individuals with KOA is Arthrogenic Muscle Inhibition (AMI). AMI is characterized by persistent neural inhibition that prevents full activation of the quadriceps, leading to muscle weakness and, potentially, muscle atrophy.^12,13 This condition likely arises from changes in sensory receptors within the damaged knee joint, which disrupts normal sensory feedback and affects various spinal and supraspinal pathways.^14–16 Notably, some patients exhibit significant deficits in quadriceps activation even in the absence of pain or visible swelling, indicating that AMI may also stem from a loss of normal excitatory sensory signals from healthy mechanoreceptors.^17,18 AMI represents a complex interplay between peripheral joint pathology and central nervous system adaptations, highlighting the need for therapies that target both mechanical and neurological aspects of muscle function.

In the realm of non-surgical management for KOA, various rehabilitation approaches are employed, including weight loss, patient education, exercise, orthotics, and neuromuscular training. ¹⁹ Among these, the use of off-loading knee braces has been recognized as an effective intervention. These braces help reduce pressure on the affected compartment, alleviating pain and enhancing functional capacity. However, long-term use of these devices can lead to decreased muscle activity and exacerbate muscle weakness, creating a vicious cycle that may accelerate disease progression.^4,20–22 This paradoxical effect emphasizes the necessity for adjunctive strategies that counteract muscle disuse while maintaining the brace’s unloading benefits.

To address these challenges, neuromuscular training techniques, such as whole-body vibration (WBV) and Local Muscle Vibration (LMV), have garnered attention for their potential to improve muscle activity and functional performance.^23–25 While WBV shows promise, its clinical application is limited due to factors such as reduced vibration energy reaching target muscles, high equipment costs, and lack of portability. Conversely, LMV is considered a feasible and cost-effective alternative, delivering targeted vibrations directly to specific muscles or tendons using lightweight, handheld devices.^26,27 LMV operates through mechanisms like the tonic vibration reflex, which enhances Ia afferent signaling to promote muscle spindle activation and subsequent motor unit recruitment, potentially overcoming AMI more effectively than generalized vibration methods.

Therefore, the present study aims to investigate the feasibility of combining off-loading knee braces with LMV to simultaneously reduce joint pressure and strengthen the quadriceps. This study will assess the impact of these combined approaches on clinical outcomes and biomechanical parameters in individuals with KOA.

Methods

Study design and participants

This randomized clinical trial was conducted on 16 participants diagnosed with medial compartment knee osteoarthritis. Participants were selected based on specific inclusion and exclusion criteria and were randomly assigned to either an intervention group (vibration-equipped knee brace) or a control group (conventional knee brace). The intervention period for both groups lasted 4 weeks. To ensure robustness, the trial incorporated standardized protocols for brace fitting, usage monitoring, and outcome assessments, with all evaluations performed under controlled laboratory conditions to minimize external variables.

Ethics statement

The study protocol received approval from the Ethics Committee of the University of Social Welfare and Rehabilitation Sciences (Approval Code: IR. USWR.REC.1403.112) in September 2024. The protocol was also registered with the Iranian Registry of Clinical Trials on February 14, 2024 (Reference Code: IRCT20240929063200N1). All participants were informed about the study by a specialist physician and provided written informed consent. Evaluations were conducted in the Biomechanics Laboratory at Sharif University of Technology, with participant recruitment starting on April 5, 2025, and concluding on September 11, 2025. Ethical considerations extended to ensuring participant confidentiality, voluntary withdrawal options, and regular monitoring for adverse events such as skin irritation or discomfort from brace use.

Sample size

A priori sample size estimation was performed using G*Power software (version 3.1.9.7). Based on an expected effect size for knee adduction moment, ²⁸ an alpha level of 0.05, and a statistical power of 80%, the required sample size was 8 participants per group. Finally, 16 individuals with medial compartment knee osteoarthritis were enrolled and randomly assigned to two groups: (1) conventional off-loading knee brace (n = 8) and (2) LMV-equipped off-loading knee brace (n = 8). The estimation accounted for potential dropout rates and variability in biomechanical measures, ensuring sufficient power to detect clinically meaningful differences. (Figure 1)

Figure 1.

CONSORT 2025 flow diagram.

Flow diagram of the progress through the phases of a randomised trial of two groups (that is, enrolment, intervention allocation, follow-up, and data analysis).

Eligibility criteria and recruitment

Participants were included in the study based on inclusion and exclusion criteria as follows:

Inclusion Criteria: Osteoarthritis of the medial compartment, graded 2 or 3 based on the Kellgren-Lawrence scale, ²⁹ age in the range of 50-70 years ³⁰ and Body Mass Index (BMI) ≤ 30 kg/m². ²⁹ Recruitment involved screening through orthopedic clinics, with radiographic confirmation of KOA grade and BMI verification via standardized measurements.

Exclusion Criteria: Knee osteoarthritis of the lateral compartment, Rheumatoid arthritis, Osteoarthritis of the ankle, hip, spine, and patellofemoral joint, any chronic or acute disorders causing an observable change in the individual’s gait pattern, History of lower limb surgery, History of knee joint injection or surgery within the past six months. These criteria were rigorously applied to homogenize the sample and reduce confounding factors related to comorbidities or prior interventions.

Randomization, blinding, and treatment allocation

Eligible participants were randomly assigned to either the intervention or control group based on the result of a coin toss. This simple randomization method was chosen for its practicality in a small sample, with allocation concealment maintained until brace assignment. A single-blinded approach was implemented, where participants were unaware of the comparative group details, and assessors remained blinded to group allocation during data collection and initial analysis to minimize bias.

Intervention

All participants received a customized off-loading knee brace tailored to their specific size and pathology. The study’s two groups underwent distinct interventions based on this brace platform.

Conventional off-loading brace

Participants in the control group were provided with a conventional off-loading knee brace. This relatively lightweight (<500 g) three-point valgus knee brace was designed to alleviate pressure on the medial compartment by applying a three-point valgus force, effectively shifting the weight-bearing load from the damaged medial compartment to the lateral compartment. The brace featured a unilateral polycentric joint on the medial side, a soft pad on the outside, and cross-over straps for fixation. Custom molding utilized the three-point pressure principle to correct varus knee angulation.

The brace was fabricated at a specialized orthotic clinic, where initial anthropometric measurements were taken, including.

• • Trochanteric length to the knee joint condyle.

• • Length from joint condyle to medial malleolus.

• • Circumference of the thigh and shank at 1/3 and 2/3 of their respective lengths.

The VB was fitted by an experienced orthotist and comprised a neoprene thigh and lower leg section, extending to approximately two-thirds of the corresponding leg area. It was adjustable and patient-controlled, ensuring that it did not exert unacceptable pressure on the knee. This group wore the standard off-loading brace without any vibration for a minimum of 4 hours daily during walking activities for 4 weeks. For comfort, foam lined the areas of the brace that contacted the body, and the orthosis was secured using elastic and non-elastic straps. The straps were made of elastic material, with their tightness adjusted based on the principle of Maximum Tolerable Pressure (MTP). This approach ensures optimal load transfer while prioritizing patient comfort. To enhance compliance, the external strap’s tension was adjusted based on the “maximum tolerable pressure” principle. ²⁸ Compliance was monitored through self-reported logs and follow-up calls to ensure consistent usage (Figure 2)

Figure 2.

Conventional off-loading knee brace.

Equipped the off-loading knee brace with a customized LMV system

The intervention group used an off-loading knee brace equipped with a customized LMV system. The brace retained the structural design of the conventional model but incorporated a portable vibration module positioned laterally on the shank and secured in a 3D-printed pocket within the brace to prevent slippage.

Participants in the LMV group followed the same brace-wearing instructions and additionally received LMV stimulation during 30-minute walking sessions, three times per week for four weeks (12 sessions in total). Participants used the LMV-equipped brace for four weeks every other day. The vibration device was applied during 12 walking sessions (three sessions per week), each consisting of a 30-minute walk on a flat surface per session at a self-selected pace. During these sessions, the vibration device was active. Outside these sessions, participants in the LMV group continued to wear the brace during daily activities without activating the LMV module. Each participant in the LMV group used the vibration module for an average of 90 minutes per week. Adherence to the intervention protocol was monitored through regular telephone follow-ups and self-reported usage logs completed by participants at the end of the intervention period. The LMV system represented an advancement over previous prototypes, utilizing a lightweight single-joint design that prioritized patient comfort.

The LMV device consisted of a portable vibration module controlled by a microcontroller (ATmega128) and an inertial measurement unit (MPU-6050) for gait phase detection. Additional components included a DC-DC boost converter, a lithium-ion battery charger module, and a push-button interface. Mechanical vibration was generated using a miniature 5-V solenoid actuator operating at a frequency of 30 Hz. ³¹ This frequency was selected based on evidence suggesting optimal stimulation of Ia afferent fibers without inducing muscle fatigue. The actuator produced oscillatory mechanical stimulation with an approximate peak-to-peak amplitude of 0.5 mm and a sinusoidal waveform.

The angular velocity signal of the shank in the sagittal plane was processed by the microcontroller to detect gait events, including mid-swing, initial contact, and toe-off, based on the algorithm described by Maqbool et al. ³² Vibration was automatically activated from the mid-swing phase until the end of the stance phase, corresponding to approximately 70% of the gait cycle, and remained inactive during the remaining phases. The gait phase detection algorithm was based on the angular velocity and acceleration signals obtained from the inertial measurement unit mounted on the shank. Because the acceleration component is aligned with the direction of motion, the sign of the acceleration signal corresponds to the direction of shank rotation. During walking, the shank rotates clockwise during the stance phase and counterclockwise during the swing phase. Therefore, the direction of acceleration differs between these two phases. Based on this principle, the algorithm used the direction of the acceleration signal to distinguish between stance and swing phases. When the stance phase was detected, the controller activated the vibration module, while vibration remained inactive during the swing phase.

Before participant testing, the vibration output of the actuator was evaluated during bench testing. The inertial measurement unit (MPU-6050) integrated into the device was used to record the acceleration signal generated by the actuator. The recorded signal was analyzed to verify the vibration frequency (30 Hz) and confirm stable oscillatory output. These tests ensured that the actuator produced consistent vibration during operation. The gait phase detection algorithm was evaluated during pilot walking trials before the main experiment. Detected gait events (mid-swing, initial contact, and toe-off) were visually compared with video recordings of the walking trials to confirm correct identification of gait phases.

All primary outcome assessments were conducted in the laboratory during two sessions: a baseline session before the intervention and a follow-up session after four weeks of brace use. Importantly, participants performed all assessments without wearing the knee brace, ensuring that measurements reflected participants’ natural performance and were not influenced by immediate mechanical effects of the LMV device (Figures 3 and 4).

Figure 3.

Equipped the off-loading knee brace with a customized LMV system.

Figure 4.

The vibrator-to-orthosis connection piece is shown. This piece, designed using SolidWorks software and manufactured by a 3D printer, can be attached to the fabric part of the load-bearing orthosis using 2 rivets below and above.

Safety and adverse events

All participants were monitored for adverse events throughout the 4-week intervention period. Adverse events included skin irritation, discomfort, pain exacerbation, or any other unexpected reaction related to brace use or LMV application. Participants with sensitive skin experienced mild redness or discomfort at the upper or lower edges of the brace, especially over bony prominences such as the tibial crest. To minimize irritation, participants were instructed to wear a thin layer of clothing under the brace, preventing direct skin contact with the brace. The brace was checked for proper fit and suspension, and minor adjustments were made if any slippage or pressure points were noted. All skin reactions were recorded, and participants were advised to report any discomfort immediately.

Outcome measures

In this study, clinical outcomes were considered primary endpoints and included pain and total WOMAC scores, which were assessed using the visual analog scale (VAS) and the validated Persian version of the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC). The WOMAC questionnaire consists of three subscales: pain (5 items, scored 0–20), stiffness (2 items, scored 0–8), and physical function (17 items, scored 0–68), with total scores ranging from 0 to 96. Higher scores indicate worse symptoms. It has demonstrated high reliability (Cronbach’s alpha >0.80) and validity in Persian-speaking populations, making it suitable for capturing patient-reported outcomes in KOA studies. As a secondary clinical outcome, pain intensity was measured using a 10-cm VAS, where participants indicated their pain level from 0 (no pain) to 10 (worst imaginable pain), providing a simple yet sensitive measure of subjective pain.

Biomechanical parameters were assessed as primary outcomes and included the first and second peak knee adduction moments (KAM1, KAM2) and the adduction moment impulse. Secondary biomechanical outcomes included step length, cadence, and knee range of motion (ROM). All biomechanical measures were obtained pre- and post-intervention using a three-dimensional motion analysis system, allowing objective quantification of joint loading and movement patterns. Shapiro–Wilk tests confirmed normality of the data, justifying the use of parametric analyses. Within-group comparisons were performed using paired t-tests, while between-group effects were evaluated using repeated-measures analysis of variance (ANOVA), controlling for relevant covariates. This structured approach, with clear distinction between primary and secondary outcomes, allows for precise interpretation of intervention effects and aligns with best practices in reporting clinical trial results.

Evaluation procedure

Eligible participants were informed of the study procedures and enrolled after providing written informed consent. Following legal approvals for the clinical trial, participants were randomly assigned to either the conventional off-loading knee brace group (control) or the LMV-equipped brace group (intervention).

At baseline, participants completed the WOMAC questionnaires and underwent gait analysis using motion capture cameras and force plates, without wearing their assigned knee brace. Participants refrained from physical activity before testing to avoid fatigue-related interference.

Before delivering the LMV-equipped brace to participants, the device was initialized and tested in the laboratory. A brief familiarization walk was performed to ensure the system could detect the gait pattern and deliver vibration correctly in synchrony with gait phases. After this initial check, participants took the device home and used it independently for three 30-minute sessions per week over four weeks. Adherence and device usage at home were based on participants’ self-reports, which represents a limitation regarding precise dose monitoring.

Over the following four weeks, participants engaged in structured walking exercises. In the intervention group, the brace provided muscle vibration stimulation synchronized with gait phases during training sessions only, while participants continued to wear the brace without activating LMV during the rest of the day. In the control group, participants used the conventional off-loading knee brace for the same duration and walking sessions, but without any vibration stimulation. Weekly check-ins were implemented for both groups to monitor adherence and address any issues.

After four weeks, all assessments—including gait, pain, and WOMAC scores—were repeated without wearing the knee brace or LMV device. A single-blinded design was maintained throughout the trial, with separate scheduling for the two groups to prevent communication and minimize bias.

Marker placement

Anthropometric measurements were used to identify the joint locations or the center of mass. The use of specialized measurement instruments for direct anthropometric data collection greatly improves the accuracy with little time investment, even though these data can be estimated from primary parameters like a person’s height and mass. Weight, height, the length and width of the lower limbs were all measured during this procedure. The distance measured along a path that passes through the knee joint between the lateral malleolus and the anterior superior iliac spine (ASIS) was determined to be the true length of the lower limb. The distance between the medial and lateral femoral epicondyles was used to define knee width, and the distance between the medial and lateral malleoli was used to define ankle width. A caliper was used to measure each width. ³³ These precise measurements ensured accurate marker placement and model calibration for biomechanical analysis.

The motion of each body segment was tracked using a Plug-In-Gait lower body model. A total of 25 reflective markers were placed on each participant. 16 markers were placed on bony landmarks as per the standard Plug-In-Gait model, with two additional markers on the anterior thigh, two on the anterior shank to improve marker location estimation, and five markers on the trunk to estimate the center of mass. ³⁴ Marker trajectories were captured at 100 Hz using an 8-camera system, synchronized with force plate data for inverse dynamics calculations.

Data analysis (statistics) section

Statistical analyses were performed using SPSS version 22. Continuous variables were summarized with medians and interquartile ranges (IQR), while categorical variables were presented as counts and proportions. The normality of data distribution was assessed using the Shapiro-Wilk test. An independent samples t-test compared functional activity levels, electrical muscle activity levels, lower limb joint angles, and joint torques between the two groups. A paired samples t-test compared these variables within each group pre- and post-intervention. The Pearson correlation coefficient was calculated to determine relationships between demographic variables and dependent variables. Non-parametric versions of these tests were applied as needed. Additionally, ANCOVA was used for inter-group comparisons post-intervention, adjusting for baseline values to account for any initial differences.

Results

Participant clinical characteristics

The clinical and demographic characteristics of the patients included in the study are presented in Table 1. No significant differences were observed between the LMV-Brace and Conventional Brace groups in age, height, weight, or BMI (all p > 0.05), indicating comparability at baseline.

Table 1.

Patient demographic information.

​	Experimental	Control	P value
N	8	8	​
Age	58.9 (6.36)	63.4 (7.76)	0.244
Height	166.7 (11.32)	161.0 (6.23)	0.239
Weight	77.3 (5.71)	75.0 (14.24)	0.702
BMI	28.0 (2.53)	28.8 (4.50)	0.673

Clinical and biomechanical outcomes

Primary outcomes included WOMAC and VAS, whereas secondary outcomes comprised biomechanical measures (stride length, cadence, KAM1, KAM2, impulse, and knee ROM). Table 2 shows the mean (M), standard deviation (SD), and 95% confidence intervals (CI) for all outcomes at baseline and after the 4-week intervention.

Table 2.

Mean (SD) of aforementioned parameters.

Variable	Group	Time	Mean ± SD	95% CI
VAS	Control	Before	2.1 ± 1.21	0.48 – 3.72
		After 4 weeks	1.4 ± 0.98	0.05 – 2.75
VAS	Experimental	Before	7.3 ± 2.29	5.13 – 9.47
		After 4 weeks	6.6 ± 0.84	5.62 – 7.58
WOMAC (total scores)	Control	Before	26.0 ± 13.61	13.06 – 38.94
		After 4 weeks	17.6 ± 9.60	8.22 – 26.98
WOMAC (total scores)	Experimental	Before	20.5 ± 8.14	13.77 – 27.23
		After 4 weeks	18.5 ± 8.75	11.47 – 25.53
Stride Length (m)	Control	Before	1.10 ± 0.16	1.02 – 1.18
		After 4 weeks	1.08 ± 0.14	1.01 – 1.15
Stride Length (m)	Experimental	Before	0.92 ± 0.12	0.86 – 0.98
		After 4 weeks	0.97 ± 0.11	0.91 – 1.03
Cadence (steps/min)	Control	Before	47.6 ± 4.30	44.87 – 50.33
		After 4 weeks	50.4 ± 3.97	47.75 – 53.05
Cadence (steps/min)	Experimental	Before	85.0 ± 6.86	80.50 – 89.50
		After 4 weeks	87.5 ± 8.50	81.71 – 93.29
KAM1 (Nm/kg)	Control	Before	555.6 ± 34.05	536.41 – 574.79
		After 4 weeks	548.2 ± 78.08	499.53 – 596.87
KAM1 (Nm/kg)	Experimental	Before	474.8 ± 175.6	371.61 – 577.99
		After 4 weeks	509.3 ± 156.19	411.88 – 606.72
KAM2 (Nm/kg)	Control	Before	513.2 ± 88.51	442.45 – 583.95
		After 4 weeks	488.3 ± 100.54	409.83 – 566.77
KAM2 (Nm/kg)	Experimental	Before	415.5 ± 171.90	303.62 – 527.38
		After 4 weeks	445.6 ± 127.95	353.36 – 537.84
Impulse (Nm·s)	Control	Before	0.29 ± 0.03	0.274 – 0.306
		After 4 weeks	0.25 ± 0.04	0.233 – 0.267
Impulse (Nm·s)	Experimental	Before	0.27 ± 0.13	0.183 – 0.357
		After 4 weeks	0.29 ± 0.10	0.214 – 0.366
Knee ROM	Control	Before	43.9 ± 5.99	40.25 – 47.55
		After 4 weeks	43.3 ± 6.03	39.63 – 46.97
Knee ROM	Experimental	Before	24.0 ± 5.96	20.34 – 27.66
		After 4 weeks	20.9 ± 6.13	17.17 – 24.63

Statistical methods

Table 3 summarizes the intra-group comparisons (pre-test vs. post-test), as well as the inter-group comparisons. Shapiro-Wilk tests confirmed the normal distribution of all variables. Accordingly, paired t-tests were used for intra-group comparisons, and ANCOVA (adjusting for baseline values) was used for inter-group comparisons. Partial eta squared is reported as a measure of effect size. Multiple comparisons were limited to primary outcomes to reduce Type I error; secondary outcomes are reported descriptively.

Table 3.

Intergroup and intragroup, with and without interventions (control).

​	Intragroup comparison		Intragroup comparison
	p value 1	p value 2	p value 3	Partial eta squared
Vas	0.011	0.140	0.246	0.102
Womac	0.001	0.024	0.552	0.028
Stride Length	0.057	0.531	0.441	0.046
Cadence	0.460	0.036	0.125	0.171
KAM1	0.230	0.817	0.599	0.022
KAM2	0.329	0.479	0.623	0.019
IMPULSE	0.533	0.022	0.050	0.257

p-value 1: Intragroup comparison between pre- and post-intervention in the Experimental group.

p-value 2: Intragroup comparison between pre- and post-intervention in the Control group.

p-value 3: Intergroup comparison between the Experimental group and Control group between pre- and post-intervention.

Intra-group comparisons

Conventional off-loading brace

Significant improvements were observed in functional status (WOMAC: p = 0.024, −32.31%, 95% CI [−38.9, −25.7]) and KAM impulse (p = 0.022, −7.68%, 95% CI [−12.1, −3.3]). Cadence also showed a small, significant increase (p = 0.036, +5.88%, 95% CI [1.2, 10.5]). Pain (VAS: p = 0.140, −33.33%, 95% CI [−40.0, −26.6]) and other biomechanical variables, including stride length, PKAM1, PKAM2, and knee ROM, did not reach statistical significance.

Equipped the off-loading knee brace with a customized LMV system

Significant improvements were observed in WOMAC (p = 0.001, −9.76%, 95% CI [−15.4, −4.1]) and VAS scores (p = 0.011, −9.59%, 95% CI [−14.2, −4.9]). Stride length showed a trend toward improvement (p = 0.057, +5.43%, 95% CI [−0.2, 11.1]). Changes in KAM impulse (p = 0.533, −24.79%), PKAM1 (p = 0.230, −18.89%), PKAM2 (p = 0.329, −14.3%), cadence (p = 0.460, +2.94%), and knee ROM (p = 0.392, −12.92%) were not statistically significant. However, numerical reductions in kinetic variables suggest a trend toward greater mechanical unloading, accompanied by rapid clinical improvements.

Intergroup comparisons

Equipped the off-loading knee brace with a customized LMV system in comparison with a conventional brace

As shown in Table 3, the LMV-Brace group exhibited a trend toward improved functional outcomes compared to the Conventional Brace group. Statistically significant differences between groups were limited. Specifically, no significant differences were observed for pain intensity measured by VAS (p = 0.246), cadence (p = 0.125), stride length (p = 0.441), peak knee adduction moments PKAM1 (p = 0.599) and PKAM2 (p = 0.623), or frontal plane knee ROM (p = 0.048). Partial eta squared values indicate small to moderate effect sizes, and 95% confidence intervals for inter-group differences are provided in Table 3.

Despite the lack of statistical significance for most measures, the LMV-Brace group demonstrated numerically greater reductions in KAM impulse (p= 0.050, −24.79% vs. −7.68%) and knee ROM compared to the Conventional Brace group. These results suggest that the LMV-Brace may provide meaningful mechanical unloading and contribute to functional improvements while changing pain and other spatio-temporal gait.

Discussion

This study aimed to evaluate the effect of equipping an off-loading knee brace with a customized LMV system compared to a conventional brace over four weeks in patients with medial compartment KOA. The results indicate that the LMV-equipped brace provides superior biomechanical and clinical effects through a distinct mechanism of action. Specifically, the integration of LMV enhances mechanical unloading by augmenting neuromuscular responses, leading to more efficient joint protection and symptom management, as reflected by notable reductions in KAM impulse (−24.79% vs. −7.68%, F = 4.66; p = 0.050) and ROM (F = 4.747; p = 0.048).

The greater reduction in KAM impulse observed with the LMV brace likely results from vibration-induced stimulation of muscle spindles during the stance phase, promoting improved quadriceps activation and load distribution. This is supported by evidence on vibration-induced reflex facilitation.^35,36 Similarly, the improvement in knee ROM suggests enhanced proprioception and periarticular muscle co-contraction, refining motor control and limiting excessive joint movements that could exacerbate cartilage wear. These findings align with prior studies demonstrating vibration’s role in cortical excitability and spinal reflex modulation. ³⁷

Intra-group analyses further highlight the synergistic effects of LMV: the LMV group exhibited rapid improvements in WOMAC (p = 0.001) and VAS (p = 0.011), suggesting that LMV may counteract arthrogenic muscle inhibition by restoring excitatory signals, complementing the passive unloading provided by the brace. ³⁷ In contrast, the Conventional Brace group reduced KAM impulse (p = 0.022) and increased cadence (p = 0.036) but did not achieve significant pain relief (VAS, p = 0.140), indicating slower clinical improvements. The accelerated response observed with the LMV brace may be attributed to analgesic mechanisms such as gate control modulation, endorphin release, and improved local blood flow, which alleviate muscle tension and allow more natural gait patterns.^37,38

No significant inter-group differences were observed in variables such as stride length or peak KAM moments, potentially due to the short duration of the intervention. Initial neuromuscular gains may require longer periods to translate into statistically significant differences. The Conventional Brace group’s improvement in cadence, consistent with Schmalz et al., ³⁹ reflects adequate unloading to enhance gait rhythm, whereas the LMV group’s trend toward increased stride length (p = 0.050) suggests that neuromuscular modulation can restore more natural gait patterns by mitigating fear-avoidance behaviors. ³⁶ Pain’s influence on kinetics—such as increased adduction moments and reduced speed/ROM—appears to be more effectively addressed by LMV, consistent with findings by Ebrahim Abadi et al. ³⁷ demonstrating immediate reductions via mechanoreceptor stimulation and metabolic improvements.

Systematic reviews by Barati et al. ⁴⁰ corroborate our findings, highlighting LMV’s efficacy in reducing pain and stiffness and improving function through enhanced reflex activity. ³⁶ Conflicting evidence regarding vibration-induced force effects^35,40 likely reflects variations in vibration frequency and duration; our 30 Hz protocol provided balanced activation without inducing fatigue. Compared to Xavier Robert et al., ⁴¹ the 24.79% KAM reduction observed here emphasizes LMV’s dynamic modulation beyond conventional static orthotics, although some combined interventions may yield more modest improvements.^42,43

Adherence remains a critical factor in brace efficacy, with discomfort often limiting long-term use (28% at one year). ⁴¹ The LMV brace’s rapid relief may foster better compliance through positive reinforcement.

Conclusion

In summary, the LMV-offloading knee brace significantly improves clinical outcomes for patients with medial compartment KOA, producing a 24.79% reduction in KAM impulse and enhancing knee joint range of motion. This brace provides rapid relief from pain and functional limitations while promoting adherence to orthotic use due to its effectiveness. Overall, combining LMV-equipped braces with conventional treatments appears to be the most beneficial strategy for managing symptoms and improving quality of life in patients with medial knee osteoarthritis.

Limitation

Study limitations include the small sample size, short intervention duration, absence of a no-treatment control group, and reliance on self-reported adherence for home-based LMV use. Future studies should involve larger cohorts, longer follow-up periods, and objective monitoring of device usage to better evaluate dose–response effects.

ORCID iDs

Mahsa Zangi https://orcid.org/0009-0008-0248-6434

Mobina Khosravi https://orcid.org/0000-0003-2947-0834

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.*