Visual-feedback Balance Platform Training Combined with Conventional Rehabilitation for Balance Function after Total Knee Arthroplasty

Summary

The goal of the presented protocol is to improve balance function after total knee arthroplasty by combining conventional rehabilitation training with the visual-feedback balance platform training.

Abstract

The aim of the protocol is to investigate the feasibility and effectiveness of combining conventional rehabilitation with visual-feedback balance platform training in elderly patients undergoing primary unilateral Total Knee Arthroplasty (TKA). The method shows a single-blind, randomized controlled trial that enrolled 60 patients undergoing primary unilateral TKA (23 males, 37 females). Participants were randomly assigned to a control group (n=30; mean age 69.60 ± 2.98 years) or an observation group (n=30; mean age 70.87 ± 4.26 years). The control group received conventional rehabilitation, while the observation group received conventional rehabilitation plus visual-feedback balance platform training. The visual feedback intervention commenced in the second postoperative week, conducted 5 times per week for 20-30 min per session, over a 5-week period. Balance function was evaluated using platform-specific metrics: Length of Movement Trajectory (LMT) and Area of Movement Ellipse (AME). Functional mobility was assessed via the Timed Up and Go Test (TUGT), Single-Leg Stance (SLS), Timed Sit-to-Stand (STS), and Timed Stair Test (TST). Data were analyzed using independent and paired t-tests with a significance level of p < 0.05. After the 6-week program, both groups showed significant improvements across all parameters compared to baseline (p < 0.05). However, the observation group exhibited significantly superior outcomes compared to the control group. Specifically, static stability metrics (LMT, AME) and TUGT performance were significantly better in the observation group (p < 0.01). Similarly, the observation group demonstrated significant improvement in SLS, STS, and TST (p < 0.05) compared to controls. Integrating visual-feedback balance platform training with conventional rehabilitation significantly enhanced static and dynamic balance function in elderly TKA patients, offering superior efficacy to conventional rehabilitation alone.

Introduction

Knee osteoarthritis (KOA) is a prevalent degenerative musculoskeletal disorder, affecting approximately 23% of individuals aged 40 and older worldwide, and is characterized by chronic pain, limited joint mobility, impaired balance and proprioception, abnormal gait, and overall functional decline¹. Despite advances in diagnostic and therapeutic techniques, KOA remains a significant public health challenge due to the aging population and rising obesity rates. Total knee arthroplasty (TKA) is an effective intervention for end-stage KOA, alleviating pain and improving function². However, evidence suggests that many patients continue to experience functional limitations postoperatively, including an average reduction in walking speed of 18% and up to 51% reduction in stair-climbing speed³. Notably, approximately 40% of TKA patients remain at high risk of falls, comparable to preoperative levels⁴, and exhibit significant difficulties in functional activities such as squatting, lateral movement, and turning⁵.

Postoperative rehabilitation following total knee arthroplasty (TKA) is crucial for restoring function, alleviating pain, and enhancing the patient's independence in daily living⁶. Key goals include maximizing range of motion (ROM) and strengthening peri-articular muscles, ultimately improving activities of daily living (ADL)⁷. Knee stability and balance control-which rely on intact proprioception and efficient neuromuscular control-are critical for optimal outcomes after TKA.

The visual-feedback balance platform training system, a visual feedback-based training device, combines quantitative balance assessment with targeted exercise⁸. This protocol targets elderly post-TKA patients with partial weight-bearing capability. However, to ensure safety and feasibility, it excludes patients with severe cognitive impairment (MMSE < 24) or those unable to stand independently. By providing real-time visual feedback on center of pressure (COP) movements, the system guides patients through multidirectional, task-oriented exercises, designed to reinforce the sensorimotor control loop. Compared to conventional subjective rehabilitation, this approach enhances motor learning by providing an external focus of attention for immediate error correction. While visual feedback training has demonstrated efficacy in neurological rehabilitation populations, such as stroke or Parkinson's disease, its specific application and standardized protocol for post-TKA balance recovery warrant further investigation. Therefore, the objective of this protocol is to demonstrate a standardized visual-feedback training method that aims to facilitate quantifiable improvements in static and dynamic balance stability for TKA patients, addressing the limitations of conventional subjective rehabilitation.

Protocol

This study protocol received approval from the ethics committee of the Affiliated Hospital of Xuzhou Medical University (Approval Number: XYFY2024-KL316-01), and informed consent was obtained from all participants.

1 . Study design and participant selection

1. Use a randomized controlled trial, employing single-blinding and a control group, to assess the feasibility and effectiveness of the program. Select a total of 60 patients who underwent primary unilateral total knee arthroplasty (TKA) at the Department of Orthopedic Surgery, Xuzhou Medical University Affiliated Hospital, from January 2024 to June 2025 as study subjects.
2. Randomly assign patients to a control group (n=30) and an observation group (n=30) using a random number table. Conceal the allocation using sequentially numbered, opaque, sealed envelopes. Details of instruments can be found in the Table of Materials.
3. Eligibility screening
   1. Inclusion criteria: Include patients undergoing unilateral TKA for the first time; age ≥65 years; alert, with normal cognitive function (Mini-Mental State Examination [MMSE] score > 24), no hearing impairment, and capable of effective communication; and have signed an informed consent form and committed to participating in follow-up throughout the study.
   2. Exclusion criteria: Exclude patients with severe failure of vital organs such as the heart, liver, lungs, or kidneys; patients diagnosed with anxiety disorder, depression, or those who have been taking psychiatric medications long-term; participants concurrently enrolled in other interventional clinical studies; and participants with other neurological conditions that may significantly impair balance function (e.g., stroke, severe peripheral neuropathy, etc.).
4. Randomization and blinding
   1. Randomly assign patients to a control group (n=30) and an observation group (n=30) using a standard random number table method to ensure full randomization and eliminate selection bias. Conceal the allocation using sequentially numbered, opaque, sealed envelopes. Note that this study employs single blinding (assessors blinded to group allocation; Figure 1).

2 . Baseline assessments

NOTE: Before the experiment, the following assessments were completed by an experienced physiotherapist for all patients: LMT (mm), AME (mm²), the Timed Up and Go Test (TUGT), Single-Leg Stance Time (SLS), Timed Sit-to-Stand Test (STS), and Timed Stair Test (TST).

1. Visual-feedback balance platform training system assessment: Use the visual-feedback balance platform training system to conduct tests in a double-foot standing position. Record COP parameters in both open-eye and closed-eye states: LMT (mm) and AME (mm²). Each test trial lasted for 30 s, repeat test 3x, and take the average value. For LMT, a greater length means a poorer balance function. For AME, a larger area means poorer balance function. A smaller value of the 2 parameters indicates a better balance function.
   NOTE: The open-eye condition assesses balance with full sensory integration, while the closed-eye condition isolates proprioceptive and vestibular contributions by eliminating visual input.
2. TUGT: Instruct the patient to stand up from a standard chair (seat height approximately 46 cm), walk in a straight line for 3 m, turn around, return, and sit back down. Allow one practice attempt. Conduct the formal test 2x and record the average time (s). Provide no physical assistance during the test.
3. SLS: Assesses single-leg static balance ability. Ask the patient to stand with both hands naturally hanging at the sides. One leg is bent at the knee and lifted off the ground by approximately 15-20 cm, with both legs avoiding contact. Record the duration (in s) during which the patient maintains stability on one leg, with the raised foot not touching the ground or the supporting leg not moving. Test the single-leg stance time of the surgical side leg with the eyes open. Perform the test 2x and use the best value for analysis. A larger value indicates better balance function.
4. TST: Record the total time (in s) required for the patient to safely and quickly ascend and descend four steps (each step approximately 17-18 cm high). This test reflects functional mobility, strength, and dynamic balance. Perform the test 1x. A smaller value indicates a better balance function.

3 . Conventional rehabilitation training after TKA

NOTE: Patients in both the control and observation groups received this training.

1. Stage 1 (1-7 days after surgery)
   1. Prepare ankle pump exercise with 5 s per repetition, 5x-10x per set, 4-5 sets per day.
   2. Joint mobility training: Perform continuous passive motion device-assisted training with knee flexion angle restriction < 90°, 20 min per session, once per day. Perform active knee extension exercises 5x-10x per set, 2-3 sets per day.
   3. Perform isometric contraction exercises for the gluteal muscles, quadriceps, hamstrings, etc., 5x-10x per set, 2-3 sets per day.
   4. Perform breathing exercises with deep breathing, coughing exercises, combined with upper limb stretching and chest expansion exercises at 5x-10x per set, 2-3 sets per day.
   5. Transfer training: Within the range of tolerance, use a walker to perform progressive weight-bearing transfer exercises at 10 min per session, 1x per day.
   6. Physical therapy: Perform Neuromuscular Electrical Stimulation (NMES) treatment, with electrodes placed on the medial and lateral thigh muscles to promote quadriceps activation at 20 min per session, 1x per day.
   7. Pain and swelling management: Apply ice to the affected knee after training at 25 min per session, 1x per day.
2. Stage 2 (2-6 weeks after surgery)
   1. Joint mobility training for patellar mobilization: Perform active knee flexion and extension exercises with gradual transition to power cycling training at 5-10 min per day.
   2. Muscle strength training: Perform isometric exercises for the gluteal muscles, quadriceps, hamstrings, gradually progressing to resistance training that combines closed-chain movements (such as micro squats and sitting to stand) with open-chain movements (such as sitting knee extensions) at 10x-20x repetitions per set, 2-3 sets per day.
   3. Standing balance training: Maintain static stability while standing on both feet (initially, moderate external support may be provided to the affected knee). Train to transition from open-eye to closed-eye state at 5-10 min per day.
   4. Basic dynamic balance training: Perform standing on one leg (healthy side/affected side); shifting the center of gravity forward, backward, left, and right; walking sideways; marching in place at 10-20 min per day.

4 . Visual-feedback balance platform training

NOTE: Only the observation group received this training.

1. Training start time and duration: Schedule intervention initiation for the 2^nd postoperative week based on safety considerations and standard clinical rehabilitation phases. Reserve the 1^st week for acute wound healing, edema management, and DVT prevention. Active Visual-feedback balance platform training requires weight-bearing, commence it in week 2 (aligning with the start of Stage 2 conventional rehabilitation) to ensure patient safety and tolerance. Conduct training 5x per week, with each time lasting 20-30 min (including rest periods between sets), for a total of 5 weeks.
2. Create an individualized profile for each patient, including the patient's full name, age, hospitalization identification number, diagnosis, and other relevant medical details.
3. Explain and demonstrate the correct operation method for each program to ensure the patient fully comprehends the purpose of the training.
4. Choose an appropriate training plan based on the patient's balance function and physical tolerance. Set parameters for each program, such as training duration, difficulty level, etc.
5. Increase the difficulty of training from simple to complex, and the duration of training from short to long. Assess pain using the visual analog scale (VAS) and fatigue using the Borg Rating of Perceived Exertion (RPE). If the VAS score > 5 or RPE > 13 during the training process, reduce the training intensity or pause the session immediately.
6. Instruct the patient to stand in front of the device (Figure 2) and stand on the balance board with both feet. Ask the patient to take the functional standing posture: The inner ankle of the foot passes through the red horizontal line (3 cm below the A3-A7 horizontal axis), both feet are parallel to the shoulders and hips, with the heel of the foot on the same horizontal axis (Figure 3).
7. Conduct the Visual-feedback balance platform training as follows.
   1. Stage 1 (approximately weeks 1-2 of training): Focus mainly on static balance control, to focus on enhancing patients' static stability function.
   2. Training mode: Use static stability training mode. Turn on the Visual feedback balance training system and follow the steps below.
   3. Access the evaluation menu, select Static Stability Evaluation. Configure training mode and parameters, and click Options to enter the settings interface. In the general settings, under the location option, select Static. Under the time option, select Training (s).
   4. In parameter selection, choose Single from the continuous options. Set training range based on the results of the patient's static stability assessment. Define a circular training area with the origin of the coordinate system as its axis and the average sway amplitude in the anterior-posterior (Y-axis) and lateral (X-axis) directions as its diameter.
   5. Generate training area by clicking the Corresponding scale on the X-axis or Y-axis, and the system will automatically generate a circular range centered at the origin with the set value as its radius, displayed within the coordinate system.
   6. Begin the training. Have the patient stand on the balance board and press the Start button. The patient controls the on-screen cursor (+) by shifting their body weight, keeping it within the circular area. If the cursor moves outside this area, the system emits a beep-beep alert. Set the parameters at 3-5 sets per session, 3 min per set, with 1 to 2 min of rest between sets. Increase the difficulty of training by reducing the circle's area.
   7. Stage 2 (Weeks 2-3 of training): Perform dynamic postural stability to transition from static control to dynamic stability using Bipedal dynamic stability mode as described below.
   8. Turn on the Visual feedback balance training system and follow the steps below.
   9. Access the evaluation menu, click the Stability Limit button in the menu. Configure training mode, which includes two options: static and dynamic training. Static training locks the balance board in place, preventing movement. Dynamic training unlocks the balance board, requiring adjustment of its firmness level within a range of 1 to 50.
   10. Begin the training. Ask the patient to stand with both feet on a balance platform and control the cursor (+) on the screen by shifting the patient's body weight to touch randomly appearing flashing target points (Figure 4). Set parameters at 8-10 sets per session with 1-2 min rest between sets. If 75% of the reference value is achieved in training results, use this criterion for advancing to the next stage.
   11. Stage 3 (Weeks 3-5 of training): Carry out task-oriented complex control to enhance neuromuscular coordination through complex, task-specific activities. Use Gamified scenarios (e.g., Simulated Skiing, Shooting Range, Flight Control, etc.).
   12. Training method: Ask the patient to perform dynamic weight shifts to navigate obstacles or capture targets within the interactive game environment. Open the Visual feedback balance test and training system, then select the training method from the game menu.
   13. Click the SKI module. The system will display the training settings page. Configure the following three parameters in sequence: Training Mode: Static = Lock the balance board, Dynamic = Unlock the balance board; Level of difficulty = Divided into three levels: Easy, Medium, and Hard.
       Includes three options: waist, single leg, and seated. Begin with static mode, gradually progressing from easy to difficult levels. For the training area, select the waist, with both feet positioned on the balance board. During training, patients control the in-game character by shifting their body weight, aiming to avoid touching the side rails and successfully reaching the finish line.
   14. Click the Shooting range module. The pop-up page functions similarly to the ski slope module. After selecting the appropriate game training parameters, begin the training. During the training session, the patient eliminates randomly appearing targets by shifting their body weight.
   15. Set parameters to 1-2 different task modules per session. Once patients can complete training from simple to difficult tasks in static mode, progress them to dynamic mode training from simple to difficult tasks. Measure the advancement in ski game training by the patient's ability to control their body's center of gravity to prevent the character in the game interface from touching the side rails. Measure the advancement in shooting range game training by the patient's ability to control their center of gravity to completely hit the target in the game interface.
8. As the patient's function recovers, based on the patient's performance and the physiotherapist's assessment, regularly recombine the training plan and adjust the difficulty and duration of training.

5 . Follow-up procedures

1. Following 6 weeks of treatment, re-evaluate all patients, including LMT and AME in both open-eye and closed-eye states, TUGT, SLS, STS, and TST. These reassessments were conducted by the same physical therapist.

6 . Data analysis

1. Analyze all data using SPSS 20.0. Normality Test: Use the Kolmogorov-Smirnov statistic to test the normality of all variables.
2. Baseline comparison: Calculate demographic and clinical characteristic differences at baseline using one-way analysis of variance (ANOVA) or the χ² test.
3. Outcome analysis: Use paired-sample t-tests to compare within-group changes (pre- vs. post-intervention). Use independent-samples t-tests to compare the control and observation groups after the intervention. In addition to p-values, calculate the mean difference (MD), 95% confidence intervals (95% CIs), and effect sizes (Cohen's d) to quantify the magnitude and precision of the treatment effects.
4. Significance: Consider a p-value < 0.05 as statistically significant.

Results

A total of 60 patients were enrolled and randomly assigned to either the control or observation group (Table 1). Baseline characteristics were strictly controlled. There was no statistically significant difference between the two groups for sex, age, disease duration, or lesion side (p> 0.05), indicating that the two groups were comparable at baseline (Table 1).

Throughout the study, 10 adverse events (e.g., falls,...

Discussion

The results of this study indicate that integrating the visual-feedback balance platform training into the conventional rehabilitation for a 6-week training program can significantly improve balance function in patients who have undergone unilateral TKA surgery, with clinical efficacy superior to conventional rehabilitation alone.

Balance function refers to the advanced neuromuscular control ability of the human body to maintain center of gravity stability, precise control posture, and prevent...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Lower limb joint rehabilitation device	Xuzhou Kangmai Medical Equipment Technology Co., Ltd	030805000067
Pro-kin system	TecnoBody Company, Italy	030899000212