Next Article in Journal
The Seismic Performance of Earthen Historical Buildings in Seismic-Prone Regions: The Church of Santo Tomás de Aquino in Rondocan as a Complex Example
Previous Article in Journal
Seaweed Pelvetia canaliculata as a Source of Bioactive Compounds for Application in Fried Pre-Coated Mackerel (Scomber scombrus) Fillets: A Functional Food Approach
Previous Article in Special Issue
Systematic Review of Mecanum and Omni Wheel Technologies for Motor Impairments
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Applied Sciences
Volume 15
Issue 13
10.3390/app15137620
applsci-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Effect of Virtual-Reality-Based Training, Including Preceding Trunk Stabilization Education, on Postural Control and Balance in Patients with Stroke: A Randomized Controlled Trial

by
SeongMin Lee
and
JongEun Yim
*
Department of Physical Therapy, Graduate School of Sahmyook University, Seoul 01795, Republic of Korea
*
Author to whom correspondence should be addressed.
Appl. Sci. 2025, 15(13), 7620; https://doi.org/10.3390/app15137620
Submission received: 8 May 2025 / Revised: 30 June 2025 / Accepted: 1 July 2025 / Published: 7 July 2025
(This article belongs to the Special Issue Assistive Technologies for Rehabilitation: Challenges and Applications)
Browse Figures
Versions Notes

Abstract

This study investigated the effects of virtual reality (VR)-based training combined with preliminary trunk stabilization education on postural control and balance in stroke patients. A single-blind randomized controlled trial enrolled 30 participants, randomly divided into a trunk stabilization group (n = 15) and a control group (n = 15). The trunk stabilization group engaged in 10 min of trunk stabilization education followed by 20 min of VR-based training, three times weekly for three weeks. The control group participated only in VR-based training. Outcomes were assessed using the Korean Trunk Impairment Scale (K-TIS), Postural Assessment Scale for Stroke (PASS), Berg Balance Scale (BBS), limit of stability (LOS), and center of pressure (COP) measurements. Both groups significantly improved in all measured outcomes post-intervention (p < 0.05). Notably, the trunk stabilization group exhibited significantly superior improvements in the K-TIS, PASS, BBS, LOS, and COP path length compared to the control group (p < 0.05). These results highlight the enhanced effectiveness of integrating trunk stabilization education with VR-based training, suggesting that it not only yields statistically significant improvements but also provides clinically meaningful benefits for functional postural control and balance recovery in stroke rehabilitation.

1. Introduction

Stroke is a neurological disorder caused by a disruption of blood supply because of cerebral infarction or hemorrhage [1]. Patients with stroke typically present with motor impairments in the upper and lower limbs, sensory deficits, and language disorders. Trunk dysfunction is frequently observed and significantly affects functional outcomes [2]. Trunk muscle function is essential for sitting balance, mobility, walking, and activities of daily living in patients with stroke [3]. Compromised trunk function often results in an asymmetrical posture, adversely affecting postural and equilibrium responses and thereby impairing balance ability [4].
Improving balance requires appropriate postural control strategies to maintain the body’s center of pressure (COP) with minimal sway [5]. Postural control is crucial not only for responding to unexpected body movements but also for initiating voluntary movements. Trunk stabilization plays a key role in the execution of these strategies [6]. Trunk stabilization education has been shown to enhance trunk control, dynamic sitting and standing balance, walking ability, and daily functioning in patients with stroke [7]. Additionally, it contributes to improved weight-bearing, muscle strength, and postural control, facilitating better dynamic balance and mobility during stroke recovery [8].
Virtual reality (VR)-based training is a therapeutic intervention that can enhance balance following stroke. This method offers a realistic and immersive environment by integrating multisensory stimuli (visual, auditory, tactile, and somatosensory inputs) and promotes functional recovery through goal-directed motivational tasks [9]. According to Lewis and Rosie [10], VR creates a rich environment for acquiring new skills and can help resolve functional issues in patients with stroke while increasing engagement and enjoyment. The motivational aspect of VR training makes it more appealing and engaging than traditional therapeutic exercises and leads to increased concentration and adherence [11,12,13]. Lee and Bae [14] demonstrated that VR-based video games offering visual and auditory feedback effectively improve postural control and gait ability in patients with chronic stroke. Similarly, Alhwoaimel et al. [15] reported improved trunk control and balance with VR-based video game interventions.
However, goal-oriented tasks, such as those in VR training, may lead patients to focus more on task completion than on movement quality, potentially reinforcing abnormal compensatory patterns and resulting in inefficient motor recovery [16]. Although functional recovery is the primary goal of post-stroke rehabilitation, compensatory strategies are often used instead of normal motor function. The repeated use of such compensation may result in chronic joint pain and dysfunctional movement patterns [16]. Patients with stroke often exhibit altered muscle recruitment sequences due to impaired selective motor control, making them susceptible to abnormal compensation during VR training [17]. This compensation may worsen trunk dysfunction and postural asymmetry, compromising balance responses and control mechanisms [4].
To address these limitations, Magee [18] emphasized the importance of trunk stabilization in their systematic review. Maintaining a neutral spine and co-contraction of the trunk muscles are critical for postural control in patients with stroke. Trunk stabilization education is effective in restoring symmetrical muscle thickness of the transversus abdominis and enhancing internal oblique activation, thereby improving trunk stability and supporting optimal movement patterns and muscle recruitment [19,20].
A commonly used clinical method for trunk stabilization is the abdominal drawing-in maneuver (ADIM), a technique in which the patient draws the lower abdomen inward without pelvic or rib cage movement. This maneuver specifically targets and strengthens the deep trunk stabilizers, including the transversus abdominis and internal oblique muscles, which are essential for maintaining core stability [21]. Based on core stability theory, ADIM is an effective approach to selectively activating deep core muscles and coordinating pelvic movements, thereby contributing to postural control and spinal alignment [22]. To facilitate accurate execution of trunk stabilization, a pressure biofeedback device—which provides real-time visual and auditory feedback on abdominal pressure changes—was employed. This device helps patients with stroke develop a better understanding of spatial positioning and muscle engagement, ensuring proper activation of the transversus abdominis during ADIM training. Visual feedback through this device helps ensure accurate pre-activation of the transversus abdominis muscle. Haruyama et al. [22] found that ADIM combined with pressure biofeedback significantly improved trunk function, balance in the standing posture, and trunk movement in patients with stroke. McGalliard et al. [23] also reported that ADIM training enhanced transversus abdominis thickness during functional tasks.
Although VR training has demonstrated considerable effects on balance, concerns remain regarding potential asymmetrical postures and maladaptive recovery due to goal-directed movements. To address these issues, we used a trunk stabilization program with a pressure biofeedback device to provide visual and auditory feedback. This study aimed to improve muscle symmetry and recruitment of the transversus abdominis and internal oblique muscles, thereby promoting trunk postural control and balance in patients with stroke. However, research combining VR-based training with pre-training trunk stabilization to facilitate normal motor recovery remains limited. Therefore, this study aimed to determine whether a combined intervention of VR training and pre-training trunk stabilization would yield more effective improvements in automatic postural control and balance than those of VR training alone.

2. Materials and Methods

2.1. Participants

This study recruited 34 individuals diagnosed with subacute stroke who were admitted to the I Hospital in Incheon, South Korea. Eligible participants had been diagnosed with stroke within the previous 4 months and demonstrated sufficient cognitive and communication abilities, as indicated by a score of 21 or higher on the Korean version of the mini-mental state examination. Participants were also required to have a score of 20 or lower on the Korean version of the Trunk Impairment Scale (K-TIS), possess intact vision, be able to sit independently for at least 1 min, and maintain a seated position for 20 min with or without support. Patients were excluded if they were >80 years of age, unable to comprehend the study procedures, or presented with other neurological or orthopedic conditions that could affect postural control. Additional exclusion criteria included the presence of unilateral neglect, pushing syndrome, or a high K-TIS score. All participants were provided a thorough explanation of the study’s objectives, procedures, and potential risks. Written informed consent was obtained before participation, and individuals were assured that they could withdraw from the study at any time without any consequences. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Institutional Review Board of Sahmyook University (IRB No. SYU 2022-06-005-001). A flow diagram of participant enrollment, allocation, follow-up, and analysis is presented in Figure 1. A priori power analysis was conducted using G*Power version 3.1.9.7 (Heinrich Heine University Düsseldorf, Düsseldorf, Germany) to determine the appropriate sample size. Assuming a moderate effect size (f = 0.25), a significance level of α = 0.05, and a power of 0.80 for a two-way repeated measures ANOVA, the required sample size was calculated to be at least 28 participants. To account for potential dropouts, 34 participants were initially enrolled. Although the a priori analysis suggested a minimum of 28 participants, a post hoc calculation based on the final sample size of 30 (15 per group) yielded an achieved power of approximately 0.76, which is considered acceptable for detecting a moderate effect.

2.2. Procedure

Baseline physical characteristics, such as age, sex, height, weight, stroke duration, side of hemiparesis, and stroke type (ischemic or hemorrhagic), were recorded before the intervention. The participants were randomly assigned to either the experimental or control group using a computerized randomization process, resulting in 17 individuals in each group. To minimize bias, outcome assessors were blinded to group allocation. In addition, all research assistants were trained in standardized intervention and assessment protocols to ensure procedural consistency. Pre-intervention assessments were conducted before the training program began, and post-intervention evaluations were performed the day after the 3-week intervention concluded. Postural control, static balance, and dynamic balance were assessed in both groups. The experimental group participated in 30 min sessions consisting of 10 min of trunk stabilization education, followed by 20 min of VR-based training. The control group received only the 20 min VR-based training. Both groups completed the training three times per week for 3 weeks. After discharge, two participants from each group withdrew, leaving 15 participants per group for the final data analysis (Figure 1).

2.3. Intervention

Both groups underwent the same VR-based training; however, the experimental group performed additional trunk stabilization education prior to each VR session. All interventions were conducted under the supervision of trained research assistants who were instructed on both the measurement and intervention protocols.

2.3.1. Trunk Stabilization Education

Trunk stabilization education began with the abdominal drawing-in maneuver (ADIM), using a pressure biofeedback unit (PBU; Chattanooga Group, Hixson, TN, USA) initially set to 70 mmHg to activate the transversus abdominis muscle. Following the method of Jeon et al. [24], participants were instructed to reduce the pressure to approximately 60 mmHg during ADIM and maintain it within a ±5 mmHg range throughout functional tasks. The PBU was placed under the lower abdomen, and the therapist visually monitored pressure maintenance during movement. A contraction was considered successful when at least 80% of the maximal ADIM pressure was sustained. Once achieved, VR-based training commenced. Weekly recordings of maximal ADIM values were used to adjust training intensity [24,25]. The trunk stabilization education program consisted of two phases: sitting and standing. Training was applied progressively based on each participant’s performance. Following the approach described by Verheyden et al. [26], exercises were introduced from basic to more complex tasks. Repetitions and duration were modified by the therapist according to each participant’s motor capacity and response. The seated phase began with basic supine activities such as pelvic tilts and bridging, progressing to dynamic sitting tasks involving trunk flexion–extension, lateral flexion, rotation, and weight shifting for segmental trunk control. The standing phase incorporated functional tasks including sit-to-stand, static standing, and multi-directional weight shifts. This progressive structure enabled individualized load progression while preserving task specificity and participant engagement. Each session lasted 10 min and was conducted three times per week for 3 weeks. Vital signs were continuously monitored, and the intervention was stopped immediately if any adverse symptoms occurred.

2.3.2. VR-Based Training

VR-based training was delivered using the BioRescue system (RM Ingenierie, Rodez, France; BioRescue version 5.7, SyCoMoRe version 8.52) in a quiet, controlled environment. Participants performed seated or standing exercises on a pressure-sensitive platform that tracked the center of pressure (COP) in real time. Task-specific programs—such as maze navigation, image erasing, and downhill skiing—were implemented to improve postural control, weight-shifting ability, and symmetry. Training difficulty progressed across five levels by increasing the speed of visual stimuli (e.g., obstacle movement) and enhancing postural response sensitivity, which required more precise weight shifts. These parameters were individually adjusted according to each participant’s functional capacity. Each 20 min session provided continuous visual, auditory, and verbal feedback to optimize motor performance. As described by Delbroek et al. [27], the maze navigation task required participants to move a cursor through a top-view maze by shifting their body weight, the image erasing task involved sponge-like movements to clear a virtual blackboard, and the downhill skiing task demanded sustained directional weight shifts to avoid obstacles. Together, these interactive exercises created dynamic sensorimotor challenges designed to enhance both static and dynamic balance, as well as cognitive–motor integration.

2.4. Outcome Measures

A comprehensive set of outcome measures was used to evaluate different aspects of postural control and balance. These included clinical scales as well as instrument-based assessments, which were categorized into three domains: postural control, static balance, and dynamic balance.

2.4.1. Postural Control

  • K-TIS
The K-TIS was developed to assess trunk motor impairment in individuals with stroke [28]. It has been validated for use across various stages of stroke recovery, including acute and subacute patients within 3 months of onset, as well as those in the chronic stage beyond 6 months [29]. The K-TIS evaluates static and dynamic sitting balance, coordination of the trunk, and postural control. The total score ranges from 0 to 23, with high scores indicating better trunk function. It has demonstrated high reliability, with inter-rater reliability coefficients ranging from r = 0.87 to 0.96 and intra-rater reliability coefficients ranging from r = 0.85 to 0.99. Owing to its strong psychometric properties, the K-TIS serves as a clinically useful tool for the qualitative assessment of trunk control and evaluation of therapeutic outcomes.
2.
Postural Assessment Scale for Stroke (PASS)
The PASS was developed to complement the BL Motor Assessment and to specifically evaluate balance and trunk control in patients with stroke [30]. The PASS consists of 12 items across two subdomains: maintaining posture (including sitting unsupported and transitioning from supine to sitting) and changing posture (including sitting to lying down). Each item is rated on a scale from 0 to 3, with a maximum total score of 36; high scores reflect a greater ability to control dynamic balance and posture. The PASS has demonstrated excellent reliability, with inter-rater reliability reported at r = 0.97 and intra-rater reliability reported at r = 0.94 [31].

2.4.2. Static Balance

  • COP
Static balance was assessed using the BioRescue system which provides COP-based measurements for evaluating balance performance. The system quantifies sway area (mm2), total sway distance (cm), and average sway velocity (cm/s) based on bilateral hip pressure distribution [32]. All participants were seated with their hips and knees positioned at 90° and instructed to gaze at a fixed central target displayed on a monitor positioned in front of them. Each measurement was conducted over a 30 s period, first with eyes open and subsequently with eyes closed, following the same procedure. The test–retest reliability of the COP measurements has been reported to be high, with an intraclass correlation coefficient of 0.84 [33].

2.4.3. Dynamic Balance

  • Berg Balance Scale (BBS)
Dynamic balance was evaluated using the BBS, which is used to determine the effects of pre-training trunk stabilization on postural strategy and reactive control in patients with stroke [6]. The BBS comprises 14 items categorized into three domains: sitting, standing, and postural transitions. Each item is rated on a scale of 0 to 4, with a total possible score of 56. High scores indicate better dynamic balance ability. The tasks include functional activities, such as rising from a seated position, sitting down, transferring between chairs, reaching forward while standing, retrieving an object from the floor, turning to look behind, turning 360°, and alternating foot placement on a stool [34]. The BBS has demonstrated excellent reliability, with an intra-rater reliability of r = 0.99 and an inter-rater reliability of r = 0.98 [35].
2.
Limit of Stability (LOS)
The LOS was assessed using the BioRescue system to evaluate the maximum displacement of the body’s COP during voluntary weight shifting. The platform (610 mm × 580 mm × 10 mm) was embedded with approximately 1600 pressure sensors and, data were sampled at 100 Hz, which were transmitted to the analysis software in real time. During testing, the participants were seated with their hips and knees at 90° and feet flat on the ground. They were instructed to follow the on-screen directional arrows by shifting their COP toward the indicated direction without lifting their feet or losing balance. The trials were repeated if the participant lost stability. Each measurement was performed three times, and the average value was used for the analysis. The LOS test has shown high test–retest reliability, with an intraclass correlation coefficient of 0.84 [33].

2.4.4. Data Analysis

All statistical analyses were performed using SPSS version 25.0 (SPSS Corp., Armonk, NY, USA). Normality and homogeneity of variance were tested to ensure group equivalence. Descriptive statistics were used to summarize the demographic and clinical characteristics. For baseline comparisons between the experimental and control groups, independent t-tests were used.
To assess the effects of the intervention, a two-way repeated measures ANOVA (group × time) was conducted for each outcome variable to examine both main effects of time and group, as well as their interaction. Statistical significance was set at p < 0.05.

3. Results

This section presents the effects of trunk stabilization education combined with VR-based training on postural control, dynamic balance, and static balance. Pre- and post-intervention outcomes are compared between the experimental and control groups to evaluate the effectiveness of the intervention.

3.1. General Characteristics and Homogeneity of Participants

A total of 30 participants completed the study, with 15 allocated to the experimental group and 15 to the control group. There were no significant differences in general characteristics between the groups, including age, height, weight, mini-mental state examination—Korean version—scores, and time since stroke onset (months), indicating baseline homogeneity (Table 1).

3.2. Pre-Test Outcome Measures and Baseline Homogeneity of Participants

No statistically significant differences were observed between the experimental and control groups in the pre-test scores of the outcome measures, including the K-TIS, PASS, BBS, and LOS. These results indicate that the two groups were comparable at baseline in terms of postural control and balance-related functions, supporting the validity of subsequent comparisons (Table 2).

3.3. Postural Control (K-TIS and PASS)

Postural control was assessed using the K-TIS and PASS (Table 3) (Figure 2). Postural control significantly improved throughout the intervention period in both groups. Furthermore, a significant time × group interaction was identified for both the K-TIS and PASS, suggesting that the experimental group exhibited greater improvements than the control group.
For the K-TIS score, the experimental group improved from 12.67 ± 3.57 at baseline to 19.27 ± 2.31 post-intervention, while the control group increased from 14.87 ± 3.87 to 18.73 ± 2.78. The mean change was significantly greater in the experimental group (6.60 ± 3.56) than in the control group (3.86 ± 2.47), as indicated by a significant interaction effect (F = 5.958, p = 0.021).
In terms of the PASS, the experimental group’s score increased from 25.00 ± 4.69 at baseline to 32.60 ± 2.92 post-intervention, while the control group improved from 27.00 ± 4.29 to 30.93 ± 2.93. The magnitude of improvement was significantly greater in the experimental group (7.60 ± 3.92) than in the control group (3.93 ± 2.63), as indicated by a significant interaction effect (F = 9.034, p = 0.006).
These results not only demonstrate statistically significant improvements but also suggest clinically meaningful gains in trunk function and sitting balance, which are essential for safe transfers and early functional independence in stroke rehabilitation.

3.4. Dynamic Balance (BBS and LOS)

Dynamic balance was assessed using the BBS and LOS (Table 4) (Figure 3). Dynamic balance significantly improved in both groups following the intervention. Additionally, a significant group × time interaction was observed for the BBS (F = 8.767, p = 0.006), suggesting that the experimental group exhibited greater improvements than the control group.
For the BBS, the scores increased from 23.07 ± 13.06 to 46.60 ± 8.63 in the experimental group and from 30.93 ± 2.93 to 43.93 ± 11.01 in the control group. The mean change was greater in the experimental group (23.53 ± 9.94) than in the control group (13.00 ± 9.53).
For the LOS, the values increased from 6673.71 ± 4430.90 mm2 to 13091.13 ± 6896.00 mm2 in the experimental group and from 6749.42 ± 4393.63 mm2 to 10014.97 ± 5086.36 mm2 in the control group. A significant group × time interaction was also found for the LOS (F = 6.415, p = 0.017), indicating a greater improvement in the experimental group (6417.42 ± 4358.72 mm2 vs. 3265.55 ± 2056.75 mm2).
Given that a change of approximately 6 points in the Berg Balance Scale (BBS) is considered clinically meaningful in individuals with stroke, the observed improvement of over 23 points in the experimental group represents a substantial enhancement in functional balance, likely contributing to reduced fall risk and improved mobility.

3.5. Static Balance (COP)

Static balance was evaluated by measuring COP parameters—sway area (mm2), sway distance (cm), and sway velocity (cm/s)—under both eyes-open and eyes-closed conditions following 3 weeks of VR-based training with or without prior trunk stabilization (Table 5).

3.5.1. Eyes-Open

Under the eyes-open condition, the experimental group showed a significant reduction in sway area, from 31.13 ± 37.63 mm2 to 9.47 ± 17.12 mm2 (p < 0.05), whereas the control group’s decrease from 18.47 ± 37.05 mm2 to 3.93 ± 7.32 mm2 was not statistically significant (p > 0.05). Among these variables, only sway distance demonstrated a significant group × time interaction, suggesting that the experimental group experienced a greater degree of improvement compared to the control group.
The main effect of time was statistically significant, while the group × time interaction effect was not (F = 0.360, p = 0.553).
Sway distance showed a reduction from 6.99 ± 2.11 cm to 4.13 ± 1.18 cm in the experimental group and from 6.26 ± 2.19 cm to 5.21 ± 1.19 cm in the control group. The improvement was statistically significant over time, and a significant group × time interaction was found (F = 7.177, p = 0.012), indicating superior gains in the experimental group.
Sway velocity declined from 0.30 ± 0.19 cm/s to 0.20 ± 0.08 cm/s in the experimental group and from 0.22 ± 0.05 cm/s to 0.18 ± 0.04 cm/s in the control group. Although the main effect of time was significant, the group × time interaction was not statistically significant (F = 1.503, p = 0.230) (Figure 4). Although sway velocity improved over time in both groups, the absence of a significant group × time interaction suggests that the observed changes may reflect general effects of repetitive task practice rather than a specific benefit of trunk stabilization.

3.5.2. Eyes-Closed

Under the eyes-closed condition, the experimental group showed a significant reduction in sway area, from 23.93 ± 35.14 mm2 to 2.13 ± 2.41 mm2 (p < 0.05), whereas the control group showed a non-significant decrease from 18.67 ± 43.43 mm2 to 1.53 ± 3.02 mm2 (p > 0.05). However, none of the variables exhibited significant group × time interactions.
Sway distance declined from 6.28 ± 1.96 cm to 4.22 ± 1.83 cm in the experimental group and from 5.64 ± 1.47 cm to 4.84 ± 1.29 cm in the control group. A significant group × time interaction was found (F = 5.613, p = 0.025), indicating a greater reduction in the experimental group.
Sway velocity was reduced from 0.24 ± 0.15 cm/s to 0.16 ± 0.06 cm/s in the experimental group and from 0.19 ± 0.05 cm/s to 0.17 ± 0.04 cm/s in the control group. The group × time interaction was not significant (F = 2.185, p = 0.150) (Figure 5). As no significant interaction effects were found for sway area or sway velocity under eyes-closed conditions, these results should be interpreted with caution. The reduction in visual input may have limited the differentiating impact of the trunk stabilization intervention.

4. Discussion

This study found that trunk stabilization education provided prior to VR-based training led to significantly greater improvements in postural control, dynamic balance, and sway distance compared to VR training alone. Specifically, the experimental group showed larger gains in K-TIS, PASS, BBS, and LOS scores as well as significant reductions in sway distance under both visual conditions. These results suggest that preparatory trunk stabilization may enhance the effectiveness of VR-based balance rehabilitation by promoting core activation and improving postural symmetry.
Postural control, which forms the foundation for basic activities of daily living in early stroke rehabilitation, involves responding to unexpected perturbations and controlling body movement during voluntary shifts of the center of mass [6,36]. In this study, postural control was assessed using the K-TIS and PASS, both of which are reliable and clinically relevant measures in stroke rehabilitation. The experimental group demonstrated significantly greater improvement in PASS scores compared to the control group. This finding is consistent with previous studies reporting that trunk stabilization education improves selective motor control and reduces trunk muscle imbalance [7,19].
K-TIS scores also significantly improved in both groups, suggesting a general effect of task-based training. However, the experimental group achieved a significantly greater change, supporting the added benefit of pre-training trunk stabilization in facilitating trunk control through co-contraction of core muscles such as the abdominals and multifidus [17]. Even in the control group, which received only VR-based training, significant improvements in K-TIS scores were observed (p < 0.05), suggesting that repetitive task-based VR training can also promote trunk function through motor learning.
Although both the Postural Assessment Scale for Stroke (PASS) and the Berg Balance Scale (BBS) include elements related to postural transitions, these tools assess distinct but complementary aspects of balance control. The PASS, as described by Benaim et al. [30], primarily focuses on the ability to maintain and change posture in early-stage stroke rehabilitation, particularly capturing static postural control and transitions. In contrast, the Berg Balance Scale (BBS) focuses on dynamic balance and functional mobility by assessing tasks that challenge the center of mass during movement [35]. Accordingly, the two scales were analyzed separately in this study to reflect the multifaceted nature of postural control and dynamic balance in stroke patients.
Dynamic balance, the ability to voluntarily adjust posture or COP in response to disturbances, was assessed using the BBS and LOS [5]. The BBS, widely used to estimate rehabilitation outcomes and fall risk, evaluates static and dynamic balance through 14 functional tasks [34]. Trunk stabilization exercises have been shown to improve postural control and dynamic balance in patients with subacute stroke [7,37]. Consistent with this, the experimental group showed significant improvements in BBS scores, suggesting that pre-training trunk stabilization enhances neuromuscular coordination and dynamic balance. The BBS is also sensitive in detecting dynamic balance impairments and correlates closely with mobility and gait performance [36].
The LOS was assessed using the BioRescue, which measures the maximum COP displacement during voluntary weight shifting. Improvements in the LOS following weight-bearing training using pressure biofeedback have been reported [25,38,39]. In this study, both groups showed significant improvements in the LOS, with the experimental group demonstrating significantly greater changes in voluntary weight shifting, suggesting enhanced voluntary weight shifting and postural control, particularly for patients with hemiplegia.
Assessment of static balance provides insights into the ability to regulate body sway, weight shifting, and postural control, all essential for functional recovery [32]. Patients with stroke often show increased postural sway due to muscle weakness, abnormal tone, and impaired movement patterns [40]. Static balance was evaluated using COP parameters—sway area, sway distance, and sway velocity—under eyes-open and eyes-closed conditions. The experimental group showed significant improvements in sway distance under both conditions, with greater improvements under both visual conditions, consistent with previous findings that trunk stabilization education on unstable surfaces reduces sway and improves trunk control [14].
Although both groups showed reduced sway velocity under eyes-open conditions, no significant between-group difference was observed, possibly due to the relatively small sway distances during sitting tasks. Under eyes-closed conditions, both groups exhibited reduced sway distance, with a significantly greater reduction in the experimental group. This may reflect the enhanced trunk control achieved through pre-training trunk stabilization, which could compensate for the lack of visual input and reduce reliance on compensatory strategies such as shoulder elevation or lateral trunk lean [37]. These results highlight the consistent effect of trunk stabilization across both visual conditions, facilitating weight shift toward the paretic side and improving postural symmetry [36].
The observed improvements in postural control and dynamic balance suggest that integrating trunk stabilization into conventional VR-based programs may enhance rehabilitation outcomes, particularly for subacute stroke patients with trunk instability or limited sitting tolerance. While these findings are encouraging, several limitations should be acknowledged. The relatively small sample size may limit the generalizability of the results, underscoring the need for future studies with larger, multicenter populations. Additionally, the short duration of the intervention may not fully reflect long-term postural adaptations, emphasizing the need for long-term follow-up to confirm sustained effects. As the intervention was applied only to subacute patients, its applicability to individuals in the chronic phase remains uncertain. Future studies comparing different recovery stages could help clarify when trunk stabilization is most beneficial. Furthermore, the lack of direct assessment of compensatory trunk strategies limits insight into how patients adjusted their movement patterns. Future studies incorporating biomechanical or neuromuscular data may better explain the mechanisms of improvement. Despite these limitations, this study offers preliminary evidence that trunk-specific stabilization training, when applied prior to VR-based balance rehabilitation, may serve as an effective strategy to enhance postural control and functional recovery in stroke patients.

5. Conclusions

This study investigated the effects of VR-based training combined with pre-training trunk stabilization on postural control and balance in patients with subacute stroke. Both groups received 20 min of VR-based training three times per week for 3 weeks, while the experimental group additionally received 10 min of trunk stabilization education using a pressure biofeedback device. Post-intervention assessments revealed statistically significant improvements in postural control (K-TIS and PASS), dynamic balance, and static balance in both groups (p < 0.05). However, the experimental group showed significantly greater improvements than the control group across all outcome measures (p < 0.05). These findings suggest that pre-training trunk stabilization enhances the effectiveness of VR-based rehabilitation and may serve as a clinically valuable preparatory strategy to improve postural control and balance in stroke rehabilitation. Further clinical application and integration of this approach into real-world settings are encouraged to maximize its therapeutic potential. In addition, future studies may consider comparing this intervention with other biofeedback modalities—such as auditory, vibrational, or multi-sensory systems—to further optimize balance rehabilitation protocols.

Author Contributions

Conceptualization, S.L. and J.Y.; methodology, S.L. and J.Y.; validation, S.L. and J.Y.; formal analysis, S.L. and J.Y.; investigation, S.L.; resources, J.Y.; data curation, S.L. and J.Y.; writing—original draft preparation, S.L.; writing—review and editing, J.Y.; visualization, S.L. and J.Y.; supervision, J.Y.; project administration, J.Y. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of Sahmyook University (IRB No. SYU 2022-06-005-001).

Informed Consent Statement

Written informed consent has been obtained from the patients to publish this paper.

Data Availability Statement

The data are not publicly available due to participant confidentiality but may be obtained from the corresponding author upon reasonable request and with a confidentiality agreement.

Acknowledgments

This research was supported by Sahmyook University Research Fund in 2024.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Roth, E.J. Rehabilitation in Stroke Syndromes. Phys. Med. Rehabil. Clin. N. Am. 2007, 18, 817–853. [Google Scholar] [CrossRef]
  2. Popkirov, S.; Stone, J.; Buchan, A.M. Functional neurological disorder: A common and treatable stroke mimic. Stroke 2020, 51, 1629–1635. [Google Scholar] [CrossRef]
  3. Bohannon, R.W. Lateral trunk flexion strength: Impairment, measurement reliability and implications following unilateral brain lesion. Int. J. Rehabil. Res. 1992, 15, 249–251. [Google Scholar] [CrossRef]
  4. Ikai, T.; Kamikubo, T.; Takehara, I.; Nishi, M.; Miyano, S. Dynamic postural control in patients with hemiparesis. Am. J. Phys. Med. Rehabil. 2003, 82, 463–469. [Google Scholar] [CrossRef]
  5. Horak, F.; Esselman, P.; Anderson, M.E.; Lynch, M. The effects of movement velocity, mass displaced, and task certainty on associated postural adjustments made by normal and hemiplegic individuals. J. Neurol. Neurosurg. Psychiatry 1984, 47, 1020–1028. [Google Scholar] [CrossRef] [PubMed]
  6. Forget, R.; Lamarre, Y. Postural adjustments associated with different unloadings of the forearm: Effects of proprioceptive and cutaneous afferent deprivation. Can. J. Physiol. Pharmacol. 1995, 73, 285–294. [Google Scholar] [CrossRef]
  7. Cabanas-Valdés, R.; Bagur-Calafat, C.; Girabent-Farrés, M.; Caballero-Gómez, F.M.; Hernández-Valiño, M.; Urrutia Cuchi, G. The effect of additional core stability exercises on improving dynamic sitting balance and trunk control for subacute stroke patients: A randomized controlled trial. Clin. Rehabil. 2016, 30, 1024–1033. [Google Scholar] [CrossRef] [PubMed]
  8. Eng, J.J.; Tang, P.-F. Gait training strategies to optimize walking ability in people with stroke: A synthesis of the evidence. Expert Rev. Neurother. 2007, 7, 1417–1436. [Google Scholar] [CrossRef]
  9. Chen, L.; Lo, W.L.A.; Mao, Y.R.; Ding, M.H.; Lin, Q.; Li, H.; Zhao, J.L.; Xu, Z.Q.; Bian, R.H.; Huang, D.F. Effect of virtual reality on postural and balance control in patients with stroke: A systematic literature review. BioMed Res. Int. 2016, 2016, 7309272. [Google Scholar] [CrossRef]
  10. Lewis, G.N.; Rosie, J.A. Virtual reality games for movement rehabilitation in neurological conditions: How do we meet the needs and expectations of the users? Disabil. Rehabil. 2012, 34, 1880–1886. [Google Scholar] [CrossRef]
  11. McNulty, P.A.; Thompson-Butel, A.G.; Faux, S.G.; Lin, G.; Katrak, P.H.; Harris, L.R.; Shiner, C.T. The efficacy of Wii-based movement therapy for upper limb rehabilitation in the chronic poststroke period: A randomized controlled trial. Int. J. Stroke 2015, 10, 1253–1260. [Google Scholar] [CrossRef] [PubMed]
  12. Webster, D.; Celik, O. Systematic review of Kinect applications in elderly care and stroke rehabilitation. J. Neuroeng. Rehabil. 2014, 11, 1–24. [Google Scholar] [CrossRef] [PubMed]
  13. Wingham, J.; Adie, K.; Turner, D.; Schofield, C.; Pritchard, C. Participant and caregiver experience of the Nintendo Wii SportsTM after stroke: Qualitative study of the trial of WiiTM in stroke (TWIST). Clin. Rehabil. 2015, 29, 295–305. [Google Scholar] [CrossRef]
  14. Lee, D.; Bae, Y. The effectiveness of driving game on trunk control and gait ability in stroke. J. Mot. Behav. 2020, 52, 33–40. [Google Scholar] [CrossRef] [PubMed]
  15. Alhwoaimel, N.; Hughes, A.-M.; Warner, M.; Burridge, J.; Brown, S.; Verheyden, G.; Thijs, L.; Wee, S.K.; Turk, R. Trunk exercise programme post-stroke using a virtual reality video game-based system: A mixed methods feasibility study. In Proceedings of the 3rd Saudi Physical Therapy Association Conference (SPTA) 2019, Riyadh International Convention & Exhibition Center, Riyadh, Saudi Arabia, 6–8 September 2019. [Google Scholar]
  16. Foreman, M.H.; Engsberg, J.R. A virtual reality tool for measuring and shaping trunk compensation for persons with stroke: Design and initial feasibility testing. J. Rehabil. Assist. Technol. Eng. 2019, 6, 2055668318823673. [Google Scholar] [CrossRef]
  17. Yu, S.-H.; Park, S.-D. The effects of core stability strength exercise on muscle activity and trunk impairment scale in stroke patients. J. Exerc. Rehabil. 2013, 9, 362. [Google Scholar] [CrossRef]
  18. Magee, D. Instability and Stabilization: Theory and Treatment, 2nd ed.; Seminar Workbook; Orthopaedic Physical Therapy Products: Minneapolis, MN, USA, 1999. [Google Scholar]
  19. Marshall, P.W.; Murphy, B.A. Core stability exercises on and off a swiss ball. Arch. Phys. Med. Rehabil. 2005, 86, 242–249. [Google Scholar] [CrossRef]
  20. Van Criekinge, T.; Truijen, S.; Verbruggen, C.; Van de Venis, L.; Saeys, W. The effect of trunk training on muscle thickness and muscle activity: A systematic review. Disabil. Rehabil. 2019, 41, 1751–1759. [Google Scholar] [CrossRef]
  21. Macedo, L.G.; Maher, C.G.; Latimer, J.; McAuley, J.H. Motor control exercise for persistent, nonspecific low back pain: A systematic review. Phys. Ther. 2009, 89, 9–25. [Google Scholar] [CrossRef]
  22. Haruyama, K.; Kawakami, M.; Otsuka, T. Effect of core stability training on trunk function, standing balance, and mobility in stroke patients: A randomized controlled trial. Neurorehabilit. Neural Repair 2017, 31, 240–249. [Google Scholar] [CrossRef]
  23. McGalliard, M.K.; Dedrick, G.S.; Brismée, J.M.; Cook, C.E.; Apte, G.G.; Sizer, P.S., Jr. Changes in transversus abdominis thickness with use of the abdominal drawing-in maneuver during a functional task. Pm&r 2010, 2, 187–194. [Google Scholar]
  24. Jeon, I.-C.; Hwang, U.-J.; Jung, S.-H.; Kwon, O.-Y. Comparison of gluteus maximus and hamstring electromyographic activity and lumbopelvic motion during three different prone hip extension exercises in healthy volunteers. Phys. Ther. Sport 2016, 22, 35–40. [Google Scholar] [CrossRef]
  25. Jung, S.; Lee, K.; Kim, M.; Song, C. Audiovisual Biofeedback-Based Trunk Stabilization Training Using a Pressure Biofeedback System in Stroke Patients: A Randomized, Single-Blinded Study. Stroke Res. Treat. 2017, 2017, 6190593. [Google Scholar] [CrossRef]
  26. Verheyden, G.; Vereeck, L.; Truijen, S.; Troch, M.; LaFosse, C.; Saeys, W.; Leenaerts, E.; Palinckx, A.; De Weerdt, W. Additional exercises improve trunk performance after stroke: A pilot randomized controlled trial. Neurorehabilit. Neural Repair 2009, 23, 281–286. [Google Scholar] [CrossRef] [PubMed]
  27. Delbroek, T.; Hansen, D.; Lemmens, R.J.; Deschuytere, J.; Gielen, E.; Vanrumste, B.; Feys, H. The effect of cognitive-motor dual task training with the biorescue force platform on cognition, balance and dual task performance in institutionalized older adults: A randomized controlled trial. J. Neuroeng. Rehabil. 2017, 14, 9. [Google Scholar] [CrossRef] [PubMed]
  28. Verheyden, G.; Nieuwboer, A.; Mertin, J.; Preger, R.; Kiekens, C.; De Weerdt, W. The Trunk Impairment Scale: A new tool to measure motor impairment of the trunk after stroke. Clin. Rehabil. 2004, 18, 326–334. [Google Scholar] [CrossRef]
  29. Ko, J.; You, Y. Reliability and responsiveness of the Korean version of the trunk impairment scale for stroke patients. J. Korean Phys. Ther. 2015, 27, 175–182. [Google Scholar] [CrossRef]
  30. Benaim, C.; Pérennou, D.A.; Villy, J.; Rousseaux, M.; Pelissier, J.Y. Validation of a standardized assessment of postural control in stroke patients: The Postural Assessment Scale for Stroke Patients (PASS). Stroke 1999, 30, 1862–1868. [Google Scholar] [CrossRef]
  31. Mao, H.-F.; Hsueh, I.-P.; Tang, P.-F.; Sheu, C.-F.; Hsieh, C.-L. Analysis and comparison of the psychometric properties of three balance measures for stroke patients. Stroke 2002, 33, 1022–1027. [Google Scholar] [CrossRef]
  32. Seo, K.; Kim, J.; Wi, G. The effects of stair gait exercise on static balance ability of stroke patients. J. Phys. Ther. Sci. 2014, 26, 1835–1838. [Google Scholar] [CrossRef]
  33. Song, G.-B.; Park, E.-C. Comparison of the effects of task-oriented training and virtual reality training on upper extremity function, balance ability, and depression in stroke patients. J. Korean Soc. Phys. Med. 2016, 11, 115–125. [Google Scholar] [CrossRef]
  34. Berg, K.O.; Maki, B.E.; Williams, J.I.; Holliday, P.J.; Wood-Dauphinee, S.L. Clinical and laboratory measures of postural balance in an elderly population. Arch. Phys. Med. Rehabil. 1992, 73, 1073–1080. [Google Scholar] [PubMed]
  35. Bogle Thorbahn, L.D.; Newton, R.A. Use of the Berg Balance Test to predict falls in elderly persons. Phys. Ther. 1996, 76, 576–583. [Google Scholar] [CrossRef]
  36. Cabanas-Valdés, R.; Bagur-Calafat, C.; Girabent-Farrés, M.; Caballero-Gómez, F.M.; du Port de Pontcharra-Serra, H.; German-Romero, A.; Urrútia, G. Long-term follow-up of a randomized controlled trial on additional core stability exercises training for improving dynamic sitting balance and trunk control in stroke patients. Clin. Rehabil. 2017, 31, 1492–1499. [Google Scholar] [CrossRef] [PubMed]
  37. Seo, D.K.; Kwon, O.S.; Kim, J.H.; Lee, D.Y. The effect of trunk stabilization exercise on the thickness of the deep abdominal muscles and balance in patients with chronic stroke. J. Phys. Ther. Sci. 2012, 24, 181–185. [Google Scholar] [CrossRef]
  38. Yang, D.J. Influence of biofeedback weight bearing training in sit to stand to sit and the limits of stability on stroke patients. J. Phys. Ther. Sci. 2016, 28, 3011–3014. [Google Scholar] [CrossRef] [PubMed]
  39. Massion, J. Postural control system. Curr. Opin. Neurobiol. 1994, 4, 877–887. [Google Scholar] [CrossRef]
  40. de Haart, M.; Geurts, A.C.; Huidekoper, S.C.; Fasotti, L.; van Limbeek, J. Recovery of standing balance in postacute stroke patients: A rehabilitation cohort study. Arch. Phys. Med. Rehabil. 2004, 85, 886–895. [Google Scholar] [CrossRef]
Applsci 15 07620 g001
Figure 1. Study flow diagram.
Figure 1. Study flow diagram.
Applsci 15 07620 g001
Applsci 15 07620 g002
Figure 2. Pre- and post-treatment comparisons of postural control measures between groups.
Figure 2. Pre- and post-treatment comparisons of postural control measures between groups.
Applsci 15 07620 g002
Applsci 15 07620 g003
Figure 3. Pre- and post-treatment comparisons of dynamic balance measures between groups.
Figure 3. Pre- and post-treatment comparisons of dynamic balance measures between groups.
Applsci 15 07620 g003
Applsci 15 07620 g004
Figure 4. Pre- and post-treatment comparisons of static balance (eyes-open) measures between groups.
Figure 4. Pre- and post-treatment comparisons of static balance (eyes-open) measures between groups.
Applsci 15 07620 g004
Applsci 15 07620 g005
Figure 5. Pre- and post-treatment comparisons of static balance (eyes-closed) measures between groups.
Figure 5. Pre- and post-treatment comparisons of static balance (eyes-closed) measures between groups.
Applsci 15 07620 g005
Table 1. Baseline characteristics of participants (N = 30).
Table 1. Baseline characteristics of participants (N = 30).
VariableExperimental Group
(n = 15)		Control Group
(n = 15)		t (p)
Age (years)62.73 ± 10.8365.67 ± 13.39−0.659 (0.515)
Height (cm)164.07 ± 7.90163.80 ± 8.330.090 (0.929)
Weight (kg)65.93 ± 9.9362.73 ± 10.490.857 (0.399)
MMSE-K score24.33 ± 3.2225.60 ± 3.481.034 (0.310)
Onset (months)2.73 ± 1.222.47 ± 1.300.578 (0.568)
MMSE-K = mini-mental state examination—Korean version.
Table 2. Baseline scores of outcome measures between groups (N = 30).
Table 2. Baseline scores of outcome measures between groups (N = 30).
VariableExperimental Group
(n = 15)		Control Group
(n = 15)		t (p)
K-TIS (score)12.67 ± 3.5714.87 ± 3.87−1.616 (0.117)
PASS (score)25.00 ± 4.6927.00 ± 4.29−1.218 (0.233)
BBS (score)23.07 ± 13.0630.93 ± 13.03−1.651 (0.110)
LOS (mm2)6673.71 ± 4430.906749.42 ± 4393.63−0.047 (0.963)
K-TIS = Korean version of the Trunk Impairment Scale; PASS = Postural Assessment Scale for Stroke; BBS = Berg Balance Scale; LOS = limit of stability.
Table 3. Pre- and post-treatment scores of postural control measures (N = 30).
Table 3. Pre- and post-treatment scores of postural control measures (N = 30).
VariableExperimental Group
(n = 15)		Control Group
(n = 15)		Group X Time
F(df), p
K-TIS (score)
Pre12.67 ± 3.5714.87 ± 3.87
Post19.27 ± 2.3118.73 ± 2.78
Change6.60 ± 3.563.86 ± 2.47F(1,28) = 5.958, p = 0.021
PASS (score)
Pre25.00 ± 4.6927.00 ± 4.29
Post32.60 ± 2.9230.93 ± 2.93
Change7.60 ± 3.923.93 ± 2.63F(1,28) = 9.034, p = 0.006
Table 4. Pre- and post-treatment scores of dynamic balance measures (N = 30).
Table 4. Pre- and post-treatment scores of dynamic balance measures (N = 30).
VariableExperimental Group
(n = 15)		Control Group
(n = 15)		Group X Time
F(df), p
BBS (score)
Pre23.07 ± 13.0630.93 ± 2.93
Post46.60 ± 8.6343.93 ± 11.01
Change23.53 ± 9.9413.00 ± 9.53F(1,28) = 8.767, p = 0.006
LOS (mm2)
Pre6673.71 ± 4430.906749.42 ± 4393.63
Post13,091.13 ± 6896.0010,014.97 ± 5086.36
Change6417.42 ± 4358.723265.55 ± 2056.75F(1,28) = 6.415, p = 0.017
Table 5. Pre- and post-treatment scores of static balance measures (N = 30).
Table 5. Pre- and post-treatment scores of static balance measures (N = 30).
VariableExperimental Group
(n = 15)		Control Group
(n = 15)		Group X Time
F(df), p
Eyes-open
Surface area (mm2)
Pre31.13 ± 37.6318.47 ± 37.05
Post9.47 ± 17.123.93 ± 7.32
Change−21.67 ± 34.84−14.53 ± 30.09F(1,28) = 0.360, p = 0.553
Length (cm)
Pre6.99 ± 2.116.26 ± 2.19
Post4.13 ± 1.185.21 ± 1.19
Change−2.86 ± 1.85−1.04 ± 1.85F(1,28) = 7.177, p = 0.012
Average speed (cm/s)
Pre0.30 ± 0.190.22 ± 0.05
Post0.20 ± 0.080.18 ± 0.04
Change−0.10 ± 0.16−0.04 ± 0.05F(1,28) = 1.503, p = 0.230
Eyes-closed
Surface area (mm2)
Pre23.93 ± 35.1418.67 ± 43.43
Post2.13 ± 2.411.53 ± 3.02
Change−21.80 ± 33.57−17.13 ± 40.59F(1,28) = 0.118, p = 0.734
Length (cm)
Pre6.28 ± 1.965.64 ± 1.47
Post4.22 ± 1.834.84 ± 1.29
Change−2.06 ± 1.76−0.80 ± 1.06F(1,28) = 5.613, p = 0.025
Average speed (cm/s)
Pre0.24 ± 0.150.19 ± 0.05
Post0.16 ± 0.060.17 ± 0.04
Change−0.07 ± 0.13−0.02 ± 0.04F(1,28) = 2.185, p = 0.150
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Lee, S.; Yim, J. Effect of Virtual-Reality-Based Training, Including Preceding Trunk Stabilization Education, on Postural Control and Balance in Patients with Stroke: A Randomized Controlled Trial. Appl. Sci. 2025, 15, 7620. https://doi.org/10.3390/app15137620

AMA Style

Lee S, Yim J. Effect of Virtual-Reality-Based Training, Including Preceding Trunk Stabilization Education, on Postural Control and Balance in Patients with Stroke: A Randomized Controlled Trial. Applied Sciences. 2025; 15(13):7620. https://doi.org/10.3390/app15137620

Chicago/Turabian Style

Lee, SeongMin, and JongEun Yim. 2025. "Effect of Virtual-Reality-Based Training, Including Preceding Trunk Stabilization Education, on Postural Control and Balance in Patients with Stroke: A Randomized Controlled Trial" Applied Sciences 15, no. 13: 7620. https://doi.org/10.3390/app15137620

APA Style

Lee, S., & Yim, J. (2025). Effect of Virtual-Reality-Based Training, Including Preceding Trunk Stabilization Education, on Postural Control and Balance in Patients with Stroke: A Randomized Controlled Trial. Applied Sciences, 15(13), 7620. https://doi.org/10.3390/app15137620

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop