Effects of Virtual Reality-Based Interventions on Neuropathic Pain in Individuals with Spinal Cord Injury: A Meta-Analysis of Randomized Controlled Trials

Chang Liu; Yamei Liu

doi:10.2147/JPR.S644420

. 2026 Sep 17;19:644420. doi: 10.2147/JPR.S644420

Effects of Virtual Reality-Based Interventions on Neuropathic Pain in Individuals with Spinal Cord Injury: A Meta-Analysis of Randomized Controlled Trials

Chang Liu ^1,^✉, Yamei Liu ²

PMCID: PMC13592339 PMID: 42770034

Abstract

Background

Neuropathic pain is a common and difficult-to-treat complication of spinal cord injury (SCI). Virtual reality (VR)-based interventions have emerged as a non-pharmacological approach to pain modulation, but their efficacy remains uncertain. This meta-analysis evaluated the effect of VR-based interventions on neuropathic pain intensity in individuals with SCI.

Methods

PubMed, Cochrane Library, Embase, Web of Science, Wanfang, and CNKI were searched through June 19, 2026. Randomized controlled trials (RCTs) comparing VR-based interventions with control conditions in adults with SCI-related neuropathic pain were included. Standardized mean differences (SMDs) with 95% confidence intervals (CIs) were pooled using a random-effects model accounting for the influence of potential heterogeneity.

Results

Ten RCTs involving 317 participants contributed 12 comparisons. VR-based interventions significantly reduced neuropathic pain intensity compared with controls (SMD = −0.50, 95% CI −0.76 to −0.25; p < 0.001; I² = 27%). Leave-one-out sensitivity analyses confirmed the robustness of the finding (SMD range, −0.42 to −0.60; all p < 0.05). No significant subgroup differences were observed according to study design, mean age, time since SCI, VR intervention type, intervention session pattern, comparator type, pain assessment scale, or risk-of-bias judgment (all p for subgroup differences > 0.05). Meta-regression analyses identified no significant effect modifiers. The certainty of evidence was moderate according to the GRADE system.

Conclusion

VR-based interventions probably reduce neuropathic pain intensity in individuals with SCI. Further well-designed, adequately powered RCTs are warranted to confirm the magnitude and durability of this effect.

Keywords: spinal cord injury, neuropathic pain, virtual reality, rehabilitation, meta-analysis

Introduction

Spinal cord injury (SCI) is a devastating neurological condition that results in permanent or long-term impairment of sensory, motor, and autonomic function below the level of injury.¹^,² Neuropathic pain is one of the most common and disabling complications following SCI, affecting approximately 40%–50% of patients and frequently persisting as a chronic condition.³^,⁴ Characterized by spontaneous burning, shooting, or electric shock-like pain, often accompanied by allodynia and hyperalgesia, SCI-related neuropathic pain substantially impairs physical function, sleep quality, emotional well-being, participation in rehabilitation, and overall quality of life.⁵^,⁶ Although pharmacological therapies remain the cornerstone of treatment, their analgesic effects are often modest and may be limited by adverse effects, tolerance, and poor long-term adherence.⁶^,⁷ Consequently, effective non-pharmacological interventions that can complement conventional pain management are urgently needed.

Virtual reality (VR) has emerged as an innovative rehabilitation technology that creates computer-generated interactive environments in which users can experience immersive or non-immersive virtual scenarios through multisensory stimulation.⁸^,⁹ In recent years, VR has been increasingly applied across neurological rehabilitation, including motor and gait training, balance rehabilitation, upper-limb recovery, cognitive rehabilitation, and pain management, because it enables repetitive, task-specific, interactive, and multisensory training within controlled virtual environments.¹⁰^,¹¹ In patients with SCI, VR-based interventions have shown potential benefits for motor function, balance, upper-limb performance, functional independence, and other rehabilitation outcomes.^12–14 The analgesic effects of VR are thought to involve several complementary mechanisms, including attentional distraction from nociceptive stimuli, restoration of congruent sensorimotor integration through virtual walking or body ownership illusions, modulation of maladaptive cortical reorganization, and promotion of neuroplasticity through repetitive multisensory feedback and motor imagery.¹⁵^,¹⁶ These characteristics make VR an attractive adjunctive therapy for managing SCI-related neuropathic pain.¹⁷

Previous systematic reviews have primarily focused on the effects of VR on functional recovery, balance, or upper-limb motor function after SCI rather than neuropathic pain specifically.^18–20 A recent systematic review and meta-analysis published evaluated the effectiveness of VR for SCI-related neuropathic pain.²¹ However, it included only seven studies, of which fewer than half were randomized controlled trials (RCTs), while the remaining evidence was derived from uncontrolled pre–post studies, limiting the strength of the conclusions.²¹ Furthermore, several recently published RCTs were unavailable or not included,^22–24 and potential sources of heterogeneity were incompletely explored. Therefore, an updated meta-analysis synthesizing evidence exclusively from RCTs is warranted to provide a more rigorous estimate of the comparative effectiveness of VR-based interventions for SCI-related neuropathic pain. Accordingly, the present meta-analysis of RCTs aimed to evaluate the effect of VR-based interventions on neuropathic pain intensity in adults with SCI and to explore potential factors influencing treatment efficacy through subgroup, sensitivity, and meta-regression analyses.

Methods

This systematic review and meta-analysis followed established methodological guidance from the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) reporting framework²⁵ and the Cochrane Handbook for Systematic Reviews of Interventions.²⁵ The review protocol was registered prospectively with the PROSPERO international prospective register of systematic reviews (CRD420261433222).

Study Inclusion and Exclusion Criteria

Eligible studies were selected based on predefined criteria structured according to the PICOS framework.

Population (P)

Adult participants (aged ≥18 years) with SCI and SCI-related neuropathic pain. Neuropathic pain could be identified using established diagnostic criteria, validated assessment instruments, or clinical assessment and explicitly described by the study authors as neuropathic pain related to SCI. No restrictions were imposed regarding sex, etiology of SCI, neurological level of injury, American Spinal Injury Association Impairment Scale (AIS) grade, completeness of injury, or time since injury.

Intervention (I)

VR-based interventions intended to reduce or modulate neuropathic pain. Eligible interventions included immersive and non-immersive VR approaches, such as virtual walking or movement, virtual limb or body embodiment, visually induced movement or walking illusions, interactive or immersive virtual environments, and VR-assisted exercise. Studies in which the VR-based intervention was combined with another treatment were eligible when the same co-intervention was provided to the comparator group, thereby allowing the effect of the VR-based component to be evaluated.

Comparator (C)

An eligible comparator could include no intervention, usual care, conventional rehabilitation, sham or control virtual/visual interventions, non-immersive or two-dimensional visual controls, non-VR exercise, or another active intervention without the therapeutic VR-based component.

Outcomes (O)

Neuropathic pain intensity or severity assessed quantitatively using the Visual Analogue Scale (VAS), Numerical Rating Scale (NRS), Brief Pain Inventory (BPI) pain severity score, or another comparable quantitative measure of neuropathic pain intensity or severity. Studies were required to provide sufficient quantitative data for effect-size calculation or data from which an effect estimate and its variance could be derived.

Study Design (S)

RCTs, including both parallel-group and randomized crossover trials.

Studies were excluded if they were non-randomized or quasi-randomized trials, single-arm studies, uncontrolled before–after studies, observational studies, case reports or case series, qualitative studies, reviews, study protocols, conference abstracts, or animal studies. Studies involving participants without SCI or without SCI-related neuropathic pain were excluded, as were studies of mixed populations when relevant data for participants with SCI-related neuropathic pain could not be extracted separately. Studies evaluating interventions without a virtual, simulated, or visually mediated illusion component, or those in which an additional active treatment was provided only to the intervention group such that the independent effect of the VR-based component could not be determined, were excluded. Studies that did not quantitatively assess neuropathic pain intensity or severity or did not provide sufficient data for effect-size calculation were also excluded. For multiple publications involving the same or overlapping participant population, the report which provided the most complete and relevant data was included to avoid double counting.

Database Search

PubMed, Cochrane Library, Embase, Web of Science, Wanfang, and China National Knowledge Infrastructure (CNKI) were systematically searched using the search strategy combined terms related to SCI, neuropathic pain, and VR-based interventions. The search syntax was adapted for each database, and the complete database-specific search strategies are provided in Supplementary File 1. We considered only peer-reviewed, English or Chinese language full-text studies conducted in human participants. In addition, the bibliographies of relevant reviews and eligible articles were manually screened to identify any additional records. The final search update was performed on June 19, 2026.

Risk of Bias Evaluation

Two reviewers independently performed the literature search, screened the retrieved records for eligibility, extracted relevant data, and evaluated the risk of bias of the included studies. Any disagreements were resolved through discussion between the two authors until consensus was reached. The methodological quality of the included randomized controlled trials was evaluated using the Cochrane Risk of Bias tool (RoB 2.0), which examines potential bias across five domains: the randomization process, deviations from intended interventions, missing outcome data, measurement of outcomes, and selection of the reported results.²⁵ Each domain and the overall risk of bias were judged as low risk of bias, some concerns, or high risk of bias, and these judgments were subsequently integrated to generate an overall risk-of-bias assessment for each study.

Data Extraction

Extracted data included general study information (first author, publication year, and country), study design (double-blind, single-blind, or open-label, crossover or parallel-group RCTs), participant characteristics (number of patients included, mean age, sex distribution, AIS grade, level of injury, and time since SCI), details of VR-based interventions and controls, and tools used to evaluate the pain intensity.

Statistical Analysis

Because pain intensity was assessed using different measurement scales across the included studies, effect sizes were pooled as standardized mean differences (SMDs) with 95% confidence intervals (CIs).²⁵ All pain scales were oriented in the same direction, with lower scores indicating lower pain intensity or severity. Therefore, a negative SMD indicated a greater reduction in neuropathic pain favoring the VR-based intervention. For studies with more than one eligible intervention group sharing a common control group, the sample size of the control group was divided equally among the relevant comparisons to avoid double counting of participants.²⁵ Between-study heterogeneity was assessed using Cochran’s Q test and quantified with the I² statistic. I² values of < 25%, 25–75%, and > 75% were considered to indicate low, moderate, and high heterogeneity, respectively.²⁶ Pooled estimates were generated using a random-effects model to account for anticipated clinical and methodological diversity among trials.²⁵ The stability of the findings was evaluated through leave-one-out analyses.²⁵ Prespecified subgroup analyses were performed based on the type of RCT (parallel-group vs crossover), mean age of the patients, time since SCI onset, type of VR-based intervention (sensorimotor/embodiment-based VR vs distraction/environment-based VR), intervention session pattern (single session vs multiple sessions), type of control (sham/virtual control vs active non-VR control), scales for evaluating pain intensity, and risk of bias according to RoB 2.0. Medians of continuous variables were used as cutoff values for subgroup analyses to ensure a balanced distribution of studies across subgroups. In addition, the univariate meta-regression analysis was also performed to explore whether study-level characteristics were associated with the effect estimates, such as the sample size, mean age of the patients, proportion of men, time after SCI, and duration of intervention. Potential publication bias was explored using funnel plot inspection and Egger’s regression test.²⁷ Statistical significance was defined as a two-sided p value < 0.05. All analyses were carried out using RevMan (version 5.3, Cochrane Collaboration, Oxford, UK) and Stata (version 17.0, StataCorp, College Station, TX, USA) software.

Certainty of Evidence

Two reviewers independently assessed the certainty of evidence using the GRADE (Grading of Recommendations, Assessment, Development and Evaluation) approach, which evaluates the level of confidence in the results based on several domains, including risk of bias, inconsistency, indirectness, imprecision, and potential publication bias.²⁸ The overall certainty for each outcome was classified as high, moderate, low, or very low. Any disagreements between reviewers were resolved through discussion until a consensus was reached.

Results

Literature Search

Figure 1 illustrates the study selection process for this meta-analysis. The database search initially identified 525 records, of which 336 remained after removing duplicates. Screening of titles and abstracts led to the exclusion of 310 articles that were not relevant to the study objective. The full texts of the remaining 26 articles were then reviewed in detail, and 16 were excluded for the reasons summarized in Figure 1. Ultimately, 10 RCTs^22–24^,^29–35 met the eligibility criteria and were included in the quantitative synthesis.

A flowchart of study identification and screening process for a review. The flowchart outlines the process of identifying studies via databases and registers. Initially, 525 records were identified from various databases: PubMed 195, Cochrane Library 47, Embase 134, Web of Science 119, Wanfang 18 and CNKI 12. After removing 189 duplicate records, 336 records were screened. During screening, 310 records were excluded due to being reviews, editorials, irrelevant studies, or meta-analyses. This left 26 reports sought for retrieval, all of which were retrieved. These reports were assessed for eligibility, resulting in the exclusion of 16 reports for reasons such as not being randomized controlled trials (5), not limited to patients with spinal cord injury (2), not virtual reality-based interventions (2), outcome of neuropathic pain not reported (4) and overlapping patients (3). Ultimately, 10 studies were included in the review. — Flowchart for the literature search and study inclusion.

Characteristics of the Included Studies

A total of 10 RCTs were included,^22–24^,^29–35 contributing 12 comparisons to the meta-analysis. Two studies contributed two comparisons each.²⁹^,³⁵ Specifically, Soler et al²⁹ evaluated the virtual illusion intervention in participants receiving either concurrent active transcranial direct current stimulation (tDCS) or sham tDCS; these two comparisons were therefore analyzed separately as the with-tDCS and without-tDCS comparisons. Tabacof et al³⁵ evaluated two distinct VR interventions, namely somatic VR and scenic VR, against a common control group. These intervention arms were treated as separate comparisons, with the sample size of the shared control group divided equally between the two comparisons to avoid double counting. The included studies were published between 2010 and 2026 and were conducted in Spain, Turkey, Switzerland, the United States, Australia, China, and India. Sample sizes were generally small, ranging from 11 to 62 participants, with the 10 included RCTs enrolling a total of 317 participants with SCI-related neuropathic pain. The mean participant age ranged from 32.3 to 54.3 years, with men accounting for 45.5% to 100% of participants. The included populations were clinically heterogeneous with respect to neurological level and severity of SCI, encompassing cervical, thoracic, and lumbar injuries, as well as both paraplegia and tetraplegia and complete and incomplete injuries. The reported time since SCI ranged from approximately 8 to 234 months, although one study reported only a minimum duration of 12 months.³⁴ The VR-based interventions included visually induced or virtual walking illusions, virtual leg embodiment, semi-immersive and immersive virtual walking, interactive three-dimensional virtual environments, virtual nature environments combined with arm cycling, somatic VR designed to promote embodiment and motor imagery, scenic VR primarily intended for distraction, and non-immersive visual walking interventions. Intervention exposure ranged from a single session lasting from approximately 1 to 20 minutes to repeated treatment programs lasting 2 to 6 weeks. Comparator conditions included control visual illusions, transcutaneous electrical nerve stimulation (TENS), asynchronous virtual stimulation, virtual wheelchair movement, no-VR conditions, conventional rehabilitation, neutral virtual environments, documentary viewing, and two-dimensional animated films. Neuropathic pain intensity was assessed using the NRS,²⁹^,^32–35 VAS,²²^,²⁴^,³⁰^,³¹ or BPI.²³ Detailed characteristics of the included studies are presented in Table 1.

Table 1.

Characteristics of the Included RCTs

Study	Country	Design	No. of Patients	Mean Age (Years)	Men (%)	AIS Grade	Level of Injury	Time Since SCI (Months)	Details of VR Intervention	Details of Control Group	Tools for NP Evaluation
Soler 2010 with tDCS²⁹	Spain	R, DB, P	20	43.3	75.0	A-B	C4-T12	85.2	Video of walking legs + vertical mirror + synchronized walking sound; patients encouraged to imagine walking; 5 sessions/week; 20 min total (5 min tDCS + 15 min video) for 2 weeks;	Control illusion: graphical illustrations, faces, landscapes (no human movement); 5 sessions/week; 20 min total (5 min tDCS + 15 min video); 2 weeks (10 sessions)	NRS
Soler 2010 no tDCS²⁹	Spain	R, DB, P	19	48.4	68.4	A-B	C4-T12	116.2	Video of walking legs + vertical mirror + synchronized walking sound; patients encouraged to imagine walking; 5 sessions/week; 20 min total (5 min sham tDCS + 15 min video) for 2 weeks;	Control illusion: graphical illustrations, faces, landscapes (no human movement); 5 sessions/week; 20 min total (5 min sham tDCS + 15 min video); 2 weeks (10 sessions)	NRS
Özkul 2015³⁰	Turkey	R, OL, CO	24	32.3	75.0	A-C	C5-L5	19	Video of walking legs projected on screen + vertical mirror (150 × 52 cm) placed 2.5 m in front; patients viewed reflection of own body on top and walking legs on bottom; synchronized walking sound; patients asked to move upper extremities and upper body in rhythm with walking; 5 sessions/week; 15 min per session; for 2 weeks;	Intelect TENS device; 5 sessions/week; 30 min per session; for 2 weeks;	VAS
Pozeg 2017³¹	Switzerland	R, OL, CO	11	47.3	90.0	A-C	T2-L2	205.2	Immersive VR with HMD; real-time video of virtual legs fed to HMD from first-person perspective; fake legs placed on chair mimicking sitting posture; 60 seconds per condition; 4 min for each session; single session	Virtual leg illusion using the same HMD with asynchronous visuotactile stimulation	VAS
Richardson 2019³²	USA	R, SB, P	59	44.8	79.7	NR, complete, 38/59 (64.4%), and incomplete, 21/59 (35.6%)	Tetraplegia, 25/59 (42.4%); paraplegia, 34/59 (57.6%)	178.8	Semi-immersive VR: 20-min video played on 3D monitor in quiet, dimly lit room; first-person view from actor’s perspective; actor walking along a path, head down with moving legs and feet in view; every ~3 min, actor scanned horizon briefly then returned to leg/foot view; 20 min, single session;	Same setup and actor, in first-person view, propelling a manual wheelchair along the same path for 20 min (no lower limb movement); single session;	NRS
Austin 2021³³	Australia	R, OL, CO	16	54.3	100.0	A-D	Lower thoracic/upper lumbar: 8 (50%); Mid-thoracic: 5 (31%); Upper thoracic: 1 (6%); Lower cervical: 2 (13%)	234	Handheld joystick for navigation, 3D VR experience) - scenic meadow environment with 360° view; participants instructed on joystick use to move around and make full use of 360° scene; 15 min single session	Participant seated in same position as VR session; 15 min single session;	NRS
Azurdia 2022³⁴	USA	R, OL, CO	11	43.3	45.5	NR	Paraplegia: 2 (18%); Tetraplegia: 9 (82%)	≥ 12	Virtual nature environments (Blue Ocean, Green Meadows, White Winter); Arm cycle ergometer at 50±5 RPM; 6-minute submaximal trial; single session;	No-VR arm cycle ergometer at 50±5 RPM; 6-minute submaximal trial; single session;	NRS
Ji 2022²²	China	R, OL, P	62	38.8	56.5	NR (complete 44, incomplete 18)	Cervical: 13 (21%); Thoracic: 19 (30.6%); Lumbar: 30 (48.4%)	10.1	Head-mounted VR glasses, 360° panoramic camera recording; male/female models in casual clothes walking along path in first-person view; legs/feet visible; every 3 min, model looks up at path for 10s then returns to leg/foot view; 30 min per session; 3 times per day; for 6 weeks	No VR, conventional rehabilitation	VAS
Tabacof 2024 somatic³⁵	USA	R, DB, P	12	36.1	73.3	A-E	Paraplegic and tetraplegic	168	360° first-person perspective videos showing somatic activities; participants instructed to embody displayed limbs as their own and engage in motor imagery; 20 min per session; 3 sessions per week for 4 weeks	Solid black 360° screen with 12 different audio tracks of nature sounds	NRS
Tabacof 2024 scenic³⁵	USA	R, DB, P	11	39.4	66.7	A-E	Paraplegic and tetraplegic	138	360° landscape environments in first-person perspective; dynamic image changes according to head position; no virtual limbs displayed; corresponding background audio; Participants exposed to landscape environments; designed for distraction; 20 min per session; 3 sessions per week for 4 weeks	Solid black 360° screen with 12 different audio tracks of nature sounds	NRS
Mollà-Casanova 2026²³	Spain	R, SB, P	29	50.2	65.8	C-D	Tetraplegic and paraplegic	74.4	Non-immersive: mirror reflecting upper body + height-adjustable screen with video projection; standing position using standing device; Videos of individuals walking (legs moving) - customized to match participant’s body size; 30 min (gait training + strength/stretching); 3 sessions/week for 6 weeks	Documentary from space (no human/animal movement) - avoids stimulation of motor brain areas; 30 min; 3 sessions/week for 6 weeks	BPI
Selvaraj 2026²⁴	India	R, DB, P	43	35.8	83.7	A-C	T1-T12	8	VR Headset with Oculus Touch Controllers; 360° videos recorded by authors featuring lower-limb activities: wheelchair propelling, stationary cycling, outdoor cycling, walking; first-person perspective; spatial audio for enhanced immersion; Participants seated in wheelchair with feet on static cycle ergometer pedals; engaged with virtual environment by moving head; designed to promote motor imagery and sensorimotor engagement; 15 min per session; 2 sessions per day for 2 weeks	2D animated films (non-immersive, no VR features); 15 min per session; 2 sessions per day for 2 weeks	VAS

Open in a new tab

Abbreviations: AIS, American Spinal Injury Association Impairment Scale; BPI, Brief Pain Inventory; CO, crossover; DB, double-blind; HMD, head-mounted display; NP, neuropathic pain; NR, not reported; NRS, Numerical Rating Scale; OL, open-label; P, parallel-group; R, randomized; RPM, revolutions per minute; SB, single-blind; SCI, spinal cord injury; tDCS, transcranial direct current stimulation; TENS, transcutaneous electrical nerve stimulation; VAS, Visual Analogue Scale; VR, virtual reality.

Risk-of-Bias Assessment

According to the Cochrane RoB 2.0 tool, all included studies were judged to have a low risk of bias arising from the randomization process, and all were considered at low risk for measurement of the outcome. Two studies (three comparisons)²⁴^,²⁹ were judged to have an overall low risk of bias. The remaining studies²²^,²³^,^30–35 were rated as having some concerns, primarily because of the inability to blind participants or intervention providers, potential selective reporting in studies without a prespecified or registered analysis protocol, or concerns related to missing outcome data. In particular, the two comparisons from Tabacof et al³⁵ had some concerns regarding missing outcome data because of substantial attrition, while Mollà-Casanova et al²³ also had some concerns related to attrition and the use of a per-protocol analysis. Nevertheless, the measurement of pain outcomes was generally based on validated instruments, and several studies incorporated blinded outcome assessment.²³^,²⁴^,²⁹^,³²^,³⁵ The detailed risk-of-bias assessments are presented in Table 2.

Table 2.

Study Risk of Bias Evaluation via the Cochrane RoB 2.0 Tool

Study	Randomization Process	Deviations from Intended Interventions	Missing Outcome Data	Measurement of the Outcome	Selection of the Reported Results	Overall
Soler 2010 with tDCS²⁹	Low risk (Computer-generated randomization list; groups similar at baseline; no baseline differences in demographics, pain intensity, or medication distribution)	Low risk (Double-blind design; patients and outcome assessors blinded; sham tDCS effectively blinded; both groups received same co-intervention (tDCS))	Low risk (No dropouts in this comparison; all 20 patients completed the study)	Low risk (Validated instruments (NRS); same researcher blind to treatment performed all evaluations; independent researcher applied interventions)	Low risk (Primary and secondary outcomes pre-specified; all reported outcomes consistent with methods)	Low risk
Soler 2010 no tDCS²⁹	Low risk (Computer-generated randomization list; groups similar at baseline; no baseline differences in demographics, pain intensity, or medication distribution)	Low risk (Double-blind design; patients and outcome assessors blinded; sham tDCS effectively blinded; both groups received same co-intervention (sham tDCS))	Low risk (1 dropout in VI group (transferred to another city); unlikely related to intervention; acceptable attrition rate (5.3%))	Low risk (Validated instruments (NRS); same researcher blind to treatment performed all evaluations; independent researcher applied interventions)	Low risk (Primary and secondary outcomes pre-specified; all reported outcomes consistent with methods)	Low risk
Özkul 2015³⁰	Low risk (Patients randomly divided into two groups using a table of random numbers; baseline characteristics comparable)	Some concerns (Not blinded (open-label design); patients and therapists knew which intervention was being applied; however, crossover design with washout minimizes some bias)	Low risk (26 patients enrolled, 24 completed (2 dropped out due to discharge from hospital); reasons reported; attrition <10%)	Low risk (Validated instruments used (VAS); same physiotherapist performed all assessments)	Some concerns (Multiple outcomes assessed; no pre-registered protocol identified; potential for selective reporting of significant findings)	Some concerns
Pozeg 2017³¹	Low risk (Randomization of VLI conditions; counterbalancing of VLI/FBI order; no baseline differences expected due to within-subjects design)	Some concerns (Not explicitly blinded; however, within-subjects crossover design and clearly defined synchronous vs asynchronous conditions minimize risk; participants may have perceived differences between conditions)	Low risk (pain data available for all 11 patients with neuropathic pain)	Low risk (Validated VAS for pain intensity (0–100); validated questionnaires for illusion (adapted from published body illusion studies))	Some concerns (Multiple outcomes assessed; no pre-registered protocol identified; some analyses performed post-hoc; correction for multiple comparisons noted but not applied to all analyses)	Some concerns
Richardson 2019³²	Low risk (Blocked randomization schedule with 1:1 allocation; generated by investigator not involved in outcome assessment; baseline characteristics comparable between groups)	Some concerns (Participants could not be blinded to condition (nature of VR); however, examiner was blinded and left room during viewing; participants instructed not to describe what they viewed)	Low risk (Only 1 participant (in control group) declined participation after randomization; no missing outcome data for remaining 59 participants)	Low risk (Validated instruments (NRS); same examiner (blinded to condition) performed all assessments; pain classification using established criteria (Bryce-Ragnarsson)	Some concerns (Multiple outcomes assessed; no pre-registered protocol identified in manuscript (though NCT01884662 registered); some analyses appear post-hoc (exploratory secondary analyses noted))	Some concerns
Austin 2021³³	Low risk (Random allocation using sealed opaque envelopes; randomization sequencing, enrollment, and assignment performed by PA)	Some concerns (Participants and researchers could not be blinded (obvious differences between 3D HMD and 2D screen); however, researcher used neutral language script to minimize bias; 60-min washout period adequate to eliminate carryover effects)	Low risk (17 enrolled, 1 excluded (no pain in previous week); complete data for remaining 16 participants)	Low risk (Validated instruments (NRS); NPRS recommended by IMMPACT; consistent assessment timing (11:00 a.m).	Some concerns (Registered trial (ACTRN12618000959279); primary outcome specified; however, multiple comparisons without correction noted)	Some concerns
Azurdia 2022³⁴	Low risk (Randomized crossover; treatment order randomized)	Some concerns (Not blinded; no explicit washout period between conditions (1 week between visits, but effect may persist))	Low risk (Complete data for all 11 participants)	Low risk (Validated pain intensity scale used; data reported as mean ± SD)	Some concerns (No pre-registered protocol identified; multiple outcomes assessed)	Some concerns
Ji 2022²²	Low risk (Random number table method; baseline characteristics comparable)	Some concerns (Blinding not specified; participants likely aware of group assignment (walking vs wheeling)	Low risk (No dropouts reported; all 62 participants completed the study)	Low risk (Validated VAS used; SAS, SDS, PSQI are validated instruments)	Some concerns (No pre-registered protocol identified; all outcomes reported appear consistent with methods)	Some concerns
Tabacof 2024 somatic³⁵	Low risk (Simple randomization; stratified by injury level (paraplegia/tetraplegia); baseline characteristics comparable)	Low risk (Double-blinded - participants and outcome assessors blinded to group allocation; environments randomized across sessions; VE order randomized)	Some concerns (32 enrolled, 22 completed (68.75% completion); 10 dropouts (31.25%) - high attrition; reasons include COVID-19 disruptions, discomfort with headset, personal reasons)	Low risk (Validated instruments (NPSI, NRS, NPS); outcome assessors blinded; consistent assessment timing)	Low risk (Registered trial (NCT04700033); outcomes specified in protocol; analysis plan clearly described)	Some concerns
Tabacof 2024 scenic³⁵	Low risk (Simple randomization; stratified by injury level (paraplegia/tetraplegia); baseline characteristics comparable)	Low risk (Double-blinded - participants and outcome assessors blinded to group allocation; environments randomized across sessions; VE order randomized)	Some concerns (32 enrolled, 22 completed (68.75% completion); 10 dropouts (31.25%) - high attrition; reasons include COVID-19 disruptions, discomfort with headset, personal reasons)	Low risk (Validated instruments (NPSI, NRS, NPS); outcome assessors blinded; consistent assessment timing)	Low risk (Registered trial (NCT04700033); outcomes specified in protocol; analysis plan clearly described)	Some concerns
Mollà-Casanova 2026²³	Low risk (Computer-generated random sequence; 1:1 allocation; Random Allocation Software; allocation concealment preserved)	Some concerns (Participants and therapists could not be blinded (nature of therapy); however, outcome assessor was blinded)	Some concerns (9 participants (23.7%) lost to follow-up; reasons reported (hospital discharge, surgery, non-response); per-protocol analysis used, not ITT; attrition >20% at follow-up)	Low risk (Validated BPI (Spanish version); single examiner with >5 years’ experience; assessor blinded)	Low risk (Registered trial (NCT04809987); outcomes specified; analysis plan described)	Some concerns
Selvaraj 2026²⁴	Low risk (Computer-generated block randomization (block size=5); serially numbered sealed opaque envelopes; 1:1 allocation)	Low risk (Double-blind - participants and outcome assessor blinded; both groups used same device; sham control (2D animated films) effectively masks intervention)	Low risk (52 enrolled, 43 completed (82.7% completion); 9 dropouts (17.3%) due to logistical reasons; reasons unrelated to intervention; per-protocol analysis)	Low risk (Validated VAS (0–100 mm); LANSS used for NP screening; independent blinded assessor; MCID threshold (13mm) specified)	Low risk (Outcomes clearly specified (VAS primary, CDS secondary); results reported comprehensively; subgroup analyses clearly identified as exploratory)	Low risk

Open in a new tab

Meta-Analysis Results

The pooled results of 12 comparisons from 10 RCTs^22–24^,^29–35 showed that compared to controls, VR-based intervention was associated with a reduced intensity of neuropathic pain in patients with SCI (SMD: −0.50, 95% CI: −0.76 to −0.25, p < 0.001; Figure 2A) with moderate heterogeneity (Cochrane Q test p = 0.18, I² = 27%). Sensitivity analyses, performed by sequentially excluding individual studies, yielded consistent results (SMD range: −0.42 to −0.60, all p < 0.05).

Three forest plots of standardized mean difference for neuropathic pain comparing virtual reality and control. Image A displays a forest plot comparing virtual reality and control groups across 12 studies. Key metrics include mean differences and 95percentage confidence intervals. Notable effects: Soler 2010 with tDCS (-0.65), Austin 2021 (-0.79), Azurdia 2022 (-1.54), Ji 2022 (-0.99). Total effect is -0.50, CI -0.76 to -0.25, with totals 195 and 184. Image B shows subgroup analysis by study design: Parallel group trials (8 comparisons) subtotal -0.55, CI -0.80 to -0.30, Crossover trials (4 comparisons) subtotal -0.55, CI -1.22 to 0.13. Overall total remains -0.50, CI -0.76 to -0.25. Image C presents subgroup analysis by mean age: <40 years (5 comparisons) subtotal -0.44, CI -0.93 to 0.04, greater than or equal to 40 years (7 comparisons) subtotal -0.53, CI -0.83 to -0.23. Overall total is consistent at -0.50, CI -0.76 to -0.25, with totals 195 and 184. — Forest plots for the meta-analysis evaluating the effect of VR-based intervention versus controls on intensity of neuropathic pain in patients with SCI. (A) overall meta-analysis; (B) subgroup analysis according to the study design; (C) subgroup analysis according to the mean age of the patients.

Subgroup analyses are presented in Figure 2B to Figure 3B. As shown in Figure 2B, similar results were obtained for parallel-group RCTs and crossover RCTs (SMD: −0.55 vs −0.55; p for subgroup difference = 0.99). In Figure 2C, consistent results were also observed for studies with mean ages of the patients < 40 years and ≥ 40 years (SMD: −0.44 vs −0.53; p for subgroup difference = 0.77). In addition, the results were similar for studies of patients with time after SCI < 96 months and ≥ 96 months (SMD: −0.48 vs −0.41; p for subgroup difference = 0.78; Figure 4A). Further subgroup analyses showed that both sensorimotor/embodiment-based VR and distraction/environment-based VR (SMD: −0.44 vs −0.82, p for subgroup difference = 0.36; Figure 4B) yielded better effects on the intensity of neuropathic pain as compared to controls. The results were similar for single-session and multiple-session VR-based interventions (SMD: −0.67 vs −0.43, p for subgroup difference = 0.41; Figure 5A) and compared to sham/virtual control or active non-VR control (SMD: −0.52 vs −0.68, p for subgroup difference = 0.74; Figure 5B). Moreover, consistent results were observed for studies with pain intensity evaluated via NRS, VAS, and BPI (SMD: −0.58, −0.45 vs −0.26, p for subgroup difference = 0.70; Figure 3A), and in low-risk studies and studies of some concerns in RoB 2.0 (SMD: −0.52 vs −0.51, p for subgroup difference = 0.97; Figure 3B).

Two forest plots comparing virtual reality intervention versus control for neuropathic pain intensity subgroups. Image A displays a forest plot categorized by time post-spinal cord injury. The x-axis, labeled Std. Mean Difference, ranges from -2 to 2. Studies are grouped by time: less than 96 months shows effect sizes of -0.65, 0.10, -0.99, -0.26, -0.62, subtotal -0.48 (95percentage CI: -0.90 to -0.06). Greater than or equal to 96 months shows -0.17, -0.20, -0.44, -0.79, -0.49, 0.05, subtotal -0.41 (95percentage CI: -0.73 to -0.09). Overall total is -0.45 (95percentage CI: -0.67 to -0.23). Image B shows a forest plot categorized by virtual reality type. The x-axis, labeled Std. Mean Difference, ranges from -2 to 2. Sensorimotor or embodiment-based VR subgroup shows -0.65, -0.17, 0.10, -0.20, -0.44, -0.99, -0.49, -0.26, -0.62, subtotal -0.44 (95percentage CI: -0.68 to -0.19). Distraction or environment-based VR subgroup shows -0.79, -1.54, 0.05, subtotal -0.82 (95percentage CI: -1.61 to -0.04). Overall total is -0.50 (95percentage CI: -0.76 to -0.25). — Forest plots for the subgroup analyses evaluating the effect of VR-based intervention versus controls on intensity of neuropathic pain in patients with SCI. (A) subgroup analysis according to scales for evaluating pain intensity; (B) subgroup analysis according to the risk of bias of RCTs.

Paired forest plots of VR vs control for neuropathic pain, overall SMD total minus 0.50. The image A showing a forest plot subgrouped by session frequency. The x-axis label is, Std. Mean Difference, with no unit, ranging from minus 2 to 2 in increments of 1. The y-axis label is, Study or Subgroup, with no unit. Single session studies and effect sizes: Pozeg 2017 minus 0.20, 95 percent confidence interval minus 1.04 to 0.64, Richardson 2019 minus 0.44, minus 0.95 to 0.08, Austin 2021 minus 0.79, minus 1.51 to minus 0.07, Azurdia 2022 minus 1.54, minus 2.52 to minus 0.57. Subtotal minus 0.67, minus 1.15 to minus 0.19. Multiple session studies: Soler 2010 with tDCS minus 0.65, minus 1.55 to 0.26, Soler 2010 no tDCS minus 0.17, minus 1.07 to 0.73, Ozkul 2015 0.10, minus 0.47 to 0.67, Ji 2022 minus 0.99, minus 1.52 to minus 0.46, Molla Casanova 2026 minus 0.26, minus 0.99 to 0.48, Tabacof 2024 scenic 0.05, minus 1.17 to 1.28, Tabacof 2024 somatic minus 0.49, minus 1.72 to 0.73, Selvaraj 2026 minus 0.62, minus 1.23 to minus 0.00. Subtotal minus 0.43, minus 0.74 to minus 0.11. Total minus 0.50, minus 0.76 to minus 0.25. The image B showing a forest plot subgrouped by control type. The x-axis label is, Std. Mean Difference, with no unit, ranging from minus 2 to 2 in increments of 1. The y-axis label is, Study or Subgroup, with no unit. Sham virtual control studies: Soler 2010 with tDCS minus 0.65, minus 1.55 to 0.26, Soler 2010 no tDCS minus 0.17, minus 1.07 to 0.73, Pozeg 2017 minus 0.20, minus 1.04 to 0.64, Richardson 2019 minus 0.44, minus 0.95 to 0.08, Ji 2022 minus 0.99, minus 1.52 to minus 0.46, Tabacof 2024 somatic minus 0.49, minus 1.72 to 0.73, Molla Casanova 2026 minus 0.26, minus 0.99 to 0.48, Tabacof 2024 scenic 0.05, minus 1.17 to 1.28, Selvaraj 2026 minus 0.62, minus 1.23 to minus 0.00. Subtotal minus 0.52, minus 0.76 to minus 0.28. Active non virtual reality control studies: Ozkul 2015 0.10, minus 0.47 to 0.67, Austin 2021 minus 0.79, minus 1.51 to minus 0.07, Azurdia 2022 minus 1.54, minus 2.52 to minus 0.57. Subtotal minus 0.68, minus 1.60 to 0.24. Total minus 0.50, minus 0.76 to minus 0.25. — Forest plots for the subgroup analyses evaluating the effect of VR-based intervention versus controls on intensity of neuropathic pain in patients with SCI. (A) subgroup analysis according to the time after SCI onset; (B) subgroup analysis according to the type of VR-based intervention.

Two forest plots comparing VR and control effects on neuropathic pain intensity across subgroups. A forest plot illustrates neuropathic pain intensity using Std. Mean Difference (-2 to 2) on the x-axis. Studies are grouped by NRS, VAS and BPI on the y-axis. NRS results: Soler 2010 with tDCS -0.65, no tDCS -0.17, Richardson 2019 -0.44, Austin 2021 -0.79, Azurdia 2022 -1.54, Tabacof 2024 somatic -0.49, scenic 0.09. NRS subtotal: -0.58. VAS results: Ozkul 2015 0.10, Pozeg 2017 -0.20, Ji 2022 -0.99, Selvaraj 2026 -0.62. VAS subtotal: -0.45. BPI result: Molla Casanova 2026 -0.26, subtotal: -0.26. Overall effect: -0.50. Another plot shows risk of bias with Std. Mean Difference (-2 to 2) on the x-axis. Low risk effects: Soler 2010 with tDCS -0.65, no tDCS -0.17, Selvaraj 2026 -0.62, subtotal: -0.52. Some concerns effects: Ozkul 2015 0.10, Pozeg 2017 -0.20, Richardson 2019 -0.44, Austin 2021 -0.79, Azurdia 2022 -1.54, Ji 2022 -0.99, Tabacof 2024 somatic -0.49, scenic 0.09, Molla Casanova 2026 -0.26, subtotal: -0.51. Overall effect: -0.50. — Forest plots for the subgroup analyses evaluating the effect of VR-based intervention versus controls on intensity of neuropathic pain in patients with SCI. (A) subgroup analysis according to the sessions of VR-based intervention; (B) subgroup analysis according to the type of control.

Univariate meta-regression analyses were performed to explore potential sources of heterogeneity (Table 3). Results showed that none of the evaluated study characteristics, including sample size, mean age of the patients, proportion of men, time after SCI, or duration of intervention, was significantly associated with the effect estimates (all p > 0.05).

Table 3.

Results of Univariate Meta-Regression Analysis

Variables	SMD for the Influence of VR-Based Intervention on NP After Patients with SCI
Variables	Coefficient	95% CI	p Values	Adjusted R²
Sample size	−0.0051	−0.0205 to 0.0102	0.47	0%
Mean age (years)	−0.0098	−0.0559 to 0.0363	0.65	0%
Men (%)	0.011	−0.010 to 0.032	0.29	7.2%
Time since SCI (months)	0.00081	−0.00300 to 0.00463	0.64	0%
Intervention duration (days)	−0.00030	−0.02037 to 0.01976	0.97	0%

Open in a new tab

Abbreviations: CI, confidence interval; NP, neuropathic pain; R², proportion of between-study variance explained; SCI, spinal cord injury; SMD, standardized mean difference; VR, virtual reality.

Publication Bias

Funnel plots assessing publication bias for the meta-analysis of the intensity of neuropathic pain in patients with SCI are presented in Figure 6. Visual inspection suggested an approximately symmetrical distribution of the included studies, indicating a low likelihood of publication bias. This was further supported by Egger’s regression test, which did not detect significant asymmetry (p = 0.83). However, these findings should be interpreted with caution, as the number of included studies was relatively small (k = 12), which may limit the statistical power to detect publication bias.

Certainty of Evidence

The certainty of evidence for the primary outcome was assessed using the GRADE approach (Table 4). Overall, the certainty of evidence was rated as moderate, primarily due to concerns regarding risk of bias, as most of the included studies were judged as having some concerns according to the RoB 2.0 assessment. Inconsistency, indirectness, and imprecision were not considered serious, given the low statistical heterogeneity (I² = 27%), direct relevance of the included studies to the review question, and a pooled estimate with a 95% CI that did not cross the null effect. No substantial publication bias was evident from visual inspection of the funnel plot. Overall, VR-based interventions probably reduce neuropathic pain intensity in individuals with SCI compared with control interventions, although further well-designed, adequately powered RCTs may influence the certainty and magnitude of this effect.

Table 4.

Summary of Findings and Certainty of Evidence (GRADE)

Outcome	No. of Participants (Studies)	Study Design	Risk of Bias	Inconsistency	Indirectness	Imprecision	Publication Bias	Relative Effect:	Certainty of Evidence (GRADE)	Comments
Neuropathic pain intensity in patients with SCI after VR-based interventions compared with controls	317 participants (10 RCTs, 12 comparisons)	RCTs	Serious – Most included trials were judged as having “some concerns” according to RoB 2.0, primarily because of lack of blinding, potential selective reporting, or missing outcome data, although all studies had a low risk of bias in the randomization process and outcome measurement.	Not serious – Statistical heterogeneity was low (I² = 27%), and the direction of the overall effect generally favored VR-based interventions; subgroup and meta-regression analyses were also performed to explore potential sources of heterogeneity.	Not serious – The included studies directly evaluated the population, interventions, comparators, and outcome relevant to the review question.	Not serious – The pooled 95% CI excluded the null effect and was sufficiently precise to support a beneficial effect of VR-based interventions.	Not serious – The funnel plot appeared approximately symmetrical, with no clear evidence of substantial publication bias.	SMD (95% CI): −0.50 (−0.76 to −0.25)	⨁⨁⨁◯ Moderate	VR-based interventions probably reduce neuropathic pain intensity in individuals with SCI compared with control interventions; however, the certainty of evidence is limited by concerns regarding risk of bias in several included trials.

Open in a new tab

Notes: GRADE certainty ratings: ⨁⨁⨁⨁, high certainty; ⨁⨁⨁◯, moderate certainty; ⨁⨁◯◯, low certainty; ⨁◯◯◯, very low certainty. Specific reasons for each GRADE domain: Risk of bias: Downgraded by one level because most included studies were judged as having some concerns according to RoB 2.0, mainly owing to lack of blinding, potential selective reporting, or missing outcome data. Inconsistency: Not downgraded because overall statistical heterogeneity was low (I² = 27%), and no substantial unexplained inconsistency was identified. Indirectness: Not downgraded because the included RCTs directly assessed VR-based interventions for neuropathic pain in individuals with SCI. Imprecision: Not downgraded because the pooled 95% CI excluded the null effect and supported a beneficial effect, although the overall evidence base remained relatively small. Publication bias: Not downgraded because visual inspection of the funnel plot did not indicate substantial asymmetry.

Abbreviations: CI, confidence interval; GRADE, Grading of Recommendations Assessment, Development and Evaluation; RCTs, randomized controlled trials; RoB 2.0, revised Cochrane risk of bias tool version 2; SCI, spinal cord injury; SMD, standardized mean difference; VR, virtual reality.

Discussion

Our meta-analysis provides updated randomized evidence supporting the use of VR-based interventions for reducing neuropathic pain intensity in individuals with SCI. By synthesizing data exclusively from RCTs, we found that VR-based interventions were associated with a moderate reduction in neuropathic pain compared with control conditions, with consistent findings across sensitivity analyses and moderate certainty of evidence according to the GRADE framework. Importantly, no significant differences in treatment effects were observed across prespecified subgroup analyses or meta-regression analyses, suggesting that the analgesic benefits of VR were generally consistent across different patient characteristics and intervention approaches. These findings strengthen the evidence supporting VR as a promising adjunctive non-pharmacological intervention for SCI-related neuropathic pain while providing a more rigorous estimate than previous reviews that combined randomized and uncontrolled studies.

Although the exact mechanisms underlying VR-mediated analgesia remain incompletely understood, several complementary neurophysiological and psychological mechanisms may explain the observed benefits. First, VR provides immersive multisensory stimulation that competes with nociceptive processing for attentional resources, thereby reducing pain perception through cognitive distraction.³⁶^,³⁷ Unlike conventional distraction techniques, immersive VR creates a strong sense of presence within a virtual environment, enabling patients to temporarily disengage from ongoing pain experiences.³⁸ Second, several included interventions incorporated virtual walking, virtual limb embodiment, or visually induced movement illusions, which may restore congruence between motor intention, visual feedback, and body representation.³⁹ Following SCI, disruption of sensorimotor pathways may contribute to maladaptive cortical reorganization and altered body representation, both of which have been implicated in the development and maintenance of neuropathic pain.⁴⁰^,⁴¹ Providing visually coherent representations of movement despite absent motor output may partially normalize cortical sensory-motor integration and reduce the mismatch between intended and perceived movement.³³ Third, VR-based interventions promote repetitive multisensory stimulation and motor imagery, both of which may facilitate adaptive neuroplasticity within sensorimotor networks.⁴²^,⁴³ Experimental studies have demonstrated that motor imagery and action observation activate cortical regions involved in movement planning and execution, even in the absence of actual movement, potentially promoting beneficial cortical reorganization.⁴⁴^,⁴⁵ Finally, VR interventions often enhance motivation, engagement, and treatment adherence compared with conventional rehabilitation by providing interactive, goal-directed, and enjoyable therapeutic experiences.⁴⁶^,⁴⁷ Improved adherence may increase treatment intensity and repetition, which are considered important drivers of neuroplastic adaptation during neurological rehabilitation.⁴⁸

Interestingly, subgroup analyses demonstrated no significant differences according to the type of VR intervention, including sensorimotor or embodiment-based interventions and distraction or environment-based interventions. Although these approaches are believed to operate through partially different mechanisms, the present findings suggest that both strategies may ultimately converge on reducing central pain processing. Sensorimotor-based VR may primarily exert its effects through restoration of body ownership, normalization of sensorimotor integration, and modulation of maladaptive cortical reorganization,⁴⁹ whereas distraction-oriented VR may predominantly reduce pain through attentional diversion, emotional engagement, and cognitive modulation of nociceptive processing.⁵⁰ Because both approaches provide immersive multisensory experiences and enhance patient engagement, their overall analgesic effects may be comparable despite mechanistic differences. Nevertheless, the relatively small number of studies within each subgroup limits statistical power to detect modest differences between intervention types, and future adequately powered comparative trials are needed to determine whether specific VR paradigms provide superior analgesic benefits.

Similarly, no significant differences were observed between single-session and multiple-session interventions. Several explanations may account for this finding. Single-session VR interventions may produce immediate reductions in pain through attentional distraction and altered pain perception during or shortly after treatment,⁵¹ whereas repeated interventions may additionally induce cumulative neuroplastic adaptations over time.⁵² The apparent similarity between subgroup estimates should therefore not be interpreted as evidence that treatment duration is unimportant. Rather, the limited number of available RCTs, variability in intervention protocols, and differences in session frequency, duration, and total treatment dose may have reduced the ability to detect dose-response relationships. Indeed, intervention protocols varied substantially across studies, ranging from brief single-session visual illusion paradigms to multi-week rehabilitation programs incorporating immersive VR training. Standardized reporting of intervention dose and fidelity will be essential for future investigations.

The absence of significant associations in the meta-regression analyses further suggests that study-level characteristics, including sample size, mean participant age, proportion of men, time since SCI, and intervention duration, did not significantly modify treatment effects. These findings should be interpreted cautiously because meta-regression analyses based on a limited number of studies are inherently underpowered and susceptible to ecological bias. Furthermore, important patient-level characteristics that may influence treatment response, such as neurological level of injury, completeness of injury, baseline pain severity, duration of neuropathic pain, concomitant analgesic medications, psychological status, and previous rehabilitation exposure, could not be evaluated because individual participant data were unavailable. Consequently, the absence of statistically significant effect modification should not be interpreted as evidence that these factors are clinically irrelevant.

Our study has several important strengths. To our knowledge, this is the first meta-analysis restricted exclusively to RCTs evaluating VR-based interventions for SCI-related neuropathic pain, thereby minimizing the bias associated with uncontrolled pre-post studies included in previous reviews. The literature search was comprehensive and up to date, incorporating several recently published RCTs that substantially expanded the available evidence. We performed extensive subgroup, sensitivity, and meta-regression analyses to evaluate the robustness of the findings and explore potential sources of heterogeneity. Risk of bias was assessed using the contemporary RoB 2.0 tool, and the certainty of evidence was evaluated using the GRADE approach, providing a transparent assessment of the strength of the available evidence. Several limitations should also be acknowledged. First, although statistical heterogeneity was relatively low, substantial clinical heterogeneity existed across studies regarding VR hardware, degree of immersion, intervention content, treatment duration, session frequency, comparator interventions, and rehabilitation settings. Such variability may have influenced treatment effects but could not be fully explored because of the limited number of available studies. Second, most included RCTs enrolled relatively small samples, increasing the possibility of imprecise effect estimates and limiting statistical power for subgroup and meta-regression analyses. Third, although all included studies were randomized, most were judged as having some concerns according to the RoB 2.0 assessment, primarily related to the inherent difficulty of participant blinding, incomplete outcome data, or potential selective reporting. Fourth, comparator interventions varied considerably, ranging from sham visual conditions to conventional rehabilitation and other active interventions, which may have influenced the magnitude of the pooled treatment effect. Finally, most studies evaluated only short-term outcomes immediately after intervention, precluding robust conclusions regarding the durability of analgesic benefits over longer follow-up periods.

The present findings have several important clinical implications. Given the modest efficacy and potential adverse effects of pharmacological therapies for SCI-related neuropathic pain, VR-based interventions may represent a valuable adjunctive treatment within multidisciplinary rehabilitation programs. Their generally good tolerability reported in the included trials, ability to enhance patient engagement, and increasing accessibility through commercially available VR technologies further support their clinical potential.⁵³ Nevertheless, considering the methodological limitations of the available evidence, the current findings should be interpreted cautiously. Future large-scale, multicenter RCTs employing standardized VR protocols, consistent control interventions, longer follow-up durations, and comprehensive reporting of intervention characteristics are needed. Comparative effectiveness studies evaluating different VR paradigms, immersion levels, treatment doses, and combinations with pharmacological or rehabilitation therapies would further clarify the optimal implementation of VR in clinical practice. In addition, mechanistic studies incorporating neuroimaging, neurophysiological assessments, and patient-reported outcomes may help elucidate the biological pathways underlying VR-mediated analgesia and identify patients most likely to benefit.

Conclusions

In conclusion, the current evidence indicates that VR-based interventions probably reduce neuropathic pain intensity in individuals with SCI compared with control interventions. Although the certainty of evidence was moderate and the findings were generally robust across multiple analyses, further high-quality, adequately powered RCTs with standardized intervention protocols and longer follow-up are warranted to confirm the magnitude, durability, and optimal delivery of VR-based analgesic interventions for SCI-related neuropathic pain.

Funding Statement

No funding was received for this study.

Data Sharing Statement

This study protocol is registered with PROSPERO (registration number: CRD420261433222). The data used and/or analyzed in this study may be obtained upon reasonable request to the corresponding author, Chang Liu.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The author declares that the research, writing, and/or publication of this paper did not receive any financial support.

References

1.Lu Y, Shang Z, Zhang W, et al. Global incidence and characteristics of spinal cord injury since 2000–2021: a systematic review and meta-analysis. BMC Med. 2024;22(1):285. doi: 10.1186/s12916-024-03514-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Anjum A, Yazid MD, Fauzi Daud M, et al. Spinal Cord Injury: pathophysiology, Multimolecular Interactions, and Underlying Recovery Mechanisms. Int J Mol Sci. 2020;21(20):7533. doi: 10.3390/ijms21207533 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Salehian F, Aarabi A, Rajai Firouzabadi S, et al. Prevalence of neuropathic pain following spinal cord injury: an updated systematic review and meta-analysis. J Clin Neurosci. 2025;142:111660. doi: 10.1016/j.jocn.2025.111660 [DOI] [PubMed] [Google Scholar]
4.Li J, Kang W, Wang X, Pan F. Progress in treatment of pathological neuropathic pain after spinal cord injury. Front Neurol. 2024;15:1430288. doi: 10.3389/fneur.2024.1430288 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Tao Z, Zhou X, Jin L, et al. Neuropathic pain after spinal cord injury: mechanisms, animal models and pain assessments. Exp Neurol. 2026;399:115678. doi: 10.1016/j.expneurol.2026.115678 [DOI] [PubMed] [Google Scholar]
6.Rugnath R, Orzechowicz C, Newell C, Carullo V, Rugnath A. A Literature Review: the Mechanisms and Treatment of Neuropathic Pain-A Brief Discussion. Biomedicines. 2024;12(1):204. doi: 10.3390/biomedicines12010204 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Widerström-Noga E. Neuropathic Pain and Spinal Cord Injury: management, Phenotypes, and Biomarkers. Drugs. 2023;83(11):1001–21. doi: 10.1007/s40265-023-01903-7 [DOI] [PubMed] [Google Scholar]
8.Lorkowski J, Raulinajtys-Grzybek M, Pokorski M. Virtual Reality-Based Rehabilitation Therapy. Physiol Res. 2025;74(4):563–569. doi: 10.33549/physiolres.935590 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Fusaro M, Lisi MP, Era V, et al. The Transformative Power of Virtual Reality: redefining Interactions in Virtual Platforms, Education, Healthcare, and Workplaces. Topoi. 2025;44(4):1071–1086. doi: 10.1007/s11245-025-10216-1 [DOI] [Google Scholar]
10.Ion LA, Săndulescu MI, Potcovaru CG, et al. Immersive Virtual Reality for Stroke Rehabilitation: linking Clinical and Digital Measures of Motor Recovery-A Pilot Study. Bioengineering. 2026;13(1):59. doi: 10.3390/bioengineering13010059 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Kong Y, Shu Z, Han YS. Global Trends in Virtual Reality Research on Motor Rehabilitation from 2005 to 2025: a Bibliometric Analysis. Healthcare (Basel). 2026;14(13). doi: 10.3390/healthcare14131976 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Yeo E, Chau B, Chi B, Ruckle DE, Ta P. Virtual Reality Neurorehabilitation for Mobility in Spinal Cord Injury: a Structured Review. Innov Clin Neurosci. 2019;16(1–2):13–20. [PMC free article] [PubMed] [Google Scholar]
13.Scalise M, Bora TS, Zancanella C, Safa A, Stefini R, Cannizzaro D. Virtual Reality as a Therapeutic Tool in Spinal Cord Injury Rehabilitation: a Comprehensive Evaluation and Systematic Review. J Clin Med. 2024;13(18):5429. doi: 10.3390/jcm13185429 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Wang L, Zhang H, Ai H, Liu Y. Effects of virtual reality rehabilitation after spinal cord injury: a systematic review and meta-analysis. J Neuroeng Rehabil. 2024;21(1):191. doi: 10.1186/s12984-024-01492-w [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Li J, Yang H, Xiao Y, et al. The analgesic effects and neural oscillatory mechanisms of virtual reality scenes based on distraction and mindfulness strategies in human volunteers. Br J Anaesth. 2023;131(6):1082–1092. doi: 10.1016/j.bja.2023.09.001 [DOI] [PubMed] [Google Scholar]
16.Masood RZ, Ali M, Manzoor SF, Sajid EU, Kalam H. Virtual reality as a transformative tool in pain management. Ann Med Surg (Lond). 2025;87(9):5368–5369. doi: 10.1097/ms9.0000000000003630 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Li Q, Du X, Zhao D, et al. Virtual Reality Interventions for Spinal Cord Injury-Related Neuropathic Pain: a Comprehensive Scoping Review of Multidimensional Outcomes. J Pain Res. 2025;18:5175–5190. doi: 10.2147/jpr.s547243 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.De Miguel-Rubio A, Rubio MD, Salazar A, Camacho R, Lucena-Anton D. Effectiveness of Virtual Reality on Functional Performance after Spinal Cord Injury: a Systematic Review and Meta-Analysis of Randomized Controlled Trials. J Clin Med. 2020;9(7):2065. doi: 10.3390/jcm9072065 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Miguel-Rubio A, Rubio MD, Salazar A, et al. Is Virtual Reality Effective for Balance Recovery in Patients with Spinal Cord Injury? A Systematic Review and Meta-Analysis. J Clin Med. 2020;9(9):2861. doi: 10.3390/jcm9092861 [DOI] [PMC free article] [PubMed] [Google Scholar]
20.De Miguel-Rubio A, Rubio MD, Alba-Rueda A, Salazar A, Moral-Munoz JA, Lucena-Anton D. Virtual Reality Systems for Upper Limb Motor Function Recovery in Patients With Spinal Cord Injury: systematic Review and Meta-Analysis. JMIR Mhealth Uhealth. 2020;8(12):e22537. doi: 10.2196/22537 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Gao F, Yu T, Song J, Chen Z, Hu D. Effectiveness of virtual reality in treating neuropathic pain in patients with spinal cord injury: a systematic review and meta-analysis. J Spinal Cord Med. 2026;2026:1–13. doi: 10.1080/10790268.2026.2626118 [DOI] [PubMed] [Google Scholar]
22.Ji HB, Li YK, Xing SX. Effects of Virtual Walking on Spinal Cord Injury-Related Neuropathic Pain. J Cervicodynia Lumbodynia. 2022;43(3):398–400. doi: 10.3969/j.issn.1005-7234.2022.03.026 [DOI] [Google Scholar]
23.Mollà-Casanova S, Muñoz-Gómez E, Sempere-Rubio N, et al. Effect of virtual walking on chronic pain, tone and spasticity in incomplete spinal cord injury. Disabil Rehabil. 2026;48(9):2814–2826. doi: 10.1080/09638288.2025.2583905 [DOI] [PubMed] [Google Scholar]
24.Selvaraj JS, Prakash H, Augustine TA, Kumar SSK, Velkumar S, Amalraj JA. Immersive virtual reality for neuropathic pain management in spinal cord injury: a randomized controlled trial. J Spinal Cord Med. 2026;49(3):543–555. doi: 10.1080/10790268.2025.2514321 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Higgins J, Thomas J, Chandler J, et al. Cochrane Handbook for Systematic Reviews of Interventions Version 6.2; The Cochrane Collaboration; 2021. Available from: www.training.cochrane.org/handbook. [Google Scholar]
26.Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med. 2002;21(11):1539–1558. doi: 10.1002/sim.1186 [DOI] [PubMed] [Google Scholar]
27.Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315(7109):629–634. [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Guyatt G, Oxman AD, Akl EA, et al. GRADE guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables. J Clin Epidemiol. 2011;64(4):383–394. doi: 10.1016/j.jclinepi.2010.04.026 [DOI] [PubMed] [Google Scholar]
29.Soler MD, Kumru H, Pelayo R, et al. Effectiveness of transcranial direct current stimulation and visual illusion on neuropathic pain in spinal cord injury. Brain. 2010;133(9):2565–2577. doi: 10.1093/brain/awq184 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Özkul Ç, Kılınç M, Yıldırım SA, Topçuoğlu EY, Akyüz M. Effects of visual illusion and transcutaneous electrical nerve stimulation on neuropathic pain in patients with spinal cord injury: a randomised controlled cross-over trial. J Back Musculoskelet Rehabil. 2015;28(4):709–719. doi: 10.3233/bmr-140573 [DOI] [PubMed] [Google Scholar]
31.Pozeg P, Palluel E, Ronchi R, et al. Virtual reality improves embodiment and neuropathic pain caused by spinal cord injury. Neurology. 2017;89(18):1894–1903. doi: 10.1212/WNL.0000000000004585 [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Richardson EJ, McKinley EC, Rahman A, Klebine P, Redden DT, Richards JS. Effects of virtual walking on spinal cord injury-related neuropathic pain: a randomized, controlled trial. Rehabil Psychol. 2019;64(1):13–24. doi: 10.1037/rep0000246 [DOI] [PubMed] [Google Scholar]
33.Austin PD, Siddall PJ. Virtual reality for the treatment of neuropathic pain in people with spinal cord injuries: a scoping review. J Spinal Cord Med. 2021;44(1):8–18. doi: 10.1080/10790268.2019.1575554 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Azurdia D, Acuña SM, Narasaki-Jara M, Furtado Jr O, Jung T. Effects of Virtual Reality-Based Aerobic Exercise on Perceptions of Pain and Fatigue in Individuals with Spinal Cord Injury. Games Health J. 2022;11(4):236–241. doi: 10.1089/g4h.2021.0151 [DOI] [PubMed] [Google Scholar]
35.Tabacof L, Salazar SI, Breyman E, et al. Immersive virtual reality for chronic neuropathic pain after spinal cord injury: a pilot, randomized, controlled trial. Pain Rep. 2024;9(6):e1173. doi: 10.1097/pr9.0000000000001173 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Li A, Montaño Z, Chen VJ, Gold JI. Virtual reality and pain management: current trends and future directions. Pain Manag. 2011;1(2):147–157. doi: 10.2217/pmt.10.15 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Preti K, Pinero M, Weitzman M, Malhotra V. Virtual reality: treating pain and anxiety. Am Nurse J. 2024;19(5):42–46. doi: 10.51256/anj052442 [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Wiederhold BK, Gao K, Sulea C, Wiederhold MD. Virtual reality as a distraction technique in chronic pain patients. Cyberpsychol Behav Soc Netw. 2014;17(6):346–352. doi: 10.1089/cyber.2014.0207 [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Dai Y, Li J, Li Y, et al. A scoping review on the role of virtual walking intervention in enhancing wellness. NPJ Digit Med. 2025;8(1):217. doi: 10.1038/s41746-025-01609-y [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Vastano R, Costantini M, Widerstrom-Noga E. Maladaptive reorganization following SCI: the role of body representation and multisensory integration. Prog Neurobiol. 2022;208:102179. doi: 10.1016/j.pneurobio.2021.102179 [DOI] [PubMed] [Google Scholar]
41.Diana L, Sritharan J, Vallesi V, Brunello N, Verma RK, Zito GA. Chronic spinal cord injury is associated with morphometric brain changes in functional networks beyond the sensorimotor system. Front Neurol. 2025;16:1672506. doi: 10.3389/fneur.2025.1672506 [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Wankhede NL, Koppula S, Ballal S, et al. Virtual reality modulating dynamics of neuroplasticity: innovations in neuro-motor rehabilitation. Neuroscience. 2025;566:97–111. doi: 10.1016/j.neuroscience.2024.12.040 [DOI] [PubMed] [Google Scholar]
43.Xie Q, Xiong X, Liu M, et al. Effects of multi-sensory virtual reality training on gait adaptability and somatomotor network remodeling in patients with stroke: a randomized controlled trial. J Neuroeng Rehabil. 2026;23(1). doi: 10.1186/s12984-026-02003-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Eaves DL, Riach M, Holmes PS, Wright DJ. Motor Imagery during Action Observation: a Brief Review of Evidence, Theory and Future Research Opportunities. Front Neurosci. 2016;10:514. doi: 10.3389/fnins.2016.00514 [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Mustile M, Kourtis D, Edwards MG, Donaldson DI, Ietswaart M. Neural correlates of motor imagery and execution in real-world dynamic behavior: evidence for similarities and differences. Front Hum Neurosci. 2024;18:1412307. doi: 10.3389/fnhum.2024.1412307 [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Eid M, Alkerdasy H, Hilmy A, Sameh A, Ashraf A. Effectiveness of virtual reality-based exercise rehabilitation for enhancing functional outcomes and quality of life in adult cancer survivors: a systematic review of randomized controlled trials. J Cancer Surviv. 2025;2025:1. doi: 10.1007/s11764-025-01923-0 [DOI] [PubMed] [Google Scholar]
47.AlHossan AM, Jahhaf RH, Alharbi AS, et al. Digital and virtual reality-based rehabilitation versus conventional therapy for rotator cuff tears and post-repair recovery: a systematic review and meta-analysis. JSES Rev Rep Tech. 2026;6(1):100584. doi: 10.1016/j.xrrt.2025.09.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Moumeni IN. Intensity over duration in neurological rehabilitation: exploring evidence for optimised recovery paradigms. Front Neurol. 2025;16:1697186. doi: 10.3389/fneur.2025.1697186 [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Adamovich SV, Fluet GG, Tunik E, Merians AS. Sensorimotor training in virtual reality: a review. NeuroRehabilitation. 2009;25(1):29–44. doi: 10.3233/nre-2009-0497 [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Indovina P, Barone D, Gallo L, Chirico A, De Pietro G, Giordano A. Virtual Reality as a Distraction Intervention to Relieve Pain and Distress During Medical Procedures: a Comprehensive Literature Review. Clin J Pain. 2018;34(9):858–877. doi: 10.1097/ajp.0000000000000599 [DOI] [PubMed] [Google Scholar]
51.Pretat T, Koller C, Hügle T. Virtual reality as a treatment for chronic musculoskeletal pain syndromes. Joint Bone Spine. 2025;92(1):105769. doi: 10.1016/j.jbspin.2024.105769 [DOI] [PubMed] [Google Scholar]
52.Goudman L, Jansen J, Billot M, et al. Virtual Reality Applications in Chronic Pain Management: systematic Review and Meta-analysis. JMIR Serious Games. 2022;10(2):e34402. doi: 10.2196/34402 [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Kouijzer M, Kip H, Bouman YHA, Kelders SM. Implementation of virtual reality in healthcare: a scoping review on the implementation process of virtual reality in various healthcare settings. Implement Sci Commun. 2023;4(1):67. doi: 10.1186/s43058-023-00442-2 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

[cit0001] 1.Lu Y, Shang Z, Zhang W, et al. Global incidence and characteristics of spinal cord injury since 2000–2021: a systematic review and meta-analysis. BMC Med. 2024;22(1):285. doi: 10.1186/s12916-024-03514-9 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0002] 2.Anjum A, Yazid MD, Fauzi Daud M, et al. Spinal Cord Injury: pathophysiology, Multimolecular Interactions, and Underlying Recovery Mechanisms. Int J Mol Sci. 2020;21(20):7533. doi: 10.3390/ijms21207533 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0003] 3.Salehian F, Aarabi A, Rajai Firouzabadi S, et al. Prevalence of neuropathic pain following spinal cord injury: an updated systematic review and meta-analysis. J Clin Neurosci. 2025;142:111660. doi: 10.1016/j.jocn.2025.111660 [DOI] [PubMed] [Google Scholar]

[cit0004] 4.Li J, Kang W, Wang X, Pan F. Progress in treatment of pathological neuropathic pain after spinal cord injury. Front Neurol. 2024;15:1430288. doi: 10.3389/fneur.2024.1430288 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0005] 5.Tao Z, Zhou X, Jin L, et al. Neuropathic pain after spinal cord injury: mechanisms, animal models and pain assessments. Exp Neurol. 2026;399:115678. doi: 10.1016/j.expneurol.2026.115678 [DOI] [PubMed] [Google Scholar]

[cit0006] 6.Rugnath R, Orzechowicz C, Newell C, Carullo V, Rugnath A. A Literature Review: the Mechanisms and Treatment of Neuropathic Pain-A Brief Discussion. Biomedicines. 2024;12(1):204. doi: 10.3390/biomedicines12010204 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0007] 7.Widerström-Noga E. Neuropathic Pain and Spinal Cord Injury: management, Phenotypes, and Biomarkers. Drugs. 2023;83(11):1001–21. doi: 10.1007/s40265-023-01903-7 [DOI] [PubMed] [Google Scholar]

[cit0008] 8.Lorkowski J, Raulinajtys-Grzybek M, Pokorski M. Virtual Reality-Based Rehabilitation Therapy. Physiol Res. 2025;74(4):563–569. doi: 10.33549/physiolres.935590 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0009] 9.Fusaro M, Lisi MP, Era V, et al. The Transformative Power of Virtual Reality: redefining Interactions in Virtual Platforms, Education, Healthcare, and Workplaces. Topoi. 2025;44(4):1071–1086. doi: 10.1007/s11245-025-10216-1 [DOI] [Google Scholar]

[cit0010] 10.Ion LA, Săndulescu MI, Potcovaru CG, et al. Immersive Virtual Reality for Stroke Rehabilitation: linking Clinical and Digital Measures of Motor Recovery-A Pilot Study. Bioengineering. 2026;13(1):59. doi: 10.3390/bioengineering13010059 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0011] 11.Kong Y, Shu Z, Han YS. Global Trends in Virtual Reality Research on Motor Rehabilitation from 2005 to 2025: a Bibliometric Analysis. Healthcare (Basel). 2026;14(13). doi: 10.3390/healthcare14131976 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0012] 12.Yeo E, Chau B, Chi B, Ruckle DE, Ta P. Virtual Reality Neurorehabilitation for Mobility in Spinal Cord Injury: a Structured Review. Innov Clin Neurosci. 2019;16(1–2):13–20. [PMC free article] [PubMed] [Google Scholar]

[cit0013] 13.Scalise M, Bora TS, Zancanella C, Safa A, Stefini R, Cannizzaro D. Virtual Reality as a Therapeutic Tool in Spinal Cord Injury Rehabilitation: a Comprehensive Evaluation and Systematic Review. J Clin Med. 2024;13(18):5429. doi: 10.3390/jcm13185429 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0014] 14.Wang L, Zhang H, Ai H, Liu Y. Effects of virtual reality rehabilitation after spinal cord injury: a systematic review and meta-analysis. J Neuroeng Rehabil. 2024;21(1):191. doi: 10.1186/s12984-024-01492-w [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0015] 15.Li J, Yang H, Xiao Y, et al. The analgesic effects and neural oscillatory mechanisms of virtual reality scenes based on distraction and mindfulness strategies in human volunteers. Br J Anaesth. 2023;131(6):1082–1092. doi: 10.1016/j.bja.2023.09.001 [DOI] [PubMed] [Google Scholar]

[cit0016] 16.Masood RZ, Ali M, Manzoor SF, Sajid EU, Kalam H. Virtual reality as a transformative tool in pain management. Ann Med Surg (Lond). 2025;87(9):5368–5369. doi: 10.1097/ms9.0000000000003630 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0017] 17.Li Q, Du X, Zhao D, et al. Virtual Reality Interventions for Spinal Cord Injury-Related Neuropathic Pain: a Comprehensive Scoping Review of Multidimensional Outcomes. J Pain Res. 2025;18:5175–5190. doi: 10.2147/jpr.s547243 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0018] 18.De Miguel-Rubio A, Rubio MD, Salazar A, Camacho R, Lucena-Anton D. Effectiveness of Virtual Reality on Functional Performance after Spinal Cord Injury: a Systematic Review and Meta-Analysis of Randomized Controlled Trials. J Clin Med. 2020;9(7):2065. doi: 10.3390/jcm9072065 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0019] 19.Miguel-Rubio A, Rubio MD, Salazar A, et al. Is Virtual Reality Effective for Balance Recovery in Patients with Spinal Cord Injury? A Systematic Review and Meta-Analysis. J Clin Med. 2020;9(9):2861. doi: 10.3390/jcm9092861 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0020] 20.De Miguel-Rubio A, Rubio MD, Alba-Rueda A, Salazar A, Moral-Munoz JA, Lucena-Anton D. Virtual Reality Systems for Upper Limb Motor Function Recovery in Patients With Spinal Cord Injury: systematic Review and Meta-Analysis. JMIR Mhealth Uhealth. 2020;8(12):e22537. doi: 10.2196/22537 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0021] 21.Gao F, Yu T, Song J, Chen Z, Hu D. Effectiveness of virtual reality in treating neuropathic pain in patients with spinal cord injury: a systematic review and meta-analysis. J Spinal Cord Med. 2026;2026:1–13. doi: 10.1080/10790268.2026.2626118 [DOI] [PubMed] [Google Scholar]

[cit0022] 22.Ji HB, Li YK, Xing SX. Effects of Virtual Walking on Spinal Cord Injury-Related Neuropathic Pain. J Cervicodynia Lumbodynia. 2022;43(3):398–400. doi: 10.3969/j.issn.1005-7234.2022.03.026 [DOI] [Google Scholar]

[cit0023] 23.Mollà-Casanova S, Muñoz-Gómez E, Sempere-Rubio N, et al. Effect of virtual walking on chronic pain, tone and spasticity in incomplete spinal cord injury. Disabil Rehabil. 2026;48(9):2814–2826. doi: 10.1080/09638288.2025.2583905 [DOI] [PubMed] [Google Scholar]

[cit0024] 24.Selvaraj JS, Prakash H, Augustine TA, Kumar SSK, Velkumar S, Amalraj JA. Immersive virtual reality for neuropathic pain management in spinal cord injury: a randomized controlled trial. J Spinal Cord Med. 2026;49(3):543–555. doi: 10.1080/10790268.2025.2514321 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0025] 25.Higgins J, Thomas J, Chandler J, et al. Cochrane Handbook for Systematic Reviews of Interventions Version 6.2; The Cochrane Collaboration; 2021. Available from: www.training.cochrane.org/handbook. [Google Scholar]

[cit0026] 26.Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med. 2002;21(11):1539–1558. doi: 10.1002/sim.1186 [DOI] [PubMed] [Google Scholar]

[cit0027] 27.Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315(7109):629–634. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0028] 28.Guyatt G, Oxman AD, Akl EA, et al. GRADE guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables. J Clin Epidemiol. 2011;64(4):383–394. doi: 10.1016/j.jclinepi.2010.04.026 [DOI] [PubMed] [Google Scholar]

[cit0029] 29.Soler MD, Kumru H, Pelayo R, et al. Effectiveness of transcranial direct current stimulation and visual illusion on neuropathic pain in spinal cord injury. Brain. 2010;133(9):2565–2577. doi: 10.1093/brain/awq184 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0030] 30.Özkul Ç, Kılınç M, Yıldırım SA, Topçuoğlu EY, Akyüz M. Effects of visual illusion and transcutaneous electrical nerve stimulation on neuropathic pain in patients with spinal cord injury: a randomised controlled cross-over trial. J Back Musculoskelet Rehabil. 2015;28(4):709–719. doi: 10.3233/bmr-140573 [DOI] [PubMed] [Google Scholar]

[cit0031] 31.Pozeg P, Palluel E, Ronchi R, et al. Virtual reality improves embodiment and neuropathic pain caused by spinal cord injury. Neurology. 2017;89(18):1894–1903. doi: 10.1212/WNL.0000000000004585 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0032] 32.Richardson EJ, McKinley EC, Rahman A, Klebine P, Redden DT, Richards JS. Effects of virtual walking on spinal cord injury-related neuropathic pain: a randomized, controlled trial. Rehabil Psychol. 2019;64(1):13–24. doi: 10.1037/rep0000246 [DOI] [PubMed] [Google Scholar]

[cit0033] 33.Austin PD, Siddall PJ. Virtual reality for the treatment of neuropathic pain in people with spinal cord injuries: a scoping review. J Spinal Cord Med. 2021;44(1):8–18. doi: 10.1080/10790268.2019.1575554 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0034] 34.Azurdia D, Acuña SM, Narasaki-Jara M, Furtado Jr O, Jung T. Effects of Virtual Reality-Based Aerobic Exercise on Perceptions of Pain and Fatigue in Individuals with Spinal Cord Injury. Games Health J. 2022;11(4):236–241. doi: 10.1089/g4h.2021.0151 [DOI] [PubMed] [Google Scholar]

[cit0035] 35.Tabacof L, Salazar SI, Breyman E, et al. Immersive virtual reality for chronic neuropathic pain after spinal cord injury: a pilot, randomized, controlled trial. Pain Rep. 2024;9(6):e1173. doi: 10.1097/pr9.0000000000001173 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0036] 36.Li A, Montaño Z, Chen VJ, Gold JI. Virtual reality and pain management: current trends and future directions. Pain Manag. 2011;1(2):147–157. doi: 10.2217/pmt.10.15 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0037] 37.Preti K, Pinero M, Weitzman M, Malhotra V. Virtual reality: treating pain and anxiety. Am Nurse J. 2024;19(5):42–46. doi: 10.51256/anj052442 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0038] 38.Wiederhold BK, Gao K, Sulea C, Wiederhold MD. Virtual reality as a distraction technique in chronic pain patients. Cyberpsychol Behav Soc Netw. 2014;17(6):346–352. doi: 10.1089/cyber.2014.0207 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0039] 39.Dai Y, Li J, Li Y, et al. A scoping review on the role of virtual walking intervention in enhancing wellness. NPJ Digit Med. 2025;8(1):217. doi: 10.1038/s41746-025-01609-y [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0040] 40.Vastano R, Costantini M, Widerstrom-Noga E. Maladaptive reorganization following SCI: the role of body representation and multisensory integration. Prog Neurobiol. 2022;208:102179. doi: 10.1016/j.pneurobio.2021.102179 [DOI] [PubMed] [Google Scholar]

[cit0041] 41.Diana L, Sritharan J, Vallesi V, Brunello N, Verma RK, Zito GA. Chronic spinal cord injury is associated with morphometric brain changes in functional networks beyond the sensorimotor system. Front Neurol. 2025;16:1672506. doi: 10.3389/fneur.2025.1672506 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0042] 42.Wankhede NL, Koppula S, Ballal S, et al. Virtual reality modulating dynamics of neuroplasticity: innovations in neuro-motor rehabilitation. Neuroscience. 2025;566:97–111. doi: 10.1016/j.neuroscience.2024.12.040 [DOI] [PubMed] [Google Scholar]

[cit0043] 43.Xie Q, Xiong X, Liu M, et al. Effects of multi-sensory virtual reality training on gait adaptability and somatomotor network remodeling in patients with stroke: a randomized controlled trial. J Neuroeng Rehabil. 2026;23(1). doi: 10.1186/s12984-026-02003-9 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0044] 44.Eaves DL, Riach M, Holmes PS, Wright DJ. Motor Imagery during Action Observation: a Brief Review of Evidence, Theory and Future Research Opportunities. Front Neurosci. 2016;10:514. doi: 10.3389/fnins.2016.00514 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0045] 45.Mustile M, Kourtis D, Edwards MG, Donaldson DI, Ietswaart M. Neural correlates of motor imagery and execution in real-world dynamic behavior: evidence for similarities and differences. Front Hum Neurosci. 2024;18:1412307. doi: 10.3389/fnhum.2024.1412307 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0046] 46.Eid M, Alkerdasy H, Hilmy A, Sameh A, Ashraf A. Effectiveness of virtual reality-based exercise rehabilitation for enhancing functional outcomes and quality of life in adult cancer survivors: a systematic review of randomized controlled trials. J Cancer Surviv. 2025;2025:1. doi: 10.1007/s11764-025-01923-0 [DOI] [PubMed] [Google Scholar]

[cit0047] 47.AlHossan AM, Jahhaf RH, Alharbi AS, et al. Digital and virtual reality-based rehabilitation versus conventional therapy for rotator cuff tears and post-repair recovery: a systematic review and meta-analysis. JSES Rev Rep Tech. 2026;6(1):100584. doi: 10.1016/j.xrrt.2025.09.003 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0048] 48.Moumeni IN. Intensity over duration in neurological rehabilitation: exploring evidence for optimised recovery paradigms. Front Neurol. 2025;16:1697186. doi: 10.3389/fneur.2025.1697186 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0049] 49.Adamovich SV, Fluet GG, Tunik E, Merians AS. Sensorimotor training in virtual reality: a review. NeuroRehabilitation. 2009;25(1):29–44. doi: 10.3233/nre-2009-0497 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0050] 50.Indovina P, Barone D, Gallo L, Chirico A, De Pietro G, Giordano A. Virtual Reality as a Distraction Intervention to Relieve Pain and Distress During Medical Procedures: a Comprehensive Literature Review. Clin J Pain. 2018;34(9):858–877. doi: 10.1097/ajp.0000000000000599 [DOI] [PubMed] [Google Scholar]

[cit0051] 51.Pretat T, Koller C, Hügle T. Virtual reality as a treatment for chronic musculoskeletal pain syndromes. Joint Bone Spine. 2025;92(1):105769. doi: 10.1016/j.jbspin.2024.105769 [DOI] [PubMed] [Google Scholar]

[cit0052] 52.Goudman L, Jansen J, Billot M, et al. Virtual Reality Applications in Chronic Pain Management: systematic Review and Meta-analysis. JMIR Serious Games. 2022;10(2):e34402. doi: 10.2196/34402 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0053] 53.Kouijzer M, Kip H, Bouman YHA, Kelders SM. Implementation of virtual reality in healthcare: a scoping review on the implementation process of virtual reality in various healthcare settings. Implement Sci Commun. 2023;4(1):67. doi: 10.1186/s43058-023-00442-2 [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Effects of Virtual Reality-Based Interventions on Neuropathic Pain in Individuals with Spinal Cord Injury: A Meta-Analysis of Randomized Controlled Trials

Chang Liu

Yamei Liu

Abstract

Background

Methods

Results

Conclusion

Introduction

Methods

Study Inclusion and Exclusion Criteria

Population (P)

Intervention (I)

Comparator (C)

Outcomes (O)

Study Design (S)

Database Search

Risk of Bias Evaluation

Data Extraction

Statistical Analysis

Certainty of Evidence

Results

Literature Search

Figure 1.

Characteristics of the Included Studies

Table 1.

Risk-of-Bias Assessment

Table 2.

Meta-Analysis Results

Figure 2.

Figure 3.

Figure 4.

Figure 5.

Table 3.

Publication Bias

Figure 6.

Certainty of Evidence

Table 4.

Discussion

Conclusions

Funding Statement

Data Sharing Statement

Author Contributions

Disclosure

References

Associated Data

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases