1 Introduction
Bilateral vestibulopathy (BVP), commonly termed bilateral vestibular hypofunction (BVH) in the rehabilitation literature, is a chronic vestibular syndrome characterized by bilateral reduction or loss of vestibular sensory input and vestibulo-ocular reflex (VOR) function. Other terms used in earlier literature include bilateral vestibular failure, deficiency, areflexia, and loss. In this guideline, the term BVH is used throughout. The cardinal manifestations are postural and gait instability, typically exacerbated in darkness, on uneven surfaces, or during head motion, together with head or body movement induced visual blurring or oscillopsia
[1,
2]. Symptoms are generally minimal while sitting or lying still. Some patients also experience impaired spatial orientation, navigation, or spatial memory
[1].
The population prevalence of BVH remains uncertain. An analysis of the 2008 US National Health Interview Survey estimated the prevalence of severe-to-profound symptomatic BVH at approximately 28 per 100,000 adults
[3]; however, because case identification was based primarily on self-reported symptoms without objective vestibular testing, the true prevalence may be underestimated. The etiologic spectrum is broad and includes ototoxic exposure, bilateral Ménière disease (MD), infectious and immune-mediated disorders, genetic conditions, neoplasms, iatrogenic injury, and neurodegenerative disease, although a substantial proportion of cases remain idiopathic
[4,
5]. Considerable heterogeneity exists in etiology, onset, disease course, residual vestibular function, symptom phenotype, comorbidity, and functional impairment. BVH can profoundly compromise gaze stability, postural control, mobility, driving, employment, and social participation, while increasing the risks of falls, injury, healthcare utilization, and productivity loss.
Vestibular rehabilitation therapy (VRT) is the cornerstone of non-pharmacological management for BVH-related symptoms, impairments, and activity limitations. Its effects are mediated through vestibular adaptation, substitution, and habituation, together with visual and somatosensory reweighting, gaze-stabilization training, balance and gait retraining, and task-specific functional practice. Existing clinical practice guidelines, systematic reviews, and clinical studies support the effectiveness of VRT in improving gaze stability, postural control, gait, functional mobility, and health-related quality of life
[6,
7]. Nevertheless, the BVH-specific evidence base remains limited by small sample sizes and substantial heterogeneity in diagnostic criteria, residual vestibular function, intervention content, treatment dose, and outcome measures. The optimal training dose, predictors of response, long-term effectiveness, and role of emerging technologies therefore remain incompletely defined.
Although expert consensus statements and technical recommendations on vestibular rehabilitation have been published in China, there is currently no evidence-based clinical practice guideline specifically addressing the rehabilitation of adults with BVH. Clinical practice remains inconsistent with respect to etiologic evaluation, selection of rehabilitation outcomes, timing of treatment initiation, exercise content and dose, progression criteria, treatment response, and long-term follow-up. To address these gaps, the editorial office of the Chinese Journal of Otorhinolaryngology Head and Neck Surgery convened a multidisciplinary guideline development group. Relevant studies were systematically identified through searches of major international and Chinese databases from inception to the prespecified cutoff date. Evidence was appraised and recommendations were graded according to the 2009 Oxford Centre for Evidence-Based Medicine framework, with additional consideration of the balance of benefits and harms, patient values and preferences, resource implications, acceptability, and feasibility. The guideline development process followed World Health Organization guideline-development principles. This guideline provides recommendations across the full continuum of care, including diagnosis and etiologic evaluation, rehabilitation assessment, design and delivery of VRT, criteria for treatment modification, suspension, or discontinuation, and the use of adjunctive and emerging interventions. It is intended to support evidence-informed, standardized, and clinically applicable decision-making by otolaryngologists, neurologists, rehabilitation physicians, general practitioners, physical and rehabilitation therapists, audiologists, nurses, and other professionals involved in the care of adults with BVH.
Terminology: The Bárány Society diagnostic consensus and the International Classification of Vestibular Disorders (ICVD) use BVP for the diagnostic entity, whereas rehabilitation literature commonly uses BVH for the overlapping population with bilaterally reduced vestibular function. In this guideline, BVP is used when referring specifically to the Bárány Society diagnostic entity, BVH is retained when describing rehabilitation evidence, and BVH/BVP denotes the target population shared by both literatures.
2 Methods
This guideline was developed in accordance with established methodological standards for clinical practice guidelines and the definition proposed by the US Institute of Medicine
[8]. The development process was guided by the 2014 World Health Organization (WHO) Handbook for Guideline Development, second edition
[9], and other relevant methodological standards. The Appraisal of Guidelines for Research and Evaluation II (AGREE II) instrument was used to enhance the methodological rigour and transparency of guideline development
[10], and the final guideline was reported in accordance with the reporting items for practice guidelines in healthcare (RIGHT) statement
[11]. A prespecified guideline protocol was developed before the evidence review and formulation of recommendations. The overall methodological framework is shown in Fig. 1.
2.1 Guideline initiation and methodological support
In July 2025, this guideline was initiated by the Eye & ENT Hospital of Fudan University. The Center for Evidence-Based Medicine, Fudan University, provided methodological guidance and technical support throughout the development process.
2.2 Guideline registration
The guideline was registered with the International Practice Guideline Registry and Transparency Platform, registration number: PREPARE-2025CN1236. The guideline protocol and registration information are provided in Suppl. Appendix 1.
2.3 Objectives and scope of the guideline
This guideline provides evidence-based recommendations for the diagnosis and phenotype-directed etiologic evaluation, rehabilitation assessment, design and delivery of VRT, monitoring of treatment response, criteria for treatment modification, temporary suspension or discontinuation, and the use of selected adjunctive or emerging interventions in adults with BVH/BVP.
The specific objectives are to facilitate central compensation and optimize functional recovery; reduce oscillopsia, motion-induced visual blurring, dizziness or unsteadiness, postural instability, and gait dysfunction; improve mobility, activities of daily living, social participation, and health-related quality of life; reduce the risks of falls, injury, physical deconditioning, and persistent functional disability; and promote the appropriate and efficient use of rehabilitation resources.
The guideline is intended for healthcare professionals involved in vestibular assessment and rehabilitation, including otolaryngologists, neurologists, rehabilitation physicians, general practitioners, physical and rehabilitation therapists, audiologists, nurses, and other relevant professionals. Diagnostic and etiologic recommendations apply to adults aged 18 years or older with suspected or confirmed BVH/BVP; rehabilitation recommendations apply primarily to adults with objectively confirmed peripheral bilateral vestibular hypofunction. The guideline does not address pediatric BVH/BVP, isolated unilateral vestibular hypofunction, or acute central vestibular syndromes. In patients with mixed peripheral and central disorders, including cerebellar ataxia, neuropathy, and vestibular areflexia syndrome (CANVAS) or other neurodegenerative disease, the recommendations should be adapted within multidisciplinary, disease-specific care.
The guideline is applicable across tertiary and secondary hospitals, specialist rehabilitation hospitals, primary-care institutions, and community-based rehabilitation settings with the expertise, facilities, and safety provisions required to provide vestibular assessment and rehabilitation.
2.4 Guideline development groups
A multidisciplinary guideline-development structure was established in July 2025, comprising a Guideline Steering Committee, a Guideline Development Group, a Guideline Secretariat, and an independent External Review Panel. The Steering Committee was co-chaired by a clinical expert and a guideline methodologist and was responsible for oversight of the guideline scope, methods, conflict-of-interest management, consensus procedures, and final approval.
The Guideline Development Group included experts in otorhinolaryngology–head and neck surgery, neurotology and vestibular medicine, neurology, rehabilitation medicine, vestibular rehabilitation, clinical epidemiology, biostatistics, evidence-based medicine, and medical editing, with representation from six geographic regions of China. The group prioritized the clinical questions and outcomes, reviewed and interpreted the evidence, formulated recommendations, and approved the final guideline.
The Guideline Secretariat coordinated protocol development, evidence searches and synthesis, preparation of evidence summaries, assessment of patient values and preferences, documentation of the development process, and manuscript drafting. The External Review Panel independently assessed the relevance, clarity, feasibility, applicability, and evidentiary basis of the draft recommendations. The full membership, affiliations, roles, and conflicts of interest of all contributors are provided in Suppl. Appendix 2.
2.5 Declaration and management of conflicts of interest
All members of the Guideline Steering Committee, Guideline Development Group, and Guideline Secretariat completed written declarations of financial and non-financial conflicts of interest before participating in guideline development and updated their declarations when necessary. The Steering Committee reviewed all disclosures and determined appropriate management measures according to the nature and relevance of each conflict.
Depending on the extent of the conflict, affected members could be restricted from participating in evidence appraisal, recommendation deliberation, consensus voting, or other relevant activities. Members without relevant conflicts were permitted to participate fully. All declarations and the corresponding management decisions are presented in a Suppl. Appendix 1 to the guideline.
2.6 Identification and prioritization of clinical questions
Potential clinical questions were generated from existing guidelines and reviews, a structured survey of 33 front-line clinicians, rehabilitation professionals, and researchers, and issues identified during preliminary stakeholder consultation. An initial list of 16 questions was rated on a 1–7 importance scale. After the first Delphi round, the 10 highest-priority questions were retained; a second Delphi round and panel discussion consolidated these into five principal clinical questions covering diagnosis and etiologic evaluation, rehabilitation assessment, VRT formulation and delivery, modification or discontinuation of treatment, and adjunctive or emerging interventions.
The final questions were structured using the population, intervention, comparator, and outcomes (PICO) framework—population, intervention or index approach, comparator when applicable, and outcomes—or an appropriate adaptation for diagnostic and etiologic questions. Candidate outcomes were prioritized according to clinical importance and relevance to patients, including oscillopsia and functional gaze stability, postural control, gait and mobility, falls, activity and participation, quality of life, treatment burden, and adverse events. Detailed information on the generation, prioritization, and consolidation of clinical questions and the Delphi process is provided in Suppl. Appendix 3.
Patient perspectives and preferences were explored through informal discussions during routine clinical encounters and stakeholder consultation. These discussions focused on symptom burden, acceptable exercise burden, preferred supervision and delivery modes, priority outcomes, safety concerns, and access to rehabilitation. No structured patient questionnaire was administered, and no patient-level data were systematically collected or analyzed for research purposes. A summary of patient and stakeholder perspectives considered during guideline development is provided in Suppl. Appendix 4.
2.7 Evidence identification, appraisal, and synthesis
2.7.1 Search strategy and study selection
Separate search strategies were developed for each of the five clinical questions. PubMed, MEDLINE, Embase, Web of Science, the Cochrane Library, SinoMed, CNKI, and Wanfang Data were searched from inception to December 31, 2025. Guideline repositories, professional-society websites, and reference lists of eligible reports were also screened. The search was updated before final recommendation approval in May 2026. Full strategies, search interfaces, limits, and update details are provided in Suppl. Appendix 5.
Eligible evidence included clinical practice guidelines, consensus statements, systematic reviews, randomized and non-randomized intervention studies, diagnostic-accuracy studies, cohort and case-control studies, cross-sectional studies, and clinically informative case series. Studies were required to address adults with suspected or confirmed BVH/BVP or a directly relevant vestibular-rehabilitation population. Indirect evidence from mixed or unilateral vestibular populations was retained only when BVH/BVP-specific evidence was unavailable and applicability was explicitly considered. Pediatric studies, animal studies, isolated central vestibular disorders, and reports without clinically relevant outcomes were excluded.
The searches identified 773 records. After removal of 317 duplicates, 456 records underwent title and abstract screening; 219 were excluded. Full texts of 237 reports were assessed, 202 were excluded with reasons, and 35 reports were included in the evidence synthesis. Two reviewers independently screened records and full texts; disagreements were resolved by discussion or adjudication by a third reviewer. The study selection process is presented in Suppl. Appendix 6.
2.7.2 Data extraction and evidence synthesis
Data were extracted using standardized forms and independently verified by a second reviewer. Extracted information included study design, population, diagnostic criteria, interventions or index tests, comparators, outcomes, follow-up, adverse events, and implementation-relevant factors. Because of substantial clinical and methodological heterogeneity, evidence was generally synthesized narratively. Quantitative pooling was undertaken only when populations, interventions, and outcomes were sufficiently comparable. Evidence tables are provided in Suppl. Appendix 7.
2.7.3 Methodological quality and risk-of-bias assessment
Systematic reviews and meta-analyses were assessed using A Measurement Tool to Assess Systematic Reviews 2 (AMSTAR 2)
[12] without calculation of an overall numerical score. Randomized trials were evaluated using the Revised Cochrane Risk-of-Bias Tool for Randomized Trials (RoB 2)
[13]; non-randomized intervention studies were assessed using the Risk of Bias in Non-randomized Studies of Interventions (ROBINS-I) when applicable; cohort and case-control studies were assessed using the Newcastle–Ottawa Scale
[14]; diagnostic-accuracy studies were assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2); and existing clinical practice guidelines were appraised using AGREE II. Two reviewers independently completed each assessment, with disagreements resolved by consensus or third-reviewer adjudication. Appraisal results informed interpretation of the evidence and are reported in Suppl. Appendix 8.
2.7.4 Evidence levels and recommendation strength
Levels of evidence were assigned according to the 2009 Oxford Centre for Evidence-Based Medicine (OCEBM) framework. Recommendation strength was classified as strong, moderate, or weak using a modified framework developed by the Guideline Development Group. Recommendation strength was determined independently of evidence level by considering methodological quality, consistency and directness, the balance of benefits and harms, patient values and preferences, resource requirements, equity, acceptability, feasibility, and applicability. Accordingly, low-level evidence could support a strong recommendation when the expected benefit or safety imperative was substantial and the burden or risk of following the recommendation was minimal. The operational definitions are presented in Table 1.
2.8 Formulation of recommendations and consensus development
For each clinical question, the Guideline Secretariat prepared structured evidence summaries and recommendation-development worksheets. Draft recommendations were formulated through a modified Delphi process and subsequently deliberated at two hybrid consensus meetings held in December 2025 and May 2026.
During the second Delphi round, panel members rated each draft recommendation using a three-category agreement scale (agree, modify, or disagree). Consensus was considered achieved when ≥ 80% of participants selected either “agree” or “modify”. Voting was anonymous; abstentions, recusals, and the number of eligible voters were documented. Recommendations that did not initially reach consensus were revised after panel discussion and subjected to a further vote. Unresolved disagreement and its rationale were recorded. The final recommendations are summarized in Table 2. Recommendation-development worksheets are provided in Suppl. Appendix 9. The final recommendations addressed four major domains: assessment and diagnosis, individualized VRT, emerging technologies, and rehabilitation outcome assessment. The overall clinical pathway integrating diagnosis, rehabilitation intervention, emerging technologies, and outcome assessment is summarized in Fig. 2.
2.9 External review, quality assurance, and approval
The complete draft was reviewed independently by the External Review Panel, which included clinical specialists, rehabilitation professionals, and guideline methodologists who had not participated in recommendation voting. Reviewers assessed clinical relevance, evidence linkage, clarity, feasibility, applicability, and safety. All comments and the corresponding responses or revisions were documented. The revised manuscript underwent a final RIGHT reporting check and AGREE II-based methodological quality review before approval by the Guideline Steering Committee.
2.10 Guideline updating
The Guideline Development Group plans to review this guideline no later than 5 years after publication. An earlier update will be initiated if major new evidence emerges, relevant diagnostic criteria change, important safety concerns are identified, or new interventions are introduced that may materially alter current recommendations.
The update will include renewed systematic literature searches, reassessment of the methodological quality and level of the evidence, reconsideration of existing recommendations, and formulation of new recommendations where appropriate, using methods consistent with those applied in the original guideline.
2.11 Dissemination and implementation
Following publication, the guideline will be disseminated through peer-reviewed journals, presentations and dedicated sessions at relevant scientific meetings, regional and multidisciplinary educational workshops, and professional digital and social-media platforms.
Implementation will be supported by the development of concise clinical algorithms, assessment checklists, exercise-prescription and progression tools, patient-education materials, and quality indicators suitable for different levels of care. Where feasible, pilot implementation, professional surveys, and clinical audits will be undertaken to identify barriers to uptake, evaluate adherence to key recommendations, and inform subsequent implementation strategies and guideline updates.
2.12 Funding and editorial independence
The guideline was supported by the “Common Disease Prevention and Control Research” Key Project of the National Key R&D Program of China (Nos. 2023YFC2508000 and 2024YFC2511100); The Win-Win Program, Eye & ENT Hospital of Fudan University. The funding body did not participate in question selection, evidence identification or appraisal, recommendation deliberation or voting, external review, manuscript approval, or the decision to publish. The Guideline Steering Committee retained final authority over all methodological and editorial decisions.
2.13 Availability of supporting materials
The protocol, detailed search strategies, Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow information, evidence tables, methodological quality and risk-of-bias assessments, patient values-and-preferences report, recommendation-development worksheets, voting results, external-review responses, and conflict-of-interest declarations are available through the PREPARE registration record and Suppl. Appendices 1–9.
3 Clinical Questions and Recommendations
3.1 How should BVH/BVP be diagnosed, and how should its etiology be evaluated?
Recommendation 1: We recommend diagnosing BVH/BVP in accordance with the Bárány Society criteria. The diagnosis should integrate a compatible chronic vestibular syndrome, objective evidence of bilaterally reduced or absent angular VOR function, and exclusion of alternative diagnoses (Level of evidence: 5; Recommendation strength: Strong).
Evidence summary
This recommendation was informed by one international expert consensus statement, one narrative review, two observational studies, and one longitudinal follow-up study. The principal evidence was derived from the Bárány Society consensus criteria and was therefore assigned Level 5 evidence. Observational evidence, graded as Levels 2b–3b, provided additional support for integrating the clinical phenotype with objective vestibular testing.
Rationale
The clinical presentation of BVH is heterogeneous. Some patients present predominantly with chronic imbalance, oscillopsia, or movement-induced visual blurring, whereas classic vestibular symptoms may be absent or underrecognized. Consequently, BVH has historically been underdiagnosed, and diagnostic criteria have varied substantially across clinical practice and research. The adoption of a standardized diagnostic framework is therefore essential to improve diagnostic consistency, reproducibility, and comparability across studies.
In 2017, the Bárány Society published international diagnostic criteria for bilateral vestibulopathy (BVP), which remain the most widely accepted framework for clinical practice and research
[1]. According to these criteria, definite BVP requires all of the following:
A. A chronic vestibular syndrome characterized by: (1) unsteadiness when walking or standing, together with at least one of the following: (2) movement-induced blurred vision or oscillopsia during walking or rapid head or body motion; (3) and/or worsening of unsteadiness in darkness and/or on uneven ground;
B. No symptoms while sitting or lying under static conditions;
C. Bilaterally reduced or absent horizontal angular vestibulo-ocular reflex (aVOR) function, documented by at least one of the following: (1) a horizontal aVOR gain of < 0.6 in both ears on the video head impulse test (vHIT); (2) a reduced caloric response, defined as a sum of bithermal maximum peak slow-phase velocities of < 6°/s on each side; (3) and/or a horizontal aVOR gain of < 0.1 during sinusoidal rotational-chair testing at 0.1 Hz, together with a phase lead of > 68° and/or a time constant of < 5 seconds;
D. The findings are not better explained by another disease.
The Bárány Society also proposed criteria for probable BVP, which require the same characteristic chronic vestibular syndrome and absence of symptoms under static conditions, together with bilaterally pathological bedside head impulse tests and exclusion of a more appropriate alternative diagnosis.
In a large cohort study, Lucieer et al.
[15] reported that patients meeting the Bárány Society criteria showed a consistent pattern of impaired postural stability, dynamic visual acuity, and gait function, supporting the clinical validity of the criteria for identifying patients with a typical BVP phenotype.
With the increasing availability of vHIT, its role in the diagnostic work-up of BVH has become increasingly prominent. vHIT assesses high-frequency semicircular-canal function and is rapid, well tolerated, and readily applicable in clinical settings. However, some patients have limited or atypical oscillopsia because of reduced habitual activity, behavioral adaptation, or compensatory strategies. In such patients, objective testing with vHIT, caloric stimulation, or rotational-chair testing may still demonstrate substantial bilateral vestibular loss.
Van Stiphout et al.
[5] emphasized that BVP should be regarded as a heterogeneous chronic syndrome with multiple potential etiologies and that diagnosis should integrate clinical manifestations, vestibular laboratory findings, and etiologic clues rather than rely on any single test. The authors further recommended evaluation for neurodegenerative conditions, including CANVAS, in patients with sensory neuropathy, cerebellar signs, or progressive gait impairment.
Van de Berg et al.
[2] reported that the combined use of vHIT, caloric testing, and rotational-chair testing improves the detection of bilateral vestibular impairment and reduces the likelihood of false-negative results associated with reliance on a single modality. Because these tests assess vestibular function across different stimulus frequencies, discordant findings may occur, particularly in patients with partial or frequency-specific residual vestibular function.
A 2024 retrospective study comparing the clinical utility of caloric testing, vHIT, and rotational-chair testing similarly found that no single vestibular test was sufficient in all patients
[16]. vHIT showed relatively high specificity but lower sensitivity, whereas combining vHIT with caloric or rotational-chair testing improved diagnostic yield.
Taken together, the available evidence supports a standardized diagnostic approach based on the Bárány Society criteria and integrating the clinical phenotype, objective assessment of bilateral aVOR function, and exclusion of alternative peripheral, central, visual, somatosensory, or musculoskeletal causes.
Recommendation 2: We recommend a structured, phenotype-directed etiologic and differential diagnostic evaluation in patients with confirmed or suspected BVH/BVP. The assessment should specifically consider ototoxic injury, bilateral Ménière disease, hereditary and neurodegenerative disorders, infectious and immune-mediated diseases, trauma, and postoperative vestibular injury. Patients in whom no cause is identified after an appropriate evaluation may be classified as having idiopathic BVH/BVP (Level of evidence: 2b; Recommendation strength: Moderate).
Evidence summary
This recommendation was informed by one expert consensus statement, one single-center retrospective case series, one retrospective cohort study, two narrative reviews, one longitudinal follow-up study, and one multicenter observational study. The evidence base consisted predominantly of consensus evidence, narrative reviews, retrospective cohorts, and multicenter observational data; no randomized controlled trials (RCTs) were identified.
The overall evidence was assigned Level 2b. Given the observational nature of the evidence and the absence of direct comparative studies evaluating alternative etiologic work-up strategies, a moderate recommendation is appropriate. A strong recommendation may be considered if the recommendation is explicitly restricted to a structured, clinically directed evaluation rather than indiscriminate laboratory or imaging screening.
Rationale
BVH/BVP represents a heterogeneous group of vestibular syndromes arising from diverse pathological processes. Etiology may influence the pattern and progression of vestibular loss, associated neurological or auditory manifestations, rehabilitation response, prognosis, and eligibility for disease-specific or emerging interventions. A structured etiologic assessment is therefore important for identifying potentially modifiable causes and guiding subsequent management.
Recognized causes include ototoxic drug exposure, bilateral Ménière disease, sequelae of vestibular neuritis or labyrinthitis, hereditary disorders, immune-mediated inner-ear disease, infection, neurodegenerative disease, trauma, and vestibular injury following otologic or neurosurgical procedures. Aminoglycoside-associated ototoxicity remains one of the most consistently identified causes. Nevertheless, despite extensive evaluation, a substantial proportion of patients remain without an established etiology.
The 2017 Bárány Society consensus emphasized that BVP is a clinical syndrome rather than a single disease entity and may result from a wide range of peripheral vestibular disorders
[1]. Once the diagnosis has been established, further investigation should therefore be directed toward identifying the underlying cause and clinically relevant comorbidities.
In a 2022 multicenter observational study of 173 patients fulfilling the Bárány Society diagnostic criteria, Van Stiphout et al.
[17] systematically examined patterns of vestibular impairment across etiologic subgroups. Ototoxic injury, hereditary disease, particularly DFNA9 (deafness, autosomal dominant 9)-associated disease, bilateral Ménière disease, infectious disorders, and neurodegenerative conditions were among the most frequently identified causes. Distinct etiologies were associated with different patterns of semicircular-canal involvement and residual vestibular function. Ototoxic injury was typically associated with severe and extensive bilateral canal loss, whereas hereditary disease more often followed a progressive course.
In a 2023 review, Van Stiphout et al.
[5] proposed a structured diagnostic algorithm for etiologic evaluation. The recommended assessment included a detailed history of ototoxic exposure; evaluation for hereditary syndromes, neurodegenerative disease, cerebellar involvement, bilateral Ménière disease, immune-mediated disease, and CANVAS; and targeted audiological, neurological, imaging, laboratory, or genetic investigations according to the clinical phenotype. The review emphasized that some etiologies have characteristic patterns of progression and specific implications for management. For example, CANVAS is typically associated with sensory neuronopathy and cerebellar ataxia, whereas hereditary BVH may warrant family screening and genetic counselling.
Moyaert et al.
[18] retrospectively analyzed the etiology and hearing status of 315 patients who fulfilled the Bárány Society criteria. A definite etiology was identified in only 37% of patients, and hereditary disorders accounted for 31% of non-idiopathic cases. Approximately three-quarters of the cohort had some degree of hearing loss, underscoring both the etiologic heterogeneity of BVH and the importance of formal audiological assessment.
A review by Hain et al.
[84] identified aminoglycoside ototoxicity as a leading recognized cause of BVH, followed by immune-mediated inner-ear disease, neurological disorders, and congenital abnormalities. However, approximately half of the reported cases remained unexplained.
Similarly, Strupp et al.
[85] reported substantial variation in etiologic distributions across studies. Approximately 51% of cases were classified as idiopathic. Among patients with an identifiable cause, toxic or metabolic factors accounted for approximately 13%–21%, with aminoglycoside-associated ototoxicity representing a major contributor. Other reported causes included infection, particularly meningitis, neurodegenerative disease, and systemic disorders.
In a retrospective observational cohort of 154 patients, Lucieer et al.
[15] further demonstrated the etiologic complexity of BVH. Ototoxic injury, particularly aminoglycoside toxicity, was among the most frequently identified causes; other causes included bilateral Ménière disease, hereditary and neurodegenerative disorders, inflammatory or immune-mediated disease, trauma, and postoperative vestibular injury. Across published cohorts, approximately 20%–51% of cases were classified as idiopathic.
Overall, a structured, phenotype-directed etiologic evaluation may identify potentially modifiable causes, reveal associated auditory or neurological disease, inform individualized rehabilitation and prognosis, and support appropriate referral for genetic counselling or specialist management. However, current evidence does not support indiscriminate use of extensive laboratory, imaging, or genetic testing in all patients. Investigations should instead be selected according to the clinical history, pattern of vestibular involvement, hearing status, neurological findings, disease course, family history, and relevant exposure history.
3.2 How should rehabilitation assessment be conducted in adults with BVH/BVP?
Recommendation 3: Adults with BVH/BVP should undergo a structured, multidomain assessment before rehabilitation. A consistent set of clinically relevant measures should be repeated during treatment, at discharge, and, when indicated, during follow-up to establish goals, guide treatment progression, and evaluate response (Level of evidence: 2b; Recommendation strength: Moderate).
Evidence summary
The supporting evidence included a clinical practice guideline, an expert consensus statement, systematic reviews, prospective cohort studies, and intervention studies. Collectively, these sources support assessment across the domains of body function, activity, participation, and relevant contextual factors. However, no validated core outcome set has yet been established specifically for BVH/BVP.
Rationale
BVH/BVP affects substantially more than peripheral vestibular function. Patients may have impaired gaze stability, postural control, gait, sensory integration, mobility, cognition, emotional well-being, and social participation. A rehabilitation assessment based solely on vestibular laboratory findings therefore fails to capture the full burden of disease or the outcomes most relevant to patients.
Consistent with the International Classification of Functioning, Disability and Health (ICF) framework, assessment should encompass body functions and structures, activity limitations, participation restrictions, and relevant personal and environmental factors
[5]. The 2022 APTA clinical practice guideline likewise supports the use of measures addressing gaze stability, balance, gait, fall risk, symptoms, and functional limitations in patients with peripheral vestibular hypofunction
[7].
Patient-reported and performance-based measures provide complementary information. Guinand et al.
[19] documented substantial limitations in mobility, daily activities, and quality of life among patients with BVP. The Bilateral Vestibulopathy Questionnaire was subsequently developed and validated to capture the broader spectrum of physical, cognitive, emotional, and participation-related consequences of BVP
[20,
21]. A systematic review further demonstrated marked heterogeneity in the outcome measures used in BVP research and identified six broad domains: vestibular function, motor function, quality of life, cognition, hearing, and mental well-being
[22]. The review did not establish a definitive core outcome set, underscoring the need to select a pragmatic and consistent battery tailored to the clinical setting.
Falls are common and cannot be predicted reliably from residual vestibular function alone. In a prospective cohort of 119 patients, 39% reported falls; Dizziness Handicap Inventory (DHI) scores greater than 47 and Oscillopsia Severity Questionnaire scores greater than 27.5 showed only moderate discrimination and should not be used in isolation
[23]. Assessment should therefore integrate fall history, functional gait, sensory and neurological comorbidities, confidence, and environmental risk.
A minimum assessment battery should address symptoms and patient-reported impact, gaze stability, postural control and sensory integration, gait and mobility, fall risk, activity and participation, and clinically relevant visual, auditory, neurological, cognitive, and psychological modifiers. The same measures should be used at serial assessments whenever feasible.
Recommendation 4: Functional gaze stability should be assessed in adults with BVH/BVP. Where feasible, standardized dynamic visual acuity testing should be used to quantify the effect of head movement on visual performance and to monitor response to rehabilitation (Level of evidence: 2b; Recommendation strength: Moderate).
Evidence summary
This recommendation was informed by an international consensus statement, RCT, and observational studies. Although randomized evidence supports the responsiveness of dynamic visual acuity (DVA) to gaze-stabilization training, evidence for its use as a routine outcome measure remains limited by small samples and variation in testing protocols.
Rationale
Bilateral impairment of the angular VOR commonly causes movement-induced visual blurring and oscillopsia. DVA is not required for the diagnosis of BVP, but it provides a functional measure of the ability to maintain visual clarity during head movement
[1].
In a matched cross-sectional study, Zhu et al.
[24] compared treadmill-based DVA in 41 adults with BVP and 41 age- and sex-matched healthy participants. Participants with BVP demonstrated significantly greater DVA loss at all walking speeds, and DVA loss increased progressively with walking speed in the BVP group. These findings support treadmill-based DVA as a functional measure of movement-related gaze instability in BVP. However, because the study did not assess patients before and after rehabilitation, it does not establish the responsiveness of DVA to treatment. The study also found that increasing age was associated with a greater likelihood of being unable to complete testing at higher walking speeds, and that handrail use affected DVA performance. Testing speed, use of external support, viewing conditions, and safety procedures should therefore be standardized and documented when DVA is used for serial assessment.
In a prospective randomized double-blind trial of 13 patients with BVH, Herdman et al.
[26] found that vestibular exercises significantly improved DVA compared with placebo exercises. Improvement in DVA generally occurred without a corresponding increase in measured VOR gain, suggesting that compensatory saccades and other centrally programmed strategies contributed to recovery. Changes in DVA also did not correlate closely with subjective oscillopsia.
DVA should therefore be interpreted as a measure of functional gaze stability rather than a direct surrogate for restoration of peripheral vestibular function. Testing should use standardized conditions, including consistent optotype presentation, head-movement direction and velocity, viewing distance, and comparison with static visual acuity. Subjective oscillopsia should be assessed separately because objective visual performance and symptom severity may not change in parallel.
Recommendation 5: Rehabilitation assessment should include postural control, sensory dependence, dynamic gait, functional mobility, and fall risk. Assessment tools should be selected according to functional capacity, presenting limitations, rehabilitation goals, and the resources and safety provisions of the clinical setting (Level of evidence: 2b; Recommendation strength: Moderate).
Evidence summary
The evidence included an expert consensus statement, systematic reviews, prospective and retrospective cohort studies, cross-sectional studies, and studies of measurement properties. Direct BVP-specific evidence was available for balance impairment and falls, whereas evidence supporting individual clinical instruments was frequently derived from broader vestibular or balance-disorder populations.
Rationale
Patients with BVH/BVP commonly experience instability during standing and walking, particularly in darkness, on uneven surfaces, during head movement, and under dual-task or visually complex conditions. A classic case series by Bronstein et al.
[27] showed that full-field visual motion could provoke excessive postural sway in patients with visually induced vertigo. Although the study was not specific to BVH/BVP, its findings support visual dependence and impaired sensory-conflict resolution as clinically relevant contributors to postural instability and justify assessment under visually complex or conflicting sensory conditions. These limitations may not be apparent during simple static testing. Assessment should therefore include tasks that challenge dynamic balance, sensory integration, gait adaptation, and functional mobility.
A systematic review by Herssens et al.
[22] demonstrated that patients with BVP have impairments across multiple balance domains, particularly under conditions of sensory conflict. Prospective data further indicate that falls are frequent. Wuehr et al.
[28] reported falls in 43% of patients over follow-up, compared with 13% of healthy controls; 70% of affected patients reported recurrent falls. In contrast, Schniepp et al.
[25] found that the severity of peripheral vestibular loss and disease duration did not distinguish fallers from non-fallers. Increased temporal gait variability, particularly at slow walking speeds, and coexisting peripheral neuropathy were more informative.
Clinical testing may include the Functional Gait Assessment (FGA), Dynamic Gait Index (DGI), Timed Up and Go test (TUG), gait speed, tandem walking, turning, walking with head movements, dual-task walking, Romberg and foam-Romberg conditions, the modified Clinical Test of Sensory Interaction on Balance, tandem stance, and single-leg stance. No single test adequately characterizes all relevant deficits.
The FGA and DGI are commonly used to assess adaptive gait. Whitney et al.
[29] found that a DGI score of 19 or lower was associated with a 2.58-fold greater likelihood of reporting a fall in a mixed vestibular-disorder population. Marchetti et al.
[30] reported greater responsiveness for the FGA than for the DGI, with minimal detectable changes of approximately 6 and 4 points, respectively. These values were not derived specifically from BVP populations and should therefore be interpreted cautiously.
Recent BVP-specific evidence indicates that more demanding gait tasks, including tandem walking and gait under complex conditions, may be more sensitive than simple mobility measures for detecting functional impairment
[31]. Tool selection should balance clinical relevance, measurement properties, patient safety, feasibility, and the risk of floor or ceiling effects. Fall assessment should additionally include falls and near-falls over the preceding 6–12 months, fear of falling, use of mobility aids, environmental hazards, visual impairment, peripheral neuropathy, and other neurological or musculoskeletal comorbidities. Recommended domains and candidate measures for multidomain rehabilitation assessment are summarized in Table 3.
3.3 How should VRT be formulated and delivered in adults with BVH/BVP?
Recommendation 6: Following objective confirmation of peripheral bilateral vestibular hypofunction and screening for contraindications and safety risks, adults with BVH/BVP who have vestibular-related impairments, activity limitations, or participation restrictions should receive individualized VRT targeted to principal symptoms, functional impairments, comorbidities, and patient-centered goals (Level of evidence: 1b; Recommendation strength: Strong).
Evidence summary
The evidence base included a high-quality clinical practice guideline, a systematic review, randomized and controlled trials, and observational studies. The evidence consistently supports VRT for improving gaze stability, postural control, gait, and functional mobility, although improvements in activity and participation and the magnitude of benefit vary among patients.
Rationale
The principal goals of VRT in BVH/BVP are to improve visual stability during head movement, postural and gait safety, mobility, participation, and health-related quality of life. In patients with severe or longstanding bilateral vestibular loss, complete restoration of peripheral vestibular function is generally not a realistic treatment target. Rehabilitation should instead emphasize functional improvement, development of effective compensatory strategies, and sustainable self-management.
The 2022 American Physical Therapy Association (APTA) clinical practice guideline strongly recommends vestibular physical therapy for patients with unilateral or bilateral peripheral vestibular hypofunction who have associated symptoms or functional limitations
[7]. A systematic review by Porciuncula et al.
[32] found moderate evidence that exercise-based vestibular rehabilitation improves gaze and postural stability in adults with BVH, although evidence for activity and participation outcomes was limited and the included studies were heterogeneous.
Small controlled studies provide direct BVH-specific evidence. Krebs et al.
[33] reported improvements in walking speed and dynamic stability after vestibular rehabilitation in a double-blind placebo-controlled study of eight patients. Herdman et al.
[26] demonstrated greater recovery of dynamic visual acuity with vestibular exercises than with placebo exercises in a randomized double-blind trial. Subsequent work by Krebs et al.
[34] indicated that vestibular rehabilitation is beneficial but not uniformly effective, supporting the need to tailor treatment and progression to individual response.
The heterogeneity of BVH/BVP, including differences in etiology, residual vestibular function, visual and somatosensory impairment, neurological comorbidity, and functional goals further supports an individualized approach
[5]. Observational studies have reported improvements in disability, mobility, balance, and activity confidence after customized VRT
[35–
37]. In the study by Amita et al.
[36], 10 patients with bilateral vestibular lesions showed improvements primarily in the functional domain of the DHI, falls efficacy, and mCTSIB sway velocity; the findings should be interpreted cautiously because of the small sample and uncontrolled design.
Studies in mixed vestibular populations also suggest that individualized assessment, professional instruction, and adherence influence treatment response
[38–
40]. Because these studies were not restricted to BVH/BVP, they provide indirect support for individualized treatment rather than direct evidence of BVH-specific efficacy.
Recommendation 7: Gaze-stabilization training that incorporates active head movement should be used to improve movement-related visual stability in adults with BVH/BVP. Saccadic or smooth-pursuit eye movements performed without head movement should not be used in isolation as the principal intervention for improving gaze stability (Level of evidence: 1b; Recommendation strength: Strong).
Evidence summary
This recommendation was informed by a guideline, an RCT and intervention studies. Direct evidence supports exercises that challenge visual fixation during active head movement, whereas evidence does not support eye-movement-only exercises as the primary treatment for gaze instability.
Rationale
Gaze-stabilization exercises should be selected according to residual VOR function, the severity of movement-induced visual blurring or oscillopsia, compensatory eye-movement capacity, symptom tolerance, and functional goals. Patients with useful residual vestibular function may emphasize adaptation exercises such as VOR×1 and, when appropriate, VOR×2. In patients with profound or near-complete vestibular loss, greater emphasis may be placed on substitution strategies, including coordinated eye–head movements, gaze shifts, predictive saccades, target acquisition, and task-specific visual-stability practice.
The 2022 APTA guideline recommends against using isolated saccadic or smooth-pursuit exercises without head movement as specific gaze-stability exercises
[7]. In a double-blind randomized crossover study, Lehnen et al.
[41] compared head-movement-emphasized rehabilitation with eye-movement-only training in two adults with chronic BVP. Dynamic visual performance improved after the head-movement intervention but not after eye-movement-only training. Because only two patients were studied, these findings should be regarded as mechanistic rather than definitive efficacy evidence.
In the randomized double-blind trial by Herdman et al.
[26], vestibular exercises improved dynamic visual acuity compared with placebo eye-movement exercises. A single-group study of 25 patients with chronic BVH also reported improvements in dizziness, balance, vestibular-related activities of daily living, sleep quality, and kinesiophobia after a structured multimodal rehabilitation program that included adaptation, eye–head coordination, balance, postural, and home exercises
[42]. Because the latter intervention was multimodal, its findings do not isolate the specific effect of gaze-stabilization training.
Recommendation 8: Balance and gait training should be individualized according to the patient’s impairments in postural control, dynamic gait, sensory integration, functional mobility, and fall risk, and should be progressively advanced toward relevant daily activities (Level of evidence: 2b; Recommendation strength: Moderate).
Evidence summary
The evidence included a guideline, intervention and observational studies.
Rationale
The 2022 APTA guideline recommends static and dynamic balance and gait exercises as central components of rehabilitation for peripheral vestibular hypofunction
[7]. For BVH/BVP, training should challenge the systems and tasks most relevant to the individual, including stance and gait under reduced visual input, compliant or uneven surfaces, head movement, turning, speed changes, obstacle negotiation, and dual-task conditions.
Eder et al.
[43] studied 23 patients with BVP who completed a standardized VRT program incorporating balance and gait exercises. FGA and TUG performance improved after training, whereas the addition of noisy galvanic vestibular stimulation conferred no clear additional benefit. These findings support standardized progressive VRT but do not identify the independent contribution of each exercise component.
Boutabla et al.
[31] found that patients with BVP performed worse than healthy controls on the FGA and tandem walking, whereas the TUG test did not consistently distinguish groups. Although this was an assessment rather than an intervention study, it indicates that more demanding gait tasks may better capture BVP-related limitations and should inform the selection and progression of task-specific training.
Recommendation 9: A multifactorial fall-prevention and safety-management strategy should be incorporated into rehabilitation for adults with BVH/BVP (Level of evidence: 2b; Recommendation strength: Moderate).
Evidence summary
This recommendation was informed by a prospective cohort study and a review of falls in BVP. The evidence establishes a high burden of falls but does not directly demonstrate that a specific fall-prevention program reduces fall incidence in this population.
Rationale
Wuehr et al.
[28] prospectively monitored 30 patients with BVH and 30 age- and sex-matched healthy controls. During six months of follow-up, 43% of patients and 13% of controls experienced at least one fall; 70% of patients who fell reported recurrent falls. Patients also showed reduced ambulatory activity, indicating that conventional clinical testing may underestimate the real-world burden of disease.
A review by Herssens et al.
[55] identified 11 studies involving 359 individuals with BVP, of whom 149 (42%) experienced falls during the assessed periods. Commonly reported circumstances included loss of balance, darkness, and uneven ground. However, fall definitions, ascertainment methods, and follow-up procedures varied substantially, and prospective intervention evidence was lacking.
Given the frequency and potential consequences of falls, rehabilitation should include a multifactorial safety strategy based on individual risk. Components may include education, prospective recording of falls and near-falls, environmental modification, adequate lighting, appropriate footwear, assistive-device assessment, strength and reactive-balance training, and graded practice under clinically relevant environmental conditions. High-risk tasks should be undertaken only with appropriate protection and supervision.
Recommendation 10: Following professional assessment, adults with BVH/BVP should undertake an adequately dosed and progressively advanced program of gaze-stabilization, balance, and gait exercises. As a general reference, gaze-stabilization exercises may be performed 3–5 times daily, for a total of 20–40 minutes per day, for approximately 5–7 weeks. Balance and gait training commonly continues for approximately 6–9 weeks; approximately 20 minutes per day may be used as a pragmatic target when tolerated, not as a mandatory minimum. Dose and progression should be individualized according to symptom tolerance, recovery after exercise, fall risk, treatment response, adherence, and goal attainment (Level of evidence: 5; Recommendation strength: Moderate).
Evidence summary
The recommended dosage should be interpreted as a reference range rather than a fixed prescription. Direct evidence supports the effectiveness of gaze-stabilization exercise, but evidence defining the optimal dose is limited. The 6–9 weeks duration proposed for balance training in BVH is based primarily on expert opinion.
Rationale
The 2022 APTA guideline recommends gaze-stabilization exercises 3–5 times daily, for a total of 20–40 minutes per day, for approximately 5–7 weeks in patients with BVH
[7]. It also suggests that balance training commonly continues for 6–9 weeks, while acknowledging that this duration is based largely on expert opinion rather than comparative dose-response trials.
In the randomized double-blind trial by Herdman et al.
[27], participants performed gaze-stabilization or placebo eye-movement exercises 4–5 times daily for a total of 20–40 minutes, together with 20 minutes of daily balance and gait exercises and weekly therapist follow-up. Dynamic visual acuity improved more in the gaze-stabilization group. The study provides a direct basis for the gaze-stabilization dose but included only 13 patients and was not designed to establish the optimal dose.
Krebs et al.
[33] used a progressively advanced program incorporating gaze-stabilization, balance, and gait exercises in eight patients with BVH. Improvements were observed in walking speed and dynamic stability, although the small sample and limited differences in patient-reported disability preclude firm dose recommendations.
Patients with severe oscillopsia, advanced age, high fall risk, visual or somatosensory impairment, cognitive limitations, or poor adherence may require a lower initial challenge, closer supervision, slower progression, or a longer treatment period. Conversely, exercise should not be continued indefinitely without reassessment. When treatment goals have been achieved, improvement has stabilized, or progress has plateaued despite adequate treatment and adherence, patients may transition to a maintenance program.
Recommendation 11: A structured home-exercise program should be combined with initial professional assessment and periodic review of exercise selection, dose, technique, progression, adherence, and safety. Telephone, video, or digital monitoring may supplement professional supervision in appropriately selected patients but should not replace direct care when safety, cognition, comorbidity, or exercise complexity requires closer supervision (Level of evidence: 1b; Recommendation strength: Strong).
Evidence summary
The strongest evidence supports professional assessment and supervision combined with home practice. Evidence for telephone, video, internet-based, and app-supported delivery is promising but is derived predominantly from unilateral, mixed vestibular, or chronic dizziness populations and should therefore be extrapolated cautiously to BVH/BVP.
Rationale
The 2022 APTA guideline strongly recommends supervised vestibular physical therapy because supervision facilitates individualized exercise prescription, correction of technique, progression, monitoring of safety, and adherence
[7]. Professional supervision need not require every session to be conducted face to face, but completely unsupervised exercise without individualized assessment or feedback is less well supported.
In a randomized trial of patients with peripheral vestibular disorders, Pavlou et al.
[44] reported a dropout rate of 55% in the unsupervised group compared with 10% in each supervised group. Supervised training was also associated with more consistent improvements in postural and functional outcomes. Comparative studies in chronic dizziness and vestibular dysfunction have likewise reported improvements with both supervised and home-based treatment, with larger or more consistent gains in some outcomes under supervision
[45].
Evidence concerning remote rehabilitation is supportive but indirect. Systematic reviews indicate that telerehabilitation may improve dizziness severity, dizziness-related disability, and anxiety, but the included populations, intervention formats, doses, and follow-up procedures are heterogeneous and rarely specific to BVH
[46,
47]. Adherence studies further suggest that patients with BVH may have particular difficulty sustaining home exercise and may benefit from individualized adherence support
[40].
A randomized study in chronic unilateral vestibular hypofunction found that periodic reassessment and telephone supervision improved several balance, gait, disability, and adherence outcomes compared with unsupervised home exercise
[48]. These findings provide indirect support for remote supervision in BVH/BVP but should not be interpreted as proof of equivalent efficacy.
Hall et al.
[49] demonstrated the feasibility and usability of a remote therapeutic-monitoring platform for adults with chronic peripheral vestibular hypofunction; the study was not designed to establish clinical efficacy. In a randomized trial involving older adults with chronic vestibular syndrome, internet-based rehabilitation produced clinically meaningful short-term improvement
[50]. At 36 months, improvements within the intervention groups were maintained, but no significant between-group differences remained, partly because participants receiving usual care were subsequently allowed to access vestibular rehabilitation.
D’Silva et al.
[51] found that a gamified mobile application improved exercise accuracy and engagement during a single session in adults with unilateral or bilateral vestibular hypofunction. These findings support digital tools as adjuncts to home training but do not establish long-term clinical effectiveness.
Accordingly, remote telephone or video follow-up and digital monitoring may supplement professional supervision after an appropriate initial assessment. Direct or more frequent supervision should be prioritized for patients with high fall risk, cognitive or communication difficulties, substantial neurological or sensory comorbidity, poor exercise technique, or limited ability to train safely at home. Practical considerations for the formulation, progression, dosage, supervision, and delivery of VRT are summarized in Table 4.
3.4 When should VRT be modified, temporarily suspended, or discontinued?
Recommendation 12: Exercise difficulty or dose should be reduced, or training temporarily suspended, when symptoms worsen persistently, recovery after exercise becomes substantially longer than the patient’s usual response, or falls or near-falls increase. Training should be stopped and urgent medical evaluation arranged when new focal neurological signs, sudden hearing loss, or an acute or rapidly progressive pattern of vestibular symptoms develops (Level of evidence: 5; Recommendation strength: Strong).
Evidence summary
This recommendation was informed by one clinical practice guideline, one international consensus statement, one review, and one prospective observational study. Direct intervention evidence is lacking because the recommendation primarily addresses patient safety. The evidence was therefore classified as Level 5. A strong recommendation was issued because the potential consequences of failing to identify a serious underlying condition are substantial, temporary suspension and reassessment are readily feasible, and the risks associated with this approach are minimal.
Rationale
Mild, transient, and readily reversible dizziness, unsteadiness, or visual blurring may occur during VRT and can represent an acceptable response to exercises intended to promote adaptation, substitution, or habituation. Such symptoms should generally return toward baseline within a clinically acceptable period.
Persistent symptom exacerbation, progressively prolonged recovery after exercise, a marked increase in falls or near-falls, or deterioration in the ability to perform exercises safely should not be regarded as an expected training response. In these circumstances, exercise intensity or complexity should be reduced, or training should be temporarily suspended pending reassessment.
The 2022 APTA guideline emphasizes that treatment decisions should take account of clinical status, exercise tolerance, comorbidities, functional capacity, cognitive and physical ability, and safety risk
[7]. BVH/BVP is independently associated with a high burden of falls. Prospective and review evidence indicates that falls and recurrent falls are frequent and commonly occur under conditions of reduced visual input or uneven support
[28,
55]. A substantial increase in falls during treatment therefore warrants immediate review of exercise selection, supervision, assistive-device use, and environmental safety.
The emergence of focal neurological signs, rapidly progressive ataxia, sudden or rapidly worsening hearing loss, or a new pattern of acute vestibular symptoms should prompt consideration of central neurological, vascular, neurodegenerative, or acute otological disease. The Bárány Society diagnostic framework requires that the presentation not be better explained by another disorder
[1]. New or changing clinical features should therefore not be attributed automatically to VRT. Referral to neurotology, neurology, emergency care, or another appropriate specialty should be arranged according to the suspected condition and its urgency.
Recommendation 13: Patients may transition from active, supervised rehabilitation to maintenance exercise and long-term self-management when goals have been achieved, symptoms and functional performance have stabilized, and the patient can exercise safely and independently, or when a genuine therapeutic plateau is reached after diagnostic review, adequate treatment dose and progression, adherence, and modifiable comorbidities have been addressed (Level of evidence: 2b; Recommendation strength: Moderate).
Evidence summary
This recommendation was informed by one clinical practice guideline, one systematic review, and longitudinal observational evidence. Evidence concerning specific discharge criteria remains limited, and decisions should be individualized according to functional goals, residual disability, comorbidity, and patient preference.
Rationale
The 2022 APTA guideline identifies achievement of primary goals, resolution or stabilization of symptoms, normalization of balance or vestibular function, and a plateau in progress as potential reasons for discontinuing active treatment
[7]. Complete normalization of peripheral vestibular function should not, however, be required in BVH/BVP because bilateral vestibular impairment frequently persists despite meaningful functional improvement.
For adults with BVH/BVP, discharge decisions should focus on clinically relevant function rather than normalization of vestibular laboratory findings. Appropriate criteria include attainment of patient-centered goals, stabilization of gaze and postural function, safe community mobility, independent performance of an appropriate home program, and the absence of further meaningful improvement despite adequate treatment progression and adherence.
Herdman et al.
[26] demonstrated that functional visual performance can improve after VRT even when peripheral vestibular deficits remain. Residual impairment after formal treatment therefore does not necessarily indicate treatment failure, but may require continued maintenance exercise, compensatory strategies, environmental modification, or assistive-device use.
In a retrospective longitudinal study of 97 patients with BVP, Loos et al.
[52] found that vestibular function, symptoms, and health-related quality of life generally did not show clinically relevant spontaneous improvement over a median follow-up of 24 months. These findings support viewing BVH/BVP as a condition requiring long-term functional management. Transition from supervised treatment should therefore be accompanied by an individualized maintenance plan and clear instructions regarding indications for reassessment.
3.5 Should adjunctive or emerging interventions be considered in adults with BVH/BVP who respond inadequately to conventional VRT?
Recommendation 14: In adults with severe BVH/BVP who remain substantially limited after optimized VRT, a time-limited, goal-directed trial of selected non-invasive sensory-augmentation or sensory-substitution technology may be considered following specialist assessment. The intervention should complement, not replace, core VRT and should be discontinued if prespecified functional goals are not met. nGVS should not be used as routine clinical treatment outside an approved research protocol or closely governed investigational pathway (Level of evidence: 2b–3b; Recommendation strength: Weak).
Evidence summary
The evidence included systematic reviews, randomized or crossover studies, and multiple small observational and experimental studies. The interventions, stimulation paradigms, outcome measures, and study populations were highly heterogeneous. Most studies were small, uncontrolled, conducted under laboratory conditions, or enriched for patients who had shown an initial response. Evidence for sustained improvement in falls, community mobility, participation, or quality of life remains insufficient.
Rationale
Some patients with severe BVH/BVP retain substantial disability after appropriately delivered VRT. Sensory-substitution and sensory-augmentation technologies seek to provide supplementary information about body orientation or movement through tactile, auditory, visual, or electrical signals. Their intended role is to support postural control and mobility when natural vestibular input and conventional compensatory strategies remain inadequate.
(1) Vibrotactile and electrotactile feedback
Vibrotactile devices typically encode trunk tilt, body orientation, or deviation from vertical and transmit this information through tactors placed around the waist or trunk. In a 2024 observational study of 121 patients who considered previous vestibular rehabilitation insufficient, Kingma et al.
[53] reported improvement in self-rated balance and mobility among a selected subgroup using a subconscious vibrotactile belt, with relatively high adherence among those who experienced early benefit. The study had no control group, relied primarily on self-reported outcomes, and selected continued users on the basis of their initial response.
An earlier study of 39 patients with severe bilateral vestibular loss used a 2-hour preselection trial before allowing initial responders to use a vibrotactile balance belt in daily life
[54]. Twenty-three of the original 39 patients ultimately reported a clear benefit. Because the study was uncontrolled and enriched for early responders, it should not be interpreted as a randomized efficacy trial.
Reviews and small experimental studies have also evaluated vibrotactile neurofeedback, virtual-reality–assisted training, computerized dynamic posturography, and electrotactile feedback delivered through the tongue
[56-
62]. Some studies reported short-term improvements in postural sway, balance-related performance, or patient-reported outcomes; however, device design, training protocols, comparator interventions, and study populations varied considerably. BrainPort and related tongue-based systems have shown feasibility in small uncontrolled cohorts, including patients with bilateral vestibular loss, but evidence for durable, clinically important benefit remains limited
[61,
62].
(2) Auditory feedback
Auditory feedback systems convert trunk acceleration, postural sway, or body orientation into sound cues. Small laboratory studies reported improved postural performance in patients with bilateral vestibular loss during challenging sensory conditions
[63,
64]. These findings support the physiological feasibility of auditory substitution but do not establish long-term effectiveness in everyday mobility or fall prevention. Auditory feedback may also interfere with environmental listening and may be unsuitable for patients with clinically important hearing loss.
(3) Noisy galvanic vestibular stimulation (nGVS)
nGVS applies imperceptible or near-threshold stochastic electrical stimulation across the mastoid processes with the aim of enhancing residual vestibular signalling through stochastic resonance
[65].
The 2022 systematic review and meta-analysis by McLaren et al.
[66] identified seven studies evaluating postural control in people with BVP. The pooled findings suggested a short-term reduction in postural sway, particularly during eyes-closed standing, but the effects were task dependent and the included studies were small and heterogeneous. The review did not establish sustained improvement in falls, participation, or quality of life.
Small experimental studies have reported immediate or short-lived improvements in gait speed, step length, postural sway, self-motion perception, or dynamic stability in selected patients with BVP
[67–
71]. Studies in community-dwelling older adults or healthy participants provide mechanistic support but should not be treated as direct evidence of clinical effectiveness in BVH/BVP
[72–
74]. Physiological studies have further suggested that baseline instability, residual vestibular function, and stimulation parameters may influence response
[69,
70].
Fujimoto et al.
[71] reported that a 30-minute subthreshold nGVS session was followed by improvements in postural sway and perceived instability lasting several hours in 13 patients with BVP. The uncontrolled design, small sample, potential placebo effects, and selection bias limit causal interpretation.
Eder et al.
[43] evaluated nGVS as an adjunct to a standardized VRT program. Functional gait and mobility improved after rehabilitation, but nGVS provided no clear additional benefit. More importantly, a 2025 multicenter, randomized, double-blind, placebo-controlled crossover trial did not demonstrate a significant overall effect of prolonged nGVS on posture or gait in patients with poorly compensated vestibulopathy
[75].
Accordingly, nGVS remains an investigational intervention. Its optimal candidates, stimulation parameters, durability of response, and effect on patient-important outcomes have not been established.
Recommendation 15: Vestibular implantation should not be offered as routine clinical care. It may be considered only within an ethics-approved clinical trial or formally governed investigational pathway at a specialist multidisciplinary center, following stringent patient selection, comprehensive assessment of potential benefits and harms, including the risk of hearing loss and fully informed consent (Level of evidence: 2b; Recommendation strength: Strong against routine clinical use).
Evidence summary
The evidence included one systematic review, one narrative review, prospective nonrandomized cohort studies, small case series, and case reports. Although vestibular implants can generate vestibular reflexes and have shown promising functional and patient-reported outcomes in selected recipients, the total number of implanted participants remains small. Surgical approaches, devices, stimulation paradigms, outcome measures, and follow-up periods differ substantially across studies.
Rationale
A vestibular implant is an implantable neuroprosthesis designed to encode head motion and electrically stimulate branches of the vestibular nerve. Its intended functions include partial restoration of the VOR and vestibulo-collic reflexes (VCR), improvement in gaze and postural stability, and reduction in movement-related visual and balance disability.
A 2025 systematic review identified 21 eligible studies involving 36 vestibular implant recipients across four centers worldwide
[76]. The review concluded that vestibular implantation can evoke vestibular reflex responses and may improve selected clinical outcomes, but emphasized the heterogeneity of devices, surgical techniques, fitting strategies, and outcome measures. Hearing preservation, long-term continuous use, durability, adverse events, and cost-effectiveness remain incompletely characterized.
A 2024 review similarly characterized vestibular implantation as an investigational technology undergoing clinical evaluation in the United States and Europe
[77]. Published studies indicate that electrical stimulation can drive vestibulo-ocular and vestibulo-collic responses and may improve dizziness-related disability and quality of life, but the technology has not reached the evidentiary or regulatory threshold for routine clinical use.
In a prospective, nonrandomized, open-label single-center cohort, 10 adults with severe BVH who received a multichannel vestibular implant reported clinically important improvements at 6 months in the Dizziness Handicap Inventory, Vestibular Disorders Activities of Daily Living scale, and 36-Item Short Form Survey (SF-36) utility score; a separate group of 10 individuals awaiting implantation did not show comparable mean changes
[78]. The small convenience sample, lack of randomization and masking, reliance on patient-reported outcomes, and potential conflicts of interest require cautious interpretation.
Studies examining stimulation parameters have demonstrated that electrical amplitude, pulse duration, frequency, and rate of change influence the magnitude and dynamic range of electrically evoked vestibulo-ocular responses
[79]. These findings are important for future device fitting but do not establish clinical effectiveness.
Small feasibility studies and case reports indicate that vestibular nerve pathways may remain responsive after longstanding vestibular deprivation
[80]. Other small studies have reported improvements in dynamic visual acuity or oscillopsia with the implant activated, although changes may also reflect compensatory saccades and nonvestibular strategies
[81].
In the clinical trial reported by Chow et al.
[82], eight adults with severe BVH underwent unilateral implantation with stimulation of three semicircular-canal nerve branches. Measures of posture, gait, and quality of life generally improved at 6 months and 1 year, and placebo-mode testing supported an effect of motion-modulated stimulation. However, implantation caused ipsilateral hearing loss: at 6 months, pure-tone thresholds increased by 3–16 dB in five participants and by 74–104 dB in three participants.
Boutros et al.
[83] reported stable, motion-modulated three-dimensional vestibulo-ocular responses during prolonged continuous use in four participants with ototoxic bilateral vestibular loss. These findings support technical feasibility and physiological durability, but the sample size was insufficient to establish clinical effectiveness or safety.
Vestibular implantation therefore remains a promising but experimental intervention. Future trials should address comparative effectiveness, long-term safety, preservation of residual hearing and vestibular function, patient selection, rehabilitation requirements after implantation, device reliability, cost-effectiveness, and outcomes that are meaningful to patients.
4 Guideline limitations and research priorities
The BVH/BVP-specific evidence base is limited by small samples, heterogeneous diagnostic criteria, variable residual vestibular function, multimodal interventions, inconsistent outcome measures, short follow-up, and frequent reliance on indirect evidence from mixed vestibular populations. The evidence-level and recommendation-strength framework used in this guideline is a transparent modification rather than the official OCEBM A–D recommendation-grade system. The guideline does not establish a validated core outcome set, and several patient-reported instruments require Chinese-language cross-cultural validation.
Priority research areas include development of a consensus core outcome set; adequately powered BVH/BVP-specific trials; dose–response studies for gaze, balance, and gait training; prospective fall-prevention trials; predictors of treatment response; long-term effectiveness and cost-effectiveness; equitable implementation in community settings; validation of remote and digital models; and comparative trials of sensory augmentation, nGVS, and vestibular implantation using patient-important outcomes and rigorous safety monitoring.
5 Supplementary files
Supplementary material is available in the online version of this article at
https://doi.org/10.15302/ENTD.2026.090006 and is accessible for authorized users.
The Author(s). This article is published by Higher Education Press at journal.hep.com.cn.
This is an open access article distributed under the Creative Commons Attribution License 4.0 (CC BY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
PDF
(5493KB)
