Abstract
Objective
To assess whether preoperative vestibular rehabilitation and/or intratympanic gentamicin (ITG) improve vestibular compensation and postoperative functional outcomes in patients undergoing vestibular schwannoma surgery.
Design
Systematic review conducted in accordance with the PRISMA 2020 statement.
Study sample
Twelve original studies were identified through PubMed, Web of Science, and Google Scholar. Eligible designs included clinical trials, cohort, case–control, and case-series studies evaluating preoperative vestibular rehabilitation and/or ITG before surgery. Methodological quality was assessed using the Joanna Briggs Institute (JBI) critical appraisal checklists and the RoB 2 tool.
Results
Three studies implemented ITG alone, whereas nine integrated ITG and/or structured vestibular rehabilitation within multimodal prehabilitation protocols. Rehabilitation components ranged from supervised physiotherapy to home-based or hybrid programs. A Bayesian random-effects meta-analysis of three studies evaluating hospital length of stay (LOS) yielded a small-to-moderate pooled effect favouring prehabilitation (μ = 0.28; 95% CrI −0.37 to 1.00), with moderate heterogeneity and wide uncertainty. Other outcomes such as posturography and dizziness handicap could not be meta-analysed due to methodological heterogeneity. Overall risk of bias was moderate-to-high.
Conclusions
Preoperative prehabilitation may support postoperative recovery, particularly when tailored to residual vestibular function and delivered through supervised or hybrid formats. Larger, high-quality randomized trials using standardized outcomes are warranted.
Keywords
Introduction
The preoperative approach in vestibular ablative surgery has become a key focus in contemporary medical research. This review provides an update and critical analysis of the existing scientific literature. Vestibular schwannoma or acoustic neuroma represents more than 80% 1 of cerebellopontine angle tumors and approximately 8% 2 of intracranial tumors. The symptoms caused by the mass effect related to gradual growth determine the quality of life of patients and include hearing loss, tinnitus, hydrocephalus, facial paresthesia, headache, and vertigo. 2 The management strategy typically involves a watch-and-wait approach, radiotherapy, or surgical intervention. 3
This systematic literature review adheres to the methodological standards and guidelines established in the PRISMA 2020 statement. 4 We evaluate the available therapeutic options before surgery that may improve vestibular compensation and thus reduce dizziness and associated disability. This review builds upon and complements recent systematic analyses by Potdar et al. 5 and Fuentealba Bassaletti et al. 6 which primarily evaluated ITG as a prehabilitation strategy. In contrast, the present work broadens this scope by integrating evidence from both pharmacological and combined (pharmacological + rehabilitative) preoperative interventions, aiming to close remaining methodological gaps and clarify their respective contributions to vestibular compensation and postoperative recovery. Vestibular compensation is the spontaneous, multi-level plastic reorganization of central vestibular pathways that rapidly re-balances resting activity in the vestibular nuclei to resolve static deficits and, over time, mitigates dynamic deficits through sensory and behavioral substitution. 7
We analyze the methodological quality of the included studies to synthesize the evidence and highlight existing needs. Answering the research question allows us to understand whether pre-surgical treatment with ITG is better than rehabilitative treatment, or whether rehabilitative treatment is better than ITG, or if we should offer combined pre-surgical treatments in patients with vestibular schwannoma. This systematic review aims to compile all available therapeutic approaches in order to offer a comprehensive perspective that supports evidence-based decision-making.
Methods
Search, protocol and register
The literature search was performed independently by two reviewers (ATT & MJDM) between 1 and 30 April 2024 in PubMed, Web of Science, and Google Scholar, following the PICO framework. Population: patients with unilateral vestibular schwannoma; Intervention: pre-surgical intervention (vestibular rehabilitation or ITG); Comparators: no pre-surgical treatment, pharmacological therapy, or rehabilitation alone; Outcomes: vestibular compensation and postoperative recovery.
Search terms were combined using Boolean operators as follows: prehabilitation vestibular neuroma OR (prehabilitation vestibular schwannoma AND rehabilitation vestibular neuroma) OR rehabilitation vestibular schwannoma OR vestibular prehab OR vestibular prehabilitation. Searches in PubMed (Advanced Search, All Fields) and Web of Science (Advanced Search, All Databases, All Collections) were executed without design or date filters, while Google Scholar used the same keywords as a free query to maximize sensitivity. The complete search strategy, including the full automatic term mapping (ATM) expanded PubMed query generated during the original April 2024 search, is provided in Supplemental Table S1. The complete process is shown in the PRISMA flowchart (Figure 1). Preferred reporting items for systematic reviews and meta-analyses (‘PRISMA’) flowchart.
The search was updated (1–30 May 2025) by re-running the same strategy in all databases without date restrictions, in order to capture newly published as well as newly indexed records. Records already evaluated in the April 2024 screening were identified and excluded during the screening phase to avoid duplication.
The protocol was registered in PROSPERO (CRD42024504365) in March 2024 and amended in August 2024, February 2025, and August 2025 to ensure transparency and methodological rigor.
Eligibility criteria
Included studies enrolled in human patients with vestibular schwannoma, with no restriction on age or publication year, who received a pre-surgical therapeutic intervention (vestibular rehabilitation, ITG, or combined). Eligible designs comprised randomized or non-randomized clinical trials, quasi-experimental studies, prospective/retrospective cohort, case–control, and comparative case series reporting outcomes related to vestibular compensation or postoperative recovery. Studies had to involve an intervention delivered before surgery; purely observational reports without an intervention were not eligible.
Exclusion criteria were defined a priori and applied uniformly. Studies were excluded if they lacked a pre-surgical intervention; presented duplicate/overlapping populations (retaining the most complete/representative report); were conference abstracts or other grey literature; had ineligible designs (systematic/narrative reviews, opinion pieces, single case reports, clinical practice guidelines, theoretical papers); were animal studies; included patients with neurodegenerative disorders or neurofibromatosis type II; or were not published in English, French, Spanish, or German. No date limits were applied; language limits were chosen for accessibility and scientific rigor.
Study selection and data collection process
Two independent reviewers (ATT & MJDM) screened all retrieved records in Rayyan using a two-phase process (Phase 1 and Phase 2). The initial classification (‘include’, ‘exclude’, ‘uncertain’) was performed blindly and independently to ensure methodological transparency. Duplicates were removed automatically and manually using Zotero prior to screening.
During Phase 1, records were excluded as clearly irrelevant based on title and abstract. During Phase 2, full-text articles were assessed for eligibility according to predefined inclusion and exclusion criteria, and specific reasons for exclusion were recorded (screening duplicates, systematic reviews, practice guidelines, narrative or book chapters, grey literature/conference abstracts, no preoperative intervention applied, overlapping/duplicated cohorts, or did not meet inclusion criteria). Discrepancies were resolved by discussion or, when necessary, by a third reviewer (ACC).
Risk of bias was assessed independently by both reviewers using design-specific tools: the JBI Critical Appraisal Checklists 8 for quasi-experimental, cohort, case-control studies, and case-series studies, and the RoB 9 tool for the single randomized controlled trial. These tools were selected for their validity across heterogeneous designs and their wide use in clinical research. Each question was judged following JBI guidance, 10 and overall risk (low, moderate, high) was determined qualitatively. Percentages of affirmative (“Yes”) responses were reported descriptively for transparency but not used to define categories (Supplemental Appendix A). Overall judgments reflected limitations in key domains such as confounding, participant retention, and statistical conclusion validity.
Data extraction and analysis
Data were extracted using a standardized Excel spreadsheet. Variables extracted included author and year, study design, risk of bias score, participant characteristics, type of pre-surgical intervention, surgical intervention, outcome measures, and main results. Data were summarized in tables for qualitative interpretation and Bayesian analysis.
Main features of the included studies.
ABC, activities-specific balance confidence scale; aVOR, angular vestibulo-ocular reflex; CA, anterior canal; CES, composite equilibrium score; CH, horizontal canal; CP, posterior canal; EEV, european evaluation of vertigo; FGA, functional gait assessment; GAD-7, generalised anxiety disorder scale (7 items); GBI, glasgow benefit inventory; GHSI, glasgow health status inventory; OKN, optokinetic; PANQOL, penn acoustic neuroma quality of life; PTA, pure tone audiometry; SNC, central nervous system; VEMPs, vestibular evoked myogenic potentials; VF, vestibular function; VNG, videonystagmography; ZUNG, zung self-rating depression scale. Summary of findings from the studies included in the systematic review.
Statistical analysis
The quantitative synthesis focused on the outcome of Length of Hospital Stay (LOS), which provided sufficient and consistent between-group data across studies. Other variables, including posturography/Sensory Organization Test (SOT) and subjective measures such as the Dizziness Handicap Inventory (DHI), were excluded because the results were reported using different outcome parameters and follow-up times, preventing standardized effect size calculation and comparability between studies. These outcomes were retained for qualitative interpretation. In this review, posturography refers to quantitative, instrumented assessments of balance control (e.g., sway or torque variance), whereas the SOT represents a standardized posturographic protocol assessing balance under six sensory conditions. Effect sizes (Cohen’s d) and standard errors (SE) were derived from reported group differences using validated formulas based on z-statistics or p-values. 11 Bayesian random-effects models were fitted with the bayesmeta package in RStudio (v2025.06.1). For the overall effect (μ), a flat non-informative prior was applied, and for heterogeneity (τ), a half-normal prior (SD = 1) was used, as recommended for small meta-analyses to avoid overestimation of variability.
Model robustness was assessed through sensitivity analyses using alternative priors (half-normal, half-Cauchy, log-normal, uniform) and a leave-one-out approach.
Bayesian analysis of hospital length of stay
A Bayesian random-effects meta-analysis was performed to estimate the effect of prehabilitation on LOS as a clinical indicator of postoperative recovery. Three studies reporting between-group differences with sufficient statistical information were included. One study by Amiraraghi et al. 12 was excluded because of its very small sample size and the inclusion of a non-vestibular schwannoma patient in the control group, which compromised comparability. Detailed numerical outputs and sensitivity tests are reported in Supplemental Appendix B and C.
Results
A total of 1909 records were identified across the three databases in April 2024 (sum of both reviewers): 500 and 509 from PubMed, 330 and 348 from Web of Science, and 111 and 111 from Google Scholar. After duplicate removal using Zotero, 1185 records were excluded. The first screening (title and abstract) was performed on the remaining 724 records, of which 693 were excluded as irrelevant.
In the second screening phase, 31 articles were reviewed in full text, and 23 were excluded for the following reasons: screening duplicates (n = 8), systematic reviews (n = 4), practice guideline (n = 2), narrative review (n = 1), book chapter (n = 1), grey literature/conference abstract (n = 1), no preoperative intervention applied (n = 1), and overlapping/duplicated cohort within the same research group (n = 5). Eight articles were included for final evaluation.12–19
The updated search conducted in May 2025 identified 1043 additional records per reviewer (2086 in total): 540 and 540 from PubMed, 369 and 369 from Web of Science, and 134 and 134 from Google Scholar. After removing 1043 duplicates in Zotero, the remaining 1043 unique records proceeded to the first screening phase in Rayyan (title and abstract). During this phase, 928 records published before 2024 were excluded because they had already been analyzed in the initial search. The remaining 115 records moved to the second screening phase for full-text eligibility review, where 113 were excluded (did not meet inclusion criteria n = 90; duplicates n = 9; systematic review n = 12; included in April 2024 n = 2). Finally, two studies were included for the final review.20,21
Three additional studies were identified via other methods (two through ResearchGate and one during peer review). One of these was excluded as grey literature/conference abstract, and two were included.22,23
As a result, the final review comprised 12 studies in total.
The selection process is detailed in the PRISMA flow diagram (Figure 1). Due to methodological and clinical heterogeneity among the 12 included studies, a full quantitative synthesis was not feasible.
A Bayesian random-effects meta-analysis was conducted for LOS, including three studies that reported intergroup (experimental vs control) comparisons with extractable standardized mean differences.15,16,21 Using a half-normal prior with a scale of 1 for the between-study heterogeneity (τ), the pooled posterior mean effect size was 0.28 (Cohen’s d; 95% credible interval [CrI]: −0.37 to 1.00), suggesting a small-to-moderate tendency toward shorter LOS in prehabilitated groups. Between-study heterogeneity was low-to-moderate (τ = 0.25; 95% CrI: 0.00–1.07), and the 95% prediction interval (−0.95 to 1.59) indicated that the true effects in future comparable studies are likely to remain within a broad range. In Bayesian inference, a 95% credible interval (CrI) represents the range within which the true effect size lies with 95% posterior probability, differing conceptually from a frequentist confidence interval (CI). Although both the credible interval for the pooled mean effect and the prediction interval included the null value, the posterior distribution remained skewed toward a beneficial direction. These findings are illustrated in Figure 2 and summarized numerically in Supplemental Appendix B and C. Bayesian random-effects forest plot for LOS. Bayesian random-effects forest plot for between-group differences in hospital length of stay (LOS) (n = 3 studies). Each study displays its original reported effect size (quoted estimate, black square) and the corresponding posterior shrinkage estimate (grey diamond) derived from the Bayesian hierarchical model. Effect sizes are expressed as standardized mean differences (Cohen’s d) with 95% credible intervals (CrI). The diamond represents the pooled posterior mean (μ = 0.28; 95% CrI: −0.37 to 1.00), and the grey bar indicates the 95% prediction interval (−0.95 to 1.59). Between-study heterogeneity was low-to-moderate (τ = 0.25; 95% CrI: 0.00–1.07).
Sensitivity analyses using alternative prior distributions for heterogeneity (half-normal, half-Cauchy, log-normal, and uniform) yielded consistent pooled estimates (μ ≈ 0.28–0.29) with overlapping CrIs, confirming the robustness of the results. Leave-one-out analyses under the HN(1) prior produced pooled means ranging from μ = 0.26 to 0.40, with Fellmann et al. 16 and Trudel et al. 15 exerting the greatest numerical influence but without altering overall conclusions. Overall, the Bayesian estimates for LOS consistently indicated a small-to-moderate effect favouring prehabilitation, although with wide uncertainty and credible intervals encompassing the null.
Assessment of risk of bias
Risk of bias was evaluated using design specific JBI Critical Appraisal Checklists and the RoB 2 tool. Most studies showed moderate to high risk of bias, primarily related to residual confounding and, in some cohort studies, incomplete follow-up. Two studies19,21 presented lower risk levels, whereas one randomized trial, 13 one quasi-experimental study 14 and one cohort study 17 showed an overall high risk of bias. The remaining studies were judged to have a moderate risk of bias. Full scoring details are provided in Supplemental Appendix A.
Evaluation using a results table
The aim was to synthesise the information extracted from each article to enhance its understanding and analysis (Table 1). A descending table layout was created, where the uppermost rows include studies with the highest level of evidence, classified by study design. This structure enables a clearer synthesis of intervention types, outcome measures, and main findings, and supports identification of studies eligible for Bayesian comparison.
Discussion
Discussion is organized by considering three complementary outcome domains reported across the included studies: objective physiological measures of vestibular function (vHIT or caloric testing), objective functional outcomes related to balance and mobility (posturography or LOS), and patient-reported clinical outcomes (DHI or balance confidence scales). These domains are discussed separately because they assess different levels of the vestibular system sensory input, observable motor performance, and perceived impact and changes at one level do not necessarily translate into improvements at another.
Given the methodological heterogeneity in study design, intervention components, and outcome assessment, results are interpreted within each outcome domain rather than combined across fundamentally different endpoints. This structured approach helps distinguish physiological effects from functional performance and patient-reported impact and explains why apparently inconsistent findings may coexist across studies.
Overview of intervention types and study designs
According to the updated results table, all twelve studies included in this review incorporated ITG in at least one study arm or as part of a combined prehabilitation protocol, although the degree of application varied. ITG remained the predominant intervention, typically delivered within a multimodal prehabilitation framework that often included structured or home-based vestibular exercises. Eight studies,12,14–16,20–23 included structured intergroup comparisons with and without ITG, allowing partial attribution of effects to the pharmacological component, though co-interventions such as balance or gaze-stabilization training were frequently applied. In contrast, Ferguson et al. 17 and Magnusson et al. 18 reported uncontrolled series combining ITG and individualized prehabilitation, making it difficult to isolate the specific contribution of the drug. Tarnutzer et al. 19 described physiological deafferentation with ITG without evaluating functional outcomes, while Cada et al. 13 explored a mixed program combining ITG and home-based exercises with limited methodological rigor.
This framework, developed in the following section, mirrors prior systematic efforts such as Potdar et al. 5 which also noted the difficulty in isolating the specific contribution of ITG due to overlapping interventions and methodological variability. Earlier reports by Magnusson et al.24,25 and Tjernström et al.26–28 were excluded from the current analysis because of potential cohort overlap yet are cited contextually to clarify the historical development of vestibular prehabilitation and the decision to retain the most representative datasets.18,22
Physiological evidence of deafferentation and clinical implications
Of the twelve included studies, only three, Magnusson et al. 18 Tarnutzer et al. 19 and Bonaventurová et al. 20 provided objective preoperative evidence of ITG induced vestibular function reduction, documented using physiological measures such as caloric testing, vHIT, and VEMP. In this review, vestibular deafferentation is used as an umbrella term to describe objectively documented reductions in peripheral vestibular input, encompassing a spectrum from partial vestibular hypofunction to near-complete vestibular loss, acknowledging that included studies operationalized this concept using different physiological markers and thresholds.
Tarnutzer et al. 19 demonstrated canal-specific reductions in angular VOR gain and increased corrective saccades, confirming targeted vestibular deafferentation. Similarly, Bonaventurová et al. 20 compared ITG preoperatively with home exercises versus postoperative optokinetic (OKN) stimulation during the first 10 days after surgery and standard rehabilitation, clarifying that only ITG constituted a pre-surgical intervention. This study also reported significant preoperative vestibular function reduction following ITG, including decreased VOR gain, increased caloric weakness, and reduced cVEMP responses, supporting partial vestibular deafferentation before surgery. Postoperative DHI outcomes at 3 months showed changes in the total score and, descriptively, in subscale scores, as reported by the original authors. Earlier studies such as Magnusson et al. 18 and Tjernström et al. 22 further supported this physiological mechanism, showing that gradual ITG-induced loss before surgery can facilitate early central compensation. Magnusson et al. 18 documented near-complete vestibular deafferentation prior to surgery, evidenced by caloric asymmetry approaching 98%, whereas Tjernström et al. 22 contrasted slow ITG-induced deafferentation with fast surgical vestibular loss, demonstrating more efficient multisensory integration after gradual vestibular deprivation. These findings suggest that pre-surgical chemical deafferentation may precondition the vestibular system, allowing smoother post-operative adaptation. However, not all studies linked physiological deafferentation to functional benefit. Fellmann et al. 16 found no significant differences in postural stability or DHI, despite confirmed ITG-induced hypofunction. In this study, vestibular involvement was defined at a functional level without physiological quantification of the magnitude of vestibular loss, reflecting partial hypofunction rather than the near-complete deafferentation described by Magnusson et al. 18 Similarly, Hrubá et at. 14 and Cada et al. 13 reported postoperative changes in posturography and subjective outcomes across groups, without a clear advantage for prehabilitated patients. These inconsistencies indicate that physiological vestibular loss alone does not guarantee better recovery, and that variability in how vestibular deafferentation was defined and measured across studies may partly explain the context-dependent and heterogeneous functional outcomes reported.
Vestibular rehabilitation and hybrid strategies
Multiple studies combined ITG with vestibular rehabilitation, implemented as supervised, home-based, or hybrid programs. Balatková et al. 23 paired ITG with home balance training, reporting improved quality of life and reduced anxiety but no significant group differences in dizziness or depression. Hrubá et al. 14 integrated ITG into a pre- and postoperative program, noting balance and confidence gains in both groups without intergroup differences. Bonaventurová et al. 20 reported that mild hearing deterioration occurred in some ITG-treated patients, consistent with its ototoxic potential. As reported by Bonaventurová et al. 20 postoperative changes in DHI scores were observed following both ITG and OKN stimulation, although the overall pattern of improvement differed between intervention groups.
Ferguson et al. 17 applied individualized exercises with ITG, showing early mobilization and good tolerance despite lacking a control group. Cada et al. 13 though limited by small sample size and high bias risk, reported similar improvements in dizziness and resilience between supervised and unsupervised groups. Trudel et al. 15 presented a structured hybrid model ITG plus supervised training reporting faster mobilization and a modest, non-significant reduction in hospital stay.
Tjernström et al. 22 observed that gradual ITG deafferentation with activity encouragement improved multisensory integration, emphasizing early movement. Fellmann et al. 16 found no additional functional benefit from vestibular prehabilitation, illustrating variability among studies. Magnusson et al. 18 described rapid postoperative mobility and preserved stability following ITG prehabilitation, supporting its compensatory potential.
Overall, functional outcomes appeared to depend more on rehabilitation intensity, structure, and supervision than on ITG itself. Structured or supervised prehabilitation was associated with more consistent functional adaptation, whereas unsupervised or observational programs yielded more variable effects.14,18,22,23
Functional recovery and hospital stay
Regarding hospital stay and early mobilization, five studies reported relevant data.12,15–17,21 Although definitions and metrics varied, most studies used LOS as a proxy for postoperative functional recovery. While Amiraraghi et al. 12 reported a significant reduction in LOS, other studies showed non-significant trends or conditional effects based on age or tumor size. Ferguson et al. 17 findings suggested feasibility of compensation post-ITG, but lacked a comparator group. Fellmann et al. 16 data showed no between-group differences in DHI, Functional Gait Assessment, or LOS, suggesting no added clinical value of prehabilitation.
Quantitative synthesis (Bayesian meta-analysis)
The Bayesian synthesis showed small and uncertain effects favouring multimodal prehabilitation strategies integrating ITG as a pharmacological component, consistent with previous systematic reviews.5,6 Both reported physiologically effective vestibular deafferentation but no consistent improvement in functional or subjective outcomes. In our model, the modest reduction in hospital stay suggests that physiological preconditioning alone may be insufficient to accelerate recovery without robust postoperative rehabilitation. These findings highlight the complex interaction between preoperative vestibular loss, compensatory capacity, and continuity of rehabilitation across the perioperative period. Overall, the evidence indicates that ITG-integrated prehabilitation is physiologically sound but clinically unproven, with potential benefits constrained by methodological heterogeneity and small sample sizes.
Inconsistent or null results
Some studies failed to demonstrate clear clinical benefits of preoperative interventions. Cada et al. 13 observed increased DHI scores post-surgery in both groups, with no added benefit from combining ITG and supervised training. Fellmann et al. 16 in a larger retrospective cohort, reported no significant differences between groups in DHI, Functional Gait Assessment, or LOS. Potdar et al. 21 similarly found no statistically significant reduction in LOS, although younger patients and those with smaller tumors appeared to benefit more from ITG-based protocols. In this study, however, ITG was applied as a stand-alone intervention rather than as part of a structured prehabilitation program, suggesting that baseline differences in age or tumor size rather than the intervention itself may have accounted for the observed trend. These negative or inconclusive findings highlight the variability in outcome sensitivity, the possible impact of individual baseline characteristics, and the importance of surgical context and rehabilitation fidelity.
Implications for research and clinical practice
The available evidence, mostly derived from observational and quasi-experimental studies, suggests that preoperative interventions, primarily pharmacological and implemented within multimodal prehabilitation programs, have the potential to improve postoperative outcomes in selected patients. However, their efficacy is not uniform and seems to depend on factors such as the type, timing, and supervision of the intervention, as well as individual characteristics like residual vestibular function and overall physical condition.
Taken together, previous research suggests that integrating individual vestibular profiles and behavioural factors into prehabilitation planning may enhance postoperative recovery, rather than applying uniform protocols. Patients with partial vestibular reserve, higher baseline activity levels, or greater adherence to preoperative exercises tend to adapt more efficiently, indicating that prehabilitation efficacy depends as much on patient-specific factors as on the intervention itself.14,18,20,22,23
From a clinical standpoint, the data point toward a tailored approach: combining ITG with structured vestibular rehabilitation could be considered for patients with sufficient vestibular reserve and an acceptable auditory-risk profile, although current evidence remains limited. This recommendation remains consistent with the cautious conclusions of Potdar et al., 5 whose systematic review focused specifically on ITG prehabilitation. Their findings highlighted the physiological rationale for chemical deafferentation but also emphasized the limited clinical evidence and small sample sizes supporting its isolated use. In line with these observations, Fuentealba Bassaletti et al. 6 found that ITG prehabilitation produced measurable physiological deafferentation on objective vestibular tests, yet no consistent improvement in subjective outcomes such as dizziness perception, anxiety, or balance confidence. This reinforces the distinction between objective vestibular changes and perceived functional recovery, emphasizing the need for multidimensional outcome measures in future trials.
The following references, although not part of the reviewed sample, are cited to contextualize these findings within broader clinical and neurophysiological research. In this framework, the seemingly opposite observations regarding residual versus advanced vestibular loss can be understood as part of a physiological continuum. Both partial vestibular reserve and near-complete but gradual preoperative deafferentation may facilitate central recalibration, supporting the concept of an optimal preoperative deafferentation window in which vestibular loss is sufficient to trigger adaptation but not abrupt enough to hinder compensation. This concept aligns with the hypothesis that gradual preoperative vestibular deafferentation allows earlier cerebellar recalibration, facilitating central compensation after surgery, as described in sequential neuroplasticity models of unilateral vestibular loss. 29 Consistent with this framework, Batuecas et al. 30 observed that patients with greater preoperative vestibular loss reflected in caloric weakness and low VOR gain tended to compensate more rapidly after surgery and improved PR scores. Together, these findings suggest that reaching a sufficient degree of preoperative deafferentation may help optimize central adaptive mechanisms.
Additionally, both preoperative vestibular rehabilitation and ITG-mediated vestibular deafferentation have been proposed to facilitate early postoperative mobility in patients undergoing vestibular schwannoma surgery. 31
Research priorities should focus on designing adequately powered randomized controlled trials to isolate intervention effects and minimize bias. The methodological appraisal using the JBI critical appraisal checklists and the RoB 2 tool informed our interpretation. Most studies presented moderate-to-high risk of bias, but directionally consistent trends were observed across designs, suggesting potential benefits that should be interpreted with caution. Standardized outcome measures particularly for functional balance, dizziness impact, and mobility are needed to enable meaningful comparisons across studies. Finally, integrating biomarkers of vestibular reserve (VOR gain, PR Score) may refine patient selection and improve the precision of prehabilitation strategies.
In summary, preoperative interventions may facilitate vestibular compensation after vestibular schwannoma surgery, but their benefits depend on careful patient selection, integration with postoperative care, and alignment with individual physiological profiles. A one-size-fits-all approach is unlikely to succeed.
Conclusions
Preoperative interventions such as ITG and structured vestibular rehabilitation may support postoperative recovery in patients undergoing vestibular schwannoma surgery. Current evidence highlights the importance of individualized patient selection based on residual vestibular function, as quantified by tools such as vHIT gain and PR Score, which may help identify candidates most likely to benefit from prehabilitation and guide the optimal timing and modality of intervention. Overall, the evidence suggests that prehabilitation should be viewed as an integrative and individualized strategy rather than a single standardized intervention, dependent on the extent of vestibular loss, timing, supervision of rehabilitation, and patient-specific physiological profiles. Future research should prioritize the development of shared methodological frameworks that allow reliable comparison between studies while retaining sufficient flexibility to adapt interventions to individual vestibular function and clinical context.
Supplemental material
Supplemental material - Preoperative interventions to improve clinical results in patients with vestibular schwannoma: A systematic review
Supplemental material for Preoperative interventions to improve clinical results in patients with vestibular schwannoma: A systematic review by Albert Torrents Torrero, Àngela Callejo Castillo, David Parés Martínez and Ma José Durà Mata in Journal of Vestibular Research
Footnotes
Acknowledgements
The author thanks colleagues from the Comissió de Fisioteràpia Vestibular of the Col·legi de Fisioterapeutes de Catalunya (CFC) and from the Sociedad Española de Rehabilitación Vestibular (SERVE) for professional exchange and inspiration. This work was carried out within the framework of the Doctorate in Surgery and Morphological Sciences of the Universitat Autònoma de Barcelona (UAB).
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Disclaimer
During the preparation of this work, the author used the support of NotebookLM, and ChatGPT5 assistant in order to contrast results previously extracted by the author. After using this tool/service, the authors reviewed and edited the content as necessary and take full responsibility for the content of the publication.
Supplemental material
Supplemental material is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
