Does vestibular function correlate with objective MRI findings after vestibular schwannoma surgery?| ACTA Otorhinolaryngologica Italica

Abstract

Objective. Vestibular schwannoma surgery leads to acute unilateral vestibular loss. In some patients, however, the process of post-operatively initiated central compensation proceeds more rapidly than in others. This study aimed to evaluate post-operative vestibular function and correlate it with morphological findings of MRI scans.
Methods. The study included 29 patients who underwent surgery for vestibular schwannoma. Vestibular function was analysed post-operatively by video head impulse test (vHIT). Subjective symptoms were evaluated using validated questionnaires. All patients underwent MRI 3 months post-operatively, and the presence of the facial and vestibulocochlear nerves in the internal auditory canal was assessed.
Results. The vestibulo-ocular reflex gain measured by the vHIT correlated positively with audiological findings. Subjective perception of vestibular disorder did not correlate with objectively measured vestibular impairment or with MRI findings.
Conclusions. After the resection of vestibular schwannoma, some patients may still have preserved vestibular function as measured by vHIT. The preserved function does not correlate with subjective symptoms. Patients with partially deteriorated vestibular function showed lower sensitivity to combined stimuli.

Introduction

Vestibular schwannoma constitutes 80-90% of all cerebellopontine angle tumours ¹. It often presents with hearing impairment, tinnitus, dizziness, or vertigo. Large tumours as well as smaller but growing or symptomatic tumours are indicated for surgical resection ². Vertigo is a dominant symptom after vestibular schwannoma resection. Compensation of the post-operative vestibular disorder is a multifactorial process ³. Age, tumour size and comorbidities are the factors that influence the compensation process. Other factors that affect vestibular compensation, such as mental state, inclination to anxiety and depressive states, are also being studied ^4-6. Subjective symptoms and mental state are usually measured with validated questionnaires. As the tumour originates from the vestibular portion of the vestibulocochlear nerve, tumour resection results in the disruption of the vestibular nerve. Transection of the vestibular nerve can be partial or radical. Several objective tests are available to measure residual vestibular function, such as the caloric test, video head impulse test (vHIT), rotational test and vestibular evoked myogenic potential (VEMP) test ⁷. In the present study, vHIT was used to objectively measure vestibular function. vHIT has recently become a popular test to determine vestibular function ^8,9. It uses high-frequency impulse stimulation corresponding to natural head movements, instead of non-physiological low frequency stimulation used in the caloric test. Both methods are complementary to each other and examine slightly different phenomena. Even though all patients with radical resection of vestibular schwannoma undergo concurrent vestibular neurectomy, an unusually high vHIT gain was observed in some of these patients as reported in a previous study ¹⁰. Therefore, in the present study, we aimed to describe changes in vestibular function in more detail and correlate them with morphological changes in the internal auditory canal (IAC) observed on post-operative MRI scans.

Methods

Patients

The study included 29 patients with unilateral vestibular schwannoma, aged 25-69 years, who had undergone surgery through a retrosigmoid approach for tumour resection. Complete surgical removal of the tumour was achieved in all patients (Fig. 1).

Vestibular function and hearing tests

All patients were examined clinically before and after surgery. vHIT (ICS Impulse, Otometrics, Denmark) was used to determine the post-operative function of each semicircular canal. vHIT was performed between the 7th and 10th post-operative day. Two patients were excluded since they did not undergo the examination in this period. The gain of the vestibulo-ocular reflex (VOR gain) for each semicircular canal was calculated as the ratio of the areas under the eye velocity curve and the head velocity curve from the start to the end of the head impulse ¹¹. The pattern of saccades was studied, and the saccades were divided into scattered and clustered patterns; the clustered pattern was subdivided into covert and overt saccades ¹². Pure tone average (PTA) was counted from hearing levels in dB at 0.5, 1, 2 and 4 kHz pre-operatively and post-operatively (3 months after the surgery). MRI findings of the vestibulocochlear nerve were correlated with the function of the semicircular canal of each side (operated and intact one). All patients completed two questionnaires before surgery and 3 months after surgery to evaluate vertigo symptoms and their impact on quality of life based on subjective perception. The first was the Dizziness Handicap Inventory (DHI) questionnaire. This questionnaire consists of 25 items, and the score ranges from 0 to +100. The higher the score, the more severe the handicap it corresponds to ¹³.

The second questionnaire ^5,6 (Tab. I) was developed by neurotologists from the Department of Otorhinolaryngology and Head and Neck Surgery of the 1st Faculty of Medicine, Charles University in Prague, Faculty Hospital Motol. Scores for each question range from 1 to 4. A higher score corresponds to more severe handicap.

MRI

The post-operative evaluation of the possible presence of residual nerve structures within the IAC was performed on an MRI machine (1.5 T Magnetom Avanto, Siemens, Erlangen, Germany) using a 3D-T2 space sequence (voxel size: 0.6 × 0.6 × 0.6 mm). The original dataset obtained in the transversal plane and multiplanar reconstruction (MPR) in the oblique sagittal plane perpendicular to the long axis of the IAC were evaluated for the presence of the facial nerve (in the anterior upper quadrant), cochlear nerve (in the anterior lower quadrat), superior vestibular nerve (in the posterior upper quadrant) and inferior vestibular nerve (in the posterior lower quadrant). MRI findings were not evaluated in two patients who underwent post-operative MRI scans at a different hospital, where special sequences for reconstruction of the IAC were missing.

Statistical analysis

Statistical analyses were performed using SPSS version 25 (SPSS Inc., Chicago, IL, USA). For processing descriptive data, we used standard measures of central tendency (e.g., mean and median), dispersion (e.g., variance, standard deviation and percentiles) and data shape statistics (kurtosis and skewness). The research sample was divided into three subgroups based on the values of VOR gain in the horizontal canal as measured by the vHIT on the operated side: 1st group: ≤ 0.40; 2nd group: 0.41 to 0.69; 3rd group: > 0.70. The between-group differences (both pre-operative and post-operative) were then examined using Kruskal-Wallis one-way analysis of variance (ANOVA), followed by post hoc comparison with Mann-Whitney U test. The within-group differences of repeated measurements (e.g., pre-operative vs post-operative values) were analysed by a Friedman test (an alternative to repeated-measures ANOVA), followed by a series of Wilcoxon signed-rank tests. Given the relatively small sample size for both between- and within-group comparisons, we used the Monte Carlo resampling procedure with n = 10,000 samples, which compensate for tied values and do not depend on asymptotic approximations for p-values ¹⁴. The level of statistical significance was set at α = 0.05. Pearson’s correlation coefficient was used to determine bivariate associations between the variables.

Results

Responses to the questionnaires and objective findings were evaluated. Changes in subjective perception of balance disorder were evaluated through responses to the questionnaires, and statistical analysis was used to compare the survey results with vHIT results. All patients were tested clinically before the surgery. None of the patients had positive head impulse test, so we assume that the gain of vHIT would be normal (0.7-1.1) in the study population ¹⁵. vHIT was performed between the 7th and 10th post-operative day and showed interesting results. Patients were divided into three groups according to the value of VOR gain in the horizontal canal on the operated side (Fig. 1). The first group had VOR gain less than 0.4 (n = 9), the second group had partially deteriorated vestibular function with VOR gain between 0.41 and 0.69 (n = 10), and the third group had VOR gain higher than 0.7 (n = 6). In most patients in group 1, a VOR gain of less than 0.4 was observed in all semicircular canals on the operated side, which corresponds with acute peripheral vestibular loss after vestibular neurectomy. Although the surgeon reported the complete resection of the vestibular nerve in all patients from this group, the radiologist described the presence of a portion of the vestibular nerve on MRI in four patients. Even after declared vestibular neurectomy the gain higher than 0.7 was observed in group 3. Similar results were reported in a previous study ¹⁰. According to the surgeon, complete resection was achieved in five patients, and in one patient the resection was incomplete and the vestibular nerves were preserved. None of the branches of the vestibular nerve were discernible on MRI in one patient in this group. Figure 1 shows the detailed results of MRI. An example of MRI finding in the IAC with all nerves preserved is shown in Figure 2. Figure 3 shows a case with only the facial nerve preserved. Group 2 patients had a borderline VOR gain between 0.41 and 0.69 (Fig. 4). Of the 10 patients in this group, the superior vestibular nerve was preserved in two patients, and complete resection was achieved in the remaining patients according to the surgeon. The superior vestibular nerve was observed on MRI in three patients, while the inferior vestibular nerve was observed in one patient. This group showed different functional characteristics than the other two groups. Comparison between VOR gain of semicircular canals on the operated and healthy sides is shown in Figure 4. Subgroups were classified according to VOR gain of the horizontal canal on the operated side. The values for vertical canals were similar to those for the horizontal canal for each subgroup. The lateral semicircular canal is commonly examined because the measurement is less affected by artefacts as compared to that for the vertical canals.

Comparison of the three groups using other observed variables revealed that post-operative hearing (Pure Tone Average, PTA) differed among the patients, which was similar to the difference in VOR gain in the horizontal canal on the operated side as measured by the vHIT. Patients with the highest post-operative gain also showed better post-operative PTA. Hearing level was post-operatively good in one patient (26 dB), two patients had serviceable hearing and the remaining patients had non-serviceable hearing according to the Gardner-Robertson scale. Post-operative PTA correlated significantly with tumour grade (p = 0.001) and size (p = 0.049) (Fig. 5).

Statistical analysis was performed to observe the correlation between the DHI result, VOR gain and post-operative morphological findings on MRI scans. Our findings suggest that the post-operative result of the DHI is not influenced by the presence of any specific nerve or combination of preserved nerves. The presence of the inferior or superior vestibular nerve on MRI or higher VOR gain measured by the vHIT in the post-operative stage did not predict a better result in the DHI. However, the DHI result had a positive correlation with better PTA level.

The study group showed a significant difference in response to question seven of the questionnaire used in our department, which evaluates the response to multisensorial stimuli (combined audio-visual stimuli). While patients in groups 1 and 3 showed deterioration in their response to multisensorial stimuli post-operatively, mild improvement was observed in patients in group 2 compared to their pre-operative status (Fig. 6).

Discussion

During resection of vestibular schwannoma through the retrosigmoid suboccipital approach, the surgeon’s aim is to preserve the nerves in the internal auditory meatus. It is usually impossible to preserve the vestibular portion of the vestibulocochlear nerve as the tumour originates from one of the vestibular nerve branches. This approach enables to preserve the cochlear portion of the vestibulocochlear nerve and thus to preserve hearing. The presence of an anatomically uninterrupted nerve can be post-operatively evaluated with MRI multiplanar reconstruction in a plane perpendicular to the internal acoustic meatus. The radiologist must be informed about the approach used for dural closure in the region of the IAC (if a muscle or an artificial material was used for closure). To the best of our knowledge, the present study is the first to investigate the correlation between MRI findings of patients with vestibular schwannoma and their vHIT results post-operatively.

Normal or only a mild decrease in VOR gain measured by the vHIT post-operatively was observed in many patients. This finding corresponds to preserved vestibular function post-operatively. However, our study shows that this preservation of vestibular function contrasted with MRI findings in some patients in whom complete resection of the vestibular nerve was performed. In group 3 patients with high VOR gain, only facial and cochlear nerves were preserved in two patients, while in group 2, the nerves were preserved in six of 10 patients. A non-zero VOR gain was, however, noted in complete unilateral peripheral vestibular loss. These non-zero values in complete unilateral peripheral loss are due to the contralateral disinhibition pathway ^16,17, but the values of gain are subnormal. In our study, we observed patients with borderline and normal gain, which is in contrast with the finding of unidentified vestibular nerve on MRI. Based on this finding, it can be hypothesised that some vestibular nerve fibres may be conducted in the cochlear nerve, or the vestibular pathway might be partially dominated by the cochlear nerve. These fibres might be responsible for the preservation of vestibular function ¹⁸. Another explanation of this phenomenon is that the vHIT itself is not sensitive enough to detect impaired function of individual semicircular canals. Even in a non-functioning semicircular canal, residual VOR gain can be measured ¹⁹.

Comparison of subjective symptoms evaluated by the questionnaires showed no significant correlation with possibly preserved nerves or combination of nerves in the IAC. No significant difference was observed in response to questionnaires between patients with resected and those with intact vestibular nerve including its branches. According to a previous study, instability or dizziness perceived subjectively by a patient does not seem to correlate with objective findings measured by vHIT ²⁰. This shows that multimodal sensory loading is particularly demanding for patients with vestibular peripheral loss. Patients with partial preservation of the VOR response were found to compensate this handicap on follow-up examination, but not those with VOR loss or good VOR response. This finding is not easy to explain; moreover, it corresponds neither to the DHI nor to other similar questions (e.g., question 6) of our questionnaire results. We therefore hypothesise that the central sensory fusion of vestibular, visual and proprioceptor inputs results in sensorial reweighting differentiated according to the final vestibular deficit ²¹.

Our data also did not show any correlation with MRI findings. Objectively measured vestibular hypofunction did not correspond to MRI findings. It may be also because MRI is not sensitive enough to display slight changes in the internal auditory canal. A higher level of visual dependency is often observed during the compensation process in these patients. The process of compensation of a vestibular lesion is probably more complicated, and is affected centrally by other brainstem centres ²². According to some previous studies, it seems that responsiveness to mental status may affect the compensation process ^5,23. The overall result of compensation may be affected by anxiety, insufficient coping behaviour, or adaptation disorder. The precise mechanisms need to be studied further.

Theoretical framework for the change in the sensitivity to various sensory inputs is laid by the sensory weighting model. Multiple estimates of head and eye movements in space are calculated by central nervous system using internal models of sensory and physical relationships. Multiple sensory inputs are complemented with internal models to yield accurate estimate of gravity, head and eye movements ²⁴. In experiments on healthy volunteers, rapid, context dependent, reweighting was observed between visual and vestibular inputs ²¹. We hypothesise, that the abovementioned processes can explain increased weight of visual stimulation in our data.

During assessment of vestibular function, we examined only the function of the semicircular canals by vHIT. Otolithic function can be tested using the VEMP test; however, this test was not included in our study. We measured the function of semicircular canals by vHIT only once in post-operative period, and this is a possible limitation of the study. All patients in the study had a normal vHIT result pre-operatively, so we assume that pre-operative gain would not be significantly alternated. According to the results in patients with vestibular schwannoma, we assume that the VOR gain measured by vHIT does not change significantly when measured on 7th-10th post-operative day or 3 months post-operatively. The results of the present study were compared with those of the study of Aw ⁷, in which post-operative MRI findings were evaluated and the preserved vestibular nerve observed on MRI was found to lead to poor outcome after vestibular neurectomy. However, in our study, the nerve was affected pre-operatively by the growth of the tumour, which is in contrast with the condition of the pre-operative status of a healthy nerve and impairment of the labyrinth in patients with Meniere’s disease in the study of Aw. This is the possible reason for the difference in the results of these two studies.

In agreement with a previous study, it was confirmed that patients are more sensitive to optokinetic stimulation and other audio-visual stimulation post-operatively. Interestingly, group 2 patients with borderline gain between 0.41 and 0.69 showed significantly lower sensitivity to combined audio-visual stimuli than the remaining patients. Patients with normal VOR gain and those with VOR lower than 0.4 showed a deterioration in this parameter. Similar results were observed in a previous study where patients pretreated with gentamicin were less sensitive to optokinetic stimulation ¹⁰. It seems that partial impairment of vestibular function leads to a greater flexibility in a patient. This might be because the process of compensation starts earlier and proceeds more quickly in patients with partial impairment of vestibular function than in patients with a complete lesion.

Conclusions

The results of the present study confirmed that even after the resection of vestibular schwannoma, patients may have preserved vestibular function as measured by the vHIT. This preservation of function did not correlate with subjective symptoms post-operatively. Vestibular function was preserved even in cases where none of the branches of the vestibular nerve were identified by MRI of the IAC. MRI may not be sensitive enough to display fine post-operative changes in the IAC. We hypothesise that some vestibular fibres may pass through the cochlear nerve or are dispersed within the IAC and not discernible by MRI. Patients with partially deteriorated vestibular function show lower visual dependency to multisensory stimuli. Neither post-operative vHIT result nor post-operative MRI finding predict functional vestibular performance status.

Conflict of interest statement

The authors declare no conflict of interest.

Funding

This work was supported by the Ministry of Health of the Czech Republic (Grant No. NU20-08-00311) and GAUK No.192121.

Author contributions

Z_Č, R_Č: conceptualization; ZB: methodology; SB, VK, VB, VS, PH, KK, ZF: investigation; MK: statistical analysis; JL: imaging methods analysis; ZB, MB: writing - original draft preparation; ZB, Z_Č, R_Č, EM: writing - review and editing; Z_Č: funding acquisition; Z_Č: supervision.

Ethical consideration

The study was approved by the Ethics Committee for Multi-Centric Clinical Trials of the University Hospital Motol. (protocol number EK-297/20).

The research was conducted ethically, with all study procedures being performed in accordance with the requirements of the World Medical Association’s Declaration of Helsinki.

Written informed consent was obtained from each participant/patient for study participation and data publication.

Figures and tables

Figure 1.Overview of patient characteristics.

Figure 2.MRI scan of the internal auditory canal on the operated side. Facial (empty arrow), cochlear nerve (full arrow), superior vestibular nerve (empty arrowhead), inferior vestibular nerve (full arrowhead), still preserved within internal auditory canal, have blurry contours corresponding with post-operative changes: 3D T2/space.

Figure 3.MRI scan of the internal auditory canal on the operated side. Usual configuration of the nerves inside the right-sided internal auditory canal is observed. Of note is the absence of the vestibulocochlear nerve and deformation of the facial nerve (arrow) within the post-operatively changed left-sided internal auditory canal: 3D T2/space.

Figure 4.Post-operative vHIT results. Values of VOR gain in the anterior (A), lateral (H) and posterior (P) semicircular canals measured by the vHIT post-operatively on the operated and healthy sides.

Figure 5.Change in pure tone average before and after the operation. The study group was divided in three subgroups (1, 2 and 3) according to the value of VOR gain in the horizontal canal measured by the vHIT on the operated side.

Figure 6.Response to question 7: Do you have instability with/or does a greater amount of auditory and visual sensations bother you (e.g., shopping malls)? The graph shows the mean value of PTA in each subgroup before and after the operation. The study group was divided in three subgroups (1, 2 and 3) according to the value of VOR gain in the horizontal canal measured by the vHIT on the operated side. The graph shows the mean value for each subgroup before and after the operation.

References

1. Halliday J, Rutherford SA, McCabe MG. An update on the diagnosis and treatment of vestibular schwannoma. Expert Rev Neurother. 2018; 18:29-39. DOI
2. Betka J, Zvěřina E, Balogová Z. Complications of microsurgery of vestibular schwannoma. Biomed Res Int. 2014; 315952DOI
3. Herdman SJ, Clendaniel RA, Mattox DE. Vestibular adaptation exercises and recovery: acute stage after acoustic neuroma resection. Otolaryngol Head Neck Surg. 1995; 113:77-87. DOI
4. Cohen HS, Kimball KT, Jenkins HA. Factors affecting recovery after acoustic neuroma resection. Acta Otolaryngol. 2002; 122:841-850.
5. Balatkova Z, Cada Z, Hruba S. Assessment of visual sensation, psychiatric profile and quality of life following vestibular schwannoma surgery in patients prehabituated by chemical vestibular ablation. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub. 2020; 164:444-453. DOI
6. Čada Z, Balatková Z, Chovanec M. Vertigo perception and quality of life in patients after surgical treatment of vestibular schwannoma with pre-treatment pre-habituation by chemical vestibular ablation. Biomed Res Int. 2016; 6767216DOI
7. Aw ST, Magnussen JS, Todd MJ. MRI of the vestibular nerve after selective vestibular neurectomy. Acta Otolaryngol. 2006; 126:1053-1056. DOI
8. MacDougall HG, Weber KP, McGarvie LA. The video head impulse test: diagnostic accuracy in peripheral vestibulopathy. Neurology. 2009; 73:1134-1141. DOI
9. McGarvie LA, MacDougall HG, Halmagyi GM. The video head impulse test (vHIT) of semicircular canal function – age-dependent normative values of VOR gain in healthy subjects. Front Neurol. 2015; 6:154. DOI
10. Cerny R, Balatkova Z, Hruba S. Residual vestibular function after vestibular schwannoma surgery. Neurochirurgie. 2020; 66:80-84. DOI
11. Jacobsen CL, Abrahamsen ER, Skals RK. Is regression gain or instantaneous gain the most reliable and reproducible gain value when performing video head impulse testing of the lateral semicircular canals?. J Vestib Res. 2021; 31:151-162. DOI
12. Rey-Martinez J, Yanes J, Esteban J. The role of predictability in saccadic eye responses in the suppression head impulse test of horizontal semicircular canal function. Front Neurol. 2017; 8:536. DOI
13. Jacobson GP, Newman CW. The development of the dizziness handicap inventory. Arch Otolaryngol Head Neck Surg. 1990; 116:424-427. DOI
14. Berry KJ, Mielke PW. Exact and Monte carlo resampling procedures for the Wilcoxon-Mann-Whitney and Kruskal-Wallis tests. Percept Mot Skills. 2000; 91:749-754. DOI
15. Korda A, Carey JP, Zamaro E. How good are we in evaluating a bedside head impulse test?. Ear Hearing. 2020; 41:1747-1751. DOI
16. Halmagyi GM, Chen LK, MacDougall HG. The video head impulse test. Front Neurol. 2017; 8:23. DOI
17. Wagner AR, Schubert MC. Evidence a shared mechanism mediates ipsi- and contralesional compensatory saccades and gait after unilateral vestibular deafferentation. J Neurophysiol. 2020; 123:1486-1495. DOI
18. Ozdogmus O, Sezen O, Kubilay U. Connections between the facial, vestibular and cochlear nerve bundles within the internal auditory canal. J Anat. 2004; 205:65-75. DOI
19. Comacchio F, Castellucci A. Posterior semicircular canal ossification following acute vestibular loss mimicking inferior vestibular neuritis: a case report. Front Neurol. 2022; 13:1015555. DOI
20. Eravci FC, Yilmaz M, Ugur MB. The relationship of vestibulo-ocular reflex with self-reported dizziness handicap in patients with vestibular deafferentation. Ear Nose Throat J. 2021; 100:NP299-NP307. DOI
21. Hwang S, Agada P, Kiemel T. Dynamic reweighting of three modalities for sensor fusion. PLoS One. 2014; 9:E88132. DOI
22. Jacob RG, Redfern MS, Furman JM. Space and motion discomfort and abnormal balance control in patients with anxiety disorders. J Neurol Neurosurg Psychiatry. 2009; 80:74-78. DOI
23. Brown CS, Cooper MW, Peskoe SB. Associations of vestibular tests with penn acoustic neuroma quality of life scores after resection of vestibular schwannoma. Otol Neurotol. 2020; 41:E241-E249. DOI
24. Zupan LH, Merfeld DM, Darlot C. Using sensory weighting to model the influence of canal, otolith and visual cues on spatial orientation and eye movements. Exp Brain Res. 2011; 210:407-422. DOI