A case series shows independent vestibular labyrinthine function after major surgical trauma to the human cochlea.

Background: The receptors for hearing and balance are housed together in the labyrinth of the inner ear and share the same fluids. Surgical damage to either receptor system was widely believed to cause certain permanent loss of the receptor function of the other. That principle, however, has been called into question because there have been anecdotal reports in individual patients of at least partial preservation of cochlear function after major surgical damage to the vestibular division and vice versa.
Methods: We performed specific objective vestibular function tests before and after surgical trauma (partial or subtotal cochlear removal) for treatment of intracochlear tumors in 27 consecutive patients in a tertiary referral center. Vestibular function was assessed by calorics (low-frequency response of the lateral semicircular canal), vestibulo-ocular reflex by video head impulse test (vHIT) of the three semicircular canals, cervical and ocular vestibular evoked myogenic potentials (cVEMP, saccule and oVEMP, utricle). Preoperative and postoperative distributions were compared with paired t-tests.
Results: Here we show that there was no significant difference between pre- and post-operative measures for all tests of the five vestibular organs, and that after major surgical cochlear trauma, the vestibular receptors continue to function independently. Conclusions: These surprising observations have important implications for our understanding of the function and the surgery of the peripheral auditory and vestibular system in general and open up new possibilities for the development, construction and evaluation of neural interfaces for electrical or optical stimulation of the peripheral auditory and vestibular nervous system.

Introduction

The perception of balance and spatial orientation is achieved and maintained by a complex set of sensorimotor control systems including input from the vestibular receptors in the inner ear. Diseases or trauma with disturbances of vestibular function result in dizziness and vertigo and exhibit a substantial impact on daily life in humans, especially for bilateral vestibular hypofunction[1]. The receptors for both the sense of hearing and balance—the cochlear and vestibular sensory systems—are housed together in the membranous labyrinth of the inner ear and share the same fluids with a tightly regulated homeostasis. The fluid bathing the receptors of both systems, endolymph, is crucial for normal receptor function and the major generation of this fluid takes place in the stria vascularis of the cochlea.

Surgical approaches to each of these systems have adhered to a principle, that damage to the membranous labyrinth of either system will cause almost certain permanent loss of the receptor function of both systems. Although the overall clinical effect of minimally invasive surgical approaches to the cochlea (like a cochleostomy during cochlear implantation, CI) on the vestibular function is considered to be non-significant[2], CI may lead to vestibular dysfunction and dizziness in some patients[2-5]. Therefore, any inner ear surgery generally aims on ‘atraumatic’ surgical techniques. That principle, however, has been called into question because there have been anecdotal reports in individual patients of at least partial preservation of cochlear function after major surgical damage to the vestibular division and vice versa[6-13].

Here we report for the first time (to the best of our knowledge) in a large case series, that after major trauma to the cochlea—cochlear removal to treat intracochlear schwannoma, removing the membranous labyrinth of the cochlea—the vestibular receptors continue to function normally, as shown by specific, objective tests of peripheral vestibular function before and after surgery.

These results demonstrate that under particular conditions, the cochlear and vestibular sensory systems can function independently, which has important implications for inner ear surgical procedures, and the development of neural interfaces of sensory prostheses for the auditory system and gene and cell-based inner ear therapies.

Methods

Study design, setting, and participants

The study comprises a large retrospective and prospective monocentric case series analyzing results from a tertiary (university) referral center. Patients included all consecutive patients with surgical resection of intracochlear schwannomas, most of them with hearing rehabilitation with a cochlear implant (CI).

Surgery

Surgical tumor removal was achieved through partial or subtotal cochleoectomy as described in detail elsewhere[14]. In short: The tumor was removed through a retroauricular-transmeatal, microscopic surgical approach. The cochlear capsule was opened exposing the tumor anterior to the RW, followed by drilling along the basal turn. A bony arch was left at the round window to later secure the electrode carrier (Fig. 1). The second turn was opened anterior to the oval window. In order to gain access to tumor parts located medially to the modiolus, the apical parts of the modiolus were removed preserving as much as possible of the modiolus, which contains the spiral ganglion cells in Rosenthal’s canal. During surgery, the base of scala tympani was blocked towards the hook region and the base of scala vestibuli and scala media with soft tissue. If patients received a CI, a device with a precurved, perimodiolar electrode carrier (CI512 or CI612; Nucleus CI, Cochlear, Sydney, Australia) was used. The cochlear defect was closed with an autologous cartilage/perichondrium compound transplant and bone pâté.

Fig. 1

Intracochlear tumor and situs after subtotal cochlear removal.

a Axial MRI (T1-w with Gd- enhancement) showing an intracochlear schwannoma (→) in the right cochlea. Although being a very rare disease, we can gain interesting insights into inner ear (patho)physiology from its diagnostics and treatment. MRI: Dr. Georg Eisele, Radiologisches Zentrum, Wangen, Germany (with permission). b Intraoperative endoscopic view after tumor resection through subtotal cochleoectomy and before cochlear implantation and defect reconstruction. During surgery, the bony arch (dashed line) of the round window is preserved for securing the electrode[14]. The Ductus reuniens courses along that bony arch of the round window and in the depth of the hook region connects the cochlear duct with the saccule in the vestibule (Fig. 2),[33]. Ch. t. chorda tympani, Gd Gadolinium, L left, M modiolus (2nd cochlear turn), MH malleus handle, OSL osseus spiral lamina, R right, RW round window, S stapes, ST scala tympani (basal cochlear turn), TT tensor tympani muscle, VII facial nerve, w- weighted.

Fig. 2

Schematic illustration of the inner ear with the five vestibular receptors and the cochlea and results of functional tests before and after surgery.

Postoperative otolith receptor function amplitudes and latencies are shown for utricle (a, b) and saccule (c, d) with data points for individual patients, means and 95% confidence intervals. Pre- and postoperative impulse responses of the anterior and posterior semicircular canals (vHIT) are shown on (e) and (g) with each line showing the data for one patient. The means and 95% confidence intervals are shown adjacent to the data points. The gray areas in the vHIT plots show abnormal results. Fast frequency (vHIT) and low-frequency (caloric) response of the lateral semicircular canal (h, i) and spontaneous nystagmus (j) are shown. ‘Ipsilateral’ refers to the tumor-affected side and ‘contralateral’ to the other (healthy) side. Word recognition scores (WRS) for monosyllables at 65 dB SPL with the cochlear implant compared to maximum word recognitions scores before surgery (WRSmax) are shown on (f). The location of the surgical blockage of the vestibular from the auditory system is indicated as a red bar. All statistical comparisons were made with paired two-tailed t-tests. Error bars in (a–i) show the 95% confidence interval. n number of participants, ns not significant, preop preoperative measurements, postop postoperative measurements, SPL: sound pressure level, VEMP cervical (c) or ocular (o) vestibular-evoked myogenic potentials. vHIT video head impulse test. Schematic illustration adapted from Spalteholz 1920[34].

Measurements of peripheral vestibular and auditory function

Prior to the surgery (<6 months) and during 12 months after surgery, the function of each vestibular sensory region was objectively assessed by specific, objective peripheral vestibular tests: Calorics (low-frequency response of the lateral semicircular canal), vestibulo-ocular reflex (VOR) by video head impulse test (vHIT) (manually induced head impulse response of lateral, anterior and posterior semicircular canals), cervical vestibular-evoked myogenic potentials (cVEMP, saccule), and ocular vestibular-evoked myogenic potentials (oVEMP, utricle). The function of the auditory system was assessed as speech recognition with the CI, if applicable.

Spontaneous nystagmus and calorics

Videonystagmography was performed using a Hortmann Vestlab videonystagmography system and the Vestlab OS software (GN Otometrics, Taastrup, Denmark). Over a recording period of 30 s, eye movements larger than 0.3°/s were identified and considered as spontaneous nystagmus (SPN) and counted. Caloric stimulation of both external ear canals was maintained by cold (30 °C) and warm (44 °C) water volumes of 75 ml within time periods of 30 s. Low-frequency responses of the lateral semicircular canal (SCC) with horizontal and vertical eye movements were recorded for 60 s. Eye movements larger than 0.75°/s were identified and considered as nystagmus. Slow-phase velocity (SPV) of all eye movements after ipsilateral and contralateral caloric irrigation was automatically calculated. Maximum absolute SPV values of the ipsilateral (SPVipsi) and contralateral (SPVcontra) sides were used to calculate the vestibular response relative to the side affected by the tumor with the formula: (SPVipsi,cold + SPVipsi,warm – SPVcontra,cold – SPVcontra,warm)/(SPVipsi,cold + SPVipsi,warm + SPVcontra, cold + SPVcontra,warm). Negative values indicated a reduced response on the ipsilateral side, and positive values indicated a reduced response on the contralateral side. Results below –25% and above 25% were considered abnormal.

Vestibulo-ocular reflex

VOR recordings were performed to assess impulse responses of all SCCs using a video head-impulse test (vHIT) system (GN Otometrics, Taastrup, Denmark). After small (10–20°) and fast unpredictable passive head turns aligned to the SCC orientations (lateral, left anterior-right posterior (LARP), and right anterior-left posterior (RALP) planes), the vestibulo-ocular response (VOR gain) was measured together with overt and covert saccades which were identified after recording by a 250-Hz high-speed eye camera that also measured head movements. Absolute mean VOR gain between eye and head movement of the affected and non-affected sides was measured.

Vestibular-evoked myogenic potentials

Otolith function was assessed by measuring cervical (cVEMP) and ocular (oVEMP) vestibular-evoked myogenic potentials using the Eclipse Platform (Interacoustics, Copenhagen, Denmark) that stimulated monaurally with tone bursts (1 or 2 cycles plateau; no rise/fall; 8 Hz stimulus repetition rate). For air-conducted VEMPs, insert earphones were used with tone bursts of 500 Hz at a stimulation level of 100 dB nHL. For bone-conducted VEMPs either a B81 transducer (Radioear, New Eagle, USA) stimulating with 500 Hz at 70 dB nHL or the recently developed B250 transducer (Chalmers, Gothenburg, Sweden) stimulating with 250 Hz at 80 dB nHL were used[15,16]. Preoperatively, only air-conducted cVEMPs and oVEMPs were measured because bone conduction was not available in most cases. Postoperative cVEMPs and oVEMPs were always measured to bone-conducted stimulation, because the incus was removed during surgical tumor removal (Fig. 1b), resulting in loss of air conduction. In addition, air-conducted VEMP testing in CI patients bears the risk of false-negative results due to (inner ear) conductive loss, and response rates to bone-conducted stimulation have been reported to be higher compared to air-conducted stimulation[17]. Due to these methodological differences, only the postoperative VEMPs were analyzed, and amplitudes and latencies of the affected ipsilateral side were compared to the contralateral side as a reference.

cVEMPs were measured using self-adhesive Neuroline 720 surface electrodes (Ambu A/S, Ballerup, Denmark) placed on the upper half of the ipsilateral sternocleidomastoid muscles, a reference electrode on the sternum, and the ground electrode on the forehead. In an upright sitting position, patients were instructed to rotate their heads toward the non-stimulated ear. The electromyogram (EMG) was measured in a –20 to 80 ms window relative to the onset of the stimulus and bandpass filtered between 10 and 1000 Hz. Visual feedback was given to the patient to maintain constant muscle tension. The first positive–negative peak (p13–n23) of the averaged EMG was defined as the cVEMP amplitude and normalized by the mean root-mean-square EMG level. For every presentation, at least 200 stimuli were averaged. For oVEMP recordings, electrode pairs were placed as bipolar montage on the infra-orbital ridge 1 cm below the lower eyelid contralateral to the stimulated side and about 2 cm caudal to the first electrode with the ground electrode placed on the forehead. The patients were asked to look maximally upwards. The first negative-positive peak (n10–p15) of the averaged EMG was defined as the oVEMP amplitude. For every presentation, at least 100 stimuli were averaged. For cVEMPs and oVEMPs, amplitudes and latencies between the tumor-affected side and the non-affected side were analyzed.

Word recognition with CI

For patients receiving a CI, hearing performance was measured at first fit, and around 1, 3, 6, and 12 months after surgery. Word recognition scores were measured at a sound pressure level of 65 dB SPL (WRS65) by the Freiburg monosyllables and the Freiburg numbers test after blocking and masking the contralateral ear with broad band noise, if applicable.

Statistics and reproducibility

If data could not be obtained in the defined postoperative test period of 12 months, they were completed by later recordings or treated as missing data. The respective n was reported. Preoperative and postoperative distributions were compared with paired t-tests. Bonferroni correction was applied if multiple comparisons were performed. The significance level was set to 5%. For all statistical calculations, GraphPad Prism software (Version 8, Graphpad Software, San Diego, CA, USA) was used.

Ethical approval

The study was approved by the responsible ethics committee (approval numbers 2019-026 and 2019-050). All relevant ethical regulations were followed, informed consent was obtained from all participants.

Table. 1

Functional test results before and after partial or subtotal cochlear removal.

Measurement	N	Mean	Lower 95% CI of mean	Upper 95% CI of mean	N	Mean	Lower 95% CI of mean	Upper 95% CI of mean
vHIT gain	Preoperative				Postoperative
Lateral semicircular canal	24	1,00	0,93	1,10	24	1	0,88	1,1
Posterior semicircular canal	23	0,75	0,66	0,84	23	0,83	0,73	0,93
Anterior semicircular canal	24	0,88	0,81	0,95	24	0,83	0,74	0,92
Videonystagmography
Spontaneous nystagmus re. affected side (°/s)	20	0,08	−0,28	0,43	20	−0,03	−0,39	0,34
Caloric irrigation (% response re. affected side)	20	−17	−26	−7,9	20	−17	−34	−1
cVEMP
p13-n23 Amplitude ipsilateral (EMG normalized)					18	1.8	1.1	2.6
p13 Latency ipsilateral (ms)					18	15	14	16
n23 Latency ipsilateral (ms)					18	25	24	25
p13-n23 Amplitude contralateral (EMG normalized)					22	1.6	1.1	2.2
p13 Latency contralateral (ms)					22	14	13	15
n23 Latency contralateral (ms)					22	25	24	25
oVEMP
n10-p15 Amplitude ipsilateral (µV)					19	8.4	5.3	12
n10 Latency ipsilateral					19	12	11	12
p15 Latency ipsilateral					19	16	15	17
n10-p15 Amplitude contralateral (µV)					18	6.1	3.6	8.7
n10 Latency contralateral					18	12	11	13
p15 Latency contralateral					18	16	15	17
Speech perception in quiet
Word Recognition Score (WRS, %), maximum WRS preoperative and at 65 dB SPL postoperative	24	10.2	0.4	20.0	12a	72.5	60.5	84.5

aat: 12 months