Vestibular Implantation Can Work Even After >20 Years of Bilateral Vestibular Hypofunction

Abstract

Objective:

To determine whether prosthetic stimulation delivered via a vestibular implant can elicit artificial sensation of head movement despite long (23 year) duration adult-onset ototoxic bilateral vestibular hypofunction (BVH).

Study Design:

Case Report

Setting:

Tertiary care center as part of a first-in-human clinical trial

Patients:

1

Interventions:

Unilateral vestibular implantation with an investigational multichannel vestibular implant in a 55-year-old male with a well-documented 23-year history of aminoglycoside-induced BVH.

Main Outcome Measures:

Electrically evoked vestibulo-ocular reflexes (eeVOR).

Results:

Vestibular implant stimulation can drive stimulus-aligned eeVOR and elicit a vestibular percept 23 years after the onset of bilateral vestibulopathy. Prosthetic stimulation targeting individual semicircular canals elicited eye movements that approximately aligned with each targeted canal’s axis. The magnitude of the eeVOR response increased with increasing stimulus current amplitude. Response alignment and magnitude were similar to those observed for implant recipients who underwent vestibular implantation <10 years after BVH onset. Responses were approximately stable over 18 months of continuous device use (24 hr/day except during sleep).

Conclusions:

Vestibular implantation and prosthetic electrical stimulation of semicircular canal afferent nerves can drive canal-specific eye movement responses >20 years after onset of ototoxic vestibular hypofunction.

Introduction:

Vestibular implants (VI) can drive vestibular reflexes by encoding head motion information and electrically stimulating the vestibular end organs. Current clinical trials are designed to study patients with bilateral vestibular hypofunction (BVH, also called bilateral vestibulopathy).¹ About 35% of adult-onset BVH cases are attributable to ototoxic hair cell injury, another ~15% are due to other known diseases of the vestibular labyrinth, ~50% of cases are idiopathic, and a minority are attributable to isolated central nervous system disorders.^2,3 Among idiopathic cases, many, and possibly most, are due to undiagnosed causes of labyrinthine dysfunction (such as genetic, rheumatologic or ischemic disease) that spare the vestibular nerve and central vestibular pathways. Therefore, a majority of adult-onset BVH cases should be amenable to VI stimulation.

Adult-onset BVH is typically a chronic disorder for which there is no cure. Standard-of-care practice involves supportive counseling, fall risk mitigation, cessation of vestibular suppressants, and vestibular rehabilitation therapy exercises. Results from first-in-human VI clinical trials are promising,^4–10 but many patients disabled by chronic BVH will have experienced symptoms for decades before presenting for VI candidacy evaluation. It is therefore important to establish whether it is possible to achieve effective VI stimulation decades after the onset of BVH. In cochlear implantation (CI), duration of deafness (DoD, defined as the delay between the diagnosis of severe to profound hearing loss and cochlear implantation) is an important determinant of hearing outcomes.^11–14 While most studies show improvements in hearing despite long DoD compared to preoperative testing, DoD longer than 10–45 years has been associated with worse outcome.^12–14 Considering similarities between cochlear and vestibular implantation and the underlying pathologies they are intended to treat; one might expect similar limitations in patient outcomes for VIs after long-duration BVH.

In this case report, we demonstrate that VI stimulation can drive electrically-evoked vestibulo-ocular reflex (eeVOR) responses and elicit vestibular perception in a patient with longstanding (>20 years) adult-onset BVH due (in this case) to aminoglycoside ototoxicity. This patient’s posture, gait, quality of life, and hearing outcomes were previously published; however, analysis of his eye motion data in response to electric vestibular stimulation is not reported in the literature.⁴

Clinical Capsule:

A 55-year-old man presented to the authors for evaluation for inclusion in a first-in-human clinical trial investigating a VI (ClinicalTrials.gov: NCT02725463). He reported a 23-year history of BVH after being treated with intravenous gentamicin in 1996 for a dental infection complicated by sternoclavicular joint septic arthritis. He described initially noticing vertigo three weeks into antibiotic therapy. The vertigo resolved, but he developed oscillopsia and chronic postural imbalance. Despite being an otherwise healthy 22-year-old at the time of onset, he described a prolonged recovery, including a four-month admission to a nursing facility, use of forearm crutches to steady his walking for years, and participating in vestibular physical therapy. He was ultimately diagnosed with BVH.

For over 20 years, he continued to experience persistently debilitating symptoms of BVH, including oscillopsia, postural instability, difficulty walking without falling when in dim lighting or on unsteady surfaces, and inability to carry on a discussion while walking (evidently due to the need to devote conscious effort to compensate for deficient vestibular reflexes that normally stabilize vision, posture and gait). A comprehensive evaluation of vestibular function was performed as part of the clinical trial to determine candidacy for VI surgery. Caloric nystagmography revealed bilateral horizontal canal hypofunction (sum maximal bithermal response of <10 ˚/s in each ear). He was also found to have minimal response to ice-water caloric stimulation, low or absent vestibular evoked myogenic potentials, and pathologically low VOR gains on 6-canal video head impulse testing and rotary chair testing.

Intervention, Methods and Results:

Comprehensive descriptions of the device (the MVI Multichannel Vestibular Implant System^™, provided by Labyrinth Devices, LLC, Baltimore MD), study design, and inclusion criteria were previously published.^4,6 Similar to a cochlear implant receiver/stimulator, the implanted part of the system includes an internal receiver/stimulator, an electrode array, an inductive coil link, and magnets to aid in fixation and alignment of the external component of the inductive link. The electrode array includes three subarrays (one for each semicircular canal), each containing three electrodes, and a stimulation reference electrode. The external component includes a power/control unit (PCU) and a head worn unit (HWU). The latter includes three fixation magnets, an inductive coil link, and a three-axis motion sensor.

Having met the clinical trial’s inclusion criteria, the patient underwent uneventful right-side VI implantation in September 2019. Three weeks after implantation, the clinical trial team activated the device. Eye movement data were recorded using a binocular 3-dimensional video-oculography (3D VOG) system during electric stimulation targeting the semicircular canals (Neurolign, Toronto, ON, Canada). Details of the electric stimulation paradigm, the 3D VOG system, and data analysis were previously described.⁶ The best stimulating electrode in each semicircular canal was selected for activation and long-term motion-modulated stimulation.¹⁵ Canal-axis-aligned components of 3D eeVOR responses were analyzed over a range of biphasic, charge-balanced 200μs/phase current pulse trains. Current pulse amplitude ranges were selected to include the level that induced a rotational motion percept around the axis of the target canal while minimizing motion percepts in other directions, facial twitching or tinnitus (signifying spurious activation of other vestibular nerve branches, the facial nerve, and the cochlear nerve, respectively). The mean axis of 3D eye movement responses was compared to the 3D axis of the targeted canal.⁶

At activation, stimulation from the device produced visible eye movements and a head motion percept. Even when the patient was instructed to hold his head still and fix his eyes on a brightly lit Earth-fixed target (which should suppress the vestibulo-ocular reflex), pitch movements of the HWU (and its motion sensor) elicited a pitch motion percept and prominent reflexive eye movements approximately aligned in axis and timing with the HWU motion (Video 1). After 18 months of continuous stimulation, electric stimulation of each of the semicircular canals elicited eye movement responses that roughly aligned in 3D with the anterior [RA], horizontal [RH], and posterior [RP] canals in the right ear that were being stimulated (Figure 1). The magnitude of eeVOR responses grew with increasing stimulus current amplitude. Response alignment and magnitude were comparable to those observed for patients who underwent vestibular implantation <10 years after BVH onset.⁶

Figure 1:

eeVOR responses remain stable after 18 mos. of continuous, motion-modulated stimulation. Peak excitatory half-cycle slow phase 3D eye velocity of the subject’s left and right eyes while stationary during stimulus modulation emulating 2Hz sinusoidal head rotation. Modulation depths of 5%–100% represent 20–400°/s head velocities elicited by stimulating the anterior (A), horizontal (B), and posterior (C) canals of the right ear. (D) Mean response axis for each canal aligns approximately with the intended canal as viewed in 3D.

He reported improvements in oscillopsia, postural stability, gait, and ability to exercise and stay active, but he also experienced a profound sensorineural hearing loss in the implanted ear. Overall, he reported that the benefits of improved balance outweighed the unilateral hearing loss, facilitating return to a more active lifestyle and yielding an overall improvement in health-related quality of life (Video 2).

Discussion:

This case demonstrates that human vestibular nerve afferents can remain sensitive to VI stimulation even 23-years after adult-onset ototoxic BVH. Additionally, canal-aligned reflexive eye movements remained stable after at least 18 months of continuous stimulation. Complementing the sparse other data published about eeVOR responses decades after the onset of BVH symptoms, this case supports the conclusion that individuals with long-duration BVH may benefit from VI.^4–7 Importantly, these responses are similar to those observed in patients implanted with this device <10 years after BVH onset.⁶

How responses to VI stimulation depend on the underlying cause of peripheral BVH is largely unknown. Since the primary target of VI stimulation is the distal aspects of vestibular afferent nerve branches, we expect that patients with BVH due to hair cell mechanoreception dysfunction will respond to electric stimulation. Some authors have suggested that etiology may play a significant role in the responsiveness of vestibular afferent neurons to electric stimulation. In three patients implanted with a cochleovestibular implant, prior labyrinthitis appeared to result in significantly smaller eeVOR responses compared to two patients with BVH from bilateral Meniere’s disease.⁵ A review of the literature shows that VIs can drive an eeVOR in patients with several BVH etiologies, including idiopathic, aminoglycoside ototoxicity, Meniere’s disease, trauma-related, and genetic (DFNA9).^4–7

The relatively small number of patients who have undergone VI surgery limits the ability to quantitatively define the effect of sensorineural vestibular loss duration on responsiveness to VI stimulation. In contrast, effects of DoD on CI outcomes is a topic of great interest among CI clinicians and researchers.^{11–14,16–18} Several studies have reported an association between long DoD and poorer hearing outcomes after CI,^11–14 while others have not found the same associations.^16–18 When considering the association between DoD and CI outcomes, the benefits from the device are often felt to outweigh the risks, even if the benefit is not as great as if the CI had been implanted earlier. In practice, prolonged DoD is not considered an absolute contraindication to cochlear implantation but rather is weighed by the clinician during preoperative counseling and consent conversations.

Many future VI candidates will have experienced BVH symptoms for decades before gaining access to vestibular implantation as a treatment option. Although data from a larger population would be needed to predict outcomes as a function of duration of loss with high certainty, this case supports the notion that a long duration of vestibular loss need not be disqualifying.

Conclusion:

The presented case suggests that longstanding adult-onset BVH (>20 years) should not be considered a contraindication for vestibular implantation. Given that many potential VI recipients will have lived with disabling BVH symptoms for decades, demonstration of VI benefit despite long duration of vestibular loss is a favorable indicator with important implications for patients considering vestibular implantation.

Supplementary Material

Supplemental 2

Video 2: Structured Interview 18mo Post Activation-23yr Hx of BVH

Supplemental 1

Video 1: eeVOR 23yr Hx BVH