Clinical manifestations

Presentation and course

	• Vestibular neuritis (ie, vestibular loss alone) and viral labyrinthitis (ie, cochleovestibular loss) may both be caused by viral infections.
	• Viral labyrinthitis can occur as part of a systemic viral illness, or it may occur with localized involvement of the labyrinth or eighth nerve without apparent systemic involvement.

Viral labyrinthitis can occur as part of a systemic viral illness, or it may occur with localized involvement of the labyrinth or eighth nerve without apparent systemic involvement (37; 07). Evidence of pharyngitis, tonsillitis, or sinusitis may precede or occur with labyrinthitis (17; 18; 53; 52; 55; 09; 08; 07). With unilateral disease, the nystagmus is typical of peripheral vestibulopathy, with a horizontal-rotatory nystagmus beating toward the healthy ear that is direction-fixed and enhanced by removal of visual fixation (55; 07; 34; 57). Curiously, peripheral vestibular nystagmus in vestibular neuritis is not active at the brainstem level during REM sleep (19).

Viral labyrinthitis typically has an acute or subacute onset with hearing loss, subjective tinnitus, intense vertigo with spontaneous nystagmus, gait unsteadiness/ataxia, and nausea/vomiting.

If the symptoms and signs are predominantly vestibular, without audiological or other neurologic symptoms and signs, the condition is called "vestibular neuronitis" or "vestibular neuritis," the term proposed in the original full description by Dix and Hallpike in 1952 (17; 18). Otoscopy and pure tone audiometry are typically normal, whereas bithermal caloric testing shows moderate to complete canal paresis (17; 18). Although a viral basis is likely in this circumstance, a very careful assessment should be done, especially in the elderly and in those with vascular disease risk factors, particularly to exclude vestibular presentations of posterior circulation ischemia.

Clinical features of vestibular neuritis include (1) a sensation of rotatory vertigo with an acute onset lasting several days; (2) associated nausea; (3) spontaneous horizontal-torsional nystagmus beating toward the unaffected ear; (4) abnormal head impulse test for the involved semicircular canals (ie, abnormal with head impulse test toward the affected ear); (5) unsteadiness with a tendency to fall toward the affected side; (6) deviation of the subjective visual vertical toward the affected side; and (7) normal hearing (07; 58; 34; 57). The head impulse test and caloric irrigation show an ipsilateral deficit of the vestibulo-ocular reflex (58). The clinical features are similar for labyrinthitis, except that hearing is impaired in the affected ear (or ears).

In a longitudinal study of 51 consecutive patients with vestibular neuritis, all patients had abnormal head-shaking nystagmus, caloric stimulation, and head thrust test results as well as at least one otolith-related test abnormality: abnormal tilt of subjective visual vertical, abnormal ocular torsion, or abnormal vestibular evoked myogenic potentials (VEMPs) (38). Less common findings included skew deviation (14%) and a complete ocular tilt reaction (4%) (38), although current diagnostic criteria would exclude cases with skew deviation (57).

There is a significant difference between horizontal vestibulo-ocular reflex (h-VOR) characteristics in vestibular neuritis (ie, vestibular loss alone) and viral labyrinthitis (ie, cochleovestibular loss) (41). Patients with vestibular neuritis exhibit a strong horizontal semicircular canal deficit but no h-VOR asymmetry between the two rotational directions, whereas patients with viral labyrinthitis typically exhibit moderate canal paresis and a marked h-VOR deficit with rotation toward the affected ear. Thus, h-VOR dynamic asymmetry after an acute unilateral inner ear lesion is not due to canal dysfunction alone but instead involves complex adaptive changes in the central VOR; this otolith-canal interaction is mainly linked to the loss of saccular function (41).

Prognosis and complications

In their original description, Dix and Hallpike noted that vestibular neuritis "generally recovers in the course of a few years" (17; 18). Recovery from vestibular neuritis is due to a combination of (a) peripheral restoration of labyrinthine function; (b) somatosensory and visual substitution; and (c) central compensation (58). Peripheral restoration of vestibular function is seldom complete but can be improved with corticosteroids (04; 59; 58). Central compensation can be improved by vestibular rehabilitation exercises (58; 34).

On follow-up, otolith test results returned to normal more rapidly than canal test results in a longitudinal study of 51 consecutive patients with vestibular neuritis (38). The head thrust test was the best predictor of symptom recovery: 80% of patients who continued to report dizziness/vertigo at the last follow-up visit had a positive head thrust test result, whereas only 10% of patients who were not dizzy/vertiginous had a positive head thrust test result.

Residual symptoms often persist months or even years after onset: more than half of affected patients report minor symptoms, such as vertigo with sudden movement of the head and imbalance in the dark (35). Static symptoms invariably resolve, although often not completely, whereas dynamic symptoms and signs (eg, impairment of vestibulo-ocular reflexes from the ipsilesional semicircular canals on head impulse testing) only improve slightly, if at all (26).

Benign paroxysmal positioning vertigo often occurs in patients who have had vestibular neuritis, presumably because of viral disruption of the otolithic membrane (03).

Biological basis

Etiology and pathogenesis

	• Multiple viruses have been implicated in viral labyrinthitis, but the evidence supporting a viral etiology is often circumstantial.
	• Viral reactivation of latent herpes simplex type 1 in vestibular ganglia is a suspected cause of vestibular neuritis.

Clinical and serologic studies of patients with vestibular neuritis suggest that the viruses are involved in the pathogenesis of this disease, but the exact pathophysiologic mechanisms are unclear. Proposed theories of causation have included viral reactivation, vascular occlusion, and immune-mediated mechanisms (25).

In vestibular neuritis, inflammation of the vestibular nerve is followed by demyelination and loss of function (10). Clinical, laboratory, and neuroradiological findings support an inflammatory component to the pathogenesis: (1) intravenous corticosteroids are useful in facilitating resolution of acute symptoms, presumably by limiting the negative consequences of the inflammatory response (04; 59); (2) higher plasma fibrinogen and CRP levels are present in the acute phase; and (3) there is increased gadolinium uptake in the vestibular nerve and Scarpa ganglion on enhanced MRI.

Multiple viruses have been implicated in viral labyrinthitis, but the evidence supporting a viral etiology is often circumstantial (eg, upper respiratory tract illness within 1 to 2 weeks before onset of vestibular symptoms) (17; 18; 35; 40; 20; 13; 33; 49) (Tables 1–2).

Table 1. Viruses Implicated in Viral Labyrinthitis and Vestibular Neuritis

	• Adenoviral vector-based COVID-19 vaccination (54)
	• Adenovirus (64)
	• Coxsackievirus
	• Cytomegalovirus (40)
	• Enterovirus (unspecified) (21)
	• Epstein-Barr virus
	• Hepatitis E virus (Orthohepevirus A) (63)
	• Herpes simplex type 1 (01; 40)
	• Human herpes virus 6/7 (20)
	• Influenza A virus (55)
	• Influenza B virus (40)
	• Lymphocytic choriomeningitis virus
	• Measles virus (MeV), (rubeola)
	• Mumps virus (MuV)
	• Parainfluenza virus (55)
	• Rubella virus (RuV)
	• SARS-CoV-2 (COVID-19) (06; 13; 23; 33; 49; 43)
	• Varicella zoster virus (37)

Table 2. Viruses Implicated in Viral Labyrinthitis and Vestibular Neuritis, Classified by Nucleic Acid and Envelope

DNA viruses
	• Enveloped, double-stranded DNA
		- Herpesviridae (herpesviruses)
			-- herpes simplex type 1
			-- Epstein-Barr virus
			-- varicella zoster virus (human herpes virus 3)
			-- human herpes virus 6/7
	• Unenveloped, double-stranded DNA
		- Adenoviridae
			-- adenoviruses
			-- adenoviral vector-based vaccines
RNA viruses
	• Enveloped (+) RNA
		- Coronaviridae (coronaviruses)
			-- SARS-CoV-2
		- Matonaviridae
			-- rubella virus
		- Picornaviridae (picornaviruses)
			-- coxsackievirus
		- Hepeviridae
			-- hepatitis E virus
	• Enveloped (-) RNA
		- Paramyxoviridae (paramyxoviruses)
			-- measles virus
			-- mumps virus
			-- human parainfluenza viruses (HPIVs)
		- Orthomyxoviridae (orthomyxoviruses)
			-- influenza A
			-- influenza B
	• Ambisense (+/-) RNA
		- Arenaviridae (arenavirus)
			-- lymphocytic choriomeningitis virus

In patients who died with viral encephalitis and labyrinthitis, labyrinthitis-related pathology was restricted to the scala media, vestibular labyrinth, and internal auditory canal (37). Histologic findings included variable degeneration of the vestibular labyrinth with sloughing of the otolithic membrane, degeneration of the organ of Corti, early encapsulation of the tectorial membrane, degeneration of the stria vascularis, and round cell infiltration of the modiolus and eighth cranial nerve.

The histologic findings in the temporal bones of individuals who had vestibular neuritis suggest viral infection as the main etiologic cause (52; 16; 09; 10). In some cases, pathology has demonstrated viral involvement of the vestibular end organs and eighth cranial nerve (52; 09), typically with maximal damage in the distal branches of the vestibular nerve (16).

Viral reactivation in temporal ganglia is the suspected cause of vestibular neuritis, as well as Bell palsy and sudden hearing loss. Latent herpes simplex type 1 (HSV-1) has been detected in vestibular (Scarpa), geniculate (a sensory ganglion of the facial nerve), and spiral (cochlear) ganglia of human temporal bones by nested polymerase chain reaction (01; 02; 03). The distribution of HSV-1 in human geniculate and vestibular ganglia does not support either viral migration along the anastomosis between the intermediate nerve and the superior vestibular nerve or a preferential latency of HSV-1 in the superior vestibular nerve (01). HSV-1 can be detected in about one third of cases involving the geniculate and vestibular ganglia and the vestibular nuclei; the various patterns of HSV-1 infection of vestibular structures are compatible with virus migration from the vestibular ganglia to the vestibular nuclei and from the ipsilateral to the contralateral vestibular nucleus via commissural fibers (02).

The superior vestibular nerve is more frequently affected than the inferior vestibular nerve, which is partially or totally spared (46; 28). The reasons for this are poorly understood, but anatomical differences and the distribution of latent viruses (eg, herpes simplex type 1) may contribute (46; 28). Latent herpes simplex virus 1 can be demonstrated in only a minority of vestibular ganglia, but infected neurons are typically located only in the superior part (28). However, other studies found no difference in the relative numbers of latency-associated transcripts or production of virions between reactivated, latently infected superior versus inferior vestibular ganglion neurons (46).

Vestibular neuritis can produce chronic deafferentation of the vestibular sensory epithelia and the vestibular nuclei (09). Postmortem examination of the brain and temporal bones of a patient with vestibular neuritis showed (1) selective neuronal loss in Scarpa ganglia on the side with absent caloric response; (2) loss of hair cells and an "epithelialization" of the utricular macule and semicircular canal cristae on the deafferented side; and (3) decreased synaptic density in the vestibular nuclei on the deafferented side compared to the normal side.

The vestibulo-cortical system, which includes the parieto-insular vestibular cortex, incorporates an extensive cortical network connected with various subcortical processing areas (eg, oculomotor, somatosensory, visual, and cerebellar regions). Consequently, sudden unilateral vestibular impairment induces a wide variety of cortical and subcortical responses, including changes in regional cerebral glucose metabolism in the vestibular cortex, decreased resting-state activity in the contralateral intraparietal sulcus (close to the human equivalent of the parieto-insular vestibular cortex), and decreased internetwork connectivity between the “default mode” network and multiple other networks (11; 44). Simultaneously, a neural cascade of acute plasticity activity occurs in the visual and multisensory cortex as well as in areas involved in spatial navigation (eg, entorhinal, supramarginal gyrus).

Animal studies. Animal studies have demonstrated that several human viruses can infect the vestibular nerve and the vestibular membranous labyrinth, including rubeola, herpes simplex, cytomegalovirus, and neurotropic strains of influenza A and mumps virus (16; 10).

An animal model of vestibular neuritis has been developed using retroauricular inoculation of herpes simplex virus in mice (30; 29; 10; 22).

Epidemiology

	• Vestibular neuritis is a common cause of peripheral vestibular vertigo, with an annual incidence of 3.5 per 100,000 population.
	• Vestibular neuritis chiefly affects those 30 to 50 years of age, without preference for sex.

Vestibular neuritis is a common cause of peripheral vestibular vertigo, with an annual incidence of 3.5 per 100,000 population (58).

As recognized by Dix and Hallpike in their initial descriptions, vestibular neuritis "chiefly affects the age group 30 to 50 without preference for sex" (17; 18); this has largely been confirmed by later studies (10).

No clear seasonal pattern has been demonstrated for incident cases of vestibular neuritis (32).

In a case-control study at a tertiary referral hospital, vestibular neuritis in the elderly was associated with cerebral small vessel disease, suggesting (though not proving) that small vessel cerebrovascular disease can contribute to incident vestibular neuritis in the elderly (48).

A genome-wide association study of vestibular neuritis identified four regions containing protein coding genes related to host factors for viral replication or associated with insulin metabolism and resistance (50). Genes related to viral processes include (1) nuclear receptor subfamily 3 group C member 2 (NR3C2), a receptor for mineralocorticoids and glucocorticoids that is a host factor for HSV-1 replication; (2) ankyrin repeat domain 30A (ANKRD30A), which encodes a host factor for HIV-1 infection, shows rapid evolution, and is induced by interferon stimulation; and (3) mediator complex 30 (MED30), which is involved in the replication of HIV-1, with a knockdown leading to impaired viral replication (50). A later case-control study by the same group identified a significant association of the rs12979860-T risk allele for herpes labialis severity with susceptibility to vestibular neuritis, providing further indirect evidence for the involvement of HSV-1 in vestibular neuritis pathogenesis (51).

Prevention

There is presently no adequate form of prevention for viral labyrinthitis or vestibular neuritis. Although vaccines exist for some of the implicated viruses, some cases of vestibular neuritis are thought to be caused by inflammatory responses to viruses or reactivation of latent viruses within neurons.

Differential diagnosis

Confusing conditions

Relevant differential diagnoses include "vestibular pseudoneuritis" due to acute pontomedullary brainstem lesions (eg, AICA, PICA) or cerebellar nodular infarctions (39; 47), vestibular migraine, monosymptomatic-onset Meniere disease, and Ramsay Hunt syndrome (07; 15; 58). Less common considerations include perilymph fistula (07), isolated labyrinthine infarction (07), and demyelinating disease (61). Perilymph fistulas have an abrupt onset associated with head trauma, barotrauma, or sudden strain (eg, heavy lifting, coughing, or sneezing) and may be associated with chronic otitis with cholesteatoma (07).

Skew deviation is a specific but insensitive (40%) sign of vestibular pseudoneuritis (15).

In patients with acute vestibular syndrome, a negative horizontal head impulse test (ie, normal VOR) strongly suggests a central lesion with a "pseudo-labyrinthine" presentation (47). Most patients with acute peripheral vestibulopathy from a stroke have a negative horizontal head impulse test, but contrary to conventional wisdom, about 10% have a positive horizontal head impulse test. In particular, patients with lateral pontine or cerebellar strokes (eg, vestibulocerebellar, pontocerebellar, and pontocerebello-labyrinthine strokes) may have a positive horizontal head impulse test. Therefore, a positive horizontal head impulse test does not prove that an acute vestibular syndrome is due to peripheral (and relatively benign) pathology (eg, vestibular neuritis). Additional clinical features (eg, directionality of nystagmus, severity of truncal instability, nature of hearing loss) and neuroimaging should be considered in patients with acute vestibular syndrome with a positive horizontal head impulse test before a central localization can be confidently excluded. Head impulse gain and saccade analysis may further help distinguish pontine cerebellar stroke and vestibular neuritis (14).

Diagnostic workup

	• Inclusion criteria for “acute unilateral vestibulopathy” (vestibular neuritis) are (1) acute or subacute onset of sustained vertigo (acute vestibular syndrome) of moderate to severe intensity that is symptomatic for at least 24 hours; (2) spontaneous peripheral vestibular nystagmus; and (3) reduced vestibulo-ocular reflex (VOR) function on the affected side.
	• Exclusions to a diagnosis of “acute unilateral vestibulopathy” (vestibular neuritis) include (1) acute central neurologic symptoms, audiological symptoms, or otologic symptoms; (2) acute central neurologic signs (eg, skew deviation or gaze-evoked nystagmus); (3) acute audiological signs; and (4) the condition is better accounted for by another disease or disorder.

Careful bedside examination of patients with suspected labyrinthitis or vestibular neuritis is important for both diagnosis and prognosis (42). In particular, bedside oculomotor findings play a critical role in differentiating vestibular neuritis from stroke and other pathologies.

The diagnostic hallmarks of vestibular neuritis are (1) rotatory vertigo with an acute onset lasting several days; (2) associated nausea; (3) spontaneous horizontal-torsional nystagmus beating toward the unaffected ear; (4) abnormal head impulse test for the involved semicircular canals (ie, toward the affected ear); (5) unsteadiness with a tendency to fall toward the affected side; (6) deviation of the subjective visual vertical toward the affected side; (7) ipsilesional caloric paresis (with electro- or videonystagmography); (8) decreased responses of vestibular-evoked myogenic potentials during stimulation of the affected ear; and (9) normal pure tone audiometry (07; 58; 34; 57). The head impulse test and caloric irrigation show an ipsilateral deficit of the vestibulo-ocular reflex (58). The diagnostic hallmarks are similar for labyrinthitis, except that hearing is impaired in the affected ear (or ears), which can be quantified with pure tone audiometry.

The Barany Society has developed diagnostic criteria for “acute unilateral vestibulopathy” (vestibular neuritis) (Table 3), variants (ie, for evolving and "probable" cases not meeting the full diagnostic criteria), and for a history of acute unilateral vestibulopathy (vestibular neuritis) (Table 4) (57).

Table 3. Barany Society Diagnostic Criteria For “Acute Unilateral Vestibulopathy” (Vestibular Neuritis)

Inclusions (all required)
	• Acute or subacute onset of sustained vertigo (acute vestibular syndrome) of moderate to severe intensity that is symptomatic for at least 24 hours
	• Spontaneous peripheral vestibular nystagmus, ie, with a trajectory appropriate to the semicircular canal afferents involved (generally horizontal-torsional), direction-fixed, and enhanced by removal of visual fixation
	• Reduced vestibulo-ocular reflex (VOR) function on the affected side (ie, opposite the direction of the fast phase of the spontaneous nystagmus)
Exclusions
	• Acute central neurologic symptoms, audiological symptoms (eg, hearing loss or tinnitus), or otologic symptoms (eg, otalgia)
	• Acute central neurologic signs, including central ocular motor or central vestibular signs (eg, skew deviation, gaze-evoked nystagmus)
	• Acute audiological signs
	• Better accounted for by another disease or disorder
	(57)

The Barany Society created a separate diagnostic category of “acute unilateral vestibulopathy in evolution” (vestibular neuritis in evolution) for symptoms that have persisted for more than 3 hours but have not yet lasted for at least 24 hours (57); the diagnostic criteria are otherwise unchanged from those for “acute unilateral vestibulopathy” (vestibular neuritis). If there is no clear evidence of reduced VOR function by bedside examination on the side opposite the direction of the fast phase of the spontaneous nystagmus, but the diagnostic criteria are otherwise met, the Barany Society categorizes this as “probable acute unilateral vestibulopathy” (probable vestibular neuritis) (57).

Table 4. Barany Society Diagnostic Criteria For “History of Acute Unilateral Vestibulopathy” (History of Vestibular Neuritis)

Inclusions (all required)
	• Acute or subacute onset of sustained vertigo (acute vestibular syndrome) that was symptomatic for at least 24 hours and then slowly decreased in intensity over days
	• Unilateral reduced VOR function
Exclusions
	• History of acute central neurologic symptoms or audiological symptoms (eg, hearing loss or tinnitus)
	• History of acute central neurologic or audiological signs
	• Better accounted for by another disease or disorder
	(57)

Acute vestibular neuritis most often affects both vestibular nerve divisions (05; 60). Canal-plane video head impulse testing (ie, horizontal vHIT) identifies superior nerve dysfunction, which is uniformly present in patients with vestibular neuritis tested acutely, whereas both cervical/vestibular evoked myogenic potentials and posterior vHIT are necessary to diagnose inferior vestibular nerve involvement (45; 60). Rare cases of isolated "inferior vestibular neuritis" have been reported with symptoms consistent with vestibular neuritis but normal caloric testing (05; 45). A slight, asymptomatic, position-dependent nystagmus may be observed with the pathological ear down in patients with inferior vestibular neuritis (45). Unilateral loss of vestibular evoked myogenic potentials is found in these cases (45).

Brainstem evoked response audiometry (BERA) shows prolonged latencies and I–III interpeak interval (10).

Higher plasma fibrinogen and CRP levels may be demonstrated in the acute phase (10).

Increased gadolinium uptake may be demonstrated in the vestibular nerve and Scarpa ganglion on enhanced MRI (36; 10). High-field-strength MRI (3.0 T) with high-dose (0.3 mmol/kg of body weight) gadolinium-pentetic acid may show isolated enhancement of the vestibular nerve on the affected side, supporting the hypothesis of a viral and inflammatory cause of acute vestibular neuritis (36).

Brain imaging is indicated to exclude a vascular basis, even in patients with typical clinical findings, particularly when the patient is elderly or has known vascular disease or significant vascular risk factors, unprecedented headache, a negative head impulse test, or severe unsteadiness (62; 34).

Management

	• The management of the acute phase of vestibular neuritis is primarily medical, whereas long-term treatment is designed to improve vestibular compensation.
	• Symptomatic medication is indicated during the acute phase to relieve the vertigo and nausea/vomiting.
	• Antihistamines, anticholinergic agents, antidopaminergic agents, and GABAergic agents are useful in acutely suppressing vertiginous symptoms.
	• Corticosteroids are beneficial, but available evidence does not support the use of antiviral agents.

The management of the acute phase of vestibular neuritis is primarily medical, whereas long-term treatment is designed to improve vestibular compensation.

Symptomatic medication is indicated during the acute phase to relieve the vertigo and nausea/vomiting. Antihistamines, anticholinergic agents, antidopaminergic agents, and GABAergic agents (eg, gabapentin, benzodiazepines) are useful in acutely suppressing vertiginous symptoms (07). Antiemetics and vestibular suppressants should be withdrawn as soon as feasible (preferably after the first several days) because, at least theoretically, their prolonged use may impede the process of central vestibular compensation (62).

Corticosteroids and antiviral agents have also been used in the treatment of vestibular neuritis, but the available evidence provides no convincing basis to use antiviral agents (07; 59; 25). Corticosteroids are useful in facilitating resolution of acute symptoms, presumably by limiting the negative consequences of the inflammatory response (04; 59; 24; 31; 12; 27). In a controlled trial, methylprednisolone was more effective than placebo in controlling acute symptoms (04). In a later trial, methylprednisolone, valacyclovir, and combined methylprednisolone plus valacyclovir were administered to three patient groups with vestibular neuritis (59). Methylprednisolone alone was effective, whereas valacyclovir had no evident efficacy in the treatment of vestibular dysfunction, either alone or in combination with methylprednisolone. In this trial, methylprednisolone was administered as a single morning dose on a tapering schedule over about 3 weeks: 100 mg on days 1 to 3; 80 mg on days 4 to 6; 60 mg on days 7 to 9; 40 mg on days 10 to 12; 20 mg on days 13 to 15; 10 mg on days 16 to 18; and 10 mg on days 20 and 22 (59).

Early resumption of normal activity should be encouraged to promote compensation (62). Directed vestibular rehabilitation further hastens recovery (56; 58; 62; 34; 24; 31; 27).