Sudden Unilateral Hearing Loss With Simultaneous Ipsilateral Posterior Semicircular Canal Benign Paroxysmal Positional Vertigo: Title and subTitle BreakA Variant of Vestibulo-Cochlear Neurolabyrinthitis?

Posterior semicircular canal (pSCC) benign paroxysmal positional vertigo (BPPV) is a common vestibular disorder,¹ characterized by paroxysmal nystagmus with an up-beating vertical component and a torsional component with the superior poles of the eyes beating toward the undermost ear in the Dix-Hallpike position. The reversal of the nystagmus on putting the patient in an upright sitting position and the adaptation of the nystagmus on repeated testing are best explained by the canalithiasis hypothesis,¹ which suggests that dislodged utricular otoconia become trapped in the pSCC so that when the head changes position relative to gravity, the otoconia move within the pSCC and act as a piston to produce endolymph flow. This hypothesis is supported by the finding of otoconial matter in the pSCC of patients with BPPV during pSCC occlusion surgery² and by the high cure rate that results after different maneuvers aimed at moving the otoconia out of the affected pSCC.¹

While most cases of pSCC BPPV are primary or idiopathic (ie, they are not attributable to any underlying inner ear disease),^{3 - 5} approximately 15% follow an acute peripheral vestibulopathy,³ which is also known as vestibular neuritis⁶ or neurolabyrinthitis.⁷ Furthermore, although between 8% and 44% of cases are associated with a chronic ipsilateral sensorineural hearing loss,^{3 ,5} an association with simultaneous sudden ipsilateral sensorineural hearing loss is rare.⁴

We describe 4 patients who simultaneously developed sudden unilateral sensorineural hearing loss and pSCC BPPV on the same side but who had intact lateral SCC function (Table 1). Two of the patients also had signs of abnormal otolith function^{7 - 10} on the same side. These observations might provide clues concerning the possible causes and mechanisms of acute cochleovestibulopathies.

A 79-year-old man suddenly developed left-sided deafness and tinnitus and then, a few hours later, vertigo and nausea on head movement. The vertigo increased, and the patient vomited several times, until he was admitted to hospital on the evening of the same day. The next day, he had no nystagmus in light but a weak right-beating spontaneous nystagmus under Frenzel glasses. The results of the head impulse test¹¹ of lateral SCC function were normal. He tended to fall to the left and backward on the Romberg test. A Dix-Hallpike test to the left side elicited paroxysmal vertigo and nystagmus typical of left-sided pSCC BPPV with an up-beating vertical component and a torsional component with the superior poles of the eyes beating toward the left ear. Audiometry revealed total loss of hearing in the left ear. Electronystamography showed a right-beating spontaneous nystagmus with eyes closed (slow-phase velocity <4°/s), while the results of caloric tests of lateral SCC function were normal, with −1% canal paresis and 7% directional preponderance. The findings of computed tomography and magnetic resonance imaging of the brain were also normal. With vestibular exercises,¹ the pSCC BPPV resolved over 2 weeks, but the total deafness and tinnitus in the patient's left ear remained unchanged.

A 48-year-old woman woke with right-sided deafness and tinnitus, as well as vertigo, nausea, and vomiting. The vertigo was severe for 3 days but slowly decreased; the right-sided hearing loss persisted and positioning vertigo developed. The audiogram showed an almost completely deaf right ear (Figure 1). A Dix-Hallpike test to the right side elicited paroxysmal vertigo and nystagmus typical of right sided pSCC BPPV with an up-beating vertical component and a torsional component with the superior poles of the eyes beating toward the right ear. An Epley type particle-repositioning maneuver¹ was performed, and 2 days later, the results of the Dix-Hallpike test were normal. Although an electronystagmogram showed no spontaneous nystagmus, caloric tests of lateral SCC function showed only an 18% canal paresis in the right ear and a 41% directional preponderance on the left side, while the results of rotary chair tests and a head impulse test of lateral SCC function were normal. These findings are consistent with right-sided pSCC BPPV.¹² Vestibular evoked myogenic potentials (VEMPs),^{7 - 8} results of a test of the subjective visual horizontal,^{9 - 10} and findings of computed tomography of the brain were all normal. Three months later, the patient's vertigo had resolved and her hearing had improved only to a pure-tone average of 83-dB hearing level, with a speech discrimination of 20% at 90 dB.

A 66-year-old man woke with right-sided deafness, as well as vertigo, nausea, and vomiting. The vertigo became worse when he rolled over to the right in bed. The continuous vertigo resolved in a few hours, but the deafness and positioning vertigo persisted. Two weeks later, a Dix-Hallpike test to the right side elicited paroxysmal vertigo and nystagmus typical of right-sided pSCC-BPPV with an up-beating vertical component and a torsional component with the superior poles of the eyes beating toward the right ear. The audiogram showed a profound right sensorineural hearing loss (Figure 2). The electronystagmogram showed no spontaneous nystagmus, and the results of the caloric test were normal, with –3% canal paresis and 3% directional preponderance. The results of rotary chair tests and a head impulse test of lateral SCC function were normal, as were the VEMPs. A test of the subjective visual horizontal showed a 10° clockwise tilt, indicating acute loss of right utricular function.^{9 - 10} A magnetic resonance imaging scan of the brain revealed no abnormalities. Three Epley-type particle-repositioning maneuvers were tried during the following weeks, without success. After a Semont maneuver,¹ the patient's vertigo resolved. An audiogram 2 months later showed some recovery of hearing, with a pure-tone average of 75-dB hearing level and a speech discrimination of 13% at 90 dB.

A 40-year-old man suddenly developed left-sided deafness and tinnitus and, a few minutes later, vertigo and nausea. He was bedridden with vertigo for 2 days, but after 1 week he was able to go back to work. He then noticed attacks of positioning vertigo when he rolled to the left in bed and on getting out of bed. The left-sided hearing loss and tinnitus remained unchanged. An audiogram obtained 1 month later showed a left-sided high-frequency sensorineural hearing loss (Figure 3). A Dix-Hallpike test to the left side elicited paroxysmal vertigo and nystagmus typical of left-sided pSCC BPPV with an up-beating vertical component and a torsional component with the superior poles of the eyes beating toward the left ear. An electronystagmogram showed no spontaneous nystagmus, and the results of caloric tests of lateral SCC function were normal, with a 17% canal paresis in the left ear and 13% directional preponderance for right-beating nystagmus. The results of the head impulse test of lateral SCC function were normal, as were those of a test of the subjective visual horizontal, but the VEMPs were reduced in the left ear.^{7 - 8} The amplitude of click response was 316 µV in the right ear and 156 µV in the left ear, while the amplitude of tap response was 217 µV in the right ear and 28 µV in the left ear. Brainstem auditory evoked potentials and a computed tomogram of the brain were normal. An Epley-type particle-repositioning maneuver was performed, and 1 week later, the results of the Dix-Hallpike test were normal and the vertigo had resolved.

Our 4 patients all had pSCC BPPV and a spectrum of ipsilateral sensorineural hearing loss from partial to total; 2 patients also had the loss of utricular or saccular function. However, none manifested what is generally considered to be the hallmark of acute peripheral vestibulopathy: a canal paresis on caloric testing. A prerequisite for the development of pSCC BPPV is that some pSCC function remains. Thus, any combination of inner ear damage that detaches otoconia and spares some pSCC function could produce pSCC BPPV. In fact, the finding of a pSCC BPPV nystagmus indicates that the pSCC is still functioning to some extent. Furthermore, the combination of sudden sensorineural hearing loss with ipsilateral pSCC BPPV suggests that the lesion is within the labyrinth itself and not within the vestibulocochlear nerve.

What might cause pSCC BPPV with sudden unilateral sensorineural hearing loss? If the utricle and the cochlea were damaged in these patients, but the pSCC was spared, then, according to the canalithiasis hypothesis, sequestered otoconia in the pSCC could cause BPPV. The findings in these 4 cases show that acute cochleovestibulopathies are not limited to structures predicted by the divisions established by innervation or blood flow.

The internal auditory artery has 3 main branches: (1) the cochlear artery, which supplies the apical part of the cochlea; (2) the superior vestibular artery, which supplies the utricle and the anterior and lateral SCCs; and (3) the vestibulocochlear artery, which supplies the basal part of the cochlea, the saccule, and the pSCC.¹³ Thus, 1 vascular lesion could not easily explain the combination of cochlear and utricular damage without damage to the lateral SCC that was found in patient 3 or the combination of saccular damage with intact pSCC function that was found in patient 4. Multiple small emboli affecting more than 1 of the inner ear arteries or anatomical variants of inner ear vascular supply¹⁴ are possible, but not very likely, explanations. Labyrinthine hypoperfusion with transient cochlear ischemia has been suggested as a possible cause of sensorineural hearing loss during prolonged cardiopulmonary bypass.¹⁵ Transient labyrinthine ischemia might cause a "patchy" loss of inner ear function if parts of the labyrinth differ in their sensitivity to ischemia, and it has been suggested as a cause of otoconial release and BPPV.¹⁶ Kim et al¹⁷ recently reported the gross and microscopic findings in a patient who had unilateral sudden sensorineural hearing loss and vestibular loss 7 years before death. A fibrotic scar in the vestibulocochlear nerve and degenerative changes in the cochlea and the anterior and lateral SCCs, with patency of the internal auditory artery and its branches, were attributed to a transient period of reduced perfusion. However, the evidence that an ischemic lesion caused the lesions is only indirect, and the same pattern, with preserved function of the saccule and the pSCC, is the most common one in acute unilateral peripheral vestibulopathy.^{6 - 7} None of our 4 patients had a history or manifestations of arteriosclerotic disease. Furthermore, the findings of computed tomography and magnetic resonance imaging of the brain were normal in all patients, without signs of significant vascular lesions.

Viral infection has long been suspected as a possible cause of both acute unilateral peripheral vestibulopathy and sudden unilateral sensorineural hearing loss. Benign paroxysmal positional vertigo occurs in about 15% of patients after acute unilateral peripheral vestibulopathy,^{3 ,7} the hallmark of which, apart from the history of sudden, spontaneous vertigo that slowly decreases over several days, is unilateral impairment of lateral SCC function. None of our patients had a unilateral canal paresis on caloric testing or abnormal head impulse test results. Irrespective of the underlying cause, the simplest explanation of the findings of normal lateral SCC function is that the function might have recovered during the time between onset of symptoms and clinical testing. A normalization of the caloric reaction has been reported to take place in 46% of patients within 1 year of the onset of an acute unilateral peripheral vestibulopathy, and in 14% of patients, the function returns within the first 1 to 10 days.¹⁸ However, as the superior vestibular nerve carries fibers both from the lateral SCC and from the utricle,⁶ a complete cochleovestibular nerve lesion due to "neuritis" cannot adequately explain the combination of utricular damage and preserved lateral SCC function, as occurred in case 3. Likewise, the finding of saccular damage together with intact pSCC function in case 4 also argues against a gross nerve lesion, as fibers from the saccule and from the pSCC travel together in the inferior vestibular nerve.⁷

The distribution of herpes simplex type 1 in geniculate, vestibular, and spiral ganglia in human temporal bones has been studied by nested polymerase chain reaction.^{19 - 20} Herpes simplex type 1–specific DNA was found in all temporal bones, with different combinations of the ganglia being positive: 56% of the geniculate, 61% of the vestibular, and 50% of the spiral ganglia. All combinations of positive and negative ganglia were found in individual temporal bones with equal frequencies,¹⁹ and all parts of the examined vestibular ganglia were equally infected.²⁰ These findings support the hypothesis of a viral origin of both acute unilateral peripheral vestibulopathy and sudden unilateral sensorineural hearing loss, and especially their occasional combinations. A reactivation of a latent viral infection is compatible both with the histories of the patients and the findings of the "patchy" pattern of labyrinthine loss in our 4 cases. However, the findings are better explained by a viral labyrinthitis or a distal cochleovestibular neuritis than by a proximal nerve lesion. A comparison can be made with common labial herpes simplex reactivation, where the viruses migrate from the trigeminal ganglion along the mandibular sensory nerve branch and the lesion is confined to the dermis (the end organ) but the nerve trunk itself remains intact. In contrast, herpes zoster may damage the nerve, with sensory disturbance and neuralgia as sequelae.

Magnetic resonance imaging scans of the cochleovestibular nerve and the labyrinth in patients with acute vertigo and hearing loss show conflicting results. While Mark et al²¹ reported contrast enhancement of the labyrinth on magnetic resonance imaging scans that correlated with clinical findings of sudden hearing loss or vertigo or both in 12 patients, Strupp et al²² found no enhancement of the labyrinth or the vestibular nerve in 60 consecutive patients with vestibular neuritis. In contrast, there are commonly facial nerve enhancement in acute facial nerve palsy and enhancement of both the facial and the cochleovestibular nerve in herpes zoster oticus.²³ It has been suggested that the difference in nerve enhancement between acute facial nerve palsy and vestibular neuritis reflects the difference in vascularization between the facial and the vestibular nerves.²² A correlation between the grade of facial paresis and the extent of nerve enhancement has been reported in cases of herpes zoster oticus.²³ It might be that vestibular neuritis, a lesion confined to the organs that are innervated by the superior vestibular nerve,⁶ represents a mild reactivation of infection, while a cochleovestibular lesion is more widespread or aggressive and thus more likely to demonstrate labyrinthine contrast enhancement. This is an area for future studies.

Böhmer²⁴ described 3 patients with sudden unilateral sensorineural hearing loss who, within hours to days after the loss of hearing, developed BPPV with bilateral geotropic, horizontal nystagmus that did not lessen on repeated provocation and was more intense when the ear with the hearing loss was positioned down on the Dix-Hallpike test; ie, they had lateral SCC BPPV.²⁵ These patients had no spontaneous nystagmus and no canal paresis, and their vertigo resolved within weeks. They probably had damage to the cochlea and the utricle, with sequestered otoconia entering the lateral SCC instead of the pSCC. Patients with pSCC BPPV sometimes report that their positional vertigo started after 1 or 2 days of intense vertigo, which was relieved if they lay still and flat on their backs. If they turned their heads to the right or to the left, the vertigo immediately increased in intensity. One could speculate that such patients initially had lateral SCC BPPV that spontaneously converted to pSCC BPPV.

Loud monaural clicks evoke myogenic potentials in the ipsilateral sternocleidomastoid muscle: VEMPs.⁷ They are generated by a vestibulocollic reflex, and the findings of neurophysiological studies in animals,²⁶ as well as indirect evidence from studies of patients with vestibular neurolabyrinthitis,⁷ indicate that they are of saccular origin. While VEMPs are not affected by sensorineural hearing loss or deafness, they are attenuated by even mild conductive hearing loss.²⁷ The attenuation can be overcome by delivering the stimulus as repeated taps to the patient's forehead with a reflex hammer.⁸ We consider a side difference in VEMPs as significant if the amplitude on one side is less than one third that on the other side.

The visual perception of gravitational horizontal or vertical can be tested by asking a subject to adjust a dimly lit bar to the horizontal (subjective visual horizontal) or vertical (subjective visual vertical) in darkness.^{9 - 10} While normal subjects can adjust the light-bar to within ±2° of the true gravitational horizontal or vertical, patients with acute unilateral peripheral vestibular lesions set the bar at a tilt toward the lesioned side by up to 20°.^{9 - 10} The test reflects tonic torsional eye position and is thus presumed to indirectly measure utricular function.⁹

The positional vertigo was cured in all 4 patients by treatment with particle-repositioning or "liberatory" maneuvers, such as vestibular habituation exercises, the Epley maneuver, or the Semont maneuver. Three treatments with the Epley maneuver were ineffective in case 3. However, the patient's condition resolved after a subsequent Semont maneuver. This outcome indicates that treatment in such cases should not rely solely on one of the different maneuvers.

Accepted for publication May 16, 2000.

This study was supported by the National Health and Medical Research Council of Australia and the trustees of the Department of Neurology of the Royal Prince Alfred Hospital, Sydney. Dr Karlberg was supported by grants from the Swedish Medical Research Council and the Wenner-Gren Foundations, Stockholm, and the Maggie Stephens Foundation, Lund, Sweden.

Reprints: G. Michael Halmagyi, MD, Department of Neuro-otology, Royal Prince Alfred Hospital, Camperdown 2050 NSW, Sydney, Australia.