Timothy C. Hain, MD • Page last modified: November 8, 2020
New technology has greatly enhanced our ability to detect vestibular neuritis. The VHIT test is both accurate and very quick, as well as unlikely to make a person dizzier than they already are. We offer this technology in our clinic in Chicago.
In vestibular neuritis, also known as vestibular neuronitis, dizziness is attributed to a viral infection of the vestibular nerve or ganglion (see figure 1). The vestibular nerve carries information from the inner ear about head movement. When one of the two vestibular nerves is infected, there is an imbalance between the two sides, and vertigo appears. Vestibular neuronitis is another term that is used for the same clinical syndrome, and reflects the idea that the damage is often to the vestibular ganglion which contains neurons rather than the vestibular nerve itself.
The various terms for the same clinical syndrome probably reflect our lack of ability to localize the site of lesion. The term "neuritis" implies damage to the nerve, and "neuronitis', damage to the sensory neurons of the vestibular ganglion. There is actually evidence for both. There is also some evidence for viral damage to the brainstem vestibular nucleus (Arbusow et al, 2000), a second potential "neuronitis". As the vestibular neurons are distinct from cochlear neurons in the brainstem, this localization (as well as the vestibular ganglion) makes more sense than the nerve in persons with no hearing symptoms. Nevertheless, if the nerve were involved after it separates from the cochlear nerve, neuritis would still be a reasonable mechanism. Prior to death and autopsy there is no way to make a clear distinction, and the present favored term is "neuritis".
Labyrinthitis, is defined as the combination of the symptoms of vestibular neuritis, with the addition of hearing symptoms. It may be due to a process that affects the inner ear as a whole or the 8th nerve as a whole. Labyrinthitis is also always attributed to an infection.
Figure 1: Cutaway of the inner ear. Movement of the head is detected by the semicircular canals, and transmitted to the brain via the vestibular nerve. Vestibular neuritis may affect the nerve itself or the vestibular ganglion (Scarpa's ganglion).
In vestibular neuritis, the virus that causes the infection is thought to be usually a member of the herpes family, the same group that causes cold sores in the mouth as well as a variety of other disorders (Arbusow et al, 2000). There is some controversy about this idea however, as there is little direct evidence for herpes infection (Matsuo, 1986). Slightly more evidence was provided by Rujescu et al (2020) who noted that vestibular neuritis is associated with a gene suggesting risk for herpes. The varacella zoster virus (the cause of Ramsey Hunt) is also thought to be a common source of vestibular neuritis.
It is also thought that a similar syndrome indistinguishable from vestibular neuritis can be caused by loss of blood flow to the vestibular system (Fischer, 1967). However, present thought is that inflammation, presumably viral, is much more common than loss of blood flow. There are still some that disagree (Fattori et al. 2003) .
In labyrinthitis, it is thought that generally viruses cause the infection, but rarely labyrinthitis can be the result of a bacterial middle ear infection. While there are several different definitions of vestibular neuritis in the literature, with variable amounts of vertigo and hearing symptoms, we will use the definition of Silvoniemi (1988) who stated that the syndrome of vestibular neuritis is confined to the vestibular system. In vestibular neuritis, by definition, hearing is unaffected. In labyrinthitis, hearing may be reduced or distorted in tandem with vertigo.
These definitions are flawed -- they depend on clinical findings and imply anatomic localization that may not always be true. It has been reported that some patients with the clinical syndrome of "vestibular neuritis", anatomically may have lesions in the labyrinth (Murofushi et al, 2003). Although anatomic data is rarely available, if diagnostic technology improves in the future, we may need to change the definition of "vestibular neuritis".
Both vestibular neuritis and labyrinthitis are rarely painful -- when there is pain it is particularly important to get treatment rapidly as there may be a treatable bacterial infection or herpes infection.
The symptoms of both vestibular neuritis and labyrinthitis typically include dizziness or vertigo, disequilibrium or imbalance, and nausea. Acutely, the dizziness is constant. After a few days, symptoms are often only precipitated by sudden movements. A sudden turn of the head is the most common "problem" motion. While patients with these disorders can be sensitive to head position, it is generally not related to the side of the head which is down (as in BPPV), but rather just whether the patient is lying down or sitting up.
Pathologic study of a single patient documented findings compatible with an isolated viral infection of Scarpa's ganglion (the vestibular ganglion). There was loss of hair cells, "epithelialization" of the utricular maculae and semicircular canal cristae on the deafferented side, and reduced synaptic density in the ipsilateral vestibular nucleus (Baloh et al, 1996).
Absent oVEMP (superior division), Courtesy of Dr. Dario Yacovino.
|Absent left anterior and lateral canal response on VHIT, Courtesy of Dr. Dario Yacovino.|
In spite of the limited pathology that would suggest involvement of the entire vestibular nerve, there is reasonable evidence that vestibular neuritis often spares part of the vestibular nerve, the inferior-division (e.g. Fetter and Dichgans, 1996; Goebel et al, 2001) although not all agree (Lu et al, 2003). Because the inferior division supplies the posterior semicircular canal and saccule, even a "complete" loss on vestibular testing (associated with a superior canal lesion) may be associated with some retained canal function.
Furthermore, it is common to have another dizziness syndrome, BPPV, follow vestibular neuritis. Presumably this happens because the utricle is damaged (supplied by the superior vestibular nerve), and deposits loose otoconia into the preserved posterior canal (supplied by the inferior vestibular nerve).
Similarly, an inferior division vestibular neuritis might be associated with a normal ENG test but an abnormal VEMP test. (Aw et al, 2001). Practically speaking, we are not sure that this is a real entity as abnormal VEMPs are common in otherwise normal persons. While the new VHIT test can theoretically measure inferior division vestibular neuritis, we are not sure right now that this is accurate. That being said, we have no doubts about VHIT diagnosing the common superior division vestibular neuritis.
About 5% of all dizziness (and perhaps 15% of all vertigo) is due to vestibular neuritis or labyrinthitis. It occurs in all age groups, but cases are rare in children. Brodsky et al (2016) reported 11 patients out of 301 evaluated at their pediatric vestibular clinic. (Brodsky et al, 2016).
In the author's clinical practice, Chicago Dizziness and Hearing, as of 2016 there were more than 700 patients with this diagnosis out of roughly 20,000 total patients. This fraction is far less than patients with Migraine, but comparable to patients with Meniere's disease.
Acutely, in uncomplicated cases, a thorough examination including video-frenzel goggles is all that is necessary. The VHIT test is particularly good for diagnosis as it is very efficient in detecting vestibular neuritis and is generally normal in strokes. Certain types of specialists, "otologists", "neurotologist", and "otoneurologists", are especially good at making these diagnoses and seeing one of these doctors early on may make it possible to avoid unnecessary testing, as well as inappropriate treatment or medication. In large part, the process involves ascertaining that the entire situation can be explained by a lesion in one or the other vestibular nerve. It is not possible on clinical examination to be absolutely certain that the picture of "vestibular neuritis" is not actually caused by a brainstem stroke, or cerebellar stroke, or an inner ear tumor, so mistakes are possible (Lee et al, 2003). A positive VHIT test helps considerably by (usually) crossing strokes off the list of possibilities. Nevertheless, strokes and tumors happen so rarely that it is not necessary to perform MRI scans or the like very often for this situation.
In severe or complex situations, we will sometimes (not always) order the following tests:
- Rotatory chair test -- acutely there is spontaneous nystagmus that overwhelms almost everything.
- Audiogram (to distinguish labyrinthitis from neuritis, and to detect Meniere's disease).
- VEMP (to detect vestibular neuritis -- cVEMP should be normal, oVEMP should be abnormal)
- MRI of brain and IAC, with and without contrast (to be sure there are no tumors, inflammation of nerve or cochlea).
- Blood testing (fasting blood sugar, Lyme, FTA, CRP, HSV-1 titer)
An example of an acutely positive Rotatory chair:
There is reduced gain at low frequencies, increased phase, a very strong asymmetry (from the spontaneous nystagmus). The gain-TC product here was 2.6, which suggests unilateral loss.
The recording here shows very powerful right-beating spontaneous nystagmus, associated with left vestibular neuritis.
After a week goes by, the spontaneous nystagmus abates but the gain/phase remain abnormal.
As first described by Fluur (1973), on positional testing, the horizontal nystagmus is usually strongest with the "bad" ear down, and weakest with the "good ear up". This is called "homolateral excitation". This is shown above using the head-prone test. This individual has strong left vestibular neuritis, with right-beating nystagmus. It is more powerful with the head-right (putting the left ear down), than head-left. This particular person's torsional nystagmus was more obvious with the right ear down. Little has been done to investigate this phenomenon since 1973.
The VHIT test has immensely simplified the differential diagnosis of Vestibular Neuritis. VN is mainly diagnosed with security when there is a strong unilateral positive on the VHIT, and symptoms that last longer than a few days. The VHIT is not yet universally available, but its adoption seems to be rapid in the country. The office "HIT" test is nearly as good, as long as the person doing the test is highly experienced. However, we prefer the paper trail of the VHIT to the more subjective nature of the HIT.
There are many medical conditions that can create roughly the same constellation of findings and symptoms as vestibular neuritis and labyrinthitis. Sorting these out usually is done by a physician who can combine clinical knowledge and experience with results of inner ear testing. A "classic" case of VN mainly relies on ascertaining that findings consist of a subacute onset (over hours but usually lasting days) of pure dizziness.
- If there are hearing symptoms with dizziness, then labyrinthitis would be the first consideration. Menieres disease is also a very reasonable consideration.
- Some persons have a"stuttering" course -- a series of abrupt onset/stopping of symptoms (usually attributed to circulation problems). A constellation of accompanying signs that would be unusual in ear disease -- such as weakness, numbness, unusually prominent unsteadiness, or skew deviation can also help make this diagnosis.
- A slowly progressive dizziness is usually attributed to a slowly growing or worsening process, such as an acoustic tumor.
- When there are hearing symptoms with spells, Meniere's disease would be considered more seriously. When there are headaches, prominent visual symptoms, and light sensitivity -- migraine.
Signs of vestibular neuritis include spontaneous nystagmus, and unsteadiness. The two movies above show VN after a week and VN more acutely at one day.
One may notice that vision is disturbed or jumpy on looking to a particular side. This usually means that the opposite ear is affected -- it is called "Alexander's Law" and is due to asymmetric gaze evoked nystagmus. Occasionally other ocular disturbances will occur such as vertical double vision -- skew deviation (Safran et al, 1994).
However if symptoms persist beyond one month, reoccur periodically, or evolve with time (see following), testing may be proposed. In this situation, nearly all patients will be asked to undergo an audiogram and an ENG. An audiogram is a hearing test needed to distinguish between vestibular neuritis and other possible diagnoses such as Meniere's disease and Migraine. The ENG test is essential to document the characteristic reduced responses to motion of one ear. An example of this is shown on the caloric test page.
A test called a VEMP may be helpful in determining the extent of damage (Lu et al, 2003). Also, VEMP can be helpful in confirming the diagnosis of vestibular neuritis as opposed to another process that has damaged the nerve as most persons with vestibular neuritis will have reduced ENG function but a present (albeit perhaps reduced) VEMP. VEMP's recover more quickly than other tests do in vestibular neuritis (Kim et al, 2008).
An MRI scan will be performed if there is any reasonable possibility of a stroke or brain tumor. In most instances, it is most cost effective to see a neurologist prior to obtaining an MRI. As illustrated in the central vertigo cases here, sometimes these can be very difficult to detect at the bedside. Occasionally with an MRI one can visualize the inflammation of the vestibular nerve or of the labyrinth. A case of cochlear inflammation is shown here. Blood tests for diabetes, thyroid disorders, Lyme disease, collagen vascular disease and syphilis are sometimes performed, looking for these treatable illnesses. However, it is rare that these are ever positive.
Acutely, vestibular neuritis is usually treated symptomatically, meaning that medications are given for nausea (anti-emetics) and to reduce dizziness (vestibular suppressants). Typical medications used are "Antivert (meclizine)", "Ativan (lorazepam) ", "Phenergan", "Compazine", and "Valium (diazepam) ". When a herpes virus infection is strongly suspected, a medication called "Acyclovir" or a relative may be used. This is particularly common when there is a recurrent pattern (see below), but the data to support this is weak (Strupp et al, 2004).. When a circulation disturbance is suspected, an agent that reduces the likelihood of stroke may be used.
Acute labyrinthitis is treated with the same medications as as vestibular neuritis, plus an antibiotic such as amoxicillin if there is evidence for a middle ear infection (otitis media), such as ear pain and an abnormal ear examination suggesting fluid, redness or pus behind the ear drum. Occasionally, especially for persons whose nausea and vomiting cannot be controlled, an admission to the hospital is made to treat dehydration with intravenous fluids. Generally admission is brief, just long enough to rehydrate the patient and start them on an effective medication to prevent vomiting.
Steroids (prednisone, methylprednisolone or decadron) were previously suggested. Strupp and others (2004) reported that steroids (methylprednisolone for 3 weeks) significantly improved the recovery of peripheral vestibular function in patients with vestibular neuritis, while valacyclovir did not. However, a meta-analysis of 4 similar studies concluded that all studies suggesting improvement had significant methodological bias, and that there is currently insufficient evidence to recommend use of steroids for treatment of vestibular neuritis (Fishman et al, 2011; Wegner et al, 2012). Furthermore, Yoo et al (2017) studied 29 patients treated with methylprednisolone and found that "In this prospective RCT, methylprednisolone had no additional benefit in patients with VN who underwent vestibular exercises and received a Ginkgo biloba". This study seems a bit underpowered to us, but still it suggests that steroids are not very helpful.
Most patients with vestibular neuritis are back to work by 2 weeks, and by 2 months are not noticing much dizziness. However, there is immense variability. There is variability in the extent of lesion (i.e. 0%-100%), variability in the recovery of function (i.e. 0%-100%), and also variability in individual compensation. Because we now have better methods of measuring vestibular function (e.g. VHIT testing), we can now document recovery much more easily.
As a general rule, persons with bigger problems (i.e. 100% loss) do worse than people with small problems (i.e. 30% loss). Roughly speaking, it takes most people about a year to recover from a 100% loss. One a nerve is "dead", recovery does not occur through restoration of function, and the "gain" of the nerve remains down forever.
Recovery of function VHIT test in January VHIT test in November
Regarding recovery of function, although not much data is available, it is generally thought that about 50% of patients recover function by one year (i.e. the damaged nerve starts to work better), and the other 50 don't. One can tell if there has been recovery of function using the VHIT test. A total loss on VHIT results in a gain of 0.5 towards the bad ear (right in the case above). Recovery is restoration of gain to 1.0. An example of good recovery is shown above. This is somewhat unusual however. In our experience, most people with a gain of 0.5, will stay that way a year later.
The plot above shows an example where the patient recovered substantially, as shown by their VHIT test, organized as a time series using our huge practice database and R-software written by Dr. Hain. Note that almost all of the recovery occurred in the first 6 months. There are also many who show no recovery at all. Recall also that the VHIT test does not measure the entire vestibular response, just the high frequency gain.
Regarding compensation, it is generally thought that nearly everyone will compensate through just activities of daily life, but those who "push", get better much faster (perhaps about 4 times faster).
Another example of a patient trajectory is above. The nerve has improved a little bit (gain from 0.63 to 0.75), but the compensation has not changed much. This person was advised to be more active.
Recovery of function VHIT test at onset VHIT test one year later. The gain is a little better, but # of predictive saccades only a little better.
Some authors have suggested that almost all of the variability in patient's symptom inventories after recovery, is due to psychological variables. We think that this misses the point -- in our opinion, almost all of the variability in symptom inventories, whether well or sick, is due to psychological variables. In other words, these surveys don't measure behavior relevant to the population, they are just individual documentation of distress.
After two-three months, testing (i.e. an ENG, audiogram, VEMP, and others) is indicated to be certain that this is indeed the correct diagnosis and a referral to a vestibular rehabilitation program, may help speed full recovery via compensation.
Bilateral vestibular neuritis (also see this link)
If a virus can affect one vestibular nerve, why not the other ? There is a fairly well recognized situation where there is vestibular neuritis on one side, and then after a fairly long gap (usually years), on the other as well, leaving the person with both ears damaged. This was first described by Schuknecht and Witt in 1985, and called "bilateral sequential vestibular neuritis". Here the diagnosis can be reasonably well established by observing two typical bouts of VN, but ending up with bilateral loss rather than recovery.
As vestibular neuritis tends to spare the inferior vestibular nerve(Goebel et al, 2001), one might expect these patients to have present "cVEMP" tests, but absent calorics and rotatory chair responses. The VHIT test provides another way to document this -- absent superior vestibular nerve responses (i.e. absent anterior canal and lateral canal), with preserved inferior vestibular nerve (i.e. Posterior canal). These situations are very unusual and the VHIT test is not very reliable either, when considering the vertical canals.
The situation where both ears are "taken out" at the same time, also seems plausible, but difficult to prove. One would think that this would result in an "idiopathic" bilateral loss presentation. There are indeed a very substantial number of "idiopathic" bilateral vestibular loss. However, as the causal diagnosis presumably would require an autopsy, it does not seem likely that we will clear this up anytime soon. This is a "medical hypothesis". Again, VHIT testing or the combination of a lateral canal test (calorics or R-chair) with a cVEMP, might be a way to make this inference.
One would expect that those with preserved portions of their vestibular system would do better, long term, than those with a total "wipe out". So there is some rehab implication.
Recurrent vestibular neuritis -- the real thing vs central causes of similar symptoms such as Benign recurrent vertigo (BRV)
Fortunately, in the great majority of cases (at least 95%) vestibular neuritis it is a one-time experience. Rarely (5%) the syndrome is recurrent, coming back at least once, and sometimes year after year. When there is clear evidence of vestibular nerve damage, it is still simply called "recurrent vestibular neuritis.When it is recurrent, but there is normal vestibular function, the same symptom complex may generate other potential diagnoses. A potential game-changer for this diagnosis are the recent availability of strong tests of the vestibular nerve, namely the HIT/VHIT tests.
For situations where there is dizziness but lacking strong evidence of vestibular nerve damage (i.e. normal HIT/VHIT test), other possibilities include benign paroxysmal vertigo in children (Basser, 1964), benign recurrent vertigo (Slater 1979, Moretti et al, 1980), or vestibular Meniere's syndrome (Rassekh and Harker, 1992).
More simply, these can all be boiled down into a catch-all of migraine associated vertigo. As is the usual case for almost anything having "migraine" in it's name, nearly anything goes, because migraine is a "wastebasket syndrome" that has no specific objective findings. One does not even have to have a headache. It may be familial for example (Oh et al, 2001). Rather than lumping BRV into migraine, it may instead be an entity by itself (Lee et al, 2006) but lacking any clear diagnostic findings that distinguish it from recurrent vestibular neuritis or acephalgic migraine. So in essence, this is a symptom complex without a clear evidence base establishing mechanism.
When labyrinthitis recurs (i.e. there is hearing and dizziness that recurs), the diagnosis is often changed from labyrinthitis to "Meniere's disease". The reason for this is that the diagnostic criteria for Meniere's are essentially those of recurrent labyrinthitis. It is the author's impression that this "conversion" process occurs far more commonly than there is recurrent vestibular neuritis.
Another recurrence pattern in vestibular neuritis is the "quick spin" pattern - -people complain of brief spells lasting seconds to minutes in which the entire world rotates at high speed, then stops, without any hearing symptoms. This may occur as often as 50 times/day. This pattern of dizziness often responds to anticonvulsants such as carbamazepine or oxcarbamazine, and in these situations, may reasonably be attributed to vestibular paroxysmia. . In this disorder, one can often recognize the patient the video-frenzel goggles. There is a paretic type spontaneous nystagmus and vibration induced nystagmus, which reverses with hyperventilation for 30 seconds.
Case example of quick spins: A middle aged administrator complained of multiple spells of spinning vertigo with nausea unaccompanied by hearing symptoms. The spells lasted 10-20 minutes, were accompanied by sweating and nausea. He has had times in which he has had three or four episodes per day. There appear to be no consistent triggers. On examination, a right-beating spontaneous nystagmus was observed. This reversed direction with hyperventilation. Hearing testing was normal as was ENG testing and MRI scan. After being started on oxcarbamazine, gradually increasing to 600 mg twice/day, his spells decreased to less than once/two weeks, and were minimal in intensity.
For the most common type of unilateral vestibular neuritis, you will probably be unable to work for one or two weeks. You may be left with some minor sensitivity to head motion which will persist for several years, and may reduce your ability to perform athletic activities such as racquetball, volleyball and similar activities. After the acute phase is over, for a moderate deficit, falls are no more likely than in persons of your age without vestibular deficit (Herdman et al, 2000). Persons in certain occupations, such as pilots, may have a greater long term impact (Shupak et al, 2003).
This may sound obvious, but oddly enough not everyone agrees about this, but the duration of your symptoms depends on the severity of damage. If you have a "unilateral wipe out", nothing left, you will not do as well as someone with a mild unilateral vestibular weakness. Other important variables are how hard you "push" yourself to recover (it is better to push), and how anxious you get about this condition. Unfortunately, sometimes the health care system ends up focusing attention on illness rather than wellness. In other words, it is usually best to push to recover a normal life, and not make caring for your vestibular neuritis your "new career". This may mean saying "no Thank you" to some well-meaning health care providers, who want to be your personal trainer for vestibular issues.
You may also have mild problems with your thinking. Even in persons who are well compensated, sensory integration seems to require more attention in persons with vestibular lesions than normal subjects (Redfern et al, 2003).
Acknowledgment: The graphic of figure 1 was originally funded by NIH.