Timothy C. Hain, MD. Page last modified: February 4, 2018
See also: Case of vertigo in CI.
|Cochlear implant from a patient of Dr. Hain's medical practice in Chicago (Chicago Dizziness and Hearing).||Cochlear implant (from Wikipedia). Note that the electrode on this picture traverses the round window.|
An exciting recent development is an ability to provide hearing to some bilaterally deafened individuals through implantation of a device which directly stimulates the hearing nerve (actually the spiral ganglion). Generally an electrode is threaded into the cochlea, often via the round window. This is called a cochlear implant, CI for short. The electrode can also be inserted through the bony capsule of the cochlea - -called a "cochleostomy".
A meta-analysis from 2018 (Hansel et al) concluded that of 116 studies, "Vertigo after cochlear implantation was reported in 9.3% of the patients with a continuous increase in patient age at surgery." We are surprised that the prevalence is so low ! In fact, we are dubious that this is correct. We wonder about the potential for bias, which is generally high, and would seem particularly high in reports of this nature.
More reasonably, vertigo or imbalance occurs in 25-74% of patients who have cochlear implants (Steenerson et al, 2001 and others below). Kubo et al (2001) suggested that about half of all CI patients develop dizziness, but in most it resolved within one month. Batuecas-Caletrio and associates (2015) reported that only 30% of CI patients have a "postoperative change in vestibular function", using a rather broad screening process.
We ourselves would expect that 100% of CI patients would have some imbalance or vertigo, as this is an invasive surgery that creates a hole in the round window. However, the benefits to hearing greatly outweigh the impact of the disturbance to balance and mild dizziness, in most.
As an overview, this is a very difficult diagnostic situation as there are too many potential causes of dizziness in people who have severe inner ear disease, who have then gotten an electrode inserted into one of their inner ears.
From the literature, nearly any inner ear problem that causes dizziness has been postulated to be a source of dizziness post implant.
- Vertigo from the disorder that caused hearing to decline in the first place.
- BPPV post implant.
- Insertional damage to the inner ear, including hydrops, and inadvertent puncture of the membrane between the scala tympani and vestibuli.
- Electrical stimulation of saccule or elsewhere (Cordes et al, 2012)
- Leakage around cochleostomy through round window (Kusuma et al, 2005)
- Sympathetic inner ear damage (Harris et al, 1985)
BPPV is easy to spot and treat. Vertigo from electrical stimulation is also easy to spot as it only occurs when the implant is activated. Misdirected implants can be seen on a CT scan of the inner ear. Otherwise, as hearing cannot be followed post CI, the diagnostic process is largely based on trial/error, and may not end up with a clear answer. Similarly, MRI scans post implant are generally impossible and cannot be used. This is a problem.
What can be done -- Chen et al (2016) found reduced VEMP tests and reduced caloric tests in CI patients, with the VEMP being most sensitive. Abouzayd et al (2016) suggested that patients with CI should have caloric tests, cervical-VEMPS, and head impulse tests, but added that that sensitivity of all of these tests was low. The cVEMP test was most often impaired (Robard et al, 2015). This is unsurprising considering that in CI the electrode is often threaded through the middle ear (i.e. cVEMP testing might be reflecting damage to the middle ear rather than vestibular damage).
In our view, a careful history and clinical examination including video Frenzel goggles would seem to us to be the most suitable diagnostic process, preferably combined with a vHIT test and imaging of the implant, ideally with a temporal bone CT scan. This is not something that one could usually do, lets say, at a community hospital. It intrinsically requires expertise, experience and resources.
Vestibular responses) are not generally not eliminated by cochlear implants, and thus the underlying process that affected hearing in the first place can continue to cause dizziness and balance.
Common causes of hearing loss requiring implants include
- age-related (presbyacusis)
- autoimmune inner ear disease (AIED)
- Meniere's disease
- idiopathic hearing loss
If one is old enough to develop age-related hearing loss, one is also old enough to have a much higher prevalence of BPPV.
AIED, in our experience, is nearly impossible to diagnose with certainty, and the diagnostic space that it occupies overlaps with Menieres, congenital and idiopathic. We would be more likely though to attribute bilateral hearing loss, without vertigo, to autoimmune. In other words, the label used to describe progressive hearing loss, is a "self fulfilling prophesy" when one considers vertigo.
Meniere's disease, of course, requires vertigo to be diagnosed, and as the CI process does not treat Meniere's disease as a whole, one would certainly expect continued vertigo after the CI.
The approach to this problem is that one needs to understand the chance of vertigo from the disease, look for new patterns after implant, make an educated guess, and attempt to treat.
BPPV, or benign paroxysmal positional vertigo, is one of the most common cause of vertigo, essentially because it is a vulnerable system in the inner ear. One would expect that it would be very common post CI.
Reports vary as to how often CI patients develop positional vertigo. Di Girolamo et al (1999) suggested that it is "unusual". Limb et al (2005) reported that BPPV occurred in 12 out of 52 CI patients, thus roughly 20%. Viccaro (2007) reported it in 8 out of 70 patients -- roughly 10%. Shetye (2012) suggested that it was "infrequent". Zanetti (2007) observed it in 4 of 62 patients. Thus, as of 2015, it appears that roughly 10-20% of patients develop BPPV post cochlear implant. We find it surprising that there are not more.
Limb et al (2005) proposed two theories -- introduction of bone dust into the labyrinth, and dislodging of otoconia during surgery.
BPPV associated with CI is treated similarly to BPPV that arises due to other ear damage, generally with good success.
When one inserts a cochlear implant, a long wire is threaded through the round window. Something like trying to insert a coat-hanger through a small hole and snag something that you can't see on the other side. This is a small area, and while one can figure out if it made it to the cochlea afterwards, one doesn't necessarily know if it ran into something on the way in. There are some things in the way (like the saccule, or the scala tympani) that might get punctured. We know from previous attempts to puncture the saccule to treat Meniere's (the "Cody tack") that this can do a lot of damage. We also know that membrane ruptures can cause dizziness.
Basta et al (2008) suggested that " chronic, persisting dizziness after cochlear implant surgery is largely based on a dysfunction of the saccular macula which is an integral component of the otolith system." While we accept that the saccule is often damaged during insertion, we are dubious that this causes chronic persistent vertigo, as saccule function (as measured by VEMP testing) is routinely lost in older individuals, who generally do not have "chronic persistent vertigo". Basta also suggested "A possible coactivation of the inferior vestibular nerve by the electrical stimulation might play an additional role in the pathogenesis of the persisting postsurgical dizziness." We are also dubious about this idea, as if so, the symptoms should vanish after the implant is turned off.
Lesinski (1998) reported a case of Tullio's, attributed to "scar tissue surrounding the ossicles after CI". They treated the Tullio's by disarticulating the ossicles. This would suggest that sound may have been stimulating a displaced saccule.
Kubo et al (2001) reported that about 15% of their patients developed delayed vertigo, accompanied by tinnitus and hearing loss, and resembled Meniere's disease. This suggests that some patients develop hydrops post implant. There is some recent support for this idea as patients with CI can lose low frequency hearing, and some recent studies suggest obliteration of the ductus reunions. See also the discussion below under the heading of "leaks" about how perilymph fistula might be confused with hydrops.
Vestibular function, somewhat surprisingly, is generally only modestly reduced in the implanted ear. (Brey et al, 1995). In other words, most cases of dizziness post implant do not seem to be due to "vestibular shutdown" of the ear in question.
Occasional patients develop meningitis post-implant. One would think that there might both be infections as well as non-infectious (sterile) meningitis, or simply inflammation of the labyrinthine system.
Although one might think that dizziness could be related to inadvertent electrical stimulation of the vestibular system, dizziness is not usually related to implant activation, (Fina et al, 2003). Cordes (2012) suggested that dizziness that depends on activation of the CI was mainly related to stimulation of the saccule, which is the vestibular structure that is must be traversed for the electrode to get to the cochlea. Electrode migration, according to van der Marel et al (2012) is common. When the electrode migrates, it may start to stimulate the saccule. Tange et al (2006) reported a patient where the implant was misdirected into the vestibular part of the inner ear, and who of course developed vertigo. This situation can be easily diagnosed from a post-surgical scan.
Although electrical stimulation can cause vertigo, practically, it is very rare that balance is affected by the electrical stimulation. In other words, vertigo is not generally triggered by "turning on" the implant. In fact, the added auditory information may be a little helpful (i.e. most of the time), according to an uncontrolled study of Shayman et al, 2017. We doubt that this effect is very significant however.
Electrodes can "fold up" during insertion (Pijl et al, 2008), and of course, this requires reimplantation. This situation is easily recognized by a failure of the implant to work, and an abnormal CT scan.
The wire for cochlear implants goes through the round window (see figure above), or through the cochlea through a "cochleostomy". Trauma to the round window can cause a perilymphatic fistula, which is a potential source of dizziness. Just looking at the mechanics of the situation, one would think this would be a huge cause of dizziness in CI patients. Choo (1984) pointed out that CI's run the risk of creating leaks in the round window.
One would think that there would be many leaks through the cochleostomy site in the round window post implant causing dizziness, as we invoke the spector of perilymphatic fistula to explain dizziness in persons who have had far less inner ear trauma than CI, but oddly enough, this has not been reported often. This may be due to the huge problem we have with detecting perilymph leaks in general, or that there is something special about CI's and leaks don't occur. In other words leaks are not happening (doubtful), or they are simply being missed (could be) or they are being ignored. At this writing (2015), we do not have a reasonable method to detect a chronic leak through the round window, so there is no way to resolve this question.
The typical signs of PLF are pressure and motion triggered dizziness - -i.e. dizziness on straining, or dizziness with loud noises due to activation of the ossicular chain. One would think that the first could be present, but the latter would be unusual in CI patients. Dizziness on straining can be detected with the Valsalva test. Our experience is that many ear conditions have a mildly positive Valsalva, so the sensitivity of this procedure in CI patients is likely small.
Modestly arguing against the hypothesis that dizziness in CI is due to a chronic round window leak is that Todt et al(2008) suggested that while dizziness was common post implant, but that insertion through the round window decreased the risk of dizziness associated with insertion.
Of course, air in the labyrinth on MRI (pneuomolabyrinth) is a sure sign of a perilymph leak. Rather than going out to do CT or MRI scans in every patient who has dizziness, it should be enough to simply examine the patient as in our experience, the air bubble patients have considerable positional nystagmus. Nevertheless, pneuomlabyrinth is extremely rare, and the absence of air does not prove the absence of a leak. (Venegas et al, 2016)
Leakage of perilymph reduces perilymph pressure, and could cause symptoms similar to Meniere's disease. As the problem is structural, salt restriction or diuretics would not be likely to help.
Karimi et al (2017) reported that "resealing" of the cochleostomy improved vertigo in 6 of 9 patients. Our thought is that there are many potential sources of dizziness in CI, and that this is likely one of them. The problem is in determining which patient will benefit from another surgery. It would be difficult to do a placebo controlled study for this problem. The newer, high-resolution CT scanners (cone beam) might provide some help.
Regarding autoimmune inner ear disease or hydrops, these conditions are difficult to identify in persons with their original hearing equipment. It is nearly impossible to diagnose them after CI. The idea is that trauma to the ear releases antigen from a protected site, and that after a few years, the body mounts a defense against itself and causes more damage. Current diagnostic technology is not able to detect this situation. It is especially hard because CI patients nearly always are deaf on both ears, and have no more hearing to lose from autoimmune sources. However, they might develop vertigo. Another unresolved issue.
Treatment is basically trial/error. One tries turning off the implant, one explores the effect of pressure (for leakage), one tries out treatments for hydrops, and if there is positional nystagmus, treat with PT. If there are headaches, one treats for migraine too. Vestibular rehab is unlikely to make the situation worse, and is generally worth a try.
According to Tutar et al (2014), transmastoid labyrinthectomy can be used to remove remaining vestibular function in an implanted ear. As this was a single case report, it seems to us that were it us, we would hope that someone else would blaze the trail to be #2, as this seems to us to be risky business.