2010年10月17日 星期日




Timothy C. Hain, MD

Last substantial content edit: 2/2003. Please read our disclaimer.

This page is no longer being updated. Click HERE to go to the more recent version.

Causes Diagnosis Treatment Education IndexSearch this site


Ear RocksIn Benign Paroxysmal Positional Vertigo (BPPV) dizziness is thought to be due to debris which has collected within a part of the inner ear.  This debris can be thought of  as "ear rocks", although the formal name is "otoconia". Ear rocks are small crystals of calcium carbonate derived from a structure in the ear called the "utricle" (figure1 ). While the saccule also contains otoconia, they are not able to migrate into the canal system. The utricle may have been damaged by head injury, infection, or other disorder of the inner ear, or may have degenerated because of advanced age. Normally otoconia appear to have a slow turnover. They are probably dissolved naturally as well as actively reabsorbed by the "dark cells" of the labyrinth (Lim, 1973, 1984), which are found adjacent to the utricle and the crista, although this idea is not accepted by all (see Zucca, 1998, and Buckingham, 1999).

BPPV is a common cause of dizziness. About 20% of all dizziness is due to BPPV. The older you are, the more likely it is that your dizziness is due to BPPV, as about 50% of all dizziness in older people is due to BPPV. In a recent study, 9% of a group of urban dwelling elders were found to have undiagnosed BPPV (Oghalai, J. S., et al., 2000).

The symptoms of BPPV include dizziness or vertigo, lightheadedness, imbalance, and nausea. Activities which bring on symptoms will vary among persons, but symptoms are almost always precipitated by a change of position of the head with respect to gravity. Getting out of bed or rolling over in bed are common "problem" motions . Because people with BPPV often feel dizzy and unsteady when they tip their heads back to look up, sometimes BPPV is called "top shelf vertigo." Women with BPPV may find that the use of shampoo bowls in beauty parlors brings on symptoms. An intermittent pattern is common. BPPV may be present for a few weeks, then stop, then come back again.


The most common cause of BPPV in people under age 50 is head injury . There is also an association with migraine (Ishiyama et al, 2000). In older people, the most common cause is degeneration of the vestibular system of the inner ear. BPPV becomes much more common with advancing age (Froeling et al, 1991). In half of all cases, BPPV is called "idiopathic," which means it occurs for no known reason. Viruses affecting the ear such as those causing vestibular neuritis , minor strokes such as those involving anterior inferior cerebellar artery (AICA) syndrome", and Meniere's disease are significant but unusual causes. Occasionally BPPV follows surgery, where the cause is felt to be a combination of a prolonged period of supine positioning, or ear trauma when the surgery is to the inner ear (Atacan et al 2001). Other causes of positional symptoms are discussed here.

What doesn't cause BPPV ?

Gacek has suggested that BPPV is due to recurrent neuritis of the inferior vestibular nerve (Gacek and Gacek, 2002). We think that this is highly unlikely as BPPV is very well explained by mechanical consequences of loose debris within the inner ear, and not at all consistent with the usual picture of vestibular neuritis. BPPV is also not caused by psychological distress, and it is not a side effect of medication.


Your physician can make the diagnosis based on your history, findings on physical examination, and the results of vestibular and auditory tests. Often, the diagnosis can be made with history and physical examination. Most other conditions that have positional dizziness get worse on standing rather than lying down (e.g. orthostatic hypotension). Electronystagmography (ENG) testing may be needed to look for the characteristic nystagmus (jumping of the eyes). It has been claimed that BPPV accompanied by unilateral lateral canal paralysis is suggestive of a vascular etiology (Kim et al, 1999). For diagnosis of BPPV with laboratory tests, it is important to have the ENG test done by a laboratory that can measure vertical eye movements. A magnetic resonance imaging (MRI) scan will be performed if a stroke or brain tumor is suspected. A rotatory chair test may be used for difficult diagnostic problems. It is possible but rather uncommon to have BPPV in both ears (bilateral BPPV).

There are some rare conditions that have symptoms that resemble BPPV. Patients with certain types of central vertigo such as the spinocerebellar ataxias may have "bed spins" and prefer to sleep propped up in bed (Jen et al, 1998). These conditions can generally be detected on a careful neurological examination and also are generally accompanied by a family history of other persons with similar symptoms.


Certain modifications in your daily activities may be necessary to cope with your dizziness. Use two or more pillows at night. Avoid sleeping on the "bad" side. In the morning, get up slowly and sit on the edge of the bed for a minute. Avoid bending down to pick up things, and extending the head, such as to get something out of a cabinet. Be careful when at the dentist's office, the beauty parlor when lying back having ones hair washed, when participating in sports activities and when you are lying flat on your back.


BPPV has often been described as "self-limiting" because symptoms often subside or disappear within six months of onset. Symptoms tend to wax and wane. Motion sickness medications are sometimes helpful in controlling the nausea associated with BPPV but are otherwise rarely beneficial. However, various kinds of physical maneuvers and exercises have proved effective. Three varieties of conservative treatment, which involve exercises, and a treatment that involves surgery are described in the next sections.

OFFICE TREATMENT OF BPPV: The Epley and Semont Maneuvers

Epley ManeuverThere are two treatments of BPPV that are usually performed in the doctor's office. Both treatments are very effective, with roughly an 80% cure rate, according to a study by Herdman and others (1993). If your doctor is unfamiliar with these treatments, you can find a list of knowledgeable doctors from the Vestibular Disorders Association (VEDA) .

The maneuvers, named after their inventors, are both intended to move debris or "ear rocks" out of the sensitive part of the ear (posterior canal) to a less sensitive location. Each maneuver takes about 15 minutes to complete. The Semont maneuver (also called the "liberatory" maneuver) involves a procedure whereby the patient is rapidly moved from lying on one side to lying on the other. It is a brisk maneuver that is not currently favored in the United States.

The Epley maneuver is also called the particle repositioning, canalith repositioning procedure, and modified liberatory maneuver. It is illustrated in figure 2. Click here for an animation. It involves sequential movement of the head into four positions, staying in each position for roughly 30 seconds. The recurrence rate for BPPV after these maneuvers is about 30 percent at one year, and in some instances a second treatment may be necessary. While some authors advocate use of vibration in the Epley maneuver, we have not found this useful in a study of our patients (Hain et al, 2000). Some authors also suggest leaving out some of the positions in the Epley maneuver, especially position 'D'. We suggest that you avoid therapy using this methodology.

After either of these maneuvers, you should be prepared to follow the instructions below, which are aimed at reducing the chance that debris might fall back into the sensitive back part of the ear.


1. Wait for 10 minutes after the maneuver is performed before going home. This is to avoid "quick spins," or brief bursts of vertigo as debris repositions itself immediately after the maneuver. Don't drive yourself home.

epley45.gif (6379 bytes)2. Sleep semi-recumbent for the next two nights. This means sleep with your head halfway between being flat and upright (a 45 degree angle). This is most easily done by using a recliner chair or by using pillows arranged on a couch (see figure 3). During the day, try to keep your head vertical. You must not go to the hairdresser or dentist. No exercise which requires head movement. When men shave under their chins, they should bend their bodies forward in order to keep their head vertical. If eyedrops are required, try to put them in without tilting the head back. Shampoo only under the shower.

3. For at least one week, avoid provoking head positions that might bring BPPV on again.

  • Use two pillows when you sleep.
  • Avoid sleeping on the "bad" side.
  • Don't turn your head far up or far down.

Be careful to avoid head-extended position, in which you are lying on your back, especially with your head turned towards the affected side. This means be cautious at the beauty parlor, dentist's office, and while undergoing minor surgery. Try  to stay as upright as possible. Exercises for low-back pain should be stopped for a week. No "sit-ups" should be done for at least one week and no "crawl" swimming. (Breast stroke is OK.) Also avoid far head-forward positions such as might occur in certain exercises (i.e. touching the toes). Do not start doing the Brandt-Daroff exercises immediately or 2 days after the Epley or Semont maneuver, unless specifically instructed otherwise by your health care provider.

4. At one week after treatment, put yourself in the position that usually makes you dizzy. Position yourself cautiously and under conditions in which you can't fall or hurt yourself. Let your doctor know how you did.

Comment: Massoud and Ireland (1996) stated that post-treatment instructions were not necessary. While we respect these authors, at this writing (2002), we still feel it best to follow the procedure recommended by Epley.


These maneuvers are effective in about 80% of patients with BPPV (Herdman et al, 1993). If you are among the other 20 percent,  your doctor may wish you to proceed with the Brandt-Daroff exercises, as described below. If a maneuver works but symptoms recur or the response is only partial (about 40% of the time according to Smouha, 1997), another trial of the maneuver might be advised. The "habituation" exercises are also sometimes useful in the situation where all other maneuvers (Epley, Semont, Brandt-Daroff) have been tried -- in essence these consist of a more intense and prolonged series of positional exercises. When all maneuvers have been tried, the diagnosis is clear, and symptoms are still intolerable, surgical management (posterior canal plugging) may be offered.

BPPV often recurs. About 1/3 of patients have a recurrence in the first year after treatment, and by five years, about half of all patients have a recurrence (Hain et al, 2000; Nunez et al; 2000). If BPPV recurs, in our practice we usually retreat with one of the maneuvers above, and then follow this with a once/day set of the Brandt-Daroff exercises.

In some persons, the positional vertigo can be eliminated but imbalance persists. In these persons it may be reasonable to undertake a course of generic vestibular rehabilitation, as they may still need to compensate for a changed utricular mass or a component of persistent vertigo caused by cupulolithiasis. Fujino et al (1994) reported conventional rehab has some efficacy, even without specific maneuvers.


BRANDT-DAROFF EXERCISES Brandt-Daroff Exercises for BPPV

Click here for an animation

The Brandt-Daroff Exercises are a method of treating BPPV, usually used when the office treatment fails. They succeed in 95% of cases but are more arduous than the office treatments. These exercises are performed in three sets per day for two weeks. In each set, one performs the maneuver as shown five times.

1 repetition = maneuver done to each side in turn (takes 2 minutes)

Suggested Schedule for Brandt-Daroff exercises
Morning5 repetitions10 minutes
Noon5 repetitions10 minutes
Evening5 repetitions10 minutes

Start sitting upright (position 1). Then move into the side-lying position (position 2), with the head angled upward about halfway. An easy way to remember this is to imagine someone standing about 6 feet in front of you, and just keep looking at their head at all times. Stay in the side-lying position for 30 seconds, or until the dizziness subsides if this is longer, then go back to the sitting position (position 3). Stay there for 30 seconds, and then go to the opposite side (position 4) and follow the same routine..

These exercises should be performed for two weeks, three times per day, or for three weeks, twice per day. This adds up to 52 sets in total. In most persons, complete relief from symptoms is obtained after 30 sets, or about 10 days. In approximately 30 percent of patients, BPPV will recur within one year. If BPPV recurs, you may wish to add one 10-minute exercise to your daily routine (Amin et al, 1999). The Brandt-Daroff exercises as well as the Semont and Epley maneuvers are compared in an article by Brandt (1994), listed in the reference section.

Home Epley Left
Home Epley (for the left ear).



The Epley and/or Semont maneuvers as described above can be done at home (Radke et al, 1999; Furman and Hain, 2004). We often recommend the home-Epley to our patients who have a clear diagnosis. This procedure seems to be even more effective than the in-office procedure, perhaps because it is repeated every night for a week.

The method (for the left side) is performed as shown on the figure to the right. One stays in each of the supine (lying down) positions for 30 seconds, and in the sitting upright position (top) for 1 minute. Thus, once cycle takes 2 1/2 minutes. Typically 3 cycles are performed just prior to going to sleep. It is best to do them at night rather than in the morning or midday, as if one becomes dizzy following the exercises, then it can resolve while one is sleeping. The mirror image of this procedure is used for the right ear.

There are several problems with the "do it yourself" method. If the diagnosis of BPPV has not been confirmed, one may be attempting to treat another condition (such as a brain tumor or stroke) with positional exercises -- this is unlikely to be successful and may delay proper treatment. A second problem is that the home-Epley requires knowledge of the "bad" side. Sometimes this can be tricky to establish. Complications such as conversion to another canal (see below) can occur during the Epley maneuver, which are better handled in a doctor's office than at home. Finally, occasionally during the Epley maneuver neurological symptoms are provoked due to compression of the vertebral arteries. In our opinion, it is safer to have the first Epley performed in a doctors office where appropriate action can be taken in this eventuality.

sitecdWe offer a home treatment DVD that illustrates the home Epley exercises.




If the exercises described above are ineffective in controlling symptoms, symptoms have persisted for a year or longer,  and the diagnosis is very clear, a surgical procedure called "posterior canal plugging" may be recommended. Canal plugging blocks most of the posterior canal's function without affecting the functions of the other canals or parts of the ear. This procedure poses a small risk to hearing, but is effective in about 90% of individuals who have had no response to any other treatment. Only about 1 percent of our BPPV patients eventually have this procedure done.  Surgery should not be considered until all three maneuvers/exercises (Epley, Semont, and Brandt-Daroff) have been attempted and failed. See the article by Parnes (1990, 1996) in the references for more information.

There are several alternative surgeries. Dr Gacek (Syracuse, New York) has written extensively about singular nerve section. Dr. Anthony (Houston, Texas), advocates laser assisted posterior canal plugging. It seems to us that these procedures, which require unusual amounts of surgical skill, have little advantage over a canal plugging procedure. Of course, it is always advisable when planning surgery to select a surgeon who has had as wide an experience as possible.Complications are rare (Rizvi and Gauthier, 2002)

There are several surgical procedures that we feel are inadvisable for the individual with intractable BPPV. Vestibular nerve section, while effective, eliminates more of the normal vestibular system than is necessary. Labyrinthectomy and sacculotomy are also both generally inappropriate because of  reduction or loss of hearing expected with these procedures.


Lateral Canal BPPV, Anterior Canal BPPV, Cupulolithiasis, Vestibulolithiasis, Multicanal patterns

There are several rarer variants of BPPV which may occur spontaneously as well as after the Brandt-Daroff maneuvers or Epley/Semont maneuvers. They are mainly thought to be caused by migration of otoconial debris into canals other than the posterior canal, the anterior or lateral canal. There is presently no data reported as to the frequency and extent of these syndromes following treatment procedures. It is the author's estimate that they occur in roughly 5% of Epley maneuvers and about 10% of the time after the Brandt-Daroff exercises. In nearly all instances, with the exception of cupulolithiasis, these variants of BPPV following maneuvers resolve within a week without any special treatment, but when they do not, there are procedures available to treat them.

In clinical practice, atypical BPPV arising spontaneously is first treated with maneuvers as is typical BPPV, and the special treatments as outlined below are entered into only after treatment failure. When atypical BPPV follows the Epley, Semont or Brandt-Daroff maneuvers, specific exercises are generally begun as soon as the diagnosis is ascertained. In patients in whom the exercise treatment of atypical BPPV fails, especially in situations where onset is spontaneous, additional diagnostic testing such as MRI scanning may be indicated. The reason for this is to look for other types of positional vertigo.

Lateral canal BPPV is the most common atypical BPPV variant, accounting for about 3-9 percent of cases (Korres et al, 2002). Most cases are seen as a consequence of an Epley maneuver. It is diagnosed by a horizontal nystagmus that changes direction according to the ear that is down. More detail about lateral canal BPPV as well as an illustration of a home exercise can be found here.

Anterior canal BPPV is also rare, and a recent study suggested that it accounts for about 2% of cases of BPPV (Korres et al, 2002). It is diagnosed by a positional nystagmus with components of downbeating and torsional movement on taking up the Dix-Hallpike position, or a nystagmus that is upbeating and torsional when sitting up from the Dix-Hallpike. There are a number of different suggestions in the literature about the direction of the torsional quick phase in anterior canal BPPV. In our view, the nystagmus during the Dix-Hallpike to one side is most likely due to excitation of the anterior canal on the opposite side. This should cause downbeating nystagmus as well as torsional nystagmus with a quick-phase towards the disturbed ear. Thus the direction of the torsional component during the down-phase of the Dix-Hallpike tells you which is the bad ear. Anterior canal BPPV can be provoked from the opposite ear to the side of the Dix-Hallpike maneuver -- in other words, if you get dizzy to the right side, the problem ear might be the left. Some authors have suggested that because the anterior canals are oriented so that parts are near the saggital plane, anterior canal BPPV can be provoked with a Dix-Hallpike maneuver to either side as well as in the "head hanging" position (Bertholon et al, 2002). The upbeating nystagmus on sitting may be very persistent as the debris settles on the cupula of the anterior canal. Anterior canal BPPV is probably rare because the anterior canal is normally the highest part of the ear. Debris would naturally tend to fall out of the posterior half of the anterior canal. From the geometry of the ear, it would seem likely that anterior canal BPPV might occasionally result as a complication of the Epley maneuver.

Debris might also be temporarily located in the common crus area, which is the shared canal between the anterior and posterior canal. Should debris be present in the common cruse, one would expect a purely torsional nystagmus. During the down phase of the Dix-Hallpike, the torsional nystagmus should beat away from the bad ear. During the up phase of the Dix-Hallpike, the torsional nystagmus should beat towards the bad ear.

Cupulolithiasis is a condition in which debris is stuck to the cupula of a semicircular canal, rather than being loose within the canal. Cupulolithiasis is not a treatment complication, but rather is part of the spectrum of BPPV. The mechanistic hypothesis is based on pathological findings of deposits on the cupula made by Schuknecht and Ruby in three patients who had BPPV during their lives (Schuknecht 1969; Schuknecht et al. 1973). Moriarty and colleagues found similar deposits in 28% of 566 temporal bones (Moriarty et al. 1992). Schuknecht pointed out that cupulolithiasis hypothesis fails to explain the usual characteristic latency and burst pattern of BPPV nystagmus as well as remissions (Schuknecht et al. 1973). Rather, cupulolithiasis should result in a constant nystagmus. This pattern is sometimes seen (Smouha et al. 1995). Cupulolithiasis might theoretically occur in any canal -- horizontal, anterior or vertical, each of which might have it's own pattern of positional nystagmus. Some authors hold that both the cupulolithiasis and canalithiasis hypotheses may be correct (Brandt et al. 1994). If cupulolithiasis is suspected, it seems logical to treat with either the Epley with vibration, or alternatively, use the Semont maneuver. There are no studies of cupulolithiasis to indicate which strategy is the most effective.

Vestibulolithiasis is a hypothetical condition in which debris is present on the vestibule-side of the cupula, rather than being on the canal side. For this theory, there is loose debris, close to but unattached to the cupula of the posterior canal, possibly in the vestibule or short arm of the semicircular canal. Pathologic studies of BPPV have found roughly equal amounts of fixed debris on either side of the cupula (Moriarty et al. 1992), suggesting that loose debris might also be found on either side. For the vestibulolithiasis mechanism, when the head is moved, stones or other debris might shift from vestibule to ampulla, or within the ampulla, impacting the cupula. This mechanism would be expected to resemble cupulolithiasis, having a persistent nystagmus, but with intermittency because the debris is movable. Very little data is available as to the frequency of this pattern, and no data is available regarding treatment.

Multicanal patterns. If debris can get into one canal, why shouldn't it be able to get into more than one ? It is common to find small amounts of horizontal nystagmus or contralateral downbeating nystagmus in a person with classic posterior canal BPPV. While other explanations are possible, the most likely one is that there is debris in multiple canals.


The Vestibular Disorders Association (VEDA) maintains a large and comprehensive list of doctors who have indicated a proficiency in treating BPPV. Please contact them to find a local treating doctor.


Literature-and DVDs


Click here for very recent, but possibly less relevant references.

Search this site

Published literature referred to above:


(c) 1997-2005 Timothy C. Hain, thain@dizziness-and-balance.com