Treatment of benign paroxysmal positional vertigo.

Treatment of Benign Paroxysmal Positional Vertigo Peripheral vestibular disorders result in vertigo, disequilibrium, and frequently nausea and vomiting. The purpose of this article is to describe the physical therapy management of one of the more common peripheral vestibular disorders--benign paroxysmal positional vertigo (BPPV). Several different approaches have been used in the treatment of BPPV. In this article, these approahces are compared, and the criteria used in choosing the appropriate approach are presented. Case studies are used to illustrate the different treatment approaches.

Characteristics of Benign

Paroxysmal Positional Vertigo

The diagnosis of BPPV is based on certain characteristic clinical findings. [1-3] Patients with BPPV experience vertigo when moved rapidly into a supine position with the head turned so that the affected ear is 30 to 45 degrees below the horizontal (Fig 1). The vertigo occurs with a latency of 1 to 40 seconds after the patient has been placed in the provoking position (usually after 1-5 seconds). Patients also develop a characterisctic nystagmus, which is torsional with the eyes directed toward the affected side and becomes more vertical ("upbeating") when the eyes are directed away from the affected side. The vertigo and the nystagmus increase in intensity and then disappear in 30 to 60 seconds. The response usually fatiques if the patient is placed, although this is a variable position, although this is a variable finding, occurring in only 87% of the cases. [3]

Rapidly positioning the patient so that the affected ear is approximately 30 degrees below the horizontal (Hallpike-Dix maneuver) (Fig 1) results in an ampullofugal deflection of the cupula of the posterior canal. Neurons innervating the ipsilateral superior oblique muscle and the contralateral inferior rectus muscle are excited, resulting in a torsional nystagmus. The Hallpike-Dix maneuver commonly produces the vertigo and nystagmus of BPPV, but any movement that excites the posterior semi-circular canal (eg, tilting the head back) may trigger a similar, although smaller, response. In BPPV, it is unusual for vertigo to be induced by movements that excite the anterior or the horizontal canals such as learning forward or turning the head in a horizontal, plane, although horizontal-canal benign paroxysmal vertigo has been reported. [3,4]

Patients with bPPV report a fairly characteristic history. Typically, they experience a sudded onset of Vertigo when rolling over in bed. The duration of the vertigo often cannot be described because the patient quickly moves out of the provoking position and then avoids that position for fear of initiating that other movements precipitate the vertigo, such as straightening up after bending over or looking up suddenly as when reaching for an object on a high shelf. Patients may describe a single occurrence of vertigo, or they may have a long history of episodic vertigo intermixed with periods in which they are symptom-free. Patients with BPPV typically have periods of remission lasting from months to years during which they experience no episodes of vertigo.

Patients with BPPV also may complain of disequilibrium, and they may have abnormal postural responses. Black and Nashner [5] found decreased postural stability in patients with BPPV and suggested that these patients rely excessively on visual cues to maintain balance. A more recent study by Voorhees [6] failed to reproduce this finding, although Voorhees did find that patients with BPPV could not use vestibular cues effectively to maintain balance. Voorhees [6] suggests that the difference may be related to the comorbidity of head injury in the patients studied by Black and Nashner, [5] which was not true of his patient sample.

The etiology of BPPV is often unknown. Baloh et al [3] report that, in a study of 240 patients with BPPV, a diagnosis could not be determined in 118 patients. Among the patients in whom a diagnosis could be made, the most frequent diagnoses were head trauma (n = 43) and viral neuronitis (n = 37). The remainder of the patients had varied histories including surgery, Meniere's disease, vertebral basilar insufficiency, migraine, and multiple sclerosis. In many cases, the diagnoses were unrelated to the BPPV. Gacek [7] suggests that some type of insult (eg, head trauma, infection, vascular injury) to the labyrinth resulting in degeneration is common to all cases of BPPV.

Baloh et al [3h found that the peak incidence of onset occurred in the sixth decade of life in the patients in whom no diagnosis could be made. In patients with a history of head trauma, there was no ocrrelation of age to onset, whereas in those patients with a history of viral neuronitis, the onset was usually between the fourth and sixth decades of life. Semont et al [8] in a study of 711 patients with BPPV, found a different relationship between etiology and age of onset. Patients with idiopathic BPPV were from 45 to 85 years of age, whereas patients in the posttrauma group were considerably younger (age 20-45 years).

Other Causes of

Positional Vertigo

Positional vertigo occurs with lesions of the central nervous system as well as with lesions of the peripheral nervous system. The characteristics of central positional vertigo and BPPV, however, are different (Table). [9] In central positional vertigo, the vertigo begins as soon as the patient is put into the provoking position; the response persists as long as the position is maintained, and it does not fatigue with repeated positional changes. Central positional vertigo may occur in a variety of disorders affecting the brain stem such as infarct, tumors, and multiple sclerosis. [10,11] Unfortunately, central positional vertigo does not respond to the treatments that can be used so effectively in BPPV. Another type of positional vertigo that does not habituate occurs when the vertibular nerve is compressed by blood vessels. This type of disorder is treated with microvascular decompression surgery. [12]

Another diagnosis that may be confused with BPPV is cervical vertigo. [13] Cervical vertigo, although poorly understood, is believed to be due to inappropriate afferent signals from joint and spindle receptors. [14] Animal studies have demonstrated that unilateral cervical rhizotomies or the injection of a local anesthetic unilaterally in the cervical region results in ataxia or nystagmus. [13] This effect has been difficult to demonstrate in human beings. De Jong and Bles [13] report that patients with cervical vertigo will tend to fall backward if they extend their neck while standing with eyes closed. Because patients with BPPV sometimes complain of vertigo induced by neck extension (head movement from flexion to extension would excite the posterior semicircular canals), these two diagnoses may be confused. In addition, patients with cervical vertigo as well as those with BPPV may have a history of trauma.

Vertigo induced by positional changes may also occur in patients with perilymph fistual (PLF). In patients with PLF, there is an abnormal communication between the inner and middle ears, usually through the round or oval window. Patients with PLF typically complain of sudden hearing loss, vertigo, disequilibrium, and nausea. [15,16] As in BPPV, this disorder can occur after head trauma, although there are many other causes such as stapedectomy and barotrauma. [15,17,18] The nystagmus that occurs with positional changes in patients with PLF may be similar to that seen in patients with BPPV, although other forms of nystagmus also occur. [16,19]

Pathogenesis

Two different theories--"cupulolithiasis" and "canalithiasis"--relate the structural relationship of the utricle and the posterior canal to the vertigo and nystagmus that occur with specific head movements in patients with BPPV. Normally, the semicircular canals respond to head movement but not to head position (ie, not to the pull of gravity). Both theories suggest that structural changes occur in which one of the posterior semicircular canals becomes gravity-sensitive. The characteristics of BPPV (ie, latency, burst, and duration) can be adequately explained by both theories.

Theory 1--Cupulolithiasis. This theory proposes that degenerative debris from the utricle (probably fragments of otoconia) fall onto the cupula of the posterior canal, making the ampulla gravity-sensitive. This phenomenon--cupulolithiasis--was first described in 1969 by Schuknecht, [20] who found basophilic deposits on the cupula of the posterior canal in patients with a history of BPPV. Positioning the head with the affected ear below the horizontal causes an inappropriate deflection of the cupula of the posterior canal, presumably because of its gravity-sensitivity, and results in vertigo, nystagmus, and nausea. The latency of the onset of the vertigo and nystagmus is related to the time required to displace the gravity-sensitive cupula. The gradual increase in vertigo and nystagmus is related to the increased deflection of the cupula. The gradual decrease in vertigo and nystagmus that occurs if the head-hanging position is maintained is due to adaptation.

Theory 2--Canalithiasis. Hall et al [21] propose a somewhat different theory concerning the mechanical factors producing BPPV. They suggest that the degenerative debris is not adherent to the cupula of the posterior canal but instead is free-floating in the endolymph. When the head is moved into the provoking position, the endolymph, moved by the falling otoconia, pulls on the cupula, thus exciting the neurons. The latency of the response is related to the time required for the cupula to be deflected by the pull of the endolymph. The increase in vertigo and nystagmus that occurs is related to the relative deflection of the cupula. The decrease in vertigo and nystagmus as the position is maintained is due to cessation of endolymph movement. As with the cupulolithiasis theory, the Hallpike-Dix maneuver is most likely to result in vertigo and nystagmus, although symptoms may be provoked by other movements in the plane of the posterior canal. Hall et al [21] argue that their model explains the fatigue of vertigo that occurs with repeated positional changes in patients with BPPV better than does the cupulolithiasis model. They hypothesize that with repeated movement of the head into the precipitating position, some of the debris moves out of the posterior canal, thereby reducing the response.

Treatment

Several approaches have been developed to treat patients with BPPV. One treatment approach is based on the idea that the debris embedded in the cupula of the posterior canal can be dislodged by repeatedly moving the patient into the position that provokes the vertigo. [22] An alternative treatment approach, the Liberatory maneuver, moves the patient through a series of positions in order to float the debris out of the posterior canal but does not distinguish between cupulolithiasis and canalithiasis. [8] A third approach, suggested by Norre and De Weerdt [23] and by Tangeman and Wheeler, [24] is based on the concept of habituating the CNS response to movement-provoked vertigo.

Treatment 1--Brandt' exercises. Proposed by Brandt and Daroff, [22] this treatment requires the patient to move into the provoking position repeatedly, several times a day. The patient is first positioned sitting and then rapidly moves into the position that causes the vertigo (Fig 2). Torsional nystagmus may occur with the onset of the vertigo. The severity of the vertigo will be directly related to how rapidly the patients moves into the provoking position. The patient stays in that position until the vertigo stops and then sits up again. Usually moving to the sitting position will also result in vertigo, although this "rebound effect" will be less severe and of a shorter duration. Nystagmus, if it reoccurs, will be in the opposite direction. The patient remains in the upright position for 30 seconds and then moves rapidly into the mirror-image position on the other side, maintains that position for 30 seconds, and then sits up. The patient then repeats the entire maneuver until the vertigo diminishes. The entire sequence is repeated every 3 hours during the day until the patient has experienced no episodes of vertigo for 2 consecutive days. It is not clear why these exercises result in a decrease in the vertigo and nystagmus. One explanation is that the debris becomes dislodged from the cupula of the posterior canal and moves to a location where it no longer affects the cupula during head movement. A second possibility is that central adaptation occurs, reducing the nervous system response to the signal from the posterior canal. Brandt and Daroff [22] argue against central adaptation as a mechanism for recovery because many patients recover abruptly.

Treatment 2--Liberatory maneuver. As with the treatment approach proposed by Brandt and Daroff, [22] the provoking position must first be identified. [8] The patient is first moved quickly from a sitting position to the provoking position and kept in that position for 2 to 3 minutes. He is then turned rapidly to the opposite eardown position with the therapist maintaining the alignment of the neck and head on the body. The patient stays in this position for 5 minutes. Typically, nystagmus and vertigo reappear in this position. The patient is then slowly returned to a seated position. He must remain in a vertical position for 48 hours (including while sleeping) and must avoid the provoking position for 1 week following the treatment. Unlike the exercises suggested by Brandt and Daroff, [22] the Liberatory maneuver usually requires only a single treatment. It purportedly works by floating the debris through the canal system to the common crus.

Treatment 3--Habituation exercises. The exercises of Norre and De Weerdt [23] differ from those described by Brandt and Daroff [22] in that the provoking positions used are specific for each patient and are not limited to the Hallpike-Dix maneuver. In a more recent article, Norre and Beckers [25] describe the specific movements used to establish the individualized treatments and discuss the results of those treatments. Tangeman and Wheeler [24] describe the three phases of their treatment protocol. Phase I is similar to the Brandt and Daroff [22] protocol and consists of having the patient move repeatedly into the Hallpike-Dix position; phases II and III include a wide variety of balance exercises that incorporate eye and head movements and that seem similar to the Cawthorne-Cooksey exercises advocated for patients with unilateral vestibular hypofunction. [26] The inclusion of specific exercises for balance in these latter approaches is appropriate for the treatment of the postural instability sometimes seen in patients with BPPV. In addition, the patients samples in these studies included patients with vestibular hypofunction or patients with BPPV combined with vestibular hypofunction who would most likely also have balance problems.

Treatment Efficacy

Studies on the efficacy of these treatments indicate that both Brandt's exercises and the Liberatory maneuver facilitate recovery. [8,22] The results of these studies are confounded, however, by the high incidence of spontaneous remission in patients with BPPV. Several authors have reported spontaneous recovery within 3 to 4 weeks [8,27] although Brandt and Daroff [22] suggest that the vertigo may not disappear for months if left untreated.

Brandt and Daroff [22] studied a series of 67 patients with histories of BPPV of 2 days' to 8 months' duration. None of these patients had evidence of other neurological or neurotological disease. They reported that 98% of the patients had no symptoms after 3 to 14 days of exercises. The only patient who did not respond to treatment had a PLF requiring surgical repair. Recurrence of BPPV was minimal, affecting only 3% of the patients. In our experience with a series of 20 patients with BPPV treated with exercises similar to those advocated by Brandt and Daroff, [22] the amount of time until the patients were symptom-free (n = 12) or had at least a moderate reduction in symptoms (n = 7) was more protracted, extending from 1 week to 6 months. Patients with only partial recovery complained most frequently of an intermittent "swimming" sensation rather than of true vertigo. One patient experienced no change in vertigo. These patients had histories of BPPV ranging from a few days to 35 years. Possibly, the longer the history of the disorder, the more resistant the BPPV is to treatment. We also observed that most patients requiring a more extended course of treatment had additional nervous system disorders that may have impeded recovery.

Semont et al [8] report a series of 711 patients with BPPV treated with the Liberatory maneuver over an 8-year period. It is not clear from their article whether the patients had other neuro-otological problems. The authors state that some of the patients had slightly increased or decreased responses on caloric testing, but they did not define their criteria for a normal response. Statistically significant abnormal responses to caloric testing have been reported to occur in up to 47% of patients with BPPV. [3] Semont et al [8] report a "cure" rate of 84% after a single treatment and 93% after two treatments. Again, recurrence of the symptoms was infrequent (4%). We have used a similar maneuver (Fig 3) on a much smaller sample of patients with BPPV (N = 14). This maneuver resulted in remission of BPPV in 11 patients; 2 of those patients required two treatments. The maneuver was ineffective in 3 patients, one of whom had multiple sclerosis; whether her vertigo was due to a peripheral nervous system lesion or a CNS lesion is not known. Neither the Liberatory maneuver nor Brandt's exercise approach ameliorated the symptoms in these 3 patients.

Norre and Beckers, [28,29] compared the efficacy of the Liberatory technique with Habituation exercises. For the Habituation treatment, patients repeated the positional changes five times and performed two or three sessions each day. In their series of 23 patients treated with the Liberatory technique, 52% were free of vertigo after one treatment. Only 32% of the 28 patients treated with Habituation exercises were free of vertigo at the end of 1 week, but the remaining patients reported a decrease in their symptoms. By the end of 6 weeks, all 28 of the patients treated with Habituation exercises had no vertigo. In addition, those patients treated with the Liberatory technique who did not improve with a single treatment were switched to the Habituation treatment protocol, and all except one was free of vertigo at the end of the 6-week treatment period. They concluded that the two treatments were equally effective in the treatment of BPPV.

Guidelines to Treatment of

Benign Paroxysmal

Positional Vertigo

The diagnosis of BPPV is usually made by a neuro-otologist or a neurologist, who would then refer the patient for physical therapy. The physical therapist should examine the patient to determine 1) what positional changes produce the vertigo, 2) whether any balance problems exist that are associated with the BPPV, and 3) whether other conditions exist that may affect treatment (ie, neck pain).

Tests to determine the provoking positions are important in order to develop the appropriate treatment protocol and to monitor the progress of the patient. The latency, duration, and intensity (scaled 1-5 or 1-10) of the vertigo should be documented for each of the position changes (Fig 4). It is important to perform the positional changes quickly in order to provoke a response. Testing must be performed consistently because a decreased response, obtained when the positional change is made too slowly, may imitate improvement. Nystagmus is observed using Frenzel glasses, which magnify the patient's eyes for the observer. These glasses prevent the patient from using visual fixation to suppress the nystagmus. The direction and duration of the nystagmus should be noted. Some of the position changes listed in Figure 4 should not result in vertigo or nystagmus in patients with BPPV because the movements do not affect the posterior canal.

For most patients, either Brandt's exercises or the Liberatory treatment approach may be used; however, the following factors should be considered:

1. Elderly patients may be less tolerant of the Liberatory maneuver than younger patients, especially if they move cautiously because of other conditions such as arthritis.

2. Patients may not wish to stay in an upright position for the 48 hours required by the Liberatory maneuver. Bending over may be difficult for some patients to avoid (eg, parents with small children, patients required to perform certain work-related activities).

3. Patients, especially those with long histories of BPPV, may have anxiety about moving into the provoking position and may be resistant to the Liberatory maneuver; Brandt's exercises may be modified so that the patient has more control over the positional change and gradually becomes less fearful of provoking the vertigo and nausea. The anxious patient, however, may tend to move out of the provoking position too quickly when attempting to perform the exercises without assistance. The extent of anxiety patients can experience should not be underestimated; one patient I observed with a long history of BPPV became so fearful of provoking the vertigo that he tied one arm down at night to keep from rolling over onto the "bad side."

4. The success of Brandt's exercises are dependent on the compliance of the patient. Some improvement may occur within a few days after initiating the treatment, but treatment may have to be continued for extended periods of time. Weekly clinical visits may help improve the compliance of the patient, but in patients with poor compaliance, the Liberatory maneuver may be the more appropriate choice.

5. Cervical and back pain may preclude the Liberatory maneuver or may be aggravated by the repeated positional changes of Brandt's exercises. The positional changes used in Brandt's exercises may be modified to enable the patient to perform them.

6. There is some risk of neck injury when performing the Liberatory maneuver or similar maneuvers. This risk is small, however, because the head is turned to the side before the patient reclines and because the head is supported at all times.

7. The patient must "work through" the vertigo and its accompanying nausea. [14]. Usually, these complaints disappear quickly when the patient is mvoed out of the provoking position or as the vertigo decreases. Repeated positional changes, as would occur with the Brandt's exercises, may cause a prolongerd and generalized dis-equilibrium with persistent nausea. These effects may be disturbing enough that the patient stops the exercises. Patients should be warned that these effects may occur and that they are temporary. Usually, these effects can be controlled by modifying the exercises (eg, decreasing the repetitions for a while) or by regulating the time during the day when the exercises are performed. These effects may also be controlled by taking medication, such as Phenergan[R], (*) half an hour before the excercises are performed. Brandt's exercises are usually performed with the eyes closed to minimize the visuovestibular conflict contributing to the nausea. Opening the eyes may result in an increase in the nausea but may also facilitate adaptation and therefore recovery.

8. The Liberatory maneuver usually should not be used in patients with bilateral BPPV. Bilateral BPPV, like BPPV affecting the labyrinth unilaterally, has been reported in idiopathic cases and after head injury. [30]

Case Studies

Case 1

The patient was a 50-year-old woman who was hit by a car 2 weeks previously while crossing a street. At the time of the accident, she was alert and oriented as to time, place, and person but had complaints of cervical pain and was taken to the emergency room of the nearest hospital. While in the radiology department, the patient had a cardiac arrest and was resuscitated. She was admitted to the intensive care unit but had no other cardiac incidents. While in the intensive care unit, the patient began to complain of vertigo when she turned her head to the left. On examination 1 week later, the patient's strength and sensation were within normal limits, her cerebellar signs were negative, and she had no pathological reflexes. The Hallpike-Dix maneuver to the left resulted in a "down-beating" and counterclockwise nystagmus and complaints of vertigo with a latency and duration consistent with BPPV. Sitting up resulted in vertigo and nystagmus (inn the opposite direction) that was less severe and of shorter duration. Down-beating nystagmus, in contrast to upbeating nystagmus, is unusual in BPPV, but it has been reported in 9% of subjects with BPPV. [3] A diagnosis was made of postconcussional BPPV. The patient was treated with meclizine and was referred to physical therapy for exercises.

At the time of her initial physical therapy visit, the patient was found to have vertigo and torsional nystagmus when placed in either the left or the right Hallpike-Dix position. In addition, the patient still complained of cervical pain and was wearing a cervical collar at the advice of the therapist who was treating her neck injury. Her neck rotation range of motion and neck lateral flexion ROM were decreased by 20% to the right and left. Brandt's exercises were used for treatment of the BPPV because of the patient's cervical complaints and the fact that the BPPV was bilateral (Fig 2). The treatment goal was to eliminate the vertigo in 8 weeks. Initially, the patient was instructed to perform the exercises wearing her cervical collar as that seemed to minimize her neck pain. On follow-up visits, the patient reported a gradual decrease in the intensity and duration of the vertigo induced by the positional changes. In 4 weeks, she had no complaints of vertigo, and nystagmus could not be observed under Frenzel lenses with the Hallpike-Dix maneuver to the left; vertigo and nystagmus occurred with the Hallpike-Dix maneuver to the right but were less severe. Seven weeks after initiation of treatment, the patient had no complaints of vertigo, and nystagmus could not be induced with any positional changes. The patient was discharged from treatment with the goal of eliminating the vertigo attained.

Case 2

The patient was a 51-year-old man who reported a history of intermittent mild vertigo occurring for several years. He described a spinning sensation that lasted for only a few seconds when he rolled from right to left or when he moved from a supine to a sitting position. He denied experiencing vertigo or disequilibrium induced by movement of the environment, movement through the environment, or pressure changes. He was still playing tennis and reported no vertigo while serving. He denied experiencing falls or unsteadiness.

On assessment, he was without spontaneous, gaze-evoked, or head-shaking nystagmus. With movement into the left Hallpike-Dix position, he complained of vertigo. A torsional nystagmus was also observed through Frenzel lenses. The latency, burst pattern, and duration of the vertigo and nystagmus was consistent with BPPV. On sitting up, the patient experienced a second episode of vertigo, although it was milder than the initial episode. No other positional changes provoked the vertigo.

This patient was treated using a modification of the Liberatory maneuver (Fig 3). He tolerated the procedure well and was instructed to remain upright for 48 hours and to avoid lying on the left side for 1 week. At the end of 1 week, he returned for reassessment. At that time, the left Hallpike-Dix maneuver was repeated. The patient subsequently experienced no vertigo, and no nystagmus was observed through Frenzel lenses. The patient was then discharged from treatment.

Conclusion

Benign paroxysmal positional vertigo is a common peripheral vestibular disorder that can be treated with physical therapy. This article has reviewed its etiology and pathology and has described several different treatment approaches. Suggestions are made as an aid for clinical decision making.

References

[1] Barany R. Diagnose von Krankheitserscheinungen im Bereiche des Otolithenapparatus. Acta Otolaryngol. 1921;2:434-437

[2] Dix MK, Hallpike CS. Pathology, symptomatology and diagnosis of certain disorders of the vestibular system. Proceedings of the Royal Society of Medicine. 1952;45:341-354

[3] Baloh RW, Honrubia V, Jacobson K. Benign positional vertigo: clinical and oculographic features in 240 cases. Neurology. 1987;37:371-378

[4] McClure JA. Horizontal canal BPV. J Ptolaryngol. 1985;14:30-35

[5] Black FO, Nashner LM. Postural disturbance in patients with benign paroxysmal positional nystagmus. Ann Otol Rhinol Laryngol. 1984;93:595-599

[6] Voorhees RL. The role of dynamic posturography in neurotologic diagnosis. Laryngoscope. 1989;99:995-1001

[7] Gacek RR. Cupulolithiasis. Adv Otorbinolaryngol. 1982;28:80-85

[8] Semont A, Freyss G, Vitte E. Curing the BPPV with a Liberatory maneuver. Adv Otorbinolaryngol. 1988;42:290-293

[9] Snow JB. Positional vertigo. N Engl J Med. 1984;310:1740-1741

[10] Fisher CM. Vertigo in cerebrovascular disease. Arch Otolaryngol. 1967;85:529-534

[11] Troost TB. Dizziness and vertigo in vertebrobasilar disease, part II: central causes and vertebrobasilar disease. Current Concepts of Cerebrovascular Disease-Stroke. November-December 1979:413-415

[12] Jennetta PJ, Moller MB, Moller AR. Disabling positional vertigo. N Engl J Med. 1984;310:1700-1705

[13] De Jong JMBV, Bles W. Cervical dizziness and ataxia. In: Bles W. Brandt TH, eds. Disorders of Posture and Gait. Amsterdan, the Netherlands: Elsevier; 1986:185-206

[14] Herdman SJ. Patients with vestibular disorders. In: Postgraduate Advances in Physical Therapy. Alexandria, VA: American Physical Therapy Association; 1987

[15] Glasscock ME, McKennan KX, Levine SC. Persistent traumatic perilymph fistulas. Laryngoscope. 1987;97:860-864

[16] Healy GB, Strong MS, Sampogna D. Ataxia, vertigo, and hearing loss. Arch Otolaryngol. 1974;100:130-135

[17] Seltzer S, McCabe BF. Perilymph fistula: The Iowa experience. Laryngoscope. 1986;94:37-49

[18] Lehrer JF, Rubin RC, Poole DC, et al. Perilymphatic fistula: a definitive and curable cause of vertigo following head trauma. West J Med. 1984;141:57-60

[19] Singleton GT, Post KN, Karlan MS, et al. Perilymph fistulas: diagnostic criteria and therapy. Annals of Otology. 1978;87:797-803

[20] Schuknecht HF. Cupulolithiasis. Arch Otolaryngol. 1969;90:765-778

[21] Hall SF, Ruby RRF, McClure JA. The mechanisms of benign paroxysmal vertigo. J Otolaryngol. 1979;8:151-158

[22] Brandt T, Daroff RB. Physical therapy for benign paroxysmal positional vertigo. Arch Otolaryngol. 1980;106:484-485

[23] Norre ME, De Weerdt W. Positional (provoked) vertigo treated by postural training vestibular habituation training. Agressologie. 1981;22:37-44

[24] Tangeman PT, Wheeler J. Inner ear concussion syndrome: vestibular implications and physical therapy treatment. Topics in Acute Care Trauma Rehabilitation. 1986;1:72-83

[25] Norre ME, Beckers A. Bestibular habituation training: exercise treatment for vertigo based upon the habituation effect. Otolaryngol Head Neck Surg. 1989;101:14-19

[26] Hecker HC, Haug CO, Herndon JW. Treatment of the vertiginous patient using Cawthorne's vestibular exercises. Laryngoscope. 1974;84:2065-2072

[27] Gyo K. Benign paroxysmal positional vertigo as a complication of postoperative bedrest. Laryngoscope. 1988;98:332-333

[28] Norre ME, Beckers A. Exercise treatment for paroxysmal positional vertigo: comparison of two types of exercises. Arch Otorbinolaryngol. 1987;244:291-294

[29] Norre ME, eckers A. Comparative study of two types of exercise treatment for paroxysmal positioning vertigo. Adv Otorbinolaryngol. 1988;42:287-289

[30] Longridge NS, Barber HO. Bilateral paroxysmal positioning nystagmus. J Otolaryngol. 1978;395-400

S Herdman, PhD, PT, is Assistant Professor, Department of Otolaryngology-Head and Neck Surgery, Johns Hopkins University, 600 N Wolfe St, Baltimore, MD 21205 (USA).