<LEAP> highlights the findings and application of Cochrane reviews and other evidence pertinent to the practice of physical therapy. The Cochrane Library is a respected source of reliable evidence related to health care. Cochrane systematic reviews explore the evidence for and against the effectiveness and appropriateness of interventions—medications, surgery, education, nutrition, exercise—and the evidence for and against the use of diagnostic tests for specific conditions. Cochrane reviews are designed to facilitate the decisions of clinicians, patients, and others in health care by providing a careful review and interpretation of research studies published in the scientific literature.1 Each article in this PTJ series summarizes a Cochrane review or other scientific evidence on a single topic and presents clinical scenarios based on real patients or programs to illustrate how the results of the review can be used to directly inform clinical decisions. This article focuses on an adult patient with benign paroxysmal positional vertigo. Can a canalith repositioning procedure help this patient?
Benign paroxysmal positional vertigo (BPPV) presents as brief periods of vertigo experienced with a change in the position of a person's head relative to gravity. Benign paroxysmal positional vertigo is a mechanical disorder of the inner ear and is caused by abnormal stimulation of 1 or more of the 3 semicircular canals (Fig. 1). The otoconia from the utricle dislodge and settle within 1 of the 3 semicircular canals, changing the fluid-filled canal dynamics from detecting rotation of the canals to detecting gravitation forces on the head. With BPPV, changing the plane of the involved canal relative to gravity causes debris to settle to the lowest part of the canal, changing the fluid pressure across the cupula, deflecting the hair cells, and generating the characteristic ocular nystagmus with associated vertigo. There are 3 mechanisms of BPPV, the most common being debris within the long arm of the canal; this mechanism is called canalithiasis.2 Debris also may be found in the short arm of the canal3 or attached to the cupula, weighting the membrane; this mechanism is called cupulolithiasis.3 The canal most commonly involved is the posterior semicircular canal.4
Mechanisms of benign paroxysmal positional vertigo (BPPV), illustrated for left posterior semicircular canal BPPV. The crista ampullaris of the fluid-filled semicircular canals contains sensory epithelium consisting of hair cells embedded within the cupula, a fine, gelatinous membrane. Rotation of the head deflects the hair cells. The macula utriculi consists of a weighted sensory membrane containing hair cells implanted within an otolithic membrane weighted with calcite particles (otoconia). Gravitational forces on the head deflect the hair cells. Reprinted with permission from American Dizziness and Balance. Copyright 2007.
The diagnosis of BPPV is based on history and findings on positional testing.5,6 Patients who complain of vertigo with rolling in bed or getting out of bed are 4.3 times more likely to have BPPV.7 The Dix-Hallpike test8 and the side-lying test9 are the standard tests5,6 for evaluating posterior canal BPPV. A comparison of the Dix-Hallpike test and the side-lying test revealed no significant difference between the 2 techniques.9 Frenzel lenses or video oculography should be used to prevent the suppression of ocular nystagmus. The diagnostic criteria for posterior canal BPPV are vertigo associated with the characteristic nystagmus—torsional (superior pole of the eye directed toward the lowermost ear [ie, the involved ear]) and upbeating,10 with a latency of 1 to 45 seconds before onset11–13 and a duration of less than 60 seconds14—and fatigue with repeated positioning.14 For the Dix-Hallpike test, the estimated sensitivity is 79% (95% confidence interval [95% CI]=65–94), and the specificity is 75% (95% CI=33–100); for the side-lying test, the estimated sensitivity is 90% (95% CI=79–100), and the specificity is 75% (95% CI=34–100).15
Once identified, posterior canal BPPV may be effectively treated with a particle repositioning maneuver, such as the canalith repositioning maneuver (CRP) described by Epley.16 The CRP is a series of 4 head positions designed to use the effects of gravity to treat canalithiasis (Fig. 2A). With each head position, the debris settles to the lowest portion of the canal, moving the debris away from the ampulla, into the common crus, and then into the utricle. With each change in position (steps 2–4), the movement of debris through the canal and away from the ampulla may create changes in pressure across the cupula, resulting in the generation of the typical torsional and upbeating nystagmus—predicting the successful outcome of the maneuver (Fig. 3A; video below).17 If, during the second or third position of the maneuver (or both), the movement of debris through the canal reverses direction, with the debris moving toward the ampulla, a reversal in the direction of nystagmus may be observed—predicting treatment failure (Fig. 3B).17 Video oculography should be used to monitor nystagmus in response to the movement of debris through the canals and to prevent the suppression of nystagmus. The procedure is repeated a minimum of 3 times within a treatment session. Repeating the CRP more than once is significantly more effective than performing the CRP only once within a treatment session.18
A video of the canalith repositioning procedure for benign paroxysmal positional vertigo in the left posterior semicircular canal.
Particle repositioning maneuvers. (A) Canalith repositioning procedure, illustrated for treatment of the right posterior semicircular canal. The clinician moves the patient through a series of 4 positions, starting with the placement of the involved canal in the head-hanging position of the Dix-Hallpike test. To begin, the patient is positioned on the treatment table in the sitting position with the legs extended. The patient's head is rotated 45 degrees toward the right. The patient is then lowered into the supine position with the neck extended 30 degrees over the edge of the treatment table. This is the head-hanging position. The head is rotated through 90 degrees of motion ending in 45 degrees of neck rotation toward the uninvolved side. This step is followed by rolling onto the uninvolved side while maintaining the position of the head in relation to the trunk and, finally, sitting up from lying on the uninvolved side. Each position is maintained for a minimum of 45 seconds or as long as the nystagmus lasts plus an additional 20 seconds. The procedure is repeated 3 times. (B) Liberatory (Semont) maneuver, illustrated for treatment of the right posterior semicircular canal. The patient sits on the edge of the treatment table. The clinician rapidly moves the patient so that he or she is lying on the involved side with the head rotated 45 degrees toward the uninvolved side. While maintaining the position of the head in relation to the trunk, the clinician swings the patient from lying on the involved side to lying on the uninvolved side. The head is then gently tapped on the treatment table. Each position is maintained for 1.5 minutes. The procedure is repeated 3 times. Reprinted with permission from American Dizziness and Balance. Copyright 2007.
Movement of debris within the right posterior semicircular canal during the canalith repositioning procedure, illustrated for treatment of the right posterior semicircular canal. The arrow indicates the direction of angular velocity and the flow of endolymph. The cupula acts as a plunger (diaphragm within the ampulla). (A) Prediction of successful outcome. With each position of the head, debris settles away from the cupula, creating an ampullofugal flow of endolymph away from the cupula, exciting the hair cells within the crista ampullaris, and generating the same direction of nystagmus (Ny). (B) Prediction of failed outcome. In the second or third position (or both), debris settles toward the cupula, reversing the direction of flow of endolymph toward the ampulla (ampullopetal), inhibiting the hair cells within the crista ampullaris, and reversing the direction of nystagmus.
Acceptable modifications of the CRP (described by Epley) include performance of the CRP without mastoid vibration (thought to prevent debris from adhering to the canal walls) and self-administration of the CRP.5,6 The short-term outcome is the same for patients with and patients without postintervention postural and activity restrictions.19,20 However, fewer treatment sessions are required to produce a negative Dix-Hallpike test result if postintervention postural and activity restrictions are maintained.21 To determine the results of the CRP, the Dix-Hallpike test should be performed optimally more than 24 hours after treatment to avoid the effects of fatigue.22 A successful outcome is defined as conversion from a positive Dix-Hallpike test result to a negative one (ie, no benign paroxysmal positional nystagmus and no associated vertigo).
In a Cochrane systematic review, Hilton and Pinder23 examined the evidence to determine the effectiveness of the CRP relative to that of other treatments or no treatment for people with posterior canal BPPV. That review was an update of a Cochrane review first published in The Cochrane Library in 2002 (issue 1) and then updated in 2004 and 2007. Randomized controlled trials were included if participants were adults (>16 years old) and had a clinical diagnosis of posterior canal BPPV based on a positive result on the Dix-Hallpike test (observation of the characteristic nystagmus and symptoms of associated vertigo). Trials compared the effectiveness of the modified CRP (without the use of vibration) with that of a sham maneuver or no treatment (control). There was no evidence comparing the effectiveness of the CRP with that of other physical, medical, or surgical therapies for posterior canal BPPV. Follow-up ranged from less than 24 hours22,24 to 1 to 2 weeks25 and 1 month26,27 after treatment. The primary outcomes included complete resolution of symptoms and conversion from a positive to a negative Dix-Hallpike test result, the only objective physiological change resulting from treatment. The search date of the review was May 19, 2010. A summary of the findings is given in the Appendix.
Take-Home Message
The review reported the results of 5 randomized controlled trials involving 292 adults with a clinical diagnosis of posterior canal BPPV based on history and a positive result on the Dix-Hallpike test. The outcome of the CRP was compared with that of no treatment (control) in 1 study27 or a sham treatment in the remaining 4 studies.22,24–26 The sham treatment consisted of lying on the affected side for 5 minutes,25 lying in the first lateral position of the liberatory maneuver,26 or CRP performed as if the opposite ear were affected.22,24 In 2 trials, participants were given instructions on postintervention restrictions25,26; in 3 trials, no postintervention restrictions were included.22,24,27 The overall risk for bias was graded low for 4 trials22,24–26 but could not be determined for 1 trial.27
Pooled data showed a statistically significant difference in symptom resolution and in conversion from a positive to a negative Dix-Hallpike test result in favor of the CRP.23 At short-term follow-up (<1 month), the rates of subjective symptom resolution for patients treated with the CRP were 32% to 80%,22,24–27 those obtained with the sham treatment were 10% to 34%,22,24–26 and that obtained with the control was 35%.27 The pooled odds in favor of complete subjective symptom resolution were 4.42 times higher (95% CI=2.62–7.44) with the CRP than with the sham treatment or the control.22,24–27 At short-term follow-up, the success rates for patients who converted to a negative Dix-Hallpike test result when treated with the CRP were 34% to 89%,22,24–27 those obtained with the sham treatment were 10% to 38%,22,24–26 and that obtained with the control was 65%.27 The magnitude of the effect of the CRP compared with that of the sham treatment or the control was significant in all 5 studies. The pooled odds in favor of conversion from a positive to a negative Dix-Hallpike test result were 6.40 times higher (95% CI=3.63–11.28) with the CRP than with the sham treatment or the control.22,24–27 The individual odds in favor of symptom resolution varied from 3.03 (95% CI=1.01–9.07) to 37.33 (95% CI=8.75–159.22). There were no reported adverse effects or serious complications of treatment. Reported complications included the inability to tolerate positions because of cervical spine dysfunction,25 emesis during treatment,22,25 nausea,22 and fainting.27
Case #24: CRP for Treatment of Posterior Semicircular Canal BPPV
Can CRP help this patient?
“Dr X” is a 40-year-old woman with a 7-year history of episodic positional vertigo, the most recent episode beginning 7 days earlier, when she rolled over in bed and woke with severe vertigo and nausea. She was evaluated by a neuro-otologist, diagnosed with BPPV, and referred to outpatient physical therapy for vestibular rehabilitation. To differentiate between BPPV and other causes of dizziness, the physical therapist evaluated self-report participation questionnaires, a history, a neurologic screen, and positional testing. The patient completed the Dizziness Handicap Inventory,28 designed to evaluate self-perceived handicap due to dizziness, and the Activities-specific Balance Confidence Scale,29 designed to evaluate self-perceived balance confidence in performing household and community activities (Table). The patient reported constant nausea; vertigo when rolling in bed toward the right, getting in and out of bed, bending forward, looking up, and moving her head rapidly; and vomiting during the physician's examination. She slept in bed propped upright on 5 pillows to avoid provoking positions, experienced difficulty concentrating at work, and felt anxious. The patient's history and findings on the Dizziness Handicap Inventory suggested severe BPPV. Due to the severity of the BPPV, to reduce the risk of emesis, and to prevent suppression of ocular nystagmus, the patient medicated with ondansetron (Zofran, GlaxoSmithKline, Research Triangle Park, North Carolina) before the evaluation, in accordance with her physician's order.
Preintervention and Postintervention Outcome Measures
The results of the neurologic screen were negative. The patient had no contraindications for further positional testing or a particle repositioning maneuver. Functional outcome measures administered included the Dynamic Gait Index,30 designed to evaluate postural stability during various walking tasks, and the Timed “Up & Go” Test,31 a measurement of walking speed (Table). The Dix-Hallpike test was performed with the use of video oculography to avoid the visual fixation that suppresses ocular nystagmus and to digitally record eye movements. In the Dix-Hallpike test, the result for the left head hanging position was negative. The result for the right head hanging position was positive; typical nystagmus—counterclockwise torsional (superior pole of the eye directed toward the lowermost ear) and upbeating—was observed, with a 2-second latency before the onset of nystagmus, a duration of nystagmus of 21 seconds, and a peak intensity of velocity of nystagmus of 8 of 10 (0=no nystagmus; 10=extreme nystagmus). The patient complained of severe vertigo associated with nystagmus and became nauseated. After the patient was returned to the upright position, strong downbeating nystagmus associated with vertigo and disorientation to vertical was noted. The patient's history and findings on the Dix-Hallpike test suggested BPPV involving the right posterior canal. The patient was identified as a good candidate for the CRP for right posterior canal BPPV. The mechanism of BPPV and treatment options were discussed with the patient. She was informed of potential treatment complications, such as canal conversion (movement of debris into another canal), canal jam (movement of debris from a wide to a narrow portion of the canal, resulting in plugging of the canal), further nausea, emesis, and further imbalance. She understood that the treatment risks were minimal and consented to CRP treatment for right posterior canal BPPV.
The physical therapy intervention consisted of performance of the CRP on the right without the use of mastoid vibration. Each position was maintained for the duration of nystagmus in the second step plus 20 seconds (Fig. 2A). Video oculography was used to observe nystagmus in each position to monitor treatment progress. For optimal effectiveness of the maneuver, 3 cycles of the CRP were performed in 1 treatment session. During the first cycle, counterclockwise torsional and upbeating nystagmus was observed in the second, third, and fourth steps, suggesting the movement of debris away from the right posterior canal ampulla and successful treatment.17 The nystagmus was associated with vertigo. The patient felt nauseated but did not have emesis. During the second and third cycles of the CRP, no nystagmus was observed, and the patient did not complain of vertigo or nausea. After the procedure, the patient sat in the clinic for 20 minutes to allow debris to settle and to be supported if she experienced sudden imbalance or vertigo. She was instructed in postural and activity restrictions to minimize the number of treatment sessions. She was instructed to sleep on the uninvolved side with her head elevated on a wedge made of 3 pillows and to avoid up and down movements of the head for 1 week. She was given information describing the mechanisms of BPPV, treatment options, posttreatment postural and activity restrictions, and the recurrence of BPPV. She was instructed to have someone drive her home.
The treatment results were evaluated at a 1-week follow-up. Success was defined as no subjective complaints of vertigo and conversion from a positive to a negative Dix-Hallpike test result. The patient reported maintaining postural and activity restrictions and experiencing no episodes of positional vertigo or symptoms of nausea during her daily routine. The total score on the Dizziness Handicap Inventory suggested no self-perceived handicap due to dizziness, and the total scores on the Activities-specific Balance Confidence Scale and the Dynamic Gait Index suggested no residual balance deficits (Table). The patient's gait speed on the Timed “Up & Go” Test was within normal limits for her age (Table). The Dix-Hallpike test was performed with video oculography. No nystagmus was observed, and the patient did not complain of vertigo. The findings suggested the successful resolution of BPPV after treatment with the CRP. The patient was educated in the probability of the recurrence of BPPV and factors associated with recurrence. Further intervention was not indicated.
How did the results of the Cochrane review apply to Dr X?
The physical therapist considered whether Dr X would be a good candidate for the use of the CRP to treat right posterior canal BPPV. Using the PICO (Patient, Intervention, Comparison, Outcome) format, the clinician asked the following question: “Will a 40-year-old woman diagnosed with right posterior semicircular canal BPPV benefit from the CRP to resolve symptoms of vertigo and to convert from a positive to a negative Dix-Hallpike test result?” On the basis of history and findings on the Dix-Hallpike test, Dr X had a diagnosis of right posterior canal BPPV. The results of the Cochrane review may be applied directly to Dr X. The participants were similar in age and sex and had similar symptoms of positional vertigo.
Our procedure matched the published protocols of articles described in the Cochrane review. The CRP consisted of a series of 4 positions without the use of mastoid vibration.22,24–27 The duration of each position was the duration of nystagmus plus 20 seconds.25,26 Three cycles of the CRP were performed, and no nystagmus was observed during the second and third cycles. The CRP was repeated until no nystagmus was observed; alternatively, a maximum of 5 cycles can be performed to optimize success.22,25,26 After the maneuver, the patient was instructed in postural and activity restrictions, with the exception of wearing a soft cervical collar,25,26 to reduce the number of treatment sessions required to produce a negative Dix-Hallpike test result.
How well do the outcomes of the intervention provided to Dr X match those suggested by the systematic review?
In 1 treatment session, Dr X achieved her goal of the resolution of all symptoms of vertigo during her daily routine. The therapist achieved her goal of conversion from a positive to a negative Dix-Hallpike test result with no subjective complaints of vertigo. The successful resolution of posterior canal BPPV with 1 CRP treatment session is consistent with the reported Cochrane review findings.23 After 1 treatment session, the average short-term rates of success of the CRP for the complete resolution of symptoms ranged from 34% to 89%,22,24–27 with an average rate of 57% (SD=18%). For conversion from a positive to a negative Dix-Hallpike test result, the average short-term rates of success of the CRP ranged from 34% to 89%,22,24–27 with an average rate of 72% (SD=23%).23
Dr X's perceived level of handicap due to dizziness and perceived level of balance confidence in performing household and community activities returned to normal. Balance control during gait activities and speed of gait returned to age-appropriate levels. Participation and functional level goals were not included in the trials and therefore were not included in the Cochrane review. Dr X's positive outcomes exceeded those that were significant in the Cochrane review.
Can you apply the results of the systematic review to your own patients?
All 292 participants in the trials included in the Cochrane review were diagnosed with posterior canal BPPV on the basis of history and findings on the Dix-Hallpike test. After 1 treatment session, the average short-term rate of success of the CRP for complete symptom resolution was 57% (SD=18%), and that for conversion from a positive to a negative Dix-Hallpike test result was 72% (SD=23%).23 There were few reported adverse effects of treatment and no serious complications. The only reported limitations were emesis and cervical dysfunction preventing positioning of the head. The results of the Cochrane review23 may be applied to patients diagnosed with posterior canal BPPV. However, long-term follow-up was lacking in all of the trials. Therefore, there is no good evidence that the CRP provides a long-term resolution of symptoms. At the time of the search date, there was no good evidence comparing the CRP with other particle repositioning maneuvers, medical intervention, or surgical intervention for posterior canal BPPV.
Two double-blind randomized sham-controlled trials assessed the short-term efficacy of the liberatory maneuver, an alternative particle repositioning maneuver, for the treatment of posterior canal BPPV32 (Fig. 2B). At short-term follow-up, the rates of subjective symptom resolution for patients treated with the liberatory maneuver were 85%33 and 87%,34 and those obtained with the sham treatment were 0%34 and 14%.33 The presence of nystagmus in the same direction in the second and third steps suggested successful treatment.35 Side effects included transient nausea, vomiting, pallor, sweating, and loss of balance.33,34 No trials have been done to directly compare the liberatory maneuver with the CRP.
Other treatments for posterior semicircular canal BPPV include Brandt-Daroff exercises (habituation exercises).36,37 Subsequent to the Cochrane review search date, 2 trials were done to compare the CRP with Brandt-Daroff exercises, the liberatory maneuver, and no treatment (control). At 1 week, the short-term success rates were 81%36 and 87%37 for the CRP, 25%36 and 35%37 for the Brandt-Daroff exercises, 53% for the liberatory maneuver,37 and 15% for the control.37 Brandt-Daroff exercises were as effective as the control at 1 week.37
What can be advised based on the results of this systematic review?
The available data are sufficient to determine that the CRP is a safe and effective short-term treatment for posterior canal BPPV. After 1 treatment session, the CRP successfully resolves symptoms of vertigo and converts a positive Dix-Hallpike test result to a negative one. At short-term follow-up, individual trials and pooled data show a statistically significant effect in favor of the CRP over the control. The CRP is safe, with no serious complications. There is no good evidence that the CRP provides a long-term resolution of symptoms. At the time of the search, there was no good evidence comparing the CRP with other physical, medical, or surgical therapies for posterior canal BPPV. Future updates of the review need to take into account subsequent studies comparing the CRP with other physical therapy treatments for posterior canal BPPV, examining the effect size stratified by symptom duration, and comparing long-term follow-up of symptom resolution with the CRP and long-term follow-up of symptom resolution with other physical therapy treatments.
Appendix.
Cochrane Review Results23 for Effectiveness of the Canalith Repositioning Maneuver (CRP) in the Treatment of Posterior Semicircular Canal Benign Paroxysmal Positional Vertigo (BPPV)
Footnotes
Special thanks are due to Janet Callahan, PT, DPT, MS, NCS, Sarah Hourihan, PT, DPT, Rene Crumley, PT, DPT, NCS, CEEAA, and Tammie Ostrowski, PT, for assisting with the acquisition of digital recordings of nystagmus. Special thanks also are due to Eric Sanderson, PT, DPT, for illustrating Figure 3 (prediction of canalith repositioning procedure success and failure).
- Received June 7, 2013.
- Accepted May 26, 2014.
- © 2014 American Physical Therapy Association
References
Issue highlights
