Keywords vertigo - positional vertigo - central origin
Introduction
Benign paroxysmal positional vertigo (BPPV) is the most common form of peripheral
vertigo, representing ∼ 25% of all vertigos of vestibular origin. An incidence of
64/100,000 inhabitants per year is estimated in the United States, with a lifetime
prevalence of 2.4%. However, due to its high rate of spontaneous remissions, it is
very difficult to estimate its real incidence.[1 ] It is characterized by a recurrent vertigo, which lasts for seconds, is caused by
specific positions of the head in space, and is not accompanied by other otoneurologic
symptoms. It usually affects patients older than 45 years of age (up to 9% of the
population over 60 years old), and has a higher incidence among women, with a ratio
of 2:1.[1 ] The exact etiology of BPPV is still a matter of debate. In any case, more than 50%
of the cases are idiopathic, primary BPPV; the remaining cases are of secondary BPPV.[2 ] Among the latter, the most frequent cause is traumatic brain injury, including those
cases of mild severity (18–20%), and viral labyrinthitis (16%). Ischemia of the anterior
vestibular artery or surgery of the middle or inner ear (especially stapedectomy)
are possible less frequent causes.[3 ] Other factors associated with BPPV have also been described: Ménière disease, migraine,
vertebrobasilar insufficiency, prolonged bed rest, ototoxicity, depression over the
past year and cardiovascular risk factors, such as hypertension, hyperlipidemia, diabetes,
coronary disease and obesity. According to the latest research, the rates of involvement
of the posterior semicircular canal (PSC), the horizontal semicircular canal (HSC),
and the superior semicircular canal (SSC) are 60 to 90%, 10 to 25% and 1 to 2% respectively.[4 ] Although first described by Barany in 1921, the specific characteristics of this
disorder were defined in 1952 by Dix and Hallpike. The proposed treatments range from
observation (considering BPPV a self-limiting disorder) to aggressive surgical procedures,
such as section of the posterior ampullary nerve or occlusion of the posterior semicircular
canal. The conservative treatments, which are currently used and recommended, include
canalicular repositioning maneuvers and habituation exercises.[5 ]
[6 ] The former, described by Brandt and Daroff[7 ] and Norré and Beckers,[8 ] aim to increase the tolerance of the vestibular system to vertigo rather than eliminate
the cause, and are used both as an initial treatment and as a treatment secondary
to the failure of the repositioning maneuvers.
The repositioning maneuvers are all based on the two currently accepted etiopathogenic
theories, the theory of cupulolithiasis and the theory of canalolithiasis,[6 ] both described by Schuknecht. The Epley maneuver (EM) and the Semont maneuver are
for the posterior semicircular canal; the Gufoni and Lempert maneuver, for the horizontal
semicircular canal; and the Yacovino maneuver, for the superior semicircular canal.
The particle replacement maneuvers are very effective, especially in the case of the
posterior duct: it is estimated that they resolve 90 to 95% of the cases with an average
of 1 to 3 maneuvers[9 ]
[10 ]
[11 ]
[12 ] This resolution rate is lower for the HSC (85%),[11 ]
[13 ]
[14 ] and it is difficult to estimate, due to the scarce casuistry, the resolution rate
for the SSC.
However, there are patients (3.5 to 12%, depending on the series)[15 ]
[16 ] in whom the symptomatology and nystagmus persist despite the reiteration of the
maneuvers, and others in whom it recurs after the symptomatology has been solved (around
25%, especially in the first 6 to 12 months).
In these cases, it is always necessary to ratify the diagnosis, excluding other causes
of positional vertigo different from BPPV.[9 ] The pathologies that can be confused with BPPV are grouped into three categories:
otological, neurological and other entities.[17 ] Among the otological pathologies are Ménière disease, unilateral acute vestibular
syndrome (neuritis, labyrinthitis), superior semicircular canal dehiscence, dilated
vestibular aqueduct, and perilymphatic fistula.[17 ]
[18 ] The neurological causes include vestibular migraine, vestibular paroxysm, lesions
in structures around the IV ventricle, cerebellar lesions, brainstem lesions (demyelinating
lesions, tumors, and ischemic and degenerative lesions) and Arnold-Chiari disease.
We must bear in mind the association between BPPV and other vestibular pathologies
such as Ménière disease, vestibular neuritis and migraine, so that, in the presence
of a patient with specific symptoms, the possibility of presenting more than one vestibular
disorder simultaneously should be considered. Other entities that can simulate BPPV
are orthostatic hypotension, panic disorder or anxiety, drugs[3 ] (primidone, carbamazepine, phenytoin, lithium, tranquilizers, antihypertensives),
cervical rotational vertigo and positional convergence spasm.[19 ]
The objective of the following work is to assess the possible factors related to the
persistence of BPPV and the importance of Magnetic Resonance Imaging (MRI) scans in
the differential diagnosis of BPPV in cases that are refractory to the therapeutic
maneuvers described.
Methods
A review of all cases of BPPV diagnosed in the Otorhinolaryngology Service of a tertiary
hospital in the past 3 years with periodic follow-ups for a period of 6 months was
performed. The canalicular repositioning maneuvers were always performed by the same
otolaryngologists. A total of 176 patients were diagnosed and treated, and 150 of
them were diagnosed with BPPV due to PSC canalolithiasis; 20 patients had BPPV from
HSC; 10 of them had BPPV due to canalolithiasis; 10 of them had BPPV due to cupulolithiasis;
3 had BPPV due to SSC canalolithiasis; and 3 patients were diagnosed with PSC multicanal
BPPV and HSC, for a total of 176 treated BPPVs. For the diagnosis, the high suspicion
of BPPV was fundamentally based on the anamnesis. In addition, a complete otoneurologic
exploration was performed, with otomicroscopy, acoumetry, exploration of spontaneous
nystagmus in the five positions of the look and provoked nystagmuys by specific maneuvers.
The Dix Hallpike (DH) maneuver was performed to explore the PSC and the SCC, while
the McClure maneuver or Roll test was performed to explore the HSC.[12 ] At the first examination, nystagmus was observed with the naked eyes (approximately
half of the patients were diagnosed and initially treated at the Emergency Department).
In all cases with difficult exploration of the nystagmus, doubts about its direction
or poor response after one correctly executed maneuver, a videonistagmography was
performed. We used the EM to treat the vertigo of the SSC (n = 3) and PSC in all cases (n = 153), including those with involvement of more than one canal: a case of cupulolithiasis
of the left HSC + canalolithiasis of the right PSC (the patient needed 4 Ems, and
had subsequent instability), a case of cupulolithiasis of the right HSC + canalolithiasis
of the right PSC (4 EMs and subsequent instability) and a case of canalolithiasis
of the left HSC + canalolithiasis of the left PSC (1 EM). The “barbecue” manoeuver
was used to treat the BPPV of the HSC, both for the cases of canalolithiasis and cupulolithiasis.
Within the refractory BPPV we included 3 categories: 1) positional vertigo with persistence
of the nystagmus and of the symptomatology after 3 therapeutic maneuvers correctly
performed; 2) those patients who maintained instability after the execution of 3 maneuvers,
in spite of the negativization of the DH maneuver or of the Roll test; and 3) positional
vertigo that initially responded to the maneuvers with resolution of the symptoms
and the nystagmus, but recurred within a period of time shorter than 3 months. Finally,
we divided the patients into two groups according to the response to the treatment:
group A was composed of the cases refractory to the repositioning maneuvers, and group
B was composed of the patients with a good response. We analyzed the otoneurologic
background of every group to investigate a possible relation with the risk of refractory
BPPV. We also considered the cases with affectation of more than one canal at the
same time to evaluate if the presence of a multicanal BPPV was associated with a higher
risk of refractory vertigo. The statistical analyses were performed by an independent
statistician. For the comparisons between the two groups, odds ratio (OR) and Fisher
exact test were used, and the statistical significance was defined at p < 0.05. In all patients in whom it was necessary to repeat the maneuver on 3 or more
occasions, we performed an imaging study through an MRI with gadolinium, to rule out
other causes of BPPV, especially central ones.
Results
The mean age of the patients was 61.3 years, with a female preponderance and femaleto
male ratio of 2.6:1 (128 females and 48 males, 72.8% and 27.2% respectively). With
respect to the affected side, the right ear was more frequently affected (102 right
ears and 77 left ears, also considering the patients with multicanal vertigo: right
PSC + left HSC; right PSC + right HSC; left PSC + left HSC).
The efficacy of the EM at the first attempt reached 76.4% (117 patients out of 153
with negative DH maneuver after an EM), and it was necessary to perform the EM twice
in 18 patients (with negativization of the DH maneuver), 3 times in 11 patients (with
negativization of the DH maneuver) and 4 times in 7 patients (with negativization
of the DH maneuver in only 3 of them; in the remaining 4, Brandt-Daroff habituation
exercises were required). At 6 months, negativization of the DH maneuver was achieved
in 97.38% of all cases (149 out of of 153 patients). In 4 patients with persistent
positive DH (2 of them with a history of vestibular neuritis and refractory to 4 maneuvers)
and in 7 patients with persistent instability after 3 EMs but with negative DH (1
female patient with a history of stapedotomy in the 2 previous weeks), it was necessary
to resort to the Brandt-Daroff exercises and vestibular rehabilitation.
The BPPV of the HSC was treated with the “barbecue” maneuver described by Lempert,
both for the cases of canalolithiasis and cupulolithiasis. Including the multicanal
BPPV, we present 11 cases of canalolithiasis and 12 of cupulolithiasis. Of the 11
patients with canalolithiasis, in 8 (72.72%) the Roll test was negativized, with the
disappearance of the symptoms with a single maneuver. It was necessary to perform
it twice in the other 3 patients (27.28%). Of the 12 patients with cupulolithiasis,
in 7 (58.33%) 1 maneuver was sufficient; in 4 patients, 2 were performed; and, in
1 patient, the maneuver had to be performed 4 times. The rate of efficacy of the Lempert
maneuver was of 72.72% in the cases of canalolithiasis, and of 58.33% in the cases
of cupulolithiasis.
In the 3 patients who presented a multicanal BPPV, PSC and HSC, 1 canal was treated,
and after solving it, the other was treated. They were a cupulolithiasis of the left
HSC with canalolithiasis of the right PSC (4 EMs, 1 Semont maneuver and 2 Lempert
maneuvers were performed); a right HSC cupulolithiasis with right PSC canalolithiasis
(1 Lempert maneuver and 4 EMs were performed), and one left HSC canalolithiasis with
left PSC canalolithiasis (1 Lempert maneuver and 1 EM were performed).[12 ]
In all of the cases that required three or more maneuvers, with complete resolution
or not of the symptomatology, we requested a brain study with an MRI with gadolinium
to rule out central causes of paroxysmal positional vertigo.
In total, MRIs were performed in 19 patients:
Seven patients diagnosed with BPPV of posterior semicircular canal. Two suffered from
multicanal vertigo: one case of left HSC cupulolithiasis + right PSC canalolithiasis
and other case of right HSC cupulolithiasis + right PSC canalolithiasis. Out of these
7, in 4 the DH maneuver was not negativized.
Eleven patients diagnosed with BPPV of the PSC who needed 3 EMs, and in 3 of them
the instability persisted after the maneuver.
One patient diagnosed with BPPV due to cupulolithiasis of the right HSC, in which
4 Lempert maneuvers were performed, with persistence of the symptoms and nystagmus
with the Roll test.
After 19 MRIs, a focal thickening of 2.4mm of the right VIII cranial nerve (CN) pair
was diagnosed, and it was compatible with a small vestibular schwannoma in a 65-year-old
female patient with BPPV due to cupulolithiasis of the right HSC refractory to 4 Lempert
maneuvers ([Fig. 1 ]).
Fig. 1 (A) Magnetic resonance imaging scan (MRI) with gadolinium T2-weighted sequence; (B)
MRI with gadolinium sequence FIESTA. The thickening of the VIII cranial nerve in the
right internal auditory canal is observed.
Considering “normal” the realization of up to 3 canalicular repositioning maneuvers,
out of a total of 19 that required 3 or more maneuvers, we actually considered 11
cases refractory ([Table 1 ]):
Table 1
Characteristics of the patients with refractory vertigo
Canal involved
Number of maneuvers
History
MRI
Resolution
1
PSC
4 Epley
Vestibular Neuritis
Negative
DH positive;
Brandt-Daroff
2
PSC
4 Epley
Vestibular Neuritis
Negative
DH positive;
Brandt-Daroff
3
PSC
4 Epley
None
Negative
DH positive;
Brandt-Daroff
4
PSC
4 Epley
None
Negative
DH negative
5
PSC
4 Epley
None
Negative
DH negative
6
PSC + HSC
cupulolithiasis
4 Epley and 1 Lempert
None
Negative
DH negative;
Roll test negative
7
PSC + HSC
cupulolithiasis
4 Epley, 1 Semont and 2 Lempert
None
Negative
DH negative;
Roll Test negative
8
PSC
3 Epley
Stapedo-tomy
Negative
Instability
9
PSC
3 Epley
None
Negative
Instability
10
PSC
3 Epley
None
Negative
Instability
11
HSC cupulolithiasis
4 Lempert
None
Right VIII neurinoma
Roll test positive;
vestibular rehabilitation
Abbreviations: DH, Dix-Hallpike maneuver; HSC, horizontal semicircular canal; MRI, magnetic resonance
imaging; PSC, posterior semicircular canal.
Seven patients with BPPV of the PSC who underwent 4 EMs, of which 3 did not negativize
the DH maneuver (2 of them with a diagnosis of vestibular neuritis and 1 without a
history). Of the 4 that negativized the DH maneuver, 2 were multiple-canal BPPVs and
2 were BPPVs without otoneurologic history.
Three patients with BPPV of the PSC who maintained instability after 3 maneuvers (1
with a history of stapedotomy in the previous 15 days, and 2 without any history).
One case of vertigo due to cupulolithiasis of the right HSC in which 4 Lempert maneuvers
were performed (right vestibular schwannoma) without negativization of the Roll test
and with persistence of the symptomatology.
A total of 3 out of these 11 patients presented otoneurologic background, 2 of them
with history of vestibular neuritis (refractory to 4 maneuvers, without negativization
of the DH maneuver), and 1 patient had undergone a stapedotomy two weeks before the
onset of the symptoms (3 EMs with subsequent instability). Two other patients presented
a multicanal vertigo, respectively a case of cupulolithiasis of the left HSC + canalolithiasis
of the right PSC, and another one of cupulolithiasis of the right HSC with canalolithiasis
of the right PSC (4 EMs and negativization of the DH maneuver). Finally, one patient
was diagnosed with a right vestibular schwannoma associated with a cupulolithiasis
of the right HSC refractory to 4 maneuvers.
The other 5 patients (1 without negativization of the DH maneuver after 4 EMs, 2 with
negativization of the DH maneuver after 4 EMs, and 2 who maintained instability after
3 maneuvers) did not present a history of interest or other otoneurologic diagnosis.
In the 3 patients with instability after 3 EMs and in the 3 patients who did not negativize
the DH maneuver after 4 EMs, the habituation treatment was proposed with Brandt-Daroff
exercises and vestibular rehabilitation, and all of them presented improvement in
their symptoms.
The performance of an MRI with gadolinium was indicated in the cases that required
the performance of 3 therapeutic maneuvers and the symptomatology was resolved (8/19).
In none of them, the pathology was found through the MRI.
We divided the patients into two groups, refractory patients and good responders to
the repositioning manoeuvers, and observed several differences.
Among the refractory patients (11/176–6.25%), 4 of them (36.36%) presented otoneurologic
background or any otoneurologic pathology, and 2 (18.18%) other patients were diagnosed
with multicanal BPPV. We observed a female to male ratio of 9:2 (77.78% and 22.22%
respectively) and a mean age of 68.2 years. On the other hand, in the group of good
responders (165/176–93.75%), 15 patients presented otoneurologic background (9.09%),
only 1 was diagnosed with multicanal BPPV (0.60%), the female to male ratio was 119:46
(61.35% and 38.65% respectively), and the mean age was 59.98 years ([Table 2 ] and [Table 3 ]).
Table 2
Comparison of the two groups of patients
Otoneurologic background
Multicanal BPPV
F:M
Mean age
Group A (refractory vertigo;
n = 11)
4/11 (36.36%)
2/11 (18.18%)
9:2 (77.78%:22.22%)
68.2 years
Group B
(good responders; n = 165)
15/165 (9.09%)
1/165 (0.60%)
119:46 (61.35%:38.65%)
59.8 years
p -value (Fisher exact test, IC 95%)
p < 0.05
p < 0.05
p > 0.05
p > 0.05
Abbreviations: BPPV, benign paroxysmal positional vertigo; F, female; M, male.
Notes: p < 0.05: statistically significant. Fisher exact test was used to compare the groups.
Otoneurologic background and multicanal affection were associated with refractory
vertigo (p < 0.05). Sex and age didn't show any relation with refractory vertigo.
Table 3
Otoneurologic background of the two groups of patients
Otoneurologic background
Group A (n = 11)
Group B (n = 165)
n
%
n
%
Vestibular neuritis
2
18.18
2
1.21
Stapedectomy
1
9.09
2
1.21
Vestibular schwannoma
1
9.09
Ménière syndrome
4
2.42
Sudden hearing loss
2
1.21
Sudden tinnitus
2
1.21
Labyrinthine fistula
1
0.60
Vestibular migraine
2
1.21
The statistical results showed that the presence of otoneurologic background was strongly
associated with a higher risk of refractory BPPV (OR = 6.0952; 95% confidence interval
[95%CI] =1.5998–23.2234; p = 0.0081; Fisher exact test = 0.0199; 95%CI). When we analyzed every otoneurologic
antecedent, we observed a statistically significant association between refractory
vertigo and vestibular neuritis (OR = 18.1111; 95%CI = 2.2814–143.7740; p = 0.0061; Fisher exact test = 0.02; 95%CI). We also observed a statistically significant
association between multicanal vertigo and a major risk of refractory vertigo (OR = 36.4444;
95%CI = 3.0140–440.6768; p = 0.0047; Fisher exact test = 0.0103; 95%CI). We didn't observe a statistically significant
association between refractory vertigo and stapedectomy or vestibular schwannoma.
Female sex and age were not associated with a higher risk of poor response to the
treatment.
Discussion
Most cases of BPPV are idiopathic and respond successfully to canalicular repositioning
maneuvers, and some even resolve without any treatment, being self-limiting episodes.[15 ] Analyzing the results of previous studies, we observed that the failure rate of
repositioning maneuvers varies between 4.5% and 32%. In our series, the negativization
of the DH maneuver was achieved in 97.38% of the cases after 3 maneuvers, with 2.62%
(n = 3) of patients continuing to have a positive DH and another 2.62% (n = 3) with residual instability. All of them performed the Brandt-Daroff habituation
exercises and underwent vestibular rehabilitation with symptomatic improvement. In
case of lack of response to 3 correctly performed maneuvers, or of frequent recurrence
after initial improvement, other causes, especially neurological (central) and otological
other than BPPV, should be ruled out. Neurological causes include vestibular migraine,
vestibular paroxysm, lesions in structures around the IV ventricle, cerebellar lesions,
brainstem lesions (demyelinating, tumor, ischemic and degenerative) and Arnold-Chiari
disease. There is a series of warning signs in the exploration that may suggest the
presence of paroxysmal positional vertigo of central origin,[20 ] and that forces us to perform a MRI:
Presence of signs or symptoms of CNS pathology (such as a gaze-evoked nystagmus).
Absence of nystagmus inversion when inverting the position.
Position-induced marked hyperemesis.
A direction of the nystagmus that does not correspond to the duct that the maneuver
explores, especially in the following cases: either persistent lower vertical positional
nystagmus without torsional component or nystagmus that takes different directions
in the diagnostic maneuvers performed repeatedly (provided that no replacement maneuvers
have been performed, because, in this case, the particles could have moved from one
conduit to another).
Changing direction nystagmus without changes in the position of the head.
Spontaneous nystagmus in the absence of positional maneuvers (discarding the positional
pseudonystagmus observed in BPPV of HSC)
Nystagmus without dizziness in the positional tests.
Poor response to the therapeutic maneuvers.
Multiple recurrences confirmed with diagnostic maneuvers.
Among the otological /otoneurologic diseases that can be confused with BPPV or that
can be associated with it are chronic otitis media, Ménière syndrome, vestibular neuritis
and perilymphatic fistula.[21 ] The Ménière syndrome can begin with a BPPV-like symptomatology and be associated
with BPPV, and, in this case, the BPPV is considered secondary.
Perez et al[22 ] describe patients with Ménière syndrome in whom episodes of BPPV may precede, accompany
or follow the classic outbreaks of Ménière syndrome. That is why a careful follow-up
of the patients with BPPV is important.
Analyzing the series reported in the literature, in the series of 27 patients of Welling
and Barness,[23 ] 4 cases of refractory BPPV were found, and the presence of a perilymphatic fistula
after stapedotomy and a brain glioma were identified as causes of the failure of the
maneuvers. In the series of 90 patients described by Rupa,[15 ] out of the 7 refractory cases (8%), 1 was diagnosed with posterior fossa meningioma.
In our series of 176 cases, among the 11 refractory patients (6.25%), we identified
a case of neurinoma of the right VIII CN. Although the causal relationship between
BPPV and brain lesions is not known, a possible mechanism would be related to the
vascular compromise of the labyrinth due to the ischemia of the anterior vestibular
artery caused by the tumor.[24 ]
It is noteworthy that 4 out of the 11 cases considered refractory presented some otoneurologic
background or otoneurologic pathology associated to BPPVs of more difficult control
(2 cases of vestibular neuritis, 1 stapedectomy and 1 vestibular schwannoma), while
only 15 (9.09%) of the 165 patients with a good response presented a similar otoneurologic
background. Although the number of patients is small, the results showed a statistically
significant association between an otoneurologic background and refractory BPPV (p < 0.05), especially for vestibular neuritis. We also demonstrated that multicanal
BPPV is more difficult to manage than simple BPPV, and it is associated with refractory
vertigo, probably because it is more difficult to diagnose and requires special attention
and knowledge of the pathophysiological patterns of positional vertigo. We must also
bear in mind that the treatment of HSC is usually more complex, especially in the
case of cupulolithiasis, and that BPPV of different canals sometimes coexist; therefore,
when treating, for example, the PSC and not the HSC, the patient continues to present
the symptoms. It is also possible to convert a BPPV of the PSC to a BPPV of the HSC,
probably by passing the otoliths from the PSC to the HSC through the utricle.[25 ] In our series we had three cases of multiple-canal BPPV with involvement of the
PSC and HSC, and two of these required several maneuvers to be solved, with a statistically
significant difference.
Finally, if a correct localization diagnosis has been made, identifying the semicircular
canal involved and differentiating between canalolithiasis and cupulolithiasis, and
if the specific maneuvers were not successful, we can use the Brandt-Daroff exercises,
which, in some cases, have been shown to be more effective than the repositioning
maneuvers.[6 ]
The present article has several methodological limitations. It is a retrospective
study, so it is possible that bias occurred during the data collection. Since BPPV
can precede, accompany or follow the classic symptoms of Ménière syndrome or vestibular
neuritis, it is possible that several patients could be affected by an undiagnosed
otoneurologic disease of which the BPPV was only the first manifestation. Another
limitation is the small number of patients with a poor response, only 11(6.25%) cases
versus 165(93.75%) of good responders, and this could influence the results observed
in the study.
Conclusions
Although the EM, as well as the “barbecue” maneuver, if correctly performed, enable
physicians to reach very high success rates, some patients are more difficult to treat,
and an accurate differential diagnosis is mandatory. We observed that the presence
of an otoneurologic background, especially vestibular neuritis and multicanal involvement,
is statistically associated with cases of refractory BPPV, and we recommend performing
an MRI with gadolinium to rule out vertigo of central origin. Despite our results,
more research is needed to confirm the relationship between otoneurologic pathology
and refractory BPPV.
