Corticosteroids for the treatment of idiopathic acute

Transcription

vestibular dysfunction (vestibular neuritis) (Review)
Fishman JM, Burgess C, Waddell A
This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library
2011, Issue 5
http://www.thecochranelibrary.com
Corticosteroids for the treatment of idiopathic acute vestibular dysfunction (vestibular neuritis) (Review)
Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
TABLE OF CONTENTS
HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SUMMARY OF FINDINGS FOR THE MAIN COMPARISON . . . . . . . . . . . . . . . . . . .
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 1.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 2.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 3.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 4.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 5.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 6.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ACKNOWLEDGEMENTS
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 1.1. Comparison 1 Corticosteroids versus control/placebo, Outcome 1 Complete caloric recovery at 1 month
(defined as lateralisation in caloric testing < 25%). . . . . . . . . . . . . . . . . . . . . .
Analysis 1.2. Comparison 1 Corticosteroids versus control/placebo, Outcome 2 Complete caloric recovery at 12 months
(defined as lateralisation in caloric testing < 25%). . . . . . . . . . . . . . . . . . . . . .
Analysis 1.3. Comparison 1 Corticosteroids versus control/placebo, Outcome 3 Vertigo at 24 hours. . . . . . .
Analysis 1.4. Comparison 1 Corticosteroids versus control/placebo, Outcome 4 Dizziness Handicap Inventory score.
Analysis 1.5. Comparison 1 Corticosteroids versus control/placebo, Outcome 5 Pathological findings on
electronystagmography at 1 month. . . . . . . . . . . . . . . . . . . . . . . . . . .
electronystagmography at 12 months. . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 1.7. Comparison 1 Corticosteroids versus control/placebo, Outcome 7 Improvement (as %) in lateralisation of
ENG caloric test at 1 month. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 1.8. Comparison 1 Corticosteroids versus control/placebo, Outcome 8 Improvement (as %) in lateralisation of
ENG caloric test at 12 months. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 1.9. Comparison 1 Corticosteroids versus control/placebo, Outcome 9 Mean time to recovery. . . . . .
APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DIFFERENCES BETWEEN PROTOCOL AND REVIEW . . . . . . . . . . . . . . . . . . . . .
NOTES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
INDEX TERMS
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1
1
2
2
5
6
6
8
10
13
14
15
15
16
17
17
18
18
19
26
27
27
28
29
30
30
31
32
32
33
35
35
35
36
36
36
36
i
[Intervention Review]
vestibular dysfunction (vestibular neuritis)
Jonathan M Fishman1 , Chris Burgess2 , Angus Waddell2
1 UCL
Institute of Child Health, London, UK. 2 ENT Department, Great Western Hospital, Swindon, UK
Contact address: Jonathan M Fishman, UCL Institute of Child Health, 30 Guilford Street, London, WC1N 1EH, UK.
[email protected].
Editorial group: Cochrane Ear, Nose and Throat Disorders Group.
Publication status and date: New, published in Issue 5, 2011.
Review content assessed as up-to-date: 27 December 2010.
Citation: Fishman JM, Burgess C, Waddell A. Corticosteroids for the treatment of idiopathic acute vestibular dysfunction (vestibular
neuritis). Cochrane Database of Systematic Reviews 2011, Issue 5. Art. No.: CD008607. DOI: 10.1002/14651858.CD008607.pub2.
ABSTRACT
Background
Idiopathic acute vestibular dysfunction (vestibular neuritis) is the second most common cause of peripheral vertigo after benign
paroxysmal positional vertigo (BPPV) and accounts for 7% of the patients who present at outpatient clinics specialising in the treatment
of dizziness. The exact aetiology of the condition is unknown and the effects of corticosteroids on the condition and its recovery are
uncertain.
Objectives
To assess the effectiveness of corticosteroids in the management of patients with idiopathic acute vestibular dysfunction (vestibular
neuritis).
Search methods
We searched the Cochrane ENT Group Trials Register; CENTRAL; PubMed; EMBASE; CINAHL; Web of Science; BIOSIS Previews;
Cambridge Scientific Abstracts; ICTRP and additional sources for published and unpublished trials. The date of the most recent search
was 28 December 2010.
Selection criteria
Randomised controlled trials comparing corticosteroids with placebo, no treatment or other active treatments, for adults diagnosed
with idiopathic acute vestibular dysfunction.
Data collection and analysis
Two authors independently selected studies from the search results and extracted data. Three authors independently assessed risk of
bias.
Main results
Four trials, involving a total of 149 participants, compared the effectiveness of oral corticosteroids against placebo. All the trials were
small and of low methodological quality. Although there was an overall significant effect of corticosteroids compared with placebo
medication on complete caloric recovery at one month (risk ratio (RR) of 2.81; 95% confidence interval (CI) 1.32 to 6.00, P = 0.007),
no significant effect was seen on complete caloric recovery at 12 months (RR 1.58; 95% CI 0.45 to 5.62, P = 0.48), or on the extent of
1
caloric recovery at either one month (mean difference (MD) 9.60%; 95% CI -20.66 to 39.86, P = 0.53) or at 12 months (MD 6.83%;
95% CI -27.69 to 41.36, P = 0.70). In addition, there was no significant difference between corticosteroids and placebo medication in
the symptomatic recovery of vestibular function following idiopathic acute vestibular dysfunction with respect to vertigo at 24 hours
(RR 0.39; 95% CI 0.04 to 3.57, P = 0.40) and use of the Dizziness Handicap Inventory score at one, three, six and 12 months.
Authors’ conclusions
Overall, there is currently insufficient evidence from these trials to support the administration of corticosteroids to patients with
idiopathic acute vestibular dysfunction. We found no trials with a low risk of methodological bias that used the highest level of diagnostic
criteria and outcome measures. We recommend that future studies should include health-related quality of life and symptom-based
outcome measures, in addition to objective measures of vestibular improvement, such as caloric testing and electronystagmography.
PLAIN LANGUAGE SUMMARY
Corticosteroids for the treatment of idiopathic acute vestibular dysfunction (vestibular neuritis)
Idiopathic acute vestibular dysfunction (vestibular neuritis) is a common condition of unknown cause. Patients with the condition
often experience dizziness, nausea or vomiting, and trouble with vision, balance or mobility, but have normal hearing and no tinnitus.
It has been proposed that a course of corticosteroids, if given early on, may improve recovery from the condition and long-term patient
outcome. However, corticosteroids can cause adverse effects (e.g. bleeding stomach ulcer, mood changes, etc.).
This review identified four randomised controlled trials including 149 adult patients with idiopathic acute vestibular dysfunction
(vestibular neuritis) treated with either corticosteroids or placebo. The studies were varied in that they used different drugs and different
treatment regimens. On the basis of these studies, there is currently insufficient evidence in favour of corticosteroids over placebo
medication in the symptomatic recovery and objective testing of vestibular function, both in the short-term and long-term. Further
studies of higher quality are needed to test the effectiveness of corticosteroids in patients with the condition.
2
3
The mean Dizziness
Handicap Inventory score
in the intervention group
was
6.48 higher
(7.52 lower to 18.51
higher)
The mean caloric extent of recovery (improvement as a % in lateralisation of electronystagmography on caloric testing)
was
9.60 higher
(20.66 lower to 39.86
higher)
Caloric extent of recovery (improvement as
a % in lateralisation
of electronystagmography on caloric testing)
Follow up: 1 month
338 per 1000
(35 to 1000)
Corticosteroids
Placebo
867 per 1000
Corresponding risk
Assumed risk
Illustrative comparative risks* (95% CI)
Dizziness Handicap Inventory score
Follow up: 1 to 12 months
Vertigo
Follow up: 24 hours
Outcomes
RR 0.39
(0.04 to 3.57)
Relative effect
(95% CI)
Patient or population: Patients with idiopathic acute vestibular dysfunction (vestibular neuritis)
Settings: Outpatient or emergency department
Intervention: Corticosteroids
Comparison: Placebo
Corticosteroids compared to placebo for idiopathic acute vestibular dysfunction (vestibular neuritis)
30
(1 study)
30
(1 study)
60
(2 studies)
No of participants
(studies)
S U M M A R Y O F F I N D I N G S F O R T H E M A I N C O M P A R I S O N [Explanation]
⊕⊕
low1,2
⊕⊕
low1,2
⊕⊕
low1,2
Quality of the evidence
(GRADE)
Comments
4
593 per 1000
(279 to 1000)
702 per 1000
(200 to 1000)
See comment
Complete caloric recov- 211 per 1000
ery
Follow up: 1 month
Complete caloric recov- 444 per 1000
ery
Follow up: 12 months
Adverse effects
Not estimable
RR 1.58
(0.45 to 5.62)
RR 2.81
(1.32 to 6)
73
(1 study)
89
(2 studies)
50
(2 studies)
89
(2 studies)
⊕⊕⊕
moderate2
⊕⊕
low1,2
⊕⊕
low1,2
⊕⊕
low1,2
1 out of 35 patients in the
corticosteroids group experienced severe adverse
effect (bleeding gastric ulcer)
2
1
Quality downgraded because of limitations in the design and implementation.
Quality downgraded because of small sample sizes and broad confidence intervals.
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the
assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: confidence interval; RR: risk ratio
See comment
The mean caloric extent of recovery (improvement as a % in lateralisation of electronystagmography on caloric testing)
was
6.83 higher
(27.69 lower to 41.36
higher)
Caloric extent of recovery (improvement as
a % in lateralisation
of electronystagmography on caloric testing)
Follow up: 12 months
BACKGROUND
Description of the condition
Vertigo is defined as a disabling sensation in which the affected
individual feels that they or their environment are in a state of constant movement, most often of a rotatory nature (Martin 1998).
Vertigo may have central nervous system or peripheral causes.
’Peripheral’ vertigo refers to vertigo that originates from pathology of the motion-sensing organs of the inner ear or the associated vestibular nerves. Causes of peripheral vertigo include benign
paroxysmal positional vertigo (BPPV), Ménière’s disease and acute
labyrinthitis. Vestibular neuritis is the second most common cause
of peripheral vertigo after BPPV (Strupp 2004). It accounts for
7% of the patients who present at outpatient clinics specialising
in the treatment of dizziness and survey data from Japan suggest
an annual incidence of around 3.5 cases per 100,000 population,
although this figure is considered likely to be an underestimate
of the true incidence of the condition (Neuhauser 2007; Sekitani
1993; Strupp 2004).
Vestibular neuritis is manifested clinically by a sudden onset of
sustained vertigo and imbalance accompanied by characteristic
rapid and involuntary eye movements referred to as nystagmus.
Additional signs of vestibular neuritis include a positive Romberg’s
sign (i.e. when the eyes are closed the patient falls towards the
affected ear) and a positive head-thrust test. Associated autonomic
features such as nausea and vomiting are often present. Audiometry (hearing tests) and brain imaging are normal in the condition. Specific tests of vestibular function (electronystagmography
and caloric testing) can document the unilateral loss of vestibular
function but they are rarely necessary (Baloh 2003).
The symptoms of vestibular neuritis usually last between three
and seven days followed by gradual resolution. These vestibular
symptoms and signs occur in the absence of any significant hearing
impairment, supporting the theory that the underlying pathology
does not affect the entire labyrinth (inner ear) and is selective for
the vestibular nerve. Although vestibular neuritis is usually unilateral, bilateral involvement may occur (Scott-Brown 1997). A
clear problem with previous studies that have attempted to evaluate treatments for vestibular neuritis is that there are no wellaccepted diagnostic criteria for the disorder. By including anyone
with isolated attacks of vertigo, these studies may include a range
of different conditions.
The aetiology of vestibular neuritis is currently unknown and
therefore we favour use of the term ’idiopathic acute vestibular
dysfunction’; a term which is used interchangeably with ’vestibular
neuritis’ throughout this review. The most likely cause is thought
to be viral (Bartual-Pastor 2005; Davis 1993). The evidence, however, remains circumstantial and others have implicated microvascular disturbances in the aetiology of the condition (vascular hypothesis) (Nadol 1995; Schulz 1998). Clinical supporting evidence that viral agents are responsible for this condition (viral hypothesis) is based on epidemiological evidence of an increased in-
cidence of this vestibular condition during an epidemic of viral
infections, and clinical evidence that an upper respiratory viral disorder frequently precedes the vestibular syndrome (Coats 1969;
Davis 1993; Shahle 1966). Postmortem studies on patients who
were thought to have suffered vestibular neuritis during their lifetime have shown atrophy of the vestibular nerve and the vestibular
sensory epithelium. These changes are similar to the histopathological findings in known viral disorders, such as herpes zoster oticus (Schuknecht 1981). In addition, herpes simplex virus type 1
(HSV-1) DNA has been detected on autopsy with the use of the
polymerase chain reaction in about two of three human vestibular ganglia (Arbusow 1999; Arbusow 2000; Furuta 1993; Schulz
1998). This indicates that the vestibular ganglia are latently infected by HSV-1.
Additional histopathological observations in this disorder include
degeneration of the vestibular (Scarpa’s) ganglion and its processes
in the presence of a normal auditory end organ and nerve. A
round cell infiltrate is also frequently observed surrounding the
vestibular nerve fibres in the internal auditory meatus (Baloh 1996;
Bartual-Pastor 2005; Davis 1993).
Recovery of functionally normal balance after vestibular neuritis
typically takes several weeks, although longer periods of recovery may be encountered. Recovery from a peripheral vestibular
lesion results from a combination of the restoration of peripheral
labyrinthine function (which is usually incomplete in the case of
vestibular neuritis) and gradual adaptation of the brain to the imbalance in vestibular function (central compensation). However,
previous studies have confirmed the lack of correlation between recovery of clinical symptoms and caloric recovery (Bergenius 1999;
Ohbayashi 1993). Furthermore, up to 20% of patients with the
condition experience continued symptoms such as impaired vision
and postural imbalance during walking and especially during head
movements, despite normalisation of caloric testing (Bronstein
2007).
Description of the intervention
Despite the assumed aetiology of vestibular neuritis, the effects of
corticosteroids on the condition and its recovery are uncertain.
How the intervention might work
The best treatment for a patient who presents with an acute episode
of peripheral vestibular loss is controversial, because the pathophysiology is uncertain. Assuming that vestibular neuritis is the
result of a viral or postviral inflammation of the vestibular nerve,
treatment aimed at stopping the inflammation has been proposed.
The anti-inflammatory effect of corticosteroids, which results in
reduced swelling and mechanical compression of the vestibular
nerve within the temporal bone, may explain why treatment with
corticosteroids results in improvement. Parens 1999 showed that
5
prednisolone and hydrocortisone can be traced in endolymph after systemic administration.
In addition, in animal experiments, corticosteroids have been reported to have an effect not only on the inflammatory process,
but also on central vestibular compensation mechanisms, resulting in enhanced vestibular compensation following acute peripheral vestibular insults that result in unilateral labyrinthine deficits
(Cameron 1999; Jerram 1995; Yamanaka 1995a; Yamanaka
1995b).
Why it is important to do this review
Vestibular neuritis is the second commonest cause of vertigo (
Strupp 2004), causes significant incapacitation for a considerable
number of patients and imposes an important financial and social
burden on patients and healthcare institutions alike.
There have been a number of randomised controlled trials. However, each has used different regimes, at different doses, in different
settings and some with different objectives. Reviews on this topic
to date are incomplete and non-systematic. There has been one
recent meta-analysis published following acceptance of our title
by the Cochrane Ear, Nose and Throat Disorders Review Group
on the effectiveness of corticosteroids as a treatment for vestibular
neuritis (Goudakos 2010). However, a clear problem with studies
that have attempted to evaluate treatments for vestibular neuritis is
that there are no well-accepted diagnostic criteria for the disorder.
By including anyone with isolated attacks of vertigo, these studies
may include a range of different conditions.
There is uncertainty as to (i) whether corticosteroids are better
than placebo or no treatment in the management of vestibular
neuritis (idiopathic acute vestibular dysfunction) and (ii) whether
any perceived benefit has any benefit on long-term functional outcomes. A Cochrane Review is therefore warranted to assess the
benefits and harms of corticosteroids for vestibular neuritis, in relation to short-term recovery and long-term functional outcomes.
OBJECTIVES
1. To study the effectiveness of corticosteroids in the recovery
of peripheral vestibular function in patients with idiopathic acute
vestibular dysfunction.
2. To determine the adverse effects of these medications.
METHODS
Criteria for considering studies for this review
Types of studies
Randomised controlled trials.
Types of participants
Inclusion criteria
Adults (age > 16 years old) of either gender, diagnosed with vestibular neuritis (idiopathic acute peripheral vestibular dysfunction/
vestibulopathy, rather than acute cochleo-vestibular dysfunction),
as defined by the following criteria:
1. first episode of sudden onset sustained vertigo measured in
days;
2. absence of auditory symptoms or findings suggestive of
alternative diagnoses;
3. absence of neurological signs other than spontaneous
nystagmus (unidirectional, horizontal, obeying Alexander’s Law
and enhancing with removal of optic fixation), a positive headthrust test or a positive Romberg’s test;
4. absence of neurological symptoms or findings suggestive of
alternative diagnoses.
Treatment must be initiated within seven days of the onset of
symptoms.
Exclusion criteria
We excluded patients with any other cause of acute vertigo (e.g.
benign paroxysmal positional vertigo, Ménière’s disease).
Types of interventions
Any corticosteroids (any timing, any dose, by oral/intravenous/intramuscular/intratympanic route and of any duration), including
medications such as prednisolone, dexamethasone, methylprednisolone, etc.
Corticosteroids were compared to placebo, no treatment, any active intervention and/or any active comparator.
Types of outcome measures
Primary outcomes
1. Proportion of patients that recover; and/or
2. Degree of recovery (if appropriate) of peripheral vestibular
function in patients with idiopathic acute vestibular dysfunction,
through subjective patient reporting of symptoms, use of
validated questionnaires, or objective evidence of recovery (e.g.
electronystagmography, caloric testing and other vestibular
function and balance tests), as defined by the authors.
6
Secondary outcomes
Searching other resources
1. Time to recovery, including a return to normal activities
and health-related quality of life measures (generic, or diseasespecific, or both).
2. Patient-reported adverse events: severe and minor. A severe
adverse event is defined as resulting in a patient discontinuing
the medication and withdrawing from the study. A minor
adverse event is defined as a side effect experienced by the patient
but where they continue to take the medication.
We scanned reference lists of identified publications for additional
trials and contacted authors as necessary. We searched PubMed,
TRIPdatabase, NHS Evidence - ENT & Audiology, and Google
to retrieve existing systematic reviews possibly relevant to this systematic review, so that we could scan their reference lists for additional trials.
Data collection and analysis
Search methods for identification of studies
We conducted systematic searches for randomised controlled trials. There were no language, publication year or publication status
restrictions. We contacted original authors for clarification and
further data if trial reports were unclear, and we arranged translations of papers where necessary. The date of the most recent search
was 28 December 2010.
Selection of studies
Two authors (JMF and CB) independently screened the results
of the search to identify studies which loosely met the inclusion
criteria of the review. We obtained these studies in full text and
applied the inclusion criteria independently. We resolved any disagreement over which studies to include by discussion, or by referral to the third author (AW).
Data extraction and management
Electronic searches
We identified published, unpublished and ongoing studies by
searching the following databases from their inception: the
Cochrane Ear, Nose and Throat Disorders Group Trials Register; the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2010, Issue 4); PubMed; EMBASE;
CINAHL; LILACS; KoreaMed; IndMed; PakMediNet; CAB Abstracts; Web of Science; BIOSIS Previews; CNKI; mRCT; ClinicalTrials.gov; ICTRP and Google.
We modelled subject strategies for databases on the search strategy
designed for CENTRAL. Where appropriate, we combined subject strategies with adaptations of the highly sensitive search strategy designed by the Cochrane Collaboration for identifying randomised controlled trials and controlled clinical trials (as described
in the Cochrane Handbook for Systematic Reviews of Interventions
Version 5.1.0, Box 6.4.b. (Handbook 2011)). Search strategies for
major databases including CENTRAL are provided in Appendix
1.
Two authors (JMF and CB) extracted data independently using
standardised, pre-piloted data extraction forms. Where data were
missing or unclear, JMF contacted the authors of the trial for
unpublished data or clarification.
Assessment of risk of bias in included studies
We assessed the quality of the included studies using the ’Risk
of bias’ tool mentioned in the Cochrane Handbook for Systematic
Reviews of Interventions (Handbook 2011, version 5.1.0) and constructed a ’Risk of bias’ table. The authors (JMF, CB, AW) independently assessed all studies. Disagreements were resolved by
discussion.
We rated individual criteria on validity as ’low’ risk of bias, ’high’
risk of bias or ’unclear’ risk of bias. If a criterion was rated as
’unclear’, we made an attempt to obtain further information from
the authors of the specific study. We summarised the criteria to
derive an overall assessment of validity and gave a grade (A, B or
C).
Risk of bias
Interpretation
Relationship to individual
criteria
Grade A: low risk of bias
Plausible bias unlikely to seriously alter the results All of the criteria met
Grade B: moderate risk of bias
Plausible bias that raises some doubt about the One or more criteria partly met
results
7
(Continued)
Grade C: high risk of bias
Plausible bias that seriously weakens confidence One or more criteria not met
in the results
The results of this bias assessment are displayed as a ’Risk of bias’
summary figure generated by RevMan 5.1 software (Handbook
2011; RevMan 2011).
Measures of treatment effect
For dichotomous data, we calculated individual and pooled statistics as risk ratio (RR) with 95% confidence intervals (CI). We
analysed continuous data using mean differences (MD) and standardised mean differences (SMD) with 95% confidence intervals.
Dealing with missing data
We performed all analyses on an intention-to-treat basis. Whenever possible, we contacted the original investigators to request
missing data, made explicit the assumptions of any methods used
to cope with missing data, performed sensitivity analyses to assess
how sensitive results were to reasonable changes in the assumptions that were made and addressed the potential impact of missing data on the findings of the review in the assessment of risk of
bias and in the Discussion section of the review.
Assessment of heterogeneity
We assessed heterogeneity by visually inspecting the overlap of
confidence intervals for the results of individual studies, as depicted
graphically using horizontal lines. More formally, we assessed the
impact of heterogeneity using the Chi2 test and I2 statistic available
in RevMan 5.1 (RevMan 2011).
Assessment of reporting biases
We conducted a sensitivity analysis, where possible, on small-study
effects. We considered use of a funnel plot to investigate the possibility of positive outcome (publication) bias if there were a sufficient number of studies available.
We sought statistical advice. If we were able to obtain data in the
form of means and standard deviations, we summarised effects
on outcomes as standardised mean differences (SMD) with their
95% confidence intervals (CI). We performed all analyses on an
intention-to-treat basis. We performed our statistical analysis using Review Manager 5.1 (RevMan 2011). Where possible, we calculated pooled estimates using a fixed-effect model. However, if
there was substantial statistical heterogeneity (an I2 value > 50%,
as specified in the Cochrane Handbook for Systematic Reviews of
Interventions), or clinical heterogeneity, we used a random-effects
model (Handbook 2011). We considered a sensitivity analysis to
compare fixed and random-effects estimates.
We produced a ’Summary of findings’ table using GRADEpro
2008.
Subgroup analysis and investigation of heterogeneity
Depending upon the available data the following further subgroup
analyses appeared of interest:
1. choice of corticosteroid;
2. dose of corticosteroid;
3. route of administration;
4. duration of treatment; and
5. timing of treatment.
The possibility of analysing such subgroups depended on the number of studies available and whether the studies were of sufficient
quality. Although the data did not allow for this on this occasion,
it will be considered in updates (if feasible).
Sensitivity analysis
If the data obtained were suitable for meta-analysis, we planned to
carry out a sensitivity analysis and present the results in a summary
table. A sensitivity analysis and a summary measure of effect might
not be appropriate if the number of included trials is low. The
data did not allow for this on this occasion, however it will be
considered in updates (if feasible).
Data synthesis
We extracted data from the included published studies and entered
them into RevMan 5.1 (RevMan 2011) for statistical analysis.
Where data were insufficient or incomplete, we made an attempt
to contact the authors to obtain further data. If appropriate we
carried out a meta-analysis.
RESULTS
Description of studies
8
See: Characteristics of included studies; Characteristics of excluded
studies; Characteristics of ongoing studies.
Results of the search
The electronic search resulted in the initial identification of 226
references. After removal of duplicates and an initial sift for relevance we were left with 49 publications. We screened titles and
abstracts of all 49 references, resulting in 12 potentially eligible
articles. We obtained the full texts of all 12 studies and assessed
them for eligibility. Four randomised controlled trials met the inclusion criteria.
In addition, we handsearched the references of all the studies for
which the full text was retrieved. However, we identified no additional studies that could provide data to answer the research question.
A study flow diagram is shown in Figure 1, according to the template described in the PRISMA statement (Liberati 2009).
9
Figure 1. Study flow diagram.
10
Included studies
Design
Four randomised controlled trials fulfilled the inclusion criteria and are included in this review (Ariyasu 1990; Rezaie 2006;
Shupak 2008; Strupp 2004). Full details are provided in the
Characteristics of included studies table. All four trials compared
corticosteroids with placebo.
Sample sizes
All four randomised controlled trials that met the inclusion criteria
contained small numbers of patients.
The study by Ariyasu 1990 contained 20 patients (10 patients
in each arm of the trial); the study by Rezaie 2006 contained 40
patients (20 in each arm of the trial); the study by Shupak 2008
contained 30 patients (15 patients in each arm of the trial) and
the study by Strupp 2004 contained 73 patients (38 patients in
the placebo arm and 35 patients in the treatment arm). In the
latter study, eight patients in the placebo arm and six patients in
the treatment arm of the trial were excluded from the trial leaving
59 patients (30 patients in the placebo arm and 29 patients in the
treatment arm).
Therefore a total of 149 patients participated in the four studies,
with a mean sample size of 37.25 and a range of 20 to 59.
Sample size calculations (power analyses) were reported by Strupp
2004 and Rezaie 2006, but not by Ariyasu 1990 and Shupak 2008.
Omission of sample size calculations from the latter, smaller studies, with probable poor statistical power, was a likely methodological flaw.
Setting
Patients were recruited from a variety of settings including emergency departments (Strupp 2004), outpatient clinics (Rezaie
2006), or both (Shupak 2008). The source of patients was not
clarified in the study by Ariyasu 1990.
Participants
The studies included patients aged between 15 and 80 years old.
The median ages were 42.45 (Ariyasu 1990) and 35 (Rezaie 2006)
years. The mean ages were 48.25 (Strupp 2004) and 47.45 (Shupak
2008) years. The study by Rezaie 2006 included a single patient
aged 15 years. Although we specified a lower limit of 16 years of
age in our inclusion criteria, we did not feel this merited exclusion
of this study from the review.
The studies differed in the diagnostic criteria used to define the
participants as suffering from idiopathic acute vestibular dysfunction.
In the study by Ariyasu 1990, the diagnosis of idiopathic acute
vestibular dysfunction was based on a history consistent with acute
rotatory vertigo suggestive of vestibular manifestations without
cochlear involvement, supported by pure-tone audiometry within
normal limits, bilateral symmetrical, clinical observable rotatory,
spontaneous or positional nystagmus confirmed by electronystagmography and a unilaterally decreased caloric response of greater
than 25%. To eliminate the possibility of other neurological disorders, a normal neurological physical examination was required,
including a test of cranial nerves. To be eligible for the study, patients required vertiginous symptoms within 72 hours of evaluation.
In the study by Rezaie 2006, a diagnosis of idiopathic acute
vestibular dysfunction was based on audiometric and other laboratory tests, as well as blood pressure measurement, examining
cardiac status, cerebral magnetic resonance imaging and caloric
tests. Patients with hearing loss or tinnitus were excluded from the
study.
In the study by Shupak 2008, a diagnosis of idiopathic acute
vestibular dysfunction was only reached if all the following symptoms and signs were present:
1. acute onset of prolonged, severe rotatory vertigo;
2. the presence of spontaneous nystagmus and postural
imbalance;
3. documentation of unilaterally reduced caloric response
(caloric lateralisation > 25%);
4. absence of new hearing loss, tinnitus or neurological
deficits; and
5. patients without a history of vestibular dysfunction.
In the study by Strupp 2004, the diagnosis of idiopathic acute
vestibular dysfunction was based on the following criteria:
1. history of acute or subacute (i.e. within minutes to hours)
onset of severe, prolonged rotatory vertigo, nausea and postural
imbalance;
2. absence of cochlear symptoms, such as tinnitus or acute
hearing loss before, during, or after the onset of vertigo;
3. no history of vestibular dysfunction before the acute onset
of symptoms;
4. on clinical examination, horizontal, spontaneous nystagmus
with a rotational component towards the unaffected ear (fast
phase), without evidence of a central vestibular lesion and a headthrust test (performed by turning the head of the patient rapidly
to the right and left to provoke compensatory eye movements)
showing an ipsilateral deficit of the horizontal semicircular canal;
5. caloric irrigation showing hyporesponsiveness or lack of
responsiveness of the horizontal canal of the affected ear (the
11
maximal slow-phase velocity during caloric irrigation with water
at 30 °C and 40 °C should be less than three degrees per second
on the affected side, and the asymmetry between the two sides
should be more than 25%);
6. rotation of the eyes towards the affected ear without
showing vertical divergence of one eye above the other;
7. absence of other signs or symptoms of brainstem or
cerebellar disorders; and
8. normal findings on cranial magnetic resonance imaging.
To be eligible for the study patients were required to have symptoms within three days of recruitment.
Interventions
All four trials compared corticosteroids with placebo, but used
different treatment regimens (different choice of corticosteroid,
different doses, different durations of treatment, different weaning
strategies and different timings of treatment in relation to the onset
of symptoms). In some studies the patients also received additional
treatments, such as vestibular sedatives or antiemetics. In all four
studies, corticosteroids were administered via the oral route. The
intratympanic route was not utilised in any of the trials.
In the study by Ariyasu 1990, 32 mg methylprednisolone was
given orally on the first day and then decreased to 4 mg gradually
over the next seven days. Placebo (lactose) contained no active
ingredient except a trace of quinine sulphate to impart a bitter
taste identical to that of the methylprednisolone. No mention was
made in the study of any other adjuvant treatments being given
(such as vestibular sedatives, or antiemetics).
In the study by Rezaie 2006, patients were treated with 18 mg
dexamethasone daily (6 mg three times daily) for three days, plus
100 mg dimenhydrinate daily for three days. The placebo group
were treated with placebo plus 100 mg dimenhydrinate daily for
three days.
In Shupak 2008, patients were treated with prednisolone 1 mg/
kg daily for five days, followed by reducing regimen for the next
15 days. Placebo consisted of 200 mg lactose for 20 consecutive
days. Famotidine (histamine H2-receptor antagonist) 20 mg daily
was provided with prednisolone in the study group to protect the
gastric mucosa, but not the placebo group. Both groups received
vestibular sedatives for the first five days after presentation (25 mg
promethazine three times daily for severe symptoms, or cinnarizine
up to three times daily for milder symptoms).
In the study by Strupp 2004, patients were treated with 100 mg
methylprednisolone (or the matching placebo, lactose) as a single
morning dose, daily for three days, then tapered off to 10 mg over
the next 19 days. Patients also received 150 mg pirenzepine (a
muscarinic M1-receptor antagonist) once a day to protect the gastric mucosa. If necessary, patients also received antiemetic agents
(50 mg to 150 mg of dimenhydrinate a day) for a maximum of
three days.
In Ariyasu 1990 and Strupp 2004, treatment was initiated within
three days of the onset of symptoms. The exact timing of the
commencement of steroids in relation to the commencement of
symptoms was unclear in the studies by Rezaie 2006 and Shupak
2008.
Outcomes
In terms of symptom-based outcomes, both Ariyasu 1990 and
Rezaie 2006 reported subjective relief of vertigo at 24 hours as
an outcome measure. Shupak 2008 reported the proportion of
patients with symptoms and signs and a Dizziness Handicap Inventory score at one, three, six and 12 months. Rezaie 2006 additionally reported the mean time (in hours) for the relief of vertigo,
nausea and nystagmus. Strupp 2004 did not assess the duration
and severity of symptoms.
Ariyasu 1990 and Shupak 2008 both reported pathological findings on electronystagmography at one month as an outcome measure, with Shupak 2008 additionally reporting data at three, six
and 12 months.
Shupak 2008 and Strupp 2004 both reported an improvement
(as a %) in the lateralisation of the electronystagmographic caloric
test at 12 months, with Shupak 2008 additionally reporting data
at one, three and six months.
Complete caloric recovery (defined as lateralisation in caloric testing < 25%) was used as an outcome measure in Ariyasu 1990,
Shupak 2008 and Strupp 2004. Ariyasu 1990 reported data at
one month and Strupp 2004 reported data at 12 months, while
Shupak 2008 reported data at one, three, six and 12 months.
Only Strupp 2004 formally addressed severe and minor patientrelated adverse outcomes. Although other trials acknowledged that
corticosteroids have side effects (and Ariyasu 1990; Rezaie 2006
and Shupak 2008 excluded patients at high risk of such adverse
effects), they did not formally report adverse outcomes.
Excluded studies
Full details are provided in the Characteristics of excluded studies
table.
Of the 12 potentially eligible studies for which the full texts were
obtained, we excluded eight. One was an editorial (Anonymous
2004), one a summary of available evidence (Chou 2009), two
were review articles (Brantberg 2008; Walker 2009) and four were
found to be prospective non-randomised studies and were therefore excluded from the review (Kitahara 2001; Kitahara 2003;
Ohbayashi 1993; Yamashita 1989). In addition to these, we identified one recent meta-analysis (published following acceptance of
our title by the Cochrane Ear, Nose and Throat Disorders Review
Group) on the effectiveness of corticosteroids as a treatment for
vestibular neuritis (Goudakos 2010).
Risk of bias in included studies
12
Risk of bias for the included studies is reported in the individual
’Risk of bias’ tables (see Characteristics of included studies) and
is summarised in a ’Risk of bias’ summary (Figure 2) and graph
(Figure 3).
Figure 2. Risk of bias summary: review authors’ judgements about each risk of bias item for each included
study.
13
Figure 3. Risk of bias graph: review authors’ judgements about each risk of bias item presented as
percentages across all included studies.
We considered that the risk of bias in Ariyasu 1990, Shupak 2008
and Strupp 2004 is moderate (Grade B for overall methodological
quality).
The trial by Ariyasu 1990 was double-blinded (“methylprednisolone and placebo tablets were identical in appearance, packaging and labelling except for code numbering”), although patients
were poorly randomised (“methylprednisolone and placebo were
paired in bottles labelled ’A’ and ’B’ and randomised accordingly;
of every ten patients, five would start treatment with placebo and
five with methylprednisolone”). It was unclear how it was decided
which five patients would receive ’drug A’ and which five would
receive ’drug B’.
The randomisation and blinding strategy employed in Shupak
2008 was unclear. Further information is required about the precise method of randomisation and blinding used, as these details
are not provided in the text.
Out of 141 patients recruited in Strupp 2004, 27 (19%) were lost
during follow up. No intention-to-treat analysis was conducted.
However, missing outcome data are balanced in numbers across
intervention groups, with similar reasons for missing data across
groups. Although reported as double-blinded, the blinding procedure in Strupp 2004 was unclear.
We considered that the risk of bias in Rezaie 2006 is high (Grade
C for overall methodological quality). Patients were poorly randomised (based on their hospitalisation date, therefore the trial
lacked adequate sequence generation and allocation concealment).
The attending staff and patients were not aware of the assignment
and therefore an attempt was made to double-blind the trial. However, further details regarding the blinding procedure are necessary
to permit full judgement regarding the blinding procedure (e.g.
were the steroid and placebo medications identical in appearance?
). In addition, only outpatient clinic (but not emergency department) patients were recruited in the trial, which might have influenced the results.
Effects of interventions
See: Summary of findings for the main comparison
Corticosteroids compared to placebo for idiopathic acute
vestibular dysfunction (vestibular neuritis)
(See also ’Summary of findings for the main comparison’).
Comparison: corticosteroids versus control (placebo)
All four trials that met the inclusion criteria compared corticosteroids with placebo (Ariyasu 1990; Rezaie 2006; Shupak 2008;
Strupp 2004).
Primary outcomes
Proportion of patients that recover
Three trials addressed complete caloric recovery (Ariyasu 1990;
Shupak 2008; Strupp 2004).
The proportion of patients with complete caloric recovery at one
month after the initiation of therapy was significantly different
between the corticosteroids group and placebo group according
to the results of two studies (RR fixed 2.81; 95% CI 1.32 to
6.00, P = 0.007) (Analysis 1.1) (Figure 4) (Ariyasu 1990; Shupak
2008). However, although the heterogeneity, as measured by the
I2 statistic, was 13% the results must be interpreted with caution
due to the clinical and methodological heterogeneity apparent
between the two studies, the small number of patients and the
14
unclear randomisation and blinding procedures detailed in these
two studies.
Figure 4. Forest plot of comparison: 1 Corticosteroids versus control/placebo, outcome: 1.7 Complete
caloric recovery at 1 month (defined as lateralisation in caloric testing < 25%).
Complete caloric recovery at 12 months was reported by two studies (Shupak 2008; Strupp 2004). When analysed using a fixedeffect model, there was a significant difference between the corticosteroids and placebo group. However, the heterogeneity as measured by the I2 statistic was 91%. When re-analysed using a random-effects model the pooled results did not show a significant
difference between the corticosteroids and placebo groups (RR
random 1.58; 95% CI 0.45 to 5.62, P = 0.48) (Analysis 1.2)
(Figure 5).
Figure 5. Forest plot of comparison: 1 Corticosteroids versus control/placebo, outcome: 1.8 Complete
caloric recovery at 12 months (defined as lateralisation in caloric testing < 25%).
15
Degree of recovery
Three trials addressed symptomatic recovery (Ariyasu 1990; Rezaie
2006; Shupak 2008).
Ariyasu 1990 and Rezaie 2006 reported subjective relief of vertigo
at 24 hours, but there was no overall significant difference between
the corticosteroids and placebo groups (RR random 0.39; 95%
CI 0.04 to 3.57, P = 0.40) (Analysis 1.3) (Figure 6). Shupak 2008
reported the Dizziness Handicap Inventory score at one, three,
six and 12 months although there was no significant difference
between the two groups (Analysis 1.4).
Figure 6. Forest plot of comparison: 1 Corticosteroids versus control/placebo, outcome: 1.1 Vertigo at 24
hours.
Degree of caloric recovery was reported as pathological findings on electronystagmography (Ariyasu 1990 and Shupak 2008)
and improvement (as a %) in the lateralisation of caloric testing
(Shupak 2008 and Strupp 2004). There was no significant difference between the corticosteroids and placebo groups with respect to pathological findings on electronystagmography at both
one month (RR random 0.36; 95% CI 0.02 to 7.85, P = 0.52)
(Analysis 1.5) and at 12 months (RR fixed 1.50; 95% CI 0.29 to
7.73, P = 0.63) (Analysis 1.6). There was also no significant difference with respect to improvement (as a %) in the lateralisation
of caloric testing at both one month (MD fixed 9.60%; 95% CI
-20.66 to 39.86, P = 0.53) (Analysis 1.7) and at 12 months (MD
random 6.83%; 95% CI -27.69 to 41.36, P = 0.70) (Analysis 1.8).
Secondary outcomes
Time to recovery
Only one trial addressed the mean time to recovery in hours, with
respect to vertigo, nausea and nystagmus (Rezaie 2006). Although
the results were significant (Analysis 1.9), the risk of bias is high
(Grade C for overall methodological quality) and the patient numbers small. Patients taking corticosteroids in Shupak 2008 were
reported to make a more rapid recovery compared with patients
taking placebo medication, although there was no difference in
long-term outcome.
Patient-reported adverse events
Although all four included trials acknowledged that corticosteroids
have side effects (and excluded patients at high risk of such adverse
effects), only Strupp 2004 formally reported severe and minor
patient-related adverse outcomes.
In the Strupp 2004 trial, one patient (1/141 or 0.71%) suffered a
severe adverse effect in the steroid group (gastric ulcer with minor
bleeding, 10 days following the commencement of steroids, which
necessitated local injection with adrenaline). This was despite the
administration of pirenzepine (a muscarinic M1-receptor antagonist which reduces gastric acid secretion and thereby helps to
protect the stomach lining). The methylprednisolone was stopped
and the patient excluded from the study.
Minor adverse effects were experienced by eight patients (8/141
or 5.67%). Three patients reported dyspepsia and five patients
reported mood swings, but all eight patients continued treatment.
16
DISCUSSION
Summary of main results
Four randomised controlled trials were identified for inclusion in
this review. We made efforts to identify all relevant studies and no
studies were excluded due to language or year of publication.
Overall, there is insufficient evidence from these trials to support the administration of corticosteroids to patients with idiopathic acute vestibular dysfunction. In addition, as demonstrated
by Strupp 2004, corticosteroids are not without risk of adverse
events. However, it is difficult to reach firm conclusions from these
data since they contain small numbers of patients (i.e. small sample sizes) and there is significant clinical, methodological and statistical heterogeneity between the studies. For example, the studies differ with respect to where the patients were recruited from,
the diagnostic criteria used to diagnose idiopathic acute vestibular
dysfunction, inclusion and exclusion criteria, treatment regimens
(choice of corticosteroid, dose of corticosteroid, duration of treatment, reducing regimen, timing of drug commencement in relation to the onset of symptoms, presence of additional treatments
in conjunction with the use of corticosteroids such as vestibular
sedatives and antiemetics), and the outcome measures used. In
addition, we considered the risk of bias to be moderate in three
studies (Grade B) and high in one study (Grade C). We found no
trials with a low risk of methodological bias that used the highest level of diagnostic criteria and outcome measures (i.e. overall
quality Grade A).
Given the above heterogeneity between the trials, it might be argued why attempt to pool together the results at all? Some of the
trials agree in their direction of effect and an argument can be
made for pooling since this will improve precision. Both pooled
and unpooled results are presented here; in some circumstances we
acknowledge that it is inappropriate, but in other circumstances
it might be beneficial.
health-related quality of life outcome measures may provide more
valuable information. Previous studies have confirmed the lack
of correlation between recovery of clinical symptoms and caloric
recovery, regardless of the treatment strategy (corticosteroids or
placebo) (Bergenius 1999; Ohbayashi 1993).
Potential biases in the review process
The authors reviewed the risk of bias of the trials independently
and in accordance with the current recommended approach for
assessing the risk of bias in Cochrane Reviews (Handbook 2011).
There was therefore a low risk of bias in the review process, and
the authors have no conflicts of interest.
Agreements and disagreements with other
studies or reviews
There has been one recent systematic review and meta-analysis
published following acceptance of our title by the Cochrane Ear,
Nose and Throat Disorders Review Group, on the effectiveness
of corticosteroids as a treatment for vestibular neuritis (Goudakos
2010). They concluded that corticosteroids improve the caloric
extent and recovery of canal paresis, but made no difference to
the recovery of clinical symptoms in patients with vestibular neuritis. One reason for the differences seen between their study and
ours comes from the way the meta-analyses were conducted. In
their meta-analyses of complete caloric recovery at 12 months
and caloric improvement at 12 months a fixed-effect model was
utilised, despite the presence of significant heterogeneity among
the data (I2 statistics of 79% and 95%, respectively). When the
data were re-analysed using a random-effects model, the difference
between corticosteroids and placebo became insignificant in both
cases.
Quality of the evidence
AUTHORS’ CONCLUSIONS
The fact that only four randomised controlled trials were identified, and all had methodological flaws particularly with respect
to selection and performance biases, means that it is not possible
to reach firm conclusions regarding the use of corticosteroids in
idiopathic acute vestibular dysfunction. In addition, revealing the
statistical significance of treatment benefit in studies of idiopathic
acute vestibular dysfunction requires large numbers of patients
due to the variable spontaneous recovery profile of patients.
In the four trials identified, there was a disproportionate emphasis placed on caloric irrigation and electronystagmography in the
assessment of vestibular recovery with corticosteroids, or placebo
medication, which cannot be directly translated into clinical terms.
The primary goal of medical practice remains the improvement of
patients’ clinical symptoms and in this regard symptom-based and
Implications for practice
Based on the currently available data, there is insufficient evidence
to support the use of corticosteroids in the management of patients
with idiopathic acute vestibular dysfunction.
Implications for research
Large-scale, adequately powered and well-designed randomised
controlled trials are needed to clarify the effectiveness of corticosteroids in the treatment of idiopathic acute vestibular dysfunction. Future studies should include health-related quality of life
and symptom-based outcome measures, in addition to objective
measures of vestibular improvement, such as caloric irrigation and
electronystagmography.
17
ACKNOWLEDGEMENTS
We are grateful to Jenny Bellorini in the Cochrane Ear, Nose and
Throat Disorders Group for her assistance and to Gemma Sandberg for the design and running of the search strategy.
REFERENCES
References to studies included in this review
Ariyasu 1990 {published data only}
Ariyasu L, Byl FM, Sprague MS, Adour KK. The beneficial
effect of methylprednisolone in acute vestibular vertigo.
Archives of Otolaryngology - Head and Neck Surgery 1990;
116(6):700–3.
Rezaie 2006 {published data only}
Rezaie AA, Hashemian F, Rezaie N. Corticosteroids effect
on vestibular neuritis symptom relief. Pakistan Journal of
Medical Sciences 2006;22(4):409–11.
Shupak 2008 {published data only}
Shupak A, Issa A, Golz A, Kaminer M, Braverman I.
Prednisone treatment for vestibular neuritis. Otology and
Neurotology 2008;29(3):368–74.
Ohbayashi 1993 {published data only}
Ohbayashi S, Oda M, Yamamoto M, Urano M, Harada K,
Horikoshi H, et al.Recovery of the vestibular function after
vestibular neuronitis. Acta Oto-Laryngologica. Supplement
1993;503:31–4.
Walker 2009 {published data only}
Walker MF. Treatment of vestibular neuritis. Current
Treatment Options in Neurology 2009;11(1):41–5.
Yamashita 1989 {published data only}
Yamashita H, Sekitani T, Okami K, Endo S. Evaluation
of steroid therapy for vestibular neuronitis. Bulletin of the
Yamaguchi Medical School 1989;36(1-2):37–41.
References to ongoing studies
Strupp 2004 {published data only}
Strupp M, Zingler VC, Arbusow V, Niklas D, Maag KP,
Dieterich M, et al.Methylprednisolone, valacyclovir, or the
combination for vestibular neuritis. New England Journal of
Medicine 2004;351(4):354–61.
NCT01231009 {unpublished data only}
The role of corticosteroids and vestibular exercises in
recovery of vestibular neuritis: a prospective randomized
controlled clinical trial. http://clinicaltrials.gov/ct2/show/
NCT01231009 (accessed 5 April 2011).
References to studies excluded from this review
Additional references
Anonymous 2004 {published data only}
Anonymous. Steroid is effective for vestibular neuritis,
valacyclovir is not. Journal of Family Practice 2004;53(11):
864–7.
Arbusow 1999
Arbusow V, Schulz P, Strupp M, Dieterich M, von
Reinhardstoettner A, Rauch E, et al.Distribution of herpes
simplex virus type 1 in human geniculate and vestibular
ganglia: implications for vestibular neuritis. Annals of
Neurology 1999;46(3):416–9.
Brantberg 2008 {published data only}
Brantberg K, Goplen F, Bråthen G, Nordahl SH,
Arnesen H. Should vestibular neuritis be treated with
corticosteroids? [article in Norwegian]. Tidsskrift for Den
Norske Laegeforening 2008;128(18):2062–3.
Chou 2009 {published data only}
Chou HC, Yen ZS. Towards evidence based emergency
medicine: best BETs from the Manchester Royal Infirmary.
BET 1: steroids for patients with vestibular neuronitis.
Emergency Medicine Journal 2009;26(11):813–5.
Kitahara 2001 {published data only}
Kitahara T, Okumura S, Takeda N, Nishiike S, Uno A,
Fukushima M, et al.Effects of steroid therapy on long-term
canal prognosis and activity in the daily life of vestibular
neuronitis patients [article in Japanese]. Nippon Jibiinkoka
Gakkai Kaiho 2001;104(11):1059–64.
Kitahara 2003 {published data only}
Kitahara T, Kondoh K, Morihana T, Okumura S, Horii A,
Takeda N, et al.Steroid effects on vestibular compensation
in human. Neurological Research 2003;25(3):287–91.
Arbusow 2000
Arbusow V, Strupp M, Wasicky R, Horn AK, Schulz P,
Brandt T. Detection of herpes simplex virus type 1 in
human vestibular nuclei. Neurology 2000;55(6):880–2.
Baloh 1996
Baloh RW, Ishyama A, Wackym PA, Honrubia V. Vestibular
neuritis: clinical-pathologic correlation. Otolaryngology Head and Neck Surgery 1996;114(4):586–92.
Baloh 2003
Baloh RW. Clinical practice: vestibular neuritis. New
England Journal of Medicine 2003;348(11):1027–32.
Bartual-Pastor 2005
Bartual-Pastor J. Vestibular neuritis: etiopathogenesis.
Revue de Laryngologie Otologie Rhinologie 2005;126(4):
279–81.
Bergenius 1999
Bergenius J, Perols O. Vestibular neuritis: a follow-up study.
Acta Oto-Laryngologica 1999;119(8):895–9.
18
Bronstein 2007
Bronstein A, Lempert T. Dizziness - A Practical Approach
to Diagnosis and Management. Cambridge: Cambridge
University Press, 2007:67.
Cameron 1999
Cameron SA, Dutia MB. Lesion-induced plasticity in rat
vestibular nucleus neurones dependent on glucocorticoid
receptor activation. Journal of Physiology 1999;518(Pt 1):
151–8.
Coats 1969
Coats A. Vestibular neuronitis. Transactions of the American
Academy of Ophthalmology and Otolaryngology 1969;73:
395–408.
Davis 1993
Davis LE. Viruses and vestibular neuritis: review of human
and animal studies. Acta Oto-Laryngologica. Supplement
1993;503:70–3.
Furuta 1993
Furuta Y, Takasu T, Fukuda S, Inuyama Y, Sato KC,
Nagashima K. Latent herpes simplex virus type 1 in human
vestibular ganglia. Acta Oto-Laryngologica. Supplement
1993;503:85–9.
Goudakos 2010
Goudakos JK, Markou KD, Franco-Vidal V, Vital V,
Tsaligopoulos M, Darrouzet V. Corticosteroids in the
treatment of vestibular neuritis: a systematic review and
meta-analysis. Otology & Neurotology 2010;31(2):183–9.
GRADEpro 2008
Brozek J, Oxman A, Schünemann H. GRADEpro
version 3.2 for Windows. Grading of Recommendations
Assessment, Development and Evaluation (GRADE)
Working Group, 2008.
Handbook 2011
Higgins JPT, Green S (editors). Cochrane Handbook
for Systematic Reviews of Interventions Version 5.1.0
[updated March 2011]. The Cochrane Collaboration,
2011. Available from www.cochrane-handbook.org.
Martin 1998
Martin EA (Ed). Concise Medical Dictionary. Fifth. Oxford:
Oxford University Press, 1998.
Nadol 1995
Nadol JB Jr. Vestibular neuritis. Otolaryngology - Head and
Neck Surgery 1995;112(1):162–72.
Neuhauser 2007
Neuhauer HK. Epidemiology of vertigo. Current Opinion
in Neurology 2007;20(1):40–6.
Parens 1999
Parens LS, Sun AH, Freeman DJ. Corticosteroid
pharmacokinetics in the inner ear fluids: an animal study
followed by clinical application. Laryngoscope 1999;109(7
Pt 2):1–17.
RevMan 2011
The Nordic Cochrane Centre, The Cochrane Collaboration.
Review Manager (RevMan). 5.1. Copenhagen: The Nordic
Cochrane Centre, The Cochrane Collaboration, 2011.
Schuknecht 1981
Schuknecht HF, Kitamura K. Vestibular neuritis. Annals of
Otology, Rhinology, and Laryngology. Supplement 1981;90:
1–19.
Schulz 1998
Schulz P, Arbusow V, Strupp M, Dieterich M, Rauch E,
Brandt T. Highly variable distribution of HSV-1-specific
DNA in human geniculate, vestibular and spiral ganglia.
Neuroscience Letters 1998;252(2):139–42.
Scott-Brown 1997
Gacek RR. Chapter 5: Pathology of the vestibular system.
In: Kerr AG editor(s). Scott-Brown’s Otolaryngology. Sixth.
Vol. 3 (Otology), Butterworth Heinemann, 1997:23–24.
Sekitani 1993
Sekitani T, Imate Y, Noguchi T, Inokuma T. Vestibular
neuronitis: epidemiological survey by questionnaire in
Japan. Acta Oto-Laryngologica. Supplement 1993;503:9–12.
Shahle 1966
Stahle J. Vestibular neuritis. In: Wolfson R editor(s). The
vestibular system and its diseases. Philadelphia: University of
Pennsylvania Press, 1966:459–70.
Jerram 1995
Jerram AH, Darlington CL, Smith PF. Methylprednisolone
reduces spontaneous nystagmus following unilateral
labyrinthectomy in guinea pig. European Journal of
Pharmacology 1995;275(3):291–3.
Yamanaka 1995a
Yamanaka T, Sasa M, Amano T, Miyahara H, Matsunaga
T. Role of glucocorticoid in vestibular compensation in
relation to activation of vestibular nucleus neurons. Acta
Oto-Laryngologica. Supplement 1995;519:168–72.
Liberati 2009
Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche
PC, Ioannidis JP, et al.The PRISMA statement for reporting
systematic reviews and meta-analyses of studies that evaluate
health care interventions: explanation and elaboration.
PLoS Medicine 2009;6(7):e1000100.
Yamanaka 1995b
Yamanaka T, Amano T, Sasa M, Matsunaga T. Prednisolone
excitation of medial vestibular nucleus neurons in cats.
European Archives of Otorhinolaryngology 1995;252(2):
112–8.
∗
Indicates the major publication for the study
19
CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]
Ariyasu 1990
Methods
Prospective, double-blinded, randomised controlled trial and cross-over study
Participants
20 patients (18 to 65 years of age), divided into study group and control group
Interventions
32 mg methylprednisolone was given orally on the first day and then decreased to 4 mg
gradually over the next 7 days
Placebo (lactose) contained no active ingredient except a trace of quinine sulphate to
impart a bitter taste identical to that of the methylprednisolone
Outcomes
Subjective relief of symptoms within first 24 hours
One month electronystagmography
Notes
Treatment initiated within 3 days of onset of symptoms
No dropouts
The study was graded B for overall methodological quality according to the stated criteria
Risk of bias
Bias
Authors’ judgement
Support for judgement
Random sequence generation (selection Unclear risk
bias)
Methylprednisolone and placebo were
paired in bottles labelled ’A’ and ’B’ and
randomly assigned to patients. Further details regarding sequence generation necessary
Allocation concealment (selection bias)
Drug containers of identical appearance
Low risk
Blinding of participants and personnel Low risk
(performance bias)
All outcomes
Double-blinded; drug containers of identical appearance
Blinding of outcome assessment (detection Low risk
bias)
All outcomes
Assessors blinded for subjective and objective outcomes
Incomplete outcome data (attrition bias)
All outcomes
Low risk
No missing outcome data
Selective reporting (reporting bias)
Unclear risk
Insufficient information to permit judgement
20
Ariyasu 1990
(Continued)
Other bias
Low risk
The study appears to be free of other
sources of bias
Rezaie 2006
Methods
Prospective randomised controlled trial
Participants
40 patients (15 to 55 years of age) from outpatient clinics, divided into placebo group
and study group
Interventions
Study group were treated with 18 mg dexamethasone daily (6 mg 3 times daily) for 3
days plus 100 mg dimenhydrinate daily for 3 days. Placebo group were treated with
placebo plus 100 mg dimenhydrinate daily for 3 days
Outcomes
Patient symptoms and calorics recorded at 24, 48 and 72 hours following initiation of
drug therapy
Mean time for relief of vertigo (hours)
Mean time for relief of nausea (hours)
Mean time for relief of nystagmus (hours)
Notes
Timing of intervention in relation to commencement of symptoms unclear
The study was graded C for overall methodological quality according to the stated criteria
Risk of bias
Bias
Random sequence generation (selection High risk
bias)
Sequence generated by algorithm based on
their hospitalisation date
Patients randomly assigned to intervention
based on their hospitalisation date
High risk
Blinding of participants and personnel Unclear risk
(performance bias)
All outcomes
Reported as double-blinded. Attending staff
and patient were not aware of assignment.
However, further details regarding blinding
procedure necessary
Blinding of outcome assessment (detection Unclear risk
bias)
All outcomes
Reported as double-blinded. Investigators
were not aware of assignment. However, further details regarding blinding procedure necessary
All outcomes
Low risk
Unclear risk
21
Rezaie 2006
(Continued)
Other bias
Unclear risk
Only outpatient clinic patients were recruited
Shupak 2008
Methods
Participants
30 patients (22 to 72 years of age) from emergency departments, or outpatient clinics,
divided into study group and control group
Interventions
Prednisolone 1 mg/kg was given daily for 5 days, followed by reducing regimen for the
next 15 days. Placebo consisted of 200 mg lactose for 20 consecutive days. Famotidine
(histamine H2-receptor antagonist) 20 mg daily was provided with prednisolone in the
study group to protect the gastric mucosa, but not the placebo group. Both groups
received vestibular sedatives for the first 5 days after presentation (25 mg promethazine 3
times daily for severe symptoms, or cinnarizine up to 3 times daily for milder symptoms)
Outcomes
Presence of symptoms and signs
Dizziness Handicap Inventory score
Caloric lateralisation on electronystagmography
Pathological finding on electronystagmography
Complete resolution
Findings recorded at 1, 3, 6 and 12 months
Notes
Unclear randomisation and blinding procedure
Timing of intervention in relation to commencement of symptoms unclear
No dropouts
Risk of bias
Bias
Random sequence generation (selection Unclear risk
bias)
Insufficient information about the sequence
generation to permit judgement
Unclear risk
(performance bias)
All outcomes
bias)
All outcomes
All outcomes
Low risk
22
Shupak 2008
(Continued)
Unclear risk
Other bias
Low risk
The study appears to be free of other sources
of bias
Strupp 2004
Methods
Participants
141 patients (18 to 80 years of age), recruited from 2 hospital emergency departments,
divided into placebo group (38 patients), methylprednisolone group (35 patients), valacyclovir group (33 patients) and methylprednisolone plus valacyclovir group (35 patients). Six patients in the steroid group and 8 in the placebo group were excluded from
the study, leaving 29 patients in the steroid group and 30 patients in the placebo group
Interventions
100 mg methylprednisolone (or the matching placebo lactose) as a single morning dose
was given daily for 3 days, then tapered off to 10 mg over the next 19 days. Patients
also received 150 mg pirenzepine (a muscarinic M1-receptor antagonist) once a day to
protect the gastric mucosa. If necessary, patients also received antiemetic agents (50 mg
to 150 mg of dimenhydrinate a day) for a maximum of 3 days
Outcomes
Vestibular function as determined by Jongkee’s formula within 3 days after the onset of
symptoms and 12 months later
Notes
Treatment initiated within 3 days of onset of symptoms
Eight patients in the placebo group (2 patients discontinued medication; 2 patients noncompliant with medication; 4 patients lost to follow up) and 6 in the steroid group (2
patients discontinued medication; 3 patients lost to follow up; 1 patient suffered a severe
adverse side effect) were excluded from the study
One patient suffered a severe adverse effect in the steroid group (gastric ulcer with minor bleeding), despite the administration of pirenzepine and methylprednisolone was
stopped. Minor adverse effects were experienced by 8 patients (3 patients reported dyspepsia; 5 patients reported mood swings), but all these patients continued treatment
Risk of bias
Bias
Random sequence generation (selection Low risk
bias)
Patients were randomly assigned by means of
computer-generated block randomisation
Computer-generated block randomisation
Low risk
(performance bias)
All outcomes
Reported as double-blinded, but exact details
regarding the blinding procedure employed
unclear
23
Strupp 2004
(Continued)
bias)
All outcomes
Reported as double-blinded, but exact details
regarding the blinding procedure employed
unclear
All outcomes
Low risk
No use of intention-to-treat analysis. 27/141
patients were lost during follow up. However,
missing outcome data balanced in numbers
across intervention groups, with similar reasons for missing data across groups
Unclear risk
Other bias
Low risk
The study appears to be free of other sources
of bias
Characteristics of excluded studies [ordered by study ID]
Study
Reason for exclusion
Anonymous 2004
ALLOCATION:
Editorial; not a randomised controlled trial
Brantberg 2008
ALLOCATION:
Review article; not a randomised controlled trial
Chou 2009
ALLOCATION:
Summary of available evidence; not a randomised controlled trial
Kitahara 2001
ALLOCATION:
Not a randomised controlled trial
Kitahara 2003
ALLOCATION:
Authors acknowledge in article that study is not a randomised controlled trial
Ohbayashi 1993
ALLOCATION:
No mention of randomisation and therefore not strictly a randomised controlled trial
Walker 2009
ALLOCATION:
Review article; not a randomised controlled trial
Yamashita 1989
ALLOCATION:
Not a randomised controlled trial
24
Characteristics of ongoing studies [ordered by study ID]
NCT01231009
Trial name or title
The role of corticosteroids and vestibular exercises in recovery of vestibular neuritis (ClinicalTrials.gov identifier: NCT01231009)
Methods
Prospective, double-blinded, randomised controlled trial (ongoing)
Participants
Inclusion criteria:
1. patients (18 to 80 years old) with a history of acute or subacute rotatory vertigo with postural
imbalance and nausea;
2. horizontal rotatory spontaneous nystagmus.
Exclusion criteria:
1. history of vestibular dysfunction before the acute onset of symptoms;
2. simultaneous hearing loss;
3. neurological disorder;
4. pregnancy or lactation;
5. history of psychiatric disorders;
6. history of glaucoma;
7. diabetes;
8. hypertension;
9. contraindication in receiving corticosteroids.
Interventions
Corticosteroids (active comparator): intravenous dexamethasone 8 mg 3 times per day tapering down for 7
days, followed by oral dexamethasone 2 mg per day tapering down over 7 days
Vestibular exercises (placebo comparator): 15 days of vestibular exercises to enhance the vestibulo-ocular reflex
under an expert physiotherapist
Outcomes
Primary outcome measures
1. Clinical recovery of vestibular function by clinical examination (nystagmus, Romberg’s test, headthrust test, head-shaking test) and questionnaires (Greek edition of Dizziness Handicap Inventory and
European Evaluation of Vertigo Scale)
2. Laboratory recovery of vestibular function (caloric test and vestibular evoked myogenic potentials)
Starting date
January 2010
Contact information
John Goudakos, Aristotle University Of Thessaloniki, Greece
Notes
Estimated study completion date: January 2013
25
DATA AND ANALYSES
Comparison 1. Corticosteroids versus control/placebo
Outcome or subgroup title
1 Complete caloric recovery at 1
month (defined as lateralisation
in caloric testing < 25%)
2 Complete caloric recovery
at 12 months (defined as
lateralisation in caloric testing
< 25%)
3 Vertigo at 24 hours
4 Dizziness Handicap Inventory
score
4.1 1 month
4.2 3 months
4.3 6 months
4.4 12 months
5 Pathological findings on
electronystagmography at 1
month
6 Pathological findings on
electronystagmography at 12
months
7 Improvement (as %) in
lateralisation of ENG caloric
test at 1 month
8 Improvement (as %) in
lateralisation of ENG caloric
test at 12 months
9 Mean time to recovery
9.1 Vertigo (mean time to
recovery in hours)
9.2 Nausea (mean time to
recovery in hours)
9.3 Nystagmus (mean time to
recovery in hours)
No. of
studies
No. of
participants
2
50
Risk Ratio (M-H, Fixed, 95% CI)
2.81 [1.32, 6.00]
2
89
Risk Ratio (M-H, Random, 95% CI)
1.58 [0.45, 5.62]
2
1
60
Mean Difference (IV, Fixed, 95% CI)
0.39 [0.04, 3.57]
Totals not selected
0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0.36 [0.02, 7.85]
Statistical method
Effect size
1
1
1
1
2
50
1
30
Risk Ratio (M-H, Fixed, 95% CI)
1.5 [0.29, 7.73]
1
30
9.60 [-20.66, 39.86]
2
89
Mean Difference (IV, Random, 95% CI)
6.83 [-27.69, 41.36]
1
1
Totals not selected
0.0 [0.0, 0.0]
1
0.0 [0.0, 0.0]
1
0.0 [0.0, 0.0]
26
Analysis 1.1. Comparison 1 Corticosteroids versus control/placebo, Outcome 1 Complete caloric recovery
at 1 month (defined as lateralisation in caloric testing < 25%).
Review:
Comparison: 1 Corticosteroids versus control/placebo
Outcome: 1 Complete caloric recovery at 1 month (defined as lateralisation in caloric testing < 25%)
Study or subgroup
Steroid group
Placebo
n/N
n/N
Risk Ratio
Weight
Ariyasu 1990
16/16
2/4
65.9 %
1.94 [ 0.80, 4.68 ]
Shupak 2008
9/15
2/15
34.1 %
4.50 [ 1.16, 17.44 ]
Total (95% CI)
31
19
100.0 %
2.81 [ 1.32, 6.00 ]
M-H,Fixed,95% CI
Risk Ratio
M-H,Fixed,95% CI
Total events: 25 (Steroid group), 4 (Placebo)
Heterogeneity: Chi2 = 1.14, df = 1 (P = 0.28); I2 =13%
Test for overall effect: Z = 2.68 (P = 0.0074)
Test for subgroup differences: Not applicable
0.01
0.1
1
Favours placebo
10
100
Favours steroids
Analysis 1.2. Comparison 1 Corticosteroids versus control/placebo, Outcome 2 Complete caloric recovery
at 12 months (defined as lateralisation in caloric testing < 25%).
Review:
Outcome: 2 Complete caloric recovery at 12 months (defined as lateralisation in caloric testing < 25%)
Study or subgroup
Steroid group
Placebo
n/N
n/N
Shupak 2008
11/15
12/15
51.8 %
0.92 [ 0.62, 1.36 ]
Strupp 2004
22/29
8/30
48.2 %
2.84 [ 1.52, 5.33 ]
44
45
100.0 %
1.58 [ 0.45, 5.62 ]
Total (95% CI)
Risk Ratio
MH,Random,95%
CI
Weight
Risk Ratio
MH,Random,95%
CI
Heterogeneity: Tau2 = 0.77; Chi2 = 11.67, df = 1 (P = 0.00063); I2 =91%
0.01
0.1
Favours placebo
1
10
100
Favours steroids
27
Analysis 1.3. Comparison 1 Corticosteroids versus control/placebo, Outcome 3 Vertigo at 24 hours.
Review:
Outcome: 3 Vertigo at 24 hours
Study or subgroup
Steroid group
Placebo
Risk Ratio
MH,Random,95%
CI
Weight
Risk Ratio
MH,Random,95%
CI
n/N
n/N
Ariyasu 1990
1/10
7/10
41.2 %
0.14 [ 0.02, 0.96 ]
Rezaie 2006
15/20
19/20
58.8 %
0.79 [ 0.60, 1.04 ]
30
30
100.0 %
0.39 [ 0.04, 3.57 ]
Total (95% CI)
0.01
0.1
Favours steroids
1
10
100
Favours placebo
28
Analysis 1.4. Comparison 1 Corticosteroids versus control/placebo, Outcome 4 Dizziness Handicap
Inventory score.
Review:
Outcome: 4 Dizziness Handicap Inventory score
Study or subgroup
Steroid group
Mean
Difference
Placebo
Mean
Difference
N
Mean(SD)
N
Mean(SD)
IV,Fixed,95% CI
IV,Fixed,95% CI
15
20.9 (21.7)
15
15.8 (12.3)
5.10 [ -7.52, 17.72 ]
15
14.8 (19)
15
6.5 (6.8)
8.30 [ -1.91, 18.51 ]
15
17.8 (14.4)
15
8.3 (7.8)
9.50 [ 1.21, 17.79 ]
15
11.1 (14.2)
15
8.1 (12.6)
3.00 [ -6.61, 12.61 ]
1 1 month
Shupak 2008
2 3 months
Shupak 2008
3 6 months
Shupak 2008
4 12 months
Shupak 2008
-100
-50
0
Favours steroids
50
100
Favours placebo
29
electronystagmography at 1 month.
Review:
Outcome: 5 Pathological findings on electronystagmography at 1 month
Study or subgroup
Steroid group
Placebo
Risk Ratio
MH,Random,95%
CI
Weight
Risk Ratio
MH,Random,95%
CI
n/N
n/N
Ariyasu 1990
0/16
2/4
41.6 %
0.06 [ 0.00, 1.04 ]
Shupak 2008
4/15
3/15
58.4 %
1.33 [ 0.36, 4.97 ]
Total (95% CI)
31
19
100.0 %
0.36 [ 0.02, 7.85 ]
0.001 0.01 0.1
1
Favours steroids
10 100 1000
Favours placebo
electronystagmography at 12 months.
Review:
Outcome: 6 Pathological findings on electronystagmography at 12 months
Study or subgroup
Steroid group
Placebo
n/N
n/N
Risk Ratio
Weight
Shupak 2008
3/15
2/15
100.0 %
1.50 [ 0.29, 7.73 ]
Total (95% CI)
15
15
100.0 %
1.50 [ 0.29, 7.73 ]
M-H,Fixed,95% CI
Risk Ratio
M-H,Fixed,95% CI
Heterogeneity: not applicable
0.01
0.1
Favours steroids
1
10
100
Favours placebo
30
Analysis 1.7. Comparison 1 Corticosteroids versus control/placebo, Outcome 7 Improvement (as %) in
lateralisation of ENG caloric test at 1 month.
Review:
Outcome: 7 Improvement (as %) in lateralisation of ENG caloric test at 1 month
Study or subgroup
Steroid group
Mean
Difference
Placebo
N
Mean(SD)
N
Mean(SD)
Shupak 2008
15
23.4 (39.6)
15
13.8 (44.8)
Total (95% CI)
15
Weight
IV,Fixed,95% CI
Mean
Difference
IV,Fixed,95% CI
15
100.0 %
9.60 [ -20.66, 39.86 ]
100.0 %
9.60 [ -20.66, 39.86 ]
Heterogeneity: not applicable
-100
-50
Favours placebo
0
50
100
Favours steroids
31
Analysis 1.8. Comparison 1 Corticosteroids versus control/placebo, Outcome 8 Improvement (as %) in
lateralisation of ENG caloric test at 12 months.
Review:
Outcome: 8 Improvement (as %) in lateralisation of ENG caloric test at 12 months
Study or subgroup
Steroid group
Mean
Difference
Placebo
Weight
Mean
Difference
N
Mean(SD)
N
Mean(SD)
Shupak 2008
15
33.4 (37.7)
15
46 (31)
45.1 %
-12.60 [ -37.30, 12.10 ]
Strupp 2004
29
62.4 (16.9)
30
39.6 (28.1)
54.9 %
22.80 [ 11.01, 34.59 ]
100.0 %
6.83 [ -27.69, 41.36 ]
Total (95% CI)
44
IV,Random,95% CI
IV,Random,95% CI
45
-100
-50
0
Favours placebo
50
100
Favours steroids
Analysis 1.9. Comparison 1 Corticosteroids versus control/placebo, Outcome 9 Mean time to recovery.
Review:
Outcome: 9 Mean time to recovery
Study or subgroup
Steroid group
N
Mean
Difference
Placebo
Mean
Difference
Mean(SD)
N
Mean(SD)
IV,Fixed,95% CI
IV,Fixed,95% CI
45.6 (15.3)
20
68.4 (11.7)
-22.80 [ -31.24, -14.36 ]
28.8 (12.5)
20
54 (20.4)
-25.20 [ -35.69, -14.71 ]
28.8 (9.8)
20
63.6 (16.1)
-34.80 [ -43.06, -26.54 ]
1 Vertigo (mean time to recovery in hours)
Rezaie 2006
20
2 Nausea (mean time to recovery in hours)
Rezaie 2006
20
3 Nystagmus (mean time to recovery in hours)
Rezaie 2006
20
-50
-25
Favours steroids
0
25
50
Favours placebo
32
APPENDICES
Appendix 1. Search strategies
CENTRAL
Cochrane Ear Nose and PubMed
Throat Group’s Trials Register (ProCite database)
EMBASE (Ovid)
#1 MeSH descriptor Vestibular
Neuronitis explode all trees
#2 MeSH descriptor Vestibular
Nerve explode all trees
#3 vestibul*
#4 #2 OR #3
#5 MeSH descriptor Neuritis
explode all trees
#6 MeSH descriptor Inflammation explode all trees
#7 neuri* OR neuroni* OR inflamm* OR dysfunction* OR
disorder* OR lesion*
#8 #5 OR #6 OR #7
#9 #4 AND #8
#10 acute NEXT vestibulopathy OR acute NEXT vertigo
OR labyrinthitis
#11 #1 OR #9 OR 10
#12 MeSH descriptor Adrenal
Cortex Hormones explode all
trees
#13 MeSH descriptor Steroids
explode all trees
#14 MeSH descriptor Glucocorticoids explode all trees
#15 MeSH descriptor Anti-inflammatory Agents explode all
trees
#16 MeSH descriptor Antiinflammatory Agents, NonSteroidal explode all trees
#17 #15 NOT #16
((vestibul* AND (neuri* OR
neuroni* OR inflamm* OR
dysfunction* OR disorder* OR
lesion*)) AND (corticosteroid*
OR glucocorticoid* OR corticoid* OR steroid* OR alclometasone OR Beclomethasone OR Betamethasone OR
Budesonide OR Clobetasol OR
clobetasone OR Desoximetasone OR Dexamethasone OR
diflorasone OR Diflucortolone
OR Flumethasone OR Fluocinolone OR Fluocortolone OR
Fluorometholone OR Fluprednisolone OR Flurandrenolone
OR Melengestrol OR Methylprednisolone OR Paramethasone OR Prednicarbate OR
Prednisolone OR Prednisone
OR Triamcinolone))
(((acute AND (vestibulopathy
OR vertigo)) OR labyrinthitis))
AND (corticosteroid* OR glucocorticoid* OR corticoid* OR
steroid* OR alclometasone OR
Beclomethasone OR Betamethasone OR
OR Flumethasone OR Fluoci-
1 vestibular neuronitis/
2 vestibular nerve/
3 vestibul*.tw.
4 exp neuritis/
5 exp inflammation/
6 (neuri* or neuroni* or inflamm* or dysfunction* or disorder* or lesion*).tw.
7 2 or 3
8 4 or 5 or 6
9 7 and 8
10 ((acute and (vestibulopathy
or vertigo)) or labyrinthitis).tw.
11 1 or 9 or 10
12 exp corticosteroid/
13 steroid/
14 exp antiinflammatory
agent/
15 exp nonsteroid antiinflammatory agent/
16 14 not 15
17 (corticosteroid* or glucocorticoid* or corticoid* or steroid*
or
alclometasone or Beclomethasone or
Betamethasone or Budesonide
or Clobetasol or clobetasone
or Desoximetasone or Dexamethasone or diflorasone or
Diflucortolone or Flumethasone or Fluocinolone or Fluocortolone or Fluorometholone
#1 “Vestibular Neuronitis”
[Mesh]
#2 (“Vestibular Nerve” [Mesh]
OR vestibul* [tiab]) AND
(“Neuritis” [Mesh] OR “Inflammation” [Mesh] OR neuri*
[tiab] OR neuroni* [tiab] OR
inflamm* [tiab] OR dysfunction* [tiab] OR disorder* [tiab]
OR lesion* [tiab])
#3 ((acute [tiab] AND (vestibulopathy [tiab] OR vertigo
[tiab])) OR labyrinthitis [tiab])
#4 #1 OR #2 OR #3
#5 “Adrenal Cortex Hormones”
[Mesh] OR “Steroids” [Mesh]
OR “Glucocorticoids” [Mesh]
#6 “Anti-inflammatory Agents”
[Mesh] NOT “Anti-inflammatory Agents, Non-Steroidal”
[Mesh]
#7 corticosteroid* [tiab] OR
glucocorticoid* [tiab] OR corticoid* [tiab] OR steroid* [tiab]
OR alclometasone [tiab] OR
Beclomethasone [tiab] OR Betamethasone [tiab] OR Budesonide [tiab] OR Clobetasol
[tiab] OR clobetasone [tiab]
OR Desoximetasone [tiab] OR
Dexamethasone [tiab] OR diflorasone [tiab] OR Diflucortolone [tiab] OR Flumethasone
33
(Continued)
#18 corticosteroid* OR glucocorticoid*
OR corticoid* OR steroid* OR
alclometasone OR Beclomethasone OR Betamethasone OR
OR Triamcinolone
#19 #12 OR #13 OR #14 OR
#17 OR #18
#20 #11 AND #19
nolone OR Fluocortolone OR
OR Triamcinolone)
[tiab] OR Fluocinolone [tiab]
OR Fluocortolone [tiab] OR
Fluorometholone [tiab] OR
Fluprednisolone [tiab] OR Flurandrenolone [tiab] OR Melengestrol [tiab] OR Methylprednisolone [tiab] OR Paramethasone [tiab] OR Prednicarbate [tiab] OR Prednisolone
[tiab] OR Prednisone [tiab] OR
Triamcinolone [tiab]
#8 #5 OR #6 OR #7
#9 #4 AND #8
or Fluprednisolone or Flurandrenolone or Melengestrol or
Methylprednisolone or Paramethasone or Prednicarbate or
Prednisolone or Prednisone or
Triamcinolone).tw.
18 12 or 13 or 16 or 17
19 11 and 18
Web of Science/BIOSIS Pre- CINAHL (EBSCO)
views (Web of Knowledge)
CAB Abstracts (Ovid)
ICTRP
#1 TS=(vestibul* AND (neuri*
OR neuroni* OR inflamm* OR
dysfunction* OR disorder* OR
lesion*))
#2
TS=((acute
AND (vestibulopathy OR vertigo)) OR labyrinthitis)
#3 #2 OR #1
#4
TS=(corticosteroid* OR glucocorticoid*
alclometasone OR Beclomethasone OR Betamethasone OR
1 vestibul*.tw.
2 exp neuritis/
3 exp inflammation/
4 (neuri* or neuroni* or inflamm* or dysfunction* or disorder* or lesion*).tw.
5 2 or 3 or 4
6 1 AND 5
7 ((acute and (vestibulopathy or
vertigo)) or labyrinthitis).tw.
8 6 or 7
9 exp corticosteroid/
10 (corticosteroid* or glucocorticoid* or corticoid* or steroid*
or
alclometasone or Beclomethasone or
Betamethasone or Budesonide
or Clobetasol or clobetasone
or Desoximetasone or Dexamethasone or diflorasone or
Diflucortolone or Flumethasone or Fluocinolone or Fluocortolone or Fluorometholone
or Fluprednisolone or Fluran-
vestibular AND neuritis OR
vestibular AND neuronitis OR
vestibular AND inflammation
OR vestibular AND dysfunction OR vestibular AND disorder OR vestibular AND lesion
S1 (MH “Vestibular Nerve”)
S2 TX vestibul*
S3 (MH “Neuritis+”)
S4 (MH “Inflammation+”)
S5 TX neuri* OR neuroni* OR
inflamm* OR dysfunction* OR
disorder* OR lesion*
S6 S1 or S2
S7 S3 or S4 or S5
S8 S6 and S7
S9 TX ( (acute AND (vestibulopathy OR vertigo)) ) or TX
labyrinthitis
S10 S8 or S9
S11 (MH “Adrenal Cortex
Hormones/TU”)
S12 (MH “Steroids”)
S13 (MH “Glucocorticoids”)
S14 (MH “Antiinflammatory
Agents+”)
S15 (MH “Antiinflammatory
Agents, Non-Steroidal+”)
S16 S14 NOT S15
S17
TX
corti-
34
(Continued)
Prednisolone OR Prednisone costeroid* OR glucocorticoid*
OR Triamcinolone)
alclometasone OR Beclometha#5 #4 AND #3
sone OR Betamethasone OR
OR Triamcinolone
S18 S11 or S12 or S13 or S16
or S17
S19 S10 and S18
drenolone or Melengestrol or
Methylprednisolone or Paramethasone or Prednicarbate or
Prednisolone or Prednisone or
Triamcinolone).tw.
11 9 or 10
12 8 AND 11
HISTORY
Protocol first published: Issue 7, 2010
Review first published: Issue 5, 2011
CONTRIBUTIONS OF AUTHORS
Jonathan M Fishman: drafting the text of the protocol and review; screening search results; organising retrieval of papers; screening
retrieved papers against inclusion criteria; contacting trial authors for additional information; data extraction; quality assessment; data
analysis and interpretation.
Chris Burgess: drafting the text of the protocol and review; screening search results; organising retrieval of papers; screening retrieved
papers against inclusion criteria; contacting trial authors for additional information; data extraction; quality assessment; data analysis
and interpretation.
Angus Waddell: appraisal of drafts at all stages of review; screening search results; interpretation of results; guarantor for the review.
35
DECLARATIONS OF INTEREST
None known.
SOURCES OF SUPPORT
Internal sources
• No sources of support provided, Not specified.
External sources
• No sources of support provided, Not specified.
DIFFERENCES BETWEEN PROTOCOL AND REVIEW
Changes have been made to reflect the latest upgrade to RevMan and the Cochrane Handbook for Systematic Reviews of Interventions
(Handbook 2011; RevMan 2011).
NOTES
BPPV = benign paroxysmal positional vertigo
CI = confidence interval
DHI = dizziness handicap inventory
ENG = electronystagmography
MD = mean difference
RCT = randomised controlled trial
RR = relative risk; risk ratio
SMD = standardised mean difference
INDEX TERMS
Medical Subject Headings (MeSH)
Acute Disease; Adrenal Cortex Hormones [∗ therapeutic use]; Caloric Tests; Dexamethasone [therapeutic use]; Methylprednisolone
[therapeutic use]; Prednisolone [therapeutic use]; Randomized Controlled Trials as Topic; Vertigo [etiology]; Vestibular Neuronitis
[complications; ∗ drug therapy]
36
MeSH check words
Humans
37

Corticosteroids for the treatment of idiopathic acute

Transcription

Similar documents

introduction to evaluation and management of patients with

COMPLETE VESTIBULECTOMY SURGERY

Corticosteroids in the Treatment of Vestibular Neuritis:

Vestibular Rehabilitation Therapy Melissa Nelson 

Otoneurology Lecture Plan (two lectures) Vestibular Physiology

Vestibular Function and Anatomy

What Is Computerized Dynamic Posturography (CDP)?