REVIEW ARTICLE
The Effectiveness of Vestibular Rehabilitation Interventions in Treating Unilateral Peripheral Vestibular Disorders: A Systematic Review Scott A. Arnold, Aaron M. Stewart, Heather M. Moor, Rita C. Karl & Jennifer C. Reneker* Division of Physical Therapy, Walsh University, North Canton, OH, USA
Abstract Background and Purpose. Various types of vestibular rehabilitation therapy are routinely used in clinical practice to treat unilateral peripheral vestibular hypofunction. The purpose of this systematic review was to compare the effectiveness of vestibular rehabilitation interventions (adaptation, substitution and habituation) in people with unilateral peripheral vestibular hypofunction, exclusionary of benign paroxysmal positional vertigo and Meniere’s disease. Methods. A search of the literature was conducted using PubMed, CINAHL and Scopus. Studies were eligible for inclusion if they were 1) a randomized controlled trial or randomized clinical trial; 2) written in English; 3) of participants with a unilateral, peripheral vestibular hypofunction; 4) of a conservative treatment approach only; and 5) with human subjects. Quality was assessed by two authors using the Physiotherapy Evidence Database scale. Effect size was calculated to determine the effect of treatment within each study group. Results. Seven papers were selected for inclusion. Physiotherapy Evidence Database scores ranged from 2/10 to 7/10. Interventions within the selected studies included combinations of adaptation, habituation, substitution or substitution by itself. Calculated effect sizes, or significance values, revealed that all interventions demonstrated effectiveness. Two studies reported improvements on the dynamic gait index, and a large difference was seen between intervention groups of the two studies. Discussion. Results suggest that vestibular therapy for unilateral peripheral vestibular hypofunction is effective. When considering all seven studies included in the review, it is difficult to determine the superiority of one intervention over another in treating unilateral peripheral vestibular hypofunction except when patient outcomes are captured by the dynamic gait index or dizziness handicap inventory. Many studies in this review demonstrate notable biases, suggesting that results should be used with caution. Future research should aim to use a common set of measures to capture outcomes. Copyright © 2015 John Wiley & Sons, Ltd. Received 2 April 2014; Revised 1 December 2014; Accepted 25 April 2015 Keywords dizziness/rehabilitation; exercise therapy; head movements; vestibular diseases/rehabilitation *Correspondence Jennifer Reneker, Division of Physical Therapy, Walsh University, 2020 East Maple St.,44720, North Canton, OH, USA. Email: [email protected]
Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/pri.1635
Introduction Dizziness is a common complaint reported in primary care practice and results in more than 6 million visits
Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
to physician offices per year in the United States alone (Brodovsky and Vnenchak, 2013). Dizziness symptoms were reported in 1 out of 5 individuals of working age (Yardley et al., 1998). It is reported that health-related
S. Arnold et al.
Effectiveness of Vestibular Rehabilitation: A Systematic Review
quality of life, both mental and physical, is significantly impaired in those with dizziness (Weidt et al., 2014). Although dizziness can be caused by a variety of medical conditions, it is estimated that 50% of cases are due to vestibular dysfunction (Hall and Cox, 2009). Many patients with a complaint of dizziness are referred to physical therapy for treatment to relieve the dizziness and symptoms associated with disequilibrium (Tee and Chee, 2005). Vestibular rehabilitation is described predominantly as a movement and exercise-based approach that includes adaptation, substitution (including postural control strategies) and habituation (Hillier and McDonnell, 2011). Unilateral peripheral vestibular hypofunction is a disorder that affects one side of the vestibular system and excludes vestibular dysfunction related to the brain (Hillier and McDonnell, 2011). Potential causes of unilateral peripheral vestibular hypofunction include vestibular neuritis, vestibular labyrinthitis, perilymphatic fistula and acoustic neuroma (Brodovsky and Vnenchak, 2013). In order to differentiate central from peripheral vestibular dysfunction, a thorough examination must be completed. The specific procedures included in vestibular rehabilitation should be aimed at the presenting symptoms to produce favourable outcomes (Deveze et al., 2014). Because of the high prevalence of vestibular dysfunction and resultant dizziness and because dizziness negatively impacts health-related quality of life, it is important that effective treatments of vestibular dysfunction are identified. Previous research on the effectiveness of vestibular rehabilitation for unilateral peripheral vestibular dysfunction has demonstrated significant differences in favour of vestibular rehabilitation when compared with control (placebo, sham, usual care or no intervention) (Hillier and McDonnell, 2011). To our knowledge, no major differences have been found in treatment effectiveness among interventions (Hillier and McDonnell, 2011). Identifying effective interventions may hasten recovery from dizziness related to vestibular dysfunction by giving clinicians insight into interventions to use during treatment. Benign paroxysmal positional vertigo (BPPV) has accepted treatment approaches (Hilton and Pinder, 2004; Hillier and McDonnell, 2011; Bruintjes et al., 2014), and Meniere’s Disease has several studied treatment approaches with varying efficacy (Pullens and van Benthem, 2011; Hillier and McDonnell, 2011; Phillips and Westerberg, 2011; Pullens et al., 2013); therefore,
these will not be the focus of the current study. The purpose of this systematic review was to compare the effectiveness of vestibular rehabilitation interventions (adaptation, substitution and habituation) in people with unilateral peripheral vestibular hypofunction, exclusionary of BPPV and Meniere’s Disease.
Methods Study design The present study utilized the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA, 2014), a prospective guide for writing systematic reviews and meta-analyses, in the research and reporting of data (PRISMA Statement). Search strategy Individualized, computer-based search strategies for PubMed (Appendix 1), Scopus and CINAHL were developed in November 2013. PubMed was searched using a comprehensive strategy that included search terms related to vestibular rehabilitation for unilateral peripheral vestibular disorders. There were no limits applied for the publication date of articles, but the following limits were applied to the search results: (1) Humans, (2) studies published in English, (3) randomized control trials and (4) controlled clinical trials. All remaining databases were searched using comparable strategies. A hand search was performed by exploring the references of included articles to identify studies not captured through the electronic searches. Eligibility criteria Studies were included in the review if the following inclusion criteria were met: 1) a randomized controlled trial or randomized clinical trial; 2) written in English; 3) of participants with a unilateral, peripheral vestibular hypofunction; 4) of a conservative treatment approach only; and 5) with human subjects. Studies omitted from the review included 1) BPPV, 2) Meniere’s Disease or 3) central vestibular disorders within the subject population, as this was beyond the scope of this review. Study selection Studies were selected for inclusion in this systematic review based on a stepwise process outlined in Figure 1. Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
S. Arnold et al.
Effectiveness of Vestibular Rehabilitation: A Systematic Review
Figure 1. Study flow for the systematic review. BPPV, benign paroxysmal positional vertigo; CNS, central nervous system; RCT, randomized controlled/clinical trial.
At each step, articles were independently screened by two members of the research team, based on the inclusion criteria, in an effort to reduce bias. Title and abstract were reviewed by two members (HM and RE), as well as full text (SA and AS). In cases of discrepancy between the two initial reviewers, a third reviewer determined the study’s inclusion or exclusion.
Data collection Data were collected from the articles accepted in this study by two members of the research team in tandem (SA and AS). Both members were required to agree upon the selected data for them to be included. Data were chosen based on their usefulness in answering the research question. The study sample characteristics, the intervention type and the outcome measures of the Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
selected studies were included in this review. Where data were not presented in the original research article, effort was made to retrieve unavailable and unclear data from the author. Vestibular rehabilitation strategies used in each study intervention were grouped according to the three most common vestibular rehabilitation intervention types including adaptation, habituation and substitution (Table 1). In this review, the following classification was used for the interventions in the included studies: adaptation exercises included exercises designed to improve vestibulo-ocular reflex gain. Typically, these interventions consist of fixing gaze on a target and keeping that target in focus while moving the head (Childs, 2010; Deveze et al., 2014). Habituation exercises included interventions with repetitious exposure of the subject to symptom-provoking
Effectiveness of Vestibular Rehabilitation: A Systematic Review
activities. This is performed until the vestibular system becomes accustomed and symptoms are no longer produced (Boyer et al., 2008; Childs, 2010). Substitution exercises included interventions to train patients to use their somatosensory and visual systems (Childs, 2010). Where available, p-values of within group changes from pre-test (reference) to post-test (endpoint) for the outcome measures were reported. Cohen’s D effect sizes from individual studies were calculated for within-group changes for the selected outcome measures from pre-test (reference) to post-test (endpoint) when such data were presented. Values of between-group comparisons were unable to be calculated because of limited or unclear data. Effect size is a measure of the magnitude of an intervention or the strength of an effect of a particular treatment (Cook, 2008). In order to understand the magnitude of an intervention, a classification system has been proposed, with values falling below 0.2 being trivial, at or above 0.2 but below 0.5 being small, at or above 0.50 but below 0.80 being medium, and at or above 0.80 being large (Cook, 2008). Cohen’s D effect sizes and/or p-values were reported in Tables 2 and 3. If the effect size was greater in one group, this was considered of value and noted. In the absence of effect sizes, p-values were used to make a determination on the effectiveness of a treatment approach for the selected group.
Quality assessment Risk of bias in all eligible studies was determined using the validated Physiotherapy Evidence Database (PEDro) scale (de Mortin, 2009). The scale is an 11-item retrospective guide for assessing internal validity of randomized controlled trials. Item 1, ‘eligibility criteria were specified’, pertains to external validity and is not included in the overall PEDro score. The other 10 items were answered ‘yes’ or ‘no’, ‘yes’ answers were summed to obtain a final score out of 10. Higher scores reflect higher study quality (PEDro, 2014). The assessment was completed independently by two members of the research team (HM and RE). Data from the assessment were recorded in Table 4. Kappa values for each item were calculated to demonstrate the level of PEDro scoring agreement between the authors.
Results/findings The database searches yielded 148 articles, with an additional nine articles identified through a hand
S. Arnold et al.
search. Eliminating duplicates reduced the total from 157 to 104 articles. There were nine studies that were selected to be included in this review. For two studies, the data that were needed to complete the summary analysis were not reported. The author was contacted through email, but the results were unable to be obtained; therefore, only seven studies met our inclusion criteria (Figure 1). The calculated Kappa scores for the inter-rater reliability of title reviews, abstract reviews and full-text reviews were 0.904 (95% confidence interval [CI] = 0.773–0.904), 0.914 (95% CI = 0.663–0.914) and 0.850 (95% CI = 0.442–0.987), respectively. Sample sizes of the seven studies ranged from 7 to 53. All study designs were randomized controlled trials or randomized clinical trials. Study descriptions and population characteristics can be found in Table 1. Risk of bias assessment with the PEDro scale resulted in the following: one study scored 2/10 (Clendaniel, 2010), two scored a 4/10 (Cohen and Kimball, 2004; Teggi et al., 2009), three scored a 5/10 (Giray et al., 2009; Pavlou et al., 2012; Marioni et al., 2013) and one article scored a 7/10 (Herdman et al., 2003). All of the studies (100%) complied with random allocation. Other more frequently satisfied criteria of the PEDro scale were providing measures for one key outcome (85.7%), baseline comparability (71.4%) and measuring 85% of initial subjects (71.4%). Assessors were blinded 28.6% while blinding of therapists and providing between-group statistical comparisons occurred 14.3%. No article met concealed allocation, blind therapist or intention to treat criteria. Table 2 provides full details of the PEDro scoring for all studies included. There was 100% agreement between authors for all items of the PEDro scale. The classification of the intervention types used within the seven chosen articles (Table 1) is of particular interest. According to the vestibular rehabilitation strategies utilized within the intervention groups in each study, the studies clustered into one of three groups: 1) adaptation and habituation compared with substitution (Cohen and Kimball, 2004), 2) adaptation and substitution compared with habituation and substitution (Herdman et al., 2003; Clendaniel, 2010; Marioni et al., 2013 and 3) adaptation, substitution and habituation compared with a control (Giray et al., 2009; Teggi et al., 2009; Pavlou et al., 2012). Outcome measures used in the included studies can be found in Tables 2 and 3. Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
†a
7
21
40
41
16
Herdman et al., 2003
Marioni et al., 2013
Giray et al., 2009
Pavlou et al., 2012
53
Sample size (n)
Clendaniel, 2010
Cohen and Kimball, 2004
Author, year
Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
IG3: 39.8 (NR)
IG2: 42 (NR)
CG: 50.38 (18.59) IG1: 42.1 (NR)
CG 1: 48 (4) CG 2: 42 (9) IG1: 52.4 (14.9)
IG1: 45 (7)
CG: 64.9 (16.2)
IG1: 52.4 (14.9)
43.9 (NR)
51.1 (13.6)
Age (SD)
Adaptation Habituation Substitution
Adaptation Habituation Substitution
Adaptation Substitution
Adaptation Substitution
Adaptation Habituation Substitution
Adaptation Habituation Substitution
Intervention type
IG3: this group was the first five participants of IG1. After receiving initial interventions, a wash out period was performed, and they received the same interventions as IG2
IG1: HEP for one group containing rapid head movements and HEP for second group containing rapid head movements with attention. Movements intended to cause mild vertigo. Encouraged to stare at stationary object during head movement IG2: HEP containing slow repetitive head movements IG1: HEP containing progressive adaptation exercises using ×1/×2 viewing gaze stability in numerous plans of motion while seating or standing IG2: HEP containing progressive habituation exercises using large amplitude head and trunk movements while seating or standing IG1: HEP containing progressive ×1/×2 viewing, gaze stability, eye/head movements between two targets and balance and gait exercises CG: HEP consisting of placebo saccadic eye movements, and balance and gait IG1: customized posturography assisted limits of stability vestibular rehab programme consisting of body weight shifting. HEP containing head movements and standing balance exercises CG1: no training or exercise CG2: spontaneous recovery IG1: individualized rehabilitation programme based on patient presentation. Treatments varied from ×1/×2 viewing, gaze stability during ambulation and foam exercises or visual desensitization. HEP including habituation, substitution and balance CG: no training or exercise IG1: ×1 viewing, head movements, standing on foam, marching and directional stepping all while viewing static virtual reality images. HEP Cawthorne–Cooksey exercise programme consisting of eye, head and trunk movements while sitting, standing or ambulating IG2: same as IG1 except pt. viewing dynamic virtual reality images
Group description
Table 1. Summary of trials included in review (n = 7): sample size, age, intervention type, group description, dosage
(Continues)
45 minutes; 2× per week; 4 weeks (IG) 15–20 minutes; 3× per week; 4 weeks (HEP) 45 minutes; 2× per week; 4 weeks (IG) 15–20 minutes; 3× per week; 4 weeks (HEP)
15–20 minutes; 3× per week; 4 weeks (HEP)
— 45 minutes; 2× per week; 4 weeks (IG)
— — 1 minute; 3× per day; 4 weeks (adaptation) 30–40; 2× per week; 4 weeks (balance) 30–40 minutes; 2× per day; 4 weeks (HEP)
30 minutes; 1× per week; 5 weeks (HEP) 3× per day; 5 weeks (HEP)
—
20–30 minutes; 4–5× per day; 4 weeks (HEP) 20 minutes; 4–5× per day; 4 weeks (balance and gait)
×3 per day; 6 weeks (IG)
×3 per day; 6 weeks (IG)
5–10 minutes; 5× per day; 24 weeks (HEP)
5–10 minutes; 5× per day; 24 weeks (HEP)
Dosage
S. Arnold et al. Effectiveness of Vestibular Rehabilitation: A Systematic Review
The authors in this study initially reported on three different intervention groups, but the intervention groups of rapid head movements and rapid head movements with attention were combined in their results. In
S. Arnold et al.
Adaptation and habituation compared with substitution In one article, the authors compared a combined intervention of adaptation and habituation exercises with an intervention utilizing substitution and assessed the interventions’ effects in treating unilateral peripheral vestibular hypofunction (Cohen and Kimball, 2004). Using the PEDro scale, this study was found to be at risk of bias. In this study, the authors reported on the time to perform a repetitive head movement task, with one endpoint value (6 months) showing a large effect size for both the adaptation and habituation group (Cohen’s D = 2.82) and the substitution group (Cohen’s D = 2.85), with a trivial difference between groups. Adaptation and substitution compared with habituation and substitution
order to match the results reported in Table 3, the intervention groups are also combined in this table.
CG1: 51.4 (9.1)
SD, standard deviation; NR, not reported; IG, intervention group; CG, control group; HEP, home exercise programme.
a
—
5× per day; 4 weeks (HEP)
45 minutes: 3× per week: 4 weeks (IG)
IG1: stabilometric platform centre of gravity exercises with visual feedback on a monitor. Standing balance by performing alternating standing heel to toe rocking. Additional interventions included staring at a target immediately after performing head movements, positional changes and head shaking CG1: performed only on daily activities Adaptation Habituation Substitution IG1: 53.5 (9.8) 40 Teggi et al., 2009
Author, year
Table 1. (Continued)
Sample size (n)
Age (SD)
Intervention type
Group description
Dosage
Effectiveness of Vestibular Rehabilitation: A Systematic Review
A combined programme of adaptation and substitution was compared with a combined programme of habituation and substitution in one study of poor quality (Clendaniel, 2010). Large effect sizes were demonstrated for both intervention groups (3.82 for adaptation and substitution and 2.65 for habituation and substitution) as measured by the dizziness handicap inventory (DHI). Active and passive dynamic visual acuity (DVA) showed greater improvements in habituation exercise combined with substitution using large amplitude cervical and trunk movements (effect size 2.01 for active and 1.13 for passive DVA) over progressive adaptation exercise combined with substitution using ×1 viewing (effect size 0.29 for active and 0.23 for passive DVA). Motion sensitivity quotient scores demonstrated greater improvement in adaptation exercise with substitution using ×1 viewing (Cohen’s D = 0.90) than in habituation exercise with substitution using large amplitude cervical and trunk movements (Cohen’s D = 0.61) (Clendaniel, 2010). Two studies (Herdman et al., 2003; Marioni et al., 2013), one of which ranked poorly on the PEDro scale (Marioni et al., 2013), compared a combined programme of adaptation and substitution with a control group. Table 2 lists the outcome measures used in these studies. Reported measures for limits of stability yielded no to small effect in the control groups and no to large effect in the intervention group for Marioni et al. (2013). Various outcomes of the modified clinical test of sensory interaction of balance were reported by Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
Marioni et al., 2013†a
Herdman et al., 2003
Giray et al., 2009
Author, year
LOS RT (s)
FOAM EC
FOAM EO
FIRM EC
mCTSIB (°·s ) FIRM EO
1
DVA active (LogMAR) DVA passive (LogMAR) Oscillopsia VAS 0.2 (0.1), 0.2 (0.1) 0.3 (0.1), 0.3 (0.2) 0.2 (0.1), 0.2 (0.1) 0.4(0.2), 0.4(0.2) 0.4(0.1), 0.4(0.1) 0.7 (0.2), 0.6 (0.2) 1.3 (0.2), 1.3 (0.2) 2.8 (0.6), 2.3 (0.4)
CG1 A: 0.6 (0.2), 0.6 (0.2) P: 0.4 (0.1), 0.4 (0.1) R: 0.6 (0.2), 0.6 (0.2) L: 0.6 (0.3), 0.6 (0.3) CG2: A: 0.7 (0.2), 0.7 (0.2) P: 0.6 (0.3), 0.6 (0.3)
CG1: CG2: CG1: CG2: CG1: CG2: CG1: CG2:
0.323 (0.117), NR (NR) 0.374 (0.085), NR (NR) 2.2 (2.1), NR (NR)
0.2 (NR), 0.40 (NR) 0.3 (NR), 0.40 (NR) 0.75 (NR), 0.8 (Herdman5 [NR]) 2.05 (NR), 2.25 (NR) 0.8 (NR), 1.10 (NR) 3.6 (NR), 2.9 (NR)
mCTSIB (d·s ) FIRM EO FIRM EC FOAM EO FOAM EC Total Unsteadiness VAS
1
58 (NR), 60 (NR) 20 (NR), 20 (NR) 19 (NR), 21 (NR) 19 (NR), 18 (NR) 2 (NR), 3 (NR) 7 (NR), 7 (NR) 12 (NR), 11 (NR) 54.5 (NR), 55 (NR)
Control group R, EP (SD)
DHI Total Emotional Functional Physical Mild Moderate Severe BBS
Outcome measure
A: 0.8 (0.3), 0.6 (0.2) P: 0.7 (0.2), 0.4 (0.2) I: 0.7 (0.2), 0.7 (0.2) C: 0.8 (0.3), 0.7 (0.2)
2.6 (0.7), 1.4 (0.2)
0.6(0.2), 0.4(0.1)
0.4 (0.2), 0.3 (0.1)
0.2 (0.2), 0.3 (0.1)
0.372 (0.155), NR (NR) 0.408 (0.173), NR (NR) 3.2 (3.2), NR (NR)
0.25 (NR), 0.20 (NR) 0.3 (NR), 0.25 (NR) 1 (NR), 0.70 (NR) 2.6 (NR), 1.85 (NR) 1.1 (NR), 0.85 (NR) 4.45 (NR), 1.35 (NR)
64 (NR), 22 (NR) 17 (NR), 5 (NR) 28 (NR), 12 (NR) 20 (NR), 9 (NR) 1 (NR), 13 (NR) 8 (NR), 4 (NR) 11 (NR), 3 (NR) 54 (NR), 56 (NR)
IG1 R, EP (SD)
A: NR, NS, NS P: NR, NS, 0.002 R: NR/I: NS, NS L: NR/C: NS, NS
NR, 0.03, 0.00004
NR, NS, 0.02
NR, NS, NS
A: 0.00, 0.00, 0.79 P: 0.00, 0.00, 1.50 R: 0.00/I: 0.00, 0.00 L: 0.00/C: 0.00, 0.39
0.00, 0.98, 2.33
0.00, 0.50, 1.27
0.00, 0.00, 0.63
0.00, 0.00, 0.63
UC, UC UC, UC UC, UC
0.07,
The Effectiveness of Vestibular Rehabilitation Interventions in Treating Unilateral Peripheral Vestibular Disorders: A Systematic Review Scott A. Arnold, Aaron M. Stewart, Heather M. Moor, Rita C. Karl & Jennifer C. Reneker* Division of Physical Therapy, Walsh University, North Canton, OH, USA
Abstract Background and Purpose. Various types of vestibular rehabilitation therapy are routinely used in clinical practice to treat unilateral peripheral vestibular hypofunction. The purpose of this systematic review was to compare the effectiveness of vestibular rehabilitation interventions (adaptation, substitution and habituation) in people with unilateral peripheral vestibular hypofunction, exclusionary of benign paroxysmal positional vertigo and Meniere’s disease. Methods. A search of the literature was conducted using PubMed, CINAHL and Scopus. Studies were eligible for inclusion if they were 1) a randomized controlled trial or randomized clinical trial; 2) written in English; 3) of participants with a unilateral, peripheral vestibular hypofunction; 4) of a conservative treatment approach only; and 5) with human subjects. Quality was assessed by two authors using the Physiotherapy Evidence Database scale. Effect size was calculated to determine the effect of treatment within each study group. Results. Seven papers were selected for inclusion. Physiotherapy Evidence Database scores ranged from 2/10 to 7/10. Interventions within the selected studies included combinations of adaptation, habituation, substitution or substitution by itself. Calculated effect sizes, or significance values, revealed that all interventions demonstrated effectiveness. Two studies reported improvements on the dynamic gait index, and a large difference was seen between intervention groups of the two studies. Discussion. Results suggest that vestibular therapy for unilateral peripheral vestibular hypofunction is effective. When considering all seven studies included in the review, it is difficult to determine the superiority of one intervention over another in treating unilateral peripheral vestibular hypofunction except when patient outcomes are captured by the dynamic gait index or dizziness handicap inventory. Many studies in this review demonstrate notable biases, suggesting that results should be used with caution. Future research should aim to use a common set of measures to capture outcomes. Copyright © 2015 John Wiley & Sons, Ltd. Received 2 April 2014; Revised 1 December 2014; Accepted 25 April 2015 Keywords dizziness/rehabilitation; exercise therapy; head movements; vestibular diseases/rehabilitation *Correspondence Jennifer Reneker, Division of Physical Therapy, Walsh University, 2020 East Maple St.,44720, North Canton, OH, USA. Email: [email protected]
Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/pri.1635
Introduction Dizziness is a common complaint reported in primary care practice and results in more than 6 million visits
Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
to physician offices per year in the United States alone (Brodovsky and Vnenchak, 2013). Dizziness symptoms were reported in 1 out of 5 individuals of working age (Yardley et al., 1998). It is reported that health-related
S. Arnold et al.
Effectiveness of Vestibular Rehabilitation: A Systematic Review
quality of life, both mental and physical, is significantly impaired in those with dizziness (Weidt et al., 2014). Although dizziness can be caused by a variety of medical conditions, it is estimated that 50% of cases are due to vestibular dysfunction (Hall and Cox, 2009). Many patients with a complaint of dizziness are referred to physical therapy for treatment to relieve the dizziness and symptoms associated with disequilibrium (Tee and Chee, 2005). Vestibular rehabilitation is described predominantly as a movement and exercise-based approach that includes adaptation, substitution (including postural control strategies) and habituation (Hillier and McDonnell, 2011). Unilateral peripheral vestibular hypofunction is a disorder that affects one side of the vestibular system and excludes vestibular dysfunction related to the brain (Hillier and McDonnell, 2011). Potential causes of unilateral peripheral vestibular hypofunction include vestibular neuritis, vestibular labyrinthitis, perilymphatic fistula and acoustic neuroma (Brodovsky and Vnenchak, 2013). In order to differentiate central from peripheral vestibular dysfunction, a thorough examination must be completed. The specific procedures included in vestibular rehabilitation should be aimed at the presenting symptoms to produce favourable outcomes (Deveze et al., 2014). Because of the high prevalence of vestibular dysfunction and resultant dizziness and because dizziness negatively impacts health-related quality of life, it is important that effective treatments of vestibular dysfunction are identified. Previous research on the effectiveness of vestibular rehabilitation for unilateral peripheral vestibular dysfunction has demonstrated significant differences in favour of vestibular rehabilitation when compared with control (placebo, sham, usual care or no intervention) (Hillier and McDonnell, 2011). To our knowledge, no major differences have been found in treatment effectiveness among interventions (Hillier and McDonnell, 2011). Identifying effective interventions may hasten recovery from dizziness related to vestibular dysfunction by giving clinicians insight into interventions to use during treatment. Benign paroxysmal positional vertigo (BPPV) has accepted treatment approaches (Hilton and Pinder, 2004; Hillier and McDonnell, 2011; Bruintjes et al., 2014), and Meniere’s Disease has several studied treatment approaches with varying efficacy (Pullens and van Benthem, 2011; Hillier and McDonnell, 2011; Phillips and Westerberg, 2011; Pullens et al., 2013); therefore,
these will not be the focus of the current study. The purpose of this systematic review was to compare the effectiveness of vestibular rehabilitation interventions (adaptation, substitution and habituation) in people with unilateral peripheral vestibular hypofunction, exclusionary of BPPV and Meniere’s Disease.
Methods Study design The present study utilized the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA, 2014), a prospective guide for writing systematic reviews and meta-analyses, in the research and reporting of data (PRISMA Statement). Search strategy Individualized, computer-based search strategies for PubMed (Appendix 1), Scopus and CINAHL were developed in November 2013. PubMed was searched using a comprehensive strategy that included search terms related to vestibular rehabilitation for unilateral peripheral vestibular disorders. There were no limits applied for the publication date of articles, but the following limits were applied to the search results: (1) Humans, (2) studies published in English, (3) randomized control trials and (4) controlled clinical trials. All remaining databases were searched using comparable strategies. A hand search was performed by exploring the references of included articles to identify studies not captured through the electronic searches. Eligibility criteria Studies were included in the review if the following inclusion criteria were met: 1) a randomized controlled trial or randomized clinical trial; 2) written in English; 3) of participants with a unilateral, peripheral vestibular hypofunction; 4) of a conservative treatment approach only; and 5) with human subjects. Studies omitted from the review included 1) BPPV, 2) Meniere’s Disease or 3) central vestibular disorders within the subject population, as this was beyond the scope of this review. Study selection Studies were selected for inclusion in this systematic review based on a stepwise process outlined in Figure 1. Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
S. Arnold et al.
Effectiveness of Vestibular Rehabilitation: A Systematic Review
Figure 1. Study flow for the systematic review. BPPV, benign paroxysmal positional vertigo; CNS, central nervous system; RCT, randomized controlled/clinical trial.
At each step, articles were independently screened by two members of the research team, based on the inclusion criteria, in an effort to reduce bias. Title and abstract were reviewed by two members (HM and RE), as well as full text (SA and AS). In cases of discrepancy between the two initial reviewers, a third reviewer determined the study’s inclusion or exclusion.
Data collection Data were collected from the articles accepted in this study by two members of the research team in tandem (SA and AS). Both members were required to agree upon the selected data for them to be included. Data were chosen based on their usefulness in answering the research question. The study sample characteristics, the intervention type and the outcome measures of the Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
selected studies were included in this review. Where data were not presented in the original research article, effort was made to retrieve unavailable and unclear data from the author. Vestibular rehabilitation strategies used in each study intervention were grouped according to the three most common vestibular rehabilitation intervention types including adaptation, habituation and substitution (Table 1). In this review, the following classification was used for the interventions in the included studies: adaptation exercises included exercises designed to improve vestibulo-ocular reflex gain. Typically, these interventions consist of fixing gaze on a target and keeping that target in focus while moving the head (Childs, 2010; Deveze et al., 2014). Habituation exercises included interventions with repetitious exposure of the subject to symptom-provoking
Effectiveness of Vestibular Rehabilitation: A Systematic Review
activities. This is performed until the vestibular system becomes accustomed and symptoms are no longer produced (Boyer et al., 2008; Childs, 2010). Substitution exercises included interventions to train patients to use their somatosensory and visual systems (Childs, 2010). Where available, p-values of within group changes from pre-test (reference) to post-test (endpoint) for the outcome measures were reported. Cohen’s D effect sizes from individual studies were calculated for within-group changes for the selected outcome measures from pre-test (reference) to post-test (endpoint) when such data were presented. Values of between-group comparisons were unable to be calculated because of limited or unclear data. Effect size is a measure of the magnitude of an intervention or the strength of an effect of a particular treatment (Cook, 2008). In order to understand the magnitude of an intervention, a classification system has been proposed, with values falling below 0.2 being trivial, at or above 0.2 but below 0.5 being small, at or above 0.50 but below 0.80 being medium, and at or above 0.80 being large (Cook, 2008). Cohen’s D effect sizes and/or p-values were reported in Tables 2 and 3. If the effect size was greater in one group, this was considered of value and noted. In the absence of effect sizes, p-values were used to make a determination on the effectiveness of a treatment approach for the selected group.
Quality assessment Risk of bias in all eligible studies was determined using the validated Physiotherapy Evidence Database (PEDro) scale (de Mortin, 2009). The scale is an 11-item retrospective guide for assessing internal validity of randomized controlled trials. Item 1, ‘eligibility criteria were specified’, pertains to external validity and is not included in the overall PEDro score. The other 10 items were answered ‘yes’ or ‘no’, ‘yes’ answers were summed to obtain a final score out of 10. Higher scores reflect higher study quality (PEDro, 2014). The assessment was completed independently by two members of the research team (HM and RE). Data from the assessment were recorded in Table 4. Kappa values for each item were calculated to demonstrate the level of PEDro scoring agreement between the authors.
Results/findings The database searches yielded 148 articles, with an additional nine articles identified through a hand
S. Arnold et al.
search. Eliminating duplicates reduced the total from 157 to 104 articles. There were nine studies that were selected to be included in this review. For two studies, the data that were needed to complete the summary analysis were not reported. The author was contacted through email, but the results were unable to be obtained; therefore, only seven studies met our inclusion criteria (Figure 1). The calculated Kappa scores for the inter-rater reliability of title reviews, abstract reviews and full-text reviews were 0.904 (95% confidence interval [CI] = 0.773–0.904), 0.914 (95% CI = 0.663–0.914) and 0.850 (95% CI = 0.442–0.987), respectively. Sample sizes of the seven studies ranged from 7 to 53. All study designs were randomized controlled trials or randomized clinical trials. Study descriptions and population characteristics can be found in Table 1. Risk of bias assessment with the PEDro scale resulted in the following: one study scored 2/10 (Clendaniel, 2010), two scored a 4/10 (Cohen and Kimball, 2004; Teggi et al., 2009), three scored a 5/10 (Giray et al., 2009; Pavlou et al., 2012; Marioni et al., 2013) and one article scored a 7/10 (Herdman et al., 2003). All of the studies (100%) complied with random allocation. Other more frequently satisfied criteria of the PEDro scale were providing measures for one key outcome (85.7%), baseline comparability (71.4%) and measuring 85% of initial subjects (71.4%). Assessors were blinded 28.6% while blinding of therapists and providing between-group statistical comparisons occurred 14.3%. No article met concealed allocation, blind therapist or intention to treat criteria. Table 2 provides full details of the PEDro scoring for all studies included. There was 100% agreement between authors for all items of the PEDro scale. The classification of the intervention types used within the seven chosen articles (Table 1) is of particular interest. According to the vestibular rehabilitation strategies utilized within the intervention groups in each study, the studies clustered into one of three groups: 1) adaptation and habituation compared with substitution (Cohen and Kimball, 2004), 2) adaptation and substitution compared with habituation and substitution (Herdman et al., 2003; Clendaniel, 2010; Marioni et al., 2013 and 3) adaptation, substitution and habituation compared with a control (Giray et al., 2009; Teggi et al., 2009; Pavlou et al., 2012). Outcome measures used in the included studies can be found in Tables 2 and 3. Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
†a
7
21
40
41
16
Herdman et al., 2003
Marioni et al., 2013
Giray et al., 2009
Pavlou et al., 2012
53
Sample size (n)
Clendaniel, 2010
Cohen and Kimball, 2004
Author, year
Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
IG3: 39.8 (NR)
IG2: 42 (NR)
CG: 50.38 (18.59) IG1: 42.1 (NR)
CG 1: 48 (4) CG 2: 42 (9) IG1: 52.4 (14.9)
IG1: 45 (7)
CG: 64.9 (16.2)
IG1: 52.4 (14.9)
43.9 (NR)
51.1 (13.6)
Age (SD)
Adaptation Habituation Substitution
Adaptation Habituation Substitution
Adaptation Substitution
Adaptation Substitution
Adaptation Habituation Substitution
Adaptation Habituation Substitution
Intervention type
IG3: this group was the first five participants of IG1. After receiving initial interventions, a wash out period was performed, and they received the same interventions as IG2
IG1: HEP for one group containing rapid head movements and HEP for second group containing rapid head movements with attention. Movements intended to cause mild vertigo. Encouraged to stare at stationary object during head movement IG2: HEP containing slow repetitive head movements IG1: HEP containing progressive adaptation exercises using ×1/×2 viewing gaze stability in numerous plans of motion while seating or standing IG2: HEP containing progressive habituation exercises using large amplitude head and trunk movements while seating or standing IG1: HEP containing progressive ×1/×2 viewing, gaze stability, eye/head movements between two targets and balance and gait exercises CG: HEP consisting of placebo saccadic eye movements, and balance and gait IG1: customized posturography assisted limits of stability vestibular rehab programme consisting of body weight shifting. HEP containing head movements and standing balance exercises CG1: no training or exercise CG2: spontaneous recovery IG1: individualized rehabilitation programme based on patient presentation. Treatments varied from ×1/×2 viewing, gaze stability during ambulation and foam exercises or visual desensitization. HEP including habituation, substitution and balance CG: no training or exercise IG1: ×1 viewing, head movements, standing on foam, marching and directional stepping all while viewing static virtual reality images. HEP Cawthorne–Cooksey exercise programme consisting of eye, head and trunk movements while sitting, standing or ambulating IG2: same as IG1 except pt. viewing dynamic virtual reality images
Group description
Table 1. Summary of trials included in review (n = 7): sample size, age, intervention type, group description, dosage
(Continues)
45 minutes; 2× per week; 4 weeks (IG) 15–20 minutes; 3× per week; 4 weeks (HEP) 45 minutes; 2× per week; 4 weeks (IG) 15–20 minutes; 3× per week; 4 weeks (HEP)
15–20 minutes; 3× per week; 4 weeks (HEP)
— 45 minutes; 2× per week; 4 weeks (IG)
— — 1 minute; 3× per day; 4 weeks (adaptation) 30–40; 2× per week; 4 weeks (balance) 30–40 minutes; 2× per day; 4 weeks (HEP)
30 minutes; 1× per week; 5 weeks (HEP) 3× per day; 5 weeks (HEP)
—
20–30 minutes; 4–5× per day; 4 weeks (HEP) 20 minutes; 4–5× per day; 4 weeks (balance and gait)
×3 per day; 6 weeks (IG)
×3 per day; 6 weeks (IG)
5–10 minutes; 5× per day; 24 weeks (HEP)
5–10 minutes; 5× per day; 24 weeks (HEP)
Dosage
S. Arnold et al. Effectiveness of Vestibular Rehabilitation: A Systematic Review
The authors in this study initially reported on three different intervention groups, but the intervention groups of rapid head movements and rapid head movements with attention were combined in their results. In
S. Arnold et al.
Adaptation and habituation compared with substitution In one article, the authors compared a combined intervention of adaptation and habituation exercises with an intervention utilizing substitution and assessed the interventions’ effects in treating unilateral peripheral vestibular hypofunction (Cohen and Kimball, 2004). Using the PEDro scale, this study was found to be at risk of bias. In this study, the authors reported on the time to perform a repetitive head movement task, with one endpoint value (6 months) showing a large effect size for both the adaptation and habituation group (Cohen’s D = 2.82) and the substitution group (Cohen’s D = 2.85), with a trivial difference between groups. Adaptation and substitution compared with habituation and substitution
order to match the results reported in Table 3, the intervention groups are also combined in this table.
CG1: 51.4 (9.1)
SD, standard deviation; NR, not reported; IG, intervention group; CG, control group; HEP, home exercise programme.
a
—
5× per day; 4 weeks (HEP)
45 minutes: 3× per week: 4 weeks (IG)
IG1: stabilometric platform centre of gravity exercises with visual feedback on a monitor. Standing balance by performing alternating standing heel to toe rocking. Additional interventions included staring at a target immediately after performing head movements, positional changes and head shaking CG1: performed only on daily activities Adaptation Habituation Substitution IG1: 53.5 (9.8) 40 Teggi et al., 2009
Author, year
Table 1. (Continued)
Sample size (n)
Age (SD)
Intervention type
Group description
Dosage
Effectiveness of Vestibular Rehabilitation: A Systematic Review
A combined programme of adaptation and substitution was compared with a combined programme of habituation and substitution in one study of poor quality (Clendaniel, 2010). Large effect sizes were demonstrated for both intervention groups (3.82 for adaptation and substitution and 2.65 for habituation and substitution) as measured by the dizziness handicap inventory (DHI). Active and passive dynamic visual acuity (DVA) showed greater improvements in habituation exercise combined with substitution using large amplitude cervical and trunk movements (effect size 2.01 for active and 1.13 for passive DVA) over progressive adaptation exercise combined with substitution using ×1 viewing (effect size 0.29 for active and 0.23 for passive DVA). Motion sensitivity quotient scores demonstrated greater improvement in adaptation exercise with substitution using ×1 viewing (Cohen’s D = 0.90) than in habituation exercise with substitution using large amplitude cervical and trunk movements (Cohen’s D = 0.61) (Clendaniel, 2010). Two studies (Herdman et al., 2003; Marioni et al., 2013), one of which ranked poorly on the PEDro scale (Marioni et al., 2013), compared a combined programme of adaptation and substitution with a control group. Table 2 lists the outcome measures used in these studies. Reported measures for limits of stability yielded no to small effect in the control groups and no to large effect in the intervention group for Marioni et al. (2013). Various outcomes of the modified clinical test of sensory interaction of balance were reported by Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
Physiother. Res. Int. (2015) © 2015 John Wiley & Sons, Ltd.
Marioni et al., 2013†a
Herdman et al., 2003
Giray et al., 2009
Author, year
LOS RT (s)
FOAM EC
FOAM EO
FIRM EC
mCTSIB (°·s ) FIRM EO
1
DVA active (LogMAR) DVA passive (LogMAR) Oscillopsia VAS 0.2 (0.1), 0.2 (0.1) 0.3 (0.1), 0.3 (0.2) 0.2 (0.1), 0.2 (0.1) 0.4(0.2), 0.4(0.2) 0.4(0.1), 0.4(0.1) 0.7 (0.2), 0.6 (0.2) 1.3 (0.2), 1.3 (0.2) 2.8 (0.6), 2.3 (0.4)
CG1 A: 0.6 (0.2), 0.6 (0.2) P: 0.4 (0.1), 0.4 (0.1) R: 0.6 (0.2), 0.6 (0.2) L: 0.6 (0.3), 0.6 (0.3) CG2: A: 0.7 (0.2), 0.7 (0.2) P: 0.6 (0.3), 0.6 (0.3)
CG1: CG2: CG1: CG2: CG1: CG2: CG1: CG2:
0.323 (0.117), NR (NR) 0.374 (0.085), NR (NR) 2.2 (2.1), NR (NR)
0.2 (NR), 0.40 (NR) 0.3 (NR), 0.40 (NR) 0.75 (NR), 0.8 (Herdman5 [NR]) 2.05 (NR), 2.25 (NR) 0.8 (NR), 1.10 (NR) 3.6 (NR), 2.9 (NR)
mCTSIB (d·s ) FIRM EO FIRM EC FOAM EO FOAM EC Total Unsteadiness VAS
1
58 (NR), 60 (NR) 20 (NR), 20 (NR) 19 (NR), 21 (NR) 19 (NR), 18 (NR) 2 (NR), 3 (NR) 7 (NR), 7 (NR) 12 (NR), 11 (NR) 54.5 (NR), 55 (NR)
Control group R, EP (SD)
DHI Total Emotional Functional Physical Mild Moderate Severe BBS
Outcome measure
A: 0.8 (0.3), 0.6 (0.2) P: 0.7 (0.2), 0.4 (0.2) I: 0.7 (0.2), 0.7 (0.2) C: 0.8 (0.3), 0.7 (0.2)
2.6 (0.7), 1.4 (0.2)
0.6(0.2), 0.4(0.1)
0.4 (0.2), 0.3 (0.1)
0.2 (0.2), 0.3 (0.1)
0.372 (0.155), NR (NR) 0.408 (0.173), NR (NR) 3.2 (3.2), NR (NR)
0.25 (NR), 0.20 (NR) 0.3 (NR), 0.25 (NR) 1 (NR), 0.70 (NR) 2.6 (NR), 1.85 (NR) 1.1 (NR), 0.85 (NR) 4.45 (NR), 1.35 (NR)
64 (NR), 22 (NR) 17 (NR), 5 (NR) 28 (NR), 12 (NR) 20 (NR), 9 (NR) 1 (NR), 13 (NR) 8 (NR), 4 (NR) 11 (NR), 3 (NR) 54 (NR), 56 (NR)
IG1 R, EP (SD)
A: NR, NS, NS P: NR, NS, 0.002 R: NR/I: NS, NS L: NR/C: NS, NS
NR, 0.03, 0.00004
NR, NS, 0.02
NR, NS, NS
A: 0.00, 0.00, 0.79 P: 0.00, 0.00, 1.50 R: 0.00/I: 0.00, 0.00 L: 0.00/C: 0.00, 0.39
0.00, 0.98, 2.33
0.00, 0.50, 1.27
0.00, 0.00, 0.63
0.00, 0.00, 0.63
UC, UC UC, UC UC, UC
0.07,
