Introduction
Ménière's disease is a common, idiopathic, inner-ear disorder which causes episodic symptoms of vertigo with associated ear pressure, tinnitus and hearing loss. Today, despite the advent of high resolution imaging, molecular diagnostics and single-cell physiology, the pathophysiology of Ménière's disease has remained unclear.Reference Jeffrey, Vrabec and Lawrence1 There are several theories regarding the pathogenesis of Ménière's disease. Some studies suggested that an abnormal endolymph drainage system may predispose patients to future development of Ménière's disease.Reference Cummings, Flint, Harker, Haughy, Richardson and Robbins2 Schuknecht postulated that ruptures in the membranous labyrinth allow leakage of potassium-rich endolymph into the perilymph. This results in a decrease in auditory and vestibular neuronal outflow consistent with hearing loss and features of acute vestibular paralysis.Reference Schuknecht3, Reference Schuknecht4
There is no diagnostic laboratory test for Ménière's disease, and complete history-taking is the only way of making the diagnosis.Reference Cummings, Flint, Harker, Haughy, Richardson and Robbins2 Various classifications of Ménière's disease patients have been developed, with the aim of facilitating treatment selection and monitoring. These classifications have been based either on audiological criteria (reflecting cochlear lesions) or balance criteria (reflecting posterior labyrinth dysfunction).Reference Shea5, 6 The most widely used audiological criteria involve audiometric hearing thresholds and tinnitus intensity assessment. Subjective vestibular parameters have also been used for Ménière's disease staging, notably the frequency of typical vertigo attacks and the incapacity scores obtained by questionnaires.Reference Soto, Labella, Santos, del Rio, Lirola and Cabanas7
Recently, computerised dynamic posturography has been introduced for objective evaluation of patients with balance disorders.Reference Havia, Kentala and Pyykko8 Computerised dynamic posturography determines the patient's centre of gravity in six different situations, with or without suppression of extravestibular sensory information. Although computerised dynamic posturography is not a specific diagnostic tool for Ménière's disease or other causes of imbalance, it supplies important information about patients' functional balance status in the course of their disease. In this study, we aimed to evaluate the use of computerised dynamic posturography as an objective and quantitative balance test for the staging of Ménière's disease patients.
Materials and methods
A total of 53 patients with a definitive diagnosis of Ménière's disease were included in the study. Patients were referred to our tertiary university hospital (Hazrat Rasoul Akram Hospital) between August 2006 and July 2007.
All patients met American Academy of Otolaryngology criteria for definitive diagnosis of Ménière's disease (i.e. two or more spontaneous and unequivocal episodes of at least 20 minutes' duration; sensory hearing loss identified audiometrically on at least one occasion; tinnitus and/or feeling of fullness in the affected ear; and exclusion of other possible causes).6 Patients who had been treated by ablative surgery and/or intratympanic gentamicin were excluded from the study.
Twenty-five males (47.2 per cent) and 28 females (52.8 per cent) were entered into the study. The patients' mean age was 41 years (range 15–62 years). Ménière's disease was bilateral in five cases and unilateral in the remainder.
All patients underwent at least one computerised dynamic posturography assessment. In all cases, the patient's centre of gravity was determined under six sensory conditions: (1) immobile surface, immobile visual surroundings, eyes open; (2) immobile surface, immobile visual surroundings, eyes closed; (3) immobile surface, mobile visual surroundings, eyes open; (4) mobile surface, immobile visual surroundings, eyes open; (5) mobile surface, immobile visual surroundings, eyes closed; and (6) mobile surface, mobile visual surroundings, eyes open.
Each of these six conditions was tested three times, giving a total of 18 tests per session. All the tests were performed in one session for each patient. The following parameters were used in the data analysis: (1) condition five scores (see above); (2) condition six scores (see above); and (3) vestibular ratios = [condition five score]/[condition one score] ×100, indicating the patient's ability to use vestibular information for maintenance of his or her balance.
In this study, we assessed the possible correlation between these three parameters and the patient's bone conduction hearing threshold, the time elapsed since the last vertigo attack and the Ménière's disease activity level.
Patient's bone conduction hearing thresholds were classified, on the basis of the American Academy of Otolaryngology system, into four stages: stage one = mean threshold <26 dB; stage two = mean threshold 26–40 dB; stage three = mean threshold 41–70 dB; and stage four = mean threshold >70 dB.6
The mean hearing threshold for each patient was calculated as the mean hearing threshold for three frequencies (500, 1000 and 2000 Hz). Twenty patients (37.7 per cent) were placed in the stage one hearing threshold group, 13 (24.5 per cent) in stage two and 20 (37.7 per cent) in stage three. None of our patients had a mean threshold of >70 dB (stage four).
Time elapsed since the last Ménière's disease attack was determined by asking the patient on the day of the dynamic posturography session and quantifying their answer in days.
Ménière's disease activity level was quantified on the basis of time elapsed since the last attack, into three levels: (1) recent post Ménière's disease attack, when the patient reported a typical vertigo attack in the previous week (23 patients); (2) late post Ménière's disease attack, when the patient reported a typical vertigo attack in the previous two months (26 patients); and (3) inactive Ménière's disease, when the patient reported no attack during the last two months (four patients).Reference Soto, Labella, Santos, del Rio, Lirola and Cabanas7
Differences in the three parameters, comparing the hearing stages and the Ménière's disease activity levels, were compared using one-way analysis of variance and the Kruskal–Wallis test. Correlation analysis was calculated to show the relation between time elapsed since the last attack and computerised dynamic posturography parameters. A p value <0.05 was considered to be statistically significant.
Results
Figure 1 shows mean scores for the computerised dynamic posturography parameters (i.e. condition five scores, condition six scores and vestibular ratio) in each hearing threshold group (i.e. stages one to three). Mean values for condition five score and vestibular ratio slightly increased from stage one to stage three, but this trend was not statistically significant (p > 0.05).
Fig. 1 Mean computerised dynamic posturography (CDP) scores in Ménière's disease patients grouped by hearing loss stage (p > 0.05). SOT5 = CDP assessment condition five; SOT6 = CDP assessment condition six; VEST = vestibular ratio.
Figure 2 plots the same computerised dynamic posturography parameters against the time elapsed since the last vertigo attack. These results suggest that two of the three parameters (condition five score and vestibular ratio) tended to increase with increasing time since last attack (p < 0.05).
Fig. 2 Logarithmic regression of (a) computerised dynamic posturography (CDP) assessment condition five (SOT5) score (r = 0.369, p = 0.007), (b) CDP assessment condition six (SOT6) score (r = −0.036, p = 0.7) and (c) vestibular ratio (VEST) score (r = 0.362, p = 0.008), plotted against time elapsed since last Ménière's disease attack.
Figure 3 shows mean scores for the same three parameters for each Ménière's disease activity level group (i.e. recent post-attack, late post-attack and inactive). Mean scores for all three parameters were lower in the recent post-attack groups compared with the late post-attack group, and lower in the late post-attack group compared with the inactive group (except condition six score), but this variation was not statistically significant (p > 0.05).
Fig. 3 Mean computerised dynamic posturography (CDP) scores (for CDP assessment condition five score (SOT5), CDP assessment condition six score (SOT6) and vestibular ratio (VEST)) in patients grouped by Ménière's disease (MD) activity level stage; p > 0.05.
Discussion
Ménière's disease is a cochleo-vestibular disorder which affects both the auditory and balance systems. There are numerous proposals for classification of Ménière's disease based on hearing threshold and/or subjective symptoms of disequilibrium. At present, there are few published studies reporting the use of objective balance tests such as computerised dynamic posturography for Ménière's disease staging. Although several reports have evaluated the value of posturography in Ménière's disease,Reference Morrison, Hawken, Kennard and Kenyon9–Reference Pyykkö, Aalto, Starck and Ishizaki12 Soto and colleagues' study was the first to use computerised dynamic posturography for objective staging of balance in Ménière's disease patients.Reference Soto, Labella, Santos, del Rio, Lirola and Cabanas7 In the present study, we aimed to undertake an objective, quantitative evaluation of the balance status of Ménière's disease patients, using computerised dynamic posturography.
Our results indicated slight, but statistically insignificant, correlations between dynamic posturography scores and audiometric hearing thresholds. Soto and colleagues' study showed moderate, statistically significant correlation between computerised dynamic posturography scores and audiometric hearing thresholds.Reference Soto, Labella, Santos, del Rio, Lirola and Cabanas7 This may have been due to their larger sample size (75 patients). Soto et al. observed that this correlation was particularly strong in patients with stage four hearing loss. We had no patient at this stage.
Logarithmic regression of posturographic scores (except condition six scores) showed significant correlation with time elapsed since last Ménière's disease attack (p < 0.05) and computerised dynamic posturography scores increase within 60 days. These findings were similar to the results of Soto and colleagues' study.Reference Soto, Labella, Santos, del Rio, Lirola and Cabanas7 Both studies suggest that maximum posturographic recovery is reached within approximately two months of an attack.
Soto et al. suggested that Ménière's disease activity level could be classified as recent post-attack (i.e. up to one week post-attack), late post-attack (between one week and two months post-attack) and inactive (i.e. more than two months post-attack). They found strongly significant differences in mean posturographic scores, comparing these three activity levels. Although our results showed an increasing trend in computerised dynamic posturography scores as patients progressed from recent post-attack to late and inactive stages, these differences were not statistically significant. This statistical insignificance appears to be due to the limited sample size of our study.
• Various classifications of Ménière's disease have been developed based upon either audiological or subjective balance criteria. There are few studies of the use of objective balance tests for Ménière's disease staging
• This study evaluated the use of computerised dynamic posturography as an objective, quantitative balance test for staging Ménière's disease patients
• Results showed an increasing trend in computerised dynamic posturography scores from recent post-attack to late and inactive stages, but these differences were not statistically significant
• Further studies with larger sample sizes are required in order to investigate the clinical use of this objective staging system for Ménière's disease
We recommend that further studies be performed with larger sample sizes, in order to evaluate the value of computerised dynamic posturography for functional balance staging and monitoring of patients with Ménière's disease. The classification of Ménière's disease into three stages according to activity level, as proposed by Soto and colleagues, was useful in our patients, although not resulting in statistically significant differences (perhaps due to our limited sample size).
Acknowledgement
We thank Dr Nader Sadigh and Mrs. Shahrzad Ziae Ahmadi for his review and advice on preparing this manuscript.
