|Ménière's disease or Ménière's syndrome can be a disabling condition where hearing loss, tinnitus, vertigo and a feeling of fullness in the ear come together in various proportions and extents.
this month has examined the evidence concerning the incidence and prevalence of the disease; we have also run a search on MEDLINE for randomised controlled studies (RCTs) of treatments published since 1990 (using the terms Ménière and random*). The only results have been for betahistine treatment, and these have been expanded by RCTs of betahistine trials in the 1980s.
What is Ménière's Disease?The constant pathological finding in Ménière's disease is a progressive distension of the endolymphatic space of the inner ear. It appears that fluid build-up in the endolymphatic space (endolymphatic hydrops), caused either by overproduction or reduced absorption, exposes hair cells responsible for sensing movement and balance to progressive damage or paralysis. The result of that paralysis is that the person has an attack of dizziness, often with nausea and vomiting.
Early on these attacks can be short, as the damage to the hair cells is temporary and they can begin to function normally when the hydrops resolves. Repetitive insults lead to irreversible changes as the hair cells die; when dead they do not regrow, and hearing loss, in particular, can become permanent. There is an excellent recent review of Ménière's disease .
What causes Ménière's Disease?The simple answer is that there is no simple answer. A review of over 300 articles published between 1983 and 1989  covers many papers which look at the aetiology of Ménière's disease. One `theme' in the reports on aetiology and pathology seems to be an increase in immunoglobulins both in the endolymphatic fluid and in serum, occurring in a high proportion of patients with Ménière's disease. While it is likely to be a multifactorial condition, some sort of underlying infection (probably viral) or auto-immune component seems likely in patients with this condition.
How is Ménière's Disease diagnosed?The usual clinical diagnosis is the triad of symptoms of vertigo, hearing loss and tinnitus, often with a feeling of fullness of the ear. The tinnitus and hearing loss are usually unilateral, and the disease at onset is often mild, with attacks lasting for an hour or so. In some patients, however, the onset can be much more serious, with trains of episodes resulting in significant nausea, vomiting and prostration.
Electrocochleography (ECOG) exposes the ear to a train of click or tonal stimuli recorded by an electrode situated close to the cochlea which records action potentials from the firing of auditory neurones. Endolymphatic hydrops results in a widening of the wave-form with multiple peaks that is quite different from the normal result [see 1]. This test is quite simple and can be carried out in most ENT centres.
How common is Ménière's Disease?There are a number of reports concerning the incidence and prevalence of Ménière's disease. The numbers vary quite considerably from study to study, and quite clearly different diagnostic criteria have been used at different times and in different places.
In Northern Ireland, Wilmot  estimated from experience over a 25-year period that the incidence was between 1 and 2 cases per 10,000 population per year (say 200 cases per million).
Stahle and colleagues  estimated the incidence of Ménière's disease in a Swedish population in 1973; this was possible because Ménière's disease was recorded on the health records of all inpatients with the diagnosis in Sweden. A computer analysis was conducted in the Uppsala and Skåne parts of Sweden, with a combined population of over 2 million. In 1973, the two regions with a joint population of 2,263,285 had 257 patients diagnosed with Ménière's disease (114 cases per million) of whom 60% were women, and most (228) cases were between 15 and 69 years.
Watanabe's  paper on the incidence of Ménière's disease in Japan also has some interesting historical data on incidence from the UK in the 1950s and 1960s; results from Oxford in the mid 1950s suggest an incidence of about 560 per million, though higher figures are also quoted.
In the 1970s nation-wide surveys on the incidence of Ménière's disease were carried out in Japan. Though they reveal much of interest about age and sex distributions, severity and other aspects of disease, the only national figure for incidence based on a one-day survey was 160 cases per million, though figures on a 1-week survey were much lower, at 35 per million.
In the United states, a very thorough study of the incidence and prevalence of Ménière's disease was carried out in the population of Rochester, Minnesota, using a centralised diagnostic index at the Mayo clinic. This study examined cases between 1951 and 1980 . The incidence of Ménière's disease in 1980 was 153 cases per million of population. Median ages of onset and diagnosis were 50 and 53 years respectively, although half the cases were diagnosed within 6 months of onset. This is a most important paper giving useful information about this difficult disease. The prevalence of the disease was 2,182 per million.
A recent report from Siena  and Latium from two hospitals serving the needs of 104,000 people, over 13 years found 111 cases, almost all being diagnosed between the ages of 10 and 70 years, with a distinct peak between 40 and 50 years. This gives an incidence of 82 cases per million. Interestingly, the incidence was some 3 times higher in hospital workers than the general population, and the authors suggested that this may reflect a higher rate of diagnosis - and that the figures for the general population may be reduced by under-diagnosis.
What is the natural history of Ménière's disease?There does not seem to be a clear answer to this question. In some patients the unilateral disease appears to "burn-out" with deafness remaining but with the vertigo and tinnitus declining. Other patients (the minority, but perhaps up to 25%) go on to develop a severe bilateral disorder where the vertigo remains - and where ablation of the inner ear becomes necessary.
Can lifestyle changes help?It is suggested that food allergy or excess caffeine, nicotine or alcohol may be in some way involved in the aetiology of Ménière's disease. This is far from being proven. It is suggested that avoidance of caffeine, etc., may help. Bandolier has found no evidence of that in any RCT or other study.
Are there surgical treatments?There are a number of surgical procedures , but Bandolier was not able to find randomised controlled trials of their effectiveness. We would be happy to report on such trials in future if any readers can bring them to our attention - computer literature searches are not infallible.
Are there medical treatments?Yes, there are a number, but the only RCTs that we could find referred to betahistine. The mechanism of action of betahistine appears to be by promoting better circulation in the microvasculature, leading to reduction in endolymphatic hydrops.
There are six RCTs with betahistine, and five involve treatment of Ménière's disease. These studies used different doses, and the outcome measures used were not usually dichotomous so a meta-analysis with numbers-needed-to-treat was not possible. Instead, Bandolier will give a brief report on each paper. The quality of the studies was generally high.
Wilmot & Menon, 1976 This was a randomised double-bind cross-over comparison of betahistine 24 mg daily and placebo for 8 or 12 weeks. Twenty-four patients began the study and analysis was performed on 22 - there were two patients with worsened symptoms who did not complete the treatments.
Patients maintained a daily symptom card for vertigo, tinnitus, deafness, fullness of ear and vomiting. Betahistine produced significantly less severe vertigo, tinnitus and fullness of ear than did placebo.
Frew & Menon, 1976 This RCT compared betahistine with placebo after a 4-week period on placebo followed by four 8-week periods with either betahistine 32 mg daily or placebo on two occasions each. Of 28 patients who began the treatment, 22 completed it, six being unable to co-operate, for reasons not given.
Symptom score cards were again used, and patient scoring showed significantly better results for vertigo and tinnitus.
Fischer and van Elferen, 1985 Patients were recruited from general practitioners lists. 83 patients were finally selected, who after one month on placebo were randomly allocated to betahistine 48 mg daily or placebo for three months. Ten patients did not complete the study for a variety of reasons.
A battery of outcome measures was used, including number of attacks per month, mean duration of attacks and intensity and accompanying symptoms.
Betahistine proved to be significantly better than placebo for number, duration and intensity of attacks, with significantly more patients asymptomatic in the last month of treatment with betahistine compared with placebo.
This paper is also important because the authors examined the effectiveness of treatment with regard to duration of Ménière's disease in a post-hoc analysis. They found that, with betahistine, the patients who had complete relief had duration of symptoms less than a third as long as those who still had symptoms. The message would seem to be start treatment early in the course of the disease for the best results.
Deering et al, 1986 Another general practice study involved 88 patients in a cross-over design with betahistine 72 mg daily compared with cinnarizine 90 mg daily for three months on each treatment. Forty-six patients completed the full six month period; the most common reason for discontinuing betahistine was feeling better.
The frequency of attacks fell from a pre-treatment mean of 9 attacks per month to about 5 attacks per month with betahistine, and the mean duration of attacks was about half or less with betahistine (down from 1.3 to 0.5 hours on average). Betahistine was significantly better than cinnarizine.
Fraysse et al, 1991 This GP study involved 55 patients treated for two months with betahistine 48 mg daily or flunarizine 10 mg daily in a parallel group design. There were six withdrawals for a variety of reasons.
Again a battery of outcome measures was used, including attack numbers, duration and intensity. Betahistine was better than flunarizine, and showed very large improvements in attacks over a period of two months, as shown below.
Betts et al, 1991 This RCT evaluated the effects of betahistine 216 mg daily, prochlorphenazine 15 mg and placebo, taken for three days, on driving skills in twelve volunteer subjects. The psychomotor effects of betahistine could not be distinguished from placebo.
In all the above studies, the incidence of adverse effects was low. There were 22 reports of adverse events in 19 patients on betahistine in study 11, where the dose of betahistine was somewhat higher than the 48 mg daily usually recommended, and drowsiness was the most frequent side-effect. The incidence of adverse events was low at 48 mg/day in study 12; drowsiness and gastrointestinal disorders occurred in 18% and 4% of patients respectively after two months of treatment.
ConclusionMénière's disease is not common, but an average GP practice with some 6000 patients would expect to see perhaps one new case per year. A Health Authority of 250,000 would see some 60 cases per year.
The disease seems to strike in middle years, most commonly between 30 and 60 years, and to affect men and women about equally. The reasons for the onset and pattern of the disease are not yet understood.
There is an effective diagnostic test that should be available at all ENT clinics, and at least one effective medical treatment. Betahistine costs about £235 a year to maintain a patient on 48 mg daily; this cost has to be set against the cost of not treating patients who, untreated, may have up to 20 attacks of long duration per month, and consume much GP and specialist clinic time.
next story in this issue