The burden of Ménière syndrome (MS) is substantial, especially when considering the significant impact on the quality of life of those affected. Reported estimates of incidence and prevalence have varied widely due to methodological differences between studies, changes in criteria for diagnosis of MS, and differences in populations studied. Reported prevalence rates for MS range from 3.5 per 100,000 to 513 per 100,000. A recent study using health claims data for more than 60 million patients in the United States found prevalence of 190 per 100,000 with a female:male ratio of 1.89:1. The prevalence of MS increases with increasing age.
Although Meniere’s syndrome (MS) has been recognized as a clinical entity for nearly 150 years since first being described by Prosper Ménière the epidemiology of the disorder is still not clearly defined. Reported prevalence rates of MS have varied widely, with estimates as low as 3.5 per 100,000 and as high as 513 per 100,000. The wide range of values are likely due to methodological differences, changes over time in criteria for the diagnosis of MS, difficulty in distinguishing MS from related conditions such as migraine-associated vertigo, and differences in the populations surveyed. The actual prevalence of MS may also be changing over time. In one report the investigators speculated that the stresses of modern society or even changes in diet have led to an increase in its occurrence over time, especially in the female population.
Published reports of the epidemiology of MS generally fall into 2 methodological categories: retrospective case series and population-based surveys. Most of the studies are retrospective series that start with known cases of MS identified from patient records for a given group of hospitals and clinics. The population served by the hospitals and clinics then serves as the dominator for calculating incidence and prevalence. This methodology introduces sampling bias in that patients in the population with the disease may not have been treated at the hospitals and clinics surveyed for various reasons. Population-based cross-sectional studies reduce sampling bias by surveying a random sample of the general population. The incidence and prevalence in the general population are inferred from the exact values in the sample group. Unfortunately, for disorders such as MS that are relatively rare at the population level, very large sample sizes are needed to achieve sufficient power to accurately estimate epidemiologic characteristics in population-based studies. Radtke and colleagues compared this to “searching for a needle in a haystack” with regard to MS.
Estimates of incidence
Incidence is defined as the number of new cases occurring over a specified period of time, usually 1 year. In 1954 Cawthorne and Hewlett attempted to estimate the incidence of MS by examining a register of clinical records for 8 clinical practices in Great Britain serving a population of 27,365 people; they arrived at an annual incidence of 157 per 100,000. As pointed out by Wladislavosky-Waserman and colleagues, this number most likely represents a combination of incidence and prevalence, as some patients may have had onset of symptoms in preceding years.
In 1973 Stahle and colleagues examined records from a standard, nationally administered records system to determine the incidence of MS in a patient population from 2 cities in Sweden; they found an annual incidence of 46 per 100,000.
Celestino and Ralli reviewed the records from 1973 to 1985 from a hospital and outpatient clinic serving a community of 103,797 people in Italy. The 1972 American Academy of Ophthalmology and Otolaryngology guidelines were applied for diagnosis of MS, and an incidence of 8.2 per 100,0000 per year was found.
Estimates of prevalence
Prevalence is defined as the proportion of individuals in a population having a disease. The most often-cited study examining prevalence of MS in the United States was performed by Wladislavosky-Waserman and colleagues. These investigators identified cases of MS by examining medical records from 1953 to 1980 for the Mayo Clinic and Olmstead Medical Group, the major health care providers for the 40,000 inhabitants of Rochester, Minnesota. A prevalence of 218 per 100,000 in 1980 was reported. As Celestino and Ralli pointed out, one-third of patients included in the Rochester study had recurrent vertigo without cochlear symptoms and would not meet current criteria for MS. Therefore, prevalence was likely overestimated. Also, the population studied was homogeneous relative to the current United States population; 99% of the subjects were white.
In 1980, Nakae and colleagues reported prevalence estimates of MS based on the results of 2 nationwide surveys in Japan. The first involved a random sample of 811 hospitals and 729 clinics, and the prevalence of MS was 73 per 100,000. In the second study, all university and general hospitals in Japan (a total of 190 hospitals) were surveyed, and the prevalence was 3.5 per 100,000. It appears this number is lower than expected due to exclusion of outpatient clinics. Later surveys of more limited geographic regions in Japan, reported by Tokumasu and colleagues and Watanabe, yielded prevalence rates of approximately 17 per 100,000. More recently, Shojaku and colleagues examined medical records from 3 hospitals in central Japan from 1980 to 2004 and found an overall prevalence of 34.5 per 100,000.
In their study of incidence of MS in Italy, Celestino and Ralli calculated prevalence by multiplying the annual incidence by a conversion factor taking into account estimated life expectancy and average age of onset of MS, as previously described by Arenberg and colleagues; they arrived at a prevalence 205 per 100,000.
Havia and colleagues surveyed a random sample of 5000 people in Southern Finland and found a prevalence of 513 per 100,000, considerably higher than all other published results. Radtke and colleagues questioned the validity of the survey used in the Havia study. These investigators suggested it may not have sufficiently discriminated vestibular from nonvestibular vertigo and did not appropriately quantify duration of attacks, leading to an overestimation of prevalence. Radtke and colleagues themselves performed a survey of a random sample of 4869 people in Germany. Subjects were screened for moderate or severe dizziness or vertigo. If present, a more thorough telephone interview was performed to assess for MS using a stepwise application of the American Academy of Otolaryngology (AAO) 1995 criteria. Using this strict approach, they found a prevalence of 120 per 100,000.
To further define the current prevalence of MS in the general United States population, the authors recently analyzed data from a large health claims database containing information for over 60 million patients. The database comprises fully adjudicated medical and pharmaceutical claims for over 60 million unique patients from over 97 health plans across the United States (almost 25% of the entire United States insured population). Patients in the database are representative of the national, commercially insured population on a variety of demographic measures including age, gender, health plan type, and geographic location. For the 3 years from 2005 to 2007, the prevalence among the entire United States insured population was estimated to be 190 per 100,000.
Estimates of prevalence
Prevalence is defined as the proportion of individuals in a population having a disease. The most often-cited study examining prevalence of MS in the United States was performed by Wladislavosky-Waserman and colleagues. These investigators identified cases of MS by examining medical records from 1953 to 1980 for the Mayo Clinic and Olmstead Medical Group, the major health care providers for the 40,000 inhabitants of Rochester, Minnesota. A prevalence of 218 per 100,000 in 1980 was reported. As Celestino and Ralli pointed out, one-third of patients included in the Rochester study had recurrent vertigo without cochlear symptoms and would not meet current criteria for MS. Therefore, prevalence was likely overestimated. Also, the population studied was homogeneous relative to the current United States population; 99% of the subjects were white.
In 1980, Nakae and colleagues reported prevalence estimates of MS based on the results of 2 nationwide surveys in Japan. The first involved a random sample of 811 hospitals and 729 clinics, and the prevalence of MS was 73 per 100,000. In the second study, all university and general hospitals in Japan (a total of 190 hospitals) were surveyed, and the prevalence was 3.5 per 100,000. It appears this number is lower than expected due to exclusion of outpatient clinics. Later surveys of more limited geographic regions in Japan, reported by Tokumasu and colleagues and Watanabe, yielded prevalence rates of approximately 17 per 100,000. More recently, Shojaku and colleagues examined medical records from 3 hospitals in central Japan from 1980 to 2004 and found an overall prevalence of 34.5 per 100,000.
In their study of incidence of MS in Italy, Celestino and Ralli calculated prevalence by multiplying the annual incidence by a conversion factor taking into account estimated life expectancy and average age of onset of MS, as previously described by Arenberg and colleagues; they arrived at a prevalence 205 per 100,000.
Havia and colleagues surveyed a random sample of 5000 people in Southern Finland and found a prevalence of 513 per 100,000, considerably higher than all other published results. Radtke and colleagues questioned the validity of the survey used in the Havia study. These investigators suggested it may not have sufficiently discriminated vestibular from nonvestibular vertigo and did not appropriately quantify duration of attacks, leading to an overestimation of prevalence. Radtke and colleagues themselves performed a survey of a random sample of 4869 people in Germany. Subjects were screened for moderate or severe dizziness or vertigo. If present, a more thorough telephone interview was performed to assess for MS using a stepwise application of the American Academy of Otolaryngology (AAO) 1995 criteria. Using this strict approach, they found a prevalence of 120 per 100,000.
To further define the current prevalence of MS in the general United States population, the authors recently analyzed data from a large health claims database containing information for over 60 million patients. The database comprises fully adjudicated medical and pharmaceutical claims for over 60 million unique patients from over 97 health plans across the United States (almost 25% of the entire United States insured population). Patients in the database are representative of the national, commercially insured population on a variety of demographic measures including age, gender, health plan type, and geographic location. For the 3 years from 2005 to 2007, the prevalence among the entire United States insured population was estimated to be 190 per 100,000.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
