Clinical Otosclerosis, Prevalence Estimates and Spontaneous Progress

Transcription

1 Acta Otolaryngol (Stockh) 1999; 119: Clinical Otosclerosis, Prevalence Estimates and Spontaneous Progress YUKIMI SAKIHARA and AGNETE PARVING From the Department of Audiology, Bispebjerg Hospital H:S, Copenhagen, Denmark Sakihara Y, Parving A. Clinical otosclerosis, pre alence estimates and spontaneous progress. Acta Otolaryngol (Stockh) 1999; 119: This retrospective study was performed in order to estimate the prevalence of clinical otosclerosis as a function of age and gender, characterize the hearing level in otosclerosis and describe the spontaneous progress of the disease as a function of age. Clinically based samples were obtained from an audiological department, including 556 subjects: n=166 (30%) males and n=390 (70%) females with a median age of 75 years (range years) at the time of examination. A subdivision of the sample into 3 age-bands, years (n=39), years (n=78) and 60 years (n=439), was performed. The overall prevalence estimate of clinical otosclerosis in the area in question was 1.41/1,000 [95% confidence interval (CI)= /1,000] with an estimate of 0.9/1,000 (95% CI= /1,000) in males and 1.85/1,000 (95% CI= /1,000) in females, with an increase in the prevalence as a function of age from 0.22/1,000 (95% CI= /1,000) to 3.53/1,000 (95% CI= /1,000) in the elderly. The estimates should be considered underestimates, as not all clinical otosclerosis in the area was included. No significant differences in the better and worse ear hearing levels averaged across khz (BEHL/WEHL khz) were found as a function of gender, and in general the impairment in the BEHL khz was fairly moderate until the age of 60 years. In ears previously subjected to surgery a significantly better hearing level of median 63 db (range db) was found than in the no-surgery ears, with a median hearing level of 71 db (range db). An estimate of the progress showed an increment in the hearing level in the elderly 60 years of 30 db over 30 years in non-operated ears, fairly similar to the 40 db progress in operated ears. It was concluded that a significantly higher prevalence of clinical otosclerosis is present in females than in males, that the overall hearing level in otosclerosis is fairly moderate until the age of 60 years, and that previously operated ears have significantly better hearing than non-operated ears. Key words: clinical otosclerosis, pre alence estimates, spontaneous progress. INTRODUCTION Population-based studies on hereditary hearing impairment in adults are extremely difficult, not to say impossible, owing to the lack of specific audiometric measurements indicating a genetic hearing disorder. In addition, hearing disorders are often the result of various both intrinsic and extrinsic factors, which act in combination, making a strict aetiological classification of inheritance highly difficult. Owing to these difficulties the prevalence of hereditary hearing impairment in adults is virtually unknown, although estimates have been made based on the genetic expression of loci for hearing impairment in various populations and samples (1). As part of an EU-concerted action on genetic hearing impairment an attempt was made to provide some prevalence estimates based on approximately 28,000 examinations in an audiological department. From this clinically based study a prevalence of hereditary hearing impairment (HHI) in adults was estimated as 3.2/1,000 [95% confidence interval (CI): ], with a significantly higher prevalence of 4.1/1,000 (95% CI: ) in females than that of 2.1/1,000 (95% CI: ) in males. The prevalence estimates should be interpreted with caution, as they are clinically based and underestimates are likely (2). Although various factors such as genetic, infectious and hormonal aspects have been suggested as the underlying mechanism for otosclerosis (3 9), this hearing disorder is usually recognized as an inherited disease, characterized by a dominant mode of transmission with reduced penetrance (3, 10). The genetic premise has recently been supported by the localization of a candidate gene, aggrecan, to chromosome 15q25 q26 in an Indian family (11). In a population-based study an estimated prevalence of presumed otosclerosis was 2.1%, with a higher prevalence in those above the age of 40 years, but without significant differences as a function of gender (12). The prevalence of clinical otosclerosis in Caucasian populations varies considerably, from 0.04 to 1% (9, 10), which may be ascribed to, for instance, differences in the diagnostic criteria of otosclerosis, differences in the cohorts and sampling of subjects, and true differences as a function of country. In the previously reported study a prevalence of 2% of clinical otosclerosis was found, with a proportion of 44% among HHI in adults, in an audiological department (2). The objectives of the present retrospective study are to estimate further the prevalence of clinical otosclerosis as a function of age and gender, characterize the hearing level in otosclerosis and describe the spontaneous progress of the disease as a function of age Scandinavian University Press. ISSN

2 MATERIALS AND METHODS All records concerning adult patients examined within a 5-year period in the department, , comprising 27,692 subjects, were scrutinized by one of the authors (AP). A 5-year period was chosen because subjects living in the local area provided with a hearing instrument are supposed to have it replaced within a 5-year period, and thus it may be anticipated that the sampling of subjects with at least a severe to profound HHI is fairly complete. From the records all subjects suffering from any hearing impairment in the right or left ear [i.e. 2 or more pure-tone frequencies 20 db HL and/or individual pure-tone levels deviating, when compared with internationally agreed standards (13)] caused by genetic factors and living in the Copenhagen City at the time of the examinations were sampled and included in the study. Among the 1,265 subjects sampled, 44% suffered from otosclerosis, with the diagnosis based on several of the following criteria: family history of otosclerosis, normal otoscopy, an air-bone gap 15 db HL averaged across khz, absent stapedial reflexes, Gellé s test showing no intensity variations, normal tympanometry, history of previous surgery for otosclerosis, knowledge on a previous air-bone gap, and speech recognition score better than predicted from the air-conduction threshold in pure sensorineural hearing impairment in the elderly. Owing to the retrospective nature of this study, no information was available on the surgical procedure or the age at which the surgery was performed. However, considering the surgical procedures used throughout the 1960 s and onwards in Denmark, it is reasonable to believe that the majority had performed stapedectomy. It should be mentioned that information on the number of subjects operated for otosclerosis at the ear, nose and throat (ENT) departments in the Copenhagen health authority area was not available because of organizational changes. Among the 556 subjects, 166 (30%) were males and 390 (70%) were females, with a median age of 75 Clinical otosclerosis, pre alence and progress 469 years (range years) at the time of examination. From the records the pure-tone air-conduction thresholds in the right and left ear at the frequency range khz were noted and registered in a database. In those who had been examined several times, the most recent audiogram was noted. It should be mentioned that during the 5-year period there were no changes in the procedures for calibration of the equipment (13) or for the measurements of the hearing thresholds (14). Data analysis The overall prevalence of otosclerosis was estimated using the local annual population statistics from 1989 (15). In order to evaluate the prevalence and hearing levels as a function of age, the sample was subdivided into 3 age-bands: years (n=39), years (n=78) and 60 years (n=439). The prevalence was then estimated for males and females based on the age- and gender-matched background population. To describe the hearing levels in subjects, the better and worse ear thresholds (BEHL/WEHL) averaged across khz analysed according to the 3 agebands in males and females were used. Since some subjects had been operated in either the right/left or both ears, a comparative analysis concerning hearing levels averaged across khz in the operated (n=276) and non-operated ears (n=653) was performed in all ears and subdivided into age-bands. For statistical descriptive purposes 95% CI were indicated, and the Mann Whitney test was used for comparative analysis, with a 5% level indicating significance. RESULTS Estimated pre alence of otosclerosis The overall prevalence estimate of clinical otosclerosis in the area in question was 1.41/1,000 (n=556/ 395,048; 95% CI= /1,000) with an estimate of 0.9/1,000 (n=166/183,843; 95% CI= /1,000) in males and l.85/1,000 (n=390/211,205; 95% CI= Table I. Pre alence estimates of clinical otosclerosis in arious age-bands and as a function of gender Population Otosclerosis (n=556) Prevalence (n=1/1,000) Age (years) Males Females Total Males Females Overall Males Females ,392 86, ,629 45, ,822 79, ,642 28, ,450 31, ,214 17, ,

3 470 Y. Sakihara and A. Par ing Table II. BEHL/WEHL khz in subjects suffering from otosclerosis, o erall and subdi ided according to age-bands BEHL khz (db) WEHL khz (db) Age (years) M F M F Overall 53 (5 114) 56 (0 119) NS 76 (18 120) 75 (14 120) NS (5 76) 23 (0 49) S 66 (33 110) 43 (19 118) S (10 93) 34 (4 78) NS 55 (18 93) 49 (14 108) NS (13 114) 61 (3 119) NS 84 (40 120) 79 (25 120) NS Data are shown as median (range). M, male; F, female; S, significant; NS, non-significant /1,000) in females, being significantly higher in females. Table I shows an outline of the background population, the number of subjects in each of the 3 age-bands suffering from otosclerosis and the estimated prevalence of otosclerosis as a function of age and gender, demonstrating a preponderance in females and an increase as a function of age from 0.22/1,000 (n=39/176,629; 95% CI= / 1,000) in the youngest age-band to 3.53/1,000 (n= 439/124,230; 95% CI= /1,000) in the elderly. As 79% (n=439) of the sample was 60 years old, additional information concerning the prevalence of clinical otosclerosis in the elderly 60 years of age was estimated by further subdivision into 10-year age-bands. This showed an increasing prevalence from 2.1/1,000 in those aged years to 6.9/1,000 in those 90 years of age. Better and worse ear hearing le els a eraged across khz The median hearing level and range is shown according to age-bands for the BEHL/WEHL khz in Table II. In each age-band males had poorer hearing than females; however, a significant difference as a function of gender was found only in the age-band years. In general, the impairment in the BEHL khz, which is essential for social hearing, is fairly moderate until the age of 60 years in both males and females. Above this age the hearing impairment increases substantially for both the BEHL khz and WEHL khz in both genders. Hearing le els in the surgery/no-surgery ears In order to evaluate the spontaneous progress of otosclerosis the hearing levels averaged across khz in 653 non-operated ears were analysed according to age-bands. In addition, the hearing levels in the non-operated ears were compared with the levels in the 276 operated ears, as it may be anticipated that an operation for otosclerosis maintains an improved hearing level longitudinally. Table III shows the hearing levels in the operated and non-operated (i.e. spontaneous progress) ears according to age-bands. A substantial longitudinal progress in the hearing level from median 41 to 79 db was found from the youngest age-band of years to those 60 years of age, and a significantly better hearing level was found in operated compared with non-operated ears in those aged 60 years. (The age-band years was omitted owing to the small number, n=9, of operated ears in this ageband.) In order to obtain more detailed information on spontaneous longitudinal hearing levels in the elderly, the non-operated ears representing the sample of subjects aged 60 years, i.e. n=439 (115 males and 324 females), at a median age of 77 years (range 60 95) at the time of examination were subdivided into 5-year age-bands and the hearing levels were analysed in each ear (n=502) as a function of age and gender (Table IV). DISCUSSION This study shows that the estimated overall prevalence of clinical otosclerosis in Copenhagen city is 1.41/1,000, with twice as many females as males. The prevalence increased as a function of age. A fairly moderate ( 40 db) median BEHL khz was maintained until the age of 60 years in both genders. Above the age of 60 years the BEHL khz was Table III. Hearing le els (db) a eraged across khz in the surgery/no-surgery ears Age (years) Surgery (n=276) No-surgery (n=653) Overall 63 (24 119) 71 (5 120) S (18 110) (24 95) 44 (5 105) NS (26 119) 79 (23 120) S Data are shown as median (range). S, significant; NS, non-significant.

4 Table IV. Hearing le els (db) a eraged across khz in the no-surgery ears in the subjects aged 60 years Age (years) Male Female Overall 80 (23 120) 78 (25 120) NS (34 101) 64 (30 113) NS (23 120) 60 (26 101) S (30 109) 70 (25 118) NS (34 115) 81 (40 119) NS (50 118) 79 (34 120) NS (79 106) 84 (43 110) NS (84 96) 95 (80 119) NS (86 111) Data are shown as median (range). S, significant; NS, non-significant. 60 db, and deteriorated additionally as a function of age. The hearing in the operated ears was better in those above 60 years of age compared with the hearing as a result of spontaneous progress. Since the present study was based on a clinical sample from an audiological department, the estimated prevalence may represent substantial underestimates, specifically in the younger age groups, as these patients may suffer from otosclerosis and be examined/operated at an ENT department or by the local ENT physician, and thus not be included in the present sample. In addition, the hearing loss in the younger subjects ( 60 years) suffering from otosclerosis may be moderate, resulting in limited hearing problems, and these subjects may be found only in a population-based study with appropriate air- and bone-conduction testing. The theory that the present prevalence represents an underestimate is supported by the prevalence of 2.1% of population-based presumptive otosclerosis reported by Browning & Gatehouse (12). The overall estimated prevalence of clinical otosclerosis of 1.4/1,000 found in the present study is fairly compatible with previous estimates in Caucasian populations, varying from 0.04 to 0.6% (10), and a mean estimate of clinical otosclerosis of 3/1,000 has been suggested (16). The increase in the prevalence as a function of age in this study contrasts somewhat with previous studies, showing the highest prevalence of clinical otosclerosis in the age range years of age and exhibiting a substantial decrease above the age of 70 years (9). In addition, the high prevalence of 6.1/1, /1,000 in the elderly, 80 years of age, in this report seems to be the highest figure reported, which probably reflects the differences in samples derived from ENT departments and an audiological department. This should be taken into account when estimating prevalences of clinical Clinical otosclerosis, pre alence and progress 471 otosclerosis, and has consequences for the provision of hearing services to the elderly. Irrespective of the varying estimates concerning the prevalence of otosclerosis, the significant preponderance in females found in clinical samples is supported by the present study. The female preponderance, especially demonstrated in the elderly aged 60 years, has been ascribed to hormonal differences as a function of gender (8, 9). Moreover, it has been suggested that genetic factors such as polyhybrid inheritance may account for the gender differences (10, 17); however, the lack of gender differences in the prevalence of otosclerosis found on a population basis is contradictory (12). The poorer hearing in females, for which help is sought, might explain the female preponderance in the prevalence of clinical otosclerosis; however, the present data cannot support this suggestion, although females were likely to have a more severe affliction than males in the population study (12). The data analysis concerning BEHL/WEHL khz as a function of gender showed no major differences; however, a deterioration as a function of age in the present study was obvious (Table II). Although the BEHL khz was moderate (median 40 and 34 db in males and females, respectively), a hearing disability demanding rehabilitation was present even in the few younger subjects included in this study. On a group basis the median BEHL/WEHL khz in the age-group 60 years represent a severe hearing impairment and considering the age of the subjects (median 77 years) audiological rehabilitation should be offered, although some individuals may benefit from surgical intervention at this high age (18). Although discrepancies in the data analysis exist, the deterioration in the hearing levels according to age in this study is fairly compatible with the findings by Gristwood (9) in the age group 60 years of age, although the hearing levels in the younger subjects were significantly worse than the present, probably reflecting the differences in the sample selections. The younger severely hearing-impaired sought surgical help in an ENT department, whereas the elderly sought audiological rehabilitation. The comparison of the hearing levels averaged across khz between the operated and non-operated ears showed significantly better hearing in the operated ears (Table III), and although no information on age at and type of surgery was available, the present findings indicate that the elderly aged 60 years have benefited from previous surgery as a group, although the majority may still need hearing instruments.

5 472 Y. Sakihara and A. Par ing In a previous true longitudinal study of 296 operated ears tested 1 12 years postoperatively an increment in the hearing level of 1 db/year was ascribed to normal ageing in the high-frequency area, whereas an increment of about 2 db/year was found in non-operated ears. The deterioration was ascribed to both progress in stapes fixation and ageing (19). The discrepancy in the age distribution, i.e. 28% 60 years reported by Nilsson (19) and 79% (n=439) 60 years in the present study, may explain why the increment in the hearing level averaged across khz of about 30 db in 30 years in non-operated ears (Table IV) was less pronounced in the present study, or may be ascribed to differences in the data analysis and/or study designs. In the operated ears (n=230) in those aged 60 years the longitudinal deterioration was 40 db hearing level from the ageband to the age-band, which is fairly similar to the deterioration in the non-operated ears. The factor(s) causing the deterioration in the hearing level as a function of age cannot be further evaluated in the present study; however, genetic factors are likely to cause the increment, as these factors are considered to be of major importance in both otosclerosis and presbyacusis (11, 20). In conclusion, the prevalence of clinical otosclerosis, based on a sample from an audiological department, showed an increase as a function of age, reaching its peak in the age group above 80 years. A significant deterioration in the hearing level was found as a function of age, showing a severe hearing impairment in those aged 60 years; in the elderly the hearing level was better in the operated than in the non-operated ears, which implies that subjects may benefit from an operation before reaching an advanced age. ACKNOWLEDGMENTS We gratefully acknowledge the meticulous support with the data analysis performed by audiotechnician Birger Christensen. REFERENCES 1. Morton NE. Genetic epidemiology of hearing impairment. In: Ruben RJ, Van der Water T, Steele KP, eds. Genetics of hearing impairment. Ann N Y Acad Sci 1991; 613: Sakihara Y, Christensen B, Parving A. The prevalence of hereditary hearing impairment in adults. J Audiol Med 1998; submitted. 3. Gordon MA. The genetics of otosclerosis. A review. Am J Otol 1989; 10: Reardon W. Genetic deafness. J Med Gen 1992; 29: Arnold W, Friedmann I. Immunohistochemistry of otosclerosis. Acta Otolaryngol (Stockh) 1990; Suppl 470: Arnold W, Niedermeyer HP, Lehn N, Neubert W, Höfler H. Measles virus in otosclerosis and the specific immune response of the inner ear. Acta Otolaryngol (Stockh) 1996; 116: McKenna MJ, Kristiansen AG, Haines J. Polymerase chain reaction amplification of a measles virus sequence from human temporal bone sections with active otosclerosis. Am J Otol 1996; 17: Gristwood RE, Venables WN. Otosclerosis in South Australia. Clin Otolaryngol 1984; 9: Gristwood RE. Otosclerosis (otospongiosis): general considerations. In: Alberti PW, Ruben RJ, eds. Otologic medicine and surgery. New York: Churchill-Livingstone, 1988: Declau F, van der Heyning P. Otosclerosis. In: Martini A, Read A, Stephens D, eds. Genetics and hearing impairment. London: Whurr, 1996: Tomek MS, Brown MR, Mani SR, et al. Localization of a gene for otosclerosis to chromosome 15q25 q26. Hum Mol Gen 1998; 7/2: Browning GG, Gatehouse S. The prevalence of middle ear disease in the adult British population. Clin Otolaryngol 1992; 17: International Organization for Standardization. ISO 389: Acoustics Standard reference zero for the calibration of pure tone audiometers. Geneva, Switzerland, International Organization for Standardization. ISO 8253/1. Acoustics Audiometric test methods. Part 1: Basic pure tone air and bone conduction threshold audiometry. Geneva, Switzerland, Statistical Year Book of Copenhagen 1989: Copenhagen: Copenhagen Statistical Office, Causse JR, Causse JB. Otospongiosis as a genetic disease. Am J Otol 1984; 5: Hernandez-Orozco F, Courtney GT. Genetic aspects of clinical otosclerosis. Ann Otol Rhinol Laryngol 1964; 73: Ghonim MR, El-Degwy AA, El-Sharabasy AE. Stapedotomy in the profoundly deaf. Oto Rhino Laryngol 1997; 59: Nilsson G. Longterm results after stapedectomy. Acta Otolaryngol 1977; 84: Willott JF. Aging and the auditory system: anatomy, physiology, and psychophysics. London: Whurr, Submitted May 29, 1998; accepted December 23, 1998 Address for correspondence: Agnete Parving Department of Audiology Bispebjerg Hospital H:S DK-2400 Copenhagen NV Denmark Tel: Fax: bbhaudap@pip.dknet.dk