Abstract
Objective
To analyse the epidemiology of otosclerosis in a British cohort collected between 2011 and 2017.
Design
Retrospective cohort study.
Setting
Five UK ENT Departments.
Patients
Patients with surgically confirmed otosclerosis.
Main Outcome Measures
Questionnaire data documented family history of otosclerosis, age of onset, medical history and information on associated risk factors for 657 patients. Pre and post-surgical pure-tone audiometry was collected for 154 of these patients.
Results
The age of onset, incidence of bilateral disease, tinnitus and vertigo, a higher prevalence of women (65%) than men (35%) are similar to those reported previously for otosclerosis cohorts. No association with measles infection was detected. Patients with a family history (40%) have an earlier age of onset and a higher incidence of bilateral disease and vertigo than non-familial subjects. Pedigree analysis is consistent with an autosomal dominant inheritance with reduced penetrance being apparent in 44/91 pedigrees studied. Women who associate their hearing loss with pregnancy have an earlier age of onset than those that do not (p=6x10-6).
Conclusions
This study confirms that otosclerosis is an early adult onset disease that is more prevalent in women than men with a large minority of patients having a family history of otosclerosis. We report new evidence to support a relationship between pregnancy and otosclerosis progression in a proportion of women. In addition, this is the first study to identify differences in severity between familial and non-familial cases of otosclerosis, highlighting the possibility that more than one aetiology may be involved.
Keywords: Otosclerosis, Middle Ear, Bone Remodeling, Hearing Loss, Epidemiology, Stapes Surgery
Introduction
Otosclerosis is a common form of adult-onset hearing impairment, typically characterised by disordered bone remodelling in the otic capsule that leads to a progressive conductive hearing loss. The remodelling process often involves the stapedio-vestibular interface and can lead to fixation of the stapedial footplate. Patients often choose to undergo surgical treatment, by means of stapedectomy or stapedotomy. Both techniques are successful in improving the hearing loss to varying degrees1. Although many patients with otosclerosis select surgical treatment, use of hearing aids to manage the hearing loss are advised prior to surgery and are an alternative for patients who are not candidates for stapes surgery2–4.
Clinical otosclerosis is relatively common in White Europeans with a reported frequency of 0.1-2.1%, and is also common among individuals of Indian extraction5–9. The age of onset is variable, although hearing loss typically begins in the third decade, with a range from first decade to the sixth decade5,7,10. In most cases, hearing loss is bilateral (70-85%) and is usually asymmetrical, developing initially in one ear before progressing to the other1,11–13. Otosclerosis has traditionally been regarded as a middle-ear disease but the inner ear can also be affected, with mixed or pure sensorineural hearing loss, tinnitus and sometimes vertigo14,15.
Otosclerosis is considered a complex disease with both genetic and environmental factors that can occur in isolated cases with no family history (non-familial), or in cases with a strong familial inheritance pattern consistent with a monogenic cause (familial). A positive family history of otosclerosis has been previously reported in between 30-70% of cases5,10,11,16,17. This wide range probably reflects a recruitment bias in these cohorts and variable ascertainment of relatives of the proband. The pattern of inheritance in familial otosclerosis is most often consistent with an autosomal dominant mutation that exhibits variable penetrance. Although reduced penetrance is common, the degree of penetrance can vary greatly between families and has been reported between 40-90%5,10,18–23. To date, otosclerosis has proved resistant to analysis by conventional genetic techniques such as linkage analysis. In 2009, a genome-wide association study identified an association between RELN and otosclerosis24, however its biological role remains unclear. Recently, mutations and altered expression of SERPINF1 have been identified in patients with familial otosclerosis25.
In addition to the genetic component of otosclerosis, various other aetiological factors have been suggested as playing a role in the pathogenesis of otosclerosis over the years. Clinical otosclerosis is consistently reported as more prevalent in females than in males, giving rise to the hypothesis that sex hormones may contribute to the development of disease26–29. Other factors which have been associated with otosclerosis incidence include measles-virus infection30–36, fluoride in drinking water37 and co-incidence with certain systemic connective tissue disorders38,39. However, despite its long history, many questions as to the aetiology of this condition remain unclear. One reason for this is a scarcity of empirical data on the epidemiology of otosclerosis in large well characterised cohorts. Although there have been several reviews of otosclerosis literature in recent decades which have included reiterations of the established epidemiology, these are largely based on a relatively limited number of studies some of which date back as far as the 1950s5,10,16,18. There is therefore a need for a new analysis of otosclerosis epidemiology in sizeable cohorts to confirm or refute the conclusions inherited from much earlier studies.
The aim of this study was to report the epidemiological findings from a British cohort recently recruited for a genetic study of otosclerosis25 and examine factors such as age of onset, pattern of inheritance, hearing loss symptoms, as well as investigate the differences between the various demographic sub-groups of the cohort including gender, familial inheritance and pregnancy characteristics to reveal any potentially distinct pathological mechanisms.
Methods
Patient Recruitment
Between 2011 and 2017 individuals with a confirmed diagnosis of otosclerosis were recruited from the Royal National Throat Nose and Ear Hospital (London), Princess Margaret Hospital (Windsor), Sunderland Royal Hospital (Sunderland), Freeman Hospital (Newcastle upon Tyne) and Ninewells Hospital (Dundee). The diagnosis of otosclerosis was made on clinical and audiometric examination, and then confirmed during surgery. Two recruitment pipelines were used, one via individuals attending for surgical treatment and another involved a retrospective recruitment of past surgical patients from the same centres. The study was approved by the London Bloomsbury NRES Ethics committee (11/LO/0489) and patients were recruited by informed consent. Patient reported outcomes were recorded using a structured questionnaire regarding medical history (see Supplemental Digital Content 1). All female subjects answered a second questionnaire detailing history of pregnancies and hormonal medication use (see Supplemental Digital Content 2).
Family Pedigrees
In 96 patients who reported a strong family history of otosclerosis a family pedigree was constructed based on further investigation with the proband and recruitment of other family members to the study. We investigated at least three generations per family and following detailed assessment, a pattern of inheritance was assigned for each pedigree. The number of affected parents of probands was used to estimate the penetrance in otosclerosis.
Audiometry
Pure-tone audiometry was performed not earlier than 3 months before surgery and 3 months after. Air-conduction (AC) and bone-conduction (BC) included frequencies 0.5, 1, 2, 3 and 4 kHz. The guidelines from the Committee on Hearing and Equilibrium regarding the four-tone pure-tone average (0.5, 1, 2 and 3 kHz) were followed40. Pure-tone average (PTA) values (0.5, 1, 2 and 3 kHz) were calculated regarding AC and BC thresholds, as well as the air-bone gap (ABG). We used the parameters of successful surgery as: post-operative ABG ≤10 dB; AC improvement ≥20 dB; or BC not worsened by ≥5 dB41.
Data Analysis
Univariate analyses were performed on questionnaire data, comparing different demographics using the Chi-sq test. Audiometric data were analysed to establish normal distribution and Student's t-tests were conducted to establish if data were significant. Results with values of p<0.05 were considered statistically significant. SigmaPlot 11.0 software was used to perform all statistical analysis.
Results & Discussion
Characteristics of the study cohort
A total of 1025 subjects with a confirmed diagnosis of otosclerosis were invited to join the study: 368 either declined to take part or did not respond to a written invitation (36%). Questionnaire data was obtained from a total of 657 subjects and the cohort demographics are summarised in Table 1.
Table 1. Demographic characteristics of otosclerosis cohort.
| Variables | n=657 | |
|---|---|---|
| Gender | Male Female |
35% 65% |
| Age at consent | Years (mean ± SD) | 50 ± 13.3 |
| Ethnicity | White/White British Asian/Asian British Black/Black British Other No entry |
85% 7% 3% 2% 3% |
| Age of onset | ≤10 years 11-20 years 21-30 years 31-40 years 41-50 years ≥51 years |
6% 11% 30% 32% 17% 4% |
| Family history | ≥2 family members 1 family member No family history |
16% 24% 60% |
| Hearing loss | Bilateral Unilateral No entry/DK (n = 13) |
62% 38% |
| Tinnitus | Yes No No entry/DK (n = 10) |
68% 32% |
| Vertigo | Yes No No entry/DK (n = 22) |
31% 69% |
SD, standard deviation. DK, Do not know.
The characteristics of this British otosclerosis cohort are largely consistent with those of other published otosclerosis cohorts and the reported epidemiology of otosclerosis5,10,11. As has been reported previously there are more females (65%) than males in this British cohort. Regarding ethnicity, the make-up of the cohort is very similar to the ethnicities reported for the UK population in the 2011 UK Census (n=63,182,178) of 86% White/White British, 7% Asian/Asian British and 3% Black/Black British42. Most participants report the age of onset of the disease as in the 3rd and 4th decades (62%, Figure 1A), with onset above the age of 50 (4%) or below the age of 10 (6%) being rare. Familial clustering has long been identified in otosclerosis with evidence of a monogenic, reduced penetrance familial form of the disease proposed to exist alongside the more common sporadic disease5,10,11,16,17, and as in other cohorts a large minority of patients (40%) report having at least one other family member diagnosed with otosclerosis, although only 16% of cases report having at least two affected family members.
Figure 1. Characteristics of otosclerosis cohort.
(A) Age of onset and (B) incidence of hearing phenotypes in otosclerosis cohort according to duration of hearing loss (n=657). (C) Age of onset and (D) incidence of bilateral hearing loss in male and female cases according to duration of hearing loss (Male, n=233; Female; n=424). (E) Age of onset and (F) bilateral hearing loss in familial and non-familial cases according to duration of hearing loss (Familial, n=265; Non-familial; n=392). Familial cases were defined as having 1 or more affected family member. ns, not significant; *, p<0.05; **, p<0.01.
Although most patients in our cohort have bilateral disease (62%) this is lower than the often quoted figure of 70-85%1,11,12. In addition, most patients report suffering from tinnitus (68%) and a significant minority of subjects also reported symptoms of vertigo (31%). The incidence of tinnitus in our cohort is consistent with figures found in the literature of 50-85%16,43,44. However, the incidence of vertigo is higher than 9-24% found in previous studies11,16,43.
We examined the progression of symptoms with increased duration of hearing loss (Figure 1B). In patients who had less than 10 years of hearing loss, 45% of subjects reported bilateral symptoms, with this figure increasing to 70-85% of subjects after more than 30 years of hearing loss, which is in line with what is found in published literature. Incidence of vertigo also increased from 25% to 45% with duration of hearing loss. The relationship with tinnitus over time is less clear, which may be a result of improved hearing after surgery.
Gender and otosclerosis
There are a greater number of females than males in the cohort with a ratio of 1.9:1, which is in keeping with previous reports of a gender bias of 1.5-2:1 in the female to male prevalence of otosclerosis11,41,45,46. Dichotomising this British cohort by gender (Table 2) shows that the age of females at recruitment and self-reported age of onset is not significantly different from males (Figure 1C). Furthermore, there is evidence that women experience greater severity of symptoms at an earlier age than men. Women report significantly higher rates of bilateral disease (p=0.03), tinnitus (p=0.02), and a greater incidence of vertigo although this latter difference is not statistically significant (p=0.06). The higher rate of bilateral disease in women was maintained throughout progression of the disease although was only statistically significant at two time-points (Figure 1D).
Table 2. Demographic characteristics of otosclerosis cohort stratified by gender and family history.
| Variables | Male (n=233) |
Female (n=424) |
p-value | Familial (n=265) |
Non-familial (n=392) |
p-value | |
|---|---|---|---|---|---|---|---|
| Gender | Male Female |
n/a | n/a | n/a | 33% 67% |
37% 63% |
0.36 |
| Age at consent | Years (mean ± SD) | 51 ± 12.7 | 50 ± 13.6 | 0.23 | 52 ± 13.5 | 49 ± 13.1 | 8x10-3 |
| Ethnicity | White/White British Asian/Asian British Black/Black British Other No entry |
85% 7% 3% 1% 4% |
84% 8% 2% 4% 2% |
0.11 |
86% 8% 2% 2% 2% |
84% 7% 3% 3% 3% |
0.85 |
| Age of onset | ≤10 years 11-20 years 21-30 years 31-40 years 41-50 years ≥51 years |
7% 10% 23% 35% 21% 4% |
6% 11% 33% 30% 16% 4% |
0.11 | 4% 14% 34% 29% 14% 5% |
8% 8% 27% 34% 20% 3% |
6x10-3 |
| Family history | Familial Non-familial |
38% 62% |
42% 58% |
0.36 | n/a | n/a | n/a |
| Hearing loss | Bilateral Unilateral |
56% 44% |
66% 34% |
0.03 | 69% 31% |
58% 42% |
6x10-3 |
| Tinnitus | Yes No |
62% 38% |
72% 28% |
0.02 | 70% 30% |
67% 33% |
0.47 |
| Vertigo | Yes No |
26% 74% |
33% 67% |
0.06 | 36% 64% |
28% 72% |
0.04 |
Familial cases were defined as having 1 or more affected family member. SD, standard deviation. Significant differences in bold.
It has previously been proposed that the greater prevalence of women in otosclerosis cohorts might be explained by women seeking earlier clinical treatment than men47. However, the consistency and degree of the female bias across different cohorts seems unlikely to be fully explained by this factor leading to the hypothesis that oestrogen may play a role in the aetiology of the disorder26–29. In our cohort the similar age of onset reported in males and females does not support an earlier attendance of women in audiological clinics and instead is more consistent with the hypothesis that women are at a higher risk of developing clinical otosclerosis than men.
Familial inheritance and otosclerosis
Otosclerosis is unusual in that in some populations it is a relatively common disorder with a familial form of the disease, that is thought to have a monogenic origin, and a non-familial form. We examined the profile of otosclerosis comparing familial and non-familial cases (Table 2). Familial cases were defined as having 1 or more family members with otosclerosis. No significant differences were observed between familial cases with 1 affected relative and familial cases with 2 or more affected relatives (Table S1, Supplemental Digital Content 3).
The age at recruitment of familial cases was older than non-familial cases by an average of 3 years (p=8x10-3). Two recruitment pipelines were used, one via individuals attending for surgical treatment and another involved a retrospective recruitment of past surgical patients. Patients recruited retrospectively are therefore likely to be older than those recruited at surgery. Therefore, we examined the characteristics of our cohort comparing surgical and retrospective recruitment (Table S2, Supplemental Digital Content 3). The age at recruitment of the retrospective cases was significantly older than surgical cases (55 and 48 years respectively, p=1x10-9). A significant difference is also observed in the reporting of family history, where 35% of current surgical cases compared to 51% of retrospective cases report a family history of otosclerosis (p=2x10-4). It is therefore possible that the older age of familial patients at recruitment may be explained by a greater motivation of patients with family history to respond to the retrospective recruitment method and hence, that 35% represents a better estimate of familial incidence. This also highlights why such a wide range is observed in the published literature, with a positive family history reported between 30-70% of cases5,10,11,16,17, and probably reflects a recruitment bias within the cohorts.
Despite the older age at recruitment familial patients report a significantly earlier age of onset than non-familial subjects with a mode age of onset in the 3rd decade compared to 4th decade for non-familial patients (p=6x10-3; Figure 1E). Familial patients also report a higher incidence of bilateral disease and vertigo than non-familial subjects consistent with an earlier, more severe disease (p=6x10-3 and p=0.04 respectively). Incidences of tinnitus were similar in both groups. As familial subjects reported significantly higher rates of bilateral disease, we looked at the correlation between the duration and bilaterality of hearing loss in familial and non-familial participants (Figure 1F). In those patients who reported hearing loss for between 11 and 20 years, familial subjects reported significantly higher rates of bilateral disease compared to non-familial (70% and 53% respectively, p=0.02). This difference was not observed after a longer duration of otosclerosis but suggests familial otosclerosis may be quicker to progress to bilateral disease.
Previous studies have suggested the possibility of a differing aetiology for the two forms of otosclerosis5,43. However, most studies have not documented family history and therefore did not compare the characteristics of familial and non-familial cases of otosclerosis. In 2001, one study that evaluated a relatively small retrospective cohort of 183 patients found no difference in the degree of clinical severity between sporadic and familial cases43. This may explain why this is the first study to report that familial patients have a significantly earlier age of onset, and report a higher incidence of bilateral disease and vertigo than non-familial subjects, consistent with an earlier, more severe disease in familial cases suggesting that distinct pathological mechanisms may be involved.
Pedigree analysis
Otosclerosis is commonly described as having an autosomal dominant mode of inheritance, with evidence of incomplete penetrance5,10,18,19, although the empirical evidence available to support this is limited. To investigate the inheritance pattern of familial otosclerosis in our cohort, family pedigrees were constructed from 96 patients who reported a strong family history of otosclerosis. In 91 of the 96 families the apparent mode of inheritance was most consistent with an autosomal dominant pattern, with evidence of incomplete penetrance observed in 44 of these 91 families. The remaining five families having insufficient information to ascribe a mode of inheritance. In estimating the degree of penetrance in otosclerosis, we examined the number of affected parents of probands. Of the 96 pedigree probands, there are 62 affected parents indicating an estimated penetrance of at least 65%. Prior to this study very little data has been published since the 1980s regarding the commonly recognised mode of inheritance found in the literature. Therefore, our study is important in establishing this further.
Pregnancy, breastfeeding and oral contraception
The preponderance of female patients with otosclerosis has led to speculation that oestrogen signalling may be involved in the aetiology of the disease26–29. The typical age of onset of otosclerosis is concurrent with typical age of pregnancy and a link between the two has been suggested although remains disputed27,29. To clarify the relationship of pregnancy and hormones with otosclerosis, female specific questionnaire data were examined in more detail (Table 3). From the 424 females in the cohort, there were 409 responses for the female only questionnaire, where 313 (77%) had experienced at least one pregnancy (Preg_Y) and 96 had no history of pregnancy (Preg_N). Age at recruitment is significantly associated with pregnancy as might be expected since younger women reported fewer pregnancies (p=1x10-6). Overall no significant differences were observed in the age of onset, family history, bilateral disease, incidence of tinnitus or vertigo between Preg_Y and Preg_N women providing no evidence for an interaction between pregnancy and risk of disease. Both Preg_Y and Preg_N women have an earlier age of onset than men, but very similar to each other (Figure 2A).
Table 3. Demographic characteristics of female otosclerosis cases stratified by the effect of pregnancy on hearing loss.
| Variables | Preg_Y (n=313) |
Preg_N (n=96) |
p-value | Preg_HC (n=100) |
Preg_NC (n=204) |
p-value | |
|---|---|---|---|---|---|---|---|
| Age at consent | Years (mean ± SD) | 52 ± 12.8 | 44 ± 14.5 | 1x10-6 | 49 ± 12.5 | 53 ± 12.8 | 6x10-3 |
| Age of onset | ≤10 years 11-20 years 21-30 years 31-40 years 41-50 years ≥51 years |
6% 11% 32% 29% 18% 4% |
8% 15% 36% 29% 10% 2% |
0.27 | 6% 14% 45% 32% 3% 0% |
6% 9% 25% 28% 26% 6% |
6x10-6 |
| Family history | Familial Non-familial |
41% 59% |
45% 55% |
0.54 | 41% 59% |
40% 60% |
0.99 |
| Hearing loss | Bilateral Unilateral |
68% 32% |
58% 42% |
0.10 | 73% 27% |
65% 35% |
0.21 |
| Tinnitus | Yes No |
69% 31% |
77% 23% |
0.19 | 72% 28% |
69% 31% |
0.72 |
| Vertigo | Yes No |
36% 64% |
27% 73% |
0.17 | 32% 68% |
37% 63% |
0.45 |
| Number of pregnancies | 1 2 3 4 5+ |
n/a | n/a | n/a | 15% 37% 26% 15% 7% |
20% 45% 25% 6% 4% |
0.07 |
| Breastfed | Yes No |
n/a | n/a | n/a | 90% 10% |
78% 22% |
0.02 |
| Time spent breastfeeding | <1 month 2-5 months 6-12 months >1 year |
n/a | n/a | n/a | 11% 21% 27% 40% |
20% 23% 28% 29% |
0.19 |
| Long-term use of oral contraception | ≤5 years >5 years |
48% 52% |
34% 66% |
0.05# | 46% 54% |
49% 51% |
0.69 |
SD, standard deviation. Preg_Y, History of pregnancy; Preg_N, No history of pregnancy; Preg_HC, Pregnant with hearing changes; Preg_NC, Pregnant no hearing changes.
p<0.05. Significant differences in bold.
Figure 2. Comparison of otosclerosis characteristics in patients when stratified by the effect of pregnancy on hearing loss.
(A) Age of onset in women who had experienced pregnancy (Preg_Y, n=313) and those that had not (Preg_N, n=96). (B) Age of onset in women who report changes in hearing associated with pregnancy (Preg_HC, n=100) and those who report no change during pregnancy (Preg_NC, n = 204). (C) Bilateral hearing loss in pregnancy sub-groups. (D) Time spent breastfeeding and (E) number of pregnancies in Preg_HC and Preg_NC women. Male data used as control comparison in panels A-C (n = 233). ns, not significant; *, p<0.05; **, p<0.01; ****, p<0.0001.
However, 33% of Preg_Y women responded <Yes> to “Did you notice any changes in your hearing ability during your pregnancy/pregnancies?” When these third of Preg_Y women who reported awareness of changes in their hearing (Preg_HC) were compared with women that reported no changes in their hearing during pregnancy (Preg_NC), Preg_HC women were on average four years younger at recruitment than Preg_NC women (p=6x10-3, Table 3). In addition, there was a highly significant association between the reported age of onset of otosclerosis and whether women reported a change in hearing with pregnancy (p=6x10-6, Figure 2B). Preg_HC women have an earlier age of onset than both males and Preg_NC women. The age of onset in Preg_NC women is very similar to that of males. Hence, 43% of males and 40% of Preg_NC females have onset of symptoms before the age of 30, in contrast this figure rises to 65% of Preg_HC females. Preg_N women also have a lower age of onset, similar to that of Preg_HC females. One possibility is that this might be related to long-term use of oral contraceptives; 66% of Preg_N women have used oral contraception for more than 5 years compared to 52% of Preg_Y (p=0.049).
In the whole study group, females reported a higher incidence of bilateral disease than males (p=0.03). Preg_HC females reported a higher incidence of bilateral hearing problems compared to Preg_N women and men (p=0.04 and p=7x10-3 respectively, Figure 2C). For Preg_HC women, a significantly higher proportion breastfed after their pregnancies compared to Preg_NC women (90% compared to 78%, p=0.02). There was no significant difference between Preg_HC and Preg_NC women in the total time of breastfeeding (Figure 2D) or in the total number of pregnancies (Figure 2E).
In previous studies, it has been reported that in a proportion of women with otosclerosis a deterioration of hearing can occur during pregnancy5,10,27. Similar to these studies, it must be noted that the reporting of a patient’s impression of deterioration of hearing during pregnancy requires a subjective response. With that in mind, our data further confirms that in a proportion of women pregnancy leads to a deterioration of hearing. Furthermore, our data suggest that in this significant minority of women pregnancy may accelerate the progression of otosclerosis with a significant shift forward in the age of onset in these women.
Measles virus as a risk factor
The reported data on the role of measles infection in the pathogenesis of otosclerosis has been contradictory. In several studies, measles virus antigens and RNA have been detected in otosclerotic foci and it was hypothesised that this may be an inflammatory trigger during the active phase of otosclerosis30,31,33. In contrast, other studies failed to demonstrate any viral antibodies in otosclerotic foci or find any evidence of measles virus RNA in stapes samples or bone cell cultures32,34,36. We examined the profile of otosclerosis comparing cases that reported measles infection and cases without measles infection (Table S3, Supplemental Digital Content 3). Age at recruitment was significantly associated with measles infection which might be expected as more older patients who did not receive vaccination reported measles infection. Interestingly, patients with measles infection report a significantly later age of otosclerosis onset than subjects with no measles infection, with a mode age of onset in the 4th decade compared to 3rd decade respectively (p=2x10-3; Figure S1, Supplemental Digital Content 3).
Hence, we find no evidence to support a relationship between measles infection and risk of otosclerosis in our British cohort. Previously, studies in German cohorts have suggested that the incidence of otosclerosis has declined since the introduction of measles vaccination35, as well as highlighting the increasing age of otosclerosis patients at the time of surgery over this period48. Similar to the contradictions with measles virus antigens and RNA studies, our data do not support previous suggestions that measles infection is an otosclerosis risk factor. However, there is still a need for a definitive study that does not rely on retrospective self-reporting of measles infections which may explain our failure to detect an association here.
Audiometry and surgery
Audiological data was obtained from 154 subjects and the demographics of this study population was consistent with our main cohort (Table S4, Supplemental Digital Content 3). The audiometric results from our study are presented in Table S5 and S6 (see Supplemental Digital Content 3). The mean pre-operative pure-tone average (PTA) air-conduction (AC) threshold was 57 dB HL, and the bone-conduction (BC) was 28 dB HL. The air-bone gap (ABG) was 29 dB. After surgery, the corresponding results were AC 31 dB HL, BC 25 dB HL and ABG 5 dB. The mean hearing gain after surgery, expressed as improvement of the AC, was 27 dB and in 75% of patients the AC threshold improved by more than 20 dB. After surgery, the BC was maintained or not worsened by more than 5 dB in 86% of the patients. In 88% of patients, the post-operative ABG was within ≤10 dB, and in 98% it was within 20 dB. Importantly, the mean pure-tone air and bone hearing thresholds of our cohort are consistent with those of other published cohorts41,46.
As highlighted above, clinical differences can be observed in the epidemiology and severity of otosclerosis between groups and we wanted to evaluate whether these differences are reflected in pre- and post-operative audiometric thresholds between groups. We compared AC, BC and ABG between male and female patients (Figure S2A-C, Supplemental Digital Content 3) and no significant differences were observed in the pre-operative data. However, in post-operative data male patients had a significantly higher ABG (7 dB male and 5 dB female; p=0.04). Even though this is significant, the surgical outcome would still be considered successful. When comparing familial and non-familial surgical patients no significant differences were observed (Figure S2D-F, Supplemental Digital Content 3).
Finally, we compared hearing thresholds between Preg_Y and Preg_N women (Figure S3A-C, Supplemental Digital Content 3) and found no significant differences, which is consistent with other published data29. Even when we compared Preg_HC and Preg_NC women no significant impact was observed (Figure S3D-F, Supplemental Digital Content 3).
Similar surgical outcomes between sub-groups is perhaps understandable considering the relatively short-term follow-up of the study and participants would have sought medical advice at a point when their hearing loss constitutes a disability. This is more likely to be the case the greater the hearing loss, rather than the duration of the hearing loss. It would therefore be more informative to undertake a prospective study beginning from the age of onset of otosclerosis to evaluate whether clinical differences observed in the epidemiology and severity of otosclerosis can be reflected in hearing thresholds between groups. Additionally, these results suggest that whilst disease processes may be distinct between groups, the effect they have on the stapes footplate and bone conduction are similar.
Conclusion
In summary, this study has identified that familial otosclerosis has a more rapid onset, with a higher incidence of bilateral disease than non-familial otosclerosis. This is the first study to identify differences in the degree of clinical severity between familial and non-familial cases of otosclerosis. Therefore, suggesting that more than one aetiology may be involved and highlighting a need for further investigation. In addition, we have found further evidence to support the hypothesis that women are at a higher risk of developing clinical otosclerosis than men and pregnancy may accelerate disease progression, at least in a proportion of women. Despite these differences in otosclerosis onset and progression, patients choose to undergo surgery at similar times, with similar outcomes. Finally, our British otosclerosis cohort also confirms several characteristics largely consistent with previous published literature, including age of onset, sex ratio, family history and hearing loss symptoms associated with otosclerosis. Otosclerosis is a common cause of adult-onset hearing loss, resulting in a significant annual surgical load. There remain several unresolved questions with respect to certain aetiological factors that continue to limit the confidence of evidence-based patient information. We believe there is a further need for modern cohort observations, including longitudinal studies, and that our present cohort adds significantly to the literature on this elusive disease.
Supplementary Material
Acknowledgements
We are indebted to the patients and their families for their willingness to participate in our research study. We are also grateful for the UK CRN staff, physicians and hospital staff, whose efforts were essential to this research. With specific acknowledgment to Naveed Kara (Consultant ENT Surgeon, Freeman Hospital, Newcastle upon Tyne), Judith Heaton (Consultant ENT Surgeon, Sunderland Royal Hospital, Sunderland) and Stephen E. Jones (Consultant ENT Surgeon, Ninewells Hospital, Dundee) for setting up additional recruitment sites. This work was funded by Action on Hearing Loss (grant references G71, G83 and 512:UEI:SD) and supported by the National Institute for Health Research University College London Hospitals Biomedical Research Centre.
This work was funded by Action on Hearing Loss (grant references G71, G83 and 512:UEI:SD) and supported by the National Institute for Health Research University College London Hospitals Biomedical Research Centre.
Footnotes
The authors disclose no conflicts of interest.
References
- 1.House HP, Hansen MR, Al Dakhail AA, House JW. Stapedectomy versus stapedotomy: comparison of results with long-term follow-up. Laryngoscope. 2002 Nov;112(11):2046–2050. doi: 10.1097/00005537-200211000-00025. [DOI] [PubMed] [Google Scholar]
- 2.Lancer H, Manickavasagam J, Zaman A, Lancer J. Stapes surgery: a National Survey of British Otologists. Eur Arch Otorhinolaryngol. 2016 Feb;273(2):371–379. doi: 10.1007/s00405-015-3560-6. [DOI] [PubMed] [Google Scholar]
- 3.Batson L, Rizzolo D. Otosclerosis: An update on diagnosis and treatment. JAAPA. 2017 Feb;30(2):17–22. doi: 10.1097/01.JAA.0000511784.21936.1b. [DOI] [PubMed] [Google Scholar]
- 4.Danesh AA, Shahnaz N, Hall JW., 3rd The Audiology of Otosclerosis. Otolaryngol Clin North Am. 2018 Apr;51(2):327–342. doi: 10.1016/j.otc.2017.11.007. [DOI] [PubMed] [Google Scholar]
- 5.Morrison AW. Genetic factors in otosclerosis. Ann R Coll Surg Engl. 1967 Aug;41(2):202–237. [PMC free article] [PubMed] [Google Scholar]
- 6.Gristwood RE, Venables WN. Otosclerosis in South Australia. Clin Otolaryngol Allied Sci. 1984 Aug;9(4):221–228. doi: 10.1111/j.1365-2273.1984.tb01501.x. [DOI] [PubMed] [Google Scholar]
- 7.Gordon MA. The genetics of otosclerosis: a review. Am J Otol. 1989 Nov;10(6):426–438. doi: 10.1097/00129492-198911000-00003. [DOI] [PubMed] [Google Scholar]
- 8.Browning GG, Gatehouse S. The prevalence of middle ear disease in the adult British population. Clin Otolaryngol Allied Sci. 1992 Aug;17(4):317–321. doi: 10.1111/j.1365-2273.1992.tb01004.x. [DOI] [PubMed] [Google Scholar]
- 9.Sakihara Y, Parving A. Clinical otosclerosis, prevalence estimates and spontaneous progress. Acta Otolaryngol. 1999;119(4):468–472. doi: 10.1080/00016489950181017. [DOI] [PubMed] [Google Scholar]
- 10.Larsson A. Otosclerosis. A genetic and clinical study. Acta Otolaryngol Suppl. 1960;154:1–86. [PubMed] [Google Scholar]
- 11.Del Bo M, Zaghis A, Ambrosetti U. Some observations concerning 200 stapedectomies: fifteen years postoperatively. Laryngoscope. 1987 Oct;97(10):1211–1213. doi: 10.1288/00005537-198710000-00017. [DOI] [PubMed] [Google Scholar]
- 12.Lippy WH, Berenholz LP, Burkey JM. Otosclerosis in the 1960s, 1970s, 1980s, and 1990s. Laryngoscope. 1999 Aug;109(8):1307–1309. doi: 10.1097/00005537-199908000-00022. [DOI] [PubMed] [Google Scholar]
- 13.Emmett JR. Physical examination and clinical evaluation of the patient with otosclerosis. Otolaryngol Clin North Am. 1993 Jun;26(3):353–357. [PubMed] [Google Scholar]
- 14.Ramsay HA, Linthicum FH., Jr Mixed hearing loss in otosclerosis: indication for long-term follow-up. Am J Otol. 1994 Jul;15(4):536–539. [PubMed] [Google Scholar]
- 15.Topsakal V, Fransen E, Schmerber S, et al. Audiometric analyses confirm a cochlear component, disproportional to age, in stapedial otosclerosis. Otol Neurotol. 2006 Sep;27(6):781–787. doi: 10.1097/01.mao.0000231500.46534.79. [DOI] [PubMed] [Google Scholar]
- 16.Cawthorne T. Otosclerosis. J Laryngol Otol. 1955 Jul;69(7):437–456. doi: 10.1017/s0022215100050933. [DOI] [PubMed] [Google Scholar]
- 17.Sabitha R, Ramalingam R, Ramalingam KK, Sivakumaran TA, Ramesh A. Genetics of otosclerosis. J Laryngol Otol. 1997 Feb;111(2):109–112. doi: 10.1017/s0022215100136606. [DOI] [PubMed] [Google Scholar]
- 18.Causse JR, Causse JB. Otospongiosis as a genetic disease. Early detection, medical management, and prevention. Am J Otol. 1984 Jan;5(3):211–223. [PubMed] [Google Scholar]
- 19.Saeed SR, Briggs M, Lobo C, Al-Zoubi F, Ramsden RT, Read AP. The genetics of otosclerosis: pedigree studies and linkage analysis. Adv Otorhinolaryngol. 2007;65:75–85. doi: 10.1159/000098675. [DOI] [PubMed] [Google Scholar]
- 20.Van Den Bogaert K, Govaerts PJ, Schatteman I, et al. A second gene for otosclerosis, OTSC2, maps to chromosome 7q34-36. Am J Hum Genet. 2001 Feb;68(2):495–500. doi: 10.1086/318185. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Chen W, Campbell CA, Green GE, et al. Linkage of otosclerosis to a third locus (OTSC3) on human chromosome 6p21.3-22.3. J Med Genet. 2002 Jul;39(7):473–477. doi: 10.1136/jmg.39.7.473. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Brownstein Z, Goldfarb A, Levi H, Frydman M, Avraham KB. Chromosomal mapping and phenotypic characterization of hereditary otosclerosis linked to the OTSC4 locus. Arch Otolaryngol Head Neck Surg. 2006 Apr;132(4):416–424. doi: 10.1001/archotol.132.4.416. [DOI] [PubMed] [Google Scholar]
- 23.Schrauwen I, Weegerink NJ, Fransen E, et al. A new locus for otosclerosis, OTSC10, maps to chromosome 1q41-44. Clin Genet. 2011 May;79(5):495–497. doi: 10.1111/j.1399-0004.2010.01576.x. [DOI] [PubMed] [Google Scholar]
- 24.Schrauwen I, Ealy M, Huentelman MJ, et al. A genome-wide analysis identifies genetic variants in the RELN gene associated with otosclerosis. Am J Hum Genet. 2009 Mar;84(3):328–338. doi: 10.1016/j.ajhg.2009.01.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Ziff JL, Crompton M, Powell HR, et al. Mutations and altered expression of SERPINF1 in patients with familial otosclerosis. Hum Mol Genet. 2016 Jun 15;25(12):2393–2403. doi: 10.1093/hmg/ddw106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Podoshin L, Gertner R, Fradis M, et al. Oral contraceptive pills and clinical otosclerosis. Int J Gynaecol Obstet. 1978;15(6):554–555. doi: 10.1002/j.1879-3479.1977.tb00755.x. [DOI] [PubMed] [Google Scholar]
- 27.Gristwood RE, Venables WN. Pregnancy and otosclerosis. Clin Otolaryngol Allied Sci. 1983 Jun;8(3):205–210. doi: 10.1111/j.1365-2273.1983.tb01428.x. [DOI] [PubMed] [Google Scholar]
- 28.Vessey M, Painter R. Oral contraception and ear disease: findings in a large cohort study. Contraception. 2001 Feb;63(2):61–63. doi: 10.1016/s0010-7824(01)00176-7. [DOI] [PubMed] [Google Scholar]
- 29.Lippy WH, Berenholz LP, Schuring AG, Burkey JM. Does pregnancy affect otosclerosis? Laryngoscope. 2005 Oct;115(10):1833–1836. doi: 10.1097/01.MLG.0000187573.99335.85. [DOI] [PubMed] [Google Scholar]
- 30.Arnold W, Friedmann I. [Detection of measles and rubella-specific antigens in the endochondral ossification zone in otosclerosis] Laryngol Rhinol Otol (Stuttg) 1987 Apr;66(4):167–171. [PubMed] [Google Scholar]
- 31.McKenna MJ, Mills BG. Immunohistochemical evidence of measles virus antigens in active otosclerosis. Otolaryngol Head Neck Surg. 1989 Oct;101(4):415–421. doi: 10.1177/019459988910100401. [DOI] [PubMed] [Google Scholar]
- 32.Roald B, Storvold G, Mair IW, Mjoen S. Respiratory tract viruses in otosclerotic lesions. An immunohistochemical study. Acta Otolaryngol. 1992;112(2):334–338. doi: 10.1080/00016489.1992.11665428. [DOI] [PubMed] [Google Scholar]
- 33.Arnold W, Niedermeyer HP, Lehn N, Neubert W, Hofler H. Measles virus in otosclerosis and the specific immune response of the inner ear. Acta Otolaryngol. 1996 Sep;116(5):705–709. doi: 10.3109/00016489609137910. [DOI] [PubMed] [Google Scholar]
- 34.Grayeli AB, Palmer P, Tran Ba Huy P, et al. No evidence of measles virus in stapes samples from patients with otosclerosis. J Clin Microbiol. 2000 Jul;38(7):2655–2660. doi: 10.1128/jcm.38.7.2655-2660.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Arnold W, Busch R, Arnold A, Ritscher B, Neiss A, Niedermeyer HP. The influence of measles vaccination on the incidence of otosclerosis in Germany. Eur Arch Otorhinolaryngol. 2007 Jul;264(7):741–748. doi: 10.1007/s00405-007-0253-9. [DOI] [PubMed] [Google Scholar]
- 36.Flores-Garcia ML, Colin-Castro CA, Hernandez-Palestina MS, Sanchez-Larios R, Franco-Cendejas R. Absence of Measles Virus Detection from Stapes of Patients with Otosclerosis. Otolaryngol Head Neck Surg. 2018 Jan;158(1):158–162. doi: 10.1177/0194599817733674. [DOI] [PubMed] [Google Scholar]
- 37.Daniel HJ., 3rd Stapedial otosclerosis and fluorine in the drinking water. Arch Otolaryngol. 1969 Nov;90(5):585–589. doi: 10.1001/archotol.1969.00770030587011. [DOI] [PubMed] [Google Scholar]
- 38.Miyajima C, Ishimoto S, Yamasoba T. Otosclerosis associated with Ehlers-Danlos syndrome: report of a case. Acta Otolaryngol Suppl. 2007 Dec;(559):157–159. doi: 10.1080/03655230701600418. [DOI] [PubMed] [Google Scholar]
- 39.Santos F, McCall AA, Chien W, Merchant S. Otopathology in Osteogenesis Imperfecta. Otol Neurotol. 2012 Dec;33(9):1562–1566. doi: 10.1097/MAO.0b013e31826bf19b. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Committee on Hearing and Equilibrium guidelines for the evaluation of results of treatment of conductive hearing loss. Otolaryngol Head Neck Surg. 1995 Sep;113(3):186–187. doi: 10.1016/S0194-5998(95)70103-6. [DOI] [PubMed] [Google Scholar]
- 41.Stromback K, Lundman L, Bjorsne A, Grendin J, Stjernquist-Desatnik A, Dahlin-Redfors Y. Stapes surgery in Sweden: evaluation of a national-based register. Eur Arch Otorhinolaryngol. 2017 Jun;274(6):2421–2427. doi: 10.1007/s00405-017-4510-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Office for National Statistics. Ethnicity and National Identity in England and Wales 2011. [Accessed December 1, 2017]; http://www.ons.gov.uk/ons/dcp171776_290558.pdf.
- 43.Mazzoli M, Rosignoli M, Martin A. Otosclerosis: are familial and isolated cases different disorders? J Audiol Med. 2001;10(1):49–59. [Google Scholar]
- 44.Gristwood RE, Venables WN. Otosclerosis and chronic tinnitus. Ann Otol Rhinol Laryngol. 2003 May;112(5):398–403. doi: 10.1177/000348940311200502. [DOI] [PubMed] [Google Scholar]
- 45.Manolidis S, Alford RL, Smith RJ, Ball C, Manolidis L. Do the genes that cause otosclerosis reduce susceptibility to otitis media? Otol Neurotol. 2003 Nov;24(6):868–871. doi: 10.1097/00129492-200311000-00007. [DOI] [PubMed] [Google Scholar]
- 46.Vincent R, Sperling NM, Oates J, Jindal M. Surgical findings and long-term hearing results in 3,050 stapedotomies for primary otosclerosis: a prospective study with the otology-neurotology database. Otol Neurotol. 2006 Dec;27(8 Suppl 2):S25–47. doi: 10.1097/01.mao.0000235311.80066.df. [DOI] [PubMed] [Google Scholar]
- 47.Smyth G. Otosclerosis. In: Kerr A, editor. Scott-Brown’s Otolaryngology: Volume 3: Otology. 6th ed. Oxford: Butterworth-Heinemann; 1997. [Google Scholar]
- 48.Niedermeyer HP, Arnold W, Schwub D, Busch R, Wiest I, Sedlmeier R. Shift of the distribution of age in patients with otosclerosis. Acta Otolaryngol. 2001 Jan;121(2):197–199. doi: 10.1080/000164801300043488. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.