Introduction
Cholesteatoma is characterised by a benign a mass of keratinising squamous epithelium, peri-matrix and keratin, usually found in the tympanic cavity, mastoid or subepithelial connective tissue, which demonstrates destructive and inflammatory properties.Reference Maniu, Harabagiu, Perde Schrepler, Cătană, Fănuţă and Mogoantă1, Reference Yung, Tono, Olszewska, Yamamoto, Sudhoff and Sakagami2 The exact pathogenic molecular mechanisms behind the formation and propagation of cholesteatoma remain unclear. Studies examining the immunohistochemistry of the matrix and perimatrix have considerably improved the knowledge of cholesteatoma pathogenesis, although its aetiology has not been established.Reference Dodson, Hashisaki, Hobgood and Lambert3 Left untreated, cholesteatoma is associated with significant morbidity through complications including recurrent ear infections, hearing loss, facial nerve paralysis and destruction of the semi-circular canals.Reference Zarandy and Rutka4, Reference Nevoux, Lenoir, Roger, Denoyelle, Ducou Le Pointe and Garabédian5 Mortality is rare, but is usually related to intracranial infection.Reference Zarandy and Rutka4
Cholesteatoma can be classified into congenital and acquired forms. Congenital cholesteatoma is mainly seen in children and presents with an intact tympanic membrane, while the acquired form is usually seen in adults and is typically associated with a defect in the tympanic membrane.Reference Persaud, Hajioff, Trinidade, Khemani, Bhattacharyya and Papadimitriou6 Cholesteatoma is uncommon, with a reported incidence of 3–15 cases per 100 000 childrenReference Tos7, Reference Kuo, Shiao, Yung, Sakagami, Sudhoff and Wang8 and 9.2–12.6 cases per 100 000 adults.Reference Tos7, Reference Olszewska, Wagner, Bernal-Sprekelsen, Ebmeyer, Dazert and Hildmann9 Geographical studies have identified higher rates of cholesteatoma within less developed countriesReference Louw10 and in areas of greater socioeconomic deprivation.Reference Khalid-Raja, Tikka and Coulson11 Familial tendencies have been reported, with cholesteatoma affecting several family members.Reference Homøe and Rosborg12, Reference Prinsley13
There have been a few studies investigating the differences between childhood and adult cholesteatoma. These have focused on various aspects of the disease, including symptom presentation, intra-operative cholesteatoma or granulation levels, hearing levels, and molecular differences. Edelstein et al. found increased neuro-otological symptoms in adults compared to children.Reference Edelstein, Parisier, Ahuja, Juarbe, Chute and Wenig14 Regarding complications, while some authors have reported fewer complications in children,Reference Darrouzet, Duclos, Portmann and Bebear15 others have not found any statistical differences between the two groups.Reference Lynrah, Bakshi, Panda and Khandelwal16 Mallet et al. reported more aggressive cholesteatoma in children based on increased levels of the monoclonal antibody MIB1, which is thought to be responsible for epithelial hyperproliferation and hence more aggressive disease.Reference Mallet, Nouwen, Lecomte-Houcke and Desaulty17 This was further confirmed by Dornelles et al., who reported an increased number of inflammatory markers, including cluster of differentiation 31, and matrix metalloproteinases 2 and 9, in childhood cholesteatoma compared to adult disease.Reference Dornelles Cde, da Costa, Meurer, Rosito, da Silva and Alves18
We reviewed our experiences of cholesteatoma surgery with the aim of answering three main questions: (1) do the intra-operative findings support the notion that more aggressive disease is present in children?; (2) do the rates of revision surgery for cholesteatoma differ between children and adults?; and (3) do differences exist between the post-operative complications seen in children and adults following cholesteatoma surgery?
Materials and methods
Data were retrospectively extracted from the Common Otology Audit,19 an online database for recording ear operations. This data represented the operative caseload of a single otology consultant and his trainees for cholesteatoma between January 2006 and May 2017. All operations took place at the Norfolk and Norwich or the James Paget University Hospitals. Patient electronic health records and clinic letters were also reviewed, and these provided a valuable source of information.
Based on age, data were classified into two groups. Group 1 (n = 58) comprised of children aged below 16 years, while group 2 (n = 260) consisted of adults aged 16 years or more. Patients’ demographic details were gathered, and data for specific parameters including cholesteatoma site, operative findings, revision surgery rates, audiometric results and post-operative follow-up duration were extracted.
Statistical analysis was performed on the two groups to assess for significant differences between the children and adults with cholesteatoma. This was conducted using GraphPad Prism software, version 6.00 for Windows (GraphPad Software, La Jolla, California, USA). Where appropriate, quantitative data were compared using Fisher's exact test or the student's unpaired t-test. A p-value of less than 0.05 was considered statistically significant.
Results
Demographics
In total, 352 operations were performed for cholesteatoma in 318 patients; 186 operations were for left-sided disease, while 166 were for right-sided disease. Bilateral cholesteatoma was seen in 13 patients (4.1 per cent).
In children, 71 operations were performed on 58 patients. The average age at operation was 11.0 years and the average length of recorded follow up during the study period was 16 months (range, 3–84 months). Left-sided disease was seen in 36 cases, compared to 35 on the right. Bilateral disease was seen in three children (5.2 per cent), all of whom underwent bilateral surgery. One child had revision surgery performed on the same ear twice for cholesteatoma.
In the adult group, 281 operations were carried out on 260 patients. The average age at operation was 46.0 years, while the average length of recorded follow up during the study period was 12 months (range, 3–84 months). Left-sided disease was seen in 150 cases, compared to 131 for right-sided disease. Bilateral cholesteatoma was seen in 10 adults (3.8 per cent), of which 9 (3.5 per cent) underwent bilateral surgery.
Extension of disease
Regarding the site of cholesteatoma, children had significantly higher rates of cholesteatoma behind the tympanic membrane (p = 0.007), and in the sinus tympani (p = 0.001), the mastoid antrum (p = 0.014) and mastoid air cells (p = 0.003). There was no statistical difference in cholesteatoma affecting the attic between both groups (p = 0.559) (Table I).
Operative site of cholesteatoma in children and adults
* Fisher's exact test
Ossicular erosion
The level of ossicular erosion was significantly higher in children compared to adults in the malleus (p = 0.041), incus (p = 0.042) and stapes superstructure (p = 0.0016). Cholesteatoma affecting the stapes footplate was not seen in either group (Table II).
Incidence of ossicular erosion due to cholesteatoma in children and adults
* Fisher's exact test
Revision surgery
Within this study, all revision mastoid surgical procedures were performed for cholesteatoma. In adults, 225 procedures (80.1 per cent) were primary operations and 56 procedures (19.9 per cent) were revision operations. In 9 adults, both primary and revision surgical procedures were undertaken by the authors’ team during the study period, with an average interval of 565.8 days between operations. In the remaining 47 adults, primary surgery was either performed outside the study period, by a different surgeon or at a different institution. The rate of revision surgery was comparable in children: 55 (77 per cent) were primary operations and 16 (23 per cent) were revision cases. Similarly, 9 children had both primary and revision surgery during the study period, with an average interval of 647.1 days between operations. There was no significant difference in the rate of revision surgery for cholesteatoma between children and adults (Fisher's exact test, p = 0.624).
Complications
In children, four complications (5.6 per cent) were seen in four different patients, which including facial nerve palsy (1.4 per cent), hypertrophic scarring (1.4 per cent), post-auricular fistula (1.4 per cent) and wound abscess (1.4 per cent) (Table III).
Incidence and outcome of complications associated with cholesteatoma surgery in children
In adults, a total of 23 complications (8.2 per cent) occurred in 18 different patients. These included cerebrospinal fluid leak (0.4 per cent), dead ear (2.1 per cent), facial palsy (1.1 per cent), taste disturbance (0.4 per cent), vertigo (3.9 per cent) and wound infection (0.4 per cent) (Table IV).
Incidence and outcome of complications associated with cholesteatoma surgery in adults
CSF = cerebrospinal fluid
No statistical difference was found between children and adults in terms of the rate of complications following cholesteatoma surgery (Fisher's exact test, p = 0.620).
Audiometric results
A variety of reconstructive techniques were used individualised according to the operative findings. Post-operative pure tone audiometry demonstrated an average post-operative hearing gain of 0.62 dB in adults (range, −92.5 to 42.5 dB) and 1.67 dB in children (range, −45 to 36.9 dB); the difference was not statistically significant (unpaired t-test, p = 0.8341).
Discussion
We found significant differences between childhood cholesteatoma and adult disease, the former of which was more extensive and destructive at the time of surgery. These results support the findings of other studies that have suggested cholesteatoma in children to be more aggressive.Reference Lynrah, Bakshi, Panda and Khandelwal16, Reference Visvanathan, Kubba and Morrissey20, Reference Glasscock, Dickins and Wiet21 Although many studies have reviewed children and adult populations with cholesteatoma, a limited number of studies have directly compared these two groups.
An Egyptian study by ElrashidiReference Elrashidi22 divided 81 patients with cholesteatoma into age-based groups: children, adolescents and adults. The overall extent of disease was significantly higher in children; however, the rate of ossicular erosion between the groups was not significantly different. Sade and colleagues presented the findings of two separate studies that investigated untreated cholesteatoma disease in 63 children and 148 adults.Reference Sadé and Fuchs23, Reference Sadé and Shatz24 Their results found similar levels of malleus and stapes erosion, while adult cholesteatoma was associated with higher rates of incus erosion.
Although the differences between children and adult cholesteatoma are not fully understood, studies have demonstrated differences in the histological constituents of cholesteatoma affecting these groups. Elevated levels of the monoclonal antibody MIB1, cluster of differentiation 31, and matrix metalloproteinases 2 and 9 have all been isolated in childhood cholesteatoma cases compared to adult disease.Reference Mallet, Nouwen, Lecomte-Houcke and Desaulty17, Reference Dornelles Cde, da Costa, Meurer, Rosito, da Silva and Alves18 The pro-inflammatory or hyperproliferative effects of these mediators may explain why cholesteatoma in children demonstrates a greater propensity to spread and cause ossicular erosion. Other theories that may account for these differences are raised within an article by Preciado, which reviewed the biology of cholesteatoma in children. These theories include increased levels of circulating growth hormone among children, a higher incidence of childhood infectious otitis media and differences in aeration of the mastoid cavity.Reference Preciado25
• Cholesteatoma is considered more aggressive in children than in adults, but few studies have directly compared these two groups
• In this study, cholesteatoma was more extensive and destructive in children compared to adults
• Rates of revision surgery and post-operative complications were similar between children and adults
• These findings support a more aggressive disease process in children
Interestingly, despite childhood cholesteatoma behaving more aggressively, no significant differences were found in revision rates and complications associated with surgery. Around one in five patients underwent revision mastoid surgery for cholesteatoma within children and adults. Adult cholesteatoma surgery was associated with more significant complications, including six incidences of post-operative dead ear. Importantly, four cases of dead ear were predictable at the time of surgery given the extent of disease found. Further analysis of adult patients who developed significant post-operative complications extends beyond the scope of this study; however, these findings may represent a subset of more aggressive adult cholesteatoma or be a consequence of delayed surgical intervention.
Conclusion
This study found significantly higher rates of ossicular erosion and extension of cholesteatoma within the middle ear in children, suggestive of a more aggressive disease process. However, this did not translate into differences in cholesteatoma recurrence, complications or post-operative hearing gain. The study findings suggest that a more pragmatic approach to managing childhood cholesteatoma is needed, in which surgery is offered early to avoid further progression of disease. It may also inform decision making regarding the need for reconstructive surgery, although this should be assessed on a case-by-case basis.
Acknowledgements
We are especially grateful to Mr Peter Ransome and Ms Debbie Reade (Library Department, James Paget University Hospital), who kindly conducted a literature review and helped to obtain original articles for this study. Lastly, I wish to dedicate this work to my dear friend Dr Oscar Cassagneau-Francis.
Competing interests
None declared.
