Introduction
Chronic suppurative otitis media with cholesteatoma is a locally aggressive cell growth disorder that has a tendency to cause temporal as well as intracranial complications by virtue of bone destruction. While the prevalence of the mucosal type of chronic suppurative otitis media is associated with malnutrition, poor hygiene and overcrowding, chronic suppurative otitis media with cholesteatoma or the squamosal variant has long been considered more common in developed countries with improved living conditions and better healthcare.Reference Bluestone1
The term ‘cholesteatoma’ is a misnomer. Although formerly considered to be a tumour, it is a cell proliferation disorder. It consists of a matrix of multi-layered squamous epithelium surrounded by connective tissue made up of granulation tissue also known as peri-matrix. The cholesteatoma core is composed of keratinous debris. The level of activity or inflammation in the peri-matrix determines the aggressiveness of the cholesteatoma.Reference Bassiouny, Badour, Omran and Osama2
The peri-matrix of cholesteatoma is rich in enzymes such as metalloproteinases and various angiogenic factors. The level of inflammation within the cholesteatoma is directly proportional to the level of metalloproteinase activity, angiogenesis and the cluster of differentiation 31 count, which in turn determines the aggressiveness of cholesteatoma.Reference Dornelles, da Costa, Meurer, Rosito, da Silva and Alves3 It is this chronic inflammation and granulation tissue formation that is responsible for osteitis and ossicular destruction associated with cholesteatoma, which subsequently influences the extent of hearing loss. Further influences include the site of disease and its extension in the middle-ear cavity at the time of presentation.Reference Albera, Canale, Piumetto, Lacilla and Dagna4
Although there are a number of surgical approaches for cholesteatoma depending upon the extent and presence of complications, canal wall down mastoidectomy is a favoured option as it guarantees more complete removal of the disease and carries a lower risk of recurrence. Nonetheless, this treatment mandates periodic cleaning of the mastoid cavity.Reference Shirazi, Muzaffar, Leonetti and Marzo5 Further, such a cavity needs to be well epithelised. This necessitates the post-operative packing of the cavity with medicated packs or substances such as bismuth iodine paraffin paste or xeroform, to promote healing.
The clinical behaviour of paediatric cholesteatoma has long been a matter of debate. While both Tos and Edelstein consider paediatric cholesteatoma to be less expansile, resulting in fewer complications, other researchers, such as Glasscock, as well as Palva, suggest that paediatric cholesteatoma is more aggressive and extensive than its adult counterpart (as described in Dornelles et al.Reference Dornelles, da Costa, Meurer, Rosito, da Silva and Alves3). Further, there are those, like Smythe and colleagues (as cited in Dornelles et al.Reference Dornelles, da Costa, Meurer, Rosito, da Silva and Alves3), who consider paediatric cholesteatoma to be a completely different disease with its own characteristics.
Although our understanding of the histological and clinical behaviour of paediatric cholesteatoma has markedly improved, there has been a dearth of comparative studies of various characteristics between paediatric and adult cholesteatoma. In addition, most of the studies are retrospective in nature, with their own sets of limitations. Furthermore, the knowledge is limited by the polarity of data, with the majority of it being region-specific. Hence, we conducted a prospective study at our centre that aimed to bridge the gap by comparing the various parameters, including pre-operative hearing level, intra-operative extent of disease and ossicular status, between paediatric and adult cases in a single study.
Materials and methods
This cohort, prospective, observational study was conducted in the Department of ENT, Government Medical College and Hospital, Chandigarh, India, over a period of two years. It involved 50 patients who were clinically diagnosed with chronic suppurative otitis media (attico-antral disease) with cholesteatoma. These patients were composed of two equally sized groups: children (aged less than 18 years) and adults (aged over 18 years).
Patients who did not give consent to be included in the study were excluded. Also excluded were those who had a history of previous surgery in the same ear for the same disease, or who were being operated for another condition, neoplastic or granulomatous lesions of the temporal bone, an immunocompromised status, or were otherwise unfit for surgery.
The selected patients were subjected to a detailed history-taking, and underwent complete general and ENT examinations. Otoscopic examination of the ears was conducted, followed by examination under the microscope. Pure tone audiogram was performed in all cases to document the patients’ hearing status. X-rays of the mastoid bilaterally, with Law's lateral view and Towne's oblique view, were taken to assess the pathology and surgical anatomy of the mastoid. Computed tomography (CT) of the temporal bone was not carried out in most patients for economic reasons (some individuals had undergone CT outside of our institution).
All patients underwent canal wall down mastoidectomy performed by a single senior consultant. The intra-operative findings, including extent of disease and ossicular status, were noted.Reference Farrior6
Post-operatively, the mastoid cavities were packed monthly with bismuth iodine paraffin paste packing. The patients were followed up for six months or until the cavity was completely epithelialised, whichever was sooner.
The chi-square test was used to evaluate the level of significance; comparisons were considered significant when p-values were less than 0.05.
Results
A total of 50 patients with chronic suppurative otitis media with cholesteatoma were selected for the study, comprising equal numbers of adults and children. The age range of the children was 5–16 years (mean ± standard deviation (SD) = 12.52 ± 2.74). The age range of the adults was 19–56 years (mean ± SD = 32.72 ± 10.39). The number of male and female patients was 16 and 9, respectively, in the paediatric group, and was 18 and 7, respectively, in the adult group. In both groups, the left side was more commonly involved than the right side, and bilateral ear involvement was seen in 16 per cent of cases.
Ear discharge was the predominant complaint, with a mean duration of 5.48 years in paediatrics and 13.60 years in adults. The average duration of decreased hearing was 5.271 years in children and 10.8 years in adults. The most common otoscopic finding in our study was attic destruction with cholesteatoma in both the groups. Attic destruction was seen in 20 of the 25 adult patients, and was evident in 17 of the 25 child patients (p < 0.05). Posterosuperior retraction pocket was more commonly seen in the paediatric group (52 per cent) than in the adult group (44 per cent). Most of the patients had more than one clinical finding as observed during examination under the microscope.
In our study, the cavity was seen in 16 of the 25 paediatric patients on X-ray of the mastoid (using Law's lateral view and Towne's view). However, this was not statistically significantly different from adults, in which the cavity was seen in 15 cases (p > 0.05). A diploic mastoid air cell system was the least common radiological finding observed.
The mean bone conduction threshold was 50.56 dB in children and 54.24 dB in adults. The mean air conduction threshold was 26.24 dB in children and 24.16 dB in adults. The average air–bone gap was 24.32 dB in paediatrics and 30.24 dB in adults (p > 0.05). In both groups, moderate hearing loss was the most common presentation.
Intra-operatively, one paediatric cholesteatoma patient presented with post-aural abscess and there was an absence of bony cortex over the antrum. Involvement of the attic, aditus and antrum was observed in all the cases in our study. Further, involvement of middle ear, sinus tympani area, hypotympanum and Eustachian tube area was found to be more common in paediatrics with cholesteatoma than in adults.
Dural plate dehiscence was more common in adults than in children (10 adults and 8 paediatrics) (p > 0.05). One adult cholesteatoma case was associated with extradural abscess. Sinus plate dehiscence was more common in paediatric patients (n = 7) as compared to adults (n = 2) (p > 0.05). Out of the 50 patients, facial canal dehiscence was seen in 12 patients (9 adults and 3 paediatrics). The incidence of facial canal dehiscence was significantly higher in adults (36 per cent) than in children (12 per cent) (p < 0.05). The horizontal segment was the most commonly involved region in both the groups, and isolated vertical segment dehiscence was least common. Lateral semicircular canal dehiscence was observed in 12 cases, being more common in adults (n = 8, 32 per cent) than in children (n = 4, 16 per cent) (p > 0.05).
Our study showed involvement of the ossicles by the disease, either partially or completely, in almost all the cases. In both the groups, the incus was the most commonly involved bone. Involvement of the stapes suprastructure by the disease was significantly more common in children (64 per cent) than in adults (32 per cent) (p < 0.05).
In our study, the mean duration for mastoid cavity healing was 3.4 months in children and 2.96 months in adults, which was comparable (p > 0.05). Forty-nine out of 50 cases healed within a period of six months.
Discussion
This prospective, cohort, observational study was conducted on 50 patients (25 adults and 25 paediatrics) attending the ENT out-patients department, who were clinically diagnosed with chronic suppurative otitis media with cholesteatoma, and who were undergoing canal wall down mastoidectomy. These patients were followed up for a period of six months post-operatively.
The age at presentation varied from 5 years to 56 years. The age range of the children varied from 5 to 16 years (mean, 12.52 years); the age range of the adults ranged from 19 to 56 years (mean, 32.72 years). This finding is comparable to observations by Aquino et al., who reported age ranges of 4–15 years in children and 16–68 years in adults.Reference Aquino, Cruz Filho and de Aquino7 Another study, by De Corso et al., reported the age ranges of cholesteatoma presentation to be 3–16 years (mean, 8.8 years) in children and 18–81 years (mean, 45.4 years) in adults.Reference De Corso, Marchese, Scarano and Paludetti8
In both the groups, disease was more common among males (72 per cent of adults and 64 per cent of paediatrics). A similar observation was made by Diom et al., who reported slight male predominance, with a sex ratio of 1.44.Reference Diom, Cisse, Tall, Ndiaye, Pegbessou and Ndiaye9 This is comparable to the 1.4:1 male:female ratio among children in a study by Kuo et al.Reference Kuo, Shiao, Yung, Sakagami, Sudhoff and Wang10
In our study, ear discharge was the predominant symptom in both the groups. In both paediatric and adult cases of cholesteatoma, the left side (56 per cent) was more commonly involved than the right side (28 per cent). The mean duration of symptoms was 5.48 years in children and 13.6 years in adults. Further, the mean duration of decreased hearing was 5.271 years in children and 10.8 years in adults. Palva et al. reported the average duration of disease to be 5.8 years in children and 21.8 years in adults.Reference Palva, Karma and Karja11
Attic destruction was the most common finding observed via otoscopy and examination under the microscope, in both groups (20 adults and 17 paediatrics). Posterosuperior retraction pocket was more common in children (13 cases) as compared to adults (11 cases). Chang and Kim also reported the retraction pocket in the attic or posterosuperior quadrant as a hallmark of acquired cholesteatoma.Reference Chang and Kim12
In our study, moderate hearing loss was the most common presentation. The mean bone conduction threshold was found to be 50.56 dB in children and 54.24 dB in adults. The mean air conduction threshold was 26.24 dB in children and 24.16 dB in adults. The average air–bone gap was 24.32 dB in paediatrics and 30.24 dB in adults. A similar observation was made by Rosito et al., who documented higher air conduction and bone conduction thresholds in adults than in children.Reference Rosito, Netto, Teixeira and da Costa13
Intra-operatively, one paediatric case (4 per cent) presented with a post-aural abscess and absent bony cortex over the antrum. A similar observation was made by Canter, who reported a higher incidence of mastoiditis and mastoid abscesses in children.Reference Canter, Kerr and Booth14 In our study, granulations with cholesteatoma was more common in children (6 out of 25) than in adults (2 out of 25). Involvement of the attic, aditus and antrum was seen in all cases. Involvement of the middle ear, sinus tympani, hypotympanum and Eustachian tube area was more common in paediatrics than in adults. A similar observation was made by Shrestha and Kafle; in their study, involvement of the whole tympanic cavity and mastoid was seen in 82 per cent of the children.Reference Shrestha and Kafle15 Edfeldt et al. also compared the disease extent between paediatrics and adults, and observed that the involvement of pars tensa by the disease was more common in children (63 per cent) than in adults (40 per cent). Middle-ear cholesteatoma was seen in 12 per cent of paediatric cases as compared to 8 per cent of adult cases.Reference Edfeldt, Kinnefors, Stromback, Kobler and Andersen16
Out of a total of 50 cases, dural plate dehiscence was seen in 18 cases in our study (10 adults and 8 paediatrics). Sinus plate dehiscence was more common in children (seven cases) than in adults (two cases). This was in contrast to a study by Shreshtha and Kafle, which found dural plate dehiscence in 40 per cent and sinus plate dehiscence in 12 per cent of paediatric cholesteatoma cases.Reference Shrestha and Kafle15
Facial canal dehiscence was more common in adults (36 per cent) than in children (12 per cent). The tympanic segment was the most commonly involved region. This is comparable to the study by Moody and Lambert, who reported dehiscence of the tympanic segment in a total of 18.8 per cent of cases (78 out of 416), and was more common in adults.Reference Moody and Lambert17 In another study, by Shinnabe et al., fallopian canal dehiscence was observed in 6 out of 37 ears (16.8 per cent) in the paediatric group (aged 4–14 years) and in 91 out of 273 ears (33.3 per cent) in the adult group (aged 15–84 years).Reference Shinnabe, Yamamoto, Hara, Hasegawa, Matsuzawa and Kanazawa18
In our study, lateral semicircular canal dehiscence was more common in adults (8 out of 25, 32 per cent) than in children (4 out of 25, 12 per cent). This is in contrast to a study by Lynrah et al., which reported an increased incidence of fistula in children (16.7 per cent) than in adults (6.7 per cent).Reference Lynrah, Bakshi, Panda and Khandelwal19 The shorter duration of history and earlier presentation to the hospital might be responsible for the lower incidence of complications in the paediatric group.
Ossicular involvement by the disease, either partial or complete, was invariably seen in both groups in our study. The incus was the most commonly involved ossicle. Involvement of the stapes suprastructure was more common in children (16 out of 25, 64 per cent) than in adults (8 out of 25, 32 per cent). A similar observation was made by Dornellas et al., who reported involvement of the ossicular chain in 100 per cent of paediatric cases and in 92 per cent of adult cases.Reference Dornelles, da Costa, Meurer and Schweiger20Another study, by Mohammadi et al., also found the most common pathology to be incus erosion, which was complete in 55.4 per cent and partial in 30.7 per cent of cases. Stapes suprastructure involvement was more common than a total loss of bone (40.9 per cent vs 25.9 per cent, respectively).Reference Mohammadi, Naderpour and Mousaviagdas21
• Paediatric cholesteatoma generally presents with extensive disease spread and ossicular destruction at presentation
• Limited data are available comparing paediatric with adult cholesteatoma, with most evidence being retrospective and from the Western world
• In this study, cholesteatoma was more extensive in paediatrics, with more ossicular destruction, while the complication rate was higher in adults
• High inflammatory mediator levels and angiogenesis in cholesteatoma peri-matrix, along with granulations, are responsible for more aggressive paediatric cholesteatoma
Following surgery, the mastoid cavity healed in all cases except one, within six months. The mean duration of healing was found to be 3.4 months in children and 2.96 months in adults. This is attributed to the epithelisation of the cavity. A similar observation was made by East et al., who reported epithelisation of the cavity in six weeks in 7 out of 10 cases, using a vascularised temporoparietal fascia flap.Reference East, Brough and Grant22
Conclusion
Our study showed that paediatric cholesteatoma was more extensively spread in the temporal bone and mastoid cavity, with a greater extent of ossicular destruction, as compared to adult cholesteatoma; this is consistent with data from other parts of the world. This can possibly be attributed to a higher incidence of granulations in paediatric cases, and the associated increase in inflammatory and enzymatic activity in the peri-matrix.
However, the same result was not reflected in terms of complication rate. Complications such as facial canal dehiscence and semicircular canal dehiscence were more common in adults, possibly owing to early presentation and a shorter duration of symptoms. As such, the author recommends further studies involving a larger sample size from a widespread stratum of the population, to provide substantiate evidence.
In addition, as the mastoidectomy cavity was found to heal within six months of canal wall down mastoidectomy in almost all cases, the average number of post-operative visits by the patients can be estimated and pre-operative counselling can be carried out accordingly.
Competing interests
None declared
