|
| |
| ORIGINAL ARTICLE |
|
| Ahead of print publication |
|
|
Comparative analysis of preoperative high-resolution computed tomography temporal bone and intraoperative surgical findings in cases of cholesteatoma
Ranjit Singh Lahel1, Shailendra Tripathi2, Amit Chail3
1 Department of Radiodiagnosis, Military Hospital, Sagar, Madhya Pradesh, India
2 Department of ENT, Military Hospital, Secunderabad, Telangana, India
3 Department of Psychiatry, Command Hospital, Chandimandir, Haryana, India
| Date of Submission | 14-Apr-2022 |
| Date of Decision | 22-Jun-2022 |
| Date of Acceptance | 05-Jul-2022 |
| Date of Web Publication | 05-Oct-2022 |
Correspondence Address:
Ranjit Singh Lahel,
Department of Radiodiagnosis, Military Hospital, Sagar, Madhya Pradesh
India
 Source of Support: None, Conflict of Interest: None DOI: 10.4103/jmms.jmms_64_22
Introduction: Cholesteatoma has been described as an accumulation of desquamated keratin debris in a pocket of squamous epithelium over a matrix within the middle ear cleft. A prospective study was carried out to establish a correlation between preoperative high-resolution computed tomography (HRCT) temporal bone findings with intraoperative findings in cases of cholesteatoma. Methods: Fifty patients of chronic otitis media, active squamous type, i.e. clinically suspected cholesteatoma, scheduled to undergo tympanomastoid exploration were selected. The positive and negative findings of the scan were documented as per the parameters to be analyzed. The important intraoperative surgical findings were then analyzed with preoperative HRCT findings. Results: Agreement between surgical and HRCT findings ranged from 86% (facial canal erosion) to 100% (external bony canal involvement). HRCT had higher detection rate as compared to surgery for scutum erosion (40% vs. 36%), tegmen erosion (6% vs. 4%), facial canal erosion (28% vs. 14%), lateral semicircular canal erosion (6% vs. 2%), incus erosion (78% vs. 64%), sinus tympani involvement (14% vs. 4%), attic involvement (44% vs. 42%), mastoid involvement (88% vs. 80%), and aditus widening (54% vs. 44%). The higher detection of scutum erosion, tegmen erosion, facial canal erosion, lateral semicircular canal erosion, and incus erosion is probably due to partial volume averaging. Conclusion: The findings of the present study showed that HRCT is an essential and indispensable tool for preoperative evaluation of cholesteatoma. It was highly sensitive and adequately specific for a significant number of findings. These findings are pivotal in planning the surgery and preparing for necessary anticipatory reconstructive procedures.
Keywords: Cholesteatoma, chronic otitis media, high-resolution computed tomography temporal bone, mastoid erosion, scutum erosion, stapes erosion
How to cite this URL:
Lahel RS, Tripathi S, Chail A. Comparative analysis of preoperative high-resolution computed tomography temporal bone and intraoperative surgical findings in cases of cholesteatoma. J Mar Med Soc [Epub ahead of print] [cited 2023 Mar 26]. Available from: https://www.marinemedicalsociety.in/preprintarticle.asp?id=357922 |
| Introduction | |  |
Abramson has elaborately defined cholesteatoma where “cholesteatoma is a three-dimensional epidermal and connective tissue structure, usually in the form of a sac and frequently conforming to the architecture of various spaces of the middle ear, attic, and mastoid. This structure has a capacity for progressive and independent growth at the expense of underlying bone and has a tendency to recur after removal.”[1] It is a non-neoplastic lesion derived from abnormal growth of the keratinizing squamous epithelium in the temporal bone.[2] Cholesteatoma is locally invasive as a result of the enzymatic activity of its matrix and hence its pathological sequence can lead to the destruction of structures in the middle ear cleft. It therefore can erode into adjacent structures and can cause intracranial and extracranial complications. Due to the potentially fatal intracranial complications, cholesteatoma remains a cause of morbidity and death for those who lack access to advanced medical care.[3] In India, the prevalence of cholesteatoma in complicated cases of chronic otitis media (COM) has been found to be 65%–75%.[4]
Cholesteatoma can be congenital or acquired. Acquired cholesteatoma can be further differentiated into primary and secondary types. The Primary acquired type is a cholesteatoma, based on retraction theory, arising from either pars tensa or pars flaccida which results in the growth of epithelium into the middle ear (mesotympanum and epitympanum) and may extend to various subsites of middle ear cleft subsequently. Secondary acquired cholesteatoma occurs as a result of a perforated tympanic membrane, surgery, or trauma where the epithelium gets trapped in the middle ear cleft or mastoid bone.[5]
High-resolution computed tomography (HRCT) scan of temporal bone confirms the presence of cholesteatoma within the temporal bone and can be used as complementary to clinical examination. HRCT can focus on specific characteristics for different etiologies and pathologies of cholesteatoma. In attic cholesteatoma, the erosion of the scutum which is the first sign of cholesteatoma can be assessed in the coronal view. Cholesteatoma of pars tensa spreads to the long process of incus and superstructure of stapes. The likelihood of attic cholesteatoma increases with the expansion of aditus ad antrum. The most common site of the labyrinthine fistula is the dome of the lateral semicircular canal and the most commonly affected segment of the facial nerve is the tympanic segment. Hence, HRCT provides diagnostic information which can be utilized for planning the appropriate surgical intervention.
HRCT scanning with a slice thickness of <1 mm provides elusive details of structures in the middle ear cleft. It helps to detect even a small soft-tissue thickening. However, it has a limitation that it cannot confirm the presence of cholesteatoma because it cannot differentiate between soft-tissue thickening and cholesteatoma, which requires surgical correlation or confirmation by magnetic resonance imaging using diffusion-weighted imaging.[6] Surgical treatment is the most common management strategy in most patients, moreover, the imaging results can be validated immediately.[7]
With this contextual information, the present study was planned with an aim to evaluate the findings in a preoperative HRCT scan among patients with cholesteatoma and to corroborate and confirm these findings with intra-operative surgical findings.
| Methods | | |
The hospital-based prospective study was conducted over a period of 3 years at the department of radiodiagnosis in collaboration with the department of otorhinolaryngology, at a tertiary care referral hospital catering to patients from varying demographic profiles.
Fifty patients with clinical suspicion of middle ear cholesteatoma undergoing HRCT temporal bone and willing for subsequent surgical treatment were included in the study. Appropriate approval for conducting the study was obtained from institutional ethics committee. Informed consent was obtained from all patients.
Scans were performed in axial planes with 0.6 mm thin sections in soft tissue and bone windows with coronal and sagittal reconstruction. Imaging characteristics of HRCT scans were recorded in all patients. The intraoperative surgical findings were correlated with preoperative HRCT findings recorded earlier.
The following parameters were evaluated:
- Mesotympanum involvement
- Mastoid involvement
- Scutum erosion
- Tegmen erosion
- Facial canal erosion
- Lateral semicircular canal erosion
- Malleus, incus, and stapes erosions
- Sinus tympani involvement
- External bony canal involvement
- Facial canal dehiscence
- Attic involvement
- Aditus widening.
Aditus widening does not have any clearly defined set of definitive objective criteria. To eliminate interobserver variation in documenting the same, the CT scans were reported by the same radiologist with a comparable assessment with the scan of a contralateral ear; similarly, the cases were operated by the same surgeon, and the findings were recorded. The extent of correlation between HRCT and surgical findings was evaluated using the kappa statistic. The diagnostic efficacy of HRCT for different parameters was evaluated in terms of sensitivity, specificity, positive predictive value, negative predictive value, and accuracy. The confidence level of the study was 95% and a P < 0.05 indicated a statistically significant association.
| Results | | |
A total of 50 patients as defined by inclusion and exclusion criteria were enrolled in the study population. The age of patients ranged from 5 to 70 years. The maximum number of patients was in the age group of 11–20 years (26%), followed by 41–50 years (24%), 21–30 years (20%), and 31–40 years (14%). There were 2 (4%) cases in the age group 0–10 years, 2 (4%) in 51–60 years, and 4 (8%) in 61–70 years. The mean age of patients was 32.22 ± 16.38 years.
There were 20 (40%) males and 30 (60%) females. The male to female ratio of the study population was 0.66. There were 24 (48%) patients with left-side involvement and 26 (52%) patients with right-side involvement.
On HRCT, the most common finding was mesotympanum involvement (90%), followed by mastoid involvement (88%), incus erosion (78%), stapes erosion (56%), aditus widening (54%), attic involvement (44%), scutum erosion (40%), facial canal erosion (28%), tegmen erosion (6%), external bony canal involvement (6%) and lateral semicircular canal erosion (6%), and malleus erosion (2%).
On surgery, the most common finding was mesotympanum involvement (92%) followed by mastoid involvement (80%), incus erosion (64%), stapes erosion (60%), aditus widening (54%), attic involvement (42%), scutum erosion (36%), facial canal erosion (14%), external bony canal involvement (6%), tegmen erosion (4%), lateral semicircular canal erosion (2%), and malleus erosion (2%).
Agreement between surgical and HRCT findings ranged from 86% (facial canal erosion) to 100% (external bony canal involvement).
HRCT had higher detection rate as compared to surgery for scutum erosion (40% vs. 36%), tegmen erosion (6% vs. 4%), facial canal erosion (28% vs. 14%), lateral semicircular canal erosion (6% vs. 2%), incus erosion (78% vs. 64%), sinus tympani involvement (14% vs. 4%), attic involvement (44% vs. 42%), mastoid involvement (88% vs. 80%), and aditus widening (54% vs. 44%). The higher detection of scutum erosion, tegmen erosion, facial canal erosion, incus erosion, and lateral semicircular canal erosion in HRCT while less so in intraoperative findings is probably explained by partial volume averaging where there is no actual erosion.
HRCT had lower detection rate as compared to surgery for stapes erosion (56% vs. 60%) and mesotympanum involvement (90% vs. 92%). The detection rate was similar for two techniques for external bony canal involvement (6%) and malleus erosion (2%).
HRCT had 100% sensitivity for all findings except for stapes erosion (93.33%) and mesotympanum involvement (97.83%). HRCT had the least specificity for mastoid involvement (60%) and 100% specificity for stapes erosion, external bony canal, malleus erosion, and mesotympanum involvement. Except for five findings, namely facial canal erosion, incus erosion, sinus tympani involvement, mastoid involvement, and aditus widening, for all the other findings, HRCT had a specificity >90%.
Illustrations of HRCT findings as seen in various patients with cholesteatoma in our study are presented as shown in [Figure 1],[Figure 2],[Figure 3] and an intraoperative image [Figure 4]. | Figure 2: Axial section showing cholesteatoma engulfing and eroding ossicles with extension into stapes footplate region and sinus tympani
Click here to view |
 | Figure 3: Coronal reformat showing cholesteatoma in right Prussak's space eroding malleus and tip of scutum
Click here to view |
 | Figure 4: Right ear canal wall down surgery done, showing: facial ridge (1), bony facial canal in the attic (horizontal part – intact) (2), and cholesteatoma engulfing the ossicles and extending into the region of stapes footplate (arrow)
Click here to view |
| Discussion | | |
Cholesteatoma has been described as an accumulation of desquamated keratin debris in a pocket of squamous epithelium over a matrix within the middle ear cleft. Cholesteatoma occurs in the middle ear and can extend to aditus ad antrum. It has the potential to cause serious petrous complications.
The current study emphasizes the role of HRCT scan in cases of chronic otitis media with suspected cholesteatoma. Most of the studies primarily cater to the local population, which severely limits the representation of diverse demographics. Our study factored in the point that being a tertiary care referral hospital, patients represented an assorted demographic profile from various states, varied ethnicity, and wide-ranging environmental exposure. Thus, it adds to the data pool available on the topic
The accomplishment of a surgical involvement depends on preoperative know-how of the extent of different pathologies and thereafter determining the approach and type of intervention required. HRCT of temporal bone provides confirmation of the extent of involvement and is pivotal for diagnosing cholesteatoma developing behind a closed tympanic membrane. Hence, surgical management is often preceded by imaging to know the extent of the underlying pathology and to plan surgical intervention accordingly.
Age
The findings of the present study are in sync with outcomes in different series from India that have reported an increased prevalence of cholesteatoma among younger patients.[8]
Location
The most common finding with respect to the location of cholesteatoma was mesotympanum involvement (90%) followed by mastoid involvement (88%), attic involvement (44%), and external bony canal involvement (6%), respectively. The involvement of attic, antrum, and sinus tympani was noticed more during HRCT and the same could not be corroborated with intraoperative findings. It is explained by the fact that the HRCT scan cannot differentiate between cholesteatoma and mucosal thickening or secretions.
Bony involvement
The most common finding was incus erosion (78%), stapes erosion (56%), aditus widening (54%), scutum erosion (40%), facial canal erosion (28%), tegmen erosion (6%), and lateral semicircular canal erosion (6%) while malleus was least involved bone (erosion seen in approx. 2% of cases).
In spite of the variability in findings in different studies, the consistent finding brought out by the majority is that the malleus is the most resistant bone in terms of erosion and the same was evident in our study too.[9]
Dashottar et al. reported involvement of mesotympanum in 48 patients (96%) as the most common finding on HRCT, followed by mastoid involvement in 40 patients (80%) and aditus widening in 24 (48%) patients, which is comparable to the results of the present study.[8] Sreedhar et al. documented findings as per anatomical factors and reported the attic involvement (80%), mastoid involvement (76%), and mesotympanum involvement (56%) as the major location and extent-related findings.[10]
In our study, as far as extent and location of disease expansion were concerned, high accuracy was obtained for mesotympanum involvement which was the most common finding and was confirmed surgically in 92% of cases. The finding was comparable to that of the observation of Sreedhar et al. who also obtained an absolute agreement between HRCT and intraoperative findings for mesotympanum involvement.[10] However, in their study, an absolute agreement was observed for mastoid involvement and aditus widening too. However, in the present study, we found this agreement to the extent of 92% and 90%, respectively, for mastoid involvement and aditus widening. For mastoid involvement and attic involvement, an accuracy level of 92% and 98%, respectively, and sensitivity as high as 100% was achieved in our study. The mastoid involvement however lacked adequate specificity (60%).
For bony and structural abnormalities, we obtained a high level of agreement positive predictive value (PPV >90) between HRCT and intraoperative findings for stapes erosion (PPV 100), external bony canal involvement (PPV 100), malleus erosion (PPV 100), incus erosion (PPV 94.12), and scutum erosion (PPV 90). For all these, the sensitivity of HRCT ranged from 93.33% to 100% and specificity ranged from 88.89% to 100%. However, for findings such as sinus tympani involvement (PPV 28.57), lateral semicircular canal erosion (PPV 33.33), facial canal dehiscence (PPV 50), and tegmen erosion (PPV 66.67), the positive predictive values were less and level of agreement with the intraoperative findings was suboptimal. Similar low sensitivity and specificity for findings such as facial canal dehiscence and stapes erosion were also reported by Ng et al. in their series of 32 patients only.[11] In another study conducted in 35 patients, Shah et al. also found low specificity (66.7%) of HRCT for the evaluation of incus bone erosion; however, their study showed 81.3% sensitivity and 94.1% specificity for malleus and 80% sensitivity and 95.7% specificity for stapes, respectively.[12] In a smaller series, Sarin et al. found the sensitivity and specificity of stapes erosion as 53.3% sensitive and 60% specific and lateral semicircular canal (SCC) erosion 50% sensitive and 92.1% specific.[13] In their study, they also reported sensitivity levels lower than 50% for facial nerve dehiscence and tegmen erosion.
| Conclusion | | |
The findings of the present study showed that HRCT is an essential and indispensable tool for preoperative evaluation of cholesteatoma. It was highly sensitive and adequately specific for a significant number of findings. These findings are pivotal in planning the surgery and preparing for necessary anticipatory reconstructive procedures during the surgery.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
| References | | |
| 1. | Abramson M. Cholesteatoma pathogenesis: Evidence for the migration theory. In: McCabe B, Sade J, Abramson M, editors. Cholesteatoma, First International Conference. Birmingham: Aesculapis; 1977. p. 176-86.  |
| 2. | Robinson JM. Cholesteatoma: Skin in the wrong place. J R Soc Med 1997;90:93-6. |
| 3. | Prescott CA. Cholesteatoma in children--the experience at the red cross war memorial children's hospital in South Africa 1988-1996. Int J Pediatr Otorhinolaryngol 1999;49:15-9. |
| 4. | Gupta A, Agarwal SR. A study of prevalence of cholesteatoma in complications of suppurative otitis media. Indian J Otolaryngol Head Neck Surg 1998;50:140-6. |
| 5. | Kuo CL. Etiopathogenesis of acquired cholesteatoma: Prominent theories and recent advances in biomolecular research. Laryngoscope 2015;125:234-40. |
| 6. | Harnsberger R, Hudgins R, Wiggins P. Diagnostic Imaging: Head and Neck. Salt Lake City, Utah: Amirsys; 2004. p. 188-89. |
| 7. | Jahromi AM, Ariani FH, Arabkhani R, Ariani SH. Evaluation of cholesteatoma frequency in patients with chronic otitis media. Iran J Otorhinolaryngol 2010;22:21-4. |
| 8. | Dashottar S, Bucha A, Sinha S, Nema D. Preoperative temporal bone HRCT and intra-operative findings in middle ear cholesteatoma: A comparative study. Int J Otorhinolaryngol Head Neck Surg 2019;5:77-82. |
| 9. | Albera R, Canale A, Piumetto E, Lacilla M, Dagna F. Ossicular chain lesions in cholesteatoma. Acta Otorhinolaryngol Ital 2012;32:309-13. |
| 10. | Sreedhar S, Pujary K, Agarwal AC, Balakrishnan R. Role of high-resolution computed tomography scan in the evaluation of cholesteatoma: A correlation of high-resolution computed tomography with intra-operative findings. Indian J Otol 2015;21:103. [Full text] |
| 11. | Ng JH, Zhang EZ, Soon SR, Tan VY, Tan TY, Mok PK, et al. Pre-operative high resolution computed tomography scans for cholesteatoma: Has anything changed? Am J Otolaryngol 2014;35:508-13. |
| 12. | Shah C, Shah P, Shah S. Role of HRCT temporal bone in pre-operative evaluation of cholesteatoma. Int J Med Sci Public Health 2014;3:69-72. |
| 13. | Sarin V, Kaur T, Sood AS. Role of pre-operative high resolution computed tomography of temporal bone in squamous chronic otitis media. Pak J Otolaryngol 2015;31:51. |
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
|
