Key Points
Question
For congenital cholesteatoma (CC) limited to the middle ear and/or mastoid antrum in children, is the use of total transcanal endoscopic resection associated with a reduced risk of residual cholesteatoma?
Findings
In this cohort study of 271 children with CC limited to the middle ear and/or mastoid antrum, the incidence of residual cholesteatoma after transcanal endoscopic ear surgery was 13.3%, with a favorable surgical outcome.
Meaning
This study’s findings suggest that transcanal endoscopic ear surgery may be an effective alternative for children with CC limited to the middle ear and/or mastoid antrum.
This cohort study assesses outcomes of endoscopic ear surgery in children with congenital cholesteatoma of the middle ear and mastoid antrum.
Abstract
Importance
Transcanal endoscopic ear surgery (TEES) provides minimally invasive transcanal access to the middle ear and improves middle ear visibility during cholesteatoma resection. However, the literature on outcomes following TEES alone for the removal of congenital cholesteatoma (CC) is lacking and limited to small series.
Objective
To assess outcomes of TEES for CC limited to the middle ear and/or mastoid antrum and to explore the risk factors associated with recidivism (ie, recurrent and/or residual cholesteatoma).
Design, Setting, and Participants
This cohort study evaluated retrospective, multicenter data for 271 children with CC who underwent TEES at 9 tertiary referral hospitals in South Korea between January 1, 2013, and December 31, 2021, and had a follow-up of at least 6 months after surgery.
Main Outcomes and Measures
Outcomes included the incidence of residual cholesteatoma and audiometric data after TEES. A multivariable analysis using Cox proportional hazards regression models was used to assess associations between cholesteatoma characteristics and recidivism, with hazard ratios (HRs) and 95% CIs reported.
Results
Of the 271 patients (mean [SD] age, 3.5 [2.9] years; 194 [71.6%] boys, 77 [28.4%] girls), 190 had Potsic stage I CC (70.1%), 21 (7.7%) had stage II, 57 (21.0%) had stage III, and 3 (1.1%) had stage IV. Thirty-six patients (13.3%) with residual cholesteatoma were found, including 15 (7.9%) with Potsic stage I, 3 (14.3%) with stage II, and 18 (31.6%) with stage III. In the multivariable analysis, invasion of the malleus (HR, 2.257; 95% CI, 1.074-4.743) and posterosuperior quadrant location (HR, 3.078; 95% CI, 1.540-6.151) were associated with the incidence of recidivism. Overall, hearing loss (>25 dB on auditory behavioral test or >30 dB of auditory evoked responses) decreased from 24.4% to 17.7% after TEES.
Conclusions and Relevance
This cohort study involved the largest known population to date of CC removed by TEES. The findings suggest that TEES may be feasible and effective for the removal of CC limited to the middle ear and/or mastoid antrum in children.
Introduction
Congenital cholesteatoma (CC) is an uncommon condition, accounting for approximately 2% to 15% of all cholesteatomas.1 The reported incidence of CC, however, has been increasing, which might be due to early identification and increased awareness of cholesteatomas.2,3 The improvement in and widespread use of diagnostic tools, such as endoscopes and microscopes, in local clinics have likely played an important role in the early identification of CCs by primary care physicians, pediatricians, and otolaryngologists. The treatment of choice for CC is complete surgical removal, avoiding damage to the normal structures, and prevention of recurrence. The traditional method of middle ear cholesteatoma surgery is performed under a microscope. The transcanal approach with a microscope provides a magnified image of excellent quality in a straight line, but it is difficult to view cholesteatomas hidden in the anterior epitympanic recess, tympanic sinus, facial recess, and hypotympanum. The approach can be technically more challenging in pediatric patients due to the narrowest segment of the ear canal and generally small size of the ear. An inability to visualize or access the limits of cholesteatoma can lead to incomplete resection of the cholesteatoma matrix from underlying bone, which contributes to residual cholesteatoma.4
With advances in endoscopic technology, the role of transcanal endoscopic ear surgery (TEES) has gained more attention in managing cholesteatoma, particularly in pediatric populations.5,6,7,8,9 The angled endoscope provides wide dynamic visualization, which aids in looking for hidden areas in the middle ear cavity during CC removal. The improved visualization offered by the endoscope may also obviate a large incision (postauricular or endaural incision), which provides minimally invasive transcanal access to the middle ear. Numerous studies have concluded that pediatric CC confined to the middle ear can be successfully managed with TEES,5,6,7,8 but these series were limited by small case numbers. Because endoscopic surgery is completed by using only 1 hand, as the other hand is required to hold the endoscope, ear surgeons must consider that the postauricular microscopic approach improves the probability of complete CC clearance by using 2 hands for cholesteatoma dissection. This study assessed the outcomes of TEES for CC to determine the clinical efficacy of TEES in pediatric patients with CC.
Methods
Study Design and Participants
This cohort study evaluated retrospective, multicenter data of patients who underwent TEES as a primary surgery for CC at 9 tertiary referral hospitals in South Korea from January 1, 2013, to December 31, 2021. Congenital cholesteatoma was defined by the presence of cholesteatoma medial to an intact tympanic membrane (TM) in a patient without a history of TM perforation or otorrhea. We did not consider the history of otitis media with effusion.10 Patients who had a follow-up of at least 6 months after surgery were included. Patients who required an operating microscope to perform mastoidectomy or atticotomy were excluded. This study was approved by the institutional review board of Dankook University Hospital, and written informed consent was waived owing to the retrospective nature of the study. The study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Data Collection
The following data were collected: demographic information (age and sex), characteristics of the CC at presentation, surgical management, incidence of residual or recurrent cholesteatoma, hearing results before and after surgery, and incidence of postoperative complications. The characteristics of CC were assessed based on operative findings. The type of CC was classified as a closed cyst (encapsulated cholesteatoma) or open infiltrative cyst (cholesteatoma matrix in direct contact with the middle ear mucosa) according to the morphology.11,12,13 Potsic stage was determined as follows: stage I, cholesteatoma confined to a single quadrant; stage II, cholesteatoma in multiple quadrants but without ossicular involvement or mastoid extension; stage III, ossicular involvement without mastoid extension; and stage IV, mastoid involvement.14 Pre- and postoperative hearing thresholds were assessed by auditory brainstem responses, auditory steady state responses (ASSRs), and pure tone audiometry (PTA). Postoperative audiometric data were collected at least 1 month after the surgery. The hearing threshold for ASSR and PTA was calculated as the average of 500-, 1000-, 2000-, and 4000-Hz frequencies. Normal hearing was defined as an air conduction (AC) threshold of 25-dB hearing loss or less for PTA15 and 30-dB normal hearing or less for auditory brainstem response and ASSR.16,17 Residual cholesteatoma was defined as cholesteatoma in a subsequent surgery in the same ear. Residual cholesteatoma was differentiated from recurrent cholesteatoma by the absence of continuity with TM retraction or a potential implantation event between the date of primary surgery and the identification of recidivism (ie, recurrent and/or residual cholesteatoma). Postoperative complications during the follow-up period, including TM perforation, TM adhesion, otorrhea, dizziness, facial paralysis, wound infection, and other complications, were also recorded.
Statistical Analysis
A Cox proportional hazards regression model was used to conduct univariable and multivariable analyses of risk factors associated with recidivism. Variables with P < .05 in the univariable analysis were included in the multivariable model, and the backward likelihood ratio method was used for analysis. Statistical significance was defined as a 2-sided P < .05. Statistical analyses were performed using SPSS, version 28.0 (IBM Corporation).
Results
Patient Characteristics
A total of 271 patients with CC met the inclusion criteria for this study. The patient characteristics are summarized in Table 1. The mean (SD) age at the time of the first surgery was 3.5 (2.9) years. There were 194 boys (71.6%) and 77 girls (28.4%) in the study population, and 150 right ears (55.4%) and 121 left ears (44.6%) were affected. Audiometric data were available for 161 patients (59.4%) preoperatively and 168 (62.0%) postoperatively. The mean (SD) length of follow-up was 29.7 (22.2) months. During the follow-up, 114 patients had a temporal bone computed tomography scan.
Table 1. Characteristics of Patients Undergoing Transcanal Endoscopic Ear Surgery (TEES).
| Characteristic | TEES, No. (%) |
|---|---|
| No. of patients | 271 |
| Age, mean (SD), y | 3.5 (2.9) |
| Sex | |
| Male | 194 (71.6) |
| Female | 77 (28.4) |
| Right ear | 150 (55.4) |
| Left ear | 121 (44.6) |
| Preoperative auditory evaluation | |
| Not done | 110 (40.6) |
| Auditory evoked responses (ABR and/or ASSR) | 94 (34.7) |
| Auditory behavioral test (PTA) | 67 (24.7) |
| Postoperative auditory evaluation | |
| Not done | 103 (38.0) |
| Auditory evoked responses (ABR and/or ASSR) | 50 (18.5) |
| Auditory behavioral test (PTA) | 118 (43.5) |
| Postoperative temporal bone CT scan | 114 (42.1) |
| Follow-up, mean (SD), mo | 29.7 (22.2) |
Abbreviations: ABR, auditory brainstem response; ASSR, auditory steady state response; CT; computed tomography; PTA, pure tone audiometry.
CC Characteristics
There were 190 cases (70.1%) of Potsic stage I, 21 (7.7%) of stage II, 57 (21.0%) of stage III, and 3 (1.1%) of stage IV disease. All CCs classified as Potsic stage IV were limited to the antrum. The locations of the CCs in the tympanic cavity are shown in Table 2. Among the CCs involving 1 quadrant, the cholesteatomas were mostly found in the anterosuperior quadrant (ASQ) (172 of 220; 78.2%). When the CCs involved multiple quadrants, the posterosuperior quadrant (PSQ) was the most frequently involved location (38 of 51; 74.5%), followed by the ASQ (36 of 51; 70.6%). Based on intraoperative morphologic data, 197 ears (72.7%) had closed-type masses, and the other 74 ears (27.3%) had open-type masses. As defined, there were no cases of ossicular involvement classified as stage I and II, while all stage III and IV cases had ossicular erosion present. Among the CCs involving the ossicular chain, incus erosion was most frequently found intraoperatively, in 52 of 60 (86.7%) patients. Two stage III cases had erosion of the fallopian canal and eroded ossicles.
Table 2. Disease Characteristics.
| Finding at surgery | Patients, No. (%) |
|---|---|
| Potsic stage of disease | |
| I | 190 (70.1) |
| II | 21 (7.7) |
| III | 57 (21.0) |
| IV | 3 (1.1) |
| Location in the tympanic cavity | |
| Single quadrant | 220 (81.2) |
| ASQ | 172 (78.2) |
| PSQ | 20 (9.1) |
| AIQ | 6 (2.7) |
| PIQ | 22 (10.0) |
| Multiple quadrants | 51 (18.8) |
| Involved quadrant | |
| ASQ | 36 (70.6) |
| PSQ | 38 (74.5) |
| AIQ | 19 (37.3) |
| PIQ | 25 (49.0) |
| Type of cholesteatoma | |
| Closed | 197 (72.7) |
| Open | 74 (27.3) |
| Ossicular chain status | |
| Intact | 211 (77.9) |
| Eroded | 60 (22.1) |
| Malleus | 26 (43.3) |
| Incus | 52 (86.7) |
| Stapes | 43 (71.7) |
| Erosion of the fallopian canal | 2 (0.7) |
Abbreviations: AIQ, anteroinferior quadrant; ASQ, anterosuperior quadrant; PIQ posteroinferior quadrant; PSQ, posterosuperior quadrant.
Surgical Treatment
Details of surgery varied by Potsic stage (Table 3). All cases were accessed through the ear canal for endoscopic dissection of CCs, and 8 patients required additional atticotomy for visualization of the epitympanum. Most patients who underwent TEES (90.0%) had the TM preserved. Graft materials included soft tissue (perichondrium or fascia), cartilage, or cellular human dermal allograft (Megaderm; L&C BIO Co, Ltd). The decision to proceed with ossiculoplasty during the primary procedure was based on the surgeon’s judgment. Among 60 patients with an eroded ossicular chain in Potsic stages III to IV, 15 (25.0%) underwent reconstruction as the primary procedure, 20 (33.3%) underwent reconstruction as the secondary procedure, and 25 (41.7%) did not undergo reconstruction. Of the 37 patients who underwent ossicular reconstruction during the primary or secondary procedures, 27 (73.0%) with stage III to IV disease required total ossicular reconstruction (TOR), and 10 underwent partial ossicular reconstruction (POR). Two patients with an intact ossicular chain in Potsic stage I to II had POR after separation of the incudostapedial joint to remove the CC in the ASQ. All but 2 patients underwent ossicular reconstruction with titanium prosthesis. Two patients underwent POR and TOR with cartilage. In most patients (190 of 271; 70.1%), the chorda tympani nerve was identified and preserved during the TEES. However, 10 patients had an injured chorda tympani.
Table 3. Surgical Procedures.
| No. (%) | |||||
|---|---|---|---|---|---|
| Total (n = 271) | Stage I (n = 190) | Stage II (n = 21) | Stage III (n = 57) | Stage IV (n = 3) | |
| Age, mean (SD), y | 3.5 (2.9) | 2.8 (1.6) | 2.9 (1.4) | 5.7 (4.6) | 10.0 (7.2) |
| Surgery length, mean (SD), min | 61.8 (40.4) | 52.6 (32.9) | 71.0 (46.0) | 86.1 (44.9) | 121.3 (97.7) |
| Endoscopic atticotomy | 8 (3.0) | 1 (0.5) | 0 | 6 (10.5) | 1 (33.3) |
| TM reconstruction | 27 (10.0) | 12 (6.3) | 2 (9.5) | 13 (22.8) | 0 |
| Perichondrium graft | 13 (48.1) | 3 (25.0) | 1 (50.0) | 9 (69.2) | 0 |
| Fascia graft | 3 (11.1) | 2 (16.7) | 0 | 1 (7.7) | 0 |
| Cartilage graft | 1 (3.7) | 0 | 0 | 1 (7.7) | 0 |
| Megaderm graft | 10 (37.0) | 7 (58.3) | 1 (50.0) | 2 (15.4) | 0 |
| Ossicular chain reconstruction | |||||
| Not reconstructed | 234 (86.3) | 189 (99.5) | 20 (95.2) | 25 (43.9) | 0 |
| Reconstructed | 37 (13.7) | 1 (0.5) | 1 (4.8) | 32 (56.1) | 3 (100) |
| Primary surgery | 16 (43.2) | 1 (100) | 0 | 13 (40.6) | 2 (66.7) |
| Secondary surgery | 21 (56.8) | 0 | 1 (100) | 19 (59.4) | 1 (33.3) |
| POR or SC | 10 (27.0) | 1 (100) | 1 (100) | 8 (25.0) | 0 |
| TOR | 27 (73.0) | 0 | 0 | 24 (75.0) | 3 (100) |
| Chorda tympani | |||||
| Not identified | 71 (26.2) | 63 (33.2) | 1 (4.8) | 7 (12.3) | 0 |
| Saved | 190 (70.1) | 126 (66.3) | 20 (95.2) | 42 (73.7) | 2 (66.7) |
| Sacrificed | 10 (3.7) | 1 (0.5) | 0 | 8 (14.0) | 1 (33.3) |
Abbreviations: POR, partial ossicular reconstruction; SC, short columellization; TM, tympanic membrane; TOR, total ossicular reconstruction.
Outcomes of TEES
A second surgery was performed for a planned second look, when there was evidence of residual cholesteatoma, or for ossicular reconstruction. The planned second looks were performed by surgeons who routinely do a second look or because the surgeon was aware of the possibility of leaving residual cholesteatoma. Overall, 56 patients (20.7%) underwent a second-look procedure, including 20 (10.5%) with Potsic stage I, 5 (23.8%) with stage II, 30 (52.6%) with stage III, and 1 (33.3%) with stage IV disease (Table 4). Most patients (50 of 56; 89.3%) underwent TEES again, but 6 (10.7%) underwent a second-look procedure via the microscopic approach. Overall, recidivism of cholesteatoma was found in 36 patients (13.3%) during the second-look procedure. There was no evidence of recurrent cholesteatoma. Residual cholesteatoma was found in 15 of 190 (7.9%) patients with stage I disease, 3 of 21 (14.3%) with stage II, and 18 of 57 (31.6%) with stage III. No ears with Potsic stage IV disease had residual or recurrent cholesteatoma. Fourteen patients (5.2%) had postoperative complications, including 5 with TM perforation, 5 with TM adhesion, 1 with otorrhea, 1 with cholesterol granuloma, 1 with dizziness, and 1 with keratoma in the ear canal. No patients had facial paralysis or wound infection.
Table 4. Outcomes of Transcanal Endoscopic Ear Surgery.
| No. (%) | |||||
|---|---|---|---|---|---|
| Total (n = 271) | Stage I (n = 190) | Stage II (n = 21) | Stage III (n = 57) | Stage IV (n = 3) | |
| Follow-up, mean (SD), mo | 29.7 (22.2) | 26.5 (19.3) | 23.0 (14.5) | 42.9 (28.6) | 25.3 (14.0) |
| Second-look operation | 56 (20.7) | 20 (10.5) | 5 (23.8) | 30 (52.6) | 1 (33.3) |
| Endoscopic approach | 50 (89.3) | 19 (95.0) | 5 (100) | 26 (86.7) | 0 |
| Microscopic approach | 6 (10.7) | 1 (5.0) | 0 | 4 (13.3) | 1 (100) |
| Recidivism of cholesteatoma | 36 (13.3) | 15 (7.9) | 3 (14.3) | 18 (31.6) | 0 |
| Postoperative complicationsa | 14 (5.2) | 6 (3.2) | 4 (19.0) | 3 (5.3) | 1 (33.3) |
| Preoperative hearing thresholds | |||||
| Unknown | 110 (40.6) | 85 (44.7) | 9 (42.9) | 16 (28.1) | 0 |
| Normal hearingb | 95 (35.1) | 74 (38.9) | 4 (19.0) | 17 (29.8) | 0 |
| Hearing loss | 66 (24.4) | 31 (16.3) | 8 (38.1) | 24 (42.1) | 3 (100) |
| AC threshold, mean (SD), dB | 27.3 (14.4) | 23.8 (12.0) | 35.9 (18.6) | 31.4 (14.6) | 56.3 (13.1) |
| Postoperative hearing thresholds | |||||
| Unknown | 103 (38.0) | 84 (44.2) | 7 (33.3) | 12 (21.0) | 0 |
| Normal hearingb | 120 (44.3) | 91 (47.9) | 8 (38.1) | 20 (35.1) | 1 (33.3) |
| Hearing loss | 48 (17.7) | 15 (7.9) | 6 (28.6) | 25 (43.9) | 2 (66.7) |
| AC threshold, mean (SD), dB hearing loss | 23.2 (15.7) | 17.7 (11.5) | 28.9 (15.4) | 34.2 (18.5) | 30.4 (7.1) |
Abbreviation: AC, air conduction.
Postoperative complications included tympanic membrane perforation, tympanic membrane adhesion, otorrhea, dizziness, granulation cholesteatoma, and keratoma in the ear canal.
Normal hearing is defined as an AC threshold of 25 dB hearing loss or less for pure tone audiometry and 30 dB normal hearing or less for auditory brainstem response and auditory steady state response.
Audiologic evaluation revealed that 66 patients (24.4%) had preoperative hearing loss, and 48 (17.7%) had postoperative hearing loss. The mean (SD) AC threshold improved from 27.3 (14.4) dB before surgery to 23.2 (15.7) dB after surgery. In Potsic stages I and II, hearing loss decreased from 18.5% before to 10.0% after surgery, with a mean (SD) postoperative AC threshold of 19.5 (12.3) dB (17.7 [11.5] dB for stage I and 28.9 [15.4] dB for stage II). However, patients with Potsic stage III and IV disease still had 45.0% hearing loss before and after surgery. The mean (SD) postoperative AC threshold was 34.2 (18.5) dB in stage III and 30.4 (7.1) dB in stage IV. Audiometric data for the postoperative bone conduction (BC) threshold were available for 100 patients. All but 2 patients had a normal BC threshold (≤25 dB hearing loss). Two patients had a 28.75-dB hearing loss of BC thresholds in the range of 500 to 4000 Hz.
Risk Factors Associated With CC Recidivism
Multivariable analysis using Cox proportional hazards regression models was performed to determine which factors were independently associated with recidivism, including age at surgery, atticotomy, type of cholesteatoma, invasion of ossicles, number of quadrants involved, and location in the tympanic cavity. Results of the univariable and multivariable Cox proportional hazards regression model are summarized in Table 5. Univariable analyses revealed that several covariates were significantly associated with recidivism. Open type of cholesteatoma (hazard ratio [HR], 2.109; 95% CI, 1.094-4.065), invasion of malleus (HR, 2.805; 95% CI, 1.341-5.867), invasion of incus (HR, 2.306; 95% CI, 1.183-4.496), invasion of stapes (HR, 2.985; 95% CI, 1.529-5.824), involvement of multiple quadrants (HR, 3.193; 95% CI, 1.623-6.281), location in PSQ (HR, 3.421; 95% CI, 1.742-6.720), and location in posteroinferior quadrant (PIQ) (HR, 2.227; 95% CI, 1.092-4.544) were significantly associated with a greater risk of recidivism. These 7 significant covariates were included in the multivariable analyses, but the type of cholesteatoma, stage, invasion of the incus, invasion of the stapes, and location in PIQ were removed with the backward likelihood ratio method. The final model revealed that invasion of the malleus and PSQ location of cholesteatoma were significantly associated with a higher hazard of recidivism from all risk factors (HR, 2.257 [95% CI, 1.074-4.743] and 3.078 [95% CI, 1.540-6.151], respectively).
Table 5. Risk Factors Associated With Recidivism After Surgery to Remove Congenital Cholesteatoma.
| Variables | No. (%) with recidivism (n = 36) | Univariable | Multivariable | ||
|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | ||
| Age at surgery, mean (SD), years | 3.5 (2.9) | 1.002 (0.909-1.105) | .96 | NA | NA |
| Atticotomy | |||||
| No | 34 of 263 (12.9) | 1 [Reference] | .35 | NA | NA |
| Yes | 2 of 8 (25.0) | 1.983 (0.475-8.283) | NA | ||
| Type of cholesteatoma | |||||
| Closed | 19 of 197 (9.6) | 1 [Reference] | .03 | NA | NA |
| Open | 17 of 74 (23.0) | 2.109 (1.094-4.065) | NA | ||
| Invasion of malleus | |||||
| No | 26 of 245 (10.6) | 1 [Reference] | .006 | 1 [Reference] | .032 |
| Yes | 10 of 26 (38.5) | 2.805 (1.341-5.867) | 2.257 (1.074-4.743) | ||
| Invasion of incus | |||||
| No | 20 of 219 (9.1) | 1 [Reference] | .01 | NA | NA |
| Yes | 16 of 52 (30.8) | 2.306 (1.183-4.496) | NA | ||
| Invasion of stapes | |||||
| No | 21 of 228 (9.2) | 1 [Reference] | .001 | NA | NA |
| Yes | 15 of 43 (34.9) | 2.985 (1.529-5.824) | NA | ||
| No. of quadrants involved | |||||
| Single quadrant | 21 of 220 (9.5) | 1 [Reference] | <.001 | NA | NA |
| Multiple quadrants | 15 of 51 (29.4) | 3.193 (1.623-6.281) | NA | ||
| Location in ASQ | |||||
| No | 10 of 63 (15.9) | 1 [Reference] | .43 | NA | NA |
| Yes | 26 of 208 (12.5) | 0.744 (0.359-1.544) | NA | ||
| Location in AIQ | |||||
| No | 31 of 246 (12.6) | 1 [Reference] | .37 | NA | NA |
| Yes | 5 of 25 (20.0) | 1.543 (0.598-3.976) | NA | ||
| Location in PSQ | |||||
| No | 20 of 213 (9.4) | 1 [Reference] | <.001 | 1 [Reference] | .001 |
| Yes | 16 of 58 (27.6) | 3.421 (1.742-6.720) | 3.078 (1.540-6.151) | ||
| Location in PIQ | |||||
| No | 25 of 224 (11.2) | 1 [Reference] | .03 | NA | NA |
| Yes | 11 of 47 (23.4) | 2.227 (1.092-4.544) | NA | ||
Abbreviations: AIQ, anteroinferior quadrant; ASQ, anterosuperior quadrant; HR, hazard ratio; NA, not applicable; PIQ, posteroinferior quadrant; PSQ, posterosuperior quadrant.
Discussion
This cohort study represents the largest series to date of CC removed by TEES and reveals a favorable surgical outcome, with a recidivism rate of 13.3% among 271 children with CC limited to the middle ear and/or mastoid antrum. The observed recidivism rate was lower than published estimates with microscopic techniques, which range from 20% to 52%.10,14,18,19 Risk of residual cholesteatoma was associated with a higher Potsic stage.14 In the early stages (I and II), only 8.5% of patients had residual cholesteatoma. Recidivism was low even among patients with advanced-stage CC (III and IV), with a recidivism rate of 30.0%. These rates compare favorably with those of residual cholesteatoma removed by the microscopic technique, ranging from a 13% risk in stage I to 67% in stage IV.14 In a review of 82 cases of microscopic surgery for CC, residual cholesteatomas were encountered in 5% of stage I, 24% of stage II, 44% of stage III, and 64% of stage IV cases.19 The TEES technique improves visualization and can reduce the risk of residual cholesteatoma.20,21 The use of both hands to dissect cholesteatoma, which is difficult while using an endoscope, does not provide an advantage for resection completeness of middle ear and attic cholesteatomas.22,23
Studies have indicated that the status of the ossicles and their reconstruction, the initial stage of CC, and the type of surgical technique are important factors for postoperative hearing restoration in cholesteatoma surgery.19,24,25 In the early stage, there were no cases of ossicular involvement, but 2 patients had PORs after separation of the incudostapedial joint for cholesteatoma eradication. The audiologic evaluation of early-stage cholesteatoma revealed excellent audiometric results. Hearing loss decreased from 18.5% to 10.0% after surgery, with a mean postoperative AC threshold of 19.5 dB (17.7 dB for stage I and 28.9 dB for stage II). Advanced-stage cholesteatoma usually causes significant ossicle destruction, especially in the incus, and consistent results were observed in this study. Among 60 patients with an eroded ossicular chain in Potsic stage III and IV, 27 underwent TOR, and 8 underwent POR. As previously noted, more extensive cholesteatoma is associated with worse hearing outcomes. Patients with Potsic stage III and IV still had 45.0% hearing loss after surgery. The mean postoperative AC threshold was 34.2 dB in Potsic stage III and 30.4 dB in Potsic stage IV. Two patients had decreased BC thresholds after surgery, with a mean hearing loss of 28.75 dB in the range of 500 to 4000 Hz. The observed hearing outcomes of advanced-stage cholesteatoma in our cohort were worse than those reported by Stapleton et al,19 which were 25 dB in stage III and 35.5 dB in stage IV. The short-term follow-up of postoperative audiometric data collected starting 1 month after TEES could explain this finding. Long-term hearing outcomes were not observed in our cohort, and 25 patients did not undergo ossicular reconstruction. In addition, 20% of advanced-stage cholesteatomas did not have postoperative audiologic evaluation.
Risk factors for residual cholesteatoma include open-type cholesteatoma, stapes involvement, surgical procedure, and multiple quadrant locations.13,18,19 To determine the risk factors of recidivism, some relevant factors were studied, including age at surgery, surgical procedure (eg, atticotomy), type of cholesteatoma, invasion of ossicles, number of quadrants involved, and location in the tympanic cavity. Of these, the multivariable Cox proportional hazard regression analysis revealed that invasion of the malleus and a PSQ location of cholesteatoma were independent risk factors associated with the incidence of cholesteatoma recidivism.
In this study, CCs most frequently developed in the ASQ of the tympanic cavity and mainly extended to the PSQ. Thus, most CCs involving multiple quadrants presented as a mass medial to the malleus in a dumbbell shape. If the cholesteatoma is strongly attached to the medial side of the malleus handle, it could be challenging to dissect the cholesteatoma from the malleus handle without overstimulating the inner ear.7,19 To ensure complete removal of the cholesteatoma from the malleus, the ossicular chain can be disarticulated and ossicular reconstruction performed. A small hearing benefit may be achieved by preserving the intact ossicular chain compared with greater postoperative morbidity in pediatric cholesteatoma surgery.8,26 Use of an endoscope can afford a close-up view around the ossicles to better visualize areas where the cholesteatoma may have extended and help with the decision of whether to remove the ossicles.
Careful dissection and inspection of the stapes are crucial to ensure complete removal of the cholesteatoma in the PSQ because cholesteatoma enveloping, abutting, or eroding the stapes may harbor residual cholesteatoma.19 If the stapes superstructure is eroded, the cholesteatoma matrix can be adhered to the footplate and over the tympanic segment of the facial nerve. Univariable Cox proportional hazards regression analysis revealed that incus and stapes invasions were significantly associated with a higher risk of residual cholesteatoma, but these factors were removed in the multivariable analysis. When a CC occupies a large portion of the tympanic cavity and envelops the ossicles, it obscures the degree of erosion. Furthermore, potential exists for variable interpretation of ossicular involvement, which may imply contact between the cholesteatoma and an ossicle, ossicular erosion, or a surgeon’s judgment that ossicle removal is required. Cholesteatoma in the PSQ, where the incus and stapes are located, can be a significant independent risk factor for recurrence.
This study included 3 cases of CC extending into the mastoid antrum, none of which had residual or recurrent cholesteatoma for at least 1 year after TEES. Congenital cholesteatoma limited to the mastoid antrum can often be removed effectively with TEES with endoscopic ear instruments. With TEES growing in popularity, curved suctions, angled dissections, ultrasonic bone curette, and flexible steering tips are being developed to access all areas visualized by endoscopy. The posterior aspect of the lateral semicircular canal is often considered a guide for an achievable limit to TEES dissection. Because the larger atticotomy or atticoantrostomy required for access can be difficult to reconstruct effectively, at least in children, further study is warranted on the comparative surgical outcome of TEES and the posterior approach with a microscope in cholesteatoma extended to the mastoid antrum.
Limitations
This study has several limitations. First, although all patients had more than 6 months of follow-up to detect residual cholesteatoma, they may have recidivism that has not yet presented or is otherwise unknown by their surgical teams due to small size. Second, the second-look procedure, the criterion standard for detection of residual cholesteatoma, was not performed for all patients, and residual cholesteatomas may therefore have been missed. This series of CCs removed by TEES includes cases from multiple surgeons across 9 institutions, and the skill and decision-making of the surgeons in this cohort could not be controlled.
Conclusions
This cohort study involves, to our knowledge, the largest population to date of patients with CCs removed by TEES. The findings suggest that TEES may be effective in treating CC limited to the middle ear and/or mastoid antrum in children. Recidivism was low even for advanced stages. Based on our findings, cholesteatoma invasion of the malleus and presence in the PSQ of the tympanic cavity may be associated with significantly higher residual rates. These results may help to guide surgeons to achieve optimal results for patients with CC.
Data Sharing Statement
References
- 1.Potsic WP, Korman SB, Samadi DS, Wetmore RF. Congenital cholesteatoma: 20 years’ experience at The Children’s Hospital of Philadelphia. Otolaryngol Head Neck Surg. 2002;126(4):409-414. [DOI] [PubMed] [Google Scholar]
- 2.Bennett M, Warren F, Jackson GC, Kaylie D. Congenital cholesteatoma: theories, facts, and 53 patients. Otolaryngol Clin North Am. 2006;39(6):1081-1094. [DOI] [PubMed] [Google Scholar]
- 3.Cho HS, Kim HG, Jung DJ, Jang JH, Lee SH, Lee KY. Clinical aspects and surgical outcomes of congenital cholesteatoma in 93 children: increasing trends of congenital cholesteatoma from 1997 through 2012. J Audiol Otol. 2016;20(3):168-173. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Doyle KJ, Luxford WM. Congenital aural cholesteatoma: results of surgery in 60 cases. Laryngoscope. 1995;105(3 Pt 1):263-267. [DOI] [PubMed] [Google Scholar]
- 5.Marchioni D, Soloperto D, Rubini A, et al. Endoscopic exclusive transcanal approach to the tympanic cavity cholesteatoma in pediatric patients: our experience. Int J Pediatr Otorhinolaryngol. 2015;79(3):316-322. [DOI] [PubMed] [Google Scholar]
- 6.Kobayashi T, Gyo K, Komori M, Hyodo M. Efficacy and safety of transcanal endoscopic ear surgery for congenital cholesteatomas: a preliminary report. Otol Neurotol. 2015;36(10):1644-1650. [DOI] [PubMed] [Google Scholar]
- 7.Park JH, Ahn J, Moon IJ. Transcanal endoscopic ear surgery for congenital cholesteatoma. Clin Exp Otorhinolaryngol. 2018;11(4):233-241. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Jenks CM, Purcell PL, Federici G, et al. Transcanal endoscopic ear surgery for congenital cholesteatoma: a multi-institutional series. Otolaryngol Head Neck Surg. 2022;167(3):537-544. [DOI] [PubMed] [Google Scholar]
- 9.Basonbul RA, Ronner EA, Kozin ED, Lee DJ, Cohen MS. Systematic review of endoscopic ear surgery outcomes for pediatric cholesteatoma. Otol Neurotol. 2021;42(1):108-115. [DOI] [PubMed] [Google Scholar]
- 10.Levenson MJ, Parisier SC, Chute P, Wenig S, Juarbe C. A review of twenty congenital cholesteatomas of the middle ear in children. Otolaryngol Head Neck Surg. 1986;94(5):560-567. [DOI] [PubMed] [Google Scholar]
- 11.Kojima H, Tanaka Y, Shiwa M, Sakurai Y, Moriyama H. Congenital cholesteatoma clinical features and surgical results. Am J Otolaryngol. 2006;27(5):299-305. [DOI] [PubMed] [Google Scholar]
- 12.McGill TJ, Merchant S, Healy GB, Friedman EM. Congenital cholesteatoma of the middle ear in children: a clinical and histopathological report. Laryngoscope. 1991;101(6 Pt 1):606-613. [DOI] [PubMed] [Google Scholar]
- 13.Bacciu A, Di Lella F, Pasanisi E, et al. Open vs closed type congenital cholesteatoma of the middle ear: two distinct entities or two aspects of the same phenomenon? Int J Pediatr Otorhinolaryngol. 2014;78(12):2205-2209. [DOI] [PubMed] [Google Scholar]
- 14.Potsic WP, Samadi DS, Marsh RR, Wetmore RF. A staging system for congenital cholesteatoma. Arch Otolaryngol Head Neck Surg. 2002;128(9):1009-1012. [DOI] [PubMed] [Google Scholar]
- 15.World Health Organization . International Classification of Impairments, Disabilities, and Handicaps: A Manual of Classification Relating to the Consequences of Disease. World Health Organization; 1980. [Google Scholar]
- 16.D’haenens W, Dhooge I, Maes L, et al. The clinical value of the multiple-frequency 80-Hz auditory steady-state response in adults with normal hearing and hearing loss. Arch Otolaryngol Head Neck Surg. 2009;135(5):496-506. [DOI] [PubMed] [Google Scholar]
- 17.Rosa LA, Suzuki MR, Angrisani RG, Azevedo MF. Auditory brainstem response: reference-values for age. Codas. 2014;26(2):117-121. [DOI] [PubMed] [Google Scholar]
- 18.Lazard DS, Roger G, Denoyelle F, Chauvin P, Garabédian EN. Congenital cholesteatoma: risk factors for residual disease and retraction pockets–a report on 117 cases. Laryngoscope. 2007;117(4):634-637. [DOI] [PubMed] [Google Scholar]
- 19.Stapleton AL, Egloff AM, Yellon RF. Congenital cholesteatoma: predictors for residual disease and hearing outcomes. Arch Otolaryngol Head Neck Surg. 2012;138(3):280-285. [DOI] [PubMed] [Google Scholar]
- 20.Bennett ML, Zhang D, Labadie RF, Noble JH. Comparison of middle ear visualization with endoscopy and microscopy. Otol Neurotol. 2016;37(4):362-366. [DOI] [PubMed] [Google Scholar]
- 21.Badr-el-Dine M. Value of ear endoscopy in cholesteatoma surgery. Otol Neurotol. 2002;23(5):631-635. [DOI] [PubMed] [Google Scholar]
- 22.James AL, Cushing S, Papsin BC. Residual cholesteatoma after endoscope-guided surgery in children. Otol Neurotol. 2016;37(2):196-201. [DOI] [PubMed] [Google Scholar]
- 23.Dixon PR, James AL. Evaluation of residual disease following transcanal totally endoscopic vs postauricular surgery among children with middle ear and attic cholesteatoma. JAMA Otolaryngol Head Neck Surg. 2020;146(5):408-413. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Prasad SC, La Melia C, Medina M, et al. Long-term surgical and functional outcomes of the intact canal wall technique for middle ear cholesteatoma in the paediatric population. Acta Otorhinolaryngol Ital. 2014;34(5):354-361. [PMC free article] [PubMed] [Google Scholar]
- 25.Lailach S, Zahnert T, Lasurashvili N, Kemper M, Beleites T, Neudert M. Hearing outcome after sequential cholesteatoma surgery. Eur Arch Otorhinolaryngol. 2016;273(8):2035-2046. [DOI] [PubMed] [Google Scholar]
- 26.Kimura E, G Leonard C, L James A. The effect of the status of the ossicular chain and choice of graft material on hearing outcomes in pediatric cholesteatoma surgery. J Int Adv Otol. 2021;17(2):127-133. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Sharing Statement