Keywords
cochlear implant - cochlear round window - temporal bone - round window - ear
Introduction
There are three key surgical steps for the cochlear implant; they include 1) cortical
mastoidectomy; 2) posterior tympanotomy; and 3) round window identification and cochleostomy.
These surgical steps may be influenced by the anatomical variations that may result
in different levels of difficulty during surgery. Accordingly, surgical procedure
should be modifiable, if necessary, to obtain adequate surgical exposure.
High-resolution computed tomography (HRCT) has always been the gold standard for the
preoperative assessment of the cochlear implant (CI) patients and can predict the
difficulty of the surgical procedure.[1]
[2]
The aim of our study is to provide a basic structured format for the preoperative
reading of the CT scan in relation to each of the key surgical steps, which can be
tailored to meet the individual requirements. The proposed reporting format, in the
form of a simple checklist, follows the order in which the ear structures are approached
during surgery and allows for the proper preoperative assessment.
Patients and Methods
Ethical Considerations
Both the research project and patients' records retrieval were approved by the institutional
review board (IRB). The patients' consents for using their data in the research project
were also obtained.
Study Design
This was a prospective cohort study of consecutive cochlear implant surgeries performed
at our tertiary implant center in the period from January 2014 to December 2016. Patients
with inner ear anomalies, previous mastoid surgery, reimplantations, revision surgeries
and patients who were implanted via alternative techniques (for example, transcanal
approach) were excluded from the study.
Subjects
Forty-seven cases have fulfilled the inclusion criteria and were included in the study.
All patients were having bilateral severe-to-profound or profound sensorineural hearing
loss.
Methods
High Resolution Computed Tomography (HRCT) scans of the temporal bone were obtained
with slice thickness of 1.0 mm and were acquired at 120 kVp, 250 mA, imaging matrix
of 512 × 512, 160 mm field of view (FOV) and voxel size of ≈ 0.3½ x 0.3½ x 0.35 mm3. They were viewed in the standard bone window setting and were analyzed and reported
according to our proposed radiology checklist ([Fig. 1]).
Fig. 1 Radiological checklist used for reporting CI cases.
All images were evaluated by the primary author. To avoid interobserver variability,
the images were re-evaluated retrospectively by another independent investigator.
When the evaluation by the two investigators was different, the final decision was
made by the two investigators reviewing the scan together. Each item of the proposed
checklist was assessed objectively as follows:
1. Mastoid evaluation: feasibility of cortical mastoidectomy depends on the location
of the sigmoid sinus (SS) and the level of the temporal dura. The SS locations were
classified into four types according to their relation to three imaginary lines; line
1) the posteriorly-extended line that joins the common crus of the posterior semi-circular
canal (PSCC) to the PSCC; line 2) the posteriorly-extended line of the tympanic segment
of the facial nerve (FN); and line 3) the posteriorly-extended line of the malleal-incudal
axis ([Fig. 2]).[3] Types 3 and 4 were considered as predictors for contracted mastoid and difficult
mastoidectomy. The dural level was measured relative to the upper border of the petrous
bone ([Fig. 3]).[4]
Fig. 2 Two axial HRCT sections. Based on the shown reference lines, the sigmoid sinus was
categorized into four types; type [1] the most protruding portion of the sigmoid sinus
is medial or posterior to line 1, type [2] the most protruding portion of the sigmoid
sinus is located between line 1 and line 2, type [3] the most protruding portion of
the sigmoid sinus is located between line 2 and line 3, and type 4) the most protruding
portion of the sigmoid sinus is lateral or anterior to line 3.
Fig. 3 Coronal cut of HRCT illustrating the distance measured from a line drawn tangentially
from the upper edge of the petrous bone to the lowest point of temporal dura.
2. Posterior tympanotomy & cochleostomy evaluation: the presence of air cells around
the facial recess (sentinel air cell) was assessed using the axial cuts. The impact
of the FN course on the feasibility of posterior tympanotomy as well as the accessibility
of the round window (RW), with subsequent feasibility of cochlear implant (CI) electrode
array insertion, was assessed using the proposed prediction line by Mandour et al[5] ([Fig. 4]).
Fig. 4 Two axial HRCT sections. The left section shows round window area (triangle). The
EAC line (solid line) was drawn between the bony-cartilaginous junction of the posterior
EAC wall and the tympanic annulus, and the RW line (dashed line) was drawn from the
posterior edge of the RWN along the anterolateral part of the FN. Notice that both
lines are nearly parallel, indicating good RW accessibility. The right image shows
that the EAC line (dashed line) and the RW line (solid line) intersect with each other
laterally, indicating limited accessibility.
The difficulties encountered with each of the three key intraoperative steps (cortical
mastoidectomy, posterior tympanotomy, and round window access) were scored subjectively
according to the following scale: 1 (easy) - indicated a classic straightforward surgical
step; 2 (moderate) - indicated some difficulty that could be simply overcome by positional
manipulation of the patient or the microscope; and 3 (difficult) - indicated definite
difficulty interfering with optimum achieving of the surgical step in spite of positional
manipulation or that required special instrumentation to accomplish the step. Scoring
was performed by the primary surgeons who were blinded to the potential predictors
of the difficulties reported on the checklist.
Statistical Analysis
Using a two-sided Z test and α level of 0.05, a power analysis showed that recruiting
a sample size of 47 patients will yield an 80% power to detect the difference in the
predictability of the surgical difficulties with and without utilizing our checklist.
The chi-square test of independence was used to find the potential relationship between
the difficulty of the mastoidectomy and the SS location as well as the potential relationship
between the difficulty in accessing the facial recess and the presence/absence of
the facial recess air cell. It was also used to determine the potential relationship
between the difficulties in RW accessibility encountered during surgery, which were
reported radiologically.
One-way analysis of variance (ANOVA) and Student t-test were used to compare the level of the tegmen between the three groups representing
the degrees of difficulties of the mastoidectomy (that is, easy, moderate, and difficult).
Results
Forty-seven patients were included in the study (29 male and 18 female). They were
41 children with a mean age of 3.7 years (range: from 9 months to 8.5 years) and 6
adults with a mean age of 22.4 years (range: from 19 to 40 years). All the children
were prelingual while five adults were postlingual, and one was perilingual. A nucleus
freedom cochlear implant system with CI422 electrode (Cochlear, Lane Cover, NSW, Australia)
was used in 32 patients, and an Advanced Bionics HiRes 90K system with Hifocus 1J
electrode (Sylmar, California, USA) was used in the remaining 15 patients. Forty-five
cases had hearing loss of unknown etiology, while in one adult hearing loss was posttraumatic
and, in another case, it was due to autoimmune inner ear disease. Forty patients received
a CI on the right side, and seven patients on the left side. There was no statistically
significant correlation between neither age nor sex and radiological measurements
or surgical findings.
Regarding the dural level, the mean distance of tegmen from the tangential line through
the upper edge of the petrous bone was different between the three groups (difficult = 7.5 ± 0.22 mm,
moderate = 5.36 ± 0.46 mm, and easy = 4.48 ± 0.35 mm). The mean distance of the difficult
group was significantly longer than that of the easy group based on one-way ANOVA
([Table 1]).
Table 1
Correlation between radiologically predicted level of temporal dura and surgical difficulty
scales encountered during mastoidectomy
|
Dural level
|
Surgical difficulties
|
|
Easy
|
Moderate
|
Difficult
|
|
High
|
17
|
10
|
0
|
|
Low
|
0
|
0
|
20
|
A chi-square test of independence showed that there was a significant association
between the difficulty encountered during the cortical mastoidectomy and the SS type,
as reported in the radiology checklist (p-value < 0.001) ([Table 2]). The absence of an air cell around the FN was associated significantly with difficulty
in accessing the facial recess (p-value < 0.001). Also, it showed a significant association between the difficulties
in RW accessibility encountered during surgery, which were reported radiologically
using the proposed prediction line (p-value < 0.001) ([Fig. 5]).
Fig. 5 Correlation between predicted and intra-operative RWN visibility (group A represents
radiologically visible RWN, group B represents radiologically invisible RWN)
Table 2
Correlation between type of mastoid pneumatization, determined radiologically, and
surgical difficulty scales encountered during mastoidectomy
|
Mastoid pneumatization type
(n. of cases)
|
Surgical difficulties
|
|
Easy
|
Moderate
|
Difficult
|
|
Type 1 (11 cases)
|
10
|
1
|
0
|
|
Type 2 (8 cases)
|
6
|
1
|
1
|
|
Type 3 (10 cases)
|
1
|
3
|
6
|
|
Type 4 (18 cases)
|
0
|
5
|
13
|
Discussion
There is a high degree of variability between surgeons in analyzing the preoperative
CT scan. Our goal is to formalize and standardize the various radiological parameters
to provide a basic structured format for reporting the preoperative CT in a checklist
form. The reported checklist allows proper systematic surgical evaluation and follows
the order in which the ear structures are approached during surgery.
In our study, we reported that contracted mastoid cavity and low-lying tegmen were
associated with a greater degree of difficulty during mastoidectomy with difficult
identification of the important landmarks required for subsequent steps (such as the
lateral semicircular canal and incus), which may result in high risks of iatrogenic
injury to the FN and labyrinth.[6]
We have noticed that the anteriorly displaced SS may interfere with proper visualization
on turning the head of the patient away from the surgeon's side, which may be needed
to expand the surgical field view to identify the various landmarks, such as the short
process of incus and the tympanic segment of FN.
Our results seem to be consistent with previous studies in the literature. Pereira
et al[4] have reported that surgical difficulties during mastoidectomy were encountered when
the distance between the anterior border of the SS and the posterior wall of the EAC
was less than 9 mm, and the distance between the temporal meninges and the upper edge
of the petrous bone was equal to or above 7 mm. Park et al[6] reported that low-lying tegmen and poor mastoid pneumatization were associated with
a greater difficulty during mastoidectomy.
Drilling in the area of sentinel air cells (air cells around the facial recess) exposes
the top end of the facial recess.[7] In our study, the absence of sentinel air cells on the preoperative CT scan was
associated with a higher degree of difficulty during posterior tympanotomy. This finding
was similar to that observed by Park et al[6]
It has been widely reported in the literature that the angle between the cortex of
the EAC and the FN, and FN location are the most important variables with strongest
correlation to round window niche (RWN) accessibility with subsequent feasibility
of electrode array insertion.[8]
[9] Apart from complex radiologic measurements, we have utilized the prediction line
method proposed by Mandour et al[5] to assess RWN accessibility. We have found a significant correlation between predicted
radiologic difficulties and actual surgical difficulties encountered during operation.
Other factors that could compromise the feasibility of electrode array full insertion,
as anomalous cochlea and abnormal cochlear patency, have been considered but have
not been reported in our proposed checklist as those cases did not fulfill the inclusion
criteria of our study. These cases were managed individually and were excluded from
the study to avoid bias.
We recognize that there are some limitations in our study, including that the other
factors that may determine round window visibility have not been tested and assessment
of the difficulty of each surgical step is subjective. These issues shall be addressed
in future studies.
Conclusion
The proposed checklist provides a systematic structured approach for preoperative
radiological evaluation. Contracted mastoid and lower tegmen position were associated
with a greater difficulty of the cortical mastoidectomy. The presence of an air cell
around the facial nerve was predictive of easier facial recess access exposure. Facial
nerve location and EAC posterior wall inclination were predictive of difficult RW
accessibility.
