The Neuroradiology Journal 21: 393-400, 2008
www. centauro. it
High Resolution Computed Tomography Analysis of the Temporal Bone HSIUNG-KWANG CHUNG*, CHIN-YUAN WANG*, CHIA-DER LIN*, YU-CHIEN LO**, MING-HSUI TSAI* * Department of Otorhinolaryngology, Head & Neck Surgery, China Medical University Hospital; Taichung City, Taiwan (R.O.C.) ** Departments of Radiology, China Medical University Hospital; Taichung City, Taiwan (R.O.C.)
Key words: temporal bone, high resolution computed tomography, internal acoustic canal, vestibular schwannoma, superior semicircular canal dehiscence
SUMMARY – Detailed radiological assessment by high resolution computed tomography (HRCT) of temporal bone is demanded before any temporal bone or skull base surgery. The aim of this study was to measure the relations between the anatomical landmarks of the temporal bone and to assist the otolaryngologist in establishing accurate preoperative evaluation. We enrolled 43 patients who underwent temporal bone HRCT between February 2004 and May 2004. Contiguous axial and coronal images at 1.0 mm thickness were obtained. Some landmarks such as the superior and inferior lips of the internal acoustic canal (IAC), the malleoincus joint, and the posterior semicircular canal were labeled in the coronal and axial views. Then we measured the distance between them. Average IAC diameter in the coronal and axial views was 5.33 mm and 6.92 mm. Average IAC length in the coronal and axial views was 12.29 mm and 11.09 mm. The thickness of the retrolabyrinthine bone was 3.78 mm. The incidence of thinning bone overlying the superior semicircular canal was 2.3%. Our data could be applied to normal distribution because there were no statistical differences between the measurements of normal ears and diseased ears. Several specific measurements can be applied to the preoperative evaluation of vestibular schwannoma including the retrosigmoid approach, the translabyrinthine approach and the middle fossa approach.
Introduction The newer image-processing technologies investigating the temporal bone structures were recently presented. Bartling et Al studied the registration and fusion of computed tomography (CT) and magnetic resonance imaging (MRI) of temporal bone by an experimental software package and the overall mean value for CRE (consistency registration error) was 0.60 mm 1. Jun et Al investigated the usefulness of a three-dimensional reconstruction of the temporal bone 2. Both studies were based on temporal bone HRCT. Although MRI has assumed mainstream significance for early detection of vestibular schwannomas, the role of temporal bone high-resolution computed tomography (HRCT) analysis cannot be substituted. Advances in medical imaging technology enable sharper resolution of temporal osseous structures and allow preoperative evaluation
of the anatomical landmarks by otolaryngologists. The surgical approaches to the treatment of vestibular schwannomas are varied; they differ according to tumor size and location, disease progression, and individual anatomical characteristics. For example, a retrosigmoid approach, which is generally acknowledged as the treatment of choice for the preservation of hearing function, is associated with inadvertent fenestration of the labyrinth and potential deafness. The translabyrinthine approach provides the surgeon with direct exposure of the internal acoustic canal (IAC); thus, there is a possibility of preserving the facial nerve anatomy. However, in most cases, hearing is sacrificed due to intraoperative labyrinthectomy. This usually occurs because the lateral semicircular canal (LSC) was opened and entered in the first instance. The middle fossa approach has the potential of preserving hearing; this approach is provided when that the tumor size 393
High Resolution Computed Tomography Analysis of the Temporal Bone
does not exceed 2 cm. The lower half of the IAC fundus cannot be visualized entirely; therefore, there is a possibility of recurrence. Moreover, the lack of anatomical landmarks on the floor of the middle cranial fossa, the predisposition to facial nerve injury, and the sequelae of temporal lobe retraction make the middle fossa approach considerably challenging. Further, individual variability in the temporal bone morphology may affect the selection of the microsurgical approach. More detailed radiological assessment is demanded before any temporal bone or skull base surgery using the modern HRCT technique. In this study, the radiologic data based on the measurements of the relations between the anatomical landmarks of the temporal bone were collected. We hope that the following data subsequent to statistical analysis will be considered the anatomical reference data and eventually assist otolaryngologists in establishing accurate preoperative evaluation. Additionally, several landmarks could serve as a guide to surgeons in identifying the IAC and the embedded tumor. Prior knowledge of the distances between the life-threatening structures and the anatomical landmarks can prevent harmful drilling during surgery. Several investigators have studied the IAC during temporal bone dissection in cadavers using medical imaging and through the study of histopathological sections. Our data are applicable to the length and physical dimensions, and the shape of the IAC. We compared our data to those of other authors in order to assess whether regional or racial differences contributed to data variation. Missing or thinning of bone overlying the superior semicircular canal (SSC) may lead to superior semicircular canal dehiscence (SCD) syndrome. HRCT can be used to detect abnormal thickness of the temporal bone. Patients with thin bone overlying the SSC remained asymptomatic in their daily life; however, in the event of head injury or barotrauma to these patients, there is a possibility of occurrence of SCD syndrome. Materials and Methods This was a retrospective study. Between February 2004 and May 2004, we enrolled patients who underwent temporal bone HRCT at our institution. We reviewed their basic data, clinical manifestations and the diagnoses of all the patients. Most of their complaints were otological, such as hearing impairment, persistent 394
Hsiung-Kwang Chung
otorrhea, and intractable vertigo. Our exclusion criteria were critical to ensure minimal influence of other diseases. First, patients with documented congenital anomaly were excluded from the analysis. Previous aural procedures might have resulted in the destruction of normal anatomical structures, so patients with history of ear surgery were excluded. There is a possibility that the ears are still in a state of postnatal development in teenagers; hence, we excluded patients who were younger than 18 years of age. Temporal bone fracture would alter the relative positions of our landmarks, so patients with a similar history were omitted. If cholesteatomas had resulted in decay of the ossicles or the landmarks, we did not include such cases. Every effort was made to keep the patients in the central head position inside the scanner so that the resulting asymmetry appearance between right and left temporal bones is minimally reduced. In order to ensure the quality of this study, we omitted the images from improper head positions. Contiguous axial and coronal images at 1.0mm thickness were obtained using multi-slice axial CT system (GE, Milwaukee, WI, U.S.A.). We set the window level of 350 HU and a window width of 3,000 HU, which provided the sharpest images. All the measurements were performed by a single otolaryngologist via the PACS system (Petavision; Asan Medical Center, Seoul, South Korea). In the coronal views, we labeled the superior and inferior lips of the IAC, the fundus of the IAC, the malleoincus joint, the superior lip of the external acoustic canal (EAC), and the roof of the SSC. Additionally, other specific landmarks were identified (figure 1). In the axial sections, the anterior and posterior lips of the IAC, the posterior semicircular canal (PSC), the LSC, the deepest part of sigmoid sinus, and the malleoincus joint were identified. Figure 2 indicates the locations of the other landmarks 3. If the landmarks we selected were on the same slice, then we measured the distance directly. If they were not on the same plane, a radiologist would reformat or rotate the images on an advanced workstation (AW 4.3, GE, Milwaukee, WI, USA), and hence we still can calculate the distances through different images. The differences between individuals or diseases were estimated using the unpaired and paired t tests. The data were analyzed using the SAS statistical software (version 6.12; SAS Institute Inc, Cary, NC). A p value of less than 0.05 was considered statistically significant.
www. centauro. it
The Neuroradiology Journal 21: 393-400, 2008
Figure 1 A) Computed tomography image of the temporal bone in the coronal views. B) Diagram of the landmarks: (1) superior lip of the IAC; (2) midpoint of the porus acusticus; (3) inferior lip of the IAC; (4) fundus of the IAC; (5) malleoincus joint; (6) margin of the petrous bone; (7) shortcut between petrous bone and (8); (8) roof of the SSC; (9) superior lip of EAC; (10) point where the line extending from (6) and (8) and the lateral skull bone met.
Figure 2 A) Computed tomography image of the temporal bone in the axial views. B) Diagram of the landmarks: (A) anterior lip of the IAC (B) midpoint of the porus acusticus; (C) inferior lip of the IAC; (D) fundus of the IAC; (E) malleoincus joint; (F) common crus; (G) shortcut of (F) and the petrous bone; (H) posterior semicircular canal; (I) shortcut of (H) and the petrous bone; (J) deepest part of sigmoid sinus.
395
High Resolution Computed Tomography Analysis of the Temporal Bone
Results A total of 43 temporal bone HRCT scans (86 ears) were reviewed. Tables 1 and 2 show the demographics and the diagnosis of all patients. There were 30 normal ears and 56 diseased ears. In the coronal views, the average vertical diameter of the IAC was 5.33 mm (standard deviation (SD), 1.1 mm; range, 3.34-8.13 mm), and the average length of the IAC was 12.29 mm (SD, 2.0 mm; range, 6.39-17.62 mm). The average distance between the roof of the SSC and the petrous bone (the thickness of the bone overlying the SSC) was 1.7 mm. The average distance from the SSC to the lateral petrous bone was 25.41 mm (SD, 2.8 mm; range, 18.9133.90 mm). Other morphometric data and the results of the statistical analyses obtained from the coronal views are presented in table 3.
Hsiung-Kwang Chung
of the bone immediately posterior to the PSC, also known as the retrolabyrinthine bone, was measured. The thickness of the retrolabyrinthine bone was 3.78 mm (SD, 2.0 mm; range 1.01-9.97 mm). Table 4 shows the other measurements from the axial views. Statistical analysis using the unpaired t test revealed that the differences between the measurements of the normal ears and those of the diseased ears were not statistically significant. The five selected measurements between the normal ears and the diseased ears are listed in table 5. All the p values were greater than 0.05. Subsequently, groupwise comparisons were performed between the other characteristics. However, there were still no differences between the diseased and normal ears. Therefore, our data could be applied to the normal distribution regardless of the aural condition, genders, ages and diagnoses of the patients.
Table 1 Demographics of all patients
Characteristics
Data
No. of ears
86 (43 patients)
Discussion
Age (years) Mean
41
Range
18~82
Men/Women
12/31
Table 2 Diagnoses of all patients
Diagnosis
No. of ears
Normal ears
30
(35%)
Chronic otitis media
20
(23%)
Cholesteatoma
16
(19%)
Idiopathic hearing loss
09
(10%)
Vertigo
04
0(5%)
Acute mastoiditis
04
0(5%)
EAC tumor
03
0(3%)
In the axial views, the average horizontal diameter was 6.92 mm (SD, 1.4 mm; range, 4.12-10.66 mm), and the average length of the IAC was 11.09 mm (SD, 2.0 mm; range, 6.9515.29 mm). The average distance between the posterior lip of the IAC and the deepest part of the sigmoid sinus was 35.55 mm (SD, 3.8 mm; range 27.76-44.83 mm). The thickness 396
The IAC length has been investigated by several authors and their results are varied. Day et Al studied 32 fixed cadaveric temporal bones using fine-cut bone window CT scans and reported that the average canal length was 10 mm (SD, 1.8 mm; range, 6.6- 14.0 mm) 4. Driscoll et Al studied 40 coronal temporal bone CT scans and observed that the average IAC length was 14.9 mm (SD, 2.9 mm; range, 9.522 mm) 5. However, there were differences in the measurement techniques. Certain authors assumed the length of the IAC floor as an indicator of the IAC length. We labeled the midpoint between the superior and inferior lips of the IAC on the coronal views and measured the distance between the midpoint and the fundus. Similarly, we obtained the IAC length on the axial views. According to our data, the IAC lengths were 12.29 mm and 11.09 mm on coronal and axial sections, respectively. We observed that the IAC lengths on the coronal views were significantly greater than those on the axial views, according to the paired t test (p < 0.05). Selesnick et Al compared the MRI and the intraoperative findings of the lateral IAC, their IAC lengths were 12.7 mm and 13.7 mm on the axial and coronal views, respectively 6. This may be attributed to the shorter distance between the posterior lip of IAC and the fundus. Sakashita et Al reconstructed the IACs of 20 patients by three-dimensional CT technology and concluded that regardless of age, the
www. centauro. it
The Neuroradiology Journal 21: 393-400, 2008
Table 3 Results of the statistical analyses obtained from the computed tomography image of the temporal bone in coronal views
Landmarks
Descriptions
Mean±SD
Range
(1)-(3)
IAC diameter
5.33±1.1 mm
3.34~8.13 mm
(2)-(4)
IAC length
12.29±2.0 mm
6.39~17.62 mm
(8)-(10)
SSC to lateral petrous bone
25.41±2.8 mm
18.91~33.9 mm
(4)-(10)
IAC fundus to lateral petrous bone
29.86±3.2 mm
42.21~19.82 mm
(4)-(5)
IAC fundus to malleoincus joint
10.93±1.7 mm
8.28~17.56 mm
(4)-(8)
IAC fundus to SSC
8.57±2.0 mm
6.22~10.93 mm
(1)-(4)
Superior IAC lip to IAC fundus
12.01±2.2 mm
6.22~17.94 mm
(3)-(4)
Inferior IAC lip to IAC fundus
13.12±2.3 mm
7.51~18.48 mm
(1)-(5)
Superior IAC lip to malleoincus joint
22.59±2.8 mm
16.04~29.89 mm
(3)-(5)
Inferior IAC lip to malleoincus joint
24.06±2.7 mm
17.89~31.8 mm
(6)-(8)
Petrous bone margin to SSC
15.25±2.9 mm
6.2~21.83 mm
(7)-(8)
Roof of SSC to petrous bone
1.70±0.9 mm
0~5 mm
(5)-(9)
Bony EAC to malleoincus joint
14.91±3.0 mm
8.31~21.95 mm
(4)-(9)
Fundus of IAC to bony EAC
25.47±3.1 mm
19.21~33.22 mm
∠(1)(4)(3)
Angle of IAC between fundus
21.77±5.3°
12~36°
∠(1)(5)(3)
Angle of IAC between malleoincus joint
11.29±2.7°
5~17°
anterior wall was the longest whereas the posterior wall was the shortest 7. The greater the distance between the posterior lip of IAC and the fundus was, the more difficult it became to perform internal labyrinthectomy via the retrosigmoid approach 8. The rates of development of the lengths and diameters of the IACs are different. The IAC diameter increased until approximately one year of age while the IAC length increased until late adolescence, i.e., approximately ten years of age 7. Hence, we excluded patients younger than 18 years of age. The average vertical diameter of the IAC (on coronal views) was 5.33 mm, and the average horizontal diameter (on axial views) was 6.92 mm. Thus, the latter was greater than the former, and the difference between them was statistically significant according to the paired t test. The shape of the IAC was flat and narrowing was observed posterolaterally. The retrosigmoid approach for the surgical treatment of vestibular schwannomas has a hearing preservation rate of 18% to 75%, the key to ensure prevention of permanent sensorineural hearing loss is to avoid labyrinthine
injury, especially during lateral IAC dissection . Miller et Al defined the center of the craniotomy during the retrosigmoid approach as 15 mm posterior to the sigmoid sinus 8,11. On an average, the PSC is situated 10.51 mm posterior to the sigmoid sinus. The average distance between the posterior lip of the IAC and the sigmoid sinus was 35.55 mm. The distance from the posterior IAC lip to the vestibulae was the maximum permissible length of the posterior IAC lip that could be removed 4. According to our reports, the distances were 10 mm (from the posterior IAC lip to the common crus) and 14.15 mm (from the posterior IAC lip to the PSC). Ekinci G et Al examined 100 normal ears in 50 subjects and measured the length from the medial lip of the posterior wall of the IAC to the medial wall of the vestibule (close to the common crus in our study) 12. The average distance was 9.7 mm, which was similar to our data. The foremost structures to be visualized after retrolabyrinthectomy were the endolymphatic system and the common crus. However, we observed that there were two definitions of the retrolabyrinthine bones. One is defined as the thickness of the petrous bone overlying 9,10
397
High Resolution Computed Tomography Analysis of the Temporal Bone
Hsiung-Kwang Chung
Table 4 Results of the statistical analyses obtained from the computed tomography image of the temporal bone in the axial views
Landmarks
Descriptions
Mean±SD
Range
(A)-(C)
IAC diameter
6.92±1.4 mm
4.12~10.66 mm
(B)-(D)
IAC length
11.09±2.0 mm
6.93~15.29 mm
(C)-(J)
Posterior IAC lip to sigmoid sinus
35.55±3.8 mm
27.76~44.83 mm
(C)-(F)
Posterior IAC lip to common crus
10.0±1.8 mm
7.26~16.52 mm
(C)-(H)
Posterior IAC lip to PSC
14.15±2.2 mm
8.24~18.71 mm
(H)-(I)
Shortcut of PSC and petrous bone
3.78±2.0 mm
1.01~9.97 mm
(A)-(D)
Anterior IAC lip to IAC fundus
13.87±1.9 mm
9.37~18.25 mm
(C)-(D)
Posterior IAC lip to IAC fundus
9.8±2.1 mm
5.67~15.06 mm
(A)-(E)
Anterior IAC lip to malleoincus joint
22.62±1.9 mm
18.41~27.66 mm
(C)-(E)
Posterior IAC lip to malleoincus joint
18.72±2.1 mm
12.91~22.95 mm
(D)-(H)
PSC to IAC fundus
9.90±1.6 mm
5.4~14.12 mm
(D)-(J)
Sigmoid sinus to IAC fundus
29.86±2.8 mm
24.2~35.45 mm
(F)-(G)
Shortcut of common crus and petrous bone
5.26±1.4 mm
3.86~8.66 mm
(E)-(J)
Malleoincus joint to sigmoid sinus
25.39±3.1 mm
20.56~37.24 mm
(H)-(J)
PSC to sigmoid sinus
10.51±1.5 mm
8.64~13.55 mm
(F)-(J)
Common crus to sigmoid sinus
11.27±1.8 mm
8.82~15.06 mm
∠(A)(D)(C)
Angle of IAC between fundus
28.95±7.7°
10~48°
∠(A)(E)(C)
Angle of IAC between malleoincus joint
15.97±3.9°
9~26°
Table 5 Differences between the measurements of normal ears and diseased ears
Landmarks and descriptions
Normal Ears
Diseased Ears
p value
Coronal views
5.38±1.0 mm
5.31±1.2 mm
0.7660
Axial views
6.83±1.4 mm
6.97±1.4 mm
0.6497
Coronal views
12.25±2.0 mm
12.31±2.0 mm
0.8795
Axial views
11.36±1.9 mm
10.95±2.0 mm
0.3645
SSC to lateral petrous bone
25.33±3.3 mm
25.45±2.6 mm
0.8636
IAC fundus to malleoincus joint
11.02±1.5 mm
10.88±1.8 mm
0.7311
Angle of IAC between malleoincus joint
11.17±2.6°
11.36±2.8°
0.7557
Posterior IAC lip to sigmoid sinus
35.67±3.7 mm
35.48±3.9 mm
0.8203
Posterior IAC lip to PSC
14.08±2.1 mm
14.19±2.2 mm
0.8251
Shortcut of PSC and petrous bone
3.99±2.4 mm
3.66±1.8 mm
0.4672
IAC diameter
IAC length
398
www. centauro. it
the PSC; the other is the thickness of the common crus 4,8. We measured both the thickness values on the axial views, and the results were 3.78 mm (thickness of the petrous bone overlying the PSC) and 5.26 mm (thickness of the common crus). The aforementioned data would provide preoperative guidelines for the surgeon to avoid any labyrinthine injury and to obtain optimal potential for hearing preservation. The fundus of the IAC can be widely exposed during translabyrinthine approach, therefore, complete tumor removal is possible regardless of the tumor size. However. the translabyrinthine approach may destroy the residual hearing in the affected ear because of labyrinthectomy. The sigmoid sinus will be decompressed prior to labyrinthectomy, so we measured the distances from the sigmoid sinus and the common crus and that to the PSC; they were 11.27 mm and 10.51 mm, respectively. The average distance between the fundus of the IAC and the sigmoid sinus was 29.86 mm. The short process of the incus can be identified as an oblique projection during translabyrinthine approach. According to our measurements, the average distance between the sigmoid sinus and the malleoincus joint was 25.39 mm and that between the PSC and the fundus was 9.9 mm. These landmarks and measurements can provide guidelines to the surgeon in outlining the fundus of the IAC. The middle fossa approach had the best potential for hearing preservation provided that the maximum dimension of the tumor did not exceed 2 cm. Factors including poor angle of view to the IAC, blind dissection, and intraoperative temporal lobe retraction made this approach technically more difficult 5. Further, the patient ran the risk of neurological sequelae such as facial nerve paralysis, seizure, and hemiparesis. Further, epidural hematoma, meningitis, and cerebrospinal fluid leakage have been reported as life-threatening complications associated with this approach 13. We drew an imaginary line from the margin of the petrous bone to the upper limit of the SSC and extended it to the lateral skull bone. The point where the line and the lateral skull bone met was defined as the lower limit of the bone flap for temporal craniotomy. On the coronal views, the average distances from this point to the upper limit of the SSC and to the IAC fundus
The Neuroradiology Journal 21: 393-400, 2008
were 25.41 mm and 29.86 mm, respectively. Fisch and Tsunoda et Al proposed that the anatomical position between the IAC and the SSC appeared to be relatively constant. Therefore, the labyrinthine bone could be drilled until the contour of the SSC was exposed 14,15. According to the results, the average distance between the upper limit of the SSC and the IAC fundus was 8.57 mm on average. According to our report, the average thickness of the bone overlying the SSC on the coronal views was 1.7 mm. Kanzaki proposed that the thickness was 2.4 mm (SD, 1.3 mm; range 0.8-5.8 mm) 3. Carey et Al investigated 100 control specimens screened from 1000 temporal bones and reported that the average thickness was 1.79 mm (SD, 1.20 mm; range, 0.13-6.24 mm) 16. The diagnosis of SCD was established on the basis of history, physical examination and confirmation of bony defect overlying the SCC on the temporal bone HRCT 17. There were two patients with zero bone thickness. Due to the limited resolution, we preferred to define them as “thinning” rather than “dehiscence.” The incidence rate of thinning was 2.3%. However, only one patient had clinical vestibulopathy, the other ear was within the normal groups. Carey et Al studied 1000 temporal bones of 596 adults. In all, there were 19 specimens with evidence of thinning or dehiscence, giving an incidence rate of 1.9% 16. Conclusions There were no major differences in the distances between the selected landmarks of the temporal bone on HRCT analysis with regard to the diagnoses, ages and genders of the patients. Our estimated data were similar to the publications by others. Several specific measurements such as the thickness of the retrolabyrinthine bone, the IAC lengths, and the distance between the sigmoid sinus and the IAC can be applied to the preoperative evaluation of vestibular schwannoma. The surgeon should choose the most appropriate surgical approach on the basis of the anatomical characteristics, their preferences, and their technical expertise. The incidence of thinning bone overlying the SSC was 2.3%; patients with such findings should be careful to avoid barotrauma and accidental head injury.
399
High Resolution Computed Tomography Analysis of the Temporal Bone
Hsiung-Kwang Chung
References 1 Bartling SH, Peldschus K, Rodt T et Al: Registration and fusion of CT and MRI of the temporal bone. J Comput Assist Tomogr 29: 305-10, 2005. 2 Jun BC, Song SW, Cho JE et Al: Three-dimensional reconstruction based on images from spiral high-resolution computed tomography of the temporal bone: anatomy and clinical application. J Laryngol Otol 119: 693-8, 2005. 3 Kanzaki J: The surgical treatments of acoustic neuromas. Oto-Rhino-Laryngology Tokyo 33: 201-528, 1990. 4 Day JD, Kellogg JX, Fukushima T et Al: Microsurgical anatomy of the inner surface of the petrous bone: neuroradiological and morphometric analysis as an adjunct to the retrosigmoid transmeatal approach. Neurosurgery 34: 1003-1008, 1994. 5 Driscoll CLW, Jackler RK, Pitts LH et Al: Is the entire fundus of the internal auditory canal visible during the middle fossa approach for acoustic neuroma? Am J Otol 21: 382-388, 2000. 6 Selesnick SH, Rebol J, Heier LA et Al: Internal auditory canal involvement of acoustic neuromas: Surgical correlated to magnetic resonance imaging findings. Otol Neurotol 22: 912-916, 2001. 7 Sakashita T, Sando I: Postnatal development of the internal auditory canal studied by computer-aided threedimensional reconstruction and measurement. Ann Otol Rhinol Laryngol 104: 469-475, 1995. 8 Miller RS, Pensak ML: An anatomic and radiologic evaluation of access to the lateral internal auditory canal via the retrosigmoid approach and description of an internal labyrinthectomy. Otol Neurotol 27: 697-704, 2006. 9 Rhoton AL Jr.: Microsurgical removal of acoustic neuromas. Surg Neurol 6: 211-219, 1976. 10 Magnan J, Barbieri M, Mora R et Al: Retrosigmoid approach for small and medium-sized acoustic neuromas. Otol Neurotol 23: 141-145, 2002. 11 Blevins NH, Jackler RK: Exposure of the lateral extremity of the internal auditory canal through the retrosigmoid approach: a radioanatomic study. Otolaryngol Head Neck Surg 111: 81-90, 1994. 12 Ekinci G, Koc A, Baltacioglu F et Al: Temporal bone measurements on high-resolution computed tomography. J Otolaryngol 33: 387-9, 2004. 13 Bryce GE, Nedzelski JM, Rowed DW et Al: Cerebrospinal fluid leaks and meningitis in acoustic neuroma surgery. Otolaryngol Head Neck Surg 104: 81-87, 1991.
400
14 Tsunoda A, Komatsuzaki A, Kobayashi M et Al: Threedimensional image for the middle fossa approach and its anatomical considerations. Laryngoscope 111: 10481052, 2001. 15 Fisch U: Transtemporal surgery of the internal auditory canal: report of 92 cases, technique, indications and results. Adv Otorhinolaryngol 17: 203-240, 1970. 16 Carey JP, Minor LB, Nager GT: Dehiscence or thinning of bone overlying the superior semicircular canal in a temporal bone survey. Arch Otolaryngol Head Neck Surg 126: 137-147, 2000. 17 Hillman TA, Kertesz TR, Hadley K et Al: Reversible peripheral vestibulopathy: the treatment of superior canal dehiscence. Otolaryngol Head Neck Surg 134: 431-436, 2006.
Chia-Der Lin, MD Department of Otorhinolaryngology, Head & Neck Surgery China Medical University Hospital No.2 Yu-Te Road Taichung city (403) Taiwan (R.O.C.) Tel.: +886-4-2205-2121 ext 4436 Fax: +886-4-2205-2121 ext 4438 E-mail: [email protected]
www. centauro. it
High Resolution Computed Tomography Analysis of the Temporal Bone HSIUNG-KWANG CHUNG*, CHIN-YUAN WANG*, CHIA-DER LIN*, YU-CHIEN LO**, MING-HSUI TSAI* * Department of Otorhinolaryngology, Head & Neck Surgery, China Medical University Hospital; Taichung City, Taiwan (R.O.C.) ** Departments of Radiology, China Medical University Hospital; Taichung City, Taiwan (R.O.C.)
Key words: temporal bone, high resolution computed tomography, internal acoustic canal, vestibular schwannoma, superior semicircular canal dehiscence
SUMMARY – Detailed radiological assessment by high resolution computed tomography (HRCT) of temporal bone is demanded before any temporal bone or skull base surgery. The aim of this study was to measure the relations between the anatomical landmarks of the temporal bone and to assist the otolaryngologist in establishing accurate preoperative evaluation. We enrolled 43 patients who underwent temporal bone HRCT between February 2004 and May 2004. Contiguous axial and coronal images at 1.0 mm thickness were obtained. Some landmarks such as the superior and inferior lips of the internal acoustic canal (IAC), the malleoincus joint, and the posterior semicircular canal were labeled in the coronal and axial views. Then we measured the distance between them. Average IAC diameter in the coronal and axial views was 5.33 mm and 6.92 mm. Average IAC length in the coronal and axial views was 12.29 mm and 11.09 mm. The thickness of the retrolabyrinthine bone was 3.78 mm. The incidence of thinning bone overlying the superior semicircular canal was 2.3%. Our data could be applied to normal distribution because there were no statistical differences between the measurements of normal ears and diseased ears. Several specific measurements can be applied to the preoperative evaluation of vestibular schwannoma including the retrosigmoid approach, the translabyrinthine approach and the middle fossa approach.
Introduction The newer image-processing technologies investigating the temporal bone structures were recently presented. Bartling et Al studied the registration and fusion of computed tomography (CT) and magnetic resonance imaging (MRI) of temporal bone by an experimental software package and the overall mean value for CRE (consistency registration error) was 0.60 mm 1. Jun et Al investigated the usefulness of a three-dimensional reconstruction of the temporal bone 2. Both studies were based on temporal bone HRCT. Although MRI has assumed mainstream significance for early detection of vestibular schwannomas, the role of temporal bone high-resolution computed tomography (HRCT) analysis cannot be substituted. Advances in medical imaging technology enable sharper resolution of temporal osseous structures and allow preoperative evaluation
of the anatomical landmarks by otolaryngologists. The surgical approaches to the treatment of vestibular schwannomas are varied; they differ according to tumor size and location, disease progression, and individual anatomical characteristics. For example, a retrosigmoid approach, which is generally acknowledged as the treatment of choice for the preservation of hearing function, is associated with inadvertent fenestration of the labyrinth and potential deafness. The translabyrinthine approach provides the surgeon with direct exposure of the internal acoustic canal (IAC); thus, there is a possibility of preserving the facial nerve anatomy. However, in most cases, hearing is sacrificed due to intraoperative labyrinthectomy. This usually occurs because the lateral semicircular canal (LSC) was opened and entered in the first instance. The middle fossa approach has the potential of preserving hearing; this approach is provided when that the tumor size 393
High Resolution Computed Tomography Analysis of the Temporal Bone
does not exceed 2 cm. The lower half of the IAC fundus cannot be visualized entirely; therefore, there is a possibility of recurrence. Moreover, the lack of anatomical landmarks on the floor of the middle cranial fossa, the predisposition to facial nerve injury, and the sequelae of temporal lobe retraction make the middle fossa approach considerably challenging. Further, individual variability in the temporal bone morphology may affect the selection of the microsurgical approach. More detailed radiological assessment is demanded before any temporal bone or skull base surgery using the modern HRCT technique. In this study, the radiologic data based on the measurements of the relations between the anatomical landmarks of the temporal bone were collected. We hope that the following data subsequent to statistical analysis will be considered the anatomical reference data and eventually assist otolaryngologists in establishing accurate preoperative evaluation. Additionally, several landmarks could serve as a guide to surgeons in identifying the IAC and the embedded tumor. Prior knowledge of the distances between the life-threatening structures and the anatomical landmarks can prevent harmful drilling during surgery. Several investigators have studied the IAC during temporal bone dissection in cadavers using medical imaging and through the study of histopathological sections. Our data are applicable to the length and physical dimensions, and the shape of the IAC. We compared our data to those of other authors in order to assess whether regional or racial differences contributed to data variation. Missing or thinning of bone overlying the superior semicircular canal (SSC) may lead to superior semicircular canal dehiscence (SCD) syndrome. HRCT can be used to detect abnormal thickness of the temporal bone. Patients with thin bone overlying the SSC remained asymptomatic in their daily life; however, in the event of head injury or barotrauma to these patients, there is a possibility of occurrence of SCD syndrome. Materials and Methods This was a retrospective study. Between February 2004 and May 2004, we enrolled patients who underwent temporal bone HRCT at our institution. We reviewed their basic data, clinical manifestations and the diagnoses of all the patients. Most of their complaints were otological, such as hearing impairment, persistent 394
Hsiung-Kwang Chung
otorrhea, and intractable vertigo. Our exclusion criteria were critical to ensure minimal influence of other diseases. First, patients with documented congenital anomaly were excluded from the analysis. Previous aural procedures might have resulted in the destruction of normal anatomical structures, so patients with history of ear surgery were excluded. There is a possibility that the ears are still in a state of postnatal development in teenagers; hence, we excluded patients who were younger than 18 years of age. Temporal bone fracture would alter the relative positions of our landmarks, so patients with a similar history were omitted. If cholesteatomas had resulted in decay of the ossicles or the landmarks, we did not include such cases. Every effort was made to keep the patients in the central head position inside the scanner so that the resulting asymmetry appearance between right and left temporal bones is minimally reduced. In order to ensure the quality of this study, we omitted the images from improper head positions. Contiguous axial and coronal images at 1.0mm thickness were obtained using multi-slice axial CT system (GE, Milwaukee, WI, U.S.A.). We set the window level of 350 HU and a window width of 3,000 HU, which provided the sharpest images. All the measurements were performed by a single otolaryngologist via the PACS system (Petavision; Asan Medical Center, Seoul, South Korea). In the coronal views, we labeled the superior and inferior lips of the IAC, the fundus of the IAC, the malleoincus joint, the superior lip of the external acoustic canal (EAC), and the roof of the SSC. Additionally, other specific landmarks were identified (figure 1). In the axial sections, the anterior and posterior lips of the IAC, the posterior semicircular canal (PSC), the LSC, the deepest part of sigmoid sinus, and the malleoincus joint were identified. Figure 2 indicates the locations of the other landmarks 3. If the landmarks we selected were on the same slice, then we measured the distance directly. If they were not on the same plane, a radiologist would reformat or rotate the images on an advanced workstation (AW 4.3, GE, Milwaukee, WI, USA), and hence we still can calculate the distances through different images. The differences between individuals or diseases were estimated using the unpaired and paired t tests. The data were analyzed using the SAS statistical software (version 6.12; SAS Institute Inc, Cary, NC). A p value of less than 0.05 was considered statistically significant.
www. centauro. it
The Neuroradiology Journal 21: 393-400, 2008
Figure 1 A) Computed tomography image of the temporal bone in the coronal views. B) Diagram of the landmarks: (1) superior lip of the IAC; (2) midpoint of the porus acusticus; (3) inferior lip of the IAC; (4) fundus of the IAC; (5) malleoincus joint; (6) margin of the petrous bone; (7) shortcut between petrous bone and (8); (8) roof of the SSC; (9) superior lip of EAC; (10) point where the line extending from (6) and (8) and the lateral skull bone met.
Figure 2 A) Computed tomography image of the temporal bone in the axial views. B) Diagram of the landmarks: (A) anterior lip of the IAC (B) midpoint of the porus acusticus; (C) inferior lip of the IAC; (D) fundus of the IAC; (E) malleoincus joint; (F) common crus; (G) shortcut of (F) and the petrous bone; (H) posterior semicircular canal; (I) shortcut of (H) and the petrous bone; (J) deepest part of sigmoid sinus.
395
High Resolution Computed Tomography Analysis of the Temporal Bone
Results A total of 43 temporal bone HRCT scans (86 ears) were reviewed. Tables 1 and 2 show the demographics and the diagnosis of all patients. There were 30 normal ears and 56 diseased ears. In the coronal views, the average vertical diameter of the IAC was 5.33 mm (standard deviation (SD), 1.1 mm; range, 3.34-8.13 mm), and the average length of the IAC was 12.29 mm (SD, 2.0 mm; range, 6.39-17.62 mm). The average distance between the roof of the SSC and the petrous bone (the thickness of the bone overlying the SSC) was 1.7 mm. The average distance from the SSC to the lateral petrous bone was 25.41 mm (SD, 2.8 mm; range, 18.9133.90 mm). Other morphometric data and the results of the statistical analyses obtained from the coronal views are presented in table 3.
Hsiung-Kwang Chung
of the bone immediately posterior to the PSC, also known as the retrolabyrinthine bone, was measured. The thickness of the retrolabyrinthine bone was 3.78 mm (SD, 2.0 mm; range 1.01-9.97 mm). Table 4 shows the other measurements from the axial views. Statistical analysis using the unpaired t test revealed that the differences between the measurements of the normal ears and those of the diseased ears were not statistically significant. The five selected measurements between the normal ears and the diseased ears are listed in table 5. All the p values were greater than 0.05. Subsequently, groupwise comparisons were performed between the other characteristics. However, there were still no differences between the diseased and normal ears. Therefore, our data could be applied to the normal distribution regardless of the aural condition, genders, ages and diagnoses of the patients.
Table 1 Demographics of all patients
Characteristics
Data
No. of ears
86 (43 patients)
Discussion
Age (years) Mean
41
Range
18~82
Men/Women
12/31
Table 2 Diagnoses of all patients
Diagnosis
No. of ears
Normal ears
30
(35%)
Chronic otitis media
20
(23%)
Cholesteatoma
16
(19%)
Idiopathic hearing loss
09
(10%)
Vertigo
04
0(5%)
Acute mastoiditis
04
0(5%)
EAC tumor
03
0(3%)
In the axial views, the average horizontal diameter was 6.92 mm (SD, 1.4 mm; range, 4.12-10.66 mm), and the average length of the IAC was 11.09 mm (SD, 2.0 mm; range, 6.9515.29 mm). The average distance between the posterior lip of the IAC and the deepest part of the sigmoid sinus was 35.55 mm (SD, 3.8 mm; range 27.76-44.83 mm). The thickness 396
The IAC length has been investigated by several authors and their results are varied. Day et Al studied 32 fixed cadaveric temporal bones using fine-cut bone window CT scans and reported that the average canal length was 10 mm (SD, 1.8 mm; range, 6.6- 14.0 mm) 4. Driscoll et Al studied 40 coronal temporal bone CT scans and observed that the average IAC length was 14.9 mm (SD, 2.9 mm; range, 9.522 mm) 5. However, there were differences in the measurement techniques. Certain authors assumed the length of the IAC floor as an indicator of the IAC length. We labeled the midpoint between the superior and inferior lips of the IAC on the coronal views and measured the distance between the midpoint and the fundus. Similarly, we obtained the IAC length on the axial views. According to our data, the IAC lengths were 12.29 mm and 11.09 mm on coronal and axial sections, respectively. We observed that the IAC lengths on the coronal views were significantly greater than those on the axial views, according to the paired t test (p < 0.05). Selesnick et Al compared the MRI and the intraoperative findings of the lateral IAC, their IAC lengths were 12.7 mm and 13.7 mm on the axial and coronal views, respectively 6. This may be attributed to the shorter distance between the posterior lip of IAC and the fundus. Sakashita et Al reconstructed the IACs of 20 patients by three-dimensional CT technology and concluded that regardless of age, the
www. centauro. it
The Neuroradiology Journal 21: 393-400, 2008
Table 3 Results of the statistical analyses obtained from the computed tomography image of the temporal bone in coronal views
Landmarks
Descriptions
Mean±SD
Range
(1)-(3)
IAC diameter
5.33±1.1 mm
3.34~8.13 mm
(2)-(4)
IAC length
12.29±2.0 mm
6.39~17.62 mm
(8)-(10)
SSC to lateral petrous bone
25.41±2.8 mm
18.91~33.9 mm
(4)-(10)
IAC fundus to lateral petrous bone
29.86±3.2 mm
42.21~19.82 mm
(4)-(5)
IAC fundus to malleoincus joint
10.93±1.7 mm
8.28~17.56 mm
(4)-(8)
IAC fundus to SSC
8.57±2.0 mm
6.22~10.93 mm
(1)-(4)
Superior IAC lip to IAC fundus
12.01±2.2 mm
6.22~17.94 mm
(3)-(4)
Inferior IAC lip to IAC fundus
13.12±2.3 mm
7.51~18.48 mm
(1)-(5)
Superior IAC lip to malleoincus joint
22.59±2.8 mm
16.04~29.89 mm
(3)-(5)
Inferior IAC lip to malleoincus joint
24.06±2.7 mm
17.89~31.8 mm
(6)-(8)
Petrous bone margin to SSC
15.25±2.9 mm
6.2~21.83 mm
(7)-(8)
Roof of SSC to petrous bone
1.70±0.9 mm
0~5 mm
(5)-(9)
Bony EAC to malleoincus joint
14.91±3.0 mm
8.31~21.95 mm
(4)-(9)
Fundus of IAC to bony EAC
25.47±3.1 mm
19.21~33.22 mm
∠(1)(4)(3)
Angle of IAC between fundus
21.77±5.3°
12~36°
∠(1)(5)(3)
Angle of IAC between malleoincus joint
11.29±2.7°
5~17°
anterior wall was the longest whereas the posterior wall was the shortest 7. The greater the distance between the posterior lip of IAC and the fundus was, the more difficult it became to perform internal labyrinthectomy via the retrosigmoid approach 8. The rates of development of the lengths and diameters of the IACs are different. The IAC diameter increased until approximately one year of age while the IAC length increased until late adolescence, i.e., approximately ten years of age 7. Hence, we excluded patients younger than 18 years of age. The average vertical diameter of the IAC (on coronal views) was 5.33 mm, and the average horizontal diameter (on axial views) was 6.92 mm. Thus, the latter was greater than the former, and the difference between them was statistically significant according to the paired t test. The shape of the IAC was flat and narrowing was observed posterolaterally. The retrosigmoid approach for the surgical treatment of vestibular schwannomas has a hearing preservation rate of 18% to 75%, the key to ensure prevention of permanent sensorineural hearing loss is to avoid labyrinthine
injury, especially during lateral IAC dissection . Miller et Al defined the center of the craniotomy during the retrosigmoid approach as 15 mm posterior to the sigmoid sinus 8,11. On an average, the PSC is situated 10.51 mm posterior to the sigmoid sinus. The average distance between the posterior lip of the IAC and the sigmoid sinus was 35.55 mm. The distance from the posterior IAC lip to the vestibulae was the maximum permissible length of the posterior IAC lip that could be removed 4. According to our reports, the distances were 10 mm (from the posterior IAC lip to the common crus) and 14.15 mm (from the posterior IAC lip to the PSC). Ekinci G et Al examined 100 normal ears in 50 subjects and measured the length from the medial lip of the posterior wall of the IAC to the medial wall of the vestibule (close to the common crus in our study) 12. The average distance was 9.7 mm, which was similar to our data. The foremost structures to be visualized after retrolabyrinthectomy were the endolymphatic system and the common crus. However, we observed that there were two definitions of the retrolabyrinthine bones. One is defined as the thickness of the petrous bone overlying 9,10
397
High Resolution Computed Tomography Analysis of the Temporal Bone
Hsiung-Kwang Chung
Table 4 Results of the statistical analyses obtained from the computed tomography image of the temporal bone in the axial views
Landmarks
Descriptions
Mean±SD
Range
(A)-(C)
IAC diameter
6.92±1.4 mm
4.12~10.66 mm
(B)-(D)
IAC length
11.09±2.0 mm
6.93~15.29 mm
(C)-(J)
Posterior IAC lip to sigmoid sinus
35.55±3.8 mm
27.76~44.83 mm
(C)-(F)
Posterior IAC lip to common crus
10.0±1.8 mm
7.26~16.52 mm
(C)-(H)
Posterior IAC lip to PSC
14.15±2.2 mm
8.24~18.71 mm
(H)-(I)
Shortcut of PSC and petrous bone
3.78±2.0 mm
1.01~9.97 mm
(A)-(D)
Anterior IAC lip to IAC fundus
13.87±1.9 mm
9.37~18.25 mm
(C)-(D)
Posterior IAC lip to IAC fundus
9.8±2.1 mm
5.67~15.06 mm
(A)-(E)
Anterior IAC lip to malleoincus joint
22.62±1.9 mm
18.41~27.66 mm
(C)-(E)
Posterior IAC lip to malleoincus joint
18.72±2.1 mm
12.91~22.95 mm
(D)-(H)
PSC to IAC fundus
9.90±1.6 mm
5.4~14.12 mm
(D)-(J)
Sigmoid sinus to IAC fundus
29.86±2.8 mm
24.2~35.45 mm
(F)-(G)
Shortcut of common crus and petrous bone
5.26±1.4 mm
3.86~8.66 mm
(E)-(J)
Malleoincus joint to sigmoid sinus
25.39±3.1 mm
20.56~37.24 mm
(H)-(J)
PSC to sigmoid sinus
10.51±1.5 mm
8.64~13.55 mm
(F)-(J)
Common crus to sigmoid sinus
11.27±1.8 mm
8.82~15.06 mm
∠(A)(D)(C)
Angle of IAC between fundus
28.95±7.7°
10~48°
∠(A)(E)(C)
Angle of IAC between malleoincus joint
15.97±3.9°
9~26°
Table 5 Differences between the measurements of normal ears and diseased ears
Landmarks and descriptions
Normal Ears
Diseased Ears
p value
Coronal views
5.38±1.0 mm
5.31±1.2 mm
0.7660
Axial views
6.83±1.4 mm
6.97±1.4 mm
0.6497
Coronal views
12.25±2.0 mm
12.31±2.0 mm
0.8795
Axial views
11.36±1.9 mm
10.95±2.0 mm
0.3645
SSC to lateral petrous bone
25.33±3.3 mm
25.45±2.6 mm
0.8636
IAC fundus to malleoincus joint
11.02±1.5 mm
10.88±1.8 mm
0.7311
Angle of IAC between malleoincus joint
11.17±2.6°
11.36±2.8°
0.7557
Posterior IAC lip to sigmoid sinus
35.67±3.7 mm
35.48±3.9 mm
0.8203
Posterior IAC lip to PSC
14.08±2.1 mm
14.19±2.2 mm
0.8251
Shortcut of PSC and petrous bone
3.99±2.4 mm
3.66±1.8 mm
0.4672
IAC diameter
IAC length
398
www. centauro. it
the PSC; the other is the thickness of the common crus 4,8. We measured both the thickness values on the axial views, and the results were 3.78 mm (thickness of the petrous bone overlying the PSC) and 5.26 mm (thickness of the common crus). The aforementioned data would provide preoperative guidelines for the surgeon to avoid any labyrinthine injury and to obtain optimal potential for hearing preservation. The fundus of the IAC can be widely exposed during translabyrinthine approach, therefore, complete tumor removal is possible regardless of the tumor size. However. the translabyrinthine approach may destroy the residual hearing in the affected ear because of labyrinthectomy. The sigmoid sinus will be decompressed prior to labyrinthectomy, so we measured the distances from the sigmoid sinus and the common crus and that to the PSC; they were 11.27 mm and 10.51 mm, respectively. The average distance between the fundus of the IAC and the sigmoid sinus was 29.86 mm. The short process of the incus can be identified as an oblique projection during translabyrinthine approach. According to our measurements, the average distance between the sigmoid sinus and the malleoincus joint was 25.39 mm and that between the PSC and the fundus was 9.9 mm. These landmarks and measurements can provide guidelines to the surgeon in outlining the fundus of the IAC. The middle fossa approach had the best potential for hearing preservation provided that the maximum dimension of the tumor did not exceed 2 cm. Factors including poor angle of view to the IAC, blind dissection, and intraoperative temporal lobe retraction made this approach technically more difficult 5. Further, the patient ran the risk of neurological sequelae such as facial nerve paralysis, seizure, and hemiparesis. Further, epidural hematoma, meningitis, and cerebrospinal fluid leakage have been reported as life-threatening complications associated with this approach 13. We drew an imaginary line from the margin of the petrous bone to the upper limit of the SSC and extended it to the lateral skull bone. The point where the line and the lateral skull bone met was defined as the lower limit of the bone flap for temporal craniotomy. On the coronal views, the average distances from this point to the upper limit of the SSC and to the IAC fundus
The Neuroradiology Journal 21: 393-400, 2008
were 25.41 mm and 29.86 mm, respectively. Fisch and Tsunoda et Al proposed that the anatomical position between the IAC and the SSC appeared to be relatively constant. Therefore, the labyrinthine bone could be drilled until the contour of the SSC was exposed 14,15. According to the results, the average distance between the upper limit of the SSC and the IAC fundus was 8.57 mm on average. According to our report, the average thickness of the bone overlying the SSC on the coronal views was 1.7 mm. Kanzaki proposed that the thickness was 2.4 mm (SD, 1.3 mm; range 0.8-5.8 mm) 3. Carey et Al investigated 100 control specimens screened from 1000 temporal bones and reported that the average thickness was 1.79 mm (SD, 1.20 mm; range, 0.13-6.24 mm) 16. The diagnosis of SCD was established on the basis of history, physical examination and confirmation of bony defect overlying the SCC on the temporal bone HRCT 17. There were two patients with zero bone thickness. Due to the limited resolution, we preferred to define them as “thinning” rather than “dehiscence.” The incidence rate of thinning was 2.3%. However, only one patient had clinical vestibulopathy, the other ear was within the normal groups. Carey et Al studied 1000 temporal bones of 596 adults. In all, there were 19 specimens with evidence of thinning or dehiscence, giving an incidence rate of 1.9% 16. Conclusions There were no major differences in the distances between the selected landmarks of the temporal bone on HRCT analysis with regard to the diagnoses, ages and genders of the patients. Our estimated data were similar to the publications by others. Several specific measurements such as the thickness of the retrolabyrinthine bone, the IAC lengths, and the distance between the sigmoid sinus and the IAC can be applied to the preoperative evaluation of vestibular schwannoma. The surgeon should choose the most appropriate surgical approach on the basis of the anatomical characteristics, their preferences, and their technical expertise. The incidence of thinning bone overlying the SSC was 2.3%; patients with such findings should be careful to avoid barotrauma and accidental head injury.
399
High Resolution Computed Tomography Analysis of the Temporal Bone
Hsiung-Kwang Chung
References 1 Bartling SH, Peldschus K, Rodt T et Al: Registration and fusion of CT and MRI of the temporal bone. J Comput Assist Tomogr 29: 305-10, 2005. 2 Jun BC, Song SW, Cho JE et Al: Three-dimensional reconstruction based on images from spiral high-resolution computed tomography of the temporal bone: anatomy and clinical application. J Laryngol Otol 119: 693-8, 2005. 3 Kanzaki J: The surgical treatments of acoustic neuromas. Oto-Rhino-Laryngology Tokyo 33: 201-528, 1990. 4 Day JD, Kellogg JX, Fukushima T et Al: Microsurgical anatomy of the inner surface of the petrous bone: neuroradiological and morphometric analysis as an adjunct to the retrosigmoid transmeatal approach. Neurosurgery 34: 1003-1008, 1994. 5 Driscoll CLW, Jackler RK, Pitts LH et Al: Is the entire fundus of the internal auditory canal visible during the middle fossa approach for acoustic neuroma? Am J Otol 21: 382-388, 2000. 6 Selesnick SH, Rebol J, Heier LA et Al: Internal auditory canal involvement of acoustic neuromas: Surgical correlated to magnetic resonance imaging findings. Otol Neurotol 22: 912-916, 2001. 7 Sakashita T, Sando I: Postnatal development of the internal auditory canal studied by computer-aided threedimensional reconstruction and measurement. Ann Otol Rhinol Laryngol 104: 469-475, 1995. 8 Miller RS, Pensak ML: An anatomic and radiologic evaluation of access to the lateral internal auditory canal via the retrosigmoid approach and description of an internal labyrinthectomy. Otol Neurotol 27: 697-704, 2006. 9 Rhoton AL Jr.: Microsurgical removal of acoustic neuromas. Surg Neurol 6: 211-219, 1976. 10 Magnan J, Barbieri M, Mora R et Al: Retrosigmoid approach for small and medium-sized acoustic neuromas. Otol Neurotol 23: 141-145, 2002. 11 Blevins NH, Jackler RK: Exposure of the lateral extremity of the internal auditory canal through the retrosigmoid approach: a radioanatomic study. Otolaryngol Head Neck Surg 111: 81-90, 1994. 12 Ekinci G, Koc A, Baltacioglu F et Al: Temporal bone measurements on high-resolution computed tomography. J Otolaryngol 33: 387-9, 2004. 13 Bryce GE, Nedzelski JM, Rowed DW et Al: Cerebrospinal fluid leaks and meningitis in acoustic neuroma surgery. Otolaryngol Head Neck Surg 104: 81-87, 1991.
400
14 Tsunoda A, Komatsuzaki A, Kobayashi M et Al: Threedimensional image for the middle fossa approach and its anatomical considerations. Laryngoscope 111: 10481052, 2001. 15 Fisch U: Transtemporal surgery of the internal auditory canal: report of 92 cases, technique, indications and results. Adv Otorhinolaryngol 17: 203-240, 1970. 16 Carey JP, Minor LB, Nager GT: Dehiscence or thinning of bone overlying the superior semicircular canal in a temporal bone survey. Arch Otolaryngol Head Neck Surg 126: 137-147, 2000. 17 Hillman TA, Kertesz TR, Hadley K et Al: Reversible peripheral vestibulopathy: the treatment of superior canal dehiscence. Otolaryngol Head Neck Surg 134: 431-436, 2006.
Chia-Der Lin, MD Department of Otorhinolaryngology, Head & Neck Surgery China Medical University Hospital No.2 Yu-Te Road Taichung city (403) Taiwan (R.O.C.) Tel.: +886-4-2205-2121 ext 4436 Fax: +886-4-2205-2121 ext 4438 E-mail: [email protected]
