- Email: [email protected]
Usefulness of Oblique Coronal Computed Tomography and Magnetic Resonance Imaging in the Endoscopic Endonasal Approach to Treat Skull Base Lesions
Accepted Manuscript Usefulness of oblique coronal CT and MRI imaging in the endoscopic endonasal approach to treat skull base lesions Shigetoshi Yano, MD., PhD., Naoki Shinojima, MD., PhD., Mika Kitajima, MD., PhD., Hiroyuki Uetani, MD., PhD., Takuichiro Hide, MD., PhD., Akitake Mukasa, MD., PhD. PII:
S1878-8750(18)30062-7
DOI:
10.1016/j.wneu.2018.01.022
Reference:
WNEU 7220
To appear in:
World Neurosurgery
Received Date: 9 August 2017 Revised Date:
1 January 2018
Accepted Date: 4 January 2018
Please cite this article as: Yano S, Shinojima N, Kitajima M, Uetani H, Hide T, Mukasa A, Usefulness of oblique coronal CT and MRI imaging in the endoscopic endonasal approach to treat skull base lesions, World Neurosurgery (2018), doi: 10.1016/j.wneu.2018.01.022. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Usefulness Usefulness of oblique coronal CT and MRI imaging in the endoscopic endonasal
RI PT
approach to treat skull base lesions lesions
Shigetoshi Yano, MD., PhD.1, Naoki Shinojima MD., PhD.1, Mika Kitajima, MD., PhD.2,
SC
Hiroyuki Uetani MD., PhD.2, Takuichiro Hide, MD., PhD.1, Akitake Mukasa, MD.,
1Department
M AN U
PhD.1
of Neurosurgery, Faculty of Life Sciences, Kumamoto University2
Naoki Shinojima
[email protected]
[email protected]
EP
Mika Kitajima
TE D
Department of radiology, Faculty of Life Sciences, Kumamoto University
AC C
Hiroyukih Uetani
[email protected]
Takuichiro Hide
[email protected]
Akitake Mukasa
[email protected]
Corresponding author: Shigetoshi Yano Address: 1-1-1 Honjo, Chuo-ku, Kumamoto 860-8556, JAPAN
ACCEPTED MANUSCRIPT
Tel: +8196-373-5219; Fax: +8196-371-8064
RI PT
Email: [email protected]
Key Words: endoscopic endonasal transsphenoidal approach, skull base tumor,
SC
preoperative diagnosis, oblique coronal
AC C
EP
TE D
M AN U
Short title: Oblique coronal imaging for endoscopic transsphenoidal surgery
ACCEPTED MANUSCRIPT
Abstract:
RI PT
Objective:
This report examines the utility of the preoperative image to orientate the surgeon in
SC
the sphenoid sinus during an endoscopic endonasal transsphenoidal approach (ETSS).
M AN U
Methods:
ETSS was performed in 100 cases of sellar lesion and used to classify the sphenoid sinus septum shape. Preoperative CT and MRI scans were taken for two types of coronal
TE D
imaging: conventional and oblique. Expected sphenoid sinus septum shape was compared with those from ETSS to estimate concordance. The confirmation rate of
EP
anatomical landmarks in the sphenoid sinus by endoscopic observation was compared
AC C
in various types of septum and the identification rate in oblique coronal imaging was also examined.
Results: The most common septum shape was single type (31%), followed by branched (26%), parallel (18%), none (12%), cross (9%), and bridge (4%) types. In oblique coronal images,
ACCEPTED MANUSCRIPT
preoperative evaluation and endoscopic findings were consistent in 93 to 100% of cases. However, with conventional coronal images, the concordance rate was 22.2 to 83.9%,
RI PT
and in the none, branched, and cross types, the concordance rate was significantly lower than that for oblique coronal images. Although confirmation of the midline through
SC
estimation of landmarks by endoscopic observation was difficult in 33 cases,
best indicated the midline.
Conclusion:
M AN U
preoperative CT and MRI showed landmarks in all cases and oblique coronal images
TE D
Use of oblique coronal images in addition to conventional images provided good orientation of anatomical structures in the sphenoid sinus. The combination of
AC C
EP
preoperative imaging and endoscopic observation could allow safer surgery in ETSS.
247words
ACCEPTED MANUSCRIPT
Introduction： Introduction： Use of the endoscopic endonasal transsphenoidal approach has grown recently. While
RI PT
endoscope provides wide and bright vision to the surgeon, a steep learning curve is necessary to master this technique due to the tight working space and difficulty in
SC
obtaining orientation (1) (2, 3).
Even an experienced surgeon can make mistakes when approaching the sella turcica,
injury to the internal carotid artery
M AN U
particularly when an endoscope is used as the sole surgical tool. In these instances, (4) (5),
dural laceration of frontal base or posterior
fossa or unexpected CSF leakage before opening the sellar floor has been reported (6-10) .
TE D
As a cause of such disastrous complications, disorientation at the time of opening the sallar floor or removal of the tumor can be considered. Therefore, careful simulation
EP
before surgery is important.
AC C
For intraoperative orientation, various supporting systems have been developed recently. Virtual endoscopy
(11-17)
or neuronavigation
(6, 7, 18-20)
are powerful assistance
technique. However, preparation of virtual endoscopy requires time and effort, and using a navigation system may be restricted in a situation where equipment is not prepared. It would be a challenge to employ these systems for all patients in every institute.
ACCEPTED MANUSCRIPT
We have been using the endoscopic endonasal transsphenoidal approach (ETSS) since the year 2001 using above supporting systems. Although these assistance devices are
RI PT
certainly useful, we have felt the necessity of images that operator can recognize more intuitively during surgery, because the endoscopic image is a 2D image and sometimes
SC
it may be different from the surgeon's expectation. Also, several landmarks are noted as being important in current literature, but few studies have shown the evaluation
M AN U
methods against actual endoscopic images (10, 21-23).
We prepared oblique coronal CT and MRI images, obtained as perpendicular slices to the penetration direction of the endoscope into the nasal cavity. While these images are
TE D
expected to be similar to two-dimensional endoscopic images, it is unclear whether the matching rate between the preoperative image and the actual surgical image is
EP
different in the type of the septum, or which type of septum is difficult to recognize the
AC C
structure of sella floor and surrounding surgical landmarks. In this study, the usefulness of preoperative simulation using oblique coronal images for endoscopic approach to the sellar floor was then evaluated.
ACCEPTED MANUSCRIPT
Materials and Methods:
Faculty of life sciences Kumamoto University.
Clinical materials
RI PT
This research is approved by Ethics Committee on Clinical Research for Humans in
SC
From January 2011 to December 2015 (excluding cases of reoperation and subjects with unclear sphenoid sinus septation due to the presence of extended tumors) 100 patients
M AN U
undergoing ETSS at Kumamoto University were included in this study. Preoperative CT, MRI and intraoperative endoscopic views were reviewed for these 100
TE D
cases retrospectively.
Coronal CT and MRI images
EP
Conventional coronal images were obtained by slicing perpendicular to the orbitomeatal
AC C
line for CT and perpendicular to the anterior commissure–posterior commissure (AC-PC) line for MRI.
Oblique coronal CT and MRI images were made using slices perpendicular to the line connecting the nasal cavity and the dorsum sellae, a path comparable to the endoscope insertion axis. CT was taken under the condition of 2 mm thickness with 2 mm gap after injection of contrast media. MRI was carried out as post contrast spin echo T1
ACCEPTED MANUSCRIPT
images with 3 mm thickness and 0.3 mm gap.
RI PT
Surgical approach
For approaching the sellar floor, a 4-mm rigid endoscope (KARL STORZ HOPKINS II
SC
rigid endoscope, 0° or 30°, Tuttlingen, Germany) was inserted into each nostril and after removing the bony septum, was used to connect the bilateral sphenoid ostia. The
(24).
M AN U
sphenoid sinus anterior wall was opened as wide as possible, as we described previously After entering the sphenoid sinus, the bony septum, sellar floor, prominence of the
internal carotid artery, clival recess, optic canal, optico-carotid recess (OCR), and
TE D
tuberculum sellae were all located as anatomical landmarks. The bony septum was drilled carefully to entirely expose the sellar floor.
EP
For tumors extending over the sella turcica, bone from the skull base surrounding the
AC C
sellar floor was opened after preparation of the septal flap (24). The neuronavigation system (Medtronic Co., Ltd., Cororado, USA) was used for cases with tumors extending in frontal, cavernous, or clival directions or tumors encasing the internal carotid artery. Surgical images were captured and recorded.
Evaluation of the utility of each coronal slice
ACCEPTED MANUSCRIPT
The type of bony septum attached to the sellar floor in the sphenoid sinus was classified according to the observed shape on the two-dimensional CT, MRI, and endoscopic view
RI PT
images as follows: none (O shape), no obvious septum attached to the sellar floor; single (I shape), one major septum runs over the sellar floor; parallel (II shape), two or three
SC
septa run side by side on the sellar floor; branched (Y shape), septum branched and attached to the sellar floor; bridge (H shape), two parallel septa connect to one
M AN U
horizontal septum; cross (+ shape), two septa cross on the sellar floor in either a + or X shape. During the operation, each surgeon judged the septum type through endoscopic observation when the sphenoid sinus was opened and landmarks surrounding the sellar
TE D
floor could be identified.
Images of oblique coronal and conventional coronal of CT and MRI prepared before
EP
surgery were randomly observed by three surgeons after hiding the patient’s name and
AC C
the pattern of the septum passing over the sellar floor was evaluated according to the classification of intraoperative endoscopic images. Analysis was performed regardless of whether the preoperative image type matched the type observed in the operative field. Surrounding landmarks were observed and evaluated as to how many of four structures (the bilateral internal carotid prominence and bilateral OCR) could be observed during
ACCEPTED MANUSCRIPT
surgery. Number of landmarks was compared with classified septum type.
RI PT
Statistical analysis
The concordance rate of septum type classified via CT or MRI imaging with that as
SC
judged through endoscopic observation was estimated by Fisher exact test or Chi square
AC C
EP
TE D
M AN U
test. Values less than 0.05 were deemed to be significant.
ACCEPTED MANUSCRIPT
Results:
RI PT
Clinical characteristics of the 100 cases included in this study are summarized in Table 1. The differences between conventional and oblique coronal images were shown in Fig.
SC
1 as a representative case. In this case, only one septum was observed on the sellar floor in the conventional coronal image (Fig. 1b). However, three complicated septa were
M AN U
observed on the oblique coronal image (Fig. 1d); two septa attached to the sellar floor, bulging to the right side, a branch from the right septum connected to the midline septum; the final septum ran towards the left optic canal.
TE D
Six types of septum were classified through surgical findings. Corresponding oblique coronal and conventional coronal images for representative cases are shown in Fig. 2
EP
and 3.
AC C
In a case of none type, oblique coronal images identified bilateral septa attached to each carotid prominence but neither were attached to the sellar leading to classification as none type. In the same case, conventional coronal images identified one septum running to right side of the sellar floor; hence, it was classified as single type preoperatively (Fig. 2a, 2b, 2c). In a case of single type, a curved septum was observed to run from left inferior to right
ACCEPTED MANUSCRIPT
superior in the oblique coronal image, but shown as running from the midline towards the right on the conventional coronal image (Fig. 2e, 2f). Although both images resulted
RI PT
in single type classification, the precise position of sellar floor attachment was different. Operative view confirmed results from the oblique coronal image (Fig. 2d).
SC
In a case of parallel type, right OCR was not observed due to attachment of the right septum, the left septum ran along the carotid prominence lateral to the left OCR in the
M AN U
operative view (Fig. 2g). In the oblique coronal image, right OCR was observed lateral to the right septum (Fig. 2h). In the conventional coronal image, attachment of the right
recognized (Fig. 2i).
TE D
septum was the same as in the oblique coronal image but the left septum was not
Results for cases with more complicated septa are shown in Fig. 3.
EP
In a case with branched type, operative view revealed a Y-shaped septum attached to
AC C
the sellar floor (Fig. 3a). The left septum was attached to the carotid prominence, this was identical to that observed in the oblique coronal image (Fig. 3b). In the conventional coronal image, a Y-shaped septum was also observed but the septum was observed to run along the midline (Fig. 3c). In a case of bridge type, operative view revealed an H shape (Fig. 3d), bilateral septa ran along bilateral carotid prominences and connected to the sellar floor. This was
ACCEPTED MANUSCRIPT
confirmed with the oblique coronal image (Fig. 3e); however, a Y shape septum was observed on the conventional coronal image (Fig. 3f).
RI PT
In a case of cross type, both oblique coronal and conventional coronal images show a + shaped septum (Fig 3h, 3i). The cross point was located operatively and on the oblique
SC
coronal image in corresponding locations on the sellar floor (Fig 3g, 3h), while the conventional coronal image located the cross point as being beneath the sellar floor (Fig.
M AN U
3i).
In intraoperative observation of 100 cases, single type was the most common (31 cases),
TE D
following this, incidence was branched type (26 cases), parallel type (18 cases), none type (12 cases), cross type (9 cases) and bridge type (4 cases) (Table 2). When we
EP
examined the concordance rates of operative view classification with either preoperative
AC C
oblique coronal images or conventional coronal images for each septum type, high concordance rates (more than 93%) were recorded for oblique coronal images for all types of septum, while relatively low concordance rates (22 to 83%) were observed for conventional coronal images. This rate was significantly low in conventional coronal images for none, branched, and cross types. Although oblique coronal images almost always corresponded to the operative view, conventional coronal images sometimes
ACCEPTED MANUSCRIPT
highlighted an underestimation of parallel and bridge types.
RI PT
We also recorded how many surgical landmarks such as the bilateral carotid prominence and bilateral OCR (a total of four structures), were observed in the surgical
SC
field by endoscopic observation (Table 3). Cases with more than three landmarks were recorded in 5 of 12 (41.7%) none type, 13 of 31 (41.9%) single type, 8 of 18 (44.4 %)
M AN U
parallel type, 14 of 26 (53.8%) branched type, 3 of 4 (75.0%) bridge type, and 3 of 9 (33.3%) cross type. Landmarks were most easily confirmed in bridge type and least
TE D
often confirmed in cross type, although this difference was not statistically significant.
Through surgical findings, confirmation of the midline was straightforward in 46 cases
EP
since these cases identified more than 3 landmarks. However, in 31 cases one or no
AC C
landmarks were observed, and there were 2 cases where landmarks appeared unilaterally. In oblique coronal images, attachment position of the septum and relationship to the landmarks were clearly observed preoperatively in all cases and showed high concordance rates with operative findings, regardless of septum type.
ACCEPTED MANUSCRIPT
Case presentation:
RI PT
Case 1 (Fig. 4)
A 68-year-old man with non-functioning pituitary adenoma. Although conventional
SC
coronal MRI imaging showed a pituitary tumor extending to the suprasellar region with two parallel septa attached to the sellar floor on the left side of the sphenoid sinus (Fig.
M AN U
4a), oblique coronal MRI and CT images indicated that the sellar floor was not enlarged and branched type septum attached to the left side of the sellar floor (Fig 4b, 4c), which was similar to the intraoperative endoscopic view (Fig. 4d). Intraoperatively, while left
TE D
OCR was identified, the right OCR was covered by septum or bone and hence, could not be identified. The midline was not easily decided through endoscopic view, but the
EP
oblique coronal image indicated the midline to be on the sellar floor or tuberculum sellar
AC C
via an existing septum. Marking was performed after confirmation by neuronavigation which verified the septum attachment position predicted by oblique coronal images (Fig. 4d).
Case 2 (Fig. 5) A 20-year-old man with acromegaly. Preoperative oblique coronal CT and MRI showed a
ACCEPTED MANUSCRIPT
tumor that had enlarged the sellar floor and extended into the right cavernous sinus. Cross type septum that run parallel beneath the sellar floor and left side septum made
RI PT
border of the sellar floor was expected (Fig. 5a, 5b), which was identical to the intraoperative view (Fig. 5c). The midline was easily identified by the position of septum
SC
and bilateral optic canals by reffurring to preoperative oblique images, although it would have been difficult to confirm the midline by endoscopic view only. Sellar floor
AC C
EP
TE D
neuronavigation system.
M AN U
was opened safely (Fig. 5d) and the tumor removed successfully without the aid of the
ACCEPTED MANUSCRIPT
Discussion： Discussion： In the endoscopic endonasal approach, the most important factor is the direction of
RI PT
endoscope entry. Since the endoscopic image has a wide angle, a slight directional change in the entry of the nasal cavity results in a large deviation of arrival point in the
SC
sphenoid or ethmoid sinuses. Further to this, the endoscope can rotate in the surgeon’s hand, so constant checking of the vertical plane and midline is critical. Indeed, there are
M AN U
some reports of disorientation-related internal carotid artery injury (5, 25-27). Confirmation of bilaterally symmetrical structures such as the carotid prominence or optic prominence is important to measure the midline. The positions of tuberculum
TE D
sellae and clivus recess can be used as guides to locate the sella turcica, and this is especially helpful in flat floor types. However, in the sphenoid sinus, some patterns of
EP
bony septum may disrupt the observation of these landmarks.
AC C
Variation of the sphenoid septation has been well known for some time
(28, 29)
and it is
clear that the most commonly presenting septum does not necessarily lie along the midline. Recently, Twigg et al.
(30)
assessed 100 cases of CT and MRI, highlighting the
significant anatomical variations in patients with pituitary disease. They stressed the need to be aware of unusual septation across the internal carotid artery at sella level and the possibility of deviation of the internal carotid artery itself. Also prevalence of
ACCEPTED MANUSCRIPT
the Onodi cell which is identified as the most posterior ethmoid cell and extend to the optic nerve or internal carotid artery was reported as 39.8% of 196 paranasal sinus (31).
It is also claimed that the risk of optic nerve injury increases
RI PT
examination by CT
when the identification of the Onodi cell during the operation.
(32) (8) (33)
We have also
SC
observed the septum running over the carotid prominence and optic prominence in many cases; hence, in this study, we examined the relationship between septation
M AN U
variation and anatomical landmarks surrounding the sella turcica.
In order to recognize the positional relationship between front, rear, left, and right
TE D
during surgery, we chose the four landmarks of bilateral carotid prominences and bilateral OCR. Under endoscopic observation, only 19% of cases had all four landmarks
EP
confirmed, and 33% of cases had less than one observable landmark. This indicates the
AC C
difficulty of midline confirmation solely through endoscopic observation. Although no significant difference was observed in the number of landmarks observed for each type of sphenoid sinus septum, there was no case in which all four landmarks could be confirmed in cross type of septum and this type showed the lowest landmark confirmation rate. These results indicate the importance of recognizing the septum attachment point to skull base structures by CT and MRI before surgery.
ACCEPTED MANUSCRIPT
Although CT clearly depicts septum, MRI can not visualize bone structures. However, it is possible to recognize the shape of the septum from the contrast of the mucous
RI PT
membrane adhering to the sinus septum wall. Oblique MRI shows the shape and extension of the tumor well in the direction of the endoscope and well depicts the
SC
relation with the internal carotid arteries and the cavernous sinus. Although contrast CT is sufficient to evaluate the shape of septum, attention is paid to the usefulness of
M AN U
simultaneously recognizing the detailed relationship between the tumor to be removed and the surrounding structure, and MRI is also used in the same imaging method.
simulation.
TE D
Combined use of both CT and MRI should improve the quality of the preoperative
While the slice direction of conventional coronal CT and MRI images may vary between
EP
different institutes, our institute adopts a method of slicing perpendicularly to OM or
AC C
AC-PC lines. Indeed, the conventional method is particularly useful for the qualitative diagnosis of pituitary gland and pituitary tumor, however, concordance rate of septum type identification with operative endoscopic findings was low. In conventional coronal images, branched type and cross type septa were significantly more commonly predicted prior to surgery and false positive septa were predicted in none type. Moreover, position of septum attachment was sometimes misread as the midline. Oblique coronal images
ACCEPTED MANUSCRIPT
gave predictions consistent with endoscopic images in more than 94% of cases. It was most reliable to predict the midline location prior to surgery using oblique
RI PT
coronal images to estimate the relationship between septum attachment location and surrounding landmarks.
SC
Even in a case where the landmark was unclear in the intraoperative endoscopic image, the oblique coronal images provided clarity and so clear identification of the structure
M AN U
behind the septum attachment. In previous reports, bone window CT scans sliced parallel to the transsphenoidal surgical route were introduced
(34),
and Twigg et al.
(30)
classified the 5 types of septation according to attachment position relative to the
TE D
internal carotid artery. Certainly, these techniques are useful preoperative prediction methods, but our oblique coronal scan has provided preoperative simulation that more
AC C
EP
closely matches intraoperative endoscopic image findings.
To improve the orientation of the surgeon during endoscopic endonasal surgery, various devices and methods have been recommended including, virtual endoscopy
(11-17),
neuronavigation (6, 7, 18-20), indocyanine green endoscopy(35), endonasal ultrasound (6, 7, 36), and intraoperative MRI (37-40). Virtual endoscopy resembles our oblique coronal imaging concept in terms of mimicking
ACCEPTED MANUSCRIPT
the endoscopic view, however, preparation of virtual images takes time and costs more than oblique coronal imaging. Neuronavigation can also be useful in confirming the
RI PT
midline and surrounding structures, however the operator needs to check the orientation at an appropriate timing unless navigation is a persistent monitor. Prior to
SC
the surgery, it is important for the surgeon to confirm the variation in structure and position of the midline. Thus, expanding the surgeon’s resources from conventional
M AN U
coronal images to include oblique coronal images, should make endoscopic endonasal
AC C
EP
TE D
skull base approach surgery safer for all.
ACCEPTED MANUSCRIPT
Figure Legends
Comparison of conventional coronal CT and oblique coronal CT images
RI PT
Fig. 1
a) Scout view of conventional coronal CT. Images were obtained perpendicular to the
SC
orbitomeatal line. b) Representative bone image of conventional CT indicating single bony septum in sphenoid sinus. c) Scout view of oblique coronal CT of the same patient.
M AN U
Images were obtained perpendicular to the line connecting the nasal cavity and the dorsum sellae. d) Representative bone image CT corresponding to panel b. Complex
Representative images of three septum types identified through operative
EP
Fig. 2
TE D
branched type septum was observed.
AC C
endoscopic images, preoperative oblique coronal CT images and preoperative conventional CT images. Septum type was classified by endoscopic findings as none type (a), single type (d), and parallel type (g). Corresponding preoperative oblique coronal images (b, e, h) and conventional coronal images (c, f, i) are shown for comparison.
ACCEPTED MANUSCRIPT
Fig. 3
Representative images of remaining three septum types identified through
RI PT
operative endoscopic images, preoperative oblique coronal CT images and preoperative conventional CT images.
SC
Remaining septum types were classified by endoscopic findings as branched type (a), bridge type (d) and cross type (g). Corresponding preoperative oblique coronal images (b,
Preoperative CT and MRI images and intraoperative image of Case 1
TE D
Fig. 4
M AN U
e, h) and conventional coronal images (c, f, i) are shown for comparison.
Preoperative Gd-enhanced MRI of conventional coronal image (a) and corresponding
EP
oblique coronal image (b). Corresponding image of oblique coronal CT (c) indicated
AC C
branched type septum in sphenoid sinus. Endoscopic image of sellar floor after opening the sphenoid sinus (d). Branched septum runs from the left side to the sellar floor midline. The OCR and carotid prominence were observed on the left side, but landmarks of the right side were not clearly observed. Midline was marked with blue dye.
ACCEPTED MANUSCRIPT
Fig. 5
Representative images of Case 2
RI PT
Preoperative oblique coronal CT (a) and MRI (b) showed that the tumor enlarged the sellar floor and extended to the right cavernous sinus with a cross type septum running
SC
along the sellar floor. Intraoperative endoscopic view confirmed this (c). After opening the sellar floor, the dura was exposed and the midline confirmed by the sphenoidal
AC C
EP
TE D
M AN U
septum (d).
ACCEPTED MANUSCRIPT
References
Leach P, Abou-Zeid AH, Kearney T, Davis J, Trainer PJ, Gnanalingham KK.
RI PT
1.
Endoscopic transsphenoidal pituitary surgery: evidence of an operative learning curve.
Neurosurgery. 2010;67:1205-1212. 2.
Shikary T, Andaluz N, Meinzen-Derr J, Edwards C, Theodosopoulos P, Zimmer LA.
SC
Operative Learning Curve After Transition to Endoscopic Transsphenoidal Pituitary Surgery. World neurosurgery. 2017;102:608-612.
Chi F, Wang Y, Lin Y, Ge J, Qiu Y, Guo L. A learning curve of endoscopic
M AN U
3.
transsphenoidal surgery for pituitary adenoma. The Journal of craniofacial surgery. 2013;24:2064-2067.
4.
Carr S, Twigg V, Mirza S. Radiological study of the anatomy of the keystone area of
the nasal septum using computed tomography to aid septal surgery. Clinical otolaryngology :
official
journal
of
ENT-UK
;
official
journal
of
Netherlands
Society
for
5.
TE D
Oto-Rhino-Laryngology & Cervico-Facial Surgery. 2016;41:317-320. Gardner PA, Tormenti MJ, Pant H, Fernandez-Miranda JC, Snyderman CH,
Horowitz MB. Carotid artery injury during endoscopic endonasal skull base surgery:
Yano S, Kawano T, Kudo M, Makino K, Nakamura H, Kai Y, Morioka M, Kuratsu J.
AC C
6.
EP
incidence and outcomes. Neurosurgery. 2013;73:ons261-269; discussion ons269-270.
Endoscopic endonasal transsphenoidal approach through the bilateral nostrils for pituitary adenomas. Neurologia medico-chirurgica. 2009;49:1-7.
7.
Saeki N, Horiguchi K, Murai H, Hasegawa Y, Hanazawa T, Okamoto Y. Endoscopic
endonasal pituitary and skull base surgery. Neurologia medico-chirurgica. 2010;50:756-764.
8.
Meybodi AT, Vigo V, Benet A. The Onodi Cell: An Anatomic Illustration. World
neurosurgery. 2017;103:950 e955-950 e956. 9.
Shin JH, Kim SW, Hong YK, Jeun SS, Kang SG, Cho JH, Park YJ. The Onodi cell:
ACCEPTED MANUSCRIPT
an obstacle to sellar lesions with a transsphenoidal approach. Otolaryngology--head and
neck surgery : official journal of American Academy of Otolaryngology-Head and Neck Surgery. 2011;145:1040-1042. Kassam AB, Prevedello DM, Carrau RL, Snyderman CH, Thomas A, Gardner P,
RI PT
10.
Zanation A, Duz B, Stefko ST, Byers K, Horowitz MB. Endoscopic endonasal skull base surgery: analysis of complications in the authors' initial 800 patients. Journal of
neurosurgery. 2011;114:1544-1568.
Neubauer A, Wolfsberger S. Virtual endoscopy in neurosurgery: a review.
SC
11.
Neurosurgery. 2013;72 Suppl 1:97-106.
Rosseau G, Bailes J, del Maestro R, Cabral A, Choudhury N, Comas O, Debergue P,
M AN U
12.
De Luca G, Hovdebo J, Jiang D, Laroche D, Neubauer A, Pazos V, Thibault F, Diraddo R. The development of a virtual simulator for training neurosurgeons to perform and perfect endoscopic endonasal transsphenoidal surgery. Neurosurgery. 2013;73 Suppl 1:85-93.
13.
de Notaris M, Solari D, Cavallo LM, Ensenat J, Alobid I, Soria G, Gonzalez JB,
Ferrer E, Prats-Galino A. The use of a three-dimensional novel computer-based model for
TE D
analysis of the endonasal endoscopic approach to the midline skull base. World
neurosurgery. 2011;75:106-113; discussion 136-140. 14.
Zymberg S, Vaz-Guimaraes Filho F, Lyra M. Neuroendoscopic training:
15.
AC C
2010;53:44-46.
EP
presentation of a new real simulator. Minimally invasive neurosurgery : MIN.
Virtual
Tolsdorff B, Pommert A, Hohne KH, Petersik A, Pflesser B, Tiede U, Leuwer R. reality:
a
new
paranasal
sinus
surgery
simulator.
The
Laryngoscope.
2010;120:420-426.
16.
Schulze F, Buhler K, Neubauer A, Kanitsar A, Holton L, Wolfsberger S.
Intra-operative virtual endoscopy for image guided endonasal transsphenoidal pituitary surgery. International journal of computer assisted radiology and surgery. 2010;5:143-154.
17.
Neubauer A, Wolfsberger S, Forster MT, Mroz L, Wegenkittl R, Buhler K.
Advanced virtual endoscopic pituitary surgery. IEEE transactions on visualization and
ACCEPTED MANUSCRIPT
computer graphics. 2005;11:497-507. 18.
Duque SG, Gorrepati R, Kesavabhotla K, Huang C, Boockvar JA. Endoscopic
endonasal transphenoidal surgery using the BrainLAB(R) Headband for navigation without
RI PT
rigid fixation. Journal of neurological surgery Part A, Central European neurosurgery. 2014;75:267-269.
19.
Kawamata T, Iseki H, Shibasaki T, Hori T. Endoscopic augmented reality
navigation system for endonasal transsphenoidal surgery to treat pituitary tumors:
20.
SC
technical note. Neurosurgery. 2002;50:1393-1397.
Mirota DJ, Wang H, Taylor RH, Ishii M, Gallia GL, Hager GD. A system for
medical imaging. 2012;31:963-976. 21.
M AN U
video-based navigation for endoscopic endonasal skull base surgery. IEEE transactions on
de Notaris M, Palma K, Serra L, Ensenat J, Alobid I, Poblete J, Gonzalez JB, Solari
D, Ferrer E, Prats-Galino A. A three-dimensional computer-based perspective of the skull base. World neurosurgery. 2014;82:S41-48.
de Notaris M, Topczewski T, de Angelis M, Ensenat J, Alobid I, Gondolbleu AM,
TE D
22.
Soria G, Gonzalez JB, Ferrer E, Prats-Galino A. Anatomic skull base education using advanced neuroimaging techniques. World neurosurgery. 2013;79:S16 e19-13.
Kim YH, Kim JE, Kim MJ, Cho JH. New landmark for the endoscopic endonasal
EP
23.
transsphenoidal approach of pituitary surgery. Journal of Korean Neurosurgical Society.
24.
AC C
2013;53:218-222.
Yano S, Hide T, Shinojima N, Hasegawa Y, Kawano T, Kuratsu J. Endoscopic
endonasal skull base approach for parasellar lesions: Initial experiences, results, efficacy, and complications. Surgical neurology international. 2014;5:51.
25.
Chin OY, Ghosh R, Fang CH, Baredes S, Liu JK, Eloy JA. Internal carotid artery
injury in endoscopic endonasal surgery: A systematic review. The Laryngoscope. 2016;126:582-590.
26.
Valentine R, Wormald PJ. Carotid artery injury after endonasal surgery.
ACCEPTED MANUSCRIPT
Otolaryngologic clinics of North America. 2011;44:1059-1079. 27.
Cathelinaud O, Bizeau A, Rimbot A, Arteaga C, Verdalle P. Endoscopic endonasal
surgery complication: new methods of intracavernous internal carotid artery injury
28.
Rhoton AL, Jr., Harris FS, Renn WH. Microsurgical anatomy of the sellar region
and cavernous sinus. Clinical neurosurgery. 1977;24:54-85.
Renn WH, Rhoton AL, Jr. Microsurgical anatomy of the sellar region. Journal of
SC
29.
neurosurgery. 1975;43:288-298.
Twigg V, Carr SD, Balakumar R, Sinha S, Mirza S. Radiological features for the
M AN U
30.
RI PT
treatment. Revue de laryngologie - otologie - rhinologie. 2008;129:305-308.
approach in trans-sphenoidal pituitary surgery. Pituitary. 2017;20:395-402.
31.
Chmielik LP, Chmielik A. The prevalence of the Onodi cell - Most suitable method
of CT evaluation in its detection. International journal of pediatric otorhinolaryngology. 2017;97:202-205.
O'Brien WT, Sr., Hamelin S, Weitzel EK. The Preoperative Sinus CT: Avoiding a
TE D
32.
"CLOSE" Call with Surgical Complications. Radiology. 2016;281:10-21.
33.
Senturk M, Guler I, Azgin I, Sakarya EU, Ovet G, Alatas N, Tolu I, Erdur O. The
EP
role of Onodi cells in sphenoiditis: results of multiplanar reconstruction of computed
34.
AC C
tomography scanning. Brazilian journal of otorhinolaryngology. 2017;83:88-93.
Saeki N, Yamaura A, Numata T, Hoshi S. Bone window CT evaluation of the nasal
cavity for the transsphenoidal approach. British journal of neurosurgery. 1999;13:285-289.
35.
Hide T, Yano S, Shinojima N, Kuratsu J. Usefulness of the indocyanine green
fluorescence endoscope in endonasal transsphenoidal surgery. Journal of neurosurgery. 2015;122:1185-1192.
36.
Caversaccio M, Langlotz F, Nolte LP, Hausler R. Impact of a self-developed
planning and self-constructed navigation system on skull base surgery: 10 years experience.
Acta oto-laryngologica. 2007;127:403-407.
ACCEPTED MANUSCRIPT
37.
Fuminari K, Hideki A, Manabu O, Mitsunori M. Extended endoscopic endonasal
surgery using three-dimensional endoscopy in the intra-operative MRI suite for
38.
RI PT
supra-diaphragmatic ectopic pituitary adenoma. Turkish neurosurgery. 2015;25:503-507.
Netuka D, Masopust V, Belsan T, Profantova N, Benes V. Endoscopic endonasal
resection of medial orbital lesions with intraoperative MRI. Acta neurochirurgica. 2013;155:455-461.
Cappabianca P, Cavallo LM, Esposito F, De Divitiis O, Messina A, De Divitiis E.
SC
39.
Extended endoscopic endonasal approach to the midline skull base: the evolving role of transsphenoidal
surgery.
Advances
40.
technical
standards
M AN U
2008;33:151-199.
and
in
neurosurgery.
Gardner PA, Kassam AB, Rothfus WE, Snyderman CH, Carrau RL. Preoperative
and intraoperative imaging for endoscopic endonasal approaches to the skull base.
AC C
EP
TE D
Otolaryngologic clinics of North America. 2008;41:215-230, vii.
ACCEPTED MANUSCRIPT
Values
Cases
100
Male vs. Female
44 vs. 56
Mean age, years (range)
48.4 (4-83)
Mean tumor size, mm (range)
24.8 (5.0-60.0)
Pathology
RI PT
Characteristics
SC
Table 1 Patient characteristics
Non-functioning adenoma Prolactinoma Meningioma Craniopharyngioma
25
M AN U
GH-secreting adenoma
27 15 9 8
ACTH-secreting adenoma
5
Rathke's cleft cyst
3 1
FSH-secreting adenoma
1
Granule cells
1
Metastatic pituitary tumor
1
Pilocytic astrocytoma
1
Pituicytoma
1
Plasmacytoma
1
TSH-secreting adenoma
1
AC C
EP
TE D
Ecchordosis Physaliphora
Surgical approach Standard ETSS
64
Extended ETSS
36
ACCEPTED MANUSCRIPT
Table 2 Comparison of concordance ratios between endoscopic findings and preoperative images of oblique coronal CT and conventional coronal CT in each septum type
type
Endoscopic Oblique coronal findings
Concordance
Conventional coronal Concordance
p value
(cases)
Case
None
12
12
100.0
6
50.0
Single
31
29
93.5
26
83.9
Parallel
18
17
94.4
13
72.2
0.08871
Branched 26
25
96.2
10
38.5
0.00001*
Bridge
4
4
100.0
1
25.0
0.07143
Cross
9
9
100.0
2
22.2
0.00113*
0.00686* 0.21241
SC
(%)
M AN U
(%)
Case
RI PT
Septum
TE D
Concordance rate of endoscopic and oblique coronal or conventional coronal CT image findings, for each septum type. Statistical analysis was performed using Fisher’s exact
AC C
EP
test (* indicates significant difference).
ACCEPTED MANUSCRIPT
Table 3 Distribution comparison of landmark verification during surgery for each
RI PT
septum type
Numbers of confirmed anatomical landmarks Case
4
3
2b*
None
12
3
2
3
Single
31
7
6
7
1
Parallel
18
5
3
3
1
Branched
26
4
10
5
Bridge
4
3
1
Cross
9
Total
100
M AN U
19
2u**
3
2
27
21
1
0
2
2
6
4
4
2
4
3
SC
Septum type
2
4 20
11
Of the four landmarks (bilateral carotid prominence and bilateral OCR), the distribution of the number of landmarks that could be confirmed intraoperatively was
TE D
compared for each septum type. *2b indicates bilateral which means that 2 landmarks include one on each side, such as right carotid prominence and left OCR etc.
AC C
EP
**2u indicates two landmarks of the same side.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Highlights
Preparation for oblique coronal images of CT and MRI provides good expectancy of
In addition to the shape of septum, midline can be confirmed easily by surrounding anatomical structures on oblique coronal images.
Oblique coronal images are helpful in endonasal endoscopic transsphenoidal
EP
TE D
M AN U
SC
approach.
AC C
RI PT
sphenoid septum.
ACCEPTED MANUSCRIPT
Abbreviations: ETSS, endoscopic endonasal transsphenoidal approach; CT, computed tomography; MRI, magnetic resonance imaging; OM line, orbitomeatal line; AC-PC line,
AC C
EP
TE D
M AN U
SC
RI PT
anterior commissure–posterior commissure line; OCR, optico-carotid recess
ACCEPTED MANUSCRIPT
Disclosure: Disclosure The authors report no conflict of interest concerning the materials and methods used in this study or the findings reported in this paper.
AC C
EP
TE D
M AN U
SC
RI PT
Shigetoshi Yano MD., PhD.
S1878-8750(18)30062-7
DOI:
10.1016/j.wneu.2018.01.022
Reference:
WNEU 7220
To appear in:
World Neurosurgery
Received Date: 9 August 2017 Revised Date:
1 January 2018
Accepted Date: 4 January 2018
Please cite this article as: Yano S, Shinojima N, Kitajima M, Uetani H, Hide T, Mukasa A, Usefulness of oblique coronal CT and MRI imaging in the endoscopic endonasal approach to treat skull base lesions, World Neurosurgery (2018), doi: 10.1016/j.wneu.2018.01.022. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Usefulness Usefulness of oblique coronal CT and MRI imaging in the endoscopic endonasal
RI PT
approach to treat skull base lesions lesions
Shigetoshi Yano, MD., PhD.1, Naoki Shinojima MD., PhD.1, Mika Kitajima, MD., PhD.2,
SC
Hiroyuki Uetani MD., PhD.2, Takuichiro Hide, MD., PhD.1, Akitake Mukasa, MD.,
1Department
M AN U
PhD.1
of Neurosurgery, Faculty of Life Sciences, Kumamoto University2
Naoki Shinojima
[email protected]
[email protected]
EP
Mika Kitajima
TE D
Department of radiology, Faculty of Life Sciences, Kumamoto University
AC C
Hiroyukih Uetani
[email protected]
Takuichiro Hide
[email protected]
Akitake Mukasa
[email protected]
Corresponding author: Shigetoshi Yano Address: 1-1-1 Honjo, Chuo-ku, Kumamoto 860-8556, JAPAN
ACCEPTED MANUSCRIPT
Tel: +8196-373-5219; Fax: +8196-371-8064
RI PT
Email: [email protected]
Key Words: endoscopic endonasal transsphenoidal approach, skull base tumor,
SC
preoperative diagnosis, oblique coronal
AC C
EP
TE D
M AN U
Short title: Oblique coronal imaging for endoscopic transsphenoidal surgery
ACCEPTED MANUSCRIPT
Abstract:
RI PT
Objective:
This report examines the utility of the preoperative image to orientate the surgeon in
SC
the sphenoid sinus during an endoscopic endonasal transsphenoidal approach (ETSS).
M AN U
Methods:
ETSS was performed in 100 cases of sellar lesion and used to classify the sphenoid sinus septum shape. Preoperative CT and MRI scans were taken for two types of coronal
TE D
imaging: conventional and oblique. Expected sphenoid sinus septum shape was compared with those from ETSS to estimate concordance. The confirmation rate of
EP
anatomical landmarks in the sphenoid sinus by endoscopic observation was compared
AC C
in various types of septum and the identification rate in oblique coronal imaging was also examined.
Results: The most common septum shape was single type (31%), followed by branched (26%), parallel (18%), none (12%), cross (9%), and bridge (4%) types. In oblique coronal images,
ACCEPTED MANUSCRIPT
preoperative evaluation and endoscopic findings were consistent in 93 to 100% of cases. However, with conventional coronal images, the concordance rate was 22.2 to 83.9%,
RI PT
and in the none, branched, and cross types, the concordance rate was significantly lower than that for oblique coronal images. Although confirmation of the midline through
SC
estimation of landmarks by endoscopic observation was difficult in 33 cases,
best indicated the midline.
Conclusion:
M AN U
preoperative CT and MRI showed landmarks in all cases and oblique coronal images
TE D
Use of oblique coronal images in addition to conventional images provided good orientation of anatomical structures in the sphenoid sinus. The combination of
AC C
EP
preoperative imaging and endoscopic observation could allow safer surgery in ETSS.
247words
ACCEPTED MANUSCRIPT
Introduction： Introduction： Use of the endoscopic endonasal transsphenoidal approach has grown recently. While
RI PT
endoscope provides wide and bright vision to the surgeon, a steep learning curve is necessary to master this technique due to the tight working space and difficulty in
SC
obtaining orientation (1) (2, 3).
Even an experienced surgeon can make mistakes when approaching the sella turcica,
injury to the internal carotid artery
M AN U
particularly when an endoscope is used as the sole surgical tool. In these instances, (4) (5),
dural laceration of frontal base or posterior
fossa or unexpected CSF leakage before opening the sellar floor has been reported (6-10) .
TE D
As a cause of such disastrous complications, disorientation at the time of opening the sallar floor or removal of the tumor can be considered. Therefore, careful simulation
EP
before surgery is important.
AC C
For intraoperative orientation, various supporting systems have been developed recently. Virtual endoscopy
(11-17)
or neuronavigation
(6, 7, 18-20)
are powerful assistance
technique. However, preparation of virtual endoscopy requires time and effort, and using a navigation system may be restricted in a situation where equipment is not prepared. It would be a challenge to employ these systems for all patients in every institute.
ACCEPTED MANUSCRIPT
We have been using the endoscopic endonasal transsphenoidal approach (ETSS) since the year 2001 using above supporting systems. Although these assistance devices are
RI PT
certainly useful, we have felt the necessity of images that operator can recognize more intuitively during surgery, because the endoscopic image is a 2D image and sometimes
SC
it may be different from the surgeon's expectation. Also, several landmarks are noted as being important in current literature, but few studies have shown the evaluation
M AN U
methods against actual endoscopic images (10, 21-23).
We prepared oblique coronal CT and MRI images, obtained as perpendicular slices to the penetration direction of the endoscope into the nasal cavity. While these images are
TE D
expected to be similar to two-dimensional endoscopic images, it is unclear whether the matching rate between the preoperative image and the actual surgical image is
EP
different in the type of the septum, or which type of septum is difficult to recognize the
AC C
structure of sella floor and surrounding surgical landmarks. In this study, the usefulness of preoperative simulation using oblique coronal images for endoscopic approach to the sellar floor was then evaluated.
ACCEPTED MANUSCRIPT
Materials and Methods:
Faculty of life sciences Kumamoto University.
Clinical materials
RI PT
This research is approved by Ethics Committee on Clinical Research for Humans in
SC
From January 2011 to December 2015 (excluding cases of reoperation and subjects with unclear sphenoid sinus septation due to the presence of extended tumors) 100 patients
M AN U
undergoing ETSS at Kumamoto University were included in this study. Preoperative CT, MRI and intraoperative endoscopic views were reviewed for these 100
TE D
cases retrospectively.
Coronal CT and MRI images
EP
Conventional coronal images were obtained by slicing perpendicular to the orbitomeatal
AC C
line for CT and perpendicular to the anterior commissure–posterior commissure (AC-PC) line for MRI.
Oblique coronal CT and MRI images were made using slices perpendicular to the line connecting the nasal cavity and the dorsum sellae, a path comparable to the endoscope insertion axis. CT was taken under the condition of 2 mm thickness with 2 mm gap after injection of contrast media. MRI was carried out as post contrast spin echo T1
ACCEPTED MANUSCRIPT
images with 3 mm thickness and 0.3 mm gap.
RI PT
Surgical approach
For approaching the sellar floor, a 4-mm rigid endoscope (KARL STORZ HOPKINS II
SC
rigid endoscope, 0° or 30°, Tuttlingen, Germany) was inserted into each nostril and after removing the bony septum, was used to connect the bilateral sphenoid ostia. The
(24).
M AN U
sphenoid sinus anterior wall was opened as wide as possible, as we described previously After entering the sphenoid sinus, the bony septum, sellar floor, prominence of the
internal carotid artery, clival recess, optic canal, optico-carotid recess (OCR), and
TE D
tuberculum sellae were all located as anatomical landmarks. The bony septum was drilled carefully to entirely expose the sellar floor.
EP
For tumors extending over the sella turcica, bone from the skull base surrounding the
AC C
sellar floor was opened after preparation of the septal flap (24). The neuronavigation system (Medtronic Co., Ltd., Cororado, USA) was used for cases with tumors extending in frontal, cavernous, or clival directions or tumors encasing the internal carotid artery. Surgical images were captured and recorded.
Evaluation of the utility of each coronal slice
ACCEPTED MANUSCRIPT
The type of bony septum attached to the sellar floor in the sphenoid sinus was classified according to the observed shape on the two-dimensional CT, MRI, and endoscopic view
RI PT
images as follows: none (O shape), no obvious septum attached to the sellar floor; single (I shape), one major septum runs over the sellar floor; parallel (II shape), two or three
SC
septa run side by side on the sellar floor; branched (Y shape), septum branched and attached to the sellar floor; bridge (H shape), two parallel septa connect to one
M AN U
horizontal septum; cross (+ shape), two septa cross on the sellar floor in either a + or X shape. During the operation, each surgeon judged the septum type through endoscopic observation when the sphenoid sinus was opened and landmarks surrounding the sellar
TE D
floor could be identified.
Images of oblique coronal and conventional coronal of CT and MRI prepared before
EP
surgery were randomly observed by three surgeons after hiding the patient’s name and
AC C
the pattern of the septum passing over the sellar floor was evaluated according to the classification of intraoperative endoscopic images. Analysis was performed regardless of whether the preoperative image type matched the type observed in the operative field. Surrounding landmarks were observed and evaluated as to how many of four structures (the bilateral internal carotid prominence and bilateral OCR) could be observed during
ACCEPTED MANUSCRIPT
surgery. Number of landmarks was compared with classified septum type.
RI PT
Statistical analysis
The concordance rate of septum type classified via CT or MRI imaging with that as
SC
judged through endoscopic observation was estimated by Fisher exact test or Chi square
AC C
EP
TE D
M AN U
test. Values less than 0.05 were deemed to be significant.
ACCEPTED MANUSCRIPT
Results:
RI PT
Clinical characteristics of the 100 cases included in this study are summarized in Table 1. The differences between conventional and oblique coronal images were shown in Fig.
SC
1 as a representative case. In this case, only one septum was observed on the sellar floor in the conventional coronal image (Fig. 1b). However, three complicated septa were
M AN U
observed on the oblique coronal image (Fig. 1d); two septa attached to the sellar floor, bulging to the right side, a branch from the right septum connected to the midline septum; the final septum ran towards the left optic canal.
TE D
Six types of septum were classified through surgical findings. Corresponding oblique coronal and conventional coronal images for representative cases are shown in Fig. 2
EP
and 3.
AC C
In a case of none type, oblique coronal images identified bilateral septa attached to each carotid prominence but neither were attached to the sellar leading to classification as none type. In the same case, conventional coronal images identified one septum running to right side of the sellar floor; hence, it was classified as single type preoperatively (Fig. 2a, 2b, 2c). In a case of single type, a curved septum was observed to run from left inferior to right
ACCEPTED MANUSCRIPT
superior in the oblique coronal image, but shown as running from the midline towards the right on the conventional coronal image (Fig. 2e, 2f). Although both images resulted
RI PT
in single type classification, the precise position of sellar floor attachment was different. Operative view confirmed results from the oblique coronal image (Fig. 2d).
SC
In a case of parallel type, right OCR was not observed due to attachment of the right septum, the left septum ran along the carotid prominence lateral to the left OCR in the
M AN U
operative view (Fig. 2g). In the oblique coronal image, right OCR was observed lateral to the right septum (Fig. 2h). In the conventional coronal image, attachment of the right
recognized (Fig. 2i).
TE D
septum was the same as in the oblique coronal image but the left septum was not
Results for cases with more complicated septa are shown in Fig. 3.
EP
In a case with branched type, operative view revealed a Y-shaped septum attached to
AC C
the sellar floor (Fig. 3a). The left septum was attached to the carotid prominence, this was identical to that observed in the oblique coronal image (Fig. 3b). In the conventional coronal image, a Y-shaped septum was also observed but the septum was observed to run along the midline (Fig. 3c). In a case of bridge type, operative view revealed an H shape (Fig. 3d), bilateral septa ran along bilateral carotid prominences and connected to the sellar floor. This was
ACCEPTED MANUSCRIPT
confirmed with the oblique coronal image (Fig. 3e); however, a Y shape septum was observed on the conventional coronal image (Fig. 3f).
RI PT
In a case of cross type, both oblique coronal and conventional coronal images show a + shaped septum (Fig 3h, 3i). The cross point was located operatively and on the oblique
SC
coronal image in corresponding locations on the sellar floor (Fig 3g, 3h), while the conventional coronal image located the cross point as being beneath the sellar floor (Fig.
M AN U
3i).
In intraoperative observation of 100 cases, single type was the most common (31 cases),
TE D
following this, incidence was branched type (26 cases), parallel type (18 cases), none type (12 cases), cross type (9 cases) and bridge type (4 cases) (Table 2). When we
EP
examined the concordance rates of operative view classification with either preoperative
AC C
oblique coronal images or conventional coronal images for each septum type, high concordance rates (more than 93%) were recorded for oblique coronal images for all types of septum, while relatively low concordance rates (22 to 83%) were observed for conventional coronal images. This rate was significantly low in conventional coronal images for none, branched, and cross types. Although oblique coronal images almost always corresponded to the operative view, conventional coronal images sometimes
ACCEPTED MANUSCRIPT
highlighted an underestimation of parallel and bridge types.
RI PT
We also recorded how many surgical landmarks such as the bilateral carotid prominence and bilateral OCR (a total of four structures), were observed in the surgical
SC
field by endoscopic observation (Table 3). Cases with more than three landmarks were recorded in 5 of 12 (41.7%) none type, 13 of 31 (41.9%) single type, 8 of 18 (44.4 %)
M AN U
parallel type, 14 of 26 (53.8%) branched type, 3 of 4 (75.0%) bridge type, and 3 of 9 (33.3%) cross type. Landmarks were most easily confirmed in bridge type and least
TE D
often confirmed in cross type, although this difference was not statistically significant.
Through surgical findings, confirmation of the midline was straightforward in 46 cases
EP
since these cases identified more than 3 landmarks. However, in 31 cases one or no
AC C
landmarks were observed, and there were 2 cases where landmarks appeared unilaterally. In oblique coronal images, attachment position of the septum and relationship to the landmarks were clearly observed preoperatively in all cases and showed high concordance rates with operative findings, regardless of septum type.
ACCEPTED MANUSCRIPT
Case presentation:
RI PT
Case 1 (Fig. 4)
A 68-year-old man with non-functioning pituitary adenoma. Although conventional
SC
coronal MRI imaging showed a pituitary tumor extending to the suprasellar region with two parallel septa attached to the sellar floor on the left side of the sphenoid sinus (Fig.
M AN U
4a), oblique coronal MRI and CT images indicated that the sellar floor was not enlarged and branched type septum attached to the left side of the sellar floor (Fig 4b, 4c), which was similar to the intraoperative endoscopic view (Fig. 4d). Intraoperatively, while left
TE D
OCR was identified, the right OCR was covered by septum or bone and hence, could not be identified. The midline was not easily decided through endoscopic view, but the
EP
oblique coronal image indicated the midline to be on the sellar floor or tuberculum sellar
AC C
via an existing septum. Marking was performed after confirmation by neuronavigation which verified the septum attachment position predicted by oblique coronal images (Fig. 4d).
Case 2 (Fig. 5) A 20-year-old man with acromegaly. Preoperative oblique coronal CT and MRI showed a
ACCEPTED MANUSCRIPT
tumor that had enlarged the sellar floor and extended into the right cavernous sinus. Cross type septum that run parallel beneath the sellar floor and left side septum made
RI PT
border of the sellar floor was expected (Fig. 5a, 5b), which was identical to the intraoperative view (Fig. 5c). The midline was easily identified by the position of septum
SC
and bilateral optic canals by reffurring to preoperative oblique images, although it would have been difficult to confirm the midline by endoscopic view only. Sellar floor
AC C
EP
TE D
neuronavigation system.
M AN U
was opened safely (Fig. 5d) and the tumor removed successfully without the aid of the
ACCEPTED MANUSCRIPT
Discussion： Discussion： In the endoscopic endonasal approach, the most important factor is the direction of
RI PT
endoscope entry. Since the endoscopic image has a wide angle, a slight directional change in the entry of the nasal cavity results in a large deviation of arrival point in the
SC
sphenoid or ethmoid sinuses. Further to this, the endoscope can rotate in the surgeon’s hand, so constant checking of the vertical plane and midline is critical. Indeed, there are
M AN U
some reports of disorientation-related internal carotid artery injury (5, 25-27). Confirmation of bilaterally symmetrical structures such as the carotid prominence or optic prominence is important to measure the midline. The positions of tuberculum
TE D
sellae and clivus recess can be used as guides to locate the sella turcica, and this is especially helpful in flat floor types. However, in the sphenoid sinus, some patterns of
EP
bony septum may disrupt the observation of these landmarks.
AC C
Variation of the sphenoid septation has been well known for some time
(28, 29)
and it is
clear that the most commonly presenting septum does not necessarily lie along the midline. Recently, Twigg et al.
(30)
assessed 100 cases of CT and MRI, highlighting the
significant anatomical variations in patients with pituitary disease. They stressed the need to be aware of unusual septation across the internal carotid artery at sella level and the possibility of deviation of the internal carotid artery itself. Also prevalence of
ACCEPTED MANUSCRIPT
the Onodi cell which is identified as the most posterior ethmoid cell and extend to the optic nerve or internal carotid artery was reported as 39.8% of 196 paranasal sinus (31).
It is also claimed that the risk of optic nerve injury increases
RI PT
examination by CT
when the identification of the Onodi cell during the operation.
(32) (8) (33)
We have also
SC
observed the septum running over the carotid prominence and optic prominence in many cases; hence, in this study, we examined the relationship between septation
M AN U
variation and anatomical landmarks surrounding the sella turcica.
In order to recognize the positional relationship between front, rear, left, and right
TE D
during surgery, we chose the four landmarks of bilateral carotid prominences and bilateral OCR. Under endoscopic observation, only 19% of cases had all four landmarks
EP
confirmed, and 33% of cases had less than one observable landmark. This indicates the
AC C
difficulty of midline confirmation solely through endoscopic observation. Although no significant difference was observed in the number of landmarks observed for each type of sphenoid sinus septum, there was no case in which all four landmarks could be confirmed in cross type of septum and this type showed the lowest landmark confirmation rate. These results indicate the importance of recognizing the septum attachment point to skull base structures by CT and MRI before surgery.
ACCEPTED MANUSCRIPT
Although CT clearly depicts septum, MRI can not visualize bone structures. However, it is possible to recognize the shape of the septum from the contrast of the mucous
RI PT
membrane adhering to the sinus septum wall. Oblique MRI shows the shape and extension of the tumor well in the direction of the endoscope and well depicts the
SC
relation with the internal carotid arteries and the cavernous sinus. Although contrast CT is sufficient to evaluate the shape of septum, attention is paid to the usefulness of
M AN U
simultaneously recognizing the detailed relationship between the tumor to be removed and the surrounding structure, and MRI is also used in the same imaging method.
simulation.
TE D
Combined use of both CT and MRI should improve the quality of the preoperative
While the slice direction of conventional coronal CT and MRI images may vary between
EP
different institutes, our institute adopts a method of slicing perpendicularly to OM or
AC C
AC-PC lines. Indeed, the conventional method is particularly useful for the qualitative diagnosis of pituitary gland and pituitary tumor, however, concordance rate of septum type identification with operative endoscopic findings was low. In conventional coronal images, branched type and cross type septa were significantly more commonly predicted prior to surgery and false positive septa were predicted in none type. Moreover, position of septum attachment was sometimes misread as the midline. Oblique coronal images
ACCEPTED MANUSCRIPT
gave predictions consistent with endoscopic images in more than 94% of cases. It was most reliable to predict the midline location prior to surgery using oblique
RI PT
coronal images to estimate the relationship between septum attachment location and surrounding landmarks.
SC
Even in a case where the landmark was unclear in the intraoperative endoscopic image, the oblique coronal images provided clarity and so clear identification of the structure
M AN U
behind the septum attachment. In previous reports, bone window CT scans sliced parallel to the transsphenoidal surgical route were introduced
(34),
and Twigg et al.
(30)
classified the 5 types of septation according to attachment position relative to the
TE D
internal carotid artery. Certainly, these techniques are useful preoperative prediction methods, but our oblique coronal scan has provided preoperative simulation that more
AC C
EP
closely matches intraoperative endoscopic image findings.
To improve the orientation of the surgeon during endoscopic endonasal surgery, various devices and methods have been recommended including, virtual endoscopy
(11-17),
neuronavigation (6, 7, 18-20), indocyanine green endoscopy(35), endonasal ultrasound (6, 7, 36), and intraoperative MRI (37-40). Virtual endoscopy resembles our oblique coronal imaging concept in terms of mimicking
ACCEPTED MANUSCRIPT
the endoscopic view, however, preparation of virtual images takes time and costs more than oblique coronal imaging. Neuronavigation can also be useful in confirming the
RI PT
midline and surrounding structures, however the operator needs to check the orientation at an appropriate timing unless navigation is a persistent monitor. Prior to
SC
the surgery, it is important for the surgeon to confirm the variation in structure and position of the midline. Thus, expanding the surgeon’s resources from conventional
M AN U
coronal images to include oblique coronal images, should make endoscopic endonasal
AC C
EP
TE D
skull base approach surgery safer for all.
ACCEPTED MANUSCRIPT
Figure Legends
Comparison of conventional coronal CT and oblique coronal CT images
RI PT
Fig. 1
a) Scout view of conventional coronal CT. Images were obtained perpendicular to the
SC
orbitomeatal line. b) Representative bone image of conventional CT indicating single bony septum in sphenoid sinus. c) Scout view of oblique coronal CT of the same patient.
M AN U
Images were obtained perpendicular to the line connecting the nasal cavity and the dorsum sellae. d) Representative bone image CT corresponding to panel b. Complex
Representative images of three septum types identified through operative
EP
Fig. 2
TE D
branched type septum was observed.
AC C
endoscopic images, preoperative oblique coronal CT images and preoperative conventional CT images. Septum type was classified by endoscopic findings as none type (a), single type (d), and parallel type (g). Corresponding preoperative oblique coronal images (b, e, h) and conventional coronal images (c, f, i) are shown for comparison.
ACCEPTED MANUSCRIPT
Fig. 3
Representative images of remaining three septum types identified through
RI PT
operative endoscopic images, preoperative oblique coronal CT images and preoperative conventional CT images.
SC
Remaining septum types were classified by endoscopic findings as branched type (a), bridge type (d) and cross type (g). Corresponding preoperative oblique coronal images (b,
Preoperative CT and MRI images and intraoperative image of Case 1
TE D
Fig. 4
M AN U
e, h) and conventional coronal images (c, f, i) are shown for comparison.
Preoperative Gd-enhanced MRI of conventional coronal image (a) and corresponding
EP
oblique coronal image (b). Corresponding image of oblique coronal CT (c) indicated
AC C
branched type septum in sphenoid sinus. Endoscopic image of sellar floor after opening the sphenoid sinus (d). Branched septum runs from the left side to the sellar floor midline. The OCR and carotid prominence were observed on the left side, but landmarks of the right side were not clearly observed. Midline was marked with blue dye.
ACCEPTED MANUSCRIPT
Fig. 5
Representative images of Case 2
RI PT
Preoperative oblique coronal CT (a) and MRI (b) showed that the tumor enlarged the sellar floor and extended to the right cavernous sinus with a cross type septum running
SC
along the sellar floor. Intraoperative endoscopic view confirmed this (c). After opening the sellar floor, the dura was exposed and the midline confirmed by the sphenoidal
AC C
EP
TE D
M AN U
septum (d).
ACCEPTED MANUSCRIPT
References
Leach P, Abou-Zeid AH, Kearney T, Davis J, Trainer PJ, Gnanalingham KK.
RI PT
1.
Endoscopic transsphenoidal pituitary surgery: evidence of an operative learning curve.
Neurosurgery. 2010;67:1205-1212. 2.
Shikary T, Andaluz N, Meinzen-Derr J, Edwards C, Theodosopoulos P, Zimmer LA.
SC
Operative Learning Curve After Transition to Endoscopic Transsphenoidal Pituitary Surgery. World neurosurgery. 2017;102:608-612.
Chi F, Wang Y, Lin Y, Ge J, Qiu Y, Guo L. A learning curve of endoscopic
M AN U
3.
transsphenoidal surgery for pituitary adenoma. The Journal of craniofacial surgery. 2013;24:2064-2067.
4.
Carr S, Twigg V, Mirza S. Radiological study of the anatomy of the keystone area of
the nasal septum using computed tomography to aid septal surgery. Clinical otolaryngology :
official
journal
of
ENT-UK
;
official
journal
of
Netherlands
Society
for
5.
TE D
Oto-Rhino-Laryngology & Cervico-Facial Surgery. 2016;41:317-320. Gardner PA, Tormenti MJ, Pant H, Fernandez-Miranda JC, Snyderman CH,
Horowitz MB. Carotid artery injury during endoscopic endonasal skull base surgery:
Yano S, Kawano T, Kudo M, Makino K, Nakamura H, Kai Y, Morioka M, Kuratsu J.
AC C
6.
EP
incidence and outcomes. Neurosurgery. 2013;73:ons261-269; discussion ons269-270.
Endoscopic endonasal transsphenoidal approach through the bilateral nostrils for pituitary adenomas. Neurologia medico-chirurgica. 2009;49:1-7.
7.
Saeki N, Horiguchi K, Murai H, Hasegawa Y, Hanazawa T, Okamoto Y. Endoscopic
endonasal pituitary and skull base surgery. Neurologia medico-chirurgica. 2010;50:756-764.
8.
Meybodi AT, Vigo V, Benet A. The Onodi Cell: An Anatomic Illustration. World
neurosurgery. 2017;103:950 e955-950 e956. 9.
Shin JH, Kim SW, Hong YK, Jeun SS, Kang SG, Cho JH, Park YJ. The Onodi cell:
ACCEPTED MANUSCRIPT
an obstacle to sellar lesions with a transsphenoidal approach. Otolaryngology--head and
neck surgery : official journal of American Academy of Otolaryngology-Head and Neck Surgery. 2011;145:1040-1042. Kassam AB, Prevedello DM, Carrau RL, Snyderman CH, Thomas A, Gardner P,
RI PT
10.
Zanation A, Duz B, Stefko ST, Byers K, Horowitz MB. Endoscopic endonasal skull base surgery: analysis of complications in the authors' initial 800 patients. Journal of
neurosurgery. 2011;114:1544-1568.
Neubauer A, Wolfsberger S. Virtual endoscopy in neurosurgery: a review.
SC
11.
Neurosurgery. 2013;72 Suppl 1:97-106.
Rosseau G, Bailes J, del Maestro R, Cabral A, Choudhury N, Comas O, Debergue P,
M AN U
12.
De Luca G, Hovdebo J, Jiang D, Laroche D, Neubauer A, Pazos V, Thibault F, Diraddo R. The development of a virtual simulator for training neurosurgeons to perform and perfect endoscopic endonasal transsphenoidal surgery. Neurosurgery. 2013;73 Suppl 1:85-93.
13.
de Notaris M, Solari D, Cavallo LM, Ensenat J, Alobid I, Soria G, Gonzalez JB,
Ferrer E, Prats-Galino A. The use of a three-dimensional novel computer-based model for
TE D
analysis of the endonasal endoscopic approach to the midline skull base. World
neurosurgery. 2011;75:106-113; discussion 136-140. 14.
Zymberg S, Vaz-Guimaraes Filho F, Lyra M. Neuroendoscopic training:
15.
AC C
2010;53:44-46.
EP
presentation of a new real simulator. Minimally invasive neurosurgery : MIN.
Virtual
Tolsdorff B, Pommert A, Hohne KH, Petersik A, Pflesser B, Tiede U, Leuwer R. reality:
a
new
paranasal
sinus
surgery
simulator.
The
Laryngoscope.
2010;120:420-426.
16.
Schulze F, Buhler K, Neubauer A, Kanitsar A, Holton L, Wolfsberger S.
Intra-operative virtual endoscopy for image guided endonasal transsphenoidal pituitary surgery. International journal of computer assisted radiology and surgery. 2010;5:143-154.
17.
Neubauer A, Wolfsberger S, Forster MT, Mroz L, Wegenkittl R, Buhler K.
Advanced virtual endoscopic pituitary surgery. IEEE transactions on visualization and
ACCEPTED MANUSCRIPT
computer graphics. 2005;11:497-507. 18.
Duque SG, Gorrepati R, Kesavabhotla K, Huang C, Boockvar JA. Endoscopic
endonasal transphenoidal surgery using the BrainLAB(R) Headband for navigation without
RI PT
rigid fixation. Journal of neurological surgery Part A, Central European neurosurgery. 2014;75:267-269.
19.
Kawamata T, Iseki H, Shibasaki T, Hori T. Endoscopic augmented reality
navigation system for endonasal transsphenoidal surgery to treat pituitary tumors:
20.
SC
technical note. Neurosurgery. 2002;50:1393-1397.
Mirota DJ, Wang H, Taylor RH, Ishii M, Gallia GL, Hager GD. A system for
medical imaging. 2012;31:963-976. 21.
M AN U
video-based navigation for endoscopic endonasal skull base surgery. IEEE transactions on
de Notaris M, Palma K, Serra L, Ensenat J, Alobid I, Poblete J, Gonzalez JB, Solari
D, Ferrer E, Prats-Galino A. A three-dimensional computer-based perspective of the skull base. World neurosurgery. 2014;82:S41-48.
de Notaris M, Topczewski T, de Angelis M, Ensenat J, Alobid I, Gondolbleu AM,
TE D
22.
Soria G, Gonzalez JB, Ferrer E, Prats-Galino A. Anatomic skull base education using advanced neuroimaging techniques. World neurosurgery. 2013;79:S16 e19-13.
Kim YH, Kim JE, Kim MJ, Cho JH. New landmark for the endoscopic endonasal
EP
23.
transsphenoidal approach of pituitary surgery. Journal of Korean Neurosurgical Society.
24.
AC C
2013;53:218-222.
Yano S, Hide T, Shinojima N, Hasegawa Y, Kawano T, Kuratsu J. Endoscopic
endonasal skull base approach for parasellar lesions: Initial experiences, results, efficacy, and complications. Surgical neurology international. 2014;5:51.
25.
Chin OY, Ghosh R, Fang CH, Baredes S, Liu JK, Eloy JA. Internal carotid artery
injury in endoscopic endonasal surgery: A systematic review. The Laryngoscope. 2016;126:582-590.
26.
Valentine R, Wormald PJ. Carotid artery injury after endonasal surgery.
ACCEPTED MANUSCRIPT
Otolaryngologic clinics of North America. 2011;44:1059-1079. 27.
Cathelinaud O, Bizeau A, Rimbot A, Arteaga C, Verdalle P. Endoscopic endonasal
surgery complication: new methods of intracavernous internal carotid artery injury
28.
Rhoton AL, Jr., Harris FS, Renn WH. Microsurgical anatomy of the sellar region
and cavernous sinus. Clinical neurosurgery. 1977;24:54-85.
Renn WH, Rhoton AL, Jr. Microsurgical anatomy of the sellar region. Journal of
SC
29.
neurosurgery. 1975;43:288-298.
Twigg V, Carr SD, Balakumar R, Sinha S, Mirza S. Radiological features for the
M AN U
30.
RI PT
treatment. Revue de laryngologie - otologie - rhinologie. 2008;129:305-308.
approach in trans-sphenoidal pituitary surgery. Pituitary. 2017;20:395-402.
31.
Chmielik LP, Chmielik A. The prevalence of the Onodi cell - Most suitable method
of CT evaluation in its detection. International journal of pediatric otorhinolaryngology. 2017;97:202-205.
O'Brien WT, Sr., Hamelin S, Weitzel EK. The Preoperative Sinus CT: Avoiding a
TE D
32.
"CLOSE" Call with Surgical Complications. Radiology. 2016;281:10-21.
33.
Senturk M, Guler I, Azgin I, Sakarya EU, Ovet G, Alatas N, Tolu I, Erdur O. The
EP
role of Onodi cells in sphenoiditis: results of multiplanar reconstruction of computed
34.
AC C
tomography scanning. Brazilian journal of otorhinolaryngology. 2017;83:88-93.
Saeki N, Yamaura A, Numata T, Hoshi S. Bone window CT evaluation of the nasal
cavity for the transsphenoidal approach. British journal of neurosurgery. 1999;13:285-289.
35.
Hide T, Yano S, Shinojima N, Kuratsu J. Usefulness of the indocyanine green
fluorescence endoscope in endonasal transsphenoidal surgery. Journal of neurosurgery. 2015;122:1185-1192.
36.
Caversaccio M, Langlotz F, Nolte LP, Hausler R. Impact of a self-developed
planning and self-constructed navigation system on skull base surgery: 10 years experience.
Acta oto-laryngologica. 2007;127:403-407.
ACCEPTED MANUSCRIPT
37.
Fuminari K, Hideki A, Manabu O, Mitsunori M. Extended endoscopic endonasal
surgery using three-dimensional endoscopy in the intra-operative MRI suite for
38.
RI PT
supra-diaphragmatic ectopic pituitary adenoma. Turkish neurosurgery. 2015;25:503-507.
Netuka D, Masopust V, Belsan T, Profantova N, Benes V. Endoscopic endonasal
resection of medial orbital lesions with intraoperative MRI. Acta neurochirurgica. 2013;155:455-461.
Cappabianca P, Cavallo LM, Esposito F, De Divitiis O, Messina A, De Divitiis E.
SC
39.
Extended endoscopic endonasal approach to the midline skull base: the evolving role of transsphenoidal
surgery.
Advances
40.
technical
standards
M AN U
2008;33:151-199.
and
in
neurosurgery.
Gardner PA, Kassam AB, Rothfus WE, Snyderman CH, Carrau RL. Preoperative
and intraoperative imaging for endoscopic endonasal approaches to the skull base.
AC C
EP
TE D
Otolaryngologic clinics of North America. 2008;41:215-230, vii.
ACCEPTED MANUSCRIPT
Values
Cases
100
Male vs. Female
44 vs. 56
Mean age, years (range)
48.4 (4-83)
Mean tumor size, mm (range)
24.8 (5.0-60.0)
Pathology
RI PT
Characteristics
SC
Table 1 Patient characteristics
Non-functioning adenoma Prolactinoma Meningioma Craniopharyngioma
25
M AN U
GH-secreting adenoma
27 15 9 8
ACTH-secreting adenoma
5
Rathke's cleft cyst
3 1
FSH-secreting adenoma
1
Granule cells
1
Metastatic pituitary tumor
1
Pilocytic astrocytoma
1
Pituicytoma
1
Plasmacytoma
1
TSH-secreting adenoma
1
AC C
EP
TE D
Ecchordosis Physaliphora
Surgical approach Standard ETSS
64
Extended ETSS
36
ACCEPTED MANUSCRIPT
Table 2 Comparison of concordance ratios between endoscopic findings and preoperative images of oblique coronal CT and conventional coronal CT in each septum type
type
Endoscopic Oblique coronal findings
Concordance
Conventional coronal Concordance
p value
(cases)
Case
None
12
12
100.0
6
50.0
Single
31
29
93.5
26
83.9
Parallel
18
17
94.4
13
72.2
0.08871
Branched 26
25
96.2
10
38.5
0.00001*
Bridge
4
4
100.0
1
25.0
0.07143
Cross
9
9
100.0
2
22.2
0.00113*
0.00686* 0.21241
SC
(%)
M AN U
(%)
Case
RI PT
Septum
TE D
Concordance rate of endoscopic and oblique coronal or conventional coronal CT image findings, for each septum type. Statistical analysis was performed using Fisher’s exact
AC C
EP
test (* indicates significant difference).
ACCEPTED MANUSCRIPT
Table 3 Distribution comparison of landmark verification during surgery for each
RI PT
septum type
Numbers of confirmed anatomical landmarks Case
4
3
2b*
None
12
3
2
3
Single
31
7
6
7
1
Parallel
18
5
3
3
1
Branched
26
4
10
5
Bridge
4
3
1
Cross
9
Total
100
M AN U
19
2u**
3
2
27
21
1
0
2
2
6
4
4
2
4
3
SC
Septum type
2
4 20
11
Of the four landmarks (bilateral carotid prominence and bilateral OCR), the distribution of the number of landmarks that could be confirmed intraoperatively was
TE D
compared for each septum type. *2b indicates bilateral which means that 2 landmarks include one on each side, such as right carotid prominence and left OCR etc.
AC C
EP
**2u indicates two landmarks of the same side.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Highlights
Preparation for oblique coronal images of CT and MRI provides good expectancy of
In addition to the shape of septum, midline can be confirmed easily by surrounding anatomical structures on oblique coronal images.
Oblique coronal images are helpful in endonasal endoscopic transsphenoidal
EP
TE D
M AN U
SC
approach.
AC C
RI PT
sphenoid septum.
ACCEPTED MANUSCRIPT
Abbreviations: ETSS, endoscopic endonasal transsphenoidal approach; CT, computed tomography; MRI, magnetic resonance imaging; OM line, orbitomeatal line; AC-PC line,
AC C
EP
TE D
M AN U
SC
RI PT
anterior commissure–posterior commissure line; OCR, optico-carotid recess
ACCEPTED MANUSCRIPT
Disclosure: Disclosure The authors report no conflict of interest concerning the materials and methods used in this study or the findings reported in this paper.
AC C
EP
TE D
M AN U
SC
RI PT
Shigetoshi Yano MD., PhD.