ORIGINAL COMMUNICATION

Endoscopic Anatomy of Sphenoid Sinus for

Pituitary Surgery
A. UNLU,
1
*
C. MECO,
2,3
H.C. UGUR,
1
A. COMERT,
4
M. OZDEMIR,
1
AND A. ELHAN
4
1
Department of Neurosurgery, Faculty of Medicine, Ankara University, Ankara, Turkey
2
Department of Otorhinolaryngology-Head and Neck Surgery, Faculty of Medicine,
Ankara University, Ankara, Turkey
3
Department of Otorhinolaryngology-Head and Neck Surgery, Salzburg Paracelsus University
Medical School, Salzburg, Austria
4
Department of Anatomy, Faculty of Medicine, Ankara University, Ankara, Turkey
Endoscopic endonasal transsphenoidal approach to the sellar region yields an
alternative to classical microsurgical approaches. Endoscopes provide images
that differ from microscopic view. This study aimed to highlight surgical land-
marks and their anatomical relationships for pituitary surgery through endo-
scopic perspective. Ten sides of ve adult cadaveric heads with red-colored la-
tex injected arteries were evaluated. Endoscopic dissections were performed
and measurements were done in the sphenoid sinuses before and after the re-
moval of bony structures in all the aspects. Endoscopic vision of the sellar
region enabled a wide panoramic perspective and detailed inspection. The
measurements, in general, indicated the variations in the bony structures and
soft tissues. The width of the pituitary, which is the distance between the
medial margins of the carotid prominences, was measured as 21 6 2.5 mm
and the distance between the medial margin of the carotid prominences at the
lower margin of the pituitary was 18 6 3.1 mm. After the bony structures
were removed, further measurements were done. The width of the pituitary,
which is the distance between the medial margins of the anterior curvature of
the ICA, was measured as 23.2 6 3 mm, while the distance between the pos-
terior curvature of the ICA was 19.7 6 4.9 mm. Endoscopic view provided
superior detailed visualization of the close relationships between pituitary
gland, internal carotid arteries, and optic nerves. This facilitated exact evalua-
tion for variations, which could result in more effective and safe surgery. How-
ever, these variations again emphasize the necessity of preoperative radiologi-
cal evaluation in each case. Clin. Anat. 21:627632, 2008. V VC
2008 Wiley-Liss, Inc.
Key words: sphenoid sinus; pituitary adenoma; anatomy; endoscopy; endo-
scopic endonasal surgery
INTRODUCTION
Since the beginning of 19th century, transsphe-
noidal approach has become the most commonly
used procedure in surgical treatment of pituitary
tumors. Transseptal, sublabial, and direct transnasal
approaches have been applied to directly see the
sellar region through a tunnel vision via a nasal
speculum (Hardy, 1969).
*Correspondence to: Agahan Unlu, Department of Neurosurgery,
Faculty of Medicine, Ankara University, 06100, Sihhiye, Ankara,
Turkey. E-mail: [email protected]
Received 11 January 2008; Revised 17 May 2008; Accepted 2
August 2008
Published online 23 September 2008 in Wiley InterScience
(www.interscience.wiley.com). DOI 10.1002/ca.20707
V VC
2008 Wiley-Liss, Inc.
Clinical Anatomy 21:627632 (2008)
Fig. 1
Fig. 2
Fig. 3. A: Endoscopical view of the frontal basal area. on, optic nerve; ch,
chiasm; , frontal lobe; p, pituitary gland; ca, carotid artery. B: Schematic view of A.
628 Unlu et al.
In the early eighties, endoscopes were introduced
for nasal and paranasal sinus surgery, which in a
very short time revolutionized the management of
sinonasal pathologies and became a powerful tool in
the hands of otorhinolaryngologists (Stammberger,
1986). Endoscopes, in conjunction with the presen-
tation of computed tomography (CT), allowed a
much better understanding of the paranasal sinus
anatomy and its relationships to vital structures. Fur-
thermore, advances in instrumentation and con-
dence gained in endoscopic techniques encouraged
surgeons of different disciplines to merge their
evolving skills to address various other pathologies
including pituitary tumors (Kassam et al., 2005; Lau-
fer et al., 2007).
First reported around the mid nineties by otorhi-
nolaryngologists, Jankowski et al. (1992) and then
by Carrau et al. (1996), endoscopic endonasal trans-
sphenoidal approach to the sellar region tumors has
continuously evolved in the last decade as a surgical
modality, and thus become a good alternative to the
widely used classical microsurgical approaches.
Endoscopes provide a widespread and detailed view
of anatomical structures, which differ from the mi-
croscopic view (Batay et al., 2002; Catapano et al.,
2006). Already being used by several centers around
the world for its distinct advantages, this approach
still signies an entirely new eld for most neurosur-
geons. In the learning curve of this endoscopic oper-
ative procedure, sphenoid sinus anatomy plays the
utmost important role as its bony walls cover several
important structures like internal carotid arteries
(ICA), optic nerves, and skull base, which should be
respected during surgery.
The sphenoid sinuses are cage-like aerated spaces
located at a central position along the midline skull
base. Because of different types of aeration, they
show large variations in the size and shape affecting
the three-dimensional anatomical features of their
surrounding walls. Behind the bony layer of the walls,
which could also be occasionally dehiscent, the ICAs
are situated at the mid-lateral walls and the optic
nerves at the border of superior and lateral walls.
The projection of the optic nerves at the superolateral
corner could sometimes be hidden by well pneuma-
tized posterosphenoid cells challenging identication.
The spatial orientation to the landmarks in sphenoid
sinuses at any moment of endoscopic dissections
during surgery is crucial, in order to avoid injuries
which could result in devastating complications like
arterial hemorrhages or visual loss. Emphasizing the
signicance of anatomical knowledge merged with
endoscopic spatial orientation, this study aimed to
highlight surgical landmarks of the sphenoid sinus
and their anatomical relationships for pituitary sur-
gery through an endoscopic view.
MATERIALS AND METHODS
Ten sides of ve adult cadaveric heads were eval-
uated. All the cadavers were adult males aged
between 42 and 68. The arteries of the heads were
injected with red-colored latex through common ca-
rotid arteries. Endoscopic dissections were per-
formed with endoscopic sinus and skull base surgery
instruments and with the guidance of 18-cm long
rigid endoscopes with 4-mm diameter and 08 and 308
angled lenses that are used for standard endoscopic
sinus surgery (Karl Storz Co., Germany). A xenon
light source and a single CCD digital video camera
were connected to the endoscopes (Karl Storz Co.,
Germany). The images acquired were displayed on
an LCD monitor to enable dissection and simultane-
ously recorded digitally on a DVD medium. For bony
dissections, a high-speed drill system was used
(Medtronic Co., USA). The measurements were done
on a millimetric scale.
In each cadaver, dissection started with the evalu-
ation of both endonasal cavities. Bilaterally, the
sphenoid sinus ostiums were located at the sphe-
noethmoid recess, medial to the superior turbinates.
The sphenoid sinus ostiums were enlarged to facili-
tate the complete removal of the anterior wall of the
sphenoid sinuses up to the level of planum sphenoi-
dale, along with the lower portion of superior turbi-
Fig. 2. A, B: Endoscopical view of the sphenoid
sinus after removal of the upper and lateral bony struc-
tures. g, the distance between the optic nerves at the
level of opticocarotid recesses; h, the height of the pitu-
itary; i, the width of the pituitary, which is the distance
between the medial margins of the anterior curvature of
the internal carotid artery at the longest axis; ts, dura
of tuberculum sellae; d, diaphragma sella; p, pituitary;
ca-ac, carotid artery anterior curvature; ca-pc, carotid
artery posterior curvature; clv, clivus; cs, cavernous
sinus; *ocr, opticocarotid recess, j, the distance
between the medial margins of the carotid artery at the
inferior margin of the pituitary; k, the distance between
the medial margins of the posterior curvature of the ca-
rotid artery at the shortest axis; l, the distance between
the lateral margins of the anterior curvature of the
carotid artery at the longest axis; m, the distance
between the lateral margins of the posterior curvature
of the carotid artery at the longest axis. C: Schematic
view of A and B.
Fig. 1. A, B: The sphenoid sinus, endoscopic view.
Pituitary prominence, carotid prominences were seen. a,
the distance between the medial margins of the optic
nerves at the level of the opticocarotid recesses; *(ocr)
b, the height of the sellar wall; c, the width of the ante-
rior sellar wall, which is the distance between the medial
margins of carotid prominences at the longest axis; d,
the distance between the medial margins of the carotid
prominences at the lower margin of the anterior sellar
wall; e, the distance between the lateral margins of the
carotid prominences at the longest axis; f, the distance
between the rostrum sphenoidale (the anterior wall of
the sphenoid sinus) and the anterior sellar wall, on, optic
nerve; p, pituitary; clv, clivus; ssw, sphenoid sinus wall;
cap, carotid artery prominences; *ocr, opticocarotid
recess. C: Schematic view of A and B.
629 Endoscopic Anatomy for Pituitary Surgery
nates, as well as the posterior part of the septum
and rostrum sphenoidale to expose the widest surgi-
cal eld under anterior endoscopic view. Thus, bilat-
eral 3 to 4 handed manipulations were facilitated.
The pattern of the intersinus septa and other addi-
tional accessory septae were recorded according to
the posterior termination structure. Then, the inter-
sinus and accessory septae were completely
removed, and the anterior wall of the sella was iden-
tied. Sphenoid sinus pneumatization and the type
of sphenoid sinus (conchal, presellar, and sellar
types) were determined. On both sides, along the
lateral wall of the sphenoid sinus, optic nerves and
ICA (anterior curvature at the superior portion and
posterior curvature at inferior portion) were identi-
ed along with the opticocarotid recesses. Presence
of Onodi cells was also recorded, as these most pos-
terior ethmoid cells pneumatize superolaterally to-
ward sphenoid sinus, pushing the superolateral as-
pect of the anterior wall of the sphenoid sinus back-
wards often leaving the path of optic nerves in those
cells. As a result, Onodi cells cause postero-infero-
medial oblique displacement of the sphenoid sinus
anterior wall narrowing the view of lateral wall in the
sphenoid sinus, which could interfere with landmark
detection.
After dening the bony landmarks of the sphenoid
sinus, the following were measured in the rst step:
a, the distance between medial margins of optic
nerves at the level of the opticocarotid recess; b, the
height of the sellar wall; c, the width of the anterior
sellar wall, which is the distance between medial
margins of carotid prominences at the longest axis;
d, the distance between medial margins of the ca-
rotid prominences at the lower margin of the anterior
sellar wall; e, the distance between the lateral mar-
gins of the carotid prominences at the longest axis;
f, the distance between the rostrum sphenoidale (the
anterior wall of the sphenoid sinus) and the anterior
sellar wall (Figs. 1A and 1B). Then, bony walls of the
sphenoid sinus were drilled and removed in all
aspects except the inferior wall of the sphenoid sinus
and clivus, leaving only the soft tissues on the lateral
walls and the roof. In this second step, following
measurements were done: g, the distance between
the optic nerves at the level of the opticocarotid
recess; h, the height of the pituitary gland; i, the
width of the pituitary, which is the distance between
the medial margins of the anterior curvature of the
internal carotid artery at the longest axis; j, the dis-
tance between the medial margins of the carotid ar-
tery at the inferior margin of the pituitary; k, the dis-
tance between the medial margins of the posterior
curvature of the carotid artery at the shortest axis; l,
the distance between the lateral margins of the ante-
rior curvature of the carotid artery at the longest
axis; m, the distance between the lateral margins of
the posterior curvature of the carotid artery at the
longest axis (Figs. 2A and 2B).
In the third step, the dura mater of the anterior
fossa along the tuberculum sellae and pituitary were
removed, and the distance between the optic chiasm
and the superior margin of the pituitary was meas-
ured (Fig. 3).
RESULTS
Endoscopic dissection enabled fast, uncompli-
cated, and direct access to the sphenoid sinus. The
panoramic large view on the monitor facilitated
straightforward recognition of the bony anatomical
landmarks. On completely removing the anterior wall
of the sphenoid sinuses up to the level of the planum
sphenoidale, it was observed that the posterior ter-
mination of the intersinus septa and other additional
accessory septae ended on the ICA prominence in
four cadavers (two on the left and two on the right
sides). One was terminating precisely on the midline.
According to sphenoid sinus pneumatization, two
cadavers had presellar type and three cadavers had
sellar type sphenoid sinuses. In all the cadavers, the
ON and ICA prominences along with the opticocaro-
tid recess (OCR) and the pituitary bulge in the mid-
line were identied on both sides. In two cadavers,
bilateral Onodi cells were encountered.
After identifying the bony landmarks, the distance
between medial margins of the ON at the level of the
OCR was measured as 12.1 6 2.4 mm (range: 9.3
16.3 mm). The width of the pituitary, which is the
distance between the medial margins of the carotid
prominences at its longest axis, was measured as 21
6 2.5 mm (range:17.724.4 mm) and the distance
between the medial margin of the carotid prominen-
ces at the lower margin of the pituitary was 18 6 3.1
mm (range: 11.224.9 mm). On sagittal plane, the
distance between the rostrum sphenoidale (the ante-
rior wall of the sphenoid sinus) to the anterior wall of
the sellar wall was 20.6 6 1.5 mm (range: 18.923
mm). After the bony structures were removed, fur-
ther measurements were done. The width of the pi-
tuitary, which is the distance between the medial
margins of the anterior curvature of the ICA at its
longest axis was measured as 23.2 6 3 mm (range:
18.225.7 mm), while the distance between the pos-
terior curvature of the ICA at its shortest axis was
19.7 6 4.9 mm (range: 13.226.2 mm). The results
of all the measurements in the rst and second steps
have been presented in Tables 1 and 2. In the third
step, the distance between the optic chiasm and the
superior margin of the pituitary was measured as 5.3
6 1.3 mm (range: 3.46.9 mm).
DISCUSSION
Earlier studies evaluated the surgical landmarks of
the sphenoid sinuses and their relationships with
TABLE 1. The Results of the Bony Landmark
Measurements (Fig. 1B)
Mean 6 SD (mm) Range (mm)
a 12.1 6 2.4 9.316.3
b 8.7 6 3.1 4.512.6
c 21 6 2.5 17.724.4
d 18 6 3.1 11.224.8
e 28 6 6.2 20.837.4
f 20.6 6 1.5 18.923
630 Unlu et al.
neurovascular structures for pituitary surgery in
detail. However, most of these anatomical studies
were conducted either by sagittal sectioning of
cadaver heads or by en-block removal of the sphe-
noid sinuses and sella turcica (Renn and Rhoton,
1975; Fujii, 1979; Lang, 1989). Afterwards, endo-
scopic surgeons started to investigate this same area
from an endonasal point of view (Jankowski et al.,
1992; Sethi et al., 1995; Carrau et al., 1996; Cappa-
bianca et al., 2002, 2004; Kassam et al., 2005;
Catapano et al., 2006; Sethi and Leong, 2006; Laufer
et al., 2007). Their ndings on the intrasphenoidal
bony landmarks and the incidence of anatomical var-
iations were similar to the results of earlier nonendo-
scopic studies. Nevertheless, it was conrmed that
use of endoscopes with their panoramic, well illumi-
nated and angled vision, facilitated an improvement
for intraoperative recognition of these variations,
which allowed altering the approach and technique
or the selection of instruments according to the spe-
cic anatomic presentation during the surgery of this
area as well as in pituitary surgery.
Both the increasing popularity of endoscopic pitui-
tary surgery and recent introduction of expanded
endonasal approaches around the sphenoid and
sellar regions have emphasized once more the indis-
pensable role of endoscopic anatomical knowledge of
the neurovascular structures beyond the bony cage
of the sphenoid sinuses (Kassam et al., 2005; Laufer
et al., 2007). Because thorough knowledge on the
surrounding structures and their relationships with
each other as well as on the variations of these struc-
tures may eventually affect the success and safety of
surgery, the measurements of neurovascular struc-
tures were done after the removal of the covering
bone. Our measurements showed that the location of
these structures (despite having a conventional pat-
tern) could substantially vary, probably related to the
extent of sphenoid pneumatization during its devel-
opmental period. The shape of the sinuses, optic
nerve, and carotid protuberances may also vary.
Sometimes there is no bony layer over either optic
nerves or carotid arteries (Renn and Rhoton, 1975).
In our study, no such bony variations were observed.
However, the bone over these structures was some-
times very thin like a translucent membrane. Another
nding was that the distance from the sinus ostium
to the sellar wall and the width of the sinus might
vary depending on different pneumatization grades
(Das et al., 2001; de Divitiis et al., 2002). Similarly,
it was determined that the distance between both ca-
rotid arteries might also vary signicantly below and
over pituitary gland. Additionally, as shown earlier,
the pituitary gland may overlap the intracavernous
carotid artery with a tongue-like projection (Rhoton
et al., 1979). Nevertheless, we did not observe such
a tongue like extension of pituitary gland into the
cavernous sinus. Our measurements also suggest
that the distance between the optic chiasm and the
superior margin of pituitary after removing the bony
layers is normally not adequate for further dissection
in between for suprasellar pathologies. Thus, caudal
suspension of the pituitary gland is a necessity to
gain space, if one desires to proceed through, as sug-
gested by some authors (Kassam et al., 2005; Laufer
et al., 2007).
Our dissections under endoscopic view proved
once again the feasibility of fully endoscopic endo-
nasal transsphenoidal approach for pituitary surgery
in this very narrow area. In spite of the anatomic
variations measured, all the important landmarks
were identied with the bright illuminated panoramic
vision of the sphenoid walls. Knowing the variations,
one should also consider alterations in the normal
anatomical structures due to pathologies in that
region that actually indicates surgery. Thus, preoper-
ative radiological workup certainly enhances the an-
atomical knowledge and consequently the safety of
the endoscopic approach by delineating soft tissue
and bony structures before surgery. This additional
information guides the surgeons like a roadmap dur-
ing dissections by giving hints about the surgical
anatomy of that specic case. Accordingly, utilization
of the same radiological information by computer-
assisted navigation systems could also be a very
helpful companion to endoscopic dissections, but
they should never be a replacement to the anatomi-
cal knowledge due to their limitations and pitfalls. In
our opinion, another key factor for optimal endo-
scopic surgery lies in the ongoing multidisciplinary
management using 34 handed surgical technique,
which merges and boosts the best skills of each dis-
cipline to achieve the ideal goals.
Endoscopic approaches are exponentially being
used in our surgical elds, because of their distinct
advantages and efciency. Hereby, we emphasize
once more the importance of anatomical dissections
for endoscopic surgery of the sellar region, as the
image of anatomical structures on the monitor may
appear slightly altered in comparison to familiar mi-
croscopic view.
Anatomical studies are essential in increasing the
anatomical knowledge and improving surgical per-
formance in line with technological progress. Our en-
doscopic dissections showed us improved, wide pan-
oramic view of the close relationship of ICA and optic
nerves with the pituitary gland, presenting with
some variations, which might be a concern during
surgery. Thus, for clinical practice, our results point
out the eminent role of preoperative radiological
mapping of the bony structures and soft tissues in
each case, even though the pure endoscopic endo-
nasal transsphenoidal approach provides excellent
view of the surgical landmarks.
TABLE 2. The Results of the Measurements of the
Carotid Artery, Optic Nerve, and Pituitary
Relationships (Fig. 2B)
Mean 6 SD (mm) Range (mm)
g 12.7 6 1.1 11.514.5
h 7.3 6 2.5 4.510.3
i 23.1 6 3 18.225.7
j 19 6 4.4 13.524.9
k 19.7 6 4.9 13.226.2
l 36.6 6 6.6 29.246.4
m 34.2 6 6.8 25.844.1
631 Endoscopic Anatomy for Pituitary Surgery
REFERENCES
Batay F, Vural E, Karasu A, Al-Mefty O. 2002. Comparison of the ex-
posure obtained by endoscope and microscope in the extended
trans-sphenoidal approach. Skull Base 12:119124.
Cappabianca P, Cavallo LM, Colao A, de Divitiis E. 2002. Surgical
complications associated with the endoscopic endonasal trans-
sphenoidal approach for pituitary adenomas. J Neurosurg
97:293298.
Cappabianca P, Cavallo LM, de Divitiis E. 2004. Endoscopic endo-
nasal transsphenoidal surgery. Neurosurgery 55:933941.
Carrau RL, Jho HD, Ko Y. 1996. Transnasal-transsphenoidal endo-
scopic surgery of the pituitary gland. Laryngoscope 106:914918.
Catapano D, Sloffer CA, Frank G, Pasquini E, DAngelo VA, Lanzino
G. 2006. Comparison between the microscope and endoscope in
the direct endonasal extended transsphenoidal approach: Ana-
tomical study. J Neurosurg 104:419425.
Das K, Spencer W, Nwagwu CI, Schaeffer S, Wenk E, Weiss MH,
Couldwell WT. 2001. Approaches to the sellar and parasellar
region: Anatomic comparison of endonasal-transsphenoidal,
sublabial-transsphenoidal, and transethmoidal approaches. Neu-
rol Res 23:5154.
de Divitiis E, Cappabianca P, Cavallo LM. 2002. Endoscopic trans-
sphenoidal approach: Adaptability of the procedure to different
sellar lesions. Neurosurgery 51:699707.
Fujii K, Chambers SM, Rhoton AL Jr. 1979. Neurovascular relation-
ship of the sphenoid sinus - a microsurgical study. J Neurosurg
50:31-39
Hardy J. 1969. Microneurosurgery of the hypophysis: A subnasal
transsphenoidal approach with television magnication and tele-
vised radiouoroscopic control. In: Rand RW, editor. Microneuro-
surgery. St. Louis, MO: CV Mosby. p 87103.
Jankowski R, Auque J, Simon C, Marchal JC, Hepner H, Wayoff M.
1992. Endoscopic pituitary tumor surgery. Laryngoscope 102:
198202.
Kassam A, Snyderman CH, Mintz A, Gardner P, Carrau RL. 2005.
Expanded endonasal approach: The rostrocaudal axis. I. Crista
galli to the sella turcica. Neurosurg Focus 19:E3.
Lang J. 1989. Clinical anatomy of the nose, nasal cavity and para-
nasal sinuses. New York: Thieme Medical Publishers, Inc. p 85
98.
Laufer I, Anand VK, Schwartz TH. 2007. Endoscopic, endonasal
extended transsphenoidal, transplanumtranstuberculumapproach
for resectionof suprasellar lesions. J Neurosurg106:400406.
Renn WH, Rhoton AL Jr. 1975. Microsurgical anatomy of the sellar
region. J Neurosurg 43:288298.
Rhoton AL Jr, Hardy DG, Chambers SM. 1979. Microsurgical anat-
omy and dissection of the sphenoid bone, cavernous sinus and
sellar region. Surg Neurol 12:63104.
Sethi DS, Leong JL. 2006. Endoscopic pituitary surgery. Otolaryngol
Clin North Am 39:563583.
Sethi DS, Stanley RE, Pillay PK. 1995. Endoscopic anatomy of the
sphenoid sinus and sella turcica. J Laryngol Otol 109:951955.
Stammberger H. 1986. Endoscopic endonasal surgeryConcepts in
treatment of recurring rhinosinusitis. II. Surgical technique. Oto-
laryngol Head Neck Surg 94:147156.
632 Unlu et al.