ORIGINAL RESEARCHSKULL BASE SURGERY

E dyfo

Is i, MZaSpten

AB

INfospatwiperMEgofivbaswa

theREacc

poligartrot7.0res

COoriana

dre

Su

Tinprotioof the ITF may lead to alternative, less invasive approachesto

poforfosIT

nicgoma

ofph

golarlaraptum

su

doITgelitetrarar

an

fraco

pro

sivan

un

dreThtatan

IT

OtolaryngologyHead and Neck Surgery (2009) 140, 861-865

019doipathology in this region.Anatomically, the superior border of the ITF is com-sed of the greater wing of the sphenoid featuring theamen ovale and spinosum (Fig 1), and the temporalsa, which contains the temporalis muscle. Medially, the

F is bounded by the lateral pterygoid plate and commu-

MATERIALS AND METHODS

Ten sides in five adult cadaver heads were dissected endoscop-ically in a position simulating that in the operating room toachieve as much real surgical information as possible. Trans-nasal dissection was performed using 4 18-cm rod-lens

Received September 5, 2008; revised November 10, 2008; accepted February 12, 2009.

4-5998/$36.00 2009 American Academy of OtolaryngologyHead and Neck Surgery Foundation. All rights reserved.ndoscopic transnasal stussa: A new orientation

lam R. Herzallah, MD, Ross Germangazig, Egypt; and Miami, FL

onsorships or competing interests that may be relevant to con-t are disclosed at the end of this article.STRACT

TRODUCTION: The medial portion of the infratemporalsa (ITF) is not infrequently involved in sinonasal and skull basehologies. However, endoscopic view of the ITF remains unclearth lack of studies addressing this region from the endoscopicspective.THODS: Using an extended endoscopic approach, the ptery-

palatine and infratemporal fossae were dissected in 10 sides ofe adult cadaver heads. A plane of dissection along the pterygoide and the infratemporal surface of the greater sphenoid wings developed. High-quality images were produced by couplingvideo camera to a digital recording system.SULTS: The foramen rotundum, ovale, and spinosum wereessed and new landmarks were described from the endoscopic

int of view. The sphenomandibularis muscle was also high-hted. Maxillary and mandibular nerves and middle meningealery were all identified. Columellar measurements to the foramenundum and ovale ranged from 6.1 to 8.0 cm for the former andto 9.1 cm for the latter, with a mean of 6.75 cm and 7.78 cm

pectively.NCLUSION: The current study provides a novel endoscopic

entation to the medial ITF. Such knowledge should provide antomical basis for experienced surgeons to endoscopically ad-ss this region with more safety and efficacy.

2009 American Academy of OtolaryngologyHead and Neckrgery Foundation. All rights reserved.

he infratemporal fossa (ITF) is a deeply seated regionbeneath the skull base. Management of lesions arising

or extending to the ITF often requires invasive surgicalcedures, a source of cosmetic and functional complica-

ns.1-6 A better understanding of the endoscopic anatomy:10.1016/j.otohns.2009.02.020of the infratemporal

D, and Roy R. Casiano, MD,

ates with the pterygopalatine fossa (PPF) via the ptery-maxillary fissure. The ITF is limited laterally by thendibular ramus, extends anteriorly to the posterior wallthe maxillary sinus, and opens inferiorly into the para-aryngeal space.7The ITF approaches are categorized as lateral (transzy-matic and lateral infratemporal), inferior (transmandibu-and transcervical), or anterior (transfacial, transmaxil-

y, transoral, and transpalatal).8 In general, transfacialproaches are indicated for sinonasal or nasopharyngeal

ors invading the ITF.Concordant with the evolution of minimally invasive

rgery in this era, many attempts have been made to eno-scopically address different pathologies extending to theF. Transnasal excision of advanced juvenile nasopharyn-al angiofibromas (JNA) is the leading example in recentrature.9-11 A handful of case reports have also describednsnasal endoscopic access to the ITF in managing thee maxillary nerve schwannomas, solitary fibrous tumors,d mucoceles.11-13 Additionally, for a small subset of in-temporal masses such as lymphomas and rhabdomyosar-mas, biopsy results before a definitive procedure shouldve useful in guiding management.Despite rapid progress in the number of minimally inva-e approaches being performed endoscopically, the ITFatomy as described from the endoscopic view remainedclear, with the literature lacking studies that fully ad-ssed this region from the unique endoscopic perspective.e aim of this work was to provide a new endoscopic orien-ion to the medial portion of the ITF and to describe theatomical details and measurement variations of some keyF landmarks, from the transnasal endoscopic approach.

en

30ligco

proco

Ltdplitutinvco

sco

us

dleturdleec

themu

ce

cre

art

disbodripoa

pema

ma

Thbeallwa

tifithaforphatetheThupV2betaktoan

tinFigprono

lacma

FigincofptethevidpteVCtioofon

thefosUlLA

862 OtolaryngologyHead and Neck Surgery, Vol 140, No 6, June 2009doscopes (Karl Storz and Co., Tuttlingen, Germany) with 0-,-, and 70-degree lenses. The endoscope was connected to aht source through a fiberoptic cable and to a video cameraupled to a 21-inch monitor. High-quality digital files wereduced utilizing a video camera connected to a digital re-

rding system ((DV-Cam, JVC, Victor Company of Japan,., Tokyo, Japan). Digital pictures were reproduced by cou-

ng the DV-Cam to a computer video capture system. Insti-ional Review Board approval was not necessary as the studyolves de-identified cadaveric specimens, and thus is not

nsidered human subject research.To accomplish an adequate access to the ITF, an endo-pic transnasal transantral transpterygoid approach was

ed. The approach was started by performing partial mid-turbinectomy, removing the inferior half of the middle

binate and preserving its olfactory mucosa. A wide mid-meatal antrostomy and anterior and posterior ethmoid-

tomies were also classically performed.A vertical incision was performed in the posterior part ofmiddle meatus just behind the posterior fontanelle. A

cosal flap was elevated to expose the thick orbital pro-ss of the vertical plate of the palatine bone and its ethmoid

ure 1 Inferior view of the skull base. BPP, base of pterygoidcess; GWS, infratemporal surface of the greater wing of sphe-

id; FO, foramen ovale; FS, foramen spinosum; FL, foramenerum; CC, carotid canal; JF, jugular foramen; FM, foramengnum.st. The sphenopalatine foramen and the sphenopalatineery were then identified and carefully dissected.

AJSoA 30-degree endoscope was consistently used for furthersection. The orbital process that forms the anteriorundary of the sphenopalatine foramen was taken off, withlling sometimes helpful to remove this thick bone. Thesterior wall of the maxillary sinus was then removed withcurette in medial-to-lateral direction, and the anteriorriosteum of the PPF was then opened. Generally, thexillary nerve (V2) lies at a higher level than the internalxillary artery (IMA) in the lateral portion of the PPF.is interrelationship of the neurovascular structures hasen further detailed in our previous study14 and shouldow safer dissection into the PPF. The fat filling the PPFs then dissected and different IMA branches were iden-ed and transected as necessary. The pterygoid processt forms the posterior wall of PPF was identified. Threeamina open into the back wall of the PPF. These are thearyngeal canal most medially, the vidian canal immedi-ly lateral to this in the base of the pterygoid process, andforamen rotundum superolateral to vidian canal (Fig 2).

e V2 passes through the foramen rotudum, crossing theper part of the PPF towards the infraorbital canal. As the

lies high up in the PPF, dissection was maintained justlow its level to avoid inadvertent injury. However, care isen that, with dissection of the fat in the PPF, the V2 tendsdroop below the level foramen rotundum as it moves

terolaterally towards the inferior orbital fissure to con-ue as the infraorbital nerve.14

ure 2 Diagrammatic anterior view of the sphenoid boneluding its pterygoid process and greater wing (GWS). The basethe pterygoid process (BPP) forms the posterior boundary of therygopalatine fossa. Three foramina open into the back wall offossa. These are the pharyngeal canal (PC) most medially, theian canal (VC) immediately lateral to this in the base of therygoid process, and the foramen rotundum (FR) superolateral to. The dashed arrow represents the course of endoscopic dissec-

n on the pterygoid base and then along the infratemporal surfacethe GWS towards the foramen ovale. The yellow and red arrowsthe right side depict the course of the maxillary nerve (V2) andtortuous internal maxillary artery (IMA) in the pterygopalatine

sa. Modified with permission from Daniels DL, Mark LP,mer JL, Mafee MF, McDaniel J, Shah NC, Erickson S, Sether, Jaradeh SS. Osseous anatomy of the pterygopalatine fossa.

NR Am J Neuroradiol. 1998 Sep;19(8):1423-32. The Americanciety of Neuroradiology.

mu

theres

thefro

latThan

pte

IThean

relthe

RE

Usofpogoma

po

bedibgeles

plahefacthesco

skthe

FigandiangreIMartTBLPfrosur

Figfosmu

the(Gprointsec

(V

FigpteLP(UitsgrefictemGetemSuSc

863Herzallah et al Endoscopic transnasal study of the infratemporal . . .The inferior and superior heads of the lateral pterygoidscle attach to the lateral surface of the pterygoid base andinfratemporal surface of the greater sphenoid wing,

pectively. Thus, further dissection was performed alongse bones by separating the lateral pterygoid muscle headsm their bony attachments.More access to the ITF was accomplished by continuederal dissection of the posterior wall of the maxillary sinus.e deep belly of the temporalis muscle was then identifiedd dissected lateral to the foramen rotundum and lateralrygoid muscle.In order to appropriately plan the dissection in the medial

F, a Fisch B approach was implemented on one cadaverad. Endoscopic transnasal dissection was also performedd the structures revealed from both approaches were cor-ated. This allowed better understanding of the anatomy inother endoscopically dissected sides.

SULTS

ing the aforementioned approach, endoscopic dissectionthe neurovascular structures in the PPF and the medialrtion of the ITF was performed. The vidian nerve, ptery-palatine ganglion, and maxillary nerve as well as thexillary artery and its branches were all identified (Fig 3).Endoscopic dissection behind the lateral portion of thesterior maxillary wall was performed as far as the deep

ure 3 Endoscopic view after removal of the left posteriorposterolateral maxillary wall. PC, posterior choana; VC, vid-

canal; SS, sphenoid sinus; PPG, pterygopalatine ganglion; GPN,ater palatine nerve; V2, maxillary nerve; ION, infraorbital nerve;A, internal maxillary artery; PSAA, posterior superior alveolarery; IOA, infraorbital artery; LBr, lateral pterygoid branch of IMA;r, branch of IMA to the deep belly of temporalis muscle (TM);M, lateral pterygoid muscle with its two heads partially separated

m the lateral surface of the pterygoid base and the infratemporalface of the greater wing of sphenoid bone (GWS).

depprolly of the temporalis muscle, or the so-called sphenoman-ularis muscle (Figs 3, 4). These vertically oriented fibers

nerally restrict further lateral endoscopic dissection un-s electrocautery is used.On the other hand, we were able to develop a dissectionne by endoscopic separation of the lower and upper

ads of the lateral pterygoid muscle from the lateral sur-e of the pterygoid base and the infratemporal surface ofgreater wing of the sphenoid bone, respectively. Endo-pic dissection, thus, proceeded posteriorly along the

ull base toward the foramen ovale, the latter transmittingmandibular division of the trigeminal nerve. Posterior to

ure 5 Endoscopic view of the left medial infratemporalsa. (A) The upper and lower heads of the lateral pterygoidscle (LPM [UH] & LPM [LH], respectively) are dissected frominfratemporal surface of the greater wing of the sphenoid

WS) and from the lateral surface of the base of pterygoidcess (BPP). FR, foramen rotundum; VC, vidian canal; IMA,

ernal maxillary artery retracted inferolaterally. (B) Further dis-tion of the LPM allows identification of the mandibular nerve3) as it emerges from the foramen ovale. The dashed black line

ure 4 Diagrammatic views of the left lateral and medialrygoid muscles and the deep belly of the temporalis muscle.M, lateral pterygoid muscles, with its upper and lower headsH and LH, respectively); MPM, medial pterygoid muscle, withsuperficial and deep heads (SH and DH, respectively); GWS,ater wing of sphenoid; MS, maxillary sinus; SUP TM, super-

ial part of the temporalis muscle; Deep TM, deep belly of theporalis muscle ( sphenomandibularis muscle). Modified from

ers C, Nyssen-Behets C, Cosnard G, et al. The deep belly of theporalis muscle: an anatomical, histological and MRI study.

rg Radiol Anat. 2005; 27:184-191. With permission of SpringerienceBusiness Media.icts an area of the lateral pterygoid base that could be drilled tovide a wider access to the medial infratemporal fossa.

thewime

me

era

foslatme

frolat(T

DI

Enwhne

doofma

pla

deov

ca

inva

tocu

wi

su

ca

ex

stuma

oriinsFushsptheagres

scr

tone

tis

plamu

theattthemu

a g

thisepofThlan

en

ITmo

FigterstrThsur

witheptema

ingtoforthema

TC(ec

H

M

864 OtolaryngologyHead and Neck Surgery, Vol 140, No 6, June 2009foramen ovale, the foramen spinosum was identifiedth a bridge of bone between both foramina. The middleningeal artery was identified passing through the fora-n spinosum posterolateral to the mandibular nerve. Lat-l to this, the mandibular condyle, laying in its articularsa, received the insertion of the endoscopically dissectederal pterygoid muscle (Figs 5, 6). Columellar measure-nts to the foramen rotundum and foramen ovale rangedm 6.1 to 8.0 cm for the former and 7.0 to 9.1 cm for theter, with a mean of 6.75 cm and 7.78 cm, respectivelyable 1).

SCUSSION

doscopic access to the ITF is not uncommonly requireden dealing with advanced or deeply located sinonasal

oplasms. The endoscopic approach, whenever applicable,es not only avoid the functional and cosmetic morbidityopen surgical approaches, but should also provide agnified, multi-angled view, with good access to the neo-stic projections at various skull base foramina.In 2001, Hartnick et al15 used the cadaveric model to

scribe a new surgical endoscopic approach to the foramenale via Gillies and lateral brow incisions. However, be-use many lesions extending to this area actually originatethe sinonasal region or at the pterygopalatine fossa, thelue of such lateral endoscopic approach might be limitedvisualization of the ITF or performing a biopsy. In therrent study, we have explored this region in continuityth the sinonasal endoscopic dissection, describing the key

ure 6 Endoscopic view of the left infratemporal fossa pos-olateral to the base of the pterygoid process (BPP). The labeleductures have been confirmed by an open lateral approach. (A)e lateral pterygoid muscle (LPM) is dissected of the lateralface of the BPP and of the infratemporal surface of the greaterng of the sphenoid bone (GWS). The mandibular nerve (V3) and

middle meningeal artery (MMA) are identified. BBP, base ofrygoid process; FR, foramen rotundum; VC, vidian canal; MC,ndibular condyle in its articular fossa. (B) A closer view show-the V3 emerging from the foramen ovale (FO). Posterolateral

the FO is a bony bridge (BB) separating the latter from theamen spinosum (FS) that transmits the MMA. Of note also isingbony spine (BS) at the posterolateral edge of the FS. MC,

ndibular condyle.rgical landmarks from an anterior approach. In the surgi-l realm, this should allow more complete endoscopiccision of advanced sinonasal lesions.To the best of our knowledge, this is the first endoscopicdy to highlight the sphenomandibularis muscle. Thissticatory muscle, first described by Dunn et al in 1996,16ginates from the greater wing of the sphenoid bone anderts distally on the coronoid process of the mandible.rther anatomical work by Geers and coworkers17 hasown that the so-called sphenomandibularis muscle corre-onds to the deep portion of the temporalis muscle, sincere is no epimysial septum between the two structures. In

reement with the endoscopic orientation provided in ourults, the authors of the last anatomical study have de-ibed the medial limit of the temporalis muscle deep bellycome close to the foramen rotundum and the maxillaryrve, with a 4- to 7-mm-wide space containing adiposesue separating the two structures.17In the present study, we have proposed a new dissectionne along the bony attachments of the lateral pterygoidscle heads in order to access the superomedial portion ofITF beneath the foramen ovale. If the dissection is

empted away from the base of the pterygoid process, bothlateral pterygoid and the deep belly of the temporalis

scle would be encountered, limiting the development ofood surgical plane.Finally, some key endoscopic landmarks are described ins report. The foramen ovale and the bony bridge thatarates it from the foramen spinosum, as well as the spinethe latter, were sequentially identified endoscopically.ese bony structures could serve as helpful endoscopicdmarks if surgical dissection in this region is warranted.In conclusion, the current work should present a new

doscopic surgical orientation to the medial portion of theF. In experience d hands, this would allow safer as well asre effective management of lesions extending to, or aris-

able 1olumellar measurements to the foramen rotundum

FR) and foramen ovale (FO) as measured by thendoscopic transnasal approach in five adultadaver heads

ead number

Columellar measurements (cm)

Right side Left side

FR FO FR FO

1 7 8.4 7 7.82 7.7 8.9 8 9.13 6.3 7.0 6.5 7.14 6.3 7.6 6.1 7.25 6.3 7.4 6.3 7.3

ean distance 6.72 7.86 6.78 7.7at, such a surgically challenging area.

AUTHOR INFORMATION

From the Department of Otolaryngology, Faculty of Medicine, ZagazigUniversity, Egypt (Dr Herzallah); and the Department of Otolaryngology,University of Miami, Miller School of Medicine, Miami, FL (Drs Germaniand Casiano).Corresponding author: Islam R. Herzallah, MD, Department of Otolaryn-gology, Faculty of Medicine, Zagazig University, Zagazig, Egypt.E-mail address: iherzallah@gmail.com.Presented at the Annual Meeting of the American Academy of Otolaryn-gologyHead and Neck Surgery, Chicago, IL, September 21-24, 2008.

AUTHOR CONTRIBUTIONS

Islam R. Herzallah, review of previous studies, performing the dissectionwork, and writing the manuscript; Ross Germani, assisting in the dissec-tion work, reviewing the written manuscript; Roy R. Casiano, guiding thestudy, providing dissection ideas based on the surgical experience, andreviewing the work.

DISCLOSURES

CoandSp

RE

1.

2.

3. Fisch U. Infratemporal fossa approach for lesions in the temporal boneand base of the skull. Adv Otorhinolaryngol 1984;34:25466.

4. Fisch U, Fagan P, Valavanis A. The infratemporal fossa approach forthe lateral skull base. Otolaryngol Clin North Am 1984;17:51352.

5. Sekhar LN, Schramm VL Jr, Jones NF, et al. Operative exposure andmanagement of the petrous and upper cervical internal carotid artery.Neurosurgery 1986;19:96782.

6. Sekhar LN, Schramm VL Jr, Jones NF. Subtemporal-preauricularinfratemporal fossa approach to large lateral and posterior cranial baseneoplasms. J Neurosurg 1987;67:48899.

7. Grant J. An atlas of anatomy. Baltimore, MD: Williams & Wilkins;1972.

8. Tiwari R, Quak J, Egeler S, et al. Tumors of the infratemporal fossa.Skull Base Surg 2000;10:19.

9. Onerci TM, Yucel OT, Ogretmenoglu O. Endoscopic surgery in treat-ment of juvenile nasopharyngeal angiofibroma. Int J Pediatr Otolar-yngol 2003;67:121925.

10. Nicolai P, Berlucchi M, Tomenzoli D, et al. Endoscopic surgery forjuvenile angiofibroma: when and how. Laryngoscope 2003;113:775 82.

11. Robinson S, Patel N, Wormald PJ. Endoscopic management of benigntumors extending into the infratemporal fossa: a two-surgeon transna-sal approach. Laryngoscope 2005;115:181822.

12. Jurado-Ramos A, Romero FR, Baos EC, et al. Minimally invasiveendoscopic techniques for treating large, benign processes of the nose,paranasal sinus, and pterygomaxillary and infratemporal fossae: soli-

13.

14.

15.

16.

17.

865Herzallah et al Endoscopic transnasal study of the infratemporal . . .mpeting interests: Roy R. Casiano, consultant for Medtronic/XomedGyrus.

onsorships: None.

FERENCES

Fisch U. Infratemporal fossa approach to tumours of the temporal boneand base of the skull. J Laryngol Otol 1978;92:94967.Fisch U, Pillsbury HC. Infratemporal fossa approach to lesions in thetemporal bone and base of the skull. Arch Otolaryngol 1979;105:99107.tary fibrous tumour. J Laryngol Otol 2008;11:15.Park CS, Park YJ, Cho JH, et al. Infratemporal fossa mucocele withtrigeminal nerve compression. Auris Nasus Larynx 2008;35:4558.Herzallah IR, Elsheikh EM, Casiano RR. Endoscopic endonasalstudy of the maxillary nerve: a new orientation. Am J Rhinol 200721:637 43.Hartnick CJ, Myseros JS, Myer CM 3rd. Endoscopic access to theinfratemporal fossa and skull base: a cadaveric study. Arch Otolaryn-gol Head Neck Surg 2001;127:13257.Dunn GF, Hack GD, Robinson WL, et al. Anatomical observation ofa craniomandibular muscle originating from the skull base: the sphe-nomandibularis. Cranio 1996;14:97103.Geers C, Nyssen-Behets C, Cosnard G, et al. The deep belly of thetemporalis muscle: an anatomical, histological and MRI study. SurgRadiol Anat 2005;27:18491.

Endoscopic transnasal study of the infratemporal fossa: A new orientationMATERIALS AND METHODSRESULTSDISCUSSIONAUTHOR INFORMATIONAUTHOR CONTRIBUTIONSDISCLOSURESREFERENCES