Extended Endoscopic Techniques for Sinonasal Resections

E x t e n d e d En d o s c o p i c Te c h n i q u e s f o r Sinonasal Resections Richard J. Harvey, MDa,*, Richard M. Gallagher, Raymond Sacks, MDc MD b , KEYWORDS  Endoscopic  Skull base  Tumor  Sinonasal  Angiofibroma  Juvenile nasopharyngeal angiofibroma  Inverted papilloma  Osteoma Endoscopic resections of benign neoplastic disease of the anterior skull base and paranasal sinuses is now widely practiced.1 Selected malignancies can also be successfully managed by an endoscopic approach.2,3 However, the approach should never dictate the surgery performed. Anatomic location and areas involved by a pathologic condition should always be the determining factor. Similarly, pathology such as inverted papilloma, should never imply a particular surgery (endoscopic medial maxillectomy or lateral rhinotomy). Although endoscopic resection has replaced many open approaches at our institutions, the authors still use a combination of techniques to remove extensive disease. The endoscopic surgeon performing extended procedures should be equally comfortable performing a similar open procedure. Endoscopic surgery should not imply conservative surgery. If a pathologic lesion is considered irresectable via an open approach then it is axiomatic that this is true for the endoscopic option. There are a variety of open approaches that can be applied and they have been well described,4,5 however, they have a limited role in the management of benign disease. The midface degloving approach is perhaps 1 open approach that is still sometimes used to manage lesions for which an endoscopic approach may not suffice. a Rhinology and Skull Base Surgery, Department of Otolaryngology/Skull Base Surgery, St Vincent’s Hospital, 354 Victoria Street, Darlinghurst, Sydney, New South Wales 2010, Australia b Rhinology and Skull Base, Department of Otolaryngology/Skull Base Surgery, St Vincent’s Hospital, Suite 1002b, 438 Victoria Street, Darlinghurst, Sydney, New South Wales 2010, Australia c Department of Otolaryngology/Head & Neck Surgery, Concord General Hospital, 354 Victoria Street, Concord, Sydney, New South Wales 2010, Australia * Corresponding author. E-mail address: richard@sydneyentclinic.com Otolaryngol Clin N Am 43 (2010) 613–638 doi:10.1016/j.otc.2010.02.016 oto.theclinics.com 0030-6665/10/$ – see front matter Crown Copyright ª 2010 Published by Elsevier Inc. All rights reserved. 614 Harvey et al The authors believe the foundations of successful extended endoscopic surgery, whether for accessing a lateral frontal mucocele or removing malignant disease, relies on 5 important concepts: preoperative planning (surgery and equipment required), obtaining appropriate surgical access, micro- and macrovascular control; reconstruction of nasolacrimal physiology; and postoperative care of the large endoscopic cavity (Table 1). PREOPERATIVE PLANNING The philosophy of complete endoscopic resection can be retained without the need for traditional en bloc surgery. The limits of the area to be resected and bone removed can often be defined before surgery begins. An attempt should be made to define the surgical margins preoperatively. This ensures that a surgical plan is adhered to and will enhance total removal. The authors believe there needs to be a shift away from the patho-etiology focus of traditional teachings and emphasize the need to resect anatomic zones or regions, therefore tailoring surgery to the exact extent of disease and preserving normal structures. This is not pathology-specific surgery but site-specific surgery. The ability to gain good visualization and access to the anatomic region of the lesion is essential. Particularly in malignant disease, being able to accurately map resection margins is vital for intra- and postoperative decision making (Fig. 1). Further resection of positive frozen section margins can be inaccurate if many (>10) biopsies are taken. Postoperatively, accurate surgical mapping aids radiation oncologists in defining treatment fields and assists focused endoscopic surveillance. ENDOSCOPIC SURGICAL ACCESS There are 4 areas notorious for recurrence and present challenging access1: 1. 2. 3. 4. Anterolateral maxilla Frontal sinus Supraorbital ethmoid cell Floor of a well-pneumatized maxillary sinus. Table 1 Foundations of extended endoscopic surgery Preoperative planning Ensure that imaging, skill, equipment, and a predefined surgical plan are created Surgical access Accessing anterolateral disease of the maxilla and within the frontal sinus requires unconventional or combination techniques Anatomic orientation Preoperatively defining a structured approach to identify fixed anatomic landmarks Vascular control Microvascular management: preoperative reduction of associated inflammatory changes, anesthetic techniques, and intraoperative vasoconstriction Macrovascular control with a structured approach to the ethmoidal, sphenopalatine, internal maxillary, and carotid artery Reconstruction Ensuring a functional lacrimal system, the formation of a final cavity that will allow relatively normal nasal physiology Reconstruction of dura or periorbita Postoperative management of the large cavity Controlling adhesions, crusting, bacterial colonization and facilitating mucosalization Extended Endoscopic Surgery Fig. 1. Systematic systems to ensure pathologic resection margins greatly aid communication between nursing staff, the pathology and radiation oncology teams. Blank template on the left and an operative example from surgery on the right. Many staging systems have been developed for benign pathologic conditions commonly managed endoscopically. Examples by Cannady and colleagues,6 Jameson and Kountakis,7 Krouse,8 and Woodworth and colleagues9 all touch on important aspects in the groupings of their patients. However, unlike malignant staging, it is fundamentally the completeness of surgical resection of the tumor that dictates the final outcome for benign disease. These staging systems reflect surgical complexity of access rather than intrinsic disease factors such as nodal or metastatic spread. Synchronous and metachronous malignant disease may occur but the effect of these events on outcome is unlikely to be reflected in these staging systems. Potentially, the difficult or higher-stage tumors are simply those lesions associated with more difficult access. Predefining regions or zones that require endoscopic access and resection has become an important process in our institution (Table 2). The limits of tissue removal may too easily align with surgeon comfort rather than anatomic boundaries defined by the presurgical clinical and radiological examination. The principles of en bloc resection, from its oncologic foundations in managing malignant disease, are often followed by some surgeons to ensure that the appropriate margins have been reached. With careful planning and preoperative evaluation of radiology, it is possible to define the zone of resection likely to be required. Table 2 outlines our current surgical approach to endoscopic resection. Accessing Anterolateral Disease Five zones were developed and are used at the Medical University of South Carolina (MUSC) Rhinology and Skull Base and St Vincent’s Rhinology and Skull Base Divisions when planning surgical access in endoscopic tumor removal (Fig. 2A; Table 3). 615 616 Harvey et al Table 2 Surgical access planning Anatomic Site Pathology Involves Surgical Access Consideration Anterolateral maxilla and infratemporal fossa Zone 2/3 Appropriate angled instruments need to be available; 40 burrs and debriders are not angled enough for zone 3; 60–75 instruments are usually required Ancillary techniques required such as maxillary trephine, maxillotomy, or transseptal access An open approach may be better Zone 3 or 4 Zone 5 Frontal sinus Supraorbital ethmoid Medial quarter of orbital roof Medial half of orbital roof or lateral posterior and anterior walls Orbital roof lateral to midpoint Frontal recess Unilateral access with a Draf 2a or 2b Draf 3 Possible trephine Anterior ethmoidal artery A dehiscent anterior ethmoidal artery may be obscured on imaging because of a nearby pathologic lesion; control is required in approach The potential for dural or periorbital injury needs to be balanced with pathology and risk of recurrence Orbitocranial cleft Maxillary floor Dental roots Low maxilla relative to nasal floor Zone 1: tumor is limited to        Septum Turbinates Middle meatus Ethmoid Frontal Sphenoid sinuses Medial orbital wall. External trephine or osteoplastic flap required Draf 2b or 3 required as reconstruction of the recess with exposed bone requires greater intervention Damage to roots likely or pathology may be of odontogenic nature with tooth extraction or endodontics required Angled instruments or ancillary access, such as maxillary trephination or modified medial maxillectomy required Extended Endoscopic Surgery Fig. 2. MUSC zones: the MUSC endoscopic resection zones (A). Zones 1 to 5 demonstrate increasing anterior and lateral disease (B–F). 1, nasal cavity; 2, medial to infraorbital nerve (ION); 3, lateral to ION and up to the zygomatic recess of the maxilla; 4, the anterior maxillary sinus wall; 5, premaxillary tissue. (From Harvey RJ, Sheahan PO, Schlosser RJ. Surgical management of benign sinonasal masses. Otolaryngol Clin North Am 2009;42(2):353–75; with permission.) Surgery may include turbinectomies, septectomy, middle meatal antrostomy (MMA), frontal, sphenoid, ethmoid surgery. Basic endoscopic sinus surgery instrumentation is required (see Fig. 2B). Zone 2: tumor extends to involve  Maxillary sinus medial to the inferior orbital nerve (ION)  Limited posterior wall or  Maxillary floor. MMA or modified endoscopic medial maxillectomy10,11 (MMM) is needed for tumor surveillance. Sinus surgery to include MMA  MMM and sphenopalatine artery management and some angled instrumentation is needed (see Fig. 2C). Zone 3: tumor involves  Maxilla lateral to ION and up to the zygomatic recess  Nasolacrimal duct or medial buttress may need resection. Surgery may require dacrocystorhinostomy (DCR), possible trans-septal approach, trephine, total medial maxillectomy (TMM)12 or maxillotomy,13 medial buttress removal. Traditional open approaches are described for tumors in this location (sublabial Caldwell-Luc type approach, open lateral rhinotomy, and midface degloving). Angled instrumentation is mandatory for ipsilateral surgery (see Fig. 2D). Zone 4: Tumor involves  Anterior maxillary wall without extension into premaxillary soft tissue. 617 618 Harvey et al Table 3 Surgical resection zones MUSC Zone Anatomic Region Surgery Techniques Instrumentation Zone 1 Tumor limited to septum, turbinates, middle meatus, ethmoid, frontal, sphenoid sinuses, medial orbital wall (inverted papilloma, hemangioma, chondroma) Surgery includes turbinectomy and septectomy Basic endoscopic sinus surgery instrumentation Zone 2 Tumor extends to involve maxillary sinus medial to the inferior orbital nerve (ION), limited posterior wall or maxillary floor (inverted papilloma, juvenile nasopharyngeal angiofibroma) Middle meatal antrostomy, frontal recess surgery (Draf 1–3, trephine, or osteoplastic), sphenoid, ethmoid surgery. Some sinus surgery to include sphenopalatine artery management or modified endoscopic medial maxillectomy needed for tumor surveillance Angled instrumentation and bipolar diathermy/endoscopic clip applicators. Maxillary trephination may be used. Rongeurs or chisel required for bone removal Zone 3 Tumor involves nasolacrimal duct, medial buttress, or maxilla lateral to ION and up to the zygomatic recess (inverted papilloma, juvenile nasopharyngeal angiofibroma) Requires dacrocystorhinostomy, possible trans-septal approach, possible endoscopic Denker maxillotomy, or open approach (sublabial Caldwell-Luc type approach) Angled instrumentation has limitations in access. Standard endoscopic sinus surgery instruments via trans-septal approach or maxillary trephine may be required Zone 4 Tumor involves anterior maxillary wall with minimal extension into premaxillary soft tissue Surgery requires trans-septal approach, endoscopic Denker maxillotomy or premaxillary endoscopic sinus surgery approach. Sublabial open type approach Open lateral rhinotomy/midface degloving Endoscopic sinus surgery instruments via trans-septal approach or maxillary trephine may be required Angled ipsilateral endoscopic instruments of little utility Zone 5 Tumor involves premaxillary tissue or skin Surgery requires open approach Open surgical instrumentation Extended Endoscopic Surgery Surgery requires trans-septal dissection with direct drilling to the anterior maxillary wall (mucosal side) or 1 of the previously described external approaches (see Fig. 2E). Zone 5: tumor involves premaxillary tissue and/or skin. Surgery requires open approach (see Fig. 2F). Modified medial maxillectomy This technique is widely used to manage access to the maxilla, infratemporal fossa, maxillary artery, and maxillary sinus floor. It is technically the same as that described as a salvage procedure for chronic maxillary sinusitis (Fig. 3).10,11 A modified medial maxillectomy also ensures dependent drainage for a final cavity that may not have normal mucocillary function. In additional, it provides excellent access for postoperative care and surveillance. Trans-septal Access Our current technique involves the creation of a large posterior based septal flap in the contralateral nasal cavity.14 This mucoperichondrial/periosteal flap is pedicled posteriorly on the septal branch of the sphenopalatine artery. The anterior incision commences at the hemitransfixion, or mucosquamous junction (Fig. 4C). The lateral incision starts well lateral on the nasal floor near the inferior meatus (see Fig. 4A). Foreshortening of the flap after elevation occurs and additional width is important for adequate reconstruction. The superomedial incision is made high, under the nasal dorsum (see Fig. 4B). The flap is then raised back to the middle turbinate and reflected between septum and middle turbinate to prevent injury during the subsequent tumor removal (see Fig. 4D). The ipsilateral mucosa over this area is raised as an inverted-U flap with a random blood supply based inferiorly from the nasal floor (see Fig. 4E). Beginning at the head of the inferior turbinate, a window of septal cartilage is removed posteriorly (see Fig. 4F). An area of 1.52 cm is removed to allows the endoscope and instrument to work comfortably through the septum. This approach is ideal for zone 3 or 4 pathologic conditions. Maxillary Trephination The development of specialized instrument sets for maxillary trephination (Fig. 5) has greatly assisted the ease with which an additional port for endoscope or instrument can be deployed. Robinson and colleagues15–17 have help to redefine the landmarks for the placement of these trephines. The safest entry point for a canine fossa puncture was where a vertical line drawn through the midpupillary line was bisected by a horizontal line drawn through the floor of the pyriform aperture. The placement of the trephine can assist access to the maxillary floor, retraction for infratemporal tumors, and early access for maxillary artery ligation and control. This is an excellent adjunct to lateral infratemporal fossa or lateral maxillary lesion. However, for those pathologic conditions involving the anterior wall itself, the trephine does not improve surgical access and will come through tumor in its approach. Maxillotomy Endoscopic maxillotomy or endoscopic Denker maxillotomy has been described and can provide similar access to the trans-septal approach.13 This procedure involves the removal of the medial buttress via osteotomies (Fig. 6). A premaxillary plane is raised and the entire medial buttress is removed. The lacrimal apparatus is disrupted as with a total medial maxillectomy. However, the additional bone removal disrupts the 619 620 Harvey et al Fig. 3. The modified medial maxillectomy. Although originally designed as a salvage procedure for recalcitrant inflammatory maxillary sinusitis, modified medial maxillectomy is a quick and simple procedure to enhance access to the maxillary sinus floor, roof, and infratemporal fossa. The stepwise approach: (A) simple antrostomy, (B) excision of turbinate up to natural os, (C) down Kerrison Rongeur divided the medial maxillary wall from os to floor, (D) osteotome runs low on floor to posterior maxillary sinus wall, (E) grasping forceps push the segment posteriorly and perpendicular to the nasal cavity, (F) scissors remove the segment close to the vertical palatine bone (G, H) bipolar diathermy of the posterior turbinate remnant, (I) finished right cavity dramatically improving access to the maxillary sinus, orbital floor, and infratemporal fossa via the posterior wall. anterior superior alveolar nerve, potentially transecting the canine root and may lead to loss of lateral support of the alar cartilage to the piriform aperture. When performed via a direct nasal or endoscopic route, the resulting alar retraction and collapse is never as severe compared with similar lateral rhinotomy approaches but can still occur. The authors prefer the trans-septal approach for most lateral pathologic conditions but maxillotomy is a good option if the medial buttress bone is directly involved in pathology. Extended Endoscopic Surgery Fig. 4. Trans-septal. The trans-septal approach to access right anterior maxillary inverted papilloma: (A) lateral incision in the left nasal floor, (B) high left septal incision under dorsum, (C) anterior incision almost at the mucosquamous junction, (D) large left septal flap reflected between septum and middle turbinate; the septal cartilage is on view, (E) a right mucosal flap based inferiorly, (F) a 1.5  2.0 cm window of septum being removed, starting at the head of the inferior turbinate, (G) access through the septum to the right anterior maxilla, (H) the left nasal cavity after closure. (From Harvey RJ, Sheehan PO, Debnath NI, et al. Transseptal approach for extended endoscopic resections of the maxilla and infratemporal fossa. Am J Rhinol Allergy 2009;23(4):426–32; with permission.) 621 622 Harvey et al Fig. 5. Specialized maxillary trephine kits allow an additional instrument for retraction, endoscopic camera access, or the ability to address the internal maxillary artery before tumor debulking when the nasal cavity is filled with pathology not allowing a simple approach. Frontal Sinus The coronal incision/osteoplastic flap and browline incision/frontal trephine are 2 important open adjuncts that are used to manage frontal sinus extension. Trans-facial incisions are rarely used. Blepharoplasty incision or orbital crease approaches have been described recently.18,19 Lateral frontal sinus and trephination The modified endoscopic Lothrop procedure (MELP, Draf 3, frontal drillout) is an established means of access to the frontal sinus. Its use in inflammatory sinus disease and as access for treatment of cerebrospinal fluid leak and benign neoplasms is well documented.20,21 In the treatment of benign lesions, particularly inverted papilloma, drilling of bone at the site of attachment rather than the use of scraping techniques Fig. 6. Endoscopic maxillotomy is an alternative to trans-septal surgery. (A) A premaxillary plan is elevated anterior to the left maxilla and osteotome used to remove the lateral buttress of the pyriform aperture through to the maxillary sinus. (B) Image-guided surgery pictures to assist orientation to the area being excised. (C) The final left maxillary cavity (1, zygomatic recess; 2, buccal fat covered with periosteum). Extended Endoscopic Surgery may reduce the risk of recurrence.1,22 Therefore, to adequately treat these lesions, access is required such that an angled drill can be used under vision with bone contact. The ability to contact the bone under vision with the head of a 70 diamond burr defines good access.23 Post Draf 3, lateral endoscopic access to the anterior and posterior walls of the frontal sinus is excellent for 95% of anatomy (Timperley D and Harvey RJ, unpublished data, 2010). Access to the orbital roof was limited (10.3  4.6 mm from medial orbital wall). Access to the orbital roof is reliable in the medial orbital quarter only. For a frontal sinus pneumatized beyond the midorbital point, only 10% of lateral orbital roofs were contacted. For lesions between these points, the anterior-posterior distance between the olfactory fossa and the outer periostium of the nasofrontal beak may help to define which lesions are amenable to endoscopic access. Access correlated with this distance between the olfactory fossa and outer periostium of the nasofrontal beak (r 5 0.6, P<.01) (Timperley D and Harvey RJ, unpublished data, 2010). Disease of the frontal sinus is often not accessible for a total resection via a transnasal only approach9 even when a Draf 3 has been performed. Other adjunctive procedures may be necessary. The frontal trephine24 and osteoplastic flap form the basis of achieving additional access. Understanding the need for these in the preoperative assessment is key. They are easy to perform but the need for them should be defined preoperatively and not discovered as unexpectedly necessary during the surgery. Use of magnetic resonance imaging and computed tomography (CT) help in this assessment. Frontal trephine is an excellent adjunct for lesions lateral to the midorbital point.25,26 The formation of a small 1- to 2-cm incision and bone window allows dissection instrument and endoscope to facilitate dissection (Fig. 7). Frontal trephine can also be used to allow an above and below visualization and dissection technique. The midorbital exit point for the supraorbital neurovascular pedicle and awareness of the supratrochlear nerve bundles is important to ensure safe dissection.27 Frontal recess Access is not the only concern. Reconstruction of the frontal recess may be necessary if the pathologic lesion has been removed from within the frontal sinus. A combination of maximal widening of the frontal recess (Draf 2a, b or 3),28 mucosal preservation, and possible sialastic sheet stenting for 7 to 21 days postoperatively may be appropriate in this circumstance. In addition, inadvertent frontal recess obstruction may occur if the surgery is performed adjacent to frontal recess. A Draf 2a28 is routinely performed for most endoscopic resections. This ensures correct localization of the frontal recess, posterior table, and aids postoperative care. Supraorbital Ethmoid Cell The supraorbital ethmoid (SOE) cell presents a unique surgical problem for the treating rhinologist. Any anterior approach (open or endoscopic) will have great difficulty in removing disease from the increasingly narrow orbitocranial cleft of the SOE, formed between the orbit roof and anterior cranial fossa, as dissection proceeds posterior. Instrumentation may simply not fit into this cleft. Even with removal of orbital bone (the medial wall and roof) and ethmoid roof, the cleft of dura and periorbita is still restrictive (Fig. 8). Only a subcranial or frontal craniotomy approach allows elevation of the anterior cranial fossa dura and removal of the superior bone; the disease in this cleft can then be addressed. Identification of disease in this area preoperatively 623 624 Harvey et al Fig. 7. A mucopyocele presenting 9 years after an attempted obliteration. (A) The imageguided pictures of the right frontal/supraorbital mucopyocele. (B) Trans-trephine endoscopic dissection and mobilization of the mass. (C) The mass being removed via a frontal trephine. (D) The final cavity demonstrating the postexcision space between posterior frontal sinus bone and the orbital roof (periorbita only). is important to balance the approach-related morbidity and need for completeness of resection. Dental Roots The adult maxillary sinus pneumatizes below the nasal floor in most adults. The bone between dental roots and sinus mucosa is on average only 2 mm for the second premolar tooth. Significant morbidity can arise from aggressive drilling in this area. Identifying the maxillary dental relationship is important for preoperative counseling. A modified medial maxillectomy (see Fig. 3) facilitates access, postoperative care, and follow-up for pathologic conditions in the maxillary sinus floor. ORIENTATION USING FIXED ANATOMIC LANDMARKS Easy disorientation can occur during open or endoscopic surgery within the complex anatomy of the skull base. However, the anatomy for endoscopic surgeons has its Extended Endoscopic Surgery Fig. 8. SOE cell problem in tumor resection. The SOE cell forms a narrow cleft between orbit (*) and anterior cranial fossa (#). The inverted papilloma (IP) can be seen in this cleft. The CT scan (A) is for reference. The orbital wall (B), anterior cranial fossa (C) and SOE (D) arrangement makes resection and especially drilling challenging in these cases. (From Harvey RJ, Sheahan PO, Schlosser RJ. Surgical management of benign sinonasal masses. Otolaryngol Clin North Am 2009;42(2):353–75; with permission.) foundations in functional endoscopic techniques.29 Uncinectomy and removal of the bulla have little meaning to those removing large bulky pathologic lesions from the paranasal sinus system. Large tumors, such as inverted papilloma or malignancy, may have significantly distorted or destroyed these functional anatomic features. Although it is important to include the natural ostia into any final endoscopic resection cavity, the steps to gain orientation for tumor resection differ from surgery for inflammatory disease. Where landmarks have been removed or altered by a pathologic lesion, the use of fixed anatomic landmarks is required. Discovering fixed anatomy allows safe dissection and completeness of removal. The nasal floor, posterior choana, eustachian tube opening, skull base, sella, and orbital wall are the fixed anatomic features that we seek out during endoscopic surgery. Finding traditional anatomic landmarks around the periphery of a tumor will always be the mainstay of endoscopic orientation. Similarly, the contralateral paranasal sinus anatomy can be used to find key landmarks, such as the sphenoid roof, for small lesions. However, for bulky tumors that span nearly orbit to orbit, these techniques may not be practical. Discovery of the maxillary sinus leads to location of the orbital floor (maxillary sinus roof) and finding the sphenoid sinus allows identification of the skull base (sphenoid sinus roof). However, significant tumor bulk can sit between these 2 key landmarks and prevent quick progress (Fig. 9). Image-guided surgery can greatly enhance our confidence and orientation in this situation.30 But image-guided surgery is an accessory not always available, accurate, or reliable. During endoscopic surgery, we follow a structured approach to the identification of fixed landmarks to allow quick and easy orientation in relation to the skull base. 625 626 Harvey et al Fig. 9. Finding normal anatomy and fixed anatomic landmarks, such as the sphenoid roof, is important to ensure surgery progresses quickly and safely. The maxillary sinus roof (or orbital floor) intersects at approximately 50% the height of the sphenoid anterior wall (A, B). A stepwise approach is followed: floor of nose and inferior turbinate, posterior choana and eustachian tube orifice, maxillary sinus roof (orbital floor) and posterior wall, and then the medial orbital wall. The next group of landmarks are in the posterior skull base, superior turbinate (defining the lateral boundary of the olfactory cleft), skull base (sphenoid roof to posterior frontal table), and a clear view of the orbital axis (optic nerve to lamina papyracea) (Fig. 10). The superior turbinate serves as a key landmark in endoscopic sinus surgery.31,32 However, when the superior turbinate is not available, previously resected or replaced by a pathologic lesion, transitioning from the anterior group to the posterior group of landmarks can be challenging. Superior dissection can potentially damage the olfactory fossa or posterior ethmoid roof. The use of the orbital floor and orbital axis as a fixed landmark is of great value in skull base surgery.33 ‘‘Stay below or at the level of the orbital floor as dissection proceeds posteriorly and one will avoid the skull base’’ (see Fig. 9). When bulky disease fills the operative area, it can assist debulking of tumor and further posterior discovery of a safe entry to the sphenoid, thus allowing identification of the skull base. Fig. 10. Defining the right orbital axis. The ability to see or localize the entire length of the axis is essential to orientation in the skull base. Orbital morbidity is significantly low if the surgeon is able to complete this task (A, B). Extended Endoscopic Surgery Using the nasal floor as a reference, the parallel line extending from the maxillary sinus roof (orbital floor) allows safe entry to the sphenoid sinus. This rule allows a safe route of entry into the sphenoid when all other anatomic features have been distorted. Once the sphenoid roof is located, the remainder of the skull base can be identified by working from posterior to anterior. The medial orbital floor was also noted to approximate to 40% of the sphenoid height. There was approximately 14 mm and no less than 10 mm between this landmark and carotid, optic nerve, ethmoid roof, and anterior ethmoidal artery.34 From radiological study, there seems to be a mean vertical distance of 11.0  2.9 mm to sphenoid roof, correlating to a maxillary roof line intersecting the anterior sphenoid face at 52  13% of its height,35 and direct distance of at least 10 mm from the orbital floor34 to critical anatomy. This distance encompasses the bite size of many commonly used surgical instruments. Orlandi and colleagues32 acknowledged that perforation of the basal lamella at the level of the maxillary sinus roof is a safe maneuver in proceeding to posterior ethmoidectomy. Defining the highest maxillary sinus roof point allows easier identification of the transition to the medial orbital wall. The authors’ alternative guides to the skull base, medial orbital wall, and sphenoid sinus are described in Table 4. VASCULAR CONTROL FOR EXTENDED PROCEDURES Vascular control is arguably the most common reason for incomplete resection, and not just for endoscopic cases. Endoscopic resection of benign and malignant tumors of the nasal fossae, paranasal sinuses, anterior skull base, and beyond requires good access and a dry surgical field. Poor hemostasis can lead to imprecise removal of tumor, increased difficulty in recognizing the most important anatomic landmarks and identifying the sinus outflow pathways. Poor hemostasis enhances the risks of Table 4 Commonly used characteristics or guides for finding fixed anatomy in endoscopic surgery Endoscopic Anatomy Identifying Feature Skull base Different color (white) No evidence of translucency (ie, air cell seen through it) Follow the back wall of the frontal sinus posterior Follow the sphenoid roof anterior Partitions become broad based on the skull base Palpate behind partitions before removing them to identify level of roof Orbital wall Balloting the eye will cause movement if lamina papyracea has been removed Manipulation of the medial wall creates mass movement (Cohen’s sign) In similar parasaggital plane as natural ostium of maxillary sinus Define the junction of roof to vertical medial wall Color change (yellow or off-white) Sphenoid sinus Ostium medial and posterior to superior turbinate Ostium located 12–15 mm superior to posterior choana arch Locate the contralateral sphenoid and remove the intersinus septum Follow the septum posterior and find the ostium or face in the submucoperiosteal layer Removal of the medial and inferior partition of the most posterior ethmoid cell (Bolger’s box) Enter the sphenoid face at or below the level of the orbital floor 627 628 Harvey et al intraoperative complications and postoperative scarring. Most importantly, it leads to an incomplete operation. It is therefore important that surgeons who perform endoscopic tumor resections have a complete understanding of the anatomy and in particular the vascular supply to the nose and paranasal sinuses. The concept of controlling microvasculature in addition to larger artery/arteriolar structures (macrovascular) is an important process to ensure a workable operative field for prolonged endoscopic surgery. A structured approach to the management of both capillary bed and major blood vessels results in complete and thorough tumor resection.36 Monopolar cautery, as a result of possible current dispersion, should not be used within the sphenoidal sinus, on the skull base, or intracranially. Bipolar cautery with endoscopic forceps is preferred or Ligge clip applicator. Microvascular Control Preoperative management of associated infective or inflammatory surrounding mucosa is important. It is common preoperative practice at our institution to give systemic glucocorticosteroids to reduce the obstructive mucosal changes associated with large tumors. This greatly improves the operative field and we believe enhances the return of normal mucosal function. The choice of anesthetic is important. Total intravenous anesthesia with remifentanyl and propofol is associated with better mucosal hemostasis.37,38 Cotton pledgets containing adrenaline 1:1000 are placed in the nasal cavity over the areas of surgical access for 10 minutes before the surgical procedure. The middle turbinate, lateral nasal wall, and septum are infiltrated with 1% naropin with adrenaline 1:100,000. Warm water irrigation has been advocated for hemostasis in nasal mucosa39,40 and frequent saline irrigation is used to control the intraoperative field.41 Reverse Trendelenburg of the operative table to 10 to 30 has a profound effect by decreasing regional mucosal blood flow by 38%42 and reduces dural venous pressure.43–45 Macrovascular Control The blood supply of the nose and paranasal sinuses is from the external (sphenopalatine artery) and internal (anterior and posterior ethmoid arteries) carotid systems. The sphenopalatine artery is the terminal branch of the maxillary artery and provides 90% of the blood supply to the nose and sinuses. It is therefore the key artery that needs to be controlled when resecting tumors. This requires an understanding of the variable branching of the sphenopalatine artery and of the anatomy of the pterygopalatine fossa allowing competent sphenopalatine artery ligation.46 The size of tumor may require dissection of the posterior wall of the maxillary antrum to expose the maxillary artery and enable more lateral vascular control (Fig. 11). A modified medial maxillectomy is usually required for access (see Fig. 3).10,11 In known vascular tumors, such as juvenile nasopharyngeal angiofibromas, this can be achieved preoperatively by angiography and embolization. The anterior and posterior ethmoid arteries are more difficult arteries to approach and identify. The ethmoid arteries are usually within the bone of the ethmoid roof and only 20% can be simply clipped.47 The key to identifying the arteries is to dissect the lamina papyracea to the level of the frontoethmoidal suture line (Fig. 12) and then to gradually dissect posteriorly elevating the periorbita.48,49 The anterior ethmoidal artery can be seen passing medially into the roof of the ethmoid (see Fig. 12). Once mobilized, ligation can then be performed. Further dissection posteriorly will identify the posterior ethmoid artery (about 10–15 mm posterior) much closer to the orbital apex.50,51 Alternative approaches are either via a mini-Lynch incision52 (Fig. 13) or Extended Endoscopic Surgery Fig. 11. Ligation of the right internal maxillary artery via removal of the right posterior maxillary sinus wall before resection. Controlled devascularization of a large pathologic lesion dramatically improves the endoscopic surgical field. transcaruncular incision and dissection within the orbit with the use of the endoscope to identify the arteries. The authors favor a complete endoscopic trans-nasal approach, which has always proved to be successful. The management of vessels related to the cavernous carotid artery and pituitary gland is more difficult. The inferior hypophyseal arteries are the most likely point of bleeding during sellar and pituitary surgery, and it is possible during methodical dissection to identify these arteries and to ligate as required. A recurrent superior hypophyseal artery may lie within the vasculature of the pituitary stalk and injury can result in optic nerve, pituitary, or hypothalamic injury.53,54 Special mention should be given to the management of injury to the cavernous carotid artery during dissection. The internal carotid artery is a robust structure and small bleeding points directly on the artery can be managed potentially with bipolar diathermy, suturing, or even a muscle patch, although the risk of subsequent false aneurysm is high (Fig. 14). Significant injury to the artery is potentially catastrophic and requires immediate packing of the operative site, cessation of surgery, stabilization of the patient, angiography, and the consideration of coiling. 629 630 Harvey et al Fig. 12. Ligation of the left anterior ethmoidal artery. (A) Removal of the medial orbital wall (*periorbita). (B) Removal of the bony anterior ethmoidal artery canal with curette or diamond drill. (C) Mobilization of a long segment. (D) This allows clips or bipolar forceps to control the vessel. In summary, complete tumor resection requires good vascular control. The only way to achieve this is for the surgical team to have a structured approach to how they are going to manage the vascular supply to the tumor and paranasal sinuses. Typically the tumor may need to be resected until the sphenopalatine artery can be identified at which point it should be immediately controlled, which will reduce most of the blood supply to the tumor. If required, dissection can be continued laterally into the pterygopalatine fossa and beyond to control the maxillary artery. Superiorly, when the lamina can be appropriately dissected, the anterior and posterior ethmoidal arteries should be ligated. This enables tumor resection in a relatively bloodless field. RECONSTRUCTION Attention should be paid to restoring nasal physiology as part of any endoscopic resection. Although this may seem pedantic after several hours of tumor removal, a working cavity results in return of function and a more satisfied patient in the Extended Endoscopic Surgery Fig. 13. Endoscopic orbital ligation of the anterior ethmoidal artery. (A) A 1- to 1.5-cm external incision with subperiosteal dissection to the anterior ethmoidal artery orbital exit point (*). (B) Further 10- to 14-mm dissection reveals the posterior ethmoidal artery exit point (#). (C) Final view with both arteries controlled before removal of a large juvenile nasopharyngeal angiofibroma. long-term. Connecting natural drainage pathways with the surgical cavity prevents mucus recirculation. Avoidance of large sump formation, particularly in the maxillary sinus ensures dependent drainage and easy access to saline irrigation to those areas in which mucocillary function may not fully return. Dural reconstruction can be successfully made30 with free grafts or pedicled flaps2,55 and is not discussed here. Managing the Nasolacrimal System The management of the lacrimal system creates distinct problems and concerns for the endoscopic surgeon involved in the treatment of neoplastic sinonasal disease. The lower nasolacrimal duct should simply be removed and reconstructed when disease is adjacent or involves this area. Working around this structure simply decreases visualization and increases the risk of positive margins. Consideration for Fig. 14. False aneurysm formation of the left internal carotid artery. Even with good local control, internal carotid artery injury (from an outside center attempt at sublabial hypophysectomy) can result in subsequent false aneurysm. Indiscriminant packing should be avoided in this situation if possible. (A) Soft tissue and (B) bone window CT images with a previously placed stent that did not prevent false aneurysm formation. 631 632 Harvey et al reconstruction is made when resection of the lacrimal system during surgical approach occurs or from tumor involvement. Need for resection of lacrimal system during surgical approach In patients with tumors in zone 3 and 4, it is often necessary to resect the nasolacrimal duct for the purpose of access and vision. The duct should never be retained and compromise the access and the dissection to avoid postoperative epiphora. There are reliable reconstructive techniques with greater than 95% lacrimal patency should the lower lacrimal system need to be removed.56–58 If the nasolacrimal duct is divided sharply, then reconstruction with marsupialization of the retained distal duct with or without stenting will suffice (Fig. 15). Endoscopic DCR can be used to provide a robust and reliable patency but is rarely warranted. Involvement of the lacrimal system by tumor The surgical approach to the resection of the lacrimal system is dependent on the extent of the involvement of the lacrimal system by the tumor. Fig. 15. Image-guided surgery pictures from a right medial maxillectomy. The membranous right nasolacrimal duct is exposed for 6 to 8 mm in its distal portion and marsupialized to the nasal cavity rather than a formal DCR. Extended Endoscopic Surgery Fig. 16. Inverted papilloma within the right lacrimal sac requires partial or subtotal sac removal. Reconstruction needs to be appropriate for the degree of sac loss. 1. Tumor involves nasolacrimal duct. The nasolacrimal duct can be completely resected along with the tumor and typically no formal reconstruction is required. After the surgical margin is defined, the remaining duct can be simply marsupialized. Should the remaining duct be less than 5 mm then a formal DCR is preferred. 2. Tumor involves the lacrimal sac. A formal endoscopic DCR with wide bone removal should be performed for access. The medial sac wall can be safely sacrificed and DCR completed using standard techniques. Occasionally the lateral sac mucosa needs to be resected and reconstruction can be achieved by the placement of a free mucosal graft (Fig. 16). A 3-mm punch biopsy cutter is used on a Blakesley forcep to preserve the common canniculus opening and mucosa. The remaining lateral sac is removed. A central perforation of the graft is created with the same 3-mm punch biopsy instrument. The graft is placed to provide near complete mucosal apposition preserving the common canniculus. Stenting is used and a Gelfoam donut dressing is used to secure the graft (Fig. 17). 3. Tumor involves the common canalliculus. The entire sac and common canalliculus need to be resected and reconstruction is required by the insertion of a glass Jones tube through the medial canthus (Fig. 18). This is best performed by an ocular plastic surgeon and an endoscopic surgeon. POSTOPERATIVE MANAGEMENT Leaving a large endoscopic cavity without good postoperative control usually results in a wound bed covered with large crusting (a combination of dry blood and mucus) and superficial bacterial colonization. High-volume positive-pressure squeeze bottle irrigations are used from the first postoperative day. The authors find that this improves patient comfort, breathing, and counterintuitively reduces bleeding.39–41 Antibiotics are given for 14 days postoperatively as there is exposed bone and foreign material in the cavity. Packing has a limited role. Extensive tight ribbon gauze is generally not necessary and is uncomfortable for the patient.59 Sialastic sheeting is used routinely to cover the anterior septum, particularly for trans-septal approaches.14 Prolonged endoscopic surgery is associated with significant excoriation and mucosal abrasion to anterior septum even with careful technique. Heavy fibrinous exudate can occur and can cause nasal obstruction, adhesion, and discomfort if not managed; 0.4-mm silastic 633 634 Harvey et al Fig. 17. Cutting out the common cannuliculus with a circular biopsy punch (A) from the right lacrimal sac (B). sheeting secured with a through and through prolene suture assists with this problem. Exposed bone is covered with Gelfoam or SurgiFlo (Johnson & Johnson Medical) (Fig. 19). A gloved Merocel dressing is placed as a middle meatal spacer or in the new cavity created, less for hemostasis and more for preventing extensive crusting. The dressing is secured with a trans-septal prolene suture. The gloved Merocel spacer is removed on day 7. This generally provides a moist environment with a soft clot that can easily be suctioned in the outpatient clinic. Pain management has evolved in the past few years in our experience. Combination acetaminophen and opiates are used initially. The extensive bone exposure often results in a secondary pain phenomenon around postoperative days 5 to 10, similar to that experienced by tonsillectomy patients undergoing secondary healing. Nonsteroidal anti-inflammatory drugs greatly help to reduce this phenomenon. IMPLICATIONS FOR RESEARCH Further research is required on optimal wound healing for large resection cavities created by the extended surgery. Long-term follow-up of sinonasal function greatly assists in defining the role for extended endoscopic resection. Health cost and economic studies are required to demonstrate a cost advantage to managing patients via an endoscopic approach. This type of surgery often involves long operating time but substantial savings in postoperative care, recovery, and inpatient stay. Fig. 18. Jones tubes. Exposure of the left medial canthus (A). The glass Jones tube in situ (B) and the final external position (C). Extended Endoscopic Surgery Fig. 19. A trans-septal approach to a left maxillary inverted papilloma. (A) The attachment site in the left zygomatic recess is drilled directly. (B, C) The large bone exposed cavity is covered with Surgiflo to encourage granulation and prevent heavy crusting. High-volume saline irrigations always accompany this postoperative care. IMPLICATIONS FOR CLINICAL PRACTICE Endoscopic techniques developed from managing inflammatory sinus disease have little relevance in large tumor resections. The focus for these cases should be on appropriate surgical access, early vascular control, and complete excision. Our greater understanding of sinonasal physiology and endoscopic evaluation should allow the creation of a functional cavity even after large resections. A postresection nasal cavity with extensive adhesions, chronic mucosal inflammation, and mucus recirculation should be a relic of days when surgery was performed under a headlight. REFERENCES 1. Harvey RJ, Sheahan PO, Schlosser RJ. Surgical management of benign sinonasal masses. Otolaryngol Clin North Am 2009;42(2):353–75. 2. Harvey RJ, Stamm AC, Vellutini E, et al. Closure of large skull base defects after endoscopic trans-nasal craniotomy. J Neurosurg 2009;111(2):371–9. 3. Lund V, Howard DJ, Wei WI. Endoscopic resection of malignant tumors of the nose and sinuses. Am J Rhinol 2007;21(1):89–94. 4. Danesi G, Panciera DT, Harvey RJ, et al. Juvenile nasopharyngeal angiofibroma: evaluation and surgical management of advanced disease. Otolaryngol Head Neck Surg 2008;138(5):581–6. 5. Weissler MC, Montgomery WW, Turner PA, et al. Inverted papilloma. Ann Otol Rhinol Laryngol 1986;95(3 Pt 1):215–21. 6. Cannady SB, Batra PS, Sautter NB, et al. New staging system for sinonasal inverted papilloma in the endoscopic era. Laryngoscope 2007;117(7):1283–7. 7. Jameson MJ, Kountakis SE. Endoscopic management of extensive inverted papilloma. Am J Rhinol 2005;19(5):446–51. 8. Krouse JH. Endoscopic treatment of inverted papilloma: safety and efficacy. Am J Otolaryngol 2001;22(2):87–99. 9. Woodworth BA, Bhargave GA, Palmer JN, et al. Clinical outcomes of endoscopic and endoscopic-assisted resection of inverted papillomas: a 15-year experience. Am J Rhinol 2007;21(5):591–600 [erratum appears in Am J Rhinol 2008;22(1):97]. 10. Cho DY, Hwang PH. Results of endoscopic maxillary mega-antrostomy in recalcitrant maxillary sinusitis. Am J Rhinol 2008;22(6):658–62. 11. Woodworth BA, Parker RO, Schlosser RJ. Modified endoscopic medial maxillectomy for chronic maxillary sinusitis. Am J Rhinol 2006;20(3):317–9. 635 636 Harvey et al 12. Wormald PJ, Ooi E, van Hasselt CA, et al. Endoscopic removal of sinonasal inverted papilloma including endoscopic medial maxillectomy. Laryngoscope 2003;113(5):867–73. 13. James D, Crockard HA. Surgical access to the base of skull and upper cervical spine by extended maxillotomy. Neurosurgery 1991;29(3):411–6. 14. Harvey RJ, Sheehan PO, Debnath NI, et al. Transseptal approach for extended endoscopic resections of the maxilla and infratemporal fossa. Am J Rhinol Allergy 2009;23(4):426–32. 15. Robinson SR, Baird R, Le T, et al. The incidence of complications after canine fossa puncture performed during endoscopic sinus surgery. Am J Rhinol 2005; 19(2):203–6. 16. Robinson S, Wormald PJ. Patterns of innervation of the anterior maxilla: a cadaver study with relevance to canine fossa puncture of the maxillary sinus. Laryngoscope 2005;115(10):1785–8. 17. Robinson S, Patel N, Wormald PJ. Endoscopic management of benign tumors extending into the infratemporal fossa: a two-surgeon transnasal approach. Laryngoscope 2005;115(10):1818–22. 18. Knipe TA, Gandhi PD, Fleming JC, et al. Transblepharoplasty approach to sequestered disease of the lateral frontal sinus with ophthalmologic manifestations. Am J Rhinol 2007;21(1):100–4. 19. Seiberling K, Floreani S, Robinson S, et al. Endoscopic management of frontal sinus osteomas revisited. Am J Rhinol Allergy 2009;23(3):331–6. 20. Chen C, Selva D, Wormald PJ. Endoscopic modified Lothrop procedure: an alternative for frontal osteoma excision. Rhinology 2004;42(4):239–43. 21. Yoon BN, Batra PS, Citardi MJ, et al. Frontal sinus inverted papilloma: surgical strategy based on the site of attachment. Am J Rhinol Allergy 2009;23: 337–41. 22. Chiu AG, Jackman AH, Antunes MB, et al. Radiographic and histologic analysis of the bone underlying inverted papillomas. Laryngoscope 2006;116(9): 1617–20. 23. Chandra RK, Schlosser R, Kennedy DW. Use of the 70-degree diamond burr in the management of complicated frontal sinus disease. Laryngoscope 2004; 114(2):188–92. 24. Zacharek MA, Fong KJ, Hwang PH. Image-guided frontal trephination: a minimally invasive approach for hard-to-reach frontal sinus disease. Otolaryngol Head Neck Surg 2006;135(4):518–22. 25. Batra PS, Citardi MJ, Lanza DC. Combined endoscopic trephination and endoscopic frontal sinusotomy for management of complex frontal sinus pathology. Am J Rhinol 2005;19(5):435–41. 26. Chiu AG, Vaughan WC. Management of the lateral frontal sinus lesion and the supraorbital cell mucocele. Am J Rhinol 2004;18(2):83–6. 27. Seiberling K, Jardeleza C, Wormald PJ. Minitrephination of the frontal sinus: indications and uses in today’s era of sinus surgery. Am J Rhinol Allergy 2009;23: 229–31. 28. Weber R, Draf W, Kratzsch B, et al. Modern concepts of frontal sinus surgery. Laryngoscope 2001;111(1):137–46. 29. Stammberger HR, Kennedy DW. Paranasal sinuses: anatomic terminology and nomenclature. The Anatomic Terminology Group. Ann Otol Rhinol Laryngol Suppl 1995;167:7–16. 30. Harvey RJ, Smith JE, Wise SK, et al. Intracranial complications before and after endoscopic skull base reconstruction. Am J Rhinol 2008;22(5):516–21. Extended Endoscopic Surgery 31. Bolger WE, Keyes AS, Lanza DC. Use of the superior meatus and superior turbinate in the endoscopic approach to the sphenoid sinus. Otolaryngol Head Neck Surg 1999;120(3):308–13. 32. Orlandi RR, Lanza DC, Bolger WE, et al. The forgotten turbinate: the role of the superior turbinate in endoscopic sinus surgery. Am J Rhinol 1999;13(4): 251–9. 33. Stamm AC, Nogueira JF Jr, Harvey RJ. Revision endoscopic skull base surgery. In: Kountakis SE, Jacobs J, Gosepath J, editors, Revision sinus surgery, vol. 1. Heidelberg (Germany): Springer; 2008. p. 289–300. 34. Casiano RR. A stepwise surgical technique using the medial orbital floor as the key landmark in performing endoscopic sinus surgery. Laryngoscope 2001; 111(6):964–74. 35. Harvey RJ, Shelton BS, Timperley D, et al. Using fixed anatomical landmarks in endoscopic skull base surgery. Am J Rhinol Allergy 2010. [Epub ahead of print]. 36. Harvey RJ, Gallagher RM. Endoscopic vascular control of the anterior skull base. Presented at Annual Scientific Conference. Melbourne, Australia, March 25, 2006. 37. Eberhart LH, Folz BJ, Wulf H, et al. Intravenous anesthesia provides optimal surgical conditions during microscopic and endoscopic sinus surgery. Laryngoscope 2003;113(8):1369–73. 38. Wormald PJ, van Renen G, Perks J, et al. The effect of the total intravenous anesthesia compared with inhalational anesthesia on the surgical field during endoscopic sinus surgery. Am J Rhinol 2005;19(5):514–20. 39. Stangerup SE, Thomsen HK. Histological changes in the nasal mucosa after hotwater irrigation. An animal experimental study. Rhinology 1996;34(1):14–7. 40. Stangerup SE, Dommerby H, Lau T, et al. Hot-water irrigation as a treatment of posterior epistaxis. Rhinology 1996;34(1):18–20. 41. Kassam A, Snyderman CH, Carrau RL, et al. Endoneurosurgical hemostasis techniques: lessons learned from 400 cases. Neurosurg 2005;19(1):E7. 42. Gurr P, Callanan V, Baldwin D. Laser-Doppler blood flowmetry measurement of nasal mucosa blood flow after injection of the greater palatine canal. J Laryngol Otol 1996;110(2):124–8. 43. Iwabuchi T, Sobata E, Suzuki M, et al. Dural sinus pressure as related to neurosurgical positions. Neurosurgery 1983;12(2):203–7. 44. Tankisi A, Cold GE. Optimal reverse Trendelenburg position in patients undergoing craniotomy for cerebral tumors. J Neurosurg 2007;106(2):239–44. 45. Haure P, Cold GE, Hansen TM, et al. The ICP-lowering effect of 10 degrees reverse Trendelenburg position during craniotomy is stable during a 10-minute period. J Neurosurg Anesthesiol 2003;15(4):297–301. 46. Simmen DB, Raghavan U, Briner HR, et al. The anatomy of the sphenopalatine artery for the endoscopic sinus surgeon. Am J Rhinol 2006;20(5):502–5. 47. Floreani SR, Nair SB, Switajewski MC, et al. Endoscopic anterior ethmoidal artery ligation: a cadaver study. Laryngoscope 2006;116(7):1263–7. 48. Camp AA, Dutton JM, Caldarelli DD. Endoscopic transnasal transethmoid ligation of the anterior ethmoid artery. Am J Rhinol Allergy 2009;23(2):200–2. 49. Pletcher SD, Metson R. Endoscopic ligation of the anterior ethmoid artery. Laryngoscope 2007;117(2):378–81. 50. Simmen D, Raghavan U, Briner HR, et al. The surgeon’s view of the anterior ethmoid artery. Clin Otolaryngol 2006;31(3):187–91. 51. Han JK, Becker SS, Bomeli SR, et al. Endoscopic localization of the anterior and posterior ethmoid arteries. Ann Otol Rhinol Laryngol 2008;117(12):931–5. 637 638 Harvey et al 52. Woolford TJ, Jones NS. Endoscopic ligation of anterior ethmoidal artery in treatment of epistaxis. J Laryngol Otol 2000;114(11):858–60. 53. van Overbeeke J, Sekhar L. Microanatomy of the blood supply to the optic nerve. Orbit 2003;22(2):81–8. 54. Krisht AF, Barrow DL, Barnett DW, et al. The microsurgical anatomy of the superior hypophyseal artery. Neurosurgery 1994;35(5):899–903 [discussion: 903]. 55. Harvey RJ, Sheahan PO, Schlosser RJ. Inferior turbinate pedicle flap for endoscopic skull base defect repair. Am J Rhinol Allergy 2009;23(5):522–6. 56. Wormald PJ. Powered endoscopic dacryocystorhinostomy. Otolaryngol Clin North Am 2006;39(3):539–49. 57. Mann BS, Wormald PJ, Mann BS, et al. Endoscopic assessment of the dacryocystorhinostomy ostium after endoscopic surgery. Laryngoscope 2006;116(7): 1172–4. 58. Tsirbas A, Davis G, Wormald PJ, et al. Revision dacryocystorhinostomy: a comparison of endoscopic and external techniques. Am J Rhinol 2005;19(3): 322–5. 59. von Schoenberg M, Robinson P, Ryan R. Nasal packing after routine nasal surgery–is it justified? J Laryngol Otol 1993;107(10):902–5.