Surgical Treatment of Nasopharyngeal Malignancies: Role of Endoscopic Endonasal Approaches
Pornthep Kasemsiri1, Daniel M. Prevedello2, Leo Ditzel2, Bradley A. Otto3, Matthew Old3, Ted Teknos3, Enver Ozer3, Amit Agrawal3,
Ricardo L. Carrau3, Amin B. Kassam4
1Department of Otorhinolaryngology, Srinagarind hospital Faculty of Medicine, Khon Kaen University, Khon Kaen, Thailand
2Department of Neurological Surgery, Wexner Medical Center, The Ohio State University, Columbus, OH, USA
3Department of Otolaryngology-Head and Neck Surgery, Wexner Medical Center, The Ohio State University, Columbus, OH, USA
4Department of Neurological Surgery, Aurora Health System, Milwaukee, WI, USA
Corresponding author: Ricardo L. Carrau, Email: Ricardo.Carrau@osumc.edu
Abstract: The development of endoscopic techniques has progressed onward with the recent enrichment of anatomical knowledge and experience, as well as the development of advanced technologies. Expansion of the clinical indications for endoscopic endonasal approaches has produced a paradigm shift in the surgical management of nasopharyngeal malignancies similar to that of other areas of the ventral skull base.
Endoscopic endonasal nasopharyngectomy is a feasible minimal access technique indicated for the resection of select recurrent and radioresistant nasopharyngeal malignancies. It bestows the advantages of lack of external incisions or scars, decreased trauma to adjacent soft tissues and bone, reduced risk of neurological damage, and superior postoperative quality of life.
A comprehensive review of the available literature shows that endoscopic surgery end-results are better than those yielded by re-irradiation or surgery via conventional open approaches. We acknowledge, however, that these results may be influenced by a bias in patient selection. Nonetheless, they suggest that, in properly selected cases, an endoscopic endonasal nasopharyngectomy affords low morbidity rates without compromising oncologic outcomes. Therefore, endoscopic endonasal approaches play a strong role in the treatment of nasopharyngeal malignancies. Nonetheless, one should be fully aware of the indications and limitations of using endoscopic endonasal approach to achieve optimal outcomes.
Keywords: Endoscopic endonasal approaches, Nasopharyngectomy, Malignancy, Nasopharynx
Squamous cell carcinomas account for the majority of malignancies arising involving nasopharynx. Nasopharyngeal carcinomas (NPC) are endemic in Southern China and Southeast Asia, with an incidence of 10-50 per 100,000 populations per year . However, other sinonasal malignancies (e.g. adenoid cystic carcinoma, adenocarcinoma, rhabdomyosarcoma, melanoma, lymphoma, and reticular cell sarcoma) may arise or extend into nasopharynx .
Radiotherapy or concurrent chemoradiotherapy is the mainstay of treatment in primary nasopharyngeal carcinomas , whereas surgery is reserved for persistent or recurrent loco-regional tumor . An overall loco-regional failure after initial treatment is approximately 10%  and could be treated by surgical salvage or re-irradiation. However, several studies have reported that salvage surgery provides better local control with 5-year survival rates between 30% and 52% [5–8]; and, with less morbidity than high-dose re-irradiation [6, 8, 9]. Therefore, surgical resection is an acceptable alternative to the various options of modern re-irradiation techniques (i.e. stereotactic, brachytherapy, proton, or intensity-modulated therapy). Additionally, surgery plays a significant role in the treatment of radioresistant malignancies, such as those of glandular and mesenchymal origin [10, 11]. Primary resection of these radio-resistant tumors seems advantageous .
Nasopharyngeal surgery is challenging due to its deep-seated position in the central part of cranial base and complex anatomical relationships. Access to the nasopharynx may be obtained using a variety of approaches that have been developed to facilitate an adequate resection. Extent and location of the tumor, as well as the skills, training and preferences of the surgical team dictate the indications and choice for any particular approach. In 1983, Fisch  reported a lateral approach through the infratemporal fossa. This approach provides an adequate exposure of the lateral nasopharynx and the petrous internal carotid artery (ICA). However, a significant limitation of this and other lateral approaches is the significant difficulty to extirpate tumor from the contralateral side of nasopharynx. In addition, its associated sequelae include conductive hearing loss and trigeminal nerve dysfunction. Tumor on the posterior wall of the nasopharynx can be addressed via inferior approaches such as the transpalatal and transmandibular approaches. Nonetheless, it is difficult to manipulate instruments through the transpalatal corridor, and exposure of the parapharyngeal space is limited. These obstacles could be circumvented with a mandibular swing technique ; however, this technique is associated with significant morbidity including trismus, dental malocclusion, and cranial nerve injuries. Anterior approaches have been proposed as alternative routes to resect nasopharyngeal tumors. A maxillary swing technique, as described by Wei et al , provides exposure of the entire nasopharynx and its surrounding area; however, aggressive tumors with intracranial or skull base invasion dictate that the approach should be combined with a craniotomy and this is more amenable when using a facial translocation. Unfortunately, both the maxillary swing and the facial translocation are also associated with significant complications including osteoradionecrosis, leakage of CSF, and flap necrosis.
In great part, the evolution toward minimal access and minimally invasive approaches were triggered as an attempt to avoid the morbidity of open approaches. Endonasal approaches were introduced as minimal access or minimally invasive procedures. To et al  used a midface degloving transnasal approach to avoid a facial scar. This technique yielded less morbidity and mortality than conventional open approaches. Roh and Park  applied a microscope with CO2 laser to extirpate three recurrent rT1 tumors via a trans-septal approach. They achieved satisfactory results with no recurrences. An alternative technique surged following the adoption of rod-lens endoscopy. Endoscopic endonasal techniques offer the significant advantage of using the pre-existent air spaces of the sinonasal tract; thus, enabling direct access to the nasopharynx while avoiding external incisions or scars, obviating the translocation of maxillofacial skeleton or the transposition of critical neurovascular structures. In addition, endoscopic visualization provides a magnified image and the possibility to explore around corners with angled lenses. These characteristics allow a better identification, definition and control of tumor margins.
In this manuscript, the current principles and techniques for the endoscopic endonasal resection of malignancies involving nasopharynx will be critically appraised considering the available literature. Surgical outcomes are highly dependent on an appropriate selection of patients, thorough planning, precise surgical technique and use of adjunctive measures; therefore, we present a brief discussion of perioperative preparations and management based on our experience.
Principles of endoscopic endonasal resection of malignancies
The selection of any surgical approach is based mainly on the goals of surgery (i.e. debulking, decompression, resection, wide resection), extent of the tumor to adjacent anatomical areas, its relationship to critical structures, and available methods of reconstruction of the consequent defect. In addition, age, previous surgeries or irradiation, comorbidities and specific needs of the patient must be regarded. All these considerations must be matched against the skills, training and experience of the surgical team.
Endoscopic endonasal approaches address the target lesions via anatomy-based modular approaches that can be categorized according to their orientation in the sagittal or coronal planes. However, isolated endoscopic endonasal approaches for malignant nasopharyngeal tumors are anatomically limited by paramedian critical neurovascular structures (i.e. orbit, optic nerve, ICA); thus, extension beyond these lateral boundaries constitute a contraindication for an endoscopic endonasal approach (or indicate the need to combine endoscopic and external approaches). In addition to this anatomical limitation, an active sinonasal bacterial or fungal infection contraindicates an elective endoscopic endonasal approach, if it requires a transdural resection; thus, any acute infection involving the surgical corridor should be treated prior to surgery that will transgress the meningeal barrier [18, 19].
A nasopharyngectomy aims toward a complete tumor extirpation with negative margins. However, one should consider that, as opposed to traditional open approaches, an endoscopic endonasal resection of a malignancy is rarely completed en bloc. This salient feature of endoscopic endonasal resection has been the subject of controversy; however, a similar dilemma has been previously faced with cancers in other regions of the head and neck including pharyngeal and laryngeal squamous cell carcinoma (i.e. transoral microscopic or robotic resections), and skin cancer (i.e. Moh’s surgery). Select cancers in these areas are currently resected in a piecemeal or layered fashion without jeopardizing outcomes. Similar findings have been reported for the resection of sinonasal malignancies with endoscopic endonasal surgery, suggesting that a piecemeal or layered resection does not compromise the oncologic resection .
All patients must undergo a complete physical examination with emphasis on the sinonasal region, status of cervical lymph nodes, and basic neurologic function (especially cranial nerve status). In addition, distant metastasis should be considered, particularly on the most common sites including lung, liver, and bones. This information is necessary to stage the nasopharyngeal carcinoma, as staging is one of most important factors to determine prognosis and for treatment planning. Endoscopy of the sinonasal region is essential for a detailed assessment of the sinonasal tract and the nasopharyngeal tumor, ascertaining any anatomical variation, absence of active infection, and maybe estimating the vascularity of the tumor.
Contrasted computed tomographic scan (CT) and magnetic resonance imaging (MRI) are complementary investigations to respectively evaluate the bony and soft tissue extensions of the tumor, including intracranial, perineural and vascular involvement, and to suggest its degree of vascularity; therefore, serving as surgical maps. Tumors that involve or are in direct contact with critical neurovascular structures are best evaluated with a CT angiography (CTA). CTA is useful for surgical planning and for image fusion with MRI during intraoperative navigation.
Samples for histological analysis and final diagnosis should be obtained before the definitive surgery. One should note, however, that the inflammation and bleeding resulting from a biopsy may alter the tumor appearance in the MRI; thus, we prefer completing the imaging first, whenever possible. In select patients, a biopsy may be obtained in the office setting. Nonetheless, sometimes this cannot be accomplished due to an uncooperative patient, a highly vascular tumor, or if ample tissue sampling is required for special histological or immunohistological studies (e.g. lymphoma). These patients require sampling in the operating room, often under general anesthesia. However, patients with significant comorbidities that increase the surgical risk deserve special consideration, as a second surgery maybe ill advised. A single stage surgery with intraoperative histological confirmation followed by definitive resection may be the most prudent approach in this clinical scenario.
Due to the risk of significant or catastrophic bleeding (e.g., ICA injury), the patient is typed and cross-matched for packed red blood cells (2-4 units). Following the same premise, insertion of large bore intravenous lines and an arterial line for continuous monitoring of the blood pressure are highly desirable. The surgical team should communicate this and other pertinent issues that may have an impact on the type and preparation for anesthesia; e.g. airway considerations, estimated length of surgery, positioning, need for cranial nerve monitoring (i.e. need for electromyography contraindicates the use of paralyzing agents). Broad-spectrum prophylactic perioperative antibiotics (third or fourth generation cephalosporin with cerebrospinal fluid penetration) are administered just before the surgery and continued through the second postoperative day (nasal packing may require an extended course of antibiotics; i.e. until it is removed).
Following orotracheal intubation, the endotracheal tube is fixed toward the left side and the patient is placed in a 3-pin head holder with the neck tilted to the left and turned to the right and the head on a neutral flexion-extension position. Although a 3-pin head holder system is not critical, it does provide a firm stabilization of the head that is advantageous during drilling. Pin fixation is also superior, and strongly advocated, in prolonged surgeries where decubitus ulcers of the scalp may be a consideration; and, to secure the head when paralysis is not possible due to the need to monitor EMGs of specific cranial nerves.
Application of nasal topical decongestants/vasoconstrictors, such as 0.05% oxymetazoline or epinephrine 1/10,000-1/20,000, aids in the visualization and hemostasis. Injection of a solution of lidocaine (0.5-1%) and epinephrine (1/50,000. -1/200,00) complements the topical application. Povidone solution is applied to the perinasal and periumbilical regions (in the event that an autologous fat free graft is required for reconstruction). If the patient presents a risk for ICA injury, or if the surgeon anticipates the need for a free fascia lata graft, the lateral thigh is also prepped (i.e. to harvest a muscle patch).
We advocate electrophysiologic monitoring for patients presenting with tumors that are in intimate contact with cranial nerves or major neurovascular structures. Somatosensory evoked potentials (SSEPs) identify early signs of brain compromise due to ischemia, edema, contusion or hemorrhage; hence, it is a useful adjunct in case of ICA injury. In addition, electromyography of pertinent cranial nerves and muscles should be considered when the resection of tumor extends into the retrobulbar orbit, superior orbital fissure, or cavernous sinus (CNs III-VI) or when extending to the hypoglossal canal (CN XII), jugular foramen or parapharyngeal space (CNs IX-XI).
A 00 rod-lens endoscope provides adequate and undistorted visualization of the surgical field and ensures that straight instrumentation is adequate during the dissection (the corridor line of sight matches the alignment of the instruments). Therefore, we favor the visualization provided by a 00 rod lens endoscope coupled to a high-definition endoscopic camera and monitor for most of the surgery. Conversely, endoscopes with angled lenses allow looking around the corners or bypass obstacles that may interfere with visualization when using a 00 lens. However, angled lenses are more difficult to use due to their distorted view and because they require angled instruments to match their line of sight. Angled lenses are also use to corroborate the absence of residual tumor confirming adequate margins of resection.
Due to the proximity of the lens to the surgical field and dissection instruments, the lens is frequently soiled. Manual irrigation with warm normal saline solution (preferred) or an irrigation-suction cleansing device may be used to clean the lens; thus, maintaining visualization.
Endoscopic endonasal surgical technique
Approach and Resection
Intuitively, the extent of the endoscopic endonasal approach and resection matches the tumor location and extensions. A small and localized lesion on the central nasopharynx may be resected via a transnasal corridor. This ranges from a direct transnasal route with or without lateralization of the inferior turbinates or an enhanced corridor that includes the removal of the posterior part of inferior turbinate and a limited infero-posterior septectomy (to facilitate bilateral instrumentation). This corridor offers the possibility to resect the posterior and superior nasopharyngeal walls including the bony floor of the sphenoid sinus. Furthermore, the resection can extend antero-superiorly to include the anterior wall and the sphenoid sinus or inferiorly to remove the lower third of the clivus or even the central anterior arch of C1 (similar to the technique used for an endonasal endoscopic odontoidectomy).
Involvement of one or both tori tubarius is common; thus, resection of the medial cartilaginous portion of the affected Eustachian tube is often required. To achieve this resection, the mucoperiosteum of the posterolateral wall of the nasal cavity (over the ascending process of the palatine bone and medial pterygoid plate) is incised just posterior to the inferior turbinate. This incision placement can be modified according to anterior extent of the tumor. A mucoperiosteal dissection follows this plane posteriorly to expose the medial aspect of the medial pterygoid. This subperiosteal plane may be followed to elevate the soft tissues covering the roof of the nasopharynx and facilitate the transection of the torus.
Lateral involvement of the fossa of Rosenmüller requires a transpterygoid approach including a medial maxillectomy and transposition of the soft tissue contents of the medial pterygopalatine fossa to expose the entire height of the pterygoid process, and to allow the removal of the medial pterygoid plate and its base (up to the level of foramen rotundum). Any anatomical variation requiring a modification of this template is planned according to the preoperative imaging.
Whenever possible, the lateral pterygoid plate is preserved, as it serves as a landmark to mark the position of V3 (just posterior to its base) and the parapharyngeal ICA (posterior to V3); therefore, these three structures align on the sagittal plane. Tumor invasion of the pterygopalatine and/or infratemporal fossa requires a lateral expansion of the transpterygoid corridor. The dimensions of this corridor depend on the geometry and pneumatization of the maxillary sinus and nasal cavities (inherent), and the extension of the tumor (to be followed and exposed by the surgeon). An endoscopic medial maxillectomy extends from the roof of the antrum (inferior orbital wall) to the floor of the nasal cavity, and from the nasolacrimal duct anteriorly to the posterior wall of the antrum (Figure 1A). This exposes the entire height of the posterior wall even when using a 00 rod-lens endoscope. Occasionally, the medial maxillectomy needs to be extended anteriorly to enhance the lateral exposure (i.e. extensive tumor involvement of the infratemporal fossa). This may even require an endoscopic Denker’s approach, which includes the removal of the remaining inferior turbinate, anterior aspect of the inferior meatus, piriform aperture and anterior wall of the maxilla. Exposure of the piriform aperture and anterior maxilla requires a vertical incision just anterior to the head of the inferior turbinate, just on the edge of the aperture (Figure 1B). This edge can be palpated with a blunt dissector to optimize the placement of the incision, which is then carried through the periosteum to enable a subperiosteal exposure of the anterior maxilla (Figure 1C). The medial maxillectomy is then extended anteriorly to remove the piriform aperture and sufficient anterior maxillary wall to expose the entire lateral wall of the antrum.
An anterior extension of the medial maxillectomy may be tailored to the specific visualization need, e.g. confining it to removal of the lateral wall inferior to the nasolacrimal opening. However, a full Denker’s approach includes removing the anteromedial aspect of the ascending process of the maxilla (piriform aperture) (Figure 1D,E), dissecting and transecting the lacrimal duct sharply. This allows visualization that extends to the most lateral aspect of the infratemporal fossa with a straight line of sight.
Figure 1. Malignant tumor involving the lateral nasopharynx and area adjacent to the paraclival ICA,requiring an endoscopic transpterygoid approach. After a total ethmoidectomy, medial maxillectomy, and sphenoidotomies(A), an endoscopic Denker’s approach facilitatesthe lateral dissection. Exposure of the piriform aperture requires a vertical incision just anterior to the head of the inferior turbinate (B). This edge can be palpated with a blunt dissector to determine the best placement for the incision, which is then carried to the boneto facilitatea subperiosteal exposure of the anterior maxilla (C). Removal of theanterior aspect of the inferior meatus, remaining inferior turbinateand anterior wall of the maxilla (D,E). Coagulated sphenopalatine artery is needed before removing the posterior wall of maxillary sinus (F). The periosteum of the posterior wall of the antrum is preserved to safeguard the vascular structures within the pterygopalatine fossa (G). The bone of the palatine canal is removed to free the descending palatine artery and the greater palatine nerve. Sequentially, vidian nerve is sacrificedallowing the lateral displacement of the pterygopalatine fossa soft tissues (H,I). After adequate exposure of the anterior aspect of the pterygoid process, the vidian canal is dissected (J). Once the ICA position is confirmed with Doppler ultrasound and/or intraoperative navigation, the superior vidian canal is completely removed. , Removing of the medial pterygoid plate exposes the cartilaginous Eustachian tube (K). The medial Eustachian tube is transected and removed to expose the fossa of Rosenmüller and the tumor within (L-P). The tumor is removed en bloc or in sequential layers according to its relationship to critical neurovascular structures (Q,R).Finally, an autologous fat free graft is inserted prior to obliterate a significant dead space (S) and then HBF is placed cover the bony defect and ICA (T).
(MS = maxillary sinus; Ant. MS = anterior wall of maxillary sinus; SS = sphenoid sinus; HBF = Hadad-Bassagaisteguy nasoseptal flap; V = vidian nerve; ICA = internal carotid artery; M Pt. = medial pterygoid plate; T = tumor; ET = Eustachian tube; F = fat graft)
After complete exposure of the posterior wall of maxillary sinus (Figure 2A), the mucoperiosteum covering the superior third of the posterior wall of the antrum and the corresponding level over the perpendicular plate of the palatine bone is elevated to identify the sphenopalatine foramen. Identifying the sphenopalatine artery and freeing its trunk from the sphenopalatine foramen (Figure 1F) allows its posterior displacement and the removal of the posterior wall of maxillary sinus. This is best accomplished using a 1-2 mm Kerrison rongeur to remove the anterior aspect of the sphenopalatine foramen and subsequently extending the resection laterally to remove the posterior wall of the antrum; thus, exposing the anterior aspect of the pterygopalatine fossa. Cautious dissection of the medial aspect of the posterior wall of the maxillary sinus is necessary, as it forms part of the descending palatine canal containing the greater palatine nerve and descending palatine artery. This exposure may be extended inferiorly to reach the level of the antral floor, and laterally to reach the inferior orbital fissure or even further to remove the lateral wall of the antrum. The periosteum of the posterior wall of the antrum is displaced and preserved securing the vascular structures within the pterygopalatine fossa (Figure 1G; Figure 2B-D).
Figure 2. Schematic illustration shows the basic steps of an endoscopic transpterygoid nasopharyngectomy (00 endoscopic perspective). The posterior wall of the maxillary sinusis fully exposed after a medial maxillectomy (A). The posterior wallof the maxillary sinus is removed preserving the periosteum of posterior wall (B); thus, protectingvascular structures in the pterygopalatine fossa, including thevenous plexus, internal maxillary artery and its branches (C). The sphenopalatine ganglionand its branches (D) (lesser and greater palatine nerves, vidian nerve)seat behind internal maxillary artery. The vidian neurovascular bundle could be sacrificed to allow the lateral displacement of the soft tissues of the pterygopalatine fossa toexpose the entire height of the pterygoid process (E). Following this exposure, the bone around the vidian canal and foramen rotundum is drilled out. The medial pterygoid plate is removed to expose the cartilaginous part of the Eustachian tube (F). Exposure of the bony Eustachian tube requires the displacement or resection of the temporalis, medial pterygoid, and lateral pterygoid muscles (G). These dissections allow exposure of for a wide resection of the entire fossa of Rosenmüller. The mandibular branch of the trigeminal nerve (CN V3) is identified anterior to the horizontal ICA and under CN V2. The cartilaginous ET could be resected following the identification of the parapharyngeal ICA(H). The infratemporal muscles could be resectedfor adequate exposure to extirpate the tumor with negative margins (I).
(SS = sphenoid sinuses; PWA = posterior wall of antrum; NP = nasopharynx; IMA = internal maxillary artery; PtP = pterygoid plate; SPG = sphenopalatine ganglion; V1= CN V1; V2 = CN V2; V3 = CN V3; VN = vidian nerve; LPM = lateral pterygoid muscle; MPM = medial pterygoid muscle; TM = temporalis muscle; bET = bonyEustachian tube;cET = cartilaginous Eustachian tube; vICA = vertical internal carotid artery; hICA = horizontal internal carotid artery; ppICA = parapharyngeal internal carotid artery)
The bone of the greater and lesser palatine canals is removed until the descending palatine artery and the greater and lesser palatine nerves are free. Subsequently, the vidian neurovascular bundle is transected and the periosteum of the pterygoid process is dissected from its anterior aspect. The soft tissue contents of the pterygopalatine fossa are left contained between the posterior periosteum of the maxilla and the periosteum of the anterior aspect of the pterygoid process. This periosteal sac is then displaced laterally and inferiorly exposing the entire height of the pterygoid process (Figure 1H,I; Figure 2E). Following this exposure, the bone around the vidian canal and foramen rotundum is drilled out in an anterior to posterior fashion (Figure 1J). The mandibular branch of the trigeminal nerve (CN V3) is identified anterior to the petrous ICA and under CN V2.
Following the identification of the foramen lacerum and ICA’s corresponding segment, the bone covering the anterior and inferior aspects of its petrous (horizontal) segment is removed laterally. When distal control of the ICA is necessary, the bony canal over the paraclival carotid (vertical segment) is also removed. Dissection of the tumor along the ICA is performed with great care, confirming the position and trajectory of the vessel with both the intraoperative navigation and acoustic Doppler sonography. When the tumor involves any of the segments of the ICA, we control the common, external and internal carotid arteries proximally through a transcervical approach. Furthermore, if the surgery requires exposure or control of the parapharyngeal ICA, we dissect the vessel distally, as close to the skull base as possible, through the transcervical approach. We then insert a sheet of blue plastic, or a thin silicone sheet, or cottonoids to mark and protect the parapharyngeal ICA during the transnasal resection. These are removed after the transnasal resection is completed.
Access to the entire fossa of Rosenmüller requires the removal of the medial pterygoid plate (Figure 1K; Figure 2F) and medial Eustachian tube (Figure 1L-P; Figure 2H). For exposure of the bony Eustachian tube, the temporalis, medial pterygoid, and lateral pterygoid muscles could be displaced (Figure 2G) or removed (if involved by tumor or requiring an expanded lateral exposure) (Figure 2I). The tumor may be removed en bloc or in sequential layers according to its relationship to critical neurovascular structures (Figure 1Q,R).
Reconstruction of the surgical defect, including coverage of the middle cranial and/or posterior cranial fossae and exposed ICAs, must be carefully planned including several alternative options. A contralateral Hadad-Bassagaisteguy flap must be harvested at the beginning of the surgery. According to the surgical needs, it may be stored in its ipsilateral maxillary antrum or sphenoid sinus, or against the ipsilateral lateral nasal wall (preferred). A Caicedo reverse flap is raised and mobilized to cover the septal cartilage donor site, which was left bare after harvesting the Hadad-Bassagaisteguy flap. One recent modification of the reverse flap includes shifting the level of its inferior incision up, so that 5 mm of mucosa remain as an inferiorly based flap. This flap is then rotated toward the contralateral side; thus, covering the exposed bone of the maxillary crest. Silicone splints are sutured transeptal to protect the remaining septal mucosa and reverse flap.
Following histological confirmation of tumor-free surgical margins, the reconstruction recreates the arachnoid layer using an inlay graft of collagen matrix (e.g. DuraGen; Integra LifeSciences Corp., Plainsboro, NJ). The nasoseptal flap is then on-laid, overlapping the defect widely (allowing for some tissue contraction). Nasopore (Stryker Corp; Kalamazoo, MI), or Stammberger Sinu-Foam (Arthrocare-ENT Corp.; Austin, TX), or Merogel (Medtronic Corporation; Jacksonville, FL) creates a non-adherent barrier between the flap and the nasal packing. Occasionally, a free fat graft is inserted to obliterate a significant dead space such as the sphenoid sinus or clival recess (Figure 1S); therefore, enabling an optimal flap positioning and longer reach (Figure 1T). However, if the extent of the critical areas of the defect (i.e. dural effect and protection of ICAs) is larger than the potential coverage of the Hadad-Bassagaisteguy flap, or this flap is not available, reconstruction proceeds with a temporoparietal fascia flap. The temporoparietal fascia flap is harvested via a hemicoronal incision and its passage into the nasal cavity is facilitated by a tunnel that connects the temporal to the infratemporal fossa, and eventually to the nasal cavity via the pterygopalatine fossa [21,22]. The geometry and dimensions of the pterygopalatine fossa, antrum and the soft tissue flap, dictate the dimensions of the tunnel, which requires varying degrees of expansion using soft tissue dissection, bougienage and partial removal of the pterygoid process. Finally, the reconstruction is bolstered in place using sponge packing (if a CSF leak was repaired) or any of the absorbable alternatives that were mentioned previously (if no CSF leak was encountered). An indwelling lumbar spinal drainage is rarely indicated even after a CSF leak repair. We reserve its use for patients who are considered at high risk for increased intra-ventricular pressure and for those with an intraoperative high-flow leak (defined as opening of the third ventricle or at least 2 cisterns). Silicone splints are left in place to protect the nasal septum, if needed.
Upon completion of the surgery, the patient is transferred to a monitored unit that facilitates early detection of complications and provides attention to fluid and metabolic disturbances that can occur during the early postoperative period. In addition, the physician should be alert to others major complications that can occur after skull base surgery, especially significant intracranial hemorrhage or tension pneumocephalus. These early complications may be identified with a postoperative non-contrasted CT scan immediately after surgery (on a stable patient). We also advocate a contrasted MRI within the first 24 hours after surgery to corroborate the completion of the resection. In addition, the uptake of contrast may provide an estimate of the vascularity of the reconstructive flap as well as the adequacy of its positioning.
During the first six postoperative weeks, the patient is evaluated in the clinic setting every 1-2 weeks for debridement of the nasal cavity. Most of the nasal cavity has remucosalized by this time. Subsequent visits are customized to the needs of the patient and to fulfill oncologic monitoring.
Salvage options for recurrent NPC include modern re-irradiation with or without chemotherapy, and surgery. The main limitation of re-irradiation is normal tissue tolerance. Leung et al. reported that 57% of their patients experienced with one or more major complications from a second course of radiation therapy. These included grade III neck fibrosis, deafness, trismus, epistaxis, nasopharyngeal soft tissue ulcer, cranial neuropathies, endocrine dysfunction, and temporal lobe necrosis.
Salvage surgery of residual or recurrent nasopharyngeal carcinoma is a difficult undertaking due to the location of the tumors that are frequently adjacent to critical neurovascular bundles. However, the risk-benefit ratio seems positive, as several retrospective studies have reported that surgical salvage yields a better local control rate (40% to 60%) than fractionated re-irradiation (15% to 35%) . Additionally, surgery plays a primary role in the treatment of radioresistant tumors such as malignancies of glandular and mesenchymal origin.
Several external approaches have been developed to access and adequately resect tumors of the nasopharynx. External surgical approaches do not vary significantly regarding oncologic outcomes. The maxillary swing technique provided a 5-years local control rate of 60% and a 5-years survival rate of 55% . A small series of seven patients undergoing a transcervical-mandibulotomy-palatal approach showed that five of seven cases (71%) showed no evidence of disease during follow up (12 to 96 months) . A transnasal approach through a midfacial degloving allowed a resection with clear margins in 12 of 15 cases (80%) . However, open approaches are associated with significant morbidity and sometimes have limited access to some subunits of the nasopharynx.
Surgical techniques for the resection of nasopharyngeal malignancies have evolved rapidly from traditional open surgery to endoscopic endonasal approach addressing both morbidity and oncologic concerns. It should be noted that endoscopic techniques vary widely and that different techniques have been developed to address different populations. In 2005, Yoshizaki et al. were the first to report an endoscopic trans-septal nasopharyngectomy that allowed the successful resection in four of five patients. Only one patient with a recurrent nasopharyngeal carcinoma with massive parapharyngeal extension could not be rendered tumor free. Chen et al. applied either a diode laser or the harmonic scalpel to extirpate six recurrent T1-2a nasopharyngeal carcinomas, achieving adequate resections (negative margins) in all cases, with no perioperative complications. During a period of follow up of 16-59 months (mean= 29 months), they encountered only one loco-regional recurrence and one death from skull base osteoradionecrosis. Ko et al., subsequently introduced the titanyl-phosphate laser to facilitate an endoscopic resection. Their series included 28 patients with recurrent NPC (12 rT1N0, 14 rT2aN0, 2 rT2aN1). Twenty-five patients underwent surgery alone and three required postoperative adjuvant radiotherapy due to positive margins. Two-year local disease-free survival rates for the 12 patients with rT1 tumor and for 16 patients with rT2a tumor were 100% and 41.7% (p = 0.007); whereas their 2-year overall survival rates were 90.9% and 38.5% (p = 0.03), respectively. Three patients (12%) had obvious osteoradionecrosis of the nasopharynx, and one patient developed hypoglossal nerve dysfunction. They suggested that this technique offers a simple, safe, and effective treatment for rT1 NPC, although its benefits were less for rT2a.
Chen et al.  reported the first large case series of endoscopic nasopharyngectomy including 37 recurrent NPC patients: 17 rT1N0M0, 4 rT2aN0M0, 14 rT2bN0M0, and 2 rT3N0M0). In 35 patients the resection achieved negative surgical margins. In the remaining two cases, the tumors were dissected into small blocks, and one had positive surgical margins. The 2-year overall survival rate, local relapse-free survival rate, and progression-free survival rate were 84.2%, 86.3%, and 82.6%, respectively. A subgroup analysis revealed that the local recurrence rate was 0% for rT1, 25 % (1/4) for rT2a, 28.6% (4/14) for rT2b.
An important factor for local control is the ability to achieve negative margins laterally controlling an involved Eustachian tube or within the parapharyngeal space. Therefore, an extended endoscopic transpterygoid nasopharyngectomy is advantageous to control the lateral extension of tumor. Al-Sheibani et al. reported the endoscopic transpterygoid nasopharyngectomy technique, used to extirpate the twenty nasopharyngeal malignancies of various histologies including nine epidermoid carcinomas, one lymphoepithelioma, five adenoid cystic carcinomas, two adenocarcinomas, two mucoepidermoid carcinomas, and one sarcoma. Ninety-five percent of these cases had negative microscopic margins. The approach includes an endoscopic endonasal extended inferomedial maxillectomy, mobilization of pterygopalatine fossa, removal of the pterygoid process and resection of the medial Eustachian tube to access the posterolateral nasopharynx. No perioperative mortality, cerebrovascular accidents, postoperative CSF leak, postoperative epistaxis, or infectious complications were encountered; however, one case with a stage IV mucoepidermoid carcinoma suffered an ICA injury that required intraoperative sacrifice without permanent sequelae. Early outcomes demonstrated an overall survival of 45 % (9/20) and a local control of 65% (13/20). These outcomes have been reproduced by Castelnuovo et al. , who recently reported their experience with endoscopic nasopharyngeal resection, similarly including patients with advanced stage. Their study included 36 consecutive patients with primary (9 patients) and locally recurrent (27 patients) nasopharyngeal carcinomas, including patients with stage I (44.4%), stage II (8.4%), stage III (41.6%), and stage IVA (5.6%) and a mean follow up period of 38 months. Overall oncologic outcomes at 5 years, survival rate, disease specific, and disease free survival were 75.1±9.13%, 80.9±7.79%, and 58.1±14.8%, respectively. However, the reproducibility and efficacy of endoscopic transpterygoid nasopharyngectomy require validation with larger case series and longer follow up.
Limitations of endoscopic transpterygoid nasopharyngectomy include the difficulty to resect tumor, encasing the parapharyngeal ICA or extending posterior to the ICA (parapharyngeal or petrous segments). These situations are best managed combining the endoscopic approaches with robotic and/or open approaches such as the preauricular or postauricular subtemporal approaches .
Recently, the use of the Da Vinci robot (Intuitive Surgical, Sunnyvale, California) has been suggested as an adjunctive tool to facilitate skull base surgery. It provides three-dimensional (3D) visualization, affords 2 to 3-handed surgery (robotic arms and co-surgeon/assistant’s), and has the advantage of wristed instruments that improve the ability to manipulate the tissue (e.g. dural resection or repair) [31,32]. However, the robot, lacks a drill; therefore, any extension of tumor requiring removal of bone is an insurmountable obstacle for a pure robotic technique. Nonetheless, in select cases, the endoscopic endonasal and transoral robotic surgery are complementary. Yin et al.  first reported the clinical use of an endoscopic endonasal approach combined with a transoral robotic resection of a small recurrence at the superior aspect of the nasopharynx, Subsequently, a combined technique, using transoral robotic surgery and an endoscopic transpterygoid approach, to remove extensive malignant tumors of the posterior skull base, nasopharynx and ITF was developed and applied by Carrau et al. 
Avoidance of complications is one of the great surgical challenges. Endoscopic endonasal techniques are sound and carry low morbidity. Common complications and sequelae include headache, otitis media with effusion, flap necrosis, and osteoradionecrosis; however, major morbidity has been noted. Al-Sheibani et al.  reported two cases that suffered an ICA blowout. One related to radionecrosis of the cervical soft tissues and the other to osteoradionecrosis of the clivus occurred 3 and 56 months after treatment respectively. These events were not directly or temporally associated with the nasopharyngectomy and, as illustrated in one of the cases, this complication may even occur years after treatment. Catastrophic bleeding carries the potential for death; therefore, techniques for the prevention and management of ICA injury or blowout should be mastered. Prevention of injury to the internal carotid artery is based on proper understanding of skull base anatomy from an endoscopic perspective, thorough reviewing of the preoperative imaging, use of adjunctive tools such as image guidance and acoustic Doppler sonography, and cautious dissection. A fundamental tenet is that any injury to the ICA must be controlled while maintaining cerebral perfusion. Therefore, neurophysiologic monitoring is helpful as it reflects cerebral perfusion.
Digital compression of the cervical carotid may diminish its flow, although the effectiveness of this maneuver is variable and its logistics not as simple as it appears (i.e. the two operating surgeons will have their hands occupied controlling the bleeding and is difficult to accommodate a third person around the head of the patient). Management by the anesthesiologist is fundamental and should include adequate and prompt resuscitation with fluids and blood products. Use of hypotensive anesthesia as an attempt to control the bleeding is contraindicated since this results in cerebral hypoperfusion. A key measure that may appear counterintuitive is the administration of heparin to avoid embolic phenomena.
Ultimately, a two-surgeons-four-hands technique with dynamic handling of the endoscope to preserve an adequate view of the surgical field and the use of two suctions offers the best opportunity to identify and control the site of bleeding. Initially the bleeding is directed into the suction tips to maintain visualization while focal pressure is applied with a cottonoid. The endoscope is advanced through the nostril with the least blood flow and visualization is maintained by cleaning the lens with saline solution (by manual irrigation or by a lens cleansing device). Concomitantly, an assistant harvests muscle from the lateral thigh (preferred) or abdomen. Some have advocated harvesting muscle from the temporal area or even the tongue; however, this is difficult to do around the surgeons that are controlling the bleeding, the volume of muscle may be inadequate; and, in the case of the tongue, it will be contaminated with oral flora and may even produce speech deficits. Muscle must be directly applied over the injury to induce hemostasis [35, 36]. It should be noted that a muscle patch may require up to 40-45 min to seal the vessel. Once hemostasis and resuscitation are successful and the patient vital signs and neurophysiologic monitoring are stable, additional packing could be placed to hold the muscle in place and the patient may be transported to the angiography suite for definitive management.
Endovascular sacrifice of the ICA is the most commonly used alternative; however, it is best performed after assessment of collateral blood flow with a balloon occlusion test (i.e. estimates the risk of ischemic stroke and/or the need for bypass). Use of a covered stent  or pipeline may preserve the patency of the ICA; however, they are not available for intracranial use in all institutions. In addition, deployment of a covered stent into the cavernous sinus segment of ICA is technically challenging and requires at least 4-6 weeks of medications that inhibit platelet aggregation. The latter is an important consideration if the tumor needs to be removed or debulked urgently. A follow-up angiography is also recommended after any intraoperative vascular injury. These patients are at risk for a delayed pseudoaneurysm and rupture that can present weeks to years after the event. 
Quality of life (QOL) is an important factor when considering the value of the therapeutic intervention. Ng and Wei  reported that after open maxillary swing approaches, postoperative trismus and oronasal fistulae were common problems that disturbed normal speech, eating, and swallowing function. Severe trismus was associated with a low QOL score. Endoscopic approaches were not associated with any masticatory deficit or impaired swallowing. Furthermore, endoscopic nasopharyngectomy affords a shorter length of hospital stay. Castelnuovo et al. reported that mean hospitalization after an endoscopic approach was 7.2 days (range 2 to 14 days).
Currently, a vascularized flap is used to reline the entire nasopharyngeal defect promoting a complete and expedient functional recovery. Chen et al.  reported that, on the average, the nasopharyngeal wound heals completely in 28 days (range 15 to 56 days) and the donor site wound recovers in 46.5 days (range 24 to 84 days). Currently, however, we reconstruct the donor site with a Caicedo reverse rotational flap or a mucoperiosteal graft  that rapidly promotes wound healing within 1 to 2 postoperative weeks .
Endoscopic approaches seem beneficial and advantageous in adequately selected patients; however, they require specialized technical expertise and significant experience. Large and highly invasive tumors, such as those with encasement of the parapharyngeal ICA or extension of the tumor posterior to the ICA (parapharyngeal or petrous segment) are best controlled using an open approach (an endoscopic approach may be adjunctive). Additionally, an unexpected tumor extension during an endoscopic approach may require conversion to an open procedure; thus, a skull base surgery team should be able to perform endoscopic, endoscopic assisted and open approaches according to the extent of the tumor and patient’s needs.
Extended endonasal endoscopic approaches provide an important addition to the surgical armamentarium for the resection of nasopharyngeal malignancies. A comprehensive review of the available literatures suggests that it yields satisfactory oncologic outcomes and limited morbidity; however, proper selection of cases is critical to the achievement of adequate outcomes.
1. Luo J, Chia KS, Chia SE, Reilly M, Tan CS, Ye W. Secular trends of nasopharyngeal carcinoma incidence in Singapore, Hong Kong and Los Angeles Chinese populations, 1973-1997 [J]. Eur J Epidemiol, 2007, 22(8):513-521.
2. Brown JJ, Fee WE. Surgical resection of the nasopharynx [J]. Oper Tech Otolaryngol-Head Neck Surg, 2010, 21(1):26-34.
3. Lin J-C, Jan J-S, Hsu C-Y, Liang W-M, Jiang R-S, Wang W-Y. Phase III study of concurrent chemoradiotherapy versus radiotherapy alone for advanced nasopharyngeal carcinoma: positive effect on overall and progression-free survival [J]. J Clin Oncol Off J Am Soc Clin Oncol, 2003, 21(4):631-637.
4. Chan JYW, Wei WI. Critical appraisal of maxillary swing approach for nasopharyngeal carcinoma [J]. Expert Opin Ther Targets, 2012, 16 Suppl 1:S111-117.
5. Yu KH, Leung SF, Tung SY, Zee B, Chua DTT, Sze WM, et al. Survival outcome of patients with nasopharyngeal carcinoma with first local failure: a study by the Hong Kong Nasopharyngeal Carcinoma Study Group [J]. Head Neck, 2005, 27(5):397-405.
6. King WW, Ku PK, Mok CO, Teo PM. Nasopharyngectomy in the treatment of recurrent nasopharyngeal carcinoma: a twelve-year experience [J]. Head Neck, 2000, 22(3):215-222.
7. Fee WE Jr, Moir MS, Choi EC, Goffinet D. Nasopharyngectomy for recurrent nasopharyngeal cancer: a 2- to 17-year follow-up [J]. Arch Otolaryngol Head Neck Surg, 2002, 128(3):280-284.
8. Hsu MM, Hong RL, Ting LL, Ko JY, Sheen TS, Lou PJ. Factors affecting the overall survival after salvage surgery in patients with recurrent nasopharyngeal carcinoma at the primary site: experience with 60 cases [J]. Arch Otolaryngol Head Neck Surg, 2001, 127(7):798-802.
9. Suárez C, Rodrigo JP, Rinaldo A, Langendijk JA, Shaha AR, Ferlito A. Current treatment options for recurrent nasopharyngeal cancer [J]. Eur Arch Oto-Rhino-Laryngol Off J Eur Fed Oto-Rhino-Laryngol Soc EUFOS Affil Ger Soc Oto-Rhino-Laryngol - Head Neck Surg, 2010, 267(12):1811-1824.
10. Chen AM, Bucci MK, Quivey JM, Garcia J, Eisele DW, Fu KK. Long-term outcome of patients treated by radiation therapy alone for salivary gland carcinomas [J]. Int J Radiat Oncol Biol Phys, 2006, 66(4):1044-1050.
11. Chen AM, Granchi PJ, Garcia J, Bucci MK, Fu KK, Eisele DW. Local-regional recurrence after surgery without postoperative irradiation for carcinomas of the major salivary glands: implications for adjuvant therapy [J]. Int J Radiat Oncol Biol Phys, 2007, 67(4):982-987.
12. Al-Sheibani S, Zanation AM, Carrau RL, Prevedello DM, Prokopakis EP, McLaughlin N, et al. Endoscopic endonasal transpterygoid nasopharyngectomy [J]. The Laryngoscope, 2011, 121(10):2081-2089.
13. Fisch U. The infratemporal fossa approach for nasopharyngeal tumors [J]. The Laryngoscope, 1983, 93(1):36-44.
14. Morton RP, Liavaag PG, McLean M, Freeman JL. Transcervico-mandibulo-palatal approach for surgical salvage of recurrent nasopharyngeal cancer [J]. Head Neck, 1996, 18(4):352-358.
15. Wei WI, Lam KH, Sham JS. New approach to the nasopharynx: the maxillary swing approach. Head Neck [J], 1991, 13(3):200-207.
16. To EW, Teo PM, Ku PK, Pang PC. Nasopharyngectomy for recurrent nasopharyngeal carcinoma: an innovative transnasal approach through a mid-face deglove incision with stereotactic navigation guidance [J]. Br J Oral Maxillofac Surg, 2001, 39(1):55-62.
17. Roh J-L, Park CI. Transseptal laser resection of recurrent carcinoma confined to the nasopharynx [J]. The Laryngoscope, 2006, 116(5):839-841.
18. Ditzel Filho LF, de Lara D, Prevedello DM, Kasemsiri P, Otto BA, Old M, et al. Expanded Endonasal Approaches to the Anterior Skull Base [J]. Int J Otorhinolaryngol Clin, 2011, 3(3):176-183.
19. Kasemsiri P, Carrau RL, Prevedello DM, Ditzel Filho LF, de Lara D, Otto BA, et al. Indications and limitations of endoscopic skull base surgery [J]. Future Neurol, 2012, 7(3):263-277.
20. Ong YK, Ong YK, Solares CA, Carrau RL, Snyderman CH. New developments in transnasal endoscopic surgery for malignancies of the sinonasal tract and adjacent skull base [J]. Curr Opin Otolaryngol Head Neck Surg, 2010, 18(2):107-113.
21. Fortes FSG, Carrau RL, Snyderman CH, Kassam A, Prevedello D, Vescan A, et al. Transpterygoid transposition of a temporoparietal fascia flap: a new method for skull base reconstruction after endoscopic expanded endonasal approaches [J]. The Laryngoscope, 2007, 117(6):970-976.
22. Kasemsiri P, Solares CA, Carrau RL, Prosser JD, Prevedello DM, Otto BA, et al. Endoscopic endonasal transpterygoid approaches: anatomical landmarks for planning the surgical corridor [J]. The Laryngoscope, 2013, 123(4):811-815.
23. Leung TW, Tung SY, Sze WK, Sze WM, Wong VY, Wong CS, et al. Salvage radiation therapy for locally recurrent nasopharyngeal carcinoma [J]. Int J Radiat Oncol Biol Phys, 2000, 48(5):1331-1338.
24. Chen C, Fee W, Chen J, Chan C, Khong B, Hara W, et al. Salvage Treatment for Locally Recurrent Nasopharyngeal Carcinoma (NPC) [J]. Am J Clin Oncol, 2012.
25. Ng RW-M, Wei WI. Quality of life of patients with recurrent nasopharyngeal carcinoma treated with nasopharyngectomy using the maxillary swing approach [J]. Arch Otolaryngol Head Neck Surg, 2006, 132(3):309-316.
26. Yoshizaki T, Wakisaka N, Murono S, Shimizu Y, Furukawa M. Endoscopic nasopharyngectomy for patients with recurrent nasopharyngeal carcinoma at the primary site [J]. The Laryngoscope, 2005, 115(8):1517-1519.
27. Chen M-K, Lai J-C, Chang C-C, Liu M-T. Minimally invasive endoscopic nasopharyngectomy in the treatment of recurrent T1-2a nasopharyngeal carcinoma [J]. The Laryngoscope, 2007, 117(5):894-896.
28. Ko J-Y, Wang C-P, Ting L-L, Yang T-L, Tan C-T. Endoscopic nasopharyngectomy with potassium-titanyl-phosphate (KTP) laser for early locally recurrent nasopharyngeal carcinoma [J]. Head Neck, 2009, 31(10):1309-1315.
29. Chen M-Y, Wen W-P, Guo X, Yang A-K, Qian C-N, Hua Y-J, et al. Endoscopic nasopharyngectomy for locally recurrent nasopharyngeal carcinoma [J]. The Laryngoscope, 2009, 119(3):516-522.
30. Castelnuovo P, Nicolai P, Turri-Zanoni M, Battaglia P, Bolzoni Villaret A, Gallo S, et al. Endoscopic Endonasal Nasopharyngectomy in Selected Cancers [J]. Otolaryngol--Head Neck Surg Off J Am Acad Otolaryngol-Head Neck Surg, 2013, 149(3):424-430
31. Ozer E, Durmus K, Carrau RL, de Lara D, Ditzel Filho LFS, Prevedello DM, et al. Applications of transoral, transcervical, transnasal, and transpalatal corridors for Robotic surgery of the skull base [J]. The Laryngoscope, 2013, 123(9): 2176-2179
32. Schneider JS, Burgner J, Webster RJ 3rd, Russell PT 3rd. Robotic surgery for the sinuses and skull base: what are the possibilities and what are the obstacles? [J]. Curr Opin Otolaryngol Head Neck Surg, 2013, 21(1):11-16.
33. Yin Tsang RK, Ho WK, Wei WI. Combined transnasal endoscopic and transoral robotic resection of recurrent nasopharyngeal carcinoma [J]. Head Neck, 2012, 34(8):1190-1193.
34. Carrau RL, Prevedello DM, de Lara D, Durmus K, Ozer E. Combined transoral robotic surgery and endoscopic endonasal approach for the resection of extensive malignancies of the skull base [J]. Head Neck, 2013, 35(11): E351-358
35. Valentine R, Wormald P-J. Carotid artery injury after endonasal surgery [J]. Otolaryngol Clin North Am, 2011, 44(5):1059-1079.
36. Valentine R, Wormald P-J. Controlling the surgical field during a large endoscopic vascular injury [J]. The Laryngoscope, 2011, 121(3):562-566.
37. Lippert BM, Ringel K, Stoeter P, Hey O, Mann WJ. Stentgraft-implantation for treatment of internal carotid artery injury during endonasal sinus surgery [J]. Am J Rhinol, 2007, 21(4):520-524.
38. Biswas D, Daudia A, Jones NS, McConachie NS. Profuse epistaxis following sphenoid surgery: a ruptured carotid artery pseudoaneurysm and its management [J]. J Laryngol Otol, 2009,123(6):692-694.
39. Chen M-Y, Wang S-L, Zhu Y-L, Shen G-P, Qiu F, Luo D-H, et al. Use of a posterior pedicle nasal septum and floor mucoperiosteum flap to resurface the nasopharynx after endoscopic nasopharyngectomy for recurrent nasopharyngeal carcinoma [J]. Head Neck, 2012, 34(10):1383-1388.
40. Caicedo-Granados E, Carrau R, Snyderman CH, Prevedello D, Fernandez-Miranda J, Gardner P, et al. Reverse rotation flap for reconstruction of donor site after vascular pedicled nasoseptal flap in skull base surgery [J]. The Laryngoscope, 2010, 120(8):1550-1552.
41. Kasemsiri P, Carrau RL, Otto BA, Tang IP, Prevedello DM, Muto J, et al. Reconstruction of the pedicled nasoseptal flap donor site with a contralateral reverse rotation flap: technical modifications and outcomes [J]. The Laryngoscope, 2013, 123(11):2601-2604.
- There are currently no refbacks.
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Asia Press is a professional Science, Technology and Medicine publisher, who owns rapid publication, Peer-Reviewed, Open Access Journals. Asia Press aims to promote “knowledge sharing”. As you know, the main barrier for free “knowledge sharing” is the cost of publishing and transfer. In order to encourage scholars and scientists to the max, and devote whole power to realize the aim of “knowledge sharing” and the benefit of “all” mankind, Asia Press performs a permanent policy of no charge for publication and access, and always open its door for authors worldwide.
© 2013-2017 by the Asia Press. All rights reserved.
© 2013-2017 by the Asia Press. All rights reserved.