INTRODUCTION PATIENTS AND METHODS. Patients. Surgery

Transcription

1 ORIGINAL ARTICLE Endonasal endoscopic surgery for squamous cell carcinoma of the sinonasal cavities and skull base: Oncologic outcomes based on treatment strategy and tumor etiology John R. de Almeida, MD, MSc, 1 Shirley Y. Su, MBBS, 2 Maria Koutourousiou, MD, 3 Francisco Vaz Guimaraes Filho, MD, 3 Juan C. Fernandez Miranda, MD, 3 Eric W. Wang, MD, 4 Paul A. Gardner, MD, 3 Carl H. Snyderman, MD, MBA 4* 1 Department of Otolaryngology Head and Neck Surgery, Princess Margaret Hospital, Toronto, Canada, 2 Department of Otolaryngology Head and Neck Surgery, MD Anderson Cancer Center, Houston, Texas, 3 Department of Neurosurgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, 4 Department of Otolaryngology Head and Neck Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania. Accepted 28 April 2014 Published online 11 July 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI /hed ABSTRACT: Background. Oncologic outcomes for sinonasal and skull base squamous cell carcinoma (SCC) treated with an endoscopic endonasal approach (EEA) needs investigation. Methods. Patients with SCC treated with EEA were stratified by treatment strategy and tumor etiology and reviewed. Results. Thirty-four patients were treated with EEA, or which 27 had definitive resection and 7 had debulking surgery. In the definitive group, 17 had de novo tumors and 10 had tumors arising from inverted papilloma. Definitive resection was associated with better 5-year diseasefree survival (DFS) and overall survival (OS) than debulking (62% vs 17%; p 5.02; and 78% vs 30%; p 5.03). Patients with de novo tumors had similar 5-year DFS and OS to those arising from inverted papilloma (62% vs 62%; p 5.75; and 75% vs 86%; p 5.24). Conclusion. Definitive resection of sinonasal SCC with EEA provides sound oncologic outcomes. SCC arising from inverted papilloma does not have prognostic significance. VC 2014 Wiley Periodicals, Inc. Head Neck 37: , 2015 KEY WORDS: endoscopic, squamous cell carcinoma, sinonasal, inverted papilloma, debulking, oncologic INTRODUCTION Sinonasal malignancies of the head and neck are rare tumors comprising only 3% of all head and neck cancers. 1 Squamous cell carcinoma (SCC) is the most common of these tumors, 2 arising either de novo or in association with inverted papillomas in approximately 10% of cases. 3 Of these, 7.1% of malignancies arise synchronous to the inverted papilloma and 3.6% arise metachronously. Studies have shown a 5-year overall survival (OS) of sinonasal SCC ranging from 43% to 59%. 4,5 There is currently no evidence to suggest a survival difference between de novo carcinomas and those arising from inverted papilloma. 6 Over the last decade, several centers have explored the role of the endoscopic endonasal approach (EEA) for resection of malignant tumors The goals of treatment with the EEA are not limited to curative resection. Other goals may include symptom palliation and debulking before definitive chemoradiotherapy. 9 The debulking approach has been shown in selected cases to improve symptoms, such as visual loss, diplopia, and headache, despite subtotal tumor resection. 13 *Corresponding author: C. H. Snyderman, Department of Otolaryngology Head and Neck Surgery, Center for Cranial Base Surgery, Eye and Ear Institute, University of Pittsburgh Medical Center, 203 Lothrop Street, Suite 500, Pittsburgh, PA snydermanch@upmc.edu This work was presented at the 23rd Annual North American Skull Base Society Meeting in Miami, Florida, February 15, The purpose of the present study was to evaluate the oncologic outcomes of endoscopic endonasal resection of sinonasal SCC and to better elucidate survival differences in patients treated with definitive resection compared to those who underwent debulking surgery, as well as differences between those who have de novo carcinomas compared to those arising from inverted papilloma. PATIENTS AND METHODS Patients Patients undergoing surgery at the University of Pittsburgh Medical Center for sinonasal SCC were reviewed from 2000 to Institutional review board ethics approval was obtained and, due to the retrospective nature of the study, patient consent was assumed. Patients were eligible for the study if they primarily underwent an EEA. Those who underwent simultaneous adjunctive open approaches, either transcranial or transfacial, were included as long as the primary approach was endoscopic. Patients undergoing primary surgery, salvage surgery, and debulking surgery for definitive biopsy or symptom palliation were all included in the analysis. Surgery The goals of surgery varied for each patient depending on the biological behavior and extent of tumor. Patients underwent either definitive resection with curative intent HEAD & NECK DOI /HED AUGUST

2 DE ALMEIDA ET AL. or debulking surgery. Debulking surgery was performed for symptom palliation or to reduce the extent of radiation fields. The goal of definitive resection was to achieve negative surgical margins. The surgical technique varied based on the location of the tumor and the proximity to vital structures. In general, for definitive resection, gross tumor was debulked in a piecemeal fashion within the sinonasal cavity. Tumors involving the skull base were devascularized by ligation of the anterior and posterior ethmoid arteries. The lamina papyracea was resected for tumors extending to the lateral nasal wall, and periorbita resected for tumors with bony erosion. Tumors infiltrating the orbital fat were treated with orbital exenteration. The bony skull base (cribriform plate and fovea ethmoidalis) was resected for tumors involving the bony skull base and dural resections were performed for tumors with skull base erosion. Frontal lobe brain tissue was resected for tumors with limited brain involvement either endoscopically or through a frontal craniotomy when needed for negative margins. Tumors involving the floor of the nose, hard palate, or maxillary sinus were treated with a maxillectomy in addition to the endoscopic approach through a transoral or lateral rhinotomy approach. Lateral extension of tumors into the pterygopalatine fossa or infratemporal fossa was approached endoscopically through a transpterygoid approach. Frozen section margins were assessed intraoperatively for definitive resections and in some debulking cases where negative margins were needed to spare radiation to vital structures (eg, optic nerve). Reconstruction of anterior skull base defects was achieved using a 2-layer closure of the dura and skull base. Duraplasty was performed using Duragen, cadaveric pericardium, or autologous fascia lata with an inlay technique. Skull base reconstruction was performed using either nasoseptal flaps or pericranial flaps as an onlay. Fibrin glue or dural sealant was placed over the reconstruction. Nasal packing and/or Foley balloon catheters were placed depending on the extent and location of the reconstruction. Outcomes Outcomes pertaining to patient information, histopathologic information, extent of tumor involvement, extent of surgery, hospital course, complications, and oncologic outcomes were retrospectively reviewed. Staging of tumors was based on the TNM staging system of the American Joint Committee on Cancer (2010) for maxillary sinus, or ethmoid sinus/nasal cavity cancers. Cancers of the sphenoid sinus were staged based on the staging system for ethmoid cancers as there is no currently wellaccepted staging system for this location. TABLE 1. Variables Demographic, pathologic, and prior treatment information. Total (N 5 34) Age, mean (SD) 57 (13) Sex, no. (%) Male 20 (59) Female 14 (41) Site, no. (%) Ethmoid 15(44) Sphenoid 7 (21) Nasal cavity 7 (21) Maxillary 5 (15) T classification, no. (%) 1 2 (6) 2 3 (9) 3 4 (12) 4a 11 (32) 4b 14 (41) N classification, no. (%) N0 34 (100) N1 0 (0) Etiology De novo SCC 21 (62) Inverting papilloma 13 (38) Synchronous 7 (54) Metachronous 5 (38) Unknown 1 (8) Disease status at presentation Primary 27 (79) Recurrent 6 (18) Residual 1 (3) Abbreviation: SCC, squamous cell carcinoma. Analysis Patients were categorized into those undergoing definitive resection with curative intent versus those undergoing debulking surgery. Those with definitive resections were further subclassified into tumors originating from inverted papillomas or de novo tumors. Statistical analysis was performed using SPSS version 20 (Chicago, IL). Oncologic outcomes were assessed using Kaplan Meier methods. Patient subgroups were compared using the log-rank test. A Cox regression model was performed with age (<60 or 60), sex, positive margins, and tumors arising from inverted papillomas to determine independent predictors of locoregional control, disease-free survival (DFS), and OS. RESULTS Patients A total of 34 patients were treated for SCC of the sinonasal cavity (Table 1). The epicenter of the majority of tumors was in the nasal cavity or ethmoid sinuses (65%). The majority of patients (85%) had advanced tumors (either T3 or T4). All 4 patients (12%) with T3 disease were staged as such because of cribriform involvement. Of the 11 patients (32%) who were staged with T4a disease, 2 had orbital involvement, 4 had involvement of the pterygoid plates, 1 had frontal sinus involvement, 3 had sphenoid involvement, and 1 had involvement of the soft tissues of the face as well as the nasal bones. Of the 14 patients (41%) who were staged with T4b disease, 6 had either clivus or nasopharyngeal involvement, 6 had dural involvement (1 with concomitant orbital apex and 1 with concomitant olfactory nerve involvement), and 2 had brain involvement. Of the patients undergoing definitive resection, 2 patients had T1, 3 patients had T2, 3 patients had T3, 9 patients had T4a, and 10 patients had T4b 1164 HEAD & NECK DOI /HED AUGUST 2015

3 ENDONASAL ENDOSCOPIC SURGERY FOR SCC OF THE SINONASAL CAVITIES AND SKULL BASE TABLE 2. Treatment details. Treatment before presentation No. of patients (%) Length of stay, d Complication rate, no. (%) Endoscopic surgery 8 (24) N/A N/A Open surgery (transfacial approach) 1 (3) N/A N/A Open surgery (transcranial approach) 1 (3) N/A N/A Previous radiotherapy 8 (24) N/A N/A Previous chemotherapy 9 (26) N/A N/A Surgical approach Purely endoscopic 25 (74) (28)* Combined endoscopic and transcranial 3 (9) Combined endoscopic and transfacial 6 (18) 4 3 (50) Neck dissection 0 (0) n/a n/a Adjuvant radiotherapy 24 (71) n/a n/a Adjuvant chemotherapy 12 (35) n/a n/a Abbreviation: N/A, not applicable. * Seven patients had the following complications: cerebrospinal fluid (CSF) leak in 4 patients, 2 had meningitis, 1 had brain abscess/subdural empyema, 1 had orbital hematoma, 1 had epistaxis, and 1 had a carotid rupture. Three patients had complications including 3 CSF leaks and 1 also had meningitis. stage. Of those undergoing debulking, 1 patient had T3, 2 had T4a, and 4 had T4b stage. Thirteen patients (38%) had carcinomas arising from inverted papillomas. Of these patients, 7 had synchronous malignancies, 5 had metachronous malignancies, and 1 was unknown. Five of 7 patients (71%) with synchronous tumors had T4 disease compared with only 2 of 5 patients (40%) with metachronous tumors. Treatment All patients were treated with primarily an endoscopic approach (Table 2). Nine patients (26%) had adjunctive open approaches, including orbital exenteration for tumors invading the orbital fat in 3 patients (9%), infrastructural maxillectomy for tumors invading the palate in 3 patients (9%), and frontal craniotomy for tumors with anterior invasion not amenable to endoscopic resection in 3 patients (9%). Twenty-seven patients (79%) were treated with curative intent. Seven patients (21%) were treated with debulking surgery for symptom palliation (2 patients with intractable facial pain) or in order to reduce the tumor volume at the time of biopsy before definitive chemoradiotherapy (5 patients). Both patients had partial relief of facial pain and pressure. Twenty-four patients (71%) underwent adjuvant radiotherapy and 12 (35%) underwent adjuvant chemotherapy. Four patients did not require radiotherapy because of early-stage disease, 1 patient died from respiratory failure before adjuvant treatment, and the remainder underwent surgery for salvage after radiation failure. Perioperative outcomes and complications The mean hospital length of stay for the cohort was 4.7 days (SD, 3.9). The length of stay for the definitive and the debulking groups were 4.4 days (SD, 2.8) and 5.7 days (SD, 6.6), respectively (p 5.10). One patient in the debulking group had a length of stay of 20 days after suffering postoperative epistaxis and a deep vein thrombosis. Ten patients (29%) had at least 1 surgical complication (Table 3); 2 in the debulking group (29%) and 8 in the definitive resection group (30%). Cerebrospinal fluid (CSF) leaks were experienced by 6 patients (18%). Of these, 3 had pericranial flap reconstruction, 2 had endoscopic repairs with either mucosal or fat grafts, and 2 resolved with lumbar drains. One patient suffered an intraoperative internal carotid artery hemorrhage while dissecting the tumor from the artery at the foramen lacerum. Control of the bleeding was achieved using aneurysm clips both distally and proximally. The patient did not suffer any neurological compromise postoperatively. Three patients suffered from meningitis, 2 of which had CSF leaks and 1 who did not, and were treated with intravenous antibiotics. One patient suffered a subdural empyema after radionecrosis of the frontal craniotomy and a hardware infection. This patient had surgical drainage of the infection, removal of hardware, and a temporoparietal flap reconstruction of the defect. The 2 patients with epistaxis were managed surgically. Oncologic outcomes Of the patients treated with definitive resection, 5 (19%) had positive margins. One had a positive margin at TABLE 3. Complications. Complication No. (%) Surgical complication Internal carotid artery hemorrhage 1 (3) CSF leak 7 (21) Meningitis 3 (9) Subdural empyema/hardware infection 1 (3)* Pneumocephalus 1 (3) Rhinosinusitis 2 (6) Epistaxis 2 (6) Orbital hematoma 1 (3) Nasocutaneous fistula 1 (3) Mucocele 1 (3) Medical complication Deep vein thrombosis 2 (6) Postoperative rapid atrial fibrillation 1 (3) Abbreviation: CSF, cerebrospinal fluid. * Subdural empyema as a result of osteoradionecrosis and titanium mesh infection. Bleeding from sphenopalatine artery. HEAD & NECK DOI /HED AUGUST

4 DE ALMEIDA ET AL. FIGURE 1. Patterns of recurrences. FIGURE 3. Locoregional control by treatment strategy. the planum sphenoidale dura, 1 at V2 in Meckel s cave, 1 at the proximal olfactory nerve, 1 at the middle cranial fossa dura, and 1 at the pterygoid plates and infratemporal fossa. A total of 13 patients developed recurrent disease (Figure 1). The 9 patients who had local recurrences were treated with stereotactic radiosurgery (4), palliative chemotherapy (1), bifrontal craniotomy and transcranial surgery (1), combined medial maxillectomy and endoscopic resection followed by stereotactic radiosurgery (1), endoscopic surgery and chemoradiotherapy (1), and palliative care (1). All other recurrences were treated with chemotherapy. The mean length of follow-up was 33 months (range, months). At last follow-up, 8 patients had died; 7 of their disease and 1 with no evidence of disease. Actuarial 5-year OS for the definitive resection group was significantly better than for the debulking surgery group (78% vs 30%; p 5.03) (Figure 2). Five-year locoregional control for the definitive resection group was 62% compared to 42% for the debulking surgery group (p 5.19) (Figure 3). Five-year DFS for the definitive resection group was significantly better than for the debulking surgery group (62% vs 17%; p 5.02) (Figure 4). In patients who had definitive resections with positive margins, the oncologic outcomes were significantly worse. Locoregional control and DFS in patients with negative margins was 74% at 5 years compared to 0% at 5 years in those with positive margins (p <.001). Overall, 5-year survival was 93% in those with negative margins compared to 0% in those with positive margins (p <.001). Patients who had definitive resections for tumors arising from inverted papilloma did not do better than those who had de novo tumors. Five-year OS was 86% for those arising from inverted papilloma compared to 75% for those arising de novo (p 5.24) (Figure 5). Five-year locoregional control was 62% for both groups (p 5.75) (Figure 6) and 5-year DFS was also 62% for both groups (p 5.58) (Figure 7). Patients with tumors arising from inverted papilloma were further subdivided into those who presented synchronously with their inverted papilloma (n 5 6) and those who presented metachronously (n 5 3). The 5-year survival was 100% in those with synchronous compared to 67% in those with metachronous tumors. The 5-year locoregional control and DFS were 60% and 67%, respectively. FIGURE 2. Overall survival by treatment strategy. FIGURE 4. Disease-free survival by treatment strategy HEAD & NECK DOI /HED AUGUST 2015

5 ENDONASAL ENDOSCOPIC SURGERY FOR SCC OF THE SINONASAL CAVITIES AND SKULL BASE FIGURE 5. Overall survival by tumor etiology. FIGURE 7. Disease-free survival by tumor etiology. In a multivariable model, none of age, sex, margin status, or tumors arising from inverted papillomas were predictive of OS. However, in similar multivariable models, only positive margins were predictive of both poorer locoregional control (p 5.003) and DFS (p 5.003). DISCUSSION SCC of the sinonasal cavity and skull base have traditionally been managed with open surgical approaches, commonly including combined transfacial and/or transcranial approaches for an en bloc resection of the tumor. 5,14 Advances in endoscopic techniques have made many sinonasal malignancies amenable to surgical resection with negative margins. Although the traditional tenet of en bloc resection is not achievable with endoscopic resection, studies have shown that piecemeal resection can have comparable oncologic outcomes. 14 Potential advantages of EEA include improved visualization of the tumor, avoidance of skin incisions, and the avoidance of brain retraction. With open surgical approaches and adjuvant therapy, sinonasal SCC have a 5-year OS ranging from 43% to FIGURE 6. Locoregional control by tumor etiology. 59% (Table 4). 4,5,15 18 Despite a similar surgical paradigm over the past several decades, there have been improvements in the OS associated with these tumors, likely because of improvements in surgical technique and adjuvant therapy. Very few reports exist regarding endoscopic resection of sinonasal SCC. 1,8,19 The existing literature has promising results with survival rates reported as high as 91%. These studies have typically included early T classification tumors. The single largest reported study of EEA for sinonasal malignancy reported 25 SCCs of which 28% were T4 tumors with a 5-year disease-specific survival of 61%. In the present series, 70% of patients had T4 tumors with a corresponding 5-year OS of 78% and 5-year DFS of 62%. The role of the EEA at our institution includes debulking surgery for symptom palliation or to minimize the radiation fields to vital structures before treatment with definitive chemoradiotherapy. The oncologic outcomes in these patients are poor relative to those undergoing definitive resection. This is likely because of the innate challenges in treating unresectable disease with nonsurgical modalities. In 2 patients who underwent debulking surgery for symptom palliation, both patients experienced some subjective improvement in their facial pain. A previous small study investigated the role of debulking surgery for symptom palliation and demonstrated symptom improvement in 4 patients with improvement in diplopia in 3, and facial numbness in Similar improvements in diplopia have been reported with patients treated with induction chemotherapy followed by concurrent chemoradiotherapy in a small series of patients with sinonasal undifferentiated carcinoma with diplopia from mechanical pressure on the orbit. 20 However, high-dose irradiation of the optic nerves, cranial nerves, or brainstem can have potentially devastating complications. In these cases, surgery may have a role, although further study is needed. In this study, no difference in oncologic outcomes was apparent for tumors arising de novo versus those arising from inverted papilloma. These results are corroborated by a recent study that pooled 76 cases of SCC arising from inverted papilloma and demonstrated a 5-year survival rate of 61%. 6 The authors concluded that these HEAD & NECK DOI /HED AUGUST

6 DE ALMEIDA ET AL. TABLE 4. Summary of previous studies. Study Number treated Approach T classification Locoregional control DFS Open approaches Patel Transcranial, transfacial NS NS 49.9% 5-y DFS 53.0% 5-y DSS 44.4% 5-y OS Ganly Transcranial, transfacial NS NS 38.0% 5-y DFS 43.6% 5-y DSS 43.0% 5-y OS Buchmann Transfacial, transcranial, endoscopic NS NS NS McKay Surgery with chemotherapy (23) or radiation NS NS 32% 3-y OS with chemotherapy (7) Lee Surgery alone, surgery 1 XRT, XRT alone, neoadjuvant chemo with surgery and XRT or with XRT alone 16% 5-y OS NS NS 59.5% 5-y OS Mine Transcranial, transfacial 49.5% 5-y OS Endoscopic approaches Schipchandler Endoscopic 36% T4 80% 31-mo LC 91% 31-mo DFS, 91% 31-mo OS Nicolai Endoscopic with or without craniotomy 28% T4 72% 5-y LC 61% 5-y DSS Lund Endoscopic 35% T3 and T4 27% local recurrences Present series 27 Endoscopic with or without transfacial/ transcranial 13% died of disease 62% 5-y LC 62% 5-y DFS 78% 5-y OS Abbreviations: DFS, disease-free survival; NS, not stated; DSS, disease-specific survival; OS, overall survival; XRT, external radiation therapy; LC, local control; LRC, locoregional control. tumors were similar prognostically to de novo tumors. These tumors may often be found in advanced stages, 6,21 but have a significant prognostic advantage if detected at an early stage. 21 In the present study, tumors that arose synchronously with inverted papilloma presented with more advanced disease than those that presented metachronously. This disparity is likely the result of surveillance programs for patients who are initially diagnosed with inverted papilloma and later develop carcinomas. Rigorous follow-up for patients with inverted papilloma may result in earlier detection of carcinomatous degeneration for those tumors with metachronous presentation. This study had limitations that affect its interpretation. It was a retrospective study. Some patients were lost to follow-up as they were referred from other centers for surgery, but received additional treatment and surveillance elsewhere. The sample size of the present study was small and posed limitations particularly for subgroup analysis. For example, in order to detect a survival difference of 20% between tumors arising de novo or from inverted papillomas in a parallel arm study, one requires 320 patients per arm. 22 For example, the debulking surgery cohort consisted of only 7 patients, thus limiting the generalizability of the findings. Surgical techniques have evolved over the duration of the cohort, particularly our institutional philosophy on debulking surgery where the present paradigm is performing a minimal morbidity surgery, with relative short hospital stay and subsequent urgent referral for definitive chemoradiotherapy. CONCLUSIONS Definitive resection of SCC of the sinonasal cavity with endoscopic endonasal surgery provides sound oncologic outcomes comparable to other approaches. Patients who are candidates for primary surgery with complete resection have a better prognosis. SCC arising from inverted papilloma does not have prognostic significance. REFERENCES 1. Lund VJ, Stammberger H, Nicolai P, et al. European position paper on endoscopic management of tumours of the nose, paranasal sinuses and skull base. Rhinol Suppl 2010;22: Bhattacharyya N. Factors predicting survival for cancer of the ethmoid sinus. Am J Rhinol 2002;16: von Buchwald C, Bradley PJ. Risks of malignancy in inverted papilloma of the nose and paranasal sinuses. Curr Opin Otolaryngol Head Neck Surg 2007;15: Ganly I, Patel SG, Singh B, et al. Craniofacial resection for malignant paranasal sinus tumors: report of an international collaborative study. Head Neck 2005;27: Lee CH, Hur DG, Roh HJ, et al. Survival rates of sinonasal squamous cell carcinoma with the new AJCC staging system. Arch Otolaryngol Head Neck Surg 2007;133: Tanvetyanon T, Qin D, Padhya T, Kapoor R, McCaffrey J, Trotti A. Survival outcomes of squamous cell carcinoma arising from sinonasal inverted papilloma: report of 6 cases with systematic review and pooled analysis. Am J Otolaryngol 2009;30: Nicolai P, Battaglia P, Bignami M, et al. Endoscopic surgery for malignant tumors of the sinonasal tract and adjacent skull base: a 10 year experience. Am J Rhinol 2008;22: Luong A, Citardi MJ, Batra PS. Management of sinonasal malignant neoplasms: defining the role of endoscopy. Am J Rhinol Allergy 2010;24: Hanna E, DeMonte F, Ibrahim S, Roberts D, Levine N, Kupferman M. Endoscopic resection of sinonasal cancers with and without craniotomy: oncologic results. Arch Otolaryngol Head Neck Surg 2009;135: Batra PS, Luong A, Kanowitz SJ, et al. Outcomes of minimally invasive endoscopic resection of anterior skull base neoplasms. Laryngoscope 2010; 120: Kim BJ, Kim DW, Kim SW, et al. Endoscopic versus traditional craniofacial resection for patients with sinonasal tumors involving the anterior skull base. Clin Exp Otorhinolaryngol 2008;1: Schipchandler TZ, Batra PS, Citardi MJ, Bolger WE, Lanza DC. Outcomes for endoscopic resection of sinonasal squamous cell carcinoma. Laryngoscope 2005;115: Ketcham AS, Wilkins RH, Vanburen JM, Smith RR. A combined intracranial facial approach to the paranasal sinuses. Am J Surg 1963;106: Wellman BJ, Traynelis VC, McCulloch TM, Funk GF, Menezes AH, Hoffman HT. Midline anterior craniofacial approach for malignancy: 1168 HEAD & NECK DOI /HED AUGUST 2015

7 ENDONASAL ENDOSCOPIC SURGERY FOR SCC OF THE SINONASAL CAVITIES AND SKULL BASE results of en bloc versus piecemeal resections. Skull Base Surg 1999;9: Patel SG, Singh B, Polluri A, et al. Craniofacial surgery for malignant skull base tumors: report of an international collaborative study. Cancer 2003; 98: Buchmann L, Larsen C, Pollack A, Tawfik O, Sykes K, Hoover LA. Endoscopic techniques in resection of anterior skull base/paranasal sinus malignancies. Laryngoscope 2006;116: McKay SP, Shibuya TY, Armstrong WB, et al. Cell carcinoma of the paranasal sinuses and skull base. Am J Otolaryngol 2007;28: Mine S, Saeki N, Horiguchi K, Hanazawa T, Okamoto Y. Craniofacial resection for sinonasal malignant tumors: statistical analysis of surgical outcome over 17 years at a single institution. Skull Base 2011;21: Esposito F, Kelly DF, Vinters HV, DeSalles AA, Sercarz J, Gorgulhos AA. Primary sphenoid sinus neoplasms: a report of four cases with common clinical presentation treated with transsphenoidal surgery and adjuvant therapies. J Neurooncol 2006;76: Parbhu KC, Galler KE, Murphy BA, Pitchford CW, Mawn LA. Primary ocular presentation of sinonasal undifferentiated carcinoma. Opthal Plast Reconstr Surg 2010;26: Kim K, Kim D, Koo Y, et al. Sinonasal carcinoma associated with inverted papilloma: a report of 16 cases. J Craniomaxillofac Surg 2012;40:e125 e Schonfeld D. Statistical considerations for a study where the outcome is a time to failure. Available at: time_to_event/para_time.html. Accessed January 17, HEAD & NECK DOI /HED AUGUST