- Email: [email protected]
Endoscopic assisted antral window approach for type III nasopharyngeal angiofibroma with infratemporal fossa extension
International Journal of Pediatric Otorhinolaryngology (2008) 72, 1855—1860
www.elsevier.com/locate/ijporl
Endoscopic assisted antral window approach for type III nasopharyngeal angiofibroma with infratemporal fossa extension Mohamed A. Khalifa, Sameh M. Ragab * Department of Otolaryngology and Head & Neck Surgery, Tanta University Hospitals, Egypt Received 20 June 2008; received in revised form 22 August 2008; accepted 2 September 2008
KEYWORDS Nasopharyngeal angiofibroma; Endoscopic; Midfacial degloving; Embolization
Summary Objectives: To assess the efficacy and safety of endoscopic assisted antral window approach in advanced nasopharyngeal angiofibroma with infratemporal fossa extension. Materials and methods: Sixteen cases diagnosed as juvenile nasopharyngeal angiofibroma type III with infratemporal fossa extension were surgically managed using endoscopic assisted antral window approach (group A) and compared with another group of similar number that were managed using endoscopic assisted midfacial degloving (group B). Inclusion criteria were type III JNA with infratemporal fossa extension and a minimum follow-up of 2 years. Operative time, blood loss, adverse events and recurrences were recorded in all cases. Results: The amount of blood lost in group A was significantly less than group B. The operative time of group A was significantly less than group B. No major complications were seen in both groups. Twenty-eight patients showed complete tumor clearance. Four recurrences were seen: two in group A and two in group B. Conclusion: Endoscopic assisted antral window approach provides a safe, reliable, effective and minimally invasive technique in management of type III JNA with infratemporal fossa extension. Preoperative embolization is a safe measure in the experienced hands that helps to reduce intraoperative blood loss and improves the quality of the surgical field. # 2008 Elsevier Ireland Ltd. All rights reserved.
1. Introduction * Corresponding author at: P.O. Box 66482, Bayan 43755, Kuwait. E-mail address: [email protected] (S.M. Ragab).
Juvenile nasopharyngeal angiofibroma (JNA) is a relatively rare tumor occurring mainly in adolescent
0165-5876/$ — see front matter # 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijporl.2008.09.012
1856 boys. It accounts for 0.5% of all head and neck tumors. It originates from the posterolateral wall of the nasal cavity at the vicinity of the sphenopalatine foramen. Typically, patients present with unilateral nasal obstruction and recurrent epistaxis. As the disease advances facial deformities, proptosis, blindness and cranial nerve palsies may occur. The diagnosis of JNA is based on a careful history, and nasal endoscopic examination, supplemented by imaging studies using computed tomogram (CT) and magnetic resonance imaging (MRI). Biopsies to establish histological diagnosis are contraindicated. Diagnostic angiography helps to localize the tumorfeeding vessels [1,2]. Surgery is the mainstay of treatment JNA. Advanced types of JNA are usually managed with extensive external surgical approaches including lateral rhinotomy, midfacial degloving and craniofacial resection. Though, the risk of bleeding, complications and recurrences were reported to be the highest among these groups [3—5]. Therefore, thorough search for a less invasive surgical approach to control such extensive tumors with less bleeding, morbidities and recurrences has become a major interest for most surgeons. This study was designed to introduce and evaluate the efficacy and safety of endoscopic assisted antral window approach in advanced nasopharyngeal angiofibroma with infratemporal fossa extension.
2. Materials and methods Sixteen cases diagnosed as juvenile nasopharyngeal angiofibroma type III with infratemporal fossa extension were surgically managed using endoscopic assisted antral window approach (group A) in our institution since January 2000. These cases were prospectively followed and compared with the last consecutive 16 cases of JNA with the same type and extension that were managed using an endoscopic assisted midfacial degloving approach (group B) before January 2000. All patients were males with a mean age of 14 3 years and a range of 8—19 years. The protocol of the study and the methods of consent had been approved by the ethics Committee. Fisch [5] staging of JNA, revised by Andrews et al. [6], was used for classification of JNA in this study (Table 1). Inclusion criteria were type III JNA with infratemporal fossa extension and a minimum follow up of 2 years. Contrast enhanced CTscans and MRI were performed in all the cases. Operative time, blood loss, adverse events and recurrences were recorded in all cases. Follow-up of the patients was done monthly in the first 6 months, every 3 months up to 2 years and yearly after that. Nasal and
M.A. Khalifa, S.M. Ragab Table 1
Andrews staging of angiofibroma.
Stage I: Tumor limited to the nasal cavity and nasopharynx Stage II: Extensions into pterygopalatine fossa, maxillary, sphenoid, ethmoid sinuses Stage IIIa: Extensions into orbit or infratemporal fossa without intracranial extensions Stage IIIb: Stage IIIa with small extradural intracranial involvement Stage IVa: Large extradural intracranial or intradural extensions Stage IVb: Involvement of the cavernous sinus, pituitary, or optic chiasm
nasopharyngeal endoscopy was performed during these visits. MRI was obtained routinely after 6 and 18 months or if a recurrence was suspected. Fig. 1 shows MRI of type III JNA with infratemporal fossa extension before and after endoscopic assisted antral window resection.
2.1. Surgical procedures 2.1.1. Group A All patients were performed under general hypotensive anesthesia. Preoperative arteriography with embolization of feeding vessels and tumor bed was done in all cases one day before surgery. After routine preparation as in endoscopic sinus surgery, the uncinate process was removed. The maxillary sinus ostium was identified and widened to obtain a wide exposure of the posterior wall of maxillary sinus. Maxillary sinus ostium was widened up to the orbital floor, inferiorly to the inferior turbinate and posteriorly to posterior wall of the maxillary sinus. An incision was made in the gingivobuccal sulcus and the periosteum over the anterolateral wall of maxillary sinus elevated. The anterolateral and posterior walls of the maxillary sinus were removed. The lateral extension of tumor in the infratemporal fossa was exposed, mobilized from the surrounding soft tissue and pushed medially to be removed transnasally. The procedure was continued endoscopically and the tumor was freed from the ethmoid, sphenoid sinus, medial pterygoid plate, and nasopharynx. Drilling medial pterygoid plate and vertical plate of palatine bone was performed for good exposure and clearance. The tumor was delivered through the nose or pushed down into the pharynx and removed transorally. The rest of the surgical cavity was meticulously examined and cleared using different angled endoscopes. In cases where the lateral extension is very large, the tumor was divided at the sphenopalatine foramen, being the narrowest part, and the lateral extension was delivered through the
Endoscopic assisted antral window approach for type III nasopharyngeal angiofibroma
1857
Fig. 1 MRI of type III JNA with infratemporal fossa extension before and after endoscopic assisted antral window resection.
gingivolabial approach. Internal maxillary artery was identified and coagulated or clipped. At the end of the procedure, a ribbon gauze pack was inserted for 2 days. 2.1.2. Group B The operation was performed under hypotensive general anesthesia. A midfacial degloving approach as described by Casson et al. [7] and revised by Howard and Lund [8] was used in all cases with some modifications. No preoperative embolization was used. The tumor was identified and progressively mobilized from its bed. The posterior wall of the maxillary sinus was removed and the tumor parts in the infratemporal and pterygopalatine fosse were mobilized and removed. The internal maxillary artery was coagulated with bipolar cautery or ligated. Removal of the vertical part of the palatine bone and medial pterygoid plate were performed for good exposure and tumor clearance. The procedure ended by meticulous clearing of the rest of the tumor from ethmoid sinus, sphenoid sinus and remaining parts of the surgical cavity using different angled endoscopes. At the end of surgery, the cavity was bounded posteriorly by the posterior wall of the sphenoid sinus and posterior wall of nasopharynx, superiorly by the cribriform plate and the roof of the ethmoids, and laterally by the coronoid process of the mandible. A ribbon gauze pack was inserted in the cavity for 48 h.
considered significant. Parametric tests such as ttest were applied for data that followed a normal distribution. Nonparametric tests such as Mann— Whitney U test and Chi-squared test were applied for data that did not follow a normal distribution.
3. Results 3.1. Blood loss The amount of blood lost in group A (575 cm3 123) was significantly (P < 0.001, two sample t-test) less than group B (1075 cm3 198), with a median difference of 450 cm3. Fig. 2 shows boxplot of intraoperative blood loss in groups A and B.
2.2. Statistical methods Analysis was done using SPSS for windows statistics software package. Data were expressed as mean standard deviation (S.D.). P-values < 0.05 were
Fig. 2 and B.
Boxplot of intraoperative blood loss in groups A
1858
Fig. 3
M.A. Khalifa, S.M. Ragab
Boxplot of operative time in groups A and B.
3.2. Operative time The operative time of group A (128 min 14) was significantly (P < 0.001, two sample t-test) less than group B (153 min 23), with a median difference of 20 min. Fig. 3 shows boxplot of operative time in groups A and B.
3.3. Adverse events No major complications were seen in both groups. Group A showed some minor adverse events due to preoperative embolization such as trismus, headaches, fever and localized pain. Group B showed some adverse events due to midfacial degloving such as nasal valve stenosis, infraorbital anesthesia and prolonged recovery times.
3.4. Recurrences Twenty-eight patients showed complete tumor clearance. Four recurrences were seen: two in group A and two in group B. Recurrences were discovered endoscopically on the 4th, 7th, 8th, and 9th months. They were detected in the ethmoidal region (one case), sphenoidal region (one case) and base of pterygoids (two cases). All of them had a second surgery and were removed endoscopically.
4. Discussion Surgery is the principal management of juvenile nasopharyngeal angiofibromas. Small type I and type II tumors have been removed via a transpalatal [9],
midfacial degloving [10] or endoscopic approaches [11]. Type IIIa tumors, with infratemporal fossa or orbital involvement but without intracranial involvement, have been classically resected via a combined transpalatal and transmaxillary approaches [12], endoscopic assisted midfacial degloving approach [13] or the infratemporal fossa approach advocated by Fisch [5]. An adjuvant second endoscopic approach through the anterior wall of the maxillary sinus has been reported whenever the lateral extension of the angiofibroma within the pterygomaxillary fossa could not be completely visualized and controlled only by the transnasal endoscopic approach [14]. The surgical resection of more extensive JNAs with intracranial extension (types IIIb and IV) has been accomplished with a variety of extensive techniques, generally involving a combination of extracranial and intracranial approaches [15]. In this study we introduced an endoscopic assisted antral window approach with preoperative embolization and compared it to endoscopic assisted midfacial degloving, regarding intraoperative blood loss, operative time, adverse events and recurrences, in the management of type III JNA with infratemporal fossa extension. Excision of advanced JNA is principally considered a high-risk surgery regarding perioperative hemorrhage and may require blood transfusion, which is not without morbidity. Perioperative hemorrhage would also significantly hinder the resection. In this study, intraoperative blood loss was significantly less in group A than that in group B, with a mean difference of 47% and a median difference of 43%. There was an average blood loss of 575 cm3 in group A and 1075 cm3 in group B. This difference between the two groups could be attributed mainly to preoperative embolization. Other factors may include less operative time and less invasive external incisions. Halving of blood loss in group A would certainly decrease the risk of the surgery, diminish the need of blood transfusion and improve the efficacy of the surgical excision. Consequently, many authors agree that preoperative embolization is a helpful measure in reducing intraoperative blood loss [16,17]. Moulin et al. [16] showed a mean blood loss of 5380 cm3 in nonembolized patients versus 1037.5 cm3 in embolized patients. Tyagi et al. [12] reported a range of 1500 cm3 of blood loss in large angiofibroma using a combined transpalatal and transmaxillary approach with preoperative embolization. El-Banhawy et al. [13] reported an average blood loss of 500 cm3 with embolization and 1000 cm3 without embolization using an endoscopic assisted midfacial degloving approach for type III JNA.
Endoscopic assisted antral window approach for type III nasopharyngeal angiofibroma The duration of surgery is one of the important factors affecting the morbidity of the procedure. Reducing the operative time would diminish the anesthetic risk, blood loss and cost of surgery. In our study, the operative time was significantly less in group A (mean = 128 min) than group B (mean = 153 min). There was a 16% decrease in the mean and 14% decrease in the median operative time when group A was compared to group B. This could be explained with the minimal external surgical incision and reduction of the amount of perioperative bleeding in group A. Despite the importance of operative time, few if any authors have discussed it. A single study discussing 9 JNAs with only one advanced type III JNA reported an operative time of 165 min for such procedure [11]. Discussion of any surgical procedure is essentially worthless without weighing its adverse events. In this study, we did not encounter any major complication in the two groups. Group A showed some minor adverse events due to preoperative embolization such as trismus, headaches, fever and localized pain. Some controversy still exists regarding the use of preoperative embolization in management of head and neck tumors. Opponents of the technique stress on the rare complications such as cerebral infarcts and vision loss which are generally secondary to abnormal collaterals from the internal maxillary artery to the intracranial and intraorbital contents [18]. On the contrary, most of the surgeons agree that preoperative embolization is a safe and effective procedure which should be performed but with an experienced radiologist [11,12,16,19]. Detailed assessment of the possible dangerous collaterals by superselective microcatheterization of the dangerous collaterals should be performed to avoid major complications [20]. Preoperative embolization has also been accused to induce incomplete excision of the tumor through impaired visualization of the tumor limits at surgery [21]. However, we did not face such a problem since the recurrence rate of group A (with embolization) and B (without embolization) was similar. Group B showed some adverse events due to midfacial degloving such as nasal valve stenosis, infraorbital anesthesia and prolonged recovery times. These morbidities have been shown to be common with such technique [3,13,22]. Unusual complications such as oroantral fistula and epiphora have also been reported [23]. The endoscopic assisted antral window approach avoided such morbidities and at the same time showed a comparative efficacy in eradication of the tumor. Recurrence of JNA is classically linked to stage of the disease. Extension to the infratemporal fossa, base of pterygoids, clivus, sphenoid sinus, cavernous sinus and cranial fossa were described to be associated with high recurrences [24,25]. In this study the
1859
recurrence rate for type III JNA with infratemporal extension was 12.5% whether endoscopic assisted antral window approach or endoscopic assisted midfacial degloving was used. We elected to treat them rather than following them up radiologically since recurrences were small and very handy to be removed endoscopically. The reported recurrence rate for types I and II JNA ranged from 0% to 15% [26—28]. The usual relapse rate for type III tumors in most of the literature ranged from 25% to 40% [24,29]. Some authors described better results. El-Banhawy et al. [13] reported 13.5% recurrence using endoscopic assisted midfacial degloving in 15 patients of type III JNA. Tyagi et al. [12] described 13.8% residual tumor in 65 cases of type IIIa JNA using a combined transpalatal and transmaxillary approach. Finally, a question may be raised about the possibility of an exclusive endoscopic approach in type III JNA with infratemporal fossa extension. Although, endoscopic surgery is becoming popular, it still has inherent limitations in far lateral access. The number of total cases of advanced JNA with infratemporal fossa extension operated exclusively endoscopically in literature is very small [11,19,30]. Lateral infratemporal extension of angiofibroma is considered an obstacle for exclusively endoscopic removal of JNA and the patient must be informed that it may be necessary to resort to an open surgery in such types of JNA [30]. Exclusive endoscopic removal of angiofibroma was done by Roger et al. [30] but only one case in their series had an infratemporal fossa extension, in which a residual tumor was left behind. The concept of the exclusive endoscopic approach proponents is that residual disease does not necessarily need a further surgery as it may remain stable or involutes [24]. Second combined surgery is needed only if the tumor remnant progresses or symptoms recur [19,24]. Robinson et al. [31], in a preliminary report of 4 JNAs, used a two-surgeon endoscopic transnasal approach to remove infratemporal extensions of JNA. In their approach, medial maxillectomy, removal of the lateral maxillary wall, ethmoidectomy along with septal dissection in the form of dislocation of the cartilaginous bony junctions, transverse mucosal incision at the side of the tumor and removal of the posterior bony septum were required to remove such angiofibromas. We believe that this is an extensive dissection and may affect facial growth. The antral window in our technique is much less invasive and allows endoscopes to be introduced transnasally and transantrally, gaining benefit from its angles to perfectly resect the tumor. Even in very large JNA, the tumor could be divided at the spheno-palatine foramen with the infratemporal portion being delivered through the sublabial incision and the nasal part through the transnasal route.
1860
M.A. Khalifa, S.M. Ragab
5. Conclusion Endoscopic assisted antral window approach provides a safe reliable effective and minimally invasive technique in management of type III JNA with infratemporal fossa extension. Preoperative embolization is a safe measure in the experienced hands that helps to reduce intraoperative blood loss and improves the quality of the surgical field.
[15]
[16]
[17]
References [18] [1] H. Glad, B. Vainer, C. Buchwald, B.L. Petersen, S.A. Theilgaard, P. Bonvin, et al., Juvenile nasopharyngeal angiofibromas in Denmark 1981—2003: diagnosis, incidence, and treatment, Acta Otolaryngol. 127 (3) (2007) 292—299. [2] A.R. Antonelli, J. Cappiello, D. Di Lorenzo, C.A. Donajo, P. Nicolai, A. Orlandini, Diagnosis, staging, and treatment of juvenile nasopharyngeal angiofibroma (JNA), Laryngoscope 97 (11) (1987) 1319—1325. [3] J.D. Browne, A.H. Messner, Lateral orbital/anterior midfacial degloving approach for nasopharyngeal angiofibromas with cavernous sinus extension, Skull Base Surg. 4 (4) (1994) 232—238. [4] L.G. Close, S.D. Schaefer, B.E. Mickey, S.C. Manning, Surgical management of nasopharyngeal angiofibroma involving the cavernous sinus, Arch. Otolaryngol. Head Neck Surg. 115 (9) (1989) 1091—1095. [5] U. Fisch, The infratemporal fossa approach for nasopharyngeal tumors, Laryngoscope 93 (1) (1983) 36—44. [6] J.C. Andrews, U. Fisch, A. Valavanis, U. Aeppli, M.S. Makek, The surgical management of extensive nasopharyngeal angiofibromas with the infratemporal fossa approach, Laryngoscope 99 (4) (1989) 429—437. [7] P.R. Casson, P.C. Bonanno, J.M. Converse, The midface degloving procedure, Plast. Reconstr. Surg. 53 (1) (1974) 102—103. [8] D.J. Howard, V.J. Lund, The midfacial degloving approach to sinonasal disease, J. Laryngol. Otol. 106 (12) (1992) 1059— 1062. [9] J.G. Spector, Management of juvenile angiofibromata, Laryngoscope 98 (9) (1988) 1016—1026. [10] A.J. Maniglia, Indications and techniques of midfacial degloving: a 15-year experience, Arch. Otolaryngol. Head Neck Surg. 112 (7) (1986) 750—752. [11] L.M. de Brito Macedo Ferreira, E.F. Gomes, J.F. Azevedo, J.R. Souza, A.R. de Paula, A.S. do Nascimento Rios, Endoscopic surgery of nasopharyngeal angiofibroma, Rev. Bras. Otorrinolaringol. (Engl. Ed.) 72 (4) (2006) 475—480. [12] I. Tyagi, R. Syal, A. Goyal, Staging and surgical approaches in large juvenile angiofibroma–—study of 95 cases, Int. J. Pediatr. Otorhinolaryngol. 70 (9) (2006) 1619—1627. [13] O.A. El-Banhawy, S.E.-D. Ael, T. Amer, Endoscopic-assisted midfacial degloving approach for type III juvenile angiofibroma, Int. J. Pediatr. Otorhinolaryngol. 68 (1) (2004) 21—28. [14] E. Pasquini, V. Sciarretta, G. Frank, C. Cantaroni, G.C. Modugno, D. Mazzatenta, et al., Endoscopic treatment of
[19]
[20]
[21] [22]
[23]
[24]
[25] [26]
[27]
[28]
[29]
[30]
[31]
benign tumors of the nose and paranasal sinuses, Otolaryngol. Head Neck Surg. 131 (3) (2004) 180—186. A.H. Marshall, P.J. Bradley, Management dilemmas in the treatment and follow-up of advanced juvenile nasopharyngeal angiofibroma, ORL J. Otorhinolaryngol. Relat. Spec. 68 (5) (2006) 273—278. G. Moulin, C. Chagnaud, R. Gras, E. Gueguen, P. Dessi, J.Y. Gaubert, et al., Juvenile nasopharyngeal angiofibroma: comparison of blood loss during removal in embolized group versus nonembolized group, Cardiovasc. Intervent. Radiol. 18 (3) (1995) 158—161. E. Garcia-Cervigon, S. Bien, D. Rufenacht, C. Thurel, D. Reizine, H.P. Tran Ba, et al., Pre-operative embolization of naso-pharyngeal angiofibromas: report of 58 cases, Neuroradiology 30 (6) (1988) 556—560. A. Casasco, E. Houdart, A. Biondi, H.S. Jhaveri, D. Herbreteau, A. Aymard, et al., Major complications of percutaneous embolization of skull-base tumors, AJNR Am. J. Neuroradiol. 20 (1) (1999) 179—181. T.M. Onerci, O.T. Yucel, O. Ogretmenoglu, Endoscopic surgery in treatment of juvenile nasopharyngeal angiofibroma, Int. J. Pediatr. Otorhinolaryngol. 67 (11) (2003) 1219—1225. M. Onerci, K. Gumus, B. Cil, B. Eldem, A rare complication of embolization in juvenile nasopharyngeal angiofibroma, Int. J. Pediatr. Otorhinolaryngol. 69 (3) (2005) 423—428. G. Lloyd, D. Howard, V.J. Lund, L. Savy, Imaging for juvenile angiofibroma, J. Laryngol. Otol. 114 (9) (2000) 727—730. O.A. El Banhawy, A. Ragab, M.M. El Sharnoby, Surgical resection of type III juvenile angiofibroma without preoperative embolization, Int. J. Pediatr. Otorhinolaryngol. 70 (10) (2006) 1715—1723. D.J. Howard, V.J. Lund, The role of midfacial degloving in modern rhinological practice, J. Laryngol. Otol. 113 (10) (1999) 885—887. P. Herman, G. Lot, R. Chapot, D. Salvan, P.T. Huy, Long-term follow-up of juvenile nasopharyngeal angiofibromas: analysis of recurrences, Laryngoscope 109 (1) (1999) 140—147. I. Tyagi, R. Syal, A. Goyal, Recurrent and residual juvenile angiofibromas, J. Laryngol. Otol. 121 (5) (2007) 460—467. J.J. Fagan, C.H. Snyderman, R.L. Carrau, I.P. Janecka, Nasopharyngeal angiofibromas: selecting a surgical approach, Head Neck 19 (5) (1997) 391—399. D. Radkowski, T. McGill, G.B. Healy, L. Ohlms, D.T. Jones, Angiofibroma: changes in staging and treatment, Arch. Otolaryngol. Head Neck Surg. 122 (2) (1996) 122—129. K. Ungkanont, R.M. Byers, R.S. Weber, D.L. Callender, P.F. Wolf, H. Goepfert, Juvenile nasopharyngeal angiofibroma: an update of therapeutic management, Head Neck 18 (1) (1996) 60—66. M. Jorissen, P. Eloy, P. Rombaux, C. Bachert, J. Daele, Endoscopic sinus surgery for juvenile nasopharyngeal angiofibroma, Acta Otorhinolaryngol. Belg. 54 (2) (2000) 201—219. G. Roger, H.P. Tran Ba, P. Froehlich, A.T. Van Den, J.M. Klossek, E. Serrano, et al., Exclusively endoscopic removal of juvenile nasopharyngeal angiofibroma: trends and limits, Arch. Otolaryngol. Head Neck Surg. 128 (8) (2002) 928—935. S. Robinson, N. Patel, P.J. Wormald, Endoscopic management of benign tumors extending into the infratemporal fossa: a two-surgeon transnasal approach, Laryngoscope 115 (10) (2005) 1818—1822.
Available online at www.sciencedirect.com
www.elsevier.com/locate/ijporl
Endoscopic assisted antral window approach for type III nasopharyngeal angiofibroma with infratemporal fossa extension Mohamed A. Khalifa, Sameh M. Ragab * Department of Otolaryngology and Head & Neck Surgery, Tanta University Hospitals, Egypt Received 20 June 2008; received in revised form 22 August 2008; accepted 2 September 2008
KEYWORDS Nasopharyngeal angiofibroma; Endoscopic; Midfacial degloving; Embolization
Summary Objectives: To assess the efficacy and safety of endoscopic assisted antral window approach in advanced nasopharyngeal angiofibroma with infratemporal fossa extension. Materials and methods: Sixteen cases diagnosed as juvenile nasopharyngeal angiofibroma type III with infratemporal fossa extension were surgically managed using endoscopic assisted antral window approach (group A) and compared with another group of similar number that were managed using endoscopic assisted midfacial degloving (group B). Inclusion criteria were type III JNA with infratemporal fossa extension and a minimum follow-up of 2 years. Operative time, blood loss, adverse events and recurrences were recorded in all cases. Results: The amount of blood lost in group A was significantly less than group B. The operative time of group A was significantly less than group B. No major complications were seen in both groups. Twenty-eight patients showed complete tumor clearance. Four recurrences were seen: two in group A and two in group B. Conclusion: Endoscopic assisted antral window approach provides a safe, reliable, effective and minimally invasive technique in management of type III JNA with infratemporal fossa extension. Preoperative embolization is a safe measure in the experienced hands that helps to reduce intraoperative blood loss and improves the quality of the surgical field. # 2008 Elsevier Ireland Ltd. All rights reserved.
1. Introduction * Corresponding author at: P.O. Box 66482, Bayan 43755, Kuwait. E-mail address: [email protected] (S.M. Ragab).
Juvenile nasopharyngeal angiofibroma (JNA) is a relatively rare tumor occurring mainly in adolescent
0165-5876/$ — see front matter # 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijporl.2008.09.012
1856 boys. It accounts for 0.5% of all head and neck tumors. It originates from the posterolateral wall of the nasal cavity at the vicinity of the sphenopalatine foramen. Typically, patients present with unilateral nasal obstruction and recurrent epistaxis. As the disease advances facial deformities, proptosis, blindness and cranial nerve palsies may occur. The diagnosis of JNA is based on a careful history, and nasal endoscopic examination, supplemented by imaging studies using computed tomogram (CT) and magnetic resonance imaging (MRI). Biopsies to establish histological diagnosis are contraindicated. Diagnostic angiography helps to localize the tumorfeeding vessels [1,2]. Surgery is the mainstay of treatment JNA. Advanced types of JNA are usually managed with extensive external surgical approaches including lateral rhinotomy, midfacial degloving and craniofacial resection. Though, the risk of bleeding, complications and recurrences were reported to be the highest among these groups [3—5]. Therefore, thorough search for a less invasive surgical approach to control such extensive tumors with less bleeding, morbidities and recurrences has become a major interest for most surgeons. This study was designed to introduce and evaluate the efficacy and safety of endoscopic assisted antral window approach in advanced nasopharyngeal angiofibroma with infratemporal fossa extension.
2. Materials and methods Sixteen cases diagnosed as juvenile nasopharyngeal angiofibroma type III with infratemporal fossa extension were surgically managed using endoscopic assisted antral window approach (group A) in our institution since January 2000. These cases were prospectively followed and compared with the last consecutive 16 cases of JNA with the same type and extension that were managed using an endoscopic assisted midfacial degloving approach (group B) before January 2000. All patients were males with a mean age of 14 3 years and a range of 8—19 years. The protocol of the study and the methods of consent had been approved by the ethics Committee. Fisch [5] staging of JNA, revised by Andrews et al. [6], was used for classification of JNA in this study (Table 1). Inclusion criteria were type III JNA with infratemporal fossa extension and a minimum follow up of 2 years. Contrast enhanced CTscans and MRI were performed in all the cases. Operative time, blood loss, adverse events and recurrences were recorded in all cases. Follow-up of the patients was done monthly in the first 6 months, every 3 months up to 2 years and yearly after that. Nasal and
M.A. Khalifa, S.M. Ragab Table 1
Andrews staging of angiofibroma.
Stage I: Tumor limited to the nasal cavity and nasopharynx Stage II: Extensions into pterygopalatine fossa, maxillary, sphenoid, ethmoid sinuses Stage IIIa: Extensions into orbit or infratemporal fossa without intracranial extensions Stage IIIb: Stage IIIa with small extradural intracranial involvement Stage IVa: Large extradural intracranial or intradural extensions Stage IVb: Involvement of the cavernous sinus, pituitary, or optic chiasm
nasopharyngeal endoscopy was performed during these visits. MRI was obtained routinely after 6 and 18 months or if a recurrence was suspected. Fig. 1 shows MRI of type III JNA with infratemporal fossa extension before and after endoscopic assisted antral window resection.
2.1. Surgical procedures 2.1.1. Group A All patients were performed under general hypotensive anesthesia. Preoperative arteriography with embolization of feeding vessels and tumor bed was done in all cases one day before surgery. After routine preparation as in endoscopic sinus surgery, the uncinate process was removed. The maxillary sinus ostium was identified and widened to obtain a wide exposure of the posterior wall of maxillary sinus. Maxillary sinus ostium was widened up to the orbital floor, inferiorly to the inferior turbinate and posteriorly to posterior wall of the maxillary sinus. An incision was made in the gingivobuccal sulcus and the periosteum over the anterolateral wall of maxillary sinus elevated. The anterolateral and posterior walls of the maxillary sinus were removed. The lateral extension of tumor in the infratemporal fossa was exposed, mobilized from the surrounding soft tissue and pushed medially to be removed transnasally. The procedure was continued endoscopically and the tumor was freed from the ethmoid, sphenoid sinus, medial pterygoid plate, and nasopharynx. Drilling medial pterygoid plate and vertical plate of palatine bone was performed for good exposure and clearance. The tumor was delivered through the nose or pushed down into the pharynx and removed transorally. The rest of the surgical cavity was meticulously examined and cleared using different angled endoscopes. In cases where the lateral extension is very large, the tumor was divided at the sphenopalatine foramen, being the narrowest part, and the lateral extension was delivered through the
Endoscopic assisted antral window approach for type III nasopharyngeal angiofibroma
1857
Fig. 1 MRI of type III JNA with infratemporal fossa extension before and after endoscopic assisted antral window resection.
gingivolabial approach. Internal maxillary artery was identified and coagulated or clipped. At the end of the procedure, a ribbon gauze pack was inserted for 2 days. 2.1.2. Group B The operation was performed under hypotensive general anesthesia. A midfacial degloving approach as described by Casson et al. [7] and revised by Howard and Lund [8] was used in all cases with some modifications. No preoperative embolization was used. The tumor was identified and progressively mobilized from its bed. The posterior wall of the maxillary sinus was removed and the tumor parts in the infratemporal and pterygopalatine fosse were mobilized and removed. The internal maxillary artery was coagulated with bipolar cautery or ligated. Removal of the vertical part of the palatine bone and medial pterygoid plate were performed for good exposure and tumor clearance. The procedure ended by meticulous clearing of the rest of the tumor from ethmoid sinus, sphenoid sinus and remaining parts of the surgical cavity using different angled endoscopes. At the end of surgery, the cavity was bounded posteriorly by the posterior wall of the sphenoid sinus and posterior wall of nasopharynx, superiorly by the cribriform plate and the roof of the ethmoids, and laterally by the coronoid process of the mandible. A ribbon gauze pack was inserted in the cavity for 48 h.
considered significant. Parametric tests such as ttest were applied for data that followed a normal distribution. Nonparametric tests such as Mann— Whitney U test and Chi-squared test were applied for data that did not follow a normal distribution.
3. Results 3.1. Blood loss The amount of blood lost in group A (575 cm3 123) was significantly (P < 0.001, two sample t-test) less than group B (1075 cm3 198), with a median difference of 450 cm3. Fig. 2 shows boxplot of intraoperative blood loss in groups A and B.
2.2. Statistical methods Analysis was done using SPSS for windows statistics software package. Data were expressed as mean standard deviation (S.D.). P-values < 0.05 were
Fig. 2 and B.
Boxplot of intraoperative blood loss in groups A
1858
Fig. 3
M.A. Khalifa, S.M. Ragab
Boxplot of operative time in groups A and B.
3.2. Operative time The operative time of group A (128 min 14) was significantly (P < 0.001, two sample t-test) less than group B (153 min 23), with a median difference of 20 min. Fig. 3 shows boxplot of operative time in groups A and B.
3.3. Adverse events No major complications were seen in both groups. Group A showed some minor adverse events due to preoperative embolization such as trismus, headaches, fever and localized pain. Group B showed some adverse events due to midfacial degloving such as nasal valve stenosis, infraorbital anesthesia and prolonged recovery times.
3.4. Recurrences Twenty-eight patients showed complete tumor clearance. Four recurrences were seen: two in group A and two in group B. Recurrences were discovered endoscopically on the 4th, 7th, 8th, and 9th months. They were detected in the ethmoidal region (one case), sphenoidal region (one case) and base of pterygoids (two cases). All of them had a second surgery and were removed endoscopically.
4. Discussion Surgery is the principal management of juvenile nasopharyngeal angiofibromas. Small type I and type II tumors have been removed via a transpalatal [9],
midfacial degloving [10] or endoscopic approaches [11]. Type IIIa tumors, with infratemporal fossa or orbital involvement but without intracranial involvement, have been classically resected via a combined transpalatal and transmaxillary approaches [12], endoscopic assisted midfacial degloving approach [13] or the infratemporal fossa approach advocated by Fisch [5]. An adjuvant second endoscopic approach through the anterior wall of the maxillary sinus has been reported whenever the lateral extension of the angiofibroma within the pterygomaxillary fossa could not be completely visualized and controlled only by the transnasal endoscopic approach [14]. The surgical resection of more extensive JNAs with intracranial extension (types IIIb and IV) has been accomplished with a variety of extensive techniques, generally involving a combination of extracranial and intracranial approaches [15]. In this study we introduced an endoscopic assisted antral window approach with preoperative embolization and compared it to endoscopic assisted midfacial degloving, regarding intraoperative blood loss, operative time, adverse events and recurrences, in the management of type III JNA with infratemporal fossa extension. Excision of advanced JNA is principally considered a high-risk surgery regarding perioperative hemorrhage and may require blood transfusion, which is not without morbidity. Perioperative hemorrhage would also significantly hinder the resection. In this study, intraoperative blood loss was significantly less in group A than that in group B, with a mean difference of 47% and a median difference of 43%. There was an average blood loss of 575 cm3 in group A and 1075 cm3 in group B. This difference between the two groups could be attributed mainly to preoperative embolization. Other factors may include less operative time and less invasive external incisions. Halving of blood loss in group A would certainly decrease the risk of the surgery, diminish the need of blood transfusion and improve the efficacy of the surgical excision. Consequently, many authors agree that preoperative embolization is a helpful measure in reducing intraoperative blood loss [16,17]. Moulin et al. [16] showed a mean blood loss of 5380 cm3 in nonembolized patients versus 1037.5 cm3 in embolized patients. Tyagi et al. [12] reported a range of 1500 cm3 of blood loss in large angiofibroma using a combined transpalatal and transmaxillary approach with preoperative embolization. El-Banhawy et al. [13] reported an average blood loss of 500 cm3 with embolization and 1000 cm3 without embolization using an endoscopic assisted midfacial degloving approach for type III JNA.
Endoscopic assisted antral window approach for type III nasopharyngeal angiofibroma The duration of surgery is one of the important factors affecting the morbidity of the procedure. Reducing the operative time would diminish the anesthetic risk, blood loss and cost of surgery. In our study, the operative time was significantly less in group A (mean = 128 min) than group B (mean = 153 min). There was a 16% decrease in the mean and 14% decrease in the median operative time when group A was compared to group B. This could be explained with the minimal external surgical incision and reduction of the amount of perioperative bleeding in group A. Despite the importance of operative time, few if any authors have discussed it. A single study discussing 9 JNAs with only one advanced type III JNA reported an operative time of 165 min for such procedure [11]. Discussion of any surgical procedure is essentially worthless without weighing its adverse events. In this study, we did not encounter any major complication in the two groups. Group A showed some minor adverse events due to preoperative embolization such as trismus, headaches, fever and localized pain. Some controversy still exists regarding the use of preoperative embolization in management of head and neck tumors. Opponents of the technique stress on the rare complications such as cerebral infarcts and vision loss which are generally secondary to abnormal collaterals from the internal maxillary artery to the intracranial and intraorbital contents [18]. On the contrary, most of the surgeons agree that preoperative embolization is a safe and effective procedure which should be performed but with an experienced radiologist [11,12,16,19]. Detailed assessment of the possible dangerous collaterals by superselective microcatheterization of the dangerous collaterals should be performed to avoid major complications [20]. Preoperative embolization has also been accused to induce incomplete excision of the tumor through impaired visualization of the tumor limits at surgery [21]. However, we did not face such a problem since the recurrence rate of group A (with embolization) and B (without embolization) was similar. Group B showed some adverse events due to midfacial degloving such as nasal valve stenosis, infraorbital anesthesia and prolonged recovery times. These morbidities have been shown to be common with such technique [3,13,22]. Unusual complications such as oroantral fistula and epiphora have also been reported [23]. The endoscopic assisted antral window approach avoided such morbidities and at the same time showed a comparative efficacy in eradication of the tumor. Recurrence of JNA is classically linked to stage of the disease. Extension to the infratemporal fossa, base of pterygoids, clivus, sphenoid sinus, cavernous sinus and cranial fossa were described to be associated with high recurrences [24,25]. In this study the
1859
recurrence rate for type III JNA with infratemporal extension was 12.5% whether endoscopic assisted antral window approach or endoscopic assisted midfacial degloving was used. We elected to treat them rather than following them up radiologically since recurrences were small and very handy to be removed endoscopically. The reported recurrence rate for types I and II JNA ranged from 0% to 15% [26—28]. The usual relapse rate for type III tumors in most of the literature ranged from 25% to 40% [24,29]. Some authors described better results. El-Banhawy et al. [13] reported 13.5% recurrence using endoscopic assisted midfacial degloving in 15 patients of type III JNA. Tyagi et al. [12] described 13.8% residual tumor in 65 cases of type IIIa JNA using a combined transpalatal and transmaxillary approach. Finally, a question may be raised about the possibility of an exclusive endoscopic approach in type III JNA with infratemporal fossa extension. Although, endoscopic surgery is becoming popular, it still has inherent limitations in far lateral access. The number of total cases of advanced JNA with infratemporal fossa extension operated exclusively endoscopically in literature is very small [11,19,30]. Lateral infratemporal extension of angiofibroma is considered an obstacle for exclusively endoscopic removal of JNA and the patient must be informed that it may be necessary to resort to an open surgery in such types of JNA [30]. Exclusive endoscopic removal of angiofibroma was done by Roger et al. [30] but only one case in their series had an infratemporal fossa extension, in which a residual tumor was left behind. The concept of the exclusive endoscopic approach proponents is that residual disease does not necessarily need a further surgery as it may remain stable or involutes [24]. Second combined surgery is needed only if the tumor remnant progresses or symptoms recur [19,24]. Robinson et al. [31], in a preliminary report of 4 JNAs, used a two-surgeon endoscopic transnasal approach to remove infratemporal extensions of JNA. In their approach, medial maxillectomy, removal of the lateral maxillary wall, ethmoidectomy along with septal dissection in the form of dislocation of the cartilaginous bony junctions, transverse mucosal incision at the side of the tumor and removal of the posterior bony septum were required to remove such angiofibromas. We believe that this is an extensive dissection and may affect facial growth. The antral window in our technique is much less invasive and allows endoscopes to be introduced transnasally and transantrally, gaining benefit from its angles to perfectly resect the tumor. Even in very large JNA, the tumor could be divided at the spheno-palatine foramen with the infratemporal portion being delivered through the sublabial incision and the nasal part through the transnasal route.
1860
M.A. Khalifa, S.M. Ragab
5. Conclusion Endoscopic assisted antral window approach provides a safe reliable effective and minimally invasive technique in management of type III JNA with infratemporal fossa extension. Preoperative embolization is a safe measure in the experienced hands that helps to reduce intraoperative blood loss and improves the quality of the surgical field.
[15]
[16]
[17]
References [18] [1] H. Glad, B. Vainer, C. Buchwald, B.L. Petersen, S.A. Theilgaard, P. Bonvin, et al., Juvenile nasopharyngeal angiofibromas in Denmark 1981—2003: diagnosis, incidence, and treatment, Acta Otolaryngol. 127 (3) (2007) 292—299. [2] A.R. Antonelli, J. Cappiello, D. Di Lorenzo, C.A. Donajo, P. Nicolai, A. Orlandini, Diagnosis, staging, and treatment of juvenile nasopharyngeal angiofibroma (JNA), Laryngoscope 97 (11) (1987) 1319—1325. [3] J.D. Browne, A.H. Messner, Lateral orbital/anterior midfacial degloving approach for nasopharyngeal angiofibromas with cavernous sinus extension, Skull Base Surg. 4 (4) (1994) 232—238. [4] L.G. Close, S.D. Schaefer, B.E. Mickey, S.C. Manning, Surgical management of nasopharyngeal angiofibroma involving the cavernous sinus, Arch. Otolaryngol. Head Neck Surg. 115 (9) (1989) 1091—1095. [5] U. Fisch, The infratemporal fossa approach for nasopharyngeal tumors, Laryngoscope 93 (1) (1983) 36—44. [6] J.C. Andrews, U. Fisch, A. Valavanis, U. Aeppli, M.S. Makek, The surgical management of extensive nasopharyngeal angiofibromas with the infratemporal fossa approach, Laryngoscope 99 (4) (1989) 429—437. [7] P.R. Casson, P.C. Bonanno, J.M. Converse, The midface degloving procedure, Plast. Reconstr. Surg. 53 (1) (1974) 102—103. [8] D.J. Howard, V.J. Lund, The midfacial degloving approach to sinonasal disease, J. Laryngol. Otol. 106 (12) (1992) 1059— 1062. [9] J.G. Spector, Management of juvenile angiofibromata, Laryngoscope 98 (9) (1988) 1016—1026. [10] A.J. Maniglia, Indications and techniques of midfacial degloving: a 15-year experience, Arch. Otolaryngol. Head Neck Surg. 112 (7) (1986) 750—752. [11] L.M. de Brito Macedo Ferreira, E.F. Gomes, J.F. Azevedo, J.R. Souza, A.R. de Paula, A.S. do Nascimento Rios, Endoscopic surgery of nasopharyngeal angiofibroma, Rev. Bras. Otorrinolaringol. (Engl. Ed.) 72 (4) (2006) 475—480. [12] I. Tyagi, R. Syal, A. Goyal, Staging and surgical approaches in large juvenile angiofibroma–—study of 95 cases, Int. J. Pediatr. Otorhinolaryngol. 70 (9) (2006) 1619—1627. [13] O.A. El-Banhawy, S.E.-D. Ael, T. Amer, Endoscopic-assisted midfacial degloving approach for type III juvenile angiofibroma, Int. J. Pediatr. Otorhinolaryngol. 68 (1) (2004) 21—28. [14] E. Pasquini, V. Sciarretta, G. Frank, C. Cantaroni, G.C. Modugno, D. Mazzatenta, et al., Endoscopic treatment of
[19]
[20]
[21] [22]
[23]
[24]
[25] [26]
[27]
[28]
[29]
[30]
[31]
benign tumors of the nose and paranasal sinuses, Otolaryngol. Head Neck Surg. 131 (3) (2004) 180—186. A.H. Marshall, P.J. Bradley, Management dilemmas in the treatment and follow-up of advanced juvenile nasopharyngeal angiofibroma, ORL J. Otorhinolaryngol. Relat. Spec. 68 (5) (2006) 273—278. G. Moulin, C. Chagnaud, R. Gras, E. Gueguen, P. Dessi, J.Y. Gaubert, et al., Juvenile nasopharyngeal angiofibroma: comparison of blood loss during removal in embolized group versus nonembolized group, Cardiovasc. Intervent. Radiol. 18 (3) (1995) 158—161. E. Garcia-Cervigon, S. Bien, D. Rufenacht, C. Thurel, D. Reizine, H.P. Tran Ba, et al., Pre-operative embolization of naso-pharyngeal angiofibromas: report of 58 cases, Neuroradiology 30 (6) (1988) 556—560. A. Casasco, E. Houdart, A. Biondi, H.S. Jhaveri, D. Herbreteau, A. Aymard, et al., Major complications of percutaneous embolization of skull-base tumors, AJNR Am. J. Neuroradiol. 20 (1) (1999) 179—181. T.M. Onerci, O.T. Yucel, O. Ogretmenoglu, Endoscopic surgery in treatment of juvenile nasopharyngeal angiofibroma, Int. J. Pediatr. Otorhinolaryngol. 67 (11) (2003) 1219—1225. M. Onerci, K. Gumus, B. Cil, B. Eldem, A rare complication of embolization in juvenile nasopharyngeal angiofibroma, Int. J. Pediatr. Otorhinolaryngol. 69 (3) (2005) 423—428. G. Lloyd, D. Howard, V.J. Lund, L. Savy, Imaging for juvenile angiofibroma, J. Laryngol. Otol. 114 (9) (2000) 727—730. O.A. El Banhawy, A. Ragab, M.M. El Sharnoby, Surgical resection of type III juvenile angiofibroma without preoperative embolization, Int. J. Pediatr. Otorhinolaryngol. 70 (10) (2006) 1715—1723. D.J. Howard, V.J. Lund, The role of midfacial degloving in modern rhinological practice, J. Laryngol. Otol. 113 (10) (1999) 885—887. P. Herman, G. Lot, R. Chapot, D. Salvan, P.T. Huy, Long-term follow-up of juvenile nasopharyngeal angiofibromas: analysis of recurrences, Laryngoscope 109 (1) (1999) 140—147. I. Tyagi, R. Syal, A. Goyal, Recurrent and residual juvenile angiofibromas, J. Laryngol. Otol. 121 (5) (2007) 460—467. J.J. Fagan, C.H. Snyderman, R.L. Carrau, I.P. Janecka, Nasopharyngeal angiofibromas: selecting a surgical approach, Head Neck 19 (5) (1997) 391—399. D. Radkowski, T. McGill, G.B. Healy, L. Ohlms, D.T. Jones, Angiofibroma: changes in staging and treatment, Arch. Otolaryngol. Head Neck Surg. 122 (2) (1996) 122—129. K. Ungkanont, R.M. Byers, R.S. Weber, D.L. Callender, P.F. Wolf, H. Goepfert, Juvenile nasopharyngeal angiofibroma: an update of therapeutic management, Head Neck 18 (1) (1996) 60—66. M. Jorissen, P. Eloy, P. Rombaux, C. Bachert, J. Daele, Endoscopic sinus surgery for juvenile nasopharyngeal angiofibroma, Acta Otorhinolaryngol. Belg. 54 (2) (2000) 201—219. G. Roger, H.P. Tran Ba, P. Froehlich, A.T. Van Den, J.M. Klossek, E. Serrano, et al., Exclusively endoscopic removal of juvenile nasopharyngeal angiofibroma: trends and limits, Arch. Otolaryngol. Head Neck Surg. 128 (8) (2002) 928—935. S. Robinson, N. Patel, P.J. Wormald, Endoscopic management of benign tumors extending into the infratemporal fossa: a two-surgeon transnasal approach, Laryngoscope 115 (10) (2005) 1818—1822.
Available online at www.sciencedirect.com