Advanced Juvenile Nasopharyngeal Angiofibroma:
Controversies in Management
Brett Cordes, M.D.
June 21, 2007

In 1841, Dr. Liston reported the first treated surgical case of JNA.  In 1847, Dr. Chelius reported the fibrous nature of the tumor and also the correlation with puberty.  Chauveau, in 1906, termed it juvenile nasopharyngeal fibroma and Friedberg, in 1940, termed it angiofibroma and also suggested the implantation of radon seeds as treatment.  Finally, Martin, in 1948, published the first series. 

JNAs are benign, vascular, non-encapsulating tumors.  They represent approximately 0.5% of head and neck neoplasms.  Mean age of onset is 14.9 years; however, the range does extend between about thirteen or fourteen years to twenty-five years in age, and incidence is between 1 in 5 of 60,000 patients.  It is seen almost exclusively in adolescent males.   

The presentation of JNA also includes the triad of nasal obstruction, which is seen between 79% and 100% of the time; epistaxis, which is seen between 50% and 71%; and, an intranasal mass, as shown in this endoscopic photo.  Advanced cases can present with facial deformity, proptosis, cranial nerve deficits, and headache. Here you can see the sphenopalatine foramen, the pterygomaxillary fissure, and the associated fossa as well as infratemporal fossa, and this is how these tumors extend laterally.  This is some radiologic anatomy. Here is the pterygopalatine fossa as it would normally be.  However, a tumor can expand the fossa, as shown here, and can greatly expand it if there is a large JNA pushing the posterior maxillary sinus wall anteriorly and extending it into the infratemporal fossa.  This is another photo around the pterygoid plate.  This is a normal radiograph, and this would be a stage IA tumor; as you can see, there is no bony destruction. 

At this point, the main treatments for advanced JNA include surgery and radiation.  In the past, modalities such as chemotherapy, hormonal therapy, and cryotherapy agents have been used.  As for staging, Dr. Sessions and Dr. Alford, in 1981, in a landmark paper, described the radiologic staging of JNA.  As you can see here, IA correlates with tumors confined to the nose and nasopharynx; IB with extension into the sinuses; IIA with minimal extension onto the pterygopalatine fossa; IIB for occupation of the pterygopalatine fossa with or without erosion of the orbital bones; IIC extending into the infratemporal fossa; and, finally, stage III with intracranial extension.  Ugo Fisch developed his own staging system a couple of years after that paper and, as you can see here, ranging from stage I, with tumor occupying the nasal cavity and nasopharynx to stage IV, representing massive invasion of the cavernous sinus, optic chiasm or pituitary fossa.  Radkowski developed another staging system in 1996.  This was taken directly from Dr. Sessions’ previous staging system.  As you can see, stages IA through IIC are identical; however, stages IIC and III in the Sessions’ system were expanded to include stages IIC through IIIB in the Radkowski system, as shown here. This was done because it was felt that it added additional prognostic factors and also suggested different approaches that could be used. 

So what exactly is advanced JNA?  An arbitrary definition could be: a tumor greater than or equal to stage IIB, which is correlated to a Fisch stage III.  Current controversies in the management of advanced JNA include the role of radiation, the optimal surgical approach, as well as how to manage for recurrent or residual disease, all the while trying to keep the principle of tumor excision (which is complete resection) balanced with minimal morbidity.  However, in the recent past, even this principle has been considered a controversial in that these tumors are benign and it may not be necessary to completely excise them. 

The origin of this tumor is felt to be in the superior border of sphenopalatine foramen, specifically at the borders of the sphenoid process of the palatine bone, the horizontal ala of the vomer, and the roof of the pterygoid process of the sphenoid.  Patterns of spread for this tumor include posterosuperior spread into the sphenoid sinus anteriorly or superiorly in the ethmoids.  It can spread posteriorly and inferiorly into the choanae or anteriorly into the nasal cavity.  They can also spread laterally, as shown here, into the pterygopalatine fossa or also more laterally in the infratemporal fossa, or superiorly in the orbit, or even in the middle cranial fossa.  Now the routes of relative intracranial extension include through the infratemporal fossa and onto the floor of the middle cranial fossa, through the pterygomaxillary fissure, infratemporal fossa, and into the inferior or superior orbital fissures.  It has been shown to erode the sphenoid sinus indirectly into the cavernous sinus, as in our patient.  Additionally, however, in a limited amount of time, it can grow along the horizontal lamina of the ethmoids, as well as the cribriform plate into the anterior cranial fossa. 

How are these tumors surgically approached?  Surgical approach is dictated by the stage of the tumor with maximized exposure of the tumor while, of course, limiting the morbidity and mortality.  In small tumors, that is stage IA, IB and IIA, there is very little debate.  Conventional approaches include transpalatal or transantral approaches; however, modern approaches include endoscopic approaches.  In advanced JNA, there is considerable debate as to the optimal surgical approach. 

The options for approaching JNA include medial maxillectomy, bilateral rhinotomy or midfacial degloving, Le Fort I osteotomy, or, of course, the endoscopic approach.  The time-honored infratemporal fossa approach, as well as Tyagi, who described his combined transmaxillary, transpalatal approach, cranial facial resection and a frontotemporal craniotomy. We will briefly go through these approaches. 

Lateral rhinotomy is appropriate for stages IIA to IIC or IIIA tumors, and these are tumors confined to the nasopharynx, nasal cavity, paranasal sinuses, pterygopalatine fossa, and specifically, the medial temporal fossa.  As you can see here, the incision is carried down to the bone, and you can see the bone window about to be opened.  It does give direct exposure of the surgical field; however, the disadvantages include intranasal crusting, which occurs in almost every patien; scarring; numbness; and nasolacrimal duct obstruction as well as a propensity for facial asymmetry and the development of problems with growth of the facial bones. 

The midfacial degloving approach is appropriate for stages IIA through IIIA.   These are tumors in the same locations, as those accessible via lateral rhinotomy approach.  The advantage of midfacial degloving is that it has no external incisions, and it does have direct exposure of the tumor.  The complications include the fact that it could affect facial development.  It also could develop vestibular stenosis and, as mentioned previously, intranasal crusting.  Additionally, Howard et al, in 2001, suggested that drilling the base of sphenoid via the pterygoid canal resulted in removal of all residual tumor and therefore minimizes recurrence rate.

The Le Fort I approach is appropriate for stages IIIA and IIIB tumors.  These are tumors with extensive skull base as well as medial cavernous sinus invasion.  It can involve unilateral or bilateral osteotomies with a resultant palatal drop and you can also resect the posterior maxillary sinus wall, allowing you to control tumor vascular supply.  The disadvantage of this approach is the extensive bony disruption that it is warranted and results in facial skeletal abnormalities. 

The infratemporal fossa approach is appropriate for stages IIC and IIIB tumors, particularly tumors that invade the lateral cavernous sinus.  As you can see here, this is a temporal modification of the infratemporal fossa approach that permits early ligation of the internal maxillary artery as well as early visualization of the internal carotid artery.  Several modifications of this approach have been given, specifically the Fisch type C or postauricular approach, as well as the Fisch type B or preauricular approach.  Mickey et al., in 1988, combined the frontotemporal craniotomy approach as well as the lateral infratemporal fossa approach to gain access to the anterior cavernous sinus and orbital apex, as well as extradural and intradural intracranial tumors. 

The temporal approach, as shown here, as taken by Browne et al., was reported in 2000.  It is a lateral preauricular infratemporal fossa approach, which they consider a modification of Fisch type C.  As you can see, this is the stage IIIA tumor on the right.  It starts with a curvilinear preauricular skin incision; the skin flap is elevated anteriorly giving exposure to the zygomatic arch, lateral orbital rim, and the temporalis muscle.  The extent of bone removal varies with tumor involvement, and an additional temporal craniectomy can be performed for dural or BII involvement.  This is a theoretical exposure of this approach.  We have tumor here in the dark.  This is BIII, this is BII, and this is the middle fossa dura here, and this is what a real life exposure might look like.  The light arrow is pointing to BII, which is being displaced superiorly by the tumor shown by the black arrow. 

A pterional craniotomy can also be performed for extensive intradural involvement.  This is the tumor being removed from the nasopharyngeal fossa, and finally the temporalis muscle fills the temporal and infratemporal fossa defects.  This combines the transpalatal and transmaxillary approach by Tyagi et al.  He did a retrospective study of eighty-five cases, all of which contained stages IIB to IIIA JNA tumors.  He uses a Lazy S incision, as shown here, and it is appropriate for tumors involving infratemporal as well as the pterygopalatine fossa and the orbit with minimal intracranial involvement.  Complications included CSF leak, palatal fistula, as well as serous otitis media, and Eustachian tube dysfunction, which were the more common complications.  Additionally, he had no facial asymmetry or cosmetic defects with these approaches. 

Craniofacial resection is appropriate for stage IIIB tumors.  Dr. Bales et al., and this includes Dr. Randy Weber when he was at Penn, described this approach for extensive intracranial involvement.  This is a stage IIIB JNA tumor invading the left cavernous sinus and displacing the internal carotid arteries laterally.  Bales combines an infratemporal fossa approach with a facial translocation and starts with a hemicoronal skin incision combined with the Weber-Ferguson incision, as shown here.  These are the bone cuts that they make, subtemporal craniectomy as well as removing the zygomatic arch, and the bone cut that you might see with a temporal transfacial approach.  This is the vascularized osteoplastic flap with the transfacial approach that has been lateralized.  Soft tissue defects are filled with rotated temporalis muscle.  Complications included nasal crusting, serous otitis media, and facial numbness.

Finally, the intranasal endoscopic approach.  This approach has recently been shown to be appropriate for stages up to IIIA, excluding stage IIC, since you do have limited exposure of the lateral infratemporal fossa.  Here you can see the view that might be obtained with a zero-degree endoscope, and these black lines show the extra view that you might obtain with a thirty-degree endoscope.  The reason why these tumors, which are somewhat extensive, are accessible via the endoscopic approach is because the tumors, although nonencapsulated, often come out in one whole total piece.  Briefly, the technique starts with a middle turbinectomy, and goes on to removing the posterior maxillary sinus wall and tumor resection.  Here you can see tumor coming out of the pterygomaxillary fossa, and finally, this inset here shows the sphenopalatine artery being clipped via this approach.  The advantages are reduced blood loss, complication rate, and hospital stay, and more importantly, this approach has been shown to have comparable recurrence rate. 

Lastly, the craniotomy is appropriate for stage IIIB tumors, as in our case.  It is often combined with anterior or lateral approaches, and it likely requires neurosurgical intervention.  It does have increased risk of surgical morbidities, specifically to the cranial nerves, and it may be performed as Tyagi suggested, as a staged approach to minimize morbidity. 

What does the evidence show in terms of the optimal surgical approach?  Hosseini et al., performed a retrospective review of thirty-seven patients with a mean followup of 46.5 months.  The total recurrence rate was 27%, five of which were symptomatic and five of which were only residual tumor, and, as you can see here, the recurrence rate is directly correlated with the stage of the tumor.  It did show a statistical risk for the tumor recurring when these tumors extended and involved only these three places in anatomy:  the orbit, neurocranial fossa and the base of pterygoids.  His presenting stages included  eight IIC tumors, two IIIA tumors, and five  IIIB tumors.  His conclusion, based solely on recurrence rate, is that the midfacial degloving approach was most appropriate for stage IIC tumors, and the Le Fort I approach was most appropriate for stages IIIA and IIIB tumors.  His other approaches, although not listed, were transpalatal. 

Tosun et al., performed another retrospective review of twenty-four patients with a mean followup of two years.  He showed no recurrences for stages IIB or IIIA tumors ,and his approaches included lateral rhinotomy for his IIB tumor, as well as those shown here, for his IIIA tumors.  Complications involved in the lateral rhinotomy and then midfacial degloving approaches were crusting and scarring, as shown before.  However, his endoscopic cases had no complications.  Granted, there were only two cases, but they were extensive tumors, and we are beginning to see progression for this minimally invasive approach. 

This is a study by Cansiz et al.  It shows his approach is dictated by tumor stage.  Stages I and II tumors were approached via the endoscopic route almost exclusively.  However, he describes the modified midfacial degloving approach for stages III and IV, as well as the craniotomy approach for stage IV.  His complications included temporofacial palsy for his combined approach as well as nasal crusting, facial paresthesias, and one ITA rupture intraoperatively.  He only had four of twenty-two recurrences, and his suggestion is that the modified midfacial degloving approach is appropriate for almost every tumor other than stage I and II.  This is a photo of one of their patients postoperatively showing no cosmetic deformity. 

Here is the changing surgical concept that we have been seeing between 1980 and 2002.  The incidence of the early stages, specifically stage S1 and S2, has increased, probably because of better technology.  However, the trend is showing that the endoscopic approach is becoming more favorable, and, in fact, between 2000 and the year 2002 they exclusively used the endoscopic approach.  The transpalatal approach was associated with 1/1 recurrence, lateral rhinotomy had zero recurrences, and in only one of fifteen for the endoscopic approach.  They stated that they had 1.8 recurrences per patient, and that this was comparable to the reported rate since 1960 and independence of tumor stage in the approach used for initial tumor removal.  There were no major complications.  Now, we have shown at least in one paper, it does depend on tumor stage.  However, it does appear that the approach itself does not necessarily correlate with recurrence.  They used the endonasal approach between 1995 and 1999 only for the stage I and II tumors; however, between 2000 and 2002, even with stage III disease, they used the endoscopic approach exclusively.  Their conclusion was that endoscopy was appropriate for up to stage III tumors, and it may also be combined with degloving approaches. 

This paper, from the Mayo Clinic, describes endoscopic versus traditional approaches over the last thirty years.  As you can see, there were many more tumors involved with the traditional approach as opposed to the endoscopic.  However, traditional approaches were used for more extensive tumors, and the six endoscopic patients in the staging were not available for either one.  The recurrence rate for the traditional approaches was 24% with a mean followup of four years.  The recurrence rate for the endoscopic approach was 0%, but follow-up was only thirteen months.  The blood loss and the hospital stay for the traditional group was 1250 cc and five days respectively, compared to that of endoscopic group, which was 225 cc and two days respectively. 

Complications, as you can see here, were fairly numerous, but minimal in the traditional approach.  There was one temporary cheek numbness as a complication of the endoscopic approach.  They concluded that the endoscopic approach is safe and effective for selective tumors up to Sessions IIC, per their paper, and with less complications. 

How do we manage recurrence of residual disease?   The incidence of recurrence that we have talked about overall was usually between 22% and 37% and does appear to be higher with increased staging of the tumors.  Some considerations in managing the recurrent disease.  First of all, the presence of persistent radiographic abnormality does not necessarily signify residual tumor.  In fact, nine of eleven patients on the two-year followup in Lee’s paper did have an abnormality on follow-up imaging.  However, biopsy was significant only for chronic inflammation.  Additionally, it has been suggested that there is no recurrence of JNA and that it is more appropriate to think of it as incomplete resection. 

Finally, Cansiz, in 2006, suggested that there is spontaneous regression of the tumor at about age twenty to twenty-five years, and this is backed up by years of anecdotal evidence that these tumors rarely present in patients older than twenty or twenty-five years.  As you can see here, although this represents chronic inflammation in our case, you might not want to take someone back to the operating room quite as aggressively when there is a possibility of spontaneous regression.

In summary, for residual disease, options include observation with periodic imaging for persistent radiographic abnormality or disease that is asymptomatic.  Of course, you can surgically excise these lesions with symptomatic disease and the approach for the revision cases is the same as that for the primary disease, relying only on anatomy in the restaging of the tumor. And, of course, radiation therapy is an option.  Radiation therapy is appropriate for unresectable disease as well as residual disease.  Typically given between 3000 and 5500 cGy over three to six weeks.  Dr. Kuppersmith and Dr. Donovan, here at Baylor, performed a small study showing that 4500 cGy of IMRT had promising results in the management of advanced JNA.  The controversy, however, with radiation therapy lies in balancing the acute morbidity and mortality of surgical resection with the late complications of radiation.  UCLA, over the last forty years, performed a study on twenty-seven of their advanced JNA cases treated with primary XRT.  As  shown here, the average follow-up was six years.  Four of their patients developed recurrences between two and five years after treatment and four of their patients developed long-term complications.  Now, even though there were only four patients who developed these complications, they are actually six complications ranging from presentation three years after treatment to presentation 10 years after treatment; and you can see here the range of complications they incurred. 

This is a table from their study.  As you can see here, over the years, they have improved or at least changed their amount of radiation that they give as well as CT conformal planning enhancing the way in which it is given,  which could lead to greater treatment benefits and less complications.  Recurrence is only 15%, which is technically less than surgery, and the complications are shown here, as we previously discussed.  Is this increased efficiency of the radiation or is it simply inadequate follow-up?  This is another chart showing radiation therapy of several large studies over the last fifty years, and again, the recurrence rate for all of their JNA cases is only 14%, which is again less than surgery for all stages.  Now we are beginning to see complications including radiation-induced malignancy, and the question is: are these complications comparable to the acute morbidity of surgery?  This is the question that needs to be discussed with the patient, radiation therapist, surgeon, etc. 

This is the chart showing radiation therapy only for advanced JNA lesions.  There are several studies shown here.  Again, recurrence rate is 27%, and this is only for advanced JNA, so technically this is still less than surgery.  Complications are also minimal, as shown in these studies.  Now, the largest study was Dr. McGahan and Dr. Parke here at Baylor in 1989, in which they illustrate a fairly important point:  4/4 tumors treated at 3200 cGy recurred,  whereas 0/16 treated at this radiation was treated successfully, both of which had no complications. 

Radiation treatment is appropriate for unresectable or residual recurrent disease.  The overall success rate is 73% to 85%, which is technically higher than with surgery.  However, it is weighed against the potential for late complications.  Additionally, success in the radiation literature is defined as having either a tumor that is not symptomatic or a tumor that is not enlarging,  and, in fact, these tumors takes several years to actually disappear.  Radiation requires longer followup to delineate these complications,  and Lee described the fact that the latency period for complications resulting from radiation can be as long as twenty years.  Finally, modern techniques should increase the efficacy of this treatment, including specifically IMRT as well as better eye shielding and shielding near the pituitary. The current recommended dose is 3600 to 4000 cGy over three to six weeks. 

As for surgery for advanced JNA, approaches are dictated by the stage, the risk of the complication, and perhaps, most importantly. the surgeon’s preference.  The endoscopic approach is gaining popularity, and one factor that really cannot be surmised in a paper is that of surgeon capability and surgeon experience, and, in fact, if the surgeon is quite capable and experienced with removing these extensive tumors, it allows for definitive and permanent treatment of this disease process.

Case Presentation:

NL is a 13-year-old boy referred to TCH for the evaluation and management of a juvenile nasopharyngeal angiofibroma.  He presented acutely to the ER for brisk, unremitting epistaxis, and was admitted to the hospital. Further questioning revealed a 5 month history of nasal congestion. He had no complaints of facial pain, edema, or vision disturbances.

Past medical, surgical, and social history were unremarkable.

Physical exam was significant for a left-sided, purple, posterior nasal mass that bled on manipulation. Cranial nerves II-XII were intact. He had no external facial deformity.

Imaging was obtained, including CT and MRI, and an intranasal mass was demonstrated to be invading the intracranial cavity. An arteriogram with embolization was also acquired, and the diagnosis of JNA was made.

The patient was taken to the operating room, in conjunction with neurosurgery, for surgical removal. The mass was extirpated without obvious sequelae, the diagnosis was confirmed to be JNA, and the patient did well post-operatively.

His follow up clinical exam and imaging show no evidence of disease, and there have been no complications.

Bibliography

Andrews JC, Fisch U, Valavanis A, Aeppli U, Makek MS. The surgical management of extensive nasopharyngeal angiofibromas with the infratemporal fossa approach. Laryngoscope 1989;99:429-437.

Bales C, Kotapka M, Loevner LA, Al-Rawi M, Weinstein G, Hurst R, Weber RS. Craniofacial resection of advanced juvenile nasopharyngeal angiofibroma. Arch Otolaryngol Head Neck Surg 2002;128:1071-1078.

Briant TD, Fitzpatrick PJ, Berman J. Nasopharyngeal angiofibroma: a twenty year study. Laryngoscope 1978;88:1247-51.

Browne JD, Jacob SL. Temporal approach for resection of juvenile nasopharyngeal angiofibromas. Laryngoscope 2000;110:1287-1293.

Cansiz H, Guvenc MG, Sekercioglu N. Surgical approaches to juvenile nasopharyngeal angiofibroma. J Cranio-Maxillofacial Surg 2006;34:3-8.

Chandler JR, Goulding R, Moskowitz L, Quencer RM. Nasopharyngeal angiofibromas: staging and management. Ann Otol Rhinol Laryngol 1984;93:322-329.

Cummings BJ. Relative risk factors in the treatment of juvenile nasopharyngeal angiofibroma. Head Neck Surg 1980;3:21-6.

Cummings BJ, Blend R, Keane T, Fitzpatrick P, Beale F, Clark R, Garrett P, Harwood A, Payne D, Rider W. Primary radiation therapy for juvenile nasopharyngeal angiofibroma. Laryngoscope 1984;94:1599-605.

Enepekides DJ. Recent advances in the treatment of juvenile angiofibroma. Curr Opin Otolaryngol Head Neck Surg 2004;12:495-499.

Fields JN, Halverson KJ, Devineni VR, Simpson JR, Perez CA. Juvenile nasopharyngeal angiofibroma: efficacy of radiation therapy. Radiology 1990;176:263-5.

Fisch U. The infratemporal fossa approach for nasopharyngeal tumors. Laryngoscope 1983;93:36-44.

Fitzpatrick PJ. The nasopharyngeal angiofibroma. Clin Radiol 1967;18:62-8.

Friedberg SA. Vascular fibroma of the nasopharynx (nasopharyngeal fibroma). Arch Otolaryngol 1940;31:313-326.

Goepfert H, Cangir A, Lee YY. Chemotherapy for aggressive juvenile nasopharyngeal angiofibroma. Laryngoscope 1985;111:285-289.

Hosseini SMS, Borghei P, Borghei SH, Ashtiani MTK, Shirkhoda A. Angiofibroma: an outcome review of conventional surgical approaches. Eur Arch Otorhinolaryngol 2005;262:807-812.

Howard DJ, Lloyd G, Lund V. Recurrence and its avoidance in juvenile angiofibroma. Laryngoscope 2001;111:1509-1511.

Kuppersmith RB, The BS, Donovan DT, Mai W-Y, Chiu JK, Woo SY, Butler EB. The use of intensity modulated radiotherapy for the treatment of extensive and recurrent juvenile angiofibroma. Int J Pediatr Otorhinolaryngol 2000;52:261-268.

Lee JT, Chen P, Safa A, Juillard G, Calcaterra TC. The role of radiation in the treatment of advanced juvenile angiofibroma. Laryngoscope 2002;112:1213-1220.

Mann WJ, Jecker P, Amedee RG. Juvenile Angiofibromas: changing surgical concept over the last 20 years. Laryngoscope 2004;114:291-293.

McGahan RA, Durrance FY, Parke RB Jr, Easley JD, Chou JL. The treatment of advanced juvenile nasopharyngeal angiofibromas. Int J Radiat Oncol Biol Phys 1989;17:1067-1072.

Naraghi M, Kashfi A. Endoscopic resection of nasopharyngeal angiofibromas by combined transnasal and transoral routes. Am J Otolaryngol 2003;24:149-154.

Paris J, Guelfucci B, Moulin G, Zanaret M, Triglia JM. Diagnosis and treatment of juvenile nasopharyngeal angiofibroma. Eur Arch Otorhinolaryngol 2001;258:120-124.

Pryor SG, Moore EJ, Kasperbauer JL. Endoscopic versus traditional approaches for excision of juvenile nasopharyngeal angiofibroma. Laryngoscope 2005;115:1201-1207.

Radkowski D, McGill T, Healy GB, Ohlms L, Jones DT. Angiofibroma: changes in staging and treatment. Arch Otolaryngol Head Neck Surg 1996;122:122-129.

Roger G, Tran Ba Huy T, Froehlich P, Van den Abbeele T, Klossek JM, Serrano E, Garabedian EN, Herman P. Exclusively endoscopic removal of juvenile nasopharyngeal angiofibroma. Arch Otolaryngol Head Neck Surg 2002;128:928-935.

Saylam G, Yucel OT, Sungur A, Onerci M. Proliferation, angiogenesis and hormonal markers in juvenile nasopharyngeal angiofibroma. Int J Pediatr Otorhinolaryngol 2006:70:227-234.

Sessions RB, Bryan RN, Naclerio RM, Alford BR. Radiographic staging of juvenile angiofibroma. Head Neck Surg 1981;3:279-83.

Sessions RB, Wills PI, Alford BR, Harrell JE, Evans RA. Juvenile nasopharyngeal angiofibroma: radiographic aspects. Laryngoscope 1976;86:2-18.

Solomon D, Isaacson G. Transoral ‘adenoidectomy’ excision of juvenile nasopharyngeal angiofibromas. Ann Otol Rhinol Laryngol 2007;116:243-7.

Tosun F, Ozer C, Gerek M, Yetiser S. Surgical approaches for nasopharyngeal angiofibroma: comparative analysis and current trends. J Craniofac Surg 2006 Jan;17(1):15-20.

Tyagi I, Syal R, Goyal A. Staging and surgical approaches in large juvenile angiofibroma—study of 95 cases. Int J Pediatr Otorhinolaryngol 2006;70:1619-1627.

Waldman SR, Levine HL, Astor F, Wood BG, Weinstein M, Tucker HM. Surgical experience with nasopharyngeal angiofibroma. Arch Otolaryngol 1981;107:677-82.