The Laryngoscope C 2013 The American Laryngological, V
Rhinological and Otological Society, Inc.
Surgical Management of Juvenile Nasopharyngeal Angiofibroma: Analysis of 162 Cases From 1995 to 2012 Yang Huang, MD; Zhuofu Liu, MD; Jingjing Wang, MD; Xicai Sun, MD; Lei Yang, MS; Dehui Wang, MD, PhD Objectives/Hypothesis: The purpose of this study was to report on a series of 162 patients presenting with juvenile nasopharyngeal angiofibroma in a single academic hospital during the past 17 years, in an effort to compare outcomes between open and transnasal endoscopic approach, and to define an ideal treatment strategy. Study Design: Patients who received either open or endoscopic surgery with a minimum follow-up of 6 months were selected. Local control and complications were compared between groups. Methods: Retrospectively, clinical data, surgical reports, pre- and postoperative images, and follow-up information were reviewed and analyzed. Results: All patients were male subjects from 8 to 41 years old. Ninety-six patients were treated by transpalatal or transmaxillary approach, and the remaining 66 patients were treated using transnasal endoscopic approach with/without labiogingival incision. When compared to the open surgery group, the endoscopic surgery group showed a lower median intraoperative blood loss (800 vs. 1100 mL, P 5.017) and a lower number of postoperative complications (one vs. 10). In addition, recurrence statistically correlated with Radkowski’s classification and patient age. Conclusions: Transnasal endoscopic approach can be successfully used for Radkowski’s stages I-IIb tumors and selective IIc-IIIb lesions, allowing for less blood loss, fewer postoperative complications, and a lower percentage of recurrence in comparison to open surgery. The management of recurrent tumor is complex, should be individually tailored, and should take into account tumor location, patient age, complications of treatment, and the possibility of spontaneous involution, to better define treatment strategy. Key Words: Juvenile nasopharyngeal angiofibroma, endoscopic surgery, skull base, treatment outcome, recurrence. Level of Evidence: 4. Laryngoscope, 00:000–000, 2014
INTRODUCTION Juvenile nasopharyngeal angiofibroma (JNA) is a rare, histologically benign but highly vascular and locally invasive tumor that shows characteristic epidemiological features and growth patterns. This lesion is seen predominantly in adolescent male subjects presenting with recurrent epistaxis and unilateral or bilateral nasal obstruction, and it accounts for 0.05% to 1% of all head and neck tumors.1,2 JNA is considered to originate from the superior margin of the sphenopalatine foramen. From its origin, the tumor grows medially into the nasopharynx and nasal sinuses, laterally into the sphenopalatine and infratemporal fossae, posteriorly to critical anatomic structures such as internal carotid artery (ICA), cavernous sinus and orbital apex, and may even
From the Department of Otorhinolaryngology (Y.H., Z.L., J.W., X.S., and Eye, Ear, Nose, and Throat Hospital, and the Department of Biostatistics (L.Y.), School of Public Health, Fudan University, Shanghai, China. Editor’s Note: This Manuscript was accepted for publication November 11, 2013. This work was supported by the Shanghai Science and Technology Commission Fund, Natural Science Foundation of China. The authors have no other funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Dehui Wang, MD, 83 Fenyang Road, Shanghai, China 200031. E-mail: [email protected] D.W.);
DOI: 10.1002/lary.24522
Laryngoscope 00: Month 2014
invade the skull base or extend intracranially. Skull base invasion is seen in approximately 20% and intracranial involvement in 4.3% to 11% of cases, although frank dural invasion is rare.1,3,4 Surgical excision is still recognized to be the first choice of treatment for JNA, especially when there is no intracranial involvement. Traditional open surgical approaches are comprised of transpalatal, transmaxillary (lateral rhinotomy or midfacial degloving), LeFort 1 osteotomy, and infratemporal fossa craniotomy.5 During the past 2 decades, the surgical approach to JNA has been evolving in relation to the improvement of endoscopic techniques. Compared to open surgery, transnasal endoscopic surgery has less morbidity, better amplifying visualization in identifying residual tumor behind corners or other inaccessible locations, lower intraoperative blood loss, less need for blood transfusion, less hospitalization, and a lower rate of recurrence.6–8 Recurrence in JNA is common and related to many factors. Radkowski found that the most important factor in determining the risk of postoperative recurrence was preoperative tumor stage.9 Reported recurrence rates vary from approximately 20% to over 50%.3,10 This article presents a retrospective, descriptive study of 162 patients with a diagnosis of JNA during the past 17 years. To the best of our knowledge, few articles have been published comparing open and endoscopic approaches for treating large, locally advanced JNA in fairly large numbers of patients. We report our surgical results with regard to Huang et al.: Surgical Management of Juvenile Angiofibroma
1
TABLE I. Radkowski Staging System for Juvenile Nasopharyngeal Angiofibroma. Stage
Ia
Limited to nose and/or nasopharyngeal vault
Ib IIa
Extension into one or more paranasal sinus Minimal extension into pterygopalatine fossa
IIb
Full occupation of pterygopalatine fossa with or without erosion of orbital bones
IIc
Extension into the infratemporal fossa or posterior to pterygoid plates Erosion of skull base—minimal intracranial extension
IIIa IIIb
Erosion of skull base—extensive intracranial extensionwith or without cavernous sinus invasion
tumor stage, surgical strategy, recurrence rate, and complications to evaluate the two techniques.
MATERIALS AND METHODS Patient Selection The medical records of patients with histologically proven JNA who received surgical treatment at Fudan University Eye, Ear, Nose, and Throat Hospital between September 1, 1995 and April 30, 2012 were reviewed retrospectively. Patients who received either open surgery or transnasal endoscopic surgery were selected, whereas combined-approach cases (patients who received open approach combined with transnasal endoscopic surgery) were excluded. Epidemiological and clinical data, surgical reports, pre- and postoperative images, details concerning complications, and follow-up information of selected cases were retrieved and analyzed. Radkowski’s classification9 (Table I) based on preoperative computed tomography (CT) and magnetic resonance imaging (MRI) was adopted. A standard follow-up protocol was instituted for all patients. Patients were required to visit the surgeon regularly (1 week, 1 month, 3 months, 6 months, 9 months, 12 months after surgery during the first postoperative year, every 6 months in the second year, and every 12 months afterward until the end of puberty), and undergo nasal endoscopy during each visit and CT or MRI every 6 months. In this study, recurrence was diagnosed as the finding of an active enhancing mass on endoscopy and/or imaging, and either surgery or radiotherapy was carried out. If follow-up imaging demonstrated that the mass was stable or showed regression, the patient was monitored every 6 months.
received surgery previously at other centers. Recurrent epistaxis and nasal obstruction were the most common presenting symptoms. Using Radkowski’s stage classification, we found 14 patients (8.6%) with Ia, 15 (9.3%) with Ib, 28 (17.3%) with IIa, 17 (10.5%) with IIb, 54 (33.3%) with IIc, 16 (9.9%) with IIIa, and 18 (11.1%) with IIIb disease. No patients had intradural invasion in our series. The median follow-up time was 55 months (range, 6–182 months). Six patients could not be contacted after discharge from our hospital and were lost to follow-up. Embolization, as a routine preoperative procedure from 2005 and onward, was performed uneventfully in 64 patients just 1 or 2 days before surgery and was carried out at Shanghai No. 9 Hospital. It was not until the year 2000 that transnasal endoscopic techniques were used to resect JNA at our hospital. Since then, endoscopic surgery has been performed increasingly, not only in cases extending to paranasal sinus and pterygopalatine fossa but also in cases in which infratemporal fossa and skull base were involved. In this study, 96 patients were treated by transpalatal or transmaxillary (lateral rhinotomy or midfacial degloving) approaches and are defined as group 1. The remaining 66 patients, defined as group 2, were treated through transnasal endoscopic approach with/without labiogingival incision. Staging of patients in the two groups was equally distributed (Table II). The median age was 17 years (range, 8–41 years) in group 1 and 16 years (range, 9–41 years) in group 2.
Recurrences According to Clinical Stage, Age, and Surgical Approach The overall recurrence rate was 31.4% (49/156) in our study. Median recurrence discovery was 16 months after the initial operation at our hospital. Patients suffering stage IIIb JNA, according to Radkowski’s staging system, had the highest recurrence rate (61.1%, 11/18), which was a significant difference compared to the other patients. The remaining patients’ recurrence rate almost reduced according to their disease stage (Table III). The odds ratio also confirmed that patients with other stage JNA were less likely to experience recurrence than patients with a stage IIIb lesion.
Statistical Analysis The influence of different factors on tumor recurrence was performed by Fisher exact test. Wilcoxon test and v2 test. Logistic regression was also performed to evaluate the effect of main factors (surgical approach and staging) on recurrence. Factor analysis of variance on log-transformed data for intraoperative blood loss was given. All statistical analysis for this article was generated using SAS/STAT software (version 9.2 of the SAS System for Windows; SAS, Inc., Cary, NC) and SPSS software (version 19 of SPSS for Windows; SPSS Inc., Chicago, IL).
TABLE II. Staging of Patients According to Radkowski’s Classification. Open Surgery (Group 1) Stage
Endoscopic Surgery (Group 2)
No. of Patients
Proportion
Stage
No. of Patients
Proportion
Ia
10
10.4%
Ia
4
6.1%
Ib IIa
10 22
10.4% 22.9%
Ib IIa
5 6
7.6% 9.1%
IIb
6
6.3%
IIb
11
16.7%
RESULTS
IIc IIIa
30 10
31.3% 10.4%
IIc IIIa
24 6
36.4% 9.1%
Patient Characteristics
IIIb
IIIb
10
15.2%
Total
66
100%
All patients were male subjects ranging in age from 8 to 41 years (mean, 17.5). Thirty patients (18.5%) had Laryngoscope 00: Month 2014
2
Total
8
8.3%
96
100%
Huang et al.: Surgical Management of Juvenile Angiofibroma
TABLE III. Recurrence Rate and OR According to Radkowski Staging System. Stage
No. of Recurrences
OR
95% Wald Confidence Limits
Ia
2/14 (14.3%)
—
—
—
Ib
1/14 (7.1%)
0.438
0.035
5.510
IIa IIb
7/28 (25.0%) 5/17 (29.4%)
1.907 2.908
0.338 0.452
10.768 18.706
IIc
15/50 (30.0%)
2.456
0.474
12.724
IIIa IIIb
8/15 (53.3%) 11/18 (61.1%)
6.267 9.807
0.989 1.578
39.722 60.956
developed laryngeal obstruction that needed tracheotomy, one patient complained of temporary aural fullness, and another patient complained of exotropia in the right eye because of dysfunction of the right medial rectus muscle. In the endoscopic surgery group, only one patient complained of epistaxis. The median intraoperative blood loss was 1,100 mL (range, 50–5,300 mL) in the open surgery group, compared to 800 mL (range, 50–5,000 mL) in the endoscopic surgery group, a statistically significant difference (P 5.017) (Table IV). The median amount of blood loss in the embolized group was 1,041 mL, compared to 1,249 mL in the nonembolized group (P 5.129).
Fisher exact test, P 5.0105 (
Rhinological and Otological Society, Inc.
Surgical Management of Juvenile Nasopharyngeal Angiofibroma: Analysis of 162 Cases From 1995 to 2012 Yang Huang, MD; Zhuofu Liu, MD; Jingjing Wang, MD; Xicai Sun, MD; Lei Yang, MS; Dehui Wang, MD, PhD Objectives/Hypothesis: The purpose of this study was to report on a series of 162 patients presenting with juvenile nasopharyngeal angiofibroma in a single academic hospital during the past 17 years, in an effort to compare outcomes between open and transnasal endoscopic approach, and to define an ideal treatment strategy. Study Design: Patients who received either open or endoscopic surgery with a minimum follow-up of 6 months were selected. Local control and complications were compared between groups. Methods: Retrospectively, clinical data, surgical reports, pre- and postoperative images, and follow-up information were reviewed and analyzed. Results: All patients were male subjects from 8 to 41 years old. Ninety-six patients were treated by transpalatal or transmaxillary approach, and the remaining 66 patients were treated using transnasal endoscopic approach with/without labiogingival incision. When compared to the open surgery group, the endoscopic surgery group showed a lower median intraoperative blood loss (800 vs. 1100 mL, P 5.017) and a lower number of postoperative complications (one vs. 10). In addition, recurrence statistically correlated with Radkowski’s classification and patient age. Conclusions: Transnasal endoscopic approach can be successfully used for Radkowski’s stages I-IIb tumors and selective IIc-IIIb lesions, allowing for less blood loss, fewer postoperative complications, and a lower percentage of recurrence in comparison to open surgery. The management of recurrent tumor is complex, should be individually tailored, and should take into account tumor location, patient age, complications of treatment, and the possibility of spontaneous involution, to better define treatment strategy. Key Words: Juvenile nasopharyngeal angiofibroma, endoscopic surgery, skull base, treatment outcome, recurrence. Level of Evidence: 4. Laryngoscope, 00:000–000, 2014
INTRODUCTION Juvenile nasopharyngeal angiofibroma (JNA) is a rare, histologically benign but highly vascular and locally invasive tumor that shows characteristic epidemiological features and growth patterns. This lesion is seen predominantly in adolescent male subjects presenting with recurrent epistaxis and unilateral or bilateral nasal obstruction, and it accounts for 0.05% to 1% of all head and neck tumors.1,2 JNA is considered to originate from the superior margin of the sphenopalatine foramen. From its origin, the tumor grows medially into the nasopharynx and nasal sinuses, laterally into the sphenopalatine and infratemporal fossae, posteriorly to critical anatomic structures such as internal carotid artery (ICA), cavernous sinus and orbital apex, and may even
From the Department of Otorhinolaryngology (Y.H., Z.L., J.W., X.S., and Eye, Ear, Nose, and Throat Hospital, and the Department of Biostatistics (L.Y.), School of Public Health, Fudan University, Shanghai, China. Editor’s Note: This Manuscript was accepted for publication November 11, 2013. This work was supported by the Shanghai Science and Technology Commission Fund, Natural Science Foundation of China. The authors have no other funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Dehui Wang, MD, 83 Fenyang Road, Shanghai, China 200031. E-mail: [email protected] D.W.);
DOI: 10.1002/lary.24522
Laryngoscope 00: Month 2014
invade the skull base or extend intracranially. Skull base invasion is seen in approximately 20% and intracranial involvement in 4.3% to 11% of cases, although frank dural invasion is rare.1,3,4 Surgical excision is still recognized to be the first choice of treatment for JNA, especially when there is no intracranial involvement. Traditional open surgical approaches are comprised of transpalatal, transmaxillary (lateral rhinotomy or midfacial degloving), LeFort 1 osteotomy, and infratemporal fossa craniotomy.5 During the past 2 decades, the surgical approach to JNA has been evolving in relation to the improvement of endoscopic techniques. Compared to open surgery, transnasal endoscopic surgery has less morbidity, better amplifying visualization in identifying residual tumor behind corners or other inaccessible locations, lower intraoperative blood loss, less need for blood transfusion, less hospitalization, and a lower rate of recurrence.6–8 Recurrence in JNA is common and related to many factors. Radkowski found that the most important factor in determining the risk of postoperative recurrence was preoperative tumor stage.9 Reported recurrence rates vary from approximately 20% to over 50%.3,10 This article presents a retrospective, descriptive study of 162 patients with a diagnosis of JNA during the past 17 years. To the best of our knowledge, few articles have been published comparing open and endoscopic approaches for treating large, locally advanced JNA in fairly large numbers of patients. We report our surgical results with regard to Huang et al.: Surgical Management of Juvenile Angiofibroma
1
TABLE I. Radkowski Staging System for Juvenile Nasopharyngeal Angiofibroma. Stage
Ia
Limited to nose and/or nasopharyngeal vault
Ib IIa
Extension into one or more paranasal sinus Minimal extension into pterygopalatine fossa
IIb
Full occupation of pterygopalatine fossa with or without erosion of orbital bones
IIc
Extension into the infratemporal fossa or posterior to pterygoid plates Erosion of skull base—minimal intracranial extension
IIIa IIIb
Erosion of skull base—extensive intracranial extensionwith or without cavernous sinus invasion
tumor stage, surgical strategy, recurrence rate, and complications to evaluate the two techniques.
MATERIALS AND METHODS Patient Selection The medical records of patients with histologically proven JNA who received surgical treatment at Fudan University Eye, Ear, Nose, and Throat Hospital between September 1, 1995 and April 30, 2012 were reviewed retrospectively. Patients who received either open surgery or transnasal endoscopic surgery were selected, whereas combined-approach cases (patients who received open approach combined with transnasal endoscopic surgery) were excluded. Epidemiological and clinical data, surgical reports, pre- and postoperative images, details concerning complications, and follow-up information of selected cases were retrieved and analyzed. Radkowski’s classification9 (Table I) based on preoperative computed tomography (CT) and magnetic resonance imaging (MRI) was adopted. A standard follow-up protocol was instituted for all patients. Patients were required to visit the surgeon regularly (1 week, 1 month, 3 months, 6 months, 9 months, 12 months after surgery during the first postoperative year, every 6 months in the second year, and every 12 months afterward until the end of puberty), and undergo nasal endoscopy during each visit and CT or MRI every 6 months. In this study, recurrence was diagnosed as the finding of an active enhancing mass on endoscopy and/or imaging, and either surgery or radiotherapy was carried out. If follow-up imaging demonstrated that the mass was stable or showed regression, the patient was monitored every 6 months.
received surgery previously at other centers. Recurrent epistaxis and nasal obstruction were the most common presenting symptoms. Using Radkowski’s stage classification, we found 14 patients (8.6%) with Ia, 15 (9.3%) with Ib, 28 (17.3%) with IIa, 17 (10.5%) with IIb, 54 (33.3%) with IIc, 16 (9.9%) with IIIa, and 18 (11.1%) with IIIb disease. No patients had intradural invasion in our series. The median follow-up time was 55 months (range, 6–182 months). Six patients could not be contacted after discharge from our hospital and were lost to follow-up. Embolization, as a routine preoperative procedure from 2005 and onward, was performed uneventfully in 64 patients just 1 or 2 days before surgery and was carried out at Shanghai No. 9 Hospital. It was not until the year 2000 that transnasal endoscopic techniques were used to resect JNA at our hospital. Since then, endoscopic surgery has been performed increasingly, not only in cases extending to paranasal sinus and pterygopalatine fossa but also in cases in which infratemporal fossa and skull base were involved. In this study, 96 patients were treated by transpalatal or transmaxillary (lateral rhinotomy or midfacial degloving) approaches and are defined as group 1. The remaining 66 patients, defined as group 2, were treated through transnasal endoscopic approach with/without labiogingival incision. Staging of patients in the two groups was equally distributed (Table II). The median age was 17 years (range, 8–41 years) in group 1 and 16 years (range, 9–41 years) in group 2.
Recurrences According to Clinical Stage, Age, and Surgical Approach The overall recurrence rate was 31.4% (49/156) in our study. Median recurrence discovery was 16 months after the initial operation at our hospital. Patients suffering stage IIIb JNA, according to Radkowski’s staging system, had the highest recurrence rate (61.1%, 11/18), which was a significant difference compared to the other patients. The remaining patients’ recurrence rate almost reduced according to their disease stage (Table III). The odds ratio also confirmed that patients with other stage JNA were less likely to experience recurrence than patients with a stage IIIb lesion.
Statistical Analysis The influence of different factors on tumor recurrence was performed by Fisher exact test. Wilcoxon test and v2 test. Logistic regression was also performed to evaluate the effect of main factors (surgical approach and staging) on recurrence. Factor analysis of variance on log-transformed data for intraoperative blood loss was given. All statistical analysis for this article was generated using SAS/STAT software (version 9.2 of the SAS System for Windows; SAS, Inc., Cary, NC) and SPSS software (version 19 of SPSS for Windows; SPSS Inc., Chicago, IL).
TABLE II. Staging of Patients According to Radkowski’s Classification. Open Surgery (Group 1) Stage
Endoscopic Surgery (Group 2)
No. of Patients
Proportion
Stage
No. of Patients
Proportion
Ia
10
10.4%
Ia
4
6.1%
Ib IIa
10 22
10.4% 22.9%
Ib IIa
5 6
7.6% 9.1%
IIb
6
6.3%
IIb
11
16.7%
RESULTS
IIc IIIa
30 10
31.3% 10.4%
IIc IIIa
24 6
36.4% 9.1%
Patient Characteristics
IIIb
IIIb
10
15.2%
Total
66
100%
All patients were male subjects ranging in age from 8 to 41 years (mean, 17.5). Thirty patients (18.5%) had Laryngoscope 00: Month 2014
2
Total
8
8.3%
96
100%
Huang et al.: Surgical Management of Juvenile Angiofibroma
TABLE III. Recurrence Rate and OR According to Radkowski Staging System. Stage
No. of Recurrences
OR
95% Wald Confidence Limits
Ia
2/14 (14.3%)
—
—
—
Ib
1/14 (7.1%)
0.438
0.035
5.510
IIa IIb
7/28 (25.0%) 5/17 (29.4%)
1.907 2.908
0.338 0.452
10.768 18.706
IIc
15/50 (30.0%)
2.456
0.474
12.724
IIIa IIIb
8/15 (53.3%) 11/18 (61.1%)
6.267 9.807
0.989 1.578
39.722 60.956
developed laryngeal obstruction that needed tracheotomy, one patient complained of temporary aural fullness, and another patient complained of exotropia in the right eye because of dysfunction of the right medial rectus muscle. In the endoscopic surgery group, only one patient complained of epistaxis. The median intraoperative blood loss was 1,100 mL (range, 50–5,300 mL) in the open surgery group, compared to 800 mL (range, 50–5,000 mL) in the endoscopic surgery group, a statistically significant difference (P 5.017) (Table IV). The median amount of blood loss in the embolized group was 1,041 mL, compared to 1,249 mL in the nonembolized group (P 5.129).
Fisher exact test, P 5.0105 (
