Meta-Analysis of Published Data on Outcome of Endoscopic versus Open Management of Juvenile Nasopharyngeal Angiofibromas

Introduction: Juvenile nasopharyngeal angiofibroma (JNA) is a vascular neoplasm of the nasopharynx that presents in adolescent males. Open surgical removal in the past could entail significant morbidity. Recently, endoscopic resection has been reported with good results. The purpose of this study was to perform a comparative analysis of JNA patients in the reported literature who have undergone endoscopic resection versus open resection, to better understand differences in outcomes. We gathered multiple data points, to attempt to control for tumor characteristics and see if there were outcome differences attributable to surgical technique.

Methods: A PubMed search was performed of journal articles in English with the terms “nasopharyngeal angiofibroma” and “endoscopic.” We selected articles containing primary data, which reported both endoscopic and open surgical techniques, as well as series from the same authors who had used one of these techniques. Articles were examined for multiple clinical variables, including preoperative embolization, need for transfusion, postoperative complications, and recurrence. Tumors were categorized as “limited” or “advanced” disease using both Fisch/Andrews and Radkowski/Sessions tumor staging systems. We performed multivariate analysis, controlling for multiple variables, to best identify the independent effect of each independent variable.

Results: Thirty-eight articles were identified with a total of 383 patients treated for JNA. Of these, 259 underwent endoscopic resection, and 124 underwent open resection. In terms of tumor stage, patients with limited disease were more likely to undergo endoscopic resection than open resection (77.2% vs. 22.8%), whereas patients with advanced disease were more likely to undergo open resection (57.0% vs. 42.0%). Among endoscopic-only patients for whom complications data were available, 8 (3.7%) postoperative complications were reported (five synechia, two paresthesias, and one infraorbital nerve anesthesia). Among open surgery patients for whom complications data were reported, 19 (26.0%) experienced postoperative complications. Of 333 patients for whom tumor recurrence data were reported, 39 (11.7%) had recurrence. Twenty-one of 259 (8.1%) endoscopic-only treated patients had a recurrence, whereas 18 of 74 (24.3%) open patients had a recurrence (P < 0.001). However, multivariate analysis, controlling for tumor stage, revealed that the endoscopic approach was not a significant variable influencing tumor recurrence (P = 0.25). Only advanced tumor stage was an independent predictor of tumor recurrence (P = 0.02).

Conclusion: Endoscopic resection of JNA in selected patients appears to be a safe technique with a very low reported risk of morbidity, mortality, and tumor recurrence. We recommend that surgical approach be selected on an individual-case basis.