Accuracy of intraoperative frozen section analysis of nasopharyngeal carcinoma resection margins

Background Frozen section analysis (FSA) is frequently used in salvage surgery for recurrent or residual nasopharyngeal carcinoma (rNPC) after radiotherapy to ensure adequate tumor removal. However, the diagnostic accuracy of FSA in rNPC has never been clearly established. We believe that this is the first study to specifically address these issues. Methods Patients with rNPC who underwent nasopharyngectomy in Queen Mary Hospital from 2006 to 2011 were identified. Clinical data, FSA results, and permanent histological results were analyzed. Results In the tissue-based analysis, the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were 70.6%, 100%, 100%, 95.2%, and 95.7%, respectively. Only 37% of inconclusive FSA turned out negative on permanent histology. Presence of inconclusive (p = .000) or positive (p = .000) FSA results in the same operation significantly lowered the NPV of FSA. Conclusion FSA is useful in ensuring clear resection margins for rNPC. Further resection is advisable in cases of inconclusive FSA results. © 2013 Wiley Periodicals, Inc. Head Neck 36: 638–642, 2014