Lincoln G L Tan1, Mark L C Khoo. 1. Head and Neck Service, Department of Otolaryngology, Tan Tock Seng Hospital, Singapore.
Abstract
INTRODUCTION: Identifying malignancy either preoperatively or intraoperatively can have a significant impact on the management of salivary gland tumours. We review our experience with fine needle aspiration cytology (FNAC) and frozen section (FS) for salivary gland lesions. We analyse the accuracy of both modalities and their influence on management. MATERIALS AND METHODS: Retrospective review of 114 patients who underwent salivary gland surgery, 91 with intraoperative FS and 68 with preoperative FNAC. Both sets of results were compared against each other and the final histopathological diagnosis. RESULTS: The accuracy of FS was 92.3%, with a sensitivity and specificity of 62.5% and 100%. Histologic concordance was 92.4% for benign lesions, and 100% for malignant tumours. The accuracy of FNAC was 89.7%, with a sensitivity and specificity of 100%. The non-diagnostic rate was 10.3%. Histologic concordance for FNAC was inferior to that for FS, with only 64.2% of benign lesions and 50% of malignant tumours correctly identified. FNAC did not alter the management of benign disease even when a correct diagnosis was obtained. CONCLUSION: Our results suggest that FNAC and FS are complementary in usefulness for malignant tumours. However, FNAC does not influence the management of benign lesions and routine FNAC for every patient may not be cost-effective.
INTRODUCTION: Identifying malignancy either preoperatively or intraoperatively can have a significant impact on the management of salivary gland tumours. We review our experience with fine needle aspiration cytology (FNAC) and frozen section (FS) for salivary gland lesions. We analyse the accuracy of both modalities and their influence on management. MATERIALS AND METHODS: Retrospective review of 114 patients who underwent salivary gland surgery, 91 with intraoperative FS and 68 with preoperative FNAC. Both sets of results were compared against each other and the final histopathological diagnosis. RESULTS: The accuracy of FS was 92.3%, with a sensitivity and specificity of 62.5% and 100%. Histologic concordance was 92.4% for benign lesions, and 100% for malignant tumours. The accuracy of FNAC was 89.7%, with a sensitivity and specificity of 100%. The non-diagnostic rate was 10.3%. Histologic concordance for FNAC was inferior to that for FS, with only 64.2% of benign lesions and 50% of malignant tumours correctly identified. FNAC did not alter the management of benign disease even when a correct diagnosis was obtained. CONCLUSION: Our results suggest that FNAC and FS are complementary in usefulness for malignant tumours. However, FNAC does not influence the management of benign lesions and routine FNAC for every patient may not be cost-effective.