FINE-NEEDLE ASPIRATION BIOPSY OF PAROTID LESIONS - COMPARISON WITH FROZEN-SECTION

Background: Opinions vary as to the necessity for fine-needle aspirati on biopsy (FNAB) in parotid rumours. The present study reflects the ex perience gained over a 12-year period and shows the accuracy of a diag nostic FNAB, and improved results with experience. Methods: Between 19 83 and 1995, 201 parotid lesions were excised by one surgeon (AGP)and a prospective database was established. Fine-needle aspiration biopsy was performed prior to surgery in 195 lesions and frozen section was p erformed in 159 lesions. Results: The FNAB was diagnostic in 129 (66%) specimens and its sensitivity for malignancy was 90% and specificity was 100% (excluding non-diagnostic FNAB, where there was insufficient cellular material for reliable diagnosis or where specific tissue diag nosis could not be given). The positive predictive value was 100% and the negative predictive value was 98%. Of interest, the positive predi ctive value of diagnostic FNAB for pleomorphic adenomas was 99%. A spe cific tumour diagnosis could not be made on the FNAB sample in 37 spec imens; 11 of these lesions were histologically confirmed as malignant after excision. Frozen section was diagnostic in 144 specimens (91%). its sensitivity for malignancy was 96% and specificity was 99%. The po sitive predictive value was 96% and negative predictive value 99%. A s pecific tumour diagnosis could not be made on frozen section in 15 spe cimens, including six cases in which malignancy was finally reported. The positive predictive value of diagnostic frozen section for pleomor phic adenomas was 99%. Conclusions: A diagnostic FNAB is an accurate a nd useful tool in the management of parotid lesions. An FNAB diagnosis of pleomorphic adenoma obviates the need for frozen section. The perf ormance of FNAB in parotid rumours does not in any way preclude the ne cessity of surgical removal of such lesions except in exceptional circ umstances.