Aspirates of eighty cases of salivary gland lesions with histologic correlation were reviewed along with the preoperative assessment in each case. Thirty-seven cases (46%) had frozen sections made. The eighty cases included 63 cases of benign
lesions
(45 cases of pleomorphic adenoma, 6 of Warthin's tumor, 2 of other benign neoplasms and 10 of nonneoplastic lesions) and 17 of primary salivary gland malignancy. The accuracies in distinguishing malignancy from other benign lesions were 87.5% for
fine
needle aspiration cytology (FNAC) and 92.5% for preoperative assessment. The accuracy of FNAC in diagnosing the specific type of lesions was 60%. The diagnostic sensitivity and specificity of FNAC for malignant lesions were 82.3% and 91.8%
respectively,
and improved to 88.2% and 93.6% at the stage of preoperative assessment. Several previously published studies of more than 100 cases were listed. A comparison of the cytodiagnosis and frozen section diagnosis (FS) was made. The accuracy of FS in
distinguishing malignant lesions was 83.7% and 78.3% was that of FNAC in the same cases. The accuracies in diagnosing the specific type of disease were 72.9% for FS and 45.9% for FNAC. The diagnostic sensitivity of FS for malignant disease was
45.4% and
specificity of FS, 100% (for FNAC, 72.7% and 80.7% respectively).
It would be advisable to work out more refined way to get better out of FNAC in that FS is not so accurate as we might expect to be and we can take a lot of advantage if we get the picture of histologic nature of the lesion preoperatively, though
FNAC
still has some problems and pitfalls and should be interpreted in the clinical context and reenforced by properly selected diagnostic imaging modalities and, if indicated, their FNAC adjuncts. (Korean J Otolaryngol 38:12, 1995)
|