Salivary gland fine-needle aspiration cytology with the application of the Milan system for risk stratification and histological correlation: A retrospective 6-year study.
2020
BACKGROUND The Milan System for Reporting Salivary Gland Cytopathology (MSRSGC) is based on risk stratification. We presented our experience with fine-needle aspiration cytology (FNAC) for the diagnosis of salivary glands lesions by applying the MSRSGC categorization to the cytological diagnoses, and determined risk of malignancy (ROM) for each category. METHODS Fine-needle aspiration cytology of salivary gland lesions performed over a 6-year period was retrieved. FNAC results were retrospectively categorized according to the MSRSGC criteria, and correlated with corresponding histologic follow-up. ROM for each diagnostic category was calculated. RESULTS A total of 208 FNAC of salivary gland lesions were reviewed and retrospectively categorized as: non-diagnostic (ND) 23 (11%), non-neoplastic (NN) 54 (26%), atypia of undetermined significance (AUS) 10 (4.8%), benign neoplasms (BN) 77 (37%), salivary gland of uncertain malignant potential (SUMP) 13 (6.3%), suspicious for malignancy (SM) 7 (3.4%), and malignant (M) 24 (11.5%). Histopathological follow-up was available for 84 of 208 cases (40.4%). Overall concordance rate between FNAC and histology was 78.8%. Sensitivity, specificity, positive predictive value, and negative predictive value were calculated as 93.3%, 94.6%, 82.4%, and 98.2%, respectively. Diagnostic accuracy to distinguish benign from malignant disease was 94.4%. ROM for each category was ND 0%, NN 0%, AUS 75%, BN 2.2%, SUMP 28.6%, SM 50%, and M 100%. CONCLUSION Fine-needle aspiration cytology continues to be an accurate diagnostic tool for most salivary gland neoplasms showing classical morphologic features. However, difficult cases with unusual or overlapping features will occur. In these situations, the use of MSRSGC risk-stratification could be helpful to define appropriate management.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
28
References
4
Citations
NaN
KQI