A retrospective six-year cytohistological correlation of fine-needle aspiration cytology with classification by the milan system for reporting salivary gland cytopathology

Abstract

Salivary gland fine needle aspiration (FNA) is invaluable pre-operative technique for clinical management of salivary gland lesions. However, precise and definite diagnosis can be challenging due to intratumoral heterogeneity and overlapping features among diverse entities. There was no standard classification for salivary gland FNA lesions. As a result, there are various reporting styles for salivary gland FNA with equivocal terminology. This can lead to miscommunication between clinicians and pathologists. Recently new classification system named “The Milan System for Reporting Salivary Gland Cytopathology” (MSRSGC) was proposed. Our retrospective study aims to evaluate the accuracy of FNAs of salivary gland lesions according to the MSRSGC and determine their associated risk of neoplasm and malignancy. A retrospective slide review and classification of salivary gland FNAs over 6 years (2013-2018) were done. All samples were processed with alcohol fixed smears with Papanicolaou stain. FNAs were classified according to the MSRSGC as follows: non-diagnostic (ND), non-neoplastic (NN), atypia of undetermined significance (AUS), benign neoplasm (BN), salivary gland neoplasm of uncertain malignant potential (SUMP), suspicious for malignancy (SM), or malignant (M). The risk of neoplasm (RON) and malignancy (ROM) were calculated for all diagnostic categories. A total of 374 FNA cases (371 patients) were performed for 6 years and 150 FNA cases had clinical and surgical follow up (40.1%). Among surgically treated cases, the distribution of each FNA’s category were as follows; ND (13.3%), NN (5.3%), AUS (6.7%), BN (50.0%), SUMP (18.0%), SM (4.0%), and M (2.7%). The overall risk of neoplasm (RON) was 70.0% for the ND, 62.5% for NN, 90% for the AUS, 100% for BN, 96.3% for SUMP, 100% for SM and M. The overall risk of malignancy (ROM) was 35% for ND, 25% for NN, 50% for AUS, 4% for BN, 33.3% for SUMP, 83.3% for SM and 75% for M. Compared to the ROM recommended by MSRSGC, ND and NN were slightly higher, and there was a significant difference in AUS. BN were within the range, SUMP and M were slightly lower, and SM was higher. Newly proposed MSRSGC is a useful system for classifying salivary gland lesions according to their associated risk of malignancy. Adoption of MSRSGC could potentially stratify ROM and remove vagueness among clinician and surgeon.