Retrospective application of the Milan System for reporting salivary gland cytopathology: A Cancer Center experience

Background: The Milan System for Reporting Salivary Gland Cytopathology (MSRSGC) was recently proposed. Herein, we retrospectively applied this nomenclature system to salivary gland lesions sampled by ultrasound-guided fine-needle aspiration (FNA). Methods: All cases of salivary gland FNA with available surgical follow-up, in the period from 2014 to 2017 at our institution were reviewed and reclassified according to one of the six categories of the MSRSGC, blind to the surgical outcome. Overall sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated, as well as risks of neoplasm (RON) and risk of malignancy (ROM) for each of the proposed categories. Results: There were 104 salivary gland lesions, with a female predominance (57.7%), most cases from the parotid gland (89.4%). Mean age was 53.2 years. Distribution of the specimens according to the Milan System was as follows: 19.2% nondiagnostic (ND), 8.7% non-neoplastic (NN), 9.6% atypia of undetermined significance (AUS), 40.4% benign neoplasm (BN), 14.4% salivary gland neoplasm of uncertain malignant potential (SUMP), 1.9% suspicious for malignancy (SFM), and 5.8% malignant. Sensitivity, specificity, PPV, and NPV using MSRSGC were calculated as 75%, 98.4%, 88.9%, and 95.3%, respectively. RON/ROM for each category were 60%/15% for ND, 44.4%/0% for NN, 90%/40% for AUS, 100%/9.5% for BN, 100%/13.3% for SUMP, 50%/50% for SFM and 100%/100% for malignant. Conclusion: The use of the Milan System proved to be a useful method to predict the risk of neoplasm and malignancy in the sample studied, with high sensitivity and specificity.