# Effectiveness of The Milan System for Reporting Salivary Gland Cytology: A 5-Year Retrospective Review

## TL;DR

This study evaluated the Milan system for reporting salivary gland cytology and found it to be highly accurate in distinguishing benign from malignant tumors.

## Contribution

The study provides empirical validation of the Milan system's diagnostic accuracy in salivary gland neoplasms.

## Key findings

- The Milan system showed 94.8% diagnostic accuracy in classifying salivary gland lesions.
- The system had 100% specificity and positive predictive value for malignancy.
- Risk of malignancy varied significantly across cytological categories.

## Abstract

Use of fine needle aspiration cytology (FNAC) in the diagnosis of salivary gland neoplasms is controversial due to the diverse morphologic patterns and overlapping features between benign and malignant lesions. The Milan system has been introduced to report salivary gland cytopathology. The present study aimed to reclassify salivary gland lesions according to the Milan system for reporting salivary gland cytopathology (MSRSGC) to determine the risk of malignancy (ROM) and to estimate the diagnostic accuracy of the Milan system.

In this retrospective cohort study, 136 salivary gland fine needle aspiration cytology samples taken from patients referred to Imam Khomeini and Amir-Aalam Hospital, Tehran, Iran, from 2016 to 2021, were retrieved along with the histopathological follow-up. Cytology smears were reviewed and reclassified based on MSRSGC. In addition, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and diagnostic accuracy were calculated.

ROM for each category was 26.7% for non-diagnostic, 12.5% for non-neoplastic, 40% for atypia of undetermined significance (AUS), 0 for benign neoplasm, 0 for salivary gland neoplasm with uncertain malignant potential (SUMP), 100% for suspicious for malignancy, and 100% for malignant group. Sensitivity, specificity, PPV, and NPV in differentiating benign from malignant neoplasms based on MSRSGC were 75.9%, 100%, 100%, and 93.8%, respectively. Diagnostic accuracy was calculated as 94.8%.

MSRSGC may be associated with a high accuracy in differentiation of benign from malignant salivary gland neoplasms, indicating its potential value as an effective classification system for reporting salivary gland cytology. The ROM for cytological categories except SUMP can be almost similar to that suggested by MSRSGC.