Fine needle aspiration (FNA) cytology of salivary glands is a well known and established pre-operative technique that helps the clinical management of the different lesions. Recently a new system for reporting salivary gland lesions observed at FNA was proposed (Milan System). The aim of this study was to evaluate the diagnostic accuracy of salivary gland FNA comparing the system used in our hospital with the Milan System. A retrospective audit of FNA specimens of salivary gland lesions reported from 2011 to 2017 was performed as well as a correlation wıth the follow-up histology wherever it was available. The aspirates were previously categorized as: non-diagnostic, benign, neoplastic, atypical, suspicious for malignancy and malignant. The aspirates were then categorized according to the Milan system as follows: nondiagnostic, nonneoplastic, atypical, benign neoplasm, neoplasm of uncertain malignant potential (NUMP), suspicious for malignancy, or positive for malignancy. A total of 388 salivary gland aspirates were evaluated and classified according the old system as: non-diagnostic (n=28), benign (n=246), neoplastic (n=57), atypical (n=36), suspicious for malignancy (n=7) and malignant (n=14). According the Milan System the distribution was: non-diagnostic (n=28), non-neoplastic (n=89), atypical (n=39), benign neoplasm (n=156), NUMP (n=55), suspicious for malignancy (n=7) and malignant (n=14). When we applied the binary system considering only the benign and malignant cases, both classifications showed the same sensitivity (62.5%) and specificity (n=100%). The diagnostıc accuracy (95.8%), positive predictive value (n=100%) and negative predictive value (95.5%) were also similar. Salivary gland FNA has a high diagnostic accuracy and helps the clinical management independently to the report system used, however in difficult cases showing overlapping features the use of Milan System could be beneficial.