BACKGROUND: In 2018, the Milan System for Reporting Salivary Gland Cytopathology was published defining a diagnostic categorization scheme with known malignancy risks and clinical follow up recommendations. Inter-observer reproducibility of the categories was not defined. METHODS: Salivary gland fine-needle aspirations (FNA) were reviewed over a 5 year period and classified by three independent observers. Inter-observer reproducibility was estimated using observed agreement and chance corrected agreement (Cohen's kappa). RESULTS: Four hundred and eight cases were reviewed with chance corrected agreement of 0.42 for the original Milan System and 0.48 for a modified system using "similar follow up" categories. Categories 4A and 6B showed substantial agreement (kappa =0.71 and 0.72). CONCLUSION: The Milan System shows moderate over all agreement between observers. Strongest inter-observer agreement was seen for categories 4A and 6B.
BACKGROUND: In 2018, the Milan System for Reporting Salivary Gland Cytopathology was published defining a diagnostic categorization scheme with known malignancy risks and clinical follow up recommendations. Inter-observer reproducibility of the categories was not defined. METHODS: Salivary gland fine-needle aspirations (FNA) were reviewed over a 5 year period and classified by three independent observers. Inter-observer reproducibility was estimated using observed agreement and chance corrected agreement (Cohen's kappa). RESULTS: Four hundred and eight cases were reviewed with chance corrected agreement of 0.42 for the original Milan System and 0.48 for a modified system using "similar follow up" categories. Categories 4A and 6B showed substantial agreement (kappa =0.71 and 0.72). CONCLUSION: The Milan System shows moderate over all agreement between observers. Strongest inter-observer agreement was seen for categories 4A and 6B.