Success of salivary gland fine-needle aspiration cytology is depending on various lesion- and sampling-related factors

Abstract

Purpose. Ultrasound-guided fine-needle aspiration cytology (FNAC) is a widely used diagnostic procedure which facilitates the differentiation of salivary gland lesions. Although the performance of salivary gland FNAC (SG-FNAC) has improved since the introduction of the Milan System for Reporting Salivary Gland Cytopathology (MSRSGC), the range of the reported performance is still wide. Therefore, the aim of this study was to determine lesion- and sampling-related factors that influence the success of SG-FNAC. Methods. All SG-FNAC cases performed in a tertiary referral hospital between September 1st, 2011, and August 31st, 2022, were retrospectively identified. Demographic, histopathological, lesion-specific, and sampling-related data were retrieved from the clinical charts. Cytopathological reports were categorized according to the MSRSGC. The risk of malignancy (ROM), the performance measures, and factors influencing the success of SG‑FNAC were calculated. Results. Overall, 1,289 cases with histopathological follow-up diagnosis (out of 1,952 SG-FNACs) were included. The ROM was: non-diagnostic = 23.9%, non-neoplastic = 4.4%, atypia of undetermined significance (AUS) = 34.5%, neoplasm-benign = 1.0%, neoplasm‑salivary gland neoplasm of uncertain malignant potential (SUMP) = 15.3%, suspicious for malignancy = 74.1%, malignant = 96.2%. The sensitivity, specificity, accuracy, positive, and negative predictive value for differentiating benign from malignant lesions (excluding lesions categorized as AUS and SUMP) were 87.5%, 97.7%, 96.3%, 85.0%, and 98.1%, respectively. A larger lesion size (OR (95% CI) =1.21 (1.06-1.39), p = 0.004), a higher number of obtained slides (OR (95% CI) = 1.31 (1.17-1.46), p < 0.001), and the physician performing the FNAC (p = 0.047) were independent predictors for a higher success, while localization of the lesion within the submandibular compared to the parotid gland (OR (95% CI) = 0.38 (0.19-0.77), p = 0.008) was an independent predictor for lower success of SG-FNAC. Conclusion. This is the largest single-center study evaluating SG-FNAC performance to date. It identified independent lesion- and sampling-related factors influencing the success of SG‑FNAC. Knowledge of those can improve performance of the procedure.