Diagnostic Accuracy of Fine-Needle Aspiration for Parotid and Submandibular Gland Lesions.

OBJECTIVE To assess the diagnostic accuracy of fine-needle aspiration (FNA) of the parotid and submandibular glands. STUDY DESIGN Case series with chart review. SETTING Tertiary referral academic center. SUBJECTS AND METHODS Retrospective analysis was performed for all parotid and submandibular gland FNAs from a single center from 2001 to 2014. There were 1283 FNAs performed for 1076 patients. Of these, 343 cases had surgical follow-up (parotid gland, n = 272; submandibular gland, n = 71). Cases were included where the patient had a preoperative FNA, followed by surgical excision. Correlation of preoperative FNA results to final surgical pathology was performed, with measures of diagnostic accuracy computed. RESULTS Malignancy was identified in 29.0% of parotid tumors and 42.3% of submandibular tumors, based on final pathology. FNA was nondiagnostic due to insufficient cellularity for evaluation in 22 of 343 cases (6.4%) and indeterminate in 39 of 343 cases (11.4%). Diagnostic accuracy in the parotid and submandibular glands for distinguishing benign from malignant pathology was determined as follows, respectively: sensitivity, 75.0% and 91.3%; specificity, 95.1% and 94.1%; positive predictive value, 84.9% and 91.3%; and negative predictive value, 91.2% and 94.4%. CONCLUSION FNA has high accuracy in identifying malignancy in parotid and submandibular gland lesions when performed at a high-volume center. Preoperative FNA results provide otolaryngologists with valuable diagnostic information that may influence the surgical management of salivary gland tumors. FNA, in conjunction with cross-sectional imaging, is useful in counseling patients for a complete informed consent.