Pitfalls in preoperative work-up of parotid gland tumours: 10-year series.

ABSTRACT

PROBLEMS/OBJECTIVES: Preoperative fine-needle aspiration cytology (FNAC) and magnetic resonance imaging (MRI) are the two most widely accepted diagnostic techniques used for the assessment of parotid gland tumours. We retrospectively evaluated the ability of FNAC and MRI to predict malignancy in parotid gland tumours. METHODOLOGY: Over a period of 10 years (2002-2011), parotidectomy for primary parotid gland tumours was performed in a consecutive series of 178 patients. Preoperative MRI was performed in 75% (133/178) of cases, and preoperative FNAC was performed in 70% of cases (124/178). Both modalities were applied in 53% (94/178) of patients. Sensitivity, specificity, and accuracy were analyzed retrospectively for each subgroup of patients. RESULTS: The sensitivity, specificity, and accuracy for predicting malignancy were 45%, 89%, and 84%, respectively, for FNAC (including only diagnostic cytology), and 40%, 88%, and 81%, respectively, for MRI. In the subgroup of patients who underwent both MRI and FNAC, sensitivity, specificity, and accuracy were 50%, 85%, and 80%, respectively. Preoperative MRI values improved significantly after introduction of diffusion-weighted (DW) acquisition in 2007 (71%, 93%, and 91%, respectively). CONCLUSIONS: Compared to previously published results, the high number of nondiagnostic smears and the low sensitivity rates in our series were disappointing, In part, this can be explained by the low percentage of malignant tumours and the high number of low-grade tumours among these. We discuss possibilities for improving preoperative performance, such as ultrasound-guided FNAC.