Do Magnetic Resonance Imaging and Ultrasound Add Anything to the Preoperative Workup of Parotid Gland Tumors?

Abstract

To assess the additional value of magnetic resonance imaging (MRI) and ultrasound (US) to physical examination (PE) and fine needle aspiration cytology (FNAC) in the preoperative determination of the location and histology of parotid gland tumors. Prospectively, 99 patients with 77 benign and 22 malignant lesions were included; 82 underwent parotidectomy. FNAC was performed in 88 patients. On PE, the location of the tumor was predicted. Eighty-nine patients had MRI and 47 US. Various characteristics and the presumed diagnosis were assessed. Furthermore, the location of the tumor in relation to the facial nerve was predicted. The results were compared with the definite histology and the location during surgery. Of the MRI characteristics, incomplete demarcation from normal parotid gland tissue showed the highest positive predictive value (PPV) for malignancy of 0.48. Of the US characteristics, enlarged lymph nodes yielded the highest PPV for malignancy of 0.5. Cytology correctly predicted the benign or malignant nature of the tumor in all cases. Superficial location was well predicted on PE with a PPV of 0.8, slightly better on MRI (PPV of 0.87), and worse on US (PPV of 0.7). FNAC is the only accurate investigation for classifying a parotid gland tumor as benign or malignant. Palpation and MRI are superior to US in predicting tumor location. Because the results for palpation and MRI are almost equal, MRI should only be reserved for specific cases, and not routinely requested.