OBJECTIVES: To determine the accuracy of using surrogate anatomic structures radiologically to predict the relation of parotid lesions to the intraparotid facial nerve. SETTING: Tertiary centre. DESIGN: Retrospective. PATIENTS AND METHODS: All patients with parotid masses over a 5-year period who undertook parotidectomy were considered. A radiologist and an otolaryngologist reviewed the images. Their decision regarding the location of the lesions using four surrogate structures was compared with intraoperative documentation. OUTCOME MEASURE: We determined the sensitivity and the specificity of using the external carotid artery, retromandibular vein, posterior belly of the digastric muscle, and tragal pointer. RESULTS: Thirty films were examined (24 magnetic resonance images [MRIs] and 6 computed tomographic [CT] scans). The sensitivity and the specificity of the retromandibular vein were 0.85 and 0.57, respectively, whereas for the external carotid artery, they were 0.94 and 0.3, respectively. It was too impractical to relate the other two structures to the lesions. CONCLUSIONS: The retromandibular vein is the most accurate surrogate structure to use on MRI or CT for predicting the location of a parotid lesion to the facial nerve. However, the substantial proportion of deep lesions misjudged limits the benefit of performing the imaging.
OBJECTIVES: To determine the accuracy of using surrogate anatomic structures radiologically to predict the relation of parotid lesions to the intraparotid facial nerve. SETTING: Tertiary centre. DESIGN: Retrospective. PATIENTS AND METHODS: All patients with parotid masses over a 5-year period who undertook parotidectomy were considered. A radiologist and an otolaryngologist reviewed the images. Their decision regarding the location of the lesions using four surrogate structures was compared with intraoperative documentation. OUTCOME MEASURE: We determined the sensitivity and the specificity of using the external carotid artery, retromandibular vein, posterior belly of the digastric muscle, and tragal pointer. RESULTS: Thirty films were examined (24 magnetic resonance images [MRIs] and 6 computed tomographic [CT] scans). The sensitivity and the specificity of the retromandibular vein were 0.85 and 0.57, respectively, whereas for the external carotid artery, they were 0.94 and 0.3, respectively. It was too impractical to relate the other two structures to the lesions. CONCLUSIONS: The retromandibular vein is the most accurate surrogate structure to use on MRI or CT for predicting the location of a parotid lesion to the facial nerve. However, the substantial proportion of deep lesions misjudged limits the benefit of performing the imaging.