RESUMO
OBJECTIVES/HYPOTHESIS: To evaluate the utility of positron emission tomography (PET)/computed tomography (CT) for staging the neck in the preoperative setting by comparing it to both CT/magnetic resonance imaging (MRI) and pathologic staging. STUDY DESIGN: Retrospective review. METHODS: Seventy-one patients with initial diagnosis of head and neck squamous cell carcinoma having preoperative MR or CT imaging and PET/CT with subsequent bilateral neck dissection as part of primary treatment (142 neck dissections). Comparison of cervical nodal metastases based on three separate staging systems: preoperative CT and/or MRI scan, preoperative PET/CT, and pathology. RESULTS: Both CT/MRI scans and PET/CT scans statistically predicted pathologic outcomes (P = 0.0001, P = 0.0001, respectively) using Chi square analysis. There was a statistically significant improvement in the prediction of true pathologic disease using PET/CT compared to CT and/or MRI alone (P = 0.005). In a subgroup analysis including only the contralateral neck, this significance persisted (P = 0.013). McNemar's test revealed that subsequent detection of bilateral neck disease by PET/CT scan was significantly superior to MRI or CT alone (P = 0.023). CONCLUSION: In pathologically positive necks, PET/CT was statistically more reliable at identifying positive disease than CT or MRI alone. Furthermore, such a result is achieved without a statistically significant difference in false positivity between PET/CT and CT or MRI. This suggests that PET/CT positivity, despite negative clinical exam and CT/MRI findings, may be more likely to signify pathologic disease and require appropriate treatment.