Purpose: To investigate the role of 18-fluorine-fluorodeoxyglucose positron emission tomography/computerized tomography ((18)F-FDG PET/CT) in the preoperative prediction of the presence and extent of neck disease in patients with oral/head and neck cancer.
Patients and methods: Seventy patients were enrolled in the study, 47 of whom had a clinically negative neck (N0), 19 of whom had a clinically positive unilateral neck (N+), and 4 of whom were negative on 1 side of the neck and positive on the other. Each patient underwent a PET/CT study before undergoing selective neck dissection for N0 disease or modified radical neck dissection for N+ disease. Tissues were submitted for histopathologic examination and were oriented for the pathologist as to the oncologic levels so as to permit correlation between histopathologic findings and the imaging results.
Results: The sensitivity and specificity of the PET/CT procedure were 79% and 82% for the N0 neck, and 95% and 25% for the N+ neck. One hundred ninety-two (11.4%) of the 1,678 nodes identified at histopathology were positive for metastases. The overall nodal sensitivity and specificity were 48% and 99%, respectively.
Conclusion: In patients with clinically negative necks, a negative test would not help the surgeon in the management strategy of the patient because of the rate of false-negative results, but a positive test can diagnose metastatic deposits with a high positive predictive value. In patients with clinically positive necks, a positive test will confirm the presence of disease, although false-negative lymph nodes were additionally identified in these clinically positive necks. With respect to nodes, the sensitivity of the imaging procedure is such that the results could not help the surgeon in deciding which level to dissect and which to spare. In the final analysis, the head and neck oncologic surgeon should not depend on the results of the PET/CT scan to determine which patients will benefit from neck dissection. Rather, time-honored principles of neck surgery should be followed, particularly with regard to the liberal execution of prophylactic neck dissections in patients with clinically N0 necks.