Ultrasonography (US) is very useful for detecting cervical neck metastases in head and neck cancer. Recently, FDG-PET is used to investigate cervical lymph nodes metastases.
We evaluated the accuracy of US and FDG-PET for identifying lymph node metastases with histopathologic correlation.
In US findings, lymph nodes were considered metastatic if their shortest diameter exceeded 7 mm in level I and II or 6 mm in level III, IV and V, and if the ratio of their shortest to longest diameters exceeded 0.5 concurrently. In FDG-PET findings, lymph nodes were classified as metastatic if their standardized uptake value (SUV) was greater than 3.0.
Cervical lymph nodes were removed by neck dissection from 31 patients with head and neck squamous cell carcinoma. Nine patients underwent bilateral neck dissection, so we evaluated 40 sides of necks in 31 patients. A total of 74 lymph nodes were diagnosed histopathologically as metastatic. Of the 74 metastatic lymph nodes, 36 (48%) were diagnosed correctly by preoperative US findings and 38 (51%) were diagnosed correctly by preoperative FDG-PET findings. The accuracy of US and FDG-PET was 87% and 92%, respectively.
The accuracy of clinical N staging by FDG-PET was superior to that by US, especially N1 and N2a were diagnosed correctly. Comparing clinical N stage with pathological N stage, 31 of 40 cases were correctly diagnosed by FDG-PET findings, while 27 of 40 cases were correctly diagnosed by US.
US and FDG-PET have the limitation that they cannot detect all histopathological metastatic lymph nodes preoperatively. US is very useful and can determine clinical N stage quickly, however, FDG-PET provides more accurate clinical N staging than US. If neck dissection will be done, it is desirable that both US and FDG-PET are done.
View full abstract