Aim: Although uptake of 2-deoxy-2-[fluorine-18]fluoro-D-glucose (FDG) in the liver is basically high, metastatic liver tumours are known to be positive on positron emission tomography (PET). The purpose of this study is to evaluate the diagnostic accuracy of FDG-PET in detecting hepatic metastases from pancreatic cancer.
Methods: Thirty-four patients with histologically proven malignant tumours of the pancreas underwent FDG-PET, computed tomography (CT) and transabdominal ultrasound (US). The findings of PET, CT and US were compared with the histopathologic findings at surgery or with clinical follow-up and the diagnostic accuracy of PET was evaluated.
Results: PET showed 28 regions of high FDG uptake (SUV = 3.3-9.1) in 13 patients. Twenty-six foci were metastatic lesions confirmed by surgery (n = 11), clinical follow-up (n = 10) or autopsy (n = 5). FDG-PET accurately differentiated seven metastatic lesions from cysts when the diagnosis by US or CT was unreliable because of the small size of the lesion. In two patients who had areas of high uptake in the liver, no metastases were detected by surgery. FDG-PET showed no areas of increased uptake in 21 patients. Surgery revealed no metastatic lesions in the liver in 20 of 21 patients, but a liver metastasis was found at surgery in one patient. The diagnostic accuracy of FDG-PET was 90%, which was comparable with that of US or CT.
Conclusion: Our data suggest that PET is reliable in detecting liver, metastases from pancreatic cancer.