Background: We compared the usefulness of positron emission tomography with the glucose analogue 2-deoxy-2-[18F]-fluoro-D-glucose (FDG-PET) and multidetector-row computed tomography (MD-CT) in diagnosing pancreatic cancer and in determining the patients' suitability for surgery.
Methods: We reviewed the clinical FDG-PET data of 103 consecutive pancreatic cancer patients between July 2004 and March 2009.
Results: The detection rates of pancreatic cancer by MD-CT (89%) and FDG-PET (91%) were similar. From the MD-CT findings, 38 patients were judged as operable, and 65, inoperable. Among the inoperable patients, noncurative factors (metastasis to the liver, peritoneum, remote lymph nodes, bones, and other organs and major arterial invasion) were detected by MD-CT and/or FDG-PET. Detection rates of liver metastasis and arterial invasion by FDG-PET were significantly inferior to those of MD-CT (neither was detected by FDG-PET alone). Remote lymph nodes and bone metastasis were detected in 20 lesions by FDG-PET alone; however, MD-CT indicated other noncurative factors in these patients. All 65 patients could be diagnosed as inoperable without FDG-PET.
Conclusions: FDG-PET is not a suitable imaging modality for either diagnosis or preoperative treatment in pancreatic cancer patients. Since it is expensive, FDG-PET as a routine diagnostic tool in pancreatic cancer patients must be used with caution.