18-Fluorodeoxyglucose Positron Emission Tomography Does Not Aid in Diagnosis of Pancreatic Ductal Adenocarcinoma

Abstract

There are no accurate and reliable tools for diagnosis of early stage pancreatic ductal adenocarcinoma (PDA) or small metastatic lesions. It is also a challenge to differentiate PDA from focal mass-forming pancreatitis (FMP). There is controversy regarding the efficacy of 18-fluorodeoxyglucose positron-emission tomography (FDG-PET) in the diagnosis of PDA. We investigated whether FDG-PET provides information that, combined with data from other imaging techniques, can aid in decision making for patients with suspected PDA. We performed a retrospective analysis of data collected from 232 consecutive patients with suspected PDA at Kobe University Hospital from January 2006 through June 2012. All patients underwent a diagnostic imaging protocol that included multidetector row computed tomography, superparamagnetic iron oxide-enhanced magnetic resonance imaging, and FDG-PET. Based on endoscopic ultrasonography, fine-needle aspiration biopsy, or endoscopic retrograde cholangiopancreatography analyses, 218 patients had PDA (89 underwent resection and 129 did not) and 14 patients had FMP (8 had focal mass-forming chronic pancreatitis and 6 had focal mass-forming autoimmune pancreatitis). FDG-PET detected 50% of stages 0 and I, 91.9% of stage II, 100% of stage III, and 96.8% of stage IV tumors. Detection was affected significantly by tumor size (P = .024) and T stage (P = .023) in resected tumors. Multidetector row computed tomography detected significantly more liver metastases than FDG-PET. Few para-aortic lymph node or peritoneal metastases were detected by FDG-PET. FDG-PET correctly identified 11 of the 14 patients with FMP (5 of 8 with focal mass-forming chronic pancreatitis and 6 of 6 with focal mass-forming autoimmune pancreatitis). FDG-PET is not effective in detecting early stage PDA and small metastases, or in differentiating PDA from FMP. Combining FDG-PET with current diagnostic techniques for PDA did not provide any decisive information, therefore it should not be included in this analysis.