The impact of PET/CT on the management of hepatic and extra hepatic metastases from gastrointestinal cancers

Abstract Purpose To investigate the efficacy of positron emission tomography/computed tomography (PET/CT) in detection and management of hepatic and extrahepatic metastases from gastrointestinal cancers. Materials and methods Between February 2008 and July 2010, patients histopathologically diagnosed with gastrointestinal cancer and showing suspected metastasis on CT screening were subsequently evaluated with PET/CT. All patients were subgrouped according to histopathological origin and localization of the primary tumor. Localization of gastrointestinal cancers was further specified as lower gastrointestinal system (GIS), upper GIS, or hepato-pancreato-biliary (HPB). Both accuracy and impact of CT and PET/CT on patient management were retrospectively evaluated. Results One hundred and thirteen patients diagnosed histopathologically with gastrointestinal cancers were retrospectively evaluated. Seventy-nine patients had adenocarcinoma and 34 patients other gastrointestinal tumors. Forty-one patients were in the upper GIS group, 30 patients in the HPB group, and 42 patients in the lower GIS group. Evaluation the diagnostic performance of PET/CT for suspected metastasis according to histopathological origin of the tumor, revealed that the sensitivity of PET/CT – although statistically not different – was higher in adenocarcinomas than in non-adenocarcinomas (90% (95% CI, 0.78–0.96) vs. 71.4% (95% CI, 0.45–0.88), P  = 0.86). The specificity was not significantly different (85.7% (95% CI, 0.70–0.93) vs. 85% (95% CI, 0.63–0.94), P  = 1.00). In the overall patient group; CT was significantly more sensitive than PET/CT for detection of hepatic metastases (94.7% vs. 78.9%, P  = 0.042), whereas PET/CT was significantly more specific than CT (48% vs. 98.7%, P P  > 0.05) but specificity was significantly higher in PET/CT than CT ( P In the overall patient group; for detection of extrahepatic metastasis, the sensitivity of CT (75%) and PET/CT (87.5%) showed no significant difference ( P  = 0.437). However, PET/CT was significantly more specific than CT (88.7% vs. 70.4%, P  = 0.007). In subgroup analysis, no significant difference was found between CT and PET/CT either in sensitivity or in specificity ( P  > 0.05). The specificity of PET/CT was highest in the lower GIS subgroup (93%). The management of 45 patients (39.8%) was revised after PET/CT evaluation. Conclusions PET/CT has a higher specificity than CT in detecting suspected hepatic and extrahepatic metastases of gastrointestinal cancers, and has an impact of nearly 40% on changing patient management strategies.