e14519 Background: Liver resection remains the only potentially curative option for a subset of patients with colorectal cancer liver metastases (CRCLM). Preoperative imaging used to determine resectability includes computed tomography (CT), magnetic resonance imaging (MRI) and positron emission tomography (PET). The objective of this study was to determine the utility of PET scanning for potentially resectable CRCLM. Methods: We retrospectively reviewed a prospectively maintained database for all patients considered for resection of CRCLM from July 2010 to July 2012 in two specialist colorectal/hepatobiliary cancer centres. We extracted and analyzed data with respect to preoperative staging imaging and definitive treatment performed on completion of staging. Results: We identified 100 patients who underwent preoperative staging investigations for potentially resectable CRCLM. The imaging techniques performed included: CT (n=99, 99%), MRI liver (n=75, 75%), PET (n= 96, 96%). In 22 (22/96, 23%) patients PET scanning added to the preoperative staging information, identifying local recurrence (n=3, 3.1%), confirming liver metastases following an inconclusive CT/MRI (n=2, 2.1%), outruling liver metastases (n=1, 1%) and identifying extrahepatic sites (EHS) suspicious for disease (n=16, 16.7%). The EHS included either lung (n=6), bone (n=2), peritoneum (n=1) or lymph nodes (n=7). There were 2 false positive results. One patient with FDG-avid mediastinal lymph nodes had no cancer on endobronchial biopsy. One patient with FDG avidity at the primary anastomosis had no evidence of disease at colonoscopy. PET definitively changed the therapeutic strategy in 16 patients (16/96, 16.6%): precluding liver resection in 10 patients (10/96, 10.4%), leading to resection of extrahepatic disease in 4 patients (4/96, 4.1%), resection of local recurrence in 1 patient (1/96, 1%) and resection of hepatic metastases in one patient (1/96, 1%). Conclusions: In this small retrospective cohort the addition of metabolic imaging altered management in 16.6% of patients with potentially resectable CRCLM. There is a need for randomized evidence to support the routine use of PET in addition to cross-sectional imaging in this setting.
Resection after preoperative chemotherapy versus synchronous liver resection of colorectal cancer liver metastases
