Survival of patients evaluated by FDG-PET before hepatic resection for metastatic colorectal carcinoma: A prospective database study

Objective To present the survival results for patients with colorectal carcinoma metastases who have undergone liver resection after being staged by [F-18] flouro-2-deoxy-D-glucose-positron emission tomography (FDG-PET). Summary Background Data Hepatic resection is standard therapy for colorectal metastases confined to the liver, but recurrence is common because of the presence of undetected cancer at the time of surgery. FDG-PET is a sensitive diagnostic tool that identifies tumors based on the increased uptake of glucose by tumor cells. To date, no survival results have been reported for patients who have actually had liver resection after being staged by FDG-PET. Methods Forty-three patients with metastatic colorectal cancer were referred for hepatic resection after conventional tumor staging with computed tomography. FDG-PET was performed on all patients. Laparotomy was performed on patients not staged out by PET. Resection was performed at the time of laparotomy unless extrahepatic disease or unresectable hepatic tumors were found. Patients were examined at intervals in the preoperative period. Results FDG-PET identified additional cancer not seen on computed tomography in 10 patients. Surgery was contraindicated in six of these patients because of the findings on FDG-PET. Laparotomy was performed in 37 patients. In ail but two, liver resection was peformed. Median follow-up in the 35 patients undergoing resection was 24 months. The Kaplan-Meier estimate of overall survival at 3 years was 77% and the lower 95% confidence limit of this estimate of survival was 60%. This figure is higher than 3-year estimate of survival found in previously published series. The 5-year disease-free survival rate was 40%. Conclusions Preoperative FDG-PET lessens the recurrence rate in patients undergoing hepatic resection for colorectal metastases to the liver by detection of disease not found on conventional imaging.