Resection of liver metastases from colorectal cancer: does preoperative chemotherapy affect the accuracy of PET in preoperative planning?

摘要

BACKGROUND: Preoperative scanning for hepatic colorectal metastases surgery remains a challenge, especially in the age of preoperative chemotherapy, which has marked biochemical and physical effects on the liver. Integrated fluoro-deoxyglucose positron emission tomography and computed tomography (FDG-PET/CT) has applications for detecting extrahepatic disease. The aim of the present study was to investigate FDG-PET/CT as a preoperative planning tool for detecting liver lesions in patients with and without preoperative chemotherapy. METHODS: Patients who had resection of hepatic colorectal metastases between January 2004 and June 2006 were included. Patients were divided into those who received preoperative chemotherapy and those who did not. Malignant hepatic lesions found on each scan were compared with those found on histopathology, intraoperative examination and/or intraoperative ultrasound. Accurate scans (scan lesions corresponded to true lesions), false positives (scan lesions detected at least one non-lesion) and false negatives (scan lesions missed at least one true lesions) were recorded. Results were also compared on a per lesion basis. RESULTS: A total of 21 patients had preoperative FDG-PET/CT scans with preoperative chemotherapy and 53 without. Accurate tests were six (29%) for the chemotherapy group versus 28 (53%) for the non-chemotherapy group (P= 0.06). Notably, there were 11 (52%) underestimations in the chemotherapy group versus 18 (34%) in the non-chemotherapy group. A total of 1.7 lesions were missed per patient in the chemotherapy group versus 0.7 in those who did not receive chemotherapy. CONCLUSION: Preoperative assessment with FDG-PET/CT is not useful for hepatic colorectal metastases, particularly when preoperative chemotherapy is used, with a trend towards underestimation of lesions.