Value of positron emission tomography with [F-18]fluorodeoxyglucose in patients with colorectal liver metastases: a prospective study.

Purpose: To assess prospectively the value of fluor-18-deoxyglucose (FDG) positron emission tomography (PET), in addition to conventional diagnostic methods (CDM), as a staging modality in candidates for resection of colorectal liver metastases.

Patients And Methods: In 51 patients analyzed for resection of colorectal liver metastases, clinical management decisions were recorded after a complete work-up with CDM. Afterward, FDG-PET scans were performed and any change of clinical management according to FDG-PET results was carefully documented. Discordances between FDG-PET and CDM results were identified and related to the final diagnosis by histopathology, intraoperative findings, and follow-up.

Results: In 10 (20%) out of 51 patients, clinical management decisions based on CDM were changed after FDG-PET findings were known. FDG-PET detected unresectable pulmonary (n = 5) and hepatic metastases (n = 1) and ruled out extrahepatic (n = 2) and hepatic disease (n = 2). Due to FDG-PET, eight patients were spared unwarranted liver resection or laparotomy and two other patients were identified as candidates for liver resection. When the results of FDG-PET were regarded as decisive in a retrospective analysis, potential change of management was 29% (15 patients). FDG-PET and CDM showed discordant extrahepatic results in 11 patients (22%) and discordant hepatic results in eight patients (16%). Compared with CDM, FDG-PET resulted in true upstaging (n = 11), true downstaging (n = 5), false upstaging (n = 1), and false downstaging (n = 2). The detection rate of liver metastases on a lesion basis was generally better for computed tomography than for FDG-PET (80% v 65%); this was related to tumor size.

Conclusion: FDG-PET as a complementary staging method improves the therapeutic management of patients with colorectal liver metastases, especially by detecting unsuspected extrahepatic disease.