Purpose: The aim of the study was to evaluate the diagnostic performance, the prognosis factors and the therapeutic impact of 18F-FDG positron emission tomography (FDG-PET) in the detection of recurrent colorectal cancers. Methods: Sixty PET/CT with 18F-FDG and CT were performed in 52 patients, at the Paul Papin cancer center between 2003 and 2005, following suspicion of colorectal cancer relapse. The FDG-PET impact on the clinical management was studied by examination of multidisciplinary concertations results. Survival analysis were realized with a mean follow up of 2.2 years. Results: Recurrence was confirmed for 50 explorations by histologic (n = 32), radiologic (n = 14) or clinical (n = 4) findings. Twenty patients died during the time of the study. On a patient based analysis, FDG-PET sensitivity, specificity and overall accuracy were 90, 90, 90% respectively compared with 74, 50 and 70% for CT. FDG-PET changed the clinical management in 18 cases (30%). A positive FDG-PET signal, more than one hepatic lesion, more than two lymph node lesions detected on FDG-PET and more than two hepatic lesions on CT were characterized as bad prognostic factors for survival. Multivariate analysis showed that the only independent bad prognostic factor was the FDG-PET detection of more than two liver lesions. Conclusion: These results confirmed the important impact of FDG-PET in the clinical management of patients with a suspected recurrence of colorectal cancer.