Purpose: Survival of patients (pts) undergoing "curative" treatment for colorectal carcinoma remains poor. Retrospective studies suggest that F-18 FDG PET is more accurate than CT for restaging suspected recurrent disease and favourably impacts management. The aim of this study was to confirm this by prospective analysis.Patient Population and Methods: 102 consecutive patients for whom the referring clinicians had prospectively committed to a treatment plan based on conventional staging investigations and who then underwent F-18 FDG PET scanning were evaluated for management change and outcome. The accuracy of PET results was assessed by surgical findings or clinical follow-up.Results: Overall PET changed management in 66/102 (65%) pts including14 pts changed from observation to active treatment after PET localized disease and 6 pts planned for local therapy to observation after negative PET. Planned surgery was cancelled in 28 pts whereas surgery was initiated in 11 pts. Radiotherapy was prevented in 5 pts, initiated in 11 and the radiation field altered in 5 others. The accuracy of the PET results could be assessed in 63/66 (95%) pts in whom management was changed. Relapse was confirmed in 50/51 pts with a positive PET but disease extent was underestimated in 4 cases. False negative PET results were confirmed in 5 cases.Conclusion: FDG PET has a high and appropriate impact on patient management in patients with suspected recurrent disease. PET can miss small volume disease but the major benefit of PET is to prevent futile attempts at local salvage therapies.