Survival improvement in patients operated on for colorectal cancer might be the consequence of early diagnosis, rather than improvement in the treatment and follow-up. Therefore the question is: how to follow-up patients operated on for colorectal cancer? And even: is it necessary to follow-up these patients? Follow-up has two intentions, to diagnose metachronous tumors (adenomas and cancers), and to reveal a relapsing disease. However more than a relapse, diagnosis of a "residual" disease is questioned. This residual disease was present after what was thought to be a "curative" treatment, but lack of sensitivity of our diagnosis tools delayed the diagnosis. Follow-up is worthwhile if prognostic of patient treated of their "residual" disease is better when diagnosis is made during this follow-up, rather than when it is postponed until the disease become symptomatic. This has never been proved. Exams which should not be performed anymore are: CA 19-9, liver tests, systematic CT scan (as a diagnosis exam, but CT scan is worthwhile when a curative treatment is expected), systematic repeated colonoscopy. Conclusions on follow-up proposed at a recent consensus conference reached consensus in that it justified exams which effectiveness was never proved....