Will Rogers revisited: prospective observational study of survival of 3592 patients with colorectal cancer according to number of nodes examined by pathologists.

摘要: To investigate the relationship between survival in colorectal cancer patients and number of lymph nodes examined by a pathologist, previously attributed to stage migration, we used data from cohort 5174 recruited September 1991 August 1994, followed-up for 5 years. We selected cases with present on all prognostic variables, stratified them into three groups examined. made multivariate comparison using Cox regression model. In all, there were 3592 variables. Patients who had >10 identified significant advantage over those 5-10 identified, turn similar 0-4 (P<0.001). This effect was whole group at Dukes' stages, although statistically only stages B (P=0.004) C (P=0.019). The remained after adjustment model which mean taken out each surgical firm did not predict survival. sub-group lymphocytic infiltration primary tumour noted prominent rather than mild (P<0.001): former also tended have more found (P=0.015). Stage migration alone cannot explain these results, as advantages are across population independent stage. Lymphocytic is prognostically important, associated found. Reactive enlargement mesentery may make easier find, reflect immune response tumour, thus indirectly impact upon