Adverse effect of blood transfusions on patient survival after resection of rectal cancer.

Studies have shown that the number of units of blood transfused perioperatively in patients operated on for colonic cancer has a progressively strong negative influence on survival. The present study, involving 198 patients with rectal cancer, was done to determine if perioperative blood transfusions had any prognostic significance. Multivariate regression analysis was applied to these patients, operated for cure of Dukes' stage A, B or C disease. Other variables analysed were age, sex, preoperative hemoglobin, albumin and lymphocyte values and the timing of transfusion. Perioperative deaths, pre- and post-operative immunodepression, neoplasia in situ, nonresections and stage D disease were excluded. It was found that the number of units of blood transfused perioperatively had a negative effect on patient survival, that was independent of the other analysed variables. It is suggested that blood transfusion perioperatively exerts an immunosuppressive effect on patients with rectal cancer.