Blood transfusions and survival after colectomy for colorectal cancer.

This study was carried out to determine the effect of perioperative blood transfusions on the survival of patients operated on for colorectal cancers. Cox's regression analysis was applied to 281 patients operated for cure of Dukes' stage A, B or C disease. Other variables studied were age, sex, tumour location, and preoperative hemoglobin, lymphocyte and albumin values. Perioperative deaths, pre- and postoperative immunodepression, neoplasia in situ, nonresections and stage D disease were excluded. It was found that the number of units of blood transfused had a strong influence on the prognosis of patients with colorectal cancer, particularly colonic cancers, but the effect could not be demonstrated when rectal cancers were studied separately, perhaps because of the small number of cases. The mechanism of action of blood transfusions seems to be independent of the other analysed variables. The authors suggest that perioperative blood transfusions may have an immunomodulatory effect in patients with colonic cancer, as already shown in recipients of transfused kidney allografts.