Several reports suggest that perioperative blood transfusion promotes recurrence and death after cancer surgery. We studied the effect of transfusion in 1599 patients who had mastectomy for carcinoma of the breast. In 228 transfused patients the disease recurred in 110 (48 per cent), compared with 647 (47 per cent) of 1371 non-transfused patients (P = 0.85). The 5-year recurrence-free survival rate was 0.54 in the transfused patients and 0.60 in the non-transfused group (P = 0.44). The transfused patients had a higher prevalence of risk factors than the non-transfused patients. Cox's multiple regression analysis confirmed that perioperative blood transfusion was of no importance. The study does not support the hypothesis that perioperative blood transfusion promotes recurrence after operation for breast cancer.