HOMOLOGOUS BLOOD-TRANSFUSION IN PATIENTS WITH PROSTATE-CANCER - NO EFFECT ON TUMOR PROGRESSION OR SURVIVAL

Objective. To determine the effect of perioperative blood transfusions in patients with prostate cancer who underwent radical prostatectomy, we analyzed 1,785 patients with a follow-up of five years or more who were treated during a twenty-one-year period (1966 to 1987). Methods. Patients were divided into three groups according to the number of un its transfused during the perioperative period: group 1, 0 units (n = 440), group 2, 1 to 2 units (n = 746), and group 3, 3 or more units (n = 599). Results. With univariate analysis, no statistically significa nt differences were found among the three groups in overall survival r ate (71%, 75%, and 71% at ten years; p = 0.48), cause-specific surviva l rate (89%, 88%, and 86% at ten years; p = 0.36), or progression-free survival rate (61%, 68%, and 68% at ten years; p = 0.83). Adjusting f or tumor grade, pathologic stage, and hormonal therapy using the Cox s tatistical model, we found no significant association between the bloo d-use group and overall survival rate (p = 0.45), cause-specific survi val rate (p = 0.17), or progression-free survival rate (p = 0.34). The estimated relative risk and 95 percent confidence interval associated with blood transfusion (three or more units versus none) were as foll ows: 1.03 and 0.76 to 1.38 for total mortality, 1.56 and 0.95 to 2.56 for cause-specific death, and 1.20 and 0.91 to 1.57 for disease progre ssion, respectively. Conclusions. According to these findings, withhol ding homologous blood transfusion, except for infectious precautions, should not be based on the suspicion that it can accelerate death from cancer in patients who undergo radical prostatectomy for prostate can cer.