The effect of perioperative blood transfusion on the survival rate of patients with gastric cancer was studied. One thousand patients with primary gastric cancer, who had curative surgery performed at the National Cancer Center Hospital in Japan from 1976 to 1981, were studied retrospectively. Overall comparison of transfused (n = 371) versus nontransfused (n = 629) patients by log rank analysis revealed a statistically significant adverse influence of blood transfusion on survival (p = 0.0001). Fifty-seven percent of transfused as compared to 80.8% of nontransfused patients survived for 5 years or more; however, after stratifying patients into stages and applying proportional regression analyses, blood transfusion did not appear to have any effect on prognosis: relative risk ratio, 1.16; p = 0.28. Similarly, comparison of patients transfused with more than 600 cc of blood and those transfused with 600 cc of blood or less revealed no statistical difference in survival time. It is postulated that the possible adverse influence of blood transfusion on the survival of patients with gastric cancer is linked to other prognostic features rather than to the immunologic sequelae of the transfusion itself.