Peri-operative blood transfusion increases length of hospital stay and number of postoperative complications in non-cardiac surgical patients

To test the hypothesis that blood transfusion alone was a significant risk factor for in-hospital morbidity in non-cardiac patients. Propensity analysis. University teaching hospital, Hong Kong. Consecutive non-cardiac patients seen in our department from 2006 to early 2009 who underwent a major procedure under general or spinal anaesthesia were included. Propensity analysis was performed to neutralise the confounding effects of preoperative variables and identify the true effects of transfusions on surgical outcomes. Receipt of intra-operative and postoperative blood transfusion was established and the difference in proportions between patients who did and did not receive donor blood tested for mortality, overall morbidity, individual complications, and number of adverse events. Transfused patients were significantly older and sicker, more likely to be male, to have lower haemoglobin values and undergo longer and more emergency surgical procedures than those not receiving a transfusion. Blood transfusion was predictive of length of postoperative hospital stay and number of complications before discharge. The amount of transfused blood was predictive of in-hospital mortality, with an odds ratio of 1.4 for each unit of blood received. The risk of a surgical wound infection was almost doubled when the patient had received a blood transfusion. After controlling for the factors associated with an increased likelihood for receiving a blood transfusion, the actual transfusion was predictive of a slower and more eventful postoperative recovery with associated costs to both the patient and health services.