Autologous blood donation

For patients subjected to more aggressive phlebotomy (up to 2 units weekly), the endogenous EPO response is more substantial [8–11]. In one clinical trial [9], a linear-logarithmic relationship was demonstrated between change in hemoglobin level and the endogenous EPO response [12]. EPO-mediated erythropoiesis in this setting is 397–568 ml (19–26% RBC expansion) [8–11], or the equivalent of 2–3 blood units. When recombinant human EPO therapy is administered during PAD, the equivalent of 5 blood units is generated [10, 13].

Preoperative autologous collections are most beneficial to those patients who are undergoing procedures with substantial anticipated blood loss, such as orthopedic joint replacement, vascular surgery, cardiac or thoracic surgery, and radical prostatectomy. Autologous blood is unnecessary for procedures that seldom require transfusion, such as transurethral resection of the prostate, cholecystectomy, herniorrhaphy, vaginal hysterectomy, and uncomplicated obstetric delivery [14]. A hospital's maximal surgical blood order schedule for blood cross-match can provide estimates of transfusion rates for specific procedures; the generally accepted cutoff at which transfusion is 'unlikely' and autologous blood procurement should not be recommended is 10% [15].

Collection of units should be scheduled as far in advance of surgery as possible for liquid blood storage (up to 42 days), to allow compensatory erythropoiesis [5] to correct the induced anemia. If the erythropoietic response to autologous blood phlebotomy is not able to maintain the patient's level of hematocrit during the donation interval, then the pre-deposit of autologous blood may actually be harmful. A study of patients undergoing hysterectomy [16] found that PAD resulted in perioperative anemia and an increased likelihood of any blood transfusion.

Even though national trends indicate a decline in PAD for all surgical patients in the USA, this practice remains a standard of care for patients undergoing total joint replacement surgery. A multicenter retrospective audit of 9482 patients undergoing these procedures [17] found that 60% underwent PAD. For non-anemic patients, PAD reduced allogeneic blood exposure by two thirds as compared with patients who did not undergo PAD. For anemic (hemoglobin <13 g/dl) patients, PAD reduced allogeneic blood exposure by only one third.

For procedures such as total joint replacement surgery, discard rates of up to 50% of collected units are common [17]. When autologous blood is collected for procedures that seldom require transfusion, such as vaginal hysterectomies, up to 90% of units collected for these procedures are wasted [16]. The additional costs associated with the collection of autologous units and the inherent 'wastage' of these units, along with advances in the safety of allogeneic blood, now render the pre-donation of autologous blood poorly cost-effective [18]. Cost-effectiveness models serve to illustrate the potential risks associated with autologous blood donation; even a very remote risk for death in patients with ischemic heart disease may entirely negate the benefits of having autologous blood available before coronary artery bypass grafting [19]. Key factors include the estimated postoperative lifespan of the patient and the likelihood of transfusion [20, 21]. In a study of autologous blood donation before coronary artery bypass grafting [19], the preoperative donation of 2 units was estimated to have a cost of US$500,000 per quality-adjusted life year. In comparison, most commonly accepted medical and surgical interventions have a cost of less than US$50,000 per quality-adjusted life year. The risk for exposure to a hepatitis virus or to HIV has declined by at least an order of magnitude since the calculation of this estimate, and the current cost-effectiveness would be significantly worse.

Some suggestions to make autologous blood programs less costly include abbreviating the donor interview for autologous collection, utilizing only whole blood and discontinuing component production, limiting the use of frozen autologous blood, applying the same transfusion guidelines for autologous and allogeneic blood, and testing only the first donated autologous blood unit for infectious disease markers. Attempts to stratify patients into groups at high and low risk for transfusion, based on the baseline level of hemoglobin and on the type of procedure, show some promise. In a study using a point score system, 80% of patients undergoing total joint replacement procedures were identified to be at low risk (<10%) for transfusion, so that autologous blood procurement for these patients would not be recommended [22].

Increased attention to the costs and safety of health care delivery has caused the relative benefits and costs of both blood transfusion and conservation to be scrutinized. The prospective identification of surgical candidates who will need transfusion and will therefore truly benefit from blood conservation must be based on factors specific to the patient, such as the baseline hematocrit and the anticipated blood loss during surgery. The decision to employ blood-sparing technology may no longer be based on the safety of the blood supply, but on evidence that blood conservation is safe and of value for individual patients.