TMU Title Bar

May, 1998


Maximizing The Effectiveness Of Preoperative Autologous Blood Donations (PABD)

Joseph E. Kiss, M.D., Medical Director, Hemapheresis and Blood Services


Dramatic improvement in the safety of the allogeneic blood supply coupled with a rigorous emphasis on cost-containment in healthcare, have prompted a re-evaluation of the use of preoperative autologous blood donations (PABD). In this issue of Update, we will discuss approaches to employ PABD more effectively, primarily by reducing the collection and wastage of units which are unlikely to be transfused.

Autologous blood remains the transfusion product of choice in patients undergoing elective surgery with established transfusion needs, since it eliminates the risk of most adverse transfusion complications including viral disease transmission, alloimmunization, allergic reactions, and graft versus host disease. It is important to remember, however, that autologous blood is not entirely risk free. Clerical errors leading to a hemolytic transfusion reaction, and bacterial contamination resulting in sepsis, even death, have been reported as rare adverse effects. In addition, the medical condition of the patients in whom autologous donation is requested may place them at greater risk from the donation itself (e.g., a hypotensive vasovagal reaction in a patient with coronary artery disease). Since the overall goal of using autologous transfusion (pre-, intra-, or post-operative) is to avoid unnecessary exposure to allogeneic blood, it should be employed in conjunction with other strategies designed for this purpose, including the use of pharmacologic agents when appropriate, adequate attention to surgical hemostasis, and conservative transfusion practices.

Cost-Effectiveness Of PABD

A recently published bibliography summarizing 60 articles involving autologous transfusion practice listed 12 (20%) whose main focus was cost-effectiveness.1 Most of the studies concluded that PABD is less cost-effective relative to other medical therapies. For example, using a standard cost-effectiveness method, Etchason et al reported that the cost of PABD ranged from $235,000 to $23,000,000 per quality-adjusted life-year (QUALY) saved, depending on the medical procedure for which PABD was performed.2 This compares unfavorably with the cost-effectiveness of other medical therapies, such as coronary bypass surgery at $7,000 per year of life saved, or renal transplantation at $18,000 (in general $50,000/QUALY is considered to be a reasonable benchmark for "medical" cost effectiveness). However, it should be recognized that transfusion medicine is practiced in a setting that is inherently risk-averse, due, of course, to heightened public awareness of HIV and other transfusion- transmissible viruses. Does this mean that PABD is passť and should never be performed? No, but analyses like these are invaluable in bringing attention to areas of the autologous donation process which need to be changed in order to utilize this technique wisely, that is, in patients who will benefit.

Clearly, the most significant factor leading to lower cost-effectiveness is collection in patients who are not likely to require transfusion. From national surveys, the overall percentage of autologous units transfused/collected is reported to be approximately 50%. In our area alone, this translates into more than 4,500 discarded units per year. The units are discarded because many do not meet the rigorous collection standards required of allogeneic donors. One of the keys to optimizing the utilization of PABD lies in better selection of candidates for this procedure.

Surgical Blood Order Schedule PABD

Two approaches have been used to provide a better fit between collection and transfusion. The first employs the hospital’s surgical blood order schedule (also called the maximum surgical blood order schedule "MSBOS"), as a guide. The SBOS is derived from analyzing each hospital’s experience (using its own surgeons and patient mix) in supplying the red blood cells needs of at least 90% of the patients undergoing a specified surgical procedure. If a given surgical procedure, for example a total hip replacement, indicates an expected transfusion of 2 units of blood, then an autologous donation of two units would be appropriate. On the other hand, if the SBOS listed a type and screen requirement, this would indicate that a transfusion would be expected in fewer than 10% of patients undergoing the procedure. Autologous donation in this setting would be very unlikely to benefit the patient, unless his or her underlying medical condition increased the risk of hemorrhage. One additional point here is since the SBOS takes into account only the immediate operative period, the probability of transfusion during the post-operative period should also be considered in the autologous donation request. However, speculative or "insurance" donations should be discouraged.

Based on SBOS criteria, surgical procedures in which PABD is likely to be of benefit include: primary and revision hip replacement, major spine surgery with instrumentation, coronary bypass surgery, major vascular surgery, selected neuro- surgical procedures (e.g., resection of AVM), hepatic resection, and radical prostatectomy. Procedures unlikely to require PABD include: hysterectomy, vaginal and cesarean deliveries, transurethral resection of the prostate, and intervertebral discectomy.

Patient-Specific PABD

A limitation of the SBOS method is that all patients are treated alike, a "one-size-fits-all" approach that neglects important individual characteristics that may more accurately predict the need for transfusions during surgery. If one assumes a constant amount of blood loss for a specific surgical procedure, for example, the patient with a lower red cell mass (the product of hematocrit and blood volume) will be more likely to reach the minimum acceptable hematocrit requiring a transfusion. In a recent study of 299 patients undergoing total hip arthroplasty, significant indicators of RBC transfusion included preoperative hemoglobin level, weight, age, estimated blood loss, and use of aspirin preoperatively.3

Cohen and Brecher have gone a step further, devising a mathematical model which quantifies the relationship between estimated surgical blood loss (EBL) to reach the minimum acceptable hematocrit.4 A person with an average 5 liter blood volume and hematocrit of 45% would need to lose about 2300 milliliters, nearly half their blood volume, to reach a transfusion threshold hematocrit of 28%. Since EBLs on this scale are unusual for most surgical procedures, autologous donation in this patient would not be necessary in most instances. In order to avoid wastage, then, PABD should be reserved for patients with either lower initial hematocrits or very high anticipated blood losses, i.e., those with less intrinsic RBC reserve above the minimum acceptable hematocrit level. By comparing the expected surgical blood loss to the patient’s blood volume and hematocrit, then setting the minimum acceptable hematocrit (which may vary depending on the patient’s medical condition), the likelihood that a transfusion will be necessary can be determined. The need for autologous donation, along with other blood conservation strategies, can then be considered. If PABD is used, it is important to complete the donation sequence with sufficient time for red blood cell regeneration to occur (at least 1 week; preferably 2) before surgery.


Much of the cost-ineffectiveness of PABD stems from overcollection and wastage in patients unlikely to require transfusion. A strict SBOS approach, choosing to collect autologous units only for those surgical procedures with a "type and crossmatch" requirement ( > 1 unit RBC), can reduce the tendency to overcollect. A customized method based on individual patient characteristics may improve cost-effectiveness even further by targeting those patients who will obtain the most benefit from PABD. The Institute’s medical staff are available to assist with PABD planning and recommendations.


Stowell CP, Giordano GF, Kiss JE, et al. Transfusion 1998;38:400.

Etchason J, Petz L, Keeler E, et al. N Engl J Med 1995; 332:719.

Nuttall GA, Santrach PJ, Oliver WC, et al. Transfusion 1996;36:144.

Cohen JA, Brecher ME. Transfusion 1995;35:640.


Copyright © 1998, Institute For Transfusion Medicine



For questions regarding Maximizing The Effectiveness Of Preoperative Autologous Blood Donations (PABD), please contact
Joseph E. Kiss, M.D. at (412) 209-7326, or by e-mail:

Visit our web page for information and past issues of the TMU newsletter at

Copies of the Transfusion Medicine Update can be obtained by contacting Deborah Small at (412) 209-7320