Correspondence
Cost Effectiveness of Autologous Blood Donation
N Engl J Med 1995; 333:461-463August 17, 1995
- Article
To the Editor:
The bleak conclusions that Etchason and colleagues offer about the limited usefulness and high costs of autologous blood donation (March 16 issue)1 seem premature and unwarranted. The ability to test and eliminate most donors with hepatitis B or C or the human immunodeficiency virus (HIV) is an important advance but does not ensure a completely safe blood supply. Just this year, new hepatitis viruses have been identified2; at least one is transmitted through blood transfusion (Alter HJ: personal communication). Many additional bacterial, viral, and protozoal infections are problems, and other complications of allogeneic blood transfusion are poorly understood. When available, autologous blood transfusion will always be a safer alternative.
At the heart of the problem of cost and documented by the analysis of Etchason et al. are attempts to make the procedures for autologous blood transfusion resemble those for allogeneic blood transfusion. One example is the cost of processing autologous blood, which is minimized when units are stored after their collection as whole blood, whereas allogeneic blood units must be separated into components (packed cells, plasma, and platelets), with additional labor and record keeping.3 The authors overlook many other costs of allogeneic blood transfusion that are not applicable to autologous blood transfusion, including donor-recruitment programs, donation-deferral registries, and maneuvers such as extended antigen typing, irradiation, and filtration for leukocyte reduction.
In the analysis by Etchason et al., a substantial part of the cost of autologous blood transfusion is associated with tests for diseases that can be transmitted by transfusion. The interim guidelines of the Food and Drug Administration (FDA) require such tests when autologous blood is shipped. Many hospitals have chosen not to perform these tests, because the health and safety of a donor are unaffected by the transfusion of untested autologous blood. Concern about the risk of inadvertent transfusion of an infected unit of autologous blood into an unintended recipient is overstated: even if 1 in 2000 autologous donors were infected with HIV,4 1 million units of autologous blood were collected each year, and the risk of the erroneous transfusion of an infected unit were as high as 1 per 80,000 collections,5 the chance of an error would be 1 in 160 million units, and such an error would occur only once every 160 years. The implementation of process control and other good manufacturing practices that the FDA now requires of blood banks makes the risk even lower.
A benefit of the article by Etchason et al. may be the attention it calls to the wastefulness of modeling autologous blood donation on allogeneic donation. I hope their data encourage the FDA to consider new, streamlined standards that will facilitate the use of autologous blood in appropriate situations.
5 ReferencesMargot S. Kruskall, M.D.
Beth Israel Hospital, Boston, MA 022151
Etchason J ,Petz L ,Keeler E , et al. The cost effectiveness of preoperative autologous blood donations. N Engl J Med 1995;332:719-724
Full Text | Web of Science | Medline2
Simons JN ,Pilot-Matias TJ ,Leary TP , et al. Identification of two flavivirus-like genomes in the GB hepatitis agent. Proc Natl Acad Sci U S A 1995;92:3401-3405
CrossRef | Web of Science | Medline3
Kruskall MS ,Yomtovian R ,Dzik WH ,Friedman KD ,Umlas J . On improving the cost-effectiveness of autologous blood transfusion practices. Transfusion 1994;34:259-264
CrossRef | Web of Science | Medline4
Kruskall MS ,Popovsky MA ,Pacini DG ,Donovan LM ,Ransil BJ . Autologous versus homologous donors: evaluation of markers for infectious disease. Transfusion 1988;28:286-288
CrossRef | Web of Science | Medline5
Linden JV . Autologous blood errors and incidents. Transfusion 1994;34:Suppl:28S-28S abstract.
Web of Science
To the Editor:
I was surprised that the risks associated with autologous blood donation were not mentioned in the article by Etchason et al. or in the accompanying editorial.1
About 3 percent of blood donations are complicated by vasovagal reactions, with severe bradycardia and undetectable blood pressure.2 Although these reactions are well tolerated by healthy donors, they may endanger those who do not meet the eligibility criteria for allogeneic donation, as in the case of most patients scheduled for coronary surgery. If we assume that 1 percent of vasovagal reactions in such patients would precipitate angina, acute myocardial infarction, or dangerous arrhythmia (which according to some studies3 is a very conservative estimate), we can infer that autologous donation carries a risk of acute life-threatening complications that is similar to the risks intended to be prevented by such a donation. Moreover, transmission of hepatitis C virus has little effect on the patient's survival, and other complications potentially prevented by autologous transfusion are extremely rare. The inclusion of the risks associated with phlebotomy in a quality-adjusted life-expectancy analysis such as the one performed by Etchason et al. would thus strengthen the argument against preoperative autologous donation, at least among patients scheduled for coronary surgery, without the inclusion of costs in the analysis.
3 ReferencesArturo Pereira, M.D.
Hospital Clínico, 08036 Barcelona, Spain1
Rutherford CJ ,Kaplan HS . Autologous blood donation -- can we bank on it? N Engl J Med 1995;332:740-742
Full Text | Web of Science | Medline2
Tomasulo PA ,Anderson AJ ,Paluso MB ,Gutschenritter MA ,Aster RH . A study of criteria for blood donor referral. Transfusion 1980;20:511-518
CrossRef | Web of Science | Medline3
AuBuchon JP ,Popovsky MA . The safety of preoperative autologous blood donation in the nonhospital setting. Transfusion 1991;31:513-517
CrossRef | Web of Science | Medline
To the Editor:
Etchason et al. cite “Bad Blood” as an example of “sensationalized reports of the dangers of blood transfusion . . . in the lay press.”1
This cover story in U.S. News and World Report involved more than 150 interviews, 16 requests under the Freedom of Information Act, and information from nine government data bases. We believe the article was fair and balanced. Our major finding, that transfusions are far riskier than patients are led to believe, is supported by many of the points raised by Rutherford and Kaplan in their editorial.
1 ReferencesAvery Comarow
, Washington, DC 20037-1196To the Editor:
Although I agree that preoperative autologous blood donation is not a cost-effective method to decrease perioperative transfusions, I do not agree with the statement in the editorial that preoperative hemodilution is an effective alternative. Several studies have shown that it is mathematically improbable that transfusion requirements will be reduced by using normovolemic hemodilution before surgery. These studies have demonstrated that multiple units of blood must be taken from a patient with a normal hematocrit in order to save the red-cell-mass equivalent of one unit of packed red cells.1-3 Clinical studies addressing this issue have been methodologically flawed. To date, there have been no well-controlled, randomized, prospective studies of patients whose blood loss exceeded the amount that would have required a transfusion in the first place. There is no solid clinical evidence on which to base a conclusion that preoperative hemodilution has any demonstrable value. Although this point remains somewhat controversial, the suggestion that the extension of the practice might be beneficial is not supported by the literature.
3 ReferencesDavid A. Lubarsky, M.D.
Duke University Medical Center, Durham, NC 277101
Brecher ME ,Rosenfeld M . Mathematical and computer modeling of acute normovolemic hemodilution. Transfusion 1994;34:176-179
CrossRef | Web of Science | Medline2
Goodnough LT ,Grishaber JE ,Monk TG ,Catalona WJ . Acute preoperative hemodilution in patients undergoing radical prostatectomy: a case study analysis of efficacy. Anesth Analg 1994;78:932-937
CrossRef | Web of Science | Medline3
Feldman JM ,Roth JV ,Bjoraker DG . Maximum blood savings by acute normovolemic hemodilution. Anesth Analg 1995;80:108-113
CrossRef | Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: Dr. Kruskall emphasizes the contribution of FDA-required testing to the costs of autologous blood donation in our analysis. We demonstrated the effect of eliminating such testing in our sensitivity analysis and agree that this important area of health policy deserves further attention. Dr. Kruskall is also concerned that we have omitted several factors. In developing our model, we excluded some factors that apply only infrequently to autologous donors (such as antigen typing) or are so rare that they would not affect our results (such as other infectious agents). We did include estimates of the costs of donor recruitment and deferral but excluded the costs of fractionating allogeneic blood into medically useful products, such as platelets and plasma. These valuable byproducts actually subsidize the social costs of obtaining allogeneic blood. We know of no evidence that the recently identified hepatitis virus, referred to by Dr. Kruskall, entails a new threat to the safety of the blood supply or a clinical outcome not already incorporated into our model.
Dr. Pereira correctly points out that the risk of donating autologous blood is not zero. He is also right to conclude that including this risk in our analysis would yield results even more unfavorable for autologous blood donation. Ignoring this risk was only one of several conservative assumptions in our model that favored autologous blood donation. Therefore, we believe that donation of autologous blood may actually be less cost effective than our estimates suggest.
Mr. Comarow objects to our characterization of his magazine's article as “sensationalized” and insists it is fair and balanced.1 If not sensational, an article entitled “Bad Blood” (in letters one third the length of the page), which includes statements, buried in 10 pages of text, acknowledging that the blood supply is safer than ever and that the risk of infection is “tiny,” certainly does not strike us as balanced.
In their editorial, Rutherford and Kaplan suggest that the “peace of mind” afforded by autologous donation is “immeasurable.” Our study demonstrates that the use of zero-risk transfusion strategies to attain peace of mind is enormously expensive and, given limited medical resources, may result in missed opportunities to provide peace of mind more effectively in other ways. Furthermore, we believe that the uncritical promotion of autologous blood donation sends the message to patients that allogeneic transfusions are not safe and erodes peace of mind for most transfusion recipients in this country who are not autologous donors.
1 ReferencesJeff Etchason, M.D.
Steven Kleinman, M.D.
University of California, Los Angeles, Los Angeles, CA 90024-1713Author/Editor Response
Dr. Lubarsky correctly points out that a number of earlier studies demonstrating reduced requirements for allogeneic transfusion with the use of acute normovolemic hemodilution were imperfect. The use of historical controls did not accurately reflect changes in decision points concerning transfusion or in surgical experience.
However, not all the studies of this procedure are similarly flawed. A prospective, well-controlled study by Ness et al.,1 involving 50 patients undergoing standardized, modified, radical suprapubic prostatectomy, compared acute normovolemic hemodilution with preoperative autologous donation. The study demonstrated that acute normovolemic hemodilution was similarly safe and effective in reducing requirements for allogeneic blood transfusion. The authors suggest that their observations are applicable to other elective procedures involving anticipated blood losses of 1000 ml or more. They note that because of the convenience of acute normovolemic hemodilution, as well as the lower cost and the shorter storage time required for blood components, this practice deserves wider application. Appropriate selection of patients, maintenance of normovolemia, and careful monitoring of patients have been cited as critical requirements for the safe use of this procedure.2
Of the three analyses that Dr. Lubarsky cites to indicate the improbability that acute normovolemic hemodilution substantially reduces requirements for allogeneic transfusion, two are mathematical models3,4 and thus subject to the constraints and assumptions inherent in modeling complex biologic systems. The third analysis5 is based on a three-year retrospective study in which only 4 percent of the cases (16 of 410) involved the use of acute normovolemic hemodilution.
Dr. Lubarsky's concern, however, demonstrates that questions about the effectiveness of acute normovolemic hemodilution remain. Further controlled, prospective clinical studies may extend and better define the circumstances in which acute normovolemic hemodilution can be used safely and effectively. Such studies may also provide further insight into the reasons for the discrepancies between clinical experience with acute normovolemic hemodilution and mathematical models of its use.
5 ReferencesHarold S. Kaplan, M.D.
Cynthia Rutherford, M.D.
University of Texas Southwestern Medical Center, Dallas, TX 752351
Ness PM ,Bourke DL ,Walsh PC . A randomized trial of perioperative hemodilution versus transfusion of preoperatively deposited autologous blood in elective surgery. Transfusion 1992;32:226-230
CrossRef | Web of Science | Medline2
Stehling L ,Zauder HL . Controversies in transfusion medicine: perioperative hemodilution: pro. Transfusion 1994;34:265-268
CrossRef | Web of Science | Medline3
Feldman JM ,Roth JV ,Bjoraker DG . Maximum blood savings by acute normovolemic hemodilution. Anesth Analg 1995;80:108-113
CrossRef | Web of Science | Medline4
Brecher ME ,Rosenfeld M . Mathematical and computer modeling of acute normovolemic hemodilution. Transfusion 1994;34:176-179
CrossRef | Web of Science | Medline5
Goodnough LT ,Grishaber JE ,Monk TG ,Catalona WJ . Acute preoperative hemodilution in patients undergoing radical prostatectomy: a case study analysis of efficacy. Anesth Analg 1994;78:932-937
CrossRef | Web of Science | Medline
- Citing Articles (4)
Citing Articles
1
Peter A. Lennox, Patricia A. Clugston, Michael E. Beasley, John Bostwick. (2004) Autologous Blood Transfusion in TRAM Breast Reconstruction:. Annals of Plastic Surgery 53:6, 532-535
CrossRef2
Eleftherios C. Vamvakas, Alvaro A. Pineda. (2000) Autologous transfusion and other approaches to reduce allogeneic blood exposure. Best Practice & Research Clinical Haematology 13:4, 533-547
CrossRef3
James Perkins, Lynne Kaminer, Margot Kruskall, Marie Cannon, Lynne Uhl, Walter Dzik, Herbert Silver, Mary O'Neill, Mark Popovsky, Karen King, Paul Ness, James AuBuchon, Arell Shapiro, Roslyn Yomtovian, Lawrence D. Petz. (2000) Should the FDA mandate that autologous units drawn and transfused within a single institution be tested for markers of infectious disease?. Transfusion 40:6, 752-753
CrossRef4
F.A. Sonnenberg, P. Gregory, R. Yomtovian, L.B. Russell, W. Tierney, M. Kosmin, J.L. Carson. (1999) The cost-effectiveness of autologous transfusion revisited: implications of an increased risk of bacterialinfection with allogeneic transfusion. Transfusion 39:8, 808-817
CrossRef
