A member requested opinions about the appropriateness of different criteria (i.e., hemoglobin concentration, hematocrit, etc.) as triggers for allogeneic versus autologous transfusions. In addition, the member wanted to know what the e-Network membership experience has been with the proportion of autologous units transfused using similar or disparate criteria for allogeneic transfusions.

The following replies were submitted in response to this question:

1. We use the same level of hemoglobin as screening criteria for autologous and allogeneic RBC's. The level used is 10 grams per deciliter, for most of cases, however, one surgical subspecialty uses a higher level for autologous, and that level is 12.5 gm/dL. We are currently working on reducing this level to 10 or 11. This higher transfusion trigger criterion is still in place since we acquired this smaller subspecialty hospital. Our goal is to have the "trigger" set at 10, i.e., the point where a chart review is not initiated. We also review cases where RBC transfusions are NOT given, but the patient hemoglobin is below 7 gm/dL.

2. There should be absolutely no difference in criteria for the appropriateness of autologous vs. allogenic transfusion.

3. At our place the same criteria apply.

4. Editor's opinion: I do not agree that the exact same transfusion trigger criteria should be used for initiating the transfusion of autologous RBCs versus allogeneic RBCs. Rather, each patient should be assessed for their medical necessity to receive a transfusion, and that medical necessity should be weighed against the potential risk of them receiving the transfusion. Since it is usually less risky to administer autologous RBCs, one should not need to demonstrate as great a benefit for the transfusion in order for that benefit to outweigh any potential risk of the transfusion. Therefore, it seems logical to argue that for any given medical necessity of transfusion, a transfusion trigger to give autologous transfusion could be set at a higher hemoglobin threshold. It does not seem logical to insist on the exact same transfusion trigger for autologous blood transfusion as would be used for allogeneic blood transfusion. Having shared this opinion, perhaps additional input will be forthcoming?

ADDENDA Feb. 27, 2001

Here are some additional comments that followed the original posting. Some of these comments have been slightly edited for sake of clarity or anonymity:

4. Linda Stehling and Toby Simon dealt with this subject in a 'Question and Answer' format article that appeared in JAMA 1992;19:2669.

5. We share the opinion of the Editor. Our hospital uses different criteria for auto and allogeneic blood, exactly for the reasons described. Autologous RBC transfusions undergo chart review only if the hemoglobin is greater than 11 gm/dL.

6. I agree with the Editor's position. Different levels of risk suggest different criteria for appropriateness.

7. We use 10 g/dL and above to review allogeneic transfusions and 11 g/dL and above for review of autologous transfusions. The review includes not only the appropriateness of the transfusion but also a review of nursing procedures to satisfy JCAHO that a representative sample of cases are reviewed to show that hospital SOPs have been followed (I believe JCAHO refers to this function as blood transfusion management review), i.e., pre transfusion vital signs taken and recorded, post transfusion documentation of amount transfused, length of time of transfusion, nurse who identified patient prior to transfusion, etc. For the hospital struggling with the autologous review at the 12.5 g/dL level, let me suggest that it may be persuasive to mention that at the 12.5 gm/dL level the surgeon is transfusing a pint at a level blood banks routinely take a pint. It would seem that we don't have to transfuse a patient to the point that the patient becomes eligible as blood donor!

8. I believe that we should at try to persuade our clinical collegues to use the same criteria for transfusion of autologous blood as are used if only allogeneic blood is available. The transfusion of the autologous unit, while less risky than the transfusion of the allogeneic unit, still carries a certain risk. Clerical errors, multiplication of occult bacteria and increased cytokines are among these risks. It would likely serve the patient better if there was a review of the expected blood use for a particular procedure, and to avoid collecting autologous units unecessarily.

9. I agree that the patient is the one being transfused and so that should determine whether or not a transfusion is to be given, not some pre-conceived criteria. However, the indications should be the same for the transfusion whether you happen to have some autologous units or have allogeneic units set up for the patient. Autologous units are not without risk, even if they are lower than the quite low risks of allogeneic units today. We should get our surgical and medical colleagues to be consistent.

10. Although it seems to make sense that it should be less risky to transfuse autologous blood, I would like to see the data that support that belief. I have been led to believe that the highest risks by far are due to bacterial contamination and human (clerical, laboratory, nursing) mistakes. I believe that the likelihood of these risks should be identical for allogeneic and autologous units. Thus, in order to obtain the same risk:benefit ratio, one would need the same benefit (as determined for that individual).

11. I agree with the Web Master that the trigger for transfusion of an autologous unit can be liberalized somewhat. While it might be inappropriate (without mitigating circumstances) to transfuse an allogeneic unit to someone with a hemoglobin of 10 g/dL, giving back an autologous unit to someone with the same hemoglobin level would not be nearly as censurable.

ADDENDA Mar. 1, 2001

12. I don't believe that the transfer triggers for autologous and allogeneic units should be any different. Once the blood leaves the donor's body, it is now a 'product' and subject to contamination and human errors. Our efforts to teach clinicians to transfuse only when clinically indicated can be easily discounted if they see that we use different hemoglobin levels for review of transfusions of allo and auto units.

Please submit comments to the e-Network Forum.

Ira A. Shulman, MD
CBBS e-Network Forum Editor & Moderator