Two colleagues from California report that according to data released by the United States Food and Drug Administration, the number of fatalities due to ABO hemolytic reactions reported annually to that agency since 2003 has declined significantly. The Notification Process for Transfusion Related Fatalities and Donation Related Deaths can be found at: http://www.fda.gov/cber/transfusion.htm.

The data set below was provided by Dr. Leslie Holness of the FDA (data and attribution used with permission).

The California colleagues are interested to know to what factors others attribute this declining trend. For example, they wonder if it is it due to more widespread use of patient identification systems, less use of operating room refrigerators to hold blood for multiple patients, or some other factors.

The following comments have been received.

ADDENDA Dec. 13, 2005

1. A transfusion medicine physician in Maryland hypothesizes that many hospitals are paying more attention to the problem of mistyping a patient's ABO status due to incorrect specimen labeling. For example, she reports that her hospital has dramatically reduced the risk of ABO hemolytic reactions since implementing a requirement that a SECOND pre-transfusion sample must be collected to confirm a patient's initial ABO blood group result. Although some senior medical staff balked at this policy change, a timely front page article on the Metro section of the Washington Post featuring an ABO hemolytic fatality due to specimen mislabeling, and discussions by the e-Network Forum in which other hospitals reported implemention of a SECOND sample requirement, enabled her to get the initiative adopted by their medical staff. She acknowledges that in order to get the initiative passed, they did need to agree to a compromise that the attending staff could waive the SECOND sample requirement provided they took full responsibility for the consequences. However, not many physicians have asked for the waiver, unless the first result is blood group O.

ADDENDA Dec. 14, 2005

2. A transfusion medicine supervisor at an academic center in Northern California reports that her hospital has not had an ABO fatality (that she is aware of), but they have had close calls. She reports that in the past few years they have instituted the following:

• They discontinued using a stationary refrigerator in the operating room in favor of portable devices and coolers that must be checked out of the blood bank for each case.
• They now require a second blood specimen be submitted for ABO/Rh confirmation (if they do not already have a historical ABO/Rh for the patient) prior to transfusion. If they cannot get a second specimen, they issue group O red cells. This policy was unpopular for a while, but after approximately 6 months, the staff adjusted, and she claims it works well.
• They dispense only group O red cells to the ER for exsanguinating patients and for all "Doe" patients.

She laments that they do not yet have a barcoded patient ID system, but they are working on it.

3. A colleague in a Sunbelt state reports that her institution implemented a second specimen requirement for all non-group O patients for whom they have no prior blood group results. They implemented this policy in June 2005, just days after one of their patients experienced a non-fatal ABO hemolytic reaction. She acknowledges that they had the new policy "in the works, but that the adverse transfusion event pushed it to implementation". She is grateful for the new policy because they have prevented two ABO incompatible transfusions since its implementation. In one case a blood specimen was mislabeled at the time of phlebotomy and in a second case there was a technical error in the lab. She comments that the hospital staff has cooperated overall, but with pockets of resistance. She adds that they have budgeted for a bedside identification/specimen labeling system for calendar year 2006.

ADDENDA Jan. 2, 2006

4. A Transfusion Medicine Service Medical Director in Arizona acknowledges that he has had first hand experience with an ABO related death that occurred in an East coast hospital during 2003, which received high visibility in the US press. He reports that the error was committed by a floor-based phlebotomist who drew one patient but put someone else's label on the specimen. Unfortunately, the phlebotomized patient had no prior ABO/Rh on file against which the current sample's typing results could be compared. Remedial actions to the transfusion error included implementation of a pre-transfusion testing system that requires ABO/Rh testing to be performed on at least two different samples (in order to verify the patient's ABO/Rh). (For a more extensive discussion of patient ABO/Rh verification see: Should pre-transfusion ABO/Rh testing be performed twice, and if so, on different samples or by different personnel?) The hospital reportedly started to consider implementation of a barcode based identification system after the fatality, but the Arizona physician is not certain if such a system is in place. However, he does state that after sitting through more than 100 hours of vendor presentations on products that could potentially prevent mistransfusion errors from occurring, in his view an RFID tag system is a very promising way to address this issue.

ADDENDA Jan. 3, 2006

5. A Transfusion Medicine physician interested in RFID technology at an East Coast academic medical center would urge e-Network Forum readers not to declare success with ABO errors too soon. The two years of lower numbers at the FDA might reflect decreased reporting of events or may simply be a statistical blip. This physician feels that RFID may prove more "user friendly" than bar coding for bedside practice. Unfortunately, hospital administrators have not created much demand from industry for bedside safety technology. As a result, only a very few companies forecast a sufficient return-on-investment to want to develop bedside safety technology. A summary of these leading edge companies can be read at: Dzik WH. Technology for enhanced transfusion safety (Am Soc Hematol Educ Program) 2005; 476-82, full-text available for free via PubMed.

Please submit comments to the e-Network Forum.

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