A colleague in Southern California reports that at her institution they batch their ABO/Rh testing and antibody screens. If a patient needs a crossmatch, they "isolate" the specimen and then only crossmatch one specimen at a time. The inquiring colleague is not certain if this meets the recommendation of the JCAHO Sentinel Event Alert (Aug. 30, 1999) in which the JCAHO suggested "Prohibiting simultaneous crossmatching of multiple patients by the same technologist". She would like to know the practice of others and how this is accomplished at other institutions.

Editor's NOTE: To my knowledge the JCAHO has not made it a requirement that a blood bank laboratory prohibit simultaneous crossmatching of multiple patients by the same technologist. If I am incorrect, I urge any e-network colleague who is affiliated with the JCAHO inspection process (and there are JCAHO inspectors within the e-network forum) to please provide feedback, ASAP. From a practical standpoint, what would you do if two patients were both bleeding out simultaneously, but only one technologist was available to perform testing? Furthermore, in the opinion of the Webmaster, one of several effective defenses against an ABO-incompatible RBC unit being issued BY THE BLOOD BANK is for the blood bank laboratory to verify each patient's ABO group before issuing RBC units that are group A, group B or group AB. See this link to an earlier e-Network discussion about how to accomplish ABO verification of transfusion recipients.  Furthermore, probably the best way to prevent ABO-incompatible transfusions from occurring is to enforce strict patient and specimen identification procedures, since more than half of ABO-incompatible transfusions occur because of mistakes at the time of phlebotomy or at the time of blood administration (Reference: BPAC - transcript of meeting on 3-15-02 - PDF).

The following responses were received.

ADDENDA Feb. 17, 2003

1. Suzanne H. Butch, MA, MT(ASCP)SBB (attribution used with permission) of the Blood Bank and Transfusion Service at the University of Michigan reports that it is true that some hemolytic transfusion reactions have been associated with technologists performing testing on more than one person at a time. However, from her reading of what JCAHO published along with its recommendations (cited above), the facility that experienced a serious transfusion error following crossmatching more than one patient at the same time did so because it was not their usual routine for a technologist to crossmatch more than one specimen at a time. When the error occurred, they were probably not following a verified process. They were also probably in a rush. In the opinion of Ms. Butch, large transfusion services would grind to a halt if testing could only be performed on one patient at a time. If there ARE processes in place designed to prevent mixups, such as having at least two blood types of the patient on file and having a computerized system to detect incompatibilities, then the risk has been reduced to a level that should allow testing of multiple samples at one time.

Her point is that there are alternatives to the suggestions the JCAHO made in its alert. What fits in one facility, may not in another. Trying to reduce error by implementing unreasonable requirements will simply lead to other serious errors.

Finally, she addresses a recommendation for blood banks to use a special armband system, by asking "Since patient identification is important every time we take specimens and give treatment to a patient, should we not have a 'special' armband for preventing medication errors as well?" She does not believe in the "blood bank" armbands. Rather, she concludes "emphasizing patient identification for ALL processes will do more than implementing a SECONDARY armbanding system for blood bank or any other area."

2. According to a JCAHO inspector (who wished to remain anonymous): "Yes, the JCAHO in an alert several years ago advised against crossmatching more than one patient at a time by a single technologist. This was based on comments by hospitals who had had transfusion-related deaths. In reality the greater danger is at the typing step, when multiple patients are tested by a single technologist. Also this is the point where transfers are made into serum and cell dilution tubes (an easy place for error). The hospital in question should have made the lab aware of the ALERT, as well as the current JCAHO expectation on alerts, and more recent relevant JCAHO notices on safety items such as patient identification. It is NOT required on any ALERT to follow the JCAHO recommendation, but it is expected that if that issue is applicable to your hospital, that you have an acceptable, rational, alternative to deal with the issue. In this instance, one would expect the matter of checking prior blood types for corroboration has been resolved, before suddenly getting a crossmatch order. In emergencies, when in doubt, or with discrepancies, issuing group O RBCs is a way to buy time; with the time gained, retesting to determine the patient's true ABO is critical."

ADDENDA Jan. 16, 2007

3. A transfusion services technologist in Iowa recalls reading several years ago that the JCAHO suggested that simultaneous crossmatching of multiple patients by the same technologist should be prohibited. Her transfusion service is currently discussing the advantages of automated testing with their hospital administration. She believes that the JCAHO position supports their request to transition from manual testing to automation. She wonders how other institutions address the situation of simultaneous crossmatching of multiple patients by the same technologist.

Please submit comments to the e-Network Forum.

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

W. Tait Stevens, MD
CBBS e-Network Forum Assistant Editor & Moderator