ADDENDA Aug. 8, 2005

15. The transfusion service medical director at an urban hospital in Los Angeles is of the opinion that while some of the critical 'values' used by her transfusion service are not necessarily indicative of life-threatening condition or require immediate action, the laboratory's notification of the patient's physician (as well as the transfusion service director) serves as an educational exercise, as well as an alert of a potential severe situation. She reports that their critical value notification policy includes the following:

1. Incompatibility found in blood issued
2. Confirmed hemolytic transfusion reaction or suspected bacterial contamination
3. Transfusion/issue of ABO/Rh discrepant blood
4. Mislabeled sample-wrong patient drawn
5. Any ABO/Rh discrepancy not resolved technically
6. Antibodies known to cause hemolytic disease of the newborn in prenatal patients
7. Positive DAT in recently transfused patient or in neonate
8. Multiple clinically significant antibodies identified in patient's serum (implication being delay in issuing blood for transfusion)
9. No compatible units available for a specific patient
10. Positive fetal cell stain test

ADDENDA Aug. 10, 2005

16. According to Dr. Kathleen Sazama who practices at M.D. Anderson Cancer Center in Houston, Texas (attribution used with permission), the concept of critical value reporting arose from the practice of ordering 'panels' of tests (rather than medically indicated individual tests), creating the opportunity for an abnormal test result that the laboratory could not determine was related to a patient's medical condition. She points out that the definition of what is called a 'critical' (older term, 'panic') value is often misunderstood. In her opinion, one suitable definition is as follows: "A critical value is a clinically unexpected laboratory result that, if not quickly recognized and acted upon, may be life-threatening to the patient." Using this definition and strictly applying the concept of a "laboratory test result" to immunohematology, she argues that it is difficult to accept that anything other than reporting the presence of hemoglobin in plasma and/or urine (usually due to sepsis or a hemolytic transfusion reaction) and/or a new positive DAT because of a hemolytic transfusion reaction (and alerting staff to look at the surrounding events (e.g., sample misidentification, another unit mistransfused to a second patient, etc.)) would be critical values in immunohematology. She continues that "Granted there can be a number of clinically unexpected immunohematology test results -- new appearance of an allo- or autoantibody, incompatibilities at crossmatch, positive DAT on cord blood or positive fetal cell screens -- but either the action is managed entirely within the laboratory so that no 'life-threatening' patient events transpire OR the test was ordered for a specific clinical indication (e.g., fetal cell screening), so it cannot be considered 'clinically unexpected.' Physicians order individual tests because they are looking for abnormalities (or confirming an existing condition), so any single test result that is abonormal, even one that could be life-threatening to the patient, cannot be considered 'clinically unexpected.' When physicians order individual laboratory tests, they (and NOT the laboratory) are responsible for managing those test results, whether or not the result is a critical value." She adds that she is troubled by the definition's limitation to just laboratory testing. In the transfusion service, she believes it to be critical if there is an unforeseen delay in providing transfusions or if only incompatible RBC units are available for transfusion. However, definitionally, the test results indicated are probably the only ones that strictly meet the definition of 'critical values.'

ADDENDA Sept. 14, 2005

17. A transfusion medicine physician at a Northern California academic medical center reports that he and his associates use the following "critical values" for their transfusion service, which provides immediate notification of the following results:

1. a positive transfusion reaction work-up for hemolysis
2. an antibody panel evaluation for a type and cross request, when the work-up is inconclusive and cross match compatible blood is not available.
3. a patient with autoimmune hemolytic anemia, multiple alloantibodies, HTLA antibody, or antibody against a frequent antigen, and cross match compatible blood is not available
4. an ABO grouping result that indicates that a specimen is from the wrong patient

ADDENDA Sept. 15, 2005

18. A Blood Bank/Hematology Coordinator in Bryan, Texas reports that in reading the above discussion on critical values, she has NOT seen mention of a critical value that is on their local list. They call as critical for a pregnant patient any titer of 16 or above for an antibody known to cause hemolytic disease of the newborn (HDN). She justifies her position quoting the AABB Manual 14th Edition, p. 500-501 "The critical titer for anti-D should be established at each facility and usually is 16 or 32 in the antihuman globulin phase." "The critical titer for anti-K1 may be lower than anti-D, typically a value of 8." She adds that they also set "Alert Values" that are important but not life-threatening at the moment, and include in their list of Alert Values a positive antibody screen.

Please submit comments to the e-Network Forum.

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