A blood banker in New York is curious to know what is considered the best time frame to repeat an antibody identification panel workup for a patient who is known to be alloimmunized. The inquiring New Yorker wants to know if there are any criteria such as an increasing strength of reactivity in a screening test for unexpected antibodies, or if every alloimmunized patient should have their alloantibodies automatically reidentified after a certain time period?

Before sending this question to the e-Network, the input of a California transfusion medicine physician expert in pretransfusion testing was solicited, who said the following: "There are no requirements as to how often one must repeat an 'antibody panel workup' for a patient who is known to be alloimmunized. Obviously, if an alloimmunized patient is in need of transfusion, and the patient has a history of recent transfusion or pregnancy, the sample used for crossmatching must have been collected within three days of the scheduled transfusion, and the donor RBCs transfused should be crossmatched using an indirect antiglobulin method and with red cells that lack all the antigens (if possible) to which the patient's past or present clinically relevant antibodies are directed. The real question is to know when to search for ADDITIONAL antibodies that have not yet been identified. Many places routinely repeat panels every 3 days for their alloimmunized patients, some places do this every week or at other time intervals. At my institution we run panels of selected reagent red cells to identify new antibodies whenever a patient's antibody screen increases in strength or changes in pattern of reactivity, or if a patient has an unexpectedly positive crossmatch with an 'antigen'-negative donor RBC unit, or if a new alloantibody is suspected to be the cause of a transfusion reaction."

The following responses were received.

1. A blood banker in Texas reports that her hospital has arbitrarily set their time limit on panel repeats at 90 days provided the screening cells are reacting the same as they did when the panel was performed. If the reaction strength is less than the previous one, that is acceptable. If the reaction strength is stronger or if a screening cell reacts which contains antigens different from the one reacting with the previously identified antibody, then a panel is repeated.

2. A blood banker from a blood center in Northern California states that there is not a thing that needs to be added to the expert's response above. The rationale his/her facility uses (in deciding when to perform red cell antibody ID panels) is exactly the same as in his own Reference Lab (and that they ask their customer hospitals to use, as well).

ADDENDA May 2, 2002

3. A blood banker reports that his Transfusion Service is in a medium-sized institution with 450 licensed beds and approximately 8,000 RBC transfusions yearly. Once an alloantibody is identified, they do not rerun a panel to search for additional antibodies until suspicion of such is prompted either by the results of the test with antibody screening cells or that of an AHG crossmatch with an antigen-negative unit. He reports that in his 18 years of using this policy, they are not aware of any instance where they failed to detect an additional antibody nor have they had a delayed reaction due to further antibody development.

Editor's note: One has to wonder what monitoring systems are in place at the responding blood banker's hospital to detect delayed hemolytic reactions (DHTRs), given that DHTRs are not uncommon nor are they completely preventable, even when using highly sensitive pretransfusion testing techniques. The discussions at the following URLs might be of interest in regard to DHTRs:

Need to investigate delayed alloantibodies like an acute hemolytic reaction?

Experience using PEG versus LISS for antibody detection

Use of phenotype-matched donor RBCs for transfusion of patients with sickle cell anemia

ADDENDA May 5, 2002

4. A blood banker with a different opinion than those expressed by the expert and in replies 1-3 above says that it her understanding of the intent of AABB Standard 5.12.3.2 is that if a patient has been transfused or pregnant in the previous 3 months a sample must be obtained within the 3 days of anticipated transfusion. To quote her, "Standard 5.12.3.3 states that in patients with previously identified clinically significant antibodies, methods of testing shall be those that identify additional clinically significant antibodies. My understanding of this is that each time you perform pretransfusion testing (frequency within 3 days of transfusion as in 5.12.3.2 or for patient's not exposed recently to allogeneic red cells the facilitity-specific interval - usually 7 days to 1 month prior to anticipated transfusion) a selected panel should be performed containing cells antigen-negative for the corresponding antibody already identified in the patient's serum and positive for antigens that would detect additional antibodies. Scenario: a facility uses a 3-cell screen. Patient has anti-e that reacts 4+. Screen cell 1 and 3 are always e+. The patient has a newly formed underlying anti-Jka that is only reacting with Jk(a+b-) cells. Cell 2 is Jk(a+b+). The patient's auto is negative because it is 3 weeks post transfusion of Jka+ donor cells and all of the cells have been destroyed. The patient's serum is not icteric because the hemolysis occured slowly during the 1st two weeks and all evidence of hemolysis has disappeared, except for a 1.5 gm decrease in hemoglobin which can easily be explained by the fact that the patient appeared in the emergency room with vomiting of blood. The crossmatches with R2R2 cells are compatible because the antibody is only reactive with Jk(a+b-) cells and both donors crossmatched are Jk(a+b+). Those facilities who are using increased reaction strength, positive autocontrol, incompatible crossmatches and clinical signs of delayed reaction are not compliant with Standard 5.12.3.3."

Editor's NOTE: While the suggestion is a good one to routinely screen all patients who have existing alloantibodies with specially designed minipanels containing red cells that are antigen-negative for the corresponding antibody already identified in the patient's serum and positive for antigens that would detect additional antibodies, in my opinion it is extreme to suggest that the failure to do so is not compliant with AABB Standards. It would seem to me that a facility that performs additional antibody panel work to identify a new unexpected alloantibody based on increased or changing reaction strength of the antibody screen, the development of a positive autocontrol or direct antiglobulin test, the finding of incompatible crossmatches and the monitoring of patients for clinical signs of delayed reaction are compliant with the spirit of AABB Standards, because such a facility is in fact employing methods of testing that do detect and identify additional clinically significant antibodies. It would be very interesting to hear from AABB Standards Committee members regarding this issue.

ADDENDA May 8, 2002

5. An AABB Reference Lab Assessor who works in Boston offers the following opinion:

"The standards for reference labs do not require any further work UNLESS there is serological or clinical evidence of a change. The rationale behind this is that a crossmatch would catch incompatible units. I agree that the crossmatch could potentially miss an antibody that reacts only with cells with a homozygous expression. However, only one cell (heterozygous or homozygous) is required for ruling out, so it could be missed there as well."

6. W. John Judd, FIBMS, MIBiol, Professor of Immunohematology Department of Pathology, University of Michigan Medical Center gave permission to share the following report and data, which are germane to this discussion. John says that he agrees whole-heartedly with comments that an antiglobulin crossmatch is a method to exclude additional antibodies (as are screening cells that lack antigens to which the patient is known to have made antibodies). He is less "sold" on the value of perceived increases in reaction strength, given the subjectivity of grading reactions and the ability of personnel to reproduce their own gradings let alone demonstrate concordance with their coworkers.

Be that as it may, he submitted an abstract several years ago to the AABB on this topic, which was rejected since a couple of the reviewers felt it contained nothing new! He says that he never published the data, but can share it now with the e-network.

Of 960 antibody studies, 385 (40%) were repeat investigations. These revealed 19 new clinically significant antibodies in 16 patients: E(6); K(5); Jka(2); Fya(2); one each of Ce, c, CD, and S. In 8 patients, the new antibodies were evident by a two-cell sample screen. Crossmatch using IAT (XM-IAT) would have found the new antibodies in 7 of the 8 other cases if the donor cells carried the relevant antigens. The exception was an anti-S masked by warm autoantibody. Repeat studies also found 26 additional new antibodies in 25 patients. These included 10 nonspecific; 5 auto; 3 cold allo; 3 passive; 2 to low incidence antigen; 2 rouleaux; 1 enzyme-only C. They were found by the antibody screen in 5 cases, but only 2 additional new antibodies would have been found by XM-IAT.

By not doing any repeat antibody identifications unless there was a reason to do one, serologically or clinically (they rely on the physicians to tell them), they significantly reduced their workload. A couple of times a year they get 'blind-sided' patient in the OR and IAT crossmatches are incompatible; they simply issue the units that are compatible and initiate a work-up. Also, if known antibodies are such that greater than 90% of random units are incompatible, they run exclusion cells, or in the case of patients with warm autoantibodies, they do adsorption studies to rule out additional antibodies on a weekly basis (not every 3 days). John says that they have not killed anyone yet, at least that they know of, in 28+ years. He now thinks that he needs to resubmit the abstract!

Please submit comments to the e-Network Forum.

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