Our Reference Lab issues "historically antigen negative units" without retyping them. I am interested in learning how many of you reconfirm the antigen types when you the units in your Blood Bank. I know that it is double work and charges, but I worry that a clerical error could have occurred at the Reference Lab.

Fortunately, our reference lab confirms the historical antigen types. When we order antigen negative blood it is generally because we don't have the anit-sera on hand.

We confirm antigen typing @ our lab for all the units we received(antigen neg or historically antigen negative) from ref. lab.

I like to confirm ag neg units rec'd from the supplier.

We order only "FDA typed" units from our supplier. Being a small facility, we carry only a minimum inventory of typing sera.

When we request work to be done by the reference lab it is because we are unable to do it ourselves - most often because we lack the necessary antisera or reagents. Obviously, we are then unable to repeat their work for confirmation. We accept their work and make sure that the report clearly states which testing was performed by us and which was performed by the reference lab being sure to include the name and address of their facility.

We confirm all antigen neg. units regardless of if they are historical or not. We actually would rather have historical units because the additional charges are less.

Even though our supplier confirms historical types, we reconfirm everything for which we have antisera.

Does anyone know of any related rules/laws/'powers that be say so' items that relate here? Are you "allowed" to not retest if you have the capability? Rules that relate to "FDA typed" (does that mean unit tested by 2 techs every donation?) Just wondering.

Bob....John...... where are youuuuuuuuuu..............

The AABB Standards state (5.14.3) "When clinically significant antibodies are detected or the receipient has a history of such....products....for transfusion that do not contain the corresponding antigen and are serologically crossmatch compatible." There is no requirement that I am aware of to type units using a reagent antisera. For example, if you're able to demonstrate the reactivity of a particular antibody in the patient's serum/plasma (e.g., X+ reactivity with "single-dose" cells in your antibody identification studies), then the serologic crossmatch can also serve as the method to select units that do not contain the corresponding antigen. If your policies/procedures define this as your process, and you follow that process, you've met the intent of the standard. There is no requirement that I am aware of to repeat any "test of record" testing performed by another laboratory (other than the ABO/D-neg confirmationof red cell products) and I suspect you likely are not repeating the antibody identification performed by the reference laboratory, or other testing performed and labeled on a unit, e.g., CMV-testing, Hemoglobin S testing, infectious disease testing.......

I am aware that some suppliers provide products that are historically negative because the hospitals will type or test the units (see above logic) to demonstrate that the corresponding antigen is not present and this is less expensive to the hospital that requesting units antigen-typed by the supplier. The other scenario when this occurs is when anti-sera (licensed or un-licensed) is currently unavailable to test a unit for a particular antigen and historically negative is probably better than unknown. This commonly occurs for low frequency antigens.

There are requriements and then there are your choices beyond the requirements.....

All our antigen negative units are confirmed by two techs. Actually we use Gel to screen the donors and all our negative donors we confirmed by tube (and by second tech).

All antigen negative units we receive from supplier, we retpye them by tube method. we keep all regular antisera except some low frequency like anti-Kpa, anti-Cw etc.

We confirm units if the reference lab relied on historical types and we have the antisera. The chance of a historical record being incorrect bothers me as we have caught a few of those from different labs.

Our reference lab sends a written report with units that are historically antigen negative. There is no label or tie tag physically attached to the unit. Does anyone place a label, tie tag or write on the unit indicating that it is historically negative for an antigen and does anyone know if there are regulations against doing this? We do not repeat the antigen type and in most cases we would not have the anti-sera.

In my Ref lab we offer the clients choice... Antigen negative units (tested and tagged) and Hist Neg units. If the client hosp chooses Hist neg units... to save money... it is their responsibility to confirm hist neg typings. We do not confirm ag types on hist neg ... that would just be ag neg blood which the client must request.

Just a point to factor into your thinking.

I feel that many immunohaematologists do not appreciate how poor phenotyping is. The EQAP that is run from Australia and has participants acroos the Asia Pacific region shows that 24% of phenotypes are incorrect (more properly thought as discrepant when compared to the referrence value). I think that you will find that this is true worldwide. Discrepancies are due to about 6 factors including users not following instructions, variable reagents, weak phenotypes etc etc....

This of course does not apply to Le typing. A very recent survey shows 50% discrepant Lea reporting as the cells were Le(a+b+).

Well that is scary, TimOz, but not something that we can change easily at this point. Antigen typing is what we have (for those of us who stock the antisera). The molecular typing is more expensive and requires a set up we don't have available to us (last I checked). In addition, I don't know that it is any less susceptible to errors in handling, process, or interpretation.

Hi Adiescast,

I think it can be improved. So many people use poor phenotyping methods and perhaps more importantly they don't follow the instructions and don't understand reagent performance and some of the science. As an example, Le phenotyping is very dodgy and the monoclonal reagents on the market show significant variation in results and are very sensitive to test conditions like centrifugation and especially temperature.

The other phenotyping that is notoriously difficult are the MNSs antigens. Choosing a good reagent, using it correctly, controlling it properly every time it is used can significantly improve the test accuracy and precision.

We confirm the ABO. We do not confirm the antigen type from reference lab whether historical or actually tested. We sent it to the reference lab because we do not have the resources to do it in house. We contract with them to provide this service. Since we do the AHG crossmatch, we would catch a clerical error, if made.

Someday, we'll be doing molecular typing on each patient and donor............. until then.....

We used to re-confirm everything that we received, but years ago switched to the following policies:

-Unconfirmed units sent by the red cross do not have an attached tie tag when the unit is received. A fax is sent to us with the "historical" antigen typing results. Once we place these into our inventory, we place a label (created in house) that lists the negative antigens. It also states that these typing results have not been confirmed and are historical only. We use these units without retyping in-house for patients that have NOT formed the corresponding antibody (ie, phenotypically similar for our sickle cell patients- C, and/or, E, K neg). Any patient who has made the antibody will have a unit that has either been confirmed by the red cross, or that has been retyped in our BB.

-Confirmed units are sent with an attached tie tag from our supplier. We DO NOT reconfirm these units after we receive them. When we add the antigen negative info into our LIS, we add a comment "Antigen types confirmed by ARC" so that we can track who has performed the antigen typing.

We have the option of ordering units confirmed or unconfirmed - obviously confirmed is more expensive, so we try not to get these unless we do not have the antisera available to type them ourselves. Hope this helps!

Someday, we'll be doing molecular typing on each patient and donor............. until then.....

We used to re-confirm everything that we received, but years ago switched to the following policies:

-Unconfirmed units sent by the red cross do not have an attached tie tag when the unit is received. A fax is sent to us with the "historical" antigen typing results. Once we place these into our inventory, we place a label (created in house) that lists the negative antigens. It also states that these typing results have not been confirmed and are historical only. We use these units without retyping in-house for patients that have NOT formed the corresponding antibody (ie, phenotypically similar for our sickle cell patients- C, and/or, E, K neg). Any patient who has made the antibody will have a unit that has either been confirmed by the red cross, or that has been retyped in our BB.

-Confirmed units are sent with an attached tie tag from our supplier. We DO NOT reconfirm these units after we receive them. When we add the antigen negative info into our LIS, we add a comment "Antigen types confirmed by ARC" so that we can track who has performed the antigen typing.

We have the option of ordering units confirmed or unconfirmed - obviously confirmed is more expensive, so we try not to get these unless we do not have the antisera available to type them ourselves. Hope this helps!

Be careful for what you wish!

Molecular typing will give you a genotype, but is only predictive of the phenotype. This is particularly true of ABO.

:eek:

We currently do molecular typing... on all donors and some patients. Its saves time, but must be confirmed in order to be considered Antigen negative! (at least until it is licensed by the good ole FDA)

My lab does confirmation for all requested antigen-negative units from our supplier.

Although the possibility of a donr developing a new antigen during the course of his/her life is almost non-existent, I think I will "sleep better at night" if I do. =Þ (I don't think a bone marrow transplant receipient can be a blood donor)

It is important to remember that, while a donor is unlikely to develop a new antigen, the original typing could have been incorrect. Also, as I recall, there are conditions that can cause an antigen to weaken in expression at times. It never hurts to confirm the historical antigen type. I am sure none of you would transfuse a patient without confirming their ABO type, no matter how many times you have typed them before!

It is important to remember that, while a donor is unlikely to develop a new antigen, the original typing could have been incorrect. Also, as I recall, there are conditions that can cause an antigen to weaken in expression at times. It never hurts to confirm the historical antigen type. I am sure none of you would transfuse a patient without confirming their ABO type, no matter how many times you have typed them before!

I agree that typing may have been incorrect, but unless this is the case (or the donor has had a stem cell transplant), I cannot think of a situation where a donor may gain an antigen (unless you are counting something like T activation).

We (that is, the NHSBT - I am not involved personally) have been looking for the exposure of neo-antigens in blood that has passed through a prion filter and, as far as I am aware, none has been found.

I can certainly cite quite a few cases where an antigen has been weakened, even to the extent that an apparent alloantibody may be made against the antigen, but, except in the case of a Lewis antigen, such a situation only occurs, as far as I am aware, in a pathological condition. I have never heard of this happening in a healthy individual. Were it found that a healthy donor has "lost" an antigen, however, I cannot see how it is likely to be clinically significant for the recipient of that blood.

The NHSBT now garuarantees that the blood in a unit is of the ABO and D type advertised on the label (the same applies to other antigens printed on the label). In the case of a new donor, each antigen is tested twice under positive sample identification, and each antigen on the label is tested at least once on that actual donation on any regular donor.

As a result, very few of the hospitals we supply with blood now actually test the group of the unit.

Hi Malcolm,

Not that it is often clinically relevant but the exceptions to your statement above regarding development of "new" antigens do exist. I can point to Lewis and HLA (mainly (BGa) antigens. We sometimes see Aboriginal women who have phenotyped as Lea positive go Lea neg during pregnancy and sometimes they even make detectable anti-Lea. After delivery the antibody disappears and they revert to an Lea pos phenotype. They are of uncharacterised clinical significance but they are treated with suspicion. If reactive at 37oC they are crossmatched. I think this is fairly commonly seem in Polynesian populations.

Bga can also come and go for odd and often undefined reasons. For those that do not know Bga is a tissue antigen that is linked to the HLA B7 tissue type. It can appear to be expressed on red cells and it is characterised that infectious diseases (typically EBV) can cause increased red cell expression and an apparent change in phenotype from Gga neg to pos. In this case, there is no real change in tissue type, just a physiological event that increase the solubilisation of tissue antigen and increased adsorption of the antigen onto red cells (probably). This is of interest to me as we sometimes see it in healthy blood donors used for antibody screening cells. They donate for many years with no problem and all of a sudden for no reason become strongly BGa pos. Maybe just a case of the flu or a common cold could be causing this.

This is not generally an issue for antigen typing donor units - BUT BGa has been implicated in increased red cell destruction post transfusion and one case report of a transfusion reactions, but the evidence is slim.