Gel Forward and reverse discrepancies

[PeterD]

I am interested to hear how institutions that use gel card testing for ABO and Rh testing deal with the apparent higher frequency of ABO forward and reverse grouping discrepancies encountered with this technology . The transfusion service in which I work has just begun using gel cards to perform typing and we have already had several of these occurences that were not reproducible in tube ABO testing. (We have been using gel for antibody screen and antibody ID for 12-15 years) At present, these situations have led to extensive antibody workups to identify nothing more than cold auto agglutinins in the patient's serum. These workups also included cold auto adsorbtions. I have a sense that this extent of workup is not necessary, and wonder what other places are doing for resolution. Keep in mind that I am not talking about cold ALLO antibodies but cold AUTO antibodies that would not have been detected using ABO tube testing. Thanks in advance for your advice.

[David Saikin]

I have not noticed that this phenomenon occurs more than with tubes . . . cold agglutinins have a rather distinictive appearance in the gel card. maybe you could reduce your workup based on your experience with and/or the appearance of the reactivity. When it's the backtype that is discrepant, you can be almost certain it is a cold (esp if your absc is negative). Yes it is a pain, but . . . I have found that those weak, junky cold reactions in gel do not even appear in tubes unless you reduce the temperature, so your IS xm should be compatible.

[Malcolm Needs ☆]

I agree with David.

We have been using these cards for many years now and have not had many problems. When we do experience a problem with a "cold" auto-antibody, we simple wash the patient's red cells with 37oC phosphate buffered saline about 4 to 6 times, and the forward group is usually fine. We then don't worry about the reverse group, on the grounds that the ABO antibodies in the gel are monoclonal and very reliable.

We have the advantage, of course, that being a Reference Laboratory, we will be performing an antibody investigation on almost all of our samples, and so can easily see if there are alloantibodies present, and will perform these investigations by pre-warmed, warm-washed LISS tube IAT if we suspect a "cold" auto-antibody is present (but we will not bother to identify the specificity of the cold auto-antibody, as it is a complete waste of time and reagents).

At best, we will test for thermal amplitude at 30oC.

Having said all that, I had a case on-call last Saturday with a lady with an incredible "cold" auto-antibody (Hb 3g/dL) and, having washed it x10 with the 37oC PBS, she still looked like an AB forward group and an O reverse group (my guess, and it is a guess, is that she was A, R1R1, K-). This kind of case is, though, disappearingly rare.

[TimOz]

Just a point to consider.

The incidence of these types of serological discrepancies in gel systems can depend on the way the test is performed and in particular the temperature. Manual methods with cold reagents in a temperature controlled lab can give very different results to automated methods. Many of the current generation instrumentation do not have reagent cooling and can have reagents warmed up to temperatures up to 30oC (sorry no idea what that is in oF). Samples may also be warmed and the temp they are tested at depends on how long they are on the instrument before they are sampled. Some Australian labs saw significant grouping discrepancies between identical manual and automated gel groups when automation was first introduced. This may be a reason that your lab is seeing a high incidence of cold autoantibodies that others do not see.

[PeterD]

Thanks for the responses, as they gave me something new to anticipate. The cases that we have seen already have been problems with the reverse grouping having unexpected reactivity that made them discrepant with the forward grouping. Additionally, when performing tube testing, the discrepancies were not present. My feeling about these cases would be to use the tube testing results and call it a day. In one of the cases the patient had been typed by tube only several weeks before without problems. I tend to think that performing a complete workup to identify a cold auto antibody at reduced temperatures and cold auto adsorbtions to get the reverse group to agree with the forward group in gel is overkill. That is what we have been doing and I was hoping someone would validate my sense that this amount of work is not necessary. I would still be interested to hear from those who agree or disagree with my method of resolution as I am guessing we will be making some policies regarding this matter and I would like to present some details of how different institutions are proceeding in these cases. Thanks again, Peter

[Malcolm Needs ☆]

Hi PeterD,

The fact that you are getting non-discrepant results in tube (I presume that any expected anti-A, anti-B and anti-A,B is being detected) serves to prove your own point and validate the technique all in one go.

You are correct in saying that all the other work is a complete and utter waste of time and resources, because the "cold" antibody is NOT going to be clinically significant under such circumstances, either in the form of an auto-antibody causing haemolysis (unless, of course, the patient is an Antarctic explorer) or in the form of an alloantibody, because it will not cause a reduction in the life-span of any transfused red cells.

[janet]

If the reaction looks like a cold or we are getting weak reverse reactions in gel but our tube method is 'no problem' we will not do any more work.....why work up what won't hurt the patient?