Since the introduction of new automation in my lab we have occassions when we come across samples that give a negative result on the antibody screen, but have historical antibody data on our computer system.

Our normal procedure is to repeat test the sample. I would be really interested to know how other labs deal with this situation.

a) Would you normally repeat test the sample ?

Assume the patient may have been transfused/ antibody was weak etc.

Also if a repeat test on the same sample then gave a 2+ to 3+ reaction- what further actions would you then take?

Thanks

The Question that must be asked is was the antibody defined as clinically significant.

It was anti-K. However further information now given suggests it was due to our lab storing the Phosphate buffererd saline in a cold room (even though this was removed a good few days to room temp prior to use- and we are the only lab in UK who do this). Apparently when PBS is cooled it develops a stratified range of PHs in the container rather than being homogenous. Potentially depending onwhere in the container the PBS is drawn may adversely affect the antibody screen, hence the difference in reproducibility.

The stratification is reversible if the containers are well mixed.

Thoughts on this please.

Some automation misses some antibodies . . . as the vendors all say - no system will pick up everything. Anti-K1 is one of those antibodies that seems to be missed sporadically by automation. I cannot answer why. There are many posts on this website that will attest to this phenomenon.

Hi David,

As I have also previously said, the system must pick the antibody up in a repeatable way on the same sample. Hopefully correction of our saline storage might resolve this problem ...watch this space!

However-I seem to have deviated from my original question which was, do labs retest their samples by same technique if the previously known antibody is not being detected on the current sample?

I do not think you should repeat the sample if historical was positive, but now becomes negative cuz probably the titer already dropped to a level that the method you are using is not sensitive enough to pick it up. Also, it wastes time and money. If it is a clinical significant alloantibody, there should be a method in you institute to alert you, and you should give the patient antigen negative blood for that antibody forever.

Hope that helps.

CK Cheng, MSc, SBB(ASCP), CQA(ASQ)

Hong Kong

Apr 23, 2009

We don't routinely retest. That's not a bad idea if you want to ensure that your test, automated or manual, was performed correctly, but the end result for a significant antibody is still the same, detectable or not: screen donor units and give antigen-negative blood. The warning bells in your computer system should go off otherwise.

If we get a negative screen on a patient with a positive antibody history, we assume that the titer is too low to detect and we transfuse with antigen negative blood.

We do not repeat the screen since some antibodies do drop below detectable levels. The patient is considered to retain that antibody or the proclivity to produce it for life so the corresponding antigen negative blood is always indicated.

We re-test mainly to ensure our technique is working consistantly. Antibody levels can drop to sub-detectable levels- but it is very unusual for this to happen on samples taken within a few weeks of each other. You need to be able to explain why you arn't detecting the antibody on the current sample, if the patient has been multi-transfused with antigen neg blood- thats a possible reason .

Regardless of the patient receiving antigen neg blood, we always need to ensure our basic critical test is working properly- and yes it does increase cost.

We only retest if something different is detected, such as an unexpected screening cell or incompatible crossmatch with antigen negative blood from previously identified antibodies.

So in a situation where a re-test was carried out on a sample with historical antibodies and this gave a positive screen compared to the result obtained 2 hrs ago- what would you do in this situation ? ( assume all weak antibody controls have worked ok).

thanks