Mixed field

[Townsend]

I'd like to know what others have seen in various methods concerning mixed field samples. We are a children's hospital who transfuses a large amount of O red blood cells. We're seeing mixed field reactions when doing blood typing on approximately 25% of our patient samples! We use gel technology for our ab screens and DATs, but are still using tube method for ABO/Rh. As we look to the future of automation or possibly changing our ABO/Rh testing to gel technique, I'd like to hear what have others experienced concerning mixed field. I liked what I saw with the ECHO demos until I discovered that it cannot read mixed-field and would kick out up to a quarter of our ABO results! Any better luck with gel - manual or using the Provue?

Stephanie

[David Saikin]

With the gel technology you should only get mf+ results for ABO when you have a truly mixed field. I have observed this in my adult pts who receive O red cells when they are not group O. You can see mf results in the absc if you load the incubation chamber in a sub-optimal manner (in which case it is artifact not a true mf).

[Malcolm Needs ☆]

I agree with David, except that we do see mixed-field reactions with A3, even though all of the red cells are really group A.

We once picked up a true chimera by ABO too. You can't get much more of a genuine mixed-field than a chimera!

[LShirley]

Gel is great at detecting mixed field reactions both manually and on the ProVue. You can get mixed field results after transfusion of only a couple of group O red cells in a non-group O patient.

[Malcolm Needs ☆]

I couldn't agree more with you LShirley, but David is absolutely correct about loading the chamber properly.

Unless you know for a fact that the individual has been transfused, it's always worthwhile repeating the test.

[JOANBALONE]

We use the Provue and it will not interpret ABO/RH mixed field reactions. However, the technologist may change or accept the results before sending the results to the LIS. Once in the LIS the technologist may still change results or make an interpretation of ABO/Rh based on mixed field reactions and pt transfusion history. I, personally, accept what the Provue is seeing and then make and interpretation of ABO/Rh in the LIS based on previous patient ABO/Rh history and transfusion history.

[Likewine99]

Ditto what LShirley said. I've worked in a pediatric BB and we used gel for ABO/Rh and loved it. If you have an out of group BMT or cord blood transplant patient you can actually start to see the mixed field as engraftment occurs.

I am now at an adult hospital with a ProVue and JOANBALONE is correct, usually one unit of O cells to a non-group O person will show up as a nice mixed field on the ProVue.

We had a B neg pt the other day that got O neg cells at one of our sister hospitals and we picked up the O cells right away.

[Eagle Eye]

We see mixfield with all our patient (mostly Rh). I have also seen MF with ABO...Gel is very sensitive we have a couple of patient A2 with anti-A1....ProVue gave 1+ with A cell and we were not able to duplicate same with tube...but patient is A2.

[janet]

With the gel technology you should only get mf+ results for ABO when you have a truly mixed field. I have observed this in my adult pts who receive O red cells when they are not group O. You can see mf results in the absc if you load the incubation chamber in a sub-optimal manner (in which case it is artifact not a true mf).

What do you mean by mf in absc if you load the incubation chamber in a sub-optimal manner .... what king of artifact do you get, how is the chamber loaded???

Thanks!

[AMcCord]

Echo will not report a mixed field, but you should be able to see it when you review the camera image for the test well. We document visual inspection results of the well when visual and Echo interps vary. We look at historic type, if available, and use tube testing as a 2nd type for all new patients.

[bbbirder]

Ditto what AMcCord said about the Echo. Depending on the strength of the reactions, you can edit the results you see on the Echo, at the Echo. If stronger, you will need to Edit in your LIS, just like the ProVue users.

We get occasional discrepant results with our Echo, just the same as tube. I think that automation is great, but it can't take the place of the (well-trained) human brain.

Linda Frederick

[PammyDQ]

THIS IS A REPOST I ENTERED ON A RELATED TOPIC, but it got no action...

When our BB LIS system was set up, there was no grading reaction "MF" created as a postive reaction, forcing us to enter a numerical result 1+-4+, then commenting that the reaction is mixed field. We use the MTS Gel system, so mixed field reactions are extremely obvious with ABO/Rh typing.

To complicate matters, we have a bone marrow/stem cell transplant unit that has been doing an increasing number of major and minor mismatched transplants. With recipients blood types changing with chimerism and with transfusion of type O blood which is then needed, we're seeing a heck of alot of mixed field. BUT NOW we're being asked to semi-quantitate the PERCENTAGE of O vs. A or B cells in the gel column!!

A project my medical director suggested is to set up samples of various percentages of red cell suspensions to use as a reference. I have yet to do this as we've been very busy and we're a "tiny" BB (only 2 techs day shift, 1 eves).

ANY SUGGESTIONS?? Any one else with a similar predicament??

[Malcolm Needs ☆]

I suppose it's out of the question to tell your medical director not to be silly? I thought so!

You could try using a FACS and labelled ABO antibodies, but you would have to be very careful because, if the recipient is a secretor, their own A and or B substance would adsorb on to the surface of the group O red cells.

In one hospital I know, they fully type the patient and the recipient, and then choose an antigen to follow that the recipient does not express and the donor does. They then only transfuse antigen negative blood, so that if the antigen is detected, it must come from the graft (but, again, for accuracy you would have to use a FACS machine and a labelled antibody).

[ldlashley]

Do you use the tube method to veirfy your edits on the ECHO? Or do you just read the camera images and edit? We are just developing our SOP and plan to go live with the ECHO June 1st. I am training the staff to look at the result log for the sample and the camera images and then if still unsure perform tube method. We are leaving Gel and going to ECHO

[PammyDQ]

Malcolm, thanks for your reply. My query is actually 2 things.

1. The transplant department follows the engraftment progress via periodic chimerism studies. This is just something we're being asked to "guesstimate" since we're seeing the changes every 72 hrs as we repeat the T&S( TS is kept current for all inpatient transplant patients). We started doing this amongst ourselves in BB then at the weekly transplant meetings it was mentioned what we're seeing in the gel column as the patients' blood types change, now they sort of expect it.

2. The biggest problem is grading MF rxns. We do not have an option to enter the result MIXED FIELD. We are forced by our computer to choose 0, 1+,2+,3+,or 4+, then comment mixed field. So, if an A+ patient recieves a unit or 2 of O+ rbc, we'll see a weak mixed field. We'd result 4+ for Anti-A with a comment MF. But with the transplants, we're seeing for example an O+ patient with a transplant that's engrafting, so we may see ~ 75% cells at the top of the Anti-A column and ~25% at the bottom...so we might call it a 3+ reaction...because we're forced to choose a grading. But this is an invention we're forced to use due to the absence of MF as a reaction strength. Of course any transfusions to such patient must be type O so we wouldn't know if residual cells are the patient's or the transfused units.

I guess my only question is anyone else having any similar issues with grading mixed field reactions such as these.

We've been told that the computer cannot be programmed for this reaction but I think it's a lazy issue by someone. It's Meditech......

[Malcolm Needs ☆]

Hi PammyDQ,

I'm sorry, I'm not making false excuses, but I can't answer you in detail until tomorrow night (UK time). My 10-year-old son has got exams next week and so I've got to get him to bed reasonably early, and I've got an absolute pile of cases to report tomorrow (including a pregnant Oh lady).

I'll answer as soon as I can tomorrow night.

Best wishes,

Malcolm

[Malcolm Needs ☆]

Hi PammyDQ,

Oh dear, it sounds like you've made a rod for your own back! That's the trouble with admitting to anything that you are doing that could be seen as "helpful" by those who do not necessarily understand the concept of what you are doing!

I can't really say much more than I did last time, but I really would advocate "choosing" an antigen the recipient lacks (let's say "S" for example) that the donor expresses, and then making sure that all cellular blood components given to the recipient are S Negative. Any red cells that are S+ must, therefore, be derived from the donor.....but how you would grade this, without the aid of a FACS, is anyone's guess, because the % of S+ red cells would be diminished by any S- cellular components that have been transfused.

Of course, you could always pull the wool over their eyes by making a guesstimate, because you couldn't be accused of actually being wrong!

[galvania]

I was at a presentation once where the different populations were quantified using molecular biology. I guess you could also use flow cytometry if you wanted to be more accurate - but that would require a lot of validation. doing it in gel really is pure quesswork.

I guess you could also do it the old-fashioned way - looking down a microscope and counting....when you have a spare half-day or so..........

[Malcolm Needs ☆]

Anna, you are quite correct that flow cytometery would take a certain amount of validation (as would any technique), but once this is over, the technique is quick, easy and accurate, whereas, as you say, estimating in a gel is pure guesswork.

Of course, it is possible to do the measurement by molecular techniques, but you would have to have not only the equipment necessary to perform the testing, but the appropriate primers for the blood group genes, and these tend to be expensive.

[rrcc1974]

PammyDQ,

We've been told that the computer cannot be programmed for this reaction but I think it's a lazy issue by someone. It's Meditech......

This is our setup. You could add any Result Code/Text you want.

AA.doc