Mabel Adams

We have never done TPs. If they aren't done at the blood center, nursing does them. The closest we get to them is answering the phone call, "Do we get the blood bags from you?" with "call central supply."

This is a sideline, but I think it takes a ton of RhIG to treat if you give Rh pos blood to a negative patient--and it seems to me I have read that it is not entirely effective. An interesting approach I heard for a young female inadvertantly transfused with Rh incompatible blood was to do a pheresis exchange transfusion with Rh negative blood then a Kleihauer to determine the number of RhIG doses appropriate to the remaining Rh pos cells. It worked for this one case reported. Anyone else have experience with similar cases? That said, it is ABO that kills so I agree with the approach.

We were told that the only way to be able to print ISBT labels interfaced to Meditech was to buy the $10,000 printer (we are also on Magic). We chose to cease aliquoting and pooling rather than purchase this for the 10 times per year we would need it. Digitrax has a standalone printer (as does another company whose name escapes me) but they are still quite expensive for the limited use we would have. We bought a 26 yr supply (minimum quantities) of the most common pre-printed irradiated labels for about $800 and are making do with that. We are not relabeling thawed FFP.

Good point, I forgot to include mistyping the patient. So if we call a patient A when he is really O, then the A unit we try to crossmatch would be strongly incompatible unless the patient had a weak reverse. We really don't care about rouleaux or cold antibodies because we accept the theory that nothing that we are going to detect in the IS xm (that we don't detect in the Ab screen) is going to be clinically significant except ABO incompatibility. If we were doing IS xms for any reason besides ABO compatibility we would not be able to substitute the e-xm. I am kind of hammering on this because I have some techs that read that IS xm practically with X-ray vision. I haven't discouraged them, but they spend a lot of time resolving very weak questionable reactions.

In a few years when barcoded armbands become common, will those that have gone to typing 2 specimens go back to only one? My hospital is already looking at barcoded armbands and we may skip the double specimen concept if it comes together soon enough. I wonder if 5 years from now double typing AND barcode armbands will be the standard of care. The barcoded armbands will only work if there is something that stops the nurse from proceeding with the transfusion--or at least warns her/him and she trusts the system enough to heed the warning.

So, lets say you titrate all these OB samples. What is the doctor going to do with the info? Is he going to forego postnatal RhIG (heaven forbid)? Is he going to panic the mother that her baby will be sick? As long as the reactions are not strong and not detectable at some other phase than IAT, why not just see if the baby has a pos DAT? What are they going to do about it anyway, if it isn't detected until delivery? The point then will be to treat the baby based on if it is affected and how much. If they need to know for next pregnancy, then they should do a repeat Ab screen on the mom about 6 mo. post delivery. You might need to do further studies in the A+ baby of an O neg mom that has a pos DAT. Even then, a titer won't really clinch the case for you. The eluate will be of more use. I believe RhIG can sometimes cause a pos DAT in the baby, although probably more likely right after it is given than 3 mo. later. Of course, we must always err on the side of giving RhIG when it is not needed, rather than withholding it when it is.

So if we pre-warm away those cold agglutinins in IS xms, what are the chances that we are pre-warming away an ABO antibody that is very weak or trying to react with a very weak antigen? Thinking aloud: The point of the IS xm is to detect ABO incompatibility. This could be caused by: 1. Mislabeled unit 2. Wrong unit or sample being tested 3. Wrong donor's segments attached to unit (this happened to us once, but it was Rh that differed--it's a good story). These would usually give the strong incompatible reactions of a reverse type--unless the patient has a weak reverse. Then maybe you could prewarm away the reaction? Especially if it is an add-on xm and you aren't aware that the patient has a weak reverse. 3. Very weak subgroup of donor not detected by donor testing. (extremely rare due to requirements of commercial antisera) 4. Unusual subgroup of donor not detected by donor testing but detected by this patient's sample. (unicorn rare) These are so rare as to have been deemed inconsequential (i.e. electronic crossmatch) Anything I missed?

We find (and were told by our Ortho trainer in 2003) that respinning the sample helps prevent topline. This certainly makes sense with respect to fibrin bits that can precipitate out in a stored specimen. It seems like platelets, being so much smaller than red cells, would spin through the gel fine, but plt clumps could probably stay on the surface and trap some red cells. And we know from Hematology that there are patients whose plts clump in EDTA. We usually respin a sample that causes topline before repeating the gel test. It would be interesting to repeat the test using an aliquot that was not respun plus one that was. Also, if you ever run a gel test on a sample you thought was spun but had only settled, you will usually get bad topline.

I voted no, but we are still trying to figure out what we plan to do about this issue.

How do you be sure that each person doesn't assume that the other person "really" checked so they are just signing off to meet the rules. This may be almost subconcious and easy to mentally justify if staff is busy. I guess it comes back to buy-in by the staff--but how do you make sure the staff three years from now are equally convinced of the value.

"We try to follow what the AABB "Guidelines for Prenatal and Perinatal Immunohematology" says... "Only when prenatal tests for Rh are unequivocal and clearly reactive (>/= 2+) should the woman be considered Rh-positive." Interestingly, John Judd, who put this together, said on the AABB member forum that he uses </= 3+ when using Gel anti-D testing. Linda Frederick" Above is a quote from one of the previous threads here on the topic. You could search for the others, or just get this AABB publication. It is very helpful in understanding current thinking. It does not harm a weak D patient to get RhIG--and, if they turn out to be a partial D, we hope it helps prevent production of anti-D directed at the epitopes they lack. This is the motivation behind no longer doing weak D testing on anyone but donors and babies of Rh neg moms (i.e. sources of sensitization). Of course there are partial D patients that type 4+ with modern reagents, so it isn't a very clear-cut area until molecular testing becomes standard.

We are not FDA inspected (we are a private lab) but one of the hospitals we serve irradiates blood for their patients in their Radiation-Oncology center, so they are FDA registered. When rewriting their protocol when converting to ISBT, I was under the impression that they would no longer need to add a barcoded facility ID label. Just to be sure, I contacted their FDA inspector. She said that they did still need to add a barcoded facility ID label to products they irradiate, place it in the lower right below the ISBT expiration, mark out the original license number, and, of course, change the product code and expiration (when indicated). I don't know if this is more her interpretation or the official FDA stance on irradiated products.

Meditech for an e-xm scares me. There are always quirks in MT that show up later that you didn't know existed. And not much of the system is put together logically, so I have a low degree of trust for something so critical. I haven't looked at their new enhancement yet.

We got re-educated quite a few years ago and quit doing LISS titers. Now it's saline, 60 min, IgG, endpoint is last tube showing 1+ or greater. Titers < 4 don't speicifically determine RhIG, but it has been used as a reasonable assumption.

Whichever specimen gives you the most positive ID of the patient would be paramount. The other factors that may be involved with testing newborns (subgroup expression, no reverse etc.) would be the same whether it is cord blood or not. I have never seen Wharton's Jelly interference, but that is the only other difference I can think of.

We use EDTA and see a fair amount of rouleaux. Wish we could go to e-xm. Anticoagulated blood has higher protein because it still has all the clotting proteins. We have a procedure for managing rouleaux. It must appear microscopically like rouleaux; first we try albumin, then incubate with the albumin, then do saline replacement if nothing less intense works. Or we can always just substitute a gel xm.

Larry, is there another Meditech ISBT group that I don't know about or are you referring to our little email club? The big issues for Meditech users in ISBT are about aliquots and modifying aliquots (like irradiating them). Also, every possible source (collecting facility) whose units you might get must be entered in your source dictionary. If your supplier doesn't import much, you are lucky. There are many many functions that they now need to change to dealing with product groups instead of product codes, but they haven't yet.

How did the baby have enough maternal anti-A,B to react with the transfused cells, but not have ABO HDFN?

I don't think there is a real standard, just the arbitrary times passed down from old lab people to younger ones that seem to provide reasonable answers in a reasonable time. I read an article once that if you wanted the patient's blood volume to totally adjust after transfusion, you would wait 24 hours--assuming they didn't have heart or kidney problems etc. That is not going to provide timely enough information for medical care, so we accept a somewhat less accurate result. Most places use something between 30 min and an hour minimum I think. It still has to be interpreted based on whether the blood was given with lots of fluids or none, plus patient's other conditions.

Another quirk of ISBT is that some pheresis plasma units that are divided by the supplier because they are big enough to make 2 products will come labeled as divided units (with A0 or B0 on the product code). Compare this to the pheresis plts that have separate product codes for each of the expected divisions from one pheresis donation. I don't dare ask why. It meant that our system (Meditech) had to be able to scan in divided products and maintain the division information even though we won't be making any aliquits ourselves.

And what antibody caused the HDFN?

They make dial and electronic thermometers on metal "spikes" that would have no risk of breaking.

Pat, could you explain Dedicated units as well? Thanks.

If you only pick up an anti-M at room temp, then it isn't clinically significant and a gel xm is fine. If you do pick it up in gel, and it is strong at immed. spin, I sometimes do an immediate spin xm using the patient's sample to tell me which units to take through gel xms. Basically using the patient's sample to antigen type the units. We don't keep commercial anti-M since it is never necessary to antigen type units for it per Issitt.

Seems logical.