DAT Negative Hemolytic Anemia

[bbbirder]

I would be interested in your thoughts on this interesting case:

84 year old man in good health, no recent change in medications. He has symptomatic anemia (HGB= 7.9). Admitted for transfusion and evaluation (occult gastric bleeding is suspected).

Patient has history of anti-M reacting in gel and at room temp tube only (negative at 37 and AHG in tube.)

3 units RBCs ('full' AHG XM compatible) transfused. HGB goes to 9.2. No evidence of GI bleeding. Retic, LDH and Bilirubin are elevated.

Patient develops "coca-cola: colored urine", also has slight fever 2 days after transfusion. DAT is negative, pre and post transfusion. HGB drops to 7.2.

Repeat antibody screen, xm's etc. don't show any different antibody or any incompatibility (in gel or tube). (One unit was M+N+ the other 2 were M neg).

Then patient then tells Hematologist that he did have dark urine at home, but didn't think too much of it or tell anyone about it until asked.

Cold LISS wash DAT is negative.

Haptoglobin is <10.

PNH is suspected. Flow cytometry shows 'downregulation of C55 in red cells (38%)', but FLAER is negative, so this is considered negative for PNH. However there some debate among the pathologists about the Flow result, due to the recent transfusions.

He also has some abnormality in a Natural Killer Cell line... I was out of my league with the Flow, but now I am really out of my league. There is some suspicion of an occult malignancy, lymphoma? with abnormal antibody production.

A Heinz body test and acid hemolysin test and some other things are pending.

Hemoglobin has been stable at around 7. Urine has been clear again.

We really don't want to transfuse him again (at least I don't at this point).

I am rambling here...

Any thoughts?

Linda Frederick

[Yanxia]

Maybe the DAT pos cells all hemolyzed , at least below our test sensitivity.

After 3 days to retest the antibody screen maybe we can get something significance.

[Bill]

On the CBC, what does the RBC histogram look like--if there is only one population of rbc's, then you know that all the cells destroyed to bring Hgb down to 7 were transfused cells. If rbc histogram show double peak, you know that patient and transfused cells were equally destroyed. If LDH & Bili return to normal and Retic remains high, let patient stabilize himself. If LDH & Bili remain high, evidence of continued increased rbc destruction and need more answers. Now I'm rambling--this is all I can offer at moment.

[Deny Morlino]

At what point was the post DAT sample drawn? As proposed above the cells may have been cleared from circulation and not picked up by the DAT testing if there was a time delay. Also agree that if the retic remains elevated the patient's bone marrow is attempting to correct the low hgb situation and our pathologist would recommend letting the patient's system correct itself. With a three unit transfusion there is the possibility of a dimorphic population of red cells showing up on the RBC scattergram. With the patient's added information of "dark urine prior to admission", I would be inclined to think the pathology group is on the right track that there is another underlying issue confusing the issue. Is patient still symptomatic for the decrease in hemoglobin as he was prior to transfusion? If not I would try to avoid further transfusion. Just some rambling thoughts, sorry.

[David Saikin]

Malcolm gave us an interesting article on DAT neg WAIHA. British Journal of Haematology, 132, 651-661. "Does a negative dat exclude waiha? A prospective sutdy of 504 cases." The abbreviations are mine.

[LaraT23]

I recently had a patient with similiar symptoms. My patient had a pos auto in the gel and negative DAT. The patient was on a maintenance dose of sulfa drugs and the gel has sulfas in it as an antibacterial. When we removed the sulfas from our screen cells ( yep in those too), we found he was negative. So, suffice to say, he was taken off of the sulfa drugs and now has less issues with chronic anemia. He does have a myledysplastic symdrome, so not all of the anemia is gone. He just doesn't have black plasma and a bili of 19 anymore.

[clmergen]

I remember a patient at a hospital I worked at that had something similiar. We never could identify an antibody. A phenotype was done and for some reason, anti-c was the suspected culprit (don't remember all the details). We transfused washed c negative units and the patient did fine. I always thought it was a little bit of voodoo medicine but it worked.

Edited April 16, 2010 by clmergen
spelling

[boonyville]

I am new to bloodbank, so forgive this tangent.

I recently started working with a few techs schooled in the <outside the USA>.One did not know if her DAT was positive, so she added a drop of saline to the spun mixture ... she then called it "negative". Is there any valid reason to do this???? It seems as if she is convienietly diluting out a weak reaction...OR if it is that weak of a reaction,does any one care?

It was a post-tranfusion reaction work-up...

Edited April 19, 2010 by Cliff
Italics text replaced by admin

[Malcolm Needs ☆]

I am new to bloodbank, so forgive this tangent.

I recently started working with a few techs schooled <outside the USA>.One did not know if her DAT was positive, so she added a drop of saline to the spun mixture ... she then called it "negative". Is there any valid reason to do this???? It seems as if she is convienietly diluting out a weak reaction...OR if it is that weak of a reaction,does any one care?

It was a post-tranfusion reaction work-up...

I think that this was a "hangover" from the "old days" when it was fairly common practice to add a drop of saline to a reaction to see if it was true agglutination, rather than rouleaux. The rouleaux would be dispersed, but the agglutination would not (allegedly)!!!!!!!!!

From what you say though, this was a DAT, in which case there is absolutely NO valid reason for doing this, and YES, very weak DAT reactions in a transfusion reaction work-up can be extremely significant.

It is a thoroughly bad (and quite possibly, dangerous) practice.

:eek::eek:

[vilma_mt]

This practice is only done at IS phase. Rouleaux is caused by abnormal proteins and 'adding' or 'replacement' of saline is used. Never used in the IgG phase since the cell washing would have done the trick.

[Eagle Eye]

What is your policy & procedure? Follow it. I always make sure my new techs follow our procedures and I tell them not to listen to all the tricks some time old timers do. My crew is fairly new so do not have that many tricks that are not consistent with our policy and procedures.

[vilma_mt]

Saline Replacement Technique though rarely used is still acceptable for use to resolve Rouleaux. Check what your policy states if there is any and make sure it is accurate. If there is none, I think one should be made to better understand rouleaux and the proper usage of saline replacement technique to avoid future errors. It's obviously a training issue. You should never turn a blind eye and ignore anything that would jeopardize patient's health whether it's policy error or employee training bring to your supervisor's attention.

[David Saikin]

I still use Saline replacement . . . it can also be used after 37C incubation.

[Malcolm Needs ☆]

I still use Saline replacement . . . it can also be used after 37C incubation.

But surely not for a DAT????????????......and particularly not when investigating a possible HTR????????

:confused::confused:

[David Saikin]

No . . . you couldn't do that for a DAT. Only to determine rouleaux.

[galvania]

Going back to the patient, although there has been no change in his drug régime, it could be that he is now beginning to react to those drugs. That would explain why he had 'dark urine' at home without a transfusion. Maybe.......

[Malcolm Needs ☆]

I would agree Anna, and it would have been a very good point, but the patient was transfused (or, at least, that's the way I read the post)! I think the original post said that he was given 3 units, had a rise in Hb, and then had the episodes of dark urine; but I could have read it incorrectly - it wouldn't be the first time by any means!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

:o:o:o:o

[rravkin@aol.com]

Are there any orders for Serum and Urine Protein Electrophoresis? I think it would be interesting to see these patterns. I would also be curious to see what the rest of the CBC results looked like and the red cell morphology. Given the negative serologic results and the appearent elevated destruction of RBC's I wonder if we may be seeing something more systemic like an osmotic fragility situation, the origin of which remains to be seen.

Edited April 20, 2010 by rravkin@aol.com

[margaretcox]

A reply to the original question: we had a delayed HTR (similar presentation to your patient, brown urine etc) that *turned out* to be (probably) an anti-Jka, even though we never found any smoking gun (DAT neg, AS and XM all neg by tube). Did a phenotype to get some possibilities, and a ref lab finally found a couple of homozygous Jka cells to react by solid phase. We continued to transfuse her with Jka-neg and she did fine after that. So my lesson from that was that it CAN be immune hemolysis even if you think it HAS to be some other reason.

[Malcolm Needs ☆]

A reply to the original question: we had a delayed HTR (similar presentation to your patient, brown urine etc) that *turned out* to be (probably) an anti-Jka, even though we never found any smoking gun (DAT neg, AS and XM all neg by tube). Did a phenotype to get some possibilities, and a ref lab finally found a couple of homozygous Jka cells to react by solid phase. We continued to transfuse her with Jka-neg and she did fine after that. So my lesson from that was that it CAN be immune hemolysis even if you think it HAS to be some other reason.

Agreed; particulalry anti-Jka (and some Dombrock antibodies).

:):)

Another thing that has just struck me is that it might be worthwhile trying with fresh serum, rather than EDTA plasma. It is a long shot, but I do know of one (fatal) anti-Vel that could only be detected using fresh serum, rather than EDTA plasma, because the complement enhanced the reaction so much.

Edited April 20, 2010 by Malcolm Needs
Faltering memory

[rravkin@aol.com]

Hey Margaret and Malcolm,

Are the posibilities you present here capable of causing the clearence of three units of PC's within a few days with no appearent outward symptoms (only a slight fever), and no appearent serologic evidence to support an immune response ( I did not see a result of an auto control mentioned). I think that this case is very interesting. The patient had "coca-cola" urine prior to the initial ER visit and upon subsequent admission. The pre-transfusion Hgb=7.9, and after three units of PC's the hgb was only 9.2 with no appearent bleeding. With a normal, yet depleted, system the hemoglobin should have reached approx 11.9 (1gram per unit). It seems that the patient is clearing the red cells at a very rapid pace; approx. one third the total circulating volume for this patient; and this is occuring with only a slight temp noted two days later and curiously the patients hemoglobin resides there after at 7.2; the approx starting hgb. Given these facts is it not posible that we are seeing an overall upset in the isotonisty of the whole blood such that only a volume of rbc's producing a hgb of approx 7 g/dl can be supported. I think that it is very interesting that the initial hgb=7.9, post tranfusion hgb= 9.2, and after clearence the hgb is back to 7.2. It appears that this concentration of rbc's is the equilibrium piont or product there of. I guess what I am trying to ask is if an immune response can destroy approx. one third the circulating rbc's with no appearent symtoms and go undetected by conventional serologic procedures or is it equally possible that we are seeing an isotonic upset of the circulating whole blood with currently an unknown origin????? It seems that the suspition of PNH is more probable than an immune response in this case.

Edited April 21, 2010 by rravkin@aol.com

[Malcolm Needs ☆]

Hey Margaret and Malcolm,

Are the posibilities you present here capable of causing the clearence of three units of PC's within a few days with no appearent outward symptoms (only a slight fever), and no appearent serologic evidence to support an immune response ( I did not see a result of an auto control mentioned). I think that this case is very interesting. The patient had "coca-cola" urine prior to the initial ER visit and upon subsequent admission. The pre-transfusion Hgb=7.9, and after three units of PC's the hgb was only 9.2 with no appearent bleeding. With a normal, yet depleted, system the hemoglobin should have reached approx 11.9 (1gram per unit). It seems that the patient is clearing the red cells at a very rapid pace; approx. one third the total circulating volume for this patient; and this is occuring with only a slight temp noted two days later and curiously the patients hemoglobin resides there after at 7.2; the approx starting hgb. Given these facts is it not posible that we are seeing an overall upset in the isotonisty of the whole blood such that only a volume of rbc's producing a hgb of approx 7 g/dl can be supported. I think that it is very interesting that the initial hgb=7.9, post tranfusion hgb= 9.2, and after clearence the hgb is back to 7.2. It appears that this concentration of rbc's is the equilibrium piont or product there of. I guess what I am trying to ask is if an immune response can destroy approx. one third the circulating rbc's with no appearent symtoms and go undetected by conventional serologic procedures or is it equally possible that we are seeing an isotonic upset of the circulating whole blood with currently an unknown origin????? It seems that the suspition of PNH is more probable than an immune response in this case.

I see exactly where you are coming from, and I must say that your explanation is more probable than mine, but......

There are cases in the literature where there has been extremely rapid clearance of transfused red cells, leading to a negative DAT (because all the transfused cells have been cleared) and no detectable antibody in the plasma; although giving (usually) Rh-matched blood leads to a sustained increase in Hb. In this case, I read it as the patient had noticed "coca-cola" urine at home in the two days after he had received the transfusion (although, as I said in a post earlier, I am quite prepared to believe I am wrong in this assumption). If I am right, however, I wonder what other symptoms he underwent and did not "admit" to, until later.

I still think you are probably correct (and I agree that this is an exceptionally interesting case), but I will hold an open mind as to the true cause until the original poster tells us the actual diagnosis!

:redface:

[rravkin@aol.com]

Hey Malcolm,

Thank you for your thoughtful response. If this case does turn out to be elicited by an immune response I will be just that much more in awe of the immune system; it is truely amazing! One side note though, I did not see a result for an auto control; this may yield more info.

[galvania]

I got the impression that he had had 'coca-cola coloured' urine after the transfusion but had ALSO had 'dark' urine at home prior to the transfusion. Bbirder - we need clarification here - sorry - and have you any new news about this patient? It's a really interesting one. Thanks

Anna

[bbbirder]

Sorry I have been off-line for a while. Yes, I have more news on this patient.

He got no more transfusion and his hemoglobin remained stable. He was discharged home (but I imagine he fatigues easily with the 7 gm HGB). Yes, he had dark urine prior to admission.

A reference lab got a positive cold-LISS wash DAT and reported a low-affinity autoantibody. They found no other allo antibody except the previously known anti-M.

Thanks for all of your comments.