Questioning our results. Patient with hemolysis on the pre-transfusion sample. Phlebotomy was non-traumatic, and later draws produced the same result. Hgb 9.0 Only transfusion was 29 years ago. No history of hemolytic anemia

Front type: A 4+; B 2+; D 4+; Rh control 2+

Looking at blood with no reagent: agglutination, large aggregates (is doing the same thing in heme lab)--warmed at 37C for 30 min, with no change

Back type: A no reaction; B +; D no reaction

Re-draw into warm, kept warm to the lab: still agglutinated Cannot be warmed away. No precipitate in plasma after 4C incubation.

Antibody screen: (solid phase): 2 cells pos, 1 cell neg

Antibody panel (solid phase): anti-e (cannot phenotype at this time, send out)

DAT polyspecific 2+: IgG 2+ and Complement 2+

Within 1 day, Hgb 6/Hct 14, LDH elevated (> 1000 IU/L); total bili went from 1.2 mg/dL to 3.4 ; haptoglobin not detectable, creatinine climbing slowly

Thoughts/comments appreciated.

Edited December 12, 201113 yr by donellda
Edited per poster's request

Sounds like a "cold" auto-antibody of wide thermal amplitude, but could also be a mixed-type cold and warm AIHA. The bad news is that there is quite a lot of work required to prove the latter.

First things first though. You try adsorbing the patient's plasma with RESt (rabbit erythrocyte stroma), or treating it with 0.01M dithiothreitol. The latter will break the J-chains of an IgM antibody, leaving, in effect, only any IgG antibody underneath unaffected, but it is important to use a monospecific anti-IgG reagent after treatment.

Much less likely, particularly given the (presumed) age of the patient, is that it could be PCH, in which case you will need to undertake a two-stage Donnath-Landsteiner test (good luck with this one - it's a pain in the neck - except that I don't actually mean neck, but I doubt whether the word I want to use would pass the test for this site!!!!!!!!!!).

Why do you say that Malcolm, I have found the Donnath-Landsteiner test to be pretty straight forward.

Only that the two-stage test takes a lot longer than the one-stage, not least because you have to find an ABO compatible source of serum (and most samples these days are in EDTA), and, if the patient is already haemolysing, there is every chance that the patient may have exhausted their own pool of active complement, otherwise, I agree with you that the test itself is straightforward. It's the fiddling around to which I was referring!

My question if the same for Malcolm's, what is the disease of this patient?

I will try to eluate the autoantibody from the cells first, at 45 C 15 min, if it not work , I will treat the patient's cells with DTT to destroy the IgM antibody( AS Malcolm says), do those I will put an A type cells to guarantee the A antigen is not destroyed by those step. Then do the forward type to get the real ABO type of this patient.

As to the free antibody, I think the panel already give you the answer.

Sorry, I don't know if this patient mix-type or warm AIHA.

Thank you, Malcolm and Yanxia. Your comments are much appreciated. Will update this post as I know more.