Neil Blumberg

Thanks Malcolm. Never say never :).

"I'm sorry Neil, but Geoff Daniels quotes some HTR's caused by anti-N reacting at 37oC," These are, if I remember correctly, fairly ancient reports and I have never seen nor heard of a case of hemolytic transfusion reaction or HDFN due to anti-N despite having had hundreds of patients with anti-N in our service over the last half century. I've never heard of anyone else seeing one. So this is very possibly a case of old reports of hemolysis due to other causes (undetected antibodies for example). Methodology for antibody detection in the 1940s and 1950s, and even 1960s, was significantly less sensitive and accurate than currently. There are reports mentioned in Mollison and other comprehensive texts such as Daniels of hemolytic reactions due to antibodies (e.g., anti-P1, anti-Leb, etc.) that have never been reported in modern literature (the last 30-40 years). This makes me suspicious that these old reports are mistaken as to the cause of hemolysis. If the mother has an anti-N and the infant is not hemolyzing, and the antibody is undetectable I would not transfuse N negative blood. If the infant is hemolyzing, that is another story, obviously. A positive DAT, hemolysis and anti-N in the mother would dictate prudence and transfusing N negative blood. But I will stand by my original comment, which is that anti-N almost never causes clinically significant hemolysis in transfusion recipients nor in affected fetuses. Absent clinical and laboratory evidence for anti-N causing the infant's anemia, there is no reason to transfuse N negative blood when the antibody is not detectable in the fetus/infant.

Anti-N does not cause hemolytic Transfusion reactions nor hemolytic disease of the newborn and fetus. So I would not give N negative blood in general, nor if the cross match is negative.

Not an approach we would use. Alloimmunization rate in autoimmune diseases is generally quite low, including aplastic anemia.

Also, were any of the transfused units antigen positive? This is the quickest way to get a negative indirect antiglobulin test ;).

Was the patient transfused? If not, hyper hemolysis is less likely. Sounds like mechanical causes are most likely.

Patient should be monitored for a delayed hemolytic transfusion reaction for about 10-14 days, not necessarily in hospital. Most common signs are fever and progressive anemia. sometimes dark urine or jaundice. Patient education before discharge if earlier than this is essential.

No, not an error.

If the donor passes your screening questions and seems clinically fine overall, there is literally no need to worry about these issues, either in terms of donor or recipient safety. The history and physical exam are by far the most important information in evaluation of patients or donors.

Not yet available. Being developed by Grifols. Probably months to a year away from FDA approval. You can contact them about becoming a testing site for licensure I'd guess. Until it's licensed you won't be able to use it in patient care, just research/validation.

Thanks Malcolm. Not pedantry at all. These exceptions are relevant and potentially important, particularly for ABO.

Ideally one avoids infusing plasma containing antibodies to recipient antigen. Of course it’s always a bad idea to transfuse antigen to which the recipient has antibodies. We often forget that after transplant all of the recipient cells except the blood cells and immune cells are of the original phenotype. We often forget that after transplant all of the recipient cells, except the blood cells and immune cells are of the original phenotype. In this case group AB cells might be safest with no incompatible antigen or antibody. Group A would be my next choice and I would wash or plasma reduce to get rid of the anti-B which will interact with the recipients endothelial cells, soluble antigen and all non-hematopoietic cells of the recipient. Some folks would use group O red cells which I think is probably the worst possible choice given the potential anti-A and anti-B. Washing or volume reduction would minimize this risk.

When you have incompatible antigen or antibody, low level hemolysis occurs, probably with complement activation. This is not at levels clinically evident, but inflammation occurs. Inflammation potentiates (increases) B cell activation and provides one mechanism by which ABO non-identical red cell transfusion (without washing) increases rbc alloimmunization to other antigens being presented. It is known to the case, both from clinical observations (referenced in my previous post) and from animal models which provide evidence that the presence of inflammation increases alloimmunization to red cells and other antigens. It's the mechanism by which adjuvants in vaccines increase immunization to microbial antigens in vaccines, by the way. Inflammation. Not a good thing if you are not infected and receiving transfused antigen :). But useful if you are trying to get a beneficial immune response to an antigen.

Whole blood only makes sense for patients with life threatening bleeding. Red cells are safer for patients who have major anemia/minor and/or slow bleeding. Whole blood will put many patients who are not hypovolemic into cardiac failure/pulmonary edema, and are, in general more toxic than red cells alone.

If you need copies of these reprints or have further questions, happy to help.

Perhaps this is one rare physician who actually reads the medical literature on the subject or has thought things through. The history of this is very simple. Based upon the experience of severe or fatal hemolytic transfusion reactions to whole blood, it was discovered that when a patient's ABO type was unknown, and urgent transfusion was life saving, group O was the least likely to result in disaster. When group O red cells became available during the middle of the last century, with modest amounts of plasma left, it was decided by the then experts that this could be used for non-urgent, routine transfusions of all patients. So-called universal donor O red cells. The problem, with the 100% accuracy of hindsight, was that we had no evidence this is was good, much less optimal practice. But it was convenient. It meant blood banks didn't have to stock all 8 Rh and ABO types, so it was good for us in the transfusion service. It wasn't good for patients. Why is that? Well, there is residual incompatible plasma with anti-A and anti-B in all group O red cells that haven't been washed or thoroughly volume depleted. Well, you might ask, and all of us have assumed for decades, that a few dozen milliliters of incompatible plasma is not a big deal. The answer, now known to some extent, is that it is a big deal for some patients who are groups A, AB and probably B. This small residual plasma can on rare occasions cause severe hemolysis. It's 100% severe if it happens to you as a patient. This has been known for decades. What is new is the data that recipients of ABO mismatched red cells (Group O in general) have a higher rate of red cell alloimmunization to other red cell antigens, (Transfusion 2012 Mar;52(3):635-40. doi: 10.1111/j.1537-2995.2011.03329.x; 2025 Mar;65(3):588-603. doi: 10.1111/trf.18135. higher rates of febrile and allergic reactions, (Transfusion 2012 Mar;52(3):635-40.doi: 10.1111/j.1537-2995.2011.03329.x.) higher rates of HLA alloimmunization, and perhaps overall higher rates of mortality (Transfusion. 2016 Mar;56(3):550-7.doi: 10.1111/trf.13376). So, if you are a recipient, you want ABO identical transfusions, or compatible red cells that have had all or almost all of the plasma removed, as by washing, for example.

We have educated our multiple myeloma specialists to send a type and screen before administering the first dose of a daratumumab (Darzalex). Our standard operating procedure is to have a panel of three cord blood cells (we have a large OB service) that is a laboratory developed test of sorts. Cord cells do not express CD38 at interfering levels. As it turns out we have made more of an issue of this than it warrants. Patients who have negative antibody screens essentially never develop new antibodies to red cells after being started on daratumumab probably because it potential inhibits B cells function. Minimal B cell function apparently yields little ability to make antibodies to red cell antigens, which are relatively weak alloantigens, especially when there is no adjuvant or inflammation in the recipient. That said, a manufacturer is making a soluble CD38 analog that will inhibit the anti-CD38 activity and make testing easier from what I've read. DTT treatment is also reasonable. But the good news is that patients on this drug do not make new antibodies. There are literature references to this, and we have probably tested about 500 patients with no new alloantibodies. Mostly non-transfused patients, obviously.

As far as I know there is no FDA requirement for two people to issue blood. Obviously some hospitals have only one person working night shift in the lab, so that isn't happening realistically. There is a traditional requirement for two people to identify the recipient and the transfused product, but this is only if positive patient identification is not used these days.

Understood, but the most important thing to realize is that minor variations are of no clinical significance when they occur. Dealing with unrealistic or irrational or foolishly rigid accreditation or regulation inspectors is a serious but separate issue :).

The good news is none of this makes any difference to clinical efficacy or safety. These temperatures are almost totally arbitrary and have no scientific basis.

For those interested in how bleeding should be treated and bleeding risk assessed, here is a 25 minutes powerpoint lecture on the topic. https://www.vumedi.com/video/evaluation-and-management-of-the-bleeding-patient-and-the-patient-at-risk-of-bleeding/ For those who would like further explanation on how we got the benefits and risks of platelet transfusion very wrong, this is a 25 minute YouTube video on the subject.

This is operationally difficult as there are all sorts of guidelines in the literature, many from professional societies recommending platelet transfusion at different platelet counts for surgery and invasive procedures. It's hard for practitioners to ignore these, for medicolegal reasons, and instead, practice what we now know is better medicine (no transfusion in most patients). I've been involved with thrombocytopenic patients and their treatment for 50 years. I'm here to tell you that, regrettably, existing expert opinion and platelet count based guidelines are almost total scientific and clinical nonsense. Strong words, but driven by recent, actual data and extensive clinical experience and research. How should we evaluate the risk of bleeding in patients undergoing surgery? It isn't the laboratory tests, although they can be useful when there is evidence from the history and physical exam that there is a hemostatic problem. Patients whose skin and mouth show no evidence of bleeding/purpura/petechia, and have no personal or family history of bleeding problems, almost never bleed unless something goes wrong during the procedure. Indeed, the bleeding rate for many high risk procedures (liver biopsy, kidney biopsy, etc.) is very low in terms of patients who need an intervention such as transfusion, surgery, etc. So we are treating 100% of patients with plasma or platelets or both in the vain hope of preventing bleeding, which happens in perhaps 1, 5 or 10% of patients or fewer. This is sub-optimal medicine, as platelet transfusion (and plasma transfusion) are high risk therapies that can, in rare instances, kill patients. Transfusion should be driven by actual bleeding and timely hemostatic evaluation (mostly things like TEG, ROTEM, Quantra with occasionally useful tests such as PT, PTT, fibrinogen, platelet count, platelet function testing such as closure time, factor XIII level). Prophylactic transfusion in this setting is unnecessary and unlikely to help, and very likely to harm. Don't do it is my advice, despite the guidelines, which have no evidence base whatever and represent a tragic, well intentioned misunderstanding of hemostasis and transfusion efficacy and safety. So what this means for half doses of platelets is that they can be used in almost everyone with equal efficacy and reduced risk of harm. We do this all the time for the last few years, and have bleeding rates that are far below those in the literature because we use only ABO identical platelets. ABO mismatched platelets actually increase rather than prevent or treat bleeding. Our bleeding rate in prophylactic transfusions is <5% compared with 70% in the PLADO study where ABO was ignored. Using ABO identical platelets reduces platelet needs in typical hematologic patients by 50% thus increasing the platelet supply overall. Prophylactic transfusion at counts <10,000 represent the only evidence based use of platelets, in general. Prophylactic use prior to paracentesis, colonoscopy, minor surgery, etc. is almost certainly of no benefit in the vast majority of patients, and leads to harm due to volume effects (250 ml of plasma increases vascular pressure) and inflammation due to the platelets themselves. References on request.

We routinely use half dose of apheresis platelets as our standard therapy. Approved by our medical staff overall. Less toxicity, lower cost, equal efficacy. What do you believe you need to do other than having a separate code?

Makes no sense to me if that's true. Hemolytic reactions have always been in a special category because they were the most common cause of acute life threatening signs and symptoms. Perhaps that's why. Then again, most professional societies and hematology/transfusion medicine textbooks/reviews completely ignore the most common serious post-transfusion toxicities. This is perhaps because there is an expert opinion (dogma) driven belief that thrombosis, infection and inflammation, sepsis and organ failure occurring more than 4-6 hours after transfusion could not possibly be due to transfusion. Wrong, but a deeply held belief based upon decades old methodologically fatally flawed meta-analyses of randomized trials which told people what they hoped was true :).

Agree that in “not recently transfused”patients, eluates are of no clinical use unless you like seeing panagglutinins.😜 Negative eluates occur in some drug dependent examples but the clinical information is paramount in such situations in any case.