Mabel Adams

We don't worry about Hgb S in neonates except for exchange transfusions for which we request it. I've heard the same about the LR filters but don't have true evidence.

I'm reactivating this topic because it has come up again. The AABB Guidelines quoted above are over 15 years old and the only reference in them that says not to use for platelets is to a specific blood warmer's operation manual. Does anyone know of any studies with modern rapid infusers like Belmont RI-2 and Level 1 regarding infusion of platelets and cryo? It seems like they should be safe and it really helps with the workflow to use them but we need evidence.

Some of the success of WB may be due to the reduction in crystalloid rather than the whole blood itself. I look forward to more good studies being published.

We are on Epic and that is one of the reasons that we have continued to use a separate blood bank banding system. Outpatient transfusions cause the same problems. Beyond that it matters greatly whether you have good institutional compliance with the electronic Epic ID scanning for inpatients. We also draw a second blood type confirmation tube on new patients. For small children we may just give them group O blood the first time but that is a quite rare event for us which would be quite common for you.

We test any sample that meets the manufacturer's specimen requirements and ours are more than 24 hours. We even validated citrate tubes so we could use them for blood types. I'm comfortable with getting a second draw on only non-O patients because we also have a BB banding system and an electronic patient ID system. The problem is the rogue humans who decide to make end runs around the systems but with belts and suspenders and duct tape, I feel pretty confident. We are also lucky in that we do the prenatal testing for most of our OB patients. That's a point to consider in making the determination of how many specimens you will have to collect.

And Admissions sometimes selects the wrong patient's record to admit--usually someone with the same name. You would hope that we have enough checks to catch this pretty quickly these days, but strange things happen.

We have a pancreatic cancer patient who swears he has never been transfused and doesn't appear to have had major surgeries that require transfusion but who has an anti-Fyb. No evidence of a source of passive antibodies. Negative antibody screen elsewhere in recent months. Are there reports of naturally occurring anti-Fyb?

Remember that the impact can be affected by A and B substance present in patients and donors. Of course, that also means more immune complexes formed.

So you could pull off an aliquot of "packed cells" from it?

We are doing pre-treatment antibody screens and sending out for full molecular typing. We started this before we were doing DTT treatment in house. We serologically K type them if we need to give blood before the molecular typing results are back. Now I am not so sure that the full typing is justified. It seems to us that those who need transfusion whilst on the drug often don't stay on the drug long-term. As mentioned above the majority of patients on it don't require transfusion. I'm not going to change policies right away but am interested in others' experience.

I am revising our procedure for HLA-Matched platelets and reviewing literature so we can have an evidence-based policy. If you are a hospital without its own HLA lab, what approach do you take? Do you start with HLA antibody testing and give units compatible with the antibodies found and only HLA type the patient if they have a high percent reactive (PRA) or do you always start with HLA typing the patient? When do you order HPA antibody testing? Do you know of evidence to support your approach? It might also be helpful to know if you are across town from your HLA Lab and blood supplier or more remote. Thanks for any input.

How do keep physicians engaged in your committees?

I found this article fascinating.

What is HC1? Can you do PBM, lab test utilization and imaging utilization with it? Does it interface with Epic so it can create Best Practice Advisories based on your data? Are you happy with it now?

Does anyone know of a remote community hospital (~250 beds) that is using WB for traumas that is not a satellite hospital for a level 1 trauma center? I just can't figure out the logistics for making the cost benefit ratio work here. We are 3.5 hours' drive from our supplier, transfuse an average of one "real" massive transfusion patient a month (one that uses over 6 units of RBCs) and cover a land area the size of the Netherlands (as I recall). The price for LTOWB is about 3X that of a RBC unit. Would it do any good if our helicopter carried 2 units of whole blood even if we didn't keep it at any of our hospitals? Also, can anyone share their QC process for packing the WB to packed cells once it is too old to use as WB? You have to prove you get the right Hct at least a percentage of the time, correct? Does the packing process require FDA registration? It didn't use to.

Dr. Blumberg, after all of your years of relentless pressure on this topic, I am finally starting to see it listed as a caveat to typical practice sometimes now so I think you are making some progress finally. Still, it is so impossible at small, remote hospitals to both have sufficient platelets available and try to give ABO identical. Will PAS platelets with less plasma help? Also, I am hoping for 7 day outdates to ease some of the inventory pressure. Maybe switching to whole blood for traumas will actually spare the platelet supply. Also, all BB computers are set to flag plasma incompatible platelets only, not major incompatible.

For IgG or polyspecific I was taught 2+ or greater a jillion years ago. Complement is less although the Quotient complement coated cells react the best I have seen.

I was thinking of a benchtop one but is there an advantage to this one? Did you buy it straight from Hettich or another supplier?

I think Siwa's is the first that is immediate spin. The others are monoclonal but must be tested at the AHG phase I believe. Please correct me if I am wrong. I would love to have choices for this purchase.

Or if you wanted to teach in an MLT/MLS program at a community college or university.

I remember John Judd once advising me to titrate an anti-M suspected to have an IgG fraction and not worry about separating the IgG from the IgM unless the titer became significant. Then we could titrate it after destroying the IgM if need be to see what the true IgG titer was. It never exceeded something like 8 so we never had to send it out for additional testing. These seems something like that--drawing a line of what is safe to save the cost of extra testing. Only do the additional testing when it is no longer safe to avoid it. You would have to determine how you will turn it out so as to not overly confuse the OB/perinatologist. "Titer against c, E, Fya, Jkb and S positive cell = 8"? Then next time when it is 16 with a cell of that phenotype, you would repeat separate titrations and results would be "titer against c & E positive cell = 8 & titer against Fya, Jkb and S positive cell = 4"? Or do you then go to separate cells for all of them but the E & c? Or turn it out as 16 and they quit using titers to monitor? I can see some logic to moving to ultrasound monitoring as soon as the cumulative titer is above 16 or so but we also like to watch how titers change over time to help us guess which antigens baby is positive for. If you already tested amniocytes for antigens that would be moot but if you have only serology to go on you could miss some clues. We titrate E & c together because they are likely to be inherited together and separate E+, c- cells are hard to find. It also depends on if you can reliably find the same phenotype of cells for the next titration (we don't have the perfect world of using the same specific donor cell for the entire pregnancy). Maybe it also matters if you know dad's phenotype/genotype. If he is R2r then baby could be c+ E- but not if he is R2R2. Sorry to ramble on; surely someone with more experience in this than I can answer with something more substantive but I've enjoyed speculating.

We called the screen positive (went ahead and reported the gel screen) because we wanted the computer to prevent electronic crossmatch if they wanted more blood while that specimen was still good on a later shift that hadn't heard about her and didn't read her comments. We were giving her units negative for the 2 antigens that she lacked (good thing that she was heterozygous for most of our "usual suspects") so we didn't want random units to be issued inadvertently. Also, I read that patients on these drugs sometimes have a drop in H&H at first dosing so I wasn't really willing to go out on a limb and call them compatible and imply that they would have normal red cell survival.

If I need to buy a new refrigerated centrifuge for spinning down units (packing whole blood, removing preservative solution for baby exchanges etc.) does anyone have any recommendations? I want the smallest size that will do the job since we aren't a blood center. The one vendor I looked at does not list blood products as one of the functions they sell accessories for.

I have a somewhat related question and would love your (and anyone else's) input. Do you have a good system for documenting annual thermometer calibration? We have a lot of thermometers and, over the year, some get moved or broken, so I am trying to find a good way to keep track of where they are and which are due and document their calibration.

Thanks for the heads up. Now if only we could get someone to tell us or tell the patient to tell us that they are on these drugs. If you have any "pull" with someone in Seattle, feel free to pass along the request.