WisKnow

Congratulations Malcolm! You undoubtedly deserve it.

Remember that you can combine the 6 elements into 1 competency exercise. You may observe the tech while they are performing testing on an unknown (2 elements covered) and ask questions related to the exercise for problem solving (another element). Have them document the unknown as they would a patient (4th element) and ask them to perform the appropriate QC for the tests they are using. Once you've reviewed any intermediate worksheets/other paperwork (5th element). They will use equipment while they work (6th element or at least part of it). Cover as much ground as possible with each observation.

From what I have been told by accrediting agencies here, different tests can only be combined into one test system if there are no unique aspects in testing or in problem solving when something goes wrong.  I interpret that as meaning the antibody ID, DAT, and elution are all separate test systems because the testing method is very different.  Different treatments, however, I could see being grouped into 1 test system (EGA, DTT, ficin) if you can "sell" them as being similar enough.  Same goes for the adsorptions, in my opinion.  I would rotate which treatment and which type of adsorption you do each year, but group them each into a test system (RBC Treatments as 1 test system and Adsorptions as a separate test system).   If you group multiple tests into a test system, only one of the tests in that system need to have the 6 elements of competency assessed each year.  However, you have to watch what you group into a test system!
It seems crazy, but by the letter of the "law," we should be doing all 6 elements of competency for every test we perform if the procedure is not exactly the same.  (I have read that even if the amount of a reagent is different, it would be a different test system.  Crazy! )  In practice, this doesn't work for our site; we would spend more time assessing competency than doing real work!  We decided to live on the edge and combine a few things into test systems, even though they are pretty different test procedures. 

Happy Independence Day to all my friends in the USA.

We considered doing this recently but because we are FDA inspected we didn't want to get into the whole "home made reagent" issue. There are a lot more hoops to jump through unlicensed sera.

Sounds like poor advice from the IRL, I think what they meant to say was that the CD38 antigen is weakly expressed on Cord cells.  The Kell system is completely removed by DTT treatment so you should give K neg blood as you have no way to exclude anti K, BTW that also means JSa, Jsb, Cellano etc. also Lu null cells do not express CD38 either.

PEG-only antibodies are well known, but it's common that they can also be detected in one of the other very sensitive methods (Gel or Solid Phase). The LISS method is less sensitive than PEG and it's not surprising that you can't detect it there.
It is possible that the PEG reagent contains a chemical that is critical to the reactivity of your patient's particular antibody. There are lots of documented examples of "chemical-X-dependent" antibodies. The absence of the specific chemical from the LISS test and the Gel system would render it nonreactive/undetectable.
Trying another manufacturer's PEG reagent (presumably with a different formulation) might clarify the situation.

Could be either. In my experience, PeG is a bit more sensitive than gel. When we had nebulous reactions in gel, we resorted to PeG to ID the antibody.
I think it's always valuable to remember that there is no one system that will detect all antibodies, even within the same specificity - that's just life (or maybe I should say, that's just biology).

I will use a patient's specimen to screen units.  When I find compatible ones I will type using the FDA approved reagents.  This is especially helpful to screen for higher incidence neg units (c, Fy, Jk, Ss).  I think that if you are using it as a sole source to screen you could be in dire straits if you missed something.

A few points, but first, a disclaimer: I am not an expert in FDA or CLIA regulations.
I suspect that many labs are using some form of a "home-brew" reagents. The number of internal controls required by various regulatory agencies increases by the day. Add the large number of "competencies" necessary for all the staff and one has to use what is available.
Personally, I don't think viral testing should really be an issue (although it would be nice to know) - you're not preparing this stuff for distribution/sale to other facilities; the "treat all materials as potentially infectious" mentality and Universal Precautions approach should cover you.
The absence of an expiration date is probably better that assigning an arbitrary one (you have no stability data). In your application, you have a in-built control - it either works or it doesn't (I'm presuming you test the untreated [positive result] and the treated cells [negative result]?).

I would not routinely use them in place of manufactured reagent antisera.  I think it would be fine for screening purposes but would confirm with manufacturer reagent antisera.

Oh, okay WisKnow.

Certainly, within the UK, using a patient's sample as antisera frowned upon, but there are occasions when the specificity is so rare that you have no option other than to use the patient's sample.
I am not thinking of something like an anti-k (such a specificity is rare enough, but it is possible to obtain an example through a commercial source, but specificities that are even rarer, such as our anti-DOLG (anti-Do8) that, as far as I know, is unique to one individual in the world, and if we didn't use her sample, there would be no anti-DOLG!

Until the genotype had been done, there was no definitive ABO type, and so group O should be given (there was never any suggestion that the patient was an Oh).  Once the genotype had been performed, and it was determined that the patient had the genotype A101/B101, then there is absolutely no problem in giving A1B.
 

This is our policy.

This is our policy also.  Almost all c negative units will be E negative so we trust to statistical luck for patients with anti-c.  We started this policy long ago when one edition of the Technical Manual recommended it.  When the wording was changed to say some recommend it, I considered our situation and continued the policy.  We are several hours away from our blood supplier so if we have caused an anti-c to develop we can't get c negative blood imported quickly like we could if we were just across the city from the supplier.  We could screen for it but would have similar results as described above after the supplier saves out the R1R1 units.  Also, our region has a lot of small hospitals that don't do Ab IDs.  I feel that this policy preserves the chance of finding crossmatch compatible blood or even to expect that O negative will be compatible with a patient with anti-E in an emergency (Rh negative is about 99% E negative).  I have never had a hemorrhaging patient with anti-E saved by this policy so maybe it is overkill, but that is our logic.  We have the added benefit now that our supplier sends us a batch of 10 historically negative units (for various random antigens) every Friday that usually includes some c negative units.  That has been a big help.

A101 and B101 are both the most common genes for the "A-transferase" and "B-transferase" respectively, meaning that there are no mutations at the molecular level of the genes.  It is still, therefore, much more likely that it is competitive inhibition between the two transferase enzymes that is causing the weak expression of the B antigen.

Hemo bioscience is selling ready to use DTT ( frozen), see our web site for details.  

Unfortunately the answer is, it depends...  If the anti CD47 is a human antibody, then it will cause a positive DAT as CD47 is expressed on most cells in the body including erythrocytes.  If the anti CD47 is mouse or goat then it will not as this antibody will bind to CD47 but will not cause the same interference we see with Daratumamab.  I believe recent therapies involve blocking some of the CD47 receptors. 

Have seen this many times with cancer patients.   They are probably immunosuppressed and not making a lot of antibodies including isoagglutinins.   I would try room temp incubation first.   Make sure you run an autocontrol.

It is the policy at our facility to give E=c= RBCs when patient has anti-E and types also as c= or if c pheno is unknown or cannot be done due to very recent transfusion, but we do not type the units for E when patient has anti-c only. In this case, we only give c= units.

I have the advantage in that I work in a Reference Laboratory in London with an enormous stock of units.
That having been said, we would give c Negative E Negative blood to all R1R1 patients with a haemoglobinopathy, or an auto-antibody (or any other condition that means they are likely to become transfusion dependent) and to females of child-bearing potential, even if they have not made an anti-c or an anti-E, as a sort of prophylaxis to stop them making these specificities (although of course, "naturally occurring" anti-E-like mimicking specificities still occur).
Any patient that falls outside this, we just give the cognate antigen negative.  Remember, the vast majority of patients who receive a transfusion either die within 12 months,or never require further transfusions throughout their lives, and so it really isn't worth giving R1R1 blood to these others.

It is the policy at our facility to give E=c= RBCs when patient has anti-E and types also as c= or if c pheno is unknown or cannot be done due to very recent transfusion, but we do not type the units for E when patient has anti-c only. In this case, we only give c= units.

Whatever the reason, do not even think of ever transfusing with group AB blood!
anna

Thanks WisKnow.
Neither the age of the patient, not the underlying condition of the patient are known reasons for weakening of A, B or H antigens, and so I would still suggest that it is as a result of direct competition between the transferase enzymes.