Sandy L

Over the years I came to realize that a lot of what we did was geared toward simply passing inspections and meeting requirement that, in reality, did little or nothing to aid the patients.  Smoke and mirrors to confound the masses. 
I've said this multiple time on this site and still believe it strongly; "Complicating a process never made it better!" yet every time something happened everyone's first response was to add more layers to the process in a effort to make it fool proof until some new fool came along. 
Human error will occur as long as humans are involved.  All we can hope for is minimizing the impact.  This statement used to make our corporate transfusion QA folks lose their minds.  
I've kinda taken a tangent for a moment.  Getting back on track, Cliff and Malcolm I agree with you completely.

It MUST be remembered that not all antibodies react by all techniques, but, equally,it MUST be remembered that not all antibodies are clinically significant.
I remember way back in the 1980's, when I was working at a hospital in Croydon, Surrey, UK, we had an anti-S that we could only detect by tube technique.  We sent this around to a whole bunch of other hospitals, who also used a mixture of techniques.  Not one of them could detect the antibody by either microplate techniques or column agglutination techniques, both thought to be more sensitive than tube techniques, but all of those who also used tube techniques were able to detect the anti-S.
I also remember having an anti-E that did not react with enzyme-treated red cells, but only reacted by IAT with untreated red cells.  This was confirmed by the International Blood Group Reference Laboratory.
Antibodies do not read books, particularly text books!
Antibodies are only clinically significant if they react strictly at 37oC, and even then, not all are clinically significant.  Think about the antibodies against antigens in either the Knops or Chido/Rodgers Blood Group Systems.

I don't know what happened, but I tried to answer three times on here, and each time it locked up, so Dansket, I have written a Word document in reply, and attach it should you want to read it.
PathLabTalk Answer.docx

The anti-D produced by a D Variant individual is slightly different to the anti-D produced by a true D Negative individual.  For example, if we look at the attached cartoon of the amino acid residues in Partial DVI Type 1 (based on a slide by Dr Geoff Daniels), the black circles represent amino acids encoded by exons derived from RHD genes, whilst the gold circles represent amino acids encoded by exons derived from RHCE genes (see also the exon map at the bottom of the cartoon)  However, the anti-D produced by such an individual will only be produced against the D epitopes missing from the Partial D Type 1 molecule, rather than the entire D molecule.  Such an "anti-D" would, of course, react with a "normal" D antigen, as the epitopes missing from the Partial DVI Type 1 would be present in the molecule of the "normal" D (and would, of course, also react with any other D Variant that expresses these epitopes).  This can also be proved by the fact that the "anti-D" produced by one individual with the Partial DVI Type 1 variant will not react with the red cells of another individual with exactly the same Partial DVI Type 1 variant.
The same applies with all other variants (i.e. the "anti-D" produced will only react against the D epitopes "missing" from the molecule).  What must also be remembered is that changes in amino acid residues may also affect the quaternary structure of the molecule as a whole, particularly if, for example, an amino acid with an acidic side chain, replaces one that has a basic or neutral side chain (or vice versa), which could have the effect of causing the entire molecule to distort, meaning that such an individual can make an "anti-D" against D epitopes that are apparently present, but are, in fact missing, because of the affect of this distortion.
As a result of all this, the "anti-D" produced is a true allo-antibody, rather than an auto-antibody, and this would explain why the DAT would be negative, as would the "auto".
I hope this rather complicated explanation helps a little bit!
Partial DVI Type 1.pptx

Fantastic.  Can't say I'm surprised, it's been a honor honour having you here.

I interpret "Transfusion Requirement" to mean the PATIENT's requirement, i.e. "this PATIENT requires Irradiated products".  Irradiation is a unit ATTRIBUTE.  So you would need to print that patient requirement on the compatibility Tag/Label along with the other required patient information.  Our tag (Cerner) would include both unit attributes and patient requirements.

We would do an extended Gel XM as long as the current screen is positive.  It's required by the LIS.  The computer system disqualifies them for electronic XM in this instance.  If the screen becomes negative in the future, they would qualify for electronic XM as the antibody is classified as not clinically significant

...attach a note to container that the blood product is in that reads that "This product must be delivered to the patient care area without delay" or something to that effect.
In the end, everyone gets trained every time they pick up a product. This would account for all the techs and volunteers that come and get blood products depending on the facility where you work.
Just a thought :-)

Perhaps there was confusion between training and competency.  CAP requires training: TRM.40900, Blood/Tissue Sign-Out 
The procedure for signing blood and tissue out of the laboratory provides adequate protection for the potential recipient.
NOTE:  A person authorized by the transfusion medicine service must perform a clerical and visual inspection of each component immediately before it is issued. Transporters of blood components and tissue must be trained in prompt delivery. Training may consist of instruction at the time the procedure is dispensed.
Evidence of Compliance:
Written  procedures for the issue of blood components and tissue AND
Written policy for the instruction of transporters on the proper handling of the product
 
 

I interpret "Transfusion Requirement" to mean the PATIENT's requirement, i.e. "this PATIENT requires Irradiated products".  Irradiation is a unit ATTRIBUTE.  So you would need to print that patient requirement on the compatibility Tag/Label along with the other required patient information.  Our tag (Cerner) would include both unit attributes and patient requirements.

I just answered this question.


My Score PASS  

I just answered this question.


My Score PASS  

I hate to say this, as it sounds so egocentric, BUT, as I said in my first ever post on PathLabTalk in 2009 (or, if it wasn't my first one, it was one of my first ones), if people used the correct nomenclature (as stated by the ISBT - and this INCLUDES the people who produce computer systems for Blood Banks (who seem to think that they are exempt from this kind of thing), this kind of problem would never happen.
Just saying!

If you are inspected by CAP you are required to have a protocol for providing blood in emergency situations. I would suspect that CLIA has a similar requirement and your lab would certainly fall under CLIA requirements.
Your plan should be written according to the resources you have. It should also address what to do when your resources have been depleted or are about to be depleted and what to do if/when the patient is transferred (do you send blood with the patient and how would you do that). Spelling out how you would deal with replacing your depleted blood supply would be a good addition to the other information.
Crazy stuff happens even at small hospitals - if somebody needs blood badly, not being prepared to deal with providing it rapidly could be a matter of life and death for the patient. I think too often we all fall into complacency and think that just because something occurs rarely, the associated policies aren't very important. In actual fact, the things that are seldom done are the things that are most likely to be screwed up. We've just spent several years making sure that our emergency release and mass transfusion protocols are up to date, realistic and (most importantly) making sure that all staff, not just lab, are aware of them. It's paid off in better performance by everyone during those uncommon events.
 

I understand that it is easier to draw the bloods when the patient comes in for pre-op, maybe many weeks before the actual op.  Can I suggest that what would be sensible would be to bleed them then for a T&S - this way you will be prepared in case the patient turns out to have anti-nasties (this is a new scientific term coined by me as it's Friday afternoon); and then to bleed them again when they come in for the op - you will be then within the time delay as well as avoiding the patient having to come in three times (pre-op, 3 days before op, and for op); but still in time to react if there is a change between the two samples

I am looking at the Participant Summary for latest CAP proficiency, anti-D titer.  For tube testing using the "uniform procedure tube method" that CAP suggests, they reported the following results: 333 participants, mode 64, consensus range 16 to 256.  For Gel testing "uniform procedure gel method", 138 participants, mode 256, consensus range 64 to 1024.  Per CAP, Consensus is determined by the Mode +/- two of the most frequent titers.
It looks like in the previous 2 surveys for the gel anti-D titers, the mode for gel was 1 to 2 dilutions higher than tube, but so was the consensus range.  It seems like there are a fair number of labs reporting gel and if you report that as your method you should be compared to other gel titer users.  I would also think as more instruments are implemented that can do gel titers, the number reporting gel will go up. 
And of course what ever you do, perform method correlation and communicate with the obstetricians any changes they may see in titer results.  We are contemplating this also. 
Also I am little confused by ""Do you want it to be faster and more hands-off or more exact?"  It seems to me that automated titers in gel should be much more reproducible. 
We are just starting to look in to performing titers on Vision.

Another reason why our computers are better at selecting ABO compatible units than serological testing is.

Another caveat about doing titrations on the Vision is that it always runs all 10 (or 12?) dilutions.  That will burn through a lot of reagent cells unnecessarily on a titer of 4!  I agree with those above that it is critical that the OB/GYNs know that you are using a method that gives different results than their textbooks are based on.  Every gel titer result should go out with a comment explaining how its results correlate to the literature for further evaluation of the pregnant person.  At least nowadays they are likely to follow with Doppler ultrasounds rather than riskier, invasive amniocentesis.  I think a review of the CAP survey results is very enlightening.

I am looking at the Participant Summary for latest CAP proficiency, anti-D titer.  For tube testing using the "uniform procedure tube method" that CAP suggests, they reported the following results: 333 participants, mode 64, consensus range 16 to 256.  For Gel testing "uniform procedure gel method", 138 participants, mode 256, consensus range 64 to 1024.  Per CAP, Consensus is determined by the Mode +/- two of the most frequent titers.
It looks like in the previous 2 surveys for the gel anti-D titers, the mode for gel was 1 to 2 dilutions higher than tube, but so was the consensus range.  It seems like there are a fair number of labs reporting gel and if you report that as your method you should be compared to other gel titer users.  I would also think as more instruments are implemented that can do gel titers, the number reporting gel will go up. 
And of course what ever you do, perform method correlation and communicate with the obstetricians any changes they may see in titer results.  We are contemplating this also. 
Also I am little confused by ""Do you want it to be faster and more hands-off or more exact?"  It seems to me that automated titers in gel should be much more reproducible. 
We are just starting to look in to performing titers on Vision.

I am looking at the Participant Summary for latest CAP proficiency, anti-D titer.  For tube testing using the "uniform procedure tube method" that CAP suggests, they reported the following results: 333 participants, mode 64, consensus range 16 to 256.  For Gel testing "uniform procedure gel method", 138 participants, mode 256, consensus range 64 to 1024.  Per CAP, Consensus is determined by the Mode +/- two of the most frequent titers.
It looks like in the previous 2 surveys for the gel anti-D titers, the mode for gel was 1 to 2 dilutions higher than tube, but so was the consensus range.  It seems like there are a fair number of labs reporting gel and if you report that as your method you should be compared to other gel titer users.  I would also think as more instruments are implemented that can do gel titers, the number reporting gel will go up. 
And of course what ever you do, perform method correlation and communicate with the obstetricians any changes they may see in titer results.  We are contemplating this also. 
Also I am little confused by ""Do you want it to be faster and more hands-off or more exact?"  It seems to me that automated titers in gel should be much more reproducible. 
We are just starting to look in to performing titers on Vision.

Sorry Malcolm.  Mea Culpa.  Can I plead old age and being very tired?

I know this post is almost 4 years old, but it was quoted a couple of hours ago and I find the highlighted part interesting.
Not sure how big of a facility you are, or how much you transfuse.  We're decent size, we transfuse about 24 - 28k RBCs a year.  I've been at my current location for 26 years.  That's approaching three quarters of a million red cells we've retyped in my time there.  I recall only one unit we received from a facility that was mislabeled.  That's pretty good evidence that retyping, while required, does not add a lot of value.  The odds that something would go wrong are beyond astronomical.  You'd have to have a mislabeled unit, you'd be on downtime, it would mistakenly get into the retyped area, then it would have to go to a patient where it could cause harm, then the odds that it actually would cause harm are still small.

Found early mornings and evenings are best.   A martini in hand doesn't hurt either.

My impression is Ortho used the "periodically" term as a CYA. I think it's sufficiently vague to be defensible by Ortho when there are problems with the reagent: "well, did you do your periodic QC?" and also vague enough that people like me can completely disregard it without being out of compliance, technically: "I define periodically as the 7th of never, unless inconsistencies are noted."

Well I did it!  With help from the forum I created my own electronic crossmatch flowchart!
 
ELECTRONIC CROSSMATCH FLOWCHART.doc