StevenB

I agree in part: In the hands of experienced techs it has its use, but.... Using the prewarm technique to eliminate stray reactivity is simply not an appropriate use of the technique. Under no circumstance would I ever recommend to my customers that they use the prewarm technique to get around 2-3 or 3-4 stray reactions on a panel. More significantly, the technique should not be used in the what would be the hospital's next step in this scenario which would be to find crossmatch compatible units using a prewarm technique. And most likely that would be the tech's next step. You ID a cold auto and want to eliminate it's interference, great. You ID an antibody whose classification as "clinically significant" is based on its ability to react at 37C, such as anti-M, anti-P1, anti-A1 and you know for a fact there are no additional common antibodies present, also great. But hospital blood banks should never use it to eliminate "stray reactions". Heck, we don't use the prewarm technique in our reference laboratory to eliminate stray reactions and combined we have over 100 yrs of experience. Using the prewarm for the right reasons...I'm "all in". Using prewarm to eliminate "stray reactions", not so much.

Prewarm? Hmm....

Yes.

Catching these patients untransfused is difficult. Submit a sample for full molecular phenotype. At a minimum, try to determine Rh and K type. If your antibody screen is negative, select crossmatch compatible units. If positive, the next step is to 0.2M DTT treat your panel cells to help distinguish that the observed reactivity is due to daratumumab. The reactivity that we've seen is usually variable in saline techniques, resembling what use to be referred to as "high titered, low advidity" or HTLA antibodies. The 0.2M DTT works very well, but I would not recommend any micro reads. Most common antibodies can be ruled-out with 0.2M DTT treatment, but not antibodies directed at the Kell system antigens. (There are other antigens "destroyed" by the DTT treatment, but they are not pertinent to this immediate discussion.) As such, if no underlying antibodies were detected, our transfusion recommendation is to select K- units if the patient has typed K-. While some recommend phenotyped matched units, we do not. Having a full phenotype does not absolve a blood bank from determining what the reactivity is due to and if any new antibodies have developed. Requesting phenotyped matched units, without the presence of antibody is a waste of a potential resource, drives up costs unnecessarily and may result in a delay that is longer than resolving the actual work-up if the patient's type is rare enough.

Without tubes, your ability to perform any problem resolution is very limited.....extremely limited. Rouleaux is a classic example: See it with Gel testing and there is nothing you can do except ship it out to a reference lab if you don't have tube testing available. Time is wasted and additional costs incurred all for what would be a simple resolution if tubes were available. Warm autos, cold autos, determining clinical significance of anti-M, anti-P1, anti-A1, rouleaux resolution...all of these can be resolved with tube testing. The real question is, what is the cost vs benefit of having a 3% panel available if you only use it sparingly. On a side note...our lab is not fond of Ortho cells. They are very "sticky". To a person, you could have a 12 cell selected panel from various manufacturers (which we routinely do) with Ortho cells intermixed, and we would be able to accurately identify Ortho cells on the first rock of the tube. We use them, but only because our customers use them and there can be issues with Ortho's 0.8% cells in Gel testing that are reagent dependent and disappear when converting Ortho 3% cells to an 0.8% concentration.

Under our organization's policy, a heterozygous crossout would be acceptable in a technique that is considered an "enhancement". We consider Solid Phase to be an enhancement technique so heterozygous crossouts are acceptable. Ideally, we would prefer 2 crossouts for a specific antigen, but that is not a hard fast rule....just good practice in my opinion.

Answering the original question, I don't believe you need to send every suspected anti-Bg antibody to a reference laboratory for confirmation (even though we'd like the business!). Just be cautious; make sure you have appropriate crossouts as I mentioned above.

From the manufacturer's insert: "The results obtained in this application should be interpreted with particular caution, as there is clear evidence that the reactivity of red blood cell surface antigens may be impaired by chloroquine treatment, which could lead to a negative reaction with a weakly reactive antibody other than one of "anti-Bg" specificity." Paraphrasing Issitt and Anstee (Applied Blood Group Serology, 4th Ed), Bg antigens can appear, or disappear over time. So a donor who had previously typed Bg- by a manufacturer, may later have the antigen, but appear on the panel as Bg- (or HLA- as they do now). The opposite may very well happen too: a cell marked Bg+ may very well be Bg- in the future. My only advice is, if you decide to use this reagent, you must be sure that you have performed an adequate rule-out of all other antibody possibilities as required by the FDA. With the exception of the K antigen (as not all panels have K+k- cells), I'd make sure that I had homozygous crossouts for all of the other required antigens prior to resorting to chloroquine diphosphate treatment. I'm not disagreeing with Malcolm, only cautioning that there is an appropriate time to utilize chloroquine.

More specifically, pgs 403 and 404 of the current Technical Manual. It's decent, but a better source would be to contact a reference laboratory. They should be able to answer any question you may have regarding antibody identification problem. And just on a side note... a positive autocontrol very rarely means a prewarm is to follow. In fact, never by itself. Again, a reference laboratory is a great resource available to you.

No problem John, I enjoy a good discussion. I agree totally with you when you say that there must have been some level of antibody reactivity for the antibody to have been detected in the first place. However when is comes to performing a titration for prenatal studies the only test of concern is the titration and it's results, not the PEG, LISS, Gel or Solid Phase tests that originally detected the antibody in question. Our physicians want to know whether their patient has an antibody, and if so how high is the titer. In performing that test, I am only concerned with: Does the antibody react in that test method and if so at what level. If the antibody is not reactive in the 1:1 dilution, then the test is simply nonreactive and reported out as such. That is a precise and accurate reporting of the test result obtained. Anything else is speculation as we don't and can't test for a titer of <1. Obviously, as you have pointed out, there is some level of antibody present, but if a result is not tested for, and titrations do not test for <1 titers, then in my opinion it should not be reported as such. Remember, this is a titration test...if there is a way to perform a titration of a sample to detect a titer of <1, I am "all in" on your position. There are other tests available that will quantify the amount of antibody present, but they are not the test in question. Thinking of this in other terms, when performing an antibody identification, it is not uncommon for us to use multiple techniques and there are times when the saline tube test is performed. If nonreactive in saline, but reactive in other test methods, we would never report out that the antibody was present in the saline test phase, but at a level that was visually undetectable. While we know the antibody is there in the patient's sample, it would not be accurate to report out something that we did not see. Essentially, that is what is being done when a result of <1 is reported in a titration test. As always in Blood Banking there are exceptions. The endpoint for prenatal titrations is the last tube (or column, I suppose) that demonstrates 1+ reactivity. That is a very specific parameter and it raises an interesting question: If a titration reveals only reactivity in the 1:1 tube, but the reactivity is less than 1+, what is the titer? In this situation, I would consider reporting out a titer of <1 because there is actually a visible result and a corresponding score would be reported. However, I would also consider an accurate report to be "a titer of zero, score____" since as previously mentioned the endpoint for prenatal titrations is a 1+ reaction. Ultimately, a lab is going to report what they feel comfortable with. In our region, we've had zero questions from physicians since we implemented reporting nonreactive titration studies in this manner. The same can not be said when a few of our techs were previously reporting the "<1" result.

As there is no such a thing as a titer of less than 1, I'm guessing the titration result was actually nonreactive in the test phase they used for their titration study....which is good news for you and your baby! Some labs report out titers as <1 when the first tube of the titration is nonreactive....not sure why as it is simply a negative result. Again, that is good news for you and your baby. I hope all goes well!

Welcome to the US's world of regulations.....

While I stick-by my answer above, as always there are exceptions and this topic is no different.... Long Answer: Maybe The AABB Technical Manual and the "Standards" state: "The serum or plasma of either the neonate or the mother may be used to perform the test for unexpected antibodies..." (5.17.1). It goes on to state: "If the initial antibody screen demonstrates clinically significant unexpected red cell antibodies, units shall be prepared for transfusion that either do not contain the corresponding antigen or are compatible by antiglobulin crossmatch until the antibody is no longer demonstrable in the neonate's serum or plasma." (5.17.1.3) In this situation, if the transfusion service were to go the ultra conservative route and choose to use ONLY the mother's serum or plasma in their pretransfusion testing, then YES, they would be forced to select c- units as that would be the only unit that would be compatible with the mom's (or mum's) serum or plasma. If they choose to use the neonate's serum or plasma, then we are back to the "short answer" answer above, and NO, c- units would not be required.

Short answer: No.

Tube testing has it's place. As "older" more experienced techs retire, the art of "problem resolution" will eventually fade away as hospitals become more reliant on automation. Heck, I'm aware of hospitals that are unable to resolve a simple rouleaux problem due to abandoning all tube testing. Of course, this results in delayed patient care and increased costs as they send them to us for resolution. Perhaps that's OK though as it helps add to my job security!

Yes Scott, the transfusion history or pregancy does have an impact on how the work on a patient is to proceed. First, as has been mentioned here, is the frequency of collection of the sample which I won't get into as it has been addressed. Second, the frequency in which you perform an antibody identification/rule-out is dependent on the patient's history. If the patient has been transfused or pregnant within the past three months, then a panel should be tested in the face of a positive antibody screen as long as a new sample is required (the date of the new test is outside of the "three days" (actually 4) of the previous test as allowed by the Standards). You do not have to re-identify a known antibody, but you must make sure a new antibody has not developed. In my opinion, a compatible crossmatch fails to meet the criteria for identifying a newly formed antibody as I am sure the donor's phenotype is not known therefore the criteria set forth in the Code of Federal Regulations can not be met. A compatible crossmatch does not equal a nonreactive antibody panel! If you have previously ID'd an antibody and the patient does not get transfused and has not been pregnant or transfused within the prior three months, then technically, that ID stands for the next admission....as long as the new screen reflects what was previously identified. Obviously, if you ID'd an anti-K and a K- cell comes-up positive in the screen, then it would be wise to perform a panel to ID the unexpected reactivity. Of course, in this scenario it would be wise to have full faith in the patient's ability to communicate accurately whether or not they have been transfused at another facility. A bit wordy...sorry.

If you have reactivity, you need to explore what is causing it. So in your hypothetical; the fact that you have negative screen cells has zero bearing on the positive reactions seen with the panel testing. You would not be able to report out a negative antibody screen in the presence of positive panel results.....at least I would never recommend it to any of my customers. So, if a tech was unfortunate enough to shotgun a test and set-up a screen and panel at the same time, the reactive panel would trump the negative screen...you can't just ignore reactivity.

When we come across these, we use the term: "Antibody of undetermined specificity and unknown clinical significance".

My guess is that all of the reactivity you have discribed is not due to Sda antibodies exclusively. I've been working in a Reference Lab since 88' and I can count on one hand...maybe a hand and a half, the number of anti-Sda antibodies I've worked on. As David mentioned, the agglutination is very distinctive: when observed microscopically, they can appear as tight (globs) refractile agglutinates "in a sea of free cells" (as the Antigen Facts Book mentions). Most that I have worked on were detected at AHG, and most often micro reactions or very weak macro. Very seldom at RT which goes against what the Antigens Facts Book states (not saying that info is wrong). The reactivity is enhanced when tested with enzyme-treated red cells and personally....I'd send it to a Reference Lab prior to collecting that guinea pig urine from your collegue's pet. Just sayin....

Not to make light of your situation, but if you do happen to transfuse a Kp(a+) cell and there is a response, you most likely will have resolved your situation for future transfusions. If that were to occur, I'd freeze whatever remaining serum or plasma you had from this patient. If in the future the antibody were to drop off again, use the frozen sample to do a "compatibility check" on the units you intend to transfuse. I was going to suggest that in the future, anytime you see a Kp(a+) cell or other lows on your panels, remove the cell from the test. But with the way the manufacturer's are stacking the panels with lows, you'd lose half of your panel! And if you are seeing lows on your screening cells, I'd recommend sending them a nasty gram expressing your displeasure at having lows on screening cells!

"sometimes the antibody has fallen to a titre too low to detect". This is a common statement regarding this topic. How often is "sometimes"? 1 in 50? 1 in 20? 1 in 10? Let's go with the latter... If 10% of the time a hospital sees a patient with a previously known antibody that is no longer detectable, that means that 90% of the time they are performing an unnecessary test as the screen will be positive and additional testing is required. Doesn't this seem a bit odd? More importantly, under this scenario, 90% of the time charges are being generated that are unnecessary. I think the 1 in 10 scenario is actually a conservative figure considering the common use of Gel screening which is one of the most sensitive test methods blood bankers use. Personally, I think my focus would be more on what is more likely to occur than what may occur. That doesn't mean I'm right, it just is the approach I'd use.

If working in a hospital, I guess I'd make my decision based on how often I end-up doing a panel, or an abbreviated selected panel, after confirming the previously known antibody is still reactive. Just taking a wild guess, I'd imagine that is the majority of the time. If so, then it seems to me that it would be more cost effective just to start with a selected panel.

Not sure if you are still confused about IgM and one's ability to detect it..... Paraphrasing Malcolm; complement is not required to be present for an IgM antibody to bind to it's cooresponding antigen. With or without complement, the IgM antibody will bind to the antigen which most likely will result in visable agglutination. If detecting hemolysis is your end goal, then yes you would need a source of complement (serum or antibody free "fresh serum") to accomplish that goal.

Per shily: "The fifth submission the results were: titer 8, score 34 for the current sample. Tested in parallel, the previous sample's titer and score were 2, and 15. Per the technical manual and many other resources, the titer result was considered clinically insignificant. StevenB,your post is marvelous, but I think titer 8 and 2 were significant, do I remember wrong? I t is more than 2 tube of titer." Thanks shily. I would have responded earlier but I'm not seeing beyond the first page in these posts. We follow the Technical Manual's recommendation when determining clinically significant increases which states: "In comparative studies, a significant difference in titer is three or more dilutions." In this instance, the first observable titer was 2. That sample was frozen and run in parallel with the next sample that demonstrated a titer of 8. That is a two tube increase and per the TM, not a significant change. I suppose though, one could argue that when compared to the initial test results, where the antibody was "nonreactive in the test phase used for titration studies" (not <1) that the titer of 8 fell into the "three or more dilutions" increase. Right or wrong, we only do the comparison to the previous sample. But it is food for thought, thanks much.

Pet peeve topic.... There is no such thing as a titer of "less than one". It either meets the reaction criteria (generally 1+ is the standard, but other strengths might be in use) or it doesn't. The first tube in a standard titration contains nothing but raw sample and cells...so I'm not sure how you would treat the raw sample to obtain a titer of "less than one". A sample that has a +w reaction in tube one should be reported out as a titer of 0 if your standard for interpreting a titer result is higher than the +w reaction observed. Ok....I feel better now.