Missed Kells on Echo

[Jeff3344]

Hola everybody!!!

We are trying to bring an Echo on board.

We have a known K (Kell) in house, so we've had plenty of sample to work with.

And the Echo misses it every time.

Sure, you have a 90% chance of getting a compatible unit, but that is more than I'm willing to wager on.

And for pre-natal issues and such...

We can't be taking this sort of chance.

I did a bit of a test today. I took our known Kell, and spiked the tar out of a sample of [X] who was of the same antigen negative, and same blood type. I put a LOT of anti-K in [X] . Both patients were drawn on the same date. That would be yesterday.

Yep! Both came out negative on the Echo!

So I did the same samples in Gel as well. Yep! Getting 2-3+ in gel on our in-house, and a good 4+ on my spiked sample. That would be [X].

I had stir-balls in all the proper reagents, all the QC was done, we passed all our initilazation tests.

Either we are doing something horridly wrong, or this thing/our lot/ect... skips Kell.

Has anyone elce tried it?

Any info would be helpful because I'm voting for "No confidence" on this as an analyser. One skipped antibody is one too much in blood bank.

[bbbirder]

So you added reagent anti-K to a specimen? Was this monoclonal?

We have an Echo now and it has never missed a K. (It has had 'difficulty' with some anti-E's...for the most part these were weaker but "eye-readable". The specimens were submitted to Immucor. Not all reports are back yet, but at least one of them was probably an IgM.)

E and K are both antibodies that can occasionally be naturally occurring.

Any chance this is an IgM? Have you submitted it to Immucor?

Linda Frederick

[AMcCord]

I've picked up 2 nice Anti-Ks with Echo over the two months. 3-4+ with Echo and 1-2+ with Gel. Echo has produced 'eye-readable' results for a couple of anti-Es and 1 anti-Fy(a) that showed nicely in Gel. I also picked up an anti-E that Gel missed entirely and an anti-Jk(a) that was almost too weak to see in Gel (and unidentifiable with Gel or PeG).

Keep in mind that no one system or method is going to catch all antibodies. That's a fact of life that we blood bankers have to live with. Select the method that you think will work well for your patient population. And also remember, you have most likely missed an antibody or antibodies with Gel and just don't know it (and with any other method you have used routinely).

[RR1]

Hi Jeff,

Could you try and see if you can detect the anti-K by Tube technique?- this data would be very useful for me.

I have had quite a few problems similar to yours and have subsequently reported them to my Medical devices body (MHRA in the UK, who are now investigating).

Whenever we change techniques, the detection of clinically significant antibodies MUST be as good as, or better than first generation techniques (tube) and good as or better than second generation techniques (gel), to ensure patient safety.

Rashmi

[johna]

Jeff,

Did you happen to test a different lot# of Capture Ready-Screen to see if it was perhaps a problem only with the lot you were using?

We've detected a couple of anti-Kells on the Echo, which is certainly not to say that we haven't missed any. But I have to totally agree with our friend from Nebraska who points out the fact that no testing system is perfect. To reach the highest level of sensitivity one should ideally run every specimen by both solid-phase and gel (and maybe throw a PEG tube screen in to boot).

[John C. Staley]

Jeff, I know Immucor would love to get their hands on a aliquot of that anti-K you are working with. I have to repeat what Ann said, no system is 100% specific and 100% sensitive. That is just a fact of life when working within biological systems. Try to get more samples of anti-K as see if you can duplicate the non-detection you are seeing. My bet is that your current anti-K is an anomally (sp?) which you will be unlikely to repeat with any other anti-K you can find.

:abduction

[RR1]

Hi Jeff,

It would be a good idea to freeze some aliquots and run them with the subsequent lot of capture plates for next few months. Run this at the start of a new lot and close to lot expiry date.

[Injun Blood Doc]

Jeff,

Any chance the anti-K has a major IgM component. Capture does not detect IgM, wherease the gel does.

[dingalls2]

We have a relatively new Kell (Nov) on a multiply transfused pt. It shows 2+ in gel and not on the Echo-or manual Capture backup. We assume it's because it is IgM, so I agree it misses some.

[macarton]

We are in the middle of validation of our Echo. We are a gel user. We have a known Kell, 1+ reaction in gel. It's negative on the Echo. We have two other patient's with weak E (one with Jka also) on gel, one was negative the other one only showed the Jka on Echo. We had another patient with an anti D, the screening cell 2 was positive on the Echo and the panel completely negative. Mary C

[RR1]

I'm trying to develop a better understanding of Capture, please could someone actually explain how the Capture screen manages to avoid detecting these 'insignificant IgM antibodies'. Is this specifically a feature of the Capture-R strips, the incubation of the plate or the AHG reagent used?

Also, if anyone uses this techniques to perform donor/ patient compatibility testing- how does this reduced ability to detect IgM antibodies ensure that ABO incompatibilities are not missed ?....may be i'm just getting a bit confused ??!!!

Many thanks

Rashmi

[AMcCord]

ABO incompatibilities are not reliably detected by AHG methods of any kind. That is where the immediate spin crossmatch comes in (or the electronic crossmatch).

[dmorrissbb]

The indicator cells are only coated with anti-IgG. An IgM antibody will attach to a corresponding antigen if present, but you will not detect it because the idicator cells are not coated with anti-IgM. Hope this helps. -Darlene M.

[RR1]

Thanks- but Gel techniques also use monospecific anti-IgG reagents- so why would an antibody be detected by Gel and not by Capture ? - what is it about capture that makes it more specific.

I have also been told that Capture may not detect certain IgG subclasses of antibodies (IgG2and IgG4)- so does this mean the indicator cells are coated only in IgG1and IgG3? - so would we all be happy to transfuse a patient with an anti-Fya or -K that was detrermined to be an IgG2 or4 subclass ?-I certainly would not be, and would consider this unsafe practice.

Also could the detection be a feature of adhering the red cell stroma to the wells ?

Thanks

Rashmi

[Linda0623]

Hi all--

One little thing I thought I would throw in the mix for all to consider.......

The fact that the ECHO has some difficulty in interpreting reactions of 1+, sometimes even 2+, applies to screens and panels as well......

What I have found, is that scrutiny of the "0" reactions compared to the negative control background on the ECHO, helps give you a clue as to what is going on........

When I was trained on manual Capture by Immucor, they specifically taught that you should compare the backgrounds on the panels to the negative control. Any shading of color that is different in the background, or fraying edges of the cell button beyond the amount evidenced in the negative control, should be suspected as a weak +.

Several times already, we have seen this, especially with Anti E's, but I recently also saw this with a Kell. A ? reading is given if the camera determines a reading greater than 5....the reading the ECHO gave was a 7.....I compared that answer to the cell that read 1+, and it was a reading of 20. Visually, comparing the 2 wells was not all that different. comparing the ? cell background to the negative control background indicated that there was indeed a weak reaction.

These are difficult, because everyone's eyes are different, and I suppose you could say that we wanted to make this work....but in several instances, for validation purposes, I would run IGG crossmatches of known E pos/neg or Kell pos/neg cells to check compatibility. The Ag neg cells, (so far) were always compatible, and the Ag pos. cells were incompatible in all cases except one........

Not too shabby......

Sorry so long, but I think part of the concerns stem from the time needed to read this methods weaker reactions with confidence and experience....like everything else.........

[RR1]

Thanks for the info...so you had a problem with one of the antigen positive crossmatches- was the reaction '?' or negative?- What if one of your techs had inadvertantly issued this unit ? - who would the responsibility lie with if the patient had a TR ?

Thanks

Rashmi

[Linda0623]

Hi Rashmi.....

I talked about this a little bit on a different ECHO thread, but I am not considering using the IgG crossmatch assay for crossmatching purposes per se...... and the reaction I got on the AG + unit was 1+.....

I am using the crossmatch assay on the ECHO to "pre-select" units for antigen typing.....if it's not compatible, I don't screen it for the required antigen, as it is actually faster and cheaper to crossmatch than to screen ransomly for certain antigens.

In the instance I mentioned, I purposely "crossmatched" known E or K units for validation purposes. I wanted to see if in fact on questionable or cannot rule-out situations, could I pick up incompatibility on the ECHO that could be potentially missed if I didn't notate the antibody specifically E or Kell. The positive E unit that I tested, would not have been normally selected to crossmatch, and our computer system would not have allowed it's selection due to the antigen typing results entered on that unit, so they couldn't have easily inadvertently used this unit for the patient.

Also, as I mentioned above, we do IS and IGG crossmatches on all patients with + antibody screens in tube on the bench.....luckily, on a daily basis we do not do that many IGG crossmatches and therefore do not see the value of having either 2 techs do one crossmatch or see the efficiency of doing only the IGG crossmatches on known Ag Neg blood for efficiency purposes.

Hope this isn't too confusing......

Linda

[RR1]

Thanks Linda, -that does clarify things. Do your trainee staff find interpreting reactions easy- or are only your experienced staff let loose on the equipment ? and are you comparing the results against the computer screen or from the printout ( we are only allowed to print using black and white toners- the image is quite poor).

Best wishes

Rashmi

[Linda0623]

Well....this is a whole topic in itself......

I am doing my best to "bring everyone along" with me as I gain better experience and understanding of how Capture works.....or doesn't. Daily I go over the current workups and their statuses and next steps with my day shift, and as feasible with my evening shift. We are predominently a specialized hospital and donor center serving Orthopedics with some neurology, oncology and cardiology thrown in for good measure, but only on adults, and we don't have a general Emergency Room or OB services.

What that does, generally speaking, is to give us time to figure things out, as most surgeries performed are elective, and planned....although we do take "stable" orthopedic traumas....(snicker)......what we do get however, are some pretty complicated cases, as we get people from all over the country and beyond that come here because of our world renowned surgeons. I've had a Bombay patient, several cartwright a patients, and a patinet withan IgG reactive York with 4 other allo antibodies for that used 12 units of blood before we gave him the York and the Fya.........

The ECHO, specifically Capture methodology, in my opinion overall has made us a better transfusion service. My techs, including myself, have to think, and evaluate what we get much more thoroughly than previously, because we have to get comfy with what the ECHO is telling us..... many things are alot harder to dismiss as ACARO (our term for all clinically significant antibodies ruled out). We have had instances where because we had done previous work in tube that was negative (pre admission screening) that when we ran the OR sample on the ECHO we would get positives we were not expecting that were sometimes strong, sometimes weak, but generally very significant. Nothing like those to inspire confidence, hee-hee---but it's part of the learning curve......

That being said, I would say that about 85% of the time at this point, my techs complete and enter the results of the work-up without my intervention. If they get a positive screen and a negative panel, they have to go over it with myself or my other super user before final decisions are made. I try to make sure as I said above, that I bring them along with me.....it doesn't benefit me to have them lag behind......

An example of this was while I was on vacation last week, they got a RS3 that was 2+ on cell 2, but the panel was negative. I had them run the Extend I panel, and one cell came up positive. I looked at the Masterlists, and had them run the third panel I had available. It also came up one cell positive. The common thread, they were the only 2 Kpa positive cells I had available. The catch?: why was the antibody screen positive? Cell #2 on the RS3 they ran was Kpa Neg........Hmmm.........Still working on this one!!!(Lucky she doesn't go to surgery until next month!!)

Did I mention the learning curve?????

Best Regards,

Linda

[RR1]

Hi Linda,

This methodology seems to require a lot of staff training- and in your situation you can afford to investigate thoroughly all results obtained by your staff. This is not realistic in a trauma care setting- when decisions have to be made quickly. Also in the case of obstetrics and urgent unplanned procedures.

Antibody screening and identification must be simple to perform and interpret for 99% cases - especially as staff at my place are lone workers at night and need to make these decisions (they are also mainly haematology folk- with basic transfusion serology experience). As staffing becomes increasingly multidisciplinary the need for simplicity is even more important, to maintain patient safety.

Best Wishes

Rashmi

[Dr. Pepper]

I concur with the consensus that no one method will detect everything every time better than all the others. If it makes you feel any better, Issitt points out that if you have to go to heroic measures to detect an antibody, it will probably will not do too much (immediately) to transfused antigen-positive cells. And what about the patients with an antibody to a low-incidence antigen such as Wr(a), Kp(a) etc. not present on the screening cells and hence undetected, who have the misfortune of receiving an electronic or I.S. compatible unit that in fact carries the antigen. These are infrequent but unavoidable risks of current transfusion practices.

That said, I still want to see my anti-Ks, too!

[RR1]

Thank you for that- and I agree there will always be differences in detection techniques. A lot of my concerns surround repeatability using the same testing system. When you see anti-K, anti-S, anti-Fya - not being detected in a reproducible way- with the same sample- worries do set in, and ultimately in a highly litigious society who would take responsibility if the patient had a TR ?

Also when less sensitive (tube) techniques were used for primary antibody screening- for us this was approx 10 years ago, we all accepted that we would see approx 2-3 DTRs / year caused by sub-detectable antibodies- it was not unusual. With the advent of newer techniques - this frequency was reduced significantly.

When users change their primary identification techniques- there needs to be a level of confidence that the test is better than what was previously used (especially as these systems allowed us to progress to electronic issue), and that we are not going to see an increase in transfusion reactions to the level we had 10 yrs ago.

Obviously only time will tell if any one technique has a greater association with DTRs, but this also relies on staff reporting adverse reactions appropriately. Our respective haemovigilance bodies need to collate and trend this data carefully with respect to different serological techniques used.

I fully agree that the numbers of patients we are talking about is few (hopefully !!) - but we all do need to question why we are even having this discussion.

Best Wishes

Rashmi

[huntilla]

Our limited history with the echo has detected two Kell antibodies in the four months we have been live. One sample was a previously identified Kell and one sample was newly discovered. We used Gel testing prior to implementation of the Echo and IMHO the sensitivity is equivalent. In our validation testing we had a JKa that the Gel technique was unable to detect. Both platforms have issues and ultimately are improvements on the subjectivity of manual techniques for antibody screening.

[RR1]

The antibodies we have had difficulty detecting, when taken back to Gel technique and Tube technique- showed good reactivity.

[Linda0623]

HI Rashmi,

You are right in saying that I am fortunate to have time in most cases.....but that is not to say that this method is difficult to interpret in general.....I just meant to indicate that I am an advocate of making all of my techs able to perform at a high level, and I do my best to encourage that whether they are students, new grads, or veterans like like my 30and 40+ yr blood bankers......

Overall, this method does what I need it to do.......detect clinically significant antibodies with more sensitivity and specificity than PEG in tube....we have caught more weak antibodies, even with the ? or having to visually call positive cells that the ECHO called neg, than what we have missed and know about...... Several times we have taken these weak antibodies and tried to demonstrate them in tube, and get everything completely negative. Would we do better with gel? I cannot say either way with certainty, as we have never used gel here.

What I can offer you from the technical perspective, is this:

When I investigated the idea of automating, I asked both Ortho and Immucor to come in and demo their methods for all of my staff, some of whom had no appreciable knowledge or understanding of either method. Overwhelmingly, the techs preferred solid phase, and my 40+ year blood banker just simply said "when do we get this"......

That has gone a long way in helping everyone along, and I am happy with my 85% independence of work-up completion at this point as we have only been LIVE with the method for about 3 months now......

Regards,

Linda