2 cell verses 3 cell screen

[PattiJoNye]

One of the hospitals that I am working with needs some good documentation so that their Medical director will let them switch to a 2 cell screen. Any suggestions where I might find some?

[Malcolm Needs ☆]

Hi PattiJoNye,

I have attached a short monogram on the subject that I wrote for one of the hospitals served by my Reference Laboratory 2 or 3 years ago, who wanted to know almost exactly what you want to know yourself.

I very much doubt if your Consultant will be swayed by the monogram itself, by I do cite 4 references that may be of use (particularly the Transfusion Editorial by George Garratty).

I hope this proves to be of, at least a little, use to you.

Best wishes,

Malcolm

Three cell screen versus two cell screen.doc

Edited September 20, 2009 by Malcolm Needs
Spelling - AGAIN!!!!!!!!!!!!!!!!

[Bill]

Last time I inquired with Ortho about this(approx 3 yrs ago), the technical person stated that the antibodies detected by the 2 cell product is the same as the antibodies detected on the 3 cell product. Why increase usage of gel tubules by 50% or the number of AHG tubes by 50% when there is no increase in the number of antibodies that can be detected?

[PattiJoNye]

Thanks. Hopefully this will help.

[PattiJoNye]

THanks

[adiescast]

Last time I looked at the two cell screen, many of them did not have homozygous expression of the Jka and/or Jkb antigens. Admittedly that was a long time ago...but it was a matter of concern to me. Has that changed with the current two cell screens?

[knittie5]

I love the 3 cell vs the 2 cell screen for positive ab screens. it is sooo helpful in ab id use with the homozygous cells.

[Malcolm Needs ☆]

Last time I looked at the two cell screen, many of them did not have homozygous expression of the Jka and/or Jkb antigens. Admittedly that was a long time ago...but it was a matter of concern to me. Has that changed with the current two cell screens?

This is certainly so in the UK. I would be surprised if it were not so everywhere now that electronic issue is becoming de rigeur.

[David Saikin]

If you are doing antibody id's then I think the 3 cell screen provides more pertinent information. If not, the 2 cell screen will probably suffice. As stated above, there is the caveat that you are not always going to get homozygosity in the Duffy's and Kidd's - and Malcolm is right on about the e-crossmatch. If I was doing that, I'd want the 3 cell version.

[BSIPHERD]

Last time I looked at the two cell screen, many of them did not have homozygous expression of the Jka and/or Jkb antigens. Admittedly that was a long time ago...but it was a matter of concern to me. Has that changed with the current two cell screens?

You can request that you receive a lot that has Jk(a+b-) and Fy(a-b+) cells (there's generally a name for it, such as primary lot).

[carolyn swickard]

2 cell screens never have cells with the f antigen (they are only R1R1and R2R2) - the 3 cell screens have a rr cell. Anti-f can be quite clinically significant and occurs more often than Cw or Dia, which are also frequently missing unless you order the special screens.

[Likewine99]

We have been on the 2 cell screen for as long as I can remember and we perform panels.

Price is a consideration, one less gel well, 2 cell screen set is cheaper than the 3 cell set.

[TimOz]

Well Australia may be different, but down here the 2-cell screen is designed to detect antibodies implicated in HDFN.

The 3-cell screen is designed to detect these antibodies but also those implicated in causing transfusion reactions. The main difference is the Jk homozygous cells. It is very difficult for any manufacturer to make a 2-cell screen with all the antigens and with the correct zygosity and most will not guarantee a 2-cell screen will have homozygous Jka and Jkb cells on every batch. Gel users should be aware that gel is not very sensitive to Jk antibodies and they are a common cause of delayed transfusion reactions (the most common cause in the UK). I don't think you should make it worse by using heterozygous screening cells.

And while a 2-cell screen is cheaper to run than a 3-cell screen, it is inferior. I like to suggest WIIWYD to people who are trying to balance test performance and reagent cost. (What If It Was Your Daughter?)

So, 2-cell screen is for antenatal testing ONLY and 3-cell screen is for antenatal and pre-transfusion testing.

[Malcolm Needs ☆]

2 cell screens never have cells with the f antigen (they are only R1R1and R2R2) - the 3 cell screens have a rr cell. Anti-f can be quite clinically significant and occurs more often than Cw or Dia, which are also frequently missing unless you order the special screens.

I'm not convinced that an anti-f is all that clinically significant.

It has only every been implicated in mild cases of HDN and mild to moderate cases of delayed haemolytic transfusion reactions.

In terms of numbers of cases, you will find quite a few in the literature that implicate anti-f, but in almost all cases the clinical sequalae are mild, and, to a certain extent, the fact that these cases are still written up strongly suggests their rarity (you do not read too many cases studies of, for example, anti-K causing HDN or a transfusion reaction, purely because it is well known to cause these problems, and does so all too frequently).

[Malcolm Needs ☆]

Well Australia may be different, but down here the 2-cell screen is designed to detect antibodies implicated in HDFN.

The 3-cell screen is designed to detect these antibodies but also those implicated in causing transfusion reactions. The main difference is the Jk homozygous cells. It is very difficult for any manufacturer to make a 2-cell screen with all the antigens and with the correct zygosity and most will not guarantee a 2-cell screen will have homozygous Jka and Jkb cells on every batch. Gel users should be aware that gel is not very sensitive to Jk antibodies and they are a common cause of delayed transfusion reactions (the most common cause in the UK). I don't think you should make it worse by using heterozygous screening cells.

And while a 2-cell screen is cheaper to run than a 3-cell screen, it is inferior. I like to suggest WIIWYD to people who are trying to balance test performance and reagent cost. (What If It Was Your Daughter?)

So, 2-cell screen is for antenatal testing ONLY and 3-cell screen is for antenatal and pre-transfusion testing.

I agree with an awful lot of what you have said here Tim. The thing is though, in the UK the "zygosity" of the screening cells is quite firmly governed by the BCSH Guidelines, and all of the 2 cell screening cells are either S+s- or S-s+, Fy(a+b-) or Fy(a-b+) and Jk(a+b-) or Jk(a-b+).

I think that it may be an advantage to us that, in terms of geography, the UK is not all that big, and so the number of vials of the screening cells that need to be produced is fewer than would be needed for a larger country. It follows, therefore, that the number of units diverted from the pool used to transfuse patients to make reagents is quite small (even when taking ratios into account).

[Oldtimer]

I agree that the zygosity of the cells is the important issue. When institutions started looking at immediate spin crossmatch in lieu of the antiglobulin phase, it became necessary to examine the makeup of the screening cells very closely. Most times it was found that 3 cells were needed to get the desired homozygous cells. That again is the important consideration for an institution using the electronic crossmatch.

[TimOz]

To Malcolm (and anyone else),

What are you thinking about the changing ethnicity of your population.

Statisticaly in Australia based on 1999 numbers we are:

70% Anglo-Celtic but dropping to 62.3 by 2025

1.4% West Asian/N. African - estimated to be 4.8% in 2025

1.3% Indian/Sinhalese - estimated to be 2.1% in 2025

2.5% S.E Asian - estimated to be 5.5% in 2025

2.7% N. Asian - estimated to be 6.6% in 2025

That means 20 - 30% of our population will NOT be Anglo/Celtic/Mediteranean.

We have recently seen a few clinical cases of serious antibodies that are not detected with our normal (read designed for Caucasians) screening cells. This is in a situation where we use electronic blood release and rely on the negative antibody screen for blood safety. Quite a few Australian scientists have been phlegmatic about this issue but as you would imagine, a few clinical cases can rapidly change people's minds.

And another thing. I think many people fail to really critically examine their antibody screening cells. People often think that a negaitve screen means no antibodies. I see publications that say things like this: "The screening cells covered the major blood group antigens and confirmed hoimozygous expression for red cells antigens, as recommended for blood transfusion services in the UK."

They then present the incidence of E, Jk etc and not mention other antibodies that are not considered important in Caucasian populations. If you don't look, you won't find them! These studies need to publish the antigenic make up of the screening cells used - in detail.

I fully recognise that there is a balance between the testing economics and the chance of patient harm (as you rightly point out in your 3-cell vs 2-cell disertation when discussing enzyme only antibodies). BUT, I think that Immunohaematology is a numbers game - and the numbers are changing. When does our population change so much that the historical data we rely on is no longer safe to extrapolate with? Certainly studies looking at antibody incidence and specificity in non-Caucasian ethnic groups using screening cells designed for Caucasians have been shown to be wildly incorrect.

Your thoughts.

[Malcolm Needs ☆]

Yes, once again I am forced to agree with you Tim (you're right, people will begin to talk)!

There are areas within England where the "native" Caucasian population (whatever that may be, when you consider that we have been invaded by the Romans, the Vikings, the Normans, et al) is in the minority.

I have often thought that we should have a panel that includes a Js(a+b-) cell (or, at the very least, a Js (a+b+) cell) and, increasingly, there is a need for such red cells as Di(a+) on the panel, BUT, we would have to have at least 4 screening cells to accommodate all possibilities.

Most of our hospitals do still cross-match (some perform electronic issue, but it is still a tiny minority) and so will tend to detect the kind of antibody to which you allude.

We are, without doubt, detecting more and more anti-Us, anti-Fy3s anti-hrSs and anti-hrBs, but, of course, these would all be detected with the present screening cells (albeit, the latter two are a pain to get a true specificity).

I feel that, "screening cell wise", we are okay at present, but you are absolutely right when you say that a watch needs to be kept and changes made when appropriate.

[eric1980]

Hi PattiJoNye,

I have attached a short monogram on the subject that I wrote for one of the hospitals served by my Reference Laboratory 2 or 3 years ago, who wanted to know almost exactly what you want to know yourself.

I very much doubt if your Consultant will be swayed by the monogram itself, by I do cite 4 references that may be of use (particularly the Transfusion Editorial by George Garratty).

I hope this proves to be of, at least a little, use to you.

Best wishes,

Malcolm

*Clicks on imaginary "Thanks" button for reading the monogram*

[Malcolm Needs ☆]

My pleasure Eric.

[Malcolm Needs ☆]

Something unusual, if not absolutely unique has occurred; I've had a thought!

For those of you who detect anti-Cw and nothing else out of the ordinary with either a two cell or three cell screen in the plasma of a dce/dce patient, would you:

a) give rr by electronic issue?

give rr by immediate spin cross-match?

c) give rr cross-match compatible blood?

d) test the rr units for the Cw antigen?

e) other?

I am just interested.

:confused::confused:

[David Saikin]

Just as an added support of Tim's comments above - years ago, while we were still doing AHG xm's on everyone, I was the night tech at a large tertiary care facility. When I got to work the evening staff was perplexed - negative absc, 3/4 units 3-4+ at coombs', routine abid panel - negative. I repeated their work and agreed with it. Setting up a panel of Low incidence ags I found anti-Kpa,-Jsa and an anti-Lua. This one would have slipped through the cracks even today. You would think that there would be some incidence of transfusion reactions with cases like this, but I have not seen much in the literature about any. Malcolm - what is your august opinion and/or experience with these LI antibodies at your reference facility? Thanks in advance

[Malcolm Needs ☆]

Hi David,

In the case of each of these antibodies, we would advise the hospital to give IAT cross-match compatible blood, rather than antigen negative blood. Whilst transfusion reactions are not unknown for anti-Kpa or anti-Jsa, they are usually extremely mild (more a case of the red cells not lasting as long as would be expected, than a clinical reaction), but if the antibody is strong enough to detected the antigen in a cross-match, then don't give that particular unit.

I have never heard of an anti-Lua causing any reaction at all.

The problem is that, when a person makes an antibody against a low incidence antigen, they tend to make several antibodies against low incidence antigens (as in the case of your patient) and so I would always advocate that a cross-match be performed (just in case the unit to be given is negative for the Lu(a), Kp(a) and Js(a) antigens, but positive for another low incidence antigen that the patient may have made an antibody against).

That having been said, there must, by the law of averages, be patients around who have received blood via electronic issue that is not compatible, but which has not caused a clinically significant transfusion reaction. Certainly, ever since the Severe Hazards of Transfusion (SHOT) Scheme has been going in the UK (about a decade now I think) there has not been a single report of a transfusion reaction due to electronic issue; there have been quite literally hundreds involving serological cross-matches, which begs the question, for those still unconvinced, which is safer?

[adiescast]

Something unusual, if not absolutely unique has occurred; I've had a thought!

For those of you who detect anti-Cw and nothing else out of the ordinary with either a two cell or three cell screen in the plasma of a dce/dce patient, would you:

a) give rr by electronic issue?

give rr by immediate spin cross-match?

c) give rr cross-match compatible blood?

d) test the rr units for the Cw antigen?

e) other?

I am just interested.

:confused::confused:

We fall back to AHG crossmatch for pretty much any antibody outside of the ABO group (for a patient with anti-A1 we go to immediate spin crossmatch of type O blood. Our computer system will not allow electronic crossmatch if there is anything unusual in the patient's history). The other exception would be passively acquired antibody.

We do not distinguish between "clinically significant" and "clinically insignificant" antibodies.

[Malcolm Needs ☆]

Thanks adiescast.

I tell everyone why I am asking after there have been a few more answers (if any).