So we are thinking of going to Electronic crossmatches. What are your opinions on this. I am so used to AHG crossmatches, it may take some getting used to mentally! What qualifies a patient for electronic XM? Neg screen, history...what else? any tips on validating? I think we are just going to look at a years worth of crossmatches and pull data from there... Although it will save on cost, it makes me nervous. I know we are behind on the times!

If you are in the US, there are pretty specific rules for who qualifies for E-XM. First you have to have a computer system whose software is designed by the vendor to do it. Units must be scanned in (especially blood type) and retyped in the system. Patients must have at least 2 blood types on record either both now or one from history. Although it is not an actual requirement for the E-XM to do the two blood types on separate specimens it is obviously a better idea than doing it twice on the same one. There must be no history of clinically significant antibodies and in most software, the current antibody screen must be negative (even if it is only passive anti-D or a cold agglutinin). The software must be capable of checking that all of those conditions are met before allowing E-XM. Validating would be to prove that it would work right if conditions were met and not be allowed if they weren't all met.

Everyone must understand that, like the Immediate Spin XM, it is simply a check on ABO compatibility and there could rarely be an antibody to a low freq Ag that will be undetected. Odds are that only one unit out of many would be incompatible so it hasn't proven a frequent problem. And obviously it can't be done if the computer system is down.

The post from Mabel pretty much reflects the position in the UK.

You can find our Guidelines in the Library section of this site, or put BCSH Guidelines into your search engine.

For US members, I just uploaded the FDA Guidance for Electronic Crossmatch April 2011 to the library.

PS it won't be there until it is approved.

Edited June 8, 201114 yr by tricore
additional info

So we are thinking of going to Electronic crossmatches. What are your opinions on this. I am so used to AHG crossmatches, it may take some getting used to mentally! What qualifies a patient for electronic XM? Neg screen, history...what else? any tips on validating? I think we are just going to look at a years worth of crossmatches and pull data from there... Although it will save on cost, it makes me nervous. I know we are behind on the times!

Wow, that is a big jump, to go from AHG straight to EXM. We just stared doing EXM a few months ago, and love it! We had been doing IS for years and years, so this wasn't much of a change, as far as thinking went. Just the time saved making cell suspensions for an IS is great.

Once you do it for a while, you will love it.

:cool::cool::cool::cool::cool::cool::cool:

I agree Elizabeth; it is a big jump - but it is not a jump about which one should be frightened.

If you remember, it was a big jump when Ottenburg first suggested cross-matching. It was a big jump to abandon minor cross-matching. It was a big jump to abandon using albumin techniques. It was a big jump to abandon enzyme screening. It was a huge jump to abandon IAT croos-match in favour of IS cross-matching. It was a big jump to abandon clotted samples, instead of using EDTA samples, because we were so frightened of missing "complement-only" antibodies, but we have, and so have the patients, survived them all (apart from the odd occasions, that usually involved laboratory or other mistakes, that could have been avoided).

As long as your computer system is thoroughly validated, go for it.

I agree Elizabeth; it is a big jump - but it is not a jump about which one should be frightened.

If you remember, it was a big jump when Ottenburg first suggested cross-matching. It was a big jump to abandon minor cross-matching. It was a big jump to abandon using albumin techniques. It was a big jump to abandon enzyme screening. It was a huge jump to abandon IAT croos-match in favour of IS cross-matching. It was a big jump to abandon clotted samples, instead of using EDTA samples, because we were so frightened of missing "complement-only" antibodies, but we have, and so have the patients, survived them all (apart from the odd occasions, that usually involved laboratory or other mistakes, that could have been avoided).

As long as your computer system is thoroughly validated, go for it.

I didn't mean to scare anyone! I agree Malcolm, it is certainly attainable, and the computer software and validation is the biggest piece.

ElinF, what prompted this change? A new BB Supervisor? Just curious. (I know, it killed the cat.)

To those of you doing ECM's: when you have a patient who does not qualify because you have only one ABO on record, what options do you give your techs to get the second one? Do you allow using a specimen from another department even if it's not properly labeled? Does it have to have been drawn at a different time? Will you have the patient restuck just to get a second ABO?

We have not implemented EXM yet, but plan to do so soon. In our hospital, blood bank specimens are labeled no differently than other lab specimens. We will allow techs to pull an EDTA specimen from hematology (as long as it is not too old to be tested) to perform a needed second type. We will require the second type to have been from a different phlebotomy from the current type. We will not insist that it be drawn by a different collector. If there is no other usable EDTA in lab, we will require the patient to be re-stuck.

A. When a type & screen is ordered for pre-transfusion testing, look the patient up in the computer system to see if a previous type is documented.

B. If there is a previous type on file, proceed with testing on the current specimen.

C. If there are no prior types on file, repeat the ABO/Rh type on the current specimen when the patient meets any one of the following:

1. Is Group O,

2. If an autologous unit is in house and has been confirm typed (and the type agrees with the specimen type)

3. If the patient has a blood bank armband and the specimen has the armband number on it.

D. If there are no prior types on file and if the patient types as A, B or AB, order a BBRPT and obtain another specimen.

1. The specimen can be from another specimen in the lab as long as it was drawn at a different time from the BB sample. For example, if there is a CBC tube drawn on a different day or time, it would be acceptable.

2. If another specimen is not available, have one drawn. Inform the nursing unit that another specimen is required.

3. If blood is needed prior to completion of the second type, issue group O Negative RBCs.

Yep to many of the posts. I used it for many years at one Institution (and hope to implement it here) and as EDibble noted; Once you do it for awhile, you will love it.

The staff at my current Hospital know this is in my plans and they are very nervous. I think that is normal. It will just take time for them to see the safety of the system.

Just wondering about your comment regarding looking at "past" records for your validation. I think you need to validate the electronic crossmatch function itself; not just look at past records of a system you know worked. For example, if you have an antibody patient and had to perform AHG crossmatches, just because your computer system noticed if you accidentally only performed an I.S. crossmatch instead of AHG, does not mean that if you tried to get away with an electronic crossmatch, that your computer would detect that and appropriately warn and/or forbid certain functions. But there probably are some things you could use from looking at previous records. Having not participated in the validation myself (yet), not sure what all we could obtain from past records vs. current records.

Good luck; do not give up because people are nervous. It WILL be ok; and wonderful!

Brenda Hutson

Providing you have a computer system approved for it. We have been doing electronic xm since 1995 (2 different computer systems). We were one of the 33 sites that applied for the variance from FDA before they figured out that it was OK and now you just have to have your validation available to show them. I still have my original corresponence. FDA issued a reviewer's checklist six months after I submitted our request for a variance.

Validation that I had to submit to FDA: crossmatch all blood types against all blood types (8x8) to show that if you try to give an A to a B it won't let you, etc.; prove that if a patient didn't qualify, .i.e, only one type on file, positive antibody screen, previous clinically significant antibody, donor unit without a typing result on file, etc., an EC would not be allowed; all messages displayed when the patient didn't qualify; all file maintenance that had anything to do with the EC. I also had to submit all the pertinent SOPs and the decision tables from the computer showing what blood type was allowed to be given to what blood type. I am probably missing something in this list.

The techs would probably revolt if they had to give it up. That said, when we first started doing it I saw a lot of tubes in the incubators at some techs workstations because they were still doing AHG xm to make themselves feel better about it. That soon stopped.

We revaliate when we bring up a new site (great training tool) and when we upgrade to a new version of software.