Computer Crossmatch: what is needed to implement?

[Barbarakym]

To go from a regular Crossmatch to computer Crossmatch what validations and training are required?

[David Saikin]

You BBIS must be (501k) approved for the e-xm. (truth tables for pt/donor types, written policy - need 2 valid ABORh types on pt - 1 may be historical). Validate that you cannot do this with only one type and incompatible ABO and that it does work when your acceptance criteria is met (2 pt types, ABO compatible).

There must be more.

[Barbarakym]

We are getting new system to us, Cerner. Supposedly this will be validated company wide. Some hospitals are up live already Can we assume the validation by company is good enough?

How about from our BB assuming computer as built is ok, what needs to be done local to switch from tube to EC? A procedure of how. I get that. Training also. But do we need to do anything else?

Switching from tube abs for example to gel we had to validate and each tech had to demonstrate. What do we need to do to validate it works as a whole ? Or cuz the company says it works we can just start handing out blood?

[tricore]

See library on this site, FDA Guidance for Electronic XM

[tbostock]

[ATTACH]622[/ATTACH]You need to do your own validation of electronic crossmatch. Your inspectors will want to see it.

Edited March 22, 2012 by tbostock

[mcgouc]

We had a company wide validation when we went on Cerner ME, but you still have to validate the computer crossmatch works in your facility using your policies. There were some options in the electronic crossmatch program that needed changing before we went live that passed the company validation. There is one option where you can't add on units using the electronic crossmatch after issuing a unit without an override. That had to be changed. We use a Blood Bank bracelet and they didn't have a spot in the electronic crossmatch to enter the bracelet number so it would print on the tag. We resolved those problems and then validated in-house. The techs were given an in-service and had to pass a written quiz. We waited a couple of months just to get everyone used to the new computer before changing to the electronic crossmatch. Some of the facilities went live without validating in-house and were calling me to see why they couldn't use the electronic crossmatch after issuing a unit.

[Barbarakym]

Thank you everyone. Another related question. Once we go to computer (electronic) crossmatch are we required to be inspected by FDA?

[tbostock]

Thank you everyone. Another related question. Once we go to computer (electronic) crossmatch are we required to be inspected by FDA?

No, just keep in mind, just because you are not collecting or modifying blood products, which would require inspections with the FDA, does not mean they can pop in anytime to inspect any Blood Bank. Just because they can.

[Auntie-D]

You need it built in as well that if a result is changed on the analyser (ie ignoring crud) then it won't ei. Also both groups must be on the analyser so you need a way of distinguishing manual grouping from electronic.

[Mabel Adams]

... both groups must be on the analyser so you need a way of distinguishing manual grouping from electronic.

Do you mean that in the UK, electronic xm is not allowed if blood typing is done manually? That is not a requirement in the US.

[Steven Jeff]

Do you mean that in the UK, electronic xm is not allowed if blood typing is done manually? That is not a requirement in the US.

Mabel, my understanding of the UK guidelines is that electronic issue (EI) is acceptable in the UK on manual typing, however it is strongly discouraged. It is unlikely that any laboratory in the UK would undertake EI without at least one group and screen being processed through an automated analyser with the results being transferred directly to the blood bank computer system following authorisation on the analyser.

Steve

:)

[Mabel Adams]

That seems like a good idea. More consistent sample ID with a barcode than with a human. Electronic xm in the US pre-dates the ready availability of automated analyzers by quite a few years so I suppose that is why the regs don't require it.

[Auntie-D]

Yeah it is strongly discouraged to the point they asked about it in the MHRA inspection... You can see why - a closed system is best. Barcode from wrist to wrist is what they hope for.

[Steven Jeff]

Yeah it is strongly discouraged to the point they asked about it in the MHRA inspection... You can see why - a closed system is best. Barcode from wrist to wrist is what they hope for.

I entirely agree with the point you have made. We can discuss the merits (or otherwise) of testing two samples or one sample twice etc. The focus must be on the whole process from sample collection to transfusion (wrist to wrist) including the use of printed barcoded PID labels on the samples provided they are generated at the bedside. The less human/manual manipulation the better. We can dream!!!!!

Steve

:)

Edited March 27, 2012 by Steven Jeff
format

[Cxm]

Barbarakym,

Although venders must validate their systems prior to gaining 510k clearance, additional testing must be performed to include site specific build and design. Cerner and other companies offer third party blood bank validation services that meet the FDA requirements. Most of this work is done remotely and the only instance that I’m aware of where additional testing must be performed on site is for Computer Crossmatch. This is due to AABB standard 5.15.2.1 (current edition).