Changing tubes used in blood bank testing

[Jane]

Ok- here is yet another question about validation. Mine is a little different. We are planning to switch from the BD pink top tube (plastic K2 edta) to the Greiner Bio-One pink top (also plastic but K3 edta) due to significant price differences. I am planning to do a small parallel study as the new tube has FDA approval for the testing we will be performing. My question is what tests should I do for the validation- we use this tube for most everything in the blood bank- ABO/Rh, DAT, Antibody screen and ID, crossmatches, etc. Will just doing tests with similar methadology be adequate- antibody screens and ABO/Rh on all the samples?

Opinions please!

[donellda]

The validation study would have to involve all of the procedures that you would be using the new tubes for. If you run parallel studies with the old tubes and the new tubes I think that this should be sufficient. When we switched from red tops to the plastic pink tops this is what we did.

[David Saikin]

If the tube is FDA approved for blood bank use, why do extra validation? If you have validated EDTA tubes previously, my feelings are make the switch. It will be difficult to validate all your circumstances, notably +DATs. So, will you not use them for that or what, if you haven't validated that aspect of testing? Valdiation is for something that is totally new, not variations on a theme.

[John C. Staley]

David, I couldn't agree with you more. I waited to see if anyone else would make this statement and was glad to see you did. It never ceases to amaze me the lengths we seem to go because we fear not doing enough.

[Lcsmrz]

I'm throwing in a dissenting opinion ...

Validation is necessary when changing any critical material -- I'm assuming specimen tubes would be a critical material -- and includes more than just checking for FDA-approval. There could be significant differences other than the anticoagulant formula and vendor, and the validation process should be designed to uncover them and to verify that your particular process accounts for them. And it needs to be documented ...

Some examples: a review of the new tube's product insert may indicate different expiration dating, post-draw handling or storage requirements, and may require only a completely-full tube be used for testing. It may be FDA-approved for tube testing, but not for gel. It may have a slightly-different color cap, requiring changes in SOP and possibly some nursing education. The company's parallel study for the approval submission may have referenced one substantially different from the one I was using.

Validation doesn't need to be as extensive as a 510(k) submission, but it needs to follow your change control procedure and to document your journey through the change, including approvals.

Too many times, supply decisions are made by bean counters for cost reasons only, and later limitations are found that compromised patient care that should have been uncovered before initiating the change. In addition to verifying the company's insert claims are possible in your lab, validation procedures should be designed to uncover any difference from your existing processes and document your decisions about the differences. I seem to remember a posting somewhere that Greiner tubes underperformed in one person's validation study; that notation would cause me to more extensively parallel test the new tube, then a literature search that uncovered only positive comments.

Just my $0.02 ...

[donellda]

I was just going to post that it is in section 3.3 of the AABB Standards but you already said what I was going to say.

[John C. Staley]

I think we are dealing with different definitons of "validation" here. I fully agree that you should thoroughly review the package insert to see if there are any significant variations from the current SOP you are using. I don't consider that validation, I consider that common sense. What I take exception to is when there is no significant changes in SOP based on package insert information and still expecting people to test in duplicate, triplicate etc in large numbers with no obvious reason other than to say you validated. I validate when I am bringing a new test or procedure on-line that we have never done before and I don't know if we can make it work here or when we are changing from one technique to a completely different one. Example; if someone were to come out with a new elution technique/kit I certainly would perform an adequate amount of validation to confirm that it was as good or better than the current one we are using and that my staff could use it successfully.

[David Saikin]

John - I'm glad you responded. I tried, but could not come up with what sounded good. Nice job.

ds

[Lcsmrz]

You certainly have to define the word "validation" when discussing a topic like this. Some people may read here that validation is not required and freely substitute "equivalent" critical material without giving it another thought, while another thinks changing the color of their test tube racks means conducting a study that qualifies for FDA submission.

Validation is documenting evidence that that the system consistently performs as expected; verification is confirming that the specifications have been met. My comfort level with a change varies with the criticalness of the process or the material and with the time I'm given to get it done. I love the phrase "... known to produce ..."

Someone at the facility is charged with the responsibility to approve new and changed processes after reviewing the submitted packet of evidence. The level of documentation could be one page or take up volumes, depending on the level of evidence required. Whatever it takes to comply with SOP and get the final signature is "necessary." If a reagent is classified as a critical materiral and the SOP says all changes in critical material will be thoroughly evaluated, tested, documented, etc, then that's what's necessary.

My SOP says validation is all activities that document the journey from initiation to implementation. It makes few exceptions and attempts to even document "common sense" that occurs along the way. To us old blood bankers, checking product inserts is second-nature, yet to a new grad given a first project, this step could be overlooked. We continually research and evaluate the available technology, yet I'm sure there are some who feel that the National Enquirer contains all the necessary information for the job.

I like to see organized documentation that the facility gave some thought to a change, arrived at a scientifically-reasonable decision for what they did bsaed on that documentation, and obtained the appropraite signatures. In this example, just saying that it would save money with no further documentation would cause me to delve further into their validation efforts, present and past.

[Art]

Lcsmrz

Well written. Your definitions and commentary are right on.

A validation assures me that the process in my lab functions properly and meets my lab’s acceptance criteria under my conditions. Validation by a manufacturer does not negate my responsibility of ensuring that changes will not affect my process. Validations also enable me to tap the knowledge of others in the facility that may see the process from a different perspective. Risk assessments (clinical/programming/security and regulatory) will dictate the extent and complexity of the validation.

Section 5.0 of the AABB Standards also addresses this issue as Process Control.