Use of Check Standards

[Legarm]

I am trying to find out what the industry standard is regarding check standards. More specifically, what calculation is performed and what are the limits that are set.

[ravenwork]

Greetings,

I think your question is too open-ended.

What industry?

I am in the environmental testing biz, and our check standards are highly regulated, method specific, and show a wide range of acceptance criteria depending upon many factors.

For HPLC methods, quantifying at the ppb level, drinking water testing, we typically have to run checks every 8 hours, with deviation of +/- 20% vs. the initial calibration usually considered acceptable.

Evan Cooper

[Legarm]

In the Pharmaceutical Industry, specifically the QC lab environment.

[MG]

The last GLP project I worked on, we used 15% error at several levels, except 20% error was allowed at the LLOQ. We used this for guidance:

http://www.fda.gov/cder/guidance/4252fnl.pdf

[tom jupille]

And if it's cGMP, they will look for 1%.

The general rule can be summarized as: for the results to be valid, you have to demonstrate that you meet system suitability requirements. These will be specific to the particular method.

That said, the most conservative approach is to run a full system suitability before both and after running your samples, so that you can demonstrate that you were (presumably) OK in between. Intermediate check samples would be run in between as appropriate to avoid the possibility of failing system suitability at the end, thereby invalidating the whole set.

If you like to live on the edge, you could save time by running a check sample ("mini system suitability) at the beginnin to confirm retention, resolution, plates, etc. and then do the full system suitability (including repeatability) only at the end.

[Alex Buske]

Tom,
I disagree,
System suitability should be testet before analysis. Usually we do not have problems with resolution and plates. it just takes a number of injections until the system runs stable. This is especially true for for short isocratic methods.
Check samples are often run at the end. They are nice, yet i have never used (or needed) them.

Alex

[tom jupille]

Alex: I was making a joke with the last paragraph ("If you like to live on the edge . . ."). I certainly do not recommend that approach. Nonetheless, it is permissible.

[Sunjay]

In my previous company, we have mentioned that sys suit once established will be applicable for next 24 hrs and after this 3 std injection shall be given. If sample ends before 24 hrs, one std injection will be given and rsd checked wrt initial injections.

In my present company, we r injecting a conrol std prep after 12 injection and after end of sequence. We r checking the RSD of control std with initial sys suit injections. It should not be more than 2.0%.

I have seen different practices in diff companies. USP has written one paragraph in the chromatography chpter about this. There are no specific criteria about it as per my knowledge.

Thanks

[rc_12321]

It is well known that bracketing system is to ensure that our system is doing well during the runs. Hence we need to use these checks not only at the beginning of the run but also in between and at the end of analysis.
In my lab the practice we follow is , if we use both standard and other systemsuitability parameters such as resolution ,plate count etc at the beginning of the run as systemsuitability then we have to use the check standard after every 3 samples and other parameters need to be checked at the end of analysis.

[Legarm]

Thank you all for your responses.