Page 1 of 1

testing stability in method validation

Posted: Thu Jul 15, 2010 1:27 pm
by Friederike
Dear all,
in our lab, when we want to determine the stability of an analyte after a period of, say, 24 hours, we do the following:

we measure QC samples at time zero (3 injections/concentration) - store that sample solutions (or prepared samples) at predefined conditions - reinject the samples after 24 hours (again 3 injections/conc.) and quantify using a freshly made calibration curve on that day - compare the mean result of these injections with that from time zero

Someone told me that I have to quantify the injections from time 24h using the original calib curve (from time zero) and that I have to compare the results with the nominal concentration.

What is the most common procedure in regulated labs?

Thank you in advance for your input.

Posted: Thu Jul 15, 2010 2:09 pm
by Blazer
I have always done it as you described, using a fresh standard to quantify the samples at each time point of the solution stability study.

Posted: Thu Jul 15, 2010 3:29 pm
by Rob Burgess
Yep I would follow a similar procedure how you have described.

Posted: Thu Jul 15, 2010 5:26 pm
by Ronaldo
i think it depends in the stability of your analyte in the standards prepared .If the analyte is very stable in the solvent that was used to prepare the stadards then use this calibration for all your samples .If the analyte is not stable as in the standards then yes prepare new calibration .

This is what i do with measuring the solubility i use only one calibration for all my test , yes it will be slightly some variation but the error will be cancel out when compare several samples at that day

Posted: Thu Jul 15, 2010 10:37 pm
by LC_labrat
Using fresh standards to determine solution stability of aged samples is the practice I've seen.