Sensitivity check in syst suit

[mojo]

What do you guys do for the impurity syst suit with regards the sensitivity? Do you do:

a) LOD only

b) LOQ only

c) both LOD & LOQ

d) neither

We use the LOQ for imp's, but LOD for the cleandown analysis. What is the justification for your approach?

[Rob Burgess]

We run a semi-pseudo LoQ at the reporting limit of our impurities for pharma compounds. This is normally 0.05% concentration of our typical sample solution.

However, this isn't a true LoQ in the strictest sense of the definition. All it means is that we have a S/N greater 10 so it gives confidence that we can adequately quantify our impurities at this level. However the true LoQ (S/N = 10) value is usually much lower.


Plus it will also depend on how you measure your S/N. S/N calculated as root mean square (RMS noise) >> than S/N calculated as EP noise. I just wish the regulators would get together and make sense of all these differences.