Chromatography Forum

We are currently running an HPLC method multiple times for an assay, but the method repeatability is being questioned. Thus, we are trying to look for parameters that we can monitor to ensure that the column that we are using is OK.

I have read several discussions on using plate count for gradient methods, and wanted to know - I know that the equation for isocratic methods is not to be used scientifically for a gradient elution, but can I use it as a qualitative measurement of column performance (i.e. measure it at the beginning of a column's lifetime, and then monitor it over column usage to determine when it can no longer be used) for a specific peak in my reference standard injections? This would be for internal use, not as scientific data. And would this information be comparable between columns/ systems that use the same method?

You are correct that the plate count cannot be measured from a single gradient run; the isocratic formula returns a number that is meaningless. The only meaningful piece of information in there is the peak width.

My recommendation for a system suitability test in this sort of situation would include:
- width of a major peak
- tailing factor of a major peak
- retention time windows for peaks of interest
- resolution between any "critical" pairs of peaks
For pharmaceutical analyses, also include precision for a reasonable number of injections (typically 5); for trace analysis, a baseline noise spec and verification of LOD & LOQ would also be a good idea.

the method repeatability is being questioned. Thus, we are trying to look for parameters that we can monitor to ensure that the column that we are using is OK.

Questioned for what reason?

My experience with gradient methods has been that folks have issues with repeatability when the equilibration time (equilibration volume) before the next injection is not enough.

Your system suitability data should demonstrate necessary repeatability.

We've cGMP-validated quite a few gradient test procedures here, all developed in house too. Typically, we use isocratic assay if possible, but gradient is a wonderful tool, and the column is pretty much cleaned off and ready for the subsequent injection.

I would agree.