Internal Standards for LC-MS

[dth]

Very grateful to you all for input. On the same subject of IS - but this time LC-MS related, I'm finding results are much better when I analyse without the IS. How can I justifiy this - or does this mean I have to find a different IS that behaves more like my analyte. Sample prep is simply removal of plasma proteins and direct injection of the S/N. All samples, calibrators, controls and test samples are treated the same and IS is added prior to protein precipitation step. Not using an IS for MS work seems to be a total 'no no' in the literature.
Thanks again
Diane
------

[tom jupille]

By "much better", I'll assume you're referring to the repeatability of the numbers. What that (strongly) suggests is that your source(s) of error are different for the analyte(s) and the IS. Possibilities that come to mind include:

• - Integration problems (are any of your peaks small enough to have S/N problems?)
  - Ionization suppression changing during the run (most likely if one of your peaks comes off early and the other comes off significantly later; ideally the IS should elute just after the analyte)
  - Different behavior during the workup, as you suspect.

Other people can probably add to the list.

[Uwe Neue]

Under the described circumstances, differences in ion suppression between the analyte and the internal standard are the most likely cause of the problem. You can verify this by running a set of analysis, where a blank water sample is treated the same way as your plasma sample. If the reproducibility problem goes away, it was due to ion suppression. If it remains (more doubtful), it is related to experimental details (such as adsorption of analyte or standard on vials etc.)

If it is ion suppression, the way around this is improved sample preparation.

[bert]

Diane,

the topic 'Quantitation using LC/MS' (http://www.sepsci.com/chromforum/viewtopic.php?t=952), dated Nov 02 2004 good give you some more information.

Regards Bert

[MG]

In my experience, you are more likely to suffer from differential ion suppression if your ISTD has a different retention time than your analyte. The ideal ISTD is an isotopically labeled version of your analyte (at least three deuterium or 13-C labels, or more). This will coelute with your analyte and also behave most similarly in the preparation steps.