interference peak

[elizabeth s]

Interference peak found in analyte retention time on actual sample run what modification I can do on the method to seperate this peak.
actural sample matrix is Human Plasma.
No interference peak in actual standards and QC's

Thanks
Elizabeth

[Uwe Neue]

1. Improve the sample preparation method!
2. Change the chromatography!

[elizabeth s]

Do you mean changing mobile phase composition to achieve separation?

[tom jupille]

At the risk of putting words into Uwe's post, there really are only two things you can do:

1. Change the sample workup procedure to eliminate the interference. This is most easily accomplished if the interference is quite different chemically from the analyte (a protein, for example).

2. Adjust the chromatography to improve separation. Assuming you are doing reversed-phase, this usually means changing the selectivity, which can be accomplished in one (or more) of six ways.
- change the mobile phase strength (or gradient steepness)
- change the temperature
- change the organic solvent
- change the column chemistry
- change the pH
- put in additives

I've listed these roughly in order of ease of use. Without knowing a lot more about your sample and separation chemistry, it's hard to be more specific.

As far as the internal standard question, ideally, the internal standard should be as similar as possible to the analyte in terms of chromatographic behavior.

Internal standards add a certain amount of error to the measurement (you have to measure two peaks). If the precision is good to begin with, this additional error makes things worse. That is the situation in most LC analyses (certainly using UV detection).

When there are sources of error that affect the IS and the analyte in the same way, however, those errors cancel when the area ratios are calculated. That's the situation in some analyses where sample workup or injection errors are involved. It's also the situation in MS detection where the ionization efficiency can vary considerably from sample to sample due to matrix effects.

Bottom line: IS is used only when it improves quantitative precision.

[Mark Tracy]

The main reason why isotopically labelled IS are not used more for LC/MS is cost. Synthesizing, assaying and certifiying isotopically labelled substances is hard work. Commercial vendors won't do it unless there are enough paying customers.

[tom jupille]

Mark, on seeing your post, I realize I misinterpreted the original question!

[Uwe Neue]

Sorry for having been so short on my first reply...

I think an improvement in the sample preparation method is the most promising approach. We commonly use solid-phase extraction for effectively removing interferences from plasma samples. The details of the method depend somewhat on your analyte, but in principle they are very straightforward. The preferred method is to use mixed-mode ion-exchangers, if your analyte can be put into an ionic form. For a cationic compound we load under acidic conditions onto a mixed-mode strong cation-exchanger, wash with water to remove polar interferences, then wash out acidic and neutral hydrophobic interferences with (acidified) methanol, and finally elute the compound with ammonia in methanol. For acidic compounds, an anion exchanger is used and the pH in the wash protocol is inverted. For large throughput, this protocol is executed in 96-well plates.

Usually, this protocol gives a drastic improvement over initial results. If there is still a problem, the protocol can be fine-tuned for the analyte at hand.