Plasticisers in biological matrices

[steve wileman]

I'm detecting plasticisers (mainly phthalate metabolites and BPA) in biological matrices using SPE - LC-MS/MS. I have been having trouble getting my QC's to pass. ie I can produce pretty nice calibration plots in one batch of urine (for example), but when I spike a different batch I get a different response - particularly at the lower end (1-5ng/ml). I am measuring what is in the matrix to start with (ie the "blank") and deducting this from the total peak area. I use C18 SPE, then either phenyl or C18 HPLC column.
Has anybody else come across this difficulty of getting "blank" biological matrices?
Thanks!
Steve

[Don_Hilton]

These compounds are quite common in the general population http://www.cdc.gov/exposurereport/pdf/FourthReport.pdf It may be worth checking the procedures used by laboratories such as the CDC to see where they get materials to use as "Blank" matrix. A look at the published papers may help - or look for authors and contact them. And the better your analytical method - the harder it is going to be to find "blank" urine.

[Steve Reimer]

And the better your analytical method - the harder it is going to be to find "blank" urine.

Or even "blank" containers to collect and process it in. We have had problems getting fish that is free of mercury for low level mercury analysis.

[steve wileman]

Yes, thanks for that. I recognise what you're saying - I have tried spiking artificial urine which works well, but I dont get a true picture when compared to the real matrix ie. the peaks are much bigger at a fixed concentration. Others have also produced cali-plots in solvent, which is not representative of the matrix at all. I can deal with non-blank blanks provided I measure what that is, and deduct it - the problem seems to be that when I attempt to replicate cali-plots in a different batch of matrix, and even after deducting the blank values for this new matrix, the concentrations work out different - especially at low concentrations - 1-5ng/ml. Having said all that, I have been experimenting with a 1mm HPLC column (as opposed to 2.1mm) and at least for some analytes the situation has improved. This may indicate that it is an integration/sensitivity problem - always tricky when working at these trace levels! Thanks again for your response.

[Alp]

One possibility is that something in some urine batches (but not others) is co-eluting with your analyte and causing matrix effect. Are you retaining your analyte on an HPLC column and if so, is it retaining well outside the void volume?

[steve wileman]

thanks very much for that, I also think that this is possible, but analytes are retained until well after void volume I would say. Things are looking a bit better with the 1mm columns and also the Phenomenex core shell column. Nearly there.. Thanks again.