Anticoagulants/Rodenticides in Liver/Whole Blood Samples

[akthmps]

I work at an animal diagnostic laboratory and one of the tests I'm developing is for the analysis of anticoagulants in liver/whole blood samples. Here are the anticoagulants of interest:

COUMARINS:
1. Warfarin
2. Brodifacoum
3. Bromodialone

INDANEDIONES:
1. Diphacinone
2. Chlorophacinone
3. Pival

My HPLC system is a Perkin Elmer Series 200, with autosampler, pump, UV-VIS detector, fluorescence detector, and column oven. I am currently using a Partisphere RTF C18 column, 4.6 x 250 mm. I have been setting the column oven at 30 Celcius.

I've had bad luck with this analysis so far. Initially, I tried an isocratic 1.2mL/min analysis with the standards, and my retention times have been highly variable, with some analytes shifting as much as 1 minute between runs. The isocratic mobile phase was composed of: 300mL water + 7.5mL CH3COOH + 32.5mL 2N NaOH + 15mL dibutylamine + 395mL ACN. The coumarin anticoagulants were resolved with fluorescence detection, excitation 285 emission 390. The indanediones were detected on the UV detector, set at 280nm. However, the retention times were somewhat variable. I would equilibrate the column for 30 minutes with mobile phase, run for 40 minutes, then re-equilibrate. I tried a gradient elution next, with variations of my original isocratic mobile phase (I just changed the amount of water/ACN in each mobile phase), but still had variable retention times. My fluorescence data also had a slight upward drift throughout the run with gradient elution.

I recently acquired a procedure from another laboratory for anticoagulant analysis. It's a gradient elution procedure, with different mobile phases for the indanediones and coumarins. I had all the necessary reagents to try this procedure for the coumarins, and here is how the elution works:

BUFFER:
3.85 grams ammonium acetate + 2mL glacial acetic acid + 2mL triethylamine. Bring up to 1000mL volume with high purity water.

Column flow at 30 degrees is 1.0mL/min, and the gradient run is as follows:

TIME (minutes) 0 = 62% methanol, 38% buffer
TIME 1 = 62% methanol, 38% buffer. START RAMP:
TIME 4 = 79% methanol, 21% buffer HOLD
TIME 14 = 79% methanol, 21% buffer START RAMP:
TIME 15 = 62% methanol, 38% buffer

The data was horrible. I completely lost my peak shape, and I'm not even sure if what I saw was any of the 3 coumarin anticoagulants. With isocratic elution, at least I was able to get good peak shapes, and somewhat good repeatability, compared to any of the gradient elution procedures I used.

Can anyone suggest a new mobile phase recipe for me to try? Or what else I can do? I would prefer an isocratic run, but I'm certainly willing to try gradient again. I would like to be able to detect all the anticoagulants in a single run, rather than eluting the indanediones and coumarins separately. And this seemed possible with my original isocratic elution procedure, I just need to make the data a bit more consistent.

Finally, any thoughts on my sample preparation?

Homogenize 5-6 grams of tissue sample (or 5-mL whole blood) in 20mL acetonitrile. Sonicate for 10 minutes. Centrifuge, remove supernatant. Condition an Alumina-N cartridge (5g) with 8mL methanol, 8mL water, then 8mL acetonitrile. Pass supernatant through column, collect this supernatant, and pass 10mL acetonitrile through, also collecting. Evaporate the combined supernatant/acetonitrile under nitrogen and reconstitute in 1mL acetonitrile.

Any help with this analysis would be greatly appreciated!

Thanks,
Aaron

[Uwe Neue]

A brief estimate of the molar amounts of acid and base in your first rpocedure indicates that the mobile phase has an alkaline pH. If this is correct, you have destroyed your column, and the lack of performance with the gradient method indicates this.

I suggest to buy a new column, specifically buy one based on a high purity silica.

For suitable sample preparation techniques, I recommend to look at the Waters web site for the generic instructions how to develop methods for plasma samples using the Oasis SPE sorbents.

[Mark Tracy]

Many years ago I too worked at a veterinary diagnostic toxicology lab, and tried to do the same thing. I finally decided that combining the coumarins and indanediones in one run was too hard. Perhaps modern column technology can make it possible. Good luck!