Distinguishing Contamination from Carryover When Using LC-MS/MS to Detect Drugs of Abuse Kristine van Natta (Presenter) Thermo Fisher Scientific

Authors: Joe Di Bussolo, Kristine Van Natta, Stephanie Samra, Yu Zhou and Xiaolei Xie
Thermo Fisher Scientific

1. Problem

An established LC-MS/MS method to detect drugs of abuse showed what appeared to be excessive carryover for five compounds: Amphetamine, BEG (Cocaine metabolite), Cocaine, EDDP (Methadone metabolite) and Methadone.

2. Method Information

• Calibrators made by spiking negative-urine and diluting 1:10 with water containing internal standards. Double blanks made of pure water. Test mixture of Amphetamine, EDDP, BEG, Cocaine and Methadone, 5 ng/mL each in water.

• 10 uL injections into the LX injector of a Thermo Scientific™ Transcend™ TLX-1 system.

• Solvent A: 10 mM ammonium formate in water; Solvent B: 10 mM ammonium formate in methanol.

• Thermo Scientific™ Accucore™ Biphenyl, 2.6 um, 50 x 2.1 mm HPLC column.

• Original method pumped at 0.5 mL/min a 4-minute gradient from 100% A to 100% B in three steps.

• Test method pumped at 0.5 mL/min a 1-minute linear gradient from 100% A to 100% B.

• In both methods the column was washed by 100% B for 1 minute and equilibrated with 100% A for 1 minute. Additional equilibration by 100% A at 0.5 mL/min occurred between injections.

• Thermo Scientific™ TSQ Altis™ mass spectrometer with heated electrospray ion source detected analytes using specific reaction monitoring (SRM) transitions from precursor to product ions.

3. Troubleshooting Steps

1. Ran batches to make 10 µL injections of a pre-blank, test mixture, then three post-blanks.

2. Pre-blanks and last two post-blanks showed evidence of contamination.

3. Allowing the column to equilibrate longer (6 minutes versus 2-3 minutes) before running a batch resulted in proportionally greater peak areas for the analytes tested in the pre-blank.

4. Replacing Solvent A bottle with a fresh one resulted in much lower peak areas in the pre-blanks.

5. Peak areas in the first post-blank were obviously due to true carryover as the following post-blanks showed progressively smaller peaks.

6. Improvements to the autosampler method reduced true carryover to acceptable levels.

4. Outcome

Contamination was quickly distinguished from carryover by injecting a pre-blank followed by a test mixture then three post-blanks. In this case, the root cause – contaminated Solvent A – was quickly identified and remedied.

Problem

Method Information

Troubleshooting Steps

Outcome