A device for automated direct sampling and quantitation from solid-phase sorbent extraction cards by electrospray tandem mass spectrometry.

Abstract

A new solid-phase extraction (SPE) device in the 96-well format (SPE Card) has been employed for automated off-line sample preparation of low-volume urine samples. On-line automated analyte elution via SPE and direct quantitation by micro ion spray mass spectrometry is reported. This sample preparation device has the format of a microtiter plate and is molded in a plastic frame which houses 96 separate sandwiched 3M Empore sorbents (0.5-mm-thickness, 8-microm particles) covered on both sides by a microfiber support material. Ninety-six discrete SPE zones, each 7 mm in diameter, are imbedded into the sheet in the conventional 9-mm pitch (spacing) of a 96-well microtiter plate. In this study one-quarter of an SPE Card (24 individual zones) was used merely as a convenience. After automated off-line interference elution of applied human urine from 24 samples, a section of SPE Card is mounted vertically on a computer-controlled X, Y, Z positioner in front of a micro ion spray direct sampling tube equipped with a beveled tip. The beveled tip of this needle robotically penetrates each SPE elution zone (sorbent disk) or stationary phase in a serial fashion. The eluted analytes are sequentially transferred directly to a microelectrosprayer to obtain tandem mass spectrometric (MS/MS) analysis. This strategy precludes any HPLC separation and the associated method development. The quantitative determination of Ritalin (methylphenidate) from fortified human urine samples is demonstrated. A trideuterated internal standard of methylphenidate was used to obtain ion current response ratios between the parent drug and the internal standard. Human control urine samples fortified from 6.6 to 3300 ng/mL (normal therapeutic levels have been determined in other studies to be between 50 and 100 ng/mL urine) were analyzed and a linear calibration curve was obtained with a correlation coefficient of 0.9999, where the precision of the quality control (QC) samples ranged from 9.6% at the 24 ng/mL QC level to 1.2% at the 3000 ng/mL QC level, and the accuracy for the four levels of QC samples ranged from 98.1% to 100.3%. The QC samples were prepared at four concentrations which included 24, 240, 1200, and 3000 ng/mL, respectively. The run time per sample in this work was 1.5 min not including the sample preparation time.