Mass spectrometric (MS) methods are today widely used in analytical toxicology, but in comprehensive quantitative drug screening it is laborious to maintain quantitative calibration and use of historic calibration is not feasible. In non-MS gas chromatographic (GC) and liquid chromatographic (LC) methods, historic calibration is widely used. However, GC is not amenable to analysis of current polar drugs and traditional LC-UV methods do not have high enough identification power to meet the requirements of screening. Ultra-performance LC (UPLC) with two consecutive detectors appears to possess high potential for simultaneous multi-component screening and quantification. In this study, a comprehensive quantitative screening method for 170 basic drugs in blood samples was developed and validated using UPLC coupled with photodiode array (PDA) and corona charged aerosol detector (CAD).

Blood samples were extracted with organic solvent (ethyl acetate: butyl acetate, 25:75) in basic conditions. After extraction, the organic solvent was evaporated to dryness and samples were reconstituted with the UPLC mobile phase. Dibenzepine was used as an internal standard. Chromatographic separation was performed at 60°C using a HSS C18 column (150mm × 2,1mm, particle size 1,8μm), and the mobile phase consisted of 0.1% trifluoroacetic acid and methanol. After a three minute isocratic phase, a linear gradient from 5% to 95% methanol in 15 minutes followed. The flow rate was 0.4mL/min. UV spectra were collected in the range of 210–400nm, and wavelength of 230nm was used for quantification. Substance identification was based on the UV spectrum, retention times on both detectors, and the response ratio of CAD and PDA at wavelength 230nm. Calibration was carried out using a single calibration point at the vicinity of the upper limit of therapeutic range of each drug. Individual calibration curves were created for both detectors and the average result was used for quantification.

The retention times were found be very stable, the relative standard deviation of retention times being <0.03% intraday and <0.3% interday. In order to simplify data processing and interpretation of the reports, the analytes were divided into five data processing methods, each containing every fifth analyte based the retention time. This approach provided a maximum of five candidates for each peak, thus simplifying manual interpretation even with more complicated samples. Linearity of one-point calibration was found to be acceptable within the therapeutic and toxic ranges of the drugs of interest, with linear ranges generally being 0.05–5,0mg/L or 0.1–5,0mg/L. Calibration was found to be stable over one month, the bias between old and new calibration being <15%.

The developed method allows the detection and quantification of 170 basic drugs in therapeutic and toxic concentrations in a single run. Due to the stability of calibration and good linearity, historic one-point calibration can be utilized. Adding the CAD detector with universal response after the PDA detector increases the reliability of both identification and quantification. In addition, drugs with poor UV absorption can be detected and quantitated. This method provides a tool for comprehensive quantitative screening for ordinary basic drugs in blood and leaves LC-MS target analysis to be applied to low-dose compounds.