Driving Under the Influence of Cannabis: Pitfalls, Validation, and Quality Control of a UPLC-MS/MS Method for the Quantification of Tetrahydrocannabinol in Oral Fluid Collected With StatSure, Quantisal, or Certus Collector

Background: Driving under the influence of drugs (DUID) has a large impact on the worldwide mortality risk. Therefore, DUID legislations based on impairment or analytical limits are adopted. Drug detection in oral fluid is of interest due to the ease of sampling during roadside controls. The prevalence of Delta(9)-tetrahydrocannabinol (THC) in seriously injured drivers ranges from 0.5% to 7.6% in Europe. For these reasons, the quantification of THC in oral fluid collected with 3 alternative on-site collectors is presented and discussed in this publication. Methods: An ultra-performance liquid chromatography-mass spectrometric quantification method for THC in oral fluid samples collected with the StatSure (Diagnostic Systems), Quantisal (Immunalysis), and Certus (Concateno) devices was validated according to the international guidelines. Small sample volumes of 100-200 mu L were extracted using hexane. Special attention was paid to factors such as matrix effects, THC adsorption onto the collector, and stability in the collection fluid. Results: A relatively high-throughput analysis was developed and validated according to ISO 17025 requirements. Although the effects of the matrix on the quantification could be minimized using a deuterated internal standard, and stability was acceptable according the validation data, adsorption of THC onto the collectors was a problem. For the StatSure device, THC was totally recovered from the collector pad after storage for 24 hours at room temperature or 7 days at 4 degrees C. A loss of 15%-25% was observed for the Quantisal collector, whereas the recovery from the Certus device was irreproducible (relative standard deviation, 44%-85%) and low (29%-80%). During the roadside setting, a practical problem arose: small volumes of oral fluid (eg, 300 mu L) were collected. However, THC was easily detected and concentrations ranged from 8 to 922 ng/mL in neat oral fluid. Conclusion: A relatively high-throughput analysis (40 samples in 4 hours) adapted for routine DUID analysis was developed and validated for THC quantification in oral fluid samples collected from drivers under the influence of cannabis.