Hair analysis for delta(9)-THC,delta(9)-THC-COOH,CBN and CBD,by GC/MS-EI. Comparison with GC/MS-NCI for delta(9)-THC-COOH

A sensitive analytical method was developed for quantitative analysis of delta(9)-tetrahydrocannabinol (delta(9)-THC), 11-nor-delta(9)-tetrahydrocannabinol-carboxylic acid (delta(9)-THC-COOH), cannabinol (CBN) and cannabidiol (CBD) in human hair. The identification of delta(9)-THC-COOH in hair would document Cannabis use more effectively than the detection of parent drug (delta(9)-THC) which might have come from environmental exposure. Ketamine was added to hair samples as internal standard for CBN and CBD. Ketoprofen was added to hair samples as internal standard for the other compounds. Samples were hydrolyzed with beta-glucuronidase/arylsulfatase for 2h at 40 degrees C. After cooling, samples were extracted with a liquid-liquid extraction procedure (with chloroform/isopropyl alcohol, after alkalinization, and n-hexane/ethyl acetate, after acidification), which was developed in our laboratory. The extracts were analysed before and after derivatization with pentafluoropropionic anhydride (PFPA) and pentafluoropropanol (PFPOH) using a Hewlett Packard gas chromatographer/mass spectrometer detector, in electron impact mode (GC/MS-EI). Derivatized delta(9)-THC-COOH was also analysed using a Hewlett Packard gas chromatographer/mass spectrometer detector, in negative ion chemical ionization mode (GC/MS-NCI) using methane as the reagent gas. Responses were linear ranging from 0.10 to 5.00 ng/mg hair for delta(9)-THC and CBN, 0.10-10.00 ng/mg hair for CBD, 0.01-5.00 ng/mg for delta(9)-THC-COOH (r(2)>0.99). The intra-assay precisions ranged from <0.01 to 12.40%. Extraction recoveries ranged from 80.9 to 104.0% for delta(9)-THC, 85.9-100.0% for delta(9)-THC-COOH, 76.7-95.8% for CBN and 71.0-94.0% for CBD. The analytical method was applied to 87 human hair samples, obtained from individuals who testified in court of having committed drug related crimes. Quantification of delta(9)-THC-COOH using GC/MS-NCI was found to be more convenient than GC/MS-EI. The latter may give rise to false negatives due to the detection limit.