Forensic Toxicol (2012) 30:59–65 DOI 10.1007/s11419-011-0128-z
SHORT COMMUNICATION
A simple quantitation method for benzoylecgonine from oral fluid, blood, and urine samples used for determining 22 illicit and licit drugs by GC–MS with liquid–liquid extraction ´ rok La´szlo´ Instito´ris • Vilmos Angyal • Zso´fia A ´ • Eva Kereszty Tibor Varga
•
Received: 22 September 2011 / Accepted: 7 December 2011 / Published online: 14 January 2012 Ó Japanese Association of Forensic Toxicology and Springer 2012
Abstract The analytical system used for determination of 22 illicit and licit drugs by gas chromatography–mass spectrometry following liquid–liquid extraction has been extended with an additional extraction step with n-butyl acetate:CH2Cl2 = 1:8 to determine benzoylecgonine from the same oral fluid, blood, or urine samples. The extraction recovery of benzoylecgonine was 39.7% for oral fluid, 14.9% for blood, and 26.7% for urine; the cutoffs were 25, 8.0, and 8.0 ng/ml, respectively. The method was fully validated and proved suitable in further proficiency tests and for analysis of 2738 oral fluid, 197 blood, and 1298 urine samples. Keywords Benzoylecgonine Liquid–liquid extraction Oral fluid Blood Urine GC–MS
Introduction In forensic cases, cocaine consumption is mainly proven by the presence of the parent compound and/or its main metabolites benzoylecgonine (BZE) and ecgonine methyl ester (EME) in different biological fluids (mainly in blood, urine, or oral fluid). For determination of a number of illicit and licit drugs from whole blood, urine, and oral fluid a relatively simple and fast liquid–liquid extraction (LLE) system followed by gas chromatography–mass spectrometry ´ rok E´. Kereszty T. Varga L. Instito´ris (&) Z. A Department of Forensic Medicine, University of Szeged, Kossuth L. sgt. 40, P.O. Box 427, Szeged 6724, Hungary e-mail:
[email protected] V. Angyal TEVA Pharmaceutical Works, Ta´ncsics Miha´ly u´t 82, Go¨do¨ll}o 2100, Hungary
(GC–MS) was developed at the National Institute for Health and Welfare (NIHW, Helsinki, Finland) and was adapted in our laboratory. This system has two different sample processing procedures: one for amphetamine-like substances [1] and another for ‘‘other drugs’’ (illicit drugs, involving cocaine, benzodiazepines, zopiclone, zolpidem, etc.) [2, 3]. In this system, amphetamines are extracted from a 0.2-ml sample with 0.5 ml of toluene, while the other drugs (including cocaine) are extracted from another 1-ml sample with 5 ml of n-butyl acetate (BuOAc) at pH 9.2. Under these conditions, however, BZE is present in the highly watersoluble zwitterion form and cannot be extracted with nonpolar solvents like BuOAc. Thus, a separate method and sample preparation was used for its determination at the NIHW. According to the literature, different sample processing and GC–MS methods have been developed for determination of BZE in biological samples [4–11]. Because BZE has a longer detection window and is present in higher concentration in the oral fluid, blood, or urine than EME [4, 12], its determination is widely used to prove cocaine consumption. It was, for example, the main reason why BZE was selected among the two main metabolites as one of the ‘‘compulsory to measure’’ substances in the ‘‘driving under the influence of drugs, alcohol and medicines (DRUID) EU-6 project’’ in blood and oral fluid. Given that methylene chloride was successfully used to extract BZE at alkaline pH [6], this study aimed to establish an additional step to extract BZE from the same sample previously used for extraction of other drugs, thereby decreasing the required sample quantity, and to make the whole procedure simpler. Using low sample volume for analysis is especially important in forensic cases when the quantity of blood, urine, or oral fluid is limited. The present procedure generally requires 2.4 ml of
123
60
Forensic Toxicol (2012) 30:59–65
blood, urine, or oral fluid (the minimum quantity can be reduced to 1.2 ml if necessary) for duplicate measurement.
Materials and methods All materials were of analytical or HPLC grade: Na2HPO4, H3PO4, and CH2Cl2 were obtained from Merck (Darmstadt, Germany); acetonitrile (ACN) and BuOAc from Scharlau (Sentmenat, Spain); N-methyl-N-(trimethylsilyl)trifluoroacetamide (MSTFA) from Sigma-Aldrich (St. Louis, USA); BZE of GC–MS purity and its deuterated analog (BZE-D3) of 1 mg/ml from Promochem (Molsheim, France). Urine and oral fluid samples were collected from volunteers. A 1-ml sample of oral fluid was mixed with 1 ml of StatSure buffer (Medford, USA) to reproduce the conditions of oral fluid collection with the StatSure device [13]. The StatSure device consists of an absorptive cellulose pad with a volume adequacy indicator and a plastic tube containing a buffer solution. The window in the stem of the collection pad turns blue when 1 ml of oral fluid is collected. The buffer contains inorganic compounds and detergent but its composition was not given by the manufacturer. Blood was taken from healthy volunteers in 5-ml Vacutainer tubes containing 0.5 ml of 0.105 M sodium citrate (Becton & Dickinson, Franklin Lakes, USA). Standard stock solutions (eight concentrations) for calibration curves and validation were prepared in the way that 25-ll spiking volume contained 10, 25, 50, 125, 250, 500, 750 and 1000 ng BZE in ACN. The internal standard (ISTD) solution contained 50 ng BZE-D3 in a 25-ll spiking volume dissolved in ACN. Sample preparation A 1-ml aliquot of blood, urine, or oral fluid was spiked with 25 ll of standard and 25 ll of ISTD, mixed with 1.0 ml of 0.5 M Na2HPO4 buffer (pH 9) and 5 ml of BuOAc in a 12-ml capped centrifuge tube (Brand, Wertheim, Germany). The mixture was extracted for 30 s with a multipulse vortexer (Glas-Col, Terre Haute, USA), centrifuged (3000 rpm, 5 min), and 4.5 ml of the organic phase was discarded. After the method was applied to the LLE system, all other compounds, including cocaine, were determined Table 1 Retention times and SIM ions of BZE-TMS in the three matrices investigated
Linearity and intraday precision Five calibration curves prepared with the eight calibration standards were made and measured within 1 day. The linearity was evaluated by the R2 value of the average calibration curve calculated by a least-squares regression model weighed according to reciprocal concentration of
Matrix
Rt (min)
BZE-D3
Oral fluid
3.567
BZE BZE-D3
Blood
BZE-D3
123
Validation procedure
Substance
BZE Rt, retention time; T1, target ion; Q1 and Q2, qualifier ions
from this 4.5 ml of BuOAc phase. The lower phase (which still contained 0.5 ml BuOAc) was re-extracted with 4 ml of CH2Cl2. After centrifugation (3000 rpm, 5 min), 3.5 ml of the CH2Cl2 phase was transferred into a clean tube and evaporated at room temperature under a flow of compressed air. The residue was redissolved in 75 ll of ACN by vortex, and 60 ll was transferred into a GC vial supplied with a 200-ll insert (ViaLab, Munich, Germany). Then 30 ll of MSTFA was added, the vial was capped and heated for 30 min at 80°C in a multiblock heater (Barnstead, Dubuque, USA). The sample was analyzed by GC–MS in electron ionization (EI) mode. The analyses were performed with an apparatus consisting of a 6890 N Network GC system, a 5975B inert XL MSD equipped with turbo pump, interchangeable EI and CI ion source, a 7683B autosampler (Agilent, CA, USA), and an Agilent MSD Chemstation data system (G 1701DA D.03.00.611). The samples were measured in EI mode using a DB-5MS column (15 m 9 0.25 mm 9 0.25 lm, Agilent J & W, CA, USA), and a double gooseneck liner (Agilent 5181-3315). The temperatures of the inlet, transfer line, ion source, and quadrupole were 280, 300, 230, and 150°C, respectively. The initial temperature of the oven was adjusted to 160°C with a 0.5-min hold time and was increased by 50°C/min to 220°C, 10°C/min to 236°C, 50°C/ min to 330°C with a final holding time of 1 min. The run time was 6.18 min. A 2-ll aliquot of the derivatized sample was injected in pulsed splitless mode (injection pulse was 158.6 kPa until 0.2 min) with a constant 1.3 ml/min He flow rate. MS spectra, retention times, and selected ion monitoring (SIM) ions of the trimethylsilyl (TMS) derivatives were determined in full scan mode (m/z 50–550); MS detection was performed in SIM mode. The selected SIM ions and retention times are presented in Table 1.
BZE
Urine
T1 (m/z)
Q1 (m/z)
Q2 (m/z)
85
243
364
3.571 3.566
240 85
82 243
361 364
3.572
82
240
361
3.565
85
243
364
3.570
82
240
361
Forensic Toxicol (2012) 30:59–65
BZE (criteria: R2 [ 0.99). Intraday precision was characterized by accuracy (bias%) and repeatability (RSD%) of the values recalculated by the average curve and related to the nominal concentrations. The limit of acceptance was 20% for repeatability and 15% for accuracy [14]. Interday precision Triplicate samples with the highest, middle, and lowest concentration standards were prepared and measured on 5 separate days. The concentrations of the samples were calculated against their daily calibration curves. Interday precision was characterized by accuracy (bias%) and repeatability (RSD%) with acceptance limits of 20 and 15%, respectively [14]. Selectivity The peak areas (AUC) of the SIM ions of 10 or 25 ng BZE were compared with those of the disturbing or overlapping matrix peaks of blank samples from six different volunteers. The limit of acceptability was set to have at least a three times difference between the target and the average matrix response, and was set to have at least two times difference for the qualifier ions. Extraction recovery Six samples, containing matrix, were spiked with 125 ng of BZE and processed as described. Six other samples containing only 0.50 ml of BuOAc and 4.0 ml of CH2Cl2 were also spiked with 125 ng of BZE. A 3.5-ml volume of these samples was evaporated, the residue was redissolved in 75 ll of ACN, and 60 ll was transferred into a GC vial. Thirty microliters of MSTFA was added and the sample derivatized and analyzed as described (‘‘total samples’’). Recovery was calculated as the percentage of ‘‘total samples’’ according to the target responses (AUC). The stability of samples spiked with 125 ng BZE and ISTD was tested for all three matrices. Eight samples of each matrix type were processed and MSTFA was added. The samples were combined, then divided into seven vials and injected at 4-h intervals. The measured and calculated concentrations were plotted against time to see if they decreased during testing (24 h). Regression analysis was used to evaluate the change in concentrations and the limit for instability was P \ 0.05.
Results Extracted ion chromatograms containing both target (T) and qualifier ions (Q1, Q2) obtained from blood (A),
61
oral fluid (B), and urine (C) samples spiked with 25 ng of BZE are presented in Fig. 1. Retention times and the ions selected for SIM measurement of BZE-TMS from the SCAN spectra for oral fluid, blood, and urine are presented in Table 1. The ion at m/z 240 was chosen as the target ion for oral fluid, because no disturbing matrix effect was found during the preliminary experiments. Intraday precision (Table 2) was characterized by accuracy (bias%) and repeatability (RSD%). The correlation coefficients of the average calibration curves were 0.996 for oral fluid, 1.000 for blood, and 0.999 for urine. The linearity ranges were 25.0–750 ng/ml for oral fluid and 10.0–1000 ng/ml for blood and urine. Based on the linearity ranges, concentrations of 25, 125, and 750 ng/ml were chosen for interday precision study for oral fluid, while concentrations of 10, 125, and 1000 ng/ml were chosen for blood and urine (Table 3). Selectivity was characterized by the ratio of peak areas obtained for the spiked blank samples at the lowest doses to the peak area of the overlapping matrix of the blank extraction. Peak area ratios were evaluated for both target and qualifier ions (Table 4) and for the three different matrices respectively. Except for the qualifier ion at m/z 240 in urine, the area ratios were higher than 3 for target and qualifier ions, thus fulfilling the selectivity validation criteria for all three matrices. The lowest dose of oral fluid used for the selectivity study was also regarded as the cutoff. To determine the cutoff values for blood and urine, six samples containing 0 (matrix only), 2, 4, 8, and 10 ng/ml of BZE were prepared and measured. The lowest concentration that fulfilled the criteria that average AUCBZE [ AUCMATRIX ? 3SD proved to be 8.0 ng/ml both for blood and urine. The cutoff for oral fluid was not investigated (see later). Significant difference between the calculated and measured concentrations in the stability studies were not found during the 24 h of investigation (P [ 0.05 for all the three matrices). Extraction recovery was calculated as the percentage of the ‘‘total samples’’ by comparison of AUCs without ISTD correction. The recovery was 39.7% for oral fluid, 14.9% for blood, and 26.7% for urine.
Discussion Due to their hydrophilic and polar character, polar solvent mixtures are often used for LLE of BZE from different biological samples [6, 7, 9, 15–17], mainly at alkaline pH. Gerlits [6] reported the extraction of BZE by CH2Cl2 from urine at pH 12.1 following a preliminary washing with an organic solvent mixture to decrease the matrix effect. The
123
62
Forensic Toxicol (2012) 30:59–65
Fig. 1 Extract ion chromatogram of a typical blood (a), oral fluid (b), and urine (c) sample spiked with 25 ng/ml BZE. T Target ion; Q1 and Q2 qualifier ions
123
Forensic Toxicol (2012) 30:59–65 Table 2 Intraday precision: accuracy (bias%) and repeatability (RSD%) for BZE in oral fluid, blood, and urine
63
Concentration (ng/ml)
Oral fluid
Blood
Bias (%)
RSD (%)
4.69
2.47
0.34
2.05
750
-4.60
12.7
1.66
2.57
0.10
0.76
500
8.73
12.4
3.10
1.09
-0.55
0.75
250
-1.29
-0.55
2.28
-0.02
2.53
125
2.52
1000
6.75
RSD (%)
Bias (%)
RSD (%)
-0.66
15.4
0.96
2.53
-0.60
7.48
16.9
8.63
4.03
-0.24
1.87
25
-4.94
18.4
-0.83
3.24
1.18
2.44
1.12
7.74
2.48
3.47
Concentration (ng/ml)
Oral fluid Bias (%)
Blood RSD (%)
1000 750
3.18
2.25
125
5.99
5.14
-1.56
4.91
25 10
Table 4 Selectivity: lowest dose/matrix response ratios for BZE-TMS and BZE-D3-TMS
Bias (%)
50 10
Table 3 Interday precision: accuracy (bias%) and repeatability (RSD%) for BZE in oral fluid, blood, and urine
Urine
Urine
Bias (%)
RSD (%)
Bias (%)
RSD (%)
0.03
0.77
2.14
3.42
-7.84
8.12
0.25
2.68
5.97
5.59
0.93
6.68
Compound
Matrix
SIM ions
BZE 25 ng/ml
Oral fluid
240
18,409
–
–
82
23,280
1,887
12
361
3,837
–
–
BZE-D3 50 ng/ml
BZE 10 ng/ml
Blood
BZE-D3 50 ng/ml
BZE 10 ng/ml
Urine
BZE-D3 50 ng/ml
cutoff was not stated (the lowest applied BZE concentration was 75 ng/ml), only that the method fulfilled the National Institute of Drug Abuse (NIDH) criteria for BZE (150 ng/ml). In our study, the cutoff values for blood and urine were 8.0 ng/ml. The same investigation for oral fluid
Response
Mean matrix response
Ratio
85
76,789
23,897
3
243
26,085
2,298
11
364
12,240
2,445
5
82
20,847
3,094
7
240
7,102
182
39
361
6,909
1,852
4
85
113,911
7,223
16
243 364
11,068 19,418
1,271 58
9 335
82
61,313
2,225
28
240
58,799
27,963
2
361
13,813
4,016
3
85
401,863
60,422
7
243
182,850
36,351
5
364
70,941
4,734
15
was not performed because the method was applied only during the DRUID EU-6 project with a recommended 25.0 ng/ml cutoff. Gerlits [6] found a higher recovery of BZE for urine (68% vs the 26.7% in our study) but this ‘‘overall recovery’’ was
123
64
calculated by correcting the BZE response with the internal standard and not by directly comparing the AUCs. The relatively low recovery in our method is not a problem because it is proved to be sufficient for qualitative and quantitative determination of BZE in all three matrices (see later). The sample processing of BZE fits well to the LLE protocol used in our laboratory. For extraction of other illicit (D9-tetrahydrocannabinol, 11-hydroxy-D9-tetrahydrocannabinol, 11-nor-9-carboxy-D9-tetrahydrocannabinol, 6-acetylmorphine, morphine, codeine, methadone, cocaine, and ketamine) and licit drugs (tramadol, zopiclone, zolpidem, clonazepam, 7-amino-clonazepam, diazepam, nordiazepam, oxazepam, temazepam, alprazolam, temazepam, midazolam, and nitrazepam), 1 ml (at least 0.5 ml) of sample is necessary. The sample is mixed with 1 ml of 0.5 M phosphate buffer (pH 9.2), spiked with 25 ll of ISTD mixture (which also contains BZE-D3), and extracted with 5 ml of BuOAC. A 4.5-ml volume of the upper phase is used for the determination of all other drugs, except amphetamines, after derivatization with MSTFA or MTBSTFA by GC–MS in EI or NCI mode. In our additional extraction step, BZE is extracted from the lower phase, which still contains 0.5 ml of BuOAc, with 4 ml of CH2Cl2. A 3.5-ml volume of the organic phase is evaporated, redissolved in ACN, MSTFA is added, and analyzed as described in the ‘‘Materials and methods’’ section. For GC–MS analysis, the same system configuration and column are applied as for other drugs measured in EI mode, except that oven heating parameters and the collected SIM ions are different. In preliminary experiments, BZE was coextracted with the other drugs using different ratios of BuOAc/CH2Cl2 mixtures. The presence of CH2Cl2, however, resulted in an increased matrix effect, and, as a consequence, a decreased sensitivity in detection of the other drugs. Thus, a two-step extraction system was necessary. To extract BZE from the water phase following BuOAc extraction, different BuOAc/ CH2Cl2 ratios were tested, but the recovery did not change or was worsened. Increasing the duration of extraction resulted in a decreased sensitivity, due to the higher matrix background, without significant improvement of recovery. During sample preparation, chemical hydrolysis of cocaine to BZE can occur [18], which leads to overestimation of the BZE amount formed by enzymatic hydrolysis. The rate of this decomposition was investigated in urine at alkaline pH [6] and was found to be negligible (about 0.5%). External quality control and authentic samples As a participant in the DRUID EU-6 project, oral fluid samples of 2738 randomly stopped car drivers and blood samples of 102 drivers that had died in traffic accidents
123
Forensic Toxicol (2012) 30:59–65
were also measured for BZE by this method. During the project, the laboratory participated in four proficiency tests for oral fluid (RTI International, Research Triangle, NC, USA) and blood (ARVECON, Walldorf, Germany). There were no false positive or negative results and the values measured were within the accepted concentration ranges in all cases. Two successful proficiency tests were also performed for urine (QualiCont, Szeged, Hungary). Since 2008, 197 blood and 1298 urine samples from autopsy and criminal cases were investigated for the mentioned licit and illicit drugs and BZE by this method. One blood (61.6 ng/ml) and 32 urine (33.3–18760 ng/ml) samples were positive for BZE. During the DRUID project, BZE was not found in the blood samples of 102 deceased drivers, and was detected only in 2 oral fluid samples (54.9 and 203 ng/ml) collected from 2738 randomly stopped car drivers. By comparison internationally, this frequency is in accordance with those European countries that are not highly infected with cocaine (under publication). Performance was checked each day with three samples spiked with 10, 125, and 1000 ng/ml BZE. The results were accepted only if the measured concentrations were within the range of nominal value ±3SD (for urine 8.96–11.1, 116–135, and 939–1062, respectively; for blood 7.6–12.4, 115–135, and 926–1074, respectively); otherwise, a new calibration was required. The method was also successfully adapted, validated, and applied for determination of BZE in more than 3000 oral fluid samples by another DRUID participant (National Institute for Health and Welfare, Helsinki, Finland) [19].
Conclusion A simple and fast method has been developed for qualitative and quantitative determination of BZE in oral fluid, blood, and urine samples with liquid–liquid extraction followed by rapid GC–MS analysis in SIM mode. The main advantage of the method is that after the extraction of 22 illicit and licit drugs with BuOAc, BZE can be easily extracted from the remaining sample with CH2Cl2 without additional sample preparation. Apart from the oven heating parameters, the amount of BZE can be determined with the same GC–MS configuration that is applied for the other drugs, saving time by eliminating methodology changes. The cutoff values proved to be 8.0 ng/ml for blood and urine, and 25 ng/ml for oral fluid collected by the StatSure device. The method fulfilled all validation criteria [1–3, 14], produced good results in proficiency tests, and has been successfully adopted by another laboratory. Acknowledgments This work was financially supported by TREN05-PF6TR-S07. 61320-518404-DRUID. The authors thank to Pirjo
Forensic Toxicol (2012) 30:59–65 Lillsunde, Kaarina Langel, Kari Ariniemi, and Teemu Gunnar (National Institute for Health and Welfare, Helsinki, Finland) for their help and advice in adaptation of the LLE and GC–MS analysis system to our laboratory. The authors also thank Tu¨nde Benk} o and Edit Kopasz for technical assistance.
References 1. Kankaanpaa A, Gunnar T, Ariniemi K, Lillsunde P, Mykkanen S, Seppala T (2004) Single step procedure for gas chromatographymass spectrometry screening and quantitative determination of amphetamine-type stimulants and related drugs in blood, serum, oral fluid and urine samples. J Chromatogr B 810:57–68. doi: 10.1016/jchromb.2004.07.035 2. Gunnar T, Ariniemi K, Lillsunde P (2005) Determination of 14 benzodiazepines and hydroxyl metabolites, zaleplon and zolpidem as tert-butyldimethylsilyl derivatives compared with other common silylating reagents in whole blood by gas chromatography–mass spectrometry. J Chromatogr B 818:175–189. doi: 10.1016/jchromb.2004.12.032 3. Gunnar T, Ariniemi K, Lillsunde P (2005) Validated toxicological determination of 30 drugs of abuse as optimized derivatives in oral fluid by long column fast gas chromatography/electron impact mass spectrometry. J Mass Spectrom 40:739–753. doi: 10.1002/jms.846 4. Kolbrich EA, Barnes AJ, Gorelick DA, Boyd SJ, Cone EJ, Huestis MA (2006) Major and minor metabolites of cocaine in human plasma following controlled subcutaneous cocaine administration. J Anal Toxicol 30:501–510 5. Kato K, Hillsgrove M, Weinhold L, Gorelik DA, Darwin WD (1993) Cocaine and metabolite excretion in saliva under stimulated and nonstimulated conditions. J Anal Toxicol 17:338–341 6. Gerlits J (1993) GC/MS quantitation of benzoylecgonine following liquid–liquid extraction of urine. J Forensic Sci 38:1210–1214 ´ lvarez I, Palos F, Bermejo AM, Ferna´ndez P, Tabernero MJ 7. A (2006) Simultaneous determination of methadone, cocaine and their metabolites in urine by GC–MS. Anal Lett 39:1393–1399. doi:10.1080/00032710600668699 8. Thompson LK, Yousefnejad D, Kumor K, Sherer M, Cone EJ (1987) Confirmation of cocaine in human saliva after intravenous use. J Anal Toxicol 11:36–38
65 9. Ca´mpora P, Bermejo AM, Tabernero MJ, Ferna´ndez P (2003) Quantitation of cocaine and its major metabolites in human saliva using gas chromatography–positive chemical ionization–mass spectrometry (GC–PCI–MS). J Anal Toxicol 27:270–274 10. Clauwaert K, Decaestecker T, Mortier K, Lambert W, Deforce D, Van Peteghem C, Van Bocxlaer J (2004) The determination of cocaine, benzoylecgonine, and cocaethylene in small volume oral fluid samples by liquid chromatography–quadrupole-time-offlight mass spectrometry. J Anal Toxicol 28:655–659 11. Cognard E, Bouchonnet S, Staub C (2006) Validation of a gas chromatography–ion trap tandem mass spectrometry for simultaneous analyse of cocaine and its metabolites in saliva. J Pharm Biomed Anal 41:925–934. doi:10.1016/jpba.2006.01.041 12. Jufer R, Walsh SL, Cone EJ, Sampson-Cole A (2006) Effect of repeated cocaine administration on detection times in oral fluid and urine. J Anal Toxicol 30:458–462 13. Langel K, Engblom C, Pehrsson A, Gunnar T, Ariniemi K, Lillsunde P (2008) Drug testing in oral fluid—evaluation of sample collection devices. J Anal Toxicol 32:393–401 14. Peters FT, Maurer HH (2002) Bioanalytical method validation and its implications for forensic and clinical toxicology—a review. Accred Qual Assur 7:441–449. doi:10.1007/s00769-0020516-5 15. Griesemer EC, Liu Y, Budd RD, Raftogianis L, Noguchi TT (1983) The determination of cocaine and its major metabolite, benzoylecgonine, in postmortem fluids and tissues by computerized gas chromatography–mass spectrometry. J Forensic Sci 28:894–900 16. Chinn DM, Crouch DJ, Peat MA, Finkle BS, Jennison TA (1980) Gas chromatography–chemical ionization mass spectrometry of cocaine and its metabolites in biological fluids. J Anal Toxicol 4:37–42 17. Needleman SB, Goodin K (1991) Liquid–liquid extraction systems for THC-COOH and benzoylecgonine. J Anal Toxicol 15:179–191 18. Jatlow P (1988) Cocaine: analysis, pharmacokinetics, and metabolic disposition. Yale J Biol Med 61:105–113 19. Langel K, Gunnar T, Ariniemi K, Rajamaki O, Lillsunde P (2011) A validated method for the detection and quantitation of 50 drugs of abuse and medicinal drugs in oral fluid by gas chromatography–mass spectrometry. J Chromatogr B 879:859–870. doi: 10.1016/j.jchromb.2011.02.027
123
