Anal Bioanal Chem (2007) 388:1415–1435 DOI 10.1007/s00216-007-1271-6

REVIEW

Bioanalytical procedures for determination of drugs of abuse in blood Thomas Kraemer & Liane D. Paul

Received: 6 February 2007 / Revised: 16 March 2007 / Accepted: 19 March 2007 / Published online: 28 April 2007 # Springer-Verlag 2007

Abstract Determination of drugs of abuse in blood is of great importance in clinical and forensic toxicology. This review describes procedures for detection of the following drugs of abuse and their metabolites in whole blood, plasma or serum: Δ9-tetrahydrocannabinol, 11-hydroxyΔ9-tetrahydrocannabinol, 11-nor-9-carboxy-Δ9-tetrahydrocannabinol, 11-nor-9-carboxy-Δ9-tetrahydrocannabinol glucuronide, heroin, 6-monoacetylmorphine, morphine, morphine-6-glucuronide, morphine-3-glucuronide, codeine, amphetamine, methamphetamine, 3,4-methylenedioxymethamphetamine, N-ethyl-3,4-methylenedioxyamphetamine, 3,4methylenedioxyamphetamine, cocaine, benzoylecgonine, ecgonine methyl ester, cocaethylene, other cocaine metabolites or pyrolysis products (norcocaine, norcocaethylene, norbenzoylecgonine, m-hydroxycocaine, p-hydroxycocaine, mhydroxybenzoylecgonine, p-hydroxybenzoylecgonine, ethyl ecgonine, ecgonine, anhydroecgonine methyl ester, anhydroecgonine ethyl ester, anhydroecgonine, noranhydroecgonine, N-hydroxynorcocaine, cocaine N-oxide, anhydroecgonine methyl ester N-oxide). Metabolites and degradation products which are recommended to be monitored for assessment in clinical or forensic toxicology are mentioned. Papers written in English between 2002 and the beginning of 2007 are reviewed. Analytical methods are assessed for their suitability in forensic toxicology, where special requirements have to be met. For many of the analytes sensitive immunological

T. Kraemer (*) Institute of Legal Medicine, Saarland University, 66421 Homburg (Saar), Germany e-mail: [email protected] L. D. Paul Institute of Legal Medicine, Ludwig Maximilians University, 80337 Munich, Germany

methods for screening are available. Screening and confirmation is mostly done by gas chromatography (GC)–mass spectrometry (MS) or liquid chromatography (LC)–MS(/MS) procedures. Basic information about the biosample assayed, internal standard, workup, GC or LC column and mobile phase, detection mode, and validation data for each procedure is summarized in two tables to facilitate the selection of a method suitable for a specific analytic problem. Keywords Drugs of abuse . Gas chromatography–mass spectrometry . Liquid chromatography–tandem mass spectrometry . Blood . Determination . Forensic toxicology Abbreviations AEME anhydroecgonine methyl ester APCI atmospheric pressure chemical ionization BZE benzoylecgonine CI chemical ionization CYP cytochrome P-450 CZE capillary zone electrophoresis DAD diode array detection EI electron ionization EME ecgonine methyl ester ESI electrospray ionization GC gas chromatography HFBA heptafluorobutyric anhydride (S)(S)-(-)-heptafluorobutyrylprolyl chloride HFBPCl HPLC high-performance liquid chromatography LC liquid chromatography LIF laser-induced fluorescence LLE liquid–liquid extraction LOQ limit of quantification LOD limit of detection 6-MAM 6-monoacetylmorphine

1416

MDA MDEA MDMA M3G M6G MRM MS MS/MS (R)-(-)MTPCl NICI OH-THC PICI SIM SPE THC THC-COO gluc THCCOOH UPLC

Anal Bioanal Chem (2007) 388:1415–1435

3,4-methylenedioxyamphetamine N-ethyl-3,4-methylenedioxyamphetamine 3,4-methylenedioxymethamphetamine morphine-3-glucuronide morphine-6-glucuronide multiple reaction monitoring mass spectrometry tandem mass spectrometry (R)-(-)-a-methoxy-a-trifluoromethylphenylacetyl chloride negative ion chemical ionization 11-hydroxy-Δ9-tetrahydrocannabinol positive ion chemical ionization selected ion monitoring solid-phase extraction Δ9-tetrahydrocannabinol 11-nor-9-carboxy-Δ9-tetrahydrocannabinol glucuronide 11-nor-9-carboxy-Δ9-tetrahydrocannabinol ultraperformance liquid chromatography

Introduction Determination of drugs of abuse in blood is of great importance in many fields; however, the list of drugs of abuse can vary, depending on the context in which the samples are analyzed. Clinical toxicology, forensic toxicology, workplace drug testing, testing of driving under the influence of drugs, doping analysis and rehabilitation programs all focus on different drugs of abuse. In this review, methods for the analysis mainly of illicit drugs in blood are covered. In particular, the following drugs and their metabolites have been included: Δ9-tetrahydrocannabinol (THC), 11-hydroxy-Δ9-tetrahydrocannabinol (HO-THC), 11-nor-9-carboxy-Δ9-tetrahydrocannabinol (THC-COOH), 11-nor-9-carboxy-Δ9-tetrahydrocannabinol glucuronide (THC-COOgluc), heroin, 6-monoacetylmorphine (6-MAM), morphine, morphine-6-glucuronide (M6G), morphine-3-glucuronide (M3G), codeine, amphetamine, methamphetamine, 3,4-methylenedioxymethamphetamine (MDMA), N-ethyl-3,4-methylenedioxyamphetamine (MDEA), 3,4-methylenedioxyamphetamine (MDA), cocaine, benzoylecgonine (BZE), ecgonine methyl ester (EME), cocaethylene, other cocaine metabolites or pyrolysis products. The methods were assessed mainly for their suitability for forensic toxicological purposes. In this field of toxicology, great demands are made on complete certainty of results, and legal requirements in addition to national and international guidelines have to be considered. The methods must provide quantitative results, e.g., if legal limits exist (e.g., for driving

under the influence of drugs) or pharmacodynamic or pharmacokinetic interpretations are necessary. Because of the multitude of papers related to this topic and the rapid developments in analytical technology, the present review is restricted to papers from 2002 up to the beginning of 2007. The state of affairs concerning determination of drugs of abuse in blood before 2002 can be gleaned in the review of Moeller and Kraemer [1]. The principal information on the chromatographic procedures is summarized in Tables 1 and 2 to simplify rapid selection of a method suitable for an actual analytical problem. The drugs are listed in the tables according to the international nonproprietary names or the common names. The kind of biosample used is given in the “matrix/sample” column (blood, plasma, serum, etc.). If other matrices were additionally investigated, they are mentioned in parentheses. Because the selection of the internal standard is of importance for the quality of the data of a method, this information is given in the “internal standard” column. For mass spectrometry (MS) procedures, stable isotopes are the most suitable internal standards, because they have the same chemical properties as the corresponding analyte. Sample preparation is concisely summarized in the “workup” column. The principal information on the gas chromatography (GC) or liquid chromatography (LC) column and mobile phase as well as on the detection mode is listed. Validation data such as limit of detection (LOD), limit of quantification (LOQ) or linearity are summarized for easy estimation of whether a procedure can be useful to solve an actual toxicological case. As accuracy and precision data were within the required limits, these data were omitted to save space. The methods used to calculate recovery values differed between the publications, making direct comparisons useless [2]. In addition, this parameter is not an issue if all other validation criteria are met, especially if the sensitivity is sufficient. Nowadays, screening procedures in blood for the most relevant drugs of abuse on the basis of immunoassays are available. Kroener et al. [3] evaluated four commonly used immunoassay kits for their efficiency in screening for drugs of abuse in whole blood. Six groups of illicit drugs (opiates, cannabinoids, amphetamines, cocaine and benzoylecgonine, benzodiazepines and methadone) were determined. Confirmation of results using chromatographic procedures was made in all cases. The authors concluded that the microtiterplate immunoassays revealed higher sensitivities and proved to be advantageous when detecting the lowest drug concentrations. However, homogenic assays such as ADx or CEDIA might be an alternative for clinical toxicology, featuring faster and easier handling [3]. Kupiec et al. [4] found ELISA tests suitable for screening for amphetamines even in postmortem samples. However, even though Spiehler et al. [5]

THC THC-COOH

THC OH-THC THC-COOH

THC THC-COOH

[19]

[13]

[12]

Opiates [37]

Codeine Morphine 6-MAM Hydrocodone Hydromophone Oxycodone Oxymorphone

THC OH-THC THC-COOH CBD CBN THC THC-COOH OH-THC

[18]

[17]

Blood, 1 mL

THC

[2]

Blood, 2 mL (postmortem)

Plasma, 1 mL

Plasma, 1 mL

Blood, 1 mL

Plasma, blood, 0.025 mL

Plasma, 1 mL

Blood, plasma, 1 g (premortem and postmortem)

Serum, 1 mL

Matrix/sample

Cannabinoids [14] THC OH-THC THC-COOH [15] THC OH-THC THC-COOH

Reference Compound

Table 1 Gas-chromatographic procedures

Protein precipitation with methanol (β-glucuronidase), derivatization with hydroxylamine, SPE, silylation with BSTFA + 1% TMCS

d3 analogues Oxycodone-d6

d3 analogues

d3 analogues

Escherichia coli β- glucuronidase hydrolysis, SPE, silylation with BSTFA + 1% TMCS Protein precipitation with acetonitrile, SPE, derivatization with TFAA/HFIP

d3 analogues

d3 analogues

THC-d3 OH-THC-d3 THC-COOH-d3

THC-d3

d3 analogues, 100 ng/mL

d3 analogues

Internal standard

Protein precipitation with cold (−20 °C) acetonitrile, SPE, silylation with BSTFA + 1% TMCS

SPE (C18), methylation (TMAH/DMSO, iodomethane, HCl, extraction by isooctane)

Automated SPE (C18 ec), silylation with BSTFA

SPE (strata C18 ec), methylation (TBAH/DMSO, iodomethane, HCl, extraction by isooctane) Protein precipitation with acetonitrile, SPE (SPEC C18AR discs), methylation with TMAH/DMSO, iodomethane, HCl and extraction with isooctane LLE (n-hexane), PFPA/PFPOH

Workup

Detection mode and conditions

Linearity (ng/mL)

EI-MS, SIM 1–100

Restek base deactivated guard column (5 m×0.25 mm) and HP-1MS (30 m×0.25 mm, 0.25 μm)

EI-MS, SIM 50–1,000 50–1,000 10–500 50–1,000 50–1,000 100–2,000 50–1,000

EI-MS, SIM 0.25–10 0.25–10 0.25–50 0.25–10 0.25–10 EI-MS, SIM 2–30 Large-volume 2–30 programmed-temperature 10–150 vaporization injection MDN-5 column (30 m×0.25 mm, 0.25 μm) 1–100 2D- GC (two chromatographic EI-MS 1–100 columns with a Deans switch) RTX-200 (20 m×0.18 mm, 0.20 μm) and DB-17 (15 m×0.25 mm, 0.25 μm) 0.5–50 HP5 MS (30 m×0.25 mm, PICI-MS, 0.25 μm) methane as 0.5–50 1.0–100 reactant gas, SIM 0.5–100 HP-DB-1 (30 m×0.32 mm, NICI-MS, 0.25 μm) methane as 2.5–100 reactant gas, SIM

HP5 MS (30 m×0.25 mm, 0.25 μm) HP5 MS (30 m×0.25 mm, 0.25 μm)

HP1 (12 m×0.2 mm, 0.33 μm) EI-MS, SIM 0.5–10 0.5–10 2.5–50 HP5 MS (30 m×0.25 mm, EI-MS, SIM 1–30 0.25 μm) 1–30 5–150

Gas chromatography

NA

0.5 2.5

5 (6-MAM) 3.5 20 (oxycodone) 2.5 00.7 3.5 3.0 2.5 3.5

NA

NA

1.0 1.0

0.5 0.5 1.0

0.24 0.15 0.26 0.29 1.1 0.7 0.8 0.7

0.5

0.52 0.49 0.65 0.5–1

Limit of detection (ng/mL)

0.80 0.51 0.88 0.95 3.9 NA

1.0

0.8–2

0.62 0.68, 3.35

Limit of quantification (ng/mL)

Anal Bioanal Chem (2007) 388:1415–1435 1417

Morphine

Codeine Morphine 6-MAM Hydrocodone Hydromophone Oxycodone Dihydrocodeine Thebaine Codeine Morphine 6-MAM Hydrocodone Hydromophone Oxycodone Oxymorphone M6G M3G

Morphine

[42, 43]

[44]

[38]

[45]

Amphetamines [61] Amphetamine Methamphetamine MDMA MDA MDEA Others (not validated)

[29]

[41]

Blood, 1 mL

Morphine 6-MAM Morphine 6-MAM Codeine

[35]

Workup

Serum, 2 mL

Plasma, 1 mL

Plasma, 1 mL

Blood, 1 mL

LLE (KOH; sodium sulfate; tert-butyl methyl ether), HCl before evaporation, aqueous KOH, LLE (toluene), on-line derivatization (MBTFA)

Protein precipitation with acetonitrile (β-glucuronidase), fourfold LLE at pH 9.3, derivatization with methoxyamine/pyridine and propionic anhydride, hexane/ chloroform/ammonium ydroxide SPE (C18), derivatization with PFBBr in DMF + diisopropylethylamine, BSTFA/ pyridine SPE (C18), derivatization with pentafluorobenzyl chloroformate/ pyridinex and MSTFA+1% TMCS/pyridine

SPE (mixed mode), derivatization with TFAA SPE (Clean Sreen, CSDAU), Blood, 3 mL codeine analysis underivatized, (femoral, postmortem) (liver, subsequent derivatization with cerebrospinal fluid, BSTFA+1% TMCS vitreous humor) (β-Glucuronidase), protein Blood, 1 mL precipitation with cold (cardial, acetonitrile, SPE (Cerex Clin II), post-mortem) derivatization with MSTFA (brain, liver, stomach contents, vitreous humor) Automated SPE (Bond Elute Blood, 3 mL Certify I) (β-glucuronidase), (postmortem) derivatization with BSTFA+ 1% (urine, liver, TMCS kidney, muscle)

Matrix/sample

Reference Compound

Table 1 (continued)

DB-5 (30 m×0.25 mm, 0.25 μm)

HP-Ultra-1 (12 m×0,2 mm, 0.33 μm)

Restek guard column, deactivated with 5% phenylmethylsilicone (1 m×0.5 mm) and DB-1 (15 m×0.25 mm, 0.25 μm), DB-5MS (15 m×0.25 mm, 0.25 μm)

HP-Ultra-1 (7 m×0.2 mm)

Morphine-d3

Codeine-d3 Morphine-d6 6-MAM-d6 Hydrocodone-d3 Hydromophone-d3 Oxycodone-d3 Dihydrocodeine-d6 d6 analogues oxymorphone-d3

d3 analogues

Morphine-d3

HP-1MS (25 m×0.2 mm, 0.33 μm)

HP-5 (30 m×0.25 mm, 0.25 μm)

Morphine-d3 Codeine-d3

Amphetamine-d5 Methamphetamine-d5 MDMA-d5 MDA-d5 MDEA-d5

HP1 (15 m)

Gas chromatography

Morphine-d3

Internal standard Linearity (ng/mL)

2–1,700 1–1,700 1–1,700

1.25–320 nM

10–1,289 mM 15–1,920 mM

EI-MS, SIM 5–100 5–100 5–100 5–100 5–100

NICI-MS, methane as reactant gas, SIM NICI-MS, methane as reactant gas, SIM

EI-MS, SIM 6.25–1,600 12.5–1,600 1.56–3,200 6.25–1,600 1.56–1,600 12.5–800 6.25–3,200 12.5–1,600 Ion-trap 10–2,000 EI-MS, full scan

EI-MS, SIM 10–1,000

EI-MS, SIM, only one ion per analyte

EI-MS, SIM 50–1,600

Detection mode and conditions

NA

1.25 nM

3 nM 1.5 nM

6.25 12.5 1.56 6.25 1.56 12.5 6.25 12.5 10

10

No validation data 2 1 1

Limit of quantification (ng/mL)

2.5 3.4 6.9 3.1 5.0

0.3 nM

10 nM 15 nM

1.56 3.12 0.78 3.13 0.78 6.25 1.56 6.25 2

5

NA

NA

Limit of detection (ng/mL)

1418 Anal Bioanal Chem (2007) 388:1415–1435

MDMA, MDEA, MDA

Amphetamine Methamphetamine Amphetamine

MDMA MDA HMMA HMA

MDMA MDA

[74]

[73]

[77]

[78]

[65]

Amphetamine Methamphetamine MDMA MDA MDEA

[75]

Amphetamine Methamphetamine MDMA MDA MDEA Others [56] Amphetamine Methamphetamine MDMA MDA MDEA Others Amphetamines, chiral methods [79] Amphetamine Methamphetamine MDMA MDA MDEA [76] Amphetamine

[57]

LLE (NaOH; n-hexane), derivatization with (S)-HFBPCI SPE (mixed-mode), derivatization with (S)-HFBPCl

Plasma, 1 mL

Plasma (urine)

Plasma (urine)

Blood, 1.0 g

Plasma, 0.2 mL

Plasma, 0.2 mL

(R/S)-MDMA-d5 (R/S)-MDA-d5 pholedrine

(R/S)-Amphetamine-d11 (R/S)-Methamphetamine-d5 (R/S)-Amphetamine-d11

(R/S)-MDMA-d5 (R/S)-MDA-d5

(R/S)-Amphetamine-d11 (R/S)- Methamphetamine-d5 (R/S)-MDMA-d5 (R/S)-MDA-d5

(R/S)-Amphetamine-d5 (R/S)-Methamphetamine-d5 (R/S)-MDMA-d5 (R/S)-MDA-d5 (R/S)-MDEA-d5 (R/S)-Amphetamine-d5

Phenomenex-5% phenyl (15 m×0.25 mm, 0.25 μm)

HP-5 MS (30 m×0.25 mm, 0.25 μm) HP-5 MS (30 m×0.25 mm, 0.25 μm) Ultra 2-5% (12 m×0.22 mm, 0.33 μm)

HP-5 MS (30 m×0.25 mm, 0.25 μm)

EI-MS, SIM 12.5–200 (R and S MDMA and HMMA) 1.25–20 (R and S MDA and HMA) EI-MS, SIM 12.5–200 (R and S MDMA and

41 4.7 (MDA)

NA

5

5–250 (R and S Amphetamine, Methamphetamine, MDMA, MDEA) 1–50 (R and S MDA) NA 5–250 (R and S MDMA, MDEA) 1–50 (R and S MDA) 5–250 (R and S) NA NICI-MS, SIM EI-MS, SIM 5–400 (R and S)

NICI-MS, SIM

NICI-MS, SIM NICI-MS, SIM

SGE-BPX5 (15 m×0.25 mm, 0.25 μm) HP-5 MS (30 m×0.25 mm, 0.25 μm)

0.049 [(R)-(-)] 0.195 [(S)-(+)] NA

0.004 μg/g

EI-MS, SIM 0.004–3 μg/g

HP-5 MS (30 m×0.25 mm, 0.25 μm)

0.006–50

5

EI-MS, SIM 5–1,000 (quant.)

HP-5 MS (30 m×0.25 mm, 0.25 μm)

Amphetamine-d5 Methamphetamine-d5 MDMA-d5 MDA-d5 MDEA-d5

25

20–5,000 EI-MS, full-scan (screening), SIM (quant.)

DB-5 MS (30 m×0.32 mm, 1 μm)

NA

EI-MS, SIM 12.5–2,000 ng/g

HP-5 MS (30 m×0.25 mm, 0.25 μm)

12.5 ng/g

EI-MS, SIM 12.5–2,000 ng/g

HP-5 MS (30 m×0.25 mm, 0.25 μm)

NA

EI-MS, SIM 0.4–30

HP-5 MS (30 m×0.25 mm, 0.25 μm)

Methylmexiletine

Methamphetamine-d5

Amphetamine-d5 Methamphetamine-d5 (Demethylselegiline-d8) Methamphetamine-d5

Enzymatic hydrolysis, SPE (mixed-mode), (R/S)-MDMA-d5 derivatization amines: (R)-MTPCl, (R/S)-MDA-d5 pholedrine

SPE (mixed-mode), derivatization with (S)-HFBPCl LLE (NaOH; isooctane), derivatization with (S)-TPC Enzymatic hydrolysis, SPE (mixed-mode), derivatization amines: (R)-MTPCl, phenols: HMDS

SPE (mixed-mode), derivatization with (S)-HFBPCl

LLE (ammonium hydroxide; 1-chlorobutane), derivatization with (R)-(-)-MTPCl

Blood, 0.50 g

Plasma, serum, 0.025–0.2 mL

SPE (mixed mode), derivatization with HFBA

LLE (KOH; isooctane), HCl before evaporation, derivatization with TFAA Extrelut (borate buffer, pH 10.5; ethyl acetate), on-column derivatization (propylchloroformate) Extrelut (borate buffer, pH 10.5; ethyl acetate), on-column derivatization (propylchloroformate) Single step extraction/derivatization, (NaHCO3/KOH; toluene; HFBA)

Plasma, 1 mL

Blood, serum, 0.2 mL (oral fluid)

Blood, 0.5 g (urine, hair)

Amphetamine Methamphetamine

[58]

[59]

Amphetamine Plasma, 0.5–2 mL Methamphetamine (hair) (Demethylselegiline) Amphetamine Blood, 0.5 g Methamphetamine

[62]

14 1.6 (MDA)

NA

NA

NA

NA

NA

NA

NA

NA

NA

5 ng/g

NA

NA

Anal Bioanal Chem (2007) 388:1415–1435 1419

[104]

[107]

[103]

Cocaine [97]

Cocaine Cocaethylene Norcocaine Norcocaethylene Ecgonine methyl ester Ecgonine ethyl ester BZE Nor-BZE m-Hydroxy-BZE p-Hydroxy-BZE Ecgonine AEME Anhydroecgonine ethyl ester Anhydroecgonine Noranhydroecgonine Cocaine Cocaethylene Norcocaine Norcocaethylene Ecgonine methyl ester Ecgonine ethyl ester BZE Nor-BZE Ecgonine m-Hydroxy-BZE AEME Cocaine Norcocaine Ecgonine methyl ester BZE m-Hydroxycocaine p-Hydroxycocaine m-Hydroxy-BZE p-Hydroxy-BZE Cocaine Cocaethylene

HMMA HMA

Reference Compound

Table 1 (continued)

Cocaine-d3 Cocaethylene-d3

SPME (PDMS, 100 μm)

HP-5 MS (12 m×0.2 mm, 0.33 μm)

HP-1MS (12 m×0.2 mm, 0.33 μm)

Cocaine-d3 Norcocaine-d3 Ecgonine methyl ester-d3 BZE-d3

Protein precipitation (acetonitrile), SPE (Clean Screen ZSDAU), derivatization with MTBSTFA + TBDMCS and BSTFA + TMCS

Plasma, 1 mL

Plasma, 1 mL

100% methyl siloxane (12 m × 0.20 mm, 0.33 μm)

Cocaine-d3 Cocaethylene-d3 Norcocaine-d3 Ecgonine methyl ester-d3 BZE-d3

Gas chromatography

Blood, 3 mL (urine, Protein precipitation (acetonitrile), muscle tissue) SPE (mixed mode), derivatization with PFPA/PFPOH after converting the analytes into hydrochloride salts

Internal standard

DB-5MS (15 m×0.25 mm)

SPE (C8 + benzenesulfonic acid), further different extraction/derivatization steps; PFP, iodopentane, DMF/dipropylacetal)

phenols: HMDS

Workup

d3 analogues, so far as available

Blood, 3 mL (urine)

Matrix/sample

HMMA) 1.25–20 (R and S MDA and HMA)

Linearity (ng/mL)

EI-MS, SIM 25–1,000

EI-MS, SIM 2.5–50 50–1,000

EI-MS, SIM 1–800 6–6,400 (ecgonine)

EI-MS, SIM 1–2,000 1–50 2–50 4–50 1–2,000 2–2,000 4–2,000 2–1,000 1–2,000 1–2,000 16–2,500 1–50 2–50 2–2,500 2–50

Detection mode and conditions

NA

2.5

2 13 (AEME) 25 (nor-BZE) 50 (OH-BZE) 800 (ecgonine)

NA

Limit of quantification (ng/mL)

19 11

NA

2 13 (AEME) 25 (nor-BZE) 50 (OH-BZE) 640 (ecgonine)

NA

Limit of detection (ng/mL)

1420 Anal Bioanal Chem (2007) 388:1415–1435

Serum, plasma, blood, 1 mL

LLE with butyl acetate, (underivatized for GC-ECD for low-dosed benzodiazepines, zaleplon, zopiclone), two-step silylation with MTBSTFA and MSTFA for GC-MS

Flurazepam (!)

ECD and (DB-35 (30 m×0.53 mm, 1.0 μm) for ECD) DB-35 ms EI-MS, SIM (30 m×0.32 mm, 0.25 μm) for MS

10–500 2–200 2–1,000 5–1,000 2–1,000 2–200 2–500

10 2 2 10 5 5 10

<2 5 (THC, codeine) 2 (cocaine)

AEME anhydroecgonine methyl ester, BSTFA N,O-bis(trimetylsilyl)trifluoroacetamide, BZE benzoylecgonine, CBD cannabidiol, CBN cannabinol, DMF dimethylformamide, DMSO dimethyl sulfoxide, ec endcapped, ECD electrochemical detection, EI electron ionization, GC gas chromatography, (S)-HFBPCl (S)-(-)-heptafluorobutyrylprolyl chloride, HFIP hexafluoroisopropanol, HMA 4-hydroxy-3-methoxyamphetamine, HMDS 1,1,1,3,3,3-hexamethyldisilazane, HMMA 4-hydroxy-3-methoxymethamphetamine, LLE liquid–liquid extraction, 6-MAM 6-monoacetylmorphine, MDA 3,4-methylenedioxyamphetamine, MDEA N-ethyl-3,4-methylenedioxyamphetamine, MBTFA N-methyl-bis(trifluoroacetamide), MDMA 3,4-methylenedioxymethamphetamine, M3G morphine-3-glucuronide, M6G morphine-6-glucuronide, MS mass spectrometry, MSTFA N-methyl-N-trimethylsilyltrifluoroacetamide, MTBSTFA N-methyl-N-(tert-butyldimethylsilyl)trifluoroacetamide, (R)-MTPCl (R)-(-)-α-methoxy-α-trifluormethylphenylacetyl chloride, NA not available, NICI negative ion chemical ionization, OH-THC 11-hydroxy-Δ9-tetrahydrocannabinol, PDMS polydimethylsiloxane, PFBBr pentafluorobenzyl bromide, PFPA pentafluoropropionic anhydride, PFPOH pentafluoropropanol, PICI positive ion chemical ionization, SIM single ion monitoring, SPE solid-phase extraction, SPME solid-phase microextraction, TBAH tetrabutylammoniumhydroxide, TBDMCS tert-butyldimethylchlorosilane, TFAA trifluoroacetic anhydride, THC Δ9tetrahydrocannabinol, THC-COOH 11-nor-9-carboxy-Δ9-tetrahydrocannabinol, TMAH tetramethylammoniumhydroxide, TMCS trimethylchlorosilane, (S)-TPC (S)-(-)-trifluoroacetylprolyl chloride

Mixed analytes [109] THC OH-THC THC-COOH Codeine Morphine 6-MAM Cocaine (and 44 others)

Anal Bioanal Chem (2007) 388:1415–1435 1421

THC-COOH THC-COO-gluc

THC CBD

[23]

[21]

Morphine M6G M3G

Morphine

Morphine

[47]

[48]

[26]

Morphine M6G M3G

THC

[22]

Opiates [46]

THC OH-THC THC-COOH (THC-COO-gluc)

[20]

SPE (HF Bond Elute Certify)

LLE with hexane

Automated SPE (C18)

Plasma, 0.5 mL

Serum, 0.2 mL LLE with THF, Zeolite Y SPE (size-exclusion chromatography)

SPE (C8)

Serum, Automated SPE 0.6 mL (from (C6) arterial blood)

Serum, 1 mL SPE (C18 ec)

Mouse (!) SPMEM blood (urine, brain)

Plasma, 0.5 mL (urine)

Plasma, 0.25 mL (oral fluid)

Plasma, 1 mL

Plasma, 0.5 mL

Cannabinoids [24] THC (THC-COOH, data not shown) Protein precipitation with acetonitrile, SPE (CN)

Matrix/sample Workup

Reference Compound

NA

NA

NA

NA

THC-d3

d3 analogues

μBondapak C18 (300 mm×4 mm, 10 μm)

Lichrocart (4 mm×4 mm) and (250 mm×5 mm)

Cp-Sper C8 (250 mm×4.6 mm)

Symmetry C18 (150 mm×4.9 mm, 5 μm)

RP C18 column (200 mm×4.6 mm) and C18 guard column (50 mm×4.6 mm)

Zorbax Eclipse XDB C8 column (150 mm×2.1 mm, 5 μm)

Luna 3u PhenylHexyl column (50 mm×2 mm, 3 μm) and Polar-RP guard cartridge (4 mm×2.0 mm) XTerra MS C18 column (150 mm×2.1 mm, 3.5 μm)

d3 analogues

THC-d3

Nucleosil C18 column (125 mm×4.6 mm, 3 μm)

Stationary phase

Lopinavir

Internal standard

Table 2 High-performance liquid-chromatographic procedures

1 mM octane sulfonic acid (as ion pairing agent) in 8% acetonitrile in water/ 30 mM phosphate buffer pH 3 0.01 M potassium dihydrogen phosphate pH 2.1/acetonitrile(89:11) and heptane sulfonic acid 0.4 g/L Acetonitrile/0.2 M potassium dihydrogen phosphate buffer (gradient) 5% methanol/3% acetonitrile/0.5 mM sodium acetate, 0.012 M potassium dihydrogen orthophosphate/0.148 mM phosphoric acid in water

Acetonitrile/ methanol/20 mM ammonium acetate pH 4.0 (41:41:18) Methanol/water (85:15)

1 mM ammonium formate/ methanol (10:90)

Methanol/acetonitrile/ 5 mM tetrabutylammonium perchlorate (as counterion) pH 3.2 (50:25:25) Acetonitrile/5 mM ammonium acetate pH 6.5 (gradient)

Mobile phase (v/v)

20–200

NA

NA

10 90 60

NA

0.6 1.1

NA

0.2 0.2 1.6

NA

No validation 10 data

10.12–202.4 10

5 5 25

5–90 5–100 25–580 UVD (210 nm) in sequence with ECD

FLD (excitation 210 nm, emission 350 nm) FLD

NA

NA

NA

NA

2.4 4.5

10–500

UVD (210 nm)

APCI-MS, SIM (only one ion per analyte)

0.5

0.8 0.8 4.3

1–100 1–100 5–500

ESI-MS/MS, positive ion mode, MRM, 2 transitions ESI-MS/MS, positive mode, MRM, 2 transitions ESI-MS/MS, positive mode, MRM

0.5–100

5

Limit of Limit of quantification detection (ng/mL) (ng/mL)

5–100

Linearity (ng/mL)

UV (215 nm)

Detection mode

1422 Anal Bioanal Chem (2007) 388:1415–1435

Morphine M6G M3G

Morphine M6G M3G

Heroin 6-MAM Morphine M6G M3G Methadone EMDP (acetylcodeine, codiene, cocaine, benzolecgonine, norcocaine) Morphine M6G M3G

[31]

[50]

[49]

[51]

Plasma, 0.1 mL (brain perfusion fluid, sheep plasma)

Plasma, 0.25 mL

Plasma, 0.25 mL

plasma, 0.25 mL

Plasma, 0.5 mL

Protein precipitation with acetonitrile

Codeine-d3 Morphine-d3 M3G-d3 Hydromorphone-d3 Oxycodone-d3

d3 analogues

5 mM ammonium formate pH 4.0/acetonitrile (gradient)

0.01% TFA in water/0.01% TFA in acetonitrile (11:89), 4 mL/min

Acetonitrile/water/TFA (90.99:9:0.01), 0.7 mL/min

Acetonitrile/methanol/ 10 mM formate pH3 (2.5:2.5:95)

10 mM potassium dihydrogen phosphate, 2 mM 1heptanesulfonic acid pH 2.5/ acetonitrile (90:10)

0.05%TFA (purification), Column switching: in-line filter (A-431, 5 μm), HYPurity C18 acetonitrile/5 mM ammonium acetate (70:30) (separation) guard column (10 mm×3 mm, 3 μm) (purification, desalting) and ZIC HILIC 50 mm×4.6 mm, 5 μm) (hydrophilic interaction chromatography) C12 MAX-RP Ammonium formate (pH (150 mm×2 mm, 4 μm), 3.5)/acetonitrile (gradient) No separation of hydrocodone, morphine, M3G and M6G

RP guard column (10 mm×3 mm) and RP Zorbax Bonus (150 mm×4.6 mm, 5 μm)

Betasil silica (50 mm×3.0 mm, 5 μm), packed silica, normal phase, high flow rate

Betasil silica (50 mm×3.0 mm, 5 μm)

d3 analogues Automated SPE (C18), 96-well plate, automatic liquid handler, 96channel programmable liquid-handling workstation d3 analogues Automated SPE, (Oasis HLB), 96-well plate, automatic liquid handler, 96-channel programmable liquidhandling workstation SPE (mixed mode) Heroin-d6 Morphine-d3 M3G-d3 Methadone-d9

Guard column (4 mm×4 mm) and RP select B (250 mm×4 mm, 5 μm)

Phenomenex C8 guard column (4 mm×2 mm), Atlantis DC18 (150 mm×2.1 mm, 5 μm)

No internal standard

96-well plate SPE (MCX, d3 analogues mixed mode)

Blood, Protein precipitation 0.150 mL with methanol, (postmortem) immunoaffinity SPE

Plasma, 1 mL Automated SPE Codeine (C18 ec) Morphine M6G M3G Hydromophone Oxymorphone Oxycodone (and other opioids)

Morphine M6G M3G

[33]

[40]

Morphine M6G M3G

[39]

ESI-MS/MS, positive ion mode, MRM

ESI-MS/MS, positive ion mode, MRM, only one transition per analyte

FLD (excitation 235 nm, emission 345 nm) Wavelengthresolved laserinduced FLD, full emission spectra ESI-MS, positive ion mode, SIM, only one ion per analyte ESI-MS/MS, positive ion mode, MRM, only one transition per analyte ESI-MS/MS, positive ion mode, SRM, flow splitter, only one transition per analyte ESI-MS/MS, positive ion mode, MRM

10–1,000 10–1,000 5–500 10–1,000 1–100 1–100 10–1,000 (intraday/ interday

4.9 4.7 8.4 6.6 1.6 2.9 5.3

1.4 1.3 2.5 1.9 0.5 0.8 1.5

NA

0.78 0.53 1.49

0.78–500 0.53–500 1.49–1,000

NA

NA

NA

1 2 20

0.5 1.0 10

0.25 0.25 1.0

<5

5 10 10

5–500 5 (qualitative)

NA

0.5–50 1.0–100 10–1,000

0.5 0.5 5.0

NA

NA

0.5–200 0.5–200 5–2,000

25

25–250

Anal Bioanal Chem (2007) 388:1415–1435 1423

Cocaine [99]

Cocaine Norcocaine

Amphetamine Methamphetamine (selegiline, norselegiline) [64] Amphetamine MDMA MDA (PMA) [70] Amphetamine Methamphetamine MDMA MDEA MDA (ephedrine) Amphetamines, chiral methods [80] MDEA MDA HME

[63]

Plasma, 1 mL SPE (Clean Screen ZSDAU)

Cocaine-d3 Norcocaine-d3

N-Ethyl-3,4methylenedioxy benzylamine

Protein precipitation with Amphetamine-d11 methanol Methamphetamine-d14 MDMA-d5 MDA-d5 MDEA-d5

Plasma, 1 mL Enzymatic hydrolysis, (urine) SPE (CX)

Plasma, 0.05 mL (oral fluid)

Plasma, 1 mL LLE (NH4OH; Amphetamine-d5 butylchloride/ Methamphetamine-d5 acetonitrile), HCl before (selegiline-d8, evaporation norselegeline-d11) Blood, 0.5 mL LLE (K2CO3, pH 9.5; Ephedrine (!) (urine, tissue) hexane/ethyl acetate), HCl before evaporation

Amphetamine Rat (!) serum, Protein precipitation with Amphetamine-d11 Methamphetamine 30–200 μL zinc sulfate, SPE (Oasis HLB)

[68]

3,4-Dihydroxy benzylamine

1-Methyl-3phenylpropylamine

Internal standard

HHMA

LLE (borate buffer, Blood, pH 10; ethyl acetate), plasma, DIB-Cl 0.1 mL (rat blood) Plasma, 1 mL acid hydrolysis (HCl), SPE (SCX)

Matrix/sample Workup

[69]

Amphetamines [72] MDMA MDA

Reference Compound

Table 2 (continued)

Inertsil ODAS-AQ (100 mm×2 mm, 5 μm)

Water with 0.1% formic acid/methanol

ESI MS/MS, positive ion

FLD (excitation 286 nm, emission 322 nm)

ESI MS/MS, positive ion mode, MRM

Sonic spray ionization MS

Acetonitrile/water containing 0.001% formic acid (gradient)

10 mM ammonium acetate/ acetonitrile, (isocratic)

APCI MS/MS, positive ion mode, MRM

ESI MS/MS, positive ion mode, MRM

ECD

FLD (excitation 330 nm, emission 440 nm)

Detection mode

Water containing 0.1% formic acid/methanol, (73:27)

10 mM citric acid, 20 mM disodium hydrogen phosphate pH 4.0/acetonitrile/methanol (50:45:5) 0.1 M sodium acetate containing 0.1 M 1octanesulfonic acid and 4 mM EDTA pH 3.1/acetonitrile (82:18) 10 mM ammonium acetate pH 3.7/acetonitrile/methanol (72.5:25:2.5)

Mobile phase (v/v)

Chiral CBH, 150 mm×4 mm, 5 μm) 20 mM sodium dihydrogenphosphate pH 6.44, 50 mM EDTA disodium salt/2-propanol (93:7)

Hypersil BDS C18 guard column (7.5 mm×2.1mm, 3 μm) and Hypersil BDS C18 (100 mm× 2.1 mm, 3 μm) Hypersil BDS C18 (100 mm× 2.1 mm, 3 μm)

Hypersil BDS C18 guard column (10 mm×2.1 mm, 3 μm) and Hypersil BDS C18 (100 mm× 2.1 mm, 3 μm) MetaSil Basic (100 mm×2 mm; 3 μm)

4.6 mm Kromasil 100 n-butyl-silane (C4) (Teknokroma).

Wakopak Handy ODS column (150 mm×4.6 mm, 6 μm)

Stationary phase

28.2

NA

2.5–750

5–800 5–100 10–2,000 (R and S)

0.5–500

10–1,000

0.2–20

2.5

NA

0.5 1 (MDA)

NA

NA

NA

NA

0.5 0.2 0.25 0.2 1

5 2.5 5

NA

NA

9.2

0.75 0.50

Limit of Limit of quantification detection (ng/mL) (ng/mL)

0.3–10 (low) 0.3 10–1,000 (high)

50–1,000

2–500 1–250

precision and accuracy >15% for some analytes)

Linearity (ng/mL)

1424 Anal Bioanal Chem (2007) 388:1415–1435

Blood, 0.025 mL (heart, peripheral) (urine, vitreous humor) MDMA-d5 MBDB-d5 Cocaine-d3 BZE-d3

d analogues

SPE (Oasis HLB)

Automated SPE (C18) for cannabinoids and basic drugs separately

Protein precipitation with d analogues acetonitrile

Ecgonine methyl ester-d3 BZE-d3

Cannabinoids: Polar-RP guard cartridge (4 mm×2.0 mm) and Luna PhenylHexyl (50 mm× 2 mm, 3 μm) Basic drugs: Polar-RP guard cartridge (4 mm×2.0 mm) and Synergi Hydro-RP (150 mm×2 mm, 2 μm)

Atlantis dC18 (100 mm× 2.1 mm, 3 μm)

Synergi Polar RP guard column (4×2.0 mm) and Synergi Polar RP (150 mm× 2.0 mm, 4 μm)

0.5 1 (morphine, 6-MAM, MDA, MDEA, MBDB)

0.2 0.2 1.6 2.3 4.0 1.2 2.4 3.1 1.5 0.3

2

0.8 0.8 4.3 3.1 6.3 5.3 6.6 8.0 4.7 0.9 1–100 1–100 5–250 1–250 2–250 1–250 2–250 2–500 2–250 1–500 ESI-MS/MS, positive ion mode, MRM 2–3 transitions per analyte

Cannabinoids: 5 mM ammonium acetate pH 6.5/ acetonitrile (gradient) Basic drugs: 4 mM ammonium acetate pH 4.6/acetonitrile (gradient), postcolumn addition of acetonitrile

NA

2–250

NA

ESI-MS, positive ion mode, SIM, only two ions per analyte

0.1–7 μM

Acetonitrile/ammonium formate buffer pH 3.0 (gradient)

10 mM aqueous ammonium formate, 0.001% formic acid pH 4.5/acetonitrile (gradient)

mode, MRM APCI MS/MS, positive mode, MRM APCI MS/MS, positive ion mode, MRM

APCI atmospheric pressure chemical ionization, FLD fluorescence detection, DIB-Cl 4-(4,5-diphenyl-1H-imidazol-2-yl)benzoyl chloride, ESI electrospray ionization, EMDP ethyl-5-methyl-3, 3diphenyl-l-pyrroline, HHMA 3,4-dihydroxymethamphetamine, MBDB 2-methylamino-1-(3,4-methylenedioxyphenyl) butaneamine, MRM multiple reaction monitoring, SPMEM solid-phase microextraction membrane, TFA trifluoroacetamide, THC-COO-gluc 11-nor-9-carboxy-Δ9-tetrahydrocannabinol glucuronide, UVD UV detection

Mixed analytes Plasma, [110] Morphine 0.2 mL 6-MAM Amphetamine Methamphetamine MDMA MDA MDEA MBDB Cocaine BZE [111] THC OH-THC THC-COOH Codiene Morphine 6-MAM M3G Normorphine Cocaine BZE (methadone)

[108]

Ecgonine methyl ester BZE Cocaine-N-oxide Cocaine Ecgonine methyl ester BZE Ecgonine

Anal Bioanal Chem (2007) 388:1415–1435 1425

1426

concluded that a negative cocaine result using a microtiterplate ELISA need not be confirmed and Schuetz et al. [6] found an “incredibly high reliability” of their cloned enzyme donor immunoassay (CEDIA) for BZE, it is still axiomatic that positive immunoassay results must be confirmed by a second independent method that is at least as sensitive as the screening test and that provides the highest level of confidence in the result. Nowadays, this is usually done by GC coupled with MS (GC-MS) or LC– (tandem) MS [LC-MS(/MS)]. If the prevalence of positive samples is high like, e.g., in clinical toxicology, specific chromatographic procedures should be used also for screening purposes [7–9]. Several papers on screening procedures for drugs of abuse in blood have been published over the time period covered by this review. Some of them are included in the 2005 review of Maurer [7] of multianalyte procedures for screening and quantification of drugs in blood, plasma or serum using LC coupled with a single-stage or a tandem mass spectrometer. Some of those procedures have also been included in the present review.

Cannabinoids Cannabis sativa L. (marijuana, hashish, hashish oil, etc.) continues to be the most widely produced and trafficked plant-based illicit drug worldwide [10] and is reported to be the illicit drug most frequently used in Europe and the USA [10, 11]. The primary psychoactive component of cannabis is Δ9-tetrahydrocannabinol (THC) [12]. THC is rapidly metabolized by cytochrome P-450 (CYP) enzymes to the equipotent psychoactive 11-hydroxy-Δ9-tetrahydrocannabinol (OH-THC) and further to the inactive 11-nor-9carboxy- Δ9-tetrahydrocannabinol (THC-COOH) [13]. These compounds are glucuronidated and mainly excreted into urine. Determination of OH-THC and THC-COOH besides the parent compound THC is recommended from a toxicological point of view, because the type of application and consumption behavior may be assessed by individual drug levels in blood or plasma/serum [14]. When quantitative analysis is conducted, it should be mentioned whether whole blood or plasma/serum had been used. The respective concentrations are not directly comparable, because THC is almost completely bound to plasma proteins (90%) and is very poorly distributed into red blood cells (10%). Giroud et al. [15] suggested an average plasma to whole blood concentration ratio of 1.6 for THC, OH-THC and THC-COOH concentrations (1.5 for THC), as determined in healthy volunteers with a relatively low coefficient of variation between individuals. The same plasma-to-blood ratio was found for 11-nor-9-carboxy-

Anal Bioanal Chem (2007) 388:1415–1435

Δ9-tetrahydrocannabinol glucuronide (THC-COO-gluc) by Skopp et al. [16] (blood-to-plasma ratio 0.62). However, the distribution in postmortem samples (peripheral femoral blood) between serum and whole blood concentrations was more variable, with the mean ratio being 2.4 (2.2 for THC) [15]. Hematocrit changes caused by blood hemolysis occurring after death, protein aggregation and degradation were given as possible explanations. As plasma or serum to whole blood ratios may be subject to changes, calculations of ratios with fix factors might be unreliable. This has to be considered especially in forensic toxicology, where degraded specimens are frequent. Care should be taken if concentrations of THC and metabolites determined in serum/plasma or whole blood are compared. Most publications dealt with the analysis of plasma or serum. Gustafson et al. [13] included a hydrolysis step by β-glucuronidase from Escherichia coli in their sample preparation to release the analyte from the glucuronide moiety for quantification of total cannabinoids. This enzyme was reported to be most effective one in cleaving ether-linked glucuronic acid of cannabinoid metabolites such as THC glucuronide and OH-THC glucuronide. Hydrolysis of the ester bond of THC-COOgluc was less affected by the source of glucuronidase. Both, liquid–liquid extraction (LLE) as well as solid-phase extraction (SPE) procedures were described for extraction of cannabinoids from biological matrices. In the case of whole blood, a protein precipitation step with, e.g., (cold) acetonitrile, should take place first. SPE on C18 cartridges for simultaneous determination of THC and THC-COOH seems to be best performed by dilution of the sample with an acidic buffer and/or by carrying out an acidic washing step with acetic acid or other weak or diluted acids (pH 4– 4.5), respectively [13, 15, 17, 18]. The low pH should convert THC-COOH into the nonionic form for the following elution with organic solvents such as ethyl acetate, hexane, acetonitrile, acetone and mixtures of them. In several reports, two separate elution steps for THC (basified) and THC-COOH (acidic) were performed [12, 19]. Automated SPE (e.g., by use of a RapidTrace workstation) was also reported [18, 20]. Yang and Xie [21] presented a method, using a solid-phase microextraction membrane coupled with LC-MS. However, the membrane (polyamide- and Tenax-based) was self-prepared and the method was approved only for qualitative analysis owing to limitations of the membrane size and membrane adsorption capacity. In addition, it was very time consuming (desorption over 6 h) and the procedure was only evaluated with mouse blood. In its current state, the method cannot be applied for forensic purposes. Methylation and silylation were chosen most often as derivatization method for GC-MS determination. GC-MS is still the most common analytical procedure for confirma-

Anal Bioanal Chem (2007) 388:1415–1435

tion and quantitative determination of cannabinoids in blood or plasma/serum with regard to forensic problems [2, 14, 15, 17–19]. Two interesting new technical GC modifications with respect to cannabinoid determination were reported within the time frame of this review. Teske et al. [17] presented a GC-MS confirmation method, based on large-volume programmed-temperature vaporization injection using only 25 μL of plasma or whole blood [17]. The reduction in sample volume was compensated by increasing the extract volume injected (20 μL) without a strong increase in background noise or a deterioration in detection limits. However, the method described recorded only two ions per analyte and, moreover, one mass track of THC (m/z 313) was disturbed by coelution of a matrix peak. Furthermore, the lower end of the calibration range was as high as 2 ng/mL for THC and the respective LOQ was not mentioned. Therefore, the method in its current form does not meet the requirements of reliable forensic toxicological analysis. Scurlock et al. [19] reduced matrix interferences by addition of a Deans switch to a standard GC oven which allowed the use of two serial chromatographic columns of differing stationary phases (2D-GC-MS). The Deans switch could be programmed to direct the output of the primary column to the secondary column for only a small window surrounding the retention time of each analyte peak. The separation of the analyte from coeluting compounds on the second column resulted in very clean chromatography regardless of the degree of degradation of the blood specimen. Besides the most widely used electron ionization (EI), two methods were described using chemical ionization (CI)–MS. Huang et al. [12] reported a negative ion CI (NICI)–MS assay for the quantitative determination of THC and THC-COOH in plasma with methane as reactant gas [12]. Gustafson et al. [13] used positive ion CI (PICI)–MS for the same purpose. NICI and PICI techniques may be employed to improve selectivity and sensitivity, resulting in very low detection limits. However, owing to lower fragmentation energy, the mass spectra contain a smaller number of prominent and characteristic ion peaks and often only a single ion mass. This results in increasing sensitivity, but adequate fragment ions as qualifiers are not provided. In both methods, only one ion could be monitored for each analyte, which might be a problem with regard to the requirements for reliable identification in forensic toxicology. Apart from the GC-MS technique, an increasing number of methods for the quantitation of THC and THCCOOH based on LC mainly combined with MS were published in the last few years. One advantage of LC-MS(/MS) is that, in contrast to GC-MS, no derivatization of the extracts is required. Abbara et al. [24] reported a method to quantify THC and/or THC-COOH in plasma by high-performance

1427

LC (HPLC)–UV analysis. A relatively time consuming sample preparation owing to a two-step extraction method, the low sensitivity (LOQ 5 ng/mL) and the use of the AIDS medicament lopinavir as an internal standard are clear drawbacks for forensic applications. Sensitive and reliable methods for the determination of cannabinoids in plasma using LC-MS/MS with positive electrospray ionization (ESI) in the multiple reaction monitoring (MRM) mode were reported [20, 22]. The LC-MS/MS methods offered the possibility to routinely determine THC-COOgluc [23]. Instabilities of cannabinoids and their metabolites in plasma or whole blood or in processed samples were often discussed. Several studies included stability tests in their validation procedures [2, 12–14, 23, 24]. Stability was given, if accuracy and precision were within the assay variability, at least within ±20% of the target concentrations. Serum and plasma samples stored at −20 °C did not show any considerable change in analyte concentrations for 1 month [13, 14] and for 2 months in borosilicate glass tubes [24]. Plasma (possibly containing 1% sodium fluoride) storage in polypropylene containers at −20 °C for 7 days also revealed no significant impact on quantitation of THC or THC-COOH [12]. Analyte concentrations were stable within the criteria in plasma after up to three freeze– thaw cycles (−20 °C/ambient temperature) [12, 13, 24], at 4 °C for 72 h [13] and at ambient temperature for 24 h [12, 24]. One study only revealed stability for 8 h, but not for 24 h at room temperature [13]. THC and THC-COOH were stable in whole blood at room temperature for 6 h [12]. Processed samples (derivatized extracts, according to the respective sample preparation) in autosampler vials were within acceptance criteria at −20 °C for up to 3 days [12], at 4 °C or at room temperature over a total period of 7 days [2], at room temperature for 72 h [13] or at 4 °C for 24 h [24]. Skopp et al. [23] investigated the stability of THC-COOgluc spiked into plasma and urine samples under different storage conditions [23]. THC-COOgluc is an ester-linked β-glucuronide and such acyl glucuronides were reported to be generally unstable. This is of importance as degradation of THC-COOgluc during storage or transport of the sample leads to the formation of THC-COOH and the concentration of unconjugated THC-COOH in plasma is used in forensic interpretations as a criterion to differentiate between infrequent and frequent drug consumption. THC-COOgluc was stable at −20 °C in plasma and urine. Concentrations remained within the limits of analytical precision at 4 °C for 7 days in plasma, but small amounts of THC-COOH formed under these conditions in vitro. At 20 °C, a marked change in concentration could be observed within 2 days of storage. Addition of fluoride/oxalate to plasma samples may decelerate hydrolysis of THC-COOgluc [25].

1428

Opiates/opioids Opioids (largely heroin) accounted for about 60% of all recorded illicit drug treatment requests in 2004 in many European countries and among these cases, 53% of users reported injecting the drug [10]. Moreover, available citybased estimates of drug-related mortality (overdose and other causes) suggest that currently 10–23% of overall mortality among adults aged 15–49 in Europe can be attributed to opioid use [10]. Toxicological analysis of opiates is of interest with respect to assessment of intoxications and culpability as well as to cause of death in cases of clinical, pathological or forensic importance and, finally, also to drug abuse control and human pharmacokinetic or pharmacodynamic studies [26, 27]. This review deals with the bioanalysis of the original opiates, morphine and codeine, and the most abused semisynthetical one, heroin, as well as with their principal metabolites. In humans, heroin is rapidly hydrolyzed to 6-monoacetylmorphine (6-MAM) and subsequently to morphine by esterases (especially the erythrocyte acetylcholinesterase) [28]. Morphine itself is also extensively metabolized, mainly by phase II glucuronidation at the 3-hydroxy and 6-hydroxy moieties [29]. The major metabolite of morphine is morphine-3-glucuronide (M3G) [30]. Plasma concentrations of M3G can exceed that of morphine [31]. The active metabolite morphine-6-glucuronide (M6G), an agonist at μopioid receptors, contributes to clinical central nervous opioid effects when it accumulates in the plasma of patients with renal failure. This should probably not happen after only short-term administration of morphine. But recently, M6G has been identified to additionally exert important peripheral opioid effects [30]. However, M3G is not an agonist at opioid receptors [30]. Nevertheless, quantitative determination of M3G may play a role in evaluation of survival time of heroin overdose victims as well as in pharmacokinetic assessment of multimorbid patients with renal dysfunction receiving morphine as medication. Acetylcodeine and codeine, common adulterants of street heroin, have been used as markers for illicit heroin use [32]. A variety of techniques have been used to quantify morphine and its metabolites [33]. Immunoassays offer simple handling but lack the specificity to distinguish opiates from their corresponding glucuronides. They should only be performed as screening procedures. Within the timeframe covered by this review, de Jong et al. [34] described a radioreceptor assay for the determination of morphine and M6G in serum without sample pretreatment. The assay was based on competitive inhibition of binding of the μ-opioid-selective radiolabeled ligand [3H]-DAMGO to the striatal opioid receptor by opioid ligands (e.g., M6G). The inactive M3G displayed a low affinity to the μ-opioid

Anal Bioanal Chem (2007) 388:1415–1435

receptor. The assay was sensitive (LOD 1.6 nM M6G equiv). However, the value of the assay for quantitative determination is limited as morphine and M6G could not be distinguished and, moreover, intraassay precision was not acceptable (above 35%). The suitability of the Neogen microtiter plate ELISA for opiates was evaluated for screening of postmortem blood [35]. No matrix effects were found for diluted whole blood in these assays. The optimal cut-off for the Neogen Opiates Group ELISA was found to be relatively high, lying between 20 and 50 ng/mL morphine equiv and the interassay precision was poor. Ethylmorphine, codeine and hydrocodone had substantially higher cross-reactivities than morphine. A competitive immunoassay for morphine in serum, using capillary zone electrophoresis (CZE) combined with laser-induced fluorescence (LIF) detection, was established by Mi et al. [36]. The linear range was 50–1,000 ng/mL with a detection limit of 40 ng/mL. Neither of the assays was sensitive enough for most forensic problems. Several workup procedures were applied for purification of biomatrices to overcome matrix interference in the course of determination of opiates. For analysis of whole blood, protein precipitation with acetonitrile or methanol was conducted by several authors [37–39]. Meatherall [38] conducted a four-step LLE, which might be too laborintensive and time-consuming for daily routine work. Immunoaffinity chromatography using immobilized specific antibodies against morphine and morphine glucuronides was applied by Hupka et al. [39] for purification and concentration of biological samples, especially of postmortem whole-blood specimens. Recent improvements in extraction techniques have focused on SPE [33]. Routinely, mixed-mode SPE columns with hydrophobic moieties and cation exchangers were used. Extraction was performed by performing an acidic wash step and elution with organic solvents containing a low amount of ammonium hydroxide [33, 35, 37]. An approach to increase throughput was the use of 96-well plates rather than individual SPE cartridges [33]. Following the same idea, Shou et al. [31] further reduced sample preparation time by extensive automatization. They developed a bioanalytical method using automated sample transferring, automated 96-well plate SPE on a liquid-handling workstation and LC-MS/MS for morphine, M6G and M3G. Bengtsson et al. [40] used column switching for purification and desalting of brain microdialysis and plasma samples. The sample was loaded on a HyPurity column and desalted for 30 s. Thereafter, the analytes were eluted onto a ZIC HILIC (hydrophilic interaction LC) column for separation. Emara et al. [27] determined morphine in human plasma by CZE and micellar electrokinetic capillary chromatography using UV absorption detection at 190 nm [27]. The sample could be applied directly into the capillary.

Anal Bioanal Chem (2007) 388:1415–1435

However, the selectivity (UV monitoring of a single wavelength) and sensitivity of this method do not appear to be sufficient for forensic purposes. Since 2002, several GC-MS methods have been described, mainly for the analysis of opiates in postmortem whole blood and using the EI mode [35, 37, 41–44]. One procedure using an ion-trap mass spectrometer was described [38]. Silylation was the most applied derivatization procedure. Special treatment was needed for the keto analogues of morphine and codeine [37]. The samples were derivatized with hydroxylamine prior to SPE to convert the C6 keto moieties to oxime derivatives preventing enolization and production of multiple analyte peaks during the subsequent silylation step. Meatherall [38] converted the ketone groups of hydrocodone, hydromorphone, oxycodone and oxymorphone to methoximes by adding methoxyamine and pyridine as a catalyst following propionylation of the hydroxyl moieties. In some cases, β-glucuronidase treatment could be included in the sample workup for determination of total morphine. Detection limits can be improved using NICI-MS, because the extent of fragmentation is reduced and the diagnostic fragment ion carries the main proportion of the total ion current. Two NICI procedures were described by Leis et al., an assay for morphine [45] and an assay for its unhydrolyzed main metabolites M3G and M6G [29]. The methods allowed highly sensitive, precise and accurate determination of the analytes. However, for the needs of forensic application, they appear not to be suitable, as only one ion m/z peak per analyte was prominent and could be recorded in the NICI mode. Analysis of morphine and its unhydrolyzed metabolites M3G and M6G without derivatization is possible by HPLC [33]. Traditionally, HPLC with UV, FL or electrochemical detection was used for quantification of morphine and conjugated metabolites [26, 39, 46–48]. Hupka et al. [39] reported a novel detection method using wavelengthresolved LIF coupled with HPLC for separation. Total fluorescence spectra were recorded by a CCD camera. Choosing isolated spectral ranges of interest adapted to the native fluorescence of the analytes and the matrix led to increased specifity and sensitivity, at least if highly specific immunoaffinity extraction was done. However, C8 solidphase extracts of spiked blood samples showed very intense fluorescence from the matrix. The new detection method has not been validated so far. LC coupled with MS (usually LC–MS/MS) has become the technique of choice for simultaneous determination of low concentrations of morphine, M3G and M6G, as it provides greater sensitivity compared with EI-GC-MS. Several methods have been described since 2002 [31, 33, 40, 49–51]. In some procedures, only one transition per analyte was monitored. Although MRM with one transition is more selective than

1429

corresponding single ion monitoring (SIM) methods with one or even two ions, current requirements for confirmatory analysis of drugs of abuse (especially in forensic toxicology) are not met. Shou et al. [50] further improved their LC-MS/MS determination in terms of high throughput. This resulted in an injection-to-injection cycle time of 48 s with a 30 s acquisition time (18-s autosampler overhead due to a rinsing step). However, chromatographic resolution of the three analytes morphine, M6G and M3G has to be achieved. Firstly, separation of the glucuronides is necessary, as they share the same precursor and product ions [40]. Secondly, the labile glucuronides could fragment to morphine in the LC-MS interface. If morphine and the metabolites are not chromatographically resolved, this will result in the overestimation of at least morphine concentrations [50]. Separation of morphine and M6G was apparently not done for the method described in [51]. To achieve separation within the 30-s acquisition time, silica (normal phase) columns operated under high flow rates of high organic/low aqueous mobile phases were used [50]. These conditions could be useful for bioanalysis of polar, ionic compounds such as the morphine glucuronides [50]. Chromatographic baseline resolution has been maintained under the ultrafast LC-MS/MS conditions. As shown in investigations on the freeze–thaw stability of morphine, M3G and M6G in plasma samples, significant sample degradation did not occur after three freeze–thaw cycles [29, 31, 33, 45]. No decomposition of standard and stock solutions was measurable after 2 months of storage at −20 °C [29, 45]. For the respective sample preparation methods, storage of processed samples at ambient temperature for up to 24 h prior to analysis appeared to have little effect on the quantitation [31, 33]. According to Rook et al. [32], hydrolysis of the unstable heroin could be reduced in plasma samples by adding serum esterase blocking fluoride in the plasma tubes, applying a low temperature and a low pH value.

Amphetamines and amphetamine-derived designer drugs According to the 2005 US National Survey on Drug Use and Health, the rates for past-month and past-year methamphetamine use did not change between 2004 and 2005, but the lifetime rate declined from 4.9 to 4.3% [11]. Among young adults across European countries, the prevalence of lifetime use of ecstasy is 5.2%, ranging from 0.5 to 14.6%, although rates of less than 3.6% are reported by half of the countries. The Czech Republic (14.6%), the UK (12.7%) and Spain (8.3%) report the highest prevalence rates [10]. Amphetamine [(R,S)-1-phenyl-2-propaneamine)] and methamphetamine [(R,S)-N-methyl-1-phenyl-2-propane-

1430

amine)] are powerful stimulants of the central nervous system by acting as indirect sympathomimetic drugs. They are drugs of abuse as well as doping agents. The S-(+) enantiomers of amphetamine and methamphetamine have 5 times more psychostimulant activity than the R-(-) enantiomers, resulting in the necessity of chiral methods for determination. The main metabolic pathways of amphetamine and methamphetamine have been well known for many years: aromatic hydroxylation, aliphatic hydroxylation, N-demethylation, oxidative deamination, N-oxidation and conjugation of the nitrogen. The phenolic metabolites are partly excreted as conjugates [52, 53]. The target analytes in blood are the parent compounds amphetamine and methamphetamine. N-alkylated or N,N-dialkylated amphetamine derivatives like amphetaminil, benzphetamine, clobenzorex, dimethylamphetamine, ethylamphetamine, famprofazone, fencamine, fenethylline, fenproporex, furfenorex, mefenorex, mesocarb, prenylamine or selegiline are therapeutically used as sympathomimetics, anorectics, non-opioid analgesics, antiparkinsonians or vasodilators. It is well known that these drugs are metabolically (bis)dealkylated to amphetamine or methamphetamine. Therefore, they are also of interest in forensic toxicology and doping control, where legitimate use of these therapeutics must be differentiated from illicit amphetamine use. In such cases, chiral analysis or inclusion of the precursors or specific metabolites may be necessary [53, 54]. The designer drugs 3,4-methylenedioxyamphetamine [MDA; (R,S)-1-(3′,4′-methylenedioxyphenyl)-2-propanamine, “love pills”], 3,4-methylenedioxymethamphetamine (MDMA; “Adam,” “ecstasy”) and N-ethyl-3,4-methylenedioxyamphetamine (MDEA; “Eve”) have gained great popularity as “rave drugs.” They produce feelings of euphoria and energy and a desire to socialize. Nichols coined the term “entactogens” for this new group of drugs. They may lead to more or less severe intoxications and impairment of ability to drive a car [53, 55]. The metabolism of amphetamine-derived designer drugs has been well studied in humans and rats. They undergo predominantly two overlapping metabolic pathways: demethylenation to dihydroxy derivatives (catechols) followed by methylation of one of the hydroxy groups, and successive degradation of the side chain to N-dealkyl and deamino oxo metabolites [53, 55]. Target analytes in blood usually are the parent compounds and for MDMA and MDEA also the N-dealkyl metabolite MDA. Determination of amphetamines can be done by gas-chromatographic procedures with mass-spectrometric detection and nowadays more and more by liquid-chromatographic procedures mostly with mass-spectrometric detection. Peters et al. [56] presented a validated GC-MS procedure for simultaneous screening and quantitation of amphet-

Anal Bioanal Chem (2007) 388:1415–1435

amines and amphetamine-derived designer drugs in plasma. The analytes were extracted using mixed-mode SPE, derivatized with heptafluorobutyric anhydride (HFBA) and determined by means of GC-MS in the SIM mode. Kankaanpää et al. [57] also used HFBA derivatization; however, they performed extraction and derivatization in one step by adding the extraction–derivatization reagent consisting of toluene, HFBA and deuterated internal standard to the alkalinized blood (or oral fluid or urine). After centrifugation, the toluene layer was used for GC-MS SIM analysis. Sample preparation was simpler than that in the method of Peters et al., but the LOQ were worse (25 vs. 5 ng/mL). Amphetamines can also be extracted using Extrelut columns. Nishida et al. [58, 59] used such columns also for on-column derivatization with propyl chloroformate. After derivatization, the propoxycarbonyl amphetamines were extracted from the columns with ethyl acetate. Using a deuterated internal standard and GC-MS detection in SIM mode, Nishida et al. [58, 59] set the LOQ for amphetamine and methamphetamine at 12.5 ng/mL. Off-line trichloroethyl chloroformate derivatization was also proposed [60]; however, this procedure did not use deuterated standards and was not validated. Traditional alkaline LLE procedures and acylation using perfluoroacylated acylation reagents followed by GC-MS detection are still in use and they work for routine analysis. Whenever evaporation steps with extraction solvents are necessary, care has to be taken that amphetamine is not coevaporated. Addition of hydrochloric acid to convert free amphetamine base into its hydrochloride may help [61–65]; however, too much hydrochloric acid may have deteriorating effects on the GC column. Because there exist many methods for amphetamine extraction without hydrochloric acid addition and the sensitivities are still good, such measures are not mandatory. With LC-MS machines becoming standard equipment in the forensic and clinical toxicology laboratory, the number of corresponding procedures dealing with determination of amphetamines in the literature has multiplied [63, 64, 66– 70]. MS detection allows the use of deuterated internal standards. They should be used whenever possible. Methods using ephedrine as an internal standard for determination of amphetamine-related substances such as that of Mortier et al. [64] cannot be accepted for routine use in the clinical or forensic laboratory. Ephedrine is a widely used medicament and it can never be excluded that the patient or defendant to be monitored has taken this drug. In addition, ephedrine is a side-chain hydroxylation product in the metabolism of methamphetamine and, finally, methamphetamine can be synthesized from ephedrine with some ephedrine remaining unchanged. In such cases, the peak area of the internal standard would be overestimated, leading to underestimation of the analyte concentration.

Anal Bioanal Chem (2007) 388:1415–1435

A simple precipitation step by adding methanol containing the deuterated internal standards can be sufficient for sample preparation. MS/MS offers superior selectivity and sensitivity even after simple LC separation on a C18 column. The MS should be performed in the MRM mode for quantitative procedures. Such a simple procedure was described by Wood et al. [70] for quantitation of amphetamine, methamphetamine, MDMA, MDEA, MDA and ephedrine in human plasma. Despite the simplicity of the procedure, the LOQ were in the range 0.5–2.0 ng/mL, which is absolutely sufficient for analysis of these drugs in clinical and forensic toxicology. Employing a LLE step, the LC–atmospheric pressure CI (APCI)–MS/MS procedure of Slawson et al. [63] achieved a limit of quantitation of 0.2 ng/mL for amphetamine and methamphetamine (and 0.1 ng/mL for selegiline and norselegiline). SPE procedures have also been used in order to get cleaner plasma extracts. Hendrickson et al. [68] extracted amphetamine and methamphetamine from rat serum using Oasis HLB columns. A deuterated internal standard and a standard C18 HPLC column were used. MS/MS was used in electrospray mode. Ion suppression studies were performed [68]. Such ion suppression (or enhancement) studies must be part of the validation of LC-MS(/MS) methods [71]. The next step in enhancing the LC-MS procedures may be the use of ultraperformance LC (UPLC), which provides significant gains in resolution and sensitivity. Apollonio et al. [66] demonstrated the power of UPLC for determination of amphetamine-type substances (and others); however, only introductory data were given. More (validation) work has to be done. HPLC procedures with detectors such as electrochemical detectors [69] or fluorescence detectors [72] are rarely used and will not be discussed in detail here. As stated above, the enantiomers of amphetamines and amphetamine-derived designer drugs have different pharmacological potencies, making it sometimes necessary to separate the enantiomers. Separation of enantiomers can be accomplished by using a chiral stationary phase or by forming diastereomers by derivatizing the enantiomers with a chiral reagent prior to their chromatographic separation. The diastereomers can then be separated using standard achiral stationary phases. In the case of capillary electrophoretic methods, chiral selectors can be used. Peters et al. used self-synthesized (S)-(-)-heptafluorobutyrylprolylchloride [(S)-HFBPCl] for converting amphetamine and methamphetamine [73] or MDMA, MDEA and MDA [74] into their diastereomers after a mixed-mode SPE. Separation could be done by achiral GC (HP-5MS). The derivatives were readily ionized in the highly sensitive NICI mode (SIM, three ions per analyte), because of the electronegativity of the heptafluorobutyryl moiety. The authors chose the (S)-HFBPCl derivatization under aqueous

1431

alkaline conditions, because of missing racemization and destruction of excess reagent, thus eliminating GC-MS interferences. Deuterated internal standards were used. Good validation data proved the suitability of the methods [73, 74]. The method was also suitable for simultaneous validated determination of amphetamine, methamphetamine, MDMA, MDEA and MDA in clinical toxicology and driving under the influence of drugs cases [75]. In order to avoid amphetamine losses during solvent evaporation, Leis et al. [76] added the (S)-HFBPCl reagent directly into the n-hexane extraction solvent after basic LLE. Also using GC-MS in the NICI mode, quantitation down to 0.006 ng/mL was possible [76]; however, only one ion could be monitored for each analyte, which might be a problem with regard to the requirements for reliable identification in forensic toxicology. If determination of the phenolic metabolites is also necessary, two-step derivatizations can be applied. Pizarro et al. [77, 78] used (R)-(-)-a-methoxy-a-trifluoromethylphenylacetyl chloride [(R)-MTPCl] for derivatizing the amine group of the analytes and to form diastereomers. The phenols were then covered by 1,1,1,3,3,3-hexamethyldisilazane derivatization. Electron-impact GC-MS in the SIM mode was used [77, 78]. Of course, sensitivities were not as good as those for the aforementioned NICI methods. GC-MS methods for amphetamine enantiomer determination after (R)-MTPCl or (S)-(-)-trifluoroacetyl prolyl chloride derivatization were also employed by Nyström et al. [65] and by Rasmussen et al. [79]. GC-MS methods work quite well for determination of amphetamine enantiomers. LC separation using a chiral protein phase (chiral-CBH) and fluorimetric detection may work, but such a procedure is quite exotic in forensic toxicology [80]. The capillary electrophoretic method of Rudaz et al. [81] is not suitable for use in the forensic laboratory as it has detection limits as high as 100– 400 ng/mL for the different amphetamines.

Cocaine Cocaine has become one of the most frequently abused drugs in many parts of the world. According to the 2005 National Survey on Drug Use and Health, an estimated 13.8% of Americans aged 12 years or older had tried cocaine at least once in their lifetimes. Approximately 5.5 million (2.3%) had used cocaine in the past year and 2.4 million (1.0%) had used cocaine within the past month [11]. Estimates of cocaine use in Europe (last-year prevalence) now place cocaine slightly ahead of amphetamine and ecstasy as Europe’s second most used illicit drug [10]. The major routes of cocaine administration are intranasal, intravenous and through smoking. Cocaine is rapidly metabolized to a multitude of metabolites. Consequently, at

1432

least the major metabolite benzoylecgonine (BZE) should be included in screening and quantitation procedures in plasma. BZE formation is catalyzed by a nonspecific liver carboxylesterase, which also catalyzes the transesterification of cocaine in the presence of ethanol, thus producing the pharmacologically active metabolite cocaethylene (benzoylecgonine ethyl ester) [82]. Enzymatic hydrolysis of cocaine by different plasma or liver esterases forms the pharmacologically inactive metabolite ecgonine methyl ester (EME, methyl ecgonine). There also are cytochrome P-450 (CYP) mediated metabolic steps in cocaine metabolism, resulting in the formation of norcocaine [83]. CYP 3A4 has been identified as the cocaine N-demethylase [84]. Subsequent N-hydroxylation via several CYP families leads to N-hydroxynorcocaine [85]. Further oxidative metabolites, such as nitroxide radical, nitrosonium, cocaine iminium and formaldehyde, are discussed as being responsible for hepatotoxic effects [86, 87]. Finally, hydroxylation in the meta or para position of the benzoyl moiety of cocaine and benzoylecgonine has been described. Of course, the meta- or para-hydroxylated cocaine can be demethylated to the corresponding hydroxylated m-hydroxybenzoylecgonine or p-hydroxybenzoylecgonine [88–90]. Heating cocaine base (crack) leads to formation of anhydroecgonine methyl ester (AEME, methylecgonidine); therefore, this pyrolysis product has been proposed as a marker to differentiate between smoking of cocaine and other routes of administration [91–94]. AEME is metabolized by esterases to anhydroecgonine (ecgonidine) [95], which seems to be a longer-persisting indicator of crack smoking than AEME [96]. In the presence of ethanol, transesterification of AEME to anhydroecgonine ethyl ester (ethyl ecgonidine) occurs [97]. Finally, N-demethylation of anhydroecgonine to the corresponding nor metabolite (noranhydroecgonine, norecgonidine) has been reported [97]. Cocaine and its pyrolysis product AEME can also be metabolized to thermolabile N-oxides [83, 98–100]. In summary, cocaine shows susceptibility to hydrolysis and pyrolysis, extensive metabolism and formation of thermolabile products, thus demanding special care during analysis at least if determination of several metabolites and/ or pyrolysis products is necessary. If determination of cocaine is required, inhibition of plasma esterases by, e.g., sodium fluoride, is necessary. For routine use in forensic laboratories, determination of cocaine and its major metabolite BZE may be sufficient. Interestingly, such simple procedures have hardly been published in the last 5 years. Fernandez et al. [101] included cocaine, BZE and cocaethylene in their HPLC–diode array detection (DAD) method for determination of opioids and cocaine in plasma. The analytes were extracted from plasma using a standard SPE procedure (Bond Elut Certify), separated using an RP8 column and measured by DAD with the wavelength

Anal Bioanal Chem (2007) 388:1415–1435

adjusted to the corresponding absorption maximum of each drug. Two years earlier the same group had shown that quantification of cocaine, BZE and EME can also be achieved by simple GC–flame ionization detection after SPE and silylation of the solid-phase extracts (Bond Elut Certify) [102]. However, there is an obvious trend towards more comprehensive methods for determination of cocaine. Besides cocaine, BZE and EME, simultaneous GC-MS determination of norbenzoylecgonine, norcocaine, ecgonine, m-hydroxybenzoylecgonine, AEME, cocaethylene, norcocaethylene and ecgonine ethyl ester in blood, urine and muscle was reported by Cardona et al. [103]. Blood or muscle proteins were precipitated with acetonitrile. Deuterated analogues of the analytes (as far as available) were used as internal standards. The different kinds of functional groups of the analytes could be derivatized in one step by using a mixture of pentafluoropropionic anhydride and pentafluoropropanol. Interestingly, analytical sensitivity could markedly be improved by conversion of the analytes into their hydrochloride salts by bubbling HCl vapor through the SPE eluates prior to derivatization. The advantage of such comprehensive methods is that concurrent ethanol intake can be detected as well as the route of administration (smoking crack vs. sniffing or intravenous injection). The analytical findings for cocaethylene, norcocaethylene or ecgonine ethyl ester are consistent with the concurrent use of cocaine and ethanol. If no further information is needed, simple GC-MS determination (SIM mode) of cocaine and cocaethylene after solid-phase microextraction (polydimethylsiloxane membrane) might be sufficient [104]. One should be aware of the fact that some of the cocaethylene is also hydrolyzed by esterases to BZE [105]. Unfortunately, interpretation of analytical findings of cocaine pyrolysis products in blood samples is not as unambiguous as that for the ethanol transesterification products at least when using GC-MS. It is well known that AEME is not only formed during crack smoking but also in the GC injection port. The amount of AEME increases with the amount of cocaine, leading to false positives in authentic samples [106]. Reducing the injector temperature to as little as 140 °C may prevent AEME formation [97], but may also result in loss of sensitivity for other analytes [103]. What may help in assessment of AEME findings is spiking a blank plasma with increasing amounts of cocaine and analyzing these samples using the same procedures as for the authentic samples. Only if AEME concentrations are greater than those formed pyrolytically from the spiked samples can smoking of cocaine be assumed. Such a method for the correction of quantitative values was proposed by Toennes et al. [106]. Cardona et al. [103] also used this approach and did actually find that the AEME concentrations measured in their authentic samples were

Anal Bioanal Chem (2007) 388:1415–1435

consistent with those produced solely by the GC inlet pyrolysis of cocaine controls in blood. These authors also stated that anhydroecgonine could not be used as a marker for the abuse of cocaine by smoking, because it was also produced pyrolytically from cocaine metabolites in the GC injector. The GC-MS procedure of Paul et al. [97] consisted of SPEs (C8 and benzene sulfonic acid) and several injections of eluates without derivatization or after combined pentafluoropropanol/pentafluoropropionic anhydride derivatization [97]. Using this complex procedure, Paul et al. [97] achieved determination of cocaine and 14 cocaine (ethanol-related) metabolites (cocaethylene, norcocaine, norcocaethylene, BZE, norbenzoylecgonine, m-hydroxybenzoylecgonine, p-hydroxybenzoylecgonine, EME, ethyl ecgonine, ecgonine) and pyrolysis products (AEME, anhydroecgonine ethyl ester, anhydroecgonine, noranhydroecgonine). The four major metabolites were benzoylecgonine, ecgonine, norbenzoylecgonine and EME, accounting for 88% (postmortem cases) and 97% (living volunteers) of all metabolites. The authors stated that at BZE concentrations below 100 ng/mL detection of ecgonine might be advantageous. GC-MS (using deuterated standards) after SPE (mixed-mode) and derivatization [N,O-bis(trimethyl)trifluoroacetamide with 1% trimethylchlorosilane and Nmethyl-N-(tert-butyldimethylsilyl)trifluoroacetamide with 1% tert-butyldimethylchlorosilane] was successfully applied for determination of cocaine and metabolites in plasma after controlled subcutaneous cocaine administration [107]. In this study, cocaine was detected within 5 min and 30–40 min after dosing. BZE and EME were first detected 5–15 min after dosing. Minor metabolites were detected much less frequently for up to 32 h. LC-MS has proven to be a valuable tool for elucidating oxidative metabolism of cocaine. Fandino et al. [98] used LC–electrospray multiple-stage MS and nanoelectrospray multiple-stage MS for identification of the N-oxide of AEME. Detection and time course of the N-oxide of cocaine in plasma were assessed by Lin et al. [99] using LC-ESI MS/MS and LC-APCI MS/MS. Less than 0.5% of cocaine N-oxide decomposed when LC-ESI MS/MS was used. As for the GC-MS methods, sample preparation has usually been done by SPE. Mass-spectrometric detection allowed the use of deuterated internal standards [99]. Duer et al. [108] examined the relationships between concentrations of cocaine and its hydrolysis products in peripheral blood, heart blood, vitreous humor and urine using LC-MS/ MS in the APCI mode [108]. They found a strong correlation between blood and vitreous humor concentrations and weak correlations between the urine and blood as well as between the urine and vitreous humor concentrations. They concluded that vitreous humor might be an important medium for toxicology testing.

1433

Perspectives Determination of drugs of abuse in blood is a growing issue in clinical and forensic toxicology. Technological progress in analytical toxicology has helped overcome most obstacles between the toxicologist and the reliable analysis of those drugs. Fortunately, validation of procedures has become a matter of course. A new trend is to develop procedures that allow for the simultaneous screening and confirmation of several groups of drugs. Such procedures can be based on GC-MS [109] or more often on LC-MS/ MS techniques [110, 111]. Replacement of immunoassay screenings by LC-MS/MS methods has already been proposed—at least for oral fluid testing [112]. A practical solution for the quantitative estimation of drug metabolites when reference standards are not available might be the use of ultra-low-flow nanospray for the normalization of conventional LC-MS data through equimolar response [113]. Finally, UPLC may become an alternative to traditional HPLC techniques [66].

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

Moeller MR, Kraemer T (2002) Ther Drug Monit 24:210–221 Chu MH, Drummer OH (2002) J Anal Toxicol 26:575–581 Kroener L, Musshoff F, Madea B (2003) J Anal Toxicol 27:205–212 Kupiec T, DeCicco L, Spiehler V, Sneed G, Kemp P (2002) J Anal Toxicol 26:513–518 Spiehler V, Isenschmid DS, Matthews P, Kemp P, Kupiec T (2003) J Anal Toxicol 27:587–591 Schutz H, Auch J, Erdmann F, Weiler G, Verhoff MA (2006) Arzneimittelforschung 56:414–420 Maurer HH (2005) Clin Biochem 38:310–318 Maurer HH (2004) Clin Chem Lab Med 42:1310–1324 Maurer HH, Peters FT (2005) Ther Drug Monit 27:686–688 European Monitoring Centre for Drugs and Drug Addiction (2006) Annual report 2006: the state of the drugs problem in Europe. Office for Official Publications of the European Communities, Luxembourg SAMHSA (2005) 2005 National survey on drug use & health: national findings. DHHS publication no SMA 06-4194 Huang W, Moody DE, Andrenyak DM, Smith EK, Foltz RL, Huestis MA, Newton JF (2001) J Anal Toxicol 25:531–537 Gustafson RA, Moolchan ET, Barnes A, Levine B, Huestis MA (2003) J Chromatogr B 798:145–154 Steinmeyer S, Bregel D, Warth S, Kraemer T, Moeller MR (2002) J Chromatogr B 772:239–248 Giroud C, Menetrey A, Augsburger M, Buclin T, Sanchez-Mazas P, Mangin P (2001) Forensic Sci Int 123:159–164 Skopp G, Potsch L, Mauden M, Richter B (2002) Forensic Sci Int 126:17–23 Teske J, Putzbach K, Engewald W, Muller RK (2002) J Chromatogr B 772:299–306 Nadulski T, Sporkert F, Schnelle M, Stadelmann AM, Roser P, Schefter T, Pragst F (2005) J Anal Toxicol 29:782–789 Scurlock RD, Ohlson GB, Worthen DK (2006) J Anal Toxicol 30:262–266 Maralikova B, Weinmann W (2004) J Mass Spectrom 39:526–531

1434 21. Yang R, Xie W (2006) Forensic Sci Int 162:135–139 22. Laloup M, Del Mar Ramirez FM, Wood M, De BG, Maes V, Samyn N (2006) Forensic Sci Int 161:175–179 23. Skopp G, Potsch L (2002) Clin Chem 48:301–306 24. Abbara C, Galy R, Benyamina A, Reynaud M, BonhommeFaivre L (2006) J Pharm Biomed Anal 41:1011–1016 25. Toennes SW, Kauert GF (2001) J Anal Toxicol 25:339–343 26. Ghazi-Khansari M, Zendehdel R, Pirali-Hamedani M, Amini M (2006) Clin Chim Acta 364:235–238 27. Emara S, Darwish I, Youssef D, Masujima T (2004) Biomed Chromatogr 18:21–27 28. Salmon AY, Goren Z, Avissar Y, Soreq H (1999) Clin Exp Pharmacol Physiol 26:596–600 29. Leis HJ, Fauler G, Raspotnig G, Windischhofer W (2002) J Mass Spectrom 37:395–400 30. Lotsch J (2005) J Pain Symptom Manage 29:S10–S24 31. Shou WZ, Pelzer M, Addison T, Jiang X, Naidong W (2002) J Pharm Biomed Anal 27:143–152 32. Rook EJ, Huitema AD, van den BW, Hillebrand MJ, van Ree JM, Beijnen JH (2006) J Anal Toxicol 30:390–394 33. Whittington D, Kharasch ED (2003) J Chromatogr B 796:95–103 34. de Jong LA, Kramer K, Kroeze MP, Bischoff R, Uges DR, Franke JP (2005) J Pharm Biomed Anal 39:964–971 35. Kemp P, Sneed G, Kupiec T, Spiehler V (2002) J Anal Toxicol 26:504–512 36. Mi JQ, Zhang XX, Chang WB (2004) J Immunoassay Immunochem 25:57–70 37. Ropero-Miller JD, Lambing MK, Winecker RE (2002) J Anal Toxicol 26:524–528 38. Meatherall R (2005) J Anal Toxicol 29:301–308 39. Hupka Y, Beike J, Roegener J, Brinkmann B, Blaschke G, Kohler H (2005) Int J Legal Med 119:121–128 40. Bengtsson J, Jansson B, Hammarlund-Udenaes M (2005) Rapid Commun Mass Spectrom 19:2116–2122 41. Wyman J, Bultman S (2004) J Anal Toxicol 28:260–263 42. Kerrigan S, Honey D, Baker G (2004) J Anal Toxicol 28:529–532 43. Crandall CS, Kerrigan S, Blau RL, Lavalley J, Zumwalt R, McKinney PE (2006) J Forensic Sci 51:413–420 44. Lewis RJ, Johnson RD, Hattrup RA (2005) J Chromatogr B 822:137–145 45. Leis HJ, Windischhofer W, Fauler G (2002) Rapid Commun Mass Spectrom 16:646–649 46. Netriova J, Blahova E, Johanesova Z, Brandsteterova E, Lehotay J, Serdt K, Mocak J (2006) Pharmazie 61:528–534 47. Bouwmeester NJ, Anderson BJ, Tibboel D, Holford NH (2004) Br J Anaesth 92:208–217 48. Peters JW, Anderson BJ, Simons SH, Uges DR, Tibboel D (2005) Intensive Care Med 31:257–263 49. Rook EJ, Hillebrand MJ, Rosing H, van Ree JM, Beijnen JH (2005) J Chromatogr B 824:213–221 50. Shou WZ, Chen YL, Eerkes A, Tang YQ, Magis L, Jiang X, Naidong W (2002) Rapid Commun Mass Spectrom 16:1613–1621 51. Musshoff F, Trafkowski J, Kuepper U, Madea B (2006) J Mass Spectrom 41:633–640 52. de la TR, Farre M, Navarro M, Pacifici R, Zuccaro P, Pichini S (2004) Clin Pharmacokinet 43:157–185 53. Kraemer T, Maurer HH (2002) Ther Drug Monit 24:277–289 54. Musshoff F (2000) Drug Metab Rev 32:15–44 55. de la TR, Farre M, Roset PN, Pizarro N, Abanades S, Segura M, Segura J, Cami J (2004) Ther Drug Monit 26:137–144 56. Peters FT, Schaefer S, Staack RF, Kraemer T, Maurer HH (2003) J Mass Spectrom 38:659–676 57. Kankaanpää A, Gunnar T, Ariniemi K, Lillsunde P, Mykkanen S, Seppala T (2004) J Chromatogr B 810:57–68

Anal Bioanal Chem (2007) 388:1415–1435 58. Nishida M, Namera A, Yashiki M, Kojima T (2003) J Chromatogr B 789:65–71 59. Nishida M, Namera A, Yashiki M, Kojima T (2002) Forensic Sci Int 125:156–162 60. Frison G, Tedeschi L, Favretto D, Reheman A, Ferrara SD (2005) Rapid Commun Mass Spectrom 19:919–927 61. Hidvegi E, Fabian P, Hideg Z, Somogyi G (2006) Forensic Sci Int 161:119–123 62. Kronstrand R, Ahlner J, Dizdar N, Larson G (2003) J Anal Toxicol 27:135–141 63. Slawson MH, Taccogno JL, Foltz RL, Moody DE (2002) J Anal Toxicol 26:430–437 64. Mortier KA, Dams R, Lambert WE, De Letter EA, Van CS, De Leenheer AP (2002) Rapid Commun Mass Spectrom 16:865–870 65. Nyström I, Trygg T, Woxler P, Ahlner J, Kronstrand R (2005) J Anal Toxicol 29:682–688 66. Apollonio LG, Pianca DJ, Whittall IR, Maher WA, Kyd JM (2006) J Chromatogr B 836:111–115 67. Dams R, De Letter EA, Mortier KA, Cordonnier JA, Lambert WE, Piette MH, Van CS, De Leenheer AP (2003) J Anal Toxicol 27:318–322 68. Hendrickson HP, Milesi-Halle A, Laurenzana EM, Owens SM (2004) J Chromatogr B 806:81–87 69. Segura M, Ortuno J, McLure JA, Pujadas M, Pizarro N, Farre M, Llebaria A, Joglar J, Segura J, de la TR (2002) J Chromatogr B 769:313–321 70. Wood M, De BG, Samyn N, Morris M, Cooper DP, Maes RA, De Bruijn EA (2003) J Anal Toxicol 27:78–87 71. Maurer HH (2005) Anal Bioanal Chem 381:110–118 72. Tomita M, Nakashima MN, Wada M, Nakashima K (2006) Biomed Chromatogr 20:1380–1385 73. Peters FT, Kraemer T, Maurer HH (2002) Clin Chem 48:1472–1485 74. Peters FT, Samyn N, Lamers CT, Riedel WJ, Kraemer T, De BG, Maurer HH (2005) Clin Chem 51:1811–1822 75. Peters FT, Samyn N, Wahl M, Kraemer T, De BG, Maurer HH (2003) J Anal Toxicol 27:552–559 76. Leis HJ, Rechberger GN, Fauler G, Windischhofer W (2003) Rapid Commun Mass Spectrom 17:569–575 77. Pizarro N, Llebaria A, Cano S, Joglar J, Farre M, Segura J, de la TR (2003) Rapid Commun Mass Spectrom 17:330–336 78. Pizarro N, Farre M, Pujadas M, Peiro AM, Roset PN, Joglar J, de la TR (2004) Drug Metab Dispos 32:1001–1007 79. Rasmussen LB, Olsen KH, Johansen SS (2006) J Chromatogr B 842:136–141 80. Buechler J, Schwab M, Mikus G, Fischer B, Hermle L, Marx C, Gron G, Spitzer M, Kovar KA (2003) J Chromatogr B 793:207–222 81. Rudaz S, Geiser L, Souverain S, Prat J, Veuthey JL (2005) Electrophoresis 26:3910–3920 82. Brzezinski MR, Abraham TL, Stone CL, Dean RA, Bosron WF (1994) Biochem Pharmacol 48:1747–1755 83. Kloss MW, Rosen GM, Rauckman EJ (1983) Mol Pharmacol 23:482–485 84. LeDuc BW, Sinclair PR, Shuster L, Sinclair JF, Evans JE, Greenblatt DJ (1993) Pharmacology 46:294–300 85. Pellinen P, Kulmala L, Konttila J, Auriola S, Pasanen M, Juvonen R (2000) Arch Toxicol 74:511–520 86. Kovacic P (2005) Med Hypotheses 64:350–356 87. Kovacic P, Cooksy AL (2005) Med Hypotheses 64:357–366 88. Cone EJ, Tsadik A, Oyler J, Darwin WD (1998) Ther Drug Monit 20:556–560 89. Watanabe K, Hida Y, Matsunaga T, Yamamoto I, Yoshimura H (1993) Biol Pharm Bull 16:1041–1043

Anal Bioanal Chem (2007) 388:1415–1435 90. Zhang JY, Foltz RL (1990) J Anal Toxicol 14:201–205 91. Jacob P III, Lewis ER, Elias-Baker BA, Jones RT (1990) J Anal Toxicol 14:353–357 92. Kintz P, Sengler C, Cirimele V, Mangin P (1997) Hum Exp Toxicol 16:123–127 93. Nakahara Y, Ishigami A (1991) J Anal Toxicol 15:105– 109 94. Toennes SW, Fandino AS, Kauert G (1999) J Chromatogr B 735:127–132 95. Fandino AS, Toennes SW, Kauert GF (2002) J Anal Toxicol 26:567–570 96. Paul BD, McWhorter LK, Smith ML (1999) J Mass Spectrom 34:651–660 97. Paul BD, Lalani S, Bosy T, Jacobs AJ, Huestis MA (2005) Biomed Chromatogr 19:677–688 98. Fandino AS, Karas M, Toennes SW, Kauert G (2002) J Mass Spectrom 37:525–532 99. Lin SN, Walsh SL, Moody DE, Foltz RL (2003) Anal Chem 75:4335–4340 100. Wang PP, Bartlett MG (1999) J Anal Toxicol 23:62–66 101. Fernandez P, Morales L, Vazquez C, Bermejo AM, Tabernero MJ (2006) Forensic Sci Int 161:31–35

1435 102. Fernandez P, Bujan L, Bermejo AM, Tabernero MJ (2004) J Appl Toxicol 24:283–287 103. Cardona PS, Chaturvedi AK, Soper JW, Canfield DV (2006) Forensic Sci Int 157:46–56 104. Alvarez I, Bermejo AM, Tabernero MJ, Fernandez P, Lopez P (2007) J Chromatogr B 845:90–94 105. Laizure SC, Mandrell T, Gades NM, Parker RB (2003) Drug Metab Dispos 31:16–20 106. Toennes SW, Fandino AS, Hesse FJ, Kauert GF (2003) J Chromatogr B 792:345–351 107. Kolbrich EA, Barnes AJ, Gorelick DA, Boyd SJ, Cone EJ, Huestis MA (2006) J Anal Toxicol 30:501–510 108. Duer WC, Spitz DJ, McFarland S (2006) J Forensic Sci 51:421–425 109. Gunnar T, Mykkanen S, Ariniemi K, Lillsunde P (2004) J Chromatogr B 806:205–219 110. Concheiro M, de CA, Quintela O, Lopez-Rivadulla M, Cruz A (2006) J Chromatogr B 832:81–89 111. Maralikova B, Weinmann W (2004) J Chromatogr B 811:21–30 112. Allen KR, Azad R, Field HP, Blake DK (2005) Ann Clin Biochem 42:277–284 113. Valaskovic GA, Utley L, Lee MS, Wu JT (2006) Rapid Commun Mass Spectrom 20:1087–1096