Analysis of forensically relevant drugs in blood and urine by conventional and advanced liquid chromatography-tandem mass spectrometry methods

Institut National de Criminalistique et de Criminologie Nationaal Instituut voor Criminalistiek en Criminologie ANALYSIS OF FORENSICALLY RELEVANT DRUGS IN BLOOD AND URINE BY CONVENTIONAL AND ADVANCED LIQUID CHROMATOGRAPHY-TANDEM MASS SPECTROMETRY METHODS Memoria presentada por María del Mar Ramírez Fernández Para optar al grado de Doctora por la Universidad de Santiago de Compostela Director: Manuel López-Rivadulla Codirector: Nele Samyn Santiago de Compostela, Septiembre 2009 This work was conducted at the National Institute of Criminalistics and Criminology, Federal Public Service Justice, Belgium. Corresponding address: María del Mar Ramírez Fernández NICC, Section Toxicology Chaussée de Vilvorde 100 1120 Brussels, Belgium Tel: +32 (0) 2 240 05 00 e-mail : mariadelmar.ramirezfer[email protected]gov.be Institut National de Criminalistique et de Criminologie Nationaal Instituut voor Criminalistiek en Criminologie D. Manuel López-Rivadulla Lamas, Catedrático de Toxicología de la Universidad de Santiago de Compostela, y Dña Nele Samyn, Jefe de Laboratorio del Departamento de Toxicología del Instituto Nacional de Criminalística y Criminología, Certifican: Que la presente memoria, titulada “Análisis de drogas relevantes desde el punto de vista forense, en sangre y orina, mediante métodos de cromatografía líquida acoplada a un doble espectrómetro de masas en tandem convencionales y avanzados”, que presenta la licenciada María del Mar Ramírez Fernández para optar al grado de Doctora ha sido realizada bajo su dirección en ambos departamentos y, a su juicio, reúne todos los requisitos exigidos por la normativa vigente para la elaboración y presentación de Tesis Doctorales. Y para que así conste, firma el presente informe en Santiago de Compostela, a 21 septiembre de 2009. M. López-Rivadulla N. Samyn No digas que fue un sueño... Acknowledgements A doctoral thesis is a combination of time, effort, motivation and training, and a way to share knowledge. First of all, thanks to the National Institute of Criminalistics and Criminology to give me the chance of being part of the Toxicology’s wonderful team. Thanks to Ms Leriche to accept me as a new member in 2004, and to Mr de Kinder to support me later on. Thanks to Gert De Boeck for trusting me during this time and to help me with the instruments problems. Thanks to Nele Samyn, for sharing with me her experience in chromatography and method validation for the analysis of drugs in conventional and alternatives matrices. Thanks to Marleen, Sarah, Malika, Bart Laeremans, Bart Viaene, Rhimou, Caroline, Evi, and Mathias who have helped me directly or indirectly in the lab and their good advices. Thanks to Vincent Areshka for his good lessons in chemistry at any time demanded. Thanks to Michelle Wood for her support in LC-MS/MS and her help and remarks when writing the articles. Mil gracias a Manolo Lopez-Rivadulla, por aceptar ser mi tutor en la distancia, y sobretodo por su gran amabilidad. Y gracias a su equipo, Ana, Marta, Oscar y Angelines, que tan bien me ha acogido cada vez que he ido a Santiago (¡sois fantásticos!). A mi familia, desde España, que aunque lejos en la distancia pero cerca en el corazón, ha aceptado y apoyado todas las decisiones que he tomado. Y a ti Vincent, compañero de mi vida, por apoyarme en los buenos y los momentos difíciles, por tu paciencia y tus ánimos. Por ti y para ti va esta tesis. ¡Gracias a todos! Table of contents Summary 1 Résumé 5 Resumen 9 I: INTRODUCTION: PHARMACOLOGY OF AMPHETAMINES, CANNABINOIDS AND HALLUCINOGENS 13 1.1. Introduction 15 1.2. Amphetamines 17 1.3. Cannabis 20 1.4. Hallucinogens and related drugs 23 1.4.1. Bufotenine 23 1.4.2. Cathinone 24 1.4.3. Chlorpheniramine 25 1.4.4. Kavain 26 1.4.5. Ketamine 27 1.4.6. Ibogaine 29 1.4.7. Lobeline 30 1.4.8. LSD 31 1.4.9. Mescaline 34 1.4.10. Psilocin 35 1.4.11. Ritalinic acid (methylphenidate) 36 1.4.12. Scopolamine 37 II: CONVENTIONAL LC-MS/MS METHODS 55 2.1. Introduction 57 2.2. History 58 2.3. Analytical background 59 2.3.1. Ionization 59 2.3.1.1. Electrospray ionization 59 2.3.2. Ion separation 64 2.3.2.1. Quadrupole 64 2.3.2.2. Tandem Mass Spectrometry 66 2.3.3. Ion detection 69 2.4. Characteristics of current LC-MS/MS applications in forensic Laboratories 70 2.4.1. Handling and robustness 70 2.4.2. Analytical limitations 73 2.4.3. Matrix effect 74 PPT: Protein PrecipiTation QC: Quality Control Qi: Quadrupole QqQ: Triple Quadrupole Quad: Quadrupole RAM: Restricted Access Materials RP: Reverse Phase RSD: Relative Standard Deviation SIL: Stable Isotope Labeled SIM: Selected Ion Monitoring S/N: Signal to Noise SPE: Solid Phase Extraction SPME: Solid-Phase Micro-Extraction SOFT: Society of Forensic Toxicologists SRM: Selected Reaction Monitoring SSI: Sonic Spray Ionization TDM: Therapeutic Drug Monitoring THC: Δ9-Tetrahydrocannabinol 11-OH-THC: 11HydroxyΔ9-TetraHydroCannabinol THC-COOH: 11-nor-9-CarboxyΔ9-TetraHydroCannabinol TIC: Total Ion Current TQ: Triple Quadrupole TOF: Time Of Flight TSI: ThermoSpray Ionization UPLC: Ultra-Performance Liquid Chromatography XLC: On-line SPE Summary 1 Summary The present work intended to evaluate the analysis of forensically relevant drugs in blood and urine by conventional and advanced LC-MS/MS methods. Chapter I gives a short review of the pharmacology of the drugs studied in this thesis (amphetamines, cannabis and hallucinogens) and their effects on human behavior and performance. Hallucinogens are psychoactive substances that powerfully alter perception, mood, and a host of cognitive processes. They are considered physiologically safe and do not produce dependence or addiction. Their origin predates written history, and they were employed by early cultures in a variety of sociocultural and ritual contexts. Nowadays, cannabis is still Europe’s most commonly consumed illicit drug as nearly a quarter of all Europeans have tried cannabis in their lifetime. Relaxation, well-being and somnolence are one of the effects of cannabis. Moreover, in parts of Europe, use of amphetamines constitutes an important part of the drug problem. They stimulate the CNS, increasing the heart rates and blood pressure and decreasing appetite, among other effects. LC-MS(/MS) is a widely used method in forensic laboratories, particularly in applications where non-volatile, labile or high-molecular-weight compounds are being analyzed. Chapter II aim to give an overview of the current LC-MS(/MS) applications in forensic laboratories and detailed information in relation to ionization, ion separation, and ion detection, together with tandem mass spectrometry. Suppression or enhancement of analyte ionization by coeluting compounds is a well known phenomenon in LC-MS(/MS) analysis mainly depending on the sample matrix, the sample preparation procedure, the quality of chromatographic separation, mobile phase additives, and ionization type. Therefore, common sample preparation procedures are protein PPT, LLE and SPE. Optimum sample preparation leads to enhanced selectivity and sensitivity. Detailed information about the recent applications of LC-MS(/MS) to the analysis of amphetamines, cannabis and hallucinogens in blood and urine are included at the end of this chapter. However, sample preparation is often regarded as time-consuming, a laborious work. Recent developments in on-line SPE aspects of high-throughput quantitative bioanalysis of drug and metabolite in biological matrices are described in Chapter III. High-throughput analysis is becoming increasingly important in forensic laboratories. One commercial automated on-line SPE system is the Symbiosis system manufactured by Spark Holland. In on-line SPE sorbents of very small particles are packed in a miniature LC column (cartridge) and higher pressures are Summary 2 applied to distribute the SPE solvents. Switching valves direct the flow to the LC column or waste, as appropriate. In contrast of traditional off-line SPE, several times consuming steps are eliminated. The extracted sample is directly injected to the analytical column by a simple valve switch. As a consequence no sample volume is lost during transfer, thus increasing the overall assay sensitivity. An extensive literature survey is given about the application of this instrument to the analysis of drugs in biological matrices. Chapter IV presents the objectives of the thesis which are an evaluation of the conventional LC-MS/MS technique and the new trend, on-line SPE-LC-MS/MS (Symbiosis), for the analysis of: a) multiple hallucinogens, chlorpheniramine, ketamine, ritalinic acid and metabolites in urine using off-line SPE, b) THC and metabolites in blood with LLE as off-line sample preparation procedure, c) THC-COOH (main metabolite of THC in urine) by on-line SPE, and d) 7 amphetamines and metabolites in blood and urine also by on-line SPE. Chapter V.I presents the development and validation of a LC-MS/MS method for the quantification of hallucinogens and other related compounds in urine. The method comprises an off-line SPE procedure, evaporation to dryness and reconstitution in mobile phase. The total run time was 20 min. External QCs containing LSD where analyzed within each series of analysis. The method was fully validated and applied to authentic urine samples (containing psilocin, ketamine, norketamine and chlorpheniramine). Chapter V.II describes the validation of the method for the analysis of THC and two of its main metabolites in blood. LLE with hexane: ethyl acetate was applied as clean-up procedure followed by centrifugation, complete evaporation, and reconstitution. The run time was 13 minutes. Two external QCs were used within each series of analysis. The method was completely validated in terms of precision, accuracy, specificity, recovery, matrix effects and stability. Finally the method was applied to authentic blood samples from forensic cases. Chapter V.III focuses on the development and validation of a method using the Symbiosis system for the analysis of THC-COOH in urine (500 µL). As the THC-COOH is glucoronized in urine, a previous hydrolysis was carried out using KOH 10 M. Then, the diluted urine was acidified in the LC vials for its direct injection. The method was fully validated and applied to authentic samples from cannabis users. Another application of the on-line-SPE-LC-MS/MS system is presented in Chapter V.IV for Summary 3 the direct quantification of 7 amphetamines and metabolites in blood and urine. The method, completely validated following international guidelines, required a minimum sample handling: the dilution of 100 µL blood or 50 µL of urine samples in the aqueous mobile phase, spiked with the IS, directly in the LC vials, previous injection. HLB cartridges were used for sample clean-up. The method was applied for the analysis of blood and urine samples from a MDMA study, and from forensic cases. The conclusions in Chapter VI show the advantages of LC-MS/MS and the new trend in on-line SPE. The results demonstrated that although conventional LCMS/MS methods are still robust and confident, on-line SPE coupled to LC-MS/MS is highly effective in terms of time safe and high throughput. Chapter VII is focused in the future perspectives in relation to LC-MS/MS applications. On-line SPE-LC-MS/MS and the UPLC-MS/MS are potent candidates for the analysis of drugs in conventional and alternative matrices (e.g. oral fluid), in such a way that the innovative chromatographic technologies are re-shaping the ways that separations and sample preparation are performed in high-throughput laboratories. Résumé 5 Résumé L’objet de cette thèse est d’évaluer l’analyse de drogues, pertinentes du point de vue médicolégal, dans le sang et l’urine par le biais des méthodes LC-MS/MS conventionnelles et avancés. Le Chapitre I présente un bref exposé de la pharmacologie des drogues traitées dans cette thèse (amphétamines, cannabis et hallucinogènes) et de leurs effets sur le comportement. Les hallucinogènes sont des substances psycoactives qui altèrent dans une large mesure la perception, l’humeur et la raison. Elles sont considérées physiologiquement sans risque parce qu’elles ne produisent pas de dépendance ou d’addiction. Leur origine remonte aux cultures ancestrales étaient utilisées dans les rituels sacrés et autres contextes socioculturels. Aujourd’hui le cannabis est la drogue illicite la plus consommée en Europe, de sorte que près d’un quart des européens l’ont goûté au moins une fois dans leur vie. Les effets du cannabis sont la relaxation, le bien-être et la somnolence. D’autres drogues problématiques en Europe sont les amphétamines. Elles stimulent le CNS en augmentant la pression artérielle et diminuant l’appétit, entre autres effets. LC-MS/MS est une technique largement utilisée dans les laboratoires médicolégaux, en particulier dans l’analyse de substances volatiles, thermiquement instables ou de poids moléculaire élevé. Le Chapitre II fourni un exposé général des applications récentes avec LC-MS(/MS) dans les laboratoires médico-légaux ainsi qu’une information détaillée sur l’ionisation, la séparation et la détection des ions, en plus de la spectrométrie de masses en tandem. Le phénomène connu comme suppression ou augmentation de l’ionisation des ions dans les analyses avec LCMS(/MS) dépend principalement de la nature de la matrice, du processus de préparation de l’échantillon, de la qualité de la séparation chromatographique, des additifs de la phase mobile et du type d’ionisation. Les procédures type de préparation de l’échantillon sont la PPT, la LLE, et la SPE. Une préparation optimale de l’échantillon entraîne une augmentation de la sensibilité et de la sélectivité. A la fin de ce chapitre, on trouvera également des information détaillées sur les applications récentes de LC-MS(/MS) pour l’analyse des amphétamines, cannabis et hallucinogènes dans le sang et l’urine. Cependant, la préparation de l’échantillon est considérée comme une étape fastidieuse et nécessitante du temps. Le Chapitre III expose les récents développements en SPE on-line à propos de productivité dans les analyses quantitatives des drogues dans les matrices biologiques. Ainsi, les analyses de Résumé 6 haute production sont en passe de devenir essentiels dans les analyses médicolégales. Un des systèmes commercialisés avec SPE on-line est le système Symbiosis produit par Spark Holland. Dans le système SPE on-line, le matériau d’extraction est emballé avec des particules très petites dans des colonnes LC (cartouches) qui travaillent à d’haute pression et auxquelles on applique les solvants d’extraction. Les valves d’échange dirigent le flux vers la colonne ou l’évacuent, si nécessaire. A la différence du SPE off-line classique, le système on-line permet d’éviter plusieurs étapes intermédiaires. Ensuite, l‘échantillon extrait est injectée directement dans la colonne analytique par le biais d’une simple valve d’échange. Par conséquent, il n’y a pas de perte de volume de l’échantillon et cela permet d’augmenter la sensitivité. Ce chapitre inclue également une révision bibliographique sur les applications actuelles de cet instrument pour l’analyse des drogues dans les fluides biologiques. Le Chapitre IV présente les objectifs de la thèse qui sont l’évaluation de la technique LC-MS/MS conventionnelle ainsi que la nouvelle tendance SPE-LCMS/MS on-line (Symbiosis), pour l’analyse de a) nombreux hallucinogènes, chlorpheniramine, ketamine, acide ritalinique et métabolites dans l’urine avec SPE off-line : b) THC et métabolites dans le sang au moyen de LLE off-line comme procédure de préparation de l’échantillon, c) THC-COOH (principal métabolite du THC dans l’urine) par vu de SPE on-line et d) 7 amphétamines et métabolites dans le sang et l’urine avec SPE on-line. Le Chapitre V.I. présente le développement et la validation d’une méthode avec LCMS/MS pour la quantification des hallucinogènes et d’autres composés analogues dans l’urine. La méthode consiste en une LLE off-line, une évaporation à sec et une reconstitution dans la phase mobile. La durée totale de l’analyse est de 20 min. Avec chaque série d’analyse nous avons analysé des QC externes qui contiennent le LSD. La méthode a été entièrement validée et appliquée à l’analyse d’urines authentiques (avec de la psilocin, ketamine, norketamine et chlorpheniramine). Le Chapitre V.II décrit la validation de la méthode pour l’analyse du THC et deux de ses métabolites dans le sang. Comme procédure de nettoyage de l’échantillon nous avons utilisé LLE avec hexane: acétate d’éthyle, ensuite une centrifugation, une évaporation complète (16 min) et une reconstitution. La durée de l’analyse a été de 13 min. Nous avons ajouté deux QC externes à chaque série d’analyse. La méthode Résumé 7 a été entièrement validée en termes de précision, d’exactitude, de spécificité, de récupération, d’effet matrice et de stabilité. Finalement la méthode a été appliquée aux analyses de sang de cas médico-légales. Le Chapitre V.III est centré sur le développement et la validation d’une méthode utilisant le système Symbiosis pour l’analyse de THC-COOH dans l’urine (500 µL). Etant donné le THC-COOH est glucuronidé dans l’urine, on a effectué une hydrolyse préalable avec KOH 10 M. Ensuite, l’urine diluée a été acidifiée directement dans les fioles LC en vue de son injection directe. La méthode a été entièrement validée et appliquée à l’analyse d’urine de consommateurs de cannabis. Une autre application du système SPE-LCMS on-line est décrite dans le Chapitre V.IV. pour l’analyse de 7 amphétamines et métabolites dans le sang et l’urine. La méthode, qui a été entièrement validée selon les normes internationales, a nécessité moindre manipulations: la dilution en phase aqueuse, contenant l’IS, de 100µL sang ou de 50 µL d’urine directement dans les fioles LC, avant l’injection. Pour le nettoyage de l’échantillon nous avons utilisé des cartouches HLB. La méthode a été appliquée à l’analyse de sang et urine dans une étude avec MDMA, et à de cas médico-légaux. Les conclusions du Chapitre VI mettent en évidence les avantages du LC-MS/MS et la nouvelle tendance SPE on-line. Les résultats montrent que même si les méthodes LC-MS/MS conventionnelles restent fiables, la SPE on-line couplée au LC-MS/MS est très efficace et permets de gains de temps et de productivité significatifs. Le Chapitre VII est centré sur les perspectives futures liées aux applications de la LC-MS/MS. Le progrès en SPE on-line et la séparation chromatographique (UPLC) sont les candidats potentiels pour l’analyse des drogues dans les matrices conventionnelles et alternatives (comme le fluide oral), de sorte qu’on est en train de métamorphoser la méthode dans laquelle la séparation et le traitement des échantillons appliqués dans les laboratoires de grande productivité. Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 15 1. 1. Introduction One of the biggest problems facing society today is the abuse and misuse of drugs. There are numerous drugs available to the community at large. Many offenders charged with violent crimes, or victims of violent crime, may have been under the influence of psychoactive drugs at the time a crime was committed. Therefore, it is important to understand the possible pharmacological effects associated with the use of these drugs. The principal drugs of concern can be divided into the categories shown in Table 1. Table 1. Principal classes of psychoactive substances (1) CNS depressants 1 Ethanol All alcoholic beverages (alcohol) 2 Benzodiazepines Alprazolam, diazepam, flunitrazepam, oxazepam, temazepam, etc 3 Opioids Codeine, heroin, methadone, morphine, oxycodone, pethidine, etc 4 Antipsychotics Chlorpromazine, clozapine, fluphenazine, haloperidol, olanzapine, etc 5 Antidepressants Amitriptyline, doxepin, dothiepin, fluoxetine, moclobemide, sertraline, etc 6 Marijuana Various forms of Cannabis sativa containing tetrahydrocannabinol 7 Barbiturates Amylobarbital, butobarbital, secobarbital, phenobarbital, thiopental, etc CNS stimulants 8 Amphetamines Speed' (methamphetamine), 'ecstasy' 9 Cocaine Free base 'crack' and hydrochloride 10 Other stimulants Ephedrine, pseudoephedrine, phentermine, fenfluramine, etc Other substances 11 Inhalants Petrol, solvents, propane (LPG), paint, butane lighter fluid, etc 12 Hallucinogens LSD, ecstasy, plant-derived substance such as mescaline, psilocybin, etc 13 Phencyclidine Usually abbreviated as PCP, and ketamine 14 Anabolic steroids Testosterone, stanozolol, etc Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 16 Cannabis in its various forms is arguably the second most commonly abused drug after alcohol. Smoked, eaten, imbibed –or just talked aboutit seems the world has a strong appetite for it. An estimated one in five European adults has tried cannabis, in such a way that globally, nearly 50 000 tones of cannabis herb or resin is produced each year. Little wonder, then, that cannabis has become a controversial cultural and commercial phenomenon (2). On the other hand, drugs such as MDMA (3,4-MethylenenDioxyMethAmphetamine), ketamine and LSD (Lysergic Acid Dietylamide), are typically used by teenagers and young adults at bars, clubs, concerts, and parties, and the use of these drugs is reported to help maintain energy levels for dancing or to enhance an altered state of consciousness (3). The use of these called ‘club drugs’ has increased significantly over the past 2 decades. One reason for their augmented use is the easy availability and low cost. In the following, the drugs studied in this thesis are described. Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 17 1.2. Amphetamines In many European countries the second most commonly used illegal substance is some form of synthetically produced drug. The use of amphetamines and related drugs among the general population is typically low, but prevalence rates among younger age groups are significantly higher, and in some social settings or cultural groups the use of these drugs may be particularly high. The amphetamines are central nervous system (CNS) stimulants, with similarities to some naturally occurring weak stimulants like ephedrine, a weak stimulant that occurs naturally in the branches of a number of an Ephedra species in concentrations up to 1.2% (4). The amphetamines are related in structure to the legal stimulants. Substitutions on the nitrogen and the ring system account for most of the structural variations and differences in stimulating and euphoric effects. Due to the chiral carbon atom adjacent to the nitrogen and to which a methyl group is attached, stereoisomerism has some serious implications in analytical chemistry and forensic toxicology (Figure 1). Figure 1. Structures of selected amphetamines O OHN CH 3 CH 3 MDMA O OH2N CH 3 MDA O OH3C N HCH 3 MDEA O NH 2 PMA OH CH 3 N H CH 3 Ephedrine NH 2 Amphetamine N H Methamphetamine Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 18 The abused amphetamine analogues are synthesized in clandestine laboratories by a variety of chemical processes from a number of precursor molecules i.e. from ephedrine or pseudoephedrine to d-methamphetamine. The presence of significant quantities of the starting product and impurities in the street samples allows for profiling of the source of drug. MDMA (‘ecstasy’) has become one of the most popular recreational drugs among young people visiting raves and mega house parties. Although the media often write about a very fast evolution in the choice of stimulants: MDEA (3,4-MethyleneDioxy-N-EhylAmphetamine) (‘Eve’), MDA (3,4MethylenenDioxyAmphetamine), MBDB (3,4-MethylBenzoDioxolylButanamine), PMA (4-Para-MethoxyAmphetamine), etc, our experience in the INCC in Belgium, is that amphetamine and MDMA are by far the most popular drugs, with other variants popping up only occasionally. Physical effects can include reduced appetite, increased/distorted sensations, hyperactivity, tachycardia, increased blood pressure, sweating, etc. Psychological effects can include anxiety and/or general nervousness, euphoria, creative of philosophical thinking, perception of increased energy, increased sense of well being, feeling of power or superiority, talkativeness, etc (1). The amphetamine-like stimulants have good oral and intranasal bioavailability and give maximum blood concentrations (0.1-0.4 µg/mL) within 1-3 h after a normal single dose. Proportionally higher concentrations are expected with higher doses, although non-linear kinetics of MDMA in humans has been described. The amphetamines are metabolized by similar pathways that involve a combination of hydroxylation of the ring and the side-chain carbon atom adjacent to the ring, and removal of the nitrogen. Drugs with alkyl groups on the nitrogen are dealkylated sometimes producing pharmacologically active metabolites: methamphetamine is metabolized to amphetamine, MDMA and MDEA are both metabolized to MDA. All metabolites with hydroxyl groups are excreted as conjugates (5,8). The half-life of amphetamines varies from 3-6 h to more than a day. Drugs with relatively long elimination half-lives will often show an accumulation of blood concentration with repeated dosing (chronic administration). Their clearance is particularly sensitive to the pH of urine. This is due to the basic nature of amphetamines, which are excreted rapidly if urine is acidic (increases ionization of drug), but only relatively slowly if urine is basic. Detection times in urine range up to 1-2 days following usual doses. In contrast to many other drugs of abuse, substantial Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 19 amounts of unchanged drug are excreted in urine. Amphetamine is excreted almost completely unchanged when urine is kept acidic, while basic urine will retard elimination and allow substantial metabolism to occur. Similar phenomena occur with other amphetamine analogues, including methamphetamine and ephedrine. (1,9-12). Thus, under normal conditions, methamphetamine is excreted largely as the parent drug (40%) following smoked and intravenous single doses, while only 7% of the dose is excreted as its main metabolite, amphetamine (13,14). Abstinence from amphetamine can still result in detection in urine for 2 days. However, highdose abusers can have urine positives for up to 9 days after last use (15,16). MDMA is metabolized to the demethylated active analogue, MDA. The MDA to MDMA urine concentrations are 0.15 or less following consumption of MDMA (17-19). Ephedrine is substantially excreted into urine unchanged (70% of dose). A single dose would be detectable in urine for 24-36 h (20). Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 20 1.3. Cannabis Cannabis is the collective term for the psychoactive substances of the Cannabis sativa plant and one of the most frequently used illicit drug in the Western World. Cannabis preparations include marijuana plant material, hashish (resin of female flowering tops) and hashish oil (extra from the resin). Cannabis is consumed by inhalation by either smoking or using a bong or a water pipe, or by ingestion of baked cannabis products (21). Cannabis products are mostly smoked in combination with tobacco. Thus, Δ9tetrahydrocannabinol (THC), the primary psychoactive analyte, is found in the plant’s flowering or fruity tops, leaves, and resin. There are over 421 different chemical compounds in cannabis, including 61 cannabinoids, chemical compounds containing 21 carbons atoms related to THC. During smoking more than 2000 compounds may be produced by pyrolysis; nitrogenous compounds, amino acids, hydrocarbons, sugars, terpenes, and simply fatty acids, are contributing to known pharmacological and toxicological properties of cannabis. The composition of cannabinoids varies depending on growing conditions, plant strain and the age of the sample. Smoking cannabis produces a much quicker absorption with maximum plasma concentrations occurring within a few minutes of smoking. The amount absorbed is also much higher than with ingestion; studies show a bioavailability of 14-50%, since THC is activated by a heating process through a decarboxylation reaction (22). The acute behavioral and physiological effects of cannabis have been well described (23-28): euphoria, relaxation, well-being, somnolence, changes in visual and auditory perception, altered perception of time and space, short-term memory loss, impaired learning, reduced performance, anxiety and panic reactions, disphoria, hallucinations, flashbacks, cardiovascular symptoms, red conjuctiva, small increase in pupil size, and reduced lacrimation There are three metabolites with significance: the 11-hydroxy (11-OH-THC), the 9carboxy (THC-COOH) and the glucuronide of THC-COOH (Figure 2). Other metabolites found in blood are the 8and 2-hydroxy-THC metabolites. Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 21 O OH HO THC Hydroxylation O OH CH 2OH O OH O OH COOH Oxydation 11-OH-THC THC-COOH O OH COOGlucoronid e THC-COOH-glucuronide O OH OH 2-OH-THC Glucuronidation Hydroxylation Hydroxylation 8-OH-THC Figure 2. Major metabolic route for THC The oral availability of cannabis is about 6%, which means that only 6% of the ingested THC is absorbed into the blood stream. The remainder is metabolized prior to entry into the blood stream, or is not absorbed. THC is measurable in plasma within seconds after inhalation of the first puff of marijuana smoke. Concentrations continue to increase rapidly and peak concentrations occur at approximately 9 min. Whole blood cannabinoid concentrations are approximately one half of the concentrations found in plasma due to the low partition coefficient of drug into erythrocytes (THC is 97-99 % protein bound in plasma). Once absorption has taken place, THC is rapidly distributed to tissues, concentrations being highest in adipose (fat) tissue due to its low water solubility and high affinity for fatty tissues. This distribution phase results in a rapid decline in blood plasma THC concentrations. THC concentrations greater than 10 ng/mL are uncommon after 1h even after moderate to high doses of cannabis (29-35). The terminal elimination half-life of THC is reported as 4-5 days, with a range of 3-13 days, although a shorter half-life of about 1 day has also been quoted. At first glance, this relatively long half-life contradicts the well-known short action of cannabis. However, this half-life is Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 22 calculated after the process of distribution to tissues has taken place and represents the terminal phase of the elimination curve. In high-dose (3.5% cigarette) users of marijuana, concentrations of THC can be detected for up to 24h after last use (>0.5 ng/mL) (32,36-38). The metabolite THC-COOH is also found in the blood of cannabis users (39). Because of difficulties in measuring THC accurately (because of the low concentrations), laboratories frequently measure also this inactive metabolite. Plasma concentrations of unconjugated THC-COOH following a 1.97% THC cannabis cigarette peak at 43 ng/mL at 20 min after smoking. By 6 h plasma concentrations fall to 13 ng/mL. 11-OH-THC is also produced from THC metabolism and also exhibits THC-like activity. Blood concentrations are rapidly detected following a 1.75% or 3.55% marijuana cigarette within 15-20 min. This metabolite is only detectable in blood for 12-24 h after normal-strength marijuana cigarettes (40,41). The main metabolite of THC in urine is THC-COOH. However, interpretation of results in urine is not simple. The amount of time that THC-COOH remains detectable in urine depends of the following factors: amount and frequency of use of cannabis, the metabolic rage, the body mass, the age, the overall health, the drug tolerance and the urine pH. Some researchers have used urinary THC-COOH concentrations in excess of 80 ng/mL to imply impairment. However, others have found little predictive value in using urinary THC-COOH concentrations, especially since 24% of users would have blood THC concentrations of less than 1 ng/mL, a concentration that does not normally cause impairment. Many persons with impaired THC concentrations, would not be impaired based on urinary THC-COOH concentrations (42,43). Gustafson et al (44) demonstrated that the terminal urinary elimination of THC-COOH following oral administration was approximately two to three days for doses ranging from 0.39 to 14.8 mg/d. Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 23 N H N HO 1.4. Hallucinogens and related compounds 1.4. 1. Bufotenine Bufotenine (Figure 3) is a hallucinogenic substance that exists naturally in some plants and in the cutaneous secretions of amphibians. (45-47). Although it is used as a therapeutic agent in China and other Asian countries, and as an aphrodisiac, ‘love stone’ in the West Indies, recent reports indicate its toxicity may carry a significant mortality rate (48,49). On the other hand, bufotenine has long been accepted as a naturally occurring component of human blood, brain and cerebral spinal fluid. However, while its biological presence at low concentrations is acknowledged, the biological function remains a mystery (50). Figure 3. Structure of bufotenine There is not so many information about the pharmacokinetics of bufotenine. Fuller et al (51) administered bufotenine subcutaneously in rats (1-100 µg/kg). It was distributed mainly to the lungs, heart and blood, and to much lesser extent, the brain and liver. It reached peak concentrations at 1 hour and it was nearly completely eliminated within 8 hours. In humans, bufotenine is rapidly absorbed following intravenous administration and it is excreted predominantly in the urine (70%). Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 24 O CH 3 NH 2 1.4.2. Cathinone Khat (Catha edulis Fosk) is an evergreen plant that grows at high altitudes in East Africa and Arabian Peninsula, being used as a stimulant plant for centuries. Chewing its fresh leaves is a widespread habit in local populations, with several million people consuming khat regularly in social sessions that often last for hours. Its current use among particular migrant communities in Europe and elsewhere has caused alarm among policy makers and health care professionals (52,53). A typical dose of 100 g fresh khat leaves contains on average 36 mg cathinone (Figure 4), with a higher content of cathinone in young leaves that are still growing (54). Users of khat report increased levels of energy, alertness and selfesteem, a sensation of elation, enhanced imaginative ability and a higher capacity to associate ideas (55-57). Figure 4. Structure of cathinone Chewing results in a high extraction of the alkaloids with only 9% remaining in the leaf residues. A chewing dose of 45 g khat leaves i.e. 0.6 g/kg of body weight results in a mean absoption dose of 45 mg of cathinone. The euphoric effects of khat start after about 1h of chewing of 60 g fresh khat leaves per subject. Blood levels of cathinone start to rise within 1h and peak plasma levels are obtained 1.5-3.5h after the onset of chewing (58). Maximum plasma levels range from 40 to 140 ng/mL after 1 h chewing. In plasma, cathinone is detectable up to 24h. The elimination half life is some 260 min. Only 2% of administered cathinone was found unchanged in the urine (53,59). Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 31 1.4.8. LSD Although drugs producing sensory distortions have been used by man for several millennia, many consider the modern era of psychedelics to have begun when the psychotropic effects of LSD were discovered by Albert Hofmann in 1943. This discovery ushered in an era of intense LSD research, with nearly 1000 articles appearing in the medical literature by 1961 (94). Most of this early research was based upon the drugs’ capacity to produce a “model psychosis” (95). By the mid-1960s, LSD and other related drugs had become associated with various counterculture movements, depicted as dangerous, and widely popularized as drugs of abuse. Accordingly, scientific interest in these drugs faded by the late 1960s, but human research with related psychedelics has recently experienced a slight renaissance (84,96-98). What the experience of the 1960s has pointed to many possible therapeutic and non-medical uses. Previous clinical experience, plus more recent informal use, has indicated other potential therapeutic uses for cluster headaches and addictions, among other conditions. The late Daniel X. Freedman made comments consistent with that assessment, stating, “one basic dimension of behavior…compel-compellingly revealed in LSD states is ‘portentousness’-the capacity of the mind to see more than it can explain, to believe in and be impressed with more than it can explicate, to believe in and be impressed with mire than it can rationally justify, to experience boundlessness and ‘boundaryless” events, from banal to the profound” (99). Although these descriptions focus on the more spectacular effects that these substances are capable of producing, low doses generally elicit less dramatic results. Typical clinical effects of LSD and related hallucinogens would include the following: • Somatic symptoms: dizziness, weakness, tremors, nausea, drowsiness, paresthesias, and blurred visions • Perceptual symptoms: altered shapes and colors, difficulty in focusing on objects, sharpened sense of hearing, and rarely synesthesias. • Psychic symptoms: alterations in mood (happy, sad, or irritable at varying times), tension, distorted time sense, difficulty in expressing thoughts, depersonalization, dreamlike feelings, and visual hallucinations (100). Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 32 LSD can induce disturbances of experience, otherwise observed only in psychoses, such as alteration of cognitive functions, and depersonalization. However, this drug does not appear to produce illness in emotionally healthy persons, but these problems seem to be precipitated in predisposed individuals. In atypical courses of intoxication, so-called “bad trips”, anxiety and excitement predominate. A bad trip is a disturbing experience whose manifestations can range from feelings of vague anxiety and alienation to profoundly disturbing states of unrelieved terror, ultimate entrapment, or cosmic annihilation. The potential causes can be a result of wrong set and settings (101,100). Figure 10. Structures of LSD and some key metabolites HN N CH 3 CON(C 2H5)2 LSD HN N CH 3 CON(C 2H5)2 OH 13-OH-LSD HN N CH 3 CON(C 2H5)2 OOH 2-Oxo-3-OH-LSD CON(C 2H5)2 N HN H N-Desmethyl-LSD N HN CH 3 CONH 2 Desethyl-LSD Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 33 LSD acts as an antagonist of peripheral 5-HT receptors, but acts as an agonist at CNS 5-HT receptors. Peak blood concentrations of LSD are typically less than 10 ng/mL. Maximum plasma concentrations following 70 µg dose range up to 2 ng/mL. The maximum blood levels occur around 30-90 min, but psychedelics effects of the drug occur within 5-10 min. The elimination half-life is estimated at 3-6h. Blood concentrations of drug are therefore quite low by 8h post-dose often less than 1 ng/mL. The duration of the action of LSD parallels approximately the blood concentrations, with ‘trips’ lasting from a few to several hours depending on the dose used (1). LSD is rapidly metabolized with only about 1-3% of an oral dose excreted in the urine as unchanged LSD. The major metabolites in urine are nor-LSD and 2-oxo-3OH-LSD (Figure 10). In addition, glucuronide conjugates of 2-oxo-3-OH-LSD are also present. Because of the low blood concentrations, LSD is most often measured in urine. Peak urine concentrations following oral ingestion of a typical street dose are normally less than 10 ng/mL and drop bellow 1 ng/mL within 12-24h. Therefore, extremely sensitive analytical methods are required to detect LSD use for more than 1 day after ingestion of the drug. 2-oxo-3-OH-LSD can be detected in urine up to 96h after administration, whereas LSD can only be detected for 12-24h postadministration (102,103). Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 34 OCH 3 H 3 CO H 3 CO NH 2 1.4.9. Mescaline Mescaline is one of the classic hallucinogens, which is known as the major alkaloid of the cactus peyote (Figure 11) (33,45). Peyote grows within a narrow strip of desert along the TexasMexico border. When properly cut, the crown (“button”) is removed from the plant leaving the root intact. Peyote is not smoked but eaten as peeled fresh buttons, dried whole buttons, dried/ground powder (sometimes also reconstituted in water), or is steeped/reconstituted into a warm tea. Peyote is most commonly consumed as a sacrament in the all-night ceremonies of the Native American Church (104,105). The effects of peyote and mescaline in humans are well studied. Native peyote cults used the cactus because it produces rich visual hallucinations. These psychoactive effects were used in psychiatric studies as a chemically induced model of mental illness (72). Figure 11. Mescaline Typical hallucinogenic doses range from 200 to 500 mg of mescaline with blood concentrations of 3.8 mg/L at 2 h and 1.5 mg/L at 7 h after ingestion (106). Following intake, mescaline is mainly excreted in the urine unchanged from (55-60%) (107). Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 35 Psilocybin O N H N CH 3 CH 3 P O HO HO N H N CH 3 CH 3 OH Psilocin 1.4.10. Psilocin In recent years, the recreational use of hallucinogenic mushrooms, so-called “magic mushroom”, has become an increasing social problem in several countries. They are not only naturally occurring but also offered as kits for cultivation, and contain the hallucinogenic indole derivatives, psilocin and psilocybin (Figure 12). Differences in the psilocin and psilocybin contents of the fruit bodies depends on the factors such as species, developmental stages, climatic conditions and the availability of soluble nitrogen and phosphorous in the soil (108). Thus, it takes about 30 fruitbodies of Psilocybe semilanceata to produce hallucinatory experience. Other large fungi which also contain psilocybin and psilocin are Psilocybe cubensis Panaeolus spp, Copelandia spp and Gymnopilus spp (109). There are several reports on the contents of psilocin and psilocybin in magic mushrooms (110-112) . Figure 12. Psilocin and psilocybin The rapid and extensive cleavage of the phosphoric ester group of psilocybin by alkaline phosphatase and unspecific esterases indicates that psilocybin acts as a prodrug and that its hydroxyl metabolite psilocin represents the true pharmacologically active agent (110,113). Hasler et al (114) compared the pharmacokinetic parameters in plasma and urine after an oral administration of psilocybin to 6 subjects. Within 24h about 0.9% of the applied dose of psilocybin was excreted as free psilocin. However, attention must be paid to the stability of psilocin. According to Tiscione et al (115), urine concentrations of psilocin decrease rapidly even kept in the refrigerator at 4°C. In conclusion, we may find psilocin in urine during 26-48 hours after administration. Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 36 OCH 3 O N H Methylphenidate O N H OH Ritalinic acid 1.4.11. Ritalinic acid (methylphenidate) Methylphenidate (Ritalin) is a phenethylamine derivative used in the treatment of depression, narcolepsy, attention-deficit disorder, and childhood hyperkinesis (116-119). Recently, methylphenidate has become a popular drug of abuse with the usual mode of administration being intravenous injection of dissolved tablets often in combination with the drug pentazocine. Pentazocine (Talwin) is a synthetic benzomorphan derivative that has properties of an analgesic and is about one third to one sixth as potent as morphine. The combination of Ritalin and Talwin is commonly referred to on the street as “poor man’s heroin’. The fact that methylphenidate is commonly prescribed in the treatment of attention-deficit disorder provides a likely source for drugs users. Figure 13. Ritalinic acid and methylphenidate Methylphenidate is reported to be absorbed quickly and completely from the gut after oral administration and it is rapidly hydrolyzed in the methyl ester linkage to its metabolite, ritalinic acid (Figure 13). Minor metabolic pathways for both these compounds include parahydroxylation of the aromatic ring, oxidation to 6-oxoderivates and glucoronide formation. Both methylphenidate and ritalinic acid are usually measured in plasma and urine (120,121). Following oral administration of methylphenidate peak plasma levels at 2 h. Because about 70% of methylphenidate is eliminated in the urine as ritalinic acid, it is obviously a better indicator (more prevalent) that the parent methylphenidate for detecting usage (118,122). Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 37 O OH O N CH 3 O 4.12. Scopolamine The Datura plants are members of the botanical family Solonaceae, which also contains common foods as tomatoes, potatoes, eggplants, peppers, and tobacco. This plant is originated from the tropical areas of Central and South America and it is now a cosmopolitan weed in temperate regions (104). The toxicity of Datura species is well known and has been linked deaths and poisonings for centuries (123). Traditional preparations include adding roots, leaves, or seeds to a fermented drink; drinking an infusion of the leaves or other parts; smoking the leaves; or chewing the fruit. When the plant material is taken orally, the effects last longer than when smoked and will also be more narcotic and hallucinogenic (124,125). The main toxic components are tropano alkaloids: hyoscyamine, which forms a diastereomeric mixture know as atropine and scopolamine (Figure 14). Scopolamine is an antimuscarinic agent (used as analgesic) and a smooth muscle relaxant. It is also an antispasmodic agent with antinauseant properties, and is extensively used in the treatment of motion sickness and in pre-operative medication (126,127). Weak infusions are used as hypnotics by the elderly and as aphrodisiac by adults. Datura species have also been used in criminal activities (128). Figure 14. Structure of scopolamine Up to now, scopolomine metabolism in man has not been verified stringently. An elucidation of the chemical structures of the metabolites extracted from human urine is still lacking. Huaixia et al (129) described the determination of scopolamine and its main metabolites in rat urine after ingesting 55 mg/kg scopolamine. Eighteen metabolites and the parent drug could be detected for up 106 h. In human, Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 38 pharmacokinetic parameters are dependent on the dosage form. According to Renner et al (130), scopolamine has a limited bioavailability if orally administered. The maximum drug concentrations occurs approximately 0.5 h after oral administration, and only 2.6% of nonmetabolized scopolamine is excreted in urine. Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 39 References 1. O. H. Drummer. The forensic pharmacology of drugs of abuse. Publishers. Arnold: (2001). 2. EMCDDA. www.emcdda.europa.eu(2009). 3. S. C. Koesters, P. D. Rogers and C. R. Rajasingham. MDMA ('ecstasy') and other 'club drugs'. The new epidemic. Pediatr. Clin. North Am. 49: 415433 (2002). 4. E. Marchei, M. Pellegrini, R. Pacifici, P. Zuccaro and S. Pichini. A rapid and simple procedure for the determination of ephedrine alkaloids in dietary supplements by gas chromatography-mass spectrometry. J. Pharm. Biomed. Anal. 41: 1633-1641 (2006). 5. K. Zaitsu, M. Katagi, H. T. Kamata, T. Kamata, N. Shima, A. Miki, H. Tsuchihashi and Y. Mori. Determination of the metabolites of the new designer drugs bk-MBDB and bk-MDEA in human urine. Forensic Sci. Int. 188: 131-139 (2009). 6. M. Rohanova and M. Balikova. Studies on distribution and metabolism of para-methoxymethamphetamine (PMMA) in rats after subcutaneous administration. Toxicology 259: 61-68 (2009). 7. M. Mueller, F. T. Peters, M. A. Huestis, G. A. Ricaurte and H. H. Maurer. Simultaneous liquid chromatographic-electrospray ionization mass spectrometric quantification of 3,4-methylenedioxymethamphetamine (MDMA, Ecstasy) and its metabolites 3,4-dihydroxymethamphetamine, 4-hydroxy-3-methoxymethamphetamine and 3,4methylenedioxyamphetamine in squirrel monkey and human plasma after acidic conjugate cleavage. Forensic Sci. Int. 184: 64-68 (2009). 8. A. H. Ewald and H. H. Maurer. 2,5-Dimethoxyamphetamine-derived designer drugs: studies on the identification of cytochrome P450 (CYP) isoenzymes involved in formation of their main metabolites and on their capability to inhibit CYP2D6. Toxicol. Lett. 183: 52-57 (2008). Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 40 9. M. R. Fuh, T. Y. Wu and T. Y. Lin. Determination of amphetamine and methamphetamine in urine by solid phase extraction and ion-pair liquid chromatography-electrospray-tandem mass spectrometry. Talanta 68: 987-991 (2006). 10. E. Gere-Paszti and J. Jakus. The effect of N-acetylcysteine on amphetaminemediated dopamine release in rat brain striatal slices by ion-pair reversed-phase high performance liquid chromatography. Biomed. Chromatogr. 23: 658-664 (2009). 11. T. Kumazawa, C. Hasegawa, X. P. Lee, K. Hara, H. Seno, O. Suzuki and K. Sato. Simultaneous determination of methamphetamine and amphetamine in human urine using pipette tip solid-phase extraction and gas chromatography-mass spectrometry. J. Pharm. Biomed. Anal. 44: 602-607 (2007). 12. C. Wang, G. Fan, M. Lin, Y. Chen, W. Zhao and Y. Wu. Development and validation of a liquid chromatography/tandem mass spectrometry assay for the simultaneous determination of D-amphetamine and diphenhydramine in beagle dog plasma and its application to a pharmacokinetic study. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 854: 48-56 (2007). 13. C. E. Cook, A. R. Jeffcoat, B. M. Sadler, J. M. Hill, R. D. Voyksner, D. E. Pugh, W. R. White and M. Perez-Reyes. Pharmacokinetics of oral methamphetamine and effects of repeated daily dosing in humans. Drug Metab Dispos. 20: 856-862 (1992). 14. R. J. Schepers, J. M. Oyler, R. E. Joseph, Jr., E. J. Cone, E. T. Moolchan and M. A. Huestis. Methamphetamine and amphetamine pharmacokinetics in oral fluid and plasma after controlled oral methamphetamine administration to human volunteers. Clin. Chem. 49: 121-132 (2003). 15. I. Kim, J. M. Oyler, E. T. Moolchan, E. J. Cone and M. A. Huestis. Urinary pharmacokinetics of methamphetamine and its metabolite, amphetamine following controlled oral administration to humans. Ther. Drug Monit. 26: 664-672 (2004). Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 47 68. A. R. Bilia, S. Gallon and F. F. Vincieri. Kava-kava and anxiety: growing knowledge about the efficacy and safety. Life Sci. 70: 2581-2597 (2002). 69. A. Capasso and L. Sorrentino. Pharmacological studies on the sedative and hypnotic effect of Kava kava and Passiflora extracts combination. Phytomedicine. 12: 39-45 (2005). 70. O. Meissner and H. Haberlein. HPLC analysis of flavokavins and kavapyrones from Piper methysticum Forst. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 826: 46-49 (2005). 71. J. Anke and I. Ramzan. Pharmacokinetic and pharmacodynamic drug interactions with Kava (Piper methysticum Forst. f.). J. Ethnopharmacol. 93: 153-160 (2004). 72. J. Beyer, O. H. Drummer and H. H. Maurer. Analysis of toxic alkaloids in body samples. Forensic Sci. Int. 185: 1-9 (2009). 73. P. H. Duffield and D. Jamieson. Development of tolerance to kava in mice. Clin. Exp. Pharmacol. Physiol 18: 571-578 (1991). 74. F. Tarbah, H. Mahler, B. Kardel, W. Weinmann, D. Hafner and T. Daldrup. Kinetics of kavain and its metabolites after oral application. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 789: 115-130 (2003). 75. M. Villain, V. Cirimele, A. Tracqui, F. X. Ricaut, B. Ludes and P. Kintz. Testing for kavain in human hair using gas chromatography-tandem mass spectrometry. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 798: 351-354 (2003). 76. L. Degenhardt and M. Dunn. The epidemiology of GHB and ketamine use in an Australian household survey. Int. J. Drug Policy 19: 311-316 (2008). 77. L. Muetzelfeldt, S. K. Kamboj, H. Rees, J. Taylor, C. J. Morgan and H. V. Curran. Journey through the K-hole: phenomenological aspects of ketamine use. Drug Alcohol Depend. 95: 219-229 (2008). Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 48 78. R. K. Siegel. Phencyclidine and ketamine intoxication: a study of four populations of recreational users. NIDA Res. Monogr119-147 (1978). 79. J. A. Hall and C. B. Moore. Drug facilitated sexual assault--a review. J. Forensic Leg. Med. 15: 291-297 (2008). 80. N. A. Anis, S. C. Berry, N. R. Burton and D. Lodge. The dissociative anaesthetics, ketamine and phencyclidine, selectively reduce excitation of central mammalian neurones by N-methyl-aspartate. Br. J. Pharmacol. 79: 565-575 (1983). 81. E. M. Kim, J. S. Lee, S. K. Choi, M. A. Lim and H. S. Chung. Analysis of ketamine and norketamine in urine by automatic solid-phase extraction (SPE) and positive ion chemical ionization-gas chromatography-mass spectrometry (PCI-GC-MS). Forensic Sci. Int. 174: 197-202 (2008). 82. S. L. Chou, M. H. Yang, Y. C. Ling and Y. S. Giang. Gas chromatographyisotope dilution mass spectrometry preceded by liquid-liquid extraction and chemical derivatization for the determination of ketamine and norketamine in urine. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 799: 37-50 (2004). 83. M. K. Huang, C. Liu, J. H. Li and S. D. Huang. Quantitative detection of ketamine, norketamine, and dehydronorketamine in urine using chemical derivatization followed by gas chromatography-mass spectrometry. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 820: 165-173 (2005). 84. K. R. Alper, H. S. Lotsof and C. D. Kaplan. The ibogaine medical subculture. J. Ethnopharmacol. 115: 9-24 (2008). 85. K. R. Alper. Ibogaine: a review. Alkaloids. Chem. Biol. 56: 1-38 (2001). 86. V. Kontrimaviciute, H. Breton, O. Mathieu, J. C. Mathieu-Daude and F. M. Bressolle. Liquid chromatography-electrospray mass spectrometry determination of ibogaine and noribogaine in human plasma and whole blood. Application to a poisoning involving Tabernanthe iboga root. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 843: 131-141 (2006). Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 49 87. V. Kontrimaviciute, O. Mathieu, J. C. Mathieu-Daude, P. Vainauskas, T. Casper, E. Baccino and F. M. Bressolle. Distribution of ibogaine and noribogaine in a man following a poisoning involving root bark of the Tabernanthe iboga shrub. J. Anal. Toxicol. 30: 434-440 (2006). 88. M. Cheze, A. Lenoan, M. Deveaux and G. Pepin. Determination of ibogaine and noribogaine in biological fluids and hair by LC-MS/MS after Tabernanthe iboga abuse Iboga alkaloids distribution in a drowning death case. Forensic Sci. Int. 176: 58-66 (2008). 89. L. P. Dwoskin and P. A. Crooks. A novel mechanism of action and potential use for lobeline as a treatment for psychostimulant abuse. Biochem. Pharmacol. 63: 89-98 (2002). 90. J. W. Sloan, W. R. Martin, M. Bostwick, R. Hook and E. Wala. The comparative binding characteristics of nicotinic ligands and their pharmacology. Pharmacol. Biochem. Behav. 30: 255-267 (1988). 91. J. E. Polston, C. S. Cunningham, K. R. Rodvelt and D. K. Miller. Lobeline augments and inhibits cocaine-induced hyperactivity in rats. Life Sci. 79: 981-990 (2006). 92. Y. Gaillard and G. Pepin. Poisoning by plant material: review of human cases and analytical determination of main toxins by high-performance liquid chromatography-(tandem) mass spectrometry. J. Chromatogr. B Biomed. Sci. Appl. 733: 181-229 (1999). 93. W. Song, Z. Peng, B. Ge, F. Han and Y. Chen. Identification of metabolites of lobeline in the rat urine by liquid chromatography-tandem mass spectrometry. Int. J. Mass. Spec. 269: 131-137 (2008). 94. E. Dyck. Flashback: psychiatric experimentation with LSD in historical perspective. Can. J. Psychiatry 50: 381-388 (2005). 95. E. Gouzoulis-Mayfrank, L. Hermle, B. Thelen and H. Sass. History, rationale and potential of human experimental hallucinogenic drug research in psychiatry. Pharmacopsychiatry 31 Suppl 2: 63-68 (1998). Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 50 96. F. Schifano, P. Deluca, A. Baldacchino, T. Peltoniemi, N. Scherbaum, M. Torrens, M. Farre, I. Flores, M. Rossi, D. Eastwood and others. Drugs on the web; the Psychonaut 2002 EU project. Prog. Neuropsychopharmacol. Biol. Psychiatry 30: 640-646 (2006). 97. P. Scartezzini and E. Speroni. Review on some plants of Indian traditional medicine with antioxidant activity. J. Ethnopharmacol. 71: 23-43 (2000). 98. G. I. Stafford, M. E. Pedersen, J. van Staden and A. K. Jager. Review on plants with CNS-effects used in traditional South African medicine against mental diseases. J. Ethnopharmacol. 119: 513-537 (2008). 99. D. X. Freedman. On the use and abuse of LSD. Arch. Gen. Psychiatry 18: 330-347 (1968). 100. D. E. Nichols. Hallucinogens. Pharmacol. Ther. 101: 131-181 (2004). 101. M. E. Attema-de Jonge, C. B. Portier and E. J. Franssen. [Automutilation after consumption of hallucinogenic mushrooms]. Ned. Tijdschr. Geneeskd. 151: 2869-2872 (2007). 102. D. Libong, S. Bouchonnet and I. Ricordel. A selective and sensitive method for quantitation of lysergic acid diethylamide (LSD) in whole blood by gas chromatography-ion trap tandem mass spectrometry. J. Anal. Toxicol. 27: 24-29 (2003). 103. J. H. Sklerov, J. Magluilo, Jr., K. K. Shannon and M. L. Smith. Liquid chromatography-electrospray ionization mass spectrometry for the detection of lysergide and a major metabolite, 2-oxo-3-hydroxy-LSD, in urine and blood. J. Anal. Toxicol. 24: 543-549 (2000). 104. J. H. Halpern. Hallucinogens and dissociative agents naturally growing in the United States. Pharmacol. Ther. 102: 131-138 (2004). 105. J. G. Bruhn, P. A. De Smet, H. R. El Seedi and O. Beck. Mescaline use for 5700 years. Lancet 359: 1866-(2002). Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 51 106. J. L. Henry, J. Epley and T. P. Rohrig. The analysis and distribution of mescaline in postmortem tissues. J. Anal. Toxicol. 27: 381-382 (2003). 107. K. Bjornstad, A. Helander and O. Beck. Development and clinical application of an LC-MS-MS method for mescaline in urine. J. Anal. Toxicol. 32: 227-231 (2008). 108. K. Tsujikawa, T. Kanamori, Y. Iwata, Y. Ohmae, R. Sugita, H. Inoue and T. Kishi. Morphological and chemical analysis of magic mushrooms in Japan. Forensic Sci. Int. 138: 85-90 (2003). 109. J. N. Edwards and J. A. Henry. Medical problems of mushrooms ingestion. Mycologist 3: 13-15 (1989). 110. F. Musshoff, B. Madea and J. Beike. Hallucinogenic mushrooms on the German market - simple instructions for examination and identification. Forensic Sci. Int. 113: 389-395 (2000). 111. J. Bigwood and M. W. Beug. Variation of psilocybin and psilocin levels with repeated flushes (harvests) of mature sporocarps of Psilocybe cubensis (Earle) Singer. J. Ethnopharmacol. 5: 287-291 (1982). 112. S. M. Sottolano and I. S. Lurie. The quantitation of psilocybin in hallucinogenic mushrooms using high performance liquid chromatography. J. Forensic Sci. 28: 929-935 (1983). 113. A. F. Grieshaber, K. A. Moore and B. Levine. The detection of psilocin in human urine. J. Forensic Sci. 46: 627-630 (2001). 114. F. Hasler, D. Bourquin, R. Brenneisen and F. X. Vollenweider. Renal excretion profiles of psilocin following oral administration of psilocybin: a controlled study in man. J. Pharm. Biomed. Anal. 30: 331-339 (2002). 115. N. B. Tiscione and M. I. Miller. Psilocin identified in a DUID investigation. J. Anal. Toxicol. 30: 342-345 (2006). Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 52 116. C. C. Huang, I. S. Shiah, H. K. Chen, W. C. Mao and Y. W. Yeh. Adjunctive use of methylphenidate in the treatment of psychotic unipolar depression. Clin. Neuropharmacol. 31: 245-247 (2008). 117. G. E. Francisco and C. B. Ivanhoe. Successful treatment of post-traumatic narcolepsy with methylphenidate: a case report. Am. J. Phys. Med. Rehabil. 75: 63-65 (1996). 118. B. L. Hungund, J. M. Perel, M. J. Hurwic, J. Sverd and B. G. Winsberg. Pharmacokinetics of methylphenidate in hyperkinetic children. Br. J. Clin. Pharmacol. 8: 571-576 (1979). 119. A. Balcioglu, J. Q. Ren, D. McCarthy, T. J. Spencer, J. Biederman and P. G. Bhide. Plasma and brain concentrations of oral therapeutic doses of methylphenidate and their impact on brain monoamine content in mice. Neuropharmacology(2009). 120. E. Marchei, J. A. Munoz, O. Garcia-Algar, M. Pellegrini, O. Vall, P. Zuccaro and S. Pichini. Development and validation of a liquid chromatographymass spectrometry assay for hair analysis of methylphenidate. Forensic Sci. Int. 176: 42-46 (2008). 121. E. Marchei, M. Farre, M. Pellegrini, S. Rossi, O. Garcia-Algar, O. Vall and S. Pichini. Liquid chromatography-electrospray ionization mass spectrometry determination of methylphenidate and ritalinic acid in conventional and non-conventional biological matrices. J. Pharm. Biomed. Anal. 49: 434-439 (2009). 122. J. Eichhorst, M. Etter, J. Lepage and D. C. Lehotay. Urinary screening for methylphenidate (Ritalin) abuse: a comparison of liquid chromatography-tandem mass spectrometry, gas chromatographymass spectrometry, and immunoassay methods. Clin. Biochem. 37: 175-183 (2004). 123. T. Vougiouklakis, V. A. Boumba and A. Mitselou. Fatal poisoning in the region of Epirus, Greece, during the period 1998-2004. J. Clin. Forensic Med. 13: 321-325 (2006). Introduction: Pharmacology of amphetamines, cannabinoids and hallucinogens 53 124. P. Kintz, M. Villain, Y. Barguil, J. Y. Charlot and V. Cirimele. Testing for atropine and scopolamine in hair by LC-MS-MS after Datura inoxia abuse. J. Anal. Toxicol. 30: 454-457 (2006). 125. L. Kursinszki, H. Hank, I. Laszlo and E. Szoke. Simultaneous analysis of hyoscyamine, scopolamine, 6beta-hydroxyhyoscyamine and apoatropine in Solanaceous hairy roots by reversed-phase highperformance liquid chromatography. J. Chromatogr. A 1091: 32-39 (2005). 126. J. I. Einarsson, B. O. Audbergsson and A. Thorsteinsson. Scopolamine for prevention of postoperative nausea in gynecologic laparoscopy, a randomized trial. J. Minim. Invasive. Gynecol. 15: 26-31 (2008). 127. N. Sah, V. Ramesh, B. Kaul, P. Dalby, K. Shestak and M. C. Vallejo. Transdermal scopolamine patch in addition to ondansetron for postoperative nausea and vomiting prophylaxis in patients undergoing ambulatory cosmetic surgery. J. Clin. Anesth. 21: 249-252 (2009). 128. P. A. Steenkamp, N. M. Harding, F. R. van Heerden and B. E. van Wyk. Fatal Datura poisoning: identification of atropine and scopolamine by high performance liquid chromatography/photodiode array/mass spectrometry. Forensic Sci. Int. 145: 31-39 (2004). 129. H. Chen, Y. Chen, H. Wang, P. Du, F. Han and H. Zhang. Analysis of scopolamine and its eighteen metabolites in rat urine by liquid chromatography-tandem mass spectrometry. Talanta 67: 984-991 (2005). 130. U. D. Renner, R. Oertel and W. Kirch. Pharmacokinetics and pharmacodynamics in clinical use of scopolamine. Ther. Drug Monit. 27: 655-665 (2005). II. CONVENTIONAL LC-MS/MS METHODS Conventional LC-MS/MS methods 63 800 150 000 ESI LC-MS (/MS) APLI LC-MS (/MS) APPI LC-MS (/MS) APCI LC-MS (/MS) GC-MS (/MS) Analyte polarity Molecular weight that the increase in the signal to noise (S/N) ratio due to the increase in the organic modifiers is not counterbalanced by an increase in flow rate (25). Alternatively, normal-phase chromatography with an aqueous/organic mobile phase enables the retention of polar species with a higher percentage of organic modifier, resulting in higher ionization efficiency when compared with RP separation (26). Figure 16 shows the range of application of LC-ESI-MS(/MS) in terms of analyte polarity and molecular weight compared with GC-MS analysis. Figure 16. Range of application of different hyphenated chromatographic and MS techniques. Conventional LC-MS/MS methods 64 2.3.2. Ion separation The mass analyzer represents the heart of a mass spectrometer, i.e. the device able to measure the m/z ratios of gas-phase ions. In order to allow a free path of the ions through the analyzer towards the detector, the analyzer must be operated under vacuum. The lower the pressure (typically in the range 10-4-10-7 torr), the longer the mean free path of the gas-phase ions. This has an obvious positive effect on sensitivity and mass resolution, although these parameters ultimately also depend on the type of mass analyzer and on the instrumental design adopted by the manufacturer. Together with sensitivity, other parameters are important in order to decide whether a particular mass analyzer fulfils the requirements of a laboratory. These are mass range, scan speed, mass resolution and mass accuracy. 2.3.2.1. Quadrupole The quadrupole is the most common mass analyzer for bench-top MS instruments. The most likely reason for this resides in the fact that quadrupoles offer a good compromise of mass range covered, reproducibility of mass spectra, mass resolution and precision for quantification purposes. A quadrupole consists of four parallel rods or poles equally spaced around a central axis. An electrical potential is applied to each rod so that each two adjacent poles have opposite polarities. By applying a precisely controlled combination of two electrostatic fields-one direct current (DC) and one at varying radiofrequency (RF)- a resonance frequency for a specific m/z ratio is obtained so that ions at that m/z ratio can reach the detector, whereas ions with lower and higher m/z rations are discarded (Figure 18). A quadrupole acts as a continuous mass filter, which means that most of the ions that are continuously transferred to the quadrupole from the ion source are lost on their way to the detector and only a few ions (the ions at resonant frequency) are translated into measurable electric signal (Figure 17). By varying the resonant frequency of the quadrupole, a complete set of masses can be scanned at speeds of up to 4000-5000 amu/s, thus obtaining a ‘full-scan’ mass spectrum. Conventional LC-MS/MS methods 65 In addition to the above scan mode (typically adopted for the detection and identification of unknown compound), a quadrupole can also be operated in selected-ion monitoring (SIM) mode. In this mode, the quadrupole, is set at one or a few resonant frequencies (depending on the number of ions to be monitored). As a result, both sensitivity and precision in quantification are improved. In fact, since in SIM mode the analyzer spends more time on a specific ion, the amount of signal belonging to this ion reaching the detector is larger than in scan mode. The increase in the S/N ratio is therefore due to a true increase in signal and should not be confused with the increase in the S/N ratio due to the elimination of the huge amount of chemical noise resulting from not monitoring the full mass range. A typical quadrupole covers a mass range of up to 1000-4000 amu at low resolution (0.7 amu FWHM) and with mass accuracy of 0.1 amu. Figure 17. A schematic quadrupole analyser Conventional LC-MS/MS methods 66 2.3.2.2. Tandem Mass Spectrometry Traditionally, MS/MS has been implemented in quadrupole instruments where two different quadrupoles, one performing mass selection of precursor (parent) ion(s) and the other mass selection of product (daughter) ion(s), are separated by a collision cell (Figure 18). Here, the precursor ion is accelerated and fragmented into product ions by collision with an inert gas. As the collision cell is also a quadrupole, although simply acting as an ion guide, the two quadrupoles mass anaysers are usually referred to as Q1 and Q3, Q2 being the collision cell, and the instrument is referred to as ‘triple quadrupole’ (QqQ). Fragmentation can be favored by increasing either the kinetic energy (collision energy) of the precursor or the gas pressure in the collision cell. A triple-quadrupole instrument can be operated in a number of different ways, the simpler one being to resemble a single-quadrupole instrument by using either Q1 or Q3 as a pass-all filter. One of the most typical MS/MS scanning modes is the socalled product-ion scan, where one precursor ion is selected by Q1 and the fragment (product) ions obtained after collision are scanned by Q3, thus obtaining a production spectrum. This differs from an in-source CID spectrum in that all fragments detected necessarily originate from the precursor ion. Other scanning modes, which are very helpful in metabolite profiling studies, are precursor-ion scan and neutral-loss scan (27-29). Both these scan modes are unique to mass spectrometers that provide MS/MS in space. Precursor-ion scan allows one to screen for homologous or similar compounds having a common fragment ion (e.g. different metabolites of a drug): Q1 scans the mass range, whereas Q3 filters a unique product ion and associates it back to the precursor ion that it originated from. The latter is useful for screening compounds with a common neutral fragment loss (e.g. loss of glucoronic acid). In this operating mode, both Q1 and Q3 are synchronously scanned with a fixed mass difference (e.g. 176), and a mass is assigned only to those precursor ions that have lost the predetermined neutral mass. MS/MS quantification is typically carried out in non-scanning mode by selecting one precursor ion in Q1 and one specific product ion in Q3. This operating mode, a reaction (precursor ion-product ion) instead of an ion is monitored (so-called selected-reaction monitoring [SRM] or, if different reactions are monitored, multiple- Conventional LC-MS/MS methods 67 MS1 MS2 Collision Cell reaction monitoring [MRM] (Figure 19). In this configuration, an MS/MS instrument associates extreme selectivity of detection (the probability that two compounds will share the same precursor AND the same product ion being quite low) with the highest sensitivity. In fact, the addition if a further MS stage implies that a lower number of analyte ions will reach the ion detector. However, as the noise is largely reduced, the resulting S/N ratio will be enhanced. Figure 18. Multiple Reaction Monitoring When different reactions are monitored consecutively, as in MRM, care should be taken in order to prevent unwanted contributions to a monitored signal (i.e. the product ion of one reaction) from other reactions monitored in the same acquisition cycle. This phenomenon, called cross-talk, occurs if a precursor ion enters the collision cell when product ions from the previous reaction monitored are still present (30). If the two precursor ions share common fragments, the signal pertaining to one precursor may be overestimated due to the contribution from the other. Different strategies allow one to control this problem, such as: -adding a pause time between two consecutive transitions (though this will obviously increase the acquisition cycle and will reduce the number of data points per unit time in the chromatogram); -avoiding the consecutive monitoring of two compounds sharing common product ion (i.e. by monitoring a third reaction in between); or, even better Conventional LC-MS/MS methods 68 -optimising LC separation in order to avoid chromatographic overlapping between the cross-talking compounds (31). Recently designed instruments take care of this problem by speeding up the exit of fragments from the collision cell. Cross-talk should always be considered when developing an analytical method as it is obviously a source of error in quantification. QTOF (Quadrupole/Time Of Flight), although rather expensive, is also an interesting configuration for MS/MS analysis as it combines the high efficiency of selected ion monitoring in the first mass analyzer (Q) with the high scan rate and high resolving power of the second (TOF). Conventional LC-MS/MS methods 69 2.3.3. Ion detection The detector is the device where ions separated by the mass analyzer are converted into a measurable electric signal (current). The analogue current is the further converted to a digital signal (counts) stored by the data system. The most important characteristics for an ion-detection device are speed, dynamic range and sensitivity (gain). The most common detection device is the electron multiplier. In an electron multiplier, ions are converted into electrons by means of a dynode electrode-when an ion strikes the dynode surface, electrons are emitted and the introduced current is recorded. A less common ion-detection system is the photo-multiplier. In a photo-multiplier, ions are initially converted by a dynode into electrons. These electrons are then converted into photons by means of a phosphorous screen. A photomultiplier, operating in a cascading mode, provides signal amplification. The photo-multiplier is sealed and kept under vacuum (photons pass through the glass), and therefore may have a longer lifespan; this is different to the electron multiplier, which is exposed to the internal environment of the mass spectrometer. TOF mass analyzers have a very high scan speed, and require fast detectors with a large and plane detection are so that many different ions can be detected at the same time. Multi-channel plate detectors providing time responses lower than 1ns and high grain (more than 50 mV per single ion) are typically used for this purpose. Each channel works similarly to a small CEM; because of the easier saturation and also the rapidity of the time-to-digital conversion required, the dynamic range of this type of detector is typically lower than that of the electron and photo-multiplier. Nevertheless, saturation can be observed with any type of ion detector and should be taken into account when a lower than linear response is observed. Apart from affecting quantification, detector saturation may also influence the appearance of the full-scan mass spectrum, thus impairing the performance of library search-based information. Conventional LC-MS/MS methods 70 2.4. Characteristics of current LC-MS/MS applications in forensic laboratories Essential strengths of the LC-MS/MS technology for forensic laboratories include: • Specificity: the potentially very high analytical specificity of tandem mass spectrometry as LC detector results from using the molecular mass of the analyte and its specific disintegration behavior as detection principle. • Wide range of applicability with good practicability. In contrast to GC-MS as the “classical” mass spectrometry technique, the application of LCMS/MS is not limited to volatile molecules (usually with molecular weights below 50 Da). Furthermore, aside from the highly polar analytes (i.e. amino acids), sample preparation is usually simple and does not include derivatization techniques. Mass spectrometry detected LC assays are generally optimized to shorter runtimes. Hence, compared to GC-MS, far higher sample throughput can be realized. • Flexibility. New assays can typically be developed in house with a high degree of flexibility and within a short run time. • Information rich detection. A large number of quantitative or qualitative results can be obtained for a single analytical LC-MS/MS run, since due to the fast ion selection electronics, multi-parametric, quasi parallel analysis can be performed with a mass spectrometer. 2.4.1. Handling and robustness Although routine handling of LC-MS/MS system is easier compared to GC-MS instruments, it is much more complex than operating modern day forensic chemistry analyzers. Incorrect use can cause substantial machine damage and training for several weeks is usually required for technicians to run an instrument. While everyday handling and basic maintenance procedures can doubtlessly be performed by skilled technicians, the main responsibility for LC-MS/MS installations is typically in the hands of an academic. Such an expert is particularly necessary for the development and validation of new LC-MS/MS methods which are –aside some exceptionsindividualized (“home brewed”) assay Conventional LC-MS/MS methods 71 setups tailored to the equipment available. A comprehensive assay validation including a detailed risk assessment has to be carried out. National and international guidelines (e.g. as published by SOFT (32)) are usually the basis of such an undertaking, especially in a forensic environment where especial legal regulations have to be met in some countries. It is a key feature of LC-MS/MS using API techniques as ESI or APCI, that ideally only a clean beam of ions is transferred into the high vacuum area of an instrument while unionized molecules (LC solvents and sample matrix residuals) do not enter the mass spectrometer. Solid contaminants typically precipitate in the ion source housing around the mass spectrometer’s vacuum area entrance orifice. In most cases, contaminated hardware components can be cleaned without venting the mass spectrometer. This is in contrast to GC-MS where essentially the entire effluent of the chromatographic procedure enters the high vacuum area, cleaning of which is very difficult and laborious. Hence, state of the art LC-MS/MS instruments are by far more robust than GC-MS instruments and allow the continuous analysis of large sample batches. Daily measurements series up to 24h duration with short, simple maintenance interventions (e.g. exchange of the LC stationary phases) after several days to weeks can be achieved. Taking into account the high analyte specificity of tandem mass spectrometry, chromatographic analyte separation prior to MS/MS can be minimized, if ion suppression effects are managed. State of the art LC-MS/MS instruments with oneor two-dimensional chromatography setups allow to run up to 20 analyses per hour. Consequently, within 24 h several hundred quantitative analyses can be performed with one LC-MS/MS system in a continuous work mode. In most forensic laboratories far smaller series are run in daily routine, especially if switching from one assay to another requires hardware changes (e.g. of LC columns) causing significant down times. Typically LC-MS/MS instruments work for months with minimal maintenance but can cause unexpected substantial problems without prior warning. In general, more down-times are related to the LC modules with its large number of mechanical parts compared to the mass spectrometer. Although the chromatographic methods in routine Applicability are kept as simple as possible (no complex gradient, no saltbuffered mobile phases), typically LC problems (e.g. clogging of capillaries, gas bubbles in the fluid system, crystallization of mobile phase additives, microbial growth, abrasion, erosion, and leakage problems) also occur in LC-MS/MS. Mass spectrometry related problems most frequently arise from the API spray capillary Conventional LC-MS/MS methods 72 (erosion, blockage) and problems within the ion source housing like matrix accumulation or salt precipitation. More severe problems result from substantial contamination within the vacuum area, problems in the vacuum system or electronic faults. Such events typically require intervention of a service engineer but may be prevented by regular planned maintenance visits. Troubleshooting, e.g. for decreasing sensitivity or insufficient precision or an individual method, has to include complex considerations. They have to encompass all aspects of the assay as the sample preparation (e.g. injected volume check, early blocking/leakage detection, errors in the composition of mobile phases), the MS/MSbased analyte detection (e.g. hampered by decreasing quality of nitrogen supply causing high background signals, an instable ion spray, source contamination, electrical noise due to a failure of the detector, or decreasing vacuum quality), and used chemicals (e.g. contamination of mobile phases causing ion suppression effects, the instability of internal standards or analytes (33,34). It might happen that after some maintenance actions or with increased contaminations of some hardware components, the reoptimization of about several MS instruments settings is necessary. Several of these tuning operations have to be performed in a more the less intuitive trial and error manner. Ion spray adjustments (e.g. re-optimization of the capillary position) are typically done manually. Consequently no software read-back is offered, and the contamination of these settings is hardly possible. Since almost all LC-MS/MS systems are customized by the individual user, not two installations are identical. This makes professional troubleshooting vial “hotline” or service engineers on site additionally difficult; especially if LC and MS are purchased from different manufacturers. With few exceptions, there is no comprehensive support available from the MS industry with covers all components of an installation or specific applicability. Conventional LC-MS/MS methods 79 2.5.3. Solid phase extraction (SPE) The general definition of SPE could be as follows: Separation of analyte(s) from a mixture of matrix compounds by selective portioning of the compounds between a solid phase (sorbent) and a liquid phase (sample&solvent). SPE is a dynamic process. During extraction, the equilibrium, between analyte concentration in the sample and the analyte concentration on the sorbent, is continuously shifting. Ideally, the analyte ‘likes’ the sorbent much better than the sample and is completely extracted from the sample. The less sorbent material is required to extract the analyte from the sample, the more selective and thus cleaner the extract will be. To create a selective method, often simple sample pre-treatment steps are required before the actual extraction. First of all, the sample sometimes contains particulates that have to be filtered out. The SPE cartridge and its frit can filter out particulates, but this may not be sufficient or cause clogging. In this case centrifugation or filtration of the sample is required. Secondly, the sample is often diluted or a buffer is added to neutralize or charge the analyte. By changing the characteristics of the sample or the analyte, the conditions to retain the analyte on the sorbent can be improved. At the same time, to compensate for mistakes occurring from sample pretreatment all the way to the detection, an internal standard is added to the sample. Sometimes analytes also interact with matrix compounds (protein binding). For the analyte to retain on the cartridge, it preferably must be free in solution. Depending on the type of binding, a solvent can be added to the sample in order to disrupt this drug-protein interaction. Bellow is described how SPE works for a reversed phase interaction. • Activation: the function of this step is to wet or activate the functional groups of the sorbent to allow proper interaction with the analyte. If extraction sorbents are hydrophobic they will not be wetted by an aqueous solution. Therefore organic solvents are used. In case the sorbent is wettable, this step may be omitted. • Equilibration: the function of this step is to create a sorbent chemistry environment similar to that of the sample. If the pH is important for the Conventional LC-MS/MS methods 80 extraction chemistry, the sample pH should be adjusted properly in both the sample pre-treatment and the sample conditioning step. It is also important the solvent is fully miscible and compatible with the sample; otherwise precipitation of sample matrix components could potentially cause blockages. • Sample loading: during sample loading the free analyte will bind the extraction solvent. In this step, the flow rate of the loading step is directly related to the residence time of the analyte on the extraction solvent. A too high flow could result in breakthrough. Some of the matrix compounds will also retain on the sorbent. Usually there is enough sorbent material so this does not negatively impact the capacity. As long as the capacity of the sorbent is sufficient, the amount extracted increases proportional with the sample volume loaded on the cartridge. • Washing: with a wash step most of the retained matrix compounds are flushed to waste. The wash solvent usually has higher elution strength (% organic solvent) than the solvent used for sample loading. Matrix compounds with a weaker interaction with the sorbent compared to the analyte will go the waste while matrix compounds with stronger interactions will still remain on the cartridge. Use the strongest possible wash solvent in order to get the cleanest extract. However, during the wash, analyte-sorbent interactions can also be partially disrupted causing the analyte to move further in the SPE cartridge. Ideally, the wash is stopped before some of the analyte starts to elute from the cartridge (breakthrough). • Elution: after the majority of the matrix compounds have been washed to waste, the analyte is eluted from the sorbent. Typically, a strong elution solvent is used to disrupt all the interactions of the analyte with the sorbent. Apart from the analyte, the matrix compounds that have not been washed away are now also eluted with the analyte. These matrix compounds can still cause interferences during detection. The elution step can also be used to even further improve clean-up. This means the analyte elutes completely, but matrix compounds with stronger interactions are retained on the cartridge. Conventional LC-MS/MS methods 81 Thus, SPE is currently used as a routine sample clean up method in forensic and toxicology laboratories for a wide range of compounds (62,77-88). 2.5.4. Pros and Cons The table below shows the advantages and disadvantages it these sample preparation technologies. Table 2. Overview of sample preparation technologies most important criteria Assay Performance Method development Workflow Clean-up Sensitivity Reproducibility Time Simplicity Validation Automation Software Integration PPT    ☺ ☺  ☺  LLE ☺  ☺   ☺   SPE ☺     ☺   Conventional LC-MS/MS methods 82 2.6 LC-MS(/MS) methods for the analysis of hallucinogens, cannabinoids, and amphetamines in blood and urine in the literature A wide variety of body fluid specimens have been utilized for analysis for the presence of drugs of abuse. Blood is widely regarded as the specimen offering the best correlation between drug levels and likely dosing and likely concomitant pharmacological, cognitive, and psychomotor effects. Drug levels found in blood are often quite low (ng/mL) and often short-lived. The analysis of drugs in blood is timeconsuming, generally requiring extraction procedures before further analysis can be performed. There have been several publications addressing the application of urine immunoassays to the analysis of blood specimens, after appropriate extractions protocols (89,90,91). Although blood is widely used for drug testing in forensic toxicology settings, the invasiveness of the collection of blood specimens does not lend itself to routine testing in other non forensic context (e.g. workplace, testing environments). Furthermore, there is much greater risk of transmission of infection disease though handling of blood samples than with other alternative matrices not discussed in this thesis (e.g. oral fluid, sweat, hair). Urine offers the advantages of large specimen volume and relatively high drug concentrations. Urine is 95% water, with sodium chloride and urea in about equal amounts as the main dissolved substances, and with smaller amounts if a wide variety of other constituents. Moreover, urine is relatively easy to collect and analyze. There are a wide variety of immunoassays available for detection of most common drugs of abuse and/or their metabolites in urine. However, one of the most important limitations of urine is the relative difficulty in correlating urine drug/and or metabolite levels with likely dosing and likely impairment. Accordingly to the scope of this thesis, below it is described the analytical LCMS(/MS) methods for the determination of hallucinogens, cannabinoids and amphetamines in blood and urine. Conventional LC-MS/MS methods 83 Table 3: LC-MS(/MS) methods for the analysis of hallucinogens in blood and urine in the literature LC Reference Analyte Sample Sample preparation Column Mobile Phase Interphase Detector Validation (92) Cathinone Mescaline Plasma (1 mL) SPE (MCX) Zorbax SCX Ammonium formate Acetonitrile ESI+ Ion Trap Linearity: 10-1000 ng/mL Recovery>72% Matrix effect<7% Inaccuracy<12% Stability (93) Chlorpheniramine Plasma (500 µL) LLE (Diethyl ether) Kromasil Water Acetonitrile Formic acid ESI+ QqQ Linearity 0.2-50 ng/mL LOD 0.2 ng/mL Imprecision <10% Inacuracy<8% Stability Recovery>75% Applicability (94) Chorpheniramine Plasma (1 mL) LLE (Diethyl ether) Cyclobond I 2000 Diethylamide Methanol Acetonitrile ESI+ Simple Quad Linearity 0.13-50 ng/mL Recovery>79% Imprecision<12% Inaccuracy<10% Applicability (95) Chlorpheniramine Plasma (0.5 mL) LLE (Diethyl ether) Develosil PhA Ammonium acetate Acetonitrile Methanol ESI+ Simple Quad Linearity 0.52-20.8 ng/mL LOQ 0.52 ng/mL Imprecision<14% Inaccuracy<92% Stability Applicability Conventional LC-MS/MS methods 84 Table 3: LC-MS(/MS) methods for the analysis of hallucinogens in blood and urine in the literature (continued) LC Reference Analyte Sample Sample preparation Column Mobile Phase Interphase Detector Validation (96) Chlorpheniramine Plasma (0.2 mL) LLE (Methyl butyl ether) Betasil Diol Zorbax-SB C18 Ammonium formate Methanol ESI+ APCI+ QqQ LOQ 0.2 ng/mL Imprecision<38% Matrix effect<30% (97) Ibogaine Urine (0.25 mL) SPE (HLB) Zorbax eclipse XDB Acetonitrile Ammonium formate ESI+ Simple Quad Linearity 1.78-358 ng/mL LOQ 1.78 ng/mL Inaccuracy<8% Recovery>70% No matrix effect Stability Applicability (98) Ibogaine Blood (1 mL) LLE (Methylene chloride: Isopropanol) ODB Uptisphere C18 Formate buffer Acetonitrile ESI+ QqQ Linearity 0.05-5000 ng/mL LOQ 0.05 ng/mL LOD 1 ng/mL Imprecision<15% Inaccuracy<15% Recovery>30% Matrix effect<47% Applicability (99) Ibogaine, Psilocin Scopolamine Urine (50µL) Hydrolysis Hypersil Gold Formic acid Acetonitrile ESI+ QqQ Linearity 5-500 ng/mL Imprecision<27% No matrix effect Stability Applicability Conventional LC-MS/MS methods 85 Table 3: LC-MS(/MS) methods for the analysis of hallucinogens in blood and urine in the literature (continued) LC Reference Analyte Sample Sample preparation Column Mobile Phase Interphase Detector Validation (100) Kavain Serum and urine (1 mL) Hydrolysis LLE (Dichloromethane: Diethylether) Polar RP HexylPropyl Ammonium Formate Acetonitrile Water TIS+ QqQ No validation Applicability (101) Ketamine Norketamine Urine (1 mL) Hydrolysis SPE (Isolut HCX) LiChroCart Purosphere Star RP18e 0.1%Formic acid Acetonitrile APCI+ Simple Quad Linearity 2-2000 ng/mL LOQ 2 ng/mL Imprecision<4% Recovery >100% Applicability (102) Ketamine Norketamine Plasma (0.5 mL) SPE (HLB) Chiral AGP Isopropanol Ammonium acetate ESI+ MSD Linearity 1-125 ng/mL LOQ 1 ng/mL Imprecision<8% Inaccuracy 0% Recovery >95% Applicability (103) Ketamine Norketamine Urine (1 mL) LLE (Hexane) - 0.1% Formic acid Toluene:Acetonitrile ESI+ MSD Linearity 0-500 ng/mL LOD 3 ng/mL Imprecision<15% Inaccuracy<15% (104) Ketamine Norketamine Urine (1 mL) SPE (SPEC Plus) Brownlee PerkinElmer RP-C18 Acetonitrile Acetone Ammonium acetate TIS+ QqQ Linearity 1-500 ng/mL LOD 1 ng/mL Recovery>90% Applicability Conventional LC-MS/MS methods 86 Table 3: LC-MS(/MS) methods for the analysis of hallucinogens in blood and urine in the literature (continued) LC Reference Analyte Sample Sample preparation Column Mobile Phase Interphase Detector Validation (105) Ketamine Urine (1.6 mL) SPE (C18) Platinum EPS C18 Ammonium Formate Acetonitrile ESI+ Ion Trap Linearity 5-1000 ng/mL Imprecision<11% LOD 5 ng/mL No matrix effect Recovery>89% Applicability (106) Ketamine Urine (1.6 mL) SPE (C18) Rocket EPS C18 Ammonium Formate Acetonitrile ESI+ Ion Trap Linearity 5-160 ng/mL LOQ 5 ng/mL Imprecision<15% Inaccuracy<95% No matrix effects (107) Ketamine Plasma (0.1 mL) LLE (Dichloromethane: Isopropanol) ACE 5C18 Formic acid Acetonitirle ESI+ QqQ Linearity 1-1000 ng/mL Applicability (108) Ketamine Norketamine LSD Urine (not specified) Filtered XTerra RPC18 Ammonium acetate 0.1%Formic acid Methanol HESI+ QqQ Linearity 1-1200 ng/mL LOD>0.6 ng/mL LOQ>2.1 ng/mL Imprecision<25% Applicability (109) Ketamine Norketamine Urine (not specified) Filtered Supelcosil 18-DB Ammonium acetate Acetonitrile ESI+ APCI+ Simple Quad Linearity 5-250 ng/mL LOD>0.5 ng/mL Imprecision<13% Inaccuracy<10% Conventional LC-MS/MS methods 87 Table 3: LC-MS(/MS) methods for the analysis of hallucinogens in blood and urine in the literature (continued) LC Reference Analyte Sample Sample preparation Column Mobile Phase Interphase Detector Validation (110) Ketamine Norketamine Urine (4 mL) SPE (MCX and C8) BEH C18 0.1%Formic acid Acetonitrile ESI+ QqQ Linearity 0.1-100 ng/mL LOD> 0.03 ng/mL LOQ 0.1 ng/mL Imprecision<12% Inaccuracy<17% Recovery>61% Dilution integrity<15% No matrix effect Applicability (111) Ketamine, norketamine Blood and urine (50µL) Hydrolysis PPT (Acetonitrile) Shiseido Capcell Pak SCX UG 80 Ammonium acetate Acetonitrile ESI+ QqQ Linearity 50-5000 ng/mL Imprecision<9.4% Recovery>90% Matrix effect<30% Applicability (112) Ketamine Norketamine Plasma (100 µL) SPE (MCX) Nucleodur C18 0.1%Formic acid Acetonitrile ESI+ Simple Quad Linearity 5-500 ng/mL LOQ 4 ng/mL Imprecision<2% Recovery>84% No matrix effects Applicability Conventional LC-MS/MS methods 88 Table 3: LC-MS(/MS) methods for the analysis of hallucinogens in blood and urine in the literature (continued) LC Reference Analyte Sample Sample preparation Column Mobile Phase Interphase Detector Validation (113) Ketamine Norketamine Urine (1 mL) Hydrolysis SPE (World Wide Monitoring )ZSDAU020) Synergi Hydro RP Formate buffer Acetonitrile ESI+ Ion Trap Linearity 0-1200 ng/mL LOD 0.6 ng/mL LOQ 1.9 ng/mL Imprecision<7% Inaccuracy<6% Matrix effect<9% Recovery>97% Applicability (114) LSD Urine (0.5 mL) Hydrolysis SPE (HLB) Atlantis dC18 Ammonium formate Acetonitrile ESI+ QqQ Linearity 0.2100ng/mL) LOD 0.1ng/mL LOQ 0.2 ng/mL Recovery>65% Imprecision<15% Inaccuracy<15% Matrix effect<35% Stability Applicability (115) LSD 2-oxo-3-OH-LSD Blood and Urine (1mL) LLE (Butyl acetate) Zorbax SB-C18 0.05% formic acid Acetonitrile ESI+ QqQ Linearity 0.01-200 ng/mL LOQ >0.01 ng/mL Imprecision<23% Inaccuracy<12% (116) 2-Oxo-3-HydroxyLSD Urine (5 mL) SPE (Anion Exchange) Eclipse XDB C18 Ammonium acetate Acetonitrile APCI+ Ion Trap Matrix effect Stability Conventional LC-MS/MS methods 95 Table 5 : LC-MS(/MS) methods for the analysis of amphetamines in blood and urine in the literature (continued) LC Reference Analyte Sample Sample preparation Column Mobile Phase Interphase Detector Validation (136) Amphetamine, methamphetamine, MDA, MDMA, ephedrine Blood (1 mL) SPE (MCX) BEH C18 Pyrrolidine Methanol ESI+ Single Quad No validation Evaluation of chromatographic separation (137) Amphetamine, methamphetamine, MDA, MDMA, MDEA Plasma (1 mL) SPE (HLB) Atlantis dC18 Ammonium formate Acetonitrile ESI+ Simple Quad Linearity 2-250 ng/mL LOD>0.5 ng/mL LOQ 2 ng/mL Recovery>50% No matrix effect Imprecision<15% Inaccuracy<15% No carryover Applicability (138) Amphetamine Methamphetamine Urine (10 mL) SPE (HLB)° Symmetry Shield RP18 0.05% Formic acid Acetonitrile ESI+ Ion Trap Linearity 5-500 ng/mL LOD 1 ng/mL LOQ 5 ng/mL Imprecision<7% Inaccuracy<6% Recovery>97% No matrix effect Applicability Conventional LC-MS/MS methods 96 Table 5 : LC-MS(/MS) methods for the analysis of amphetamines in blood and urine in the literature (continued) LC Reference Analyte Sample Sample preparation Column Mobile Phase Interphase Detector Validation (106) Amphetamine, Methamphetamine, MDA, MDMA Urine (1.6 mL) SPE (C18) Rocket EPS C18 Ammonium Formate Acetonitrile ESI+ Ion Trap Linearity 5-160 ng/mL LOQ 5 ng/mL Imprecision<15% Inaccuracy<95% No matrix effects (114) Amphetamine, methamphetamine, MDA, MDMA, PMA Urine (0.5 mL) LLE (Diethyl ether) Atlantis dC18 Ammonium formate Acetonitrile ESI+ QqQ Linearity (1-1000 ng/mL) Recovery>80% Imprecision<18% Inaccuracy<18% LOD>0.2 ng/mL LOQ>1 ng/mL No matrix effects Applicability (139) Amphetamine, Methamphetamine, MDA, MDMA Urine (0.5 mL) Hydrolysis SPE (HLB) Atlantis dC18 Ammonium formate Acetonitrile ESI+ QqQ Linearity 12000ng/mL) LOD>0.5ng/mL LOQ>1 ng/mL Recovery>80% Imprecision<15% Inaccuracy<15% Matrix effect<30% Stability Applicability Conventional LC-MS/MS methods 97 Table 5 : LC-MS(/MS) methods for the analysis of amphetamines in blood and urine in the literature (continued) LC Reference Analyte Sample Sample preparation Column Mobile Phase Interphase Detector Validation (140) Amphetamine, methamphetamine, MDA, MDMA, MDEA Blood and urine (not specified) Toxitube A Utisphere ODB C18 Ammonium formate Acetonitrile ESI+ QqQ Linearity 0.1-50 ng/mL LOQ 0.1 ng/mL Imprecision<13% Inaccuracy<20% No matriz effects Applicability (109) Methamphetamine Urine (1.6 mL) SPE (C18) Platinum EPS C18 Ammonium formate Acetonitrile ESI+ Ion Trap Linearity 505000ng/mL LOD 10 ng/mL Imprecision<12% Inaccuracy<13% No matrix effects Applicability (111) MDMA, MDA, methamphetamine, amphetamine Blood and urine (50µL) Hydrolysis PPT (Acetonitrile) Shiseido Capcell Pak SCX UG 80 Ammonium acetate Acetonitrile ESI+ QqQ Linearity 50-5000 ng/mL Imprecision<9.4% Recovery>90% Matrix effect<30% Applicability (141) Amphetamine, methamphetamine, MDA, MDMA Urine (20 µL) Dilution Luna C18 0.1%Formic acid Acetonitrile ESI+ QqQ Linearity 010000ng/mL LOD>2 ng/mL LOQ>7 ng/mL Imprecision<16% No matrix effect Applicability Conventional LC-MS/MS methods 98 References 1. P. J. Taylor. High-performance liquid chromatography-mass spectrometry in the clinical laboratory. Ther. Drug Monit. 27: 689-693 (2005). 2. M. Vogeser. Liquid chromatography-tandem mass spectrometry--application in the clinical laboratory. Clin. Chem. Lab Med. 41: 117-126 (2003). 3. L. D. Bowers. High-performance liquid chromatography/mass spectrometry: state of the art for the drug analysis laboratory. Clin. Chem. 35: 12821287 (1989). 4. T. R. Covey, E. D. Lee and J. D. Henion. High-speed liquid chromatography/tandem mass spectrometry for the determination of drugs in biological samples. Anal. Chem. 58: 2453-2460 (1986). 5. D. H. Chace, D. S. Millington, N. Terada, S. G. Kahler, C. R. Roe and L. F. Hofman. Rapid diagnosis of phenylketonuria by quantitative analysis for phenylalanine and tyrosine in neonatal blood spots by tandem mass spectrometry. Clin. Chem. 39: 66-71 (1993). 6. M. Constanzer, C. Chavez and B. Matuszewski. Development and comparison of high-performance liquid chromatographic methods with tandem mass spectrometric and ultraviolet absorbance detection for the determination of cyclobenzaprine in human plasma and urine. J. Chromatogr. B Biomed. Appl. 666: 117-126 (1995). 7. M. S. Rashed, M. P. Bucknall, D. Little, A. Awad, M. Jacob, M. Alamoudi, M. Alwattar and P. T. Ozand. Screening blood spots for inborn errors of metabolism by electrospray tandem mass spectrometry with a microplate batch process and a computer algorithm for automated flagging of abnormal profiles. Clin. Chem. 43: 1129-1141 (1997). 8. P. J. Taylor, A. Jones, G. A. Balderson, S. V. Lynch, R. L. Norris and S. M. Pond. Sensitive, specific quantitative analysis of tacrolimus (FK506) in blood by liquid chromatography-electrospray tandem mass spectrometry. Clin. Chem. 42: 279-285 (1996). Conventional LC-MS/MS methods 99 9. E. Brewer and J. Henion. Atmospheric pressure ionization LC/MS/MS techniques for drug disposition studies. J. Pharm. Sci. 87: 395-402 (1998). 10. P. Kebarle. A brief overview of the present status of the mechanisms involved in electrospray mass spectrometry. J. Mass Spectrom. 35: 804-817 (2000). 11. K. A. Mortier, G. F. Zhang, C. H. van Peteghem and W. E. Lambert. Adduct formation in quantitative bioanalysis: effect of ionization conditions on paclitaxel. J. Am. Soc. Mass Spectrom. 15: 585-592 (2004). 12. P. Galindo-Iranzo, J. E. Quintanilla-Lopez, R. Lebron-Aguilar and B. Gomara. Improving the sensitivity of liquid chromatography-tandem mass spectrometry analysis of hexabromocyclododecanes by chlorine adduct generation. J. Chromatogr. A 1216: 3919-3926 (2009). 13. O. J. Pozo, P. Van Eenoo, K. Deventer and F. T. Delbeke. Ionization of anabolic steroids by adduct formation in liquid chromatography electrospray mass spectrometry. J. Mass Spectrom. 42: 497-516 (2007). 14. M. Jemal, R. B. Almond and D. S. Teitz. Quantitative bioanalysis utilizing high-performance liquid chromatography/electrospray mass spectrometry via selected-ion monitoring of the sodium ion adduct [M+Na]+. Rapid Commun. Mass Spectrom. 11: 1083-1088 (1997). 15. W. Weinmann, M. Stoertzel, S. Vogt and J. Wendt. Tune compounds for electrospray ionisation/in-source collision-induced dissociation with mass spectral library searching. J. Chromatogr. A 926: 199-209 (2001). 16. R. Dams, T. Benijts, W. Gunther, W. Lambert and A. De Leenheer. Influence of the eluent composition on the ionization efficiency for morphine of pneumatically assisted electrospray, atmospheric-pressure chemical ionization and sonic spray. Rapid Commun. Mass Spectrom. 16: 10721077 (2002). Conventional LC-MS/MS methods 100 17. R. Franski, A. Tezyk, R. Wachowiak and G. Schroeder. Influence of mobile phase composition on the high-performance liquid chromatographic/electrospray ionization mass spectrometric analysis of 11-nor-9-carboxy-delta9-tetrahydrocannabinol (THC-COOH) and its glucuronide in urine. J. Mass Spectrom. 39: 458-460 (2004). 18. L. Peng and T. Farkas. Analysis of basic compounds by reversed-phase liquid chromatography-electrospray mass spectrometry in high-pH mobile phases. J. Chromatogr. A 1179: 131-144 (2008). 19. S. Zhou and K. D. Cook. Probing solvent fractionation in electrospray droplets with laser-induced fluorescence of a solvatochromic dye. Anal. Chem. 72: 963-969 (2000). 20. S. Zhou, B. S. Prebyl and K. D. Cook. Profiling pH changes in the electrospray plume. Anal. Chem. 74: 4885-4888 (2002). 21. R. Dams, M. A. Huestis, W. E. Lambert and C. M. Murphy. Matrix effect in bio-analysis of illicit drugs with LC-MS/MS: influence of ionization type, sample preparation, and biofluid. J. Am. Soc. Mass Spectrom. 14: 1290-1294 (2003). 22. Y. Hsieh, M. Chintala, H. Mei, J. Agans, J. M. Brisson, K. Ng and W. A. Korfmacher. Quantitative screening and matrix effect studies of drug discovery compounds in monkey plasma using fast-gradient liquid chromatography/tandem mass spectrometry. Rapid Commun. Mass Spectrom. 15: 2481-2487 (2001). 23. S. Souverain, S. Rudaz and J. L. Veuthey. Matrix effect in LC-ESI-MS and LC-APCI-MS with off-line and on-line extraction procedures. J. Chromatogr. A 1058: 61-66 (2004). 24. J. Zrostlikova, J. Hajslova, J. Poustka and P. Begany. Alternative calibration approaches to compensate the effect of co-extracted matrix components in liquid chromatography-electrospray ionisation tandem mass spectrometry analysis of pesticide residues in plant materials. J. Chromatogr. A 973: 13-26 (2002). Conventional LC-MS/MS methods 101 25. L. Politi, L. Morini, A. Groppi, V. Poloni, F. Pozzi and A. Polettini. Direct determination of the ethanol metabolites ethyl glucuronide and ethyl sulfate in urine by liquid chromatography/electrospray tandem mass spectrometry. Rapid Commun. Mass Spectrom. 19: 1321-1331 (2005). 26. W. Naidong. Bioanalytical liquid chromatography tandem mass spectrometry methods on underivatized silica columns with aqueous/organic mobile phases. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 796: 209224 (2003). 27. J. Canezin, A. Cailleux, A. Turcant, A. Le Bouil, P. Harry and P. Allain. Determination of LSD and its metabolites in human biological fluids by high-performance liquid chromatography with electrospray tandem mass spectrometry. J. Chromatogr. B Biomed. Sci. Appl. 765: 15-27 (2001). 28. D. Q. Liu and C. E. Hop. Strategies for characterization of drug metabolites using liquid chromatography-tandem mass spectrometry in conjunction with chemical derivatization and on-line H/D exchange approaches. J. Pharm. Biomed. Anal. 37: 1-18 (2005). 29. R. Kostiainen, T. Kotiaho, T. Kuuranne and S. Auriola. Liquid chromatography/atmospheric pressure ionization-mass spectrometry in drug metabolism studies. J. Mass Spectrom. 38: 357-372 (2003). 30. B. K. Matuszewski, M. L. Constanzer and C. M. Chavez-Eng. Matrix effect in quantitative LC/MS/MS analyses of biological fluids: a method for determination of finasteride in human plasma at picogram per milliliter concentrations. Anal. Chem. 70: 882-889 (1998). 31. M. Gergov, I. Ojanpera and E. Vuori. Simultaneous screening for 238 drugs in blood by liquid chromatography-ion spray tandem mass spectrometry with multiple-reaction monitoring. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 795: 41-53 (2003). 32. The Society of Forensic Toxicologists. SOFT/AAFS Forensic Laboratory Guidelines - 2006 Version(2006) www.soft.org. Conventional LC-MS/MS methods 102 33. T. M. Annesley. Methanol-associated matrix effects in electrospray ionization tandem mass spectrometry. Clin. Chem. 53: 1827-1834 (2007). 34. K. L. Napoli. Organic solvents compromise performance of internal standard (ascomycin) in proficiency testing of mass spectrometry-based assays for tacrolimus. Clin. Chem. 52: 765-766 (2006). 35. M. Vogeser, R. Zachoval, U. Spohrer and K. Jacob. Potential lack of specificity using electrospray tandem-mass spectrometry for the analysis of mycophenolic acid in serum. Ther. Drug Monit. 23: 722-724 (2001). 36. M. Vogeser and U. Spohrer. Pitfall in the high-throughput quantification of whole blood cyclosporin A using liquid chromatography-tandem mass spectrometry. Clin. Chem. Lab Med. 43: 400-402 (2005). 37. S. Cristoni, L. R. Bernardi, P. Gerthoux, P. Mocarelli, M. Brambilla, E. Gonella, F. Guidugli and M. Cantu. Surface-activated chemical ionization and high-flow gradient chromatography to reduce matrix effect. Rapid Commun. Mass Spectrom. 20: 2376-2382 (2006). 38. M. M. De Zan, G. Gil, M. J. Culzoni, R. G. Siano, H. C. Goicoechea and G. M. Martinez. Solving matrix-effects exploiting the second order advantage in the resolution and determination of eight tetracycline antibiotics in effluent wastewater by modelling liquid chromatography data with multivariate curve resolution-alternating least squares and unfoldedpartial least squares followed by residual bilinearization algorithms I. Effect of signal pre-treatment. J. Chromatogr. A 1179: 106-114 (2008). 39. I. Fernandez-Figares, L. C. Rodriguez and A. Gonzalez-Casado. Effect of different matrices on physiological amino acids analysis by liquid chromatography: evaluation and correction of the matrix effect. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 799: 73-79 (2004). 40. S. J. Hattan, J. Marchese, M. Albertinetti, S. Krishnan, N. Khainovski and P. Juhasz. Effect of solvent composition on signal intensity in liquid chromatography-matrix-assisted laser desorption ionization experiments. J. Chromatogr. A 1053: 291-297 (2004). Conventional LC-MS/MS methods 103 41. E. Rogatsky and D. Stein. Evaluation of matrix effect and chromatography efficiency: new parameters for validation of method development. J. Am. Soc. Mass Spectrom. 16: 1757-1759 (2005). 42. M. Villagrasa, M. Guillamon, E. Eljarrat and D. Barcelo. Matrix effect in liquid chromatography-electrospray ionization mass spectrometry analysis of benzoxazinoid derivatives in plant material. J. Chromatogr. A 1157: 108-114 (2007). 43. S. T. Wu, D. Schoener and M. Jemal. Plasma phospholipids implicated in the matrix effect observed in liquid chromatography/tandem mass spectrometry bioanalysis: evaluation of the use of colloidal silica in combination with divalent or trivalent cations for the selective removal of phospholipids from plasma. Rapid Commun. Mass Spectrom. 22: 2873-2881 (2008). 44. A. Van Eeckhaut, K. Lanckmans, S. Sarre, I. Smolders and Y. Michotte. Validation of bioanalytical LC-MS/MS assays: evaluation of matrix effects. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 877: 21982207 (2009). 45. E. Chambers, D. M. Wagrowski-Diehl, Z. Lu and J. R. Mazzeo. Systematic and comprehensive strategy for reducing matrix effects in LC/MS/MS analyses. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 852: 2234 (2007). 46. T. M. Annesley. Ion suppression in mass spectrometry. Clin. Chem. 49: 10411044 (2003). 47. S. A. Gustavsson, J. Samskog, K. E. Markides and B. Langstrom. Studies of signal suppression in liquid chromatography-electrospray ionization mass spectrometry using volatile ion-pairing reagents. J. Chromatogr. A 937: 41-47 (2001). 48. C. G. Huber and M. R. Buchmeiser. On-line cation exchange for suppression of adduct formation in negative-ion electrospray mass spectrometry of nucleic acids. Anal. Chem. 70: 5288-5295 (1998). Conventional LC-MS/MS methods 104 49. P. J. Larger, M. Breda, D. Fraier, H. Hughes and C. A. James. Ionsuppression effects in liquid chromatography-tandem mass spectrometry due to a formulation agent, a case study in drug discovery bioanalysis. J. Pharm. Biomed. Anal. 39: 206-216 (2005). 50. F. T. Peters, O. H. Drummer and F. Musshoff. Validation of new methods. Forensic Sci. Int. 165: 216-224 (2007). 51. R. Bonfiglio, R. C. King, T. V. Olah and K. Merkle. The effects of sample preparation methods on the variability of the electrospray ionization response for model drug compounds. Rapid Commun. Mass Spectrom. 13: 1175-1185 (1999). 52. B. K. Matuszewski, M. L. Constanzer and C. M. Chavez-Eng. Strategies for the assessment of matrix effect in quantitative bioanalytical methods based on HPLC-MS/MS. Anal. Chem. 75: 3019-3030 (2003). 53. M. Urpi-Sarda, O. Jauregui, R. M. Lamuela-Raventos, W. Jaeger, M. Miksits, M. I. Covas and C. Andres-Lacueva. Uptake of diet resveratrol into the human low-density lipoprotein. Identification and quantification of resveratrol metabolites by liquid chromatography coupled with tandem mass spectrometry. Anal. Chem. 77: 3149-3155 (2005). 54. N. Van Hoof, D. Courtheyn, J. P. Antignac, W. M. Van De, S. Poelmans, H. Noppe and H. De Brabander. Multi-residue liquid chromatography/tandem mass spectrometric analysis of beta-agonists in urine using molecular imprinted polymers. Rapid Commun. Mass Spectrom. 19: 2801-2808 (2005). 55. J. Bhatt, A. Jangid, R. Shetty, B. Shah, S. Kambli, G. Subbaiah and S. Singh. Rapid and sensitive liquid chromatography-mass spectrometry method for determination of ropinirole in human plasma. J. Pharm. Biomed. Anal. 40: 1202-1208 (2006). Conventional LC-MS/MS methods 111 97. V. Kontrimaviciute, H. Breton, O. Mathieu, J. C. Mathieu-Daude and F. M. Bressolle. Liquid chromatography-electrospray mass spectrometry determination of ibogaine and noribogaine in human plasma and whole blood. Application to a poisoning involving Tabernanthe iboga root. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 843: 131-141 (2006). 98. M. Cheze, A. Lenoan, M. Deveaux and G. Pepin. Determination of ibogaine and noribogaine in biological fluids and hair by LC-MS/MS after Tabernanthe iboga abuse Iboga alkaloids distribution in a drowning death case. Forensic Sci. Int. 176: 58-66 (2008). 99. K. Bjornstad, O. Beck and A. Helander. A multi-component LC-MS/MS method for detection of ten plant-derived psychoactive substances in urine. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 877: 11621168 (2009). 100. F. Tarbah, H. Mahler, B. Kardel, W. Weinmann, D. Hafner and T. Daldrup. Kinetics of kavain and its metabolites after oral application. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 789: 115-130 (2003). 101. P. Adamowicz and M. Kala. Urinary excretion rates of ketamine and norketamine following therapeutic ketamine administration: method and detection window considerations. J. Anal. Toxicol. 29: 376-382 (2005). 102. M. E. Rodriguez Rosas, S. Patel and I. W. Wainer. Determination of the enantiomers of ketamine and norketamine in human plasma by enantioselective liquid chromatography-mass spectrometry. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 794: 99-108 (2003). 103. A. C. Lua and H. R. Lin. A rapid and sensitive ESI-MS screening procedure for ketamine and norketamine in urine samples. J. Anal. Toxicol. 28: 680-684 (2004). 104. K. C. Wang, T. S. Shih and S. G. Cheng. Use of SPE and LC/TIS/MS/MS for rapid detection and quantitation of ketamine and its metabolite, norketamine, in urine. Forensic Sci. Int. 147: 81-88 (2005). Conventional LC-MS/MS methods 112 105. J. Y. Cheng and V. K. Mok. Rapid determination of ketamine in urine by liquid chromatography-tandem mass spectrometry for a high throughput laboratory. Forensic Sci. Int. 142: 9-15 (2004). 106. W. C. Cheng, T. S. Yau, M. K. Wong, L. P. Chan and V. K. Mok. A highthroughput urinalysis of abused drugs based on a SPE-LC-MS/MS method coupled with an in-house developed post-analysis data treatment system. Forensic Sci. Int. 162: 95-107 (2006). 107. D. P. Lankveld, B. Driessen, L. R. Soma, P. J. Moate, J. Rudy, C. E. Uboh, P. van Dijk and L. J. Hellebrekers. Pharmacodynamic effects and pharmacokinetic profile of a long-term continuous rate infusion of racemic ketamine in healthy conscious horses. J. Vet. Pharmacol. Ther. 29: 477-488 (2006). 108. C. Y. Chen, C. C. Shen, T. J. Yang, Y. Z. Chang and M. R. Lee. High throughput screening various abused drugs and metabolites in urine by liquid chromatography-heated electrospray ionization/tandem mass spectrometry. Talanta 77: 1351-1357 (2009). 109. C. Y. Chen, M. R. Lee, F. C. Cheng and G. J. Wu. Determination of ketamine and metabolites in urine by liquid chromatography-mass spectrometry. Talanta 72: 1217-1222 (2007). 110. M. C. Parkin, S. C. Turfus, N. W. Smith, J. M. Halket, R. A. Braithwaite, S. P. Elliott, M. D. Osselton, D. A. Cowan and A. T. Kicman. Detection of ketamine and its metabolites in urine by ultra high pressure liquid chromatography-tandem mass spectrometry. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 876: 137-142 (2008). 111. K. Kuwayama, H. Inoue, T. Kanamori, K. Tsujikawa, H. Miyaguchi, Y. T. Iwata, S. Miyauchi and N. Kamo. Analysis of amphetamine-type stimulants and their metabolites in plasma, urine and bile by liquid chromatography with a strong cation-exchange column-tandem mass spectrometry. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 867: 78-83 (2008). Conventional LC-MS/MS methods 113 112. T. Legrand, S. Roy, C. Monchaud, C. Grondin, M. Duval and E. JacqzAigrain. Determination of ketamine and norketamine in plasma by micro-liquid chromatography-mass spectrometry. J. Pharm. Biomed. Anal. 48: 171-176 (2008). 113. N. Harun, R. A. Anderson and E. I. Miller. Validation of an enzyme-linked immunosorbent assay screening method and a liquid chromatographytandem mass spectrometry confirmation method for the identification and quantification of ketamine and norketamine in urine samples from malaysia. J. Anal. Toxicol. 33: 310-321 (2009). 114. M. Concheiro, S. M. Simoes, O. Quintela, A. de Castro, M. J. Dias, A. Cruz and M. Lopez-Rivadulla. Fast LC-MS/MS method for the determination of amphetamine, methamphetamine, MDA, MDMA, MDEA, MBDB and PMA in urine. Forensic Sci. Int. 171: 44-51 (2007). 115. S. S. Johansen and J. L. Jensen. Liquid chromatography-tandem mass spectrometry determination of LSD, ISO-LSD, and the main metabolite 2-oxo-3-hydroxy-LSD in forensic samples and application in a forensic case. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 825: 21-28 (2005). 116. C. K. Horn, K. L. Klette and P. R. Stout. LC-MS analysis of 2-oxo-3-hydroxy LSD from urine using a Speedisk positive-pressure processor with Cerex PolyChrom CLIN II columns. J. Anal. Toxicol. 27: 459-463 (2003). 117. G. K. Poch, K. L. Klette, D. A. Hallare, M. G. Manglicmot, R. J. Czarny, L. K. McWhorter and C. J. Anderson. Detection of metabolites of lysergic acid diethylamide (LSD) in human urine specimens: 2-oxo-3-hydroxyLSD, a prevalent metabolite of LSD. J. Chromatogr. B Biomed. Sci. Appl. 724: 23-33 (1999). 118. S. A. White, T. Catterick, M. E. Harrison, D. E. Johnston, G. D. Reed and K. S. Webb. Determination of lysergide in urine by high-performance liquid chromatography combined with electrospray ionisation mass spectrometry. J. Chromatogr. B Biomed. Sci. Appl. 689: 335-340 (1997). Conventional LC-MS/MS methods 114 119. H. Hoja, P. Marquet, B. Verneuil, H. Lotfi, J. L. Dupuy and G. Lachatre. Determination of LSD and N-demethyl-LSD in urine by liquid chromatography coupled to electrospray ionization mass spectrometry. J. Chromatogr. B Biomed. Sci. Appl. 692: 329-335 (1997). 120. K. Bjornstad, A. Helander and O. Beck. Development and clinical application of an LC-MS-MS method for mescaline in urine. J. Anal. Toxicol. 32: 227-231 (2008). 121. D. R. Doerge, C. M. Fogle, M. G. Paule, M. McCullagh and S. Bajic. Analysis of methylphenidate and its metabolite ritalinic acid in monkey plasma by liquid chromatography/electrospray ionization mass spectrometry. Rapid Commun. Mass Spectrom. 14: 619-623 (2000). 122. E. Marchei, M. Farre, M. Pellegrini, S. Rossi, O. Garcia-Algar, O. Vall and S. Pichini. Liquid chromatography-electrospray ionization mass spectrometry determination of methylphenidate and ritalinic acid in conventional and non-conventional biological matrices. J. Pharm. Biomed. Anal. 49: 434-439 (2009). 123. T. Kamata, M. Nishikawa, M. Katagi and H. Tsuchihashi. Optimized glucuronide hydrolysis for the detection of psilocin in human urine samples. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 796: 421-427 (2003). 124. H. Chen, Y. Chen, H. Wang, P. Du, F. Han and H. Zhang. Analysis of scopolamine and its eighteen metabolites in rat urine by liquid chromatography-tandem mass spectrometry. Talanta 67: 984-991 (2005). 125. W. Weinmann, M. Goerner, S. Vogt, R. Goerke and S. Pollak. Fast confirmation of 11-nor-9-carboxy-Delta(9)-tetrahydrocannabinol (THCCOOH) in urine by LC/MS/MS using negative atmospheric-pressure chemical ionisation (APCI). Forensic Sci. Int. 121: 103-107 (2001). Conventional LC-MS/MS methods 115 126. R. Yang and W. Xie. Determination of cannabinoids in biological samples using a new solid phase micro-extraction membrane and liquid chromatography-mass spectrometry. Forensic Sci. Int. 162: 135-139 (2006). 127. M. Wood, M. Laloup, M. M. Ramirez Fernandez, K. M. Jenkins, M. S. Young, J. G. Ramaekers, G. De Boeck and N. Samyn. Quantitative analysis of multiple illicit drugs in preserved oral fluid by solid-phase extraction and liquid chromatography-tandem mass spectrometry. Forensic Sci. Int. 150: 227-238 (2005). 128. H. Teixeira, A. Verstraete, P. Proenca, F. Corte-Real, P. Monsanto and D. N. Vieira. Validated method for the simultaneous determination of Delta9THC and Delta9-THC-COOH in oral fluid, urine and whole blood using solid-phase extraction and liquid chromatography-mass spectrometry with electrospray ionization. Forensic Sci. Int. 170: 148-155 (2007). 129. S. B. Grauwiler, A. Scholer and J. Drewe. Development of a LC/MS/MS method for the analysis of cannabinoids in human EDTA-plasma and urine after small doses of Cannabis sativa extracts. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 850: 515-522 (2007). 130. C. Jamey, E. Szwarc, A. Tracqui and B. Ludes. Determination of cannabinoids in whole blood by UPLC-MS-MS. J. Anal. Toxicol. 32: 349-354 (2008). 131. N. Stephanson, M. Josefsson, R. Kronstrand and O. Beck. Accurate identification and quantification of 11-nor-delta(9)- tetrahydrocannabinol-9-carboxylic acid in urine drug testing: evaluation of a direct high efficiency liquid chromatographic-mass spectrometric method. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 871: 101108 (2008). 132. M. Wood, G. De Boeck, N. Samyn, M. Morris, D. P. Cooper, R. A. Maes and E. A. De Bruijn. Development of a rapid and sensitive method for the quantitation of amphetamines in human plasma and oral fluid by LCMS-MS. J. Anal. Toxicol. 27: 78-87 (2003). Conventional LC-MS/MS methods 116 133. H. P. Hendrickson, A. Milesi-Halle, E. M. Laurenzana and S. M. Owens. Development of a liquid chromatography-tandem mass spectrometric method for the determination of methamphetamine and amphetamine using small volumes of rat serum. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 806: 81-87 (2004). 134. T. Y. Wu and M. R. Fuh. Determination of amphetamine, methamphetamine, 3,4-methylenedioxyamphetamine, 3,4methylenedioxyethylamphetamine, and 3,4methylenedioxymethamphetamine in urine by online solid-phase extraction and ion-pairing liquid chromatography with detection by electrospray tandem mass spectrometry. Rapid Commun. Mass Spectrom. 19: 775-780 (2005). 135. T. S. Ho, J. L. Reubsaet, H. S. Anthonsen, S. Pedersen-Bjergaard and K. E. Rasmussen. Liquid-phase microextraction based on carrier mediated transport combined with liquid chromatography-mass spectrometry. New concept for the determination of polar drugs in a single drop of human plasma. J. Chromatogr. A 1072: 29-36 (2005). 136. L. G. Apollonio, D. J. Pianca, I. R. Whittall, W. A. Maher and J. M. Kyd. A demonstration of the use of ultra-performance liquid chromatographymass spectrometry [UPLC/MS] in the determination of amphetaminetype substances and ketamine for forensic and toxicological analysis. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 836: 111-115 (2006). 137. M. Concheiro, A. de Castro, O. Quintela, M. Lopez-Rivadulla and A. Cruz. Determination of drugs of abuse and their metabolites in human plasma by liquid chromatography-mass spectrometry. An application to 156 road fatalities. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 832: 81-89 (2006). 138. M. R. Fuh, T. Y. Wu and T. Y. Lin. Determination of amphetamine and methamphetamine in urine by solid phase extraction and ion-pair liquid chromatography-electrospray-tandem mass spectrometry. Talanta 68: 987-991 (2006). Conventional LC-MS/MS methods 117 139. M. Concheiro, A. de Castro, O. Quintela, A. Cruz and M. Lopez-Rivadulla. Determination of illicit drugs and their metabolites in human urine by liquid chromatography tandem mass spectrometry including relative ion intensity criterion. J. Anal. Toxicol. 31: 573-580 (2007). 140. M. Cheze, M. Deveaux, C. Martin, M. Lhermitte and G. Pepin. Simultaneous analysis of six amphetamines and analogues in hair, blood and urine by LC-ESI-MS/MS. Application to the determination of MDMA after low ecstasy intake. Forensic Sci. Int. 170: 100-104 (2007). 141. M. Andersson, E. Gustavsson, N. Stephanson and O. Beck. Direct injection LC-MS/MS method for identification and quantification of amphetamine, methamphetamine, 3,4-methylenedioxyamphetamine and 3,4-methylenedioxymethamphetamine in urine drug testing. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 861: 22-28 (2008). III. NEW ADVANCES IN LC-MS/MS METHODS New advances in LC-MS/MS methods 127 Table7: On-line SPE applications with the Symbiosis system SPE LC Reference Analytes Sample On-line system Cartridge Column Mobile Phase Detection Validation (14) 10 drug candidates (names not done) Plasma (500 µL) Prospekt IST Isolute solidphase CN Zorbax C18 RX Acetonitrile Water TFA QqQ Linearity 2.5-1000 ng/mL LOD: 2.5-5 ng/mL Imprecision <20% Recovery:50-100% Applicability (15) Codeine Plasma (700 µL) Prospekt Bond-Elut C2 Hypersil BDS C18, Phosphate buffer Acetonitrile UV Linearity 2-140 ng/mL LOD 0.5ng/mL Imprecision < 5.03% Inaccuracy 1.82% Recovery: 91.9% Applicability (16) Piroxicam Plasma (200 µL) Prospekt Bond Elut C8 Kromasil C18, Acetonitrle Phosphate Buffer. UV Imprecision<9.05% Inaccuracy<7.8% Recovery>90% LOQ: 50 ng/mL Applicability New advances in LC-MS/MS methods 128 Table7: On-line SPE applications with the Symbiosis system (continued) SPE LC Reference Analytes Sample On-line system Cartridge Column Mobile Phase Detection Validation (17) Camptothecin and metabolites Plasma (100 µL) Prospekt C18 analytichem Symmetry C18 Phosphate buffer Acetonitrile Fluorescence Linearity 2.5-25000 ng/mL LOQ: 2.5 and 5 ng/m Imprecision<14.2% Inaccuracy<10.5% Recovery>96% Applicability (7) Sulfadiazine, sulfamerazine, taxol, propanolol, carbamazepine, procainamide, caffeine, ranitidine, theophillyne, theobromine, acetaminophen Serum or plasma (100 µL) Prospekt Hysphere Resin GP Hypersil ODS (C18) Acetonitrile (0.1% Formic acid) Acetate buffer QqQ Linearity 2.5-40 ng/mL LOQ: >0.2 ng/mL Imprecision<15% Inaccuracy<15% Recovery>95% Carryover<0.1% Applicability New advances in LC-MS/MS methods 129 Table7: On-line SPE applications with the Symbiosis system (continued) SPE LC Reference Analytes Sample On-line system Cartridge Column Mobile Phase Detection Validation (10) Propanolol, ketoconazole Plasma (5, 50, 100 µL) Symbiosis Hysphere C18 HD Luna C18 Acetate buffer Acetonitrile QqQ Recovery>90% Linearity 0.1-100 ng/mL Imprecision<10% Inaccuracy<9% Applicability (10) Propanolol, ketoconazole, Diclofenac, ibuprofen Plasma (1 mL) Symbiosis Hysphere C18 HD Luna C18 Chromolith C18 Acetate buffer Acetonitrile QqQ Recoveries:100% Range:1-1000 ng/mL CV%:>90 Bias%<14 Applicability (18) Clozapine, desmethylclozapine, clozapine-N-oxide Serum (50 µL) Prospekt2 Hysphere C18 HD Zorbax Eclipse XDB C18 Acetate buffer Methanol MS Linearity 10-1000 ng/mL LOQ: 50 ng/mL LOD> 0.15 ng/mL Imprecision<20% Inaccuracy<10% Applicability New advances in LC-MS/MS methods 130 Table7: On-line SPE applications with the Symbiosis system (continued) SPE LC Reference Analytes Sample On-line system Cartridge Column Mobile Phase Detection Validation (19) Amitriptyline, nortriptyline, imipramine, desipramine, trazodone, fluoxetine, norfluoxetine, paroxetine, fluvoxamine, sertraline, venlafaxine, norclomipramine, citalopram, clomipramine Plasma (50 µL) Symbiosis Oasis MCX Gemini C18 Bicarbonate buffer Acetonitrile QqQ Recoveries>99% Matrix effect<18% Range: 10-1000 ng/mL CV%<20 Bias%<20 LOQ: 10 ng/mL Stability in the autosampler and freeze/thaw cycles Applicabilty New advances in LC-MS/MS methods 131 Table7: On-line SPE applications with the Symbiosis system (continued) SPE LC Reference Analytes Sample On-line system Cartridge Column Mobile Phase Detection Validation (20) Guvacine, kojic acid, theobromide, codeine, emetine, theophylline, hyoscyamine, quinine, catechin, chlorogenic acid, dihydrorobinetin, harmine, caffeic acid, khellol glucoside, berberine, coumaric acid, ellagic acid, etc Stock solutions (10 µL) Symbiosis HysPhere Resin GP Luna C18 0.1% Formic acid Acetonitrile DAD Evaluation of trapping efficiency Influence of flow rate of loading solvent Influence of acetonitrile content in loading solvent Influence of analyte content Capacity of GP cartridges (21) 8 drug compounds (names not done) Blood (5-10 µL) Symbiosis HysPhere C18 HD X-Bridge C18 0.1% Formic acid Methanol QqQ Recovery>50% Stability Applicability New advances in LC-MS/MS methods 132 Table7: On-line SPE applications with the Symbiosis system (continued) SPE LC Reference Analytes Sample On-line system Cartridge Column Mobile Phase Detection Validation (11) Cocaine, ecgonine, ecgonine methyl ester, benzoylecgonine, cocaethylene Whole blood (500 µL) Symbiosis Hysphere MM Anion Exchange Gemini C6-Phenyl 0.1%Formic acid Acetonitrile QqQ Linearity: 4-500 ng/mL LOQ : 8-47 ng/mL LOD : 3-16 ng/mL Imprecision<9% Inaccuracy<7% Applicability (12) Bennzoylecgonine, ecgonine methyl ester, ecgonine, cocaethylene Urine (1 mL) Symbiosis Hysphere MM Anion Exchange Gemini C6-Phenyl 0.1% Formic acid Acetonitrile QqQ Linearity 7-1000 ng/mL Imprecision<9% Inaccuracy<5% LOD: 3-23 ng/mL LOQ : 7-69 ng/mL Applicability New advances in LC-MS/MS methods 133 References 1. S. M. Wille and W. E. Lambert. Recent developments in extraction procedures relevant to analytical toxicology. Anal. Bioanal. Chem. 388: 1381-1391 (2007). 2. D. T. Rossi and N. Zhang. Automating solid-phase extraction: current aspects and future prospects. J. Chromatogr. A 885: 97-113 (2000). 3. S. Rodriguez-Mozaz, M. J. Lopez de Alda and D. Barcelo. Advantages and limitations of on-line solid phase extraction coupled to liquid chromatography-mass spectrometry technologies versus biosensors for monitoring of emerging contaminants in water. J. Chromatogr. A 1152: 97-115 (2007). 4. J. L. Veuthey, S. Souverain and S. Rudaz. Column-switching procedures for the fast analysis of drugs in biologic samples. Ther. Drug Monit. 26: 161-166 (2004). 5. X. S. Xu, K. X. Yan, H. Song and M. W. Lo. Quantitative determination of a novel dual PPAR alpha/gamma agonist using on-line turbulent flow extraction with liquid chromatography-tandem mass spectrometry. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 814: 29-36 (2005). 6. X. Zang, R. Luo, N. Song, T. K. Chen and H. Bozigian. A novel on-line solidphase extraction approach integrated with a monolithic column and tandem mass spectrometry for direct plasma analysis of multiple drugs and metabolites. Rapid Commun. Mass Spectrom. 19: 3259-3268 (2005). 7. A. Schellen, B. Ooms, L. D. van de, R. Vreeken and W. D. van Dongen. Generic solid phase extraction-liquid chromatography-tandem mass spectrometry method for fast determination of drugs in biological fluids. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 788: 251-259 (2003). 8. Y. Alnouti, M. Li, O. Kavetskaia, H. Bi, C. E. Hop and A. I. Gusev. Method for internal standard introduction for quantitative analysis using on-line solid-phase extraction LC-MS/MS. Anal. Chem. 78: 1331-1336 (2006). 9. T. Koal, M. Sibum, E. Koster, K. Resch and V. Kaever. Direct and fast determination of antiretroviral drugs by automated online solid-phase extractionliquid chromatography-tandem mass spectrometry in human plasma. Clin. Chem. Lab Med. 44: 299-305 (2006). New advances in LC-MS/MS methods 134 10. Y. Alnouti, K. Srinivasan, D. Waddell, H. Bi, O. Kavetskaia and A. I. Gusev. Development and application of a new on-line SPE system combined with LCMS/MS detection for high throughput direct analysis of pharmaceutical compounds in plasma. J. Chromatogr. A 1080: 99-106 (2005). 11. E. Jagerdeo, M. A. Montgomery, M. A. Lebeau and M. Sibum. An automated SPE/LC/MS/MS method for the analysis of cocaine and metabolites in whole blood. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci.(2008). 12. E. Jagerdeo, M. A. Montgomery, M. Sibum, T. A. Sasaki and M. A. Lebeau. Rapid analysis of cocaine and metabolites in urine using a completely automated solid-phase extraction-high-performance liquid chromatography-tandem mass spectrometry method. J. Anal. Toxicol. 32: 570-576 (2008). 13. M. Li, Y. Alnouti, R. Leverence, H. Bi and A. I. Gusev. Increase of the LC-MS/MS sensitivity and detection limits using on-line sample preparation with large volume plasma injection. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 825: 152-160 (2005). 14. D. A. McLoughlin, T. V. Olah and J. D. Gilbert. A direct technique for the simultaneous determination of 10 drug candidates in plasma by liquid chromatography-atmospheric pressure chemical ionization mass spectrometry interfaced to a Prospekt solid-phase extraction system. J. Pharm. Biomed. Anal. 15: 1893-1901 (1997). 15. J. A. Pascual and J. Sanagustin. Fully automated analytical method for codeine quantification in human plasma using on-line solid-phase extraction and highperformance liquid chromatography with ultraviolet detection. J. Chromatogr. B Biomed. Sci. Appl. 724: 295-302 (1999). 16. M. Yritia, P. Parra, J. M. Fernandez and J. M. Barbanoj. Piroxicam quantitation in human plasma by high-performance liquid chromatography with onand off-line solid-phase extraction. J. Chromatogr. A 846: 199-205 (1999). New advances in LC-MS/MS methods 135 17. A. Kurita and N. Kaneda. High-performance liquid chromatographic method for the simultaneous determination of the camptothecin derivative irinotecan hydrochloride, CPT-11, and its metabolites SN-38 and SN-38 glucuronide in rat plasma with a fully automated on-line solid-phase extraction system, PROSPEKT. J. Chromatogr. B Biomed. Sci. Appl. 724: 335-344 (1999). 18. H. A. Niederlander, E. H. Koster, M. J. Hilhorst, H. J. Metting, M. Eilders, B. Ooms and G. J. de Jong. High throughput therapeutic drug monitoring of clozapine and metabolites in serum by on-line coupling of solid phase extraction with liquid chromatography-mass spectrometry. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 834: 98-107 (2006). 19. A. de Castro, M. M. Ramirez Fernandez, M. Laloup, N. Samyn, G. De Boeck, M. Wood, V. Maes and M. Lopez-Rivadulla. High-throughput on-line solid-phase extraction-liquid chromatography-tandem mass spectrometry method for the simultaneous analysis of 14 antidepressants and their metabolites in plasma. J. Chromatogr. A 1160: 3-12 (2007). 20. C. Clarkson, M. Sibum, R. Mensen and J. W. Jaroszewski. Evaluation of on-line solid-phase extraction parameters for hyphenated, high-performance liquid chromatography-solid-phase extraction-nuclear magnetic resonance applications. J. Chromatogr. A 1165: 1-9 (2007). 21. B. A. Ingelse, G. Vogel, M. Botterblom, D. Nanninga and B. Ooms. Direct injection of whole blood for liquid chromatography/tandem mass spectrometry analysis to support single-rodent pharmacokinetic studies. Rapid Commun. Mass Spectrom. 22: 834-840 (2008). adjusted to optimise the signal for the most abundant product ions, which were subsequently used for multiple reaction monitoring (MRM) analysis. The transitions were m/z 315.2 !193.1 and m/z315.2 !259.3 for THC. The former (and most prominent precursor-product transition) was used for quantification and the latter transition used as a qualifier. The transition for THC-d3 was m/z318.2 !196.1. 2.3. Data analysis The experimental data were processed to obtain the following parameters: the true positives (TP, positive oral fluid samples matching a positive plasma sample; positive drug tests matching a positive oral fluid/plasma sample), true negatives (TN, negative oral fluid samples matching a negative plasma sample; negative drug tests matching a negative oral fluid/ plasma sample), false positives (FP, positive oral fluid samples that were not confirmed in plasma; positive drug tests that were not confirmed in oral fluid/plasma) and false negatives (FN, oral fluid samples that were negative but corresponding to a positive plasma result; drug tests that were negative but corresponding to a positive oral fluid/plasma result). Based on these results, the sensitivity, specificity and accuracy were calculated. In addition, the LC–MS–MS results for oral fluid and plasma were used to establish receiver operating characteristic (ROC) curves: the sensitivity was plotted versus 100-specificity for each cut-off concentration using MedCalc 1 (version 7.3) statistical software. The optimal cut-off for oral fluid to ‘predict’ a positive plasma result is the value corresponding with the highest accuracy (minimal FN and FP results). For the analysis of the results, the applied cut-off for oral fluid was always the LOQ of the LC–MS–MS method for oral fluid (0.5 ng/mL). For plasma, two different cut-off values were used: the LOQ of the LC–MS–MS method for plasma (0.5 ng/mL) and the legal limit in Belgium for DUID (2 ng/mL for THC). 2.4. The on-site Dra ¨ger DrugTest 1 system for oral fluid As part of the Belgian participation in the Rosita-2 project, the Dra ¨ger DrugTest 1 for the on-site detection of THC in oral fluid was tested. This system combines the test strip method of immunological drug detection with an innovative signal technology known as UPT (Up-Converting Phosphor Technology). It consists of two main components: the Dra ¨ger DrugTest 1 Kit for Oral Fluids (comprising the collection device and the test cassette with inserted buffercartridge) and the Dra ¨gerDrugTest 1 Analyzer (a portable analyzer for reading the test cassette and for data management). Different panel cassettes are available for detecting up to six substance classes in a single cassette. The test subject collects an oral fluid sample by gently moving the collection device in the mouth for about a minute. The sampling sponge swells in size as it soaks up the oral fluid, signaling to the tester that the sampling process is complete. The collection device is first inserted into the test cassette and pushed down to release some of the oral fluid into the buffer cartridge. The handle of the collection device is removed and the buffer cartridge is sealed. After a 4 min reaction time, the buffer cartridge is turned and lowered, triggering the immunological detection reaction within the test cassette. After 8 min, the test cassette can be inserted into the Dra ¨ger DrugTest 1 Analyzer, which displays the results of the analysis, distinct for each class of drugs, in the form of a qualitative reading (positive or negative). According to the manufacturer, the cut-off value for THC is 20 ng/mL. 3. Results and discussion 3.1. Validation of the analytical method for plasma samples Selectivity of the method was achieved by a combination of retention time, precursor and product ions. Quantification was based on the most prominent product ion (i.e. quantifier); confirmation of THC was evaluated through the presence of the second product (i.e. qualifier). At the LOQ the qualifier had a signal to noise ratio (S:N) >10:1. The acceptance range for the peak area ratio quantifier/qualifier was 2.34 0.28 for all analyses. Calibration curves for THC were prepared in blank plasma for each batch and ranged from 0.5 to 100 ng/mL. Negative controls and low and high QC samples containing 2.5 and 25 ng/mL THC respectively, were included in each batch. In each case, a weighted (1/x) linear regression line was applied. Linearity with correlation coefficients r 2 >0.999 were achieved within the range investigated. The back-calculated concentrations of all calibrators were compared with their respective nominal values and were within 100 15% of the nominal value. The LOQ, defined as the concentration of lowest calibrator which was calculated to be within 20% of the nominal value and with a % relative standard deviation (R.S.D.) less than 20%, was 0.5 ng/mL. The intra-assay precision (repeatability) and inter-assay precision (reproducibility) are shown in Table 1. Intra-assay variation was evaluated by M. Laloup et al. / Forensic Science International 161 (2006) 175–179 177 Table 1 Precision and accuracy data for the LC–MS–MS analysis of THC in plasma Concentration of QC (ng/mL) Intra-assay precision (n= 5) Inter-assay precision (n= 10) Mean concentration found (ng/mL) R.S.D. (%) Bias (%) Mean concentration found (ng/mL) R.S.D. (%) Bias (%) 2.5 2.5 3.4 1.1 2.6 11.9 2.0 25.0 24.9 1.1 0.5 25.3 11.1 1.1 Samples were prepared by the liquid–liquid extraction method as described in the text. QCs were prepared for every run in blank plasma. Intra-assay variation was evaluated by replicate (n= 5) analysis of both QC samples in a single run. Inter-assay variation was evaluated by replicate analysis of the QC samples in several experiments performed on 10 different days and by two operators. A comparison of the calculated concentrations of the QC samples to their respective nominal values, was used to assess the bias of the method. replicate (n= 5) analysis of both QC samples in a single run. Inter-assay variation was evaluated by replicate analysis of these QC samples in several experiments performed on ten different days and by two operators. A comparison of the calculated concentrations of the QC samples to their respective nominal values, was used to assess the bias of the method. Results indicated that the bias of the assay was 2.0% with all R.S.D.s <12%. The recovery of the method was estimated by comparing the response of a 5 ng/mL calibrator when the nondeuterated compound was added before the extraction step (n= 3) with the response obtained when the non-deuterated analyte was added after sample preparation (n= 3). THC-d3 was added before the extraction step in both conditions. The recovery was 79.2 0.9%. Matrix effects were evaluated by post-column infusion as described by Bonfiglio et al. [12].No ion suppression or enhancement could be detected at the elution time of THC (n= 3). 3.2. Comparison of analytical data in oral fluid and plasma The detection of the inactive metabolite 11-nor-9-carboxyTHC in plasma, which is considered not to be present in oral fluid [13], was not included in the method since only the indication of recent cannabis use (through the presence of THC) was of interest. In total, 139 combined oral fluid–plasma samples were obtained. These plasma samples included 114 positive samples (82.0%) for THC using the LOQ of the LC– MS–MS method at a cut-off value of 0.5 ng/mL. Using the legal limit for plasma as a cut-off value (2 ng/mL), 95 cases (68.3%) were positive for THC. The median concentration in plasma (8.8 ng/mL) clearly exceeded the legal cut-off level; concentrations ranged from 0.6 to 51.3 ng/mL. The analysis of the oral fluid samples from the same subjects showed a median THC concentration of 23.0 ng/mL with concentrations ranging from 0.5 to 1462 ng/mL. Table 2 presents the sensitivity, specificity and accuracy when comparing the analytical data obtained for oral fluid with those for plasma (i.e. the reference sample), using both cut-off values for the detection of THC in plasma. When using the LOQ of the LC–MS–MS method as cut-off value for plasma samples, the sensitivity and accuracy were >95%. A somewhat lower specificity of 84% was noted. These results indicate a good correlation of oral fluid and plasma samples and are in concordance with previous published reports, although other collection protocols and cut-off values were used [14–17]. However, when applying the legal cut-off value for plasma samples, the specificity decreased dramatically (51.2%) due to a high number of FP results. In this case, a high sensitivity was calculated since there were no FN results. An accuracy of 84.9% was noted. These results indicate a high probability of detecting recent cannabis use, however, a certain number of positive oral fluid samples may not be confirmed in plasma. The usefulness of oral fluid testing for THC is dependent on its ability to serve as a diagnostic indicator of recent cannabis use. In this study, we determined what would be the optimal cut-off for THC in oral fluid, in order to ‘predict’ that the plasma sample would be positive for THC. This optimal cut-off was calculated using ROC curve analysis, with different cut-off values for plasma as reference sample: (A) the LOQ of the analytical method and (B) the legal limit in Belgium. Using the LOQ of the method, the analysis showed an optimal cut-off value at 1.2 ng/mL oral fluid, corresponding to a sensitivity of 94.7% and a specificity of 92%. This cut-off value was somewhat lower in comparison with the SAMHSA proposed confirmation cut-offs in oral fluid, i.e. 2 ng/mL THC [18]. When applying the legal cut-off for plasma samples, a higher optimal cut-off value of 5.2 ng/mL was calculated with a sensitivity of 91.6% and a specificity of 88.6%. 3.3. Evaluation of the on-site Dra ¨ger DrugTest 1 system for the detection of THC in oral fluid The sensitivity, specificity and accuracy for on-site oral fluid analyses with the Dra ¨ger DrugTest 1 for THC are shown in Table 3, with either oral fluid or plasma (using the LOQ of the method or the legal cut-off value) as reference sample. In all cases, a low sensitivity and accuracy were noted (<66%), due to a high number of FN results. However, there is no indication that the FN results correspond to the oral fluid samples with the lowest concentrations of THC. The application of higher confirmation cut-off values will therefore not alter the results significantly. On the other hand, the high number of FN results could be due to a state of dryness in the oral cavity, leading to insufficient sample volume in the sampling sponge. Based on previous reports, M. Laloup et al. / Forensic Science International 161 (2006) 175–179178 Table 2 Prediction of the presence of THC in plasma by LC–MS–MS analysis of the corresponding oral fluid samples (collected with the Intercept 1 device) Plasma:LOQ (%) Plasma:legal cut-off (%) Sensitivity 98.2 100 Specificity 84.0 51.2 Accuracy 95.7 84.9 The sensitivity, specificity and accuracy were calculated based on the number of true positives, true negatives, false positives and false negatives, using plasma as the reference sample. The applied cut-off values were the LOQ of the method for oral fluid (0.5 ng/mL) and either the LOQ of the LC–MS–MS method for plasma (0.5 ng/mL) or the legal limit in Belgium for DUID for THC in plasma (2.0 ng/mL). Table 3 Sensitivity, specificity and accuracy of the on-site Dra ¨ger DrugTest 1 system during roadside controls, calculated based on the number of true positives, true negatives, false positives and false negatives Oral fluid (%) Plasma:LOQ (%) Plasma:legal cut-off (%) Sensitivity 49.5 50.9 57.8 Specificity 100 92.9 87.5 Accuracy 55.0 55.7 65.6 Either oral fluid or plasma were used as reference sample; the applied cut-off values were the LOQ of the method for oral fluid (0.5 ng/mL) and either the LOQ of the LC–MS–MS method for plasma (0.5 ng/mL) or the legal limit in Belgium for DUID for THC in plasma (2.0 ng/mL). the detection of cannabis use in oral fluid appears to be a general problem for most on-site drug tests, due to low sensitivities [8–10,14]. 4. Conclusions The aim of the study was to evaluate (A) oral fluid as a predictor of positive plasma samples and (B) the use of the onsite Dra ¨ger DrugTest 1 system for the detection of THC in oral fluid at roadside controls. Oral fluid (collected with the Intercept 1 device) and plasma samples were analyzed with a validated LC–MS–MS method. The overall results of the study indicated a good accuracy (84.9–95.7% depending on the cutoff used for plasma analysis) when correlating THC detection in oral fluid and plasma, suggesting that oral fluid is a good predictor of actual cannabis influence. However, the Dra ¨ger DrugTest 1 system cannot be recommended for on-site oral fluid tests for THC due to the low accuracy (<66%). The search for an on-site screening method that can provide acceptable accuracies for the detection of THC in oral fluid remains a major hurdle. Acknowledgements We thank the technical staff Malika Bouazzati, Bart Viaene and Rhimou Sebbagh for their practical support during the course of the project. References [1] Annual Report 2004: The state of the drugs problem in the European Union and Norway, European Monitoring Centre for Drugs and Drug Addiction (EMCDDA), http://www.annualreport.emcdda.eu.int/en/homeen.html. [2] J.G. Ramaekers, H.W.J. Robbe, J.F. O’Hanlon, Marijuana, alcohol and actual driving performance, Hum. Psychopharmacol. Clin. Exp. 15 (2000) 551–558. [3] O.H. Drummer, J. Gerostamoulos, H. Batziris, M. Chu, J. Caplehorn, M.D. Robertson, P. Swann, The involvement of drugs in drivers of motor vehicles killed in Australian road traffic crashes, Accid. Anal. Prev. 36 (2004) 239–348. [4] R.S. Niedbala, K.W. Kardos, D.F. Fritch, S. Kardos, T. Fries, J. Waga, J. Robb, E.J. Cone, Detection of marijuana use by oral fluid and urine analysis following single-dose administration of smoked and oral marijuana, J. Anal. Toxicol. 25 (2001) 289–303. [5] E.J. Cone, L. Presley, M. Lehrer, W. Seiter, M. Smith, K.W. Kardos, D. Fritch, S. Salamone, R.S. Niedbala, Oral fluid testing for drugs of abuse: positive prevalence rates by Intercept immunoassay screening and GC– MS–MS confirmation and suggested cut-off concentrations, J. Anal. Toxicol. 26 (2002) 541–546. [6] M.A. Huestis, E.J. Cone, Relationship of D 9 -tetrahydrocannabinol concentrations in oral fluid and plasma after controlled administration of smoked cannabis, J. Anal. Toxicol. 28 (2004) 394–399. [7] A.G. Verstraete, Oral fluid testing for driving under the influence of drugs: history, recent progress and remaining challenges, Forensic Sci. Int. 150 (2005) 143–150. [8] J.M. Walsh, R. Flegel, D.J. Crouch, L. Cangianelli, J. Baudys, An evaluation of rapid point-of-collection oral fluid drug-testing devices, J. Anal. Toxicol. 27 (2003) 429–439. [9] D.J. Crouch, J.M. Walsh, R. Flegel, L. Cangianelli, J. Baudys, R. Atkins, An evaluation of selected oral fluid point-of-collection drug-testing devices, J. Anal. Toxicol. 29 (2005) 244–248. [10] M. Laloup, M.R. Fernandez, M. Wood, G. De Boeck, C. Henquet, V. Maes, N. Samyn, Quantitative analysis of D 9 -tetrahydrocannabinol in preserved oral fluid by liquid chromatography–tandem mass spectrometry, J. Chromatogr. A 1082 (2005) 15–24. [11] P. Kintz, W. Bernhard, M. Villain, M. Gasser, B. Aebi, V. Cirimele, Detection of cannabis use in drivers with the Drugwipe device and by GC– MS after Intercept 1 device collection, J. Anal. Toxicol. 28 (2005) 724– 727. [12] R. Bonfiglio, R.C. King, T.V. Olah, K. Merkle, The effects of sample preparation methods on the variability of the electrospray ionization response for model drug compounds, Rapid Commun. Mass Spectrom. 13 (1999) 1175–1185. [13] N. Samyn, A. Verstraete, C. van Haeren, P. Kintz, Analysis of drugs of abuse in saliva, Forensic Sci. Rev. 11 (1999) 1–19. [14] A.G. Verstraete, M. Puddu, Evaluation of different roadside drug tests, in: A.G. Verstraete (Ed.), Rosita. Roadside Testing Assessment, Rosita Consortium, Gent, 2001, pp. 167–232. [15] N. Samyn, G. De Boeck, A.G. Verstraete, The use of oral fluid and sweat wipes for the detection of drugs of abuse in drivers, J. Forensic Sci. 47 (2002) 1380–1387. [16] S.W. Toennes, G.F. Kauert, S. Steinmeyer, M.R. Moeller, Driving under the influence of drugs-evaluation of analytical data of drugs in oral fluid, serum and urine, and correlation with impairment symptoms, Forensic Sci. Int. 152 (2005) 149–155. [17] S.W. Toennes, S. Steinmeyer, H.J. Maurer, M.R. Moeller, G.F. Kauert, Screening for drugs of abuse in oral fluid—correlation of analysis results with serum in forensic cases, J. Anal. Toxicol. 29 (2005) 22–27. [18] Substance Abuse and Mental Health Administration, Mandatory guidelines for Federal workplace drug testing programs, Fed. Regist. 59 (1994) 29908–29931. M. Laloup et al. / Forensic Science International 161 (2006) 175–179 179 Journal of Chromatography A, 1130 (2006) 3–15 Review Recent applications of liquid chromatography–mass spectrometry in forensic science Michelle Wooda,b,∗, Marleen Laloup c, Nele Samync, Maria del Mar Ramirez Fernandezc, Ernst A. de Bruijnb, Robert A.A. Maesb, Gert De Boeckc aWaters Corporation, MS Technologies Centre, Micromass UK Ltd, Atlas Park, Simonsway, Wythenshawe, Manchester M22 5PP, UK bUtrecht Institute of Pharmaceutical Sciences (UIPS), Department of Human Toxicology, University of Utrecht, Sorbonnelaan 16, 3584 CA Utrecht, The Netherlands cNational Institute of Criminalistics and Criminology (N.I.C.C.), Section Toxicology, Vilvoordsesteenweg 98, 1120 Brussels, Belgium Available online 22 May 2006 Abstract Recent years have seen the development of powerful technologies that have provided forensic scientists with new analytical capabilities, unimaginable only a few years ago. With liquid chromatography–mass spectrometry (LC–MS) in particular, there has been an explosion in the range of new products available for solving many analytical problems, especially for those applications in which non-volatile, labile and/or high molecular weight compounds are being analysed. The aim of this article is to present an overview of some of the most recent applications of LC–MS (/MS) to forensic analysis. To this end, our survey encompasses the period from 2002 to 2005 and focuses on trace analysis (including chemical warfare agents, explosives and dyes), the use of alternative specimens for monitoring drugs of abuse, systematic toxicological analysis and high-throughput analysis. It is not the intention to provide an exhaustive review of the literature but rather to provide the reader with a ‘flavour’ of the versatility and utility of the technique within the forensic sciences. © 2006 Elsevier B.V. All rights reserved. Keywords: LC–MS; Forensic; Trace chemicals; Alternative specimens; Drugs of abuse; Systematic toxicological analysis Contents 1. Introduction............................................................................................................... 4 2. Trace chemicals ........................................................................................................... 4 2.1. Chemical warfare agents ............................................................................................. 4 2.2. Explosives.......................................................................................................... 5 2.3. Dyes ............................................................................................................... 6 3. Drugs of abuse in alternative matrices ....................................................................................... 6 3.1. Hair ............................................................................................................... 7 3.2. Oral fluid ........................................................................................................... 8 3.3. Meconium.......................................................................................................... 8 3.4. Post-mortem alternative specimens .................................................................................... 9 4. Systematic toxicological analysis using liquid chromatography–mass spectrometry............................................... 9 4.1. Targeted screening using liquid chromatography–tandem mass spectrometry .............................................. 9 4.2. Liquid chromatography–mass spectrometry with reference libraries ..................................................... 10 4.3. Liquid chromatography–(tandem) mass spectrometry with exact mass ................................................... 10 4.4. Data (or information)-dependent acquisition .......................................................................... 11 4.5. Sample preparation: considerations for systematic toxicological analysis................................................. 11 ∗Corresponding author. Tel.: +44 161 435 4100; fax: +44 161 435 4444. E-mail address: michelle [email protected] (M. Wood). 0021-9673/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2006.04.084 4M. Wood et al. / J. Chromatogr. A 1130 (2006) 3–15 5. High-throughput liquid chromatographic–(tandem) mass spectrometric analysis ................................................ 12 6. Conclusion .............................................................................................................. 13 References .............................................................................................................. 13 1. Introduction The term “forensic science” covers those professions which are involved in the application of the social and physical sciences to the criminal justice system. Forensic experts are required to explain the smallest details of the methods used, to substantiate the choice of the applied technique and to give their unbiased conclusions—all under the critical and often mistrustful gaze of the servants of the justice, as well as the general public and the media. The final result of the work of the forensic scientist exerts a direct influence on the fate of a given individual. This burden is a most important stimulus, and one which determines the way of thinking and acting in forensic sciences. Consequently, the methods applied in forensic laboratories should assure a very highlevel of reliability and must be subjected to extensivequality assurance and rigid quality control programs. The legal system is based on the belief that the legal process results in justice. This has come under some question in recent years. Of course, the forensic scientist cannot change scepticism and mistrust singlehandedly. He or she can, however, contribute to restoring faith in the judicial processes by using science and technology in the search for facts in civil, criminal and regulatory matters. Recent years have seen the development of powerful technologies that have provided forensic scientists with new analytical capabilities which were unimaginable only a few years ago. The ability of mass spectrometry (MS) to extract chemical fingerprints from microscopic levels of analyte is invaluable in this quest, enabling the legally defensible identification and quantification of a wide range of compounds. Gas chromatography (GC)–MS, liquid chromatography (LC)–MS, isotope ratio (IR)- MS and inductively coupled plasma (ICP)-MS have become routine tools to enable detection and characterization of minute quantities in what can often be very complex matrices. In the case of LC–MS, the last two decades have seen some significant developments and improvements in instrumentation design. Particularly noteworthy has been the introduction of robust, user-friendly interfaces such as those based on atmospheric pressure ionisation techniques, e.g. electrospray (ESI) and atmospheric pressure chemical ionisation (APCI). Consequently, many analysts and laboratories are finally at the point where they are considering the acquisition of LC–MS capabilities. According to Willoughby et al. [1] LC–MS has progressed from the “innovators” stage through the “early adaptors”, to the “early majority” stage and is now open to specialists from a variety of disciplines, especially for those applications where involatile, labile and/or high molecular weight compounds are being analyzed. The purpose of this article is to review some of the most recent applications of LC–MS (/MS) to forensic analysis with special focus on the following; trace analysis, the use of alternative specimens for monitoring drugs of abuse, systematic toxicological analysis and high-throughput analysis. 2. Trace chemicals 2.1. Chemical warfare agents Determining the use of chemical warfare agents (CWAs) in times of war or in acts of terrorism requires rapid and reliable methods. The sarin gas attacks by a Japanese cult in Matsumoto city (1994) and the Tokyo subway system (1995) represented the first cases in which a CWA was indiscriminately released against a civilian population [2]. The latter incident resulted in the deaths of 12 people and led over 5000 to seek medical attention. Nerve agents are extremely potent organophosphorus compounds that cause biological effects by irreversibly inhibiting the enzyme acetylcholinesterase (AChE). To confirm exposure, biological samples, e.g. urine, can be analysed for the agents themselves, their metabolites or their degradation products. Nerve agents are rather volatile compounds, thus analysis by GC–MS might be considered the obvious choice. However, in an aqueous environment, these agents readily hydrolyse to produce alkyl alkylphosphonates (RMPAs), these in turn can be further hydrolysed to methyl phosphonate (MPA) (Fig. 1). LC–MS is increasingly being used for these low molecular weight, highly polar compounds whilst exploiting the benefits over GC–MS, of reduced sample handling and no requirement for derivatisation [3–5]. Hayes et al. [6] recently developed LC–tandem MS (LC–MS/MS) methods for the analysis of the short-lived metabolites of several CWAs including; sulfur mustard, sarin, soman, cyclohexyl methylphosphonofluridate (GF) and O-ethyl S-2-diisopropylamino ethyl methylphosphonothioate (VX) in urine. These methods were also used to determine the feasibility of using saliva as a complementary or alternative matrix to urine; this could be a particularly valuable approach to assess the exposure of young children, where collection of a urine sample on demand is often difficult. VX comprises a mixture of two enantiomers which demonstrate significant differences in the rate of AChE inhibition and overall toxicity. Thus, the ability to distinguish between them is desirable for toxicological studies and for the development of antidotes. Smith [7] has used normal-phase LC in conjunction with MS detection for this purpose. LC–MS has also been used to investigate the longer-lived metabolites. Several groups have used LC–MS to determine the metabolites of sulfur mustard, i.e. the ␤-lyase metabolites in urine samples from human casualties after sulphur mustard poisoning [8,9]. In the case of large-scale attacks, analysis of the environment and other materials may also be required. Hancock and D’Agostino have developed a LC–ESI-MS (/MS) procedure which allows the identification of a munitions grade sample of tabun, sarin, soman, GF and the nerve agent stimulant triethyl phosphate (TEP) on manmade fibres [10]. Although this technique uses only minimal sample preparation the same group M. Wood et al. / J. Chromatogr. A 1130 (2006) 3–15 5 Fig. 1. Hydrolysis pathway of nerve gases and the alkyl alkylphosphonates (RMPAs) namely, isopropyl (IMPA), pinacolyl (PMPA) and ethyl (EMPA) methyl phosphonates. In turn, these can be hydrolysed to methyl phosphonate (MPA). have more recently experimented to omit sample preparation completely and to allow the direct analysis of TEP collected on solid-phase microextraction (SPME) fibres [11]. The biotoxin ricin originates from the seeds (castor beans) of the Ricinus communis plant and is extremely toxic (human LD50 estimated at 3–30 ␮g/kg by inhalation or ingestion, respectively) [12]. It has the unique position of being the only protein listed under the Chemical Weapons Convention and is of forensic interest due to its potential for terrorist use or as a homicide agent [13]. Due to the high molecular weight of this compound (66 kDa) absolute structural elucidation of the intact protein is not possible using nominal mass analysis. However, several groups have used a preliminary enzymatic digestion to convert the protein into intermediate molecular weight peptides followed by LC–MS (/MS) using a hybrid quadrupole time-of-flight (QTOF) instrument [12,14]. The methods were used to characterise purified ricin from several different varieties of R. communis and also from crude castor bean extracts. 2.2. Explosives The analysis of trace levels of explosives is critical in crime scene forensic investigations, homeland security and environmental analysis. LC–MS is a well-established technique for explosives in associated complex matrices such as post-blast residues and in environmental samples such as soil and plant material extracts [15,16]. Although these compounds have a low vapour pressure they tend to be heat labile and can degrade at the high temperatures typically used in GC injectors. Thus, LC–MS is particularly well-suited to the analysis of these relatively polar molecules, heat labile compounds. Many of the methods rely on the formation of cluster or adduct ions for identification. Gapeev et al. [17] studied the formation of cluster ions of 1,3,5-trinitro1,3,5-triazacyclohexane (RDX), one of the most commonly used military explosives in both ESI and APCI. Results showed that in ESI, self-decomposition of RDX did not play a role in adduct formation; the adducts were produced from impurities present in the mobile phase at ppm levels. In contrast, with APCI, part of the RDX molecule decomposes yielding a NO2−species; this in turn clusters with other RDX molecules. More recently, Mathis and McCord presented a comprehensive method to allow the screening of a panel of high explosives including; RDX, 2,4,6,-trinitrotoluene (TNT), pentaerythritol tetranitrate (PETN), 1,3,5,7-tetramethylene-2,4,6,8tetranitramine (HMX), nitrogycerine (NG) and ethylene glycol dinitrate (EGDN). This method was based on the competitive formation of adducts following infusion of the high explosives with a mixture of four anions; chloride, formate, acetate and nitrate. Information relating to the relative extent of adduct formation (based on intensity ratios) in addition to adduct stability, was used to provide a multiplexed detection scheme (Fig. 2)[18]. Anti-personnel (AP) mines are currently in place in over seventy countries and are designed to maim or kill humans. In addition to the lives that are lost, the mere suspicion that they may be present, can prevent the use of large areas which could otherwise be utilised for agriculture or social infrastructure. Removal of landmines from such areas is known as humanitarian de-mining and relies on the accurate detection of the explosive. A potentially useful approach and one which is currently under investigation, is the detection of the chemical vapours which arise from the explosives and are transported into the surrounding atmosphere. High sensitivity is required since the 6M. Wood et al. / J. Chromatogr. A 1130 (2006) 3–15 Fig. 2. (a) Mass spectrum of a mixture of high explosives containing EGDN, NG, TNT, PETN, RDX and HMX in 50% MeOH/50% aqueous mixture with 0.3 mM ammonium chloride, ammonium formate and ammonium nitrate. (b) Relative stability of high explosive adducts of chloride, formate, acetate and nitrate using negative ion ESI-MS. From Ref. [18] with permission. concentration of molecules expected to reach the gas phase is low. Sanchez et al. [19] have developed a method for the sampling and identification of nitroaromatic explosives. Air was sampled at flow rates of up to 15 L/min using a holder fitted with a C18 solid-phase extraction (SPE) membrane. After sampling, trapped analytes were desorbed on-line and analysed by LC–MS/MS using an APCI interface. Storage stability studies indicated that the captured analytes were stable for 1 week or 3 weeks, when membranes were stored at room temperature or at −4◦C, respectively. The method allowed the identification and separation of most of the isomers of TNT and 2,4-dinitrotoluene (DNT); limits of detection were in the range of femtogram/L. The method is suitable for the chemical profiling of militarygrade explosives and is valuable for both forensic identification and for de-mining purposes. 2.3. Dyes Textile fibres found at a crime scene can be used as chemical evidence in a wide range of crimes; dye identification and comparison can be of particular importance. Recently, Huang et al. [20] have used LC–MS to enable unambiguous differentiation between structurally related textile dyes which were previously indistinguishable by UV–vis absorption profile or by microspectrophotometry. They concluded that where singlestage LC–MS fails to differentiate, analysis should be extended to include LC–MS/MS of the extracted dye mixture. The group of colour additives known as the Sudan dyes are synthetically produced azo-dyes. Their degradation products are considered to be carcinogens and teratogens. Due to this fact their use as food additives is banned in the USA and the European Union (EU). However, in some countries they are still used to enhance the colour of bell pepper and chilli powders. The discovery of a batch of chilli powder contaminated with Sudan I in February 2005 resulted in the largest product recall in British history [21]. The widespread use of this batch of chilli powder led to the withdrawal of hundreds of food products including Worcester sauce, pizza and seafood sauces. Calbiani et al. [22] reported a LC–MS/MS (nominal mass/low resolution triple quadrupole) method for the simultaneous analysis of four Sudan dyes in foodstuffs (Fig. 3). More recently, this group have used capillary LC in conjunction with high-resolution MS instrumentation to further distinguish between isobaric ions and to further increase confidence by providing elemental composition [23]. Using exact mass in both MS and MS/MS experiments, they were able to provide unambiguous confirmation of Sudan I in authentic food samples. Pepper sprays are readily available to law-enforcement personnel and to the general public for a variety of uses including riot control and self-defence. In these cases, the presence of pepper sprays, on clothing for example, may help to determine the facts of an incident. A common active ingredient of these sprays is capsaicin, an oily resin extracted from capsicum fruits. Some of the pepper sprays also contain a coloured dye or a UV-activated fluorescent marker to permit the localisation of the product. However, there are now a number of products on the market that do not contain such visible aids to analysis. Cavett et al. [24] have developed a method to initially visualise colourless pepper sprays on fabric and to subsequently confirm the presence of naturally occurring and synthetic capsaicinoid molecules. Visualisation was achieved by chemical derivatisation of the capsaicinoids using a diazonium salt. Identification of the capsaicinoids and their derivatives was then accomplished following methanolic extraction from the garment. Extracts were analysed within 6.5 min, using a YMC Basic column in conjunction with LC–APCI-MS detection. Work is on-going to confirm the spectra and proposed fragment ions of the derivatives via MS/MS and exact mass determination. 3. Drugs of abuse in alternative matrices For the detection of illicit drugs, plasma and urine are currently the most common matrices investigated. However, over the past few years there has been an increased interest in the use of more convenient, less invasive specimens, e.g. hair, oral fluid and sweat, to document drug use and exposure [25,26]. Indeed in April 2004, the US Department of Health and Human Services proposed new guidelines for the use of these alternative specimens as an adjunct to urine, for the testing of employees in a number of situations including; pre-employment, random, reasonable cause and post-accident testing [27]. For these samples, collection is relatively easy to perform and requires no M. Wood et al. / J. Chromatogr. A 1130 (2006) 3–15 7 Fig. 3. LC–ESI–MS/MS MRM traces obtained from 125 ␮g/L standard solution of (a) Sudan I; (b) Sudan II; (c) Sudan III; (d) Sudan IV; (e) Disperse Orange 13 internal standard (100 ␮g/L; left column) and from a blank chilli tomato and cheese sauce sample spiked with 125 ␮g/L each (1685 ␮g/kg sample) of (f) Sudan I; (g) Sudan II; (h) Sudan III; (i) Sudan IV; (j) Disperse Orange 13 internal standard (100 ␮g/L; right column). From Ref. [22] with permission. special equipment or facilities. Furthermore, collection can be supervised, thus reducing the opportunity for sample adulteration. One of the main disadvantages however, of using these alternatives is that the volume or amount of sample is usually limited, consequently highly sensitive confirmatory techniques such as LC–MS/MS become a necessity. 3.1. Hair In addition to the convenience of sample collection, any drugs and metabolites incorporated into hair, tend to persist much longer than in conventional specimens. Recently, hair has been used to document drug exposure in a variety of scenarios such as forensic and workplace testing [28–30], to monitor compliance to drug therapy [31,32] and particularly for investigating cases of drug-facilitated crimes (DFC) [33–43]. The availability of standard reference materials for drugs of abuse in hair is vital and enables those laboratories performing hair analysis to check the accuracy of their methods [44]. Over the last few years DFC, e.g. sexual assault and robbery, have been increasing; these crimes are often difficult to prove due to factors such as the low concentrations of drugs used, or their rapid clearance from the body. In addition, many victims of DFC do not report an incident until several days later, often due to the amnesia caused by the drug. Hence, conventional specimens such as blood or urine may have limited value. Hair samples have been successfully used to document cases of DFC involving a variety of drugs including; benzodiazepines and the hypnotics (zolpidem and zopiclone), methadone and buprenorphine [33–43]. Kintz and co-workers concluded that due to the extremely low concentrations of drugs typically encountered in hair analysis (low pg/mg) the “sensitivity of LC–MS/MS appears to be a pre-requisite to document any case involving drug-facilitated sexual assault”. However, they also added the caveat that hair analysis should not simply be considered as an alternative to blood and urine testing but as a complementary technique where possible. The importance of this was revealed in a controlled study to investigate the window of detection for lorazepam in urine, oral fluid and hair [39]. Following a single (2.5 mg) dose, the drug could be still be detected in urine and oral fluid for 144 and 8 h, respectively, after dosing. However, they were unable to detect lorazepam in hair samples collected 4 weeks after administration. Cheze et al. [42] used LC–MS/MS to conduct a survey into the drugs most commonly used to commit DFC in Paris over the period from June 2003 to May 2004. Out of the total of 128 cases investigated, 18% were proven DFC cases and they found a high prevalence of zolpidem and clonazepam, followed by bromazepam, nordazepam and midazolam (Fig. 4). Laloup et al. [43] recently reported a LC–MS/MS method for the simultaneous analysis of 26 benzodiazepines and metabolites, zolpidem and zopiclone in blood, urine and hair. The Fig. 4. Distribution of benzodiazepines and benzodiazepine-like hypnotics in 23 cases of proved DFC. From Ref. [42] with permission. 8M. Wood et al. / J. Chromatogr. A 1130 (2006) 3–15 method was applied to authentic samples from both clinical and forensic cases, including the analysis of hair from a woman who claimed to have been drugged and sexually abused over a period of several years. Thirty-three centimetre lengths of hair were submitted for analysis and cut into 1–3 cm sections; all segments were found to be positive for more than one benzodiazepine, indicating multiple drug exposure, with higher concentrations closer to the root. These results demonstrated the utility of hair to provide a long-term drug history. 3.2. Oral fluid The use of oral fluid as an alternative specimen is also increasing in popularity especially for monitoring recent drug use within the workplace, at the roadside, in prisons and to check compliance to medication. Concheiro et al. [45] developed a method for the quantification of the active constituent of cannabis, i.e. 9tetrahydrocannabinol (9-THC) in oral fluid. Samples were collected by spitting into polypropylene tubes. Two hundred microlitres of sample was processed using liquid/liquid extraction (LLE) with hexane followed by analysis using LC–MS. Limits of detection of 2 ␮g/L were achieved. Wood et al. [46] reported a validated method for the simultaneous analysis of six amphetamines in oral fluid (also collected by expectoration). The procedure required only 50 ␮L of sample to achieve limits of detection of 2 ␮g/L or better and comprised rapid and simple sample preparation, i.e. protein precipitation (PPT) using methanol followed by LC–MS/MS. Dams et al. [47] described a method for methadone and multiple illicit drugs in addition to their metabolites in oral fluid. Their method also involved PPT using acetonitrile followed by LC–MS/MS analysis. The method proved useful for determining methadone concentrations in pregnant opiate and/or cocaine addicts. Although the methods referenced above utilized oral fluid that has been collected by expectoration, it should be noted that the increased interest in oral fluid has also been accompanied by an increase in the availability of specialized collection devices; these promise a simplified, more controllable collection and sample stability. The final choice of oral fluid collection system, however, has been shown to have serious implications on drug analysis [48–50]. The Intercept is a US Food and Drugs Administration (FDA) approved sampling device that is used on a large scale in the USA for workplace drug testing and is one of the devices currently under investigation in a joint roadside study between the EU and the USA to detect driving under the influence of drugs [51]. The collection system contains additives which can cause problems, e.g. ion suppression during LC–MS/MS analysis in the absence of a suitable cleanup method. Several groups have employed LLE (with hexane) to prepare the so-called ‘preserved oral fluid’ specimen prior to analysis; drugs of interest have included 9-THC, benzodiazepines and hypnotics [52,53]. A SPE method has also been developed, which is combined with LC–MS/MS to allow the simultaneous determination of a panel of common basic illicit drugs [50]. Work is underway to extend the current panel of analytes to include 9-THC [54]. Other groups have used different devices to collect samples for the purpose of drug monitoring. Wylie et al. [55] developed a method for the analysis of 49 licit and illicit drugs in oral fluid collected using the Omni-Sal device. Samples were extracted using SPE and then analysed by LC–MS/MS and GC–MS. Recently, Teixeira et al. [49] used the Salivette device to collect samples and to quantify 9-THC in oral fluid samples following SPE and LC–MS. A method was developed for the separation of the enantiomers of methadone and its metabolite EDDP in saliva [56]. Methadone is administered therapeutically as a racemic mix, i.e. a 50:50 mix of the enantiomers. There are significant differences between the enantiomers in terms of receptor affinity, analgesic potency and pharmacokinetic profiles. Thus, therapeutic monitoring of this agent and its metabolite requires an enantioselective technique. Samples were collected using the Salivette device. Following centrifugation, analytes were separated using an immobilized ␣1-acid glycoprotein chiral stationary phase (AGP-CSP) in conjunction with MS detection. The optimized and validated method was applied to the analysis of samples collected from patients following a methadone maintenance program. 3.3. Meconium Drug abuse during pregnancy is a major problem and has been associated with prenatal complications and high morbidity and mortality rates of newborns. Some birth defects are thought to be related to fetal exposure to drugs. Detection of in utero drug exposure has traditionally been accomplished by urine drug testing. However, this only reflects maternal drug use over the last 3–4 days and abstinence of the mother for several days prior to delivery, may produce a negative result. Monitoring exposure through testing of alternative matrices, such as neonatal meconium and hair, offers advantages including non-invasive collection and detection earlier in gestation [57–59]. Meconium is the first fecal matter produced by the neonate typically within the first 5 days after birth. Since the formation of meconium starts between the 12th and 16th week of gestation and accumulates in the fetal bowel until birth, use of this specimen can extend the window of drug detection considerably, i.e. to approximately the last 20 weeks of pregnancy. Pichini et al. have described methods for the analysis of opiates and cocaine and respective metabolites [60] and for the analysis of amphetamine derivatives in this specimen [61].In both cases samples were prepared by SPE and analysis was achieved using LC–MS (three qualifying ions per compound). Sensitivity was sufficient to allow the detection of all drugs in the low nanograms per gram meconium. Another report describes the application of LC–MS/MS for the simultaneous quantification of methadone and its metabolites in meconium after methanolic extraction followed by SPE [62]. This method represents an improvement over previous methods in terms of sensitivity and specificity and was successfully applied for the quantification of these compounds in meconium from infants whose mothers were maintained on methadone during pregnancy. M. Wood et al. / J. Chromatogr. A 1130 (2006) 3–15 9 3.4. Post-mortem alternative specimens Clearly, alternative specimens can prove invaluable for the documentation of drug use in the living person. This can also be true for post-mortem investigations. Toxicological analysis of the usual post-mortem specimens can often pose special difficulties. This may because of the decomposed nature of the specimens themselves and/or the presence of putrefactive compounds. In the absence of any suitable tissues or fluids, insects have been proposed as reliable alternate specimens and indeed have been used to identify the presence of various drugs within the cadaver [63,64]. Although the involvement and contribution of the identified drugs to the actual death may be questionable, the insect tissues have, nevertheless, proved a useful sample. Wood et al. [65] presented a method for the simultaneous analysis of 10 benzodiazepines in larvae and puparia of the Calliphora vicina (Diptera, Calliphoridae). Benzodiazepines are the most widely prescribed psychoactive active drugs in the world. However, they are frequently misused and are consequently often encountered in post-mortem analysis. Larvae were prepared by homogenization followed by precipitation using acetonitrile. Puparia were pulverized in a ball mill and then extracted by ultrasonification in methanol. All extracts were subsequently analysed using LC–MS/MS. The utility of this method was confirmed through its application to the analysis of larvae and puparia that had been reared on media spiked with a range of concentrations of nordiazepam. The concentrations were equivalent to those expected in skeletal muscle following fatal human overdoses. Both the parent drug and its metabolite oxazepam could be detected in single larvae or puparia. Pien et al. [66] extended these preliminary studies to investigate the effects of different concentrations of nordiazepam on larval development and growth. Larval development can be used in the estimation of post-mortem interval. In some cases, the presence of drugs has been shown to affect development of the insect, consequently these disturbances can have serious implications on the accuracy of post-mortem interval calculations. 4. Systematic toxicological analysis using liquid chromatography–mass spectrometry The ability to screen for a large number of unknown analytes in human samples is of interest to many areas of society including hospital emergency departments, forensic pathologists, police/prison officials and employers. Currently, the ‘gold standard’ for screening is GC–MS, owing in part, to the specificity and sensitivity of the technique, but also as a result of the availability of large libraries of standardized electron ionisation (EI) mass spectra. However, since GC–MS is limited to the analysis of volatile and thermally stable compounds, and because the technique usually requires a specific derivatisation for polar analytes, alternative methods have been investigated. One of the most important questions at present is: To what extent can LC–MS be applied to the search for unknown substances? Although its use as a confirmatory tool is becoming more widely accepted, its use for screening purposes is still not fully established. This is reflected in the many different LC–MS strategies currently being investigated for systematic toxicological anaysis (STA). 4.1. Targeted screening using liquid chromatography–tandem mass spectrometry One approach for screening is to use LC-multiple (or selected) reaction monitoring (MRM or SRM) analysis. The clear advantage of MRM analysis is in its specificity (sensitivity); a precursor ion from the targeted molecule is selected by the first quadrupole or mass filter, fragmented in the collision cell and then a structurally significant or specific product ion (or ions) selected by the second quadrupole filter. For confidence in identification, several MRM channels should be included per compound and their ratios calculated (and compared to standards). The main drawback in terms of screening, is that the technique relies on the selection of a pre-defined precursor ion in the first quadrupole. Clearly, this is not applicable to the analysis of complete unknowns but rather is a method targeted against a panel of known/expected compounds. Nevertheless, this approach has been used successfully for single or multiple drug classes in a variety of biological matrices. However, due to the relatively low number of analytes often included, its utility for screening of real toxicological samples may be limited. In an attempt to circumvent this, investigators have attempted to create targeted methods for much larger panels of drugs. In 2003, Gergov et al. [67] reported a qualitative screening method for 238 drugs in blood. Following a two-step LLE procedure, compounds were separated on a C18 column. For each compound, identification was based on a single MRM channel and retention time (tR). For 80% of the drugs investigated, the analytical sensitivity was sufficient to detect at therapeutic concentrations in blood. Although clearly a useful method, the high number of analytes meant that dwell times had to be low (25 ms), typically resulting in ∼4 data points across a chromatographic peak; whilst suitable for qualitative purposes, this is insufficient for accurate quantification which really requires 10–15 data points. Furthermore, such limitations on dwell times also meant that any attempts to try to increase the panel of drugs further, would be problematic without providing better chromatographic separation between compounds or by using instrumentation with faster scan capabilities. M¨ ueller et al. [68] recently presented a novel targeted method for the analysis of 301 compounds. In this procedure an initial ‘survey’ scan was performed which involved monitoring 301 MRM channels. An intensity threshold was set and if any MRM exceeded this pre-set threshold, the instrument was instantly switched into ‘enhanced parent ion’ (EPI) scan mode. Product ions were generated by acquisition at three different collision energies, i.e. 20, 35 and 50 eV and the resultant spectra subsequently searched against a database. The developed method was successfully applied to blood and urine from forensic cases. This procedure also met with the same limitations as the previous method, i.e. utilisation of the short dwell times means that this technique is currently limited to ∼300 analytes; better chromatographic separation will be required if this is to be increased in the future. A liquid chromatography–tandem mass spectrometry method was developed for the simultaneous quantification of 26 benzodiazepines and metabolites, zolpidem and zopiclone, in blood, urine, and hair. Drugs were extracted from all matrices by liquid–liquid extraction with 1-chlorobutane. Chromatography was achieved using a XTerra MS C18 column eluted with a mixture of methanol and formate buffer. Data were acquired using positive electrospray ionization and multiple reaction monitoring using one precursor ion/product ion transition per compound. Quantification was performed using 13 deuterated analogues. Further confirmation of the identity of the compounds was achieved through a second injection of positive samples, monitoring two transitions per compound. The limits of quantification for all benzodiazepines ranged from 1 to 2 ng/mL in blood, 10 to 25 ng/mL in urine, and 0.5 to 10 pg/mg in hair. Linearity was observed from the limit of quantification of each compound to 200 ng/mL, 1000 ng/mL, and 1000 pg/mg for blood, urine, and hair, respectively (r 2 > 0.99). Precision for quality control samples, spiked at three concentrations, was calculated (CV < 20% in most cases). Extraction recoveries for the three matrices ranged from 25.1 to 103.8%, except for one compound (cloxazolam in urine). Ion suppression was studied for all matrices. The validated assay was applied to authentic blood, urine, and hair samples from forensic cases. Introduction Benzodiazepines are a large class of prescribed psychoactive drugs used widely for different medical conditions such as the symptomatic treatment of anxiety and sleep disorders, the treatment of anxiety-related conditions, and as anti-convulsants (1). Benzodiazepines bind with high affinity to the - s u bunit of the GABA A receptor in the central nervous system. Zolpidem and zopiclone are structurally different from the benzodiazepines, but also bind to the GABA A receptor. They provoke the same (side-) effects as the benzodiazepines, including sedation and hypnosis. Chronic use can generate tolerance to the effects. Because of their wide therapeutic index, benzodiazepines have a low risk of serious adverse reactions and toxicity. Unf o r t u n a t e l y, misuse of these compounds is often reported (1–4), and they are often implicated in connection with various types of crime, such as murder and drug-facilitated assault (5), in suicide attempts (6), and, due to their interference with cognitive and psychomotor functions, in road traffic accidents (7,8). Consequently, benzodiazepines are frequently encountered in clinical and forensic toxicology cases. Blood and urine are the conventional specimens to document drug exposure. Hair analysis has proven to be a reliable indicator of past drug abuse, as a complement to blood or urine analysis, for proving or excluding chronic drug use, or, at least, exposure to drugs; therefore it becomes highly useful in monitoring long-term histories of drug abuse (9). As a biological matrix, hair offers particular advantages: it can be easily obtained without violating individual privacy and, due to its stability, it can be stored and transported without requiring specific precautions. However, the extraction and detection of benzodiazepines from hair samples is relatively difficult due to the low levels of drugs that are incorporated into the hair. Therefore, highly sensitive analytical techniques are required for trace-level quantification of benzodiazepines. A number of studies have been reported on the qualitative and quantitative analysis of benzodiazepines and their metaboValidation of a Liquid Chromatography–Ta n d e m Mass Spectrometry Method for the Simultaneous Determination of 26 Benzodiazepines and Metabolites, Zolpidem and Zopiclone, in Blood, Urine, and Hair Marleen Laloup 1, *, Maria del Mar Ramirez Fernandez 1 , Gert De Boeck 1 , Michelle Wood 2 , Viviane Maes 3 , and Nele Samyn 1 1 Federal Public Service Justice, National Institute of Criminalistics and Criminology (N.I.C.C.), Brussels, Belgium; 2 Waters Corporation, MS Technologies Centre, Manchester, U.K.; and 3 Department of Clinical Chemistry-Toxicology, Academic Hospital, Free University of Brussels, Brussels, Belgium Reproduction (photocopying) of editorial content of this journal is prohibited without publisher’s permission. 616 Journal of Analytical Toxicology, Vol. 29, October 2005 * Author to whom correspondence should be addressed: Marleen Laloup, National Institute of Criminalistics and Criminology (NICC), Section Toxicology, Vilvoordsesteenweg 98, 1120 Brussels, Belgium. E-mail: [email protected]. Abstract Journal of Analytical Toxicology, Vol. 29, October 2005 617 lites in different biological matrices. Several techniques have been used for the toxicological analysis of benzodiazepines in blood, urine, or hair, including immunoassays (10–14), gas chromatography (GC) (11,14–19), and high performanceliquid chromatography (HPLC) (11,15,20–27). However, most methods covered only a single substance or mixtures of a few benzodiazepines in only one biological matrix. Until now, only three reports dealing with the simultaneous analysis of a large series of benzodiazepines have been published. Pirnay et al. were the first to publish a GC–mass spectrometry (MS)–MS ion trap method covering 22 benzodiazepines in urine and blood extracts after trimethylsilylation of the drugs (28). H o w e v e r, due to partial or full thermal degradation of some benzodiazepines during the injection step [a phenomenon already reported by Japp et al. and Weston et al. (29,30)], decomposition products had to be taken into account in the method. Detection thresholds for this method were in the range of 10–500 ng/mL, except for the triazolo-benzodiazepines (alprazolam, estazolam, and triazolam), for which the detection threshold was 1000 ng/mL, due to poor chromatographic resolution. The ability of HPLC to separate a large range of underivatized substances, coupled with the milder working conditions of the technique, makes it particularly valuable for the analysis of some of the more thermolabile benzodiazepines. Kratzsch et al. reported on a validated LC–atmospheric pressure chemical ionization-MS method for the screening of 23 benzodiazepines (their antagonist flumazenil and zaleplon), zolpidem, and zopiclone in plasma using liquid–liquid extraction (31). After screening and identification in the scan mode using an LC–MS library, the analytes were quantified in the selected-ion monitoring mode. The limit of quantification (LOQ) ranged from 0.5 to 200 ng/mL. Smink et al. described an LC–MS(–MS) ion trap method after solid-phase extraction for the determination of 33 benzodiazepines, metabolites, and benzodiazepine-like substances in whole blood (32). The LOQ for this method ranged from 0.4 to 41.9 ng/mL. In this report, a validated and highly sensitive LC–electrospray ionization Table I. MRM Transitions and Conditions for all Compounds and Their Deuterated Analogues* Precursor Product Cone Collision Compound ion (m/z) ion (m/z) voltage (V) energy (eV) 7-Aminoclonazepam 286.00 120.80 40 28 250.10 40 20 7-Aminoflunitrazepam 284.00 134.80 40 28 227.00 40 25 Bromazepam 315.90 181.90 40 30 209.00 40 25 Clonazepam 315.90 214.00 38 35 270.00 38 25 Flunitrazepam 314.00 239.10 43 35 268.10 43 25 Clobazam 301.00 224.00 35 33 259.00 35 20 Desmethylflunitrazepam 300.00 198.00 40 38 254.10 40 25 Estazolam 295.00 205.00 35 40 267.10 35 23 Nitrazepam 282.00 180.00 38 35 236.10 38 25 Alprazolam 309.00 205.00 45 43 281.10 45 43 Chlornordiazepam 304.90 139.80 45 30 206.00 45 35 Temazepam 300.90 255.00 28 20 283.00 28 15 Desalkylflurazepam 289.00 139.80 43 28 226.00 43 28 Oxazepam 287.00 241.00 27 22 269.00 27 15 Nordiazepam 271.00 139.80 43 28 164.80 43 28 Brotizolam 394.90 279.10 40 28 314.00 40 23 Triazolam 342.90 308.10 45 25 315.00 45 30 Lormetazepam 334.90 176.90 25 40 289.00 25 22 Lorazepam 320.90 229.00 30 30 275.00 30 22 Prazepam 325.00 139.80 40 38 271.10 40 22 Clotiazepam 319.00 153.80 40 28 291.10 40 23 Tetrazepam 289.00 225.10 45 28 253.10 45 25 Diazepam 285.00 153.80 40 28 193.00 40 30 Loprazolam 465.00 84.70 48 23 110.80 48 25 Flurazepam 388.00 288.10 35 25 315.10 35 23 Cloxazolam 349.00 139.80 40 38 164.80 40 35 Zolpidem 308.10 91.70 45 50 235.10 45 35 continued * Underlined transitions were used for quantification. Journal of Analytical Toxicology, Vol. 29, October 2005 618 (ESI)-MS–MS method is described for the quantification of 26 commonly encountered benzodiazepines and their metabolites, zolpidem and zopiclone, in three different biological matrices (i.e., blood, urine, and hair). This method covers more than 85% of all benzodiazepines and benzodiazepine-like substances registered for the Belgian market and was applied to authentic blood, urine, and hair samples from clinical and forensic cases. Experimental Chemicals Individual ampoules of 7-aminoclonazepam, 7-aminoflunitrazepam, bromazepam, clonazepam, flunitrazepam, clobazam, desmethylflunitrazepam, estazolam, nitrazepam, alprazolam, Zopiclone 277.00 111.80 35 43 217.00 35 25 7-Aminoclonazepam-d 4 290.00 120.80 40 30 7-Aminoflunitrazepam-d 7 291.10 137.90 40 28 Clonazepam-d 4 320.00 274.10 40 25 Flunitrazepam-d 7 321.10 275.20 40 25 Alprazolam-d 5 314.00 286.10 45 25 Temazepam-d 5 305.90 260.10 25 23 Desalkylflurazepam-d 4 293.00 139.80 45 30 Oxazepam-d 5 292.00 246.10 40 23 Nordiazepam-d 5 276.00 139.80 45 28 Triazolam-d 4 349.00 314.10 40 28 Lorazepam-d 4 326.90 281.00 30 20 Prazepam-d 5 330.00 276.10 40 22 Diazepam-d 5 290.00 153.80 40 28 * Underlined transitions were used for quantification. Table I (continued). MRM Transitions and Conditions for all Compounds and Their Deuterated Analogues* Precursor Product Cone Collision Compound Ion (m/z) Ion (m/z) Voltage (V) Energy (eV) Figure 1. MRM chromatograms obtained following the analysis of a spiked hair sample with 10 pg/mg of flurazepam (1), cloxazolam (2), zolpidem (3), 7-aminoclonazepam (4), 7-aminoflunitrazepam (5), zopiclone (6), loprazolam (7), clonazepam (8), bromazepam (9), flunitrazepam (10), clobazam (11), desmethylflunitrazepam (12), estazolam (13), nitrazepam (14), lorazepam (15), alprazolam (16), chlornordiazepam (17), temazepam (18), desalkylflurazepam (19), oxazepam (20), nordiazepam (21), lormetazepam (22), triazolam (23), brotizolam (24), diazepam (25), tetrazepam (26), clotiazepam (27), and prazepam (28). Journal of Analytical Toxicology, Vol. 29, October 2005 619 temazepam, desalkylflurazepam, oxazepam, nordiazepam, triazolam, lormetazepam, lorazepam, prazepam, tetrazepam, diazepam, and flurazepam (at a concentration of 1 mg/mL in methanol) and 7-aminoclonazepam-d 4 , 7-aminoflun i t r a z e p a m - d 7 , clonazepam-d 4 ,flunit r a z e p a m - d 7 , alprazolam-d 5 , temazepamd 5 , desalkylflurazepam-d 4 , oxazepam-d 5 , n o r d i a z e p a m - d 5 , triazolam-d 4 , lorazepamd 4 , prazepam-d 5 ,anddiazepam-d 5 ( 0 . 1 mg/mL in methanol) were purchased from LGC Promochem (Molsheim, France). Brotizolam, clotiazepam, chlornordiazepam, loprazolam, cloxazolam, zolpidem, and zopiclone were a gift from Dr. V. Maes (pure standards obtained from respective manufacturers). All solvents were HPLC grade. Specimens Pooled blank blood samples were used for development and validation of the procedure for blood and were obtained from a local blood bank. Blank urine and hair samples for the development and validation of the procedure were obtained from drug-free volunteers. Authentic blood, urine, and hair samples were collected during roadside controls for drugged driving, from drug-facilitated sexual assault (DFSA) cases (submitted as a part of the sexual assault kit), suicide cases, and (attempted) murder cases. Preparation of standard solutions An internal standard (IS) stock solution of each of the deuterated analytes was prepared (0.4 μg/mL in methanol), which was further diluted with methanol to yield appropriate concentrations to add to samples, calibrators, and quality control (QC) samples. A benzodiazepine stock solution of all analytes was prepared (4 μg/mL in methanol) and further diluted with methanol to yield working solutions at appropriate concentrations to add to calibrators and QC s a m p l e s . Working solutions were prepared monthly and stored at 4°C. Sample preparation and extraction Preparation of blood samples. B l o o d samples (250 μL) were mixed with 50 μL of a 1:4 dilution of the IS stock solution. Preparation of urine samples. U r i n e samples (250 μL) mixed with 50 μL of Table II. Intra-assay and Interassay Precision and Bias of the QC Samples, Prepared in Blood, Spiked at a Concentration of 4.0, 12, and 100 ng/mL Concentration of Intra-assay Precision Interassay Precision Compound QC (ng/mL) RSD Bias (%) RSD Bias (%) 7-Aminoclonazepam 4 2.3 –2.8 3.1 –1.9 12 1.0 1.3 2.0 0.0 100 1.6 2.3 3.1 0.5 7-Aminoflunitrazepam 4 2.6 0.0 2.6 2.7 12 0.4 6.5 1.2 9.3 100 1.1 2.2 2.5 4.0 Bromazepam 4 5.9 –5.4 10.4 –1.7 12 2.8 –8.4 4.7 –4.0 100 5.3 –7.6 4.9 –1.1 Clonazepam 4 1.0 –4.7 2.2 –0.2 12 2.3 2.3 2.8 2.6 100 1.6 4.4 2.5 3.7 Flunitrazepam 4 2.4 –0.9 5.1 –2.0 12 2.4 0.9 2.9 –0.2 100 2.1 1.7 2.1 0.6 Clobazam 4 4.0 –1.9 5.1 1.2 12 2.6 6.3 4.7 5.6 100 3.1 4.5 4.4 3.7 Desmethylflunitrazepam 4 2.8 –18.0 10.6 –11.0 12 2.5 –12.9 10.6 –8.1 100 3.4 –13.7 9.6 –7.6 Estazolam 4 4.0 –3.6 4.2 –1.4 12 3.9 3.0 5.1 2.2 100 2.2 2.8 3.0 3.0 Nitrazepam 4 1.7 –8.8 7.7 –4.0 12 1.6 –6.2 5.8 –1.9 100 5.0 –11.6 6.9 –4.9 Alprazolam 4 6.3 6.0 6.5 14.5 12 3.8 –4.6 4.4 –1.9 100 3.2 0.8 3.0 0.2 Chloornordiazepam 4 3.2 –0.3 5.1 –1.3 12 4.9 4.2 6.0 0.5 100 2.1 7.6 4.3 3.1 Temazepam 4 3.1 0.6 5.1 –1.0 12 3.0 1.8 3.7 –0.3 100 2.1 2.1 2.0 0.7 Desmethylflurazepam 4 1.1 –1.6 4.6 –1.5 12 3.2 –0.6 3.5 –0.9 100 1.3 1.4 2.2 0.9 Oxazepam 4 3.4 –2.8 5.3 1.0 12 3.0 1.4 5.6 0.8 100 1.7 0.9 2.7 1.4 Nordiazepam 4 5.4 –17.2 4.3 –2.8 12 4.3 –2.6 5.2 –0.2 100 3.7 1.8 4.1 1.4 Brotizolam 4 3.6 –1.8 7.1 0.3 12 3.1 –0.1 9.4 –1.3 100 1.3 1.0 6.2 1.6 Triazolam 4 3.4 0.8 6.3 –2.2 12 6.1 2.1 5.4 0.5 100 4.2 2.3 4.4 0.4 Lormetazepam 4 3.8 4.2 10.3 3.4 12 3.6 3.5 9.5 3.0 100 2.6 8.9 9.7 6.8 continued Journal of Analytical Toxicology, Vol. 29, October 2005 620 the IS stock solution were buffered to pH 4.6 with 200 μL acetate buffer (3M) and then incubated for 1 h at 56°C with 25 μL of -glucuronidase (Helix pomatia, 127,300 U/mL) (Sigma, St. Louis, MO). Preparation of hair samples. After decontamination (twice with dichloromethane, once with water, and once with methanol, 15 min. each, under ultrasonication), hair samples were dried and cut in segments of 1–3 cm each. Approximately 20 mg was powdered in a ball mill, which allowed simultaneous pulverization of 48 segments (Precellys 48, Bertin Technologies, Montigny-Le-Bretonneux, France), and then 50 μL of a 1:20 dilution of the IS stock solution was added. After incubation of the pulverized samples with 1 mL of methanol at 45°C for 2 h with orbital shaking, samples were centrifuged. The supernatants were subsequently transferred to 10-mL disposable screw-top vials and concentrated under nitrogen to 100–200 μL. Extraction. After sample preparation, samples were extracted with 4 mL of 1-chlorobutane after the addition of a saturated ammonium chloride buffer (pH 9.2); 500 μL was used for blood and hydrolyzed urine samples and 1 mL was used for the pulverized hair samples. After mechanical shaking (10 min) and centrifugation (10 min at 3000 g), the organic phase was transferred to a 5-mL disposable screw-top vial and then evaporated to dryness at 40°C in a vacuum centrifuge. For the extracted urine and blood samples, the residue was reconstituted in 100 μL of 0.1% formic acid in water/methanol (70:30, v/v), and 10 μL was injected into the LC–MS–MS system. The residue of the hair samples was reconstituted in 80 μL of 0.1% formic acid in water/methanol (70:30, v/v), and 20 μL was injected into the LC–MS–MS system. LC–MS–MS C h r o m a t o g r a p h y . LC was performed using a Waters Alliance 2690 separation module (Waters, Milford, MA). Analytes were separated on a XTerra MS C18 column (150- 2.1-mm, 3.5 μm) (Waters), using a gradient elution with 0.1% formic acid (A) and methanol (B), at a flow rate of 0.2 mL/min. A gradient was applied starting from 10% B, and increased to 50% over the first 5 min. From 5 min to 20 min, B was linearly increased to 70% before returning to the initial conditions within 0.1 min and equilibrating for 14.9 min, which resulted in a total run time of 35 min. M S . A Quattro Premier tandem MS (Waters) was used for all analyses. Ionization was achieved using electrospray in positive mode (ESI+). The optimum conditions were capillary voltage, 1.0 kV; source block temperature, 120°C; desolvation gas (nitrogen) heated to 270°C, and delivered at a flow rate of 700 L/h. In order to establish the appropriate multiple reaction monitoring (MRM) conditions for the individual compounds, solutions of standards [200 ng/mL, in 0.1% formic acid in water/methanol (70:30, v/v)] were infused into the MS and the cone voltage (CV) optimized to maximize the intensity of the protonated molecular species [M+H] + . Collision-induced dissociation of each protonated molecule was performed. The collision gas (argon) pressure was maintained at 0.35 Pa (3.5 1 0 – 3 mbar) and the collision energy adjusted to optimize the signal for the most abundant product ions, which were subsequently used for MRM analysis. All aspects of data acquisition were controlled using MassLynx NT 4.0 software with automated data processing using the QuanLynx program (Waters). LC–MS–MS assay validation L i n e a r i t y, L OQ, pr ec is io n, ac cu rac y, and recovery. Quantification was performed by integration of the area under the specific MRM chromatograms in reference to the integrated area of the deuterated analogues. The different IS were assigned to the different analytes in Lorazepam 4 2.4 –2.0 4.2 –3.7 12 4.5 1.4 5.4 –0.3 100 1.5 2.2 2.8 1.9 Prazepam 4 1.6 –8.9 2.3 –6.7 12 2.7 –2.3 5.4 –4.2 100 2.3 3.9 4.8 1.3 Clotiazepam 4 1.5 –2.4 4.9 –2.7 12 2.9 –1.4 6.4 –3.3 100 3.1 3.5 5.2 2.3 Tetrazepam 4 2.1 –5.5 4.2 –3.3 12 3.1 –3.0 4.4 –4.6 100 3.9 0.9 5.1 –0.3 Diazepam 4 2.7 –11.8 3.5 –3.0 12 2.0 –0.5 4.0 3.3 100 3.6 2.7 5.0 4.1 Loprazolam 4 2.0 –5.0 15.4 –9.7 12 8.2 –15.6 19.4 –16.0 100 9.6 –4.3 19.9 –3.2 Flurazepam 4 9.7 –3.2 12.1 –0.8 12 7.3 –4.5 10.8 –5.5 100 9.3 –11.8 9.4 –4.3 Cloxazolam 4 11.7 –11.3 11.5 –6.9 12 13.4 –9.9 12.0 –9.1 100 2.5 –6.6 16.6 –9.2 Zolpidem 4 1.3 –5.2 16.9 0.2 12 3.4 –9.8 16.8 –5.3 100 3.7 –4.5 13.1 1.7 Zopiclone 4 14.2 –12.5 17.3 –20.0 12 12.6 –16.7 13.2 –15.0 100 8.1 –15.0 9.7 –16.3 Table II (continued). Intra-assay and Interassay Precision and Bias of the QC Samples, Prepared in Blood, Spiked at a Concentration of 4.0, 12, and 100 ng/mL Concentration of Intra-assay Precision Interassay Precision Compound QC (ng/mL) RSD Bias (%) RSD Bias (%) Journal of Analytical Toxicology, Vol. 29, October 2005 621 the following combinations: 7-aminoflunitrazepam-d 7 was used for the calculations of peak-area ratios and concentrations of 7-aminoflunitrazepam, loprazolam, flurazepam, cloxazolam, zolpidem, and zopiclone; 7-aminoclona z e p a m - d 4 for 7-aminoclonazepam; f l u n i t r a z e p a m - d 7 for bromazepam and flunitrazepam; clonazepam-d 4 for clonazepam, clobazam, desmethylflunitrazepam, estazolam and nitrazepam; t e m a z e p a m - d 5 for temazepam; des a l k y l f l u r a z e p a m - d 4 for desalkylflurazepam and lormetazepam; nord i a z e p a m - d 5 for chlornordiazepam, nordiazepam and brotizolam; triazolamd 4 for triazolam; lorazepam-d 4 for lorazepam; alprazolam-d 5 for alprazolam; prazepam-d 5 for prazepam; diazepam-d 5 for clotiazepam, tetrazepam, and diazepam; oxazepam-d 5 for oxazepam. Calibration curves ranged from 1 to 200 ng/mL (1, 2, 5, 10, 20, 40, 80, 160, and 200 ng/mL) for blood samples, from 10 to 1000 ng/mL (10, 25, 50, 100, 250, 500, 750, and 1000 ng/mL) for urine, and from 0.5 to 1000 pg/mg (0.5, 1, 2, 5, 10, 50, 100, 200, 500, and 1000 pg/mg) for hair samples. Standard response curves were generated daily using a weighted (1/x) least-squares linear regression model. The LOQ was defined as the concentration of the lowest calibrator, which was calculated to be within ±20% of the nominal value and with a relative standard deviation (RSD) less than 20% (33,34). QCs were prepared for every run in blank matrix at a concentration of 4, 12, and 100 ng/mL for blood samples, at a concentration of 40, 150, and 600 ng/mL for urine and at a concentration of 7.5, 75, and 750 pg/mg for hair samples. Intra-assay precision was evaluated by analysis of 5 sets of the QC samples in one run for each of the three biological matrices. Interassay precision was evaluated by replicate analysis of one set of QC samples in several experiments performed on five different days. The precision was expressed as the RSD. A comparison of the calculated concentrations of the QC samples to their respective nominal values was used to assess the accuracy (bias) of the method. For the three matrices, relative recoveries were estimated by comparing the ratio of the peak area of the medium QC sample when the non-deuterated compounds were added before any pretreatTable III. Intra-assay and Interassay Precision and Bias of the QC Samples, Prepared in Urine, Spiked at a Concentration of 40, 150, and 600 ng/mL Concentration of Intra-assay Precision Interassay Precision Compound QC (ng/mL) RSD Bias (%) RSD Bias (%) 7-Aminoclonazepam 40 1.5 0.3 3.8 5.6 150 2.5 –4.6 3.0 –3.3 600 3.8 –4.5 4.6 –5.9 7-Aminoflunitrazepam 40 2.3 8.4 2.8 14.1 150 2.3 3.2 4.0 11.7 600 1.6 –1.9 3.2 5.0 Bromazepam 40 5.8 3.5 5.3 7.7 150 9.4 4.8 10.0 –7.5 600 8.1 5.0 7.6 –1.1 Clonazepam 40 2.0 5.4 1.9 4.8 150 0.9 –0.6 1.2 1.8 600 1.4 –1.3 3.0 –5.2 Flunitrazepam 40 2.7 3.9 2.6 3.7 150 2.9 –1.5 3.0 –1.7 600 0.9 –0.8 3.5 –0.5 Clobazam 40 1.8 8.5 1.6 6.8 150 3.5 4.3 3.2 1.1 600 2.8 –0.3 2.8 –4.1 Desmethylflunitrazepam 40 4.9 –1.1 6.5 10.0 150 4.3 –0.1 4.2 6.2 600 7.9 –5.0 7.7 –1.9 Estazolam 40 1.8 4.0 2.6 9.0 150 2.3 –2.8 1.9 –2.6 600 0.8 0.4 3.2 –7.6 Nitrazepam 40 2.8 1.9 2.3 3.3 150 1.2 –3.3 1.0 –2.9 600 1.6 –2.7 3.2 –9.8 Alprazolam 40 5.0 1.9 4.1 1.0 150 3.9 –4.9 3.4 –5.7 600 2.3 3.4 3.4 –2.2 Chloornordiazepam 40 2.9 5.2 3.6 10.7 150 2.2 0.1 2.3 1.8 600 3.4 –1.9 4.0 1.6 Temazepam 40 2.2 1.2 2.5 5.4 150 2.4 –4.5 2.2 –5.3 600 2.1 –3.9 3.9 1.4 Desalkylflurazepam 40 1.9 2.2 2.6 3.2 150 1.7 –2.6 1.7 –1.1 600 3.3 2.4 3.2 –1.8 Oxazepam 40 3.5 –0.1 4.1 0.3 150 3.1 –6.5 3.6 –8.6 600 3.1 –2.0 2.8 0.5 Nordiazepam 40 0.9 3.7 2.3 6.6 150 3.5 –1.1 3.0 –0.4 600 1.6 1.0 2.0 –0.2 Brotizolam 40 3.1 3.8 4.1 8.2 150 3.1 –1.8 3.0 –1.5 600 3.2 3.8 4.1 4.7 Triazolam 40 2.1 3.5 3.2 –1.1 150 1.3 –4.3 1.5 –6.3 600 2.7 –1.2 3.0 –5.4 Lormetazepam 40 2.1 1.6 2.9 –0.2 150 2.7 –4.0 3.6 –10.8 600 3.6 5.0 3.7 –1.3 continued Journal of Analytical Toxicology, Vol. 29, October 2005 622 ment or extraction (n= 3) divided by the peak area of the internal standards with the ratio of the peak area obtained when the nondeuterated analytes were added after the extraction (n= 3) divided by the peak area of the internal standards. The deuterated standards were added before the extraction in all experiments. Assessment of matrix effects. To assess any potential suppression or enhancement of ionization from components present in the extracted biological matrix, a continuous post-column infusion was performed using a mixture of all benzodiazepines and deuterated analogues (10 ng/mL at a flow rate of 10 μL/min) to produce a constant elevated response in both MRM channels. The interference of this constant response was monitored following the injection of extracted samples and compared to the response following the injection of mobile phase only. Results and Discussion The applied gradient ensured the elution of all the drugs examined within 20 min and produced chromatographic peaks of acceptable symmetry. Selectivity of the method was achieved by a combination of retention time, precursor, and transitions. During the first injection of an extracted sample, the most prominent precursor-product transitions (except for alprazolam, for which an elevated background was noticed in this transition) were used for quantification. Further confirmation of the identity of the compounds was achieved through a second injection of positive samples and by monitoring two transitions (i.e., a quantifier and a qualifier) per compound. For the deuterated internal standards, a single MRM transition was used. Table I summarizes the MRM transitions and conditions of all quantifiers and qualifiers for all analytes and IS. For all compounds investigated, peak-area ratios quantifier/qualifier were found to be very reproducible with variation (as RSD) less than 10%. The method was validated for linearity, LOQ, precision, acc u r a c y, a n d an a l y t i c a l r e c ov e r y b y t h e a n a l y s e s o f s p i k e d blood, urine, and hair samples. In each case, a weighted ( 1 / x) linear regression line was applied. Linearity with correlation coefficients r 2 > 0.99 were achieved in the range investigated (i.e., from the LOQ to 200 ng/mL for blood samples, from the LOQ to 1000 ng/mL for urine samples, and from LOQ to 1000 pg/mg for hair samples). The back-calculated concentrations of all calibrators were compared with their respective nominal values and were within 100 ± 20% of the nominal value. The obtained LOQ in blood was 1.0 ng/mL for all analytes, except for lorazepam, loprazolam, zolpidem, and zopiclone, where an LOQ of 2.0 ng/mL was observed. For urine samples, the LOQ was established at 10 ng/mL for all analytes, except for brotizolam and tetrazepam with an LOQ of 25 ng/mL. Finally, for hair samples, LOQs ranging from 0.5 pg/mg (prazepam) to 10 pg/mg (tetrazepam, loprazolam, and zopiclone) were observed. Figure 1 shows the MRM chromatograms obtained following the analysis of a hair sample spiked with 10 pg/mg of each of the compounds. At the LOQ, the qualifier had a signal-to-noise ratio > 10:1. The intra-assay precision (repeatability), interassay precision (reproducibility), and bias were < 20% in most cases for all matrices tested (Tables II–IV). The obtained LOQ was comparable with previous reports dealing with a large series of benzodiazepines (28,31,32). The reported sensitivities for hair samples were comparable with the values for other LC–MS–MS methods dealing with only one or a limited number of benzodiazepines (35–39). The LOQs were slightly higher than the values obtained with previous negative-ion chemical ionizationGC–MS methods detecting one or a few benzodiazepines in hair (40,41). Lorazepam 40 2.5 2.8 3.0 0.5 150 3.5 –3.9 3.1 –0.9 600 2.8 –1.7 2.7 –5.4 Prazepam 40 2.2 5.2 2.4 6.4 150 1.1 –2.2 1.0 –1.3 600 1.0 0.4 1.7 –1.4 Clotiazepam 40 3.0 5.9 2.5 5.3 150 1.0 –2.1 1.7 –4.5 600 1.5 4.3 5.1 –8.9 Tetrazepam 40 7.3 9.8 17.8 12.7 150 4.9 5.8 18.0 –20.6 600 6.5 –0.2 20.6 –6.5 Diazepam 40 2.7 3.7 2.9 7.1 150 2.1 –1.7 1.8 –1.7 600 1.2 –0.1 2.0 –0.6 Loprazolam 40 10.2 19.0 12.0 13.0 150 8.3 18.6 9.1 –12.5 600 8.6 10.1 13.0 –6.8 Flurazepam 40 9.7 19.4 10.9 –10.1 150 4.5 6.7 9.7 –19.3 600 3.4 5.0 5.7 –7.9 Cloxazolam 40 4.5 11.2 16.5 –3.8 150 2.3 –0.4 20.8 –5.9 600 4.9 –3.1 19.5 –9.7 Zolpidem 40 9.6 13.5 10.4 –18.1 150 1.3 6.3 6.3 –2.7 600 2.4 4.5 2.3 3.7 Zopiclone 40 9.1 18.0 12.6 –7.8 150 2.5 7.1 10.7 7.9 600 4.1 1.2 10.3 11.2 Table III (continued). Intra-assay and Interassay Precision and Bias of the QC Samples, Prepared in Urine, Spiked at a Concentration of 40, 150, and 600 ng/mL Concentration of Intra-assay Precision Interassay Precision Compound QC (ng/mL) RSD Bias (%) RSD Bias (%) Journal of Analytical Toxicology, Vol. 29, October 2005 623 After sample preparation, the same extraction procedure was applied for the three biological matrices. Reproducible recoveries were obtained, ranging from 49.2 to 103.7% in blood, from 25.1 to 98% for urine (except for cloxazolam, with an extraction recovery of 3.8%), and from 53.0 to 103.8% for hair. The extraction recoveries are similar for the three matrices, except for tetrazepam and cloxazolam, where significantly lower recoveries were noted for the extraction of spiked urine samples. To demonstrate that these compounds show a reduced stability when incubating at higher temperatures in aqueous solutions (during the hydrolysis of urine samples), an additional experiment was performed, adding the nondeuterated compounds after sample preparation, but before extraction with 1-chlorobutane. In this experiment, these extraction recoveries for tetrazepam and cloxazolam were similar as those obtained for blood, indicating a decomposition during the preparation of urine samples. The extraction recoveries obtained are comparable with previous reports ( 1 5 , 2 2 , 2 6 – 2 8 , 3 1 , 3 2 ) . Insufficient sample clean up can result in matrix effects, leading to either suppression or enhancement of the analyte response (42–44). This can lead to variable sensitivities and decreased precision and accuracy. Consequently, in the development of any LC–MS(–MS) method, the potential for any such ion suppression or enhancement should be assessed. To test this, post-column infusion experiments [based on the method described by Bonfiglio et al. (42)] were performed for the three biological matrices to provide information of the effect of matrix throughout the course of the detection w i n d o w. A decrease in response ranging from 20 to 90% was observed starting from 1.7 min to maximal 6.0 min for the three matrices tested (data not shown). H o w e v e r, t h i s s u p p r e s s io n h a d d i m i n - ished and normal baseline responses were restored before the elution time of the compounds. In the rest of the chromatographic run, no significant changes in responses were observed. The results confirm the usefulness of the liquid– liquid extraction procedure as a sample clean up before chromatography to obtain reproducible and reliable quantitaTable IV. Intra-assay and Interassay Precision and Bias of the QC Samples, Prepared in Hair, Spiked at a Concentration of 7.5, 75, and 750 pg/mg Concentration of Intra-assay precision Interassay precision Compound QC (ng/mL) RSD Bias (%) RSD Bias (%) 7-Aminoclonazepam 7.5 5.5 9.4 6.1 5.0 75 2.5 0.8 2.2 2.5 750 4.4 –2.7 4.2 –2.3 7-Aminoflunitrazepam 7.5 4.4 7.9 4.4 6.3 75 1.6 0.1 2.7 2.2 750 0.4 2.3 2.5 2.2 Bromazepam 7.5 16.8 –7.8 13.6 8.6 75 8.1 12.7 7.4 9.8 750 13.1 12.7 10.6 3.8 Clonazepam 7.5 3.5 12.7 4.3 9.7 75 2.2 10.3 1.9 10.0 750 2.0 4.9 4.5 2.8 Flunitrazepam 7.5 3.3 5.4 3.4 5.2 75 1.1 2.2 5.3 5.6 750 1.3 3.7 3.9 6.8 Clobazam 7.5 2.9 3.6 4.9 –1.3 75 2.8 9.5 13.9 1.2 750 5.6 5.2 14.5 –3.8 Desmethylflunitrazepam 7.5 9.1 10.3 9.0 15.1 75 9.9 13.1 11.1 8.5 750 9.0 4.9 9.0 10.0 Estazolam 7.5 5.0 10.4 9.9 15.2 75 4.7 3.0 16.6 14.2 750 2.9 0.4 12.6 10.2 Nitrazepam 7.5 5.6 8.8 4.9 11.8 75 3.6 8.3 8.3 15.9 750 2.6 2.2 3.4 8.1 Alprazolam 7.5 17.8 19.4 16.6 17.9 75 3.8 –0.5 3.6 –0.3 750 7.3 6.9 1.2 4.5 Chlornordiazepam 7.5 6.4 3.5 6.2 2.7 75 2.7 6.5 4.5 5.8 750 5.8 3.2 6.7 1.9 Temazepam 7.5 8.2 4.0 7.9 2.6 75 0.8 3.7 2.1 5.0 750 4.5 4.7 5.4 4.0 Desalkylflurazepam 7.5 7.4 8.9 6.7 9.1 75 1.1 5.2 2.3 7.3 750 2.7 6.5 2.3 8.0 Oxazepam 7.5 7.2 8.6 9.3 10.8 75 3.5 2.7 16.5 9.1 750 2.7 0.5 5.0 2.5 Nordiazepam 7.5 7.0 4.3 6.6 3.6 75 3.4 4.1 4.0 7.2 750 5.7 6.4 4.9 7.0 Brotizolam 7.5 7.0 5.8 8.2 6.3 75 7.0 4.8 6.2 4.2 750 8.6 4.2 7.6 3.0 Triazolam 7.5 12.3 2.4 11.0 4.7 75 1.9 –1.6 3.9 2.0 750 4.3 3.0 3.8 4.8 Lormetazepam 7.5 7.1 10.2 8.1 2.4 75 8.0 8.9 7.5 1.6 750 4.2 9.4 4.3 4.1 continued Journal of Analytical Toxicology, Vol. 29, October 2005 624 tive results for all compounds without major interference of matrix compounds. The validated LC–MS–MS method has been successfully applied to the analysis of external quality control samples in serum (organized by the Gesellschaft für Toxikologische und Forensische Chemie). A certificate was obtained for all compounds tested (diazepam, nordiazepam, bromazepam, clonazepam, flunitrazepam, desmethylflunitrazepam, 7-aminoflunitrazepam, and oxazepam). In addition, the method was used for the analysis of authentic blood, urine, and hair samples. Blood and urine samples were collected during roadside controls for drugged driving, from DFSA cases, suicide cases, and (attempted) murder cases. Samples with a concentration above the linear range of the calibration curve were appropriately diluted in water (before extraction), and reanalyzed. To check the validity of the quantification for this modified sample preparation, five different added concentrations above the calibration range for blood and urine (n= 3 for each concentration) were analyzed with different dilutions. The bias was always < 20%. Hair analysis was applied to two authentic cases. In the first case, a young woman claimed to have been sexually abused over a period of several years while being drugged. Because of the increased detection window, toxicological analysis of hair yields long-term information about drug consumption and allows the personal history of drug use to be established. As such, the laboratory was requested to analyze the victim’s hair (33 cm). Starting from the root, 16 segments of 1–3 cm were analyzed. All segments were positive for more than one benzodiazepine, with generally higher concentrations in segments closer to the root. Benzodiazepines detected above the LOQ were 7-aminoclonazepam (in 5 segments, ranging from 2.2 to 11.7 pg/mg), bromazepam (in all segments, ranging from 19.1 to 1405 pg/mg), clonazepam (in 2 segments, ranging from 16.0 to 32.6 pg/mg), clobazam (in 1 segment, 11.8 pg/mg), oxazepam (1 segment, 8.5 pg/mg), nordiazepam (in 5 segments, ranging from 2.3 to 6.9 pg/mg), lormetazepam (in 5 segments, ranging from 5.0 to 28.1), lorazepam (in 4 segments, ranging from 7.6 to 13.9 pg/mg), tetrazepam (in 2 segments, 55.6 and 123.6 pg/mg), loprazolam (in 1 segment, 24.7 pg/mg), and zolpidem (in all segments, ranging from 8.5 to 177.5 pg/mg). These results indicate long-term multiple drug exposure. In the second case, a middle-aged woman ingested one tablet of Lexotan (containing 3 mg of bromazepam) on two consecutive days, followed by one tablet on three consecutive days a month later. Hair (7 cm) was cut 6 weeks later and 5 segments of 1–1.5 cm were analyzed. Figure 2 shows the chromatogram obtained after the analysis of the first two proximal segments, positive for bromazepam at 24.2 pg/mg (segment 1, 1.5 cm) and 11.7 pg/mg (segment 2, 1.5 cm). In the third segment (1 cm) traces of bromazepam were detected, probably because of the variability in the incorporation of this drug into the hair shaft and axial migration after incorporation, leading to a distribution over a small region (35). This result is consistent with the doses taken and is in accordance with concentrations of bromazepam found in hair in earlier reports (35,37). Conclusions In this report, a validated and highly sensitive LC–ESI-MS–MS method is described for the simultaneous quantification of 26 commonly encountered benzodiazepines and their metabolites, zolpidem and zopiclone, in three different biological matrices (i.e., blood, urine, and hair). Because of the low LOQs, the method was demonstrated appropriate for Lorazepam 7.5 7.4 –0.8 15.0 –5.3 75 4.0 3.3 16.0 –1.7 750 6.4 4.8 18.0 –1.1 Prazepam 7.5 3.2 7.7 4.2 6.3 75 1.3 3.8 2.4 7.1 750 4.2 6.2 5.8 6.5 Clotiazepam 7.5 12.4 2.1 15.7 0.2 75 3.1 –5.8 13.6 –4.8 750 7.1 7.4 17.6 5.7 Tetrazepam 7.5 NA* NA NA NA 75 6.8 –5.5 19.4 –10.4 750 7.1 6.4 16.3 0.2 Diazepam 7.5 4.7 5.2 4.8 5.2 75 2.0 3.1 4.0 6.1 750 7.5 1.2 7.1 2.6 Loprazolam 7.5 NA NA NA NA 75 8.9 15.1 19.7 –1.1 750 7.9 14.1 6.3 9.8 Flurazepam 7.5 25.0 –6.1 20.0 18.1 75 8.9 –15.2 14.3 11.2 750 15.7 –4.3 17.3 15.2 Cloxazolam 7.5 12.6 4.3 11.9 11.5 75 6.3 9.5 5.0 –3.9 750 10.4 18.2 8.6 –0.4 Zolpidem 7.5 17.2 8.6 15.7 –3.1 75 7.6 0.0 17.9 –3.4 750 12.9 –6.2 14.2 10.8 Zopiclone 7.5 NA NA NA NA 75 11.9 –3.2 13.1 15.9 750 11.2 14.0 10.2 15.0 *NA = not analyzed < LOQ. Table IV (continued). Intra-assay and Interassay Precision and Bias of the QC Samples, Prepared in Hair, Spiked at a Concentration of 7.5, 75, and 750 pg/mg Concentration of Intra-assay Precision Interassay Precision Compound QC (ng/mL) RSD Bias (%) RSD Bias (%) Journal of Analytical Toxicology, Vol. 29, October 2005 625 the quantification of low doses of these compounds in authentic blood, urine, and hair samples collected from forensic toxicology cases. Acknowledgment We would like to acknowledge the Belgian Federal Office for Scientific, Technical, and Cultural Affairs. References 1. O.H. Drummer and M. Odell. Benzodiazepines and related drugs. The Forensic Pharmacology of Drugs of Abuse. Arnold, London, U.K., 2001, pp 103–175. 2. D.J. Garretty, K. Wolff, A.W. Hay, and D. Raistrick. Benzodiazepine misuse by drug addicts. Ann. Clin. Biochem. 34: 68–73 (1997). 3. C . P. O’Brien. Benzodiazepine use, abuse, and dependence. J. Clin. Psychiatry 66: 28–33 (2005). 4. J. Sareen, M.W. Enns, and B.J. Cox. Potential for misuse of sedatives. Am. J. Psychiatry 161: 1722–1723 (2004). 5. A. Negrusz and R.E. Gaensslen. Analytical developments in toxicological investigation of drug-facilitated sexual assault. Anal. Bioanal. Chem. 376: 1192–1197 (2003). 6. D. Gunnell, D. Ho, and V. Murray. Medical management of deliberate drug overdose: a neglected area for suicide prevention? Emerg. Med. J. 21: 35–38 (2004). 7. E. Kelly, S. Darke, and J. Ross. A review of drug use and driving: e p i d e m i o l o g y, impairment, risk factors and risk perceptions. D r u g Alcohol Rev. 23: 319–344 (2004). 8. K.L. Movig, M.P. Mathijssen, P.H. Nagel, T. van Egmond, J.J. de Gier, H.G. Leufkens, and A.C. Egberts. Psychoactive substance use and the risk of motor vehicle accidents. Accid. Anal. Prev. 3 6 : 631–636 (2004). 9. P. Kintz. Value of hair analysis in postmortem toxicology.F o r e n s i c Sci. Int. 142: 127–134 (2004). 10. A.A. Elian. ELISA detection of clonazepam and 7-aminoclonazepam in whole blood and urine. Forensic Sci. Int. 134: 54–56 (2003). 11. L. Kroener, F. Musshoff, and B. Madea. Evaluation of immunochemical drug screenings of whole blood samples. A retrospective optimization of cutoff levels after confirmation-analysis on GC–MS and HPLC–DAD. J. Anal. Toxicol. 27: 205–212 (2003). 12. P. Kemp, G. Sneed, T. Kupiec, and V. Spiehler. Validation of a microtiter plate ELISA for screening of postmortem blood for opiates and benzodiazepines. J. Anal. Toxicol. 26: 504–512 (2002). 13. V. Spiehler. Hair analysis by immunological methods from the beginning to 2000. Forensic Sci. Int. 107: 249–259 (2000). 14. P.H. Wang, C. Liu, W.I. Ts a y, J.H. Li, R.H. Liu, T.G. Wu, W.J. Cheng, D.L. Lin, T.Y. Huang, and C.H. Chen. Improved screen and confirmation test of 7-aminoflunitrazepam in urine specimens for monitoring flunitrazepam (Rohypnol) exposure. J. Anal. Toxicol. 26: 411–418 (2002). 15. K.S. Scott and Y. Nakahara. A study into the rate of incorporation of eight benzodiazepines into rat hair. Forensic Sci. Int. 1 3 3 : 47–56 (2003). 16. H.H. Maurer. Role of gas chromatography–mass spectrometry with negative ion chemical ionization in clinical and forensic toxi c o l o g y, doping control, and biomonitoring. Ther. Drug Monit. 2 4 : 247–254 (2002). 17. A. Negrusz, C.M. Moore, K.B. Hinkel, T.L. Stockham, M. Ve r m a , M.J. Strong, and P.G. Janicak. Deposition of 7-aminoflunitrazepam and flunitrazepam in hair after a single dose of Rohypnol. J. Forensic Sci. 46: 1143–1151 (2001). 18. S. Paterson, N. McLachlan-Troup, R. Cordero, M. Dohnal, and S. Carman. Qualitative screening for drugs of abuse in hair using GC–MS. J. Anal. Toxicol. 25: 203–208 (2001). 19. T. Keller, A. Schneider, and E. Tu t s c h - B a u e r. GC/MS determination of zolpidem in postmortem specimens in a voluntary intoxication. Forensic Sci. Int. 106: 103–108 (1999). 20. M.H. Andraus, A. Wong, O.A. Silva, C.Y. Wada, O. To f f l e t o , C . P. Azevedo, and M.C. Salvadori. Determination of bromazepam in human plasma by high-performance liquid chromatography with electrospray ionization tandem mass spectrometric detection: application to a bioequivalence study. J. Mass Spectrom. 3 9 : 1348–1355 (2004). 21. A. Bugey and C. Staub. Rapid analysis of benzodiazepines in whole blood by high-performance liquid chromatography: use of a monolithic column. J. Pharm. Biomed. Anal. 3 5 : 5 5 5 – 5 6 2 (2004). 22. N. Jourdil, J. Bessard, F. Vincent, H. Eysseric, and G. Bessard. Automated solid-phase extraction and liquid chromatography–electrospray ionization-mass spectrometry for the determination of flunitrazepam and its metabolites in human urine and plasma samples. J. Chromatogr. B 788: 207–219 (2003). 23. A. El Mahjoub and C. Staub. Semiautomated high-performance liquid chromatographic method for the determination of benzodiazepines in whole blood. J. Anal. Toxicol. 2 5 : 209–214 (2001). Figure 2. Chromatograms of a hair extract [segment 1 (A) and segment 2 (B)] from a subject exposed to one tablet of bromazepam on two consecutive days, followed by one tablet on three consecutive days a month later (top trace: quantifier, middle trace: qualifier, bottom trace: flunitrazepamd 7 ). Bromazepam concentrations were 24.2 pg/mg (A) and 11.7 pg/mg (B). 20 M. Laloup et al. / J. Chromatogr. A 1082 (2005) 15–24 Table 2 Precisionaand accuracy data for THC for the extraction of 100 and 500 ␮L of spiked preserved oral fluid samples Volume oral fluid (␮L) Concentration of QC (ng/mL) Intra-assay precision Inter-assay precision Mean concentration found (ng/mL) RSD (%) Bias (%) Mean concentration found (ng/mL) RSD (%) Bias (%) 100 2.5 2.5 3.6 −1.0 2.4 2.9 −2.5 25.0 24.8 5.4 −0.7 24.0 5.4 −4.1 500 0.5 0.5 2.5 −2.4 0.5 4.1 −5.5 5.0 4.9 0.4 −2.0 4.7 3.8 −6.8 aIntra-assay precision was evaluated by the preparation and analysis of four replicates of a low and a high QC in a single assay for both volumes of oral fluid used. Inter-assay precision was evaluated by the preparation and analysis of each QC over eight consecutive days. The Intercept collector contains a variety of chemicals, i.e. sodiumchloride,sodiumbenzoate,potassiumsorbate,bovine gelatin, Tween 20, chlorhexidine digluconate and a blue dye, some of which can interfere with the LC–MS-MS detection signal. To assess this, we compared peak area responses obtained when THC was added after the extraction of blank preserved oral fluid with the responses obtained when THC was added to an extract where the preserved oral fluid was substituted with water. No statistically significant different peak areas were observed. Post-column infusion experiments (based on the method described by Bonfiglio et al. [22]) were performed to provide information of the effect of matrix throughout the course of the whole chromatographic run and not just at the elution time for the analytes. The effect on THC response obtained following the injection of a mobile phase control is shown in Fig. 2A. As expected, no changes in response were observed. The effects on THC response obtained following the injection of a sample prior extraction and after extraction of 100 and 500␮L of preserved oral fluid are given in Fig. 2B, C and D, respectively. The results confirm the usefulness of the liquid–liquid extraction as a sample clean-up before chromatography: a decrease of 100% in response starting from ∼1.7min was observed when no sample clean-up was performed. A reduction of 50% was still noted at the moment of elution of THC, probably due to the elution of endogenous components. When injecting extracted samples, this suppression was still apparent but restored by the elution time of THC. In addition to THC, cannabidiol and cannabinol are two components that are also present in the Fig. 2. Evaluation of the effect on THC response of an injection of a mobile phase control (A), a blank sample prior to extraction (B) and the same sample following the extraction of 100 and 500␮L of preserved oral fluid (C and D, respectively). The shaded area indicates the elution position of THC. M. Laloup et al. / J. Chromatogr. A 1082 (2005) 15–24 21 Fig. 3. LC–MS-MS analysis of an extracted 100␮L blank oral fluid sample enriched with 5ng/mL THC-d3(top trace), THC and cannabidiol (middle trace) and cannabinol (bottom trace). Peak intensity is shown in the top right-hand corner of each trace. Cannabis sativa plant and may also be detected in oral fluid. To evaluate their potential for interference, standards were analysed using the developed LC–MS-MS method. This is particularly important in the case of cannabidiol since this component has the same molecular mass (and thus the same protonated species) as THC and shows the same product ions after CID. Cannabidiol eluted at 3.28min and was chromatographically resolved from THC. In contrast, cannabinol did not produce any response in the monitored MRM channel due to a different molecular mass. The appropriate MRM transition for this component was m/z311.2→223.1, as determined by direct infusion experiments. Cannabinol was demonstrated to elute at 4.38min. Fig. 3 shows the MRM chromatograms obtained following LC–MS-MS analysis of an extracted 100␮L blank oral fluid sample enriched with 5ng/mL THC-d3, THC, cannabidiol and cannabinol. The validated LC–MS-MS method was applied to the analysis of 102 oral fluid samples collected with the Intercept from volunteers who had received either a placebo cigarette or a marijuana cigarette. THC concentrations obtained after smoking a single marijuana cigarette are shown in Fig. 4. For these cases only the presence of THC had to Fig. 4. Boxand whisker plots of THC levels in preserved oral fluid samples from nine healthy volunteers following smoking of a single marijuana cigarette. Oral fluid samples were taken 0.5h prior to smoking and at 0.25, 0.5, 1, 1.25 and 1.5h after smoking. Concentrations plotted on the Y-axis are expressed as ng/mL. The central box represents the values from the lower to upper quartile (25–75 percentile). The middle line represents the median. The horizontal line extends from the minimum to the maximum value, excluding “outside” (not present) and “far out” values (cross marker) which are displayed as separate points. 22 M. Laloup et al. / J. Chromatogr. A 1082 (2005) 15–24 Fig. 5. Typical MRM chromatograms obtained following the analysis of two authentic preserved oral fluid specimens obtained from drivers in a roadside setting. Concentrations were 5.7ng/mL (A) and 50.8ng/mL (B). The figure shows the response for THC-d3(top trace) and for the two transitions of THC (quantifier and qualifier; middle and bottom traces respectively). Peak intensity is shown in the top right-hand corner of each trace. be confirmed, thus 500␮L of oral fluid was used for the analysis. For samples where the response exceeded the upper limit of the standard curve, reanalysis of only 100␮L was performed. At −0.5h all specimens were negative for THC, except for three subjects in which low concentrations were found (0.2, 0.4 and 2.2ng/mL). However, it should be noted that in both the placebo and marijuana condition, THC could be detected, probably due to incomplete removal of THC for the preparation of the placebo cigarette. Mean peak (±1 SD) THC concentration in the marijuana condi- M. Laloup et al. / J. Chromatogr. A 1082 (2005) 15–24 23 Table 3 Results obtained applying the method to 48 preserved oral fluid samples collected by the police at the roadside Sample identity THC (ng/mL) Sample identity THC (ng/mL) 1 5.7 25 60.2 2 7.0 26 3.9 3 4.6 27 52.2 4 18.5 28 25.4 5 2.5 29 193.5 6 95.8 30 111.2 7 <LOQ 31 7.3 8 84.7 32 14.6 9 <LOQ 33 1.9 10 0.5 34 4.7 11 4.5 35 100.0 12 3.9 36 23.0 13 31.9 37 57.1 14 50.8 38 88.6 15 34.6 39 3.9 16 56.0 40 375.8 17 81.1 41 3.7 18 11.9 42 4.4 19 107.4 43 4.2 20 92.1 44 4.2 21 10.0 45 4.2 22 17.6 46 4.1 23 94.8 47 4.0 24 37.2 48 4.4 tion occurred at the first specimen collection (0.25h) and was 30.6ng/mL (±21.6 ng/mL). Thereafter, THC concentrations declined steadily to mean concentrations of 2.6ng/mL (±2.3ng/mL). Overall, concentrations were quite variable; this has also been reported by other authors [8] and may be duetothelackofexactvolumemeasurementofthecollection device. The Intercept device is a collection device on which the specimen is absorbed onto a matrix, leading to variable absorbed volumes. The mean peak concentration is lower than the one reported by Niedbala et al. using the same collection device [8]. This could be due to the fact that the samples were only analysed several months after sampling. During this time the samples were conserved at −20◦C on the pad, i.e. without prior centrifugation. However, no stability studies on this aspect were available from the manufacturer. During roadside controls for drugged driving, the police collected 48 authentic oral fluid samples for a confirmatory analysis in the laboratory. In these cases only 100␮Lof preserved oral fluid was used due to limited sample volume. Fig. 5 shows typical MRM chromatograms of Intercept samples obtained from two marijuana users. In Fig. 5A, the presence of cannabidiol (at a retention time of 3.28min) was also noted. A summary of the quantitative results for the positive samples is presented in Table 3. In these samples, the median THC concentration was 13.3ng/mL with a range from 0.5 to 375.8ng/mL. The measured THC concentrations varied considerably and some samples had to be reanalysed after dilution (one in five dilution with blank oral fluid). 4. Conclusions A fully validated LC–MS-MS method for the determination of THC in preserved oral fluid, collected with the Intercept device, was developed. The method offers the combination of a very simple liquid–liquid extraction to avoid ion suppression, a high recovery and excellent precision and accuracy, when using either 100 or 500␮L of collected sample. The method was successfully applied to Intercept samples collected at the roadside and collected after a controlled study with cannabis. Acknowledgement We would like to acknowledge the Belgian Federal Office for Scientific, Technical and Cultural Affairs. References [1] P. Kintz, N. Samyn, Therap. Drug Monit. 24 (2002) 239. [2] E. Cone, Forensic Sci. Int. 21 (2001) 7. [3] N. Samyn, G. De Boeck, A.G. Verstraete, J. Forensic Sci. 47 (2002) 1380. [4] D.A. Kidwell, J.C. Holland, S. Athanaselis, J. Chromatogr. B 713 (1998) 111. [5] A. Verstraete, in: J. Oliver, P. Williams, A. Clayton (Eds.), Proceedings of the 17th International Conference on Alcohol, Drugs and Traffic Safety, CD-ROM. [6] E.J. Cone, L. Presley, M. Lehrer, W. Seiter, M. Smith, K.W. Kardos, D. Fritch, S. Salamone, S. Niedbala, J. Anal. Toxicol. 26 (2002) 541. [7] M.A. Huestis, E.J. Cone, J. Anal. Toxicol. 28 (2004) 394. [8] R.S. Niedbala, K.W. Kardos, D.F. Fritch, S. Kardos, T. Fries, J. Waga, J. Robb, E.J. Cone, J. Anal. Toxicol. 25 (2001) 289. [9] W. Schramm, R.H. Smith, P.A. Craig, D.A. Kidwell, J. Anal. Toxicol. 16 (1992) 1. [10] N. Samyn, A. Verstraete, C. van Haeren, P. Kintz, Forensic Sci. Rev. 11 (1999) 1. [11] R.A. Gustafson, E.T. Moolchan, A. Barnes, B. Levine, M.A. Huestis, J. Chromatogr. B 798 (2003) 145. [12] M.H. Chu, O.H. Drummer, J. Anal. Toxicol. 26 (2002) 575. [13] P. Kintz, V. Cirimele, B. Ludes, J. Anal. Toxicol. 24 (2000) 557. [14] W. Weinmann, S. Vogt, R. Goerke, C. Muller, A. Bromberger, Forensic Sci. Int. 113 (2000) 381. [15] W. Weinmann, M. Goerner, S. Vogt, R. Goerke, S. Pollak, Forensic Sci. Int. 121 (2001) 103. [16] B. Maralikova, W. Weinmann, J. Mass Spectrom. 39 (2004) 526. [17] S. Valiveti, A.L. Stinchcomb, J. Chromatogr. B 803 (2004) 243. [18] M. Concheiro, A. de Castro, O. Quintela, A. Cruz, M. LopezRivadulla, J. Chromatogr. B 810 (2004) 319. [19] V.P. Shah, K.K. Midha, S. Dighe, I.J. McGilveray, J.P. Skelly, A. Yacobi, T. Layloff, C.T. Viswanathan, C.E. Cook, et al., Eur. J. Drug Metab. Pharmacokinet. 16 (1991) 249. [20] V.P. Shah, K.K. Midha, J.W. Findlay, H.M. Hill, J.D. Hulse, I.J. McGilveray, G. McKay, K.J. Miller, R.N. Patnaik, M.L. Powell, A. Tonelli, C.T. Viswanathan, A. Yacobi, Pharm. Res. 17 (2000) 1551. 24 M. Laloup et al. / J. Chromatogr. A 1082 (2005) 15–24 [21] Substance Abuse and Mental Health Administration. Mandatory guidelines for Federal workplace drug testing programs, Fed. Regist. 59 (1994) 29908–29931. [22] R. Bonfiglio, R.C. King, T.V. Olah, K. Merkle, Rapid Commun. Mass Spectrom. 13 (1999) 1175. [23] B.K. Matuszewski, M.L. Constanzer, C.M. Chavez-Eng, Anal. Chem. 70 (1998) 882. [24] R. Dams, M.A. Huestis, W.E. Lambert, C.M. Murphy, J. Am. Soc. Mass Spectrom. 14 (2003) 1290. Quantitative analysis of multiple illicit drugs in preserved oral fluid by solid-phase extraction and liquid chromatography–tandem mass spectrometry Michelle Wood a , Marleen Laloup b , Maria del Mar Ramirez Fernandez b , Kevin M. Jenkins c , Michael S. Young c , Jan G. Ramaekers d , Gert De Boeck b , Nele Samyn b, * a Waters Corporation, MS Technologies Centre, Manchester, UK b Federal Public Service Justice, National Institute of Criminalistics and Criminology (NICC), Vilvoordsesteenweg 100, 1120 Brussels, Belgium c Waters Corporation, Milford, MA, USA d Experimental Psychopharmacology Unit, Brain and Behaviour Institute, Maastricht University, Maastricht, The Netherlands Received 4 October 2004; received in revised form 13 November 2004; accepted 14 November 2004 Available online 18 April 2005 Abstract We present a validated method for the simultaneous analysis of basic drugs which comprises a sample clean-up step, using mixed-mode solid-phase extraction (SPE), followed by LC–MS/MS analysis. Deuterated analogues for all of the analytes of interest were used for quantitation. The applied HPLC gradient ensured the elution of all the drugs examined within 14 min and produced chromatographic peaks of acceptable symmetry. Selectivity of the method was achieved by a combination of retention time, and two precursor-product ion transitions for the non-deuterated analogues. Oral fluid was collected with the Intercept 1 ,a FDA approved sampling device that is used on a large scale in the US for workplace drug testing. However, this collection system contains some ingredients (stabilizers and preservatives) that can cause substantial interferences, e.g. ion suppression or enhancement during LC–MS/MS analysis, in the absence of suitable sample pre-treatment. The use of the SPE was demonstrated to be highly effective and led to significant decreases in the interferences. Extraction was found to be both reproducible and efficient with recoveries >76% for all of the analytes. Furthermore, the processed samples were demonstrated to be stable for 48 h, except for cocaine and benzoylecgonine, where a slight negative trend was observed, but did not compromise the quantitation. In all cases the method was linear over the range investigated (2–200 mg/L) with an excellent intraassay and inter-assay precision (coefficients of variation <10% in most cases) for QC samples spiked at a concentration of 4, 12 and 100 mg/L. Limits of quantitation were estimated to be at 2 mg/L with limits of detection ranging from 0.2 to 0.5 mg/L, which meets the requirements of SAMHSA for oral fluid testing in the workplace. The method was subsequently applied to the analysis of Intercept 1 samples collected at the roadside by the police, and to determine MDMA and MDA levels in oral fluid samples from a controlled study. #2005 Elsevier Ireland Ltd. All rights reserved. Keywords: LC–MS/MS; Oral fluid; SPE; Ion suppression www.elsevier.com/locate/forsciint Forensic Science International 150 (2005) 227–238 * Corresponding author. Tel.: +32 2 240 05 00; fax: +32 2 242 47 61. E-mail address: [email protected]v.be (N. Samyn). 0379-0738/$ – see front matter #2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.forsciint.2004.11.027 1. Introduction A variety of body specimens other than urine, such as saliva (oral fluid), sweat and hair have been used to document drug exposure for pre-employment screening, in forensic toxicology laboratories and in clinical applications [1–4]. The advantage of these samples over traditional matrices like urine and blood is that collection is almost non-invasive, relatively easy to perform, and may be achieved under close supervision to prevent adulteration or substitution of the sample. Tools for the detection of drugs in alternative specimens often utilise traditional technology, though some limitations are imposed which require special attention: the specimen volume or mass is often small, the target analytes are different from urine and the analyte concentration is lower than in urine [1–5]. Oral fluid can be extracted and analysed like other biological fluids such as blood. In general, there will be less interference from endogenous compounds than with blood or urine [1]. Laboratory immunoassay procedures (EIA) to screen for drugs of abuse in oral fluid have been validated for cocaine [6,7] opiates [8,9] and cannabinoids [10]. Many reports on oral fluid refer to the common GC–MS procedures for certain classes of drugs in blood using electron impact mode. These methods utilise the deuterated analogues for parent drugs, e.g. D 9 -tetrahydrocannabinol (THC) and cocaine and also for relevant metabolites, e.g. 6-acetylmorphine (6-AM) [1,4,5]. Due to the small sample volume of oral fluid specimens, chromatographic procedures using tandem mass spectrometry, either GC or LC, have been developed for the detection of a single class of analytes [11,12] or for the analysis of multiple classes of illicit drugs simultaneously [13,14]. The high specificity and the increased signal-to-noise in combination with short chromatographic run times and a potential to reduce sample preparation because there is no need for derivatization, make LC–MS/MS the technique of choice for high-throughput confirmation of multiple illicit drugs in oral fluid samples. One major issue that needs to be addressed is the choice of sampling method and the influence of the collected matrix on the LC–MS/MS analysis. Mortier et al. used liquid chromatography quadrupole-time-of-flight mass spectrometry with electrospray ionisation to successfully determine morphine, codeine, cocaine, benzoylecgonine and amphetamines in oral fluid samples obtained by spitting and subjected to solid-phase extraction (SPE). However, when authentic samples collected with a specific device were analysed, interferences were noticed compromising the quantitative analysis [13]. Dams et al. investigated the influence of oral fluid matrix components (samples collected with a neutral Salivette 1 ) during post-column infusion of morphine, after different sample preparation steps [15]. The Intercept 1 is a FDA approved sampling device that is used on a large scale in the US for workplace drug testing [11]. It is also used to collect oral fluid samples for confirmation analyses in the joint roadside study between the European Union and the US to detect driving under the influence of drugs [16]. The collection system contains stabilizing salts, non-ionic surfactants for surface wetting and antibacterial agents, and guarantees a good stability for most illicit drugs and their metabolites during storage at 48C. However, these ingredients can also cause interferences, e.g. ion suppression during LC–MS/MS analysis in the absence of a suitable clean-up method. We have validated a newly developed LC–ESI–MS/MS method combined with a routine SPE clean-up for the simultaneous quantitation of the major analytes which can be detected after consumption of basic illicit drugs; the oral fluid was collected using the Intercept 1 device. The method was applied to Intercept 1 samples collected at the roadside and after a controlled administration of MDMA to volunteers. 2. Materials and methods 2.1. Materials Ammonium formate and ammonium hydrogen carbonate were purchased from Sigma–Aldrich (Steinheim, Germany). Tetrahydrofuran (chromatographic grade) and hydrochloric acid (fuming, 37%) were from the same supplier. Formic acid, ammonia solution (32%, extra pure), hydrochloric acid solution (0.1N) and all other solvents (HPLC-grade) were purchased from Merck (Darmstadt, Germany). Solid-phase extraction (SPE) cartridges Oasis 1 MCX (30 mg, 1 cm 3 ) were from Waters (Milford, MA). Individual stock solutions of the drugs and their deuterated analogues were purchased from LGC Promochem (Molsheim, France). Stock solutions of amphetamine-d 11 , methamphetamine-d 5 , MDMA-d 5 , MDA-d 5 , cocaine-d 3 , benzoylecgonine-d 8 , morphine-d 3 , codeine-d 6 and 6-acetylmorphine-d 6 were obtained at concentrations of 0.1 g/L in methanol or acetonitrile. Amphetamine, methamphetamine, MDMA, MDA, cocaine, benzoylecgonine, morphine, codeine and 6-AM were certified at a concentration of 1 g/L in methanol or acetonitrile. Separate working solutions of the drugs, for tuning and selectivity experiments, were prepared in the laboratory at a concentration of 1 mg/L in methanol. A mixed working solution of non-deuterated compounds at 4 mg/L in methanol was used for the preparation of calibrators and QC samples within each run. A mixed internal standard working solution of 1 mg/L was prepared in methanol. Working solutions were stored at 20 8C, and were prepared monthly. To obtain the lower concentrations needed for internal standardization and validation of each experiment, further dilutions in water were prepared the same day. 2.2. Specimens Oral fluid used for the preparation of blanks, calibrators and QC samples was obtained from healthy volunteers and M. Wood et al. / Forensic Science International 150 (2005) 227–238228 collected with the Intercept 1 collection device (OraSure Technologies, Bethlehem, PA) according to the manufacturer’s instructions. Briefly, after gently wiping the collector pad between gum and cheek for approximately 2 min (as a kind of toothbrush), the device is placed in the supplied vial, which contains a stabilizing buffer solution, and sealed. After centrifugation, the recovered fluid was transferred in cryotubes and represents a mixture of the collected oral fluid and the buffer in a proportion of approximately 1:2. The device collects an average of 0.38 0.19 mL of oral fluid and a dilution factor of 1 in 3 is arbitrarily accepted [11]. The tubes were sealed and stored at 20 8C prior to analysis. Authentic oral fluid samples were collected by the police at the roadside during roadblocks to intercept drivers under the influence of drugs, using the same procedure as described for the blank samples. A third series of oral fluid samples were obtained with a similar protocol from 18 volunteers who received either placebo or a high (100 mg) or a low (75 mg) dose of MDMA. Oral fluid samples were collected at 1.5 and 5.5 h after administration of the drug. The study protocol was approved by the ethics committee of the University Hospital of Maastricht in The Netherlands. 2.3. Sample preparation Twenty-five microliters of concentrated hydrochloric acid, 50 mL of an internal standard working solution (at 0.2 mg/L) and 750 mL of water were added to 250 mL of oral fluid specimen collected with the Intercept 1 device. After conditioning with 1 mL of methanol and 1 mL of 0.1N hydrochloric acid, the diluted oral fluid samples were applied onto the SPE columns. Clean-up was accomplished with successive 1 mL washes of 0.1N HCl, tetrahydrofuran and a mixture of methanol and water (50:50, v/v). The cartridges were dried by applying full vacuum for 5 min before elution with 0.5 mL of 5% ammonia in methanol. After the extraction, the elution solution was treated according to a variety of different protocols: a simple dilution with 1 mL of water which had been previously proposed [17]; a fairly rapid and controllable concentration step to 50–100 mL, performed in a vacuum centrifuge (Jouan RC 10.22) at 40 8C, followed by addition of 950 mL of ammonium formate buffer (10 mM, with 0.01% formic acid, pH 4.2) before injection onto the LC system; complete evaporation of the elution solution performed with and without adding 50 mL of a 5% hydrochloric acid solution in methanol before taking to dryness. In the latter case thedry residue was then reconstituted in 1 mL of a mixture of the ammonium formate buffer pH 4.2 and methanol (95/5, v/v). 2.4. Chromatographic conditions LC was performed using a Waters Alliance 2695 separation module. All aspects of system operation and data acquisition were controlled using MassLynx NT 3.5 software (Micromass UK Limited, UK). Analytes were separated on a XTerra MS C 18 column (2.1 mm 150 mm, 3.5 mm) (Waters) using a gradient elution with 10 mM ammonium bicarbonate (pH 10) (A) and methanol (B), at a flow rate of 0.25 mL/min. A gradient was carried out starting from 30% B at 3 min, B was then increased to 50% over the next 1 min. From 4 min to 12 min, B was linearly increased to 75%. At 12 min, B was increased to 90% in 1 min before returning to its initial conditions within 0.1 min and equilibrating for 6.9 min, which resulted in a total run time of 20 min. An injection volume of 20 mL was used. 2.5. Mass spectrometry A Quattro Ultima tandem mass spectrometer (Micromass UK Limited, UK) fitted with a Z-Spray ion interface was used for all analyses. Ionisation was achieved using electrospray in the positive ionisation mode (ES+). The following conditions were found to be optimal for the analysis: capillary voltage, 1.0 kV; source block temperature, 120 8C; desolvation gas (nitrogen) heated to 350 8C and delivered at a flow rate of 800 L/h. The appropriate multiple reaction monitoring (MRM) conditions for the individual analytes and their respective deuterated analogues, were determined by direct infusion into the mass spectrometer. The cone voltage (CV) was adjusted to maximise the intensity of the protonated molecular species [M + H] + and collision induced dissociation of each protonated molecule was performed. Collision gas (argon) pressure was maintained at 2.7 10 3 mbar and the collision energy (eV) adjusted to optimise the signal for the most abundant product ions, which were subsequently used for MRM analysis. 2.6. Method validation 2.6.1. Selectivity, stability, recovery, assessment of matrix effects The ability of the analytical method to differentiate and quantify the analyte in the presence of other components in the matrix and of other target analytes was assessed by including blank specimens from a different origin in every run, and by injection of single analyte solutions and evaluation of the MRM transition signal. Analyte stability in the final extract was checked by repeated injections of an extracted calibrator at 80 mg/L over 48 h, and plotting of the absolute peak areas as a function of time. Recoveries were estimated by comparing the responses of an 80 mg/L calibrator when the non-deuterated compounds were added before the extraction step with those obtained when the non-deuterated analytes were added after sample preparation. To assess any potential suppression or enhancement of ionisation due to the sample matrix, two types of experiments were performed. In the first experiment, standards were added after the sample pre-treatment, i.e. just before M. Wood et al. / Forensic Science International 150 (2005) 227–238 229 injection and the peak responses were compared to those obtained from a methanolic standard diluted in the same volume of ammonium formate buffer. In addition, for an assessment of the effects of untreated samples, the diluted oral fluid samples that would normally be applied to the SPE cartridge were also directly injected into the LC. The second type of experiment involved a continuous post-column infusion of a mixture of the analytes of interest and their deuterated analogues (10 mg/L at a flow rate of 10 mL/min) to produce a constant elevated response in each MRM channel. The interference of this constant response was monitored following the injection of samples either prior to or after SPE clean-up. 2.6.2. Linearity, intra-assay and inter-assay precision, accuracy Quantitation was performed by integration of the area under the specific MRM chromatograms in reference to the integrated area of its respective deuterated analogue which was added before the extraction procedure. Freshly prepared working solutions of 0.02, 0.1, 0.4 and 1 mg/L in water were used to prepare oral fluid calibrators at a concentration of 2, 10, 20, 40, 80, 120 and 200 mg/L. Standard curves were freshly prepared with each batch of QC and authentic samples. Standard curves were generated using a least-squares linear regression, with a 1/xweighting factor. Quality control samples (QC) were prepared for every run in blank oral fluid M. Wood et al. / Forensic Science International 150 (2005) 227–238230 Table 1 MRM transitions and conditions for all compounds and their deuterated analogues Compound Precursor ion (m/z) Product ions (m/z) Cone voltage (V) Collision energy (eV) Amphetamine 136.10 119.10, 91.00 20 9, 17 Amphetamine-d 11 147.10 98.00 35 18 Methamphetamine 150.10 119.00, 91.00 20 9, 20 Methamphetamine-d 5 155.10 92.00 20 20 MDA 180.05 105.00, 77.00 20 22, 30 MDA-d 5 185.00 168.10 20 10 MDMA 194.05 163.05, 105.10 40 12, 25 MDMA-d 5 199.10 165.10 40 13 Cocaine 304.15 182.10, 82.10 20 18, 28 Cocaine-d 3 307.15 185.10 22 20 Benzoylecgonine 290.15 168.10, 105.00 45 20, 30 Benzoylecgonine-d 8 298.10 171.00 45 20 Morphine 286.10 165.10, 152.00 75 40, 57 Morphine-d 3 289.00 165.10 75 47 6-AM 328.10 165.00, 152.00 80 40, 70 6-AM-d 6 334.20 165.10 75 40 Codeine 300.10 165.10, 128.10 70 43, 58 Codeine-d 6 306.20 165.10 70 45 Italicised transitions were used for quantitation. Table 2 Stability of 6-AM and cocaine in oral fluid samples following a variety of post-SPE protocols Post-SPE protocols Results pH Stability over 48 h 6-AM Cocaine 1 Dilution with water 11 Unstable Unstable 2 Concentration to 50–100 mL and dilution with water 8–9 Stable Unstable 3 Concentration to 50–100 mL and reconstitution in 0.95 mL ammmonium formate buffer 10 mM; 0.01% formic acid) a 4–5 Stable Stable 4 Complete evaporation and reconstitution in 0.95 mL ammmonium formate buffer 10 mM:0.01% formic acid) b 4-5 Stable Stable 5 Concentration to 100 mL, addition of 50 mL MeOH:HCl (95:5), complete evaporation and reconstitution in 0.95 mL ammmonium formate buffer 10 mM:0.01% formic acid) c 4–5 Stable Stable a Regression analysis showed a slight negative trend for cocaine and, to a lesser extent, for benzoylecgonine over a period of 48 h. The decrease in peak area was significantly different from zero (t-test, p<0.05). For all other compounds, there was no significant change of the peak area. b Loss of amphetamines has been reported in the evaporation process. c This procedure extends the sample preparation considerably. at a concentration of 4, 12 and 100 mg/L. Intra-assay precision was evaluated by replicate (n= 4) analysis of the three QC samples in one run. Inter-assay precision was evaluated by replicate analysis of the QC samples in several experiments performed on four different days by two operators. Comparing the calculated concentrations of all calibrators and QC samples to their respective nominal values, provided data on the accuracy of the method. 2.6.3. Limit of detection and limit of quantitation The limit of quantitation (LOQ) was defined in this study as the lowest calibrator with an acceptable relative uncertainty (coefficient of variation 20% and an accuracy of 100 20%). The limit of detection (LOD) was estimated from extracted oral fluid samples, spiked with decreasing concentrations of the analytes, where the response of the quantitative ion was equal to 10 times the response of the blank extract. 3. Results and discussion 3.1. Method validation The applied gradient ensured the elution of all the drugs examined within 14 min and produced chromatographic peaks of acceptable symmetry. Selectivity of the method was achieved by a combination of retention time, precursor and product ions. With the exception of MDA, the most prominent precursor-product transitions were used for quantitation of the non-deuterated compounds and the next most abundant, used as qualifiers. For MDA an elevated background response was noted when using the MRM transition based on the most prominent product, i.e. m/z180 >163. Improved sensitivity (based on signal-to-noise) was achieved when the MRM transition utilised an alternative product ion (Table 1). For the corresponding deuterated analogues, only one transition was monitored. Injection of single analyte solutions did not produce interference in the other MRM channels. One of the limiting factors of LC–MS(-MS) applications is the potential presence of a matrix effect, leading to suppression or enhancement of the analyte response. This typically occurs as a result of insufficient clean-up of the matrix, and although partly overcome by the use of deuterated internal standards, it leads to variable sensitivities, and decreased precision and accuracy. Dams et al. [15] observed no signal suppression for morphine in oral fluid collected with a Salivette 1 after a simple dilution step, and a small ion suppression effect after SPE (maximum 10–15%) in the ESI mode. Protein precipitation resulted in suppression of 50– 70% in some areas of the chromatogram. The Intercept 1 collector contains a variety of chemicals, i.e. sodium chloride, sodium benzoate, potassium sorbate, bovine gelatin, tween 20, chlorhexidine digluconate and a blue dye, some of which can interfere with the LC–MS/MS detection signal. After direct injection of diluted spiked oral fluid samples collected from seven individuals and analysed without SPE clean-up, relatively small variations in the responses for cocaine and benzoylecgonine (ranging from 12% suppression to 24% enhancement) were observed. In all cases there was an enhancement in the responses for MDMA and methamphetamine, which was very variable ranging from 2 to 62%. The most dramatic effect was the suppression of the analyte signal for morphine and 6-AM (68–87%), and to a lesser extent for codeine, amphetamine and MDA (33– 67%). The Oasis 1 MCX (30 mg, 1 cm 3 ) SPE cartridges utilise mixed-mode (cation-exchange) sorbents, which proM. Wood et al. / Forensic Science International 150 (2005) 227–238 231 Table 3 Extraction recovery and matrix suppression Compound Percent recovery (mean 1S.D.) Estimated effect after SPE (%) Amphetamine 90.7 4.4 6.2 Methamphetamine 83.7 4.1 4.6 MDA 82.2 2.9 7.8 MDMA 76.6 2.3 1.9 Cocaine 93.1 0.7 +5.1 Benzoylecgonine 93.9 2.0 +12 Morphine 99.0 0.9 13 6-AM 91.8 1.6 9.8 Codeine 94.4 2.0 +3.4 Data represent the mean of four experiments with an 80 mg/L calibrator. Table 4 Equation of a typical calibration curve with coefficient of determination (r 2 ), and the estimated limits of quantitation (LOQ) of the method Compound Equation r 2 LOQ (mg/L) Amphetamine y= 1.0306x+ 0.1952 0.999735 2.0 Methamphetamine y= 1.8859x+ 0.0032 0.999279 2.0 MDA y= 0.4206x0.0843 0.999731 2.0 MDMA y= 1.1554x+ 0.0752 0.999598 2.0 Cocaine y= 1.0043x+ 0.8846 0.999952 2.0 Benzoylecgonine y= 1.4007x+ 0.2487 0.999781 2.0 Morphine y= 1.4769x+ 0.0991 0.999467 2.0 6-AM y= 0.9387x+ 0.1621 0.999717 2.0 Codeine y= 1.0836x+ 0.0802 0.999288 2.0 Leenheer, Simultaneous, quantitative determination of opiates, amphetamines, cocaine and benzoylecgonine in oral fluid by liquid chromatography quadrupole-time-of-flight mass spectrometry, J. Chromatogr. B 779 (2002) 321–330. [14] R. Dams, C.M. Murphy, R.E. Choo, W.E. Lambert, A.P. De Leenheer, M.A. Huestis, LC-APCI-MS/MS analysis of multiple illicit drugs, methadone, and their metabolites in oral fluid following protein precipitation, Anal. Chem. 75 (2003) 798– 804. [15] R. Dams, M.A. Huestis, W.E. Lambert, C.M. Murphy, Matrix effect in bio-analysis of illicit drugs with LC-MS/MS: influence of ionization type, sample preparation, and biofluid, J. Am. Soc. Mass Spectrom. 14 (2003) 1290–1294. [16] http://www.rosita.org. [17] K.M. Jenkins, M.S. Young, M. Wood, Solid-phase extraction and tandem LC/MS determination of drugs of abuse in preserved saliva, in: Proceedings of the 52nd ASMS meeting, May 2004, Nashville, Tennessee, published on CD rom, THPF: poster 101. [18] N. Samyn, G. De Boeck, M. Wood, C.T.J. Lamers, D. De Waard, K.A. Brookhuis, A.G. Verstraete, W.J. Riedel, Plasma, oral fluid and sweat wipe ecstasy concentrations in controlled and real-life conditions, Forensic Sci. Int. 128 (2002) 90–97. [19] M. Navarro, S. Pichini, M. Farre ´, J. Ortuno, P.N. Roset, J. Segura, R. de la Torre, Usefulness of saliva for measurement of 3,4-methylenedioxymethamphetamine and its metabolites: correlation with plasma drug concentrations and effect of salivary pH, Clin. Chem. 47 (2001) 1788–1795. [20] L. Presley, M. Lehrer, W. Seiter, D. Hahn, B. Rowland, M. Smith, K.W. Kardos, D. Fritch, S. Salamone, S. Niedbala, E.J. Cone, High prevalence of 6-acetylmorphine in morphinepositive oral fluid specimens, Forensic Sci. Int. 133 (2003) 22–25. [21] R. Dams, Bio-analysis of opioids by liquid chromatography– tandem mass spectrometry. Doctoral thesis, Ghent University, 2003. M. Wood et al. / Forensic Science International 150 (2005) 227–238238