Screening and Confirmation of 62 Drugs of Abuse and Metabolites in Urine by
Ultra-High-Performance Liquid Chromatography – Quadrupole Time-of-Flight Mass
Spectrometry
I-Lin Tsai1,2, Te-I Weng3,4, Yufeng J. Tseng1,2,5,6, Happy Kuy-Lok Tan7, Hsiao-Ju Sun7and Ching-Hua Kuo1,2,3,8* 1
School of Pharmacy, College of Medicine, National Taiwan University, No. 1, Section 1, Jen-Ai Road, Taipei 100, Taiwan,2The Metabolomics Core Laboratory, Center of Genomic Medicine, National Taiwan University, Taipei, Taiwan,3Forensic and Clinical Toxicology Center, National Taiwan University College of Medicine and National Taiwan University Hospital, Taipei, Taiwan,4Institute of Toxicology, College of Medicine, National Taiwan University, Taipei, Taiwan,5Graduate Institute of Biomedical Electronics and Bioinformatics, National Taiwan University, Taipei, Taiwan,6Department of Computer Science and Information Engineering, National Taiwan University, Taipei, Taiwan,7Department of General Psychiatry, Taoyan Mental Hospital, Taoyan, Taiwan and8Department of Pharmacy, National Taiwan University Hospital, Taipei, Taiwan
*Author to whom correspondence should be addressed. Email: [email protected]
An ultra-high-performance liquid chromatography – quadrupole time-of-flight mass spectrometry (UHPLC – QTOF-MS) method for the screening and confirmation of 62 drugs of abuse and their metabo-lites in urine was developed in this study. The most commonly abused drugs, including amphetamines, opioids, cocaine, benzodia-zepines (BZDs) and barbiturates, and many other new and emerging abused drugs, were selected as the analytes for this study. Urine samples were diluted 5-fold with deionized water before analysis. Using a superficially porous micro-particulate column and an acetic acid-based mobile phase, 54 basic and 8 acidic analytes could be detected within 15 and 12 min in positive and negative ionization modes, respectively. The MS collision energies for the 62 analytes were optimized, and their respective fragmentation patterns were constructed in the in-house library for confirmatory analysis. The coefficients of variation of the intra- and inter-day precision of the analyte responses all were <17.39%. All analytes, except barbital, showed matrix effects of 77 – 121%. The limits of detection of the 62 analytes were between 2.8 and 187.5 ng/mL, which were lower than their respective cut-off concentrations (20 – 500 ng/mL). Ten urine samples from patients undergoing methadone treatment were analyzed by the developed UHPLC – QTOF-MS method, and the results were compared with the immunoassay method.
Introduction
Drug abuse is a major cause of social problems worldwide. The prevalence of drugs of abuse, including marijuana, cocaine, heroin, hallucinogens and prescription-type drugs used nonmedically in the USA from 2002 to 2011, was 7.9–8.9% (20 million people) among individuals who were at least 12 years old (1). Among these abused drugs, marijuana is the most commonly used, with a prevalence of 6–7%. The prevalence of marijuana abuse in European countries was as high as 10% within the recent decade. Cocaine, amphetamines and lysergic acid diethylamide (LSD) also showed high abuse rates in Europe (2). Other abused substances, such as sedative-hypnotics, could induce suicide, murder, sexual assault and traffic accidents. Identifying abused drugs in biological samples provides scientific evi-dence in court against criminals and victims and could improve the quality of clinical management in emergencies. Therefore, develop-ing a sensitive and comprehensive analytical system to analyze abused drugs is important in clinical and forensic toxicology.
Immunoassays are commonly used as a first-line screening method in the detection of abused drugs in urine or other
biofluids (3,4). Although these methods are convenient, insuffi-cient specificity and limited coverage of drugs remain major lim-itations of immunoassays. For example, compounds with structures similar to the target drug will influence the result of such tests. Therefore, a second analytical procedure, such as gas chromatography – mass spectrometry (GC – MS) or liquid chro-matography – mass spectrometry (LC – MS), is always necessary for drug confirmation.
GC – MS and LC – MS have high sensitivity and good separation power, and they have been used as not only secondary analytical tools, but also independent analytical methods to identify drugs of abuse. The mass spectra obtained from GC–MS can be compared with large databases of reference compounds, which facilitate the identification of abused drugs. However, most compounds of inter-est need to be derivatized with acetylating or alkylating agents before GC–MS analysis (5,6). LC is an alternative technique that is especially useful for polar and thermolabile compounds. The ad-vantage of simple sample preparation makes LC–MS a rapidly growing field in the detection of abused drugs. Among the differ-ent mass spectrometry platforms, triple quadrupole mass spec-trometry with multiple reaction monitoring (MRM) is currently the most commonly adapted technique for the quantitative ana-lysis of abused drugs (7,8). Methods using LC–MS-MS for targeted and nontargeted screening approaches have also been developed in recent years (9). However, only compounds with pre-established transition ion pairs can be detected, and the sensitivity of triple quadrupole mass spectrometry decreases as the number of tran-sition ion pairs increases in the MRM mode (10).
High-resolution mass spectrometry, such as time-of-flight mass spectrometry (TOF-MS), has become an emerging tech-nique for high throughput toxicological screening in recent years (11–15). It offers accurate mass measurement with relative accuracy at the parts per million ( ppm) level, and the absolute ac-curacy is typically in the mDa range. TOF-MS also provides excel-lent full-scan sensitivity, and comprehensive drug screening can be performed without predefined target analytes. The high mass ac-curacy enables the use of exact monoisotopic masses and isotopic patterns for compound identification. This advantage allows for the detection of rare chemicals or metabolites, which have stan-dards that are difficult to acquire. Moreover, when screening for additional target drugs or designer compounds in previously ana-lyzed samples, the samples do not need to be reanaana-lyzed because Journal of Analytical Toxicology 2013;37:642 – 651
the TOF-MS analytical data can be reprocessed. The successful ap-plication of TOF-MS in toxicological screening has been achieved by Ojanpera¨, Kolmonen, Pelander, and Gergov et al., who devel-oped several screening methods for doping agents and forensic drugs in biological samples by high-performance liquid chromato-graphy (HPLC)–TOF-MS (12,16–18). Other groups also applied LC–TOF-MS to screen for abused drugs in various biological samples (19–22). One problem that has been reported for the application of TOF-MS to abused-drug screening is the inevitable false-positive results when only accurate mass and isotopic pattern matching are applied for identification. Confirmation with fragmen-tation spectra is an effective method to minimize false identification (9). Therefore, the integration of screening and the confirmation of abused drugs by quadrupole TOF-MS (QTOF-MS) were recently investigated. Saleh et al. recently used nine abused drugs as ana-lytes to compare the screening performance of LC–TOF-MS and immunochemical methods. The results showed that LC– TOF-MS has potential as a replacement of the immunoassay method for drug screening, which provides advantages in terms of universal-ity, sensitivity and the selective detection of abused drugs (10). With the integration of screening and the confirmation on the same ultra-high-performance liquid chromatography–QTOF-MS (UHPLC–QTOF-MS) analytical platform, the simplicity of an ana-lytical workflow in a forensic lab could be further improved.
To increase the coverage of the abused drugs, this study used the most commonly abused drugs, including amphetamines, opioids, cocaine, BZDs, ketamine (KET), LSD, cannabis and many other new and emerging abused drugs, as our analytes. Metabolites, including norketamine, norephedrine, 7-aminoflunitrazepam, nordiazepam and 11-nor-9-carboxy-tetrahydrocannabinol (THCCOOH), were in-cluded because the presence of metabolites in urine provides stron-ger evidence of drug use. We aimed to develop a simple and effective UHPLC–QTOF-MS method for the comprehensive screen-ing and confirmation of abused drugs. The screenscreen-ing step was achieved by identification with the retention time, accurate mass and isotopic pattern, and the candidate compounds were further analyzed in an MS tandem mode to compare their fragment mass spectra with the library data to confirm the candidates’ results using the same analytical platform. The validated UHPLC–QTOF-MS method was applied to analyze 10 urine samples from patients undergoing methadone therapy, and the results were compared with the test results obtained by the immunoassay method.
Experimental
Standards and reagents
Amphetamine, alprazolam, 7-aminoflunitrazepam, amobarbital, aminorex, bromazepam, buprenorphine, butalbital, butorphanol, 4-bromo-2,5-dimethoxyphenethylamine (2C-B), butabarbital, clonazepam, chlordiazepoxide, clobazam, dihydrocodeine, diaze-pam, ephedrine, estazolam, fentanyl, flurazediaze-pam, flunitrazepam (FM2), glutethimide, heroin, KET, lorazepam, lormetazepam, LSD, methamphetamine, 4-methoxyamphetamine (PMA), 3,4-methyle-nedioxyamphetamine (MDA), 3,4-methylenedioxymethampheta-mine (MDMA), 3,4-methylenedioxy-N-ethylampheta3,4-methylenedioxymethampheta-mine (MDEA), para-methoxymethamphetamine (PMMA), meperidine, metha-done, midazolam, meprobamate, methylephedrine, methylphen-idate, norketamine, norephedrine, nitrazepam, nordiazepam, nalorphine, 11-nor-9-carboxy-THC (THCCOOH), oxazepam,
pentazocine, phentermine, prazepam, pseudoephedrine, seco-barbital, triazolam, temazepam, tramadol and zolpidem were purchased from Cerilliant (Round Rock, TX, USA). Cocaine hydrochloride, codeine, morphine, pentobarbital, barbital and phenobarbital were purchased from Sigma-Aldrich (St. Louis, MO, USA). Acetonitrile and acetic acid were purchased from Merck (Darmstadt, Germany). Formic acid was purchased from Riedel-de Hae¨n (Sigma-Aldrich). Methanol (MeOH) and water were purchased from Scharlau (Spain). All reagents and solvents used were of analytical or chromatographic grade.
Drug-free urine samples were donated by healthy volunteers. The drug-free urine samples were verified to not contain drugs before use. Urine samples with abused drugs were collected from patients undergoing methadone maintenance treatment (MMT) in the Taoyan Mental Hospital. The study procedure was approved by the institutional review board and all participants provided written informed consent before receiving screening tests. The samples were screened by an immunoassay method after they were collected in the hospital. The competitive im-munoassay can detect 12 types of drugs in urine, including acet-aminophen, amphetamines, methamphetamines, barbiturates, BZDs, cocaine, methadone, opiates, phencyclidine, THC, tricyc-lic antidepressants and KET (ketamine: Firstep Bioresearch, Tainan, Taiwan; other drugs: Triagew
Tox Drug Screen, Biosite, San Diego, CA, USA).
Sample preparation
One hundred microliters of the urine sample were diluted with 400 mL of water. The diluted sample was then centrifuged at 15,000 g for 5 min, and the supernatant (200 mL) was subjected to UHPLC – QTOF-MS analysis.
Ultra-high-performance liquid chromatography – quadrupole time-of-flight mass spectrometry
The Agilent 1290 UHPLC system consisted of a degasser and a quaternary solvent pump (Agilent Technologies, Waldbronn, Germany). A Poroshell EC-C18 column (2.1 100 mm, 2.7 mm, Agilent) was used for compound separation. The mobile phase was composed of 0.1% acetic acid (Solvent A) and MeOH (Solvent B). The mass spectrometer was an Agilent 6540 QTOF-MS equipped with an electrospray ion source. Mass spectrometry cali-bration was performed daily before analysis by the infusion of a low concentration of a tuning mix (Agilent Technologies, USA). The gradient profile used for positive ionization detection was as follows: 0 –1 min, 2% B; 1– 10 min, 2– 50% B; 10–15 min, 50–90% B; 15–17 min, 90% B and then re-equilibration of the column for 3 min. The gradient for negative ionization detection was as follows: 0– 8 min, 25–35% B; 8– 10 min, 35– 85% B; 10– 15 min, 85% B and then re-equilibration of the column for 3 min. The flow rate was maintained at 0.4 mL/min, and the injection volume was 5 mL. The parameters of the mass spectrometer for positive and negative ionization mode detection were as follows: sheath gas temperature, 3258C; sheath gas flow, 11 L/min; nebulizer, 45 psi; capillary voltage, 3,000 V; gas temperature, 3258C ; drying gas flow, 6 L/min for positive ionization mode detection and 5 L/min for negative ionization mode detection; nozzle voltage, 1,000 V; fragmentor voltage, 120 V; skimmer 1, 65 V; octopole radio fre-quency (RF) peak, 750 V and TOF-MS scan range, 50–1000 m/z.
During the analysis, reference masses of 121.0509 and 922.009798 as well as 112.9856 and 966.0007 m/z were used for positive and negative mode mass accuracy correction, respectively.
Validation
Method validation was performed in terms of selectivity, intra-and inter-day precision of the analyte response intra-and retention times, matrix effect, limit of detection (LOD) and carryover. The cut-off concentrations listed in TableIfor the 62 analytes were defined according to previous reports and the regulations reported from the Department of Health, Executive Yuan of Taiwan (23–25).
Selectivity
Six drug-free urine samples were donated from three healthy females and three healthy males. Aliquots of each drug-free urine sample were treated according to the sample preparation method described in the section Sample preparation. The six drug-free urine samples were used to determine whether there was any endogenous interference with the same retention time and the exact mass of the abused drugs and their metabolites. In addition, 62 analytes were spiked into drug-free urine samples at the cut-off concentrations to evaluate the ability of the devel-oped method to distinguish the 62 different analytes.
Intra- and inter-day precision of the analyte response and retention times
For intra-day precision of the analyte response, urine samples were spiked with 62 analytes at a concentration that was 10 times the cut-off concentration (high concentration level, n ¼ 6) and at the cut-off concentration (low concentration level, n ¼ 6). Inter-day precision was also evaluated for high and low concentration-spiked urine samples. The samples were injected six times at three differ-ent time points. The intra- and inter-day precision of the analyte re-sponse was expressed as a coefficient of variation (CV, %). The intra- and inter-day precision of the retention times for all the 62 analytes were also calculated from the same data set and were expressed as CV, %.
Matrix effect
Drug-free urine samples from six healthy volunteers were used to evaluate matrix effects. After urine pretreatment with a 5-fold dilution, analytes were spiked into samples at concentrations that were 20 and 200% of the cut-off concentrations. Analytes with the same concentrations were prepared in deionized water. The matrix effect was calculated from the peak area ratio of compounds that were spiked in the matrix to those prepared in water and multiplied by 100%. The average and CV (%) of the matrix effect for 62 analytes at high and low concentrations were calculated.
Limit of detection
The LOD of each analyte was determined as the concentration at which the signal-to-noise ratio equals to 3 (S/N ¼ 3).
Carryover
Drug-free urine samples were spiked with analyte at a concen-tration that was 20-fold higher than the cut-off concenconcen-tration. Three blank water samples were injected following the spiked
urine sample to investigate carryover. Analytical data from the blank water samples were screened by the software with the para-meters described in the section Analysis of human urine samples. The peak heights of the detected analytes from the blank water sample were divided by that from the spiked sample and multi-plied by 100% to perform quantitative results of carryover.
Analysis of human urine samples
Ten urine samples with abused drugs were collected from patients undergoing MMT in the Taoyan Mental Hospital. The urine samples were first screened by immunoassay, and the results were compared with that from the developed UHPLC – QTOF-MS method. Before sample analysis, QC samples with 62 analytes at the cut-off concentrations were used to confirm the accuracy of retention times and signal responses before sample analysis. After pretreatment, the samples were first analyzed in the full-scan mode and then re-injected to collect MS – MS spectra for confirmation after data screening. Between each sample analysis, blank water was injected to prevent carryover. The analytical results from the first full-scan mode were automat-ically screened by Mass Hunter (Agilent Technologies) using the ‘Find by formula’ function. We inputted a list of 62 analytes with the compound name, retention time and molecular formula using a default document provided by Agilent Technologies in a comma-separated value file format. The screening match toler-ances were set as retention time tolerance, +0.2 min; relative mass accuracy (mass tolerance), +20 ppm and peak intensity, 1,500 counts. The relative mass accuracy was calculated by the following equation and given in ppm:
ðm=zexperimental m=ztheoreticalÞ m=ztheoretical
106;
where m/zexperimental is the m/z value provided by the mass spectrometry and m/z theoretical is the m/z value calculated through its empirical formula.
In addition, an overall score was calculated to screen out com-pounds: the weighted average of different matching factors, in-cluding retention time (weighting, 100), mass match (weighting, 100), isotope abundance (weighting, 60) and isotope spacing (weighting, 50). Higher overall scores indicated better matching results. Only compounds with an overall score .60 were sub-jected to MS–MS confirmation. In the confirmation step, the product ion spectrum of the candidate compound was compared with the spectrum of the standard. At least three diagnostic ions, including the parent ion, were used for spectrum confirmation. Tramadol is the only compound with two diagnostic ions because of its limited fragment ions. The relative abundance of the diag-nostic ion (% of the base peak) should not differ by .25% (rela-tive) compared with the standard spectrum (26).
Results and discussion
Sample pretreatment method development
Solid-phase extraction with a mixed-mode cation extraction cartridge is typically used as the sample pretreatment method for abused-drug screening (16,17). However, the multiple ex-traction steps involved are time consuming and labor intensive.
Table I
Validation results of 62 abused drugs and metabolites including intra- (n ¼ 6) and inter-day (three different occasions) precision for peak response and retention time, LOD and matrix effect (n ¼ 6, urine matrix from six healthy volunteers)
No ID Formula Cut-off (ng/
mL)
Rt (min)
Precision for peak response Precision for Rt LODa(ng/ mL)
Matrix effect-high Matrix effect-low Intra-day (CVb, %) Inter-day (CV, %) Intra-day (CV, %) Inter-day (CV, %)
Highc Lowd High Low Average
(%) CV (%) Average (%) CV (%) Basic compounds 1 Morphine C17H19NO3 200 2.60 1.19 1.23 3.25 5.47 0.18 0.31 2.8 103.63 9.50 106.64 13.18 2 Norephedrine C9H13NO 300 3.37 1.74 1.92 5.04 6.49 0.14 0.26 6.6 107.75 11.75 108.51 7.40 3 Ephedrine C10H15NO 300 4.01 1.00 2.07 5.52 5.21 0.05 0.22 11.8 101.01 9.27 110.82 19.75 4 Aminorex C9H10N2O 300 4.16 1.71 4.52 6.88 7.38 0.12 0.13 15.3 103.59 7.77 106.78 5.84 5 Pseudoephedrine C10H15NO 300 4.20 1.32 1.92 5.83 5.20 0.05 0.22 11.1 101.14 9.26 110.34 19.72 6 Nalorphine C19H21NO3 300 4.24 1.10 1.65 3.53 4.21 0.12 0.17 8.1 103.84 11.83 93.55 15.90 7 Methylephedrine C11H17NO 300 4.25 1.18 2.31 3.75 5.61 0.13 0.19 17.8 103.72 20.15 118.72 3.91 8 Dihydrocodeine C18H23NO3 300 4.30 0.86 2.21 4.12 3.76 0.08 0.08 7.8 101.99 9.66 111.90 10.65 9 Codeine C18H21NO3 300 4.35 1.34 1.40 6.78 6.41 0.10 0.11 5.1 107.32 12.96 109.73 4.48 10 Amphetamine C9H13N 300 4.80 1.32 4.26 6.42 14.46 0.06 0.16 8.9 96.60 19.25 103.44 4.11 11 Methamphetamine C10H15N 300 4.99 0.90 2.23 5.55 6.71 0.06 0.13 15.9 99.48 14.46 102.81 15.21 12 MDA C10H13NO2 500 5.13 1.18 12.49 5.60 14.84 0.05 0.18 35.5 95.39 15.38 98.60 19.25 13 MDMA C11H15NO2 500 5.22 0.65 1.91 5.43 4.05 0.04 0.34 17.5 99.63 7.28 105.15 7.71 14 PMA C10H15NO 300 5.42 1.23 1.58 7.00 6.77 0.08 0.26 21.4 101.72 9.83 101.71 7.45 15 PMMA C11H17NO 300 5.57 1.21 1.99 5.01 3.00 0.08 0.24 10.5 101.89 8.16 107.53 5.64 16 MDEA C12H17NO2 500 5.78 0.86 3.70 3.56 3.49 0.06 0.23 14.2 98.99 8.38 100.85 4.99 17 Phentermine C10H15N 300 6.00 1.12 2.18 6.53 6.17 0.04 0.25 24.1 101.39 9.97 103.67 7.56 18 Norketamine C12H14ClNO 100 6.52 1.47 3.60 6.89 7.08 0.07 0.08 4.9 100.62 5.69 104.94 6.38 19 Ketamine C13H16ClNO 100 6.55 1.62 1.61 4.56 7.26 0.09 0.10 11.5 101.44 8.99 109.19 6.91 20 Tramadol C16H25NO2 300 7.05 1.42 3.26 4.55 4.45 0.34 0.34 9.3 100.75 6.64 106.58 7.55 21 Heroin C21H23NO5 300 7.15 1.62 5.86 12.05 14.67 0.09 0.14 13.0 100.78 12.87 119.81 11.06 22 Cocaine C17H21NO4 300 7.15 2.68 2.77 6.63 6.56 0.09 0.22 15.7 98.61 9.28 112.77 7.47 23 Methylphenidate C14H19NO2 300 7.24 1.27 2.31 3.75 5.61 0.10 0.39 11.3 107.50 11.73 121.06 4.75 24 Meperidine C15H21NO2 200 7.70 1.08 1.39 4.11 4.50 0.38 0.41 8.8 103.15 8.93 109.46 7.22 25 2C-B C10H14BrNO2 300 7.85 1.30 4.43 9.62 8.32 0.09 0.12 10.4 105.52 7.95 101.08 15.21 26 Zolpidem C19H21N3O 300 8.02 0.91 1.90 3.58 4.18 0.13 0.11 9.6 101.68 9.06 108.61 7.63 27 7-Aminoflunitrazepam C16H14FN3O 300 8.12 1.18 1.05 7.57 3.92 0.05 0.07 9.2 105.72 7.84 98.97 10.57 28 LSD C20H25N3O 300 8.28 1.09 0.90 4.82 5.02 0.04 0.08 10.3 103.70 5.98 105.40 6.18 29 Butorphanol C21H29NO2 300 8.49 3.32 2.10 7.29 5.41 0.05 0.33 3.1 101.04 6.48 116.19 6.43 30 Pentazocine C19H27NO 200 8.57 1.38 3.77 7.08 8.58 0.06 0.18 5.3 108.27 5.53 103.78 4.90 31 PCP C17H25N 300 8.73 2.38 4.10 6.39 6.43 0.04 0.51 16.3 103.63 6.22 115.43 7.35 32 Meprobamate C9H18N2O4 300 9.08 3.49 14.26 7.98 14.32 0.04 0.08 187.5 103.48 7.52 115.17 9.97 33 Fentanyl C22H28N2O 200 9.17 0.55 1.61 5.75 6.10 0.06 0.07 7.1 91.48 13.90 110.60 12.68 34 Flurazepam C21H23ClFN3O 300 9.77 0.81 1.52 5.52 7.30 0.13 0.13 8.4 109.56 10.18 120.54 11.46 35 Midazolam C18H13ClFN3 300 10.21 2.48 3.06 7.41 8.97 0.08 0.07 16.1 97.08 4.95 99.44 4.91 36 Buprenorphine C29H41NO4 20 10.47 1.98 14.60 12.50 12.48 0.07 0.08 5.1 95.05 18.43 101.08 15.21 37 Bromazepam C14H10BrN3O 300 10.75 2.34 10.73 4.03 11.06 0.03 0.07 9.8 93.64 9.55 94.84 11.03 38 Glutethimide C13H15NO2 300 10.83 1.42 2.24 8.23 8.04 0.05 0.02 140.6 103.18 7.97 113.73 6.36 39 Chlordiazepoxide C16H14ClN3O 300 11.02 3.28 3.09 5.68 6.08 0.05 0.04 13.7 99.93 8.42 108.84 7.04 40 Nitrazepam C15H11N3O3 300 11.26 0.69 2.10 7.34 5.86 0.03 0.06 5.6 101.87 7.37 102.48 5.34 41 Clonazepam C15H10ClN3O3 300 11.37 3.24 6.45 17.39 8.32 0.03 0.03 5.4 95.37 11.88 112.08 5.84 42 Methadone C21H27NO 200 11.41 1.19 3.95 9.53 10.16 0.02 0.05 7.4 106.49 7.03 121.45 6.35 43 Flunitrazepam C16H12FN3O3 300 11.48 1.90 1.91 5.98 7.34 0.02 0.05 4.6 97.78 9.18 116.27 5.51 44 Estazolam C16H11ClN4 300 11.70 0.64 1.66 8.45 8.26 0.02 0.04 4.9 103.32 7.22 103.52 7.12 45 Clobazem C16H13ClN2O2 300 11.80 2.32 2.72 5.83 6.92 0.02 0.03 155.2 102.82 7.18 113.77 7.57 46 Oxazepam C15H11ClN2O2 300 12.04 0.51 2.47 7.90 7.55 0.02 0.04 9.2 99.02 8.88 106.10 7.16 47 Triazolam C17H12Cl2N4 300 12.06 2.15 1.98 6.00 7.25 0.02 0.03 8.5 101.46 8.36 113.07 7.19 48 Alprazolam C17H13ClN4 300 12.06 0.74 2.66 8.38 5.93 0.03 0.04 6.1 102.23 6.62 109.02 5.79 49 Lorazepam C15H10Cl2N2O2 300 12.07 0.38 3.12 7.34 10.11 0.03 0.04 12.9 102.24 7.87 105.83 7.05 50 Temazepam C16H13ClN2O2 300 12.33 2.64 2.05 7.04 7.57 0.03 0.04 97.7 101.07 8.20 111.96 4.62 51 Lormetazepam C16H12Cl2N2O2 300 12.50 1.31 3.15 10.49 8.91 0.02 0.03 4.1 102.66 6.44 107.23 7.54 52 Nordiazepam C15H11ClN2O 300 12.71 0.92 2.14 7.66 7.18 0.04 0.03 4.5 100.41 10.01 106.02 5.98 53 Diazepam C16H13ClN2O 300 12.99 2.11 3.20 6.74 8.06 0.02 0.03 4.0 101.44 7.61 111.43 6.18 54 Prazepam C19H17ClN2O 300 14.00 0.89 1.92 10.71 10.75 0.02 0.02 2.9 95.13 4.90 104.71 7.79 Acidic compounds 55 Barbital C8H12N2O3 500 2.08 1.60 1.39 2.16 4.32 0.15 0.26 77.3 39.30 19.26 29.54 21.71 56 Phenobarbital C12H12N2O3 500 5.12 3.40 0.79 2.83 4.29 0.08 0.38 17.9 96.87 7.48 91.05 5.15 57 Butabarbital C10H16N2O3 500 7.10 1.36 1.52 3.38 5.28 0.15 0.32 4.7 100.18 5.09 96.70 6.74 58 Butalbital C11H16N2O3 500 8.51 2.27 2.15 3.06 6.66 0.11 0.33 19.8 95.12 5.61 94.83 5.07 59 Pentobarbital C11H18N2O3 500 9.90 1.06 1.92 2.23 2.95 0.03 0.04 16.3 90.61 12.92 77.95 16.50 60 Amobarbital C11H18N2O3 500 9.91 3.91 3.05 3.96 4.47 0.01 0.04 17.9 91.90 11.61 79.55 14.16 61 Secobarbital C12H18N2O3 500 10.18 3.66 1.76 2.81 2.53 0.03 0.03 11.7 79.89 17.83 77.62 19.62 62 THCCOOH C21H28O4 50 11.66 4.52 13.01 6.03 11.45 0.03 0.05 22.1 99.17 9.02 101.32 13.28 Rt: retention time.
aLOD, limit of detection, defined as the concentration with a signal-to-noise ratio equal to 3 (S/N ¼ 3). bCV, coefficient of variation ¼ (standard deviation/mean) 100%.
cHigh concentration-spiked samples, the spiked concentration was 10 times the cut-off concentration. dLow concentration-spiked samples, the spiked concentration was one time the cut-off concentration.
Furthermore, because the target compounds have a wide variety of physical and chemical properties, their extraction recoveries when using only one type of solid phase extraction (SPE) cart-ridge will vary greatly. In this study, we used a sample dilution method instead of SPE as the sample pretreatment. With proper urine dilution in water, matrix effects can be reduced, and there will be no loss of analytes during sample preparation. Although higher-fold dilutions can decrease matrix effects, detection sensitivity will also be sacrificed. Therefore, dilution factors of 5 and 10 were tested, and the signal intensity and matrix effects were investigated to select the most suitable dilution factor. Buprenorphine, with a 20-ng/mL cut-off concentration, could not be detected in the urine sample if it was diluted 10 times with water. Because all the 62 analytes could be detected using a dilution factor of 5, this dilution was used to further investigate the matrix effects of the 62 analytes.
The matrix effects of 62 analytes were studied at high and low spiked concentrations, and the results are summarized in TableI. From the results of positive and negative ionization detection, only barbital, which had the shortest retention time (2.08 min), showed a matrix effect of30–40% at low and high spiked con-centrations. This was due to the coelution of hydrophilic en-dogenous interferences at the beginning of the separation. Because the signal intensity of barbital was relatively high, signal reduction caused by the matrix effect did not result in false-negative detection. Considering the sensitivity and matrix effect, a 5-fold dilution was chosen as the optimum dilution factor.
UHPLC – QTQF-MS method development
Most of the investigated abused drugs are weak bases with nitro-gen groups that can be detected by QTOF-MS in the positive ion-ization mode. However, barbiturates are acidic compounds that can only be detected by mass spectrometry in the negative ion-ization mode. Among all the 62 abused drugs and metabolites, 54 analytes were basic compounds and 8 analytes were acidic com-pounds. Therefore, positive and negative ionization detections were used in this study for the comprehensive screening of abused drugs. Although modern instruments provide fast polar-ity switching to detect positive and negative ions in a single ana-lytical run (20), we observed a significant decrease in the life span of the instrument component that controls the MS polarity switch (sampling capillary). As a result, basic and acidic drugs were analyzed in separate runs.
The LC column used in this study was a superficially porous micro-particulate packing column that provided analytical per-formance similar to a sub-2 mm particle column with lower backpressure. With this advantage, it was possible to use a mobile phase with higher viscosity or a higher flow rate to achieve efficient screening. Additionally, it can be used in HPLC and UHPLC, which improve the flexibility of the instrument’s use. To improve the method’s sensitivity, single and combination buffers consisting of 5 mM ammonium acetate, 0.1% formic acid and 0.1% acetic acid were tested for their effect on the signal in-tensities of the selected analytes. Owing to the different chem-ical properties of the 62 analytes, it was difficult to identify one mobile phase composition that was suitable for all of the com-pounds. From our results, most of the basic analytes showed high peak intensities when different buffers were used. However, barbiturates (acidic analytes) showed low peak
intensities when a mobile phase with 0.1% formic acid or 5 mM ammonium acetate with 0.1% formic acid was used. In contrast, the peak intensities of barbiturates improved significantly when using 0.1% acetic acid. The type of organic solvent did not sig-nificantly affect the peak intensity for most of the compounds. Considering the observation of peak intensity and cost effi-ciency, the final mobile phase composition was 0.1% acetic acid in water and MeOH.
This study used a simple dilution method for sample prepar-ation to improve the simplicity of the method. Because endogen-ous components were not removed during sample preparation, LC chromatographic conditions were optimized to minimize the matrix effect caused by endogenous components. For basic drug screening, the gradient was started from 2% MeOH to prevent early eluters, such as morphine, from coeluting with hydrophilic interferences. Under the developed gradient conditions described in the section Ultra-high-performance liquid chromatography– quadrupole time-of-flight mass spectrometry, the retention times of 54 basic abused drugs and metabolites were within 15 min. For acidic compounds, barbiturates were relatively more lipophilic, and the gradient was started at 25% MeOH. Under the developed gradient conditions, eight acidic drugs could be eluted within 12 min. The retention times of the investigated 62 analytes are listed in TableI, and the extracted ion chromatographs of 62 ana-lytes at their cut-off concentrations are presented in Figure1.
The mass spectrometry parameters, such as drying gas flow rate, sheath gas flow rate, drying gas temperature and fragmentor voltage, were all optimized. Because it is difficult to choose one set of optimized conditions for all 62 analytes, analytes with low cut-off concentrations, including buprenorphine and THCCOOH, were used to select mass spectrometry parameters in positive and negative ionization modes, respectively. Buprenorphine showed a higher intensity with a drying gas flow rate of 6 L/min, a sheath gas flow rate of 11 L/min and a drying gas temperature of 3258C. THCCOOH showed a higher intensity with a drying gas flow rate of 5 L/min, a sheath gas flow rate of 11 L/min and a drying gas temperature of 3258C. Additionally, the fragmentor voltage was set at 120 V because higher voltage caused frag-mentation of low-molecular-weight analytes, such as amphet-amine and norephedrine. Other MS conditions are described in the section Ultra-high-performance liquid chromatography – quadrupole time-of-flight mass spectrometry.
Validation
The qualitative validation parameters included selectivity, intra-and inter-day precision of the analyte responses intra-and retention times, matrix effect, LODs and carryover. For the selectivity test, urine samples from three healthy females and three healthy males were used. After a 5-fold dilution of the drug-free urine, each drug-free sample was analyzed by the UHPLC – QTOF-MS method in positive and negative modes. The results showed that there was no endogenous interference with the same exact masses and retention times as the 62 abused drugs and their metabolites. Spiked samples were used to test the ability of the method to distinguish structural isomers. Four sets of structural isomers were used in this study: (1) ephedrine/pseudoephe-drine/PMA with m/z 166.1226, (2) methamphetamine/phenter-mine with m/z 150.1277, (3) clobazem/temazepam with m/z 301.0738 and (4) pentobarbital/amobarbital with m/z
225.1245. Isomers in groups (1 – 3) could be distinguished by their specific retention times (TableI). Although pentobarbital and amobarbital had the same retention times under the devel-oped LC gradient, they could be separated with an isocratic LC condition; however, the isocratic condition resulted in a longer
analytical time. In this study, we used a step gradient to provide analytical efficiency, and the detected results of the two com-pounds were expressed as ‘pentobarbital/amobarbital’.
The intra- and inter-day precision of the analyte responses were evaluated at concentrations that were the same as the
cut-off concentration and 10 times the cut-off concentration. The CVs of the intra- and inter-day precision of the analyte re-sponse for low concentration-spiked samples were ,14.60 and 14.84%, respectively. The CVs of the intra- and inter-day precision were ,4.52 and 17.39% for the high concentration-spiked samples. The CVs of the intra- and inter-day precision of the reten-tion times were ,0.38 and 0.51%, respectively. Because there is no extraction step in the proposed experimental procedure, only the matrix effect was evaluated. The matrix effect (n ¼ 6) was cal-culated under the optimal conditions, and the values are summar-ized in Table I. All analytes showed matrix effects of 77–121%, except barbital (only 30–40%). The CVs of the matrix effects between six individual samples for low and high spiked concen-trations were ,21.71 and 20.15%, respectively.
The LODs of the 62 analytes are listed in TableI. Among the 62 analytes, 57 analytes provided good sensitivity with LODs ,40 ng/mL, which was much lower than their cut-off concen-trations. Meprobamate, glutethimide, temazepam, clobazem and barbital were analytes with LODs .70 ng/mL (77.3 – 187.5 ng/ mL), but their LODs were still lower than their respective cut-off concentrations.
For carryover investigation, only analyte MDEA showed 0.05% carryover (the ratio of peak height from the blank water to that from the spiked sample) in the first water blank injection. Therefore, one injection of blank water was used between each sample analysis.
Construction of a fragmentation library for 62 abused drugs and metabolites
When analyzing abused drugs in a biological matrix, false posi-tives might occur when using TOF-MS for drug screening due to other compounds that have identical empirical formulas. To provide accurate analytical results, it is essential to perform a confirmation step after the initial screening. Owing to the diver-sity of the structures of the 62 analytes, the collision energy was optimized for each drug to generate the most informative frag-mentation patterns. For each analyte, low, medium and high col-lision energies were tested, and the optimal colcol-lision energies and dominant fragment ions are summarized in TableII.
Automatic screening and confirmation of abused drugs Automatic screening of 62 analytes was performed using the Mass Hunter Software – Qualitative Analysis Software provided by Agilent. The compound name, molecular formula and retention times of the 62 analytes were inputted into the software. Screening match tolerances were retention time tolerance, +0.2 min; masses tolerance, +20 ppm and peak height, 1500 counts. The output of the screening results includes the mass difference between the measured mass and the theoretical mass in ppm, peak area, peak height and the overall score (Figure2). The overall score is the weighted average of different matching factors, including retention time (weighting, 100), mass match (weighting, 100), isotope abun-dance (weighting, 60) and isotope spacing (weighting, 50). Higher scores were correlated with a higher possibility that the specific peak was a true positive. We used high and low concentration-spiked samples to determine the suitable overall score, and the results showed that all the spiked analytes could be detected with an overall score .60. If the threshold of the screening score was
set too high, false-negative rates increased. Setting a low threshold value was also unnecessary, because it increased the burden of the confirmation step. Therefore, only compounds with an overall score .60 were introduced to MS–MS confirmation.
After the screening step, the result was confirmed by compar-ing the pattern of product ions with the in-house fragmentation Table II
MS – MS information for 62 abused drugs and metabolites at a specific collision energy No. ID Collision energy (V) Parent ion Product ion 1 Product ion 2 Product ion 3 Basic compounds 1 Morphinea 30 286.1438 201.0906 58.0649 165.0699 2 Norephedrine 20 152.1070 134.0961 117.0697 91.0542 3 Ephedrine 20 166.1226 148.1123 117.0699 56.0499 4 Aminorex 30 163.0866 103.0544 77.0386 120.0804 5 Pseudoephedrine 20 166.1226 148.1119 117.0696 56.0493 6 Nalorphinea 30 312.1594 201.0901 70.0659 270.1131 7 Methylephedrine 20 180.1383 162.1280 117.0698 147.1043 8 Dihydrocodeinea 30 302.1751 199.0767 227.1079 245.1172 9 Codeinea 30 300.1593 58.0662 215.1081 183.0814 10 Amphetamine 10 136.1121 119.0866 91.0557 – 11 Methamphetamine 10 150.1277 119.0855 91.0554 – 12 MDA 20 180.1019 105.0695 135.0437 163.0749 13 MDMA 20 194.1176 105.0698 135.0440 163.0754 14 PMA 20 166.1226 121.0512 149.0822 91.0414 15 PMMA 20 180.1383 121.0510 149.0814 91.0413 16 MDEA 20 208.1332 105.0703 135.0444 163.0758 17 Phentermine 10 150.1277 91.0541 133.1007 105.0694 18 Norketamine 20 224.0837 125.0149 179.0616 207.0569 19 Ketamine 20 238.0993 125.0156 179.0626 220.0894 20 Tramadol 20 264.1958 58.0650 – – 21 Heroin 30 370.1649 58.0653 268.1336 211.0757 22 Cocaine 30 304.1543 182.1187 82.0657 105.0341 23 Methylphenidate 30 234.1489 84.0815 56.0503 – 24 Meperidine 30 248.1645 70.0655 220.1341 174.1283 25 2C-B 30 260.0281 227.9791 212.9541 134.0732 26 Zolpidem 30 308.1757 235.1232 263.1181 – 27 7-Aminoflunitrazepam 30 284.1194 135.0922 227.0985 256.1250 28 LSD 20 324.2070 223.1243 281.1662 208.0772 29 Butorphanol 30 328.2276 310.2179 124.1120 – 30 Pentazocine 30 286.2165 69.0713 218.1550 175.1128 31 PCP 20 244.2060 86.0964 91.0545 159.1167 32 Meprobamate 10 219.1339 158.1171 97.1012 55.0545 33 Fentanyl 30 337.2274 188.1435 105.0697 – 34 Flurazepam 30 388.1586 315.0704 100.1124 288.0594 35 Midazolam 30 326.0855 291.1171 244.0321 209.0632 36 Buprenorphine 50 468.3108 55.0548 396.2173 187.0746 37 Bromazepam 30 316.0080 182.0839 209.0949 80.0496 38 Glutethimide 10 218.1176 98.9753 157.0168 190.1226 39 Chlordiazepoxide 20 300.0898 282.0798 227.0499 57.0451 40 Nitrazepam 30 282.0873 236.0955 180.0810 207.0919 41 Clonazepam 30 316.1483 270.0554 214.0425 – 42 Methadone 20 310.2165 265.1586 105.0338 57.0337 43 Flunitrazepam 30 314.0935 268.1010 239.0974 211.0786 44 Estazolam 30 295.0745 267.0565 205.0763 138.0100 45 Clobazem 30 301.0738 259.0645 98.9756 224.0943 46 Oxazepam 20 287.0582 241.0530 269.0477 104.0491 47 Triazolam 30 343.0512 308.0826 315.0332 239.0397 48 Alprazolam 30 309.0902 281.0738 205.0782 165.0222 49 Lorazepam 20 321.0192 275.0153 303.0100 229.0537 50 Temazepam 30 301.0738 255.0706 193.0901 228.0588 51 Lormetazepam 10 335.0363 289.0304 317.0254 – 52 Nordiazepam 30 271.0633 140.0265 208.1001 91.0546 53 Diazepam 30 285.0789 193.0894 154.0423 222.1159 54 Prazepam 30 325.1102 271.0649 140.0267 208.1002 Acidic compounds 55 Barbital 20 183.0775 140.0692 136.9360 96.0807 56 Phenobarbital 20 231.0775 59.0146 144.0801 85.0055 57 Butabarbital 30 211.1088 136.9486 196.9086 59.0119 58 Butalbital 10 223.1088 180.1035 141.0166 59.0140 59 Pentobarbital 20 225.1240 182.1171 138.1280 59.0137 60 Amobarbital 20 225.1245 182.1185 138.1286 85.0039 61 Secobarbital 30 237.1245 158.9541 114.9664 59.0119 62 THCCOOH 30 343.1915 59.0156 300.2087 246.1626 a
library. Samples with the anticipated results were reanalyzed by UHPLC – QTOF-MS in the MS tandem mode using optimal colli-sion energy to generate product ion mass spectra. By comparing
the diagnostic ions of compounds in the urine sample with the in-house library data (Table II), we were able to confirm the result as either a true or false positive.
Figure 2. The screening result and product ion spectra of candidate peaks and target standards for the urine sample from Case 17. (A) Automatic screening result, (B) true positive result of amphetamine and (C) false-positive result of temazepam.
We applied the developed method to 10 urine samples col-lected from patients undergoing MMT. The results of the UHPLC – QTOF-MS method were compared with that from the immunoassay test, which are reported in Table III. Methadone was detected and confirmed in all the 10 samples, because it was the prescribed drug for the 10 patients. Morphine was detected and confirmed from eight samples by the developed method, which was also detected by immunoassay as the opioid drug. Codeine is also an opioid drug that was detected and confirmed by the UHPLC – QTOF-MS method in the samples from Cases 6 and 10. Four of the 10 specimens showed different analytical results between the developed method and the immunoassay. Amphetamine was only detected and confirmed by the UHPLC – QTOF-MS method in Cases 11 and 17. However, it was not detected by the immunoassay. In Case 3, KET was detected by the UHPLC – QTOF-MS method and the immunoassay, but norke-tamine, the metabolite of KET, was only identified by the UHPLC – QTOF-MS method. The detection of drug metabolites could support the consumption of KET. In addition, 7-aminoflunitrazepam, which is the metabolite of flunitrazepam, was only detected by the UHPLC – QTOF-MS method in Case 18. In contrast, the immunoassay did not detect any BZD in this spe-cimen. Although the immunoassay method already includes screening item of BZDs, drug consumption might not be
detected if most of the parent form is converted to its metabo-lites. Compared with the immunoassay method, the UHPLC – QTOF-MS is advantageous in that a comprehensive screening of different classes of drugs could be completed on a single plat-form with improved selectivity, whereas different immunoassay reagents need to be used to increase the coverage in immuno-assay tests. This finding also demonstrated that UHPLC – QTOF-MS has the potential to replace immunoassays, which has been recently proposed also by Saleh et al. (10).
One example, Case 17, showed the importance of the con-firmation step (Figure2). In the first step, morphine, amphet-amine, methamphetamphet-amine, methadone and temazepam were detected as candidate analytes (score .60). However, with product ion spectrum confirmation, the temazepam signal was determined to be a false positive (Figure2C). With the integra-tion of screening and confirmaintegra-tion in the same UHPLC – QTOF-MS platform, 62 abused drugs and metabolites could be accurately determined with high confidence.
Conclusions
In this study, an UHPLC method coupled with QTOF-MS for the screening and confirmation of 62 abused drugs and metabolites in urine was developed. A 5-fold dilution with water was applied to simplify the sample pretreatment. Positive and negative ioniza-tion modes were used to provide better sensitivity for basic and acidic analytes. The validated results showed good selectivity, pre-cision and detection sensitivity, and the matrix effects were between 77 and 121% for 98% of the analytes. Fragmentation pat-terns and diagnostic ions were generated for 62 abused drugs and metabolites in this study, and the screening results could be con-firmed by QTOF-MS with MS– MS detection on the same platform to improve the test accuracy. With the integration of the screen-ing and confirmation steps in the same platform, better accuracy can be achieved by UHPLC– QTOF-MS.
Acknowledgments
The authors thank the NTU Integrated Core Facility for Functional Genomics of the National Research Program for the Genomic Medicine of Taiwan for technical assistance.
Funding
This study was sponsored by the Food and Drug Administration, Department of Health, Executive Yuan (DOH99-FDA-71017 and DOH99-FDA-71020).
References
1. Substance Abuse and Mental Health Services Administration (2012) Results from the 2011 National Survey on Drug Use and Health: Summary of National Findings. http://www.samhsa.gov/data/nsduh/ 2k11results/nsduhresults2011.pdf (28 March 2013, date last accessed). 2. European Monitoring Centre for Drugs and Drug Addiction (2009)
General Population Surveys of Drug Use-3: Last Year Prevalence of Drug Use among all Adults (Aged 15 to 64 Years) in Nationwide Surveys among the General Population. http://www. emcdda.europa.eu/stats09/gpstab3 (28 March 2013, date last accessed).
Table III
Screening results of 10 urine samples by UHPLC – QTOF-MS and immunoassay methods Case number QTOF-MS screen result QTOF-MS MS/MS confirm Immunoassaya
Case 1 7-Aminoflunitrazepam þ BZD
Methadone þ MTD
Lorazepam 2
Case 2 Morphine þ OPI
Chlordiazepoxide 2 2 Methadone þ MTD Case 3b Morphine þ OPI Norketamine þ NA Ketamine þ Ketamine Methadone þ MTD
Case 4 Morphine þ OPI
Codeine 2
Methadone þ MTD
Case 6 Morphine þ OPI
Codeine þ OPI
Methadone þ MTD
Temazepam 2 2
Case 9 Morphine þ OPI
Methadone þ MTD
Case 10 Morphine þ OPI
Codeine þ OPI
Methadone þ MTD
Case 11 Morphine þ OPI
Amphetamine þ 2
Methamphetamine þ mAMP
Methadone þ MTD
Case 17 Morphine þ OPI
Amphetamine þ 2 Methamphetamine þ mAMP Methadone þ MTD Temazepam 2 2 Case 18 7-Aminoflunitrazepam þ 2 Methadone þ MTD
NA: not available.
a
In immunoassay, the targets included acetaminophen (APAP), amphetamines (AMPs), methamphetamines (mAMPs), barbiturates (BARs), benzodiazepines (BZDs), cocaine (COC), methadone (MTD), opiates (OPIs), phencyclidine (PCP), tetrahydrocannabinol (THC), tricyclic antidepressants (TCAs) and ketamine (KET).
b
The case numbers indicated with bold formatting are samples that showed different findings with the use of UHPLC – QTOF-MS and immunoassay methods.
3. Gonzalez-Marino, I., Quintana, J.B., Rodriguez, I., Gonzalez-Diez, M., Cela, R. (2012) Screening and selective quantification of illicit drugs in wastewater by mixed-mode solid-phase extraction and quadrupole-time-of-flight liquid chromatography-mass spectrometry. Analytical
Chemistry,84, 1708–1717.
4. Pesce, A., Rosenthal, M., West, R., West, C., Crews, B., Mikel, C. et al. (2010) An evaluation of the diagnostic accuracy of liquid chromatography-tandem mass spectrometry versus immunoassay drug testing in pain patients. Pain Physician,13, 273– 281.
5. Weinmann, W., Renz, M., Vogt, S., Pollak, S. (2000) Automated solid-phase extraction and two-step derivatisation for simultaneous ana-lysis of basic illicit drugs in serum by GC/MS. International Journal of Legal Medicine,113, 229 – 235.
6. Marchei, E., Colone, P., Nastasi, G.G., Calabro, C., Pellegrini, M., Pacifici, R. et al. (2008) On-site screening and GC-MS analysis of cocaine and heroin metabolites in body-packers urine. Journal of
Pharmaceutical and Biomedical Analysis,48, 383–387.
7. Saussereau, E., Lacroix, C., Gaulier, J.M., Goulle, J.P. (2012) On-line liquid chromatography/tandem mass spectrometry simultaneous de-termination of opiates, cocainics and amphetamines in dried blood spots. Journal of Chromatography B-Analytical Technologies in the Biomedical and Life Sciences,885, 1 – 7.
8. Bueno, M.J.M., Ucles, S., Hernando, M.D., Fernandez-Alba, A.R. (2011) Development of a solvent-free method for the simultaneous identifi-cation/quantification of drugs of abuse and their metabolites in
en-vironmental water by LC-MS/MS. Talanta,85, 157 – 166.
9. Maurer, H.H. (2012) What is the future of (ultra) high performance liquid chromatography coupled to low and high resolution mass
spectrometry for toxicological drug screening? Journal of
Chromatography A, doi:10.1016/j.chroma.2012.08.069.
10. Saleh, A., Stephanson, N.N., Granelli, I., Villen, T., Beck, O. (2012) Evaluation of a direct high-capacity target screening approach for urine drug testing using liquid chromatography-time-of-flight mass spectrometry. Journal of Chromatography B-Analytical Technologies in the Biomedical and Life Sciences,909, 6–13.
11. Lee, H.K., Ho, C.S., Iu, Y.P.H., Lai, P.S.J., Shek, C.C., Lo, Y.C. et al. (2009) Development of a broad toxicological screening technique for urine using ultra-performance liquid chromatography and time-of-flight
mass spectrometry. Analytica Chimica Acta,649, 80–90.
12. Kolmonen, M., Leinonen, A., Kuuranne, T., Pelander, A., Ojanpera, I.
(2009) Generic sample preparation and dual polarity liquid
chromatography- time-of-flight mass spectrometry for high-throughput screening in doping analysis. Drug Testing and Analysis,1, 250–266. 13. Dominguez-Romero, J.C., Garcia-Reyes, J.F., Molina-Diaz, A. (2011)
Screening and quantitation of multiclass drugs of abuse and pharmaceu-ticals in hair by fast liquid chromatography electrospray time-of-flight
mass spectrometry. Journal of Chromatography B-Analytical
Technologies in the Biomedical and Life Sciences,879, 2034–2042. 14. Broecker, S., Herre, S., Wust, B., Zweigenbaum, J., Pragst, F. (2011)
Development and practical application of a library of CID accur-ate mass spectra of more than 2,500 toxic compounds for
systematic toxicological analysis by LC-QTOF-MS with data-dependent acquisition. Analytical and Bioanalytical Chemistry, 400, 101– 117.
15. Ojanpera, I., Kolmonen, M., Pelander, A. (2012) Current use of high-resolution mass spectrometry in drug screening relevant to clinical and forensic toxicology and doping control. Analytical and
Bioanalytical Chemistry,403, 1203 – 1220.
16. Kolmonen, M., Leinonen, A., Pelander, A., Ojanpera, I. (2007) A general screening method for doping agents in human urine by solid phase extraction and liquid chromatography/time-of-flight mass
spectrometry. Analytica Chimica Acta,585, 94 – 102.
17. Pelander, A., Ojanpera, I., Laks, S., Rasanen, I., Vuori, E. (2003) Toxicological screening with formula-based metabolite identifica-tion by liquid chromatography/time-of-flight mass spectrometry. Analytical Chemistry,75, 5710 – 5718.
18. Gergov, M., Boucher, B., Ojanpera, I., Vuori, E. (2001) Toxicological screening of urine for drugs by liquid chromatography/time-of-flight mass spectrometry with automated target library search based on elemental formulas. Rapid Communications in Mass Spectrometry, 15, 521 – 526.
19. Birkler, R.I.D., Telving, R., Ingemann-Hansen, O., Charles, A.V., Johannsen, M., Andreasen, M.F. (2012) Screening analysis for medi-cinal drugs and drugs of abuse in whole blood using ultra-performance liquid chromatography time-of-flight mass spectrom-etry (UPLC-TOF-MS)-toxicological findings in cases of alleged sexual assault. Forensic Science International,222, 154 – 161.
20. Marin, S.J., Hughes, J.M., Lawlor, B.G., Clark, C.J., McMillin, G.A. (2012) Rapid screening for 67 drugs and metabolites in serum or plasma by accurate-mass LC-TOF-MS. Journal of Analytical Toxicology,36, 477 – 486.
21. Crews, B.O., Pesce, A.J., West, R., Nguyen, H., Fitzgerald, R.L. (2012) Evaluation of high-resolution mass spectrometry for urine toxicol-ogy screening in a pain management setting. Journal of Analytical Toxicology,36, 601 – 607.
22. Guale, F., Shahreza, S., Walterscheid, J.P., Chen, H.H., Arndt, C., Kelly, A.T. et al. (2013) Validation of LCTOF-MS screening for drugs, meta-bolites, and collateral compounds in forensic toxicology specimens. Journal of Analytical Toxicology,37, 17 – 24.
23. Moriya, F. (2009) Urine levels of drugs for which Triage DOA screen-ing was positive. Legal Medicine,11, S434 – S436.
24. Enno, F., Joseph, V.L. (2009) Pharmacology and abuse cocaine, amphetamines, ecstasy and related designer drugs. Dordrecht/ Heidelberg/London/New York: Springer, pp. 253 – 260.
25. Department of Health, Executive Yuan (2008) Regulations Governing Drug Abuse Urine Testing Operations. Taiwan, ROC: Department of Health, Executive Yuan. http://dohlaw.doh.gov.tw/Chi/EngContent. asp?Msgid=114&KeyWord= (28 March 2013, date last accessed). 26. Rivier, L. (2003) Criteria for the identification of compounds by liquid
mass spectrometry and liquid chromatography-multiple mass spectrometry in forensic toxicology and doping analysis.
