- Email: [email protected]
Evaluation of two immunoassay procedures for drug testing in hair samples
Forensic Science International 215 (2012) 60–63
Contents lists available at ScienceDirect
Forensic Science International journal homepage: www.elsevier.com/locate/forsciint
Evaluation of two immunoassay procedures for drug testing in hair samples§ F. Musshoff a,*, K.M. Kirschbaum a, K. Graumann a, C. Herzfeld a, H. Sachs b, B. Madea a a b
Institute of Forensic Medicine, University Bonn, Stiftsplatz 12, 53111 Bonn, Germany FTC GmbH Mu¨nchen, Bayerstr. 53, 80335 Munich, Germany
A R T I C L E I N F O
A B S T R A C T
Article history: Received 29 September 2010 Received in revised form 9 March 2011 Accepted 24 March 2011 Available online 23 May 2011
A preliminary initial enzyme-linked immunosorbent assay (LUCIO1-Direct ELISA kit) and a preliminary DRI1 enzyme immunoassay were evaluated for drug detection in head hair with respect to lowered cutoff values recommended in Germany for the control of abstinence in cases of re-granting of drivers’ licences. Following drug classes were included: cannabinoids, opiates, cocaine like substances, amphetamine, methamphetamine (and methylenedioxyamphetamines), methadone, and benzodiazepines. 759 analyses were performed using LUCIO1-Direct ELISA kits and 936 analyses using DRI1 enzyme immunoassay tests. Sample size for each drug group and immunoassay test reached from 74 to 178. The LUCIO1-Direct ELISA kit revealed a sensitivity of 91% for amphetamine up to 98% for methadone (methamphetamine 92%, cocaine 94%, opiates 94%, benzodiazepines 96%) and values of specificity of 72% for methadone up to 89% for amphetamine and benzodiazepines. The test was not useful for a preliminary screening for tetrahydrocannabinol (sensitivity of 65%) in consideration of a suggested cutoff of 0.02 ng/mg. The DRI1 enzyme immunoassay test was only useful for morphine and cocaine testing at low recommended new cutoff values (0.1 ng/mg) revealing sensitivities of 94% and 99%, respectively. ß 2011 Elsevier Ireland Ltd. All rights reserved.
Key words: Hair analysis Drugs Immunoassay Chromatography Driving under influence (DUI) Re-granting of drivers’ licences
1. Introduction Analytical methods are getting more sensitive and enable detection of drugs at low concentrations. Cutoff values for the control of abstinence in cases of drivers’ licence re-granting were lowered and harmonized in Germany in 2009 [1]. The new values recommended for hair analyses for such purposes in Germany are quiet lower compared to other recommendations from the Society of Hair Testing (SoHT) [2], the Substance Abuse and Mental Health Services Administration (SAMHSA) [3], or for European workplace drug testing [4] (Table 1). When using immunochemical methods for a preliminary test according to the guidelines of the Society of Forensic and Toxicological Chemistry (GTFCh) the relevant analyte concentration (cutoff) obtained by chromatographic method must show positive immunochemical result for the relevant substance class which must be documented by appropriate validation procedure [5]. Sensitivity should be nearly 100% and the rate of false-positives simultaneously should be as low as possible due to cost and time.
§ ‘‘48th Annual Meeting of the International Association of Forensic Toxicologists (TIAFT). Joint meeting with the society of Toxicological and Forensic Chemistry (GTFCh)’’. August 29th to September 2, 2010, Bonn, Germany, Guest edited by Thomas Kraemer, Hans. H. Maurer and Frank Musshoff. * Corresponding author. Tel.: +49 228 738316; fax: +49 228 738368. E-mail address: [email protected] (F. Musshoff).
0379-0738/$ – see front matter ß 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.forsciint.2011.03.030
In the present study a preliminary enzyme-linked immunosorbent assay (LUCIO1-Direct ELISA kit) and a preliminary DRI1 enzyme immunoassay were evaluated for drug detection in hair with respect to new cutoff values in Germany. 2. Methods and samples 2.1. Immunoassays For drug testing in hair Thermo Fisher (Passau, Germany) suggested the DRI1 reagents for amphetamine, ecstasy, THC, opiates, cocaine, and methadone, respectively. For sample preparation and calibration of an autoanalyzer special reagents were allocated. Reagents for the microtiter plate LUCIO1-Direct ELISA kit for the analysis of amphetamine, methamphetamine/ecstasy, delta-9-terahydrocannabinol (THC), opiates, cocaine, benzodiazepines, and methadone were provided by NAL von Minden (Regensburg, Germany) who distributes the reagents of Immunanalysis Corporation (Pomona, CA, USA) in Germany. 2.1.1. DRI1 enzyme immunoassay Hair samples were cut with scissors and about 33 mg (1 mg) were transferred to a glass test tube provided with a hermetic seal cap, then 1000 mL of SLV-VMA-T washing solution was added. The mixture was shaken for about 20–30 s and the SLVVMA-T solution was discarded. To eliminate the SLV-VMA-T solution the pipette tip was placed directly on the pellet of the keratinic matrix compressed onto the test tube bottom and at least 940 mL of the solution was recovered. 400 mL of VMA-T reactive reagent was added to the test tube containing the sample under examination. After mixing in order to ascertain that hair is completely soaked in the reagent it was digested for 1 h at 100 8C and mixed every 20–30 min. Then the sample was cooled at room temperature and the supernatant was placed into a cuvette, centrifuged and left for about 15 min before carrying out the immunoassay tests. Extracts were analyzed by immunoassay tests using methods specifically calibrated for keratinic matrix with
F. Musshoff et al. / Forensic Science International 215 (2012) 60–63
61
Table 1 Recommended cutoff values for drug testing in hair. Drug/Drug class
Society of Hair Testing (SoHT)
Substance Abuse and Mental Health Services Administration (SAMHSA)
European Workplace Drug Testing
German Guid elines for driving licence regranting ‘
THC Opiates Cocaine Amphetamines Methadone Benzodiazepines Ethylglucuronide
0.1 0.2 0.5 0.2
THC metabolite 0.00005a 0.2 1 0.3
0.05 0.2 0.5 0.2
0.02 0.1 0.1 0.1 0.1 0.05 0.007
0.05 0.007 (0.03*)
0.03b
a
Delta-9-tetrahydrocannabinol-9-carboxylic acid. The cutoff for ethylglucuronide in hair to strongly suggest chronic excessive alcohol consumption is proposed at 30 pg/mg scalp hair measured in the 0–3 cm proximal segment. A concentration equal or greater than 7 pg/mg indicates alcohol consumption and provides evidence to refute a claim of abstinence. b
CAL-VMA-T calibrators (see Table 2 for composition and concentration range) on a CDx90 autoanalyzer. The assays were run according to manufacturer’s instructions. For all procedures reactions were read at 340 nm.
recommended cutoff values for hair analysis required for control of abstinence in cases of re-granting of drivers’ licences (Table 4). 2.3. Samples
2.1.2. LUCIO1-Direct ELISA kit The ELISA technology was based upon the competitive binding to antibody of enzyme labelled antigen and unlabelled antigen in proportion to their concentration in the reaction well. Hair samples were cut with scissors into small pieces and an aliquot was weighed (10 mg). Hair was washed with methanol (2 mL, 10 min) and allowed to dry. HEB hair extraction buffer (pH 4.5, 0.5 mL) was added and the tubes were capped and incubated for 2 h in an ultrasound bath at 75 8C. The total extract was removed and after cooling neutralization buffer was added (50 mL). An aliquot of 25 mL was used for amphetamine, methamphetamine, opiate, cocaine, methadone, and benzodiazepine ELISA and an aliquot of 100 mL was used for THC test. The assays were run according to manufacturer’s instructions. The microplate reader employed was an ELx 800 at wavelength of 450 nm (BioTek Instruments Inc., Winooski, USA) (Table 3). 2.2. Chromatographic analyses Drugs were quantified using gas chromatography in combination with mass spectrometry (GC/MS) or high performance liquid chromatographic methods coupled to mass spectrometry (LC/MS). Sample preparation for GC/MS comprehended a methanolic ultrasonication followed by solid-phase-extraction (cocain like compounds and opiates) or liquid-liquid extraction (THC and amphetamine/methamphetamine/methylenedioxyamphetamines) for clean-up and derivatization of the analytes before injection into the chromatographic systems. For LC/MS analysis the extract of methanolic ultrasonication was reconstituted in mobile phase and directly injected. All methods are standard procedures used in the routine laboratories in Bonn or Munich, fully validated according to guidelines of GTFCh [5] and accredited according to DIN EN ISO 17025. More details according chromatographic procedures were previously described [6–8]. The limits of detection (LODs) were below the
Table 2 Concentrations of calibrators (CAL-VMA-T) for the DRI1 tests (and CEDIA cannabinoid-OFT) according to the manufacturer’s recommendations [ng/mg].
Opiates Cocaine Methadone Amphetamine Ecstasy Cannabinoids
Level 1
Level 2
Level 3
Level 4
Level 5
0.00 0.00 0.00 0.00 0.00 0.00
0.48 0.50 1.00 0.80 0.50 0.13
0.96 1.00 2.00 1.60 1.00 0.26
1.92 2.00 4.00 3.20 2.00 0.52
3.84 4.00 8.00 6.40 4.00 1.04
Table 3 Concentrations of calibrators [ng/mg] for the Direct ELISA kit according to the manufacturer’s recommendations.
Opiates Cocaine Methadone Amphetamine Ecstasy Cannabinoids
Level 1
Level 2
Level 3
Level 4
Level 5
0.00 0.00 0.00 0.00 0.00 0.00
0.05 0.05 0.05 0.05 0.05 0.01
0.10 0.10 0.10 0.10 0.10 0.02
0.15 0.15 0.15 0.15 0.15 0.03
0.20 0.20 0.20 0.20 0.20 0.04
Authentic head hair samples were collected from cases of re-granting of drivers’ licences and other criminal cases, hair samples from other sites of the body were not enrolled. Only a few coloured hair samples were included, but no bleached samples. It was ensured that positive samples were preferentially enrolled in the present study with concentrations about recommended new lowered cutoff values but less with high drug concentrations. Few samples were diluted with blank hair to obtain concentration close to required limits. 2.4. Statistics Immunoassay cutoff values were calculated using contingency tables and receiver operating characteristic (ROC) curves [9] considering Youden-Indices (sensitivity + specificity 1) [10] with respect to measured chromatographic concentrations as ‘‘true’’ results. Sensitivity (true positives/(true positives + false negatives)), specificity (true negatives/(true negatives + false positives)), positive predictive value (PPV) (true positives/(false positives + true positives)), negative predictive value (NPV) (true negative/(true negative + false negative)) and the overall misclassification rate (OMR) ((false positive + false negatives)/all) were calculated by personal computer and use of Microsoft Office Excel 2003. Because of the fact that sensitivity and specificity are probabilities the standard error was calculated according to the following formula: standard error = (sensitivity or specificity (1 sensitivity or specificity)/sample size)1/2. Receiver operating characteristic (ROC) curves (sensitivity vs. (1 specificity)) were generated using
Table 4 Limits of quantification (LOQs) of the routine LC/MS and GC/MS procedures. Analyte
GC/MS
LC/MS
Tetrahydrocannabinol Anhydroecgoninmethylester Cocaine Norcocain Cocaethylene Benzoylecgonin Ecgoninmethylester Diacetylmorphin (heroin) 6-Monoacetylmorphine Morphine Codeine Dihydrocodeine Methadone EDDP Amphetamine Methamphetamine MDMA MDA MDE Diazepam Nordazepam Oxazepam Flunitrazepam 7-Aminoflunitrazepam Bromazepam Alprazolam Lorazepam
0.02
0.013 0.084 0.016 0.02 0.018 0.015 0.035 0.046 0.027 0.037 0.032 0.027 0.034 0.037 0.023 0.011 0.013 0.018 0.02 0.037 0.039 0.034 0.019 0.019 0.029 0.011 0.018
0.05
0.05
0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05
F. Musshoff et al. / Forensic Science International 215 (2012) 60–63
62
SPSS version 17.0 for Windows (SPSS Inc., Chicago, USA) to calculate cutoff values for each drug class.
3. Results and discussion 759 analyses were performed using LUCIO1-Direct ELISA kits and 936 analyses using DRI1 enzyme immunoassay tests. Sample size for each drug group and immunoassay test reached from 74 to 178. Immunoassay data were analyzed by the use of contingency tables that allowed calculation of sensitivity, specificity, positive and negative predictive value and the overall misclassification rate. ROC curves were used for all analytes to evaluate discrimination power of the tests and to confirm cutoff values (Figs. 1 and 2 for example). The Youden-Index was an additional tool to test quality of immunoassay tests in this study. Data for each test of correct negative, correct positive, false negative and false positive samples are given in Tables 5 and 6. The recommended cutoff value of 0.02 ng/mg for the test of THC revealed a high rate of false-negative samples in both tests. When increasing the cutoff value to 0.1 ng/mg as recommended by SoHT sensitivity of 92% and specificity of 87% were achieved with LUCIO1-Direct ELISA test. A sensitivity of at least 90% is recommended by the GTFCh [5]. However, a sensitivity of only 92% means to allow 8 individuals every 100 to seem negative even if using scheduled substances. Analyses of other drug classes tested with LUCIO1-Direct ELISA kits showed results with values of sensitivity of 91% for amphetamine up to 98% for methadone and values of specificity of 72% up to 89% (Table 7). The cutoff value of 0.1 ng/mg for the test of amphetamine, ecstasy, and methadone revealed high rates of false-negative and false-positive samples using DRI1 enzyme immunoassay tests resulting in values for sensitivity of 71–87% and values for specificity of 50–74% (Table 8). Only analyses of opiates and cocaine using the aforementioned test showed perhaps satisfying results with values of sensitivity of 94% and 99% and values of specificity of 73% and 91%. Considering the median values in Tables 7 and 8 it was ensured that positive samples were enrolled preferentially in the present study with concentrations about the recommended new lowered cutoff values but less with high drug concentrations. In contrast to urine or blood drug testing immunoassays for hair analysis are not common in daily routine due to low concentrations required for this purpose. Immunoassay evaluation for hair analysis is described for single substance classes [8,11–16] or as preliminary screening procedures considering higher cutoff values [17]. Also both procedures evaluated in the present study were recently tested by others and revealed satisfying results using calibration about recommended cutoff values of SoHT [18–20].
Fig. 2. ROC curve of amphetamine test using the DRI1 immunoassay test.
Fig. 1. ROC curve of amphetamine test using the Direct ELISA kit.
It has to be taken into consideration that sensitivity for preliminary screenings is required to approach as much as possible 100%. Acceptable results were not obtained with both assays with respect to new cutoff values recommended in Germany. A direct chromatographic measurement seems to be necessary.
Table 5 Summary of contingency tables for the Direct ELISA kit. Chromatographic analysis
+
Amphetamine 66 8
Methamphetamine + 4 41
Cannabinoids
+ 3 35
67 15
63 5
Opiates + 22 40
42 12
Cocaine + 5 78
Methadone + 2 33
29 10
+ 1 41
54 21
Benzodiazepines 78 10
+ 1 22
Table 6 Summary of contingency tables for the DRI1 immunoassay test. Chromatographic analysis
Amphetamine
+
52 31
+ 14 56
Methamphetamine 49 49
+ 6 41
Cannabinoids
46 20
+ 31 76
Opiates
41 4
Cocaine + 1 74
66 25
Methadone + 5 82
69 24
+ 21 53
F. Musshoff et al. / Forensic Science International 215 (2012) 60–63
63
Table 7 Summary of results for the LUCIO1-Direct ELISA kit.
Concentration range Median Chromat. cutoff value Immunoassay cutoff value Sensitivity (%) Specificity (%) PPV NPV OMR Youden
Amphetamine
Methamphetamine
THC
THC
Cocaine
Opiates
Methadone
Benzodiazepines
0.05–5.00 0.12 0.1 0.076 91 89 84 94 10 82
0.01–4.00 0.10 0.1 0.037 92 82 70 96 15 74
0.006–0.80 0.06 0.02 0.008 65 93 89 74 21 57
0.006–0.80 0.06 0.1 0.008 92 87 73 96 12 79
0.01–114.5 0.20 0.1 0.083 94 78 87 89 12 72
0.02–0.83 0.17 0.1 0.004 94 74 77 94 16 69
0.02–2.60 0.12 0.1 0.088 98 72 66 98 19 70
0.01–0.98 0.04 0.05 0.079 96 89 69 99 10 84
Table 8 Summary of results for the DRI1 test.
Concentration range Median Chromat. cutoff value Immunoassay cutoff value Sensitivity (%) Specificity (%) PPV NPV OMR Youden
Amphetamine
Methamphetamine
THC
THC
Cocaine
Opiates
Methadone
0.05–5.0 0.30 0.1 0.22 80 63 64 79 29 29
0.05–5.0 0.20 0.1 0.23 87 50 46 89 38 38
0.01–1.63 0.10 0.02 0.04 71 70 79 60 29 29
0.01–1.63 0.10 0.1 0.04 87 65 63 88 26 26
0.05–11.9 1.33 0.1 0.08 99 91 95 98 4 4
0.05–4.0 0.19 0.1 0.24 94 73 77 93 17 17
0.20 0.1 0.18 72 74 69 77 27 27
Acknowledgements Reagents and technical equipment needed for the present evaluation were provided by Thermo Fisher and NAL van Minden, respectively. Thanks to Mrs. Werthebach and Dr. Dormeier (Thermo Fisher) as well as to Dr. Kosovac (NAL von Minden) for their technical support. References [1] W. Schubert, R. Mattern, Beurteilungskriterien: Urteilsbildung in der Medizinisch-Psychologischen Fahreignungsdiagnostik 2. Aufl, Kirschbaum Verlag, Bonn, 2009. [2] Society of Hair Testing, Recommendations for hair testing in forensic cases, Forensic Sci. Int. 145 (2004) 83–84. [3] Substance Abuse and Mental Health Services Administration (SAMHSA), Mandatory Guidelines for Federal Workplace Drug Testing Programs. Draft #4, 2001, http://www.workplace.samhsa.gov/ResourceCenter/Dt/FA/GuidelinesDraft4. html (accessed 12/09/2010). [4] R. Agius, P. Kintz, Guidelines for European workplace drug and alcohol testing in hair, Drug Test. Anal. 2 (2010) 367–376. [5] Society of Forensic and Toxicological Chemistry (GTFCh), Richtlinien der GTFCh zur Qualita¨tssicherung bei forensisch-toxikologischen Untersuchungen, 2009, https://www.gtfch.org/cms/files/GTFCh_Richtlinie_For-Tox_Version%201.pdf (accessed 12/09/2010). [6] D. Thieme, H. Sachs, Improved screening capabilities in forensic toxicology by application of liquid chromatography–tandem mass spectrometry, Anal. Chim. Acta 483 (2003) 171–179. [7] F. Musshoff, F. Driever, K. Lachenmeier, D.W. Lachenmeier, M. Banger, B. Madea, Results of hair analyses for drugs of abuse and comparison with self-reports and urine tests, Forensic Sci. Int. 156 (2006) 118–123. [8] K. Lachenmeier, F. Musshoff, B. Madea, Determination of opiates and cocaine in hair using automated enzyme immunoassay screening methodologies followed by gas chromatographic-mass spectrometric (GC–MS) confirmation, Forensic Sci. Int. 159 (2006) 189–199.
[9] M.H. Zweig, G. Campbell, Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical medicine, Clin. Chem. 39 (1993) 561–577. [10] R. Fluss, D. Faraggi, B. Reiser, Estimation of the Youden Index and its associated cutoff point, Biom. J. 47 (2005) 458–472. [11] C. Moore, D. Deitermann, D. Lewis, B. Feeley, R.S. Niedbala, The detection of cocaine in hair specimens using micro-plate enzyme immunoassay, J. Forensic Sci. 44 (1999) 609–612. [12] G. Cooper, L. Wilson, C. Reid, D. Baldwin, C. Hand, V. Spiehler, Validation of the Cozart microplate ELISA for detection of opiates in hair, J. Anal. Toxicol. 27 (2003) 581–586. [13] G. Cooper, L. Wilson, C. Reid, D. Baldwin, C. Hand, V. Spiehler, Comparison of Cozart microplate ELISA and GC–MS detection of methadone and metabolites in human hair, J. Anal. Toxicol. 29 (2005) 678–681. [14] E. Han, E. Miller, J. Lee, Y. Park, M. Lim, H. Chung, F.M. Wylie, J.S. Oliver, Validation of the immunoanalysis microplate ELISA for the detection of methamphetamine in hair, J. Anal. Toxicol. 30 (2006) 380–385. [15] H. Miyaguchi, H. Takahashi, T. Ohashi, K. Mawatari, Y.T. Iwata, H. Inoue, T. Kitamori, Rapid analysis of methamphetamine in hair by micropulverized extraction and microchip-based competitive ELISA, Forensic Sci. Int. 184 (2009) 1–5. [16] P. Lopez, S. Martello, A.M. Bermejo, V.E. De, M.J. Tabernero, M. Chiarotti, Validation of ELISA screening and LC–MS/MS confirmation methods for cocaine in hair after simple extraction, Anal. Bioanal. Chem. 397 (2010) 1539–1548. [17] M.L. Pujol, V. Cirimele, P.J. Tritsch, M. Villain, P. Kintz, Evaluation of the IDS onestep ELISA kits for the detection of illicit drugs in hair, Forensic Sci. Int. 170 (2007) 189–192. [18] R. de la Torre, E. Civit, F. Svaizer, A. Lotti, M. Gottardi, M. Miozzo, High throughput analysis of drugs of abuse in hair by combining purposely designed sample extraction compatible with immunometric methods used for drug testing in urine, Forensic Sci. Int. 196 (2010) 18–21. [19] C. Coulter, M. Taruc, J. Tuyay, C. Moore, Quantitation of tetrahydrocannabinol in hair using immunoassay and liquid chromatography with tandem mass spectrometric detection, Drug Test. Anal. 1 (2009) 234–239. [20] C. Coulter, J. Tuyay, M. Taruc, C. Moore, Semi-quantitative analysis of drugs of abuse, including tetrahydrocannabinol in hair using aqueous extraction and immunoassay, Forensic Sci. Int. 196 (2010) 70–73.
Contents lists available at ScienceDirect
Forensic Science International journal homepage: www.elsevier.com/locate/forsciint
Evaluation of two immunoassay procedures for drug testing in hair samples§ F. Musshoff a,*, K.M. Kirschbaum a, K. Graumann a, C. Herzfeld a, H. Sachs b, B. Madea a a b
Institute of Forensic Medicine, University Bonn, Stiftsplatz 12, 53111 Bonn, Germany FTC GmbH Mu¨nchen, Bayerstr. 53, 80335 Munich, Germany
A R T I C L E I N F O
A B S T R A C T
Article history: Received 29 September 2010 Received in revised form 9 March 2011 Accepted 24 March 2011 Available online 23 May 2011
A preliminary initial enzyme-linked immunosorbent assay (LUCIO1-Direct ELISA kit) and a preliminary DRI1 enzyme immunoassay were evaluated for drug detection in head hair with respect to lowered cutoff values recommended in Germany for the control of abstinence in cases of re-granting of drivers’ licences. Following drug classes were included: cannabinoids, opiates, cocaine like substances, amphetamine, methamphetamine (and methylenedioxyamphetamines), methadone, and benzodiazepines. 759 analyses were performed using LUCIO1-Direct ELISA kits and 936 analyses using DRI1 enzyme immunoassay tests. Sample size for each drug group and immunoassay test reached from 74 to 178. The LUCIO1-Direct ELISA kit revealed a sensitivity of 91% for amphetamine up to 98% for methadone (methamphetamine 92%, cocaine 94%, opiates 94%, benzodiazepines 96%) and values of specificity of 72% for methadone up to 89% for amphetamine and benzodiazepines. The test was not useful for a preliminary screening for tetrahydrocannabinol (sensitivity of 65%) in consideration of a suggested cutoff of 0.02 ng/mg. The DRI1 enzyme immunoassay test was only useful for morphine and cocaine testing at low recommended new cutoff values (0.1 ng/mg) revealing sensitivities of 94% and 99%, respectively. ß 2011 Elsevier Ireland Ltd. All rights reserved.
Key words: Hair analysis Drugs Immunoassay Chromatography Driving under influence (DUI) Re-granting of drivers’ licences
1. Introduction Analytical methods are getting more sensitive and enable detection of drugs at low concentrations. Cutoff values for the control of abstinence in cases of drivers’ licence re-granting were lowered and harmonized in Germany in 2009 [1]. The new values recommended for hair analyses for such purposes in Germany are quiet lower compared to other recommendations from the Society of Hair Testing (SoHT) [2], the Substance Abuse and Mental Health Services Administration (SAMHSA) [3], or for European workplace drug testing [4] (Table 1). When using immunochemical methods for a preliminary test according to the guidelines of the Society of Forensic and Toxicological Chemistry (GTFCh) the relevant analyte concentration (cutoff) obtained by chromatographic method must show positive immunochemical result for the relevant substance class which must be documented by appropriate validation procedure [5]. Sensitivity should be nearly 100% and the rate of false-positives simultaneously should be as low as possible due to cost and time.
§ ‘‘48th Annual Meeting of the International Association of Forensic Toxicologists (TIAFT). Joint meeting with the society of Toxicological and Forensic Chemistry (GTFCh)’’. August 29th to September 2, 2010, Bonn, Germany, Guest edited by Thomas Kraemer, Hans. H. Maurer and Frank Musshoff. * Corresponding author. Tel.: +49 228 738316; fax: +49 228 738368. E-mail address: [email protected] (F. Musshoff).
0379-0738/$ – see front matter ß 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.forsciint.2011.03.030
In the present study a preliminary enzyme-linked immunosorbent assay (LUCIO1-Direct ELISA kit) and a preliminary DRI1 enzyme immunoassay were evaluated for drug detection in hair with respect to new cutoff values in Germany. 2. Methods and samples 2.1. Immunoassays For drug testing in hair Thermo Fisher (Passau, Germany) suggested the DRI1 reagents for amphetamine, ecstasy, THC, opiates, cocaine, and methadone, respectively. For sample preparation and calibration of an autoanalyzer special reagents were allocated. Reagents for the microtiter plate LUCIO1-Direct ELISA kit for the analysis of amphetamine, methamphetamine/ecstasy, delta-9-terahydrocannabinol (THC), opiates, cocaine, benzodiazepines, and methadone were provided by NAL von Minden (Regensburg, Germany) who distributes the reagents of Immunanalysis Corporation (Pomona, CA, USA) in Germany. 2.1.1. DRI1 enzyme immunoassay Hair samples were cut with scissors and about 33 mg (1 mg) were transferred to a glass test tube provided with a hermetic seal cap, then 1000 mL of SLV-VMA-T washing solution was added. The mixture was shaken for about 20–30 s and the SLVVMA-T solution was discarded. To eliminate the SLV-VMA-T solution the pipette tip was placed directly on the pellet of the keratinic matrix compressed onto the test tube bottom and at least 940 mL of the solution was recovered. 400 mL of VMA-T reactive reagent was added to the test tube containing the sample under examination. After mixing in order to ascertain that hair is completely soaked in the reagent it was digested for 1 h at 100 8C and mixed every 20–30 min. Then the sample was cooled at room temperature and the supernatant was placed into a cuvette, centrifuged and left for about 15 min before carrying out the immunoassay tests. Extracts were analyzed by immunoassay tests using methods specifically calibrated for keratinic matrix with
F. Musshoff et al. / Forensic Science International 215 (2012) 60–63
61
Table 1 Recommended cutoff values for drug testing in hair. Drug/Drug class
Society of Hair Testing (SoHT)
Substance Abuse and Mental Health Services Administration (SAMHSA)
European Workplace Drug Testing
German Guid elines for driving licence regranting ‘
THC Opiates Cocaine Amphetamines Methadone Benzodiazepines Ethylglucuronide
0.1 0.2 0.5 0.2
THC metabolite 0.00005a 0.2 1 0.3
0.05 0.2 0.5 0.2
0.02 0.1 0.1 0.1 0.1 0.05 0.007
0.05 0.007 (0.03*)
0.03b
a
Delta-9-tetrahydrocannabinol-9-carboxylic acid. The cutoff for ethylglucuronide in hair to strongly suggest chronic excessive alcohol consumption is proposed at 30 pg/mg scalp hair measured in the 0–3 cm proximal segment. A concentration equal or greater than 7 pg/mg indicates alcohol consumption and provides evidence to refute a claim of abstinence. b
CAL-VMA-T calibrators (see Table 2 for composition and concentration range) on a CDx90 autoanalyzer. The assays were run according to manufacturer’s instructions. For all procedures reactions were read at 340 nm.
recommended cutoff values for hair analysis required for control of abstinence in cases of re-granting of drivers’ licences (Table 4). 2.3. Samples
2.1.2. LUCIO1-Direct ELISA kit The ELISA technology was based upon the competitive binding to antibody of enzyme labelled antigen and unlabelled antigen in proportion to their concentration in the reaction well. Hair samples were cut with scissors into small pieces and an aliquot was weighed (10 mg). Hair was washed with methanol (2 mL, 10 min) and allowed to dry. HEB hair extraction buffer (pH 4.5, 0.5 mL) was added and the tubes were capped and incubated for 2 h in an ultrasound bath at 75 8C. The total extract was removed and after cooling neutralization buffer was added (50 mL). An aliquot of 25 mL was used for amphetamine, methamphetamine, opiate, cocaine, methadone, and benzodiazepine ELISA and an aliquot of 100 mL was used for THC test. The assays were run according to manufacturer’s instructions. The microplate reader employed was an ELx 800 at wavelength of 450 nm (BioTek Instruments Inc., Winooski, USA) (Table 3). 2.2. Chromatographic analyses Drugs were quantified using gas chromatography in combination with mass spectrometry (GC/MS) or high performance liquid chromatographic methods coupled to mass spectrometry (LC/MS). Sample preparation for GC/MS comprehended a methanolic ultrasonication followed by solid-phase-extraction (cocain like compounds and opiates) or liquid-liquid extraction (THC and amphetamine/methamphetamine/methylenedioxyamphetamines) for clean-up and derivatization of the analytes before injection into the chromatographic systems. For LC/MS analysis the extract of methanolic ultrasonication was reconstituted in mobile phase and directly injected. All methods are standard procedures used in the routine laboratories in Bonn or Munich, fully validated according to guidelines of GTFCh [5] and accredited according to DIN EN ISO 17025. More details according chromatographic procedures were previously described [6–8]. The limits of detection (LODs) were below the
Table 2 Concentrations of calibrators (CAL-VMA-T) for the DRI1 tests (and CEDIA cannabinoid-OFT) according to the manufacturer’s recommendations [ng/mg].
Opiates Cocaine Methadone Amphetamine Ecstasy Cannabinoids
Level 1
Level 2
Level 3
Level 4
Level 5
0.00 0.00 0.00 0.00 0.00 0.00
0.48 0.50 1.00 0.80 0.50 0.13
0.96 1.00 2.00 1.60 1.00 0.26
1.92 2.00 4.00 3.20 2.00 0.52
3.84 4.00 8.00 6.40 4.00 1.04
Table 3 Concentrations of calibrators [ng/mg] for the Direct ELISA kit according to the manufacturer’s recommendations.
Opiates Cocaine Methadone Amphetamine Ecstasy Cannabinoids
Level 1
Level 2
Level 3
Level 4
Level 5
0.00 0.00 0.00 0.00 0.00 0.00
0.05 0.05 0.05 0.05 0.05 0.01
0.10 0.10 0.10 0.10 0.10 0.02
0.15 0.15 0.15 0.15 0.15 0.03
0.20 0.20 0.20 0.20 0.20 0.04
Authentic head hair samples were collected from cases of re-granting of drivers’ licences and other criminal cases, hair samples from other sites of the body were not enrolled. Only a few coloured hair samples were included, but no bleached samples. It was ensured that positive samples were preferentially enrolled in the present study with concentrations about recommended new lowered cutoff values but less with high drug concentrations. Few samples were diluted with blank hair to obtain concentration close to required limits. 2.4. Statistics Immunoassay cutoff values were calculated using contingency tables and receiver operating characteristic (ROC) curves [9] considering Youden-Indices (sensitivity + specificity 1) [10] with respect to measured chromatographic concentrations as ‘‘true’’ results. Sensitivity (true positives/(true positives + false negatives)), specificity (true negatives/(true negatives + false positives)), positive predictive value (PPV) (true positives/(false positives + true positives)), negative predictive value (NPV) (true negative/(true negative + false negative)) and the overall misclassification rate (OMR) ((false positive + false negatives)/all) were calculated by personal computer and use of Microsoft Office Excel 2003. Because of the fact that sensitivity and specificity are probabilities the standard error was calculated according to the following formula: standard error = (sensitivity or specificity (1 sensitivity or specificity)/sample size)1/2. Receiver operating characteristic (ROC) curves (sensitivity vs. (1 specificity)) were generated using
Table 4 Limits of quantification (LOQs) of the routine LC/MS and GC/MS procedures. Analyte
GC/MS
LC/MS
Tetrahydrocannabinol Anhydroecgoninmethylester Cocaine Norcocain Cocaethylene Benzoylecgonin Ecgoninmethylester Diacetylmorphin (heroin) 6-Monoacetylmorphine Morphine Codeine Dihydrocodeine Methadone EDDP Amphetamine Methamphetamine MDMA MDA MDE Diazepam Nordazepam Oxazepam Flunitrazepam 7-Aminoflunitrazepam Bromazepam Alprazolam Lorazepam
0.02
0.013 0.084 0.016 0.02 0.018 0.015 0.035 0.046 0.027 0.037 0.032 0.027 0.034 0.037 0.023 0.011 0.013 0.018 0.02 0.037 0.039 0.034 0.019 0.019 0.029 0.011 0.018
0.05
0.05
0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05
F. Musshoff et al. / Forensic Science International 215 (2012) 60–63
62
SPSS version 17.0 for Windows (SPSS Inc., Chicago, USA) to calculate cutoff values for each drug class.
3. Results and discussion 759 analyses were performed using LUCIO1-Direct ELISA kits and 936 analyses using DRI1 enzyme immunoassay tests. Sample size for each drug group and immunoassay test reached from 74 to 178. Immunoassay data were analyzed by the use of contingency tables that allowed calculation of sensitivity, specificity, positive and negative predictive value and the overall misclassification rate. ROC curves were used for all analytes to evaluate discrimination power of the tests and to confirm cutoff values (Figs. 1 and 2 for example). The Youden-Index was an additional tool to test quality of immunoassay tests in this study. Data for each test of correct negative, correct positive, false negative and false positive samples are given in Tables 5 and 6. The recommended cutoff value of 0.02 ng/mg for the test of THC revealed a high rate of false-negative samples in both tests. When increasing the cutoff value to 0.1 ng/mg as recommended by SoHT sensitivity of 92% and specificity of 87% were achieved with LUCIO1-Direct ELISA test. A sensitivity of at least 90% is recommended by the GTFCh [5]. However, a sensitivity of only 92% means to allow 8 individuals every 100 to seem negative even if using scheduled substances. Analyses of other drug classes tested with LUCIO1-Direct ELISA kits showed results with values of sensitivity of 91% for amphetamine up to 98% for methadone and values of specificity of 72% up to 89% (Table 7). The cutoff value of 0.1 ng/mg for the test of amphetamine, ecstasy, and methadone revealed high rates of false-negative and false-positive samples using DRI1 enzyme immunoassay tests resulting in values for sensitivity of 71–87% and values for specificity of 50–74% (Table 8). Only analyses of opiates and cocaine using the aforementioned test showed perhaps satisfying results with values of sensitivity of 94% and 99% and values of specificity of 73% and 91%. Considering the median values in Tables 7 and 8 it was ensured that positive samples were enrolled preferentially in the present study with concentrations about the recommended new lowered cutoff values but less with high drug concentrations. In contrast to urine or blood drug testing immunoassays for hair analysis are not common in daily routine due to low concentrations required for this purpose. Immunoassay evaluation for hair analysis is described for single substance classes [8,11–16] or as preliminary screening procedures considering higher cutoff values [17]. Also both procedures evaluated in the present study were recently tested by others and revealed satisfying results using calibration about recommended cutoff values of SoHT [18–20].
Fig. 2. ROC curve of amphetamine test using the DRI1 immunoassay test.
Fig. 1. ROC curve of amphetamine test using the Direct ELISA kit.
It has to be taken into consideration that sensitivity for preliminary screenings is required to approach as much as possible 100%. Acceptable results were not obtained with both assays with respect to new cutoff values recommended in Germany. A direct chromatographic measurement seems to be necessary.
Table 5 Summary of contingency tables for the Direct ELISA kit. Chromatographic analysis
+
Amphetamine 66 8
Methamphetamine + 4 41
Cannabinoids
+ 3 35
67 15
63 5
Opiates + 22 40
42 12
Cocaine + 5 78
Methadone + 2 33
29 10
+ 1 41
54 21
Benzodiazepines 78 10
+ 1 22
Table 6 Summary of contingency tables for the DRI1 immunoassay test. Chromatographic analysis
Amphetamine
+
52 31
+ 14 56
Methamphetamine 49 49
+ 6 41
Cannabinoids
46 20
+ 31 76
Opiates
41 4
Cocaine + 1 74
66 25
Methadone + 5 82
69 24
+ 21 53
F. Musshoff et al. / Forensic Science International 215 (2012) 60–63
63
Table 7 Summary of results for the LUCIO1-Direct ELISA kit.
Concentration range Median Chromat. cutoff value Immunoassay cutoff value Sensitivity (%) Specificity (%) PPV NPV OMR Youden
Amphetamine
Methamphetamine
THC
THC
Cocaine
Opiates
Methadone
Benzodiazepines
0.05–5.00 0.12 0.1 0.076 91 89 84 94 10 82
0.01–4.00 0.10 0.1 0.037 92 82 70 96 15 74
0.006–0.80 0.06 0.02 0.008 65 93 89 74 21 57
0.006–0.80 0.06 0.1 0.008 92 87 73 96 12 79
0.01–114.5 0.20 0.1 0.083 94 78 87 89 12 72
0.02–0.83 0.17 0.1 0.004 94 74 77 94 16 69
0.02–2.60 0.12 0.1 0.088 98 72 66 98 19 70
0.01–0.98 0.04 0.05 0.079 96 89 69 99 10 84
Table 8 Summary of results for the DRI1 test.
Concentration range Median Chromat. cutoff value Immunoassay cutoff value Sensitivity (%) Specificity (%) PPV NPV OMR Youden
Amphetamine
Methamphetamine
THC
THC
Cocaine
Opiates
Methadone
0.05–5.0 0.30 0.1 0.22 80 63 64 79 29 29
0.05–5.0 0.20 0.1 0.23 87 50 46 89 38 38
0.01–1.63 0.10 0.02 0.04 71 70 79 60 29 29
0.01–1.63 0.10 0.1 0.04 87 65 63 88 26 26
0.05–11.9 1.33 0.1 0.08 99 91 95 98 4 4
0.05–4.0 0.19 0.1 0.24 94 73 77 93 17 17
0.20 0.1 0.18 72 74 69 77 27 27
Acknowledgements Reagents and technical equipment needed for the present evaluation were provided by Thermo Fisher and NAL van Minden, respectively. Thanks to Mrs. Werthebach and Dr. Dormeier (Thermo Fisher) as well as to Dr. Kosovac (NAL von Minden) for their technical support. References [1] W. Schubert, R. Mattern, Beurteilungskriterien: Urteilsbildung in der Medizinisch-Psychologischen Fahreignungsdiagnostik 2. Aufl, Kirschbaum Verlag, Bonn, 2009. [2] Society of Hair Testing, Recommendations for hair testing in forensic cases, Forensic Sci. Int. 145 (2004) 83–84. [3] Substance Abuse and Mental Health Services Administration (SAMHSA), Mandatory Guidelines for Federal Workplace Drug Testing Programs. Draft #4, 2001, http://www.workplace.samhsa.gov/ResourceCenter/Dt/FA/GuidelinesDraft4. html (accessed 12/09/2010). [4] R. Agius, P. Kintz, Guidelines for European workplace drug and alcohol testing in hair, Drug Test. Anal. 2 (2010) 367–376. [5] Society of Forensic and Toxicological Chemistry (GTFCh), Richtlinien der GTFCh zur Qualita¨tssicherung bei forensisch-toxikologischen Untersuchungen, 2009, https://www.gtfch.org/cms/files/GTFCh_Richtlinie_For-Tox_Version%201.pdf (accessed 12/09/2010). [6] D. Thieme, H. Sachs, Improved screening capabilities in forensic toxicology by application of liquid chromatography–tandem mass spectrometry, Anal. Chim. Acta 483 (2003) 171–179. [7] F. Musshoff, F. Driever, K. Lachenmeier, D.W. Lachenmeier, M. Banger, B. Madea, Results of hair analyses for drugs of abuse and comparison with self-reports and urine tests, Forensic Sci. Int. 156 (2006) 118–123. [8] K. Lachenmeier, F. Musshoff, B. Madea, Determination of opiates and cocaine in hair using automated enzyme immunoassay screening methodologies followed by gas chromatographic-mass spectrometric (GC–MS) confirmation, Forensic Sci. Int. 159 (2006) 189–199.
[9] M.H. Zweig, G. Campbell, Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical medicine, Clin. Chem. 39 (1993) 561–577. [10] R. Fluss, D. Faraggi, B. Reiser, Estimation of the Youden Index and its associated cutoff point, Biom. J. 47 (2005) 458–472. [11] C. Moore, D. Deitermann, D. Lewis, B. Feeley, R.S. Niedbala, The detection of cocaine in hair specimens using micro-plate enzyme immunoassay, J. Forensic Sci. 44 (1999) 609–612. [12] G. Cooper, L. Wilson, C. Reid, D. Baldwin, C. Hand, V. Spiehler, Validation of the Cozart microplate ELISA for detection of opiates in hair, J. Anal. Toxicol. 27 (2003) 581–586. [13] G. Cooper, L. Wilson, C. Reid, D. Baldwin, C. Hand, V. Spiehler, Comparison of Cozart microplate ELISA and GC–MS detection of methadone and metabolites in human hair, J. Anal. Toxicol. 29 (2005) 678–681. [14] E. Han, E. Miller, J. Lee, Y. Park, M. Lim, H. Chung, F.M. Wylie, J.S. Oliver, Validation of the immunoanalysis microplate ELISA for the detection of methamphetamine in hair, J. Anal. Toxicol. 30 (2006) 380–385. [15] H. Miyaguchi, H. Takahashi, T. Ohashi, K. Mawatari, Y.T. Iwata, H. Inoue, T. Kitamori, Rapid analysis of methamphetamine in hair by micropulverized extraction and microchip-based competitive ELISA, Forensic Sci. Int. 184 (2009) 1–5. [16] P. Lopez, S. Martello, A.M. Bermejo, V.E. De, M.J. Tabernero, M. Chiarotti, Validation of ELISA screening and LC–MS/MS confirmation methods for cocaine in hair after simple extraction, Anal. Bioanal. Chem. 397 (2010) 1539–1548. [17] M.L. Pujol, V. Cirimele, P.J. Tritsch, M. Villain, P. Kintz, Evaluation of the IDS onestep ELISA kits for the detection of illicit drugs in hair, Forensic Sci. Int. 170 (2007) 189–192. [18] R. de la Torre, E. Civit, F. Svaizer, A. Lotti, M. Gottardi, M. Miozzo, High throughput analysis of drugs of abuse in hair by combining purposely designed sample extraction compatible with immunometric methods used for drug testing in urine, Forensic Sci. Int. 196 (2010) 18–21. [19] C. Coulter, M. Taruc, J. Tuyay, C. Moore, Quantitation of tetrahydrocannabinol in hair using immunoassay and liquid chromatography with tandem mass spectrometric detection, Drug Test. Anal. 1 (2009) 234–239. [20] C. Coulter, J. Tuyay, M. Taruc, C. Moore, Semi-quantitative analysis of drugs of abuse, including tetrahydrocannabinol in hair using aqueous extraction and immunoassay, Forensic Sci. Int. 196 (2010) 70–73.