Forensic Science International - Science Direct

Forensic Science International 84 (1997) 271-279

ELSEVIER

Forensic Science International

Influence of sample preparation on analytical results: drug analysis [GC/MS] on hair snippets versus hair powder using various extraction methods. Hans P. Eser”y*, Lucia Pijtschb, Gisela Skopp’, Manfred R. Moeller” aInstitute of Legal Medicine, University of Saarland, D-66421 Homburg/Saar, Germany bInstitute of Legal Medicine, Johannes-Gutenberg-University,D-55131 Mairu, Germany ‘Institute of Legal Medicine, Ruprecht-ffirls-University, D-69115 Heidelberg, Germany

Abstract The comparison of aqueous extraction methods and hair extraction by organic solvents performed on hair powder as well as on hair snippets of the same sample revealed different qualities of the procedures. Qualitative and quantitative results by the same derivatization step and GC/MS detection demonstrated, that the risk of missing a drug substance is higher using hair snippets than after drug extraction on pulverised hair. Drug recovery for opiates, cocaine and benzoylecgonine from hair was found to be best in aqueous solvents or in methanol extracts. The results are discussed under the aspects of solid-phase extraction, the hair sample representing an inhomogenous material. The localisation of drug molecules in hair, the hair swelling and penetration behavior of the particular extraction medium as well as the partition coefficient of solvent/hair phase for a particular drug substance are considered to influence the results. 0 1997 Elsevier Science Ireland Ltd. All rights reserved Keywords: Hair analysis; Hair extraction; Comparison of extraction procedures; Drug recovery; Partition coefficient; Solvent/hair phase

1. Introduction

Sample preparation and sample extraction are the most important steps of an * Corresponding author. 0379-0738/97/$17.00 0 1997 Elsevier Science Ireland Ltd. All rights resewed PfI SO379-0738(96)02071-3

272

H.P. Eser /Forensic

Science International 84 (1997) 271-279

analytical procedure. Especially in analysis of inhomogenous materials, implications on the analytical results seem obvious. Previously, in 1981, Valente et al. [l] reported on a comparison of procedures for extracting morphine and cocaine from hair. They found that morphine can be extracted with aqueous solutions with similar yields to hot methanol, a finding which was confirmed by other study groups [2-41. They suggested however, crushing of hair did not improve drug recovery. To our knowledge, studies on drug recoveries comparing the extraction matrices hair powder versus hair snippets have not been published to date. Drug extractions on both specimens are commonly used for expertise. Recently, Drummer and coworkers [5] called attention to the problem that analytical results in hair analysis might be more influenced by sample preparation and extraction procedures than hitherto assumed. They stressed aspects of drug stability during various extraction procedures and compared the yields of psychotic drugs from hair using different extraction methods. In most forensic cases mainly the qualitative results are evident. Thus, the method applied to hair samples should offer reliable, overall detection of illicit drugs. For all current methods, that do not solubilise hair material completely, drug elution from hair fibers as well as from powdered hair may be regarded as solid phase extractions. The aim of this pilot study was to elucidate some influences of sample preparation on the analytical results in hair analysis. 2. Materials

and methods

2.1. Chemicals, reagents and materials

Solutions of drug reference standards containing 0.1 g/l d,-cocaine, d,-benzoylecgonine, 6-monoacetylmorphine, d,-6-monoacetylmorphine, d,-morphine, and d,-codeine were purchased from Radian Corp. (Austin, TX, USA) as well as solutions of benzoylecgonine and morphine of 1 g/l. Standard solutions of cocaine and codeine at 1 g/l were purchased from Sigma (Deisenhofen, Germany). All reagents used were of analytical grade. PBS-tabs, acetic acid, arylsulfatase/p-glucuronidase as well as analytical grade potassium hydroxide, sodium bicarbonate, urea, acetone, methanol, dichloromethane and 2-propanol were purchased by Merck (Darmstadt, Germany). Pentafluoropropionic anhydride and 2,2,3,3,3-pentafluoro-1-propanol were purchased from Aldrich (Steinheim, Germany). Chromabond” C,, ec SPE columns (200 mg, 3 ml> were from Macherey and Nagel (Dueren, Germany). The GC-column was a HP-l fused-silica capillary column (12 m x 0.2 mm i.d., 0.33 pm film thickness) from Hewlett-Packard (Waldbronn, Germany). 2.2. Collection and preparation of the hair samples

Hair samples from five persons with a known history of drug abuse were collected at autopsy. The hair strands were washed in 10 ml ethanol/methanol

H.P. Eser /Forensic


273

(9:1, v/v) at room temperature for 3 min. Then the proximal 4-cm-segments of each strand were divided longitudinally into 10 portions. Five portions of a particular hair sample were cut into pieces of approximately l-3 mm in length, the other five portions were pulverised for 10 min in a ball mill (Retsch, Haan, Germany). 2.3. Extraction and derivatization Each of the following extraction methods (A-E) was applied to the corresponding extraction matrix (hair powder and hair snippets) of a particular hair sample. 2.3.1. Method A To 50 mg of hair powder or snippets, 5 ml of methanol and 100 ng of internal standard were added, respectively. The mixture was ultrasonicated for 5 h at approximately 40°C. 2.3.2. Method B To 20 mg of pulverised/cut hair 2 ml of acetate buffer (pH 41, 100 ng of internal standard and 60 ~1 of arylsulfatase/P-glucuronidase were added. The mixture was incubated at 42°C for 1.5 h. After incubation, the mixture was neutralised with sodium bicarbonate and applied to the preconditioned SPE-column. The column was washed with 3 ml of distilled water, 3 ml of 5% sodium hydrogen carbonate and again with 3 ml of distilled water. The analytes were eluted by acetone/dichloromethane (3:1, v:v). 2.3.3. Method C To 20 mg of hair powder or hair snippets, respectively 2 ml of 8 M aqueous urea solution and 100 ng of internal standard were added and incubated at room temperature overnight. The extracts were worked up with SPE as described above. 2.3.4. Method D To 50 mg of hair, 5 ml of acetone and 100 ng of internal standard were added. The mixture was ultrasonicated for 5 h at approximately 40°C. 2.3.5. Method E To 50 mg of hair powder or snippets, 5 ml of 2-propanol and 100 ng of internal standard were added. The mixture was ultrasonicated for 5 h at approximately 40°C. All organic solvents were evaporated at 40°C under a slight flow of nitrogen. Derivatization was performed with 100 ~1 of PFPA and 75 ~1 of PFPOH at 60°C for 30 min. After evaporation under nitrogen at 60°C the residue was dissolved in 100 ~1 ethyl acetate for methods A, D and E and in 30 ~1 for methods B and C. 2.4. Instrumentation and chromatographic co, Zitions The analyses were performed using a Hewlett-Packard 5971A mass selective detector in electron impact-selected ion monitoring (EI-SIM) mode equipped with

274

H.P. Eser /Forensic


a Hewlett Packard 5890 Series II gas chromatograph, a Hewlett-Packard 7673 auto sampler and Hewlett-Packard B.02.04 Chem Station software installed at an IBM-compatible 486 PC. Chromatographic separation was achieved on a HP-l fused-silica capillary column (12 m X 0.2 mm i.d., 0.33 ,um film thickness) from Hewlett-Packard (Waldbronn, Germany) using helium as carrier gas. Temperature program started with 70°C for 2 min, increased at 30”C/min to 200°C at 2”C/min to 215”C, then at 30”C/min to 300°C. The final temperature was held for 10 min.

Comparison of the qualitative results on hair snippets [n=5]

Comparison of the qualitative results on hair powder [n=5] rcslts

cm hair

coc~yns (4x);BZE (lx);wdeine

Fig. 1. Comparison of the qualitative

(2x)

results on hair snippets versus hair powder.

H.P. Eser /Forensic ScienceInternational 84 (1997)271-279

n Snippets

275

IPowder

Fig. 2. Analytical results of case 1. Comparison of hair snippets (1st column) versus hair powder (2nd column) for method B. COC, cocaine; BZE, benzoylecgonine; MOR, morphine; MAM, 6monoacetylmorphine; COD, codeine; DHC, dihydrocodeine.

The injector temperature was maintained at 260°C. MS-interface temperature was 280°C. 3. Results

The results clearly demonstrated that qualitative and quantitative drug recovery from hair powder is higher than from hair snippets (Figs. l-4). The analytical results were also strongly influenced by the extraction solvent. Generally, the following graduation for drug recovery was observed: Hair powder > hair snippets Aqueous solutions = methanol > acetone > isopropanol The best results were obtained by performing method A, B or C on hair powder. The best qualitative drug recovery resulted from method C. However, in most of the cases drug concentrations were lower than using method A or B. Studies on drug stability revealed, that shortening the incubation time of method C will increase the quantitative results. Further studies to optimise this method are under current investigation. Although differences of drug concentrations in methanol extracts from hair snippets and hair powder were also present, fewer differences resulted compared to the other methods. Methanol extracts showed very high matrix loading. Derivatization by trifluoroacetic acid as well as TMS-derivates were found even to increase the problems of interpretation of the chromatograms. We agree with Cone [2], that methanol extracts should be cleaned up by solid-phase extraction prior to GC/MS. Overall, in view of all analytical results as well as the purity of the extracts, the best results in this preliminary study were obtained by method B.

276

H.P. Eser /Forensic

Fig. 3. Analytical


results on hair snippets comparing methods A-E.

4. Discussion

Numerous methods to extract drugs from hair have been published [6-151. Besides drug stability, one of the requirements for an appropriate extraction procedure is that the extraction solvent must gain access to the various localisations of drug substances in hair. Complete dissolution of the hair components seems to be best. However, for most drugs, solubilisation of hair is problematic, in most cases even impossible, due to the instability of the drug substances under the chemical conditions required to dissolve hair. Solid-phase extraction procedures on mechanically desintegrated fibers present an alternative. Destruction of the hair structure can be produced by pulverizing the hair fibers or to a lesser extent by ultrasonication of the hair sample in organic solvents [16]. Rothe et al. investigated the capability of various solvents differing in polarity and hydrophilicity for direct extraction of opiates from hair snippets but not on hair powder [17]. Concerning

H.P. Eser /Forensic

Science Intemaiional84

Dihydmmd~

Morphlnc 1

10 9 I 3

277

(1997) 271-279

6 hG

’

5

I’ a3 r’ L

I 0

0 L

2.

,

.

cur

Fig. 4. Analytical

5

I

I

1

1

cm

results on hair powder comparing methods A-E.

hair snippets, their results were confirmed by our findings. From the present results however, it is also evident, that the risk of missing a drug substance is much higher in direct extraction from hair snippets than from hair powder, no matter what solvent is used. It could be demonstrated, that the swelling properties of hair in the particular extraction solvent are highly important. This effect has been already stressed by Baumgartner and his coworkers for the application of various washing procedures [18]. It had been already shown, that there is a graduation in the rate, in which alcohols and other organic solvents enter a keratinized fiber. Isopropanol swells the hair to a lesser extent than water, methanol or acetone and was found rarely to enter the hair. In contrast, urea is known to produce swelling of hair far beyond that of water, it breaks down hydrogen bonds and denatures proteins [19]. Thus, due to its nature, it promotes both, improved access to drug molecules in hair

278

H.P. Eser /Forensic


snippets and in hair powder, as well as facilitating drug diffusion out of hair. The present study confirms these reflections. The following requirements for drug elution are summarised: The solvent must reach the sites where the particular drug molecules are located. It should not decompose the analytes and according to the principles of solid-phase extractions, it should have an optimal partition coefficient for the solvent/hair phase. Overall, aqueous drug elution from hair was found, as already known, to be similar to methanol, but better than using other organic solvents. These findings might be explained by the physicochemical properties and the different localisations of drug molecules in hair. Dye diffusion studies in human hair already provided evidence, that for given time periods, organic solvents will not penetrate keratinized hair in the same magnitude as water molecules [20]. Although most non-ionized drugs dissolve better in methanol, water molecules are able to gain accessto all structures in the keratinized hair and enable sufficient diffusion from the hair phase. In contrast, organic solvents preferentially elute drug substances from the hair powder surface as well as from the peripheral interior areas of the hair fragments, provided that the solvent can enter keratinized tissue at all. Considering the biochemical concept of drug incorporation [21], the binding of drug molecules may occur to the cell membrane complex, to the melanin granules as well as to the hair proteins. The cell membrane complex (CMC) is highly exposed in hair powder and most affected by ultrasonication. From all alcohols, methanol is known to penetrate keratinized material best. Thus, similar drug recovery from hair by methanol and by aqueous solutions can be understood. Aqueous extraction procedures from hair powder as well as methanol elution allow a reasonable screening for evident drugs of abuse. Hair analysis done in this way provides semiquantitative rather than quantative results. It must be stressed, that for complete drug recovery from hair it has to be considered that different, highly specific extraction procedures for any particular substance are required. References

111 D.

Valente, M. Cassini, M. Pilagopochi and G. Vanssetti, Hair as asample In assessing morphine or cocaine addiction. Clin. Chem., 27 (1981) 1952-1953. El E.J. Cone, Testing human hair for drugs of abuse. J. Anal. Toxicol., 14 (1990) l-7. im Kopfhaar ohne [31 G. Kauert, L. V. Meyer and I. Herrle, Drogen- und Medikamentennachweis Estraktion des Haaraufschlusses mittels GC/MS. Zentralbl. Rechrsmed., 38 (1992) 33. [41 H. Sachs and I. Raff, Comparison of quantitative results of drugs in human hair using GC/MS. Forensic Sci. Int., 63 (1993) 207-216. 151 F.J. Couper, I.M. McIntyre and O.H. Drummer, Extraction of psychotic drugs from human scalp hair. J. Forensic Sci., 40 (1995) 83-86. bl A.M. Baumgartner, P.J. Jones, W.A. Baumgartner and C.T. Black, Radioimmunoassay of hair for determining opiate abuse histories. J. Nucl. Med., 20 (1979) 748-752. [71 W. Arnold and K. Piischel, Experimental studies on hair as an indicator of post or present drug use. J. Forensic Sci., 21 (1981) 83. B1 M. Marigo, F. Tagharo, C. Poiesi, S. Lafsca and C. Neri, Determination of morphine in the hair of heroin addicts by high performance liquid chromatography with fluorimetric detection. J. Anal. Toxicol., 10 (19861 158-161.

H.P. Eser /Forensic

Science International

84 (1997) 271-279

279

[91 B.A. Goldberger, Y.H. Caplan, T. Maguire and E.J. Cone, Testing human hair for drugs of abuse. II. Identification of heroin and 6-acetylmorphine as indicator of heroin abuse. J. Anal. Toxicol., 15 (1991) 226-231. [lOI Y. Nakahara, K. Takahashi, M. Shimamine and A. Saitoh, Hair analysis for drugs of abuse. IV. Determination of total morphine and confirmation of 6-actetylmorphine in monkey and human hair by GC/MS. Arch. Toxicol., 66 (1992) 669-614. illI M.R. Moeller, Drug detection in hair by chromatographic procedures. J. Chrornatogr., 580 (1992) 125-134. [121 M.J. Welch, L.T. Sniegoski, C.C. Allgood and M. Habram, Hair analysis for drugs of abuse: evaluation of analytical methods, environmental issues and development of reference materials. J. Anal. Toxicol., 17 (1993) 389-398. iI31 M. Chiariotti, Overview on extraction procedures. Forensic Sci. Int., 63 (1993) 161-170. 1141 C. Offidani, S. Strano Rossi and M. Chiarotti, Improved enzymatic hydrolysis of hair. Forensic Sci. ht., 63 (1993) 171-173. [151 W.A. Baumgattner and V.A. Hill, Sample preparation techniques, Forensic Sci. Int., 63 (1993) 121-135. M L. Pdtsch, G. Skopp, H.P. Eser and M.R. Moeller, Influence of sample preparation on the analytical results. I. Studies on the extraction matrix by electron microscopy [SEM/TEM]. 1st European Meeting on Hair Analysis, 17-19 June (1996), Genoa, Italy. 1171 M. Rothe and F. Pragst, Solvent optimization for the direct extraction of opiates from hair samples. J. Anal. Toxicol., 19 (1995) 236-240. [181 W.A. Baumgartner and V.A. Hill, Hair analysis for drugs of abuse: decontamination issues. In I. Sunshine (ed.), Recent Developments in Therapeutic Drug Monitoring and Clinical Toxicology. Marcel Dekker, New York, 1992, pp. 555-563. [191 C.R. Robbins, Chemical and physical behavior of human hair, Springer, New York, 1988. DO1 L. Potsch and M.R. Moeller, On pathways for small molecules into and out of human hair fibers. J. Forensic Sci., 41 (1996) 121-125. ml L. Potsch, G. Skopp and M.R. Miiller, Zum Suchtmittelnachweis in Haaren. II. Eine biochemische Erkliirung fiir den endogenen Einbau von Pharmaka und illegalen Drogen ins Haar wahrend der Anagenphase. Rechtsmedizin, 7 (1996) l-5.