Simultaneous Detection and Quantification of Amphetamines Diaz by mikeholy


									Journal of Analytical Toxicology, Vol. 32, September 2008

Simultaneous Detection and Quantification of
Amphetamines,Diazepam and its Metabolites,
Cocaine and its Metabolites, and Opiates in Hair by
LC–ESI-MS–MS Using a Single Extraction Method
Eleanor I. Miller*, Fiona M. Wylie, and John S. Oliver
Forensic Medicine and Science, University of Glasgow, University Place, Glasgow, G12 8QQ, Scotland

 Abstract                                                                                        opiate abuse histories. It has since been investigated as an al-
                                                                                                 ternative biological matrix to more conventional matrices such
A liquid chromatography–tandem mass spectrometry method was                                      as blood and urine, and advantages to using hair have been
developed and validated for the simultaneous identification and                                  identified. These include the ability of hair testing to reveal
quantification of amphetamines, diazepam and its metabolites,
                                                                                                 chronic drug use, which is desirable for drug monitoring; non-
cocaine and its metabolites, and opiates from hair using a single
                                                                                                 invasive sample collection that can be easily supervised, re-
extraction method. As part of the method development, Gemini
C18, Synergi Hydro RP, and Zorbax Stablebond-Phenyl LC columns                                   ducing the risk of sample adulteration; and the application of
were tested with three different mobile phases. Analyte recovery                                 segmental analysis to determine the approximate time of drug
and limit of detection were evaluated for two different solid-phase                              exposure. A restriction of hair testing is the limited sample
extraction methods that used Bond Elut Certify™ and Clean Screen®                                weight provided for testing and the potentially low drug con-
cartridges. Phosphate buffer (pH 5.0) was chosen as the optimum                                  centrations. A single extraction method that can be used for the
hair incubation medium because of the high stability of cocaine                                  analysis of multiple drug groups is particularly useful for sam-
and 6-monoacetylmorphine using this method and faster sample                                     ples of low weight.
preparation. The optimized method was fully validated. Linearity                                    There are, however, still some controversies with respect to
was established over the concentration range 0.2–10 ng/mg hair,                                  the exact mechanisms of drug incorporation, passive contam-
and the correlation coefficients were all greater than 0.99. Total
                                                                                                 ination and the effectiveness of wash procedures, and the po-
extraction recoveries were greater than 76%, detection limits were
                                                                                                 tential for bias of cosmetically treated hair and of hair from in-
between 0.02 and 0.09 ng/mg, and the intra- and interday
imprecisions were generally less than 20% in spiked hair. The intra-                             dividuals of different cultures. Until these issues are resolved,
and interbatch imprecision of the method for a pooled authentic                                  hair testing should not be a single source of evidence in a tox-
hair sample ranged from 1.4 to 23.4% relative standard deviation                                 icological investigation, but where possible should be used in
(RSD) and 8.3 to 25.4% RSD, respectively, for representative                                     combination with blood and urine analysis to provide infor-
analytes from the different drug groups. The percent matrix effect                               mation on both chronic and acute drug use or to rule out the
ranged from 63.5 to 135.6%, with most analytes demonstrating ion                                 use of drugs. The consensus opinion of the Society of Forensic
suppression. Sixteen postmortem samples collected from suspected                                 Toxicologists (SOFT), which currently proposes that hair evi-
drug-related deaths were analyzed for the 17 drugs of abuse and                                  dence is a useful specimen in forensic investigations only when
metabolites included in the method. The method was sufficiently                                  supported by additional evidence of drug use, has not changed
sensitive and specific for the analysis of drugs and metabolites in
                                                                                                 since 1990 (2).
postmortem hair samples. There is scope for the inclusion of other
                                                                                                    Reports on hair testing in the scientific literature generally
target drugs and metabolites in the method.
                                                                                                 restrict analysis to one or two drug groups (3–9). Routine hair
                                                                                                 samples are initially screened by immunoassay (particularly in
                                                                                                 workplace testing where there are large numbers of negatives)
Introduction                                                                                     and then positives are confirmed with a more sensitive and
                                                                                                 specific technique such as gas chromatography–mass spec-
                                                                                                 trometry (GC–MS) or liquid chromatography–tandem mass
  Hair was first analyzed for drugs almost 30 years ago by
                                                                                                 spectrometry (LC–MS–MS). The Society of Hair Testing (SoHT)
Baumgartner et al. (1) in 1979 for the purpose of determining
                                                                                                 proposed guidelines in 2004 for immunoassay screening and
* Author to whom correspondence should be addressed.                                             confirmation cut-offs for amphetamines, cocaine and its

                                     Reproduction (photocopying) of editorial content of this journal is prohibited without publisher’s permission.           457
                                                                                 Journal of Analytical Toxicology, Vol. 32, September 2008

metabolites, and opiates in hair (10). GC–MS and LC–MS–MS          hydroxide, ethyl acetate, formic acid, dichloromethane,
methods have been reported for the simultaneous qualitative        propan-2-ol, and potassium dihydrogen phosphate were pur-
analysis of common drugs of abuse in hair (11–13) and the          chased from BDH (Poole, U.K.) and were of analytical grade.
simultaneous quantification of two or three target drug groups     Ammonium formate and sodium dodecyl sulfate were pur-
(3–9,14–19). Recently, methods using capillary zone elec-          chased from Sigma-Aldrich (Dorset, U.K.). d,l-Amphetamine,
trophoresis–MS have been reported for the quantitative anal-       d,l-methamphetamine, d,l-MDA, d,l-MDMA, d,l-MDEA, di-
ysis of three drug groups in hair (16,17).                         azepam, nordiazepam, oxazepam, temazepam, cocaine, ben-
   The simultaneous quantification of amphetamines, benzo-         zoylecgonine, ecgonine methyl ester, cocaethylene, mor-
diazepines, cocaine and its metabolites, and opiates in hair by    phine, 6-monoacetylmorphine (6-MAM), codeine, and
GC–MS has been reported by one group (20). The method in-          dihydrocodeine were obtained from Promochem (Teddington,
volved an overnight acid extraction at 50°C followed by a mixed    U.K.). Deuterated internal standards, including d,l-am-
mode solid-phase extraction (SPE) clean-up and a single GC–        phetamine-d5, d,l-methamphetamine-d5, d,l-MDA-d5, d,l-
MS injection, operated in full scan mode. An earlier study by      MDMA-d5, d,l-MDEA-d5, diazepam-d5, nordiazepam-d5, ox-
the same group developed a qualitative screening method for        azepam-d5, temazepam-d5, cocaine-d3, benzoylecgonine-d3,
the same drugs reported in their more recent quantitative          ecgonine methyl ester-d3, cocaethylene-d3, morphine-d3, 6-
method. The qualitative method used an overnight incuba-           MAM-d3, codeine-d3, and dihydrocodeine-d6, were obtained
tion in methanol for 18 h at 45°C with no clean-up step. Sig-      from LGC Promochem. World Wide Monitoring Clean
nificant baseline noise was observed using this method and the     Screen® SPE columns were purchased from Kinesis (Cam-
authors noted that this could potentially cause difficulty in      bridgeshire, U.K.). Bond Elut Certify™ SPE columns were
the identification of low drug concentrations. A qualitative       purchased from Crawford Scientific (Strathaven, U.K.).
LC–MS–MS screening method has been reported for the anal-
ysis of amphetamines, benzodiazepines, cocaine and its             Standard solutions
metabolites, and opiates (12). The drugs were extracted from         Individual drug stock standard solutions and deuterated
the hair by incubation in acidic mobile phase. An aliquot of ex-   drug standards were obtained as 100 µg/mL prepared in
tract was injected into the LC–MS–MS system. The authors           methanol. A combined working drug solution of d,l-am-
proposed that this method could be used as an alternative to       phetamine, d,l-methamphetamine, d,l-MDA, d,l-MDMA, d,l-
immunoassay screening. This method was not, however, de-           MDEA, diazepam, nordiazepam, oxazepam, temazepam, co-
veloped for quantitative purposes.                                 caine, benzoylecgonine, ecgonine methyl ester, cocaethylene,
   This present study reports the development and validation of    morphine, 6-MAM, codeine, and dihydrocodeine was prepared
a single extraction procedure for the quantification of am-        at 1 µg/mL by 100-fold dilution with methanol. This was
phetamines, diazepam, and its metabolites, cocaine and its         achieved by adding 25 µL of each 100 µg/mL drug solution
metabolites, and opiates in hair collected from suspected drug-    into a 25-mL volumetric flask and making up the volume to
related deaths. The drugs included in the method were those        the 25-mL mark with methanol. A combined working deuter-
that are commonly encountered in our laboratory in blood           ated internal standard solution of d,l-amphetamine-d5, d,l-
samples as identified by the in-house database. The method was     methamphetamine-d5, d,l-MDA-d5, d,l-MDMA-d5, d,l-MDEA-
validated for 17 drugs of abuse and metabolites and applied to     d5, diazepam-d5, nordiazepam-d5, oxazepam-d5, temazepamd5,
postmortem samples generally in the range of 10–30 mg for          cocaine-d3, benzoylecgonine-d3, ecgonine methyl ester-d3,
segments excluding roots, but also to root samples, some of        cocaethylene-d3, morphine-d3, 6-MAM-d3, codeine-d3, and di-
which weighed less than 1 mg. A single extraction method           hydrocodeine-d6 was also prepared at 1 µg/mL in a similar
that could be applied in the analysis of a wide range of drugs     way to the working drug solution.
would prove to be very useful, particularly for the analysis of
weight-limited hair samples taken from polydrug users, in-         Instrumentation
cluding drug-related death cases where several drug classes           LC–MS–MS analysis was carried out using a Surveyor HPLC
could be present in each sample.                                   system with an LCQ Deca XP Plus™ ion trap MS (Thermo
                                                                   Finnigan, San José, CA). During method development, three
                                                                   different LC columns were tested. The Gemini and Synergi
                                                                   Hydro RP columns were purchased from Phenomenex (Tor-
Experimental                                                       rance, CA). Both were C18 columns but the Synergi Hydro RP
                                                                   column also had polar end-capping. The Zorbax Stablebond
Samples                                                            (SB) Phenyl column was purchased from ChromTech
   All samples were postmortem head hair samples submitted         (Cheshire, U.K.). This was a non-endcapped phenyl modified
to the toxicology laboratory of the Forensic Medicine and Sci-     silica column. The columns (and their dimensions) used were
ence Section at the University of Glasgow for routine testing.     the Gemini C18 (150 mm × 2.0 mm, 3-µm particle size), the
The samples were tested for diagnostic purposes, and the re-       Synergi Hydro RP (150 mm × 2.0 mm, 4-µm particle size), and
sults were reported to Procurators Fiscal.                         the Zorbax Stablebond (SB) (50 mm × 2.1 mm, 3.5-µm particle
                                                                   size). These columns were fitted with guard columns with
Chemicals                                                          identical packing material which was purchased from the same
  Methanol, acetone, acetonitrile, acetic acid, ammonium           companies as the LC columns. Guard column dimensions were

Journal of Analytical Toxicology, Vol. 32, September 2008

4 mm × 2.0 mm, 5-µm particle size for the Gemini column; 4         investigated for the simultaneous extraction of the drugs of
mm × 2.0 mm, 4-µm particle size for the Synergi Hydro RP           abuse from hair. The Bond Elut Certify method was previously
column; and 12.5 × 2.1 mm, 5-µm particle size for the Zorbax       published, having been applied in the simultaneous screening
SB Phenyl column.                                                  of acidic, neutral, and basic drugs from oral fluid (21). The car-
                                                                   tridges were conditioned with 2 mL methanol and 2 mL phos-
                                                                   phate buffer (pH 6.0). After the samples were applied to the car-
                                                                   tridges, the cartridges were washed with 1 mL deionized water
Method Development                                                 followed by 0.5 mL 0.01 M acetic acid. The cartridges were then
                                                                   air dried under full vacuum for 10 min, and 50 µL methanol
Initial tuning and ion identification                              was added; the cartridges were dried for an additional 2 min.
  The optimum tuning parameters, precursor, and product            The elution step for the basic analytes was modified slightly
ions were identified for each analyte. This was achieved           from this published method and involved 1.5 mL ethyl acetate
through a combination of automatic and manual tuning.              with 2% aqueous ammonium hydroxide and 1.5 mL
                                                                   dichloromethane/isopropanol/aqueous ammonium hydroxide
Optimum LC and mobile phase combination                            (78:20:2, v/v/v) with a 2-min drying step between the two so-
   Initially, three mobile phase systems and three LC columns      lutions.
were investigated to determine which mobile phase system              The Clean Screen SPE method intended for the extraction of
and column produced the greatest peak area response for the        drugs of abuse in urine and provided by the SPE manufac-
target analytes (5,12,21). The Gemini C18, Synergi Hydro RP,       turers was adapted for hair extracts. Clean Screen (ZSDAU020)
and Zorbax SB columns were each tested with 3 mM ammo-             extraction cartridges were conditioned sequentially with 3 mL
nium formate + 0.001% formic acid and acetonitrile, 10 mM          methanol, 3 mL deionized water, and 1 mL phosphate buffer
ammonium acetate + 0.001% formic acid and acetonitrile,            (0.1 M, pH 5.0). The vortex mixed samples were loaded in 2 mL
and 10:10:80 methanol/acetonitrile/20 mM formate buffer and        phosphate buffer (0.1 M, pH 5.0) and allowed to drip through
35:35:65 methanol/acetonitrile/20 mM formate buffer gradient       without vacuum. The columns were then washed with 3 mL
systems.                                                           phosphate buffer (0.1 M, pH 5.0) and 1 mL acetic acid (1.0 M)
   Three 20-µL injections of 50 ng drug/200 µL mobile phase        and dried for 5 min under full vacuum. The drugs were eluted
were run for each column and mobile phase system combina-          using 2 mL methanol/2% aqueous ammonium hydroxide.
tion. The retention times of the analytes were also compared
for each system.                                                   Comparison of SPE methods
                                                                      SPE clean-up. Hair was collected from five members of lab-
Stability of cocaine and 6-MAM                                     oratory personnel. Ten milligrams ± 0.1 mg hair was washed
   The stability of cocaine and 6-MAM was tested in a variety of   and subsequently extracted using the Bond Elut Certify and
incubation media including methanol, phosphate buffer pH           Clean Screen methods. The chromatograms for each blank hair
5.0, 0.1 M, 0.05 M, and 0.1 M hydrochloric acid and methanol       extract using both SPE methods were compared visually to de-
containing 1, 2, 5, 10, and 25% ammonium hydroxide (v/v).          termine which method produced the cleanest extracts overall.
   Three samples were prepared by spiking 1.5 mL of the par-          Total extraction recovery from spiked hair samples. In the
ticular incubation medium with 50 ng of cocaine and 6-MAM.         spiked hair experiment, 10 mg ± 0.1 mg blank hair was
The ammoniated methanol incubations were subsequently left         weighed out into three vials and was spiked with 50 ng analyte
to incubate for 18 h at room temperature. The aqueous acidic       mix. Two unextracted samples were prepared at the same con-
media, methanol, and phosphate buffer incubations were sub-        centration and kept refrigerated at 4°C during the extraction.
sequently left for 18 h at 45°C. After incubation, 50 ng of co-    The samples were then incubated in 1.5 mL phosphate buffer
caine-d3 and 6-MAM-d3 were added to the vials and with the ex-     (pH 5.0) for 18 h at 45°C and extracted by the Bond Elut Cer-
ception of the phosphate buffer extract, the vial contents were    tify and Clean Screen methods. Fifty nanograms of deuterated
evaporated to dryness at room temperature under nitrogen.          internal standard was added to the eluant after the extraction
The phosphate buffer extracts were extracted by SPE , and the      and also to the unextracted samples. The samples were evapo-
results from the extracts were compared to three unextracted       rated to dryness at room temperature under nitrogen and sub-
standards to determine stability. Three unextracted standards      sequently reconstituted in 200 µL of mobile phase initial con-
were prepared at the same concentration and kept in the fridge     ditions. Twenty microliters was injected for each sample.
at 4°C during the incubation period. Fifty nanograms of deuter-       Incubation recovery from authentic hair samples. Three
ated internal standard was added to the unextracted standards      postmortem hair samples which had drug-positive blood re-
at the same time as the incubated samples, and the vial con-       sults were washed once with 0.1% sodium dodecyl sulfate
tents were evaporated to dryness under nitrogen, with the ex-      wash, twice with deionized water washes, and twice with
ception of the phosphate buffer extract. The samples, once         dichloromethane washes. Each wash involved a 10-min ultra-
dry, were reconstituted in 200 µL of mobile phase initial con-     sonication. The root–0.5 cm segment was removed, and the
ditions, and 20 µL was injected.                                   0.5–3.5 cm segment was cut up into smaller segments of 2–3
                                                                   mm in length. This cut-up segment was subsequently split to
SPE methods                                                        compare the methanol and phosphate buffer incubation
  Bond Elut Certify and Clean Screen DAU SPE columns were          methods. After the incubation, the samples were extracted

                                                                                     Journal of Analytical Toxicology, Vol. 32, September 2008

using the Bond Elut Certify method described in the following         LOD and LOQ
section.                                                                These data had already been established as part of the com-
                                                                      parison of SPE methods.
Limits of detection (LOD) and quantification (LOQ)
  The LOD and LOQ were determined for each drug using                 Imprecision of the procedure using authentic hair samples
spiked hair. Ten milligrams of blank washed hair was spiked              A pooled hair sample was tested to determine the impreci-
with 0.1, 0.2, 0.3, 0.4, 0.5, 0.75, 1, 2, and 5 ng of each drug and   sion of the procedure using authentic hair samples which con-
50 ng of deuterated internal standard. The samples were then          tained incorporated drugs. Imprecision experiments using
incubated, extracted by both SPE methods, and analyzed by             spiked hair do not truly represent the imprecision of the pro-
LC–MS–MS. The LOD and LOQ were calculated at a signal-to-             cedure for authentic hair samples. An intraday imprecision
noise ratio of 3 and 10, respectively.                                between extracts was calculated for n = 3. The interday im-
                                                                      precision was calculated by analyzing the pooled hair sample in
                                                                      duplicate on two days (n = 4).

Method Validation                                                     Case samples
                                                                        Sixteen postmortem case samples were tested for diagnostic
  The optimized method was fully validated. The extracts were         purposes. The root–0.5 cm segment was removed from each
reconstituted in 150 µL of mobile phase for the validation            sample. The remaining hair was cut into 3-cm segments. The
work to allow a greater quantity of analyte to be injected onto       segments were subsequently washed according to the proce-
the LC column, ultimately improving the assay sensitivity.            dure described previously. After removing the second
                                                                      dichloromethane wash, the hair samples were left to dry
Linearity                                                             overnight at room temperature. The samples were then incu-
  Linearity was determined over the concentration range 0.2–          bated and extracted using the validated method.
10 ng/mg spiked hair. This sample weight was chosen for the
validation because the hair case samples were analyzed in seg-
ments and the sample weights were generally low.
                                                                      Method Development Results and Discussion
Total extraction recovery
  The percent total extraction recovery was determined for            Initial tuning and ion identification
each analyte at 0.5, 1, and 5 ng/mg. This recovery takes into ac-        Optimum tuning parameters, precursor and product quan-
count a combination of the incubation step, SPE step, and             titation ions are shown in Table I for the drugs tested in the
matrix effect because the reference was unextracted standards.        method. There was only one product ion produced during
                                                                      fragmentation for the amphetamines because of their low
Matrix effect                                                         molecular weights.
   Matrix effect was assessed by comparing the mean peak-area
ratio of product ion/internal standard of extracted blank hair        Optimum LC column and mobile phase combination
samples spiked with standard solution following SPE (A) (n =            The mobile phase system producing the greatest average
6) versus the mean peak-area ratio of product ion/internal            peak area response for each drug was recorded as 100% and the
standard of unextracted standards prepared in mobile phase at         average peak area responses for the other two mobile phase sys-
equivalent concentrations (B) (n = 6). Matrix effect (%) was cal-     tems were calculated as a percent of the optimum mobile
culated for a low standard concentration of 0.5 ng/mg, ac-            phase system response as shown in Table II.
cording to the calculation proposed in a previously published           The Zorbax SB-Phenyl column and mobile phase system B
article (22).                                                         produced the greatest average peak area response for the
                                                                      greatest number of analytes compared to the other two
Accuracy and imprecision                                              columns and mobile phase systems. Although the Zorbax SB-
   Intrabatch accuracy and imprecision were assessed over the         Phenyl column was initially selected for further work, a pres-
linear range for five extracted 10 mg blank hair samples spiked       sure problem was encountered during further method devel-
at low, medium, and high concentrations (0.5, 1, and 5 ng/mg).        opment. The column pressure would not stabilize and exceeded
The intrabatch imprecision was expressed as a percent relative        the LC–MS–MS maximum pressure setting of 5800 psi. The
standard deviation (RSD) calculated using the five individual         manufacturers recommend using such mobile phases as ace-
values obtained on the same day. Accuracy was calculated by di-       tonitrile/water mixtures because the column is moderately
viding the mean measured concentration of five extracts by the        non-polar in nature, so the mobile phase selection was theo-
theoretical spike concentration and was expressed as a per-           retically suitable. However, it was not feasible to use this
centage of the theoretical spike concentration.                       column in practice. The Synergi Hydro RP column and mobile
   Interbatch imprecision was evaluated for five replicate hair       phase system A were therefore selected for further work be-
extracts spiked at these concentrations on five separate days         cause this column and mobile phase combination was stable
(ntotal = 25). It was expressed as a % RSD value, which was cal-      and produced a higher relative average percent peak area re-
culated using all 25 individual values.                               sponse for more analytes compared with the Gemini column

Journal of Analytical Toxicology, Vol. 32, September 2008

and mobile phase system C.                                                   The results obtained for the aqueous alkaline media showed
   As expected, the retention times and elution order of the an-          an increased cocaine and 6-MAM hydrolysis with increasing
alytes were different depending on the LC column and mobile               concentration of aqueous ammonium hydroxide. Both the
phase system used. However, the number of scan events re-                 acidic and alkaline incubation media produced significant hy-
quired for each retention window was comparable for each                  drolysis of 6-MAM. Cocaine showed significant hydrolysis using
system. Furthermore, it was not possible to analyze all 17 an-            25% aqueous ammonium hydroxide in methanol. It was there-
alytes in 1 injection because of the nature of the ion trap MS.           fore decided to compare the recovery of the methanol and
Therefore, the analytes were analyzed in three separate injec-            phosphate buffer incubation media for future work in au-
tions. This was done to minimize the number of scan events                thentic hair samples because both cocaine and 6-MAM were rel-
per time window to achieve maximum sensitivity and to max-                atively stable using these methods. The methanol and phos-
imize the number of points across the peak to obtain more ac-             phate buffer incubation media produced 6-MAM recoveries of
curate and reproducible analyte quantifications. The first in-            88.6% and 102.9%, respectively, and cocaine recoveries were
jection was for cocaine and its metabolites and opiates. The              99.4% and 102.3%, respectively.
second injection was for amphetamines, and the third was for
diazepam and its metabolites. Figures 1–4 show the results of
some positive hair samples which were extracted using the
method used in the validation work (i.e., Clean Screen extrac-            Comparison of SPE methods
tion cartridge and the Synergi Hydro RP column and mobile
phase system A).                                                          SPE clean-up
                                                                            The Clean Screen method produced the cleanest extracts
Stability of cocaine and 6-MAM                                            overall for all drug classes and for hair from all five individuals.
  A stability experiment for cocaine and 6-MAM in various in-
cubation media was carried out with a view to selecting the in-           Total extraction recovery
cubation method that would yield the highest cocaine and 6-                  The Clean Screen method produced higher recoveries for a
MAM recoveries. The extent of cocaine and 6-MAM degradation               greater number of analytes compared to the Bond Elut Certify
was negligible when using the methanol and phosphate buffer               method for spiked hair. All of these were 76% or greater. Two
incubations. This supports the findings of another paper that             analytes of significant difference were morphine and MDMA,
evaluated cocaine and 6-MAM hydrolysis following a methanol               where there was approximately 20% difference in recovery
and phosphate buffer (pH 5.0) incubation (23). In contrast, all           with Clean Screen, producing higher recovery. In contrast,
concentrations of aqueous acidic media resulted in significant            oxazepam had a higher recovery using the Bond Elut Certify
6-MAM hydrolysis, whereas cocaine degradation was negli-                  extraction by approximately 13%. In the U.K., oxazepam is
gible. This result was in agreement with previously published             likely to be present in hair as a result of metabolism rather than
results (24).                                                             oxazepam use; therefore, the Clean Screen method would de-

  Table I. Optimum Tuning Parameters and Precursor and Product ions for Each Analyte

                             Sheath            Auxillary      Capillary      Collision      Precursor         Product            Retention
                               Gas               Gas        Temperature       Energy           Ion              Ion                Time
  Analyte                     (AU)              (AU)            (°C)           (%)           (m/z)             (m/z)               (tR )

  EME                          10                 10           210              29             200            182*, 168             2.8
  Cocaine                      10                 10           300              30             304            182*, 150            19.4
  Cocaethylene                 10                 10           210              30             318            196*, 150            20.8
  Benzoylecgonine              10                 10           300              28             290            168*, 150            10.5
  Morphine                     20                 10           200              33             286            201*, 229            10.3
  6-MAM                        20                 10           210              34             328            211*, 268            15.2
  Codeine                      10                  5           240              34             300            215*, 243            14.7
  Dihydrocodeine               10                  5           240              33             302            245*, 201            13.8
  Oxazepam                     20                 20           300              29             287            269*, 241            19.2
  Temazepam                    20                 20           300              29             301            283*, 255            20.4
  Nordiazepam                  20                 15           300              41             271            243*, 140            20.6
  Diazepam                     20                 20           300              42             285            257*, 222            21.9
  Amphetamine                  15                  5           210              23             136            119*                 10.8
  Methamphetamine              15                  5           210              28             150            119*                 12.1
  MDA                          15                 15           220              21             180            163*                 11.7
  MDMA                         15                 15           280              25             194            163*                 13.2
  MDEA                         15                 15           220              20             208            163*                 13.6

  * Quantitation ion.

                                                                                            Journal of Analytical Toxicology, Vol. 32, September 2008

tect diazepam and other breakdown products that had a higher               flect real cases, the previous recovery experiment was a means
recovery than oxazepam.                                                    of testing the total extraction recovery of the analytes using a
                                                                           particular SPE method in the presence of hair matrix. The
Incubation recovery from authentic hair samples                            experiment was used as a means of determining the optimum
using the methanol and phosphate buffer incubation                         extraction cartridge. It was difficult to obtain sufficient real
  Although extraction recoveries using spiked hair do not re-              samples that contained all 17 analytes. However, the extraction

 Table II. Relative Percent Peak Area Response for the Three Optimum LC Column and Mobile Phase Combinations

                        Gemini C18 + Mobile Phase System C   Synergi Hydro RP + Mobile Phase System A    Zorbax Phenyl + Mobile Phase System B
                        Relative %PAR               tR          Relative %PAR            tR                Relative %PAR              tR
 Analyte                    (n = 3)                (min)            (n = 3)             (min)                  (n = 3)               (min)

 EME                        100.0                   1.8             59.5                 4.9                    62.9                   3.2
 Cocaine                     46.6                  11.5             67.8                19.4                   100.0                  21.7
 Cocaethylene                48.1                  15.9            100.0                20.8                    71.0                  23.8
 Benzoylecgonine             42.4                   5.7            100.0                10.5                    72.3                  11.7
 Morphine                   100.0                   2.3             89.7                10.3                    47.3                   9.5
 6-MAM                      100.0                   4.8             28.2                15.2                    33.5                  16.1
 Codeine                    100.0                   4.1             34.2                14.7                    31.2                  14.3
 Dihydrocodeine             100.0                   3.4             43.2                13.8                    43.3                  13.3
 Oxazepam                    45.1                  23.6            100.0                19.2                    79.9                  19.4
 Temazepam                   37.8                  25.1             59.0                20.4                   100.0                  20.5
 Nordiazepam                  4.0                  25.8            100.0                20.6                    83.4                  20.5
 Diazepam                     4.3                  27.2            100.0                21.9                    49.5                  21.7
 Amphetamine                 85.2                   4.2             68.7                10.8                   100.0                   9.6
 Methamphetamine             31.3                   4.9             71.4                12.1                   100.0                  12.3
 MDA                         89.2                   4.5             33.8                11.7                   100.0                  12.1
 MDMA                        46.2                   5.1             57.5                13.2                   100.0                  14.5
 MDEA                        54.0                   6.2             53.6                13.6                   100.0                  16.7

 Figure 1. Hair sample positive for amphetamine.

Journal of Analytical Toxicology, Vol. 32, September 2008

recoveries of cocaine and its metabolites, amphetamine, and                       quantitative results overall, extracting greater concentrations
opiates were tested for three real cases.                                         of EME, benzoylecgonine, 6-MAM, codeine, and amphetamine.
  With the exception of the cocaine concentration detected in                     Morphine was not detected in sample 3 by either incubation
sample 1, the phosphate buffer extraction produced higher                         method. A possible explanation for this is the low levels of 6-

  Figure 2. Hair sample positive for diazepam (A) and nordiazepam (B).

  Figure 3. Hair sample positive for morphine (A), codeine (B), dihydrocodeine (C), and 6-MAM (D).

                                                                                                       Journal of Analytical Toxicology, Vol. 32, September 2008

 Figure 4. Hair sample positive for EME (A), cocaine (B), cocaethylene (C), and benzoylecgonine (D).

 Table III. Analyte Recovery from Spiked Hair (n = 3)                               Table IV. % Matrix Effect for 0.5 ng/mg Spiked Hair
                                                                                    Extracts (n = 6)
                                       % Recovery (n = 3)
                                                                                                                              % Matrix Effect
 Analyte                 0.5 ng/mg         1 ng/mg           5 ng/mg                     Analyte                                (% RSD)

 EME                     99.2 (6.6)        96.4 (3.7)        97.7 (1.0)                  EME                                    119.2 (5.9)
 Cocaine                 87.2 (8.9)        91.8 (4.9)        97.3 (3.1)                  Cocaine                                 96.4 (13.7)
 Cocaethylene            79.4 (13.5)       97.8 (6.6)        92.5 (2.8)                  Cocaethylene                            82.9 (66.1)
 Benzoylecgonine         95.3 (8.5)       100.7 (4.5)        99.4 (3.2)                  Benzoylecgonine                         85.1 (21.6)
 Morphine                91.2 (2.2)        80.1 (6.7)        92.3 (1.1)                  Morphine                                85.4 (11.1)
 6-MAM                  109.6 (7.7)       104.5 (4.6)        94.3 (2.5)                  6-MAM                                  110.6 (18.9)
 Codeine                 99.6 (2.3)        94.4 (5.4)        99.3 (0.2)                  Codeine                                 92.6 (9.5)
 Dihydrocodeine          86.5 (7.5)        80.9 (3.2)       100.2 (0.7)                  Dihydrocodeine                         102.8 (5.1)
 Oxazepam               108.0 (8.8)        96.1 (3.8)       100.6 (3.3)                  Oxazepam                                67.6 (43.8)
 Temazepam               76.3 (7.2)        96.5 (9.4)        95.7 (6.4)                  Temazepam                               63.5 (65.1)
 Nordiazepam             97.9 (5.6)        94.7 (3.7)        93.2 (2.7)                  Nordiazepam                            130.3 (22.6)
 Diazepam                96.9 (6.0)        98.3 (2.1)        94.7 (0.6)                  Diazepam                                95.5 (13.9)
 Amphetamine            108.1 (17.0)       99.6 (6.0)        91.0 (0.6)                  Amphetamine                             97.9 (3.4)
 Methamphetamine         93.6 (2.2)       106.0 (0.8)        94.2 (2.3)                  Methamphetamine                        135.6 (10.8)
 MDA                     82.8 (7.9)       106.2 (12.5)       97.6 (6.1)                  MDA                                     76.4 (17.2)
 MDMA                    98.3 (12.6)      100.0 (9.7)        98.2 (0.6)                  MDMA                                    90.4 (8.7)
 MDEA                   102.5 (13.2)      104.3 (10.3)       90.8 (1.6)                  MDEA                                    91.1 (11.9)

MAM detected in the samples. The 6-MAM/morphine ratios                            qualitative and quantitative results compared to the methanol
observed in the case samples that were tested for this paper                      incubation. Benzodiazepines were not detected in any of the
ranged from 0.14 to 28.00 (5.47 mean). If the mean 6-                             samples, so it was not possible to compare the two incubations
MAM/morphine ratio was applied to calculate a predicted mor-                      for this drug group.
phine concentration from the 6-MAM concentration found, it                           Analyte recoveries for the methanol incubation were gener-
would be 0.15 ng/mg, which is lower than the LOQ of the                           ally lower for most analytes, as reported in various studies
method.                                                                           (23–25). It has been proposed that the generally higher recov-
  Overall, the phosphate buffer incubation produced better                        eries obtained using a phosphate buffer incubation is a result

Journal of Analytical Toxicology, Vol. 32, September 2008

  Table V. Intraday Accuracy and Precision

                            Intrabatch Precision (% RSD, n = 5)   Interbatch Precision (% RSD, n = 25)             Accuracy (% of Target, n = 5)

  Analyte                  0.5 ng/mg      1 ng/mg       5 ng/mg   0.5 ng/mg     1 ng/mg       5 ng/mg           0.5 ng/mg     1 ng/mg     5 ng/mg

  EME                          5.5         4.9              3.8     18.3          9.9           5.6               106.0        99.0        109.6
  Cocaine                      7.2         5.8              3.0     10.7          7.7           7.1                82.0        96.0         93.6
  Cocaethylene                10.0         5.0              3.7     19.0*        15.8          10.2                96.0       103.0         96.2
  Benzoylecgonine              7.1         6.3              3.6     22.5         12.0           8.8                92.0        92.0         99.6
  Morphine                     8.4         7.5              6.9     16.4         15.5*         11.0                94.0        91.0        100.6
  6-MAM                       11.4         5.2              4.8     20.5         13.6           8.2                90.0        87.0        101.6
  Codeine                      8.9         6.5              2.0     18.7         17.5           9.3                94.0       107.0        108.4
  Dihydrocodeine               7.0         5.3              1.4     18.2         12.9           8.5                98.0       116.0        100.4
  Oxazepam                    11.6         8.9              6.0     16.8         14.2*         10.8*              100.0       106.0        101.6
  Temazepam                   18.7        13.9              9.4     20.2*        18.9          10.6                88.0        94.0        105.8
  Nordiazepam                 14.4         9.6              6.6     24.3*        22.9*         10.6               114.0       111.0         97.6
  Diazepam                     6.6         5.6              2.9     13.3         12.2           9.2                96.0        98.0         97.2
  Amphetamine                  9.2         6.7              7.0     16.9         15.7          12.2               112.0        80.0        104.2
  Methamphetamine              6.0         6.1              4.4     20.7          8.9           8.6               116.0        92.0         88.6
  MDA                          9.1         5.0              7.1     20.8         19.4          19.6                88.0        95.0         99.2
  MDMA                        10.6         6.6              5.9     15.9         10.8          10.0               104.0        89.0        101.6
  MDEA                         6.7         6.6              3.5     15.1         11.0           9.8               114.0        88.0        110.0

  * n = 24.

of the water molecules penetrating the keratinized hair to a                    Table VI. LOD and LOQ Values for Bond Elut Certify
greater extent than an organic solvent (26). Furthermore, the                   and Clean Screen Methods
non-keratinous hair regions could potentially provide diffusion
channels, both into and out of the hair, for small drug                                                  Bond Elut Certify         Clean Screen
molecules in the presence of water (26).                                                                  LOD         LOQ         LOD         LOQ
                                                                                Analyte                 (ng/mg)     (ng/mg)     (ng/mg)     (ng/mg)
   The Clean Screen SPE method produced lower detection                         EME                      0.03         0.11       0.03        0.08
and quantitation limits for the majority of analytes (Table III).               Cocaine                  0.10         0.34       0.04        0.13
The detection limits ranged from 0.02 to 0.09 ng/mg, and the                    Cocaethylene             0.04         0.12       0.04        0.14
quantitation limits ranged was 0.05 to 0.31 ng/mg. This may be                  Benzoylecgonine          0.03         0.11       0.02        0.05
                                                                                Morphine                 0.10         0.32       0.07        0.26
a consequence of the higher recoveries and reduced matrix in-
                                                                                6-MAM                    0.11         0.38       0.09        0.31
terference observed using the Clean Screen method compared
                                                                                Codeine                  0.06         0.23       0.06        0.22
to the Bond Elut Certify method. The Clean Screen method was                    Dihydrocodeine           0.03         0.11       0.05        0.16
therefore validated in further work.                                            Oxazepam                 0.09         0.31       0.02        0.06
                                                                                Temazepam                0.04         0.12       0.08        0.26
                                                                                Nordiazepam              0.08         0.27       0.07        0.23
                                                                                Diazepam                 0.03         0.11       0.06        0.16
Method Validation Results                                                       Amphetamine              0.08         0.26       0.03        0.10
                                                                                Methamphetamine          0.04         0.13       0.04        0.13
Linearity                                                                       MDA                      0.10         0.32       0.02        0.07
   All regression lines had R2 values > 0.99 over the concen-                   MDMA                     0.02         0.08       0.04        0.13
                                                                                MDEA                     0.03         0.09       0.03        0.09
tration range 0.2–10 ng/mg. This was the acceptance criterion
used for the determination of linearity.

Total extraction recovery                                                     given in Table III. All gave high recoveries (> 76%) at all three
  The percent total extraction recovery was calculated as the                 concentrations and % RSD values for these were ≤ 17%.
mean peak-area ratio of product ion/internal standard for the
samples in which the standard solution was added before SPE                   Matrix effect
(n = 3), divided by the mean peak-area ratio of product ion/in-                 The percent matrix effect ranged from 63.5 to 135.6%. Most
ternal standard for the samples in which the standard solution                analytes (12 out of 17) demonstrated ion suppression, some
was added after SPE (n = 3).                                                  being affected to a greater extent than others. Oxazepam and
  The recovery values for all the analytes in spiked hair are                 temazepam product ions were suppressed the most, 36.5%

                                                                                             Journal of Analytical Toxicology, Vol. 32, September 2008

and 32.4%, respectively. With the excep-
                                              Table VII. LC–MS–MS Results for Postmortem Cases
tion of nordiazepam, methamphetamine,
and MDA, all of the other analytes were                                    Weight         LC–MS–MS Results
suppressed or enhanced by ± 20%, which        Hair Sample Number           (mg)           (ng/mg)                   Cause of Death
is the value which is currently regarded
as acceptable for accuracy and impreci-       (1) Root–0.5 cm                2.85         0.91 MOR*
sion experiments in forensic toxicology                                                   0.97 6-MAM
(2). At present, no criteria have been pub-                                               1.38 COD
lished regarding acceptance criteria for                                                  2.93 DHC
matrix effects. ESI is reported to be par-    (1) 0.5–3.5 cm               23.80          2.02 MOR                  Bronchopneumonia due to
ticularly prone to matrix effects com-                                                    1.31 6-MAM                heroin intoxication
pared to other ionization techniques                                                      0.37 COD
such as atmospheric pressure chemical                                                     1.22 DHC
ionization (27). This may account for the                                                 0.10 DZ
high variation in some of the ion sup-                                                    0.37 NDZ
pression data. In addition, hair from six     (2) Root–0.5 cm                2.05         1.67 MOR
individuals was used in the matrix effect                                                 1.96 6-MAM
evaluation where in reality a larger                                                      0.34 DZ
sample group should have been used,                                                       2.26 NDZ                  No information available
consisting of a range of hair colors, hair    (2) 0.5–3.5 cm               25.45          1.23 MOR
from people of different races, and a                                                     0.12 6-MAM
range of dyed and chemically treated hair.                                                0.45 COD
A larger sample group could reduce the                                                    0.13 DHC
average matrix effect or increase it from                                                 0.06 COC
the one quoted in this paper. The results                                                 0.42 BZE
of the % matrix effect evaluation are                                                     1.03 MDMA
given in Table IV.                                                                        0.22 OXAZ
                                                                                          0.10 TMZ
                                                                                          0.52 DZ
Accuracy and imprecision                                                                  0.67 NDZ
   The intrabatch accuracy and impreci-
sion values were acceptable by SOFT           (3) Root–0.5 cm                2.74         NEG for all
guidelines of ± 20% for the three con-                                                    drug groups               Heroin and alcohol intoxication
centrations tested (2). The results are       (3) 0.5–3.5 cm               19.09          0.09 MOR
given in Table V. Intraday accuracy                                                       0.52 6-MAM
ranged from 80 to 116%. Imprecision (%                                                    0.21 COD
RSD) ranged from 1.4 to 18.7% for ana-                                                    0.29 COC
lytes. As expected, lower precision was                                                   0.16 BZE
observed at the lower concentrations.                                                     1.26 MDMA
                                                                                          0.12 DZ
   Similarly, the interbatch imprecision
                                                                                          0.36 NDZ
was generally acceptable within SOFT
guidelines of ± 20% for the three con-        (3) 3.5–6.5 cm               18.60          0.53 6-MAM
centrations tested (2). Interbatch preci-                                                 0.96 COC
sion for nordiazepam was slightly higher                                                  0.45 BZE
at the lower concentrations; however, the                                                 1.60 MDMA                 Heroin and alcohol intoxication
guidelines also state that ± 25–30% may       (4) Root–0.5 cm                3.21         0.58 MOR
be acceptable for some analytes. Impre-                                                   0.77 6-MAM
cision (% RSD) ranged from 5.6 to 24.3%                                                   0.32 DHC
for all analytes. As found in the intra-                                                  1.40 NDZ                  Bronchopneumonia due to
batch imprecision experiment, the %                                                                                 aspiration of gastric contents
RSD was higher at lower concentrations.                                                                             due to morphine intoxication
The % RSD range was higher for inter-         (4) 0.5–3.5 cm               21.12          0.16 MOR
batch imprecision compared to intra-                                                      0.58 6-MAM
batch imprecision, due to instrumental                                                                                     Table continues on next page
and extraction interbatch variation and
also a greater number of data values.
                                              * Abbreviations: MOR, morphine; 6-MAM, 6-monoacetylmorphine; COD, codeine; DHC, dihydrocodeine;
Some of the analyte data shown in Table         COC, cocaine; BZE, benzoylecgonine; COCAETH, cocaethylene; EME, ecgonine methyl ester;
IV for interbatch precision was calculated      AMP, amphetamine; DZ, diazepam; NDZ, nordiazepam; OXAZ, oxazepam; and TMZ, temazepam.
for n = 24. This was due to the removal of

Journal of Analytical Toxicology, Vol. 32, September 2008

                                                                                                             an outlier in the data sets.
  Table VII. LC–MS–MS Results for Postmortem Cases (Continued)

                            Weight      LC–MS–MS Results                                                     LOD and LOQ
  Hair Sample Number         (mg)           (ng/mg)                      Cause of Death                        The LOD and LOQ values for the Clean
                                                                                                             Screen method are provided in Table VI.
                                              0.20 COD*                                                      The LOD values ranged from 0.02 to 0.09
                                              0.09 DHC                                                       ng/mg hair and the LOQ ranged from
                                              0.86 COC                                                       0.05 to 0.31 ng/mg hair.
                                              0.11 BZE
                                              0.19 AMP
                                                                                                             Imprecision of the procedure using
                                              1.42 MDMA
                                              0.04 DZ                                                        authentic hair samples
                                              0.29 NDZ                                                          The intrabatch imprecision for n = 3
                                                                                                             aliquots of the pooled authentic hair
  (5) Root–0.5 cm               1.71          1.39 6-MAM                Heroin intoxication                  sample extracted on the same day re-
  (5) 0.5–3.5 cm               12.62          0.22 MOR                                                       ported here as ng/mg ± SD (% RSD) were
                                              0.49 6-MAM                                                     morphine 0.29 ± 0.01 (4.2%); 6-MAM 4.2
                                              0.32 COD                                                       ± 0.57 (13.7%); cocaine 4.6 ± 0.26 (5.6%);
                                              10.18 COC                                                      benzoylecgonine 0.93 ± 0.01 (1.4%); am-
                                              3.11 BZE                                                       phetamine 3.3 ± 0.47 (14.2%); and di-
                                              0.43 COCAETH                                                   azepam 0.62 ± 0.15 (23.4%).
                                              0.24 EME
                                                                                                                The interbatch imprecision for n = 2
                                              0.11 MDMA
                                              0.17 DZ
                                                                                                             aliquots of the pooled authentic hair
                                              0.52 NDZ                                                       sample extracted on two different days
                                                                                                             (i.e., n = 4), reported here as ng/mg ± SD
  (6) Root–0.5 cm               1.03          NEG for all               Gastrointestinal hemorrhage          (% RSD) were morphine 0.30 ± 0.03
                                              drug groups               due to esophageal ulcers             (9.0%); 6-MAM 3.3 ± 0.84 (25.4%); co-
                                                                        due to chronic alcohol abuse
                                                                                                             caine 4.7 ± 0.40 (8.4%); benzoylecgonine
  (6) 0.5–3.5 cm               18.49          2.32 COD                                                       0.90 ± 0.13 (14.6%); amphetamine 3.6 ±
  (7) Root–0.5 cm               1.49          0.13 COD                  Heroin intoxication
                                                                                                             0.30 (8.3%); and diazepam 0.58 ± 0.06
                                                                                                             (11.1%). The interbatch imprecision for
  (7) 0.5–3.5 cm                8.38          1.10 MOR                                                       diazepam was for n = 3 as a result of a bad
                                              4.55 6-MAM                                                     injection by the instrument.
                                              0.11 COD
  (8) Root–0.5 cm               1.05          NEG for all               Heroin intoxication
                                              drug groups
                                                                                                             Case Sample Results
  (8) 0.5–3.5 cm                9.24          0.58 6-MAM
                                              0.08 COD
                                              0.21 DHC                                                          The case sample results are provided
                                                                                                             in Table VII. The acceptance criteria used
  (9) Root–0.5 cm               3.16          0.71 6-MAM
                                                                                                             for qualifier ratios was ± 20% as recom-
                                              2.05 COC
                                              0.31 BZE                                                       mended by SOFT (2). Because the am-
                                              0.25 COCAETH              Morphine (heroin) intoxication       phetamines have relatively low molecular
                                                                                                             weights and only fragment to produce
  (9) 0.5–3.5 cm               28.55          0.04 MOR                                                       one major product ion, the total ion
                                              1.12 6-MAM
                                                                                                             count to major product ion ratio was used
                                              13.99 COC
                                              4.73 BZE                                                       for identification of amphetamines. At
                                              0.21 EME                                                       least one drug class was detected in every
                                              0.54 COCAETH                                                   sample. Some of the cocaine, dihy-
                                              0.21 AMP                                                       drocodeine, and amphetamine results
                                              0.18 DZ                                                        were greater than the highest point on
                                              0.17 NDZ                                                       the calibration graph. These concentra-
                                                                                                             tions were calculated by extrapolation be-
                                                                              Table continues on next page   cause there was an insufficient amount
                                                                                                             of sample to repeat the analysis. The hair
                                                                                                             samples were reported as positive for a
 * Abbreviations: MOR, morphine; 6-MAM, 6-monoacetylmorphine; COD, codeine; DHC, dihydrocodeine;
   COC, cocaine; BZE, benzoylecgonine; COCAETH, cocaethylene; EME, ecgonine methyl ester;                    particular drug group according to the
   AMP, amphetamine; DZ, diazepam; NDZ, nordiazepam; OXAZ, oxazepam; and TMZ, temazepam.                     SoHT guidelines for opiates, cocaine, and
                                                                                                             amphetamines (10). There are currently

                                                                                                   Journal of Analytical Toxicology, Vol. 32, September 2008

no published guidelines for diazepam and its metabolites in       one drug group. Furthermore, four of these root samples tested
hair.                                                             positive for two drug groups. These root samples were of very
  Amphetamine, benzoylecgonine, cocaethylene, cocaine,            low weight (≤ 3.27 mg) and the high sensitivity of the LC–MS–
codeine, diazepam, dihydrocodeine, morphine, 6-MAM, and           MS instrument has been highlighted by the positive results. A
nordiazepam were detected in some of the root–0.5 cm seg-         greater number of analytes were detected in the segments ex-
ments. Ten of the 16 root samples tested positive for at least    cluding roots compared to the root samples. Also, the ranges
                                                                                       found were much broader than the ranges
                                                                                       obtained for the root segments. This may
 Table VII. LC–MS–MS Results for Postmortem Cases (Continued)                          have been due to greater segment
                                                                                       weights, or alternatively, the individual
                    Weight    LC–MS–MS Results                                         may have used drugs in the past which
 Hair Sample Number  (mg)         (ng/mg)             Cause of Death                   they had stopped using. Oxazepam,
                                                                                       temazepam, ecgonine methyl ester, and
 (10) Root–0.5 cm       3.27      0.60 COC*                                            MDMA were detected in addition to the
                                  0.18 BZE
                                                                                       compounds found in the roots. The drug
                                  0.24 COCAETH
                                  0.31 DZ
                                                                                       concentrations found in this present
                                  0.34 NDZ           Cocaine, heroin, and              study were within the ranges reported in
                                                     diazepam intoxication             another study which used the same phos-
                                                                                       phate buffer incubation (23).
 (10) 0.5–3.5 cm       29.47      0.08 6-MAM                                              The cause of death is also reported for
                                  2.50 COC
                                                                                       each of the cases in Table VII. The hair
                                  1.88 BZE
                                  0.05 COCAETH
                                                                                       analysis provided the pathologist with
                                  0.15 EME                                             extra information on prior drug use his-
                                  0.26 OXAZ                                            tory such as chronic or naïve use. The
                                  0.14 TMZ                                             root–0.5 cm segment was cut from the
                                  0.24 DZ                                              bulk of the hair because this segment
                                  0.64 NDZ                                             could be contaminated with drug-posi-
 (11) 0.5–3.5 cm        8.55      0.49 MDMA          Heroin and alcohol intoxication
                                                                                       tive blood circulating in the body at the
                                                                                       time of death. This segment was analyzed
 (11) 3.5–6.5 cm        8.69      0.02 COD                                             as an indicator of recent drug use. The
 (12) Root–0.5 cm       0.35      14.48 DHC          Sertraline, methadone, and        0.5–3.5 cm segment therefore should not
                                                     morphine overdose                 be contaminated with blood from this
                                                                                       source. The 0.5–3.5 cm segment was
 (12) 0.5–3.5 cm       10.40      0.87 MOR
                                                                                       tested to provide information on drug use
                                  0.12 6-MAM
                                  16.18 DHC
                                                                                       in the 3.5 months prior to death, as-
                                  0.46 MDMA                                            suming the generally accepted growth
                                  3.51 OXAZ                                            rate of 1 cm/month.
                                             0.68 DZ
                                             0.92 NDZ
 (13) Root–0.5 cm               0.21         NEG for all               Heroin intoxication
                                             drugs groups                                                Conclusions
 (13) 0.5–3.5 cm                3.79         0.71 6-MAM
                                                                                                            The developed and validated LC–MS–
 (14) 0.5–1.5 cm                1.25         NEG for all               Cerebro-vascular                  MS method is capable of simultaneously
                                             drug groups               accident (“stroke”)               identifying and quantifying am-
 (15) Root–0.5 cm               1.75         3.54 MOR                                                    phetamines, diazepam and its metabo-
                                             1.31 6-MAM                Heroin intoxication               lites, cocaine and its metabolites, and opi-
                                                                                                         ates from one hair sample of 8–30 mg for
 (15) 0.5–3.5 cm              18.66          1.03 MOR
                                             1.32 6-MAM
                                                                                                         segments excluding roots.
                                             0.31 COD                                                       It is also capable of detecting and quan-
                                             10.72 DHC                                                   tifying these drug groups in root seg-
                                                                                                         ments of low weight (in one case, < 1 mg).
 (16) Root–0.5 cm               2.49         5.59 AMP                  Amphetamine intoxication          The method proved to be sufficiently sen-
 (16) 0.5–3.5 cm              18.21          7.29 AMP                                                    sitive and specific for the analysis of 17
                                                                                                         drugs and metabolites in postmortem
 * Abbreviations: MOR, morphine; 6-MAM, 6-monoacetylmorphine; COD, codeine; DHC, dihydrocodeine;
   COC, cocaine; BZE, benzoylecgonine; COCAETH, cocaethylene; EME, ecgonine methyl ester;
                                                                                                         hair samples. There is scope for the in-
   AMP, amphetamine; DZ, diazepam; NDZ, nordiazepam; OXAZ, oxazepam; and TMZ, temazepam.                 clusion of other target drugs and metabo-
                                                                                                         lites into the method. Maximum infor-

Journal of Analytical Toxicology, Vol. 32, September 2008

mation is obtained from one hair sample which is extremely                      S. Carman. Qualitative screening for drugs of abuse in hair using
useful when the sample weight is limited. It is currently not                   GC–MS. J. Anal. Toxicol. 25: 203–208 (2001).
                                                                          12.   R. Kronstrand, I. Nyström, J. Strandberg, and H. Druid. Screening
possible to relate the quantity of drug detected in hair with the               for drugs of abuse in hair with ion spray LC–MS–MS. Forensic Sci.
amount of drug ingested or the frequency of drug use; a qual-                   Int. 145: 183–190 (2004).
itative screening method may provide all the required infor-              13.   S. Gentili, M. Cornetta, and T. Macchia. Rapid screening proce-
mation for as accurate an interpretation as possible at this                    dure based on headspace solid-phase microextraction and gas
time. However, the quantitative method can be used to deter-                    chromatography–mass spectrometry for the detection of many
                                                                                recreational drugs in hair. J. Chromatogr. B 801: 289–296 (2004).
mine if an individual is a chronic drug user. It can be used to           14.                          ˇi´      ˇi´                c
                                                                                L. Skender, V. Karac c, I. Brc c, and A. Bagari´ . Quantitative de-
compare the segments within an individual’s hair to give an in-                 termination of amphetamines, cocaine, and opiates in human
dication of drug use over a period of time. However, care must                  hair by gas chromatography–mass spectrometry. Forensic Sci.
be taken with this for longer segments where some leaching of                   Int. 125: 120–126 (2002).
drug from the hair might have occurred through hygienic                   15.   C. Girod and C. Staub. Analysis of drugs of abuse in hair by
                                                                                automated solid-phase extraction, GC–EI-MS, and GC ion trap–
procedures and environmental exposure. The quantitative                         CI-MS. Forensic Sci. Int. 107: 261–271 (2000).
method is also useful where further analysis is required on               16.   R. Gottardo, A. Fanigliulo, F. Bortolotti, G. De Paoli, J.P. Pascali,
drug users’ hair, and where possible, to analyze known users’                   and F. Tagliaro. Broad spectrum toxicological analysis of hair
hair to establish an in-house database of results for this par-                 based on capillary zone electrophoresis time-of-flight mass spec-
ticular method and hair types.                                                  trometry. J. Chromatogr. A 1159: 190–197 (2007).
                                                                          17.   R. Gottardo, F. Bortolotti, G. De Paoli, J.P. Pascali, I. Mikšík, and
                                                                                F. Tagliaro. Hair analysis for illicit drugs by using capillary zone
                                                                                electrophoresis–electrospray ionization-ion trap mass spectrom-
                                                                                etry. J. Chromatogr. A 1159: 185–189 (2007).
                                                                          18.   M.R. Moeller, P. Fey, and R. Wennig. Simultaneous determination
References                                                                      of drugs of abuse (opiates, cocaine, and amphetamine) in human
                                                                                hair by GC–MS and its application to a methadone treatment
 1. A.M. Baumgartner, P.F. Jones, W.A. Baumgartner, and C.T. Black.             program. Forensic Sci. Int. 63: 185–206 (1993).
    Radioimmunoassay of hair for determining opiate-abuse histories.      19.   M. Klys, S. Rojek, J. Kulikowska, E. Bo´ ek, and M. ´
                                                                                                                             z            Scislowski.
    J. Nucl. Med. 20: 748–752 (1979).                                           Usefulness of multiparameter opiates-amphetamines-cocainics
 2. SOFT/AAFS Forensic Laboratory Guidelines 2006. http://www.soft              analysis in hair of drug users for the evaluation of an abuse pro- Accessed                        file by means of LC–APCI-MS–MS. J. Chromatogr. B 854: 299–307
    July 2007.                                                                  (2007).
 3. K. Lachenmeier, F. Musshoff, and B. Madea. Determination of opi-      20.   R. Cordero and S. Paterson. Simultaneous quantification of
    ates and cocaine in hair using automated enzyme immunoassay                 opiates, amphetamines, cocaine, and metabolites and diazepam
    screening methodologies followed by gas chromatographic–mass                and metabolite in a single hair sample using GC–MS. J. Chro-
    spectrometric (GC–MS) confirmation. Forensic Sci. Int. 159: 189–            matogr. B 850: 423–431 (2007).
    199 (2006).                                                           21.   F.M. Wylie, H. Torrance, R.A. Anderson, and J.S. Oliver. Drugs in
 4. W.E. Brewer, R.C. Galipo, K.W. Sellers, and S.L. Morgan. Analysis           oral fluid. Part I. Validation of an analytical procedure for licit and
    of cocaine, benzoylecgonine, codeine, and morphine in hair by               illicit drugs in oral fluid. Forensic Sci. Int. 150: 191–198 (2005).
    supercritical fluid extraction with carbon dioxide modified with      22.   B.K. Matuszewski, M.L. Constanzer, and C.M. Chavez-Eng. Strate-
    methanol. Anal. Chem. 73: 2371–2376 (2001).                                 gies for the assessment of matrix-effect in quantitative bioana-
 5. K.B. Scheidweiler and M.A. Huestis. Simultaneous quantifica-                lytical methods based on HPLC–MS–MS. Anal. Chem. 75: 3019–
    tion of opiates, cocaine and metabolites in hair by LC–APCI-                3030 (2003).
    MS–MS. Anal. Chem. 76: 4358–4363 (2004).                              23.   F.S. Romolo, M.C. Rotolo, I. Palmi, R. Pacifici, and A. Lopez . Op-
 6. B.A. Goldberger, Y.H. Caplan, T. Maguire, and E.J. Cone. Testing            timized conditions for simultaneous determination of opiates,
    human hair for drugs of abuse, III. Identification of heroin and 6-         cocaine and benzoylecgonine in hair samples by GC–MS.
    acetylmorphine as indicators of heroin use. J. Anal. Toxicol. 15:           Forensic Sci. Int. 138: 17–26 (2003).
    226–231 (1991).                                                       24.   A. Polettini, C. Stramesi, C. Vignali, and M. Montagna. Determi-
 7. E. Cognard, S. Rudaz, S. Bouchonnet, and C. Staub. Analysis of              nation of opiates in hair: effects of extraction methods on re-
    cocaine and three of its metabolites in hair by gas chromatog-              covery and on stability of analytes. Forensic Sci. Int. 84: 271–279
    raphy–mass spectrometry using ion-trap detection for CI–MS–                 (1997).
    MS. J. Chromatogr. B 826: 17–25 (2005).                               25.   V. Cirimele, P. Kintz, and P. Mangin. Comparison of different ex-
 8. E. Han, W. Yang, J. Lee, Y. Park, E. Kim, M. Lim, and H. Chung.             traction procedures for drugs in hair of drug addicts. Biomed.
    Correlation of methamphetamine results and concentrations be-               Chromatogr. 10: 179–182 (1996).
    tween head, axillary, and pubic hair. Forensic Sci. Int. 147: 21–     26.   G. Skopp, L. Potsch, and R. Aderjan. Experimental investigations
    24 (2005).                                                                  on hair fibres as diffusion bridges and opiates as solutes in solu-
 9. E.I. Miller, F.M. Wylie, and J.S. Oliver. Detection of benzodi-             tion. J. Forensic Sci. 41: 117–120 (1996).
    azepines in hair by LC–ESI-MS–MS. J. Anal. Toxicol. 30(7): 441–       27.   Applications of LC–MS in Toxicology, A. Polettini, Ed. Pharma-
    448 (2006).                                                                 ceutical Press, London, U.K., 2006.
10. Society of Hair Testing. Recommendations for hair testing in
    forensic cases. Forensic Sci. Int. 145: 83–84 (2004).                                Manuscript received December 17, 2007;
11. S. Paterson, N. McLachlan-Troup, R. Cordero, M. Dohnal, and                            revision received March 20, 2008.


To top