You are on page 1of 7

Clinical Biochemistry 46 (2013) 11181124

Contents lists available at SciVerse ScienceDirect

Clinical Biochemistry
journal homepage: www.elsevier.com/locate/clinbiochem

Simultaneous detection of 19 drugs of abuse on dried urine spot by liquid


chromatographytandem mass spectrometry
Yungkang Lee a,, Keane K.Y. Lai a, b, c, d, S.M. Hossein Sadrzadeh a, 1
a

Department of Pathology and Laboratory Medicine, Cedars-Sinai Medical Center, Los Angeles, CA 90048, United States
Department of Pathology, University of Southern California, Los Angeles, CA 90033, United States
Southern California Research Center for ALPD and Cirrhosis, University of Southern California, Los Angeles, CA 90033, United States
d
USC Norris Comprehensive Cancer Center, University of Southern California, Los Angeles, CA 90033, United States
b
c

a r t i c l e

i n f o

Article history:
Received 5 January 2013
Received in revised form 22 March 2013
Accepted 27 March 2013
Available online 11 April 2013
Keywords:
LCMS/MS
Extraction
EMIT
Reex GCMS
SAMHSA
iMethod
Cross-reactivity
Cost analysis

a b s t r a c t
Background: The lack of specicity of immunoassays for drugs of abuse testing (DAT), and concerns over
its cost in conjunction with reex conrmatory tests prompted us to investigate the combinatorial use of
dried urine spot (DUS) and LCMS/MS as an alternative.
Methods: The method development and validation were performed in accordance with the guidelines
published by FDA and CLSI.
Results: In this study we established and validated the precision, accuracy, and linearity of our DUSLCMS/
MS method, and assessed the recovery, interference, and carryover as well. The linearity check for all 19 analytes
demonstrated slopes between 0.94 and 1.04, and R2 always greater than 0.99. Between-batch CV for QC at 4 difference levels ranged from 1.1% to 10%, where CV of LLOQ ranged from 1.2% to 12.8% and CV of ULOQ ranged from
0.8% to 5.1%. A concordance study with patient specimens between our method and GCMS demonstrated 80.8%
to 100% agreement. Stability of DUS specimens was assessed up to 30 days and the measured concentrations
ranged from 94% to 114% of the 100 ng/mL urine calibrator used for this assessment.
Conclusions: We established and validated a DUSLCMS/MS method for DAT that conforms to the guidelines dictated by FDA, CLSI, and SAMHSA. While our method with high sensitivity and specicity provides an
alternative diagnostic utility to EMIT immunoassays, it also offers superior solutions in specimen transportation,
preservation, and storage. The benets of our method are apparent in reducing turnaround time and test costs
that result in better patient care.
2013 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.

Introduction
Liquid chromatographytandem mass spectrometry (LCMS/MS)
has become an essential part of routine clinical laboratory testing.
Although the application of mass spectrometry in conjunction with
gas chromatography (GC) for clinical diagnosis dates back to several
decades, it is not until the past ve to ten years that LCMS/MS has
become a premier choice for the analysis of a wide variety of analytes
[1]. Unlike GCMS, the detection of analytes by LCMS/MS does not
Abbreviations: DAT, drug of abuse testing; DUS, dried urine spot; LCMS/MS, liquid
chromatographytandem mass spectrometry; FDA, Food and Drug Administration; CLSI,
Clinical and Laboratory Standards Institute; CV, coefcient of variation; QC, quality
control; LLOQ, lower limit of quantication; ULOQ, upper limit of quantication; GCMS,
gas chromatographymass spectrometry; SAMHSA, Substance Abuse and Mental Health
Services Administration; EMIT, enzyme multiplied immunoassay technique.
Corresponding author at: Department of Pathology and Laboratory Medicine, CedarsSinai Medical Center, 8700 Beverly Boulevard, Los Angeles, CA 90048, United States.
E-mail addresses: hugolee@alumni.usc.edu (Y. Lee), hossein.sadrzadeh@cls.ab.ca
(S.M.H. Sadrzadeh).
1
Clinical Biochemistry Section, Calgary Laboratory Services, University of Calgary,
Calgary, Alberta, Canada.
1
Clinical Biochemistry Section, Calgary Laboratory Services, University of Calgary,
Calgary, Alberta, Canada.

require them to be volatile at elevated temperature, nor require laborious derivatization. In addition, LCMS/MS preserves structural integrity of thermolabile analytes better than GCMS analysis [2].
The innovation and renement of atmospheric pressure ionization
(API) techniques, most commonly available as electrospray ionization
(ESI) and atmospheric pressure chemical ionization (APCI), have dramatically improved analyte ionization efciency while better preserving
the molecular structure and integrity of small molecule chemicals to
macromolecular proteins and lipids. For such reasons these soft ionization techniques coupled with liquid chromatography have emerged as
the preferred ionization methods employed in LCMS/MS quantitative
analysis [36].
Dried blood spot (DBS) has been used in clinical laboratories for
many years, especially to detect inborn errors of metabolism. DBS offers
a very convenient way of collecting specimens on lter papers. DBS
eliminates the need for phlebotomists and oftentimes requires only
one or two drops of blood to be used for analysis. Most importantly,
DBS can be easily mailed to clinical laboratories for analysis, as specimens on DBS are less prone to degradation and hydrolysis, thus offering
an advantage during specimen storage and transport. Although DBS has
been used for many years, dried urine spot (DUS) has not been introduced into clinical laboratories for routine use [79].

0009-9120/$ see front matter 2013 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.clinbiochem.2013.03.027

Y. Lee et al. / Clinical Biochemistry 46 (2013) 11181124

Here we report the successful development and validation of a simple DUSLCMS/MS method that simultaneously detects and quanties
a panel of 19 drugs of abuse that encompasses nine natural and semisynthetic opiates, three synthetic opiates, two sedative-hypnotics, and
ve stimulants. While this method provides excellent sensitivity and
specicity, it also offers advantages in specimen transport, preservation,
and storage. This method drastically reduces turn-around time and test
costs which leads to better patient care.
Materials and methods
Calibrators and controls
Certied reference materials acquired from Cerilliant (Round Rock,
TX) were used to prepare calibrators and deuterated materials, the latter of which were spiked in extraction buffer at 200 ng/mL as internal
standards. Quality control samples were prepared by spiking pooled
drug-free urine with certied reference materials at the concentrations
indicated in the Results section. The detailed protocol for preparing calibrator and QC mixtures is available upon request.
Chromatography
Chromatography was performed on a Shimadzu Prominence UFLC
XR system (Columbia, MD) with Phenomenex Luna PFP 2.1 50 mm
column (Torrance, CA). Mobile phase A consisted of 0.1% formic acid
and 1 mM ammonium formate (Sigma-Aldrich, St. Louis, MO). Mobile
phase B was LCMS grade acetonitrile (Sigma-Aldrich, St. Louis, MO)
with 0.1% formic acid. A gradient program with a ow rate of 0.5 mL/
min as indicated in Supplemental Table 1 was used for analyte separation. LC ow was diverted to waste for the rst 0.8 min to prevent
source contamination at the mass spectrometer.
Mass spectrometry
Mass spectrometric detection was performed using an AB Sciex
3200 QTRAP triple-quadrupole-ion trap hybrid instrument (Foster
City, CA). Electrospray ionization was performed in positive mode
with an ionization voltage of 2000 V and a TEM temperature of
500 C. Transition for each analyte was chosen by tuning the instrument
with 1 g/mL solution of each analyte in methanol. The declustering potential, entrance potential, cell entrance potential, collision energy, and
cell exit potential were optimized individually for each analyte (Supplemental Table 2). Analyst software by AB Sciex (Foster City, CA) was used
for MS/MS data acquisition and post-acquisition processing. Scheduled
MRM (sMRM) method was used to monitor the analyte ion traces, and
was set for the detection windows of 60 s and the target scan times of
0.4 s. Unit resolution was set for both Q1 and Q3. LCMS/MS chromatogram was analyzed to ensure that all MS/MS scans were at least 50 ms
in dwell time.
Sample preparation
Whatman 903 Protein Saver Cards (GE Healthcare, Piscataway NJ)
were chosen for testing where 15 L was center-spotted on each sample circle. Cards were air dried in a laminar ow hood for 2 h before
being stored away in air-tight bags with desiccant at ambient temperature without exposure to light. Analytes on DUS cards were extracted by
rst punching two 6 mm circles out of two sample circles per card,
then mixing the two circles with 450 L of extraction buffer (water:
methanol:acetonitrile = 1:4:4) with deuterated internal standards at
200 ng/mL. After vortexing for 10 min, samples were cleared by centrifugation at 14,000 g for 5 min. Extractants were transferred to clean
tubes followed by evaporating the organic phase using heated nitrogen
gas. The remaining aqueous extractants were mixed with 50 L of 5%
acetonitrile followed by analysis of 20 L by LCMS/MS.

1119

Method validation
Matrix effects were assessed by injecting the DUS extractants of
drug-free urine and of 5% mobile phase B with a constant post-column
infusion of a neat solution containing 19 drugs at 1 g/mL. Extracted
ion chromatograms from both samples were examined against each
other for augmentation and suppression of ion counts at the retention
time of each analyte.
Calibration curves were constructed by measuring calibrators in duplicate at a minimum of six concentrations for each analyte, by spiking
drug-free urine with certied reference materials. Three calibrations
were assessed each day for each analyte, for three consecutive days.
MQ III of the Analyst software was chosen for quantication of analyte
peaks with the following parameters: minimum peak height, 1000; RT
window, 120 s; smoothing width, 3; noise percent, 50; base sub window, 1 min; and peak splitting factor, 2. Regression analysis of the
peak areas and the nominal concentrations was performed in linear
mode with 1/x weighting. Successful calibrations were contingent
upon the following criteria: (1) at least two thirds of the calibrators
had to have inaccuracy and imprecision less than 15%, except LLOQ for
which both had to be less than 20%, (2) the response (the ratio of the
ion counts of the analyte to that of its internal standard) at LLOQ had
to be at least ve times that of the blank samples, and (3) the regression
coefcient R 2 of the calibrations had to be greater than 0.99 [10,11].
To assess the assay precision, both the positive and the negative
quality controls were run six times within each batch and the corresponding coefcient of variation (CV) was calculated to determine
within- and between-batch precision at different concentrations. A
chromatography validation kit consisting of Luna PFP columns from
three different lots was used for this method validation.
Statistics
Regression analyses on measured concentrations of drugs and
their nominal concentrations were performed in linear mode by
GraphPad Prism. Calculations for mean, standard deviation, CV, and
bias were performed by Microsoft Excel.
Results
In order to make this assay most sensitive, ion transitions and
compound parameters including declustering potential, entrance
potential, cell entrance potential, collision energy, and cell exit potential were optimized for each drug and its deuterated internal
standard by manual tuning as shown in Supplemental Table 2 and
Supplemental Fig. 1. Optimal source parameters including curtain
gas ow, source temperature, ion source gas 1, ion source gas 2,
and ion spray were also chosen for the highest total ion counts for
19 drugs (data not shown).
The liquid chromatography was optimized to ensure suitable retention and adequate separation on 19 analytes (Fig. 1A). Codeine and
hydrocodone have the same nominal molecular weight of 299, and
morphine and hydromorphone have the same nominal molecular
weight of 285. In order to eliminate cross-talk from their isobaric counterparts, baseline separations for codeine and hydrocodone (Fig. 1B)
and for morphine and hydromorphone were successfully achieved.
Scheduled MRM function was selected to monitor the analyte ion traces
with detection window of 120 s and target scan time of 0.4 s. During MS
parameter optimization we realized that consistent scans could only be
achieved with an MS dwell time of at least 50 ms. Therefore each MS/
MS scan in this method was at least 50 ms in order to ensure that chromatographic peak intensity was accurately recorded so that peaks could
be properly constructed for quantication while assay sensitivity was
not compromised (Supplemental Fig. 2).
Our chromatography study on interferences by constant postcolumn infusion of 19 analytes revealed strong ion suppression

1120

Y. Lee et al. / Clinical Biochemistry 46 (2013) 11181124

Fig. 1. A. Representative LCMS/MS chromatogram of 19 drugs of abuse. Dried urine spots were prepared with urine samples containing 19 drugs of abuse at 1 g/mL. Extraction
and analysis were done as described in the Materials and Methods section. THCA: 11-nor-9-carboxy-9-THC. B. Baseline separation of isobaric codeine and hydrocodone (molecular
weight 299) at their respective retention times of 2.6 and 3.9 min.

between 0.3 and 0.6 min as shown in Fig. 2A. This early elution was
diverted to waste in our LCMS/MS method to prevent source contamination. Modest ion suppression between 2.5 and 2.8 min was
also identied. However analyses on amphetamine and codeine in
this chromatographic region did not suffer from this phenomenon.

Chromatographic carryover was assessed by running drug-free


urine samples immediately after running urine calibrators at
20,000 ng/mL, which is twenty times higher than ULOQ of all 19
analytes (1000 ng/mL). There was no appreciable carryover observed
for amphetamine as shown in Fig. 2B.

Fig. 2. A. Chromatography study on interferences by constant post-column infusion of 19 analytes. Chromatography on DUS extractant of drug-free urine (top panel) and on initial mobile phase
(bottom panel) was performed with post-column infusion of 19 analytes at 1 g/mL. Strong ion suppression was observed between 0.3 and 0.6 min where the early elution was diverted to
waste to prevent source contamination, and modest ion suppression was observed between 2.5 and 2.8 min whereas analyses of amphetamine and codeine were not affected by this phenomenon. B. Chromatographic carryover of amphetamine analysis. Carryover was assessed by examining the extracted ion chromatogram of a blank DUS sample (bottom panel) immediately after
running a DUS sample at 20 g/mL, or 20 times greater than ULOQ (top panel). Extraction and analysis were done as described in the Materials and Methods section.

Y. Lee et al. / Clinical Biochemistry 46 (2013) 11181124

1121

1122

Y. Lee et al. / Clinical Biochemistry 46 (2013) 11181124

Fig. 3. Linearity for amphetamine. Dried urine spots were prepared with urine calibrators at 0, 10, 25, 50, 100, 250, 500, and 1000 ng/mL. Extraction and analysis were done as described in the Materials and Methods section. Correlation coefcients of the regression analyses of their peak areas and nominal concentrations were always greater than 0.99.
Please refer to Supplemental Fig. 3 for the rest of the 19 analytes.

Linearity was established for all 19 analytes by measuring six replicates each day for each calibrator at 10, 25, 50, 100, 250, 500, and
1000 ng/mL for three consecutive days. Fig. 3 and Supplemental
Fig. 3 show the regression analyses of all 19 drugs with their peak
areas versus their nominal concentrations within a batch. As shown
in Table 1, the correlation for urine calibrators of all 19 analytes between their measured concentrations and their nominal concentrations was high with slopes between 0.94 and 1.04, and R 2 always
greater than 0.99.
Extraction recovery was assessed by comparing triplicate DUS extractions from a combination of two 6 mm (diameter) DUS punches
of urine calibrators at 10 (except analytes with LLOQ of 25 ng/mL),
25, 50, 100, 250, 500, and 1000 ng/mL, to 7.5 L of respective urine calibrators extracted by the same extraction buffer. Since 15 L of urine
calibrators was spotted on each 12 mm (diameter) sample circle, the
equivalent of the drug amount from two 6 mm punches was deduced
by: 15 L (6 mm/12 mm) 2 2 punches = 7.5 L. The extraction recovery and the linearity were assessed within each batch and between
batches. The within-batch extraction recovery is shown in Table 1. It
was noted that several analytes like diazepam, oxazepam, fentanyl,
methadone, propoxyphene, and norpropoxyphene exhibited extraction
recovery greater than 100% with 6 mm punches, and yet demonstrated
high linearity, precision, and accuracy. Nevertheless the recovery of
these analytes reverted to the close proximity to 100% when 12 mm
punches were used for extraction and analyses (data not shown).
Quality control samples were made by spiking drug-free urine with
reference materials. Four levels of QC samples negative, low, medium,
and high were assessed for each analyte where their concentrations
were set at 75% of cutoff, 125% of cutoff, 50th percentile between LLOQ
and ULOQ, and 85th percentile between LLOQ and ULOQ, respectively, except for 6-MAM where the measurement of its negative QC was not possible because it would be below the LLOQ for this analyte. Performance of
QC samples was examined with six replicates at each level, each day, for
three consecutive days. As shown in Table 2, CV for all QC samples at
four different levels ranged from 1.1% to 10%, where CV of LLOQ ranged
from 1.2% to 12.8% and CV of ULOQ ranged from 0.8% to 5.1%.
We also compared the results generated by our DUSLCMS/MS
method to that of GCMS performed at a commercial laboratory,

using randomly selected patient specimens. By using the cutoffs listed


in Table 1, we determined that agreement ranged from 80.8% to 100%
between these two methods for all 19 analytes as shown in Table 3.
Stability of DUS specimens was assessed by storing the specimens
at room temperature without exposure to air and light for up to
30 days. Supplemental Table 3 shows that the stability of 19 spotted
analytes at 100 ng/mL ranged between 94% and 114%. No drift or
erratic pattern was observed for any of these analytes during this period and the variances in analyte concentration along the test course
were well within the allowed imprecision of 15%.
Discussion
We have developed and validated an LCMS/MS assay using dried
urine spot that simultaneously detects and quanties 19 drugs of
abuse and metabolites. This method meets or exceeds the cutoff requirements set forth in the Mandatory Guidelines for Federal Workplace Drug Testing Programs published by the Substance Abuse and
Mental Health Services Administration, of the Department of Health
and Human Services [12,13]. Automated immunoassay systems are
fast and sensitive enough for screening purposes in emergent and primary care settings. However, due to their lack of specicity, immunoassays cannot differentiate particular drugs within the same drug
classes [2,14]. In addition, cross-reactivity in immunoassays toward
metabolites and structurally similar compounds (e.g., pseudoephedrine vs. amphetamine) often complicates the interpretation of results,
and may yield false positive results.
One of the challenges in using dried specimens like DUS and DBS is
the dilution of samples following the extraction process. Although
7.5 L of the urine specimens spotted on the DUS cards was diluted
to 100 L after the sample extractions, we demonstrated that high
sensitivity was achieved for all analytes, with LLOQ at 10 ng/mL or
25 ng/mL, which is in line with the sensitivity reported previously
from non-DUS methods with similar LCMS/MS congurations
[4,6,15,16]. A previous study on the recovery with DUS cards reported
that 6 mm punches could soak up 16.5 L of urine (compared with
7.5 L of urine observed in our study) [17]. This increased capacity
might be attributable to their use of a different DUS card and/or the

Y. Lee et al. / Clinical Biochemistry 46 (2013) 11181124

1123

Table 1
Assay quantication limits, cutoff, within-batch assay extraction recovery, and linearity. LLOQ, ULOQ, and cutoff are listed in ng/mL. Extraction recovery was assessed by comparing
triplicate extractions from urine calibrators at 10 (except analytes with LLOQ of 25 ng/mL), 25, 50, 100, 250, 500, and 1000 ng/mL, with 7.5 L of the urine calibrators extracted by
the same extraction buffer without spotting. Within-batch linearity statistics was summarized with the results from three individual calibrations, each of which comprised duplicate calibrators (n = 6) at 10, 25, 50, 100, 250, 500, and 1000 ng/mL, except analytes with LLOQ of 25 ng/mL.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19

Analyte

LLOQ

ULOQ

Cutoff

6-MAM
Codeine
Hydrocodone
Hydromorphone
Morphine
Oxycodone
Oxymorphone
Amphetamine
MDA
MDEA
Methamphetamine
MDMA
Diazepam
Oxazepam
Benzoylecgonine
Fentanyl
Methadone
Propoxyphene
Norpropoxyphene

10
10
10
10
25
25
25
10
25
10
10
10
10
10
10
10
10
10
10

1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000

10
100
100
100
100
100
100
100
100
100
100
100
50
50
50
50
50
100
100

% Recovery

Slope

Intercept

R2

Mm

Max

Mean

SD

Mean

SD

Set 1

Set 2

Set 3

84.4
78.1
88.7
89.5
82.5
88.7
86.5
73.3
94.8
93.3
77.4
85.4
193.8
174.2
71.8
108.1
130.3
181.7
111.2

112.7
104.8
98.2
109.2
102.0
98.8
99.0
85.5
106.9
97.2
84.6
96.6
264.2
210.9
87.5
146.5
169.6
217.4
174.8

0.97
0.99
1.04
0.99
0.99
0.98
1.00
1.00
1.01
0.98
1.04
1.01
0.99
0.99
0.97
0.98
0.94
1.01
1.00

0.01
0.01
0.01
0.00
0.03
0.00
0.01
0.00
0.01
0.00
0.01
0.00
0.02
0.01
0.02
0.02
0.02
0.01
0.01

7.25
1.02
8.31
3.46
6.65
6.21
0.51
0.06
2.72
5.41
11.19
1.60
3.47
2.35
8.49
1.74
8.68
1.48
0.72

1.60
1.40
3.97
1.41
8.81
2.06
2.09
0.66
1.54
0.70
2.51
1.44
2.78
1.83
3.39
3.03
1.88
1.87
4.01

0.9977
0.9983
0.9955
0.9987
0.9945
0.9974
0.9988
0.9995
0.9972
0.9989
0.9953
0.9940
0.9985
0.9998
0.9931
0.9918
0.9931
0.9973
0.9970

0.9978
0.9976
0.9982
0.9995
0.9975
0.9983
0.9986
0.9994
0.9988
0.9992
0.9978
0.9964
0.9976
0.9966
0.9981
0.9971
0.9985
0.9970
0.9969

0.9978
0.9982
0.9984
0.9992
0.9981
0.9965
0.9994
0.9999
0.9993
0.9995
0.9951
0.9988
0.9992
0.9992
0.9972
0.9976
0.9970
0.9994
0.9982

dipping method employed in that study. Nonetheless, we demonstrated that DUS spotting with an equivalent of 7.5 L urine specimen
is sufcient to provide the necessary sensitivity even for drug-ofabuse conrmatory purposes.
For several decades, GCMS technology has been used as the reference method to measure and conrm the presence of various
drugs of abuse in clinical laboratories. However, due to the limitations
that have been discussed previously [2], LCMS is becoming the
method of choice in modern clinical laboratories. As expected, overall
our LCMS/MS method correlated highly with GCMS (Table 3). The
percent agreement was over 90% for the majority of these 19 analytes.
Minor discrepancies might be due to the following reasons. First, hydrolysis of drug-glucuronide conjugates in urine specimens, not DUS
specimens, might have articially increased the concentrations that

were measured by the dilute-and-shoot GCMS method. Therefore,


DUS is very advantageous when there is a need to preserve the drug
metabolites to identify exactly what drugs are taken or when drugs
are taken, such as for toxicological and forensic purposes. Second,
these discrepancies might have arisen randomly in situations where
drug levels were near the assay cutoffs. Moreover, by employing our
extraction technique, DUS can also be used as the specimen of choice
for GC/MS analysis/conrmation of drugs of abuse.
While the upfront acquisition cost for an LCMS/MS system may be
a critical factor for community and mid-size laboratories to adopt
this technology, it is important to note that on the cost per test basis,
LCMS/MS is very competitive compared with the EMIT immunoassays.
Based on previous studies, drugs of abuse testing costs approximately
$9.4 per test by EMIT and $2.5 per test by LCMS/MS [18,19]. Integrating

Table 2
Between-batch QC validation. Assay precision with QC samples at four levels negative,
low, medium, and high was validated with six replicates at each level, each day, for
three consecutive days (n = 18), where the concentrations of negative, low, medium,
and high samples were made of 75% of LLOQ, 125% of LLOQ, 50th percentile between
LLOQ and ULOQ, and 85th percentile between LLOQ and ULOQ, respectively.

Table 3
Comparison of results by our DUSLCMS/MS method with those by external GCMS. Total
number of samples compared included random numbers of positive and negative samples,
albeit in similar numbers. For a sample to be called positive on methamphetamine,
amphetamine must be detected at at least 100 ng/mL in addition to methamphetamine.

Precision (CV)

6-MAM
Codeine
Hydrocodone
Hydromorphone
Morphine
Oxycodone
Oxymorphone
Amphetamine
MDA
MDEA
Methamphetamine
MDMA
Diazepam
Oxazepam
Benzoylecgonine
Fentanyl
Methadone
Propoxyphene
Norpropoxyphene

QC
Negative

Low

Med

High

ND
3.2
6.4
2.7
4.4
5.2
5.1
8.5
8.1
3.1
2.9
8.3
2.8
7.1
3.4
3.7
3.3
3.6
3.8

6.3
4.7
6.0
3.1
3.7
2.9
3.7
3.4
4.1
1.4
3.7
6.7
1.4
4.7
1.3
1.5
3.1
6.6
2.4

2.7
3.8
3.1
2.3
3.7
2.8
2.0
2.7
4.0
1.1
3.1
5.6
2.9
3.0
2.8
1.4
3.3
10.0
2.1

3.7
4.2
4.8
2.3
3.8
3.7
2.7
2.8
4.8
2.7
4.7
6.3
2.5
2.0
3.0
1.8
1.4
2.2
4.7

LLOQ

ULOQ

12.8
10.1
3.7
8.5
7.1
4.5
4.8
2.5
8.1
1.2
1.4
3.0
3.0
6.8
5.8
3.4
2.8
6.3
12.1

1.2
1.8
2.3
1.2
4.5
0.8
1.6
1.6
2.9
0.8
2.1
5.1
3.2
3.5
4.1
4.7
2.2
2.9
3.5

6-MAM
Codeine
Hydrocodone
Hydromorphone
Morphine
Oxycodone
Oxymorphone
Amphetamine
MDA
MDEA
Methamphetamine
MDMA
Diazepam
Oxazepam
Benzoylecgonine
Fentanyl
Methadone
Propoxyphene
Norpropoxyphene

Total number
of samples
compared

Number of
results that
agreed

Number of
results that
disagreed

Percentage of
results that
agreed

26
52
55
57
51
53
59
46
28
16
42
26
40
38
43
7
6
6
5

23
48
53
52
49
50
55
45
26
1.5
41
21
37
36
42
7
6
5
5

3
4
2
5
2
3
4
1
2
1
1
5
3
2
1
0
0
1
0

88.5
92.3
96.4
91.2
96.1
94.3
93.2
97.8
92.9
93.8
97.6
80.8
92.5
94.7
97.7
100.0
100.0
83.3
100.0

1124

Y. Lee et al. / Clinical Biochemistry 46 (2013) 11181124

a xed cost of $375,000 in setting up an LCMS/MS system with a comprehensive service contract and a reagent cost of $1 per test, a laboratory running 35 samples per day can offset this upfront cost in 5 years by
merely the cost savings. It is important to realize here that we leave out
all costs pertaining to the conrmatory tests, because reex testing policy varies between institutions. Also, we do not consider differences in
test reimbursement. At our medical center all positive screening results
by EMIT are reexed for GCMS conrmation and the associated costs
are substantial. We hope that our newly developed DUSLCMS/MS
method, which detects and quanties drugs of abuse in a single run,
will obviate the need for screening with reex GCMS conrmation in
the near future. Not only will this improve patient care by reducing
turn-around time, it will also signicantly reduce the overall cost.
The chromatographic component of this DUSLCMS/MS method
was built with expandability for more analytes in a single run. For example, 11-nor-9-carboxy-9-THC can be incorporated into the analysis
as shown in Fig. 1A. Unfortunately, at the time of our testing, the scheduled MRM function of the Analyst software did not allow polarity
switching within a run, which prevented us from further validation.
Here, we demonstrated a reliable DUSLCMS/MS method which
simultaneously detects and quanties 19 drugs of abuse. Excellent
sensitivity and specicity were achieved with only 15 L of urine, a
cost-efcient HPLC-triple quadrupole mass spectrometer, and regular
HPLC columns. It is particularly worth noting that our DUS extraction
method offers great sample stability and convenience in sample transport and storage, making it an ideal choice for drug screenings at
schools, workplaces, penitentiaries, forensic ofces, and pain management clinics.

Acknowledgments
The authors would like to thank Kathy Hurst of Cedars-Sinai Medical
Center for her administrative support, Jessica Bertain of Phenomenex
for the generous supply of the chromatography validation kit, and
Julie Hilton of GE Healthcare for providing the DMPK and the 903
cards for our initial testing.

Appendix A. Supplementary data


Supplementary data to this article can be found online at http://
dx.doi.org/10.1016/j.clinbiochem.2013.03.027.

References
[1] Eichhorst JC, Etter ML, Rousseaux N, Lehotay DC. Drugs of abuse testing by tandem
mass spectrometry: A rapid, simple method to replace immunoassays. Clin Biochem
2009;42:153142.
[2] Peat M, Davis AE. Analytical considerations and approaches for drugs. In: Karch
SB, editor. Drug Abuse Handbook. Boca Baton, FL: CRC Press; 1998.
[3] Badawi N, Simonsen KW, Steentoft A, Bernhoft IM, Linnet K. Simultaneous screening
and quantication of 29 drugs of abuse in oral uid by solid-phase extraction and
ultraperformance LCMS/MS. Clin Chem 2009;55:200418.
[4] de Castro A, Concheiro M, Shakleya DM, Huestis MA. Development and validation
of a liquid chromatography mass spectrometry assay for the simultaneous quantication of methadone, cocaine, opiates and metabolites in human umbilical cord. J
Chromatogr B Analyt Technol Biomed Life Sci 2009;877:306571.
[5] Siuzdak G. In: Siuzdak G, editor. The expanding role of mass spectrometry in biotechnology. 2nd ed. San Diego, CA: MCC Press; 2006.
[6] Rook EJ, Hillebrand MJ, Rosing H, van Ree JM, Beijnen JH. The quantitative analysis
of heroin, methadone and their metabolites and the simultaneous detection of cocaine, acetylcodeine and their metabolites in human plasma by high-performance
liquid chromatography coupled with tandem mass spectrometry. J Chromatogr B
Analyt Technol Biomed Life Sci 2005;824:21321.
[7] Edelbroek PM, van der Heijden J, Stolk LM. Dried blood spot methods in therapeutic
drug monitoring: methods, assays, and pitfalls. Ther Drug Monit 2009;31:32736.
[8] Keevil BG. The analysis of dried blood spot samples using liquid chromatography
tandem mass spectrometry. Clin Biochem 2011;44:1108.
[9] Li W, Tse FL. Dried blood spot sampling in combination with LCMS/MS for quantitative analysis of small molecules. Biomed Chromatogr 2010;24:4965.
[10] U.S. Department of Health and Human Services FaDA, Center for Drug Evaluation,
Research (CDER), Center for Veterinary Medicine (CVM). Guidance for industry:
bioanalytical method validation; 2001.
[11] Mass spectrometry in the clinical laboratory: general principles and guidance;
approved guideline. Clinical and Laboratory Standards Institute; 2007.
[12] The Substance Abuse and Mental Health Services Administration (SAMHSA).
Mandatory guidelines for federal workplace drug testing programs; Federal
Register November 25, 2008 (73 FR 71858); 2008. p. 71858907.
[13] The Substance Abuse and Mental Health Services Administration (SAMHSA).
Mandatory guidelines for federal workplace drug testing programs; Federal
Register April 30, 2010 (75 FR 22809); 2010. p. 2280910.
[14] Verstraete AG, Heyden FV. Comparison of the sensitivity and specicity of six
immunoassays for the detection of amphetamines in urine. J Anal Toxicol 2005;29:
35964.
[15] Feng J, Wang L, Dai I, Harmon T, Bernert JT. Simultaneous determination of multiple
drugs of abuse and relevant metabolites in urine by LCMSMS. J Anal Toxicol
2007;31:35968.
[16] Yuan C, Heideloff C, Kozak M, Wang S. Simultaneous quantication of 19 drugs/
metabolites in urine important for pain management by liquid chromatography
tandem mass spectrometry. Clin Chem Lab Med 2012;50:95103.
[17] Meany DL, Clarke W. Opiate dau screening using dried urine specimens. Clin Chim
Acta 2009;401:1889.
[18] Sasaki TA. Forensic technology widens net for drugs of abuse: The rise of
LC/MS/MS for toxicology testing. Forensic Magazine 2007.
[19] Lewandrowski K, Flood J, Finn C, Tannous B, Farris AB, Benzer TI, et al. Implementation
of point-of-care rapid urine testing for drugs of abuse in the emergency department
of an academic medical center: impact on test utilization and ED length of stay. Am
J Clin Pathol 2008;129:796801.