| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
a Address correspondence to this author at: Forensic Toxicology, AFIP Annex, 1413 Research Blvd., Rockville, MD 20850. Fax 301-319-0628; e-mail paul{at}afip.osd.mil
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResults and DiscussionConclusionReferences |
|---|
Methods: Urine samples containing opiates were extracted, derivatized, and analyzed using gas chromatography–mass spectrometry with selective ion monitoring.
Results: The limits of detection for codeine, morphine, and 6-AM were 6, 5, and 0.5 µg/L, respectively. Recoveries were >90%. Quantification was linear over the concentration range of 6–1000 µg/L for codeine, 5–5000 µg/L for morphine, and 0.5–800 µg/L for 6-AM. Cutoff concentrations for confirmation of opiates were 100, 100, and 10 µg/L for free codeine, free morphine, and 6-AM.
Conclusion: The proposed cutoff concentrations for free morphine and 6-AM provide better detection windows for morphine and heroin use than the cutoff concentrations for total morphine and 6-AM used at present. Detection of free codeine, instead of total codeine, simplifies interpretation of codeine use. The single-extraction method enables simultaneous, less labor-intensive analysis of morphine, codeine, and 6-AM.© 1999 American Association for Clinical Chemistry
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionConclusionReferences |
|---|
Initial screening tests by radioimmunoassay followed by gas chromatography (GC)1 confirmatory tests were introduced by the Department of Defense (DoD) in 1983 for monitoring the use of controlled substances by military personnel. The cutoff concentrations for both tests for opiates were 300 µg/L. Specimens found positive in the screening test were tested by a GC confirmation procedure for codeine and morphine. In confirmation, total (conjugated and unconjugated) codeine and total morphine were tested after acid hydrolysis of the corresponding conjugated compounds. In 1984, the GC confirmation procedure was replaced by a more reliable gas chromatography–mass spectrometry (GC-MS) confirmation procedure (2).
In September 1984, the DoD opiate detection procedure was challenged when the urine concentrations of total codeine and total morphine of a volunteer who had ingested ~25 g of Indian poppy seeds were found to be 832 and 1458 µg/L, respectively. Almost immediately, all opiate-positive specimens were reported with a caution that explained the effect of poppy seed on test results. In 1986, DoD introduced detection of 6-acetylmorphine (6-AM), a unique metabolite of heroin, as a tool to confirm heroin use (3). The cutoff concentration for 6-AM was 10 µg/L. Specimens that were positive for morphine in GC-MS confirmation were tested again for 6-AM. However, the 300 µg/L morphine confirmation cutoff was too low because it produced a large number of specimens that contained either no detectable 6-AM or 6-AM below the cutoff concentration. To improve correlation between cutoff concentrations of total morphine and 6-AM and to alleviate positive results from poppy seed ingestion, in 1988, the morphine confirmation cutoff was increased from 300 to 4000 µg/L. In 1995, the immunoassay cutoff concentration was also increased from 300 to 2000 µg/L to eliminate a large number of immunoassay-positive specimens that were negative in the confirmation test.
In 1988, the Department of Health and Human Services (DHHS) introduced guidelines for testing Federal Agency and Department of Transportation regulated specimens (4). The opiate screening and confirmation cutoff concentrations were established at 300 µg/L. As part of the protocol, all results were required to be reviewed by a Medical Review Officer to ensure that the positive results were not because of legitimate use of codeine and morphine or by ingestion of dietary poppy seeds. The Medical Review Officer examined donors for evidence of drug abuse by taking a clinical history and performing a physical examination, and often requested a test for 6-AM to confirm heroin use. In the guidelines, no cutoff was mandated for 6-AM confirmation. Generally, the limit of detection (LOD) of a procedure was used to confirm the presence of 6-AM in urine. Because the LOD varied considerably between laboratories, the 6-AM results were misleading in some forensic investigations.
In 1995 and 1997, the DHHS revised the opiate guidelines
(5)(6) with an effective date of May 1, 1998.
Later in a separate memorandum, the DHHS postponed the implementation
date until December 1, 1998, because additional time was needed to
validate immunoassay test kits and 6-AM confirmatory procedures. In the
guidelines, the screening cutoff for opiate testing was increased from
300 to 2000 µg/L. The confirmation cutoff concentrations for both
codeine and morphine were also increased from 300 to 2000 µg/L. The
guidelines require all morphine-positive specimens to be tested again
for 6-AM at a cutoff concentration of 10 µg/L. Because 6-AM is a
unique metabolite of heroin, its presence in urine would be used to
confirm heroin use. If the 6-AM concentration is <10 µg/L, the
specimen will be reported as positive for morphine (
2000 µg/L)
only. A specimen will be reported as positive for codeine if the
codeine concentration is 2000 µg/L. The objective of the new cutoff
concentrations was to eliminate a considerable number of specimens that
may be positive from poppy seed use.
Both DoD and DHHS currently require two confirmation tests of urine to verify use of heroin, a test for total morphine and a separate test for 6-AM. The objective of this study was to determine appropriate cutoff concentrations for free codeine, free morphine, and 6-AM that could be substituted for the cutoff concentrations for total morphine, total codeine, and 6-AM, using retrospective published data (7)(8)(9), and to develop a confirmatory assay that would allow simultaneous measurement of free morphine, free codeine, and 6-AM. The combined objective was to propose a method for identifying codeine, morphine, and heroin use with one urine immunoassay and one confirmation test.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionConclusionReferences |
|---|
instrument
A Hewlett-Packard (HP) GC-MS system consisting of an HP 5890
series II Plus GC and 5972 quadrupole mass selective detector (MSD),
and Vectra XM2 4/100i computer workstation was used. An HP 18593B
autoinjector was used to inject the samples into the GC-MS. The MSD was
operated under electron impact mode. The ion window and dwell time were
0.2 atomic mass units and 50 ms, respectively. The electron multiplier
was set at the autotune voltage for codeine and morphine, and 600 V
above autotune for 6-AM. The flow rate of helium through a DB-5MS
capillary column [5:95 phenyl:methyl siloxane, 15 m x 0.25 mm
(i.d.); J&W Scientific] was 1.4 mL/min.
specimen analysis at the navy laboratory (1986)
Urine specimens (422 237) were tested at the Navy Drug Screening
Laboratory, Norfolk, VA during 1986 as part of the Department of the
Navy random urinalysis program. The specimens were tested initially by
radioimmunoassay at a cutoff concentration of 300 µg/L. The
confirmation procedures for total morphine and 6-AM were same as the
published procedures, using a cutoff of 300 and 10 µg/L,
respectively, to report samples as positive for heroin
(2)(3). In brief, the samples for morphine
analysis were hydrolyzed with acid. After an acid-base separation, the
morphine was extracted by solvent and then detected by GC-MS as acetyl
derivative. Acid hydrolysis was avoided for 6-AM analysis. The compound
was extracted using a procedure similar to morphine and was detected as
the propionyl derivative. Linear ranges for morphine and 6-AM were
25–800 and 1–100 µg/L, respectively. All positive samples were
quantified, allowing a retrospective application of different cutoffs
for study purposes.
proposed analytical procedure for confirmation
All 6-AM and d6-AM stock solutions were
prepared in acetonitrile because the compounds were unstable in
methanol or ethanol solutions. When stored at 2–4 °C in
acetonitrile , the compounds were stable for at least 3 years. For
extraction, a mixture of internal standards,
d6-codeine, d3-morphine,
and d6-AM (15, 15, and 0.5 mg/L in 0.1 mL of
acetonitrile) were added to 5.0 mL of urine specimens and calibrator
and control solutions containing known amounts of codeine, morphine,
and 6-AM. The specimens were clinical samples collected in 1993 and
stored frozen at -18 °C. The concentrations of 6-AM, morphine, and
codeine in the calibrators were 10, 300, and 300 µg/L, respectively.
Two control solutions at concentrations twofold lower than the
calibrators (5, 150, and 150 µg/L) and twofold higher than the
calibrators (20, 600, and 600 µg/L) for 6-AM, morphine, and codeine,
respectively, were used to verify the quantitative results of the batch
analysis. A negative control was used to verify contamination during
analysis. Calibrators and controls were used in each batch analysis.
Phosphate buffer (2 mL of 0.1 mol/L, pH 6.0) was added to the samples.
The pH values of the solutions were 6.0 ± 0.5. At this pH range,
the opiates were in cationic forms. The solutions were poured into
solid-phase extraction columns prewashed with methanol (3 mL),
deionized water (3 mL), and phosphate buffer (1 mL of 0.1 mol/L, pH
6.0). The solutions were allowed to pass through the columns by gravity
flow. The columns were washed with deionized water (2 mL), HCl (2 mL of
0.001 mol/L, pH 3.0), and methanol (3 mL). Methanol removed most of the
nonionic compounds. The columns were dried for 2 min using suction. The
opiates were extracted from the sorbent with 3 mL of a mixture of
methylene chloride–isopropanol–14.8 mol/L ammonium hydroxide
(8:2:0.2, by volume), using gravity flow. When the eluting solvent
passed through the column, mild suction was used to collect the last
drop. The collected solutions in 10-mL glass tubes were
evaporated to dryness at 50 °C under a stream of nitrogen. The
extracts were then derivatized.
derivatization
Pentafluoropropionylation (10).
In routine
analysis, this derivatization is recommended for detection of codeine,
morphine, and 6-AM. Approximately 50 µL of pentafluoropropionic
anhydride was added to the extract in 10-mL glass tubes. The tubes were
capped with polyethylene caps, vortex-mixed, and heated at 70 °C for
15 min in a sand bath. The excess reagent was evaporated to dryness at
50 °C under a stream of nitrogen.
Propionylation (3).
The extracts in the 10-mL glass
tubes were dissolved in 50 µL of propionic anhydride and 50 µL of
dry pyridine. The tubes were capped with polyethylene caps,
vortex-mixed, and heated at 70 °C for 15 min in a sand bath. The
excess reagents were evaporated to dryness at 50 °C under a stream
of nitrogen.
Silylation.
The extracts in the 10-mL glass tubes were
dissolved in 50 µL of BSTFA with 10 g/L trimethyl-chlorosilane.
The tubes were capped with polyethylene caps, vortex-mixed, and
heated at 70 °C for 15 min in a sand bath. The compounds were
injected into the GC-MS instrument with the excess derivatizing
agents as solvents.
gc-ms analysis
The pentafluoropropionylated and propionylated compounds were
dissolved in 50 µL of acetonitrile and transferred into autoinjector
vials. The vials were sealed with Teflon-coated rubber discs, and
~1–2 µL of the samples for codeine and morphine and ~3–4 µL
of the samples for 6-AM were introduced into the GC-MS for analysis,
using an autoinjector.
GC-MS conditions for pentafluoropropionyl derivatives.
The
instrument was operated in splitless and temperature program mode. The
split valve was turned on at 0.3 and 0.5 min after the codeine/morphine
and 6-AM injections, respectively. The oven temperature was increased
from 120 to 235 °C at 30 °C/min. The oven was held at 120 and
235 °C for 0.3 min and 2 min, respectively. Both injector and
detector temperatures were 280 °C. The following ions were
monitored: for morphine, m/z 577, 430, and 414; for
d3-morphine, m/z 580 and 433; for
codeine, m/z 445, 283, and 282; and for
d6-codeine, m/z 451 and 288. When 6-AM
was analyzed, the monitored ions were m/z 473, 414, and 361
for 6-AM; and m/z 479 and 417 for
d6-AM. For 6-AM analysis, the MSD was turned on
immediately after the retention time (RT) of morphine. When
semisynthetic opiates interfered in the detection of 6-AM, the ion
m/z 414 was changed to m/z 474 and the oven
temperature was increased from 150 to 240 °C at 10 °C/min. The
oven was held at 150 and 240 °C for 0.3 min and 2 min, respectively.
GC-MS conditions for propionyl derivatives.
The split valve
was turned on at 0.5 min after the codeine/morphine and 6-AM
injections. The oven temperature was increased from 180 to 290 °C at
15 °C/min. The oven was held at 180 °C for 0.5 min. Both injector
and detector temperatures were 280 °C. The following ions were
monitored: for morphine, m/z 397, 341, and 324; for
d3-morphine, m/z 400 and 344; for
codeine, m/z 355, 282, and 229; and for
d6-codeine, m/z 361 and 288. When 6-AM
was analyzed, the ions monitored were m/z 384, 383, and 324
for 6-AM; and m/z 389 and 333 for
d6-AM. For 6-AM analysis, the MSD was turned on
immediately after the RT of morphine.
GC-MS conditions for silyl derivatives.
The split valve was
turned on at 0.5 min after the codeine/morphine and 6-AM injections.
The oven temperature was increased from 200 to 310 °C at
35 °C/min. The oven was held at 200 and 310 °C for 1.0 min and
0.5 min, respectively. Both injector and detector temperatures were
280 °C. The ions monitored were as follows: for morphine,
m/z 429, 430, and 401; for
d3-morphine, m/z 432 and 433; for
codeine, m/z 371, 234, and 229; and for
d6-codeine, m/z 377 and 378. When 6-AM
was analyzed, the ions monitored were m/z 399, 400, and 340
for 6-AM; and m/z 405 and 406 for
d6-AM. For 6-AM analysis, the MSD was turned on
immediately after the RT of morphine.
|
Results and Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionConclusionReferences |
|---|
2000 µg/L, and 17 of these showed a detectable amount of
6-AM (LOD, 1 µg/L). Analytical results of 16 specimens were reported
in a separate publication (3). The concentrations of 6-AM
and total morphine in two other specimens were 1 and 6793 µg/L, and
463 and 29 355 µg/L, respectively. The number of specimens positive
for 6-AM with concentrations of total morphine 2000 µg/L are
summarized in Table 1
. Only 9 specimens were found to contain 6-AM 10 µg/L with
the total morphine 4000 µg/L compared with 13 specimens containing
6-AM 10 µg/L with the total morphine 2000 µg/L. Two more
specimens (total of 15) were found to contain 6-AM between 5 and 10
µg/L. If the 4000 µg/L cutoff was used for total morphine, ~30%
of the specimens would be designated as negative for heroin when the
6-AM concentrations were above the cutoff concentration.
|
In a study conducted by the DHHS (6), approximately 1.1 x 106 specimens were tested for opiates by five certified laboratories, and 7294 specimens (0.66%) were found positive for codeine and/or morphine. The 300 µg/L cutoff was used in this study. The specific GC-MS procedures used to confirm the presence of the drugs were not reported. When 848 specimens were tested for 6-AM, only 16 specimens were found positive for 6-AM (at or above the LOD). In 14 of these 16 specimens, total morphine was >2000 µg/L. The prevalence of 6-AM (16 of 848) was similar to the results we observed in specimen analyses at the Navy laboratory (17 of 926).
review of published data from clinical studies of heroin
Considerable information about the relationship between total
morphine, free morphine, and 6-AM is available from two sets of
clinical experiments (7). In the experiments, doses of 3 and
6 mg of heroin were administered separately to six human subjects.
Ninety-two urine specimens were tested, and 24 specimens showed
detectable amounts of 6-AM. The relationships between total morphine,
free morphine, and 6-AM are summarized in Table 2
. A total of 16 specimens were found to contain 6-AM 10
µg/L. An additional specimen (total of 17) was found to contain 6-AM
between 5 and 10 µg/L. When 5 and 10 µg/L cutoff concentrations
were compared, the results of the clinical studies (16 vs 17) were
similar to the results of specimen analyses conducted by us at the Navy
laboratory (13 vs 15). However, if the proposed DHHS cutoffs of 2000
µg/L for total morphine and 10 µg/L for 6-AM were applied, only 10
of the 16 6-AM-positive specimens would be designated as positive for
heroin. The reason for missing the positive specimens could be
attributed to the detection window for 6-AM, i.e., the first 4 h
after heroin use preceded the time window for peak total-morphine
concentration. The number of specimens containing 6-AM that would be
identified as negative for heroin use would be even greater using the
4000 µg/L cutoff in effect in the DoD drug testing program. On the
basis of the results of this study, the total-morphine cutoff is the
parameter that limits the number of heroin-positive specimens.
|
Further investigation of the clinical results showed that all 16
specimens that contained 6-AM 10 µg/L also contained free morphine
100 µg/L. The results of 27 specimens that showed free morphine
100 µg/L, total morphine 2000 µg/L, and 6-AM 1 µg/L were
analyzed statistically [Table 2
in Ref. (7)]. The
correlation coefficient (r) between the concentrations of
6-AM and free morphine was 0.8336 compared with 0.5657 between 6-AM and
total morphine. Approximately 69% (r2
= 0.69) of the free-morphine concentrations, compared with only 32%
(r2 = 0.32) of the total-morphine
concentrations, were related to 6-AM concentrations. If the present
DHHS cutoff for total morphine of 2000 µg/L was used, only 10 of 16
specimens would be tested for 6-AM, although the 6-AM concentrations
for all 16 specimens were 10 µg/L. The probability (P)
that all 16 6-AM specimens would be positive for total morphine was
calculated using the standard gaussian distribution method. The
P (n 16) was found to be only 0.0013 (z 3.1).
Similarly, the P (n 16) for 6-AM using the DoD cutoff for
total morphine of 4000 µg/L was 3.0 x 10-5
(z 4.5). Therefore, when a cutoff for total morphine of
2000 or 4000 µg/L is used, few specimens that contain 6-AM 10
µg/L would not be tested for 6-AM and would be designated as negative
for heroin use. Present DoD and DHHS guidelines require acid or enzyme
hydrolysis of conjugated morphine to measure the total morphine. The
6-AM cannot be analyzed under these hydrolytic conditions. A second
confirmation procedure must be conducted to test 6-AM in urine. If the
present guidelines were modified to test for free morphine and 6-AM, a
method could be developed that could detect both compounds
simultaneously.
review of published data from clinical studies of morphine
If morphine is consumed, the same 100 µg/L cutoff for free
morphine may be applied to investigation of morphine use. In a clinical
experiment, a dose of 20 mg of morphine was administered to four
subjects (8). At a cutoff concentration of 2000 µg/L for
total morphine or 100 µg/L for free morphine, specimens were positive
up to 36 h after drug administration (Table 3
). When the free-morphine cutoff of 100 µg/L and the
total-morphine cutoff of 2000 µg/L were used, the number of specimens
positive for morphine were 18 and 16, respectively. The P (n
18) of the total-morphine specimens was 0.0668 (z
1.5). Similar to the heroin results, these two specimens (specimen E)
were collected within 1.2 h after drug administration. It appeared
that the metabolism of free morphine to the conjugated compound was
lowest at the early stage of excretion. Typically, free morphine was
~26–34% of total morphine within the first hour of excretion and
gradually decreased to ~5% over a period of 36 h. Therefore,
100 µg/L free morphine and 2000 µg/L total morphine correlate well
later in excretion but differ considerably during the initial phase of
excretion. Throughout the excretion period of morphine, the free
morphine at a cutoff concentration of 100 µg/L produced fewer
false-negative results than the total morphine at a cutoff
concentration of 2000 µg/L.
|
review of published data from clinical studies of codeine
In two separate experiments, 60- and 120-mg doses of codeine
phosphate were administered to four subjects (9). When a
cutoff for free codeine of 100 µg/L and a cutoff for total codeine of
2000 µg/L were used, the numbers of specimens positive for codeine
were 44 and 43, respectively (Table 4
). The P (n 44) of the total-codeine specimens was
0.1562 (z 1.01). The difference between the number of
positive specimens (43 vs 44) was not significant. Like total morphine,
the detection of total codeine requires hydrolysis of the conjugated
compounds. The detection of free codeine is advantageous because the
compound can be extracted simultaneously with free morphine and 6-AM.
Free codeine can be detected up to 24 h after drug administration.
It is noteworthy that only 4 of 44 free-codeine-positive specimens
showed free morphine >100 µg/L. The specimens were collected within
6.3 h after administration of 120 mg of codeine. In all four
specimens, the concentrations of free codeine were 52- to 428-fold
higher than the concentrations of free morphine. Generally,
pharmaceutical doses of codeine phosphate are in the range of 10–60
mg. It may take more than a 60-mg dose of codeine for a specimen to be
positive for free morphine. If a specimen is positive for both free
codeine and free morphine and the concentration of free codeine is
substantially higher than (>50-fold) that of free morphine, the
specimen would be considered positive for opiate from codeine use. In
the majority of specimens, free morphine was negative (<100 µg/L)
when free codeine was 100 µg/L, simplifying the interpretation of
codeine use.
|
analytical procedures for detection of 6-am, free morphine, and
free codeine
The procedure described in Materials and Methods was
designed to detect free codeine, free morphine, and 6-AM simultaneously
as the pentafluoropropionyl derivatives. Because the concentrations of
free morphine and free codeine generally are much higher than that of
6-AM, the derivatized extract was analyzed by two separate injection
modes. The codeine and morphine were analyzed together by one GC-MS
method, whereas the 6-AM was analyzed by another method. Although the
methods were different, the entire batch can be programmed in a
sequence table and injected by an autoinjector. After a solvent
injection, each specimen was injected twice and two different sets of
mass ions were monitored. For 6-AM, ~3–4 µL of sample from a
specimen vial was injected, and the 6-AM and
d6-6-AM ions were monitored. The multiplier was
set at 600 V above autotune voltage. For codeine/morphine, ~1–2 µL
of sample was injected from the same specimen vial, and ions of
codeine, morphine, d6-codeine, and
d3-morphine were monitored. The multiplier
voltage was set at the same voltage that the instrument autotuned.
After two injections, a solvent was injected to ensure that there was
no drug carryover to the subsequent samples.
Although the monitored ions were different for 6-AM and codeine/morphine, the GC conditions used in these two methods were the same. The RTs for 3,6-dipentafluoropropionylmorphine, 6-pentafluoropropionylcodeine, and 3-pentafluoropropionyl-6-acetylmorphine were 4.84, 5.09, and 5.48 min, respectively. For 6-AM analysis, the MS detector was turned on after the morphine peak eluted. Otherwise, common ions in morphine may have suppressed the peak of the 6-AM in the chromatogram.
The identification of each drug was based on comparing RTs (± 2%) and relative ion abundances (± 20%) with the reference compound. The overall recoveries were determined by adding internal standards at the beginning and end of the extraction procedure and were 93–97% for codeine, 90–92% for morphine, and 98–100% for 6-AM. Excellent linearity was observed over the concentration ranges: 6–1000 µg/L for codeine, 5–5000 µg/L for morphine, and 0.5–800 µg/L for 6-AM. The slope, intercept, and correlation coefficient were 0.98, 4.1, and 0.9987 for codeine; 0.99, 5.9, and 0.9991 for morphine; and 1.11, 0.16, and 0.9993 for 6-AM, respectively. Below the lowest limit of linearity or LOD, the qualifying ion ratios exceeded ± 20% of that of reference compounds.
In specimen analysis, the LOD may vary considerably with the wide
variation of urine matrices. When this procedure was used, 6-AM as low
as 0.5 µg/L was detected in six different urine samples. Similarly,
both codeine and morphine as low as 5 µg/L were also detected. In
physiological samples, when the 6-AM concentrations were in the range
of 0.8–30 µg/L, the concentrations of free and total morphine were
in the range of 43–841 and 372-8721 µg/L, respectively
(7). Reference solutions at physiological concentrations of
6-AM and morphine were used in this analysis. A chromatogram of a
sample fortified with free morphine, free codeine, and 6-AM at
concentrations of 12.5, 12.5, and 0.5 µg/L, respectively, is
presented in Fig. 1
. The method was used successfully to analyze a group specimens
frozen at -18 °C for ~4 years (1993–1997). To monitor the
stability of the compounds, reference solutions were also stored frozen
with the specimens. No loss of compounds was observed when the results
were compared with results of freshly prepared reference solutions.
|
Chromatographic assays that measure multiple analytes simultaneously
increase demands for resolution and sensitivity. Therefore, we
investigated two other alternate procedures, propionylation and
silylation. The LOD after propionylation of 6-AM with propionic
anhydride and pyridine was found to be 3 µg/L compared with 0.5
µg/L for the pentafluoropropionyl derivative. The monitored ions for
6-AM were m/z 384 (M+ + 1), 383
(M+), and 324; and for
d6-AM, the monitored ions were m/z 389
(M+) and 333. Urine free of 6-AM was fortified
with 1000 µg/L free morphine. When the urine was analyzed for 6-AM,
surprisingly, a small peak at the RT of 6-AM was detected. Although the
ion ratio of 384:383 [(M+ + 1):M+] was the same as
that of reference 6-AM, the ion ratio of 324:383 was significantly
lower. The identical RT and the ratio of molecular ion to the isotopic
ion of the molecule suggested that the compound may be
3-acetyl-6-propionylmorphine (Fig. 2
, I), an isomer of 3-propionyl-6-acetylmorphine (Fig. 2
, II). We
studied the mechanism of mass fragmentation of 3,6-diacetylmorphine
(Fig. 2
, III), 3-propionyl-6-acetylmorphine (Fig. 2
, II), and
3,6-di[2H6]acetylmorphine
(Fig. 2
, IV). The
3,6-di[2H6]acetylmorphine
was synthesized from morphine treated with
[2H6]acetic anhydride and
pyridine. During fragmentation, the hydrogen atom at the 3
-position
of 3,6-diacetylmorphine and 3-propionyl-6-acetylmorphine migrated to
the oxygen atom at the 3-position, producing characteristic fragment
ions [M -
(CH2=CO)]+ and [M -
(CH3-CH=CO)]+,
respectively. Both ions showed the same m/z 327 as the base
peak. The mechanism of fragmentation was further confirmed when
fragment ion m/z 331 [M -
(C2H2=CO)] was
formed as the base peak from the M+ 375 of
3,6-di[2H6]acetylmorphine
(Fig. 2
, IV). The identity of the 3-acetyl-6-propionylmorphine (Fig. 2
, I) produced from morphine and propionic anhydride was further confirmed
when fragment ion m/z 341 [M+ -
(CH2=CO)], characteristic of morphine with a
3-acetyl group, was monitored. Like many alkyl esters of morphine, the
ion m/z 341 after 3-deacetylation also appeared as the base
peak. The source of the acetyl group that produced the minute amount of
3-acetyl-6-propionylmorphine appeared to be the propionic anhydride
reagent. The small amount of acetyl group, present in the reagent as
mixed anhydride, interfered with the detection of 6-AM when the
morphine concentration was >50 µg/L and the 6-AM concentration was
<3 µg/L.
|
Codeine, morphine, and 6-AM were silylated with BSTFA and analyzed by
the GC-MS. Morphine and 6-AM as pentafluoropropionyl, propionyl, or
silyl derivatives exhibited some common ions at different RTs. In
physiological samples, morphine concentrations are generally much
higher than the 6-AM concentrations. Therefore, the strong morphine
peak may easily overlap with the weak 6-AM peak and interfere with the
identification of 6-AM. The separation of the chromatographic peaks of
all three derivatives was evaluated. The separation of RT between 6-AM
and morphine expressed as a percentage was calculated from the
following formula:
 |
The percentage of separation for silyl (5.1%), propionyl (7.6%), and pentafluoropropionyl (11.6%) derivatives showed that the chromatographic separation between 3-pentafluoropropionyl-6-acetylmorphine (from 6-AM) and 3,6-dipentafluoropropionylmorphine (from morphine) was better than the separation of the other two derivatives. Thus, the pentafluoropropionyl derivatives of codeine, morphine, and 6-AM produced better results than the propionyl or silyl derivatives of the same compounds.
Interferences from structurally related semisynthetic opiates were also evaluated. The detection of codeine and morphine as pentafluoropropionyl derivatives had no interference from 1000 µg/L hydrocodone, hydromorphone, and norcodeine. Ion m/z 414 from the pentafluoropropionyl derivative of 6-AM showed interference from norcodeine. When the ion m/z 414 was changed to m/z 474 (M+ + 1) and the GC oven temperature was increased from 150 to 240 °C at 10 °C/min, all three qualifying ions showed no interference from norcodeine. The RT of 3-pentafluoropropionyl-6-acetylmorphine was 7.53 min. The interfering compound was also resolved when 6-AM was tested as a propionyl derivative.
|
Conclusion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionConclusionReferences |
|---|
10 µg/L
(positive) with total morphine <2000 µg/L (negative). Under the
regulatory guidelines, these specimens would not be tested for 6-AM and
would be considered as negative. These 16 6-AM-positive specimens (10
µg/L) contained free morphine 100 µg/L (positive). In the early
phase of excretion after morphine consumption, free morphine was >100
µg/L, whereas total morphine was <2000 µg/L. Therefore, throughout
the excretion period of morphine, detection of free morphine at a
cutoff concentration of 100 µg/L was better than detection of total
morphine at a cutoff concentration of 2000 µg/L for identifying
morphine use. When codeine was consumed, free codeine at a cutoff
concentration of 100 µg/L showed a good correlation with total
codeine at a cutoff concentration of 2000 µg/L. However, testing
of free codeine is advantageous because the compound can be tested
simultaneously with free morphine and 6-AM. In most
free-codeine-positive specimens, the free-morphine concentrations were
<100 µg/L. However, when the concentrations of both free codeine and
free morphine were 100 µg/L, free codeine was excreted at
concentrations higher (>50-fold) than the free morphine, allowing
forensic investigators to distinguish codeine from morphine use.
Detection of 6-AM and free morphine at cutoff concentrations of 10 and
100 µg/L, respectively, provided detection windows that were better
than those provided by total morphine at the cutoff concentration of
2000 µg/L, followed by detection of 6-AM at a cutoff concentration of
10 µg/L. The regulatory cutoff concentrations for the drugs are
summarized in Table 5
. In addition, to improve concordance between cutoffs using free
morphine and free codeine with 6-AM, the procedure presented detected
all three compounds using a single extraction instead of the two
separate labor-intensive procedures currently required. The procedure
can detect codeine, morphine, and 6-AM as low as 6, 5, and 0.5 µg/L,
respectively.
|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
The opinions expressed herein are those of the authors and are not to be construed as official or as reflecting the views of the Department of the Army, the Department of the Navy, or the Department of Defense.
1 Nonstandard abbreviations: GC-MS, gas chromatography–mass spectrometry; DoD, Department of Defense; 6-AM, 6-acetylmorphine; DHHS, Department of Health and Human Services; LOD, limit of detection; d6-AM, N-desmethyl-N-[2H3]methyl-6-[2H3]acetylmorphine; BSTFA, bis(trimethylsilyl)trifluoroacetamide; HP, Hewlett-Packard; MSD, mass-selective detector; and RT, retention time. 
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionConclusionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  J. L Zacher and D. M Givone False-Positive Urine Opiate Screening Associated with Fluoroquinolone Use Ann. Pharmacother., September 1, 2004; 38(9): 1525 - 1528. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. R. Baden, G. Horowitz, H. Jacoby, and G. M. Eliopoulos Quinolones and False-Positive Urine Screening for Opiates by Immunoassay Technology JAMA, December 26, 2001; 286(24): 3115 - 3119. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Ceder and A. W. Jones Concentration Ratios of Morphine to Codeine in Blood of Impaired Drivers as Evidence of Heroin Use and not Medication with Codeine Clin. Chem., November 1, 2001; 47(11): 1980 - 1984. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
