Download Distribution of Triazolam and α-Hydroxytriazolam in a Fatal

Journal of Analytical Toxicology, Vol. 26, January/February 2002
[CaseReport
Distribution of Triazolam and (z-Hydroxytriazolam in a
Fatal Intoxication Case
Barry tevine 1,~,*, Alison Grieshaber 2, Joseph Pestaner 1, Karla A. Moore 1, and John E. Smialek 1
70ffice of the Chief Medical Examiner, State of Maryland, 111 Penn Street, Baltimore, Maryland 21201 and 2Division of Forensic
Toxicology, Armed Forces Institute of Pathology, Washington, D.C, 20306-6000
[Abstract
The case of a 77-year-old woman who was found dead in her bathtub
with her head clearly above the water line is presented. The decedent
had a medical history of depression, liver disease,spinal stenosis,and
diabetes mellitus. An empty medication bottle of triazolam was found
in the trashcan. At autopsy, no injury or evidence of drowning was
found. Toxicological analysis identified triazolam at a concentration
of 0.12 mg/L in the heart blood. Triazolam and cc-hydroxytriazolam
were quantitated in the specimens received. The medical examiner
ruled that the cause of death was triazolam intoxication and the
manner of death was suicide.
Introduction
Triazolam is a short-acting triazolobenzodiazepine used to
treat insomnia. It is sold under the trade name Halcion | and is
available in 0.125-mg or 0.25-rag tablets (1). A number of pharmacokinetic studies indicate peak blood concentrations up to
0.017 mg/L following doses up to 1 mg (2). Triazolam is extensively metabolized in humans by hydroxylation and conjugation. Its major metabolite, 1-hydroxymethyltriazolam (o~-OHtriazolam), is pharmacologically active. The major urinary
metabolites are glucuronide conjugates of the hydroxylated
metabolites; only trace amounts of parent drug are detected in
the urine (3). A number of fatalities due to triazolam with and
without alcohol and other drugs have been reported (4--8).
The following is a report of triazolam intoxication in which
triazolam and r
were quantitated in the biological specimens.
Case History
A 77-year-old woman was found dead in her bathtub. She
was wearing a bathing suit, and her head was clearly above the
water line. Police were called to the house by a neighbor, who
* Disclaimer: The opinions or assertionscontained herein are the private views of the authors
and are not to be construed as official or as reflecting the views of the Department of Defense
or of the Army, Navy, or Air Force.
Acknowledgment: This work was supported in part by the American Registryof Pathology.
* Author to whom correspondence should be addressed.
52
also unlocked the door to allow the police to enter. The house
was secure, and the decedent was in a second floor bathtub.
There was an empty medication bottle of triazolam found in the
trashcan with its cap on the sink counter; a note demonstrating
increased somnolence was also at the scene. The decedent's
medical history included depression, liver disease, spinal
stenosis, and diabetes mellitus.
A complete autopsy was performed, and no injury was noted.
In addition, no water was identified in the sinuses, airway, or
stomach, and no temporal bone hemorrhage was noted. Other
than emphysema, there were no significant pulmonary findings.
Moderate atherosclerotic cardiovascular disease and hepatic
cirrhosis were found. The spleen was enlarged, and the sclerae
were not icteric. No medication was seen in the gastric contents,
which consisted of 160 mL of fluid and digested material. Based
on the investigation and autopsy findings, the cause of death
was pended for toxicological analysis.
Experimental
Triazolam and oc-OH-triazolam analysis
Specimen preparation. Tissue specimens were prepared according to a previously presented method (9). TWo grams of
tissue was mixed with 0.3 mL 1.0N acetic acid and centrifuged.
The supernatant was removed, and the pH was adjusted to 6 by
adding base. Two internal standards were used: 0.2 mg alprazolam for triazolam quantitation and 0.1 mg oc-hydroxyalprazolam-d5 for (z-OH-triazolam analysis. Internal standards were
added directly to the blank, calibrators, control, and fluids prior
to buffering; internal standards were added prior to the preparation of tissue homogenates and buffering.
Five calibrators in drug-free blood were prepared at the following concentrations: 0.02, 0.05, 0.1, 0.2, and 0.5 rng/L. In addition, a separate 0.1-mg/L blood control was prepared. The
0.02-mg/L calibrator served as the lower limit of quantitation;
the 0.5-mg/L calibrator served as the upper limit of quantitation.
One milliliter of 100raM pH 6 phosphate buffer was added to
2 mL blank, calibrator, control, fluid, or tissue supernatant.
After mixing and centrifuging, each solution was added to a
Reproduction (photocopying) of editorial content of this journal is prohibited without publisher's permission.
Journal of Analytical Toxicology, Vol. 26, January/February 2002
10-mL Clean Screen| extraction column previously conditioned by sequential additions of 3 mL methanol, 3 mL deionized water, and 2 mL 100raM pH 6 phosphate buffer. The
columns were washed with 3 mL deionizedwater and 2 mL 20%
acetonitrile in 100mM pH 6 phosphate buffer. The columns
were then dried for 5 rain and rinsed with 1 mL hexane. The
compounds were eluted with 2 mL 2% ammonium hydroxide in
ethyl acetate. The eluants were evaporated to dryness.
Enzymatic hydrolysis. Internal standard and 1 mL of 100mM
pH 6 phosphate buffer were added to urine samples. Hydrolysis
was perfomedby adding ~-glucuronidase(Helixpomatia; Sigma,
type H-2 crude solution) at a concentration equivalent to
2500-5000 U/mL of urine used. The mixture was vortex mixed.
After ensuring that the pH was between 5.5 and 6.5, the tubes
were placed in a 56~ water bath for 1.5 h. After centrifugation,
the supernatant was removed and extracted as described.
Derivatization. The residues were reconstituted in 50 IJL
BSTFA and heated at 90~ for 15 min to produce the
trimethylsilyl derivatives of the hydroxylated compounds. After
cooling, 2 tJL was injected into the gas chromatograph-mass
spectrometer (GC-MS).
Instrumentation. A Hewlett-Packard 6890 GC attached to a
5973 mass selectivedetector.The followingconditions were used:
GC column, DB-1 15-m x 0.25-ram column, 0.25-1Jmthickness;
injector, splitless; injection port temperature, 255~ transfer
line temperature, 255~ oven parameters, initial temperature
140~ rate #1, 25.0~
to 225~ rate #2, 12.0~
to
245~ rate #3, 30.0~
to 300~ run time, 8.9 min; carrier
gas, helium; and constant pressure, 10.0 psi.
The mass spectrometer was operated in the electron ionization mode. The following ions were monitored: m/z 279 for alprazolam; m/z 386 for ~-hydroxyalprazolam-d~-TMS;m/z 313,
342, and 238 for triazolam; and m/z 415, 417, and 430
for a-hydroxytriazolam-TMS. The quantitating ions for
triazolam and ~-hydroxytriazolam-TMSwere m/z 313 and 415,
respectively.
Results
A urine specimen from the presented case was tested for
volatile substances, therapeutic, and abused drugs. Volatile
Table I. Triazolam (T) and c~-Hydroxytriazolam (OH-T)
Concentrations in the Presented Case
Specimen
Blood, heart (rag/L)
Blood, peripheral (mg/L)
Kidney (mg/kg)
Liver (mg/kg)
Cerebrospinal fluid (mg/L)
Urine, unhydrolyzed (mg/L)
Urine, hydrolyzed (mg/L)
* NF = nonefound
+QNS = quantity not sufficient
T
OH-T
0.12
0.091
0.071
0.092
0.023
0.024
QNS ~
NF*
NF
0.22
0.043
NF
0.030
1.37
testing for methanol, ethanol, acetone, and isopropanol was
accomplished using headspace GC. Acidic/neutral and alkaline
drug testing were accomplished by GC-nitrogen-phosphorus
detection. The urine was tested for morphine by radioimmunoassay and for acetaminophen, ethchlorvynol, and salicylate by color tests. No other volatile substances or therapeutic
or abused drugs were detected using these techniques. Because
of the suspicions of a triazolam intoxication, the urine was
screened for benzodiazepines by fluorescence polarization immunoassay at a 0.1-mg/L nordiazepam cutoff. This screening
test was positive. As a result, the specimens were tested for triazolam and ~-OH-triazolam. Table I lists the quantitative results
in the presented case.
Discussion
The heart blood concentration of 0.12 mg/L in the presented
case, is consistent with blood concentrations previously reported in triazolam fatalities. Bal et al. (4) reported three cases
of drug intoxication involving triazolam. Blood concentrations
ranged from 0.016 to 0.026 mg/L; other drugs were present in
each case. Steentoft and Worm (5) measured triazolam in blood
in five cases of triazolam intoxication. In one case with no
ethanol present, the blood concentration was 0.11 mg/L. In
four cases with ethanol greater than 0.10 g/dL present, the
blood triazolam concentrations ranged from 0.04 to 0.22 mg/L.
A blood triazolam concentration as high as 0.30 mg/L has been
reported in an intoxication with a 0.10-g/dL blood-alcohol concentration (BAC) (6). Joynt (7) published blood data on four
deaths due to triazolam alone. The blood concentrations ranged
from 0.01 to 0.04 mg/L. Another nine deaths were identified
with triazolam where the BAC exceeded 0.10 g/dL. The blood
triazolam concentrations in these cases ranged from 0.01 to
0.053 mg/L. Kovesand Wells (8) reported three additional cases
with a BACgreater than 0.10 g/dL; blood triazolam concentrations ranged from 0.022 to 0.065 mg/L.
In the presented case, both heart blood and peripheral blood
from the subclavian vein were collected. The ratio of heart
blood to peripheral blood triazolam concentrations was 1.61.
Dalpe-Scott et al. (10) reported ratios of heart blood to femoral
blood triazolam concentrations in four postmortem cases. The
average heart blood to femoral blood ratio was 2.8 with a range
of 1.0 to 7.5. No actual blood concentrations were provided.
Postmortem redistribution of benzodiazepines has not been
widely reported. However, the volume of distribution of the
basic drug triazolam (1.1-2.7 L/kg [2]), may suggest the possibility of postmortem redistribution.
This is the first reported triazolam fatality where the major
metabolite of triazolam, ~-OH-triazolam, was also measured.
~-OH-triazolam was not detected in either postmortem blood
specimen at a limit of quantitation of 0.02 rag/L, but was detected in the liver, kidney, and urine. This is not surprising because the liver is the location of ~-OH-triazolam formation.
Moreover, the urine was analyzed with and without glucuronide
hydrolysis. As expected, most of the r
was in the
conjugated form. This is consistent with previously reported
53
Journal of Analytical Toxicology,Vol. 26, January/February2002
urine excretion studies involving triazolam (11). One unexpected finding was the high concentration of r
detected in the kidney relative to the blood and liver. In a previously reported fatality due to alprazolam, a triazolobenzodiazepine like triazolam, no r
was detected in
the kidney (12). Because this is the only reported triazolam fatality where r
is quantitated, it cannot be determined whether or not this observation is an outlier.
Based on the investigation, autopsy, and toxicology findings,
the medical examiner ruled that the cause of death was triazolam intoxication and the manner of death was suicide.
References
1. Physicians' Desk Reference, 55th ed. Medical Economics Co.,
Montvale, NJ, 2001.
2. R.C. Baselt. Disposition of Toxic Drugs and Chemicals in Man, 5th
ed. Chemical Toxicology Institute, Foster City, CA, 2000.
3. F.S. Eberts, Jr., Y. Philopoulos, L.M. Reineke, and R.W. Vliek. Triazolam disposition. Clin. Pharm. ?'her. 29:81-93 (1981).
4. T.S. Bal, B. Johnson, and E.A. Kilner. Three deaths involving triazolam-analytical aspects.J. Forensic ScL Soc. 29:119-123 (1989).
54
5. A. Steentoft and K. Worm. Cases of fatal triazolam poisoning.
J. Forensic 5ci. 5oc. 33:45-48 (1993).
6. J.J.O'Dowd, RR Spragg, and RA. Routledge. Fatal triazolam poisoning. Br. Med. J. 297:1048 (1988).
7. B. Joynt. Triazolam blood concentration in forensic cases in
Canada. J. Anal. Toxicol. 17:171-177 (1993).
8. G. Koves and J. Wells. The quantitation of triazolam in postmortem
blood by gas chromatography/negative ion chemical ionization
mass spectrometry. J. Anal ToxicoL 10:241-244 (1986).
9. K.K. Martin, J. Magluilo, Jr., and K.A. Moore. Solid phase extraction
on the Zymark Rapid Trace Workstation for the screening of basic
drugs. Presented at the 28th Society of Forensic Toxicologists
Meeting, Albuquerque, NM, 1998
10. M. Dalpe-Scott, M. Degouffe, D. Garbutt, and M. Drost. A comparison of drug concentration in postmortem cardiac and peripheral blood in 320 cases. Can. 5oc. Forensic 5ci. J. 28:113-121
(1995).
11. A.D. Fraser,W. Bryan, and A.F. Isner. Urine screening for s-OH triazolam by FPIA and EIA and confirmation by GC/MS. J. Anal.
Toxicol. 16:347-350 (1992).
12. A.J. Jenkins, B. Levine, J.L. Locke, and J.E. Smialek. A fatality due
to alprazolam intoxication. J. Anal. Toxicol. 21:218-220 (1997).
Manuscript received April 16, 2001;
revision received June 26, 2001.