| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
Department of Pathology, Princess Margaret Hospital, Lai Chi Kok, Hong Kong, China.
a Author for correspondence. Fax 852 23700969; e-mail cc16cdb{at}hkstar.com.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The reported toxic effects of THP include depression of neurological, respiratory, and cardiac function in pediatric poisonings (4), as well as acute or chronic hepatitis after regular use in adults (5)(6)(7). One fatal case linked to suicide by massive THP ingestion has been documented (8); however, co-ingestion with other drugs was also evident. The offending agent in those episodes was traced to a proprietary health product, "Jin Bu Huan Anodyne Tablets", which contains purified, concentrated THP. Consequently, THP was entered into the update of Poisindex®, an international toxicological database. The Jin Bu Huan Anodyne Tablets was thus banned by the US Food and Drug Administration; in other countries (e.g., Singapore), THP is classified as a controlled substance, and a license is required to sell it.
The published cases of THP toxicity have been supported mainly by clinical histories and product analyses; limited data are available concerning the positive identification of THP in body fluids from poisoned subjects. In one major case study (4)(5)(6), the analytical results were supplied only in brief. Gas chromatography–mass spectrometry (GC-MS) and nuclear magnetic resonance (NMR) characterized the products retrieved from patients in that study as highly concentrated LEVO-THP (25–28 mg/tablet) without the presence of other plant alkaloids. However, extensive toxicology screens reported negative findings in urine and serum collected promptly from two severely ill children after ingestion of Jin Bu Huan Anodyne Tablets (estimated to be equivalent to 500-1700 mg of THP).
Here, we define and validate the analytical parameters for the measurement of THP in standardized toxicology screens by HPLC with diode-array detection (HPLC-DAD) and GC-MS. We also review the analytical and clinical findings of several acutely intoxicated adults in which THP has been identified.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
materials
The chemicals and organic solvents for HPLC-DAD were described in
our previous publication (9). For GC-MS, ethyl acetate
(chromatographic grade) and anhydrous sodium sulfate (analytical grade)
were obtained from Sigma Chemical; the silylation reagent
N,O-bis(trimethylsilyl)trifluoroacetamide with 10
mL/L trimethylchlorosilane was purchased from AllTech Associates, Inc.
Quality-control urine "LyphochekTM toxicology" was obtained from Bio-Rad Laboratories, ECS. The THP used as the standard was a gift from the Chinese Medicinal Material Research Center, the Chinese University of Hong Kong; it was originally produced by the National Institute for the Control of Pharmaceutical and Biological Products, Ministry of Health, Beijing, China.
A reference pharmaceutical Jin Bu Huan Anodyne Tablets, which had been analyzed by three independent institutes and found to consistently contain 300 mg/g LEVO-THP and 700 mg/g starch (3)(6)(10), was obtained from Bose Drug Manufactory.
The proprietary herbal hypnotic labeled "Sleeping Pill", as declared by a patient, was obtained from a local health store. Tablets of another proprietary product, "Rohypson", were obtained from a partially emptied package from another patient.
hplc-dad screen
Our previously validated HPLC-DAD method (9) was used
routinely with slight modification for first-line toxicology
screening,. A model LC Star liquid chromatograph (Varian Chromatography
System) was equipped with a diode-array detector, an automated sampler,
a ternary pump, and a workstation. The analytical column of 150 x
4.6 mm with a replaceable guard cartridge was packed with 5 µm
base-deactivated Hypersil C-18 phase (Shandon HPLC). We used a gradient
elution generated by the proportional mixing of two solvents: solvent A
contained 50 mL/L acetonitrile, and solvent b contained 500 mL/L
acetonitrile, both in 50 mmol/L phosphate buffer, pH 3.0, containing 3
mL/L triethylamine and 375 mg/L sodium octyl sulfate. The gradient
conditions were as follows: solvent B increased from 12% to 100% in a
25-min linear gradient, was maintained at 12% for 3 min, and returned
to 12% in 3 min; the waiting time between analytical runs was 7 min.
The flow rate was 1.0 mL/min, and the column was operated at ambient
temperature. Drug identifications were achieved by computerized library
searches based on retention and spectral data from 330 records of
common drugs, poisons, and metabolites. Comparable spectral and
retention parameters were verified against the published database
(9).
The validated protocol for mixed-mode solid-phase extraction (9) was followed for the uniform isolation of drugs from serum and urine. However, specimens were diluted in a modified solution (0.4 mol/L phosphate buffer, pH 6.0, containing 2.5 mg/L pinazepam as an internal standard) before extraction. For sample introduction to the Varian HPLC by partial loop filling, 35 µL of the extract was injected.
gc-ms follow up
The GC-MS protocol was set up initially for urine metabolic
profiling (11). The original method used solvent extraction
at acidic pH and trimethylsilyl (TMS) derivatization, which
identified not only a comprehensive range of organic acids, but also a
number of acidic drugs. When the acidic extraction was replaced with
the commercially available extraction tube, Toxi-Tube A (i.e., basic
and neutral extraction), we could analyze unidentified substances from
the HPLC toxicology screen. Analyses were performed on a
Hewlett-Packard bench-top GC-MS system consisting of a 5890 Series II
Plus gas chromatograph, a 5972 mass selective detector, and a G1034C MS
Chemstation. An HP-5MS cross-linked 5% phenyl, methylsilicone
fused-silica capillary column (30 m x 0.25 mm i.d., 0.25 µm
film thickness) was used. The operative temperatures were as follows:
injector (splitless, 2 min), 250 °C; oven, 60 °C (1.5 min), with
a linear change at 8 °C/min to 300 °C (3 min); transfer line,
280 °C. The carrier gas was helium (1.0 mL/min). The mass detector
was operated in scan mode in the range 50–550 atomic mass units
(~1.5 s/scan). Library searches were based on the external databases
(also from Hewlett-Packard), which included the Wiley library
(>200 000 records of general compounds) and the Pfleger-Maurer-Weber
library (>4000 records of drugs, poisons, pesticides, and
metabolites).
A 4-mL urine sample was added to an extraction tube containing 20 µL of the internal standard pinazepam (0.5 g/L in methanol). The tube was extracted for 10 min by gentle inversion and then centrifuged, and the upper organic layer was evaporated. The dried residue was reconstituted with 100 µL of derivatization agent and incubated at 65 °C for 30 min. After cooling, the derivatized sample was diluted with 500 µL of dry ethyl acetate (stored over anhydrous sodium sulfate). The derivatized sample (1-µL ) was then injected into the GC-MS.
preparation of reference thp materials
Because of the limited supply of the official chemical standard,
we also made our own reference materials of THP from the Jin Bu Huan
Anodyne Tablets, using the following method: 60 Jin Bu Huan tablets
were dissolved in 50 mL of 900 mL/L ethanol at 50 °C. The warm
suspension was filtered, and the filtrate was allowed to stand
overnight at 4 °C. White, needle-like crystals were harvested and
dried in an oven at 50 °C. The dried crystals were reconstituted in
10 mL of 900 mL/L ethanol, and the crystallization step was repeated.
The refined crystals were dissolved in 700 mL/L acetonitrile at 1
g/L. The reference solution was calibrated by HPLC at 278 nm
against the official standard as 0.94 g/L THP (average of duplicates),
without impurities that absorb in the ultraviolet (UV) region. The
reference solution was added to the commercial urine control material
for validation of routine identification parameters. It was also
added to a drug-free serum and used to establish response factors
in semiquantitative analyses.
analysis of thp content in proprietary medications
Drug tablets were homogenized in 5 mL of 700 mL/L acetonitrile by
incubation at 37 °C for 15 min and sonication for 5 min. After
centrifugation, the clear supernatant was diluted accordingly in 700
mL/L acetonitrile for HPLC injection. The THP content was determined by
external standardization against the reference THP solution.
study of glucuronide metabolites
Urine from patients was subjected to glucuronidase enzyme
digestion, according to a standardized procedure (9) before
extractions for HPLC-DAD or GC-MS.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. The UV spectrum of THP (Fig. 1B
) is characterized by an
absorption band at 278 nm and a shoulder at ~225 nm; the chromophores
can be tracked to the two dimethoxy-benzoyl moieties in the substituted
isoquinoline molecule. Repeated HPLC analyses (n = 8) of the
THP-fortified control showed reproducible retention times (mean ±
SD, 16.6 ± 0.14 min) and detector response at 210 nm (mean
± SD, 395 ± 37 mA).
|
The mass spectrum of THP (Fig. 1C
) features a prominent molecular ion
(m/z 355) and a retro-Diels-Alder type cleavage
(12) (yielding the m/z 164 and 190 clusters
illustrated in Fig. 1D
); mass fragments produced by the neutral loss of
a methyl group and a methoxy group were also noteworthy. Under our
GC-MS conditions, the THP peak exhibited a retention index of 2968,
calibrated against that of a widely adopted database (13).
acute poisonings screened thp-positive by hplc
Thus, in a retrospective analysis of the HPLC-DAD raw data in the
16-month period, we identified nine cases of acute poisoning in which
THP was detected in urine and/or serum. Brief descriptions of the
clinical histories of those cases are summarized in Table 1
. On the basis of clinical judgment and circumstantial grounds,
the suspected agents were mostly unspecified hypnotics; herbal
preparations available over-the-counter were implicated in some cases.
During the same 16-month period, positive identification of one or more
drugs was found in 67% of all requests (n = 492).
Sedative-hypnotics were responsible for 37% of the confirmed
drug-related poisonings. Given the laboratory evidence of THP in body
fluids of those nine cases, acute THP poisoning is not uncommon in our
hospital settings.
|
clinical summary and product analyses
After initial central nervous system disturbances on
admission, all patients were responsive to conservative treatment and
were discharged within 1–3 days. A partially emptied package of
Rohypson (unidentified source, labeled "natural herbal for
sleeping") was retrieved from case 8. Case 7 reported the ingestion
of another proprietary product, Sleeping Pill (in the brand of "Kwei
Feng" from Kwang Si, China). Duplicated analyses of these two
products by HPLC confirmed pure THP in concentrated form: the Rohypson
averaged 26 mg of THP per tablet; the Sleeping Pill averaged 25 mg of
THP per tablet. Coincidentally, the appearance of both tablets was
almost identical (e.g., size, color, and weight) to the Jin Bu Huan
Anodyne Tablets. Given the recommended daily oral doses of THP as
60–480 mg (10), these two patients were apparently taken
excessive THP (i.e., case 8 was ingesting ~2000 mg, and case 7 was
ingesting ~1500 mg).
paired analyses of serum and urine
Serum THP was estimated by a response-factor established from a
THP-fortified serum by the HPLC method. Duplicated analyses at three
levels (0.5, 2.4, and 9.4 mg/L) showed a linear response (average, 40
mA per 1.0 mg/L at 210 nm). A detection limit of 0.1 mg/L (4 mA) was
calculated by extrapolation, corresponding to a signal-to-noise
ratio of 5. The serum concentrations of the five cases (cases 1, 2, 7,
8, and 9) were thus retrospectively estimated as <0.1–1.2 mg/L (Table 1
). Combined analyses of both urine and serum appeared to improve the
screening yield in cases of THP exposure. A typical THP overdose with
both serum and urine screened is represented in Fig. 2
. Additional peaks with UV spectra matching that of THP
were consistently eluted at earlier retention times; two major
peaks were highlighted as M1 and M2 in Fig. 2
. Very often, the relative
signal of THP to these two peaks was inverted from serum to urine
(exact values listed in Table 1
); thus the two peaks were tentatively
identified as polar metabolites of THP.
|
follow-up analyses of urine by gc-ms
Urine samples from four cases of THP poisoning were subsequently
tested by the GC-MS procedure. Concordant identifications of THP were
achieved in case 5, in which excessive THP was found by both screens.
GC-MS failed to detect THP in two urine samples established by HPLC-DAD
(cases 7 and 9). No THP could be detected in the urine from case 8 by
both methods, whereas THP was detected in the patient's serum by HPLC.
In urine samples (e.g., from cases 7, 8, and 9), GC-MS detected two
additional peaks other than THP, albeit at variable amounts (Fig. 3
: PK1 and PK2). On the basis of the fragmentation patterns in
mass spectrometry (Fig. 4
), the two peaks were tentatively identified as TMS derivatives
of O-desmethylated metabolites of THP. However, additional validation
was not possible because reference materials for these O-desmethylated
derivatives of THP were not available.
|
|
investigation of presumptive metabolites of thp by glucuronidase
digestion
The effect of glucuronidase treatment on the urine THP profiles by
GC-MS and HPLC-DAD provided more clues to THP metabolism. In the GC-MS
profile, the tentatively identified O-desmethylated metabolites of THP
were much enhanced by glucuronidase treatment. Enzyme digestion
also yielded an additional but small peak tentatively identified as
di-O-desmethylated THP-di-TMS derivatives (Fig. 4
). In the HPLC screen,
M1 and M2 were digested by glucuronidase, producing additional peaks
(labeled as M3 and M4 in Fig. 5
). The UV spectra of M3 and M4 were in better agreement with THP
than those of M1 and M2.
|
Taken together, in the glucuronidase-digested urine, both GC-MS and HPLC-DAD were probably measuring the same phenolic metabolites of THP. Importantly, HPLC-DAD with mixed-mode solid-phase extraction at pH 6 detected intact glucuronides, which were not extracted under the alkaline condition in the sample clean-up for GC-MS.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We observed rapid metabolism and relatively low serum concentration of THP in most intoxicated subjects. Our enzyme digestion experiment outlined a possible path of THP metabolism by O-desmethylation and subsequent conjugation of the exposed phenolic groups with glucuronate. The disposition and metabolism of THP in humans has not been studied (2). In animal experiments, oral doses of THP were rapidly and almost completely absorbed; THP penetrated the blood-brain barrier easily and tended to be concentrated in the fat, tissues, and various organs. Urinary excretion appeared inconsistent among different animal species; recovery of administrated doses was 1/300 from rabbits (intraperitoneal injection) but >80% from rats (subcutaneous injection). No metabolite identification was attempted in those few studies (2). Although we have preliminary data on the metabolic patterns relevant for quick toxicological screening, the information is not complete because of the limitations imposed by the methods, subjects, and dosing in the settings of emergency poisoning. Proper investigation should involve controlled human studies with measurements capable of optical isomer quantification.
Our case series suggests no evidence of severe toxicity in those transient and benign poisonings, although we are cautious about the missing clinical details in this retrospective study. The finding of quick recoveries in most THP-poisoned patients was consistent with the observed rapid onset but short-lived action of THP in animals (2). Additional prospective studies, with laboratory support by the established toxicology screens described above, will better define the relationship between the dose, the blood concentration, and the toxicity of THP in acute poisonings. The need for such investigation is pressing because we also demonstrated that additional over-the-counter products with concentrated THP were involved. This potent alkaloid could cause major health effects at excessive or prolonged doses (4)(5)(6)(7)(8).
In conclusion, our study provides metabolic evidence of THP involvement in a number of unrelated adult cases of acute intoxication. The practical procedures described here support prospective assessment of the toxicity risks and safety margins for this neuroactive alkaloid.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  F. K. Li, C.-K. Lai, W. T. Poon, A. Y. W. Chan, K. W. Chan, K. C. Tse, T. M. Chan, and K. N. Lai Aggravation of non-steroidal anti-inflammatory drug-induced hepatitis and acute renal failure by slimming drug containing anthraquinones Nephrol. Dial. Transplant., July 1, 2004; 19(7): 1916 - 1917. [Full Text] [PDF]
|
|
|
|
 |
|
 K K Lau, C K Lai, and A Y. Chan Phenytoin poisoning after using Chinese proprietary medicines Human and Experimental Toxicology, July 1, 2000; 19(7): 385 - 386. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
