Scientific and technical
A case of fatal poisoning with the aconite plant:
quantitative analysis in biological fluid
Regional Laboratory for Toxicology, City Hospital N.H.S. Teaching Trust, Dudley Road. Birmingham 818 7QH, United Kingdom
Science & Justice 2002 42 11 1 - 115 Received 16 January 2002 accepted 23 April 2002
In recent years recorded cases of plant poisoning have become
rare, this may in part be due to the possibility of plant ingestion
not being indicated at the beginning of an investigation.
Aconitum napellus (aconite, Wolfsbane, Monkshood) is one of
the most poisonous plants in the UK. It contains various potent
alkaloids such as aconitine, isoaconitine, lycaconitine and
napelline. Ingestion of Aconitum plant extracts can result in
severe, potentially fatal toxic effects. This paper describes the
analytical findings in a recent death in the UK. resulting from
deliberate ingestion of Aconitum napellus extract. The
concentrations of aconitine measured by HPLC-DAD in the post
mortem femoral blood and urine were 10.8 pg/L and 264 p g k ,
respectively. The aconitine concentration in the ante mortem
urine was 334 pg/L and was estimated to be 6 pg/L in the ante
mortem serum. Hence, accidental, suicidal or homicidal
poisoning due to the ingestion of plant material remains a
possibility and should be borne in mind when investigating
sudden or unexplained death.
Ces dernieres annCes, le nombre de cas d'empoisonnement par
les plantes est devenu rare. Cela peut &tredQen partie au fait que
la possibilitC de I'ingestion de plantes n'a pas CtC indiquCe au
dCbut de I'investigation. Aconitum napellus (aconite) est I'une
des plantes les plus vCnCneuses du Royaume-Uni. Elle contient
plusieurs alcaloi'des puissants, tels que I'aconitine,
I'isoaconitine, la lycaconitine et la napeline. L'ingestion
d'extraits de plante Aconitum peut dCboucher sur des effets
toxiques graves, potentiellement mortels. Cet article dCcrit les
rksultats analytiques dans un cas de mort au Royaume-Uni,
provoquC par la consommation dClibCrCe d'extrait d'Aconitum
napellus. La concentration d'aconitine mesurke par HPLC-DAD
dans le sang fCmoral postmortem et I'urine Ctait de 10.8 pg/L et
de 264 pg/L respectivement. La concentration d'aconitine dans
I'urine antemortem Ctait de 334 yg/L et a CtC estimCe a 6 pg/l
dans le sCrum antemortem. C'est pourquoi I'empoisonnement
accidentel, suicidaire ou homicidaire par la consommation de
matCriel de plante reste une possibilitC et doit &tre toujours a
l'esprit lors de I'investigation d'une mort soudaine ou
inexpliquCe.
sciencetkjtlstice Volume 42 No.2 (2002) 111 - 115
SP Elliott
In den letzten Jahren sind Berichte von Falle uber Vergiftungen
durch Pflanzen selten geworden. Dies mag teilweise daran
liegen, dass es zu Beginn einer Untersuchung keinen Hinweis
auf pflanzliches Materials gegeben hat. Aconitum napellus
(Aconitum, Eisenhut, Monchskappe) ist eine der giftigsten
Pflanzen in England. Sie enthalt mehrere stark wirksame
Alkaloide wie Aconitin, Isoaconitin, Lycaconitin und Napellin.
Die Einnahme von Aconitin-Pflanzenextracten kann zu
schweren, moglicherweise todlichen toxischen Effekten fuhren.
Diese Arbeit beschreibt die analytischen Ergebnisse in einem
Todesfall der kurzlich in England nach der absichtlichen
Einnahme eines Aconitum Napellus Extracts untersucht wurde.
Die mittels HPLC-DAD in femoralem Blut und in Urin nach
Todeseintritt gemessenen Aconitin Konzentrationen lagen bei
10.8 pg/L respektive 264 pg/L. Der Wert im Serum vor
Todeseintritt wurde auf 6 pg/L geschatzt. Bei der Untersuchung
eines plotzlich eingetretenen Todes oder einer ungeklaen
Todesursache sollte immer auch die Moglichkeit der Vergiftung
durch die Aufnahme von Pflanzenmaterial - sei es zufallig,
selbstmorderisch oder durch Mord - im Auge behalten werden.
En la actualidad 10s casos de enevenenamiento con plantas son
raros, esto puede ser en parte a la posibilidad de que la ingestion
de plantas no es indicada en el inicio de las investigaciones.
Aconitum napellus (aconite, wolfsbane, Monkshood) es una de
las plantas mas venenosas en el Reino Unido. Contiene vkios
alcaloides potentes como la aconitina, la isoaconitina, la
licaconitina e la napellina. La ingestion de extractos de Aconitum
pueden resultar en casos severos y fatales de intoxicacion. Este
trabajo describe 10s resultados analiticos de un caso de muerte
reciente por ingestion deliberada de extract0 de Aconitum
napellus en el Reino Unido. Las concentraciones de aconitina
medidas por HPLC-DAD en la sangre femoral e en orina fue
10.8 pg/L y 264 yg/L respectivamente. La concentracion de
aconitina en la orina antemorten fue 334 pg/L y fue estimada
como 6 pg/L en el suero antemorten. Como causas de
accidentes, suicidios u homicidios, la ingesti6n de plantas es una
posibilidad que debe sospecharse en la investigacidn de muerte
slibita o inexplicada.
O The Forensic Science Society 2002
Key words Forensic science, aconite, HPLC-DAD,
alkaloids, aconitine, plant poisoning.
Page 111
SP Elliott
A case of fatal poisoning with the aconite plant: quantitative analysis in biological fluid
Introduction
For many thousands of years the toxicological properties of
particular plants have been exploited for both medicinal and
sometimes criminal purposes [I]. However, most cases of
poisoning typically involve pharmaceuticals or drugs of abuse
and routine toxicological analyses do not generally detect plant
alkaloids. Therefore, in recent years recorded cases of plant
poisoning have become a rare occurrence and instances may be
missed unless sufficient case history or scene evidence is
available.
Aconitum napellus (also known as aconite, Wolfsbane,
Monkshood, etc) is native throughout Europe. It is a perennial
plant and can grow up to 1 metre high with light green leaves and
blue, mauve or white flowers [2]. Although now seldom found
wild, it is thought to be one of the most poisonous plants in the
United Kingdom and is still cultivated in domestic gardens [2].
It is also related to the other Aconitum plants of the
Ranunculaceae family that are widely distributed across North
America and northern Asia [3]. Various Aconitum species are
also used in Chinese herbal products mainly for the treatment of
musculoskeletal disorders, however, several cases of poisoning
have been reported [4-71.
The Aconitum plants contain various potent neurotoxic
alkaloids, in particular the polycyclic diterpene alkaloid
aconitine (Figure 1). The alkaloids are contained in all Aconitum
plant tissue but particularly in the root. It has been reported that
the qualitative and quantitative presence of alkaloids depends on
the species of plant and ecological factors (i.e. a particular
geographical region may have a particular alkaloid profile [3]. In
Japan, Aconitum japonicum grown in the Hokkaido region
contains jesaconitine as its main alkaloid with additional
alkaloids present (including aconitine) [3]. In Europe, Aconitum
napellus contains aconitine in addition to other alkaloids such as
isoaconitine, lycaconitine and napelline [2]. Aconitine can cause
conduction block at voltage-sensitive sodium channels in the
exons [8]. This may result in various toxic symptoms including
burning of the mouth, abdominal pains, diarrhoea, vomiting,
headache, delirium, coldness, reduced heart rate, paralysis,
convulsions, coma and respiratory arrest [2,7]. If sodium
channels in the heart are affected, potentially fatal ventricular
tachycardia and premature tachycardia can occur [3].
Due to their potency, only small amounts of aconitine and related
alkaloids are required to produce toxic symptoms. Furthermore,
it has been found that aconitine is relatively unstable and
undergoes non-enzymatic hydrolysis [8]. It has been reported
that the hydrolysis rate increases at temperatures above freezing
(e.g. 25C and 37C) and at pH greater than 7 [9]. In addition to
in vitro instability, both hydrolysis and de-esterification occur
metabolically in vivo [9]. Hence, only trace amounts of aconitine
may be detected in biological fluid following ingestion of
Aconitum plant extracts [3,9]. In two reported cases of poisoning
following ingestion of the Aconitum plant found in Japan, gas
chromatography with mass spectrometry (GC-MS) analysis
revealed that the greatest concentration of alkaloids occurred in
the urine [3,9]. In a non-fatal case, an aconitine serum
Page 112
concentration of 1.7 yg/L was attained shortly after ingestion
[9]. In a fatal case, a post mortem aconitine blood concentration
of 1.1 pgIL was measured [3].
Detection of such alkaloids has been achieved using both
GC-MS (following BSTFA derivatisation) and high
performance liquid chromatography (HPLC) with UV detection
and mass spectrometry [10,11].
This paper describes the analytical findings in a recent death in
the UK resulting from deliberate ingestion of Aconitum napellus
extract. Aconitine analysis was performed using high
performance liquid chromatography with diode array UV
detection (HPLC-DAD).
Case History
A 60 year old male with personal problems had produced an
aconite extract for oral ingestion by liquidising tissue from
Aconitum napellus (possibly self-cultivated). Shortly afterwards
he became agitated, nauseous and began to vomit violently. An
ambulance was called and he was taken to a nearby hospital. He
remained conscious for a while and informed the staff of the
ingestion of aconite and his intentions. He was treated
supportively but died following cardio-respiratory arrest,
approximately two hours after ingestion. A post-mortem
examination one day after death revealed no evidence of natural
disease.
Various plant tissues and books pertaining to plant
pathology/toxicology were found at the original scene.
Specimens of blood/serum and urine obtained at post mortem
and taken upon admission (ante mortem), along with a sample of
the apparent aconite extract (liquid), were submitted for
toxicological analysis.
Experimental
Chromatographic equipment
HPLC-DAD analysis was performed using a P580 low pressure
pump, STH 585 column oven, ASI-100 autosampler and a
UVD340S diode array detector all from Dionex (Camberley,
UK). A Waters Spherisorb SSODICN 4.6 mm x 150 mm
cartridge column (Elstree, UK), protected by a 4 mm x 10 mm
guard column of Spherisorb S50DS2 was used for the analysis.
Data acquisition was handled by a Dionex Chromeleon software
package with the diode-array detector recording spectral data
between 200 nm and 595 nm.
Materials
Triethylammonium phosphate buffer (TEAP, 1.OM, pH 3.0) was
supplied by Fluka, Dorset, UK, the HPLC-grade acetonitrile was
supplied by Rathburns Chemicals Ltd, Walkerburn, UK. The
HPLC-grade l-chlorobutane was obtained from Fisher Scientific
International, Loughborough, UK. Clomipramine was kindly
donated by Novartis Pharmaceuticals (Camberley, UK). Pure
aconitine was supplied by Sigma-Aldrich, Dorset, U.K. and was
used to prepare reference and calibration standards for the
formal identification and quantitation of aconitine in the plant
extract and specimens analysed. 12.5, 25, 50, 100 and 200 yg/L
science&justice Volume 42 No.2 (2002) 111 - 115

A case of fatal poisoning with the aconite plant: quantitative analysis in biological fluid
aconitine calibration standards were prepared in blank equine
plasma.
Extraction methods for biological specimens
For HPLC-DAD analysis, 3 ml of 500 ygll clomipramine
internal standard (in 0.2M Na2C03 solution, pH 10) was added
to 3 ml of samplelstandard followed by 5 ml of 1-chlorobutane
extraction solvent in a polypropylene 12 ml tube. After three
minutes mechanical shaking and centrifugation at 4500rpm for
three minutes, the upper solvent layer was transferred to a
second tube. A further 5 ml of 1-chlorobutane was added to the
first tube followed by three minutes shaking and centrifugation
at 4500rpm for three minutes. The upper solvent layer was
transferred to the second tube (10 ml solvent in all). Drugs were
extracted into 100 y1 of added 0.05M H2S04 and after three
minutes shaking and centrifugation at 3500rpm for three
minutes, the upper solvent layer was aspirated and 100 yL of the
acid layer was transferred into a vial for injection. The injection
volume was 50 yl.
For analysis of the plant extract no sample extraction was
performed. 2 ml of liquid was centrifuged at 4500 rpm for
Sfiveminutes. 100 yL of the supernatant was transferred into a
vial for injection. The injection volume was 5 p1.
Chromatography conditions
For HPLC-DAD screening, a previously described method was
used based on single step gradient elution (0-70% acetonitrile in
15 minutes, holding at 70% acetonitrile for three minutes) [12].
The quantitation procedure was based on 40% acetonitrile
isocratic elution conditions at a flow rate of 2 mllmin.
Results
Qualitative analyses
Prior to sample analysis, pure aconitine reference material was
used to obtain HPLC-DAD analytical data, including retention
Figure 1 Chemical structure of aconitine.
science&justice volume 4 2 No.2 (2002) 111 - 1 1 5
SP Elliott
index value (RI=352) and UV spectral data (UV maxima = 233
and 274 nm).
General drug screening for common pharmaceuticals and drugs
of abuse using gas chromatography with nitrogen phosphorus
detection (GC-NPD), enzyme immunoassay (EMITII) and
HPLC-DAD analysis of the post mortem and ante mortem
bloodserum and urine specimens detected only aconitine. Gas
chromatography with mass-spectrometry (GC-MS) analysis was
not available.
HPLC-DAD analysis of the liquid plant extract also detected
aconitine. No pure reference standards were available for
definitive analysis of any other alkaloids present.
Using HPLC-DAD, aconitine (RI=352) was identified by
retention index value (RI) and by UV spectral matching (Figure
2) [12,13].
Quantitative analyses
Analysis for alcohols by headspace gas chromatography with
flame ionisation detection (GC-FID) detected an ethanol
concentration of 166 mgldl in the ante mortem serum.
Using 40% acetonitrile (and 60% 25mM TEAP buffer) isocratic
conditions, aconitine eluted at 3.78 minutes and the 500 pgll
clomipramine internal standard eluted at 4.74 minutes (Figure
3a). Clomipramine was chosen as the internal standard based on
the extraction behaviour of the drug and its favourable elution
characteristics. For optimum sensitivity a 233 nm detection
wavelength was used (the limit of detection was determined to
be approximately 2 ygll). A linear calibration curve was
produced from the 12.5, 25, 50, 100 and 200 ygll aconitine
standards (Figure 3b). Post mortem femoral blood and urine
samples in addition to an ante mortem urine sample were
analysed. Due to the expected higher concentrations compared
Figure 2 Spectral matching of aconitine library entry and
aconitine detected in the case post mortem
blood (99.1% match).
Page 113
SP Elliott
A case of fatal poisoning with the aconite plant: quantitative analysis in
to blood, urine specimens were analysed with and without two-
fold dilution in blank equine plasma. Unfortunately, there was
insufficient volume to accurately measure aconitine in the ante
mortem serum.
The mean aconitine concentrations measured in the post mortem
femoral blood and urine were 10.8 ygll and 264 ygll,
respectively. The aconitine concentration in the ante mortem
urine was 334 pgll and was estimated to be 6 ygll in the serum.
However, due to the unstable nature of the compound the
concentrations of aconitine may have been even higher.
Additional stability study
Previously, Mizugaki et a1 have published information regarding
the extensive in vitro production of benzoylaconine and aconine
from aconitine in pH 7.4 buffer at 37C with time (91. Hence,
further study was performed to assess the presence of any
aconitine related products resulting from in vitro instability. 20
mL of a 50 mgll aconitine standard solution in deionised water
was analysed under isocratic conditions (40% acetonitrile, 60%
buffer) including clomipramine as an internal standard. Figure
4a shows the resultant chromatogram with aconitine eluting at
3.8 mins and clomipramine eluting at 4.7 mins (time 0). The
aconitinelclomipramine solution was then incubated at 37C for
3 days and 20 yL injected on to the column and analysed as
before. It can be seen in Figure 4b that the aconitine peak height
was significantly reduced but the clomipramine internal standard
peak height was identical to that obtained at time 0. Additional
peaks were observed at 1.8, 2.0 and 2.9 mins (Associate peak 1,
2, and 3 respectively). Unfortunately, as no pure reference
standard for additional aconitine related alkaloids were
available, definitive identification of these compounds was not
possible. However, as the UV spectra were very similar to that
of aconitine it appears likely they are aconitine-associated
compounds. Associates 1 and 2 were the predominant peaks
observed. Comparison with time 0 indicates that Associates 2
and 3 were present in the solution albeit at much lower
concentrations. Comparison with the ante mortem and post
mortem samples did not detect any of these Associated
Figure 3 (a) Chromatogram showing elution of aconitine and clomipramine (internal standard) under 40% acetonitrile
isocratic conditions. Samples were extracted using 1-chlorobutane and back extracted into 100 pl of 0.05M
H2S04. Injection volume was 50 PI. (b) Linear calibration curve obtained during quantitative analysis.
233nm
Clomipramine
Solvent (ISTD)
- front
5.0 -
I I
Y
Awnitine
& I
Page 114
biological fluid
compounds thus it can be concluded that the samples analysed
had undergone little in vitro instability, if at all, prior to analysis.
The study also showed that such instability products do not
interfere with the HPLC-DAD method described.
Discussion
Although not as sensitive as previous reported GC-MS methods
[9], the HPLC-DAD method presented in this paper has been
successfully applied for the qualitative and quantitative analysis
of aconitine in human biological fluid.
A review by Chan of aconitine poisoning between 1955 and
1994 showed a total of 30 reported deaths, with 23 occurring in
China [4]. However, there are very little published data
regarding aconitine concentrations measured in biological fluid.
Two recent papers presented extensive data for many Japanese
Aconitum plant alkaloids [3,9]. In these non-fatal and fatal cases,
although jesaconitine was determined to be the main alkaloid
present, aconitine was also detected and concentrations of
1.7 ygI1 and 1.1 ygll were measured in the ante mortem serum
and post mortem blood, respectively. In both instances poisoning
was attributed to aconite ingestion. In the case presented here
there was overwhelming clinical and circumstantial evidence of
Aconitum napellus intoxication. As this aconite species has a
different alkaloid profile to that ingested in the Japanese cases, it
was not appropriate to compare aconitine concentrations.
Nevertheless, based on published information, the
concentrations measured (particularly in the post mortem
femoral blood; 10.8 ygn) supported aconite poisoning leading to
cardiorespiratory arrest as the cause of death. The additional
presence of ethanol (166 mgldl) may have also exacerbated
toxicity.
Plant alkaloids are not usually identified during routine
toxicology screening. However, as in this case, at high
concentrations aconitine may be detectable during routine
screening for chemically basic drugs, particularly when using
HPLC-DAD or GC-MS. Due to the vast numbers of possible
alkaloids, difficulty in obtaining pure analytical standards and
science&justice Volume 42 No.2 (2002) 111 - 115
 SP Elliott
A case of fatal poisoning with the aconite plant: quantitative analysis in biological fluid

the lack of comparative cases it is difficult to compile and
maintain relevant data. The only indication for potential plant
poisoning will therefore probably be based upon evidence at the
scene or witness statements. Provision of material in such cases
is therefore vital in order to determine the relevant toxicological
analyses required.
Conclusions
Aconitum napellus (Wolfsbane, aconite) is one of the most
poisonous plants in the UK. This paper presented a case of fatal
aconite poisoning in which the Coroner recorded a verdict of
suicide. The high aconitine concentrations measured in blood
and urine were consistent with previous published data involving
aconite intoxication [3,9].
In recent decades, reported instances of plant poisoning have
become rare. Nevertheless, accidental, suicidal or homicidal
poisoning due to the ingestion of plant material and/or extracts
remains a possibility and should be borne in mind when
investigating sudden or unexplained death.
Acknowledgements
The author would like to thank Mr Bernard Smith, Dr Robin
Braithwaite and Dr Stephen George for their assistance and
discussions during this investigation.
Toxicology 1994; 36(5): 452-455.
Dickens P, Tai TT, But PP, Tomlinson B, Ng HK, Yan KW. Fatal accidental
aconitine poisoning following ingestion of Chinese herbal medcine: a report of two
cases. Forensic Science International 1994; 67: 55-58.
Chan N,Tomlinson B, Critchley JA. Aconitine poisoning follow~ngthe ingestion of
Chinese herbal medicines: a report of eight cases. Australian and New Zealand
Journal of Medicine 1993; 23: 268-271.
Muroi M, Kimura I and Kimura M. Blocking effects of hypaconitine and aconitine
on nerve action potentials in phren~c nerve-diaphragm muscles of mice.
Neuropharmacology 1990; 29: 567-572.
Mizugaki M, Ito K, Ohyama Y, Konishi Y, Tanaka S, Kurasawa K. Quantitative
analys~sof Aconltum alkalo~dsIn the urine and serum of a male attempting su~c~de
by
oral intake of aconite extract. Journal of Analytical Toxicology 1998; 22: 33e-340.
Hikino H et al. Determination of Aconitum alkaloids by high performance lkquid
chromatography. Journal of Chromatography 1981; 21 1: 123-128.
Ohta H, Seto Y and Tsunoda N. Determination of Aconitum alkaloids In blood and
urlne samples. High performance liquid chromatographic separation, sol~dphase
extraction and mass spectrometric confirmation.Journal of Chromatography B 1997;
691: 351-356.
Elliott SP and Hale KA. Development of a HPLC retent~onindex scale for
toxicological drug screening. Journal of Chromatography B 1997; 694: 99-1 14.
Elliott SP and Hale KA. Appl~cations
of a HPLC-DAD drug screening system based
on retention indices and UV spectra. Journal of Analytical Toxicology 1998; 22:
28&290.

References
1 Fiddes FS. Po~sonlngby aconitine: report of two cases. British Medical Journal
1958; 2: 779-780.
2 Cooper MR and Johnson AW. Poisonous Plants and Fungi in Britain. Second
Edition. Nowvich: Stationery Office, 1998; 188-1 89.
3 Ito K, Tanaka S, Funayama M and Mizugaki M. Distribution of Aconitum alkaloids
in body flu~dsand tissues in a suic~dalcase of aconite ingest~on.
Journal of Analytical
Toxicology 2000; 24: 348353.
4 Chan N. Aconitine poisoning: a global perspective. Veterinary and Human
Toxicology 1994; 36(4):32-28,
5 Chan TY, Tomlinson B, Tse LK, Chan JC, Chan WW, Critchley JA. Aconitine
poisoning due to Chinese herbal medic~nes:a review. Veterinary and Human

Figure 4 Chromatograms showing elution of 50 mg/L aqueous aconitine solution and clomipramine (internal standard)
under 40% acetonitrile isocratic conditions. Aconitine-associated instability compounds are also indicated.
Following time 0 analysis (Figure a), the solution was incubated at 37C for 3 days (Figure b), Injection volume
was 20 PI.

I
Assoc 2

Assoc. 1

20.0
Clomipramine
Assoc. 3 (ISTD)

-10 4
0.00
, , , ,
1.OO
, , , ,
2.00
, , , ,
3.00
, , ,
4.00
, , , , ,
5.00
, , rnin
,
6.00

science&justicc volume 42 No.2 (2002) 11 1 - 115 Page 115