Forensic Science International, 28 (1985) 109-113 109

Elsevier Scientific Publishers Ireland Ltd.

A FATAL POISONING WITH LSD

R.R. FYSH, M .C.H. OON, K.N. ROBINSON, R.N. SM ITH, P.C. W HITE and
M .J. W HITEHOUSE

Metropolitan Police Forensic Science Laboratory, 109 Lambeth Road, London SE1 7LP
(U.K.)

(Received January 3,1985)

(Revision received February 28, 1985)
(A ccepted February 28,1985)

Summary

Radioimmunoassay, high-performance liquid chromatography and capillary gas chro-

matography-mass spectrometry were used to detect and measure LSD in the first reported
case of fatal poisoning by LSD. The levels found in ante-mortem serum and plasma and
in post-mortem blood, liver blood and stomach contents are given.

Key words: Fatal LSD poisoning; Toxicological analysis of LSD; A nte- and post-mortem
LSD levels.

Introduction

Fatalities associated with lysergide (LSD) are usually due to injuries

received whilst under the influence of the drug. In a recent case, however,
a 25-year-old male died 16 h after being admitted to hospital, and a
Coroner’s enquiry concluded, on the basis of the medical and toxicological
evidence, that the actual cause of death was poisoning by LSD. This paper
describes the toxicological aspects of the case in which LSD was analysed
in ante- and post-mortem samples by various techniques.
Ante-mortem serum and plasma and post-mortem blood, liver blood and
stomach contents were available for toxicological examination. The samples
were analysed quantitatively for LSD by radioimmunoassay (RIA). High-
performance liquid chromatography (HPLC) and capillary gas chromato-
graphy-mass spectrometry (capillary GC-MS) were used to confirm the
presence of LSD and to verify some of the quantitative results.
Routine toxicological examination of the liver was carried out elsewhere
prior to submission of the samples to this laboratory; no drugs were detect-
ed. RIA of the ante-mortem serum in this laboratory demonstrated the
absence of amphetamine [l] and cannabinoids [2], thus LSD was the only
drug detected in the case.

0379-0738/85/$03.30
o 1985 Elsevier Scientific Publishers Ireland Ltd.
Printed and Published in Ireland
110

Materials and Methods

1. Sample preparation
Samples (100~~1) for RIA were vortexed with methanol (300+1), allowed
to stand for 15-30 min and centrifuged for 2 min (12,000 g). Aliquots
(200+1) of the methanolic supernatants were evaporated to dryness in
silanised glass tubes under a stream of air at room temperature and taken
up in 0.05 M barbitone buffer (200 ~1) (pH 8.6) containing 0.1% w/ v bovine
serum albumin and 0.1% w/ v sodium azide.
Plasma (1.0 ml) for HPLC analysis was mixed with 2 N ammonium
hydroxide (100 ~1) and extracted with diethyl ether (2 X 3 ml). The ex-
tracts were pooled, evaporated to dryness at 60°C under nitrogen in dark-
ness and reconstituted in HPLC eluent (150 ~1).
Stomach contents (4.0 g) for HPLC and capillary GC-MS analysis were
mixed with concentrated ammonium hydroxide (1.0 ml) and extracted
with ether (2 X 3 ml). The extracts were pooled, evaporated to dryness
at 60°C under nitrogen in darkness, reconstituted in 0.1 N hydrochloric
acid (100 ~1) and washed with ether (2 X 200 ~1). The acid solution was
made alkaline with concentrated ammonium hydroxide (100 ~1) and ex-
tracted with ether (2 X 3 ml). The extracts were pooled, evaporated to
dryness at 60°C under nitrogen in darkness and reconstituted in HPLC
eluent (2 ml). This reconstituted extract was analysed by HPLC. For capil-
lary GC-MS analysis, the extract was evaporated to dryness at 60°C under
nitrogen in darkness and reconstituted in water (200 ~1). The pH was ad-
justed to 10 with concentrated ammonium hydroxide and the solution
was extracted with ether (2 X 3 ml). The extracts were combined, evapo-
rated to dryness at 60°C under nitrogen in darkness and reconstituted
in methanol (20 ~1).

2. Radioimmunoassay (RIA)
The reconstituted extracts were assayed for LSD by the method of
Ratcliffe et al. [ 31. The antiserum (provided by courtesy of the Central
Research Establishment, Home Office Forensic Science Service) was raised
in sheep using an LSD-bovine serum albumin conjugate in which the indolic
nitrogen of LSD was linked to the protein.
The recovery of LSD from blood samples spiked at 25 ng/ ml was found
to be 88% with a coefficient of variation of 14.5% (n = 10). The detection
limit of the assay was 2.8 ng/ ml (mean response of 24 blank blood extracts
plus 3 standard deviations, multiplied by 4 to allow for dilution of the
blood samples in preparing the extracts).

3. High- performance liquid chromatography (HPLC)

HPLC was carried out on a 24 cm X 4.9 mm i.d. column of Spherisorb
S5W silica (Phase Separations, Queensferry, Clwyd). The eluent was 0.01 M
ammonium perchlorate in HPLC-grade methanol adjusted to pH 6.7 with
 111

0.1 M sodium hydroxide in methanol. The flow-rate was 1.0 ml/ min. A
Fluoromonitor III fluorescence detector fitted with a zinc lamp was used
(Laboratory Data Control, Stone, Staffordshire). The excitation wavelength
was 308 nm and the emission was measured through a 370-700-nm filter.
Injections were made under continuous flow conditions via an injection
valve (Negretti and Zambra, Southampton, Hampshire) fitted with a 25+1
loop.

4. Capillary GC-M S
The stomach contents extract was analysed by capillary GC-MS using a
Carlo Erba Fractovap Series 4160 GC (Erba Science, Swindon, Wiltshire)
coupled to a VG Micromass 12-12F quadrupole MS (VG Masslab, Altrin-
cham, Cheshire). The GC column was a 12-m bonded phase BP-l capillary
(SGE, Milton Keynes, Buckinghamshire) which led directly into the ion
source of the MS. The carrier gas was helium at an inlet pressure of 1.3
kg/ cm2. The injector temperature was 250°C. Splitless injections (0.5 ~1)
were made and the injection port was vented after 50 s. The column was
held at 100°C for 1 min after injection and was then raised to 280°C at
approximately 40” C/ min. The MS was operated in the electron impact
mode with an accelerating voltage of 70 eV. The interface temperature
was 250°C and the source temperature was 200°C. Multiple ion detection
was used to monitor LSD ions of m/ e 323 (molecular ion) and 181.

Results and discussion

LSD was detected by RIA in all the samples. The levels found are given
in Table 1 and may include a response to cross-reacting metabolites in
addition to unchanged LSD.
Ante-mortem plasma (Fig. 1) and stomach contents extracts were ana-
lysed for LSD by HPLC. A peak with a retention identical to that of authen-
tic LSD was detected in each extract. Comparison of the peak heights with
those of known concentrations of LSD gave the levels shown in Table 1.

TABLE 1

LSD LEVELS FOUND BY RIA AND HPLC

-
Sample LSD (n&ml) found by:

RIA HPLC
--
Ante-mortem serum 14.4 -
Ante-mortem plasma 14.8 8
Post-mortem blood 4.8 -
Stomach contents 55.2 60
Liver blood 7.2 -
112

LSD

pi
I N J ECT I ON

14
-.-
-----r--_

10

M I N U T ES
‘II
6
I

27

Fig. 1. HPLC of ante-mortem plasma extract. Conditions as described in the text; 25 ~1

injected. The unlabelled peak (solvent front) is typical of that observed with control
samples.

The results compared reasonably well with those obtained by RIA. Co-
extractives or insufficient sample precluded HPLC analysis of the other
samples.
The stomach contents extract was analysed by capillary GC-MS. Ions
of m/e 323 and 181 were monitored by multiple ion detection and gave
peaks with a retention of 10.0 min. Identical results were obtained with
authentic LSD. The other samples were not examined by capillary CC-MS.
It is difficult to assess the significance of the results in Table 1 since no
fatal levels of LSD in humans are given in the literature. The highest plasma
level of LSD previously found [4] was 9.5 ng/ml, and was the mean plasma
level of 5 subjects 5 min after intravenous injection of 2 pg LSD/kg. Indi-
vidual levels were not given. In the same study, the half-life of LSD in
plasma was calculated to be 175 min. In another study [5], 13 subjects
were each given 160 pug LSD orally and plasma levels of LSD were measured
at intervals up to 5 h. The highest level found in any subject was 8.8 ng/ml
after 130 min. The other plasma levels in these studies were generally much
lower, as were the plasma levels found in other cases of LSD ingestion
[3,6,7]. In one previous case in this laboratory (unpublished results), we
 113

found LSD in both blood and stomach contents at levels of 7.6 and 11.6
ng/ml, respectively. In a further five cases in which we found LSD in
stomach contents, the blood levels were below the RIA detection limit
(2.8 ng/ml). In four cases in which LSD was detected in blood, but stomach
contents were not available for analysis, the levels ranged from 4.4 to 8.0
ng/ml. The levels found in the ante-mortem samples in the present case
(Table 1) are thus nearly double the highest levels previously found in
blood.

References

R.N. Smith, Immunoassays in forensic toxicology. In J.S. Oliver (ed.), Forensic

Toxicology, Croom Helm, London, 1080, pp. 34-47.
B. Law, P.A. Mason, A.C. Moffat and L.J. King, A novel ‘zSI-radioimmunoassay
for the analysis of a9-tetrahydrocannabinol and its metabolites in human body
fluids. J. Anal. Toxicol., 8 (1984) 14-18.
W.A. Ratcliffe, SM. Fletcher, A.C. Moffat, J.G. Ratcliffe, W.A. Harland and T.R.
Levitt, Radioimmunoassay of lysergic acid diethylamide (LSD) in serum and urine
using antisera of different specificities. Clin. Chem., 23 (1977) 169-l 74.
G.K. Aghajanian and O.H.L. Bing, Persistence of lysergic acid diethylamide in the
plasma of human subjects. Clin. Pharmacol. Ther., 5 (1964) 611-614.
D.G. UpshaIl and D.G. Wailling, The determination of LSD in human plasma follow-
ing oral administration. Clin. Chim. Acta, 36 (1972) 67-73.
B. Widdop, The detection of LSD in biological fluids. Bull. Int. Assoc. Forensic
Toxicol., 7 (1971) 6-7.
P.J. Twitchett, S.M. Fletcher, A.T. Sullivan and A.C. Moffat, Analysis of LSD in
human body fluids by high-performance liquid chromatography, fluorescence spectro-
scopy and radioimmunoassay. J. Chromatog., 150 (1978) 73-84.