Professional Documents
Culture Documents
Barry Levine
State of Maryland, Office of the Chief Medical Examiner, Baltimore, Maryland
David Purser
Hartford Environmental Research, Hatfield, United Kingdom
Abstract
The toxicological and postmortem analysis of fire victims blood
and tissue can disclose the type and quantity of toxic species, such
as carbon monoxide or hydrogen cyanide, that they inhaled prior
to death. For fire cases, these toxicological data can reveal
objective data about the nature and circumstances of a fire, and
thus assist both the Medical Examiner and the Fire Investigator in
their investigations. Assigning a level of significance to cyanide
concentrations found in the blood and tissue of fire victims is
often hampered by the fact that cyanide is inherently unstable in
cadavers and in stored tissue samples. Numerous researchers have
provided insight into and characterized the stability of cyanide in
the body and in collected biological specimens. Based on studies
by these researchers, the rate of transformation of cyanide in
blood and tissue specimens is dependent on the initial cyanide
concentration in the sample at time of death, the length of time
that a sample remains in the cadaver, the length of time that a
sample remains in storage, and the preservation (e.g., addition of
sodium fluoride to sample) and storage conditions (e.g.,
temperature) of the sample.
Introduction
The measurement and interpretation of cyanide (and to a
lesser extent thiocyanate) in blood and other tissues is obviously important in cases of suspected poisoning by cyanide or
cyanogenic compounds such as in certain foodstuffs, but a
major consideration by the authors of this paper is in relation
to the role of hydrogen cyanide (HCN) in injuries and deaths
of fire victims. Many fuels involved in fires have a significant
nitrogen content, and studies of experimental fires have shown
* Author to whom correspondence should be addressed. Combustion Science &
Engineering, Inc., 8940 Old Annapolis Road, Suite L, Columbia, MD 21045.
E-mail: jmcallister@csefire.com.
612
Reproduction (photocopying) of editorial content of this journal is prohibited without publishers permission.
An important parameter in relation to rate and extent of decrease is temperature, with greater and more rapid decreases
occurring at higher temperatures. For fire victims, tissues
may be subjected to heating above body temperature for significant periods before specimens can be obtained. In some
localities, a cadaver may then remain at ambient temperature
for periods of hours to several days before autopsy. Similarly,
specimens such as blood samples may subsequently be stored
at ambient temperature for a period of time, then refrigerated
(4C) or frozen (20C) for periods of up to months before
analysis.
Another complicating factor is the possibility of formation of
cyanide in tissue during storage prior to analysis, which may
somewhat offset transformation losses or even result in an increase in cyanide concentrations under certain circumstances.
In this context, it is important to distinguish between the relatively high blood cyanide concentrations occurring in cases
involving cyanide poisoning and the low, background level
concentrations occurring naturally and in subjects such as
smokers, in those on diets high in cyanogens, and in patients
receiving cyanogenic drugs. According to Ballantyne (13) and
Vesey and Wilson (14), a small amount of cyanide can be
formed in samples that were initially negative for cyanide,
which is most likely due to conversion of plasma thiocyanate
(SCN) to cyanide on acidification of whole blood in the presence of free hemoglobin. This conversion may represent a significant artifact during the study of such conditions, but it
would be considered less relevant at levels of lethal significance. However, on rare occasions in some other studies
(15,16), considerably greater increases in cyanide concentration have been reported to occur following storage of initially
cyanide-negative blood.
This paper will review the published literature in an attempt
to establish a foundation for the proper interpretation of postmortem cyanide concentration changes.
Specimen
Mean Percentage of
Original CN Concentration
Detectable after 1 h at
Ambient Temperature
Human urine
Human blood serum
Rabbit blood serum
Rabbit whole blood
73%
39%
33%
78%
Mean Percentage
of SCN Appreciation
after 1 h at
Ambient Temperature
30%
33%
26%
Not measured
613
work of Ballantyne et al. (10). Furthermore, the Ballantyne experiment revealed one possible mechanism of decrease in
whole blood via transformation of cyanide to thiocyanate. In
one of their studies, Ballantyne et al. (10) measured the
changes of cyanide concentrations in human serum and urine,
and in rabbit serum and whole blood, all spiked with potassium
cyanide. SCN, a metabolite of cyanide, was also measured in
the urine and serum specimens. To quantify the significance of
the thiocyanate development in the spiked samples, Ballantyne
and co-workers (10) determined the thiocyanate concentration
in control samples and found that human urine and serum and
rabbit serum were at concentrations of 2.49 0.81 mg/L, 5.80
1.74 mg/L, and 0.64 0.65 mg/L, respectively. Although
the authors did not report the storage temperatures of the
samples, all samples are assumed to have been stored at ambient room temperature. Potassium cyanide was added to each
substrate for a starting concentration of 1.2 mg CN/L. The
concentration of cyanide in the samples was tested at increments of 4, 10, 20, 30, 40, 50, and 60 min after addition of the
potassium cyanide. Table I depicts a summary of the results
from the Ballantyne et al. (10) testing. It is notable that, al-
614
though there were significant decreases in cyanide content, especially in serum, there were increases in thiocyanate.
Based on this research, the authors concluded that the most
rapid decrease in cyanide concentration in human and rabbit
serum occurs in the first 20 min after the sample was spiked,
but that only a small amount of the lost cyanide was measured
as SCN in the serum specimens. Specifically, only approximately 26% of the rabbit and human serum cyanide was recovered as thiocyanate. The rate of depreciation of cyanide in
the spiked rabbit blood was less significant than in serum;
however, the most significant decrease of cyanide in the blood
occurred within the first 20 min, as was also seen in serum.
The rate of depreciation of cyanide in urine was also less than
in serum, and the deviation in measurements of cyanide urine
concentration was notable, ranging from 49% to 88%.
The authors further explored the transformation of cyanide
in whole blood removed from rabbits injected with 8 mg
CN/kg of body weight. Blood was drawn from the animals 4
min after death, and the cyanide concentration of the blood
was then measured at various time increments up to 24 h
after death. Although the authors did not report the storage
temperatures of the samples, all samples are assumed to have
been stored at ambient room temperature. At 4 min, the
cyanide blood concentration was approximately 15.9 mg
CN/L. After 1 h, only 84% of this concentration was present in
the collected blood. After a 24-h period, a mean of 67% of the
initial concentration was present in the blood. Therefore, the
same trend was found in the blood recovered from the rabbits
injected with cyanide as in the spiked samples.
Ballantyne et al. (11) further expanded their study in a second
publication that compared the rate of transformation of cyanide
in blood and tissue removed from rabbits immediately after
death with the rate of transformation of cyanide in blood and
tissue removed from rabbits at various times after death. The
specimens measured included blood, lung, brain, kidney, and
liver. The rabbits were killed by injection of 8 mg CN/kg of body
weight. Within 30 min of death, specimens were removed from
the rabbits. Three sets of specimens were immediately analyzed, followed by analysis of samples at 1, 3, 7, 14, and 21
days. All samples were stored between 10C and 15C. Three additional sets of specimens were removed at 30 min and analyzed, followed by the removal and analysis of samples at 1, 3,
7, 14, and 21 days after death. The rabbit carcasses were stored
at 10C and 15C in sealed polyethylene bags during the test period. Figure 1 depicts the results from measurements of samples removed from the animals immediately after death and
tested at various time increments. No detectable levels of
cyanide were found in the liver at 1 day, the kidney at 3 days,
and the brain and lungs at 14 days. The cyanide concentration
in the blood at 14 days was approximately 38% of the original
value measured immediately after death and increased by 8%
from 14 days to 21 days of storage. A slight increase in the
concentration of cyanide in the lung tissue was also documented between days 1 and 3. Figure 2 depicts the results from
testing of organs and blood samples removed from the animals at various times after death. No detectable levels of cyanide
were found in the liver and kidney at 3 days and the brain and
lung at 14 days. The cyanide concentration in the blood at 21
Sample
Storage
Temperature
Storage
Time
1
2
3
4
4C
20C
4C
20C
8 weeks
8 weeks
8 weeks
8 weeks
Initial Cyanide
% Loss of
Concentration (mg/L) Cyanide
5.0
5.0
2.5
2.5
30%
31%
17%
17%
Sample
Time
Group A
High MetHb
%Loss of Cyanide
24 h
72 h
85
90
Group B1
Group B2
Low MetHb
Low MetHb
% Loss of Cyanide % Loss of Cyanide
47
63
45
100
615
616
were administered the same dose of cyanide, and Moriya and coworkers (20) noted that Group A rabbits survived longer (16
min) than Group B rabbits (6 min). This change in survival time
was presumably due to the protective reaction of some of the
cyanide with methemoglobin to form cyanomethemoglobin.
Although the author does not comment on the cause of the
higher concentrations of cyanide in Group A, the pretreatment
of the Group A rabbits with sodium nitrite may have allowed a
large proportion of cyanide to be sequestered into the blood and
bound to MetHb, while more cyanide in the Group B rabbits was
distributed into the tissue and other body fluid compartments.
Moriya commented that the rate of disappearance of cyanide
was greater in the blood containing higher concentration of
cyanide (Group A) than in Group B. Ballantyne and Bright
showed a strong relationship between cyanide concentration
and % depreciation; therefore, it is likely that this relationship
may have played a role in concentration changes. Because the
Group A rabbits had higher MetHb and higher cyanide concentrations, the independent role of each variable in the
changes in cyanide concentrations in the sample is unclear.
Based on the Moriya et al. (20) results, it appears that in the
high MetHb concentration group, there was increased cyanide
depreciation in samples stored under elevated temperature and
room temperature conditions. However, under refrigerated
conditions, the subsequent percentage depreciations of the
high MetHb samples were similar to those with low MetHb
(Figure 3), representing approximately 30% of that remaining
after 4 h. Additionally, when samples were stored at room temperature for a significant period of time (e.g., 3 days), almost all,
if not all, of the cyanide disappeared, as was seen in Group B2.
In a study by Narayanaswami et al. (21), human stomach
tissue samples were spiked with cyanide and stored at room
temperature for 1, 4, 8, 10, 15, and 20 days. The tissues were
maintained at a pH of 6.5 and spiked with 1 mL of sodium
cyanide ranging in concentration from 50 to 150 mg/L. An
additional set of samples maintained at a pH of 8.0 was spiked
with 1 mL of 100 mg/L sodium cyanide. Researchers found that
75% to 80% of the cyanide in the tissue at the 6.5 pH was lost
in the first week, after which point no significant changes were
seen. Sample initial concentration did not appear to have a significant effect on percent decrease in cyanide concentration.
free cyanide was only produced in the samples at temperatures greater than 50C. The highest total cyanide (bound plus
free) was found to be produced in both hemoglobin and blood
at 75C. A total cyanide content of approximately 8 ng was
produced in hemoglobin, and 12 ng was produced in blood at
75C. Seto and co-workers (22) concluded that the maximum
heat-induced cyanide production, which was less than 0.050
mg/L in blood, did not exceed that which was toxicologically
significant and, therefore, would have a negligible influence on
confirmation of cyanide poisoning in fire victims.
In a study published by Chikasue et al. (23), blood and tissue
samples from five cyanide poisoning cases (non-fire related)
were collected and measured at the time of autopsy. The samples were again measured for cyanide concentrations between
1 day and 3 weeks after collection and storage at refrigerated
(4C) and frozen (20C) conditions. The times from death to
autopsy were 8 h, 13 h, 16 h, 13 h, and 1820 h for Cases 15.
The summaries of test results for Cases 1 and 2 are presented
in Figures 4 and 5, respectively. The abbreviations LHB and
RHB in Figures 4 and 5 represent the left and right heart
blood. The researchers found that the ratio of blood cyanide
concentrations after storage to that before storage ranged from
0.71 to 1.46 under refrigerated and frozen storage conditions.
617
A larger range of ratios from 0.2 to 8.8 was measured for liver,
kidney, brain, and stomach content cyanide concentrations
under refrigerated and frozen conditions. Cases 1 and 2 are the
only cases for which full data was presented. In general, the researchers found increases in cyanide concentrations in Cases
1 and 2 after 20 days of freezing in all samples, with the exception of the stomach contents from Case 2, which showed a
slight decrease. In Case 1, after 24 days of refrigeration, an
overall increase was found in the right (+ 31%) and left
(+ 20%) heart blood, kidney (+ 87%), and brain (+ 60%) samples, and a decrease was found in the stomach contents (37%).
No change was seen in the liver sample. In Case 2, after 24 days
of refrigeration, an overall increase was measured in the right
(+ 26%) and left (+ 29%) heart blood, and decreases were measured in the liver (78%), kidney (53%), brain (43%), and
stomach contents (54%).
The research of Curry et al. (15) and Lokan et al. (16) focused
on the development of cyanide in negative samples stored
under various temperature conditions. In a publication by
Curry et al. (15), research on cyanide development in postmortem negative samples arose from a case in which a positive
cyanide sample was found where no cyanide exposure was possible. Curry et al. (15) found that serum and red cell samples
showed no cyanide production during a two-week period at
room temperature, with the exception of one sample that measured 0.25 mg/L cyanide. However, in transfusion blood samples, stored in test tubes closed with cotton wool plugs at room
temperature for two months, high cyanide concentrations
were found on the order of 100 mg/L. In one case stored at 4C
for 3 months, cyanide levels in the sample apparently increased
from 3 to 40 mg/L. Another sample, initially containing no detectable cyanide, showed a level of 5 mg/L after 2.5 months of
storage at 4C, but this level then decreased to 0.1 mg/L after
6 additional months of storage at 4C. The authors found no
cyanide in a sample preserved with 1% sodium fluoride, a
compound which the authors stated will completely inhibit
cyanide production. Nevertheless, the authors concluded that,
although it was not a constant phenomenon, cyanide production could occur in bacteriologically sterile stored blood. The
significance of this work is difficult to fully evaluate, because
this research was carried out under variable conditions using
a semi-qualitative method of analysis.
In contrast, Ballantyne (24), in a series of detailed studies of
fresh pooled blood, found a slow transformation in the cyanide
concentration when the samples were stored at room temperature and 4C for up to 15 weeks, with increases in negative
samples seen only in those samples stored at 20C. The increases were small and only reached a maximum peak cyanide
level of 0.20 mg/L. Ballantyne (13) and Vesey et al. (14) considered this increase to be due to conversion of approximately
3% of plasma thiocyanate to cyanide. Egekeze and Oehme (25)
also reported only minor changes over 7 days at 4C and 25C.
Lokan et al. (16) developed an interest in negative sample
cyanide production, because they identified three postmortem
samples that had high positive cyanide concentrations, but
where no sources of victim cyanide exposure could be identified. However, they pointed out that these three unusual cases
should be set against a total 331 cases routinely screened for
618
Conclusions
If appropriate procedures are followed to preserve samples,
fire victim cyanide and COHb levels can provide investigators
with objective data for use in hypotheses development. Fuels
containing nitrogen as well as carbon produce high concentrations of both CO and HCN under vitiated combustion conditions occurring in many enclosure fires (such as many fires in
buildings or transport vehicles), whereas smoldering or nonflaming fire produce mainly CO and rapidly growing well-ventilated fires produce low yields of both CO and HCN. At
concentrations typical of vitiated fires, HCN can produce incapacitation significantly more rapidly than CO. For these reasons, the levels of both cyanide and COHB in the blood of fire
victims can provide important information relating to the materials involved during critical phases of fire development and
(especially when used in conjunction with fire experiments or
fire modeling), can contribute to evaluating the time course and
causes of incapacitation and death in fatal fire incidents.
As an example of how postmortem blood cyanide information can be used in a fire investigation, consider a case where
an investigator arrives on a fire scene and finds fire damage
confined to a living room, with a fire victim dead in an upstairs
bedroom. The fire was determined to have started in an overstuffed cotton chair and eventually spread to adjacent
polyurethane-treated wood paneling. The fire investigators
needed to determine whether the victim died early in the fire
when only the chair was involved, or later in the fire during full
room involvement. Although the victim in this case clearly
died of CO poisoning with a COHb level greater than 70%,
the victims blood was found to have nearly 2.0 g/mL of
cyanide The presence of significant cyanide in the victims
blood allowed the fire investigators to determine that the fire
spread to adjacent polyurethane-treated wood paneling. There-
fore, the fact that the victim did not die until late in the fire,
after involvement of the wood paneling, was determined using
blood cyanide toxicological data.
The material presented in this paper shows that the rate of
transformation of cyanide in blood and tissue specimens is dependent on four criteria: 1. initial sample concentration at time
of death; 2. length of time that sample remains in the cadaver
after death; 3. length of time that sample remains in storage before analysis; and 4. conditions of sample preservation and
storage (i.e., temperature, pH, sodium fluoride addition).
The basic findings from experimental work carried out under
laboratory conditions are that when cyanide concentrations of
near-lethal significance (i.e., approximately 3 mg/L) are present
in a cadaver at death or in postmortem blood or tissue samples,
a subsequent decrease in concentration occurs over periods of
hours to months. The greatest rate of cyanide decrease occurs
in cadavers, and to a slightly lesser extent in blood samples, at
ambient or elevated temperatures. Serum and plasma cyanide
levels decrease rapidly (within 20 minutes) when stored at ambient temperatures. The rate of cyanide decrease in whole blood
is slower: it can be significant at 4C, but is small when samples
are frozen at 15C to 20C. Under conditions where cyanide
decreases occur, these decreases tend to be greater with higher
initial blood cyanide concentrations. When blood is frozen, a
small increase tends to occur upon thawing (approximately 0.2
mg/L), which is considered to be due to the formation of small
amounts of cyanide resulting from acidification of plasma or
serum thiocyanate in the presence of hemoglobin.
Most studies, as discussed here, show a decrease in cyanide
concentration in postmortem cadaver, blood, and tissue samples. However, there are a limited number of reports from
studies of cyanide-negative transfusion blood and a small
number of cases considered not to have involved cyanide exposure [around 1% of routine cases examined by Lokan et al.
(16)] in which very high concentrations of cyanide have been
measured in blood samples stored under room temperature, refrigerated, or frozen conditions for periods of a month or more.
In several of these cases, measured cyanide levels were more
than 30 times the lethal concentration. No obvious reasons
have been identified for the very high levels of cyanide formation in these samples, although microbiological contamination
is a possible consideration. The authors (15,16) issue a caveat
that considerable care should be exercised in attempting to interpret high blood cyanide concentrations in unpreserved
blood specimens in isolation. The authors reporting these findings have found them not to occur in blood samples preserved
with 1% sodium fluoride; therefore, it is recommended that at
least one blood sample should be preserved with 1% sodium
fluoride at postmortem collection. The lack of reports of similar exceptionally high blood levels in several major blood
cyanide studies of fire fatalities and survivors indicates that
such instances are relatively uncommon.
The main recommendations from this review are that samples of blood and other tissues should be taken as soon as possible after death and analyzed for cyanide as soon as possible
after collection. Storage of samples at 4C provided relatively
stable results; however, samples are best preserved in deepfreezer (20C). Additionally, the time intervals between death,
619
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
References
1. D.A. Purser. Toxicity assessment of combustion products. In The
SFPE Handbook of Fire Protection Engineering, 3rd ed. National
Fire Protection Association, Quincy, MA, 2002, pp 2/832/171.
2. D.A. Purser. Toxic product yield and hazard assessment for fully
enclosed design fires involving fire retarded materials. Polymer
Int. 49: 12321255 (2000).
3. F.J. Baud, P. Barriot, V. Toffis, B. Riou, E. Vicaut, Y. Lecarpentier,
R. Bourdon, A. Astier, and C. Bismuth. Elevated blood cyanide
concentrations in victims of smoke inhalation. New Engl. J. Med.
325: 17611766 (1991).
4. D.A. Purser and W.D. Woolley. Biological effects of combustion
atmospheres. J. Fire Sci. 1: 118144 (1982).
5. D.A. Purser, P. Grimshaw, and K.R. Berrill. Intoxication by cyanide
in fires: a study in monkeys using polyacrylonitrile. Arch. Environ.
Hlth. 39: 394400 (1984).
6. D.A. Purser. Behavioural impairment in smoke environments.
Toxicology 115: 2540 (1996).
7. B. Ballantyne. The forensic diagnosis of acute cyanide poisoning.
In Forensic Toxicology. Wright, Bristol, U.K., 1974, pp 99112.
8. R.A. Anderson, I. Thomson, and W.A. Harland. The importance
620
24.
25.
26.
27.
28.
29.