FIRE, SMOKE, SUDDEN DEATH AND THE MISSED DIAGNOSIS
Cyanide poisoning is generally viewed as an extremely rare event, presenting as a dramatic form of suicide or sometimes as the result of an industrial accident in the mining, metallurgy, or plating industries. Indeed, these forms of “pure” cyanide poisoning are quite rare. There is, however, another form of cyanide poisoning which is often unappreciated in its presentation and because of associated pathology, even the astute clinician may let the diagnosis of cyanide poisoning go undetected.
The most common of cause of cyanide poisoning is smoke inhalation. It is also the most underappreciated (and thus under-treated) source of cyanide poisoning. The literature is replete with references equating smoke inhalation with carbon monoxide poisoning. The toxicological truth is far more complex. Clearly, carbon monoxide plays a significant role in the morbidity and mortality of smoke inhalation, but it is not the only important toxicant in this setting and in some cases, it is far less important than other associated toxicants, including cyanide.
The notion that cyanide might be important in smoke inhalation dates back at least to 1966, when Wetherell noted the presence of significant amounts of cyanide in the blood of fatal fire victims studied by his service. (1) Numerous exposure chamber studies of animals followed, which almost invariably showed that the concomitant exposure of animals to carbon monoxide and cyanide led to more rapid incapacitation (inability to escape) and death, with the toxic interaction being either additive or in some cases synergistic. (2-4) In 1991, Baud et al undertook the first prospective evaluation of fire victims where samples of blood for both carbon monoxide and cyanide analysis were collected at the scene of the fire.
The findings with respect to mortality and blood cyanide concentrations are
telling: the mean blood cyanide concentrations among patients who died was
116.5 ± 89.6 µmol/l (toxic > 39 µmol/l, lethal > 100 µmol/l), while
survivors had a mean of 21.6 ± 36.4 µmol/l. Carbon monoxide concentrations
(1 mmol/l is toxic and 5 mmol/l potentially fatal) were 2.8 ± 2.0 and 0.7 ±
0.7 mmol/l respectively in fatal and surviving victims.
The importance of these findings has been challenged by Barillo, who, on the
basis of a study of cadavers, whose blood was sampled for cyanide at time of
death and at autopsy found that ~15% of victims had blood cyanide
concentrations in the toxic range and 7% in the lethal range and yet
concluded that, specific assay and treatment for cyanide poisoning is rarely
necessary in the treatment of victims of smoke and fire.
It seems tenuous to advance such an important clinical conclusion based on
an autopsy study. Furthermore, it is important to remember that cyanide has
a relatively short half-life and that if autopsy is delayed before cyanide
sampling, the values may be markedly decreased, so that Barillo’s findings
are almost certainly significant underestimates of the actual peak blood
cyanide concentrations. More recently, we have examined the effect of
treatment of smoke inhalation victims with the specific cyanide antidote
hydroxocobalamin, which binds cyanide to form cyanocobalamin (vitamin B12).
Comparing data in the same hospital, with the same recruitment patterns
before and after the institution of hydroxocobalamin therapy, the observed
survival more than doubled (manuscript in preparation).
Granted, there are problems with the use of historical controls, but very
little else changed in the management of these patients over the 10 or so
years duration of the two studies. Numerous other authors have signaled the
importance of cyanide in smoke inhalation (a review of the evidence for
cyanide in smoke inhalation is provided by Alarie (5)). Ignoring a
potentially lethal toxicant because of the presence of another appears
illogical, so long as one can be treated without adversely affecting the
efficacious treatment of the other.
Another potential source of easily missed cyanide poisoning is “occult”
cyanide poisoning. How many patients who die in the emergency department
with cardiac arrest leave with an accurate etiological diagnosis? Was it a
myocardial infarction, a pulmonary embolus, a sudden dysrhythmia or… cyanide
poisoning? While the latter probably represents a very small subset of the
cohort, it is safe to say that we have no idea how often we are faced with
sudden death from occult poisoning. It is rare that patients who fall into
the age group for coronary artery disease have complete toxicological
evaluations or autopsy.
It may well be assumed that they died a "natural death." Yet how often are
these arrests due to undetected suicides? How many of those are from
cyanide? This question may never have an answer, but one thing is clear. If
we don’t consider cyanide in the differential diagnosis of patients
presenting in extremis or in cardiac arrest, the diagnosis and specific
treatment will surely not follow.
What are the signs and symptoms of cyanide poisoning? We are currently
reviewing the published literature and our own clinical experience in a
total of approximately 150 cases of cyanide poisoning to glean the most
helpful clues to the presence of cyanide. Some interesting findings can be
reported: common signs include an irregular respiratory pattern (tachypnea,
bradypnea), coma or altered mental status, seizures and dilated pupils. Of
interest is that while the presence of coma is often associated with lethal
outcome, mydriasis has no prognostic value (manuscript in preparation).
The definitive diagnosis of cyanide poisoning is the blood cyanide
concentration (BCN). A BCN > 39 µmol/l is considered toxic and >100 µmol/l
potentially fatal. Unfortunately, BCN is rarely obtainable from the
laboratory in time to assist with the diagnosis. Lactate is a useful
surrogate when the diagnosis is suggested by the clinical history. While
carbon monoxide poisoning can raise plasma lactate, it rarely reaches the
concentrations seen with cyanide poisoning in smoke inhalation (10 mmol/l).
(6, 7) Plasma lactate can also be helpful in pure cyanide poisoning. (8)
In summary, cyanide poisoning may well be rare, but probably not as rare as
we believe. Until more systematic toxicological studies are done in smoke
inhalation victims and patients presenting in cardiac arrest or extremis, we
will likely continue to underestimate and under-treat such victims. At least
considering the diagnosis may lead to useful screening tests (lactate) which
serve to orient us toward or away from such a diagnosis(6, 8). A prospective
trial of smoke inhalation victims, planned for implementation in the U.S.
next year may help to answer some of these questions. The development of
cyanide antidotes associated with few serious side effects, such as
hydroxocobalamin, will facilitate the treatment of suspected cyanide
poisoning.
Stephen W. Borron, MD, MS, FACEP, FACMT
Associate Clinical Professor of Emergency Medicine, Medicine, and
Occupational and Environmental Health
The George Washington University
Washington, DC 20037
and
Medical and Toxicological Critical
Care
Unit
Lariboisiere
Hospital
Paris,
France
REFERENCES
1.
Wetherell HR. The occurrence of cyanide in the blood of fire victims. J
Forensic Sci 1966;11(2):167-73.
2. Norris JC, Moore SJ, Hume AS. Synergistic lethality induced by the
combination of carbon monoxide and cyanide.
Toxicology 1986;40(2):121-9.
3. Levin BC, Paabo M, Gurman JL, Harris SE. Effects of exposure to single or
multiple combinations of the predominant toxic gases and low oxygen
atmospheres produced in fires. Fundam Appl Toxicol 1987;9(2):236-50.
4. Purser D. Behavioural impairment in smoke environments. Toxicology
1996;115(1-3):25-40.
5. Alarie Y. Toxicity of fire smoke. Crit Rev Toxicol 2002;32(4):259-89.
6. Baud FJ, Barriot P, Toffis V, Riou B, Vicaut E, Lecarpentier Y, et al.
Elevated blood cyanide concentrations in victims of smoke inhalation. N Engl
J Med 1991;325(25):1761-6.
7. Benaissa ML, Megarbane B, Borron SW, Baud FJ. Is elevated plasma lactate
a useful marker in the evaluation of pure carbon monoxide poisoning?
Intensive Care Med 2003;29(8):1372-5.
8. Baud FJ, Borron SW, Megarbane B, Trout H, Lapostolle F, Vicaut E, et al.
Value of lactic acidosis in the assessment of the severity of acute cyanide
poisoning. Crit Care Med 2002;30(9):2044-50.