Acute Cyanide Poisoning Complicated
by Lactic Acidosis and Pulmonary Edema
David L. Graham; David Laman, MD; James Theodore, MD;
Massive cyanide poisoning occurred
in a 21-year-old man who had ingested
600 mg of potassium cyanide. The clinical
course was marked by acute pulmonary
edema and lactic acidosis. Because the
poison was unidentified until nine hours
after ingestion, the patient received only
supportive treatment which included diu-
resis, oxygen, bicarbonate, and assisted
ventilation. A review of the literature
shows that many case reports are poorly
documented and do not provide a firm
basis for evaluating therapy. To our
knowledge, only four patients, including
ours, have had blood levels of cyanide
measured. In the absence of a suitable
history, diagnosis of cyanide poisoning is
difficult. A simple chemical test which can
be performed on gastric aspirate is avail-
able. Hydroxocobalamin may be used as a
nontoxic specific antidote. Nonspecific
supportive therapy is of great impor-
tance.
problem cyanide poisoning
The
the
has been
of
recognized at least since
time of the famous chemist
Scheele who first synthesized HCN
and then "was killed by the vapor set
free in the breaking of a flask of this
fluid" in 1786.' Much of the clinical
literature has been concerned with
case reports of cyanide poisoning, in
which currently accepted antidotes
have been applied often with apparent
success.
To our knowledge, in only four of
sixty-one cases reported in the last 100
years has the magnitude of poisoning
been documented quantitatively.
Some patients may have had such mild
poisoning that recovery without treat¬
ment would have occurred. However, a
Accepted for publication Sept 28, 1976.
From the Department of Medicine, Stanford
(Calif) University School of Medicine. Mr
Graham is a medical student.
Reprint requests to Department of Medicine,
Stanford University School of Medicine, Stan-
ford, CA 94305 (Dr Robin).
Downloaded From: http://archinte.jamanetwork.com/ by a DALHOUSIE UNIVERSITY-DAL-11762 User on 05/18/2015
Eugene D. Robin, MD
causal relationship has been inferred
between antidotal use and successful
outcome. This forms the basis for
advocacy of nitrite and 3 thiosulfate
therapy in modern texts.2
We treated a patient with massive
cyanide poisoning who recovered with
only supportive treatment. Lactic aci-
dosis and pulmonary edema were
important manifestations of cyanide
For editorial comment see p 993.
intoxication. Changes in arterial blood
gas concentrations and acid-base val¬
ues were analyzed. A laboratory test
to confirm the diagnosis of oral
cyanide poisoning is available. A
review of current therapeutic ap¬
proaches is undertaken and the
rationale for the use of hydroxocobal-
amin is described. Some of the asso¬
ciations between cyanide and human
disease are outlined.
REPORT OF A CASE
A 21-year-old man was brought to the
hospital because of unconsciousness. He
was stuporous, cyanotic, and with evidence
of recent emesis. Gasping respirations (24/
min) were present, blood pressure was 168/
112 mm Hg, and temperature was 36.5 C.
Pulse rate was 68 beats per minute and
irregular. Pupils were midposition and
reactive to light. Diffuse bilateral rales and
rhonchi were noted. The point of maximum
impulse was in the fifth intercostal space
at the midclavicular line. No murmurs,
gallops, or rubs were noted. The nailbeds
were cyanotic. Urine reaction for protein
was 1 +, for glucose 2 +, and was negative
for ketones. Hemogram was normal. Labo¬
ratory values were as follows: sodium, 136
mEq/liter; potassium, 3.6 mEq/liter; chlo¬
ride, 104 mEq/liter; and total carbon diox¬
ide, 7.0 (anion gap of 35 mEq/liter [normal
15 to 20]). The BUN level was 21 mg/100 ml
and glucose concentration was 245 mg/100
ml. Arterial blood gas values while the
patient was receiving 5 liters of oxygen per
minute were Po2 115 mm Hg, Pco2 12 mm
Hg, bicarbonate 5.6 mEq/liter, and pH
7.27. Creatinine phosphokinase was 220 IU.
The chest x-ray film demonstrated diffuse
bilateral infiltrates consistent with pulmo¬
nary edema. Diuresis produced prompt
roentgenographic clearing of alveolar infil¬
trates. Screening for toxic substances in
blood and gastric contents was reported as
negative. The ECG showed sinus arrhyth¬
mia with multifocal premature ventricular
contractions (PVCs) and right axis devia¬
tion. Central venous pressure (CVP) was 9
to 11 cm H,0 for the first 6 hours after
admission. Assisted ventilation was insti¬
tuted and diuresis occurred when the
patient was given 80 mg of furosemide
intravenously. Treatment consisted of ox¬
ygen and intravenous fluids, including
sodium bicarbonate and potassium chlo¬
ride. Three hours after admission, while
the patient was receiving 80% oxygen, the
Pao2 was 64 mm Hg, Pco, 35 mm Hg,
bicarbonate 14.5 mEq/liter, and pH 7.23.
Subsequent blood gas measurements are
shown in Table 1. Nine hours after admis¬
sion, we discovered that the patient had
ingested potassium cyanide, and because
at least ten hours had elapsed, no treat¬
ment with sodium nitrite or sodium thio-
sulfate was undertaken. Ten hours after
admission, CVP was 11 cm H,0, and the
blood pressure was normal. Bicarbonate
concentration increased from 5.6 on admis¬
sion to 19 mEq/liter at 20 hours. The
creatinine level was 1.3 mg/100 ml. During
the next 24 hours, the blood gas values
continued to improve; extubation was
performed and an adequate fraction of
inspiratory oxygen (FIo2) pressure was
maintained. The chest roentgenogram
cleared, although the ECG revealed a rate
of 70 beats per minute with PVCs and
right axis of 90°.
Blood drawn 12 hours after admission
showed a cyanide level of 2.0 /¿g/ml; at 22
hours, it was 1.6 /¿g/ml, and at 84 hours 1.2
jug/ml. Blood sent 17 hours after admission
for ketoacids analysis revealed /J-hydroxy-
butyrate 1.2 mEq/liter (normal, 0.6 ± 0.3
mEq/liter), acetoacetate 1.6 mEq/liter
(normal, 0.8 ± 0.4 mEq/liter), and lactic
acid level 3.3 mEq/liter (normal, 0.8 ± 0.4
 of glycolysis is markedly increased
Table 1.—Selected Blood Gas Measurements Following Cyanide Ingestion Pasteur effect. This, in
through the
Day & HCO Anion Gap turn, leads to lactic acidosis. In experi¬
Time Pao,, mm Hg Pco., mm Hg mEq/Llter,
pH Fio, mm Hg mm/Liter mental cyanide poisoning, lactic acido¬
1 7 PM 115 12 5.6 7.27 35 sis is invariably present, and the
10:30 PM_64_35 14.0 7.23
80%_
...

degree of lactic acidosis correlates

2 1:30
am_139_31 UFO" T33 80%
17.1
with the intensity of cyanide poison¬
_3_AM_194_32 7.35 80%
ing.-1 Additionally, decreases in perfu¬
5 AM
_21 sion associated with the circulatory
8:45 AM 100 25 ÜFÖ 7.40 30%
effects of cyanide would accentuate
~

_5 PM_96_33 2Ö~3 7~42 28%

3 12:05 am 86
6 8:25 AM 82
mEq/liter).
Analysis of two capsules
found on his
person showed that each contained approx¬
imately 200 mg potassium cyanide. Follow¬
ing recovery the patient stated that he had
ingested three capsules. The commonly
accepted lethal dose for cyanide is 50 to 200
mg4 so that he had been exposed to approx¬
imately three to six times the lethal dose.
Five days after admission, the chest
x-ray film.was normal and the patient was
discharged well on the ninth day.
COMMENT
This patient ingested a massive
dose of cyanide and entered the
hospital with severe pulmonary edema
and metabolic acidosis which proved
to be lactic acidosis. Blood cyanide
levels were markedly elevated even at
12 and 22 hours after admission with
blood levels of 2.0 jug/ml and 1.6 pg/
ml, respectively, implying a very
severe intoxication (see below). Be¬
cause the nature of the poison was not
recognized initially, treatment was
supportive, and he recovered com¬
pletely. Two of the more interesting
manifestations found in the patient
were pulmonary edema and lactic
acidosis. The pathogenesis of pulmo¬
nary edema in this patient could be
based on several mechanisms.
Cyanide affects a variety of intra-
cellular metabolic processes (vide in¬
fra) and could possibly injure alveolar
epithelium and capillary endothelium
directly, producing capillary leak syn¬
drome. Against this possibility is the
rapid rate of resolution following
diuresis. Additionally, inhalation of
100% nitrogen (which like CN~ de¬
presses oxidative phosphorylation)
does not cause pulmonary edema.5
Injuries to the CNS have been
described in the pathogenesis of acute
pulmonary edema, even in the absence
of left ventricular disease."8 Acute
28 19~3 7.47 28% 12
41 25~Ö 7~42 RÄ
injury to the brain especially of the
cerebellum and corpus callosum re¬
lated to cyanide has been shown histo-
logically.911 Olsen and Klein have
shown KCN produces increased lac-
tate and decreases of adenosine
triphosphate (ATP) in brain, separate
from the effects of CNS hypoperfu-
sion.1' It is possible, therefore, that an
acute CNS insult by cyanide produced
neurogenic pulmonary edema
(NPE)." However, the rate of resolu¬
tion of pulmonary edema would be
unusual in NPE.
Most likely is a direct toxic effect of
KCN on the myocardium leading to
left ventricular failure and increased
pulmonary capillary pressure. (Unfor¬
tunately, wedge pressures were not
measured.) This patient had addi¬
tional evidence of cardiac dysfunction
manifested by sinus bradycardia and
multiple premature ventricular con¬
tractions. Other investigators have
reported evidence of conduction ab¬
normalities and left ventricular dys¬
function following cyanide poison¬
ing.1415 An increased central venous
pressure was noted during the first 24
hours following admission. Although
cardiac enlargement was not present,
its absence has been noted in other
instances of cardiogenic pulmonary
edema.16
Another facet of the illness was the
severe metabolic acidosis present on
admission. This was associated with a
large anion gap (35 mEq/liter) with¬
out acetonemia or ketonemia, sug¬
gesting lactic acidosis, and this was
confirmed by a blood lactate level
which was still elevated 17 hours after
admission.
Significant cyanide poisoning
should be invariably associated with
lactic acidosis. As oxidative phosphor-
ylation is blocked by cyanide, the rate
lactic acidosis.
It is surprising that lactic acidosis
has not been more commonly reported
in clinical cyanide poisoning. Trapp
described blood gas abnormalities con¬
sistent with metabolic acidosis.22
Naughton reported two cases of acute
KCN poisoning with severe metabolic
acidosis. Both of Naughton's patients
required large amounts of bicarbonate
to correct the metabolic acidosis which
in all likelihood represented lactic
acidosis.23
These considerations suggest that
measurements of blood pH, plasma
bicarbonate, and blood lactic acid
levels should be obtained in CN~
poisoning. These provide important
therapeutic and prognostic informa¬
tion.
METABOLIC EFFECTS
Cyanide inhibits various metabolic
processes. The most important effect
is the intramitochondrial inhibition of
oxidative phosphorylation at cyto-
chrome oxidase. By combining with
the aa, complex, cyanide prevents 0,
from reoxidizing reduced cytochrome
a:), thus inhibiting electron transport,
mitochondrial oxygen utilization, and
cellular respiration.24
As a consequence of inhibition of
oxidative phosphorylation, several
phenomena occur. Mitochondrial 02
utilization ceases and arteriovenous
0. differences are abolished. The loss
of ATP generation in the mitochon¬
drial electron transport chain evokes a
Pasteur effect. This increases lactic
acid generation and leads to lactic
acidosis. The buffering of lactic acid
leads to a progressive fall in plasma
bicarbonate concentration.
The brain is obviously the key organ
involved in cyanide poisoning. It has
been shown that cyanide significantly
increases brain lactate concentration
and decreases brain ATP.12-25 The
changes in brain metabolites after
cyanide administration in mice were

Downloaded From: http://archinte.jamanetwork.com/ by a DALHOUSIE UNIVERSITY-DAL-11762 User on 05/18/2015

 shown to be due to the direct effect of
cyanide on the electron transport
chain rather than the effect of
decreased perfusion alone.12
An excellent animal model which
illustrates the translation of biochem¬
ical mechanisms into altered physi¬
ology is the freshwater turtle, Pseu-
demys scripta elegans. This animal
can survive long periods in the total
absence of ATP generation by oxida¬
tion phosphorylation (0, consump¬
tion zero). Therefore, the effects of
=
massive cyanide poisoning can be
followed in an intact and surviving
animal. Following doses of sodium
cyanide which are approximately 15
times the lethal dose in mice, arterial
and venous Po2 equilibrate with
ambient Po, (~ 150 mm Hg) as tissue
0, utilization ceases, and the animal
continues to breathe. Lactic acid
concentrations increase sharply and
plasma bicarbonate falls.21
REVIEW OF PREVIOUS
CASE REPORTS
Table 2 summarizes 60 reported
cases from the literature. It shows
that patients differ widely with
respect to validity of diagnosis, clin¬
ical manifestations, modes of therapy,
and outcome. Specific tests for CN~
were not done in most patients and
blood levels are available in only four
patients. A number of the patients
may have recovered spontaneously.
BLOOD LEVELS OF CYANIDE
The relationship between blood lev¬
els of cyanide and ultimate outcome is
of obvious importance. There are
extensive data in experimental ani¬
mals but data in man are quite sparse.
To our knowledge, our report is only
the fourth in 100 years in which blood
levels have been reported.
The levels of 2.0 fig/ml (ppm) at 12
hours and 1.6 jug/ml at 20 hours after
admission clearly demonstrate that
there was significant poisoning. Nor¬
mal values are less than 0.20 ftg/ml.211
Fatal cyanide poisoning has been
reported with blood levels of > 3 fig/
ml.3"-32 Extrapolation of blood cyanide
levels in our patient based upon the
known exponential decay of cyanide
in vivo would suggest that the initial
or early postadmission level was well
within the lethal range.31
When possible, blood levels should
Downloaded From: http://archinte.jamanetwork.com/ by a DALHOUSIE UNIVERSITY-DAL-11762 User on 05/18/2015
be measured and the intensity of
specific therapy adjusted according-
ly.
DIAGNOSIS OF CYANIDE
POISONING
Under appropriate circumstances,
in the absence of a suitable history,
several physical findings may call
attention to cyanide poisoning. The
scent of bitter almonds or a "pleasant
almondy odor," is one classical
sign.15-32-13 Unfortunately, many indi¬
viduals are unable to recognize this
smell. In a controlled study, only six of
19 physicians, after smelling a sample
of cyanide indicated a "yes" answer to
the question: "Could it be cyanide?"
Five of 30 were unable to detect any
smell when confronted by a sample of
sodium sulfate containing essence of
almond.34 Thus, the presence of cya¬
nide may be difficult to detect by
odor.
Alterations in the appearance of the
fundal vessels on fundoscopic exami¬
nation has been suggested as a useful
diagnostic sign. In one patient it was
noted that the retinal arteries and
veins were equally red (abolition of
arteriovenous 0, difference).34 How¬
ever, this finding might only be
present in massive poisoning. Both
pulmonary edema and lactic acidosis
are, of course, found in many disor¬
ders and thus are nonspecific. Objec¬
tive diagnosis requires some type of
simple chemical test. Such a test has
been proposed by Lee-Jones et al.35
This method can indicate the presence
of cyanide in five to ten minutes. A
few crystals of ferrous sulfate are
added to 5 to 10 ml of gastric aspirate.
Then approximately 4 to 5 drops of
20% NaOH is added to precipitate the
iron. The solution is boiled, cooled, and
made barely acidic by the addition of 8
to 10 drops of 10% HC1. If cyanide is
present, a greenish-blue precipitate or
color which intensifies on standing,
will appear. The test is applicable only
with ingested cyanide. The reaction is
negative with poisoning by barbitu¬
rates, phenothiazines, benzodiaze-
pines, and tricyclic antidepressants.
One notable exception is salicylate
poisoning; the sample initially turns
blue-green, then becomes purple.
THERAPY
A large number of specific anti-
dotes have been suggested. These are
summarized in Table 3. Some general
comments are in order. The clinical
use of many antidotes is based on
animal experiments. This is appro¬
priate because a critical analysis of
case reports (Table 2 [available from
author]) indicates that most human
studies do not lend themselves to
precise therapeutic conclusions. How¬
ever, animal studies also have inher¬
ent limitations. In most animal stud¬
ies, the treatment protocols were not
designed to resemble the usual clinical
setting. Antidotes were virtually al¬
ways given prior to or simultaneous
with cyanide administration. The
route of administration of cyanide
was usually different from that of
clinical cyanide poisoning. The intrin¬
sic toxicity of the antidotes was not
carefully considered, especially should
the diagnosis of cyanide poisoning
prove to be erroneous. For example, a
child treated with nitrite because of
presumed CN poisoning died because
of overwhelming methemoglobine-
mia.36
Given these uncertainties, the exact
role of various antidotes is' not clear.
The use of relatively nontoxic anti¬
dotes is obviously preferable to the
use of more toxic agents.
These uncertainties have not been
dealt with in standard pharmacology
and medical texts.23 Modern texts
often summarize noncritically evalu¬
ated data using outdated concepts.
For example, the nitrite-thiosulfate
treatment for cyanide poisoning in
dogs,17 was quickly popularized when
one manufacturer produced a cyanide
antidote kit (Cyanide Antidote Pack¬
age) and a number of case reports
attesting to its effectiveness in appar¬
ently rescuing victims from the
effects of fatal cyanide poisoning
appeared.38"4" (Table 2). Their treat¬
ment is described in modern pharma¬
cology texts and based on the report
that "48 out of 49 cases of acute HCN
poisoning were treated successfully
with such therapy."2 It could be noted,
however, that no quantitative tests
for cyanide in blood or gastric aspirate
were ever reported in either Chen or
Rose's cases, or in any other reports
attesting to the combination antidotes
they advocate (Table 2). Regarding
the case reports so collected as
evidence for antidotal efficacy, one
 Table 2.—Summary of Proposed Antidotes for Cyanide Poisoning*
Current Estimated
Agent Mechanism Potential Toxicity Availability References Utllityt
Amyl nitrite Forms methemoglobin Difficult to achieve anti¬ Cyanide Antidote Pack¬ 39 -(A)
(Met-Hb)as step 2 dotal levels without car¬ age 42 -(H)
diovascular collapse
Sodium nitrite NaNo. + Hb ^i Met-Hb Tachycardia, vomiting, Cyanide Antidote Pack¬ 39 + + + (A)
Met-Hb + CN ji hypotension, severe age 42 ?(H)
Cyanmet-Hb methemoglobinemia, 43
hypoxia, vascular col¬
lapse
Aminopropiophenone Forms Met-Hb As with nitrite, but action 43 -(A)
too slow to be of use -(H)
Sodium thiosulfate Na2S20, + CN rhodanese, None known Cyanide Antidote Pack¬ 2 + + (A)
SCN + NaL.SO, age ? Pharmacy 39 + ?(H)
-

Oxygen Increase O, content Oxygen toxicity unlikely- 46 + + + (A)

of arterial blood when used < 48 hr 47 + + + (H)
May reverse CN
binding with cytochromes
Potentiates activity
of sodium nitrite
and sodium thiosulfate
Cobalt salts Chelates cyanide Significant loss of Ca *
Edetate cobalt (Kelocya- 42 + (A)
including edetate Mg**, plus intense pur¬ nor [France]) 48 ?(H)
gation 84
Hydroxocobalamin OH-B,, + CN CN-B,, None known See Nonproprietary 48 + + + (A)
'

Names and Trademarks 53-56 + + (H)

listing
"Since the molecular weight of hydroxocobalamin is 1346 gm/mole, and one mole of cyanide is neutralized by one mole of hydroxocobalamin, large
amounts are needed for antidotal use. Thus 100 mg of KCN would require approximately 1,800 mg of hydroxocobalamin.
t(H) signifies Human studies or cases; (A), animal studies.

might wonder: (1) whether there was a
significantly improved outcome in
these patients as a result of antidote
therapy, and (2) whether what they
were suffering from was in reality,
life-threatening intoxication or occa¬
sionally whether the patient was even
suffering from CN poisoning.
"
nism of action is based on its combina¬
tion with cyanide in the presence of
the enzyme thiosulfate transulferase
(rhodanese) and 0, to produce rela¬
tively nontoxic thiocyanate.3 Thiosul¬
fate is said to be relatively nontoxic2
and has generally been used in combi¬
nation with sodium nitrite. Its effi¬
cacy as a single agent has been shown
py is indicated, it should not be used
alone. Because its benefits may be
based on a supernormal Po2, it should
be used even in the face of a normal
(or supernormal) Pao2.
A variety of cobalt salts have been
advocated.454849 Their basic mecha¬
nism is to bind cyanide by chelation.
One preparation, edetate cobalt (Kelo-

Three of the proposed antidotes

(amyl nitrite, sodium nitrite, and
aminopropiophenone) act primarily by
increasing the amount of circulating
methemoglobin (M-Hb). In turn,
M-Hb binds the cyanide ion, thus
forming cyanmethemoglobin and de¬
creasing CN- combination with cy-
tochrome oxidase.'1 Amyl nitrite is
probably unsatisfactory because it is
difficult to achieve adequate levels of
M-Hb without producing cardiovas¬
cular collapse.'2 Aminopropiophenone
produces M-Hb so slowly that it is not
of use in the clinical setting.43 Sodium
nitrite has been extensively used and
is part of the Cyanide Antidote
Package found in many emergency
rooms. It has considerable intrinsic
toxicity, and its use in patients with
cardiovascular collapse or cerebral
vascular hemorrhage is hazard¬
ous.4245
Sodium thiosulfate is a commonly
used antidote and is also part of the
Cyanide Antidote Package. Its mecha-
to be about that of sodium nitrite,
when used prophylactically in dogs.17
Given the present state of the art,
accepted practice would be to use the
combination of sodium nitrite and
thiosulfate in patients in which the
diagnosis of overwhelming CN- poi¬
soning is clearly established. As clin¬
ical experience accumulates with new¬
er therapeutic approaches, this treat¬
ment may well become obsolete.
The inhalation of 100% oxygen
provides several advantages. It in¬
creases arterial Po, and thus may
increase tissue 0, delivery. It may
reverse the binding of cyanide with
cytochrome oxidase.46 Oxygen may
help increase the conversion of cya¬
nide to thiocyanate by thiosulfate.
(Recent work suggests less sensitivity
of thiosulfate for 02 than previously
thought.47.) Its low toxicity and ease of
administration suggest that it should
be used in every patient with cyanide
poisoning. However, if specific thera-
cyanor [Britain]; no comparable US
product) is commercially available in
Europe and Australia. However, high
concentrations of cobalt salts have
their own intrinsic toxicity.5"51 This
dictates great caution in their use.
Probably the most promising anti¬
dote is hydroxocobalamin (vitamin
Bi2a), which reverses CN" toxicity by
combining with cyanide to form
cyanocobalamin (vitamin B12).525:1 In
animal studies, Evans found this
agent to be especially useful when
combined with thiosulfate.48 The com¬
bination increased the lethal dose for
50% survival of the group (LD,„) three¬
fold in mice and rabbits. He also found
that hydroxocobalamin, when given
intravenously in equimolar quantities
or greater, was effective in rescuing
14 of 16 mice even after ventilation
had failed, following cyanide dose of
200 to 500 fig/kg (2.0 to 4 x LD,„).48
Posner et al were able to show a
significant antidotal effect of hydrox-

Downloaded From: http://archinte.jamanetwork.com/ by a DALHOUSIE UNIVERSITY-DAL-11762 User on 05/18/2015

 ocobalamin in treating sodium cya¬
nide poisoned guinea pigs receiving
lethal doses, as compared to saline
treated controls. They found that in
the groups treated with hydroxocobal¬
amin, all six of six were rescued, but
the controls all died with ECG confir¬
mation of death in less than 30
minutes. Furthermore, they tested
eight animals for hydroxocobalamin
toxicity and found that even at the
rather large dose of 1 mg/gm which is
five times the dose rate used for
successful treatment, no untoward
effects were noted after observation
for 48 hours.54
The successful use of hydroxocobal¬
amin has now been reported in acute
human cyanide poisoning. Two of
three patients with massive cyanide
poisoning (proven by blood level mea¬
surements) recovered after hydroxo¬
cobalamin administration.5" It is cur¬
rently undergoing investigation as a
promising treatment in tobacco am-
blyopia57 (see below). It offers the
overwhelming advantage that it is
essentially nontoxic. Thus, even in the
face of an erroneous diagnosis or
dosage, the patient is not at increased
risk because of therapy. In summary,
then, the use of oxygen together with
intravenous hydroxocobalamin and/or
sodium thiosulfate are probably the
specific therapies of choice.
Nonspecific supportive therapy is
an essential part of the treatment.
This is emphasized by the patient
treated by Liebowitz and Schwartz58
as well as by our patient. Both recov¬
ered from massive cyanide poisoning
without specific treatment. Lactic
acidosis may require bicarbonate ad¬
ministration. Ventilatory failure (res¬
piratory acidosis) may require as-
1. Sollman T: Hydrocyanic Acids and Cya-
nides, in: A Manual of Pharmacology. Philadel-
phia, WB Saunders, 1942, p 826.
12. Olsen NS, Klein RJ: Effect of cyanide on
the concentration of lactate and phosphates in
brain. J Biol Chem 167:739, 1947.
18. Robin ED, Vester JW, Murdaugh HV, et al:
Prolonged anaerobiosis in a vertebrate: Anaer-
obic metabolism in the freshwater turtle. J Cell
Comp Physiol 63:287, 1964.
22. Trapp WG: Massive cyanide poisoning with
recovery. Can Med Assoc J 102:517, 1970.
29. Halstr\l=o/\mF, M\l=o/\llerKO: The content of
cyanide in human organs from cases of cyanide
taken by mouth. Acta Pharmacol 1:18, 1945.
33. Doyle AC: The adventure of the veiled
lodger, in The case book of Sherlock Holmes, in
the Complete Sherlock Holmes, vol 2. New York,
Doubleday & Co, 1953, pp 1290-1299.
Downloaded From: http://archinte.jamanetwork.com/ by a DALHOUSIE UNIVERSITY-DAL-11762 User on 05/18/2015
sisted ventilation. Careful attention
to myocardial depression cardiacor
arrhythmias may be necessary. Care¬
ful fluid and electrolyte treatment are
mandatory. Renal function should be
monitored.
CYANIDE IN HUMAN DISEASE
In addition to acute cyanide poison¬
ing related to accidental or suicidal
exposure, there are a number of other
circumstances in which cyanide is
associated with human disease.
The most important is the relation¬
ship of acute and chronic cyanide
poisoning to cassava (Manihot escu-
lenta crantz). This root is one of the
world's most important staple food
crops providing a major source of calo¬
ries to perhaps 300 million people.39
Cassava contains a number of cyano-
genic glucosides such as linamarin and
lotaustralin.™ Recent work has indi¬
cated that in Africa, high cassava
intakes are associated with a common
disorder, "2tropical ataxic neuropathy
(TAN)."1 In this disorder, high plas¬
ma and urinary cyanide levels are
associated with lesions of the skin,
mucous membranes, optic and au¬
ditory nerves, spinal cord, and periph¬
eral nerves. It is probable that this
entity is a form of chronic cyanide
poisoning. It has also been suggested
that in the presence of marginal
iodine and low protein intakes, diets,
high in cassava may be a factor in the
development of goiter and cretinism.63
In certain parts of Africa where
cassava intakes are high, TAN and
disturbances in thyroid metabolism
are endemic."4
A number of these diseases have
been linked (some by circumstantial
evidence) to chronic cyanide poisoning
References
34. Buchanan IS, Dhamee MS, Griffiths FED,
et al: Abnormal fundal appearances in of
a case
poisoning by a cyanide capsule. Med Sci Law
16:29, 1976.
40. Chen KK, Rose CL: Treatment of acute
cyanide poisoning. JAMA 162:1154, 1956.
41. Albaum HG, Tepperman J, Bodansky O:
Competition of methemoglobin and cytochrome
oxidase for cyanide in vitro. J Biol Chem 163:641,
1946.
46. Isom GE, Way JL: Effect of O2 on cyanide
intoxication: Reactivation of cyanide-inhibited
glucose metabolism. J Pharmacol Exp Ther
189:235, 1974.
48. Evans CL: Cobalt compounds as antidotes
for hydrocyanic acid. Br J Pharmacol 23:455,
1964.
52. Conn JB, Norman SL, Wartman JG: The
equilibrium between vitamin B12 (cyano-cobala-
(mostly derived from tobacco smoke).
These include tobacco amblyopia, ret-
robulbar neuritis in pernicious ane¬
mia, subacute combined degeneration
of the cord in vitamin B12 deficiency,
Leber's hereditary optic atrophy, and
dominantly inherited optic atro¬
phy.65-6" These associations may not
withstand critical analysis, but are of
interest. Cyanide is found in high
concentration in a number of foods in
addition to cassava. Sporadic cases of
acute cyanide poisoning have occurred
because of ingestion of apricot pits67
and choke cherries.68
An interesting etiological cause of
acute CN- poisoning is related to the
therapeutic use of sodium nitro-
prusside. One of the metabolic by¬
products of nitroprusside is CN-.
Several recent deaths reported from
its use could have been acute CN~
poisoning."971
The medical importance of acute
and chronic CN" poisoning would,
therefore, seem to be expanding.
Table 2 and the complete list of references are
available from Dr Robin.
This investigation wijs supported in part by a
grant (1-76C-094) from the California Lung Asso¬
ciation and by grant 1-72-N024 from the North
Foundation. Dr Laman is the recipient of Public
Health Service young pulmonary investigator
award 1 R23 HL-17165. Dr Theodore is the
recipient of Public Health Service pulmonary
academic award HL 1 K07 H HL-00129.
Nonproprietary Names
and Trademarks of Drugs
Hydroxocobalamm-AlphaRedisol, Alpha-
Ruvite, Belvedrox, Codroxomin, Droxo-
vite, Neo-Betalin, Neo-Vitwel, Sytobex-
H.
Sodium nitroprusside-M-pWde.
min) and cyanide ion. Science 113:658, 1951.
53. Mushett CW, Kelly KL, Boxer GE, et al:
Antidotal efficacy of vitamin B-12a (hydroxoco-
balamin) in experimental cyanide poisoning. Proc
Soc Exp Biol Med 81:234, 1952.
61. Osuntokun BO: Ataxic neuropathy asso-
ciated with high cassava diets in West Africa, in
Chronic Cassava Toxicity: Proceedings of an
Interdisciplinary Workshop, London, Jan 29-30,
1973. London, International Development Re-
search Centre Monograph IDRC-010e, 1973,
pp 127-138.
65. Wilson J: Cyanide and human disease, in
Chronic Cassava Toxicity: Proceedings of an
Interdisciplinary Workshop, London, Jan 29-30,
1973. London, International Development Re-
search Centre Monograph IDRC-010e, 1973,
pp 121-125.