TOXICOLOGY/REVIEW ARTICLE

Sodium Thiosulfate or Hydroxocobalamin for the Empiric

Treatment of Cyanide Poisoning?
Alan H. Hall, MD From the Toxicology Consulting and Medical Translating Services, Inc., Elk Mountain, WY (Hall);
Richard Dart, MD, PhD the Rocky Mountain Poison and Drug Center, Denver Health, Denver, CO (Dart, Bogdan); and the
Departments of Preventive Medicine and Biometrics (Hall), Surgery [Emergency Medicine] (Dart),
Gregory Bogdan, PhD
and Pharmaceutical Sciences (Bogdan), the University of Colorado Health Sciences Center,
Denver, CO.

Cyanide poisoning must be seriously considered in victims of smoke inhalation from enclosed space
fires; it is also a credible terrorism threat agent. The treatment of cyanide poisoning is empiric
because laboratory confirmation can take hours or days. Empiric treatment requires a safe and
effective antidote that can be rapidly administered by either out-of-hospital or emergency department
personnel. Among several cyanide antidotes available, sodium thiosulfate and hydroxocobalamin have
been proposed for use in these circumstances. The evidence available to assess either sodium
thiosulfate or hydroxocobalamin is incomplete. According to recent safety and efficacy studies in
animals and human safety and uncontrolled efficacy studies, hydroxocobalamin seems to be an
appropriate antidote for empiric treatment of smoke inhalation and other suspected cyanide poisoning
victims in the out-of-hospital setting. Sodium thiosulfate can also be administered in the out-of-hospital
setting. The efficacy of sodium thiosulfate is based on individual case studies, and there are
contradictory conclusions about efficacy in animal models. The onset of antidotal action of sodium
thiosulfate may be too slow for it to be the only cyanide antidote for emergency use.
Hydroxocobalamin is being developed for potential introduction in the United States and may represent
a new option for emergency personnel in cases of suspected or confirmed cyanide poisoning in the
out-of-hospital setting. [Ann Emerg Med. 2007;49:806-813.]

0196-0644/$-see front matter

Copyright © 2007 by the American College of Emergency Physicians.
doi:10.1016/j.annemergmed.2006.09.021

SEE EDITORIAL, P. 814.
INTRODUCTION
Cyanide has been used as an agent for suicide, homicide,
chemical warfare, and genocide and can contribute to death of
victims of smoke inhalation in enclosed space fires.1-4 In the
United States, smoke inhalation causes as many as 10,000
fatalities each year.5 Of US annual burn fatalities, 60% to 80%
are likely due to smoke inhalation.6 Fire smoke contains a
complex mixture of gases that contribute to smoke inhalation-
associated death, including hydrogen cyanide that is released
from natural and synthetic materials on combustion.1,4 Cyanide
and cyanogenic compounds are also potential weapons of
terrorism.7,8 Because cyanide can be released from both
synthetic and natural materials, terrorist acts with explosives or
incendiaries, resulting in enclosed space fires, could result in
poisoning, even if cyanide or cyanogenic compounds are not used.
Cyanide reacts with high affinity with metals such as ferric
iron (Fe3⫹) and cobalt and binds to numerous critical enzyme
systems in the body. Cyanide inhibits oxidative phosphorylation
and causes central nervous system and cardiovascular dysfunction
because of cellular hypoxia, primarily by binding to and
inactivating the enzyme cytochrome oxidase (cytochrome a3).5,9
Cyanide poisoning requires immediate and aggressive
treatment. The emergency diagnosis is usually difficult because
there are no pathognomonic symptoms or signs,10 and
laboratory confirmation of cyanide poisoning takes hours to
days to obtain. A confirmed diagnosis of cyanide poisoning is
not possible in out-of-hospital emergency situations. Therefore,
the decision to administer specific cyanide antidotes must be
made on clinical characteristics. These can include hypotension,
altered mental status, elevated plasma lactate concentrations
(ⱖ10 mmol/L), and relatively normal oxygen saturation. In
smoke inhalation victims, the presence of soot in the mouth and
nose, with concurrent altered mental status and hypotension,
suggests the possibility of cyanide poisoning.5
Because of diagnostic uncertainty, a cyanide antidote may be
administered to patients who may not be poisoned. A favorable
risk:benefit ratio is thus highly desirable.11-15 There are 4
antidotes used in various countries for treatment of cyanide

806 Annals of Emergency Medicine Volume , .  : June 

Hall et al
poisoning: the Cyanide Antidote Kit (amyl nitrite, sodium
nitrite, and sodium thiosulfate), hydroxocobalamin (Cyanokit),
dicobalt– ethylenediamine tetraacetic acid (EDTA)
(Kelocyanor), and 4-dimethylaminophenol (DMAP).
Unfortunately, dicobalt-EDTA, DMAP, and the sodium nitrite
component of the cyanide antidote kit have serious adverse
effects (including hypotension and the formation of
methemoglobin that can exacerbate carbon monoxide–induced
hypoxemia) that make them less desirable for empiric use,
especially outside a health care facility.12 Only 2 cyanide
antidotes, hydroxocobalamin and a component of the Cyanide
Antidote kit, sodium thiosulfate, have been proposed for
empiric therapy in cyanide poisoning (P. Edelman, written
communication, 2003).16
Sodium thiosulfate has been proposed for use alone in
victims of chemical terrorism by the US Department of Health
and Human Services (P. Edelman, written communication,
2003). However, this idea has been challenged because of its
slow onset of action.17 Lang first demonstrated the antidotal
efficacy of sodium thiosulfate for cyanide poisoning in 1885,
and before 1929, sodium thiosulfate was the only known
specific cyanide antidote.18 There is little recent information
available about the efficacy of sodium thiosulfate for treatment
of cyanide poisoning.19
Hydroxocobalamin (vitamin B12a), the natural form of
vitamin B12, is a hemelike molecule with a complexed cobalt
(Co) atom.20 It is used for cyanide poisoning and smoke
inhalation in France,21 where it is administered by the out-of-
hospital system and has been in clinical use for more than
30 years. Hydroxocobalamin is also used in other countries,
including Sweden, Denmark, Spain, Japan, and Hong
Kong.17, 22-25
The purpose of this analysis is to evaluate and review the
literature on hydroxocobalamin and sodium thiosulfate
monotherapy so that emergency personnel and clinical
toxicologists are better able to develop practice for the empiric
treatment of cyanide poisoning.
EVALUATION OF MEDICAL LITERATURE
A review of published and recently presented studies on
hydroxocobalamin and sodium thiosulfate was performed in
several languages (English, French, Spanish, Danish, Swedish,
German, and Japanese; professional translations were obtained)
and with electronic and hardcopy literature searches from the
1930s to the present. Searches were conducted at the National
Library of Medicine, Bethesda, MD, with MEDLINE,
TOXLINE, and other reference resources. Search terms
included “hydroxocobalamin,” “sodium thiosulfate,” “cyanide
poisoning,” “cyanide antidotes,” “toxic terrorism,” and “smoke
inhalation.” The extensive collection of relevant books and
articles of one of the authors (A.H.H.) was also reviewed. More
than 250 references were reviewed, as well as Google searches
for recent media reports of cyanide poisoning, toxic terrorism, and
smoke inhalation incidents. References involving chronic cyanide
dietary exposure from cyanogenic plants were not reviewed.
Volume , .  : June 
Empiric Treatments for Cyanide Poisoning
EFFICACY OF HYDROXOCOBALAMIN
The effectiveness of hydroxocobalamin as a cyanide antidote
was first demonstrated in 1952.26 The mechanism of action is
direct binding to cyanide to form nontoxic cyanocobalamin
(vitamin B12) that is excreted in the urine.11 In an in vitro study
using human fibroblasts incubated in a cyanide solution,
addition of hydroxocobalamin resulted in a 75% decrease in
intracellular cyanide concentrations and formation of
intracellular cyanocobalamin, indicating that hydroxocobalamin
penetrates cells and can act intracellularly.27
Animal Studies
Hydroxocobalamin crosses the blood-brain barrier and enters
the cerebrospinal fluid in experimental animals.28 It has been
tested as a cyanide antidote in animal models using mice,21,26
rabbits,29,30 guinea pigs,31,32 dogs,33-38 and baboons.39 The
majority of these studies demonstrated antidotal efficacy of
hydroxocobalamin even if administered after cyanide. No
animal study compared hydroxocobalamin and sodium
thiosulfate by using the same experimental protocol.
The efficacy of hydroxocobalamin was compared to saline
solution vehicle for the treatment of dogs administered
potassium cyanide (0.4 mg/kg per minute, intravenously).36 In
vehicle-treated animals, the overall mortality rate was 82% (14/
17), and death occurred within 4 hours to 4 days. In contrast,
79% (15/19) and 100% (18/18) of animals treated with
hydroxocobalamin 75 mg/kg or 150 mg/kg, respectively,
survived through 14 days after poisoning, with no neurologic or
other sequelae and no significant effects on clinical chemistry or
hematology tests. Administration of hydroxocobalamin was
associated with beneficial increases in blood pressure beginning
1 to 3 minutes after initiation of infusion. The
hydroxocobalamin doses used were designed to mimic
recommended doses (5 to 10 g) administered to humans.36
Human Studies, Case Series, and Reports
In humans, the published efficacy data for hydroxocobalamin
consist of 1 prospective and 1 retrospective study in victims of
smoke inhalation,40,41 as well as case reports and case series.
Preclinical studies in heavily smoking human volunteers showed
clearing of the small amounts of cyanide (⬍2 ␮mol/L)
detectable in the blood of heavy smokers.42
A prospective open-label noncomparative trial in victims of
enclosed space fire-smoke inhalation (defined as extrication
from an enclosed space fire scene; soot in the nose, mouth, or
throat; and some degree of neurologic impairment) was
performed in Paris from 1989 to 1994.40 Results have been
published in abstract form. Of 69 patients, 37 were comatose
and 14 were initially in cardiopulmonary arrest (apneic and
pulseless). In addition to standard resuscitation measures,
hydroxocobalamin was administered in the out-of-hospital
setting (5 to 15 g intravenously). The survival rate was 72%
(50/69). There were no neuropsychiatric sequelae in the
majority (41/50) of the survivors (82%). The main causes of
Annals of Emergency Medicine 807
Empiric Treatments for Cyanide Poisoning
death were infectious complications (unrelated to cyanide
poisoning) and decerebration.
Fortin et al41 retrospectively reported effects of
hydroxocobalamin in 101 smoke inhalation victims. Among 72
patients for whom survival status was known, survival rate was
41.7% after administration of hydroxocobalamin. Of the 38
patients found in cardiac arrest, 21 had a return of spontaneous
circulation during out-of-hospital care. Twelve patients were
initially hemodynamically unstable (systolic blood pressure 0
to ⱕ90 mm Hg), and 9 recovered systolic blood pressure
approximately 30 minutes after the start of hydroxocobalamin
infusion. Those with significant neurologic impairment had
little improvement after hydroxocobalamin administration, as
assessed by Glasgow Coma Scale scores.
Survival of severe acute cyanide poisoning from sources
other than smoke inhalation after administration of
hydroxocobalamin alone has been documented.43-46 In a case
series, 9 adult patients were poisoned by ingestion of cyanide
salts (n⫽7), ingestion of acetonitrile (n⫽1), and exposure to
cyanogen bromide gas (n⫽1).43 All received hydroxocobalamin
(average dose 8.1 g). Five of the 9 patients were comatose,
and 6 had a systolic blood pressure less than 90 mm Hg.
Administration of hydroxocobalamin was associated with
improved blood pressure in patients presenting with
hypotensive shock. Six patients survived, and 3 patients who
were in cardiac arrest ultimately died. All 3 patients who died
were admitted several hours after the onset of the poisoning,
when neurologic impairment seemed irreversible.
Espinoza et al47 reported 8 pediatric patients (aged 8 to 11
years) with acute cyanide poisoning from ingestion of
improperly prepared bitter cassava (Manihot esculenta) in
Venezuela. These children presented to the hospital with
vomiting, excessive weakness, respiratory failure, bradycardia,
hypotension, and cardiovascular collapse. Because of limited
drug supply, only the 4 most acutely ill children were treated
with sodium nitrite; the remaining 4 children were each treated
with 500 mg of hydroxocobalamin. Both groups of children
improved within a few minutes, remained asymptomatic
thereafter, and were discharged the following day, with normal
neurologic assessment results.
Hydroxocobalamin became available in France on an
investigational use basis in about 1970 as the Anphar-Rolland
trousse anticyanure containing 4 g of lyophilized
hydroxocobalamin to be reconstituted in 8 g of 10% sodium
thiosulfate solution stabilized with sodium sulfite. Between
1970 and 1984, 10 patients with acute cyanide poisoning
treated with the trousse anticyanure were reported, as well as 1
patient treated with 200 mg of intravenous hydroxocobalamin
without sodium thiosulfate.48-55 In only 2 cases was the
hydroxocobalamin/sodium thiosulfate antidote combination
the only cyanide antidote administered.52,54
SAFETY OF HYDROXOCOBALAMIN
In humans, a transient reddish-brown discoloration of the
skin, mucous membranes, and urine occurs from the color of
808 Annals of Emergency Medicine
Hall et al
the medication itself. The discoloration clears during several
days in poisoned patients administered an antidotal dose14 and
appears to be harmless. In burn patients, discoloration of the
skin and exudates can complicate the subsequent evaluation of
lesion depth if this is not done before hydroxocobalamin
administration.56
In the serum, the peak light absorption of hydroxocobalamin
at 325, 364, 525, and 564.5 nm can interfere with colorimetric
blood chemistry analyses for liver enzymes, bilirubin, creatinine,
creatine kinase, phosphorus, glucose, magnesium, and iron,
which are autoanalyzer-dependent. Either falsely high or low
results may be obtained, but none appear to be of major clinical
significance.57-60
After chronic administration, rare cases of anaphylaxis,
anaphylactoid reactions (including bronchospasm and
oropharyngeal angioedema), pruritus, or urticaria have been
reported in patients receiving low-dose hydroxocobalamin (1 to
10 mg intramuscularly per month) for vitamin B12 deficiency or
other indications.61-67 One case of transient facial acne49 and 1
case of urticaria and Quinke’s edema50 have been reported in
patients receiving antidotal doses. Both patients were
administered an older formulation of 4 g of lyophilized
hydroxocobalamin dissolved in 10% sodium thiosulfate solution
stabilized with sodium sulfite, and dicobalt-EDTA was also
administered. The current hydroxocobalamin formulation
available in France contains lyophilized hydroxocobalamin
alone.
In normal human volunteers administered 5 g of
hydroxocobalamin intravenously during 20 minutes, mild,
transient, self-limited hypertension (mean increase of 13.6%
systolic and 25.9% diastolic blood pressures) accompanied by
reflex bradycardia (mean decrease of 16.3% in pulse rate) was
noted.42 There were no changes on ECG and no clinically
significant abnormalities of CBC counts, liver and kidney
function test results, serum electrolyte levels, prothrombin and
partial thromboplastin times, or blood glucose levels. All
subjects remained asymptomatic and all findings returned to
baseline without treatment.
A randomized, double-blind, placebo-controlled, ascending
dose study was conducted to evaluate the safety of
hydroxocobalamin in healthy volunteers.68 One hundred
thirty-six volunteers were randomized to receive either
hydroxocobalamin (2.5 g, 5 g, 7.5 g, 10 g) or placebo by
intravenous infusion during 7.5 to 30 minutes. The most
common hydroxocobalamin-related effects were chromaturia
and reddening of the skin. Other common adverse events were
pustular or papular rash, headache, erythema at the injection
site, decrease in lymphocyte percentage, and chest pressure.
Hydroxocobalamin was associated with a modest increase in
blood pressure that was self-limiting. The clinical relevance of
the increase in blood pressure in this and other studies is unclear
but may be desirable in cyanide poisoning because it avoids the
hypotension associated with nitrite-containing cyanide
antidotes.12 Two allergic reactions occurred within minutes of
Volume , .  : June 
Hall et al
initiation of infusion in volunteers assigned to receive the 5-g
and 10-g doses, respectively, and were successfully managed
with dexamethasone, dimethindene maleate, or both. This
finding is similar to isolated reports of allergic reactions to
hydroxocobalamin during chronic intramuscular administration
for vitamin B12 deficiency.61-64 More severe manifestations,
including wheezing, hypotension, or stridor, did not occur.
No cases of allergic reactions have been reported after
administration to 170 acutely poisoned patients.40,41
Breton et al69 reported 2 children who experienced smoke
inhalation and presented comatose, covered in soot, but without
burns. Both patients received hydroxocobalamin in the out-of-
hospital setting. The 11-month-old child was cyanotic but not
acidotic. The 4 1/2-year-old child was not cyanotic and had
an initial pH of 7.21; however, the arterial blood gas values
suggested a respiratory rather than a lactic acidosis. Neither had
plasma lactate concentrations suggestive of a significant cyanide
poisoning component to their smoke inhalation (plasma lactate
concentrations ⬍10.0 mmol/L). Both children developed
chronic respiratory sequelae. Although neither child appeared
to have significant cyanide poisoning, out-of-hospital
administration of the recommended hydroxocobalamin dose
was not associated with any adverse effects.
EFFICACY OF SODIUM THIOSULFATE
Sodium thiosulfate removes cyanide from the blood through
the action of the enzyme rhodanese.10 Sodium thiosulfate has
limited distribution into the brain, an organ highly susceptible
to the effects of cyanide-induced histotoxic anoxia, and has
limited penetration into the mitochondria, where the
endogenous cyanide-detoxifying enzyme rhodanese is located.18
No clinical trials of sodium thiosulfate are available, and efficacy
has been extrapolated from case studies and series of acute
cyanide poisoning.
Animal Studies
Most experimental animal studies have compared sodium
thiosulfate alone to its combination with other antidotes, most
often sodium nitrite or hydroxocobalamin. In dog studies with
a continuous infusion of hydrocyanic acid at 0.1 mg/kg per
minute, Hug and Marenzi70 reported that both methylene blue
and sodium nitrite were clearly more efficacious than sodium
thiosulfate alone.
Kinetic studies have shown that administration of sodium
thiosulfate accelerates by more than 3 times the conversion of
cyanide to much less toxic thiocyanate by the endogenous
rhodanese enzyme.71 However, sodium thiosulfate has generally
been thought to have only a preventive action and not a curative
action (eg, it has to be administered before the onset of cyanide
poisoning symptoms and signs to be efficacious). In studies of
rabbits administered potassium cyanide by gastric instillation,
Hug and Marenzi70 found that 7 of 10 animals died despite
administration of 1 g of sodium thiosulfate intravenously after
cyanide administration. They also noted that sodium
Volume , .  : June 
Empiric Treatments for Cyanide Poisoning
thiosulfate, which was practically nontoxic in normal animals,
caused shock and decreased the time to cardiac arrest and death
when subsequently administered to the poisoned animals. These
effects were also noted when sodium thiosulfate was mixed with
sodium nitrite and administered. The effects were exacerbated
when the sodium thiosulfate dose was increased. The best
results were obtained when sodium nitrite was administered
first, followed by sodium thiosulfate. Hug72 concluded that
sodium thiosulfate “. . . requires a certain latent period to
exert its antidotal action . . . it has a preventive [eg, must be
administered prophylactically before the cyanide] but not a
curative [eg, efficacious after poisoning symptoms develop]
effect on cyanide poisoning.”
In a sodium cyanide continuous infusion dog model, sodium
thiosulfate was not effective if administered any later in the
poisoning than the beginning of the phase of primary apnea.73
Sodium nitrite with sodium thiosulfate, hydroxocobalamin with
sodium thiosulfate, 4-DMAP, and dicobalt-EDTA were all
effective at later phases of the poisoning.73
In a recent study, rats were administered potassium cyanide
intraperitoneally, followed 6 to 16 minutes later (when
blood cyanide concentrations were expected to peak) by
intraperitoneal sodium thiosulfate or normal saline solution at
225 mg/kg.74 In the sodium thiosulfate–treated animals, there
were significantly lower arterial blood cyanide concentrations,
arterial and venous lactate concentrations, and venous PO2s,
indicating curative efficacy for improving these cyanide toxicity
markers.74 The authors observed that caution should be used in
extrapolating these results to humans.74 The intraperitoneal
route is an unlikely route of cyanide exposure and would not be
used for sodium thiosulfate administration in humans. The dose
used was somewhat higher than that recommended in humans
(approximately 179 mg/kg), and survival or death was not an
endpoint.
Human Studies, Case Series, and Case Reports
No clinical trials in humans that assess the efficacy of sodium
thiosulfate alone were found. Case reports have associated
sodium thiosulfate with survival. Chin and Calderon75
described a 19-year-old woman who drank an unknown
substance containing a “high concentration” of cyanide. Coma,
hypotension, and dilated pupils were noted. Initial arterial
blood gases showed a severe acidosis, with a pH of 7.01, and the
plasma lactate level was 10.0 mmol/L. Initial treatment with
12.5 g of sodium thiosulfate intravenously was followed by
improvement of the pH to 7.17. The patient survived with
sodium nitrite and further sodium thiosulfate plus supportive
measures, but sequelae of short-term memory difficulties and
bradykinesia developed.
Perrson19 described 5 cases of cyanide poisoning, 4 treated
with sodium thiosulfate alone and 1 treated initially with
sodium thiosulfate followed by dicobalt-EDTA. For some of
these patients, blood cyanide concentrations were reported.
Potentially toxic concentrations of cyanide are 1 mg/L
(39 ␮mol/L), and potentially lethal concentrations are
Annals of Emergency Medicine 809
Empiric Treatments for Cyanide Poisoning
approximately 2.7 mg/L (100 ␮mol/L), as reported in forensic
medicine.44
Twin brothers aged 2 1/2 years were found unconscious
during an enclosed space fire. Both were severely acidotic on
presentation. Carbon monoxide concentrations were not
significantly elevated. Treatment with oxygen, hyperbaric
oxygen, and approximately 400 mg/kg of sodium thiosulfate
was followed by recovery without sequelae. Blood cyanide
concentrations were 1.15 mg/L and 1.1 mg/L.19
A 28-year-old man ingested an unknown amount of a
cyanide solution. On presentation, he was talkative, anxious,
and hyperventilating. Sodium thiosulfate, 15 g, was
administered by slow intravenous infusion. This patient never
lost consciousness but became calm and mentally clear, and the
mild acidosis resolved. The blood cyanide concentration was
9.9 mg/L.19 This cyanide concentration is inconsistent with the
clinical presentation and course.
A 32-year-old man was deeply comatose and slightly cyanotic
after ingesting an unknown amount of a potassium cyanide
solution. There was improvement 30 minutes after
administration of oxygen and 8 g of sodium thiosulfate.
Dicobalt-EDTA was then administered, resulting in return of
normal consciousness. The blood cyanide concentration was
3.7 mg/L.19
A 54-year-old man was comatose and in respiratory arrest
after ingesting 3 g of potassium cyanide. Metabolic acidosis was
present (pH 7.31). An initial dose of 12 g of sodium thiosulfate
was administered, followed by an additional 3-g dose.
Spontaneous respiration resumed 30 minutes later, and he was
alert at 6 hours after admission. There were no sequelae. The
blood cyanide concentration was 0.26 mg/L.19
Of 4 previously published cases reviewed by Perrson,19
3 involved cyanide toxicity from infusion of sodium
nitroprusside, which improved after administration of sodium
thiosulfate. The fourth was a 30-year-old man who ingested an
unknown amount of a cyanide-containing insecticide and was
found comatose, apneic, and cyanotic 30 minutes later.
Supportive measures and administration of sodium thiosulfate
(dose not specified) was followed by return of consciousness,
although persistent neurologic symptoms developed.
A 31-year-old male Japanese patient was found in bed
comatose and having seizures after ingesting potassium
cyanide.76 The patient had severe metabolic acidosis, an
elevated central venous PO2, and normal oxygen saturation level,
consistent with cyanide poisoning. Three pearls of amyl nitrite
were administered, followed by 10 g sodium thiosulfate. The
patient recovered during the following 30 minutes. Cyanide
concentration in stored plasma was found to be approximately
1.2 mg/L when measured 6 months later.76
Two Malaysian children developed acute cassava poisoning
after eating “tapioca” cake (M utilissima).77 Both had vomiting,
drowsiness, and weakness. The 2 1/2-year-old boy also had
reactive mydriasis, and the 1 1/2-year-old girl had dyspnea
and cyanosis. Potential metabolic acidosis was reflected as a
810 Annals of Emergency Medicine
Hall et al
decreased alkali reserve (7.2 mmol/L) in both children. Both
recovered with supportive care and intravenous infusion of
50 mL of 25% sodium thiosulfate and made full recoveries.
One of 7 patients exposed to hydrogen cyanide by inhalation
in a chemical trading company office, a 19-year-old woman was
discovered unconscious and hypotensive (blood pressure 77/65
mm Hg), with severe metabolic acidosis.78 The patient received
amyl nitrite, and sodium thiosulfate 12.5 g was administered
intravenously. The patient had repeated seizures, requiring
anticonvulsant therapy. She was extubated 15 hours after
admission. At 1-year follow-up, complaints of mild impairment
of recent memory and concentration ability were confirmed by
neuropsychological testing.
A 23-year-old German man attempted suicide by ingesting
1,500 mg of potassium cyanide.20 Six hours after hospital
admission, his blood cyanide level was 6 mg/L. Supportive
treatments, administration of supplemental oxygen, correction
of metabolic acidosis, and administration of 1 g of sodium
thiosulfate per hour for 24 hours were associated with survival.
SAFETY OF SODIUM THIOSULFATE
Dogs administered sodium thiosulfate at 3,000 mg/kg
intravenously developed metabolic acidosis, hypoxemia,
hypernatremia, ECG abnormalities, and changes in blood
pressure.19 Dogs and rabbits administered sodium thiosulfate
at 500 to 4,000 mg/kg intravenously developed hypotension.31
For comparison, the recommended adult dose of sodium
thiosulfate from the Cyanide Antidote kit is 12.5 g, or about
179 mg/kg for a 70-kg person.
In normal volunteers administered 4 g of hydroxocobalamin
plus 12.5 g of sodium thiosulfate intravenously, nausea,
retching, vomiting, and injection site pain, irritation, and a
burning sensation were observed.42,79 When sodium thiosulfate
was removed from the study protocol, these effects were no
longer observed, suggesting that they were due to sodium
thiosulfate. In another normal volunteer study of sodium
thiosulfate infusion alone, nausea and vomiting were also
observed.80
CONCLUSION
The treatment of suspected cyanide poisoning presents a
dilemma for first-response emergency personnel. It is important
to initiate treatment as soon as possible, but some available
antidotes have undesirable safety profiles and cannot be
administered empirically in an out-of-hospital setting.
The evidence available to assess the 2 cyanide antidotes that
can be used empirically, sodium thiosulfate or
hydroxocobalamin, is incomplete. Some practitioners in the
United States have used sodium thiosulfate, but there are no
clinical trials to assess efficacy. In France and other countries,
hydroxocobalamin is used for empiric treatment, and there are
retrospective and prospective clinical studies. Because of ethical
considerations, prospective, controlled clinical trial efficacy data
in humans are not available for either agent, and there are no
animal or human comparative studies.
Volume , .  : June 
Hall et al
Based on recent safety and efficacy studies in animals, safety
studies in healthy volunteers, and uncontrolled efficacy studies
in humans, hydroxocobalamin seems to be an appropriate
antidote for empiric treatment of smoke inhalation and other
suspected cyanide poisoning for victims in the out-of-hospital
setting. It is not associated with life-threatening hypotension or
methemoglobinemia found with nitrite-containing cyanide
antidotes and can be administered in cases of suspected cyanide
poisoning and when concurrent carbon monoxide poisoning
occurs. Sodium thiosulfate can also be administered in the out-
of-hospital setting. The efficacy of sodium thiosulfate is based
on individual case studies, and there are contradictory
conclusions about sodium thiosulfate efficacy in animal models.
The onset of antidotal action of sodium thiosulfate may be too
slow for it to be the only cyanide antidote for emergency use.
Because there is a concern of chemical interactions when these 2
cyanide antidotes are administered simultaneously,81 they
should not be administered in the same intravenous line.
Hydroxocobalamin is currently provided in the United
States only in a dilute formulation that is not practical for use in
cyanide poisoning. A more appropriate formulation is being
developed for introduction in the United States. If this drug is
approved, emergency physicians will have a new option for
empiric treatment of cyanide poisoning in the out-of-hospital
setting.
The authors wish to thank Patrice C. Ferriola, PhD,
KZEPharmAssociates, LLC, Chapel Hill, NC, for assistance with
article formatting and preparation.
Supervising editor: Donald M. Yealy, MD
Funding and support: Dr. Hall’s work on this article was
supported in part by EMD Pharmaceuticals, Inc., Durham, NC,
an affiliate of Merck KGaA and the US developer of the
cyanide antidote hydroxocobalamin (Cyanokit). Drs. Hall, Dart,
and Bogdan are members of the Cyanokit Advisory Board of
EMD Pharmaceuticals, Inc.
Publication dates: Received for publication April 13, 2006.
Revision received August 31, 2006. Accepted for publication
September 25, 2006. Available online November 13, 2006.
Address for reprints: Richard Hart, MD, PhD, Rocky Mountain
Poison and Drug Center, 770 Bannock Street, MC 0180,
Denver, CO 80204.
Address for correspondence: Alan H. Hall, MD, TCMTS, Inc.,
PO Box 184, Mile 5.0 Pass Creek Road, Elk Mountain, WY
82324; 307-348-7371, fax 307-348-7372; E-mail
ahalltoxic@msn.com.
REFERENCES
1. Alarie Y. Toxicity of fire smoke. Crit Rev Toxicol. 2002;2:259-289.
2. Canfield D, Chatuverdi A, Dubowski K. Carboxyhemoglobin and
blood cyanide concentrations in relation to aviation accidents.
Aviat Space Environ Med. 2005;76:978-980.
3. Chatuverdi A, Sanders D. Aircraft fires, smoke toxicity, and
survival. Aviat Space Environ Med. 1996;67:275-278.
Volume , .  : June 
Empiric Treatments for Cyanide Poisoning
4. Stefanidou M, Athanaselis S. Toxicological aspects of fire. Vet
Hum Toxicol. 2004;46:196-199.
5. Borron SW, Baud FJ. Toxicity, cyanide: updated May 18, 2006.
Available at: http://www.emedicine.com/emerg/topic118.htm.
Accessed June 26, 2006.
6. Lee-Chiong TL Jr. Smoke inhalation injury. Postgrad Med. 1999;
105:55-62.
7. Rotenberg JS. Cyanide as a weapon of terror. Pediatr Ann. 2003;
32:236-240.
8. Keim ME. Terrorism involving cyanide: the prospect of improving
preparedness in the prehospital setting. Prehospital Disaster
Med. 2006;21:s56-s60.
9. Agency for Toxic Substances and Disease Registry, US
Department of Health and Human Services, Public Health Service.
Cyanide toxicity. Am Fam Physician. 1993;48:107-114.
10. Hall AH, Rumack BH. Clinical toxicology of cyanide. Ann Emerg
Med. 1986;15:1067-1074.
11. Bowden CA, Krenzelok EP. Clinical applications of commonly used
contemporary antidotes: a US perspective. Drug Saf. 1997;16:
9-47.
12. Hall A, Kulig K, Rumack B. Suspected cyanide poisoning in
smoke inhalation: complications of sodium nitrite therapy.
J Toxicol Clin Toxicol. 1989;9:3-9.
13. Hall AH, Rumack BH. Hydroxycobalamin/sodium thiosulfate as a
cyanide antidote. J Emerg Med. 1987;5:115-121.
14. Megarbane B, Delahaye A, Goldgran-Toledano D, et al. Antidotal
treatment of cyanide poisoning. J Chin Med Assoc. 2003;66:193-
203.
15. Brouard A, Blaisot B, Bismuth C. Hydroxocobalamine in cyanide
poisoning. J Toxicol Clin Exp. 1987;7:155-168.
16. Sauer SW, Keim ME. Hydroxocobalamin: improved public health
readiness for cyanide disasters. Ann Emerg Med. 2001;37:635-
641.
17. Santiago I. Gas poisoning. An Sist Sanit Navar. 2003;26(suppl
1):173-180.
18. Baskin SI, Horowitz AM, Nealley EW. The antidotal action of
sodium nitrite and sodium thiosulfate against cyanide poisoning.
J Clin Pharmacol. 1992;32:368-375.
19. Perrson H. Sodium thiosulfate. In: Meredith TJ, Haines JA, Berger
J-C, et al, eds. IPCS/CEC Evaluation of Antidotes Series, Vol. 2:
Antidotes for Poisoning by Cyanide. Cambridge, UK: Cambridge
University Press; 1993:43-64.
20. Bismuth C. Hydroxocobalamin. In: Meredith TJ, Haines JA, Berger
J-C, et al, eds. IPCS/CEC Evaluation of Antidotes Series, Vol. 2:
Antidotes for Poisoning by Cyanide. Cambridge, UK: Cambridge
University Press; 1993:65-77.
21. Hantson P, Benaissa L, Baud F. Smoke poisoning. Presse Med.
1999;28:1949-1954.
22. Personne M, Kulling P. Vitamin B12-variant in cyanide poisoning:
hydroxycobalamin is a new effective antidote. Lakartidningen.
1997;94:877-878.
23. Lau FL. Emergency management of poisoning in Hong Kong. Hong
Kong Med J. 2000;6:288-292.
24. Kuroki Y, Endo Y, Masatono K, et al. Non-approved antidotes in
Japan: foreign antidotes imported for poisoning countermeasures
at the Summit 2000. Chudoku Kenkyu. 2001;14:259-267.
25. Møeller S, Hemmingsen C. Science and practice: case reports:
cyanide poisoning. Ugeskr Laeger. 2003;165:2579-2580.
26. Mushett CW, Kelley KL, Boxer GE, et al. Antidotal efficacy of
vitamin B12a (hydroxo-cobalamin) in experimental cyanide
poisoning. Proc Soc Exp Biol Med. 1952;81:234-237.
27. Astier A, Baud FJ. Complexation of intracellular cyanide by
hydroxocobalamin using a human cellular model. Hum Exp
Toxicol. 1996;15:19-25.
Annals of Emergency Medicine 811
Empiric Treatments for Cyanide Poisoning
28. Van den Berg MP, Merkus P, Romeijn SG, et al.
Hydroxocobalamin uptake into the cerebrospinal fluid after nasal
and intravenous delivery in rats and humans. J Drug Target.
2003;11:325-331.
29. Faure J, Vincent M, Benabid A, et al. Treatment with
hydroxocobalamin for hypercyanemia and hypercyanuria following
compression of the sciatic nerve in the rabbit. C R Soc Biol
(Paris). 1977;171:1221-1226.
30. Vincent M, Vincent F, Marka C, et al. Cyanide and its relationship
to nervous suffering: physiopathological aspects of intoxication.
Clin Toxicol. 1981;18:1519-1527.
31. Mizoule J. Study of the Antidotal Action of Hydroxocobalamin.
Paris, France: Université de Paris: Faculté de Pharmacie; 1966.
32. Posner MA, Tobey RE, McElroy H. Hydroxocobalamin therapy of
cyanide intoxication in guinea pigs. Anesthesiology. 1976;44:
157-160.
33. Krapez JR, Vesey CJ, Adams L, et al. Effects of cyanide antidotes
used with sodium nitroprusside infusions: sodium thiosulphate
and hydroxocobalamin given prophylactically to dogs. Br J
Anaesth. 1981;53:793-804.
34. Rose CL, Chen KK, Harris PN. Mercuric cyanide poisoning and its
treatment in dogs. Proc Soc Exp Biol Med (New York). 1964;116:
371-373.
35. Ivankovich AD, Braverman B, Kanuru RP, et al. Cyanide antidotes
and methods of their administration in dogs: a comparative
study. Anesthesiology. 1980;52:210-216.
36. Paulet G, Olivier M. On the subject of the therapeutic value of
hydroxocobalamin in hydrocyanic acid poisoning. Ann Pharm Fr.
1963;21:133-137.
37. Paulet G. On the subject of the treatment of cyanide poisoning
with chelates of cobalt. Urgence Med Chirurg. 1965;II:611-613.
38. Borron S, Stonerook M, Reid K. Efficacy of hydroxocobalamin for
the treatment of acute cyanide poisoning in adult beagle dogs.
Clin Toxicol. 2006;44:5-15.
39. Posner MA, Rodkey FL, Tobey RE. Nitroprusside-induced cyanide
poisoning: antidotal effect of hydroxocobalamin. Anesthesiology.
1976;44:330-335.
40. Borron S, Barriot P, Imbert M, et al. Hydroxocobalamin for empiric
treatment of smoke inhalation-associated cyanide poisoning:
results of a prospective study in the prehospital setting. Ann
Emerg Med. 2005;46:S77.
41. Fortin JL, Ruttiman M, Domanski L, et al. Prehospital use of
hydroxocobalamin for smoke inhalation in the Paris fire brigade.
Clin Toxicol. 2006;44:37-44.
42. Forsyth JC, Mueller PD, Becker CE, et al. Hydroxocobalamin as a
cyanide antidote: safety, efficacy, and pharmacokinetics in heavily
smoking normal volunteers. J Toxicol Clin Toxicol. 1993;31:277-
294.
43. Borron SW, Megarbane B, Baud F. Hydroxocobalamin is an
effective antidote in severe acute cyanide poisoning in man. Int J
Toxicol. 2004;23:399-400.
44. Bromley J, Hughes BG, Leong DC, et al. Life-threatening
interaction between complementary medicines: cyanide toxicity
following ingestion of amygdalin and vitamin C. Ann
Pharmacother. 2005;39:1566-1569.
45. Fortin J, Waroux S, Ruttiman M. Hydroxocobalamin for poisoning
caused by ingestion of potassium cyanide. Clin Toxicol. 2005;43:
731.
46. Weng TI, Fang CC, Lin SM, et al. Elevated plasma cyanide level
after hydroxocobalamin infusion for cyanide poisoning. Am J
Emerg Med. 2004;22:492-493.
47. Espinoza OB, Perez M, Ramirez MS. Bitter cassava
poisoning in eight children: a case report. Vet Hum Toxicol.
1992;34:65.
812 Annals of Emergency Medicine
Hall et al
48. Bismuth C, Cantineau JP, Pontal P, et al. Priority of oxygenation
in cyanide poisoning. J Toxicol Med. 1984;4:107-121.
49. Motin J, Bouletreau P, Rouzioux JM. Severe cyanide poisoning
treated successfully with hydroxocobalamin. J Med Strasbourg.
1970;1:717-722.
50. Yacoub M, Faure J, Morena H, et al. Acute cyanide poisoning:
current data on the metabolism of cyanide and treatment with
hydroxocobalamin. Eur J Toxicol Environ Hyg. 1974;7:22-29.
51. Lutier F, Dusoleil P, de Montgros J. Action of a massive dose of
hydroxocobalamin in acute cyanide poisoning: apropos of a case.
Arch Mal Prof. 1971;32:683-686.
52. Racle JP, Chausset R, Dissait F. Acute cyanide poisoning:
biological, toxicological, clinical, and therapeutic aspects. Rev
Med Clermont Ferrand. 1976;5:371-382.
53. Bourrelier J, Paulet G. Hydrocyanic poisoning following severe
burns by fused sodium cyanate: 3 cases treated with EDTACO2.
Presse Med. 1971;79:1013-1014.
54. Jouglard J, Nava G, Botta A, et al. Regarding an acute poisoning
with potassium cyanide treated with hydroxocobalamin. Marseille
Med. 1974;12:617-624.
55. Bismuth C, Baud FJ, Djeghout H, et al. Cyanide poisoning from
propionitrile exposure. J Emerg Med. 1987;5:191-195.
56. Perro G, Cutillas M, Maachi B, et al. Plasma lactate levels and
the calculated anion gap are not predictive of cyanide poisoning
in fire burns. Brûlures. 2002;3:13-15.
57. Curry SC, Connor DA, Raschke RA. Effect of the cyanide antidote
hydroxocobalamin on commonly ordered serum chemistry studies.
Ann Emerg Med. 1994;24:65-67.
58. Gourlain H, Caliez C, Laforge M, et al. Study of the mechanisms
involved in hydroxocobalamin interference with determination of
some biochemical parameters. Ann Biol Clin (Paris). 1994;52:
121-124.
59. Vest P, Renaudeau C, Tellal S, et al. Interference of
hydroxocobalamin treatment on common biochemical
determinations. Ann Biol Clin (Paris). 2002;60:57-64.
60. Beaudeux JL, Flourie F, Peynet J, et al. Hydroxocobalamin and
sodium thiosulfate interfere negatively with measurement of
creatine kinase activity. Clin Chem. 1994;40:2120-2121.
61. Auzepy P, Veissieres JF, Deparis M. Anaphylactic shock due to
hydroxocobalamine [letter]. Nouv Presse Med. 1974;3:152.
62. Branco-Ferreira M, Clode MH, Pereira-Barbosa MA, et al. Anaphylactic
reaction to hydroxycobalamin. Allergy. 1997;52:118-119.
63. Dally S, Gaultier M. Anaphylactic shock caused by
hydroxocobalamin. Nouv Presse Med. 1976;5:1917.
64. Hovding G. Anaphylactic reaction after injection of vitamin B12.
BMJ. 1968;3:102.
65. James J, Warin RP. Sensitivity to cyanocobalamin and
hydroxocobalamin. BMJ. 1971;2:262.
66. Heyworth-Smith D, Hogan PG. Allergy to hydroxycobalamin, with
tolerance of cyanocobalamin. Med J Aust. 2002;177:162-163.
67. Vidal C, Lorenzo A. Anaphylactoid reaction to hydroxycobalamin
with tolerance of cyanocobalamin. Postgrad Med J. 1998;74:702.
68. Uhl W, Nolting A, Golor G, et al. Safety of hydroxocobalamin in
healthy volunteers in a randomized, placebo-controlled study. Clin
Toxicol. 2006;44:17-28.
69. Breton D, Jouvet P, de Blic J, et al. Toxicity of fire smoke:
apropos of 2 pediatric cases. Arch Fr Pediatr. 1993;50:43-45.
70. Hug E, Marenzi A. Binding of hydrocyanic acid by red blood cells
which contain methemoglobin. C R Soc Biol. 1933;114:84-86.
71. Paulet G. Current state of treatment for cyanide poisoning. Arch
Mal Prof. 1984;45:185-192.
72. Hug E. L’intoxication par l’acide cyanhydrique [Poisoning by
hydrocyanic acid: antidotal action of methylene blue, sodium
nitrite, and sodium sulfide (Na2S)]. Comptes Rend Soc Biol.
1932;111:89-90.
Volume , .  : June 
Hall et al Empiric Treatments for Cyanide Poisoning

73. Paulet G, Dassonville J. The value of dimethylaminophenol
(DMAP) in treatment of cyanhydride poisoning: experimental
study. J Toxicol Clin Exp. 1985;5:105-111.
74. Renard C, Borron SW, Renaudeau C, et al. Sodium thiosulfate
for acute cyanide poisoning: study in a rat model. Ann Pharm Fr.
2005;63:154-161.
75. Chin RG, Calderon Y. Acute cyanide poisoning: a case report.
J Emerg Med. 2000;18:441-445.
76. Nakatani T, Kosugi Y, Mori A, et al. Changes in the parameters of
oxygen metabolism in a clinical course recovering from potassium
cyanide. Am J Emerg Med. 1993;11:213-217.
77. Cheok SS. Acute cassava poisoning in children in Sarawak. Trop
Doct. 1978;8:99-101.
78. Lam KK, Lau FL. An incident of hydrogen cyanide poisoning. Am
J Emerg Med. 2000;18:172-175.
79. Osterloh J, Hall AH. Hydroxocobalamin analysis and
pharmacokinetics [letter]. Clin Toxicol. 1997;35:409-411.
80. Ivankovich AD, Braverman B, Stephens TS, et al. Sodium
thiosulfate disposition in humans: relation to sodium
nitroprusside toxicity. Anesthesiology. 1983;58:11-17.
81. Evans CL. Cobalt compounds as antidotes for hydrocyanic acid.
Br J Pharmacol Chemother. 1964,23:455-475.

Request for Abstracts for ACEP’s Research Forum (non-moderated)

Researchers have a unique opportunity to showcase emergency medicine research published or presented in other
specialties’ journals or meetings in the past year.
ACEP’s Research Forum is providing emergency medicine researchers with another opportunity this year to present
scientific emergency medicine research at the 2007 conference, which will be held October 8-9 in conjunction with
Scientific Assembly in Seattle, WA.
Abstracts from emergency physicians who have presented or published in non-emergency medicine specialty meetings or
journals within the past 12 months will be considered. Case reports or subject reviews are not considered original
research. These abstracts will be accepted on a space available basis as non-moderated posters. If accepted the
presenter is obligated to be available to discuss their poster(s) with Research Forum attendees.
This is an excellent venue to showcase outstanding emergency medicine research from competing meetings or journals.
Please submit your research absract(s) to the academic affairs department by September 7, 2007 at
academicaffairs@acep.org, or by fax at 972-580-2816. For questions, please call 800-798-1822, ext. 3291.

Volume , .  : June  Annals of Emergency Medicine 813