Article, Toxicology

Lethal acute poisoning with potassium ferrocyanide

Case Report

Lethal acute poisoning with potassium ferrocyanide Abstract

We report the first case ever published of death after acute poisoning with potassium ferrocyanide. A 56-year-old retirED pharmacist who was practicing oenology called the emergency telephone service to tell that he had just ingested 2 glasses of potassium ferrocyanide. He had vomiting and then collapsed within less than 1 hour. Later on, aggressive cardiopulmonary resuscitation rapidly restored stable hemo- dynamics. Although laboratory findings evidenced the hallmark of Cyanide poisoning, which is dramatically elevated lactate concentrations, Blood cyanide level (0.70 mg/L) was below lethal values, which typically exceed 2.5 to 3.0 mg/L. However, this finding should be interpreted taking into account the late blood sampling, the prior administration of the antidote hydroxocobalamine, and the delayed assay. Hyperkalemia, which was rapidly suspected on electrocardiogram findings, was confirmed and found to be moderate (6.5 mmol/L), but after a quick onset of potassium-lowering therapy. The Resuscitation procedure was stopped after 3 days because of cerebral death, which was quickly suspected the day after intoxication. The death was pronounced 4 days after poisoning. The respective role of cyanide or potassium release from potassium ferrocyanide is discussed.

Hexacyanoferrate compounds are widely considered not to be cyanogenic. Some of them have even been approved for use as antidotes in cesium and thallium intoxications. Nevertheless, very few data are available on their actual cyanide-releasing properties, or toxicity profile. We report the first case of death ever published after acute human poisoning with potassium ferrocyanide.

On September 30, 2008, a 56-year-old retired pharmacist who was practicing oenology called the emergency telephone service telling that he had just ingested 2 glasses of potassium ferrocyanide as well as an unknown amount of clopidogrel, zopiclone, and slow-release morphine sulfate at 9:00 AM. His medical history included obesity (101 kg body weight), chronic alcoholism, and stent surgery for coronary disease. His current drug treatment consisted of clopidogrel only. Because telephone contact

was rapidly lost, a paramedic unit was immediately sent out. He was found in cardiorespiratory arrest, lying in bed with vomiting and clopidogrel tablets all around him. On a nearby table, there was a plastic container labeled potassium ferrocyanide with a yellow crystalline powder inside. A semiautomated external defibrillator was used, but no Electric shock was delivered because of asystolia. Cardio- pulmonary resuscitation was given for 10 minutes until the arrival of the mobile Emergency medical care unit. An electrocardiographic recording confirmed asystolia, and the patient presented with a reactive mydriasis. Endotracheal intubation and mechanical ventilation with 100% oxygen were installed. Gastric aspiration evidenced blood in the stomach. Cardiac activity was restored after intravenous injection of 2 mg epinephrine. Initially widened QRS complexes rapidly normalized. During transfer to the hospital, 2 episodes of bradycardia required 2 further intra- venous injections of 2 mg epinephrine. The patient was admitted in the intensive care unit at 10:00 AM. Shortly after admission, asystolia recurred and the patient presented with profuse diarrhea. Stable hemodynamics was achieved within 30 minutes after repeated administrations of intravenous epinephrine (totaling 29 mg), bicarbonate 4.2% (3 x 250 mL), and Fluid replacement. Hydroxocoba- lamin (5 g) was given intravenously without inducing any clear improvement. Hyperkalemia was suspected from electrocardiogram recordings, and a bolus of insulin 15 UI was administered followed by a 500-mL infusion of 30% dextrose. At 11:00 AM, arterial blood gas findings were as follows: pH 6.82; PaCO2, 76 mm Hg; and PaO2, 84 mm Hg. Blood laboratory tests revealed the following: lactate, above 15 mmol/L; sodium, 142 mmol/L; potassium, 6.5 mmol/L; chloride, 108 mmol/L; bicarbonate, 15 mmol/L; anion gap, 32 mmol/L; creatinine, 129 umol/L; calcium, 1.91 mmol/L; phosphorus, 2.75 mmol/L; magnesium, 1.13 mmol/L; glu- cose, 46.2 mmol/L; white blood count, 9.2 x 109/L; hemoglobin, 160 g/L; activated partial thromboplastin time, 50 seconds; prothrombin ratio, 51%; and fibrinogen, 2 g/L. A late blood sampling after stabilization of hemodynamics and after completion of hydroxocobalamin administration revealed blood cyanide level at 0.70 mg/L (normal, b0.1 mg/L) using a Head-Space gas chromatography/Mass Spectrometry (HS-GC/MS) assay performed 3 days after sampling. Toxicologic screening (high-performance liquid chromatography/ultraviolet spectrophotometry) only

0735-6757/$ – see front matter (C) 2010

642.e4 Case Report

revealed zopiclone at therapeutic serum levels (0.014 mg/L) and atracurium, which was given for mechanical ventilation. Neither morphine nor clopidogrel was evidenced. Alcohol blood level was 0.6 g/L. Controlled hyperventilation and intravenous bicarbonate failed to improve acid-base imbal- ance and hyperkalemia. Therefore, continuous veno-venous hemodiafiltration (CVVHDF) with bicarbonate-buffered fluid was used. Because of hemofilter clotting during heparin anticoagulation, CVVHDF was continued using citrate- buffered fluid (from 8:00 PM). Epinephrine (2 mg/h) was progressively reduced, and stable hemodynamics was persisting after epinephrine withdrawal at 8:00 PM. At 10:00 PM, lactic acidosis dramatically improved (lactate level, 5.1 mmol/L, and pH 7.33), and normalization was seen on the following day. However, anoxic brain damage was evidenced on electroencephalography on October 1 and 3, 2008. Therefore, CVVHDF was stopped and the patient declared dead on October 4. Potassium ferrocyanide identification was done by adding ferric chloride to the water dissolution of the yellow crystalline material: the color changed immediately to the characteristic blue of Prussian blue (PB).

To the best of our knowledge, this is the first human case of lethal acute poisoning with potassium ferrocyanide ever reported. It has initially been considered as most probably cyanide poisoning because of elevated lactate blood levels, the hallmark of such a poisoning.

Potassium ferrocyanide [K4Fe(CN)6] is a hexacyanofer- rate compound as well as potassium ferricyanide [K3Fe (CN)6] and ferric hexacyanoferrate (Fe4[Fe(CN)6]3). De- spite the presence of cyanide ions, hexacyanoferrate compounds are generally considered to be nontoxic because of the tight binding between the cyanide ions and the metal that prevents the release of free cyanide [1]. They are used in photographic processing, pigment production, and as laboratory reagents. Because hexacyanoferrate precipitates most metal ions including iron, copper, and lead, potassium ferrocyanide is also used for removing iron from wine. Other compounds, such as potassium ferricyanide (“solu- ble” or colloidal PB) and ferric hexacyanoferrate (insoluble PB), are used for the oral treatment of poisonings with radioisotopes of cesium or thallium [1,2].

Similar cyanide release from either potassium ferrocya-

nide or potassium ferricyanide enhanced by strong acids has been evidenced in vitro [3]. Although hexacyanoferrate compounds seem to be very poorly absorbed through the gastrointestinal tract [4], a blue discoloration of sweat and tears in patients treated with repeated oral doses of PB has been described [5]. The release of cyanide ions is actually minimal at physiologic blood pH. In dogs, intravenous doses of 100 mg/kg ferrocyanide were totally recovered in the urine, and they were well tolerated [6]. They only released hydrogen cyanide under the influence of a strong acid. Therefore, after oral administration, cyanide release may be facilitated by gastric and duodenal juice as shown in in vivo and in vitro studies [4,7,8]. Sustained elevation of blood

cyanide levels has been described in a patient who remained free of any biologic or clinical symptoms of cyanide poisoning after Acute ingestion of potassium ferrocyanide, but this was attributed to analytic interference in the cyanide assay [3]. In 3 male human volunteers, 0.03 mg CN/kg body weight was absorbed after the oral administration of 500 mg of potassium ferricyanide, which is considered to be approximately 20 to 100 times below the lethal human dose (0.5-3.5 mg CN/kg) [4]. Taking into account this result, doses of ferricyanide above 10 g may be harmful. However, these results are not consistent with the recommended daily dose of PB when used as an antidote, which are generally in the range of 3 to 20 g [2]. No case of cyanide poisoning has seemingly ever been reported after the therapeutic use of PB up to a daily dose of 5 g q.i.d. This discrepancy may be due to endogenous cyanide detoxification, which is not over- whelmed when divided doses are used. Furthermore, the form of PB is often not precisely defined in the literature; importantly, ferric hexacyanoferrate (insoluble PB) has the lowest CN releasing potency [7]. Thus, toxicity resulting from cyanide exposure is not expected after PB administra- tion at the recommended therapeutic oral dose, but no conclusion can be drawn regarding the human toxic dose [8]. Only 1 case of acute ingestion of 1 coffee spoon of potassium ferricyanide and 1 of an unknown amount of potassium ferrocyanide have been published [3,9]. Both patients remained asymptomatic. In our case, the acute ingested dose of potassium ferrocyanide was probably greater than the doses involved either in previous cases of acute poisoning or in the therapeutic use of hexacyanoferrate. According to the results of an in vitro study, CN release is linear with time owing to the low percentage of decomposition of the hexacyanoferrate compounds. A residence time of 1 hour in the stomach of 20 g of potassium ferrocyanide would release only 0.56 mg of cyanide, whereas assuming a residence time of 1 hour in the stomach and 23 hours in the intestine, the whole CN release would be 2.8 mg [7]. Because our patient presented with a rapid onset of symptoms, CN release in the stomach only has to be considered. From a theoretical point of view and assuming that CN release is linear with dose, the required oral dose for a rapid onset of symptoms would be at least 1795 g of potassium ferrocyanide, which is extremely unlikely.

Although our patient collapsed within less than an hour after ingestion and laboratory findings evidenced the hallmark of cyanide poisoning, namely, dramatically elevat- ed lactate Blood concentrations, his blood cyanide level was below lethal levels, which typically exceed 2.5 to 3.0 mg/L [10]. However, the interpretation of this result should take into account the late blood sampling performed after completion of hydroxocobalamin administration and the delayed analysis. Indeed, the meaningfulness of Blood cyanide levels depends on early sampling as well as careful storage and analysis because of the rapid metabolism and instability of cyanide in blood samples and the vulnerability of cyanide assays to multiple sources of interference [11].

Case Report 642.e5

Because cyanide release was likely to be low at doses compatible with the patient’s statement, the involvement of potassium released from potassium ferrocyanide should also be considered. K4Fe(CN)6 hydrolysis is a 2-step reaction. The first reaction consists of K+, Fe2+ and [Fe(CN)6]4- release. The second reaction with a dissociation constant of 10-35 mmol at physiologic pH is the hydrolysis of [Fe (CN)6]4- to Fe2+ and CN [12]. Therefore, potassium release was likely to be involved in the hyperkalemia noticed in this patient. Nevertheless, acidosis per se can be a contributing factor by driving potassium extracellularly. However, despite prompt treatment to decrease hyperkalemia initially suspected from electrocardiogram recording, kalemia was still at 6.5 mmol/L. Short onset of Severe hyperkalemia can lead to cardiac arrest, Cardiovascular collapse, and anoxia with lactic acidosis. Finally, the possible failure of hydro- xocobalamin as an antidote must be taken into account when evaluating the respective role of cyanide and hyperkalemia in the lethal outcome of this poisoning.

Christine Payen MD

Poison Center and Pharmacovigilance Department

69424 Lyon cedex 03, France E-mail address: [email protected]

Christian Combe MD

Resuscitation Unit Hospital of Villefranche-Sur-Saone, France

Catherine Le Meur PharmD

Laboratory of Medical Analysis (LAT-LUMTOX)

Lyon, France

Yvan Gaillard PharmD, PhD

Laboratory of Medical Analysis (LAT-LUMTOX)

La Voulte Sur Rhone, France

Corine Pulce MD Jacques Descotes MD, PharmD, PhD

Poison Center and Pharmacovigilance Department

69424 Lyon cedex 03, France

doi:10.1016/j.ajem.2009.09.007

References

  1. Pearce J. Studies of any toxicological effects of Prussian blue compounds in mammals–a review. Food Chem Toxicol 1994;32: 577-82.
  2. Thompson DF, Callen E. Soluble or insoluble Prussian blue for radiocesium and thallium poisoning? Ann Pharmacother 2004;38(9): 1509-14.
  3. Laforge M, Gourlain H, Fompeydie, et al. Ferrocyanide ingestion may cause false positives in cyanide determination. Clin Toxicol 1999;37 (3):337-40.
  4. Nielsen P, Dresow B, Fischer R, et al. Bioavailability of iron and cyanide from oral potassium ferric hexacyanoferrate (II) in humans. Arch Toxicol 1990;64:420-2.
  5. Hoffman RS. Thallium toxicity and the role of Prussian blue in therapy. Toxicol Rev 2003;22:29-40.
  6. Kleeman CR, Epstein FH, Rubini ME, et al. Initial distribution and fate of ferrocyanide in dogs. Am J Physiol 1955;182:548-52.
  7. Verijl JM, Joore HCA, van Dijk A, et al. In vitro cyanide release of four Prussian blue salts used for the treatment of cesium contaminated persons. Clin Toxicol 1993;31(4):553-62.
  8. Yang Y, Brownell C, Sadrieh N, et al. Quantitative measurement of cyanide release from Prussian blue. Clin Toxicol 2007;45(7): 776-81.
  9. Hantson P, N’Geye P, Laforge M, et al. Suicide attempt by ingestion of potassium ferricyanide. Clin Toxicol 1996;34(4):471-3.
  10. Ellenhorn MJ. Cyanide. Ellenhorn’s medical toxicology diagnosis and treatment of human poisoning. 2nd ed. Baltimore: Williams and Wilkins; 1997. p. 1476-84.
  11. Borron SW. Recognition and treatment of acute cyanide poisoning. J Emerg Nurs 2006;32:S12-8.
  12. Crawley FEH, Goddard EA. Internal dose from carbon-14 labelled compounds. The metabolism of carbon-14 labelled potassium cyanide in the rats. Health Phys 1977;32:135-42.