Clinical Toxicology (2010) 48, 477–484 Copyright © Informa UK, Ltd. ISSN: 1556-3650 print / 1556-9519 online DOI: 10.3109/15563650.2010.505573
REVIEW LCLT
Does amyl nitrite have a role in the management of pre-hospital mass casualty cyanide poisoning? OPHIR LAVON1 and YEDIDIA BENTUR1,2 Amyl nitrite in cyanide mass casualty incidents
1 2
Rambam Health Care Campus, Israel Poison Information Center, Haifa, Israel The Rappaport Faculty of Medicine, Technion-Israel Institute of Technology, Haifa, Israel
Context. Amyl nitrite has been recommended as a cyanide antidote for several decades. Its antidotal properties were initially attributed to induction of methemoglobin and later to a nitric oxide mediated hemodynamic effect. The ease of administration and alleged rapid clinical effect would recommend its wide use in the pre-hospital management of mass casualty cyanide poisoning; yet there are concerns regarding the use of amyl nitrite for this indication. Objective. Review the data on amyl nitrite in cyanide poisoning and evaluate its efficacy and safety in mass casualty cyanide poisoning. Methods. A literature search utilizing PubMed, Toxnet, textbooks in toxicology and pharmacology, and the bibliographies of the articles retrieved identified 17 experimental studies and human reports on the use of amyl nitrite in cyanide poisoning, and 40 additional articles on amyl nitrite’s properties and adverse effects. One paper was excluded as it was a conference abstract with limited data. Mechanisms of action. The antidotal properties of amyl nitrite were attributed initially to induction of methemoglobinemia and later to nitric oxide mediated vasodilation. Efficacy: experimental studies. Animal studies on the use of amyl nitrite in cyanide poisoning are limited, and their results are inconsistent, which makes their extrapolation to humans questionable. Efficacy: human studies. Clinical reports are limited in number and the part played by amyl nitrite relative to the other treatments administered (e.g. life support, sodium nitrite, and sodium thiosulfate) is unclear. Adverse effects. Amyl nitrite can be associated with potentially serious adverse reactions such as hypotension, syncope, excessive methemoglobinemia, and hemolysis in G6PD deficient patients. These effects are more pronounced in young children, in the elderly, and in patients with cardiac and pulmonary disorders. Dose regimen. The method of administration of amyl nitrite (breaking pearls into gauze or a handkerchief and applying it intermittently to the victim’s nose and mouth for a few minutes) is not easily controlled, might result in under- or over-dosing, can prevent the caregiver from administering life support, and possibly expose him/her to amyl nitrite’s adverse effects. Conclusions. Administration of amyl nitrite in mass casualty cyanide poisoning can result in unnecessary morbidity and may interfere with the proper management of the incident and the required supportive treatment and rapid evacuation. In the authors’ opinion these drawbacks make the use of amyl nitrite in pre-hospital mass casualty cyanide poisoning unwarranted. Keywords
Amyl nitrite; Cyanide; Poisoning; Antidote; Pre-hospital
Introduction Mass casualty cyanide poisoning resulting in high and rapid morbidity and mortality can occur in several scenarios such as fire, industrial accident, or terrorist attack.1 Appropriate preparedness to such incidents requires an efficient, safe, and easily administered antidote that can be utilized in the prehospital setting.2 Amyl nitrite was first suggested as an antidote for cyanide poisoning in 1888 by Pedigo.3 It’s simple use and alleged beneficial clinical effect suggest it as a preferred pre-hospital antidote for mass cyanide poisoning.2 However, the clinical experience and evidence based data on amyl nitrite in cyanide poisoning are limited and its potential serious
adverse effects should not be under-estimated. The objective of this article is to review the literature on amyl nitrite in cyanide poisoning and evaluate its efficacy and safety in mass casualty cyanide poisoning.
Methodology We searched PubMed between 1966 and 2009 using the search terms amyl nitrite, cyanide, poisoning, toxicity, and intoxication. In addition, we searched the database Toxnet, several textbooks of toxicology and pharmacology, and the bibliographies of the articles retrieved. The search identified 17 experimental studies and human reports on the use of amyl nitrite in cyanide poisoning, and 40 additional articles on amyl nitrite’s properties, mechanism of action, dose regimen, and adverse effects. One paper was excluded as it was a conference abstract with limited data.
Clinical Toxicology vol. 48 no. 6 2010
478 Cyanide poisoning: mechanisms, features, and management Cyanide binds to the ferric iron of cytochrome oxidase and inactivates it. This leads to uncoupling of mitochondrial oxidative phosphorylation, regardless of adequate oxygen supply, and to cellular hypoxia. As a result, cellular metabolism shifts from aerobic to anaerobic metabolism with accumulation of lactic acid. Vulnerable tissues of the brain and heart are rapidly and gravely affected.4 Prompt loss of consciousness and hemodynamic imbalance are observed. Animal studies demonstrated coronary and pulmonary vasoconstriction, decrease in arterial blood pressure, and myocardial depression shortly after cyanide exposure.5,6 Acute exposure to cyanide can result in severe clinical manifestations and even death. Sudden collapse, seizures, coma, and cardiopulmonary depression are characteristics of severe poisoning.1 The mainstays of treatment are supportive care and specific antidotes. Successful outcome in patients with severe cyanide poisoning was reported after intensive respiratory and circulatory support alone.7,8 Several antidotes including complexing agents (e.g. hydroxocobalamin), methemoglobin-inducers (e.g. sodium nitrite and amyl nitrite), and sulfur donors (e.g. sodium thiosulfate) have been used.2,9 Cyanide antidotes should be given intravenously (except for amyl nitrite) as early as possible together with intensive supportive treatment, and preferably in a medical facility.8
Chemical and pharmacological properties of amyl nitrite Amyl nitrite is an alkyl nitrite (an aliphatic ester of nitrous acid) with a molecular weight of 117.15 and a specific gravity of 0.877. It is a highly volatile clear-yellowish liquid, with a fruity odor and a vapor pressure of 85 mmHg at 24°C (75°F). It is highly flammable with a flash point at 21°C (73°F) and auto-ignition at 210°C (410°F). It is readily soluble in alcohol and ether and less soluble in water. Amyl nitrite is readily absorbed into the circulation from mucous membranes. The rate and extent of absorption is greatest from the lungs and poor from the gastrointestinal tract because of rapid hydrolysis.10 The greater part of amyl nitrite (60%) is rapidly inactivated in the liver by hydrolysis to isoamyl alcohol and nitrite;11 the rest is excreted unchanged in the urine. The elimination of amyl nitrite follows first order kinetics. Amyl nitrite has two major pharmacodynamic effects, vasodilatation (through formation of nitric oxide and production of cGMP) and methemoglobin formation (by oxidation of hemoglobin iron). Its vasodilatatory property was first reported almost a century and a half ago and a few years later it was used for the first time in the treatment of angina pectoris.12
O. Lavon and Y. Bentur
Mechanisms of action of amyl nitrite Traditionally the antidotal effect of nitrites in cyanide poisoning was attributed to methemoglobin formation.1,2,13,14 Cyanide has higher affinity to ferric iron (as in methemoglobin) than to ferrous iron (as in cytochrome oxidase), thus favoring the formation of cyanmethemoglobin. The goal of nitrite therapy has been to achieve a methemoglobin concentration of 20–30%, which is the maximal tolerated level in a healthy individual (not cyanide poisoning data). Animal studies demonstrated a rapid and marked development of methemoglobinemia after exposure to amyl nitrite. Inhalation of 0.112% v/v amyl nitrite in air by mice for a few minutes resulted in 60% methemoglobin formation.15 In cats, inhalation of amyl nitrite (0.06 and 0.12% v/v) resulted in methemoglobin concentrations of 30 and 70%, respectively. Dogs inhaling the content of a 0.3 mL ampoule of amyl nitrite for 3 min developed methemoglobinemia of up to 32%.16 Contrarily, a low concentration methemoglobinemia was observed in humans after inhalation of amyl nitrite. Six healthy volunteers who inhaled 0.1 mL amyl nitrite applied to gauze, 10 times for 20 s at 1 min intervals, developed methemoglobin concentrations of 3.45–6%. Their average diastolic blood pressure decreased from 75 to 60 mmHg, and heart rate increased from 75 to 110/min.15 Several other studies showed similar results.17,18 Methemoglobin concentrations of 3.6–9.2% were reported to be associated with favorable response in cyanide poisoned patients treated with sodium nitrite.18–20 These data are inconclusive due to the possibility of having missed the peak concentration of methemoglobin, possible interference with cyanmethemoglobin, and the concomitant use of sodium thiosulfate.1,2,21,22 The above findings imply that animal data on amyl nitriteinduced methemoglobinemia cannot be extrapolated to humans. It could be hypothesized that methemoglobin concentrations attained after administration of nitrites are insufficient to treat cyanide poisoning efficaciously, or that another mechanism of action is involved. Clues to this hypothesis may be found in the better efficacy of nitrites compared to non-nitrite methemoglobin forming agents such as 4-dimethylaminophenol and the efficacy of nitrites in the presence of methylene blue (which prevents methemoglobin formation) in experimental cyanide poisoning.4,21,23,24 Vasodilatation was proposed as an alternative mechanism of action of nitrites.4 Amyl nitrite counteracts the early negative circulatory effect of cyanide by raising cerebral and myocardial perfusion.25,26 Inhalation of amyl nitrite by cyanide-poisoned dogs resulted in a rise in arterial blood pressure.5 The hemodynamic effect of improved perfusion by nitrites is mediated by activation of cGMP which produces the potent vasodilator nitric oxide.27 This conversion occurs mainly in the presence of low oxygen concentrations.28,29 A similar antidotal effect was observed in animal studies with other vasodilators, such as phenoxybenzamine, isosorbide dinitrate, and chlorpromazine.5,30–33
Clinical Toxicology vol. 48 no. 6 2010
Amyl nitrite in cyanide mass casualty incidents
479
Efficacy: experimental studies
Dose regimen
There are few animal studies on amyl nitrite, most of them performed on dogs and mice.5,9,16,34–36 A summary of the animal studies is shown in Table 1. An additional animal study was published as a conference abstract but not as a peer reviewed article.37 The data presented in the abstract of this study is scanty and incomplete, and therefore we did not include it in Table 1. The quoted studies have several limitations. Most of them provide scanty information, and were done decades ago with different designs and set-ups unrelated to clinical scenarios. The number of animals was generally small. Different doses of amyl nitrite and different cyanide compounds were used. End-points were usually limited to survival with no laboratory determination of cyanide or amyl nitrite levels. No dose–response or concentration–response correlation is available. Moreover, the concentration of methemoglobin in the studied animals was higher than that reported in humans, as discussed earlier. Limited data were provided on respiratory and hemodynamic parameters. Finally, in most studies statistical analysis was not performed. In conclusion, the results of these studies cannot be considered good scientific evidence for the efficacy of amyl nitrite in cyanide poisoning. Caution should be exercised when extrapolating these studies to human exposures, especially to mass casualty incidents.
Amyl nitrite is available in pure form in fragile glass pearls containing 0.3 mL. It is recommended to break the pearl into a gauze or handkerchief and apply it to the nose and mouth intermittently for 30–60 s every 30–60 s, for ∼ 5 min.2 This practice has several limitations: 1. Uncontrolled application with potential inadequate dosing; that is too frequent or prolonged administration causing adverse effects (e.g. headache, hypotension, syncope), or under-dosing resulting in inefficacy. This is of special concern in inadequately trained health care providers and during stressful events such as mass incidents. 2. Potential risk to caregivers who are inadvertently exposed to this highly volatile substance.12 3. Amyl nitrite is extremely flammable.12 Its uncontrolled use in a pre-hospital setting exposed to fire hazards is potentially dangerous. 4. Out-of-hospital storage. Amyl nitrite pearls should be kept refrigerated and protected from light to prevent decomposition and loss of potency.12,46 5. Abuse concern.47–49 Amyl nitrite and other alkyl nitrites have been known for decades as aphrodisiacs, mainly used by homosexual men for recreational purposes; known street names include ‘poppers’ and ‘snappers’.50,51 Wide availability of amyl nitrite in emergency kits can be a source of inappropriate use and abuse.
Efficacy: human studies
Adverse reactions and vulnerable populations
A summary of the human reports on the use of amyl nitrite in cyanide poisoning is shown in Table 2. Eleven publications on the use of amyl nitrite in cyanide poisoning were found; three small series and eight case reports on 31 patients altogether.13,18,35,38–45 The information provided therein is often limited and lacks substantiated evidence for the efficacy of amyl nitrite. In some cases these were reports by first responders, usually co-workers with unclear professional medical training and experience. The majority of the victims were young healthy subjects with accidental industrial cyanide exposure. The time to administration of amyl nitrite varied in different reports and the dose was variable, uncontrolled, and not always detailed. Only in three cases was amyl nitrite given alone. In all other cases amyl nitrite was given along with or after supportive treatment and other antidotes. Serum amyl nitrite was not determined in any of the reports. Methemoglobin was not measured routinely. Blood pressure was not recorded consistently or immediately prior to and after amyl nitrite administration. Adverse drug reactions were not routinely recorded. Based on these reports, it is difficult to draw firm conclusions on the role of amyl nitrite in acute cyanide poisoning and particularly in mass incidents.
Exposure to amyl nitrite can result in serious adverse effects. The main reactions are presented in Table 3. Headache, dizziness, nausea, and vomiting were frequent complaints in case reports of amyl nitrite administration to cyanide poisoned patients.35,44 Hypotension is commonly observed after exposure to amyl nitrite in volunteers and cardiac patients.52,53 Moody et al.52 reported hypotension in 27% of patients undergoing elective cardiac catheterization who received amyl nitrite as a vasodilator. Marked hypotension was also reported shortly after amyl nitrite administration in cyanide poisoning.44 In this case the patient was exposed repeatedly to amyl nitrite and each time developed hypotension necessitating the discontinuation of the administration. Vasodilatation and pooling of blood develop rapidly after amyl nitrite administration and can result in shock and syncope, especially when abused.54,55 Transient reflex tachycardia is often seen due to amyl nitrite-induced vasodilatation.6,18 Extensive exposure to amyl nitrite (e.g. abuse, repeated or prolonged administration) can result in symptomatic methemoglobinemia.39,47,48,56,57 Amyl nitrite administration to patients after exposure to fire smoke can be dangerous and is considered contraindicated by some clinicians. Fire smoke contains nitrogen oxides and carbon monoxide and their inhalation can result in carboxyhemoglobinemia and
480
Dogs: 6 control, 0 treated
Animals
Inh. CNCl, 3300–6300 and 6,900–11,800 mg.min/m3, for 1–2.5 min
Inh. HCN 1,005–2,100 mg.min/m3.
SC NaCN, 6–30 mg/kg
Cyanide exposure
IV NaCN, 2.5 mg/kg
IV KCN, 3–4 mg/kg
IV NaCN, 2.5 mg/kg
Dogs: 20 control, 30 treated
Dogs: 11 treated, no control
Dogs: 10 control, 15 treated
Vick and Froehlich5
Klimmek and Krettek36
CNCl high concentration: 11/11 control and 8/11 treated died 161/198 control, 93/250 treated died
CNCl low concentration: 12/13 control and 3/13 treated died.
4/6 control died within 103–300 min, 4/10 treated died within 107–720 min HCN: 6/9 control and 4/9 treated died.
Outcome
Survival only Inh. 30 s post-poisoning in None a chamber, 9–13 mg/L, 2.5–4.5 min None LD50 of NaCN Inh. immediately postLD50 of control: 5.36 ± 0.28 poisoning, 0.3 ml every mg/kg, LD50 of treated: 24.5 3–5 min up to 20 min ± 1.2 mg/kg IV 0.6 ml immediately Mech. ventilation, Survival, BP, HR, 20/20 control, 10/10 IV AN, and 0/20 inh. AN died. post-poisoning (n = 10) no O2 (only for RR, ECG, Hb, or Inh. 0.3 ml within 3 AN treated MtHb min of poisoning (n = 20) dogs) MtHb rose from 4.8% to a maximum of 31% after AN BP, HR, RR, ECG normalized in surviving dogs IV 0.3–1.2 ml immediately Mech. ventilation, IV AN: BP, HR, IV treated dogs followed up to post-poisoning (n = 4), no O2 RR, Hb, lactate. 60 min: ↓BP, ↑HR, ↑lactate, Inh. 0.3 ml within 3 min survival not reported. of poisoning (n = 7) Inh. AN: survival Inh. AN: 6/7 died (4 within 20 min) IV 0.6 ml immediately Survival, BP, HR, 10/10 control and 0/15 post-poisoning, Inh. 0.3 Mech. ventilation, RR treated died. BP, HR, RR no O2 ml within 3 min of normalized in survivors poisoning
Mech. ventilation, Survival only no O2
Inh. 45 s after poisoning, 0.3 ml for 5 min
Survival only
Monitoring parameters
None
Other treatments
Inh. immediately post-CN, 0.3 ml every 3–5 min
Amyl nitrite administration
SC = subcutaneous, IV = intravenous, Inh. = inhalation, AN = amyl nitrite, CN = cyanide, NaCN = sodium cyanide, KCN = potassium cyanide, HCN = hydrogen cyanide, CNCl = cyanogen chloride, BP = blood pressure, HR = heart rate, RR = respiratory rate, ECG = electrocardiogram, Hb = hemoglobin, MtHb = methemoglobin, LD50 = median lethal dose, min = minute, mech. = mechanical.
Vick and Froehlich9
SC NaCN, 5–24 mg/kg
Dogs: 16 control, 13 treated
Chen and Rose35
Jandorf Mice: 198 control, Inh. CNCl, 2.5–3 mg/L and Brodansky16 250 treated for 2–2.5 min
Jandorf Dogs: 33 pairs of and Brodansky16 control and treated
Chen et al.34
Study
Table 1. Efficacy: experimental studies
481
Cyanide exposure
Co-treatments
Coma, seizures, hypoventilation, acidosis
Coma
Unconscious, hypotension, hypoventilation, seizures
Mech. ventilation, O2, SN, ST, Bic., naloxone, glucose, diazepam, AC, GL Mech. ventilation, O2, SN, ST Mech. ventilation, O2, ST, AC, GL
30 min after admission, 3 pearls, 1 min each Within 5 min, Mech. ventilation (6), unclear dose O2 (7); treated only on scene
Not specified
4 h postingestion, unclear dose
Inh. HCN; industrial Apnea, cardiac arrest accident
Not specified
furosemide
Mech. ventilation, O2, SN, ST, mannitol,
Unconscious (4), semi-comatose (3), depressed breathing (2) 1 female, 19 years Inh. HCN; industrial Unconscious, After 4 h, 1 pearl Mech. ventilation, accident hypotension, seizures O2, Bic., ST
1 male, 37 years
Outcome
No response to AN, extubated after 15 h, neurologic impairment at 1 year Improved within several hours
All recovered within 3–6 h and returned to work
Improved within several hours Improved after 90 min
Improved after 15 h
Not reported
Not reported
Headache (3)
Hypotension (systolic BP 80 mmHg)
Not reported
Not reported
Not reported
Not reported
Nausea (2), vomiting (2), headache (2) MtHb 40%
Not reported
Adverse reactions
HCN = hydrogen cyanide, CNCl = cyanogen chloride, CaCN = calcium cyanide, KCN = potassium cyanide, BP = blood pressure, MetHb = methemoglobin, inh. = inhalation, mech. = mechanical, O2 = oxygen, AN = amyl nitrite, SN = sodium nitrite, ST = sodium thiosulfate, Bic. = sodium bicarbonate, AC = activated charcoal, GL = gastric lavage, min = minute, h = hour.
Jian et al.45
Lam and Lau13
Wurzburg44
Nakatani et al.43
7 males, unknown Inh. HCN; age industrial accident
1 female, 24 years Ingested KCN; suicide 1 male, 31 years Ingested KCN; suicide
Johnson et al.18
Hall et al.42
1 male, unknown Ingested KCN; age accident 1 female, 32 years Ingested Laetrile; suicide 1 male, 4 years Ingested Laetrile; accident
Not specified
AN treatment
9 survived: 2 improved Mech. ventilation (9), after AN, 7 improved O2 (1), SN (3), ST (3) later All recovered, only 1 Within 5–60 min Mech. ventilation (4), improved after AN up to 20 pearls O2 (1), SN (1), epinephrine (1) 45 min postMech. ventilation, Improved after 75 min ingestion, O2, GL 1 pearl/min, 5 times Abdominal pain, Not specified SN, ST, Pro-Banthine, Improved, time not cyanosis, hypotension chlorpromazine specified Hypoxemia Not specified GL, SN, ST Improved within 24 h
Clinical manifestations
Inh. HCN; industrial Unconscious (9), accident semi-comatose (2), apnea (8) 5 males, 1 female, Inh. HCN (5) Unconscious (3), 22–67 years and CNCl (1); semi-comatose (2), accident apnea (2) 1 male, 18 years 2 g CaCN; suicide Unconscious, seizures, tachypnea, tachycaredia
11 males, 26–48 years
Patients
Thomas and Brooks40 Moss et al.41
Mascarenhas et al.39
Chen and Rose35
Wolfsie38
Study
Table 2. Efficacy: human studies
Clinical Toxicology vol. 48 no. 6 2010
482
O. Lavon and Y. Bentur
Table 3. Adverse reactions of amyl nitrite reported in humans* System affected
Adverse effect
References
Cardiovascular
Hypotension Syncope, shock Reflex tachycardia Headache, dizziness, syncope Nausea, vomiting Methemoglobinemia** Hemolytic anemia Elevated intra-ocular pressure Irritation*** Irritation, dermatitis*** T-cell impairment***
52–55 54,55 6,18 35,44 35 39,47,48,56,57 60–62,67 47,48,51,64 51 51 69
Neurological Gastrointestinal Hematological Ocular Dermatological Immunological
* Data derived from case series and case reports; ** Abuse; *** Repeated use.
methemoglobinemia which impairs oxygen delivery capacity and to some extent also tissue oxygen utilization. Thus, administration of amyl nitrite to smoke inhalation victims can result in additional impairment of oxygen delivery and worsen their outcome.58 Amyl nitrite can cause hemolytic anemia, particularly in G6PD-deficient patients.59–62 Cardiovascular and pulmonary patients are especially at risk as they have limited reserve to handle amyl nitrite-induced hypotension and methemoglobinemia.47 There is concern regarding potential rise in intracranial pressure secondary to amyl nitrite exposure. This was observed in rats treated with amyl nitrite and is probably due to vasodilatation and pooling of blood in the brain.63 Amyl nitrite was reported to elevate intraocular pressure and aggravate glaucoma,47,48,51,64 the specific mechanism is unclear. There are reports of amyl nitrite causing irritation to the eyes and skin that resulted in ocular and dermal inflammation, especially after repeated or prolonged exposure.51 The safety of amyl nitrite was not substantiated in extreme age groups. Administration to children is seldom reported and there are no pediatric dosage recommendations. Elderly patients are at higher risk of developing adverse reactions from amyl nitrite because they are more likely to suffer from cardiovascular and pulmonary diseases. Reports on the use of amyl nitrite in this age group are limited. Chen and Rose35 reported the use of amyl nitrite in three cyanide-poisoned patients aged 61–67. No information regarding co-morbidities was provided. One of the patients suffered nausea and vomiting after exposure to amyl nitrite. The use of amyl nitrite during pregnancy is controversial. Data on potential teratogenicity in humans and animals is lacking. Amyl nitrite was used as a muscle relaxant during Caesarean section.65 Yet, fetal toxicity secondary to maternal exposure to amyl nitrite is possible. The fetus is sensitive to hypotension, and the risk of methemoglobinemia in fetal blood is high.66 There is some evidence pointing to mutagenic properties of amyl nitrite.67,68 Male homosexuals who developed Kaposi’s
sarcoma were also heavy nitrite users. Nitrites were found to be mutagenic in bacteria.67 Amyl nitrite was positive in the mouse lymphoma TK+/− and Salmonella typhimurium mutagenicity assays.68 Impairment of the function of T-lymphocytes was demonstrated in humans after repeated exposure to amyl nitrite.69 The use of amyl nitrite and other nitrites and nitrates together with anti-phosphodiesterases (e.g. sildenafil) is contraindicated due to the risk of severe hypotension secondary to an augmented effect of nitric oxide.70
Role in mass casualty cyanide poisoning Mass casualty cyanide poisoning is a potential threat for a chemical terrorist attack or war, industrial accident, and fires.71,72 Dispersion of gaseous cyanide compound in a crowded space can cause a large number of casualties. The simplicity of administration of amyl nitrite enables first responders to give it by inhalation to patients on scene and allegedly improve outcome. Some clinicians advocate the wide use of amyl nitrite in such events.2,73 This recommendation is based on the unproven and debatable assumption that amyl nitrite is an effective and relatively safe antidote in mass casualty cyanide poisoning. The expected distribution of severity is related to the concentration of inhaled cyanide which is related to the distance from the dispersing source and to meteorological conditions. The minimal lethal concentration of HCN is estimated to be 180–200 ppm.74,75 At this concentration even a brief inhalation usually results in unpreventable rapid death and the exposed patients do not benefit from antidotal treatment. Inhalation of 50 ppm HCN or less for 30–60 min is not expected to result in serious acute clinical manifestations.76 Victims exposed to these low concentrations can be safely transported to a medical facility while receiving supportive treatment, and there is no need for any urgent administration of antidotes on scene. Only patients exposed to the range of concentrations lower than the immediately lethal levels and higher than those not associated with immediate serious toxicity need intensive treatment and antidotes on scene. Therefore, administration of amyl nitrite to all exposed people on scene is not mandatory and may result in unnecessary adverse effects. Moreover, the suggested protocol of amyl nitrite administration neutralizes the caregiver for at least a few minutes, diverts his efforts and attention from essential supportive treatment and rapid evacuation, and exposes him/her to amyl nitrite and its potential adverse effects. This can be a major obstacle in the management of any mass casualty incident. The victim population in a civilian mass casualty incident usually comprises people of widely different ages and health conditions. The use of a medication with unproven efficacy and potential serious adverse reactions is of major concern. Amyl nitrite was reported to be associated with hypotension in a cyanide poisoned patient. It can also cause syncope and
Clinical Toxicology vol. 48 no. 6 2010
Amyl nitrite in cyanide mass casualty incidents shock, especially in patients with respiratory and cardiovascular disorders. Its use in children, elders, and pregnant women is controversial. Its safety for caregivers is not substantiated. In our opinion, the risk from possible wide injudicious use of amyl nitrite in the pre-hospital management of mass casualty cyanide poisoning outweighs its unproven benefit. Other antidotes with a better safety and efficacy profile such as hydroxocobolamin may be considered, provided simple modes of administration become available.77,78
Conclusions The efficacy of amyl nitrite in human cyanide poisoning has not been proven. The mechanism of its antidotal properties is not entirely clear. The reported serious adverse reactions of amyl nitrite limit its use. The recommended protocol of amyl nitrite in cyanide poisoning is problematic. The safety of its administration to vulnerable populations and the potential neutralization of caregivers are of major concern. The effectiveness of cyanide antidotes administered on scene is limited to patients exposed to a relatively narrow range of cyanide concentrations. Casualties exposed to concentrations which are immediately lethal and those which are not associated with immediate serious toxicity do not require urgent prehospital antidotal therapy. Administration of amyl nitrite in mass casualty cyanide poisoning can result in unnecessary morbidity and may interfere with the proper management of the incident and the required supportive treatment and rapid evacuation. In our opinion these drawbacks make the use of amyl nitrite in pre-hospital mass casualty cyanide poisoning unwarranted.
Declaration of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this paper.
References 1. Vogel SN, Sultan CH, Jeneyck RP. Cyanide poisoning. J Toxicol Clin Toxicol 1981; 18:361–383. 2. Marrs TC. Antidotal treatment of acute cyanide poisoning. Adv Drug React Acute Poisoning Rev 1988; 4:179–206. 3. Pedigo LG. Antagonism between amyl nitrite and prussic acid. Tr M Soc Virginia 1888; 19:124–131. 4. Way JL, Sylvester D, Morgan RL, Isom GE, Burrows CB. Recent perspectives on the toxicodynamic basis of cyanide antagonism. Fundam App Toxicol 1984; 4:S231–S239. 5. Vick JA, Froehlich HL. Studies of cyanide poisoning. Arch Int Pharmacodyn 1985; 273:314–322. 6. Haley TJ. Review of the physiological effects of amyl, butyl and isobutyl nitrites. Clin Toxicol 1980;16:317–329. 7. Brivet F, Delraissy JF, Duche M, Bertrand P, Dortmont J. Acute cyanide poisoning: recovery with non-specific supportive therapy. Intensive Care Med 1983; 9:33–35.
483 8. Bismuth C, Cantineau JP, Pontal P, Baud FJ, Garnier R. Cyanide intoxication: primacy of symptomatic treatment. Presse Med 1984; 13:2493–2497. 9. Vick JA, Froehlich HL. Treatment of cyanide poisoning. Mil Med 1991; 156:330–339. 10. Cohen S. The volatile nitrites. JAMA 1979; 241:2077–2078. 11. Moshage H, Kok B, Huizenga JR, Jansen PLM. Nitrite and nitrate determination in plasma. Clin Chem 1991; 41:892–896. 12. Seifert SA. Nitrates and nitrites. In: Dart RC, ed. Medical Toxicology. 3rd ed. Philadelphia: Lippincot Williams & Wilkins; 2004:1177–1178. 13. Lam KK, Lau FL. An incident of hydrogen cyanide poisoning. Am J Emerg Med 2000; 18:172–174. 14. Hall AH, Doutre WH, Ludden T, Kulig KW, Rumack BH. Nitrite/thiosulfate treated acute cyanide poisoning: estimated kinetics after antidotes. J Toxicol Clin Toxicol 1987; 25:121–133. 15. Bastian G, Mercker H. The efficacy of amyl nitrite inhalation in the treatment of cyanide poisoning. Naunyn-Schmiedebergs Arch Exp Pathol 1959; 237:285–295. 16. Jandorf BJ, Brodansky O. Therapeutic and prophylactic effect of methemoglobinemia in inhalation poisoning by hydrogen cyanide and cyanogen chloride. J Ind Hyg Toxicol 1946; 28:125–132. 17. Mathes K, Gross F. The determination of methemoglobin and cyanohemoglobin in circulatory blood. J Naunyn-Schmiedebergs Arch Exp Pathol Pharmakol 1939; 191:701. 18. Johnson WS, Hall AH, Rumack BH. Cyanide poisoning successfully treated without “therapeutic methemoglobin levels.” Am J Emerg Med 1989; 7:437–440. 19. DiNapoli J, Hall AH, Drake R, Rumack BH. Cyanide and arsenic poisoning by intravenous injection. Ann Emerg Med 1989; 18:308–311. 20. Johnson RP, Mellors JW. Arteriolization of venous blood gases: a clue to the diagnosis of cyanide poisoning. J Emerg Med 1988; 6:401–404. 21. Holstege CP, Isom EG, Kirk AM. Cyanide and hydrogen sulfide. In: Flomenbaum NE, Goldfrank LR, Hoffman RS, Howland MA, Lewin NA, Nelson LS, eds. Goldfrank’s toxicologic emergencies. 8th ed. New York: McGraw-Hill; 2006:1717–1725. 22. Erdman AR. Cyanide antidote package. In: Dart RC, ed. Medical Toxicology. 3rd ed. Philadelphia: Lippincott Williams & Wilkins; 2004:173. 23. Klimmek R, Roddewig C, Fladerer H, Krettek C, Weger N. Effects of 4-dimethylaminophenol, Co2EDTA, or NaNO2 on cerebral blood flow and sinus blood homeostasis of dogs in connection with acute cyanide poisoning. Toxicology 1983; 26:143–154. 24. Marrs TC, Bright JE, Woodman AC. Species differences in methaemoglobin production after addition of 4-dimethylaminophenol, a cyanide antidote, to blood in vitro: a comparative study. Comp Biochem Physiol B 1987; 86:141–148. 25. Mathew RJ, Wilson WH, Tant SR. Regional cerebral blood flow changes associated with amyl nitrite inhalation. Br J Addict 1989; 84:293–299. 26. Murad F. Drugs used for treatment of angina: organic nitrates, calcium channel blockers and beta-adrenergic antagonists. In: Gilman AG, Rall TW, Nies AS, Taylor P, eds. Goodman and Gilman’s the pharmacological basis of therapeutics. 8th ed. New York: Pergam Press; 1990:764–773. 27. Fiscus RR. Molecular mechanisms of endothelium-mediated vasodilation. Semin Thromb Hemost 1988; 14(Suppl):12–22. 28. Cosby K, Partov KS, Crawford JH. Nitrite reduction to nitric oxide by deoxyhemoglobin vasodilates the human circulation. Nat Med 2003; 9:1498–1505. 29. Modin A, Bjorne H, Herulf M. Nitrite-derived nitric oxide: a possible mediator of “acidic-metabolic” vasodilatation. Acta Physiol Scand 2001; 171:9–16. 30. Sun P, Borowitz JL, Kanthasamy AG, Kane MD, Gunasekar PG, Isom GE. Antagonism of cyanide toxicity by isosorbide dinitrite: possible role of nitric oxide. Toxicology 1995; 104:105–111. 31. Dodson RA, Burrows GE, Isom GE, Way JL. Mechanism of chlorpromazine antagonism of cyanide intoxication. Proc West Pharmacol Soc 1975; 18:348–350. 32. Way JL, Burrows G. Cyanide intoxication: protection with chlorpromazine. Toxicol Appl Pharmacol 1976; 36:93–97.
Clinical Toxicology vol. 48 no. 6 2010
484 33. Burrows GE, Way JL. Antagonism of cyanide toxicity by phenoxybenzamine. Fed Proc 1975; 35:250–251. 34. Chen KK, Rose CL, Clowes GHA. Amyl nitrite and cyanide poisoning. JAMA 1933; 100:1920–1922. 35. Chen KK, Rose CL. Nitrite and thiosulfate therapy in cyanide poisoning. JAMA 1952; 146:113–119. 36. Klimmek R, Krettek C. Effects of amyl nitrite on circulation, respiration and blood homeostasis in cyanide poisoning. Arch Toxicol 1988; 62:161–166. 37. Groff W, Hassett CC, Johnson W, Vick J. Recent studies on the therapy of cyanide poisoning. Fed Proc Fed Amer Soc Exp Biol 1973; 32:372. 38. Wolfsie JH. Treatment of cyanide poisoning in industry. Am Med Assoc Arch Ind Hyg 1951; 4:417–425. 39. Mascarenhas BR, Geller AC, Goodman AI. Cyanide poisoning: medical emergency. NY State J Med 1969; 69:1782–1784. 40. Thomas TA, Brooks JW. Accidental cyanide poisoning. Anaesthesia 1970; 25:110–114. 41. Moss M, Khalil N, Gray J. Deliberate self-poisoning with laetrile. CMAJ 1981; 125:1126–1128. 42. Hall AH, Linden CH, Kulig KW, Rumack BH. Cyanide poisoning from laetrile ingestion: role of nitrite therapy. Pediatrics 1986; 78:269–272. 43. Nakatani T, Kosugi Y, Mori A, Tajimi K, Kobayashi K. Changes in the parameters of oxygen metabolism in a clinical course recovering from potassium cyanide. Am J Emerg Med 1993; 11:213–217. 44. Wurzburg H. Treatment of cyanide poisoning in industrial setting. Vet Human Toxicol 1996; 38:44–47. 45. Jian X, Guo G, Ruan Y, Lin D, Zhao B. Severe keloids caused by hydrogen cyanide injury: a case report. Cutan Ocul Toxicol 2008; 27:97–101. 46. Guidotti T. Acute cyanide poisoning in prehospital care: new challenges, new tools for intervention. Prehosp Diast Med 2006; 21:s40–s48. 47. Maickel RP. The fate and toxicity of butyl nitrites. NIDA Res Monogr 1988; 83:15–27. 48. Haverkos HW, Dourherty J. Health hazards of nitrite inhalants. Am J Med 1988; 84:479–482. 49. Balster RL. Neural basis of inhalant abuse. Drugs Alcohol Depend 1998; 51:207–214. 50. Lowry TP. Neurophysiological aspects of amyl nitrite. J Psychodelic Drugs 1980; 12:73–74. 51. Fisher AA. “Poppers” or “snappers” dermatitis in homosexual men. Cutis 1984; 34:118–122. 52. Moody JM Jr, Baily SR, Rubal BJ. Subtle features of the hemodynamic response to amyl nitrite inhalation: new aspects of an old tool. Clin Cardiol 1993; 16:331–338. 53. Ma S, Long JP. Central noradrenergic activity and the cardiovascular effects of nitroglycerin and amyl nitrite. J Cardiovasc Pharmacol 1992; 20:826–836. 54. Wilkins RW, Haynes FM, Weiss S. The role of venous system circulatory collapse induced by sodium nitrite. J Clin Invest 1937; 16:85–91. 55. Turchen SG, Manoguerra AS, Whitney C. Severe cyanide poisoning from the ingestion of an acetonitrile-containing cosmetic: Am J Emerg Med 1991; 9:264–267.
O. Lavon and Y. Bentur 56. Berlin CM Jr. The treatment of cyanide poisoning in children. Pediatrics 1970; 46:793–796. 57. Machabert R, Testud F, Descotes J. Methemoglobinemia due to amyl nitrite inhalation: a case report. Hum Exp Toxicol 1994; 13:913–914. 58. Moore SJ, Norris JC, Walsh DA, Hume AS. Antidotal use of methemoglobin forming cyanide antagonism in concurrent carbon monoxide/ cyanide intoxication. J Pharmacol Exp Ther 1987; 242:70–73. 59. Brnades JC, Bufill JA, Pisciotta AV. Amyl nitrite induced hemolytic anemia. Am J Med 1989; 86:252–254. 60. Graves TD, Mitchell S. Acute haemolytic anaemia after inhalation of amyl nitrite. J R Soc Med 2003; 96(12):594–595. 61. Costello C, Pourgourides E, Youle M. Amyl nitrite induced acute haemolytic anaemia in HIV-positive man. Int J STD AIDS 2000; 11:334. 62. Romeril KR, Concannon AJ. Heinz body haemolytic anaemia after sniffing volatile nitrites. Med J Aust 1981; 1:302–303. 63. Nocross NC. Intracerebral blood flow. Arch Neurol Psychiatry 1938; 40:291. 64. Grant WH. Toxicology of the eye. 4th ed. Springfield, IL: Charles C Thomas; 1993. 65. Hendricks SK, Ross B, Colvard MA, Cahill D, Shy K, Benedetti TJ. Amyl nitrite: use as a smooth muscle relaxant in difficult preterm cesarean section. Am J Perinatol 1992; 9:289–292. 66. Tarburton JP, Metcalf WK. Kinetics of amyl nitrite-induced hemoglobin oxidation in cord and adult blood. Toxicology 1985; 36:15–21. 67. Dunkel VC, Rogers-Back AN, Lawlor TE, Harbell JW, Cameron TP. Mutagenicity of some alkyl nitrite used as reacreational drugs. Environ Med Mutagen 1989; 14:115–122. 68. Balimandawa M, de Meester C, Léonard A. The mutagenicity of nitrite in the Salmonella/microsome test system. Mutat Res 1994; 321:7–11. 69. Goedert JJ, Neuland CY, Wallen WC, Greene MH, Mann DL, Murray C, Strong DM, Fraumeni JF Jr, Blattner WA. Amyl nitrite may alter T-lymphocytes in homosexual men. Lancet 1982; 1:412–416. 70. James JS. Viagra warning re “poppers” and notice re protease inhibitors. AIDS Treat News 1998; 294:1. 71. Maniscalco PM. From smoke inhalation to chemical attacks: acute cyanide poisoning in the prehospital setting. Prehosp Diast Med 2006; 21:s38–s39. 72. Keim ME. Terrorism involving cyanide: the prospect of improving preparedness in the prehospital setting. Prehosp Diast Med 2006; 21:s56–s60. 73. Krivoy A, Finkelstein A, Rotman E, Layish I, Tashma Z, Hoffman A, Schein O, Yehezkelli Y, Dushnitsky T, Eisenkraft A. Cyanides—treatment beneath the shade of terror. Harefuah 2007; 146:228–234, 244. 74. Bacroft J. The toxicity of atmospheres containing hydrocyanic acid gas. J Hyg 1931; 31:1–34. 75. Gettler AO, Baine JO. The toxicology of cyanide. Am J Med Sci 1938; 195:182–198. 76. Bonsall JL. Survival without sequelae following exposure to 500 mg/m3 of hydrogen cyanide. Hum Toxicol 1984; 3:57–60. 77. Shepherd G, Velez LI. Role of hydroxocobalamin in acute cyanide poisoning. Ann Pharmacother 2008; 42:661–669. 78. DesLauriers CA, Burda AM, Wahl M. Hydroxocobalamin as a cyanide antidote. Am J Ther 2006; 13:161–165.
Clinical Toxicology (2010) 48, 485–496 Copyright © Informa UK, Ltd. ISSN: 1556-3650 print / 1556-9519 online DOI: 10.3109/15563650.2010.506876
REVIEW LCLT
Occupational toxicology of asbestos-related malignancies MARCELLO LOTTI1, LORENZO BERGAMO1, and BRUNO MURER2 Diagnostic issues in asbestos-related malignancies
1 2
Dipartimento di Medicina Ambientale e Sanità Pubblica, Università degli Studi di Padova, Padova, Italy Dipartimento di Anatomia Patologica, Azienda ULS 12 Veneziana, Mestre, Italy
Introduction. Asbestos is banned in most Western countries but related malignancies are still of clinical concern because of their long latencies. This review identifies and addresses some controversial occupational and clinical aspects of asbestos-related malignancies. Methods. Papers published in English from 1980 to 2009 were retrieved from PubMed. A total of 307 original articles were identified and 159 were included. Assessment of exposure. The retrospective assessment of exposure is usually performed by using questionnaires and job exposure matrices and by careful collection of medical history. In this way crucial information about manufacturing processes and specific jobs can be obtained. In addition, fibers and asbestos bodies are counted in lung tissue, broncho-alveolar lavage, and sputum, but different techniques and interlaboratory variability hamper the interpretation of reported measurements. Screening for malignancies. The effectiveness of low-dose chest CT screening in exposed workers is debatable. Several biomarkers have also been considered to screen individuals at risk for lung cancer and mesothelioma but reliable signatures are still missing. Attribution of lung cancer. Exposures correlating with lung cancer are high and in the same range where asbestosis occurs. However, the unresolved question is whether the presence of fibrosis is a requirement for the attribution of lung cancer to asbestos. The etiology of lung cancer is difficult to define in cases of low-level asbestos exposure and concurrent smoking habits. Mesothelioma. The diagnosis of malignant mesothelioma may also be difficult, because of procedures in sampling, fixation, and processing, and uses of immunohistochemical probes. Conclusions. Assessment of exposure is crucial and requires accurate medical and occupational histories. Quantitative analysis of asbestos body burden is better performed in digested lung tissues by counting asbestos bodies by light microscopy and/or uncoated fibers by transmission electron microscopy. The benefits of screenings for asbestos-related malignancies are equivocal. The attribution of lung cancer to asbestos exposure is difficult in a clinical setting because of the need to assess asbestos body burden and the fact that virtually all these patients are also tobacco smokers or former smokers. Given the premise that asbestosis is necessary to causally link lung cancer to asbestos, it follows that the assessment of both lung fibrosis and asbestos body burden is necessary. Keywords
Asbestos exposure; Lung cancer; Mesothelioma; Diagnosis; Screening
Introduction Educated guesses often need to be made on certain clinical problems of asbestos-related malignancies, inasmuch as unresolved controversial and difficult issues surround our understanding of these diseases. The use of asbestos has been banned or under strict control for more than a decade in most Western countries, but asbestos-related malignancies are still of clinical concern because of the long latency period between the initial exposure and the onset of diseases. In addition, the majority of produced asbestos is still used in Eastern Europe, Latin America, and Asia, and the World Health Organization has estimated that 125 million people worldwide are currently exposed to asbestos.1 This review is intended neither to be a comprehensive discussion of asbestos-related diseases because many sources Received 26 March 2010; accepted 4 July 2010. Address correspondence to Marcello Lotti, Dipartimento di Medicina Ambientale e Sanità Pubblica, Università degli Studi di Padova, Via Giustiniani, 2, 35128 Padova, Italy. E-mail:
[email protected]
are available,2–8 nor to offer a viewpoint on the medico-legal aspects that are found elsewhere.9,10 Rather, we wish to raise and make plain several issues that should concern those interested in the occupational and clinical aspects of asbestosrelated malignancies, including the assessment of asbestos exposures, the screening for malignancies in current and past exposed workers, the attribution of lung cancer to asbestos exposures, and the pathology of mesothelioma.
Methodology Review papers were retrieved from PubMed with the following limits: written in English, published between 1980 and 2009, and humans. Two or more keywords with the Boolean operator “AND” were used. A total of 4,106 results were obtained as follows: asbestos AND exposure: 560; lung cancer AND asbestos: 375; lung cancer AND CT screening: 776; lung cancer AND biomarker: 1,137; lung cancer AND asbestosis: 139; asbestos AND smoking: 148; mesothelioma AND histopathology: 491; mesothelioma AND cytology: 480.
Clinical Toxicology vol. 48 no. 6 2010
486 Reviews were first screened according to the appropriateness of the title and duplicates were removed, leaving 1,066 items. By reading the abstracts, 217 reviews were then selected. From quoted references, 307 original articles were identified and 152 were included in our article. By using the same keywords (but not the limit “review”), the 2009 updating identified a total of 2,226 papers that were screened for appropriateness and duplicates. An additional 7 papers were selected, making a total of 159 papers. The final inclusion criteria were based upon our judgment on the quality of the papers and their appropriateness for the specific topic to be discussed.
Assessment of exposure Given the long latency of asbestos-related malignancies – that is the time between the beginning of exposure and the onset of disease – the retrospective assessment of asbestos exposures represents a challenge for both epidemiology and clinical medicine. Questionnaires and job exposure matrices are used in epidemiology,11–18 whereas a careful medical history and an estimate of asbestos body burden represent the cornerstones in a clinical setting. Medical history Medical history requires information about the manufacturing process and the specific job where exposure to asbestos occurred. Asbestos is a family of natural hydrated silicates with a fibrous geometry (length:width ratio ≥3). Asbestos is divided into two groups differing in mineralogic properties and chemical composition: amphiboles and serpentines. The amphibole family consists of crocidolite [Na2Fe2(FeMg)3Si8O22(OH)2], amosite [(FeMg)7Si8O22(OH)2], anthophyllite [Mg7Si8O22(OH)2], and tremolite [Ca2Mg5Si8O22(OH)2]. Chrysotile [Mg3[Si2O5](OH)4] is the only serpentine and accounts for over 90% of all commercial asbestos. Amphibole fibers have a relatively small crosssectional diameter with a needle-like shape and tend to be readily transported to the periphery of the lung. These characteristics are thought to account for the higher pathogenicity of amphiboles, as compared with that of chrysotile. Given the different risks related to the varieties of occupational exposures, the intensity of exposure and the type of fiber should be understood. The length of exposure must also be recorded to assess the cumulative dose, given the nature of dose–response relationships between asbestos, asbestosis, and lung cancer. A panel of experts suggested in the Helsinki Report19 examples of exposures that may be associated with an increased risk of lung cancer: 1 year of exposure in the manufacture of asbestos products, asbestos spraying, insulation work, and demolition of old buildings; or from 5 to 10 years of moderate exposure such as in construction work and shipbuilding. Possible indirect and remote exposures to
M. Lotti et al. asbestos should also be investigated, particularly in the case of mesothelioma because even short exposures far back in time may be etiologically relevant.20 The minimal latency is considered to be 10 years for lung cancer and 15–20 years for mesothelioma. Smoking habits should be carefully assessed because the vast majority of asbestos workers are current or former smokers (Table 1) and interactions between asbestos and smoke have been shown to increase the incidence of lung cancer. Recording the end of exposure is important when counting asbestos bodies and/or fibers to assess body burden, given the variable clearance from the lung of different asbestos fibers.29–32 For example, the estimated half-life of crocidolite and amosite fibers in the lung is in the range of several years and that of crysotile is in the range of months. However, in lung cancer patients with past high exposures to crysotile the count of fibers was still fairly high when measured several years after the end of exposure.27 Asbestos bodies and fibers Direct evidence of exposure is given by counting fibers and asbestos bodies in lung tissue (surgical and postmortem), broncho-alveolar lavage, and sputum. However, several problems arise when assessing body burden by means of these methods. Transmission electron microscopy has a higher sensitivity than scanning electron microscopy when chrysotile fibers are to be detected and fibers of <5 μm length are to be counted.33 Interlaboratory variability is large34 because standardized analytical methods are not available.19 For example, when asbestos fibers were measured with transmission electron microscopy in lung tissues of the general population, the results were inconsistent (Table 1). Similar variability was found in the general population where the concentrations of different fibers in the lung were measured in the range of <100 – 2,500/g wet tissue with the scanning electron microscopy technique.35 This wide range of variability may be related to the examined population (urban vs. rural), to the age of subjects, and to the processing of lung samples. Asbestos bodies can be observed and counted with light microscopy. Median values of 2.9 asbestos bodies/g wet lung tissue (range = 0.2–22)35 and a range between <1 up to 70 asbestos bodies per milliliter of broncho-alveolar lavage36 were reported in the general population. Asbestos bodies can be observed occasionally in histological slides of the lung from unexposed individuals, and it is recommended that for the diagnosis of asbestosis two or more asbestos bodies per section area of 1 cm2 should be observed together with fibrosis.19 The Helsinki Report suggests criteria to identify individuals with probability of occupational exposure. These include over 0.1 million anfibole fibers (>5 μm)/g dry lung or over 1 million anfibole fibers (>1 μm)/g dry lung or over 1,000 asbestos bodies/g dry lung (100 asbestos bodies/g wet lung) or over 1 asbestos body per milliliter of broncho-alveolar
Clinical Toxicology vol. 48 no. 6 2010
Diagnostic issues in asbestos-related malignancies
487
Table 1. Transmission electron microscopy count of asbestos fibers in lung tissue from general populationa and lung cancer patients with asbestos exposureb (selected references)
Population Generala
Millions of uncoated fibers/ g dry weight of lung tissue (range)
Rurala
7c (0–50.4); 0.4 (0–1.7); 0.2 (0–1.5) 2.8c (0–11.7); 0.14 (0–1.7); 0.009 (0–1.0); 0.02 (0–0.4) 0.09d (<0.04–0.98)
Asbestos towna Construction workersb
2.63c (0.1–14.6) 5.79c (0.3 – 49.9)
Various occupationsb
1,141.7c
Textile asbestos workersb Various occupationsb
61.1d 7.82c (3.91–14.82)
Various occupationsb
0.78c (0.1–4.1)
Generala
Type of fiber (%) from all cases
Number of cases
Chrysotile (92); crocidolite (4); amosite (4) Chrysotile (91,8); crocidolite (4,6); amosite (3); tremolite (0,6) Chrysotile (46,6); amphiboles (23,3); others (30,1) Chrysotile (100) Anthophyllite (40); crocidolite (51); chrysotile (6); amosite (3) Chrysotile (6,9); crocidolite (56,7); amosite (30,4); others (6) N/A Chrysotile (25); amphiboles (2); others (73) Chrysotile (100)
5621 5521 2222 2323 2224e 1425f 1026 827 2028g
c
Mean. Median. e Twenty-one cases were smokers. f Patients with both lung cancer and asbestosis. g Seventeen cases were smokers. N/A, Information not available. d
lavage.19 Nevertheless, counts in sputum and broncho-alveolar lavage should be cautiously considered when assessing lung burden of asbestos.28,37 Similarly, a quantitative assessment of asbestos bodies on histological sections is a matter of debate because variables such as staining, number of sections, and magnification are often not specified.38 Epidemiological studies show a strong correlation between size of exposure and incidence of both asbestosis and lung cancer.39 Therefore, the counts of asbestos bodies and fibers is expected to be in the same range. In patients with both asbestosis and lung cancer, asbestos bodies count was found in the range 150–343,000 asbestos bodies/g wet lung tissue and that of uncoated fibers (>5 μm) in the range 18,500–7,800,000/g wet lung when measured with scanning electron microscopy.38 In patients with asbestos-related lung cancer, the variability of fiber counts (Table 1) and of asbestos bodies is also high.40 The presence of asbestos bodies in histological sections showing fibrosis suggests asbestosis because in their absence the detection of tissue asbestos fibers typical of asbestosis is unlikely.41 In fact, fibrosis appearance and its severity correlate with pulmonary concentration of uncoated fibers, equal or greater than 5 μm in length, exceeding 1 million/g dry lung tissue.38 To add more complexity, when exposures were assessed by medical history, job exposure matrices, or questionnaires, estimates did not match with asbestos bodies and fiber counts because cumulative exposures consistent with higher risk for lung cancer have been associated with low asbestos bodies and fiber counts.42 Possible explanations include the type of
fibers, their clearance from the lung, and an underestimate of smoking as an etiological factor. The assessment of asbestos body burden for the diagnosis of mesothelioma is considered of less importance because asbestos is the only recognized risk factor for this neoplasia and cases have been reported even after remote, short, and small exposures to asbestos.20 Nevertheless, asbestos bodies and fiber counts have also been reported in mesothelioma cases38 and they might be useful in distinguishing sporadic mesotheliomas.
Screening for malignancies Mesothelioma is an aggressive cancer with poor prognosis when diagnosed either at the onset of symptoms or occasionally and irrespectively of treatment.43 Moreover, the level of asbestos exposure associated with the development of mesothelioma might be rather low and not precisely estimated.20 On the contrary, heavy occupational exposures are associated with the development of all types of lung cancer,19 and the detection of lung tumors at an early stage could potentially allow early treatment resulting in improved morbidity and mortality rates.
Imaging Whether chest low-dose CT screening has a major impact on survival of lung cancer patients or not is a matter of controversy,44,45
Clinical Toxicology vol. 48 no. 6 2010
488 because two major studies reached opposite conclusions. A case–control observational study on 31,567 asymptomatic individuals at risk for lung cancer concluded that annual spiral CT screening can detect curable lung cancer.46 Among patients who received a diagnosis of lung cancer, 85% had clinical stage I lung cancer with an estimated 10-year survival rate corresponding to 88%. However, a longitudinal study showed a discrepancy between a good lung cancer-specific 2-year survival in individuals with early-stage cancer (90%), comparable with that estimated with the previous study, but no decline in the number of advanced lung cancer (42 patients vs. 33.4 expected) and of deaths (38 deaths vs. 38.8 expected) in the group as a whole.47 The latter results suggest that CT screening does not intercept lung tumors that will later progress to clinical disease and death, otherwise a relevant proportion of early-stage cancers would have lowered mortality. Thus, the overall benefits of CT screening were challenged and the authors concluded that asymptomatic individuals should not be screened outside clinical research studies. Some authors have recommended waiting for the outcome of ongoing prospective randomized multi-center trials,45 whereas others hold a divergent opinion on the social responsibility of screening for lung cancer. On the one hand, there is the compelling reason that lung cancer is curable if diagnosed early,48 whereas on the other, there is caution because it is uncertain if screening improves mortality and outweighs risks and costs.49 CT screening in asbestos workers will suffer from the same biases of any similar program, including lead time, length time, and over-diagnosis biases.50 In addition, when a population is stimulated by advertisement to undergo CT scanning, as might occur in asbestos workers, the crossover biases might have a substantial prevalence.51 This occurs when subjects undergo CT scan outside the trial, changing in this way the disease-specific mortality of either controls or screened population. Nevertheless, given the strong association between asbestos exposure and adenocarcinoma,52 it is likely that <1 cm peripheral tumors will be easier to detect by CT scan than other tumors located centrally, such as small cell and squamous cell cancers. A few studies addressing the feasibility of chest CT screening of asbestos-exposed workers have produced conflicting results.53–57 Varying prevalence of lung cancer has been reported ranging from 0.457 to 4.28,54 but follow-up was of short duration hampering an assessment of the effect on mortality. In one of these studies,57 the incidence of lung cancer was equal to that of the general population of similar age. In addition, most studies give warnings about the large number of incidental findings detected by chest CT. Nevertheless, though the costs of screening are high,57 this may not be an issue if society believes it has a social responsibility to these workers. Therefore, waiting for the results of ongoing studies might deprive these workers of potential benefits that outweigh risks. However, there are no specific reasons related to asbestos risk that would change the above uncertainties concerning CT screening of lung cancer.
M. Lotti et al. Biomarkers of lung cancer Biomarkers are aimed at predicting individuals at risk of developing cancer and/or detecting the disease at an early stage, assuming that cancer development is preventable and/ or effectively treated. The landscape of lung tumor biomarkers is changing rapidly but effective screening tools are not yet available and the implementation of their use in patient care is at its infancy because larger data sets are required for validation.58,59 Promising candidates include tests that investigate host susceptibility, tissue injury, gene expression, and proteomic profiling in accessible tissue. Genomewide association studies investigated the links between various types of diseases, including cancer, and single-nucleotide polymorphisms.60 The susceptibility to lung cancer was associated with variations of 15q25 gene encoding subunits of nicotine receptors,61,62 with a reduced DNA repair capacity63 and with the intensity of smoking that was associated with a variant of chromosome15q24.64 Circulating tumor cells, tumor-derived cytokines, and angiogenic factors have also been proposed as early diagnostic makers of lung cancer.65,66 Tumor-specific genetic and epigenetic changes, such as circulating DNA, tumor-specific DNA alterations, and promoter methylation, can be detected in blood and sputum from lung cancer patients and they could prove to be useful for either early diagnosis or predicting survival.67–70 Serum proteomic pattern analysis is also an active field of research,71 though its utility requires validation.72 In conclusion, although many of these preliminary studies are helpful in reducing the number of candidate biomarkers, only retrospective cohort studies on stored specimens will be useful in evaluating the performance of a given biomarker.73 Biomarkers have long been considered for the secondary prevention of asbestos-related malignancies but substantial progress has not been made. Asbestos-related molecular signatures such as specific changes in certain chromosome regions have been detected in both lung tumors and in vitro74–76 but reliable biomarkers for screening or early detection of lung tumors associated with asbestos are still missing. Biomarkers of mesothelioma Several diagnostic markers of mesothelioma have been suggested, but no single marker is available for biomonitoring use. Thus, gains and losses of several chromosome regions have been detected77–79 and, like many common tumors, malignant mesothelioma shows changes of tumor suppressor genes and oncogenes that affect distinct cell-signaling pathways, including p16INK4a/p14ARF.80,81 However, the mutation/deletion of neurofibromatosis type 2 gene,82,83 the target gene of 22q12 loss, suggests the need of a unique dysregulation for malignant mesothelioma development.84 Blood concentrations of osteopontin85 and of mesothelin family proteins86 allowed mesothelioma to be identified in exposed workers but it is unknown as to whether early
Clinical Toxicology vol. 48 no. 6 2010
Diagnostic issues in asbestos-related malignancies detection would change the outcome.87 In addition, high concentrations of mesothelin in blood and pleural effusion are associated with mesothelioma, with a specificity of 98% and a sensitivity of 67%;88 glucose transporter (GLUT-1) has also been considered.89
The attribution of lung cancer Lung cancer is a leading cause of cancer death worldwide and comprises two major forms: non-small cell and small cell cancers. The former accounts for about 85% of cases and includes adenocarcinomas, squamous cell carcinomas, and large-cell carcinomas. Other tumor types are rare. For unknown reasons, adenocarcinoma has replaced squamous cell carcinoma over the past 25 years as the most frequent histological subtype.90,91 Cigarette smoking is associated with 90% lung cancers of all histological types,92 developing in approximately 10–15% of active smokers.91,93 Causes of lung cancer in never smokers have yet to be identified and the disease is thought to be a distinct entity based upon differences in epidemiological, clinical, and molecular features.94,95 Occupational exposure Epidemiological studies indicate that asbestos exposure is associated with an increased risk of lung cancer,96–98 accounting for an estimated 2–5% of new lung cancers.99–101 An increased proportion of lower-lobe tumor localization102 and of adenocarcinomas50 has been reported among workers exposed to asbestos, but these associations have not been confirmed in other studies.103 The risk of developing lung cancer is linearly related to cumulative asbestos exposure, with an estimated increase of 1% for each fiber/mL-year of exposure.104 However, the strength of the association and the slope of the cumulative dose–response relationships vary considerably across studies and occupations.3 The risk appears to be smaller in miners105 and friction product manufacturers,106 intermediate in asbestos cement107 and product manufacturers,108 and highest in textile workers.109,110 The slope is steeper in asbestos cement manufacturers,107 much less in friction product manufacturers,106 and intermediate in the textile industry.109,110 The Helsinki Report estimates a cumulative exposure of 25 fibers/mL-years to increase the risk of lung cancer by twofold.19 However, a subsequent reevaluation concluded that probably the exposures associated with increased risk of lung cancer are higher.111 A recent risk analysis of 14 studies identified a cumulative “no effect” exposure to crysotile for lung cancer in the range of 25–1,000 fibers/mL-years.112 However, these dose estimates have been challenged for accuracy, mainly because of crude surrogates of exposure.113–115 Environmental and lower-level occupational exposures are not associated with an increased risk of lung cancer.39,98,116
489 Asbestosis and lung cancer A major controversy on asbestos-related lung cancer concerns its relationship with asbestosis. At issue is whether cancer arises from exposure levels required to cause asbestosis but independently from fibrosis or is there a mechanism shared with asbestosis pathophysiology that leads to cancer. Several authors hold the notion of an increased risk of lung cancer in asbestos-exposed workers in the absence of radiographic evidence of asbestosis117,118 but others conclude that the risk only increases if radiographic asbestosis is present.119,120 A high correlation between asbestosis and lung cancer rates was observed in 38 cohorts of workers,39 and the risks of developing asbestosis and lung cancer are in the same range of cumulative dose.121 However, no data are available to ascertain whether or not pulmonary fibrosis is an independent risk factor for lung cancer, although progression of asbestosis was shown to predict the development of lung cancer.122 A review of nine epidemiological studies supporting either proposition concluded that the question as to whether or not cancer arises in the presence of pulmonary fibrosis is unanswerable epidemiologically.6 Nevertheless, asbestosis is a good indication that exposure of the patient was high enough to put him at risk of lung cancer. Diagnostic criteria of nonmalignant asbestos-related diseases have been proposed123 and include the following: the evidence of structural pathology consistent with asbestosis as documented by imaging or histology, the evidence of exposure as documented by occupational and environmental history, markers of exposure (usually pleural plaques), recovery of asbestos bodies/fibers, and the exclusion of alternative plausible causes. However, several potential pitfalls should be considered in the interpretation of findings including uses, values, and limitations of plain chest radiographs, CT, and HRCT in the diagnosis of asbestosis, rounded atelectasis, and asbestos-related pleural diseases.123 Interstitial fibrosis and pleural thickening is easily detected with conventional CT and HRCT that are always necessary when assessing lung cancer. However, profusion of small irregular opacities associated with smoking might confound the radiological diagnosis of minimal asbestosis.123,124 The suggestion has been made that in the presence of minimal radiographic findings, a diagnosis of asbestosis could be made on clinical grounds if rales and a reduced diffusing capacity of the lung for carbon monoxide are also present, though these criteria are debatable.125 In addition, even a normal HRCT cannot completely exclude asbestosis, as shown when HRCT findings were compared with histopathology.126 Moreover, thin-section CT features may not distinguish other types of pulmonary fibrosis such as idiopathic pulmonary fibroses.127 The severity of interstitial fibrosis detected with chest radiograph correlates with asbestos cumulative exposure,128 but the presence/size of pleural plaques does not,129 although their presence was suggested as an additional criterion of
Clinical Toxicology vol. 48 no. 6 2010
490 attribution for lung cancer.19 Pleural plaques have been associated with greater risk of lung cancer123,130 but this conclusion should be challenged because plaques manifest themselves at asbestos exposures much smaller than those associated with increased risk of cancer.131 Histopathology is characterized by diffuse peribronchiolar fibrosis and presence of asbestos bodies, with or without alveolar septal fibrosis, and in more severe cases honeycomb changes may also be present.37 In conclusion, when both high asbestos body burden and fibrosis are present, then attribution of lung cancer is clearcut but it is also acceptable in the absence of fibrosis.19 Detection of pleural plaques indicates past exposure and it is not a substitute for estimating high asbestos body burden that must coexist with lung tumors. Smoking and asbestos interaction Epidemiological evidence indicates that smoking and asbestos exposures taking place together have more than additive and less than multiplicative effects on lung cancer incidence.132 However, epidemiology has little to say about the nature of the combined carcinogenetic effects of asbestos and smoking. Several hypotheses have been made,133 but one should recognize that no conclusions about the mechanism of either asbestos8 or smoking tumorigenicity have been reached. Hypotheses consider the mutual influences of asbestos and smoke in the delivery to epithelial cells of smoke carcinogens and fibers, and there is some evidence in humans that smoking increases the penetration of fibers into the bronchial mucosa where the effect is greater for chrysotile than for the anfiboles.134 Therefore, given the evidence of this interaction – that is drawn from epidemiology – extrapolations can hardly be made at levels of asbestos exposure that do not increase the risk of cancer per se.
Pathology of mesothelioma Histopathology Malignant mesothelioma is divided into three basic histological types – epithelial, sarcomatous, and biphasic.135 However, a myriad of histological patterns are observed within these main types that could be relevant in the differential diagnosis with other neoplasms.136 Epithelial mesothelioma is the most common and consists of epithelioid mesothelial cells arranged in tubulopapillary, microglandular, or sheet-like patterns. Less common variants include deciduoid, signet ring, adenoid cystic, small cell, and clear cell. Sarcomatous mesothelioma is characterized by a spindle cell proliferation resembling fibrosarcoma or malignant fibrous histiocytoma. Sarcomatous mesothelioma rarely contains areas of osteosarcomatous differentiation. The lymphohistiocytoid and desmoplastic variants of sarcomatous mesothelioma should not be confused with inflammatory processes and the lymphohistiocytoid
M. Lotti et al. variant must be differentiated from lymphomas. Biphasic mesothelioma is a mixture of epithelial and sarcomatous patterns in which each histological type must make up for at least 10% of the neoplasm. Immunohistochemical stains support the diagnosis of mesothelioma.137 The International Mesothelioma Panel recommends the use of at least two positive – calretinin, keratin 5/6, D2-40, mesothelin, and Wilms tumor 1 protein (WT-1) – and two negative markers (carcinoembryonic antigen CEA, MOC-31, B72.3, BerEP4, and Claudin-4) as an initial panel for the epithelial and biphasic types138 making distinction between epithelioid mesothelioma and carcinoma possible. In the diagnosis of the sarcomatoid type and its variants, the stains for keratins are useful, whereas other markers such as calretinin and WT-1 are expressed in a smaller percentage of cases.139 However, about 10–20% of sarcomatoid mesotheliomas do not stain for keratins and additional panel of antibodies should be used depending on differential diagnoses together with clinical, radiographic, and histopathological findings.138,139 Nevertheless, expert panels refrain from advocating a specific set of antibodies because variable expression of these markers was found in many studies. The reasons for conflicting results occurring with the same antibodies in up to 30% of cases include differences in sampling, fixation, processing, use of antigen-retrieval techniques, and use of different clones. For instance, not all recommended antibodies seem to work the same in all hands. An example is calretinin that is available in different clones. When certain polyclonal antibodies are used, staining is observed in more than 95% of epitheliod mesotheliomas,140 whereas with others, the expression varies between 40 and 70%.141,142 In the case of monoclonal antibodies, its expression is observed in less than 80% of cases.143 No differences were detected between different calretinin antibodies in the diagnosis of sarcomatous malignant mesothelioma.143 Nuclear staining for calretinin is also expressed in metastatic urothelial carcinoma and metastatic granulosa cell tumor. Additional problems arise from small-size biopsies, obtained using video-assisted thoracoscopy or blind core needle biopsies, making multiple deep biopsies of both visceral and parietal pleura desirable. A distinction between benign mesothelial processes and malignant mesothelioma especially on small biopsy may be difficult, when assessing true stromal invasion that is the most accurate indicator of malignancy. The diagnostic criteria to discriminate benign from malignant proliferation have been described,144,145 but difficulties can be encountered in the presence of entrapped mesothelial cells or when the pleural biopsies are poorly oriented, giving a false impression of invasion. In these cases, the presence of unequivocal invasion of the underlying tissues, of obvious cellular nodules with stromal expansion, of proliferating cells throughout the full thickness of the pleura, and of mild necrosis, all favor the diagnosis of malignant mesothelioma.
Clinical Toxicology vol. 48 no. 6 2010
Diagnostic issues in asbestos-related malignancies A surgical specimen showing an atypical mesothelial proliferation confined to the serosal surface may lead to suspicion of an early diffuse malignant mesothelioma with minimal invasion or of other “pre-invasive” lesions including mesothelioma in situ. These lesions may also be an occasional finding in biopsies in which diffuse malignant mesothelioma is not suspected.138,146 The diagnostic criteria for these lesions is under investigation by the International Mesothelioma Panel that recommended the term “atypical mesothelial proliferation” for such lesions.138 The histological features include an abnormal linear or tubulopapillary proliferation of mesothelial cells, the presence of atypical cells often showing prominent nucleoli, cell-to-cell variation, and the absence of exudative inflammation and/or desmoplastic underlying fibrosis.147 However, the diagnosis of mesothelioma in situ should be considered proven only when unequivocal invasion is identified in a different area of the pleura or at a different time, although the finding of atypical proliferation could identify patients at risk when observed in conjunction with positive fluid cytology. Cytology The cytological diagnosis of mesotheliomas is difficult and controversial.148 The epithelial variant of malignant mesotheliomas is usually associated with pleural effusion, whereas the sarcomatous and biphasic types rarely present in this way. Thus, the epithelial type may be diagnosed by cytology in pleural fluid, whereas the sarcomatous or biphasic variants are usually recognized with fine needle biopsies. Reported sensitivities of a cytological diagnosis of malignant mesothelioma range from 4 to 77%149 but the preparation of cellblock and the application of ancillary techniques such as cytochemistry and immunocytochemistry increase sensitivity to 98%.150 The characteristic cytological appearance is that of numerous large three-dimensional clusters of atypical cells with irregular, scalloped edge, sometimes presenting as a complex papillary pattern. However, in some cases single cells, cells in long chains, cell-in-cell arrangement, and cellular clasping characterize the pattern. In addition, the cytological diagnosis of epithelial mesothelioma in pleural fluid often requires a distinction from reactive effusions. In reactive mesothelial hyperplasia, the specimen lacks the high cellularity and the complex cellular arrangement seen in mesothelioma. However, hyperplasia may also be cellular showing atypical and variably enlarged nuclei. The most common problem is the differential diagnosis between malignant mesothelioma and metastatic adenocarcinoma. The presence of two cell populations can be helpful in the identification of metastatic neoplasm, even though cases with a pure population of adenocarcinomatous cells are frequent. A panel of immunostains on cell-block or on liquidbased preparations can be useful for differential diagnosis.151,152 These immunohistochemistry panels include positive and
491 negative markers that identify both mesothelioma and adenocarcinoma, as described above for histological diagnosis.
Conclusions This review suggests how physicians might decide what kind of information should be used in the diagnosis of asbestosrelated malignancies. Assessment of exposure is crucial and requires accurate medical and occupational histories. Quantitative analysis of asbestos body burden is better performed in digested lung tissues by counting asbestos bodies by light microscopy40 and/or uncoated fibers by transmission electron microscopy.19 These counts should possibly be in the range recorded for a given disease by the same laboratory. The benefits of screenings for asbestos-related malignancies are equivocal. A periodic assessment of individuals who are, or have been, exposed to asbestos still rely on clinical judgment by weighting specific risk factors (i.e., intensity of exposure to asbestos, asbestos bodies in the sputum, smoking, etc.), procedure-related risks (e.g., radiation, biopsies), and clinical conditions (e.g., COPD, lung function). The attribution of lung cancer to asbestos exposure may be difficult in a clinical setting.2,19,111,113–115,153 Although a distinction between risk of disease and actually having the disease because of that risk is obvious,154 proposed diagnostic criteria do not often underline the difference.19 Thus, the risk of lung cancer due to asbestos exposure is described and can be calculated from epidemiological correlations, whereas the etiological diagnosis on a patient remains at least in part judgmental. Indeed, in both circumstances limitations are similar and include the difficulties in assessing asbestos body burden, the fact that virtually all these patients are also tobacco smokers or former smokers, and, finally, that we ignore the mechanisms of asbestos-related diseases. Given the premise that asbestosis is present in non-neoplastic tissue, lung cancer can be causally linked to asbestos. However, because of the notion that a sufficient asbestos exposure or body burden is enough by themselves for causality, it follows that the assessment of both lung fibrosis and asbestos body burden is necessary. Practical problems also arise when epidemiological evidence of the interaction between asbestos and smoking has to be translated into individual cases. Unresolved issues include the lowest asbestos dose needed for interaction to occur and the type of exposure, given the varying risk slopes in different industrial settings.155 Consequently, no criteria have been suggested to assess the relative contribution of asbestos and smoking in the development of lung cancer.19 The diagnosis of malignant mesothelioma may be complex and controversial, although a diagnostic accuracy of all variants of mesothelioma greater than 95% is possible.137,143,156–158 Cytological examination of serous effusions permits the diagnosis of malignant mesothelioma when ancillary techniques are employed.
Clinical Toxicology vol. 48 no. 6 2010
492 In summary, an agreement among experts on asbestosrelated malignancies is far from consensus. The final conclusion cannot be different from the premise: there are no valid substitutes for clinical judgment when meaningful diagnoses and follow-up of formerly exposed individuals are to be made.159
Declaration of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this paper.
References 1. World Health Organization. Elimination of asbestos-related diseases. 2006; Geneva; WHO/SDE/OEH/06.03. 2. Henderson DW, Rodelsperger K, Woitowitz HJ, Leigh J. After Helsinki: a multidisciplinary review of the relationship between asbestos exposure and lung cancer, with emphasis on studies published during 1997–2004. Pathology 2004; 36:517–550. 3. Mossman BT, Gee JB. Asbestos-related diseases. N Engl J Med 1989; 320:1721–1730. 4. Robinson BW, Chahinian AP. Mesothelioma. London: Martin Dunitz; 2002. 5. Roggli VL, Oury TD, Sporn TA. Pathology of Asbestos-Associated Disease. New York: Springer; 2004. 6. Hessel PA, Gamble JF, McDonald JC. Asbestos, asbestosis, and lung cancer: a critical assessment of the epidemiological evidence. Thorax 2005; 60:433–436. 7. Becklake MR, Bagatin E, Neder JA. Asbestos-related diseases of the lungs and pleura: uses, trends and management over the last century. Int J Tuberc Lung Dis 2007; 11:356–369. 8. Kamp DW. Asbestos-induced lung diseases: an update. Transl Res 2009; 153:143–152. 9. Motley RL, Patrick CWJ, Kearse AM. Medicolegal aspects of asbestos-related diseases: a plaintiff’s attorney’s perspective. In: Roggli VL, Oury TD, Sporn TA, eds. Pathology of Asbestos-Associated Disease. New York: Springer; 2004:355–376. 10. Parnell AH. Medicolegal aspects of asbestos-related disease: a defendant’s attorney’s perspective. In: Roggli VL, Oury TD, Sporn TA, eds. Pathology of Asbestos-Associated Disease. New York: Springer; 2004:377–401. 11. Holmes E, Garshick E. The reproducibility of the self-report of occupational exposure to asbestos and dust. J Occup Med 1991; 33:134–138. 12. Ahrens W, Jockel KH, Brochard P, Bolm-Audorff U, Grossgarten K, Iwatsubo Y, Orlowski E, Pohlabeln H, Berrino F. Retrospective assessment of asbestos exposure–I. Case-control analysis in a study of lung cancer: efficiency of job-specific questionnaires and job exposure matrices. Int J Epidemiol 1993; 22(Suppl 2):S83–S95. 13. Orlowski E, Pohlabeln H, Berrino F, Ahrens W, Bolm-Audorff U, Grossgarten K, Iwatsubo Y, Jöckel KH, Brochard P. Retrospective assessment of asbestos exposure–II. At the job level: complementarity of job-specific questionnaire and job exposure matrices. Int J Epidemiol 1993; 22(Suppl 2):S96–S105. 14. Burdorf A, Swuste P. An expert system for the evaluation of historical asbestos exposure as diagnostic criterion in asbestos-related diseases. Ann Occup Hyg 1999; 43:57–66. 15. Rice C, Heineman EF. An asbestos job exposure matrix to characterize fiber type, length, and relative exposure intensity. Appl Occup Environ Hyg 2003; 18:506–512.
M. Lotti et al. 16. Nam JM, Rice C, Gail MH. Comparison of asbestos exposure assessments by next-of-kin respondents, by an occupational hygienist, and by a job-exposure matrix from the National Occupational Hazard Survey. Am J Ind Med 2005; 47:443–450. 17. Dement JM, Kuempel ED, Zumwalde RD, Smith RJ, Stayner LT, Loomis D. Development of a fibre size-specific job-exposure matrix for airborne asbestos fibres. Occup Environ Med 2008; 65:605–612. 18. Swuste P, Dahhan M, Burdorf A. Linking expert judgement and trends in occupational exposure into a job-exposure matrix for historical exposure to asbestos in the Netherlands. Ann Occup Hyg 2008; 52:397–403. 19. Asbestos, asbestosis, and cancer: the Helsinki criteria for diagnosis and attribution. Scand J Work Environ Health 1997; 23:311–316. 20. Hillerdal G. Mesothelioma: cases associated with non-occupational and low dose exposures. Occup Environ Med 1999; 56:505–513. 21. Gibbs AR, Pooley FD. Analysis and interpretation of inorganic mineral particles in “lung” tissues. Thorax 1996; 51:327–334. 22. Lim HS, Kim JY, Sakai K, Hisanaga N. Pulmonary asbestos and nonasbestos fiber concentrations in autopsied inhabitants in Pohang, Korea. Ind Health 2004; 42:163–170. 23. Tossavainen A, Kovalevsky E, Vanhala E, Tuomi T. Pulmonary mineral fibers after occupational and environmental exposure to asbestos in the Russian chrysotile industry. Am J Ind Med 2000; 37:327–333. 24. Anttila S, Karjalainen A, Taikina-aho O, Kyyronen P, Vainio H. Lung cancer in the lower lobe is associated with pulmonary asbestos fiber count and fiber size. Environ Health Perspect 1993; 101:166–170. 25. Wagner JC, Newhouse ML, Corrin B, Rossiter CE, Griffiths DM. Correlation between fibre content of the lung and disease in east London asbestos factory workers. Br J Ind Med 1988; 45:305–308. 26. Green FH, Harley R, Vallyathan V, Althouse R, Fick G, Dement J, Mitha R, Pooley F. Exposure and mineralogical correlates of pulmonary fibrosis in chrysotile asbestos workers. Occup Environ Med 1997; 54:549–559. 27. Rödelsperger K, Mándi A, Tossavainen A, Brückel B, Barbisan P, Woitowitz HJ. Inorganic fibres in the lung tissue of Hungarian and German lung cancer patients. Int Arch Occup Environ Health 2001; 74:133–138. 28. Teschler H, Thompson AB, Dollenkamp R, Konietzko N, Costabel U. Relevance of asbestos bodies in sputum. Eur Respir J 1996; 9:680–686. 29. de Klerk NH, Musk AW, Williams V, Filion PR, Whitaker D, Shilkin KB. Comparison of measures of exposure to asbestos in former crocidolite workers from Wittenoom Gorge, W. Australia. Am J Ind Med 1996; 30:579–587. 30. Churg A, Vedal S. Fiber burden and patterns of asbestos-related disease in workers with heavy mixed amosite and chrysotile exposure. Am J Respir Crit Care Med 1994; 150:663–669. 31. Berry G. Models for mesothelioma incidence following exposure to fibers in terms of timing and duration of exposure and the biopersistence of the fibers. Inhal Toxicol 1999; 11:111–130. 32. Finkelstein MM, Dufresne A. Inferences on the kinetics of asbestos deposition and clearance among chrysotile miners and millers. Am J Ind Med 1999; 35:401–412. 33. Dodson RF, Hammar SP, Poye LW. A technical comparison of evaluating asbestos concentration by phase-contrast microscopy (PCM), scanning electron microscopy (SEM), and analytical transmission electron microscopy (ATEM) as illustrated from data generated from a case report. Inhal Toxicol 2008; 20:723–732. 34. Gylseth B, Churg A, Davis JM, Johnson N, Morgan A, Mowe G, Rogers A, Roggli V. Analysis of asbestos fibers and asbestos bodies in tissue samples from human lung. An international interlaboratory trial. Scand J Work Environ Health 1985; 11:107–110. 35. Roggli VL, Sanders LL. Asbestos content of lung tissue and carcinoma of the lung: a clinicopathologic correlation and mineral fiber analysis of 234 cases. Ann Occup Hyg 2000; 44:109–117. 36. Sartorelli P, Scancarello G, Romeo R, Marcianò G, Rottoli P, Arcangeli G, Palmi S. Asbestos exposure assessment by mineralogical analysis of bronchoalveolar lavage fluid. J Occup Environ Med 2001; 43:872–881.
Clinical Toxicology vol. 48 no. 6 2010
Diagnostic issues in asbestos-related malignancies 37. Sporn TA, Roggli VL. Asbestosis. In: Roggli VL, Oury TD, Sporn TA, eds. Pathology of Asbestos-Associated Disease. New York: Springer; 2004:71–103. 38. Roggli VL, Sharma A. Analysis of tissue mineral fiber content. In: Roggli VL, Oury TD, Sporn TA, eds. Pathology of Asbestos-Associated Disease. New York: Springer; 2004:309–354. 39. Weiss W. Asbestosis: a marker for the increased risk of lung cancer among workers exposed to asbestos. Chest 1999; 115:536–549. 40. Mollo F, Magnani C, Bo P, Burlo P, Cravello M. The attribution of lung cancers to asbestos exposure: a pathologic study of 924 unselected cases. Am J Clin Pathol 2002; 117:90–95. 41. Churg A. Asbestos-related disease in the workplace and the environment: controversial issues. Monogr Pathol 1993; 36:54–77. 42. Everatt RP, Smolianskiene G, Tossavainen A, Cicenas S, Jankauskas R. Occupational asbestos exposure among respiratory cancer patients in Lithuania. Am J Ind Med 2007; 50:455–463. 43. Steele JP, Klabatsa A, Fennell DA, Palläska A, Sheaff MT, Evans MT, Shamash J, Rudd RM. Prognostic factors in mesothelioma. Lung Cancer 2005; 49(Suppl 1):S49–S52. 44. Yau G, Lock M, Rodrigues G. Systematic review of baseline low-dose CT lung cancer screening. Lung Cancer 2007; 58:161–170. 45. Postmus PE. Screening for lung cancer, an ongoing debate. Ann Oncol 2008; 19(Suppl 7):25–27. 46. The International Early Lung Cancer Action Program Investigators, Henschke CI, Yankelevitz DF, Libby DM, Pasmantier MW, Smith JP, Miettinen OS. Survival of patients with stage I lung cancer detected on CT screening. N Engl J Med 2006; 355:1763–1771. 47. Bach PB, Jett JR, Pastorino U, Tockman MS, Swensen SJ, Begg CB. Computed tomography screening and lung cancer outcomes. J Am Med Assoc 2007; 297:953–961. 48. Yankelevitz DF. Screening for a cancer: acting on social responsibility. Am J Roentgenol 2007; 188:1171–1172. 49. Lee CI, Forman HP. CT screening for lung cancer: implications on social responsibility. Am J Roentgenol 2007; 188:297–298. 50. Patz EF Jr, Goodman PC, Bepler G. Screening for lung cancer. N Engl J Med 2000; 343:1627–1633. 51. Gohagan JK, Marcus PM, Fagerstrom RM, Pinsky PF, Kramer BS, Prorok PC, Ascher S, Bailey W, Brewer B, Church T, Engelhard D, Ford M, Fouad M, Freedman M, Gelmann E, Gierada D, Hocking W, Inampudi S, Irons B, Johnson CC, Jones A, Kucera G, Kvale P, Lappe K, Manor W, Moore A, Nath H, Neff S, Oken M, Plunkett M, Price H, Reding D, Riley T, Schwartz M, Spizarny D, Yoffie R, Zylak C, Lung Screening Study Research Group. Final results of the Lung Screening Study, a randomized feasibility study of spiral CT versus chest X-ray screening for lung cancer. Lung Cancer 2005; 47:9–15. 52. Mollo F, Pira E, Piolatto G, Bellis D, Burlo P, Andreozzi A, Bontempi S, Negri E. Lung adenocarcinoma and indicators of asbestos exposure. Int J Cancer 1995; 60:289–293. 53. Tiitola M, Kivisaari L, Huuskonen MS, Mattson K, Koskinen H, Lehtola H, Zitting A, Vehmas T. Computed tomography screening for lung cancer in asbestos-exposed workers. Lung Cancer 2002; 35:17–22. 54. Das M, Mühlenbruch G, Mahnken AH, Hering KG, Sirbu H, Zschiesche W, Knoll L, Felten MK, Kraus T, Günther RW, Wildberger JE. Asbestos Surveillance Program Aachen (ASPA): initial results from baseline screening for lung cancer in asbestos-exposed high-risk individuals using low-dose multidetector-row CT. Eur Radiol 2007; 17:1193–1199. 55. Fasola G, Belvedere O, Aita M, Zanin T, Follador A, Cassetti P, Meduri S, De Pangher V, Pignata G, Rosolen V, Barbone F, Grossi F. Low-dose computed tomography screening for lung cancer and pleural mesothelioma in an asbestos-exposed population: baseline results of a prospective, nonrandomized feasibility trial – an Alpe-adria Thoracic Oncology Multidisciplinary Group Study (ATOM 002). Oncologist 2007; 12:1215–1224. 56. Vierikko T, Järvenpää R, Autti T, Oksa P, Huuskonen M, Kaleva S, Laurikka J, Kajander S, Paakkola K, Saarelainen S, Salomaa ER,
493
57.
58. 59. 60. 61.
62.
63.
64.
65.
66.
67.
68.
69.
Tossavainen A, Tukiainen P, Uitti J, Vehmas T. Chest CT screening of asbestos-exposed workers: lung lesions and incidental findings. Eur Respir J 2007; 29:78–84. Mastrangelo G, Ballarin MN, Bellini E, Bizzotto R, Zannol F, Gioffrè F, Gobbi M, Tessadri G, Marchiori L, Marangi G, Bozzolan S, Lange JH, Valentini F, Spolaore P. Feasibility of a screening programme for lung cancer in former asbestos workers. Occup Med (Lond) 2008; 58:175–180. Herbst RS, Heymach JV, Lippman SM. Lung cancer. N Engl J Med 2008; 359:1367–1380. Dalton WS, Friend SH. Cancer biomarkers – an invitation to the table. Science 2006; 312:1165–1168. Hardy J, Singleton A. Genomewide association studies and human disease. N Engl J Med 2009; 360:1759–1768. Amos CI, Wu X, Broderick P, Gorlov IP, Gu J, Eisen T, Dong Q, Zhang Q, Gu X, Vijayakrishnan J, Sullivan K, Matakidou A, Wang Y, Mills G, Doheny K, Tsai YY, Chen WV, Shete S, Spitz MR, Houlston RS. Genome-wide association scan of tag SNPs identifies a susceptibility locus for lung cancer at 15q25.1. Nat Genet 2008; 40:616–622. Hung RJ, McKay JD, Gaborieau V, Boffetta P, Hashibe M, Zaridze D, Mukeria A, Szeszenia-Dabrowska N, Lissowska J, Rudnai P, Fabianova E, Mates D, Bencko V, Foretova L, Janout V, Chen C, Goodman G, Field JK, Liloglou T, Xinarianos G, Cassidy A, McLaughlin J, Liu G, Narod S, Krokan HE, Skorpen F, Elvestad MB, Hveem K, Vatten L, Linseisen J, Clavel-Chapelon F, Vineis P, Bueno-de-Mesquita HB, Lund E, Martinez C, Bingham S, Rasmuson T, Hainaut P, Riboli E, Ahrens W, Benhamou S, Lagiou P, Trichopoulos D, Holcátová I, Merletti F, Kjaerheim K, Agudo A, Macfarlane G, Talamini R, Simonato L, Lowry R, Conway DI, Znaor A, Healy C, Zelenika D, Boland A, Delepine M, Foglio M, Lechner D, Matsuda F, Blanche H, Gut I, Heath S, Lathrop M, Brennan P. A susceptibility locus for lung cancer maps to nicotinic acetylcholine receptor subunit genes on 15q25. Nature 2008; 452:633–637. Yu D, Zhang X, Liu J, Yuan P, Tan W, Guo Y, Sun T, Zhao D, Yang M, Liu J, Xu B, Lin D. Characterization of functional excision repair cross-complementation group 1 variants and their association with lung cancer risk and prognosis. Clin Cancer Res 2008; 14:2878–2886. Thorgeirsson TE, Geller F, Sulem P, Rafnar T, Wiste A, Magnusson KP, Manolescu A, Thorleifsson G, Stefansson H, Ingason A, Stacey SN, Bergthorsson JT, Thorlacius S, Gudmundsson J, Jonsson T, Jakobsdottir M, Saemundsdottir J, Olafsdottir O, Gudmundsson LJ, Bjornsdottir G, Kristjansson K, Skuladottir H, Isaksson HJ, Gudbjartsson T, Jones GT, Mueller T, Gottsäter A, Flex A, Aben KK, de Vegt F, Mulders PF, Isla D, Vidal MJ, Asin L, Saez B, Murillo L, Blondal T, Kolbeinsson H, Stefansson JG, Hansdottir I, Runarsdottir V, Pola R, Lindblad B, van Rij AM, Dieplinger B, Haltmayer M, Mayordomo JI, Kiemeney LA, Matthiasson SE, Oskarsson H, Tyrfingsson T, Gudbjartsson DF, Gulcher JR, Jonsson S, Thorsteinsdottir U, Kong A, Stefansson K. A variant associated with nicotine dependence, lung cancer and peripheral arterial disease. Nature 2008; 452:638–642. Ma PC, Blaszkowsky L, Bharti A, Ladanyi A, Kraeft SK, Bruno A, Skarin AT, Chen LB, Salgia R. Circulating tumor cells and serum tumor biomarkers in small cell lung cancer. Anticancer Res 2003; 23:49–62. Carpagnano GE, Spanevello A, Curci C, Salerno F, Palladino GP, Resta O, Di Gioia G, Carpagnano F, Foschino Barbaro MP. IL-2, TNF-alpha, and leptin: local versus systemic concentrations in NSCLC patients. Oncol Res 2007; 16:375–381. Gautschi O, Bigosch C, Huegli B, Jermann M, Marx A, Chasse E, Ratschiller D, Weder W, Joerger M, Betticher DC, Stahel RA, Ziegler A. Circulating deoxyribonucleic acid as prognostic marker in nonsmall-cell lung cancer patients undergoing chemotherapy. J Clin Oncol 2004; 22:4157–4164. Chen XQ, Stroun M, Magnenat JL, Nicod LP, Kurt AM, Lyautey J, Lederrey C, Anker P. Microsatellite alterations in plasma DNA of small cell lung cancer patients. Nat Med 1996; 2:1033–1035. Belinsky SA, Liechty KC, Gentry FD, Wolf HJ, Rogers J, Vu K, Haney J, Kennedy TC, Hirsch FR, Miller Y, Franklin WA, Herman
Clinical Toxicology vol. 48 no. 6 2010
494
70.
71.
72.
73.
74.
75.
76.
77.
78.
79. 80.
81.
82.
83.
84. 85.
86.
87.
88.
M. Lotti et al. JG, Baylin SB, Bunn PA, Byers T. Promoter hypermethylation of multiple genes in sputum precedes lung cancer incidence in a high-risk cohort. Cancer Res 2006; 66:3338–3344. Belinsky SA, Grimes MJ, Casas E, Stidley CA, Franklin WA, Bocklage TJ, Johnson DH, Schiller JH. Predicting gene promoter methylation in nonsmall-cell lung cancer by evaluating sputum and serum. Br J Cancer 2007; 96:1278–1283. Patz EF, Jr, Campa MJ, Gottlin EB, Kusmartseva I, Guan XR, Herndon JE, 2nd. Panel of serum biomarkers for the diagnosis of lung cancer. J Clin Oncol 2007; 25:5578–5583. Bharti A, Ma PC, Salgia R. Biomarker discovery in lung cancer– promises and challenges of clinical proteomics. Mass Spectrom Rev 2007; 26:451–466. Baker SG, Kramer BS, McIntosh M, Patterson BH, Shyr Y, Skates S. Evaluating markers for the early detection of cancer: overview of study designs and methods. Clin Trials 2006; 3:43–56. Nymark P, Wikman H, Ruosaari S, Hollmen J, Vanhala E, Karjalainen A, Anttila S, Knuutila S. Identification of specific gene copy number changes in asbestos-related lung cancer. Cancer Res 2006; 66:5737–5743. Wikman H, Ruosaari S, Nymark P, Sarhadi VK, Saharinen J, Vanhala E, Karjalainen A, Hollmén J, Knuutila S, Anttila S. Gene expression and copy number profiling suggests the importance of allelic imbalance in 19p in asbestos-associated lung cancer. Oncogene 2007; 26:4730–4737. Andujar P, Wang J, Descatha A, Galateau-Salle F, Abd-Alsamad I, Billon-Galland MA, Blons H, Clin B, Danel C, Housset B, LaurentPuig P, Le Pimpec-Barthes F, Letourneux M, Monnet I, Régnard JF, Renier A, Zucman-Rossi J, Pairon JC, Jaurand MC. p16(INK4A) inactivation mechanisms in non-small-cell lung cancer patients occupationally exposed to asbestos. Lung Cancer 2010; 67:23–30. Björkqvist AM, Tammilehto L, Anttila S, Mattson K, Knuutila S. Recurrent DNA copy number changes in 1q, 4q, 6q, 9p, 13q, 14q and 22q detected by comparative genomic hybridization in malignant mesothelioma. Br J Cancer 1997; 75:523–527. Sandberg AA, Bridge JA. Updates on the cytogenetics and molecular genetics of bone and soft tissue tumors. Mesothelioma. Cancer Genet Cytogenet 2001; 127:93–110. Ladanyi M. Implications of P16/CDKN2A deletion in pleural mesotheliomas. Lung Cancer 2005; 49(Suppl 1):S95–S98. Yang CT, You L, Uematsu K, Yeh CC, McCormick F, Jablons DM. p14(ARF) modulates the cytolytic effect of ONYX-015 in mesothelioma cells with wild-type p53. Cancer Res 2001; 61:5959–5963. Hirao T, Bueno R, Chen CJ, Gordon GJ, Heilig E, Kelsey KT. Alterations of the p16(INK4) locus in human malignant mesothelial tumors. Carcinogenesis 2002; 23:1127–1130. Bianchi AB, Mitsunaga SI, Cheng JQ, Klein WM, Jhanwar SC, Seizinger B, Kley N, Klein-Szanto AJ, Testa JR. High frequency of inactivating mutations in the neurofibromatosis type 2 gene (NF2) in primary malignant mesotheliomas. Proc Natl Acad Sci USA 1995; 92:10854–10858. Baser ME. Contributors to the International NF2 Mutation Database. The distribution of constitutional and somatic mutations in the neurofibromatosis 2 gene. Hum Mutat 2006; 27:297–306. Sekido Y. Genomic abnormalities and signal transduction dysregulation in malignant mesothelioma cells. Cancer Sci 2010; 101:1–6. Pass HI, Lott D, Lonardo F, Harbut M, Liu Z, Tang N, Carbone M, Webb C, Wali A. Asbestos exposure, pleural mesothelioma, and serum osteopontin levels. N Engl J Med 2005; 353:1564–1573. Robinson BW, Creaney J, Lake R, Nowak A, Musk AW, de Klerk N, Winzell P, Hellstrom KE, Hellstrom I. Mesothelin-family proteins and diagnosis of mesothelioma. Lancet 2003; 362:1612–1616. Cullen MR. Serum osteopontin levels – is it time to screen asbestosexposed workers for pleural mesothelioma? N Engl J Med 2005; 353:1617–1618. Creaney J, Yeoman D, Naumoff LK, Hof M, Segal A, Musk AW, De Klerk N, Horick N, Skates SJ, Robinson BW. Soluble mesothelin in
89.
90.
91.
92. 93. 94. 95. 96.
97.
98.
99.
100.
101.
102. 103.
104. 105.
106.
107. 108.
109.
110.
effusions: a useful tool for the diagnosis of malignant mesothelioma. Thorax 2007; 62:569–576. Kato Y, Tsuta K, Seki K, Maeshima AM, Watanabe S, Suzuki K, Asamura H, Tsuchiya R, Matsuno Y. Immunohistochemical detection of GLUT-1 can discriminate between reactive mesothelium and malignant mesothelioma. Mod Pathol 2007; 20:215–220. Minna JD. Neoplasms of the lung. In: Kasper DL, Braunwald E, Fauci A, Hauser SL, Longo DL, Jamesson JL, eds. Harrison’s Principles of Internal Medicine. New York: McGraw-Hill Medical Publishing Division; 2005:506–516. Alberg AJ, Ford JG, Samet JM, American College of Chest Physicians. Epidemiology of lung cancer: ACCP evidence-based clinical practice guidelines (2nd edition). Chest 2007; 132(Suppl 3):S29–S55. Khuder SA. Effect of cigarette smoking on major histological types of lung cancer: a meta-analysis. Lung Cancer 2001; 31:139–148. Dubey S, Powell CA. Update in lung cancer 2007. Am J Respir Crit Care Med 2008; 177:941–946. Sun S, Schiller JH, Gazdar AF. Lung cancer in never smokers–a different disease. Nat Rev Cancer 2007; 7:778–790. Subramanian J, Govindan R. Molecular genetics of lung cancer in people who have never smoked. Lancet Oncol 2008; 9:676–682. Lash TL, Crouch EA, Green LC. A meta-analysis of the relation between cumulative exposure to asbestos and relative risk of lung cancer. Occup Environ Med 1997; 54:254–263. Hodgson JT, Darnton A. The quantitative risks of mesothelioma and lung cancer in relation to asbestos exposure. Ann Occup Hyg 2000; 44:565–601. Berry G, Gibbs GW. An overview of the risk of lung cancer in relation to exposure to asbestos and of taconite miners. Regul Toxicol Pharmacol 2008; 52(Suppl 1):S218–S222. Doll R, Peto R. The causes of cancer: quantitative estimates of avoidable risks of cancer in the United States today. J Natl Cancer Inst 1981; 66:1191–1308. Gustavsson P, Ahlbom A, Andersson T, Schéele P. Calculation of fractions of lung cancer incidence attributable to occupational exposure to asbestos and combustion products in Stockholm, Sweden. Eur J Epidemiol 2003; 18:937–940. Marinaccio A, Scarselli A, Binazzi A, Mastrantonio M, Ferrante P, Iavicoli S. Magnitude of asbestos-related lung cancer mortality in Italy. Br J Cancer 2008; 99:173–175. Craighead JE, Mossman BT. The pathogenesis of asbestos-associated diseases. N Engl J Med 1982; 306:1446–1455. Lee BW, Wain JC, Kelsey KT, Wiencke JK, Christiani DC. Association of cigarette smoking and asbestos exposure with location and histology of lung cancer. Am J Respir Crit Care Med 1998; 157:748–755. Boffetta P. Health effects of asbestos exposure in humans: a quantitative assessment. Med Lav 1998; 89:471–480. McDonald J. Asbestos related disease: an epidemiological review. In: Wagner JC, ed. Biological Effects of Mineral Fibers. Lyon: International Agency for Research on Cancer; 1980:811–817 (Scientific publication no. 30). McDonald AD, Fry JS, Woolley AJ, McDonald JC. Dust exposure and mortality in an American chrysotile asbestos friction products plant. Br J Ind Med 1984; 41:151–157. Finkelstein MM. Mortality among long-term employees of an Ontario asbestos-cement factory. Br J Ind Med 1983; 40:138–144. Weill H, Hughes J, Waggenspack C. Influence of dose and fiber type on respiratory malignancy risk in asbestos cement manufacturing. Am Rev Respir Dis 1979; 120:345–354. McDonald AD, Fry JS, Woolley AJ, McDonald J. Dust exposure and mortality in an American chrysotile textile plant. Br J Ind Med 1983; 40:361–367. McDonald AD, Fry JS, Woolley AJ, McDonald JC. Dust exposure and mortality in an American factory using chrysotile, amosite, and crocidolite in mainly textile manufacture. Br J Ind Med 1982; 39:368–374.
Clinical Toxicology vol. 48 no. 6 2010
Diagnostic issues in asbestos-related malignancies 111. Tossavainen A. International expert meeting on new advances in the radiology and screening of asbestos-related diseases. Scand J Work Environ Health 2000; 26:449–454. 112. Pierce JS, McKinley MA, Paustenbach DJ, Finley BL. An evaluation of reported no-effect chrysotile asbestos exposures for lung cancer and mesothelioma. Crit Rev Toxicol 2008; 38:191–214. 113. Gibbs A, Attanoos RL, Churg A, Weill H. The “Helsinki criteria” for attribution of lung cancer to asbestos exposure: how robust are the criteria? Arch Pathol Lab Med 2007; 131:181–184. 114. Greenberg M. The “Helsinki Criteria” for attribution of lung cancer to asbestos exposure: how robust are the criteria? Arch Pathol Lab Med 2007; 131:1630. 115. Roggli VL, Hammar SP, Maddox JC, Henderson DW. Re: The “Helsinki Criteria” for attribution of lung cancer to asbestos exposure: how robust are the criteria? Arch Pathol Lab Med 2008; 132:1386–1387. 116. Camus M, Siemiatycki J, Meek B. Nonoccupational exposure to chrysotile asbestos and the risk of lung cancer. N Engl J Med 1998; 338:1565–1571. 117. Egilman D, Reinert A. Lung cancer and asbestos exposure: asbestosis is not necessary. Am J Ind Med 1996; 30:398–406. 118. Reid A, de Klerk N, Ambrosini GL, Olsen N, Pang SC, Berry G, Musk AW. The effect of asbestosis on lung cancer risk beyond the dose related effect of asbestos alone. Occup Environ Med 2005; 62:885–889. 119. Hughes JM, Weill H. Asbestosis as a precursor of asbestos related lung cancer: results of a prospective mortality study. Br J Ind Med 1991; 48:229–233. 120. Jones RN, Hughes JM, Weill H. Asbestos exposure, asbestosis, and asbestos-attributable lung cancer. Thorax 1996; 51(Suppl 2):S9–S15. 121. Loomis D, Dement JM, Wolf SH, Richardson DB. Lung cancer mortality and fiber exposures among North Carolina asbestos textile workers. Occup Environ Med 2009; 66:535–542. 122. Oksa P, Klockars M, Karjalainen A, Huuskonen MS, Vattulainen K, Pukkala E, Nordman H. Progression of asbestosis predicts lung cancer. Chest 1998; 113:1517–1521. 123. American Thoracic Society. Diagnosis and initial management of nonmalignant diseases related to asbestos. Am J Respir Crit Care Med 2004; 170:691–715. 124. Adesina AM, Vallyathan V, McQuillen EN, Weaver SO, Craighead JE. Bronchiolar inflammation and fibrosis associated with smoking. A morphologic cross-sectional population analysis. Am Rev Respir Dis 1991; 143:144–149. 125. Ross RM. The clinical diagnosis of asbestosis in this century requires more than a chest radiograph. Chest 2003; 124:1120–1128. 126. Gamsu G, Salmon CJ, Warnock ML, Blanc PD. CT quantification of interstitial fibrosis in patients with asbestosis: a comparison of two methods. Am J Roentgenol 1995; 164:63–68. 127. Copley SJ, Wells AU, Sivakumaran P, Rubens MB, Lee YC, Desai SR, MacDonald SL, Thompson RI, Colby TV, Nicholson AG, du Bois RM, Musk AW, Hansell DM. Asbestosis and idiopathic pulmonary fibrosis: comparison of thin-section CT features. Radiology 2003; 229:731–736. 128. McDonald JC, Sebastien P, Armstrong B. Radiological survey of past and present vermiculite miners exposed to tremolite. Br J Ind Med 1986; 43:445–449. 129. Van Cleemput J, De Raeve H, Verschakelen JA, Rombouts J, Lacquet LM, Nemery B. Surface of localized pleural plaques quantitated by computed tomography scanning: no relation with cumulative asbestos exposure and no effect on lung function. Am J Respir Crit Care Med 2001; 163:705–710. 130. Hillerdal G, Henderson DW. Asbestos, asbestosis, pleural plaques and lung cancer. Scand J Work Environ Health 1997; 23:93–103. 131. Weill D, Weill H. Diagnosis and initial management of nonmalignant diseases related to asbestos. Am J Respir Crit Care Med 2005; 171:527–528. 132. Wraith D, Mengersen K. A Bayesian approach to assess interaction between known risk factors: the risk of lung cancer from exposure to asbestos and smoking. Stat Methods Med Res 2008; 17:171–189.
495 133. Nelson HH, Kelsey KT. The molecular epidemiology of asbestos and tobacco in lung cancer. Oncogene 2002; 21:7284–7288. 134. Churg A, Stevens B. Enhanced retention of asbestos fibers in the airways of human smokers. Am J Respir Crit Care Med 1995; 151:1409– 1413. 135. Travis WD, Brambilla E, Harris CC, Muller-Hermelink HK. World Health Organization Classification of Tumors. Pathology and Genetics: Tumors of the Lung, Pleura, Thymus and Heart. Lyon: IARC, 2004:128–136. 136. Corson JM. Pathology of mesothelioma. Thorac Surg Clin 2004; 14:447–460. 137. Ordonez NG. Immunohistochemical diagnosis of epithelioid mesothelioma: an update. Arch Pathol Lab Med 2005; 129:1407–1414. 138. Galateau-Salle F, Brambilla E, Cagle P. Pathology of Malignant Mesothelioma. International Mesothelioma Panel. London: SpringerVerlag; 2006. 139. Attanoos RL, Dojcinov SD, Webb R, Gibbs AR. Anti-mesothelial markers in sarcomatoid mesothelioma and other spindle cell neoplasms. Histopathology 2000; 37:224–231. 140. Doglioni C, Dei Tos AP, Laurino L, Iuzzolino P, Chiarelli C, Celio MR, Viale G. Calretinin: a novel immunocytochemical marker for mesothelioma. Am J Surg Pathol 1996; 20:1037–1046. 141. Suster S, Moran CA. Applications and limitations of immunohistochemistry in the diagnosis of malignant mesothelioma. Adv Anat Pathol 2006; 13:316–329. 142. Ordonez NG. Value of calretinin immunostaining in differentiating epithelial mesothelioma from lung adenocarcinoma. Mod Pathol 1998; 11:929–933. 143. Granville LA, Younes M, Churg A, Roggli VL, Henderson DW, Cagle PT. Comparison of monoclonal versus polyclonal calretinin antibodies for immunohistochemical diagnosis of malignant mesothelioma. Appl Immunohistochem Mol Morphol 2005; 13:75–79. 144. Churg A, Colby TV, Cagle P, Corson J, Gibbs AR, Gilks B, Grimes M, Hammar S, Roggli V, Travis WD. The separation of benign and malignant mesothelial proliferations. Am J Surg Pathol 2000; 24:1183–1200. 145. Cagle PT, Churg A. Differential diagnosis of benign and malignant mesothelial proliferations on pleural biopsies. Arch Pathol Lab Med 2005; 129:1421–1427. 146. Kerrigan SA, Cagle P, Churg A. Malignant mesothelioma of the peritoneum presenting as an inflammatory lesion: a report of four cases. Am J Surg Pathol 2003; 27:248–253. 147. Henderson DW, Shilkin KB, Whitaker D. Reactive mesothelial hyperplasia vs mesothelioma, including mesothelioma in situ: a brief review. Am J Clin Pathol 1998; 110:397–404. 148. Triol JH, Conston AS, Chandler SV. Malignant mesothelioma. Cytopathology of 75 cases seen in a New Jersey community hospital. Acta Cytol 1984; 28:37–45. 149. Renshaw AA, Dean BR, Antman KH, Sugarbaker DJ, Cibas ES. The role of cytologic evaluation of pleural fluid in the diagnosis of malignant mesothelioma. Chest 1997; 111:106–109. 150. el-Habashi A, el-Morsi B, Freeman SM, el-Didi M, Marrogi AJ. Tumor oncogenic expression in malignant effusions as a possible method to enhance cytologic diagnostic sensitivity. An immunocytochemical study of 87 cases. Am J Clin Pathol 1995; 103:206–214. 151. Bedrossian CW, Bonsib S, Moran C. Differential diagnosis between mesothelioma and adenocarcinoma: a multimodal approach based on ultrastructure and immunocytochemistry. Semin Diagn Pathol 1992; 9:124–140. 152. Ko EC, Jhala NC, Shultz JJ, Chhieng DC. Use of a panel of markers in the differential diagnosis of adenocarcinoma and reactive mesothelial cells in fluid cytology. Am J Clin Pathol 2001; 116:709–715. 153. Attanoos RL, Gibbs AR. An approach to industrial post mortems. Histopathology 2009; 54:134–142. 154. Cagle PT. Criteria for attributing lung cancer to asbestos exposure. Am J Clin Pathol 2002; 117:9–15.
Clinical Toxicology vol. 48 no. 6 2010
496 155. Case BW. Asbestos, smoking, and lung cancer: interaction and attribution. Occup Environ Med 2006; 63:507–508. 156. Zellos LS, Sugarbaker DJ. Multimodality treatment of diffuse malignant pleural mesothelioma. Semin Oncol 2002; 29:41–50. 157. King JE, Hasleton PS. Immunohistochemistry and the diagnosis of malignant mesothelioma. Histopathology 2001; 38:471–476. 158. Facchetti F, Lonardi S, Gentili F, Bercich L, Falchetti M, Tardanico R, Baronchelli C, Lucini L, Santin A, Murer B. Claudin 4 identifies
M. Lotti et al. a wide spectrum of epithelial neoplasms and represents a very useful marker for carcinoma versus mesothelioma diagnosis in pleural and peritoneal biopsies and effusions. Virchows Arch 2007; 451:669–680. 159. Banks DE, Shi R, McLarty J, Cowl CT, Smith D, Tarlo SM, Daroowalla F, Balmes J, Baumann M. American College of Chest Physicians consensus statement on the respiratory health effects of asbestos. Results of a Delphi study. Chest 2009; 135:1619–1627.
Clinical Toxicology (2010) 48, 497–508 Copyright © Informa UK, Ltd. ISSN: 1556-3650 print / 1556-9519 online DOI: 10.3109/15563650.2010.506877
REVIEW LCLT
Management of phosgene-induced acute lung injury CHRISTOPHER GRAINGE1,2 and PAUL RICE2 Management of phosgene-induced acute lung injury
1 2
Department of Military Medicine, Royal Centre for Defence Medicine, Edgbaston, Birmingham, United Kingdom Biomedical Sciences Department, Defence Science and Technology Laboratory, Salisbury, Wiltshire, UK
Context. Phosgene is a substance of immense importance in the chemical industry. Because of its widespread industrial use, there is potential for small-scale exposures within the workplace, large-scale accidental release, or even deliberate release into a built-up area. Objective. This review aims to examine all published studies concerning potential treatments for phosgene-induced acute lung injury and incorporate them into up-todate clinical guidance. In addition, it aims to contrast the approaches when dealing with small numbers of patients known to be exposed (possibly with dose information) with the presentation of a large and heterogeneous population of casualties following a significant industrial accident or deliberate release; no published guidelines have specifically addressed this second problem. Methods. PubMed and Embase were searched for all available years till April 2010 and 584 papers were identified and considered. Experimental studies. Because of the nature of the injury, there have been no human trials of patients exposed to phosgene. Multiple small and large animal studies have been performed to examine potential treatments of phosgene-induced acute lung injury, but many of these used isolated organ models, pretreatment regimens, or clinically improbable doses. Recent studies in large animals using both realistic time frames and dosing regimens have improved our knowledge, but clinical guidance remains based on incomplete data. Management of a small-scale, confirmed exposure. In the circumstance of a small-scale, confirmed industrial release where a few individuals are exposed and present rapidly, an intravenous bolus of high-dose corticosteroid (e.g., methylprednisolone 1 g) should be considered, although there are no experimental data to support this recommendation. The evidence is that there is no benefit from nebulized steroid even when administered 1 h after exposure, or methylprednisolone if administered intravenously ≥6 h after exposure. Consideration should also be given to administration of nebulized acetylcysteine 1–2 g, though there is no substantive evidence of benefit outside a small animal, isolated lung model and there is a possibility of adverse effects. If the oxygen saturation falls below 94%, patients should receive the lowest concentration of supplemental oxygen to maintain their SaO2 in the normal range. Once patients require oxygen, nebulized β-agonists [e.g., salbutamol (albuterol) 5 mg by nebulizer every 4 h] may reduce lung inflammation if administered within 1 h of exposure. Elective intubation should be considered early using an ARDSnet protective ventilation strategy. Management of a large-scale, non-confirmed exposure. In the circumstances of a large-scale industrial or urban release, not all patients presenting will have been exposed and health services are likely to be highly stretched. In this situation, patients should not be treated immediately as there is no evidence that delaying therapy causes harm, rather they should be rested and observed with regular physical examination and measurement of peripheral oxygen saturations. Once a patient’s oxygen saturation falls below 94%, treatment with the lowest concentration of oxygen required to maintain their oxygen saturations in the normal range should be started. Once oxygen has been started, nebulized β-agonists [e.g., salbutamol (albuterol) 5 mg by nebulizer every 4 h] may reduce lung inflammation if administered within 1 h of exposure, though delayed administration which is likely following a large-scale release has not been tested formally. There is no benefit from nebulized steroid even when administered 1 h after exposure, or high-dose corticosteroid if administered intravenously ≥6 h after exposure. Although there are no experimental data to support this recommendation, an intravenous bolus of high-dose corticosteroid (e.g., methylprednisolone 1 g) may be considered if presentation is <6 h and resources allow. Depending on the numbers of casualties presenting, invasive ventilation should be initiated either electively once symptoms present (especially where there is a short latent period, indicating likelihood of more significant injury), or delayed until required. Ventilation should be with high positive end expiratory pressure, ARDSnet recommended ventilation. Conclusions. The mechanisms underlying the phosgene-induced acute lung injury are not well understood. Future experimental work should ensure that potential treatments are tested in a large animal model using realistic dosing regimens and clinically relevant timings, such as those that might be found in a mass casualty situation. Keywords
Phosgene; Animals; Humans; Inhalation exposure; Lung diseases
Introduction The treatment of phosgene exposure has been difficult since its first widespread use in World War I.1,2 Since then, many Received 1 June 2010; accepted 4 July 2010. Address correspondence to Paul Rice, Biomedical Sciences Department, Defence Science and Technology Laboratory, Dstl Porton Down, Salisbury, Wiltshire SP4 0JQ, UK. E-mail:
[email protected]
studies have been performed with the aim of improving the management of casualties from phosgene inhalation; as a result of these studies, clinical guidance has been issued previously.3–5 Recently, several important studies have been published which have yet to be incorporated into clinical advice.6–10 In addition, previous guidance has generally focused on the immediate treatment of individuals or small numbers of people definitely exposed to phosgene in an industrial environment. There is currently no guidance on the
Clinical Toxicology vol. 48 no. 6 2010
498 management of mass casualties following a large, and possibly deliberate, release in an urban situation where inhaled doses are likely to be unknown and where there will be a large medical and logistic burden from the worried well. This review aims to examine all published studies concerning potential treatments for phosgene-induced acute lung injury and incorporate them into up-to-date clinical guidance with sufficient detail for attending physicians to weigh up potential risks and benefits from different treatment options. In addition, it aims to contrast the approaches when dealing with small numbers of patients known to be exposed (possibly with dose information) with the presentation of a large and heterogeneous population of casualties following a significant industrial accident or deliberate release; no published guidelines have specifically addressed this second problem.
C. Grainge and P. Rice and in polymers and polycarbonates. Current estimates indicate that phosgene is produced in quantities of approximately 5 million metric tonnes per year, with the vast majority being produced and consumed on site.15 Although phosgene is now recognized as an economically important industrial chemical precursor, its first widespread use was as a chemical weapon in World War I.16 The gas was used extensively during this campaign, either alone or in combination with chlorine gas; total casualties from gas attacks in this war are estimated at 1.2 million people, with over 90,000 deaths.2 Because of its widespread industrial use, there is potential for small-scale exposures within the workplace,17 large-scale accidental release such as by a boiling liquid, expanding vapor explosion,18 or even deliberate release into a built-up area.19 Because of the pathophysiology of phosgene exposure, such exposures present difficulties in both diagnosis and management.20
Methods The electronic databases PubMed and Embase were searched for all available years to April 2010. The principal search term, “phosgene” was used for each database; no other search terms or subheadings were used. A total of 584 papers were identified and all were examined manually for controlled trials of therapeutic interventions; only controlled animal or human trials of phosgene therapy (18 studies) are formally reviewed here. Additional data from submitted but as yet unpublished studies performed in our institution have also been examined.6
Background Phosgene was first synthesized by John Davy in 1812 from chlorine provided by his more famous brother, the chemist Sir Humphrey Davy.11 John Davy combined chlorine with carbon monoxide in the presence of sunlight and described inhaling the resultant chemical: Its odour was different from that of chlorine, something like that which one might imagine would result from the smell of chlorine combined with that of ammonia, yet more intolerable and suffocating than chlorine itself, and affecting the eyes in a peculiar manner, producing a rapid flow of tears and occasioning painful sensations.12
Phosgene (COCl2), CAS number 75-44-5, also called carbonyl chloride or carbonic acid dichloride, is a colorless nonflammable gas at normal temperature and pressure, and when pressurized or refrigerated, forms a colorless liquid with boiling temperature of 8.2°C.13 Phosgene is present in minute concentrations as an atmospheric pollutant, its concentration increasing in urban areas compared to rural settings.14 Phosgene is a substance of immense importance in the chemical industry, primarily being used to produce organic solvents; it is used in the production of many common compounds including dyes, agro-chemicals, synthetic foams, resins, pharmaceuticals
Pathophysiology Chemically, phosgene reacts with human tissue in at least two ways: hydrolysis and acylation. When combined with water, phosgene produces hydrogen chloride and carbon dioxide, although as the gas is poorly soluble in water, only small amounts of hydrochloric acid are produced under normal physiological conditions. It is thought that this is only relevant in causing mucus membrane and eye symptoms when phosgene is present at relatively high concentrations.21 Acylation is probably the more important reaction, leading to lung damage characteristic of exposure. Acylation is the chemical reaction whereby phosgene loses its reactive carbon and oxygen atoms to nucleophilic biological components such as hydroxyl, thiol, amine, and sulphydryl groups present in molecules such as proteins, carbohydrates, and lipids.15 These reactions result in wide-ranging (though incompletely understood) modifications to important cellular machinery. The reactions include direct damage to lung surfactant and peroxidation of lipids, including membrane phospholipids.22–26 This damage then leads to the downstream release of arachidonic acid mediators such as the leukotrienes and the upregulation of oxidative response enzymes.27,28 The chemical changes induced by phosgene inhalation have been shown both in isolated lung preparations, in animal models and in human cases to cause increases in vascular permeability, alveolar leakage, and pulmonary edema.29–32 In animal models, isolation of the lungs from any potential changes in the cardiac system demonstrates that the pulmonary edema is noncardiogenic due to the increased permeability of the alveolar membrane.33
Clinical features The clinical picture resulting from phosgene exposure has been fully explored in previous reviews, most notably by
Clinical Toxicology vol. 48 no. 6 2010
Management of phosgene-induced acute lung injury Borak and Diller in 2001.3 Briefly, exposure to phosgene gas and its resultant symptoms may be divided into immediate and delayed onset symptoms.
499 these cases, the patients’ eyes became sensitive to light, with excessive lacrimation, which has not been reported previously.17
Small-scale versus large-scale releases Immediate symptoms Immediate symptoms are determined by the concentration of phosgene gas to which an individual is exposed and are mostly because of irritation of the mucus membranes via the hydrolysis of phosgene into hydrochloric acid in surface water.15 A recent case series of three patients in China following an industrial accident reported that the first symptom of exposure was eye irritation, which was adequately treated with eye washes and irrigation.17 The severity of immediate symptoms is related to vapor concentration, rather than the exposure dose. A concentration of >3 parts per million (ppm) produces throat irritation, at >4 ppm eye irritation, and at >4.8 ppm cough and chest tightness.3,34 The odor threshold for phosgene is significantly below these concentrations at around 0.4–1.5 ppm, though not all individuals are able to smell phosgene and its odor can be easily masked by other scents in the environment. Exposure concentration is less important in causing toxicity than the total dose to which an individual is exposed. This dose varies with minute ventilation, but can be approximated using the formula of C × t, where C is concentration (in milligrams per cubic meter) and t is time (in minutes). Highconcentration but short-duration exposures may result in initial symptoms but no long-term effects, whereas lowconcentration but longer exposures may have no immediate symptoms or even apparent smell but still result in fatal outcomes.3 It is fundamental to appreciate that initial symptoms, including the presence or absence of smell, give no indication of prognosis following potential phosgene exposure. In the absence of an indicator badge or sophisticated monitoring equipment at the time of phosgene release, there is no method of determining who has been exposed until the end of any latent period.20
Delayed symptoms Depending on the inhaled dose (rather than the exposure concentration) there may be a symptom-free period of up to 48 h following acute exposure. The duration of this period is inversely proportional to the exposure dose and, if the time of exposure is known, is indicative of the severity of injury. Higher exposure doses, therefore, result in shorter symptomfree periods, though they may still not induce any immediate symptoms; even high-exposure doses may be essentially asymptomatic initially. Following any symptom-free period, symptoms include cough, dyspnea, tachypnea, and respiratory distress caused by pulmonary edema. In the case series referred to above, the three patients had symptom-free periods of 8, 12, and 14 h with onset of symptoms of cough, tight chest, and shortness of breath at these times. Additionally, in
In the circumstances of a small contained industrial release, there is likely to be an indication of the exposure concentration from which an exposure dose can be estimated by multiplying the concentration by the exposure time. In these circumstances, there are likely to be relatively few casualties who present immediately following exposure and they will be accommodated readily within local healthcare systems. Following a large-scale industrial or deliberate release, there is less likelihood of an accurate assessment of inhaled concentration or dose and larger numbers of casualties are likely. In addition to patients who present late following occult exposure, there will probably be large numbers of persons who attend for immediate emergency care that have not been significantly exposed. Following the Tokyo subway sarin attack in 1995, approximately 5,000 people presented to hospital with the number of true casualties around a fifth of these.19 Sarin exposure is relatively easy to diagnose: following a phosgene release, large numbers of people would have to be monitored for at least 24 h. In these circumstances, local healthcare provision will be highly stressed and possibly overwhelmed. There is a need for treatment guidance for each of the two distinct scenarios above: the first where exposure is rapidly identified and patients present quickly in small numbers to medical care; the second where exposure is less certain, and a heterogeneous population of patients present both early and late in large numbers. Both scenarios would ideally have treatment guidance based on current evidence; for phosgene exposure, treatment studies have only been performed in animals.
Experimental studies Many studies have been performed examining potential treatments for phosgene-induced acute lung injury and these are summarized in Table 1. These studies have been performed by several groups using various models. Broadly, the trials can be divided into those performed on small animals (such as mice, rats, guinea pigs, and rabbits) and those performed on large animals (such as dogs and pigs). Small animals allow investigation of underlying biochemical and pathophysiological pathways involved in phosgene-induced lung injury; however, the role of small animal data in respiratory investigation, especially modeling acute lung injury, is debated.47,48 There are known species-specific responses to phosgene exposure, even between small mammals.49 In addition, there are challenges of animal genetic diversity,50,51 lung size, and ultrastructure,52 and differences in mediator production including interleukins and nitric oxide.47 Important markers of injury such as shunt fraction and the arterial
500 Table 1. Experimental studies conducted since 1989 Route
Preexposure Postexposure
Animal model
Treatment
Year
Furosemide
2010 Neb
×
Pig
Oxygen
2010 Inh
×
Pig
×
Pig
Neb
×
Pig
Salbutamol (albuterol)
2009 Neb
×
Pig
Protective ventilation
2007
×
Pig
Colchicine
2005 ip
×
Rat
n-Propyl gallate (nPG) Vitamin E
2001 Diet supplementation Diet supplementation 2000 ip and Perfusate
×
Mice
×
Mice
Corticosteroids
Eicosatetraynoic acid
2009 iv
×
Guinea pig (isolated lungs)
N
Primary outcome
16 Survival to 24 h 30 Survival to 24 h
13 Survival to 24 h 11 Survival to 24 h 12 Survival to 24 h
19 Survival to 24 h
Result
Secondary outcome
No change
LWW:BW BAL inflammation Improved Lung survival histopathology with delayed LWW:BW low-dose oxygen therapy No change BAL inflammation LWW:BW No change BAL inflammation LWW:BW No change Hemodynamic measures, LWW:BW BAL inflammation Improved Oxygenation Shunt fraction LWW:BW Lung histopathology Improved Neutrophil influx BAL protein
120 Airway hyperreactivity 80 Survival to Improved with 24 h 0.75% nPG 80 Survival to No change 24 h 34 LWW:DW Improved
LWW:DW GSH levels LWW:DW GSH levels GSH levels
Result
Reference
No change
6
Improved
7
Improved
No change
9
No change Worsened
Improved Improved Improved Improved Improved Improved Improved Improved Improved Improved Improved Improved
8
10
35
36 36 30, 37
Butylated hydroxyanisole Isoprenaline (isoproterenol)
Aminophylline
1999 Dietary supplementation 1998 it and Perfusate it only
1996 ip ip ip
Dibutyryl cAMP
1996 it or Perfusate
Acetylcysteine
1995 it
Colchicine
2010 ip 1991
Rabbit (isolated lungs)
120 Survival to 24 h 59 Lung weight gain
Rabbit (isolated lungs)
45 Lung weight gain
Improved
× × ×
Rat
45 LWW:BW 29 LWW:BW 24 LWW:BW
×
Rabbit (isolated lungs)
Lung weight gain
×
Rabbit (isolated lungs)
29 Lung weight gain
× ×
Rat Rat/Mouse
50 LWW:DW 320 Survival
Improved No change Improved, no better than ibuprofen alone Improved (it > Pulmonary Artery Perfusate) Pressure Perfusate Leukotrienes Improved Pulmonary Artery Pressure Perfusate Leukotrienes Improved Lipid peroxidation Improved Lung Neutrophil (Colchicine Influx only)
×
Rabbit (isolated lungs)
30 Lung leak index
×
× × ×
× × ×
1990 iv or ip
1989 Dibutyryl cAMP Aminophylline Isoprenaline (isoproterenol)/ terbutaline
Mice ×
1997 Perfusate
Ibuprofen Pentoxifylline Ibuprofen + Pentoxifylline
Cyclophosphamide AA861 Ibuprofen
×
×
×
Improved
LWW:BW
Improved
38
Improved in both Rx groups
Pulmonary Artery Pressure Perfusate Leukotrienes Pulmonary Artery Pressure Perfusate Leukotrienes
Improved
39
Improved
Malondialdehyde level
Improved Improved Improved 41
Improved
42
Improved Improved Improved
43
Improved Improved
44 45
Improved
46
60 Rabbit (isolated lungs)
40
33 Lung weight gain
Improved
Pulmonary Artery Pressure
No change
Abbreviations: ip, intraperitoneal; iv, intravenous; it, intratracheal; neb, nebulized; inh, inhaled; LWW:BW, lung wet weight to body weight ratio; LWW:DW, lung wet weight to dry weight ratio; BAL, bronchoalveolar lavage.
501
Clinical Toxicology vol. 48 no. 6 2010
502 partial pressure of oxygen (PaO2) are difficult to measure in small animals. All of these factors significantly compromise the direct extrapolation of results from small animals to humans. Several groups have investigated preexposure prophylaxis to ameliorate the severity of injury with subsequent phosgene exposure. This approach is useful in the understanding of pathophysiology and to suggest directions for future work on postexposure treatment, but it is not directly applicable to clinical guidance following phosgene exposure in humans. Many studies have been performed using isolated, ex vivo lung preparations where animals are exposed to phosgene gas, the animal killed, and the lungs ventilated and perfused artificially. This model enables direct measurement of lung weight gain (a measure of pulmonary edema formation) and cytokines in the perfusate. In addition, treatments can be given into the perfusate to mimic intravenous medication or intratracheally to mimic inhaled therapy. The direct translation of results from these small animal, isolated lung preparations to human treatments is not without risk; however, in the study of phosgene-induced lung injury it is often the only available evidence. Using these models, treatment can be initiated either before exposure or very rapidly after. Again the relevance of these data to clinically realistic time delays, especially in a mass casualty situation, makes firm conclusions difficult. Clinicians are primarily interested in patient survival, once this is certain, morbidity can be considered. Many small animal studies do not consider survival as an end point but rather use lung weight gain, lung wet weight (LWW) to body weight (BW) ratio (LWW:BW), or lung wet weight to dry weight (DW) ratio (LWW:DW) as makers of the severity of pulmonary edema. Multiple regression analysis in large animal models suggests that LWW:BW is related to 24 h survival, so this may be an appropriate surrogate marker.7 In small animals, it is extremely difficult to measure intrapulmonary shunt fraction, which has also been shown by logistic regression to be related to survival.7 The best animal studies for treatment options following phosgene exposure would involve large numbers of large animals followed over significant time periods examining survival as a primary end point, with markers of morbidity in surviving animals. No such studies have been performed because of resource and ethical considerations. Bearing in mind the preceding discussion, the following paragraphs summarize the salient data that have been obtained from a variety of models following phosgene exposure.
Biochemical pathways affected by phosgene and approaches to management Antioxidant pathway At least part of the pathological effect of phosgene exposure appears to be mediated by oxidant damage to the respiratory
C. Grainge and P. Rice epithelium.27,28 Therapy with antioxidants has been investigated extensively. Pretreatments have included n-propyl gallate (nPG), vitamin E, and butylated hydroxyanisole (BHA) as dietary supplementation in mice for 23 days prior to phosgene exposure.38,36 Standard laboratory mouse diets were supplemented with nPG at 0.75 and 1.5% (w/w), vitamin E at 1 and 2% (w/w), and BHA at 0.75 and 1.5% (w/w). Animals were then exposed to phosgene resulting in 23% survival in untreated animals at 24 h. Survival was increased at 24 h by the BHA 0.75% diet [survival 55% (p < 0.05)], BHA 1.5% diet [survival 92% (p < 0.0001)], and nPG at low [0.75% diet; survival 55% (p < 0.05)], but not at high dose (1.5% diet; survival 25%). Vitamin E supplementation did not improve survival. Although dietary supplementation with antioxidants is not suitable as a treatment of exposure in humans, these data show that increasing antioxidant defenses is of benefit prior to exposure. Acetylcysteine Acetylcysteine, an antioxidant routinely administered intravenously following paracetamol (acetaminophen) overdose53 and trialed as an inhaled treatment in cystic fibrosis,54 has been investigated for potential benefit following phosgene exposure.43,44 In the first study, acetylcysteine was administered via intratracheal bolus in a rabbit isolated lung preparation at a dose of 40 mg/kg approximately 50 min after phosgene exposure in five animals. Outcome measures were lung weight gain, which was significantly reduced by acetylcysteine treatment, as were perfusate leukotrienes and lipid peroxidation; these parameters were measured up to 150 min following exposure.43 The dose used of 40 mg/kg equates to a dose of 3,200 mg in an 80 kg human; the maximum dose licensed by the manufacturer of Mucomyst (an inhaled preparation of acetylcysteine) is 10 mL of a 20% solution (containing 2,000 mg) which can be delivered at a minimum interval of 2 h. Although bronchoconstriction has been reported with nebulized acetylcysteine, it was an uncommon finding in a large series of patients with cystic fibrosis.54 However, it should be considered that phosgene injury induces airway hyperresponsiveness and this may increase the likelihood of a bronchoconstrictive response with acetylcysteine.35 As a result of this single, ex vivo study, treatment with nebulized acetylcysteine has been noted in some guidance despite no survival data, large animal modeling, or confirmatory studies.3,4,13 Acetylcysteine in these studies was administered approximately 50 min postexposure. Although this might be achievable clinically as a first-aid measure in a small-scale industrial scenario, it would be impossible to implement following a largescale release resulting in mass casualties. The most recent animal study investigated intraperitoneal administration of acetylcysteine in rats immediately following phosgene exposure.44 Although this study was primarily designed to investigate the mechanisms underlying the
Clinical Toxicology vol. 48 no. 6 2010
Management of phosgene-induced acute lung injury actions of acetylcysteine following phosgene exposure, and showed that glutathione reconstitution was fundamental to its therapeutic action, it additionally demonstrated an improvement in lung wet weight to dry weight ratio in the treatment group. The dose of acetylcysteine used varied from 50 to 200 mg/kg (equivalent to 4,000–16,000 mg in humans) in comparison to a standard human dose of 600–1,200 mg given intravenously in the prevention of contrast-induced nephropathy.55 Non-steroidal anti-inflammatory drugs Non-steroidal anti-inflammatory drugs inhibit cyclo-oxygenase pathways, but following phosgene exposure this is unlikely to be relevant as there is no increase in cyclo-oxygenase metabolites.46,41 They also act to chelate iron and possibly scavenge reactive hydroxyl metabolites which could be of benefit following phosgene exposure. They have been tested in both an isolated perfused rabbit lung model46 and in a rat model.41 In the isolated perfused rabbit lung model,46 animals were pretreated with 25 mg/kg of intravenous ibuprofen for 30 min prior to phosgene exposure. Once the lungs were perfused, the perfusate was supplemented with 50 mg/L of ibuprofen (n = 5 in the treatment and two control groups). Additionally, in another group designed to determine if postexposure treatment was beneficial, rabbits were exposed then given ibuprofen 12.5 mg/kg intravenously at 10 min, 2 h, and 4 h. The lungs were removed and perfused at this point (n = 5 in the treatment group). Both pre- and posttreated animals showed a significant reduction in lung weight gain secondary to a reduction in alveolar membrane permeability. Combining the ibuprofen with iron prior to administration prevented the beneficial effects; this suggests that it is the iron-chelating effects that prevent the ongoing lipid peroxidation.46 Using the rat model, ibuprofen was administered by intraperitoneal injection 30 min before and 60 min after phosgene exposure. LWW:BW was improved 5 h postexposure at both low and high doses of ibuprofen.41 In the isolated rabbit lung model, the dose of ibuprofen used in the pretreatment experiment is equivalent to 2,000 mg in an 80 kg human, and in the posttreatment, 3,000 mg given in divided doses over 4 h. In the rat model, the lowest effective dose was equivalent to 9,200 mg preexposure and 4,650 mg postexposure. Pretreatment will not be possible with an accidental phosgene release; however, oral administration of ibuprofen would be possible and is likely to result in similar drug availability as that obtained intravenously.56 These doses are very substantially higher than would be given to humans to treat other conditions and could impact on renal function and cause gastrointestinal bleeding. In addition, treatment would have to start very soon after exposure. This might be possible in a small-scale release but not in a largescale accident that resulted in mass casualties. An earlier review3 points out that these studies have been performed but does not specifically recommend the use of ibuprofen following phosgene exposure.
503 Eicosatetraynoic acid Eicosatetraynoic acid, an arachidonic acid analog with antioxidant properties, has been examined using a guinea pig isolated lung model as a postexposure therapy. Eicosatetraynoic acid administered 5 min after phosgene exposure by intraperitoneal injection and by supplementing the lung perfusate for 180 min following exposure significantly reduced the LWW:DW ratio compared to controls. It was demonstrated that eicosatetraynoic acid had reduced lipid peroxidation and maintained concentrations of glutathione in the treated animals, suggesting that eicosatetraynoic acid had acted by an antioxidant effect.37 Furosemide In light of these findings that postexposure treatment of phosgene-induced lung injury with antioxidants might be beneficial, the effects of nebulized furosemide were investigated in a large animal model at realistic doses and a clinically meaningful time frame. Furosemide, as well as acting at the Na+/K+/Cl– cotransporter systems in the ascending limb of the loop of Henle when used as a diuretic, also has multiple actions in the lungs including the prevention of bronchoconstriction,57,58 inhibition of mast cell degranulation,59 and decreasing mucosal permeability.60 Furosemide has also been shown to act as a dose-dependent antioxidant.6,61,62 Using a well-established pig model with survival to 24 h as the primary outcome measure, furosemide was administered at a dose of 40 mg by nebulizer 1, 3, 5, 7, 9, 12, 16, and 20 h postexposure to phosgene. There were eight animals in each of the treatment and control groups. There was no improvement in survival and a worsening of PaO2:FiO2 ratio between 19 and 24 h after exposure.6 Treatment of phosgene-induced acute lung injury with antioxidants appears to be a worthwhile aim, though all positive data have only come from experiments on small animals. Other mechanisms of injury are also likely to play an important role following phosgene exposure. Cyclic 3,5-adenosine monophosphate pathway Phosgene exposure causes a decrease in lung tissue cyclic 3,5-adenosine monophosphate (cAMP); increasing cAMP may offer a therapeutic option for phosgene-induced lung injury.23 Many drugs upregulate intracellular cAMP concentrations including β-agonists (e.g., isoprenaline, terbutaline, and salbutamol) and phosphodiesterase inhibitors such as aminophylline. Direct supplementation of cAMP concentrations using dibutyryl cyclic adenosine monophosphate (DBcAMP) is also possible. This pathway has been extensively investigated using both small and large animal models. Initially DBcAMP, isoprenaline (isoproterenol), terbutaline, and aminophylline were investigated in a rabbit isolated lung model.33 Using five animals in each treatment group, each treatment was given 30 min prior to phosgene exposure, and then continued in the lung perfusate
Clinical Toxicology vol. 48 no. 6 2010
504 for a further 60 min. Each treatment improved lung weight gain. A follow-up study, examining intratracheal administration of DBcAMP given 60–80 min postexposure also reduced lung weight gain, so raising the possibility that postexposure treatment could be beneficial, at least in an isolated lung model.42
C. Grainge and P. Rice extrapolation from small animal models to large animal models and hence human therapy cannot be made. This has important ramifications for the development of clinical guidelines for treatment of chemically induced lung injury. Anti-inflammatory pathway
Aminophylline As dibutyryl cAMP is not used clinically, the phosphodiesterase inhibitor aminophylline was investigated for its potential effects following phosgene injury in the same rabbit and isolated lung models.40 Aminophylline was administered into the lung perfusate 80–90 min following exposure and this significantly reduced lung weight gain between 130 and 150 min postexposure. Aminophylline would be difficult to administer to large numbers of casualties intravenously soon after exposure but could be considered in a small-scale confirmed exposure.
Following any initial phosgene-mediated injury, there is an influx of inflammatory cells into the lungs which may further worsen the injury.45,64
Isoprenaline Intratracheal administration of isoprenaline was examined.39 An intratracheal bolus of isoprenaline given at 8 μg/kg 50–60 min after phosgene in the isolated rabbit lung model caused a significant reduction in lung weight gain up to 150 min following exposure (n = 6 treated animals). In addition, there were improvements in tracheal and pulmonary artery pressures and perfusate levels of leukotrienes.
Colchicine Colchicine decreases neutrophil recruitment by a variety of mechanisms;65 its effects on preventing neutrophil migration following phosgene exposure were investigated using both mouse and rat models. When administered intraperitoneally in rats at 1 mg/kg 1 h prior to phosgene exposure, colchicine significantly decreased airway hyperreactivity, lung lavage protein, and lung neutrophils.35 Administered to mice 30 min after phosgene exposure colchicine 1 mg/kg increased survival from 20 to 43% as well as improving other markers of injury severity.45 Although these data suggest that colchicine may be potentially helpful if administered soon after phosgene exposure, the recommended human adult dose is only 1 mg (maximum 6 mg over 3 days) because of its significant systemic toxicity. It is unknown whether such small doses of colchicine would have any therapeutic benefit.
Salbutamol (albuterol) In light of the demonstrated benefits of mediators of the cAMP pathway, including intratracheal isoprenaline, and in view of other trials showing benefit of salbutamol (albuterol) in human lung injury,63 we examined the effects of repeated nebulized salbutamol following phosgene exposure in a large animal model using clinically feasible timings and dosing regimens. Salbutamol was administered at 2.5 mg per dose (equivalent to 4 mg/dose in 80 kg human) 1, 5, 9, 13, 17, and 21 h following phosgene exposure. There was no improvement in survival with salbutamol treatment; indeed there was a worsening of measured physiological parameters such as arterial oxygenation and shunt fraction. There was, however, a significant reduction in lung inflammation with the percentage of neutrophils in bronchoalveolar lavage samples falling from 24% in controls to 12% in treated animals.8 The dissociation between small and large animal models may reflect the animal model chosen, the drug used (isoprenaline is a β1 and β2 agonist, whereas salbutamol is selective for β2 receptors), or the dosing regimen employed. In the large animal model used above, clinically realistic time frames and doses were used, suggesting that either very rapid administration or supranormal doses will be required when treating patients if any benefit is to be obtained. It is possible that there is no benefit to be had from these drugs following phosgene injury. If nothing else, these data suggest that direct
Corticosteroids Many guidelines include the suggestion that corticosteroids be administered as soon as possible after injury,13 whereas others state that its role is unproven.4,5 To address this question, a pig model was used to examine the effect of treatment with inhaled or intravenous corticosteroids delivered at a normal therapeutic dose in a clinically realistic time frame following phosgene inhalation. The study was designed to address potential treatments following a release of phosgene affecting a large number of individuals with unknown exposures. Hence, the steroids were given either intravenously (12.5 mg/kg methylprednisolone, equivalent human dose 1,000 mg) once symptoms would have developed (6 h after exposure) or prior to symptoms developing by nebulizer (1 mg budesonide) starting 1 h after exposure with doses repeated at 6, 12, and 18 h postexposure. There was no difference in survival to 24 h or in any secondary outcome measures in either treatment group.9 It was concluded on the basis of this study that corticosteroids administered at normally effective doses in a realistic time frame following phosgene injury are of no benefit. These data, however, do not preclude potential benefit of early (<6 h) or higher dose intravenous steroids or very early (<1 h) or higher dose inhaled steroids. The lack of any significant improvement in secondary outcome measures including markers of inflammation, lung weight, or pulmonary shunt
Clinical Toxicology vol. 48 no. 6 2010
Management of phosgene-induced acute lung injury fraction would suggest that if steroids were to be effective they would have to be delivered considerably earlier and at significantly higher doses than administered in this study.
Supportive treatments Protective ventilation In 2000, the National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome (ARDS) Network (ARDSnet) demonstrated that a protective ventilation (PV) strategy improved mortality by 22% compared with traditional mechanical ventilation in a stage 3 multi-national, multi-center trial of some 300 patients with acute lung injury/ARDS.66 The ARDSnet study incorporated low tidal volume ventilation with increased positive end expiratory pressure (PEEP) and FiO2; in some patients with acute lung injury, increasing FiO2 may not maintain arterial oxygenation because of the amount of shunt occurring and there is a risk of oxygen toxicity further exacerbating the lung injury.67 PEEP improves arterial oxygenation in pulmonary edema by the recruitment of alveoli through a process of redistributing lung water and by opening and stabilizing atelectatic alveoli.10 Recent work in our laboratory has examined the effects of the ARDSnet-recommended ventilation strategy on survival to 24 h in a large pig model exposed to a dose of phosgene causing 70% mortality in the control group.10 At 6 h following exposure (to mimic the time at which clinical signs are likely to be present), pigs were divided into three groups as follows. Group 1 was conventionally ventilated: FiO2, 0.24; tidal volume (TV) 10 mL/kg; PEEP, 3 cm of H2O; frequency (f), 20 breaths/min (n = 10). Group 2 was subjected to protective ventilation (PVA): FiO2, 0.4; TV, 8 mL/kg; PEEP, 8 cm of H2O; f, 20 breaths/min (n = 5). Group 3 was also protectively ventilated (PVB): FiO2, 0.4; TV, 6 mL/kg; PEEP, 8 cm of H2O; f, 25 breaths/min (n = 4). These parameters were set and remained unchanged until the end of the 24 h experiment. In the conventionally ventilated group, only 3 of 10 animals survived to 24 h; all animals from both PV treated groups survived to the end of the 24 h experiment (p < 0.05). Not only was survival significantly increased, but PV resulted in an improvement in LWW:BW ratio compared to controls (controls 17.5 ± 1.1; PVA 13.1 ± 1.3; PVB 10.4 ± 0.6). This improvement was significant (p < 0.05) in the PVB group. There was also a significant improvement in lung histopathology in both PV groups at the end of the 24 h study period. In the instance of a small-scale phosgene release, it would be possible to electively intubate and ventilate the exposed individuals with ARDSnet-recommended ventilator settings, which our data show would improve both survival and the underlying lung injury, at least to the 24 h time point. We do not have data for maintaining ventilation beyond this time but we feel it is unlikely that the injury would subsequently
505 worsen. Appropriate weaning strategies have not been investigated but routine weaning for ARDS would seem appropriate. In the case of a large-scale industrial leakage, or a deliberate phosgene release into a built-up area, it is unlikely that the local healthcare facilities would be able to electively ventilate all those who have been significantly exposed, though in some countries, planning is in place for such an eventuality.68,69 Oxygen therapy Oxygen is widely advocated as a supportive therapy in the early treatment of phosgene-induced lung injury and has been since World War I.1,4,70 However, because increasing oxygenation of the tissues above normal levels can result in the production of harmful reactive oxygen species,71 oxygen therapy, especially in the early asymptomatic phase of phosgene poisoning, could be potentially harmful and some guidance reflects this uncertainty.72 To investigate whether immediate high-dose oxygen therapy could be harmful following phosgene exposure and, in addition, to explore whether delayed oxygen therapy improved outcome, a large animal model was treated with various oxygen therapy regimens following phosgene exposure.7 The experimental protocol involved a phosgene-exposed, untreated control group (n = 10); a group treated immediately following exposure with 80% O2 (n = 5); a third group that had 80% O2 treatment but only from 6 h following exposure to reflect the asymptomatic phase of the injury (n = 5). Two additional groups were treated with 40% oxygen to reflect the possibility of limited oxygen supplies following a mass casualty situation. These groups were treated from 6 or 12 h following exposure (n = 5 in each). All animals were monitored for up to 24 h following phosgene exposure with survival to this time the primary outcome measure. Mortality at 24 h was decreased from 70% in the control group to 20% (ns) in the immediate 80% oxygen treatment group; mortality was decreased to zero in all delayed oxygen therapy groups (p < 0.05). In addition, there was an improvement in LWW:BW ratio and lung histology scores in all oxygen treated groups. There were no significant differences in these data between oxygen treatment groups, though low-dose delayed oxygen showed a nonsignificant improvement in histology compared to immediate high-dose oxygen therapy (p = 0.052). Logistic regression modeling identified two factors, increased LWW:BW ratio and percentage increase in intrapulmonary shunt, that were associated with a reduced probability of survival.7 Our findings7 suggest that there is a threshold concentration of oxygen required to improve survival which is at or below the 40% administered in the low-oxygen therapy group. Oxygen also reduces the severity of the underlying lung injury, even when significantly delayed following exposure. Our data7 also suggest that the administration of oxygen can be safely delayed until the delayed signs and symptoms of phosgene inhalation become apparent. Immediate oxygen therapy was not demonstrably inferior
Clinical Toxicology vol. 48 no. 6 2010
506 to delayed oxygen therapy, though the data did show a tendency toward that conclusion. The pragmatic approach knowing that delayed oxygen therapy is at least not inferior to immediate therapy is to delay therapy until oxygen saturation falls so saving potentially scarce resources and removing any risk that immediate oxygen therapy will worsen the injury. This is in keeping with current oxygen administration guidelines.73
Treatment recommendations Small-scale confirmed exposure In cases where there is evidence of definite exposure, the following clinical management strategy is proposed, though there are no experimental data currently to support some of these interventions: 1. Patients should be rested and observed (repeated physical examination and SaO2) and a chest X-ray should be performed some 12–24 h after exposure, or earlier if clinically indicated, and repeated as appropriate. 2. The evidence is that there is no benefit from nebulized steroid even when administered 1 h after exposure or highdose corticosteroid if administered intravenously ≥6 h after exposure. Although there are no experimental data to support this recommendation, an intravenous bolus of high-dose corticosteroid (e.g., methylprednisolone 1 g) may be considered if presentation is <6 h. 3. Nebulized acetylcysteine 1–2 g (5–10 mL of 20% solution) may be considered, although there is no substantive evidence of benefit outside a small animal, isolated lung model and there is a possibility of adverse effects. 4. If the SaO2 falls below 94%, patients should receive the lowest concentration of supplemental oxygen to maintain their SaO2 in the normal range. 5. Once patients require oxygen, nebulized β-agonists [e.g., salbutamol (albuterol) 5 mg by nebulizer every 4 h] may reduce lung inflammation if administered within 1 h of exposure, though delayed administration has not been tested formally. 6. Consider elective intubation early using the ARDSnet PV strategy as it lessens injury severity and significantly improves survival. Large-scale nonconfirmed exposure A major problem in a mass release of phosgene will be the prevalence of persons who are concerned they have been exposed, when they have not. There are good logistical reasons for recommending that patients should not be treated until there is firm evidence of significant phosgene exposure such as chest X-ray changes, crepitations (rales) on respiratory examination, or a fall in arterial oxygen saturation. In a mass casualty situation, it is likely that the numbers of pre-
C. Grainge and P. Rice senting patients will preclude the use of repeated chest X-rays; repeated physical examinations combined with regular monitoring of peripheral oxygen saturations should be used as the screening tools. There is no evidence that delaying any therapy causes harm. The experimental data provide evidence for the following clinical management strategy: 1. Patients should be rested and observed (repeated physical examination and peripheral SaO2). 2. The evidence is that there is no benefit from nebulized steroid even when administered 1 h after exposure or highdose corticosteroid if administered intravenously ≥6 h after exposure. Although there are no experimental data to support this recommendation, an intravenous bolus of high-dose corticosteroid (e.g., methylprednisolone 1 g) may be considered if presentation is <6 h and resources allow. 3. If the SaO2 falls below 94%, patients should receive the lowest concentration of supplemental oxygen to maintain their SaO2 in the normal range. 4. Once patients require oxygen, nebulized β-agonists [e.g., salbutamol (albuterol) 5 mg by nebulizer every 4 h] may reduce lung inflammation if administered within 1 h of exposure, though delayed administration has not been tested formally. 5. Ventilation should be initiated either electively once symptoms present (especially where there is a short latent period, indicating the likelihood of more significant injury) or delayed until required. Ventilation should be with high PEEP, ARDSnet-recommended ventilation.
Conclusions The mechanisms underlying the acute lung injury both during and after phosgene exposure are not well understood, though models which should improve our understanding are in development. To develop new treatments or identify already licensed drugs that may be used in the treatment of phosgeneinduced lung injury, the mechanisms underlying the injury, including those at a cellular level, should be understood more clearly. In addition, future experimental work should ensure that even if potential treatments are identified in small animals, likely candidates are taken to a large animal model using realistic dosing regimens and clinically relevant timings, such as those that might be found in a mass casualty situation.74 Until such work is completed, the above guidance is the best that can be offered from the evidence that is available.
Declaration of interest The authors have no declaration of interest to declare. The views expressed are those of the authors and do not represent those of Dstl or the Ministry of Defence.
Clinical Toxicology vol. 48 no. 6 2010
Management of phosgene-induced acute lung injury
References 1. Grainge C. Breath of life: the evolution of oxygen therapy. J R Soc Med 2004; 97:489–493. 2. Fitzgerald GJ. Chemical warfare and medical response during World War I. Am J Public Health 2008; 98:611–625. 3. Borak J, Diller WF. Phosgene exposure: mechanisms of injury and treatment strategies. J Occup Environ Med 2001; 43:110–119. 4. Blain PG. Expert Group on the Management of Chemical Casualties Caused by Terrorist Activity. Department of Health 2003. 5. Russell D, Blain PG, Rice P. Clinical management of casualties exposed to lung damaging agents: a critical review. EMJ 2006; 23:421–424. 6. Grainge C, Smith AJ, Jugg B, Jenner J, Rice P. Furosemide therapy in the treatment of phosgene induced acute lung injury. J R Army Med Corps. In Press. 7. Grainge C, Jugg BJ, Smith AJ, Brown RFR, Jenner J, Parkhouse DA, Rice P. Delayed low-dose supplemental oxygen improves survival following phosgene-induced acute lung injury. Inhal Toxicol 2010; 22:552–560. 8. Grainge C, Brown R, Jugg BJ, Smith AJ, Mann TM, Jenner J, Rice P, Parkhouse DA. Early treatment with nebulised salbutamol worsens physiological measures and does not improve survival following phosgene induced acute lung injury. J R Army Med Corps 2009; 155:105–109. 9. Smith A, Brown R, Jugg B, Platt J, Mann T, Masey C, Jenner J, Rice P. The effect of steroid treatment with inhaled budesonide or intravenous methylprednisolone on phosgene-induced acute lung injury in a porcine model. Mil Med 2009; 174:1287–1294. 10. Parkhouse DA, Brown RF, Jugg BJ, Harban FM, Platt J, Kenward CE, Jenner J, Rice P, Smith AJ. Protective ventilation strategies in the management of phosgene-induced acute lung injury. Mil Med 2007; 172:295–300. 11. Davy H. On a combination of oxymuriatic gas and oxygen gas. Philos Trans R Soc 1811; 101:155–162. 12. Davy J. On a gaseous compound of carbonic oxide and chlorine. Philos Trans R Soc 1812; 102:144–151. 13. Registry AfTSaD. Phosgene - Agency for Toxic Substances and Disease Registry guidelines. 2007. 14. Gift JS, McGaughy R, Singh DV, Sonawane B. Health assessment of phosgene: approaches for derivation of reference concentration. Regul Toxicol Pharmacol 2008; 51:98–107. 15. Pauluhn J, Carson A, Costa DL, Gordon T, Kodavanti U, Last JA, Matthay MA, Pinkerton KE, Sciuto AM. Workshop summary: phosgene-induced pulmonary toxicity revisited: appraisal of early and late markers of pulmonary injury from animal models with emphasis on human significance. Inhal Toxicol 2007; 19:789–810. 16. Smart J. History of chemical and biological warfare: an American perspective. In: Sidell FR, Takafuji ET, Franz DR, eds. Medical Aspects of Chemical and Biological Warfare. Washington, DC: Borden Institute, Walter Reed Army Medical Center; 2000:9–86. 17. Shi J. Three patients with acute phosgene poisoning. Zhonghua Lao Dong Wei Sheng Zhi Ye Bing Za Zhi 2006; 24:254. 18. Abbasi T, Abbasi SA. The boiling liquid expanding vapour explosion (BLEVE): mechanism, consequence assessment, management. J Hazard Mater 2007; 141:489–519. 19. Woodall J. Tokyo subway gas attack. Lancet 1997; 350:296. 20. Black RM. An overview of biological markers of exposure to chemical warfare agents. J Anal Toxicol 2008; 32:2–9. 21. Diller WF. Pathogenesis of phosgene poisoning. Toxicol Ind Health 1985; 1:7–15. 22. Jugg B, Jenner J, Rice P. The effect of perfluoroisobutene and phosgene on rat lavage fluid surfactant phospholipids. Hum Exp Toxicol 1999; 18:659–668. 23. Guo YL, Kennedy TP, Michael JR, Sciuto AM, Ghio AJ, Adkinson NF, Gurtner GH. Mechanism of phosgene-induced lung toxicity: role of arachidonate mediators. J Appl Physiol 1990; 69:1615–1622.
507 24. Sciuto AM, Stotts RR, Chittenden V, Choung E, Heflin MD. Changes in absorbance at 413 nm in plasma from three rodent species exposed to phosgene. Biochem Biophys Res Commun 1996; 226:906–911. 25. Duniho SM, Martin J, Forster JS, Cascio MB, Moran TS, Carpin LB, Sciuto AM. Acute changes in lung histopathology and bronchoalveolar lavage parameters in mice exposed to the choking agent gas phosgene. Toxicol Pathol 2002; 30:339–349. 26. De Curtis V, Gemma S, Sbraccia M, Testai E, Vittozzi L. The contribution of electrophilic and radicalic intermediates to phospholipid adducts formed by halomethanes in vivo. J Biochem Toxicol 1994; 9:305–310. 27. Sciuto AM, Cascio MB, Moran TS, Forster JS. The fate of antioxidant enzymes in bronchoalveolar lavage fluid over 7 days in mice with acute lung injury. Inhal Toxicol 2003; 15:675–685. 28. Sciuto AM, Phillips CS, Orzolek LD, Hege AI, Moran TS, Dillman JF. Genomic analysis of murine pulmonary tissue following carbonyl chloride inhalation. Chem Res Toxicol 2005; 18:1654–1660. 29. Frosolono MF, Currie WD. Response of the pulmonary surfactant system to phosgene. Toxicol Ind Health 1985; 1:29–35. 30. Sciuto AM, Stotts RR. Posttreatment with eicosatetraynoic acid decreases lung edema in guinea pigs exposed to phosgene: the role of leukotrienes. Exp Lung Res 1998; 24:273–292. 31. Sciuto AM, Moran TS, Narula A, Forster JS. Disruption of gas exchange in mice after exposure to the chemical threat agent phosgene. Mil Med 2001; 166:809–814. 32. Brown RF, Jugg BJ, Harban FM, Ashley Z, Kenward CE, Platt J, Hill A, Rice P, Watkins PE. Pathophysiological responses following phosgene exposure in the anaesthetized pig. J Appl Toxicol 2002; 22:263–269. 33. Kennedy TP, Michael JR, Hoidal JR, Hasty D, Sciuto AM, Hopkins C, Lazar R, Bysani GK, Tolly E, Gurtner GH. Dibutyryl cAMP, aminophylline, and beta-adrenergic agonists protect against pulmonary edema caused by phosgene. J Appl Physiol 1989; 67:2542–2552. 34. Glass D, McClanahan M, Koller L, Adeshina F. Provisional advisory levels (PALs) for phosgene (CG). Inhal Toxicol 2009; 21(Suppl 3):73–94. 35. Ghio A, Lehmann J, Winsett D, Richards J, Costa D. Colchicine decreases airway hyperreactivity after phosgene exposure. Inhal Toxicol 2005; 17:277–285. 36. Sciuto AM, Moran TS. Effect of dietary treatment with n-propyl gallate or vitamin E on the survival of mice exposed to phosgene. J Appl Toxicol 2001; 21:33–39. 37. Sciuto AM. Posttreatment with ETYA protects against phosgeneinduced lung injury by amplifying the glutathione to lipid peroxidation ratio. Inhal Toxicol 2000; 12:347–356. 38. Sciuto AM, Moran TS. BHA diet enhances the survival of mice exposed to phosgene: the effect of BHA on glutathione levels in the lung. Inhal Toxicol 1999; 11:855–871. 39. Sciuto AM, Strickland PT, Gurtner GH. Post-exposure treatment with isoproterenol attenuates pulmonary edema in phosgene-exposed rabbits. J Appl Toxicol 1998; 18:321–329. 40. Sciuto AM, Strickland PT, Kennedy TP, Gurtner GH. Postexposure treatment with aminophylline protects against phosgene-induced acute lung injury. Exp Lung Res 1997; 23:317–332. 41. Sciuto AM, Stotts RR, Hurt HH. Efficacy of ibuprofen and pentoxifylline in the treatment of phosgene-induced acute lung injury. J Appl Toxicol 1996; 16:381–384. 42. Sciuto AM, Strickland PT, Kennedy TP, Guo YL, Gurtner GH. Intratracheal administration of DBcAMP attenuates edema formation in phosgene-induced acute lung injury. J Appl Physiol 1996; 80:149–157. 43. Sciuto AM, Strickland PT, Kennedy TP, Gurtner GH. Protective effects of N-acetylcysteine treatment after phosgene exposure in rabbits. Am J Respir Crit Care Med 1995; 151:768–772. 44. Ji L, Liu R, Zhang XD, Chen HL, Bai H, Wang X, Zhao HL, Liang X, Hai CX. N-acetylcysteine attenuates phosgene-induced acute lung injury via up-regulation of Nrf2 expression. Inhal Toxicol 2010; 22:535–542.
Clinical Toxicology vol. 48 no. 6 2010
508 45. Ghio AJ, Kennedy TP, Hatch GE, Tepper JS. Reduction of neutrophil influx diminishes lung injury and mortality following phosgene inhalation. J Appl Physiol 1991; 71:657–665. 46. Kennedy TP, Rao NV, Noah W, Michael JR, Jafri MH, Gurtner GH, Hoidal JR. Ibuprofen prevents oxidant lung injury and in vitro lipid peroxidation by chelating iron. J Clin Invest 1990; 86:1565–1573. 47. Matute-Bello G, Frevert CW, Martin TR. Animal models of acute lung injury. Am J Physiol Lung Cell Mol Physiol 2008; 295:L379–L399. 48. Persson CGA. Con: mice are not a good model of human airway disease. Am J Respir Crit Care Med 2002; 166:6–7; discussion 8. 49. Sciuto AM. Assessment of early acute lung injury in rodents exposed to phosgene. Arch Toxicol 1998; 72:283–288. 50. Prows DR, Daly MJ, Shertzer HG, Leikauf GD. Ozone-induced acute lung injury: genetic analysis of F(2) mice generated from A/J and C57BL/6J strains. Am J Physiol 1999; 277:L372–L380. 51. Leikauf GD, McDowell SA, Wesselkamper SC, Hardie WD, Leikauf JE, Korfhagen TR, Prows DR. Acute lung injury: functional genomics and genetic susceptibility. Chest 2002; 121:70S–75S. 52. Pauluhn J. Acute head-only exposure of dogs to phosgene. Part III. Comparison of indicators of lung injury in dogs and rats. Inhal Toxicol 2006; 18:609–621. 53. Heard KJ. Acetylcysteine for acetaminophen poisoning. N Engl J Med 2008; 359:285–292. 54. Nash EF, Stephenson A, Ratjen F, Tullis E. Nebulized and oral thiol derivatives for pulmonary disease in cystic fibrosis. Cochrane Database of Syst Rev 2009; 1:CD007168. DOI: 10.1002/14651858.CD007168.pub2. 55. Trivedi H, Daram S, Szabo A, Bartorelli AL, Marenzi G. High-dose N-acetylcysteine for the prevention of contrast-induced nephropathy. Am J Med 2009; 874:e9–e15. 56. Martin W, Koselowske G, Töberich H, Kerkmann T, Mangold B, Augustin J. Pharmacokinetics and absolute bioavailability of ibuprofen after oral administration of ibuprofen lysine in man. Biopharm Drug Dispos 1990; 11:265–278. 57. Polosa R, Lau LC, Holgate ST. Inhibition of adenosine 5′-monophosphate- and methacholine-induced bronchoconstriction in asthma by inhaled frusemide. Eur Respir J 1990; 3:665–672. 58. O’Connor BJ, Chung KF, Chen-Worsdell YM, Fuller RW, Barnes PJ. Effect of inhaled furosemide and bumetanide on adenosine 5′-monophosphate- and sodium metabisulfite-induced bronchoconstriction in asthmatic subjects. Am Rev Respir Dis 1991; 143:1329–1333. 59. Anderson SD, He W, Temple DM. Inhibition by furosemide of inflammatory mediators from lung fragments. N Engl J Med 1991; 324:131.
C. Grainge and P. Rice 60. Inoue T, Shigeta M, Mochizuki H, Shimizu T, Morikawa A, Suzuki H, Watanabe N, Tateno M, Oriuchi N, Hirano T, Tomiyoshi K, Endo K. Effect of inhaled furosemide on lung clearance of technetium-99mDTPA. J Nucl Med 1995; 36:73–77. 61. Lahet JJ, Lenfant F, Courderot-Masuyer C, Ecarnot-Laubriet E, Vergely C, Durnet-Archeray MJ, Freysz M, Rochette L. In vivo and in vitro antioxidant properties of furosemide. Life Sci 2003; 73:1075–1082. 62. Kang MY, Tsuchiya M, Packer L, Manabe M. In vitro study on antioxidant potential of various drugs used in the perioperative period. Acta Anaesthesiol Scand 1998; 42:4–12. 63. Perkins GD, McAuley DF, Thickett DR, Gao F. The beta-agonist lung injury trial (BALTI): a randomized placebo-controlled clinical trial. Am J Respir Crit Care Med 2006; 173:281–287. 64. Ghio AJ, Hatch GE. Tolerance to phosgene is associated with a neutrophilic influx into the rat lung. Am J Respir Crit Care Med 1996; 153:1064–1071. 65. Molad Y. Update on colchicine and its mechanism of action. Curr Rheumatol Rep 2002; 4:252–256. 66. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. The Acute Respiratory Distress Syndrome Network. N Engl J Med 2000; 342:1301–1308. 67. Knight PR, Kurek C, Davidson BA, Nader ND, Patel A, Sokolowski J, Notter RH, Holm BA. Acid aspiration increases sensitivity to increased ambient oxygen concentrations. Am J Physiol Lung Cell Mol Physiol 2000; 278:L1240–L1247. 68. Malatino EM. Strategic national stockpile: overview and ventilator assets. Respir Care 2008; 53:91–5; discussion 5. 69. Wilgis J. Strategies for providing mechanical ventilation in a mass casualty incident: distribution versus stockpiling. Respir Care 2008; 53:96–100; discussion 3. 70. Diller WF, Zante R. A literature review: therapy for phosgene poisoning. Toxicol Ind Health 1985; 1:117–128. 71. Manning AM. Oxygen therapy and toxicity. Vet Clin North Am Small Anim Pract 2002; 32:1005–1020. 72. American Chemistry Council. Phosgene: information on options for first aid and medical treatment. Arlington, VA, USA: American Chemistry Council; 2006. 73. O’Driscoll BR, Howard LS, Davison AG, Society BT. BTS guideline for emergency oxygen use in adult patients. Thorax 2008; 63(Suppl 6):vi1–vi68. 74. National Institute of Health. NIH strategic plan and research agenda for medical countermeasures against chemical threats. Bethesda, MD: National Institute of Health; 2007.
Clinical Toxicology (2010) 48, 509–515 Copyright © Informa UK, Ltd. ISSN: 1556-3650 print / 1556-9519 online DOI: 10.3109/15563650.2010.496371
ARTICLE LCLT
Radiolabeling and dose fixation study of oral alpha-ketoglutarate as a cyanide antidote in healthy human volunteers GAURAV MITTAL1, THAKURI SINGH1, NEERAJ KUMAR1, ASEEM BHATNAGAR1, RAJENDRA PRASAD TRIPATHI1, RAJKUMAR TULSAWANI2, RAJAGOPALAN VIJAYARAGHAVAN3, and RAHUL BHATTACHARYA3 Dose fixation of alpha-ketoglutarate
1
Institute of Nuclear Medicine and Allied Sciences, Delhi, India Defence Institute of Physiology and Allied Sciences, Delhi, India 3 Defence Research and Development Establishment, Gwalior, India 2
Context. Radiolabeling and dose fixation study of alpha-ketoglutarate (A-KG). Objective. A-KG is a potential oral antidote for cyanide poisoning. Its protective efficacy in animals was best exhibited at a dose of 2.0 g/kg body weight, which when extrapolated to human is very high. The objective of this study was to reduce the dose of A-KG in humans with concomitant increase in its bioavailability, employing pharmacoscintigraphic techniques to assess kinetics in man. Materials and methods. A-KG was radiolabeled with technetium-99m pertechnetate (Tc-99m) and its purity, labeling efficiency, and stability in vitro were determined by instant thin layer chromatography. Time-dependent bio-absorption of the drug in rats and rabbits was assessed by gamma scintigraphy after oral administration of a tracer dose of 99mTc-A-KG mixed with nonradioactive A-KG at a concentration of 0.1–2.0 g/kg in the presence or absence of aqueous dilution. Furthermore, scintigraphy and radiometry studies were performed in healthy human volunteers using 5–20 g of A-KG, given in single or split doses followed by different quantity of water. Drug bioavailability was estimated periodically. Results. High radiolabeling (>97%) of A-KG with a stability of 24 h in vitro was obtained. Less than 1% absorption of the drug occurred within 20 min after A-KG was administered in animals at a concentration of 2.0 g/kg body weight. One-tenth reduction in dose increased the bioavailability to 15%. Significant improvement in gastric emptying of the drug was achieved when the drug was administered along with 1–5 mL of water. In humans, two doses of 10 g A-KG given at an interval of 10 min, followed by 300 mL of water, increased the drug bioavailability to 40% as compared to a single dose of 20 g. Discussion. Significant reduction in A-KG dose was achieved in humans as compared to the recommended dose in animals. Conclusion. Aqueous dilution improves the bioavailability of A-KG in humans. Keywords
Alpha-ketoglutarate; Cyanide antidote; Radiolabeling; Dose fixation
Introduction Cyanide is a highly toxic and fast-acting mitochondrial poison, which is abundantly present in the environment. Cyanide poisoning may occur due to accidental (industrial, fire smoke, iatrogenic, dietary, and so on) or deliberate (suicidal, homicidal, military, and so on) exposures.1–3 Human–cyanide interaction is very common due to its wide industrial applications. Surreptitious use of cyanide for various unlawful activities is also a potential threat to human.4 The prognosis of the victims exposed to cyanide largely depends on the termination of further exposure, supportive care, and institution of immediate and aggressive specific treatment. There are several antidotes available for the treatReceived 27 March 2010; accepted 24 May 2010. Address correspondence to Rahul Bhattacharya, Division of Experimental Therapeutics, Defence Research and Development Establishment, Jhansi Road, Gwalior 474 002, India. E-mail: rbhattacharya41 @rediffmail.com
ment of cyanide poisoning including sodium nitrite (SN) and sodium thiosulfate (STS).5,6 Cyanide antagonism by nitrites is mediated through the induction of methemoglobin that binds free cyanide to form cyanmethemoglobin. Furthermore, STS, a sulfane–sulfur donor facilitates the enzymatic conversion of cyanide into nontoxic thiocyanate.6 Although, the combination of SN and STS is very effective, it has certain disadvantages as well.7–10 Cobalt-containing compounds such as dicobalt edetate and hydroxocobalamin are other cyanide antidotes in Europe and United States.5,10– 12 Dicobalt edetate chelates cyanide as cobalticyanide, whereas hydroxocobalamin (vitamin B12a) interacts with cyanide to form nontoxic cyanocobalamin (vitamin B12). These drugs are also not free from certain side effects.7,11 More effective cyanide antidotes are needed.3 Many such compounds like ketocarboxylic acids namely, sodium pyruvate and alpha-ketoglutarate (A-KG), carbonyl compounds and their metabolites, and nutrients have been shown to afford significant protection against experimental cyanide poisoning.13–16 Cyanide detoxification by these compounds has
Clinical Toxicology vol. 48 no. 6 2010
510 been attributed to the interaction of cyanide with carbonyl moieties like aldehydes and ketone to yield cyanohydrin intermediates.17,18 Extensive research efforts have been expended to develop A-KG as an oral treatment for cyanide poisoning.19,20 Most of the cyanide antidotes are administered intravenously but the utility of oral A-KG is envisaged as a treatment for personnel engaged in evacuation or decontamination operations like the fire fighters and chronic occupational or dietary exposures.21 In oral cyanide poisoning, the gastric acid environment favors the formation of the unionized form of hydrogen cyanide, which facilitates rapid absorption.22 The alkaline environment provided by oral A-KG minimizes the absorption of hydrogen cyanide across the gastrointestinal mucosa.21 Oral administration of A-KG confers maximum protection when cyanide poisoning occurs through the same route, but oral A-KG also affords protection against cyanide given by other routes as well.23 Simultaneous treatment and pretreatment of 2.0 g/kg A-KG (oral) alone provided five- to sixfold protection against acute cyanide poisoning in rodents. This protection was further potentiated by adjunction of STS.19,20 However, extrapolation of this dose of A-KG for a 70-kg man would work out to be quite enormous (∼140 g). As such, A-KG is quite safe for human use as its oral LD50 in rodents is >5.0 g/kg body weight.24 In addition, A-KG is sold in United States as an over-the-counter nutritional supplement to improve various metabolic disorders, an use which does not warrant FDA approval.25 However, to exploit A-KG at a relatively higher dose as an oral antidote for cyanide, regulatory clearance is required. Therefore, to increase the safety margin of A-KG, we were interested to reduce its dose in humans with concurrent improvement in its bioavailability. This study addresses the dose fixation of oral A-KG in animals and healthy human volunteers employing pharmacoscintigraphic techniques. This technique is widely applied for the evaluation and development of drugs and their formulations.26,27 It is considered as the gold standard for imaging and quantifying the in vivo distribution and bio-absorption pattern of various formulations.28–30 To carry out the dose fixation studies employing pharmacoscintigraphic techniques, the drug has to be radiolabeled to a suitable radiotracer. We also report here a method for radiolabeling A-KG with technetium-99m (Tc-99m) and demonstrate its utility in drug development to study the kinetics of a potential pre-exposure cyanide antidote.
Material and methods A-KG disodium and stannous chloride dihydrate (SnCl2·2H2O) were procured from Sigma Chemical Company, St. Louis, MO, USA. All other chemicals and reagents (Analytical Reagent Grade) were purchased from Merck Ltd. (Mumbai, India). Technetium-99m was obtained from BRIT, BARC, Mumbai, India.
G. Mittal et al. Radiolabeling A-KG with Tc-99m In a sterile vial, A-KG was dissolved in distilled water, followed by the addition of sodium bicarbonate and gentisic acid. The contents were dissolved by gentle shaking. SnCl2·2H2O solution (2.0 mg/mL in 0.1 N HCl) was then added to the mixture and pH was adjusted to 7.0. After an incubation with 74 Mega Becquerel (MBq), Tc-99m pertechnetate for a period of 20 min at room temperature (approx. 25°C), the radiolabeling efficiency was checked by instant thin layer chromatography (ITLC) with the help of a gamma counter (Capintec Inc., NJ, USA).31
Radiochemical purity of radiolabeled drug The radiochemical purity and labeling efficiency of 99mTc-AKG was assessed by ascending ITLC using silica gel-coated fiber glass sheets (ITLC-SG, Gelman sciences Inc., Ann Arbor, MI, USA) as stationary phase and solvent systems, namely, 100% acetone and a solvent mixture of pyridine, glacial acetic acid, and water (3 : 5 : 1.5, v/v) as mobile phases. The radioactive contaminants were identified as reduced/hydrolyzed (R/H) Tc-99m and free Tc-99m pertechnetate.
Stability studies of radiolabeled drug Stability of 99mTc-A-KG was carried out by mixing normal human serum with 99mTc-A-KG and incubating the contents at 37°C. Aliquots were withdrawn at different time intervals until 24 h and subjected to ascending ITLC as described earlier. Any increase in free pertechnetate was considered to be due to the degradation of the radiolabel from the drug.
Dose fixation studies of A-KG using gamma scintigraphy In animals All the animal experiments conducted were approved by Institute’s Animal Ethical Committee. Male albino rats (2–3 months; 200–250 g) and New Zealand white rabbits (1.75–2 years; 2–2.25 kg) were selected for the study. Gamma scintigraphy was performed at different time intervals after oral administration of a tracer dose of 99mTc-A-KG mixed with nonradioactive A-KG (2.0 g/kg body weight). All the images were recorded on a computer system assisted with the inbuilt Entegra Version-2 software and percentage absorption of A-KG in the body from gastrointestinal tract (GIT) mucosa was estimated. In follow-up experiments, amount of A-KG was reduced to 0.1, 0.2, 0.5, and 1.0 g/kg body weight, respectively, and in addition, animals were made to drink specified quantity of water (1–5 mL) to study the effect on bio-absorption of the drug. In healthy human volunteers Human pharmacokinetic and toxicity–efficacy study was conducted in healthy human volunteers after obtaining approval from Institutional Human Ethical Committee (INM/
Clinical Toxicology vol. 48 no. 6 2010
Dose fixation of alpha-ketoglutarate
Assessments performed at start and during the study Demographic data, age, and weight were recorded with respect to each human volunteer in the case report form at the time of enrolment. History of any disease, blood pressure, and heart rate were noted. Routine laboratory investigations for hematology, biochemistry, and ECG were done to make sure the volunteer was healthy. Informed consent form was obtained from the volunteers in vernacular language or English. Based on the results of animal experiments using scintigraphy and radiometry, 5–20 g of A-KG was chosen as the range that was expected to give maximum absorption of the drug from GIT. A total of eight volunteers were recruited for the study. Three volunteers each were given 5.0 g of A-KG mixed with tracer quantity of 99mTc-AKG with 200 mL of water. Blood was drawn periodically for drug concentration and count estimation. On the second day, the procedure was repeated with 400 mL of water instead of 200 mL. Five other volunteers were given 20 g A-KG with 200 mL of water and the drug bioavailability was estimated by periodic blood examination as described above. In another set of experiment, the same five subjects were given either 20 g A-KG with 400 mL of water (n = 2) or 10 g A-KG twice at an interval of 10 min (n = 3) with 300 mL of water each. Dynamic scintigraphy images were taken to assess any difference in bio-absorption of the drug from GIT into the blood pool. A-KG assay in plasma A-KG was estimated in human plasma at indicated time points as per the protocol discussed elsewhere.32 Briefly, A-KG was derivatized with orthophenylene diamine to form a quinoxalinol derivative, which was estimated by HPLC equipped with fluorescent detector. A reference standard of A-KG was prepared for its quantitation. Statistics Four to six subjects were used for each study and the results were expressed as mean ± SD. Unpaired t-test was applied for the calculation of significance at p < 0.05 using GraphPad Instat version 3.00 for Windows XP, GraphPad Software, San Diego, CA, USA available at www.graphpad.com
Results Radiolabeling and stability of 99mTc-A-KG Table 1 shows the optimization of strength of reducing agent for radiolabeling of A-KG. A high percent radiolabeling
efficiency of 95.87 ± 0.67 was obtained when A-KG was radiolabeled with Tc-99m at pH 7.0, using stannous chloride dihydrate as reducer, gentisic acid as stabilizer, and sodium bicarbonate as binder. Radiochemical purity of the product was evaluated by ITLC, which successfully resolved labeled product from R/H and free Tc-99m. The use of two solvent systems was found to be an accurate method for distinguishing and quantifying the relative amounts of free Tc-99m pertechnetate, R/H Tc-99m, and 99mTc-A-KG. Figure 1 shows that although there was a decline in the stability of 99mTc-A-KG in serum, as compared to the corresponding saline treatment beyond 20 h, the stability was very satisfactory in vitro even up to 24 h when the percent of labeled product was 89% as determined by ITLC. Radiolabeled A-KG was thus used to obtain important information regarding drug characterization in vitro and in vivo. Table 1. Optimization of strength of reducing agent for radiolabeling of A-KG Strength of SnCl2·2H2O as reducing Labeling Reduced/hydrolyzed agent (μg/mL) efficiency (%) colloids 25 50 100 200 400 600
69.94 ± 1.15 76.21 ± 0.83 94.44 ± 0.63 95.87 ± 0.67 82.47 ± 1.35 76.4 ± 0.88
9.82 ± 0.39 7.45 ± 0.52 5.3 ± 0.43 2.62 ± 0.40 15.32 ± 0.38 22.3 ± 1.25
Free TcO4− 20.24 ± 1.10 16.34 ± 1.25 0.26 ± 0.34 1.51 ± 0.84 2.21 ± 1.21 1.3 ± 1.24
Labeling efficiency of Tc-99m with A-KG at different concentration of reducing agent by ITLC-SG using 100% acetone and a solvent mixture of pyridine, glacial acetic acid, and water (3 : 5 : 1.5, v/v) as the mobile phases. The values are expressed as percentage of 99mTc-A-KG in the preparation. Values are mean ± SD (n = 6).
% Labeling efficiency
TS/IEC/006–017/07) duly constituted for the purpose. Adult healthy males (age 24–45 years) were chosen for the study and informed consent was obtained in all cases as per institutional rules and practice. Individual Case Report Forms were used to record clinical and pharmacokinetic data.
511
* *
Time (h)
Fig. 1. In vitro stability of 99mTc-A-KG at different time intervals by ITLC-SG using 100% acetone and a solvent mixture of pyridine, glacial acetic acid, and water (3 : 5 : 1.5, v/v) as the mobile phases. The values are expressed as percentage of 99mTc-A-KG in the preparation. Values are mean ± SD (n = 6). *Significantly different as compared to corresponding saline treatment (at p < 0.05).
Clinical Toxicology vol. 48 no. 6 2010
512
G. Mittal et al.
Dose fixation studies of A-KG using gamma scintigraphy In animals Scintigraphy was performed in experimental rats and rabbits after oral administration of a tracer dose of 99mTc-A-KG mixed with nonradioactive A-KG at a concentration of 2.0 g/kg body weight. Figure 2A shows the scintigraphy images of rat fed orally with highly concentrated radiolabeled A-KG. It was found that very little counts per milliliter blood occurred at any interval suggesting very low absorption of the drug from the GIT into the systemic circulation. The radioactivity tended to remain within the stomach region suggesting that the gastric emptying was retarded very significantly. Less than 1% absorption of the drug occurred in 20 min after normalizing for total blood as 7% of body weight and the same amount of extravascular fluid. Figure 2B depicts the scintigraphy images of the rat fed with A-KG orally with and without aqueous dilution. On reducing the A-KG dose to 200 mg/kg body weight (i.e., 10% of the prescribed dose), about
15% drug was transferred to the body in 30 min. Scintigraphy clearly showed greater than 3 times faster gastric emptying after aqueous dilution. In healthy human volunteers In all, eight subjects participated in dose fixation study, receiving the drug twice in 2 days. Irrespective of the dose given, all volunteers found the drug tasted salty and was unpalatable when taken with small amounts of water. Three subjects received 5.0 g of A-KG twice, followed by five subjects who received 20 g twice (as a single dose and in two equally divided doses, respectively). Apart from vomiting in two subjects who were given concentrated A-KG solution, and nausea in others, tolerability was high in diluted product, although mild tachycardia was observed in five subjects. There was practically no nausea feeling in any of the subjects given A-KG in two divided doses, whereas two out of the five subjects who received a single 20 g dose with 200 mL of
1 min
15 min
30 min
(A)
1 min
15 min
30 min
(B)
Fig. 2. Anterior and posterior scintigraphy images showing a rat fed orally with highly concentrated radiolabeled A-KG. Practically all radioactivity is in stomach after 30 min. (A) With 1 mL of water, 2 g/kg A-KG given orally. (B) On reducing the dose to one-tenth, that is, 200 mg/kg A-KG given orally without and with aqueous dilution (5 mL of water) to two different rats. Scintigraphy showing greater than 3 times faster gastric emptying after aqueous dilution.
Clinical Toxicology vol. 48 no. 6 2010
Dose fixation of alpha-ketoglutarate
513
Blood plasma cone of A-KG (µg/mL)
water vomited whereas the remaining three had nausea for 1–2 h. The blood levels of A-KG at different time intervals, following various treatments of oral A-KG in human volunteers, are shown in Fig. 3. Those who did not vomit showed mean AKG concentration of 12 μg/mL blood in the first 15 min, with a reducing trend thereafter. The range was 8–15 μg/mL in the 5–30 min interval. Heart rate of the volunteers rose by 30 ± 4%, while blood pressure remained normal. When 20 g of A-KG was taken with 400 mL of water, no vomiting occurred and the subjective tolerance was better. The drug concentration in blood at 5–30 min interval was 10–25 μg/mL and was significantly higher as compared to the corresponding 5 and 20 g dose taken with 200 mL of water, respectively (p < 0.05). When 10 g A-KG was given twice at an interval of 10 min, followed by excessive dilution, mean and peak concentrations were higher than a single dose of 20 g (15–36 μg/mL).
45 40 35 30 25 20 15 10 5 0
# ∗
0
10 5g
20
∗
30
∗
40 50 60 Time (min)
20 g + 200 mL water
70
80
20 g + 400 mL water
90
Furthermore, there were distinctly two peaks, the second peak being always higher than the first one and more sustained (about 125% higher, 200–300% times more prolonged). Thereafter, the slope was reducing in nature. The drug concentration in blood at 30 min interval was significantly higher as compared to the corresponding 20 g doses with 200 and 400 mL of water, respectively (p < 0.05). Overall, the area under the curve of drug concentration in blood did not appear to increase after renormalization but the shift to higher amplitude in early phase appeared to be quite effective and useful. Significant gastric retention occurred when highly concentrated A-KG was administered orally (Fig. 4). It was so significant that more than 90% liquid was still retained in the stomach after 50–60 min (normal gastric emptying t1/2 is 30–35 min). Scintigraphy using 99mTc- A-KG indicated that taking the drug with plenty of water improved the bioavailability of the drug from 4 to 8% in the first 30 min. It also caused faster gastric clearance and early appearance of liquid in intestines. Dilution and dose renormalization (two divided doses) significantly improved the bioavailability (Fig. 5). The results indicate that doubling of dose resulted in nearly 100% increase in peak bioavailability, as one ascends from 5 to 10 g dose, 70% increase from 10 to 20 g dose, and 40% increase from 20 g dose to two equally divided doses of 10 g each, suggesting that a plateau would most likely be reached at 30–40 g adult dose (even if no vomiting takes place) after which no significant rise in peak blood concentration is expected even after dose escalation.
100
Discussion 10 + 10 g
Fig. 3. Pharmacokinetic profile of oral A-KG in human volunteers at different doses and aqueous dilution. Values are mean ± SD (n = 4). *Significantly different as compared to corresponding 5 g and # significantly different as compared to corresponding 20 g + 200 mL and 20 g + 400 mL (at p < 0.05).
Several effective antidotes are available for the treatment of cyanide poisoning, but there is no consensus on their efficacy and safety.7 A-KG has long being recognized as a promising treatment for cyanide intoxication, and its protective efficacy is attributed to its interaction with cyanide to form cyanohydrin
Fig. 4. Dynamic human scintigraphy images with concentrated oral 99mTc-A-KG showing practically no radioactivity transfer to the blood, a significant retention in stomach and an insignificant transfer to intestine by 30 min. Graph on the right showing release rate of 99mTc-A-KG from stomach of the same volunteer, confirming significant gastric retention with highly concentrated A-KG.
Clinical Toxicology vol. 48 no. 6 2010
514
G. Mittal et al.
Fig. 5. Dynamic human scintigraphy images showing that when 99mTc-A-KG was given in two divided doses along with 300 mL of water each time, bio-transfer improved by 100%, stomach was practically empty in 30 min and most of the activity was in intestines. Graph on the right shows the release rate of 99mTc-A-KG from stomach of the same volunteer, confirming significantly improved gastric clearance after dose renormalization and aqueous dilution.
complex.14–20 Many studies have shown that glutamine plays a fundamental role in the metabolic adaptation to injury, and A-KG is a precursor of glutamine which in turn is a precursor of glutathione.33 This indicates a possible antioxidant property of A-KG against cyanide-induced oxidative stress.34 Further, there is a possibility that protective effect of A-KG is facilitated by its integration into the malate-aspartate shuttle operative between the cytoplasm and the mitochondrial matrix, leading to ATP generation.35 Also, A-KG is considered as a potent natural detoxifying agent in the body and is known to alleviate hyperammonemia and hyperaminoacidurias. Other than cyanide detoxification, it is also presumed to neutralize the toxic effects of nitrogenous chemicals like ammonium compounds, amines, hydrazines, and so on.25 Role of A-KG as a nutritional supplement to bolster the citric acid cycle is well known.25 This explains a wide utility of A-KG, particularly in cyanide antagonism.21 As a nutritional supplement, the therapeutic dose of A-KG ranges between 500 and 2,500 mg per day. However, for detoxification of amines and cyanide, for which no controlled human studies have been carried out so far, the dose required may be higher.25 Because of potential limitations of popular cyanide antidotes, particularly the need for parenteral doses, oral treatment of A-KG is being considered as prophylaxis for the evacuation operations in contaminated areas, occupational and chronic dietary poisoning, and as therapy for accidental, suicidal, and homicidal cyanide exposures.21 Significant protection by pretreatment of A-KG against acute cyanide poisoning was observed at a dose between 0.50 and 2.0 g/kg.19,20 At 2.0 g/kg A-KG, the distribution half-life, elimination half-life, and C max of A-KG were found to be 0.35, 0.53 h, and 36.9 μg/mL, respectively, indicating that bioavailability of A-KG given orally did not commensurate with the dose administered.36 Only a small percentage of A-KG administered lumenally to pigs appeared in the portal circulation. This was attributed to mucosal metabolism, and possibly due to limited absorption.37
Intestinal absorption of A-KG is also limited in young pigs largely due to substantial first-pass gastrointestinal metabolism.38 In view of the large dose of A-KG required for human, and its anticipated poor absorption, this study was designed to improve its bioavailability with a reduction in its dose. The purity and stability of 99mTc-A-KG prepared in this study was acceptable as per nuclear medicine principles.28–30 All the volunteers found the drug to be quite unpalatable because of its salty taste but dilution with excess of water improved the taste. The study also showed that gastric retention of concentrated A-KG was more, but its bioavailability improved when it was given in two divided doses of 10 g each followed by 300 mL of water at an interval of 10 min. Based on our previous animal studies, we expected a dose of approximately 140 g of A-KG for cyanide antagonism in humans.19–21 However, this study suggests that the dose of A-KG could be dramatically reduced to only 20 g, a sevenfold reduction. The blood A-KG level was now almost similar to what we got with 2.0 g/kg A-KG in rats.36 This dose of A-KG would be more amenable to human without compromising with its protective efficacy. This study was limited to a small number of subjects as it was basically a dose fixation study, succeeding the protection studies in animals and preceding the human clinical studies. In fact, 28 more subjects have been taken up for the Phase-I clinical trial to elucidate the pharmacokinetics and pharmacodynamics. This study provides significant information on dose fixation and bioavailability of A-KG, a promising antidote for cyanide poisoning, prior to clinical trials in human volunteers.
Declaration of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this paper.
Clinical Toxicology vol. 48 no. 6 2010
Dose fixation of alpha-ketoglutarate
References 1. Cummings TF. The treatment of cyanide poisoning. Occup Med 2004; 54:82–85. 2. Baud FJ. Cyanide: critical issues in diagnosis and treatment. Hum Exp Toxicol 2007; 26:191–201. 3. Bhattacharya R, Flora SJS. Cyanide toxicity and its treatment. In: Gupta RC, ed. Handbook of Toxicology of Chemical Warfare Agents. New York: Elsevier Inc.; 2009:255–270. 4. Rotenberg JS. Cyanide as a weapon of terror. Pediatr Ann 2003; 32:236–240. 5. Bhattacharya R. Antidotes to cyanide poisoning: present status. Indian J Pharmacol 2000; 32:94–101. 6. Baskin SL, Horowitz AM, Nealley EW. The antidotal action of sodium nitrite and sodium thiosulphate against cyanide poisoning. J Clin Pharmacol 1992; 32:368–375. 7. Van Heijst ANP, Douze JMC, Van Kesteren RG, Van Bergen J, Van Dijk A. Therapeutic problems in cyanide poisoning. Clin Toxicol 1987; 25:383–398. 8. Vick JA, Von Bredow JD. Effectiveness of intramuscularly administered cyanide antidotes on methemoglobin formation and survival. J Appl Toxicol 1998; 16:509–516. 9. Baumeister RGH, Schievelbein H, Zickgraf-Rudel G. Toxicological and clinical aspects of cyanide metabolism. Drug Res 1975; 25:1056–1063. 10. Van Heijst ANP, Meredith JJ. Antidotes for cyanide poisoning. In: Volanis GN, Sims J, Sullivan F, Turner P, eds. Basic Science in Toxicology. Brighton, UK: Taylor & Francis; 1990:558–566. 11. Hall AH, Dart R, Bogdan G. Sodim thiosulphate or hydroxocobalamin for the empiric treatment of cyanide poisoning? Ann Emerg Med 2007; 49:806–813. 12. Borron SW, Baud FJ, Megarbane B, Bismuth C. Hydroxocobalamin for severe acute cyanide poisoning by ingestion or inhalation. Am J Emerg Med 2007; 25:551–558. 13. Schwartz C, Morgan RL, Way LM, Way JL. Antagonism of cyanide intoxication with sodium pyruvate. Toxicol Appl Pharmacol 1979; 50:437–441. 14. Moore SJ, Norris JC, Ho IK, Hume AS. The efficacy of α- ketoglutaric acid in the antagonism of cyanide intoxication. Toxicol Appl Pharmacol 1986; 82:40–44. 15. Niknahad H, Khan S, Sood C, O’Brien P. Prevention of cyanideinduced cytotoxicity by nutrients in isolated rat hepatocytes. Toxicol Appl Pharmacol 1994; 128:271–279. 16. Bhattacharya R, Tulsawani RK. In vitro and in vivo evaluation of various carbonyl compounds against cyanide toxicity with particular reference to alpha-ketoglutaric acid. Drug Chem Toxicol 2008; 31:149–161. 17. Morrison RT, Boyd RN. Organic Chemistry. Boston, MA: Allyn and Bacon, Inc.; 1976:637–639. 18. Norris JC, Utley WA, Hume AS. Mechanism of antagonizing cyanide induced lethality by a-ketoglutaric acid. Toxicology 1990; 64:275–283. 19. Bhattacharya R, Vijayaraghavan R. Promising role of a-ketoglutarate in protecting the lethal effects of cyanide. Hum Exp Toxicol 2002; 21:297–303. 20. Bhattacharya R, Lakshmana Rao PV, Vijayaraghavan R. In vitro and In vivo attenuation of experimental cyanide poisoning by alpha-kotoglutarate. Toxicol Lett 2002; 128:185–195.
515 21. Bhattacharya R. a-Ketoglutarate: a promising antidote to cyanide poisoning. In: Flora SJS, Romano JA, Baskin SI, Sekhar K, eds. Pharmacological Perspectives of Toxic Chemicals and Their Antidotes. New Delhi, India: Narosa Publishing House; 2004:411–430. 22. Ballantyne B. Toxicology of cyanides. In: Ballantyne B, Marrs TC, eds. Clinical and Experimental Toxicology of Cyanides. Bristol, UK: Wright Publishers; 1987:41–126. 23. Tulsawani RK, Kumar D, Bhattacharya R. Effect of pre-treatment of a-ketoglutarate on cyanide-induced toxicity and alterations in various physiological variables in rodents. Biomed Environ Sci 2007; 20:56–63. 24. Bhattacharya R, Kumar D, Sugendran K, Pant SC, Tulsawani RK, Vijayaraghavan R. Acute toxicity studies of a-ketoglutarate: a promising antidote for cyanide poisoning. J Appl Toxicol 2001; 21:495–499. 25. Pangborn JB, Hicks JT, Chambers JR, Shahani KM, eds. A Monograph on Alpha-Ketoglutaric Acid. Lisle, IL: Bionostics, Inc.; 1990:1–3. 26. Bhavna Ahmad FJ, Mittal G, Jain GK, Malhotra G, Khar RK, Bhatnagar A. Nano-salbutamol dry powder inhalation: a new approach for treating broncho-constructive conditions. Eur J Pharm Biopharm 2009; 71:282–291. 27. Rajpal S, Mittal G, Sachdeva R, Chhillar M, Ali R, Agrawal SS, Kashyap R, Bhatnagar A. Development of atropine sulphate nasal drops and its pharmacokinetic and safety evaluation in healthy human volunteers. Environ Toxicol Pharmacol 2009; 27:206–211. 28. Davis SS, Hardy JG, Newman SP, Wilding IR. Gamma scintigraphy in the evaluation of pharmaceutical dosage forms. Eur J Nucl Med 1992; 19:971–986. 29. Wilding IR, Coupe AJ, Davis SS. The role of gamma-scintigraphy in oral drug delivery. Adv Drug Deliv Rev 2001; 46:103–124. 30. Singh AK, Bhardwaj N, Bhatnagar A. Pharmacoscintigraphy: an unexplored modality in Indian. J Pharm Sci 2004; 66:18–25. 31. Singh AK, Verma J, Bhatnagar A, Sen S, Bose M. Tc-99m Isoniazid: a specific agent for diagnosis of tuberculosis. World J Nucl Med 2003; 2:292–305. 32. Desai H. Development and method validation of analytical method for estimation of a-ketoglutarate in technical and serum by HPLC. A project report submitted by Jai Research Foundation, Valsad, India to DRDE, Gwalior. Personal communication (Study No. 4596); 2004:1–110. 33. Cynober L. The use of a-ketoglutarate salts in clinical nutrition and metabolic care. Curr Opin Clin Nutr Metab Care 1999; 2:33–37. 34. Andrae U, Singh J, Zeigler-Skylakakis K. Pyruvate and related alphaketoacids protect mammalian cells in culture against hydrogen peroxide-induced cytotoxicity. Toxicol Lett 1985; 28:93–98. 35. Varma SD, Hegde KR. The effect of a-ketoglutarate against selenite cataract formation. Exp Eye Res 2004; 79:913–918. 36. Patra D. Serum kinetics study of a-ketoglutarate in Wistar rats following single oral administration. A project report submitted by Jai Research Foundation, Valsad, India to DRDE, Gwalior. Personal communication (Study No. 1573); 2005:1–27. 37. Buddington RK, Pajor A, Buddington KK, Pierzynowski S. Absorption of alpha-ketoglutarate by the gastrointestinal tract of pigs. Comp Biochem Physiol A Mol Integr Physiol 2004; 132:215–220. 38. Lambert BD, Filip R, Stoll B, Junghans P, Derno M, Hennig U, Souffrant WB, Pierzynowski S, Burrin DG. First-pass metabolism limits the intestinal absorption of enteral alpha-ketoglutarate in young pigs. J Nutr 2006; 136:2779–2784.
Clinical Toxicology (2010) 48, 516–521 Copyright © Informa UK, Ltd. ISSN: 1556-3650 print / 1556-9519 online DOI: 10.3109/15563650.2010.490534
ARTICLE LCLT
Dapsone intoxication: clinical course and characteristics KYUNG HYE PARK1, HYUN KIM2, CHRISTOPHER CHONGSEO LEE3, KYUNG CHUL CHA2, SEUNG MIN PARK2, HO JIN JI2, HAN HO DO2, KANG HYUN LEE2, SUNG OH HWANG2, and ADAM J. SINGER3 Dapsone intoxication
1
Department of Emergency Medicine, Jeju National University Hospital, Jeju, Republic of Korea Department of Emergency Medicine, Institute of Lifelong Health, Wonju College of Medicine, Yonsei University, Wonju, Republic of Korea 3 Department of Emergency Medicine, Stony Brook University Hospital, Stony Brook, NY, USA 2
Background and objectives. Dapsone is used as an antibiotic for leprosy and for dermatological disorders and may cause methemoglobinemia. The aims of this study are to analyze the clinical characteristics of patients presenting to the emergency department (ED) with dapsone ingestion to identify risk factors associated with mortality. Methods. We conducted a retrospective observational study of adult ED patients with methemoglobinemia because of dapsone intoxication admitted to a tertiary care hospital from September 2003 to December 2008. Data collected included demographic, clinical, and laboratory characteristics, as well as survival to discharge. Characteristics of young (less than or equal to 55 years) and older (greater than age 55) patients were compared. The main outcome was inhospital mortality. Results. There were 46 patients included in the study. The minimum intoxication dose was two 100 mg tablets and the maximum was 100 tablets. Changes in mental status were more common in the older patients. Methemoglobin levels were slightly higher in the younger patients, but both groups were treated with similar doses of methylene blue. Shock and death were more common in the older patients. Conclusions. Late presentation to medical care and an altered mental status at the time of presentation were predictive of death after dapsone intoxication. Methemoglobin levels tended to be higher in those who died. Keywords Poisoning; Methemoglobinemia; Age group
Introduction Dapsone is a therapeutic agent used to treat leprosy and other dermatological disorders.1 The use of dapsone has been associated with chronic adverse effects such as neutropenia, thrombocytopenia, eosinophilic pneumonia, aplastic anemia, neuropathy, hepatitis, agranulocytosis, and acute adverse effects like nausea, vomiting, abdominal pain, methemoglobinemia, seizure, and coma.2,3 Although leprosy is relatively rare, access to dapsone is comparatively easy in the vicinity of our medical center because of the presence of a leper colony in Wonju, South Korea, in the 1960s and 1970s. As a result it is not uncommon for us to see patients with acute and chronic dapsone overdoses. It was possible to buy dapsone without prescription before August 1, 2001, when pharmaceutical dispensing was separated from medical practice in Korea. The objective of the current study was to describe a relatively large case series of patients presenting to a tertiary care
Received 9 September 2009; accepted 29 April 2010. Address correspondence to Hyun Kim, Department of Emergency Medicine, Institute of Lifelong Health, Wonju College of Medicine, Yonsei University, 162 Ilsandong, Wonju 220-701, Republic of Korea. E-mail:
[email protected]
center with acquired methemoglobinemia secondary to improper ingestion of dapsone. Specifically we compared the clinical characteristics and outcomes in the young and older patients and explored the association between several potential risk factors and mortality.
Methods The study was conducted at the Wonju Christian Hospital, a tertiary care urban hospital in Gangwon province, South Korea, between September 2003 and December 2008. A retrospective observational study design was used to describe the characteristics of patients with methemoglobinemia because of dapsone intoxication. Patients presenting to the emergency department (ED) with methemoglobinemia secondary to acute, chronic, and acute on chronic ingestion of dapsone and mixed overdose were eligible for inclusion in our study. Patients with incomplete or unclear records were excluded. Data were extracted from the electronic medical records by trained data abstracters using standardized data collection forms. Data collected included demographic, clinical, and laboratory data. These included blood gases, pulse oximetry, methemoglobin levels, liver function, and renal function test characteristics. All patients had serial arterial blood gas analysis and pulse oximetry during the course of
Clinical Toxicology vol. 48 no. 6 2010
Dapsone intoxication
517
their admission, every 8 h for the first day and every morning after that, or if needed clinically. This allowed the calculation of the difference between measured blood oxygen saturation and pulse-oximeter readings (the “oxymeter gap”). Shock was defined as a systolic blood pressure below 90 mmHg; patients in this category had features of impaired end-organ perfusion. The main outcome was in-hospital mortality.
Results Fifty-four cases of dapsone overdose induced methemoglobinemia were detected over the 5-year study period. Only 46 patients were included in the study because of incomplete data for the other 8 patients. There was a wide range in age, from 21 to 93 years. All patients were admitted through the ED. There was no difference in gender between the two study groups (young and older patients). The dose ingested varied from a minimum of two to a maximum of hundred 100-mg tablets, and the mean ingested dose was not significantly different between groups. There were no differences between groups in the cause of intoxication (intentional or accidental), period from ingestion to ED arrival, and frequency of coingestion of alcohol (Table 1). Four patients ingested dapsone with other medications or toxic materials: two patients took doxylamine, one magnesium hydroxide, and one bleaching agent of sodium hypochlorite. Patients with alcohol co-ingestion survived. The time interval between ingestion and ED presentation was greater in patients who died than in patients who survived (p = 0.003) (Table 1). More than half of all patients presented with cyanosis and shortness of breath as their chief
Data analysis Continuous data are presented as means and standard deviations and compared with the t-test. Nominal data are presented as the percent frequency of occurrence and compared with the c2-test. Patients were dichotomized into two groups based on age. Young patients included patients aged 55 and younger and older patients were those over 55 years. Comparison of the clinical and laboratory characteristics between young and older patients was performed with the t-test and the c2-test as appropriate. Multiple logistic regression was used to explore the association between potential risk factors and mortality. Data were computed with the Statistical Program for Social Science 15.0 (SPSS, Chicago, IL, USA).
Table 1. Comparison of demographic and clinical data of patients According to age
Age (years) Sex (M : F) ratio Intoxication dose (tablets, 100 mg each) Cause of intoxication (intentional : accidental) Period elapsed after ingestion (min) Initial symptoms and signs (N) Cyanosis Dyspnea Mental status changes Abdominal pain Alcohol co-ingestion (N) Irrigation (cc) Charcoal dose (g) Methemoglobin (g/dL) Oxygen saturation (%) Pulse oximetry Arterial blood gas analysis ICU admission (N) Complications (N) Pneumonia Shock Renal failure Mortality (N)
According to mortality
Total (N = 46)
>55 years
(N = 23)
≤55 years (N = 23)
(N = 37)
Mortality (N = 9)
60.91 ± 18.62 14 : 32 43.97 ± 31.78 34 : 12 543.9 ± 588.4
76.91 ± 9.37a 6 : 17 33.14 ± 30.92 14 : 9 694.5 ± 737.4
44.91 ± 9.29 8 : 15 53.44 ± 30.32 20 : 3 431.1 ± 420.9
58.54 ± 17.60 13 : 24 43.30 ± 31.24 28 : 9 407.5 ± 362.7b
70.67 ± 20.60 1:8 50.00 ± 43.60 6:3 1,203.3 ± 973.7
30 (65.2%) 25 (54.3%) 12 (26.1%) 13 (28.3%) 12 (26.1%) 4,556 ± 2,121 69.57 ± 63.49 30.89 ± 12.12
14 (60.9%) 11 (47.8%) 9 (39.1%)a 5 (21.7%) 3 (13.0%) 4,667 ± 1,506 59.09 ± 20.23 29.35 ± 9.14
16 (69.6%) 14 (60.9%) 3 (13.0%) 8 (34.8%) 9 (39.1%) 4,500 ± 2,431 79.17 ± 86.49 32.30 ± 14.39
26 (70.3%) 20 (54.1%) 7 (18.9%)b 10 (27.0%) 12 (32.4%) 4,625 ± 2,125 72.5 ± 67.81 29.93 ± 12.80
4 (44.4%) 5 (55.6%) 5 (55.6%) 3 (33.3%) 0 (0.0%) 4,000 ± 2,828 50.0 ± 0.0 35.94 ± 5.91
87.04 ± 7.12 96.58 ± 2.84 34 (73.9%)
85.47 ± 7.60 96.00 ± 3.72 18 (81.8%)
88.68 ± 6.34 97.17 ± 1.39 16 (72.7%)
88.08 ± 7.42b 97.00 ± 2.06 25 (71.4%)
82.88 ± 3.73 94.83 ± 4.68 9 (100%)
8 (17.4%) 10 (21.7%) 6 (13%) 9 (19.6%)
3 (13.6%) 8 (36.4%)a 5 (22.7%) 8 (34.8%)a
5 (21.7%) 2 (8.7%) 1 (4.5%) 1 (4.3%)
5 (13.9%) 2 (5.6%)b 1 (2.9%)b
3 (33.3%) 8 (88.9%) 5 (55.6%)
Data are represented as mean ± standard deviation unless otherwise indicated. a p < 0.05 when compared with patients ≤55 years. b p < 0.05 when compared with mortality patients.
Survival
Clinical Toxicology vol. 48 no. 6 2010
518
K.H. Park et al.
complaints. Mental status changes on ED presentation were more frequent in older patients (p = 0.044) and in those who did not survive (p = 0.025) (Table 1). The initial oxygen saturation on pulse oximetry was higher in patients who survived than in patients who died (p < 0.05) (Table 1). The base excess started to recover in survivors but failed to recover in those who died (Fig. 1). The base excess at admission was similar between the two groups, but in patients who died it was significantly lower than those in the survivors at days 2 (−7.61 ± 9.78 vs. −2.47 ± 2.75, p = 0.023) and 8 (−3.65 ± 7.47 vs. 3.02 ± 3.06, p = 0.043) (Fig. 1). The methemoglobin levels of the patients who died were higher than those in the survivors at day 9 (9.48 ± 8.58 vs. 3.64 ± 3.91, p = 0.45) (Fig. 2). From the seventh day, the hemoglobin levels of the survivors were higher than those who died (9.74 ± 1.14 g/dL vs. 8.15 ± 1.52 g/dL, p = 0.009). The initial ED management was similar in most patients including the use of gastric lavage and activated charcoal, and there was no effect of treatment on survival (Table 1). Methylene blue was administrated to 45 patients based on their symptoms, signs, and methemoglobin levels. It was normally given intravenously twice a day. The only patient who did not receive methylene blue died as soon as dapsone-induced
(mmol/L) 4 2
Surival Mortality
0 –2
0.25 0.5 0.75 1
2
3
4
5
6
7
8
9
–4 –6 –8 Time from presentation (days)
Fig. 1. Comparison of base excess between survival patients and mortality patients according to hospitalization day (*p < 0.05).
(%) 40 35 Surival Mortality
30 25 20 15 10 5 0 0.25 0.5 0.75 1
2
3
4
5
6
7
8
9
Time from presentation (days)
Fig. 2. Comparison of methemoglobin levels between survival patients and mortality patients according to hospitalization day (*p < 0.05).
methemoglobinemia was diagnosed. The total dosage of given methylene blue and the average daily dosage were not significantly different between the two age groups. The development of complications, especially shock, was more common in the older patients. The mortality was higher in the older than the younger patients (p = 0.009) (Table 1). Most deaths were due to multiple organ failure (Table 2).
Discussion Methemoglobinemia is caused by a genetic defect in hemoglobin structure, red cell metabolism, or the ingestion of a variety of drugs and toxins. Dapsone-induced methemoglobinemia occurs in accidental or suicidal poisoning4,5 or in the setting of treatment.6 After oral administration the drug is slowly absorbed, with the maximum plasma concentration reached at about 4 h, with an absorption half-life of about 1.1 h. The elimination half-life of dapsone is about 30 h. As enterohepatic circulation occurs, the elimination half-life of dapsone is decreased by administration of oral activated charcoal.1 Few studies have reported on patients with high dose ingestions of dapsone or have analyzed factors affecting poor prognosis. Carrazza and coworkers divided patients into four age groups: under age 5, ages 5–12, ages 13–18, and ages 19–50.7 Patients aged 50 and older were not included in this study. In our study, patients older than age 55 more often presented with mental confusion, developed shock, and died, even though they took smaller doses of dapsone than patients aged 55 and younger. Furthermore, the patients who did not survive presented later after ingestions were more confused and were more likely to develop shock and renal failure. Most deaths were secondary to multiple organ dysfunction. As the older patients often presented with mental status changes, history taking was often impossible delaying the diagnosis. As a result locally we recommend that methemoglobin levels should be considered in patients over age 55 with cyanosis, dyspnea, or mental status changes with unknown cause. Rachel and colleagues have also suggested that physicians and respiratory therapists be trained to look for the characteristic chocolate-brown color of blood containing methemoglobin and order a co-oximetry test for such patients.8 Although methemoglobinemia is a potentially treatable disorder, it is often fatal without adequate early treatment. Several factors affect the severity of dapsone intoxication. They include dapsone dose and serum concentration, methemoglobin level, and clinical manifestations. It has been thought that the methemoglobin level is the most important parameter to judge the severity of dapsone intoxication. Although the methemoglobin level is useful to confirm the diagnosis, it is not as important as the patient’s clinical status in determining early treatment.9 According to others, the clinical manifestations of dapsone overdose are directly related to the concentration of
519
21/F
62/F
68/M
71/F
77/F
80/F
83/F 83/F
91/F
No.
1
2
3
4
5
6
7 8
9
Unknown
Unknown Unknown
30
Unknown
20
100
Unknown
Unknown
Dose (tablets)
Mental change
Mental change Cyanosis, mental change, abdominal pain
Cyanosis, dyspnea
Mental change
Cyanosis, dyspnea, abdominal pain Dyspnea, abdominal pain Cyanosis, dyspnea, mental change Dyspnea
Chief complaints
Shock, renal failure
Shock, renal failure Pneumonia, shock Pneumonia, shock, renal failure Pneumonia, shock, renal failure None Shock
HD25
Shock, renal failure Shock
HD12
HD2 HD7
HD16
HD7
HD2
HD16
HD8
Date of death
Complications
87
84 87
86
80
80
77
85
79
SpO2 (%)
Initial
12.1
6.8 11.4
11.9
4.0
13.6
21.3
8.5
18.0
Saturation gap
MetHb: methemoglobin, HD: hospitalization day, MODS: multiple organ dysfunction syndrome, ARF: acute renal failure.
Age (years)/sex
Table 2. Analysis of the characteristics of the patients who died
25.90
50.50 29.90
37.30
36.90
38.50
41.90
40.80
41.20
Initial MetHb (%)
0.96
3.00 1.79
2.22
6.38
2.67
10.14
4.70
4.49
Initial lactate (mmol/L)
340.0
0.0 677.0
600.0
460.0
100.0
450.0
634.2
987.7
Methylene blue dose (mg/day)
MODS
MODS Metabolic acidosis, hemolytic anemia
MODS
MODS
MODS
ARF
MODS
MODS
Cause of death
Clinical Toxicology vol. 48 no. 6 2010
520 methemoglobin and fatality reported when more than about 30% of hemoglobin is converted to methemoglobin.10 The ingestion dose does not appear to be the most important factor in determining the severity of the dapsone intoxication in our series. Two patients taking two tablets of dapsone were included, whose levels of methemoglobin were over 30%. Although two tablets (200 mg) may be a therapeutic dose (1–2 mg/kg/day) for leprosy, they took dapsone without the advice and supervision of a doctor. They were 62 and 88 years old, and their initial presentations were with headache and cyanosis with dyspnea. In the present study, four cases of chronic intake and two cases of acute on chronic ingestion were included. Their presentation was similar to those with an acute ingestion. This is probably because of the fact that the typical signs and symptoms are related to the level of methemoglobinemia.2 Altered mental status is generally present in previously healthy patients with a normal hemoglobin and a methemoglobin concentration exceeding 50%.2 In the present study, mental status changes were present in older patients even with methemoglobin levels less than 30%. It is possible that the depression in mental status noted in the older patients was in part because of cerebral hypoperfusion. In the current study the presence of a depressed mental status and late presentation to medical care were predictors of death. Carrazza et al. found a significant correlation between methemoglobinemia and the time elapsed after dapsone ingestion.7 In the current study, the time elapsed from ingestion to ED arrival was notably longer in patients who subsequently died. Thirtyfour patients (73%) in our study took dapsone intentionally in an attempt to commit suicide. In the study by Carrazza et al. these patients tended to have higher methemoglobin levels and dapsone plasma concentrations.7 In our study, the doses of methylene blue were similar in those patients who died and those who survived, in both the young and older groups. This is likely because of the fact that the administered dose of methylene blue was determined by a combination of oxygen saturation, levels of methemoglobin, and clinical manifestations. Although the methemoglobin levels and oxygen saturation gaps were almost always higher in the young group of patients, the clinical manifestations in the older patients were generally more severe. Most cases developed hemolysis following methemoglobinemia. We tested for reticulocyte counts, elevated creatine kinase and lactate dehydrogenase, elevated bilirubin, and Heinz body presence. Oxidative stress due to methemoglobin formation presence of methylene blue is likely to cause hemolysis.11 Although oxygen saturation gap has been suggested as a means to judge the response to treatment, our findings did not confirm this. As methemoglobin and methylene blue have similar light absorption wavelengths an oximeter cannot be used to measure methemoglobin levels after beginning therapy with methylene blue.12,13 At presentation, in our study, pulse oximetry saturation was not correlated with methemoglobin level. We were unable to show a relationship between the gap between arterial oxygen measured by blood gas analyzer and
K.H. Park et al. pulse oximetry and methemoglobin level. This finding emphasizes the fact that pulse oximetry should not be used routinely in patients with methemoglobinemia. Pulse-oximeter readings are typically 80–85% in patients with significant methemoglobinemia. Several limitations of this study should be considered. First, our study was retrospective in nature with all the limitations of this study design. Second, we did not measure dapsone levels in our study, which may be an important determinant of both therapy and prognosis.
Conclusions Dapsone is the most important cause of acquired methemoglobinemia in South Korea. We found that late presentation to medical care and an altered mental status at the time of presentation were predictive of death after dapsone intoxication. In contrast the degree of methemoglobinemia was less reliable. Physicians should be aware of the seriousness of dapsone overdose because it may result in significant morbidity and mortality.
Acknowledgement This work was supported by a grant from KHIDI (Korea Health Industry Development Institute).
Declaration of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this paper.
References 1. Zuidema J, Modderman H, Merkus FWHM. Clinical pharmacokinetics of dapsone. Clin Pharmacokinet 1986; 11:299–315. 2. Coleman MD. Dapsone-mediated agranulocytosis: risks, possible mechanisms and prevention. Toxicology 2001; 16:53–60. 3. Flomenbaum NE, Goldfrank LR, Hoffman RS, Howland MA, Lewin NA, Nelson LS. Goldfrank’s Toxicologic Emergencies. 8th ed. New York: McGraw-Hill; 2006:883. 4. Shadnia S, Rahimi M, Moeinsadat M, Vesal G, Donyavi M, Abdollahi M. Acute methemoglobinemia following attempted suicide by dapsone. Arch Med Res 2006; 37:410–414. 5. Thunga G, Sam KG, Patel D, Khera K, Sheshadhri S, Bahuleyan S, Vansalan R, Pandit VR, Manohar C. Effectiveness of hemodialysis in acute dapsone overdose – a case report. Am J Emerg Med 2008; 26:1070.e1–1070.e4. 6. O’Dwyer D, McElvaney NG. A case of dapsone induced methemoglobinemia. Ir J Med Sci 2008; 177:273–275. 7. Carrazza MZ, Carrazza FR, Oqa S. Clinical and laboratory parameters in dapsone acute intoxication. Rev Saude Publica 2000; 34:396–401. 8. Rachel AB, Robert W, Scott M. Acquired methemoglobinemia: a retrospective series of 138 cases at 2 teaching hospitals. Medicine 2004; 83:265–273.
Clinical Toxicology vol. 48 no. 6 2010
Dapsone intoxication 9. Viccellio P. Emergency Toxicology. 2nd ed. Philadelphia: LippincottRaven; 1998:595–598. 10. Prasad R, Das BP, Singh R, Sharma KK. Dapsone induced methemoglobinemia, sulfhemoglobinemia and hemolytic anemia: a case report with a note on treatment strategies. Indian J Pharmacol 2002; 34:283–285.
521 11. McGoldrick MD, Baillie GR. Severe accidental dapsone overdose. Am J Emerg Med 1995; 13:414–415. 12. Kessler MR, Eide T, Humayun B, Poppers PJ. Spurious pulse oximeter desaturation with methylene blue injection. Anesthesiology 1986; 65:435–436. 13. Rieder HU, Frei J, Zbinden AM, Thomson DA. Pulse oximetry in methemoglobinemia. Anaesthesia 1989; 44:326–327.
Clinical Toxicology (2010) 48, 522–527 Copyright © Informa UK, Ltd. ISSN: 1556-3650 print / 1556-9519 online DOI: 10.3109/15563650.2010.491483
ARTICLE LCLT
Does wearing CBRN-PPE adversely affect the ability for clinicians to accurately, safely, and speedily draw up drugs? NICK CASTLE1, JAMES BOWEN2, and NEIL SPENCER3 Does CBRN-PPE adversely affect the draw up drugs
1 Emergency Department, Frimley Park Hospital, Camberley, UK; Department of EMC&R, Durban University of Technology, South Africa 2 Department of Paramedic Science, University of Hertfordshire, Hatfield, UK; EMS, Hamad Medical Corporation, Qatar 3 Business School, University of Hertfordshire, Hatfield, UK
Objective. Following a Chemical, Biological, Radiation, or Nuclear (CBRN) incident, the attending rescuers will be required to administer drugs while wearing the CBRN Personal Protective Equipment (CBRN-PPE). Little is known regarding the impact of the CBRN-PPE on the ability to speedily, safely, and accurately draw up drugs for subsequent administration. Design. A randomized control trial examining the ability of rescuers to draw up drugs from four different drug presentations (a Minijet prefilled syringe, an Aurum prefilled syringe, and glass and plastic ampoules with needle and syringe) was undertaken with participants wearing and not wearing the National Health Service (NHS) CBRN-PPE. Each participant underwent the procedure once in “normal clothes” and twice while wearing the CBRN-PPE. The speed and accuracy of each participant were measured as well as the participant’s perceived risk of suit damage associated with the four different drug presentations. Participants. The participants constituted a non-homogenous group of emergency nurses, nurse lecturers, paramedic lecturers, and student paramedics. A total of 81 participants were recruited; one participant withdrew due to claustrophobia. All participants had used all four drug preparations prior to being enrolled into the study. Results. Wearing the CBRN-PPE had a negative effect on drawing up drugs in the first attempt while wearing the CBRN-PPE, typically taking 63.8% longer time than without (95% confidence interval: 55.3–72.9%). Improvements were noted on second attempts. The choice of drug presentation had an effect on the time taken to draw up the drugs, with Aurum being overall the fastest and glass ampoules the slowest (p-values < 0.001). All participants rated the prefilled syringes as the easiest and the safest to use, and the glass ampoules with needles and syringes as the most difficult to use and the ones most likely to puncture the CBRN-PPE. During the drawing-up process, varying amounts of “drug” were lost, although in twothirds of the attempts all 10 mL was drawn up. The lowest volume was lost from the prefilled syringes and the maximum volume of fluid was lost from the glass ampoules. Conclusions. The NHS CBRN-PPE has a negative effect on the drawing up of drugs especially from glass ampoules. Glass ampoules represent a poor choice of drug preparation when considering speed, safety, and accuracy of drawing up of drugs while wearing protective clothing. Keywords CBRN; CBRN-PPE; Antidotes and drugs
Introduction The risk of a Chemical, Biological, Radiation, or Nuclear (CBRN) incident remains ever present. Regardless of whether it is due to a terrorist incident,1 an industrial accident,2,3 or a suicide attempt,4 the responding rescue personnel will be faced with contaminated casualties requiring decontamination. The majority of the casualties will have nontime-critical injuries/illnesses,5 but a number of them will potentially be critically ill/injured requiring immediate intervention (during decontamination) to prevent loss of life.1,5
Received 4 March 2010; accepted 4 May 2010. Address correspondence to Nick Castle, Emergency Department, Frimley Park Hospital, Portsmouth Road, Surrey, Camberley GU16 7UJ, UK. E-mail:
[email protected]
This presents a clinical challenge as the responding medical rescuers will be required to wear a CBRN Personal Protective Equipment (CBRN-PPE) to prevent rescuer contamination,6–10 but the CBRN-PPE is known to impair fine motor skills.11,12 Despite this, prompt resuscitation measures remain a key element of casualty management.1,13 The United Kingdom’s National Health Service (NHS) has issued to all emergency departments and ambulance services a standardized CBRN-PPE12 as well as prepositioned casualty treatment pods for dispatch to an incident scene.14 These pods contain a number of antidotes for intravenous (IV) or intramuscular (IM) administration presented in glass ampoules ranging in sizes from 1 to 20 mL. The effect of the CBRN-PPE on preparing drugs for administration has not been widely evaluated, although Udayasiri et al.15 partially examined this issue by evaluating the effect of the CBRNPPE on the assembly of the “Minijet” prefilled syringes.
Clinical Toxicology vol. 48 no. 6 2010
Does CBRN-PPE adversely affect the draw up drugs
523
To date no study has been undertaken to evaluate the impact of the CBRN-PPE on the drawing up of drugs from glass ampoules, plastic ampoules, or other makes of prefilled syringes.
Design Data from Udayasiri et al.15 along with timing from five pretrail test runs by non-participating clinicians were used to inform the sample size. Power calculation identified a minimum requirement of 48 participants. Eighty-one participants were recruited (Table 1); one withdrew due to claustrophobia while wearing the CBRN-PPE (not previously known to be claustrophobic). We elected to use 1% as our level of significance (p-value = 0.01) rather than the traditional 5% (p-value = 0.05%) to minimize the risk of type-I error. Emergency nurses, nurse lecturers, paramedic lecturers, and part-time student paramedics were recruited following internal university adverts. All participants work in the NHS, including the paramedic students who worked for statutory ambulance services as ambulance crew. All participants were competent at drawing up drugs for IV/IM administration, but had not previously drawn up drugs while wearing the CBRNPPE. University of Hertfordshire granted ethics approval and all participants gave written consent. The NHS level-C CBRN-PPE was used (Picture 1) during this study. The NHS CBRN-PPE is a fully encapsulated suit with integral butyl gloves that provide filtered air and has been described previously.12 The Aurum (Picture 2) and Minijet (Picture 3) are both widely used within the NHS, as are glass and plastic ampoules. Atropine (a commonly used CBRN antidote) is available in all these presentations. Participants drew up “drugs” using the four different techniques: Aurum prefilled syringe (A), Minijet prefilled syringe (MJ), glass ampoule with syringe and needle (GA), and plastic ampoule with syringe and needle (PA). From the standpoint of assembly and administration these devices can be divided into two groups: A and MJ, and GA and PA. Each skill was performed once without wearing the CBRN-PPE (participants acting as their own control) and twice while wearing the CBRN-PPE to monitor for learning effect. The order of the skill attempts and the wearing of the CBRN-PPE Table 1. Number of participants at each level of experience Level of experience
Number of participants
Emergency nurses 24 Nurse lecturers 4 Paramedic 14 Paramedic lecturers 6 Paramedic student 32 Twenty-five Emergency nurses were recruited, but 1 withdrew because of claustrophobia while wearing the CBRN-PPE Total 80
Picture. 1. The NHS CBRN-PPE.
Picture. 2. An Aurunm prefilled syringe.
was randomized. Timing commenced from the point at which the participant first touched the syringes, needles, packaging, or ampoules and ended when the “drug” was ready for administration. All prefilled syringes, syringes, and needles were presented in their original packaging, but plastic and glass ampoules were removed from their external packaging. Once the “drug” was drawn up [10 mL of water for injection (PA or GA) or assembly of a 10 mL prefilled syringe], the participant handed the syringe to the researcher. The researcher then measured the drawn up volume using a beaker with 0.1-mL incremental markings, to detect wastage. Following completion of all four skills, the participants completed a questionnaire to ascertain the ease of drawing up
Clinical Toxicology vol. 48 no. 6 2010
524
N. Castle et al.
300
Time
200
100
0 Without CBRN-PPE suit First use of CBRN-PPE suit Second use of CBRN-PPE suit CBRN
Fig. 1. Box plots of time by CBRN-PPE use. Picture. 3. A Minijet prefilled syringe. Table 2. Distribution of time taken to draw up drugs by device Time (s)
each drug presentation as well as their perceived risk of sharp injuries with each method.
Results The CBRN-PPE had a negative effect on skill performance (Fig. 1) in the first attempt while wearing the CBRN-PPE, typically taking 63.8% longer time than without (95% confidence interval: 55.3–72.9%). The participants did, however, improve by repeating the skill, typically taking only 38.9% longer time than the non-CBRN-PPE controls in the second attempt to skill completion (95% confidence interval: 31.7–46.6%). The choice of drug presentation had an effect on the time taken to draw up the drugs (Table 2), with A being the fastest overall, but with MJ being faster than either PA or GA (all p-values < 0.001). The improved speed obtained during drawing up drugs from prefilled syringes reflects the participants’ assessment of the ease of skill completion (Table 3), with 97.5 and 88.8%, respectively, of the participants stating that A or MJ was either very easy or easy to use (wearing the CBRN-PPE). This compared with 30 and 15% of participants rating PA or GA the same.
Aurum Minijet Plastic Glass
Minimum
Lower quartile
Median
Upper quartile
Maximum
0.93 4.60 13.91 17.03
3.95 14.22 33.05 49.13
6.00 19.10 51.02 71.10
17.92 42.67 138.62 171.10
40.57 112.61 235.88 255.10
Within the different professional subgroups (Table 1) no particular drug presentation was within the 1% significance value with regard to the speed of drawing up apart from PA (range 14.40–73.7) and GA (17.03–101.01) in the hands of frequently practicing nurses where GA took 69.5% longer time (p-value < 0.001, 95% confidence interval: 44.4–98.9%). Emergency nurses were significantly faster than all the other groups when drawing up drugs from either PA or GA (Table 4). During the drawing up process, varying degrees of volume of “drug” were lost, although in two-thirds of the attempts all 10 mL was drawn up. The lowest volume lost was from the prefilled syringes (Table 5). The highest volume was lost from the GA.
Table 3. Opinion on ease of carrying out procedure in suit by device Ease of drawing up drugs while wearing the CBRN-PPE Device
Very easy
Easy
Neither easy nor difficult
Aurum Minijet Plastic Glass Total
74 (92.5%) 32 (40.0%) 7 (8.8%) 3 (3.8%) 116 (36.2%)
4 (5.0%) 39 (48.8%) 17 (21.2%) 9 (11.2%) 69 (21.6%)
4 (5.0%) 25 (31.2%) 9 (11.2%) 40 (12.5%)
2 (2.5%)
Difficult
Very difficult
Total
0 (0.0%) 3 (3.8%) 23 (28.8%) 24 (30.0%) 50 (15.6%)
0 (0.0%) 2 (2.5%) 8 (10.0%) 35 (43.8%) 45 (14.1%)
80 (100.0%) 80 (100.0%) 80 (100.0%) 80 (100.0%) 320 (100.0%)
Clinical Toxicology vol. 48 no. 6 2010
Does CBRN-PPE adversely affect the draw up drugs
525
Table 4. Distribution of time for drawing up drugs by different device and professional group while wearing the CBRN-PPE Time (s) to draw up drug Device
Experience
Minimum
Lower quartile
Median
Upper quartile
Maximum
Aurum
Emergency nurse Nurse lecturers Paramedic Paramedic lecturer Paramedic student Emergency nurse Nurse lecturers Paramedic Paramedic lecturer Paramedic student Emergency nurse Nurse lecturers Paramedic Paramedic lecturer Paramedic student Emergency nurse Nurse lecturers Paramedic Paramedic lecturer Paramedic student
2.14 2.90 0.93 1.90 1.55 5.00 8.10 7.70 4.60 10.22 14.30 38.70 21.41 23.10 13.91 17.03 39.00 19.42 40.05 21.91
7.05 3.29 2.70 4.37 3.88 11.52 23.12 11.10 15.91 17.35 23.76 64.02 32.06 48.91 47.06 40.05 48.64 48.77 65.29 67.36
9.00 3.93 3.89 5.50 4.95 17.45 27.94 16.06 18.41 21.10 33.23 67.60 47.57 64.52 66.45 51.15 67.84 64.11 88.41 96.12
12.04 5.00 6.01 6.71 7.10 22.53 32.44 23.19 22.42 28.08 40.44 73.98 69.11 89.71 92.99 77.96 79.76 88.01 107.11 134.76
23.01 28.68 20.00 10.10 40.57 43.60 49.10 32.98 33.90 112.61 73.70 168.57 129.71 179.10 235.88 101.01 125.53 191.10 255.10 245.12
Minijet
Plastic
Glass
Table 5. Distribution of amount of volume drawn up by device Volume (mL) Device
Minimum
Fifth percentile
Aurum 9.20 Minijet 1.00 Plastic 0.00 Glass 0.00 Total volume to be drawn up = 10 mL.
Lower quartile
Median
Upper quartile
Maximum
10.00 9.90 9.70 9.45
10.00 10.00 10.00 9.90
10.00 10.00 10.00 10.00
10.00 10.00 10.00 10.00
9.90 8.65 8.50 7.65
Table 6. Opinion on likelihood of CBRN-PPE damage by device Likelihood of suit damage Device
Highly likely
Likely
Neither likely nor unlikely
Unlikely
Very unlikely
Total of completed questionnaires
Aurum Minijet Plastic Glass Total
0 (0.0%) 0 (0.0%) 15 (18.8%) 71 (88.8%) 86 (26.9%)
0 (0.0%) 0 (0.0%) 26 (32.5%) 5 (6.2%) 31 (9.7%)
2 (2.5%) 6 (7.5%) 23 (28.8%) 3 (3.8%) 34 (10.6%)
6 (7.5%) 14 (17.5%) 12 (15.0%) 1 (1.2%) 33 (10.3%)
72 (90.0%) 60 (75.0%) 4 (5.0%) 0 (0.0%) 136 (42.5%)
80 (100.0%) 80 (100.0%) 80 (100.0%) 80 (100.0%) 320 (100.0%)
When participants were asked about the technique that was most likely to cause damage to the CBRN-PPE (Table 6), the GA was deemed to pose the greatest risk. A and MJ were deemed the safest. During data collection, one participant had
a needle stick injury while wearing the CBRN-PPE and three participants cut their fingers when breaking the GA while not wearing the CBRN-PPE. Inspection of the CBRN-PPE after each skill demonstrated no obvious damage.
Clinical Toxicology vol. 48 no. 6 2010
526
Discussion The administration of antidotes following a CBRN incident can be lifesaving; following the Tokyo Sarin gas attack, one receiving hospital administered over 2,800 ampoules of 0.5 mg atropine to 640 patients.16 There is no maximum dose of atropine13 as treatment is repeated until respiratory symptoms diminish. Okumura et al.16 state that following exposure to an organophosphate-based compound (to include nerve agents) casualties would require between 2 mg IM (mild symptoms) to more than 10 mg (severe symptoms) of IM atropine during their immediate management. This dosing regime resembles that used by the U.K. military,17 which is initially based on self/buddy aid with IM atropine. Even higher doses of atropine where used during the Iran–Iraq war, with atropine commonly being administered as a continuous IV infusion.15 During data collection, three participants while wearing normal clothes cut their fingers on the GA and one candidate had a needle stick injury while wearing the CBRN-PPE. The risk of suit penetration from either a hypodermic needle or the sharp part of the glass ampoules highlights the potential of rescuer contamination. The risk of needle stick injury can be reduced by using “blunt” drawing-up needles for fluid that can be successfully drawn up from plastic ampoules without the use of a needle. Regardless of the drawing-up technique that is adopted, a hypodermic needle will still be required prior to IM administration. IM drug administration remains the mainstay for the immediate management of “exposed casualties.” Although the onset of symptom relief following IM administration is significantly slower than that following IV administration (2 mg atropine IV onset 1 min vs. 8 min in IM administration)16 and absorption can be unpredictable in the hypotensive or hypothermic patient.13 Critically ill/injured patients may require higher doses of antidotes (e.g., atropine)16 and therefore the availability of prefilled syringes may confer a clinical benefit as both of the prefilled syringes were faster than either the glass or plastic ampoules (Tables 2 and 4). It is noteworthy that at the upper quartile participants took over 2 min to draw up drugs from PA and nearly 3 min from GA. Both types of prefilled syringes proved to be faster, easier to use, and more accurate with regard the total volume drawn up. They were also perceived as being safer to use. This ease and speed of assembly is due to the design of prefilled syringes, which minimizes fine motor movement, a motor skill adversely affected by wearing the CBRN-PPE.11,12 There are no studies to benchmark the time periods for drawing up drugs using Aurum, glass, or plastic ampoules while wearing the CBRN-PPE, although our data regarding drawing up drugs from an MJ prefilled syringe reflect the findings of Udayasiri et al.15 The accuracy of drawing up all of 10 mL varied from device to device, although the difference lacked statistical significance. We elected to use 10 mL ampoules/prefilled syringes across all four devices as this reflects the size of the
N. Castle et al. majority of prefilled syringes. While observing the participants attempting to pick up the PA/GA, it was noted that loss of dexterity due to the integral butyl gloves made this task difficult. Butyl gloves are known to reduce fine motor movement by 55%.11 We chose not to assess the effect of the CBRNPPE on drawing up from a 1 mL ampoule (atropine is presented as a 1 mL GA in the United Kingdom mass casualty pods), but it is likely to be more difficult to achieve than with the larger 10 mL ampoule. The effect of losing even small volumes of drug present in the more concentrated 600 mcg/1 mL ampoule would have a greater effect on the amount of drug administered than the same amount of volume lost from a more dilute solution of, for example, 1 mg atropine 10 mL. This potential variation in loss of drug volume could therefore become clinically significant when using the more concentrated drug presentation. All participants had experience of using all of the drug presentations evaluated during this study and yet a clear learning effect was also noted during the repetition of preparing drugs for administration. This learning effect was seen across all four techniques. A similar learning effect has been demonstrated with other resuscitation skills.12 The presence of a learning effect is an important observation as it highlights the importance of practicing skills such as drawing up of drugs while wearing the CBRN-PPE instead of just concentrating on decontamination drills. Frequently practicing nurses, all from emergency departments, performed significantly better than other professionals with regard to drawing up drugs from both types of ampoules. The reason for this was not explored but this maybe due to the fact that drawing up drugs is a skill frequently practiced by emergency nurses.
Limitation Our research design required all ampoules to be of the same size and therefore we elected to use 10 mL prefilled syringes and PA/GA. Had we also elected to evaluate the 1 mL GA we may have detected additional issues.
Conclusion The current reliance on GA as the mainstay of the mass casualty drug pod stock piles needs to be reviewed as clinicians found drawing up 10 mL from a GA while wearing the CBRB-PPE challenging and perceived GA, needle, and syringe as the most likely to damage the CBRN-PPE. This increased risk of accidental contamination of a rescuer secondary to suit damage should be avoided wherever possible. A combination of prefilled syringes, PA, and, where no alternative exists, GA should be considered for mass antidote storage for responding to casualties following a CBRN incident. In addition, all staff that may be required to instigate treatment while wearing the CBRN-PPE should practice drawing up drugs while wearing the CBRN-PPE.
Clinical Toxicology vol. 48 no. 6 2010
Does CBRN-PPE adversely affect the draw up drugs
Declaration of interest This Project was supported by UBC Pharma through an educational grant. UBC Pharma had no prior notification of the outcome of this research ahead of publication nor have they had any editorial input. UBC Pharma produces the Minijet prefilled syringe.
References 1. Okumara T, Takasu N, Ishimatsu S, Miyanoki S, Mitsuhashi A, Kumada K, Tanaka K, Hinohara S. Report on 640 victims of the Tokyo subway sarin attack. Ann Emerg Med 1996; 28:129–136. 2. Saunders P, Jeffery B. Chemical Incidents in England and Wales, 2005. Health Protection Agency; 2007. http://www.hpa.org.uk/web/ HPAwebFile/HPAweb_C/1194947314743. Accessed 10 February 2010. 3. Clarke SFJ, Chilcott RP, Wilson JC, Kamanyire R, Baker DJ, Hallett A. Decontamination of multiple casualties who are chemically contaminated: a challenge for acute hospitals. Prehosp Disaster Med 2008; 23:175–181. 4. National Poisons Information Service. Annual Report 2008–2009. Health Protection Agency; 2009. http://www.hpa.org.uk/web/HPAwebFile/ HPAweb_C/1252326271299. Accessed 10 February 2010. 5. Byers M, Russell M, Lockey DJ. Clinical care in the “hot zone”. Emerg Med J 2008; 25:108–112. 6. Nozaki H, Hori S, Shinozawa Y, Fujishima S, Takuma K, Sagoh M, Kimura H, Ohki T, Suzuki M, Aikawa N. Secondary exposure of medical staff to sarin vapour in the emergency room. Intensive Care Med 1995; 21:1032–1035. 7. Nakajima T, Sato S, Morita H, Yanagisawa N. Sarin poisoning of a rescue team in the Matsumoto sarin incident in Japan. Occup Environ Med 1997; 54:697–701.
527 8. Geller RJ, Singleton KL, Tarantino ML, Drenzek CL, Toomey KE. Nosocomial poisoning associated with emergency department treatment of organophosphate toxicity – Georgia. 2000. J Toxicol Clin Toxicol 2001; 39:109–111. 9. Horton DK, Berkowitz Z, Kaye WE. Secondary contamination of ED personnel from hazardous material events, 1995–2001. Am J Emerg Med 2003; 21:199–204. 10. Stacey R, Morfey D, Payne S. Secondary contamination in organophosphate poisoning: analysis of an incident. Q J Med 2004; 97:75–80. 11. Kruger G. Psychological and performance effects of chemical biological protective clothing and equipment. Mil Med 2001; 166(Suppl 2):41–43. 12. Castle N, Owen R, Hann M, Clark S, Reeves D, Gurney I. Impact of Chemical, Biological, Radiation, and Nuclear Personal Protective equipment on the performance of low- and high-dexterity airway and vascular access skills. Resuscitation 2009; 80:1290–1295. 13. Weinbroum AA, Rudick V, Paret G, Kluger Y, Ben Abraham R. Anaesthesia and critical care considerations in nerve agent warfare trauma casualties. Resuscitation 2000; 47:113–123. 14. Emergency Planning: UK Reserve National Stock for Major Incidents. Department Health December; 2001. http://www.dh.gov.uk/en/ Publicationsandstatistics/Lettersandcirculars/Dearcolleagueletters/ DH_4003208. Accessed 3 February 2010. 15. Udayasiri R, Knott J, Taylor D, Papson J, Leow F, Hassan FA. Emergency department staff can effectively resuscitate in level C personal protective equipment. Emerg Med Australas 2007; 19:113–121. 16. Okumura T, Hitomi T, Hirahara K, Itoh T, Iwamura T, Nagashima F, Nakashima A, Motomura T, Motomura A, Ariyoshi K, Tominaga T, Tsuruwa M, Suzuki K, Taki K. Effective use of drugs to counter chemical terrorism. Curr Drug Ther 2009; 4:139–143. 17. Nerve agents. J R Army Med Corps 2002; 148:344–357. http:// www.ramcjournal.com/. Accessed 8 June 2010.
Clinical Toxicology (2010) 48, 528–532 Copyright © Informa UK, Ltd. ISSN: 1556-3650 print / 1556-9519 online DOI: 10.3109/15563650.2010.497150
ARTICLE LCLT
Cardiac effects of “mad honey”: a case series ERTUGRUL OKUYAN1, AHMET USLU2, and MUSTAFA OZAN LEVENT2 Cardiac effects of mad honey
1
Department of Cardiology, Bagcilar Education and Research Hospital, Istanbul, Turkey Serife Baci State Hospital, Kastamonu, Turkey
2
Background. Grayanotoxins (GTX), also known as andromedotoxins, are produced by plants of the Ericacae family. This toxin is responsible for “mad honey” intoxication, which can present with fatal cardiac bradyarrhythmias and circulatory collapse. GTXs lead to cardiac toxicity because they increase sodium channel permeability and activate the vagus nerve. Objective. We evaluated 42 patients (33 males) prospectively who had been hospitalized with diagnosis of “mad honey” intoxication in a state hospital setting. Methods and results. The median age of patients was 48.5 years and all patients were admitted with complaints of nausea, vomiting, dizziness, fainting, and sweating. Five of the patients had syncope before admission. On admission, the mean systolic blood pressure was 73.1 ± 12.7 mmHg, the mean diastolic blood pressure was 52.1 ± 11.3mmHg, mean heart rate was 38 ± 7 bpm. On initial electrocardiograms, 18 patients had sinus bradycardia, 15 patients had complete atrioventricular block, and 9 patients had nodal rhythm. All patients were monitored in a coronary care unit and treated sympomatically with atropine, intravenous fluids, and dopamine. None of the patients needed temporary pacing and all were discharged without complications. Conclusion. “Mad honey,” which is produced widely in northern parts of Turkey can be toxic. This intoxication should be considered in patients admitted to emergency department with bradycardia and hypotension especially in regions where this honey is produced. Keywords Mad honey; Grayanotoxin; Cardiotoxicity
Introduction
Methods
Grayanotoxin (GTX; andromedatoxin, acetylandromedol, or rhodotoxin) is a toxin found in Rhododendron species and other Ericaceae.1,2 It may be found in honey made in areas where Ericaceae is present and causes poisoning called “mad honey” intoxication.1–5 GTX is a polyhydroxylated cyclic diterpene. It binds to specific sodium ion channels on cell membranes.1–4 More than 60 toxic diterpenoids with the GTX skeleton have been isolated,6 and may be detected in samples of Rhododendron material by using gas chromotography.1 Turkish honey from the Black Sea coast of Turkey may contain GTX and cause poisoning. Gross physical symptoms occur after a dose-dependent latent period of minutes to hours. Despite the potential cardiac problems the condition is rarely fatal and generally lasts less than a day.1–5 Honey from Japan, Brazil, the United States, Nepal, and British Columbia may also be contaminated with GTXs.1–5,7–9 We report our recent clinical experience.
The study was conducted at an urban state hospital with more than 100,000 emergency department visits annually. We prospectively and consecutively enrolled 44 patients with the diagnosis of “mad honey” intoxication admitted to our emergency department between January 2005 and July 2007 in Kastamonu in the Black Sea region. At the time of hospitalization, key demographic and clinical characteristics were collected, a detailed history was obtained, and a thorough physical examination was performed. A diagnostic evaluation was performed for each patient; this included blood chemistry, hematology, erythrocyte sedimentation rate, fibrinogen levels, cardiac markers, including creatine kinase and MB fraction and troponin T levels, coagulation panel, ECG, chest X-ray, and transthoracic echocardiography. Patients were excluded from the study if they had known coronary artery disease history or chest pain and cardiac enzyme elevations. Data were entered using a form prepared by the researchers that was completed during initial presentation, and at the time of patient discharge. Emergency department staff were informed about the study protocol to facilitate inclusion of all patients with “mad honey” intoxication. All initial physical examinations were performed by a cardiologist and he also abstracted the data from the medical records. The diagnosis of “mad honey” poisoning was based on a history of ingestion of unprocessed locally obtained honey and typical signs
Received 18 March 2010; accepted 26 May 2010. Address correspondence to Ertugrul Okuyan, Department of Cardiology, Bagcilar Education and Research Hospital, Istanbul, Turkey. E-mail:
[email protected]
Clinical Toxicology vol. 48 no. 6 2010
Cardiac effects of mad honey of dizziness, ataxia, bradyarrhythmias, diaphoresis, and hypotension. Normally distributed data are shown as arithmetic mean and standard deviation and data not normally distributed as median values. The Mann–Whitney U-test was used for the comparison of nonparametric data and t-test was used to compare continuous variables. All tests were performed with SPSS (Statistical Package for Social Sciences, Chicago, IL, USA) for Windows 10.0. The study complies with the Declaration of Helsinki and informed consent has been obtained from all included subjects. As this is an observational and descriptive study, ethical committee approval by our institution was not required.
Results Forty-four patients were enrolled. Two patients with coronary artery disease history and cardiac marker elevations were excluded. Thirty-three patients were male and nine were female, median age was 48.5 years (range 32–73). Patients had ingested an estimated 20–200 g of mad honey before admission. Symptoms came on as quickly as 10 min after ingestion of honey, but in some were delayed up to 4 h. Almost all patients (40/42) were admitted with complaints of nausea, vomiting, dizziness, fainting, and sweating. Five of the patients had syncope before admission and one had a generalized seizure. On admission, the mean systolic blood pressure was 73.1 ± 12.7 (50–100) mmHg and the mean diastolic blood pressure was 52.1 ± 11.3 (40–70) mmHg. Mean heart rate was 38 ± 7 (25–59) bpm. On initial electrocardiograms, 18 patients had sinus bradycardia, 15 had complete atrioventricular block, and 9 had a nodal rhythm. The clinical
529 characteristics of the patients are shown in Table 1. All patients were monitored in the coronary care unit. Most patients (33) responded well to 0.5–1 mg of intravenous atropine, but in seven dopamine in vasopressor doses, and administration of parenteral saline was needed. None of the patients needed temporary cardiac pacing and all were discharged within 12–36 h of admission without complication, except one patient with previous coronary artery bypass graft surgery. Detailed history of the patients revealed that 6 patients had hypertension, 7 had both hypertension and diabetes mellitus, 10 had dyspepsia, 5 had coronary artery disease, 1 had chronic lymphocytic leukemia, and 2 had been suffering from rheumatic musculoskeletal pain. Eight patients were on betablocker therapy before hospital admission and 10 patients were using antihypertensive medication, including angiotensinconverting enzyme inhibitors and amlodipine. The mean heart rate and arterial blood pressure levels of patients on betablocker therapy were significantly lower than patients without known beta-blocker usage (Table 2). No significant difference was observed for patients on antihypertensive therapy. Most of these patients claimed that they consumed “mad honey” as a natural medication for their illnesses. Eight of 33 male patients said that they believed mad honey was a sexual stimulant and had bought “mad honey” with this in mind.
Discussion GTX is produced by Ericaceae family plants. Rhododendron species Rhododendron luteum and Rhododendron ponticum belong to this family and mad honey is produced by these
Table 1. Summary of clinical characteristics of patients Age (years) Gender (male/female) Syncope Mean systolic BP (mmHg) Mean diastolic BP (mmHg) Mean heart rate (bpm) Sinus bradycardia Complete AV block Nodal rythm History of hypertension History of diabetes mellitus History of coronary artery disease Atropine administered Dopamine administered IV fluids administered Temporary pacing Seizure Mean troponin T levels Mean left ventricular ejection fraction (LVEF) (%) Mean hospitalization duration (hours) Figures in brackets are ranges or percentages.
48.5 (32–73) 33/9 5 (12%) 73.1 ± 12.7 (50–100) mmHg 52.1 ± 11.3 (40–70) mmHg 38 ± 7 (25–59) beats/min 18 (43%) 15 (36%) 9 (21%) 13 (31%) 7 (17%) 5 (12%) 40 (95%) 7 (17%) 20 (48%) 0 1 (2%) 0.019 ± 0.007 (0.010–0.038) 58 ± 11 (50–67) 34 ± 20 h (24–70)
Clinical Toxicology vol. 48 no. 6 2010
530
E. Okuyan et al.
Table 2. Heart rate and blood pressure values at admission and hospitalization duration in mad-honey-poisoned patients with and without known beta-blocker usage Mean heart rate
Mean systolic BP
Mean diastolic BP
Patients on beta-blockers (n = 8) 35 ± 6 (25–54 bpm) 68 ± 14 (50–90 mmHg) 50.4 ± 9 (40–70 mmHg) Patients not using 43 ± 8 (28–59 bpm) 78.2 ± 11.14 (50– 53.8 ± 13.6 (40– beta-blockers (n = 34) 100 mmHg) 70 mmHg) p-value 0.02 0.03 Nonsignificant
plants.1–4,8,9 GTX-containing plants are native to ecosystems throughout the world, including Turkey, Japan, Nepal, Brazil, and parts of Europe.10,11 There are a number of toxic species native to the United States. Of particular importance are Rhododendron occidentale, Rhododendron macrophyllum, and Rhododendron albiflorum, species found in a wide swath of North American territory from British Columbia to Oregon and southern California.11 The principal toxic isomer of GTX in Rhododendron is GTX III although others, GTX I and GTX II, are present in lesser amounts. GTX I is also toxic and GTX II is less toxic.1 A variety of other Rhododendrons, Azaleas, and other members of the Ericaceae also contain these toxins. The GTXs bind to sodium channels on cell membranes. The binding unit is the group II receptor site, localized on a region of the sodium channel that is involved in the voltagedependent activation and inactivation.1,9,12,13 These compounds prevent inactivation; thus, excitable cells (nerve and muscle) are maintained in a state of depolarization, during which entry of calcium into the cells may be facilitated. This action is similar to that exerted by the alkaloids of veratrum and aconite.14 All of the observed responses of skeletal and heart muscles, nerves, and the central nervous system are related to the membrane effects.15 It is proposed that sites of cardiac and respiratory actions of GTXs are within the central nervous system and that bradycardic effects of GTX are mediated by vagal stimulation at the periphery.16 A nonselective muscarine antagonist atropine and a selective muscarine antagonist AF-DX 116 have been suggested as antidotes.16 In general, beekeepers sell their own unprocessed honey in local or regional markets. This honey is a natural, unprocessed, and unregulated product. They reach the consumer directly, with no intermediate processing. When Rhododendrons bloom in May and June, beekeepers take their hives to higher altitudes where these flowers are plentiful. It is during that season that mad honey is produced from valleys where rhododendron flowers are abundant. Local people can distinguish that honey from other varieties. It causes a sharp burning sensation in the throat and is thus also referred to as bitter honey.3 Toxic honey has a brown color and it smells like a “linden flower.” GTX poisoning is well described and is the subject of several recent case series. Many of the previous reports were retrospective or single-case reports and controversies exist about care and follow-up of these patients. It is not clear
Hospitalization duration 38 ± 20 (24–70 h) 30 ± 20 (24–60 h) Nonsignificant
where these patients should be managed, in the emergency department or coronary intensive care unit, or what is the safe period of observation for patients with a history of ingestion of “mad honey” before discharge. To clarify these issues, after initiation of this study, some have tried to answer these questions.8,12,13 “Mad honey” intoxication is usually a benign condition and is rarely fatal. Initial symptoms are excessive salivation, perspiration, vomiting, dizziness, weakness, and parasthesia in the extremities and around the mouth, low blood pressure, and sinus bradycardia. Symptoms may come on as quickly as within 10 min and be delayed up to 4 h. With higher doses, symptoms can include loss of coordination, severe and progressive muscular weakness, complete atrioventricular block, nodal rhythm or Wolff–Parkinson–White syndrome, and shock.3–5,17,18 Especially in older patients, acute myocardial infarction may complicate the scenario because of low cardiac output and decreased coronary reserve.7 Despite the potential cardiac problems, the condition generally lasts less than a day. Death associated with mad honey poisoning has not been reported so far in the literature. All of the complaints were temporary and responded to atropine and saline infusion with close observation of vital signs.1–5,9 However, one of our patients had a generalized seizure. In our study, the mean heart rates and arterial blood pressure levels of patients on admission were extremely low. This point is important, because it may reflect the possible serious consequences of the intoxication and it would seem that the patients in our series had more severe toxicity than has previously been reported.2,5,8,13 Nevertheless, most patients were discharged on the second day without any complication, but hypotension persisted in a patient with history of CABG surgery who was also on beta-blocker therapy. We also found that blood pressure and heart rate of patients on beta-blocker therapy before admission was significantly lower (Table 2). Therefore, it seems that patients using betablocking agents and antihypertensive therapies were more vulnerable to the cardiac effects of GTX. Mad honey is traditionally used in the Black Sea region as an alternative medicine for the treatment of hypertension and heart failure. Therefore, physicians in this area should warn their patients about untoward effects of “mad honey” and antihypertensive drug combinations. In this study all of the reported features are quite suggestive of a transient vagal hypertony responsible for vasodilation, depression of natural pacemaker’s activity, and gastroenteric
Clinical Toxicology vol. 48 no. 6 2010
Cardiac effects of mad honey reflexes. As a matter of fact, all symptoms seem to disappear upon administration of vagolytic and sympathomimetic drugs. Sung-Eun Kim et al. reported that GTX increases Ca2+ influx through voltage-dependent Ca2+ channels secondary to activation of voltage-dependent Na+ channels in inhibitory and excitatory nerve terminals synapsing on ventromedial hypothalamus neurons, and the subsequent increased release of GABA and glutamate from these terminals may be responsible for the autonomic symptoms of GTX intoxication.19 The centrally acting properties of GTX may be an explanation for the seizure noted in one of our patients, but this point should be further studied. In general, the severity of the honey poisoning depends on the amount ingested. Yilmaz et al. reported that the amount of honey causing poisoning is between 5 and 30 g.2 The concentration of GTX ingested may differ greatly from case to case. As GTXs are metabolized and excreted rapidly, patients generally regain consciousness and feel better within hours, and heart rate and blood pressure usually return to normal within 2–9 h.3 The amount of ingested honey was 20–200 g in our study. This value also seems to be greater than previously reported series.2,3,12,13 This can also partly explain the more severe toxicity in our patients. Gunduz et al. reported that 6 h monitoring was sufficient for stable emergency department patients.12 From our experience we suggest 24 h monitoring, particularly for older patients, those with coexisting cardiovascular pathologies, and patients receiving beta-blocking agents. Physicians should exclude acute coronary syndromes and other pathologies (neurologic, metabolic) before diagnosing “mad honey” intoxication. In our study, more than half of the patients were aware of the possible toxic nature of “mad honey.” They had consumed honey as a natural therapy for their illness or as a folk cure. In a recent report from Demircan et al., local beekeepers ranked sexual performance enhancement as the most common reason for therapeutic mad honey consumption in middle-aged men.8 Because of the increasing preference for natural products, intoxication induced by consumption of honey is also increasing worldwide. According to the Korea Food and Drug Administration, 8,048 kg of “mad honey” were imported to Republic of Korea in recent years.18 Even though importation of mad honey was prohibited in 2005, covert distribution is still possible. As the means of travel and transport systems are developing rapidly, natural honeys are distributed worldwide. For this reason, new reports from nonendemic areas can be seen and documented mad honey intoxication cases would be reportable from central Europe and North America.9,20,21 Turkey, with its unique geographic location, joins two continents: Europe and Asia. Because of this geographic position, Turkey may serve as a biological bridge in the migration of plant species. Easy and rapid transport of foodstuffs has resulted in reports of mad honey poisoning in locations far from the Black Sea region.8
531
Conclusion GTX poisoning generally respond well to atropine and supportive therapy and generally resolves in 24 h without complication. But, in patients with decreased coronary reserve, patients on anti-ischemic medications, and patients on antihypertensive therapies, more monitoring is warranted, and these patients should be followed until clinical condition completely resolves. We suggest at least 24 h of observation in a coronary care unit for patients with coexisting cardiovascular pathologies. All patients should be closely observed until other possible cardiac etiologies will be completely ruled out. As people increase use of “natural” products, “mad honey” intoxication cases may be seen more frequently than previously in nonendemic areas. Therefore, physicians should be aware of the possibility of mad honey ingestion in previously healthy patients with sudden unexplained hypotension and bradyarrhythmias.
Limitations The main limitation of this study is the inability to test the contaminated honey. In our facility, we are not able to perform such tests. We should rely on patients and relatives’ statements about amount ingested.
Declaration of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this paper.
References 1. Koca I, Koca FA. Poisoning by mad honey: a brief review. Food Chem Toxicol 2007; 45:1315–1318. 2. Yilmaz O, Eser M, Sahiner A, Altintop L, Yesildag O. Hypotension, bradycardia and syncope caused by honey poisoning. Resuscitation 2006; 68:405–408. 3. Gunduz A, Turedi S, Uzun H, Topbas M. Mad honey poisoning. Am J Emerg Med 2006; 24:595–598. 4. Yavuz H, Özel A, Akkus I, Erkul I. Honey poisoning in Turkey. Lancet 1991; 337:789–790. 5. Özhan H, Akdemir R, Yazici M, Gündüz H, Duran S, Uyan C. Cardiac emergencies caused by honey ingestion: a single centre experience. Emerg Med J 2004; 21:742–744. 6. Katakawa J, Tetsumi T, Sakaguchi K, Katai M, Terai T. Crystal and molecular structure of 10,20-epoxy-grayanotoxin-II. J Chem Crystallogr 2004; 34:311–315. 7. Akinci S, Arslan U, Karakurt K, Cengel, A. An unusual presentation of mad honey poisoning: acute myocardial infarction. Int J Cardiol 2008; 129:56–58. 8. Demircan A, Keles A, Bildik F, Aygencel G, Dogan NO, Gómez HF. Mad honey sex: therapeutic misadventures from an ancient biological weapon. Ann Emerg Med 2009; 54:824–829. 9. Gunduz A, Turedi S, Russell RM, Ayaz FA. Clinical review of grayanotoxin/ mad honey poisoning past and present. Clin Toxicol 2008; 46:437–442.
Clinical Toxicology vol. 48 no. 6 2010
532 10. Onat FY, Yegen BC, Lawrence R, Oktay A, Oktay S. Mad honey poisoning in man and rat. Rev Environ Health 1991; 9:3–9. 11. Lampe KF. Rhododendrons, mountain laurel, and mad honey. JAMA 1988; 259:2009. 12. Gunduz A, Meriçe SE, Baydin A, Topbas M, Uzun H, Türedi S, Kalkan A. Does mad honey poisoning require hospital admission? Am J Emerg Med 2009; 27:424–427. 13. Bostan M, Bostan H, Kaya AO, Bilir O, Satiroglu O, Kazdal H, Karadag Z, Bozkurt E. Clinical events in mad honey poisoning: a single centre experience. Bull Environ Contam Toxicol 2010; 84:19–22. 14. Eller P, Hochegger K. Honey intoxication and the Bezold-Jarisch reflex. Int J Cardiol 2009. DOI:10.1016/j.ijcard.2008.12.206. 15. Seyamal I, Yamaoka K, Yakehiro M, Yoshioka Y, Morihara K. Is the site of action of grayanotoxin in the sodium channel gating of squid axon? Jpn J Physiol 1985; 35:401–410.
E. Okuyan et al. 16. Onat FY, Yegen BC, Lawrance R, Oktay A, Oktay S. Site of action of grayanotoxin in mad honey in rats. J Appl Toxicol 1991; 11:199–201. 17. Dilber E, Kalyoncu M, Yarifi N, Aysenur O. A case of mad honey poisoning presenting with convulsion: intoxication instead of alternative therapy. Turk J Med Sci 2002; 32:361–362. 18. Choo YK, Kang YH, Lim SH. Cardiac problems in mad-honey intoxication. Circ J 2008; 72:1210–1211. 19. Kim SE, Shin MC, Akaike N, Kim CJ. Presynaptic effects of grayanotoxin III on excitatory and inhibitory nerve terminals in rat ventromedial hypothalamic neurons. Neurotoxicology 2010; 31:230–238. 20. Weiss WT, Smetana P, Nurnberg M, Huber K. The honey man-second degree heart block after honey intoxication. Int J Cardiol DOI:10.1016/ j.i.jcard.2008.11.116. 21. Eller P, Hochegger K, Tancevski I, Pechlaner C, Patsch JR. Sweet heart block. Circulation 2008; 118:319.
Clinical Toxicology (2010) 48, 533–538 Copyright © Informa UK, Ltd. ISSN: 1556-3650 print / 1556-9519 online DOI: 10.3109/15563650.2010.492526
ARTICLE LCLT
Hydrogen peroxide ingestion associated with portal venous gas and treatment with hyperbaric oxygen: a case series and review of the literature LOREN KEITH FRENCH, B. ZANE HOROWITZ, and NATHANAEL J. MCKEOWN Hydrogen peroxide ingestion causing portal gas
Department of Emergency Medicine, Oregon Health & Science University, Portland, OR, USA
Introduction. Ingestion of concentrated hydrogen peroxide (H2O2) has been associated with venous and arterial gas embolic events, hemorrhagic gastritis, gastrointestinal bleeding, shock, and death. Although H2O2 is generally considered a benign ingestion in low concentrations, case reports have described serious toxicity following high concentration exposures. Hyperbaric oxygen (HBO) has been used with success in managing patients suffering from gas embolism with and without manifestations of ischemia. Methods. Poison center records were searched from July 1999 to January 2010 for patients with H2O2 exposure and HBO treatment. Cases were reviewed for the concentration of H2O2, symptoms, CT scan findings of portal gas embolism, HBO treatment, and outcome. Results. Eleven cases of portal gas embolism were found. Ages ranged from 4 to 89 years. All but one ingestion was accidental in nature. In 10 cases 35% H2O2 was ingested and in 1 case 12% H2O2 was ingested. All abdominal CT scans demonstrated portal venous gas embolism in all cases. Hyperbaric treatment was successful in completely resolving all portal venous gas bubbles in nine patients (80%) and nearly resolving them in two others. Ten patients were able to be discharged home within 1 day, and one patient had a 3.5-day length of stay. Conclusions. HBO was successful in resolving portal venous gas embolism from accidental concentrated H2O2 ingestions. Keywords
Hydrogen peroxide; Portal venous gas; Hyperbaric oxygen; Embolic gas; H2O2
Introduction The medical profession has encountered the occasional case of arterial gas embolism following ingestion or injection of hydrogen peroxide (H2O2). Literature from as early as the 1960s1 has documented this phenomenon. Fortunately, the incidence of serious toxicity appears to be low, and only a handful of deaths appear in the literature. Harm appears to occur from direct caustic injury to tissue, embolic obstruction of arterial blood flow, lipid peroxidation,2 or perforation of a hollow organ. It has been suggested that first-line treatment of acute arterial oxygen embolism following H2O2 exposure is hyperbaric oxygen (HBO).3 There are no standard guidelines about when to perform diagnostic imaging looking for venous embolism nor is there consensus as to which diagnostic modality should be utilized [i.e., plain films vs. noncontrast computerized tomography (CT)]. Furthermore, there is currently no consensus about what the Received 17 February 2010; accepted 9 May 2010. Address correspondence to Loren Keith French, Department of Emergency Medicine, Oregon Health & Science University, 3181 SW Sam Jackson Park Road, Mail Code CB550, Portland, OR 97239-3098, USA. E-mail:
[email protected]
standard treatment should be if acute venous oxygen embolism is found following H2O2 exposure. Our goal is to describe a series of patients who ingested H2O2 and had evidence of portal venous gas on diagnostic imaging. All patients underwent subsequent treatment with HBO before the development of any ischemic signs of symptoms (Table 1).
Methods We searched computerized data from a single poison center (PC) from July 1999 to January 2010 looking for patients with H2O2 exposure and HBO treatment. We searched cases where “H2O2” or “hydrogen peroxide” was listed in the substance field. We eliminated cases that were subsequently confirmed as nonexposures. HBO was then entered into the treatment field and after a repeat search, 11 cases were found. These 11 cases were reviewed and assessed for concentration of H2O2, estimated amount of ingestion, reason for exposure, presenting signs and symptoms, documentation of portal venous air on abdominal CT, estimated time to HBO, and resolution of portal venous air following HBO. If available, hospital records were reviewed, however the majority of information
Clinical Toxicology vol. 48 no. 6 2010
534
L.K. French et al.
Table 1. Brief summary of cases Age (years) Patient and gender
Exposure
1
86, M
2
57, F
3
66, M
4
89, F
5
5, M
6
65, M
7
38, M
8
4, F
9
40, F
10
67, M
35%, 1 mouthful 35%, 2 mouthfuls 12%, 1 cup
11
33, F
35%, 30 mL
Reason for exposure
Symptoms
35%, unknown Accidental ingestion Oral edema and amount burns, epigastric pain, vomiting 35%, sip Accidental ingestion N/V, throat burns, mild abdominal pain 35%, 8 oz Misuse, ingestion Abdominal pain, N/V 35%, 1 Accidental ingestion CP, N/V tablespoon 35%, 1 Accidental ingestion N/V, hematemesis mouthful Accidental ingestion N/V, abdominal 35%, 1/2 cup pain Accidental ingestion Sore throat 35%, 1/2 cup
Time to Hospital length of stay HBO (h) (LOS) and disposition
Resolution of embolic burden?
5
25 h, DC home
Yes
6.5
16 h, DC home
Yes
6
24 h, left AMA
2
unknown LOS, DC home 22 h, DC home
No, small amount of residual portal air Yes
4
Accidental ingestion N/V
2.5
unknown LOS, DC home unknown LOS, DC home <12 h, DC home
Accidental ingestion N/V, skin burns, seizure Accidental ingestion Nausea, hematemesis, abdominal pain Accidental ingestion Hematemesis, abdominal pain
5
3.5 days, DC home
6
< 24 h, DC home
No, small amount of residual air Yes
< 24 h, DC home
Yes
was gathered from PC data. The study was approved by the institutional review board at the PC’s associated facility.
Results
6.5
Yes
6.5
3
Yes Yes Yes
local hospital by the PC. Presenting symptoms were throat burning, mild abdominal pain, belching, and persistent nausea. Abdominal CT demonstrated several portal venous emboli. She was treated with HBO approximately 6.5 h after exposure. Repeat abdominal CT was normal and the patient was discharged home in less than 24 h after initial contact with the PC.
Case 1 An 86-year-old male inadvertently ingested an unknown amount of 35% H2O2 at his home. His presenting symptoms included vomiting, swelling in his mouth and lips, and oral burns; he was described as “very symptomatic.” Noncontrast abdominal CT scan showed air embolus in his hepatic veins. He was given decadron for airway edema but did not require intubation. He was treated with a single round of HBO (starting 5 h after ingestion), then was admitted to the ICU for observation. He required replacement of his already existing tympanostomy tubes; otherwise there were no complications with HBO. Repeat abdominal CT after HBO showed resolution of gas emboli. He was discharged home the following day.
Case 3 A 66-year-old male drank about one cup of 35% H2O2 apparently in a therapeutic misadventure. His friend called the PC because the patient reported abdominal pain, nausea, and vomiting. The PC referred the patient to the hospital where he was noted to be stable on arrival. An abdominal CT was obtained and reported to be “very positive” with portal gas “everywhere.” He was treated with a single HBO treatment, which started 6 h after ingestion. Although his repeat abdominal CT scan showed a few residual gas bubbles, he was asymptomatic and tolerating fluids. The next morning he left the hospital against medical advice.
Case 2 A 57-year-old female unintentionally drank approximately one sip of 35% H2O2 while in a dark room. She immediately experienced nausea and vomiting. She was referred to the
Case 4 An 89-year-old female unintentionally ingested about 1 tablespoon of 35% H2O2. She called 911 because of chest pain
Clinical Toxicology vol. 48 no. 6 2010
Hydrogen peroxide ingestion causing portal gas and two episodes of vomiting shortly after ingestion and was taken to a nearby hospital for evaluation. An abdominal CT scan was obtained within 20 min of arrival and noted large amounts of gas in both lobes of the liver. She underwent HBO shortly thereafter (approximately 2 h after exposure). Although in no distress, she had persistent throat and epigastric pain as well as bloating. Repeat CT scan showed resolution of gas emboli. Because of persistent symptoms, she underwent endoscopy 36 h after ingestion that was negative for erosions or perforation. She was discharged from the hospital but returned to the emergency room several days later. She was diagnosed with gastritis and discharged home.
Case 5 A 5-year-old boy ingested a mouthful of 35% H2O2 while at his grandparent’s house. He was brought to the nearest hospital via ambulance. On arrival, the patient had nausea without vomiting and was described as very cooperative. Noncontrast abdominal CT scan revealed a liver “full of gas emboli.” He was transferred via life flight 250 miles to the nearest hyperbaric chamber. While awaiting transport he vomited a small amount of mucous and blood. He required myringotomy before HBO treatment. Following decompression (initiated 4 h after exposure) he was admitted for observation. He remained asymptomatic after the HBO treatment and repeat abdominal CT scan showed no gas. He was discharged home less than 24 h after initial contact with the PC.
Cases 6 and 7 A frantic woman called the PC after two of her family members both drank about half a cup of 35% H2O2. The first patient (patient no. 6) was a 65-year-old male, and the second (patient no. 7) was a 38-year-old male. Both men were taken to the nearest hospital where abdominal CT scans revealed portal venous gas in both patients. The older patient reported abdominal pain, nausea, and vomiting, whereas the younger patient reported a sore throat. The two men were transferred to the nearest hyperbaric chamber, and underwent uneventful decompression (both starting 6.5 h after exposure). They were admitted to the hospital for observation, and repeat CT scans were negative in both. Both men were discharge home several days after admission.
535 abdominal CT demonstrated substantial portal venous air. She was taken to the local HBO chamber for HBO treatment (2.5 h after ingestion), and then returned to the emergency room where repeat abdominal CT scan was normal. She was asymptomatic and discharged home within 12 h of initial contact with the PC.
Case 9 A 40-year-old female called the PC immediately after accidentally drinking one to two large mouthfuls of 35% H2O2. She was experiencing mouth and chest pain and complained of burns on her skin. She was taken to the hospital by ambulance. Abdominal CT showed multiple tiny bubbles in her portal system. She was transferred to the HBO chamber where she underwent HBO treatment (treatment started 5 h after ingestion). While in the hyperbaric chamber she suffered a 2 min spontaneously resolving seizure. She was admitted to the ICU following the HBO treatment. Head CT was negative. Repeat abdominal CT in the morning showed some residual air. No further HBO treatment was administered. The patient recovered uneventfully and was discharged home 3 days later. Case 10 The wife of a 67-year-old man was using 12% H2O2 at home. The container it was kept in started to leak, so she transferred the solution to a water bottle and placed it in the refrigerator. The husband consumed about one cupful thinking it was water. Within 30 min he was vomiting blood and had abdominal discomfort. He was taken to the emergency room where he was noted to have mild epistaxis and an erythematous throat. Abdominal CT showed multiple venous gas embolisms (Fig. 1). He was transferred to the hyperbaric chamber and underwent therapy (starting 6 h after exposure). Repeat CT scan in the morning demonstrated resolution of the portal gas (Fig. 2). He was discharged later that day.
Case 8 The mother of a 4-year-old girl called the PC 5 min after her daughter accidentally ingested one mouthful of 35% H2O2. Vomiting was noticed immediately after consumption. She was taken to the nearest emergency room where unenhanced
Fig. 1. Portal venous gas before HBO in patient 10.
Clinical Toxicology vol. 48 no. 6 2010
536
Fig. 2. Resolution of portal venous gas after HBO in patient 10.
Case 11 The father of a 33-year-old woman contacted the PC 5 min after the accidental ingestion of a mouthful of 35% H2O2, which his daughter mistook for spring water. The solution was used primarily by the father to soak his feet. Soon after the ingestion she experienced hematemesis. She was taken to a nearby emergency room and treated supportively. After discussion with the PC, noncontrast CT of the abdomen was performed and demonstrated air throughout her hepatic tree and gastric wall. The patient underwent HBO that started 3 h after exposure. Repeat abdominal CT after HBO was normal. The patient was discharged home the next morning. She declined endoscopy.
Discussion Hydrogen peroxide, also known as dioxidane, is a colorless to light-blue compound, which is liquid at room temperature. The viscosity of H2O2 is slightly higher than water, it has weakly acidic properties (pKa 11.62), and is miscible in water. It is considered a strong oxidizer (oxidation potential of 1.8) and potent bleaching agent. Reasons for its use include both medicinal and nonmedicinal purposes. Internet testimonials tout H2O2 as a treatment for a canker sores, athlete’s foot, yeast infections, and sinus congestion. Several brand-name toothpastes contain H2O2 to help brighten teeth. The Merck Manual lists 1.5% H2O2 as a treatment for trench mouth.4 Hydrogen peroxide has been used to clean laundry and bleach hair. Concentrated H2O2 has been used as a rocket propellant. Recently, use of small amounts of 35% H2O2 has been promoted for nutritional use. The highly reactive oxygen species H2O2 is a by-product of oxygen metabolism within most cells. Humans utilize catalase to decompose hydrogen peroxide via the following reaction: 2H2O2 → 2H2O + O2 + heat. After ingestion or external application, one volume of 6% H2O2
L.K. French et al. may produce 20 volumes of oxygen gas. This gas may dissipate through tissues into the circulation. In the case of ingestion, oxygen gas can permeate through the stomach and into the portal venous system. Once the concentration of oxygen gas exceeds its solubility in blood, bubbles are formed. These bubbles may occlude vascular flow causing tissue ischemia in a fashion similar to other embolic diseases. Although most serious cases seem to involve exposure to solutions >30%, portal venous emboli, hemorrhagic gastritis, and death have all been reported with ingestion of 3% solutions.5,6 In theory, arterial emboli may result from right to left movement across the heart through an atrial septal defect, directly through pulmonary aspiration, or absorption of H2O2 across the GI tract and catabolism within the arterial vasculature.7 Compression therapy allows for higher concentrations of gases to be dissolved in blood, reducing the likelihood of bubble formation. If decompression is performed gradually, the dissolved gas may be exchanged within the alveoli and eliminated via respiration. This is essentially the therapeutic mechanism used for elimination of nitrogen bubbles in decompression sickness. We refer to this compressive/decompressive therapy as HBO as they are often used interchangeably. Ingestion appears to be the most common route for venous gas embolism though not an absolute requirement. Vidil8 reports a case of severe oxygen embolism following wound irrigation with H2O2 resulting in hypoxia and cardiac arrhythmia, treated with subsequent HBO therapy and recovery without cardiac or neurological sequelae. The authors concluded that pressure injection of hydrogen peroxide into a closed cavity should be avoided. There are at least six case reports that describe acute neurological events (seizures, altered mental status or CVA symptoms) following the ingestion of hydrogen peroxide.3,9–11 Three adult cases had complete resolution after HBO. Mullins9 appears to be the first author to publish such a case. In his report, complete resolution of ischemic stroke symptoms occurred following HBO treatment after accidental ingestion of 35% H2O2. Rider3 reported a case in 2007 where an ingestion of a 33% solution resulted in left-sided hemiplegia, confusion, and left homonymous hemianopia. Vander10 reports a case of temporally associated altered mental status and subsequent recovery after HBO from ingestion of concentrated H2O2. The interesting feature in these cases is that time to HBO was 11, 18, and 20 h, respectively. We are not aware of any cases in which a patient developed ischemic symptoms in the setting of hydrogen peroxide poisoning and failed to respond to HBO. Although a favorable outcome is possible even with delay in treatment, permanent disability has been reported in the absence of HBO. Ashdown11 describes a series of three patients who developed severe symptoms following ingestion of 35% H2O2. One patient died, one patient had a permanent dense hemiparesis, and one patient, a 4-year-old boy, developed a spastic paresis and seizure disorder. None of these patients
Clinical Toxicology vol. 48 no. 6 2010
Hydrogen peroxide ingestion causing portal gas underwent hyperbaric therapy. In our database, we are aware of only one case in which a patient had a sudden cardiovascular collapse and subsequent death following a possible ingestion of concentrated H2O2. Exposure could not be confirmed, however, and CT scan of the abdomen was not obtained so no conclusion regarding the cause of death can be made as no autopsy was done. Severe injury is not limited to adults. Cina6 provides a case series of six deaths in children. In all cases, exposure was from ingestion. The amount and concentration was not known in each case, though four children had autopsies which demonstrated features consistent with hydrogen peroxide intoxication. In one case, the death occurred after ingestion of less than 8 oz of a 3% solution. That a child is at risk for accidental ingestion should not be surprising. In our experience, many accidental exposures occur when the solution is poured into an unlabeled bottle and placed into the refrigerator to keep cool. A toddler could easily mistake an inconspicuously appearing clear liquid as something palatable leading to experimentation and potential injury. Hydrogen peroxide ingestion followed by arterial occlusion or neurological injury with the possibility for complete recovery with HBO has been clearly established. What is less clear is if prophylactic treatment with HBO before neurological or ischemic symptoms develop is a cost-effective approach to this scenario. In our series of patients, 80% had complete resolution of portal venous gas on subsequent imaging after HBO. In the two that did not have complete resolution, only a small residual amount of gas was reported and no severe symptoms occurred. Whether this prevented neurological or cardiac sequelae cannot be concluded, and these patients may have recovered spontaneously in time without intervention. We have only one patient in our records who developed a moderated amount of portal venous gas embolism after ingestion of 35% H2O2 and did not receive HBO. We recommended transfer to the hyperbaric chamber, but this intervention was unavailable. Rather than transfer to the HBO center, he was admitted to the local hospital and did well overnight. Repeat CT scan the following day showed resolution of portal gas. Luu12 published a similar case where a 40-year-old female who ingested concentrated H2O2 was treated conservatively after abdominal X-ray demonstrated portal gas. She was released 5 days after admission without major sequelae. Given that the literature describes severe neurological injury and death, our approach has evolved and we are now recommending diagnostic imaging looking for portal venous gas in patients who are symptomatic with early nausea, vomiting, or pain. We favor noncontrast CT of the abdomen given the relatively high sensitivity for portal or peritoneal air in the context of viscous rupture.13 Given the potential for serious sequelae, including one case followed through our PC,9 once embolic gas has been detected our practice is to pursue HBO treatment.
537 In this series, there were no injuries from transportation. The complications from HBO in this series were typmanostomy tubes placement and a brief, self-terminating seizure.
Limitations Our database, like most other PCs, requires passive reporting, therefore we do not know the total number of unreported, minor exposures; symptomatic patients presented to care providers without subsequent CT imaging; or unexplained sudden deaths that are potentially attributable to oxygen gas embolism. Furthermore, because the majority of information was retrieved using information acquired through the PC database, and not hospital records, we have a somewhat limited view of the entire hospital course. It is difficult to quantify the exact volume of solution ingested, as patients rarely have the means or interest in measuring the volume of consumed fluid. We are not able to establish the exact amount of hydrogen peroxide that results in complications, although our cases do illustrate that a patient report of a “sip” or “mouthful” has resulted in portal venous embolism. The factors that predict the development of arterial embolization following venous embolism are unclear. Although the presence of a patent foramen ovale may increase this risk, we do not know whether the relationship between the amount of ingestion, concentration of solution, volume of portal gas, age, gender, or time since ingestion may portend a neurologic deterioration or death. There is only one patient in our search who developed portal venous gas embolism without subsequent HBO. We therefore lack a substantial comparison group and we cannot conclude that HBO improved outcomes. Rather, we observed that most patients did not have residual portal venous gas on imaging after HBO. Finally, HBO therapy is not a widely available service. Transportation risks and costs must be weighed against potential benefits; however, most of the patients in this series were able to be discharged home within 24 h and were able to avoid the cost of prolonged hospitalization. Managing a critically ill or psychiatrically unstable patient, should these complicate such an otherwise eligible patient, may be beyond the scope or comfort level of some hyperbaric centers.
Conclusion Ingestion of concentrated H2O2 has the potential to cause portal venous embolism. Arterial gas embolism is a potentially dreadful consequence of this exposure, but one that is amenable to HBO therapy. We are yet to uncover the patients who will spontaneously recover versus those who will develop ischemic manifestations following portal venous embolism. Until more definitive, prospective studies
Clinical Toxicology vol. 48 no. 6 2010
538 are done, we feel a prophylactic approach utilizing HBO therapy is reasonable.
Declaration of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this paper.
References 1. Anonymous. Hydrogen peroxide and gas embolism. Lancet 1967; 2(7521):875. 2. Watt BE, Proudfoot AT, Vale JA. Hydrogen peroxide poisoning. Toxicol Rev 2004; 23:51–57. 3. Rider SP, Jackson SB, Rusyniak DE. Cerebral air gas embolism from concentrated hydrogen peroxide ingestion. Clin Toxicol 2008; 46:815–818. 4. Beers MH. The Merck Manual. 18th ed., Section 8, Chapter 95. Whitehouse Station, NJ: Merck; 2006:855.
L.K. French et al. 5. Moon JM, Chun BJ, Min YI. Hemorrhagic gastritis and gas embolism after ingestiong 3% hydrogen peroxide. J Emerg Med 2006; 30:403–406. 6. Cina SJ, Downs JC, Conradi SE. Hydrogen peroxide: a source of lethal oxygen embolism. Case report and review of the literature. Am J Forensic Pathol 1994; 15:44–50. 7. Ciechanowicz R, Sein Anand J, Chodorowski Z, Kujawska-Danecka H. Acute intoxication with hydrogen peroxide with air emboli in central nervous system—a case report. Przegl Lek 2007; 64:339–340. 8. Vidil L, Racioppi L, Biais M, Revel P, Sztark F. Iatrogenic gas embolism after the use of hydrogen peroxide. Ann Fr Anesth Reanim 2008; 27:735–737. 9. Mullins ME, Beltran JT. Acute cerebral gas embolism from hydrogen peroxide ingestion successfully treated with hyperbaric oxygen. J Toxicol Clin Toxicol 1998; 36:253–256. 10. Vander Heide SJ, Seamon JP. Resolution of delayed altered mental status associated with hydrogen peroxide ingestion following hyperbaric oxygen therapy. Acad Emerg Med 2003; 10:998–1000. 11. Ashdown BC, Stricof DD, May ML, Sherman SJ, Carmody RF. Hydrogen peroxide poisoning causing brain infarction: neuroimaging findings. AJR 1998; 170:1653–1655. 12. Luu TA, Kelley MT, Strauch JA, Avradopoulos K. Portal vein gas embolism from hydrogen peroxide ingestion. Ann Emerg Med 1992; 21:1391–1393. 13. Langell JT, Mulvihill SJ. Gastrointestinal perforation and the acute abdomen. Med Clin North Am 2008; 92:599–625.
Clinical Toxicology (2010) 48, 539–544 Copyright © Informa UK, Ltd. ISSN: 1556-3650 print / 1556-9519 online DOI: 10.3109/15563650.2010.494610
ARTICLE LCLT
Acute intentional toxicity: endosulfan and other organochlorines VIJU MOSES1 and JOHN VICTOR PETER2 Acute endosulfan poisoning
1 2
Department of Medicine, Christian Medical College, Vellore, Tamil Nadu, India Medical Intensive Care Unit, Christian Medical College & Hospital, Vellore, Tamil Nadu, India
Introduction. Organochlorine pesticides continue to be used in several developing countries despite concerns regarding their toxicity profile. Endosulfan is an organochlorine recognized as an important agent of acute toxicity. Methods. In this retrospective study, the clinical features, course, and outcomes among patients with acute endosulfan poisoning requiring admission to the hospital during an 8-year period (1999–2007) were reviewed. Results. Among 34 patients hospitalized during this study period for alleged organochlorine poisoning, 16 patients with endosulfan poisoning were identified. The majority (75%) received initial treatment at a primary or secondary center. Neurological toxicity predominated, particularly low sensorium (81%) and generalized seizures (75%), including status epilepticus (33%). Other features observed included hepatic transaminase elevation, azotemia, metabolic acidosis, and leukocytosis. Mechanical ventilation was required in 69% and vasoactive agents in 19%. In-hospital mortality was 19%. There were no gross neurological sequelae at discharge. In three other patients who presented with organochlorine poisoning, the compounds ingested were lindane, endrin, and dicofol (n = 1 each). The course and outcomes in these patients were unremarkable and all three patients survived. Conclusions. Endosulfan is capable of high lethality and significant morbidity. The commonest manifestations are neurological although other organ dysfunction also occurs. In the absence of effective antidotes, restriction of its availability, along with prompt treatment of toxicity, including preemptive anticonvulsant therapy are suggested. Keywords
Mortality; Pesticide; Seizure; Organ failure
Introduction Organochlorine compounds are popular agricultural pesticides in Asian countries. Use of these compounds, although inexpensive and effective, has been restricted in North America and Europe due to concerns regarding their environmental persistence, accumulation in the food chain, and toxicity.1 Acute toxicity is due to accidental or suicidal exposure; organochlorine compounds are absorbed from the skin, lungs, and gastrointestinal tract. Cyclodienes are a class of organochlorines causing life-threatening acute toxicity, predominated by convulsions. They include endosulfan, endrin, aldrin, dieldrin, heptachlor, and chlordane.2 Endosulfan is widely used as an agricultural pesticide in countries where it has not been banned and has been identified as an important cause of pesticide-related mortality and morbidity.3,4 Endosulfan and other chlorinated cyclodienes exert their toxicity in humans by causing central nervous system stimulation through antagonism of gamma aminobutyric acidmediated inhibition. It is rapidly absorbed from the intestinal tract and, due to its lipophilic nature when formulated for
Received 10 March 2010; accepted 17 May 2010. Address correspondence to Viju Moses, Department of Medicine, Christian Medical College, Vellore, TN 632004, India. E-mail:
[email protected]
pesticidal use, is also toxic by absorption through the skin.2 Clinical and pathological reports and experiments have also highlighted injury to other organs due to acute endosulfan exposure, including hepatic and renal dysfunction, hemodynamic instability, and metabolic derangement.5–7 Although endosulfan and other organochlorine compounds have been in use since the 1940s,2 the high case fatality of acute poisoning by these compounds in India has received attention only in recent years.3 There is sparse literature on the course and outcome of this poisoning particularly in light of better intensive care over the last couple of decades. Our study was conducted to characterize the clinical presentation, course, and outcome in patients presenting with moderate and severe endosulfan toxicity needing in-patient care. We also report the data on toxicity due to other organochlorine compounds that required in-patient management during the study period.
Methods We identified the records of all patients with organochlorine poisoning admitted to our tertiary referral center in South India between May 1999 and July 2007. Documentation of acute organochlorine exposure in the Emergency Department (ED) case sheet was usually dependent on the patient or his/ her relatives a) bringing the used pesticide container to the
Clinical Toxicology vol. 48 no. 6 2010
540 hospital, b) naming the compound involved, or c) identifying one of the organochlorine containers kept among the insecticide containers on display in the ED. Patients with moderateto-severe toxicity likely to need more than 24 h of observation were admitted to the medical ward or the ICU. Records of in-patients with toxicity due to endosulfan were selected for detailed evaluation. Other case notes where the specific organochlorine compound was not recorded and patients with organochlorine poisoning other than endosulfan were scrutinized only for important clinical features and outcome variables. This segregation was done because different classes of organochlorines exert toxicity by different pathways, and endosulfan is the commonest organochlorine causing acute severe toxicity.3 Patients discharged after observation in the ED were excluded. Given the retrospective nature of the study, it was exempt from assessment by the Institutional Review Board. Data on clinical, laboratory, electrocardiographic, and radiological findings, toxins responsible, treatment, and outcome were extracted from the ED and inpatient charts and electronic medical records. Organ involvement was identified from the clinical and laboratory findings based on the following definitions: 1) central nervous system involvement: seizures or low sensorium (Glasgow coma scale 12 or less);8 2) hepatic dysfunction: alanine aminotransferase elevation more than twice the upper limit of normal, or a combined increase in aspartate aminotransferase, alkaline phosphatase, and total bilirubin, provided one of them was twice the upper limit;9 3) renal impairment: serum creatinine greater than 1.4 mg/dL.10 A reasonable temporal association was sought before attributing an organ dysfunction to endosulfan exposure. The time of consumption of the toxin was obtained from the patient if sensorium permitted, or from their attendants who brought them to our center. The time of loss of consciousness and seizures was similarly obtained from the attendants. Although this data had the potential to be inaccurate, it was felt useful to estimate the latency to the onset of neurological features needing urgent intervention, an issue of practical importance. As organophosphate poisoning is more common than organochlorine poisoning in our center, every effort was made to exclude patients with possible organophosphate poisoning. Patients who were brought with alleged organochlorine poisoning who manifested components of the “SLUDGE” (salivation, lacrimation, urination, defecation, gastrointestinal upset, emesis) symptoms11 were excluded if pseudocholinesterase levels were low. Patients who did not have any SLUDGE symptoms and who manifested symptoms and signs of organochlorine poisoning generally did not have a pseudocholinesterase level measured.
Statistical aspects Numerical data were processed to obtain relevant rates and ratios. Tests of significance were applied to identify associations between various factors and outcome variables. Fisher’s
V. Moses and J.V. Peter exact test was used for dichotomous variables, and the Wilcoxon rank-sum test for continuous variables as the data was nonparametric. For all statistical tests, a p-value <0.05 was considered to be statistically significant. The R statistical package12 was used.
Results Thirty-four patients were admitted with a diagnosis of organochlorine poisoning. Eight patients with clinical features suggesting organophosphate poisoning and a low pseudocholinesterase level, seven patients listed as organochlorine poisoning without mention of the specific compound, and three patients with documentation of an organochlorine other than endosulfan were excluded. The remaining 16 patients had documentary evidence of endosulfan poisoning and fulfilled the inclusion criteria. The study cohort predominantly consisted of young patients (median age 24.5 years) and had a female:male ratio of 0.78:1. All were suicidal ingestions and agricultural workers (31.3%) were the commonest victims (Table 1). Twelve (75%) patients were referred from a primary or secondary facility, whereas four (25%) presented directly to our center. Survival was similar in these two groups (odds ratio 0.62; 95% confidence interval: 0.02–47; p = 1). Treatment Table 1. Baseline characteristics Variablea Median age (years) Sex ratio, female:male Occupation Agricultural laborer Housewife Student Unemployed Not specified Treated elsewhere prior to referral Median time to presentation (hours) At primary center, N = 5 Range (hours) At our center, N = 12 Range (hours) Treatment given elsewhere, N = 12 Gastric lavage Induced emesis Induced fluids Atropine Pralidoxime Anticonvulsant Median volume of OC consumed, mL, N = 7 Range (mL)
Rateb 24.5 (22–30) 0.78:1 5 (31.25%) 3 (18.75%) 2 (12.5%) 2 (12.5%) 4 (25%) 12 (75%) 2 (2–2) 1–4 3 (2.69–4.25) 1–6 6 (50%) 1 (8.33%) 1 (8.33%) 3 (25%) 2 (16.67%) 1 (8.33%) 100 (50–100) 30–100
DDT, dichlorodiphenyltrichloroethane; OC, organochlorine. a Denominators are specified where less than 16. b Continuous variables are presented as median (interquartile range) and rates are presented as number of patients (percentages).
Clinical Toxicology vol. 48 no. 6 2010
Acute endosulfan poisoning
541
outside included induced emesis, gastric lavage (using a nasogastric feeding tube), intravenous fluids, atropine, and pralidoxime. Treatment with atropine was not associated with increased mortality (odds ratio 0.17; 95% confidence interval: 0.002–4.38; p = 0.21) whereas both patients who received pralidoxime elsewhere before referral survived to discharge from hospital. Gastric lavage using a wide bore orogastric tube (with endotracheal intubation when indicated for protection of the airway) was performed at our center for eight (50%) patients, and activated charcoal was administered to nine (56.3%) patients. Neurological features were the commonest manifestations of endosulfan poisoning (Table 2). Depressed sensorium occurred in 13 (81.3%) patients about 2.5 h after ingestion. Endosulfan ingestion was associated with seizures in 12 (75%) patients. The first seizure occurred about 2 h after ingestion with a range of 0.5–6 h. All seizures described were generalized, and two-thirds were either multiple or status epilepticus. Among patients who had seizures, 66.7% received two or more anticonvulsants, whereas 25% received three or more. Phenytoin (83.3%) and benzodiazepines (58.3%) were the most frequently used agents. Two (16.7%) patients with seizures also received the nondepolarizing neuromuscular blocker pancuronium (Table 3). Besides pancuronium, one received phenytoin, phenobarbital, and diazepam, and the other received phenytoin, midazolam, propofol, and thiopen-
Table 2. Neurologic findings among patients with endosulfan poisoning Variablea Low sensorium (GCS ≤ 12) Median time to loss of consciousness (hours), N = 12 Range (hours) Seizures Seizures at or prior to presentation, N = 12 Median time to first seizure (hours), N = 11 Range (hours) Number of seizures, N = 12 Single Multiple Status epilepticus Neuro-ocular findings at presentation Miosis Mydriasis Other neurologic manifestations Hypertonia Tremor Fasciculations Urinary retention
Rateb 12 (81.25%) 2.5 (1.88–3.25) 0.5–6 12 (75%) 12 (100%) 2 (1–3) 0.5–6 4 (33.33%) 4 (33.33%) 4 (33.33%) 1 (6.25%) 6 (37.5%) 0 0 0 1 (6.25%)
GCS, Glasgow coma scale. a Denominator indicated where less than 16. b Continuous variables are presented as median (interquartile range) and rates as number of patients (percentages).
Table 3. Anticonvulsants and muscle relaxants for seizures Drug Anticonvulsants Phenytoin Phenobarbital Valproate Propofol Thiopentone Any benzodiazepineb Midazolam Diazepam Lorazepam Clobazam Muscle relaxant Pancuronium Number of anticonvulsants per patienta Zero One Two Three Four Five Median anticonvulsants per patient Range a
Frequencya 10 (83.33%) 2 (16.67%) 0 (0%) 1 (8.33%) 1 (8.33%) 7 (58.33%) 3 (25%) 3 (25%) 3 (25%) 1 (8.33%) 2 (16.67%) 1 (8.33%) 3 (25%) 5 (41.67%) 1 (8.33%) 1 (8.33%) 1 (8.33%) 2 (1–2.5) 0–5
N = 12 (total number of patients who had seizures). refers to patients who received one or more benzodazepines.
b
tone. Six (37.5%) patients (54.6% of those ventilated) required ventilatory support due to seizures or low sensorium. There were no gross neurological sequelae at discharge. Cardiovascular findings included sinus tachycardia in 11 (68.8%) patients and hypotension in 3 (18.8%). Hepatic involvement was characterized by serum transaminase and bilirubin elevation in four (36.4%) and one (9%) patients, respectively, of the 11 tested. Five (31.3%) patients had renal impairment during their hospital stay, including four (25%) at presentation. Metabolic acidosis was a common finding, manifesting in 10 (76.9%) of 13 patients. Leukocytosis [14 (87.5%) patients] was the commonest hematological abnormality (Table 4). Nosocomial infections complicated the course of 43.8% of patients and ventilator-associated pneumonia (23.1%) was the commonest cause. Endotracheal intubation and ventilatory support were necessary in 11 (68.8%) patients, either for cardiorespiratory decompensation (n = 5) or for low sensorium including seizures (n = 6). Eleven (68.8%) patients needed ICU admission. The median [interquartile range (IQR)] duration of ventilatory support, intensive care, and hospital stay were 4 (IQR 2–9), 4 (IQR 2–8), and 4.5 (IQR 3–10) days, respectively, whereas the mortality in hospital was 18.8% (Table 5). Important clinical features and treatment modalities were examined for associations with mortality; however, no signif-
Clinical Toxicology vol. 48 no. 6 2010
542
V. Moses and J.V. Peter Table 4. Nonneurologic manifestations of endosulfan poisoning a
Variablea
Rateb
b
Variable
Cardiovascular Hypotension needing inotropes At presentation Any time during hospitalization Sinus tachycardia at presentation Sinus bradycardia at presentation Ventricular extrasystole ST–T wave changes Hepatic Any hepatic involvement, N = 11 Raised bilirubin, N = 1 Raised hepatic enzymes, N 11 Renal Azotemia at presentation Azotemia at any point in hospital Metabolic acidosis, N = 13 Haematologic Elevated hemoglobin at presentation Elevated WBC count at presentation Any lab coagulopathy, N = 8 Raised PT, N = 8 Raised PTT, N = 8
Table 5. Events in hospital, iatrogenic complications, outcome
Rate
1 (6.25%) 3 (18.75%) 11 (68.75%) 1 (6.25%) 2 (12.5%) 1 (6.25%) 4 (36.36%) 1 (9.09%) 4 (36.36%) 4 (25%) 5 (31.25%) 10 (76.9%) 2 (12.5%) 14 (87.5%) 1 (12.5%) 0 (0%) 1 (12.5%)
PT, prothrombin time; PTT, partial thromboplastin time; WBC, white blood cell. a Denominator indicated where less than 16. b Rates are presented as number of patients (percentages).
icant differences between survivors and nonsurvivors were identified in these parameters (Table 6). Besides the 16 patients who consumed endosulfan, the specific organochlorine was not named in the records of seven patients, among whom five had seizures, one died, and one
Gastric lavage Activated charcoal Atropine Ventilatory support Indication for mechanical ventilation, N = 11 Cardiorespiratory decompensation Neurologicc Median days on ventilator, N = 11 Range (days) Number of patients needing ICU care Median days in ICU, N = 11 Median days in ICU hospital Range (days) Mortality Iatrogenic complications Any VAP UTI CRBSI DVT or PE
8 (50%) 9 (56.25%) 11 (68.75%) 11 (68.75%) 5 (45.45%) 6 (54.55%) 4 (2–9) 2–12 11 (68.75%) 4 (2–8) 4.5 (3–10) 1–19 3 (18.75%) 7 (43.75%) 5 (31.25%) 3 (18.75%) 1 (6.25%) 0 (0%)
CRBSI, catheter related blood stream infection; DVT, deep vein thrombosis; ICU, intensive care unit; PE, pulmonary embolism; UTI, urinary tract infection; VAP, ventilator associated pneumonia. a Denominator is indicated where less than 16. b Continuous variables are presented as median (interquartile range) and rates as number of patients (percentages). c includes low sensorium due to organochlorines or durgs for seizures.
was discharged home on request in a moribund state. In addition, there was one patient each of dicofol, endrin, and lindane poisoning, all of whom survived. The patient with lindane toxicity had one episode of generalized tonic–clonic seizure at presentation.
Table 6. Comparison of survivors and non-survivors Variablea Low sensorium (GCS ≤ 12) Time to loss of consciousness (h), N = 12 Status epilepticus Azotemia at presentation Time to presentation (h), N = 12 Treatment elsewhere prior to referral Use of atropine Activated charcoal Gastric lavageb Hypotension needing inotropes Nosocomial infections Ventilator associated pneumonia a
Non-survivors, n = 3 2 (66.67%) 2.25 (1.88–2.63) 2 (66.67%) 0 (0%) 2.38 (2.06–2.69) 2 (66.67%) 1 (33.33%) 0 (0%) 1 (33.33%) 1 (33.33%) 1 (33.33%) 2 (66.67%)
Survivors, n = 13
Odds ratio
95% Cl
P-Value
11 (84.62%) 2.5 (2–3.75) 2 (15.38%) 4 (36.36%) 3.5 (3–4.75) 10 (76.92%) 10 (76.92%) 9 (69.23%) 10 (76.92%) 2 (18.18%) 6 (46.15%) 3 (23.08%)
0.39 N/A 8.86 0 N/A 0.62 0.17 0 0.17 2.55 0.6 5.76
0.13–32.08 –4.5–2.5 0.33–700.9 0–7.91 –4.25–2 0.02–47 0.002–4.38 0–1.64 0.002–4.39 0.31–78.65 0.008–14.46 0.23–429.86
0.5 0.74 0.14 0.53 0.38 1 0.21 0.06 0.21 0.49 1 0.21
Denominator is indicated where less than 16. Includes gastric lavage at primary or referral centre. Fisher's exact test is used for dichotomous variables, with proportions shown as percentages (e.g. low sensorium was present in 66.67% of non-survivors and 84.62% of survivors). Wilcoxon rank-sum test is used for continuous variables, with median and interquartile range displayed. Cl, Confidence interval. b
Clinical Toxicology vol. 48 no. 6 2010
Acute endosulfan poisoning
Discussion Our study confirms the high morbidity and mortality of moderate-to-severe endosulfan poisoning.3,4 Endosulfan was the most common organochlorine used for deliberate self-harm. There is wide variation in the incidence of acute endosulfan poisoning, from 1.8% of all poisonings in South Korea4 to 13.3% in Andhra Pradesh, India.3 Our cohort’s mortality rate was 18.8% whereas other investigators report varying figures ranging from nil in Turkey13 to 30.7% in South Korea.4 Another study from India reported a mortality of 28% with endosulfan.3 The variation in mortality and other statistics among studies can be explained by the severity of cases (district hospital versus referral center), the categories of patients included in the study (all patients versus only in-patients), and the mean quantity of toxin ingested in each series. In the Korean study with a higher mortality (30.7%), 40.4% of patients had consumed more than 100mL of endosulfan.4 In our study, although a large proportion of patients (75%) were first taken to the primary centers close to the site of exposure, only one patient among the 12 attending a primary center received anticonvulsant drugs at the initial point of care. Activated charcoal was not administered at the primary health care centers. At our center, some patients received gastric lavage beyond 1h post-ingestion. Given their apparent lack of efficacy in this cohort (albeit not randomized), the current skepticism toward gastric lavage,14 and the caution advised with activated charcoal,15 gastric lavage, and activated charcoal administration should probably be restricted to patients presenting early (<1h). Gastric lavage must be avoided at centers where facilities to secure an unprotected airway are lacking, which is a common situation in rural India. The use of atropine (25% at primary centers; 68.8% at our center) and pralidoxime (16.7% at primary centers) in our cohort is not surprising. This may be due to a) the failure to distinguish organochlorine from organophosphate poisoning based on clinical features and the names on pesticide containers or b) clinicians generalizing the treatment of organophosphate poisoning to all pesticide poisonings, because organophosphate poisoning is so common in our country. The use of these antidotes in the setting of endosulfan poisoning has not received attention from other investigators. In this cohort, the use of atropine had no adverse effect on survival, length of hospital stay, risk of seizures, or the need for ventilatory support, and both patients who had received pralidoxime survived with no adverse effects. Data on the doses used for these drugs were not available. We are unaware of any reports of seizures precipitated by either drug in humans or animals. However other theoretical concerns (particularly muscle weakness and renal impairment with pralidoxime) necessitate caution in the use of atropine and pralidoxime. Since early therapy of organophosphate poisoning with atropine may reduce the likelihood of respiratory failure,16 atropine use at a primary health center may be justified in situations where the clinical features are equivocal and the exact toxin ingested is
543 not known. This situation is probably not uncommon because eight patients in our original cohort, who were brought in with alleged “organochlorine poisoning,” manifested some symptoms and signs of organophosphate poisoning and had suppressed pseudocholinesterase levels. However, awareness of the clinical features of organochlorine poisoning may reduce the inappropriate use of these drugs with their attendant side effects (e.g., delirium and tachycardia with atropine). The cyclodiene endosulfan was the commonest compound ingested and was responsible for all the three deaths where a specific compound was documented. Although endosulfan is listed as the World Health Organization (WHO) chemical hazard category II based on experimental toxicity in rats,17 Srinivas Rao et al. reported a mortality in humans of 28% with endosulfan compared to 5% with endrin, a WHO Class Ib compound.3 In our study, a median dose of 100 mL (range 30–100 mL) of endosulfan produced the moderate-to-severe toxicity that we encountered. The concentration of endosulfan used in India is 35g/100 mL. Records of two of the three nonsurvivors had no data on the volume of ingested compound, probably because low sensorium precluded self-reporting. The incompleteness of these data prevented us from correlating the volume of toxin with survival or clinical manifestations. Moon et al. identified the ingestion of more than 35g (100 mL) of endosulfan to be an independent risk factor for death.4 In our study, all seizures were generalized starting around 2h after the reported time of ingestion. Considering the high probability of seizures (75%) and the high probability of patients with seizures needing at least two or three anticonvulsants (66.7% and 25%, respectively), prophylactic therapy with an anticonvulsant may be advisable for patients with symptomatic organochlorine poisoning, with aggressive escalation of anticonvulsant therapy with second or third line drugs if seizures are not controlled. This approach might reduce the incidence of secondary complications such as brain injury, renal dysfunction, and acidosis. Our data were not suitable for comparison among anticonvulsants with regard to efficacy and adverse effects: hence, the choice of the first and subsequent anticonvulsants remains to be determined. Neuromuscular blocking drugs were used in two cases of refractory seizures in conjunction with anticonvulsants like thiopentone and propofol along with respiratory support and monitoring, probably to limit renal and metabolic complications from muscle injury, or for securing the airway; however, neuromuscular blockers are not favored by recent authors on status epilepticus, as they may mask on going seizures.18 Renal failure (25%) is often related to rhabdomyolysis from seizures as well as hypotension. Therefore, aggressive hydration and control of seizures might prevent or mitigate renal failure. Hypotension in endosulfan poisoning has been hypothesized to be due to a direct toxic effect on the myocardium,5 the use of anticonvulsant drugs,4 or a result of prolonged acidosis. Although some of our patients had hypotension, it did not appear to impact outcome. Ventricular premature beats reported by Moon et al.4 and also evidenced in two of our patients may be related to endosulfan increasing myocardial
Clinical Toxicology vol. 48 no. 6 2010
544 sensitivity to endogenous catecholamines and predisposing to arrhythmias.19 Moon et al. also reported QTc prolongation in 43.6% and ST segment abnormalities in 30.8% not observed in our patients. The relatively early presentation among their cases (mean, 1h post-ingestion), and their higher dose of toxin (40.4% ingested more than 100 mL of endosulfan) may explain this difference.4 The need for mechanical ventilation has ranged from 0% to 76.9% in various studies3,4,13 and probably reflects the spectrum of severity of poisoning. In our study, 68.8% required mechanical ventilation for a median duration of 4 days. Our study has the following limitations. The number of patients in the cohort was small. Because of this, statistical tests may have been underpowered to reveal the associations that were actually present. Multivariate analysis also could not be done for the same reason. Analysis of the relation between survival and interventions like gastric lavage and activated charcoal is more reliably done in randomized controlled trials, and our report of these estimates must be considered preliminary (Table 6). Deficiencies in documentation limited the analysis of the response (or lack of it) to each anticonvulsant drug. Assay by gas–liquid chromatography20 would have enabled the evaluation of toxic and lethal blood levels; as such data in humans are lacking.21
Conclusions Endosulfan is an underrecognized agent for deliberate selfharm, capable of multiorgan toxicity and high lethality. The high incidence of refractory seizures, hypotension, respiratory failure, and nosocomial infections contribute both to morbidity and the case fatality rate, and need to be considered when evaluating the burden of endosulfan poisoning. Since no effective antidote is available, restriction of access to toxic compounds or substitution with less toxic ones must be explored. Such restriction of the sale of endosulfan in Sri Lanka resulted in lower case fatality from pesticide poisoning over time.22 Prophylactic anticonvulsant therapy for symptomatic patients and aggressive treatment for those manifesting seizures may limit morbidity. Further study is needed to determine the optimal regimen of anticonvulsant drugs for this condition.
Declaration of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this paper.
Acknowledgements We are grateful to the heads of units of the Department of Medicine, Dr OC Abraham, Dr K Thomas, and Dr P Mathews
V. Moses and J.V. Peter for permitting the review of patient records necessary for this work.
References 1. Peter JV, Cherian AM. Organic insecticides. Anaesth Intensive Care 2000; 28(1):11–21. 2. Smith AG. Toxicity of organochlorine insecticides. In: Pesticide Toxicity and International Regulation. West Sussex, UK: John Wiley and Sons; 2004:27–84. 3. Srinivas Rao C, Venkateswarlu V, Surender T, Eddleston M, Buckley NA. Pesticide poisoning in south India: opportunities for prevention and improved medical management. Trop Med Int Health 2005; 10(6):581–588. 4. Moon JM, Chun BJ. Acute endosulfan poisoning: a retrospective study. Hum Exp Toxicol 2009; 28(5):309–316. 5. Eyer F, Felgenhauer N, Jetzinger E, Pfab R, Zilker TR. Acute endosulfan poisoning with cerebral edema and cardiac failure. J Toxicol Clin Toxicol 2004; 42(6):927–932. 6. Kucuker H, Sahin O, Yavuz Y, Yürümez Y. Fatal acute endosulfan toxicity: a case report. Basic Clin Pharmacol Toxicol 2009; 104(1):49–51. 7. Yavuz Y, Yurumez Y, Kücüker H, Ela Y, Yüksel S. Two cases of acute endosulfan toxicity. Clin Toxicol (Phila) 2007; 45(5):530–532. 8. Teasdale G, Jennett B. Assessment of coma and impaired consciousness: a practical scale. Lancet 1974; 2(7872):81–84. 9. Bénichou C. Criteria of drug-induced liver disorders. Report of an international consensus meeting. J Hepatol 1990; 11(2):272–276. 10. Salmanullah M, Sawyer R, Hise MK. The effects of acute renal failure on long-term renal function. Ren Fail 2003; 25(2):267–276. 11. Kwong TC. Organophosphate pesticides: biochemistry and clinical toxicology. Ther Drug Monit 2002; 24(1):144. 12. R Development Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2009. http://www.R-project.org. Accessed 13 January, 2010. 13. Karatas AD, Aygun D, Baydin A. Characteristics of endosulfan poisoning: a study of 23 cases. Singapore Med J 2006; 47(12):1030–1032. 14. Vale JA, Kulig K. Position paper: gastric lavage. J Toxicol Clin Toxicol 2004; 42(7):933–943. 15. Chyka PA, Seger D, Krenzelok EP, Vale JA. Position paper: singledose activated charcoal. Clin Toxicol (Phila) 2005; 43(2):61–87. 16. Eddleston M, Buckley NA, Eyer P, Dawson AH. Management of acute organophosphorus pesticide poisoning. Lancet 2008; 371(9612):597–607. 17. WHO. The WHO Recommended Classification of Pesticides by Hazard. http://www.who.int/ipcs/publications/pesticides_hazard/en/. Accessed 30 January 2010. 18. Prasad K, Krishnan PR, Al-Roomi K, Sequeira R. Anticonvulsant therapy for status epilepticus. Br J Clin Pharmacol 2007; 63(6):640–647. 19. Holland M. Insecticides: organic chlorines, pyrethrins/pyrethroids, and DEET. In: Flomenbaum N, Goldfrank L, Hoffman R, Howland M, Lewin N, Nelson L, eds. Goldfrank’s Toxicologic Emergencies. New York: McGraw-Hill; 2006. 20. Reigart JR, Roberts JR. Recognition and Management of Pesticide Poisonings. 5th ed. Washington, DC: United States Environmental Protection Agency; 1999. http://www.epa.gov/pesticides/safety/healthcare. Accessed 1 February 2010. 21. Bernardelli BC, Gennari MC. Death caused by ingestion of endosulfan. J Forensic Sci 1987; 32(4):1109–1112. 22. Roberts DM, Karunarathna A, Buckley NA, Manuweera G, Sheriff MHR, Eddleston M. Influence of pesticide regulation on acute poisoning deaths in Sri Lanka. Bull World Health Organ 2003; 81(11):789–798.
Clinical Toxicology (2010) 48, 545–549 Copyright © Informa UK, Ltd. ISSN: 1556-3650 print / 1556-9519 online DOI: 10.3109/15563650.2010.495346
ARTICLE LCLT
Ingestion of a newly described North American mushroom species from Michigan resulting in chronic renal failure: Cortinarius orellanosus BRYAN S. JUDGE1, JOSEPH F. AMMIRATI2, GARY H. LINCOFF3, JOHN H. TRESTRAIL III4, and P. BRANDON MATHENY5 Toxinology
1
Department of Emergency Medicine, Michigan State University College of Human Medicine, Spectrum Health-Toxicology Services, Grand Rapids, MI, USA 2 Department of Biology, University of Washington, Seattle, WA, USA 3 The New York Botanical Garden, Bronx, NY, USA 4 Center for the Study of Criminal Poisoning, Grand Rapids, MI, USA 5 Department of Ecology and Evolutionary Biology, University of Tennessee, Knoxville, TN, USA
Background. Some mushrooms in the genus Cortinarius are well known to cause acute and chronic renal failure. Until now, there have been no confirmed cases of renal failure due to the ingestion of a Cortinarius mushroom in North America. We describe a case of a woman who ingested mushrooms found under an oak tree in western Michigan and developed chronic renal failure. Methods. Phylogenetic analysis of the internal transcribed spacer (ITS) regions of nuclear-encoded ribosomal RNA was performed between an unconsumed sample of the Michigan specimens, a control sample of Cortinarius orellanus (JFA9859) from Europe, and other closely related ITS sequences of Cortinarius retrieved from GenBank. An additional gene region, rpb2, was also sequenced for comparison. Results. Phylogenetic analysis revealed the Michigan material to be closely related to, but distinct from, other ITS sequences of the Orellani clade in Cortinarius. Divergence is less at the rpb2 locus. No historical taxa from North America are known to match the identification of the Michigan material. Conclusion. The mushrooms ingested by the patient were confirmed to be a new species of Cortinarius closely related to C. orellanus. We introduce a newly described North American species, Cortinarius orellanosus, capable of causing renal failure after ingestion. Keywords
Cortinarius; Mushroom; Renal failure; Poisoning; Orellanine
Introduction Orellanine-containing mushrooms in the genus Cortinarius, including Cortinarius orellanus Fries and Cortinarius rubellus Cooke, can result in acute and chronic renal failure after ingestion. Numerous human poisonings from Cortinarius have been documented in Europe.1,2 However, confirmed cases of renal failure in North America attributable to the ingestion of Cortinarius mushrooms have not been well documented. A case of mushroom ingestion in the Pacific Northwest causing renal failure has been previously ascribed to C. orellanus,3 since C. orellanus and C. rubellus were the only-known nephrotoxic mushrooms. However, one of the authors of that report felt that the ascription was erroneous.4 It is likely that the mushroom responsible was Amanita smithiana, which is endemic to west-
Received 26 February 2010; accepted 18 May 2010. Address correspondence to Bryan S. Judge, Toxicology Services, Grand Rapids Medical Education Partners, Michigan State University Program in Emergency Medicine, 1900 Wealthy Street SE, Grand Rapids, MI 49506, USA. E-mail:
[email protected]
ern North America, and has been implicated in several cases of delayed-onset renal failure.5 Another report from Canada describes a patient who developed renal failure after eating “magic” mushrooms. Although the patient’s renal failure was attributed to the ingestion of Cortinarius mushrooms, no mycological evidence was provided to confirm this association.6 Finally, several cases have been mentioned in the National Poison Data System about exposures to mushrooms containing orellanine; however, detailed information regarding the exposure and mushroom(s) involved is not readily available from these reports.7,8
Case report A 53-year-old female presented to the emergency department (ED) with an inability to urinate. Nine days before presentation, she had ingested the caps and stems of six mushrooms that she had picked in the summertime under an oak tree in her backyard in western Michigan. The mushrooms were cooked in vegetable oil with some onions and shared with a male companion, who did not consume as many of the mushrooms
Clinical Toxicology vol. 48 no. 6 2010
546 as the patient did. Three days after consumption she developed vomiting and diarrhea and was treated with promethazine suppositories that were prescribed by her physician via telephone. Her friend had diarrhea that resolved, but no other symptoms. The patient denied having abdominal pain, back pain, or myalgias. She experienced diminishing urine output over a period of 4 days before coming to the ED for evaluation. Upon presentation, the patient had normal vital signs and physical examination findings. She denied any prior medical problems, had no known drug allergies, did not smoke, rarely drank alcohol, took a multivitamin on a daily basis, and was a strict vegetarian. Four years before the ingestion, the patient’s creatinine and blood urea nitrogen were 1.0 and 20 mg/dL, respectively. Pertinent initial serum chemistries included sodium 126 mEq/L, potassium 6.4 mEq/L, bicarbonate 18 mEq/L, chloride 90 mEq/L, anion gap 18 mEq/L, glucose 109 mg/dL, creatinine 13.8 mg/dL, blood urea nitrogen 88 mg/dL, AST 15 IU/L, and ALT 8 IU/L. Urinalysis revealed specific gravity 1.010, pH 8.0, glucose 70 mg/dL, sodium 113 mEq/L, creatinine 18.3 mg/dL, protein 1,088.0 mg/dL, 38 white blood cells/HPF, and 18 red blood cells/HPF. An electrocardiogram demonstrated no findings consistent with hyperkalemia. The patient was treated with 50 mEq of sodium bicarbonate intravenously and 15 g of sodium polystyrene sulfonate orally. She was admitted in the hospital for urgent hemodialysis, further evaluation, and treatment. Intact samples of the mushrooms ingested were brought to the hospital, photographed, and dried (Fig. 1). Based on the photograph, the mushrooms were preliminarily identified as a Cortinarius mushroom species by a mycologist. Dried specimens were sent to the University of Washington and the University of Tennessee for further analysis. On the fifth day of hospitalization (14 days post-ingestion), a renal biopsy was performed and demonstrated severe interstitial edema, moderate interstitial nephritis, and acute tubular
Fig. 1. Cortinarius orellanosus – aspect and coloration of mature specimens.
B.S. Judge et al.
Fig. 2. Renal biopsy 14 days post-ingestion: Tubulointerstitial lesions with tubular degeneration, interstitial edema, and inflammatory infiltrates (Masson’s trichrome × 200 magnification).
necrosis (Fig. 2). Nine days after presentation, the patient was discharged from the hospital but remained anuric, and on hemodialysis. One year after ingestion her creatinine was 6.9 mg/dL, and she reported that she was suffering from depression and requiring peritoneal dialysis 5 times a week.
Materials and methods An unconsumed sample of the Michigan specimens and the control sample of C. orellanus (JFA9859) collected in Italy were sent to the University of Tennessee for comparative DNA sequence analysis. The fungal collections reside at the University of Washington Herbarium. DNA extraction was performed by removing 10–20 mg of dried material (a pileus wedge) for grinding in a 1.5-mL microtube with 80–100 mg of sterilized sand, a micropestle, and liquid nitrogen. DNA was extracted using a fungal DNA extraction kit manufactured by Omega Bio-Tek Inc. (Norcross, Georgia). Genomic DNA was serially diluted in two successive 1 : 10 dilutions with sterile water. PCR amplifications of the internal transcribed spacer (ITS) region, functionally the bar code locus of choice in fungal molecular systematics, were performed on a C1000 thermocycler manufactured by Bio-Rad (Hercules, CA, USA). A mixture of sterile water and 5X buffer, GoTaq, and dNTPs supplied by Invitrogen Corp. (Carlsbad, CA, USA) was prepared for each dilution of DNA and controls following manufacturer protocols. ITS1F and ITS4 primers were used for PCR amplification.9,10 Products of a single-copy nuclear protein-coding gene, rpb2, were also produced using primers b6F and b7.1R.11 Amplicons were screened in 1% agarose gels with ethidium bromide and viewed on a UV transilluminator.
Clinical Toxicology vol. 48 no. 6 2010
Toxinology Products were screened and purified using QIAGEN PCR purification columns (Valencia, CA, USA). A BigDye Terminator 3.1 Cycle Sequencing kit (Applied Biosystems, Foster City, CA, USA) was used to produce the sequence reactions that were subsequently purified in 96-well Sephadex G-50 columns (General Electric Healthcare, Piscataway, NJ, USA) using separator strips manufactured by Princeton Separations (Freehold, NJ, USA). Sequencing was performed on an ABI 3730 48-capillary electrophoresis genetic analyzer at the Molecular Biology Resource Facility at the University of Tennessee. Sequence chromatograms were inspected and edited using Sequencher 4.8 software (Gene Codes Corp, Ann Arbor, MI, USA). GenBank (NCBI) accession numbers for ITS sequences of both the sample involved in the poisoning case (FJ263280) and the C. orellanus control (FJ263279) were released immediately to the public. BLASTn comparisons of query sequences were performed with sequences available at GenBank, a database of publicly available DNA sequences. Seven taxonomic and environmental ITS sequences were downloaded that were found to correspond to members of the Orellani clade12 or were highest matches to our query sequences: these included accessionslabeled C. rubellus, Cortinarius orellanoides Rob. Henry, and C. orellanus, species involved in delayed-onset renal failure.13 Two sequences of C. rubellus from Scotland were also downloaded from the UNITE database.14 Sequences of Cortinarius bolaris (Pers.: Fr.) Fries were used as outgroups for rooting purposes in phylogenetic analyses of the ITS regions, which were performed using Bayesian Inference in MrBayes 3.1.2, maximum parsimony (MP) in PAUP* [Phylogenetic Analysis using Parsimony (*and Other Methods)] portable version 4.0b10 for Unix, and maximum likelihood (ML) using GARLI (Genetic Algorithm for Rapid Likelihood Inference) v0.951.15–17 Model test 3.7 was used to determine a best-fit model of nucleotide evolution under the Akaike information criterion for Bayesian and ML analyses.18 Thirteen ITS sequences of the Orellani clade, Cortinarius canarius (E. Horak) G. Garnier and two accessions of C. bolaris were aligned using ClustalX 2.0.9 and default alignment settings.19 Output was saved as a nexus file and minor manual alignment adjustments made in MacClade 4.08.20 The final version of the nexus file included 728 nucleotide sites and is available upon request. After excluding the first 52 sites due to unevenness of character sampling, the model best-fit to the alignment was a HKY+I model (Hasegawa–Kishino–Yano model with a proportion of invariable sites), a nucleotide transformation matrix that allows a different substitution rate for transitions relative to transversions (Ti : tv ratio = 1.5469), a proportion of invariable sites (I) = 0.6241, an equal rates parameter for all sites (gamma), and estimated base frequencies. In the Bayesian analysis, tree sampling occurred for one million generations logging trees every 100 generations from two independent runs. The average SD of split frequencies between the two runs reached less than 0.01 after 90,000 generations, an indicator used to determine similarity of tree sampling from different runs. The first 20% of
547 trees sampled (2,000) were burned, which left 8,001 trees from each run. A total of 16,002 trees was combined and used to reconstruct a summary tree and calculate posterior probabilities for each internode. An internode recovered more than 95% of the time has a posterior probability >0.95, an indicator of statistical significance. Statistical strength for internodes was also assessed using MP and ML bootstrapping. In the former, 1,000 bootstrapped pseudoreplicate datasets were created. A heuristic algorithm under the parsimony criterion was used to search for the optimal tree for each replicate with 10 random-addition-sequence replicates using stepwise addition to obtain starting trees for branch swapping. The tree bisection–reconnection algorithm was used for branch swapping with MulTrees set to no, which saves only one of the best trees found during branch swapping. One hundred ML bootstraps were also performed using Genetic Algorithm for Rapid Likelihood Inference following a previously outlined procedure.21 Consensus trees were then created from bootstrapped trees, and 70% recovery for a given internode was considered significant.
Molecular taxonomical and phylogenetical results The ITS sequence of FJ263279 C. orellanus JFA9858 from Italy matched GenBank accession AF389164 C. orellanus from Austria at all nucleotide sites except for one gapped position (99% similarity). The two-spacer regions of the Michigan material implicated in the poisoning case differed at 16 sites (97% similarity) with respect to Austrian C. orellanus. Eight of the site differences represent nucleotide substitutions and eight represent insertion–deletion (indels) events. No closely related rpb2 sequences were detected after BLASTn searches. A pairwise comparison of rpb2 sequences of the Michigan material [GU462004] and C. orellanus from Italy [GU462003] reveals four nucleotide differences between them at three coding sites (all synonymous) and one intron site (in intron4). The Bayesian summary tree of ITS data (Fig. 3) illustrates that the Michigan Cortinarius sample clusters with other species of the Orellani clade and occupies a branch sister to European species of C. orellanus, from which it is genetically distinct.
Taxonomy As per the International Code of Botanical Nomenclature, for the name of a new species to be validly published, it must be printed in hard copy and described in Latin. A holotype must be designated and its place of deposit given. Please see the text available at http://informahealthcare.com/doi/suppl/ 10.3109/15563650.2010.495346 for English version. Cortinarius orellanosus Ammirati and Matheny, sp. nov. Pileus 3–5 cm, umboconvexus, disco latescente et scutescente, superficie sicca, fibrillosus vel squamaceofibrillosus,
Clinical Toxicology vol. 48 no. 6 2010
548
Fig. 3. Phylogenetic tree estimate of evolutionary relationships of isolates in the Orellani clade of Cortinarius inferred by Bayesian analysis. Thick branches indicate significant measures of branch support inferred by Bayesian posterior probabilities and MP and ML bootstrapping.
aurantiacobrunneus vel rubrobrunneus vel flavobrunneus, atrescens cum aetate. Lamellae subdistantes vel distantes, sane crassae, subauranticobrunneae vel brunneae, atrescentes cum aetate. Stipes 6.5–7.5 × 0.8–1.3 cm, basi aucta, superficie sicca, quibusdam fibrilibus superficialibus ad vela, pallidoflavus vel pallido-ochraceus, brunneus vel rubrobrunneus cum aetate. Contextus albidoflavus vel languidochraceus. Sporae (9) 9.5–10 (10.5) × (5.5) 6.5–7 μm, ellipsoideae vel late ellipsoideae, manifeste verrucosae. Terrestris sub quercu. Typus: USA. Michigan. Kent County. Ada, under oak, July 11, 2008, John H. Trestrail III, 07112008 (Holotypus, WTU).
Discussion Cortinarius orellanosus is a medium size gill mushroom that is characterized by a dome-shaped, dry, orange red brown to yellow brown cap; thick, well-spaced, orange brown gills; and a dry, yellowish to reddish brown stalk that is tapered below with some veil fibrils on the surface. The flesh of the mushroom is whitish to yellowish. Species in this group have rust brown spore prints and the spores are ellipsoid and distinctly ornamented. Both C. orellanus and C. orellanosus occur on the ground in association with oak trees, but C. orellanus also occurs with beech and hazel. Additional study is needed to determine the distribution of C. orellanosus in North America. It has likely been previously collected by other workers but not published to date. There are no collections of North American material in the University of Michigan herbarium under the name C. orellanus, and it has not been reported from North
B.S. Judge et al. America. An excellent description and photograph of European C. orellanus can be found in Cortinarius, Flora Photographica.22 Cortinarius rubellus Cooke has been reported from North America23 and is also featured in Cortinarius, Flora Photographica.22 This report details a confirmed case in North America of the ingestion of a Cortinarius mushroom species resulting in renal failure. Phylogenetic analysis revealed that the mushroom ingested is genetically similar to C. orellanus but distinct at two loci. Thus, the Orellani clade comprises at least three species involved in delayed-onset renal failure: a widespread species in Europe and western North America, C. rubellus; a European species, C. orellanus; and a unique North American species, C. orellanosus, implicated in the mushroom poison case reported here. The clinical features described in this case are similar to the main characteristics of Cortinarius species poisoning summarized by Danel et al.2 Our patient experienced vomiting and diarrhea 3 days after ingestion and undoubtedly developed nephrotoxicity in the preceding days before coming to ED. Typically gastrointestinal symptoms develop a few days (median 3 days) after ingestion, followed by delayed acute renal failure that occurs 4–15 days (median 8.5 days) following consumption. It is unclear what role, if any, the patient’s usage of a daily multivitamin had on the development of her renal failure in conjunction with the consumption of the mushrooms. Glucosuria, hematuria, leukocyturia, proteinuria, increased serum creatinine and serum potassium, and metabolic acidosis were present upon admission. A renal biopsy performed 14 days after ingestion showed interstitial edema, interstitial nephritis, and tubular necrosis. These laboratory and histopathological abnormalities are consistent with the primary biological features of the renal phase of Cortinarius species poisoning.2 Additionally, the patient had renal failure 1 year after ingestion that necessitated peritoneal dialysis. Although complete recovery can occur, many patients who have been poisoned by a Cortinarius mushroom species have developed chronic renal failure requiring intermittent renal replacement therapy or a renal transplant.1 It is unknown whether the patient’s companion developed nephrotoxicity. We assume that he did not since he consumed fewer mushrooms than the patient presented here and to our knowledge never sought medical care. The toxic effects of Cortinarius mushroom species seem to be dose-dependent;2 however, there is considerable inter-individual variability in the susceptibility to the toxic effects of these mushrooms.24 Interestingly, an animal study demonstrated that females appear to be more resistant than males to the toxic effects of Cortinarius speciosissimus.25 Our report is limited by the fact that we did not attempt to isolate orellanine from the sample of mushrooms. We felt that this was unnecessary for several reasons. First, most of the available analytical methods to detect this toxin are difficult to perform. Second, these methods are subject to false positives and lack sensitivity.2 Third, we had a limited
Clinical Toxicology vol. 48 no. 6 2010
Toxinology amount of material and felt that it was necessary to preserve the remaining specimens; additional C. orellanosus mushrooms need to be collected before analysis of orellanine can be performed. Finally, the genetic similarity of C. orellanosus to other Cortinarius mushrooms within the Orellani clade implicates orellanine as the putative toxin in this case. Description of C. orellanosus as a new species is supported by the degree of dissimilarity (3%) at the ITS locus. A 3% threshold is regarded as an approximate minimum indicator for species recognition.26 Several but fewer nucleotide differences exist at the rpb2 locus as well. Although it would be desirable to assess the degree of interspecific variation in C. orellanosus, we were not permitted to collect additional specimens from the patient’s backyard.
Conclusions The mushrooms ingested by the patient were confirmed to be a new species of Cortinarius closely related to C. orellanus. We introduce a newly described North American species, C. orellanosus, capable of causing renal failure after ingestion. The suspected toxin responsible for the patient’s renal failure is orellanine because of the genetic similarity of C. orellanosus and other mushrooms within the Orellani clade. Further study is necessary to determine the distribution of C. orellanosus in North America.
Acknowledgments This work was supported partly by funding from Daniel E. Stuntz Memorial Foundation. The Latin description was prepared by Owen M. Ewald, Assistant Professor, Foreign Languages and Literature, Seattle Pacific University. We thank Aaron Wolfenbarger for producing the rpb2 sequences. Molecular laboratory work was funded by the University of Tennessee.
Declaration of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this paper.
References 1. Karlson-Stiber C, Person H. Cytotoxic fungi–an overview. Toxicon 2003; 42:339–349. 2. Danel VC, Saviuc PF, Garon D. Main features of Cortinarius spp. poisoning: a literature review. Toxicon 2001; 39:1053–1060. 3. Moore B, Burton BT, Lindgren J, Rieders F, Kuehnel E, Fisher P. Cortinarius mushroom poisoning resulting in anuric renal failure. Vet Hum Toxicol 1991; 33:369. 4. West PL, Lindgren J, Horowitz BZ. Amanita smithiana mushroom ingestion: a case of delayed renal failure and literature review. J Med Toxicol 2009; 5:32–38.
549 5. Warden C, Benjamin D. Acute renal failure associated with suspected Amanita smithiana ingestions: a case series. Acad Emerg Med 1998; 5:808–812. 6. Raff E, Halloran PF, Kjellstrand CM. Renal failure after eating “magic” mushrooms. Can Med Assoc J 1992; 147:1339–1341. 7. Litovitz TL, Smilkstein M, Felberg L, Klein-Schwartz W, Berlin R, Morgan JL. 1996 annual report of the American Association of Poison Control Centers Toxic Exposure Surveillance System. Am J Emerg Med 1997; 15(5):447–500. 8. Bronstein AC, Spyker DA, Cantilena LR, Green JL, Rumack BH, Giffin SL. 2008 annual report of the American Association of Poison Control Centers’ National Poison Data System (NPDS): 26th annual report. Clin Toxicol 2009; 47(10):911–1084. 9. White TJ, Bruns TD, Lee S, Taylor JW. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis MA, Gelfand DH, Sninsky JJ, White TJ, eds. PCR Protocols: A Guide to Methods and Applications. New York: Academic Press; 1990:315–322. 10. Gardes M, Bruns TD. ITS primers with enhanced specificity for basidiomycetes–application to the identification of mycorrhizae and rusts. Mol Ecol 1993; 2:113–118. 11. Matheny PB. Improving phylogenetic inference of mushrooms using RPB1 and RPB2 sequences (Inocybe, Agaricales). Mol Phylogenet Evol 2005; 35:1–20. 12. Peintner U, Moncalvo J-M, Vilgalys R. Toward a better understanding of the infrageneric relationships in Cortinarius (Agaricales, Basidiomycota). Mycologia 2004; 95:1042–1058. 13. Benjamin DR. Mushrooms Poisons and Panaceas: A Handbook for Naturalists, Mycologists, and Physicians. New York: W.H. Freeman & Company; 1995. 14. Kõljalg U, Larsson K-H, Abarenkov K, Nilsson RH, Alexander IJ, Eberhardt U, Erland S, Høiland K, Kjøller R, Larsson E, Pennanen T, Sen R, Taylor AFS, Tedersoo L, Vrålstad T, Ursing B. UNITE: a database providing web-based methods for the molecular identification of ectomycorrhizal fungi. New Phytol 2005; 166:1063–1068. 15. Ronquist F, Huelsenbeck JP. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 2003; Aug 12:1572–1574. 16. Swofford DL. PAUP*. Phylogenetic Analysis Using Parsimony (* and other Methods). Version 4. 4th ed. Sunderland, MA: Sinauer Associates; 2003. 17. Zwickl DJ. Genetic algorithm approaches for the phylogenetic analysis of large biological sequence datasets under the maximum likelihood criterion [Ph.D. dissertation]. Austin, TX: The University of Texas at Austin; 2006. 18. Posada D, Crandall KA. Modeltest: Testing the model of DNA substitution. Bioinformatics 1998; 14(9):817–818. 19. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG. The ClustalX windows interface: Flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 1997; 25:4876–4882. 20. Maddison DR, Maddison WP. MacClade 4.08: Analysis of Phylogeny and Character Evolution. Sunderland, MA: Sinauer Associates; 2005. 21. Matheny PB, Moreau P-A. A rare and unusual lignicolous species of Inocybe (Agaricales) from eastern North America. Brittonia 2009; 61:163–171. 22. Brandrud TE, Lindström H, Marklund H, Melot J, Muskos S. Cortinarius, Flora Photographica. Matfors, Sweden: Cortinarius HB; 1998. 23. Robertson CP, Wright L, Gamiet S, Machnicki N, Ammirati J, Birkebak J, Meyer C, Allen A. Cortinarius rubellus Cooke from British Columbia, Canada and Western Washington, USA. Pac Northwest Fungi 2006; 1:1–7. 24. Bouget J, Bousser J, Pats B, Ramee MP, Chevet D, Rifle G, Giudicelli CP, Thomas R. Acute renal failure following collective intoxication by Cortinarius orellanus. Intensive Care Med 1990; 16:506–510. 25. Nieminen L, Pyy K. Sex differences in renal damage induced in the rat by the Finnish mushroom, Cortinarius speciosissimus. Acta Pathol Microbiol Scand 1976; [A] 84:222–224. 26. Ryberg M, Nilsson RH, Kristiansson E, Töpel M, Jacobsson S, Larsson E. Mining metadata from unidentified ITS sequences in GenBank: a case study in Inocybe (Basidiomycota). BMC Evol Biol 2008; 8:50.
Clinical Toxicology (2010) 48, 550–558 Copyright © Informa UK, Ltd. ISSN: 1556-3650 print / 1556-9519 online DOI: 10.3109/15563650.2010.491484
ARTICLE LCLT
Systematic differences between healthcare professionals and poison information staff in the severity scoring of pesticide exposures RICHARD DOUGLAS ADAMS, AMANDA L. GIBSON, ALISON MARGARET GOOD, and DAVID NICHOLAS BATEMAN Differences in poisoning severity scoring
NPIS Edinburgh, Scottish Poisons Information Bureau, Royal Infirmary of Edinburgh, Edinburgh, UK
Context. Severity scores are used in triage and for data comparison in cases of poisoning. Exposure severity scores have not been generally validated and their utilization by healthcare staff other than specialists in poison information (SPIs) is untested. Objective. To compare the poisoning severity grading allocated in pesticide exposure cases by healthcare professional enquirers and poison information staff. Methods. Pesticide exposures reported to the U.K. National Poisons Information Service (NPIS) systems in a prospective study were graded for severity by healthcare professional enquirers and NPIS SPIs who used established poisons severity-grading algorithms. The scores were compared in children and adults, for the two professional groupings, both overall and for separate pesticides. Results. Overall SPIs graded severity resulting from pesticide exposure at a lower level than the enquirer. For children, enquirer mean severity score was 1.62 (95% confidence interval (CI) 1.57–1.66) and SPIs mean severity score was 1.16 (95% CI 1.13–1.19) (p < 0.001). For adults, enquirer mean severity score was 1.91 (95% CI 1.84–1.97) and SPIs mean severity score was 1.74 (95% CI 1.69–1.79) (p < 0.001). Importantly, the differences in the scores between the two professional groups were greater in children [+0.46 (95% CI 0.41–0.51)] than in adults [+0.17 (95% CI 0.11–0.24)] (p < 0.001). Findings for individual pesticides were less consistent but in general showed similar trends. The exception was glyphosate for which severity grading by poison information staff was higher for children [SPIs 1.68 (95% CI 1.38–1.96) than the enquirers 1.26 (95% CI 1.08–1.44), p < 0.02]. Conclusions. Our findings suggest inherent differences in the perception of pesticide toxicity between healthcare professionals and SPIs. There was also a difference in the scoring approach depending on the pesticide involved. Additional investigations are required to define the role and accuracy of severity scoring in different types of poisoning and the applicability to different types of severity assessors. Keywords Specialists in poison information; Pesticides; Children; Poisoning; Severity grading
Introduction Poison severity scoring has been advocated as a means of standardizing data sets from different healthcare environments, and may be used as a means of triaging cases at presentation. Originally developed for use internationally so that case data from different countries could be compared,1 an international standard has been published but there is little data on the practical application of this methodology as a harmonization tool. As originally conceived, scores are normally applied at case resolution, and “intended to be an overall evaluation of the case, taking into account the most severe clinical features. Use of the Poisoning Severity Score normally requires a follow-up of all the cases, but may be used Received 5 March 2010; accepted 4 May 2010. Address correspondence to Richard Douglas Adams, NPIS Edinburgh, Scottish Poisons Information Bureau, Royal Infirmary of Edinburgh, 51 Little France Crescent, Edinburgh, EH16 4SA, UK. E-mail:
[email protected]
on admission or at other times during the course of poisoning if this is clearly stated when the data are presented.”1 Pesticides are widely perceived to present a toxic hazard, and management of exposures to pesticides presents a challenge to healthcare professionals, particularly in childhood when there is relatively little information on acute outcomes from these exposures.2 In the United Kingdom, the true health impact of pesticides is unclear. In 2004, the U.K. Health and Safety Executive (HSE) therefore commissioned the National Poisons Information Service Edinburgh Unit (NPISE) to develop systems to monitor U.K. pesticide exposure enquiries to the NPIS. In the United Kingdom, poison enquiries to the NPIS only come from healthcare professionals who may be doctors, nurses, pharmacists, or ambulance personnel. They are from two main sites: emergency departments or public health telephone enquiry-line nurse advisors (NHS Direct in England and Wales or NHS 24 in Scotland). All use the NPIS database TOXBASE or a single national telephone number to triage and manage cases.3
Clinical Toxicology vol. 48 no. 6 2010
Differences in poisoning severity scoring The purpose of this study was, for pesticide exposures, to examine the concordance of severity grading between poison information staff and the healthcare enquirers who use the NPIS. Also it aimed to explore the possibility of systematic differences in the grading of cases as assessed by the different groups of healthcare personnel while using similar patient information.
Methods All U.K. pesticide enquiries to NPISE are followed up by a questionnaire (paper or online). In this study we analyzed returned questionnaires regarding incidents of accidental exposure occurring from July 1, 2006 to June 30, 2009 (3 years). These returns included a severity grading by the responder, based on the data entered into the questionnaire. The question posed was “how serious was the exposure considered at the time of admission/assessment?” This assessment was probably contemporaneous for healthcare professionals using electronic returns, but was based on case records if a postal response was used. The enquiring healthcare professional’s clinical observation was graded by them using a questionnaire based on that of Leverton et al.4 provided by the commissioners of the research (see Appendix NPIS Pesticide Surveillance Questionnaire, QB6). The enquirers were requested to grade exposure severity at admission/presentation in five categories as: Not at all serious, Minor, Moderate, Major, or Uncertain. Specialists in poison information (SPIs) then graded the same cases based on the questionnaire data using the Poisons Severity Score (PSS). These scores were applied to the symptom descriptions on the report forms, applying to the clinical condition as described on the form. The PSS also consists of five categories: “None,” “Minor,” “Moderate,” “Severe,” and “Uncertain,” together with a sixth category – “Fatal.”1 The terms used in the gradings are not identical but sufficiently similar, apart from “Fatal,” which only occurs in PSS to allow comparison. For this study, patients in the PSS “Fatal” and “Severe” groups were amalgamated to equate with the Leverton grade “Major.” Three SPIs from NPISE with scientific degrees were involved in reviewing questionnaires and grading cases using the PSS. An internal validation of the allocated PSS was undertaken to establish uniformity and validate this approach. A sample of 100 cases was assessed blind by two further SPIs and the severity grades were compared against the original PSS using the kappa statistic. To allow statistical analysis, each severity grading was allocated a numeric value of 1–5. Thus, a grading of 1 indicates absence of any symptoms as classified by the PSS, or classified “Not at all serious” on the Leverton scale; 2 indicates mild (transient, spontaneously resolving) symptoms or “Minor;” 3 indicates moderate (pronounced or prolonged) symptoms or “Moderate;” and 4 indicates severe (life-threatening/fatal) symptoms or “Major.” Severity scores for accidental pesticide exposures were assessed separately for adult (>12 years) and childhood (≤12
551 years) exposures. This age distinction was used as it is generally considered that children over 12 years of age are old enough to knowingly perform deliberate self-harm. In addition, for appropriately large groups, exposures were also analyzed by pesticide type or active ingredient. These groups were ant killers, rodenticides, and glyphosate-containing herbicides. The results were analyzed using the GraphPad Prism software version 5.00 for Windows and the Minitab Statistical Software, release 15, for Windows. Means were calculated for the gradings allocated by the SPIs and healthcare professional enquirers in adults and children for each type of pesticide exposure. Mann–Whitney and Spearman’s rank tests were used to compare severity assessments and correlate scores between graded types. In addition, mean differences were compared between enquirers and SPIs to assess the extent to which poison severity assessed in adults and children varied between observers untrained in using severity scores on scene and trained SPIs remote from the patient.
Results A total of 1,567 completed questionnaires regarding accidental human pesticide poisoning were submitted to the NPISE during July 1, 2006 to June 30, 2009. Of these, 280 were graded as “Uncertain” by either SPIs (n = 79) or enquirers (n = 236), and in five enquiries the age of the patient was not specified. These were excluded from this study. Thus, 1,284 questionnaires were available for analysis, with 745 exposures involving children and 539 involving adults. Some exposures involved multiple pesticides and/or multiple patients, but the majority of the cases involved a single acute exposure in a single patient. The majority of the exposures in children were graded by the enquirers (n = 690, 93%) and SPIs (n = 738, 99%) as severity score 1 or 2; they were thus considered to be cases of no or only minor toxicity. Similarly in adults both enquirers (n = 448, 83%) and SPIs (n = 509, 94%) allocated scores of 1 or 2. Nevertheless, there were significant differences between the overall grading determined by the enquirers and the SPIs for both adult and childhood exposures (Table 1). SPIs graded severity resulting from pesticide exposure to be at a lower level than by the enquirer in both children and adults. In children, the enquirer mean severity score was 1.62 (95% confidence interval (CI) 1.57–1.66); the SPI mean severity score was 1.16 (95% CI 1.13–1.19) (p < 0.001). In adults, the enquirer mean severity score was 1.91 (95% CI 1.84–1.97); the SPI mean severity score was 1.74 (95% CI 1.69–1.79) (p < 0.001). Adult pesticide exposures were considered more serious than childhood exposures by both enquirers and SPIs. However, when the mean difference between scoring in adult exposures and childhood exposures is examined (Table 1), it is clear that this difference was greater in children (+0.46, 95% CI 0.41–0.51, p < 0.001) than in adults (+0.17, 95% CI
Clinical Toxicology vol. 48 no. 6 2010
552
R.D. Adams et al.
Table 1. Overall pesticide poisons severity scores (mean and 95% CI) as classified by enquirers and SPIs, difference in the mean scores and statistical significance Category All pesticides: Children (745) All pesticides: Adults (539) Mann–Whitney U value
Enquirer
SPIs
Difference in means
Mann–Whitney U value
Spearman rank (r value)
1.62 (1.57–1.66)
1.16 (1.13–1.19)
+0.46 (0.41–0.51)
169,311 p <0.001
0.14 p < 0.001 (0.06–0.21)
1.91 (1.84–1.97)
1.74 (1.69–1.79)
+0.17 (0.11–0.24)
129,194 p < 0.001
0.32 p < 0.001 (0.24–0.40)
158,251 p < 0.001
95,912 p < 0.001
159,538 p < 0.001
Table 2. Summary of rodenticide and ant killer poison severity scores (mean and 95% CI) as classified by the enquirers and the SPIs; difference in the mean scores; and statistical significance Category Rodenticides: Children (297) Rodenticides: Adults (30) Ant killer: Children (114) Ant killer: Adults (46)
Enquirer
SPIs
Difference in means
Mann–Whitney U value
1.59 (1.51–1.66) 1.57 (1.33–1.80) 1.58 (1.47–1.69) 1.63 (1.45–1.81)
1.05 (1.03–1.08) 1.30 (1.10–1.50) 1.18 (1.11–1.25) 1.72 (1.58–1.85)
+0.54 (0.45–0.62) +0.27 (−0.03–0.56) +0.40 (0.29–0.52) −0.09 (−0.26–0.09)
24,623 p < 0.001 344.5 p = 0.07 n/s 4,158 p < 0.001 946.5 p = 0.31 n/s
0.11–0.24, p < 0.001). This indicates a systematic difference in exposure score assessment in children as compared with that in adults. The Spearman’s rank correlation coefficient showed a statistically positive correlation between the enquirer and SPIs score (Table 1), suggesting a consistency of approach to the scoring systems but a shift in severity assessment, with SPIs tending to assign a lower score in all exposures. However, this difference was more marked for childhood exposures. Interestingly, findings for individual pesticides were less consistent. The rodenticide and ant killer exposures in general shared the trends shown in the overall data for children, with SPIs grading these as less severe (Table 2). In adults, the numbers were smaller and differences not significant, although the trend was toward higher gradings by SPIs. In contrast, children exposed to glyphosate herbicides (n = 27) were found to have a statistically higher severity grading calculated by the SPIs than the enquirers (1.68, 95% CI 1.38–1.96; 1.26, 95% CI 1.08–1.44, respectively) (p < 0.025). For adults (n = 68) exposed to glyphosphate herbicides, the severity score assessed by the SPIs was similar to that assessed by the enquirers and not significantly different (1.94, 95% CI 1.76–2.12; 1.74, 95% CI 1.59–1.88, respectively) (p = 0.14). An internal validation of the SPIs’ ability to assess severity scores was calculated by comparing the scores obtained by two independent assessors on 100 blinded reports against the original PSS. The overall Kendall coefficient between the SPIs (0.93, p < 0.0001), and the overall Fleiss’ Kappa value (0.85, p < 0.0001), indicates that there is a high level of concordance between the SPIs.
Discussion A standardized scheme for grading the severity of poisoning allows qualitative evaluation of the morbidity caused by poisoning and facilitates comparative data analysis.1 Grading systems have been used in clinical toxicology for many years. They are useful in triage and comparison of outcomes in poisoning in order to identify hazards and changes in severity over time.5 The PSS is a severity-grading scale developed by the International Programme on Chemical Safety, the Commission of the European Union, and the European Association of Poison Centres and Clinical Toxicologists (IPCS/EC/EAPCCT) for the assessment of telephone enquiries to poison information services. It was developed to take account of the overall clinical picture1 and has been used to generate a valid comparison regarding severity and outcome among poison centers.6 Recent work however suggests that initial scores may not be as accurate as later assessments in judging the actual severity and in particular underestimate the ultimate toxicity in severe cases.7 This was acknowledged by the authors of the PSS system who recognized its limitations as an outcome predictor.1 In the current work, we have used the PSS to assess severity based on the return data provided by the healthcare professionals. In the present study, the severity score arrived at by the enquirers might be expected to be equivalent to that assigned by the SPIs as both were based on symptom data provided on questionnaires completed by the enquirers. Pesticide exposures in adults tend to be more severe than those in children as scored by the enquirers and SPIs. The likely reason underlying the overall difference in the severity score between the
Clinical Toxicology vol. 48 no. 6 2010
Differences in poisoning severity scoring adults and children is that in general exposures in children are brief, do not take place during application, and take place at home where pesticide products available in the United Kingdom are of lower toxicity. As the majority of the returns were provided on paper, they post-date the original clinical presentation and may therefore include the healthcare professional’s more global assessment rather than a strict point of time assessment. If this were consistently the case, this might potentially bias the healthcare professional’s scoring and thus in part account for the lower severity scores determined by the SPIs as compared with those assessed by the healthcare professionals. The difference would, however, be expected to be of similar magnitude and direction in adults and children. These data show that this is not the case and that scoring of childhood severity was overall systematically greater than adult severity by healthcare professionals than SPIs (Table 1). For the categories of specific pesticides, in which we had sufficient numbers of cases to perform comparisons, we found evidence of differences in scoring depending on the agent involved. Thus for glyphosate herbicides, the SPIs recorded higher toxicity than the enquirers. We have no specific explanation for this finding. We are unable to assess whether it reflects the grading of different professional groupings, and one possibility is that physicians returned scores on more severely poisoned patients. The returns were anonymyzed and data on professional group were not collected. Nevertheless, this raises the possibility that approaches to severity scoring may be influenced by prior knowledge of either the enquirer or the SPI about the particular product involved. It is interesting to consider why there is a systematic difference between the severity scores recorded by two independent groups of observers. Clearly, the SPIs are remote from the patient and may therefore not be as involved in the overall management of the case. They may thus downgrade the symptoms deemed important by the healthcare enquirers and effectively underestimate the true toxicity profile of the compounds involved. In contrast, an alternative explanation is that enquirers overestimate the potential toxicity to a patient. These potential influences might be expected to be reduced as the majority of the questionnaires were completed after patient discharge and involved postal follow-up and case note review. The fact that the difference between severity scoring by the SPIs and the enquirers is far greater in children than in adults tends to suggests that an inherent systematic difference in the perception of severity in different patient groups may be the reason for the differences we have observed. We do not suggest that either group of grader is necessarily better at assessment, but merely that they differ in their approach. Such differences have not been previously described and are likely to be important considerations if the PSS is to be used for harmonization comparisons as originally conceived.1 We are uncertain whether these differences have important implications for acute triage of individual patients, but they obviously have implications if used in epidemiological studies
553 where differences in the severity score might be assumed to indicate real differences in a health effect. One potential difficulty is that the scoring systems used did not have precisely similar terminologies. The Leverton score4 was developed by epidemiologists for use by clinicians reviewing case notes. The PSS was developed by clinical toxicologists and poison information scientists for scoring poison information calls. Nevertheless the words used are identical in the two central categories (“minor” and “moderate”) and only differ at the two extremes. There is obviously concordance on the use of the scoring as reflected in a highly significant Spearman’s rank correlation (Table 1). While the fact that the terminologies might be interpreted differently in children and adults is a possibility, we believe that these findings suggest that there is a systematic difference in the approach of those managing children with pesticide exposures, when assessing their severity, and that this severity is considered more severe than it might have been in the same exposure in an adult.
Limitations This study has obvious limitations. It compares the perception of the severity of the case on the part of the enquirer with the SPIs grading based on the symptoms reported by the enquirer. Most case data were analyzed retrospectively. The results should be interpreted cautiously as the SPIs assessment score is based on information provided by the healthcare enquirer. There was no independent observation of the symptoms reported to the SPIs and seldom any laboratory evidence to confirm exposure. There were 280 questionnaires that could not be assessed for grading severity and were classified as “Uncertain.” It seems unlikely that these (out of a total of 1,567) would skew the results, but this cannot be absolutely refuted.
Conclusions This study compared the severity score as assessed by the enquirers to the U.K. NPIS with the PSS severity score assessed by the SPIs, based on the observations, symptoms, and signs reported by the enquirers using a post hoc questionnaire. These findings suggest a systematic difference in the approach to assessment of severity adopted by the healthcare professional enquirers as compared with that used by the SPIs, which is particularly marked in children. In addition, severity assessment score differences were not uniform for all pesticides, as illustrated by findings for glyphosate, where exposures were rated more serious by the SPIs than the healthcare professionals. Further work is required to define the role and application of severity scoring by different users and across a variety of toxins if this methodology is to be more widely adopted in epidemiological studies of poisoning.
Clinical Toxicology vol. 48 no. 6 2010
554
R.D. Adams et al.
Acknowledgments
References
We acknowledge the support of our colleagues in the NPIS units who contributed by helping to collect the data that were used in the formulation of this article. The work was conceived by DNB, analyzed by the authors, and the initial manuscript was drafted by RDA and AG. DNB is the guarantor of the data. The U.K. Health and Safety Executive and the Chemicals Regulation Directorate funded this study. The funding sponsor had no role in the design of the study other than to request a range of pesticides about which they were specifically interested in gathering exposure data.The study sponsor also played no role in writing the report or the decision to submit the article for publication.
1. Persson HE, Sjoberg GK, Haines JA, Pronczuk de Garbino J. Poisoning Severity Score. Grading of acute poisoning. J Toxicol Clin Toxicol 1998; 36:205–213. 2. Adams RD, Lupton D, Good AM, Bateman DN. UK childhood exposures to pesticides 2004–2007: a TOXBASE toxicovigilance study. Arch Dis Child 2009; 94:417–420. 3. Anonymous. NPIS Annual Report 2008/2009. ISBN 978-0-85951-650–1. http://www.spib.scot.nhs.uk/NPIS%20Annual%20Report%2008-09.pdf. Accessed 9 May 2010. 4. Leverton K, Cox V, Battershill J, Coggon D. Hospital admission for accidental pesticide poisoning among adults of working age in England, 1998–2003. Clin Toxicol 2007; 45:594–597. 5. Marchi AG, Bet N, Peisino MG, Vietti-Ramus M, Raspino M, Di Pietro P, Bernini G, Cantini L, Chiandetti L, Da Dalt L, Crichiutti G, Nocerino A. Severity grading of childhood poisoning: the multicentre study of poisoning in children (MSPC) score. J Toxicol Clin Toxicol 1995; 33:223–231. 6. Casey PB, Dexter EM, Michell J, Vale JA. The prospective value of the IPCS/EC/EAPCCT Poisoning Severity Score in cases of poisoning. J Toxicol Clin Toxicol 1998; 36:215–217. 7. Poynton MR, Bennett HKW, Ellington L, Crouch BI, Caravati EM, Jasti S. Specialist discrimination of toxic exposure severity at a poison control centre. Clin Toxicol 2009; 47:678–682.
Declaration of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this paper.
Clinical Toxicology vol. 48 no. 6 2010
Differences in poisoning severity scoring
555
Appendix: NPIS Pesticide Surveillance Questionnaire Your name: Your Dept/GP practice: Name of the contact for the Dept: Ward or hospital to which patient was transferred: A A1
CONFIRMATION OF HES DATA Please confirm the sex and year of birth of the patient Male
Female Year of birth (not date)
Pregnant? A2
Yes
No
Number of weeks
Please confirm the dates of admission and discharge/death (NB these should refer to the first admission following the poisoning incident. Please ignore any readmissions at this stage). Date of admission d
d
m
m
y
y
y
y
d
d
m
m
y
y
y
y
Date of discharge
B
EXPOSURE ACUTE
B1
CHRONIC
To what pesticide(s) was the patient thought to have been exposed? (Please give the full name or as much detail as you can even if the exact name is not recorded – e.g. rat poison-bromadiolone) or not known
B2
To what type of product was the patient thought to have been exposed? An amateur product for use in the home (inc head lice treatments) An amateur product for use in the garden A professional product for use in agriculture/horticulture A professional product for use in timber treatment A professional product for other use (please specify) A veterinary product Other type of product (please specify) Not known
Clinical Toxicology vol. 48 no. 6 2010
556
R.D. Adams et al.
B3
Specifically, please mark one of the following categories for the type of product involved? Rat/mouse killer
Fly/wasp killer
Slug/snail killer
Ant killer
Head lice treatment
Not known
Other insecticide
Herbicide
Other
B4
How certain was the exposure? Definite
B5
Possible
What was the route of exposure (please tick all that apply) Amount (mL ) Ingestion
B6
Duration (h)
Time since (h)
N/A
Inhalation
N/A
Skin contact
N/A
Eye
N/A
Other
N/A
Not Known
N/A
N/A
How serious was the exposure considered at the time of admission? Not at all serious
Minor
Moderate
Major Uncertain
B7
Was the patient exposed as a result of deliberate self-harm? No
If yes please go to Q B13
Yes
Clinical Toxicology vol. 48 no. 6 2010
Differences in poisoning severity scoring
B8
557
Was the patient using the pesticide themselves at the time he/she was exposed? No
B9
Yes
Was someone else using the pesticide at the time the patient was exposed? No
B10
Yes
Yes
Not known
Was the patient exposed as a consequence of unsatisfactory storage or transport of the pesticide? No
B12
Not known
Was the patient exposed as a consequence of use of the pesticide, but not exposed at the time of the application (e.g. after timber treatment)? No
B11
Not known
Yes
Not known
Was the patient exposed in the course of his/her job? No
Yes
Not known
B13
Briefly describe in a sentence or more the circumstances of the exposure (e.g. where, why and how).
B14
Were there any factors or circumstances other than the pesticide exposure that may have contributed to the illness of the patient? (e.g. viral infection) No
Yes
(please give details)
Not known C
CLINICAL DETAILS
C1
Was a specific biochemical test carried out? (e.g. paraquat or organophosphate assay) No What were the results?
Yes
Clinical Toxicology vol. 48 no. 6 2010
558
R.D. Adams et al.
C2
What were the main clinical abnormalities at the time of first admission? (with brief details of any other test results or investigations if available) No abnormalities
C3
Were there any treatments required related to the pesticide exposure? No
C4
Observation only
Yes
Yes
Yes
(please give details)
Was the patient subsequently readmitted to hospital in relation to the poisoning? No
C8
(please give details)
Were there any long-term effects following discharge that could be attributed to this pesticide exposure? No
C7
(please give details)
Were there any other significant complications during the admission? No
C6
(please give details)
Was there any need for ventilation and/or intensive care? No
C5
Yes
Yes
(please give details)
Death Did the patient die? No
Yes
Cause of death?
Thank you for answering our questions. Please return this form in the FREEPOST envelope provided
Clinical Toxicology (2010) 48, 559–562 Copyright © Informa UK, Ltd. ISSN: 1556-3650 print / 1556-9519 online DOI: 10.3109/15563650.2010.497149
ARTICLE LCLT
Validation of the American Association of Poison Control Centers out of hospital guideline for pediatric diphenhydramine ingestions VIKHYAT SUGYANI BEBARTA1, HOLLY W. BLAIR2, DAVID L. MORGAN3, JOSEPH MADDRY4, and DOUGLAS J. BORYS2 Validation of AAPCC guideline for diphenhydramine
1
Department of Emergency Medicine, Wilford Hall Medical Center, U.S. Air Force, San Antonio, TX, USA Central Texas Poison Center, Temple, TX, USA 3 Department of Emergency Medicine, Scott and White Memorial Hospital, Temple, TX, USA 4 Wilford Hall Medical Center, SAUSHEC Residency, San Antonio, TX, USA 2
Context. In 2006, the American Association of Poison Control Centers (AAPCC) published an out of hospital guideline for diphenhydramine overdoses in children. This guideline has not been validated. Objective. Our objective was to determine the incidence of serious clinical effects or use of medical treatments after unintentional diphenhydramine ingestions in children. We sought to determine if patients with less than 7.5 mg/kg ingestions developed medical complications of diphenhydramine toxicity. Materials and methods. In our observational case series, we searched 7 years of data (2000–2006) in the Texas Poison Center Network for diphenhydramine using the AAPCC generic codes. We included only acute, single ingestions of diphenhydramine in children under 6 years old. We included only patients with a recorded weight, known amount of ingestant, and known follow-up. We defined “serious clinical effects” as hallucinations, seizure, wide QRS on electrocardiogram, wide complex dysrhythmia, any conduction block, hypotension, hypertension, rhabdomyolysis, pyrexia, dystonia, coma, respiratory depression, or death. One trained abstractor reviewed the data and entered it into an electronic data collection form. Twenty percent of the charts were audited for abstractor agreement. Results. Our search resulted in 928 cases. Of these, 305 were included in our study. Of the patients who ingested doses less than 7.5 mg/kg, 99.7% (299/300) did not require critical treatments or were without serious clinical effects. One child was admitted. Five children ingested doses of more than 7.5 mg/kg. All five were observed in the emergency department and discharged home. Two patients had serious clinical effects of hallucinations, one of which ingested more than 7.5 mg/kg. No child required critical treatments. Our agreement on chart review for 20% of the cases was very good for “serious clinical effects” (kappa, 0.79; 95% CI, 0.39–1.0) and excellent for “critical treatments” (kappa, 1.0). Conclusion. Based on our observational case series, 99.6% of patients who reportedly ingested doses less than 7.5 mg/kg did not develop serious clinical effects or require admission. Pediatric ingestions over 7.5 mg/kg were uncommon in our study population. Serious clinical effects, critical treatments, and admission from diphenhydramine were rare in children under 6 years old. Keywords
Diphenhydramine; Overdose; Children; Poison center; Guideline
Introduction In 2008, more than 25,700 ingestions of diphenhydramine were reported to poison centers in the United States.1 Of these, 15,300 were under 6 years old.1 Although most cases result in sedation in children, large doses can cause seizures, wide complex dysrhythmias, coma, and death.2–5 In 2006, the American Association of Poison Control Centers (AAPCC) published an out of hospital guideline for diphenhydramine.6 The guideline was based on a review of peerreviewed literature and expert opinion panel. The guideline is a triage tool intended to be used by poison center personnel. Received 18 January 2010; accepted 25 May 2010. Address correspondence to Vikhyat Sugyani Bebarta, Department of Emergency Medicine, Wilford Hall Medical Center, U.S. Air Force, 23239 Crest View Way, San Antonio, TX 78261, USA. E-mail:
[email protected]
The panel advised that children under 6 years old who ingest at least 7.5 mg/kg should be referred to a health-care facility.6 This guideline has not been validated in a cohort of patients. Our objective was to determine the incidence of serious clinical effects or use of critical medical treatments after unintentional diphenhydramine ingestions in children. We evaluated 7 years of exposures at six poison centers within the Texas Poison Center Network (TPCN). We sought to determine if patients with ingestions less than 7.5 mg/kg developed serious medical complications of diphenhydramine.
Methods We performed a retrospective study within the TPCN. Our observational case series was approved by The Scott and White Memorial Hospital institutional review board. The
Clinical Toxicology vol. 48 no. 6 2010
560 TPCN receives voluntary phone calls from patients and health-care providers. The network is composed of six poison centers that provide clinical consultative services throughout Texas 24 h per day. The network covers a population of 24 million people. Nurses or pharmacists trained in toxicology receive each call for the purpose of managing the exposure and documenting the consultation. All cases are recorded on Toxicall, a standardized database. These cases are periodically uploaded and combined with data from other poison centers. The aggregate national data are maintained by the AAPCC. Cases are followed until the outcome of the event is known or as long as possible if the outcome is not known. We searched 7 years of data (January 1, 2000–December 31, 2006) in the TPCN for human cases of diphenhydramine with the AAPCC generic codes. All data for the six TPCN poison centers are stored in one database at the Central Texas Poison Center in Temple, Texas, and the TPCN staff maintains it. We identified and included only acute, single ingestions of diphenhydramine in children under 6 years old. We only included patients with a recorded weight, known amount of ingested drug, and known follow-up as identified by the poison center chart. The TPCN did not have a standard clinical protocol for these exposures during the years of the study. The trained poison information specialists enter clinical notes and document standardized codes for common symptoms, signs, and treatments. Free text clinical notes are also entered. If a patient is not referred to a health-care facility, the poison specialist will call the patient back within 1–3 h to obtain follow-up and determine resolution of symptoms. If the patient goes to the hospital, the specialist continues to follow the case by phone and collects medical information and course from the patient’s health-care provider. We abstracted the clinical notes for our outcomes. Our study involved cases that were managed prior to the publication of the AAPCC diphenhydramine guideline.
Data collection and processing Symptoms, signs, treatments, and outcomes were extracted from the database through chart review of the clinical notes and coded diagnoses and treatments. Outcomes coded as “No effect,” “Minor,” “Moderate,” “Major,” and “Death” according to the American Association of Poison Centers’ National Poison Data System outcome criteria were not used to categorize patients in our study.7 We saved the charts electronically and then printed and reviewed them. One trained abstractor (JM) reviewed the charts and entered the data into a secure spreadsheet. The abstractor reviewed the clinical notes for symptoms, signs, and treatments not coded. One of the investigators (VB) trained the abstractor. The abstractor trained on 50 charts not included in the study cohort. He was given feedback on data abstraction and corrected mistakes. We defined “serious clinical effects” as hallucinations, one or more seizures, wide QRS (>120 ms) on electrocardiogram,
V.S. Bebarta et al. wide complex dysrhythmia, any conduction block, hypotension (systolic blood pressure <80 mmHg), hypertension (systolic blood pressure >115 mmHg), rhabdomyolysis, pyrexia, dystonia, coma, respiratory depression (synonyms – respiratory distress, respiratory failure, “trouble breathing”), or death.8,9 Sedation, lethargy, or agitation were not included as “serious clinical effects” and would not be detected by our study. For a case in which the outcomes were unclear to the primary abstractor, two authors reviewed the case and determined if the clinical effect was present to ensure the broadest capture of cases.8 We defined “critical treatments” as endotracheal intubation, benzodiazepine or other sedative, antiseizure medications (phenytoin, fosphenytoin, phenobarbital, propofol, parenteral valproic acid, or levetiracetam), vasopressor (dopamine, norepinephrine, epinephrine, or vasopressin), physostigmine, antidysrhythmic (lidocaine, amiodarone, procainamide), cardioversion or defibrillation, sodium bicarbonate, or reversal agents (naloxone or flumazenil). We also recorded age, gender, weight, drug ingested, formulation, dose, dose per kilogram, management site, clinical effects, disposition, and if the ingestion was iatrogenic (given by an adult incorrectly). Definitions and outcomes were defined prior to abstraction. Once the chart review began, periodic meetings were done to resolve discrepancies. In addition, the investigator (VB) audited a random sample of 20% of the charts to assess for accuracy. He reviewed the outcomes for the presence of serious clinical effect, critical treatment, or admission. The auditor used the clinical notes of each poison center chart, and was blinded to the National Poison Data System outcome and the abstractors’ data. Our primary outcome was to determine if patients with less than 7.5 mg/kg developed serious clinical effects of diphenhydramine, were administered critical treatments, or were admitted. Our secondary outcomes were the proportion of patients who ingested doses of more than 7.5 mg/kg, highest dose when no serious effects were reported, and which critical clinical effects occurred most commonly. All data were entered into a password-protected electronic spreadsheet (Microsoft Excel; Microsoft Corporation, Redmond, WA, 2007). We performed descriptive statistics.
Results Our search resulted in 928 cases under 6 years old with diphenhydramine ingestion. Of these, 632 had a known weight coded in the chart. Eighty-four of these had an unknown dose and were excluded. We concluded 305 cases that met our remaining inclusion criteria of known follow-up. The mean age was 2.4 years (median 2, range 1 month to 5.5 years). About 55% of patients (167/305) were aged 2 or younger and 52% were female. No deaths were reported. Of the patients who ingested doses less than 7.5 mg/kg, 99.7% (95% CI 98.1–99.9%) (299/300) were without serious clinical effects and did not receive critical treatments. The
Clinical Toxicology vol. 48 no. 6 2010
Validation of AAPCC guideline for diphenhydramine
561
Table 1. Demographics and outcomes of children who ingested more than 7.5 mg/kg Age
Gender
Dose (mg/kg)
3 years 2.5 years 3 years 5 years 3 years
Male Female Female Male Male
18.75 9 7.84 23 8.05
Location of management
Gastric decontamination
Significant clinical effects
Significant treatments
Disposition
HCF HCF HCF HCF HCF
Charcoal Charcoal Charcoal Charcoal Charcoal
Hallucinations (and tremor) None None None None
None None None None None
Observation and discharge Observation and discharge Observation and discharge Observation and discharge Observation and discharge
HCF, health-care facility.
median dose of patients with ingestions less than 7.5 mg/kg was 1.6 mg/kg (mean 1.85 mg/kg, range 0.16–7.35 mg/kg). One child was admitted. His dose was iatrogenic, given by the patient’s father who was a physician. The dose ingested was recorded as 1.6 mg/kg. He did not have serious clinical effects and only mild sedation was recorded. Five children ingested more than 7.5 mg/kg. One had serious clinical effects of hallucinations (Table 1). All were observed in the emergency department and discharged home. None required critical treatments. None of the children managed at non-health-care facilities (282/305), such as home or day care, were reported to the poison center as having been sent in later to a health-care facility for evaluation. Of the patients who were evaluated at a health-care facility (24/305, 7.8%), 42% (10/24) received oral decontamination with activated charcoal. One other child had serious clinical effects (hallucinations) and the child was managed at home. The child ingested 0.92 mg/kg based on the parent’s history. No patient in our study received critical treatments. We also evaluated iatrogenic ingestions as a subset of the 305 patients. In these acute ingestions, another person gave diphenhydramine to the child. Five cases were iatrogenic. The mean ingested dose was 2.4 mg/kg (median, 2.5 mg/kg; range, 1.6–2.9 mg/kg); none ingested a dosegreater than 7.5 mg/kg. One case was from a physician error, one was from an incorrectly filled prescription, and one was from an incorrect dose given by a clinical nurse. As already described, this child was admitted for sedation after ingesting a dose of 1.6 mg/kg. The remaining were observed in the emergency department and none received oral decontamination. Our agreement on chart review for 20% of the cases was very good for “serious clinical effects” (kappa, 0.79; 95% CI, 0.39–1.0) and excellent for “critical treatments” (kappa, 1.0). No patients in the audited sample were admitted.
Discussion In our study, we found that using an ingested diphenhydramine dose of less than 7.5 mg/kg was safe to keep a child at home. Of the 300 children with a dose of more than 7.5 mg/kg, 1 patient (0.3%; 95% CI, 0.01–1.84) developed serious clinical
effects. One child below this dose developed hallucinations, which we categorized as a serious clinical effect. In addition, one child who received less than 7.5 mg/kg was admitted. However, the child was the son of a physician, and the physician had administered the excessive dose. Thus, the treatment team may have changed their typical care plan. Thus, 99.6% of cases with ingestion less than 7.5 mg/kg were kept at home with reported serious complications. Only five cases received more than 7.5 mg/kg orally. None of the patients were admitted or received critical treatments, although one developed serious clinical effects of hallucinations. One child was reported to have ingested 23 mg/kg and did not develop symptoms. As it is only one case, no conclusion regarding the dose can be drawn from this ingestion. In addition, the dose estimate was possibly an error. If the guideline has been adhered to for all patients, 19 of the 24 cases evaluated at a hospital would have been kept at home, resulting in an 80% reduction in cases referred to a health-care facility. Although many of these patients were already at the hospital, the poison center was called. Although serious clinical effects and admission were rare, few cases had any symptoms or ingested doses more than 7.5 mg/kg. The only serious effect reported was hallucination. Thus, the AAPCC guideline of not sending children who ingested less than 7.5 mg/kg to a health-care facility is likely safe based on our study. Although a higher threshold could be considered, more cases with ingestions greater than 7.5 mg/kg and cases with serious clinical effects are needed to challenge the current guideline sufficiently. In addition, our data are limited by information reported to poison centers, thus the weight or dose ingested could vary from the reported information. Three of the five reported iatrogenic dosing occurred by medical care providers. In two of these cases, the incorrect unit was dispensed (“mg” vs. “mL”). Education of healthcare providers or diphenhydramine manufacturers could reduce this error. Our study had several limitations. First, the dose ingested by the child, if any, is usually estimated by the parents and cannot be verified by medical staff. Our study is a retrospective database review and contains the inherent limitations of this type of study.10 However, we attempted to mitigate the retrospective study design flaws with strict methodology using a
Clinical Toxicology vol. 48 no. 6 2010
562 trained abstractor, holding scheduled meetings, using a standardized data form, and measuring interrater agreement.10 Our study is also limited by the use of TPCN records and strict categories for clinical effects. Only patients and healthcare providers who called the poison center were evaluated, possibly limiting the spectrum of cases reviewed. Patients may have visited a health-care facility and not called a poison center. Some of these cases may have serious clinical effects or required admission. Hospital and clinic medical records were not available for review, limiting conclusions on systemic findings, severe sedation, electrocardiogram changes, and length of observation in the emergency department. In addition, with poison center records we cannot be certain about exam findings, abnormal vital signs, and whether treatments were not missed. However, we used the clinical effect categories listed by the poison control center as well as the clinical notes to determine outcomes and clinical course. This method has been used for other studies.11 Many cases were managed at home and not evaluated by a health-care provider, thereby limiting physical assessment. Some of these patients may have had an adverse outcome after initial consultation with the poison center and they did not report it. However, if a serious clinical effect or admission did occur, it is likely rare. Patients may have visited a health-care facility and received treatment despite a recommendation by the poison center staff to stay home. We would not have captured these treatments and outcomes in our study if the patient visited the facility after the poison specialists’ follow-up phone call.
Conclusions Based on our retrospective study from six poison centers over 7 years, 99.6% of patients who ingested less than 7.5 mg/kg of diphenhydramine did not develop serious clinical effects or require admission. The results may not be generalizable to other health-care settings. Adherence to the AAPCC diphenhydramine guideline by the TPCN could have reduced the patients sent to a health-care facility by 80%. Pediatric ingestions over 7.5 mg/kg are uncommon in our study population. Serious clinical effects, critical treatments, and admission are rare in children under 6 years old with acute diphenhydramine ingestions.
V.S. Bebarta et al.
Declaration of interest The authors have no financial support or financial interest in the subject matter discussed. The views expressed in this article are those of the authors and do not reflect the official policy or position of the Department of the United States Air Force, Department of Defense, or the U.S. government.
References 1. 2008 Annual Report of the American Association of Poison Control Centers’ National Poison Data System (NPDS): 26th Annual Report. http://www.aapcc.org/dnn/Portals/0/2008annualreport.pdf. Accessed 14 January 2010. 2. Clark RF, Vance MV. Massive diphenhydramine poisoning resulting in a wide-complex tachycardia: successful treatment with sodium bicarbonate. Ann Emerg Med 1992; 21:318–321. 3. Aaron FE. A case of acute diphenhydramine hydrochloride poisoning. Br Med J 1953; 2:24. 4. Hestand HE, Teske DW. Diphenhydramine hydrochloride intoxication. J Pediatr 1977; 90:1017–1018. 5. Baker AM, Johnson DG, Levisky JA, Hearn WL, Moore KA, Levine B, Nelson SJ. Fatal diphenhydramine intoxication in infants. J Forensic Sci 2003; 48:425–428. 6. Scharman EJ, Erdman AR, Wax PM, Chyka PA, Caravati EM, Nelson LS, Manoguerra AS, Christianson G, Olson KR, Woolf AD, Keyes DC, Booze LL, Troutman WG. Diphenhydramine and dimenhydrinate poisoning: an evidence-based consensus guideline for out-of-hospital management. Clin Toxicol (Phila) 2006; 44:205–223. 7. Bronstein AC, Spyker DA, Cantilena LR Jr, Green JL, Rumack BH, Heard SE, American Association of Poison Control Centers. 2007 Annual report of the American Association of Poison Control Centers’ National Poison Data System (NPDS): 25th Annual Report. Clin Toxicol (Phila) 2008; 46:927–1057. 8. AAPCC National Poison Data System (NPDS) Reference Manual. Washington, DC: June 2007. 9. National High Blood Pressure Education Program Working Group on High Blood Pressure in Children and Adolescents. The fourth report on the diagnosis, evaluation, and treatment of high blood pressure in children and adolescents. Pediatrics 2004; 114:555–576. 10. Gilbert EH, Lowenstein SR, Koziol-McLain J, Barta DC, Steiner J. Chart reviews in emergency medicine research: where are the methods?. Ann Emerg Med 1996; 27:305–308. 11. Ngo A, Ciranni M, Olson KR. Acute quetiapine overdose in adults: a 5-year retrospective case series. Ann Emerg Med 2008; 52:541–547.
Clinical Toxicology (2010) 48, 563–565 Copyright © Informa UK, Ltd. ISSN: 1556-3650 print / 1556-9519 online DOI: 10.3109/15563650.2010.490222
SHORT REPORT LCLT
Asymptomatic congenital lead poisoning − case report NACHAMMAI R. CHINNAKARUPPAN1 and STEVEN MATTHEW MARCUS2 Case report: congenital lead poisoning
1
Neonatal Intensive Care Unit, Department of Pediatrics, Lehigh Valley Health Network, Allentown, PA, USA New Jersey Poison Information and Education System, New Jersey Medical School, University of Medicine and Dentistry of New Jersey, Newark, NJ, USA
2
Context. Congenital lead poisoning is uncommon and there is no consensus on the management of the newborn. Case Details. A female infant was born to a lead-burdened woman identified by screening just prior to delivery. Maternal blood lead level (BLL) was 58 µg/dL. The infant’s BLL on the second day of life was 72 µg/dL with a free erythrocyte protoporphyrin level of 175 µg/dL. The child was managed by an exchange transfusion followed by chelation. The BLL 6 h after exchange transfusion was 11.4 µg/dL. Follow-up 2 years later showed a BLL of 9 µg/dL and normal development. Discussion. We present the details of a case of congenital lead poisoning treated aggressively which appears to have resulted in a favorable outcome. Keywords
Lead poisoning; Pregnancy; Newborn; Exchange transfusion; Management
Introduction Lead poisoning is the most common, preventable, environmental cause of neurotoxicity in children. Lead toxicity is associated with impaired cognitive, motor, behavioral, and physical abilities.1 Children are not usually identified as being lead burdened until they are screened, generally at 12 months of age, unless there are overt signs of lead poisoning. There is no consensus on the management of these infants. We report a case of congenital lead poisoning and delineate our management strategies.
Case report A 23-year-old woman presented for the first time to the obstetrics clinic at 36 weeks gestation. A blood lead level (BLL) was obtained together with her routine prenatal screening because of her immigrant status. Her BLL was 58 μg/dL and hemoglobin 9.5 g/dL. She was prescribed ferrous sulfate and seen again in a week when her BLL was noted to be 57 μg/dL. She was not referred for evaluation or treatment. She spontaneously delivered a 38-week infant weighing 3,030 g (growth percentile 50%), length 50 cm (growth percentile 75%), and head circumference of 33 cm (growth percentile 50%). The baby’s Apgar scores were 9 at 1 min and 9 at 5 min
Received 21 January 2010; accepted 28 April 2010. Address correspondence to Nachammai R. Chinnakaruppan, Neonatal Intensive Care Unit, Department of Pediatrics, Lehigh Valley Health Network, Cedar Crest &I-78, PO Box 689, Allentown, PA 18105-1556, USA. E-mail:
[email protected]
of life. The general physical and neurological exams were normal. The newborn received routine care. The newborn’s BLL, drawn on the second day of life, was 72 μg/dL. Further testing revealed the hemoglobin was 17 g/ dL and free erythrocyte protoporphyrin level 175 μg/dL; the peripheral smear showed no basophilic stippling. Using the analogy of hyperbilirubinemia in the presence of what was believed to be a potentially encephalopathogenic level of blood lead, a decision was made to rapidly lower the circulating level of lead in the newborn by performing a double-volume exchange transfusion on the fourth day of life. The repeat BLL was not immediately available after the exchange transfusion and a decision was made to institute parenteral chelation therapy in an effort to decrease whatever total body burden of lead the newborn may have had and prevent rebound elevation of her BLL. It was subsequently learned that her BLL 6 h after the exchange transfusion was 11.4 μg/dL. Considering the controversy about potential for translocation of lead with chelation, dimercaprol (BAL) was chosen as the first-line chelating agent. She received six doses of intramuscular BAL (3 mg/kg/ dose). With the second dose of BAL, a continuous infusion of intravenous calcium disodium ethylenediaminetetraacetate (CaNa2EDTA 50 mg/kg/day or 0.7 mg/h) was started and continued for 2 days. The newborn was then started on oral succimer 50 mg three times daily for 19 days once the CaNa2EDTA was discontinued. The preparation of succimer was made at bedside by mixing the preweighed drug in water which was then titrated with sodium bicarbonate until it dissolved visually. Serum electrolytes were monitored twice weekly and showed no abnormalities. The infant continued to feed well and gain weight and showed no neurological deficits. We discouraged breastfeeding to decrease exposure to lead.2
Clinical Toxicology vol. 48 no. 6 2010
564
Discussion The BLL in U.S. children aged 1–5 years has decreased from 15 to 1.9 μg/dL in the past three decades because of the legislative policies that have decreased environmental sources of lead.1 However, cases of lead poisoning continue to be reported. Reports of neonatally discovered lead poisoning, such as ours, have been reported in the literature and management in these cases varied. In a case series resulting from a 25-year literature review, Shannon4 identified 15 women with elevated lead levels during pregnancy. The mean maternal BLL was 55 ± 19 μg/dL and corresponding mean neonatal BLL was 74 ± 44 μg/dL. Thirteen of these babies received CaNa2EDTA, BAL, and succimer or a combination therapy. There are two other reports of exchange transfusion for neonatal lead poisoning in the literature by Mycyk et al.5 and Hamilton et al.6 We opted to treat this newborn with a double-volume exchange transfusion despite the risks of the procedure, to rapidly decrease the lead burden. This decision was made as the result of consultation between a neonatologist and a toxicologist. The decision was based upon the fact that although a given blood lead may not represent a good surrogate marker for total body lead burden, it most likely does represent the lead that is highly toxic to the individual. Thus, there is an apparent similarity between lead and bilirubin in the neonates. A standard two-volume exchange would remove approximately 85% of the red cells in circulation
80 70 Blood lead level (mcg/dL)
The health department found multiple sources of lead including cooking utensils, a kettle, and a herbal remedy which the mother drank during pregnancy. The mother had pica evidenced by eating dirt from unwashed beans, chewing on flip tops from beer cans, and chewing on costume jewelry. She admitted to eating Mexican candy that tested positive for lead. These items were removed by the health department and replaced with lead-free items. The infant was discharged home on day 25 of life, with a BLL of 15 μg/dL. Discharge medication included ferrous sulfate. On her 2-year follow-up visit, she was found to be healthy, not taking any medications. There was originally a question of unilateral neural deafness at 1 year but retesting at 2 years found normal hearing. Her physical exam showed weight at 10%, length 50%, and head circumference 25%. Her tone and reflexes were normal. Her Bayley Scales of Infant and Toddler Development-111 assessment showed cognitive score of 135 and motor score of 103 showing normal development to date. The Bayley-111 is a norm-referenced developmental scale of cognitive and motor development used during the first 48 months, which has good psychometric properties and has been extensively used in the follow-up of preterm babies. A score of 85–115 is 1 SD around the mean and is considered normal.3 The institutional review board gave verbal approval for the manuscript to be published.
N.R. Chinnakaruppan and S.M. Marcus
60 50 40 30 20 10 0 0
1
2
3
4
5 6 7 Age (months)
8
9
12
24
Fig. 1. Infant blood lead levels.
before the exchange and reduce the serum indirect bilirubin level by one half. We hoped to achieve the same reduction in BLLs. The risks associated with an exchange transfusion include infection, bleeding, hyperkalemia, hypocalcemia, hypothermia, hypoglycemia, and arrhythmias and risks associated with insertion of umbilical lines (thrombosis, poor perfusion of extremities). We felt that the benefits of optimizing neurodevelopment by acutely lowering the baby’s BLL outweighed the risks of the procedure. We recognize the limitation in our management in that we opted not to measure trace metals because of the volume of blood which would be needed and the potential need to transfuse the baby if that route had been taken. As the monitoring of urinary excretion of lead by the baby needed an indwelling catheter, we also elected not to do this because of risks of infection and relied on the drop in BLL as a surrogate marker for drop in total body lead burden. The infant’s BLL (see Fig. 1) shows a brief elevation post-chelation presumably because of the BLL reequilibrating in the body between the bone, blood, and soft tissue compartments.
Sources of lead As this child was found to have an elevated lead level at birth, the external environment could not have been directly implicated in the child’s lead poisoning. The logical explanation for the child’s exposure was that it occurred transplacentally. Thus, we understood the importance of locating the source of mother’s exposure and an attempt was made to define the mother’s source of lead. In the review by Shannon,4 the top three causes of lead poisoning during pregnancy were identified as intentional pica, home renovations, and the use of dietary supplements. In our case report, the mother initially denied all of these exposures. The health department then performed a home inspection for lead and did find multiple sources of lead in the family’s home, including cooking utensils, a kettle, and herbal remedy that the mother took during pregnancy.
Clinical Toxicology vol. 48 no. 6 2010
Case report: congenital lead poisoning
Lead in pregnancy Although the mean BLL in the United States has decreased drastically in the past three decades, 0.5% women of childbearing age still have BLLs above 10 μg/dL. Lead is freely transferred from maternal to fetal circulation through the placenta by passive diffusion. The fetal membranes absorb and eliminate lead from maternal circulation and this accumulates with time in the fetal compartment.7 A high lead burden in pregnant women can result in babies that are small for gestational age from intrauterine growth restriction and adverse neurological outcomes. We do acknowledge that some of the studies that report this association may not have adequately controlled for other compounding variables.7 Gardella7 in a literature review proposed an argument for routine prenatal screening of pregnant women to remove lead sources during the prenatal period to prevent maternal and neonatal morbidity. Unfortunately, there are no uniform guidelines on the appropriate management of elevated lead levels found in pregnant women. Chelation during pregnancy appears well tolerated by mothers but it may mobilize maternal lead stores and potentially increase transmission to the fetus.8 Currently, there are no national guidelines for screening pregnant women.
Conclusion There is no consensus on the management of lead toxicity in newborns. We treated this newborn with a double-volume exchange transfusion to optimize neurodevelopment.
565 Although it is rewarding to know that she is doing well thus far, her long-term outcome remains to be determined.
Declaration of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this paper.
References 1. Binns HJ, Campbell C, Brown MJ, Advisory Committee on Childhood Lead Prevention. Interpreting and managing blood lead levels of less than 10 μg/dl in children and reducing childhood exposure to lead: recommendations of the Centers for Disease Control and Prevention Advisory Committee on Childhood Lead Poisoning Prevention. Pediatrics 2007; 120(5):e1285–e1298. 2. Baum CR, Shannon MW. Lead in breast milk. Pediatrics 1996; 97:932. 3. Bayley N. Bayley Scales of Infant and Toddler Development III. San Antonio, TX: Harcourt Assessment; 2005. 4. Shannon M. Severe lead poisoning in pregnancy. Ambul Pediatr 2003; 3(1):37–39. 5. Mycyk MB, Leikin JB. Combined exchange transfusion and chelation therapy for neonatal lead poisoning. Ann Pharmacother 2004; 38:821–824. 6. Hamilton S, Rothenberg SJ, Khan FA, Manalo M, Norris KC. Neonatal poisoning from maternal pica behavior during pregnancy. J Natl Med Assoc 2001; 93:317–319. 7. Gardella C. Lead exposure in pregnancy: a review of the literature and argument for routine prenatal screening. Obstet Gynecol Surv 2001; 56(4):231–238. 8. Horowitz BZ, Mirkin DB. Lead poisoning and chelation in a mother-neonate pair. Clin Toxicol 2001; 39(7):727–731.
Clinical Toxicology (2010) 48, 566–568 Copyright © Informa UK, Ltd. ISSN: 1556-3650 print / 1556-9519 online DOI: 10.3109/15563650.2010.496730
SHORT REPORT LCLT
Use of a lipid emulsion in a patient with refractory hypotension caused by glyphosate-surfactant herbicide SANG KYOON HAN1, JINWOO JEONG2, SEOKRAN YEOM1, JIHO RYU3, and SUNGWOOK PARK1 Lipid emulsion in glyphosate-surfactant poisoning
1
Department of Emergency Medicine, Pusan National University Hospital, Seo-Gu, Busan, Republic of Korea Department of Emergency Medicine, Medical Research Institute of the Pusan National University Hospital, Seo-Gu, Busan, Republic of Korea 3 Department of Emergency Medicine, Pusan National University Yangsan Hospital, Seo-Gu, Busan, Republic of Korea 2
Context. Circulatory shock is a major cause of mortality in glyphosate-surfactant herbicide (GlySH) poisoning, and this condition responds poorly to conventional therapies. We report a case of GlySH poisoning with shock that was refractory to vasopressors but responsive to intravenous fat emulsion (IFE). Case details. A 52-year-old man was brought to the emergency department by ambulance. He was found unconscious in his living room along with an empty bottle of GlySH herbicide, which contained glyphosate, polyoxyethyleneamine (POEA) surfactant, and water. He was drowsy at presentation. His heart rate was 44 beats/min, his blood pressure could not be measured with an arm cuff, but he had a palpable femoral pulse. After about 2.5 h of supportive care after admission, he remained hypotensive, and his systolic blood pressure was 80 mmHg. A 500 mL bottle of 20% IFE product was prepared. As a bolus, 100 mL of IFE was injected, and the remaining 400 mL was then infused. His blood pressure was 100/60 mmHg 1 h after the bolus injection. At 5 h after IFE injection, his blood pressure reached 160/100 mmHg and vasopressors were tapered. Conclusion. IFE should be considered in cases of refractory hemodynamic instability caused by GlySH after aggressive fluid and vasopressor support. Keywords Acute poisoning; Antidote; Glyphosate; Surfactant; Fat emulsion
Glyphosate is a nonselective herbicide commonly used in agriculture worldwide.1 Although glyphosate has a low toxicity in mammals, mortality following the ingestion of the glyphosatesurfactant herbicide (GlySH) product is significant.1–3 Circulatory shock is a major cause of mortality in GlySH poisoning,1 and this condition responds poorly to conventional therapies, such as fluid resuscitation or vasopressors.4,5 No specific antidote exists that is effective against shock caused by GlySH poisoning. Emerging evidence suggests that intravenous fat emulsion (IFE) can reverse hemodynamically significant poisonings, particularly when the toxin is lipophilic, and IFE can be used as an adjunct therapy in hemodynamically compromised patients.6 We report a case of GlySH poisoning with shock that was refractory to vasopressors but responsive to IFE.
Case report A 52-year-old man was brought to the emergency department by ambulance. He was found unconscious in his living room Received 4 May 2010; accepted 24 May 2010. Address correspondence to Jinwoo Jeong, Department of Emergency Medicine, Pusan National University Hospital, 1-10 Ami-Dong, Seo-Gu, Busan 602-739, Republic of Korea. E-mail:
[email protected]
along with an empty 300 mL bottle of GlySH herbicide, which contained 41% glyphosate as an isopropylamine salt, 15% polyoxyethyleneamine (POEA) surfactant, and water. He was drowsy at presentation, with a Glasgow coma scale score of 11 (E3, V4, M4). His heart rate was 44 beats/min, his blood pressure could not be measured with an arm cuff, but he had a palpable femoral pulse, and his respiratory rate was 15 breaths/min. He was intubated and a mechanical ventilator was used. A central venous catheter was inserted and vasopressor support was initiated with the infusion of dopamine. Atropine was administered to treat bradycardia. Gastric lavage was performed and 50 g of activated charcoal was given through a gastric tube. A chest radiograph showed normal findings and an electrocardiogram revealed atrial rhythm with junctional premature complexes. The results of the arterial blood gas analysis were as follows: pH 7.298; PaCO2, 26.9 mmHg; PaO2, 240.2 mmHg; and HCO3, 13.3 mEq/L. Serum electrolyte concentrations were as follows: Na, 134.4 mEq/L; K, 3.79 mEq/L; and Cl, 94.6 mEq/L. The serum ethanol concentration was 152.8 mg/dL. Other chemistry tests and complete blood count results were within normal limits. After about 2.5 h of supportive care after admission, he remained hypotensive, and his systolic blood pressure was 80 mmHg. The central venous pressure was 30 mmHg and fluid resuscitation was sufficient. Because he was refractory to conventional medical therapies including infusion of dopamine
Clinical Toxicology vol. 48 no. 6 2010
Lipid emulsion in glyphosate-surfactant poisoning and dobutamine, and remained in critical condition, an intravenous lipid emulsion was initiated. A 500 mL bottle of 20% fat emulsion product (Smoflipid; Fresenius Kabi, Bad Homburg, Germany) was prepared. As a bolus, 100 mL of IFE was injected, and the remaining 400 mL was then infused at a rate of 1.5 mL/min. His radial pulse strengthened immediately after the bolus injection. His blood pressure was 100/60 mmHg 1-h after the bolus injection and he extubated himself. At 5 h after IFE injection, his blood pressure reached 160/100 mmHg and vasopressors were tapered. He remained stable for 6 days after hospital admission and was discharged. Other than a sore throat, he experienced no complication or complaint.
Discussion The usual formulation of glyphosate herbicide products used in Asian countries is 41% glyphosate as an isopropylamine salt, water, and a variable amount of surfactant, most commonly POEA.5 A recent prospective observational study reported that mortality after ingestion of GlySH was 3.2%, and 5.5% of cases had symptoms requiring intervention, such as hypotension, respiratory failure, dysrhythmia, cardiac arrest, marked sedation, seizures, or oliguria.4 Although the mechanism of GlySH toxicity is not yet completely understood, POEA is primarily or partially responsible for the cardiovascular toxicity, possibly through mitochondrial dysfunction.5,7 Cardiovascular collapse is a major cause of death after GlySH exposure,1–3 and patients respond poorly to conventional fluid and vasopressor therapy.4,5 Moon et al. suggested that removal of toxins by hemodialysis can reverse the hemodynamic suppression caused by GlySH,8 but reported cases have demonstrated other indications for hemodialysis, such as acidosis or hyperkalemia. The decision to use hemodialysis in patients with hemodynamic instability and without the typical indications is very challenging. IFE is gaining attention as an antidote to lipophilic toxins that cause hemodynamically significant poisonings.6 Promising experimental and anecdotal evidence exists suggesting that IFE can be effective in the treatment of toxicities caused by substances such as local anesthetics, calciumchannel blockers, and tricyclic antidepressants,6,9 –12 but no report regarding IFE and GlySH herbicide exists to our knowledge. The lipid solubility of the offending toxin is an important factor in determining the efficacy of IFE.6 Glyphosate is a water-soluble compound with an octanol/water partition coefficient (log P) of −3.40,13 and it is unlikely that IFE acts on the toxicity of glyphosate itself. POEA is a nonionic surfactant, which is a mixture of polyethoxylated long-chain amines synthesized from animal-derived fatty acids, and is capable of solubilizing lipids.14 Although the log P of the specific POEA product was not available, POEA can be considered lipidsoluble because, generally, log P of nonionic surfactants
567 ranges between 1.39 and 6.40.15 IFE has the potential to reduce symptoms of GlySH poisoning by lowering free POEA concentration, which blunts its cardiovascular toxicity. IFE is not used as a first-line therapy for poisoned patients, except for cases of local anesthetic toxicity, but IFE could be used as an adjunct treatment for poisoning by lipid-soluble substances.6 Our patient was in critical condition, and conventional therapy was not very effective; thus, IFE was used as a rescue therapy and was associated with hemodynamic improvement. More data are needed to establish the effects and potential side effects of IFE on GlySH poisoning, and plasma POEA concentrations before and after IFE administration would clarify any causal relationship. For the present, IFE should be considered in cases of refractory hemodynamic instability caused by GlySH after aggressive fluid and vasopressor support.
Declaration of interest The authors report no declaration of interest. The authors alone are responsible for the content and writing of this paper.
References 1. Chen YJ, Wu ML, Deng JF, Yang CC. The epidemiology of glyphosatesurfactant herbicide poisoning in Taiwan, 1986–2007: a poison center study. Clin Toxicol (Phila) 2009; 47:670–677. 2. Lee CH, Shih CP, Hsu KH, Hung DZ, Lin CC. The early prognostic factors of glyphosate-surfactant intoxication. Am J Emerg Med 2008; 26:275–281. 3. Lee HL, Chen KW, Chi CH, Huang JJ, Tsai LM. Clinical presentations and prognostic factors of a glyphosate-surfactant herbicide intoxication: a review of 131 cases. Acad Emerg Med 2000; 7:906–910. 4. Roberts DM, Buckley NA, Mohamed F, Eddleston M, Goldstein DA, Mehrsheikh A, Bleeke MS, Dawson AH. A prospective observational study of the clinical toxicology of glyphosate-containing herbicides in adults with acute self-poisoning. Clin Toxicol (Phila) 2010; 48:129–136. 5. Lee HL, Kan CD, Tsai CL, Liou MJ, Guo HR. Comparative effects of the formulation of glyphosate-surfactant herbicides on hemodynamics in swine. Clin Toxicol (Phila) 2009; 47:651–658. 6. Jamaty C, Bailey B, Larocque A, Notebaert E, Sanogo K, Chauny JM. Lipid emulsions in the treatment of acute poisoning: a systematic review of human and animal studies. Clin Toxicol (Phila) 2010; 48:1–27. 7. Jelinek A, Klocking HP. In vitro toxicity of surfactants in U937 cells: cell membrane integrity and mitochondrial function. Exp Toxicol Pathol 1998; 50:472–476. 8. Moon JM, Min YI, Chun BJ. Can early hemodialysis affect the outcome of the ingestion of glyphosate herbicide?. Clin Toxicol (Phila) 2006; 44:329–332. 9. Cave G, Harvey M. Lipid emulsion therapy in lipophilic drug toxicity. Ann Emerg Med 2008; 51:449–450. 10. Cave G, Harvey MG, Castle CD. The role of fat emulsion therapy in a rodent model of propranolol toxicity: a preliminary study. J Med Toxicol 2006; 2:4–7. 11. Moore N, Kirton C, Bane J. Lipid emulsion to treat overdose of local anaesthetic. Anaesthesia 2006; 61:607–608.
Clinical Toxicology vol. 48 no. 6 2010
568 12. Picard J, Ward SC, Zumpe R, Meek T, Barlow J, Harrop-Griffiths W. Guidelines and the adoption of ‘lipid rescue’ therapy for local anaesthetic toxicity. Anaesthesia 2009; 64:122–125. 13. LOGKOW Database. A databank of evaluated octanol-water partition coefficients (Log P). http://logkow.cisti.nrc.ca/logkow/search.html. Accessed 2 May 2010.
S.K. Han et al. 14. Williams GM, Kroes R, Munro IC. Safety evaluation and risk assessment of the herbicide roundup and its active ingredient, glyphosate, for humans. Regul Toxicol Pharmacol 2000; 31:117–165. 15. Uppgård L-L, Lindgren Å, Sjöström M, Wold S. Multivariate quantitative structure-activity relationships for the aquatic toxicity of technical nonionic surfactants. J Surfactants Detergents 2000; 3:33–41.
Clinical Toxicology (2010) 48, 569–571 Copyright © Informa UK, Ltd. ISSN: 1556-3650 print / 1556-9519 online DOI: 10.3109/15563650.2010.492350
SHORT REPORT LCLT
Fomepizole fails to prevent progression of acidosis in 2-butoxyethanol and ethanol coingestion TAWNY HUNG1, CHRISTOPHER R. DEWITT2, WALTER MARTZ2, WILLIAM SCHREIBER2, and DANIEL THOMAS HOLMES3 2-Butoxyethanol and ethanol coingestion
1
Department of Pathology and Laboratory Medicine, University of British Columbia, Vancouver, BC, Canada British Columbia Provincial Toxicology Centre, Vancouver, BC, Canada 3 Department of Pathology and Laboratory Medicine, St. Paul’s Hospital, Vancouver, BC, Canada 2
Introduction. 2-butoxyethanol (2BE) is a solvent commonly incorporated into household and industrial cleaning products. Its ingestion causes rapid central nervous system depression, hypotension, and metabolic acidosis attributable to metabolism of the parent compound to butoxyacetic acid (BAA) by alcohol dehydrogenase. Lactic acidosis is also reported to develop in some cases. Published treatment strategies include the use of ethanol infusion, ethanol with concomitant dialysis, dialysis alone, and fomepizole. Case Report: We present an unusual case of a coingestion of ethanol and 150–250 mL of pure 2BE, which resulted in rapid obtundation, severe airway edema, hypotension, and prolonged acidosis despite the coingestion of ethanol and the administration of a loading dose of fomepizole. Continuous veno-venous hemodialysis was employed to treat the acidosis. Ingestion was confirmed by gas chromatography and mass spectrometric determinaiton of 2BE and BAA. The patient recovered without sequelae. Conclusion. Alcohol dehydrogenase inhibitors may not be adequate to prevent acidosis in significant ingestions to 2BE and extracorporeal treatments may be necessary. Keywords
Acute poisoning; Hydrocarbons; Intoxication; Management; Acid–base disorders
Introduction 2-Butoxyethanol (2BE), also known as ethylene glycol monobutyl ether, is a solvent commonly employed in commercial cleaning products. Human ingestion is characterized by hypotension, central nervous system depression, and metabolic acidosis.1–6 Other reported clinical features include liver enzyme abnormalities, renal injury, hematuria,2,4–6 disseminated intravascular coagulation,5 and acute respiratory distress syndrome.3 Although hemolytic anemia occurs in rodents,7 clinical evidence of hemolysis in human ingestion is an inconsistent finding.1,3,8 Metabolism of 2BE occurs in the liver and displays zeroorder Michaelis–Menten kinetics.1,4 The parent compound is metabolized to butoxyacetic acid (BAA) by alcohol and aldehyde dehydrogenases.1,4 In rats, alcohol dehydrogenase (ADH) inhibition decreases BAA production by diverting metabolism to conjugation pathways.7 Whereas 2BE is hypothesized to be responsible for central nervous system manifestations, BAA is thought to cause the other toxicities.1,5,6 By way of treatment, the successful use of ethanol,1 dialysis,2,3,5 a Received 19 August 2009; accepted 6 May 2010. Address correspondence to Daniel Thomas Holmes, Department of Pathology and Laboratory Medicine, St. Paul’s Hospital, 1081 Burrard Street, Vancouver, BC, V6Z 1Y6 Canada. E-mail:
[email protected]
combination of ethanol and dialysis,6 and fomepizole8 have been described. Here we report a case of inadvertent 2BE ingestion requiring hemodialysis despite coingestion of ethanol and treatment with fomepizole.
Case report A 53-year-old construction worker, consumed an unknown quantity of ethanol before accidentally ingesting approximately 150–250 mL of 99% 2BE stored in a Gatorade® bottle in his workplace, believing he was drinking alcohol. Emergency health services were called by a coworker about 20 min post-ingestion because he had vomited and become unexpectedly drowsy. A witness reported that the patient had consumed “ethylene glycol” (EG). Upon arrival of emergency health services (30 min post-ingestion), vitals were: Glasgow Coma Score, 5; heart rate, 77; and blood pressure, 96/70. Extrication from the workplace was prolonged because of the difficulties with intubation caused by airway edema. Successful intubation was achieved using a combitube®. By 1 h post-ingestion the Glasgow Coma Score was 3. He arrived at the emergency department 75 min postingestion and, after laryngoscopic endotracheal intubation, he was administered 1,125 mg of fomepizole (15 mg/kg) based on the report that he had consumed EG. However, several hours post-admission, the liquid was identified as 99% 2BE
Clinical Toxicology vol. 48 no. 6 2010
570
T. Hung et al.
from a material safety data sheet provided by another coworker. On examination, the heart rate was 80 in normal sinus rhythm with occasional premature ventricular contractions, the respiratory rate was 16, and the blood pressure was 94/67. After transfer to the intensive care unit at 120 min post-ingestion, the blood pressure had dropped to 74/61 necessitating epinephrine infusion. Initial laboratory chemistries were notable for: pH = 7.31, lactate = 25 mg/dL (2.8 mmol/L), ethanol = 138 mg/dL (30 mmol/L), and osmolal gap = 43 mmol/kg but 7 mmol/kg after accounting for ethanol using the formula employed by our laboratory (Table 1 caption). Methanol, EG, isopropanol, acetone, salicylate, and ketones were undetectable. By 10 h post-admission, he had developed a severe anion-gap metabolic acidosis (Table 1) with further lactate elevation. Continuous veno-venous hemodialysis was therefore initiated using citrate anticoagulation and Prism0cal (calcium free) solution for both replacement and dialysate at a rate of 4,000 mL/ h:2,000 mL/h for replacement and 2,000 mL/h for dialysis. The patient’s acidosis resolved within 24 h whereupon continuous veno-venous hemodialysis was discontinued. There was no clinical or laboratory evidence of renal injury, liver compromise, or hemolysis. Two serum specimens were recovered for 2BE and BAA analysis, which were, respectively, determined by gas chromatography and gas chromatography/mass spectrometry using a modification of a published method.9 Following extubation, the patient confirmed that the ingestion was unintentional and was discharged without sequelae.
Discussion 2-BE is rapidly absorbed in the gastrointestinal tract and eliminated rapidly with a reported half-life of 40 min1 (low-dose inhalation exposure). Produced by the action of ADH on 2BE, the BAA metabolite undergoes urinary excretion. The half-life of BAA during dialysis was reported to be 5.3 h in a previous case.2 Unfortunately, a robust calculation of 2BE and BAA half-lives was not possible in this case because of limited data. The maximum observed 2BE and BAA concentrations were 35 mg/dL (3 mmol/L) and 106 mg/dL (8 mmol/L), respectively. Similar to previous reports,1,6 after accounting for ethanol, our patient never displayed a significant elevation in osmolal gap. Although it has been shown that glycol ethers demonstrate a linear relationship with measured osmolality,10 levels of these compounds high enough to result in toxicity (5–10 mmol/L) often do not elevate the osmolal gap above commonly employed decision thresholds.10 Lactic acidosis has been observed in other cases of 2BE ingestion2,3,5,6 and in the present case, where the highest lactate was 66.6 mg/dL (7.4 mmol/L). Burkhart et al. reported a case of combined 2BE, propylene glycol, and monoethanolamine ingestion resulting in a lactate level of 45 mg/dL (5.0 mmol/L), which was attributed to the metabolism of propylene glycol.5 As, in this case, lactic acidosis appeared without propylene glycol coingestion, elevation in lactate may be an intrinsic consequence of 2BE metabolism. Other casespecific factors, such as the hypotension suffered on admission, may have played a role.
Table 1. Laboratory results from the time of admission till discontinuation of dialysis Time from admission (h) Na (mmol/L) K (mmol/L) Cl (mmol/L) CO2 (mmol/L) Glucose (mg/dL) Creatinine (mg/dL) Urea nitrogen (mg/dL) Ethanol (mg/dL) 2-Butoxyethanol (mg/dL) Butoxyacetic acid (mg/dL) Lactate (mg/dL) Anion gap (mmol/L) Osmolality (mmol/kg) Osmolal gap (mmol/L) pH pCO2 (mmHg) pO2 (mmHg) CHCO3 (mmol/L)
RI 135–148 3.6–4.7 110–112 22–31 65–198 0.70–1.10 7.0–22.0 <9 ND ND 4.5–18.9 <10 281–297 <10 7.36–7.44 35–45 >75 21–25
0
4
140 3.7 110 22 180 0.64 10.6 138 – – 25.2 8 337 43a 7.31 43 239 21
140 3.9 117 16 169 0.59 9.0 – – – 9.9 7 315 22 7.25 35 91 15
7
10
12
16
19
24
33
– – – – – – – – – – – – – – 7.23 32 117 13
142 3.5 116 <10 301 0.81 9.8 – – – 36.9 16 319 15 7.20 27 138 10
– – – – – – – – – – 66.6 – – – 7.16 24 166 8
144 3.4 114 12 126 – 8.4 <9 35 106 – 18 302 4 7.30 25 135 12
– – – – – – – – – – – – – – 7.40 31 121 19
– – – – – – – – – – 9.9 – – – 7.44 36 150 24
145 3.7 109 26 99 0.53 9.5 – ND 29 9.9 10 288 −11 7.42 40 100 25
Abbreviations: RI, reference interval; CHCO3−, calculated bicarbonate concentration. aResult calculated without taking into account the ethanol concentration. Using the formula for calculated osmolality developed by Purssell et al.,11 which takes into account the unexpectedly large contribution of ethanol concentration to the plasma osmolality (calculated osmolality = 2[Na] + [glucose] + [urea] + 1.25[ethanol], concentrations in mmol/L), the result is 7 mmol/kg.
Clinical Toxicology vol. 48 no. 6 2010
2-Butoxyethanol and ethanol coingestion Our patient did not develop the hemolytic anemia as reported in animal studies.7 No red cell dysmorphology, or definitive laboratory evidence of hemolysis was observed (lactate dehydrogenase remained stable). A drop in hemoglobin from 13.1 g/dL at admission to 10.7 g/dL at 33 h post-admission is attributable to hemodilution. There is currently no standard of treatment for 2BE. Published cases report the use of hemodialysis,2,3,5 a combination of ethanol and hemodialysis,6 ethanol-only treatment,1 and fomepizole-only treatment.8 In this case, metabolic acidosis occurred after 2BE ingestion, despite ethanol coingestion and fomepizole therapy. Although ADH inhibitors should theoretically be sufficient therapy for early treatment of significant 2BE ingestions, in this case, neither ethanol nor fomepizole appeared to have the desired effect. There may be other mechanisms by which acidosis develops in 2BE ingestions. Although we cannot ascertain the reason for the extended acidosis, the possible inefficacy of ADH inhibitors to prevent BAA formation suggests that their use for 2BE ingestion requires further evaluation.
Conclusion Although ADH inhibitors should be sufficient therapy for early treatment of significant 2BE ingestions, in this case an acidosis partially attributable to BAA developed nevertheless. This suggests that ADH inhibition alone for the treatment of 2BE ingestion may not always be sufficient and that extracorporeal treatments may be required.
571
Declaration of interests The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this paper.
References 1. McKinney PE, Palmer RB, Blackwell W, Benson BE. Butoxyethanol ingestion with prolonged hyperchloremic metabolic acidosis treated with ethanol therapy. J Toxicol Clin Toxicol 2000; 38:787–793. 2. Gijsenbergh FP, Jenco M, Veulemans H, Groeseneken D, Verberckmoes R, Delooz HH. Acute butylglycol intoxication: a case report. Hum Toxicol 1989; 8(3):243–245. 3. Bauer P, Weber M, Mur JM, Protois JC, Bollaert PE, Condi A, Larcan A, Lambert H. Transient non-cardiogenic pulmonary edema following massive ingestion of ethylene glycol butyl ether. Intensive Care Med 1992; 18:250–251. 4. Rambourg-Schepens MO, Buffet M, Bertault R, Jaussaud M, Journe B, Fay R, Lamiable D. Severe ethylene glycol butyl ether poisoning. Kinetics and metabolic pattern. Hum Toxicol 1988; 7:187–189. 5. Burkhart KK, Donovan JW. Hemodialysis following butoxyethanol ingestion. J Toxicol Clin Toxicol 1998; 36:723–725. 6. Gualtieri JF, DeBoer L, Harris CR, Corley R. Repeated ingestion of 2-butoxyethanol: case report and literature review. J Toxicol Clin Toxicol 2003; 41:57–62. 7. Ghanayem BI, Burka LT, Matthews HB. Metabolic basis of ethylene glycol monobutyl ether (2-butoxyethanol) toxicity: role of alcohol and aldehyde dehydrogenases. J Pharmacol Exp Ther 1987; 242:222–231. 8. Butera R, Lapostolle F, Astier A, Baud FJ. Metabolism and toxic effects of ethylene glycol butyl ether in a case of poisoning treated with 4-metylpyrazole. Toxicol Lett 1996; 88:49. 9. Shih TS, Chou JS, Chen CY, Smith TJ. Improved method to measure urinary alkoxyacetic acids. Occup Environ Med 1999; 56:460–467. 10. Browning RG, Curry SC. Effect of glycol ethers on plasma osmolality. Hum Exp Toxicol 1992; 11:488–490. 11. Purssell RA, Pudek M, Brubacher J, Abu-Laban RB. Derivation and validation of a formula to calculate the contribution of ethanol to the osmolal gap. Ann Emerg Med 2001; 38:653–659.
Clinical Toxicology (2010) 48, 572–573 Copyright © Informa UK, Ltd. ISSN: 1556-3650 print / 1556-9519 online DOI: 10.3109/15563650.2010.492352
SHORT REPORT LCLT
A case of life-threatening rectal administration of moist snuff KAI KNUDSEN1 and MORTEN STRINNHOLM2 Moist snuff poisoning
1
Surgical Sciences, Anesthesia and Intensive Care Medicine, Sahlgrenska University Hospital, Gothenburg, Sweden Surgical Sciences, Anesthesia and Intensive Care Medicine, Värnamo County Hospital, Värnamo, Sweden
2
Case report. We report a case of self-administration of 75 sachets of moist snuff rectally in a previously healthy, 42-year-old man. He presented with symptoms of nausea, discomfort, and dizziness. He had dry and warm skin, a pulse rate of 53 bpm, a mean arterial blood pressure of 135 mmHg and fluctuations in consciousness. The patient was treated with mechanical ventilation because of respiratory insufficiency. No specific anti-nicotinergic treatment was given. Plasma levels of the nicotine metabolite cotinine were 8,691 μg/L 7 h after admittance and 9,814 μg/L after 12 h. Levels of cotinine in the urine were above >50,000 μg/L. The patient developed a mild pneumonia, but he was uneventfully extubated after 12 h of mechanical ventilation. All physiological parameters were restored and he was discharged from hospital after 36 h. Conclusion. Excessive rectal administration of moist snuff may be life threatening. Patients may require intensive care. Long-term sequelae were not seen in this case. Keywords Moist snuff; Nicotine; Poisoning; Cotinine
Introduction Oral use of moist snuff is common among young males in Scandinavia, especially among active athletes. It is used as an alternative to smoking tobacco, but as much as 40% of regular users also smoke.1 At least 23% of Swedish men use moist snuff on a regular basis. Moist snuff contains tobacco, water, different flavorings, and added ingredients. Moist snuff sachets are normally placed under the upper lip for buccal absorption. Each sachet contains about 10 mg of nicotine and oral administration usually leads to a systemic dose of 2–3.5 mg from each sachet.2,3 The release of unbound nicotine is highly pH dependent. Absorption is more complete in an alkaline medium and thus rectal administration may increase toxicity because of higher pH in the rectum.4,5 At a pH 8, approximately 50% of the nicotine is unbound, at pH 8.5 around 75%, and at pH 8.9 around 90%. In excessive doses, nicotine may cause agitation, nausea, and vomiting and in rare cases unconsciousness.6–8 We describe a case of nicotine poisoning via rectal administration of moist snuff with the intention to treat severe migraine.
Case A previously healthy, 42-year-old man, arrived at the emergency ward with symptoms of nausea, discomfort, and dizziness. The Received 1 April 2010; accepted 6 May 2010. Address correspondence to Kai Knudsen, Surgical Sciences, Anesthesia and Intensive Care Medicine, Blå Stråket 7, Sahlgrenska University Hospital, Gothenburg, 41345 Sweden. E-mail:
[email protected]
patient had no previous medical history other than untreated migraine. He was a nonsmoker but using moist snuff on a regular basis. He now suffered from severe migraine and intended to treat himself with rectal administration of moist snuff. According to the patient, because of poor response, as much as 75 sachets were administered over 1–2 h. Soon after, the patient felt distressed and tried to remove the snuff himself using a shower pipe, but failed. His relatives called the ambulance. The patient was nauseated and vomited in the ambulance. He presented to the hospital with dry and warm skin, a pulse rate of 53 bpm and increased blood pressure. He was agitated and confused. Initially the patient received administration of a water enema; thereafter a rectoscopy was performed without any significant return of moist snuff. The condition gradually worsened and the patient became comatose and developed anisocoria. The patient was anesthetized with atropine, propofol, and fentanyl and endotracheally intubated with assisted ventilation. Soon after, he was examined with a CT scan of the brain, which was normal. An electrocardiogram was normal and arterial blood gases during mechanical ventilation were normal at arrival as well as after 12 h. No urine drug screening was performed at first but blood and urine samples were taken for analysis of the nicotine metabolite cotinine. Plasma levels of cotinine were 8,691 μg/L 7 h after admittance and 9,814 μg/L after 12 h. Levels of cotinine in the urine were above determination levels 7 h after admittance (>50,000 μg/L). The patient was mainly treated symptomatically. No specific anti-nicotinergic treatment was given and only 0.5 mg of atropine was given during the initial intubation procedure. No specific treatment against hypertension was needed.
Clinical Toxicology vol. 48 no. 6 2010
Moist snuff poisoning Soon after admission to the ICU, the patient had a mean arterial blood pressure of 115 mmHg, and a pulse rate of 58 bpm. Within a few hours blood pressure and pulse rate were normalized. The patient was initially extubated after 2 h but because of impaired consciousness and hypoventilation the patient needed to be reintubated. The patient’s clinical course was complicated by a mild pneumonia, but he was uneventfully extubated after another 12 h of mechanical ventilation. All physiological functions were completely restored and he was discharged from the hospital after 36 h. No further contacts with health-care providers after the incident are known.
Discussion Moist snuff has central nervous system stimulating properties and is addictive as it contains 0.5–1% of nicotine. A common use is about 6–10 sachets per day, mostly as an alternative to smoking tobacco. Nicotine is a cholinergic alkaloid acting on central and peripheral nicotinic and muscarinic receptors.9 It exerts dose-dependent effects on the CNS as well as on autonomic and neuromuscular systems.9 In low doses, nicotinergic stimulation results in sympathetic agonism but at toxic levels nicotine also acts as a nicotinergic antagonist, with parasympathetic and neuromuscular-blocking effects. It also inhibits muscarinic agonism further augmenting parasympathetic stimulation with symptoms such as bradycardia, salivation, bronchorrhea, and bronchospasm. Furthermore, nicotine passes the blood–brain barrier, stimulating release of several neurotransmitter substances such as dopamine, adrenocorticotropic hormone, norepinephrine, acetylcholine, serotonin, and β-endorphin.9 Terminal half-life of nicotine is about 2 h and the main metabolite is cotinine, which is more stable in vitro and available for quantification.9 Nicotine has been used as an analgesic against headache and it could have an antimigraine effect by vasoconstriction and μ-opioid stimulation.10 This patient had positive experiences using moist snuff to treat migraine. However, more nicotine is released in an alkaline medium and thus rectal administration may increase toxicity because of higher pH in the rectum. At least 150 mg of nicotine, probably more, was adsorbed in this case, leading to extremely high concentrations of s-cotinine. About 150 mg of nicotine is potentially lethal, and this patient’s condition was life threatening.8 The patient developed respiratory insufficiency and compromised upper airways because of impaired consciousness secondary
573 to nicotine toxicity. Blood pressure increased and pulse rate decreased but not to critical levels, suggesting invasive monitoring of arterial blood pressure in these cases. The patient also presented with a high hemoglobin value (17.7 g/dL), possibly because of dehydration. Dehydration could trigger migraine but it may also have been worsened by the effects of nicotine with nausea and vomiting. Rehydration seems to be important in the treatment of nicotine poisoning. In conclusion, excessive rectal administration of moist snuff may lead to a life-threatening nicotine poisoning. Severe cases should be monitored in an intensive care unit with invasive monitoring of blood pressure and possibilities to assist ventilation when needed. Public information on the harm of excessive and rectal administration of moist snuff seem to be needed.
Declaration of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this paper.
References 1. Luo J, Ye W, Zendehdel K, Adami J, Adami H-O, Boff Etta P, Nyrén O. Oral use of Swedish moist snuff (“snus”) and risk for cancer of the mouth, lung, and pancreas in male construction workers: a retrospective cohort study. Lancet 2007; 369:2015–2020. 2. Lunell E, Lunell M. Steady-state nicotine plasma levels following use of four different types of Swedish moist snuff compared with 2-mg Nicorette chewing gum: a crossover study. Nicotine Tob Res 2005; 7:397–403. 3. Adrian CL, Ohlin H, Dalhoff K, Jacobsen J. In vivo human buccal permeability of nicotine. Int J Pharm 2005; 311:196–202. 4. Dart RC. Medical Toxicology. 3rd ed. Philadelphia, USA: Lippincott Williams & Wilkins. 2004:601–603. 5. Wendelien M. A study of the buffering capacity of the human rectum. Pharm Weekbl Scient Ed 1989; 11:9–12. 6. American Heart Association (website): http://www.americanheart.org/ presenter.jhtml?identifier=4753. Accessed 13 May 2010. 7. Davies P, Levy S, Pahari A, Martinez D. Acute nicotine poisoning associated with a traditional remedy for eczema. Arch Dis Child 2001; 85:500–502. 8. Moriya F, Hashimoto Y. A fatal poisoning caused by methomyl and nicotine. Forensic Sci Int 2005; 149:167–170. 9. Houezec JL, Benowitz NL. Basic and clinical psychopharmacology of nicotine. Clin Chest Med 1991; 12:681–699. 10. Gupta VK. Antimigraine action of nicotine: theoretical basis and potential clinical application. Eur J Emerg Med 2007; 14:243–244.
Clinical Toxicology (2010) 48, 574–576 Copyright © Informa UK, Ltd. ISSN: 1556-3650 print / 1556-9519 online DOI: 10.3109/15563650.2010.492351
LETTERS TO THE EDITOR LCLT
Child cyanide poisoning after ingestion of bitter almonds Letters to the editor
To the Editor: Many species in the Rosaceae family are cultivated and eaten as culinary fruits throughout the world. Although the outer fleshy part of fruits such as apricots, peaches, and numerous types of plums is indeed delicious, the stone contains toxic cyanogenic heterosides. The best known of these poisons is amygdalin. Hydrolysis of these molecules in the digestive tract of mammals can release hydrocyanic acid that can lead to cyanide poisoning.1 Several cases of severe poisoning have been reported after the ingestion of apricot stones from Prunus armeniaca that appears to be the most toxic species.2,3 Another widely cultivated drupe in temperate climates is the almond. There are several varieties of almond trees (Prunus amygdalus). The most common variety, P. amygdalus dulcis, produces small quantities of cyanogenic heterosides. However, the bitter almond variety (P. amygdalus amara) produces fruit with stones containing high amounts of amygdalin as apricot pits. There have been few reports describing severe poisonings after the ingestion of bitter almonds.4–6 The purpose of this report is to describe a case of severe poisoning after the ingestion of bitter almonds with confirmed ingested dose and measured cyanide blood level.
Determination of cyanides in blood Cyanide ions were assayed by high-performance liquid chromatography linked to a fluorimeter (HPLC-Fluo). Based on the method previously described by Chinaka,7 we developed and fully validated a rapid assay (less than 2 h), which can be performed using a small-volume whole blood sample. Briefly, a 50-μL blood sample is deproteinized by the addition of 2.5 mL of a methanol–water solution. Cyanide ions potentially present in the supernatant were derivatized by the addition of a 100-μL solution containing taurine (5 mmol/L) and naphthalene dialdehyde (1 mmol/L) (1 : 1, v/v) in the dark. Chromatographic separation (Surveyor Thermo Fisher Scientific, Waltham, MA, USA) was performed on a UP BP2 column (150 × 1 mm; 5 μm) (Interchim). The mobile phase was methanol/acetonitrile (80 : 20) and ammonium formate buffer 2 mM pH 3.0 configured in the gradient mode. The emission/excitation wavelengths of the fluorometer (Waters 2475) were set at 460 and 420 nm, respectively. Cyanide retention time was 4.5 min. The calibration model was linear from 0.2 to 3 mg/L (r2 > 0.980). Validation tests showed that this technique was specific with the following characteristics: limit of detection 0.05 mg/L, limit of quantification 0.2 mg/L, intraday precision CV < 7%, interday precision CV < 25%, and accuracy < 7%. These parameters were determined at two concentrations levels: 0.5 and 2.5 mg/L.
Received 8 March 2010; accepted 7 May 2010. Address correspondence to Luc de Haro, Centre Antipoison, Hôpital Salvator, 249 Boulevard Sainte Marguerite, 13009 Marseille, France. E-mail:
[email protected]
Case report The patient is a 30-month-old girl weighing 14 kg with no particular medical history whose grandfather gave her a bitter almond to taste in December 2009. The almond had been picked directly from an almond tree near Beziers in southern France (Languedoc region) and shelled the night before. The child liked the taste and asked for more. According to the statements by both the mother and grandfather, she ate a total of five bitter almonds. Between 10 and 15 min after ingestion, the child became pale and hypotonic with difficulty holding her head up. The concerned parents decided to take the child to the emergency room and called the emergency helpline. During the trip in the car, the girl experienced fluctuating consciousness followed by general seizures. On the road, the family met the emergency team that observed the following findings: miosis, obnubilation, rightward fixation of eyes, and tachycardia. These clinical findings were compatible with a post-seizure state. Intravenous diazepam (5 mg) was prescribed and the child was taken to the local hospital emergency room. On arrival at the emergency room (1 h and 30 min after ingestion) and after consulting the poison control center, a diagnosis of cyanide poisoning was evoked (clinical feature at this moment: drowsiness, hand tremor, hypersialorrhea, blood pressure 105/55 mmHg). After collecting a blood sample, clonazepam (0.3 mg) was administered by the direct intravenous route to prevent further seizure followed by hydroxocobalamin (70 mg/kg) in a 15-min perfusion (infusion beginning 1 h 50 min after ingestion). At the end of perfusion, neurological findings were normal and the child rapidly recovered consciousness. She was transferred overnight to the children’s intensive care unit at the University Hospital Center where brain scan was carried out under sedation with midazolam showed no sequelae. During surveillance, no clinical or laboratory abnormalities were found (no acidosis observed at the intensive care unit – arterial blood gas pH 7.39 and plasmatic bicarbonate level 24 mmol/L – but these biological analyses were performed between 30 and 50 min after the antidote perfusion). The delayed biological results do not allow excluding the development of metabolic acidosis before specific treatment. The child was discharged the next morning with no additional treatment. Cyanide measurement of the pretreatment sample of whole blood based on high-performance liquid chromatography linked to a fluorimeter showed a concentration of 2.33 mg/L (toxic level > 0.5).
Discussion There are no data in the medical literature relating to the quantity of bitter almonds consumed to potential toxicity. Current information suggests that 6–10 almonds can cause potentially fatal poisoning.1,6 In our case in which two adult family members stated that the child had consumed five almonds, the clinical picture involved severe signs of cyanide poisoning with loss of consciousness and seizure that regressed rapidly after perfusion of hydroxocobalamin, the most useful cyanide antidote.8 Indeed, neurological manifestations disappeared by the end of the 15-min perfusion with complete normalization of the child’s consciousness and cognition. Early diagnosis was a key factor in prompt implementation of the use of hydroxocobalamin.8 Circulating cyanide levels just prior to the use of hydroxocobalamin were compatible with severe intoxication. Usual concentrations are lower than 0.02 mg/L in nonsmokers and 0.05 mg/L in smokers. Concentrations between 1.0 and 2.5 mg/L are considered as potentially toxic and more
Clinical Toxicology vol. 48 no. 6 2010
Letters to the editor
575
than 2.5 mg/L as potentially lethal.8 For the child described in this report, the cyanide concentration was near the life-threatening level. Based on these findings, it can be concluded that consumption of five bitter almonds is dangerous for a child and that greater quantities might be fatal.1 Rami Nader Emergencie Unit, Centre Hospitalier Général,
Béziers, France Jean-Claude Mathieu-Daudé Pharmacologie Médicale et Toxicologie, Hôpital Lapeyronie, Montpellier, France Marc Deveaux Laboratoire Toxlab, Paris, France Kristell Faure, Maryvonne Hayek-Lanthois, and Luc de Haro Centre Antipoison, Hôpital Salvator, Marseille, France
References 1. Boustié J, Caubet A, Paris M. Atlas des intoxications d’origine végétale. Encycl Méd Chir, Toxicologie-Pathologie professionnelle, 16-065-A-10, 2002;29. 2. Sayre JW, Kaymakcalan S. Cyanide poisoning from apricot seeds among children in central Turkey. N Engl J Med 1964; 270:1113–1115. 3. Akyildiz BN, Kurtoglu S, Kondolot M, Tunc A. Cyanide poisoning caused by ingestion of apricot seeds. Ann Trop Paediatr 2010; 30:39–43. 4. Pack WK, Raudonat HW, Schmidt K. Lethal poisoning with hydrocyanic acid after ingestion of bitter almonds. Z Rechtsmed 1972; 70:53–54. 5. Shragg TA, Albertson TE, Fisher CJ. Cyanide poisoning after bitter almond ingestion. West J Med 1982; 136:65–69. 6. Kovalevsky P, Bengler C. Intoxication cyanhydrique par ingestion d’amandes amères: mise au point à propos d’un cas. Réanim Soins Intens Méd Urg 1996; 12:149–153. 7. Chinaka S, Takayama N, Michigami Y, Ueda K. Simultaneous determination of cyanide and thiocyanate in blood by ion chromatography with fluorescence and ultraviolet detection. J Chromatogr B Biomed Sci Appl 1998; 713:353–359. 8. Shepherd G, Velez LI. Role of hydroxocobalamin in acute cyanide poisoning. Ann Pharmacother 2008; 42:661–669. 1556-9519 1556-3650 LCLT Clinical Toxicology, Toxicology Vol. 1, No. 1, Jul 2010: pp. 0–0
Management of Datura poisoning Letter J.J. Rasimas to the editor and J. Ward Donovan
To the Editor:
Burns and Linden highlighted the superiority of physostigmine to benzodiazepines in managing toxic anticholinergic delirium.3 In patients with delirium from Datura poisoning who are given appropriate therapy (2 mg per dose, in adults), physostigmine is effective and not normally associated with adverse effects. Thus, in 16 case reports and case series chronicling physostigmine use in 90 patients with anticholinergic plant poisoning that we reviewed, agitated delirium was reversed in 85 patients (94%), without any reports of adverse effects.4 Physostigmine has been used effectively and without adverse effects in two case series of anticholinergic plant poisonings totaling 49 patients with delirium, ranging in age from 11 to 28 years,5,6 and in younger children.7 The use of repeated doses of physostigmine decreased the severity of the toxidrome, cleared cognition, avoided the need for other sedative medications, and shortened the hospital course of jimson weed-poisoned patients without adverse events.8 We believe that this evidence, and our clinical experience, supports a more positive approach to the use of physostigmine for Datura-induced anticholinergic syndrome. J.J. Rasimas and J. Ward Donovan PinnacleHealth Toxicology Center, Penn State College of Medicine, Harrisburg, PA, USA
References 1. Krenzelok EP. Aspects of Datura poisoning and treatment. Clin Toxicol 2010; 48:104–110. 2. Schneir AB, Offerman SR, Ly BT, Davis JM, Baldwin RT, Williams SR, Clark RF. Complications of diagnostic physostigmine administration to emergency department patients. Ann Emerg Med 2003; 42:14–19. 3. Burns MJ, Linden CH, Graudins A, Brown RM, Fletcher KE. A comparison of physostigmine and benzodiazepines for the treatment of anticholinergic poisoning. Ann Emerg Med 2000; 35: 347–381. 4. Donovan JW. Anticholinergic plants. In: Brent J, Wallace KL, Burkhart KK, Phillips SD, Donovan JW, eds. Critical Care Toxicology: Diagnosis and Management of the Critically Poisoned Patient. Philadelphia: Elsevier Mosby; 2005. 5. Levy R. Jimson seed poisoning—a new hallucinogen on the horizon. JACEP 1977; 6:58–61. 6. Klein-Schwartz W, Oderda GM. Jimsonweed intoxication in adolescents and young adults. Am J Dis Child 1984; 138:737–739. 7. Ceha LJ, Presperin C, Young E, Allswede M, Erickson T. Anticholinergic toxicity from nightshade berry poisoning responsive to physostigmine. J Emerg Med 1997; 15:65–69. 8. Burkhart KK, Magalski AE, Donovan JW. A retrospective review of the use of activated charcoal and physostigmine in the treatment of jimson weed poisoning. J Tox Clin Tox 1999; 7:389. 1556-3650 LCLT Clinical 1556-9519 Toxicology Toxicology, Vol. 1, No. 1, Jun 2010: pp. 0–0
1
We read with interest the review of Datura poisoning, however, we question the restriction of physostigmine use to only the most severe or life-threatening cases. Excessive use of physostigmine or its administration to patients without evidence of an anticholinergic toxidrome may, indeed, sometimes produce cholinergic effects. However, they rarely comprise a cholinergic crisis but only mild, short-lived episodes of diaphoresis, vomiting, or diarrhea.2
Received 14 April 2010; accepted 29 April 2010 Address correspondence to J.J. Rasimas, Penn State College of Medicine, PinnacleHealth Toxicology Center, Brady Building, 10th Floor, Suite B, 111 South Front Street, Harrisburg, PA 17101, USA. E-mail: jrasimas@ pinnaclehealth.org
Datura poisoning and the use of physostigmine To the Editor: LetterKrenzelok E.P. to the editor
The cornerstone of optimal patient care is to do no harm. That mantra applies to the use of pharmacological antagonists and all too often the existence of an “antidote,” such as physostigmine, seems to beg for it to
Received 3 June 2010; accepted 4 June 2010. Address correspondence to Edward P. Krenzelok, Pittsburgh Poison Center, University of Pittsburgh Medical Center, 200 Lothrop Street, BIR 010701, Pittsburgh, PA 15213, USA. E-mail:
[email protected]
Clinical Toxicology vol. 48 no. 6 2010
Letters to the editor
575
than 2.5 mg/L as potentially lethal.8 For the child described in this report, the cyanide concentration was near the life-threatening level. Based on these findings, it can be concluded that consumption of five bitter almonds is dangerous for a child and that greater quantities might be fatal.1 Rami Nader Emergencie Unit, Centre Hospitalier Général,
Béziers, France Jean-Claude Mathieu-Daudé Pharmacologie Médicale et Toxicologie, Hôpital Lapeyronie, Montpellier, France Marc Deveaux Laboratoire Toxlab, Paris, France Kristell Faure, Maryvonne Hayek-Lanthois, and Luc de Haro Centre Antipoison, Hôpital Salvator, Marseille, France
References 1. Boustié J, Caubet A, Paris M. Atlas des intoxications d’origine végétale. Encycl Méd Chir, Toxicologie-Pathologie professionnelle, 16-065-A-10, 2002;29. 2. Sayre JW, Kaymakcalan S. Cyanide poisoning from apricot seeds among children in central Turkey. N Engl J Med 1964; 270:1113–1115. 3. Akyildiz BN, Kurtoglu S, Kondolot M, Tunc A. Cyanide poisoning caused by ingestion of apricot seeds. Ann Trop Paediatr 2010; 30:39–43. 4. Pack WK, Raudonat HW, Schmidt K. Lethal poisoning with hydrocyanic acid after ingestion of bitter almonds. Z Rechtsmed 1972; 70:53–54. 5. Shragg TA, Albertson TE, Fisher CJ. Cyanide poisoning after bitter almond ingestion. West J Med 1982; 136:65–69. 6. Kovalevsky P, Bengler C. Intoxication cyanhydrique par ingestion d’amandes amères: mise au point à propos d’un cas. Réanim Soins Intens Méd Urg 1996; 12:149–153. 7. Chinaka S, Takayama N, Michigami Y, Ueda K. Simultaneous determination of cyanide and thiocyanate in blood by ion chromatography with fluorescence and ultraviolet detection. J Chromatogr B Biomed Sci Appl 1998; 713:353–359. 8. Shepherd G, Velez LI. Role of hydroxocobalamin in acute cyanide poisoning. Ann Pharmacother 2008; 42:661–669. 1556-9519 1556-3650 LCLT Clinical Toxicology, Toxicology Vol. 1, No. 1, Jul 2010: pp. 0–0
Management of Datura poisoning Letter J.J. Rasimas to the editor and J. Ward Donovan
To the Editor:
Burns and Linden highlighted the superiority of physostigmine to benzodiazepines in managing toxic anticholinergic delirium.3 In patients with delirium from Datura poisoning who are given appropriate therapy (2 mg per dose, in adults), physostigmine is effective and not normally associated with adverse effects. Thus, in 16 case reports and case series chronicling physostigmine use in 90 patients with anticholinergic plant poisoning that we reviewed, agitated delirium was reversed in 85 patients (94%), without any reports of adverse effects.4 Physostigmine has been used effectively and without adverse effects in two case series of anticholinergic plant poisonings totaling 49 patients with delirium, ranging in age from 11 to 28 years,5,6 and in younger children.7 The use of repeated doses of physostigmine decreased the severity of the toxidrome, cleared cognition, avoided the need for other sedative medications, and shortened the hospital course of jimson weed-poisoned patients without adverse events.8 We believe that this evidence, and our clinical experience, supports a more positive approach to the use of physostigmine for Datura-induced anticholinergic syndrome. J.J. Rasimas and J. Ward Donovan PinnacleHealth Toxicology Center, Penn State College of Medicine, Harrisburg, PA, USA
References 1. Krenzelok EP. Aspects of Datura poisoning and treatment. Clin Toxicol 2010; 48:104–110. 2. Schneir AB, Offerman SR, Ly BT, Davis JM, Baldwin RT, Williams SR, Clark RF. Complications of diagnostic physostigmine administration to emergency department patients. Ann Emerg Med 2003; 42:14–19. 3. Burns MJ, Linden CH, Graudins A, Brown RM, Fletcher KE. A comparison of physostigmine and benzodiazepines for the treatment of anticholinergic poisoning. Ann Emerg Med 2000; 35: 347–381. 4. Donovan JW. Anticholinergic plants. In: Brent J, Wallace KL, Burkhart KK, Phillips SD, Donovan JW, eds. Critical Care Toxicology: Diagnosis and Management of the Critically Poisoned Patient. Philadelphia: Elsevier Mosby; 2005. 5. Levy R. Jimson seed poisoning—a new hallucinogen on the horizon. JACEP 1977; 6:58–61. 6. Klein-Schwartz W, Oderda GM. Jimsonweed intoxication in adolescents and young adults. Am J Dis Child 1984; 138:737–739. 7. Ceha LJ, Presperin C, Young E, Allswede M, Erickson T. Anticholinergic toxicity from nightshade berry poisoning responsive to physostigmine. J Emerg Med 1997; 15:65–69. 8. Burkhart KK, Magalski AE, Donovan JW. A retrospective review of the use of activated charcoal and physostigmine in the treatment of jimson weed poisoning. J Tox Clin Tox 1999; 7:389. 1556-3650 LCLT Clinical 1556-9519 Toxicology Toxicology, Vol. 1, No. 1, Jun 2010: pp. 0–0
1
We read with interest the review of Datura poisoning, however, we question the restriction of physostigmine use to only the most severe or life-threatening cases. Excessive use of physostigmine or its administration to patients without evidence of an anticholinergic toxidrome may, indeed, sometimes produce cholinergic effects. However, they rarely comprise a cholinergic crisis but only mild, short-lived episodes of diaphoresis, vomiting, or diarrhea.2
Received 14 April 2010; accepted 29 April 2010 Address correspondence to J.J. Rasimas, Penn State College of Medicine, PinnacleHealth Toxicology Center, Brady Building, 10th Floor, Suite B, 111 South Front Street, Harrisburg, PA 17101, USA. E-mail: jrasimas@ pinnaclehealth.org
Datura poisoning and the use of physostigmine To the Editor: LetterKrenzelok E.P. to the editor
The cornerstone of optimal patient care is to do no harm. That mantra applies to the use of pharmacological antagonists and all too often the existence of an “antidote,” such as physostigmine, seems to beg for it to
Received 3 June 2010; accepted 4 June 2010. Address correspondence to Edward P. Krenzelok, Pittsburgh Poison Center, University of Pittsburgh Medical Center, 200 Lothrop Street, BIR 010701, Pittsburgh, PA 15213, USA. E-mail:
[email protected]
Clinical Toxicology vol. 48 no. 6 2010
576 be used, whether it is indicated or not. Not all patients who are symptomatic following an exposure to Datura stramonium require the use of this reversal agent. As stated in my review,1 “. . . if significant anticholinergic toxicity develops, the mainstay of therapy is physostigmine” and “... the administration of physostigmine would appear to be the perfect reversal agent to use in patients who are suffering from Daturainduced anticholinergic toxicity,” which is in agreement with the expressed opinions of doctors Rasimus and Donovan. Also consistent with their letter I stated that “. . . physostigmine has been demonstrated to be more effective than benzodiazepines in the management of anticholinergic toxicity.” The difference of opinion may stem from their interpretation of my statement that limits its use to “. . . a diagnostic or therapeutic intervention for life-threatening effects or profound delirium....” I believe that most would agree that physostigmine should not be used as a universal panacea in the management of mild anticholinergic toxicity, but that its use should be reserved for clinical situations when a patient is suffering from significant anticholinergic toxicity. In a chapter authored by Dr. Donovan,2 he recommended intensive care unit admission for patients who suffer from anticholinergic plant poisoning and meet any of the following criteria: • • • • •
Delirium unresponsive to mild sedatives Hallucinations Physostigmine administration required Associated respiratory failure Associated renal failure
Letters to the editor • Airway compromise • Seizures or symptomatic dysrhythmias In my opinion, an intensive care unit admission is generally reserved for those who are seriously ill. Furthermore, the stated criteria are consistent with “severe” anticholinergic toxicity and supported by those criteria, my recommendation to use physostigmine as a “. . . therapeutic intervention for life-threatening effects or profound delirium” seems appropriate. Edward P. Krenzelok Pittsburgh Poison Center, University of Pittsburgh Medical Center, Pittsburgh, PA, USA; School of Pharmacy, University of Pittsburgh, Pittsburgh, PA, USA
References 1. Krenzelok EP. Aspects of Datura poisoning and treatment. Clin Toxicol 2010; 48:104–110. 2. Donovan JW. Anticholinergic plants. In: Brent J, Wallace KL, Burkhart KK, Phillips SD, Donovan JW, eds. Critical care toxicology: diagnosis and management of the critically poisoned patient. Philadelphia, PA: Elsevier Mosby; 2005.