Search

The question to be addressed is whether respiratory acidosis favors the cerebral distribution of cyanide, and conversely, if respiratory alkalosis limits its distribution. The pharmacokinetics of a non-toxic dose of cyanide were first studied in a group of 7 rats in order to determine the distribution phase. The pharmacokinetics were found to best fit a three-compartment model with very rapid distribution ([T.sub.1/2 [Alpha]] = 21.6 [+ or -] 3.3 s). Then, the effects of the modulation of arterial pH on the distribution of a non-toxic dose of intravenously-administered cyanide into the brain of rats were studied by means of the determination of the permeability-area product (PA). The modulation of arterial blood pH was performed by variation of Pa[CO.sub.2] in 3 groups of 8 anaesthetized mechanically-ventilated rats. The mean arterial pH measured 20 min after the start of mechanical ventilation in the acidotic, physiologic, and alkalotic groups were 7.07 [+ or -] 0.03, 7.41 [+ or -] 0.01, and 7.58 [+ or -] 0.01, respectively. The mean PAs in the acidotic, physiologic, and alkalotic groups, determined 30 s after the intravenous administration of cyanide, were 0.015 [+ or -] 0.002, 0.011 [+ or -] 0.001, and 0.008 [+ or -] 0.001 [s.sup.-1], respectively (one-way ANOVA p [is less than] 0.0087). At alkalotic pH the mean permeability-area product was 43% of that measured at acidotic pH. This effect of pH on the rapidity of cyanide distribution does not appear to be limited to specific areas of the brain. We conclude that modulation of arterial pH by altering Pa[CO.sub.2] may induce significant effects on the brain uptake of cyanide., Djerad A, Monier C, Houze P, Borron SW, Lefauconnier MJ, Baud FJ. INSERM U26, Universite Paris 7, Hopital Fernand Widal, Paris, France; Laboratoire de Biochimie A. Hopital Saint-Louis, Paris, France; [...]

Objectives: High dose buprenorphine, a potent semisynthetic agonist-antagonist for opiate receptors, is now used in substitution treatment of human heroin addiction. Deaths have been reported in addicts (mis)using buprenorphine. We determined the median lethal dose and studied the effects of high doses of intravenous buprenorphine on arterial blood gases in rats. Methods: Male Sprague-Dawley rats were administered buprenorphine intravenously to determine the LD50 using the up-and-down method. Catheterized groups of 10 rats received either no-drug, saline, acid-alcohol aqueous solvent (required to dissolve buprenorphine at a high concentration), or 3, 30 or 90 mg/kg of buprenorphine intravenously. Serial arterial blood gases were obtained over 3 h. Results: The LD50 determined in triplicate was 146.5 mg/ kg (range: 142.6-176.5). The mean dose received by surviving animals was 96.9 [+ or -] 46.7 mg/kg. There was a significant effect of the acid-alcohol aqueous solvent on arterial blood gases. Excluding the solvent effect, 3, 30 and 90 mg/kg buprenorphine doses had no significant effects. No animal in the arterial blood gas study died, even at intravenous doses up to 90 mg/kg. Conclusion: The toxicity of intravenous buprenorphine in adult rats, assessed by the LD50, is low. Buprenorphine in doses ranging from 3 to 90 mg/kg had no significant effects on arterial blood gases in adult rats compared with animals receiving the aqueous solvent. These data are consistent with the low toxicity of intravenous buprenorphine in rats. The mechanism of death after the intravenous administration of a lethal dose of buprenorphine remains to be determined., Gueye PN, Megarbane B, Borron SW, Risede P, Monier C, Debray M, Baud FJ. INSERM U26, Hopital Fernand Widal, Reanimation Medicale et Toxicologique, Hopital Lariboisiere, Paris, [...]

Cyanide (CN) poisoning in fire victims is frequent and rapidly fatal. Several antidotes to cyanide are available. However, their safety has not been assessed in the complex setting of smoke inhalation. In a prospective study we tried to assess the biological tolerance of a high dose of hydroxocobalamin (HOCo) administered at the scene of the fire in fire victims suspected of CN poisoning. Methods: Inclusion criteria: fire victims were included according to the presence of the following criteria: soot deposits in mouth or sputum plus any degree of neurological impairment. Exclusion criteria: children, pregnant women, burns of total surface body area [is greater than] 20%, and multiple trauma. Protocol design: following examination and the collection of a blood sample in dry heparin, a 5 g dose of HOCo (10 g in case of cardiovascular collapse) was administered intravenously over 15 min. Blood cyanide concentrations were measured using the method of Rieders. The following parameters were collected on admission and then daily during three days: plasma lactate concentration, blood glucose, serum creatinine, liver functions tests, prothrombin time, serum CPK, blood cell counts. Results: 42 (22 females, 20 males) of 63 patients enrolled were found to be intoxicated by cyanide, with a median blood cyanide concentration of 96.1 [micro]mol/L (range: 52-149). The median carbon monoxide concentration was 2.86 mmol/L (range: 2.06-4.10). Fourteen deaths were observed, 9 of these by decerebration, 4 due to septic shock, and one by hypoxemic pneumopathy. The median dose of hydroxocobalamin was 5 g, with a range of 5 to 15 g. Excluding the patients with initial cardiorespiratory arrest, there were no obvious side effects on blood glucose and serum creatinine concentrations and liver function tests. There was only a trend towards a decrease in hemoglobin level and platelet count over the 3-day period which, however, remained within the normal range. The lone side effects were a deep red coloration of both plasma and urine that disappeared within 7 days after injection. The deep red coloration of plasma may induce some interference with clinical chemistry. Conclusion: Hydroxocobalamin is well tolerated and is not associated with a significant effect on renal or hepatic function in fire victims. These results may be viewed as an indicator of the safety of use of hydroxocobalamin at the fire scene in smoke inhalation victims., Borron SW, Vicaut E, Ruttimann M, Baud FJ. Reanimation Medicale et Toxicologique, Hopitaux Lariboisiere et Fernand Widal, Laboratoire de Biophysique, Universite Paris VII, Paris, France; Service Medical de la Brigade [...]

Background: Cyanide (CN) poisoning in fire victims is frequent and rapidly fatal. Several antidotes to cyanide are available. However, their safety has not been assessed in the complex setting of smoke inhalation. In a prospective study we tried to assess the clinical tolerance of a high dose of hydroxocobalamin (HOCo) administered at the scene of the fire in fire victims suspected of CN poisoning. Methods: Inclusion criteria: fire victims were included according to the presence of the following criteria: soot deposits in mouth or sputum plus any degree of neurological impairment. Exclusion criteria: children, pregnant women, burn of total surface body area [is greater than] 20%, multiple trauma were excluded. Protocol design: following examination and the collection of a blood sample in dry heparin, a 5 g dose of HOCo (10 g in case of cardiovascular collapse) was administered intravenously over 15 min. Blood cyanide concentrations were measured using the method of Rieders. The heart rate and systolic blood pressure were monitored before and after the administration of HOCo, and one hour later. A particular attention was paid to the occurrence of allergic reactions: Quincke edema, cutaneous rash, and bronchospasm. Results: There were 28 females and 22 males. The mean blood CN concentration was 83 [+ or -] 73 [micro]mol/L. The mean blood carbon monoxide was 3.2 [+ or -] 2.1 mmol/L. Nineteen fire victims eventually died. In the 27 non-CN-intoxicated patients (blood CN [is less than] 40 [micro]mol/L), there was no significant change in arterial blood pressure. In the 33 CN-intoxicated patients (blood CN [is greater than] 40[micro]mol/L) a significant increase in blood pressure was observed both immediately (p [is less than] 0.001) and 1 hour later (p [is less than] 0.001) after the administration of HOCo. No allergic reactions were observed. The lone side effects were a deep red coloration of the urine that disappeared within 7 days after injection. Conclusions: In fire victims with cyanide poisoning, the administration of a high dose of hydroxocobalamin was associated with an improvement in systolic blood pressure. Hydroxocobalamin is well tolerated in fire victims without CN poisoning. No allergic reactions were observed both in cyanide poisoned and non-cyanide poisoned fire victims., Baud FJ, Favier C, Borron SW, Ruttimann M. Reanimation Medicale et Toxicologique, Hopitaux Fernand Widal et Lariboisiere, Universite Paris VII; Departement d'Anesthesie Reanimation, Hopital St Roch, Nice; Service Medical d'Urgence [...]

Background: Cyanide quickly binds to the ferric ion of cytochrome [aa.sub.3] inducing a noncompetitive inhibition of the mitochondrial cytochrome c oxidase activity. The inactivation of cytochrome c oxidase results in a shift of aerobic to anaerobic metabolism eventually leading to cellular ATP depletion and lactic acidosis. Due to the shift from aerobic to anaerobic metabolism, significant cyanide poisoning is invariably associated with lactic acidosis. Methods: Retrospective study in a university hospital ICU. Repeated blood samples were simultaneously collected for blood cyanide and plasma lactate determination in acute cyanide poisonings, excluding fire victims. The correlations were studied using the Spearman test. Results: Eleven patients were studied. Before antidotal treatment (hydroxocobalamin [+ or -] thiosulfate), the median plasma lactate concentration was 18.6 mmol/L, the median blood cyanide concentration was 156 [micro]mol/L (4.0 mg/L). There was a significant correlation between plasma lactate and blood cyanide concentrations (r = 0.745, p = 0.0174). There were significant inverse correlations between plasma lactate concentrations and systolic blood pressure (r = -0.875, p = 0.0016), spontaneous respiratory rate (r = -0.867, p = 0.0123), and arterial pH (r = -0.870, p = 0.0045). There was a significant positive correlation between plasma lactate concentrations and anion gap (r = 0.833, p = 0.0083). The correlation with blood glucose concentration did not reach the level of statistical significance (p = 0.0589). There was no significant correlation of plasma lactate concentrations with heart rate, Glasgow coma score, Pa[CO.sub.2] or Pa[O.sub.2]. Before antidotal treatment, there were significant inverse correlations of blood cyanide concentrations with systolic blood pressure (r = -0.717, p = 0.0234), and arterial pH (r = -0.912, p = 0.0013). There was a significant positive correlation with anion gap (r = 0.767, p = 0.0214), and blood glucose concentration (r = 0.783, p = 0.0172). There was no significant correlation of blood cyanide concentration with pulse rate, respiratory rate, Glasgow coma score, Pa[O.sub.2] or Pa[CO.sub.2]. During the course of cyanide poisonings, a plasma lactate concentration [is greater than or equal to] 8 mmol/L was sensitive (94%) and moderately specific (70%) for a toxic blood cyanide concentration ([is greater than or equal to] 40 [micro]mol/L or 1.0 mg/L). The specificity was substantially improved in patients not receiving catecholamines (85%). Interpretation: Several factors may contribute to the onset of lactic acidosis in cyanide poisoning including mitochondrial inhibition, cardiovascular collapse, respiratory depression, and catecholamine rush while seizures seems rare. The serial measurement of plasma lactate concentrations is useful in assessing the severity of cyanide poisoning., Baud FJ, Borron SW, Megarbane B, Bismuth C. Reanimation Medicale et Toxicologique, Universite Paris VII-INSERM U26, Hopital Lariboisiere, Paris, France; Department of Emergency Medicine, George Washington University, Washington, DC, [...]

Objectives: Opiates and benzodiazepines are commonly used conjointly, both licitly (in the clinic and operating room) and illicitly (by drug abusers). It is well known that these two drug classes may interact to provoke untoward reactions such as respiratory depression, particularly in the setting of overdose. However, the mechanisms and quantitative relationships of these interactions remain poorly understood. Given the frequency of combined use of these drugs, particularly in the setting of drug abuse, a better understanding of the interactions between benzodiazepines and opiates appears essential. We thus undertook a randomized study of the interactions of the commonly abused benzodiazepines, flunitrazepam, with three opiates, morphine, buprenorphine, and methadone, in a median lethal dose animal model. Methods: We performed randomized, operator-blinded intravenous median lethal dose (MLD) studies in rats of morphine, buprenorphine, and methadone both alone, and proceeded by a single intraperitoneal dose of flunitrazepam. We employed the up-and-down method of Dixon and Bruce, performed in quadruplicate, also comparing time to death following opiate injection. Results are expressed as median (IQR). Results: The MLDs of morphine, buprenorphine, and methadone alone were 64.0 (34.3), 234.6 (74.1) and 22.5 (7.5) mg/kg, respectively, given alone. Their MLDs were 60.6 (30.2), 38.4 (17.7) and 13.0 (2.2) mg/kg, respectively, after pretreatment with flunitrazepam 40 mg/kg ip. Time to death for morphine, buprenorphine, and methadone alone were 2.5 (3.0), 0.02 (0.15), and 2.0 (3.5) h, respectively, and 13.5 (24.0), 24.0 (65.9), and 0.0 (0.4) h, respectively, after pretreatment with flunitrazepam 40 mg/kg ip. Flunitrazepam significantly altered methadone (p = 0.02) and buprenorphine lethality, (p = 0.02). Morphine lethality was not significantly altered (p = 0.77). Time to death was only altered for buprenorphine (p [is less than] 0.01). Discussion: Flunitrazepamopiate drug-drug interactions appear complex. Flunitrazepam augmented lethality two-fold in methadone-treated rats and six-fold in buprenorphine-treated rats, without significant effect on morphine lethality. Time to death was altered in the buprenorphine plus flunitrazepam group, suggesting an alteration in the mechanism of death. A review of the pharmacology of these drugs provides possible explanations for the discrepancies observed in these interactions. Particular attention is warranted with respect to differential drug-drug interactions in terms of absorption, distribution and tissue uptake, metabolism and excretion, as well as toxicodynamic interactions at opiate and benzodiazepine receptor sites. Alteration of the metabolism of one drug group by the other appears inadequate to explain the variable observations in our study. Action at various opiate receptor sites ([Mu], [Kappa], [Delta]) by the same drug may have conflicting effects, the predominant effect depending perhaps not only on the inherent propensity of the specific opiate for binding to that receptor type, but also to the dose administered. Additionally, there is a growing body of knowledge which suggests that benzodiazepines and opiates, while each having their own specific receptor sites, may modify the effects of the other drug group's specific binding sites. These interactions and their possible role in overdose deaths will be discussed., Borron SW, Monier C, Risede P, Baud FJ. INSERM U26, Hopital Fernand Widal, Paris, France and International Toxicology Consultants, LLC, Washington, DC, [...]

Objective: Carbon monoxide and cyanide have the propensity to produce delayed neuroradiological findings and a clinical extrapyramidal syndrome following acute intoxication. In smoke inhalation, where both carbon monoxide and cyanide may be present, the relative contributions of each gas is unknown. We present a case of a smoke inhalation victim with a relatively low blood carbon monoxide and elevated blood cyanide concentrations who developed both characteristic neurological images by MRI and an extrapyramidal syndrome. Case Report: A 74-year-old female was discovered in respiratory arrest in an apartment fire. She was intubated, artificially ventilated with 100% oxygen and immediately given 5 g of hydroxocobalamin. The patient remained in a profound coma, GCS = 3, with reactive and symmetric pupils. On ICU arrival, she remained comatose, GCS = 7, with brisk and symmetric deep tendon reflexes, and a left Babinski sign. She had no cutaneous burns. Blood drawn at the scene revealed a carbon monoxide concentration of 0.99 mmol/L (~11% carboxyhemoglobin) and a blood cyanide concentration of 68 [micro]mol/L (1.77 mg/L) A brain CT on day 2 revealed no edema, ventricles of normal size, and small hypodensities in the head of the caudate nucleus and white matter of both frontal lobes. An MRI on day 8 revealed hypersignals in T2 in both hemispheres in the subcortical and paraventricular regions, of ischemic nature. The neurological course improved, with awakening permitting extubation on day 10. Thereafter, an extrapyramidal syndrome involving all 4 limbs appeared. On day 30, choreoathetotic movements of the upper extremities, head, and trunk appeared in the patient, who remained dysarthric. These movements diminished then disappeared over one month. An extrapyramidal hypertonia of all four limbs persisted. Repeat MRI performed on day 60 showed hypersignals in T1 of both lenticular nuclei, representing hemorrhagic lesions. In the T2 sequence, a hypersignal of the two lenticular nuclei was noted, predominantly at the level of the putamen, as well as nodular hypersignals in T2 of the white matter. Subcortical and cortical atrophy were noted as well. Seen in follow-up 6 months later, the clinical examination revealed no neurological deficit of motor or sensory functions. However, an incapacitating extrapyramidal hypertonia persisted, predominantly of the left upper arm and the face. Repeat MRI revealed a bilateral, but predominantly right hemispheric hypersignal on T2 weighting in the putamen, and a lesser signal in the globus pallidus. Cortical atrophy had increased. Conclusion: Cyanide should be considered as possible etiology for delayed neurological manifestations in smoke inhalation victims. Guerin JM, Boukobza M, Borron SW, Baud FJ. Medical and Toxicological Intensive Care Unit, and Department of Neuroimaging, Hopital Lariboisiere, University Paris-VII, INSERM U26, Paris, France; George Washington University, Washington, DC, USA

Objective: Buprenorphine (BUP) and methadone (MTD) are the most commonly employed substitution products for the treatment of heroin abuse. The safety of these products in overdose, relative to that of heroin, is poorly appreciated. We thus undertook an experimental study to compare the lethality of BUP and MTD in comparison with morphine sulfate (MS), used as a surrogate for heroin. Methods: Opiate-naive Sprague-Dawley rats of 200-300 grams were employed. An open-label preliminary study was performed to determine the intravenous median lethal dose (MLD) of the three products. Concentrations were then adjusted to allow for MLDs of approximately equal volume. Animals were then randomized and dosed intravenously in a blinded manner using one of the three products. The rats were observed carefully until death or until 7 days had passed. The MLD was determined for each blinded product in quadruplicate to assure reproducibility, using the up-and-down method of Dixon and Bruce. A mean median lethal dose (xMLD) was then calculated. The results are expressed as mean + SD Maximal therapeutic doses (MTD) in humans were obtained from the product package insert. A safety index (SI) was calculated for each drug, using the formula SI = x[MLD.sub.(rat)]/[MTD.sub.(human)]. Assigning a safety factor in overdose of 1 to morphine, a relative safety factor (RSF) was calculated for BUP and MTD using the formula RSF = SI(substitution product x)/SI(morphine). Results: The results are shown in the Table. Conclusions: Buprenorphine appears to be significantly safer in overdose, compared with morphine sulfate and methadone in the opiate-naive rat model, using human maximal therapeutic doses as a reference dose. Methadone appears to have a lower margin of safety than morphine sulfate in the rat. Further studies are needed to determine the relative safety of these substitution products in morphine-dependent rats., Relative Safety of Opioids Mean Median Standard Maximal Lethal Dose Deviation Therapeutic Dose Drug x[MLD.sub.(rat)] SD [MTD.sub.(human)] Morphine 45.7 10.4 0.28 mg/kg Buprenorphine 230.7 49.2 0.22 mg/kg Methadone 23.8 5.2 [...]

Objective: Acute methanol poisoning may produce serious clinical events or even be lethal in cases of non-treatment or delayed treatment. Toxicity is due to the enzymatic metabolism of methanol by alcohol dehydrogenase that yields two toxic metabolites, formaldehyde and formic acid. Recognized treatment includes sodium bicarbonate, hemodialysis and ethanol therapy as a competitive inhibitor of methanol metabolism. 4-Methylpyrazole (4-MP) is now an established treatment for ethylene glycol poisoning. The aim of this study was to determine the efficacy and safety of 4-MP in methanol intoxication. Methods: Retrospective data collection from patients (pts) hospitalized in ICU during 1992-1998 for methanol intoxication, confirmed by laboratory assessments and treated with 4-MP; descriptive analysis (median [range]) and report of 4-MP side effects. Results: 11 pts were treated for documented methanol intoxication [cooking alcohol (7 pts), windshield washing fluid (1 pt), antifreeze (1 pt) and undetermined (2 pts)]: 7M/4F; 46 years [18-58]; 5 pts with a past history of chronic alcoholism; 8 suicide attempts and 3 misuses. Six pts were severely and 5 mildly intoxicated. On admission, 6 pts were awake, 1 pt inebriated, 2 pts lethargic and 1 pt comatose requiring mechanical ventilation. Two pts presented bilateral blindness, 2 pts color vision disorder, 2 pts vomiting, 1 pt hypotension and 3 pts tachypnea. Serum methanol level on admission was 51 mg/dL [6-518]. Six pts had co-ingested ethanol with blood concentrations of 14.5 mg/dL [0-35]. Arterial pH was 7.32 [7.10-7.51], serum bicarbonate 17.0 mmol/L [2.5-25.0], Pa[CO.sub.2] 3.6 kPa [1.1-7.2], lactates 2.2 mmol/L [1.1-6.9] and creatinine 83 [micro]mol/L [53-128]. All pts were treated with 4-MP, 9 intravenously and 2 orally, with a loading dose of 700 mg [500-1200], 2 [1-7] multiple doses and a cumulative dose of 1300 mg [500-6000]. Hemodialysis was performed in the 2 pts who presented bilateral blindness. No patient died and all were alive without sequelae, except the 2 blind pts. Three patients received 5, 6 and 7 doses of 4-MP with respectively a total dose of 4000, 6000 and 3100 mg. The main adverse experiences with 4-MP were: nausea and headache in 1 pt, elevation of eosinophil cells, lymphangitis and burning sensation in 1 pt (the one who received 7 doses) and fever in 2 pts (of whom one received 5 doses). Conclusion: 4-MP treatment was an effective and safe therapeutic antidote in preventing or diminishing methanol toxicity in 7 poisoned pts. Hemodialysis was only performed in case of ocular impairment on admission. Megarbane B(1), Borron SW(1), Krencker E(2), Hantson Ph(4), Flesch F(2), Jaeger A(3), Baud FJ(1). (1) Reanimation Medicale, Hopital Lariboisiere, Paris, France (2) Centre antipoison (3) Reanimation Medicale; Hopital Universitaire, Strasbourg, France (4) Service des Soins Intensifs, Universite de Louvain, Bruxelles, Belgium

Introduction: The value of PK-PD relationships in clinical pharmacology is now well recognized. However, the potential interest of Toxicodynamic-Toxicokinetic (TK-TD) relationships in medical toxicology has been poorly investigated. The aim of this study was to correlate the depth of coma with plasma meprobamate concentrations. Materials and Methods: Plasma meprobamate concentrations were measured using a colorimetric assay. The depth of coma was assessed using the Glasgow coma scale (GCS). Non-linear regression was used for modeling TK-TD relationships. Results: TK-TD relationships were studied in 7 acute meprobamate poisonings. Three patients were previously treated with meprobamate. Previous treatment was unknown in 2 patients. Mixed drug poisoning was noted in the 7 patients. The mean GCS at the time of hospital admission was 4 [+ or -] 1, the mean plasma meprobamate concentration was 1054 [+ or -] 318 [micro]mol/L. The mean plasma meprobamate concentration associated with a GCS of 3 was 714.6 [+ or -] 277.7 [micro]mol/L (95% CI: 546.8-882.4). The TK-TD relationship was well fitted with the sigmoidal Emax model. In the 7 patients, the mean Hill coefficient (m [+ or -] SD) was 6.9 [+ or -] 4.0 (95% CI: 3.2-10.7), the mean [C.sub.50] was 487.9 [+ or -] 318.8 [micro]mol/L (95% CI: 193.0-782.7). Two patients exhibited tolerance to the sedative effect of meprobamate. In the 5 non-tolerant patients, the mean Hill coefficient (m [+ or -] SD) was 8.0 [+ or -] 4.4 (95% CI: 2.5-13.4), the mean [C.sub.50] was 315.8 [+ or -] 148.9 [micro]mol/L (95% CI: 130.9-500.7). Discussion: A maximal toxic effect, i.e.: GCS of 3, was associated with a wide range of plasma concentrations. During the course of acute human meprobamate poisoning, the relationship between the depth of coma and the corresponding plasma concentrations is of a sigmoidal shape. The high value of the Hill coefficient in non-tolerant patients showed that a small decrease in plasma meprobamate concentrations near the [C.sub.50] was associated with a dramatic improvement in their level of consciousness. In non-tolerant patients, the mean [C.sub.50] was close to the upper limit of the therapeutic plasma concentration of meprobamate given by our toxicological laboratory ([is less than or equal to] 200 [micro]mol/L). Baud FJ, Galliot R, Borron SW, Debray M, Bismuth C. Reanimation Medicale et Toxicologique-INSERM U26--University Paris 7 Laboratoire de Biomathematiques, University Paris V, France, George Washington University, Washington DC, USA

Background: Acute toxicity is the expression of dose-effect relationships, toxicokinetics, toxicodynamics, and their interactions. Some compounds (benzodiazepines) appear to involve single mechanism toxicity. Others (paraquat, ethylene glycol) induce toxic effects through multistage mechanisms. We propose `dynetics' as a general term to describe the chronology of toxic effects, their individual duration and magnitude, and the dependence of one effect on another. In other words, dynetics represents the kinetics of dynamic events. Attention to dynetics may allow the refinement of our understanding of the contribution of individual mechanisms of toxicity demonstrated in experimental studies to the clinical presentation of poisoning in humans. The medical interests of such a concept are manifold. In acetaminophen poisoning, the diagnosis and prognosis within the first 24 h rest on serum acetaminophen concentrations, while late in the course of poisoning they depend on serum ASAT and ALAT. Obtaining the right diagnostic test depends on an understanding of dynetics. Specific treatments (antidotes, extracorporeal elimination) cannot be adequately evaluated without considering this concept. Experimental and clinical data show that fomepizole can prevent life threatening intoxication at a time when plasma ethylene glycol concentrations are high and glycolate concentrations are low. Fomepizole appears to be devoid of any therapeutic value, however, when plasma glycolate and oxalate concentrations are high and plasma ethylene glycol concentrations are low. The efficacy of many antidotes will be strongly dependent on the dynetic phase at which they are administered. Dynetics demands that we classify patients in clinical trials according to the dynetic phase of the poisoning in order to more precisely assess treatment efficiency. Conclusion: Dynetics belongs to the area of integrated physiological systems. Dynetics attempts to integrate the mechanisms of toxicity in order to explain the pathophysiology of a poisoning and the corresponding signs and symptoms observed throughout its course. The consequence of this approach is to assess more precisely the efficacy of specific treatments. Baud FJ, Borron SW, Bismuth C. Reanimation Medicale et Toxicologique, Hopital Lariboisiere, INSERM U26, Universite Paris VII, 75010 Paris, France

Digitalis poisoning is a rare life-threatening event with a mortality rate before the availability of specific treatment of about 15%. Digitalis poisoning may result from an overdose of either digoxin or digitoxin. The nature of the cardiac glycoside will dramatically modify the duration of life-threatening events, in the range of 36 hours for digoxin and of 5 days for digitoxin. Furthermore, digitalis poisoning may result either from the ingestion of a single high dose of digitalis or from a too high dose during a chronic treatment. This presentation will be focused on the prognosis and the management of acute digitalis poisoning resulting from the ingestion of a single high dose of cardiac glycoside. At the time of presentation several prognostic factors can be determined in order to assess the likelihood of life-threatening cardiac events. These prognostic factors include: the age, the sex, a past history of cardiac disease, an atrioventricular block whatever its degree, and hyperkalemia. According to these prognostic factors, the probability of death ranges from about 2% to 75%. In our experience a serum potassium level greater than or equal to 6.4 mmol/L, in a digitalis poisoning without any factors of potassium overload is associated with a mortality rate of about 90%. The death of acute digitalis poisoning is mainly related to the occurrence of cardiac events either conduction abnormalities or ventricular arrhythmias. Supportive treatment including atropine, endocardial pace-maker, and anti-arrhythmics are efficient. However, their use was not associated with an improvement in the prognosis of digitalis poisoning. Digoxin-specific Fab fragments have been shown efficient in digitalis poisoning regarding the reversal of conduction abnormalities, ventricular arrhythmias, and hyperkalemia. However, there are no prospective studies comparing the mortality rate of acute digitalis poisonings treated with or without Fab fragments. Furthermore, the mortality rate of digitalis poisonings treated with specific Fab fragments in previous large series appears quite similar to that observed without Fab fragments. In a retrospective study we observed an improvement in the prognosis of digitalis poisoning only when Fab fragments were used as first line antidote in patients without life-threatening cardiac disturbances. Fab fragments are expensive and the minimal efficient dose of Fab fragments remains a matter of debate. Our data suggest that a curative dose (equimolar neutralization) is recommended for patients with life threatening intoxication while a prophylactic dose (half-equimolar neutralization) should be recommended if the patient had poor prognosis factors without acute life-threatening toxicity. However, some patients may require an additional dose of Fab fragments more especially if digitalis poisoning results from the ingestion of digitoxin. Baud FJ, Lapostolle F, Borron SW, Bismuth C. Reanimation Medicale et Toxicologique-INSERM U26--University Paris 7, SAMU 93

The questionable reliability of the conventional procedures for detection of ingested drug packages triggered us to evaluate the accuracy of a method of contrast study of the bowel in 23 nonsurgically managed cocaine body-packers. A single dose (60 mL) of a water-soluble contrast compound (amidotrizoate + meglumine) was given orally after initial clinical examination and drug detection in urine. Thereafter, roentgenograms were performed daily after spontaneous passage until obtaining two packet-free stools and negative views. Roentgenograms showed packages when performed at least 3 h after the ingestion of the contrast compound. Sensitivity and specificity of the method with respect to the detection of residual packets in the body, assessed by subsequent examination of stools, was good and did not diminish as the number of packages decreased during the time spent in ward. No side-effects were observed. We conclude that oral administration of a water-soluble contrast compound is an easily performed, efficient, and safe method for the nonsurgical management of cocaine body-packers.

1 The half-life of carbon monoxide (CO) in blood was studied retrospectively in 26 fire victims and in 19 cases of CO poisoning. Normobaric oxygen therapy was administered via mechanical ventilation in 19 fire victims, and by facial mask to the rest of the casualties.2 Arterial pH was significantly lower (P < 0.05) and PaO2significantly greater (P<0.01) in the mechanically ventilated fire victims compared to the spontaneously breathing fire victims.3 The blood CO half-lives were 91 ± 38 min for the 26 fire victims and 87±40 min for the 19 pure CO intoxica tions.4 The blood CO half-lives were 92 ± 40 min for the 19 mechanically ventilated fire victims and 87 ± 37 min for the 26 spontaneously breathing subjects.5 We conclude that the elimination of CO from blood was a slow process with no significant effects on the blood CO half-life of either the cause of the CO poisoning or the mode of normobaric oxygen therapy. These data suggest that enhancement of the elimination of carbon monoxide by normobaric oxygen in both pure CO poisoning and fire victims is more difficult to achieve and more complex than has previously been reported.

Letter