The role of hydrogen cyanide and carbon monoxide in fire casualties: a prospective study

Determinations of blood cyanide and carboxyhemoglobin concentrations were performed in 18 victims found dead in buildings after fires during a 2-year period. The results indicated that 50% of the victims had been exposed to toxic levels of hydrogen cyanide and 90% to toxic levels of carbon monoxide. Lethal concentrations of carbon monoxide were found in 83% of the victims. In one case a lethal blood cyanide but a non-toxic blood carboxyhemoglobin value was found. It is concluded that carbon monoxide appears to be more important than hydrogen cyanide as a toxic agent in the fire atmosphere, but cyanide poisoning without carbon monoxide poisoning may, under certain circumstances, be the cause of death in fire victims.     

Analysis of carboxyhemoglobin and cyanide in blood from victims of the Dupont Plaza Hotel fire in Puerto Rico

Ninety-seven people died from a fire that occurred in the Dupont Plaza Hotel in Puerto Rico on 31 Dec. 1986. All, except four who died later in the hospital, were found dead at the scene. All of the fatalities at the hotel (except for eight) were burned beyond recognition. Blood from seventy-eight of the victims was screened for carboxyhemoglobin at the Institute for Forensic Sciences in Puerto Rico and was then sent to the National Institute of Standards and Technology, Gaithersburg, Maryland, for analysis of carboxyhemoglobin and cyanide concentrations. The blood data indicated that carbon monoxide and hydrogen cyanide, singly or combined, were probably not responsible for the majority of the deaths that occurred in the badly burned victims. On the other hand, the significantly higher carboxyhemoglobin in the nonburned victims indicated that carbon monoxide alone or combined with hydrogen cyanide probably played a major role in the cause of their deaths.     

Significance of hydrocyanic acid formation during fires]

Cyanide concentrations of blood samples from fire victims autopsied in the Institute of Legal Medicine, Munich, have been determined. In 25% of 48 analyzed cases cyanide concentrations from 0.52 microgram to 6.24 microgram Cyanide/ml blood have been detected. These results are compared to former studies and the higher mean level in our collective is emphasized. The importance of hydrocyanid acid in the toxicity of fire gases is evidently greater, than assumed. Hydrocyanic acid may be produced from nitrogen continaing polymers during combustion. The quote of these polymers in clothing, furniture, and also in equipment of cars is increasing. Therefore, it is necessary to take more notice of the formation of hydrocyanic acid during combustion, even though carbon monoxide is in general the main toxic agent in fire gases.     

Potential for error when assessing blood cyanide concentrations in fire victims.

The present study explores toxicologic significance of blood cyanide concentrations in fire victims. Headspace gas chromatography was used for cyanide detection. Analysis of blood samples from ten fire victims (postmortem interval = 8 h to 3 to 5 d) detected zero to 11.9 mg/L of cyanide and a large difference in cyanide concentrations among victims. Carboxyhemoglobin (COHb) saturation was in the range of 24.9 to 84.2%. To examine the effects of methemoglobinemia and postmortem interval on blood cyanide concentrations in fire victims, an experiment was carried out using rabbits as the animal model. The rabbits were sacrificed by intramuscular injection of 1 mL/kg 2% potassium cyanide 5 min after intravenous injection of 0.33 mL/kg of 3% sodium nitrite (Group A, n = 3) or physiological saline (Group B, n = 6). Average methemoglobin contents immediately before potassium cyanide administration were 6.9 and 0.8% in Groups A and B, respectively. Average cyanide concentrations in cardiac blood at the time of death were 47.4 and 3.56 mg/L, respectively. When blood-containing hearts of the rabbits (n = 3 for Group B) were left at 46 degrees C for the first 1 h, at 20 to 25 degrees C for the next 23 h and then at 4 degrees C for 48 h, approximately 85 and 46% of the original amounts of blood cyanide disappeared within 24 h in Groups A and B, respectively. After the 72-h storage period, 37 and 10%, respectively, of the original amounts of cyanide remained in the blood. When the other three hearts in Group B were left at 20 to 25 degrees C for the last 48 h without refrigeration, cyanide had disappeared almost completely by the end of the experiment. The present results and those published in the literature demonstrate that the toxic effects of cyanide on fire victims should not be evaluated based solely on the concentration in blood.     

A study on house fire victims: age, carboxyhemoglobin, hydrogen cyanide and hemolysis.

Correlation among age, concentrations of carboxyhemoglobin and hydrogen cyanide, oxygen density and hemolysis were studied in 120 house fire victims. Victims aged over 60 years comprised approximately 50% of the pooled subjects. Blood samples were mainly collected from the left ventricle, but sometimes from both the right and left ventricles. The concentration of carboxyhemoglobin ranged from 1-95%, of which 71 persons (59.7%) died with carboxyhemoglobin concentrations below 60%. Carboxyhemoglobin concentrations below 10% were found in 9 persons (7.5%). Most of these cases involved the elderly persons. In this paper, we report on the death of elderly victims as a result of low carboxyhemoglobin concentrations. A significant correlation of blood carboxyhemoglobin concentrations existed between the right and left ventricles. The concentration of carboxyhemoglobin in the left ventricle was significantly higher than that in the right. Two out of 31 victims whose hydrogen cyanide concentrations were determined, succumbed to hydrogen cyanide poisoning, having a high concentration of hydrogen cyanide and a low concentration of carboxyhemoglobin. On analysis, oxygen density was found to be low in 13 persons. A negative correlation was shown between carboxyhemoglobin concentration and hemolysis. Inasmuch as hemolysis may indicate the extent of heat dissociation, hemolysis should provide an index of carbon monoxide dissociation from carboxyhemoglobin. In the present study of victims, possible causes of death i.e., carbon monoxide gas poisoning, hydrogen cyanide poisoning, oxygen deprivation, burning, shock due to burns and others were estimated. The survival time for elderly victims was considered to be short.     

High cyanide level in a homicide victim burned after death: evidence of post-mortem diffusion.

Elevated levels of carbon monoxide and cyanide serve as evidence of intravital burning in fire victims. Hydrogen cyanide is released by combustion of nitrogen-containing organic material such as plastics and wool. We present a case of a man who died of haemopneumothorax caused by a stab wound. According to several eye witnesses the body was wrapped in a plastic sheet and burned 2 days after death with the aid of gasoline. No coal pigment was observed in the mucosa of the upper airways at autopsy. The blood sample taken from the pulmonary vessels 6 days after death disclosed a level of blood carboxyhaemoglobin of 4% and of blood cyanide of 10 mg/l. The low carboxy-haemoglobin level was consistent with the smoking habits of the victim. The thoracic cavity had been opened by burning of the intercostal soft tissue. This allowed hydrogen cyanide gas to enter the thoracic cavity and diffuse into the blood probably causing the high blood-cyanide level.     

The causes of death in fire victims

In 169 consecutive cases of autopsied fire victims about 50% had lethal levels of carboxyhemoglobin. Soot in the respiratory tract was found in about 90% of the cases. The age distribution of the fire victims showed significantly less persons in the 15-35-year group than should be expected according to the age distribution of the population, presumably due to greater agility of younger people. More than half of the fire victims had alcohol in the blood exceeding 0.05%, and alcohol intoxication should be considered accessary to many deaths in fire. The characteristic biphasic distribution of carboxyhemoglobin in fire victims together with other observations suggest that the principal causes of death are carbon monoxide followed by carbon dioxide poisoning and/or oxygen deficiency, while the influence of heat is considered to be of minor importance.     

The effect of sodium fluoride on the stability of cyanide in postmortem blood samples from fire victims

Assigning a level of significance to cyanide concentrations found in the blood of fire victims is often hampered by the fact that cyanide is inherently unstable in cadavers and in stored blood samples. A few researchers have proposed that sodium fluoride can be used to minimize the instability of cyanide in blood samples; however, controlled studies have not been performed to support validation of this hypothesis. To test the sodium fluoride hypothesis, both treated and control blood samples from 14 autopsied fire victims were tested over a 25-30 day period. A 2% concentration of sodium fluoride was added to the blood samples at the start of testing and the samples were refrigerated between testing intervals. Cyanide concentrations in the treated and control samples were measured between 9 and 11 days post treatment and between 25 and 30 days post treatment. A statistically significant difference was not present between blood cyanide concentrations in treated and control samples between 9 and 11 days. During this time period, although there were small statistically significant increases in both treated and untreated samples the fluctuations were minor. Since the treated and control samples did not exhibit instability between 9 and 11 days, it is not surprising that the sodium fluoride appeared to have no effect. However, a statistically significant difference between blood cyanide concentrations in treated and control samples was observed between 25 and 30 days. Those samples treated with sodium fluoride showed a reduction in blood cyanide variability with virtually no overall change, over a 25-30 day period when compared to control samples, while unconditioned samples showed a significant, average increase of 35%. Based on the findings of this study, it is recommended that 2% sodium fluoride be added to blood samples obtained from fire victims to reduce cyanide instability due to bacteriological activity.     

Cyanide, carboxyhemoglobin and blood acid-base state in animals exposed to combustion products of various combinations of acrylic fiber and gauze

In order to examine the usefulness of blood cyanide concentrations as an indicator of whether or not a victim was alive at the start of a fire, blood cyanide concentrations were measured in the bodies that we autopsied in our institute between January 1986 and March, 1987. In the present study, bodies with advanced decomposition were excluded. Thirty-six bodies were included: cyanide as well as carboxyhemoglobin (COHb) were detected in four charred bodies found at the scene of a fire. On the other hand, cyanide was not detected in any of the remaining 30 bodies except in two cases suspected of having ingested a cyanide compound. Rats and rabbits were made to inhale the combustion products of various combinations of acrylic fiber (hydrogen cyanide generating material when heated) and gauze (carbon monoxide generating material when heated). The exposure to the combustion products was continued until death in the rat and until apnea in the rabbit. The concentration of hydrogen cyanide in the exposure chamber and that of blood cyanide, at the time the animal died, correlated with the amount of acrylic fiber heated. In addition to differences in blood COHb and cyanide concentrations, there were also differences in blood gas concentrations between the acrylic fiber and the gauze groups. When the rabbits were switched to room air after the occurrence of apnea, the blood gas value began to normalize.     

Fire and suicide: a three-year study of self-immolation deaths.

Thirty-two self-immolation deaths by fire, representing about 1% of suicides, occurred in the province of Ontario (population 9 million), Canada, from 1986 through 1988. The victims, mostly male (male/female ratio, 26:6), were between 21 and 71 years old (mean age, 38 years). Although the scene of self-immolation was usually familiar to the deceased, some chose remote locations. Eleven were found dead in motor vehicles. An accelerant, usually gasoline, was used in most cases. Many of these individuals had, at some time, indicated their intent to commit suicide, a few by self-immolation, but only about half had a diagnosed psychiatric illness. Most of the victims had a reason to kill themselves, but the factors that motivated them to chose self-immolation by fire were uncertain. Fourteen individuals died in hospitals from severe burn complications. The remainder were found dead at the scene. The postmortem findings of soot in the airway and elevated carbon monoxide in the blood of most of these victims [the carboxyhemoglobin (COHb) concentration was in one case less than 10%, in ten cases greater than or equal to 10 to 50%, and in seven cases greater than 50%] were helpful in determining that the individuals were not only alive at the time of the fire but also that a significant number died from smoke inhalation and carbon monoxide poisoning. The highest levels of carbon monoxide were observed in victims discovered in motor vehicles.     

Hydrogen cyanide and carbon monoxide in blood of convicted dead in a polyurethane combustion: a proposition for the data analysis

Carbon monoxide is a well-known toxic component in fire atmospheres. However, the importance of hydrogen cyanide as a toxic agent in fire causalities is under discussion. A tragic polyurethane mattress fire provoked death of 35 convicts in a prison (Unit I, Olmos, Penitenciary Service of Buenos Aires Province, Argentina), in 1990. There is no report of any investigation carried out with such a large amount of victims in Argentina. Carboxihemoglobin (COHb) and hydrogen cyanide (HCN) were quantified in victims blood to elucidate the cause of the death. Saturation of COHb ranged between 4 and 18%, and HCN 2.0-7.2mg/l. These latter values were higher than the lethal levels reported in literature. Other toxic components routinely measured (ethanol, methanol, aldehydes and other volatile compounds) gave negative results on the 35 cases. Neither drugs of abuse nor psychotropics were detected. Statistical chi(2) analysis was applied to find differences between HCN and COHb concentrations. Saturation of COHb and HCN in blood were not independent variables (chi(2)=8.25). Moreover, the ratio COHb/HCN was constant (0.47+/-0.04). In order to evaluate the contribution of each toxic to the diagnosis, a lethal index was defined for each toxic (LI(CO) and LI(HCN)). The most probable cause of death could be inferred by a suitable plot of both indexes. The results indicated that death in the 35 fire victims was probably caused by HCN, generated during the extensive polyurethane decomposition provoked by a rapid increase of temperature.     

The application of NIST's Fire Dynamics Simulator to the investigation of carbon monoxide exposure in the deaths of three Pittsburgh fire fighters

A case is reported in which computer fire modeling was used to reevaluate a fire that killed three fire fighters. NIST's Fire Dynamics Simulator (FDS) was employed to model the fire in order to estimate the concentration of carbon monoxide present in the dwelling, which was the immediate cause of death of two of the fire fighters, who appear to have removed their face pieces in order to share available air. This estimate, along with an assumed respiration volume and known blood carboxyhemoglobin, was plugged into a standard equation to estimate the time of exposure. The model indicated that 27 min into the fire, the carbon monoxide concentration had already reached approximately 3600 ppm. At this concentration, and a respiration of 70 L/min, an estimated 3 to 8 min of exposure would have been required to accumulate the concentrations of carboxyhemoglobin (49, 44, and 10%) measured on the fire fighters at autopsy.     

Accidental poisonings involving carbon monoxide, heating systems, and confined spaces

Eleven incidents of carbon monoxide (CO) intoxication resulting in sixteen fatalities are reported. All of the cases involved heating systems as either the source or the means of distributing the CO. Blood samples were analyzed for ethanol and CO. Elevated blood CO saturations were found in 14 of the 16 victims while none of the victims had a blood ethanol concentration greater than 0.10% (w/v).     

Expert significance of evaluating volatile non-metal combustion products in analysis of blood from subjects dead in fire

Qualitative and quantitative composition and time course of formation of toxic substances during burning of polymers have been studied for two polymers, P-53-P and STBO. The most toxic gaseous products of nitrogen combustion are listed: hydrogen cyanide, nitrites, acid nitriles, aldehydes, aromatic carbohydrates, alkanes, and carbon oxide. Two expert conclusions are presented, including analysis of blood and viscera from subjects who died during fire. The following toxic substances were detected: carbon oxide, hydrogen cyanide, nitrites, acrylonitrile, acetonitrile, benzene, and methane in one case and the same products except carbon oxide in the other. The author considers that the list of tested toxic substances should be extended, particularly in cases with negative results of blood testing for carbon oxide and carboxyhemoglobin.     

Survival time and acting capability after fatal injury by sharp weapons

Autopsy reports from 109 victims of fatal injury from sharp weapons were investigated with regard to type and number of lesions, blood loss, degree of alcohol intoxication, survival time and acting capability after the injury. Of the 13 who died immediately, nine had penetrating lesions of the heart. This group also had the highest number of lesions. Sixty-four victims survived for some time. The survival time increased with decreasing number of lesions. The greatest blood loss and the highest blood alcohol concentrations were found in those who survived between 0.5 h and 1 h. Twenty-four victims were able to make physical efforts after the injury and the movements varied from a few steps to the running of several hundred meters. Decisive factors for decrease in survival time and acting capability are penetrating lesions to the heart and the great vessels, and multiplicity of injuries.     

Evaluation of human brain damage in fire fatality by quantification of basic fibroblast growth factor (bFGF), glial fibrillary acidic protein (GFAP) and single-stranded DNA (ssDNA) immunoreactivities

Burns and inhalation of toxic gases, including carbon monoxide (CO) and cyanide, which are produced by combustion, are major factors involved in fire death. The present study immunohistochemically investigated basic fibroblast growth factor (bFGF), glial fibrillary acidic protein (GFAP) and single-stranded DNA (ssDNA) in the brains of fire fatalities (n=49) to examine the differences between fatal burns and CO intoxication, compared with those in cardiac deaths (n=24) and mechanical asphyxiation cases (n=23). In acute fire fatality, neuronal ssDNA immunopositivity in the cerebral cortex of the parietal lobe was high in both fatal burns and fatal CO intoxication, but that of the pallidum was higher for CO intoxication than for burns. The number of neurons was decreased in prolonged fire deaths, irrespective of the severity of burns or CO intoxication, but glias were increased in cases of fatal burns. Prolonged deaths due to burns had a higher glial bFGF immunopositivity in the cortex and white matter, higher and lower glial GFAP immunopositivity in the cortex and white matter, respectively, and a low neuronal ssDNA immunopositivity in the cerebral cortex and hippocampus. In prolonged deaths due to CO intoxication, however, glial bFGF and GFAP immunopositivities were low at each site, but neuronal ssDNA immunopositivity showed a higher value. These observations suggest increased cerebral neuronal ssDNA immunopositivity to be a finding of vitality in acute fire death, and a neuronal loss accompanied by active glial responses after severe burns, and a neuronal loss and progressive apoptosis without glial responses after CO intoxication to be characteristic in prolonged death.     

Quantitative evaluation of volatile hydrocarbons in post-mortem blood in forensic autopsy cases of fire-related deaths

Volatile hydrocarbons in post-mortem blood from victims of fires were analyzed quantitatively by headspace gas chromatography mass spectrometry. The benzene and styrene concentrations in the blood were positively correlated with the carboxyhemoglobin (CO-Hb) concentration, which is evidence that the deceased inhaled the hydrocarbons and carbon monoxide simultaneously. By contrast, the concentrations of toluene and CO-Hb in the blood were not significantly correlated. This lack of correlation could be explained by two different sources of toluene, with low blood concentrations of toluene arising when the deceased inhaled smoke and high blood concentrations of toluene arising when the deceased inhaled petroleum vapor or other unknown vapors. The quantity of soot deposited in the respiratory tract was classified into four grades (-, 1+, 2+, 3+). The mean CO-Hb concentration in the 1+ soot group was significantly lower than those in the 2+ (p<0.05) and 3+ (p<0.01) soot groups. The blood CO-Hb concentrations in the 1+ soot group were all below 30%. Those indicated that the deceased aspirated smoke that contained both soot and carbon monoxide. The wide variation in CO-Hb concentrations for each soot classification could be caused by the different types of smoke produced by different materials. For example, petroleum combustion with a limited supply of oxygen, like in a compartment fire, may produce a large volume of dense black smoke that contains a large quantity of soot. Soot deposits in the airways and the blood CO-Hb concentration are basic and essential autopsy findings that are used to investigate fire-related deaths. The quantitative GC-MS analysis of blood volatile hydrocarbons can provide additional useful information on the cause of the fire and the circumstances surrounding the death. In combination, these three findings are useful for the reconstruction of cases.     

Scene investigation, identification, and victim examination following the accident of Galaxy 203: disaster preplanning does work

Galaxy Airlines Flight 203 crashed following takeoff from Reno-Cannon International Airport on 21 Jan. 1985. Sixty-eight persons on board the aircraft perished in the initial crash and resultant fire which followed. Two victims expired as a result of crash injuries within subsequent days and one passenger survived. A community disaster response plan was in place and had been practiced by local government agencies before this incident. The successes of this preplanned response, as well as methods of actual recovery, identification, and examination of the victims is presented.     

Toxicological findings in victims of traumatic deaths.

The toxicological findings from 6037 analyses of viscera obtained from victims of traumatic death are used to correlate the relative incidence of carbon monoxide, ethyl alcohol, narcotics, hypnotics, analgesics, and tranquilizers-antidepressants in deaths occurring under the following circumstances: fire related, asphyxia by hanging, by use of plastic bags, from physical obstruction of trachae, and by drowning; traumatic injury from impact of moving train, fall from height, and occupational accident; traumatic injury to pedestrian, driver, and passenger from vehicular accidents; and from violent death by shooting, stabbing, strangulation, and beating. The influence of alcohol, narcotic drugs, and tranquilizers on carbon monoxide can be seen in some of these traumatic deaths. Ethanol alone and in combination with other drugs was present in 42.3% and 19.5% of driver and pedestrian victims, respectively, of vehicular accidents in the year 1974. Comparative analysis is presented for the toxicological data obtained on victims of homicide (shooting, stabbing, strangulation, and beating) in New York City and similar data reported for victims of homicide in Detroit. In New York City 45.9% of such victims died while under the influence of alcohol or narcotic drugs, or both, with methadone predominating in the latter category. Tissue concentrations of drugs found in victims of traumatic death are presented. Diphenylhydantoin, diazepam, meperidine, and slow-acting barbiturates were found in normal therapeutic levels. Higher concentrations of amitriptyline, chlorpromazine, propoxyphene, short-acting barbiturates, and methadone were observed. The concentration of methadone in blood and brain (0.13 +/- 0.14 mg/100 ml) and in liver (0.53 +/- 0.42 mg/100 ml) in cases of traumatic death are not different from those observed in deaths classified as due to methadone overdose.     

Free radical production from controlled low-energy fires: toxicity considerations

Most fire departments respond within the first 5 min of notification of a fire. If fire victims are found at that stage by the firefighters, then incapacitation or death has occurred during the initial low-energy phase where smoke is being produced. Studies have shown that during this initial low-energy phase of the fire, gases commonly thought responsible for incapacitation or death are frequently not present in concentrations adequate to cause this result. In the current study free radicals, measured by electron spin resonance spectroscopy, were trapped in concentrations that we consider incapacitating, thus providing an explanation for "incapacitation without cause." This finding points the way to the design of more efficient temporary protective equipment for those who are in a high fire hazard environment, such as airline passengers, and suggests the idea of establishing a thermodynamic marker for the relative toxicity of building materials.