Forensic Science International 244 (2014) e30–e33
Contents lists available at ScienceDirect
Forensic Science International journal homepage: www.elsevier.com/locate/forsciint
Case Report
An amazing case of fatal self-immolation Veronique Alunni a,*, Gilles Grevin b,1, Luc Buchet b,2, Yvan Gaillard c,3, Ge´rald Quatrehomme a,4 a Laboratoire de Me´decine Le´gale et d’Anthropologie me´dico-le´gale, Universite´ de Nice Sophia Antipolis, Faculte´ de Me´decine, 28 Avenue de Valombrose, 06107 Nice Cedex 2, France b CEPAM (Cultures et Environnements, Pre´histoire, Antiquite´, Moyen Aˆge), UMR7264-CNRS Universite´ Nice Sophia Antipolis, Campus Saint-Jean d’Ange´ly 3, 24, avenue des Diables Bleus, 06357 Nice Cedex 4, France c Laboratory of Analytical Toxicology LAT LUMTOX, 800 avenue marie Curie, La Voulte sur Rhoˆne, France
A R T I C L E I N F O
A B S T R A C T
Article history: Received 16 May 2014 Received in revised form 9 July 2014 Accepted 26 August 2014 Available online 3 September 2014
We present a surprising case of suicide by self-immolation. A surveillance camera filmed the victim’s agony. We were able to ascertain that he survived 13 min after ignition. This case was the starting point for a review of the literature of the forensic elements required to correctly analyze cases of suicide by self-immolation and to establish the causal link between the burn lesions and death. The authors will focus on the forensic and medical aspects in favor of suicide by self-immolation and on the forensic aspects required to understand the ignition process, the circumstances surrounding the fire with a particular emphasis on determining whether an accelerant was doused on the victim before ignition. ß 2014 Elsevier Ireland Ltd. All rights reserved.
Keywords: Self-immolation Forensic medicine Accelerant Suicide Death Gasoline
1. Introduction In forensic medicine, self-inflicted burns are a frequent source of admissions to burn units [1,2]. Some cases of self-immolation are fatal, although statistically speaking deaths by accidental fire are more frequent. Criminal immolations are seldom observed [3,4]. All deaths where fire is involved must be considered suspicious and require specific investigations in order to identify the victim, determine intent and establish the causes and manner of death. This requires determining on the one hand whether the victim was burned before or after death, and, on the other hand, what the exact method of ignition and fire type were. This relies on a certain number of forensic elements: the degree of carbonization of the body, the presence or absence of soot in the respiratory tract,
* Corresponding author. Tel.: +33 612550583; fax: +33 492034827. E-mail addresses: [email protected] (V. Alunni), [email protected] (G. Grevin), [email protected] (L. Buchet), [email protected] (Y. Gaillard), [email protected] (G. Quatrehomme). 1 Tel.: +33 672713810, +33 492034827. 2 Tel.: +33 489881519; fax: +33 489881555. 3 Tel.: +33 475620524; fax: +33 475855458. 4 Tel.: +33 611675806; fax: +33 492034827. http://dx.doi.org/10.1016/j.forsciint.2014.08.030 0379-0738/ß 2014 Elsevier Ireland Ltd. All rights reserved.
the determination of blood and tissue concentrations of toxic substances (in particular HbCO and cyanide), and the presence of traumatic lesions. A review of the literature shows that the study of these parameters is not always straightforward. Discrepancies between findings are possible, making the determination of the causes and manner of death more difficult. In this paper, we present a case of suicide by self-immolation that was partially filmed by surveillance camera, enabling the investigators to follow the agony of the victim in the flames. This case is exceptional both because of the manner of ignition and the lengthy survival time. The goal of this paper is to help the forensic pathologist in his diagnostic approach with victims of suicide by self-immolation. The discussion will first focus on the medical and forensic aspects of the case that led to the conclusion of suicide by self-immolation. Secondly, the forensic aspects required to understand the ignition process will be reviewed here, with a particular emphasis on determining whether an accelerant was doused on the victim before ignition. 2. Case report The body of a man was found at night on the side of the road, next to a cemetery. He was on his back, naked and extensively
[(Fig._1)TD$IG]
V. Alunni et al. / Forensic Science International 244 (2014) e30–e33
e31
on any clothes or shoes. He finally came to a rest on his back 1 min later. Total survival time, from ignition to immobilization, is 13 min. Police investigation did not uncover any evidence in favor of a homicide. The car ran on diesel fuel and gasoline was found inside the car, which argued in favor of intentional placement before ignition. Based on the police investigation and the autopsy findings, the cause of death was declared to be extensive burn wounds incompatible with life, and the manner of death was determined to be suicide. 3. Discussion
Fig. 1. Body of the victim extensively covered in burn wounds.
covered in burn wounds (Fig. 1). No item of clothing and no shoes were found at the scene. The remains of a badly burned car were found 3 m from the body, with an open driver-side door. There was a trail of fuel from underneath the car down the slope of the road to the calves of the body. The external examination of the body showed third degree burns to 100% of the body surface. The skin was a brown-black color and showed signs of having retracted due to the heat. The desquamation of the epidermis revealed the red dermis below. Almost all of the hair had disappeared from the body, save for that at the top of the cranium. The left side of the face, including his eyelashes and eyebrows, was slightly more burnt than the right. The lower limbs were bent at the hips and at the knees. The burns were homogenous in both depth and distribution all over the body, save for the backside of the ankles and calves, which showed the start of carbonization (with destruction of all of the layers of skin and part of the muscle). These areas of carbonization were level with the fuel trail on the road. His face and neck were covered in soot. No other trauma lesion was found upon external examination. The full-body scan done before the autopsy found no evidence of traumatic or medical pathologies. The autopsy did not find evidence of internal lesion, projectile or bone fracture. The respiratory tract contained an important amount of soot and the tissues had a reddish hue to them. The toxicology assay found a 2% level of HbCO (spectrophotometry) and a blood cyanide level of 0.29 mg/L (GC–MS after head space injection, derivatization using chloramine T and the use of 13C15N as internal standard) while cyanide concentration in the lung was 0.58 mg/kg. No trace of accelerant (head space GC–MS) was found on the skin and lung samples. A city surveillance camera filmed the victim going around his car, alone, before the start of the fire. The ignition of the car was filmed, as was the rapid and violent development of the fire. The exploitation of this video coupled with the investigation data led to the conclusion that the subject was seated in his car when the fire was lit. Unfortunately, the view of the car was partially obstructed by a tree. He probably used the spark caused by a cigarette lighter or a match to ignite the gasoline contained in open gas tanks placed beside him. The clothes ignited first. The victim got out of the car, explaining the open car door, before falling to the ground not far from the fire center. For the next 3 min the victim cannot be seen on the video. He reappears on the video for 4 min, crawling along the road with great difficulty, in immediate proximity to the fire center, flames visible on both of his arms. He then disappears again, probably crawling on the ground behind a tree, before appearing again 8 min later, slowly turning over on himself on the ground, flames visible on both his arms and upper body. He no longer had
Self-immolation is an uncommon method of suicide [5] especially in developed countries. It is more frequent in Asia and Africa [6,7] and in particular on the Indian subcontinent where immolation by fire has a ritual connotation [8]. A review of the literature did not yield any answers as to why an individual would chose to die this way. It is possible that certain victims have an abnormal indifference to pain [9], as is the case with a number of schizophrenic individuals. Even though most of the victims of selfimmolation suffer from some type of mental disorder [6,10–12], cases of suicide by self-immolation have been observed in patients without a history of mental disorder [10,13,14]. In some cases, social and ethno cultural processes are a better predictor of selfimmolation than mental illness [7]. Immolation can be linked to the notion of purification by fire, or sacrifice [15]. Self-immolation is defined as the act of pouring a flammable liquid onto oneself and igniting it. There are many ways of achieving this. Accelerants are the main agents chosen both in selfimmolation victims and victims of assault [16], because it is difficult to set fire to a living individual. Gasoline is the most often used accelerant [17,18] because it is the easiest to obtain [19]. Alcohol and methylated spirits can also be used [13], as well as kerosene [10,11,14], benzene [20] and paraffin [5], products that are available to populations of lower socioeconomic status. Some self-immolation victims chose to forego liquid accelerant [5,14,20,21] in favor of propane–butane gas [20]. Burns due to electricity, dangerous liquids and caustic agents are excluded from this discussion. In cases of immolation, assessment of the cause and manner of death is based on autopsy, toxicological, histological examinations and police investigation records. It is important to differentiate the homicides disguised as suicides [22], and the cases of suicide that initially appeared to be homicide. These are rare cases of selfimmolation combined with other methods of suicide [21,23–25], where the investigations are complicated by the presence of multiple fatal lesions. Certain scenarios are so complex that the case cannot be resolved if the manner of ignition is not elucidated. This requires establishing whether an accelerant was used, and whether this accelerant was poured on the victim before the fire was started. To this end, analyzing the burn pattern of the clothing and of the body yields essential information. On the one hand, heavier fabrics such as jeans can be difficult to completely consume, and remnants of clothing are often left after the fire dies out. If the carbonization of the clothing is incomplete, black streaks can be observed on the clothing, resulting from the ignition of the fabric in contact with the accelerant. The distribution of the black streaks indicates the way it was poured on the victim before ignition. On the other hand, synthetic fabrics such as nylon are easily flammable and can thus disappear completely, releasing cyanide. Blood cyanide levels above 0.20 mg/ L are considered toxic [26]. High blood cyanide levels combined with the presence of high levels of cyanide in the lung tissue is consistent with the inhalation of toxic vapors either from the combustion of synthetic fabrics (i.e. the victim’s clothing) or with
e32
V. Alunni et al. / Forensic Science International 244 (2014) e30–e33
the combustion of the car seats and plastics. In our case, the video shows that the subject quickly got out of the car and continued to burn in the open, thus making it less likely that his cyanide levels are related to the combustion of the car seats and plastics. In cases where the clothes are completely consumed by the fire, it is then necessary to examine the burn distribution on the body. In our case, the area shaped like underwear, which appeared less burnt than the rest of the body, led us to think that the victim was initially covered in clothes that did protect his body from the flames. The body was extensively burnt, but not excessively so. Only his calves showed signs of carbonization. This was where a pool of accelerant had sloped down the side of the road and come to rest, creating a localized source of more intense combustion. According to a review of the literature, the use of an accelerant is responsible for a higher rate of mortality [27] and an uneven distribution of the depth of the burn wounds, and in particular with deeper burn wounds on the front of the body where the flammable liquid is poured [10]. Our victim had a harmonious burn distribution, the top of his head was not particularly burnt, and some hair was still visible. The toxicological analysis of the skin and lung fragments did not reveal the presence of accelerant. This means that no accelerant was poured onto the body or over the head. The police investigation revealed that the car used diesel fuel but that gasoline was found inside the car. This is what led the investigators to conclude that a spark caused the ignition of the gasoline vapors, setting both the victim and his clothing on fire. The strong heat produced by the ignition may have singed the eyelashes and eyebrows [28]. However, the fact that the victim remained in contact with the heat and the flames for close to 15 min certainly explains the near-complete combustion of the eyebrows and eyelashes. After the manner of death has been elucidated, the other aspect of this discussion is the determination of the cause of death. In cases of suicide by self-immolation, a number of pathological processes can lead to death [29]. These are: - Flash fire resulting from the use of accelerant, leading to laryngospasm and bronchospasm with the inhalation of burning hot air. - Intoxication by carbon monoxide and cyanide [26] if certain materials such as rigid polyurethane burned in close proximity to the victim. - Extensive burn wounds leading to heat shock or heat rigor. It is often difficult to ascertain to what degree the thermal injury was fatal, since the victim usually continues to burn after death. In our case, video surveillance data confirms that the burn injuries where sustained while the victim was still alive. It is therefore possible to conclude that thermal injury was a major cause of death, which is rare in the literature. Other causes of death, including trauma, must also be ruled out before concluding to suicide by self-immolation. It is also required to establish whether the victim was indeed alive when set on fire. To determine whether an individual was alive at the time of ignition, it is important to record whether any soot is found in the respiratory tract. According to many authors, the majority of self-immolation victims have soot in their respiratory tracts [4,10,30]. However, the presence of soot is not mandatory [31], especially in cases of flash fires, because death occurs very quickly [29]. The absence of soot does not mean that the victim died before ignition; it can just mean that other pathological processes were more prominent. The autopsy of our victim revealed a considerable amount of soot in the respiratory tract, which we attributed to his chosen method of self-immolation, affording him a surprisingly long survival time.
The results of the carbon monoxide assay must also be interpreted. Positive post-mortem blood carbon monoxide levels are associated with individuals who were alive at the time of the fire. However, null blood HbCO levels have been described in the literature [10,13,31], in cases where there is evidence that the victim was alive when the fire was lit. These findings have been attributed to open-air fires [14,30], which are more frequent than indoor or closed space immolations [1,5,13,30]. As a general rule, in deaths by self-immolation, levels of HbCO are rarely very high [17,31,32], especially in fragile subjects [33]. Low levels of HbCO are an indirect argument in favor of the use of accelerant [29–34], and/or flash fires. In these cases, death occurs rapidly probably because of the severity of the burn lesions and the asphyxia [35,36]. In cases where the victim is in a vehicle when the fire is lit (second most common scene for immolations [17]), HbCO levels described have been quite variable. Certain authors have described levels of HbCO superior to 50% due to the confined atmosphere [10,14,17,29,30], while others have reported much lower levels in cases of rapid death [22,37]. Cases where the victim succumbed to lesions such as accident trauma before the fire ignited will not be discussed here. In our case, the victim had low HbCO levels, because he quickly exited the burning vehicle, and continued to burn in the open air from the radiant heat, despite there being few flames. So, it appears that the relationship between the presence of HbCO and the manner of death is an interesting parameter to consider [30], however interpretation must remain prudent. The toxicological analysis must also look for alcohol and drugs [6,38]. However, these assays are of little interest in establishing the manner of death [4]: victims of suicide by self-immolation are often alcohol and drug-free at the time of death [10,13]. The survival time of the victim, on top of the unusual manner of ignition, make this case stand out from other published cases of suicide by self-immolation. We have not found other cases in the literature where the survival time can be ascertained with precision (in the absence of medical intervention). Generally speaking, survival time is correlated to the degree and extensiveness of the burn lesions [29], and victims having used accelerant died quickly from burn injury and asphyxia [10]. In our case, it was possible to determine that the victim was in contact with the fire for 13 min before dying, which is a very long survival time, with consequently a heavy cover of soot on the respiratory tract. This lengthy survival time is due to the fact that the victim did not douse himself in accelerant. The analysis of the surveillance video and the photographs taken on the scene show that the victim remained on the ground for 13 min, crawling and rotating on himself, singed by low-intensity flames and burned by the very high heat radiating from the fire in close proximity, but without being in the center of the fire itself. This combination has been rarely described [39]. Black marks, associated with traces of skin and fatty tissue on the ground, trace the different positions of the burning body. These show that he made little progress along the ground from the moment he fell to the moment that he died. This highlights the importance of the initial assessment of the scene, and the consideration that must be given to the black marks indicating where the victim burned. Being able to ascertain whether the victim was doused in accelerant before ignition is very important to the overall comprehension of the manner and causes of death in cases of immolation. Arguments in favor of this dousing in accelerant are short survival times, minimal soot in the airways, low HbCO levels (or even null HbCO levels), disharmonious burn injuries with areas of more advanced carbonization on the doused areas of the body. Traces of accelerant can sometimes be found on the skin and lung samples taken during the autopsy, on the clothing (if clothing remains), and on the black streaks corresponding to accelerant projections. None of these criteria were present in our case; death
V. Alunni et al. / Forensic Science International 244 (2014) e30–e33
occurred slowly, with relatively superficial lesions distributed harmoniously on the body of the victim. There was no indication of foul play, which enabled us to conclude to suicide by selfimmolation and the cause of death was determined to be the result of extensive thermal injury. 4. Conclusion In the context of self-immolation, certain particular elements are essential to understanding how the fire was lit. It is important to be able to determine whether an accelerant was used and whether it was doused on the victim before the fire was lit. The use of accelerant is associated with black streaks on the clothing corresponding to splashes of accelerant, a disharmonious distribution of the burn lesions on the body, very little soot in the airway and low levels of HbCO. Survival time is considerably shortened by the use of accelerant. Suicide is strongly suggested by the absence of other fatal lesions, evidence that the subject was alive upon ignition, and death resulting from burn injury, flash burn, cyanide or carbon monoxide poisoning. The investigation must also show that the victim acting alone could have carried out the suicide. The elements outlined above should be considered when determining whether accelerant was used. The police investigation does not usually yield any indication of foul play. Understanding the circumstances surrounding ignition and collecting evidence in favor of the use of accelerant are fundamental to determining whether the immolation is a suicide or a homicide. The case presented here is exceptional because of the manner of ignition and the surprisingly long survival time. Both of these elements could have been deducted from the detailed forensic and police investigation, but the surveillance video brought additional information of the type that had not been described in the literature before. Acknowledgment The authors would like to thank very much Cathy Francino for her kind English language editing of the manuscript. References [1] J.M. O’Donoghue, J.L. Panchal, S.T. O’Sullivan, M. O’Shaughnessy, T.P.F. O’Connor, H. Keeley, M.J. Kelleher, A study of suicide and attempted suicide by self immolation in an Irish psychiatric population: an increasing problem, Burn 24 (1998) 144–146. [2] D.R. Cameron, S.P. Pegg, M. Muller, Self-inflicted burns, Burns 23 (6) (1997) 511–522. [3] L. Fanton, K. Jdeek, S. Tilhet-Coartet, D. Malicier, Criminal burning, Forensic Sci. Int. 10 (158(2–3)) (2006) 87–93. [4] Y. Bu¨yu¨k, Y. Koc¸ak, Fire-related fatalities in Istanbul, Turkey: analysis of 320 forensic autopsy cases, J. Forensic Leg. Med. 16 (2009) 449–454. [5] A. Sukhai, C. Harris, R.G. Moorad, M.A. Dada, Suicide by self immolation in Durban, South Africa: a five-year retrospective review, Am. J. Forensic Med. Pathol. 23 (3) (2002) 295–298. [6] A. Ahmadi, Suicide by self-immolation: comprehensive overview, experiences and suggestions, J. Burn Care Res. 28 (1) (2007) 30–41. [7] A. Ahmadi, R. Mohammadi, D. Stavrinos, A. Almasi, D.C. Schwebel, Self-immolation in Iran, J. Burn Care Res. 29 (2008) 451–460. [8] T. Kanchan, A. Menon, R.G. Menezes, Methods of choice in completed suicides: gender differences and review of literature, J. Forensic Sci. 54 (2009) 938–942.
e33
[9] D.O. Topp, Fire as a symbol and as a weapon of death, Med. Sci. Law 13 (1973) 79–86. [10] P. Leth, M. Hart-Madsen, Suicide by self-incineration, Am. J. Forensic Med. Pathol. 18 (1) (1997) 113–118. [11] A.R. Mabrouk, A.N.M. Omar, K. Massoud, M.M. Sherif, N. El Sayed, Suicide by burns: a tragic end, Burns 25 (1999) 337–339. [12] S.A. Wiechman, D.M. Ehde, B.L. Wilson, D.R. Patterson, The management of selfinflicted burn injuries and disruptive behavior for patients with borderline personality disorder, J. Burn Care Rehabil. 21 (2000) 310–317. [13] M.A. Rothschild, H.J. Raatschen, V. Schneider, Suicide by self-immolation in Berlin from 1990 to 2000, Forensic Sci. Int. 124 (2001) 163–166. [14] M.J. Shkrum, K.A. Johnston, Fire and suicide: a three-year study of self-immolation deaths, J. Forensic Sci. 37 (1) (1992) 208–221. [15] S.C. Inamdar, R.A. Oberfiel, E.B. Darrel, A suicide by self-immolation-psychosocial perspectives, Int. J. Psychiatry 29 (1983) 130–133. [16] C.C. Malic, R.O. Karoo, O. Austin, A. Phipps, Burns inflicted by self or by others – an 11 years snapshot, Burns 33 (1) (2007) 92–97. [17] A.R. Copeland, Suicidal fire deaths revisited, Z. Rechtsmed. 95 (1) (1985) 51–57. [18] N. Franchitto, C. Faurie, L. Franchitto, V. Minville, N. Telmon, D. Rouge´, Selfinflicted burns: the value of collaboration between medicine and law, J. Forensic Sci. 56 (3) (2011) 638–642. [19] A. Rashid, J.P. Gowar, A review of the trends of self-inflicted burns, Burns 30 (2004) 573–576. [20] O. Hadjiiski, P. Todorov, Suicide by self-inflicted burns, Burns 22 (5) (1996) 381–383. [21] M. Bohnert, M.A. Rothschild, Complex suicide by self-incineration, Forensic Sci. Int. 131 (2003) 197–201. [22] J.R. Gaur, S.K. Sangwan, I. Singh, K. Thukral, Evaluation of physical evidences in a burn case, Med. Sci. Law 33 (1993) 75–78. [23] E.E. Tu¨rk, S. Anders, M. Tsokos, Planned complex suicide. Report of two autopsy cases of suicidal shot injury and subsequent self-immolation, Forensic Sci. Int. 139 (2004) 35–38. [24] T.W. Adair, A. Fisher, Suicide with associated acts of arson: two cases from Colorado, J. Forensic Sci. 51 (4) (2006) 893–895. [25] G. Cave Bondi, L. Cipolloni, E. Parroni, R.R. Cecchi, A review of suicide by burning in Rome between 1947–1997 examined by the Pathology Department of the Institute of Forensic Medicine, University of Rome La Sapienza, Burns 27 (2001) 227–231. [26] Y. Caplan, R. Altman, Microdetermination of cyanide in fire fatalities, in: Presented at the Annual Meeting of the American Academy of Forensic Sciences, 18 February, Washington, DC, 1976. [27] S. Pegg, J.J. Gregory, P.G. Hogan, I.W. Mottarelly, L.F. Walker, Epidemiological pattern of adult burn injuries, Burns 5 (1979) 326–334. [28] M. Bohnert, D. Ropohl, S. Pollak, Clinical findings in the medico-legal investigation of arsonists, J. Clin. Forensic Med. 6 (1999) 145–150. [29] I. Gerling, C. Meissner, A. Reiter, M. Oehmichen, Death from thermal effects and burns, Forensic Sci. Int. 115 (2001) 33–41. [30] F. Makhlouf, J.C. Alvarez, G.L. de la Grandmaison, Suicidal and criminal immolations: an 18-years study and review of the literature, Leg. Med. 13 (2) (2011) 98–102. [31] M. Bohnert, C.R. Werner, S. Pollak, Problems associated with the diagnosis of vitality in burned bodies, Forensic Sci. Int. 135 (2003) 197–205. [32] H. Gormsen, N. Jeppesen, A. Lund, The causes of death in fire victims, Forensic Sci. Int. 24 (1984) 107–111. [33] M. Yoshida, J. Adachi, T. Watakiki, Y. Tatsuno, N. Ishida, A study on house fire victims: age, carboxyhemoglobin, hydrogen cyanide and hemolysis, Forensic Sci Int. 52 (1991) 13–20. [34] W. Schwered, E. Schultz, Carboxyhaemoglobin and methaemoglobin findings in burnt bodies, Forensic Sci. Int. 12 (1978) 233–235. [35] B. Marc, P. Blanchet, L. Boniol, Domestic aerosol and flash fire, Burns 27 (2001) 783–784. [36] E. Wilk, D. Linder, R. Vock, Untersuchungen zum inhalations-hitzeschock, Arch. Kriminol. 203 (1999) 159–169. [37] C.S. Hirsch, R.O. Bost, S.R. Gerber, M.E. Cowan, L. Adelson, I. Sunshine, Carboxyhemoglobin concentrations in flash fire victims: report of six simultaneous fire fatalities without elevated carboxyhemoglobin, Am. J. Clin. Pathol. 68 (1977) 317–320. [38] F. Zor, M. Deveci, M. Bozkurt, S. Dikkatli, H. Duman, M. Sengezer, Psychological evaluation of self-inflicted burn patients: suicide or parasuicide? Burns 31 (2005) 178–181. [39] D.P. Wirthwein, J.E. Pless, Carboxyhemoglobin levels in a series of automobile fires, Am. J. Forensic Med. Pathol. 17 (1996) 117–123.
©2014 Elsevier
Contents lists available at ScienceDirect
Forensic Science International journal homepage: www.elsevier.com/locate/forsciint
Case Report
An amazing case of fatal self-immolation Veronique Alunni a,*, Gilles Grevin b,1, Luc Buchet b,2, Yvan Gaillard c,3, Ge´rald Quatrehomme a,4 a Laboratoire de Me´decine Le´gale et d’Anthropologie me´dico-le´gale, Universite´ de Nice Sophia Antipolis, Faculte´ de Me´decine, 28 Avenue de Valombrose, 06107 Nice Cedex 2, France b CEPAM (Cultures et Environnements, Pre´histoire, Antiquite´, Moyen Aˆge), UMR7264-CNRS Universite´ Nice Sophia Antipolis, Campus Saint-Jean d’Ange´ly 3, 24, avenue des Diables Bleus, 06357 Nice Cedex 4, France c Laboratory of Analytical Toxicology LAT LUMTOX, 800 avenue marie Curie, La Voulte sur Rhoˆne, France
A R T I C L E I N F O
A B S T R A C T
Article history: Received 16 May 2014 Received in revised form 9 July 2014 Accepted 26 August 2014 Available online 3 September 2014
We present a surprising case of suicide by self-immolation. A surveillance camera filmed the victim’s agony. We were able to ascertain that he survived 13 min after ignition. This case was the starting point for a review of the literature of the forensic elements required to correctly analyze cases of suicide by self-immolation and to establish the causal link between the burn lesions and death. The authors will focus on the forensic and medical aspects in favor of suicide by self-immolation and on the forensic aspects required to understand the ignition process, the circumstances surrounding the fire with a particular emphasis on determining whether an accelerant was doused on the victim before ignition. ß 2014 Elsevier Ireland Ltd. All rights reserved.
Keywords: Self-immolation Forensic medicine Accelerant Suicide Death Gasoline
1. Introduction In forensic medicine, self-inflicted burns are a frequent source of admissions to burn units [1,2]. Some cases of self-immolation are fatal, although statistically speaking deaths by accidental fire are more frequent. Criminal immolations are seldom observed [3,4]. All deaths where fire is involved must be considered suspicious and require specific investigations in order to identify the victim, determine intent and establish the causes and manner of death. This requires determining on the one hand whether the victim was burned before or after death, and, on the other hand, what the exact method of ignition and fire type were. This relies on a certain number of forensic elements: the degree of carbonization of the body, the presence or absence of soot in the respiratory tract,
* Corresponding author. Tel.: +33 612550583; fax: +33 492034827. E-mail addresses: [email protected] (V. Alunni), [email protected] (G. Grevin), [email protected] (L. Buchet), [email protected] (Y. Gaillard), [email protected] (G. Quatrehomme). 1 Tel.: +33 672713810, +33 492034827. 2 Tel.: +33 489881519; fax: +33 489881555. 3 Tel.: +33 475620524; fax: +33 475855458. 4 Tel.: +33 611675806; fax: +33 492034827. http://dx.doi.org/10.1016/j.forsciint.2014.08.030 0379-0738/ß 2014 Elsevier Ireland Ltd. All rights reserved.
the determination of blood and tissue concentrations of toxic substances (in particular HbCO and cyanide), and the presence of traumatic lesions. A review of the literature shows that the study of these parameters is not always straightforward. Discrepancies between findings are possible, making the determination of the causes and manner of death more difficult. In this paper, we present a case of suicide by self-immolation that was partially filmed by surveillance camera, enabling the investigators to follow the agony of the victim in the flames. This case is exceptional both because of the manner of ignition and the lengthy survival time. The goal of this paper is to help the forensic pathologist in his diagnostic approach with victims of suicide by self-immolation. The discussion will first focus on the medical and forensic aspects of the case that led to the conclusion of suicide by self-immolation. Secondly, the forensic aspects required to understand the ignition process will be reviewed here, with a particular emphasis on determining whether an accelerant was doused on the victim before ignition. 2. Case report The body of a man was found at night on the side of the road, next to a cemetery. He was on his back, naked and extensively
[(Fig._1)TD$IG]
V. Alunni et al. / Forensic Science International 244 (2014) e30–e33
e31
on any clothes or shoes. He finally came to a rest on his back 1 min later. Total survival time, from ignition to immobilization, is 13 min. Police investigation did not uncover any evidence in favor of a homicide. The car ran on diesel fuel and gasoline was found inside the car, which argued in favor of intentional placement before ignition. Based on the police investigation and the autopsy findings, the cause of death was declared to be extensive burn wounds incompatible with life, and the manner of death was determined to be suicide. 3. Discussion
Fig. 1. Body of the victim extensively covered in burn wounds.
covered in burn wounds (Fig. 1). No item of clothing and no shoes were found at the scene. The remains of a badly burned car were found 3 m from the body, with an open driver-side door. There was a trail of fuel from underneath the car down the slope of the road to the calves of the body. The external examination of the body showed third degree burns to 100% of the body surface. The skin was a brown-black color and showed signs of having retracted due to the heat. The desquamation of the epidermis revealed the red dermis below. Almost all of the hair had disappeared from the body, save for that at the top of the cranium. The left side of the face, including his eyelashes and eyebrows, was slightly more burnt than the right. The lower limbs were bent at the hips and at the knees. The burns were homogenous in both depth and distribution all over the body, save for the backside of the ankles and calves, which showed the start of carbonization (with destruction of all of the layers of skin and part of the muscle). These areas of carbonization were level with the fuel trail on the road. His face and neck were covered in soot. No other trauma lesion was found upon external examination. The full-body scan done before the autopsy found no evidence of traumatic or medical pathologies. The autopsy did not find evidence of internal lesion, projectile or bone fracture. The respiratory tract contained an important amount of soot and the tissues had a reddish hue to them. The toxicology assay found a 2% level of HbCO (spectrophotometry) and a blood cyanide level of 0.29 mg/L (GC–MS after head space injection, derivatization using chloramine T and the use of 13C15N as internal standard) while cyanide concentration in the lung was 0.58 mg/kg. No trace of accelerant (head space GC–MS) was found on the skin and lung samples. A city surveillance camera filmed the victim going around his car, alone, before the start of the fire. The ignition of the car was filmed, as was the rapid and violent development of the fire. The exploitation of this video coupled with the investigation data led to the conclusion that the subject was seated in his car when the fire was lit. Unfortunately, the view of the car was partially obstructed by a tree. He probably used the spark caused by a cigarette lighter or a match to ignite the gasoline contained in open gas tanks placed beside him. The clothes ignited first. The victim got out of the car, explaining the open car door, before falling to the ground not far from the fire center. For the next 3 min the victim cannot be seen on the video. He reappears on the video for 4 min, crawling along the road with great difficulty, in immediate proximity to the fire center, flames visible on both of his arms. He then disappears again, probably crawling on the ground behind a tree, before appearing again 8 min later, slowly turning over on himself on the ground, flames visible on both his arms and upper body. He no longer had
Self-immolation is an uncommon method of suicide [5] especially in developed countries. It is more frequent in Asia and Africa [6,7] and in particular on the Indian subcontinent where immolation by fire has a ritual connotation [8]. A review of the literature did not yield any answers as to why an individual would chose to die this way. It is possible that certain victims have an abnormal indifference to pain [9], as is the case with a number of schizophrenic individuals. Even though most of the victims of selfimmolation suffer from some type of mental disorder [6,10–12], cases of suicide by self-immolation have been observed in patients without a history of mental disorder [10,13,14]. In some cases, social and ethno cultural processes are a better predictor of selfimmolation than mental illness [7]. Immolation can be linked to the notion of purification by fire, or sacrifice [15]. Self-immolation is defined as the act of pouring a flammable liquid onto oneself and igniting it. There are many ways of achieving this. Accelerants are the main agents chosen both in selfimmolation victims and victims of assault [16], because it is difficult to set fire to a living individual. Gasoline is the most often used accelerant [17,18] because it is the easiest to obtain [19]. Alcohol and methylated spirits can also be used [13], as well as kerosene [10,11,14], benzene [20] and paraffin [5], products that are available to populations of lower socioeconomic status. Some self-immolation victims chose to forego liquid accelerant [5,14,20,21] in favor of propane–butane gas [20]. Burns due to electricity, dangerous liquids and caustic agents are excluded from this discussion. In cases of immolation, assessment of the cause and manner of death is based on autopsy, toxicological, histological examinations and police investigation records. It is important to differentiate the homicides disguised as suicides [22], and the cases of suicide that initially appeared to be homicide. These are rare cases of selfimmolation combined with other methods of suicide [21,23–25], where the investigations are complicated by the presence of multiple fatal lesions. Certain scenarios are so complex that the case cannot be resolved if the manner of ignition is not elucidated. This requires establishing whether an accelerant was used, and whether this accelerant was poured on the victim before the fire was started. To this end, analyzing the burn pattern of the clothing and of the body yields essential information. On the one hand, heavier fabrics such as jeans can be difficult to completely consume, and remnants of clothing are often left after the fire dies out. If the carbonization of the clothing is incomplete, black streaks can be observed on the clothing, resulting from the ignition of the fabric in contact with the accelerant. The distribution of the black streaks indicates the way it was poured on the victim before ignition. On the other hand, synthetic fabrics such as nylon are easily flammable and can thus disappear completely, releasing cyanide. Blood cyanide levels above 0.20 mg/ L are considered toxic [26]. High blood cyanide levels combined with the presence of high levels of cyanide in the lung tissue is consistent with the inhalation of toxic vapors either from the combustion of synthetic fabrics (i.e. the victim’s clothing) or with
e32
V. Alunni et al. / Forensic Science International 244 (2014) e30–e33
the combustion of the car seats and plastics. In our case, the video shows that the subject quickly got out of the car and continued to burn in the open, thus making it less likely that his cyanide levels are related to the combustion of the car seats and plastics. In cases where the clothes are completely consumed by the fire, it is then necessary to examine the burn distribution on the body. In our case, the area shaped like underwear, which appeared less burnt than the rest of the body, led us to think that the victim was initially covered in clothes that did protect his body from the flames. The body was extensively burnt, but not excessively so. Only his calves showed signs of carbonization. This was where a pool of accelerant had sloped down the side of the road and come to rest, creating a localized source of more intense combustion. According to a review of the literature, the use of an accelerant is responsible for a higher rate of mortality [27] and an uneven distribution of the depth of the burn wounds, and in particular with deeper burn wounds on the front of the body where the flammable liquid is poured [10]. Our victim had a harmonious burn distribution, the top of his head was not particularly burnt, and some hair was still visible. The toxicological analysis of the skin and lung fragments did not reveal the presence of accelerant. This means that no accelerant was poured onto the body or over the head. The police investigation revealed that the car used diesel fuel but that gasoline was found inside the car. This is what led the investigators to conclude that a spark caused the ignition of the gasoline vapors, setting both the victim and his clothing on fire. The strong heat produced by the ignition may have singed the eyelashes and eyebrows [28]. However, the fact that the victim remained in contact with the heat and the flames for close to 15 min certainly explains the near-complete combustion of the eyebrows and eyelashes. After the manner of death has been elucidated, the other aspect of this discussion is the determination of the cause of death. In cases of suicide by self-immolation, a number of pathological processes can lead to death [29]. These are: - Flash fire resulting from the use of accelerant, leading to laryngospasm and bronchospasm with the inhalation of burning hot air. - Intoxication by carbon monoxide and cyanide [26] if certain materials such as rigid polyurethane burned in close proximity to the victim. - Extensive burn wounds leading to heat shock or heat rigor. It is often difficult to ascertain to what degree the thermal injury was fatal, since the victim usually continues to burn after death. In our case, video surveillance data confirms that the burn injuries where sustained while the victim was still alive. It is therefore possible to conclude that thermal injury was a major cause of death, which is rare in the literature. Other causes of death, including trauma, must also be ruled out before concluding to suicide by self-immolation. It is also required to establish whether the victim was indeed alive when set on fire. To determine whether an individual was alive at the time of ignition, it is important to record whether any soot is found in the respiratory tract. According to many authors, the majority of self-immolation victims have soot in their respiratory tracts [4,10,30]. However, the presence of soot is not mandatory [31], especially in cases of flash fires, because death occurs very quickly [29]. The absence of soot does not mean that the victim died before ignition; it can just mean that other pathological processes were more prominent. The autopsy of our victim revealed a considerable amount of soot in the respiratory tract, which we attributed to his chosen method of self-immolation, affording him a surprisingly long survival time.
The results of the carbon monoxide assay must also be interpreted. Positive post-mortem blood carbon monoxide levels are associated with individuals who were alive at the time of the fire. However, null blood HbCO levels have been described in the literature [10,13,31], in cases where there is evidence that the victim was alive when the fire was lit. These findings have been attributed to open-air fires [14,30], which are more frequent than indoor or closed space immolations [1,5,13,30]. As a general rule, in deaths by self-immolation, levels of HbCO are rarely very high [17,31,32], especially in fragile subjects [33]. Low levels of HbCO are an indirect argument in favor of the use of accelerant [29–34], and/or flash fires. In these cases, death occurs rapidly probably because of the severity of the burn lesions and the asphyxia [35,36]. In cases where the victim is in a vehicle when the fire is lit (second most common scene for immolations [17]), HbCO levels described have been quite variable. Certain authors have described levels of HbCO superior to 50% due to the confined atmosphere [10,14,17,29,30], while others have reported much lower levels in cases of rapid death [22,37]. Cases where the victim succumbed to lesions such as accident trauma before the fire ignited will not be discussed here. In our case, the victim had low HbCO levels, because he quickly exited the burning vehicle, and continued to burn in the open air from the radiant heat, despite there being few flames. So, it appears that the relationship between the presence of HbCO and the manner of death is an interesting parameter to consider [30], however interpretation must remain prudent. The toxicological analysis must also look for alcohol and drugs [6,38]. However, these assays are of little interest in establishing the manner of death [4]: victims of suicide by self-immolation are often alcohol and drug-free at the time of death [10,13]. The survival time of the victim, on top of the unusual manner of ignition, make this case stand out from other published cases of suicide by self-immolation. We have not found other cases in the literature where the survival time can be ascertained with precision (in the absence of medical intervention). Generally speaking, survival time is correlated to the degree and extensiveness of the burn lesions [29], and victims having used accelerant died quickly from burn injury and asphyxia [10]. In our case, it was possible to determine that the victim was in contact with the fire for 13 min before dying, which is a very long survival time, with consequently a heavy cover of soot on the respiratory tract. This lengthy survival time is due to the fact that the victim did not douse himself in accelerant. The analysis of the surveillance video and the photographs taken on the scene show that the victim remained on the ground for 13 min, crawling and rotating on himself, singed by low-intensity flames and burned by the very high heat radiating from the fire in close proximity, but without being in the center of the fire itself. This combination has been rarely described [39]. Black marks, associated with traces of skin and fatty tissue on the ground, trace the different positions of the burning body. These show that he made little progress along the ground from the moment he fell to the moment that he died. This highlights the importance of the initial assessment of the scene, and the consideration that must be given to the black marks indicating where the victim burned. Being able to ascertain whether the victim was doused in accelerant before ignition is very important to the overall comprehension of the manner and causes of death in cases of immolation. Arguments in favor of this dousing in accelerant are short survival times, minimal soot in the airways, low HbCO levels (or even null HbCO levels), disharmonious burn injuries with areas of more advanced carbonization on the doused areas of the body. Traces of accelerant can sometimes be found on the skin and lung samples taken during the autopsy, on the clothing (if clothing remains), and on the black streaks corresponding to accelerant projections. None of these criteria were present in our case; death
V. Alunni et al. / Forensic Science International 244 (2014) e30–e33
occurred slowly, with relatively superficial lesions distributed harmoniously on the body of the victim. There was no indication of foul play, which enabled us to conclude to suicide by selfimmolation and the cause of death was determined to be the result of extensive thermal injury. 4. Conclusion In the context of self-immolation, certain particular elements are essential to understanding how the fire was lit. It is important to be able to determine whether an accelerant was used and whether it was doused on the victim before the fire was lit. The use of accelerant is associated with black streaks on the clothing corresponding to splashes of accelerant, a disharmonious distribution of the burn lesions on the body, very little soot in the airway and low levels of HbCO. Survival time is considerably shortened by the use of accelerant. Suicide is strongly suggested by the absence of other fatal lesions, evidence that the subject was alive upon ignition, and death resulting from burn injury, flash burn, cyanide or carbon monoxide poisoning. The investigation must also show that the victim acting alone could have carried out the suicide. The elements outlined above should be considered when determining whether accelerant was used. The police investigation does not usually yield any indication of foul play. Understanding the circumstances surrounding ignition and collecting evidence in favor of the use of accelerant are fundamental to determining whether the immolation is a suicide or a homicide. The case presented here is exceptional because of the manner of ignition and the surprisingly long survival time. Both of these elements could have been deducted from the detailed forensic and police investigation, but the surveillance video brought additional information of the type that had not been described in the literature before. Acknowledgment The authors would like to thank very much Cathy Francino for her kind English language editing of the manuscript. References [1] J.M. O’Donoghue, J.L. Panchal, S.T. O’Sullivan, M. O’Shaughnessy, T.P.F. O’Connor, H. Keeley, M.J. Kelleher, A study of suicide and attempted suicide by self immolation in an Irish psychiatric population: an increasing problem, Burn 24 (1998) 144–146. [2] D.R. Cameron, S.P. Pegg, M. Muller, Self-inflicted burns, Burns 23 (6) (1997) 511–522. [3] L. Fanton, K. Jdeek, S. Tilhet-Coartet, D. Malicier, Criminal burning, Forensic Sci. Int. 10 (158(2–3)) (2006) 87–93. [4] Y. Bu¨yu¨k, Y. Koc¸ak, Fire-related fatalities in Istanbul, Turkey: analysis of 320 forensic autopsy cases, J. Forensic Leg. Med. 16 (2009) 449–454. [5] A. Sukhai, C. Harris, R.G. Moorad, M.A. Dada, Suicide by self immolation in Durban, South Africa: a five-year retrospective review, Am. J. Forensic Med. Pathol. 23 (3) (2002) 295–298. [6] A. Ahmadi, Suicide by self-immolation: comprehensive overview, experiences and suggestions, J. Burn Care Res. 28 (1) (2007) 30–41. [7] A. Ahmadi, R. Mohammadi, D. Stavrinos, A. Almasi, D.C. Schwebel, Self-immolation in Iran, J. Burn Care Res. 29 (2008) 451–460. [8] T. Kanchan, A. Menon, R.G. Menezes, Methods of choice in completed suicides: gender differences and review of literature, J. Forensic Sci. 54 (2009) 938–942.
e33
[9] D.O. Topp, Fire as a symbol and as a weapon of death, Med. Sci. Law 13 (1973) 79–86. [10] P. Leth, M. Hart-Madsen, Suicide by self-incineration, Am. J. Forensic Med. Pathol. 18 (1) (1997) 113–118. [11] A.R. Mabrouk, A.N.M. Omar, K. Massoud, M.M. Sherif, N. El Sayed, Suicide by burns: a tragic end, Burns 25 (1999) 337–339. [12] S.A. Wiechman, D.M. Ehde, B.L. Wilson, D.R. Patterson, The management of selfinflicted burn injuries and disruptive behavior for patients with borderline personality disorder, J. Burn Care Rehabil. 21 (2000) 310–317. [13] M.A. Rothschild, H.J. Raatschen, V. Schneider, Suicide by self-immolation in Berlin from 1990 to 2000, Forensic Sci. Int. 124 (2001) 163–166. [14] M.J. Shkrum, K.A. Johnston, Fire and suicide: a three-year study of self-immolation deaths, J. Forensic Sci. 37 (1) (1992) 208–221. [15] S.C. Inamdar, R.A. Oberfiel, E.B. Darrel, A suicide by self-immolation-psychosocial perspectives, Int. J. Psychiatry 29 (1983) 130–133. [16] C.C. Malic, R.O. Karoo, O. Austin, A. Phipps, Burns inflicted by self or by others – an 11 years snapshot, Burns 33 (1) (2007) 92–97. [17] A.R. Copeland, Suicidal fire deaths revisited, Z. Rechtsmed. 95 (1) (1985) 51–57. [18] N. Franchitto, C. Faurie, L. Franchitto, V. Minville, N. Telmon, D. Rouge´, Selfinflicted burns: the value of collaboration between medicine and law, J. Forensic Sci. 56 (3) (2011) 638–642. [19] A. Rashid, J.P. Gowar, A review of the trends of self-inflicted burns, Burns 30 (2004) 573–576. [20] O. Hadjiiski, P. Todorov, Suicide by self-inflicted burns, Burns 22 (5) (1996) 381–383. [21] M. Bohnert, M.A. Rothschild, Complex suicide by self-incineration, Forensic Sci. Int. 131 (2003) 197–201. [22] J.R. Gaur, S.K. Sangwan, I. Singh, K. Thukral, Evaluation of physical evidences in a burn case, Med. Sci. Law 33 (1993) 75–78. [23] E.E. Tu¨rk, S. Anders, M. Tsokos, Planned complex suicide. Report of two autopsy cases of suicidal shot injury and subsequent self-immolation, Forensic Sci. Int. 139 (2004) 35–38. [24] T.W. Adair, A. Fisher, Suicide with associated acts of arson: two cases from Colorado, J. Forensic Sci. 51 (4) (2006) 893–895. [25] G. Cave Bondi, L. Cipolloni, E. Parroni, R.R. Cecchi, A review of suicide by burning in Rome between 1947–1997 examined by the Pathology Department of the Institute of Forensic Medicine, University of Rome La Sapienza, Burns 27 (2001) 227–231. [26] Y. Caplan, R. Altman, Microdetermination of cyanide in fire fatalities, in: Presented at the Annual Meeting of the American Academy of Forensic Sciences, 18 February, Washington, DC, 1976. [27] S. Pegg, J.J. Gregory, P.G. Hogan, I.W. Mottarelly, L.F. Walker, Epidemiological pattern of adult burn injuries, Burns 5 (1979) 326–334. [28] M. Bohnert, D. Ropohl, S. Pollak, Clinical findings in the medico-legal investigation of arsonists, J. Clin. Forensic Med. 6 (1999) 145–150. [29] I. Gerling, C. Meissner, A. Reiter, M. Oehmichen, Death from thermal effects and burns, Forensic Sci. Int. 115 (2001) 33–41. [30] F. Makhlouf, J.C. Alvarez, G.L. de la Grandmaison, Suicidal and criminal immolations: an 18-years study and review of the literature, Leg. Med. 13 (2) (2011) 98–102. [31] M. Bohnert, C.R. Werner, S. Pollak, Problems associated with the diagnosis of vitality in burned bodies, Forensic Sci. Int. 135 (2003) 197–205. [32] H. Gormsen, N. Jeppesen, A. Lund, The causes of death in fire victims, Forensic Sci. Int. 24 (1984) 107–111. [33] M. Yoshida, J. Adachi, T. Watakiki, Y. Tatsuno, N. Ishida, A study on house fire victims: age, carboxyhemoglobin, hydrogen cyanide and hemolysis, Forensic Sci Int. 52 (1991) 13–20. [34] W. Schwered, E. Schultz, Carboxyhaemoglobin and methaemoglobin findings in burnt bodies, Forensic Sci. Int. 12 (1978) 233–235. [35] B. Marc, P. Blanchet, L. Boniol, Domestic aerosol and flash fire, Burns 27 (2001) 783–784. [36] E. Wilk, D. Linder, R. Vock, Untersuchungen zum inhalations-hitzeschock, Arch. Kriminol. 203 (1999) 159–169. [37] C.S. Hirsch, R.O. Bost, S.R. Gerber, M.E. Cowan, L. Adelson, I. Sunshine, Carboxyhemoglobin concentrations in flash fire victims: report of six simultaneous fire fatalities without elevated carboxyhemoglobin, Am. J. Clin. Pathol. 68 (1977) 317–320. [38] F. Zor, M. Deveci, M. Bozkurt, S. Dikkatli, H. Duman, M. Sengezer, Psychological evaluation of self-inflicted burn patients: suicide or parasuicide? Burns 31 (2005) 178–181. [39] D.P. Wirthwein, J.E. Pless, Carboxyhemoglobin levels in a series of automobile fires, Am. J. Forensic Med. Pathol. 17 (1996) 117–123.
©2014 Elsevier
