Forensic Science International

Forensic Science International 207 (2011) 122 126 Contents lists available at ScienceDirect Forensic Science International journal homepage: www.elsevier.com/locate/forsciint Fracture of the neck structures in suicidal hangings: A Retrospective study on contributing variables Renaud Clément a,b, Jean-Pierre Guay c, Anny Sauvageau a,d, * a Laboratoire de Sciences Judiciaires et de Médecine Légale, Montreal, Canada b Laboratoire de Médecine Légale, Faculty of Medicine, Nantes, France c School of Criminology, University of Montréal & Montreal Philippe-Pinel Institute, Canada d Office of the Chief Medical Examiner, Edmonton, Canada ARTICLE INFO ABSTRACT Article history: Received 22 October 2009 Received in revised form 3 September 2010 Accepted 19 September 2010 Available online 16 October 2010 Keywords: Asphyxia Hanging Thyroid cartilage Fracture Introduction: Several factors may play a role in the development of fractures of the neck structures in hanging. It has been repetitively demonstrated that the incidence of fractures increases with age. The role of other variables is less clear, different studies presenting contradictory results on the role of gender, the type of suspension, or the type of ligature. However, most of these studies evaluated these factors independently of the age of the victims. Considering that age is probably the most important factor in the development of neck structure fractures, all other contributing factors should be studied in relation to age. The aim of the present study is to evaluate the role of contributing factors to the development of neck structure fractures, taking age categories into account. Materials and methods: A total of 206 cases were analysed for the presence and localization of thyroid fracture. For each case, the following information was also compiled: the presence and localization of other neck structure fractures, gender and age, height and weight, body mass index () type of suspension (complete or incomplete), type of ligature used (rope, wire, clothes, sheet or lace) and localization of the knot (anterior, right, left or posterior). Results: The incidence of neck structure fractures increased with age (x 2 = 21.85; p <.001) and is significantly higher in male victims (31.4%) compared to female victims (11.8%) (x 2 = 5.41; p =.02). The incidence of fractures varied significantly with the height (t = 2.19; p =.031; D =.33), weight (t = 4.38; p <.001; D =.89) and (t = 3.84; p <.001; D =.60). The incidence of fractures did not vary significantly with the type of suspension (i.e. complete hanging with feet off the ground or incomplete hanging with body parts partially supporting the weight of the body) (x 2 = 3.12; p =.077; Phi =.077) and the type of ligature (i.e. narrow vs wide) (x 2 =.05; p =.828; Phi =.015). However, when taking the age of the victims into account, a different picture was revealed: in individuals aged 40 years or more, victims with complete suspension of the body presented with a significantly higher incidence of fractures (63.2%) compared to victims with incomplete suspension (31.0%) (x 2 = 6.79; p =.009; Phi =.318). Conclusion: Several variables contribute to the development of neck structure fractures in hanging. Age is probably the most important one. Other contributing factors are gender, height, weight, and the type of suspension. ß 2010 Elsevier Ireland Ltd. All rights reserved. 1. Introduction Fractures of the neck structures figure among the classic autopsy findings in suicidal hangings [1 3]. The incidence of such fractures varies from one study to the next, from 0 to 76.8% [4,5].In a recent paper, a meta-analysis of all English studies revealed an incidence of neck structure fractures of 36.6% 24.7 [6]. The * Corresponding author at: Office of the Chief Medical Examiner, 7007, 116 Street, Edmonton, Alberta, Canada T6H 5R8. Tel.: +1 780 427 4987; fax: +1 780 422 1265. E-mail address: anny.sauvageau@gmail.com (A. Sauvageau). majority of these fractures are isolated thyroid cartilage fracture (15.3% 12.1), followed by isolated hyoid bone fracture (11.5% 12.1) and combined thyroid cartilage and hyoid bone fracture (7.5% 6.1). Cricoid fracture is virtually non existent in suicidal hangings. Several factors may play a role in the development of fractures of the neck structures in hanging. It has been repetitively demonstrated that the incidence of fractures increases with age [7 17]. The role of gender is less clear: some authors found a male predominance of fractures [9,15,18] whereas other observed a female predominance [7,13] or no significant difference between genders [11,19]. Similarly, studies on the role of several other 0379-0738/$ see front matter ß 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.forsciint.2010.09.016

R. Clément et al. / Forensic Science International 207 (2011) 122 126 123 factors have shown opposite results: the type of suspension (i.e. complete hanging with feet off the ground or incomplete hanging with body parts partially supporting the weight of the body) [7,9,13,14,20], the type of ligature [7,9,16], the location of the knot or the highest suspension point [9,10,12,13,19] and the suspension time [13,15]. However, most of these studies evaluated these factors independently of the age of the victims. Considering that age is probably the most important factor in the development of neck structure fractures, all other contributing factors should be studied in relation to age. The aim of the present study is to evaluate the role of contributing factors to the development of neck structure fractures, taking age categories into account. 2. Materials and methods The province of Quebec (Canada) operates under a coroner system. A coroner may or may not, in performing his death investigation, order a forensic autopsy. In Quebec, a single centralized forensic laboratory covers the entire population of 7.5. Over an 8.5-year period (2000 mid 2009), all autopsy cases performed at this laboratory were retrospectively reviewed for suicidal hanging deaths. During the study period, six forensic pathologists worked at one time or another in the forensic laboratory. For hanging deaths, they all were performing autopsies with dry-neck dissection. All autopsy cases performed in Quebec during the study period are compiled in an Excel database. This database was manually searched for hanging cases and the complete files of selected cases were then consulted. These files include autopsy reports, autopsy notes, autopsy sketches and photographs, toxicological reports, preliminary information from the coroner s request for autopsy and preliminary police reports. A total of 309 suicidal hanging deaths were found. Of these, 1 case was excluded since it was not a typical hanging but a hanging from height, with dislocation of neck vertebrae (hanging after jumping from a bridge). Additionally, 50 cases were excluded from the analysis because post-mortem changes interfered with the evaluation of (significant decomposition, skeletal and charred bodies). Finally, 52 cases were also excluded because the type of hanging was not specified in the autopsy files, thus making their analysis not applicable to the present study. Overall, a total of 206 cases were analysed for the presence and localization of thyroid fracture. For each case, the following information was also compiled: the presence and localization of other neck structure fractures, gender and age, height and weight, type of suspension (complete or incomplete), type of ligature used (rope, wire, clothes, sheet or lace) and localization of the knot (anterior, right, left or posterior). The types of ligatures were regrouped into two broad categories: narrow and wide. Examples of narrow ligatures include rope, electrical cords and shoe strings, whereas the term wide ligatures encompasses pieces of clothing and bed sheets. The localization of the knot was defined by its relation to the sterno-cleidomastoid muscles: a knot localized anteriorly to the sterno-cleido-mastoid muscle was designated as anterior, a knot localized posteriorly to the sterno-cleidomastoid muscle was described as posterior whereas as knot localized on the sternocleido-mastoid muscles were labeled as lateral. The was subsequently calculated according to the WHO international formula (weight (kg)/height 2 (m 2 )) and adapted for children and adolescents. The reported data was statistically analysed using the SPSS Statistics 16.0 software. The effect sizes were measured using Cohen [21], a measure of the strength of the relationship between two variables in a statistical population. Cohen proposed the following guidelines for interpreting d: d =.20 is small effect size, d =.50 is moderate and d =.80 is large. 3. Results 3.1. Incidence and distribution of fractures of the neck structures Of the 206 suicidal hangings, most cases did not present any fracture of the neck structures (72%). Of the 28% of victims presenting with a fracture of the neck structure, the majority presented with an isolated fracture of the thyroid cartilage. The distribution of fractures is depicted in Table 1. Table 1 Distribution of fractures of the neck structures. Number of cases Percentage Absence of fracture 148 72 Presence of fracture 58 28 Isolated thyroid fracture 45 22 Isolated hyoid fracture 3 1 Combined thyroid and hyoid fracture 10 5 Cricoid fracture 0 0 Table 2 Incidence of fractures in relation to age. Age Absence of fracture Presence of fracture n % n % Less than 40 114 82 25 18 40 and more 34 51 33 49 Pearson Chi-square = 21.851; p =.000. 3.2. Incidence of fracture in relation to age and gender The incidence of neck structure fractures increased with age (x 2 = 21.85; p <.001). Victims of less than 40 years of age presented an incidence of fracture of 18% whereas this incidence increased significantly to 49% in victims of 40 years or more (Table 2). The average age of victims without fractures of the neck structures was 31.7 compared to 42.6 for victims presenting fractures (t (204) = 5.66, p <.001; D =.88) (Table 3). As for gender, the incidence rate of fracture is significantly higher in male victims (31.4%) compared to female ones (11.8%) (x 2 = 5.41; p =.02). Results are shown in Table 4. 3.3. Incidence of fracture in relation to the height, weight and The incidence of fractures varied significantly with the height (t = 2.19; p =.031; D =.33), weight (t = 4.38; p <.001; D =.89) and (t = 3.84; p <.001; D =.60) (Table 3). The average height of hanging victims with fractures of the neck structures was of 1.74 m compared to 1.71 m for victims without fractures. As for the average weight and of victims with fractures, it was of 78.2 kg and 25.6 respectively, compared to 68.6 kg and 23.2 in victims without fractures. Further analyses of covariance with age of the victim as a covariate indicate that differences remained significant. Table 5 shows mean height, weight and in relation to age and the presence of fracture, whereas Table 6 estimates marginal Table 3 Analysis of age, height, weight and in relation to fractures of the neck structures. Variable Absence of fracture mean Presence of fracture mean t p D a Age 31.7 42.6 5.66 <.001.88 1.71 1.74 2.19.031.33 68.6 78.2 4.38 <.001.89 23.2 25.6 3.84 <.001.60 a Cohen s d. Table 4 Incidence of fractures in relation to gender. Mean age % (n) Mean age t p Male victims 35.4 13.2 Absence of fracture 68.6 (118) 32.2 12.2 4.99 <.001 Presence of fracture 31.4 (54) 42.4 12.6 Female victims 31.4 13.8 Absence of fracture 88.2 (30) 29.5 13.5 2.32.032 Presence of fracture 11.8 (4) 45.5 4.7 Pearson Chi-square = 5.408; p =.020.

124 R. Clément et al. / Forensic Science International 207 (2011) 122 126 Table 5 Mean height, weight and in relation to age and the presence of fracture. Age category N Mean Std. deviation Std. error mean Less than 40 years Absence of fracture 114 1.72.0881.0082 Presence of fracture 25 1.76.0813.0163 Absence of fracture 114 68.395 14.3664 1.3455 Presence of fracture 25 79.906 15.1977 3.0395 Absence of fracture 114 22.913 3.9987.3745 Presence of fracture 25 25.369 3.4394.6879 40 years and more Absence of fracture 34 1.687.0713.0122 Presence of fracture 33 1.727.0814.0142 Absence of fracture 34 69.496 12.9523 2.2213 Presence of fracture 33 76.926 13.676 2.3807 Absence of fracture 34 24.2 3.8923.6675 Presence of fracture 33 25.689 4.0655.7077 means and standard error (SE) for subjects with and without fractures. 3.4. Incidence of fracture in relation to the type of suspension and the type of ligature The incidence of fractures did not vary significantly with the type of suspension (i.e. complete hanging with feet off the ground or incomplete hanging with body parts partially supporting the weight of the body) (x 2 = 3.12; p =.077; Phi =.077) and the type of ligature (i.e. narrow vs wide) (x 2 =.05; p =.828; Phi =.015). Results are detailed in Table 7. However, when taking the age of the victims into account, a different picture was revealed (Table 8): in individuals aged 40 years or more, victims with complete suspension of the body presented with a significantly higher incidence of fractures (63.2%) compared to victims with incomplete suspension (31.0%) (x 2 = 6.79; p =.009; Phi =.318). This difference was not present in individuals of less than 40 years of age (x 2 =.52; p =.471; Phi =.061). As for the type of ligature, no significant difference was found in individuals less than 40 years of age (x 2 =.11; p =.737; Phi =.028) as well as in older victims (x 2 =.01; p =.936; Phi =.010) 3.5. Incidence of fracture in relation to the localization of the knot The incidence of fractures did not vary significantly with the localization of the knot (x 2 = 4.11; p =.250; Phi =.141) (Table 7). The side lateralization of fracture in relation to the position of the knot was then studied (Table 9). No significant relation was found Table 6 Mean height, weight and in relation to the presence of fracture after controlling for age. Estimated marginal mean Absence of fracture 1.71 (.007) Presence of fracture 1.75 (.012) Absence of fracture 68.9 (1.18) Presence of fracture 77.4 (1.95) Absence of fracture 23.4 (.326) Presence of fracture 25.1 (.538) Table 7 Incidence of fractures in relation to the type of hanging, type of ligature and localization of the knot. Presence of x 2 p Phi/Cramer s V between the localization of the knot (back, right, left or anterior) and the lateralization of fracture (x 2 = 4.52; df =4; p =.341). However, when the knot was neither posterior nor anterior, a medium-large effect according to Cohen s criteria [21] was observed (Phi =.394, p =.094), with the fracture tending to be on the opposite side from the knot. 4. Discussion fracture (%) Type of hanging.05.828.015 Complete 28.7 Incomplete 27.3 Type of ligature 3.12.077.077 Narrow 34.0 Wide 22.9 Localization of the knot 4.11.250.141 Back 25.0 Right 31.6 Left 39.4 Front.00 4.1. Incidence of fracture in relation to age In the present study, the incidence of fractures of the neck structures increases with age. The average age of victims without fractures of the neck structures was 31.7 compared to 42.6 for victims presenting with fractures. This result comes as no surprise: it has been demonstrated in several studies that the incidence of fractures increases with age [7 17] because neck structures become calcified and more brittle in middle and later life [3,22 24]. Bony fusion of the greater horn and body of the hyoid bone is rare in an individual under 20 years old and increases with advancing age [22]. The ossification of the thyroid cartilage is also increasing with the aging process, though there seems to be no direct correlation between the degree and frequency of ossification with increasing age [23]. 4.2. General incidence of fracture In the present retrospective study, an incidence of fracture of 28% was found for suicidal hanging. In the literature, the incidence varies widely from one study to the next, from 0 to 76.8% [4 6]. This variation is not surprising considering that the expected Table 8 Incidence of fracture in relation to the type of ligature and the type of ligature. Incidence of fracture (%) x 2 p value Phi Less than 40 years old Type of ligature.11.737.028 Narrow 17.1 Wide 19.3 Type of hanging.52.471.061 Complete 15.3 Incomplete 20.0 40 years and more Type of ligature.01.936.010 Narrow 48.9 Wide 50.0 Type of hanging 6.79.009.318 Complete 63.2 Incomplete 31.0

R. Clément et al. / Forensic Science International 207 (2011) 122 126 125 Table 9 The side lateralization of the fracture in relation to the localization of the knot. Localization of the knot Absence of fracture Presence of fractures Total Right Left No lateralization Posterior 99 11 16 6 132 75.00% 8.30% 12.10% 4.50% 100.00% Right 26 3 6 3 38 68.40% 7.90% 15.80% 7.90% 100.00% Left 20 8 3 2 33 60.60% 24.20% 9.10% 6.10% 100.00% Anterior 3 0 0 0 3 100.00%.00%.00%.00% 100.00% Total 148 22 25 11 206 71.80% 10.70% 12.10% 5.30% 100.00% incidence will vary significantly depending on the proportion of older people in the series. It is therefore strongly recommended to stop describing general incidence of fracture in forensic research since this data is somehow meaningless. The incidence from now on should be presented in relation to age categories. By doing so, it is probable that the incidence found in different papers will present a lesser degree of variation. 4.3. Incidence of fracture in relation to gender Regarding the effect of gender on the incidence of neck structure fractures, several studies have reached contradictive results: some found a male predominance of fractures [9,15,18], whereas others observed a female predominance [7,13] or no significant difference between genders [11,19]. It should be pointed out however that these studies did not take age into account while evaluating the effect of gender on the incidence of fracture. In Morild, the apparent female predominance is undermined when considering that the average age of women in the study was higher than the average age of men (42.7 for women compared to 38.7 for men) [13]. In Paparo and Siegel, the apparent female predominance of fracture is also undermined by the observation that 40% of the male victims were less than 40 years compared to only 30% of female victims [7]. The two papers that failed to find any difference between genders are not detailed enough to allow a retrospective correction for age [11,19]. The remaining three studies, showing a male predominance of fractures, were comparing male and female victims of similar age distribution [9,15,18]. This is in keeping with the male predominance of fracture also demonstrated in the present study. The male predominance of fracture is easily explained by the ossification pattern of males and females: in general, the degree of ossification of the thyroid cartilage is lesser in females than in males [23]. 4.4. Incidence of fracture in relation to height, weight and A previous study by Luke et al. did not find a relationship between the incidence of fracture and height and weight, but age was not taken into account in the comparison [16]. In the present study, the incidence of fractures increased significantly with the height, the weight and the. The hanging victims with fractures are taller and heavier (average height of 1.74 m, weight of 78.2 kg) than the victims without fracture (average height of 1.71 m, weight of 68.6 kg). In fact, the hanging victims with fractures are often overweight (average of 25.6) whereas the victims without fracture are usually of normal weight (average of 23.2). This difference remains significant when covariate with age. As far as we know, this is the first study to demonstrate this effect. Considering that it is the weight of the body that causes traction on the hanging ligature, it makes sense that tall, heavy and overweight individuals are more prone to neck structure fractures. Of course, the height, weight and are not three totally independent variables. The effect sizes of each variable can be compared by comparing the d of Cohen: the effect size of weight is large (D =.89), as was the effect size of age (D =.88), whereas the effect size of is moderate (D =.60) and the effect size of height is small (D =.33). 4.5. Incidence of fracture in relation to the type of suspension Since the weight of the body on the hanging ligature is an important factor in the causation of neck structure fractures, another factor should also play an important role: the type of suspension, either complete hanging with feet off the ground and the totality of the body weight pressing on the ligature, or incomplete hanging, with partial support of the body weight (hanging in a standing, sitting, kneeling or lying down position). The type of hanging was not found to be a contributing variable in the development of neck structure fractures in three previous studies [7,11,14]. However, these studies did not take age into account. In two other studies, fractures were more frequent in atypical complete hanging (an atypical hanging is a hanging with the knot elsewhere than the nape of the neck) [9,13]. These studies were not only uncorrected for age but also lack separate evaluation of the variables typical/atypical and complete/incomplete hangings. In the present study, the incidence of fractures is higher in complete hanging compared to incomplete hanging in victims aged of 40 years or more. This result supports the findings of Samarasekera and Cooke: in their study, the incidence of hanging in victims aged 40 years or more was 75% in complete hangings compared to 58% in incomplete hangings [20]. 4.6. Incidence of fracture in relation to the type of ligature Luke et al. did not find any association between the incidence of fractures and the specific composition, width, and number of wraps of the ligature [16]. Feigin compared the incidence of fractures with narrow and wide ligatures and did not demonstrate a relationship between the incidence of fractures and the width of the ligature [11]. Similarly, the type of ligature was not found to play a role in the incidence of fractures in Paparo and Siegel [7]. In Sharma et al, fractures were more common with tough materials such as rope or electric cord compared to soft material such as saree [9]. In all these studies, the age was not taken into account in evaluating the effect of the type of ligature. In the present study, the type of ligature (narrow vs wide ligatures) was not found to be an important variable in the development of fractures. 5. Conclusion Several variables contribute to the development of neck structure fractures in hanging. Age is probably the most important

126 R. Clément et al. / Forensic Science International 207 (2011) 122 126 one. Other contributing factors are gender, height, weight, and the type of suspension. References [1] W.U. Spitz, Asphyxia, in: W.U. Spitz, D.J. Spitz (Eds.), Spitz and Fisher s Medicolegal Investigation of Death: Guidelines for the Application of Pathology to Crime Investigation, 4th ed., Springfield, IL, Charles C Thomas, 2006, pp. 783 845. [2] V.J. DiMaio, D. DiMaio, Asphyxia, in: V.J. Geberth (Ed.), Forensic Pathology, 2nd ed., CRC Press, Boca Raton, FL, 2001, pp. 229 277. [3] B. Knight, P. Saukko, Fatal Pressure on the Neck in: Knight s Forensic Pathology, 3th ed., Arnold Publishers, London, England, 2004, pp. 368 394. [4] M.A. Elfawal, O.A. Awad, Deaths from Hanging in the Eastern Province of Saudi Arabia, Med. Sci. Law 34 (1994) 307 312. [5] D. Azmak, Asphyxial deaths: a retrospective study and review of the literature, Am. J. Forensic Med. Pathol. 27 (2006) 134 144. [6] A. Godin, C. Kremer, A. Sauvageau, Fracture of the cricoid as a potential pointer to homicide: A 6-year retrospective study of neck structures fractures in hanging victims, Am. J. Forensic Med. Pathol. (2010), doi:10.1097/paf.0- b013e3181d3dc24. [7] G.P. Paparo, H. Siegel, Neck markings and fractures in suicidal hangings, Forensic Sci. Int. 24 (1984) 27 35. [8] I. Uzün, Y. Büyük, K. Gürpinar, Suicidal hanging: fatalities in Istanbul retrospective analysis of 761 autopsy cases, J. Forensic Leg. Med. 14 (2007) 406 409. [9] B.R. Sharma, V.P. Singh, D. Harish, Neck structure injuries in hanging: comparing retrospective and prospective studies, Med. Sci. Law. 45 (2005) 321 330. [10] S. Nikolic, J. Micic, T. Atanasijevic, V. Djokic, D. Djonic, Analysis of neck injuries in hanging, Am. J. Forensic Med. Pathol. 24 (2003) 179 182. [11] G. Feigin, Forensic of neck organ fractures in hanging, Am. J. Forensic Med. Pathol. 20 (1999) 128 130. [12] P. Betz, W. Eisenmenger, Frequency of throat-skeleton fractures in hanging, Am. J. Forensic Med. Pathol. 17 (1996) 191 193. [13] I. Morild, Fractures of the neck structures in suicidal hanging, Med. Sci. Law. 36 (1996) 80 84. [14] R. James, P. Silcocks, Suicidal hanging in Cardiff: a 15-year retrospective study, Forensic Sci. Int. 56 (1992) 167 175. [15] J. Simonsen, Patho-anatomic findings in neck structures in asphyxiation due to hanging: a survey of 80 cases, Forensic Sci. Int. 38 (1988) 83 91. [16] J.L. Luke, D.T. Reay, J.W. Eisele, H.J. Bonnell, Correlation of circumstances with pathological findings in asphyxial deaths by hanging: a prospective study of 61 cases from Seattle, WA, J. Forensic Sci. 30 (1985) 1140 1147. [17] C.J. Polson, D.J. Gee, B. Knight, Fatal pressure on the neck, in: C.J. Polson (Ed.), The Essentials of Forensic Medicine, 4th ed., Pergamon Press, Oxford, England, 1985, pp. 368 394. [18] H. Green, R.A. James, J.D. Gilbert, R.W. Byard, Fractures of the hyoid bone and laryngeal cartilages in suicidal hanging, J. Clin. Forensic Med. 7 (2000) 123 126. [19] J.M. Suarez-Penaranda, T. Alvarez, X. Miguens, M.S. Rodriguez-Calvo, B. Lopez de Abajo, M. Cortesao, C. Cordeiro, D.N. Vieira, J.I. Munoz, Characterization of lesions in hanging deaths, J. Forensic Sci. 53 (2008) 720 723. [20] A. Samarasekera, C. Cooke, The pathology of hanging deaths in Western Australia, Pathology 28 (1996) 334 338. [21] J. Cohen, Statistical Power Analysis for the Behavioral Sciences, 2nd ed., Erlbaum, Hillsdale, NJ, 1988. [22] D.H. Ubelaker, Hyoid fracture and strangulation, J. Forensic Sci. 37 (1992) 1216 1222. [23] H.M. Garvin, Ossification of laryngeal structures as indicators of age, J. Forensic Sci. 53 (2008) 1023 1027. [24] M. Cerny, Our experience with estimation of an individual s age from skeletal remains of the degree of thyroid cartilage ossification, Acta Univ. Palacki. Olomuc. 3 (1983) 121 144.