CASE REPORT
Craniofacial Identification by Comparison of Antemortem and Postmortem Radiographs Two Case Reports Dealing With Burnt Bodies Carlo P. Campobasso, PhD, MD,* Alessandro S. Dell’Erba, MD, PhD,† Maura Belviso, MD, PhD,‡ and Giancarlo Di Vella, MD, PhD†
Abstract: Two cases of positive identification of burnt bodies by radiographic comparison are reported. They emphasize that antemortem radiographs of the head are an important but sometimes overlooked source of information which can frequently provide useful objective data for comparison purposes. A positive identification can easily be achieved by medical examiners through visual comparison of the antemortem with the postmortem cranial and facial structures, even of bodies severely damaged by fire. In these bodies the radiographs of the skull can graphically depict structures which are often unique to the individual, such as the frontal sinus pattern and the morphology of dental restorations. However, the process of identification through radiographs is appropriate only in burnt bodies in which antemortem radiographs of the alleged deceased are available for comparison and unique craniofacial structures are still present on the body in a wellpreserved state or at least not completely destroyed by fire, depending on the extent of the burn injury. Matching of corresponding features seems preferable to other methods of personal identification such as skull-photo superimposition, morphometric analysis, and/or other computer-aided methods since these techniques need trained personnel, as well as expensive equipment which is not invariably available in the medical examiner’s office or department of anthropology. Key Words: craniofacial identification, frontal sinuses, dental restorations, burnt injuries (Am J Forensic Med Pathol 2007;28: 182–186)
P
ersonal identification of human remains is usually achieved by matching specific features detected on the cadaver with data recorded during the life of the individual. Manuscript received December 12, 2005; accepted May 17, 2006. From the *Department of Health Sciences, University of Molise, Campobasso, Italy; †the Section of Legal Medicine, University of Bari, Policlinico, Bari, Italy; ‡Institute of Legal Medicine, University Cattolica del S. Cuore of Rome, Policlinico A. Gemelli, Rome, Italy. Reprints: Carlo Pietro Campobasso, PhD, MD, Department of Health Sciences (S.pe.S), University of Molise, Via De Sanctis, Snc., 86100 Campobasso, Italy. E-mail:
[email protected] or
[email protected]. Copyright © 2007 by Lippincott Williams & Wilkins ISSN: 0195-7910/07/2802-0182 DOI: 10.1097/PAF.0b013e31806195cb
182
Since radiography is a common diagnostic tool for a great variety of dental and medical conditions, comparison of antemortem and postmortem radiographs is one of the methods most frequently adopted in personal identification.1,2 This procedure usually concentrates on visible similarities and differences between the 2 images, including the presence or absence, shape, and size of distinctive dental and bone peculiarities.3–5 There are numerous descriptions of cranial and dental radiographic features serving for identification, among which frontal sinuses and radiographically visible restoration materials are the most common. The frontal sinuses in humans, together with even a single dental restoration, may characterize the radiographic silhouette sufficiently to confer uniqueness. However, in burnt bodies the availability of frontal sinuses and/or dental restorations depends essentially on the extent of the burn injury because the skull may be badly fragmented and the teeth missing. Use of a classification model to describe the extent of burn damage in victims of fire, such as the scale proposed by Glassman and Crow6 in 1996, can provide consistent information for professionals engaged in the identification process. Morphometric techniques such as skull-photo superimposition7 or digital imaging methods8 can be valuable aids for identification. Digital pattern tracings can be examined and superimposed by commonly available computer programs such as Adobe Photoshop,9 enabling quantitative assessment. In fact, morphometric applications make it possible to convert the radiographic image (or its parts) into a series of numeric values corresponding to the contour lines of the objects of interest. However, these procedures require trained personnel and computerized devices that are not necessarily available in the legal medicine or anthropology department investigating the case.10 Furthermore, interoperator variability plays a role in any quantitative forensic assessment where different operators have taken the antemortem and postmortem radiographs, and possible errors in measurement can easily occur.11 Such procedures can definitively require more time and resources, which may make them significantly less appealing. In fact, there are still no professional standards established for the application of such techniques even if relevant efforts have been recently made.12
The American Journal of Forensic Medicine and Pathology • Volume 28, Number 2, June 2007
The American Journal of Forensic Medicine and Pathology • Volume 28, Number 2, June 2007
We describe 2 cases of radiologic identification of burnt bodies found inside their vehicles, in which the positive identification was achieved by comparison of antemortem and postmortem radiographs using only morphologic methods13 depicting dental restorations and frontal sinuses features. The visual comparison method of radiographs, even if largely subjective, can be performed quickly and easily by forensic pathologists and medical examiners.
Case 1: Suicide by Fire The burnt body of a 36-year-old white man was recovered in September 2000 in a partially burned-out car. The extent of burn damage was classified as level 2 of the Crow-Glassman Scale (CGS).6 The head was well preserved, with blackened but still present tissue of the scalp and of the face, but unrecognizable for the purposes of identification. Anteroposterior and lateral radiographs of the skull were retrieved from a local hospital by the relatives of the suspected deceased. The antemortem x-rays taken 8 years before death showed some restorations of the teeth, but only a few of them occupied the same position in the postmortem dental chart, due to the loss of some teeth in later years. In particular, the first left upper molar and the second right lower molar showed some radiopaque restorations, but
Craniofacial Identification
both teeth had been extracted antemortem, consistent with the finding of advanced healing of both the alveolar cavities. Only restorations located at the first right upper molar and the third right lower molar showed similar radiopaque images in terms of size, shape, and position in the postmortem radiographs. Although these data fully corresponded, we judged it necessary to proceed with examination of the frontal sinuses, which were undamaged by the fire. Septation and lobulation of the frontal sinuses in the antemortem x-rays showed a scalloped upper border of the sinuses with 3 arches, no supraorbital cells, 1 partial septum, slightly arched with rightward convexity in the left sinus, and slight deviation of the nasal septum. Upper and lateral limits of the frontal sinuses were easily traced onto a tracing paper and used for comparison by superimposing each original radiograph. Comparison of the skull radiographs showed substantial evidence linking the person in the original radiograph to the corpse. The outlines and edges of the frontal sinuses, as well as the nasal septum deviation, were a positive match, as had been the morphology of the dental restorations still present (Fig. 1). The correspondence of all the morphologic features observed in the comparison of the dental elements and frontal sinuses was sufficient to yield a positive identification, and no further investigations were judged necessary.
Case 2: Homicide by Multiple Gunshot Wounds to the Chest
FIGURE 1. Comparison of antemortem (A) and postmortem (B) frontal sinus pattern.
The charred remains of a 20-year-old white man were found in October 2001 in a burned-out car in a rural area of Bari (southern Italy). The extent of burn injury to the body was defined as CGS level 3.6 The skull was largely fragmented and devoid of flesh due to the soft tissue and bone having been burnt away (Fig. 2). The frontal sinuses were not available for comparison, and several of the teeth had burst apart and fallen out during the fire. Only small fragments of the charred jaw provided enough material to undertake an identification. The lower left third molar was relatively well preserved. The clinical crown of the first left lower molar had burst due to the extreme heat of the fire, exposing the roots filled with some restoration
FIGURE 2. Damage to the facial tissues and bones in burned body with CGS level 3: lateral view (A) and frontal view (B). © 2007 Lippincott Williams & Wilkins
183
Campobasso et al
The American Journal of Forensic Medicine and Pathology • Volume 28, Number 2, June 2007
FIGURE 3. Antemortem (A) and postmortem (B) radiographs used for comparison.
material. Postmortem x-rays of the lower left molar roots exhibited radiopaque images of a dental restoration. In particular, the anterior root showed a linear, slightly arched image, indicating treatment of the 2 nutrient canals of the mesial root, while the only remaining canal of the posterior root showed a virtually compact, linear radiologic image. This restoration was an exact match in terms of structural details, shape, and size with those observed in the antemortem radiographs, taken 1 year before death (Fig. 3). Another corresponding element was the partial inclusion of the lower left third molar (impacted tooth), with parallel roots and the apical ends still open consistent with the young age of the deceased. In view of the lack of other elements available for comparison, these dental features were considered sufficient to confirm the identity of the body, without undertaking further morphometric or digital imaging investigations.
DISCUSSION The techniques used to achieve personal identification are determined by the presenting condition of the fire victim’s remains. The extent of damage to the victim is primarily dependent upon the temperature and length of exposure to the fire14: recent studies on cremation at 800°C (1472°F) have demonstrated that after approximately 10 minutes, the skull cap is entirely exposed and free from any soft tissue but not yet calcined, after 20 minutes the skull vault is still present, after 30 minutes the tabula externa has disintegrated into fragments and the facial bones are calcined, and after 50 minutes the facial bones have virtually been destroyed, exposing the skull base.15 In many cases craniofacial identification of burnt bodies can be achieved using only the cranial structures not destroyed by thermal injuries. Adoption of a standardized scale, like the CGS model, to describe the extent of burn injury of human remains can provide useful information for the identification process. In our first case of a body with CGS level 2 burn damage, the identification was made mainly by the forensic pathologist on the basis of matching of the corresponding
184
features, among which the frontal sinuses, although fragile by nature, were well preserved and still available for comparison. Radiographic comparison of the dental elements could confirm only some of the dental treatments shown in the postmortem images due to later extraction of previously restored teeth (the first left upper molar and the second right lower molar) in the intervening years (the x-rays had been taken 8 years before death). However, the correspondences observed at the level of the frontal sinuses, as well as some dental elements, were deemed sufficient for positive identification without needing to carry out further morphometric and/or superimposition techniques. Frontal sinus pattern matching has proven to be a very useful tool for identification purposes16 –19 as it is a reliable method for confirming or rejecting an identification. The reliability of comparisons (or the uniqueness of individual outlines) was quantitatively assessed using elliptic Fourier analysis (EFA).12 The configuration of the frontal sinus is unique for each individual as variations have even been reported among monozygotic and dizygotic twins.20 Yoshino et al21 demonstrated the wide variability of the frontal sinuses by means of analysis of the profile of the upper margin, measurement of the sinus area, ascertainment of the presence or absence of supraorbital cells, and, finally, of the presence or absence of septation. Some identification problems have been reported only with comparison of antemortem and postmortem radiographs in “subadults” (⬍18 years) because development of the frontal sinuses continues until the age of 18 –20 years,22 or in the elderly due to morphologic changes deriving from bone reabsorption and widening of the cavities.23 Quatrehomme et al13 suggested an alternative procedure of comparison involving performance of a postmortem radiograph resembling the available antemortem x-ray as closely as possible, as regards size, exposure, and angle. Although these procedures are certainly justified, comparison of the radiographs only by means of morphologic matching of the frontal sinus features has been also proven,24 and sometimes it can be superior. For example, among 35 cases of © 2007 Lippincott Williams & Wilkins
The American Journal of Forensic Medicine and Pathology • Volume 28, Number 2, June 2007
positive identification using frontal sinus comparison, only 16 cases yielded metric quantitative matches.11 Therefore, even if less objective, based mainly on the knowledge, experience, or ability of the examiner, the visual assessment of radiographs is still a very easy and quick method to apply. In our first case, comparison between antemortem and postmortem radiographs of the frontal sinus pattern showed distinctive analogies, which were further supported by the finding of fully corresponding restorations of the right molars. No further morphometric or digital imaging investigations were necessary. Except in cases of subsequent modification due to trauma and/or dental treatment, the dental picture is undoubtedly a valid basis for identification, as it has been amply documented in the literature, especially in badly decomposed bodies or those that have been severely mutilated by fire.25,26 In fact, the high resistance to heat of dental enamel and dentin, as also of the manufactured dental prosthesis materials, is well known. Teeth have been shown to resist heat up to peaks of 1000 –1200°F (537.6 – 648.7°C); for gold crowns, the melting point ranges from 1600 to 2000°F (870.9 – 1093.2°C); for porcelain structures, resistance is as high as 2000°F (1093.2°C); amalgam fillings can withstand heat up to 1600°F (870.9°C). Orthodontic braces can tolerate temperatures up to 1300 –1800°F (704.3–982.1°C).27 However, charred and desiccated teeth are very fragile and tend to fall apart very easily. Color is often useful as a guide to the relative fragility of dental tissues, blackened remains usually being less fragile than remains that are ashen gray in color. The enamel on the anterior surface of the anterior teeth (those most exposed to the fire) is known to simply crack off when touched. Once a tooth has crumbled into pieces, it is difficult to reconstruct it sufficiently to be able to take radiographs, photographs, or impressions to serve as postmortem evidence.28 In cases of loss of comparative radiographic elements such as missing or defective key dental restorations, it is still possible to recreate a radiopaque image using a recent method based on amalgam powder and calcium hydroxide.29 However, even if the anterior teeth are usually badly damaged by fire, the posterior teeth (commonly restored) tend to remain intact, with some scorching of the dentition as in our second case, in which the only element available for personal identification was the lower left first molar. In the second case, the body had suffered much more severe damage, being classified as CGS level 3, with a largely fragmented skull without any soft tissue, in which most of the facial bones had burnt away. The frontal sinuses were consequently not available for photographic comparison, and most of the anterior teeth had also burst apart and fallen out during the fire. Only some posterior teeth, although partially damaged by the fire, were still present. The roots of the lower left first molar showed enough material for a positive personal identification. In particular, the anterior root showed a linear, slightly arched image showing the outcome of treatment of the 2 nutrient canals of the mesial root, while the only canal of the posterior root showed a virtually linear, compact radiologic image. This restoration showed an exact match of © 2007 Lippincott Williams & Wilkins
Craniofacial Identification
the structural details, shape, and size with those observed in the antemortem radiographs, as well as the partial inclusion of the lower left third molar. The radiopaque images were considered to provide sufficient anatomic features for dental identification of this badly burnt body, in which most of the remains had been completely destroyed by fire.
CONCLUSIONS The positive identification of human remains requires matching of the physical characteristics of the deceased with records of the same physical characteristics taken before death (antemortem), but the degree of reliability of the different identification methods is highly variable. In some cases, several means may be used to produce cumulative evidence increasing the level of reliability. For example, the combined use of the frontal sinus patterns (aplasia, symmetry, left or right asymmetry, number of lobulations) and the nasal septum deviation patterns (straight, left, right sigmoid, reverse sigmoid, and rare types) has been suggested to enhance the efficacy for systematic radiographic identification of unknown human remains.30 Radiology is usually the primary and often the only tool for positive identification, especially of badly decomposed bodies or those extensively damaged by fire. Teeth and bones are commonly durable and can withstand considerable decompositional and destructive forces; any single bone or tooth can potentially be used as a match identifier in radiologic examinations. Dental examination and visual assessment of antemortem and postmortem dental records and radiographs produce results with a high degree of reliability and relative simplicity.31 The uniqueness of the frontal sinus patterns is well known, and comparison of the frontal sinus configuration is easy using frontal view radiographs of the skull, even if there are considerable variations in angle or projection between the 2 radiographs. Even fragments of frontal sinuses, as well of dental restorations, may contain sufficiently distinctive patterns for comparative identification.2 Depending on the availability of features according to the stage of decay and extent of the burn injury, a meticulous morphologic analysis of the dental elements and/or frontal sinuses can thus enable positive identification, obviating the need for other methods, such as craniofacial superimposition,7 morphometric analysis, or digital imaging,9 since these latter are demanding procedures requiring trained personnel and costly equipment, which are not always available at the medicolegal department investigating the case. REFERENCES 1. Fierro MF. Identification of human remains. In: Spitz WU, Fisher RS, eds. Medicolegal Investigation of Death: Guidelines for the Application of Pathology to Crime Investigation. 3rd ed. Springfield, IL: Charles C. Thomas; 1993:71–117. 2. Brogdon BG. Radiological identification of individual remains. In: Brodgon BG, ed. Forensic Radiology. Boca Raton, FL: CRC Press; 1998. 3. Mertz CA. Dental identification. Dent Clin North Am. 1977;21:47– 67. 4. De Vore DT. Radiology and photography in forensic dentistry. Dent Clin North Am. 1977;21:69 – 83.
185
Campobasso et al
The American Journal of Forensic Medicine and Pathology • Volume 28, Number 2, June 2007
5. Jablonski NG, Shum SF. Identification of unknown human remains by comparison of antemortem and postmortem radiographs. Forensic Sci Int. 1989;42:221–230. 6. Glassman DM, Crow RM. Standardization model for describing the extent of burn injury to human remains. J Forensic Sci. 1996;41:152– 154. 7. Pesce Delfino V, Colonna M, Vacca E, et al. Computer-aided skull/face superimposition. Am J Forensic Med Pathol. 1986;7:201–212. 8. Wood RE, Nigel JK, Sweet DJ. Digital dental radiographic identification in the paediatric, mixed and permanent dentitions. J Forensic Sci. 1999;44:901–916. 9. Bowers CM, Johansen RJ. Digital imaging methods as an aid in dental identification of human remains. J Forensic Sci. 2002;47:354 –359. 10. Angyl M, Dercczy K. Personal identification on the basis of antemortem and postmortem radiographs. J Forensic Sci. 1998;43:1089 –1093. 11. Kirk NJ, Wood RE, Goldstein M. Skeletal identification using the frontal sinus region: a retrospective study of 39 cases. J Forensic Sci. 2002;47: 318 –323. 12. Christensen AM. Testing the reliability of frontal sinuses in positive identification. J Forensic Sci. 2005;50:18 –22. 13. Quatrehomme G, Fronty P, Sapanet M, et al. Identification by frontal sinus pattern in forensic anthropology. Forensic Sci Int. 1996;83:147– 153. 14. Spitz WU. Thermal injuries. In: Spitz WU, Fisher RS, eds. Medicolegal Investigation of Death: Guidelines for the Application of Pathology to Crime Investigation. 3rd ed. Springfield, IL: Charles C. Thomas; 1993: 413– 443. 15. Bohnert M, Rost T, Pollak S. The degree of destruction of human bodies in relation to the duration of the fire. Forensic Sci Int. 1998;95:11–21. 16. Culbert WL, Law FM. Identification by comparison of roentgenograms of nasal accessory sinuses and mastoid processes. JAMA. 1927;88: 1634 –1636. 17. Camps FE. Radiology and its forensic application. In: Camps FE, ed. Recent Advances in Forensic Pathology. London: Churchill; 1969:149 – 160. 18. Cheevers LS, Ascencio R. Identification by skull superimposition. Int J Forensic Dent. 1977;13:14 –16.
186
19. Atkins L, Potsaid MS. Roentgenographic identification of human remains. JAMA. 1978;240:2307–2308. 20. Asherson N. Identification by Frontal Sinus Prints: A Forensic Medical Pilot Survey. London: Lewis and Co; 1965. 21. Yoshino M, Miyasaka S, Sato H, et al. Classification system of frontal sinus patterns by radiography: its application to identification of unknown skeletal remains. Forensic Sci Int. 1987;34:289 –299. 22. Marek Z, Kus´miderski J, Lisowski Z. Radiogramme der Stirnho¨hlen als Grundlage fu¨r die Identifizierung von Katastrophenopfern und von unbekannten Skeletten. Arch F Kriminol. 1983;172:1– 6. 23. Ubelaker DH. Positive identification from the radiographic comparison of frontal sinus patterns. In: Rathbun TA, Buikstra JE, eds. Human Identification. Springfield, IL: Charles C. Thomas; 1984:399 – 411. 24. Marlin DC, Clark MA, Standish M. Identification of human remains by comparison of frontal sinus radiographs: a series of four cases. J Forensic Sci. 1991;36:1765–1772. 25. Woolridge EDJr. Forensic odontology. In: Tedeschi CG, Eckert WG, Tedeschi LG, eds. Forensic Medicine. Philadelphia, PA: WB Saunders Company; 1977:1116 26. Silverstin HA. Comparison of antermortem and postmortem findings. In: Bowers CM, Bell GL, eds. Manual of Forensic Odontology. 3rd ed. Ontario, Canada: Manticore, American Society of Forensic Odontology; 1997:31–35. 27. Bonifacio A, Belleli E. L’interesse dell’odontostomatologo ai problemi dell’identificazione personale. Odontostomatol Implantoprotesi. 1985; 22:7–10. 28. Delattre VF, Stimson PG, Burned beyond recognition: odontological identification of the charred remains of motor vehicle accident victims. Proc Am Acad Forensic Sci. 1999;5:152–155. 29. Shiroma CY. The use of amalgam powder and calcium hydroxide to recreate a radiopaque image of a lost dental restoration. J Forensic Sci. 2002;47:609 – 613. 30. Taniguchi M, Sakoda S, Kano T, et al. Possible use of nasal septum and frontal sinus patterns to radiographic identification of unknown human remains. Osaka City Med J. 2003;49:31–38. 31. Fixott RH. The Dental Clinics of North America. Philadelphia, PA: WB Saunders Company; 2001:45.
© 2007 Lippincott Williams & Wilkins
