- Email: [email protected]
Spinal hyperostosis as an important sign indicating spine injuries on postmortem computed tomography
Legal Medicine xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Legal Medicine journal homepage: www.elsevier.com/locate/legalmed
Spinal hyperostosis as an important sign indicating spine injuries on postmortem computed tomography Toru Oshima a,⇑, Mitsumasa Hayashida b, Maki Ohtani a, Manabu Hashimoto c, Satoshi Takahashi c, Koichi Ishiyama c, Takahiro Otani c, Makoto Koga c, Makoto Sugawara c, Sohtaro Mimasaka a a b c
Department of Forensic Sciences, Akita University Graduate School of Medicine, Akita 010-8543, Japan Department of Orthopaedic Surgery, Graduate School of Medical Sciences, Kyushu University, Fukuoka 812-8582, Japan Department of Radiology, Akita University Graduate School of Medicine, Akita 010-0825, Japan
a r t i c l e
i n f o
Article history: Received 16 March 2014 Received in revised form 19 March 2014 Accepted 22 March 2014 Available online xxxx Keywords: Ankylosing spondylitis Autopsy Forensic medicine Multiplanar reformat reconstruction Postmortem computed tomography Postmortem examination Spinal cord injury Spinal hyperostosis
a b s t r a c t Although spine injuries are not always detectable on postmortem computed tomography (PMCT), spinal hyperostosis, an important risk factor for spine injury, is relatively easily detectable on PMCT. We therefore examined the utility of the detection of spinal hyperostosis on PMCT as an indicator of spine injury. Full-body PMCT images of 88 autopsy cases with a bruise on the face or forehead but no identifiable skull fracture were reviewed prior to autopsy for the identification and classification of spinal hyperostosis. Spine injuries were observed in 56.0% of cases with spinal hyperostosis and 1.6% of cases without spinal hyperostosis. Among the cases with spinal hyperostosis, spine injuries were observed in 66.7% of cases at stage 2 or 3 and in 88.9% of cases at stage 3. Spine injuries were diagnosed on PMCT in 33.3% of cases prior to autopsy. A significant association was found between spinal hyperostosis and presence of spine injury that cannot be detected on PMCT, indicating that the identification of spinal hyperostosis on PMCT may assist in detecting spine injuries. This finding suggests that investigation of the presence of spine injury based on the identification of spinal hyperostosis on PMCT may assist in determining the correct cause of death by autopsy. Ó 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction In recent years, diagnostic imaging techniques, including computed tomography (CT) and magnetic resonance imaging (MRI), have become commonly used tools in the field of forensic medicine in many industrialized nations [1–4]. The primary objective of postmortem imaging diagnoses is to supplement data obtained from autopsy and other forms of forensic examination to more accurately establish the cause of death, improve the precision of individual identification, and identify victims after catastrophic disasters [5]. However, there remains considerable controversy regarding whether postmortem imaging diagnoses should be used for determining cause of death from external sources, especially ⇑ Corresponding author. Address: Department of Forensic Sciences, Akita University Graduate School of Medicine, 1-1-1 Hondo, Akita, Akita 010-8543, Japan. Tel.: +81 18 884 6094; fax: +81 18 836 2610. E-mail addresses: [email protected] (T. Oshima), mitsu-38@ortho. med.kyushu-u.ac.jp (M. Hayashida), [email protected] (M. Ohtani), [email protected] (M. Hashimoto), [email protected] (S. Takahashi), [email protected] (K. Ishiyama), [email protected]. akita-u.ac.jp (T. Otani), [email protected] (M. Koga), masugawa@ doc.med.akita-u.ac.jp (M. Sugawara), [email protected] (S. Mimasaka).
trauma [6]. Postmortem CT (PMCT) is not always able to detect laceration of the intervertebral disk or incomplete fracture of the vertebral bone without dislocation [7]. When PMCT is used for screening cases of trauma, caution must be observed regarding this imaging modality’s limitations, which include inability to detect laceration of the intervertebral disk or incomplete fracture of the vertebral bone without dislocation [7]. Individuals with spinal hyperostosis, including ankylosing spondylitis and diffuse idiopathic skeletal hyperostosis, are more prone to fracture than those with normal spines [8]. This condition has been described under several names, making accurate description and categorization of patients with different forms of spinal hyperostosis difficult [9–11]. Here, we refer to ‘‘spinal hyperostosis’’ as a purely descriptive term, without implying its use as a specific nosological status. As the force causing fracture of the spine in patients with hyperostosis is often minor, emergency physicians treating these patients are sometimes unaware that the spine has incurred sufficient trauma to have sustained fracture [8,12]. Research in the field of forensic medicine has revealed that minor trauma can incur fracture of the cervical spine in patients with ankylosing spondylitis [13]. In accordance with this finding, we have observed fracture of the cervical spine in several patients with ankylosing spondylitis resulting from minor trauma that
http://dx.doi.org/10.1016/j.legalmed.2014.03.006 1344-6223/Ó 2014 Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: Oshima T et al. Spinal hyperostosis as an important sign indicating spine injuries on postmortem computed tomography. Leg Med (2014), http://dx.doi.org/10.1016/j.legalmed.2014.03.006
2
T. Oshima et al. / Legal Medicine xxx (2014) xxx–xxx
could not be detected on PMCT. As a means of overcoming this diagnostic challenge, we evaluate the utility of detection of spinal hyperostosis on PMCT as an indicator of spine injuries. 2. Material and methods 2.1. Subject selection The mechanism of injury in blunt cervical spine injury involves 2 forces: (1) an acceleration-deceleration force or change in velocity causing significant head and neck movement and resulting in a flexion–extension injury pattern and (2) a direct force to the head or face against an immovable object that is transmitted down the cervical spine [14]. It is often difficult to identify the sustaining of an indirect force that had resulted in a flexion–extension injury pattern by autopsy. Thus, we focused on identifying the sustaining of a direct force to the head or face in our investigation of autopsy cases in this study. Among all autopsy cases that did not involve decomposed or burned bodies, we examined those that met the inclusion criteria of (1) a bruise on the face or frontal region of head and (2) age over 50 years, the age range of most individuals with ankylosing spondylitis. As the aim was identification of trauma inflicted by minor force as described in a previous report [8], the exclusion criterion was severe head trauma with skull fracture, which would likely have been inflicted by major force. The 88 cases that met the inclusion criteria consisted of 47 men and 41 women ranging from 50 to 95 years of age, with an average age of 71.9 years. Complete autopsies were performed in all cases and spine injuries were diagnosed by autopsy. The data were gathered during all forensic autopsy examinations in a manner that conformed to the privacy policy on forensic autopsy established by the Japanese Society of Legal Medicine [15].
complete three-dimensional (3D) reconstruction of the vertebral columns and to obtain MPR and 3D reconstruction images. At the first stage, the anterior surface of the vertebral bodies shows a very slight laminated thickening, with the onset of spinal hyperostosis in front of the disks characterized by a well-defined, often triangular nucleus. At the second stage, the pre-vertebral thickening becomes more accentuated and tends to be lengthened by the development of a thick spur generally assuming the shape of a sharp point resembling the flame of a candle. At the third stage, the fusion of the pre-discal nucleus is completed. For MPR and 3D reconstruction images, the head, neck, and upper thorax CT images were reconstructed at a 1.25 mm interval. 2.5. Statistical analysis The association between spine injury and spinal hyperostosis was assessed using the Fisher’s exact test. The association between the severity of spinal hyperostosis and spine injury was assessed using Cochran–Armitage trend analysis. A value of p 6 0.05 was considered an indication of statistical significance. 3. Results 3.1. Relationship between spinal hyperostosis and spine injury
It is routine practice at our institute to perform full-body PMCT prior to autopsy. All CT scans were performed using an ECLOS 16 slice CT scanner (Hitachi Medical Systems, Tokyo, Japan). The scan parameters used were as follows: (1) head and neck: 120 kV, 250 mAs, a helical pitch of 0.94, and a 0.63 mm slice thickness with a 2.5 mm reconstruction interval; (2) thorax: 120 kV, 180 mAs, a helical pitch of 1.06, and a 1.25 mm slice thickness with a 2.5 mm reconstruction interval; (3) abdomen: 120 kV, 180 mAs, a helical pitch of 1.06, and a 1.25 mm slice thickness with a 2.5 mm reconstruction interval; and (4) lower limbs: 120 kV, 140 mAs, a helical pitch of 1.31, and a 1.25 mm slice thickness with a 2.5 mm reconstruction interval. The PMCT parameters used are those used in standard practice by our institute’s non-forensic radiologists.
After initial review of the pre-autopsy PMCT images, the subjects were divided into the spinal hyperostosis (n = 25) and control (n = 63) groups based on identification of spinal hyperostosis at the C1-Th4 vertebral levels. In the spinal hyperostosis group, spine injuries were identified in 14 of the 25 (56.0%) cases, all of which (100%) had sustained trauma at the level of spinal hyperostosis or the next vertebral level (Table 1). In contrast, spine injuries were identified in only 1 of the 63 (1.6%) subjects in the control group. Review of the autopsy images revealed the difference in the extent of spine injury sustained by each group. Review of Fig. 1a–c, which shows the autopsy photos and head-and-neck PMCT scans of a 79-year-old woman found dead in a canal, revealed that the subject had spinal hyperostosis and had sustained lacerations of the intervertebral disk at C3-4 and C6-7. In contrast, review of Fig. 1d–f, which shows the autopsy photos and head-and-neck PMCT scans of a 67-year-old woman also found dead in a canal, revealed that the subject had not had spinal hyperostosis and had not sustained cervical fracture or laceration of the intervertebral disk despite sustaining trauma resulting in bleeding of the cervical muscles. Analysis of the association between spine injury and spinal hyperostosis revealed a statistically significant association between the 2 variables (Fisher’s exact test P < 0.0001) with a sensitivity of 0.95 and specificity of 0.93.
2.3. Radiological review
3.2. Relationship between spinal hyperostosis severity and spine injury
Prior to autopsy, the results of each subject’s full-body PMCT scans were reviewed by non-forensic radiologists. To aid in blinding these readers to outside influences, the radiologists were not provided with data regarding the subject and only minimal data regarding the death scene. Autopsy data were not communicated to the radiologist until after all of a subject’s CT scans had been reviewed.
The severity of each case of spinal hyperostosis was classified into one of the 3 stages described previously. The incidence of spine injury was 0% (0 of 4) at stage 1, 50.0% (6 of 12) at stage 2, and 88.9% (8 of 9) at stage 3 (Fig. 2). Statistical analysis revealed a significant positive relationship between stage of spinal hyperostosis and risk of spine injury, with the higher stage the higher the risk of spine injury (Cochran-Armitage trend analysis P = 0.0014; Table 2).
2.2. Image acquisition
2.4. Classification of spinal hyperostosis 3.3. Detection of spine injury by PMCT According to Forestier and Lagier [10], the severity of spinal hyperostosis can be classified into 3 stages based on a review of classical X-ray scans. This classification methodology was applied to perform multiplanar reformat reconstruction (MPR) and
Review of the PMCT images by a non-forensic radiologist prior to autopsy had resulted in identification of only 5 spine injuries in the 15 cases (33.3%) later diagnosed with spine injury. All cases
Please cite this article in press as: Oshima T et al. Spinal hyperostosis as an important sign indicating spine injuries on postmortem computed tomography. Leg Med (2014), http://dx.doi.org/10.1016/j.legalmed.2014.03.006
3
T. Oshima et al. / Legal Medicine xxx (2014) xxx–xxx Table 1 Description of the 15 spine injury cases assessed in the present study. No.
Age
Sex
Level of spine injuries (C1-Th4)
Level of hyperostosis (C1-Th4)
Severity of hyperostosis
Obvious dislocation
Cause of injuries
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
61 84 55 77 69 83 79 74 81 85 67 65 62 60 52
Male Male Male Female Male Male Female Male Male Male Female Male Male Male Male
C3-5 C3-4 C4 Th4 C5-6 C5,Th1 C3-4,C6-7 C3-5,C7-Th1 C4-5, C7-Th1 C3-6 C5-7 C6-7 C3-4,C6-7 C6-7 C3-6
C5-7, Th1-4 C3-Th4 C3-Th4 Th3-4 C6-7, Th3-4 C3-4, Th2-4 C3-Th4 C2-7,Th2-4 C4-5,Th2-3 C4-5 Th1-2 C6-7, Th3-4 C4-6 C4-6 None
3 3 3 3 3 3 3 3 3 2 2 2 2 2 –
+ +
Motor vehicle collision Bicycle accident Run over by vehicle Motor vehicle collision Fall Fall Fall Bicycle accident Bicycle accident Fall Fall Fall Fall Fall Fall
+ +
+
Fig. 1. Upper figures (a–c) show the autopsy photos and head-and-neck postmortem computed tomography (PMCT) images of a 79-year-old woman with spinal hyperostosis found dead in a canal (a) Lacerations of the intervertebral disk at C3-4 and C6-7 (black arrows) with spinal hyperostosis (b) Three-dimensional (3D) reconstruction from PMCT (c) MPR reconstruction from PMCT Lower figures (d–f) show the autopsy photos and head-and-neck PMCT images of a 67-year-old woman without spinal hyperostosis found dead in a canal (d) Lack of cervical fracture or laceration of the intervertebral disk despite the sustaining of trauma resulting in bleeding of the cervical muscles (white arrows). (e) 3D reconstruction from PMCT (f) Multiplanar reformat reconstruction from PMCT.
diagnosed by review of PMCT images were characterized by obvious dislocation (Table 1).
4. Discussion The spine injuries identified in the subjects examined in this study were often caused by common accidents such as a fall or
bicycle accident. Most subjects who had sustained spine injuries were found to have had moderate or severe spinal hyperostosis, whereas most subjects who had not sustained cervical fracture were found not to have had spinal hyperostosis. In addition, all subjects who had sustained spine injury and had spinal hyperostosis had been injured at the level of spinal hyperostosis or the next vertebral level. These results suggest that the spinal ankylosis caused by spinal hyperostosis often leads to the infliction of spine
Please cite this article in press as: Oshima T et al. Spinal hyperostosis as an important sign indicating spine injuries on postmortem computed tomography. Leg Med (2014), http://dx.doi.org/10.1016/j.legalmed.2014.03.006
4
T. Oshima et al. / Legal Medicine xxx (2014) xxx–xxx
stage1
stage2
stage3
spine injury(+)
0
6
8
spine injury(-)
4
6
1
total
4
12
9
Cochran- Armitage trend analysis P=0.0014 Fig. 2. Classification of severity of spinal hyperostosis and number of spine injuries in the subjects assessed in the present study.
References Table 2 Association between spinal hyperostosis and spine injury.
Spine injury(+) Spine injury( ) Total
Spinal hyperostosis(+)
Spinal hyperostosis( )
Total
14 11 25
1 62 63
15 73 88
Fisher’s exact test P < 0.0001.
injury by slight trauma, as had been concluded in previous reports [8,16]. Previous reports have also reported that review of PMCT images does not always allow for detection of laceration of the intervertebral disk or incomplete fracture of the vertebral bone without dislocation [7]. This research demonstrates that many cases of spine injury are not accompanied by dislocation, leading to difficulty in their detection on PMCT. In accordance with this finding, only 33.3% of spine injuries could be accurately diagnosed on PMCT in the present study. In contrast, spinal hyperostosis is known to be relatively easy to detect by radiological examination [10]. The results of the statistical analysis indicate that spine injury is highly associated with spinal hyperostosis. Of the 21 subjects classified at an advanced stage (2 or 3) of spinal hyperostosis, spine injuries were detected in 14 (66.7%) on PMCT. Among those at stage 3, spine injuries were detected in 8 of 9 (88.9%). These results suggest that diagnosis of high-stage spinal hyperostosis is an effective screening tool in the detection of spine injuries. Despite their utility in helping detect spine injury, we understand that classification and review of PMCT images cannot be used for diagnosis of spine injuries, which can be confirmed by autopsy only. Nevertheless, our study’s findings indicate that identification of spinal hyperostosis on PMCT may assist in determining cause of death, both now and in the future.
[1] O’Donnell C, Woodford N. Post-mortem radiology – a new sub-speciality? Clin Radiol 2008;63:1189–94. [2] Rutty GN, Morgan B, O’Donnell C, Leth PM, Thali M. Forensic institutes across the world place CT or MRI scanners or both into their mortuaries. J Trauma 2008;65:493–4. [3] Thali MJ, Yen K, Schweitzer W, Vock P, Boesch C, Ozdoba C, et al. Virtopsy, a new imaging horizon in forensic pathology: virtual autopsy by postmortem multislice computed tomography (MSCT) and magnetic resonance imaging (MRI) – a feasibility study. J Forensic Sci 2003;48:386–403. [4] Poulsen K, Simonsen J. Computed tomography as routine in connection with medico-legal autopsies. Forensic Sci Int 2007;171:190–7. [5] Iwase H, Yajima D, Hayakawa M, Yamamoto S, Motani H, Sakuma A, et al. Evaluation of computed tomography as a screening test for death inquest. J Forensic Sci 2010;55:1509–15. [6] Hoey BA, Cipolla J, Grossman MD, McQuay N, Shukla PR, Stawicki SP, et al. Postmortem computed tomography, ‘‘CATopsy’’, predicts cause of death in trauma patients. J Trauma 2007;63:979–85. [7] Iwase H, Yamamoto S, Yajima D, Hayakawa M, Kobayashi K, Otsuka K, et al. Can cervical spine injury be correctly diagnosed by postmortem computed tomography? Leg Med (Tokyo, Japan) 2009;11:168–74. [8] Hunter T, Forster B, Dvorak M. Ankylosed spines are prone to fracture. Can Fam Physician 1995;41:1213–6. [9] Houk RW, Hendrix RW, Lee C, Lal S, Schmid FR. Cervical fracture and paraplegia complicating diffuse idiopathic skeletal hyperostosis. Arthritis Rheum 1984;27:472–5. [10] Forestier J, Lagier R. Ankylosing hyperostosis of the spine. Clin Orthop Relat Res 1971;74:65–83. [11] Hunter T, Dubo H. Spinal fractures complicating ankylosing spondylitis. Ann Intern Med 1978;88:546–9. [12] Caron T, Bransford R, Nguyen Q, Agel J, Chapman J, Bellabarba C. Spine fractures in patients with ankylosing spinal disorders. Spine 2010;35:E458–64. [13] Ueno Y, Nishimura A, Ogawa Y, Nakagawa K, Yanagida Y, Mizoi Y, et al. [Fracture of the cervical spine caused by blow in patient with ankylosing spondylitis – a report of an autopsy case]. Nihon Hoigaku Zasshi 1992;46:321–6. [14] Kulvatunyou N, Friese RS, Joseph B, O’Keeffe T, Wynne JL, Tang AL, et al. Incidence and pattern of cervical spine injury in blunt assault: it is not how they are hit, but how they fall. J Trauma Acute Care Surg 2012;72:271–5. [15] Yoshioka N, Tuji T, Kajii E, Suzuki K, Tanaka N, Kato N, et al. The privacy policy of forensic autopsy in Japan. Nihon Hoigaku Zasshi 2002;56:319–22. [16] Corke CF. Spinal fracture and paraplegia after minimal trauma in a patient with ankylosing vertebral hyperostosis. BMJ 1981;282:2035.
Please cite this article in press as: Oshima T et al. Spinal hyperostosis as an important sign indicating spine injuries on postmortem computed tomography. Leg Med (2014), http://dx.doi.org/10.1016/j.legalmed.2014.03.006
Contents lists available at ScienceDirect
Legal Medicine journal homepage: www.elsevier.com/locate/legalmed
Spinal hyperostosis as an important sign indicating spine injuries on postmortem computed tomography Toru Oshima a,⇑, Mitsumasa Hayashida b, Maki Ohtani a, Manabu Hashimoto c, Satoshi Takahashi c, Koichi Ishiyama c, Takahiro Otani c, Makoto Koga c, Makoto Sugawara c, Sohtaro Mimasaka a a b c
Department of Forensic Sciences, Akita University Graduate School of Medicine, Akita 010-8543, Japan Department of Orthopaedic Surgery, Graduate School of Medical Sciences, Kyushu University, Fukuoka 812-8582, Japan Department of Radiology, Akita University Graduate School of Medicine, Akita 010-0825, Japan
a r t i c l e
i n f o
Article history: Received 16 March 2014 Received in revised form 19 March 2014 Accepted 22 March 2014 Available online xxxx Keywords: Ankylosing spondylitis Autopsy Forensic medicine Multiplanar reformat reconstruction Postmortem computed tomography Postmortem examination Spinal cord injury Spinal hyperostosis
a b s t r a c t Although spine injuries are not always detectable on postmortem computed tomography (PMCT), spinal hyperostosis, an important risk factor for spine injury, is relatively easily detectable on PMCT. We therefore examined the utility of the detection of spinal hyperostosis on PMCT as an indicator of spine injury. Full-body PMCT images of 88 autopsy cases with a bruise on the face or forehead but no identifiable skull fracture were reviewed prior to autopsy for the identification and classification of spinal hyperostosis. Spine injuries were observed in 56.0% of cases with spinal hyperostosis and 1.6% of cases without spinal hyperostosis. Among the cases with spinal hyperostosis, spine injuries were observed in 66.7% of cases at stage 2 or 3 and in 88.9% of cases at stage 3. Spine injuries were diagnosed on PMCT in 33.3% of cases prior to autopsy. A significant association was found between spinal hyperostosis and presence of spine injury that cannot be detected on PMCT, indicating that the identification of spinal hyperostosis on PMCT may assist in detecting spine injuries. This finding suggests that investigation of the presence of spine injury based on the identification of spinal hyperostosis on PMCT may assist in determining the correct cause of death by autopsy. Ó 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction In recent years, diagnostic imaging techniques, including computed tomography (CT) and magnetic resonance imaging (MRI), have become commonly used tools in the field of forensic medicine in many industrialized nations [1–4]. The primary objective of postmortem imaging diagnoses is to supplement data obtained from autopsy and other forms of forensic examination to more accurately establish the cause of death, improve the precision of individual identification, and identify victims after catastrophic disasters [5]. However, there remains considerable controversy regarding whether postmortem imaging diagnoses should be used for determining cause of death from external sources, especially ⇑ Corresponding author. Address: Department of Forensic Sciences, Akita University Graduate School of Medicine, 1-1-1 Hondo, Akita, Akita 010-8543, Japan. Tel.: +81 18 884 6094; fax: +81 18 836 2610. E-mail addresses: [email protected] (T. Oshima), mitsu-38@ortho. med.kyushu-u.ac.jp (M. Hayashida), [email protected] (M. Ohtani), [email protected] (M. Hashimoto), [email protected] (S. Takahashi), [email protected] (K. Ishiyama), [email protected]. akita-u.ac.jp (T. Otani), [email protected] (M. Koga), masugawa@ doc.med.akita-u.ac.jp (M. Sugawara), [email protected] (S. Mimasaka).
trauma [6]. Postmortem CT (PMCT) is not always able to detect laceration of the intervertebral disk or incomplete fracture of the vertebral bone without dislocation [7]. When PMCT is used for screening cases of trauma, caution must be observed regarding this imaging modality’s limitations, which include inability to detect laceration of the intervertebral disk or incomplete fracture of the vertebral bone without dislocation [7]. Individuals with spinal hyperostosis, including ankylosing spondylitis and diffuse idiopathic skeletal hyperostosis, are more prone to fracture than those with normal spines [8]. This condition has been described under several names, making accurate description and categorization of patients with different forms of spinal hyperostosis difficult [9–11]. Here, we refer to ‘‘spinal hyperostosis’’ as a purely descriptive term, without implying its use as a specific nosological status. As the force causing fracture of the spine in patients with hyperostosis is often minor, emergency physicians treating these patients are sometimes unaware that the spine has incurred sufficient trauma to have sustained fracture [8,12]. Research in the field of forensic medicine has revealed that minor trauma can incur fracture of the cervical spine in patients with ankylosing spondylitis [13]. In accordance with this finding, we have observed fracture of the cervical spine in several patients with ankylosing spondylitis resulting from minor trauma that
http://dx.doi.org/10.1016/j.legalmed.2014.03.006 1344-6223/Ó 2014 Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: Oshima T et al. Spinal hyperostosis as an important sign indicating spine injuries on postmortem computed tomography. Leg Med (2014), http://dx.doi.org/10.1016/j.legalmed.2014.03.006
2
T. Oshima et al. / Legal Medicine xxx (2014) xxx–xxx
could not be detected on PMCT. As a means of overcoming this diagnostic challenge, we evaluate the utility of detection of spinal hyperostosis on PMCT as an indicator of spine injuries. 2. Material and methods 2.1. Subject selection The mechanism of injury in blunt cervical spine injury involves 2 forces: (1) an acceleration-deceleration force or change in velocity causing significant head and neck movement and resulting in a flexion–extension injury pattern and (2) a direct force to the head or face against an immovable object that is transmitted down the cervical spine [14]. It is often difficult to identify the sustaining of an indirect force that had resulted in a flexion–extension injury pattern by autopsy. Thus, we focused on identifying the sustaining of a direct force to the head or face in our investigation of autopsy cases in this study. Among all autopsy cases that did not involve decomposed or burned bodies, we examined those that met the inclusion criteria of (1) a bruise on the face or frontal region of head and (2) age over 50 years, the age range of most individuals with ankylosing spondylitis. As the aim was identification of trauma inflicted by minor force as described in a previous report [8], the exclusion criterion was severe head trauma with skull fracture, which would likely have been inflicted by major force. The 88 cases that met the inclusion criteria consisted of 47 men and 41 women ranging from 50 to 95 years of age, with an average age of 71.9 years. Complete autopsies were performed in all cases and spine injuries were diagnosed by autopsy. The data were gathered during all forensic autopsy examinations in a manner that conformed to the privacy policy on forensic autopsy established by the Japanese Society of Legal Medicine [15].
complete three-dimensional (3D) reconstruction of the vertebral columns and to obtain MPR and 3D reconstruction images. At the first stage, the anterior surface of the vertebral bodies shows a very slight laminated thickening, with the onset of spinal hyperostosis in front of the disks characterized by a well-defined, often triangular nucleus. At the second stage, the pre-vertebral thickening becomes more accentuated and tends to be lengthened by the development of a thick spur generally assuming the shape of a sharp point resembling the flame of a candle. At the third stage, the fusion of the pre-discal nucleus is completed. For MPR and 3D reconstruction images, the head, neck, and upper thorax CT images were reconstructed at a 1.25 mm interval. 2.5. Statistical analysis The association between spine injury and spinal hyperostosis was assessed using the Fisher’s exact test. The association between the severity of spinal hyperostosis and spine injury was assessed using Cochran–Armitage trend analysis. A value of p 6 0.05 was considered an indication of statistical significance. 3. Results 3.1. Relationship between spinal hyperostosis and spine injury
It is routine practice at our institute to perform full-body PMCT prior to autopsy. All CT scans were performed using an ECLOS 16 slice CT scanner (Hitachi Medical Systems, Tokyo, Japan). The scan parameters used were as follows: (1) head and neck: 120 kV, 250 mAs, a helical pitch of 0.94, and a 0.63 mm slice thickness with a 2.5 mm reconstruction interval; (2) thorax: 120 kV, 180 mAs, a helical pitch of 1.06, and a 1.25 mm slice thickness with a 2.5 mm reconstruction interval; (3) abdomen: 120 kV, 180 mAs, a helical pitch of 1.06, and a 1.25 mm slice thickness with a 2.5 mm reconstruction interval; and (4) lower limbs: 120 kV, 140 mAs, a helical pitch of 1.31, and a 1.25 mm slice thickness with a 2.5 mm reconstruction interval. The PMCT parameters used are those used in standard practice by our institute’s non-forensic radiologists.
After initial review of the pre-autopsy PMCT images, the subjects were divided into the spinal hyperostosis (n = 25) and control (n = 63) groups based on identification of spinal hyperostosis at the C1-Th4 vertebral levels. In the spinal hyperostosis group, spine injuries were identified in 14 of the 25 (56.0%) cases, all of which (100%) had sustained trauma at the level of spinal hyperostosis or the next vertebral level (Table 1). In contrast, spine injuries were identified in only 1 of the 63 (1.6%) subjects in the control group. Review of the autopsy images revealed the difference in the extent of spine injury sustained by each group. Review of Fig. 1a–c, which shows the autopsy photos and head-and-neck PMCT scans of a 79-year-old woman found dead in a canal, revealed that the subject had spinal hyperostosis and had sustained lacerations of the intervertebral disk at C3-4 and C6-7. In contrast, review of Fig. 1d–f, which shows the autopsy photos and head-and-neck PMCT scans of a 67-year-old woman also found dead in a canal, revealed that the subject had not had spinal hyperostosis and had not sustained cervical fracture or laceration of the intervertebral disk despite sustaining trauma resulting in bleeding of the cervical muscles. Analysis of the association between spine injury and spinal hyperostosis revealed a statistically significant association between the 2 variables (Fisher’s exact test P < 0.0001) with a sensitivity of 0.95 and specificity of 0.93.
2.3. Radiological review
3.2. Relationship between spinal hyperostosis severity and spine injury
Prior to autopsy, the results of each subject’s full-body PMCT scans were reviewed by non-forensic radiologists. To aid in blinding these readers to outside influences, the radiologists were not provided with data regarding the subject and only minimal data regarding the death scene. Autopsy data were not communicated to the radiologist until after all of a subject’s CT scans had been reviewed.
The severity of each case of spinal hyperostosis was classified into one of the 3 stages described previously. The incidence of spine injury was 0% (0 of 4) at stage 1, 50.0% (6 of 12) at stage 2, and 88.9% (8 of 9) at stage 3 (Fig. 2). Statistical analysis revealed a significant positive relationship between stage of spinal hyperostosis and risk of spine injury, with the higher stage the higher the risk of spine injury (Cochran-Armitage trend analysis P = 0.0014; Table 2).
2.2. Image acquisition
2.4. Classification of spinal hyperostosis 3.3. Detection of spine injury by PMCT According to Forestier and Lagier [10], the severity of spinal hyperostosis can be classified into 3 stages based on a review of classical X-ray scans. This classification methodology was applied to perform multiplanar reformat reconstruction (MPR) and
Review of the PMCT images by a non-forensic radiologist prior to autopsy had resulted in identification of only 5 spine injuries in the 15 cases (33.3%) later diagnosed with spine injury. All cases
Please cite this article in press as: Oshima T et al. Spinal hyperostosis as an important sign indicating spine injuries on postmortem computed tomography. Leg Med (2014), http://dx.doi.org/10.1016/j.legalmed.2014.03.006
3
T. Oshima et al. / Legal Medicine xxx (2014) xxx–xxx Table 1 Description of the 15 spine injury cases assessed in the present study. No.
Age
Sex
Level of spine injuries (C1-Th4)
Level of hyperostosis (C1-Th4)
Severity of hyperostosis
Obvious dislocation
Cause of injuries
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
61 84 55 77 69 83 79 74 81 85 67 65 62 60 52
Male Male Male Female Male Male Female Male Male Male Female Male Male Male Male
C3-5 C3-4 C4 Th4 C5-6 C5,Th1 C3-4,C6-7 C3-5,C7-Th1 C4-5, C7-Th1 C3-6 C5-7 C6-7 C3-4,C6-7 C6-7 C3-6
C5-7, Th1-4 C3-Th4 C3-Th4 Th3-4 C6-7, Th3-4 C3-4, Th2-4 C3-Th4 C2-7,Th2-4 C4-5,Th2-3 C4-5 Th1-2 C6-7, Th3-4 C4-6 C4-6 None
3 3 3 3 3 3 3 3 3 2 2 2 2 2 –
+ +
Motor vehicle collision Bicycle accident Run over by vehicle Motor vehicle collision Fall Fall Fall Bicycle accident Bicycle accident Fall Fall Fall Fall Fall Fall
+ +
+
Fig. 1. Upper figures (a–c) show the autopsy photos and head-and-neck postmortem computed tomography (PMCT) images of a 79-year-old woman with spinal hyperostosis found dead in a canal (a) Lacerations of the intervertebral disk at C3-4 and C6-7 (black arrows) with spinal hyperostosis (b) Three-dimensional (3D) reconstruction from PMCT (c) MPR reconstruction from PMCT Lower figures (d–f) show the autopsy photos and head-and-neck PMCT images of a 67-year-old woman without spinal hyperostosis found dead in a canal (d) Lack of cervical fracture or laceration of the intervertebral disk despite the sustaining of trauma resulting in bleeding of the cervical muscles (white arrows). (e) 3D reconstruction from PMCT (f) Multiplanar reformat reconstruction from PMCT.
diagnosed by review of PMCT images were characterized by obvious dislocation (Table 1).
4. Discussion The spine injuries identified in the subjects examined in this study were often caused by common accidents such as a fall or
bicycle accident. Most subjects who had sustained spine injuries were found to have had moderate or severe spinal hyperostosis, whereas most subjects who had not sustained cervical fracture were found not to have had spinal hyperostosis. In addition, all subjects who had sustained spine injury and had spinal hyperostosis had been injured at the level of spinal hyperostosis or the next vertebral level. These results suggest that the spinal ankylosis caused by spinal hyperostosis often leads to the infliction of spine
Please cite this article in press as: Oshima T et al. Spinal hyperostosis as an important sign indicating spine injuries on postmortem computed tomography. Leg Med (2014), http://dx.doi.org/10.1016/j.legalmed.2014.03.006
4
T. Oshima et al. / Legal Medicine xxx (2014) xxx–xxx
stage1
stage2
stage3
spine injury(+)
0
6
8
spine injury(-)
4
6
1
total
4
12
9
Cochran- Armitage trend analysis P=0.0014 Fig. 2. Classification of severity of spinal hyperostosis and number of spine injuries in the subjects assessed in the present study.
References Table 2 Association between spinal hyperostosis and spine injury.
Spine injury(+) Spine injury( ) Total
Spinal hyperostosis(+)
Spinal hyperostosis( )
Total
14 11 25
1 62 63
15 73 88
Fisher’s exact test P < 0.0001.
injury by slight trauma, as had been concluded in previous reports [8,16]. Previous reports have also reported that review of PMCT images does not always allow for detection of laceration of the intervertebral disk or incomplete fracture of the vertebral bone without dislocation [7]. This research demonstrates that many cases of spine injury are not accompanied by dislocation, leading to difficulty in their detection on PMCT. In accordance with this finding, only 33.3% of spine injuries could be accurately diagnosed on PMCT in the present study. In contrast, spinal hyperostosis is known to be relatively easy to detect by radiological examination [10]. The results of the statistical analysis indicate that spine injury is highly associated with spinal hyperostosis. Of the 21 subjects classified at an advanced stage (2 or 3) of spinal hyperostosis, spine injuries were detected in 14 (66.7%) on PMCT. Among those at stage 3, spine injuries were detected in 8 of 9 (88.9%). These results suggest that diagnosis of high-stage spinal hyperostosis is an effective screening tool in the detection of spine injuries. Despite their utility in helping detect spine injury, we understand that classification and review of PMCT images cannot be used for diagnosis of spine injuries, which can be confirmed by autopsy only. Nevertheless, our study’s findings indicate that identification of spinal hyperostosis on PMCT may assist in determining cause of death, both now and in the future.
[1] O’Donnell C, Woodford N. Post-mortem radiology – a new sub-speciality? Clin Radiol 2008;63:1189–94. [2] Rutty GN, Morgan B, O’Donnell C, Leth PM, Thali M. Forensic institutes across the world place CT or MRI scanners or both into their mortuaries. J Trauma 2008;65:493–4. [3] Thali MJ, Yen K, Schweitzer W, Vock P, Boesch C, Ozdoba C, et al. Virtopsy, a new imaging horizon in forensic pathology: virtual autopsy by postmortem multislice computed tomography (MSCT) and magnetic resonance imaging (MRI) – a feasibility study. J Forensic Sci 2003;48:386–403. [4] Poulsen K, Simonsen J. Computed tomography as routine in connection with medico-legal autopsies. Forensic Sci Int 2007;171:190–7. [5] Iwase H, Yajima D, Hayakawa M, Yamamoto S, Motani H, Sakuma A, et al. Evaluation of computed tomography as a screening test for death inquest. J Forensic Sci 2010;55:1509–15. [6] Hoey BA, Cipolla J, Grossman MD, McQuay N, Shukla PR, Stawicki SP, et al. Postmortem computed tomography, ‘‘CATopsy’’, predicts cause of death in trauma patients. J Trauma 2007;63:979–85. [7] Iwase H, Yamamoto S, Yajima D, Hayakawa M, Kobayashi K, Otsuka K, et al. Can cervical spine injury be correctly diagnosed by postmortem computed tomography? Leg Med (Tokyo, Japan) 2009;11:168–74. [8] Hunter T, Forster B, Dvorak M. Ankylosed spines are prone to fracture. Can Fam Physician 1995;41:1213–6. [9] Houk RW, Hendrix RW, Lee C, Lal S, Schmid FR. Cervical fracture and paraplegia complicating diffuse idiopathic skeletal hyperostosis. Arthritis Rheum 1984;27:472–5. [10] Forestier J, Lagier R. Ankylosing hyperostosis of the spine. Clin Orthop Relat Res 1971;74:65–83. [11] Hunter T, Dubo H. Spinal fractures complicating ankylosing spondylitis. Ann Intern Med 1978;88:546–9. [12] Caron T, Bransford R, Nguyen Q, Agel J, Chapman J, Bellabarba C. Spine fractures in patients with ankylosing spinal disorders. Spine 2010;35:E458–64. [13] Ueno Y, Nishimura A, Ogawa Y, Nakagawa K, Yanagida Y, Mizoi Y, et al. [Fracture of the cervical spine caused by blow in patient with ankylosing spondylitis – a report of an autopsy case]. Nihon Hoigaku Zasshi 1992;46:321–6. [14] Kulvatunyou N, Friese RS, Joseph B, O’Keeffe T, Wynne JL, Tang AL, et al. Incidence and pattern of cervical spine injury in blunt assault: it is not how they are hit, but how they fall. J Trauma Acute Care Surg 2012;72:271–5. [15] Yoshioka N, Tuji T, Kajii E, Suzuki K, Tanaka N, Kato N, et al. The privacy policy of forensic autopsy in Japan. Nihon Hoigaku Zasshi 2002;56:319–22. [16] Corke CF. Spinal fracture and paraplegia after minimal trauma in a patient with ankylosing vertebral hyperostosis. BMJ 1981;282:2035.
Please cite this article in press as: Oshima T et al. Spinal hyperostosis as an important sign indicating spine injuries on postmortem computed tomography. Leg Med (2014), http://dx.doi.org/10.1016/j.legalmed.2014.03.006