Post-mortem computed tomography for forensic applications: A systematic review of gunshot deaths

Abstract

The first reported application of computed tomography (CT) to the field of forensic medicine was in 1977 for the study of gunshot injuries to the head. Since then, CT scanning has shown increased uses within the field of forensic practice. In this study a systematic literature search was carried out by consulting 17 electronic scientific databases. This systematic review considers studies published to date concerning the use of post-mortem CT (PMCT) in gunshot injury cases; this review shows variability between the PMCT and autopsy examination findings. Furthermore we identify the technical differences between the selected studies, emphasizing the specific findings obtainable with PMCT, in order to support forensic research towards shared and validated practices, whose results are used as evidence in the courtroom.

Keywords

Forensic science post-mortem computed tomography autopsy injuries death virtopsy scanning procedures gunshot systematic review evidence

Introduction

Radiology has played a role in autopsy practice since the year of the discovery of X-rays, 1895, when it was used to assist the investigation of a gunshot injury to the leg of a Canadian man called Tolson Cunning. The following year, Professor Schuster used radiology for the first time in Manchester, United Kingdom to investigate the homicidal shooting of Elizabeth Hartley.¹ The first reported application of computed tomography (CT) to the field of forensic medicine was in 1977 for the study of gunshot injuries to the head.² Since then there has been a slow but steady rise in publications related to so-called post-mortem CT (PMCT) within the field of forensic practice.³

Today, as a result of an increasing professional acceptance of the field of practice,⁴ along with technical improvements and decreasing costs, the application of both PMCT and post-mortem magnetic resonance (MR) to autopsy practice has steadily increased (Figure 1). There is a growing literature which supports its use not only as an adjunct to autopsy practice, but increasingly as a replacement to the invasive autopsy, depending upon the nature of the case investigated.^5–10 In a recent paper by Garetier et al.,¹¹ PMCT was described a useful adjunct to autopsy especially in suicide cases, to confirm the suicidal modality and exclude other causes of death.

Figure 1.

Flowchart showing the number of published papers, from 1977 to 2018, found on a simple PubMed search with the following keywords: virtopsy, CT-scan, post-mortem, tomography, gunshot.

One area of practice where PMCT has the ability to enhance the invasive autopsy, if not limit the extent of an invasive autopsy, is that of gunshot injuries. Traditionally plain film radiology and then fluoroscopy have been used to identify the presence and location of metal projectiles within bodies. However, the use of PMCT brings a new dimension to the radiological investigation of the case, not only allowing for the accurate three-dimensional location of the projectile but the consideration of soft tissue, organ and boney injuries along the projectile’s path from the entrance to the exit wound, or its final resting position within the body.

Despite the constantly growing reputation of radiological imaging applied to gunshot injury cases in forensic pathology, a systematic review investigating the efficiency and limits of the different radiological techniques, in comparison to autopsy, was still missing. The aim of this study is to provide a comprehensive systematic review considering studies published to date concerning the use of PMCT in gunshot injury cases in order to verify the variability between the PMCT and autopsy examination findings and the occurrence of technical differences between the selected studies, emphasizing the specific and useful findings obtainable with PMCT.

Materials and methods

The sources used for guidance were the Cochrane Handbook¹² and the PRISMA guidelines.¹³

Literature search

A systematic literature search was carried out by consulting 17 electronic scientific databases, including MEDLINE, SCOPUS, The Cochrane Library, Web of Science, and the Directory of Open Access Journals; electronic editorial networks such as BMJ, Blackwell, Elsevier, Karger, Nature Publishing Group, and Springer; and electronic distributors such as OVID Journals and Swetswiss, applying the following primary search strategy: (gunshot OR gunshot wound OR gunshot injur* OR ballistic forensic) AND (radiolog* OR imaging). The latest update of the electronic search was performed on 1 July 2019.

Inclusion criteria

We considered all studies that reported autopsy findings compared with CT examination. No language, publication date, publication type or status restrictions were imposed. We also used the PICOS approach (acronym of Population, Interventions, Comparator, Outcomes, Study Design, PICOS is a structured approach for framing relevant and precise questions, which are often complex and time-consuming, that can be answered in a systematic review) to select studies, according to the objective of this review, to determine the state of the art of post-mortem virtual autopsy, specifically concerning gunshot wounds, by investigating the differences among the scanning tomography techniques.

The number of studies that emerged from the literature search was 380. These were first selected by considering the abstract contents and were subsequently managed according to the study selection process.

Study selection

The study selection process was performed independently but uniformly by two authors (board-certified radiologist and pathologist), using all the above-mentioned inclusion criteria, and it included screening, determining eligibility and including the eligible studies in the systematic review. Eligibility was determined according to the description of CT scan post-mortem parameters compared with autopsy findings and was limited to publications on gunshot deaths (for further information, see Figure 2).

Figure 2.

Flow diagram of study selection process regarding CT scans used in gunshot cases.

We considered papers but not monographic scientific publications. We rejected all studies that were not focused on the subject of this review, such as those concerning other causes of death, different imaging techniques (MR imaging, X-ray, ultrasounds, etc.), and non-forensic applications of CT scanning. Furthermore, we considered only publications written in English.

We also excluded studies that completely or partially lacked readable data, such as articles reporting radiologic results that were not clearly attributable to a CT scan (compared with other radiologic techniques such as MR angio-CT) or results that summarized different causes of death, and all publications lacking autopsy results after CT examination.

No divergent opinions in including or excluding studies were encountered between the reviewers, resulting in 15 studies that were eligible for the systematic review.

Data extraction

This process was performed by two reviewers independently, and disagreements were resolved by discussion. Repetitions of data from serial publications were rejected, and the last update was retained. No authors were contacted for further information. All of the eligible included studies were written in English.

The following information was extracted from each included article:

Characteristics of the publication and the study, including authors, study design, year, journal and number of gunshot death cases (see Table 1);

Table 1.

ID and features of papers included in the review (n=15). nd=not determined.* ID number matches with the number in the references.

ID*	Authors	Year	Journal	Study design	N° of cases
11	Thali MJ, Schweitzer W, Yen K, et al.	2003	Am J Forensic Med Pathol	case report	1
12	Thali MJ, Yen K, Vock P, et al.	2003	Forensic Sci Int	case series	8
13	Oehmichen M, Gehl HB, Meissner C, et al.	2003	Acta Neuropathol	case series	17
14	Levy AD, Abbott RM, Mallak CT, et al.	2006	Radiology	case series	13
15	Harcke HT, Levy AD, Abbott RM, et al.	2007	Am J Forensic Med Pathol	original article	13
16	Harcke HT, Levy AD, Getz JM, Robinson SR.	2008	AJR Am J Roentgenol	case series	nd
17	Andenmatten MA, Thali MJ, Kneubuehl BP, et al.	2008	Leg Med	case series	22
18	Thomsen AH, Jurik AG, Uhrenholt L, Vesterby A	2009	Forensic Sci Int	case series	10
19	Ruder TD, Ross S, Preiss U, Thali MJ	2010	Leg Med	case report	1
20	Makhlouf F, Scolan V, Ferretti G, et al.	2013	Leg Med	case series	47
21	Abdul Rashid SN, Martinez RM, Ampanozi G, et al.	2013	J Forensic Radiol and Imaging	case report	1
22	Hasegawa I, Heinemann A, Tzikas A, et al.	2014	Leg Med	case report	1
23	Maiese A, Gitto L, De Matteis A, et al.	2014	Leg Med	case report	2
24	Kirchhoff SM, Scaparra EF, Grimm J, et al.	2015	Int J Legal Med	original article	51
25	Scaparra E, Peschel O, Kirchoff C, et al.	2016	Eur J Med Res	original article	41

Characteristics of tomographic techniques, such as slice thickness, mAmpere, voltage, pitch, post-processing reconstruction techniques, type and features (see Table 2);

Table 2.

Characteristics of tomographic techniques used in the included papers (n=12).

	DATA				RECONSTRUCTION SOFTWARE
ID	SLICE THICKNESS	mA	KV	PITCH		VR	MIP	SSD	MPR
11	1 mm (head) 2.5 mm (chest)	nd	nd	nd	Advantage Windows 3.1 Workstation	x		x	x
12	1–5 mm	nd	nd	1.5	Advantage Windows 4.0 Workstation			x	x
13	5 mm	175	150	1.5	Magic View, Siemens		x	x	x
14	1.25–5 mm	nd	nd	1.75	Advantage 4.2 GE, Medical System	x			x
15	1.25 mm	nd	nd	1.75	Advantage 4.2 GE, Medical System	x			x
16	1.25 mm	nd	nd	0.935	Advantage 4.2 GE, Medical System	x	x	x	x
17	1.25 mm	nd	nd	nd	Siemens Leonardo	x		x	x
18	0.62–1.5 mm	nd	120	0.5				x	x
19	1.25 mm	nd	nd	nd	Siemens Leonardo	x	x
20	0.8–1 mm	nd	nd	nd		x			x
21	nd	nd	nd	nd	Sectra Pacs Workstation
22	1.6 mm	210	130	1.5	Osirix 3.9	x			x
23	nd	nd	nd	nd	Osirix – Poser Debut
24	3.75 mm (head–cervical spine)1.25 mm (chest–abdomen)	nd	nd	nd	PACS (Picture Archiving Computer System)
25	3.75 mm (head–cervical spine)1.25 mm (chest–abdomen)	nd	nd	nd	PACS (Picture Archiving Computer System)

KV: voltage; mA: milliampere; VR: volume rendering; MI: maximum intensity projection; SSD: surface shaded display; MPR: multiplanar reconstruction.

Characteristics of gunshot lesions observed on the deceased body (see Table 3).

Table 3.

Yes/no table on gunshot lesion comparison among autopsy and post-mortem computed tomography. Y=reported, N=not reported.

ID	Track wound	Lesion direction	Site of wound	Comparison of CT vs. autopsy findings	CT vs. autopsy% agreement/disagreement	Autopsy timing
11	Y*	Y*	Y (head, chest, abdomen)	Y*	N	N
12	Y	Y	Y (head)	Y	N	N
13	Y*	Y*	Y (head)	Y*	N	Y
14	Y*	Y*	Y (head, chest)	Y*	Agreement:Lethal wounds 100%Tracks 87%Direction 69%	Y
15	Y*	Y*	Y (head, neck, chest)	Y	N	N
16	Y	Y	Y (head, neck, chest)	Y	N	Y
17	Y*	Y*	Y (head, neck, abdomen, limbs)	Y**	Agreement:N. of lesions 100%Tracks 100%Bullet detection 100%Fragments detection 100%Major injuries 100%Cause of death 77,3%	N
18	N	N	Y (head, chest)	Y	N	N
19	Y	Y	Y (chest)	Y	N	Y
20	Y*	N	Y (head, chest, abdomen)	Y**	Agreement:Entrance wounds 69.2%Exit wounds 52.2 %Tracks 72.1%	N
21	Y	Y	Y (chest)	Y	N	N
22	Y	Y	Y (head)	Y	N	Y
23	Y	Y	Y (head, neck, chest, abdomen, limbs)	Y	N	N
24	Y	Y	Y (head, cervical spine, chest, abdomen)	Y	***Disagreement:Group I: 23,4%Group II: 15,6%	N
25	Y	Y	Y (head, cervical spine, chest)	Y	Agreement:detecting blood aspiration in the airways 70.7%	Y

*Data reported in the study are cumulative for all cases. Thus, it is not possible to infer all single data for each autopsy/CT scan.

**CT scan was not performed in all death cases.

***This study offers an evaluation of the PMCT/autopsy compared data made by two groups, one formed by board-certified radiologists (group I) and the other consisting of one board-certified forensic pathologist.

Percentage agreement PMCT vs. autopsy (see Table 3).

Results

Fifteen studies were included in this review (see Table 1), on the basis of the eligibility criteria described above. From now on, we will refer to each study using the ID number from Table 1.

Focusing on the considered parameters (see Table 2), with an exception made for two studies,^14,15 we found that all of the remaining studies reported the same CT scan slice thickness value. However, they did not actually use the same setting, which ranged from a minimum of 0.62 mm¹⁶ to a maximum of 5 mm.^17,18

The amperage and voltage settings were not explicitly stated in most studies, except for those by Oehmichen et al.,¹⁸ Thomsen et al.¹⁶ and Hasegawa et al.¹⁹

The pitch value was mentioned in only nearly half of the studies, and it was not described in Thali et al.,²⁰ Andenmatten et al.,²¹ Ruder et al.,²² Makhlouf et al.,²³ Abdul Rashid et al.,¹⁴ Maiese et al.,¹⁵ Kirchhoff et al.²⁴ and Scaparra et al.²⁵

With the exception of four studies,^16,23–25 the majority clearly reported the type of software used for the reconstruction phase: the related technique of MPR (Multi Planar Reconstruction, which allows for reconstruction on coronal, sagittal and oblique projections) was the most frequently used choice (used by 10 of the 14 studies), followed by VR (Volume Rendering, assigning a different colour and a different transparency to each tissue), which was chosen in eight of the 14 studies. The least used were the SSD (Surface Shaded Display, a three-dimensional reconstruction technique that shows the surface of an organ/bone with density detected in Hounsfield units), which was used in six of the 14 studies, and the MIP (Maximal Intensity Projection, useful for enhancing the hyperdense elements of the scan and particularly suitable for the study of blood vessels), which was used in only three studies.

All of the studies reported a comparison between CT scan and autopsy findings (see Table 3), although these comparisons were not homogeneous; most studies used only synthesized but not raw data, thus preventing a significant level of accordance among the studied parameters.

In detail, a comparison between CT scan and autopsy findings for gunshot-related death cases was reported in 12 of the 15 studies^{14,15,17–20,22,24–28} (see Table 3).

Moreover, two of the 15 studies performed this comparison by analysing only a portion of fatal gunshot wounds.^21,23 The paper by Harcke et al.²⁸ did not report any comparison.

All of the selected studies noted and described the trajectory of the bullet through the body and the direction of the shot, with the exception of Thomsen et al.¹⁶

Regarding the anatomical area involved in the injury, the head was the area most frequently investigated, being described in 13 of the 15 studies.^{15–21,23–28} The chest area was involved in the injury pattern in 11 of the 15 studies,^{14–17,22,24–28} whereas the abdominal area was involved in five of the 15 studies^{15,17,21,23,24} and the neck area in six;^{15,21,24,25,27,28} limb injuries were reported in Andenmatten et al.¹⁷ and in Maiese et al.¹⁵ Six of the 15 studies^{18,19,22,25–28} carefully reported the time elapsed between death and autopsy performance, but the remaining eight made no certain reference to it.

Overall, from data extraction the percentage agreement found by PMCT and autopsy was performed in five out of the 15 studies.^{21,23–25,26} From these data it is noteworthy that there is an important variability in the interpretation of ballistic results and the use of these information to establish the cause of death. For instance, only two studies^21,26 measured the percentage of agreement on the cause of death, which ranged from 77.3%²¹ to 100%²⁶ (see Table 3).

Discussion

Systematic reviews of data from heterogeneous case reports and case series, as in this instance regarding the use of post-mortem CT scan in gunshot-related deaths, can be complex because of the particular nature of the subject.²⁰

An overall analysis of the 15 studies that we found to be eligible for this review offered some meaningful hints, starting with the differences in settings used for executing the CT scans. Even if these settings did not appear to significantly influence the diagnostic results, these 15 studies do not show consistency in either the approach or the techniques adopted, thus making it difficult to accurately match the settings used (e.g. slice thickness, pitch, milliAmpere) with the results obtained according to the anatomical area or tissue examined.

This comparison would have been useful and helpful for providing a scale of diagnostic power and agreement among the various CT scan techniques that are currently being developed, to permit greater awareness about when (in which cases) and how (the settings) post-mortem imaging can be used as the gold standard in post-mortem diagnosis.

Specifically, the CT scan is frequently successful in detecting both the entry and exit wounds as well as the presence of foreign bodies (projectiles) and the trajectory of the bullet through the body. Furthermore, it can supply valuable information about organ damage, thus allowing the forensic practitioner to establish the cause of death.

On the skin, the entry wound is detectable by using SSD reconstruction techniques,^17,20,21,26 although the achromatic sight provided by the CT limits the analysis of the classical elements of entry wounds, such as the abrasion rim, tattooing and soot deposit (for more about this issue, see Makhlouf et al.,²³ who have found that CT scans aided in detecting entry wounds in 69.2% and exit wounds in 52% of penetrating wounds).

For bone, the detection of entry wounds is also complemented by the complete display of the accessory fractures that usually spread out from them, as afforded by 3D reconstruction techniques (e.g. SSD, MPR). In particular, secondary fractures, which are primarily produced by the transmission of kinetic energy from the bullet to the body at the moment of impact (e.g. in point-blank shots, wide secondary fractures usually occur not only in the skull but even in maxillofacial bones), are difficult to assess using traditional autopsy techniques. Instead, analysis of such fractures is greatly enhanced by CT scan, as asserted in most of the studies examined, with a few exceptions.^19,22,23

Regarding the trajectory of the bullet through the body, which was described similarly in all of the studies, when a bullet has passed through the bone, CT examination may show beveling of the bone in the direction along which the bullet travels. Beveled edges are directed inward at the bone margin of the entry wound and directed outward at the exit wound. Metallic particles and bone fracture fragments are additional indicators of directionality, because they are usually carried along the direction of projectile travel.

In contrast, passage through the soft tissue is usually characterized by a linear high-attenuation area that contains numerous small locules of gas.^21,24–27

Regarding organ damage, CT scan of the head area, depicted through MPR and SSD reconstruction techniques, allows the bullet’s trajectory to be traced through the brain as a hyperdense channel determined by bone or metallic fragments (comet-like tail) and/or by blood collection. For this kind of lesion, it is even possible to observe roundish structures that are isodense with air (‘air trail’), owing to tissue discontinuation and air penetration immediately after the explosion (pneumoencephalus, PNC).

The presence of hemothorax and pneumothorax,^{17,19,24–28} as well as hemoperitoneum,^24,25,27 dislocates mobile organs from their respective cavities, thereby complicating the proper detection and interpretation of the trajectory. However, a CT scan gives meaningful results to supplement the autopsy ones.

In thoracic injuries, entry and exit wounds appear as cutaneous discontinuations with areas of sub-cutaneous emphysema.²⁵ Depending on the organs involved in the injury, there will be pulmonary consolidation areas, owing to bleeding and/or pneumothorax if the bullet pierces the lung. Instead, if large vessels and/or the heart are involved in the trajectory, the resulting images will be characterized by hemomediastinum or hemothorax, with blood effusion making trajectory detection a difficult challenge.

Another interesting feature that could derive from the use of CT scans in gunshot wounds is detecting, with high reliability, the signs of blood aspiration in gunshot injuries;²⁵ this is usually helpful in determining whether an injury was inflicted during life or after death.

Limitations of the study

This study has several limitations. First of all we need to take into account that only the papers selected for characteristics of tomographic techniques were analysed to evaluate the comparison of autopsy and CT gunshot wound findings. In fact, few variables have been useful in order to provide a panoramic view of this complex and multivariate process. Moreover, this study failed to evaluate other important variables (such as study design, injury codes, etc.) because of the intrinsic limitation of the existing information obtained by the selected studies. In addition, there have been dramatic improvements in both software and hardware over time, and so these factors may directly affect the validity of our results. For these reasons further research is needed to confirm the conclusions drawn this work. Last, it should be noted that several of the studies processed for data extraction were conducted by the same teams. At least four are from the Swiss Virtopsy team, and at least three are from another team. This could impact and affect our results, since it is likely that these authors would use similar methods in their own studies with some modifications as technology advanced.

Conclusions

The results of the 15 selected studies on gunshot injuries showed that all of the different techniques used and described in the literature to date are rather similar in terms of diagnostic findings in the forensic field. In particular, despite their differences in terms of device model, image-acquisition technique (slice thickness, milliamperage and voltage settings) and reconstruction model used (e.g. MPR, SSD), the results of the studies here examined were mostly interchangeable in terms of output. In fact, in gunshot injury cases CT scanning plays a significant role in detecting entry and exit wounds, in locating the trajectory and in providing precise information about organ damage and, hence, about the cause of death. However, CT scanning alone may not be helpful to discriminate an entrance wound from an exit wound in all cases; this determination, which is invariably important in court, is actually assisted by the autopsy and aided by the classical criteria from forensic pathology. The CT scan does not provide sufficient information about critical issues such as the extent of lesion invasiveness among different anatomical tissues.

The absolute advantages of using a CT scan can be exploited to detect foreign bodies, even those of minimal size (either of metallic or bony kinds), or to investigate gaseous findings, such as pneumoencephalus or pneumothorax. In such scenarios, it is well known that the CT scan is indeed superior to the traditional autopsy and supplements the common scanning procedure with 3D reconstruction techniques¹⁵ to describe fractures (even secondary ones) or bone splinters with a more precise localization. Similarly, the CT scan can reveal gaseous findings that are scarcely recognizable during autopsy.

However, the information that can be derived from the studies examined here does not allow the various techniques to be compared in terms of the concordance between CT and autopsy results. Overall, from these data is noteworthy that there is an important variability in the interpretation of ballistic results and the use of this information to establish the cause of death. Few studies performed systematically the percentage agreement on the cause of death found by PMCT and autopsy.

Indeed, some anatomical areas (such as the head) have been more studied than others, thus potentially allowing for a high level of concordance between autopsy and CT findings; however, such a result may not be comparable to those from other studies, given the lack of standardized reference points at each level, even for the autopsy techniques. The CT scan was used differently in each of the examined studies, even though their purpose was the same. This inconsistency hinders the scientific validation and standardization of this method.

Therefore, because the CT scan has demonstrated flexible applications in the forensic field over the past 15 years, gaining great appeal in the scientific community, it would be worth focusing subsequent studies on conducting a more precise examination and comparative description of the settings used, or even aiming at replacing the common autopsy in select cases. This suggestion has also recently been made by Eriksson et al.⁶

If post-mortem topography in gunshot injuries is standardized and validated, it might be even stronger than common autopsy and be suitable as evidence in the courtroom. In order to increase the level of scientific evidence of forensic imaging methods, we believe that there is an urgent need for validation through consistent studies, which should be designed as blinded and cross-sectional and should be based on a comparison to proper reference standards (e.g. forensic autopsy or a priori knowledge of the experimental reality).

At the moment, there is still a need for an integration of the two approaches (autopsy and PMCT); innovative experimental approaches may be necessary to evaluate the potential CT techniques to improve the accuracy of forensic analysis when they are performed independently or in combination with a traditional autopsy.

Footnotes

Declaration of conflicting interests

The authors declare no potential conflict of interest with respect to the research, authorship, and/or publication of this article. Previous presentation: none.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Giovanni Battista Paliani

Contributorship statement

Fidelia Cascini and Riccardo Rossi are responsible of the concept, design and writing of the study. Matteo Polacco reviewed the radiological aspects. Giovanni Battista Paliani and Francesca Cittadini performed the revision of the literature. Antonio Oliva is responsible of the final check and revision of the work.

Guarantor of the work

Prof. Antonio Oliva, Associate Professor of Legal Medicine, Institute of Public Health, Section of Legal Medicine, Fondazione Policlinico A. Gemelli IRCCS, Università Cattolica del Sacro Cuore, L.go F. Vito 1, 00168 Roma, Italy.

References

Eckert

Garland

The history of the forensic applications in radiology.

Am J Forensic Med Pathol 1984; 5(1): 53–56.

Wullenweber

Shneider

Grummer

TH.

A computer-tomographical examination of cranial bullet wounds.

Z Rechtsmed 1977; 80: 227–246.

Baglivo

Winklhofer

Hatch

, et al. The rise of forensic and post-mortem radiology-Analysis of the literature between the year 2000 and 2011. J Forensic Radiol Imaging 2013; 1(1): 3–9.

Poulsen

Simonsen

Computed tomography as routine in connection with medico-legal autopsies.

Forensic Sci Int 2011; 163: 190–197.

Rutty

Morgan

Robinson

, et al. Diagnostic accuracy of post-mortem CT with targeted coronary angiography versus autopsy for coroner-requested post-mortem investigations: Prospective, masked, comparison study. Lancet 2017; 390(10090): 145–154.

Eriksson

Gustafsson

and Hoistad M. Diagnostic accuracy of postmortem imaging vs autopsy—A systematic review. Eur J Radiol 2017; 89: 246–249.

Brodgon

BG.

Forensic radiology. 2nd rev. ed. Boca Raton: CRC Press, 2011.

Polacco

Sedati

Arena

, et al. Visualization of myocardial infarction by post-mortem single-organ coronary computed tomography: a feasibility study. Int J Legal Med 2015; 129(3): 517–524.

Filograna

Tartaglione

Filograna

, et al. Computed tomography (CT) virtual autopsy and classical autopsy discrepancies: radiologist's error or a demonstration of post-mortem multi-detector computed tomography (MDCT) limitation? Forensic Sci Int 2010; 195(1–3): e13–e17.

10.

Oliva

Flores

Merigioli

, et al. Autopsy investigation and Bayesian approach to coronary artery disease in victims of motor-vehicle accidents. Atherosclerosis 2011; 218(1): 28–32.

11.

Garetier

Deloire

Dédouit

, et al. Postmortem computed tomography findings in suicide victims. Diagn Interv Imaging 2017; 98(2): 101–112.

12.

Cochrane Handbook for Systematic Reviews of Interventions, Version 5.0.2, 2009. The Cochrane Collaboration. Available at www.cochrane-handbook.org

13.

Liberati

Altman

Tetzlaff

, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: Explanation and elaboration. BMJ 2009; 339: b2700.

14.

Abdul Rashid

Martinez

Ampanozi

, et al. A rare case of suicide by gunshot with nasal entry assessed by classical autopsy, post-mortem computed tomography (PMCT) and post-mortem magnetic resonance imaging (PMMR). J Forensic Radiol Imaging 2013; 1(2): 63–67.

15.

Maiese

Gitto

De Matteis

, et al. Postmortem computed tomography: Useful or unnecessary in gunshot wounds deaths? Two case reports. Leg Med 2014; 16: 357–363.

16.

Thomsen

Jurik

Uhrenholt

, et al. An alternative approach to computerized tomography (CT) in forensic pathology. Forensic Sci Int 2009; 183(1–3): 87–90.

17.

Thali

Schweitzer

Yen

, et al. New horizons in forensic radiology: The 60-second digital autopsy-full-body examination of a gunshot victim by multislice computed tomography. Am J Forensic Med Pathol 2003; 1: 22–27.

18.

Oehmichen

Gehl

Meissner

, et al. Forensic pathological aspects of postmortem imaging of gunshot injury to the head: Documentation and biometric data. Acta Neuropathol 2003; 105(6): 570–580.

19.

Ruder

Ross

Preiss

, et al. Minimally invasive post-mortem CT-angiography in a case involving a gunshot wound. Leg Med (Tokyo) 2010; 12(3): 154–156.

20.

Thali

Yen

Vock

, et al. Image-guided virtual autopsy findings of gunshot victims performed with multi-slice computed tomography and magnetic resonance imaging and subsequent correlation between radiology and autopsy findings. Forensic Sci Int 2003; 138(1-3): 8–16.

21.

Andenmatten

Thali

Kneubuehl

, et al. Gunshot injuries detected by post-mortem multislice computed tomography (MSCT): A feasibility study. Leg Med (Tokyo) 2008; 10(6): 287–292.

22.

Hasegawa

Heinemann

Tzikas

, et al. Criminal gunshot wound and iatrogenic tension pneumothorax detected by post-mortem computed tomography. Leg Med 2014; 16(3): 154–156.

23.

Makhlouf

Scolan

Ferretti

, et al. Gunshot fatalities: Correlation between post-mortem multi-slice computed tomography and autopsy findings: A 30-months retrospective study. Leg Med (Tokyo) 2013; 15(3): 145–148.

24.

Kirchhoff

Scaparra

Grimm

, et al. Postmortem computed tomography (PMCT) and autopsy in deadly gunshot wounds – a comparative study. Int J Legal Med 2015; 130(3): 819–826.

25.

Scaparra

Peschel

Kirchhoff

, et al. Detection of blood aspiration in deadly head gunshots comparing postmortem computed tomography (PMCT) and autopsy. Eur J Med Res 2016; 21(1): 43.

26.

Levy

Abbott

Mallak

, et al. Virtual autopsy: Preliminary experience in high-velocity gunshot wound victims. Radiology 2006; 240(2): 522–528.

27.

Harcke

Levy

Abbott

, et al. Autopsy radiography: Digital radiographs (DR) vs multidetector computed tomography (MDCT) in high-velocity gunshot-wound victims. Am J Forensic Med Pathol 2007; 28(1): 13–19.

28.

Harcke

Levy

Getz

, et al. MDCT analysis of projectile injury in forensic investigation. AJR Am J Roentgenol 2008; 190(2): W106–W111.