Post-Mortem CT Imaging

Post Mortem CT
for Non-Suspicious
Adult Deaths
An Introduction
Ayeshea Shenton
Peter Kralt
S. Kim Suvarna
123
Post Mortem CT for Non-Suspicious
Adult Deaths
Ayeshea Shenton • Peter Kralt
S. Kim Suvarna
Post Mortem CT
for Non-Suspicious Adult
Deaths
An Introduction
Ayeshea Shenton Peter Kralt
Department of Radiology Department of Radiology
iGene London Ltd. iGene London Ltd.
Sheffield Sheffield
UK UK
S. Kim Suvarna
Department of Histopathology
Royal Hallamshire Hospital
Sheffield
UK
ISBN 978-3-030-70828-3 ISBN 978-3-030-70829-0 (eBook)

https://doi.org/10.1007/978-3-030-70829-0
© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature
Switzerland AG 2021
This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether
the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and
transmission or information storage and retrieval, electronic adaptation, computer software, or by similar
or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the
editors give a warranty, expressed or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
Post mortem computed tomography (PMCT) is an evolving diagnostic arena deal-

ing with autopsy practice. It has only been the last decade or so that has seen it
gaining traction in the United Kingdom although it has a more established role
elsewhere in the world.
This book was written to address a need for an introductory text for those enter-
ing the world of post mortem imaging. It is based upon the cases encountered during
the authors’ first few years of PMCT practice and their learning experiences. As
principally a radiology text, it is populated with many examples of both normal
appearances and structural pathology on PMCT. We hope that it addresses the needs
of the trainee in radiology, through to the established practitioner starting PMCT
work and others working in mortuaries. We also envisage that it will be a ready
resource for autopsy pathologists and Coroners in the United Kingdom.
We are extremely grateful to our many colleagues who have helped us with
images and provided critical commentary. We are particularly indebted to the dedi-
cated radiographers, administrative and information technology staff at iGene,
without whom this work would not have been possible. We also are conscious of the
support given by the mortuary and medicolegal staff at multiple sites across the
country, and to the pathologists who have validated our scans and occasionally
shown the limitations of radiology of the deceased. We are lucky to have had the
support and understanding of Springer, as a publishing team, and lastly salute our
families for their significant understanding while this project progressed.
We would like this text to be a positive-step forward for further research and
understanding in the field of PMCT and that it may serve the needs of the deceased
and their families into the future.
Sheffield, UK Ayeshea Shenton

Sheffield, UK Peter Kralt
Sheffield, UK S. Kim Suvarna
v
Contents
1 Introduction to the Investigation of Death and Post Mortem

Computed Tomography�� 1
Introduction�� 1
What Is an Autopsy?�� 1
The Arrival of Post Mortem Imaging�� 5
The Emerging Subspecialty of PMCT �� 7
Interpretation of PMCT�� 7
Collaborative Working �� 9
The Future of PMCT�� 10
References�� 11
2 Practical Considerations of Post Mortem Computed
Tomography and Report Writing�� 15
Factors Governing the Choice of PMCT �� 15
PMCT Scanning Facility Options�� 17
The Radiographer in the PMCT Unit�� 18
Scanning Technique �� 19
Body Handling and Positioning �� 21
Contrast Media and PMCT Angiography�� 24
Ventilated PMCT �� 25
The Radiologist and PMCT �� 25
The PMCT Report�� 26
Quality Assurance and Audit �� 31
References�� 32
3 Death, Post Mortem Changes and Decomposition
on Post Mortem Computed Tomography�� 35
Death and its Broad Causes �� 35
Post Mortem Interval �� 37
Initial Changes Seen After Death�� 38
Pallor Mortis�� 38
Algor Mortis�� 38
vii
viii Contents
Livor Mortis or Hypostasis�� 38

Rigor Mortis�� 41
Later Changes in the Body After Death�� 43
Decomposition from the Radiology Perspective �� 47
Post Mortem Gas�� 50
Post Mortem Clot�� 56
Animal Predation and Post Mortem Changes�� 57
Embalmed and Previously Autopsied Bodies�� 57
Reporting Post Mortem Change and Decomposition:
Pearls and Pitfalls�� 66
References�� 66
4 External Findings, Tubes and Devices on Post Mortem Computed
Tomography�� 69
The External Examination: The Pathologist’s Perspective�� 69
External Findings on PMCT�� 70
Tubes Seen on PMCT�� 72
General Comments�� 72
Airway Adjuncts�� 80
Vascular Access�� 83
Intraosseous Needles�� 86
Devices Seen on PMCT �� 88
Implantable Cardiac and Other Electrical Devices�� 88
Non-electrical Implants�� 92
Reporting External Findings and Devices: Pearls and Pitfalls�� 96
References�� 97
5 Post Mortem Computed Tomography of the Brain and Spinal Cord �� 99
Autopsy of the Brain and Spinal Cord: The Pathologist’s Perspective�� 99
Normal PMCT Findings�� 101
Brain�� 101
Intracranial Vessels�� 105
Spinal Cord�� 107
Abnormal PMCT Findings�� 109
Cranial and Cervical Traumatic Injury�� 109
Intracranial Infection and Masses�� 111
Intracranial Haemorrhage�� 115
Cerebral Infarction �� 119
Global Ischaemia �� 120
Ventricular Obstruction and Hydrocephalus�� 123
Spinal Cord Injury�� 124
Reporting Brain and Spinal Cord Findings: Pearls and Pitfalls�� 127
References�� 127
Contents ix
6 Post Mortem Computed Tomography of the Extra-Cranial

Head and Neck �� 131
Autopsy of the Extracranial Head and Neck: The Pathologist’s
Perspective �� 132
Soft Tissues�� 133
Orbits�� 134
Paranasal Sinuses �� 138
Trauma �� 140
Choking and Aspiration �� 142
Infections of the Head and Neck�� 143
Special Circumstances: Hanging �� 147
Types of Hanging �� 147
Imaging Findings �� 148
Fractures and Dislocations�� 151
Other Findings Resulting from Hanging�� 152
Hanging or Strangulation? �� 155
Reporting Head and Neck Findings: Pearls and Pitfalls�� 158
7 Post Mortem Computed Tomography of the Chest�� 161
Autopsy of the Chest: The Pathologist’s Perspective�� 161
Thoracic Airways �� 162
Lungs�� 163
Pleural Spaces�� 167
Mediastinal Vessels�� 169
Acute Airway Obstruction �� 171
Chest Trauma �� 172
Rib Fractures�� 175
Thoracic Aortic Rupture�� 175
Parenchymal Opacity and Consolidation�� 178
Pneumothorax and Pneumomediastinum�� 181
Pulmonary Embolism�� 181
Chest Malignancy and Lymphadenopathy�� 186
Industrial/Occupational Related Lung Disease�� 190
Special Circumstances: Drowning �� 190
The Mechanism of Death in Drowning Cases �� 191
PMCT Findings�� 191
x Contents
Fluid in the Major Airways�� 191

Lung Findings in Drowning�� 192
Fluid in the Paranasal Sinuses and Gastrointestinal Tract �� 193
Dry Drowning�� 194
Reporting Chest Findings: Pearls and Pitfalls �� 194
8 Post Mortem Computed Tomography of the Heart�� 199
Cardiac Disease as a Cause of Death�� 200
Autopsy Examination of the Heart: The Pathologist’s Perspective �� 201
Post Mortem CT of the Heart: Specific Additional Techniques�� 202
Calcium Scoring of the Coronary Arteries�� 203
Post Mortem CT Coronary Artery Angiography �� 205
Pericardial Sac �� 210
Basic Coronary Anatomy�� 210
Heart Walls and Contents�� 213
Pericardial Disease�� 213
Heart Size�� 220
Coronary Artery Disease�� 222
Myocardial Disease�� 227
Valvular Heart Disease�� 229
Uncommon Cardiac Conditions�� 230
Reporting Cardiac Findings: Pearls and Pitfalls�� 234
9 Post Mortem Computed Tomography of the Abdomen and Pelvis �� 239
Autopsy of the Abdomen: The Pathologist’s Perspective�� 239
Solid Abdominal Viscera�� 241
Autolysis and Gastromalacia �� 242
Pelvic Viscera�� 244
Intra-abdominal Gas Patterns�� 246
Intra-Abdominal Fluid Patterns �� 249
Ruptured Abdominal Aortic Aneurysm �� 252
Hepato-Biliary Pathology�� 256
Intra-abdominal Inflammatory Change and Infection�� 258
Abdominal Neoplasia�� 261
The Pancreas�� 261
Bowel Volvulus and Perforation�� 261
Bowel Ischaemia�� 264
Gastrointestinal Tract Haemorrhage�� 264
Contents xi
Abdominal Trauma�� 265

Bilateral Adrenal Haemorrhage �� 268
Reporting Abdominal and Pelvic Findings: Pearls and Pitfalls �� 268
10 Post Mortem Computed Tomography of the Bones and Soft Tissues �� 271
Autopsy of the Bone and Soft Tissues: The Pathologist’s Perspective�� 272
The Skin, Subcutaneous Tissues and Muscles�� 274
The Skeleton�� 275
Degenerative Changes and Arthritis�� 281
Appendicular Fractures�� 282
Vertebral Fractures �� 285
Spinal Cord Injury�� 289
Soft Tissue Injury�� 290
Musculoskeletal Infections�� 291
Peripheral Vascular Disease �� 292
Musculoskeletal Neoplasia�� 294
Special Circumstances: Fatal Major Trauma �� 295
Interpretation of Fatal Trauma �� 295
Fatal Haemorrhage as a Consequence of Injury�� 301
Reporting Bone and Soft Tissue Findings: Pearls and Pitfalls�� 302
11 Findings Related to Attempted Cardiopulmonary Resuscitation on Post
Mortem Computed Tomography�� 305
Autopsy Following CPR: The Pathologist’s Perspective �� 305
Cardiopulmonary Resuscitation�� 306
Skeletal Findings on PMCT�� 308
Rib and Cartilage Fractures �� 308
Sternal Fractures�� 313
Vertebral Fractures �� 316
Soft Tissue Findings on PMCT�� 318
Haemothorax and Pneumothorax�� 319
Pre-Sternal Haematoma �� 321
Intravascular Gas and Dilated Right Atrium�� 322
Abdominal Findings Following CPR�� 323
Reporting CPR-Related Findings: Pearls and Pitfalls �� 325
Introduction to the Investigation
of Death and Post Mortem Computed 1
Tomography
Introduction
Post mortem computed tomography (PMCT, Fig. 1.1) has an evolving role in the
investigation of non-suspicious adult death and offers the potential to avoid open
autopsy in many cases. Although this book is written from the perspective of work-
ing within the medicolegal coronial system of England and Wales, many of the
issues are common elsewhere within the United Kingdom and in many parts of
world. The methodology and considerations of PMCT should be fully understood
by involved medicolegal parties, radiologists, pathologists, relatives and society.
What Is an Autopsy?
In England and Wales, the majority of deaths are expected and fully understood,
allowing a cause of death to be provided by the family/hospital doctor. These cases
do not require any investigation, having often had numerous investigations in life
which would confirm the disease/s responsible for the death of the individual.
Currently, most autopsy cases, at the behest of HM Coroner (‘coronial cases’),
proceed because there is no clear cause of death or where there are issues of a medi-
colegal nature. This latter group includes deaths during surgery/anaesthesia, mater-
nal deaths, deaths in custody or in relation to occupational dust exposures—to name
but a few [1].
Of the deaths registered in 2018, 41% (220,600 deaths) were reported to coro-
ners [2], mostly being referred by a family or hospital doctor or, less commonly, the
police. The coroner decides if, and when, a post mortem examination of the body is
needed [1]. The coroner also has the power to decide what type of investigation is
most appropriate, whether by open dissection or whether to consider imaging solu-
tions, where available and suitable. The ability of imaging to wholly provide the
post mortem examination result is dependent on the circumstances of death [3], so
© The Author(s), under exclusive license to Springer Nature 1

Switzerland AG 2021
A. Shenton et al., Post Mortem CT for Non-Suspicious Adult Deaths,
https://doi.org/10.1007/978-3-030-70829-0_1
2 1 Introduction to the Investigation of Death and Post Mortem Computed Tomography
Fig. 1.1 Coronal PMCT

on soft tissue windows
offers a comprehensive
cross-sectional
examination of the
whole body
What Is an Autopsy? 3
cases should ideally be assessed on an individual basis, although some centres

prefer to scan all bodies as a first-line assessment [4].
There are also (non-coronial) consent/hospital autopsies, being hospital-based
investigations, requested by the clinical team to further consider the pathology
involved in the deceased. These investigations often revolve around the efficacy of
therapies applied. They may seek to investigate unusual findings seen before death
which did not achieve a clear answer in life. The relatives can agree to, or deny, the
opportunity for this autopsy investigation. Once being the majority of autopsies in
the United Kingdom, these consent examinations have declined in number over
recent decades. Reasons include a higher diagnostic performance during life due to
increasing use of clinical imaging and possibly also increasing public concerns
about the nature of an autopsy [5].
The last type of autopsy considered here is that of the forensic examination,
which is normally performed by specialist autopsy pathologists with training in
medicolegal and criminal pathologies. These cases deal with possible homicides,
manslaughter and other criminal acts.
Of all the autopsies currently performed in England and Wales, the majority are
within the coronial medicolegal framework, which will be the substrate for the cases
considered in this chapter and onwards in the book.
The coroner’s investigation covers four basic questions. The first three are rela-
tively straightforward, posing little difficulty for those considering the case. They
include the (rarely disputed) following questions:
1. who has died

2. where the individual died
3. when the individual died
It is the fourth question, namely how they died, which is variably complex
to answer.
In some cases, the triangulation of ante mortem data and conversations with
those who attended the deceased provide reassurance as to the case being a natural
death. However, if the circumstances and information do not provide sufficient con-
fidence in terms of the cause of death being natural, or merit an autopsy examination
automatically (as above), these will come to further coronial investigation.
There is insufficient space in this book to fully describe the variable methods of
conducting an open autopsy examination, in the traditional style that has been prac-
ticed in the United Kingdom for more than 150 years. However, if asked, the general
population would consider an autopsy examination to involve direct knife incisions
into the body and examination of the internal tissues. This examination can be a
matter of direct inspection only, with the results simply being written into a report.
In some cases, specialist investigations will also be applied, with these variably
including toxicology, microscope examination of tissues, microbiology, serology,
photography and other specialist tests [6]. A brief description of techniques and
approach used for the individual body cavities is given from the pathologist’s per-
spective in each relevant chapter.
In 2018, post mortem examinations were carried out in 39% of cases referred to
the coroner [2] amounting to 85,600 deaths. Legally, the coroner does not require
family or other consent to order a post mortem examination, but, where possible, the
relatives’ religious and cultural beliefs will always be considered [7].
The purpose of the autopsy could be argued to revolve solely around finding the
medical cause of death. However, this could be argued as too simplistic, as the
autopsy will look at the body as a whole, attempting to rationalise all the pathologi-
cal disorders and factors that led to the death of the individual. This analysis, tradi-
tionally provided by pathologists, allows the coroner to ascertain whether the death
was natural, or not, and thereby enables accurate registration of the death.
In some circumstances, the investigation may highlight aspects of imperfect
social, nursing or clinical care, with legal directions subsequently being made, or
could influence future health policies. Yet, the coroner may not be seen to directly
criticise the witnesses, or any institution, where later legal considerations are
possible.
On an individual level, the identification of diseases with a particular genetic risk
within the family can allow recommendation of clinical assessment of close family
members, for example those with a risk of sudden cardiac death [8].
In England and Wales, the leading cause of death for older adults (age over 80)
in 2018 was dementia (often described as Alzheimer’s disease), although this covers
a variety of degenerative and neurovascular pathologies. For men aged 50–80, the
leading cause remained ischemic heart disease, and for women aged 50–80, malig-
nancies of the breast, trachea, bronchus or lung were cited. Younger adults (gener-
ally under the age of 50) tended to die through other mechanisms such as suicide,
injury or other unexpected acute organic pathology [9].
In the investigation of a coronial non-suspicious death, the cause of death must
be ascertained on the ‘balance of probabilities’. This is a legal test that may be sum-
marised as being ‘more probable than not’! Thus, one might regard it as only just
over the 50:50 likelihood ratio. This is clearly a lesser legal test than that required in
criminal cases (and probably less certainty than required in most clinical practice)
but has been both pragmatic and realistic in terms of managing the coronial casel-
oad for many decades.
Looking at coronial-directed autopsy examinations, most cases are found to be
natural deaths and non-suspicious in nature, allowing appropriate paperwork to be
completed with no further medicolegal consideration. However, some cases will
require a court-based discussion, commonly described as the inquest. This non-
adversarial court room investigation (inquisition) has the opportunity to question
the relatives, the pathologist, medical practitioners and other parties (potentially
including the radiologist) in order to derive a cause of death. The inquest may
involve a jury, but the majority are managed by the coroner alone.
The Arrival of Post Mortem Imaging 5
The Arrival of Post Mortem Imaging
Diagnostic imaging has become extremely important in the routine clinical manage-
ment of the living, particularly with the advent of computed tomography and mag-
netic resonance imaging. Indeed, the results of the various scans performed during
life, mapped against other investigations, will often show the patterns of disease and
permit a clear cause of death to be defined later.
It also has to be understood that imaging has had a role in autopsy investigations
performed for many decades now. This has been relatively basic, generally revolv-
ing around plain radiograph identification of foreign objects (e.g. bullets) and frac-
tures. It is also noted that the drive towards more modern radiological techniques
being used in autopsy examinations has actually been increasing across the United
Kingdom for the last two to three decades.
The proportions of current radiology autopsy examinations vary greatly by juris-
diction [2], originally driven by faith groups in Manchester and other areas. By
2018, the proportion of post mortem examinations involving less-invasive tech-
niques such as PMCT was rising (3326 cases), compared to the prior year (1671
cases), [2] and more centres are planning to adopt imaging techniques in the next
few years.
In the hospital setting, PMCT can improve diagnosis of the cause of death over
clinical diagnosis alone [10, 11]. Potentially coupled with image-guided biopsy, the
radiology-centred autopsy may be argued to provide similar diagnoses to open
autopsy [12, 13]. Such biopsy techniques could be incorporated into PMCT imag-
ing with additional resource. This is not currently a common practice but could
perhaps be one of the aspirations for autopsy examinations in the next decades.
The demand for non-invasive or minimally interactive post mortem imaging has
increased in recent years for many reasons. There has been some public concern
about open autopsy with ethical and religious objections, possibly furthered by
organ retention scandals [14]. Furthermore, the numbers of autopsy pathologists
available to perform routine open autopsies are diminishing, with more pathologists
opting for less stressful surgical pathological disciplines.
Once radiological autopsy (PMCT) investigations started, it was quickly evident
that additional benefits could be realised. These include the permanent record of
anatomical findings, which unlike open autopsy with its necessarily destructive
technique means that pathological concerns can be seen by many without ongoing
tissue degradation [15]. Furthermore, the common use of radiological techniques
means that there is an instant familiarity with the technique, with this potentially
being of benefit when discussing cases with relatives or the court.
Broadly speaking, PMCT is particularly good at finding internal haemorrhage,
bony injury, foreign bodies, gas patterns and calcification (e.g. in coronary arteries).
It is therefore generally suited to investigate the relevant pathologies of adult deaths.
PMCT is particularly effective in the investigation of sudden death, where a cata-

strophic structural event such as aortic rupture or intracranial haemorrhage is confi-
dently visualised. By contrast, for example when used to investigate hospital deaths,
where a cause of death was not already clinically or radiologically apparent, PMCT
is often less definitive.
PMCT can be supplemented with angiography, a vascular infusion of radiopaque
contrast which may be ‘whole body’ or targeted locally. Typical indications for post
mortem CT angiography (PMCTA) are trauma, vascular pathologies [16] and deaths
after medical interventions [17]. In our practice of this technique, we have used
targeted CT coronary angiography in selected cases (see Chap. 8).
Very few centres use post mortem magnetic resonance (PMMR) as the mainstay
for autopsy investigation, despite the advantages in assessing soft tissue, such as
detecting oedema in the setting of myocardial infarction [15]. Minimally invasive
autopsy with PMMR has proven to be as accurate as invasive autopsy to detect
major pathological findings in paediatric cases (fetus, neonate and infant) [18]. The
investigation of paediatric death by post mortem imaging techniques is beyond the
scope of this book, with the reader being directed to specialist texts [19].
It would be fair to say that the facilities for PMMR in the autopsy arena are lim-
ited, being relatively more costly and time-consuming to perform. It is therefore not
widely practiced in the United Kingdom. Yet, a combined (PMCT, PMCTA and
PMMR) approach may be most comprehensive as has been used with the Virtopsy®
technique in Switzerland [20].
It has been suggested that a two-thirds reduction in the number of invasive
coronial autopsies may be achieved by the use of PMCT, possibly augmented by
coronary angiography [21]. Within the United Kingdom, one service reports a
cause of death could be given on the ‘balance of probabilities’ following PMCT
(with coronary angiography and ventilation) in 97.1% of cases [22], although it
should be understood that these examinations are mostly preselected for suitabil-
ity. In another centre, where all (i.e. no-selected) cases entering the mortuary are
scanned, it has been possible to achieve a cause of death in 55.6% of cases by
PMCT alone [4].
Coronial guidance recognises that post mortem imaging is a developing field,
and therefore the results from PMCT should be used cautiously [23]. The role of
PMCT radiologist is primarily to indicate pathologies leading to a medical cause of
death. Collaborative working with the pathologist is considered of paramount
importance in order to establish and run a post mortem imaging service. Generally,
both the pathologist and the radiologist prepare a report for the coroner, with the
radiology data usually being available first. The pathologist then performs an exter-
nal examination and considers the results of the radiology, balanced against the
previous medical history and circumstances surrounding the death of the individual.
At present, the UK Chief Coroner has recommended that the pathologist states the
cause of death or (if the cause of death cannot be defined) instructs the pathologist
to proceed to full or focussed autopsy [4, 23].
Interpretation of PMCT 7
The Emerging Subspecialty of PMCT
As described, plain film radiography has long been used in post mortem and foren-
sic imaging, with cross-sectional imaging practice growing for several decades in
coronial and forensic cases [24]. What was initially an extrapolation of image inter-
pretation, derived from the radiologists’ experience of scans in the living, post mor-
tem radiology is now a rapidly growing and distinct subspecialty [25], evidenced by
the increase in scientific publications that deal with this subject [26].
PMCT is well established in many countries such as Switzerland, Japan and
Australia. The widespread introduction of a non-invasive (or minimally invasive)
alternative to open autopsy is a realistic concept in the United Kingdom. This is well
underway, although the United Kingdom has been variably slow to follow this
uptake [27]. Proposals for national implementation have been made, as it is recog-
nised that there may no longer be the need for invasive examination in certain types
of death [28].
As can be predicted, change is not easy and may be hindered by a lack of scan-
ners, radiologists, and financial constraints. For imaging to become routine, cultural
adaptation will be required, especially for pathologists [13]. Resistance has variably
been encountered although this could serve to challenge advocates of PMCT to
develop a clear understanding and justification for the service [27] and increase the
wealth of relevant scientific evidence available.
Interpretation of PMCT
The technique and equipment required for the radiological autopsy are the same as
those used in hospital medicine. Yet, it should be understood from the start that
PMCT is different to clinical imaging. There are a wide range of post mortem
changes that evolve from the time immediately after death onward, to those seen in
the days and weeks following the cessation of life. It is apparent that the radiologist
cannot simply move from reporting cases of the living to describing the pathology
of the dead in one easy step. There are a variety of changes that reflect processes of
normal decomposition, which have to be mapped against the variable pathologies
that have caused the death of the individual. Consequently, appropriate training and
exposure to a range of cases is required if one wishes to achieve a good understand-
ing of PMCT [29].
As is often seen in the elderly, multiple structural comorbidities may be present.
For example, it is vital that the radiologist does not simply focus on the presence of
coronary artery calcification/stenosis or pneumonia. Such pathologies can be pres-
ent in the deceased and yet not be part of the pathophysiology of death. For exam-
ple, one might see a case who has died from significant cranial injury with
coincidental significant coronary disease. Of course, there is a possibility that the
coronary artery disease was involved in the circumstances leading up to the head
trauma. The presence of pneumonia is also common in the final stages of life. In this
regard, it may be a readily expected process rather than the primary disease. It may
be commonly seen in cases of disseminated malignancy or cerebrovascular disease,
where the final stage of the patient’s journey involves several days of palliative
bed rest.
Conversely, in other cases, there may be little pathology to find on imaging,
despite there being a strong hint from the circumstances. One example could be a
history of drug misuse and the finding of the deceased in the presence of drug para-
phernalia, but the PMCT does not show any significant structural pathology and
might be considered ‘inconclusive’, although the absence of structural findings is an
expected reality.
It is recognised that radiology is somewhat limited in terms of identifying condi-
tions such as drug overdose, metabolic derangement, sepsis, various dementias and
a variety of abdominal disorders that lead to death. In these cases, the role of PMCT
is mainly to exclude other structural pathologies, with this being of general help for
potential subsequent invasive autopsy.
This leaves a group of other inconclusive PMCT scans, which may be frustrating
for the radiologist as well as the pathologist. Scans with non-specific structural
changes and minor variations in terms of architecture are found regularly in
PMCT. They are generally more likely to be encountered in younger population (i.e.
less than 50 years). In these cases, without a diagnostic radiological pathology, it
may be impossible to state the definite cause of death from PMCT without supple-
mentary investigations. These possibilities range from progressing to an invasive
autopsy, toxicology investigations, sampling for microbiology, accessing spleen for
DNA studies and so on. This is not a failure of the PMCT technique. Rather, PMCT
can help streamline the further steps in the process of the investigation, thereby
decreasing the use of resources and time taken, alongside limiting the extent of
autopsy investigation from the relatives’ perspective.
Some centres request the radiologist to arrive at a defined cause of death, whereas
others simply request the radiologist to list any relevant findings, there is no single
best process. Care should be taken when defining the cause of death to separate the
cause of death from the ‘mode’ of dying (e.g. an abnormal physiological state such
as cardiac or respiratory arrest, syncope or coma) [30]. The cause of death should
not indicate a mode of death, but should be precise with regard to the pathology that
has ended the life of the person (e.g. cerebral infarction, pneumonia). It goes with-
out saying that the radiologist should be cautious about independently offering any
commentary that might take a cause of death assessment from the natural into the
realm of non-natural death (e.g. homicide, suicide, accident).
There are established guidelines in England and Wales for completing medical
certificates of cause of death [31], and these are useful in providing examples of
how medical causes of death are written.
One should always be mindful that deaths referred to the coroner, which have
been declared non-suspicious, may still include unnatural causes (such as acci-
dental trauma, suicide or interaction with other parties). Pathologists are trained
specifically to look at all cases with a view to suspicious circumstances, although
Collaborative Working 9
this is perhaps not something that comes automatically to the radiologist through
their training [32].
It is important for the radiologist to keep an open mind about the possibility of
concealed third-party involvement or negligence in care and discuss concerns—if
needed. Clearly, any findings that may be interpreted as signs of a violent or non-
natural death could change the nature of the case investigation. The radiologist
should not feel inhibited from reporting any concerns directly to the pathologist or
coroner who instructed the PMCT. In some circumstances, a forensic investigation
may immediately follow.
Even if, after PMCT, an open autopsy is needed, a clear interpretation of the
absence of pathology in certain body regions has an advantage, as it allows a ‘lim-
ited autopsy’ of the remaining parts. The PMCT images would serve as a permanent
record of the body before dissection or further decomposition and can be used to
‘cross-reference’ findings. Overall, a combination of PMCT, especially when aug-
mented with targeted PMCTA and a limited autopsy, is becoming accepted as the
‘gold standard’ in death investigation rather than a full open (traditional or invasive)
autopsy alone [3, 11, 14, 21, 33–35].
In time, with experience and collaboration between pathologists and radiolo-
gists, the true extent of the role of PMCT in death investigation will certainly further
evolve and improve.
Collaborative Working
Ideally, direct case discussion between the radiologist and the pathologist would
allow the best degree of collaboration [25]. It would also allow each to understand
the techniques and limitations of both imaging and open autopsy [36]. This com-
munication is comparable to the relationship between physician and radiologist in a
clinical setting [36]. However, it is clear that the variable pressures upon radiolo-
gists, pathologists, and other staff means that text (e.g. written reports and email)
communication of data is often the medium utilised. Overall, the guidance from the
UK Chief Coroner and Royal Medical Colleges suggests that the pathologist should
retain a central coordinating role in the investigation of deaths [37], although this
could potentially change in the future.
It remains a challenge to develop effective communication when in a remote
working setting, without face-to-face professional discussion, as perhaps would
happen in a clinical multidisciplinary team meeting, although multiple platforms to
facilitate this have been introduced in recent times. When the pathologist’s report is
available sometime after the radiologist’s report, alternative arrangements to pro-
vide case feedback may also need to be made to enhance interprofessional commu-
nication and improve learning and diagnostic outcomes. These can also be
augmented by subsequent reviews and ongoing audit.
As has been pointed out, radiologists with little experience in reporting PMCT
are at risk of misinterpretation of cases if they rely on their clinical experience of the
living alone [25]. This could have significant medicolegal implications and
potentially adversely impact on their personal professional accreditation. It is

imperative that appropriate training is provided if the radiologist wishes to begin
reporting PMCT given that there are differences in image analysis and a new scope
of ‘normal’ to appreciate. Furthermore, any training will have to deliver an under-
standing of the appearances of those dying with a range of progressive disorders that
are seen in community death investigations as well as those dying in hospitals
(which may also be in fields outside of the radiologist’s normal clinical practice).
Learning should therefore be sought from both radiology and pathology colleagues
in this respect.
It is equally possible for autopsy pathologists to learn and undertake PMCT anal-
ysis independently. This process is being pushed by some forensic pathologists
[38]. Yet, a radiologist still adds particular value to the team because of their general
cross-sectional and three-dimensional anatomic understanding, knowledge of CT
technique, anatomical/other artefacts and image pattern recognition skills.
The Future of PMCT
It is likely that the relatively new technique of PMCT investigation may alter death
statistics in autopsy cases. For example, currently PMCT has a relative lack of sen-
sitivity in diagnosing pulmonary emboli, which may see a decrease in these regis-
tered as the cause of death. By contrast, it may uncover other pathologies (hip
fractures for example) that might have previously gone unrecognised.
Apart from a ‘routine’ non-contrast–enhanced PMCT, there are various addi-
tional techniques such as angiography, ventilation, PMMR and image-guided biop-
sies that could find a place in post mortem and forensic imaging. Similar to the
concept of personalised clinical medicine, one may expect to see a more tailored
approach to the deceased in post mortem investigation. Indeed, it is likely that a
combination of these techniques may become the norm in UK deaths, although the
publicly funded coronial service may find it challenging to meet the additional cost.
One evolving technique that may benefit PMCT is dual-energy CT (DECT).
Images are acquired at different energy levels, allow differentiation of materials and
better tissue characterisation. Applications that might be useful include metallic
dental artefact reduction to improve the image quality of dental CT used for identi-
fication purposes [39]. Various other possible applications have also been suggested,
more relevant to the coronial setting, for example differentiation of arterial and
venous clots, improved characterisation of coronary artery plaques and detection of
foreign bodies [40].
Successful evaluation and implementation of any new techniques, mapped to the
collaboration between radiologist and pathologist, is likely to be the best way for-
ward. Yet, it is certain that further change lies ahead…
References 11
References
1. Dorries C. Coroners’ courts: a guide to law and practice [Internet]. 3rd ed. Oxford:
Oxford University Press; 2014. https://global.oup.com/academic/product/coroners-
courts-9780199566112.
2. Ministry of Justice. Coroners statistics annual 2018 England and Wales [Internet]. 2019.
https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_
data/file/810303/Coroners_Statistics_Annual_2018.pdf.
3. Morgan B, Rutty GN. How does post-mortem imaging compare to autopsy, is this a relevant
question? J Forensic Radiol Imaging [Internet]. 2016;4:2–6. https://linkinghub.elsevier.com/
retrieve/pii/S2212478015300277.
4. Suvarna SK, Kitsanta P, Burton JL. The effects of postmortem CT scanning all cases entering
a UK public mortuary: a 3-month pilot. J Clin Pathol [Internet]. 2017;70(10):903–5. http://jcp.
bmj.com/lookup/doi/10.1136/jclinpath-2017-204505.
5. Turnbull A, Osborn M, Nicholas N. Hospital autopsy: endangered or extinct? J Clin Pathol
[Internet]. 2015;68(8):601–4. http://jcp.bmj.com/lookup/doi/10.1136/jclinpath-2014-202700.
6. Burton J, Rutty G. In: Burton JL, Rutty G, editors. The hospital autopsy [Internet]. 3rd ed.
London: CRC Press; 2010. https://www.routledge.com/The-Hospital-Autopsy-A-Manual-of-
Fundamental-Autopsy-Practice-Third-Edition/Burton-Rutty/p/book/9780340965146.
7. Ministry of Justice. Guide to coroner services for bereaved people. [Internet]. 2020. https://
assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/
file/859076/guide-to-coroner-services-bereaved-people-jan-2020.pdf.
8. Suvarna SK. National guidelines for adult autopsy cardiac dissection and diagnosis—are they
achievable? A personal view. Histopathology [Internet]. 2008;53(1):97–112. http://doi.wiley.
com/10.1111/j.1365-2559.2008.02993.x.
9. Office for National Statistics. Deaths registered in England and Wales: 2018 [Internet]. 2019.
https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/
bulletins/deathsregistrationsummarytables/2018.
10. Inai K, Noriki S, Kinoshita K, Sakai T, Kimura H, Nishijima A, et al. Postmortem CT is more
accurate than clinical diagnosis for identifying the immediate cause of death in hospitalized
patients: a prospective autopsy-based study. Virchows Arch [Internet]. 2016;469(1):101–9.
http://link.springer.com/10.1007/s00428-016-1937-6.
11. Sonnemans LJP, Kubat B, Prokop M, Klein WM. Can virtual autopsy with postmortem CT
improve clinical diagnosis of cause of death? A retrospective observational cohort study in a
Dutch tertiary referral centre. BMJ Open [Internet]. 2018;8(3):e018834. http://bmjopen.bmj.
com/lookup/doi/10.1136/bmjopen-2017-018834.
12. Blokker BM, Wagensveld IM, Weustink AC, Oosterhuis JW, Hunink MGM. Non-invasive or
minimally invasive autopsy compared to conventional autopsy of suspected natural deaths in
adults: a systematic review. Eur Radiol [Internet]. 2016;26(4):1159–79. http://link.springer.
com/10.1007/s00330-015-3908-8.
13. Blokker BM, Weustink AC, Wagensveld IM, von der Thüsen JH, Pezzato A, Dammers R,
et al. Conventional autopsy versus minimally invasive autopsy with postmortem MRI,
CT, and CT-guided biopsy: comparison of diagnostic performance. Radiology [Internet].
2018;289(3):658–67. http://pubs.rsna.org/doi/10.1148/radiol.2018180924.
14. Roberts ISD, Benamore RE, Benbow EW, Lee SH, Harris JN, Jackson A, et al. Post-mortem
imaging as an alternative to autopsy in the diagnosis of adult deaths: a validation study.
Lancet [Internet]. 2012;379(9811):136–42. https://linkinghub.elsevier.com/retrieve/pii/
S0140673611614839.
15. Smith AP, Traill ZC, Roberts IS. Post-mortem imaging in adults. Diag Histopathol [Internet].
2018;24(9):365–71. https://linkinghub.elsevier.com/retrieve/pii/S1756231718301208.
16. Grabherr S, Egger C, Vilarino R, Campana L, Jotterand M, Dedouit F. Modern post-mortem
imaging: an update on recent developments. Forensic Sci Res [Internet]. 2017;2(2):52–64.
https://www.tandfonline.com/doi/full/10.1080/20961790.2017.1330738.
17. Heinemann A, Vogel H, Heller M, Tzikas A, Püschel K. Investigation of medical interven-

tion with fatal outcome: the impact of post-mortem CT and CT angiography. Radiol Med
[Internet]. 2015;120(9):835–45. http://link.springer.com/10.1007/s11547-015-0574-5.
18. Thayyil S, Sebire NJ, Chitty LS, Wade A, Chong W, Olsen O, et al. Post-mortem MRI versus
conventional autopsy in fetuses and children: a prospective validation study. Lancet [Internet].
19. Cohen MC, Scheimberg I. The pediatric and perinatal autopsy manual [Internet]. Cambridge:
Cambridge University Press; 2014. http://ebooks.cambridge.org/ref/id/CBO9781139237017.
20. Thali MJ, Dirnhofer R, Vock P. In: Thali M, Dirnhofer R, Vock P, editors. The virtopsy
approach [Internet]. 1st ed. Boca Raton: CRC Press; 2009. https://www.taylorfrancis.com/
books/9780849381898.
21. Roberts ISD, Traill ZC. Minimally invasive autopsy employing post-mortem CT and tar-
geted coronary angiography: evaluation of its application to a routine coronial service.
Histopathology [Internet]. 2014;64(2):211–7. http://doi.wiley.com/10.1111/his.12271.
22. Robinson C, Deshpande A, Richards C, Rutty G, Mason C, Morgan B. Post-mortem computed
tomography in adult non-suspicious death investigation—evaluation of an NHS based ser-
vice. BJR Open [Internet]. 2019;1(1):20190017. https://www.birpublications.org/doi/10.1259/
bjro.20190017.
23. Chief Coroner. Guidance No 1. The use of post-mortem imaging (adults) [Internet]. 2016.
https://www.judiciary.uk/wp-content/uploads/2013/09/guidance-no-1-use-of-port-mortem-
imaging.pdf.
24. Bolliger SA, Thali MJ, Ross S, Buck U, Naether S, Vock P. Virtual autopsy using imaging:
bridging radiologic and forensic sciences. A review of the virtopsy and similar projects. Eur
Radiol [Internet]. 2008;18(2):273–82. http://link.springer.com/10.1007/s00330-007-0737-4.
25. O’Donnell C, Woodford N. Post-mortem radiology—a new sub-speciality? Clin
Radiol [Internet]. 2008;63(11):1189–94. https://linkinghub.elsevier.com/retrieve/pii/
S0009926008002122.
26. Baglivo M, Winklhofer S, Hatch GM, Ampanozi G, Thali MJ, Ruder TD. The rise of forensic
and post-mortem radiology—analysis of the literature between the year 2000 and 2011. J
Forensic Radiol Imaging [Internet]. 2013;1(1):3–9. https://linkinghub.elsevier.com/retrieve/
pii/S2212478012000044.
27. Rutty JE, Morgan B, Rutty GN. Managing transformational change: implementing cross-
sectional imaging into death investigation services in the United Kingdom. J Forensic
Radiol Imaging [Internet]. 2015;3(1):57–60. https://linkinghub.elsevier.com/retrieve/pii/
S2212478014001191.
28. NHS Implementation Sub-Group of the Department of Health. Can cross-sectional imaging as
an adjunct and/or alternative to the invasive autopsy be implemented within the NHS? [Internet].
2012. https://www.aaptuk.org/downloads/Cross-Sectional-Imaging-October-2012.pdf.
29. Roberts I, Traill Z. The radiological autopsy. In: Suvarna SK, editor. Atlas of adult autopsy
[Internet]. Cham: Springer International Publishing; 2016. p. 362. http://link.springer.
com/10.1007/978-3-319-27022-7_13.
30. Saukko P, Knight B. Knight’s forensic pathology [Internet]. 4th ed. Boca Raton: CRC
Press; 2015. https://www.routledge.com/Knights-Forensic-Pathology/Saukko-Knight/p/
book/9780340972533.
31. HM Passport Office. Guidance for doctors completing medical certificates of cause of death
in England and Wales [Internet]. 2018. https://assets.publishing.service.gov.uk/government/
uploads/system/uploads/attachment_data/file/757010/guidance-for-doctors-completing-
medical-certificates-of-cause-of-death.pdf.
32. Leadbeatter S, Lucas S, Lowe J. Standards for coroners’ pathologists in post-mortem
examinations of deaths that appear not to be suspicious [Internet]. The Royal College of
Pathologists, London; 2014. https://www.rcpath.org/uploads/assets/1b02cfb9-000a-4b2f-
b6b80256b719a5ee/Standards-for-Coroners-pathologists-in-post-mortem-examinations-of-
deaths-that-appear-not-to-be-suspicious.pdf.
References 13
33. Rutty GN, Morgan B, Robinson C, Raj V, Pakkal M, Amoroso J, et al. Diagnostic accuracy
of post-mortem CT with targeted coronary angiography versus autopsy for coroner-requested
post-mortem investigations: a prospective, masked, comparison study. Lancet [Internet].
34. Le Blanc-Louvry I, Thureau S, Duval C, Papin-Lefebvre F, Thiebot J, Dacher JN, et al. Post-
mortem computed tomography compared to forensic autopsy findings: a French experience. Eur
Radiol [Internet]. 2013;23(7):1829–35. http://link.springer.com/10.1007/s00330-013-2779-0.
35. Clarke M, McGregor A, Robinson C, Amoroso J, Morgan B, Rutty GN. Identifying the correct
cause of death: the role of post-mortem computed tomography in sudden unexplained death. J
pii/S2212478014001075.
36. Flach PM, Thali MJ, Germerott T. Times have changed! Forensic radiology—a new challenge
for radiology and forensic pathology. Am J Roentgenol [Internet]. 2014;202(4):W325–34.
http://www.ajronline.org/doi/10.2214/AJR.12.10283.
37. Maskell G, Wells M. RCR/RCPath statement on standards for medico-legal post-mortem
cross-sectional imaging in adults [Internet]. The Royal College of Radiologists and The Royal
College of Pathologists, London; 2012. https://www.rcr.ac.uk/system/files/publication/field_
publication_files/FINALDOCUMENT_PMImaging_Oct12.pdf.
38. Bedford PJ. Should pathologists be reporting forensic CT scans? Acad Forensic Pathol
[Internet]. 2012;2(2):198–201. http://journals.sagepub.com/doi/10.23907/2012.028.
39. Alkadhi H, Leschka S. Dual-energy CT: Principles, clinical value and potential applications in
forensic imaging. J Forensic Radiol Imaging [Internet]. 2013;1(4):180–5. https://linkinghub.
elsevier.com/retrieve/pii/S2212478013000956.
40. Persson A, Jackowski C, Engström E, Zachrisson H. Advances of dual source, dual-energy
imaging in postmortem CT. Eur J Radiol [Internet]. 2008;68(3):446–55. https://linkinghub.
elsevier.com/retrieve/pii/S0720048X08002507.
Practical Considerations of Post Mortem
Computed Tomography and Report 2
Writing
Introduction
With the increasing use of post mortem computed tomography (PMCT), there will
be more call to commission services in bulk fashion, often as part of a service con-
tract rather than an intermittent or ad hoc service, or paid for by relatives. There are
many considerations, some of which will be familiar to those working in a clinical
imaging department. Others will be specific to the post mortem setting and may
therefore be unfamiliar unless appropriate training has been delivered. Some aspects
of practice will interface with national or local guidelines, that need to be applied.
Of note, there are the standards for medicolegal post mortem cross-sectional imag-
ing in adults, written by the UK Royal College of Pathologists and the Royal College
of Radiologists [1]. At the time of writing this book, new guidelines are being
developed.
Factors Governing the Choice of PMCT
Case selection and how PMCT fits into death investigations should be an agreed
process with the relevant medicolegal authorities, the pathologists, the mortuary and
the relatives of the deceased. The process of using PMCT should be procured in
stages. The first question is whether the case is suitable for PMCT, or whether a
standard open autopsy without PMCT is the solution. It is noted that PMCT is a
useful tool to confirm many specific pathological lesions and to exclude certain
findings, but it is neither perfect nor all-encompassing. Some aspects of industrial
lung disease, sepsis and metabolic processes may require open autopsy investiga-
tion, with the argument applied that adding in a PMCT study is just delaying mat-
ters. Others argue that knowledge from a scan before considering an open autopsy
is always valuable [2].

Switzerland AG 2021
https://doi.org/10.1007/978-3-030-70829-0_2
16 2 Practical Considerations of Post Mortem Computed Tomography and Report…
Based on local preferences, and also given that there is some variation in coronial
decision-making and different realities for the procurator fiscal, the case mix of
PMCT studies can differ between services in different parts of the country. In some
jurisdictions, all cases entering the mortuary will be scanned, including hospital
deaths. In other centres, hospital deaths are excluded because the yield of PMCT to
provide a cause of death is lower in this patient group. In some other jurisdictions,
the progress towards PMCT is based upon the willingness and ability of the family
or state to pay a fee for the radiological investigation. In short, there is no current
standard rule or system of which case should progress to PMCT.
The coroner (or other medicolegal party) often has a specific set of cases that
require active autopsy consideration. It also has to be understood that the informa-
tion available to those considering these cases is often limited. This can make the
decision of whether PMCT might be a suitable type of autopsy examination quite
difficult. Examples of cases that almost always require some form of investigation
include:
• The adult found deceased with no overt cause of death or appropriate supportive
history, whether in hospital, nursing home or community.
• Witnessed collapse in an individual with no significant preceding history of ill-
ness or similar collapse events.
• Trauma, whether of a non-suspicious nature or non-natural type, including work-
place injuries, road traffic incidents, injuries whilst inebriated and so on.
• Those not seen alive in the hospital or emergency department but certified
deceased upon arrival at hospital in a state of cardiorespiratory arrest.
• In-hospital deaths, particularly in cases undergoing medical, obstetric, anaes-
thetic or surgical interaction, and where the underlying pathophysiology is
unclear.
• Suicides, including hanging, drug overdose, various self-directed trauma, etc.
• Drowning in sea and fresh water.
• Industry-related deaths with these commonly implicating asbestos, coal, silica
dust exposures, although this is not an exclusive list.
• Cases that have complaints and/or concerns regarding the medical/nursing care
afforded to the deceased beforehand.
When PMCT is being offered and discussed with the relatives, the bereaved fam-
ily should be informed of the potentials and limitations of imaging. This might
mean that if no clear diagnosis is achieved, a traditional open autopsy may still be
required [3]. It is worth noting that cases with areas of concern will often merit
additional open autopsy procedures anyway. In many other cases, however, the
radiological information obtained will allow a cause of death to be defined by the
pathologist and to permit appropriate registration of death paperwork.
Factors Governing the Choice of PMCT 17
PMCT Scanning Facility Options
There are both advantages and disadvantages of using dedicated rather than public
health service facilities. Co-located mortuary and CT facilities offer easy body
transfer from storage fridges to the scanner and reduces scan turnaround times. The
benefit of having experienced mortuary staff is vital when learning how to handle
bodies and when angiography techniques are being employed. The mortuary staff
can also assist with the timely aspiration of toxicology samples for any cases that
might require this investigation (Figs. 2.1, 2.2, and 2.3).
Fig. 2.1 Post mortem

sampling of blood can be
accomplished by direct
needle aspiration from the
femoral vessels
Fig. 2.2 The vitreous can

be sampled by needle
aspiration of fluid from the
eye, with subsequent
re-filling of the eye
afterwards with water
Fig. 2.3 The bladder is

usually easy to sample by
needle passed centrally just
above the pubis, although
this requires the bladder to
contain some urine. The
PMCT can be used to
assess bladder filling prior
to aspiration
The disadvantages of this co-location are that there will likely be an initial set-up
capital investment and (probably) no opportunity to use the scanner for potentially
profitable clinical activity to offset running costs. Radiographic staff will need to be
employed to work in co-located units. Radiologists may work on-site with the added
value of providing individualised protocols, reviewing images before the body is
returned from the scanner and being directly available for case discussions. However,
it may be more practical for a larger group of radiologists to be remotely accessible
through teleradiology, email and equivalent on-line communications.
Using local health service facilities for PMCT by special arrangement may be
more simple and cheaper overall, but this should not conflict with imaging of the
living. Radiographers might then be more accessible throughout the day, but some
will need to be trained and available for angiogram procedures or other appropriate
personnel be available if required. Another factor is the sensitivity element of bring-
ing dead bodies into a clinical area, where staff (or visitors) may not be accustomed
to such investigations or indeed handling the deceased.
There is no requirement by the UK Human Tissue Authority (HTA) that radio-
logical imaging of a body (including angiography) needs to occur on licensed prem-
ises. This contrasts with traditional open autopsy where tissue sampling would
likely occur. Imaging sites do however need to be licensed if there is any potential
removal of tissue, such as blood for toxicology or needle biopsy.
The Radiographer in the PMCT Unit
The conventional practice of PMCT requires a trained radiographer working with

the body of the deceased in a CT scanner unit. The radiographer undertakes the
imaging with the digital images being stored for later consideration, or rarely imme-
diate reporting by a radiologist or imaging-trained pathologist. Internationally, other
systems exist with mortuary staff or forensic technologists being trained to provide
day to day use of equipment [4, 5].
The Radiographer in the PMCT Unit 19
An experienced radiographer can also offer a wealth of knowledge when setting

up services and how to use the equipment to achieve the best images possible. They
will be able to find solutions to ‘work around’ difficult patient positioning, deal with
various post mortem artefacts and compensate for scanner limitations. If the scanned
images are sent remotely to the radiologist, then a more autonomous role of the
radiographer will be necessary to undertake additional image series, perhaps as per
agreed protocols or as judged appropriate to the circumstances. Any PMCT service
will be dependent on the quality of images produced. In the United Kingdom, the
Society of Radiographers has published standards of radiographic practice [6].
One early worry was the potentially distressing nature of the service, dealing
with the dead, with possible psychological risks to the staff involved. Appropriate
education and support should be made available for any staff working in this envi-
ronment if they have never interacted with the deceased beforehand [6]. However,
our experience to date is that the radiographers adapt well with appropriate and
ongoing support.
Scanning Technique
There are many options for scanning technique, and these will be tailored to local
facilities and preferences. One option is to scan the whole body, from head-to-toes
in one acquisition, capturing all the data, rather than potentially having to return the
body to the scanner for additional assessment (Fig. 2.4).
The total body scan depends on the scanner available; factors that should be
considered include the room size, gantry bore size and table length. If scanning
rooms have size limitations, then body handling will be an issue, which has implica-
tions for the radiographer as well as the mortuary staff.
Alternative protocols split the body into sections such as head and neck, chest to
pelvis and pelvis to toes. Not all PMCT protocols include full leg length imaging, as
this part of the body (thigh downwards) rarely has any unexpected pathology of
significance in terms of the cause of death.
Suggested scan parameters will vary for different machines although they do not
differ much from clinical applications to ensure the best images. Dose reduction
techniques are not an essential consideration. However, it does not automatically
mean that increasing the radiation dose will result in better pictures. Scans may be
reconstructed to variable slice thicknesses, but to make best use of multi-planar
reconstructions (MPR), 1 mm slices are recommended. When scanning thin slices
over such a length, however, X-ray tube overload may become an issue, particularly
if numerous cases are being examined in a relatively short period of time.
If whole-body scanning is required but technically not possible on the available
scanner, then two acquisitions may be undertaken. This solution involves moving
the body from head-first to feet-first, although this has significant time and manual
handling implications. Rarely, only one area of body scanning is required, such as
the head. In these cases, there is often cross-sectional imaging available taken in
hospital shortly before death.
Fig. 2.4 Scanogram/scout

image of a whole body
with clothing, shoes and
jewellery still in place. The
body is positioned with
knee flexed and head
turned due to rigor mortis
The issues of radiation dose, renal function (for contrast administration) and
subject movement artefact do not apply in the PMCT setting, but there are other
issues to consider. Contrast administration (angiography) into a non-flowing vascu-
lar system, body positioning problems and decomposition factors may be present.
Looking at the safety requirements surrounding use of ionising radiation [7],
there are regulations governing radiation protection, equipment maintenance and
calibration, also health and safety for handling the deceased, infection control and
cleaning factors to be considered, notwithstanding the practical aspects of manual
handling. These matters will need careful evaluation by those working in the mortu-
ary and PMCT suite, with potential variations to solutions compared to clinical
settings.
Wherever the scan takes place, it is imperative that image acquisition is of a high
standard, ideally being performed as soon as practicable so that any decomposition
change is minimised. The process of imaging should not delay the investigation and
progression of the body towards funeral arrangements. Factors of privacy and dig-
nity should always be preserved at all times when providing this sensitive service.
Body Handling and Positioning
During transportation to the mortuary and within the building itself, the body will
normally remain within a body bag (Fig. 2.5). This bag protects mortuary staff, the
radiographer and the scanner from potential leakage of body fluids. The scan can be
performed without opening the bag, which is advisable in cases of decomposition
(Fig. 2.6), severe trauma with body fluid leakage or in cases which have a potential
infection risk. One is also mindful of the reality that radiographers are not generally
used to direct inspection or interaction with deceased bodies, and it is an aesthetic
principle to keep the bag closed during the interaction with the scanner suite on
most occasions.
Fig. 2.5 A PMCT scanner

with a body awaiting scan.
The body is enclosed in a
securely sealed bag for
safe and hygienic case
transfer
Fig. 2.6 Coronal view of

the head to pelvis on lung
windows shows extensive
soft tissue gas due to
decomposition
If possible, body bags should be free from metallic or other items, as this may
cause artefacts. This does not mean that the body will arrive in an unclothed state or
covered solely by a shroud. It follows that, on occasion, the bag may require open-
ing to permit removal of such items if this will enhance the scan or prevent artefacts
from interfering with the images obtained. Permission for this interaction may be
sought from the pathologist or medicolegal representative instructing the scan.
Other aspects of interaction with the body may include better positioning in the
scanner or to lift the arms over the head when scanning the torso. The radiographer
may perform such tasks, although the mortuary staff may be better placed to deliver
this requirement or to assist the radiographer. As always when dealing with the
deceased, no interaction should be made unless requested and providing it is safe.
Rigor mortis (see Chap. 3) can result in abnormal positioning (Figs. 2.4 and 2.7).
This presents a particular problem when body parts are positioned in such a way
that they do not readily fit through the aperture of the CT scanner. It may be suitable
to teach the operator to ‘break’ rigor mortis (a method of firmly stretching the tis-
sues adjacent to a joint to realign the limb) to allow better positioning of the body.
A large scanner bore size is helpful when scanning bodies in such nonconventional
positions, either due to rigor or other pathology (Fig. 2.8). Such scanners are also
useful in dealing with those bodies with a raised body mass index (Fig. 2.9).

the head, neck and chest on
bone windows shows
abnormal body positioning
due to rigor mortis
Fig. 2.8 Sagittal view of

the head to pelvis on soft
tissue windows shows
non-conventional
positioning due to fixed
flexion deformities in a
patient with multiple
sclerosis
Fig. 2.9 Axial view of the

chest on soft tissue
windows shows the arms to
be not fully included in the
field of view for this obese
patient, despite a large
gantry bore. There is also a
reduction in image quality
due to X-ray beam
hardening
Contrast Media and PMCT Angiography
In general, one knows that clinical CT without contrast is more limited in the detail
it can offer, particularly regarding the vascular system and soft tissues. The same
reality applies to PMCT examinations, although plain CT is the norm for PMCT in
the United Kingdom.
The use of contrast may potentially increase the yield of the examination [8, 9],
assuming it is successfully placed into the vasculature. However, one must appreci-
ate that it automatically changes the nature of the test from non-invasive to mini-
mally invasive. Some centres may prefer to have a decision algorithm for when to
progress to angiography, whereas others have devolved this decision to either the
radiologist and/or the pathologist.
For PMCT with angiography, cannulation of the deceased is commonly under-
taken by a mortuary technician or pathologist. It may prompt consideration of
whether toxicology, microbiology, and other tests are required at this point. The use
of targeted contrast has not been shown to affect subsequent toxicology analysis,
but medicolegal considerations usually mandate that all removal of samples for
toxicological analysis (Figs. 2.1, 2.2, and 2.3) and other tests should occur prior to
angiography to avoid any potential legal challenge to test results.
There are different options for post mortem CT angiography (PMCTA), such as
whole-body or targeted techniques [10], which will be chosen according to the pre-
ceding history, the question being asked and local skill set of the radiographer and
facilities available. The technique and application of targeted coronary angiography
are discussed further in Chap. 8.
The Radiologist and PMCT 25
Ventilated PMCT
The collapse of lung tissue following death and the secondary pooling of blood into
the pulmonary circulation may limit radiological assessment of the lung paren-
chyma of the deceased. Consequently, methods of artificially ventilated PMCT have
been developed in order to improve the diagnostic quality of lung imaging using a
number of different techniques [11]. The technique of inflating the lung can mimic
inspiration, thereby clearing some post mortem atelectasis and hypostasis without
altering any true pathology [11]. As will be appreciated, any such technique requires
additional time and resource. It should certainly not be used where there is risk of
transmitting any respiratory infection.
When already present (e.g. following resuscitation attempts) an existing airway
can be used. However, if none is present, an airway can be introduced or instead
lung inflation can be achieved by using a continuous positive airway pressure
(CPAP) mask. In adults, the airway pressure can be provided and maintained by a
portable ventilator. If rigor mortis prevents oral airway insertion, a tracheostomy
can be performed, although this is clearly an invasive procedure that should not
occur without due consideration and appropriate permission.
The Radiologist and PMCT
The radiological skills required to interpret PMCT are broadly the same as those
required to interpret cross-sectional imaging in the living [1] although the subject
knowledge differs. The interpretation should be undertaken by either a qualified radi-
ologist or a medical practitioner with equivalent competencies in cross-sectional
imaging. However, even for radiologists with many years of experience, it is clear that
some training is required when dealing with the deceased, as there are different con-
siderations to be made if one wishes to avoid significant image misinterpretation.
It is recommended that specific training is supplied by an experienced practitio-
ner, rather than the ad hoc method of simply learning ‘as one goes along’. For
example, a knowledge of the wide range of normal appearances after death and
particularly the appearance of decomposition is required from the beginning. This
is not always obvious and cannot be extrapolated solely from knowledge of scan-
ning the living or a few post mortem cases. Post mortem changes vary greatly
depending on the time since death occurred and many other factors.
Most clinical radiologists will also require some training in the mechanisms and
language of death used in the medicolegal arena and by pathologists. They will
require some understanding of how the process of death investigation takes place.
This has particular pertinence, as many of the cases will ultimately be debated in
court. It should be understood that lawyers and other medicolegal practitioners may
take particular interest in the wording of a radiology report, particularly if it is at

variance with the preceding clinical history or any open post mortem pathology.
There is currently no mandatory training or examination in PMCT, although
many radiology training centres are now starting to teach this method of investiga-
tion as part of standard radiology learning. In the United Kingdom, there are also
several groups providing dedicated teaching for established practitioners. There is a
wealth of scientific literature available [12]. It is noted that the UK Royal College of
Pathologists autopsy exam now has questions dealing with PMCT cases. However,
the pathologists are not required to read the radiological images but rather to inter-
pret the radiology reports and to debate possible causes of death.
If the radiologist has no prior experience of working in the autopsy arena, then
going to see open autopsies, if possible, with local pathologists that use PMCT in
their practice, is invaluable for understanding the examination. Alternatively, there
are books that describe and illustrate the open autopsy procedure [13, 14]. The con-
verse is true for pathologists, with their practice being enhanced if they develop an
understanding of the methodology of imaging and how reports are produced. This
will benefit from targeted teaching and reading relevant texts. This enhanced under-
standing between the different disciplines will naturally benefit both the radiologist
and the autopsy pathologist.
The PMCT Report
There is no single, fixed and perfect way to write a PMCT report. There are many
different models that are used across the United Kingdom and indeed further afield.
However, all reports start in the same way, with the referral of the case for examina-
tion, usually with accompanying information. The request for imaging will come
from the coroner’s office (or equivalent) and will be accompanied by a medicolegal
report and/or clinical history from the general practitioner or hospital clinicians.
The full medical history should always be made available for use.
As ever, the history is of paramount importance to the interpretation of any imag-
ing findings. Any circumstantial information surrounding the death or medical his-
tory will need to be set against whether the death was expected or not; whether it
was sudden or prolonged; witnessed or un-witnessed. The history data must be con-
sidered against the known past medical history for the individual whilst appreciat-
ing that any previous medical history may actually be irrelevant to the cause of
death. How and where the body was found may also have bearing upon the interpre-
tation of findings. The data provided may also guide the need for additional tests,
such as toxicology and microbiology.
In addition to this background data, it can be helpful to know the time since death
(post mortem interval or PMI), as this may affect the interpretation of appearances/
decomposition. Knowledge of whether cardiopulmonary resuscitation (CPR) was
attempted is important, particularly in the context of broken ribs from chest com-
pressions. The reporter should read any information made available but bear in mind
The PMCT Report 27
that sometimes the information is incomplete or preliminary in its nature and on

occasion almost completely wrong!
The PMCT scan should be reviewed in the same systematic manner as any clini-
cal examination. The analysis needs to be undertaken with appropriate time being
available, without distractions and with the opportunity to interact with other parties
as deemed appropriate. Unlike an open autopsy, in PMCT there exists the opportu-
nity for further independent review of unaltered images and therefore possible
future case debate and challenges. The reporter needs to be mindful that adversarial
criticism of the report may occur later in court. However, looking at this from a
positive perspective, the ability to store images and have additional opinion later
also allows for quality improvement and learning, such as from double reporting
and audit. It is also relevant in terms of case discussion between colleagues before
any PMCT report is issued to the pathologist.
In general, it is important to document all existing pathology that may, or may
not, have contributed to death. It is only by taking a holistic perspective on the case
that any determination can be made. However, not all cases have a clear interpreta-
tion or diagnosis. The radiologist’s conclusion in terms of the likely cause of death
might cross quite a range of possibilities and probabilities. These might be expressed
as ‘definite’, ‘probable’ or ‘possible’ causes of death.
When reporting PMCT as an adjunctive technique, (such that full or limited open
autopsy follows), the report should allow the pathologist to plan the autopsy. Perhaps
the most useful aspect of an ‘inconclusive’ PMCT analysis is permitting the pathol-
ogist to focus on one part of the body. Arbitrarily, the body compartments may be
divided into the cranial cavity, the neck, the chest, the abdomen and the pelvis.
Information as to the normality of tissues in these regions may avoid time being
wasted on unnecessary dissection, which would have no pathology or relevance to
the cause of death. In addition, limiting the dissection may also improve the accept-
ability of the open autopsy to the relatives.
The PMCT analysis should also provide information with regard to any potential
open autopsy hazards (e.g. bony fragments) as well as the position of any foreign
bodies (orthopaedic implants, pacemakers, bullets etc.). The scan may also provide
important information about anatomic anomalies such as aberrant vessels and any
suspicion of infection risks such as tuberculosis. Furthermore, the consequences of
any surgical intervention previously may be apparent on PMCT, acting to guide a
specific dissection approach during an invasive autopsy.
Turning to the substance of the report, one is mindful that this depends on vari-
ous factors. The first is the time available to produce the report, which is often
reflected in the length of the report data presented. This includes whether focal or
whole-body studies are reviewed, along with any additional series or angiography.
Then, there is the amount of detail provided. Overall, the quality of the report is not
to be measured by the word count but rather by the data presented and the content/
style of discussion. It is accepted that different radiologists will work at different
paces and in different ways, reflecting the individual reporter’s experience and con-
fidence. Therefore, the time taken to write a report, and its structure, will inevitably
be variable.
Broadly speaking, straightforward cases with catastrophic findings and a history

commensurate with the radiological views can be dealt with quite promptly by an
experienced reporter. Rapid reporting is possible, particularly when using struc-
tured reports or proformas, along with voice dictation software. These reports will
take considerably less time than complex cases with multiple comorbidities, which
often have competing pathophysiological events prior to death. Nevertheless, it is
reasonable to expect a PMCT report to be provided within 4–24 h of the scan being
completed. There may need to be some very fast progression of reporting, where
religious or family social issues prompt urgent case completion for funeral pur-
poses. One should be as helpful as possible in these cases but always within accept-
able professional standards.
As with all autopsy matters, unless authorised by the coroner, the person making
a post mortem radiological examination must not communicate the full, or part of
their report, to any person other than the coroner, their officers or the pathologist
[15]. This confidentiality is mandated by medicolegal realities, and inappropriate
release of information may be tested in court. The very basic level of confidentiality
is at least equivalent to clinical practice. Yet, it is also worth noting that the family
may later be given copies of the final post mortem report by the coroner or procura-
tor fiscal, which may include a copy or full transcript of the PMCT report.
It should be remembered that pathologists are required to specifically mention
negative findings when reporting a post mortem examination [16]. The same is
advised for the radiological report. Important negatives should be included to a
standard and structure agreed between radiologist and pathologist to aid in report
confidence and understanding. However, as with clinical imaging, wasting time
detailing multiple irrelevant findings whilst missing important data results in a
report that is often of little practical use. Any limitations of PMCT study, if these
apply, should always be made clear. A thorough and useful report is always required,
as the pathologist may cite the radiological report as reason for their conclusions
when formulating a cause of death. It is currently usual practice for the pathologist
to retain a central coordinating role in the establishment of the cause of death [1]
although in the future other arrangements may develop.
Now turning to the report construct itself, one appreciates that all reporters
(trainees and established specialists/consultants) will usually, or at least initially,
benefit from a defined report structure. This often uses a systematic approach to the
body compartments and organs, which should avoid missing important findings.
The use of a proforma template can be adapted and refined to local and/or personal
preference.
One may then choose to follow a free text style of structured reporting, or poten-
tially follow a sequential checklist of items, to be assessed as present or absent. This
latter approach is favoured in the forensic setting [17, 18]. Both are acceptable.
When assembling the data for the report, the following items are normally required
to achieve a full summary. Not all need to be present on every occasion, but (serving
as an example checklist) they are there to remind the reporter of the totality of the
examination:
The PMCT Report 29
• General introduction
–– Background information, from the coroner, and occasionally from the ambu-
lance service or police.
–– Past medical history, including medications of relevance to the case.
–– Whether resuscitation was attempted.
• External findings
–– Vascular lines, drainage tubes and implanted devices.
–– Ligatures or clothing affecting the imaging.
–– External trauma.
• Internal findings
–– Decomposition assessment, indicating if the quality of the study is not satisfactory.
–– Head (brain, cranial bones and local soft tissues).
–– Neck (soft tissues, cervical spine, airway, thyroid).
–– Chest (soft tissues, lungs, pleural tissues, mediastinum, heart, bones).
–– Coronary artery calcification, possibly augmented by the Agatston cal-
cium score.
–– CT coronary angiography findings if applicable.
–– Abdomen and pelvis (soft tissues, major organs, the bowel, the pelvic tis-
sues, bones).
–– Retroperitoneum (aorta, kidneys, soft tissues).
–– Musculoskeletal/limbs (soft tissues, bones).
• Clinicoradiological correlation
–– Free text description.
–– Cause of death (definite, probable, possible, unascertained or defined in stan-
dard format: 1a, 1b, 1c; and contributory factor/s 2)
–– Potential autopsy hazards such as fractures, foreign bodies and aberrant vessels.
As indicated at the start of the chapter, there is no single solution to reporting a

PMCT scan. However, for the benefit of the beginner, one example of a complete
report might be as given below:
Clinical Data
This 75-year-old male had a history of emphysema and lung basal scarring,
along with recurrent chest infections. There was a history of previous myocar-
dial infarction 2 years ago, and hyperlipidaemia. There had been a previous
hip replacement and history of osteoporosis. He was found deceased at home.
CPR attempts were unsuccessful.
External Findings
No pacemaker or other implanted device.
Minimal expected post mortem decomposition changes.
rain, Head and Neck

B
Normal ventricular system and extra-axial spaces. Diminished differentiation
between grey and white matter is a normal post mortem finding. No intracra-
nial haemorrhage, acute large vessel infarct or space-occupying lesion.
Background of periventricular small vessel ischemic changes.
Normal skull vault and cervical spine.
Normal soft tissues of the neck. Normal intra-orbital contents.
Thorax
No supraclavicular, axillary or mediastinal lymphadenopathy.
Heart enlarged with a cardiothoracic ratio of 0.63. No haemopericardium.
Significant coronary calcification (Agatston score total 2238). Marked mitral
valve annulus calcification. Normal sedimentation in the pulmonary arteries.
Normal calibre thoracic aorta.
Small volume bilateral pleural effusion. Background of centrilobular and
paraseptal emphysema. Symmetrical ground-glass opacities displaying a
dependent gradient in keeping with normal fluid hypostasis. No focal air-
space consolidation to suggest pneumonia.
Abdomen and Pelvis

No free fluid. No free gas. Normal post mortem appearances of the liver and
gallbladder. Normal pancreas. Normal spleen. Bilateral, uncomplicated cysts
up to 2.0 cm in both kidneys. Normal urinary bladder. Enlarged prostate mea-
sures 5 cm transverse.
Incidental right adrenal adenoma measuring 1.5 cm. Normal left adre-
nal gland.
Sigmoid diverticulosis with no acute feature. No focal bowel mass or
obstruction.
No abdominal or pelvic lymphadenopathy.
Incidental aneurysm of the abdominal aorta—4.0 cm. No retroperitoneal
haematoma.
Musculoskeletal
There is generalised osteopenia. Acute, bilateral, inner cortex rib fractures
are judged consistent with attempted cardiopulmonary resuscitation. Superior
endplate fractures of L3 and L4 appear old. There is generalised peripheral
vascular atherosclerosis.
Quality Assurance and Audit 31
Clinicoradiological Correlation
1. Extensive atherosclerosis of the coronary arteries (calcium score of 2238
equates to a high risk of a significant coronary artery stenosis). In the
absence of another demonstrated pathology, an acute cardiac event is con-
sidered a possible cause of death.
2. Background of paraseptal and centrilobular lung emphysema. No features
of (superimposed) chest infection.
3. No acute or suspicious intracranial/intra-abdominal findings.
[Note: Some PMCT centres provide a cause of death, with the radiologist stating
the cause of death as per medical certificate of cause of death (MCCD) format at this
point. Many pathologists prefer not to have such analysis provided in case there is
any disparity that could provoke confusion in a court setting.]
When coming to conclusions, the radiologist should always remain mindful that
many pathologies can be identified on the scans that might have no bearing upon the
cause of death. For example, there is a high prevalence of coronary artery calcifica-
tion seen on PMCT images. Coronary artery disease is a common reality (reflecting
the Western diet and other risk factors), but this does not mean that it is the cause of
death, unless there is other corroborative data and the clear absence of alternate
pathologies.
The authors would recommend using a common lexicon to define level of cer-
tainty in a diagnosis or the suggested cause of death, for example the term ‘proba-
ble’ being a certainty above 75% and ‘possible’ above 50% [19]. An agreed manner
of describing pathology to assist those instructing the autopsy and/or the pathologist
will also aid case analysis.
Nevertheless, despite high-quality imaging and experienced interpretation, in
some cases, and especially when there is a lack of clinical history, the cause of death
may remain unclear or, in medicolegal terminology, ‘unascertained’.
Quality Assurance and Audit
All services providing post mortem imaging data and diagnoses should be subject
to review of findings and audit [1] in the same manner as would be expected in a
clinical setting. Local arrangements may also include occasional double reporting
by radiologists, discussion of diagnoses and potential review of the radiology by the
pathologists and analysis correlation of PMCT results against the final cause of
death provided. These review policies will naturally benefit the family, medicolegal
representatives and society. Ultimately, multidisciplinary working and review can
be invaluable for discussion of findings, learning, continuous professional develop-

ment and decision-making. This is unfortunately not always practical or achievable
with the differing and complex working patterns of radiologists and pathologists,
particularly if working in separate locations. As with any aspect of clinical practice,
continuing professional development (CPD), in the form of literature, courses and
online resources, is important for appraisal and revalidation.
References
1. Maskell G, Wells M. RCR/RCPath statement on standards for medico-legal post-mortem
cross-sectional imaging in adults [Internet]. The Royal College of Radiologists and The Royal
College of Pathologists, London; 2012. https://www.rcr.ac.uk/system/files/publication/field_
publication_files/FINALDOCUMENT_PMImaging_Oct12.pdf.
[Internet]. Cham: Springer International Publishing; 2016. p. 362. http://link.springer.com/
10.1007/978-3-319-27022-7_13.
3. Chief Coroner. Guidance no. 1. The use of post-mortem imaging (adults) [Internet]. 2016.
https://www.judiciary.uk/wp-content/uploads/2013/09/guidance-no-1-use-of-port-mortem-
imaging.pdf.
4. Bedford PJ, Oesterhelweg L. Different conditions and strategies to utilize forensic radiology
in the cities of Melbourne, Australia and Berlin, Germany. Forensic Sci Med Pathol [Internet].
2013;9(3):321–6. http://link.springer.com/10.1007/s12024-013-9424-8.
5. Bedford PJ. Should pathologists be reporting forensic CT scans? Acad Forensic Pathol [Inter-
net]. 2012;2(2):198–201. http://journals.sagepub.com/doi/10.23907/2012.028.
6. Standards of radiographic practice for post-mortem cross-sectional imaging (PMC-SI) [Inter-
net]. The Society and College of Radiographers and the International Association of Forensic
Radiographers, London; 2015. https://www.sor.org/Learning-advice/Professional-body-guid-
ance-and-publications/Documents-and-publications/Policy-Guidance-Document-Library/
Standards-of-Radiographic-Practice-for-Post-Mortem.
7. The ionising radiations regulations 2017 no. 1075 [Internet]. http://www.legislation.gov.uk/
uksi/2017/1075/contents/made.
8. Ross SG, Bolliger SA, Ampanozi G, Oesterhelweg L, Thali MJ, Flach PM. Postmortem CT
angiography: capabilities and limitations in traumatic and natural causes of death. Radio-
Graphics [Internet]. 2014;34(3):830–46. http://pubs.rsna.org/doi/10.1148/rg.343115169.
9. Grabherr S, Heinemann A, Vogel H, Rutty G, Morgan B, Woźniak K, et al. Postmortem CT
angiography compared with autopsy: a forensic multicenter study. Radiology [Internet].
2018;288(1):270–6. http://pubs.rsna.org/doi/10.1148/radiol.2018170559.
10. Grabherr S, Grimm J, Dominguez A, Vanhaebost J, Mangin P. Advances in post-mortem CT-
angiography. Br J Radiol [Internet]. 2014;87(1036):20130488. http://www.birpublications.
org/doi/10.1259/bjr.20130488.
11. Rutty GN, Morgan B, Germerott T, Thali M, Athurs O. Ventilated post-mortem computed
tomography—A historical review. J Forensic Radiol Imaging [Internet]. 2016;4:35–42. https://
linkinghub.elsevier.com/retrieve/pii/S2212478016300028.
12. Baglivo M, Winklhofer S, Hatch GM, Ampanozi G, Thali MJ, Ruder TD. The rise of forensic
and post-mortem radiology—analysis of the literature between the year 2000 and 2011. J
pii/S2212478012000044.
References 33
13. Burton JL and Rutty G. The hospital autopsy [Internet]. 3rd ed. London: CRC Press; 2010.
https://www.routledge.com/The-Hospital-Autopsy-A-Manual-of-Fundamental-Autopsy-
Practice-Third-Edition/Burton-Rutty/p/book/9780340965146.
14. Suvarna SK, editor. Atlas of adult autopsy [Internet]. 1st ed. Cham: Springer International
Publishing; 2016. http://link.springer.com/10.1007/978-3-319-27022-7.
15. The coroners rules 1984 No. 552 PART III rule 10 [Internet]. http://www.legislation.gov.uk/
uksi/1984/552/article/10/made.
16. Leadbeatter S, Lucas S, Lowe J. Standards for coroners’ pathologists in post-mortem
examinations of deaths that appear not to be suspicious [Internet]. The Royal College of
Pathologists, London; 2014. https://www.rcpath.org/uploads/assets/1b02cfb9-000a-4b2f-
b6b80256b719a5ee/Standards-for-Coroners-pathologists-in-post-mortem-examinations-of-
deaths-that-appear-not-to-be-suspicious.pdf.
17. Schweitzer W, Bartsch C, Ruder TD, Thali MJ. Virtopsy approach: structured reporting ver-
sus free reporting for PMCT findings. J Forensic Radiol Imaging [Internet]. 2014;2(1):28–33.
https://linkinghub.elsevier.com/retrieve/pii/S2212478013001251.
18. Reporting forensic PMCT cases [Internet] (accessed 2019 Nov 22). https://virtopsy.com/
virtopsy-education-2-cas/.
19. Panicek DM, Hricak H. How sure are you, doctor? A standardized Lexicon to describe the
radiologist’s level of certainty. Am J Roentgenol [Internet]. 2016;207(1):2–3. http://www.
ajronline.org/doi/10.2214/AJR.15.15895.
Death, Post Mortem Changes
and Decomposition on Post Mortem 3
Computed Tomography
Introduction
This chapter sets out to consider ‘death’ and the terminology used in assessing bod-
ies after death. These are specified with the various pathology terms and features,
explained in chronological sequence to aid the appreciation of related radiology
changes. Factors that may increase the rate of body decay (decomposition) are pre-
sented along with some artefacts affecting bodies, which are not seen in the radiol-
ogy of the living.
Death and its Broad Causes
It is reasonable to first consider what is meant by ‘death’. One could look at it sim-
ply as the endpoint of life, but in most cases, death is said to have occurred after ces-
sation of cardiac and respiratory effort. Alternatively, in hospital settings, one might
regard death as having occurred after loss of higher brain function [1], or one might
define it as somatic and cellular death [2]. However, even if one addresses this mat-
ter from a scientific/medical standpoint, one should be aware that society, religion
and the relatives often have varying views on this matter [3]. These are beyond the
scope of this chapter, although various literary and online texts on the subject exist.
There are a variety of reasons why death may occur. Many are natural in type,
reflecting standard pathological deaths in the community and in hospitals. Indeed,
these cases will make up the majority of the post mortem computerised tomography
(PMCT) service workload. Put simply, most autopsy radiology is concerned with
natural death processes, such as ageing and complications of various metabolic and
structural diseases. Examples would include cardiovascular disease, chest infec-
tions, cancers and so on [4]. The radiologist should remember that many cases, at
the point of death, have had a period of attempted resuscitation, usually involving

Switzerland AG 2021
https://doi.org/10.1007/978-3-030-70829-0_3
36 3 Death, Post Mortem Changes and Decomposition on Post Mortem Computed…
chest compressions and ventilatory support—which may also have effects on the
ultimate PMCT appearances.
However, one must be aware that death can also follow an episode of intoxica-
tion, starvation, dehydration, suicide or trauma. Any and all of these may cause an
individual to be admitted to hospital or may cause death within a medically super-
vised background, rather than in the community. There are also deaths that are the
consequence of adverse nursing and/or medical (iatrogenic) interactions. These
may involve errors of diagnosis and treatment, aspects of neglect and potentially
negligence. Far less common are deaths caused by physical assaults, poisoning,
homicides and animal predation. Suspicious deaths are covered in forensic pathol-
ogy texts [2].
Society places great importance on the date and time of death, yet one is aware
that the physical body persists after this event. The body no longer remains stable,
as it did until the point of death, losing its various homeostatic biochemical and
physiological processes. After death these reactions stop, the cells start to autolyse
(break down) and the characteristic post mortem (decomposition) features start to
develop. In addition, there is microbial interaction enhancing tissue breakdown,
usually beginning in the gut (Fig. 3.1) and leading to putrefaction. Decomposition,
Fig. 3.1 Early-stage

decomposition shows as
green discolouration of the
abdomen near the caecum
Post Mortem Interval 37
the overall body process of autolysis, is a progressive (yet variable) process often
reflecting ambient temperature. Consequently, interpretation of post mortem imag-
ing requires an appreciation of the range of possible appearances of the decay-
ing body.
All of the broad causes of death and factors surrounding the deceased may have
bearing on how the body is seen on PMCT, with this being complicated by tissue
autolysis and breakdown of the body. An understanding and appreciation of normal/
expected post mortem changes is crucial to avoid erroneous interpretation of any
structural pathological changes [5], as it can be more difficult to define true pathol-
ogy in and among the ongoing decomposition realities.
Nevertheless, there is still a role for scanning even in cases of moderate and
advanced decomposition and putrefaction, as PMCT may reveal significant pathol-
ogy such as trauma, cerebral or other haemorrhage [6, 7]. If no catastrophic event is
demonstrated, a scan can also be complementary to invasive autopsy in demonstrat-
ing foreign bodies/devices and bony findings (e.g. fractures). Comparison of PMCT
with clinical imaging can also play a role in body identification [8], particularly
from dental assessment.
Post Mortem Interval
The date and time of death should ideally be provided in any supporting information
by the person/s requesting the PMCT scan. If death was not witnessed or recorded,
this may be an estimate of the likely time of death or just the time and date when the
body was discovered.
The time between death and another timepoint such as the scan is known as the
post mortem interval (PMI). It is sometimes a useful datum, as it allows some antici-
pation of the variable post mortem decomposition changes, although decomposition
is often more reflective of the local circumstances. Thus, a death in hospital, with
prompt refrigeration of the corpse, will have had a stable ambient temperature
applied in a sealed environment. The degree of post mortem autolysis will be differ-
ent for this body when compared to another body left at room temperature for sev-
eral days or when the body is in an exposed environment. One should also be aware
that it is possible for quite florid decomposition to be present even with a known
short PMI, as may be seen in cases that involve sepsis.
Determination of an unknown PMI by PMCT is a forensic application, beyond
the scope of the routine coronial workload and this text. It is a difficult task, reflect-
ing the numerous factors involved [9–12]. At present, there is general consensus that
there is no reliable and consistent imaging method for this task.
It is generally accepted that there are a variety of stages that follow death, mostly
in sequential pattern. These are dealt with in the following section.
Initial Changes Seen After Death
Pallor Mortis
This is the earliest change identified following cessation of breathing, circulation

and neurological function. Pallor mortis reflects the external skin colour change,
whereby the body appears to lose its normal colour density. This process reflects the
loss of blood circulating to the skin. Often the body will still be warm, flexible at the
joints and potentially may be mistaken for one still in life. One should always be
aware that shock, significant cardiac failure and neurological dysfunction could
mimic this state. However, pallor mortis is not an issue for PMCT, as cases are not
usually taken for scanning immediately after death—reflecting time for paperwork
issues, etc. It is plainly evident that pallor mortis cannot be detected by PMCT.
Algor Mortis
The next stage seen after death, with loss of normal catabolism, is the gradual cool-
ing of the body. This progressive decrease (depending on climate) in body tempera-
ture is known as algor mortis. Ultimately, the body will come into equilibrium with
the surrounding environment, whether this be temperate, polar or tropical. It also
cannot be defined by PMCT.
The difference in the body temperature and the ambient room/surrounding tem-
perature will determine how fast the body may cool and has been used to give a guide
to PMI (i.e. establishing the time of death). There is often a misconception that the
time of death can be accurately predicted from such body/vicinity temperature obser-
vations, but this is generally accepted as imprecise at best. Clearly, if a body has been
discovered long after death, then the body temperature calculations will not assist.
If the body has been actively refrigerated, at about 4 °C in a mortuary, then defin-
ing the PMI may be entirely unrealistic.
Livor Mortis or Hypostasis
Over time, the non-circulating blood and other fluids will settle with gravity towards
the dependent parts of the body. Soon after death the position of this livor mortis/
hypostasis can be affected by moving the body; however, at around 6–8 h after
death, it becomes fixed [13]. This may be of particular forensic interest in cases
where bodies have been moved following death.
The settling of blood is a readily determined feature (Fig. 3.2). It may help assess
the PMI, being a commonly visible early post mortem change [14]. It occurs
throughout the body, the various tissues appearing darker at open autopsy when in
more dependent positions (Fig. 3.3) with the difference in the colour of organs
reflecting how the body was stored [2].
Initial Changes Seen After Death 39
Fig. 3.2 Hypostasis is

seen externally with blood
settling to dependent parts
of the body
Fig. 3.3 Hypostasis

(congestion) is seen
internally with blood
pooling to the lower side of
the heart
On PMCT, hypostasis can also be readily appreciated. Assuming a supine posi-

tion for the body after death, this is seen as a dependent gradient of increasing
attenuation of tissues and organs from anterior to posterior and is usually well dem-
onstrated in the lungs (Fig. 3.4). When death has occurred and the body remains in
a lateral position, the gradient will also be lateral (Fig. 3.5).
The skin of dependent regions can display appreciable thickening and subcuta-
neous fluid accumulation (Figs. 3.6, 3.7, and 3.8). In larger, blood filled structures
such as the major vessels and cardiac chambers, there is separation of the blood
components (Figs. 3.9 and 3.10) resulting in ‘fluid–fluid’ levels on imaging [15].
Hypostasis within the vessels should not be mistaken for vascular wall dissection,
pathological thrombus or emboli.

chest on lung windows
shows normal hypostasis
of fluid (arrows) in the
lungs with horizontal
demarcation against
aerated lung

shows hypostasis of fluid
in the right lung, the body
was found lying in a right
lateral position

head on soft tissue
windows shows fluid
collecting in the posterior
scalp (arrow) with skin
thickening due to
hypostasis in the supine
position
Initial Changes Seen After Death 41
Fig. 3.7 Axial view of

both feet on soft tissue
windows shows skin
thickening and dependent
subcutaneous fluid
accumulation in both heels
due to hypostasis in the
supine position

both feet on soft tissue
windows shows dependent
hypostatic fluid in the right
lateral foot as the body was
found lying on the
right side
Rigor Mortis
In basic terms, rigor mortis is stiffening of muscle. The process involves the muscles
becoming fixed in position, due to calcium leakage from intracellular muscle cell
stores, causing fixed actin–myosin filament cross bridging. This stiffness remains,
as there is no oxidative metabolism to create adenosine triphosphate (ATP), which
is normally required for muscle relaxation. Yet the muscles do not remain perma-
nently fixed, as later decomposition enzymatic activity degrades the muscle fila-
ment binding complexes in the cells and eventually allows release.
Initially, however, in a body seen immediately after death, there is actually a
general muscle flaccidity, which may last for a few hours. One might broadly sum-
marise the process as taking around 12 h for general rigor mortis to establish, 12 h
to remain, and 12 h to disappear [13], although this timescale is subject to variation
due to environmental factors such as temperature [2]. In warm environments, the
onset of rigor mortis is earlier than for bodies in cold settings.

windows showing normal
post mortem layered
separation of blood
components in the large
vessels due to hypostasis
(arrows)

the chest on soft tissue
windows showing
separation of blood
products in the great
vessels and right atrium
with small volume of
normal post mortem air
rising to outline the right
coronary artery origin
It should also be understood that rigor mortis starts in different parts of the body
at different times such that the face and neck muscles are earliest affected, with the
torso and limbs following later. This same pattern is also seen with the release of
the rigor.
Rigor mortis can also be ‘broken’. This manipulation is commonly performed by
mortuary staff and undertakers [2] usually to aid body transport and storage if the
body has become rigid in an awkward position. This process involves firm stretch-
ing of the muscle, thereby achieving a normal joint alignment. It must be remem-
bered that excessive force may cause physical rupture of muscles or detachment
from their insertions and hence should not be performed by any person without
appropriate training. Apart from the obvious physical challenge of scanning a body
held in various irregular fixed positions, and the subsequent image reconstructions
required to interpret the study, the rigor itself does not affect PMCT appearances.
Later Changes in the Body After Death 43
Later Changes in the Body After Death
Decomposition is progressive, reflecting a combination of processes including

autolysis, putrefaction and occasionally animal predation. It is specific to the post
mortem state and is unlikely to have been encountered by a clinical radiologist who
normally deals with scans of the living.
Post mortem changes on PMCT seem to increase in a regular pattern over time
[16], although the timeframe itself and degree of change is highly variable.
Influencing factors may be external, such as the environment in which the deceased
expired (temperature, humidity, animal predation, trauma), and internal (body habi-
tus, microbial environment and sepsis).
Most bodies in the community will be discovered within a short time after death,
others will have death confirmed literally within a few moments of life having
ceased (e.g. hospital environments, palliative care settings). Decomposition may
have started, without there being significant anatomical change. Even for bodies
discovered after several days, significant tissue degradation may not always be evi-
dent, unless there is heat in the environment or where the body is exposed to nature.
Bodies that have been discovered after a significant period of time following
death may show more structural changes of a potentially confusing nature. The
changes involve the progressive tissue lysis that widely affects the body, often
driven by the body’s own bacterial components. These changes are broadly described
as decomposition, although significant tissue destruction, breakdown and loss are
commonly referred to as putrefaction. Putrefaction is a process of decomposition
caused by microbial activity and fermentation, resulting in gas and fluid production.
It is usually seen earliest in the right iliac fossa/lower abdomen (Fig. 3.1) and is
accompanied by gaseous tissue/organ distension. It eventually spreads throughout
the body, often via blood vessels in a process described as ‘marbling’ (Fig. 3.11).
A green/grey discolouration of skin can develop as decomposition progresses
through putrefaction (Fig. 3.12). There may be blistering with serous fluid (Fig. 3.13)
and then loosening of the skin and ‘slippage’, where skin comes away from the
body (Fig. 3.14).
Fluids may leak from any orifice (‘purging’) that can result in misleading histo-
ries of apparent vomiting or haemorrhage at the scene of discovery [2]. To reduce
purging, mortuary staff may insert gauze material into the upper aerodigestive tract
(Fig. 3.15). This could be a source of misinterpretation if not recognised.
Internal organs decompose naturally at different rates. The pancreas and adrenals
undergo relatively early autolysis (Fig. 3.16), and brain changes are also rapid in
onset. The heart, prostate and uterus are comparatively resistant to decomposition
[14]. The urinary bladder also may appear ‘normal’ for a considerable period of
time (Fig. 3.17).
With time, in cases with unimpeded decomposition, all the soft tissues will be
progressively degraded. This is particularly so if the body is in an open environ-
ment, where insects and indeed other animals may interact with, or remove body
tissues. This tissue loss will be a confounding issue for PMCT but does not auto-
matically imply that a cause of death cannot be defined. Ultimately, if left unchecked,
Fig. 3.11 ‘Marbling’ of

the skin showing how
bacterial decomposition
spreads along blood
vessels
Fig. 3.12 Advanced

putrefaction showing
pronounced tissue loss and
skeletonisation. Numerous
small holes are present on
the residual tissues
reflecting maggot damage
the consequence of autolytic and decomposition phenomena is that one may be left
with some skin, thick ligaments and skeleton only, even if the PMI is only a few
weeks (Fig. 3.12). Skeletonised bodies will be a particular problem for any PMCT
assessment, although the exclusion or discovery of non-accidental osseous injuries
may occasionally be valuable.
In some cases, decomposition may occur in a cool and dry environment. Here the
tissues may desiccate and become shrivelled, yet structurally may be preserved for
many years. These changes are often referred to as ‘mummification’ but are not akin
to the ancient Egyptian practice. Mummified cases are rarely seen in PMCT units.
Clearly, knowledge of extreme ambient temperatures at the scene of death is
important, as cases of hypothermia, with freezing of body parts, can result in areas
Later Changes in the Body After Death 45

the left thigh in soft tissue
soft tissue decomposition
gas (hence the general
‘blackness’ of the image),
dependent skin thickening
and skin blistering with
fluid (arrow)
Fig. 3.14 Skin slippage,

seen here on the forearm,
is commonly seen as
decomposition progresses
of low attenuation change on PMCT. In the brain, for example, this may appear
similar to infarction but is seen in a non-vascular territory and may be accompanied
by preservation of cerebral structure [17].
Refrigeration is thought to have little effect on PMCT, as the density of water
remains almost constant when it is between 0 °C and normal body temperature. In
contrast, freezing temperatures do affect CT attenuation values. It has been shown
ex vivo that the attenuation of frozen water reduces to around −70 to −80 HU [18].
The importance of this in PMCT has not been well researched, but general knowl-
edge of this effect should result in caution when interpreting radiological findings in
this unusual setting.

the face and neck on lung
windows showing gauze
material in the upper
aerodigestive tract (arrow)

the upper abdomen on soft
peri-pancreatic haziness
(arrows) in keeping with
autolysis, in a patient who
died from a myocardial
infarct
Finally, one should be aware that whilst whole-body or targeted angiography

(PMCTA) may generally be considered a viable and worthwhile technique, if
applied early after death [19, 20], its use will be less helpful with advancing decom-
position and precluded when vessels eventually lose their integrity.
Decomposition from the Radiology Perspective 47

the pelvis on soft tissue
windows shows a relatively
preserved urinary bladder
centrally among
surrounding tissues with
decomposition gas
Decomposition from the Radiology Perspective
The radiologist is not normally required to examine the body directly. This means that the
initial external clues of post mortem change and decomposition may be difficult to appre-
ciate, since the body is normally contained within an opaque, sealed body bag.
Nevertheless, the presence of gas within the body tissues, dissolution of brain paren-
chyma and other post mortem degradation effects may indicate that the body is not one
that has been secured shortly after death, or one that has not been stored appropriately.
Specific radiological findings that can be attributed to decomposition are also
discussed in the subsequent relevant chapters, but broadly include accumulation
of intravascular gas (Figs. 3.18, 3.19, and 3.20) and extravascular gas (Figs. 3.21,
3.22, 3.23, and 3.24), fluid settling or ‘hypostasis’ (Figs. 3.4, 3.25, 3.26), gas
and fluid accumulation in cavities (Fig. 3.27), soft tissue collapse (Fig. 3.28) and
eventually liquefaction of organs. Teeth may come loose if unsupported by soft
tissue (Fig. 3.29), and eventually either mummification or skeletonisation occurs
(Figs. 3.30, 3.31, and 3.32).
It is generally appreciated that tissue degradation may proceed variably, with
factors as discussed earlier but also relating to natural disease processes (e.g. sepsis)
and other metabolic realities (e.g. deaths after a pyrexial illness). Furthermore,
aspects of medical intervention, such as invasive or surgical interactions, cancer
treatments and immunosuppression, may alter the patterns of sepsis and cause
bodily changes. These are potential traps for the unwary radiologist and potentially
hinder diagnostic endeavour.
Any signs of decomposition should be described at PMCT. Decomposition can
significantly affect the sensitivity of the scan in demonstrating pathology and the
degree of certainty with which findings can be reported. Allowing for the general
degree of decomposition is of particular value in some circumstances, such as when
trying to differentiate air embolism or pneumothorax from decomposition-related
gas accumulation.

the brain on brain windows
shows intracranial,
intravascular gas (black)
due to normal
decomposition

decomposition gas in the
hepatic and portal veins in
the liver, mostly in the
anterior segments

the chest on lung windows
(accentuating the visibility
of gas) shows
heart chambers, great and
superficial vessels

extensive decomposition
gas in the liver and soft
tissues, note the liver is
barely perceptible
Fig. 3.22 Same case as

Fig. 3.21, lung windows
better demonstrate the
decomposition gas
allowing the liver to be
visualised
Fig. 3.23 Axial view of the upper thighs on lung windows shows normal decomposition gas in the
soft tissues and distending the scrotum
The radiologist may broadly consider whether the degree of decomposition is

appropriate for the PMI and given circumstances. The use of the terms such as
‘early’ or ‘late’ decomposition is rather subjective and may be misleading if not
clarified (does ‘early’ mean ‘minimal’ or ‘earlier than expected’?). For a non-
forensic study, it is more helpful to specifically describe the changes that are believed
to be attributable to decomposition and then indicate to what degree, if any, they
limit the examination. As expected, with minimal decomposition changes, positive
findings on the scan are usually trustworthy. When there are moderate changes,
more care should be taken in interpretation. Cases with advanced decomposition
changes may have very limited reliability in terms of diagnosing soft tissue pathol-
ogy (Figs. 3.33, 3.34, and 3.35).
Post Mortem Gas
As gas patterns are well demonstrated at PMCT, it is important to understand their

nature in the post mortem setting in order to use them as a diagnostic feature. There
is debate about the exact origin and cause of gas in the body of the deceased,
although gas production generally relates to putrefaction of tissues and starts to
develop in the first few days. During this process, gas in the vessels (Figs. 3.18,

the thoracolumbar spine on
bone windows shows
normal decomposition gas
(black) in the soft tissues,
vertebral canal and marrow

showing the typical post
mortem appearance of fluid
hypostasis in the lungs and
tiny bilateral pleural
effusions (arrows)
Fig. 3.26 Axial view of both thighs on soft tissue windows demonstrates prominent dependent
subcutaneous oedema (arrows) in an obese hospital patient with poor mobility

shows extensive soft tissue
gas, free gas and free fluid
(arrows) in the pleural
spaces due to progressive
decomposition

the orbits on soft tissue
windows shows flattened
globes (arrows)
Fig. 3.29 Axial view at

the level of the oropharynx
shows displaced teeth
(arrow) in the setting of
extensive surrounding soft
tissue decomposition

the head on bone windows
shows near-skeletonisation
of the cranium with
minimal scalp soft tissue
remaining

the pelvis on bone
windows shows a
persistent heavily calcified
aorta (arrow) in a
near-skeletonised pelvis

the pelvis on bone
windows shows a
dislocated right hip joint
due to loss of soft tissue
structural support in a
near-skeletonised body

the torso and upper legs on
lung windows shows
advanced decomposition
with extensive soft tissue
and free gas, the diagnostic
ability of this study is very
limited
3.19, and 3.20) is expected before free cavity gas such as pneumothorax (Fig. 3.36)
and pneumoperitoneum (Fig. 3.37). With advancing decomposition, gas tends to
appear at about the same time in the peritoneal and pleural cavities [21]. When free
cavity gas is seen independent of vascular gas (or judged to be out of proportion),
careful thought must be given as to whether this signifies a primary pathology.

shows advanced
decomposition with soft
tissue gas and
pneumothoraces, the
diagnostic ability of this
study is very limited

the lower pelvis on lung
windows shows advanced
decomposition with
extensive soft tissue and
bone marrow gas, soft
tissue assessment would be
very limited

shows bilateral anterior
pneumothoraces (arrows)
judged to be due to
decomposition as they are
in keeping with the degree
of vascular and soft tissue
gas seen elsewhere

the upper abdomen on lung
windows shows a
pneumoperitoneum and
falciform ligament sign
(arrow) judged to be due to
decomposition rather than
primary pathology, given
the degree of
decomposition gas seen in
the soft tissues
It should also be appreciated that the volume of decomposition gas can be gener-
ally, or focally, increased in the setting of sepsis, trauma, when lines have been
introduced into the body for medical purposes and/or when resuscitation attempts
have been made (see Chap. 11).
One method devised to quantify bodily decomposition changes is the ‘radiologi-
cal alteration index’ [21]. This objective process, considering the volume of gas at
different anatomic locations, is used to derive a numerical score. Cases with a high
score (more decomposition) need more questioning of the nature of the findings.
The use of such scoring lies outside our routine PMCT practice, and we tend to use
a descriptive assessment of any decomposition, with relevance given to the particu-
lar case being assessed.
Post Mortem Clot
Post mortem ‘clotting’ is also highly variable and may be linked to the length of
the dying process as well as the underlying pathologies that surround death. Longer
deaths tend to exhibit more intravascular clot than rapid deaths. Post mortem clotting
may appear prominent in the right heart, pulmonary trunk and great vessels [22],
and the large capacity veins may often have coagulum in the post mortem state. It is
difficult to reliably distinguish this normal post mortem clot from acute pulmonary
embolism (PE), which remains one of the limitations of PMCT (see Chap. 7).
The post mortem separation of blood components can generally result in either
this heterogeneous appearance of clots in the great vessels or a simple physical
layering of low upon high-density products (hypostasis, Figs. 3.9 and 3.10). When
Embalmed and Previously Autopsied Bodies 57
separation of blood products has occurred note that the whole volume is blood and
not just the dependent, more dense component.
Animal Predation and Post Mortem Changes
If death occurs outdoors or in an uncontrolled environment, particularly in warmer

months, there is a likelihood of animal predation. Examples in the United Kingdom
commonly reflect fly maggots, rats and foxes. This predation results in faster tissue
loss, and opening the body may accelerate decomposition [2]. This is seen espe-
cially in parts of the body that may be exposed, such as the face and hands, and in
open wounds or moist areas, such as the eyes, lips and nostrils (Figs. 3.12, 3.38,
3.39, 3.40, 3.41, and 3.42).
Maggot colonies can occasionally present as internal soft tissue masses [23], and
this should be considered a differential within the decomposed body. They appear
on the scan as discrete soft tissue opacities much the same size and shape as grains
of rice. They can add significant bulk to some tissues seen on PMCT and may even
appear to fill the upper respiratory tract and/or oral tissue cavities.
Embalmed and Previously Autopsied Bodies
Coronial PMCT will normally be undertaken prior to embalming, and therefore this
is not normally an issue to consider. However, occasionally, a body will be repatri-
ated after death has occurred abroad. When transporting a body for repatriation, it is
normal practice for the body to be embalmed. Whilst a primary death investigation

the skull (which is turned
90° to the right due to
patient positioning) on
lung windows shows
multiple maggots as small
discrete densities (example
at tip of arrow) both
external and internal to an
otherwise empty cranial
cavity

the level of the nasal cavity
on lung windows shows
multiple maggots in the
nasal cavities, maxillary
sinuses and deeper soft
tissues

in the setting of marked
decomposition shows
extensive soft tissue loss
due to maggot activity;
multiple maggots are seen
in the soft tissues and
externally within the body
bag (arrow)

windows shows a
mass-like maggot
infestation of the
abdominal soft tissues
through the chest wall
anteriorly

shows marked
decomposition and
evidence of predation by
the house cat leading to
focal right posterior chest
wall tissue loss (arrow)
is often conducted by the authorities in that country, a further investigation may be

instructed when the body is returned to the United Kingdom.
Embalming is a process of body preservation generally done through chemical
agents. This can reduce health (infection) hazards associated with moving the body.
Embalming is usually performed by the infusion of fluids through large cannulae in
the neck or groin, where gas tracks may then later be seen on PMCT (Fig. 3.43). It
has been suggested that the process itself may have a detrimental effect on the qual-
ity of subsequent PMCT images [24], although this is not always the case and an
individual case assessment will need to be made. If the embalming is suboptimal,
there may be further confusion with partially preserved and partially decomposed
tissues [25].
Some bodies are returned after an initial autopsy performed abroad (Figs. 3.44
and 3.45). These autopsies are of variable quality, may be partial or complete
Fig. 3.43 Axial view of the upper thighs on soft tissue windows shows focal gas in the anterior
superficial right groin (arrow) due to cannulation for embalming, with little evidence of decompo-
sition elsewhere. Note also the subtle dependent thickening of the skin due to hypostasis

the head and neck on bone
windows showing skull
cap fractures from autopsy
craniotomy, gauze in oral
cavity, absent brain and an
intracranial fluid level
(arrow—this appears
vertical as the study has
been reconstructed)
Fig. 3.45 Volume

rendered three-dimensional
PMCT image showing
craniotomy fractures
performed during open
autopsy

the abdomen on soft tissue
windows following open
autopsy shows mal-
positioning of a calcified
larynx in the abdomen
(arrow) and no discernible
normal abdominal viscera
invasive examinations, often with no accompanying information given back to the

UK authorities. Furthermore, the prior autopsy practitioners may have placed any or
all the tissues in a variably dissected fashion into a sealed bag placed into the abdo-
men (Fig. 3.46), and it is common to see gauze used to pack any resultant cavities
(Figs. 3.47, 3.48, and 3.49). Organs placed in a bag are not accessed by embalming
fluids, with resultant pronounced tissue degradation/autolysis. It is further recog-
nised that major organs of importance may be retained in the original mortuary.
Indeed, many pathologists would have seen cases where there are few or no internal
organs within the returned body!

the upper body windowed
to show absent cranial,
neck and thoracic
structures following
autopsy and the resultant
contiguous cavity packed
with extensive gauze
material (arrows)

the head on soft tissue
windows shows an
apparently ‘empty’ cranial
cavity following an open
autopsy
PMCT interpretation of such previously autopsied cases is often futile if one

wants to establish a medical cause of death. The role of PMCT is therefore dif-
ferent here. First, it can be used to exclude that a fraudulent autopsy has taken
place—a finding of only a stitched incision without underlying dissection of
organs [2]. Second, it can reveal the extent of the performed invasive autopsy,
for example, some parts of the body may not have been examined. Limb and
bony pathology should still be apparent, and the scan can be used to reveal any
retained implants or potentially dangerous foreign bodies (Figs. 3.50 and 3.51).
In such situations, PMCT is usually considered an adjunct to further autopsy
investigation.

readily reveal the
intracranial gauze packing
highlighting the need to
assess the body with
multiple window settings

the chest following autopsy
(note discontinuous
anterior thoracic wall and
midline incision)
windowed to highlight
linear radio-dense foreign
body in the left paraspinal
thoracic cavity (arrow)

Fig. 3.50, volume rendered
PMCT image demonstrates
the foreign body to be a
retained scalpel blade
therefore requiring a
cautious approach to any
subsequent open
examination
Reporting Post Mortem Change and Decomposition:

Pearls and Pitfalls
There is almost always evidence of post mortem change on routine PMCT,

and therefore common findings should be considered to be ‘normal for the
nature of the study’. Certainly, hypostasis, putrefactive gas and fluids should
not be mistaken for pathology.
As a general rule, the degree of decomposition should be similar through-
out the body. Asymmetrical or peripheral putrefactive changes (perhaps due
to body position near a heat source or partial exposure) can be normal, but the
reporter should assess this on an individual basis, correlating with the envi-
ronment in which the body was found if necessary.
When reporting the scan of a heavily decomposed body, there are limita-
tions as to the detail that can be given regarding the soft tissues, and this
should be clearly conveyed in the report.
Implanted medical devices can still easily be seen, and a quantification of
coronary artery calcification remains possible, even in cases of advanced
decomposition (see Chap. 8).
Bony findings such as fractures can be defined giving value to the study,
despite soft tissue limitations.
Example PMCT report phrases:
• The post mortem interval at time of scan is (if known).

• Appearances are consistent with normal post mortem change.
• Appearances are consistent with minimal/moderate/advanced
decomposition.
• Extensive vascular gas and a small pneumoperitoneum secondary to
decomposition.
• The presence of decomposition change severely limits the diagnostic capa-
bility of this study.
References
1. Schofield GM, Urch CE, Stebbing J, Giamas G. When does a human being die? QJM
[Internet]. 2015;108(8):605–9. https://academic.oup.com/qjmed/article-lookup/doi/10.1093/
qjmed/hcu239.
book/9780340972533.
3. Kastenbaum R. Death, society, and human experience. 11th ed. Abingdon: Routledge; 2011.
4. Benbow E. Ageing and death. In: Cross S, editor. Underwood’s pathology: a clinical
approach [Internet]. 7th ed. Elsevier; 2019. p. 219–28. https://www.elsevier.com/books/
underwoods-pathology-a-clinical-approach/cross/978-0-7020-7212-3.
References 67
5. Offiah CE, Dean J. Post-mortem CT and MRI: appropriate post-mortem imaging appear-
ances and changes related to cardiopulmonary resuscitation. Br J Radiol [Internet].
2016;89(1058):20150851. http://www.birpublications.org/doi/10.1259/bjr.20150851.
6. Takahashi Y, Sano R, Kominato Y, Takei H, Kobayashi S, Shimada T, et al. Usefulness of post-
mortem computed tomography for demonstrating cerebral hemorrhage in a brain too fragile
for macroscopic examination. J Forensic Radiol Imaging [Internet]. 2013;1(4):212–4. https://
linkinghub.elsevier.com/retrieve/pii/S221247801300083X.
7. Ruder TD, Zech W-D, Hatch GM, Ross S, Ampanozi G, Thali MJ, et al. Still frame from the
hour of death: acute intracerebral hemorrhage on post-mortem computed tomography in a
decomposed corpse. J Forensic Radiol Imaging [Internet]. 2013;1(2):73–6. https://linkinghub.
8. Hatch GM, Dedouit F, Christensen AM, Thali MJ, Ruder TD. RADid: a pictorial review
of radiologic identification using postmortem CT. J Forensic Radiol Imaging [Internet].
9. Takahashi N, Satou C, Higuchi T, Shiotani M, Maeda H, Hirose Y. Quantitative analysis
of intracranial hypostasis: comparison of early postmortem and antemortem CT findings.
AJR Am J Roentgenol [Internet]. 2010;195(6):W388–93. http://www.ncbi.nlm.nih.gov/
pubmed/21098169.
10. Jackowski C, Thali M, Aghayev E, Yen K, Sonnenschein M, Zwygart K, et al. Postmortem
imaging of blood and its characteristics using MSCT and MRI. Int J Legal Med [Internet].
11. Ishida M, Gonoi W, Hagiwara K, Takazawa Y, Akahane M, Fukayama M, et al. Hypostasis in
the heart and great vessels of non-traumatic in-hospital death cases on postmortem computed
tomography: relationship to antemortem blood tests. Leg Med [Internet]. 2011;13(6):280–5.
https://linkinghub.elsevier.com/retrieve/pii/S134462231100112X.
12. Ishikawa N, Nishida A, Miyamori D, Kubo T, Ikegaya H. Estimation of postmortem time
based on aorta narrowing in CT imaging. J Forensic Leg Med [Internet]. 2013;20(8):1075–7.
https://linkinghub.elsevier.com/retrieve/pii/S1752928X13002655.
13. Kori S. Time since death from rigor mortis: forensic prospective. J Forensic Sci Crim Investig
[Internet]. 2018;9(5). https://juniperpublishers.com/jfsci/JFSCI.MS.ID.555771.php.
14. Levy AD, Harcke HT. Essentials of forensic imaging [Internet]. Boca Raton: CRC Press; 2010.
https://www.taylorfrancis.com/books/9781420091120.
15. Ishida M, Gonoi W, Okuma H, Shirota G, Shintani Y, Abe H, et al. Common postmortem com-
puted tomography findings following atraumatic death: differentiation between normal post-
mortem changes and pathologic lesions. Korean J Radiol [Internet]. 2015;16(4):798. https://
www.kjronline.org/DOIx.php?id=10.3348/kjr.2015.16.4.798.
16. Klein WM, Kunz T, Hermans K, Bayat AR, Koopmanschap DHJLM. The common pat-
tern of postmortem changes on whole body CT scans. J Forensic Radiol Imaging [Internet].
2016;4:47–52. https://linkinghub.elsevier.com/retrieve/pii/S2212478015300289.
17. Kawasumi Y, Usui A, Ikeda T, Ishibashi T, Funayama M. Post-mortem computed tomography
findings of the frozen brain. J Forensic Radiol Imaging [Internet]. 2017;10:37–40. https://
18. Hyodoh H, Ogura K, Sugimoto M, Suzuki Y, Kanazawa A, Murakami R, et al. Frozen (iced)
effect on postmortem CT—experimental evaluation. J Forensic Radiol Imaging [Internet].
19. Franckenberg S, Flach PM, Gascho D, Thali MJ, Ross SG. Postmortem computed
tomography-angiography (PMCTA) in decomposed bodies—a feasibility study. J Forensic
S2212478015300265.
20. Ruder TD, Schulze K, Ross S, Ampanozi G, Gascho D, Laberke P, et al. Into the decom-
posed body—feasibility of post-mortem CT angiography in a decomposed cadaver. J Forensic
S2212478014000793.
21. Egger C, Vaucher P, Doenz F, Palmiere C, Mangin P, Grabherr S. Development and valida-
tion of a postmortem radiological alteration index: the RA-index. Int J Legal Med [Internet].
22. Ross SG, Bolliger SA, Ampanozi G, Oesterhelweg L, Thali MJ, Flach PM. Postmortem
CT angiography: capabilities and limitations in traumatic and natural causes of death.
Radiographics [Internet]. 2014;34(3):830–46. http://pubs.rsna.org/doi/10.1148/rg.343115169.
com/10.1007/978-3-319-27022-7_13.
24. Balta JY, Twomey M, Moloney F, O’Connor OJ, Murphy KP, Cronin M, et al. Assessing
radiological images of human cadavers: is there an effect of different embalming solutions?
J Forensic Radiol Imaging [Internet]. 2017;11:40–6. https://linkinghub.elsevier.com/retrieve/
pii/S2212478017300564.
25. Williams EJ, Davison A. Autopsy findings in bodies repatriated to the UK. Med Sci Law
[Internet]. 2014;54(3):139–50. http://journals.sagepub.com/doi/10.1177/0025802413499325.
External Findings, Tubes and Devices
on Post Mortem Computed Tomography 4
Introduction
The external examination of a body forms an essential part of the overall post mor-
tem investigation. It provides the opportunity to spot features that point to underly-
ing pathology, alongside allowing one to consider the possibilities of unnatural
death. The latter could stop any routine open autopsy and prompt further specialist
investigation or forensic involvement.
Standard diagnostic imaging has a limited role in the examination of the external
surface of a body. Indeed, post mortem CT (PMCT) cannot depict skin colour, super-
ficial bruising, tattoos, scars or abrasions to name but a few findings. Yet, these could
be potentially significant. Therefore, both a review of external findings on the PMCT
scan/report and a visual external inspection are necessary tasks for the pathologist.
In addition, usually following deaths in hospital or where cardiopulmonary
resuscitation has been attempted, various tubes and devices may be encountered on
post mortem scans. Examples are given and their importance in relation to the cause
of death is discussed. Some devices also have relevance to handling of the body,
along with issues in relation to cremation.
The External Examination: The Pathologist’s Perspective
An external examination occurs prior to evisceration for open autopsies and is also
required in those cases that will be certified without invasive studies. For routine
coronial cases, this is usually performed by a pathologist. A radiologist would not
usually externally examine the physical body.
The external examination provides the pathologist with information about likely
internal disease. It should also be remembered that personal adornments such as
piercings and rings may be part of case identification. The presence of jaundice,
petechial haemorrhages or peripheral oedema may indicate liver failure, infections/

Switzerland AG 2021
https://doi.org/10.1007/978-3-030-70829-0_4
70 4 External Findings, Tubes and Devices on Post Mortem Computed Tomography
terminal hypoxia and cardiac dysfunction. Natural disease can also be indicated by
clubbing, tobacco tar staining, joint deformities and drug patches. Accelerated mar-
bling of the skin (whereby the pattern of underlying vessels is clearly visible, see
Chap. 3) may suggest sepsis as would also be supported by diffuse petechial
haemorrhages.
Vascular lines, drainage tubes, airways, feeding tubes and other devices can indi-
cate medical intervention alongside review of medical tattoos. The position of these
items should be confirmed as correct and should also be mapped against the back-
ground history. Furthermore, the identification of scars (fresh/healed) should sup-
port the medical history provided.
Livor mortis will point to the position of the body after death. Non-natural
pathology may be indicated by superficial self-inflicted injuries, crease marks
around the neck from ligatures, superficial skin ulcers and sinuses reflecting intra-
venous drug use and unusual bruises in cases of abuse. The degree of body nutrition
(undernourished/emaciated, through to obesity) may be relevant in terms of assess-
ment, particularly in terms of evaluation of care before death.
Criminal act pathology may be indicated by incised wounds, abrasions and pen-
etrating injuries along with the pattern of bruising around vital structures, such as
the neck. Rarely, skin colour can be a clue as to the pathology—such as pink disco-
louration in carbon monoxide poisoning, grey discolouration in methemoglobin-
emia and general pallor in association with pronounced blood loss. Examples and
full discussion on the value of the external examination can be found in autopsy
pathology texts [1].
External Findings on PMCT
Ideally, visual external examination should be performed in advance of PMCT in

order to better inform the imaging interpretation. This is not always practical, due to
mortuary workflow realities, and will need to be locally negotiated.
When performed prior to the scan, the results of the external examination, tubes
removed and sites of any toxicological sampling (Figs. 4.1 and 4.2 and see Chap. 2)
should be indicated to the radiologist. Any/all remaining devices should be noted in
the PMCT report such that the information can be correlated at a later date if
necessary.
To review the external body on PMCT, the scout views may be reviewed, stan-
dard axial images can be ‘windowed’ to better demonstrate external items (Figs. 4.3
and 4.4) or volume-rendered surface reformats can be produced (Figs. 4.5 and 4.6).
Some forensic centres use three-dimensional surface scanning or photogrammetry
to obtain a detailed picture of the body surface to aid forensic reconstructions, for
example to correlate injury with a suspected instrument [2]. These data may also be
of value in public court settings, rather than using actual photographs of the body.
External Findings on PMCT 71
Fig. 4.1 Axial view of the upper thighs on soft tissue windows shows left groin soft tissue gas
track (arrow) following toxicological blood sampling from the femoral vessels

the pelvis on lung windows
to highlight suprapubic gas
track (arrow) from urine
sampling
In the non-forensic setting, common external findings of note on PMCT include

the presence of medical equipment from attempted cardiopulmonary resuscitation
(Figs. 4.5, 4.7, 4.8, 4.9, and 4.10, see also later in this chapter and Chap. 11), nutri-
tional support (Fig. 4.11), traumatic injuries (Figs. 4.12 and 4.13) or, in cases of
suicidal hanging, the ligatures (Fig. 4.14, see also Chap. 6).

pelvis on soft tissue
windows shows streak
artefact (arrows) from a
dense metallic object on
the left forearm
Fig. 4.4 Same image as

Fig. 4.3 on bone windows
reveals the object to be a
wristwatch
Particularly in community sudden death cases, items of clothing, jewellery or

other objects can be seen to remain with the body (Figs. 4.15, 4.16, 4.17, 4.18, and
4.19). They may occasionally be valuable in terms of patient identification but may
cause artefacts that merit consideration. With agreement from the Coroner’s office,
these items may need to be removed for the scan if they render aspects of the study
non-diagnostic.
It has been suggested that after a detailed external examination of the physical
body, the PMCT can be used to ‘triage’ the need to proceed to open autopsy. If the
circumstances, clinical information, findings on visual inspection and PMCT pro-
vide enough information to establish a cause of death on the ‘balance of probabili-
ties’, then one should be able to derive a cause of death formulation and release the
body without further investigation [3].
Tubes Seen on PMCT
General Comments
After deaths in hospital, medical care or following resuscitation attempts, various

‘tubes’ may be left in place. These include airway adjuncts, central venous or arte-
rial lines, urinary catheters, nasogastric tubes and cavity drains (Figs. 4.20 and 4.21).
Tubes Seen on PMCT 73
Fig. 4.5 Volume-rendered

PMCT image showing
defibrillator pads in place
on the anterior chest wall
One benefit of PMCT is the ability to assess the body ‘as it is’ and observe poten-
tial complications such as tube misplacement (Fig. 4.22), underlying tissue trauma,
pneumothorax or air embolism [4] prior to destructive dissection. PMCT in poly-
trauma patients can identify a badly positioned tube or line to provide the medical
team with feedback and improve training [5]. Such poor placements may also be
debated as pertinent to the cause of death, for example an airway incorrectly posi-
tioned in the oesophagus.
The radiologist should however keep in mind that tube positions may have
altered after death, such as during body handling or transport. Whilst it is impor-
tant to correctly identify any malposition of a tube or other device, it should not
be assumed that this is relevant to the cause of death without other supportive
data [4].
Fig. 4.6 Volume-rendered

image of the lower legs
showing traumatic soft
tissue injuries and left leg
deformity, the defibrillator
pad wires are also visible

windows shows a
hyperdense right anterior
chest wall defibrillator pad
(arrow)

Fig. 4.6, axial view of the
upper abdomen shows
hyperdense left lateral
(arrow), note the
underlying traumatic soft
tissue and rib cage shape
deformity

lower neck on soft tissue
windows shows an
endotracheal tube (arrow)
traversing the larynx

windows shows an
endotracheal tube entering
the right main bronchus
(arrow). This may have
been pushed further down
post mortem. Also shown
are a defibrillator pad over
the right anterior chest
wall, left anterolateral rib
fracture and extensive lung
changes

windows shows an anterior
abdominal wall
percutaneous gastrostomy
(arrow) with external
flange and balloon
correctly inflated in the
stomach lumen

the left shoulder on bone
windows showing a
comminuted open fracture
of the left humerus with
skin and soft tissue
disruption (arrow)
following a road traffic
collision suicide
Fig. 4.13 Axial view

through the chest on lung
windows shows rib
fractures causing thoracic
wall distortion (arrow) and
a pneumothorax, judged to
be due to CPR chest
compressions

the neck on lung windows
shows a rope ligature
(arrow) and thin folds of
the body bag in a case of
suicide by hanging
Fig. 4.15 Axial view of both feet on bone windows shows that the body has been imaged wearing
clothes and steel toe cap shoes following death at the workplace

the left lower leg on soft
tissue windows showing an
electronic monitoring tag
which is causing some
artefact at this level

Fig. 4.16, the electronic
monitoring tag itself is
better visualised on bone
windows and the
underlying bones can be
assessed
Fig. 4.18 Axial view at the level of the knees on bone windows shows that the body has been
scanned with wallet and watch (arrows) between the legs in the body bag
Fig. 4.19 Axial view at the level of the upper thighs on soft tissue windows, scanned with ‘arms
by the sides’ shows streak artefact from multiple metallic rings on the fingers of both hands

the lower chest/upper
abdomen on soft tissue
windows shows two large
right-sided intercostal
chest drains (arrows)
inserted to treat a known
empyema. In addition,
there is a hyperdense
irregular clot in keeping
with haemothorax

same patient in Fig. 4.20
demonstrates apical (down
arrow) and basal (up
arrow) placements of the
chest drains. There is also
an endotracheal tube and
left internal jugular vein
catheter, as death occurred
in hospital
Airway Adjuncts
A wide variety of airway adjuncts exist, including oropharyngeal and nasopharyn-

geal airways, supraglottic airways, tracheostomies and endotracheal (ET) tubes
(Figs. 4.23, 4.24, 4.25, 4.26, and 4.27). The type and position of the airway adjunct
at the time of scanning should be factually documented. The reporter should be
aware of spurious foreign bodies such as dentures dislodged during resuscitation
attempts (Fig. 4.28).
ET tubes enter the trachea via the mouth and larynx. After death, they are com-
monly pushed further inwards by nursing or mortuary staff, so that the face of the
deceased can be viewed with the mouth closed. This procedure causes the tube to

windows showing a
suprapubic catheter
entering the lower anterior
abdominal wall (horizontal
arrow) but which has been
misplaced with the balloon
(vertical arrow) inflated in
the urethra. Note extensive
faecal loading and partly
visualised peripherally
calcified abdominal aortic
aneurysm

the oral cavity on bone
windows showing an
oropharyngeal airway
(arrow) and also a partially
visualised C2
anteroinferior corner
fracture

the neck on soft tissue
windows showing a
supraglottic laryngeal
airway device and widely
patent laryngeal inlet
(arrow)

the neck windowed to
show laryngectomy and
tracheostomy (arrow)

the level of the lung apices
windowed to show an
endotracheal tube (arrow
pointing to the
radiographic marker) with
inflated cuff distending the
tracheal wall. In addition,
there is a left
pneumothorax and a right

Fig. 4.26, coronal view
again shows endotracheal
tube with inflated cuff and
the left pneumothorax
move distally into the central airways, usually with the tip in the right main bron-
chus (Fig. 4.29). This should therefore not be automatically assumed as misplace-
ment. If, however, the ET tube is seen in the oesophagus, it is unlikely to have been
the result of any post mortem movement. Failure of correct endobronchial intuba-
tion may go unnoticed in the pre-hospital setting with oesophageal intubation being
unrecognised [4].
Supraglottic airways are increasingly used in emergency settings owing to their
relatively quick and easy placement. This type of airway also enters the mouth and
usually has an elliptical, inflatable, or malleable plastic cuff that should sit in the
hypopharynx (Figs. 4.30 and 4.31).
Vascular Access
Peripheral vascular cannulae can be difficult to identify on PMCT due to their small
calibre but are unlikely to be of any significance. Potential local complications

the upper aerodigestive
tract on soft tissue
windows shows an
‘omega’ shaped foreign
body (arrow) consistent
with dislodged dentures

windows showing
misplaced endotracheal
tube tip in the right main
bronchus (arrow), fluid in
the tube lumen is a
common post mortem
finding
include extravasation including that of iodinated contrast, peripherally injected for

clinical CT (Fig. 4.32).
Central venous access catheters are more important as they indicate significant
medical intervention and are more complex to place. Complications include haema-
toma around the catheter track (Fig. 4.33) although some blood is often seen around
central lines. The tip of a central venous catheter should lie centrally in the lumen of
the vessel, usually the superior vena cava. It can unfortunately be difficult to pre-
cisely assess the position of catheter (or wire) tips with respect to lumen or wall as
post mortem vessels are often collapsed, but a lack of perivascular stranding or
haematoma argues against vessel rupture.

the level of the
supraglottis, windowed to
show a supraglottic airway
device cuff (arrow)

Fig. 4.30, sagittal view of
the neck shows the
supraglottic airway device
and a patent upper airway

the chest and arms at the
level of the heart, on soft
tissue windows, shows
dense, streaky iodinated
contrast extravasation in
the right upper arm and
chest wall (arrows)

windows shows a right
internal jugular vein
haematoma (large arrow)
with loss of clear fat planes
and mass effect deviating
the larynx away, following
vascular catheter
placement (small arrow).
Note also an endotracheal
tube in the upper airway

the right tibia on bone
windows shows an
intra-osseous needle with
tip correctly positioned in
the marrow cavity
Intraosseous Needles
Intraosseous (IO) needles may be used during resuscitation attempts when vascular
access is insufficient or has failed. They are commonly inserted into the proximal
tibia (Fig. 4.34), as well as the proximal humerus (Fig. 4.35) but also distal femur,
distal tibia and sternum. Due to the cross-sectional nature of CT, needle position can
be easily assessed. The tip should lie within the medullary cavity of the bone, not in

the level of the gleno-
humeral joints on bone
windows showing a right
proximal humeral
tip in the marrow cavity

the right upper tibia on
bone windows shows an
tip into the distal cortex;
this needle was
non-flushing
the cortex (Fig. 4.36), and the shaft should not be bent (Fig. 4.37). When entirely
misplaced (Fig. 4.38) there can be extravasation into surrounding tissues and, of
course, a limited or non-therapeutic result. Venous air embolism has also been asso-
ciated although the exact mechanism is undetermined [4]. When removed prior to
the scan, these needles can leave an intra-osseous gas track (Fig. 4.39) and should
not be mistaken for fractures.

the left upper tibia on bone
windows shows an
bent metal shaft; this
needle was non-flushing

the right upper tibia on
bone windows shows a
misplaced intra-osseous
needle which does not
penetrate the bone cortex
Devices Seen on PMCT
Implantable Cardiac and Other Electrical Devices
It is important to identify these battery-operated internal devices which include

implantable loop recorders, pacemakers, re-synchronisation therapy devices and
cardiac defibrillators (Figs. 4.40, 4.41, 4.42, 4.43, and 4.44) as they must be removed
prior to cremation, due to their risk of exploding [6].
Devices Seen on PMCT 89

the left upper tibia on soft
intra-osseous gas track
(arrow) from a removed
intra-osseous needle

windows showing an
implantable loop recorder
device in the subcutaneous
left anterior chest wall
(arrow)

the upper chest on bone
windows showing a
pacemaker unit in the left
anterior chest wall (large
arrow) and two leads/
metallic wires in the left
subclavian vein (small
arrow)

the chest on bone windows
showing the conventional
position of a pacemaker
device in the left anterior
chest wall

showing an implantable
cardiac defibrillator and
pacemaker combination
device (larger than
pacemaker only) with
thicker lead containing the
‘shock coil’ at the level of
the superior vena cava
(arrow)
Implantable defibrillators also require deactivation prior to removal in order to

avoid risk of injury to anyone handling the body. Such devices are used to prevent
and treat cardiac arrhythmias, which might otherwise lead to sudden cardiac death.
Conversely, there is a risk of discharge and inducing a fatal arrhythmia to mortuary
staff when removing a device after death [7, 8]. Some of these devices record data
that may be retrieved for evaluation of potential arrhythmias or even device mal-
function, if this is in question.
On PMCT, the wires of such devices may cause localised streak artefact
(Fig. 4.45). This can hamper assessment of the heart, in particular accurate coronary
calcium scoring and can also make luminal assessment more difficult on a CT coro-
nary angiogram.
Other electrical devices one may encounter include vagal or sacral nerve root
stimulators, baclofen or other battery-operated pumps (Figs. 4.46 and 4.47). These
should all be mentioned as they mostly require removal prior to cremation.

Fig. 4.43, para-sagittal
view of the chest showing
the superficial placement
of the implantable cardiac
defibrillator/pacemaker
device and its multiple
leads

windows shows that there
is considerable streak
artefact from implantable
cardiac defibrillator wires.
This can make the
coronary calcium score
difficult to measure

the pelvis on bone
windows shows a right
lower anterior abdominal
wall placement of a
baclofen pump, the tubing
usually exits laterally and
traverses the subcutaneous
plane before entering the
vertebral canal (Fig. 4.47)

Fig. 4.46, a sagittal view of
the thoracolumbar spine on
soft tissue windows shows
the baclofen pump tubing
ascending the vertebral
canal (arrows)
Non-electrical Implants
Beyond those mentioned, there are a myriad of other implanted devices, including
stents, grafts and prostheses, some more unusual than others (Fig. 4.48) that should
be mentioned in the report. They may be relevant in indicating underlying disease
Fig. 4.48 Axial view at the level of the lower legs on soft tissue windows showing bilateral dense
surgical calf implants

windows shows the
radio-opaque markers
(arrows) of a collapsed
intra-aortic balloon pump
or recent illness (Figs. 4.49, 4.50, and 4.51). Others require removal or specialist
handling. They may have item numbers to permit patient identification—in cases of
decomposed bodies.

showing a transcatheter
aortic valve implantation/
TAVI (arrow)

windows shows a metallic
aortic valve (arrow). Note
also a healed surgical
sternal fracture and
moderate bilateral pleural
effusions
Whilst metallic joint prostheses and fracture fixations (Fig. 4.52) are usually of
no consequence, an uncommon type of intramedullary nail (Fixion®) has been
reported to have a risk of explosion during cremation due to an internal saline com-
ponent over-expanding [9]; we have not however come across this device. It is
expected that not all types of specialised implant will be familiar to every radiolo-
gist, but the reporter should mention any and all implanted devices as a matter of
routine and seek clarification from medical notes if there are queries or concerns as
to the nature or function of the device.

the left proximal femur on
bone windows showing a
dynamic hip screw and
remodelled, healed
intertrochanteric fracture
Finally, if there are implanted radioactive seeds (such as used in the prostate or
breast for the treatment of cancer), these should also be specifically mentioned. The
radiation risk is low to those working in the PMCT suite, owing to the locally acting
nature of such implants, however, appropriate radiation protective wear may be
advised. For high-radiation devices one may need to delay scanning or open autopsy.
Reporting External Findings and Devices: Pearls and Pitfalls
Make note in the report if you have performed or received details of any exter-
nal examination including knowledge of sites of toxicological sampling.
Review any scout imaging and window standard images to demonstrate
any external or superficial items of interest.
It is suggested that the presence of any/all implanted tubes and devices is
mentioned in each report as a matter of routine.
State clearly when there is an absence of such devices, particularly those
that may preclude cremation or if they were indicated as present in the clini-
cal record.
Of note, the following list of implants merit consideration before crema-
tion is permitted, as they are considered potentially dangerous [10]:
• Pacemakers
• Implantable Cardioverter Defibrillators (ICDs)
• Cardiac resynchronization therapy devices (CRTDs)
• Implantable loop recorders
• Ventricular assist devices (VADs)
• Implantable drug pumps including intrathecal pumps
• Neurostimulators (including for pain & Functional Electrical Stimulation)
• Bone growth stimulators
• Hydrocephalus programmable shunts
• Fixion® nails
• Any other battery powered or pressurised implant
• Radioactive implants
• Radiopharmaceutical treatment (via injection)
Finally, note any finding that seems out of the ordinary even, if the sig-
nificance is unknown, as relevance may potentially become apparent later.
• I have not performed an external examination of the body or received

details of such an examination.
• An endotracheal tube is present with its tip in the mid oesophagus. This is
unlikely to have migrated in the post mortem setting.
• An intra-osseous needle is placed in the proximal right tibia and the tip is
correctly sited in the medullary cavity.
• There is a pacemaker/other device in place at the time of PMCT.
• There is no implantable cardiac or other battery-powered device in place.
• A soft tissue gas track in the groin suggests there has been prior aspiration
for toxicology, please correlate with mortuary records.
References 97
References
1. Lloyd KL, Suvarna SK. External examination. In: Suvarna SK, editor. Atlas of adult autopsy
[Internet]. Cham: Springer International Publishing; 2016. p. 13–45. http://link.springer.
com/10.1007/978-3-319-27022-7_2.
2. Grabherr S, Egger C, Vilarino R, Campana L, Jotterand M, Dedouit F. Modern post-mortem
imaging: an update on recent developments. Forensic Sci Res [Internet]. 2017;2(2):52–64.
https://www.tandfonline.com/doi/full/10.1080/20961790.2017.1330738.
3. Chatzaraki V, Heimer J, Thali M, Dally A, Schweitzer W. Role of PMCT as a triage tool between
external inspection and full autopsy—case series and review. J Forensic Radiol Imaging
[Internet]. 2018;15:26–38. https://linkinghub.elsevier.com/retrieve/pii/S2212478018300601.
4. Bolster F, Ali Z, Fowler D, Daly B. Imaging of resuscitation and emergency resuscitation
devices—Lessons learned from post mortem computed tomography. J Forensic Radiol Imaging
5. Lotan E, Portnoy O, Konen E, Simon D, Guranda L. The role of early postmortem CT in
the evaluation of support-line misplacement in patients with severe trauma. Am J Roentgenol
[Internet]. 2015;204(1):3–7. http://www.ajronline.org/doi/10.2214/AJR.14.12796.
6. Johnson C, Lowe J, Osborn M. Guidance for pathologists conducting post-mortem examina-
tions on individuals with implanted electronic medical devices [Internet]. The Royal College
of Pathologists, London; 2015. https://www.rcpath.org/uploads/assets/4f04f871-257e-446b-
b94f38095defaf0d/guidance-for-pathologists-conducting-post-mortem-examinations-on-indi-
viduals-with-implanted-electronic-and-medical-devices.pdf.
7. Ackerman MJ, Giudicessi JR. Post-mortem cardiovascular implantable electronic device
interrogation: Clinical indications and potential benefits. J Am Coll Cardiol [Internet].
8. Mitchell LB, Pineda EA, Titus JL, Bartosch PM, Benditt DG. Sudden death in patients with
implantable cardioverter defibrillators: The importance of post-shock electromechanical dis-
sociation. J Am Coll Cardiol [Internet]. 2002;39(8):1323–8. https://linkinghub.elsevier.com/
retrieve/pii/S0735109702017849.
9. Phillips AW, Patel AD, Donell ST. Explosion of Fixion® humeral nail during cremation: novel
“complication” with a novel implant. Inj Extra [Internet]. 2006;37(10):357–8. https://linking-
hub.elsevier.com/retrieve/pii/S1572346106000365.
10. Ministry of Justice. The cremation (England and Wales) regulations 2008. Guidance to appli-
cants. [Internet]. 2018. https://www.cremation.org.uk/content/files/2018-guidance-to-appli-
cants.pdf.
Post Mortem Computed Tomography
of the Brain and Spinal Cord 5
Introduction
As with cranial computed tomography (CT) in the living, non-contrast–enhanced

post mortem computed tomography (PMCT) can readily identify significant haem-
orrhage, mass effect, hydrocephalus, large vessel territory infarction and, addition-
ally, fatal trauma. Cranial PMCT may therefore permit a cause of death solution
without open autopsy being needed.
One should also be mindful of further benefit in cases where (more commonly)
no cranial pathology is demonstrated. In these scenarios, opening the head at inva-
sive autopsy can usually be avoided [1]. This advantage focuses the pathologist onto
areas likely to be relevant to the cause of death, such as the heart and lungs (see
Chaps. 7 and 8).
However, before reporting cranial PMCT, it is imperative to become familiar
with normal post mortem findings in order not to mistake decomposition changes
for genuine pathology. PMCT of the brain can generally be considered to be reli-
able, although its value may fall quickly in cases with significant tissue
decomposition.
utopsy of the Brain and Spinal Cord: The Pathologist’s

A
Perspective
The post mortem examination of the brain and spinal cord is technically difficult,
since these soft structures are very well protected by cranial and vertebral bone [2].
Such examination is potentially disfiguring to the body and thereby often a particu-
lar concern for relatives of the deceased. Indeed, many pathologists and Coroners
prefer not to engage with head and nervous system tissues unless there is a good
reason requiring open access.

Switzerland AG 2021
https://doi.org/10.1007/978-3-030-70829-0_5
100 5 Post Mortem Computed Tomography of the Brain and Spinal Cord
Should the brain need to be removed, then a coronal slice through the scalp tis-
sues down to the skull is accomplished from behind the ear on both sides, with the
scalp and deeper soft tissues being reflected anteriorly and posteriorly. The calvar-
ium of the skull is removed using a circumferential saw cut, including a step in the
cut, usually made in order to facilitate reassembly of the skull and reconstitution of
the head tissues following examination. In this manner, the top of the skull is
removed, allowing direct inspection of the dura, leptomeninges and underlying
brain with the cranial nerves, tentorium cerebelli and vasculature being transected
to facilitate extraction. The pituitary can be accessed at this juncture if desired.
Once the brain has been removed and weighed, there are two possibilities for the
examination. First, one could progress through direct inspection of the meninges,
removal of the brainstem and cerebellum at the mid-brain level and a section of the
cerebellum through the peduncles, thereby exposing the fourth ventricle. Serially
slicing through the mid-brain, pons and medulla in transverse fashion allows good
inspection of the brainstem tissues. The cerebellum is traditionally cut centrally in a
sagittal fashion through the vermis, with oblique sections taken to expose the den-
tate nucleus and cortical grey ribbon tissues.
The cerebral hemispheres are examined by coronal slices, commencing anteri-
orly and generally passing progressively backwards in 1 cm steps. This exposes the
outer grey cortical ribbon, the inner white matter, the deep nuclei, the ventricles and
choroid plexus tissues. It also facilitates identification of the pineal gland.
If one is interested in the arterial vasculature, then this is usually resected in one
piece from the under surface of the brain before the brain tissue slicing. This is
generally reserved for complex vascular malformations and confirmation of
thrombosis.
It has to be recognised that examination of the unfixed brain is complex because
the brain parenchyma is so soft. Block sampling is complicated, as some lesions
may be difficult to identify in the unfixed state.
The second and alternative solution to brain tissue examination is to suspend the
intact brain in a large bucket of formalin for approximately 6 weeks, allowing the
tissues to fix and harden. This is particularly useful for complex neuropathology
cases, as it allows smaller step sections to be taken through the brain tissues. Clearly,
retaining the brain at the end of an autopsy for a period of time may be less accept-
able to relatives compared to examination of the fresh tissues, with this having to be
balanced against the need to acquire good histology.
The examination of the spinal cord can be accomplished in two ways. The first is
to turn the body into a prone position and make a longitudinal slice from the occiput
down towards the sacrum with dissection of underlying soft tissue and musculature
down to the posterior bony components of the vertebral canal. These need to be
sawn in a stepwise fashion on both sides with the ligamentous tissues as well as the
vertebral spine/posterior arch elements then being removed. This allows exposure
of the spinal cord, which can be removed intact, after transecting the relevant spi-
nal nerves.
Alternatively, the body can be left in its supine position, with the vertebral bony
tissues being exposed. The anterior vertebral arch bone is cut sequentially. One then
Normal PMCT Findings 101
removes the vertebral bodies and anterior arch tissues as well as the intervertebral
discs in one piece. This also allows exposure of the spinal cord and its removal.
Both techniques are time-consuming, labour intensive and complicated, requir-
ing good-quality mortuary staff assistance and a good reason to perform these tasks.
It has to be remembered that simply fragment autopsy sampling of parts of the brain
tissues is often an unrewarding experience unless one knows where lesions reside.
Often, it is the totality of the brain tissues and spinal cord examination that allows
one to make a value judgement as to any neuropathology. It should also be remem-
bered that brain tissues can be a particular hazard in certain circumstances. The
most important of these is that of prion diseases, with the infectious agent being
potentially aerosolised during cranial examination. Indeed, there is still deemed to
be a hazard for fixed tissue and even slide material according to some sources [3].
As a consequence, many pathologists prefer to start with the thoracic and abdom-
inal tissues in order to try to define a cause of death. If examination has been
achieved adequately beforehand, by imaging, then there is often no need to examine
the brain or related tissues. It has been shown that there is a very low frequency of
positive diagnoses in cases when there is no prompt to open the head—and opening
the head should therefore not be an automatic protocol [4].
The advent of PMCT has shown confidence in assessing the brain and to a lesser
degree the vertebral tissues, permitting the avoidance of unnecessary head and spi-
nal cord tissue examination in many cases.
Normal PMCT Findings
Brain
Very early post mortem imaging appearances (within approximately 6 h after death)
are comparable to those of the living. However, as time progresses, there are normal
changes of brain autolysis, characterised by a loss of grey–white matter differentia-
tion, decreased cerebral attenuation and mild diffuse swelling (Figs. 5.1 and 5.2)
[5–7]. Whilst perhaps less reliable than in the living for subtle change, a gross esti-
mate of cerebral volume and/or significant atrophy can still be made. If present,
significant periventricular ischaemic changes can also remain visible.
The decomposition process results in the study becoming progressively less sen-
sitive for the detection of brain abnormalities, although pathology such as signifi-
cant haemorrhage can be seen for some time. Subtle brain changes are potentially
critical findings in imaging the living, but they are unlikely to be fatal.
As decomposition progresses, gas begins to accumulate in the tissues. Initially
this is within the blood vessels, eventually becoming free within the cranium. Over
time the brain ‘slumps’ or ‘settles’ in a dependent (gravity-based) position, initially
maintaining recognisable architecture. Later it becomes soft and eventually lique-
fies, resulting in a dependent fluid level with enlarging putrefactive pneumocepha-
lus (Figs. 5.3, 5.4, 5.5, 5.6, and 5.7). At this point, the brain parenchyma can no
longer be reliably assessed by imaging.

brain cranial to the
ventricles, on brain
windows, shows the early
normal post mortem
changes of brain autolysis.
There is loss of grey–white
matter differentiation,
diffuse swelling and loss of
sulcal visualisation

brain at the level of the
lateral ventricles shows
normal post mortem
changes. There is loss of
grey–white matter
differentiation, generally
decreased cerebral
attenuation and mild
diffuse swelling

brain, on brain windows,
shows early intracranial
decomposition evidenced
by vascular gas (black in
the image) and a very tiny
pneumocephalus anteriorly

brain shows an increasing
pneumocephalus anteriorly
(arrow) compared to
Fig. 5.3 and more
extensive vascular gas

brain shows further
evolution of intracranial
decomposition with
enlarging pneumocephalus
and dependent brain
settling. Some of the gyral
outlines and the ventricles
can still be appreciated

brain shows a large
decomposition
pneumocephalus and loss
of normal brain features in
keeping with liquefaction.
The parenchyma is
unrecognisable

brain (cranium rotated
towards the right due to
positioning), on soft tissue
windows shows a
near-empty cranial vault
with small volume of
residual brain tissue
dependently, bounded by
meninges

shows multiple discrete
densities in the nasal and
oral cavities (example at
tip of arrow), some within
the cranial cavity, and a
large decomposition
pneumocephalus. Note the
folds of multiple body bags
external to the body
Occasionally, as the face is usually exposed, localised decomposition can be fur-

ther accelerated by maggot activity (Fig. 5.8, see also Chap. 3) or another animal
predation.
Intracranial Vessels
The cerebral venous sinuses, cortical veins and intracranial arteries are often hyper-
dense at PMCT due to post mortem clotting (Fig. 5.9). If there is generalised and

brain, windowed to
demonstrate normal post
mortem vessel
hyperdensity of the
cerebral venous sinuses
(large arrow) and
intracranial arteries (small
arrow). This can mimic a
contrast-enhanced
examination or thrombus,
especially when
compounded by the normal
decreased cerebral
attenuation
symmetrical vascular density, this finding can be disregarded as a normal post mor-
tem change [5, 8] although it is noted that if the body has been in a lateral position
for some time there may be asymmetry in the density—perhaps mimicking pathol-
ogy. The cerebral venous sinuses (in particular the large superior sagittal sinus) may
also show a ‘fluid–fluid level’ or an apparent ‘filling defect’ due to separation of
blood products and hypostasis (Fig. 5.10). This should not be interpreted as patho-
logical venous sinus thrombosis.
Owing to this normal post mortem vessel hyperdensity, the post mortem falx has
been described as having a ‘nodular’ appearance [8] due to visualisation of adjacent
venous structures. With knowledge of these normal post mortem appearances, such
a finding should not be mistaken for an abnormal falx or parafalcine subarachnoid
haemorrhage.
Decomposition results in the gradual accumulation of intravascular gas (Figs. 5.3
and 5.4) and is the usual explanation for gas presence. In other appropriate circum-
stances, this should not be confused with the sequela of infection, trauma or air
embolus. Arterial and venous air emboli may occur in the setting of penetrating
trauma, attempted cardiopulmonary resuscitation and iatrogenic interventions.
PMCT is superior to open autopsy in identifying both normal and pathological gas
collections [9]. For the pathologist to demonstrate intracranial gas emboli, they

the brain windowed to
demonstrate the normal
post mortem separation of
blood products in a larger
vessel. This can result in
low density of the anterior
superior sagittal sinus
(small arrow) and
high-density posteriorly
(large arrow)
would have to undertake complicated/specialised autopsy techniques, such as open-

ing the head under water [7]. It is important not to overcall this common finding on
imaging.
Spinal Cord
As with clinical CT, PMCT is not generally suitable or reliable to detect intrinsic
changes in the spinal cord, although fortunately such pathology is not often in ques-
tion. This is therefore a potential ‘blind spot’ of PMCT, much like routine open
autopsy.
Gas or blood collecting in the vertebral canal may occasionally be seen to outline
the cord and allow a gross assessment of cord integrity, most relevant when there is
a history of trauma (see also Chaps. 6 and 10). When gas collects in the vertebral
canal it is referred to as pneumorachis. Whilst this can reflect trauma, in the post
mortem setting, this is more commonly caused by advancing decomposition
(Figs. 5.11, 5.12, and 5.13).
Fig. 5.11, 5.12 and 5.13 Sagittal view of the whole spine presented on bone, lung and soft tissue
windows respectively showing pneumorachis—air in the vertebral canal which outlines the spinal
cord (arrows). At PMCT this is usually seen secondary to decomposition (in this example there is
vertebral body and soft tissue gas also due to decomposition)
Abnormal PMCT Findings 109
Abnormal PMCT Findings
Cranial and Cervical Traumatic Injury
Fatal traumatic injuries are usually accompanied by relevant history, often with
police reports and prior exclusion of a suspicious nature. In such circumstances, a
reasonably detailed PMCT report should allow the cause of death to be formulated
without further invasive investigation, aside perhaps from toxicology studies.
PMCT will easily demonstrate traumatic cranial and intracranial injuries suffi-
cient to have caused death. Indeed, findings include extensive haemorrhage, crush
fractures, brain herniation, pneumocephalus and vascular gas emboli (Figs. 5.14,
5.15, 5.16, 5.17, 5.18, and 5.19). Some appearances may seem more challenging,
such as a significant burn injury that results in destruction of brain tissues (Fig. 5.20),
although correlation with the history and external features will normally resolve this
matter quickly.
Occasionally, gunshot injury may be encountered in non-suspicious circum-
stances, following accidental or suicidal deaths (Figs. 5.21 and 5.22). Determination
of the injuries sustained (for example damage to vital structures or overwhelming
haemorrhage) will inform the cause of death, although there may be devastating
injury and no defined singular cause. In such cases, the cause of death may be
recorded as ‘multiple injuries’ secondary to gunshot, ‘or gunshot injuries to head’.
For such non-forensic cases, the information regarding the weapon and events sur-
rounding the gunshot should be known and available. For interest, and for informing
correct terminology, the reader may refer to further texts, usually of a forensic
nature [10–13].

windows showing multiple
displaced fractures,
resulting in cranial
deformity. There is a
traumatic pneumocephalus,
brain parenchymal injury
and external haemorrhage,
overall totalling a fatal
cranial injury

the head on brain windows
following crush injury
shows a traumatic
pneumocephalus, external
brain herniation (arrow)
and high-density
haemorrhagic fluid in the
body bag beneath the head

head on bone windows
shows multiple, severely
displaced cranial fractures
Catastrophic skull vault fractures, especially those resulting from crush injury
(Figs. 5.14, 5.15, and 5.16), will be clearly evident on the visual external examina-
tion. PMCT however has the additional distinct advantage of also easily demon-
strating additional skull base and cervical spine injury without the need for extensive
open dissection (Figs. 5.17, 5.18, and 5.19).

the skull on bone windows
accident shows
significantly displaced
basal skull fractures which
involve the clivus (arrow)
and occiput, with a
traumatic pneumocephalus

Fig. 5.17, soft tissue
windows demonstrate
extensive hyperdense
subarachnoid haemorrhage
in the posterior fossa,
outlining the cerebellum
and revealing traumatic
pontomedullary transection
(arrow)
Intracranial Infection and Masses
Findings relating to intracranial infection and mass lesions are more difficult to
confidently diagnose on PMCT without a known history or previous imaging to cor-
relate. As found in the clinical setting, PMCT would also usually be normal in cases

Fig. 5.17, volume-rendered
image of the skull
demonstrating the
contiguous skull base and
occipital fractures

shows extensive
intracranial and
extracranial irregular tissue
destruction resulting from
a fatal burn injury
(accidental house fire)

windows shows multiple
displaced skull fractures,
intracranial metallic shot
particles (with linear beam
hardening artefact) and
debris from an intra-oral
shotgun suicide

Fig. 5.21, at a different
level, showing the
unsurvivable brain tissue
injury with multiple
projectile tracks, fractures,
layered haemorrhage and
pneumocephalus
of meningitis or early cerebritis. Such potential infection will often be missed as a

PMCT diagnosis [14], even in the setting of sepsis/cranial symptoms [15].
An intracranial abscess, however, may be seen as a mass lesion (Fig. 5.23), often
accompanied by peri-lesional oedema, although it cannot reliably be distinguished
from an intracranial tumour or metastasis (Fig. 5.24) on imaging alone. Careful

shows an autopsy-
confirmed left frontal brain
abscess with ring-like
appearance (arrow). There
is hypoattenuating
perilesional oedema and
positive mass effect with
midline shift toward the
right

shows an autopsy-
confirmed left frontal
metastasis (arrow) with
perilesional oedema and
localised mass effect; this
was secondary to known
lung malignancy
correlation with the supporting information/clinical data and a thorough assessment

of the remaining body should be undertaken to seek out other signs of infection or
malignancy.
Intracranial Haemorrhage
As for clinical imaging, intracranial haemorrhage appears hyperdense on PMCT

(Fig. 5.18). Blood collections may visibly persist even in cases of advanced decom-
position and brain liquefaction [16–18]. Indeed, it has been described that PMCT
demonstrates a ‘still frame’ regarding haemorrhage, as it must have been at or
around the time of death [19].
PMCT is particularly valuable at demonstrating bleeding within the brain sub-
stance (Figs. 5.25 and 5.26), although a potential underlying pathology (such as
tumour or vascular malformation) may remain obscure [18]. Occasionally, as with
clinical imaging, the location or morphology of the haemorrhage may suggest the
underlying cause, such as hypertension or amyloid angiopathy (Fig. 5.27). Mass
effect, midline shift and asymmetric sulcal effacement should all be visible with
such large bleeds, until decomposition obscures these secondary effects.

shows a fatal extensive
brainstem haemorrhage
(arrow) with layered
intraventricular extension
and enlarged lateral
ventricles in keeping with
obstructive hydrocephalus

shows an acute left
parenchymal haemorrhage
(arrow) with mass effect,
midline shift toward the
right and intraventricular
extension, in a patient with
known hypertension. Post
mortem brain swelling and
loss of grey–white matter
differentiation are
compounded by the
additional mass effect

shows an acute left frontal
intracranial haematoma
(arrow) with finger-like
projections, in keeping
with cerebral amyloid
angiopathy associated
haematoma. Note also
superficial subarachnoid
extension of the
haemorrhage
In terms of sensitivity, PMCT has been shown to identify small areas of intracra-
nial haemorrhage, over a size of about 5 mm [18], which would be potentially
missed at open autopsy [15]. Conversely, it is also possible to miss small haemor-
rhages [7, 8], which may be overlooked amongst the normal post mortem hyperden-
sity of small cortical vessels. Such small haemorrhages, without secondary effects,
are however often judged unlikely to have been fatal in isolation (but might be pres-
ent as part of a constellation of other findings).
It is important to be aware that a true subarachnoid haemorrhage can occasion-
ally be difficult to distinguish from the misleading, relatively hyperdense basal cis-
terns and cortical sulci due to the decreased attenuation of the normal post mortem,
ischaemic or oedematous brain. This misleading appearance is termed ‘pseudo-
subarachnoid haemorrhage’ and is also seen in clinical imaging.
A true subarachnoid haemorrhage tends to be of striking higher attenuation [20]
(Figs. 5.28 and 5.29) and, in the absence of trauma, can be associated with underly-
ing pathology such as aneurysms (Figs. 5.30 and 5.31). Of note, normal decomposi-
tion will never lead to true subarachnoid haemorrhage [8]. Occasionally, when
extensive, the haemorrhage can track caudally into the spinal subarachnoid space
and outline the cord (Fig. 5.32).
In cases of true pathological cerebral oedema, with or without suspected ‘pseudo
subarachnoid haemorrhage’, one may see supporting signs of narrowed temporal
horns and herniation of the cerebral tonsils [21], but distinguishing such findings from
normal post mortem swelling and decreased attenuation can be extremely challenging.

shows striking high-density
haemorrhage in multiple
bilateral cerebral sulci
(example arrows) in
keeping with acute
with asymmetric right
occipital horn
intraventricular extension,
the patient was taking
anticoagulation

shows extensive
hyperdense intraventricular
and subarachnoid
haemorrhages (arrows)

shows a focal rounded
hyperdense mass in
relation to the left middle
cerebral artery in keeping
with an incidental
aneurysm. No feature of
was seen on this scan

the brainstem, windowed
to demonstrate an ovoid
hyperdense mass,
suspected to be an
aneurysm in the setting of
acute subarachnoid
haemorrhage (a tiny focus
of calcification at its
inferior margin (arrow)
suggests this to be a
vascular structure). There
is associated mass effect
distorting the brainstem
Fig. 5.32 Axial view of a

thoracic vertebral body on
high-density vertebral
canal haematoma outlining
the centrally placed,
relatively low density
thoracic spinal cord. This
blood was an extension of
a large intracranial
in a hypertensive patient
who had a sudden collapse
Cerebral Infarction
Small hyper-acute infarcts can be easily be missed on PMCT, just as in routine clini-
cal practice. The hyperdense vessel sign is generally less helpful, as high-density
vessels are a common finding and PMCT cannot differentiate ante mortem from
post mortem clot [22].

shows an asymmetric
hyperdense left middle
cerebral artery (arrow) and
associated extensive
surrounding parenchymal
hypodensity with mass
effect, consistent with
acute territory infarct
However, if there is asymmetric vessel hyperdensity and/or vascular territory

hypodense parenchyma, along with a suitable history, then the diagnosis of
infarction can be made with confidence (Figs. 5.33, 5.34, 5.35, and 5.36). In such
cases, if the PMCT does not show any other significant extracranial pathological
events, it can usually be taken that the extensive cerebral infarction is the cause
of death.
Global Ischaemia
In cases where there has been prolonged attempted cardiopulmonary resuscita-

tion, evidence of global hypoxic ischaemic injury can sometimes be identified
(Fig. 5.37). The radiologist needs to realise that, in these circumstances, hypoxic
damage is not automatically the cause of death but is rather a consequence of the
arrest. Conversely, for example, those deaths with a history such as epilepsy and
fitting should receive special consideration, as this may indeed be a relevant find-
ing in this setting.

Fig. 5.30, an axial view
more toward the vertex
demonstrates the extensive
hypodense parenchyma
and mass effect, with
midline shift bowing the
falx toward the right
(arrows)

shows right occipital
hypodensity (large arrow)
with mild mass effect
effacing the occipital horn
of the right lateral ventricle
(small arrow), in keeping
with acute/recent infarct.
Note the normal
hyperdensity of the choroid
plexus in both lateral
ventricles and posterior
superior sagittal sinus

shows a right middle
cerebral artery territory
hypodensity with no
significant overall positive
or negative mass effect, in
keeping with an evolving/
subacute infarct. The
vascular gas and small
pneumocephalus are from
decomposition changes

shows features of global
hypoxic ischaemic injury
with markedly hypodense
basal ganglia (arrows) and
insular cortex. The diffuse
cerebral swelling may be
due to a combination of
hypoxia and further post
mortem change
Ventricular Obstruction and Hydrocephalus
Although mild cerebral swelling is relatively common on PMCT, and evidence of

involutional changes may occasionally be noted, dilated ventricles are not a normal
finding. If ventricular dilatation cannot be attributed to significant central atrophy,
then cerebrospinal fluid (CSF) obstruction needs to be carefully considered,
although the underlying cause may be occult.
With such appearances, a rare but important finding in the context of sudden
death is a colloid cyst of the third ventricle. These are benign lesions sometimes
found following investigation for headaches and pressure symptoms (suggesting
raised intracranial pressure) but occasionally first identified at autopsy. As on clini-
cal imaging, they present as a focal rounded hyperdense midline mass (1–2 cm
diameter) related to the third ventricle (Fig. 5.38) and can cause acute hydrocepha-
lus and sudden unexpected death. Such secondary hydrocephalus is easier to dem-
onstrate and document on PMCT, compared to open autopsy, where brain retention
and prolonged fixation may be required [23].

shows a focal midline
hyperdense rounded mass
in relation to the third
ventricle (arrow) in
keeping with a colloid cyst.
Hydrocephalus is not seen
in this example

the upper thoracic spine on
significant lateral
displacement of the
vertebral column
secondary to traumatic
fractures resulting in a
transection of the spinal
cord (arrows)
Spinal Cord Injury
In general, spinal cord injury may be inferred when vertebral fractures are signifi-
cantly displaced or there is obliteration of the vertebral canal (Figs. 5.39 and 5.40).
Vertebral trauma is also discussed in Chap. 10.
Injuries to the cranio-cervical region and high cervical cord are particularly
important due to their potential effect on critical neurological centres. These may
prove fatal even when other injuries sustained are not extensive. Injury to the cord
should be particularly considered in the setting of chronic spinal stenosis, as this
may predispose to cord compression [24].
Following acute trauma, air may enter the vertebral canal and intracranial thecal
space from direct injury to the vertebral column or to the chest via a pleural fistula
(Figs. 5.40, 5.41, and 5.42). In this context, the air is particularly useful as it may
outline a significant spinal cord injury such as transection [25] (Fig. 5.43). Blood
can also outline the craniovertebral structures (Figs. 5.18 and 5.32). This effect has
been termed the ‘pseudo-CT myelogram’ sign and may also help reveal significant
anatomic distortion or injury [26].

Fig. 5.39, an oblique
sagittal view of the
cervicothoracic spine on
bone windows again shows
the significantly displaced
thoracic fracture with
obliteration of the vertebral
canal (arrow) consistent
with inevitable cord injury.
Also shown is traumatic
pneumorachis,
pneumothorax and soft
tissue gas

the cervical spine on bone
windows shows (arrows
top to bottom) pre-
vertebral gas,
pneumorachis and dorsal
soft tissue gas in the setting
of a penetrating chest
injury, note the absence of
decomposition gas
elsewhere

the neck on bone windows
shows a traumatic
pneumorachis (large
arrow) outlining the spinal
cord and nerve roots. A left
mandibular body fracture
and focal right
submandibular surgical
emphysema are also
demonstrated (small
arrows)

the head and neck on bone
windows showing spinal
cord transection outlined
by air (arrow), in the
setting of trauma. There
are extensive additional
skull base and skull vault
fractures
References 127
A small haemorrhage within the brainstem has previously been reported to be

visualised on PMCT although not in the context of trauma [27]; such a finding would
be important as it may cause central respiratory failure. CT is however not sensitive
enough to exclude the presence of intrinsic cord haemorrhage, particularly if there is
beam hardening due to surrounding bony structures or obesity degrading the images.
In summary, given that there will not always be gas or blood to outline spinal
structures, it is important to appreciate that spinal cord injuries cannot be excluded
purely based on PMCT [24]. As in the clinical setting, one should be aware that
disco-ligamentous injuries and small perivertebral haemorrhages can also remain
radiologically occult [28].
Reporting Brain and Spinal Cord Findings: Pearls and Pitfalls
Loss of grey–white matter differentiation, mild cerebral swelling and

generally hyperdense vessels should usually be reported as normal post
mortem appearances.
Significant haemorrhage, mass effect, major territory infarction, traumatic
injury and pathologic gas collections can reliably be visualised or excluded on
most PMCT scans, if substantial decomposition has not occurred.
A clear comment regarding the absence of pathology may help avoid the
need to open the cranial cavity should open autopsy be required.
If intrinsic spinal cord, malignant or infective pathology is suspected, and
no evidence is seen, then the limitations of PMCT should also be expressed.
• Loss of cerebral grey–white matter differentiation with mild diffuse cere-

bral swelling are considered to be normal post mortem findings.
• Hyperdense appearance of the major dural venous sinuses is a normal post
mortem finding.
• Diffuse intravascular gas is judged in keeping with decomposition.
• Large pneumocephalus and loss of normal brain architecture in keeping
with advanced decomposition. This significantly limits PMCT assessment
of the brain.
• No intracranial haemorrhage, space-occupying mass or large vessel territory
infraction. No mass effect, midline shift or hydrocephalus demonstrated.
References
com/10.1007/978-3-319-27022-7_13.
2. Burton JL, Suvarna SK. The central nervous system, with eye and ear. In: Suvarna SK, editor.
Atlas of adult autopsy [Internet]. Cham: Springer International Publishing; 2016. p. 271–97.
http://link.springer.com/10.1007/978-3-319-27022-7_9.
3. Burton JL. Health and safety at necropsy. J Clin Pathol [Internet]. 2003;56(4):254–60. http://
www.ncbi.nlm.nih.gov/pubmed/12663635.
4. Suvarna SK, McNamara MH, Scholes AL, Strickland SS. Autopsy cranial tissue exami-
nation rarely provides valuable information unless there is a specific prompt to open the
head. J Clin Pathol [Internet]. 2016;69(7):647–8. http://jcp.bmj.com/lookup/doi/10.1136/
jclinpath-2016-203657.
5. Wagensveld IM, Blokker BM, Wielopolski PA, Renken NS, Krestin GP, Hunink MG, et
al. Total-body CT and MR features of postmortem change in in-hospital deaths. PLoS One
[Internet]. 2017;12(9):e0185115. https://dx.plos.org/10.1371/journal.pone.0185115.
7. Panda A, Kumar A, Gamanagatti S, Mishra B. Virtopsy computed tomography in trauma: normal
postmortem changes and pathologic spectrum of findings. Curr Probl Diagn Radiol [Internet].
8. Smith AB, Lattin GE, Berran P, Harcke HT. Common and expected postmortem CT obser-
vations involving the brain: mimics of antemortem pathology. Am J Neuroradiol [Internet].
2012;33(7):1387–91. http://www.ajnr.org/lookup/doi/10.3174/ajnr.A2966.
9. Flach PM, Thali MJ, Germerott T. Times have changed! Forensic radiology—a new challenge
for radiology and forensic pathology. Am J Roentgenol [Internet]. 2014;202(4):W325–34.
http://www.ajronline.org/doi/10.2214/AJR.12.10283.
10. Wilson AJ. Gunshot injuries: what does a radiologist need to know? RadioGraphics [Internet].
1999;19(5):1358–68. http://pubs.rsna.org/doi/10.1148/radiographics.19.5.g99se171358.
11. van Kan RAT, Haest IIH, Lahaye MJ, Hofman PAM. The diagnostic value of forensic imag-
ing in fatal gunshot incidents: a review of literature. J Forensic Radiol Imaging [Internet].
12. Abdul Rashid SN, Martinez RM, Ampanozi G, Thali MJ, Bartsch C. 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
[Internet]. 2013;1(2):63–7. https://linkinghub.elsevier.com/retrieve/pii/S2212478013000464.
13. Jeffery AJ, Rutty GN, Robinson C, Morgan B. Computed tomography of projectile injuries.
Clin Radiol [Internet]. 2008;63(10):1160–6. https://linkinghub.elsevier.com/retrieve/pii/
S000992600800130X.
14. Roberts ISD, Benamore RE, Benbow EW, Lee SH, Harris JN, Jackson A, et al. Post-mortem
imaging as an alternative to autopsy in the diagnosis of adult deaths: a validation study.
Lancet [Internet]. 2012;379(9811):136–42. https://linkinghub.elsevier.com/retrieve/pii/
S0140673611614839.
16. Takahashi Y, Sano R, Kominato Y, Takei H, Kobayashi S, Shimada T, et al. Usefulness of post-
mortem computed tomography for demonstrating cerebral hemorrhage in a brain too fragile
for macroscopic examination. J Forensic Radiol Imaging [Internet]. 2013;1(4):212–4. https://
linkinghub.elsevier.com/retrieve/pii/S221247801300083X.
17. Sano R, Hirasawa S, Awata S, Kobayashi S, Shimada T, Takei H, et al. Use of postmortem
computed tomography to reveal acute subdural hematoma in a severely decomposed body with
advanced skeletonization. Leg Med [Internet]. 2013;15(1):32–4. https://linkinghub.elsevier.
com/retrieve/pii/S1344622312001320.
References 129
18. Tappero C, Thali MJ, Schweitzer W. The possibility of identifying brain hemorrhage in putre-
fied bodies with PMCT. Forensic Sci Med Pathol [Internet]. 2020;16(4):571–6. http://link.
springer.com/10.1007/s12024-020-00283-8.
19. Ruder TD, Zech W-D, Hatch GM, Ross S, Ampanozi G, Thali MJ, et al. Still frame from the
hour of death: acute intracerebral hemorrhage on post-mortem computed tomography in a
decomposed corpse. J Forensic Radiol Imaging [Internet]. 2013;1(2):73–6. https://linkinghub.
20. Given CA, Burdette JH, Elster AD, Williams DW. Pseudo-subarachnoid hemorrhage: a poten-
tial imaging pitfall associated with diffuse cerebral edema. AJNR Am J Neuroradiol [Internet].
2003;24(2):254–6. http://www.ncbi.nlm.nih.gov/pubmed/12591643.
21. Berger N, Ampanozi G, Schweitzer W, Ross SG, Gascho D, Ruder TD, et al. Racking the
brain: detection of cerebral edema on postmortem computed tomography compared with
forensic autopsy. Eur J Radiol [Internet]. 2015;84(4):643–51. https://linkinghub.elsevier.com/
retrieve/pii/S0720048X15000030.
22. Sutherland T, O’Donnell C. The artefacts of death: CT post-mortem findings. J Med Imaging
Radiat Oncol [Internet]. 2018;62(2):203–10. http://doi.wiley.com/10.1111/1754-9485.12691.
23. Andersen AM, Frost L, Thorup Boel LW. Colloid cysts of the third ventricle at post-mortem
CT and at autopsy: a report of two cases. J Forensic Radiol Imaging [Internet]. 2015;3(1):96–9.
24. Makino Y, Yokota H, Hayakawa M, Yajima D, Inokuchi G, Nakatani E, et al. Spinal cord inju-
ries with Normal postmortem CT findings: a pitfall of virtual autopsy for detecting traumatic
death. Am J Roentgenol [Internet]. 2014;203(2):240–4. http://www.ajronline.org/doi/10.2214/
AJR.13.11775.
25. Berger N, Ross SG, Ampanozi G, Majcen R, Schweitzer W, Gascho D, et al. Puzzling over
intracranial gas: disclosing a pitfall on postmortem computed tomography in a case of fatal
blunt trauma. J Forensic Radiol Imaging [Internet]. 2013;1(3):137–41. https://linkinghub.else-
vier.com/retrieve/pii/S2212478013000737.
26. Bolster F, Ali Z, Daly B. The “pseudo-CT myelogram sign”: an aid to the diagnosis of underly-
ing brain stem and spinal cord trauma in the presence of major craniocervical region injury on
post-mortem CT. Clin Radiol [Internet]. 2017;72(12):1085.e11–1085.e15. https://linkinghub.
28. 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 [Internet].
of the Extra-Cranial Head and Neck 6
Introduction
Significant fatal pathology in the extracranial head and neck, seen at post mortem
computed tomography (PMCT), predominantly deals with trauma from various
sources. Of particular note in the autopsy arena is the pathology of hanging.
Consequently, this topic is presented in more depth as a specific circumstance.
Occasionally, PMCT is performed after an episode of choking, with the scan poten-
tially able to identify airway obstruction.
Other pathologies such as infections or malignancy might also be found in
PMCT, often as part of systemic disease. However, in terms of being the cause of
death, they are infrequent findings compared to cardiac and lung disease.
The extracranial head and neck includes the deep facial tissues and paranasal
sinuses. These structures are rarely examined during an invasive autopsy (see
below), as they have a low yield of unexpected relevant positive findings. PMCT has
the advantage of being able to readily visualise the entirety of this anatomic region
and may occasionally reveal surprising pathology. Clearly, it also has the advantage
of avoiding destructive facial interactions.
As with most CT, the depiction of bony detail and gas distribution is excellent,
but soft tissue evaluation is limited unless structures are outlined by contrast or clear
fat planes. Streak artefact from jewellery, dental appliances or amalgam can signifi-
cantly obscure the neck and oral cavity views (Figs. 6.1 and 6.2). Where possible,
before the scan commences, such items should be removed to enhance imaging. As
with all PMCT, discussed earlier in this book, the changes of decomposition can
also significantly limit assessment (see Chap. 3).

Switzerland AG 2021
https://doi.org/10.1007/978-3-030-70829-0_6
132 6 Post Mortem Computed Tomography of the Extra-Cranial Head and Neck
Fig. 6.1 Axial view at the

level of the thyroid on soft
streak artefact from
metallic jewellery (arrow)
obscuring the neck tissues

neck at the level of the oral
cavity shows multiple
streak artefacts from dental
amalgam (arrows) which
obscure the tissues
utopsy of the Extracranial Head and Neck:

A
The Pathologist’s Perspective
The pathologist is aware that the face is one of the few areas commonly viewed by
relatives after death. It causes distress to the family and friends if there are disfigur-
ing marks or sutures. Consequently, many pathologists prefer not to open the head
tissues, particularly if PMCT has excluded any pathology at these sites.
However, if required, the dissection of the head and neck is normally accom-
plished by a Y-shaped incision running across the upper chest up towards both mas-
toid processes. This allows the skin and underlying fatty soft tissue to be reflected
upwards. It exposes the strap muscles, blood vessels, airway, thyroid, parathyroids,
salivary glands and oesophagus, up to the angle of the mandible [1].
In a conventional autopsy, the knife incision runs along the inner aspect of the
mandible and then allows incision onto the front of the cervical spine to release/
remove the neck contents.
Conventionally, the neck arteries are explored up to the bifurcation of the carotid
vessels but only after the neck tissues had been removed from the body. Rarely, the
vertebral bodies are removed in order to consider the spinal cord as discussed in the
previous chapter. Confirmation of the course and tissue interaction for artificial air-
ways, long lines and electronic devices should be undertaken as part of the routine
assessment.
In forensic cases, close attention to the strap muscles, hyoid bone and laryngeal
cartilages is undertaken in order to look for bruising and fractures of the airway
framework tissues.
Exposure of the facial tissues is not normally undertaken unless an assault is
under consideration, again in a forensic setting.
The inner nose (exposed via the soft and hard palate), the inner ear (examined by
means of focused temporal bone resection and decalcification) and eye (often
approached from the bone of the anterior cranial fossa) are not normally seen in
routine cranial practice [2].
It would be fair to say that there is rarely significant natural pathology in any of
these anatomic areas unless there is airway obstruction, primary/metastatic disease
or infection.
Soft Tissues
If sufficient fat planes are present, a reasonable assessment of the soft tissues can be
made compared to those individuals lacking in body fat (Figs. 6.3 and 6.4). This
allows the exclusion of significant masses (that obstruct the airway) or substantial
haemorrhage. The outlines of the thyroid, salivary glands and muscles also allow for
a gross assessment.
The thyroid is readily visualised due to its inherent hyperdensity (Fig. 6.5) but is
rarely of significance unless a neoplastic or a large goitre narrows the airway. The
laryngeal cartilages are generally non-calcified when young (Fig. 6.6) but become
variably calcified with increasing age.
As the face is often exposed in the open, compared to other body parts, earlier
decomposition and maggot infestation may be present (see Chap. 3). Maggots are
seen on PMCT as multiple tiny soft tissue densities in and around the facial tissues,
nasal cavities, orbits, ears and paranasal sinuses and may destroy much of the soft
tissues of the head and neck (Figs. 6.7 and 6.8).

neck on soft tissue
windows at the level of the
submandibular glands
showing good soft tissue
structure definition due to
the prominent fat planes in
this obese body, despite the
absence of intravenous
contrast

neck at the level of the
larynx in this body with
minimal body fat shows
poor definition of the
anterior neck structures
and tissue planes but no
gross asymmetry or
hyperdense haemorrhage
Orbits
In the post mortem state, over time there is loss of ocular volume leading to crum-
pling of the globes (Figs. 6.9 and 6.10), with vascular gas accumulation (Fig. 6.11)
and occasionally dislocation of the lens (Fig. 6.12). With more advanced decompo-
sition, the globes become unrecognisable (Fig. 6.13), but one should also be aware
that corneal tissues may have been harvested for tissue donation (Fig. 6.14).

neck on soft tissue
windows shows a normal
size and morphology of the
thyroid (arrows), which
has increased density
relative to surrounding
tissues owing to its
intrinsic iodine content

neck at the level of the
larynx shows normal,
non-calcified thyroid
cartilage (arrows) in a
16-year-old
It has been shown that body fluids such as vitreous humour and cerebrospinal
fluid slightly increase in density over time [3] but only by a few Hounsfield units.
This may in the future help forensic investigation with estimation of an unknown
post mortem interval (PMI). If frankly high-density intra-ocular fluid is seen, it is
more likely to relate to haemorrhage [4]. Orbital implants and retinal detachments
are also occasionally seen (Figs. 6.15 and 6.16), usually reflected in the past medical
history. These are often unrelated to the cause of death, particularly if there is no
history of trauma.

head on soft tissue
windows shows soft tissue
destruction of the right
pinna (arrow) secondary to
maggot activity, note also a
small decomposition
pneumocephalus
Fig. 6.8 Axial view at the

level of the paranasal
sinuses on lung windows
shows multiple discrete
soft tissue densities
in-and-around the facial
tissues in keeping with
maggot infestation and soft
tissue destruction. Small
decomposition
pneumocephalus also
present

orbits windowed to show
normal post mortem
crumpling of the globe
margins (arrows)

windows shows normal
post mortem ocular
collapse (arrows) and small
pneumocephalus due to
decomposition

windows shows vascular
gas due to decomposition
in the superior ophthalmic
veins (arrows)

windows shows post
mortem dislocation of the
right globe lens (arrow)
and partially collapsed
globes. Note also soft
tissue decomposition gas

the orbits windowed to
show almost complete loss
of globe tissue and
surrounding soft tissue
decomposition gas

the orbits showing absence
of both globes, harvested
for transplant with
post-procedure orbital gas
yet no surrounding
evidence of soft tissue
decomposition
Paranasal Sinuses
Fluid in the paranasal sinuses and nasopharynx is a common finding on PMCT [5]
and should usually be considered as normal (Figs. 6.16 and 6.17). One can speculate
whether unilateral collections point to localised infection or neoplasia, but the ori-
gin and density of such fluid is often multifactorial. Indeed, collections may relate

the orbits showing a dense
(white) spherical right
orbital implant and anterior
eye prosthesis

the orbits windowed to
show chronic bilateral
hyperdense retinal
detachments (arrows) and
right-side cataract in a
person who was registered
blind. Incidental partial
opacification of the
paranasal sinuses is also
demonstrated

the level of the
nasopharynx showing a
homogenous, gravity-
dependent fluid level in the
nasopharynx (arrows),
generally considered to be
a normal post mortem
finding
to respiratory secretions, decomposition, regurgitated stomach contents, haemor-

rhage and/or aspirated material to name a few. Without prior imaging, or a relevant
clinical history, it may well be impossible to differentiate the exact nature of this
fluid on PMCT.
Trauma
Facial, skull vault/base and cervical spine fractures (see also Chaps. 5 and 10) are
well demonstrated on PMCT, and they are easier to define compared to open
autopsy. Such fractures range from the trivial through to the fatal; particularly, if the
airway is compromised, there is significant neurological injury or massive
haemorrhage.
The cervical spine should always be examined for fracture, dislocation or sub-
luxation (bearing in mind that post mortem muscular laxity or rigor may result in
unusual positioning such as a cervical rotational subluxation). Whilst the spectrum
of trauma includes that similar to clinical practice, in the post mortem setting it is
more common to see severe fractures with implied cord injury which might be con-
sidered incompatible with life (Figs. 6.18, 6.19, 6.20, and 6.21). One should note
that bone fragments, haemorrhage or debris may be actively or passively transferred
from the head and neck into the lower airways, acting either as pathology or as
artefacts.
Following trauma, surgical emphysema (Fig. 6.22) or non-decomposition vascu-
lar air (Fig. 6.23) is readily visualised on PMCT. Significant soft tissue disruption

windows showing
significantly displaced,
comminuted fractures of
C2 resulting in bony
cervical canal stenosis
(arrow) and therefore
potential cord injury. A
background of
degenerative change is
noted

Fig. 6.18, an axial view
again shows the severely
comminuted C2 fracture
with multiple bony
fragments

windows showing an
extension-type fracture
through the C6/7
disc-space (arrow) and C6/
C7 posterior elements with
significant residual fracture
distraction. A background
of degenerative change is
noted
and haematoma will also be visualised, but the reporter should be aware that super-
ficial bruising or small volumes of haemorrhage are much more difficult to appreci-
ate and cannot be excluded on the basis of PMCT alone, demonstrating the need for
additional, good-quality external examination.

the head, neck and upper
windows showing
extensive cranio-cervical
trauma, incompatible
with life
Choking and Aspiration
Whilst, as discussed earlier, fluid in the nasopharynx, paranasal sinuses and airways
is usually a normal post mortem finding, in the setting of a witnessed episode of
coughing or choking, the presence of mixed density debris (potential food matrix)
in the upper aerodigestive tract raises the possibility of fatal airway obstruction
(Figs. 6.24, 6.25, 6.26, 6.27, and 6.28). Other findings linked to (potentially chronic)
aspiration include lung changes reflecting the foreign matter and/or secondary
inflammation with pneumonia, these may add confidence to the diagnosis (see also
Chap. 7). Nevertheless, it is stressed that without an appropriate history, any such
PMCT findings are probably best regarded as indeterminate.

windows showing
pre-vertebral surgical
emphysema (arrows) and a
small pneumorachis from
cranial trauma (not seen on
this slice). Note the lack of
decomposition gas
elsewhere indicating that
these gas collections are
traumatic in nature
Infections of the Head and Neck
One appreciates that there are many systemic consequences of infective processes
that may result in death. Considering the head and neck specifically, infections can
be implicated in the person’s death directly by means of airway compromise and
large infective processes damaging normal tissue function. The external clues of
sepsis such as generalised erythema, skin necrosis, a greenish tinge or marbling (see
Chap. 3) cannot be seen on PMCT. Consequently, any ante mortem data should
always be considered from a potential infection perspective. When reviewing the
PMCT, the possibility of microbial pathology may be considered when there is a
focal mass, inflammatory stranding, lymphadenopathy (reactive) and localised gas
production, causing surgical emphysema [6].
Potential sites of infective pathology include the tonsils, salivary glands, teeth/
gums with extension to the para- or retro-pharyngeal space, the floor of mouth or
upper mediastinum. Spondylodiscitis may also present with upper aerodigestive
tract swelling [7], further increasing the importance of reviewing the spine.

showing gas embolus in
the carotid artery lumen
(arrow) resulting from a
chest injury sustained in a
road traffic collision. Note
the lack of decomposition
gas elsewhere

windows showing a
discrete mixed density
suspected food bolus in the
hypopharynx and laryngeal
inlet (arrow) in a case of
witnessed choking

windows showing mixed
density debris in the
hypopharynx. This was a
case of witnessed choking
on food in patient with a
history of Parkinson’s
disease

density suspected food or
vomit in the upper
aerodigestive tract in a
patient found deceased at
home. The lack of a
supporting history makes
the imaging finding alone
less reliable

Fig. 6.26, the axial view of
the neck shows that the
mixed density debris
slightly distends and
obstructs the airway

shows similar debris
partially filling the right
main bronchus (arrow) and
patchy nodular lung
opacities due to aspiration
(as well as normal
hypostasis)
A rare complication of sinusitis or oto-mastoiditis is of bone destruction and

intracranial extension of infection. This may lead to meningitis, empyema or
abscess. The latter two may be appreciated on non-contrast PMCT if sizeable [8].
As with clinical imaging, meningitis alone would not be appreciated on PMCT (see
Chap. 5).
Special Circumstances: Hanging 147
It would be nearly impossible on PMCT alone to exclude a previously unknown

tumour as a potential cause of infection or haemorrhage or indeed to exclude sec-
ondary vascular thrombosis. If such pathology is suspected, then a limited open
autopsy or needle biopsy may be indicated, with the PMCT used to comprehen-
sively assess the bones and exclude pathology in the other body compartments.
Special Circumstances: Hanging
Suicide by hanging is unfortunately a common cause of death across the world. This
distressing reality for the relatives is potentially made worse if they have concerns
about an open autopsy. A detailed PMCT may be sufficient to avoid such investiga-
tion. The role of the radiologist is to assess the neck for hanging-related injury and
to consider the body for other injuries, disorders, and alternate causes of death or
co-existing pathology.
Where possible, owing to the complexity and small size of neck structures, thin-
slice CT should be acquired in a small field of view, with both bone and soft tissue
algorithms applied. This may occasionally require additional acquisitions per-
formed specifically for these circumstances.
One should be mindful that if subsequent open autopsy is performed, then the
scan provides a permanent record of appearances before potentially difficult and/or
destructive pathological examination.
Types of Hanging
Hanging may vary in format. First, hanging can involve a body dropped from a
height with a noose applied around the neck. In this circumstance, the sudden halt
of the fall causes a violent jerking of the head at the neck. This can lead to cervical
spine fracture dislocations (C1/C2 level), with spinal cord injury compromising
brainstem vital centres and causing death almost instantly.
Second, hanging can involve the application of a noose around the neck without
any drop. This latter reality may involve full or partial suspension of the body. Full
suspension implies that no body parts are in contact with the ground as implied with
a rope/noose holding the body aloft. Partial suspension implies some contact of the
body with the ground. Commonly, the body is partly suspended with a neck ligature
attached to a door handle, coat hook or other room structure. This partial suspension
still causes compression of neck structures and death. It involves interference with
blood flow to/from the brain as well as impeding gas flow in/out of the trachea [9].
There is also an effect from the direct pressure on carotid baroreceptors, which may
drop heart rate and blood pressure. Furthermore, there may also be a surge in cate-
cholamine release, which can exacerbate the risk of arrhythmia and subsequent car-
diac arrest.
Imaging Findings
Looking at suicidal hangings, the radiologist will often note that the standard PMCT
is normal. The absence of other abnormal findings, along with appropriate circum-
stances, will support the interpretation of hanging as the cause of death. The pathol-
ogist will arrange toxicology assessment and a thorough external assessment to
exclude other injuries and features of third-party involvement. It is tempting for the
radiologist to declare suicidal death directly after PMCT, but it is advised that this
be left to the pathologist in case additional data later becomes available.
Many cases of hanging have skin or soft tissue distortion and compression from
a ligature. The ligature/noose may still be attached to the body and evident on
PMCT (Fig. 6.29) but often with the tension released. Scanning with the ligature in
maintained tension is helpful, as it allows assessment of any anatomic distortion or
airway narrowing that may be present (Figs. 6.30, 6.31, 6.32, and 6.33). There may
be cranial displacement of the hyoid or larynx, resulting in occlusion of the soft tis-
sues of the pharynx. Constriction of these soft tissues is more likely than compres-
sion of the stronger laryngeal or tracheal cartilage [9], but there may also be injury
to these structures.
Even if removed, the ligature may have left an externally visible mark on the
body. If causing deformity of the tissues, this can be seen on volume-rendered sur-
face imaging or multiplanar reconstructions. When a body is found in full suspen-
sion, the ligature mark is usually symmetrical and slopes cranially toward the back
of the neck [10]. This contrasts with partial suspension [11], where the head may be
turned or flexed to one side, with asymmetry of the ligature marks.

following a hanging in
suspension. The thick rope
ligature remains in place
externally (arrow)

following a hanging in
suspension shows an
external rope ligature
(arrows) in tension
between the chin and neck
tissues. There is
anteroposterior
compression of the cricoid
cartilage and airway

Fig. 6.30, the sagittal view
of the neck shows the
external rope ligature
anteriorly and posteriorly
around the neck (arrows)
in tension. There is
resultant airway narrowing
at the level of the ligature
and cranial displacement
of the larynx causing soft
tissue occlusion of the
oropharynx

windows in a case of
hanging by shoelace (still
in place but not well seen
due to a thin calibre). This
lies at the level of the
thyrohyoid membrane
(arrow) and is causing
airway occlusion at the
same level

showing an electrical cable
ligature (metallic density)
in tension at the level of
the thyrohyoid membrane.
There is upward
displacement of the hyoid
bone with soft tissue
airway occlusion at the
level of the oropharynx
Fractures and Dislocations
It is unusual for suicidal hangings to result in cervical spine fractures, unless jump-
ing or falling from a height [9]. In full suspension hangings, fractures may be sym-
metrical or asymmetrical in pattern. In contrast, in partial suspension hangings there
a more likely to be asymmetry in any fractures. While interesting to correlate with
the hanging mechanism, in the non-suspicious post mortem setting, extensive detail
regarding the fracture pattern is not necessary as might be the case for clinical man-
agement or forensic cases.
When present, vertebral fractures and dislocations should be described with the
same terminology as for clinical CT, with consideration whether this has led to spi-
nal cord trauma (Figs. 6.34 and 6.35).
In addition, evidence of fracture or distortion of the laryngeal cartilages
(Figs. 6.36, 6.37, 6.38, 6.39, and 6.40) and hyoid bone (Figs. 6.41, 6.42, 6.43, and
6.44) should be sought as part of the overall review. Injuries to the superior horns of
the thyroid cartilage are generally most common in cases of hanging [11]. When the
hyoid is fractured this generally involves the greater horns [11].

showing angulation and
disc space widening at
C2/3 and a tiny displaced
postero-inferior corner
fracture of C2 (arrow)
suggesting ligamentous
disruption. These injuries
resulted from a hanging in
suspension

Fig. 6.34, the axial view of
C2 shows bilateral
transverse fractures
through the pedicles
(arrows) also known as a
‘Hangman type fracture’

windows shows a buckled,
non-calcified thyroid
cartilage following hanging
(arrow). There is also
soft tissues
Other Findings Resulting from Hanging
Soft tissue injury may be subtle at PMCT, with open autopsy and PMMRI being
considered more sensitive in detecting strap muscle trauma [12, 13]. However, even
with open autopsy there may also be surprisingly little to find macroscopically [9],
although there may be histological changes if sampled.
Subcutaneous gas collections in the head and neck, out of proportion to decom-
position changes, might possibly be caused by airway rupture from gasps for breath

the neck windowed to
show a minimally calcified
thyroid cartilage in a
young adult, buckled from
hanging injury (arrow)

shows a subtle buckle
injury of a calcified left
thyroid cartilage lamina
(arrow) secondary to
hanging in full suspension.
Additional facet joint and
vertebral degenerative
changes are noted
[12] and are better demonstrated on PMCT than open autopsy. This may just be a
subtle ‘gas bubble sign’, a very tiny focus of gas in the peri-laryngeal soft tissues
indicative of adjacent laryngeal fracture [11] or a small volume of gas within the
cartilage itself (Fig. 6.40).
When hanging vertically for some time, the effects of post mortem hypostasis
may be seen ‘dependently’ in the lower torso and legs [9]. There may be a

showing the thyroid
cartilage with a buckle
fracture of the left greater
horn (arrow) and a loose
rope ligature externally
around the neck

shows a compressed
cricoid cartilage (arrow)
following hanging injury.
Gas within the cartilage
(and not elsewhere)
suggests possible
barotrauma
craniocaudal gradient of ground-glass change in the lungs, although with later

supine positioning of the body this may partly or wholly redistribute. In younger
people with no significant co-morbidities, the lungs may be almost completely
clear, which is generally unusual for PMCT (Fig. 6.45).

showing a cranially
orientated fracture of the
greater horn of the hyoid
(arrow) following hanging
Hanging or Strangulation?
In the declared ‘non-suspicious’ setting, radiologists rely on the history supplied

regarding circumstances and how the body was found in order to interpret the find-
ings. The findings of hanging on PMCT may be difficult to differentiate from liga-
ture strangulation or manual compression (i.e. homicide, caused by another person).
On external inspection, there may be more superficial abrasion or bruising present
in such cases, as the forces involved are likely to have been higher. This highlights
the need for an experienced pathologist external assessment and questioning of the
information provided in cases of hanging.
Fig. 6.42 Sagittal

maximum intensity
projection reconstruction
on bone windows showing
fracture of the greater horn
of the hyoid from hanging
in full suspension (arrow).
The metallic high-density
electrical cable ligature
remains present with
visible soft tissue
indentation

shows a traumatic
disruption at the
fibrocartilaginous joint
(which is sometimes fused)
between the greater horn
and body of hyoid (arrow)
following hanging. Note
the left greater horn is
angled out-of-plane

showing a caudally
orientated fracture of the
greater horn of the hyoid
(arrow) following a
hanging in suspension

shows very minimal
dependent ground glass
change (arrow) in keeping
with hypostasis, in the
lungs of a young adult
following hanging
Reporting Head and Neck Findings: Pearls and Pitfalls
The soft tissues and bones of the head and neck should be routinely reviewed
to ensure there is no significant mass, airway occlusion or trauma in all
PMCT cases.
Fluid in the paranasal sinuses and nasopharynx is a common finding on
PMCT and should usually be considered as normal.
Findings consistent with choking and airway occlusion may be correlated
when there is an appropriate supporting history.
In cases of hanging, assessment should be made of the cervical spine,
hyoid, laryngeal cartilages, trachea and surrounding soft tissues. A descrip-
tion of any injuries, with relevant negative findings should be included. In
many cases, there may be no specific findings on imaging to confirm, or
refute, hanging as the cause of death other than the supplied history. This
conclusion should be made clear in the report.
• Fluid in the nasopharynx and paranasal sinuses is judged to be a normal

post mortem finding.
• No cervical spine fractures. Normal alignment of the cranio-cervical junc-
tion, cervical vertebral bodies and facet joints.
• A ligature remains in place, with tension released.
• Anteroposterior compression of the neck by a ligature at the level of the
thyrohyoid membrane is causing complete airway obstruction.
• No fractures of the cervical spine, hyoid or laryngeal cartilages.
• No surgical emphysema or large soft tissue haematoma.
References
1. Coe MS, Suvarna SK. Evisceration. In: Suvarna S, editor. Atlas of adult autopsy
[Internet]. Cham: Springer International Publishing; 2016. p. 47–63. http://link.springer.
com/10.1007/978-3-319-27022-7_3.
2. Burton JL, Suvarna SK. The central nervous system, with eye and ear. In: Suvarna SK, editor.
Atlas of adult autopsy [Internet]. Cham: Springer International Publishing; 2016. p. 271–97.
http://link.springer.com/10.1007/978-3-319-27022-7_9.
postmortem changes and pathologic spectrum of findings. Curr Probl Diagn Radiol [Internet].
References 159
5. Biljardt S, Brummel A, Tijhuis R, Sieswerda-Hoogendoorn T, Beenen LF, van Rijn RR. Post-
mortem fluid stasis in the sinus, trachea and mainstem bronchi; a computed tomography study
in adults and children. J Forensic Radiol Imaging [Internet]. 2015;3(3):162–6. https://linking-
6. Baumeister R, Gauthier S, Schweitzer W, Thali MJ, Mauf S. Small—but fatal: postmor-
tem computed tomography indicated acute tonsillitis. J Forensic Radiol Imaging [Internet].
pii/S2212478014001075.
8. Hyodoh H, Matoba K, Murakami M, Saito A, Okuya N, Matoba T. Lethal complication in
Pott’s puffy tumor: a case report. J Forensic Radiol Imaging [Internet]. 2018;14:12–5. https://
book/9780340972533.
10. Kawasumi Y, Hosokai Y, Usui A, Sato M, Takane Y, Saito H, et al. Hanging: postmortem com-
puted tomography. Poster session presented at: European Congress of Radiology; 2011 March
3–7; Vienna, Austria. [Internet]. https://doi.org/10.1594/ecr2011/C-1846.
11. Schulze K, Ebert LC, Ruder TD, Fliss B, Poschmann SA, Gascho D, et al. The gas bubble
sign—a reliable indicator of laryngeal fractures in hanging on post-mortem CT. Br J Radiol
[Internet]. 2018;20170479. http://www.birpublications.org/doi/10.1259/bjr.20170479.
12. Elifritz J, Hatch GM, Kastenbaum H, Gerrard C, Lathrop SL, Nolte KB. 1.8. PMCT findings in
hanging. J Forensic Radiol Imaging [Internet]. 2014;2(2):97. https://linkinghub.elsevier.com/
retrieve/pii/S2212478014000227.
13. Gascho D, Heimer J, Tappero C, Schaerli S. Relevant findings on postmortem CT and post-
mortem MRI in hanging, ligature strangulation and manual strangulation and their addi-
tional value compared to autopsy—a systematic review. Forensic Sci Med Pathol [Internet].
2019;15(1):84–92. http://link.springer.com/10.1007/s12024-018-0070-z.
of the Chest 7
Introduction
The causes of sudden death in an adult often relate to the chest, making this a crucial
cavity to thoroughly examine. As this is an important region and a large topic, it is
split over two chapters. This first chapter covers non-cardiac findings in the lungs
and mediastinum, with the heart being covered in its own right in the subsequent
chapter.
The lungs can be challenging to interpret on post mortem computed tomography
(PMCT). Despite the absence of movement artefact, there may be partial lung col-
lapse due to a non-inspiratory phase and variable ‘ground-glass’ parenchymal opac-
ity due to normal fluid hypostasis after death. In addition, there are a myriad of
background pathologies that may be encountered as incidental, contributory and/or
directly relevant to the cause of death.
This chapter discusses various normal and pathological chest findings and also
covers drowning as a special circumstance, which may be unfamiliar to those who
work in general clinical practice.
Autopsy of the Chest: The Pathologist’s Perspective
It would be fair to state that the majority of pathologies causing death reside within
the thorax, affecting the heart and/or lungs.
The pathologist approaches the thorax initially from the external perspective,
examining the chest for features of hyper-expansion, deformity, injuries, scars and
symmetry. The internal aspects of the chest are considered by reflecting the skin and
soft tissues from the rib cage and then removing the chest plate of anterior ribs and
sternum in one piece.

Switzerland AG 2021
https://doi.org/10.1007/978-3-030-70829-0_7
162 7 Post Mortem Computed Tomography of the Chest
Fluid collections in the pleural cavities and pericardium can easily be identified
and measured. At this point, the examination should also consider features of con-
genital anatomy variation, particularly with regard to the heart (see Chap. 8).
One can remove the chest content either with the mouth, pharynx and neck struc-
tures and/or the abdominal tissues down to the pelvic compartment in one or mul-
tiple fragments. Alternatively, one can transect the mediastinal tissues at the thoracic
inlet and cut across the superior aspect of the diaphragm to assist release of the heart
and lungs with the mediastinal component. Once removed, the thorax should be
considered, looking for fractures, metastatic neoplasia and infections, with addi-
tional usually brief review of the vertebral body alignment.
After opening the pericardium and removing the heart (see Chap. 8), the lungs
are normally removed separately by cutting through the pulmonary hilum (vessels,
airways) so that the lungs may be examined sequentially. As the cardiac tissues are
removed, it is important to check for pulmonary embolism by direct palpation of the
pulmonary artery content and visual inspection of the vasculature.
Once isolated, the lung tissues can be examined in two ways. They can be sliced
longitudinally (in the parasagittal plane) to provide an overview of the architecture,
somewhat akin to sequential radiological slices. Many pathologists also have an
alternate approach that is initially to dissect along the pulmonary artery to exclude
small pulmonary emboli and then to turn the lung tissues over and dissect along the
bronchi to exclude obstructions, infections and neoplasia. If these two sequential
examinations are performed, then the lung tissues would have been thoroughly
examined and samples can be reserved for histology as deemed appropriate.
The mediastinum rarely poses any pathological process for consideration of a
cause of death, although tumours of the thymus and mediastinal lymph nodes should
always be considered at the same time as pathology of the major airways and large
vessels are being reviewed.
Thoracic Airways
The upper respiratory tract (pharynx, larynx, trachea and main bronchi) is normally
well preserved following death and easy to identify on PMCT. Quite often there is
fluid in the trachea and main bronchi which may partially or completely fill these
structures (Figs. 7.1 and 7.2). When low-density and homogenous, this should usu-
ally be considered a normal finding [1].
More unusually, in cases found in warmer months or exposed circumstances,
maggots may have crawled down the airways from the nose and mouth and appear
as filling defects or an irregular soft tissue mass [2]. This must be considered in
cases of decomposition in order to avoid misinterpretation as pathological airway
obstruction (see Chap. 3).

upper chest on soft tissue
windows shows fluid
filling the upper trachea
(arrow) and moderate
bilateral pleural effusions

shows a fluid filled trachea
(arrow) with a tiny locule
of gas anteriorly. Note also
small pleural effusions and
patchy ground-glass
opacity
Lungs
On PMCT, the lungs almost always appear ‘abnormal’, compared to clinical imag-
ing. Ideally, the lungs are best examined as soon after death as possible in order to
reduce the effects of fluid accumulation, hypostasis and decomposition, which
increase over time.
Hypostasis is a commonly encountered post mortem change affecting the lung
parenchyma. Generally, it presents as approximately symmetrical ground-glass
opacification with a gradient of increasing density toward the dependent area. Often,
this gradient has a distinct horizontal ‘fluid-level’ demarcation (Figs. 7.3, 7.4, 7.5,
7.6, and 7.7). If the body has been lying in a position other than supine, then the

showing dependent,
bilateral ground-glass lung
opacity with horizontal
demarcation (arrows) in
keeping with normal post
mortem fluid hypostasis

showing dependent,
bilateral fluid hypostasis
and additional normal post
mortem tiny pleural
effusions (arrows)

upper chest on lung
windows shows more
extensive post mortem
fluid hypostasis in the
lungs compared to
previous examples

shows normal post mortem
fluid hypostasis of the
lungs on a background of
emphysema

Fig. 7.6, a sagittal view
again shows the dependent
(given the supine position
of the body) fluid
hypostasis of the lung
(arrows) on a background
of emphysema
direction of the gradient may reflect the position of the body at death (Figs. 7.8, 7.9,
7.10, and 7.11). With ground-glass changes, the vessels are seen ‘through’ the
density.
In some cases, the lung parenchyma may be partially collapsed at the bases.
Factors increasing this basal lung density include passive atelectasis from small
effusions and a variable ‘pushing’ effect from the diaphragm, as intra-abdominal
organs decompose and expand against the diaphragm. Paradoxically, this same

shows left lateral
ground-glass lung density
(arrows) in keeping with
fluid hypostasis which has
settled whilst the body was
in a left lateral position

showing right postero-
lateral lung hypostasis. The
body was found lying on
the right side and then
moved supine

shows anterior lung
hypostasis (arrows) in a
body found face down.
Vascular gas and
pneumorachis are noted,
resulting from
decomposition
abdominal expansion may occasionally push the sternum ventrally and result in
lung volumes that apparently increase with decomposition [3].
In order to improve the diagnostic quality of PMCT for lung pathology, tech-
niques for mechanically expanding the lungs by means of external ventilation have

shows both anterior and
posterior hypostasis of the
lungs (arrows) in a body
found face down and then
turned supine. There is a
small right and tiny left
pleural fluid collection

windows shows bilateral,
small volumes of pleural
fluid (arrows), commonly
seen post mortem and
considered to be normal
been described [4]. These should be considered where feasible, although this
technique is not within our practice.
Pleural Spaces
Early post mortem changes include the appearance of small volumes of pleural fluid
(Figs. 7.4, 7.11, and 7.12) which increase slightly in volume over the first few days
[5]. These should usually be considered as normal. Any large or asymmetric effu-
sions should raise the suspicion of infection, traumatic or neoplastic pathology.
Apparent large pleural fluid collections should be carefully reviewed, as they can be
difficult to separate from the similar density of densely consolidated lung bases,
when fluid also obliterates the expected air bronchograms (Figs. 7.13 and 7.14).
Pneumothoraces are easily detected on PMCT and are usually associated with
advanced decomposition (Fig. 7.15). They are also sometimes seen following car-
diopulmonary resuscitation (CPR) attempts with or without rib fractures (Fig. 7.16,
see also Chap. 11).

initially suggests a large
right and moderate left
pleural effusion (arrows)

Fig. 7.13d, careful
inspection of the soft tissue
windows shows subtle
hyperdense vessels in the
right lung base (arrows)
indicative of right lower
lobe consolidation. Note
that an expected ‘air
bronchogram’ appearance
is obscured by fluid in the
distal airways

showing bilateral
secondary to advancing
decomposition (significant
volume of gas seen
throughout the body
tissues)

the right anterior chest wall
on bone windows shows a
pneumothorax secondary
to rib and costal cartilage
fractures (arrows)
sustained during chest
compressions as part of
cardiopulmonary
resuscitation attempts

windows shows normal
post mortem layered
separation of the blood
components in the heart
and great vessels (arrows)
Mediastinal Vessels
Early post mortem changes include the sedimentation or ‘layering’ of blood in the
heart and great vessels (Fig. 7.17) due to the separation of cellular blood compo-
nents (erythrocytes, leukocytes and platelets, below the plasma) resulting in a
‘fluid–fluid’ level [6]. This pattern of sedimentation is notable in the main pulmo-
nary arteries and the aorta, but it is not the only recognised post mortem appearance.
A heterogeneous appearance representing normal post mortem clot formation is
possible; this may be more pronounced in cases with a longer agonal period.
The aortic wall often appears noticeably ‘hyperdense’ on PMCT compared to
clinical imaging (Fig. 7.18). This is thought to reflect the lack of movement artefact
of the wall itself and a relatively lower density of its contained, often separated,
blood products. The aorta is sometimes partly collapsed or crumpled (Fig. 7.19),
and if severe enough it may be difficult to assess for aneurysms or dissection flaps.

the mediastinum on soft
tissue windows shows the
normal hyperdensity of the
motionless aortic wall on
PMCT (arrows)

the mediastinum showing a
slightly collapsed
ascending aorta,
considered to be normal
post mortem, with normal
hyperdense wall
Acute Airway Obstruction
This term usually refers to blockage of the airway, between the pharynx and the
bifurcation of the trachea. Choking, with obstruction of the upper aerodigestive
tract, is also considered in Chap. 6. This obstruction leads to hypoxia, although
death may be caused by neurogenic cardiac arrest [7].
In sudden death cases, the obstructing item is typically a food bolus. On PMCT,
this may be difficult to separate from commonly regurgitated stomach content or
may even be partially obscured by dental streak artefact [8]. Upper airway obstruc-
tion is often supported by a history of choking and/or background neurological
disorder, and the diagnosis may be reached when there is a discrete, heterogeneous
‘mass’ within the upper airway or heterogeneous debris filling the trachea and main
bronchi (Figs. 7.20, 7.21, and 7.22). Lung findings may also be present and indi-
rectly support the diagnosis of aspiration (Figs. 7.23 and 7.24).

density debris in the
trachea and main bronchi
(arrows), extending into
the peripheral bronchi of
the lower lobes. There was
a witnessed episode of
choking in a patient with
dementia and known
swallowing difficulty

the chest of an epileptic
person found deceased at
home with food in the
mouth, there is similar
heterogeneous debris in
both the trachea and the
oesophagus

mediastinum again shows
similar heterogeneous
debris in the airways
(small arrows) and the
oesophagus (large arrow)
Chest Trauma
A wide range of thoracic traumatic injury may be readily demonstrated on PMCT,

including chest wall injuries, haemothorax/pneumothorax, lung and cardiac injury
[9] (Figs. 7.25, 7.26, 7.27, and 7.28).
On routine PMCT, it can be difficult to determine the exact origin of haemor-
rhage when there is extensive haematoma or haemothorax obscuring the anatomy.

Fig. 7.22, an axial view of
more clearly shows the
nodular and ground-glass
density lung changes,
bilateral but more at the
right base and consistent
with aspiration

the neck and chest on soft
tissue windows showing
mixed density debris in the
left main bronchus (large
arrow) and left basal
consolidation, in a patient
with dementia and known
recurrent aspiration
pneumonia. Note also
similar debris lodged in the
left pyriform fossa (small
arrow) which could be due
to swallow dysfunction

shows a left costal cartilage
fracture (arrow) following
a road traffic collision.
These are also sometimes
seen following CPR
attempts

windows shows large,
bilateral haemothoraces
(with separation of blood
products due to hypostasis)
following a fatal thoracic
crush injury

shows displaced left rib
fractures, a large left
pneumothorax and
mediastinal shift toward
the right following a road
traffic collision

shows right pneumothorax
and haemorrhagic
pneumatocoele (arrow)
collision
In such traumatic cases, the cause of death is generally either a great vessel or a
cardiac injury. A potential source of bleeding may be directly identified on angiog-
raphy or, more simply, by correlation ‘on the balance of probabilities’ of the mecha-
nism of trauma and likely site of injury.
Significant chest trauma often presents a constellation of findings. Lung contu-
sions may manifest as focal areas of increased lung density, akin to consolidation,
with surrounding ground-glass change. Contusions or lacerations can also lead to
venous fistulae and gas entering the systemic circulation. This may be suspected
when there is gas in the heart and systemic arteries, without other decomposition
changes. Caution should be taken with this interpretation in the setting of attempted
CPR (especially after positive pressure ventilation), or onset of decomposition, as
these can also explain the presence of vascular gas.
Rib Fractures
In a routine autopsy, the individual ribs are often not dissected/separated or closely
examined, unlike the assessment that normally accompanies forensic testing. By
contrast, PMCT offers a more thorough routine skeletal assessment and can be used
to give a broad suggestion of fracture age (acute, sub-acute or healed). It cannot
however reliably distinguish between recent ante mortem, agonal/CPR or post mor-
tem rib fractures.
Rib fractures should be described as incomplete (such as buckle or single cortex,
Fig. 7.29) or complete (Fig. 7.16). Fractures consistent with CPR attempts (antero-
lateral and bilateral) are further discussed in Chap. 11. Lateral and posterior rib
fractures suggest an alternate trauma.
Thoracic Aortic Rupture
Sudden death can be caused by aortic rupture, with rapid blood loss into the pericar-
dial sac (see Chap. 8) or pleural spaces (Figs. 7.30, 7.31, and 7.32). Aortic rupture

the right anterolateral chest
wall on bone windows
shows an incomplete rib
fracture, commonly seen
following chest
compressions/CPR

windows shows a ruptured
thoracic aorta (arrow) with
large haemothoraces. The
aortic wall is difficult to
visually separate from the
haematoma but is partially
identified by mural
calcification

windows shows a fatal
aortic rupture with massive
left side mixed density
haemothorax and
secondary mediastinal shift
toward the right

windows shows a large,
layered right haemothorax
with mediastinal shift
towards the left.
Haematoma is contiguous
with the ruptured thoracic
aorta, identified by its
partially collapsed,
calcified wall (arrow)

shows an aortic dissection
flap (arrow), outlined by
decomposition gas. There
was little history available
to correlate the finding
may be seen in the setting of hypertension and/or atherosclerosis, but other factors
predisposing the condition include dissection, aneurysm, trauma, inflammatory
conditions and connective tissue disease.
Aortic dissection, without rupture, may have the blood extend along the middle
plane of the vessel wall. This may cause sudden death due occlusion of the coronary
or carotid arteries. However, some dissections are chronic and in the absence of
appropriate acute symptoms may be judged as incidental (Fig. 7.33).
In cases of fatal aortic rupture, there may be a striking collapse of cardiac cham-
bers and great vessels due to hypovolaemia (Fig. 7.34). The exact point of rupture
may not be apparent on non-contrast PMCT, due to the large volume of adjacent
haematoma. Angiography in these cases may be helpful, for example using a tar-
geted coronary angiogram technique, if root dissection is suspected [10] with the
catheter balloon slightly higher in the ascending aorta than the predicted dissection
point. Whole-body angiography [11] and direct cardiac puncture [12] techniques

tissue windows shows a
flattened IVC (arrow) and
partially collapsed
abdominal aorta, ‘propped’
open by calcification, in a
case of a ruptured thoracic
aorta

shows complete left lower
lobe consolidation in the
setting of a clinical history
of chest infection
have also been described for the investigation of thoracic aortic rupture. However,
in the context of a non-suspicious sudden death, these additional techniques are not
deemed necessary if imaging features are consistent.
Parenchymal Opacity and Consolidation
The lungs are often difficult to confidently assess at PMCT. When post mortem
changes are minimal, and the lungs are reasonably aerated, abnormal parenchymal
findings may be quite obvious (Fig. 7.35). Asymmetric, patchy or segmental
increased density usually indicates a pathologic finding [6] (Figs. 7.36, 7.37, 7.38,
and 7.39). Consolidation and other non-hypostatic changes can be reported as seen,

shows a middle lobe
consolidation on a
background of emphysema
in a known smoker found
deceased

the chest showing
segmental consolidation
(arrows) within a region of
ground-glass density/fluid
hypostasis, in a patient
with a clinically diagnosed
pneumonia
but in isolation they are of indeterminate aetiology without a correlated history, such
as of cough, fever, known malignancy, resuscitation attempts (Fig. 7.40) or trauma.
Difficulty further arises in interpretation, as normal post mortem changes can
mask pathology such as inflammation, basal consolidation, nodules, masses or pul-
monary oedema. As such, it is possible to miss (or misdiagnose) pneumonia and/or

shows patchy consolidation
of the right lung base, in
keeping with
bronchopneumonia. As
there was absent history,
this was confirmed at open
autopsy

shows bilateral, patchy
consolidation in an
unexpected death,
therefore indeterminate at
PMCT. This was confirmed
as pneumonia at open
autopsy

windows shows dense
symmetrical peri-hilar
ground-glass opacity and
patchy consolidation
following a sudden
collapse and prolonged
hospital CPR. The lung
changes likely relate to
CPR and fluid resuscitation
but may hide underlying
pathology

the chest showing near
complete lung
opacification, considered
indeterminate in nature
without correlative history
other significant findings on PMCT [13]. Occasionally, the lungs are completely
opacified, and it is difficult to radiologically discriminate between infection,
oedema, other pathology and background normal post mortem changes (Fig. 7.41).
Pneumothorax and Pneumomediastinum
When considering pathological gas patterns on PMCT, it is imperative to exclude

decomposition as the cause, as this is a reasonably common post mortem reality.
Pathological pneumothorax and pneumomediastinum have many clinical causes,
but in the post mortem setting, these are likely to be seen following trauma, includ-
ing resuscitation attempts or mechanical ventilation. It has been described that,
rarely in a diabetic patient, a pneumomediastinum can be a feature suggestive of a
diabetic ketoacidosis (DKA), which is often fatal in cases with a low pH [14].
When a very large or ‘tension’ type pneumothorax is seen, with lung parenchy-
mal collapse, diaphragmatic depression and mediastinal shift (leading to respiratory
and cardiovascular compromise), this is considered to be a reasonable cause for
death (Figs. 7.42, 7.43, and 7.44). Nevertheless, correlation data should preferably
include a history of trauma, respiratory distress, hypoxia and/or chest pain.
Pulmonary Embolism
Acute pulmonary embolism (PE) is a common cause of a sudden death. PE should

always be considered if there are underlying risk factors, such as deep vein throm-
bosis, recent surgery, obesity, immobility or disseminated malignancy.
It is accepted that PE is a difficult PMCT diagnosis, compared to open autopsy
[15], and this is a subject in which further research is required. At open autopsy,

shows a unilateral,
right-side pneumothorax
and no significant
decomposition change. The
deceased was found in bed
with no available history

shows a large right-side
pneumothorax, considered
to be under tension owing
to marked mediastinal shift
to the left. The presentation
was of sudden collapse

shows bilateral
pneumothoraces and no
obvious underlying lung
disease. The deceased was
found at home
pathological embolus is readily appreciated as being different to normal post mor-

tem clot, which is often described as having a ‘chicken-fat’ appearance [16], a
smooth surface and vascular cast-like shape (replicating the vessel of origin). This
post mortem clot may be seen continuously from the right ventricle through into the
main pulmonary arteries [17]. At open autopsy, the pulmonary arteries are normally
dissected, both centrally and distally, along the pulmonary artery branches.
In comparison, PMCT without contrast is rather limited, as only the larger, cen-
tral pulmonary arteries are readily assessed, and the appearance of normal post mor-
tem clot is variable. A shower of small emboli into the periphery of the pulmonary
artery tissues would almost certainly evade detection. Very early scanning after
death may help avoid scan contamination from any central post mortem clot [18],
although this may only be achievable for in-hospital deaths with a local PMCT
service.
When PE is suspected from the history, the features that support the diagnosis on
PMCT include visualisation of discrete central clot/s, with an irregular appearance
or attenuation, possibly within a dilated pulmonary artery (Figs. 7.45, 7.46, 7.47,
and 7.48). This is different from the layered pattern of blood sedimentation reflect-
ing normal hypostasis. However, such normal layering can still mask peripheral
emboli (Fig. 7.49) [19].
Indirect features supportive of PE include right heart dilatation, interventricular
septal straightening or bowing into the left ventricle and wedge-shaped pulmonary
infarcts. The legs may demonstrate thigh/calf circumference disparity and/or peri-
vascular oedema, possibly indicative of deep vein thrombosis [19]. The inferior
vena cava (IVC) is often collapsed at PMCT, so an increased short-axis diameter has
been suggested as evidence of PE [20]. However, there are other potential causes for
a distended IVC such as right heart failure and acute myocardial infarction.
Visualisation of supporting findings along with an appropriate history (or suspicion)

windows shows
heterogeneous, irregular
clots in the main and left
pulmonary arteries
suggestive of PE (arrows).
This was confirmed at a
limited chest autopsy. Note
the different, simple
layered separation of blood
products in the ascending
aorta

windows shows ‘saddle
type’ heterogeneous
irregular clot in the main
pulmonary arteries (arrow),
not contiguous into the
pulmonary trunk or right
ventricle. Pulmonary
thromboembolism was
confirmed at limited open
autopsy of the chest

windows shows
heterogeneous attenuation
clot without the normal
separation of blood
products in a distended
right pulmonary artery
suggesting a pathological
filling defect (arrow). This
was confirmed as PE at
limited open autopsy. Note
that other large vessels
appear relatively collapsed
and no other cause of death evident should allow the diagnosis to be made (Figs. 7.50,
7.51, and 7.52), at least ‘on the balance of probability’. Depending on confidence
levels, a limited open post mortem examination of the pulmonary arteries might still
be considered.
Post mortem pulmonary angiography is achievable as part of whole-body tech-
niques [17], or by a peripheral contrast injection with subsequent chest compres-
sions, to opacify the pulmonary arteries [18]. These techniques are recognised to be
time-consuming and if performed routinely without a supportive history or other
supportive imaging findings may still suffer similar indeterminate results (due to
post mortem clot presence) as routine PMCT.

windows shows discrete
pulmonary arterial clots,
confirmed as PE secondary
to DVT at open autopsy.
Layered separation of
blood products in the
ascending aorta is noted,
representing normal
hypostasis

windows shows matched
horizontal layered
sedimentation of blood
products in both the aorta
and main pulmonary
arteries (arrows). PE was
not suggested by imaging,
yet was revealed to be the
cause of death at open
autopsy (performed as no
cause of death found)

windows. There is
hyperdense, asymmetric
clot distending the left
main pulmonary artery
(arrow) which was proven
to be pathological embolus
at open autopsy. The
distended main pulmonary
artery is different to the
partially collapsed aorta
which has a normal
sedimentation level

Fig. 7.50, a case of open
autopsy proven PE. Axial
view of the chest showing
a dilated right atrium
(arrows), a supportive
feature of PE on PMCT

Fig. 7.50 a case of
pathologically proven
PE. Axial view of the
upper abdomen showing a
distended IVC (arrows), a
supportive feature of PE
on PMCT
Chest Malignancy and Lymphadenopathy
A diagnosis of malignancy will often be known before death and so death certifica-
tion is normally straightforward. By contrast, incidental primary malignancy in the
lungs can be difficult to identify on PMCT, especially if there is post mortem change
or other pathology. Yet, sudden deaths are unlikely to relate to malignancy unless
associated with complications such as PE (see earlier), haemorrhage (Figs. 7.53,
7.54, 7.55, and 7.56) or airway obstruction (Figs. 7.57 and 7.58). ‘Review areas’
such as the breasts should be routinely examined for suspicious lesions (Fig. 7.59).

windows shows subtle
high-density fluid in the
trachea (arrow) consistent
with haemorrhage/clot.
The patient had evidence
of haematemesis/
haemoptysis when found
and a history of recent
weight loss

Fig. 7.53, the high-density
clot extends into both main
bronchi (arrows)

windows reveal a
cavitating lung mass in the
right lower lobe, consistent
with tumour

reveal background
emphysema and the lung
mass in right lower lobe,
confirmed to be
malignancy with vascular
invasion and haemorrhage
at open autopsy

windows showing a large,
previously unidentified
mediastinal tumour
(arrows) with right
bronchial obstruction. The
patient was intubated after
an episode of choking with
the endotracheal tube
entering the left main
bronchus

same case as Fig. 7.57
showing the large,
confluent mediastinal
tumour, seen obliterating
normal tissue planes
without the need for IV
contrast. It is causing
narrowing of the right
bronchus (arrow) and the
circular endotracheal tube
with radio-dense marker is
seen in the left main
bronchus
Disseminated neoplasia may be more easily evident. Correlation pathology find-

ings include multiple pulmonary nodules and/or suspicious bone lesions (Fig. 7.60).
If there is sufficient body fat, any axillary, aorto-pulmonary window, peri-cardiac
and peri-tracheal adenopathy may be seen.

windows showing an
irregular left side breast
mass, consistent with
malignancy (arrow). The
tiny pleural effusions are
considered normal in the
post mortem setting

showing widespread
sclerotic bony metastases
in both the thoracic spine
and sternum
Calcified nodes (and lung nodules) are often well demonstrated, potentially
pointing to a differential diagnosis of granulomatous disease, for example mycobac-
terial infection or sarcoid [21]. As with clinical imaging, the cause for any lymph-
adenopathy should be sought from the medical history.

easily reveals multiple,
bilateral calcified linear
pleural plaques (arrows), in
a known case of asbestos
exposure
Industrial/Occupational Related Lung Disease
This mainly reflects asbestos exposures, with silica and coal pathology being less
frequent nowadays. One key aspect of the PMCT autopsy in putative industrial dis-
ease is to identify any pathology to support such exposures as well as making com-
ment on the extent and severity of the pathology found [22].
Starting with asbestos disease, PMCT can readily reveal calcified pleural plaques
(Fig. 7.61), although plaques alone do not normally qualify as supporting a potential
post mortem legal claim. In isolation, plaque disease must not be stated as support-
ing asbestosis, unless there is clear fibrotic change in the scan data. Carcinoma and
mesothelioma may also be linked to asbestos exposure, with these often presenting
as mass lesions. Persistent pleural effusion and pleural fibrosis are asbestos-linked
pathologies, but are very difficult to confirm as not reflecting other disease.
Reference to any previous clinical imaging is very helpful.
The changes in coal and silica exposure generally are similar to each other, with
small and large parenchymal nodules being found. In coal exposure, there may be dust-
related emphysema without significant fibrosis. Contrastingly, silica exposure is vari-
ably linked with diffuse fibrosis and may also show calcification of mediastinal nodes.
In almost all cases, and certainly if a claim is being considered, an open autopsy
(perhaps limited to the chest) will be necessary. This open autopsy allows histologi-
cal sampling, with it being recognised that later legal claims and/or defence may run
onwards for many years! Overall, in cases of possible industrial disease, PMCT can
be helpful in planning targeted tissue sampling as well as excluding alternate
pathologies that are wholly or partly responsible for death.
Special Circumstances: Drowning
Not all bodies discovered in water have drowned. Furthermore, drowning may occur
away from obvious water sources. In the non-suspicious setting, a history of the
mode of death from drowning circumstances should be made available in order to
consider this diagnosis and guide how the radiologist considers the case.
Special Circumstances: Drowning 191
In this scenario, PMCT can be used to document findings consistent with drown-
ing and any other injuries present. It may reveal natural disease (which may really
be the underlying cause of death, with the body incidentally ending up in water).
The Royal College of Pathologists has issued guidelines on autopsy practice for
bodies recovered from water [23]. These guidelines state ‘if the history, scene exam-
ination, external examination and laboratory results as well as the [PMCT] images
support a diagnosis of drowning, then there is no reason that such a cause of death
cannot be provided, without the need for an invasive post mortem’. Indeed, many
drowning cases do not need open autopsy, although toxicology should always be
obtained when bodies are recovered from water. However, toxicology results are
rarely available at the time of scanning.
The Mechanism of Death in Drowning Cases
This is complex and multifactorial, involving aspiration, hypoxia, sudden osmolar

and electrolyte changes, pulmonary oedema and neural mechanisms (such as the
‘diving reflex’ and fear). These effects lead to apnoea, bradycardia and peripheral
vasoconstriction with the associated risk of developing a fatal arrhythmia.
Alternatively, the process of water immersion itself (especially if cold) can, in its
own right, precipitate such an arrhythmia. This mechanism cannot be appreciated
on a PMCT scan (or open autopsy), and the coronary arteries may appear com-
pletely normal.
PMCT Findings
In cases of drowning, PMCT findings are generally non-specific and may not allow
a definitive conclusion from the scan alone, especially when there is decomposition,
trauma or any animal predation. Open autopsy may also face a similar diagnostic
quandary [7]. There may however be a pattern of findings which, together with a
clear supporting history, fits the diagnosis [24]. These include the following findings:
Fluid in the Major Airways
Passive regurgitation of fluid into the major airways is common at PMCT, making
this finding alone generally indeterminate. In cases of drowning, the fluid in the
airways may have a low attenuation (water being generally less dense than respira-
tory tract secretions or regurgitated stomach content) yet this is considered unreli-
able as a sign in isolation (Fig. 7.62).
To add to potential confusion, the ingestion of water that contains sediment or
debris (such as sand, soil, shell fragments) may paradoxically result in increased
airway fluid density [25], although this presence of sediment in the sinuses, airways
and stomach may be a helpful sign of drowning if recognised as such [26].

windows shows
homogenous low-density
fluid filling the trachea
(arrow), in a case of fresh
water drowning. Note how
this is essentially no
different from Fig. 7.1
which was not a case of
drowning

shows extensive, bilateral
nodular ground-glass
densities in the lungs. This
was a case of fresh water
drowning
Occasionally, the fluid in the major airways may have a ‘frothy’ appearance with
a ‘plume’ from the nose or mouth, although this can also be seen in other circum-
stances such as drug intoxication or acute cardiac failure. In essence, the fluid pat-
tern in the major airways is highly variable and often non-specific.
Lung Findings in Drowning
A diffuse, mosaic or patchy pattern of ground-glass attenuation may be seen in

drowned lungs, sometimes with air space consolidation or nodular densities [24, 25,
27, 28] (Figs. 7.63, 7.64, and 7.65).
There may be an overall increase in lung volume due to air trapping, with a resul-
tant lower position of the diaphragm and a decrease in lung density [29]. At autopsy,
the lungs of drowned bodies may appear overinflated and, when cut, are wet with
foamy fluid, termed ‘emphysema aquosum’.
Special Circumstances: Drowning 193

showing extensive,
bilateral patchy and mosaic
ground-glass density in a
case of known drowning

shows extensive bilateral
airspace ground-glass
opacity and consolidation
in a young epileptic found
submerged in bath water
In fresh water drowning, the water is hypotonic and may readily pass into the
blood resulting in rapid haemodilution and a reduction in osmolality. The reverse
occurs in salt water, resulting in pulmonary oedema. In practice, these factors are
difficult to evaluate, given the abundant background variability in lung findings on
PMCT, yet fortunately they are also of little practical value in the non-suspicious
setting,
Fluid in the Paranasal Sinuses and Gastrointestinal Tract
Generally, more sinus fluid is seen in cases of drowning compared to non-drowning,

and fluid in the frontal sinuses (as opposed to than the dependent sphenoid sinuses)
is slightly more supportive of this diagnosis [30]. However, similar to the major
airways, the presence of sinus fluid, in isolation, is a non-specific finding. Caution
should be taken in drawing any conclusions from its presence.
Increased fluid in the stomach and gastrointestinal tract is also associated with
drowning [27]. This may be lower in density than ‘usual’ stomach content [28],
although once again this finding should be interpreted with caution. The absence of
fluid in the stomach and bowels does not exclude drowning, as death may have been
rapid (before significant ingestion of water) or may have occurred prior to
submersion [27].
Dry Drowning
In contradiction to the preceding descriptions of potential PMCT findings, and mak-

ing the exclusion of drowning almost impossible, is the concept of ‘dry drowning’.
In this situation, laryngospasm is thought to prevent water entering the major air-
ways and lungs, which consequently remain aerated. An alternate theory maintains
that the ‘dry’ airways are a consequence of fluid being absorbed rapidly through the
alveolar walls [7].
Reporting Chest Findings: Pearls and Pitfalls
First, an assessment of decomposition should be made, as this may signifi-

cantly affect PMCT interpretation. Note should also be made whether resus-
citation attempts, with or without chest compressions, were attempted—as
these may explain some findings.
Fluid in the airways/sinuses and small pleural effusions can largely be dis-
regarded, as these are common, normal post mortem appearances.
The extent and nature of any lung opacity should be described, and indica-
tion made whether the pattern suggests normal post mortem change (sym-
metrical ground-glass opacity with a dependent gradient) or pathology. When
suggested to be pathological, this should ideally correlate with other findings
or available history.
Pulmonary embolus is a difficult diagnosis to make on PMCT, although a
constellation of findings along with a supporting history may allow the diag-
nosis to be suggested on the balance of probabilities. It is however much more
difficult to exclude the diagnosis on PMCT.
Given the difficulties in interpreting PMCT (and open autopsy) in cases of
drowning, the history is paramount. It is anticipated that, in the non-suspi-
cious setting, this is already known and that there is no suggestion from the
scan otherwise. Correlation with any positive findings can be made, but one
should note that their absence does not exclude this diagnosis.
References 195
• Fluid seen in the airways is a common and non-specific finding on post

mortem imaging.
• Small pleural effusions are considered normal in the post mortem setting.
There is no haemothorax or pneumothorax.
• There is a pattern of normal sedimentation of blood in the main pulmonary
arteries, with no dilatation of the right heart or IVC to suggest a central
pulmonary embolus.
• Symmetrical, ground-glass opacification in both lungs shows a dependent
gradient in keeping with normal fluid hypostasis.
• Given the clinical details of infection, superimposed parenchymal consoli-
dation is in keeping with bronchopneumonia.
• Small anterior pneumothorax and multiple anterolateral rib fractures
judged secondary to attempted cardiopulmonary resuscitation.
• Almost complete/complete opacification of both lungs. It is not possible to
assess the lung parenchyma. Pre-existing/underlying lung pathology could
be missed by this study.
References
1. Biljardt S, Brummel A, Tijhuis R, Sieswerda-Hoogendoorn T, Beenen LF, van Rijn RR. Post-
mortem fluid stasis in the sinus, trachea and mainstem bronchi; a computed tomography study
in adults and children. J Forensic Radiol Imaging [Internet]. 2015;3(3):162–6. https://linking-
com/10.1007/978-3-319-27022-7_13.
4. Rutty GN, Morgan B, Germerott T, Thali M, Athurs O. Ventilated post-mortem computed
tomography—a historical review. J Forensic Radiol Imaging [Internet]. 2016;4:35–42. https://
5. Hyodoh H, Watanabe S, Okazaki S, Mizuo K, Inoue H. Postmortem computed tomography
findings in the thorax. J Forensic Radiol Imaging [Internet]. 2014;2(2):100. https://linkinghub.
book/9780340972533.
8. Iino M, Hayakawa H, Kobayashi T, Shiotani S. Asphyxia from choking on a piece of persim-
mon. J Forensic Radiol Imaging [Internet]. 2015;3(2):139–40. https://linkinghub.elsevier.com/
retrieve/pii/S2212478014001257.

postmortem changes and pathologic Spectrum of findings. Curr Probl Diagn Radiol [Internet].
10. Roberts ISD, Benamore RE, Peebles C, Roobottom C, Traill ZC. Diagnosis of coronary
artery disease using minimally invasive autopsy: evaluation of a novel method of post-mortem
coronary CT angiography. Clin Radiol [Internet]. 2011;66(7):645–50. https://linkinghub.else-
vier.com/retrieve/pii/S0009926011000675.
11. Grabherr S, Grimm J, Dominguez A, Vanhaebost J, Mangin P. Advances in post-mortem
CT-angiography. Br J Radiol [Internet]. 2014;87(1036):20130488. http://www.birpublica-
tions.org/doi/10.1259/bjr.20130488.
12. Zhou S, Wan L, Shao Y, Ying C, Wang Y, Zou D, et al. Detection of aortic rupture using
post-mortem computed tomography and post-mortem computed tomography angiography
by cardiac puncture. Int J Legal Med [Internet]. 2016;130(2):469–74. http://link.springer.
com/10.1007/s00414-015-1171-9.
13. Sonnemans LJP, Kubat B, Prokop M, Klein WM. Can virtual autopsy with postmortem CT
improve clinical diagnosis of cause of death? A retrospective observational cohort study in a
Dutch tertiary referral centre. BMJ Open [Internet]. 2018;8(3):e018834. http://bmjopen.bmj.
com/lookup/doi/10.1136/bmjopen-2017-018834.
14. Yamamoto T, Hayashi T, Murakami T, Hayashi H, Murase T, Abe Y, et al. Postmortem imaging
identified pneumomediastinum in two cases of diabetic ketoacidosis. J Forensic Radiol Imaging
16. O’Donnell C, Woodford N. Post-mortem radiology—a new sub-speciality? Clin
S0009926008002122.
17. Burke MP, Bedford P, Baber Y. Can forensic pathologists diagnose pulmonary thromboem-
bolism on postmortem computed tomography pulmonary angiography? Am J Forensic Med
Pathol [Internet]. 2014;35(2):124–31. https://journals.lww.com/00000433-201406000-00013.
18. Pichereau C, Maury E, Monnier-Cholley L, Bourcier S, Lejour G, Alves M, et al. Post-
mortem CT scan with contrast injection and chest compression to diagnose pulmonary embo-
lism. Intensive Care Med [Internet]. 2015;41(1):167–8. http://link.springer.com/10.1007/
s00134-014-3520-4.
19. Ampanozi G, Held U, Ruder TD, Ross SG, Schweitzer W, Fornaro J, et al. Pulmonary
thromboembolism on unenhanced postmortem computed tomography: feasibility and find-
ings. Leg Med [Internet]. 2016;20:68–74. https://linkinghub.elsevier.com/retrieve/pii/
S1344622316300293.
20. Mueller SL, Thali Y, Ampanozi G, Flach PM, Thali MJ, Hatch GM, et al. Distended diam-
eter of the inferior vena cava is suggestive of pulmonary thromboembolism on unenhanced
post-mortem CT. J Forensic Radiol Imaging [Internet]. 2015;3(1):38–42. https://linkinghub.
22. Osborn M, Lowe J, Attanoos R, Gibbs A. Guidelines on autopsy practice. Industrial/occu-
pational-related lung disease deaths including asbestos [Internet]. The Royal College of
Pathologists, London; 2017. https://www.rcpath.org/uploads/assets/527983bd-2820-46
0d-92409b48da8928b1/Industrialoccupational-related-lung-disease-deaths-including-
asbestos.pdf.
References 197
23. Osborn M, Taylor M, Whibley M, Lawler W, Grieve J, Hamilton S. Guidelines on autopsy

practice: autopsy for bodies recovered from water [Internet]. The Royal College of
Pathologists, London; 2018. https://www.rcpath.org/uploads/assets/a0eab7db-454b-4556-
b9961ecfd8356307/Guidelines-on-autopsy-practice-Autopsy-for-bodies-recovered-from-
water.pdf.
24. Christe A, Aghayev E, Jackowski C, Thali MJ, Vock P. Drowning—post-mortem imaging find-
ings by computed tomography. Eur Radiol [Internet]. 2008;18(2):283–90. http://link.springer.
com/10.1007/s00330-007-0745-4.
25. Bolliger SA, Ross S, Marino L, Thali MJ, Schweitzer W. Shell fragment aspiration seen at
post-mortem computed tomography indicating drowning. J Forensic Radiol Imaging [Internet].
26. Raux C, Saval F, Rouge D, Telmon N, Dedouit F. Diagnosis of drowning using post-
mortem computed tomography—state of the art. Arch Forensic Med Criminol [Internet].
2014;2(64):59–75. http://www.termedia.pl/doi/10.5114/amsik.2014.47744.
27. Lo Re G, Vernuccio F, Galfano MC, Picone D, Milone L, La Tona G, et al. Role of virtopsy
in the post-mortem diagnosis of drowning. Radiol Med [Internet]. 2015;120(3):304–8. http://
link.springer.com/10.1007/s11547-014-0438-4.
28. Mishima S, Suzuki H, Nishitani Y, Fukunaga T. Usefulness and limitation of postmortem
computed tomography in bath-related death: four case reports. J Forensic Radiol Imaging
29. Leth PM, Madsen BH. Drowning investigated by post mortem computed tomography and
autopsy. J Forensic Radiol Imaging [Internet]. 2017;9:28–30. https://linkinghub.elsevier.com/
retrieve/pii/S2212478016300739.
30. Lundemose SB, Jacobsen C, Jakobsen LS, Lynnerup N. Exact volumetric determina-
tion of fluid in the paranasal sinuses after drowning. J Forensic Radiol Imaging [Internet].
of the Heart 8
Introduction
Cardiovascular disease is the main cause of death in the United Kingdom.

Consequently, careful and judicious examination of the heart, including the coro-
nary arteries, is imperative for post mortem computed tomography (PMCT) to be
considered a viable alternative to open autopsy in non-suspicious adult deaths. This
chapter is presented as an introduction to cardiac PMCT for the general radiologist,
considering possible techniques for imaging the heart and a range of findings, from
normal variants to potentially critical pathological lesions.
The role of cardiac autopsy in general is to consider the nature of any present
cardiac disease, whether it is directly related to death or part of systemic disease [1].
Further considerations are whether the cardiac disease is inherited (important for
surviving relatives) or possibly related to non-natural pathology (e.g. illicit drug use).
With regard to imaging, a ‘routine’, non-contrast PMCT can certainly assess the
contents of the pericardial sac, consider coronary, valvular and other cardiac calci-
fications and in many cases estimate the heart size. However, it cannot assess the
lumens of the coronary arteries and is limited in the assessment of myocardium or
other soft tissue abnormalities. To enhance arterial visualisation, techniques have
been developed for targeted post mortem coronary angiography (in this chapter
referred to as PMCTA). PMCT with PMCTA can potentially provide a cause of
death in up to 92% of selected cases [2] and may reduce the number of open autop-
sies needed by up to two-thirds [3]. Despite these promising figures, it is important
to understand and appreciate that PMCT assessment of the heart still has
limitations.
Clinical cardiac radiology is a specialised field. Indeed, unless the reporter is
trained in such techniques of imaging, reporting comparable cardiac PMCT find-
ings and coronary angiography may be challenging. As with the other body tissues
in the post mortem setting, PMCT may be confounded by the additional consider-
ations of tissue autolysis and decomposition. However, careful and reasoned

Switzerland AG 2021
https://doi.org/10.1007/978-3-030-70829-0_8
200 8 Post Mortem Computed Tomography of the Heart
consideration of the imaging should still allow the general radiologist, or patholo-
gist, to appreciate the range of cardiac normality and many pathologies.
Cardiac Disease as a Cause of Death
Despite the prevalence of cardiac pathology in the population (principally ather-

oma, hypertensive cardiac disease and valvular pathologies), proving death is due to
cardiac pathology can be less than straightforward. Indeed, one must recognise that
many people die with cardiac disease present, but have succumbed from non-car-
diac causes. It is too simplistic to attribute death to coronary disease, simply because
there is coronary atheromatous calcification. One must consider the history and
exclude pathology of a non-cardiac disease group if one desires to define death as
reflecting cardiac processes.
It is specifically appreciated that cardiovascular disease is most common in those
over 50 years, even when they have no preceding cardiovascular disease symptoms
or relevant medical history. Sadly, many cases have cardiac sudden death as the first
presentation of their disease. Such deaths can occur during exercise, at rest or even
whilst asleep.
The history given may simply be that of sudden death without antecedent symp-
toms. However, other corroborate data such as shortness of breath, palpitations and/
or central angina-like chest pain is useful. Other symptoms may include ‘indiges-
tion’, atypical (neck/abdominal) pain, peripheral oedema and fatigue or symptoms
such as these several days earlier.
In the PMCT arena, cases involving a traumatic death may require special con-
sideration, as cardiac pathology may have played a part in events leading up to the
death, such as a loss of consciousness or fall, whilst driving/operating machinery,
working at a height, etc.
Classic causes of cardiac sudden death (bold = most prevalent) include [1]:
• Occlusive coronary artery disease

• Ischaemic heart disease (myocardial infarction, zonal scarring and ventric-
ular dysfunction)
• Dissection of the coronary arteries and/or aortic root
• Valve disease (stenosis, prolapse, rheumatic scarring, infective and non-
infective endocarditis)
• Myocardial disease (myocarditis, cardiomyopathies, left ventricular hyper-
trophy in the context of hypertension, amyloid, connective tissue disease)
• Congenital arterial anomaly
• Coronary vasculitis
• Coronary spasm
• Structural congenital heart disease (e.g. septal defect, Fallot’s tetralogy, etc.)
• Cardiac tumour (myxoma, sarcoma, etc.)
• Conduction system pathologies (e.g. Brugada syndrome etc.)
• Cardiac trauma
Autopsy Examination of the Heart: The Pathologist’s Perspective 201
Some of these entities can be demonstrated on PMCT, whereas others will evade
radiological confirmation. It is consequently important that the radiologist appreci-
ates the limitations of the PMCT and the possible need for focused open autopsy
examination.
Image-guided cardiac biopsy may enhance diagnostic rates by providing histo-
logical evidence, with minimal body interaction. This is potentially pertinent to
cases of myocardial infarction and myocarditis [4] but is neither widely available
nor practiced. When tissue is required, open autopsy focusing on the heart (i.e. chest
limited) is probably much easier, with PMCT being used to limit the need for the
dissection of other cavities.
utopsy Examination of the Heart: The Pathologist’s

A
Perspective
The open autopsy examination of the heart is generally accomplished at the same
time as the examination of the chest, incorporating the lungs, mediastinum and
other thoracic contents (see previous chapter). Indeed, without considering the other
tissues, it is often difficult to make a value judgement as to any cardiac disease liable
to cause death.
When open autopsy of the heart is required after PMCT, our local pathologists
tend to favour chest-limited examinations, focusing upon the heart and lungs, often
with trans-diaphragm sampling of the liver and one kidney, in order to provide max-
imum information with minimal tissue dissection/disruption. This approach appears
particularly appreciated by families, as minimally invasive.
The heart can be examined in situ by opening the pericardium and directly
inspecting the organ and its connections, but is usually removed together with the
mediastinum, lungs and related tissues. This may be performed either in isolation,
or as grouped tissues with abdominal and/or throat structures.
Whichever dissection solution is applied, the heart is extracted from the pericar-
dium after finger palpation of the opened pulmonary artery has been accomplished
(in order to exclude pulmonary embolism). Transection of the aorta and pulmonary
artery followed by the pulmonary veins and venae cavae serves to free the heart
from the mediastinum.
The heart is considered initially in terms of possible congenital architectural
variation by looking closely at the great vessels, the coronary tributary pattern,
appendages and chambers. If these are abnormal, it may prompt specialist patholo-
gist referral for congenital or inherited heart disease issues.
If the coronary artery architecture is appropriate, then 3 to 5 mm slices along the
length of the coronary arteries are undertaken to exclude thrombosis and to consider
the level of atheromatous stenosis (generally graded in 10% increments). The degree
of coronary calcification and eccentricity of plaque is also often described as part of
the coronary system macroscopic review.
Clearly, devices alter dissection. If there are coronary stents or heavy coronary
calcification, then the coronary arteries may require removal en bloc with
subsequent decalcification for full assessment. Likewise, calcified valves may

require decalcification. Furthermore, the placement of pacemaker wires may require
a modification of the dissection protocol. If there are prosthetic valves present, then
these should be approached in a schematic way such that the inferior and superior
surfaces of the valve tissues are inspected for sepsis, thrombosis or misalignment.
The cardiac chambers are normally considered initially by three transverse slices
across the ventricles, ending at a mid-level ventricular slice, thereby showing the
arrangement of the right and left ventricles and any scarring, fibrosis or acute infarc-
tion. This examination also demonstrates cardiac tissue asymmetry and endocardial
thrombosis.
The ‘top’ part of the heart (i.e. upper part of the ventricles together with the atria)
is examined ‘along the flow of blood’. In short, the technique starts by opening the
right atrium, with incision posteriorly into the right ventricle, exiting along the front
of the heart through the pulmonary outflow tract, valve and pulmonary artery. Next,
one repeats a similar set of incisions to demonstrate the left atrium, running down-
wards into the left ventricle. The examination then runs across the front of the heart
upwards into the left ventricle outflow tract, across the valve and into the aorta. This
is the common standard, although different variants of examination exist which may
reflect the pathology under consideration [5].
Consideration of septal defects, calcification of valve tissues, along with stenosis/
regurgitant features is normally part of any considered cardiac tissue examination.
As part of the examination, the heart mass can be measured and compared to
standard heart weight tables. Furthermore, one can measure items of the internal
substructure (valve circumferences, wall thickness, chamber diameter and so on).
This may give insight into the status of heart before death.
Sampling for histology is normally guided by the fact that microscopic detail
may provide additional information pertinent to the cause of the disease and thereby
the death of the individual. Histology sampling is normally targeted, with small
numbers of tissue specimens in most cases, through to extensive sampling in those
with sudden deaths involving the young. Occasionally, bacterial and virology stud-
ies can be simultaneously taken to consider overtly septic processes and the possi-
bility of viral myocarditis, respectively. Deaths involving possible inherited
cardiomyopathies normally prompt sampling of the spleen so that appropriate DNA
extraction can take place for late considerations [6].
Post Mortem CT of the Heart: Specific Additional Techniques
A familiarity with the range of non-contrast cardiac PMCT appearances is required

in all cases, discussed further on in this chapter, as well as an understanding of when
further techniques such as coronary calcium scoring, PMCTA or recommendation
of open autopsy may be helpful.
In this section, these common techniques for enhancing the imaging assessment
of the heart are presented, starting with the consideration of a calcium score (from
Post Mortem CT of the Heart: Specific Additional Techniques 203
the non-contrast images) and then targeted coronary angiography. Some centres
may choose to apply these techniques routinely to all PMCT cases, whereas others
will apply them on a case-by-case basis.
Calcium Scoring of the Coronary Arteries
A non-contrast PMCT may be supplemented with an objective assessment of the

burden of coronary calcification by means of calculating a ‘calcium score’, some-
times referred to the Agatston score (as below) depending on the technique used.
This numerical score, as in clinical practice, can be used as a proxy for the presence
of coronary atheroma and thereby also for stenosis. Coronary artery calcification is
arguably more clearly and easily detected on PMCT, compared to open autopsy [7]
(Figs. 8.1 and 8.2).
In the setting of an appropriate clinical history (such as chest pain with sudden
collapse/death), a high calcium score can be used to suggest a relevant, significant
coronary stenosis. Death resulting from a fatal arrhythmia or myocardial infarction
is thus inferred on the coronial ‘balance of probabilities’, assuming the lack of alter-
nate competing fatal pathology or confounding features.
With appropriate software (semi-automated), the score is rapidly calculated and
is deemed accurate, given the lack of motion artefacts. However, it is, to some
degree, operator dependent, and care must be taken not to include aortic root, valvu-
lar, pericardial or mediastinal calcifications. Coronary stents also present a chal-
lenge as they can sometimes be difficult to visualise, especially if nestled within
calcific plaques, or if there is in-stent calcification. If not recognised, and excluded,
they will significantly elevate the score due to their high density. Of particular value,

coronary calcification in
the left anterior descending
coronary artery (arrow),
the individual vessel score
was 1349

Fig. 8.1, there is also
extensive coronary
calcification in the
circumflex and right
coronary arteries (arrows),
individual vessel scores
were 939 and 1815
respectively
the calcium score may be calculated even in the presence of moderate decomposi-
tion, although the exclusion of other competing and potentially fatal pathologies
may not be possible in decomposed bodies.
In the authors’ centre, the calcium score is calculated using the Agatston method
[8]. Simplified, the score is based on the density of calcification multiplied by its
area, taken from a non-contrast axial study at 3-mm slices. It is calculated for each
main artery and given as an overall total. The total score, originally derived from
those in clinical settings (i.e. the living) can be categorised with regard to the risk of
coronary artery disease (CAD) as follows [9]:
0—Very low probability of significant CAD, generally <5%

1–10—Very unlikely probability of significant CAD, <10%
11–100—Mild or minimal coronary stenoses likely
101–400—Non-obstructive CAD highly likely, obstructive disease possible
>400—High likelihood (≥90%) of at least 1 “significant” coronary stenosis
Whilst the categories suggest a score of >400 to be highly significant, for con-
text, scores may occasionally be seen in native vessels (without stents) of more
than 3000!
It is to be appreciated that, whilst accurate for calcification, soft plaque is not
identified by this method. It has been shown, in the living, that half of the patients
undergoing assessment for high coronary risk, atypical symptoms or abnormal
stress test who had a normal calcium score actually had non-calcified plaque on
coronary angiography. Indeed, it was found that 1.5% of these cases had a severe
stenosis! [10]. Thus, a low or even normal calcium score does not exclude a death
from high-grade coronary stenosis. Indeed, it is recognised that potentially unstable
or vulnerable plaque is often characterised histologically by a high lipid content
rather than calcification.
On a practical level, pathologists vary considerably in their threshold for ‘signifi-

cance’, modified by the clinical setting and case in front of them but will not usually
accept deaths as coronary pathology in those with scores of 400 or less, despite the
ante mortem correlates.
In summary, a high calcium score may indicate/support, but does not confirm, a
coronary death. Conversely, a low score does not exclude coronary death. For these
reasons, careful interpretation of the wider picture, focusing on the circumstances of
the death, any reported symptoms and the prior medical history is required.
Progression to PMCTA, a limited or full open autopsy may be the appropriate
solution.
Post Mortem CT Coronary Artery Angiography
To enhance the information and confidence provided by non-contrast scans, the

coronary artery lumens may be directly assessed with PMCT angiography (PMCTA).
Whilst the techniques of injecting the contrast may vary to the clinical setting, the
results are assessed in much the same way as clinical cardiac CT angiography.
Several methods of coronary PMCTA of varying complexity have been described
[2, 4, 11, 12]. The method familiar to the authors involves cannulation of the left
carotid artery via a small incision (importantly thus making the examination ‘mini-
mally invasive’ rather than ‘non-invasive’) [11]. A ‘male length’, three-way urinary
Foley catheter is passed through to the ascending aorta so that the tip lies above the
aortic valve and coronary ostia. The 30 mL catheter balloon is inflated with water
(Fig. 8.3) to achieve a seal with the aortic wall. Priming of the catheter with water
before inserting may reduce the amount of air introduced and flushing before con-
trast injection can help dislodge post mortem clot from the aorta. Dilute (5%)
Fig. 8.3 Coronal

maximum intensity
projection of the chest
showing an inflated urinary
catheter balloon in the
ascending aorta during
targeted coronary
PMCTA. Contrast is
present in the catheter
lumen, aortic root,
coronary arteries and
myocardium
iodinated contrast is injected manually, steadily in volumes of approximately

100–200 mL at a time. Radiological scanning then takes place in a standard manner.
If the aortic valve is competent, then flow of injected contrast material into the
left heart is limited, and contrast fills the coronary arteries resulting in a targeted
angiogram. However, if there is aortic regurgitation, then the injection pressure may
not be sufficient to opacify the full length of each vessel. Contrast should not nor-
mally travel into the right side of the heart unless there is a structural defect, reflect-
ing myocardial rupture after infarction, congenital deformity or trauma (Fig. 8.4).
Another potential cause of a non-diagnostic study is the inadvertent protrusion of
post mortem clot from the aorta into the coronary arteries [7]. If a coronary artery is
not opacified with contrast from its ostium, then ‘pseudo-occlusion’ by this normal
post mortem clot (rather than pathological thrombus) is likely. This is most often
seen in relation to the left coronary ostium as, in the conventional supine position, it
lies more dependently (Fig. 8.5). Turning the patient prone, or right lateral decubi-
tus, can be helpful in clearing the clot and confirming vessel patency. Non-ostial
occlusions are much more likely to be genuine thrombus, as it is rare to find intra-
coronary post mortem clot.
Negative contrast (usually air) is sometimes inadvertently injected via the catheter
or may already be present due to decomposition, when it is usually found in the non-
dependent RCA (Figs. 8.6, 8.7, and 8.8). Some methods of angiography routinely
include an air-contrast run [2]. A combination of negative and positive contrast is
quite acceptable for manual luminal assessment but can make using vascular analysis
software more cumbersome—as the air is not tracked automatically. If necessary,
poor opacification or air in the RCA can be overcome by turning the patient prone
[11, 12]. However, this adds manual handling issues and time to the examination.

chest during coronary
PMCTA after a crush
injury. This shows contrast
in the pulmonary artery
(arrow) which is not a
normal finding. Whilst this
can arise in the setting of
congenital septal defects,
in this case open autopsy
confirmed a traumatic
ventricular septal rupture.
Large bilateral traumatic
haemothoraces are also
evident, yet no rib fractures
were seen in this young
adult
Fig. 8.5 Axial PMCTA

view of the heart shows
contiguous clot extending
from the dependent aorta
into the left coronary artery
ostium (arrow) and main
stem, likely protruded post
mortem clot from the
aorta—although there is
distal vessel contrast
implying incomplete
‘pseudo-occlusion’

mediastinum on soft tissue
windows shows non-
dependent air at the right
(arrow). There is also
normal blood
sedimentation in the aorta
and normal positions of the
right and left coronary
ostia
Once contrast has filled the arteries, a ‘myocardial blush’ of contrast is some-
times seen, reflecting capillary backfilling and/or interstitial leakage. This may aid
in revealing segments of non-perfused, infarcted myocardium (compare Figs. 8.8
and 8.9). However, if excessive, it can also make discrimination of the arteries more
difficult.

aortic root following
PMCTA contrast injection.
The normal, anteriorly
positioned right coronary
origin is filled with air,
inadvertently injected.
Note is made of a normal
tricuspid aortic valve
without any calcifications
Fig. 8.8 Oblique PMCTA

view of the heart to show
the right coronary artery
(arrows) in the anterior
atrioventricular groove.
The RCA contains contrast
and locules of
decomposition gas. There
is also gas in the cardiac
chambers from
decomposition and a left
ventricular myocardial
blush of contrast
The choice of when to undertake angiography is largely down to local agreement

but could, for example, be limited to cases of suspected coronary death, where the
Agatston score is below a defined level, such as the clinical calcium score threshold
of 400 [13]. This is because, ‘on the balance of probabilities’, a score of over 400
(in the absence of another cause of death or conflicting history) would support a
cause of death due to coronary artery disease, without requiring further investiga-
tion such as angiography [13]. In addition, as the coronary calcium score further
rises, coronary luminal visualisation becomes more difficult due to calcium bloom-
ing artefacts.
Fig. 8.9 Short axis view

of the heart (mid
ventricles) following
PMCTA shows a well
demarcated perfusion
defect of the lateral
segments of the left
ventricle (arrows)
suggesting infarct in the
distribution of the left
circumflex artery. The LCx
artery demonstrated
occlusive thrombus on
PMCTA (not shown in this
image) and the setting was
sudden death following a
few days of heartburn type
symptoms and
hypertension
One should however be mindful that the pathologist’s approach or thresholds for
considering PMCTA may differ. If the pathologist is the one sanctioning the test,
then he/she may prefer to go straight to open cardiac autopsy review for uncertain
cases. Cases where the history clearly suggests another unrelated cause of sudden
death (such as suicide or ruptured aortic aneurysm) will not usually need PMCTA. If
angiography is undertaken, it is preferable to obtain toxicological sampling in
advance to avoid any potential for contamination. Cases with an infection risk may
require exclusion from angiographic assessment. As with all specialised tests, it
should be appreciated that PMCTA increases the time, cost and invasiveness of the
overall post mortem study.
Considering now the potential results, PMCTA can provide diagnostic informa-
tion about coronary narrowing, which is at least comparable to a standard open
autopsy [12]. As with open autopsy, a 70% reduction of intra-luminal diameter is
generally taken as a significant stenosis, with 90% considered high grade and usu-
ally sufficient (in appropriate circumstances) to assign the cause of death as isch-
aemic heart disease [1]. Lesser degrees of stenosis (<70%) may be important, but
these must be tested against the absence of other pathology [1].
Arterial wall remodelling (a reaction to plaque formation) may be demonstrated
and perhaps more easily appreciated on PMCTA, compared to open autopsy mac-
roscopy, but is inferior to histology review. PMCTA also lacks the ability to provide
microscopic information about a stenosis (such as intraplaque rupture or haemor-
rhage), inflammation, evidence of prior vascular dissection and so on. However, it is
recognised that this microscopic assessment of atheromatous disease is not rou-
tinely performed at open autopsy—making PMCTA a viable alternative solution for
many cases.
Pericardial Sac
There is often a visible outline or trace of fluid in the pericardial sac, more obvious
than on clinical imaging due to the absence of cardiac pulsation artefact. This find-
ing is considered normal (Fig. 8.10). The pericardium should be thin and smooth.
Basic Coronary Anatomy
It is important to be familiar with normal and variant coronary anatomy [14] if one
desires to interpret non-contrast scans as well as PMCTA. In the aortic root, there
are right and left sinuses of Valsalva from which the coronary arteries arise
(Figs. 8.11 and 8.12). There is also a third, ‘non-coronary’ aortic sinus situated right
posterior.
The right coronary artery (RCA) travels anteriorly to the right of the pulmonary
artery and along the anterior atrioventricular groove and has branches named acute
marginals.
The left coronary artery arises in the form of the left main stem (LMS). It is a
short trunk (generally 5–10 mm) that passes between the left atrial appendage and
the pulmonary trunk.
The LMS bifurcates into the left anterior descending (LAD) and left circumflex
(LCx) arteries. Occasionally, the LAD and LCx arise separately from the left coro-
nary sinus, with this being generally considered to be a normal/benign variant,

normal thin pericardium
(arrows), outlined by fat

windows shows the normal
position of the right
coronary artery (arrow),
arising from a partially
collapsed, normal post
mortem aorta and
travelling anteriorly
through the right
atrioventricular groove

Fig. 8.11, the normal
position of the left main
coronary origin is shown
with the proximal left
anterior descending and
circumflex arteries also
demonstrated (arrows).
Multiple foci of vessel
calcification are seen on
this non-contrast study
rather than a pathological anomaly. Sometimes the LAD trifurcates with an anoma-
lous artery arising between the LAD and the LCx being called a ramus intermedius.
The LAD travels in the anterior interventricular sulcus, and its branches are
called the diagonal arteries. The LCx travels in the posterior atrioventricular groove,
and its branches are the obtuse marginals.
‘Coronary dominance’ is a term which denotes the artery that supplies the distal
posterior descending artery (PDA) and the posterolateral branch (PLB), which in
turn supply the infero-septal and inferior aspect of the left ventricle. There is right
side dominance in 80–85% of adult cases [14], with the other, non-dominant artery
expected to be smaller in calibre. All of the coronary vessels should demonstrate a
smooth tapering from their origins (Figs. 8.13 and 8.14) if they are normal.
Fig. 8.13 PMCTA curved

reconstruction of a normal
left anterior descending
artery from its origin
showing a normal, smooth
tapering of the vessel and
no stenosis
Fig. 8.14 PMCTA curved

reconstruction of a normal
right coronary artery with
vessel tracking software
‘centre line’ and points of
reference (A and B) which
allow calculation of
relative stenosis between
different points
Heart Walls and Contents
Within the heart, there may be a normal fluid–fluid level from the sedimentation of
blood, as well as intracardiac gas largely secondary to resuscitation attempts or
decomposition. A relative dilatation of the right heart is a common observation on
PMCT [15, 16], caused by the pooling of blood on the right side reflecting equalised
intravascular pressures when the circulation ceases.
It is difficult to accurately measure myocardial thickness on routine PMCT, since
inadvertent inclusion of papillary muscle and epicardial fat can lead to over-
estimation. The left ventricle is usually the chamber of most pathological interest.
Measurement of wall thickness can be made by reconstructing images into a cardiac
short-axis view (Fig. 8.9) and then measuring perpendicular to the endocardial sur-
face. Such measurement may be easier when contrast outlines the cardiac chambers
(such as when during angiography there is reflux of contrast through the aor-
tic valve).
Once measured, the meaning of the figure obtained needs careful consideration.
Bodies scanned very soon after death have the heart walls can appear artefactually
thicker on PMCT compared to ante mortem CT due to rigor mortis [17]. The ven-
tricular thickness will also vary depending on the cardiac phase (systole or diastole)
at the time of death, further reducing confidence in the measurement. Other factors
such as age and ‘athleticism’ of the deceased may factor into the assessment, and so
a ‘one-size-fits-all’ approach to defining a normal post mortem myocardial thick-
ness is inadvisable.
A method for estimating heart weight from PMCT, based on measuring the left
ventricular circumferential area, has been proposed [18], although this is not vali-
dated in the presence of decomposition or trauma and is not currently undertaken in
our own practice.
As so many factors are at play, it is very difficult to be certain of the value of
measuring LV thickness or estimating heart weight at the present time. There is cur-
rently more research needed into the application of post mortem CT cardiac mea-
surements, particularly for the post mortem period of several days to weeks
after death.
Pericardial Disease
ericardial Effusion and Calcification

P
More than a normal trace of fluid (i.e. a true effusion, Fig. 8.15) may indicate peri-
cardial disease or more likely reflect other systemic disease. If large, the effusion
may be linked with poor cardiac function, and flattening of the right ventricle sug-
gests a tamponade effect before death (Fig. 8.16). However, in the post mortem
setting one must also consider that flattening of the right ventricle may also be seen
due to flaccidity of the heart muscle during decomposition.

windows shows a small
pericardial effusion in a
case of known pulmonary
hypertension and lung
fibrosis

windows showing a very
large, non-haemorrhagic
pericardial effusion (large
arrows). There is subtle
right ventricular flattening
(small arrow), suggesting a
degree of tamponade. Note
also pacemaker wires with
associated artefact,
traversing the right atrium
Pericardial calcification is easily identified on PMCT and may be seen at sites of

previous infection (viral or tuberculous), inflammation, intervention/surgery or
trauma. The significance is higher if it is visibly constrictive or associated with a
large effusion.
Haemopericardium
A haemopericardium is seen as a hyperdense pericardial collection. In the absence
of external trauma, this is usually from either a ruptured aortic root (reflecting mural
dissection) or a ruptured ventricular free wall (following a recent myocardial
infarct). Rare causes of this pathology include coronary vasculitis, dissection or

injury during intervention.
There are two distinct patterns to a post mortem haemopericardium. First is a
horizontal layering of density (Fig. 8.17), similar to that commonly seen when
blood products separate in the great vessels during normal hypostasis. Second, and
fairly unique to this anatomic location is the ‘hyperdense ring’ appearance. Here,
there is a concentric ring of high-density matrix around the heart (representing clot)
in turn surrounded by a ring of low-density serum (Figs. 8.18 and 8.19). It has been
suggested that the concentric ring pattern forms when haemorrhage occurs around

windows shows a very
large haemopericardium
with layered separation of
blood (arrows indicate the
separation of components
yet the whole volume is
‘blood’). An intact aortic
root (not seen on this
image), high coronary
calcium score and history
of chest pain suggest this
to be from a ruptured
myocardial infarct

windows shows a
circumferential ring of
high-density clot (arrows)
around the heart,
surrounded by lower
density. This is
representative of a
haemopericardium. This
patient had a high coronary
calcium score (2603), neck
and shoulder pain followed
by sudden collapse,
altogether in keeping with
ruptured myocardial infarct

windows, (different case to
Fig. 8.18) shows the true
circumferential nature of
hyperdense clot in this ring
pattern of
haemopericardium
(arrows). This was a
known hypertensive patient
who suffered chest pain
and a sudden collapse

windows showing a large
haemopericardium with
combined patterns;
concentric hyperdensity
around the heart and also
layering of haemorrhage.
This suggests a
combination of ante
mortem and post mortem
haemorrhage. Note slight
flattening of the right side
of the heart anteriorly
suggesting tamponade and
anterolateral rib fractures
from CPR attempts
a beating heart, i.e. blood loss initiated prior to the time of death. The layering pat-
tern more likely forms in the post mortem phase, for example originating from
cardiac rupture secondary to chest compressions [19].
The absolute reliability of this categorisation can be debated [20], as there may
be other factors to consider. These include the presence of a coagulopathy (patho-
logical or pharmaceutical), continued post mortem oozing from a primary defect
and resuscitation injury following a true pathological aortic or ventricular rupture.
Occasionally, both patterns may be seen together (Figs. 8.20 and 8.21), suggest-
ing a combination of both ante mortem and peri/post mortem haemorrhage. The
emphasis should be on the ring pattern, as this is considered a vital reaction and
therefore relevant to the events leading up to death [19].

windows shows both
concentric hyperdensity
around the heart and
additional layered
hyperdensity within the
pericardial sac in keeping
with a combined pattern
haemopericardium.
Flattening of the right
(anterior) heart suggests
tamponade effect
Fig. 8.22 Axial view of the chest on soft tissue windows shows a combined concentric ring and
layered pattern of haemopericardium. A ring of hyperdensity extends around the aortic root (large
arrow), arch (not seen) and more subtly around the descending aorta (small arrow) in keeping with
an aortic dissection which has ruptured into the pericardial space
The origin of a haemopericardium is sometimes directly evident on

PMCT. For example, a markedly irregular aortic root with adjacent haematoma
is in keeping with aortic rupture or occasionally a dissection plane is evident
(Fig. 8.22). By contrast, ventricular wall defects resulting from infarction are
often occult unless revealed during angiography (Figs. 8.23 and 8.24). Arguably,
in the presence of a normal aortic root, further investigations may not be required
as, ‘on the balance of probability’, myocardial rupture is the most likely cause
of tamponade.
Of note, rupture of the left ventricle is most likely to be secondary to infarc-
tion. However, whilst a right ventricular rupture may reflect infarction, it may

concentric ring pattern of
haemopericardium in a
patient who was found
deceased in bed with
recent history of heartburn
symptoms. The aortic root
(not shown) appeared
normal and the coronary
calcium score was low.
Coronary angiography was
subsequently performed
(Fig. 8.24)
Fig. 8.24 Same patient as

Fig. 8.23, a targeted
coronary angiogram
confirmed an intact aortic
root (not shown) and reflux
of contrast into the left
ventricle demonstrated
extravasation into the
pericardial sac through a
free-wall defect (arrow).
This permitted diagnosis of
a ruptured myocardial
infarct
also be associated with traumatic injury, for example chest compressions during
resuscitation. Open autopsy with histology sampling would be required to
prove/refute each interpretation yet by careful consideration of the history and
imaging, interpretation can usually be made radiologically, again on the balance
of probabilities.
Fig. 8.25 Volume

rendered image of the
anterior chest wall
following a ‘clamshell
thoracotomy’ performed
on this penetrating chest
trauma patient to evacuate
haematoma in the
pericardial cavity
Fig. 8.26 (Same case as

Fig. 8.25) Coronal view of
volume of residual
haemorrhage in the
supra-pericardial recess
(small arrow) and large
soft tissue defects from the
clamshell thoracotomy
(large arrows). Multiple
locules of gas in the soft
tissues result either from
the primary trauma and/or
secondary to the
thoracotomy
In the emergency or resuscitation setting, when a haemopericardium with tampon-

ade is suspected (usually due to penetrating thoracic injury, suspected blunt cardiac
injury in a shocked patient or identified on bedside ultrasound), then aspiration or
drainage may be attempted by means of a thoracotomy (Figs. 8.25, 8.26, and 8.27).
Such extensive intervention however results in a scan that is very difficult to subse-
quently interpret.
Fig. 8.27 (Same case as

Fig. 8.25) Sagittal view of
the chest shows the large
anterior chest wall defect
from clamshell
thoracotomy (arrow).
Subsequent collapse of the
chest tissues/lungs and soft
tissue gas make
interpretation of the
underlying primary injury
difficult
Heart Size
ardiomegaly and the Cardiothoracic Ratio

C
The assessment of heart size on PMCT is not straightforward. Traditionally, the
term ‘cardiomegaly’ refers to an overall increase in size of the heart and/or the
weight of the heart as defined by the pathologist at autopsy. This matter is important,
as cardiac hypertrophy may be an important risk factor for sudden death. Thus, it is
helpful for the radiologist to give an indication of the heart size, or state when this
is not possible, for example in the setting of decomposition or traumatic disruption.
One should be aware that the heart was dynamic and may remain in a state of sys-
tolic/diastolic contracture after death, akin to rigor mortis.
In clinical practice, the commonly used radiographic assessment of cardiac size
is the cardiothoracic ratio (CTR), measured on a postero-anterior (PA) chest radio-
graph. A normal CTR is 0.5 or less, when the transverse diameter of heart is up to
50% of the inner thoracic cage diameter. The CTR can be measured on PMCT by
reconstruction of the image into the coronal plane and performing transverse mea-
surements, as if assessing plain film [21] (Figs. 8.28 and 8.29).
Fig. 8.28 Coronal image

of the chest on soft tissue
windows showing a
simplified example of how
to measure the transverse
diameters of the heart and
thorax in order to calculate
the cardiothoracic ratio, in
this example the heart size
is considered to be normal
(11.29/22.09 = 0.51). Note
however that the maximum
transverse diameters may
not always be in the same
coronal plane
Fig. 8.29 Coronal image

of the chest on soft tissue
windows showing a
simplified example of how
to measure the transverse
diameters of the heart and
thorax in order to calculate
the cardiothoracic ratio, in
this example cardiomegaly
is demonstrated
(15.41/25.79 = 0.6)
It has been suggested that a CTR of more than 0.5 (or more than 130 mm cardiac
diameter) might indicate cardiomegaly early after death [21]. This has been
described with sensitivity/specificity of 89%/71% for the CTR and 89%/93%
respectively for heart diameter. However, a body laid supine and in a post mortem
state will have some progressive heart ‘flattening’ with a corresponding potential
increase in CTR such that a normal post mortem CTR value has alternatively been
judged as 0.54 or more [22]. For a very high specificity (>95%) an even higher CTR
threshold of 0.57 may be more appropriate [23, 24], especially if scans are not
undertaken within 24h of death. In general, care must be taken not to overcall mild
cardiomegaly in the post mortem setting, with CTRs between 0.5 and 0.57 remain-
ing debatable.
Certainly, cardiothoracic ratio may not be an appropriate measurement in cases
where there are congenital variations to the thoracic cage, extremes of age or back-
ground lung pathologies, such as emphysema [25]. The heart size is difficult to
accurately measure and trust if there is visible decomposition (gaseous distension or

chamber collapse), pericardial fluid or cardiac injury. Chest wall deformity from
trauma (including from chest compressions during resuscitation) may also alter the
cardiothoracic diameters and ratio. When these confounding factors exist, it may
not be possible to make a confident measurement of heart size on PMCT.
One approach is that measurements (transverse cardiac diameter and CTR) may
be calculated from the images and reported in a factual manner that will allow the
pathologist to consider the significance in relation to the other findings and scenario.
CTR values between 0.5 and 0.57 might be considered ‘borderline’. The radiologist
should however make it clear when the values are likely to be less reliable owing to
the factors highlighted in the previous paragraph. Other attempts to enhance the
value of autopsy CTR alone include an adjusted CTR-based score (accounting for
body-mass index, age and gender), used to predict cardiac hypertrophy at PMCT,
available as an online tool [26], yet this also does not account for the confounding
factors mentioned earlier. Reference to any clinical imaging (if available) would be
extremely helpful in these borderline circumstances.
Coronary Artery Disease
oronary Artery Stenosis

C
There are a variety of pathologies liable to cause coronary lumen narrowing,
although one recognises the majority of UK cases reflect atheroma. Stenosis (nar-
rowing) of the coronary arteries may be due to soft plaque, calcific plaque or may
be mixed in nature. Any stenosis can be broadly described in terms of location along
the vessel (proximal, middle and distal) and to a quantitative degree in terms of the
degree of narrowing [27]:
• Minimal stenosis: <25%

• Mild stenosis: 25–49%
• Moderate stenosis: 50–69%
• Severe stenosis: >70%
• Occluded: 100%
A visual aid, originally developed to aid pathologists, may help when becoming
familiar with reporting degrees of stenosis [28], although most dedicated angiogra-
phy software solutions will provide a numeric assessment through vessel recon-
structions. Further morphological description of the arteries and stenotic lesions is
beyond the scope of this introductory text, and the reader is directed to clinical
cardiac radiology and pathology texts and relevant courses [29, 30].
Coronary artery stenosis can result in myocardial ischaemia, with deaths from
myocardial infarction or fatal dysrhythmia. Pathological studies have shown that
stenoses of more than 85% are linked with a risk of sudden death [31], although, as
above, over 70% is referred to as ‘severe’ [27] and in other sources, 90% is consid-
ered ‘high-grade’ [1]. It is suggested that the degree of significance (enough to
attribute cause of death, in the absence of confounding factors) be discussed and
agreed between radiologist and pathologist to ensure concordance in interpreta-

tions. From the pathologists’ perspective, severe stenosis (i.e. over 70–85%) with a
fitting history is usually considered sufficient to confirm coronary artery disease as
the cause of death. This analysis could be equally extrapolated to radiology
(Figs. 8.30, 8.31, and 8.32).
Faced however with a mild-to-moderate stenosis, on the balance of probability,
this narrowing being the cause of death is difficult to satisfy, without well-fitting
Fig. 8.30 Axial view of the heart following targeted coronary PMCTA, performed to investigate
a sudden unexpected death (found deceased). The calcium score was 92 (but all in the LAD).
PMCTA reveals severe stenosis of the proximal LAD (arrows). Note the vessel tracking software
markers through the stenosis, used to form curved reconstructions of the vessel (Fig. 8.31). The
PMCTA catheter tubing is seen in the opacified aortic root and pacemaker wires noted in the SVC

Fig. 8.30, curved PMCTA
reconstruction of the LAD
demonstrates the stenosis
to be secondary to a mixed
density plaque with a
calcific focus (arrow). This
plaque results in a segment
of severe (>85%) luminal
stenosis but there is
contrast ‘run-off’ distally
(i.e. it is not occlusive)

Figs. 8.30 and 8.31,
another curved
reconstruction at a
different rotation point
again demonstrates the
severe stenosis of the
proximal vessel (large
arrow). Note that a further,
eccentric mild mixed
plaque stenosis is now also
seen more distally (small
arrow)
circumstances and careful exclusion of alternate pathology. However, it is again

emphasised that any even severe coronary disease may be present but may not be the
cause of death.
It is also worth noting that complete vessel occlusion is not automatically the
cause of death, as fully blocked arteries are sometimes seen in autopsies with other
(non-cardiac) causes of death. Indeed, individuals may suffer from myocardial
infarction from such complete coronary occlusion and yet live on with this cardiac
disease. It therefore follows that finding complete occlusion should not be defined
automatically as the cause of death, without triangulated ante mortem data and
exclusion of alternate causes.
oronary Artery Bypass Grafting

C
Targeted coronary angiography will not readily assess cardiac bypass morphology
or patency. It is probably not meaningful to perform coronary calcium scores or
coronary angiography in patients who have had bypass grafting, as the results can-
not reliably be interpreted in the same manner as natural disease. Whole-body angi-
ography, if available, may allow assessment of the bypass vessels [12], but this is
rarely performed in the United Kingdom owing to the additional time and financial
resource required.
Open autopsy examination of bypass vessels is appreciated as complex to dis-
sect, due to the variable anatomy, fragile nature, local fibrosis and post-surgical
scarring. One has to consider whether such assessment (open or radiological) is best
for to the overall post mortem examination, if one is assessing a case on the balance
of probability. A history of cardiac bypass surgery indicates a significant back-
ground risk from heart disease, even with good previous revascularisation outcomes.
Again, considering an appropriate history, in the absence of confounding features
on external examination and imaging, ischaemic heart disease may be considered
the probable cause of death.
Myocardial Bridging
A ‘myocardial bridge’ is defined as an anomalous course of a major coronary artery,
commonly the left anterior descending, where there is overlying myocardium (the
‘bridge segment’) contrasting with a normal epicardial coronary artery position.
The importance of this finding is controversial, as myocardial bridges are common
(although prevalence is variably reported) and often asymptomatic [14]. The impor-
tance of the finding probably increases with increasing length and depth of the
involved segment. In life, relative stenosis of this segment during systolic compres-
sion of the “buried” artery can result in pre-stenotic dilatation, retrograde flow and
plaque formation at the bridge entrance. Nevertheless, they have been associated
with ischaemia, infarction and sudden death.
This finding can be seen on non-contrast PMCT [32], although it is better appre-
ciated following contrast administration [12]. Once seen, the significance of the
finding must be interpreted in the context of the clinical history, as it may be entirely
incidental (Fig. 8.33).
Fig. 8.33 Curved PMCTA

reconstruction of the left
anterior descending artery
demonstrates incidental
bridging of the proximal to
mid vessel segments
through the myocardium
(arrows, blushed with
contrast) and a resultant
underlying mild stenosis
nomalous Coronary Arteries

A
Anomalous coronary anatomies are seen in approximately 1% of the general popu-
lation, with varying degrees of significance. Whilst most clearly demonstrated on
PMCTA, it is usually possible to assess the coronary artery origins and the course
of the proximal segments on a non-contrast study, as they are usually outlined by
fat. Indeed, this could be considered a standard ‘review area’ in PMCT.
Some anomalous configurations are of doubtful clinical significance, such as a
separate origin of the LCx and LAD from the left coronary sinus. This alone would
not usually be considered as a cause for sudden cardiac death.
However, if an anomalous coronary artery passes between two arterial structures
(commonly the aorta and pulmonary artery, referred to as a ‘malignant’ inter-arterial
course), there is a risk of sudden death (Figs. 8.34 and 8.35). This will usually be
associated with exertion but can occur during rest and at any age. Myocardial isch-
aemia is thought to result from vessel compression during increased demand (hence
exacerbated by exercise), leading to infarction or death from arrhythmia.
Examples of such a ‘malignant’ course include an anomalous right coronary
arising from the left aortic sinus and an anomalous left coronary artery arising from
the right aortic sinus. Both configurations are recognised to be associated with sud-
den and exercise-related death [33, 34].
Anomalous origins can also be associated with slit-like ostia, oblique intramural
segments and high take-off origins, where the coronary vessel exits the aorta irregu-
larly or at an acute angle to travel a short distance through the aortic wall itself.
During exercise, as the aorta expands, it is thought that the ostium is closed like a
valve [33].
As with other cardiac findings, it is important to appreciate that anomalous arte-
rial courses may be incidental findings at PMCT [35]. Their presence needs to be
considered against the circumstances and other case findings before being sug-
gested to be related to the cause of death.
Fig. 8.34 PMCTA

oblique axial view at the
level of the coronary ostia
showing an aberrant right
(arrow) which has acute
angulation as it travels
between the aorta and (gas
filled) pulmonary artery—a
so-called malignant course

Fig. 8.34, a coronal MIP
view shows the aberrant,
mildly compressed right
coronary artery winding
around the aortic root
(arrow)
Myocardial Disease
Myocardial Infarction
It is not possible to directly visualise an acute myocardial infarction on non-
enhanced PMCT [36]. The diagnosis of acute myocardial ischaemia or infarction
may occasionally be suggested on PMCTA by a geographic perfusion defect cor-
responding to an arterial territory (Fig. 8.9), along with a stenosis or occlusion of the
relevant vessel and appropriate history (perhaps also correlating with ante mortem
electrocardiogram/ECG evidence). This complete pattern of findings is rarely seen
even in cases of histologically proven infarct [12]. One should be aware that a perfu-
sion abnormality may also be seen as an artefact of PMCTA technique (e.g. ostial
post mortem clot or air bubble ‘occlusion’) or could result from the variables of
decomposition.
One should also appreciate that, in sudden coronary occlusion deaths, there may
not be enough time for a macroscopic visible myocardial infarct to develop and to
be seen at open autopsy. If required, histological sampling from open autopsy may
be used to confirm early ischaemia, sometimes using immuno-histology, although
this is rarely undertaken in routine deaths.
Old/established infarcts can sometimes be seen on PMCT as regions of fatty
replacement, myocardial thinning and calcification (Figs. 8.36 and 8.37). Such pre-
vious infarcts and myocardial scars incur a potential arrhythmogenic risk and should
be mentioned in the report as they are potentially relevant as a cause of death.
Cardiomyopathy
Cardiomyopathies are an important cause of cardiac death and must always be a
differential diagnosis, especially in sudden unexpected deaths of the young or unex-
plained cardiac failure. Historically, they were considered in terms of primary and

the heart on soft tissue
windows shows a well
demarcated region of fatty
replacement in the left
ventricular free wall
(arrow) in a patient with a
known previous
myocardial infarct

windows shows curvilinear
calcification of a thinned
cardiac apex (arrow)
consistent with a known
previous myocardial infarct
secondary disorders. It is clear that most of the primary conditions reflect specific
gene mutations and often affect the young. This group of cardiomyopathies includes
dilated cardiomyopathy (DCM), hypertrophic cardiomyopathy (HCM) and arrhyth-
mogenic (right ventricular) cardiomyopathy (ARVC). Other variants and degenera-
tive cardiomyopathies also exist, and the reader is referred to specialist texts on this
matter [37].
By contrast, ‘secondary’ causes of cardiomyopathy (perhaps better termed as
cardiac disease reflecting systemic conditions) include amyloid deposition, hyper-
tension, alcohol misuse, sarcoid and a variety of storage disorders. Their diagnoses
may be suggested by the history, but PMCT does have a role in confirming cardiac
structural change, features of heart failure and in excluding other pathologies.
Clues for the existence of cardiomyopathy may be apparent, such as profound
cardiac enlargement, hence the need to at least consider heart size at PMCT. The
thickness of the left ventricle might point to HCM or a storage disorder (such as
Fabry disease). Likewise marked fatty change potentially suggests ARVC. A large
and dilated heart, seen in cases of DCM, is not specific and can be seen after myo-
cardial infarction, following myocarditis and in a variety of other conditions. Thus,
tissue sampling for enzyme assessment, gene/DNA review and histology are all
important, making the open autopsy a beneficial and often mandatory protocol in
this context.
Given the importance of making this diagnosis, the radiologist should have a low
threshold for advising/considering focused invasive autopsy, since the inheritance
pattern of these lesions means other family members may be at risk of sudden
deaths. This is particularly so as there are an increasing range of genetic subtypes
described. The coroner and families are often grateful to have this issue explored by
means of open autopsy and potential gene testing.
Valvular Heart Disease
Assessing this mixed group of disorders by routine PMCT has many limitations; in
terms of clinical imaging, they would usually be assessed by real-time ultrasound or
ECG-gated CT/MRI. More information may be gathered by an experienced cardiac
radiologist reporting angiographic images; however, there are findings on the rou-
tine non-contrast images that may still be of value.
Valvular calcification is readily apparent on post mortem imaging and provides
important information to indicate valve disease. Care should be taken however to
separate the valve from aortic root, mitral annular, coronary, myocardial or pericar-
dial calcifications [38, 39].
Aortic valve disease is the most common cardiac valve disease in the Western
population, particularly in older age, and the degree of calcification of the valve
leaflets correlates with the severity of stenosis. Whilst this can be estimated on
PMCT, the true functional significance cannot be assessed. Yet, some understanding
of the effects of the valve disease can be inferred from any associated chamber dila-
tation (e.g. left ventricle in relation to aortic stenosis or left atrium in relation to the
mitral valve), mural hypertrophy or aortic root dilatation. These features however
may not be measurable/reliable in the post mortem setting and need to be judged by
the radiologist for the scan in front of them. There is commonly some post mortem
vascular or cardiac collapse that may preclude these estimations (Fig. 8.38). A nor-
mal aortic valve has three leaflets, and these cannot usually be seen on routine
PMCT but may be evident on PMCTA (Fig. 8.7). The diagnosis of a bicuspid aortic
valve would be important, as significant stenosis may be present here, even if the
calcific burden is mild.

the superior mediastinum
showing normal post
mortem vessel collapse—
which makes the diagnosis
of dilatation difficult/
impossible, although the
surrounding fat planes are
preserved mitigating
against rupture
We are aware of the potential clinical use of the Agatston score to objectively
calculate the calcific burden of the aortic valve [40], but this is not within our current
PMCT practice. For the non-cardiac radiologist, a simpler categorisation, just as
that suggested for clinical practice [39], of ‘none, mild, moderate and severe’ may
be more achievable and appropriate.
Mitral valve leaflet calcification is less common but often seen in the setting of
rheumatic disease or advanced renal impairment. It is important to differentiate this
from mitral annular calcification (normally seen on the posterior and outer ring of
the valve, Fig. 8.39) which is more common, can be extensive, is degenerative in
nature and is associated with normal valve function [38].
Non-calcific valvular soft tissue pathology such as mucoid degeneration causing
incompetence (floppy mitral valve) and vegetations (infective endocarditis) are
unlikely to be identified on PMCT. The replacement of valve by metal prostheses is
readily seen on CT, yet with the associated streak artefact it would be difficult/
impossible to appreciate whether any thrombus or vegetation is present.
Uncommon Cardiac Conditions
Cardiac Tumours
When there is no ante mortem diagnosis, primary cardiac tumours are rarely seen at
PMCT and normally do not feature as autopsy findings, apart from being incidental
pathologies. Pragmatically, only the cardiac myxoma and cardiac sarcoma are gen-
erally apparent and if identified as abnormalities on imaging, PMCT would have to

windows shows incidental
calcification of the mitral
annulus (arrow) in an
elderly patient

thickened, lobulated
pericardium (arrows) in a
case of known
disseminated left bronchial
malignancy
include a very broad range of differentials. As such, these soft tissue lesions would
normally require open autopsy for confirmation.
However, disseminated malignant disease from other sites may terminally
involve the pericardium and cardiac tissues with lymphoma, mesothelioma and lung
cancer, to name but a few commonly implicated (Fig. 8.40).
Cardiac Trauma
Significant trauma to the heart is usually an unequivocal cause of death, due to
induced arrhythmia, haemopericardium with tamponade and/or rapid exsangui-
nation (Figs. 8.41 and 8.42). Mechanisms include both penetrating injuries and
blunt force/crush injury to the chest. Large haemorrhagic collections can be seen
on non-contrast imaging and may be enough to confirm the traumatic cause of
death in the setting of the known trauma history. Coronary angiography can add
to the description of the sites of cardiac and vascular disruption if further detail
is needed (Figs. 8.4, 8.43, and 8.44).

demonstrates a linear
metallic object consistent
with a knife blade
penetrating the anterior
chest wall and pericardium.
The tip now lies at the
right atrial wall

windows at a more cranial
level show a hyperdense
haemopericardium and
massive left haemothorax
(with layering of blood
products) resulting from
fatal penetrating cardiac
injury

Fig. 8.4 (PMCTA after
chest crush injury with
cardiac disruption), shows
further leakage of the
dense angiographic
contrast into the
pericardium and right
hemithorax (arrows)
indicating multiple sites of
internal injury

Fig. 8.4 and 8.43, a more
caudal slice shows the
exact site of contrast
extravasation from the IVC
(arrow), indicating a
venous tear communicating
with the pleural space/large
right haemothorax. This
was confirmed at open
autopsy
Intra-cardiac air embolus should be considered when the history is appropriate,

for example after trauma to the body (not necessarily to the heart itself), childbirth
or a therapeutic procedure where air may enter the vessels. It is estimated that a
volume of 100–250 mL of air is potentially sufficient to cause death [41]. However,
in our practice, it is much more common to see air in vascular structures secondary
to decomposition or following CPR attempts.
In non-trauma cases, care must be taken not to over-call injury related to cardio-
pulmonary resuscitation (Fig. 8.45), which can result in multiple rib, sternal and
even thoracic spinal fractures, cardiac injury, haemopericardium and haemothora-
ces. These features are further discussed in Chap. 11.

shows a complete
transverse fracture of the
sternum with minimal
displacement (arrow), seen
after 2 h of CPR chest
compressions in a
non-traumatic death
Finally, for interest, one must also be mindful of the issue of ‘commotio cordis’
in which sudden blunt force applied to the chest is associated with cardiac dysrhyth-
mia and sudden death. Normally, this is the preserve of forensic pathology cases,
but could possibly feature as part of the routine PMCT workload. Usually, no car-
diac pathology is identified on imaging [34].
Reporting Cardiac Findings: Pearls and Pitfalls
The level of detail to which cardiac findings and their relevance is reported
will depend somewhat on whether the reporter undertakes cardiac imaging in
the living and their confidence in this subspecialty.
Cardiac PMCT may variably include a non-contrast study, coronary calci-
fication scoring and whole-body or targeted coronary PMCTA as per local
agreements or case-by-case assessment.
Any prior cardiac surgery (including valves, stents, bypass grafts, pace-
maker or defibrillator devices) should be reported and considered.
An assessment of the features of decomposition should be made in order to
inform the confidence of subsequent findings.
The volume/nature of pericardial fluid and presence of pericardial calcifi-
cation should be made in all cases. An assessment of heart size (transverse
measurement or CTR for example) may be made with appropriate caution.
References 235
Coronary origins may be assessed for normal or aberrant location, without the
need for PMCTA. Valve and/or other cardiac calcifications should be noted.
A calcium score may be calculated, and an interpretation of its significance
can be given, remembering that sudden death from coronary artery disease is
difficult to diagnose with absolute certainty and reference to the available his-
tory should always be made.
The quality of any angiographic study should be noted, for example ‘excellent,
good, average or poor’ [27] and then the findings detailed for each major vessel.
A coronary cause of death is often described with the more global term
‘ischaemic heart disease’ and usually made on the balance of probabilities if
a relevant significant coronary stenosis is demonstrated (or inferred from cal-
cium scoring). This highlights that the full implications of reporting stenosis
are not fully understood, and this is a research direction needed for adult post
mortem imaging [42].
One should note that the terms ‘cardiac failure’ or ‘cardiac arrest’ should
not feature as a given cause of death as these are ‘modes of death’. They
require correlation to an underlying pathological mechanism and it is this that
the autopsy seeks to describe, such as aortic stenosis.
• The heart is considered to be enlarged with a cardiothoracic ratio of 0.6

(or 60%).
• There is a large haemopericardium with right ventricular flattening, in
keeping with tamponade. The aortic root appears normal and given the his-
tory a ruptured myocardial infarction is most likely.
• Normal coronary artery origin morphology.
• Coronary stents noted. These have been excluded from the calcium score.
The total coronary calcium score is XX; calculated using the Agatston
method. This suggests a XX risk of a significant coronary artery stenosis.
• Excellent quality coronary PMCTA with full opacification of all vessels,
there is no coronary occlusion or stenosis.
• High-grade (>90%) stenosis of the left anterior descending (or other)
artery secondary to mixed plaque.
• Severe aortic valve calcification with associated left ventricular
hypertrophy.
References
1. Osborn M, Lowe J. Guidelines on autopsy practice: sudden death with likely cardiac pathol-
ogy [Internet]. The Royal College of Pathologists, London; 2015. https://www.rcpath.org/
uploads/assets/823dfcf4-8eba-40f7-81b7e174675ecdd9/Guidelines-on-autopsy-practice-
Sudden-death-with-likely-cardiac-pathology.pdf.
3. Roberts ISD, Traill ZC. Minimally invasive autopsy employing post-mortem CT and tar-
geted coronary angiography: evaluation of its application to a routine coronial service.
Histopathology [Internet]. 2014;64(2):211–7. http://doi.wiley.com/10.1111/his.12271.
5. Suvarna SK. National guidelines for adult autopsy cardiac dissection and diagnosis—are they
achievable? A personal view. Histopathology [Internet]. 2008;53(1):97–112. http://doi.wiley.
com/10.1111/j.1365-2559.2008.02993.x.
6. Suvarna SK. The heart at autopsy, including radiological autopsy of the heart. In: Suvarna S,
editor. Cardiac pathology [Internet]. 2nd ed. Cham: Springer International Publishing; 2019.
p. 93–126. http://link.springer.com/10.1007/978-3-030-24560-3_5.
7. Michaud K, Grabherr S, Doenz F, Mangin P. Evaluation of postmortem MDCT and MDCT-
angiography for the investigation of sudden cardiac death related to atherosclerotic coronary
artery disease. Int J Cardiovasc Imaging [Internet]. 2012;28(7):1807–22. http://link.springer.
com/10.1007/s10554-012-0012-x.
8. Agatston AS, Janowitz WR, Hildner FJ, Zusmer NR, Viamonte M, Detrano R. Quantification
of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol [Internet].
1990;15(4):827–32. http://www.ncbi.nlm.nih.gov/pubmed/2407762.
9. Rumberger JA, Brundage BH, Rader DJ, Kondos G. Electron beam computed tomographic
coronary calcium scanning: a review and guidelines for use in asymptomatic persons.
Mayo Clin Proc [Internet]. 1999;74(3):243–52. https://linkinghub.elsevier.com/retrieve/pii/
S0025619611638603.
10. Kelly JL, Thickman D, Abramson SD, Chen PR, Smazal SF, Fleishman MJ, et al. Coronary CT
angiography findings in patients without coronary calcification. Am J Roentgenol [Internet].
2008;191(1):50–5. http://www.ajronline.org/doi/10.2214/AJR.07.2954.
11. Roberts ISD, Benamore RE, Peebles C, Roobottom C, Traill ZC. Diagnosis of coronary artery
disease using minimally invasive autopsy: evaluation of a novel method of post-mortem coro-
nary CT angiography. Clin Radiol [Internet]. 2011;66(7):645–50. https://linkinghub.elsevier.
com/retrieve/pii/S0009926011000675.
12. Ross SG, Thali MJ, Bolliger S, Germerott T, Ruder TD, Flach PM. Sudden death after chest
pain: feasibility of virtual autopsy with postmortem CT angiography and biopsy. Radiology
[Internet]. 2012;264(1):250–9. http://pubs.rsna.org/doi/10.1148/radiol.12092415.
13. Robinson C, Deshpande A, Rutty G, Morgan B. Post-mortem CT: is coronary angiography
required in the presence of a high coronary artery calcium score? Clin Radiol [Internet].
14. Kini S, Bis KG, Weaver L. Normal and variant coronary arterial and venous anatomy on high-
resolution CT angiography. Am J Roentgenol [Internet]. 2007;188(6):1665–74. http://www.
ajronline.org/doi/10.2214/AJR.06.1295.
16. Shiotani S, Kohno M, Ohashi N, Yamazaki K, Nakayama H, Watanabe K, et al. Dilatation of
the heart on postmortem computed tomography (PMCT): comparison with live CT. Radiat
Med [Internet]. 2003;21(1):29–35. http://www.ncbi.nlm.nih.gov/pubmed/12801141.
17. Okuma H, Gonoi W, Ishida M, Shintani Y, Takazawa Y, Fukayama M, et al. Heart wall is
thicker on postmortem computed tomography than on ante mortem computed tomography: the
first longitudinal study. PLoS One [Internet]. 2013;8(9):e76026. https://dx.plos.org/10.1371/
journal.pone.0076026.
References 237
18. Hatch GM, Ampanozi G, Thali MJ, Ruder TD. Validation of left ventricular circumferen-
tial area as a surrogate for heart weight on postmortem computed tomography. J Forensic
S2212478013000634.
19. Yamaguchi R, Makino Y, Chiba F, Torimitsu S, Yajima D, Shinozaki T, et al. Fluid-fluid level
and pericardial hyperdense ring appearance findings on unenhanced postmortem CT can dif-
ferentiate between postmortem and antemortem pericardial hemorrhage. Am J Roentgenol
[Internet]. 2015;205(6):W568–77. http://www.ajronline.org/doi/10.2214/AJR.15.14808.
20. Mychajlowycz M. The armored heart: differentiating the etiology of hemopericardium on
postmortem computed tomography. J Forensic Radiol Imaging [Internet]. 2017;9:6–7. https://
21. James P, Morgan B, Rutty GN, Brough A. Cardiothoracic ratio (CTR) measured on post-
mortem computed tomography (PMCT)—pre- and post-ventilation. J Forensic Radiol Imaging
22. Okuma H, Gonoi W, Ishida M, Shirota G, Kanno S, Shintani Y, et al. Comparison of the
cardiothoracic ratio between postmortem and antemortem computed tomography. Leg Med
23. Winklhofer S, Berger N, Ruder T, Elliott M, Stolzmann P, Thali M, et al. Cardiothoracic ratio
in postmortem computed tomography: reliability and threshold for the diagnosis of cardiomeg-
aly. Forensic Sci Med Pathol [Internet]. 2014;10(1):44–9. http://link.springer.com/10.1007/
s12024-013-9504-9.
24. Suvarna SK. Teaching and examining for post-mortem CT-scanned autopsies. Diagnostic
Histopathol [Internet]. 2020;26(8):343–9. https://linkinghub.elsevier.com/retrieve/pii/
S1756231720300815.
25. Screaton N. The cardiothoracic ratio—an inaccurate and outdated measurement: new data
from CT. Eur Radiol [Internet]. 2010;20(7):1597–8. http://link.springer.com/10.1007/
s00330-010-1721-y.
26. Jotterand M, Faouzi M, Dédouit F, Michaud K. New formula for cardiothoracic ratio for the
diagnosis of cardiomegaly on post-mortem CT. Int J Legal Med [Internet]. 2020;134(2):663–7.
http://link.springer.com/10.1007/s00414-019-02113-1.
27. Raff GL, Abidov A, Achenbach S, Berman DS, Boxt LM, et al. SCCT guidelines for the
interpretation and reporting of coronary computed tomographic angiography. J Cardiovasc
Comput Tomogr [Internet]. 2009;3(2):122–36. https://linkinghub.elsevier.com/retrieve/pii/
S1934592509000707.
28. Champ CS, Coghill SB. Visual aid for quick assessment of coronary artery stenosis at necropsy.
J Clin Pathol [Internet]. 1989;42(8):887–8. http://jcp.bmj.com/cgi/doi/10.1136/jcp.42.8.887.
29. Lim T-H, editor. Practical textbook of cardiac CT and MRI [Internet]. Berlin, Heidelberg:
Springer Berlin Heidelberg; 2015. http://link.springer.com/10.1007/978-3-642-36397-9.
30. Michaud K. Ischaemic heart disease. In: Suvarna SK, editor. Cardiac pathology [Internet].
2nd ed. Cham: Springer International Publishing; 2019. p. 137–51. http://link.springer.
com/10.1007/978-3-030-24560-3_7.
31. Davies MJ, Popple A. Sudden unexpected cardiac death? A practical approach to the forensic
problem. Histopathology [Internet]. 1979;3(4):255–77. http://doi.wiley.com/10.1111/j.1365-
2559.1979.tb03008.x.
32. Ampanozi G, Gascho D, Hatch G, Schulze C, Thali MJ, Ruder TD. What is unsought will
go undetected—myocardial bridging on postmortem computed tomography. J Forensic
S2212478013001238.
33. Taylor AJ, Rogan KM, Virmani R. Sudden cardiac death associated with isolated congenital
coronary artery anomalies. J Am Coll Cardiol [Internet]. 1992;20(3):640–7. http://www.ncbi.
nlm.nih.gov/pubmed/1512344.
34. Suvarna SK. Sudden cardiac death. In: Suvarna SK, editor. Cardiac pathology [Internet].
2nd ed. Cham: Springer International Publishing; 2019. p. 277–311. http://link.springer.
com/10.1007/978-3-030-24560-3_14.
35. Martinez RM, Flach PM, Ebert LC, Bartsch C, Thali MJ, Ampanozi G. Anomalous left coronary
artery origin on postmortem imaging in correlation with autopsy. J Forensic Radiol Imaging
[Internet]. 2014;2(3):146–8. https://linkinghub.elsevier.com/retrieve/pii/S2212478014000586.
36. Wagensveld IM, Blokker BM, Pezzato A, Wielopolski PA, Renken NS, von der Thüsen JH,
et al. Diagnostic accuracy of postmortem computed tomography, magnetic resonance imag-
ing, and computed tomography-guided biopsies for the detection of ischaemic heart disease
in a hospital setting. Eur Hear J Cardiovasc Imaging [Internet]. 2018;19(7):739–48. https://
academic.oup.com/ehjcimaging/article/19/7/739/4883380.
37. Bunning CR, Suvarna SK. Cardiomyopathies. In: Suvarna SK, editor. Cardiac pathology
[Internet]. 2nd ed. Cham: Springer International Publishing; 2019. p. 205–25. http://link.
springer.com/10.1007/978-3-030-24560-3_11.
38. Kanza RE, Allard C, Berube M. Cardiac findings on non-gated chest computed tomography:
a clinical and pictorial review. Eur J Radiol [Internet]. 2016;85(2):435–51. https://linkinghub.
elsevier.com/retrieve/pii/S0720048X15301807.
39. Williams MC, Abbas A, Tirr E, Alam S, Nicol E, Shambrook J, et al. Reporting incidental
coronary, aortic valve and cardiac calcification on non-gated thoracic computed tomography, a
consensus statement from the BSCI/BSCCT and BSTI. Br J Radiol [Internet]. 2021;94(1117).
https://www.birpublications.org/doi/10.1259/bjr.20200894.
40. Pawade T, Clavel M-A, Tribouilloy C, Dreyfus J, Mathieu T, Tastet L, et al. Computed tomog-
raphy aortic valve calcium scoring in patients with aortic stenosis. Circ Cardiovasc Imaging
[Internet]. 2018;11(3). https://www.ahajournals.org/doi/10.1161/CIRCIMAGING.117.007146.
41. Burton J, Rutty G. In: Burton JL, Rutty G, editors. The hospital autopsy [Internet]. 3rd ed.
London: CRC Press; 2010. https://www.routledge.com/The-Hospital-Autopsy-A-Manual-of-
Fundamental-Autopsy-Practice-Third-Edition/Burton-Rutty/p/book/9780340965146.
42. Morgan B, Adlam D, Robinson C, Pakkal M, Rutty GN. Adult post-mortem imaging in
traumatic and cardiorespiratory death and its relation to clinical radiological imaging. Br J
Radiol [Internet]. 2014;87(1036):20130662. http://www.birpublications.org/doi/10.1259/
bjr.20130662.
of the Abdomen and Pelvis 9
Introduction
The combined abdominal and pelvic cavities can present a challenge to interpret
on post mortem computed tomography (PMCT), as there are many false-positive
and false-negative findings to consider [1, 2]. Indeed, there is considerable over-
lap between the normal (decomposition-related) gas patterns and the findings
that indicate true pathology. It is important to appreciate that, as decomposition
progresses, the diagnostic potential of the scan also reduces in a progressive
manner. This situation can result in decreasing confidence with interpretation
and requires a degree of caution and judgement from the reporter. Scanning as
early as possible after death is preferred as this will mitigate some of the post
mortem effects.
As known from clinical practice, abdominal viscera are difficult to assess with-
out intravenous (IV) contrast. As such, it is important to realise that a number of
findings in the abdomen may be difficult to appreciate or even completely over-
looked at PMCT. This stands true even without decomposition changes. In addition,
if there is paucity of intra-abdominal body fat, the contrast between normal body
tissues is reduced and further diminishes visceral definition (Fig. 9.1).
Causes of death in this cavity are however less frequent than those seen in the
chest and particularly uncommon with no history of abdominal symptoms. However,
when sudden or unexpected, they are often visibly catastrophic and unequivocal,
such as a fatal rupture of the abdominal aorta.
Autopsy of the Abdomen: The Pathologist’s Perspective
Pathologists tend to work in terms of cavities, often regarding the abdomen and the
pelvis as two discrete zones of tissues. The oesophagus clearly falls into the thoracic
compartment but may be left intact with the abdominal block. Dissection of all these

Switzerland AG 2021
https://doi.org/10.1007/978-3-030-70829-0_9
240 9 Post Mortem Computed Tomography of the Abdomen and Pelvis

upper abdomen on soft
poor intra-abdominal
definition secondary to
both the lack of intra-
venous contrast and
intra-abdominal fat. Note
the ‘scaphoid’ cross-
sectional profile of the
anterior abdominal wall
tissues varies depending on the needs of the case. For example, cases limited to the
chest will include the oesophagus but not any of the abdominal viscera.
However, if the abdomen and pelvis do require dissection, then these tissues are
removed in one piece along with retroperitoneal compartment elements. Clearly,
peritonitis, fluid collections and disseminated cancer (omental cake) should be evi-
dent at the point the initial examination is made.
Once the abdomen is open, the jejunum is transected and the bowels are removed
progressively by incisions into the fatty mesentery permitting tissue removal en
bloc. Visual inspection and palpation are usually all that is required, as tumours with
stenosis and diverticular disease are usually quite evident, requiring only localised
opening of the bowel lumen. The bowels rarely require complete opening and wash-
out unless the cases involve diffuse mucosal disease.
The remaining tissues can be dealt with in various ways. One could start with the
pelvic content after checking the aorta and inferior vena cava. The prostate and
bladder (males) or uterus, tubes, ovaries and bladder (females) require direct inci-
sion and inspection.
The kidneys are incised along their long axis so that the pelvis can be explored
and the ureter traced, if necessary. At this point, the adrenals are normally checked
and weighed if significantly large or small.
The upper gastro-intestinal tissues generally are examined initially from the pos-
terior/inferior aspect so that the gall bladder is identified and opened. The common
bile duct may be explored in obstruction cases. The spleen is often removed at the
time of studying the liver, with consideration of the size and cut surface paren-
chyma. Any lymphadenopathy should be sampled for histology and may require
microbiological testing. Subsequently, the liver can be removed and weighed sepa-
rately. This allows the stomach (sometimes with the oesophagus still attached) to be
opened along the greater curve, through the pylorus into the duodenum. Stomach
content may be removed for toxicology analysis at this point. The mucosal content
should be considered in terms of haemorrhage and mass lesions. Lifting the stom-
ach upwards and cranially allows the pancreas to be checked and explored, usually
by serial transverse slices.
At this point, it should be remembered that the vertebral bone is exposed.

Sampling of the marrow compartment is possible by means of sawcut into the ver-
tebral block and removal of a specific section for decalcification and histology.
Histological sampling for the various solid organs and viscera is variably untaken
as part of the autopsy, although kidney and liver are common biopsy sites.
Photography may be of benefit in some cases as part of the record.
Solid Abdominal Viscera
In a body with a reasonable amount of intra-abdominal fat, inherent tissue contrast

allows for pragmatic assessment of visceral size and contour (Figs. 9.2, 9.3, and
9.4). This analysis may also demonstrate a surrounding abnormality, such as haem-
orrhage, inflammatory change or fibrosis. The solid viscera can be recognised for
some time post mortem, although they gradually fill with decomposition gas until

intra-abdominal fat
outlining normal tissues
such as the liver, adrenal
glands and collapsed aorta

mid-abdomen on soft
intra-abdominal fat
such as the kidneys and
retroperitoneum, collapsed
IVC and aorta

lower abdomen on soft
intra-abdominal fat
such as large and small
bowel loops, abdominal
wall muscles, collapsed
IVC and aorta
the parenchyma breaks down and becomes one with decomposition fluid. Given that
the abdomen and pelvis contain the bacteria-rich bowels, early signs of decomposi-
tion in the viscera are common.
It may be helpful to include maximum diameters or length measurements of
clearly enlarged or atrophic organs, such as the spleen or kidneys, as a surrogate for
organ weight (as might be obtained in open autopsy). Measuring organ density and
volume in order to ‘estimate’ weight is possible with suitable software and may cor-
relate with causes of death such as fatal haemorrhage [3]. However, it is time-
consuming, subject to marked variability and probably not reliable enough to be
informative in this setting. For example, in one small study it was shown that the
liver can decrease in volume by up to 30% by 36 h post mortem, presumably due to
passive outflow of blood and compression from the expanding bowels and lungs [4].
Autolysis and Gastromalacia
All tissues autolyse after death. Due to its early autolysis, the pancreas commonly
demonstrates a surrounding ‘haziness’ on PMCT (Fig. 9.5), which may mimic true
pancreatitis (Fig. 9.6). The pancreas is rapidly replaced with decomposition gas and
soon becomes imperceptible in relation to its surroundings. The adrenals and spleen
also undergo early autolysis, although their PMCT appearances remain ‘normal’
for longer.
Even in life, the stomach can be difficult to assess on contrast-enhanced CT, and
the challenge persists in the post mortem setting. One additional post mortem issue
is of gastromalacia (decomposition-related ‘softening’ of the stomach). This can
present as stranding around the stomach and appear pathological (Fig. 9.7).
Eventually, this causes gastric rupture, resulting in leaked content and a pneumo-
peritoneum, which is therefore potentially a normal (late) post mortem finding.
Gastromalacia may also be the cause of small pleural effusions or pneumothoraces,
as the diaphragm is not an absolute boundary, with the pleural space and peritoneum
being linked [5].

(arrows) secondary to
autolysis

(similar to Fig. 9.5) but
instead due to pancreatitis,
confirmed at limited
abdominal open autopsy

low-density (compared to
the spleen) fatty liver in a
known alcoholic. Normal
splenic size and
appearances. Faint
haziness around the
stomach (arrow) is in
keeping with autolysis
Pelvic Viscera
In comparison to the upper abdominal viscera, the pelvic tissues (prostate in males,
uterus/tubes/ovaries in females) and urinary bladder are often well preserved at
PMCT (Fig. 9.8). As with clinical imaging, a basic assessment of size and gross
appearances is probably satisfactory to exclude significant (cause of death related)
pathology, although an empty bladder is always more difficult to assess. These vis-
cera are rarely implicated a cause of death, unless they are the seat of malignancy—
which is usually known from the history.
If a urinary catheter is present it should be noted, although the reporter should
consider that its position may have altered post mortem. The presence of a catheter
(Fig. 9.9) may well be an incidental observation but could represent a potential
infective focus. In the non-catheterised bladder, it is useful to report the approxi-
mate bladder volume. Should toxicology be required, it assists the pathologist to
know if the bladder is empty as suprapubic aspiration will be futile in this situation.

pelvis on soft tissue
windows shows the relative
preservation of the urinary
bladder (arrow) compared
to the surrounding soft
tissues which contain
decomposition related gas

windows showing an
empty urinary bladder
(arrow) collapsed around a
catheter balloon and tube
If aspiration has already occurred, there may be a residual, but sometimes striking,
gas track in the anterior abdominal wall (Figs. 9.10, 9.11, and 9.12) not to be con-
fused with unexpected traumatic injury!
The presence of a pregnancy should always merit comment, being both normal
and yet potentially relevant to the death of the mother. Fetal measurements may aid
as a resource is assessing gestation stage. One should remember that non-pregnancy
pathology, such as trauma and suicide, may be the cause of death. Generally, all
maternal deaths will require open autopsy.

the pelvis on bone
windows shows a
suprapubic, subcutaneous
gas collection (arrow)
following bladder
aspiration for toxicology

Fig. 9.10, a sagittal view
again shows the suprapubic
gas (arrow) to track down
to the partially collapsed
bladder, following

the pelvis on lung windows
(different case to Figs. 9.10
and 9.11) most clearly
demonstrates a suprapubic
subcutaneous gas track
following bladder
with no significant
decomposition gas
elsewhere

normal post mortem
peripheral left lobe of the
liver (arrows)
Intra-abdominal Gas Patterns
Due to its high native bacterial load, the abdomen is usually the first location in the
body to exhibit changes of putrefaction, such as gas accumulation. At external
inspection, early putrefaction may be seen as bloating and a green tinge to the skin,
commonly of the right iliac fossa (overlying the caecum), before becoming more
generalised. Putrefaction may be rapid in states of infection or sepsis, appearing
more prominent than expected for the post mortem interval and environmental con-
ditions (see Chap. 3).
Hepatic gas is a common and normal early decomposition finding on PMCT,
seen in the hepatic veins, arteries, portal veins, or a combination of vessels. It is
usually seen first in the non-dependent (assuming supine position) left lobe
(Fig. 9.13). With smaller volumes it can be difficult to localise, and so gas elsewhere
(right heart, main portal vein, systemic veins or arteries) may help confirm location
(Fig. 9.14) although, if judged to be due to decomposition or as a consequence of
assisted ventilation/resuscitation attempts, its exact location is probably not of sig-
nificance. The location of any gas may be of importance when pathology is sus-
pected, for example, gastrointestinal distension or traumatic air embolism [6].

tissue windows shows left
portal vein decomposition
gas, this could be traced
into the main portal vein

the mid abdomen on soft
gaseous distension of the
bowel (arrows) due to
decomposition. Note also a
calcified abdominal aorta,
partially ‘propped open’ by
calcification but without
evidence of rupture
Gaseous post mortem distension of the bowel is also very common (Fig. 9.15),
and the volume of gas here can more than double in the first few days after death [4].
Intramural bowel gas can also be a normal post mortem finding, most likely to be
related to decomposition (Fig. 9.16) but may relate to failed cardio-pulmonary
resuscitation [2]. If bowel wall gas is present but seems out of proportion to decom-
position changes elsewhere and unrelated to the history, the possibility of existing
primary pneumatosis intestinalis should be considered although this is considered
rare (Fig. 9.17).
Free intra-peritoneal gas is commonly seen with advanced decomposition,
although this should always follow obvious visceral and vascular gas accumulation
(Figs. 9.18 and 9.19) to avoid mis-interpreting true pathology.

the mid abdomen on lung
windows shows gas in the
bowel wall (arrows) and
mesenteric vessels most
likely due to
decomposition

pneumatosis of the colon
(arrows) without
significant decomposition
gas in the vessels or
elsewhere. The cause of
death in this case was an
unrelated acute pneumonia

windows shows a
moderate-sized
pneumoperitoneum
outlining the thin falciform
ligament (arrow). This is
judged secondary to
decomposition given the
extensive generalised
visceral and vascular gas

windows shows a large
pneumoperitoneum with
significant abdominal
distension/bloating
(arrows), bowel wall,
vascular and soft tissue gas
all secondary to advanced
decomposition
Intra-Abdominal Fluid Patterns
A small volume of intra-abdominal or pelvic free fluid may be physiological (in a

young female) or due to progressing decomposition following expected organ autol-
ysis (Fig. 9.20). These post mortem collections should not be misdiagnosed as asci-
tes or haemorrhage [1], with the latter being hyperdense (Fig. 9.21). Fatal
intraperitoneal or retroperitoneal haemorrhage is usually extensive and unmistak-
ably identified. Smaller haemorrhagic intra-abdominal collections may result as a
consequence of chest compressions (see Chap. 11).
The presence of hyperdense fluid within the gastro-intestinal (GI) tract is notori-
ously difficult to interpret, as it can be highly variable and non-specific (Fig. 9.22).
The bowel content varies in density from multiple factors such as food, medications
(Figs. 9.23 and 9.24), previous oral contrast (usually in-hospital deaths) and, of
course, haemorrhage [2]. Conversely, when the GI tract fluid is fatty or water dense,
another potential artefact arises—the bowel wall can appear abnormally hyper-
dense. This may be a false-positive finding or, rarely and within appropriate circum-
stances, indicative of true intramural haemorrhage [2]. As such, much caution
should be taken when assessing the bowel wall and its content, especially in the
absence of anticipated pathology.

windows in a moderately
decomposed body shows a
horizontal fluid level in the
pelvis (arrow) due to
decomposition, the density
of the fluid measured a
mean of −97HU i.e. fatty
density

tissue windows showing a
small haemoperitoneum,
indicated by crescentic
hyperdensity on the left
(arrows), also note normal
decomposition gas
anteriorly in the liver.
Findings were judged to be
secondary to resuscitation
attempts

non-specific hyperdense
dependent stomach content
(arrow). The given history
suggested the possibility of
drug intoxication. Findings
could indicate partially
digested medications but
are non-specific,
toxicological sampling
would usually be required

multiple discrete densities
in the stomach (arrows)
indicating undigested
tablets after an intentional
overdose. There is
moderate decomposition
gas in the liver

heterogeneous
hyperdensity in the
stomach following suicide
by overdose of
antipsychotic medication.
Whilst non-specific this
may relate to partially
digested medications
It is not defined exactly how the density of blood (which separates due to post
mortem hypostasis) and decomposition fluid (due to variable cellular and visceral
breakdown) changes over time. Therefore, the reliability of measuring the density
of intra-abdominal fluids to accurately determine their nature is somewhat question-
able and should be cautiously correlated with the clinical history and circumstances
of the body after death.
Ruptured Abdominal Aortic Aneurysm
This catastrophic cause of sudden death is readily revealed on PMCT. Acute haem-
orrhage from a ruptured abdominal aorta appears as heterogeneous (but generally
high-density) peri-aortic stranding and retro-peritoneal haematoma (Figs. 9.25,
9.26, 9.27, 9.28, and 9.29). Occasionally, there may also be intra-peritoneal
(Fig. 9.30) or intra-thoracic extension of the haemorrhage (Fig. 9.31).

hyperdense peri-aortic
haemorrhagic stranding
and right retro-peritoneal
haematoma (arrows) from
a ruptured abdominal
aortic aneurysm. The aorta
is outlined by mural
calcification

windows showing bilateral
retro-peritoneal
haematoma (arrows)
around a heavily calcified,
ruptured abdominal aortic
aneurysm
The observation that other major vessels are collapsed might also indirectly sup-
port a diagnosis of significant haemorrhage, but it must be remembered that on
PMCT the vessels, including the aorta are very often at least partially, if not com-
pletely, collapsed (Figs. 9.2, 9.3, 9.4, and 9.32). In some cases, vessels and aneu-
rysms partially maintain their shape and size due to the presence of mural
calcification (Figs. 9.15, 9.33, and 9.34). The anteroposterior aortic diameter is thus
not a reliable measurement to prove an aneurysm in the post mortem setting (as it
would be in life), but in the absence of rupture, the exact measurement is of little
significance (Fig. 9.33). An incidental aneurysm, highlighted by calcium and throm-
bus or fibrous tissue, should not be taken as fatally ruptured without very obvious
secondary signs.

hyperdense peri-aortic and
right retro-peritoneal
haematoma around a partly
collapsed, calcified aortic
aneurysm

aneurysm with extension
of haemorrhage posterior
to the ascending colon
(arrow)

extensive retroperitoneal
haemorrhage extending
through peri-renal and
para-renal spaces,
secondary to rupture of a
now collapsed, crumpled
aortic aneurysm

large left retro-peritoneal
haematoma with intra-
peritoneal extension
resulting from a ruptured
abdominal aortic aneurysm

windows shows bilateral,
dependent haemothoraces
with layered separation
(arrows), following
aneurysm (not seen on this
slice)

collapsed abdominal aorta
(arrow), there is a small
focus of mural calcification
but no aneurysm or
evidence of rupture

partially collapsed
abdominal aortic
aneurysm, diagnosed in
life as measuring 44 mm
anteroposterior. Post
mortem the measurements
were 22 mm
anteroposterior (arrows)
and 45 mm transverse

partially collapsed
abdominal aortic aneurysm
with multiple mural foci of
calcification (arrows)
Hepato-Biliary Pathology
On PMCT, as for clinical non-contrast studies, the densities of the solid viscera are
usually similar to each other. This is a recognised limitation in defining various
pathologies.
Fatty infiltration of the liver, simple or hyperdense cysts and haemorrhage may
still be appreciated (Figs. 9.35 and 9.36). More unusually, a diffuse increase in liver
attenuation may be seen due to amiodarone use (Fig. 9.37), glycogen storage dis-
ease or mineral deposition (e.g. haemochromatosis).
Established liver cirrhosis is normally evident as a shrunken and irregular organ.
Gallstones may also be visible on PMCT, with thickened gallbladder wall and sur-
rounding inflammatory stranding suggesting cholecystitis. Rarely, resuscitation-
related trauma can be seen with liver tears and local haemorrhage.

generalised low density of
the liver parenchyma
(compared to the spleen),
indicating fatty infiltration/
hepatic steatosis in a
known alcoholic

large simple cyst arising
from the right kidney
(kidney not seen on this
slice) and adjacent small
volume of right para-colic
gutter haemorrhage (arrow)
judged to have resulted
from resuscitation attempts

generalised high density of
the liver due to amiodarone
use, correlated with the
drug history

tissue windows showing an
enlarged, subtly
heterogeneous liver
suggestive of multiple
metastases (given that
there was also a small
volume of ascites and a
suspected bowel primary).
The patient had refused
investigations in life
Unfortunately, pathology such as biliary dilatation, vascular thromboses, vis-

ceral infarcts, tumours or metastases may remain undetectable, unless extensive or
sizeable (Figs. 9.38, 9.39, and 9.40). This should be kept in mind if such pathology
is anticipated from the clinical scenario and the limitations of the study in excluding
such pathology should be communicated in these circumstances.

large slightly
hypoattenuating known
metastatic lesion in the
right lobe of liver (arrows)
in a case with a known
rectal primary

the kidneys on soft tissue
windows shows a left renal
infarct (arrow), clearly
identified due to the
presence of iodinated
contrast enhancing the
normal renal parenchyma.
Contrast had been
administered just before
death, during emergency
radiological intervention
for aortic trauma
Intra-abdominal Inflammatory Change and Infection
In a body with a reasonable volume of fat outlining tissues and no significant

confounding decomposition changes, the secondary findings that suggest
inflammation and infection include visceral oedema (Fig. 9.41), peri-visceral fat
stranding, fascial thickening, causative calculi (Figs. 9.42, 9.43, and 9.44) and
frank abscess (Fig. 9.45).
As with any imaging though, infection may be apparent but clearly a causative
organism cannot be defined without obtaining supportive microbiological evidence,
or preceding in vivo cultures. With such findings, if there is an appropriate clinical
history to support (or at the very least not contradict) an infective pathology, one
may only infer ‘infection’ and/or ‘sepsis’ as the cause of death without detailing the
specific responsible organism.

thickened gallbladder wall
(arrows), in keeping with
cholecystitis. There is a
calcified gallstone in the
gallbladder neck

right-side perinephric
inflammatory stranding
and mild hydronephrosis.
The patient had abdominal
pain and a rapid
deterioration

Fig. 9.42, a coronal view
of the abdomen and pelvis
shows a swollen right
kidney with perinephric
inflammatory stranding
and thickening of the
fascia (arrows) compared
to the left

Figs. 9.42 and 9.43, axial
view of the lower abdomen
demonstrates a right-side
ureteric calculus (arrow) as
the cause of obstruction
and likely secondary
infection. At limited open
PM the right kidney was
full of pus. Cause of death
was therefore sepsis from
pyelonephritis, secondary
to obstructing renal
calculus

right retroperitoneal,
abdominal wall and
subcutaneous mixed fluid
and gas collection (arrows)
in keeping with extensive
abscess. The exact origin
was not identified
on PMCT
Abdominal Neoplasia
In our experience, it is rare to come across unknown malignancies as a cause of sud-

den or unexpected death, and so this is an unusual diagnosis to make on PMCT
without further investigations (Fig. 9.38). Bowel malignancy may present acutely
with an obstructing mass lesion or perforation. Mesenteric and omental thickening
is variably identified by PMCT as a marker for disseminated malignancy in the
abdomen. This may be supported by a large mass within the pancreas, stomach or
ovary as common primary sources. The presence of a large fluid collection would
also support the diagnosis diagnosis of malignancy or hepatic dysfunction. Other
findings in widespread malignancy may include liver deposits (Fig. 9.39) from lung,
breast, bowel or pancreatic primaries, adrenal metastases, commonly from the lung
and adenopathy (suggesting lymphoma or secondary cancers).
Correlation against the medical history is invaluable. Limited sampling (fluid
aspiration, needle core biopsy or focused small autopsy incision and biopsy) serves
to assist the final diagnosis.
The Pancreas
Given its rapid autolysis (Fig. 9.5), there is often minimal data derived from assess-
ment of this tissue on PMCT, and it is important to be aware that normal changes
can mimic acute pancreatitis (Fig. 9.6). However, benign and malignant tumours
can persist for longer periods to allow consideration. Chronic damage (e.g. alcohol-
mediated calcifications) may support background clinical data and tie in with cir-
rhosis in those misusing alcohol. The identification of cysts and pseudocysts may be
variably confirmed, depending on their size and the post mortem interval.
Bowel Volvulus and Perforation
Bowel volvulus (e.g. small bowel, sigmoid or caecal) can be difficult to clinically
diagnose, reflecting the variable presentations and yet can be rapidly fatal [7]. The
supporting clinical history may be rather vague, equally applied to the post mortem
setting and especially for community deaths. In the setting of suspected bowel
pathology such as volvulus, free fluid suggests ‘transmigration peritonitis’, which in

turn may have resulted in sepsis and death.
Bowel imaging in the post mortem setting is generally challenging, but, when
present, the PMCT features of volvulus are helpfully specific and similar to clinical
imaging. There is distension of the affected bowel segment (localised and more
prominent than due to decomposition) possibly with associated ‘coffee bean’, ‘beak’
or ‘whorl’ imaging signs (Fig. 9.46). There may be associated rupture, leading to free
fluid and pneumoperitoneum (Figs. 9.47, 9.48, and 9.49), although in such cases the
distended bowel may decompress making the exact diagnosis difficult. Free air and
free fluid are readily visualised on PMCT but are non-specific.

windows shows mildly
distended bowel loops, free
fluid and a central
‘whorled’ mesenteric root
(arrows) in a patient who
presented with acute
abdominal symptoms.
Imaging findings are in
keeping with ruptured
small bowel volvulus

tissue windows shows an
anterior pneumoperito-
neum. The lack of
decomposition gas
accumulation elsewhere
suggests this to be
secondary to a pathological
perforation

windows shows a
pathological large
pneumoperitoneum,
distending the abdomen
(arrows), with minimal
features of decomposition
in the surrounding tissues
(compare to Fig. 9.19
which shows a
decomposition related
pneumoperitoneum)

pathological
pneumoperitoneum with
the falciform ligament
outlined anteriorly.
Moderate volume of free
fluid lateral to the liver
(arrow) also suggests true
perforated viscus rather
than decomposition
Bowel Ischaemia
In clinical imaging, intra-venous contrast can be used to demonstrate bowel vitality,

with abnormal, non-enhancing loops clearly seen. On PMCT, without such contrast,
it is difficult (if not impossible) to appreciate bowel ischaemia unless there are sec-
ondary signs and/or a very high clinical suspicion.
Mural oedema and bowel dilatation seen in a defined vascular territory (e.g.
superior mesenteric artery) and/or associated pathological gas pattern (bowel wall
gas, portal gas and eventually pneumoperitoneum) may support this interpretation.
Yet, these clinical hallmarks of bowel ischaemia significantly overlap with the nor-
mal and more common changes in decomposition. Consequently, making this diag-
nosis with confidence is difficult on PMCT alone. Without compelling history and
imaging, a limited abdominal compartment invasive autopsy may remain necessary
to confirm the diagnosis.
Gastrointestinal Tract Haemorrhage
Massive gastro-intestinal (GI) tract haemorrhage, seen as extensive hyperdense hae-

matoma potentially with separation of blood products due to hypostasis (Figs. 9.50,
9.51, and 9.52), may lead to sudden death from exsanguination or precipitate an
acute cardiac event from hypotension. Sites of origin include peptic ulcers, varices,
tumours, aorto-oesophageal or aorto-enteric fistulae and may be suggested by the
past medical history (dyspepsia, alcohol, medications) or presenting features (such
as haematemesis or melaena). Consideration of any available clinical data and the
distribution of visualised haemorrhage may allow a broad suggestion of the site of

the lower chest on soft
haemorrhagic layering in
the mildly distended
oesophagus (arrow),
confirmed haemorrhage at
open autopsy

large volume of irregular,
solid hyperdense gastric
content in keeping with
haematoma (arrows). This
was confirmed at open
autopsy. An incidental
large simple liver cyst is
noted

mid abdomen shows
further irregular
haematoma in the distal
stomach (arrows)
origin (gastric, duodenal etc.), on the balance of probabilities, although the exact
site of haemorrhage may not be demonstrated on routine PMCT. A limited abdomi-
nal invasive autopsy may proceed although, even then, it is not always possible to
identify the point of haemorrhage.
Abdominal Trauma
Compared to clinical imaging with contrast, PMCT generally has a low sensitivity
for detecting intrinsic solid organ injury such as contusions. Significant/large injuries
are detected with higher sensitivity and specificity (due to anatomic distortion, asso-
ciated gas tracks and haemorrhage). Most life-threatening liver injuries, supported by
the history and external features, can be detected by PMCT [8] making it, overall, a
suitable technique for the post mortem examination of trauma.
Early post mortem scanning is helpful to avoid the confusion of traumatic gas
patterns with normal changes in decomposition (Fig. 9.53). For interpretation of
trauma cases, it is also worth noting that putrefactive decomposition occurs more
rapidly at sites where bacteria have been introduced into the body by an injury.
Abdominal trauma, including that sustained during cardiopulmonary resuscita-

tion, (see Chap. 11) may result in injuries of the liver (owing to its size) followed by
injuries of the spleen, kidneys and other viscera (Figs. 9.54, 9.55, 9.56, and 9.57).
Lacerations appear as linear low attenuation defects with, or without, visceral con-
tour disruption, gas tracks or haemorrhage (which may be minimal in cases of exter-
nal exsanguination).
Renal injuries often lead to retroperitoneal haematoma, which can tamponade,
given the limited anatomic space. By contrast, hepatic and splenic injury can haem-
orrhage freely into the much larger intraperitoneal space with subsequent internal
exsanguination. Pelvic fractures raise the possibility of bladder and iliac vessel
ruptures.
Abdominal organ positions are usually well demonstrated within body fat, and
so a traumatic diaphragmatic hernia may be readily appreciated. However, this may
be more difficult to confirm if there is basal lung collapse, pleural effusion or
haemothorax, especially on the right where liver density may match congested
lungs and is not outlined by fat as the spleen or stomach may be [9].

pneumoperitoneum (arrow)
collision. This was
however judged most
likely due to
decomposition given the
gas presence elsewhere
(scanned 12 days post
mortem)

haemoperitoneum
secondary to large
traumatic liver laceration
(arrow)

small peri-splenic
haematoma (arrow). This
resulted from
cardiopulmonary
resuscitation with multiple
rib fractures sustained (not
illustrated). The cause of
death was acute pulmonary
embolus

the upper abdomen
windowed to show an
extensive layered
haemoperitoneum which
also outlines a splenic
laceration (arrow)

the kidneys on soft tissue
windows showing a right
perinephric haematoma
(arrow). There was a
presumed history of
unwitnessed trauma in this
anticoagulated alcoholic.
Moderate background
decomposition gas noted
Bilateral Adrenal Haemorrhage
This is a rare pathology which has a variable and often non-specific clinical presen-
tation [10]. It is associated with states of sepsis, anticoagulation, antiphospholipid
syndrome, trauma and surgery. Imaging findings are usually straightforward with
hyperdense, swollen adrenal glands (Fig. 9.58), but the finding needs to be corre-
lated with the medical history to form clear conclusions.

the adrenal glands on soft
bilateral hyperdense
adrenal haemorrhages
(arrows) in a patient who
also had acute pneumonia
(not shown) and therefore
likely sepsis
Reporting Abdominal and Pelvic Findings: Pearls and Pitfalls
An initial assessment of the quality and reliability of the imaging, in relation

to body fat and decomposition factors, defines the degree of confidence with
which subsequent findings can be held.
There is significant overlap between normal decomposition-related gas
patterns and those representing true pathology.
When the scan is adequate, fatal events such as aortic rupture, significant
traumatic injury or pathological bowel perforation should clearly be evident.
Subtle findings, of both the viscera and bowel, can often be difficult to
confirm to clinical standards and require careful correlation with all available
data. Remember that causes of sudden or unexpected death in this cavity are
unusual, compared to the chest.
• There is a small volume of retroperitoneal haemorrhage. In the absence of

a history of trauma, this most likely relates to the attempted cardiopulmo-
nary resuscitation.
• Peri-pancreatic stranding is noted, in keeping with autolysis.
References 269
• Generalised moderate gaseous bowel distension is seen, related to decom-

position. No focal mass or transition point to indicate obstruction.
• Small pneumoperitoneum, small volume of fluid in the pelvis, visceral and
vascular gas are all in keeping with decomposition.
• There is a moderate pneumoperitoneum, more than expected for decompo-
sition, given the relative lack of gas accumulation elsewhere and therefore
judged to be pathological.
• There is advanced decomposition with extensive intra-abdominal soft tis-
sue gas largely replacing the viscera. There is no gross diagnostic feature
in relation to the solid viscera or bowel, but no significant haemorrhage or
osseous trauma is present.
• There is extensive retroperitoneal haematoma surrounding a partially col-
lapsed, partially calcified abdominal aortic aneurysm. Findings are in
keeping with a fatal aortic rupture.
• Generalised omental thickening and ascites, in keeping with the known
history of disseminated malignancy.
References
1. Charlier P, Carlier R, Roffi F, Ezra J, Chaillot PF, Duchat F, et al. Postmortem abdominal CT:
assessing normal cadaveric modifications and pathological processes. Eur J Radiol [Internet].
2012;81(4):639–47. https://linkinghub.elsevier.com/retrieve/pii/S0720048X11000830.
2. Ishida M, Gonoi W, Okuma H, Shirota G, Shintani Y, Abe H, et al. Common postmortem
computed tomography findings following atraumatic death: differentiation between normal
postmortem changes and pathologic lesions. Korean J Radiol [Internet]. 2015;16(4):798.
https://www.kjronline.org/DOIx.php?id=10.3348/kjr.2015.16.4.798.
3. Chatzaraki V, Verster J, Tappero C, Thali MJ, Schweitzer W, Ampanozi G. Spleen measure-
ments with reference to cause of death and spleen weight estimation: a study on postmortem
computed tomography. J Forensic Radiol Imaging [Internet]. 2019;18:24–31. https://linking-
5. Bolster F, Ali Z, Daly B. Postmortem gastromalacia. J Forensic Radiol Imaging [Internet].
2016;5:70. https://linkinghub.elsevier.com/retrieve/pii/S2212478015000386.
6. Jackowski C, Sonnenschein M, Thali MJ, Aghayev E, Yen K, Dirnhofer R, et al. Intrahepatic
gas at postmortem computed tomography: forensic experience as a potential guide for in vivo
trauma imaging. J Trauma Inj Infect Crit Care [Internet]. 2007;62(4):979–88. https://insights.
ovid.com/crossref?an=00005373-200704000-00025.
7. Baumeister R, Gauthier S, Bolliger SA, Thali MJ, Ross SG. Forensic imaging in an unusual
postmortem case of sigmoid volvulus. J Forensic Radiol Imaging [Internet]. 2015;3(3):186–8.
8. Christe A, Ross S, Oesterhelweg L, Spendlove D, Bolliger S, Vock P, et al. Abdominal
trauma—sensitivity and specificity of postmortem noncontrast imaging findings compared
with autopsy findings. J Trauma Inj Infect Crit Care [Internet]. 2009;66(5):1302–7. https://
insights.ovid.com/crossref?an=00005373-200905000-00006.
9. Panda A, Kumar A, Gamanagatti S, Mishra B. Virtopsy computed tomography in trauma:

normal postmortem changes and pathologic Spectrum of findings. Curr Probl Diagn
S0363018815000420.
10. Fatima Z, Tariq U, Khan A, Sohail MS, Sheikh AB, Bhatti SI, et al. A rare case of bilateral
adrenal hemorrhage. Cureus [Internet]. 2018. https://www.cureus.com/articles/13145-a-rare-
case-of-bilateral-adrenal-hemorrhage.
of the Bones and Soft Tissues 10
Introduction
This chapter considers the skeleton and its related soft tissues in more depth, rather
than as part of a body cavity. One should differentiate this group of tissues as the
skeleton or osseous tissue (axial/appendicular) and the soft tissues of the limbs
(mainly muscle, but not forgetting the skin and subcutis, nerves and vessels), con-
sidering each and also their relation to the body as a whole.
Post mortem computed tomography (PMCT) has the advantage of incorporating
a complete body assessment of the bones, joints and soft tissues. This gives signifi-
cant insight into the range and extent of related pathology, traumatic or other, in
addition to the previously discussed body cavity assessment.
Causes of death relating to these tissues are commonly traumatic. Although
trauma is considered ‘unnatural’ in relation to death, it can be ‘non-suspicious’. It
will therefore be commonly encountered as part of the investigation in a coronial/
medico-legal setting. Examples include a witnessed accidental fall, suicide or a
medical event with collapse that results in subsequent injury. If the deaths are not
witnessed, one might reasonably initially regard such cases as suspicious (i.e. poten-
tially needing forensic input) until proven otherwise.
It is important to have broad familiarity with assessment of the musculoskeletal
system, as in contrast to most clinical CT ranges, the post mortem study usually
includes from vertex to toes. Even without a history of trauma, the imaged skeleton
and extremities should be examined. There may be unexpected findings that alter
the evaluation of the case, for example those relating to trauma, but also infection or
neoplasia.
Thin slices and multiplanar reconstructions are especially important, particularly
when the body is not scanned in a perfect orthogonal plane (Fig. 10.1), although
most bodies can be reasonably aligned in the scanner without issue. Reconstruction
software will allow for manipulation of images to confirm bone integrity and align-
ment—particularly important for the spine (Fig. 10.2).

Switzerland AG 2021
https://doi.org/10.1007/978-3-030-70829-0_10
272 10 Post Mortem Computed Tomography of the Bones and Soft Tissues
Fig. 10.1 Volume-

rendered image of the
whole skeleton
demonstrating multiple
fractures following major
trauma. This gives an
overview of how the body
is positioned within the
body bag at the time of
scanning
The musculoskeletal and soft tissues are only assessed to a basic level in a rou-
tine open autopsy (as highlighted in the following section), as more detailed review
requires time-consuming and potentially disfiguring dissection. Often, such investi-
gations have a low yield of significant pathology to justify this.
PMCT may be sufficient to allow the omission of subsequent skeletal or limb
dissection. Alternatively, when such autopsy is planned, it may provide a compli-
mentary assessment or indeed guide the dissection.
utopsy of the Bone and Soft Tissues: The Pathologist’s

A
Perspective
The osseous, muscular and soft tissue compartments are evaluated at open autopsy
to a minor extent, when considering the standard autopsy technique which focuses
on the main organ tissues (see previous chapters). Opening the body, by the standard
incisions, requires the anterior rib plate to be removed by saw/shear cuts along with
some of the chest musculature, in association with a longitudinal incision into the
abdominal wall. This allows a macroscopic review of the alignment of the vertebral
Autopsy of the Bone and Soft Tissues: The Pathologist’s Perspective 273
Fig. 10.2 Reconstructed

view along the long axis of
the vertebral column, on
bone windows. This
reconstruction allows easy
assessment of vertebral
body heights and
alignment, even in cases of
decomposition with
abnormal positioning such
as here
column, once organs from the chest and abdomen have been removed—and is nor-
mally all that is required in terms of bone assessment.
The ribs can be inspected individually by slicing through the musculature and
soft tissue between them, thereby looking for fractures, haemorrhage and/or distor-
tion of tissues (Paget’s disease, tumours, etc.). Consideration of the marrow com-
partment, since most cases involve adults and the elderly, tends to focus on the
vertebrae. This compartment is opened by means of bone chisel excavating through
the coronal plane of the vertebral bodies, producing a vertebral bone strip.
It is uncommon to need to interact with bone fractures or undertake widespread
skeletal dissection, unless there is a particular clinical issue to be addressed. An
example might be a fractured neck of femur, with possible sepsis or poor fixation of
a prosthetic device, prompting resection of the upper femur and the artificial hip
joint in one piece. One should always be mindful of the issues of reconstruction in
cases where body viewing after post mortem will take place. External fixators need
to be removed during autopsy.
Some other devices with regard to bones also require removal. These include
certain orthopaedic implants that have explosive potential during body cremation.
There are similar considerations as to the removal of pacemakers, nerve stimulator
units and other electronic devices, which also have a small explosion risk in crema-
toria (see Chap. 4). One, often forgotten, benefit of modern prosthetic devices is the
unique device serial number and identification code, potentially permitting identifi-
cation of a very decomposed body.
Small fragments of bone and marrow can be taken and subject to fixation, decal-
cification and histology, for example in cases of metabolic bone disease. However,
since most cases of bone pathology revolve around standard osteoporosis and
malignancy, this is rarely an issue. One should also be mindful of potential distal
effects of bone injuries requiring special histology (e.g. stains on snap-frozen lung
tissue to look for fat embolism).
The soft tissues (skin and musculature) are rarely considered beyond macros-
copy, unless there is a specific issue that merits attention. Such cases may have
atypical ulceration in the extremities, superficial injuries or possible neoplasia. In
addition, one may incise the calf and thigh veins in order to consider if there are
residual deep vein thrombotic elements in cases of pulmonary embolism.
Ultimately, each case deserves individual attention and consideration with the
clear understanding that, unlike for imaging, returning to the body is rarely possible,
one should always try to achieve all necessary tests during the autopsy.
The Skin, Subcutaneous Tissues and Muscles
Although directly observable to a viewing pathologist, the radiologist may forget

that the skin and subcutis are relevant soft tissues, which should be considered at the
same time as the bone and muscle. Hypostasis is a normal early post mortem change
that can be seen in the body and limbs on PMCT as dependent skin thickening and
subcutaneous oedema (Figs. 10.3 and 10.4). Generally, such changes are symmetri-
cal. As decomposition progresses, the skin can subsequently blister (Fig. 10.5) and
eventually slip off, see also Chap. 3.
Sometimes, drawing attention, these decomposition changes may be asymmetric
if one side of the body has been more exposed in the open or nearer a heat source
(Figs. 10.6 and 10.7). A close review to exclude potential unilateral pathology (such
as infection, malignancy or venous thrombosis) should be made, as well as consid-
eration of the environment, before judging such findings.
The muscles may be broadly assessed for a normal bulk and symmetry, usually
straightforward when there is sufficient body fat to outline them. Atrophy, hypertro-
phy, traumatic or mass lesions may be noted and correlated with the clinical history.
Significant limb asymmetry may be commonly explained by a known dense hemi-
plegia, congenital abnormalities or unilateral degenerative changes.
Fig. 10.3 Axial view at the level of the distal femurs, on soft tissue windows, showing dependent
thickening of the skin (arrows) and subcutaneous oedema, in keeping with normal post mortem
hypostasis
Fig. 10.4 Axial view of the lower legs on soft tissue windows shows dependent subcutaneous
fluid accumulation with early blistering of the skin (arrows), due to decomposition. The legs are
not seen at exactly the same level due to asymmetric post mortem positioning
The Skeleton
Unlike soft tissues, the skeleton remains generally unaltered for some years after
death. Initially, decomposition causes gas to accumulate in the intraosseous vessels
and marrow (Figs. 10.8 and 10.9), and this is not to be confused with fractures. The
mineralized structure of the bones otherwise remains intact such that they can be
assessed for pathologies akin to the clinical setting (Fig. 10.10).
There may be orthopaedic implants relating to joint replacements or fixation of
previous fractures, unrelated to the cause of death but which should be mentioned
Fig. 10.5 Axial view of both feet on soft tissue windows shows thickened dependent skin due to
hypostasis with additional left-side skin blistering due to decomposition (arrow)
Fig. 10.6 Axial view of the thighs on soft tissue windows shows asymmetric decomposition,
more prominent in the left leg where there is more soft tissue and bone marrow gas resulting in
swelling of the limb

the legs on soft tissue
windows shows swelling,
subcutaneous oedema and
extensive skin blistering of
the lateral left leg (arrows)
due to asymmetric
decomposition (the left
side was nearer to a heat
source)

the spine on bone windows
showing gas in the
vertebral vessels and
marrow (arrows),
potentially simulating
fracture/s. Decomposition
gas in the surrounding
tissues is also present

vertebral body on bone
windows shows linear
vertebral venous plexus
and marrow vessels
(arrows), potentially
mimicking fracture in the
setting of trauma
Fig. 10.10 Coronal view

of the proximal tibias on
bone windows showing
incidental, bilateral bone
infarcts of doubtful acute
significance
for completeness and correlation (Fig. 10.11). Other commonly encountered skele-
tal devices left in place following attempted cardiopulmonary resuscitation (CPR)
are intra-osseous needles placed for emergency vascular access (Fig. 10.12). For
more detail regarding implanted devices see Chap. 4.
Over time, the soft tissues surrounding the bones will decay (Fig. 10.13), eventu-
ally resulting in ‘skeletonisation’. Furthermore, as the ligaments decompose there
will be a variable physical disarticulation at the joints (Fig. 10.14), not to be assumed
traumatic. The process and significant variability of decomposition are discussed in
Chap. 3.
Fig. 10.11 Axial view at the level of the pelvis on soft tissue windows. The body has been scanned
with its ‘arms by side’. There is streak artefact from a distal radial fracture fixation plate (arrow),
unrelated to the cause of death and not severely detrimental to image assessment of the pelvis.
When the artefact is more significant, a further ‘arms up’ acquisition may be made if necessary to
clear this artefact

the right proximal tibia
showing an intra-osseous
needle, placed during
resuscitation attempts, its
tip is in the marrow cavity

shows decomposition
related loss of soft tissue
around a preserved skeletal
structure
Fig. 10.14 Oblique

sagittal view (due to body
positioning) on bone
windows shows advanced
decomposition, almost
skeletonisation, with
multiple spinal
disarticulations (arrows).
There was no history/
suspicion of trauma
Degenerative Changes and Arthritis
While arguably ‘normal’ for aging populations, degenerative osseous findings and
also those associated with the various arthritides are commonly seen on imaging,
often with no relevance to the events surrounding death. PMCT is effective in show-
ing established joint changes. Significant joint pathology should be documented, as
occasionally it may have a relationship with systemic disease (e.g. rheumatoid
arthritis and cardiac pathology) and prior trauma (Figs. 10.15 and 10.16).
Gas is sometimes seen clinically in degenerative intervertebral discs and is also
seen on PMCT (Fig. 10.17). On PMCT, however, such gas should be considered
alongside the gas pattern elsewhere to exclude decomposition. In the setting of
trauma, if gas is seen in a single disc, without evidence of degeneration or decom-
position elsewhere, it may also reflect a traumatic disruption.
Fig. 10.15 Coronal view of the pelvis on bone windows showing left-side hip joint degeneration
(arrow). The left proximal femur is not fully seen as there was flexion at the left hip joint taking the
bone out of plane. Note the normal right hip joint appearances

the pelvis on bone
windows shows severe
bilateral hip joint
degeneration with loss of
joint space, subchondral
cystic and sclerotic
changes, marked
osteophytosis and bony
remodelling
Fig. 10.17 Sagittal partial

view of the thoracolumbar
spine on bone windows
shows small foci of gas in
a disc (large arrow) and
also facet joints (small
arrows), consistent with
degenerative changes
Appendicular Fractures
Any fracture should be documented in terms of bone involved, position of the frag-
ments, and extent of soft tissue injury. Fractures of the proximal femur (Figs. 10.18
and 10.19) may particularly compromise mobility and certainly have associated
morbidity and mortality.
Long bone fractures (Fig. 10.20) may also cause significant pain and haemor-
rhage but are unlikely to be fatal in isolation. Even relatively minor fractures
(Fig. 10.21) may have had an impact on mobility and morbidity and highlight the
value of the ‘whole-body’ approach to the post mortem assessment.
Even when not fatal by itself, the radiologist should bear in mind that a fracture
may be an indirect cause of death. The blood loss may be sufficient, in those with
significant cardiovascular disease, to precipitate an acute cardiac ischaemia and/or
dysrhythmias. The possibility of fat embolism is also often overlooked as a cause of
death in the first 2 days after long bone injury, especially in those fractures which
are surgically treated. Alternatively, resultant immobility may precipitate deep vein
thrombosis and pulmonary embolism.

of the pelvis on soft tissue
windows shows an
unexpected comminuted,
intertrochanteric fracture
of the left proximal femur
(arrows) with minimal
surrounding haematoma

Fig. 10.18, a coronal view
of both legs shows the left
leg to be shortened and
externally rotated
secondary to the hip
fracture
Fig. 10.20 Sagittal view

of the knee showing a
severe fracture-dislocation
with shortening, following
a road traffic collision.
Large fragments of the
femoral condyles are seen
in the popliteal region
(arrows)

both feet on bone windows
shows a healing fracture of
the right second metatarsal
with callus (arrow), seen at
the very lower limits of the
study range
Fig. 10.22 Axial view at the level of the gleno-humeral joints shows an incidental left humeral
head fracture dislocation of unknown/unexplained cause (presumed to have been sustained around
the time of death) in a case of intra-abdominal sepsis. A congruent right gleno-humeral joint is noted
One should also be open to the possibility of a fatal medical event resulting in a col-
lapse, which might result in peri mortem, incidental fracture/s or even that perhaps
injury has been sustained during the post mortem handing of the body (Fig. 10.22).
Vertebral Fractures
Fractures of the spine usually directly involve bones (Figs. 10.23 and 10.24) but can
also relate to discs and/or ligaments with variable disruption to vertebral alignment.
One should be aware that vertebral collapses of an osteoporotic origin/pattern are a
common finding in the elderly and rarely of significance, despite there often being
considerable symptoms in life.
By contrast, fractures of the high cervical spine, or those with vertebral separa-
tion, are much more likely to be relevant to the cause of death, owing to potential
associated brainstem or cord injury. Signs of vertebral separation include straight-
forward widening (Fig. 10.25), yet one should be aware that fractures can re-align.
The position of the bones at the time of scanning may be significantly different to
the bony alignment at the time of death. More subtle misalignment, spur/teardrop
fracture configuration, visible haematoma and intervertebral gas are also relevant
signs and yet can be missed on CT if not meticulous in assessment. Furthermore,
fractures through the disco-ligamentous complex without associated bony injury
may remain occult [1–3] yet have associated cord injuries [4].
A higher suspicion for subtle or occult injuries should be held when assessing a
‘rigid’ spine (e.g. in the setting of diffuse idiopathic skeletal hyperostosis (DISH) or
ankylosing spondylitis, Fig. 10.26). In these conditions, the discs become the rela-
tive points of weakness. Disc heights should be compared to adjacent levels to avoid
missing subtle widening. If there is no history of trauma, one should bear in mind
that a rigid or fragile spine may more readily incur post mortem injury during the
process of body handling/transport [5] or following attempted cardiopulmonary
resuscitation (Fig. 10.27).
As with clinical CT, it can be difficult/impossible to age vertebral fractures, espe-
cially endplate collapse or wedge/compression types (Fig. 10.28). Previous imaging
and/or reports can aid in PMCT assessment, although these are not always available.

of the thoracolumbar spine
on bone windows shows a
comminuted acute fracture
of T12 vertebral body
(large arrow) with lucent
fracture lines. There is no
posterior loss of height or
retropulsion. An anterior
wedge compression
fracture at T7 (small
arrow), is considered
indeterminate in age

of the acute T12 fracture
seen in Fig. 10.23 shows a
vertical fracture line and
upper endplate depression
(arrow)

of the cervical spine on
soft tissue windows
showing fracture through
the C6–7 disc with residual
anterior widening/
separation (arrow)

on bone windows shows
intact, flowing ossification
of the anterior longitudinal
ligament (arrows) resulting
in a rigid spine. The body
is otherwise in a severely
decomposed state

on bone windows showing
a displaced fracture
through the T10–11 disc
space (arrow) in a person
with known ankylosing
spondylitis. This was
judged to be secondary to
chest compressions during
CPR, as there was no other
history of trauma

of the thoracolumbar
junction on bone windows
showing multiple severe
wedge compression
fractures (arrows)
The presence of haematoma or swollen para-spinal musculature suggests that

injury occurred shortly before/about the time of death. Such soft tissue findings are
more easily considered at open autopsy, which has the potential benefit of histology
sampling to aid assessment but are found to varying extent (including not at all) on
PMCT [2]. The PMCT assessment may therefore focus on correlating a known his-
tory of trauma with the relevant findings, on the balance of probabilities.
Spinal Cord Injury
As might be predicted, intrinsic cord lesions, cord haemorrhage, complete and par-
tial transections are more clearly appreciated at open autopsy with histology sam-
pling, compared to PMCT, although this autopsy protocol involves complicated and
extensive dissection. Open autopsy consideration of the spinal cord tissues is not
routinely undertaken, unless specifically indicated by the case data.
PMCT cannot usually directly assess spinal cord injury [6], although it can be
inferred through injury patterns (Fig. 10.29 and see Chap. 5). If cord injury can be
confidently predicted from the available history and correlative imaging, then a
cause of death may be given without the need for open autopsy dissection.

of the thoracic spine on
significant lateral fracture
dislocation at T2–3
(arrows) and inferred
spinal cord injury/
transection, following a
road traffic collision
Soft Tissue Injury
Large peripheral lacerations and soft tissue haematomas can be visualised on PMCT
(Fig. 10.30), although minor soft tissue injuries are better demonstrated at direct
external examination and/or open autopsy [7]. These lesser bleeds and injuries are
unlikely to be of significance in the non-suspicious setting yet, if seen, should
always be suitably explained by the history and circumstances. Unexpected and/or
atypical bruising of the limbs and neck may be signs of neglect or criminality, which
could merit forensic assessment. Many of the specific soft tissue injuries of the vari-
ous body cavities are otherwise considered in their relevant chapters earlier in
the book.
In trauma, angiography has been shown to be helpful in the depiction of trau-
matic lacerations [8]. As isolated injuries of the extremities are not usually the pri-
mary cause of death (unless catastrophic and therefore clearly visualised), it is
difficult to justify the extra cost and resource in undertaking non-targeted/peripheral
angiography in our practice.
Fig. 10.30 Axial view of both upper thighs on soft tissue windows showing asymmetric right leg
swelling due to intramuscular haematoma, evidenced by slightly hyperdense swelling and strand-
ing. This resulted from a fall 1 week prior, on a background history of alcohol misuse with liver
failure and clotting dysfunction
Musculoskeletal Infections
While unusual to come across unexpectedly, the reporter should generally be mind-
ful of infections being present, in case one needs to take measures to protect anyone
that may come into contact with the body. Infective pathology may relate to notifi-
able organisms (e.g. mycobacteria) although confirmation requires microbiology
sampling techniques (potentially via a targeted needle biopsy).
In relation to a potential cause of death, PMCT may reveal established infective
bony pathology, such as osteomyelitis, or vertebral destruction and/or collapse from
spondylo-discitis, which may otherwise have been missed at routine open autopsy
[9]. Yet, as with clinical CT, it cannot reliably exclude such infections. Spread of
infection may also be a cause of death, such as from a systemic bacteraemic sepsis.
Furthermore, one should always consider the possibility of infection in relation to
recent medical/surgical procedures or implanted prostheses, although, just as it is
with clinical imaging, it can be difficult to find imaging evidence of such infective
processes and, for example, to separate a sterile joint effusion from a septic joint.
In the peripheral skeleton, a soft tissue abscess may be appreciated if it is size-
able or if there is associated bone destruction. Chronic ulceration with associated
infection, particularly of the lower limbs, can be identified, commonly seen in the
setting of diabetes and/or peripheral neurovascular disease (Figs. 10.31, 10.32,
10.33, and 10.34). This requires correlation with the clinical history and/or external
findings.

both feet on bone windows
showing soft tissue loss
and ulceration of the right
toes (arrows) in the setting
of severe peripheral
vascular disease
Fig. 10.32 Axial view of the forefeet on soft tissue windows showing asymmetric left-side sub-
cutaneous soft tissue thickening with multiple locules of gas in the tissues, consistent with local-
ised infection and abscess, correlated with the clinical details
Fig. 10.33 Same case as Fig. 10.32, lung windows show the gas more clearly, and the relative
absence of these changes in the right lower limb
Peripheral Vascular Disease
Vessel wall calcification is easily observed on PMCT (Fig. 10.35) and correlates
with established peripheral vascular disease. Whilst a formal limb angiogram might
better assess luminal patency, it is not essential here, as calcification indicates
chronic arterial disease and is usually reflected in the past medical history. Other
vascular findings relevant to significant underlying medical disease include bypass
grafts and haemodialysis fistulae (Fig. 10.36), again usually reflected in the history.
Fig. 10.34 Axial view of both hindfeet on bone windows shows thinned soft tissue with subcuta-
neous gas (infection/localised accelerated decomposition) over the posterior left heel tip (large
arrow). There is cortical bone destruction consistent with osteomyelitis at a clinical site of ulcer-
ation laterally (small arrow)

of the lower legs on soft
heavily calcified 3-vessel
infra-popliteal peripheral
vascular disease

the level of the upper
abdomen, scanned with
‘arms by sides’ on soft
left antecubital fossa
brachiocephalic fistula for
dialysis (arrow). Note also
atrophic left kidney with
vascular calcification

of the spine on bone
sclerotic metastases (but no
vertebral collapse) from
known breast cancer (Note
breast prosthesis in situ)
Musculoskeletal Neoplasia
The diagnosis and characterisation of soft tissue or bone tumour pathology is gener-
ally considered to be limited on clinical CT when compared to MRI. However,
PMCT remains an appropriate, realistic solution for the assessment of primary and
metastatic tumours after death, where the appearances (and limitations) are similar
to clinical CT (Figs. 10.37 and 10.38). A known history of malignancy assists inter-
pretation greatly, but, if this is not given, a broad search for the likely primary should
be undertaken.
Special Circumstances: Fatal Major Trauma 295
Fig. 10.38 Axial view of the lumbar spine on bone windows shows a large, destructive lesion
partially replacing a vertebral body (arrows) with no known history of malignancy. Extensive gas
in soft tissues from decomposition prevented the identification of a primary malignancy on PMCT
Special Circumstances: Fatal Major Trauma
PMCT offers a comprehensive examination of a body that has been subject to cata-
strophic multiple injuries. These may result from circumstances such as a vehicle
(car/train etc.) collision (Figs. 10.39, 10.40, 10.41, and 10.42), a significant fall
(Figs. 10.43 and 10.44) or extensive burns (Figs. 10.45, 10.46, and 10.47). In non-
suspicious (but clearly unnatural) settings, these are likely to relate to suicide or be
witnessed accidents.
Imaging is recognised as a useful adjunct to the post mortem trauma case evalu-
ation [7, 10] but is increasingly being used to replace open autopsy in circumstances
such as this. It provides a permanent and reviewable record of the injuries sustained.
Despite these possibilities, since a pathological external inspection has to occur
in each case (and is necessary for toxicology samples to be obtained), the additional
value of imaging when a body is severely fragmented is questionable in terms of
adding meaningful information to the external assessment.
Interpretation of Fatal Trauma
For the radiologist encountering multiple severe injuries, there can be an over-
whelming assortment of findings to describe, interpret and record, especially if
approached from a clinical perspective. Occasionally, disarticulated body parts may
be placed alongside a body in a body bag (Figs. 10.48 and 10.49), not necessarily in
their correct anatomic location, and so review of the scout studies or volume-
rendered imaging is helpful to provide an overview.
The key is to identify the injury/injuries which, on the balance of probability,
caused death. This may be externally obvious, for example decapitation. Alternatively,
the fatal injury may be internal, such as an aortic transection. When there are multi-
ple and substantial injuries, it may be difficult to ascertain a single, overarching cause
of death. In such cases, following listing of the contributing findings, a summation of
‘multiple traumatic injuries’ may be the best solution.

of the head and cervico-
thoracic spine on bone
severe traumatic injuries
which are incompatible
with life. These resulted
from a road traffic collision

windows shows
catastrophic traumatic head
injury with multiple
fractures and loss of brain
tissues following a road
traffic collision

showing multiple,
extensive crush injuries to
the chest, incompatible
with life, following a road
traffic collision

showing a severely
distracted intervertebral
fracture (with inferred
spinal cord transection)
and herniation of the liver
through the defect (arrow),
collision

windows showing
extensive, unsurvivable
cranio-facial injuries
following a fall from a
high-rise building

shows large bilateral
pneumothoraces secondary
to multiple rib fractures
following a fall from a
high-rise building
The scanned injuries should undoubtedly correlate with the mechanism of injury.
In a non-suspicious case, these data will almost certainly be known and the focus of
the PMCT investigation is to document injuries, seek underlying pathology if pos-
sible, and to confirm trauma as the cause of death.
Seeking to interpret injury patterns without a known mechanism is beyond the
scope of this book, implying a suspicion of criminality and therefore forensic

of the chest and arms on
lung windows showing
extensive superficial soft
tissue loss and a right-side
pneumothorax due to
severe burn injury

Fig. 10.45, axial view of
anterolateral superficial
soft tissue loss due to burn
injury
investigation. This is detailed in other texts [5, 11]. This pattern interpretation is
more concerned with understanding the events surrounding the death, rather than
identify the cause of death itself.
Undoubtedly PMCT can depict skeletal injuries with more ease and accuracy
than a non-forensic open autopsy, but it clearly also has a role in enhancing and
directing forensic skeletal assessment. PMCT can demonstrate a wide range of
major injuries but is reported to be generally less useful for abdominal findings [6],
minor soft tissue and aortic injuries [7]. As such, PMCT is unlikely to be sufficient
alone for forensic cases, as even minor findings may be of critical relevance to the
investigation.
Fig. 10.47 Same case as Fig. 10.45, axial view of both thighs on soft tissue windows shows
extensive superficial soft tissue loss due to burn injury
Fig. 10.48 Non-

orthogonal view of an
amputated foot (which was
also placed in the body
bag) on bone windows
showing multiple fractures
and severe soft tissue
injury following a road
traffic crush injury
Fig. 10.49 Volume

rendered image of multiple
skeletal fractures and
disarticulated body parts
scanned within a body bag
after a train collision
Fatal Haemorrhage as a Consequence of Injury
Determining haemorrhage/exsanguination as a cause of death can be difficult at

both open autopsy and PMCT. This diagnosis may be considered when trauma is
less extensive than some of the examples described earlier, yet where there has been
critical injury to a vascular structure. Factors to consider include the history, docu-
mentation a large volume of haemorrhage at the scene or presence of profuse inter-
nal haemorrhage. Police and paramedic reports may occasionally offer this detail,
although the descriptions vary in completeness (and accuracy) in the non-suspicious
setting.
In terms of imaging findings, as expected, haemorrhage is generally hyper-
dense, although the blood can separate with hypostasis, resulting in a layered
‘fluid–fluid’ level (Fig. 10.50). One should appreciate that, with such appearances,
it is the whole volume of fluid that is ‘blood’, not just the dependent hyperdense
component which only represents the sedimented cell fraction. Most patients who
die of fatal haemorrhage will show a general collapse of their vessels [12]. This
feature will support the diagnosis but is unfortunately both non-specific and com-
mon on PMCT.

extensive intra-abdominal
haemorrhage with layered
separation (small arrow)
possibly secondary to the
splenic laceration (large
arrow), although
potentially further
parenchymal and vascular
injury may be obscured
Adding to the difficulty, when a large volume of internal haemorrhage is present,

the origin may be speculative as haematoma can obscure the underlying vascular
anatomy. Interpretation of the history and ‘central’ or ‘maximal’ location of blood
may reveal the most likely site. Where available, post mortem angiography
(PMCTA) may permit a more specific localisation [8], but it should be understood
that this adds significantly to the complexity and cost of the PMCT.
Reporting Bone and Soft Tissue Findings: Pearls and Pitfalls
The role of the radiologist (in a non-suspicious trauma case) is to interpret and
document the injuries and thereby deduce a likely medical cause of death.
This may be a single directly fatal injury, a summation of multiple injuries or
a related cause such as exsanguination.
In general, any fractures, soft tissue asymmetry/pathology, significant vas-
cular calcification, degenerative changes, joint replacements and implants
should be mentioned and be correlated with clinical data and/or external
findings.
Commentary of osteopenia may be of relevance in considering fractures.
It is also suggested to routinely document the absence of fractures, as this
bony assessment is a key advantage of PMCT over routine open autopsy. This
also obviates the need for speculative extensive body dissection should an
invasive examination proceed.
• No acute fractures demonstrated.

• No destructive or sinister bone lesions.
• No soft tissue mass, swelling or secondary signs of deep venous thrombo-
sis in the lower extremities.
References 303
• There are multiple catastrophic, unsurvivable cranial and vertebral injuries.

• Destructive lucency of the bone is seen with associated acute fracture, sus-
picious of pathological fracture.
• Age indeterminate vertebral wedge fractures are seen with diffuse osteope-
nia (likely osteoporosis).
• Mid thoracic vertebral fracture, in the absence of a traumatic history, is
judged to be injury from attempted cardiopulmonary resuscitation.
References
1. Kudo S, Kawasumi Y, Usui A, Arakawa M, Yamagishi N, Igari Y, et al. Post-mortem com-
puted tomography of cervical intervertebral separation: Retrospective review and comparison
of the autopsy results of 57 separations. J Forensic Radiol Imaging [Internet]. 2018;12:57–63.
2. Kawasumi Y, Usui A, Hosokai Y, Sato M, Hayashizaki Y, Saito H, et al. PMCT findings of
intervertebral separation. J Forensic Radiol Imaging [Internet]. 2014;2(4):182–7. https://link-
inghub.elsevier.com/retrieve/pii/S2212478014001051.
3. 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 [Internet].
4. Makino Y, Yokota H, Hayakawa M, Yajima D, Inokuchi G, Nakatani E, et al. Spinal cord inju-
ries with normal postmortem CT findings: a pitfall of virtual autopsy for detecting traumatic
death. Am J Roentgenol [Internet]. 2014;203(2):240–4. http://www.ajronline.org/doi/10.2214/
AJR.13.11775.
book/9780340972533.
6. Panda A, Kumar A, Gamanagatti S, Mishra B. Virtopsy computed tomography in trauma: Nor-
mal postmortem changes and pathologic Spectrum of findings. Curr Probl Diagn Radiol [Inter-
net]. 2015;44(5):391–406. https://linkinghub.elsevier.com/retrieve/pii/S0363018815000420.
7. Jalalzadeh H, Giannakopoulos GF, Berger FH, Fronczek J, van de Goot FRW, Reijnders UJ,
et al. Post-mortem imaging compared with autopsy in trauma victims—a systematic review.
Forensic Sci Int [Internet]. 2015;257:29–48. https://linkinghub.elsevier.com/retrieve/pii/
S0379073815003047.
pii/S2212478014001075.
10. Scholing M, Saltzherr TP, Fung Kon Jin PHP, Ponsen KJ, Reitsma JB, Lameris JS, et al.
The value of postmortem computed tomography as an alternative for autopsy in trauma vic-
tims: a systematic review. Eur Radiol [Internet]. 2009;19(10):2333–41. http://link.springer.
com/10.1007/s00330-009-1440-4.
11. Levy AD, Harcke HT. Essentials of forensic imaging [Internet]. Boca Raton: CRC Press; 2010.
https://www.taylorfrancis.com/books/9781420091120.
12. Aghayev E, Sonnenschein M, Jackowski C, Thali M, Buck U, Yen K, et al. Postmortem radiol-
ogy of fatal hemorrhage: measurements of cross-sectional areas of major blood vessels and vol-
umes of aorta and spleen on MDCT and volumes of heart chambers on MRI. Am J Roentgenol
Findings Related to Attempted
Cardiopulmonary Resuscitation on Post 11
Mortem Computed Tomography
Introduction
As indicated throughout this book, there are certain findings on post mortem com-
puted tomography (PMCT) which are commonly seen after attempted cardiopulmo-
nary resuscitation (CPR). It is important to appreciate that CPR itself is a form of
trauma and so findings could significantly overlap with trauma from other causes.
This range of features, particularly those found in the thorax, are important to
recognise in order to avoid misinterpretations that might lead to an incorrect cause
of death or misidentification of pre-existing disease [1, 2]. It is understood that, even
with experienced radiological interpretation, it may not be possible to completely
differentiate the possible aetiologies of the features seen.
In addition to the investigation of cause of death, PMCT can also be used to pro-
vide post-resuscitation feedback (often in terms of case audit) to the paramedic and
medical teams involved in order to aid learning [3, 4]. Rarely, errors identified by
PMCT can be instrumental in pointing to training needs, for example practitioners
failing to correctly intubate the airway. On occasion, despite history of attempted
CPR (often bystander), there are no appreciable imaging findings related to the
efforts.
In many circumstances, however, it is of some reassurance to the CPR practitio-
ners and families to document that vigorous efforts were made to render assistance
to the deceased.
Autopsy Following CPR: The Pathologist’s Perspective
It would be fair to state that many of the cases that require open autopsy have had
variable resuscitation measures, including assisted ventilation and chest compres-
sions, alongside other medical strategies. The assessment of such cases deals with
the background to the cardiorespiratory arrest and death with its underlying

Switzerland AG 2021
https://doi.org/10.1007/978-3-030-70829-0_11
306 11 Findings Related to Attempted Cardiopulmonary Resuscitation on Post Mortem…
pathologies, but also has to address the impact and adequacy of any resuscitation
efforts. This is covered in greater detail in other autopsy reference texts [5].
Open autopsy on the body starts with the external perspective, noting the pres-
ence of lines, drains, artificial airways, etc. It is generally recommended that no
devices or lines are removed from the body following failed resuscitation, before
autopsy, as these items may be of material significance in assessment the case, along
with medicolegal impact.
Any device entering the body needs to be checked for position and local compli-
cation. The pathologist may cut such lines flush with the body, perhaps pushing
inward slightly in order to avoid displacement during the body handling. This pro-
cess is particularly important for any item that enters the thoracic or abdominal
cavity, with the need to check for associated trauma and/or infection.
The autopsy approach is standard for all cases [5]. In post-resuscitation cases,
most interest focuses on the chest and to a degree the abdomen/retroperitoneum.
The general access point for the chest, abdomen and related tissues is through a
Y-shaped incision around the upper neck with a longitudinal incision from the ster-
num down to the pubis allowing reflection of skin, soft tissue, muscle and removal
of the rib plate (sternum and anterior ribs). This process allows good inspection of
the cavities and immediate evaluation of issues that are pertinent to the resuscita-
tion—or the cause of the cardiorespiratory arrest.
Rib fractures are common in resuscitation cases and generally not considered in
detail unless there is underlying mediastinal, cardiac or lung injury. On occasion,
some rib (or other bone) may be retained for histology, in order to consider osteopo-
rosis or underlying metabolic bone disease. However, this investigation is very rare.
Macroscopic photographs are vital in cases of perceived complications from
resuscitation, such as liver tears and haemorrhage or misplacement of airways/chest
drains. There should be a low threshold for autopsy histology in this scenario.
Cardiopulmonary Resuscitation
This broad term is used to describe a variety of emergency, potentially lifesaving,

activities that centre on chest compressions and artificial/assisted ventilation. With
medical (rather than bystander) intervention, there may also be the placement of
airway adjuncts and vascular access lines, administration of drugs and/or fluids and
occasionally cavity drain insertions. It is to be remembered that cardiac defibrilla-
tion is variably available in the community but is standard in hospital/medical
settings.
As always, the background case data are vital, including the prior medical history
supplemented by ambulance or hospital notes. One particular piece of information
needed for the correlation of findings on PMCT is whether chest compressions took
place, although this is usually assumed when the abbreviation term ‘CPR’ is used.
Some of the CPR interventions will be physically evident (e.g. when defibrillator
pads, intraosseous/vascular lines or artificial airways are left in place, Figs. 11.1 and
11.2, see also Chap. 4). Other interventions leave secondary evidence (such as
Cardiopulmonary Resuscitation 307
fractures resulting from chest compressions, Fig. 11.3) and are further discussed in
this chapter.
During CPR, the correct placement and position of relevant tubes and devices is
crucial to ensure effectiveness. By the time of PMCT, their positions may have
altered and may no longer be the same as when CPR was in progress (for example
artificial airways may be pushed further inwards, Fig. 11.4). Considerable caution is
advised when the radiologist is tempted to use the term ‘misplaced’ with respect to
any medical device, especially if this was unlikely to have changed the final out-
come. Such comment might be misunderstood as indicating imperfect medical
treatment and cause unnecessary distress to relatives.

the upper chest on bone
windows shows a thin,
radio-dense defibrillator
pad on the skin of the right
anterior chest wall (arrow)

the right proximal tibia,
windowed to show an
its tip in the marrow cavity

shows a defibrillator pad
over the right anterior chest
wall, bilateral rib fractures
(arrows), tiny volume of
intra-cardiac gas and
extensive lung opacity
following CPR attempts
for 1 hour

windows following CPR
attempts shows an
endotracheal tube in the
right main bronchus
(arrow)
Skeletal Findings on PMCT
Rib and Cartilage Fractures
A very common finding after CPR attempts with chest compressions is of multiple
rib fractures. These are usually bilateral, anterior or anterolateral and involve the
second to seventh ribs [3] or more commonly the third to sixth ribs [6].
These rib fractures may be ‘complete’ (Figs. 11.5 and 11.6) or ‘incomplete’
(often buckle type), usually involving the inner cortex, as this side is compressed,
(Figs. 11.7 and 11.8). Occasionally, a combination of types may be seen (Figs. 11.3,
11.9, and 11.10). In addition, or sometimes instead of fractured ribs, fractures of the
costal cartilages may be seen (Figs. 11.11, 11.12, and 11.13).
When fractures are complete there may be displacement of the bone/cartilage
ends, relating to a combination of the intensity of the resuscitation, the pre-existing
bone quality and the background chest compliance. Yet, the ribs do not have to be
visibly displaced on the scan to have caused underlying injuries to the mediastinum,
lungs or upper abdomen. During the multiple, rapid physical compressions of the
Skeletal Findings on PMCT 309

right anterior rib on bone
windows shows a complete
rib fracture following CPR
attempts

left anterior rib on bone
windows shows a complete
rib fracture following CPR
attempts

right anterior rib on bone
windows shows an
incomplete buckle fracture
of the inner cortex

left anterior rib on bone
windows shows an
incomplete buckle fracture
of the inner cortex

showing bilateral anterior
rib fractures resulting from
CPR attempts, buckle type
on the right and displaced,
complete on the left

showing bilateral anterior
rib fractures resulting from
CPR attempts, complete
with displacement on the
right and buckle on the left
with small underlying
pneumothorax

of the anterior chest wall
shows multiple, bilateral,
vertically orientated costal
cartilage fractures resulting
from CPR attempts
Fig. 11.12 Coronal

maximum intensity
projection of the anterior
chest wall shows multiple,
undisplaced bilateral costal
cartilage fractures resulting
from CPR attempts
chest, their displacement may have been more considerable. Occasionally, there are
many displaced fractures that do not return to a normal chest architecture. This
results in a residual deformity, such as a depressed sternum and/or anterior chest
wall (Figs. 11.13, 11.14, and 11.15).

windows shows a
depressed left parasternal
costal cartilage fracture
(arrow) resulting in chest
wall deformity and
compression of mediastinal
structures

of the chest on bone
sternal fractures (arrows)
and depression of the
sternum
By contrast, and important to bear in mind, posterior rib fractures are generally
considered to be inconsistent with CPR (Fig. 11.16), although these might be attrib-
uted to resuscitation attempts if an external mechanical chest compression device
has been used [6]. With use of such devices, the number of fractures demonstrated
may also generally be higher [6].
It has been reported that PMCT has a low sensitivity for rib fractures compared
to a forensic autopsy where ribs are individually dissected [7] and potentially

windows showing sternal
depression secondary to
bilateral rib fractures and
flattening of the anterior
chest wall

showing an acute right
posterior rib fracture
(arrow) following a road
traffic collision
subject to histology. However, this is not considered to be the case when compared
to a routine coronial autopsy—where the skeleton is only minimally reviewed and
dissected. PMCT also offers a record of skeletal appearances prior to removal of the
chest plate or other dissection at open autopsy. This can be reviewed later, poten-
tially days and weeks (or years!) following the autopsy for correlation.
Sternal Fractures
Sternal fractures are also sometimes seen following CPR attempts. These are usu-
ally of the mid-sternum and can either be complete (Figs. 11.17 and 11.18) or
incomplete (Figs. 11.19, 11.20, and 11.21). The PMCT report should describe such
fractures and the residual degree of displacement (which, just as for the ribs, may
have been significantly more during chest compressions) to correlate with any asso-
ciated mediastinal injury.

of the sternum on bone
windows showing a
complete but minimally
displaced sternal fracture
(arrow) following CPR
attempts

windows showing a
displaced mid-sternal
fracture (arrow) following
CPR attempts

windows showing an
incomplete, anterior cortex
sternal fracture (arrow)

windows showing an
incomplete, anterior cortex
sternal fracture (small
arrow) following CPR
attempts and an incidental,
well-corticated sternal
foramen inferiorly (large
arrow)

windows showing a
buckled but incomplete
sternal fracture (arrow)
Vertebral Fractures
More unusually, fractures of the mid thoracic spine (often around the T6 level) have
also been reported following CPR [8]. The incidence appears higher with the use of
mechanical CPR devices and more likely if there is osteopenia/osteoporosis, rigid-
ity of the spine or existing kyphosis (Figs. 11.22 and 11.23). To make the interpreta-
tion of CPR-related vertebral fractures, one will need verification of the absence of
trauma before the cardio-respiratory arrest and exclusion of a post mortem handling
or transport-related injury. This means that cases that sustained trauma before death
are more difficult to interpret (Fig. 11.24). One may consider that the absence of
significant haematoma around a vertebral fracture supports the interpretation of a
post mortem nature.

of the thoracic spine on
fracture of an intervertebral
disc space (arrow) in the
setting of ankylosis and
kyphosis. There was no
known history of trauma
other than chest
compressions performed
during CPR attempts

of the chest on bone
windows showing a sternal
fracture (small arrow) and
a corresponding level
fracture through the
intervertebral disc space
(large arrow), both
sustained following chest
compressions in the setting
of a degenerate, partially
rigid spine and background
of osteopaenia

of the thoracic spine
windowed to show a
vertebral canal haematoma
(small arrows) centred
around an acute, transverse
T7 fracture (large arrow),
involving the posterior
elements. The patient was
found unresponsive at the
bottom of the stairs
(presumed fall) and
cardiopulmonary
resuscitation was
attempted. Note also a
sternal fracture at the same
level of the vertebral
fracture. In this case it is
difficult to determine
whether the vertebral
fracture and/or haematoma
are related to the initial
trauma or chest
compressions
Soft Tissue Findings on PMCT
Soft tissue findings relating to CPR are commonly secondary to the chest wall frac-
tures described earlier, although it is possible to have isolated soft tissue injury
resulting from CPR attempts, especially in the younger population.
Small haemo/pneumothoraces, pneumomediastinum, lung contusions or lacera-
tions, small haemopericardium and even peri-hepatic, peri-splenic and retro-perito-
neal haemorrhages have all been reported to potentially be secondary to CPR-related
soft tissue injury [9].
It can be difficult to distinguish between CPR-related lung contusion and other
pathology such as pulmonary oedema, infection or post mortem atelectasis. Fluid
resuscitation may also complicate the picture. Often, following CPR, the lungs
appear very congested (Fig. 11.25) with non-specific diffuse ground-glass opacity
and areas of collapse and consolidation. These broad, overlapping differential diag-
noses often limit PMCT diagnostic interpretation.
Soft Tissue Findings on PMCT 319
Haemothorax and Pneumothorax
PMCT is sensitive to even tiny pneumothoraces (Fig. 11.25), more so than traditional
autopsy methods that require time-consuming, special techniques to accurately dem-
onstrate gas in the chest cavities. When reporting a pneumothorax, it is imperative
that one must also consider the level of decomposition, so as not to falsely attribute
autolytic gas production to that indicative of trauma/pathology (see Chap. 3).
The volume of a CPR-related post mortem haemo/pneumothorax is variable and
dependent on multiple factors such as the manual or mechanical vigour of CPR and
the length of resuscitation attempts. Any underlying coagulopathy may have bear-
ing on local blood loss. Resuscitation-linked blood and air collections (Fig. 11.26),
however, tend to be smaller in volume when compared to those resulting directly
from fatal pathologies.
That being said, it is important to appreciate that a pneumothorax may be exac-
erbated by artificial ventilation devices. The use of external mechanical chest com-
pression devices is generally associated with greater soft tissue injury and
haemorrhage [6] (Figs. 11.27 and 11.28).

showing tiny bilateral
resulting from rib fractures
sustained during attempted
CPR. The lungs are
opacified with mixed
ground glass density and
patchy consolidation and
there is also an
endotracheal tube with cuff
inflated in the trachea

the lower chest on soft
small layered density in the
right pleural space,
indicative of separated
blood products/
haemothorax and judged to
be secondary to CPR
related injury. Note the
right anterolateral buckled
rib fracture and
asymmetric chest wall
Fig. 11.27 Axial view of the chest on soft tissue windows shows a moderate size, right side lay-
ered haemothorax (arrow) after prolonged cardiopulmonary resuscitation attempts using an exter-
nal mechanical chest compression device. The right atrium is dilated and contains a similar layered
separation of blood products

shows a right anterior
pneumothorax after
prolonged
cardiopulmonary
resuscitation attempts
using an external
mechanical chest
compression device and
ventilation via
endotracheal intubation
(not seen on this image)
It is emphasised that rib fractures after CPR are extremely unlikely to cause a
post mortem aortic rupture, leading to massive haemothoraces. Aortic injury is by
far more in keeping with true pathology, suggested by massive trauma/penetrating
chest injury or primary rupture in the setting of chest pain with sudden collapse.
Pre-Sternal Haematoma
If an external mechanical chest compression device has been used during CPR,
there is a higher incidence of subcutaneous pre-sternal haematoma formation [9]
(Figs. 11.29 and 11.30). Otherwise, these are rarely seen.

windows shows a
pre-sternal soft tissue
haematoma (arrow)
using an external
mechanical chest
compression device

Fig. 11.29, sagittal view of
the sternum shows a
pre-sternal haematoma
(large arrow) and a fracture
of the anterior cortex of the
sternum (small arrow) after
CPR attempts using an
external mechanical chest
compression device
Intravascular Gas and Dilated Right Atrium
Theories to explain gas in the vasculature following CPR (outside the setting of fatal
trauma) include that it results from intravenous catheterisation, pulmonary injury
from chest compressions or possibly pneumatisation of gas that was dissolved in the
blood. This gas can be seen particularly in the cardiac chambers (Figs. 11.3 and
11.31) and liver [1, 2]. The differential is commonly seen decomposition gas, yet
both of these origins should be distinguished from gas relating to pathology, for
example in the abdomen.
Dilatation of the right atrium is also a feature seen on PMCT after CPR attempts,
possibly due to right heart ‘congestion’ and/or increased intravascular fluid admin-
istered during resuscitation (Figs. 11.27 and 11.31). Fluid shifts within the circula-
tion (realignment of blood) may also reflect the equalisation of pressures in the
various vascular compartments after death. This finding is variable and difficult to
quantify against potentially pre-existing atrial dilatation without comparative imag-
ing [1, 10].

locules of intracardiac gas
(arrows), dilated right
atrium, right anterolateral
rib fracture and extensive
lung congestion seen after
1 hour of in-hospital CPR
following a suicide by
hanging
Abdominal Findings Following CPR
Small to moderate volumes of intra-peritoneal haemorrhage (Fig. 11.32) and retro-

peritoneal haemorrhage (Figs. 11.33 and 11.34) may be seen following CPR
attempts, when no other explanation (such as abdominal trauma or a ruptured aorta)
is found. Such collections are presumed to reflect tiny visceral lacerations of the
liver and/or spleen although these may not be clearly appreciated on routine PMCT.
In hospital, when resuscitation attempts follow radiological investigations or
procedures involving iodinated contrast, haemorrhage may be even more hyper-
dense than usual owing to contrast material held in the circulation prior to cardiore-
spiratory arrest (Fig. 11.35).
Distension of the stomach and gastrointestinal (GI) tract can be seen following
CPR with artificial respiration using a facemask or poorly fitting laryngeal mask
(often used prior to intubation), as gas passes into both the trachea and the oesopha-
gus (Fig. 11.36). CPR is also reported to cause intra-mural GI tract or intra-hepatic
gas [10] although, to confidently report this finding, it would have to be seen as out
of proportion to decomposition gas elsewhere and without any prior abdominal
symptoms that might suggest acute GI pathology.

small volume of high-
density fluid dependently
around the liver and spleen
(arrows) judged to be
intraperitoneal
haemorrhage following
CPR attempts

small volume of streaky
right retroperitoneal space
CPR attempts. There was
no history of abdominal
trauma and no
haemorrhage around the
collapsed aorta

small volume of streaky
left retroperitoneal space
CPR attempts. There is no
haemorrhage around the
non-aneurysmal aorta
Fig. 11.35 Axial view of the upper abdomen on soft tissue windows showing peri-hepatic haem-
orrhage (arrow) judged to be secondary to liver injury during in-hospital CPR. This appears more
dense than expected due to the presence of iodinated contrast from preceding radiological inter-
vention. Contrast also ‘enhances’ the kidneys to reveal several simple cysts

windows shows a markedly
distended, gas filled
stomach following
prolonged resuscitation
attempts with multiple
airway interventions
References 325
Reporting CPR-Related Findings: Pearls and Pitfalls
A history of whether CPR was attempted (and specifically chest compres-

sions) should be available, or sought, when considering PMCT. If these data
are not available, then caution should be taken when interpreting the scan
findings.
A description of the positions of lines/tubes etc. may be given in the under-
standing that their position may have altered after death.
Posterior rib fractures are generally considered inconsistent with CPR.
Knowledge of the use of external chest compression devices is important,
as this may result in more CPR-associated injury.
When CPR has been performed, a description of the findings can be made
with a statement to suggest these are ‘in keeping’ or ‘consistent with’ injury
from attempted CPR.
• It is noted from the supporting information that cardiopulmonary resusci-

tation (CPR) including chest compressions was attempted.
• Endotracheal tube with tip located just above the carina at the time of
scanning.
• Bilateral anterior rib buckle fractures / sternal fractures in keeping with
attempted CPR.
• Small haemothorax/pneumothorax is in keeping with CPR related injury.
• Small volume of upper abdominal haemorrhage judged most likely to have
resulted from the prolonged CPR attempts.
References
1. Murakami T, Uetani M, Ikematsu K. Postmortem CT in emergency department: influence of
cardiopulmonary resuscitation. Poster session presented at: European Congress of Radiology;
2012 March 1–5; Vienna, Austria. [Internet]. https://doi.org/10.1594/ecr2012/C-1440.
2. Offiah CE, Dean J. Post-mortem CT and MRI: appropriate post-mortem imaging appear-
ances and changes related to cardiopulmonary resuscitation. Br J Radiol [Internet].
2016;89(1058):20150851. http://www.birpublications.org/doi/10.1259/bjr.20150851.
3. Bolster F, Ali Z, Fowler D, Daly B. Imaging of resuscitation and emergency resuscitation
devices—Lessons learned from post mortem computed tomography. J Forensic Radiol Imaging
4. Lotan E, Portnoy O, Konen E, Simon D, Guranda L. The role of early postmortem CT in
the evaluation of support-line misplacement in patients with severe trauma. Am J Roentgenol
5. Suvarna SK, editor. Atlas of adult autopsy [Internet]. 1st ed. Cham: Springer International
Publishing; 2016. http://link.springer.com/10.1007/978-3-319-27022-7.
6. Koga Y, Fujita M, Yagi T, Nakahara T, Miyauchi T, Kaneda K, et al. Effects of mechanical
chest compression device with a load-distributing band on post-resuscitation injuries i dentified
by post-mortem computed tomography. Resuscitation [Internet]. 2015;96:226–31. https://link-

inghub.elsevier.com/retrieve/pii/S030095721500386X.
7. Schulze C, Hoppe H, Schweitzer W, Schwendener N, Grabherr S, Jackowski C. Rib frac-
tures at postmortem computed tomography (PMCT) validated against the autopsy. Forensic
Sci Int [Internet]. 2013;233(1–3):90–8. https://linkinghub.elsevier.com/retrieve/pii/
S0379073813004027.
8. Bohara M, Ohara Y, Mizuno J, Matsuoka H, Hattori N, Arita K. Cardiopulmonary resuscitation-
induced thoracic vertebral fracture: a case report. NMC Case Rep J [Internet]. 2014;2(3):106–8.
https://www.jstage.jst.go.jp/article/nmccrj/2/3/2_cr.2014-0383/_article.
9. Baumeister R, Held U, Thali MJ, Flach PM, Ross S. Forensic imaging findings by post-
mortem computed tomography after manual versus mechanical chest compression. J Forensic
S2212478015300095.
10. Ishida M, Gonoi W, Okuma H, Shirota G, Shintani Y, Abe H, et al. Common postmortem
computed tomography findings following atraumatic death: differentiation between normal
postmortem changes and pathologic lesions. Korean J Radiol [Internet]. 2015;16(4):798.
https://www.kjronline.org/DOIx.php?id=10.3348/kjr.2015.16.4.798.

Post-Mortem CT Imaging

Uploaded by

Copyright:

Available Formats

Post-Mortem CT Imaging

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Post-Mortem CT Imaging

Uploaded by

Copyright:

Available Formats

Post Mortem CT

ISBN 978-3-030-70828-3 ISBN 978-3-030-70829-0 (eBook)

Post mortem computed tomography (PMCT) is an evolving diagnostic arena deal-

Sheffield, UK Ayeshea Shenton

1 Introduction to the Investigation of Death and Post Mortem

Livor Mortis or Hypostasis�������������������������������������������������������������������� 38

6 Post Mortem Computed Tomography of the Extra-Cranial

Fluid in the Major Airways�������������������������������������������������������������������� 191

Abdominal Trauma�������������������������������������������������������������������������������� 265

© The Author(s), under exclusive license to Springer Nature 1

Fig. 1.1 Coronal PMCT

cases should ideally be assessed on an individual basis, although some centres

1. who has died

The Arrival of Post Mortem Imaging

PMCT is particularly effective in the investigation of sudden death, where a cata-

The Emerging Subspecialty of PMCT

potentially adversely impact on their personal professional accreditation. It is

The Future of PMCT

17. Heinemann A, Vogel H, Heller M, Tzikas A, Püschel K. Investigation of medical interven-

Factors Governing the Choice of PMCT

© The Author(s), under exclusive license to Springer Nature 15

PMCT Scanning Facility Options

Fig. 2.1 Post mortem

Fig. 2.2 The vitreous can

Fig. 2.3 The bladder is

The Radiographer in the PMCT Unit

The conventional practice of PMCT requires a trained radiographer working with

An experienced radiographer can also offer a wealth of knowledge when setting

Fig. 2.4 Scanogram/scout

Body Handling and Positioning

Fig. 2.5 A PMCT scanner

Fig. 2.6 Coronal view of

Fig. 2.7 Coronal view of

Fig. 2.8 Sagittal view of

Fig. 2.9 Axial view of the

Contrast Media and PMCT Angiography

The Radiologist and PMCT

take particular interest in the wording of a radiology report, particularly if it is at

The PMCT Report

that sometimes the information is incomplete or preliminary in its nature and on

Broadly speaking, straightforward cases with catastrophic findings and a history

As indicated at the start of the chapter, there is no single solution to reporting a

 rain, Head and Neck

Abdomen and Pelvis

Quality Assurance and Audit

be invaluable for discussion of findings, learning, continuous professional develop-

Death and its Broad Causes

© The Author(s), under exclusive license to Springer Nature 35

Fig. 3.1 Early-stage

Post Mortem Interval

Initial Changes Seen After Death

This is the earliest change identified following cessation of breathing, circulation

Livor Mortis or Hypostasis

Fig. 3.2 Hypostasis is

Fig. 3.3 Hypostasis

On PMCT, hypostasis can also be readily appreciated. Assuming a supine posi-

Fig. 3.4 Axial view of the

Fig. 3.5 Axial view of the

Fig. 3.6 Axial view of the

Sheffield, UK Ayeshea Shenton

1 Introduction to the Investigation of Death and Post Mortem

Livor Mortis or Hypostasis�� 38

6 Post Mortem Computed Tomography of the Extra-Cranial

Fluid in the Major Airways�� 191

Abdominal Trauma�� 265

The Arrival of Post Mortem Imaging

The Emerging Subspecialty of PMCT

The Future of PMCT

Factors Governing the Choice of PMCT

PMCT Scanning Facility Options

The Radiographer in the PMCT Unit

Body Handling and Positioning

Contrast Media and PMCT Angiography

The Radiologist and PMCT

The PMCT Report

rain, Head and Neck

Abdomen and Pelvis

Quality Assurance and Audit

Death and its Broad Causes

Post Mortem Interval

Initial Changes Seen After Death

Livor Mortis or Hypostasis

Later Changes in the Body After Death

Decomposition from the Radiology Perspective

Post Mortem Gas

Post Mortem Clot

Animal Predation and Post Mortem Changes

Embalmed and Previously Autopsied Bodies

The External Examination: The Pathologist’s Perspective

External Findings on PMCT

Tubes Seen on PMCT