Forensic Neuropathology [2 ed.] 2020045920, 2020045921, 9781498706162, 9781003158035, 9780367744601

Citation preview

FORENSIC NEUROPATHOLOGY

FORENSIC NEUROPATHOLOGY Second Edition

Edited by

Helen Whitwell

Formerly Professor of Forensic Pathology University of Sheffield and Neuropathologist West Midlands, UK

Christopher Milroy

Forensic Pathologist and Professor Department of Pathology and Laboratory Medicine University of Ottawa, Canada

Daniel du Plessis

Consultant Neuropathologist Neuropathology Unit Department of Cellular Pathology and Greater Manchester Clinical Neurosciences Centre Salford Royal Hospital NHS Foundation Trust Liverpool, UK

Second edition published 2021 by CRC Press 6000 Broken Sound Parkway NW, Suite 300, Boca Raton, FL 33487-2742 and by CRC Press 2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN © 2021 Taylor & Francis Group, LLC First edition published by Edward Arnold (Publishers) Ltd. 2005 CRC Press is an imprint of Taylor & Francis Group, LLC This book contains information obtained from authentic and highly regarded sources. While all reasonable efforts have been made to publish reliable data and information, neither the author[s] nor the publisher can accept any legal responsibility or liability for any errors or omissions that may be made. The publishers wish to make clear that any views or opinions expressed in this book by individual editors, authors or contributors are personal to them and do not necessarily reflect the views/opinions of the publishers. The information or guidance contained in this book is intended for use by medical, scientific or healthcare professionals and is provided strictly as a supplement to the medical or other professional's own judgement, their knowledge of the patient's medical history, relevant manufacturer's instructions and the appropriate best practice guidelines. Because of the rapid advances in medical science, any information or advice on dosages, procedures or diagnoses should be independently verified. The reader is strongly urged to consult the relevant national drug formulary and the drug companies’ and device or material manufacturers’ printed instructions, and their websites, before administering or utilizing any of the drugs, devices or materials mentioned in this book. This book does not indicate whether a particular treatment is appropriate or suitable for a particular individual. Ultimately, it is the sole responsibility of the medical professional to make his or her own professional judgements, so as to advise and treat patients appropriately. The authors and publishers have also attempted to trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained. If any copyright material has not been acknowledged, please write and let us know so we may rectify in any future reprint. Except as permitted under US Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. For permission to photocopy or use material electronically from this work, access www.copyright.com or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. For works that are not available on CCC please contact [email protected] Trademark Notice: Product or corporate names may be trademarks or registered trademarks and are used only for identification and explanation without intent to infringe. Library of Congress Cataloging‑in‑Publication Data Names: Whitwell, Helen L., editor. | Milroy, Christopher (Christopher M.), editor. | Du Plessis, Daniel (Daniel G.), editor. Title: Forensic neuropathology / edited by, Helen Whitwell, Christopher Milroy, Daniel du Plessis. Other titles: Forensic neuropathology (Whitwell) Description: Second edition. | Boca Raton : CRC Press, 2021. | Includes bibliographical references and index. | Summary: “Forensic Neuropathology is written by experts for specialists and trainees alike and succeeds in addressing the concerns of the forensic pathologist and the neuropathologist by tackling the overlapping problems that arise during the postmortem examination and subsequent legal proceedings. Totally updated and revised, the new edition contains many new chapters, including much more information on non-accidental injury in pediatric cases, alcohol, sudden death, imaging, sports and transportation injury, long term effects of chronic brain injury and anoxal death”– Provided by publisher. Identifiers: LCCN 2020045920 (print) | LCCN 2020045921 (ebook) | ISBN 9781498706162 (hardback) | ISBN 9781003158035 (ebook) Subjects: MESH: Forensic Pathology–methods | Craniocerebral Trauma–pathology | Central Nervous System–pathology Classification: LCC RA1147 (print) | LCC RA1147 (ebook) | NLM W 825 | DDC 614/.1–dc23 LC record available at https://lccn.loc.gov/2020045920 LC ebook record available at https://lccn.loc.gov/2020045921 ISBN: 978-1-498-70616-2 (hbk) ISBN: 978-0-367-74460-1 (pbk) ISBN: 978-1-003-15803-5 (ebk) Typeset in Utopia Std by KnowledgeWorks Global Ltd.

Contents Preface Contributors

vii viii

11 Traumatic axonal injury Daniel du Plessis 12 Brain swelling, raised intracranial pressure and hypoxia-related brain injury Daniel du Plessis

116

1 Anatomy of the head and neck Peter Dangerfield

1

2 Clinical aspects of head injury Graham Flint

17

13 Sudden unexpected death in epilepsy Christopher Milroy and Daniel du Plessis

3 Imaging of head trauma Calvin Soh

25

14 Contact sport and blast-related neuropathology 145 Daniel du Plessis and Christopher Milroy

4 Biomechanics of primary traumatic head injury Michael Jones

45

15 Head injury in the child Helen Whitwell and Christopher Milroy

151

5 Techniques Helen Whitwell

55

16 Spinal injuries Graham Flint and Helen Whitwell

167

6 Scalp, facial and gunshot injuries Christopher Milroy

66

178

7 Adult skull fractures Helen Whitwell and Philip Lumb

80

17 Difficult areas in forensic neuropathology: Homicide, suicide or accident Christopher Milroy and Helen Whitwell

8 Intracranial haematomas – Extradural and subdural Helen Whitwell

87

18 Non-traumatic neurological conditions in m ­ edico-legal work Colin Smith

186

9 Subarachnoid haemorrhage and cerebrovascular traumatic pathology Daniel du Plessis and Paul Johnson 10 Contusional brain injury and intracerebral ­ haemorrhage – Traumatic and non-traumatic Helen Whitwell

94

128 139

19 Alcohol, drugs, toxins and post-mortem toxicology Colin Smith and Christopher Milroy

195

20 The role of the expert witness Paul Watson and Christopher Milroy

203

Index

208

107

v

Preface This second edition builds on the previous one with the addition of two further co-editors, Christopher Milroy, forensic pathologist from Ottawa who contributed much to the first edition and Daniel du Plessis, neuropathologist from Manchester. As before, this book is not intended to be a fully comprehensive forensic pathology or neuropathology text as there are many available. The aim is to concentrate on those difficult and emerging areas of forensic neuropathology. These often impact in the medico-legal setting. In updating, we have added a number of new chapters, including biomechanics, neuroradiology and the important topics of chronic traumatic encephalopathy and sudden unexpected death in epilepsy where knowledge has considerably advanced over recent years.

Other chapters especially those on hypoxia, clinical aspects of head injury, spinal injury, subarachnoid haemorrhage as well as infant head injury have been considerably revised. The latter now includes a section devoted to ocular pathology. New photographs and references are further additional features. We hope, as with the previous edition, that readers, including forensic pathologists, neuropathologists and general pathologists, will find this latest text informative and practically useful. Neuroscience clinicians and others will also find areas of interest. Members of the legal profession found the last edition of assistance – including for cross-examination purposes – as some of us have experienced!

vii

Contributors Peter Dangerfield Clinical Anatomist University of Liverpool Liverpool, UK Graham Flint Neurosurgeon Queen Elizabeth Hospital Birmingham, UK Paul Johnson Forensic Pathologist Royal Liverpool University Hospital Liverpool, UK Michael Jones Clinical, Trauma and Orthopaedic Engineer Cardiff University Wales, UK

viii

Philip Lumb Forensic Pathologist Manchester, UK Colin Smith Neuropathologist University of Edinburgh Scotland, UK Calvin Soh Neuroradiologist Manchester Royal Infirmary Manchester, UK HH Judge Paul Watson QC Circuit Judge Middlesbrough, UK

1

Anatomy of the head and neck Peter Dangerfield

The surface anatomy and features of the head and neck are derived from skeletal and soft tissue structures covered by skin and connective tissue that covers the underlying bony skull. In some areas, particularly the face, the skin is thin, allowing easy palpation of underlying skeletal features. It is also highly mobile as a result of the presence of a number of small but extensive subcutaneous muscles of facial expression, all supplied by a single cranial nerve (facial). The scalp is, in contrast, relatively tough and hidden from view in most individuals by hair. Within the skull, which acts as a protective shield, are the brain and brainstem, with its associated covering of tissue layers and blood vessels. The anatomy is complex and this chapter can only provide an overview of the principal features, with craniofacial anatomy included to link anatomically with the underlying structures, including the neuroanatomy, which may be of particular significance in penetrating or blunt head trauma.

The external face The anatomical surface features of the face are never totally symmetrical. As facial expressions have evolved as a communication method, the underlying anatomy has evolved from the functional need. The skin is often marked by moles and freckles and may also present with scars consequential to cuts and other trauma. The shape and size of the hairline vary with race and the eyebrows can also be highly variable. The nose comprises underlying cartilaginous and bony structures. In the coronal section, the nose is triangular in shape. The external nares (nostrils) are protected by coarse hairs (vibrissae) and serve to filter air entering the nose. The anterior part of the nose is composed of flexible underlying fibrocartilage. Inferior to the nose is the mouth, surrounded by the lips. Here, the size and shape of the mouth are very variable, both within races and between different racial groups. The lips have non-keratinised epithelium and thus appear pink as a result of the underlying blood vessels. The eyes are set within the bony orbits of the skull but protected by the rim of bone. Their individual position relative to the nose is variable and can be close-set or wide set in normal individuals. Attached within the orbits are the muscles that control eye movement while, superficially, the paired eyelids cover and protect the eye from potential damage. The lids normally permit only a portion of white sclera to appear laterally, with the transparent conjunctiva and cornea that cover the pigmented iris seen medially. The upper lid normally overlaps the iris, but the sclera may be seen between the iris and the lower lid. The shape of the lids can vary between individuals; in particular, the elevator of the upper lid can be weak or damaged, leading to a drooping appearance. Ethnic differences are often prominent. Mongoloid epicanthic folds and other minor folds of skin in the medial aspect of the orbit should be noted. Conditions such as exophthalmos

associated with hyperthyroidism can result in prominent eyes. Facial fractures affecting the maxilla and inferior margin of the orbit can lead to a sinking of the eyeball. The shape of the lids themselves can lead to a wide range of different appearances of the eye within the orbit.

The internal facial structures Internally, the muscles of facial expression and their nerve and vascular supplies contribute to the facial structures. In addition, the parotid gland is located within the lateral parts of the cheeks. The facial expression muscles are supplied by the facial (cranial VII) nerve. Their function is to control and support the structures and openings in the face, such as the eyes and mouth. In humans, their functionality serves an important role in non-verbal communication as well as aiding actions such as screwing up the eyes and chewing. The mouth is surrounded by the sphincteric orbicularis oris muscle into which merge the fibres of the buccinator, the muscle of the cheek. The buccinator contracts during chewing and serves to prevent trapping of food within the space between the gums and teeth; it also acts to raise the pressure of air expelled by musicians playing wind instruments or by whistling. The orbicularis oculi surrounds the eye and serves to function in two ways. First, fibres that surround the eye serve to screw the eye up because they are attached to the bone on the medial aspect of the orbit. Second, the palpebral fibres attach to the lateral palpebral raphe and serve to close the eye when blinking. Additional fibres are attached to the lacrimal sac and serve to dilate the sac and keep the puncta in contact with the eyeball. The facial nerve enters the face by passing through the tough fibrous capsule of the parotid gland and can be damaged during surgical procedures to that gland.

The nose The nose, as the upper part of the respiratory tract, is located superior to the hard palate and contains the organ of smell. It is divided into right and left nasal cavities by the nasal septum, with each nasal cavity having an olfactory and a respiratory area. The external nose varies considerably in size and shape in individuals and races because of differences in the nasal cartilage structure. The inferior aspect is composed of two openings called the nares (nostrils), each separated from the other by the nasal septum. The nasal bones, the frontal processes of the maxillae, the nasal part of the frontal bone and the bony part of the nasal septum form the skeletal components of the nose, whereas five main cartilages form the cartilaginous nose. These are two lateral cartilages, two alar cartilages and a septal cartilage that articulates with the bony septum. The nasal cavities open through the choanae into the nasopharynx at the posterior. The nasal mucosa is bound closely to the periosteum and perichondrium of the nasal bones and

2

Anatomy of the head and neck

Olfactory nerve fibres

Sphenoid air sinus

Middle concha

Clivus

Inferior concha

Auditory tube

Hard palate

Spinal cord

Premaxilla

Dens Uvula

Tongue Lower lip

Epiglottis Mandible Oesophagus

Figure 1.1 The nasal cavity and associated structures. cartilages, and lines the nasal cavities, except for the vestibule which is lined with skin. The olfactory area lies superior within the cavity and is the organ of smell, with its nerve fibres passing through the cribriform plate to enter the olfactory bulbs, which lie against the inferior surface of the frontal lobe of the brain. The narrow, curved roof of the nasal cavity is divided into frontonasal, ethmoidal and sphenoidal parts, named by adjacent bones. The wide floor is formed by the horizontal plate of the palatine bone and the palatine process of the maxilla. Medially, the wall is the nasal septum, comprising the vomer, perpendicular plate of the ethmoid, septal cartilage and the nasal crests of the maxillary and palatine bones. The lateral walls of the nasal cavity are made up of three nasal conchae or scroll bones, each forming a roof over a meatus connecting the nasal cavity to a sinus or the orbit. The superior meatus is between the superior and middle conchae, into which orifices from the posterior ethmoidal sinuses open. The middle meatus, inferior to the middle conchae, communicates with the frontal sinus via the frontonasal duct and the maxillary sinus at its posterior end. The inferior meatus is inferolateral to the inferior conchae and receives the nasolacrimal duct from the lacrimal sac into its anterior portion. The nose receives arterial blood from many branches, including the sphenopalatine artery, ethmoidal arteries and the facial artery. Kiesselbach’s area, found on the anterior nasal septum, is rich in capillaries and is the site of profuse nose bleeding. The nerve supply of the nasal mucosa is by the maxillary nerve, nasal branches of the greater palatine nerve and the anterior ethmoidal nerves, and branches of the nasociliary nerve. The paranasal sinuses are air-filled extensions of the nasal cavity within the frontal, maxillary, sphenoid and ethmoid bones and are named according to each bone. The ethmoidal sinuses consist of ethmoidal cells located within the ethmoid bone between the orbit and nose. The sphenoid air sinuses are unevenly divided like the frontal air sinuses and separated by a bony septum. They occupy the body of the sphenoid bone and

are separated by thin bone from the optic chiasma, the pituitary gland, the internal carotid arteries and the cavernous sinuses. The maxillary sinuses are large pyramidal cavities within the maxillae. Their floor is formed by the alveolar part of the maxilla, with the roots of the maxillary teeth, particularly the first two molars, creating conical elevations (Figure 1.1).

The oral cavity The oral cavity consists of the oral vestibule and the oral cavity. The vestibule is the space between the lips and cheeks and the teeth and gums, and communicates with the exterior through the orifice of mouth, the size of which is controlled by muscles, including the orbicularis oris. The oral cavity lies posterior and medial to the upper and lower dental arches and is limited posteriorly by the terminal groove of the tongue and palatoglossal arches and anteriorly and laterally by the maxillary and mandibular arches containing the teeth. The roof is formed by the hard and soft palate, which also forms the floor of the nasal cavities. Posteriorly, the oral cavity communicates with the oropharynx. If the mouth is closed, the tongue fills the space of the oral cavity. The hard palate forms the anterior component of the roof of the oral cavity, with its cavity filled by the resting tongue when it is at rest and formed by the palatine processes of the maxillae and the horizontal plates of the palatine bones. The incisive fossa and the greater and lesser palatine foramina open on the oral aspect of the hard palate. The soft palate is the muscular posterior part, attached to the posterior border of the hard palate and extending as a posteroinferiorly curved free margin that terminates in the uvula. It is strengthened by a palatine aponeurosis formed by the expanded tendon of the tensor veli palatini and is attached to the posterior margin of the hard palate. Laterally, it is continuous with the wall of the pharynx and joined to the pharynx and tongue by the palatopharyngeal and palatoglossal arches.

The scalp The masses of lymphoid tissue forming the palatine tonsil lie within the tonsillar fossa, bounded by the palatoglossal and palatopharyngeal arches and the tongue.

The orbit The orbit is a pyramidal, bony cavity in the face. It contains and protects the eye with its associated muscles, nerves and vessels and the lacrimal apparatus. The roof is formed by the orbital part of the frontal bone, separating the orbit from the anterior cranial fossa and containing a small fossa for the lacrimal gland. The lesser wing of the sphenoid contributes to the roof at its apex. The medial wall is formed by the thin bone of the ethmoid, frontal, lacrimal and sphenoid bones. It is indented by the fossa of the lacrimal sac and nasolacrimal duct. The lateral wall comprises the frontal process of the zygomatic bone and the greater wing of the sphenoid, and is vulnerable to direct trauma. It serves to separate the orbit from the temporal and middle cranial fossae. The floor is made up from the maxilla, zygomatic and palatine bones, with a thin inferior wall partly separated from the lateral wall by the inferior orbital fissure. At the apex of the orbit lies the optic canal, located medial to the superior orbital fissure, which carries the optic nerve and associated structures into the orbit.

the chest anteriorly, and to a rather ill-defined border defined as a line between the acromia of the shoulder posteriorly. Laterally, the neck is clearly marked by the sternocleidomastoid muscle, which is attached between the mastoid process and adjacent nuchal line on the occipital bone and the sternum. The posterior neck region presents a muscular appearance as a result of the paraspinal cervical muscles, which lie deep to the trapezius muscle. Two folds are apparent within the hairline of the head formed by the nuchal ligaments. The skin of the neck also has tension or Langer’s lines, orientated in a horizontal slightly downward direction. These represent the orientation of collagen under the skin and are important for scar formation after skin penetration and may affect the appearance of a stab wound.

Posterior structures of the neck

The arterial supply to the lower face is via the facial artery, a branch of the external carotid artery. This artery enters the face by looping across the mandible, almost to the midpoint of the ramus where it can be located by finding the small notch on the margin of the mandibular ramus in which it lies. It passes upwards and medially towards the margin of the mouth where it divides to give rise to superior and inferior labial arteries, supplying the lips. A further branch extends upwards towards the medial aspect of the eye and orbit alongside the nose. The upper part of the face and scalp is supplied by the terminal branches of the external carotid artery. The deep facial structures are supplied by the maxillary artery, which passes deep to the mandible. The superficial temporal artery passes upwards to supply the temporal region. The transverse facial artery is a branch of this artery that runs medially across the face, supplying the cheek structures. Small supraorbital and supratrochlear arteries, branches of the ophthalmic branch of the internal carotid artery, supply the forehead and anterior scalp. Posterior to the external ear can be found the occipital and posterior auricular arteries, which supply the posterior of the scalp and ear region. The extensive arterial supply to the face is highly anastomotic, and lacerations can and do bleed extensively from what might at first be assumed to be small vessels. Venous return follows the same basic pattern as the arterial supply, except that the veins drain into either the internal or the external jugular veins. Extensive lymph node distributions receive the lymph vessels, which follow the pattern of drainage of the views. Nodes can be located in the submental, mastoid, submandibular, parotid and occipital regions, and these in turn give rise to lymph vessels draining mainly into the deep cervical nodes.

A layer of deep cervical fascia encloses the whole neck, attached superiorly to the superior nuchal lines, the mastoid processes and the lower border of the mandible, and inferiorly to the acromion process of the scapula, the clavicle and the upper border of the manubrium of the sternum. In the midline posteriorly, the investing layer attaches to the ligamentum nuchae. The investing facial layer splits to enclose the trapezius and sternocleidomastoid muscles. Laterally, anatomical schemes create the anatomical posterior triangle, the posterior boundary of which is formed from part of the trapezius muscle, a very large, flat muscle located in the back of the neck, extending into the thorax, and having a long, linear attachment to structures in the midline. Its upper fibres are attached to the medial third of the superior nuchal line of the occipital bone, the external occipital protuberance of the occipital bone and the tips of all the spinous processes of the cervical vertebrae, except superiorly where it attaches to the fibrous ligamentum nuchae. The trapezius muscle receives its motor nerve supply from the spinal root of the accessory nerve. The inferior boundary of this triangle is formed by the middle third of the clavicle and its roof is formed by the deep cervical fascia of the neck. Within the floor is the prevertebral fascia which in turn covers the semispinalis capitis, splenius capitis, levator scapulae, scalenus posterior, scalenus medius and scalenus anterior muscles. Semispinalis capitis and splenius capitis are part of a group of deep muscles of the back, functioning when movement of the vertebral column and the head on the cervical spine is initiated. It is important to appreciate that they are parts of the upper end of a long column of muscle that extends up the back of the abdomen, the thorax and the neck. This column occupies the vertical hollow, or groove, on either side of the spine, collectively referred to as the erector spinae muscle, a highly complex grouping in both its structure and function. The levator scapulae muscles act to elevate the scapula. They, together with the three scalene muscles, are attached at their superior ends to the cervical vertebrae. The scalene muscles attach to the first and/or second rib. The scalenus anterior is a key to understanding the anatomy of the root of the neck as a result of its important relationships to other structures. These include the phrenic nerve, the subclavian artery and vein, the brachial plexus and the cervical part (i.e. dome) of the pleura.

The neck

The scalp

The neck extends from the inferior boarder of the mandible to the superior borders of the clavicles and sternal manubrium in

The scalp covers the cranium and comprises a thick hairy skin layer with a tough fibroadipose hypodermis and a deeper thick

Blood supply to the face

3

4

Anatomy of the head and neck fibrous aponeurotic layer containing the neurovascular structures. The fibrous nature of this tissue prevents vessel constriction in lacerations, which can lead to extensive bleeding. A thin fascial areolar connective tissue layer deep to the aponeurosis covers the skull periosteum. This layer is only loosely held together and can allow the scalp to be stripped off from the bone of the skull with relative ease. Infection may also lead to scalp swelling.

The skull bones The skull itself has two distinct parts. The vault is composed of relatively thin plates of bone that articulate with one another at a series of jagged-edged suture lines. In the neonates, these are relatively wide, with marked fontanelles delineating the junction of adjacent bones. The anterior of the skull is made up of the frontal bone, which has left and right parts in the child, although these normally fuse by the end of growth. The frontal bones contribute to the roof of the orbit. Irregular frontal air sinuses are located in the medial aspect of the frontal bones. Laterally, the vault is composed of the parietal and temporal bones whereas the posterior part is made up from the occipital bone. The occipital bone extends inferiorly into the base of the skull and is pierced by the large foramen magnum, the entry point of the spinal cord into the internal skull. The bones of the skull articulate with each other by fibrous joints called sutures, namely the coronal, sagittal and lambdoid sutures. Large surface palpable mastoid processes are located posterior to the pinna of the ear and are a prominent part of the occipital bone. The base of the skull is a complex structure. The view typically seen at post-mortem examination is illustrated in Figure 1.2a. The cranial vault is illustrated in Figure 1.2b.

The brain and nervous tissues Our understanding of the functional anatomy of the brain and nervous system is advancing rapidly today with the use of magnetic resonance imaging (MRI). Such images serve to highlight the anatomy of the head in some detail, while functional images are beginning to allow a fully understanding of brain activity of its various components. Anatomically, the brain is protected from mechanical forces by the overlying cranial bones, the membrane layers called the meninges and the circulating cerebrospinal fluid (CSF). The Squamous part of occipital bone

Squamous part of temporal bone

Petrous part of temporal bone

nervous tissue is itself insulated from the general circulation by the so-called blood-brain barrier which serves to protect it from the majority of infectious agents and chemicals, including drugs. However, these can and do breach this barrier from time to time.

The meninges The brain and spinal cord share the same layers, which are continuous with one another. Within the skull are found the dura mater, arachnoid and pia mater layers. The dura mater has two distinct layers, an outer layer that is fused to the periosteum lining the inner surface of the skull and an inner fibrous layer. Consequently, no epidural space exists superficial to this layer, unlike the situation found within the spinal canal of the vertebral column. However, there is a potential space, termed the ‘extradural space’, present that can serve as a reservoir for blood if the meningeal vessels become ruptured by trauma. The tightly adherent skull-dura layers serve to prevent spread of blood. Both layers of dura are separated by a thin gap layer, in which are found the major blood sinuses and other blood vessels. Arachnoid granulations project through the dura into the venous sinuses and serve to absorb CSF back into the venous system. All the venous sinuses drain eventually into the internal jugular veins of the neck. The arachnoid layer is subdivided into a further three layers. The arachnoid membrane covers the brain and serves to smooth out the underlying gyri and sulci. This is underlaid by an epithelial layer, and the cells and fibres of the arachnoid trabeculae that cross the subarachnoid space, giving it the appearance of a spider web. These trabeculae join the arachnoid to the underlying pia mater, the deepest brain layer. All nerves and vessels passing into or out from the brain must traverse the subarachnoid space, which is filled with CSF. The pia mater layer is firmly adherent to the surface of the brain and extends into all the folds of the brain surface; in addition, it is adherent to cerebral blood vessels as they enter the substance of the brain. The pia mater is anchored by astrocyte processes. Tough, fibrous extensions of the dura mater, the dural folds, hold the brain firmly in position and these, with the CSF in the subarachnoid space, serve to protect the brain from sudden shocks and other deceleration movements that often accompany cerebral trauma. Dural venous sinuses are located between layers of these folds. In trauma, traction on the small

Parietal bone

Frontal bone

Cribriform plate of ethmoid

Internal occipital crest (a)

Foramen magnum

Orbital plate of frontal bone Sagittal suture   (b) Figure 1.2 (a) The base of the skull and (b) the cranial vault.

Coronal structure

Arterial blood supply

Falx cerebri

Tentorium cerebelli Transverse sinus

Frontal sinus

Falx cerebelli

Figure 1.3 The tentorium cerebelli. tributary superior cerebral veins as they enter the sagittal sinus may lead to their rupture, resulting in bleeding into the subdural space. The dural folds also form a large fibrous sheet that extends deep within the brain and acts to support parts of the cortex and brainstem by providing additional stabilisation. These are the falx cerebri, tentorium cerebelli and falx cerebelli. The falx cerebri is located within the longitudinal fissure separating the left and right cerebral hemispheres. Inferiorly, it is attached to the crista galli and the internal occipital crest of the inner surface of the occipital bone. Both the superior and inferior sagittal venous sinuses lie within the falx cerebri. Its posterior margin is continuous with the tentorium cerebelli, which separates the cerebellar hemispheres from the cerebrum, but is at right angles to the falx cerebri. The transverse sinus is located within the tentorium cerebelli. Finally, the falx cerebelli separates the two cerebellar hemispheres in the midline but inferior to the tentorium cerebelli (Figure 1.3).

Arterial blood supply Arterial blood reaches the internal cranial cavity via two major vascular supplies: the internal carotid artery and the vertebral artery. The common carotid arteries are both branches of the brachiocephalic artery on the right and aortic arch on the left. They are enclosed within the tough protective fibrous carotid sheath that also encloses the internal jugular vein and vagus nerve to the posterior of the artery, and which lies lateral to the trachea and oesophagus. Both right and left common carotid arteries terminate at the level of the upper border of the thyroid cartilage, bifurcating to give rise to internal and external carotid arteries. A swelling at the division of the artery is the location of the carotid sinus and carotid body containing chemo- and baroreceptors. The internal carotid artery has no branches within the neck and ascends deep to the styloid process of the skull and adjacent muscles, to enter the carotid canal of the petrous temporal bone. This canal passes forwards and medially to enter the internal cavity of the skull as the foramen lacerum. The internal carotid artery is now located to the lateral side of the sphenoid bone, enclosed within the cavernous sinus. Its path now

proceeds anteriorly within the sinus and then it passes out of the sinus upwards to lie adjacent to the anterior clinoid process, where it enters the subarachnoid space by piercing the dura mater. This path is clearly seen on suitable radiographs and is known as the carotid siphon. At this point, the artery gives a branch called the ophthalmic artery, which passes through into the orbit, supplying the retina via the central artery of the retina and branches to the lacrimal gland and adjacent orbital structures. The right and left internal carotid arteries are interconnected by an anterior communicating artery and also posteriorly by the posterior communicating artery to the basilar artery system, forming an important component of the circle of Willis on the inferior surface of the brain. The carotid artery terminates within the subarachnoid space of the brain as the middle cerebral and anterior cerebral arteries. These supply the lateral aspect of the brain and the upper parts and medial aspect of the anterior of the brain. The anterior cerebral artery passes above the optic chiasma and lies on the medial aspect of the cerebral hemisphere, where it forms an arch around the genu of the corpus callosum. It gives off a significant branch adjacent to the anterior communicating branch of the circle of Willis, which supplies the internal capsule. The middle cerebral artery is the major terminal branch of the internal carotid artery, giving off deep branches to the internal structure of the brain, supplying the corpus striatum, internal capsule and thalamus. Occlusion of the lateral striate branch of this artery gives rise to classic stokes. The middle cerebral artery continues along the lateral fissure reaching the insula, where it divides into divisions that supply the frontal and parietal lobes, the temporal lobe and the mid-part of the optic radiation. The vertebral artery is a branch of the first part of the subclavian artery, itself a branch of the brachiocephalic artery on the right and of the arch of the aorta on the left. This artery passes superiorly and posteriorly to enter the foramen transversarium of the sixth cervical vertebra, lying at the apex of the triangle formed by longus cervicis medially and scalenus anterior laterally. The artery passes upwards within a canal in the transverse processes, giving rise to small twigs of vessels that supply the spinal cord in the neck.

5

6

Anatomy of the head and neck

Anterior cerebral artery Anterior communicating artery Middle cerebral artery Internal carotid artery Posterior communicating artery Posterior cerebral artery Superior cerebellar artery

Basilar artery

Anterioinferior cerebellar artery

Anterior spinal artery Posterioinferior cerebellar artery Vertebral artery

Figure 1.4 Circle of Willis. The artery then passes across the superior aspect of the posterior arch of the atlas vertebra and enters a groove lateral to posterior atlantooccipital membrane to gain access to the vertebral canal. It pierces the dura mater and continues upwards through the foramen magnum, joining its fellow from the opposite side and creating the basilar artery. This is located in the anterior aspect of the brainstem. The basilar artery has a number of branches: the anterior spinal artery runs inferiorly in the anteromedian groove of the spinal cord; the posteroinferior cerebellar artery runs in the lateral aspect of the medulla and the cerebellum; and two

posterior spinal arteries lie medial to the dorsal nerve roots and can arise variably from either the vertebral or the posterior cerebellar vessels. Lying on the ventral surface of the brainstem and the pons, the basilar artery supplies the brainstem and cerebellum through anteroinferior and superior cerebellar arteries. At the superior margin of the pons, the vessel divides into the posterior cerebral arteries, which supply the temporal and occipital lobes of the cortex. The posterior communicating artery (see above) forms the posterior part of the circle of Willis and links the basilar to the internal carotid artery system (Figures 1.4 and 1.5).

Anterior cerebral artery

Posterior cerebral artery Middle cerebral artery

Anterior cerebral artery

Middle cerebral artery

Posterior cerebral artery

Figure 1.5 Arterial supply to the brain.

The cerebrum

Venous drainage The venous drainage of the brain is important because thromboses can give rise to many clinical syndromes. The cerebral hemispheres are drained by the superficial and deep cerebral veins. They lack valves. The superficial veins lie within the subarachnoid space and empty into the venous sinuses. The upper part of the cortex drains into the superior sagittal sinus whereas the mid-part normally drains into the cavernous sinus via the superficial middle cerebral vein. The lower parts drain into the transverse sinus. The deep veins drain the choroid plexus, thalamus and corpus striatum. The various component branches drain into the great cerebral vein of Galen, which is located within the midpoint of the tentorium cerebelli, where it unites with the inferior sagittal sinus to form the straight sinus that in turn is a branch of the left transverse sinus. The dural folds contain the major venous sinuses. The falx cerebri encloses the superior sagittal sinus within its upper part adjacent to the skull, whereas its lower free edge, lying within the longitudinal fissure of the brain, contains the inferior sagittal sinus. The straight sinus lies within the attachment of the falx to the tentorium and unites with the superior sagittal sinus at the confluence of the sinuses.

The cerebrum The cerebrum is the largest region of the brain. Paired cerebral hemispheres form the superior and lateral surfaces of the cerebrum and have cortical surfaces characterised by the presence of elevated ridges and grooves, serving to increase the surface

area of the brain. These are the gyri, shallow grooves called sulci and deeper fissures. This folding is less in the neonate and develops into its characteristic pattern during childhood. Although the entire brain enlarged during human evolution, the cerebral hemispheres enlarged at a much faster rate than the other parts of the brain. Each cerebral hemisphere is subdivided into regions, or lobes, named after the overlying cranial bones. On each hemisphere, the deep groove of the central sulcus divides the anterior frontal lobe from the posterior parietal lobe. The lateral surface of the frontal lobe comprises the precentral gyrus with the precentral sulcus in front. The inferior surface is marked by orbital gyri and is in direct contact with the forward extending olfactory tract and bulb. Internally, the cerebral hemispheres contain the lateral ventricles (Figure 1.6) on the left and right and a central third ventricle. These are the site of CSF production. On the lateral surface, a horizontally aligned lateral sulcus separates the frontal lobe from the temporal lobe. By retracting the lips (known as the opercula) of the lateral sulcus, the insula can be exposed. This effectively lies deep to the temporal lobe and is invisible from the surface. A parieto-occipital sulcus posteriorly separates the parietal lobe from the occipital lobe. Functionally, each lobe has less clearly defined regions dedicated to a range of motor and sensory roles, with very indistinct boundaries and considerable overlap. The two cerebral hemispheres are almost completely separated by a deep longitudinal fissure containing the falx cerebri, remaining connected by the thick band of white matter called the corpus callosum, which forms a distinctive white C-shaped shape on sagittal MRI of the brain. By cutting the corpus

Forceps minor

Frontal horn of lateral ventricle Middle cerebral artery branch Corona radiata

Posterior horn of lateral ventricle

Figure 1.6 Lateral ventricle.

7

8

Anatomy of the head and neck

Corpus callosum Lateral ventricle Thalamus

Pineal gland Superior colliculus

Hypothalamus

Cerebral aqueduct

Pituitary gland

Fourth ventricle Pons Cerebellum

Medulla

Figure 1.7 Sagittal section through the head. callosum, the brain can be divided into two hemispheres, exposing the medial surface (Figure 1.7). The corpus callosum is a massive band of white matter comprising a trunk, an anterior genu, a posterior splenium and a narrow rostrum which extends from the genu to the anterior commissure. Although the surface of the cerebrum is composed of grey matter, its interior consists primarily of nerve axon pathways or white matter. Association fibres interconnect areas of the cortex within each cerebral hemisphere (Figure 1.8). Shorter association or arcuate fibres pass from one gyrus to another,

whereas longer association fibres are arranged in bundles, or fasciculi. Longitudinal fasciculi link the lobes of each cerebral hemisphere. Commissural fibres link the two hemispheres and allow communication between them. These fibres are formed into densely packed bands of nerve axons, such as the corpus callosum and the anterior commissure (Figure 1.8a-g). The cerebral cortex is also linked to other parts of the brain and brainstem by projection fibres. These fibres traverse the diencephalon, where axons passing up to sensory areas of the cerebral cortex mix with axons descending from motor areas

Longitudinal fissure

Superior frontal gyrus Middle frontal gyrus

Cingulate gyrus

Lateral orbital gyrus

(a)

Middle orbital gyrus

Gyrus rectus

Medial orbital gyrus

The cerebrum

Longitudinal fissure

Superior frontal gyrus Middle frontal gyrus

Cingulate gyrus Genu of the corpus callosum Anterior horn of lateral ventricle Inferior frontal gyrus

(b)

Gyrus rectus Cingulate gyrus

Cingulum

Corpus callosum Septum pellucidum Lateral ventricle External capsule Insula

Caudate nucleus Internal capsule Putamen Claustrum Column of fornix

Temporal lobe

(c)

Corpus callosum Caudate nucleus Internal capsule Putamen Claustrum Third ventricle Mamillary body

(d)

9

10

Anatomy of the head and neck

Longitudinal fissure

Lateral ventricle

Cingulate gyrus

Caudate nucleus

Body of the corpus callosum

Body of fornix

Thalamus

Third ventricle Red nucleus

(e)

Substantia nigra

Cingulate gyrus Body of the corpus callosum Inferior frontal gyrus Retrolenticular limb of the internal capsule Hippocampus Superior colliculus

Lateral ventricle Fornix Pulvinar Inferior horn of the lateral ventricle Temporal lobe (f)

Longitudinal fissure Precentral gyrus Postcentral gyrus Inferior parietal gyrus

Inferior horn of the lateral ventricle

Middle temporal gyrus Inferior temporal gyrus

Optic radiation Tapetum Lingual gyrus

(g)

Figure 1.8 (a)–(g) Coronal sections through the cerebral hemispheres.

The diencephalon of the cortex within an area called the internal capsule. These ascending and descending nerve fibres look alike. The central sulcus separates the motor and sensory areas of the cortex, with the precentral gyrus of the frontal lobe forming the anterior border of the sulcus. This region is the primary motor cortex. Specialist pyramidal cells within the primary motor cortex are responsible for voluntary movements by controlling somatic motor neurons in the brainstem and spinal cord. The primary sensory cortex is located in the postcentral gyrus of the parietal lobe, which forms the posterior border of the central sulcus. Neurons in this region receive somatic sensory information from sensory receptors responsible for pain, temperature, touch, pressure, vibration or taste. The thalamus relays this sensory information to the primary sensory cortex. Other regions of the cortex have developed special sensory roles. The occipital lobe contains the visual cortex whereas the temporal lobe is responsible for olfactory and auditory sensations. A specialist area of the frontal lobe and the adjacent anterior part of the insula is the gustatory cortex. This processes information from taste receptors of the tongue and pharynx.

The brainstem The brainstem comprises the medulla oblongata, itself an upward continuation of the spinal cord, the pons and the midbrain; it is continuous with the diencephalon component of the forebrain. It is related to the basiocciput or clivus and is connected to and overlaid by the posterior and lateral aspects by the cerebellum. The stem itself can be viewed anatomically only if this overlying cerebellum is removed by cutting through the three pairs of peduncles, bundles of nerve fibres attached to either side of the stem. The dorsal surface of the brainstem is marked by a continuation of the median sulcus extending upwards from the spinal cord into the medulla. Internally, within its caudal two-thirds is a closed central canal. This canal moves to a more a posterior position in the more rostral medulla, eventually opening as the fourth ventricle, deep to the cerebellum. Its floor is the dorsal surface of the rostral medulla and, dorsolaterally, it is dominated by the inferior cerebellar peduncle. Nerve fibres pass through the inferior cerebellar peduncle from the medulla to the cerebellum. The fourth ventricle is rhomboid in shape and is at its widest in the region of the junction between the medulla and pons. Nuclei of the vagus and other cranial nerves are located below the fourth ventricle, some of which may be identified using appropriate staining techniques. The postrema is an area where the blood-brain barrier is absent. It forms the most caudal aspect of the floor of the fourth ventricle. It is also the site of action of emetics. The fourth ventricle narrows superiorly and becomes the narrow cerebral aqueduct, which passes throughout the length of the midbrain. A small lateral aperture at the widest part of the fourth ventricle (also known as the foramen of Luschka) allows CSF to pass out into the subarachnoid space. The pons is the upward continuation of the brainstem from the medulla and is divisible into anterior and posterior parts. The anterior part contains the pontocerebellar fibres, which arise from the pontine nuclei. These pass through the middle cerebellar peduncle to the contralateral side of the cerebellum. In the rostral part of the pons, the lateral wall of the fourth

ventricle is composed of paired superior cerebellar peduncles, with a thin superior medullary velum connecting them and forming its roof. These peduncles converge towards the midline as they pass into the midbrain and contain cerebellar afferent and efferent fibres. The brainstem contains many ascending and descending nerve tracts, some of which terminate in brainstem nuclei located in this region. Functionally, important centres controlling respiration, cardiovascular function and levels of consciousness, forming the reticular system, are located within the brainstem. The cranial nerves III-XII are attached to the brainstem, with their fibres either originating from, or terminating in, the appropriate cranial nerve nuclei. On the dorsal surface of the midbrain can be identified four paired swellings called the superior and inferior colliculi, which form important parts of the visual and auditory systems.

The midbrain The midbrain is divided into anterior and posterior portions at the level of the cerebral aqueduct. The anterior portion is termed the ‘tegmentum’ and is bounded by the crus cerebri. The posterior portion is the tectum, made up of the inferior and superior colliculi (corpora quadrigemina). The inferior colliculus forms part of the ascending auditory pathway whereas the superior colliculus is part of the visual system. The cerebral aqueduct traverses the length of the midbrain ventral to the colliculi, with the trochlear and oculomotor cranial nerve nuclei located adjacent to it within the periaqueductal grey matter. At the level of the inferior colliculus, the superior cerebellar peduncles are related to the central portion of the tegmentum. Ventrally, the midbrain tegmentum is the site of the substantia nigra, consisting in part of pigmented, melanin-containing neurons that synthesise dopamine as their transmitter. It is degeneration of the substantia nigra that is associated with Parkinson’s disease and problems secondary to drug abuse. Anterior to the substantia nigra are the large crus cerebri, composed entirely of the descending cortical efferent fibres that have passed through the internal capsule after leaving the cerebral hemispheres. Sections through the brainstem are illustrated in Figure 1.9a–e.

The diencephalon Above the midbrain lies the forebrain (termed the ‘prosencephalon’). This comprises the paired cerebral hemispheres on each side of the centrally located tissue called the diencephalon. The diencephalon is the site of the epithalamus, thalamus, subthalamus and hypothalamus. As the cerebral hemispheres overlie the midline structures within the diencephalon, only the anterior part of the hypothalamus can be viewed on the base of the brain. Both the thalamus and hypothalamus are nuclear groups forming the lateral wall of the centrally located third ventricle. The left and right thalami are often linked together across the third ventricle by an interthalamic adhesion. The hypothalamus also contributes to the floor of the third ventricle. It is separated from the inferior aspect of the thalamus by a shallow hypothalamic sulcus and extends ventrally and medially to the subthalamus. The epithalamus is a relatively small structure comprising the pineal gland and the habenular nuclei; it is located in the caudal and dorsal region of the diencephalon, immediately above the superior colliculus of the midbrain.

11

12

Anatomy of the head and neck

Superior colliculus Central grey

Cerebral aqueduct Posterior longitudinal fasciculus

Reticular formation

Oculomotor nucleus

Medial lemniscus

Red nucleus Oculomotor nerve

Substantia nigra Parietopontine, occipitopontine and temporopontine fibres

Interpeduncular nucleus

Frontopontine fibres (a)

Fourth ventricle

Cut cerebellar peduncle

Medial lemniscus

Pontine nuclei

Corticospinal and corticonuclear fibres

(b)

Basilar groove

The diencephalon

Inferior cerebellar peduncle Reticular formation

Hypoglossal nucleus

Anterolateral system Medial lemniscus

Olivary nucleus and olivocerebellar fibres Pyramid

Basilar artery (c)

Gracile nucleus Cuneate nucleus Spinotrigeminal tract and nucleus

Hypoglossal nucleus Medial longitudinal fasiculus Reticular formation Postolivary sulcus

Olivary nucleus Medial accessory olivary nucleus

Medial lemniscus Pyramid

(d)

Gracile fasciculus and nucleus

Posterior median sulcus

Cuneate fasciculus Trigeminal nucleus

Posterior spinocerebellar tract

Pyramidal decussation Anterior spinocerebellar tract Pyramid

Vertebral artery (e)

Basilar artery

Figure 1.9 (a) Section through the midbrain at the level of the superior colliculus. (b) Transverse section at the level of the pons. Section through the medulla at the level of (c) the olive, (d) the sensory decussation and (e) the pyramids.

13

14

Anatomy of the head and neck

Dentate nucleus

SUPERIOR

Fourth ventricle Dentate nucleus Vermis Cerebellar cortex (a)

(b)

Cerebellar peduncle

INFERIOR   Figure 1.10 (a) Horizontal section and (b) cerebellum – to illustrate superior-inferior surface orientation.

Inferior to the thalamus and dorsolateral to the hypothalamus is the subthalamus, with its anterolateral aspect closely related to the internal capsule. The internal capsule separates the thalamus from the lentiform nucleus and is a common site of strokes. It contains fibres running from the thalamus to the cortex and from the cortex to the brainstem and spinal cord. These fibres form the corona radiata that can be seen between the internal capsule and the cortex. The internal capsule has a medially facing ‘V’ shape when examined in a horizontal section of the brain, with an anterior limb between the lentiform nucleus and caudate nucleus, a middle section known as the genu, and a posterior limb lying between the lentiform nucleus and the thalamus. Lateral to the thalamus, the capsule is known as the retrolentiform part. The lentiform nucleus is itself lens shaped and is formed from the putamen and globus pallidus, whereas the corpus striatum links the putamen and globus across the internal capsule. Inferiorly, the diencephalon is related to the midline optic chiasma, caudal to which is a small midline elevation called the tuber cinereum. From its apex extends the infundibulum or pituitary stalk, which attaches to the pituitary gland. Further caudal to the tuber cinereum, a pair of rounded eminences, the mamillary bodies, is located on either side of the midline. These contain the mamillary nuclei of the hypothalamus.

The cerebellum The cerebellum is part of the hindbrain that overlies the fourth ventricle on its dorsal aspect. It is connected to the brainstem by three stout pairs of fibre bundles called the inferior, middle and superior cerebellar peduncles. It comprises two laterally located hemispheres linked in the midline by the vermis, with its superior surface related to the deep surface of the tentorium cerebelli. Although the superior vermis forms a midline ridge, the inferior vermis lies in a deep groove between the hemispheres. The cerebellum has a highly folded and convoluted surface, with the folds oriented transversely. Fissures of varying depths lie between the folds. Some act as landmarks and subdivide the cerebellum into three lobes. The primary fissure is on the superior surface and separates the anterior lobe from the posterior lobe. On the inferior surface, a posterolateral fissure separates the flocculus and vermis, which together form the flocculonodular lobe. Internally, the cerebellum consists of an outer layer of grey matter, the cerebellar cortex, and an inner core of white matter, composed of afferent and efferent fibres that run to and from the cortex. Irregular projections also extend up towards

the cortex. Deep within the white matter are located four pairs of cerebellar nuclei, which have nerve connections with the cerebellar cortex and nuclei of the brainstem and thalamus. Section through the cerebellar hemispheres is illustrated in Figure 1.10a and b.

The vertebral column The vertebral column’s function is to support the trunk and to protect the underlying spinal cord. Its structure undergoes progressive growth and developmental changes in the postnatal period. These continue in adulthood and lead to decline in senescence. The bony morphology is influenced internally by genetic, hormonal and metabolic factors and externally by mechanical and environmental factors; this affects the column’s ability to react to the dynamic forces of everyday life such as compression, traction and shear. The forces vary in magnitude and are influenced by occupation, locomotion and posture. The column comprises 33 vertebral segments, each separated by a fibrocartilaginous intervertebral disc. Although the usual number of vertebrae is 7 cervical, 12 thoracic, 5 lumbar, 5 sacral and 4 coccygeal, this total is subject to frequent variations of between 32 and 35 bones. In adults, the cervical region is curved convexly forwards and is the least marked; this is a lordosis. The thoracic curve is kyphotic, i.e. concave forwards, and is a consequence of the increased posterior depth of the thoracic vertebral bodies. The lumbar curve is lordotic, i.e. convex forwards, having a greater curve in the female; the lordosis is caused by the greater anterior depth of the intervertebral discs and some anterior wedging of the vertebral bodies. The vertebral canal follows these vertebral curves. Free mobility is present in the cervical and lumbar regions, and the canal is large with a triangular shape. Where movement is reduced in the thoracic region, the canal is small and circular. These canal size differences are matched by variations in the diameter of the spinal cord and its enlargements. The lumbar spinal canal has a gradual decrease in measurement between L1 and L5 with a greater relative width in the female. Each vertebra has distinctive shapes to its body and spinous process as well as to the transverse processes and the intervertebral joints, although similarities between groups in the same region can aid identification of individual bones. In older people, it should be noted that, as bone undergoes age-related changes in its structure, this can lead to broadening and loss of height of the vertebral bodies. Such changes are more severe in women (Figure 1.11).

Internal structure of the spinal cord

Spinous process

Transverse process Subrachnoid space

Dorsal root of spinal nerve

White matter

Dura mater

Vertebral body

Figure 1.11 Spinal column.

The spinal cord The spinal cord serves to link the brainstem to the appropriate segmental nerves of the peripheral nervous system as they emerge between each vertebra. In the adult, it extends from the lower end of the medulla oblongata at the level of the foramen magnum to the level of the upper border of the second lumbar vertebra. Below this level, it tapers and forms the conus medullaris. A band of pia mater called the filum terminale continues downwards from the end of the cord to the coccyx. During early pregnancy, the cord in the foetus extends the entire length of the vertebral canal. As pregnancy progresses, the foetal vertebral column grows more rapidly in length than the spinal cord, with the result that it extends only to the level of the third lumbar vertebra at birth; it then slowly attains its adult level in childhood. The spinal cord lies fully within the vertebral canal of the column and is thus normally protected from external trauma. Damage can thus only occur either as a consequence of massive external forces that compress this bony cage or by penetrating wounds, where an instrument or other penetrating agent passes between the vertebrae in the narrow space occupied by the intervertebral disc. The three meningeal layers, as described surrounding the brain, also enclose the spinal cord. The dura mater layer extends down to the level of the second sacral vertebra and is lined on its inner surface by the arachnoid mater. The pia mater layer firmly adheres to the cord’s surface. Laterally, the cord is suspended within the dural sheath by serrated denticulate ligaments, which effectively create a shelf between the dorsal and ventral roots of the spinal nerves. The CSF is enclosed by the subarachnoid space, which also extends down to the level of the second sacral vertebra. External to the dura mater is the epidural space. This contains fat and the extensive vertebral venous plexus. The cord is grooved on its anterior surface by the median fissure and on its posterior surface by the shallower median sulcus. Laterally, the posterolateral sulcus is present, to which the dorsal roots of the spinal nerves are attached. The cord is divided into 31 spinal cord segments, from which arise the paired spinal nerves. Sensory fibres enter the cord via its dorsal (posterior) aspect, which also has a clearly defined dorsal root ganglion containing the cells of the nerve axons.

Motor fibres leave the cord on the anterolateral aspect via the ventral (anterior) root. The sensory and motor nerve roots fuse and merge into a single spinal nerve that leaves the vertebral canal via the intervertebral foramen. Leaving the foramen, the emerging nerves divide into anterior and posterior rami, each containing both motor and sensory fibres. The length of the spinal nerve within the vertebral canal increases progressively down the length of the cord, until after termination of the cord itself at the level of the second lumbar vertebra. Below this level, the nerves alone form a bundle known as the cauda equina within the vertebral canal. The blood supply of the spinal cord is maintained by the anterior and posterior spinal arteries. The anterior spinal artery is located within the anterior median fissure and is formed by the union of a branch vessel from each vertebral artery. The posterior spinal arteries are normally branches of the posteroinferior cerebellar arteries but can arise directly from the vertebral arteries. They are present on both sides of the posterior aspect of the cord. A series of radicular arteries also enters the vertebral canal through the intervertebral foramina. Surgery of aortic aneurysms may compromise these vessels. Between the vertebrae and the dura mater is the epidural space, which contains the arteries supplying the spinal cord as well as the vertebral venous plexuses. These vertebral venous plexuses are connected to a network of veins, termed Batson’s veins, which have no internal valves. These connect deep pelvic veins draining the bladder, prostate and rectum. Clinically, this is an important point since they may allow infections or metastases from malignant prostate tumours to reach the vertebrae as the vertebral venous plexus are connected to the veins draining these organs.

Internal structure of the spinal cord A transverse section of the cord exhibits an H-shaped area of grey matter containing sensory and motor nerve cells. Within the grey matter, a central canal communicates superiorly with the fourth ventricle of the brainstem. The posterior horn of the grey matter contains the termination of the sensory nerve fibres whereas the larger anterior horn contains motor cells that supply fibres within anterior roots.

15

16

Anatomy of the head and neck

Posterior spinocerebellar tract Lateral corticospinal tract Rubrospinal tract Anterolateral system Anterior corticospinal tract (a)

Gracile fasciculus Cuneate fasciculus Posterior grey horn Anterior spinocerebellar tract Anterior grey horn Anterior median fissure

Posterior median sulcus

Gracile fasciculus

Posterior spinocerebellar tract

Cuneate fasciculus

Lateral corticospinal tract

Posterior grey horn

Anterior spinocerebellar tract

Nucleus dorsalis

Anterior median fissure (b)

Anterior grey horn Anterior corticospinal tract

Gracile fasciculus Lateral corticospinal tract Rubrospinal tract Medial longitudinal fasciculus Anterolateral system (c)

Posterior grey horn

Anterior grey horn Anterior corticospinal tract Anterior median fissure

Figure 1.12 Section through the spinal cord at the level of (a) T6, (b) L3 and (c) C7.

In the thoracic and upper lumbar regions are lateral horns containing the cells of origin of the preganglionic sympathetic fibres. The grey matter of the cord is surrounded by white matter containing the ascending and descending nerve tracts (Figure 1.12a–c).

Park, J.S. 2018. Cross-Sectional Atlas of the Human Head: With 0.1-mm Pixel Size Color Images. Springer. Peris-Celda, M., Martinez-Soriano, F. and Rhoton, A.L. 2017. Rhoton’s Atlas of Head, Neck, and Brain, 1st edn. Thieme.

References and further reading

Bear, M.F., Connors, B.W. and Paradison, M.A. 2015. Neuroscience Exploring the Brain, 4th edn. Wolters Kluwer. Berkowitz, A. 2017. Clinical Neurology and Neuroanatomy: A LocalizationBased Approach. Lange McGraw Hill. Crossman, A.R. and Neary, D. 2019. Neuroanatomy: An Illustrated Colour Text, 6th edn. Elsevier. Haines, D.E. 2018. Neuroanatomy Atlas in Clinical Context: Structures, Sections, Systems, and Syndromes, 10th edn. Wolters Kluwer. Moore, K.L., Dalley, A.F. and Agur, M.R. 2017. Clinically Oriented Anatomy. 8th edn. Wolters Kluwer. Mtui, E., Gruener, G. and Dockery, P. 2020. Fitzgerald’s Clinical Neuroanatomy and Neuroscience E-Book. Elsevier. Spratt, J.D., Salkowski, L.R., Loukas, M., Turmezei, T., Weir, J. and Abrahams, P.H. 2020. Weir & Abrahams’ Imaging Atlas of Human Anatomy E-Book. Elsevier. Standring, S.S. (ed) 2020. Gray’s Anatomy, 42nd edn. Elsevier.

Atlases of anatomy and neuroanatomy Abrahams, P.H., Spratt, J., Loukas, M. and van Schoor, A.-N. 2019. McMinn and Abrahams’ Clinical Atlas of Human Anatomy, 8th edn. Elsevier. Anderson, M.W. and Fox, M.G. 2016. Sectional Anatomy by MRI and CT, 4th edn. Elsevier. Dean, D. and Herbener, T.E. 2000. Cross-Sectional Human Anatomy. Lippincott Williams & Wilkins. Dixon, A.K., Bowden, D.J., Ellis, H. and Logan, B.M. 2015. Human Sectional Anatomy: Atlas of Body Sections, CT and MRI Images, 4th edn. CRC Press. Logan, B.M., Reynolds, P., Rice, S. and Hutchings, R. 2016. McMinn’s Colour Atlas of Head and Neck Anatomy, 5th edn. Elsevier. Osborn, A.G., Hedlund, G.L. and Salzman, K.L. 2017. Osborn’s Brain. Elsevier.

Textbooks of anatomy and neuroanatomy

2

Clinical aspects of head injury Graham Flint

The forensic pathologist will often be asked to investigate a head injury, including as part of polytrauma. The purpose of this chapter is to provide a clinical context for such post-­ mortem examinations. Many cases may become the subject of court proceedings and the pathologist may be the only medical expert in the courtroom. As such, he or she may be asked to explain aspects of the medical care and to account for other events that preceded the victim’s demise, but should not stray outside his or her area of expertise.

Incidence Head injuries are very common events, particularly in modern urban society. All age groups are affected, although the nature of the resultant lesions varies somewhat between the very young, the healthy adolescent or adult, and the more elderly patient. Textbooks and reviews commonly quote annual incidences per 100,000 population, with breakdowns by severity and age group. Unfortunately, because of different study methods used, there is significant variation between estimates, sometimes as much as five or even tenfold (Brazinova et al. 2016; Roozenbeek et al. 2013). The commonest causes are road traffic collisions, falls, assaults and sports injuries. Military trauma constitutes a special category (Roberts et al. 2016; Carr et al. 2017).

Classification There are several different ways in which head injuries can be categorised, including by severity and type of damage and causative mechanisms, but even then different systems are adopted (Friedland and Hutchinson 2013). Commonly used measurements are the conscious level at presentation, the duration of coma and the period of post-traumatic amnesia.1 Radiological appearances are also important (Marshall et al. 1992); neuroradiology is covered in Chapter 3. Traditional divisions into mild, moderate and severe have been modified in some classification systems to permit meaningful retrospective classification even if there is incomplete data for individual cases (Table 2.1). An alternative approach is to consider clinical management needs. Patients who are less than fully conscious, when first attending hospital, will require admission. Most undergo observation in the first instance but some may require immediate surgical intervention and others undergo delayed surgery.

Primary injuries These include scalp injury, skull fractures, brain contusions and lacerations, as well as subarachnoid haemorrhage, arterial tears and primary axonal injury (Figure 2.1a and b). These topics are covered from the neuropathological aspect in the relevant chapter. All such mechanical damage, caused by the initial 1

The interval elapsed between the injury and the return of continuous memory.

injuring forces, cannot be reversed by medical intervention and the early management of head injuries is therefore directed at preventing the development of secondary brain damage.

Secondary damage In the first few days after injury a variety of mechanisms act at the tissue/cellular level propagating the primary damage and causing secondary injury. These include oxygen and free radical activity, calcium toxicity and neuroinflammation, as well as local ischaemia and hypoxia (see also Chapters 11 and 12). Further, secondary insults may result from systemic biochemical disturbances such as hyperglycaemia or from episodes of sepsis and pyrexia. The onset of seizures may also add to brain damage, particularly if prolonged and associated with hypoxia caused by ventilatory impairment. Secondary traumatic brain injury can also be caused by extradural haematoma, acute and chronic subdural haematomas, intracerebral haemorrhage and oedematous brain swelling. The neuropathology of these lesions is covered in the relevant chapters.

Table 2.1 Grading of head injury by severity Definite evidence of injury, moderate to severe in nature If one or more of the following features are recorded: • • • • • •

The cause of the victim’s death Loss of consciousness for more than 30 minutes Post-traumatic amnesia of more than 24 hours Worst recorded GCS of less than 13, in first 24 hours* Penetrating skull injury Presence of extradural, subdural, subarachnoid or intracerebral haemorrhage, or of cerebral contusion

Probable injury, albeit mild in nature In the absence of all of the above but in the presence of one or more of the following: • Any loss of consciousness but of less than 30 minutes’ duration • Any post-traumatic amnesia of less than 24 hours’ duration • Any skull fracture but with the dura intact Possible head injury, as judged by recorded symptoms In the absence of all of the above but in the presence of one or more of the following: • Altered mental state • Focal neurological symptoms • Headache, nausea, dizziness or visual disturbances With any effects of sedation, alcohol or other drugs having been excluded. Source: Based on the Mayo classification (Malec et al. 2007). *

18

Clinical aspects of head injury

Figure 2.1 Non-surgical lesions. a. Traumatic subarachnoid haemorrhage. This CT image of a relatively atrophic brain shows small quantities of subarachnoid blood in the right Sylvian fissure (short arrows) and in the occipital horns of both lateral ventricles (long arrows). b.  A xonal shearing injury. This gradient-echo MR scan reveals numerous, widespread low signal (arrows) areas, consistent with microhaemorrhages and indicative of shearing injuries within the white matter. c.  Carotid artery dissection. A 3D CT reconstruction, showing both internal jugular veins (JV) and both internal carotid arteries (ICA). The arrow points to a discontinuity in flow along the left internal carotid artery, caused by a traumatic dissection. Extradural (epidural) haematoma is particularly important because it is not usually associated with any significant primary brain damage and yet can prove lethal if large enough and not evacuated promptly. It typically presents with the so-called ‘lucid’ interval, but by no means invariably. This period can be of varying length, dependant on the rapidity of accumulation of the haematoma. With increase in clot size, drowsiness, confusion, ipsilateral pupillary dilatation and hemiparesis may be seen, followed by coma. Eventually, without surgical intervention, lethal brainstem compression will follow. Acute subdural haematomas, whilst also often requiring surgical attention, are commonly associated with some degree of primary brain damage and the latter may be of greater influence in determining the eventual outcome. Chronic subdural haematoma, intracerebral haemorrhage, brain swelling and oedema are covered in the relevant

chapters. Complications of head injury are covered later in this chapter.

Diagnosis and management Initial care at the scene is focused on maintenance of the airway and control of bleeding, to help prevent systemic hypoxia. In the accident and emergency (A&E) department, for the same reasons, examination is made for chest, abdominal or pelvic injuries. Intracranial lesions that might cause secondary brain damage are unlikely to have developed at this very early stage. Medical history for underlying disease should be obtained if possible. Guidelines exist to advise on which patients require computerised tomography (CT) scanning and which need admission for further observation (Table 2.2) (National Institute for Health and Care Excellence 2014).

Non-surgical treatment

Table 2.2 Indications for admission and CT scanning • • • • • • • • • • • • • • •

Altered conscious level at any time after injury Amnesia for the event Persistent vomiting Severe headaches Seizures Altered mental state Focal neurological deficit Compound skull fractures or penetrating injuries Evidence of base of skull fracture* Large scalp haematoma Anticoagulant usage Alcohol intoxication High-energy mechanism of injury Young children or elderly patients Suspected non-accidental injury

* Mastoid bruising (Battle’s sign); periorbital bruising (racoon eyes).

Glasgow Coma Scale The most important single clinical observation is the conscious level, which is now almost universally recorded as the Glasgow Coma Scale (GCS) (Table 2.3) (Teasdale and Jennett 1974). Whereas division of conscious level into the three components of the GCS is valid, and whilst an overall numerical score is of great value in studies of outcome, communication between health professionals is perhaps better served by verbal description of the GCS component parts, e.g. the patient opens his eyes to pain, localises the stimulus and utters incomprehensible sounds. Of the three GCS components, the best motor response is probably the most useful indicator of overall cerebral function at any one moment. The best GCS is 15 (6 + 5 + 4), the worst 3 (1 + 1 + 1). A patient is described as comatose when there is no speech, no eye opening and the best motor response is that of localising to pain. Table 2.3 Glasgow Coma Scale Eye opening 4 Spontaneously 3 To speech 2 To pain 1 None Best motor response 6 To command 5 Localising pain 4 Normal flexion 3 Abnormal flexion 2 Extension 1 None Speech 5 Orientated 4 Confused 3 Inappropriate words 2 Incomprehensible sounds 1 None Note: The minimum score is 3. Fully conscious and orientated scores 15. A score of 8 or less is classified as coma.

Investigations After haemodynamic stabilisation and airway control (the latter may necessitate ventilation), one or more of the following may be performed; further details of radiological examinations are covered in Chapter 3. CT head scan is performed, commonly as part of a wholebody, screening CT survey. Inclusive examination of the cervical spine is particularly important with cases of head injury. With the establishment, in many developed countries, of major trauma units, to which patients can be transported rapidly and in which there is ready access to CT scanning at all times, there is a potential drawback of early cranial imaging. An intracranial mass lesion may not have developed within the first hour or so after the injury and the initial CT scan may appear reassuringly normal. Repeat imaging may therefore be indicated, after about 4 hours, if the patient remains less than fully conscious or needs to be ventilated. CT angiography may be used if intracranial vascular damage is suspected, or if there is concern that the head injury might have been the result of a primary vascular episode, such as an aneurysmal subarachnoid haemorrhage. CT perfusion studies may be used occasionally for prognostic purposes, when the extent of cerebral infarction needs to be known. MR imaging is more sensitive than CT in detecting damage at the microscopic level, including traumatic axonal injury. Its main role is therefore more prognostic, in patients who fail to improve at an expected rate. Intracranial pressure (ICP) monitoring is now used in the management of a head injury where sedation and ventilation are required for more than a short period. Despite a lack of clear evidence that its use improves outcome (Chesnut et al. 2012), ICP monitoring is now widely adopted in the management of any head injury that requires pharmacological sedation and ventilation for more than a short period. In the absence of observable clinical signs, other than pupillary reactions, continuous measurement of the ICP may indicate the need for follow-up CT scanning, thereby identifying any potential surgical targets that may have developed. Even in the absence of such secondary lesions, non-surgical treatments may be directed by ICP measurements. Other monitoring modalities include microdialysis, which provides a means of monitoring the local brain biochemistry,2 and measurement of cerebral blood flow and of brain oxygenation. Such techniques are used mainly in research programmes and have not yet become part of routine head injury management.

Non-surgical treatment Observation For the conscious or mildly confused patient, admitted for nursing observation, standard measurements of blood pressure, pulse, respiratory rate and temperature are carried out, along with recordings of the GCS and more detailed notes about limb movements. The frequency of such observations and the length of time for which they are carried out is a matter to be decided by the supervising medical staff, in consultation with the local neurosurgical team. A young, healthy male, with a modest frontal lobe contusion, may appear very well and keen 2

Measuring levels of glucose, pyruvate, lactate, glutamate and glycerol.

19

Clinical aspects of head injury

20

Acute surgical procedures

Decompressive craniectomy A particularly contentious issue is the role of surgery for generalised brain swelling in the absence of a focal mass lesion. Although the removal of a generous calvarial bone flap is commonly performed as an adjunct to evacuation of an acute subdural haematoma, the role of such external decompression alone, for progressive and diffuse brain swelling, remains a subject of debate. Two major studies have addressed this question and produced varying results (decompressive craniectomy in diffuse traumatic brain injury study – DECRA, and the Randomised Evaluation of Surgery with Craniectomy for Uncontrollable Elevation of Intracranial Pressure trial – RESCUE ICP). The number of good recoveries is, overall, the same with decompressive surgery and with best medical treatment alone (Hutchinson et al. 2016). Notwithstanding these study results, the younger patient, who displayed good motor responses before being ventilated, and who then develops progressively rising ICP, may well benefit from a craniectomy. In an older patient, in a person with significant co-morbidity and for the individual who was comatose from the outset, such surgical intervention may reasonably be deemed futile. CSF drainage is sometimes performed to reduce ICP. Such external ventricular drains may be very effective but the surgical target (the lateral ventricle) will usually be small in the early stages after head injury. Skill is required if it is to be entered successfully, via a burr hole and modern image guidance techniques can be very helpful in these circumstances. Lumbar drainage may be used as an alternative, provided that there are no mass lesions developing intracranially. The latter would pose a risk of causing craniospinal pressure differentials to develop, leading to the creation of potentially lethal transtentorial (uncal) or medullary (tonsillar) pressure cones.

Scalp lacerations are commonly dealt with in A&E units. More complex scalp trauma may require plastic surgical procedures.

Interventional radiological procedures

to return home but a single seizure could lead to significant intracranial decompensation and deterioration in conscious level. Prophylactic anticonvulsants may be indicated in such a patient.

Critical care More serious head injuries will very likely be admitted to a critical care unit where more active intervention, as well as close observation, is required (Brain Trauma Foundation 2016). If the patient requires ventilation, to deal with associated chest injuries or to safeguard the airway, clinical neurological assessment will be limited, which is when ICP monitoring is used. In health, the upper limit of normal ICP is about 15 mm of mercury but readings up to 20 mm would be tolerated in most head injury victims. Pressure rising above this level, for sustained periods, would necessitate follow-up imaging, looking for surgically correctable causes. Otherwise, measures are followed to try and reduce the ICP and maintain cerebral perfusion. These, for the most part, consist of optimising blood gases, mean arterial pressure and blood chemistry. Manoeuvres such as hyperventilation and the use of osmotic 3 and diuretic therapy reduce ICP in the short term but are not effective over longer periods (Berger-Pelleiter et al. 2016; Burgess et al. 2016). Steroids are of no benefit and may be harmful (CRASH Trial Collaborators 2005). Measures aimed at reducing cerebral metabolic demand, such as the use of barbiturate coma (Roberts and Sydenham 2012) or induced hypothermia (Andrews et al. 2015), whilst theoretically attractive, have likewise not proven to be of benefit in the management of severe head injuries.

Skull fractures Linear vault fractures do not require treatment but depressed fractures may need to be elevated. This might be for cosmetic reasons or, occasionally, because the skull depression is sufficient to produce significant brain compression or midline shift. Compound depressed fractures usually require wound toilet or debridement, combined with elevation and reconstruction. Basal skull fractures causing CSF leakage are usually managed expectantly initially but persisting or recurring leaks may require input from a skull base surgeon. Associated injures to the facial skeleton may need maxillofacial input. Ophthalmic surgeons may be involved if one or both eyes are damaged. Intracranial surgery can be of benefit for extradural and acute subdural haematomas, as well as focal cerebral contusions. Not all such collections require removal but if they are of sufficient size, and especially if they involve the frontal or temporal regions and threaten to cause tentorial pressure cones and brainstem compression, they will need to be removed. Dilemmas may arise when CT scans reveal a significantly sized acute subdural haematoma in an elderly patient, who is confused and perhaps a little drowsy. Craniotomy and evacuation of the clot carries the risks associated with anaesthesia in someone likely to have co-morbidities, and for whom there is always a risk of post-operative bleeding. 3

Mannitol or hypertonic saline.

Once mainly a post-mortem room finding, intracranial arterial dissections, affecting the internal carotid or vertebral arteries, are increasingly detected by routine imaging techniques (Figure 2.1c). Treatment with anti-platelet medication aims to discourage secondary clot formation but traumatic aneurysms may form, after an interval of a few days. There may then be a role for endovascular repair procedures.

Complications of head injury Post-traumatic epilepsy A particular problem following head injury is the risk of epilepsy. Some 5 per cent of closed head injury victims will suffer a seizure within the first week and another 5 per cent will do so thereafter (Jennet and Teasdale 1981). Early-onset seizures, developing within the first 24 hours after injury, are regarded as being provoked and they do not require long-term anticonvulsant treatment. Whether they increase the likelihood of delayed-onset seizures developing is a matter of debate. Following one delayed-onset seizure, however, the chances of further fits occurring is increased. In the longer term, following recovery from head injury, the relative risk4 of seizures developing is low for mild injuries (1–2 per cent), and for moderate injuries (2–4 per cent), but rises significantly, to 12–24 per cent, for severe traumatic brain injury (Annegers et al. 1998). The risk 4

Compared to a similar, normal population.

Complications of head injury

Figure 2.2 Post-traumatic hydrocephalus. a. CT appearances within twenty-four hours of injury. A moderate-sized acute subdural haematoma (short arrows) and area of frontal contusions (long arrow), in a young male, lead to rising ICP and the need or decompressive craniectomy. The cerebral ventricles (asterisks) are of a normal size at this stage. b.  Two months after injury. The craniectomy is slack and the septum pellucidum is midline in position (long arrows). A moderate area of porencephaly is evident (short arrows). Some ventricular enlargement is evident, with rounding of the frontal horns (asterisks) but this is managed conservatively as clinical improvement is occurring. c.  Developments after 12 months. Rehabilitation has plateaued. The ventricles (asterisks) are now seen to have enlarged dramatically. As a result, a volume of oedematous white matter (short arrows) and overlying cerebral cortex (long arrows) have herniated through the craniectomy defect. is particularly high if the victim has suffered a cerebral contusion or intracerebral haemorrhage, or in the cases of compound depressed skull fracture where the dura is breached. Penetrating battlefield injuries are particularly likely to lead to the development of seizures (Salazar et al. 1985). The risk of delayed seizure has clear implications for people who hold driving licences. In the United Kingdom, the Drivers Vehicle and Licensing Agency issues clear guidelines as to restrictions that are placed on driving after head injury, with and without early seizures (Driver and Vehicle Licensing Agency 2016). Hydrocephalus seldom develops in the early stages following head injury, unless there is a posterior fossa haematoma obstructing the outflow of CSF from the fourth ventricle. From about a month onwards, however, at a stage when ICP will be reducing, there is a possibility that the ventricles will begin to enlarge. Blood and other cellular debris partially block the arachnoid granulations, leading to the development of a communicating hydrocephalus (Figure 2.2). Lumbar punctures may tide the patient over until CSF absorption improves but permanent CSF diversion may be indicated, in the form of a ventriculoperitoneal shunt. Reconstructive surgery may be indicated, to make good a skull vault defect created by earlier, emergency removal of a calvarial bone flap. Normally, an interval of 3 to 6 months is allowed to pass, during which time ICPs settle, inflammation subsides and a neo-dura forms between the brain and the overlying scalp. Thereafter, a cranioplasty can be performed, either by replacing the bone flap that was removed during the emergency decompression5 or by inserting a prosthetic substitute 5

Sometimes the surgeon will ‘store’ this in the subcutaneous fat of the patient’s abdominal wall.

such as titanium. Smaller defects can be filled with acrylic cement. The commonest complication is that of chronic wound infection (Acciarri et al. 2016) but benefits include cosmesis and the provision of protection for the underlying brain. Another argument in favour of restoring the skull vault is that, by allowing the brainstem to adopt a less distorted position inside the head, the reticular activating system becomes more active. A dependent patient may then become more responsive and, at the very least, somewhat easier to nurse. CSF fistulas, presenting with rhinorrhoea or otorrhoea, may develop when a skull fracture, involving the paranasal sinuses or the middle ear, also damages the underlying dura. If the leak persists, intervention may be required, to prevent meningitis developing.

Cranial nerve injuries A number of cranial nerves may be damaged by a head injury, with resultant clinical features associated with their function. Anosmia (loss of smell) may result from damage to olfactory nerve fibres passing through a fractured cribriform plate. Endocrine disturbances, the result of hypothalamic or pituitary damage, are easily overlooked, unless specifically sought. Diabetes insipidus is a not uncommon complication in the acute stages of more major head injuries but will usually settle with time, as hypothalamic axons in the posterior part of the pituitary gland regenerate. Between one in five and a quarter of victims of serious head injury are, however, likely to develop hypopituitarism in the long term, involving at least one hormone deficiency (Schneider et al. 2007; Tanriverdi et al. 2015).

21

Clinical aspects of head injury

22

Long-term sequels Long-term complications of any head injury may include motor and neuropsychological impairments,6 leading to varying degrees of disability 7 and handicap.8 Many survivors of serious or even moderate head injury will never return to work or education and will be dependent to a varying degree. Prognostic models have been developed, aiming at predicting the probability of a patient surviving a traumatic brain injury: www.crash2.lshtm.ac.uk/Risk%20calculator; www.tbi-impact. org/?p=impact/calc (MRC CRASH Trial 2008; Steyerberg et al. 2008). The eventual outcome from traumatic brain injury, taking into account cognitive, motor, sensory and other neurological deficits, as well as emotional and psychiatric disturbances, can be measured on various functional outcome scales, such as the extended Glasgow Outcome Scale (Table 2.4), (Jennett and Bond 1975; Jennett et al. 1981). Using tools such as these, a number of prognostic indicators have been identified (Table 2.5) (Murray et al. 2007). These outcome scores apply for the most part to single episodes of brain trauma. The cumulative effects of repeated, low-level trauma, as seen in chronic traumatic encephalopathy, are covered in Chapter 14. Post-concussion syndrome is a recognised complication of mild-to-moderate head injury, when the individual recovers without major physical deficits but continues to be plagued by a variety of symptoms (Table 2.6). The relative contribution of organic brain damage and psychological factors remains uncertain. Table 2.4 Glasgow Outcome Scale (extended) Good recovery Upper: normal life, without deficit or disability Lower: normal living, despite minor residual deficits Moderate disability Upper: independent, albeit with less marked disability Lower: independent but with significant residual deficits Severe disability Upper: independent at home but not elsewhere Lower: conscious but dependent upon others for care Vegetative state Unable to obey commands or communicate Dead Table 2.5 Indicators of poor outcome • • • •

Increasing age Low motor component of GCS Absent pupillary reactions Significant midline shift, effaced basal cisterns or extensive subarachnoid blood seen on CT scan • Periods of hypoxia, hypotension or anaemia • Sustained rises of intracranial pressure • Genetics – the presence of apolipoprotein E4 allele (Panenka et al. 2017) 6 7 8

Impairments are specific neurological or other physical deficits. Disability refers to the functional loss resulting from an impairment. Handicaps are the restrictions on daily living resulting from a disability.

Table 2.6 Post-concussion syndrome • • • • • • • • • • • •

Headaches Dizziness Tinnitus Photo- and phonophobia Poor short-term memory Poor concentration Irritability Disturbed sleep pattern Fatigue Depression and anxiety Lack of initiative and drive Reduced libido

Persisting vegetative state A combination of local brain injuries, periods of reduced cerebral perfusion and shearing forces, which ‘disconnect’ the cerebral cortex from lower brainstem centres, leaves the vital brainstem centres functioning, providing homeostatic support for other body organs. Unfortunately, there may be little or no evidence of cognitive response or communication, even though brainstem reflexes, such as breathing and even swallowing, remain intact. If the victim displays some awareness of his or her surroundings, albeit inconsistently, the term minimally conscious state is used. The expression ‘’prolonged disorders of consciousness” encompasses both conditions (Royal College of Physicians 2013). Locked-in syndrome develops when both the upper ­brainstem and higher centres, including the cerebral cortex, remain functional but are disconnected from lower centres. The lesion involves the corticospinal tracts at the level of the pons. There is no voluntary motor movement but communication is possible by way of eye movement – the third nerve nuclei being higher in the brainstem. Long-term survival is possible but most affected individuals are left dependent upon modern technological aids. Recovery of some additional motor function is seen in some cases (Chisholm and Gillett 2005).

Brainstem death Once the brain is dead, an individual cannot exist as a conscious being. Furthermore, once the brainstem has infarcted, consciousness can never return, even if cortical electrical activity persists for a while. The other body organs begin to fail within a matter of days, or a few weeks at the outside. Relatives will normally wish to see their loved-one before artificial support is withdrawn but they should never be asked to make the decision to take this step, or even to give their consent to it. They are not qualified to do so. Before tests for brainstem death are carried out, certain requirements need to be satisfied (Table 2.7A) (Academy of Medical Royal Colleges 2010). When individual reflexes are tested (Table 2.7B), strong stimuli are needed, including a bright light for pupillary responses and ice-cold water for caloric tests.9 The final and most fundamental reflex tested is that of respiratory drive. Ventilatory support is suspended, allowing the blood carbon dioxide level to rise to a level normally more than sufficient to initiate breathing. Adequate oxygenation is maintained 9

The external auditory meatuses need to be inspected and cleared of wax, if necessary, to make this test valid.

References

Table 2.7 Brainstem death A. Pre-requisites • Pathology known to lead to brainstem death (e.g. traumatic brain injury, spontaneous intracranial haemorrhage), this pathology being irremediable • No residual effects of paralysing or sedative drugs • Normal body temperature, biochemistry and endocrine function • Haemodynamic stability B.  Reflexes tested • Pupillary reactions to light (afferent: optic nerve; efferent: oculomotor nerve) • Corneal blink reflex (afferent: trigeminal nerve; efferent: facial nerve) • Supraorbital nerve pressure and limb or facial movement (afferent: trigeminal nerve; efferent: spinal motor tracts and nuclei) • Nail bed pressure and facial grimace (afferent: peripheral nerves; efferent: spinal sensory tracts and facial nerve) • Caloric testing (afferent: vestibular nerve; efferent: oculomotor and abducent nerves) • Dolls eye movements* (afferent: vestibular nerve; efferent: oculomotor and abducent nerves) • Pharyngeal stimulation (gag reflex) (afferent: glossopharyngeal nerve; efferent: vagus nerve) • Tracheal stimulation (cough reflex) (afferent: vagus nerve; efferent: vagus nerve) • Respiratory drive (direct stimulation of the medulla; PaCO2 ≥ 6 kPa and then rising by at least 0.5 kPa) * Optional; omit in presence of cervical spine injury. during this process by passage of a continuous flow of oxygen down the endotracheal tube. Neurological tests are then carried out by two adequately experienced practitioners10 and repeated after an interval, as checks against initial observation errors. Time of death is then given as being at the conclusion of the first set of tests.

Prevention The adoption of national guidelines for the management of head injuries and their treatment in neurosurgical units is expected to lead to a decrease in mortality from head injuries (Fuller et al. 2001; Fountain et al. 2017). Reduction in the overall burden created by so many head injuries will, however, come about not so much as a result of medical advances as by technological advances, supported by enlightened legislation (Lieutaud et al. 2016). The dangers of drink driving are well established. Car designs continue to provide ever more protection but mandatory introduction of speed regulators could achieve more. There is increasing interest in injuries sustained on the sports field (McKee et al. 2014). Measures aim, in particular, at preventing the rare but catastrophic development of acute brain swelling in the player who returns to the game after an apparently mild concussion. Contact sports injury is covered in Chapter 14. Non-accidental injuries, in infants, young children and vulnerable adults, make distressing headlines whenever they come 10

Registered for more than five years and competent to conduct and interpret the test findings.

to public attention. Their identification and prevention involve not only the medical profession but also a variety of others, in a multidisciplinary setting. This topic is covered in Chapter 15.

Acknowledgement The author is indebted to Tony Belli, Professor of Neurotrauma at the University of Birmingham, UK, for his helpful observations and suggestions in reviewing the contents of this chapter.

References Academy of Medical Royal Colleges. 2010. A code of practice for the diagnosis and confirmation of death. http://www.aomrc.org.uk/ publications/reports-guidance/code-practice-diagnosis-confirmationdeath/. Acciarri, N., Nicolini, F. and Martinoni, M. 2016. Cranioplasty: routine surgical procedure or risky operation? World J Surg Res, 5, 22–33. Andrews, P.J.D., Sinclair, H.L., Rodriguez, A., Harris, B.A., Battison, C.G., Rhodes, J.K.J. and Murray, G.D. 2015. Hypothermia for intracranial hypertension after traumatic brain injury. N Engl J Med, 373, 2403–2412. Annegers, J.F., Hauser, W.A., Coan, S.P. and Rocca, W.A. 1998. A populationbased study of seizures after traumatic brain injuries. N Engl J Med, 338, 20–24. Berger-Pelleiter, E., Émond, M., Lauzier, F., Shields, J.-F. and Turgeon, A.F. 2016. Hypertonic saline in severe traumatic brain injury: a systematic review and meta-analysis of randomized controlled trials. CJEM, 18, 112–120. Brain Trauma Foundation. 2016. Guidelines for the Management of Severe Traumatic Brain Injury, 4th edn. http://www.braintrauma.org/guidelines/guidelines-for-the-management-of-severe-tbi-4th-ed#/. Brazinova, A., Rehorcikova, V., Taylor, M.S., Buckova, 1V., Majdan, M., Psota, M., Peeters, W., Feigin, V., Theadom, A., Holkovic, L. and Synnot, A. 2016. Epidemiology of traumatic brain injury in Europe: a living systematic review. J Neurotrauma, 33, 1–30. Burgess, S., Abu-Laban, R.B., Slavik, R.S., Vu, E.N. and Zed, P.J. 2016. A systematic review of randomized controlled trials comparing hypertonic sodium solutions and mannitol for traumatic brain injury: implications for emergency department management. Ann Pharmacother, 50, 291–300. Carr, D.J., Lewis, E. and Horsfall, I. 2017. A systematic review of military head injuries. BMJ Military Health, 163, 13–19. Chesnut, R.M., Temkin, N., Carney, N., Dikmen, S., Rondina, C., Videtta, W. and Petroni, G. et al. 2012. A trial of intracranial-pressure monitoring in traumatic brain injury. N Engl J Med, 367, 2471–2481. Chisholm, N. and Gillett, G. 2005. The patient’s journey: living with locked-in syndrome. BMJ, 331, 94–97. CRASH Trial Collaborators. 2005. Final results of MRC CRASH, a randomised placebo-controlled trial of intravenous corticosteroid in adults with head injury—outcomes at 6 months. The Lancet, 365, 1957–1959. Driver and Vehicle Licensing Agency. 2016. Assessing fitness to drive: guide for medical professionals. http://www.gov.uk/guidance/assessingfitness-to-drive-a-guide-for-medical-professionals. Fountain, D.M., Kolias, A.G., Lecky, F.E., Bouamra, O., Lawrence, T., Adams, H., Bond, S.J. and Hutchinson, P.J. 2017. Survival trends after surgery for acute subdural hematoma in adults over a 20-year period. Ann Surg, 265, 590. Friedland, D. and Hutchinson, P. 2013. Classification of traumatic brain injury. Adv Clin Neurosci Rehabil, 4, 12–13. Fuller, G., Bouamra, O., Woodford, M., Jenks, T., Patel, H., Coats, T.J., Oakley, P., Mendelow, A.D., Pigott, T., Hutchinson, P.J. and Lecky, F. 2001. Temporal trends in head injury outcomes from 2003 to 2009 in England and Wales. Br J Neurosurg, 25, 414–421. Hutchinson, P.J., Kolias, A.G., Timofeev, I.S., Corteen, E.A., Czosnyka, M., Timothy, J. and Anderson, I. et al. 2016. Trial of decompressive craniectomy for traumatic intracranial hypertension. N Engl J Med, 375, 1119–1130. Jennett, B., Snoek, J., Bond, M.R. and Brooks, N. 1981. Disability after severe head injury: observations on the use of the Glasgow Outcome Scale. J Neurol Neurosurg Psychiatry, 44, 285–293.

23

24

Clinical aspects of head injury Jennett, B. and Teasdale, G. (eds) 1981. Management of Head Injuries, 3rd edn. FA Davis Company. Jennett, B. and Bond, M. 1975. Assessment of outcome after severe brain damage: a practical scale. Lancet, 305, 480–484. Lieutaud, T., Gadegbeku, B., Ndiaye, A., Chiron, M. and Viallon, V. 2016. “The decrease in traumatic brain injury epidemics deriving from road traffic collision following strengthened legislative measures in France.” PLoS One, 11(11), p. e0167082. Malec, J.F., Brown, A.W., Leibson, C.L., Flaada, J.T., Mandrekar, J.N., Diehl, N.N. and Perkins, P.K. 2007. The Mayo classification system for traumatic brain injury severity. J Neurotrauma, 24, 1417–1424. Marshall, L.F., Bowers Marshall, S., Klauber, M.R., Van Berkum, C.M., Eisenberg, H., Jane, J.A., Luerssen, T.G., Marmarou, A. and Foulkes, M.A. 1992. The diagnosis of head injury requires a classification based on computed axial tomography. J Neurotrauma, 9, S287–S292. McKee, A.C., Daneshvar, D.H., Alvarez, V.E. and Stein, T.D. 2014. The neuropathology of sport. Acta Neuropathol, 127, 29–51. MRC CRASH Trial Collaborators, Perel, P., Arango, M., Clayton, T., Edwards, P., Komolafe, E. and Poccock, S. et al. 2008. Predicting outcome after traumatic brain injury: practical prognostic models based on large cohort of international patients. BMJ, 336, 425–429. Murray, G.D., Butcher, I., McHugh, G.S., Lu, J., Mushkudiani, N.A., Maas, A.I.R., Marmarou, A. and Steyerberg, E.W. 2007. Multivariable prognostic analysis in traumatic brain injury: results from the IMPACT study. J Neurotrauma, 24, 329–337. National Institute for Health and Care Excellence. 2014. Head injury: assessment and early management. https://www.nice.org.uk/guidance/cg176. Panenka, W.J., Gardner, A.J., Dretsch, M.N., Crynen, G.C., Crawford, F.C. and Iverson, G.L. 2017. Systematic review of genetic risk factors for sustaining a mild traumatic brain injury. J Neurotrauma, 34, 2093–2099.

Roberts, I. and Sydenham, E. 2012. Barbiturates for acute traumatic brain injury. Cochrane Database Syst Rev Issue 12. Art. No.: CD000033. Roberts, S.A.G., Toman, E., Belli, A. and Midwinter, M.J. 2016. Decompressive craniectomy and cranioplasty: experience and outcomes in deployed UK military personnel. Br J Neurosurg, 30, 529–535. Roozenbeek, B., Maas, A.I.R. and Menon, D.K. 2013. Changing patterns in the epidemiology of traumatic brain injury. Nat Rev Neuro, 9, 231–236. Royal College of Physicians. 2013. Prolonged Disorders of Consciousness: National Clinical Guidelines. https://www.rcplondon.ac.uk/guidelines-policy/prolonged-disorders-consciousness-national-clinicalguidelines. Salazar, A.M., Jabbari, B., Vance, S.C., Grafman, J., Amin, D. and Dillon, J.D. 1985. Epilepsy after penetrating head injury. I. Clinical correlates: a report of the Vietnam Head Injury Study. Neurology, 35, 1406–1414. Schneider, H.J., Kreitschmann-Andermahr, I., Ghigo, E., Stalla, G.K. and Agha, A. 2007. Hypothalamopituitary dysfunction following traumatic brain injury and aneurysmal subarachnoid hemorrhage: a systematic review. JAMA, 298, 1429–1438. Steyerberg, E.W., Mushkudiani, N., Perel, P., Butcher, I., Lu, J., McHugh, G.S. and Murray, G.D. et al. 2008. Predicting outcome after traumatic brain injury: development and international validation of prognostic scores based on admission characteristics. PLoS Med, 5(8), e165. Tanriverdi, F., Schneider, H.J., Aimaretti, G., Masel, B.E., Casanueva, F.F. and Kelestimur, F. 2015. Pituitary dysfunction after traumatic brain injury: a clinical and pathophysiological approach. Endocr Rev, 36, 305–342. Teasdale, G. and Jennett, B. 1974. Assessment of coma and impaired consciousness: a practical scale. Lancet, 304, 81–84.

3

Imaging of head trauma Calvin Soh

Introduction

MRI

Each year approximately 1.4 million people are seen in emergency departments in England and Wales with a recent head injury, with about 200,000 being admitted. About 20 per cent will have evidence of brain damage or skull fracture. Only about 0.2 per cent of all attendees will die as a result of the head injury. As brain injury cannot be physically determined except for external signs, imaging is crucial in assessing extent of intracranial injury. Cross-sectional imaging is necessary in the form of CT and MRI. CT remains the mainstay for initial assessment as it is widely available, fast and is very good for assessing soft ­tissue and bony injury, facilitating any emergent or urgent neurosurgical intervention. MRI is reserved for subsequent detailed evaluation of parenchymal injury as well as subtle microhaemorrhages for diffuse traumatic axonal injury. MRI can also detect evidence of previous head injury by identifying areas of encephalomalacia at classic sites of injury. A brief description of the principles of CT and MRI imaging is discussed as a prelude to a specific description of the various types of head injury.

MRI utilises a different principle in image acquisition. Hydrogen, being present in water molecules and hence in all body tissues, behaves like micro-compass with a single proton. When placed inside a large static magnet like an MRI scanner, these are partially aligned along the main magnetic field through the bore of the scanner. Radiofrequency pulses are deployed to disrupt the alignment of these micro-compasses. Due to interaction of these hydrogen protons with each other in neighbouring tissues, the restoration of equilibrium in realignment, on cessation of the radiofrequency pulse, will emit a radio signal which is then captured by the antennas in surface and body coils to reconstruct an image. The complex physics of MR imaging allows different imaging parameters to generate imaging of different pulse sequences to maximise the contrast resolution of different soft tissues. The common sequences in clinical practice include the following:

CT imaging of the brain CT utilises radiation to generate the information by measuring the attenuation of an X-ray beam by the brain tissues and skull vault on a bank of detectors on the contralateral side of the gantry ring. The data is then reconstructed into an axial image. Modern multislice CT scanners can now acquire a head scan in 6–7 seconds with slice resolution of 0.625 mm while a 640 slice scanner can acquire it in 0.3 seconds with a resolution of 0.3 mm. Imaging is then reconstructed on both soft tissue windows to assess for brain injury, swelling and haemorrhage, and on a bone algorithm to assess for skull vault and skull base fractures. Choice of window levels and window widths is key to interpretation as these alter the contrast between tissues. On CT, greyscale units are measured in arbitrary Hounsfield unit which is a linear scale, which assigns a value of 0 HU to water and  −1000 HU to air. For very dense bone, the value is around +2000 HU. Window width describes the spread of greyscale from black to white. Window level is the centre of the window width with equal spread of CT numbers on both sides (see Figure 3.1). For differentiating soft tissues of slightly different brain density, a narrower window width is utilised, whilst assessment of fractures and intracranial air, the wider range of densities will require a wider window width but the soft tissues are indiscernible (see Figure 3.2). Typical CT window settings for reviewing the brain: • • • • •

Brain Subdural Stroke Temporal bones Soft tissues

W:80 L:40 W:130-300 L:50-100 W:8 L:32 or W:40 L:40 W:2800 L:600 W:350–400 L:20–60

• T1-weighted imaging: which is very good for anatomy where the grey matter appears grey and white matter appears white; T1-hyperintense tissues include methaemoglobin (blood degradation product), fat, melanin, contrast agents, proteinaceous fluid and microcalcification. • T2-weighted imaging: which is generally utilised to look for pathology as water is hyperintense and most acute processes involve some increase in fluid-like oedema. Unlike T1W, grey matter is relatively white, whilst white matter is relatively grey, a reverse of the pattern seen in T1W imaging. • FLAIR (FLuid Attenuation Inversion Recovery) imaging: a similar principle to T2W imaging except that water signal is suppressed, utilised to look for pathology with increased fluid and is good for assessing tissues adjacent to CSF, e.g. periventricular white matter. Normal grey matter remains relatively hyperintense to white matter. • T2* or T2 gradient-echo sequence: similar to T2W sequence except technically different in acquisition, the imaging appearances are similar to T2 but the imaging is less sharp. The advantage is that it is sensitive to altered blood products and exaggerates the low T2 signal of blood products. Haemosiderin can remain visible as stigmata of previous haemorrhage. • Susceptibility-weighted imaging (SWI or SWAN): another sequence which is highly sensitive to heavy metals, calcification and blood products. • Diffusion-weighted imaging (DWI): employs the principle of measuring diffusivity of water molecules in extracellular CSF spaces. In acute cytotoxic oedema from infarction and acutely injured cells, cell swelling reduces the volume of extracellular CSF space, resulting in restricted diffusivity. Tissue oedema with increase in extracellular CSF volume results in increase in free diffusivity. DWI generates two sets of imaging, an isotropic diffusion

26

Imaging of head trauma

Figure 3.1 Diagram to depict the spread of greyscale used to adjust the contrast resolution in an image. Window width is the spread of black to white, window level is the mid-grey.

Figure 3.2 Brain and bone windows. Brain window (W:80 L:40) demonstrates brain tissue as grey, CSF as dark grey, bone is white and fat and air are both black. Bone window (W:3000 L:800) has a wider range of contrasts and hence subtle density difference between different soft tissues is lost, but bone is less white, allowing the marrowfat to be seen as lower density in the diploic space.

Blood products trace and an ADC (apparent diffusion coefficient) map that must be read in conjunction for correct interpretation. High signal on diffusion trace and corresponding low signal on ADC map indicates diffusion restriction. This is seen in acute brain infarction, pus in empyemas and abscesses, as well as in epidermoids and very cellular tissues like cerebral lymphoma. However, as DWI is an echo-planar imaging technique, it is likened to T2* imaging, and the diffusion trace image has T2 properties and exhibits ‘T2-shine-through’ and ‘T2-blackout’ effects on blood products. Oxyhaemoglobin (hyperacute bleed) and extracellular methaemoglobin (late subacute bleed) appear bright on the diffusion trace, mimicking true diffusion restriction and can be mistaken for an abscess or an acute infarct. ADC is generally reduced on all stages of blood products. Note: Contrast on CT is described as hyperdensity, isodensity and hypodensity, whilst on MRI is described as hyperintensity, isointensity and hypointensity.

Blood products Intracerebral haematoma evolves into different products according to a timeline: • • • •

Hyperacute (first 6–12 hours): Oxyhaemoglobin Acute (24–48 hours): Deoxyhaemoglobin (Figure 3.3) Early subacute (2–7 days): Intracellular methaemoglobin Late subacute (7–14 days): Extracellular methaemoglobin (Figure 3.4) • Chronic (14–28 days): Haemosiderin (but in collapsing haematoma cavities, the centre is liquefied whilst the outline is stained with haemosiderin, until a thin haemosiderin lined cleft remains) The time course of a haematoma is not absolute and can vary slightly with individuals, local environment and size of haematoma. The signal pattern merely characterises the stage of evolution. Extra-axial haemorrhage in the subarachnoid and subdural space does not evolve along the same timeline. This time course of different blood products affects the MRI signal but is less relevant to CT which relies on density.

Figure 3.3 MRI signal characteristics of an acute intracerebral haematoma with deoxyhaemoglobin: T2-hypointense, T1-isointense, T2*-hypointense, FLAIR-hypointense, DWI trace-hypointense and ADC-hypointense.

27

28

Imaging of head trauma

Figure 3.4 MRI signal characteristics of a late subacute intracerebral haematoma. Twenty-five days after the scan of the same bleed in Figure 3.4. Extracellular methaemoglobin is T2-hyperintense, T1-hyperintense, T2*-hyperintense, FLAIRhyperintense, DWI trace-hyperintense and ADC-hypointense. The configuration of the haematoma is the same but the surrounding oedema had resolved. CT density of acute haematoma is high. The density of extravasated blood increases up to 60–80 HU initially as the clot forms, and up to 80–100 HU as the clot retracts. The haematoma remains dense up to a week. This gradually reduces in density with time due to breakdown in the blood products but the mass effect can remain. Weeks or months later, the haematoma gradually collapses from a defined cavity into a slit-like cavity. Occasionally, a dependent horizontal fluid-fluid level is seen with hyperdensity of sedimented blood cells layering below and a hypodense supernatant above due to blood serum. MRI signal of blood products is highly complex as blood products evolve, each stage of evolution is reflected in the alteration of signal intensity, which also differs with the different imaging sequences (see Figure 3.5). Detailed description of the evolution of different blood products on MRI is beyond the scope of this review and is referred to a comprehensive review by Parizel et al. (2001). There are three important points about haematomas:

1. Acute blood is hyperdense on CT. 2. T1-hyperintensity suggests blood in the form of methaemoglobin as very few abnormalities are intrinsically hyperintense on T1. 3. Haematomas can mimic an abscess on DWI where there is high signal on the diffusion trace and low signal on ADC. In this situation, if CT shows hyperdensity, it is a hyperacute haematoma, and if it is hyperintense on T1, it is a late subacute haematoma due to extracellular methaemoglobin. It is important to note that the progression of the blood products from one stage to another is slower in subdural and extradural haematomas when compared with intra-axial haematomas in view of the good vascularisation and high oxygen tension in the dura. The signal characteristics of the different blood products remain the same. Rebleeding into a subdural haematoma (SDH) can be aged according to the signal intensity of the different stages of blood products.

Subdural haemorrhage and effusions

Figure 3.5 MRI signal characteristics of blood products at different stages of evolution. Up arrows denote increased signal, down arrows denote decreased signal and horizontal arrows denote signal intensity similar to normal brain. Intensity is also emphasised in greyscale. The thickened black outline of white arrows indicates rim hypointensity with central hyperintensity of haematoma.

Skull fractures Skull vault and skull base fractures are identified on a CT scan when reviewed on bone windows. Identifying skull vault fractures itself is of no clinical relevance but it directs attention to the site of injury allowing detailed examination of the subjacent brain, as well as the diametrically opposite sites for contrecoup injury when reviewed on soft-tissue window settings. Similarly, the relevance of identification of skull base fractures is to localise injury to the skull base foramina which transmit neurovascular structures. Fracture lines through these sites indicate significant transmission of forces through the neurovascular structures, potentially resulting in arterial dissection or neurapraxia. MRI has no role in fracture identification but any marrow oedema seen as signal abnormality may again localise site of injury.

Types of head injury: primary versus secondary

Secondary injuries result from subsequent physiological responses to the initial insult. These include diffuse brain swelling and haematomas with mass effect resulting in abnormal brain shift, herniation across compartments and hypoxic brain injury. The following injuries and the imaging will be discussed: 1. 2. 3. 4. 5. 6. 7. 8.

Subdural haemorrhage and subdural hygromas Extradural haemorrhage Subarachnoid haemorrhage Intraventricular haemorrhage Contusional brain injuries Diffuse traumatic axonal injury Herniation syndromes Hypoxic brain injuries

Subdural haemorrhage and effusions

Subdural haemorrhage is derived from tearing of the bridging veins as the cortical veins course anteriorly at an angle to Head injury is broadly categorised into primary and secondary drain into the superior sagittal sinuses (Figure 3.6). These veins injuries. are prone to rupture with occipital head injury, with angular Primary injuries are direct injuries which include acute acceleration-deceleration in the sagittal plane. Associated haemorrhage, intra-axial or extra-axial (subarachnoid, sub- skull vault fractures are uncommon. It is generally crescentic dural and extradural), or contusions, haemorrhagic or non-­ over the cerebral or cerebellar convexity. It does not cross dural haemorrhagic. Diffuse injuries include diffuse traumatic reflections of the falx cerebri and tentorial leaves. It can cross axonal injury resulting from shearing forces from rapid accel- sutural lines. Parafalcine and tentorial SDHs look like asymeration, deceleration and rotational forces. metric thickened dural folds.

29

30

Imaging of head trauma

Figure 3.6 Acute subdural haematoma. A venogram demonstrates the normal cortical veins draining into the superior sagittal sinus at an angle, prone to disruption in sagittal plane. CT scan shows an acute crescentic hyperdense haematoma overlying the left frontal lobe convexity.

SDHs are occasionally biconvex and can mimic extradural haematomas, occurring in those who may have had previous neurosurgical procedures or radiotherapy. Four proposed CT features that differentiate in favour of a SDH included a crescentic tail, an obtuse angle at the margin of the haematoma, a dural line above the haematoma and a direct communication to an underlying intracerebral haematoma. SDHs are radiologically classified into acute (within 1 week), subacute (1–3 weeks) and chronic (over 3 weeks), based on the evolution of the blood products. On CT, acute SDH is hyperdense to brain tissue due to the formation of a clot (Figure 3.6). A haematocrit level, seen as a horizontal fluid-fluid level, may be present in an acute large SDH, in patients on anticoagulation therapy or a bleed into a chronic SDH. However, it may also be seen as part of the normal evolution of a haematoma as clot lysis leads to separation of the red blood cells from the serum and gravitate dependently (Figure 3.8). The density of the haematoma will reduce with time, becoming isodense with brain and subsequently hypodense and resorb. Isodense SDH (Figure 3.7) is seen as a band of similar density uniformly displacing the brain surface, and the cerebral sulci do not reach the inner table of the skull vault. In the subacute stage, SDH s can expand resulting in mass effect with abnormal brain shift, becoming clinically symptomatic at this stage (Figure 3.8). In the subacute stage, neomembranes may develop in the haematoma with friable neovascularisation. These membranes are seen as curvilinear serpiginous strands within the collection (Figure 3.9). These vessels can rupture and result in acute on chronic SDHs. On CT, the acute blood appears as heterogeneous hyperdensity in a pre-existing hypodense collection.

Figure 3.7 CT shows asymmetry with mass effect compressing the left lateral ventricle. A crescentic band of grey-matter isodensity overlying the left cerebral hemisphere is due to acute/ subacute subdural haemorrhage.

Subdural haemorrhage and effusions

Figure 3.8 Acute subdural haematoma expansion. (a) Initial CT shows an acute left cerebral convexity subdural haematoma with early mass effect on subjacent brain. (b) CT scan 8 days later shows evolution of the subdural haematoma, becoming hypodense and with swelling, resulting in significant mass effect and midline shift. A shallow rim of remaining hyperdense blood is sedimenting posteriorly. (c) Subdural haematoma with haematocrit level – clot lysis leads to separation of red cells and debris leading to a haematocrit effect.

31

32

Imaging of head trauma Occasionally, the subacute SDH continues to enlarge with no significant reduction in density due to repeated haemorrhage from the friable vessels in the neomembranes (Figure 3.10). Subdural hygromas are hypodense fluid rims overlying the cerebral convexity. These can develop rapidly and resolve rapidly, from tears in arachnoid membranes with communication between the subarachnoid and subdural space. CSF-isodense rims developing soon after trauma are more likely to be hygromas rather than chronic SDHs. On MR, the signal intensity is similar to CSF on all imaging sequences (Figure 3.11). One pitfall on MRI is the high T2 signal rim around the cerebral convexities seen in intracranial hypotension which can be mistaken for chronic SDHs and subdural hygromas. Other features that support intracranial hypotension include brain sagging on sagittal imaging, fullness of the pituitary gland, engorgement of the major dural venous sinuses and smooth pachymeningeal (dural) enhancement on the outer wall of the ‘collection’ (Figure 3.12).

Extradural haemorrhage

Figure 3.9 Acute on chronic subdural haematoma with neomembranes. The hyperdense blood represents an acute rebleed into a pre-existing chronic subdural haematoma. There is a loculation of the acute blood by the neomembranes separating it from the hypodense chronic subdural haematoma.

Extradural haemorrhage (Figure 3.13) is usually associated skull vault fractures with tearing of the meningeal arteries. Occasionally, the haemorrhage is venous in origin, resulting from tearing of meningeal veins or tearing of the dural venous sinuses. The haemorrhage elevates the periosteal layer of the dura and hence does not cross sutural lines (Figure 3.14). However, it can cross the attachment of the dural ref lections, e.g. across the midline at the base of the falx cerebri (Figure 3.15). Extradural haemorrhage may be venous resulting from fracture extending across the major dural venous sinus.

Figure 3.10 Progressive enlargement of subdural haematoma. (a) Initial subdural haematoma was hyperdense. (b) Three weeks later, haematoma was hypodense and slightly increased in volume. (c) At 5.5 weeks, SDH is thicker and slightly increased in density. (d) At 8 weeks, SDH is increasing in density. (e) At 12 weeks, the SDH is swelling with more mass effect, necessitating surgical evacuation. The sequence of changes suggests repeated bleeding as the density should reduce with swelling of the haematoma.

Extradural haemorrhage

Figure 3.11 (a) Subdural effusions. CT scan on initial presentation shows subarachnoid haemorrhage in the left Sylvian fissure as curvilinear white streaks. Scalp soft tissue swelling over the right frontotemporal scalp. (b) CT scan 2 days later shows a rim of hypodense fluid overlying the left frontal lobe convexity. (c) CT scan 18 days later showed that the left frontal convexity subdural collection had resolved but a new thick rim of hypodense subdural fluid overlies the right frontal lobe convexity. Throughout the series of imaging, the subdural collections were always hypodense, almost isodense with CSF.

Figure 3.12 Intracranial hypotension. (a) CT shows bilateral thin rim of hypodense subdural fluid, mimicking subdural effusion/chronic haematoma. (b) Axial T2 shows bilateral smooth and thin rims of fluid. (c and d) Post-contrast axial and coronal T1 show smooth dural enhancement around cerebral convexities as well as in the tentorial leaves.

33

34

Imaging of head trauma

Figure 3.13 Acute extradural haematoma on CT. (a) Hyperdense biconvex lenticular-shaped extradural haematoma related to an underlying skull vault fracture seen as a lucent line on bone windows (b). Note the overlying subgaleal haematoma indicates external sign of local trauma here. There is a contralateral intra-axial haematoma in the right frontal lobe.

Figure 3.14 Left frontal extradural haematoma. Acute EDH is lenticular shape, the posterior margin respecting the coronal suture.

Subarachnoid haemorrhage and intraventricular haemorrhage

Figure 3.15 Frontal extradural haematoma crossing dural reflection. (a) CT shows an acute extradural haematoma crossing the midline. (b) CT bone windows show associated frontal bone fracture. CT venogram on (c) coronal and (d) sagittal reformats shows that the superior sagittal sinus is displaced by the extradural haematoma.

Beware of the ‘swirl sign’ within an acute haematoma which indicates active bleeding giving rise to the appearance (Figure 3.16). Anterior temporal extradural haematoma in the middle cranial fossa is recognised as a benign entity resulting from disruption of the sphenoparietal sinus. It is lenticular in shape, limited laterally by the sphenotemporal suture, medially by the orbital fissure and below the lesser sphenoid wing. These are usually of small volume and with low risk of significant mass effect to

cause uncal herniation and tend to remain stable or smaller on follow-up imaging.

Subarachnoid haemorrhage and intraventricular haemorrhage Traumatic subarachnoid haemorrhage is seen as hyperdense blood in sulcal and cisternal CSF spaces. This may arise from direct injury to small arteries and veins or extension from

35

36

Imaging of head trauma

Figure 3.16 ‘Swirl sign’ of active bleeding. The central component with higher density with some interspersed hypodensity indicates active bleeding giving rise to the ‘swirl’ appearance.

adjacent cerebral haemorrhagic contusions. More extensive subarachnoid haemorrhage in the basal cisterns may arise from direct arterial dissections of the supraclinoidal internal carotid (Figure 3.19) or the vertebrobasilar arteries. Caution must be exercised to suspect a spontaneous subarachnoid haemorrhage from an aneurysmal rupture as a prelude to the head injury, and if doubt, a CT angiogram can be helpful. On CT, acute subarachnoid blood is visible as hyperdensity in the sulcal and cisternal spaces (Figures 3.17 and 3.18). With time, the blood is diluted by CSF and reduces in density. Redistribution of blood with CSF flow can result in sulcal hyperdense blood at the vertex a few days later. Subtle traces of subarachnoid blood may be identified in dependent locations, including a hyperdense fluid level in the occipital horns or trigones, the posterior limit of the circular cistern, the ambient and quadrigeminal cisterns and at the level of the foramen magnum. MRI is less sensitive in detecting subarachnoid blood, depending on the stage of the blood products. FLAIR imaging is sensitive as blood products are seen as hyperintensity in the CSF spaces which are normally of low signal due to fluid signal suppression (Figure 3.20). Gradient-echo sequence may also detect blood as sulcal hypointensity (Figure 3.21). Traumatic intraventricular haemorrhage is uncommon. It may occur as a result of intraventricular extension of parenchymal bleed or reflux of subarachnoid blood from the cisternal and sulcal blood (Figure 3.22). As blood is relatively dense to CSF, intraventricular blood may gravitate dependently in the supine position (Figure 3.23).

Figure 3.17 Subarachnoid haemorrhage and intraventricular haemorrhage. Streaks of hyperdensity in the sulcal spaces between the gyri represent acute subarachnoid haemorrhage. The small focus of hyperdensity-dependent posteriorly in the right trigone is due to layering of intraventricular blood.

Figure 3.18 Subarachnoid haemorrhage. Sulcal subarachnoid blood can be very subtle, seen as hyperdense streaks in the right-sided cerebral sulci.

Subarachnoid haemorrhage and intraventricular haemorrhage

Figure 3.19 Traumatic subarachnoid haemorrhage. (a) Extensive hyperdense subarachnoid blood in basal cisterns. Large left ­subfrontal lobe haematoma suggesting a ruptured anterior communicating artery aneurysm, precipitating the head injury. (b) CT on bone windows shows extensive skull base fracture through the planum sphenoidale across the base of the anterior clinoid ­process, towards the cavernous sinus. (c) CT angiogram confirms a left cavernous carotid artery injury, the artery is displaced laterally. (d) A digital subtraction angiogram of the left internal carotid artery shows irregularity of the supraclinoidal internal carotid artery ­confirming the traumatic dissection.

In severe head trauma from road traffic accidents, shearstrain injuries may cause tears of the subependymal vein, fornix, septum pellucidum and choroid plexus. In a large series by Hashimoto et al., these were associated with more severe head injury and high mortality, 20 out of 32 with IVH died. The fornix

and septum pellucidum were weak points for shearing injuries (Figure 3.24). In extreme cases, the corpus callosum may be disrupted (Figure 3.22b). Patients with isolated intraventricular haemorrhage without other major associated injuries on CT tend to do well. It is

37

38

Imaging of head trauma

Figure 3.22 Intraventricular haemorrhage on CT and disruption of corpus callosum. These injuries were sustained from a headstomping injury. (a) Hyperdense acute intraventricular blood fills the trigones of the lateral ventricles, with frontal lobe haemorrhagic contusions. (b) Midline haematoma is due to disruption of the corpus callosum.

pressure changes induced by a head impact from the rupture of the choroid plexus or tela choroidea vessels or of vessels in the ventricular walls. Figure 3.20 Traumatic subarachnoid haemorrhage. On coronal FLAIR, sulcal hyperintensity in the right parietal lobe is due to subarachnoid blood, as normal sulcal CSF is dark on FLAIR due to fluid signal suppression.

Figure 3.21 MRI of subarachnoid blood. Streaky s­ ulcal subarachnoid haemorrhage in the cerebellar sulci seen as T2-hypointensity, exaggerated T2*-hypointensity, T1-hyperintensity and FLAIR-hypointensity.

Brain contusion injuries, haemorrhagic versus non-haemorrhagic Brain contusions occur at sites either at sites of direct impact against the skull vault and skull base or at contralateral sites as a contrecoup injury. Cortical injury results in focal area

Figure 3.23 Intraventricular haemorrhage on CT. A horizontal fluid-fluid level layers in the right lateral ventricle as dense blood gravitates in the CSF.

Brain contusion injuries, haemorrhagic versus non-haemorrhagic

Figure 3.24 Traumatic haemorrhage in septum pellucidum and ventricles. (a) Hyperdense foci of blood in the midline in the septum pellucidum. (b) Hyperdense blood in the trigones of both lateral ventricles and a trace of subarachnoid blood in left posterior temporal lobe. of localised oedema and swelling, with or without haemorrhagic change. Typically, these contusions damage the crests of the gyri resulting in wedge-shaped changes with a broader cortical base. The typical sites of contusions are in the anterior frontal and temporal poles, and the orbitofrontal and inferior temporal cortex where the brain impacts against the irregular bone surface of the skull base (Figures 3.25 and 3.26). Unusual sites of contusional injury should raise suspicion (Figure 3.27).

Figure 3.25 Contusions with haemorrhage. Focal areas of hypodensity in the medial subfrontal lobes bilaterally and in the left anterior temporal lobe with patchy areas of haemorrhage are typical appearances of traumatic contusional injuries in typical locations.

Figure 3.26 Traumatic haemorrhagic contusion in the left ­temporal lobe just above the petrous ridge.

Figure 3.27 ‘Typical’ and ‘atypical’ contusional injuries. (a) Typical areas of contusional brain injury in both subfrontal lobes and left anterior temporal lobe. (b) Right posterior frontal lobe ill-defined hypodensity was ascribed to traumatic head injury. (c) Axial T2 and (d) coronal FLAIR confirm an intrinsic glioma.

39

40

Imaging of head trauma rates of different tissues. Please note that the use of the term ‘diffuse axonal injury’ (or the acronym DAI) is very common in radiology practice as a synonym for dTAI. Neuropathologists prefer the use of dTAI as it is recognised that diffuse axonal injury as a pathological feature may have many causes (see Chapter 11). These stretching injuries occur at some common locations, in increasing order of severity:

Figure 3.28 Enlarging haemorrhagic contusion. (a) Initial CT scan shows haemorrhagic contusions in left frontal lobe and right cerebellum with SDH. (b) CT scan 7 days later showed that the haematoma in the left gyrus rectus is much larger. On CT, contusions appear as areas of cortical hypodensity involving the grey matter at the recognised sites. Punctate or macroscopic haemorrhagic foci and sulcal subarachnoid blood in the vicinity or tracking into the Virchow-Robin spaces as thin linear hyperdense streaks may be seen. These changes may not be evident on the initial scan but may develop over the next 24–48 hours as ensuing damage leads to tissue necrosis and haemorrhage, producing mass effect as these changes evolve and swell (Figure 3.28). MRI is more sensitive in detecting contusional injuries as areas of cortical T2- and FLAIR-hyperintensity, with or without swelling, the punctate haemorrhages appear as hypointense foci. Gradient-echo sequence detects areas of haemorrhage as low signal corresponding with the T2-hypointensity but as it is more sensitive than spin-echo T2W sequence, it may detect more areas of hypointensity.

Diffuse traumatic axonal injury Diffuse traumatic axonal injury (dTAI) arises from shear-strain injuries due to rotational acceleration of brain tissue resulting in axonal disruption as a result of the differential acceleration

1. at the grey-white matter junctions due to differential inertia of brain matter of different densities, usually due to acceleration in the sagittal plane (Figure 3.29). 2. at the corpus callosum due to acceleration in the coronal plane, the falx cerebri restricting movement of one hemisphere whilst unimpeded movement of the contralateral hemisphere results in stretching injury (Figure  3.29); the splenium is particularly vulnerable in view of the broader falx posteriorly. 3. at the dorsolateral midbrain due to stretching of the superior cerebellar peduncles exerted by the ­cerebellum. The shear-strain injury causes disruption to axons resulting in cell death. A total of 80 per cent of dTAI are non-haemorrhagic and 20 per cent present as punctate microhaemorrhages at the site of injury, the latter are sensitive biomarkers of dTAI as these may remain visible on MRI for years after the initial injury. CT is insensitive in detecting dTAI unless there is associated punctate microhaemorrhage at recognised sites of insult (Figure 3.30). MRI is far more sensitive in identifying dTAI. The acute disruption of axons results in cytotoxic oedema and may exhibit restricted diffusivity, with high signal on the diffusion trace and low signal on the ADC map (Figure 3.31). Standard T2W and FLAIR sequences are insensitive but may reveal punctate areas of high signal. Gradient-echo sequence and SWI are very sensitive in detecting the associated microhaemorrhage (Figure 3.32). As these areas of punctate microhaemorrhage may persist as haemosiderin staining as the blood products evolve, these are useful stigmata of previous diffuse axonal injury in late scans.

Figure 3.29 Tractography images to illustrate the sites of diffuse traumatic axonal injury from shear-strain stresses on fibre tracts. The left image shows the stress at the grey-white matter junction of the parasagittal lobes from anteroposterior movement, whilst the right image shows the stress points at the corpus callosum from side-to-side movements limited by the falx cerebri.

Diffuse traumatic axonal injury

Figure 3.30 dTAI on CT. The left image shows a focus of microhaemorrhage in the left dorsolateral midbrain, grade III dTAI. The right image shows a punctate focus of microhaemorrhage at the grey-white matter junction of the right superior frontal gyrus.

Figure 3.31 dTAI on diffusion-weighted imaging. (a) Diffuse trace shows high signal in the left side of the splenium of the corpus callosum and at the grey-white matter junction of the right superior frontal gyrus. (b) Corresponding images on the ADC map which shows that the lesions are dark, indicating diffusion restriction.

Figure 3.32 dTAI on gradient-echo sequence (T2* imaging). Punctate microhaemorrhages are identified at the left dorsolateral midbrain, right side of the splenium of the corpus callosum, and bilaterally at the grey-white matter junction of the superior frontal gyri.

41

42

Imaging of head trauma The  sensitivity of detecting these microhaemorrhages is increased with field strengths of static magnetic field of the MR scanners, a 3T scanner is more sensitive than a 1.5T scanner.

Secondary effects Herniation syndromes Brain swelling, intra- and extra-axial haematomas can result in significant brain herniation. The types of herniation are based on direction of the brain shift. Asymmetry between the two sides or effacement of normal CSF planes aid recognition. This topic is discussed in a separate chapter. The syndromes include the following: 1. 2. 3. 4. 5. 6.

Subfalcine herniation (Figure 3.33) Descending transtentorial herniation (Figure 3.34) Uncal herniation (Figure 3.34) Tonsillar herniation Ascending transtentorial herniation Extracranial herniation

Hypoxic-ischaemic brain injury Hypoxic-ischaemic brain injury may occur as a result of the initial failed resuscitation of an out-of-hospital cardiorespiratory arrest following the major trauma or as a result of consequent significant brain swelling resulting in reduced brain perfusion. This is different from hypoperfusion which may result in arterial watershed infarction. On CT, global hypoxic-ischaemic brain injury results in diffuse reduction in density of the brain with swelling (Figure 3.35). As the basal ganglia is metabolically most active, the initial findings may be very subtle with reduction in density of the deep grey matter with loss of definition of the caudate nucleus and lentiform nucleus, both of which are normally of higher density relative to the internal capsule. With increasing severity, the whole brain may become generally hypodense with loss of all grey-white matter differentiation, sulcal effacement and obliteration of the basal cisternal CSF spaces, and with collapsed ventricles. At the later stage, the cerebellum appears dense relative to the cerebral hemispheres, the so-called white

Figure 3.33 Left cerebral convexity subdural haemorrhage results in left to right subfalcine herniation and transtentorial herniation. A recognised complication is compression of the ipsilateral posterior cerebral artery against the tentorial hiatus resulting in infarction in the medial parieto-occipital lobes seen on the second image as hypodensity with loss of grey-white matter differentiation.

Figure 3.34 Duret haemorrhage. Significant mass effect resulting in descending transtentorial herniation may result in Duret haemorrhage in the upper brainstem, seen here as acute haemorrhage in the paramedian ventral midbrain. There is also dilatation of the temporal horns indicating hydrocephalus. There is also uncal herniation on the left, distorting the midbrain.

Figure 3.35 Hypoxic brain injury. A comparison between a normal and a hypoxic brain shows that there is generalised sulcal effacement, with complete loss of differentiation between grey and white matter, collapse of the lateral and third ventricles.

Secondary effects Watershed infarction is due to hypoperfusion at the border zones between arterial territories. These typically occur as parasagittal wedge-shaped hypodensities at the junction between the anterior and the middle cerebral artery supplies, the middle and posterior cerebral artery supplies or at the junction of all three.

Paediatric non-accidental head injury

Figure 3.36 White cerebellum and pseudo-SAH signs. The left image shows increased density of the cerebellum relative to the supratentorial brain where there is also loss of definition of the deep grey matter. The right image shows hyperdensity in the right Sylvian fissure mimicking subarachnoid haemorrhage but the rest of the visualised brain is rather featureless and hypodense. cerebellum sign or the reversal sign, which indicates irreversible brain damage and has a very poor prognosis. Pseudosubarachnoid haemorrhage sign also indicates severe hypoxia and a tight brain where the middle cerebral arteries appear hyperdense in the Sylvian fissures (Figure 3.36).

Multiple interhemispheric, convexity and posterior fossa haemorrhages (multicompartmental SDH) are associated with NAHI (Figure 3.37). Accompanying hypoxic-ischaemic injury and generalised cerebral oedema are also significantly associated with NAHI, while focal parenchymal injury is not a discriminatory feature (Figure 3.39). SDH of low attenuation is more common in NAHI than in non-NAHI. This may reflect the additional presence of traumatic subdural hygromas rather a component of chronic subdural haemorrhage. The commonest site for SDH is over the cerebral convexities with isolated convexity collections being the least discriminatory feature. SDH in the interhemispheric fissures and posterior and middle cranial fossa is otherwise only seen following severe accidental trauma (road traffic accidents) and coagulopathy. Paediatric spinal subdural haemorrhage is strongly associated with non-accidental injury rather than accidental injury (in the absence of direct spinal trauma). It is found in around 50 per cent of cases of non-accidental head injury (Figure 3.40).

Figure 3.37 Images taken from CT scan at presentation in a 5-month-old child with acute collapse and encephalopathy. Axial image (a) shows subdural haemorrhage overlying the right and left frontal lobes (close and open arrows) as well as in the falx (open arrowhead). Subarachnoid bleeding is also seen (solid arrowhead). The coronal image (b) demonstrates similar findings. A coronal view taken more posteriorly (c) to that shown in panel B demonstrates the right side subarachnoid bleeding (solid arrowhead) and left subdural bleeding (open arrow). It also shows bleeding in the right tentorium (open arrowhead). The sagittal image (d) shows this tentorial bleeding (open arrow) as well as bleeding in the posterior fossa (solid arrow), overlying the cerebellum.

43

44

Imaging of head trauma

Figure 3.38 Susceptibility weighted image taken from a child aged 3 months presenting with alleged choking and collapse whilst feeding. There was no reported history of traumatic injury of any kind. This MRI sequence shows blood as being black. Bleeding is seen in the subdural space overlying the right occipital lobe and in the posterior falx (solid arrows). This image also demonstrates multiple haemorrhagic clefts extending into the brain (white open arrows – see Chapter 15).

Figure 3.40 Sagittal T2-weighted image of the spine in the child presented in Figures 3.38 and 3.39 showing extensive bleeding (arrows) as black signal involving the posterior spinal thoracolumbar subdural space. (Figures 3.37–3.40: Courtesy of Professor S Stivaros.) acute setting, CT is the best imaging modality for initial assessment as it allows identification of acute blood, brain swelling and abnormal brain shift, as well as bone injury of the skull vault and the skull base. MRI has the benefit of better tissue differentiation due to the inherent natural tissue contrast resolution and is a problem-solving tool in situations unresolved by CT, e.g. in dTAI and in delayed complications.

Further reading

Figure 3.39 Diffusion-weighted image in the same child presented in Figure 3.38. This image shows extensive hypoxicischaemic injury as being bright signal, involving both the right and left sides of the brain (arrows). Spinal subdural bleeding is rare (and associated with very specific injury mechanisms) in accidental trauma. Intracranial subdural bleeding can track down into the spinal subdural compartment and can therefore account for spinal subdural bleeding in the absence of direct spinal trauma.

Conclusion Cross-sectional imaging is crucial in assessment of traumatic head injury. CT and MRI have complementary roles. In the

Atzema, C., Mower, W.R., Hoffman, J.R., Holmes, J.F., Killian, A.J. and Wolfson, A.B., National Emergency X-Radiology Utilization Study (NEXUS) II Group. 2006. Prevalence and prognosis of traumatic intraventricular haemorrhage in patients with blunt head trauma. J Trauma, 60, 1010–1017. Datta, S., Stoodley, N., Jayawant, S., Renowden, S. and Kemp, A. 2005. Neuroradiological aspects of subdural haemorrhages. Arch Dis Child, 90, 947–951. Gentry, L.R., Godersky, J.C. and Thompson, B. 1988. MR imaging of head trauma: review of distribution and radiopathologic features of traumatic lesions. Am J Roentgenol, 150, 663–672. Gean, A.D., Fischbein, N.J., Purcell, D.D., Aiken, A.H., Manley, G.T. and Stiver, S.I. 2010. Benign anterior temporal epidural haematoma: indolent lesion with a characteristic CT imaging appearance after blunt head trauma. Radiology, 257, 212–218. Hashimoto, T., Nakamura, N., Ke, R. and Ra, F. 1992. Traumatic intraventricular haemorrhage in severe head injury. No Shinkei Geka, 20, 209–215. Kemp, A.M., Jaspan, T. and Griffiths, J. et al. 2011. Neuroimaging: what neuroradiological features distinguish abusive from non-abusive head trauma? A systematic review. Arch Dis Child, 96, 1103–1112. Koumellis, P., McConachie, N.S. and Jaspan, T. 2009. Spinal subdural haematomas in children with non-accidental head injury. Arch Dis Child, 94, 216–219. LeRoux, P.D., Haglund, M.M., Newell, D.W., Grady, M.S. and Winn, H.R. 1992. Intraventricular hemorrhage in blunt head trauma: an analysis of 43 cases. Neurosurgery, 31, 678–684 (discussion 684–685). Parizel, P.M., Makkat, S., Van Miert, E., Van Goethem, J.W., van den Hauwe, L. and De Schepper, A.M. 2001. Intracranial haemorrhage: principles of CT and MRI interpretation. Eur Radiol, 11, 1770–1783. Silvera, S., Oppenheim, C., Touze, E., Ducreux, D., Page, P., Domigo, V., Mas, J.-L., Roux, F.-X., Fredy, D. and Meder, J.-F. 2005. Spontaneous intracerebral hematoma on diffusion-weighted images: influence of T2-shinethrough and T2-blackout effects. AJNR Am J Neuroradiol, 26, 236–241. Su, I.C., Wang, K.C., Huang, S.H., Li, C.H., Kuo, L.T., Lee, J.E., Tseng, H.M. and Tu, Y.K. 2010. Differential CT features of acute lentiform subdural hematoma and epidural hematoma. Clin Neurol Neurosurg, 112, 552–556.

4

Biomechanics of primary traumatic head injury Michael Jones

Introduction Establishing whether a traumatic head injury is a result of an accidental or non-accidental cause is a fundamental question in forensic investigations. Often practitioners are provided with only a brief third-party description of a causal event and struggle to establish a sufficiently detailed understanding of a cause and effect relationship with which to make a differentiation. Current medical understanding, acquired by training, anecdote and experience, is supplemented with scientific evidence and drawn specialities such as pathology, radiology and population-based epidemiology. The head and central nervous system may be injured by many different mechanisms; therefore, developing a necessary understanding of the cause from practical experience and epidemiology alone is a significant challenge, since there are very many biomechanical variables that require consideration. These include the magnitude of an applied force; the duration of the application of the force; for contact head injuries – the location of the application of force, whilst for non-contact head injuries – what body segment and through what linkages is the force applied and transferred; and what is the impacted or impacting surface’s material and mechanical response during impact. The applied force may produce deformation and/or acceleration (or deceleration), which may produce injury by compression – pushing materials together, tension – pulling materials apart or shear – sliding movable areas of material over other relatively immovable areas. It is noteworthy that during an injurious loading exposure (typically of a duration measured in milliseconds [tens of thousandths of a second]), deformation, acceleration, compression, tension and shear may be produced at the same time or in succession. Biomechanical engineers, particularly in automotive and aerospace safety research, have since the 1930s investigated the causal relationships between traumatic loading scenarios and injury, routinely addressing the same biomechanical variables that arise during forensic investigations. This uniquely equips them with the skills, approach and understanding required to address forensic characterisation of primary injury mechanisms. A retrospective biomechanical engineering analysis can assist a forensic investigation by providing cause and effect understanding with regard to a stated or inferred injury-causing event. This can be undertaken by characterising the biomechanical loading environment during the event in question, quantifying the physical loading environment and evaluating its potential to produce injury by, where possible, drawing comparisons with injury tolerance and/or epidemiological data.

Characterising the biomechanical loading environment Given the wide range of velocities, locations, angles and materials associated with head injury mechanics, it is unrealistic to anticipate that a single metric can exist for injury risk for every

given scenario. Specific to the head, one primary reason is the very many different motions that can occur when a head is struck with an object or when a head strikes a surface and/or whiplashed, since the complex variety of potential responses makes each injury-causing event potentially unique. General characterisation of the biomechanical loading environment can, however, assist in developing better approximations for a mechanism of injury in question. A simple schematic, provided in Figure 4.1, illustrates a typical approach taken by a biomechanical engineer when developing an understanding of the relationship between loading and head injury.

Mechanical loading environment Head injury can be caused by both contact loading and noncontact (inertial) loading, that is setting the head in motion. Both contact and inertial forces occur during impact loading, where the head is struck with (or strikes) a surface. Only inertial loading occurs in the absence of head contact. A critical common denominator in assessing the biomechanics of head injury can be found by defining the conditions that cause responses within the head and understanding the combination of external input conditions that produce the conditions. Head injury is associated with the tissues of the head being subjected to pressures and strains beyond their structural/functional limits, such that damage (injury) occurs. Pressure is the force per unit area (pressure, P = F/A is the force per unit area applied to a material, N/m 2), external to a material. Strain is defined as the deformation per unit length of a material, due to an applied stress (stress, s = F/A is the force per unit area within a material, N/m 2) (Figure 4.2). Strain is a dimensionless unit, since it is the ratio of the change in length to the original length (ΔL/L). If a material is in its unloaded state, its strain is 0. If a material is stretched by 100 per cent of its original length, the strain is 1 and if it is stretched by 50 per cent of its original length, the strain is 0.5. Contact head loading includes crush loading and impact loading. Crush loading (static or quasi-static loading) describes when the head is exposed to an applied force over a relatively long time period (in excess of 200 ms). Impact loading occurs when the head strikes a fixed object or is struck by a moving object over a relatively short time (durations less than 50 ms). Non-contact head loading (inertial loading) is produced by forces applied over relatively long periods to body segments, other than the head, being transmitted through its linkages (joints), including the neck to the head, such as whiplash or shaking.

Crush loading The skull has been demonstrated experimentally to withstand considerable compression in any direction without fracture and that a decrease in diameter in one direction is accomplished by an increase in diameter in other directions. Crush loading

46

Biomechanics of primary traumatic head injury

Figure 4.1 Relationship between the mechanical head loading environment, tissue response and injury.

Figure 4.2 Schematic of head injury loading scenario–related strain. Strain is given as a fractional change in either length (subject to a tensile stress) or volume (subject to bulk stress) or geometry (subject to shear stress). produces significant head shape change see Figure 4.3, and is typically associated with skull fractures, particularly basilar fractures (since the bones at the base of the skull are thicker and less flexible than the calvaria) and cranial nerve damage. Crush-related traumatic brain injury (TBI) is associated with very significant applied loads. It is relatively uncommon and when it does occur, its effects are often secondary to localised

Figure 4.3 Schematic of crush loading of the head: (a) before loading and (b) after loading. Solid lines show externally applied forces, and dashed lines show internally distributed reaction forces.

skull fracture. Crush loading is associated with very little movement, largely because the brain is incompressible and has a high bulk modulus between the head and loading surfaces as the external compressive pressure (force per unit area) produces internal stresses (force per unit area) and strains (change in length/original length) in the skull and brain tissues. Crush loading in infancy is often associated with bilateral skull fractures, localised about the site of loading, the mechanism is illustrated by the solid arrows, shown in Figure 4.3b. A published case series established that symmetrical bilateral skull fractures were associated with an infant’s head being compressed between two relatively firm unyielding surfaces (Hiss and Kahana 1995). Single impacts are commonly disregarded as a cause of bilateral fractures, the suspicion being that the cause is either crush loading or multiple impacts. However, a witnessed case of an infant, subjected to a single midline cranial impact, during an accidental short fall through 60–90 cm onto a concrete step, reported bilateral, parietal, highly symmetrical, linear skull fractures (Arnholz et al. 1998). Dynamic loading is very much more common and can be either predominantly translational (straight line) from a force applied perpendicular to the head, predominantly rotational as a result of an oblique (tangential) force, or angular due to a direct

Characterisation of head impact response – physical measures

Figure 4.4 Schematic of (a) translational, (b) rotational and (c) angular head accelerations, before loading (white), after loading (blue).

translational force imparting motion through the neck about a fixed point or an inertial mechanism applied to the torso acting through the neck, such as ‘whiplash’ (see Figure 4.4a–c). Although during an impact either mechanism, linear or angular, can predominate, it is unlikely that they can exist exclusively (other than where the contact line of force coincides with the centres of mass of the head and brain). The head (including the facial skeleton, skull, soft tissue and brain) has a centre of mass, which is different to that of the brain alone, such that any applied force will produce a different motion between them. When subjected to an impact, the head has a different point of anchorage and angulation (cervical spine of the neck) to the brain (brainstem), again any applied force will produce a different motion between them. Put simply, during impact, one cannot simply describe the motion of the head and brain as purely linear, purely rotational or purely angular; a combination will always be in evidence, even though one may invariably predominate.

Characterisation of head impact response – physical measures There are a range of variables produced during head loading that can be correlated with head injury risk, which can assist forensic investigation, including acceleration, kinetic energy, force, stress (force per unit area) and strain (change in length [deformation]).

Acceleration Impact-induced acceleration (deceleration) is the rate at which a body or object is brought to rest. In practical terms, it is dependent on the pre-impact velocity, the mechanical and material response of the impacting bodies, the time over which the body is brought to rest and the area over which the contact occurs. This can be illustrated by comparing the examples of a falling head impacting a soft, deformable, energy-absorbing surface (e.g. a mattress) with a falling head impacting a firm, unyielding, energy non-absorbing surface (e.g. concrete). A head falling onto a mattress is brought to rest over a relatively longer time period, compared with a firm surface, thus producing a low rate of change of velocity (acceleration). The kinetic energy (a product of the mass and velocity squared) is dissipated over a wide area of the head and is absorbed during the structural deformation of principally the mattress and some relatively minor deformation of the impacting head. A head falling through the same distance onto concrete, however, is brought to rest over a relatively short period, producing a high acceleration. Its energy is dissipated over a concentrated area of the head and is absorbed by producing structural deformation of

the head, since there will be a negligible structural deformation of the concrete.

Acceleration measures of head injury vulnerability and risk The relationship between acceleration and time has proven to be the most valuable predictor for head impact injury and head injury thresholds. Whilst linear, rotational and angular accelerations contribute to head impact kinematics and injury, see Figure 4.4, two acceleration components, linear and angular, are frequently isolated in injury studies, due to the considerable complexities associated with their interrelationship. It is vital, therefore, that the sensitivities of regional brain strain–related response to the kinematic variables of linear and angular accelerations be quantified, such that their relative contributions to overall brain injury can be understood. An understanding of the relationship between acceleration and time, and head injury has been developed by experimentation, principally by subjecting human volunteers, cadavers and animals to accelerations and evaluating their corresponding injury tolerances and injury patterns. Observed injuries have been correlated with acceleration, in terms of the kinematic (motion) variables from which head injury thresholds have been developed. Researchers at Wayne State University conducted human adult volunteer impact sled and cadaver drop testing, which has proved to be of profound significance in the development of head injury understanding (Gurdjian et al. 1966) – establishing a relationship between acceleration and human tolerance to structural damage and demonstrating that the severity of head injury is dependent on both the magnitude and duration of head acceleration. Relationships were established between a single point linear acceleration, time and injury severity, demonstrating the complexities associated with establishing cause and effect relationships. For example, an impact force of 100g (acceleration expressed in units of gravitational acceleration = 9.81 m/s2) over a duration of 6 ms represents the same injury risk as 200g over 3 ms, whilst an impact force of 75g for 10 ms represents a danger to life and 150g for 3 ms is considered tolerable. Although, in reality, the accelerations acting on the head rise and fall during the total duration of an impact, rather than producing a single peak value at any given duration, the study is valuable, both in defining the head impact tolerance levels and for developing injury criteria to estimate the different levels of head injury risk (Gurdjian et al. 1966). The practical application of the acceleration time relationship, established by the Wayne State and other studies, has necessitated the development of mathematical expressions that can provide a standardised method, capable of estimating the severity of a head injury established from external data. In the

47

Biomechanics of primary traumatic head injury

48

forensic context, peak g and the Head Injury Criterion (HIC) have been applied by biomechanical engineers in establishing cause and effect head injury relationships. Problems exist, however, with the peak g method, since it is somewhat insensitive to the corresponding effects of time. That time should be a consideration can be illustrated by reference to early attempts to mitigate motor vehicle occupant’s crash-related head injury risk. Historically, the most frequent fatal head injury, sustained during automotive collisions, was acute subdural haematoma (ASDH), due largely to high head accelerations produced during contacts with unpadded steel vehicle dashboards. To mitigate the high accelerations, padding was applied to the dashboards, and whilst the peak acceleration and ASDH reduced, the mortality rate remained approximately constant. The impacting head, rather than developing ASDH, developed diffuse axonal injury (DAI). The padded interior structure, whilst reducing the acceleration magnitude, increased the impact time duration. The head, therefore, rather than being subjected to localised deformation and reformation and the propagation of predominantly longitudinal waves in its bony shell structure, transferred the impact energy into propagating transverse waves within the brain tissues. That ASDH is associated with a high impact acceleration and short impact duration and DAI, relatively lower impact acceleration (though still above the threshold for DAI) and longer impact duration, therefore illustrates that in addition to peak loading, threshold levels of force and exposure time are critically important in assessing injury vulnerability. To assist in the application of the Wayne State data (Gurdjian et al. 1966), the HIC was developed to measure the injurious region of the linear acceleration-time curve, irrespective of the waveform shape, shown in the following equation: 2.5  t2    1   HIC = (t 2 − t1 )  adt   HIC equation.  (t 2 − t1 )  t1     MAX

∫



Whilst a is again the peak linear acceleration during impact, the time (t 2−t1) is more sensitively considered to include only the injurious time duration, either side of the peak accelerationtime waveform, rather than the total duration. A further series of experimental studies was conducted on adult cadavers, collating skull fracture and brain injury data with corresponding HIC values a specific injury risk curve developed, which was originally intended for use during production vehicle impact safety tests but now is used for other applications, including sports and forensics. During crash testing, regulated vehicle/barrier interactions, a crash test dummy head HIC value of 1000 (associated with a 16–18 per cent risk of severe or fatal head injury) is considered as the pass/fail threshold. HIC safety thresholds for the avoidance of head injury are in the region of 570, 723, 779, 700 and 670 for 3 year olds, 6 year olds, small females, midsized males and large males, respectively (Klinich et al. 2002) child dummies, a lower HIC value of 800 is applied and for infant dummies, 377, 390 and 440, corresponding to ages 6, 12 and 18 months, respectively (Mertz et al. 1997). These values have been immensely important in informing forensic investigations.

Translational kinematic – induced head injuries Translational (linear) acceleration may produce focal skull and/ or brain injuries, see Figure 4.5, due to compressive and tensile forces producing focal skull and brain tissue deformations, for example strains (changes in length). Diffuse effects are relatively uncommon, unless the linear accelerations are high. When the skull receives a focal impact, a momentary distortion of the shape is produced, the area under the point of impact bends inwards and since the brain is virtually incompressible, there is a compensatory bulging in other areas, see Figure 4.6. Bulging establishes stresses in the inner and outer tables, see Figure 4.6. Compression forces are produced at the peak of the concavities and tensile forces at the convexities. Since bone has a greater tolerance to compressive forces than tensile and

Figure 4.5 Schematic illustrating (a) a perpendicular impact direction and (b) a kinematic response to a perpendicular impact producing a translational, (straight line) force through the centre of gravity of the head, which may produce skull fracture, contusion (secondary to fracture), subdural haematoma (secondary to fracture) and epidural haematoma.

Impact-induced head injuries An impact to the head may produce significant inertial loading and translational (straight line) and/or angular (rotational) accelerations. Which kinematic predominates has a significant effect of the injury risk.

Figure 4.6 Arrows showing the direction of compressive (c) (solid lines) and tensile forces (t) (dashed lines) on bone during impact loading.

Impact-induced head injuries

Figure 4.7 Illustrating (a) struck loop and (b) impact against a firm unyielding surface. should the tensile forces exceed the elastic threshold, then bone passes into a plastic (non-elastic) phase before fracturing. The inner table will fracture where the skull is indented and the outer table will fracture at the margins of the deformed area. During an impact demonstrated at Figure 4.7(a), a region of inbending, generally circular, oval or star-shaped, is associated with the point of application of a blow, no matter where it is struck. Relevant to impacts shown in Figure 4.7(a) and (b), at areas where the skull curves sharply, the extent of the inbending is not as great as a less curved region. Outbending of bone may occur at a significant distance from the point of application of the blow and a contrecoup type of outbending is occasionally produced at a position approximately diagonally opposite to the point of impact (Gurdjian et al. 1950).

Skull fracture tolerance A number of scientific studies have focused on determining the tolerance of the human head to skull fracture. The complexity associated with human skull fracture tolerance and fracture ‘cause and effect’, can be illustrated by the study of Gurdjian et al. (1953) during which adult cadaver heads were impacted from heights between approximately 1 and 1.5 m at the middle of the occipital bone, onto a steel plate. Whilst individual heads were subjected to similar impact exposures, fracture patterns varied widely in both location and type, demonstrating linear, multiple linear and stellate fracture patterns. Furthermore, the fractures were located both at the site of the impact and distant from the impact and the author speculated that this may be a result of impact related inbending and outbending of the skull; thus, fracture location(s) may or may not directly correlate with impact site. Additional studies were performed with impacts at the posterior parietal or parietal-occipital locations and similar observations were made. Compared to adults, paediatric head impact cause and effect studies are relatively rare; however, those examining the occurrence and pattern of skull fractures in the paediatric population show some consistency with the adult study, demonstrating

that falls of approximately 1–1.5 m are sufficient to result in skull fractures and that fracture patterns are highly variable under similar conditions between different individuals (Weber 1984, 1985; Loyd 2011). In contrast to that of an adult, the skull of an infant is a relatively soft pliant structure, which when subjected to an applied force will deform readily. In infancy, this response has many advantages, since it allows the skull to deform during birth, so that the infant head can pass through the birth canal. The forces that the head is subjected to are applied over a long time period and are crushing force. The response of the infant’s head has been assumed to be the same when forces of short time durations are applied, such as an impact. However, research has indicated that this is not the case, and in fact, the infant skull material may actually stiffen when subjected to short duration forces. This phenomenon is believed to be due to the fact that many biological tissues, rather than being purely elastic in nature, that is responding to an applied force by instantly by deforming (stress [F/A] and strain [∆L/L] being in phase, changing stress effects strain immediately), the head responds in a viscoelastic manner, that is exhibiting both viscous properties (strain [∆L/L] lags stress [F/A] by 90 degrees) and elastic properties. Therefore, viscoelastic materials exhibit behaviour with a phase difference in between that of elastic and viscous materials, that is, exhibiting some time lag in the development of strain. Thus, the impact response and tolerance of viscoelastic biological materials is rate sensitive. Adult head impact experiments have established that relatively high contact forces and large linear accelerations increase stress (F/A) in the skull bone and, therefore, the risk of skull fracture. Relative fracture tolerances have been demonstrated between regions of the skull, for example a study by Mertz et al. (1997) estimated a 5 per cent risk of skull fractures at a peak acceleration of 180g and a 40 per cent risk of fractures at 250g. Frontal bone fractures are associated with forces of 4.8–5.8 kN and temporoparietal fractures 3.5–3.6 kN (Nahum et al. 1968; Allsop et al. 1988; Schneider and Nahum 1972).

Contusions (secondary to skull fracture) Cerebral contusions at the site of impact in the presence of skull fracture are likely produced by the direct transference of the energy released by the fracturing skull to the underlying meninges and brain tissue and, therefore, as for skull fracture are well predicted by linear acceleration.

Subdural haematoma (secondary to fracture) Similarly, subdural haematoma (SDH) can be produced by the direct release by fracture of energy stored within the deforming skull to adjacent areas, including the underlying meninges.

Acute subdural haematoma (with or without skull fracture) ASDH can occur from rotational accelerations produced during translational impact, angular impact or purely inertial accelerations, i.e., whiplash/shaking or a combination. The primary cause of this injury is the differential acceleration between the skull and the brain producing gliding contusions.

Epidural haematoma Impact related epidural haematoma is commonly associated with an impact to the temporoparietal region, although related

49

50

Biomechanics of primary traumatic head injury to translational kinematics, it is relatively rare, occurring in only 0.2–6 per cent of cases (Kleiven 2003).

Angular kinematic – induced head injuries That the human brain is sensitive to rotational motion was first suggested by Holbourn (1943), who observed shear strain patterns in gel models and claimed that translational loading is not injurious, while rotation could explain the majority of traumatic brain injuries, due to the nearly incompressible properties of brain tissue. This can be explained by the fact that the bulk modulus of the brain (applied pressure/ fractional change in volume) is approximately six orders of magnitude greater than the shear modulus (shear stress/ shear strain), such that for a given impact, the brain tends to deform primarily in shear (McElhaney et al. 1976). This produces a large strain sensitivity of the brain to rotational loading and a small sensitivity to linear kinematics and, therefore, a better indicator of TBI risk (Kleiven 2006). Also, it has been shown that the most common severe injuries, such as subdural haemorrhage and diffuse traumatic axonal injury (dTAI), are more easily caused by rotational head motion (Gennarelli et al. 1972, 1987). Gurdjian (1975), however, suggested that a combination of skull deformation, pressures and inertial brain lag could present a clearer picture of head injury. As illustrated in Figure 4.8a,b, oblique impacts produce both linear and rotational head kinematics. Gennarelli et al. (1982) stated that all types of brain injury can be produced by angular acceleration, whilst Ommaya (1985) commented that rotation can produce both focal and diffuse brain injuries, while linear translation is limited to focal effects. Rotational and angular velocities and accelerations (impact and inertial) commonly produce more widespread and diffuse forces as the skull and brain experience differential accelerations. Angular kinematics produce compound translational and rotational kinematics. Significant angular velocities (angular velocity ω [rad/s] = v/r) and accelerations (angular acceleration α [rad/s2] = Δω/Δt) can be produced during oblique, tangential, impacts to the head and/or as a result of forces transferred through the neck setting the head into angular motion.

That angular acceleration is associated with more diffuse and severe shear and tension stress (internal force/area)-related injury patterns and can be understood by considering Newton’s first law. During an oblique (tangential) impact, the impacted area of the skull rotates around the axis of the cervical spine. The skull and brain have an angular inertia (property of matter causing it to resist changes in position) such that when the head is subjected to an external force, the relatively low mass skull produces less resistance to movement than the relatively high mass brain. The skull, therefore, moves relatively quickly compared to the brain, which is lagging behind and the differential movement produces shear forces at the boundary between the two bodies. Furthermore, the heterogeneity of the brain can also predispose those areas of different densities (mass per unit volume) to differential accelerations, for example the interfaces between grey and white matter (in shear, grey matter cell bodies demonstrate an isotropic response and white matter fibre bundles an anisotropic response). Axons subjected to shearing forces (a force applied perpendicular to an opposing offset force) may be subject to shear strain and tension. What dictates the severity of the injuries is the magnitude of the angular accelerations and the duration of the exposure time.

Concussion Cerebral concussion is the most common head injury, accounting for around 70 per cent of the all head injury hospital admissions (Kleiven 2003). Experimentally, primates subjected to controlled translational head motions (straight line from point of impact through the centre of gravity of the head) did not produce cerebral concussion. In contrast, those animals subjected to head rotations were all concussed (Gennarelli et al. 1972). Greater frequency and severity of rotation produced brain lesions in both translational and rotational groups. Angular accelerations above 4500 rad/s2 and angular velocities above 33 rad/s have been reported to produced concussions involving loss of consciousness lasting longer than 1 minute in contact field sports (Patton et al. 2012), and accelerations and velocities of 6383 rad/s2 and 28.3 rad/s, respectively, have been proposed as thresholds for a 50 per cent risk of concussion (Rowson et al. 2012).

Cerebral contusions Cerebral contusion, one of the most frequently found lesions following head injury, consists of heterogeneous areas of necrosis, pulping, infarction, haemorrhage and oedema (Melvin et al. 1993). Contusions generally occur at the site of impact (coup contusions), see Figure 4.9a, and at remote sites from the impact (contrecoup contusions), see Figure 4.9b. In the absence of skull fracture, contusions are believed to be induced by compression, tension, shearing or cavitation of the brain tissue as a result of excessive head rotational loading (Löwenhielm 1975). Moreover, Shreiber et al. (1997) derived a threshold of 0.19 in strain (principal logarithmic) in the cortex for a 50 per cent risk of cerebral contusions induced by vacuum. As previously discussed, this strain is sensitive only to the rotational kinematics and not the translational motion (Ueno and Melvin 1995).

Effect of cavitation Figure 4.8 Illustrations of how (a) an oblique impact ­direction produces (b) a translational (straight line) and rotational kinematic response, which may produce concussion, diffuse traumatic axonal injury, contusion, subdural haematoma and intracerebral haematoma.

Associated with contusions, cavitation is primarily caused by high impulse loading propagating shock waves through the head, see Figure 4.9a,b. A decoupling between the brain and deforming skull occurs and a negative pressure in the Cerebro Spinal Fluid (CSF) produced. Small bubbles precipitate out of the CSF;

Angular kinematic – induced head injuries

Figure 4.9 Schematic representation of the biomechanics of impact-induced (a) coup contusion, (b) contra coup contusion (c) bridging vein shear–related subdural haematoma, concussions, contusions, intracerebral haematomas and (d) tissue shear–related diffuse traumatic axonal injuries. when the environmental pressure returns to normal the bubbles burst, releasing energy, such that injurious levels of strain are produced in the cerebral cortex and periventricular tissues.

Subdural haematoma The most common mechanism of SDH is tearing of veins that traverse the subdural space to the dural sinuses, see Figure 4.9c (Gennarelli and Thibault 1982). Threshold values for angular acceleration–induced gliding contusions were established by tensile testing of human bridging veins rupture (4500 rad/s2, Löwenhielm 1974a) and human cadaver impact testing (10,000 rad/s2, Depreitere et al. 2006). Primate experiments suggested that SDH was produced by short duration and high amplitude of angular accelerations Gennarelli (1983). Lee and Haut (1989) studied the effects of strain rate on tensile failure properties of human bridging veins and determined the ultimate strain to be about 0.5, which was found to be independent of the strain rate (ε = 0.1–250 s−1), whilst an earlier study by Löwenhielm (1974b) suggested strain rate sensitivity by showing that failure strain reduced from approximately 0.8–0.2 as the rate increased. Lee et al. (1987) and Huang et al. (1999) found that the contribution of angular acceleration to tearing of bridging veins was greater than translational acceleration. Substantially larger relative motions between the skull and the brain and higher strain in the bridging veins were produced when switching from a translational to a rotational mode of motion.

Intracerebral haematomas Intracerebral haematomas (ICH), homogeneous collections of blood within the cerebral parenchyma, are associated with strains of 0.4–0.5, which has been established experimentally to be in the region of the thresholds for rupture of cerebral veins and arteries, indicating that the risk of ICH can be predicted by the pattern and magnitude of peak strain.

Diffuse traumatic axonal injury dTAI is associated with mechanical disruption (shear strain) of many axons in the cerebral hemispheres and subcortical white matter and is, from a biomechanical engineering perspective, typically associated with differential movement of adjacent regions of the brain during acceleration events. Brain tissue studies have suggested that different variables of tissue deformation, such as maximum principal strain, maximum shear strain, shear stress and pressure, are capable of predicting injurious outcomes. However, because brain is a viscoelastic material, tolerance criteria may also depend on the rate of deformation, or strain rate. The maximum strain tolerance, suggested as threshold, for dTAI is estimated at 0.3. A brain strain threshold of 0.2 (maximal principal strain) has been suggested, for the onset of neuronal malfunction, considered a first stage for dTAI. Maximum principal strains of 0.2 (Green-Lagrange strain) have also been shown to correlate with cell death and neuronal dysfunction associated with DAI (Morrison et al. 2003).

Inertial loading, shaking and whiplash High-energy inertial loading–related injuries in adults, children and infants (shown in Figure 4.9a–d), for example automotive crash scenarios, are associated with significant whiplash loading (mechanics of inertial loading, shown in Figure 4.10). There is, therefore, general agreement that inertial loading (whiplash) can produce the diffuse distribution of ASDH, commonly associated with a shaking scenario. What is not agreed is that human-induced manual shaking alone, in the absence of an associated impact, can produce sufficient loading to cause ASDH (Duhaime et al. 1987; Prange et al. 2003). Indeed, Duhaime concluded ‘that severe head injuries commonly diagnosed as shaking injuries require impact to occur and that shaking

51

52

Biomechanics of primary traumatic head injury

Figure 4.10 Mechanics of inertial head loading (‘whiplash’ or ‘shaking’), heavy arrows show torso movement (acceleration). (a) extension of the arms (b) arms fully contracted or extended, (c) contraction of the arms, light arrows show the compensatory angular movement (acceleration) of the head. alone in an otherwise normal baby is unlikely to cause the shaken baby syndrome’. A number of biomechanical engineering analyses concur that shaking alone is incapable of producing the loading (acceleration, velocity, force etc.) required to exceed the SDH thresholds established from adult primate data scaled (for mass and geometry) up to human adults and subsequently scaled down to human infants (Gennarelli et  al. 1982). The injury thresholds have in many instances been applied without question, yet, how relevant the automotive safety thresholds are to the investigation of head injury in the living infant remains largely unknown.

Infant vulnerability to inertial loading Compared to an adult, the infant head has larger extracerebral spaces, greater laxity of the meninges and inertial differences between myelinated and unmyelinated, differentiated and undifferentiated, white and grey brain tissue, potentially increasing their sensitivity to brain injury from linear and angular acceleration–induced rotational accelerations. These factors are believed to contribute to a vulnerability to shearing of bridging veins between the meninges and brain and damage to axons. Injury tolerance limits, which originate from primate studies, have been scaled up (by mass and geometry) to represent injury limits for adults and down (by the same process) to children and infants to establish tolerance limits for rotational accelerations and rotational velocities associated with the production of cerebral concussion, DAI and SDH.

Characterisation of shaking Angular kinematics Several infant-specific mechanical and computational models have been developed in an attempt to simulate the biomechanics of human-induced shaking of an infant. Mechanical anthropomorphic test device studies have shown that during shaking, adults are incapable of producing angular accelerations and velocities that exceed the injury thresholds currently attributed to the production of ASDH 10, 11, 12 & 13. Thoracic shaking frequencies range between 2.5 and 5 cycles per second (Hz) across all studies, which have produced maximum translational head

accelerations, rotational head accelerations and rotational head velocities of 1737 m/s2, 10,216 rad/s2 and 61 rad/s, respectively, see Figure 4.11. No study, to date, has demonstrated that shaking alone, without an associated impact, exceeds the injury thresholds associated with SDH. The majority of the results were found to be clustered below all predicted infant-specific injury thresholds (rotational head accelerations of ∼1000 rad/s2 and rotational head velocities of ∼20 rad/s), see Figure 4.11.

Linear kinematics Similar to angular kinematics, linear kinematics have typically produced upper peak head accelerations values in the range of 10–15g, and since a lower threshold for injury is commonly considered to be in the 50–60g range, shaking has been dismissed as a likely cause of acute brain injury. Head impact, however, has been established experimentally to produce levels of acceleration in excess of 50–60g. If the impact hypothesis is accepted, there may be an alternative explanation. Cory and Jones (2003), whilst conducting infant dummy shaking experiments and producing similar 10–15g peak linear accelerations, during the excursion of the head backwards and forwards, observed significant periodic end point impulses (above 50g), between areas of the dummy corresponding to the chin and the chest and the back of the head (occipital region) and upper back (at around the C-7 level). Such that when the head injury measure – HIC – was applied to assess the acceleration-time impulses, 70 per cent exceeded impact parameters for serious head injury, including subdural haemorrhages (critical load value, HIC = 800, Stürtz 1980), and all impacts exceeded the HIC thresholds of 377–440 for infants (Mertz et al. 1997).

Conclusion The study of the mechanics of head injury began at a time when measurement technologies were in their infancy and the only practically measurable parameter was linear acceleration, which became the benchmark for quantifying and correlating head impact severity. Whilst angular acceleration was proposed as a competing brain injury parameter, technological measurement

References

Figure 4.11 Rotational accelerations and velocities across all simulations ×, showing the scaled injury tolerance levels for a 1 month old: 50 per cent concussion likelihood tolerance level (Ommaya 1985 and Klinich et al. 1996). - - - - Cerebral concussion tolerance level (Duhaime et al. 1987). -. -. - Cerebral concussion tolerance level (Thibault and Margulies 1998). ………. Subdural haematoma tolerance level (Duhaime et al. 1987). _______

limitations curbed its mainstream consideration. The two competing acceleration-based theories resulted in the flawed perception that head injury is either a consequence of purely linear acceleration or purely angular acceleration. Regardless of the fact that neither can exist exclusively during real-life traumatic head exposures, researchers generally pursued either a linear acceleration or angular acceleration approach. Biomechanical engineers working in the forensic environment have routinely relied on the heavily funded automotive safety research injury standards and thresholds. The HIC, which is based on linear acceleration, has been largely vindicated by the significant reduction in the number of automotive-related head injuries sustained over the last 50 years. However, it continues to draw criticism from the safety engineering community for conspicuously failing to take into account those injuries that result from the angular acceleration vulnerability of the human head. In response, research into the effects of angular acceleration has received greater emphasis than linear acceleration in an attempt to establish tolerance limits for angular acceleration exposure. The fact that researchers have continued to focus on established acceleration-based research has resulted in a delay in investigations into other more difficult to measure parameters that may provide a more detailed understanding of the causes of brain injury. Increasingly, and largely as a result of an investment in contact sport–related TBI research, however, other head injury measures and tolerances are being investigated. These include the combined linear and translational acceleration, tensile, compressive and shear stresses and strains, strain rates and cavitation. Researchers are currently investigating head injury across the full spectrum of length scales, from nanoscale to macroscale. Localised, regional and global tissue response measurements are increasingly being applied to physical and computational modelling methods for head injury cause and effect investigations across a range of different loading scenarios. In parallel, biomechanical engineering–led epidemiological studies are defining the real-world scenarios, during which head injury does and does not occur. These surveillance data

provide critical input for the computational tools that will continue to improve the resolution and accuracy of simulations measuring tissue responses within specific anatomic regions. With new techniques to map these deformations, a wealth of information from nano- and microscale studies will merge with information from macroscale studies to provide increased insights into head injury cause and effect.

References Allsop, D.L., Warner, C.Y., Wille, M.G., Schneider, D.C. and Nahum, A.M. 1988. Facial impact response—a comparison of the hybrid III dummy and human cadaver. SAE Trans, 1224–1240. Arnholz, D., Hymel, K.P., Hay, T.C. and Jenny, C. 1998. Bilateral pediatric skull fractures: accident or abuse? J Trauma Acute Care Surg, 45(1), 172–174. Cory, C.Z. and Jones, M.D. 2003. Can shaking cause fatal head injury? Med Sci Law, 43(4), 317–333. Depreitere, B., Van Lierde, C., Vander Sloten, J., Van Audekercke, R., Van Der Perre, G., Plets, C. and Goffin, J. 2006. Mechanics of acute subdural hematomas resulting from bridging vein rupture. J Neurosurg, 104(6), 950–956. Duhaime, A.C., Gennarelli, T.A., Thibault, L.E., Bruce, D.A., Margulies, S.S. and Wiser, R. 1987. The shaken baby syndrome: a clinical, pathological, and biomechanical study. J Neurosurg, 66(3), 409–415. Gennarelli, T.A., Thibault, L.E. and Ommaya, A.K. 1972. Pathophysiologic responses to rotational and translational accelerations of the head (No. 720970). SAE Technical Paper. Gennarelli, T.A. and Thibault, L.E. 1982. Biomechanics of acute subdural hematoma. J Trauma, 22(8), 680–686. Gennarelli, T.A., Thibault, L.E., Adams, J.H., Graham, D.I., Thompson, C.J. and Marcincin, R.P. 1982. Diffuse axonal injury and traumatic coma in the primate. Ann Neurol, 12(6), 564–574. Gennarelli, T.A. 1983. Head injury in man and experimental animals: clinical aspects. In Trauma and Regeneration. Springer, 1–13. Gennarelli, T.A., Thibault, L.E., Tomei, G., Wiser, R., Graham, D. and Adams, J. 1987. Directional dependence of axonal brain injury due to centroidal and non-centroidal acceleration. SAE Trans, 1355–1359. Gurdjian, E.S., Webster, J.E. and Lissner, H.R. 1950. The mechanism of skull fracture. Radiology, 54(3), 313–339. Gurdjian, E.S., Webstef, J.E. and Lissner, H.R. 1953. Observations on prediction of fracture site in head injury. Radiology, 60(2), 226–235.

53

54

Biomechanics of primary traumatic head injury Gurdjian, E.S., Roberts, V.L. and Thomas, L.M. 1966. Tolerance curves of acceleration and intracranial pressure and protective index in experimental head injury. J Trauma, 6, 600–604. Gurdjian, E.S. 1975. Re-evaluation of the biomechanics of blunt impact injury of the head. Surg, Gynecol Obstet, 140(6), 845–850. Hiss, J. and Kahana, T. 1995. The medicolegal implications of bilateral cranial fractures in infants. J Trauma Acute Care Surg, 38(1), 32–34. Holbourn, A.H.S. 1943. Mechanics of head injuries. Lancet, 242(6267), 438–441. Huang, H.M., Lee, M.C., Chiu, W.T., Chen, C.T. and Lee, S.Y. 1999. Threedimensional finite element analysis of subdural hematoma. J Trauma Acute Care Surg, 47(3), 538–544. Kleiven, S., 2003. Influence of impact direction on the human head in prediction of subdural hematoma. J Neurotrauma, 20(4), 365–379. Kleiven, S., 2006. Evaluation of head injury criteria using a finite element model validated against experiments on localized brain motion, intracerebral acceleration, and intracranial pressure. Int J Crashworthiness, 11(1), 65–79. Klinich, K.D., Saul, R.A., Auguste, G., Backaitis, S. and Kleinberger, M. 1996. Techniques for developing child dummy protection reference values. NHTSA Event Report, Docket Submission No. 74-14. Klinich, K.D., Hulbert, G.M. and Schneider, L.W. 2002. Estimating infant head injury criteria and impact response using crash reconstruction and finite element modeling (No. 2002-22-0009). SAE Technical Paper. Lee, M., Melvin, J. and Ueno, K. 1987. Finite element analysis of subdural hematoma. In: Proceedings of the 31st Stapp Car Crash Conference, pp 6777. Lee, M.C. and Haut, R.C. 1989. Insensitivity of tensile failure properties of human bridging veins to strain rate: implications in biomechanics of subdural hematoma. J Biomech, 22(6–7), 537–542. Löwenhielm, P. 1974a. Strain tolerance of the vv. Cerebri sup. (Bridging veins) calculated from head-on collision tests with cadavers. Z Rechtsmed, 75(2), 131–144. Löwenhielm, P. 1974b. Dynamic properties of the parasagittal bridging beins. Z Rechtsmed, 74(1), 55–62. Löwenhielm, P. 1975. Mathematical simulation of gliding contusions. J Biomech, 8(6), 351–356. Loyd, A.M. 2011. Studies of the human head from neonate to adult: an inertial, geometrical and structural analysis with comparisons to the ATD head (Doctoral dissertation). McElhaney, J.H., Roberts, V.L. and Hilyard, J.F. 1976. Handbook of Human Tolerance. Japan Automobile Research Institute, Incorporated (JARI).

9781498706162_Ch04.indd 54

Melvin, J.W., Lighthall, J.W. and Ueno, K. 1993. Brain injury biomechanics. In Nahum, A.M. and Melvin J.W. (eds), Accidental Injury. Springer-Verlag, pp 269–290. Mertz, H.J., Prasad, P. and Nusholtz, G. 1996. Head injury risk assessment for forehead impacts. SAE Trans, 26–46. Mertz, H.J., Prasad, P. and Irwin, A.L. 1997. Injury risk curves for children and adults in frontal and rear collisions. SAE Trans, 3563–3580. Morrison, B.C. III, Cater, H.L., Wang, C.C.-B., Thomas, F.C., Hung, C.T. and Ateshian, G.A., et al. 2003. A tissue level tolerance criterion for living brain developed with an in vitro model of traumatic mechanical loading. Stapp Car Crash J, 47, 93–105. Nahum, A.M., Gatts, J.D., Gadd, C.W. and Danforth, J. 1968. Impact tolerance of the skull and face (No. 680785). SAE Technical Paper. Ommaya, A.K. 1985. Biomechanics of trauma. Appleton-Century-Crofts, Eat Norwalk, CT, Biomechanics of Head Injuries: Experimental Aspects. Patton, D.A., McIntosh, A.S., Kleiven, S. and Frechede, B. 2012. Injury data from unhelmeted football head impacts evaluated against critical strain tolerance curves. Proc Inst Mech Eng, P, 226(3–4), 177–184. Prange, M.T., Coats, B., Duhaime, A.C. and Margulies, S.S. 2003. Anthropomorphic simulations of falls, shakes, and inflicted impacts in infants. J Neurosurg, 99(1), 143–150. Rowson, S., Duma, S.M., Beckwith, J.G., Chu, J.J., Greenwald, R.M., Crisco, J.J., Brolinson, P.G., Duhaime, A.C., McAllister, T.W. and Maerlender, A.C. 2012. Rotational head kinematics in football impacts: an injury risk function for concussion. Ann Biomed Eng, 40(1), 1–13. Schneider, D.C. and Nahum, A.M. 1972. Impact studies of facial bones and skull (No. 720965). SAE Technical Paper. Shreiber, D.I., Bain, A.C. and Meaney, D.F. 1997. In vivo thresholds for mechanical injury to the blood-brain barrier. SAE Trans, 3792–3806. Stürtz, G., 1980. Biomechanical data of children. In: 24th Stapp Car Crash Conf. Proc., SAE. Paper No. 801313. Thibault, K.L. and Margulies, S.S. 1998. Age-dependent material properties of the porcine cerebrum: effect on pediatric inertial head injury criteria. J Biomech, 31(12), 1119–1126. Ueno, K. and Melvin, J.W. 1995. Finite element model study of head impact based on hybrid III head acceleration: the effects of r­otational and translational acceleration. J Biomech Eng, 117(3), 319–328. Weber, W. 1984. Experimental studies of skull fractures in infants. Z Rechtsmed, 92(2), 87–94. Weber, W. 1985. Biomechanical fragility of the infant skull. Z Rechtsmed, 94(2), 93–101.

08/03/21 1:42 PM

5

Techniques Helen Whitwell

This chapter covers in outline the approach to post-mortem examination with emphasis on the neuropathological aspects. It is not intended to be a fully comprehensive, detailed guide but concentrates rather on specific points of note from a practical aspect in forensic neuropathological cases. More details of the post-mortem examination in a case of suspected paediatric non-accidental injury are covered in Chapter 15.

Before the post-mortem examination In the United Kingdom and in many other jurisdictions, postmortem examinations on head injury or other forensic neuropathological cases will be at the request of the legal authorities. It should be good practice that this authority is confirmed in writing. In practice, this will usually be a fax/email from the relevant office.

General post-mortem examination A full post-mortem examination should be undertaken in all forensic cases. The joint Royal College of Pathologists (United Kingdom) and Forensic Science Regulator (England and Wales) published Code of Practice and Performance Standards for forensic pathologists (2018) covers forensic cases in England and Wales. There is a separate code for forensic pathologists in Scotland (Royal College of Pathologists 2016a). Neuropathology autopsy practice: post-mortem examination in patients with traumatic brain injury (Royal College of Pathologists 2010) also provides useful guidelines specifically relating to neuropathology. Standard texts include Waters (2010), Dawson and Neal (2003) and Esiri (2006).

Neuropathological examination In many cases of head trauma, there will be hospital records including radiological results. It is essential these are reviewed prior to the examination, if possible in conjunction with the neuroradiologist. Particular attention should be paid to any medical procedures, including neurosurgery, that may have been undertaken. This avoids later confusion about whether or not an injury is genuine (Figure 5.1), e.g. insertion of a ventricular shunt or an intracranial pressure monitoring device (see also Chapter 18). Apart from the above, details of any known history of the circumstances relating to the case are essential. It is, however, recognised that in the initial stages of any investigation these may be scanty and further review may be necessary at a later date, sometimes at a considerable time distance. A statement to this effect should be added to the post-mortem report – review may amend/alter the initial opinions. Radiological examination is essential in some adult cases, in particular gunshot wounds, burnt bodies and decomposed bodies, and where there is extensive head/facial trauma. Post-mortem angiography may be indicated in certain circumstances – (see Chapter 9). CT-based post-mortem examinations are becoming increasingly established. Whilst there is

a potential role for avoiding a post-mortem examination in sudden natural death (Rutty et al. 2017), it is not usually acceptable on its own in suspicious deaths (Royal College of Pathologists 2016b).

Recording information Notes taken by the pathologist, including handwritten ones, and any diagrams, as well as tapes if dictation is used, should be retained as evidence. Photographs are essential in any case where criminal proceedings may take place. Diagrams of any injuries, together with depiction of any fractures, can also be undertaken. It may be necessary to produce body plans for court use – this is often preferred for juries rather than photographs. Computer technology has aided in the production of such plans. Other investigations relating to the head before examination depend on the circumstances but may include the following: • Cerebrospinal fluid: microbiology • Collection of hair: for toxicology, this should include roots if possible and be of pencil thickness from the crown – see also Chapter 19. Hair may also be taken for forensic analysis relating to the scene or potential weapon.

External examination Part of the interpretation of any head injury case is the evaluation of external wounds. This is covered in detail in Chapter 6.

Internal examination Subscalp bruising/haematomas Following reflection of the scalp, all bruises should be documented with their location and size recorded, including involvement of muscles, e.g. temporalis muscle. Description should include an assessment of the age, i.e. recent, days. Over-dogmatic interpretation about age should not be done because it is well recognised that only estimates can be given. In elderly people, bruising can persist for many weeks/months. Re-examination of the external aspect may be necessary to coordinate any findings. Scalp bruising may be much more extensive than the injuries noted externally. Kicking tends to produce marked deep bruising (Figure 5.2). Neurosurgical procedures may mask or cause bruising or soft tissue haemorrhage (Figure 5.3). Artefactual bruising may occur over the back of the head when an individual has been lying in a prolonged comatose state. Scalp petechiae are well recognised as an artefact in many situations and should not be taken as a sign of asphyxia (Figure 5.4).

Removal of the skull This can be done either with an electric or hand saw. The latter method is indicated in suspected cases of Creutzfeldt-Jakob disease. It also tends (in experienced hands) to produce less in

56

Techniques

Figure 5.3 Bone flap with soft tissue haemorrhage caused by neurosurgery.

Figure 5.1 Bruising in association with a craniotomy wound. the way of damage to the underlying dura/brain. Care should be taken to avoid artefactual skull fracturing. Skull fractures are covered in Chapter 7. The dura should always be stripped. In addition, examination of the facial structures to assess bruising/fractures should be undertaken. Bruising to the jaw may be evident only on deep dissection, e.g. after a punch or slap to the face.

Immediate neuropathological examination Immediate inspection on removal of the skull will assess the presence of intracranial haematomas – extradural and subdural – as well as subarachnoid haemorrhage (Figure 5.5). In the latter condition, if traumatic subarachnoid haemorrhage is suspected as a result of the history and circumstances, the appearance of the brain may alert to the presence of this, with subarachnoid haemorrhage being just visible over the lower hemispheres (Figure 5.6). This is covered in detail in Chapter 9. Other conditions including meningitis and brain swelling may also be immediately apparent. Once the brain has been removed, before fixation, other findings should be documented, including, • Assessment of the volume of any extradural haemorrhage and/or subdural haemorrhage lacerating and contusional injury, including pontomedullary tears.

Figure 5.2 Haemorrhage into temporalis muscle caused by kicking.

Figure 5.4 Scalp petechiae – of no diagnostic significance.

Retention of the brain • Aneurysms: This should be done before fixation which makes identification at a later stage difficult, if not impossible; blood clot can be washed away followed by dissection along the paths of the arteries. • Infections, including meningitis, which may be predominantly basal. • Intracerebral haematomas may also be identified and care should be taken that they do not disrupt artefactually before fixation.

Figure 5.5 Acute extradural haematoma identified on removal of skull. • Sinus thrombosis. • SDH: most often only an approximate estimation of volume can be given. • Identification of items associated with surgical intervention such as aneurysm clips or coils.

The brain should be weighed fresh – standard tables are given in Appendices 1–3. It should be remembered, however, that these can be used only as a guide. After removal of the brain, taking note of the external appearance in particular in relation to swelling (see Chapter 12) the dura should be stripped for detailed assessment of any bone pathology. Additional examinations include removal of the pituitary gland. Exposure of the middle and inner ears, orbit and air sinuses may be indicated, for example, in cases of infection. Removal of the spinal cord again depends on clinical circumstances. This can be done via either the anterior or posterior approach using standard techniques. The posterior approach is preferred in exposure of the high cervical region/craniocervical junction. Removal of the spinal cord en bloc with the spinal column may be useful in some cases, e.g. compression fractures, vertebral artery trauma (see Chapter 9).

Removal of the brain in newborns and young infants External examination should include examination for congenital abnormalities as well as external injury. The skull should be opened along the sutures. This exposes the brain, the removal of which is aided if performed under water so that it can ‘float’. Fixation can be accelerated using 20 per cent formalin in a 37°C oven for 2–3 days before section if return to the body is necessary. In cases of any posterior cranial fossa abnormality, a posterior approach is recommended (see also Chapter 15).

Retention of the brain

Figure 5.6 Appearances of traumatic subarachnoid haemorrhage visualised on removal of the skull.

It is advisable and considered good practice to retain the brain where there is any issue relating to the neuropathology. This is particularly so where medico-legal proceedings may take place. Full written documentation should be done including consent (if applicable) when a brain is retained for examination. It may also be necessary to have signed continuity documentation regarding location and possession not only of the brain but also of blocks and slides. Although in some cases of head injury, the relevance of formal neuropathology may not be immediately apparent (e.g. in cases where there may be other types of assault such as strangulation or stabbing), issues may arise at a later date that were unknown at the time of the original examination, which then cannot be answered to any degree of satisfaction. A number of issues may also be raised by those acting for the defence, see later. These include length of survival after head injury and underlying pre-existing disease. There have been considerable issues raised over organ retention in recent years in the United Kingdom and other countries, as a result of retention of organs after post-mortem examination where the legal/consent issues were unclear (Burton and

57

58

Techniques Wells 2001; Redfern et al. 2001). This has caused difficulties in retention of brains for formal neuropathology even in medicolegal cases. There is also considerable pressure for the return of the brain before burial of the body. The Forensic Science Regulator addresses the issue in the document titled Legal Obligations in Forensic Pathology and Tissue Retention, 2020a. This document also covers the Human Tissue Act 2004 and the relation with the Criminal Justice System (England and Wales). Separate regulations exist in England and Wales for tissue and the defence (Forensic Science Regulator, Home Office, England and Wales 2020b). If there is no consent for prolonged retention of the brain, even a few days’ fixation in formalin produces significantly better results than immediate sectioning and blocking. This is often possible and practicable, because in many cases, burial or other disposal of the body takes several days to organise. In potentially criminal cases, this is often longer. It should be noted that information may be lost if a brain is not retained. This has been highlighted in paediatric cases, where there has been a marked reduction in formal neuropathology with consequent loss of significant information (Walsh and Moore 2003). Microwave fixation has been tried. This has some advantages where the brain needs to be returned to the body at the same time as the post-mortem examination – the method is given in Appendix 4. However, with complex cases, fixation is not optimal. In addition, histological sections prepared from such brains leads to a degree of artefactual change, which may make interpretation difficult against standard reference texts using established neuropathological techniques (personal experience). The pathologist may, however, have to return the brain immediately. This is particularly so in cases where cause of death is known and permission for brain retention has not been given. In these cases, it may be possible to obtain permission for small samples. It is important that the pathologist documents the extent of examination and the reasons preventing formal detailed dissection in the report, in case medico-legal issues arise at a later date.

Examination after fixation After fixation of the brain, which in standard neuropathological practice is for 3–4 weeks, sectioning can be done. Further details of circumstances and other information may also be available. Further recording of findings on brain cutting is essential, of both the external appearance of the brain before cutting and subsequently on sectioning. If the dura has been retained with the brain, this should be sampled, if necessary in multiple locations, particularly in the case of extra- or subdural haemorrhage. Further examination of the basal cerebral vessels should be undertaken before sectioning of the brain. The cerebellum and stem should be separated from the hemispheres. The stem should be sectioned at approximately 5-mm intervals and in some cases embedded in its entirety (head injury in infancy). Either the cerebellum can be cut in the horizontal plane or more detailed examination can be done by first dividing the hemispheres in the midline followed by horizontal cuts at approximately 5 mm through each hemisphere. The standard method for the hemispheres is for coronal sections – the initial slice is through the mamillary bodies, then at 1-cm intervals through the hemispheres. Details of anatomy are given in Chapter 1.

Blocking of the brain This depends on the circumstances of the individual case. Details of blocking for axonal injury with additional blocks for infant head injury and hypoxic-ischaemic brain damage are given in Appendix 5. Details for sudden unexpected death in infancy (SUDI) cases are outlined in Appendix 6 and for suspected dementia cases in Appendix 7. Details for SUDEP cases are given in Chapter 13 and details for CTE cases in Chapter 14.

Examination of the spinal cord The dura should be examined externally for any pathology. The dural sheath can then be opened both anteriorly and posteriorly for examination. Sampling for histology should be guided by the circumstances. In traumatic injury, transverse sections from above and below the level of injury may be helpful on frozen sections for neutral lipids and myelin staining. This may be an aid to dating injury (Dawson and Neal 2003).

Histological stains In forensic neuropathology practice, routinely used stains include haematoxylin and eosin and beta-amyloid precursor protein (β-APP) and CD68 or Iba-1 immunocytochemistry. In addition, Perls’ staining for iron is useful to assess haemosiderin. Additional stains of relevance include Masson trichrome for collagen, EVG (elastic van Gieson), Alcian blue-PAS and Movat for blood vessels and Congo red or Beta-A4-amyloid immunocytochemistry for amyloid angiopathy. Other stains required will depend on the clinical history, e.g. immunocytochemistry for neurodegenerative disorders, myelin stains for demyelinating disorders.

Provision of tissue to the defence or those instructed by the defence Pathologists acting on behalf of the defence have the right to examine samples and review results obtained by those instructed by the coroner, police and/or prosecution. Guidance is included in the document Provision of Human Tissue to the Defence (England and Wales) by the Home Office and Forensic Science Regulator 2020b. Samples held on the authority of the police can only be transferred to those acting for the defence if the appropriate authorisation has been obtained (police or prosecuting authority) or a court order (2020b).

Findings relating to hospital therapy • Scalp bruising: This may be extensive in association with surgery (see Figure 5.3). • Herniation of the brain through craniotomy/burr holes: The severely swollen brain may herniate through a craniotomy site, producing areas of haemorrhagic pressure necrosis at the edge of the hernia. Swelling with oedema of brain tissue occurs followed by variable haemorrhagic or ischaemic necrosis (see also Chapter 12). A burr hole may be associated with protrusion of smaller amounts of brain tissue (Figure 5.7). • Ventricular catheters: These may produce a variety of haemorrhages including cortical and white matter as well as injury to the corpus callosum. • Non-perfused brain: This is covered in Chapter 12.

Outline of additional procedures for particular clinicopathological entities

(a)

Figure 5.7 Burr hole with protrusion of brain tissue.

Artefacts • Fixation artefacts may occur if care is not taken in suspending the brain. Flattening of the hemispheres with gross distortion may make subsequent interpretation difficult (Figure 5.8). If fresh slices are fixed, these may also become extremely distorted. Reforming the brain with stout absorbent paper between each slice, placing each half of the brain so that the mid-coronal slices are on the base of the bucket, will reduce this (Dawson and Neal 2003).

(b)

Figure 5.9 (a) Post-mortem bacterial changes with cystic spaces seen in the right basal ganglia; (b) histological appearances with post-mortem bacterial colonies. • Changes associated with decomposition: Where there has been a delay in fixation, post-mortem bacterial activity produces a ‘Swiss cheese’ artefact (Figure 5.9a). Histology shows bacterial colonies (Figure 5.9b). • Spinal cord artefact: Excessive handling may cause localised softening/squashing with, if severe, the ‘toothpaste’ effect where there is distortion of internal architecture. • Histological artefacts: Neuronal shrinkage, particularly near the surface of the brain, may be seen with ‘corkscrew’ twisting of neurons and dark staining nuclei. This may be confused with hypoxic-ischaemic cell damage (see Chapter 12) (Figure 5.10).

Outline of additional procedures for particular clinicopathological entities Sudden unexpected death in infancy (SUDI)

Figure 5.8 Distortion of brain as a result of poor suspension/ fixation.

This group of deaths is particularly difficult and challenging. The term is used for deaths in infancy (the first year of life) where death is sudden and unexpected. This topic is covered further in Chapter 17. In some, a cause will be determined such as congenital heart disease or infection. In others, despite full investigations no definite cause is identified.

59

Techniques

60

the base of the skull and subsequent examination of the intracranial portion of the carotid arteries including the cavernous sinus should be undertaken. This may necessitate fixation and decalcification (Langlois and Little 2003). Tissue for genetic analysis may be indicated in young people (snap-frozen brain, skeletal muscle and blood).

Infectious cases

(a)

(b)

Figure 5.10 (a) Neuronal shrinkage with ‘corkscrew’ neurons not to be confused with (b) genuine hypoxic-ischaemic cell change.

There has been considerable variation in the terminology in these cases with sudden infant death syndrome, unascertained and SUDI being in use. The joint Royal College of Pathologists and Royal College of Paediatrics (2016c) includes details of the post-mortem protocol for sudden non-expected deaths in infancy (non-suspicious). Inflicted injury must be excluded in such cases – see Chapter 15. The issue of brain retention should be done on a case by case basis by the pathologist. The details of blocking for neuropathological examination are given in Appendix 6.

Examination in cases of suspected stroke In addition to the procedures outlined above, assessment of the following should be undertaken. Some of these come within the general examination and include a detailed assessment of the heart for left ventricular hypertrophy, myocardial infarction and mural or atrial thrombus, and assessment of the atheroma of the aortic arch to the carotid arteries with regard to severity of atheroma, together with evidence of other disease processes such as diabetes and hypertension. Specific neuropathological procedures related to the CNS include examination of the vertebral arteries (the procedure is the same as for examination in basal subarachnoid haemorrhages outlined in Chapter 9). Examination of the carotid arteries from the origins of the carotid artery to

All forensic cases are potentially high risk with a significant proportion from the intravenous drug abuser population. With the increasing rise in drug abusers and communicable diseases associated with this, as well as the emergence of infections including tuberculosis and hepatitis viruses including C and HIV, post-mortem examinations may be performed in a high-risk mortuary, although with appropriate precautionary measures a standard mortuary may be sufficient. Most of the viruses, and in addition the infections that may be opportunistic in AIDS cases, are readily inactivated by formaldehyde. Additional investigations for virology, including polymerase chain reaction analysis, may also be indicated. Suspected or known cases of Creutzfeldt-Jakob including the variant form identified in young adults should be carried out in specialist mortuaries equipped to deal with the additional decontamination procedures required. Details of post-mortem examination in suspected cases of Creutzfeldt-Jakob disease can be found in specialist neuropathological texts (Dawson and Neal 2003). Most recently post-mortem examinations have been undertaken in COVID-19 deaths. The neuropathology is covered in Chapter 18.

Sudden unexpected death in epilepsy (SUDEP) The post-mortem examination and neuropathology are covered in detail in Chapter 13.

Miscellaneous • CNS tissue may be useful in estimation of diatoms in cases of suspected drowning, although other tissues including long bones may also be used. Diatoms are aquatic unicellular plants with an extracellular coat (frustale) composed of silica (Figure 5.11a and b). There are many morphologically distinct varieties. They are found in naturally occurring water such as lakes, seas and rivers. The method for diatom analysis is given in Appendix 8. • Toxicological analysis may be done for carbon monoxide and alcohol on any extracranial haematoma. In specialist toxicological hands, estimation of other substances may be possible depending on the circumstances. Expert toxicological advice should be sought in such cases. • In cases of suspected excited delirium/restraint deaths, snap-frozen brain tissue may be retained. This should include the basal ganglia at the level of the anterior temporal poles with a block thickness of 0.5–1 cm snap-frozen in liquid nitrogen and stored at −70°C. This enables neurochemical examination to be undertaken, in particular for dopamine receptors; however, the diagnostic value in such cases is open to question – see also Chapter 19 (Mash and Staley 1999; Stephens et al. 2004).

Appendix 2 Infant and child brain weights

(a)

Royal College of Pathologists. 2016c. Sudden unexpected death in infancy and childhood. Rutty, G.N., Morgan, B. and Robinson, C. et al. 2017. Diagnostic accuracy of post-mortem CT with targeted coronary angiography versus autopsy for coroner-requested post-mortem investigations: a prospective, masked comparison study. Lancet 390 145–154. Stephens, B.G., Jentzen, J.M. and Karch, S. et al. 2004. Criteria for the interpretation of cocaine levels in human biological samples and their relation to the cause of death. Am J Forensic Med Pathol, 25(1), 1–10. Walsh, Z. and Moore, I.E. 2003. Changes in paediatric post mortem examination practice – are there benefits for the parents? J Pathol, 201(suppl 44A). Waters, B.L. (ed) 2009. Handbook of Autopsy Practice. 4th edn. Totowa, New Jersey: Humana Press.

APPENDICES Appendix 1 Normal adult brain weight with leptomeninges and dimensions Male

Female

Adult whole brain (g)

1400 (normal range 1100–1700)

1275 (normal range 1050–1550)

Adult cerebellum (g)

147 (normal range 136–149)

133 (normal range 127–137)

C:C ratio (per cent)

10–11

10–11

Sagittal diameter (cm)

16–17

15–16

Vertical diameter (cm)

12.5

12.5

(b)

Figure 5.11 (a) Diatom (from lung); (b) diatom preparation after acid digestion of tissue (phase contrast illumination).

References Burton, J.L. and Wells, M. 2001. The Alder Hey affair: implications for pathology practice. J Clin Pathol, 54, 820–823. Dawson, T.P. and Neal, J.W. (eds) 2003. Neuropathology Techniques. Arnold. Dekaban, A.S. and Sadowsky, D. 1978. Changes in brain weights during the span of human life: relation of brain weights to body heights and body weights. Ann Neurol, 4, 345–356. Esiri, M. and Perl, D. 2006. Oppenheimer’s Diagnostic Neuropathology; A Practical Manual. 3rd edn. London: Hodder Arnold. Forensic Science Regulator, Home Office, England and Wales (2020a). Legal obligations in Forensic Pathology and Tissue Retention. Forensic Science Regulator, Home Office, England and Wales (2020b). Provision of Human Tissue to the Defence. Geddes, J.F., Whitwell, H.L. and Graham, D.I. 2000. Traumatic axonal injury; Practical issues for diagnosis in medicolegal issues. Neuropathol Appl Neurobiol, 26, 105–116. Langlois, E.I. and Little, D. 2003. A method for exposing the intraosseous portion of the carotid arteries and its application to forensic case work. Am J Forensic Med Pathol, 24, 35–40. Mash, D.C. and Staley, J.K. 1999. D3 dopamine and kappa opioid receptor alterations in human brain of cocaine-overdose victims. Ann NY Acad Sci, 877, 507–22. Redfern, M., Keeling, J.W. and Powell, E. (eds) 2001. The Royal Liverpool Childrens Inquiry Report, 3rd edn. HMSO. Royal College of Pathologists and Forensic Science Regulator. 2018. Code of practice for forensic pathology. Royal College of Pathologists. 2010. Neuropathology autopsy practice: post mortem examination in patients with traumatic brain injury. Royal College of Pathologists. 2016a. Code of practice and performance standards for forensic pathologists dealing with suspicious deaths in Scotland. Royal College of Pathologists. 2016b. The Hutton review of forensic pathology, imaging-based autopsies and the future of the coronial autopsy service—a commentary.

Source: Reprinted from Dawson, T.P. and Neal, J.W. 2003. Neuropathology Techniques. Arnold.

Appendix 2 Infant and child brain weights Age (months)

Weight (g)

Term

362

14

 927

1

415

16

 985

2

469

18

1021

3

520

20

1054

4

558

24

1098

5

619

36

1184

6

641

4 years

1228

7

680

5 years

1251

8

717

6 years

1267

Age (months/years)

Weight (g)

9

758

7 years

1281

10

797

8 years

1292

11

830

9 years

1275

12

860

10 years

1314

Note: Gender-dependent weight difference becomes evident at around 5 years. These are average values (metadata compiled from several sources). Source: Reprinted from Dawson, T.P. and Neal, J.W. 2003. Neuropathology Techniques. Arnold.

61

62

Techniques

Appendix 3 Adult brain weights Age (years)

Male (g)

Female (g)

10

1440

1260

15

1440

1280

20

1450

1310

30

1440

1290

40

1430

1290

50

1410

1280

60

1370

1240

70

1350

1230

80

1310

1171

Source: Adapted from Dekaban and Sadowsky (1978). Reprinted from Dawson, T.P. and Neal, J.W. 2003. Neuropathology Techniques. Arnold.

Appendix 4 Rapid microwave fixation method for brain fixation • Remove brain as usual • Place in plastic brain bucket in 10 per cent formalin, covering the brain • Place lid on loosely • MICROWAVE – normal domestic • 20 minutes – reheat • 20 minutes – low • Leave to cool (about 20 minutes) • Rinse in cold water • Cut

In addition, in suspected non-accidental infant head injury • Pons and medulla • All the upper cervical cord segments

For hypoxic brain damage • Samples of all cortical areas (frontal, parietal, temporal, occipital, to include arterial boundary zones) • A block of deep grey matter • The hippocampus • A block of cerebellar hemisphere • A block of brainstem Note that if small (3 in. x 1 in.) slides are used, two samples of each of the above areas should be taken. Representative blocks should also be taken of any focal pathology seen. Modified from Geddes, Whitwell and Graham 2000, by kind permission of Blackwell Publishing Ltd. In addition any visible lesions should be sampled. Special stains β-APP, CD68, Perls (to be guided by circumstances).

Appendix 6 Blocks in SUDI cases See Figure 5.13. • • • • • • • • • • •

Note: Infant brain times – 10 minutes – reheat, 10 minutes – low.

Appendix 5 Minimum sets of blocks recommended for determining the distribution and amount of microscopic brain damage in head injury See Figure 5.12.

For traumatic axonal injury • Corpus callosum and parasagittal posterior frontal white matter • Splenium of the corpus callosum • Deep grey matter to include posterior limb of the internal capsule • Cerebellar hemisphere • Midbrain (to include decussation of superior cerebellar peduncle) • Pons (to include superior or middle cerebellar ­peduncles)

Mid frontal gyrus Mid corpus callosum Basal ganglia Parietal cortex Caudate Thalamus Hippocampi Calcarine sulcus Superior cerebellum Mid pons, whole medulla Plus any visible lesion and if trauma suspected as per Appendix 5

Appendix 7 Minimum set of blocks recommended for the assessment of dementia Blocks: cerebrum; hippocampal block including hippocampus, entorhinal cortex and inferior temporal gyrus • • • • •

Superior and middle temporal gyri Anterior tip of the temporal lobe Middle frontal gyrus Basal ganglia at the level of the mamillary bodies Thalamus

Cerebellum: Section of cerebellum including cerebellar cortex and dentate nucleus Brainstem: Midbrain including substantia nigra • Pons including locus coeruleus (pigmented nucleus) • Medulla In addition, sample any areas that macroscopically appear ­atrophic and any focal lesions (such as lacunar infarcts)

Appendix 7 Minimum set of blocks recommended for the assessment of dementia Corpus callosum, posterior frontal cortex and parasagittal white matter

Watershed zone in deep grey matter Posterior limb of the internal capsule

Splenium of the corpus callosum

Anterior watershed

Parietal cortex and posterior watershed

Anterior corpus callosum

Hippocampus and temporal cortex

Midbrain ( + decussation of SCP)

Occipital and calcarine cortex

Pons

Cerebellum ( + MCP)

Figure 5.12 Trauma sampling protocol.

63

64

Techniques

Figure 5.13 SUDI protocol.

Appendix 7 (Cont) Special stains Immunohistochemistry for the following antibodies should be undertaken: Tau and β-amyloid: midbrain, pons, hippocampus and entorhinal cortex, cerebellum, basal ganglia, superior and middle temporal gyri, striate and peristriate occipital lobe α-Synuclein: midbrain, pons, hippocampus and adjacent collateral sulcus, middle frontal gyrus, superior and middle temporal gyri

pTDP-43: hippocampus, amygdala, middle frontal gyrus, superior and middle temporal gyri

Appendix 8 Procedure for diatom analysis Requirements: Fresh lung*, Liver*, Kidney, Spleen or Brain. (*Preferred choice of tissue.) Amount of fresh tissue required: = 100 g NB. A water sample should be taken from the scene because without this qualitative and quantitative analysis cannot be made.

Appendix 8 Procedure for diatom analysis Methodology • Observe the water sample using phase-contrast microscopy. • Record the types and the number of diatoms present diagrammatically and via digital photography. • Only if diatoms are present in the scene water proceed to extract possible diatoms from fresh tissue. • In a class 1 cabinet place the tissue sample in a clean flask using forceps rinsed in distilled water. If the sample is too large use a scalpel, rinsed in distilled water, to cut into smaller sections. • Transfer the flask/s to a fume cupboard and add 100 cm3 concentrated nitric acid. • Cover the flask/s with clean loose lid/s and leave in a container in the fume cupboard for 48 hours. • When tissue has dispersed into the acid, heat slowly using a hotplate until the volume has reduced to 30 cm 3 and allow to cool. • Add 20 cm3 of concentrated sulphuric acid and heat slowly until the solution turns black. • Partly fill a separating funnel with concentrated nitric acid and set the tap to drip slowly into the charred solution. • Continue to heat the solution slowly whilst dripping in the nitric acid until the solution turns a pale straw colour.

• If a large amount of residue is visible hydrolyse at 60°C in hydrochloric acid. • Allow to cool, and then centrifuge aliquots of the solution at 3000 rpm for 10 minutes, removing the supernatant via suction. • Repeat until the flask is empty. • Rinse the flask out with distilled water to remove any remaining residue, and centrifuge as before. • If the pellet is from a sample of lung tissue – remove final supernatant and add 0.5 cm 3 distilled water to the pellet. • Resuspend the pellet using a plastic disposable pipette and transfer to a stoppered microfuge tube (1 cm3 volume). • Transfer one drop to a clean slide, coverslip, and observe using phase-contrast microscopy. • If the pellet is a sample of liver or other tissues – reduce the residue to one large drop and observe all the material on the slide. • Record types and the number of diatoms present both diagrammatically and via digital photography. • Match the diatoms found in the water sample against any found in the tissue extracts. Courtesy of Ian Newsome, Department of Forensic Pathology, University of Sheffield.

65

6

Scalp, facial and gunshot injuries Christopher Milroy

Surface injuries are central to forensic medicine. The pattern of injury provides information on the mechanism of causation of both external and internal injuries. Although this chapter deals predominantly with injuries to the scalp and face, it should be remembered that clues to the causation of injury may be present on other parts of the body. A systematic approach should be made to recording injuries, including type, dimensions, shape and colour, with the anatomical position, which can be linked to appropriate anatomical landmarks. Medical descriptions of wounds and injuries can be at variance with how legal systems define wounds. Gunshot wounds are lacerations with specific patterns of injury and are dealt with under a separate section in this chapter. As the effects of penetrating injuries are intrinsically linked to the surface injuries, this chapter also includes the effects of penetrating trauma.

Legal definitions Black’s Law Dictionary (Garner 2019) describes wounding as ‘an injury’, especially one involving rupture of the skin. Bodily injury is described as ‘physical damage to person’s body’, but the two are not necessarily the same. In English law, it is now established case law that a wound must involve the breaching of the full-thickness of the skin, i.e. the epidermis and dermis (Moriarty v Brookes 1834). In the case of an airgun pellet that struck an eye causing internal haemorrhage but no breach of the surface, it was held not to be a wound [C (a minor) v Eisenhower 1984]. Similarly, an abrasion is not a wound in law (M’Loughlin 1838). A fracture of the clavicle without breach of the overlying skin was also not a wound (Wood 1830). Although the legal definition of a wound is restricted, alternative legal charges usually exist to take account of this apparent anomaly. However, in describing a wound, some knowledge of the legal definition (if any) in the jurisdiction in which the report is to be used is helpful. Terms such as ‘cut’, ‘superficial laceration’ and ‘deep abrasion’ are confusing and may result in legal challenges.

External examination For a clear demonstration of injuries on the head, the hair may need to be shaved during post-mortem examination. It is surprising how frequently thick hair covers significant injuries and prevents identification. Photographs of the injuries should be made in colour. Injuries should be photographed with and without a measuring scale and at right angles, not obliquely. Where necessary, e.g. with bite marks and patterned bruising, a right-angled ruler should be used (Figure 6.1). Techniques such as use of ultraviolet light should be considered; they may enhance the features of an injury (Hempling 1981; Krauss and Warlen 1985). With some patterned injuries such as those caused by footwear, forensic scientists currently prefer black and white photographs, so that they can compare the pattern with footwear held on databases. Many practitioners make notes of injuries by drawing on prepared body plans.

Injuries may alter in appearance in the post-mortem period. Abrasions may dry out and appear more prominent, although they will not significantly change in size, unlike bruises, which may increase in size and prominence. A second examination of the body may therefore reveal bruises that were not initially evident. However, caution must be exercised that the injury is not an early change of putrefaction or an artefact of the initial postmortem examination. If necessary, microscopic examination of the skin can be undertaken. Microscopy can also be helpful in dating an injury, as detailed later.

Classification of injuries Abrasions A pure abrasion is an injury involving only the epidermis. Abrasions heal without scarring. An abrasion may result from a tangential movement of skin in comparison with an abrading material, or by direct imprint. In the head and face, abrasions commonly occur when a person falls. When a person falls forwards, the prominent areas of the face will be injured if the person is unable to break the fall (Figure 6.2). Falls on to the back of the head may result in occipital or parieto-occipital region abrasions, along with bruising or lacerations (Figure 6.3). Scuffed kicks may abrade the skin. Where there is a significant tangential movement of skin against an abrading surface, the direction of movement can be determined by the position of the skin tags in the abrasion. Imprint abrasions are seen when a patterned object impacts on skin. A typical example is a ligature mark. Abrasions, as well as bruises and lacerations, may be seen when a spectacle wearer is struck in the face. If such injuries are seen on the bridge of the nose, or around the eyes, the possibility of the victim wearing spectacles should be considered, because this information is not always immediately available (Figure 6.4). Abrasions may be inflicted after death, when they appear yellow, particularly during post-mortem movement of the body. Abrading-type injuries may be seen around the mouth when gastric acid from vomitus has spilt around it (Figure 6.5).

Bruises Bruises are caused by trauma to the skin that ruptures blood vessels and blood escapes into the tissues. Particularly in deeper tissues, bruises are often called contusions. The older term ‘ecchymosis’ is also sometimes used. In forensic practice, findings must be described in terminology that can be understood by non-medical lay people and if simple terminology can be used, it is to be preferred. It is more likely that a jury will understand ‘black eye’ than periorbital haematoma. The smallest pattern of bleeding is a petechial haemorrhage or a petechia. Petechiae are pinhead-sized areas of bleeding found in the skin and internal organs. Although often described as a sign of asphyxia, they are by no means specific and are seen in a number of different causes of death, as well as non-fatal

Classification of injuries

Figure 6.1 Bite mark: photographed with right-angled scale. cases. In the head, they can be seen in the eyes (Figure 6.6a) and eyelids, on the face and behind the ears, as well as sometimes on mucosal surfaces (Figure 6.6b). Petechial-type haemorrhages on the undersurface of the scalp found following dissection are a well-recognised artefact and should not be interpreted as arising ante-mortem (see Chapter 5). In the absence of other

Figure 6.2 Abrasions over bony prominence of cheek.

Figure 6.3 Linear laceration with surrounding abrasion and bruising. external petechiae such as may be seen in external pressure to the neck, they are meaningless in determining the mode of death. A bruise may vary in shape and size depending on how, where and when it was inflicted. Bruises are well recognised to undergo colour changes after infliction. These colour changes have been used in an attempt to date the time of infliction. Forensic pathologists try to differentiate fresh bruises from older injuries, although studies have shown that considerable caution must be exercised in attempting to date a specific bruise (Grossman et al. 2011). Indeed an analysis of standard forensic pathology textbooks showed a surprising variation in stated age of a bruise as assessed by colour (Wilson 1977). Bruises that are fresh are often blue or red-purple. They go through a series of changes through brown and green to yellow. This sequence, however, will vary and depends on a number

Figure 6.4 Abrasions on bridge and side of nose from spectacle damage. The black eyes are secondary to underlying orbital fractures.

67

68

Scalp, facial and gunshot injuries

Figure 6.5 Skin damage caused by vomitus post-mortem.

of factors such as size and site, age of victim and medical condition present, as well as idiosyncratic factors in the victim. Maguire and colleagues (2005) have stated that in childhood, there is no scientific basis in trying to date a bruise based upon clinical assessment or photographs and this should not be done in child protection proceedings. Although a red or blue swollen bruise is likely to be fresh, often defined as less than 24–36 hours old, red and blue colouration may be seen in older bruises. A brown bruise may be fresh or old. In a study of bruising, Langlois and Gresham (1991) found that the only reliable finding was that a yellow bruise was at least 18 hours old. Stephenson and Bialas (1996) examining children found that a yellow bruise was at least 24 hours old. However, Mosqueda and colleagues (2005) found yellow bruises occurring in less than 24 hours in a number of cases in patients over 65 years of age. Caution must therefore be exercised in determining the age of a bruise, although as a rule bruises of a similar size and site but of significantly differing colour are likely to have been inflicted at different times. A large collection of blood under the skin is known as a haematoma. Haematomas are an important finding in head-injured patients, particularly young children. In these cases, bruising on the external surface of the scalp may not be discernible, whereas reflection of the scalp at post-mortem examination may reveal significant areas of bruising. The risk of an underlying skull fracture and intracranial injury has been shown to increase with the size of scalp haematoma and haematoma location in infants, where temporal and occipitoparietal haematomas were strongly associated with skull fracture, although frontal haematomas were not. Skull fracture without associated haematoma was common in infants aged under 3 months (Greenes and Scutzman 2001). Bruises may sometimes be patterned, replicating the weapon or implement that inflicted them. A relatively common example on the head is where the victim has been stamped on (Figure 6.7). Careful documentation of these injuries is important because

(a)

(b)

Figure 6.6 (a) Petechiae in the subconjunctiva and (b) in the mouth.

Figure 6.7 Patterned bruising caused by stamping with a shod foot.

Classification of injuries by blows with a mallet. Caution should therefore be exercised in ascribing a bruise to the back of the head as being caused by ante-mortem trauma if a person has been lying face up for days with a head injury, or where the rough handling of a body postmortem cannot be excluded.

Lacerations

Figure 6.8 Tramline bruising caused by blows with a baseball bat. the type of shoe may be identified by comparison with databases. Another characteristic pattern is caused by a cylindricalshaped object such as a baton, baseball bat or similar-shaped object. These weapons often cause a tramline bruise, where there is a central zone of pallor with a band of bruising either side (Figure 6.8). Another type of patterned bruise that may be seen on the face is a bite mark. In such marks, there may be abrasion and laceration (see Figure 6.1). When such a mark is suspected, expert forensic odontological advice should be sought. ‘Love’ bites, commonly seen on the head and neck, are caused by suction, rather than biting. It is important to look inside the mouth, because this is a common site for bruising after a blow to the area, often associated with damage to teeth (Figure 6.9). Bruises may track through the tissues, so a bruise on the forehead may settle under gravity around the eyes. Similarly, bleeding from a fractured nose may result in bilateral black eyes. It is generally accepted that bruises are ante-mortem injuries. However, post-mortem bruising can be produced but requires much more force than when a circulation is present. Polson et al. (1985) produced post-mortem bruises of the occiput

Figure 6.9 Bruising and laceration to inside of lips with associated damage to the teeth caused by a blow to the mouth.

Many doctors as well as non-medically qualified people call all wounds lacerations, not differentiating a laceration from an incised wound (see next section). A laceration is a splitting or tearing of the skin caused by blunt trauma. The head is a particularly common area for lacerations to occur because the skin is closely associated with underlying bone in this region and, therefore, when blunt force is impacted on the skin, the skin is rapidly compressed against the bone. In comparison, considerably more force is required to inflict a laceration in the chest or abdomen. Lacerations on the face may result from a punch, but a punch would not be expected to cause a laceration on the scalp, unless a ring, knuckleduster or similar object was worn. On the face, lacerations occur over bony prominences, including the orbital ring, and over the maxillae. With severe damage to the underlying skeleton, the bones may break through the skin, lacerating it in the process. An unusual example of this mechanism was described by Ainsworth and Hunt (1993) where a blow to the nose, which pushed the nose upwards, resulted in the nasal bone penetrating the skin, before returning to its original position. This resulted in a curved laceration of the tip of the nose. Punches to the lower jaw do not typically cause a laceration of the skin, even with a fractured jaw, although the assailant’s hand may be lacerated from the blow, especially if he strikes the teeth of the victim. Lacerations may be linear, or slightly curved on the scalp where the blow is from a linear object (Figure 6.10). Falls onto the head may also result in a linear laceration (see Figure 6.3). Unlike an incised wound, a laceration has irregular edges, with bruising and abrasion of the margins. Tissue may bridge the wound and hair across the wound may be uncut, unlike an incisional wound. Dirt may have been forced into the wound. Lacerations, bruises and abrasions to relatively inaccessible

Figure 6.10 Multiple linear lacerations with associated bruising caused by blows with a blunt implement.

69

Scalp, facial and gunshot injuries

70

(a)

Figure 6.12 Complex laceration caused by hammer blows. concerns that the lacerations are the result of an attack but do not prove an attack. An intoxicated person, for example, may fall, get up and fall again, sustaining more than one laceration (see Chapter 17). Punches to the mouth can cause bruising and laceration to the inside of the lips, when no external injury is visible (see Figure 6.9). There may be associated with damage to the teeth. A common pattern of external injury in accelerated

(b)

Figure 6.11 (a) Laceration under the eyebrow caused by kicking and (b) with more superficial lacerations and bruising. areas where a blunt agent is normally used, such as under the chin or beneath the eyebrow, are often the result of a blow with a shod foot (Figure 6.11a,b), also causing a black eye (Teare 1961). Blows with large implements, such as a lump hammer or sledgehammer, may produce complex lacerations (Figure 6.12). The shape of a laceration may reflect the shape of an implement used in an attack. A characteristic example is a blow by a hammer that produces a crescent-shaped laceration (Figure 6.13). Lacerations on the back of the head are a common finding after a fall and there may be associated with abrasion and bruising (see Figure 6.3). More than one laceration on the scalp raises

Figure 6.13 Crescent-shaped lacerations caused by blows with a ball-pein hammer.

Classification of injuries falls, producing a significant head injury, is bruising with or without laceration of the inside of the lip from a punch, associated with bruising, laceration and abrasion to the occiput. A torn frenulum is an important injury in child abuse (see Chapter 15).

Incised wounds and stab wounds Injuries caused by sharp-edged objects produce a clean-edged wound. The paradigm of an incised wound is the surgical wound produced by a scalpel. However, any sharp-edged object may produce an incised wound, and facial injuries inflicted by broken glass are an important cause of morbidity. Although incised wounds to the head are frequently seen, they are rarely fatal, although, as with lacerations, occasional deaths from haemorrhage occur. Self-inflicted incised wounds to the head may be suicidal or inflicted for personal gain. Where they are inflicted for personal gain, they are typically superficial and do not involve sensitive areas of the face such as the eyes or lips. Most fatal stab wounds are inflicted to the trunk, particularly the chest, but penetrating wounds to the head are occasionally encountered (Figure 6.14). Stab wounds should be recorded by site, external dimensions, direction of travel and depth. Although some knives leave injuries that indicate whether they are single or double edged, this is often not possible to determine. Movement of the knife in relation to the victim may cause variation in the size of the surface wound. Some weapons have serrated edges, but this is often not possible to determine from the wound. Stab wounds aimed at the head may penetrate the skull, but if they do not the wound may be significantly affected by movement against the skull. Some stabbing weapons, notably scissors, may produce more irregular wounds. Blunter stabbing weapons, such as screwdrivers, may also produce penetrating wounds. In these cases, the screwdriver can produce a small

Figure 6.15 Stab wound caused by a screwdriver. slit-like laceration, belying the degree of damage caused by the passage of the weapon into the brain (Figures 6.15 and 6.16).

Chop injuries Chop injuries occur with a relatively blunt-edged heavy object such as an axe, machete, meat cleaver or sword. When they occur, they are commonly found on the head. The skin shows features of both sharp force and blunt force, that is a mixed incised wound and laceration and there is often damage to the underlying skull when delivered to the head (Figure 6.17). Similar injuries are seen from propeller strikes, often seen as large parallel chop injuries.

Burns Burns may be the result of the application of dry or moist heat, or of the contact of a chemical with the body. Burns can be divided by depth of injury into three groups: superficial (first degree), deep dermal (second degree) and full-thickness (third degree).

Figure 6.14 Stab wound penetrating the eye.

Figure 6.16 Cerebral damage caused by penetration with a screwdriver.

71

72

Scalp, facial and gunshot injuries

Figure 6.18 Fire death with artefactual heat fractures of the skull.

Figure 6.17 Injuries to the scalp caused by a machete. Moist heat characteristically causes scalds to the skin (first degree), whereas dry heat may cause any degree of damage. With extensive burning, as is commonly encountered in house fires, or after deliberate attempts to destroy the body, extensive destruction involving the body may be seen with exposure of muscle and bone. However, complete destruction of the body is rare and requires prolonged high temperatures. With extensive burning of the skin of the head, artefactual splits may occur and these must not be confused with ante-mortem injury. Bodies subject to significant burning of the head frequently have artefactual heat fractures (Figure 6.18) and extradural heat haematoma (see Chapter 8). Electrical injury causes damage to the skin, which can vary from a small, localised injury that may be difficult to identify to extensive destruction with high-tension current, such as when victims have climbed electricity pylons, the burns often involving the head, as well as other areas of the body. Deliberately inflicted burns, where a weapon such as a cigarette is pressed against the skin, may be encountered, particularly in child abuse, where such injuries are pathognomonic of abuse.

the barrel. These grooves are spiralled, with varying numbers, running in a clockwise or anticlockwise direction. The metal ridges left between the grooves are called lands. This rifling imparts gyroscopic stability on the bullet in flight. Examination of the wound characteristics in firearm injuries provides information on type of weapon used and distance discharged from. Accurate determination of range requires test firing of the gun with the appropriate ammunition. Before starting a post-­ mortem examination in suspected firearm deaths, radiographs of the body should be taken (Figure 6.19) (DiMaio 2016; BesantMatthews 2000). Multiple wounds do not exclude self-infliction (Kury et al. 2000)

Rifled weapon injuries Rifled weapons usually discharge a single bullet, unlike shotguns. Handheld weapons are of two main types: revolvers and self-loading pistols. Revolvers usually have a muzzle velocity of around 700 ft/s (200 m/s). The self-loading pistol loads bullets in a spring-loaded magazine. Each time a bullet is fired, a

Firearm injuries Firearms are the most frequent cause of penetrating wounds of the head. There are two principal categories of firearms: rifled weapons and smooth bored shotguns. Rifled weapons may be short or long barrelled and have a series of grooves made into

Figure 6.19 Skull radiograph of 0.22 gunshot wound to head (contact).

Rifled weapon injuries replacement bullet is moved into the firing chamber until the bullets are all discharged. The spent cartridge case is expelled, unlike the revolver, where the cartridge cases are retained until emptied by hand. In a semi-automatic setting, a bullet is advanced into the firing chamber singly each time the trigger is fired, whereas in fully automatic mode, bullets are continually discharged while the trigger is depressed until all are used up or pressure is released. Self-loading pistols typically have a higher muzzle velocity than revolvers, being around 300–360 m/s (1000–1200 ft/s). All 0.22 rifles usually have lower muzzle velocities, similar to those of handguns. Other long-barrelled rifled weapons usually have a high muzzle velocity, over 300 m/s (1000 ft/s). High-powered rifled weapons, such as military rifles, may have muzzle velocities of 600–1200 m/s (2000–4000 ft/s), although all these quoted velocities will be modified depending on the type of cartridge used.

Contact wounds In the head, the contact is often firm, the skin being in close approximation to the bone. Muzzle gases enter the wound, along with other ejecta, including primer and propellant residues and the bullet. Externally, little soot and powder burning may be evident and may require microscopic examination to confirm that it is a contact wound. As muzzle gases contain a high percentage of carbon monoxide, the tissues may appear pink, and chemical analysis reveals the presence of carboxyhaemoglobin and carboxymyoglobin. When the hard contact is over bone, most characteristically seen in the head, the muzzle gases pass into the skin and the pressure causes splitting of the skin as they exit, producing a stellate or cruciate laceration (Figure 6.20). Close examination of the edges of the wound will reveal powder burns and soot. The amount of splitting of the skin will depend on the amount of gas formed after firing of the gun. Bullets with lower charges will produce less splitting, if any. Splitting of the skin may be seen in intermediate and distance shots, when the bullet strikes a bony

Figure 6.21 Contact wound with muzzle abrasion.

prominence. When the muzzle is in hard contact with skin, the muzzle may become imprinted on to skin as a muzzle abrasion (Figure 6.21). When the contact is not hard, a muzzle abrasion is less likely, but a ring of soot and powder will occur around the entrance, because a small gap is present between muzzle and skin during discharge. In angulated contact wounds, the gun is pushed into the skin so that the muzzle is only partially in contact with the skin. Muzzle gases and powder will escape between the gap and give an eccentric pattern of staining on the skin. If the gap is sufficiently large, unburnt powder can escape and powder tattooing may be seen on the opposite side to where the barrel was in contact with the skin.

Near contact and intermediate wounds

Figure 6.20 Contact gunshot wound of head with stellate entrance wound.

Near contact wounds occur when the muzzle of the gun is close to, but not in contact with, the skin. The characteristic appearance is of an entrance wound surrounded by soot with blackened, heated skin. Unburnt powder may be present. If the barrel is angled to the skin, the distribution of the soot will be eccentric to the entrance wound. No muzzle abrasion will be present, because the muzzle has not been in contact with the skin. When the muzzle of the gun is sufficiently away from the skin, unburnt powder produces stippling (‘tattooing’) of the skin (Figure 6.22). In addition to stippling, soot will be discharged from the barrel and deposited on the skin. The pattern of stippling and soot staining will depend on a number of factors, including type of weapon and ammunition, length of barrel and distance fired from the body. With longer barrels, soot deposition occurs for a greater distance from the target than a shorter barrel. In general, soot and powder deposition can be seen up to a range of 20–30 cm with handguns, although soot may occur at a greater range depending on the powder used in the cartridge. Ball powder travels farther than modern flake powder. As the range increases, the powder tattoo pattern will

73

74

Scalp, facial and gunshot injuries

Figure 6.22 Close range gunshot wound with powder burns. become less dense and more widely dispersed. Shorter barrelled weapons will produce a more widely dispersed pattern than a longer barrelled weapon for the same ammunition and range. Clothing will of course interfere with these patterns.

Distant wounds Distant wounds are characterised by the absence of soot deposition and powder tattooing. With handguns, this is of the order of 60–100 cm, although, as with all firearm injuries, test firing will be required to determine any range with accuracy. However, if the shot has been fired from a range beyond which powder tattooing and soot deposition would occur, no maximum range can be given because the distance may be a few metres or hundreds of metres. Clothing may modify the picture. If clothing is present and the person shot from intermediate range, all soot and powder grains may be absorbed, giving the entrance wound the appearance of a distance shot. The wound will consist of an entrance hole with or without an abrasion collar, but no powder tattooing or soot deposition. Small tears radiating from the entrance hole may be seen. Other intermediate targets such as glass may modify the wound appearance, as will ricocheting bullets causing instability in the bullet and an irregular entrance wound (Burke and Rowe 1992).

Figure 6.23 Atypical entrance gunshot wound of occiput.

Skull damage Bullets penetrating the skull will characteristically produce a bevelled wound with the inner table hole larger than the external table – so-called internal bevelling (Figure 6.24). The opposite effect is seen with exit wounds, where there is typically external bevelling (Figure 6.25). However, on occasions external bevelling may be seen in some entrance wounds. This most commonly involves contact wounds, but distance shots may also produce this phenomenon. Tangential wounds of the skull have been called ‘gutter wounds’. There may be various degrees of penetration of the skull. In first-degree gutter wounds, the outer table is damaged.

Tangential wounds If a bullet hits the skin at a shallow angle, a number of different wounds may result. In tangential entrance gunshot wounds, sometimes called atypical entrance gunshot wounds, the bullet passes superficially through skin causing multiple splits. The direction of the bullet may not be obvious and a failure to appreciate that a bullet can cause such an injury may result in the wound being interpreted as a non-firearm injury. If the firearm is discharged close to the skin, powder and soot staining may be seen at one end of the wound (Figure 6.23). If the bullet passes very superficially across the skin, an abrading injury or superficial splitting of the skin may occur.

Figure 6.24 Inside of skull showing internal bevelling caused by buckshot.

Shotgun injuries

Figure 6.25 Outside of skull showing external bevelling caused by buckshot. In second-degree wounds, pressure waves fracture the inner table. In third-degree wounds, the outer table is fragmented and the inner table fractured. Fragments of bone may be driven into the brain. When the bullet superficially perforates the skull, there may be an entrance and exit wound in the skull close together. When the bullet strikes the skull at a shallow angle, a keyhole wound may be produced. The bullet most commonly breaks as it penetrates bone so that part of the bullet travels under the scalp, with the bulk of the bullet penetrating through the skull. This process produces a keyhole-shaped wound with one side of the wound being a typical entrance wound and the other end having external bevelling. Fracture lines may be seen radiating from the entrance hole. Secondary fractures occur as a result of raised intracranial pressure. As with other head injuries, these are most commonly seen in the orbital plates. Temporary cavitation will also produce secondary fractures. When the skull has been damaged after death, reconstruction of the skull with superglue will allow identification of an entrance wound.

Exit wounds Bullets passing through tissue tend to become unstable. As they exit, they will therefore tend to produce an irregular wound in comparison with the entrance wound (Figure 6.26), although this is not always the case. As the bullet exits through the skin, it will be pushed outwards, splitting the skin. Exit wounds may be stellate or slit-like. They may on superficial examination mimic a wound caused by another weapon.

Shotgun injuries When a shotgun is fired, the contents of the cartridge, i.e. the shot and wadding, will be discharged from the barrel, along with muzzle gases, burning and unburnt powder, soot, detonator and cartridge material. These all affect the characteristics of the wound, depending on the specific contents of the cartridge and the distance from which the gun is fired (Breitenecker

Figure 6.26 Slit-like exit wound.

1969). The shot may vary from a single slug to multiple small pieces of shot, typically made of lead. The site of the injury also has considerable importance in determining the characteristics of the wound.

Contact wounds Contact shotgun wounds to the head are characteristically selfinflicted and the head is a common site for suicidal wounds. The gun may be placed under the chin or in the mouth or to the temple region. Occasionally, the gun may be placed behind the ear or on the cheek. These wounds to the head have a devastating effect and are often associated with major disruption of the skull and brain, because of the effects of both the shot and the muzzle gases produced. The brain may be extruded through the resulting disrupted scalp and skull (Harruff 1995). When the mouth is the entrance wound, laceration of the lips and bruising may be seen. Examination of the inside of the mouth will typically reveal an entrance wound in the hard palate with soot and powder blackening. If the forehead is chosen as the entrance wound a tangential wound may be the result, with either an exit wound slightly higher on the forehead or even a wedge-shaped wound formed by the coalescence of both entrance and exit wounds. An entrance wound of the occiput is very suspicious of being inflicted by another party, although self-inflicted wounds to this site are encountered. When the muzzle of the gun is held in contact with the skin, a muzzle abrasion will result. When disruption of the skin is extensive and heavily blood stained, it may be difficult to identify features of a contact wound. Suturing together of the wound may allow the features to be observed after the reconstruction.

Near contact and intermediate wounds When a shotgun is discharged beyond 1–2 cm from the body powder, tattooing will occur. As with rifled weapons, the presence or absence of powder tattooing depends on the type of

75

76

Scalp, facial and gunshot injuries

Figure 6.27 Shotgun wound to neck showing petal-shaped ­damage from wadding.

ammunition and length of the barrel. The wadding may cause additional damage. With plastic wadding, after firing it opens into a ‘petal’ formation that may be replicated on skin, depending on the range (Figure 6.27). At close range, the wadding will enter the wound and no identifiable wadding injury will be present. All 0.410 ammunition has three petals and the resulting damage can be seen at a range of 7.5–12.5 cm (3–5 in.). With 12- to 20-bore shotgun ammunition has four petals, marks from which can be seen between 30 and 90 cm (1–3 ft). Wadding may produce damage up to 6 m (20 ft). As the distance between the gun and the target increases, the presence of soot and powder tattooing decrease until they are no longer present and shot will begin to widen, producing a larger entrance hole. The shot fired from a 0.410 will begin to widen at a shorter distance than larger bores. At a distance of about 1 m, the entrance wound from a 12 bore will be of the order of 2.5–4 cm. This will depend partly on the choke of the barrel, as the more choked a barrel, the more together the shot will remain. By 1.3 m the entrance wound is likely to have a ragged appearance and satellite pellets will penetrate the skin around the wound.

Figure 6.28 Shotgun wound with spread shot, including buckshot.

The energy that a bullet releases having passed through the body is therefore determined by the following equation: E=

1 M(V1 − V2)2 2

where V1 is the initial velocity on entering the body and V2 is the velocity of the bullet on exiting the body. Although the velocity of the bullet is an important component of its injury potential, the type of damage done to the body also depends on

Distant wounds Shot will disperse as the distance increases and an increasing peppering of shot will surround the central entrance wound, until there is no central entrance wound, but a spread of shot (Figures 6.28 and 6.29). At these distances, smaller shot is less likely to penetrate, unless it strikes a vulnerable area, such as the orbit.

Wound ballistics Traditionally, wounds have been described as high- or lowvelocity wounds, principally based on the speed of the bullet. Bullets travelling at above 330 m/s (1100 ft/s), the approximate speed of sound in air, are high-velocity bullets, those below this figure, low velocity. The energy imparted is determined by the following equation: E=

1 MV 2 2

Figure 6.29 Skull radiograph of Figure 3.28, showing different sizes of shot (home-made ammunition).

Unusual weapons the construction of the bullet, how intact the bullet remains and on what tissues are involved by the wound tract. Wounds are now referred to as low- and high-energy transfer wounds. Low-energy transfer wounds pass through the skin, lacerating and crushing the skin and the other tissue in its path, but only those tissues that come into immediate contact with the bullet’s path are damaged. No significant damage occurs by energy transfer to adjacent tissues outside the bullet path (Mahoney et al. 2005). In high-energy transfer wounds, as well as the damage caused by the immediate track of the bullet, there are two other methods of damage: shock waves and cavitation injury. With high-energy transfer, as the bullet passes through body tissues, the bullet compresses the tissue in front, and a spherical shock wave is transmitted before the bullet. This shock wave travels at a speed of around 450 m/s (1500 ft/s), the approximate speed of sound in water. Solid tissues such as muscle, brain and liver are particularly susceptible to damage by these shock waves. The shock waves pass along blood vessels, thus transmitting the damage some distance from the track. The third cause of damage in high-energy transfer wounds is temporary cavitation. As the missile passes through the body, the energy is released into local tissues, which are displaced forwards and outwards, forming a large cavity. The cavity exists only for a few milliseconds and is formed under negative pressure and clothing, so bacteria and other external materials can be sucked into the cavity. The brain, liver and muscle, as solid organs, are particularly susceptible to cavitation injury, whereas the lungs, being less dense, are less damaged. At speeds over 760 m/s (2500 ft/s), the tissue loses its elastic quality and the cavitation process becomes truly explosive, because the tissues can no longer recoil after the expansion. Tissue destruction therefore becomes much more severe with bullets above this speed. A stable bullet that passes right through a limb may impart only 10–20 per cent of its energy. An unstable bullet may impart 60–70 per cent of its energy. A bullet that does not exit will impart 100 per cent of its energy to the body. The pattern formation of a temporary cavity depends on the type of bullet. Experiments using material such as soap and gelatine blocks provide information on when cavity formation occurs. With full metal-jacketed bullets such as the standard NATO 7.62 mm military round, a narrow channel is produced, before cavitation begins. In the case of the NATO 7.62 mm, this is of the order of 15 cm (Kneubuehl 2011). A full metaljacketed bullet can pass through a limb before the cavitation phenomenon has developed. With soft-point bullets, such as a soft-point 7.62 mm used for hunting, cavitation will start immediately after penetration of the skin. The first soft-point bullets, produced from the British factory at Dum Dum in India in the 1890s, caused significant damage to the opposing armies. As a result of their controversial nature, such bullets were banned for use in warfare under the Hague Convention of 1899.

Appearance of the brain The wound track in the brain may be associated with contusional damage at the site of entry into the brain. Contusions may also be seen at the site of exit. Raised intracranial pressure, evidenced by gyral flattening and cerebellar tonsillar grooving, can be seen (Kirkpatrick and DiMaio 1978). Three zones of macroscopic damage may be seen in the track of a bullet (Figure 6.30). The centre of the track is a blood-filled

Figure 6.30 Coronal section of cerebral hemispheres showing gunshot wound track. permanent cavity, outside which is an area of haemorrhagic necrotic tissue, with a marginal zone of pinkish-grey tissue. In addition, more remote areas of haemorrhage may occur and subdural and extradural haematoma formation can develop. Microscopic examination of the brains of people who have survived for more than 90 minutes has revealed axonal damage with b-amyloid precursor protein (APP) positivity throughout the cerebral hemispheres and the brainstem (Koszyca et al. 1998). An examination of 20 brains for the effects of temporary cavitation was performed on victims who survived for less than 90 minutes. They had been shot with bullets with a muzzle energy of less than 500 J, 17 being bullets from handguns and 3 from low-velocity rifles. Microscopic examination revealed a zone of astrocyte damage in an area of haemorrhagic extravasation surrounding the permanent track. Further from the permanent track the damage was lessened, with damage seen for a distance of 18 mm (Oehmichen et al. 2000). After passing through the brain, a bullet may exit the skull. If it does not, it may ricochet back into the brain, or more commonly travel around the skull, parallel to the inner table, resulting in a shallow gutter wound track in the cortex of the brain (DiMaio 2016). Any bullet recovered should be carefully handled for subsequent ballistics examination. Metal forceps should not be used because they may mark the bullet, so plastic forceps should be used.

Unusual weapons Air-powered weapons Deaths from air weapons occasionally occur. The head is the most common part of the body penetrated. The entrance site may be through the eye, the temple or the forehead. The wound will be the approximate diameter of the pellet. There may be a small rim of abrasion, but powder burning will not be present, as it would be in a rimfire 0.22 rifle fired at near contact or contact range (Figure 6.31). No skin splitting is seen. If the weapon has been placed against the skin, an impression of the muzzle of the gun may be left. The muzzle impression may vary but, if the gun is available, it should be compared with the wound appearance (Milroy et al. 1998). Air-powered weapons may be modified illegally to become firearms, as may other weapons. These home-made firearms are sometimes referred to as ‘country’ guns or ‘zip guns’ (Book and Botha 1995).

77

78

Scalp, facial and gunshot injuries and 7 from plastic bullets. The rubber was separated from the steel in nine cases. Fatal penetrating injuries were to the head and chest, but in three cases, there was no penetration, the victims being struck in the head and thoracic spine. Mahanja and colleagues (2002) reported on the use of rubber and plastic bullets by Israeli forces against Arab Israelis in October 2000. Of 152 injured patients, 3 died, 2 from penetrating ocular injuries. Fractures of the skull and facial skeleton were recorded.

Humane killers Humane killers are used by veterinary surgeons and in abattoirs to kill larger animals. They are of two main types, either firing a captive bolt or a bullet with a low velocity. Typically, these weapons are used in suicides and show features of a contact injury, including contact abrasion. The captive bolt may leave an injury similar to penetration by a non-firearm-pointed weapon (Hunt and Kon 1962). Figure 6.31 Entrance wound from a 0.22 air rifle pellet.

Crossbow injuries

Rubber bullets

A number of reports have appeared detailing injuries from crossbows (Gresham 1977; Claydon 1993; Downs et al. 1994; Byard et al. 1999). Crossbows can produce fatal injuries that may simulate firearm injuries, especially if the bolt is removed. They also have the ability to produce fatal injuries at considerable distances. Crossbows may fire an arrow at 61 m/s (200 ft/s) with a range of around 270 m. Crossbow arrows may be straight bolts or the arrowhead may have a number of blades attached to it. These two types of arrowhead give different wound appearances. With bolts with blades in the arrow tips, the wound shape can be characteristic, depending on the number of blades present, which may be two, three or four. These blades form incised wounds, which with edges opposed will indicate the nature of the arrow. Crossbow bolts with a circular head may produce circular entrance wounds that mimic a bullet entry wound. If the bolt is removed, considerable difficulty may be encountered, with an entrance wound resembling a bullet wound.

With the increasing problem of urban unrest and the requirement to pacify demonstrators with non-lethal force, various weapons have been developed to hurt, but not significantly injure. In 1970, the British security forces in Northern Ireland introduced the rubber baton (often called the rubber bullet) as a method of riot control. The baton consisted of firm rubber constructed into a cylindrical mass 15 cm long with a rounded head, base 3.5 cm in diameter and weighing approximately 140 g. The baton was designed to be fired at a range of 20–40 m. The object was to strike the lower limbs of the rioters, but the weapons were not accurate. Between 1970 and 1975, 55,000 rubber batons were fired. Three fatalities were attributed to the baton during this period. Rubber batons were replaced as a riot method by plastic baton rounds. These rounds are made of Teflon, constructed as a cylindrical round without tapering, 10 cm long with a flat head and base 3.7 cm in diameter. The round weighs 135 g and has a muzzle velocity of 71 m/s and energy of 325.1 J. Since its first use in 1975, over 60,000 rounds have been discharged in Northern Ireland. Despite its greater stability, the plastic baton has been associated with 14 deaths in Northern Ireland. Its stability tends to result in the baton striking end on and may also cause damage by ricocheting. Ten victims were hit in the head and four in the chest. Typically, these wounds cause an annular abrasion with surrounding bruising. This pattern has apparently resulted in allegations that a ‘D’ or ‘LR20’ battery has been used as a projectile (Crane 2000). The Israeli military have used both plastic and rubber bullets that are of a different construction to those used by the British. Hiss and colleagues have reported on these weapons (Hiss and Kahana 1994; Hiss et al. 1997). The plastic bullet weighs 0.85 g in a 5.56 × 45 mm cartridge. The bullet is made of polyvinyl chloride and is discharged with a muzzle velocity of 1250 m/s and energy of 663.7 J. Rubber bullets have also been used. There are four types: two spherical and two cylindrical. The spherical bullets are 1.8 cm in diameter. One is wholly made of rubber, as is one of the cylindrical bullets. The other two types are made of steel with a rubber shell. The rubber bullets weigh 8.3 g and have a muzzle velocity of 75–100 m/s and a kinetic energy of 23.3–41.5 J. The rubber and steel bullets weigh 15.4 g with a muzzle velocity of 100 m/s and energy of 77 J. Hiss and colleagues reported 10 deaths from these rubber bullets

Defence injuries Defence injuries are wounds to the hands, arms and less commonly legs, produced when the victim tries to ward off blows with a weapon or with fists or feet. Firearms may also be associated with defence wounds. The identification of these wounds is important because they indicate that the victim was able to defend him- or herself for at least part of the attack. These injuries are often inflicted when the hands and arms are used to cover the head and may indicate that more blows were aimed at the head than is apparent from the number of injuries found there.

Histology of skin wounds Considerable effort has been invested into using microscopic examination to date injuries (Raekallio 1980; Betz 2003; Parai and Milroy 2015). After an initial open injury, the first change seen is margination of neutrophils, which can occur within 30 minutes or so and continues for a few hours. By 4 hours, neutrophils begin to infiltrate tissue, followed by mononuclear cells. For the first 24 hours, neutrophils predominate, but over the next 24–28 hours, mononuclear cells increase. Fibroblast proliferation begins after 48–72 hours. Fibrin is seen, which at 16 hours will stain red with Martius scarlet blue; before then

Table of cases it stains yellow. At 2–4 days, a wound will have fibroblasts at the wound periphery and new capillary growth is seen at about 4–5 days. The epithelialisation of an abrasion or small wound may be complete in 3 days. Over the next few days, fibroblasts decrease as collagen is laid down. The epithelium decreases in thickness by 5–7 days. Healing is completed by decrease in vascularity, cellular regression and restoration of collagen. The epithelium regains a basement membrane at 14 days. Bruises typically have a neutrophil inflammatory response after about 4 hours. Red blood cells may be seen intact in wounds at this and later stages, so the presence of intact red blood cells should not be interpreted as a fresh injury. Haemosiderin begins to be seen in wounds on day 3, and this can be a useful indicator of age, although previous injury at the same site should be considered, when haemosiderin might still be present. Haematoidin is occasionally found but does not appear until around day 9. A considerable amount of work has been published on the histochemical and immunohistochemical findings in wound healing. Antibodies to a variety of antigens have been used. Among these are fibronectin, which can be seen in a wound as early as 20 minutes, but routinely after 4 hours, the interstitial cell adhesion molecule (ICAM-1) after 90 minutes and selectins, which can be positive at 1 hour. Antigens marking later stages, such as fibroblast proliferation and collagen formation, can be used. Grellner (2002) examined the role of pro-inflammatory cytokines in human skin wounds. Interleukin-1b (IL-1b) and IL-6 showed enhanced expression after 15–20 minutes with increase in epidermal activity, with a marked increase after 30–60 minutes with IL-1b and 60–90 minutes with IL-6. Leukocytes reacting to these ILs were seen after 2 hours. Tumour necrosis factor alpha (TNF-α) showed changes at 15 minutes with marked enhancement at 90 minutes. These techniques remain essentially research tools rather than of practical value for casework, although they provide potential for more accurate delineation of timing. For a detailed discussion, the reader should consult the original works (Betz 2003).

References Ainsworth, R.W. and Hunt, A.C. 1993. The enemy within. Med Sci Law, 33, 358. Besant-Matthews, P.E. 2000. Examination and interpretation of rifled firearm injuries. In Mason, J. and Purdue, B.N. (eds), The Pathology of Trauma, 3rd edn. Arnold. Betz, P. 2003. Pathophysiology of wound healing. In Payne-James, J., Busuttil, A. and Smock, W. (eds), Forensic Medicine. Clinical and Pathological Aspects. Greenwich Medical Media, pp 83–89. Book, R.G. and Botha, B.C. 1995. Zulu zip-guns and an unusual murder. Am J Forensic Med Pathol, 15, 319–324. Breitenecker, R. 1969. Shotgun wound patterns. Am J Clin Pathol, 52, 269–285. Burke, T.W. and Rowe, W.F. 1992. Bullet ricochet: a comprehensive review. J Forensic Sci, 37, 1254–1260. Byard, R.W., Koszyca, B. and James, R. 1999. Crossbow suicide: mechanisms of injury and neuropathologic findings. Am J Forensic Med Pathol, 20, 347–353. Claydon, S.M. 1993. A bolt from the blue. Med Sci Law, 33, 349–350. Crane, J. 2000. Violence associated with civil disorder. In Mason, J. and Purdue, B.N. (eds), The Pathology of Trauma, 3rd edn. Arnold. DiMaio, V.J.M. 2016. Gunshot Wounds. Practical Aspects of Firearms, Ballistics and Forensic Techniques, 3rd edn. CRC Press. Downs, J.C.U., Nichols, C.A., Scala-Barnett, D. and Lifeschultz, B.D. 1994. Handling and interpretation of crossbows injuries. J Forensic Sci, 39, 428–445. Garner, B.A. (eds) 2019. Black’s Law Dictionary, 3rd edn. (Standard Edition). Thomson Reuters.

Grellner, W. 2002. Time-dependent immunohistochemical detection of proinflammatory cytokines (IL-1b, IL-6, TNF-α) in human skin wounds. Forensic Sci Int, 130, 90–96. Greenes, D.S. and Schutzman, S.A. 2001. Clinical significance of scalp abnormalities in asymptomatic head-injured infants. Pediatr Emerg Care, 17, 88–92. Gresham, G.A. 1977. Arrows of outrageous fortune. Med Sci Law, 17, 239–240. Grossman, S.E., Johnston, A., Vanezis, P. and Perrett, D. 2011. Can we assess the age of bruises? An attempt to develop an objective technique. Med Sci Law, 51(3), 170–176. Harruff, R.C. 1995. Comparison of contact shotgun wounds of the head produced by different gauge shotguns. J Forensic Sci, 40, 801–804. Hempling, S.M. 1981. The application of ultraviolet photography in clinical forensic medicine. Med Sci Law, 21, 215–222. Hiss, J. and Kahana, T. 1994. The fatalities of the intifada (uprising): the first five years. J Forensic Sci Soc, 34, 225–229. Hiss, J., Hellman, F.H. and Kahana, T. 1997. Rubber and plastic ammunition lethal injuries: the Israeli experience. Med Sci Law, 37, 139–144. Hunt, A.C. and Kon, V.M. 1962. The pattern of injury in humane killers. Med Sci Law, 2, 197–202. Kirkpatrick, J.B. and DiMaio, V.J.M. 1978. Civilian gunshot wounds of the brain. J Neurosurg, 49, 185–198. Kneubuehl, B.P. (ed) 2011. Wound Ballistics: Basics and Applications. Berlin: Springer Science & Business Media. Koszyca, B., Klombergs, P.C. and Manavis, J. et al. 1998. Widespread axonal injury in gunshot wounds to the head using amyloid precursor protein as a marker. J Neurotrauma, 15, 675–683. Krauss, T.C. and Warlen, S.C. 1985. The forensic science use of reflective ultraviolet light. J Forensic Sci, 30, 262–268. Kury, G., Weiner, J. and Duval, J. 2000. Multiple self-inflicted gunshot wounds to the head: report of a case and review of the literature. Am J Forensic Med Pathol, 21, 32–35. Langlois, N.E.I. and Gresham, G.A. 1991. The ageing of bruises. A review and study of the colour changes with time. Forensic Sci Int, 50, 227–238. Mahanja, A., Aboud, N. and Harbaji, I. et al. 2002. Blunt penetrating injuries caused by rubber bullets during the Israeli-Arab conflict in October 2000: a retrospective study. Lancet 359, 1795–1800. Mahoney, P.F., Ryan, J., Brooks, A.J. and Schwab, C.W. (eds) 2005. Ballistic Trauma: A Practical Guide. London: Springer Science & Business Media. Maguire, S., Mann, M.K., Sibert, J. and Kemp, A., 2005. Can you age bruises accurately in children? A systematic review. Arch Dis Child, 90(2), 187– 189. Milroy, C.M., Clark, J.C., Carter, N., Rutty, G. and Rooney, N. 1998. Air weapon fatalities. J Clin Pathol, 51, 525–529. Mosqueda, L., Burnight, K. and Liao, S. 2005. The life cycle of bruises in older adults. J Am Geriatr Soc, Aug; 53(8), 1339–1343. Oehmichen, M., Meissner, C. and Konig, H.G. 2000. Brain injury after gunshot wounding: morphometric analysis of cell destruction caused by temporary cavity. J Neurotrauma, 17, 155–162. Parai, J.L. and Milroy, C.M. 2015. Histological aging of bruising: a historical and ongoing challenge. Acad Forensic Pathol, 5(2), 266–272. Polson, C.J., Gee, D.J. and Knight, B. 1985. The Essentials of Forensic Medicine, 4th edn. Pergamon Press, 102–104. Raekallio, J. 1980. Histological estimation of the age of injuries. In Perper, J.A. and Wecht, C.H. (eds), Microscopic Diagnosis in Forensic Pathology. Charles C. Thomas. Stephenson, T. and Bialas, Y. 1996. Estimation of the age of bruising. Arch Dis Child, 74, 53–55. Teare, R.D. 1961. Blows with the shod foot. Med Sci Law, 1, 429–436. Wilson, E.F. 1977. Estimation of the age of cutaneous contusions in child abuse. Pediatrics, 60, 750–752.

Table of cases C (a minor) v Eisenhower [1984] QB 331, 78 Cr App Rep 48, DC. M’Loughlin [1838] 8 C & P 635. Moriarty v Brookes [1834] 6 C & P 684. Wood [1830] 1 Mood CC 278.

79

7

Adult skull fractures Helen Whitwell and Philip Lumb

A fracture can be defined as an abnormal break in the continuity of a structure such as a bone produced by stress and strain. Despite a ‘lay view’, skull fractures do not generally cause death directly but are more often an indicator of force applied. It is the underlying brain injury that causes death. Skull fractures are most commonly caused by direct force and less commonly by indirect force such as the ring fracture to the posterior fossa caused by a fall from a height. The biomechanics of skull fractures is complex. Deformation of the skull occurs with force and if the limit of the elasticity of the skull is exceeded, then fractures will occur. With a focal impact, this leads to a depressed skull fracture and with a broader impact, linear fracturing occurs. As a generalisation, the more severe the fracture, the greater the force and the greater the injury to the underlying brain (Leestma 2014). The biomechanics of skull fractures is covered in detail in Chapter 4.

Documentation of skull fractures Precise documentation is essential. When considering skull fractures, there are several features that must be observed. Scene information and a scene visit with the body in situ, and interpretation as to the assault/accident scenario with identification of weapons or sites of potential impact, may greatly assist. Assessment of clothing, in particular items worn on the head, which may alter the appearances of scalp/skull injuries, should also be done. Radiology is indicated in certain circumstances, e.g. in firearms injuries, penetrating wounds or ‘hit-and-run’ fatalities. It is especially useful in identification of foreign bodies and evidence of bony injury. In those individuals with survival in hospital, review of medical records, including radiology, should be done. As with all injuries, fractures should be photographed, from both an extracranial and an intracranial viewpoint with a scale. After photography of all external injuries associated with both the head and elsewhere to the body, the scalp should be reflected both anteriorly and posteriorly. Before the temporalis muscles are dissected and the periosteum scraped, the skull should be photographed from four quadrants, including two left and two right, front and back images from both sides. This enables documentation of deep bruising associated with fracturing. Preparation of the fracture sites can then be made for more detailed close-up photography. The periosteum should be scraped away with a sturdy post-mortem knife (such as a PM40) and cleaned of blood. Temporalis muscles should be removed in order to allow fracture extensions beneath this muscle to be readily identified. Where necessary, the scalp should be reflected as far as is possible to reveal the course of the fracture. This may involve removal of the facial skin to show continuity with facial fractures or further reflection of the occipital scalp into the neck. It may be difficult to photograph skull fractures that extend into the occipital region and is facilitated when the body is laid on its

front. All photography should be done with and without measurement scales. After adequate photographic documentation, it is the author’s preference to draw and measure the fractures either diagrammatically or using computer graphics. During the removal of the skull cap, there may be collapse of certain areas and fall-out of bone segments, which should all be documented. The skull should be drilled so that the dura mater is kept intact and if possible fracture lines avoided. A quick photograph just before the skull is removed can identify any haematoma in its original position as they may slip out as soon as the cap is removed. If the dura/meninges are still intact, the location of a haematoma can be readily assessed and compared with the location of the fracture site and underlying brain injury. After removal of the brain, the dura around the base of the brain and inside the remaining calvarium should be removed carefully. In younger individuals, traction with soft tissue paper can do most of the stripping. In elderly individuals, the dura mater tends to adhere more to the skull and dural strippers may be employed. Once all the dura has been removed, the inside of the cranium should be carefully examined and again the fracture lines documented and photographed. Gentle traction on the skull will open up any fracture lines and make them more readily visible. After documentation, measurements of skull thickness may be taken, although its value is questionable. The adult cranium varies in thickness with thin/weaker areas supported by the ‘buttresses’, e.g. occipital protuberance and sagittal ridge. Most pathologists take measurements from the frontal and parietal (6-10mm) bones, temporal bone (4mm) and occipital bone (15 mm) (Saukko and Knight 2016). The thicknesses can be documented along with photography in the post-mortem report along with normal variables. In complex cases of skull fracture, reconstruction may aid interpretation. The method for this is given in Appendix 1 and illustrated both prior to and following reconstruction (Figure 7.14a and b).

Linear skull fractures Linear fractures occur with a broad impact such as ground contact or from a flat object. They account for around 70 per cent of skull fractures. Following impact there is inbending of the skull at the site of impact with outbending peripheral to the impact site. A linear fracture is initiated by outbending of the skull which then usually extends to the impact point as well as in the opposite direction towards areas of structural weakness, in particular towards the base of the skull. Gurdjian studied deformation patterns following blunt impact using a ‘stresscoat’ method which utilised a strain sensitive lacquer applied to the skull which cracks as a result of tension stress (Gurdjian et al. 1950). He identified the first area of outbending to occur as the area of primary stress level. With increasing force of impact, additional linear fractures may occur at secondary and tertiary stress areas.

Linear skull fractures

Occipital Simple linear fractures of the occipital skull are most commonly seen in falls which may be accelerated or unprotected such as when an individual is intoxicated or unconscious from a direct blow to the front of the face. In the elderly age group, who may be at risk of ‘syncope’, this may result in an unprotected fall on to the back of the head. In forensic practice, the acceleration is usually provided by means of a blow to the front of the face such as a punch. The point of contact of the blow should be investigated by examination of the injuries to the face and also deep dissection of the facial tissues. Facial fractures may occasionally be identified as a result of this direct blow, in association with a variety of other features of blunt force trauma including lacerations and bruising. Acceleration to a fall is provided by the delivery of the blow to the face. Acceleration may also occur as a result of individuals pulling on an object, such as a handbag, with loss of grip on the object providing sudden acceleration. Typically the head is accelerated on to a firm surface such as the pavement or a tarmac road. An elderly individual is more at risk of developing intracranial haematomas, particularly subdural, most probably associated with an age-related decrease in cerebral volume (Leestma 2014, Hartshorne et al. 1997). Externally, the scalp may show evidence of blunt force impact with combinations of laceration, bruising and abrasion. A variety of linear fractures may be produced in the occipital bone – vertical fractures extending into the occipital suture line or horizontal fractures curving towards the base of the brain (Figures 7.1 and 7.2). The fractures tend to avoid the strengthened buttresses of the internal occipital protuberance (Saukko and Knight 2016). The linear fracture may continue straight through the occipital suture and extend into the base of the skull into the posterior fossa, where it often follows a complex path to the anterior fossa across the occipital suture line with dissemination of the energy of the impact away from the primary impact site. When springing of the sutures is identified, further fracturing may be seen extending into the posterior fossa away from the primary impact site. Typically in such accelerated falls the brain shows a contre-coup pattern of contusion and laceration (see Chapter 10). Contre-coup orbital plate fractures are not infrequently seen in this scenario (see later).

Figure 7.1 Occipital fracture after a fall on the back of the head.

Figure 7.2 An unusual occipital fracture involving the internal occipital protuberance.

Linear squamous temporal bone fractures The squamous temporal bone is particularly vulnerable to fracturing as a result of its thin, unsupported structure. This most often occurs as a result of direct trauma such as a blow by an object (golf ball, bat) or a fall. Fractures to this region may injure the middle meningeal artery, which courses within the cranial cavity on the deep aspect of this bone (Figure 7.3). Injury to this artery causes extradural haematoma formation (see Chapter 8).

Linear frontal fractures Linear frontal fractures are also encountered in the context of accelerated falls to the front of the head. As with occipital

Figure 7.3 Linear squamous temporal fracture in association with underlying acute extradural haematoma.

81

82

Adult skull fractures fractures may also involve the hard palate. Again, as with occipital fractures, it is important to establish the point of contact for the initial blunt force trauma. Occasionally a frontal fracture may represent a secondary impact, after injury to the posterior scalp. These accelerating forces may be associated with a depressed fracture which may be comminuted.

Front-to-back hinge fracture

Figure 7.4 Frontal fracture extending into the anterior cranial fossa. fractures, they are commonly associated with external injuries such as lacerations and bruising. Frontal fractures are most commonly vertically oriented and extend towards the face inferiorly and the parietal skull posteriorly. Most fractures remain ipsilateral to the point of contact with the ground. However, it is not uncommon to find that the fractures curve posteriorly and cross the midline away from the region of impact. They may also terminate within the sutures, causing these sutures to spring (frontal or sagittal). With linear fractures of the frontal bone, shorter emanating fractures may be identified. Inferiorly the fracture may extend in a variety of directions: into the roof of the orbit through the supraorbital ridge, posteriorly across the anterior cranial fossa (Figure 7.4). The fracture may also extend down the inside of the orbit across its medial wall, through the lacrimal bone and down into the maxilla (Figure 7.5). These

Figure 7.5 Involvement of facial bones, with a fracture extending from a frontal fracture into the orbit continuing to the maxilla.

A frontal fracture may form part of an injury complex creating a ‘front-to-back’ hinge fracture (Figure 7.6). This occurs when there is a high-energy impact over the occipital skull, which drives the front of the head rapidly into a firm surface. This may occur in pedestrian road traffic incidents. If a pedestrian is hit by a car bumper or side mirror travelling at speed, the primary impact will create a large comminuted, depressed or mosaic fracture (see below). The energy from this primary impact may drive the front of the face with considerable velocity into the pavement or roadside. This causes the secondary impact and creates a frontal fracture. With the high energies involved, both the primary impact fracture and the secondary impact (frontal) fracture may extend towards the base of the skull. Both fractures may terminate within the foramen magnum. Removal of the skull cap post-mortem allows the two halves of the skull to form a ‘front-to-back’ hinge fracture (see below). The primary and secondary impact sites may also communicate within the calvarium as the posterior extension

Figure 7.6 Front-to-back fringe fracture.

Puppe’s rule of the frontal fracture (secondary impact) communicates with the anterior extensions of the primary fracture complex. When there is a primary and secondary impact site, such as in the situation described above, the secondary impact site with the frontal fracture is usually on the contralateral side to the primary impact site, because the primary impact site offset from the midline drives the skull away from the region in which the forces apply.

Puppe’s rule Puppe’s rule can be used to assess the order in which fracturing occurred. Evidence for two separate episodes of blunt force trauma should be corroborated by external features. After injury, the first fracture is formed. Subsequent fractures caused by additional blunt force trauma will terminate and not cross in this first fracture line (Saukko and Knight 2016). Figure 7.7 gives an example of such an injury.

Depressed skull fractures These are caused by focal impact to the skull. Indentation of the skull may occur without fracturing – with increasing degrees of force then the inner table may fracture, then both the inner and outer tables and with greater force both comminuted fractures and depressed fractures may occur (Shapiro et al. 1988). The bone fracture may rupture the underlying meninges and force may be transmitted directly to the brain with cerebral contusion or laceration (Figure 7.8). The force is delivered over a small well-defined area and may mirror the shape of the impacting object, although with tangential blows they may appear wedge shaped. The pattern of external injury to the scalp may also be useful in defining the properties of the object used. In forensic practice, blunt implements such as hammers, rocks and baseball bats are among the more common weapons. Depressed fractures may also be seen in vehicular incidents, either where there is contact with a protruding object in the vehicle or a pedestrian is struck by a part of a vehicle.

Figure 7.8 Depressed skull fracture (hammer).

Pond fracture A pond fracture is a shallow depressed fracture forming a ‘concave pond’. It is more common in the infant’s pliable skull where true fracturing may not actually occur (Saukko and Knight 2016), and the skull shows a concave depression.

Mosaic fracture A mosaic fracture comprises a comminuted, depressed area of fracturing with linear fractures radiating from an apparent central region and with intercepting fractures between the radiating arms of the fracture, which give the complex appearance described as a spider’s web. The degree of depression may be minimal or absent. Plates of free-floating bone are created by the intersection of radiating fractures and intersecting fractures. At the centre of the region of impact, these intersections are closer together, producing smaller free-floating areas. Further away from the epicentre of impact, the free-floating segments are somewhat larger and eventually disappear as the radiating arms of the fracture extend outwards. Radiating arms of the fracture may be extensive and extend into the base of the skull and may also communicate with secondary fractures caused by secondary impacts. Fractures of this nature are caused by severe impact with objects causing focal and general skull deformation. This may include vehicle parts such as bumpers, wing mirrors or other presenting parts of a vehicle impact. This type of fracture has also been reported in injury due to a drone (Cheung et al. 2017). It may also be identified as the primary impact site in a fall from a height. It is not usually encountered as a consequence of an accelerated fall such as that caused by a punch to the face, unless the head contacts with such an object as described above. As a result of the broad nature of the impact, the scalp injury associated with these extensive fractures may be non-specific and include small lacerations and abrasions. However, on the deep aspect of the scalp, more extensive bruising is usually present.

Ring fractures Figure 7.7 Puppe’s rule: the second, later fracture, (vertical) terminates in the first earlier fracture.

Ring fractures are encountered in the context of falls from a height. They represent the delivery of energy to the base of the

83

84

Adult skull fractures brain through the upper cervical vertebrae. Ring fractures can also be caused by impact to the top of the head. The fracture encompasses the foramen magnum, occasionally making this free-floating, although most fractures are partial. Examination of the spinal column may reveal fracturing to the vertebral bodies, and it is essential to examine the lower limbs and pelvis thoroughly to assess (if possible) that the primary impact site was either on the feet or about the buttocks. Ring fractures may be encountered in those individuals who sustain impact to the vertex of the skull. They are also encountered occasionally in the context of road traffic incidents, particularly involving motor cyclists. However, they are usually combined with a hinge fracture in this context (Mason and Purdue 2000). Isolated ring fractures may also be caused by hyperflexion/ hyperextension of the cervical spine. The ligamentous attachments about the base of the skull are biomechanically stronger than the base alone. Violent swinging of the skull can cause the base of the brain to fracture, forming a ring around the foramen magnum. The mechanism of fracture may be enhanced by a heavy helmet, such as that worn by motorcyclists (Konrad et al. 1996). Post-mortem CT imaging has assisted in identifying detailed fractures associated with motorcycle incidents (Bell et al. 2017).

Hinge fractures The rare ‘front-to-back’ hinge fracture is described above. The classic hinge fracture involves an injury that extends across the base of the skull from left to right (Figure 7.9). After removal of the calvarium, the anterior and posterior parts of the skull may be parted to form two halves. Typically the hinge fracture extends through the petrous temporal bones, across the pituitary fossa and into the opposite petrous temporal bone. These fractures are most commonly encountered in road traffic incidents (see Chapter 17) including injury from an airbag (Perez and Palmatier 1996). They may also be encountered within the context of falls from a height with primary impact about the vertex as well as encountered in the context of blunt force trauma by an object or implement driven into the skull.

Transmitted fractures/contre-coup fractures Transmitted fractures or contre-coup fractures are away from the primary impact site and comprise isolated linear or

Figure 7.9 Hinge fracture.

Figure 7.10 Contre-coup fractures as a result of a fall on to the occiput. occasionally complex areas of fracturing (Figure 7.10). These are most commonly encountered about the cribriform plate or orbital roofs (Hein and Schultz 1990; Asano et al. 1995). Other sites include thin bone over pneumatised spaces, such as the lacrimal bones, sphenoid and frontal bones. They are most frequently encountered in falls on to the back of the head and associated with a fracture at the primary site of impact. Intracranial pressure produced by missiles such as bullets travelling across the skull can also produce the contre-coup fracture. The differential diagnosis of the isolated anterior cranial fossa fracture includes penetrating injuries through the orbital roof as well as direct trauma. Fragments of glass, after a road traffic accident, may penetrate the orbit and orbital roof. Radiological examination may identify glass, but one should be aware that not all glass fragments are identified by this means.

Compound elevated skull fractures These are rare. Causes include assault by a sharp weapon and road traffic incidents (Prasad and Anmol 2017).

Histology of skull fractures The healing of skull fractures proceeds differently to that of long bone fractures. Little information is currently available in the medical literature on the histology of skull bone healing. The internal periosteum is replaced with dura mater, which contains few osteoblasts. No significant external callus forms in the healing of skull fractures. This may be an evolutionary adaptation, because a large callus may create a space-occupying lesion, displacing the cerebral contents. Dating the age of a fracture can be difficult and usually only an estimate can be given based on histology alone (Leestma 2014). Healing of fractures may be influenced by a variety of factors including infection, hypoxia, fracture stability, interposition of tissue and severity of injury. Dating the age of the injury may be aided by examination of associated injuries such as underlying haematomas or scalp bruising/laceration (Figure 7.11). After a fracture, bleeding occurs immediately, from the injured marrow cavity, torn dura, external periosteum and cortical vessels. Bone necrosis occurs about the edges of the fracture within a few hours of the fracture occurring. The haematoma organises and may disappear after a week or so (Figure 7.12) and fibrosis occurs (Robbins et al. 2020). Callus

Medico-legal issues

Figure 7.11 Haemosiderin staining over healing fracture – may be more useful in dating the injury. Figure 7.13 Bruising above the mastoid process, Battle’s sign.

This includes bruising about the mastoid process (Battle’s sign) (Figure 7.13) and also bilateral periorbital bruising. Bleeding may also be identified emanating from the nose and both ears. These features should not be mistaken for other primary impact sites. If there is doubt that bruising about the eye is caused by direct trauma and if clinically appropriate (such as survival for a period of time), the eye should be formally examined for evidence of direct trauma. Features include commotio retinae and traumatic changes to the lens.

Post-traumatic epilepsy Figure 7.12 Haematoma within a skull fracture showing early organisation.

develops, which comprises a mixture of fibrosis, osteoblasts and osteoclasts. Active bone remodelling, with the presence of bone resorption and osteoblastic synthesis of bone matrix, can be identified. New bone may be visible at approximately 1–2 weeks. After bone remodelling, it may not be possible to identify a healed fracture site. CT scanning has been shown to be useful over conventional radiology in dating. However, histology has been shown to be superior and a combination of radiology and histology most accurate (Cappella et al. 2019).

Medico-legal issues Bruising Fractures to the base of the brain are commonly associated with additional bruising away from the scalp impact sites.

Epilepsy particularly occurs with depressed fractures and may be delayed in onset for survival months or longer. Detailed neurological review may be necessary to link a head injury with the epilepsy which may cause sudden death at some time distance following injury. General complications of head injury and post-traumatic epilepsy are covered in Chapter 2.

Infections Infections including meningitis (most commonly pneumococcus or staphylococcus) and cerebral abscess may occur either as a direct result of contamination from an external wound such as a penetrating gunshot wound – in particular high-velocity wounds which cause extensive tissue damage, via fractures of the anterior cranial fossa which may be contre-coup fractures or from air sinuses. Infection is less common with simple linear fractures as opposed to compound fractures including compound depressed fractures. In some cases, presentation may be delayed for several weeks or even longer. See also Chapters 2 and 18.

85

Adult skull fractures

86

(a)

(b)

  Figure 7.14 Complex skull fracture as a result of a gunshot wound to the head. (a) prior to reconstruction. (b) following reconstruction. (Courtesy of Professor J Crane.)

Appendix 1 Reconstruction of skull (Courtesy of Professor J Crane) 1. Fix all fragments of bone plus any scalp, dura or blood clot in 10 per cent formalin. 2. Rinse carefully in running water. Take care not to lose small fragments. 3. Carefully scrape all fragments clean using a scalpel blade. Palpate all tissues for possible bone fragments. 4. Wash in water. 5. Immerse in dilute hydrogen peroxide to bleach bone white (approximately 5 per cent 20 volume strength). Leave until bone is white – usually up to 1 week. 6. Wash in water. 7. Immerse in absolute alcohol for 2–4 hours to dehydrate. 8. Dry on tissue and place in chloroform to degrease. Change chloroform regularly and check bone for degreasing. Time needed varies depending on amount of fat present. On average the process will take up to 3 weeks. 9. Dry bone with tissue and place in 60°C incubator for 1 hour. Wet areas indicate presence of fat. 10. When degreased and dried, remove any remaining fragments of tissue, blood etc. 11. Initially reassemble skull using sellotape to tack pieces together. In severe cases of fragmentation, additional aids can be used such as drawing a map plan of the pieces or marking the interior of each piece with a reference mark, letter or number. 12. The pieces, if necessary, can be joined using minute amounts of cyanoacrylate glue but only spot join with this or the fractures cannot be filled in later on. Make use of blood vessel tracks, suture lines or bone thickness to aid reassembly. 13. Use self-hardening model plastic to join the fragments and show the fractures. Make up a watery solution and introduce into the fractures using a pipette. When set (in a few minutes) the excess can be scraped away.

14. When the skull has been reconstructed rub down with sandpaper. 15. Finally the skull is given several coats of clear matt ­varnish. A fume cupboard with appropriate extraction should be used as necessary.

References Asano, T., Ohno, K. and Takada, Y. et al. 1995. Fractures of the floor of the anterior cranial fossa. J Trauma, 39, 702–706. Bell, C., Prickett, T.R.A. and Rutty, F.S.I.G.N. 2017. PMCT images of a motorcycle helmet-associated fracture. Forensic Sci Med Pathol, 13(4), 511–514 Cappella, A., Cammilli, P. and de Boer, H.H. et al. 2019. Histologic and radiological analysis on bone fractures: estimation of post-traumatic survival time in skeletal trauma. Forensic Sci Int, 302, 109909. Cheung, Y., Chung, L.K. and Lagman, C. et al. 2017. Skull fracture with effacement of the superior sagittal sinus following drone impact. A case report. Childs Nerv Syst, 33(9), 1609–1611. Gurdjian, E.S., Webster, J.E. and Lissner, H.R. 1950. The mechanism of skull fracture. J Neurosurg, 7, 106–114. Hartshorne, N.J., Harruff, R.C. and Alvord, E.C. 1997. Fatal head injuries in ground-level falls. Am J Forensic Med Pathol, 18, 258–264. Hein, P.M. and Schultz, E. 1990. Contrecoup fractures of the anterior cranial fossae as a consequence of blunt force caused by a fall. Acta Neurochir, 105, 24–29. Konrad, C.J., Fieber, T.S., Schuepfer, G.K. and Gerber, H. 1996. Are fractures to the base of the skull influenced by the mass of the protective helmet? A retrospective study in fatally injured motorcyclists. J Trauma, 41, 854–858. Leestma, J.E. 2014. Forensic Neuropathology. Baca Raton: CRC Press. Mason, J.K. and Purdue, B.N. (eds) 2000. The Pathology of Trauma, 3rd edn. Arnold. Perez, J. and Palmatier, T. 1996. Air bag-related fatality in a short, forward positioned driver. Ann Emerg Med, 28, 722–724. Prasad, G.L. and Amnol, N. 2017. Compound elevated skull fractures: review of literature. [Review].Brain Inj, 31(4), 434–439. Robbins, S.L., Cotran, R.S.A., Kumar, V., Abbas, A. and Aster, J. 2020. Pathologic Basis of Disease, 10th edn. Philadelphia: Elsevier Saukko, P. and Knight, B. 2016. Knight’S Forensic Pathology, 4th edn. CRC Press. Shapiro, H.A., Gordon, I. and Benson, S.D. 1988. Forensic Medicine – A Guide to Principles, 3rd edn. Churchill Livingstone.

8

Intracranial haematomas Extradural and subdural Helen Whitwell

This chapter covers the various types of intracranial haematomas including extradural (epidural) haematoma (EDH) and subdural haematoma (SDH). Subarachnoid haemorrhage is covered in Chapter 9, intracerebral traumatic and non-­t raumatic haemorrhages in Chapter 10 and forensic anatomy in Chapter 1.

Extradural haematoma This is also known as an epidural haematoma. Bleeding in this condition takes place between the skull and the dura. This type of haemorrhage is found in between 5 and 22 per cent of fatal head injuries (Smith et al. 2015). Most commonly it occurs in association with a laceration or tear of the posterior branch of the middle meningeal artery beneath the squamous part of the temporal bone in association with a skull fracture. Less commonly bleeding may be venous in origin particularly in the posterior fossa. In children, skull fracture is less common. It is thought that this relates to the flexibility of the child’s skull, causing deformation rather than a fracture, with consequent evulsion of the dura from its undersurface and tearing of the blood vessel (Mealey 1960; Mazza et al. 1982). A significant minority of cases of EDH (between 20 and 30 per cent) occur elsewhere, including the frontal region and the posterior fossa. Children have a higher incidence of EDH in these locations (Koc et al. 1998). With increased use of neuroradiology, it is now recognised that a ‘pure’ EDH is less common than previously supposed. Imaging studies have shown that around 50 per cent of cases of EDH are associated with SDH and/or contusional brain injury (Phonprasert et al. 1980; Servadei et al. 1995a). The morbidity/mortality is affected not only by the presence of other intracranial pathology – in particular contusional injury and/or traumatic axonal injury – but also by delay in diagnosis/operative intervention (Servadei et al. 1995b; see Chapters 10 and 11).

Circumstances and clinical features EDH occurs as a result of focal impact to the head. The most frequent causes are falls, road traffic incidents, in which there is a higher incidence of other brain injuries, and increasing assaults. The clinical presentation is classically a history of head injury followed immediately by a period of concussion – this may be brief. This is followed by recovery – a lucid interval. It is during this period that blood accumulates within the cranial cavity to give rise to the features of a space-occupying lesion, with features of raised intracranial pressure and eventual lapse into coma. The time course for the lucid interval is variable ranging from several minutes to hours to several hours but may rarely be of day’s duration (see Chapter 2). The presence of other brain injury may also modify the clinical picture. Death occurs in EDH primarily as a result of increasing pressure within the cranial cavity and brainstem compression. It is,

however, becoming increasingly recognised that the morbidity and mortality are affected not only by the presence of other intracranial pathology (shown earlier) but also by delay in diagnosis and surgical intervention (Bricolo and Pasut 1984). Early scanning immediately after head injury may fail to pick up the development of an EDH (Mandavia and Vilagomez 2001; Chapter 2).

External injury This can be variable depending on the nature of the impact. The hair may both hide and alter the external injury. Abrasions are classically seen with ground contact. Lacerating injury may be seen where injury has occurred with an implement such as a stick, hammer or other weapon. Unless there are specific characteristics relating to the implement in question, it is usually impossible to be dogmatic in linking a particular weapon to the injury from the wound appearances alone. It may also be impossible to differentiate ground contact and a single weapon impact if the wound is a simple bruise or laceration (see Chapter 6).

Pathological appearances As this results from a focal impact, the external injury should correlate well with bruising of the scalp tissues and any fracture beneath. Examination of the skull and brain most commonly shows a fracture in the squamous part of the temporal bone. The skull in this area is thin and fracturing occurs relatively easily. In around 15 per cent of cases, no associated fracture is present. Fractures are less commonly seen in other locations. The usual site for EDH is the temporoparietal area. The dura is stripped from the inner aspect of the skull and becomes separated by the haematoma. As the haematoma enlarges, it forms an ovoid ‘bun’-shaped mass (Figure 8.1); this progressively indents the adjacent brain leaving, on removal, a concave depression. The volume of the clot varies – in fatal cases, it is usually at least 75–100 mL but can be greater (Leestma 2014). Measurement, particularly in unsuspected cases, may be difficult. With progression the typical features of raised intra-cranial pressure with herniations and brainstem compression occur. If assisted ventilation has taken place, the changes of ‘respirator’ brain may be superadded. Areas of contusional injury may be seen either in association with a fracture site or beneath an area of impact. In cases where there is additional brain injury, SDH with other areas of contusional/lacerating injury may be identified.

Histology Histological examination of the dura and haematoma is rarely of assistance in cases of EDHs. However, it is recognised that between 13 and 15 per cent of these may have a more chronic time course (Tatagiba et al. 1989; Viljoen and Wessels 1990). In such cases, histological assessment may show organising

88

Intracranial haematomas weeks. Again this clearly has implications in the question of timing of injury.

Spontaneous extradural haematoma Rarely, EDH may occur spontaneously (Henderson et al. 2001) or as a result of an underlying vascular malformation (Benzel 1995; Mustafa and Subramanian 1996; Matsumoto et al. 2001). Other rarities include association with craniofacial infection (Griffiths et al. 2002), after temporomandibular joint arthrocentesis (Carroll et al. 2000) and in association with Langerhans’ giant cell histiocytosis (Lee et al. 2000). In addition, anticoagulant therapy is known to be associated with development of an EDH including within the spinal canal (Rodriguez et al. 1995).

Heat haematoma

Figure 8.1 Acute extradural haematoma as a result of a fall from a bicycle. features relating to the haematoma, e.g. inflammatory cell infiltrate and haemosiderin deposition, as well as granulation tissue with features of organisation.

Problem areas Missed EDH

This is well recognised in the forensic setting in association with fire-related deaths. The head is exposed to intense heat and fracturing of the skull and heat haematomas occur in the extradural space. The mechanism is unknown – the distribution closely follows the pattern of charring of the skull. Blood may arise from the skull itself or the venous sinuses. These haematomas are more diffusely spread over the hemispheres, commonly bilateral with a somewhat spongy or honeycomb appearance as a result of the blood boiling (Figure 8.2). Carbon monoxide estimation in difficult cases can be done on the clot to estimate the level and correlate with evidence of the deceased being alive at the time of the fire (Saukko and Knight 2016). If the haematoma occurred during the fire then carbon monoxide should be present in the haematoma correlating with the

Although rare, this still occurs. It may be that the individual gives an unclear history in terms of an injury and subsequent loss of consciousness. A skull fracture may not be identified either because radiological examination of the patient was not carried out or the radiograph was misinterpreted. Alcohol, as well as other drugs, may lead to diagnostic problems. Particularly difficult is the EDH in children where fractures are less common and the history may be less clear. In the criminal setting, the defence may arise that, had diagnosis and treatment been instituted, the individual need not have died. Expert advice from a neurosurgeon over these issues may well be necessary. However, unless a novus actus intervenienes (new intervening act) has been shown to take place, this is unlikely to win the argument although it could play a role in terms of jury ‘sympathy’.

Timing Again within the medico-legal context timing of injury may be crucial in pinpointing the episode, particularly if only a vague history is available or indeed there is evidence of multiple episodes. Unfortunately, from the pathological appearances alone, one is unlikely to be able to give more than a rough timeframe in terms of hours. Again it may well be that in these cases more detailed clinicopathological assessment is necessary with a neurosurgical opinion rather than the pathologist straying outside his or her area of expertise.

Rarities A slow evolution/resolving EDH It is recognised that rarely bleeding may take place very slowly over 10–14 days, with spontaneous resolution over several

Figure 8.2 Heat haematoma; note more diffuse spread over hemispheres and granular appearance in contrast to the acute extradural haematoma shown in Figure 8.1 and Figure 5.5, Chapter 5.

Subdural haematoma blood level. However, in grossly damaged bodies, difficulty may still arise in identifying ante-mortem injury from post-mortem injury, especially in the presence of heat fractures.

Subdural haematoma In this situation, bleeding occurs between the dura and the surface of the brain. Classification in the past has been into acute, subacute and chronic SDH (CSDH), but there has been variation in this and unsatisfactory histological attempts at ageing. A more clinical approach has been for classification based on the appearances of the haematoma with acute comprising clotted blood and subacute a mixture of clotted blood and fluid, and chronic when the haematoma is fluid. Increased use of scanning has led to high-quality definition of the components from the radiological perspective (see Chapter 3).

Acute subdural haematoma Circumstances and clinical features SDH occurs as a result of a different mechanism of injury from that in EDH, which is a focal impact injury. SDH is caused by tearing of the bridging veins between the cortical surface and the dural sinuses. The distribution of bridging veins is more widespread than generally appreciated. They are more prone to damage within the dural border (Smith et al. 2015). Arterial bleeding from a cortical artery may also give rise to acute SDH. Post-mortem analysis has indicated that those caused by arterial rupture are more usually temporoparietal, whereas bridging vein rupture is typically in the frontoparietal, parasagittal region (Maxeiner and Wolf 2002). Most occur as a result of a fall or assault (72 per cent), with a lower incidence occurring in vehicular incidents (24 per cent) (Smith 2016). Experimental studies in non-human primates have shown that acute SDH occurs with injury where there are high rates of acceleration or deceleration as in a fall or assaults. In a road traffic accident situation, the deceleration rate is much slower (Gennarelli and Thibault 1982). The biomechanics is covered in Chapter 4. Violent shaking during interrogation has been reported to cause SDH and dTAI (Pounder 1997); however, review of the case has indicated that impact certainly played a role in an individual with poorly developed neck structures and in any event the diagnosis of dTAI in the presence of severe brain swelling could not be confirmed (Geddes and Whitwell 2003). The location of the SDH without consideration of other factors should not be taken as indicating the site of impact. Assessment of other findings, including external wounds, scalp bruising and skull fractures, is necessary. Contusional brain injury and cerebral swelling are commonly associated features with traumatic axonal injury including dTAI at the most severe end. The associated parenchymal damage is a major factor in the clinical presentation and prognosis (Wilberger et al. 1991). SDHs are ‘pure’ with no associated brain injury in 13 per cent. In these cases, the clinical presentation is of an expanding space occupying lesion. There is a more variable time course than that of EDH and clinicopathological correlation with an episode of injury may be difficult. A number of underlying conditions may predispose to acute SDH. These include underlying brain atrophy, including dementias and alcoholic brain atrophy, as well as underlying coagulopathies including anticoagulant therapy injury. Other conditions may be associated with spontaneous SDH, including arteriovenous malformations, neoplastic lesions and following

rupture of an intracranial aneurysm (O’sullivan et al. 1994; Bromberg et al. 1998; Han et al. 1999). Ethanol may be sequestered in an SDH and its estimation in a haematoma may aid in interpretation of any intoxicating state (Riggs et al. 1998).

Macroscopic pathology This is shown as fresh blood diffusely spread over one hemisphere or both hemispheres (Figure 8.3). In approximately 20 per cent of adult cases, the SDH is bilateral (Whitwell 2001). The pure acute SDH causes a mass lesion and death may occur rapidly (DiMaio and DiMaio 2001). Assessment of volume should be done although this can be difficult. Clots of 50 mL or more are likely to cause symptoms and of over 100 mL death (Davis and Robertson 1997). Interhemispheric subdural haemorrhage in adults is suggestive of an underlying coagulopathy (haemophilia, anticoagulation therapy) (Bartels et al. 1995). Examination for intracranial aneurysm should be done before fixation because this makes dissection of the circle of Willis much more difficult. Assessment of cerebral atrophy in the presence of oedema may be difficult in the fresh state, but the brain should be weighed and subsequently reweighed after fixation. Post-mortem studies have been performed in an attempt to identify bridging vein damage – these have demonstrated venous contrast leakage in the subdural space (Maxeiner 1997). The technique, however, requires radiological input and

(a)

(b)

Figure 8.3 (a) Acute subdural haematoma with associated ­cerebral swelling. (b) Acute subdural haematoma – histology showing collections of red cells, fibrin and leucocytes.

89

90

Intracranial haematomas is time-consuming, so it is not currently a readily available postmortem investigation. Other findings include contusional brain injury, macroscopic markers of traumatic axonal injury and subarachnoid haemorrhage.

Histology For histology of acute SDH, see later.

Problem areas Previous injury/disease Previous head injury/assault and underlying disease that may have significance in the medico-legal situation include atrophy including alcoholic-associated, previous contusional injury and epilepsy. Detailed assessment of previous injury history as well as history of seizures may be important in terms of causation. Rare underlying conditions may be identified including vascular malformations and neoplasia (shown earlier). In an individual case, it may be difficult to arrive at a firm conclusion on the relative roles of the traumatic and underlying natural or other pathology in the brain.

Mechanism of injury In the assault situation, it may be argued that a fall was the cause of the SDH or other brain pathology rather than directly as a result of a blow. Detailed clinicopathological analysis, as well as details of the alleged assault, is essential, see also Chapter 17. Despite this, it may be impossible to differentiate between the various components of the incident. Infant SDH is covered in Chapter 15.

A significant proportion of CSDHs is seen in association with trauma, although this may be of a minor nature and is reported in around two-thirds of patients (Sambasivan 1997; Missori et al. 2000; Asghar et al. 2002). However in a significant minority of patients, there is no history of trauma. It is likely that trauma needs to be minimal because in the atrophic brain the bridging veins leaving the cortical surface to enter the venous sinuses are longer and weaker and more vulnerable to tearing. Other predisposing causes include anticoagulation therapy – most studies, however, fail to differentiate acute, subacute and CSDH. The risk of SDH increases proportionally to the level of anticoagulation as measured by the prothrombin time ratio (Hylek and Singer 1994). Intracranial hypotension as a result of shunting or cerebrospinal fluid (CSF) leakage, either iatrogenic or spontaneous, is also recognised to cause SDH (Giamundo et al. 1985; Chen and Levy 2000). In the former, increased traction on bridging veins, as well as the large negative intracranial pressures generated when there is large distance between the ventricular catheter and shunt terminus, have been put forward as a contributing factor (Chen and Levy 2000). Other pre-existing conditions include epilepsy (Fogelholm and Waltimo 1975) and alcohol abuse (Fogelholm and Waltimo 1975, Stroobandt et al. 1995; Sambasivan 1997). Factors such as cortical atrophy and coagulation disorders, as well as alcohol, as risk factors for trauma may all play a role (Soderstrom et al. 1997). Vascular abnormalities as well as primary and secondary tumours have also been reported in association with CSDH (Popovic et al. 1994; Cinalli et al. 1997; Alimehmeti and Locatelli 2002).

Chronic subdural haematoma

Macroscopic appearances

Pathologically, CSDH is characterised by the formation of organising membranes over the surfaces of the brain, with the potential to increase enlargement of the haematoma over time; histologically, these membranes, together with the vascular channels, are highly typical. The capillaries are very thin walled and prone to spontaneous rupture without trauma giving repeated episodes of fresh bleeding. A small CSDH may be an incidental finding at post-mortem examination. With increasing volume signs and symptoms such as confusion with features similar to a dementia, stroke-like features as well as the effects of a space-occupying lesion may occur. CSDH is a specific entity, and the relation with acute SDH is unclear, although it may develop from an acute SDH as well as a subdural hygroma or be spontaneous (Smith et al. 2015). The membranes have been identified as a source of fluid exudation and haemorrhage. The fragile blood vessels are created by angiogenic stimulation by such factors as angiopoietins and vascular endothelial growth factor and associated factors. There is continued fibrinolysis with high levels of fibrinogen degradation products identified in the fluid, preventing clot formation and with continued bleeding. In addition, there is an inflammatory response within the membranes and fluid (Edlmann et al. 2017). The neomembranes develop from the dural border cells which are able to proliferate (Wittschieber et al. 2015). There is a higher incidence in the elderly population (Fogelholm and Waltimo 1975; Kudo et al. 1992; Asghar et al. 2002). There appears to be racial variation – a higher incidence is reported in Japanese people. Epidemiological figures have shown a higher incidence in males – probably related to increased trauma in this group.

The typical location is over the convexities. The inner arachnoidal membrane is initially thin – it is visible at around 7–10 days. The colour of the haematoma may be uniform or show areas of recent haemorrhage. The membranes are well developed at around 3 weeks or so. Later the membranes become thicker (‘rubber-like’) and the contents solid with fluid areas of varying colours (Figure 8.4). The liquefied clot resembles motor oil after several weeks. Other features present include indentation of the brain with or without midline shift. This is often severe in well-developed cases. There may be discolouration of the underlying arachnoid.

Figure 8.4 Chronic bilateral subdural haematoma.

Subdural haematoma

(a)

Figure 8.5 Early neomembrane with thin layer of fibroblasts. Associated conditions include atrophy, related to age, dementia or alcohol, as well as other pathologies such as old cerebral infarction or previous head injury. Bruising of the scalp in older cases may no longer be visible, particularly if there is no definite or a relatively minor traumatic event. However, subscalp bruising can persist for a considerable time, particularly in elderly people (Perper and Wecht 1980), and may lead to haematoma formation within the subscalp tissues. Histology of any bruise should also be undertaken.

Histology of SDH The literature on ageing SDH is limited. A guide published by Munro and Merritt in 1936 is still useful, as is the summary outline in Knight’s Forensic Pathology (Saukko and Knight 2016). A summary of the major features is shown in Table 8.1. Table 8.1 Ageing of subdural haematoma Up to 24–48 hours

Fresh red blood cells, fibrin

2–5 Days

Some red cell lysis with macrophages. Fibroblasts start to form membrane on the dural aspect.

7 Days

Fibroblasts invade the clot. Dural layer of fibroblasts visible macroscopically – approximately 12 cells in thickness. Few cells on the arachnoidal side.

14 Days

Progressive red cell lysis with the appearance of giant capillaries. Dural fibroblastic layer approximates half the thickness of the dura. Fibroblastic membrane on the arachnoidal side.

28 Days

Clot becomes liquefied with areas of fresh haemorrhage. Fibroblastic layer adjacent to the dura is of similar thickness to the dura with a well-formed arachnoidal layer.

1–3 Months

Giant capillaries with secondary haemorrhages. Hyalinisation of the membranes.

Later

Ageing becomes difficult. The membranes become thick and calcification may occur.

(b)

Figure 8.6 Chronic subdural haematoma with neomembranes and areas of multiloculated haematoma and some bleeding (3 months). (a) Macroscopic appearance. (b) Low power of elastic Van Geisson stain. Multiple samples of dura should be taken. In acute SDH, the haematoma comprises fresh clot with fresh red cells and fibrin to varying degrees. Within 24–48 hours neutrophil polymorphs infiltrate, followed by macrophages within a few day. Immunocytochemistry for MHC Class 2 and monocytes/macrophage antigens may be of assistance (Al-Sarraj et al. 2004). Erythrocyte lysis may be variable and intact red cells can persist for several days or more. Fibroblasts enter the clot after a few days. Early CSDHs possess a recognisable neomembrane. This is delicate, often only the thickness of a few cells (Figure 8.5). By 4 weeks, the well-formed inner and outer membranes become gradually hyalinised over 1–3 months (Figure 8.6a and b). Giant capillaries with secondary re-bleeding may be seen (Figure 8.7). Late appearances are of thickened dura caused by membranes fusing into a fibrous layer. This may calcify or ossify. These appearances can be taken only as a general guide. It is usually possible to comment that the findings could have

Figure 8.7 Giant capillaries with secondary bleeding.

91

92

Intracranial haematomas developed as a result of a particular episode. There is variation in evolution from individual to individual.

Subdural hygromas These are collections of clear, pink tinged or xanthochromic fluid within the subdural space. Neomembranes are not identified. They are commonly bilateral, and trauma is identified in a number of cases. The mechanism of injury is understood to be leakage of CSF into the subdural space, caused by a tear in the arachnoid, which is valve-like in nature (Smith et al. 2015). Rupture of an arachnoid cyst after minor trauma has been reported to cause subdural hygromas (Parsch et al. 1997; Gelabert-González et al. 2002). These are not necessarily restricted to the side of the brain where there was an acute SDH, thus supporting that they may represent a separate entity (Wittschieber et al. 2019).

Problem areas in CSDH Clinically relating trauma to the development of this type of haematoma may be difficult. It is essential to have a history of the event in question. A classic time course from inception to presentation is 3–4 months with the gradual expansion of the lesion. In the medico-legal setting, this is commonly an assault, such as a mugging in an elderly victim with direct trauma to the head, or an assault that results in a fall, such as when a handbag is snatched. Photographic evidence as well as other documentation may be available from the time of the assault. Evidence of pre- and post-event clinical state is also necessary. This may well demonstrate a decline in cerebral function after the episode. Defence counsel may argue for a different episode being responsible for the CSDH, although in practice a well-documented episode would need to be identified. Pathological examination at most can only be ‘consistent with’ any episode. The diagnosis of CSDH may be difficult, particularly in elderly people. If not diagnosed, this may well raise the issue of diagnostic/therapeutic delay in a potentially treatable condition, although in law this is unlikely as such to be of relevance.

References Al-Sarraj, S., Mohamed, S. and Kibble, M. et al. 2004. Subdural hematoma (SDH): assessment of macrophage reactivity within the dura mater and underlying hematoma. Clin Neuropathol, 23, 62–75. Alimehmeti, R. and Locatelli, M. 2002. Epidural B cell non-Hodgkin’s lymphoma associated with chronic subdural hematoma. Surg Neurol, 57, 179–182. Asghar, M., Adhiyaman, V. and Greenway, M.W. et al. 2002. Chronic subdural haematoma in the elderly – a North Wales experience. J R Soc Med, 95(6), 290–292. Bartels, R.H.M.A., Verhagen, W.I.M. and Prick, M.J.J. et al. 1995. Interhemispheric subdural hematoma in adults: case reports and a review of the literature. Neurosurgery, 36, 1210–1214. Benzel, E.C. 1995. Cervical epidural haematoma secondary to an extradural vascular malformation. Neurosurgery, 36, 585–588. Bricolo, A.P. and Pasut, M.D. 1984. Extradural hematoma: toward zero mortality. A prospective study. Neurosurgery, 14, 8–12. Bromberg, J.E., Vandertop, W.P. and Jansen, G.H. 1998. Recurrent subdural haematoma as the primary and sole manifestation of chronic lymphocytic leukaemia. Br J Neurosurg, 12, 373–376. Carroll, T.A., Smith, K. and Jakubowski, J. 2000. Extradural haematoma following temporomandibular joint arthrocentesis and lavage. Br J Neurosurg, 14, 152–154. Chen, J.T. and Levy, M.L. 2000. Causes, epidemiology, and risk factors of chronic subdural hematoma. Neurosurg Clin N Am, 11, 399–406.

Cinalli, G., Zerah, M. and Carteret, M. 1997. Subdural sarcoma associated with chronic subdural haematoma. Report of two cases and review of the literature. J Neurosurg, 86, 553–557. Davis, R.H. and Robertson, D.M. (eds) 1997. Textbook of Neuropathology, 3rd edn. Williams & Wilkins. DiMaio, V.J. and DiMaio, D. 2001. Forensic Pathology, 2nd edn. CRC Press. Edlmann, E., Giorgi-Coll, S., Whitfield, P.C., Carpenter, L.H. and Hutchinson, P.J. 2017. Pathophysiology of chronic subdural haematoma: inflammation, angiogenesis and implications for pharmacotherapy. J Neuroinflammation, 14, 1–13. Fogelholm, R. and Waltimo, O. 1975. Epidemiology of chronic subdural haematomas. Acta Neurochir (Wien), 32, 247–250. Geddes, J.F. and Whitwell, H.L., 2003. Shaken adult syndrome revisited.Am J Forensic Med Pathol, 24(3), 310–311. Gelabert-González, M., Fernández-Villa J. and Cutrín-Prieto, J. et al. 2002. Arachnoid cyst rupture with subdural hygroma: report of three cases and literature review. Childs Nerv Syst, 18, 609–613. Gennarelli, T.A. and Thibault, L.E. 1982. Biomechanics of acute subdural haematoma. J Trauma, 22, 680–686. Giamundo, A., Benvenuti, D. and Lavano, A. 1985. Chronic subdural haematoma after spinal anaesthesia. Case report. J Neurosurg Sci, 29, 153–155. Griffiths, S.J., Jatavallabhula, N.S. and Mitchell, R.D. 2002. Spontaneous extradural haematoma associated with craniofacial infections: case report and review of literature. Br J Neurosurg, 16, 188–191. Han, P.P., Theodore, N. and Porter, R.W. et al. 1999. Subdural haematomas from a type 1 spinal arteriovenous malformation. Case report. J Neurosurg, 90, 255–257. Henderson, R.D., Pittock, S., Piepgras, D.G. and Wijdicks, E.F. 2001. Acute spontaneous spinal epidural hematoma. Arch Neurol, 58, 1145–1146. Hylek, E.M. and Singer, D.E. 1994. Risk factors for intracranial hemorrhage in outpatients taking warfarin. Ann Intern Med, 120, 897–902. Koc, R.K., Pasaoglu, A. and Menku, A. et al. 1998. Extradural hematoma of the posterior cranial fossa. Neurosurg Rev, 21, 52–57. Kudo, H., Kuwamura, K. and Izawa, I. 1992. Chronic subdural hematoma in elderly people: present status on Awaji Island and epidemiological prospect. Neurol Med Chir (Tokyo), 32, 207–209. Lee, K.W., McLeary, M.S., Zuppan, C.W. and Won, D.J. 2000. Langerhans’ cell histiocytosis presenting with an intracranial epidural hematoma. Pediatr Radiol, 30, 326–328. Leestma, J.E. 2014. Forensic Neuropathology. New York: Raven Press. Mandavia, D.P. and Villagomez, J. 2001. The importance of serial neurologic examination and repeat cranial tomography in acute evolving epidural hematoma. Pediatr Emerg Care, 17, 193–195. Matsumoto, K., Akagi, K., Abekura, M. and Tasaki, O. 2001. Vertex epidural hematoma associated with traumatic arteriovenous fistula of the middle meningeal artery: a case report. Surg Neurol, 55, 302–304. Maxeiner, H. 1997. Detection of ruptured cerebral bridging veins at postmortem examination. Forensic Sci Int, 89, 103–110. Maxeiner, H. and Wolf, M. 2002. Pure subdural hematomas: a postmortem analysis of their form and bleeding points. Neurosurgery, 50, 503–508 (discussion 508–9). Mazza, C., Pasqualin, A. and Ferriotti, G. et al. 1982. Traumatic extradural haematoma in children. Acta Neurochirurg, 65, 67–80. Mealey, J. 1960. Acute extradural haematomas without demonstrable skull fractures. J Neurosurg, 17, 27–34. Missori, P., Maraglino, C. and Tarantino, R. et al. 2000. Chronic subdural haematomas in patients aged under 50. Clin Neurol Neurosurg, 102, 199–202. Munro, D. and Merritt, H.H. 1936. Surgical pathology of subdural haematoma. Based on a study of one hundred and five cases. Arch Neurol Psychiatry, 35, 64–78. Mustafa, M. and Subramanian, N. 1996. Spontaneous extra-dural haematoma causing spinal cord compression. Int Orthop, 20, 393–394. O’sullivan, M.G., Whyman, M. and Steers, J.W. et al. 1994. Acute subdural haematoma secondary to ruptured intracranial aneurysm: diagnosis and management. Br J Neurosurg, 8, 439–445.

References Parsch, C.S., Krass, J. and Hoffmann, E. et al. 1997. Arachnoid cysts associated with subdural hematomas and hygromas: analysis of 16 cases, longterm follow-up, and review of literature. Neurosurgery, 40, 482–490. Perper, J.A. and Wecht, C.H. (eds) 1980. Microscopic Diagnosis in Forensic Pathology. Thomas, p 12. Phonprasert, C., Suwanwela, C. and Hongsaprabhas, C. et al. 1980. Extradural haematoma: analysis of 138 cases. J Trauma, 20, 679–683. Popovic, E.A., Lyons, M.K. and Scheithauer, B.W. 1994. Mast cell-rich convexity meningioma presenting as a chronic subdural hematoma: case report and review of the literature. Surg Neurol, 42, 8–13. Pounder, D.J. 1997. Shaken adult syndrome. Am J Forensic Med Pathol, 18, 321–324. Riggs, J.E., Schochet, S.S. Jr and Frost, J.L. 1998. Ethanol level differentiation between post mortem and sub dural haematoma. Mil Med, 163, 722–724. Rodriguez, Y., Baena, R. and Gaetani, P. et al. 1995. Spinal epidural haematoma during anticoagulant therapy. A case report and review of the literature. J Neurosurg Sci, 39, 87–94. Sambasivan, M. 1997. An overview of chronic subdural hematoma: experience with 2300 cases. Surg Neurol, 47, 418–422. Saukko, P. and Knight, B. 2016. Knight’s Forensic Pathology, 4th edn. CRC Press. Servadei, F., Nanni, A. and Nasi, M.T. et al. 1995a. Evolving brain lesions in the first 12 hours after head injury: analysis of 37 comatose patients. Neurosurgery 37, 899–906. Servadei, F., Vergoni, G. and Staffa, G. et al. 1995b. Extradural haematomas: how many deaths can be avoided? Acta Neurochir (Wien), 133, 50–55.

Smith, C., Margulies, S. and Duhaime, A.-C. 2015. Trauma. In Love, S., Budka, H., Ironside, J.W. and Perry, A. (eds), Greenfields Neuropathology, 9th edn, Vol 1. CRC Press. Soderstrom, C.A., Smith, G.S. and Dischinger, P.C. et al. 1997. Psychoactive substance use disorders among seriously injured trauma center patients. JAMA, 277, 1769–1774. Stroobandt, G., Fransen, P., Thauvoy, C. and Menard, E. 1995. Pathogenic factors in chronic subdural haematoma and causes of recurrence after drainage. Acta Neurochir (Wien), 137, 6–14. Tatagiba, M., Sepehrnia, A. and El Ax, M. et al. 1989. Chronic epidural haematoma – report of eight cases and review of the literature. Surg Neurol, 32, 453–458. Viljoen, J.J. and Wessels, L.S. 1990. Subacute and chronic extradural haematomas. S Afr J Surg, 28, 133–137. Whitwell, H.L. 2001. Head injury. In Rutty, G.N. (ed), Essentials of Postmortem Examination Practice, Vol 1. Springer-Verlag. Wittschieber, D., Karger, B., Neiderstadt, T., Pfeiffer, H. and Hahnemann, M.L. 2015. Subdural hygromas in abusive head trauma: pathogenesis, diagnosis, and forensic implications. Am J Radiol, 36, 432–439. Wittschieber, D., Karger, B., Pfeiffer, H. and Hahnemann, M.L. 2019. Understanding subdural collections in pediatric abusive head trauma. Am J Neuroradiol, 40, 388–395. Wilberger, J.E., Harris, M. and Diamond, D.L. 1991. Acute subdural hematoma: morbidity, mortality and operative timing. J Neurosurg, 74, 212–218.

93

9

Subarachnoid haemorrhage and cerebrovascular traumatic pathology Daniel du Plessis and Paul Johnson

This chapter focuses on forensic aspects of traumatic and natural subarachnoid haemorrhage. Traumatic vascular pathology mimicking natural vascular disease and vice versa will also be addressed in this chapter.

Traumatic subarachnoid haemorrhage Subarachnoid haemorrhage in association with traumatic brain injuries Subarachnoid haemorrhage is a common finding in association with primary traumatic intracranial injuries such as extra-axial bleeds and brain contusions. It is typically patchy subarachnoid haemorrhage over the surface of the brain (Figure 9.1) resulting from shearing injury to small blood vessels. Isolated patches may also represent secondary pathology in situations involving raised intracranial pressure with ischaemic effects on blood vessels promoting extravasation of blood. The location of bleeding cannot be taken to indicate the site of impact and other factors need to be taken into account such as scalp bruising, skull fractures and contusional injury. Subarachnoid haemorrhage may stimulate a very rapid neutrophilic inflammatory response (within 1–4 hours). This is followed by a chronic inflammatory cell infiltrate with macrophage activity after around 12 hours. Haemosiderin deposition may be seen within 16–32 hours and lasts for months or years (Hammes 1944; Saukko and Knight 2016). Fibrosis may develop with the potential long-term complication of hydrocephalus (Weller 1995).

formulated after the work of Simonsen and Contostavlos (Wilks 1859; Ford 1956; Simonsen 1963, 1967, 1984; Contostavlos 1971; Cameron and Mant 1972; Tatsuno and Lindenberg 1974; Harland et al. 1983; Contostavlos 1995). The suggestion in early reports that the source of bleeding related to rupture of the extracranial portion of the vertebral arteries is untenable on anatomical grounds and has been disproved by subsequent more thorough investigative techniques (Cooper et al. 1999; Gray et al. 1999; Koszyca et al. 2003; Leadbeatter 1994). Rarely, an extracranial vertebral dissection extending into the intracranial part of the vessel may cause rupture of the latter.

Circumstances and mechanism of injury

The classic situation is that of sudden collapse, being rendered comatose or very commonly subject to rapid cardiorespiratory arrest, within the context of an attack involving direct blunt trauma to the head or upper neck. The most common sites of impact are to the region of the ear, mastoid, angle of jaw and/ or upper side of the neck (Figure 9.2). In many cases, external bruising may not be apparent despite careful inspection. Deeper dissection of the areas above though very commonly (in our experience almost invariably) reveals evidence of an impact. It has been pointed out that causative impacts may occur to any point of the head, including the chin and back of the head. It may thus be difficult to differentiate, depending on the findings, whether the haemorrhage was initiated by an action such as a kick or subsequent fall, including a fall unrelated to the attack. In our experience an impact to the vulnerable sites to the side of the head can nearly always be demonstrated. Additional impact sites found to other parts of the head are often more Traumatic basal subarachnoid readily explained by a consequent fall to the ground. Ultimately haemorrhage though, any blow which can lead to twisting and tilting of the True traumatic basal subarachnoid haemorrhage is a well-­ head causing strain on the vertebrobasilar arterial system could validated forensic entity, although not infrequently underrec- reasonably be implicated in a traumatic tear of one of the verteognised leading to suboptimal investigation. The term should brobasilar intracranial vessels. be reserved for the specific entity of damage to the vertebroThe degree of trauma is not necessarily severe, but the dembasilar vasculature in association with trauma, most commonly onstration of deep paravertebral bruising or a C1 fracture rules a blow, to the head and neck, resulting in neck hyperextension out minor trauma. The investigative approach should include and rotation, effectively twisting and tilting the head on the neck. assessment of the arterial system for any predisposing anaTraumatic tears of the distal internal carotid arteries, proximal tomical abnormalities and conditions potentially contributmiddle cerebral arteries, anterior cerebral arteries and pericallo- ing to greater susceptibility for rupture and/or dissection. This sal arteries may very rarely also cause major traumatic subarach- includes short upper cervical loop segments (Medcalf 2016) and noid bleeds with a basal component. Exploration of these vessels conditions such as fibromuscular dysplasia, segmental arterial should therefore not be neglected and should be considered in mediolysis (SAM), vascular Ehlers-Danlos disease (type IV cases where the bulk of subarachnoid haemorrhage appears to be EDS) and Marfan’s syndrome. In our experience, pre-existing supratentorial, accentuated around the circle of Willis and within vascular degenerative disease is only identified in a very small the Sylvian fissures within an appropriate traumatic context. minority of cases. Post-mortem genetic testing for vascuThe recognition that trauma may result in fatal basal sub- lopathies can be undertaken and Pickup and Pollanen (2011) arachnoid haemorrhage was first suggested by Ford, and may demonstrated mutations in the COL3A1 gene associated with have been described in the 19th century by Wilks, but it was vascular Ehlers-Danlos syndrome (type IV EDS) in two cases of

Traumatic basal subarachnoid haemorrhage

Figure 9.1 Patchy subarachnoid haemorrhage as seen in severe head injury. traumatic subarachnoid haemorrhage. A common role for this condition as a susceptibility factor though has so far not been confirmed. Whilst exceptionally a tear appears to develop at the edge of an atheromatous plaque, good evidence that atherosclerotic cerebrovascular disease predisposes to traumatic tears and dissections is wanting, but a bias towards younger age

group victims may be responsible for this observation. In properly investigated cases, arterial dissections may also be found in the terminal part of the extracranial vertebral artery, usually discontinuous with an intracranial site of rupture. Tears are typically identified on the same side of the causative impact. Rapid hyperextension/flexion of the neck without blunt impact has been reported to lead to tears on occasion (Gee 1982) and the vessels are known to be susceptible to longitudinal extension (Johnson et al. 2000). Elegant flow dynamic studies (Farag et al. 1988) showed that sudden occlusion of the vertebral arteries can induce reverse blood flow and induce tearing in the typical locations found in case work. Alternatively or additionally, impacts to the upper neck have been suggested to cause rapid compression of one of the major arteries of the neck (carotid or vertebral) causing injury to more distal continuation of the struck vessel through a ‘water hammer effect’ due to sudden forceful displacement of blood through the vessel. Severe traumatic basal subarachnoid haemorrhage is almost invariably immediately followed by a profoundly unconscious, collapsed state (closely preceding or coincident with cardiac and/ or respiratory arrest) and usually leads to death. Resuscitation may restore circulation in a minority of cases, but the outcome remains almost invariably fatal with death usually ensuing within hours to a few days. The mechanism of cardiorespiratory arrest remains speculative and has been attributed to a percussive effect on the brainstem, an abrupt exposure of the low-pressure cerebrospinal fluid-filled subarachnoid space to high intra-arterial pressure or to instantaneous near instantaneous vasospasm of blood vessels supplying the brainstem (Lindenberg and Freytag 1970; Leadbeatter 1994). In our experience, blood is very commonly found in the fourth ventricle (explained by the proximity of the usual location of tears to the lateral exit foramina) and acute distention of the fourth ventricle by blood may have a role in precipitating cardiorespiratory instability with arrest.

Post-mortem examination

(a)

(b)

Figure 9.2 (a) Kick to side of neck in traumatic subarachnoid haemorrhage. (b) Internal jaw bruise as a result of a blow causing a traumatic subarachnoid haemorrhage.

Post-mortem CT imaging may be extremely useful in demonstrating a basal subarachnoid haemorrhage, thus directing appropriate investigative techniques in advance. Post-mortem angiography may also be of value (Karhunen et al. 1990; Vanezis 1979, 1986), but negative angiography does not exclude the possibility of vascular injury. Plain radiological examination of the craniocervical region has also been recommended as an adjunct. This has the advantage that congenital bony abnormalities may be identified (Gross 1990). This may also assist in identifying fractures of the transverse processes of C1 in advance of a neck dissection. Dense basal subarachnoid haemorrhage may be an unexpected finding in an anticipated routine post-mortem and it is important that the pathologist exercises a high index of suspicion to rule out the possibility of unknown trauma. This may require modifying routine post-mortem techniques to investigate the possibility of bruising to the vulnerable sites described above and to remove the extracranial vertebral arteries. If in doubt, the case should be referred without further potential compromise of vital evidence, to a forensic pathologist. An autopsy method reported by McCarthy et al. (1999) is to ligate the distal basilar artery and then remove the cerebral hemispheres, leaving the brainstem and cerebellum in situ. The origin of each vertebral artery is identified, the thoracic organs having been left in situ. This method may permit the

95

96

Subarachnoid haemorrhage and cerebrovascular traumatic pathology passing of a cannula into one of the vertebral arteries. Liquid (water, dyed or undyed and even milk) can be injected into the artery while observing the vasculature at the base of the brain. This may demonstrate the site of rupture, focusing histological assessment on that area. Care must be taken not to inject fluid under too much pressure as this may dislodge plugging thrombi, rupture false or traumatic aneurysms and potentially even introduce full-thickness tears in areas of partial tearing or dissection. Post-mortem room angiography can also be effective using a similar cannulation approach (Johnson et al. 2017). In order to minimise artefact, it is advisable to take the forebrain out before carefully removing the hindbrain, cutting the vertebral arteries under direct vision just above the dura. It is essential to visualise the transected stumps of the intracranial vertebral arteries within the foramen magnum and then dissect these out with a cuff of surrounding dura to allow serial transverse section embedding of these segments as well. A next essential step is that of careful clearing of the basal subarachnoid haemorrhage from the vertebrobasilar arterial system, permitting exposure of these vessels prior to fixation. Such clearing should also include the circle of Willis and its major proximal branches. Clearing should be carried out using gentle irrigation by water. It must be emphasised that the posterior inferior cerebellar arteries must be preserved as part of this exercise. In those regrettable cases where a brain has been referred fixed without adequate or any clearing of the basal subarachnoid haemorrhage, the relevant vessels encased in firm blood clots require exposure, a far more difficult exercise which can be assisted by immersion (which may require several hours) and irrigation by water containing domestic dishwashing soapy detergents which assist in softening formalin-fixed blood clots. The vertebrobasilar arterial system, including the proximal posterior inferior cerebellar arteries, should be carefully dissected off the brainstem (preferably after fixation). Further examination before and after fixation should be undertaken for the presence of an aneurysm or other vascular abnormality. Following removal of the relevant arterial vessels (preferably also including removing and sampling of some of the anterior circulation vessels), the brain can be dissected following normal trauma protocols. The demonstration of the site of arterial rupture has wrongly been regarded as a very difficult exercise. This is not the case if appropriate, diligent post-mortem examination techniques are followed which, in our experience, will identify the site of haemorrhage (or at least proof evidence of mechanical vascular injury) in over 90 per cent of cases. Much focus has been placed on dissection techniques allowing for the exposure and preservation of the extracranial vertebral arteries. Whilst in our opinion still a mandatory component of the investigation of potential traumatic basal subarachnoid haemorrhage (as it assists in assessing deep vertebral traumatic impact-related injury, may reveal dissections within the course of the extracranial vertebral arteries and provide information on pre-existing, predisposing vascular disease), the main focus of attention should be the intracranial vertebrobasilar arterial system. Both intracranial vertebral arteries should be embedded in full as serial transverse around 2 mm thick annuli, from and including the point of penetrance of the dura in the foramen magnum and including the posterior inferior cerebellar artery junctions with embedding of the proximal parts of both of these vessels as well. The whole of the basilar artery up to its apex should also be embedded fully in similar manner. This is not an onerous task (that being the fate of laboratory staff who have to ensure correct

orientation when embedding), which can be accomplished within mere minutes using microcassettes or tissue wraps. Each of the intracranial vertebral arteries should be split into two or preferrably three cassettes (for example proximal third, mid-third, including posterior inferior cerebellar artery (PICA) junction and distal third) with the basilar artery split between two cassettes (proximal half and distal half). This approach (supported by at least elastic van Gieson [EVG] or Movat staining) is absolutely essential in our opinion to identify the source of rupture and/or other confirmatory evidence of mechanical overstretching in almost all cases in our experience. It is important to include the proximal posterior inferior cerebellar arteries as these may be involved by tears and dissections as well. These vessels are further the most common site of false or traumatic aneurysms in those who initially survived the traumatic event precipitating arterial injury (including those who remained conscious and those with a delayed symptomatic presentation) (Purgina and Milroy 2015). The removal of the extracranial vertebral arteries for histological assessment can be achieved by en bloc excision of the cervical spine with the base of skull (Vanezis 1979), with subsequent decalcification and dissection or serial ‘bread’ slicing. However, this is destructive and time-consuming and the arteries can be successfully exposed and excised (with practice) by a variety of described techniques. Bromilow and Burns (1985) presented an anterior cervical approach using dental wire cutting scissors to remove the anterior walls of the transverse processes, which is effective in practice at exposing the vessel from the subclavian origin to the C1 transverse foramen. A modified posterior approach such as described by Kim et al. (2015) is though recommended for the more difficult segment between the transverse process of C1 and the dural perforation. Direct observation of the location and extent of any bleeding around the arteries and the documentation of any potentially relevant anatomical variations such as short upper cervical protective loops (Medcalf 2016) is also facilitated.

Macroscopic neuropathology External examination shows predominantly basal subarachnoid haemorrhage (Figure 9.3a). Adherent clots along the course of vessels may contain traumatic aneurysms (Figure 9.3b). Occasionally direct inspection of the arterial vessels may identify an area of tearing which usually has a longitudinal orientation (Figure 9.4). Transverse defects should be viewed as suspicious of potential artefact, but histology should provide more conclusive evidence of the nature of the focus of disruption. On sectioning of the brain, intraventricular haemorrhage is commonly seen. Intraventricular haemorrhage usually represents retrograde spread from the fourth ventricle, the latter subject to a high-pressure arterial jet of blood directed into one of the lateral exit foramina, a location very often close to or at the point of vertebral arterial rupture. Other features of head injury may be present such as subdural bleeding or contusions. Such additional primary traumatic pathology is relatively uncommon and is usually explained by a fall-related impact following on from a blow which precipitated the arterial rupture. There are rare cases of rupture of an underlying aneurysm occurring with a direct time relationship to head injury (see below).

Histology A number of histological studies have been performed on the vertebral arteries. Wilkinson (1972) described the vertebral artery in 20 individuals aged 60–75 years. The extracranial

Traumatic basal subarachnoid haemorrhage

Figure 9.4 Longitudinal tear in mid-third of intracranial ­vertebral artery. (a)

(b)

Figure 9.3 (a) Traumatic basal subarachnoid haemorrhage. (b) Organising blood clot over ruptured right vertebral artery in a case of prolonged survival. portion is described as having a well-developed adventitia and elastic lamina with a broad media of collagen, smooth muscle and elastic fibres. About 1 cm after it penetrates the dura, the artery changes in structure, becoming thinner with loss of adventitial collagen and a fragmented or absent external elastic lamina. These findings were confirmed by Coast and Gee (1984). Further studies by Johnson et al. (2000, 2001) have demonstrated the development of collagenous scarring

with increasing age as well as fragmentation of elastic fibres. Changes are often focal with sparing of the intracranial portions. Atheromatous degeneration also increased with age. It is important not to misinterpret such changes as evidence of primary degenerative vascular diseases such as FMD, SAM, EDS or other degenerative arteriopathies. We advocate not only sampling the extracranial vertebral arteries to assess for predisposing significant degenerative primary vascular disease, but also taking a sample of one of the renal arteries in any case of potential traumatic basal subarachnoid haemorrhage. Most previously published cases of traumatic subarachnoid haemorrhage do not include histology which should be mandatory as it provides key, potentially pathognomonic evidence in respect of a traumatic bleed. Tears can be identified which may be partial or full and potentially plugged by fibrin or complicated by false or traumatic aneurysms (Figure 9.5a). These may be accompanied by micro- or macro-vascular dissections (Figure 9.5b). The edges of the tears or dissections usually show evidence of pathognomonic mechanical stretch–related change in the form of medial smooth muscle nucleolysis and unruffling or straightening of the internal elastic lamina (Figure 9.5c). The former is a more reliable feature caused by digestion of the nuclear envelope and chromatin resulting in eosinophilic homogenisation (rendering hyaline appearances) of segments of overstretched media and leaving eosinophilic ghost outlines of the nuclei. This is thought to be caused by mechanical overstretching which leads to enzymatic digestion of smooth muscle nuclei, a process which could conceivable continue for a short while peri- or even post-mortem. Special stains such as an EVG or smooth muscle actin immunostaining (Figure 9.5d) can be used to distinguish such change from true collagenous hyalinisation/scarring as may be found in an aneurysm wall (Pollanen et al. 1996). Such mechanical stretch–related changes can further be identified in vascular segments away from the site of rupture/dissection. It should further be born in mind that vascular traumatic injury can be multifocal, involving the opposite vertebral artery as well as the basilar artery. This motivates sampling of the whole of the vertebrobasilar arterial system as the identification of such mechanical stretch–related changes provides compelling evidence of mechanical overstretching and therefore by implication of a traumatic aetiology even in those rare cases where the actual tear implicated in the

97

98

Subarachnoid haemorrhage and cerebrovascular traumatic pathology

 

  Figure 9.5 Histology of arterial trauma (a) acute tear of intracranial vertebral artery plugged by fibrin. (b) Microdissections of the media in proximity to tear. (c) Mechanical stretch–related changes in the edge of the tear consisting of medial smooth muscle nucleolysis (imparting eosinophilic homogenisation of the media) and unruffling of the internal elastic lamina. (d) Smooth muscle actin immunostaining showing preserved media in edge of tear. fatal subarachnoid haemorrhage could not be identified. Slight faint homogenisation with fading nuclear profiles may be seen as an artefact in the vessel wall where the vessel has been subject to most bending post-mortem due to cutting of the specimen. Here such changes should be disregarded unless very overt and involving adjacent non-bent segments. Care should also be taken in not over-interpreting splitting of the internal elastic lamina from the intima or media (allowing for the introduction of some red blood cells into the splits as an artefact) as true traumatic dissections. Convincing dissections should really only be identified in the vascular media. Identifying mechanical stretch–related changes (or fibrin/neutrophils in cases who survive for several hours) in the margins of the dissections serves to further support a true dissection, but these are not invariably present. Changes consistent with pre-existing vascular disease have been described in some cases, (Dowling and Curry 1988;

Opeskin and Burke 1998). Atherosclerosis could potentially undermine the structural integrity of vessels, predisposing to traumatic rupture, although a clear correlation has not been identified by the authors experience based on a very large number of cases subject to extensive histology. A traumatic cause for an area of disruption within a plaque is best verified relying on stretch-related mechanical changes in the margins of the tear. In those subject to immediate to near immediate cardiorespiratory arrest with no resuscitation, examination of the brain reveals limited additional significant pathology. With a period of survival, the pathology is usually predominated by features consistent with secondary hypoxic-ischaemic brain injury. β-APP immunostaining in such cases is almost invariably restricted to demonstrating vascular axonal injury unless secondary fall-related head trauma was also incurred. Some non-specific congestion may be seen of periventricular thin-walled vessels. These may be associated on occasion with small perivascular

Spontaneous subarachnoid haemorrhage foci of red cell extravasation, again a very non-­specific feature and not constituting a plausible cause for significant intraventricular haemorrhage. In cases where the physical appearances of the deceased and/or where histology of the vessels suggests or confirms an underlying primary degenerative vasculopathy, appropriate genetic studies should be considered.

Medico-legal issues Alcohol and traumatic subarachnoid haemorrhage Various series show that almost all individuals who die of this condition have variable levels of blood alcohol (Simonsen 1963; Hillbom and Kaste 1981; Gray et al. 1999). The laxity of the neck muscles is thought to play a major role in that the neck is more easily jerked in an unconscious or intoxicated individual. It is further recognised that ethanol has direct effects on blood vessels, decreasing vasospasm after injury and causing vasodilatation (Barry and Scott 1979). In addition, there is delayed muscle reaction and poor coordination. However, alcohol is a feature of many homicides and other violent deaths (Whittington 1980) and concurrent intoxication is conversely not invariably the case with traumatic basal subarachnoid haemorrhage.

Trauma severity Fatal arterial rupture can be caused by a single blow or impact (Leadbeatter 1994; Gray et al. 1999). The blow may be profound enough to cause deep paraspinal bruising and may, on occasion, fracture of the transverse process of C1, but cases may only be associated with more superficial soft tissue bruising. It is also recognised to occur (extremely rarely) with sudden movement of the neck without the occurrence of impact (Gee 1982). If multiple sites of trauma are present, it can be difficult in an individual case to attribute a particular injury to the fatal haemorrhage, unless an ipsilateral deep classical lateral head/ neck bruise is documented. Correlation with other evidence, including CCTV coverage, is always valuable.

Delayed presentations The very large majority of cases are associated with sudden immediate collapse and cardiorespiratory arrest. Relying on a handful of cases reported in the literature and our own experience, such a presentation may uncommonly be delayed, usually for a matter of seconds to minutes (as backed up by CCTV evidence) with the victim being able to continue standing, remonstrating/fighting before collapsing. Short delayed presentations may be due to a dissection or partial rupture progressing to a full-thickness rupture. More delayed presentations (which could be as long as several days to weeks) are explained on the basis of a traumatic aneurysm with delayed rupture. Clearly, if an artery is not subject to an initial full-thickness disruption, the onset of haemorrhage may be delayed (Marek 1981; Deck and Jagadha 1986; Leadbeatter 1994).

in association with drug abuse (Chen et al. 2003). Associated conditions include fibromuscular dysplasia (Cloft et al. 1998), various connective tissue disorders (Stehbens 1989), polycystic kidneys (Rinkel et al. 1993) and vascular abnormalities, including arteriovenous malformations (Weller 1995). There also is a familial tendency (Alberts 1999) in some early-onset cases. There is a recognised association with alcohol and cigarette smoking for subarachnoid haemorrhage (Juvela et al. 1993). It is less clear whether these factors can be definitely implicated in the formation of aneurysms as such. Increased alcohol consumption has been shown to impair outcome with increased rebleeding and delayed ischaemia (Juvela 1992). Other risk factors for intracranial aneurysms and subarachnoid haemorrhage include older age, female gender, posterior circulation aneurysms, “non-white” ethnicity, previous symptoms, low body mass index and hypertension (Kalaria et al. 2015). Seasonal variation in aneurysmal subarachnoid haemorrhage has been noted (Weller 1995), with rupture in males peaking in autumn and in females in spring (Chyatte et al. 1994). Rupture is also more common during the day and in association with physical stress. Intoxication with drugs causing blood pressure surges such as cocaine may also precipitate aneurysmal rupture. In 15–20 per cent of cases, the cause of subarachnoid haemorrhage remains unknown even after two or more angiographic studies. So-called non-aneurysmal subarachnoid haemorrhage is divided into two groups with different clinical outcomes. Perimesencephalic subarachnoid haemorrhage is characterised by blood restricted to the perimesencephalic cisterns anterior to the brainstem which could extend to the ambient cisterns and basal parts of the Sylvian fissures. Non-perimesencephalic subarachnoid haemorrhage has a more diffuse blood distribution beyond the aforementioned regions. Non-perimesencephalic subarachnoid haemorrhage exhibits a more aggressive clinical course, although the prognosis is better overall compared to aneurysmal subarachnoid haemorrhage (Coelho et al. 2016).

Pathology Cerebral aneurysms are saccular in nature, occurring particularly at the branching points of the intracranial cerebral arteries: 85–90 per cent occur in the territory of the terminal internal carotid arteries or the anterior part of the circle of Willis and around 5–10 per cent occur on the posterior circulation (Figure 9.6). The size varies from a millimetre or so to a much larger with 85 per cent presenting as rupture of less than 10 mm (Forget et al. 2001). Giant aneurysms are classified as such with a diameter of greater than

Spontaneous subarachnoid haemorrhage Ruptured intracranial aneurysms account for about 85 per cent of cases of subarachnoid haemorrhages (Kalaria et al. 2015). Other causes include ruptured vascular malformations, mycotic aneurysm rupture and subarachnoid haemorrhage in conjunction with an intracerebral haemorrhage. Aneurysms may also rarely be atherosclerotic and infective, and caused

Figure 9.6 Middle cerebral artery berry aneurysm.

99

100

Subarachnoid haemorrhage and cerebrovascular traumatic pathology

Figure 9.8 Histological appearances of a berry aneurysm.

Figure 9.7 Giant aneurysm arising from the internal carotid/ middle cerebral artery. 25 mm (Figure 9.7). A total of 20–30 per cent of patients in whom an aneurysm is identified will have one or more aneurysms elsewhere within the main arterial circulation. Identification of an unruptured aneurysm in a case of significant subarachnoid haemorrhage may therefore be predictive of an aneurysm elsewhere in the circulation. Post-mortem identification of these aneurysms is optimal in the unfixed brain assisted by exposure of the course of the various intracranial arteries after careful removal of obscuring subarachnoid blood. Aneurysms may though be difficult to identify as they can be obliterated by rupture. Aneurysms may show calcification, atherosclerosis and evidence of previous thrombus with organisation. There may be evidence of previous leakage from the aneurysm with orange-brown subarachnoid staining. The site of rupture may be indicated by the location of haemorrhage: basal cisterns, (circle of Willis aneurysm), haematoma between the frontal lobes (anterior cerebral or anterior communicating artery aneurysm) and Sylvian fissure (middle cerebral artery aneurysm). Rupture into brain parenchyma can occur which may extend into the ventricular system. On rare occasions, aneurysms may rupture into the subdural space (even causing fatal subdural haemorrhage in the absence of any significant subarachnoid haemorrhage). In one of the authors’ experience, this has been found in a number of cases involving anterior communicating and anterior cerebral artery aneurysms.

Histological features The site of rupture may be identifiable with variable acute inflammation and other features depending on the period of survivability. The wall comprises fibrous tissue of varying thickness lacking a media with smooth muscle with atheromatous degeneration commonly present (Figure 9.8). Evidence of previous rupture and haemorrhage may also be seen.

Medico-legal issues Relationship of rupture of a saccular aneurysm and trauma This can be a difficult area and requires detailed clinicopathological assessment in relation to the alleged assault. This situation is

uncommon but recognised as evidenced by the detailed series of Bowen (1984) and Gonsoulin et al. (2002). Knight (1979) reports a case that illustrates the legal issues of ‘beyond reasonable doubt’ in English law where the victim collapses within moments of trauma to the head and neck and dies as a result of rupture of a ‘berry’ aneurysm on the posterior communicating artery. The authors have seen rare isolated cases where the time relationship between assault and collapse was instantaneous. It may be postulated that a predictable transient hypertensive surge in a situation of anxiety/agitation may precipitate aneurysmal rupture. It may also be impossible to exclude the possibility that direct trauma such as a heavy blow to the head or neck could cause rupture of a thin-walled aneurysm. Extensive histological sampling as for traumatic basal subarachnoid haemorrhage should be carried out as the demonstration of mechanical stretch–related pathology in proximity or elsewhere may serve to support the argument for traumatic aneurysmal rupture.

Mechanism of death Potential mechanisms of death pertaining to severe traumatic basal subarachnoid haemorrhage as discussed above also pertain to basal subarachnoid haemorrhage of natural causes. Fatal complications (some acting in a contributory rather than primary manner) of subarachnoid haemorrhage include obstructive hydrocephalus and severe brain swelling. The latter can assume the form of very rapid, dramatic perfusion swelling (so-called malignant brain swelling). Neurogenic pulmonary oedema, cardiac dysrhythmias or myocardial injury may also contribute to or cause death (Schievink et al. 1995). Acute neurogenic pulmonary oedema may be found in as many as 90 per cent of those who die suddenly from spontaneous subarachnoid haemorrhage (Walder et al. 2002). Morbidity related to neurogenic pulmonary oedema though is more within the range of 40–50 per cent and the reported mortality of neurogenic pulmonary oedema is approximately 7 per cent. Patient outcome in cases of neurogenic pulmonary oedema is usually determined by the underlying neurological insult that led to neurogenic pulmonary oedema and not by the pulmonary oedema per se.

Other traumatic vascular injuries Other types of traumatic vascular injury are recognised, including traumatic dissections, arteriovenous fistulae and traumatic aneurysms.

Arterial dissection Trauma to the vertebral artery causing dissection is well described and commonly involves the third segment at the level

Other traumatic vascular injuries of the axis and atlas, extending intracranially or proximally. This may be complicated by superadded thrombosis or aneurysm formation (Martin and Enevoldson 1998). Thrombosis associated with extracranial vertebral artery dissection is most often identified at the site of trauma. This usually involves the third segment at C1/C2 level. Incomplete tears of the basilar artery have also been described following trauma not necessarily associated with massive basal subarachnoid haemorrhage (Djokic et al. 2003). Penetrating injury or spinal fracture may also cause dissection. Situations known to be associated with cervical arterial dissection include chiropractic neck manipulation, head injury, including road traffic accidents, spinal fracture, direct blunt neck trauma, hanging and strangulation and a number of sporting activities (Vanezis 1986; Hinse et al. 1991; Lee et al. 1995; Prabhu et al. 1996; Norris et al. 2000; Rothwell et al. 2001). Bilateral vertebral artery dissection may occur (Takahara et al. 2019). Minor trauma associated with dissection have included yoga exercises, painting a ceiling, coughing, sneezing, vomiting, receiving an anaesthetic, resuscitation, turning the head to reverse a car and chiropractic manipulation of the neck. Examination in those cases where there is a traumatic history does not usually show underlying disease. There is an association with fibromuscular dysplasia and other heritable disorders of collagen and elastin such as Ehlers-Danlos syndrome type IV, and Marfan syndrome, autosomal dominant polycystic disease and osteogenesis imperfecta type I. There also appears to be an increased risk in cases of hypertension, atherosclerosis, smoking, oral contraception and migraine, although the literature is somewhat unclear (Haldeman et al. 1999; Mitchell 2002). There may be segmental mediolytic arteriopathy on histology (Fantaneanu et al. 2011). In medico-legal terms, each case will need careful assessment of all factors, including a history and detailed pathology. The latter may necessitate serial sectioning of the entire vessel (Leadbeatter 1994). Dissection of the extracranial major cervical arterial vessels (carotid and vertebral) is a likely underrecognised complication of traumatic head and neck injuries (Figure 9.9). True traumatic versus spontaneous dissections require distinction. True traumatic dissections are seen in a setting of severe head and neck trauma with spontaneous dissections often associated with underlying predisposed arteriopathy such as primary degenerative vasculopathies. In the latter situation minor, including unrecognised trauma probably plays some precipitating role (such as sudden twisting or overextension of the neck). Cervical arterial dissections, whether truly traumatic or spontaneous, contribute to a significant number of strokes in young adults. In the older population, the picture is often complicated by underlying arteriosclerosis. Dissections may be responsible for clinical signs rather than direct brain trauma (Rommel et al. 1999). Carotid dissections are usually extracranial above the bifurcation of the common carotid artery and usually caused by overstretching due to hyperextension and rotation of the head. These may be complicated by infarcts in up to 60 per cent of cases. The vessels may become critically stenosed or even occluded by dissecting bleeding within the arterial wall. Occlusion may also follow on thrombosis (Figure 9.9c). The majority of related infarcts occur on the basis of local thrombosis or thromboembolic events. This may have a delayed presentation, including an occasional presentation weeks to even months after the initial injury (Mokri et al. 1988; Nunnink 2002). A confident diagnosis of such a delayed stroke will require diligent histological sampling to identify both historic arterial wall injury as well

as superimposed fresh thrombosis combined with the exclusion of alternative plausible causes of a non-dissection-related stroke. Direct or penetrating trauma may also cause dissection (de Recondo et al. 1995). Subadventitial bleeds may result in pseudoaneurysms (Schievink 2001). Traumatic dissection of intracranial arteries is also recognised as a cause of infarction (Rutherfoord et al. 1996). In such cases, the dissection tends to be subintimal rather than medial or adventitial.

Traumatic aneurysms This type of aneurysm forms less than 1 per cent of all aneurysms and is more common in children and males, the latter probably reflecting the increased incidence of trauma (Larson et al. 2000).

Aetiology and pathogenesis Traumatic aneurysms occur after trauma, either blunt or penetrating in nature. Causes of blunt trauma include road traffic accidents, falls and assaults. Penetrating injuries include penetrating stab wounds and missile injury (Holmes and Harbaugh 1993; Cogbill and Sullivan 1995; Saito et al. 1995; Aarabi 1988; Amirjamshadi et al. 1996; Kieck and de Villiers 1984). There are also reports of aneurysms occurring as a complication of surgery (Ventureyra and Higgins 1994; Bavinski et al. 1997). The mechanism of injury involves either direct vascular injury or stretching of the vessel by adjacent forces and relates to the anatomical location of the involved artery. They are not infrequently related to skull fractures. Anatomically classification is as follows: • Proximal to the circle of Willis, which includes infra- and supraclinoid carotid artery aneurysms as well as vertebrobasilar aneurysms (Figure 9.10a). • Distal to the circle of Willis, which includes cortical and subcortical aneurysms. There is a variable time from trauma to presentation that relates to acute intracranial haemorrhage – an average of around 21 days, although this is variable. Presentation may also be the result of local effects of a mass lesion such as cranial nerve palsies, visual disturbance, hydrocephalus and growing skull fracture (Endo et al. 1980). Traumatic aneurysms may be true or false although some authors also include a mixed group (Larson et al. 2000). True aneurysms are as a result of localised weakening of the wall with local disruption of the intima, and variable intimal elastic layer involvement and media but leaving the adventitia intact. False aneurysms (or pseudoaneurysms) arise as a result of fullthickness laceration to the arterial wall with haematoma in continuity with the lumen. The wall is formed by organisation of blood to granulation tissue and eventually fibrous tissue. False aneurysms are the most common histological type. Differentiation from natural ‘berry’ aneurysms may be difficult. The following factors may aid distinction: peripheral location, site away from a branching point, irregular outline of the sac, absence of a neck, delayed filling and emptying and location adjacent to the falx edge (Opeskin 1995). Histological examination can be identical to saccular aneurysms, particularly in the case of true aneurysms following scarring of the vessel wall. A false aneurysm is most usually traumatic in origin. Histological features suggestive of trauma include dissection of the arterial wall and foci of medial and intimal fibrosis away from the site of the aneurysm (Paul et al. 1980). A history of trauma, which, as commented above, may be years

101

102

Subarachnoid haemorrhage and cerebrovascular traumatic pathology

(a)

(b)

Figure 9.9 (a) Carotid artery thrombosis occurring following trauma to the neck. (b) Bruising to face/neck after assault with ­subsequent carotid artery thrombosis. (c) Carotid dissection with thrombosis of vessel.

previously, should be sought. Aneurysms in childhood, particularly if peripheral, are commonly traumatic or inflammatory in origin. Aneurysms secondary to penetrating trauma occur most commonly in teenage boys suffering gunshot wounds. Those secondary to non-penetrating trauma occur at the skull base or in the periphery with motor vehicle accidents or falls being the most common cause of injury. Peripheral traumatic aneurysms can further be divided into distal anterior cerebral artery aneurysms secondary to trauma against the falcine edge and distal cortical artery aneurysms associated with an overlying skull fracture (Paul et al. 1980, Buckingham et al. 1988). Traumatic paediatric anterior cerebral artery aneurysms may commonly have a delayed presentation and may not necessarily involve major trauma with predisposition to this form of injury likely explained by anatomical consideration such as the edge of the anterior falx or sphenoidal ridge (Laurent et al. 1981).

Traumatic arteriovenous fistula These are rare and mostly found associated with skull base fractures; most commonly, they are carotid-cavernous fistulae where the internal carotid artery passes adjacent to the venous cavernous sinuses. These most commonly occur in young men and may be delayed sometime after the trauma (Kalimo et al. 2002; Tsutsumi et al. 2002). A method for exposing the intraosseous portion of the carotid arteries has been described by Langlois and Little (2003) as an aid to post-mortem examination. Most traumatic carotid-cavernous fistulas are Type A or high-flow shunts. Other sites for traumatic arteriovenous fistula include the middle meningeal artery with a meningeal vein and the extradural vertebral artery. Where allied with other traumatic vascular lesions, consideration again needs to be given to arteriopathic disorders of collagen.

Strokes and trauma

 

Figure 9.10 Multivessel trauma due to kicks delivered to the head (a) posterior inferior cerebellar artery false traumatic aneurysm accounting for more limited basal traumatic subarachnoid haemorrhage with initial survival. (b) Death followed delayed MCA ­territory stroke due to (c) traumatic internal carotid dissection complicated by occlusive thrombosis.

Strokes and trauma Ischaemic and haemorrhagic strokes may complicate traumatic head injuries as secondary pathology. The anatomy and the timing of secondary traumatic strokes are of assistance in determining the underlying aetiology. Extrinsic compression due to herniation may cause not only posterior cerebral arterial strokes but also anterior cerebral artery/ pericallosal territorial strokes and ganglionic strokes due to compromise of lenticulostriate arterial vessels. Upwards herniation of the posterior fossa structures may further compromise the superior cerebellar arterial vascular territories within the cerebellum. Traumatic subarachnoid haemorrhage may lead to vasospasm-associated strokes.

This typically has a delayed onset between 3 and 15 days, most frequently between 7 and 10 days and may resolve by 21 days post-haemorrhage (Daou et al. 2019). Radiographic vasospasm usually develops between 5 and 15 days after the original haemorrhage and is associated with a clinically apparent delayed ischaemic neurological deficit in onethird of those affected (Keyrouz and Diringer 2007). Arterial dissections may be complicated by local occlusive thrombosis or by thromboembolic phenomena which may lead to haemorrhagic infarcts. Foreign bodies such as shotgun pellets may also cause strokes, including multiple occlusive strokes. Primary traumatic ganglionic bleeds require distinction from non-traumatic, e.g. hypertensive bleeds.

103

104

Subarachnoid haemorrhage and cerebrovascular traumatic pathology

Cerebral venous thrombosis and infarction The major cerebral venous system is regrettably too often neglected during routine examination of brain despite the ease with which the major venous sinuses can be examined. Unexplained severe brain swelling or haemorrhagic infarcts not conforming to typical arterial vascular territories should prompt attention to the venous circulation of the brain. Cerebral venous thrombosis most often occurs spontaneously. In neonates sepsis, congenital heart disease and dehydration constitute common causes whereas infections are a more common cause in older children. Conditions such as nephrotic syndrome, malignancy, DIC, pregnancy, the use of oral contraceptives, leukemias or genetic or acquired clotting abnormalities may promote clinically significant cerebral venous thrombosis. Head injuries may also be complicated by cerebral venous thrombosis. Here it may occur in association with skull fractures, but it may also complicate closed head injuries (Yokota et al. 2006; Ferrera et al. 1998). The superior sagittal sinus is most commonly involved, resulting in parasagittal congestion and parenchymal haemorrhage involving both cortex and underlying white matter. Microscopically the appearances are of haemorrhagic infarction although in venous infarction neutrophils are typically abundant.

External pressure to the neck External pressure to the neck such as occurs with strangulation may lead to neuropathological complications in survivors. These are most commonly related to either the effects of global hypoxia-ischaemia or due to arterial pathology with thrombosis and arterial territory infarction. Milligan and Anderson (1980) reported two cases of stroke within days of attempted strangulation with thrombotic occlusion of the internal carotid artery. Global hypoxia-ischaemia has been reported giving rise to bilateral basal ganglia pathology (Simpson et al. 1987; Kirton and Riopelle 2001). The time relationship to trauma as well as any underlying disease need to be evaluated (Figure 9.11).

Figure 9.11 Haemorrhagic change in the right caudate nucleus and both putaminal nuclei, following survival of 24–48 hours after attempted strangulation.

References Aarabi, B. 1988. Traumatic aneurysms of brain due to high velocity missile head wounds. Neurosurgery, 22(6 Pt 1), 1056–1063. Alberts, M.J. 1999. Subarachnoid hemorrhage and intracranial aneurysms. In Alberts, M.J. (ed), Genetics of Cerebrovascular Disease. Futura, pp 237–259. Amirjamshadi, A., Rahmat, H. and Abbassioun, K. 1996. Traumatic aneurysms and arteriovenous fistulas of intracranial vessels associated with penetrating head injuries occurring during war: principles and pitfalls in diagnosis and management. A survey of 31 cases and review of the literature. J Neurosurg, 84, 769–780. Barry, K.J. and Scott, R.M. 1979. Effect of intravenous ethanol on cerebral vasospasm produced by subarachnoid blood. Stroke, 10, 535–537. Bavinski, G., Viller, M. and Knoop, E. et al. 1997. False aneurysms of the ­intracavernous carotid artery – report of 7 cases. Acta Neurochir, 139, 37–43. Bowen, D.A.L. 1984. Ruptured berry aneurysms: a clinical, pathological and forensic review. Forensic Sci Int, 26, 227–234. Bromilow, A. and Burns, J. 1985. Technique for removal of the vertebral arteries. J Clin Pathol, 38, 1400–1402. Buckingham, M.J., Crone, K.R. and Ball, W.S. et al. 1988. Traumatic intracranial aneurysms in childhood: two cases and a review of the literature. Neurosurgery, 22, 398–408. Cameron, J.M. and Mant, A.K. 1972. Fatal subarachnoid haemorrhage associated with cervical trauma. Med Sci Law, 12, 66–70. Chen, H.-J., Liang, C.-L., Lu, K. and Lui, C.-C. 2003. Rapidly growing internal carotid artery aneurysm after amphetamine abuse. Case report. Am J Forensic Med Pathol, 24, 32–34. Chyatte, D., Chen, D.L., Bronstein, K. and Brass, L.M. 1994. Seasonal fluctuations in the incidence of intracranial aneurysm rupture and its relationship to changing climatic conditions. J Neurosurg, 81, 525–530. Cloft, H.J., Kallmes, D.F. and Kallmes, M.H. et al. 1998. Prevalence of cerebral aneurysms in patients with fibromuscular dysplasia: a reassessment. J Neurosurg, 88, 436–440. Coast, G.C. and Gee, D.J. 1984. Traumatic subarachnoid haemorrhage; an alternative source. J Clin Pathol, 37, 1245–1248. Coelho, L.G.B.S.A., Costa, J.M.D. and Silva, E.I.P.A. 2016. Non-aneurysmal spontaneous subarachnoid haemorrhage: perimenencephalic vs non-perimesencephalic. Rev Bras Ter Intensiva, 28, 141–146. Cogbill, T.H. and Sullivan, H.G. 1995. Carotid artery pseudoaneurysm and pellet embolism to the middle cerebral artery following a shotgun wound of the neck. J Trauma, 39, 763–767. Contostavlos, D.L. 1971. Massive subarachnoid haemorrhage due to laceration of the vertebral artery associated with fracture of the transverse process of the atlas. J Forensic Sci, 16, 40–56. Contostavlos, D.L. 1995. Isolated basilar traumatic subarachnoid haemorrhage: an observer’s 25 year re-evaluation of the pathogenetic possibilities. Forensic Sci Int, 73, 61–74. Cooper, P.N., Sunter, J.P. and McCarthy, J.H. 1999. Traumatic subarachnoid haemorrhage: a 10 year case study. J Pathol, 187(S), 7A. de Recondo, A., Woimant, F. and Ille, O. et al. 1995. Post-traumatic ­common carotid artery dissection. Stroke, 26, 705–706. Daou, B.J., Koduri, S. and Thompson, B.G. et al. 2019. Clinical and experimental aspects of aneurysmal subarachnoid haemorrhage. CNS Neurosci Ther, 25, 1096–1112. Deck, J.H. and Jagadha, V. 1986. Fatal subarachnoid hemorrhage due to traumatic rupture of the vertebral artery. Arch Pathol Lab Med, 110, 489–493. Djokic, V., Savic, S. and Atanasijevic, T. 2003. Medicolegal diagnostic value and clinical significance of traumatic incomplete tears of the basilar artery. Am J Forensic Med Pathol, 24, 208–213. Dowling, G. and Curry, B. 1988. Traumatic basal subarachnoid haemorrhage: report of six cases and a review of the literature. Am J Forensic Med Pathol, 9, 23–31. Endo, S., Takaku, A. and Aihara, H. et al. 1980. Traumatic cerebral aneurysm associated with underlying skull fracture. Report of two infancy cases. Child’s Brain, 6, 131–139.

References Fantaneanu, T., Veinot, J.P., Torres, C., Alhazzaa, M. and Stotts, G. 2011. Cervical arterial dissections due to segmental mediolytic arteriopathy. Neurology, 77, 295–297. Farag, A.M., Franks, A. and Gee, D.J. 1988. Simple laboratory experiments to replicate some of the stresses on vertebro-basilar arterial walls – an investigation of possible mechanisms of traumatic subarachnoid haemorrhage. Forensis Sci Int, 38, 275–284. Ferrera, P.C., Pauze, D.R. and Chan, L. 1998. Sagittal sinus thrombosis after closed head injury. Am J Emerg Med, 16, 382–385. Ford, R. 1956. Basal subarachnoid haemorrhage and trauma. J Forensic Sci, 1, 117–126. Forget, T.R., Benitez, R. and Veznedaroglu, E. et al. 2001. A review of size and location of ruptured intracranial aneurysms. Neurosurgery, 49, 1322–1325. Gee, D.J. 1982. Traumatic subarachnoid haemorrhage. Proceedings of the 12th Congress of the International Academy of Forensic and Social Medicine. Vienna: Egermann, 495–498. Gonsoulin, M., Barnard, J.J. and Prahlow, J.A. 2002. Death resulting from ruptured cerebral artery aneurysm 219 cases. Am J Forensic Med Pathol, 23, 5–14. Gray, J.T., Peutz, S.M., Jackson, S.L. and Green, M.A. 1999. Traumatic subarachnoid haemorrhage: a 10 year case study and review. Forensic Sci Int, 105, 13–23. Gross, A. 1990. Traumatic basal subarachnoid haemorrhages: postmortem examination material analysis. Forensic Sci Int, 45, 53–61. Haldeman, S., Kohlbeck, F.J. and McGregor, M. 1999. Risk factors and precipitating neck movements causing vertebrobasilar artery dissection after cervical trauma and spinal manipulation. Spine, 24, 785–794. Hammes, E.M. Jr. 1944. The reaction of the meninges to blood. Arch Neurol Psychiatry, 52, 505–514. Harland, W.A., Pitts, J.F. and Watson, A.A. 1983. Subarachnoid haemorrhage due to upper cervical trauma. J Clin Pathol, 36, 1335–1341. Hillbom, M. and Kaste, M. 1981. Does alcohol intoxication precipitate aneurysmal subarachnoid haemorrhage? J Neurol Neurosurg Psychiatry, 44, 523–526. Hinse, P., Thie, A. and Lachenmayer, L. 1991. Dissection of the extracranial vertebral artery: report of four cases and review of the literature. J Neurol Neurosurg Psychiatry, 54, 863–869. Holmes, B. and Harbaugh, R.E. 1993. Traumatic intracranial aneurysms: a contemporary review. J Trauma-Injury Infect Crit Care, 35, 855–860. Johnson, C.P., How, T. and Scraggs, M. et al. 2000. A biochemical study of the human artery with implications for fatal arterial injury. Forensic Sci Int, 109, 169–182. Johnson, C.P., Baugh, R., Wilson, C.A. and Burns, J. 2001. Age related changes in the tunica media of the vertebral artery: implications for the assessment of vessels injured by trauma. J Clin Pathol, 54, 139–145. Johnson, C.P., Murphy, M., Johnson, G.A., Wills, S.M. and Medcalf, J.E. 2017. A simple post mortem room angiography method for the investigation of traumatic basal subarachnoid haemorrhage. Forensic Sci Med Pathol, 13, 4–9. Juvela, S. 1992. Alcohol consumption as a risk factor for poor outcome after aneurysmal subarachnoid haemorrhage. BMJ, 304, 1663–1667. Juvela, S., Hillbom, M., Numminen, H. and Koskinen, P. 1993. Cigarette smoking and alcohol consumption as risk factors for aneurysmal subarachnoid hemorrhage. Stroke, 24, 639–646. Kalaria, R., Ferrer, I. and Love, S. 2015. Vascular disease, hypoxia and related conditions. In Love, S. Budka, H., Ironside, J.W. and Perry, A. (eds), Greenfields Neuropathology, 9th edn, Vol 1. CRC Press. Karhunen, P.J., Kauppila, A., Penttila, A. and Erkinjuntti, T. 1990. Vertebral artery rupture in traumatic subarachnoid haemorrhage detected by postmortem angiography. Forensic Sci Int, 44, 107–115. Keyrouz, S.G. and Diringer, M.N. 2007. Clinical review: prevention and therapy of vasospasm in subarachnoid haemorrhage. Crit Care, 11(4), 220–229. Kieck, C.F. and de Villiers, J.C. 1984. Vascular lesions due to transcranial stab wounds. J Neurosurg, 60, 42–46.

Kim, S., Kim, M., Lee, B.W., Kim, Y., Choi, Y. and Seo, J.S. 2015. Investigation of bleeding focus in the intracranial vertebral artery with the use of posterior neck dissection in traumatic basal subarachnoid haemorrhage. J Forensic Leg Med, 34, 151–154. Kirton, C.A. and Riopelle, R.J. 2001. Meige syndrome secondary to basal ganglia injury: a potential cause of acute respiratory distress. Canadian J Neurol Sciences, 28, 167–173. Knight, B. 1979. Trauma and ruptured cerebral artery aneurysm. BMJ, i, 1430–1431. Koszyca, B., Gilbert, J.D. and Blumbergs, P. 2003. Traumatic subarachnoid hemorrhage and extracranial vertebral artery injury. A case report and review of the literature. Am J Forensic Med Pathol, 24, 114–118. Langlois, N.E. and Little, D. 2003. A method for exposing the intraosseous portion of the carotid arteries and its application to forensic case work. Am J Forensic Med Pathol, 24, 35–40. Larson, P.S., Reisner, A. and Morassutti, D.J. et al. 2000. Traumatic intracranial aneurysms. Neurosurg Focus, 8, 1–6. Laurent, J.P., Cheek, W.R., Mims, T. and McCluggage, C.W. 1981. Traumatic intracranial aneurysm in an infant: case report and review of the literature. Neurosurgery, 9, 303–306. Leadbeatter, S. 1994. Extracranial vertebral artery injury – evolution of a pathological illusion. Forensic Sci Int, 67, 35–40. Lee, K.P., Carlini, W.G., McCormick, G.F. and Albers, G.W. 1995. Neurologic complications following chiropractic manipulation: a survey of California neurologists. Neurology, 45, 1213–1215. Lindenberg, R. and Freytag, E. 1970. Brainstem lesions characteristic of traumatic hyperextension of the head. Arch Pathol, 90, 509–515. McCarthy, J.H., Sunter, J.P. and Cooper, P.N. 1999. A method for demonstrating the source if bleeding in cases of traumatic subarachnoid haemorrhage. J Pathol, 187, 30A. Marek, Z. 1981. Isolated sub-arachnoid hemorrhage as a medic-legal problem. Am J Forensic Med Pathol, 2, 19–22. Martin, P.J. and Enevoldson, T.P. 1998. Vertebral aneurysm due to isolated spontaneous dissection of the intracranial vertebral artery. J Neurol Neurosurg Psychiatry, 65, 946. Medcalf, J.E., Johnson, C.P., Taktak, A. and Grabherr, S. 2016. Variations in the anatomy of the vertebral artery loop segment – a potential factor for traumatic basal subarachnoid haemorrhage? Forensic Sci Med Pathol, 12, 159–165. Milligan, N. and Anderson, M. 1980. Conjugal disharmony: a hitherto unrecognized cause of strokes. BMJ, 281, 421–422. Mitchell, J. 2002. Vertebral artery atherosclerosis: a risk factor in the use of manipulative therapy? Physiother Res Int, 7, 122–135. Mokri, B., Piepgras, D.G. and Houser, O.W. 1988. Traumatic dissections of the extracranial internal carotid artery. J Neurosurg, 68, 189–197. Norris, J.W., Beletsky, V. and Nadaereishvili, Z.G. 2000. Sudden neck movement and cervical artery dissection. CMAJ, 163(1), 38–40. Nunnink, L. 2002. Blunt carotid artery injury. Emerg Med (Fremantle), 14, 412–421. Opeskin, K. 1995. Traumatic pericallosal artery aneurysm. Am J Forensic Med Pathol, 16, 11–16. Opeskin, K. and Burke, M.P. 1998. Vertebral artery trauma. Am J Forensic Med Pathol, 19, 206–217. Paul, G.A., Shaw, C.-M. and Wray, L.M. 1980. True traumatic aneurysm of the vertebral artery: case report. J Neurosurg, 53, 101–105. Pickup, M.J. and Pollanen, M.S. 2011. Traumatic subarachnoid hemorrhage and the COL3A1 gene: emergence of a potential causal link. Forensic Sci Med Pathol, 7, 192–197. Pollanen, M.S., Deck, J.H.N. and Blenkinsop, B. 1996. Injury of the tunica media in fatal rupture of the vertebral artery. Am J Forensic Med Pathol, 17, 197–201. Prabhu, V., Kizer, J. and Patil, A. et al. 1996. Vertebrobasilar thrombosis associated with nonpenetrating cervical spine trauma. Trauma, 40(1), 130–137. Purgina, B. and Milroy, C.M. 2015. Fatal traumatic aneurysm of the posterior inferior cerebellar artery with delayed rupture. Forensic Sci Int, 247, E1–E5.

105

106

Subarachnoid haemorrhage and cerebrovascular traumatic pathology Rinkel, G.J.E., Gijn, J.V. and Wijdicks, E.F.M. 1993. Subarachnoid haemorrhage without detectable aneurysm. A review of causes. Stroke, 24, 1403–9. Rommel, O., Niedeggen, A. and Tegenthoff, M. et al. 1999. Carotid and vertebral artery injury following severe head or cervical spine trauma. Cerebrovasc Dis, 9, 2002–2009. Rothwell, D.M., Bondy, S.J. and Williams, J.I. 2001. Chiropractic manipulation and stroke: a population-based case-control study. Stroke, 32, 1054–1060. Rutherfoord, S.G., Dada, M.A. and Nel, J.P. 1996. Cerebral infarction and intracranial arterial dissection in closed head injury. Am J Forensic Med Pathol, 17(1), 53–57. Saito, K., Baskaya, M.K. and Shibuya, M. et al. 1995. False traumatic aneurysm of the dorsal wall of the supraclinoid internal carotid artery. Neurol Med Chir (Tokyo), 35, 886–891. Saukko, P. and Knight, B. 2016. Knight’s Forensic Pathology, 4th edn. CRC Press. Schievink, W.I., Wijdicks, E.F. and Parisi, J.E. et al. 1995. Sudden death from aneurysmal subarachnoid haemorrhage. Neurology, 45, 871–874. Schievink, W.I. 2001. Spontaneous dissection of the carotid and vertebral arteries. N Engl J Med, 344, 898–906. Simpson, R.K. Jr, Goodman, J.C. and Rouah, E. et al. 1987. Late neuropathological consequences of strangulation. Resuscitation, 15, 171–185. Simonsen, J. 1963. Traumatic subarachnoid haemorrhage in alcohol intoxication. J Forensic Sci, 8, 97–116. Simonsen, J. 1967. Fatal subarachnoid haemorrhage in relation to minor head injuries. J Forensic Med, 14, 146–155. Simonsen, J. 1984. Fatal subarachnoid haemorrhage in relation to minor injuries in Denmark from 1967 to 1981. Forensic Sci Int, 24, 57–63. Stehbens, W.E. 1989. Etiology of intracranial berry aneurysms. J Neurosurg, 70, 823–31.

Takahara, M., Ogata, T., Abe, H., Higashi, T., Morishita, T., Takano, K. and Inoue, T. 2019. The Comparison of Clinical Findings and Treatment Between Unilateral and Bilateral Vertebral Artery Dissection. J Stroke Cerebrovasc Dis. 28, 1192–1199. Tatsuno, Y. and Lindenberg, R. 1974. Basal subarachnoid hematomas as sole intracranial traumatic lesions. Arch Pathol, 97, 211–215. Tsutsumi, M., Kazekawa, K. and Tanaka, A. et al. 2002. Traumatic middle meningeal artery pseudoaneurysm and subsequent fistula formation with the cavernous sinus: case report. Surg Neurol, 58, 325–328. Vanezis, P. 1986. Vertebral artery injuries in road traffic accidents: a postmortem study. J Forensic Sci, 26, 281–291. Vanezis, P. 1979. Techniques used in the evaluation of vertebral artery trauma at postmortem. Forensic Sci Int, 13, 159–165. Ventureyra, E.C. and Higgins, M.J. 1994. Traumatic intracranial aneurysms in childhood and adolescence. Case reports and review of the literature. Child’s Nerv Syst, 10, 361–379. Walder, B., Bründler, M.A. and Totsch, M. et al. 2002. Inference of the type and rate of subarachnoid fluid infusion on lethal neurogenic pulmonary oedema in rats. J Neurosurg Anaesthesiol, 14, 194–203. Weller, R.O. 1995. Subarachnoid haemorrhage and myths about saccular aneurysms. J Clin Pathol, 48, 1078–1081. Whittington, R.M. 1980. Alcohol related deaths. Birmingham Coroner’s records. BMJ, 284, 1162. Wilkinson, I.M.S. 1972. The vertebral artery. Extracranial and intracranial structure. Arch Neurol, 27, 392–396. Wilks, S. 1859. Sanguineous meningeal effusion. Guy’s Hosp Rep, 5, 119. Yokota, H., Eguchi, T. and Nobayashi, M. et al. 2006. Persistent intracranial hypertension caused by superior sagittal sinus stenosis following depressed skull fracture. Case report and review of the literature. J Neurosurg, 104, 849–852.

10

Contusional brain injury and intracerebral haemorrhage Traumatic and non-traumatic Helen Whitwell

Contusional and lacerating brain injury are characteristic of blunt head trauma. A contusion is ‘bruising’ of the surface of the brain as a result of damage to the small blood vessels. The pia arachnoid is not breached in contusional injury; however, with lacerating injury both the pia arachnoid and the underlying brain substance are torn. Contusional injury may produce varying sizes of intracerebral haematoma. Intracerebral haemorrhage (ICH) is most commonly natural rather than traumatic. This chapter covers both natural and traumatic ICHs concentrating, in particular, on diagnostic issues on neuropathological examination.

Contusional brain injury Macroscopic appearances Acute contusions typically occur over the crests of the gyri and have a conical appearance with the widest part in the superficial position and the apex deeper (Figure 10.1). They have a red or red-purple appearance when fresh. Larger contusions extend into white matter and can give rise to significant ICH. There is usually associated subarachnoid haemorrhage. Complications such as burst lobe and delayed ICH may occur (see later). Old contusions appear gold-brown (Figure 10.2). They are not infrequently seen as an incidental finding. Methods for quantification of the amount of contusional injury have been developed, initially by Adams et al. (1985). Their method is based on the extent and depth of the contusional injury in various regions of the brain. A contusion index can be given for the whole brain as well as for various anatomical sites. A further quantitative approach by Ryan et al. (1994) uses a sector scoring method. Both of these studies have provided considerable data for both individual cases and series of cases which have been extremely useful in pathological and biomechanical studies. However, for most medico-legal cases, description with diagrams and photographs will suffice.

Microscopic appearances Perivascular haemorrhages with larger areas of haemorrhage are seen in the early stages (Figure 10.3). Infiltration by inflammatory cells/leukocyte polymorphs is the first stage to be seen, initially with intravascular accumulation, but subsequently in the perivascular location (Holmin et al. 1998; Oehmichen et al. 2003) (Figure 10.4a). The time course is variable – some authors report intraparenchymal leukocyte polymorphs within a few hours (Anderson and Opeskin 1998). Within days, the intraparenchymal infiltrate comprises a mixture of T lymphocytes, reactive microglia and monocytes/macrophages, as well as the neutrophil polymorphs (Holmin et al. 1998; Engel et al. 2000). Generally, these appear from around 72 hours. Haemosiderin

may be seen as early as 48 hours or so, although detailed pathological studies are lacking. Capillary proliferation, astrocytosis and accumulation of haemosiderin macrophages are identifiable with increasing survival, after about 4–5 days (Figure 10.4b). Neuronal changes, either as neuronal cytoplasmic eosinophilia (red cell change) or are seen in hypoxic-ischaemic injury, can be observed in adjacent neurons, see also Chapter 12. It should be said that the time frames can only be approximate and factors such as age may play a role. Old contusions histologically resemble areas of previous infarction with meningeal fibrosis and gliosis, as well as haemosiderin macrophages (Figure 10.5a and b). The underlying white matter shows variable demyelination. Various macrophage immunohistochemical markers have been assessed to identify timing of histological changes (Oehmichen et al. 2009). However, for most practical circumstances, they have limited use.

Delayed intracerebral haemorrhage in association with contusional injury It is well recognised that intracerebral haematomas in association with contusional injury may be delayed. Presentation is usually within 48 hours of diagnosis but occasionally longer (Baratham and Dennyson 1972; Nanassis et al. 1989). This may mean that a patient with relatively minor contusional injury presents with a rapidly deteriorating picture with features of a space-occupying lesion. Computed tomography (CT) or magnetic resonance imaging (MRI) will reveal the development of such a haematoma. The relationship between development of such a haematoma and pre-existing disease, e.g. hypertension, is unclear. Other complicating issues include the role of alcohol, drugs such as cocaine and amphetamines, as well as decreased blood coagulation as a result of either anticoagulant therapy or disease such as chronic liver disease (Mackenzie 1996).

Burst lobe This occurs when a haematoma, most commonly in the temporal region, but also seen in the frontal region, is in continuity with the cerebral surface and adjacent acute subdural haematoma (Figure 10.6). This occurs in the setting of contusional injury and represents a severe end of the spectrum that may be seen in both coup and contre-coup injuries.

Coup contusions These are contusions that occur beneath the site of impact. They are less frequent than contre-coup contusions (Dawson et al. 1980). Experimental work has indicated that tissue damage is related to the mechanical loading (Shreiber et al. 1999) see also Chapter 4. The common sites are over the hemispheres or

Contusional brain injury and intracerebral haemorrhage

108

Figure 10.3 Perivascular haemorrhages – early contusional injury.

(a)

cerebellum where the brain is exposed to the relevant object (Figure 10.7). This contrasts with contre-coup injury where the base of the brain and temporal/frontal poles are the typical sites. This type of injury is most commonly seen where a relatively small object strikes the skull, causing the skull to ‘bend’ (transient deformation) into the brain and resulting in ‘suction’ forces as well as direct compressive injury (DiMaio and DiMaio 2001). A classic example would be as the result of a blow from an object

(b)

Figure 10.1 (a) Bilateral temporal lobe contusions. (b) Close-up of an area of temporal lobe contusion.

(a)

(b)

Figure 10.2 Old contusions of the frontal and temporal regions.

Figure 10.4 (a) Neutrophil polymorphs in contusional injury 2 days in age. (b) CD68 – macrophages in an area of contusion after several days.

Contusional brain injury

(a)

Figure 10.7 Coup contusion beneath area of multiple hammer blows.

Contre-coup contusions

(b)

Figure 10.5 (a) Small old contusion. (b) Perls’ stain showing haemosiderin macrophages. such as a hammer. In general terms, there is correlation between the degree of injury sustained and the mass/velocity relationship of the object (Leestma 2014). The external appearance of the wound may indicate the implement (e.g. hammer, baton), but it should be recognised that the features of any wound may not always be specific. Thus, a hammer blow may simply leave a bruise depending on the force used. Further aid to the identification of a particular weapon may be the characteristics of any skull fracture and external wounds (see Chapter 6).

Figure 10.6 ‘Burst’ lobe haematoma in continuity with cerebral surface.

These are generally said to occur at sites opposite to the site of impact. The location of any scalp injuries is a good aid to identify the primary site of impact (Ratnaike et al. 2011). The characteristic locations are the frontal poles, the orbital surfaces of the frontal lobes and the temporal lobes including the temporal poles (Figure 10.8). However, it is now recognised that frontal and temporal contusions may be found when the impact site is not directly opposite any impact. The classic situation in which the pattern of contre-coup contusional injury is seen is when the moving head strikes the ground. This may produce a combination of an occipital laceration with surrounding abrasion, scalp bruising, underlying skull fracture, subdural haematoma and contusional brain injury. In forensic practice, this may follow a punch or blow to the face. The mechanism of contre-coup contusional injury has been the subject of considerable work. Biomechanics are covered in Chapter 4. It is recognised that the brain rebounds in the skull with linear acceleration after ground impact, with the anterior parts coming into contact with the rough floor of the anterior fossa, wing of sphenoid and petrous temporal bone (Saukko and Knight 2016, DiMaio and DiMaio 2001). Other work has also identified that a ‘vacuum’ effect occurs at the site of a contrecoup lesion, with pressure momentarily increased at the site of impact and decreased diametrically opposite (Yanagida et al. 1989). It is also recognised that movement of the brain towards the impact site causes opposite tensile strains. If these forces are larger than vascular tolerance, contusional injury will occur. Other factors include the anatomy of the skull with rough irregular bony areas in the orbital roofs and temporal fossae. Where there is a side impact to the head, contusions are seen in the opposite side of the brain. Contre-coup contusional injury is rare in the occipital or parietal region. Contusional injury is more severe in the presence of a skull fracture. Larger objects such as baseball bats, tree branches and other heavy objects may involve much greater kinetic energy and produce contre-coup as well as coup injuries (Leestma 2016;

109

Contusional brain injury and intracerebral haemorrhage

110

(b)

(a)

(c)

Figure 10.8 (a) Undersurface of brain with contre-coup contusional/lacerating injury. (b) Frontal contre-coup contusions. (c) Extensive temporal contre-coup contusions.

Morrison et al. 1998). This is particularly so when the fixed head is struck, i.e. when the victim is on the ground or against an unyielding surface. It would, however, be exceptional in this context not to see some coup injury. In the assault situation, which not infrequently involves a number of participants, the contribution of an individual injury to the final picture may be difficult to establish from the pathology alone.

Fracture contusions These occur directly beneath the site of a skull fracture as a direct result of such (Figure 10.9). If the fracture is linear, the contusion tends to follow a similar pattern. Fracture contusions may not relate to the point of impact because the fracture may originate some distance away. The frontal region tends to be an atypical site for fracture contusional injury; the majority of the cases are contre-coup contusions. Posterior fossa contusional injury is almost entirely related to skull fractures. These may expand to produce haematomas (Vrankovic et al. 2000). Contrecoup posterior fossa contusions are extremely rare because the cerebellum and occipital lobes appear to be well protected possibly as a result of the relatively smooth surface of the skull, as well as the ability of the frontal structures to absorb energy from an impact.

Figure 10.9 Inferior cerebellar hemisphere showing contusion in association with occipital fracture.

Lacerating injury Lacerating injury may be seen in either coup or contre-coup contusional injuries. However, lacerating injury is well recognised as occurring in the absence of contusional injury,

Other traumatic intracerebral haematomas

Other traumatic intracerebral haematomas These include the types listed below. The pathogenesis of the first three listed is outlined in Chapter 11. These are characteristic markers for traumatic axonal injury., Brainstem haemorrhages are seen in association with effects of raised pressure within the cranial cavity as well as in traumatic axonal injury (Chapters 11 and 12). Not infrequently in an individual case, a variety of findings may be present:

(a)

1. Petechial white matter haemorrhages (diffuse vascular injury) 2. Tissue tear haemorrhages 3. Ganglionic haematoma/deep basal haematoma 4. Traumatic intraventricular haemorrhage 5. Brainstem haemorrhages

Petechial white matter haemorrhages (diffuse vascular injury) These occur as a result of very severe head injury with short survival. The pattern is seen macroscopically and more readily microscopically with small haemorrhages occurring around blood vessels (see Chapter 11). These are particularly prominent in the frontal and temporal lobes, adjacent to midline structures and in the stem adjacent to the aqueduct and floor of the fourth ventricle. They are often associated with ‘tissue tear’ haemorrhages (see below). The precise aetiology is unclear but is probably related to vascular damage at the time of injury in association with primary axotomy. The majority of cases occur after road traffic incidents. Macroscopic differential diagnosis includes rare conditions such as acute haemorrhagic leukoencephalopathy (see Chapter 18). (b)

Figure 10.10 (a) Lacerating injury in association with extensive skull fracture as a result of a fall from a height. (b) Histology of acute lacerating injury. particularly where falls from significant heights are involved with open skull fractures. The location of these lacerations does not follow the pattern of coup or contre-coup injuries as described above and may involve any part of the brain. Histology shows tissue disruption and acute haemorrhage (Figure 10.10a,b).

Herniation contusions These are either damage to the parahippocampal gyrus impacting against the tentorium cerebelli or the cerebellar tonsils impacting against the foramen magnum at the time of injury. They are usually associated with significant injury such as impact with a vehicle. They are also seen in gunshot wounds. Internal hernia is covered in Chapter 12.

Cerebellar haemorrhages These most commonly are either traumatic in origin, usually in association with cerebellar contusional injury, particularly in the presence of a fracture contusion (Vrankovic et al. 2000), or primarily hypertensive. This latter condition affects the central cerebellum including the dentate nucleus. Other rarer causes include malformations (see later).

Tissue tear haemorrhages These are associated with traumatic axonal injury and are macroscopic indicators of that diagnosis. They are areas of ICH of varying size caused by injury to vascular structures and axons as a result of shearing rather than impact. They represent maximal areas of acceleration-induced brain injury and are discussed in detail in Chapter 11. Larger linear or flame-shaped haemorrhages are known as gliding contusions and occur at the severe end of injury. Locations are predominantly in the parasagittal white matter, corpus callosum, centrum semiovale and deep grey matter as well as the rostral brainstem and pons. Not infrequently in lesser degrees of traumatic axonal injury foci may only be seen in the corpus callosum (Figure 10.11). Histological examination shows variable haemorrhage, often in a perivascular location, and associated axonal pathology with changes in b-amyloid precursor protein staining as covered in Chapter 11. Axonal pathology is almost invariably widespread elsewhere, further supporting the diagnosis.

Ganglionic haematomas The occurrence of these has been better documented with the advent of CT and MRI. They occur deep within the brain tissue in the basal ganglia region, varying in size from being relatively small (0.1 cm) to considerably larger. This type of haemorrhage is frequently associated with diffuse traumatic axonal injury (dTAI) with a poor prognosis (Adams et al. 1986). They are most commonly identifiable shortly after injury with

111

112

Contusional brain injury and intracerebral haemorrhage

Figure 10.11 Haemorrhages in the corpus callosum in traumatic axonal injury as a result of severe blunt head injury during an assault. a period of lucidity being rare. This contrasts with haematomas in association with contusional injury, in which there may be variable periods of lucidity before presentation (see below). There is some evidence to indicate that they enlarge over a period of time after injury (Boto et al. 2001). The importance of these, in the context of a motor vehicle incident, is that they may resemble primary hypertensive ICHs and bring into question the cause of the accident. Distinguishing features include the following: evidence of other traumatic brain injury such as corpus callosum, brainstem and white matter haemorrhages in the locations associated with traumatic axonal injury (Figure 10.12). Impact injury may also be seen with skull fractures as well as contusional brain damage. Spontaneous ICH tends to occur in an older age group (Siddique et al. 2002). It is difficult to substantiate a diagnosis of a primary hypertensive haemorrhage where there is no documented evidence of hypertension or post-mortem findings to support a diagnosis,

Figure 10.13 Intraventricular haemorrhage with disruption of the ventricular lining in a case of rapid death with head injury. in particular left ventricular hypertrophy (Hangartner et al. 1985) or microscopic vascular changes to indicate hypertension (Mackenzie 1996; Chapter 18). Full toxicological analysis should be undertaken because it is increasingly recognised that alcohol as well as other drugs may cause/contribute to a primary ICH, as well as coagulation disorders including anticoagulant therapy (Yajima et al. 2001; see Chapter 18). Examination of available medical records as well as correlation of the circumstances of the accident may also aid in interpretation.

Traumatic intraventricular haemorrhage This is seen in a variety of situations: • In association with the entity of traumatic subarachnoid haemorrhage caused by rupture or tear to the vertebrobasilar system (see Chapter 9). • In association with severe head injury – the precise source cannot usually be identified macroscopically but histology may show haemorrhages in association with vessels in the interventricular septum or choroid plexus. Rupture of ependymal vessels has also been identified (Makino et al. 2001). These tend to be in the posterior aspects of the ventricular system (Figure 10.13). • Extension of haemorrhage from within the brain parenchyma secondary to extension of ICH.

Brainstem haemorrhages These may occur as part of dTAI and the location of these differs from those associated with raised intracranial pressure and primary hypertensive haemorrhage. However, cerebral swelling with dTAI occurs and precise identification may be difficult. Haemorrhages secondary to raised pressure are commonly centrally placed within the medulla and pons. Primary traumatic haemorrhages classically involve the dorsolateral quadrant of the rostral brainstem (see Chapter 11).

Natural intracerebral haemorrhage Figure 10.12 Ganglionic haematoma caused by diffuse traumatic axonal injury. Note evidence of other areas of traumatic haemorrhage, which help to differentiate from a primary hypertensive haemorrhage.

Spontaneous ICH together with subarachnoid haemorrhage accounts for around 15 per cent of all strokes (Mendelow 1991). A stroke is defined as a sudden onset neurological deficit of vascular origin lasting more than 24 hours or causing death. The mortality is high and related to site, size and presenting Glasgow Coma Scale (GCS) score (Mackenzie 1996). Brainstem

ICH in association with cerebral amyloid angiopathy haematomas are almost uniformly fatal, whereas there is a lower mortality for basal ganglia or lobar haematomas. Hypertension is still the major cause of natural ICH, although the incidence has declined with the advent of therapeutic drugs. Both hypertension and amyloid angiopathy are also covered in Chapter 18. This section addresses medico-legal issues that may arise.

Hypertensive intracerebral haemorrhage This is the most common cause of ICH and accounts for about 50 per cent of all causes of ICH.

Macroscopic findings Hypertensive ICH occurs most commonly in the putamen and thalamus with a lesser proportion in the hemispheres, cerebellum and pons (Figure 10.14a). The common vessels to rupture in the hemispheres are small vessels such as a lenticulostriate vessel or pial perforating artery. Bleeding is usually a one-off event in contrast to the rebleeding that occurs with aneurysms, vascular malformations and amyloid angiopathy.

Histology Examination of the small vessels shows thickened fibrotic walls in long-standing hypertension with eosinophilic

(a)

homogenisation as well as medial accumulation of macrophages. (Figure 10.14b). Differentiation between collagen and fibrinoid necrosis may be made using stains for fibrin and collagen. Perivascular microinfarcts are commonly associated (Leestma 2016). Examination of the haematoma shows initially fresh red cells with adjacent neuronal hypoxic-ischaemic change developing after several hours, with variable acute inflammatory infiltrate over 24–48 hours. Later changes include macrophage accumulation with haemosiderin deposition. Intact red cells may persist for days to weeks (Perper and Wecht 1980).

Problem areas • Any attempt to time the age of a haemorrhage is difficult and for the most part may not be necessary in the context of a medico-legal case. • The issue of aetiology of, in particular, a ganglionic haematoma in association with dTAI has been discussed above. • The role of ‘stress’ is complex. One hypothesis is that acute rise in blood pressure may cause rupture of perforating cerebral vessels (Caplan 1988). This is reported in hypertensive as well as non-hypertensive individuals in the context of emotional or physical stress. This may be mediated in part by catecholamine release. Lammie et al. (2000) reported a case of ICH in an elderly man with known hypertension after severe emotional upset. Small-vessel fibrinoid necrosis of recent origin was identified adjacent to red cell extravasation and remote from the haematoma. It appears likely that altered cerebral vasomotor reactivity or autoregulation that occurs as part of the ageing process and chronic hypertension may predispose a vessel to rupture as well as to increased blood vessel fragility. A recent paper has also identified triggers to intracerebral haemorrhage of varying aetiology. There is some evidence to suggest that events such as excessive physical activity and other stressful events may lead to increase in such haemorrhages especially in combination (Sallinen et al. 2020). In any individual case detailed clinicopathological correlation is essential (Black and Graham 2002). However, these cases can be extremely problematic in the context of a criminal case.

ICH in association with cerebral amyloid angiopathy

(b)

Figure 10.14 (a) Hypertensive haemorrhage with intraventricular extension. (Courtesy of Dr Colin Smith). (b) Histology of microvascular pathology in hypertension showing arteriosclerotic change and hyaline vessel wall thickening.

Cerebral amyloid angiopathy (CAA) is characterised by extracellular deposition of b-amyloid (Ab) in the walls of blood vessels in the meninges and brain. ICH in association with amyloid accounts for around 12 per cent of all causes of ICH. The incidence of amyloid angiopathy increases with age. The blood vessels affected become rigid and friable which is why rebleeding is not infrequent. There is a complex association with both APOEε4 and -ε2 alleles (Greenberg et al. 1995; McCarron et al. 1999a, 1999b). APOEε4 is associated with asymptomatic CAA whereas APOEε2 is over-represented in patients who have CAA haemorrhage, often in association with anticoagulant therapy. The condition is often associated with Alzheimer-type pathology but may be seen in isolation. CAA is associated with

113

Contusional brain injury and intracerebral haemorrhage

114

hypertension in 30 per cent of cases, which may exacerbate the tendency to haemorrhage (Vintners 1987).

Macroscopic findings The location of this type of haemorrhage is commonly subcortical in a lobar pattern in the parietal and occipital regions which may be a guide, at brain examination, that this is not a primarily hypertensive haemorrhage or a traumatic ICH (Figure 10.15a). The haemorrhage may extend into the subarachnoid space and be seen as the dura is reflected. Haemorrhages may be multiple and in some cases petechial in nature.

Histology The blood vessels most frequently affected are the small-tomedium size arterioles in the cerebral cortex and leptomeninges. It is usually most severe in the occipital and parietal cortex as well as meningeal vessels. There is usually sparing of the hippocampus, basal ganglia and white matter. These are easily identifiable on haematoxylin and eosin examination, with the pink hyaline appearance typical and with amyloid involving the tunica media and adventitia. Congo red can be used to confirm this (Figure 10.15b) although immunohistochemical staining for Ab protein is the most sensitive.

Problem areas • The relationship of haemorrhage to ‘stress’ is either physical or emotional. This may predispose to ICH, possibly along a similar mechanism to that outlined above with regard to hypertensive haemorrhage. In the case of physical trauma, this may not necessarily be severe in the background of vascular fragility, although the circumstances of any potential incident need to be taken in to account. (Schmitt and Barz 1978; Dada and Rutherfoord 1993; Opeskin 1996; Pittella and da Silva Gusmao, 2016). • Association with warfarin treatment: CAA is an important contributor to ICH in patients on warfarin therapy. This has been demonstrated by the histological diagnosis as well as the elevated frequency of APOE 2 allele in patients with CAA haemorrhage (Rosand et al. 2000).

Other causes of intracerebral haematoma Vascular malformations There are various types of vascular malformations that may cause ICH and, to a varying degree, subarachnoid haemorrhage. Precise classification into a specific group may be difficult at times.

Arteriovenous malformations These generally present in patients aged 20–40 years, affecting 0.01–0.5 per cent of the population (Fleetwood and Steinberg 2002). These mainly arise on the surface of the cerebral hemispheres, particularly in the middle cerebral territories, and consist of abnormal arterial and venous structures usually in association with rather gliotic, hyalinised brain (Figure 10.16). They also occur in deep locations such as the thalamus or basal ganglia where there is a higher incidence of ICH (Fleetwood et al. 2003). They are an unpredictable cause of haemorrhage with a significant morbidity and mortality rate of between 1 and 4 per cent (Mackenzie 1996). They may be associated with ­a neurysms. (a)

Cavernous angiomas These occur most often in children. Identification may be difficult angiographically. However, on MRI, a cortical mass of abnormal vessels surrounded by cerebral parenchyma with haemosiderin is identifiable. Histologically, these consist of vessels with collagenous walls with no muscle or elastic tissue. Presentation as an ICH is less common than that of fits, headache or other neurological deficits.

Venous angiomas These consist of veins surrounded by normal brain tissue, which drains either into the leptomeninges or into the central venous system. They may present with seizures or haemorrhage.

Capillary telangiectasias (b)

Figure 10.15 (a) Old haematoma occurring in the background of cerebral amyloid angiopathy. (b) Congo red showing amyloid angiopathy.

These are normally incidental findings at post-mortem examination presenting as areas of haemorrhagic blush frequently in the pons (Challa et al. 1995) and bleeding rarely occurs. They comprise dilated small vessels randomly distributed throughout the brain parenchyma.

References

Conclusion ICHs have a variety of causes. Particularly problematic is the relationship between natural and traumatic disease, as well as drugs and alcohol.

References Adams, J.H., Doyle, D. and Graham, D.I. et al. 1985. The contusion index: a re-appraisal in man and experimental non-missile head injury. Neuropathol Appl Neurobiol, 11, 299–308. Adams, J.H., Doyle, D. and Graham, D.I. 1986. Deep intracerebral (basal ganglia) haematomas in fatal non-missile head injury in man. J Neurol Neurosurg Psychiatry, 49, 1039–1043. Anderson, R.M.D. and Opeskin, K. 1998. Timing of early changes in brain trauma. Am J Forensic Med Pathol, 19, 1–9. Baratham, G. and Dennyson, W.G. 1972. Delayed traumatic intracerebral haemorrhage. J Neurol Neurosurg Psychiatry, 35, 698–706. Black, M. and Graham, D.I. 2002. Sudden unexplained death in adults caused by intracranial pathology. J Clin Pathol, 55, 44–50. Boto, G.R., Lobato, R.D., Rivas, J.J., Gomez, P.A., de la Lama, A. and Lagares, A. 2001. Basal ganglia hematomas in severely head injured patients: clinicoradiological analysis of 37 cases. J Neurosurg, 94, 224–232. Caplan, L.R. 1988. Intracerebral haemorrhage revisited. Neurology, 38, 624–627. Challa, V.R., Moody, D.M. and Brown, W.R. 1995. Vascular malformations and the central nervous system. J Neuropathol Exp Neurol, 54, 609–621. Dada, M.A. and Rutherfoord, G.S. 1993. Medico-legal aspects of cerebral amyloid angiopathy. A case report. Am J Forensic Med Pathol, 14(4), 319–322 Dawson, S.L., Hirsch, C.S., Lucas, F.V. and Sebek, B.A. 1980. The contrecoup phenomenon. Reappraisal of a classic problem. Human Pathol, 11, 155–166. DiMaio, V.J. and DiMaio, D. 2001. Trauma to the skull and brain: craniocerebral injuries. In Forensic Pathology, 2nd edn. CRC Press. Engel, S., Schluesener, H. and Mittelbronn, M. et al. 2000. Dynamics of microglial activation after human traumatic brain injury are revealed by delayed expression of macrophage-related proteins MRP8 and MRP14. Acta Neuropathol, 100, 313–322. Fleetwood, I.G. and Steinberg, G.K.. 2002 Arterio-venous malformations. Lancet. 359; 863–873 Fleetwood, I.G., Marcellus, M.L. and Levy, R.P. et al. 2003. Deep arteiovenous malformations of the basal ganglia and thalamus: natural history. J Neurosurg, 98, 747–750 Greenberg, S.M., Rebeck, G.W. and Vonsattel, J.P.G. et al. 1995. Apolipoprotein E «4 and cerebral haemorrhage associated with ­amyloid angiopathy. Ann Neurol, 38, 254–259. Hangartner, J.R.W., Marley, N.J., Whitehead, A., Thomas, A.C. and Davies, M.J. 1985. The assessment of cardiac hypertrophy at postmortem examination. Histopathology, 9, 1295–1306. Holmin, S., Söderland, J., Biberfeld, P. and Mathiesen, T. 1998. Intracerebral inflammation after human brain contusion. Neurosurgery, 42, 291–298. Lammie, G.A., Lindley, R., Weir, S. and Wiggam, I.M. 2000. Stress-related primary intracerebral haemorrhage: postmortem examination clues to underlying mechanism. Stroke, 31, 1426–1428. Leestma, J.E. 2016. Forensic Neuropathology. New York, NY: Raven Press. McCarron, M.O., Nicholl, J.A.R. and Ironside, J.W. et al. 1999a. Cerebral amyloid angiopathy-related haemorrhage. Interaction of APOEe2 with putative risk factors. Stroke, 30, 1643–1646. McCarron, M.O., Hoffmann, K.L., DeLong, D.M., Gray, L., Saunders, A.M. and Alberts, M.J. 1999b. Intracerebral hemorrhage outcome: apolipoprotein E genotype, hematoma, and edema volumes. Am Acad Neurol, 53, 2176–2179.

Mackenzie, J.M. 1996. Intracerebral haemorrhage. J Clin Pathol, 49, 360–364. Makino, Y., Sannohoe, S., Kita, T. and Kuroda, N. 2001. Morphological changes of cerebral ventricular wall in head injuries observed with large histological specimens – Can it be the evidence of the impact to the brain? 7th Indo-Pacific Congress on Legal Medicine and Forensic Sciences, Melbourne. Mendelow, A.D. 1991. Spontaneous intracerebral haemorrhage. J Neurol Neurosurg Psychiatry, 54, 193–195. Morrison, A.L., King, T.M. and Korell, M.A. et al. 1998. Acceleration–deceleration injuries to the brain in blunt force trauma. Am J Forensic Med Pathol, 19, 109–112. Nanassis, K., Frowein, R.A. and Karimi, A. 1989. Delayed post-traumatic intracerebral bleeding. Neurosurg Rev, 12(suppl 1), 243–251. Oehmichen, M., Walter, T., Meissner, C. and Friedrich, H.J. 2003. Time course of cortical hemorrhages after closed traumatic brain injury: statistical analysis of post-traumatic histomorphological alterations. J Neurotrauma, 20(1), 87–103. Oehmichen, M., Jakob, S. and Mann, S. et al. 2009. Macropahge subsets in mechanical brain injury (MBI)—a contribution to timing of MBI based on immunohistochemical methods: a pilot study. Legal Med, 11(3), 118–124 Opeskin, K. 1996. Cerebral amyloid angiopathy: a review. Am J Forensic Med Pathol, 17, 248–254. Perper, J.A. and Wecht, C.H. (eds) 1980. Microscopic Diagnosis in Forensic Pathology. Thomas, p 12. Pitella, J.E. and da Silva Gusmao, S.N. 2016. Intracerebral haeomorrhage due to cerebral amyloid angiopathy after head injury: report of a case and review of the literature. Neuropathology, 36, 566–572 Ratnaike, T.E., Hastie, H., Gregson, B. and Mitchell, P. 2011. The geometry of brain contusion: relationship between site of contusion and direction of injury. Br J Neurosurgey, 25(3), 410–413 Rosand, J., Hylek, E.M., O’Donnell, H.C. and Greenberg, S.M. 2000. Warfarinassociated hemorrhage and cerebral amyloid angiopathy: a genetic and pathologic study. Am Acad Neurol, 55, 947–951. Ryan, G.A., McLean, A.J. and Vilenius, A.T.S. 1994. Brain injury patterns in fatally injured pedestrians. J Trauma, 36, 469–476. Sallinen, H., Putaala, J. and Strbian, D. 2020. Triggering factors in nontraumatic intra cerebral haemorrhage. J Stroke Cerebrovasc Dis, 29(8), 104921 Schmitt, H.P. and Barz, J.. 1978. Cerebral massive haemorrhage in congophilic angiopathy and its medicolegal significance Forensic Sci Int, 12(3), 187–201 Saukko, P. and Knight, B. 2016 Knight’s Forensic Pathology, 4th edn. CRC Press. Shreiber, D.I., Bain, A.C. and Ross, D.T. et al. 1999. Experimental investigation of cerebral contusion: histopathological and immunochemical evaluation of dynamic cortical deformation. J Neuropathol Exp Neurol, 58, 153–164. Siddique, M.S., Gregson, B.A. and Fernandes, H.M. et al. 2002. Comparative study of traumatic and spontaneous intracerebral hemorrhage. J Neurosurg, 96, 86–89. Vintners, H.V. 1987. Cerebral amyloid angiopathy. A critical review. Stroke, 18, 311–324. Vrankovic, D., Splavski, B. and Hecimovic, I. et al. 2000. Anatomical cerebellar protection of contrecoup hematoma development. Analysis of the mechanism of 30 posterior fossa coup hematomas. Neurosurg Rev, 23, 156–160. Yajima, Y., Hayakawa, S. and Mimasaka, M. et al. 2001. Intracerebral haematoma: traumatic or non-traumatic. J Clin Forensic Med, 8, 163–165. Yanagida, Y., Fjiwara, S. and Mizoi, Y. 1989. Differences in the intracranial pressure caused by a ‘blow’ and/or a ‘fall’ – an experimental study using physical models of the head and neck. Forensic Sci Int, 41, 135–145.

115

11

Traumatic axonal injury Daniel du Plessis

Diffuse traumatic axonal injury The fact that the axons might be damaged in a head injury was first demonstrated nearly 50 years ago by Strich (1956, 1961), who performed a neuropathological study of the brains of patients who had been vegetative after a head injury. She described widespread microscopic damage or ‘diffuse degeneration of the cerebral white matter’ and suggested that it resulted from shearing of axons at the time of injury. It was not until the 1980s, however, that the conditions necessary to cause such brain damage were known; working on primates, Gennarelli et al. (1981, 1982) were able to show that diffuse traumatic axonal injury (dTAI) was the result of angular or rotational acceleration of high magnitude, i.e. that it was primarily a non-impact phenomenon. This explained the typical situations in which dTAI occurred in human head injury – high-speed road traffic accidents and falls from a height – where excessive movement of the brain was likely to occur (Adams et al. 1991). Meticulous neuropathological studies by the Glasgow group of Adams and Graham (Adams et al. 1984, 1989a, 1989b; Graham et al. 1983, 1989, 1992, 1993, 1995) helped to expand the concept of this potentially very severe type of primary traumatic brain pathology. Diffuse axonal injury or DAI (as it was then termed) was defined primarily as a clinicopathological entity resulting from trauma, where the victim was unconscious from the time of injury – typically without structural changes on the scan to account for his or her loss of consciousness. The widespread nature of the damage caused prolonged coma, with survivors remaining either vegetative or with severe neurological disability.

Grading of diffuse traumatic axonal injury Initial studies attempting to formulate dTAI as a distinct clinicopathological entity concentrated on the severe end of the clinicopathological spectrum (essentially on what is now regarded as Grade 3 dTAI and patients rendered comatose). Such studies were also limited by reliance on non-immunohistochemical staining techniques which only demonstrated axonal injury after 8–12 hours of survival and with a further requirement of the demonstration of ‘large numbers of axonal bulbs throughout the white matter of the cerebral hemispheres and the brainstem’ (Adams et al. 1989a). Subsequently, it became clear that there were less severe forms of dTAI which led to the well-known and commonly applied grading scheme devised by Adams et al. (1989a). According to this scheme, Grade 1 dTAI was diagnosed relying on histological evidence of axonal injury in the white matter of the cerebral hemispheres, the corpus callosum, the brainstem and, less commonly, the cerebellum, further qualified by noting ‘large numbers of axonal bulbs’ throughout these white matter areas. The diagnosis of Grade 2 dTAI requires the additional demonstration of a focal tissue-tear lesion in the corpus callosum (see below), whereas the most severe form of dTAI (Grade 3) requires the presence of a focal tissue-tear lesion in one or more dorsolateral quadrants of the rostral brainstem.

Such tissue-tear lesions relied upon for grading purposes were commonly only identified at microscopic level (Adams et al. 1989a). It should be born in mind that the original definition of Grade 3 dTAI requires the demonstration of tissue tear lesion in both corpus callosum and upper brainstem, whereas it is often now diagnosed in more relaxed manner relying on brainstem lesions alone. Once the conditions necessary for dTAI were established, it was possible to develop cheap and reproducible non-primate models, to enable axonal damage to be studied at a cellular level. A number of such models have been used and have helped greatly in the understanding of what happens to axons after injury (Gennarelli 1996). Such models have shown that the severity of the axonal injury depends principally on the magnitude and duration of the acceleration and on the direction of head motion. With sagittal acceleration Grade 1, but not Grade 2 and Grade 3, dTAI can be produced. Acceleration of the same magnitude in the coronal plane produces more severe dTAI, usually Grade 3 whilst rotation in a horizontal plane produces dTAI of intermediate grade similar to that seen in man. This understanding has in turn led to the development of immunocytochemical markers of axonal damage, of which the most used in diagnostic neuropathology is an antiserum to beta-amyloid precursor protein (β-APP) (Sherriff et al. 1994). Systematic neuropathological studies employing this antibody have shown that a degree of damage to axons probably occurs in the vast majority of head injuries (Gentleman et al. 1995) and that the subsequent clinical picture is dictated by the amount and distribution of that damage.

Clinical and forensic correlates It is now understood that traumatic axonal injury (TAI) comprises a spectrum (Geddes et al. 2000). At the mild end of the spectrum, there is good experimental evidence to suggest that some axonal damage may be reversible (Tomei et al. 1990; Maxwell et al. 1994), and indeed that mild reversible axonal injury may be the neuropathological substrate of concussion (Gennarelli et al. 1998). At the other end of the spectrum lies the more severe grades of dTAI, where strains on axons are so severe that they lead to irreversible damage in a widespread (‘diffuse’) distribution, in the brainstem as well as in the cerebral hemispheres, leading to coma and potentially a persistent vegetative state. In between these two extremes of TAI lies a range of damage, for which we cannot predict the precise clinical correlates. However, rare opportunities to study systematically the brains of individuals who suffered a recent documented mild head injury, but died of an unrelated cause, suggest that scattered traumatic axonal damage, usually hemispherical, may be a common finding after mild head injury in patients who may have had few neurological or psychological symptoms (Blumbergs et al. 1994, 1995; unpublished personal observations). Biomechanical work supports the experience of neuropathologists by showing that proportionally greater force is required to produce the axonal

Tissue-tear lesions damage in the brainstem which is the cardinal feature of dTAI (Smith et al. 2000). The recognition of less severe grades of dTAI also included the recognition that loss of consciousness is not inevitable in cases of dTAI despite large numbers of axonal bulbs in the white matter of the brainstem and forebrain relying on nonimmunohistochemical staining techniques. Relying on such specific diagnostic features, a lucid interval (defined as being able to converse clearly and rationally immediately after injury) was restricted to cases showing Grade 1 dTAI where it may be observed in as many as 20 per cent of cases (Adams et al. 1989a). Some cases of Grade 2 dTAI may experience a partial/ semi-lucid interval before death (up to 20 per cent). Amongst these, macroscopically obvious focal lesions in the corpus callosum are very rare with the majority (94 per cent) revealing such lesions only at microscopic level. There appears to be no documented case of Grade 3 dTAI diagnosed by the recognition of macroscopically evident focal lesions in the corpus callosum and brainstem who experienced a semi-lucid or lucid interval, i.e. individuals suffering the most severe form of dTAI are rendered comatose immediately with no intervening lucid or semilucid interval, an attribute of potential forensic significance (see Pitfalls ­section). It should be noted that rare cases of Grade 3 dTAI with no macroscopic focal lesions may experience a semilucid interval, but none a lucid interval. The investigation of TAI utilising β-APP immunostaining has had the benefit of allowing the recognition of TAI much earlier than 12–18 hours and in much more subtle form (both in respect of staining quality and density). Given the reliance on focal tissue-tear lesions in the corpus callosum and dorsolateral rostral brainstem to diagnose Grades 2 and 3 dTAI, the detection of only minimal though diffuse accompanying axonal injury (or even no axonal injury in short survival situations) usually does not undermine the diagnosis of these very severe forms of primary traumatic brain injury and its potential forensic usefulness on commenting on the likely level of associated force and likely associated adverse clinical effect. It has though made Grade 1 dTAI a somewhat more nebulous clinico-pathological entity of much lesser reliable potential forensic informative value. Grade 1 dTAI is now much more freely diagnosed in the absence of any guidelines on a minimum threshold of axonal bulbs and threads in the forebrain and brainstem. The establishment of readily applied minimum thresholds is substantially compromised though by the dynamic evolving nature of secondary axotomy over several hours, which allows for dTAI to be present even in the absence of any detectable TAI. The possibility that other conditions may account for β-APP imunopositive axonal injury in a low density further confounds assessment (see pitfalls below). The far more relaxed diagnosis of dTAI introduced by the routine use of β-APP immunostaining has undermined the forensic reliability of Grade 1 dTAI as an inevitably severe form of primary traumatic brain injury, both in terms of the level of force required and associated clinical ill-effect. Grade 1 dTAI has now become permissive of a very wide spectrum of associated symptomology which may vary from a maintained lucid state through various severities of concussion to profound coma. Extreme caution should therefore be exercised when relying on Grade 1 dTAI alone to inform on the level of causative force and associated symptomology. Cases of Grade 1 dTAI corresponding to those more historically insisted upon in the age of non-immunohistochemical investigative techniques, namely with extensive axonal bulbs throughout the forebrain and brainstem white matter after several hours of survival may carry more predictive weight in terms of severity of force and

associated ill-effect, especially so if accompanied by tissue-tear lesions at sites other than the corpus callosum or rostral brainstem. In cases involving the demonstration of only subtle, early axonal injury in both forebrain and brainstem white matter in situations of short survival, it may be more prudent to qualify such axonal injury as potentially diffuse rather than certainly so in the absence of further substantiating, predictive pathology in the form of characteristic tissue-tear type bleeds. Experimental work shows that extreme forces are necessary for axons to rupture immediately and that in virtually all head injuries, axotomy is likely to be a secondary, delayed event after injury as a result of damage to the axonal cytoskeleton. In those dying instantaneously to near-instantaneously after a severe head injury, whose brains show widespread white matter petechial and larger haemorrhages (referred to as diffuse traumatic vascular injury), it is generally assumed that widespread primary axotomy has occurred and that this is incompatible with life (Figure 11.1). This is an assumption made on the basis of widespread microvascular shearing (vessels having a higher tolerance for shearing than axons, thus implying widespread accompanying axonal injury as well). Apart from such circumstances dTAI is rarely a cause of death on its own. It is, however, a cause of prolonged coma and an important cause of vegetative state (Graham et al. 1983; Kinney and Samuels 1994; Adams et al. 1999) or severe post-traumatic neurological disability. TAI localised to the brainstem may be a cause of respiratory failure – indeed, this is probably an important mechanism of death in infant head injury (Geddes et al. 2001) and in those who have high cervical injuries.

Tissue-tear lesions Biomechanical conditions conducive to dTAI may (if of sufficient magnitude) also cause vascular shearing of the microvasculature of the white matter in both the forebrain and brainstem. This gives rise to haemorrhagic focal lesions referred to as tissue-tear lesions, which are of particular assistance as potential surrogate markers of dTAI. Such bleeds are relied upon by radiologists to recognise dTAI in life and can assist post-mortem in drawing attention to the likelihood of dTAI, even in situations where survival was too short to permit the detection of axonal injury. These lesions may be apparent macroscopically or microscopically. Tissue-tear lesions in the corpus callosum and dorsolateral rostral brainstem assist in the grading of dTAI. • Such lesions in the corpus callosum are typically haemorrhagic, asymmetric and inferior, but may extend to the midline and involve the interventricular septum and the pillars of the fornix (Figure 11.2a). • Lesions in the dorsolateral quadrants of the rostral brainstem are not always haemorrhagic with foci of neuropil vacuolation and rarefaction also qualifying as lesions in this specific location. Grades 2 and 3 dTAI characterised by macroscopic tissue-tear lesions are further regarded as more severe forms of Grades 2 and 3 dTAI. (Adams et al. 1986, 1989a; Graham et al. 2002; Geddes and Graham 2004) (Figure 11.3). • In the parasagittal regions of the frontal and parietal lobes, severe acceleration produces typical linear or flame-shaped uni- or bilateral traumatic haemorrhages in the subcortical white matter known as ‘gliding contusions’ (Figures 11.1a and 11.2b). The term is misleading because they, in contrast to true contusions, do not involve the cortex and are caused by movement, rather than impact.

117

Traumatic axonal injury

118

(a)

(a)

(b)

Figure 11.2 (a) and (b) Macroscopic markers of diffuse axonal injury (DAI) at brain cut: gliding contusions in the ­p arasagittal white matter of both slices; There is a lesion in the corpus ­c allosum at both levels and also a typical small deep grey matter haemorrhage, not exerting mass effect. Such a patient would be likely to have brainstem lesions as well. (Reproduced from Geddes and Graham 2004, by kind ­p ermission of Butterworth Heinemann.)

(b)

Figure 11.1 Diffuse vascular injury in the brain of a road traffic accident victim who died of a head injury almost immediately after the accident. (a) A gliding contusion is present in the frontal lobe of the hemispheric slice, adjacent to the midline, and there were numerous petechial haemorrhages scattered through the central white matter. (b) The lesions in the brainstem are in a distribution typical of diffuse axonal injury (DAI), i.e. in the dorsal part of the brainstem. Damage as widespread as this is assumed to be incompatible with life. (Reproduced from Geddes and Graham 2004, by kind permission of Butterworth-Heinemann.) • It is surprising how often pathology texts omit reference to cerebral cortex/white matter junction bleeds commonly relied upon by radiologists to diagnose dTAI. Such bleeds should not be associated with immediately overlying intracortical bleeds as this may just represent white matter extension of contusions (Figure 11.4).

• Primary traumatic ganglionic bleeds are another example (see Chapter 8) of a tissue-tear bleed associated with dTAI (Figure 11.2b). • Midline and paramidline periventricular/subependymal bleeds may also be associated with dTAI in the author’s experience. This appears related to shearing injury of subependymal small veins which may act as a source of intraventricular haemorrhage, another type of bleed which should alert the examiner to the possibility of dTAI where no obvious alternative source such as an intracerebral haematoma or retrograde flow of basal subarachnoid haemorrhage could have accounted for intraventricular haemorrhage. Shearing-related injury of periventricular veins is sometimes recognised by evidence of thrombosis of affected subependymal veins. This rare feature (as an isolated finding with no surrounding obvious parenchymal injury otherwise) may lead to thalamic venous infarcts which may be bilateral. Tissue-tear bleeds in structures such as the rostral brainstem and corpus callosum may, on occasion, be difficult, if

How to recognise traumatic axonal injury or into the cerebellar peduncles, making distinction from primary traumatic tissue-tear bleeds often impossible. Recourse to review of the neuro-imaging carried out in-life (if available) and to β-APP staining may be of assistance in this regard. The radiological identification of discrete, isolated bleeds in corpus callosum or dorsolateral rostral brainstem before the intervention of substantial brain swelling with mass effect may add substantial evidence weight to interpretation of an anatomically corresponding focus of haemorrhage as a true tissue-tear bleed at post-mortem. Secondary herniation-related bleeds may often represent a terminal complication of severely raised intracranial pressure and may prove unassociated with any intimately associated surrounding enhanced axonal injury as a consequence.

Sampling protocols

Figure 11.3 Diffuse axonal injury (DAI): dorsolateral quadrant haemorrhages in the brainstem. not impossible to distinguish from secondary bleeds related to mass effect related to distortion and compression of these parts of the brain. Whilst the classic secondary herniationrelated ‘Duret’ haemorrhages are typically situated in the midline of the midbrain and rostral pons, such bleeds may not uncommonly extend into tectum and tegmentum and/

At least some TAI is likely to have been sustained in most clinically significant traumatic head injuries regardless of the ability to confirm it histologically. The cause of death is further often dependent on other primary and secondary traumatic pathology, especially so in the acute post-traumatic phase. Whether or not TAI is present may, therefore, be practically irrelevant. However, in medico-legal work, it is seldom just a question of documenting pathology, and histology may be essential in order to provide answers to questions posed by others, which may arise some considerable time later in the course of legal proceedings. With respect to TAI, this most often involves questions around survival intervals and the mode of injury (see pitfalls below). As TAI may be very widespread, any attempt to diagnose axonal damage must be based on extensive systematic sampling of the brain. As with most other forms of generalised cerebral pathology, areas of the brain differ in their susceptibility and one or two samples can give little indication of what is going on in the brain as a whole (Graham et al. 2004). In trauma, it is the midline and paramidline structures that are particularly vulnerable to TAI, which means that parasagittal cerebral white matter (such as posterior and anterior frontal parasagittal white matter), corpus callosum (anterior, body and splenium), posterior limb of the internal capsule and upper brainstem (midbrain including the decussation of the superior cerebellar peducles) and upper pons (including the middle cerebellar peduncles) must all be sampled as a minimum with representation of both sides in the cerebrum. It should be appreciated that TAI is accentuated in the more posterior corpus callosum. It could be stated that the absence of TAI in the splenium effectively rules out dTAI subject to sufficient survival (Leclercq et al. 2002). Sampling protocols also have to comprehensively assess pathology associated with secondary hypoxic-ischaemic injury, including herniation-related injury given the confounding effects both may have on recognising TAI and tissue-tear lesions (see below). A sampling protocol covering all types of diffuse brain injury in a case of head injury is provided in Chapter 5.

How to recognise traumatic axonal injury Figure 11.4 Images such as these, obtained by magnetic resonance imaging, confirm a clinical diagnosis of severe diffuse traumatic axonal injury. There are lesions in the corpus callosum, deep grey matter/posterior limb of the internal capsule, cortical grey-white mater junction and in the rostral midbrain. Compare the brainstem image with Figure 11.3. (Reproduced from Graham et al. 2002, by kind permission of Arnold.)

The use of immunohistochemistry has greatly facilitated the recognition of TAI but also introduced some difficulties. First, antibodies to proteins such as β-APP may detect damaged axons very rapidly after injury regardless of cause, well before other potentially further qualifying pathology can be detected by light microscopy. The second is that β-APP immunoreactivity can be seen in axonal damage of any aetiology, e.g. severe hypoglycaemic damage and carbon monoxide poisoning, as well as multiple sclerosis plaques, HIV encephalitis, cerebral

119

120

Traumatic axonal injury malaria and Binswanger-type white matter lesions (Graham et al. 2004). In the context of trauma, however, the principal differential diagnosis is hypoxic-ischaemic axonal damage resulting from raised intracranial pressure. Hypoxia alone without impaired perfusion does not cause detectable axonal injury (Dolinak et al. 2000). Once intracranial pressure is raised, penetrating blood vessels are liable to become compressed, leading to ischaemia and infarction in the areas that they supply. Foci of vascular compromise caused by brain swelling are common in head injury and axons around such areas of incipient infarction will be damaged and so are positive with anti-β-APP. Most brains in fatal head injury are swollen by the time of death, so both traumatic and ischaemic axonal damages are often present in the same brain. The challenge is to work out which is which. It does not help that the same anatomical areas are vulnerable (Figure 11.5) to TAI and mass effect–related vascular compromise. As a general rule, geographical expression of β-APP Figure 11.6 A low-power view of a section of pons, from the brain of a patient who died of raised intracranial pressure after a severe head injury. The stain is an immunohistochemical one for β-APP and, although faint, clearly demarcates a large area of the pontine basis. Such staining is not the result of traumatic axonal damage, but of early ischaemic damage secondary to brain swelling and shift. Similar patterns of immunoreactivity, often involving the midline and periphery of the brainstem, can be seen in patients who have not had a head injury.

Figure 11.5 Schematic representation of the effects of swelling and shift on axons. A Sylvian fissure haematoma (in red) has resulted in pronounced lateral and transtentorial shift of the affected hemisphere. The shaded grey areas indicate sites at which it is common to see incipient ischaemic damage to white matter on β-APP immunohistochemistry. Note that the sites are the same as those in which traumatic axonal damage may be seen. (Reproduced from Geddes et al. 2000, by kind permission of Blackwell Publishing Ltd.)

tends to be vascular; this is particularly true in the base of the pons, where linear β-APP immunoreactivity, particularly involving the midline and periphery, tends to outline areas of early ischaemia (Figure 11.6). Not infrequently neurons inside such areas show strong cytoplasmic eosinophilia or β-APP expression, both of which reinforce the impression of vascular damage – although, once again, no systematic studies of somatic β-APP positivity have been performed and one can only assume that it indicates acute neuronal stress of some sort, probably hypoxic-ischaemic rather than traumatic. With several days survival after injury, it might be possible to appreciate axonal damage of differing ‘ages’, i.e. clearly evolved TAI characterised by large, intensely staining bulbs and threads and histologically more recent damage in a typical vascular pattern, which is the result of terminal brain swelling and resultant foci of ischaemia. That said, the distinction is often difficult to make, despite suggestions that vascular and TAI differ in their staining characteristics on immunohistochemistry (Lambri et al. 2001). Axonal injury associated with raised intracranial pressure and secondary hypoxic-ischaemic brain injury is more readily distinguishable when manifest as geographic zones of axonal injury. These are characterised by irregular (sometimes described as zig-zag patterned) zones of axonal injury varying from enhanced linear granular axonal staining to more overtly evolved bulbs, threads and varicosities, delineating areas of ischaemia. In hyperacute form, this may be seen as patches of enhanced linear granular axonal staining alternating with non or lesser staining zones (the latter often containing a central microvessel), imparting striking mottled appearances to affected white matter (Figure 11.7). Vascular axonal injury has also been described as a ‘granular pattern with a dirty background’. The confident recognition of TAI can be very problematic when admixed with vascular axonal injury, especially so in

The evolution of traumatic axonal injury author’s experience that features consistent with hyperacute vascular axonal injury (characterised by geographic patches of enhanced linear granular axonal staining lacking discrete bulbs and threads) may manifest within less than half an hour of survival and it further appears that such changes may even emerge or develop during periods of sustained optimal resuscitation despite the latter proving unsuccessful. Finally, it is necessary to heed the reader not to misinterpret granular eosinophilic bulb-like profiles occasionally found around the microvasculature in the cerebral white matter as evidence of axonal injury. These structures are characterised by the absence of β-APP or neurofillament immunopositivity and are argyrophilic. These appear to reside within perivascular spaces rather than along the course of axons (Geddes et al. 2000).

The evolution of traumatic axonal injury Figure 11.7 Hyperacute vascular axonal injury characterised by geographic patches of enhanced linear granular β-APP axonal immunostaining. situations involving significantly raised intracranial pressure with mass effect. TAI has been described as constituting discrete, dispersed well-formed and intensely staining axonal bulbs, threads and varicosities with a ‘clean background’. Fascicle selective axonal injury where axonal injury appears confined to multiple fibre tracts showing a specific orientation has also been attributed to TAI. A study comparing β-APP immunohistochemistry in all grades of dTAI with a variety of non-traumatic conditions including hypoxic-ischaemic brain injury identified three patterns of axonal staining. A pattern described as multifocal (diffuse) was characterised by axonal swellings and bulbs scattered throughout white matter, these showing a more dispersed distribution in lesser grades of dTAI, but also in cases of carbon monoxide poisoning and hypoglycaemia. The multifocal (diffuse) pattern was associated with a higher density of such affected axons (often showing clustering in certain parts of the brain) in the more severe grades of dTAI. A second pattern was characterised by irregular, often ‘Z’-shaped profiles thought to identify the boundary of zones of ischaemia. This pattern was characteristic of cases with evidence of raised intracranial pressure in both traumatic and non-traumatic situations but never as an isolated feature. It was always found as part of the third pattern which was described as a mixed diffuse and vascular pattern. This pattern was identified in Grade 2 and Grade 3 dTAI but also in cases of post-cardiac arrest hypoxicischaemic brain injury, status epilepticus and hypoglycaemia. A common factor in cases showing the vascular pattern was that of evidence of raised intracranial pressure (Graham et al. 2004). It would be fair to stay that reliable distinction is often a frustrating exercise confounded by uncertainty and the lack of consistently reliably discriminators. Recourse to a combination of morphology and topography is best practise. Ultimately, the dictum that should apply to all forensic interpretation is imperative – when in doubt, say so. Fortunately interpretation of axonal damage is often irrelevant to the outcome. There is little or no information available about the temporal evolution of vascular axonal pathology (VAI) caused by ischaemia. No formal studies have been published. Anecdotal accounts suggest that in some cases VAI may be detected before TAI becomes manifest, i.e. in less than 35 minutes. It is the

Neuropathological and animal studies have shown clearly that axonal damage starts at the time of injury and continues to progress in the hours, days, weeks and months that follow (Povlishock 1993; Geddes et al. 1997, 2000; Gennarelli 1997; Maxwell et al. 1997; Gennarelli et al. 1998; Graham et al. 2002). They have also established that few, if any, axons are transected at the time of head injury. The primary event is damage to the axonal membrane and the underlying cytoskeleton. This varies according to the type of strains to which the axon is subjected but in most cases results in rapid interruption of axonal transport. Immunohistochemistry for axonal damage capitalises on this, by using antibodies to proteins that are normally transported down the axon in very small quantities, which then accumulate at any point at which transport is impeded. β-APP is one such molecule, which has found routine application in forensic diagnostic practice. Its use has been systematically studied by a number of different workers and it is usually detectable within 1–2 hours after injury (Geddes et al. 1997; Maxwell et al. 1997; Reichard et al. 2005; Wilkinson et al. 1999) (Figure 11.8). The temporal evolution of axonal damage in terms of morphology and cellular reaction has been well described in adults (Geddes et al. 1997; Graham et al. 2002), although it must be

Figure 11.8 Within 1–2 hours of survival after injury, damaged axons can be detected with β-APP immunostaining. The axonal swellings are not fully formed and would be very difficult to find on staining with haematoxylin and eosin (H&E). (Reproduced from Geddes and Graham 2004, by kind permission of Butterworth Heinemann.)

121

122

Traumatic axonal injury emphasised that the time course given in such publications is only an outline and takes no account of factors such as age or axonal size and for that reason should not be used to give more than an approximate indication of the timing of the pathology. The situation is further confounded by the fact that there is some indication that axons of different sizes may react at different rates and that secondary axotomy may continue to occur well beyond the first 24 hours after injury (Maxwell et al. 2003; Stone et al. 2004). For this reason alone, the suggestion that measuring an axonal bulb can give an indication of the timing of the injury (Wilkinson et al. 1999) is unlikely to prove reliable (Leclercq et al. 2002). It is nevertheless the case that hyperacute, very early axonal injury has a different quality of β-APP immunostaining compared to more evolved axonal injury (see above). It is necessary to point out a commonly underappreciated difference in the evolution of primary versus secondary axotomy, the latter pertaining to diffuse traumatic injury but the former to primary traumatic lesions involving immediate to near immediate axonal disruption such as tears or lacerations (as may for example occur within the corpus callosum due to impaction against the free edge of the falx). In such a situation, axonal bulbs may evolve very quickly, becoming detectable in less than an hour. This applies to H&E and silver staining as well as β-APP immunostaining (Anderson and Opeskin 1998). There further appears to be rare anecdotal accounts of clearly demonstrable axonal injury (presumably involving swollen threads and bulbs) at the transected ends of axons in lesions associated with instantaneous to near instantaneous death (for example transected severed spinal cord in cases of decapitations). In such situations, a short interval of some ongoing axonal transport following death may permit accumulation of β-APP at the severed end of affected axons. The demonstration of axonal bulbs and threads in such lesions should therefore not be relied upon to suggest a more extended period of survival. • Hortobágyi et al. (2007), utilising β-APP immunostaining, have demonstrated TAI in adult brains at a minimum of 35 minutes of survival. These were identified as small globules and granules with occasional thin, short filaments. A single study of fatal paediatric head injury caused by road traffic accidents has also documented finding damaged (APP-positive) axons at short survival times in the internal capsule, thalamus and brainstem in an infant of 3 months and in the internal capsule of a 2-year-old boy. Times of death were well documented for both cases, survival being only 35 minutes for the infant and 45 minutes for the older child (Gorrie et al. 2002). It is the author’s experience that the brains of young infants may show no evidence of axonal injury (either traumatic or vascular) in situations undoubtedly involving traumatic brain injury, further complicated by severe secondary hypoxic-ischaemic injury despite more than sufficient survival (in excess of several hours) and despite positive staining of appropriate control tissue and of spinal cord and nerve root tissue in the same individual. Such cases almost invariably involve a non-perfused brain with the very likely early intervention of brainstem death. It is postulated that this phenomenon not only reflects the immaturity of axons in such cases, but also a near immediate ‘stunning’ of axonal transport mechanisms preventing the detection of evolving secondary axotomy. A similar surprising absence of any axonal injury has also been encountered in extremely rare adult head injury cases.

Figure 11.9 Immunoreactivity for β-APP tends to start fading from the axonal swellings or ‘bulbs’ from about 5 days after injury, although strong expression continues to be seen in axons and varicosities. This high-power field from the internal capsule of a man who had a 10-day survival shows a large bulb on the right, which is completely negative for APP (arrowhead), whereas adjacent damaged axons still stain.

•

•

•

•

•

Whatever the age group, care should be taken when interpreting findings in cases of very short survival. By about 12–18 hours after injury, the damaged axons demonstrated by anti-β-APP will have swollen greatly, as a result of accumulation of substances that have continued to be transported down from the cell body. At this stage, the axonal swellings or bulbs are sufficiently large to detect by routine methods such as H&E or silver preparation. By 18–24 hours, many of these swollen axons will have become severed from their distal portion, i.e. have undergone secondary axotomy – and the damage is irreversible. By 7 days survival after injury, β-APP staining of bulbs becomes much less uniform and begins to fade, until eventually they are no longer immunoreactive, although some axons adjacent to the bulbs remain strongly positive for β-APP (Figure 11.9) (Geddes et al. 1997). In a situation in which there are many bulbs not staining with anti-β-APP while nearby axons are still positive, it is probably safe to say that injury occurred at least a week before death. Effete axonal bulbs are further characterised by irregular zones of loss of staining and irregular, sometimes crenulated outlines. β-APP immunopositive TAI very rarely persists beyond 6 months, then mainly seen in the form of scattered granular extracellular debris (Geddes et al. 1997). Similar granular debris may be seen at an earlier stage after 1–2 months within macrophages or as small clumps. In the authors’ experience, rare scattered axonal bulbs characterised by irregular profiles and irregular APPimmunostaining may though be identified some years after a major head injury leaving subjects in a vegetative state.

Whilst β-APP currently enjoys the widest application as a routine diagnostic tool to investigate TAI, antibodies directed against other protein (some serving as biomarkers for traumatic brain injury) has more recently gained attention given the potential benefit to detect axonal injury not highlighted by β-APP (which appears to identify only a subset of potential injured axons) and/or to demonstrate TAI even earlier than β-APP.

The evolution of traumatic axonal injury Immunostaining for SNTF, a proteolytic fragment of alpha-II spectrin has been shown to demonstrate a subpopulation of degenerating axons undetected by β-APP. Similarly, RMO-14 and NF-68 immunopositive axonal pathology may exist independent of SNTF and β-APP, which demonstrates that multiple pathological axonal phenotypes exist post-traumatic brain injury. Labelling of injured axons with some such markers may have promise in identifying severe forms of axonal damage not highlighted by β-APP (such as in conditions associated with catastrophic failure of axonal transport, preventing the transport of β-APP). Complete failure of transport accompanied by calcium-mediated proteolysis may induce rapid disintegration undetected by β-APP – a situation potentially explaining those cases showing a surprising absence of β-APP immunostaining (see above) (Johnson et al. 2016). Clusterin (apoplipoprotein J), a complement inhibitor, may show axonal expression in taumatic brain injury in less than 30 minutes following injury with an increased intensity and frequency of accumulation with increase in survival up to 24 hours, after which it wanes in intensity over several weeks after injury (Troakes et al. 2017). The ‘persistence’ of axonal bulbs and threads some years after traumatic brain injury is hardly surprising given evidence that inflammation and white matter degeneration may persist for years after a single traumatic brain injury, as such rather representing an ongoing ‘wear and tear’ process (Johnson et al. 2013). Microglial activation is the other feature typical of evolving axonal damage. These cells, which can be well demonstrated by antibodies to CD68 (the PG-M1 epitope demonstrates microglia better than KP1), are an important feature of the brain’s reaction to injury. For this reason, routine staining for CD68 is helpful in cases where there is survival of more than a few days. Iba-1 immunostaining can also be used to demonstrate microglial activation in response to axonal injury but has the drawback of demonstrating of both resting and activated populations of microglia.

(a)

• The initial increase in microglial numbers after injury is subtle and not constant: non-specific upregulation (hyperplasia) in the white matter after head injury can be detected from about 48 hours (although to be certain of this a number of blocks is needed as well as familiarity with the density of microglia normally present in the nontraumatised brain). • By around 5 days, aggregation of microglia may occur around axonal bulbs (Geddes et al. 1997). • However, at about 10 days, microglia start collecting round damaged axons in clusters or nodules, which are well shown by CD68 immunohistochemistry. • By 14 days, microglial clusters (also referred to as microglial stars) may be present in a high density. • Microglial clusters remain long after the axonal swellings have disappeared and can be seen scattered through the white matter (Figure 11.10) several months after the original injury as tombstones, marking the site of axonal damage. Note, however, if microglial nodules are found scattered through the white matter in someone who has died of a very recent head injury (or, indeed, of any other cause), it cannot be assumed that the nodules are the result of previous traumatic damage. If there was a definite history of trauma, they may indeed be that, but microglial aggregates occur in a variety of conditions, e.g. severe global hypoxic-ischaemic injury (Figure 11.11) and viral encephalitis. Isolated microglial clusters, encountered on their own, are commonly seen in routine neuropathological practice and should be ignored. However, in the absence of a history of an alternative insult, microglial nodules scattered throughout the white matter would be compatible with a previous head injury. As the central white matter degenerates, secondary long tract degeneration starts. This takes place in the months following injury and is seen in descending, particularly corticospinal, tracts in the brainstem. The traditional neuropathological method of showing long tract degeneration in such cases is to

(b)

Figure 11.10 Microglial clusters in a patient with diffuse axonal injury (DAI) who survived 5 months. (a) Aggregates of microglia in the medulla can be detected on H&E staining, but are more easily demonstrated by (b) immunohistochemistry (anti-CD68/PG-M1, in the hemispheric white matter of the same patient).

123

124

Traumatic axonal injury thalamus are most at risk of damage (neuronal loss with gliosis) with profound hypoxic-ischaemic brain injury, whereas in dTAI, there is particular involvement of the lateral and ventral nuclei (Adams et al. 1999, 2000; Graham et al. 2005).

Pitfalls in the interpretation of axonal damage

Figure 11.11 Microglial clusters scattered through the brain are not specific for trauma. The image comes from the hemispheric white matter of a patient with no previous history of head injury but who was severely disabled after an episode of global hypoxia some months earlier. demonstrate the tell-tale breakdown products of myelin in long tracts by the Marchi technique, although immunohistochemistry for CD68 (PG-M1) is quicker and serves just as well, because it will show the large numbers of foamy macrophages involved in the degeneration of the axons, anatomically confined to the descending brainstem tracts (Figure 11.12). Histological examination of the thalami and hippocampi in longer-term survivors can possibly assist in determining whether a persistent vegetative state following a head injury was contributed to by dTAI, hypoxic-ischaemic brain injury or a combination of these. The CA1 and CA4 sections of the hippocampus and the anterior and dorso-medial nuclei of the

(a)

• TAI versus other causes of axonal injury The confident and reliable recognition of TAI becomes mandatory if the presence of TAI (whether only focal, multifocal or truly diffuse) is to be relied upon to inform forensic issues around survival times. This exercise requires the exclusion of vascular axonal injury. If such discrimination can be reliably made, recourse to the microscopic evolution of TAI as detailed above can be carried out. Conditions other than VAI may also confound interpretation. Many of those subject to head injuries may have a long-term history of alcohol and drug abuse with some evidence of potentially related brain atrophy. Those suffering head injuries or other forms of unnatural death may also have been subject to previous head injuries. The elderly may further be subject to neurodegenerative pathology. In this context, the identification of sparse, isolated axonal threads, grains and bulbs could potentially just reflect low grade ‘wear and tear’ axonal injury. Non-swollen brains lacking evidence of ischaemic injury, but showing evidence of microglial activation including clustering certainly predating a known or alleged head injury, could also be subject to such a process. A very prudent approach is warranted in such situations. • Mode of injury, level of force and associated clinical effect The confident recognition of Grades 2 and 3 dTAI is more reliably informative on these issues. It is recognised that both of these constitute severe forms of primary traumatic brain injury requiring a substantial degree

(b)

Figure 11.12 Long tract degeneration in a patient with diffuse axonal injury (DAI) who survived 5 months after his head injury. (a) The breakdown of the corticospinal tracts and, to a lesser extent, the medial lemniscus, is well demonstrated by the Marchi impregnation technique (note that the darker staining round the edges and in the superior cerebellar peduncle, is an artefact often seen in such preparations.). (b) Immunohistochemistry for CD68 (PG-M1) has been performed on a section from a block taken slightly more rostrally in the brainstem. Once again, the corticospinal tracts are outlined by the immunoreactivity. (Reproduced from Geddes and Graham 2004, by kind permission of Butterworth Heinemann.)

Acknowledgements of force, an interpretation supported by its reliable and more common recognition in situations such as falls from a substantial height and within the context of road traffic accidents. It is further recognised that those diagnosed on the basis of macroscopically evident tissue-tear bleeds in the corpus callosum and/or brainstem must be regarded as representing the worst end of the spectrum within both grades. Both these grades of dTAI likely require very substantial angular acceleration/deceleration of the head within the horizontal or coronal plane or both. Definite tissue-tear bleeds affecting the corpus callosum and rostral brainstem have, in the author’s experience, not been found in heavy falls from a standing height precipitated by a punch (and not followed up by subsequent kicking or stamping whilst on the ground). Such bleeds have though been found in occasional cases following heavy falls down several stairs. There appears to be no verified case of a fully lucid interval following injury in those suffering Grades 2 and 3 dTAI. This attribute, especially so the attribute of severe Grade 3 dTAI to render the subject immediately profoundly unconscious, may have forensic significance. In someone subject to serial head injuries separated by reliably witnessed lucid behaviour, the violent act rendering the subject immediately and profoundly unconscious and subsequently persistently comatose has to be regarded as the one culpable for causing Grade 3 dTAI. This may exonerate an earlier assault in situations involving proximate incidents with different perpetrators. Grade 1 dTAI is more problematic as it now appears to accommodate a very wide spectrum of associated causative physical force and clinical ill-effect as discussed previously. The more ‘historic’ Grade 1 dTAI, recognised on the basis of a high density of axonal injury showing a widespread distribution to relevant parts of the forebrain and hindbrain white matter, is likely more reliably predictive of a significant level of force and clinically significant ill-effect, but this can only be recognised in situations involving sufficiently prolonged survival for such axonal injury to become manifest. With known survival periods of less than 2 hours, axonal injury may be less readily apparent and in the absence of any history of profound unconsciousness without an alternative explanation and tissue-tear bleeds (other than in the corpus callosum or rostral brainstem), prudence again needs to be exercised on matters such as causation and associated symptomology. It has previously been suggested that falls from a standing height may involve insufficient biomechanical conditions to cause dTAI. This has been used to suggest that a victim may have been subject to additional modes of injury following a one punch precipitated heavy fall to the ground (for example kicks or stamping). Such a position is not tenable as it appears to have been informed by an earlier experience base with a greater emphasis on pathologically more severe forms of dTAI predating the wide application of β-APP immunostaining.

Axonal damage localised to the craniocervical junction Injuries that cause hypertension/hyperflexion of the neck cause direct damage to the brainstem, which at its most extreme

involves transection of the pontomedullary junction (Keane 1978; Gennarelli et al. 1981, Simpson et al. 1989; Ezzat et al. 1995). Full pontomedullary rents are rare (and, it should be said, easily created when inappropriate force is used to remove the brain post-mortem); partial transection of the lower brainstem is even rarer. However, although complete transection is immediately fatal, survival after partial damage has been described (Pilz et al. 1982; Pilz 1983). If there has been any survival, the damage can be confirmed as genuine by the presence of associated bleeding, noting though that primary axotomy might be detectable even in situations of potential instantaneous death. Lesser degrees of stress or damage centred on the craniocervical junction may lead to non-disruptive injury to the neuraxis in that region, only detectable histologically, i.e. localised TAI. This is a separate phenomenon from the more widespread axonal damage caused by acceleration of the head, described above and it may occur in the absence of axonal injury elsewhere in the brain. The mechanism of damage is likely to be stretch to the lower brainstem, where long tracts are particularly vulnerable, principally the corticospinal tracts in the lower pons and medulla. Damage to long tracts has been reported in cases from all age groups – in infants presumed to have suffered inflicted head injury, possibly shaking and in adults with high cervical dislocation or hyperextension injuries to the neck (Lindenberg and Freytag 1970; Hardman 1979; Riggs and Schochet 1995; Geddes et al. 2000, 2001). Blocks of several levels of lower brainstem may be needed to detect it, because the pathology varies greatly in severity from case to case (Figure 11.13).

Acknowledgements The author greatly acknowledges the contribution of Jennian Geddes to this chapter.

(a)

125

126

Traumatic axonal injury

References

(b)

(c)

Figure 11.13 Localised TAI at the craniocervical junction, ­demonstrated by immunohistochemistry for β-APP. The low power views show the anatomical restriction of the axonal damage. (a) Damage in the corticospinal tracts in a 32-yearold man, with fracture-dislocation of C1–C2. Localised TAI at the cranio-cervical junction. (b) Immunoreactive axonal swellings in the same tracts in the pons of a 2-month-old boy, who died of an inflicted injury. Though fewer than in (a) the swellings are still strikingly restricted to the descending long tracts. Localised TAI at the cranio-cervical junction. (c) A further example of localised TAI. Damaged axons in the left pyramid of a 31-year-old man who died after falling 25 ft. There was no axonal damage elsewhere in the brain of any of the three cases shown in this figure, and in each the damage was bilateral. The severity of damage seen in (a) is unusual.

Adams, J.H., Doyle, D., Graham, D.I., Lawrence, A.E. and McLellan, D.R. 1984. Diffuse axonal injury in head injuries caused by a fall. Lancet, ii, 1420–1422. Adams, J.H., Doyle, D., Graham, D.I., Lawrence, A.E. and McLellan, D.R. 1986. Deep intracerebral (basal ganglia) haematomas in fatal non-missile head injury in man. J Neurol Neurosurg Psychiatry, 49, 1039–1043. Adams, J.H., Doyle, D., Gennarelli, T.A., Graham, D.I. and McLellan, D.R. 1989a. Diffuse axonal injury in head injury: definition, diagnosis and grading. Histopathology, 15, 49–59. Adams, J.H., Doyle, D., Graham, D.I., McGee, M. and McLellan, D.R. 1989b. Brain damage in fatal non-missile head injury in relation to age and type of injury. Scott Med J, 34, 399–401. Adams, J.H., Graham, D.I., Gennarelli, T.A. and Maxwell, W.L. 1991. Diffuse axonal injury in non-missile head injury. J Neurol Neurosurg Psychiatry, 54, 481–483. Adams, J.H., Jennett, B., McLellan, D.R., Murray, L.S. and Graham, D.I. 1999. The neuropathology of the vegetative state after head injury. J Clin Pathol, 52, 804–806. Adams, J.H., Graham, D.I. and Jennett, B. 2000. The neuropathology of the vegetative state after acute brain insult. Brain, 123, 1327–1338. Anderson, R.M. and Opeskin, K. 1998. Timing of early changes in brain trauma. Am J Forensic Med Pathol, 19, 1–9. Blumbergs, P.C., Scott, G., Manavis, J., Wainwright, H., Simpson, D.A. and McLean, A.J. 1994. Staining of amyloid precursor protein to study axonal damage in mild head injury. Lancet, 344, 1055–1056. Blumbergs, P.C., Scott, G., Manavis, J., Wainwright, H., Simpson, D.A. and McLean, A.J. 1995. Topography of axonal injury as defined by amyloid precursor protein and the sector scoring method in mild and severe closed head injury. J Neurotrauma, 12, 565–572. Dolinak, D., Smith, C. and Graham, D.I. 2000. Global hypoxia per se is an unusual cause of axonal injury. Acta Neuropathol, 100, 553–560. Ezzat, W., Ang, L.C. and Nyssen, J. 1995. Pontomedullary rent. A specific type of primary brainstem traumatic injury. Am J Forensic Med Pathol, 16, 336–339. Geddes, J.F., Vowles, G.H., Beer, T.W. and Ellison, D.W. 1997. The diagnosis of diffuse axonal injury: implications for forensic practice. Neuropathol Appl Neurobiol, 23, 339–347. Geddes, J.F., Whitwell, H.L. and Graham, D.I. 2000. Traumatic axonal injury: practical issues for diagnosis in medicolegal cases. Neuropathol Appl Neurobiol, 26, 105–116. Geddes, J.F., Vowles, G.H., Hackshaw, A.K., Nickols, C.D., Scott, I.S. and Whitwell, H.L. 2001. Neuropathology of inflicted head injury in children. II. Microscopic brain injury in infants. Brain, 124, 1299–1306. Geddes, J.F. and Graham, D.I. 2004. Trauma to the central nervous system. In Gray, F., Di Girolami, U. and Poirier, R. (eds), Escourolle and Poirier’s Manual of Basic Neuropathology, 4th edn. Butterworth-Heinemann. Gennarelli, T.A., Adams, J.H. and Graham, D.I. 1981. Acceleration induced head injury in the monkey. I. The model, its mechanical and physiological correlates. Acta Neuropathol Suppl, 7, 23–25. Gennarelli, T.A., Thibault, L.E., Adams, J.H., Graham, D.I., Thompson, C.J. and Marcincin, R.P. 1982. Diffuse axonal injury and traumatic coma in the primate. Ann Neurol, 12, 564–574. Gennarelli, T.A. 1996. The spectrum of traumatic axonal injury. Neuropathol Appl Neurobiol, 22, 509–513. Gennarelli, T.A. 1997. The pathobiology of traumatic brain injury. Neuroscientist, 3, 73–81. Gennarelli, T.A., Thibault, L.E. and Graham, D.I. 1998. Diffuse axonal injury: an important form of traumatic brain damage. Neuroscientist, 4, 202–215. Gentleman, S.M., Roberts, G.W. and Gennarelli, T.A. et al. 1995. Axonal injury: a universal consequence of fatal closed head injury? Acta Neuropathol, 89, 537–543. Gorrie, C., Oakes, S., Duflou, J., Blumbergs, P. and Waite, P.M. 2002. Axonal injury in children after motor vehicle crashes: extent, distribution and size of axonal swellings using β-APP immunohistochemistry. J Neurotrauma, 19, 1171–1182.

References Graham, D.I., McLellan, D., Adams, J.H., Doyle, D., Kerr, A. and Murray, L.S. 1983. The neuropathology of the vegetative state and severe disability after non-missile head injury. Acta Neurochir Suppl, 32, 65–67. Graham, D.I., Lawrence, A.E., Adams, J.H., Doyle, D., McLellan, D. and Gennarelli, T.A. 1989. Pathology of mild head injury. In Hoff, J.T., Anderson, T. and Cole, T. (eds), Mild to Moderate Head Injury. Blackwell Scientific, pp 63–75. Graham, D.I., Clark, J.C., Adams, J.H. and Gennarelli, T.A. 1992. Diffuse axonal injury caused by assault. J Clin Pathol, 45, 840–841. Graham, D.I., Adams, J.H. and Gennarelli, T.A.. 1993. Pathology of brain damage in head injury. In Cooper, P.R. (ed), Head Injury. Williams & Wilkins, pp 91–113. Graham, D.I., Adams, J.H., Nicoll, J.A.R., Maxwell, W.L. and Gennarelli, T.A. 1995. The nature, distribution and causes of traumatic brain injury. Brain Pathol, 5, 397–406. Graham, D.I., Gennarelli, T.A. and McIntosh, T.K. 2002. Trauma. In Graham, D.I. and Lantos, P.L. (eds), Greenfield’s Neuropathology, 7th edn.  Arnold, pp 767–1052. Graham, D.I., Smith, C., Reichard, R., Leclercq, P.D. and Gentleman, S.M. 2004. Trials and tribulations of using β-amyloid precursor protein immunohistochemistry to evaluate traumatic brain injury in adults. Forensic Sci Int, 146, 89–96. Graham, D.I., Adams, J.H., Murray, L.S. and Jennett, B. 2005. Neuropathology of the vegetative state after head injury. Neuropsychol Rehabil, 15, 198–213. Hardman, J.M. 1979. The pathology of traumatic brain injuries. In Thompson, R.A. and Green, J.R. (eds), Complications of Nervous System Trauma. Raven Press, pp 15–50. Hortobágyi, T., Wise, S., Hunt, N., Cary, N., Djurovic, V., Fegan-Earl, A., Shorrock, K., Rouse, D. and Al-Sarraj, S. 2007. Traumatic axonal damage in the brain can be detected using β-APP immunohistochemistry within 35 min after head injury to human adults. Neuropathol Appl Neurobiol, 33, 226–237. Johnson, V.E., Stewart, J.E., Begbie, F.D., Trojanowski, J.Q., Smith, D.H. and Stewart, W. 2013. Inflammation and white matter degeneration persist for years after a single traumatic brain injury. Brain, 136, 28–42. Johnson, V.E., Stewart, W., Weber, M.T., Cullen, D.K., Siman, R. and Smith, D.H. 2016. SNTF immunostaining reveals previously undetected axonal pathology in traumatic brain injury. Acta Neuropathol, 131, 115–135. Keane, J.R. 1978. Intermittent see-saw eye movements. Report of a patient in coma after hyperextension head injury. Arch Neurol, 35, 173–174. Kinney, H.C. and Samuels, M.A. 1994. Neuropathology of the persistent vegetative state. A review. J Neuropathol Exp Neurol, 53, 548–558. Lambri, M., Djurovic, V., Kibble, M., Cairns, N. and Al-Sarraj, S. 2001. Specificity and sensitivity of betaAPP in head injury. Clin Neuropathol, 20, 263–271. Leclercq, P.D., Stephenson, M.A., Murray, L.S., McIntosh, T.K., Graham, D.I. and Gentleman, S.M. 2002. Simple morphometry of axonal swellings cannot be used in isolation for dating lesions after traumatic brain injury. J Neurotrauma, 19, 1183–1192. Lindenberg, R. and Freytag, E. 1970. Brainstem lesions characteristic of traumatic hyperextension of the head. Arch Pathol, 90, 509–515.

Maxwell, W.L., Islam, M.N., Graham, D.I. and Gennarelli, T.A. 1994. A qualitative and quantitative analysis of the response of the retinal ganglion cell soma after stretch injury to the adult guinea-pig optic nerve. J Neurocytol, 23, 379–392. Maxwell, W.L., Povlishock, J.T. and Graham, D.I. 1997. A mechanistic analysis of nondisruptive axonal injury: a review. J Neurotrauma, 14, 419–440. Maxwell, W.L., Domleo, A., McColl, G. and Graham, D.I. 2003. Post-acute alterations in the axonal cytoskeleton after traumatic axonal injury. J Neurotrauma, 20, 151–168. Pilz, P., Strohecker, J. and Grobovschek, M. 1982. Survival after traumatic ponto-medullary tear. J Neurol Neurosurg Psychiatry, 45, 422–427. Pilz, P. 1983. Survival after ponto-medullary junction trauma. Acta Neurochir Suppl, 32, 75–78. Povlishock, J.T. 1993. Pathobiology of traumatically induced axonal injury in animals and man. Ann Emerg Med, 22, 980–986. Reichard, R.R., Smith, C. and Graham, D.I. 2005. The significance of β-APP immunoreactivity in forensic practice. Neuropathol Appl Neurobiol, 31, 304–313. Riggs, J.E. and Schochet, S.S. 1995. Spastic quadriparesis, dysarthria and dysphagia following cervical hyperextension: a traumatic pontomedullary syndrome. Mil Med, 160, 94–95. Sherriff, F.E., Bridges, L.R., Gentleman, S.M., Sivaloganathan, S. and Wilson, S. 1994. Markers of axonal injury in post mortem human brain. Acta Neuropathol, 88, 433–439. Simpson, D.A., Blumbergs, P.C., Cooter, R.D., Kilminster, M., McLean, A.J. and Scott, G. 1989. Pontomedullary tears and other gross brainstem injuries after vehicular accidents. J Trauma, 29, 1519–1525. Smith, D.H., Nonaka, M. and Miller, R. et al. 2000. Immediate coma following inertial brain injury dependent on axonal damage in the ­brainstem. J Neurosurg, 93, 315–322. Stone, J.R., Okonkwo, D.O., Dialo, A.O., Rubin, D.G., Mutlu, L.M., Povlishock, J.T. and Helm, G.A. 2004. Impaired axonal transport and altered axolemmal permeability occur in distinct populations of damaged axons following traumatic brain injury. Exp Neurol, 190, 59–69. Strich, S.J. 1956. Diffuse degeneration of the cerebral white matter in severe dementia following head injury. J Neurol Neurosurg Psychiatry, 19, 163–185. Strich, S.J. 1961. Shearing of nerve fibres as a cause of brain damage due to head injury. Lancet, 2, 443–448. Tomei, G., Spagnoli, D. and Ducati, A. et al. 1990. Morphology and neurophysiology of focal axonal injury experimentally induced in the guinea pig optic nerve. Acta Neuropathol, 80, 506–513. Troakes, C., Smyth, R., Noor, F., Maekawa, S., Killick, R., King, A. and Al Sarraj, S. 2017. Clusterin expression is upregulated following acute head injury and localised to astrocytes in old head injury. Neuropathology, 37, 12–24. Wilkinson, A.E., Bridges, L.R. and Sivaloganathan, S. 1999. Correlation of survival time with size of axonal swellings in diffuse axonal injury. Acta Neuropathol, 98, 197–202.

127

12

Brain swelling, raised intracranial pressure and hypoxia-related brain injury Daniel du Plessis

Brain swelling with raised intracranial pressure and secondary hypoxic-ischaemic brain injury has a considerable adverse impact on the morbidity and mortality of traumatic head injuries.

Brain swelling Brain swelling (also known as cerebral swelling) refers to volumetric expansion of the brain which may be recognised (1) as an increased brain weight (relative to the normal baseline brain weight for the individual in question) and (2) changes in the contour and shape of the brain reflecting expansion of the brain and (if severe enough) consequent displacement and distortion of brain tissue (herniation). Brain swelling may be caused by a variety of mechanisms. The two major causes of brain swelling highly relevant to forensic neuropathology practice are brain (cerebral) oedema and congestive swelling.

Cerebral oedema There are various pathophysiological forms of cerebral oedema. The most common and most relevant to this discussion are cytotoxic oedema and vasogenic oedema. Cytotoxic oedema involves abnormal intracellular accumulation of water. Operating alone, it is (at least initially) associated with a reduced extracellular space and an intact blood-brain barrier (BBB) (Klatzo 1967). It is most commonly promoted by conditions involving energy failure such as hypoxic-ischaemic conditions. Should conditions conducive to cytotoxic oedema persist long enough, vasogenic oedema will intervene (Bullock et al. 1994). This is associated with breakdown of the BBB. As neurons are most demanding in terms of energy requirements, cytotoxic oedema preferentially affects grey matter. Isolated cytotoxic oedema without any vasogenic oedema does not cause the brain to swell as it causes a redistribution of water from the extracellular space to the intracellular space. Combined cytotoxic and vasogenic oedema leads to brain swelling. Vasogenic oedema is the most common form of cerebral oedema. This involves breakdown of the BBB with increased vascular permeability. Vasogenic oedema is seen as a local phenomenon around lesions within the brain such as tumours, abscesses, major bleeds and strokes (Kalaria et al. 2015). Vasogenic oedema may lead to a more disproportionate accumulation of water within white matter relative to grey matter in contrast to cytotoxic oedema. Hypoxic-ischaemic conditions may promote such a leaky state and account for vasogenic oedema. Other physiological factors, which may contribute to the development of vasogenic oedema include hypertension, hypercapnia and vasodilatory agents (Atkinson et al. 1998). Hydrostatic oedema is caused by an increase in intravascular pressure in an intact vascular bed. This results in the

accumulation of protein-poor fluid in the extracellular spaces. Hydrostatic oedema is typically seen in severe hypertension or sudden impedance of venous drainage of blood. Interstitial oedema involves an increase in water content in the periventricular white matter as a result of acute obstructive hydrocephalus.

Congestive brain swelling Under certain circumstances, an increase in the volume of blood within the vessels within the brain and/or the leptomeninges causes engorgement of blood vessels, referred to as congestion. This can lead to substantial brain swelling and an increased brain weight. Congestion may be caused by an influx of blood from the arterial side of the circulation (such as may occur in a hypertensive crisis or loss of cerebral autoregulation). Alternatively, factors which may impede drainage of blood from the brain may result in a venous, more passive form of congestion. Congestive brain swelling can be very rapid, severe and life threatening in situations involving loss of cerebral autoregulation where the increase in blood volume appears to occur predominantly in the capillary and postcapillary bed. Brainstem stimulation through acceleration/deceleration type injury may increase cerebral blood flow (CBF) without increasing cerebral metabolism. This effect may on occasion be delayed and independent of the degree of primary injury, being more reliant on the mechanism of trauma. Alternatively, tissue injury may release chemical mediators into cerebrospinal fluid resulting in the stimulation of brainstem areas implicated in the regulation of CBF, causing sudden vasodilatation. Trauma may also have a direct effect on cerebral blood vessels, producing alterations in cerebrovascular tone. It is suggested that these vessels have not lost autoregulation, but simply have a new resting tone and remain responsive to carbon dioxide levels. The swelling appears to be mainly due to an increase in intracerebral blood volume, either as a true increase in cerebral blood volume or as a redistribution of intracranial blood from the blood vessels within the membranes surrounding the brain to vessels within the substance of the brain. With severe congestive brain swelling, it is possible that hydrostatic oedema may also intervene to contribute to swelling, high vascular pressures forcing water out of vessels into brain tissue. With more prolonged severe congestive swelling vasogenic oedema may supervene once the BBB becomes defective.

The post-mortem recognition of brain swelling and brain oedema Volumetric expansion of the brain (brain swelling) will manifest as (1) an increase in the brain weight from its normal baseline

Brain swelling weight and (2) external and internal alterations in the size and shape of the brain.

Brain weight Normal paediatric and adult brain weights are provided in Appendices 1, 2 and 3 in Chapter 5. An assessment of brain weight as a surrogate measure of brain swelling requires consideration of the subject’s age and sex, but also body height and weight. In practice, higher than average brain weights serve to suggest rather than confirm brain swelling (viewed in isolation without recourse to the appearances of brain), save in situations where the brain weight is substantially elevated beyond the normal expected range.

Macroscopy Advance correlation with neuroimaging may provide the most objective evidence of significant swelling and critically raised intracranial pressure. The removal of the brain should preferably be done by the pathologist (or review carried out of the original post-mortem photographs in referred cases). This allows appreciation of a ‘tight’ brain with dura stretched over the brain surface and technical difficulty in removing of the brain, including the ability to cut through the upper cervical spinal cord due to downward displacement of brainstem structures. The brain will show flattening of the gyri and obliteration of sulci (Figure 12.1). Recesses other than sulci, such as the basal cisterns, may also become less pronounced (space compromised) and the ventricles become compressed. More severe brain swelling will show shift and displacement of brain tissue referred to as herniation, although in infants before closure of the sutures the typical picture may not be seen. Acute lateral displacement of the midbrain and hypothalamus may be fatal in the absence of discernible cerebral herniation (Miller-Fisher 1995). The lesser wing of the sphenoid bone may produce a groove on the orbital surface of the frontal lobe. With asymmetric mass effect, obliteration of the foramina of Monro contralateral to the source of mass effect may lead to enlargement of the contralateral ventricle. Temporal lobe lesions may cause shift of the third ventricle and pressure on the ipsilateral Sylvian fissure compromising the middle cerebral artery in extreme situations.

Figure 12.2 Marked left-to-right shift with subfalcine herniation.

Internal examination of the brain may reveal ventricular compression. Internal anatomical landmarks can also be utilised to confirm significant mass effect. This includes caudal displacement of the posterior part of the floor of the third ventricle below the level of the tentorial incisure and displacement of the tips of the mammillary bodies 5 mm or more below the Greenhall line.

Subfalcine herniation In this situation, the ipsilateral cingulate gyrus herniates under the falx (Figure 12.2) due to mass effect in one cerebral hemisphere. This may lead to infarction in the parasagittal cortex due to compression of the pericallosal arteries. A wedge of necrosis where the gyrus is in contact with the falx may occur. This is mostly seen in frontal or parietal regions because the posterior falx is tethered. The latter though may cause micro-tearing of the dorsal midline of the posterior corpus callosum, which may leave a gliotic scar with survival.

Transtentorial uncal (parahippocampal) herniation A supratentorial mass may produce herniation of the ipsilateral uncus of the temporal lobe as well as the medial part of the parahippocampal gyrus (Figure 12.3). This is displaced through the tentorial opening. It is also known as lateral

Figure 12.1 Severe brain swelling with gyral flattening.

Figure 12.3 Left uncal herniation viewed from base of brain.

129

130

Brain swelling, raised intracranial pressure and hypoxia-related brain injury the cerebellar peduncles. Stretching and narrowing of the perforating branches of the basilar artery may cause brainstem infarction and haemorrhage (Hassler 1967). Susceptibility to this phenomenon may be increased by sudden alleviation of severely raised intracranial pressure leading to reperfusion of previously ischaemic brainstem (Klintworth 1966).

Tonsillar herniation (cerebellar coning)

Figure 12.4 Posterior cerebral artery territory infarction.

transtentorial herniation. Compression of the contralateral cerebral peduncle against the free edge of the tentorium may result in pressure-related secondary injury with or without haemorrhage in the dorsal part of the peduncle and adjacent tegmentum, a lesion referred to as Kernohan’s notch. Ipsilateral cerebral peduncular compression may lead to contralateral limb weakness as a more common associated feature of severe asymmetric tentorial herniation. The oculomotor nerve may show kinking around the ipsilateral posterior cerebral artery and pressure related bleeding. Necrosis may occur along the parahippocampal gyrus. Compression of the anterior choroidal arteries may cause infarction in the medial part of the globus pallidus, internal capsule and optic tract. A not uncommon complication is compression of the posterior cerebral artery leading to infarction in the posterior inferior temporal lobe including the hippocampus, together with the medial and inferior surfaces of the occipital cortex and even thalamus. This is characteristically haemorrhagic in nature (Figure 12.4). Cerebellar infarction may also occur due to compression of a superior cerebellar artery (SCA). These infarctions are most commonly ipsilateral to the side of the mass lesion but can be bilateral.

Displacement of the tonsils through the foramen magnum causes these structures to partly envelope the lower brainstem, tightly hugging the latter and assuming a funnel or inverted cone shaped appearance – hence the term coning or tonsillar herniation (synonymous terms). Coning is further recognised by smoothing of the folds (folia) of the cerebellar tonsils accompanied by surrounding circular impression by the rim of the foramen magnum (referred to as tonsillar grooving) and by grooving of the ventral medullary surface as a result of compression against the anterior border of the foramen magnum. With severe, sustained compression of the tonsils within the foramen magnum these structures may undergo necrosis with or without haemorrhage. Sloughed necrotic tonsillar tissue may circulate in the subarachnoid space around the spinal cord. Severe tonsillar herniation may further cause occlusion of branches of the posterior inferior cerebellar artery (PICA) (Figure 12.5). Prolapse of haemorrhagic choroid plexus through the medullary exit foramina also reflects tonsillar herniation.

Upward tentorial herniation (reversed tentorial herniation) This occurs due to herniation of the superior cerebellar surface through the tentorial opening caused by a posterior fossa mass. Conditions which may produce this include cerebellar tumours, haemorrhage or other space occupying cerebellar lesion as well as posterior fossa subdural or extradural haemorrhage. This may result in haemorrhagic infarction of the SCA territories.

External cerebral herniation Brain tissue may herniate externally through any defect – these may be traumatic or surgical. Small amounts of cortex may

Central transtentorial herniation This may be preceded by lateral tentorial herniation or intervene de novo due to rapidly expanding fronto-parietal lesions or bilateral expanding lesions such as a chronic subdural haematoma. It produces a circular hernia involving both parahippocampal gyri which is most prominent posteriorly and may compress the tectal plate. Accompanying diencephalic ‘down thrust’-related distortion and displacement of adjacent structures may lead to infarction of the mamillary bodies and the anterior pituitary. Infarctions within the vascular territories of the anterior choroidal, posterior cerebral and SCAs are frequently encountered. Secondary infarcts and haemorrhages may be seen in the upper brainstem in the most severe cases. These tend to show a midline to paramidline distribution in the midbrain tegmentum and basal and tegmental parts of the pons. More eccentric bleeds may also be found in the dorsolateral quadrants of the upper brainstem, these requiring distinction from primary traumatic tissue-tear bleeds associated with diffuse traumatic axonal injury (see Chapter 11). Such bleeds can also extend further into the floor of the fourth ventricle and into

Figure 12.5 Cerebellar tonsillar herniation showing haemorrhagic necrosis of the tonsils.

Brain swelling

Figure 12.6 Herniation of cortex in relation to craniotomy site.

Figure 12.7 Histological appearances of oedema – particularly prominent in white matter.

Microscopy protrude through burr holes (see Chapter 5, Figure 5.7) and large areas through decompressive craniectomy sites (Figure 12.6).

Paradoxical herniation Following some craniectomies, the forces of atmospheric pressure and gravity overwhelm intracranial pressure causing the brain to appear sunken. In more extreme form, this can lead to paradoxical herniation (Fields et al. 2006). Other synonyms for this condition/complication are so-called sinking skin flap syndrome, brain sag or the syndrome of the trephined. The MonroKellie doctrine defines the physiological relationship between the intracranial volumes contributing to intracranial pressure, namely brain tissue, blood and cerebrospinal fluid. These close relationships take place within a closed system. Under normal conditions cerebrospinal fluid leakage is quickly compensated by an increase in intracranial venous volume, thus maintaining physical support of the brain. When this system is ‘open’ to atmospheric pressure, the Monro-Kellie principles do not apply and a loss in the CSF compartment leads to serious difficulties maintaining the brain afloat/supported, which can lead to socalled paradoxical or internal herniation. In craniectomy, a negative pressure gradient is created between the intracranial and atmospheric compartments (temporarily or permanently) which makes cerebral herniation possible even the absence of increased intracranial pressure. The risk of paradoxical herniation in the setting of a craniectomy appears enhanced by the intervention of a CSF drainage procedure such as the insertion of a ventriculo-peritoneal shunt (VP shunt) or by carrying out a lumbar puncture/drainage procedure (Marquez-Romero et al. 2010; Jung et al. 2012). Other complications of decompressive craniectomies intervene in sequential manner (Stiver 2009). Expansion of contusions, new subdural and extradural haematomas contralateral to the decompressed hemisphere and external cerebral herniation typify the early perioperative period following decompressive surgery. Within the first week following decompression, cerebrospinal fluid circulation derangements manifest commonly as subdural hygromas. During the later phases of recovery patients may develop a new cognitive, neurological or psychological deficit termed ‘syndrome of the trephined’. A persistent vegetative state may even develop in the longer term as the most devastating of outcomes of a decompressive craniectomy.

Histological features of oedema include myelin pallor, rarefaction, swelling and vacuolation (Figure 12.7). Swollen and beaded myelinated fibres may appear splayed, especially around blood vessels. Immunohistochemistry can demonstrate the occupation of enlarged extracellular spaces by protein-rich fluid (utilising antibodies to albumin and immunoglobulins). After a day or so, the astrocytes appear swollen with gliosis occurring later. If oedema persists, macrophages infiltrate with breakdown of myelin. Oligodendroglial numbers may also become reduced. Small cystic spaces may result in the white matter. Foci of scarring at sites susceptible to herniation such as the anterior cingulate gyri, parahippocampal gyri and the posterior dorsal midline corpus callosum should alert the pathologist to a previous episode of significantly raised intracranial pressure. Points to note: • Some non-severe swelling of the brain is a relatively common finding in deaths encountered in medico-legal practice. The pathophysiology is not clearly understood, but this usually related to an increase in intravascular blood. Conditions in which this is recognised to occur include ‘asphyxial’-type deaths and others where there may be a prolonged congestive phase. Histologically, congestion of intracerebral vessels with perivascular haemorrhages (Figure 12.8) may be seen. Whist these may be seen in conditions such as suffocation or overlying in an infant, as well as ‘asphyxial’ deaths in adults, such features should not be taken as diagnostic for these conditions. • It is important not to overinterpret minor uncal grooving and tonsillar prominence as isolated features alone or in combination as reliable evidence of significant mass effect. It is well recognised that the cerebellar tonsils descend and ascend within the foramen magnum with normal changes in respiration and blood pressure. It is not uncommon to see prominent cerebellar tonsils at post-mortem examination in the absence of other features consistent with severe brain swelling. Prominent tonsils should further prompt consideration of Chiari type 1 malformations. Minor uncal grooving may also be seen at post-mortem where there is no cerebral pathology and again care should be taken in its interpretation (Shapiro et al. 1988).

131

132

Brain swelling, raised intracranial pressure and hypoxia-related brain injury perivascular pool of aquaporin 4 (AQP4), which is rate limiting for water influx during cerebral oedema formation. Points to note:

Figure 12.8 Petechial perivascular haemorrhages.

Rate of swelling Initially it was thought that congestive swelling primarily caused post-traumatic brain swelling with a lesser role for oedema. It is now established that brain oedema is usually the major contributor to traumatic brain swelling and that cerebral blood volume is reduced in proportion to CBF reduction following severe traumatic brain injury. Vasogenic oedema is initially effective within the first few hours of traumatic brain injury, the BBB becoming maximally permeable at 4–6 hours after traumatic brain injury before commencing to close and becoming differentially permeable to smaller molecules over a 7 day period. Oedema is maximal at 2–3 days after brain trauma, which is also when intracranial pressure usually peaks. It is thought that the initial transient opening of the BBB is associated with a brief period of ‘pure’ vasogenic oedema and that slower, but more widespread cytotoxic oedema subsequently develops with replacement of the initial vasogenic phase by a mixed cytotoxic/vasogenic phase caused by cytotoxic injury as more cells become dysfunctional and die (Barzó et al. 1997; Bullock et al.1994; Cernak et al. 2004). In experimental setting, cerebral oedema can contribute to brain swelling as early as 30 minutes after closed head trauma causing contusions. This may be accompanied by transient, minimal cortical oedema perhaps related to ischaemia (Tornheim and McLaurin 1981). Vasodilatation and hyperaemia (increased blood flow) have also been demonstrated in humans (usually in a paediatric setting) within 30 minutes of trauma (Bruce et al. 1981; Yoshino et al. 1985). The detection of brain swelling in a post-traumatic setting may therefore not imply a period of prolonged survival, but it should be emphasised that in the event of very short survival the main driving force will be congestive swelling on the basis of vasodilatation and not vasogenic oedema (Byard et al. 2009, 2012; Byard and Fink 2013). Appreciable brain swelling due to post-cardiac arrest hypoxic-ischaemic injury mandates a period of survival. Computerised tomography (CT) imaging may demonstrate sulcal effacement as soon as 20 minutes after restitution of spontaneous circulation in survivors of out-of-hospital cardiac arrest (Inamasu et al. 2010). Diffuse brain oedema early after cardiac arrest is often associated with poor neurology and outcome. The mechanisms of cardiac arrest-induced cerebral oedema are likely multifactorial with a possible significant role for the

• There have been case reports of individuals suffering instantaneous to near instantaneous death (within seconds) with evidence of substantial brain swelling in the absence of any non-traumatic explanation (Byard and Fink 2013). Brain evisceration in catastrophic head trauma may though provide for an explanation purely on a biomechanical basis, i.e. sudden decompression rather than an in vivo physiological response. In such cases, the presence of brain swelling and parenchymal micro-bleeds is of further forensic value in that it appears to indicate that the subject was alive at the time of injury (Z´ivkovic´ et al. 2020). • The author’s observations gained from review of the neuropathology of those who died in the Hillsborough Stadium Disaster suggest that venous congestion might also contribute substantially to brain swelling in deaths related to an asphyxial event combined with impeded cerebral venous drainage where survival was too short to permit appreciable oedematous swelling; this may be operative not only in cases involving static or dynamic crush asphyxia, but also hangings. • The phenomenon of very rapid, severe brain swelling (malignant brain swelling) is well described in children (Graham et al. 1989) but may occur in any age group. The CT appearance of diffuse swelling may though develop more readily in children due to the lack of cerebrospinal fluid available for displacement. In children, diffuse swelling may have a relatively benign course, unless there is a severe primary injury or secondary hypotensive insult (Lang et al. 1994). Other factors such as age and sex have also been implicated in the severity of brain swelling after head injury (Farin et al. 2003). It is well recognised that rapid neurological deterioration can follow within minutes or be delayed after head injury (Bruce 1984). A number of these cases follow a benign course (Snoek et al. 1984) and the syndrome can occur with relatively trivial trauma. More recently, it has been reported that delayed cerebral oedema can occur in the familial hemiplegic migraine syndrome associated with the CACNA-1A gene (Kors et al. 2001) and with apolipoprotein E deficiency in the experimental situation (Lynch et al. 2002). The pathology shows diffuse generalised brain swelling with little evidence of brain injury. It appears that the diffuse swelling may, in a number of these cases, be related to vascular congestion with hyperaemia rather than true cerebral oedema (Bruce et al. 1981; Sakas et al. 1997). Features of pathologically elevated intracranial pressure in children may differ from those validated in adults. Basal cisternal volume has been found to correlate significantly with opening intracranial pressure in paediatric head injuries in contrast to ventricular volume and extra-axial haematoma volume.

Brain injury related to hypoxia and impaired perfusion Hypoxic-ischaemic brain injury Neuronal integrity and normal functioning are highly dependent on adequate brain perfusion maintaining not only the

Brain injury related to hypoxia and impaired perfusion delivery of sufficient oxygen and glucose to cells but also the essential removal of neurotoxic metabolites such as lactate (Miyamoto and Auer 2000). Hypoxia describes a reduction of oxygen supply or content (in blood or tissue) but may also be applied to impaired metabolic utilisation of oxygen. Various types of hypoxia are listed in Table 12.1. Ischaemia refers to cessation of blood flow to tissue, although this term is commonly used to denote insufficient blood supply below the level required to maintain normal tissue function (Kalaria et al. 2015). Pure hypoxia alone with maintained adequate perfusion does not lead to discernible pathology at light microscopic level and reference to ‘hypoxic brain injury’ should be dissuaded. Structural, detectable pathology is introduced by hypoxia in combination with other adverse pathophysiological conditions (effectively creating a two-hit situation), most commonly combined with ischaemia, but also in combination with other metabolic stresses such as hypoglycaemia or underlying metabolic disorders such as mitochondrial disease. If implicated in tissue injury, hypoxia should be qualified by its cause as well as the company it keeps. In this respect, the term hypoxic-ischaemic brain injury is commonly used (although effectively a tautology), which is synonymous with stagnant hypoxia. Acute neuronal injury thus results from ischaemia (with attendant hypoxia) which prevents the removal of neurotoxic metabolites with a resultant drop in tissue pH (Siesjö 1992) and not from hypoxia alone.

Focal cerebral ischaemia and infarction Thrombotic occlusion of large extracranial and intracranial arterial vessels is usually associated with atheromatous plaques. Stenosis exceeding 90 per cent though may be sufficient to cause infarction (Kalaria et al. 2015). Most infarcts are due to thromboembolic events (60–80 per cent), either due to artery-to-artery embolism or cardioembolic strokes. The bifurcation of the common carotid is the most common source of artery-to-artery emboli. Thrombosed and severely stenosed arteries with reduced flow is usually of the fibrin-dependent ‘red’ variety, whereas ‘white’ plateletfibrin thrombi form on plaques in vessels with brisker flow. Most cardiogenic thrombo-emboli occlude parts of the middle cerebral arteries.

Figure 12.9 Widespread petechial haemorrhages secondary to fat embolism. A variety of other rarer forms of embolism may also be encountered in forensic practice.

Fat embolism Injury of fat-containing tissues may release fat emboli in the circulation. This includes bone fractures, pancreatitis, burns or trauma to viscera or subcutaneous, deep soft tissue. Neurological and respiratory symptoms usually become manifest 12–72 hours after injury. Macroscopically the brain is swollen, and there are widespread petechial perivascular haemorrhages (Figure 12.9) occasionally accompanied by perivascular micro-infarcts best demonstrated with myelin stains. Frozen sections of brain tissue should be examined to confirm the fat emboli (Kamenar and Burger 1980) (Figure 12.10). Although white matter appears to bear the brunt of injury, fat globules may be more prevalent in grey matter blood vessels.

Air emboli These may be associated with decompression of deep-sea divers (Caisson’s disease) or with cardiac surgery. Microinfarcts may be seen in the spinal cord (Palmer et al. 1987) and cerebrum.

Table 12.1 Classification of hypoxia Type of hypoxia

Pathophysiology

Clinical example

Structural changes in the brain

Hypoxic

Reduced oxygen tension in pulmonary alveoli results in hypoxaemia and brain hypoxia

Acute: pulmonary oedema Chronic: interstitial lung disease

None

Anaemic

Hypoxaemia caused by low levels of haemoglobin or by competitive binding of oxygen sites on the haemoglobin molecule

Carbon monoxide poisoning

None if not accompanied by ischaemia

Histotoxic

The neuron is unable to use the oxygen as a result of poisoning of the metabolic pathways within the cell

Cyanide poisoning

None

Stagnant (associated with oligaemia/ ischaemia)

There is reduced blood flow to the brain caused by either reduced cardiac output or disruption of local perfusion

Reduced cardiac output: cardiac arrest Disrupted local perfusion: thromboembolic infarct

Neuronal necrosis and if prolonged tissue infarction

133

134

Brain swelling, raised intracranial pressure and hypoxia-related brain injury

Figure 12.10 Fat embolus seen in a section of frozen brain tissue (oil red O, ×40).

Figure 12.11 Section of cerebellar cortex from a case of case of cardiac arrest with survival of several weeks. There is loss of Purkinje cells with reactive Bergmann gliosis (H&E, ×40).

Fibrocartilaginous emboli Intervertebral disc material has been described as a cause of cerebral infarction after minor trauma (Toro-Gonzalez et al. 1993). It is also rarely associated with pulmonary emboli.

Foreign body emboli Penetration of extracranial major arterial vessels by shotgun pellets may cause cerebral infarcts (Dada et al. 1993). Acrylic cements and other substances used in orthopaedic spinal surgery may undergo venous migration and result in pulmonary embolism as well as cerebral arterial embolisation, the latter in the presence of a patent foramen ovale permitting paradoxical embolisation of the cement (Scroop et al. 2002).

Global ischaemia Global ischaemia may be absolute in situations where CBF ceases for a period of time, such as during cardiac arrest, or may be variable in situations where the CBF is compromised, such as with systemic hypotension or raised intracranial pressure.

Figure 12.12 Bilateral watershed infarcts secondary to hypotension.

Brain damage after cardiac arrest In this situation, there is cessation of blood flow to the brain. It is important to remember that the pathological changes associated with cardiac arrest require successful resuscitation, with subsequent survival for a period of hours before it can be identified. The period of cessation of CBF may last only 5–10 minutes for irreversible brain damage to develop. Macroscopic identification of brain damage, however, requires a survival of approximately 36–48 hours. Cortical discolouration may be identified (early laminar necrosis), particularly in the parietal and occipital cortices, or may be accentuated within the depths of gyri. Microscopic diffuse neuronal ischaemic damage can be seen after 4–10 hours, initially with a pattern of selective vulnerability. Selective vulnerability refers to the fact that early in global ischaemia neurons in certain regions will be affected before others (Pulsinelli 1997) (Figure 12.11). The selective vulnerability associated with ischaemia differs from that seen in hypoglycaemia (see Table 18.1, Chapter 18).

Brain damage after hypotension In this situation, there is reduced perfusion of the brain. Mild hypotension or brief cardiac arrest injury may be restricted to

the posterior triple watershed regions. The anterior cerebral artery (ACA)/middle cerebral artery (MCA) boundary zone is affected in more severe global cerebral ischaemia (Figure 12.12). If the hypotension is severe and/or prolonged, the situation begins to mimic that seen in cardiac arrest with extensive cortical injury. Watershed infarcts may be unilateral due to asymmetry of the circle of Willis, either due to anatomical factors or compromise by acquired disease such as atherosclerosis. The cerebellum also has boundary zones, the most involved being between the SCAs and the PICAs. If hypotension is mild or quickly reversed, damage may be limited to Purkinje cell ischaemic damage or focal infarction within the boundary zones. If more severe or prolonged, the injury is extensive within the cerebellar cortex.

Timing of ischaemic changes The pathological changes seen at various stages of infarct evolution are detailed in Table 12.2 (Itabashi et al. 2007; Kalaria et al. 2015) and illustrated in Figures 12.13 and 12.14. The ‘dictum’ that red neurons confirm survival beyond 4 hours or more should not be relied upon as absolute proof of such survival. Classic, overt red neuronal change has been identified

Brain injury related to hypoxia and impaired perfusion

Table 12.2 Timing of changes in thromboembolic infarcts Time since infarct (hours)

Macroscopic appearance

Light microscopic appearance

1–12

None

Irreversible ischaemic cell change

12–24

Early loss of grey– white matter interface

Neutrophilic infiltrate although may be relatively inconspicuous

48

Cerebral oedema established, with associated mass effect. There may be tissue splitting

Activated macrophages present in damaged tissues

1–3 weeks

Cavitation begins to develop

Gliosis and neovascularisation identified

Months

Gliotic scar often golden-brown in colour as a result of haemosiderin staining

Gliotic scar

under hypoxic-ischaemic conditions earlier than 4 hours. Red neurons have been described in infarcts within 2–5 hours (Itabashi et al. 2007; Kitamura 1994). Leestma (2014) though described the evolution of red neuronal change ‘in as little as an hour or two and possibly slightly less’, an experience shared by the author. Such very early intervention of red neuronal change may represent very acute ischaemic episodes in previously healthy individuals such as in suicidal or accidental hangings with prompt rescue and attempted but failed resuscitation, acute cardiac arrest with attempted resuscitation followed by short survival and anaesthetic accidents. Firm conclusions in terms of potential survival therefore require a cautionary approach.

Red neurons versus dark neurons The recognition of true red neurons can be a challenging exercise. In forensic practice, it is essential to be very careful when diagnosing the presence of red neurons given implications both in terms of causation of injury and duration of survival. Red neurons should be distinguished from so-called dark neurons, a less specific neuronal change which may represent a postmortem artefact but also potentially an early and reversible stage of acute neuronal injury due to ischaemia, hypoglycaemia and epileptic seizures. It is very difficult, if not impossible, to distinguish truly pathological dark neurons as a precursor

Figure 12.13 Macroscopic appearances of cerebral infarcts at different ages. (a) Recent infarct (24–48 hour) with swelling and discolouration of the infarcted tissue. (b) Recent infarct (48+ hour) with cerebral swelling and mass effect. The splitting of the tissue is artefact and represents the infarcted tissues separating from viable tissues. (c) Old infarct (months) with loss of the insular cortical tissue and associated white matter. The discolouration is caused by haemosiderin deposition.

135

Brain swelling, raised intracranial pressure and hypoxia-related brain injury

136

(a)

Figure 12.15 Red neuronal change (H&E, ×40).

(b)

versus more generalised impaired blood and oxygen supply to the brain as may follow cardiac arrest). H&E staining may differ in quality (intensity and colour) between laboratories and within the same laboratory on different occasions. Post-mortem autolysis can also compromise interpretation; increased cytoplasmic eosinophilia in the deep grey matter may occur with poor fixation). The following features should be identified (Figure 12.15): • Bright red to pink and not magenta cytoplasm. • Pyknotic or disintegrating nuclei (pyknosis refers to nuclear shrinkage with dark intense staining and irregular profiles). • Indistinct nucleoli and Nissl bodies. • Prominent cell shrinkage. • Red neuronal change should occur in groups of cells and not simply in rare, isolated cells interspersed with normal appearing neurons. It is acknowledged that metabolic differences between cells may allow some to develop these changes earlier than others and it is therefore possible that isolated cells surrounded by normal cells could represent cells possibly (not definitely) reflecting an early stage of injury in the form of dark neurons.

(c)

Figure 12.14 Microscopic appearances of cerebral infarcts at different ages. (a) Recent infarct (hours) with neutrophilic infiltration. Note the neuronal eosinophilia consistent with ischaemic neuronal damage (arrow) (H&E, ×40). (b) Recent infarct (48+ hour) with macrophage infiltration (H&E, ×40). (c) Old infarct (months) with gliosis at the edge of the area of infarction (H&E, ×40).

to red neurons from artefactual dark neurons. Delayed fixation of brain tissue following death (in excess of 10 hours) promotes artefactual dark neuronal change as does manipulation of brain tissue prior to fixation. Age, general health and the abruptness of the terminal illness may influence the appearance of red neurons (as indicated previously, the minimum period of survival required for the development of red neurons may differ according to the specific hypoxic-ischaemic situation, e.g. strokes

A key point made by Itabashi et al. (2007) is that ‘a more convincing totality of findings is demanded for forensic purposes’. Subtle cytoplasmic increased eosinophilia or a magenta tint to the cytoplasm in a minority of neurons should not be considered sufficient for diagnosis of definite hypoxic-ischaemic neuronal injury. Dark neurons in contrast have the following attributes (Figure 12.16): • A monotonous, often shrunken appearance. • Basophilic with H&E staining, having a dark blue perikaryal and dendritic cytoplasm. • Slight eosinophilia may be superimposed giving a dark blue-red tint with H&E staining stain. • Apical dendrites may have an irregular, corkscrewshaped appearance. • The shrunken, dark-stained nucleus may be indistinct within the cell body since it blends into the compacted perikaryal cytoplasm.

References

(a)

Figure 12.16 Dark neurons (H&E, ×40). • Nucleoli though still discernable. • Affected neurons may be separated from adjacent neuropil, especially with paraffin embedded tissue and are often scattered among histologically normal neurons (Jortner 2006).

The non-perfused brain (respirator brain) This occurs when perfusion to the brain ceases when the mean arterial blood pressure is insufficient to overcome the intracranial pressure to allow blood to flow through the brain. Blood stagnates in the cerebral circulation. Macroscopically, there is swelling of the brain with evidence of herniation and darkgrey dusky discolouration. Fixation is poor and central parts of the brain such as the basal ganglia may disintegrate. There is blurring of the grey-white junction of the cortex including the cerebellar cortex (Figure 12.17a). Fragments of cerebellar tissue may be found around the spinal cord. Histologically, there is general tissue pallor, lysed red cells with stagnation and perineuronal vacuolation with neuronal pallor (Figure 12.17b). Tissue reaction is lacking. This may make interpretation of timing of injuries difficult. These changes take around 12 hours to develop, becoming more obvious with the passage of time (Black 1978). Bacterial overgrowth which is seen in post-mortem autolysis is usually not seen.

Delayed posthypoxic leukoencephalopathy Delayed posthypoxic leukoencephalopathy (DPHL) is a rare and under recognised entity characterised by neurological relapse following a period of clinical stability or improvement after an episode of hypoxia. Patients usually present between 1 and 4 weeks after the initial event with relatively lucid intervening periods of variable lengths. The majority of DPHL cases reported to date have been associated with carbon monoxide intoxication. However, delayed neurological sequelae have also been documented in other causes of hypoxia, particularly due to respiratory failure in drug overdoses. All patients show diffuse, extensive and confluent white matter signal abnormalities including prominent restricted diffusion, extending to the subcortical white matter and respecting the U-fibres. There is no gyral oedema or contrast enhancement. The histopathology consists of patchy subcortical myelin loss with sparing of

(b)

Figure 12.17 (a) Non-perfused brain showing grey cortical discolouration. (b) Lysis of cerebellar granule cells in non-perfused brain. U-fibres and prominent macrophage/microglial inflammation with extensive axonal injury but relative axonal preservation in most cases (Zamora et al. 2015).

References Atkinson, J.L.D., Anderson, R.E. and Murray, M.H. 1998. The early critical phase of severe head injury: importance of apnea and dysfunctional respiration. J Trauma, 45, 941–945. Barzó, P., Marmarou, A. and Fatouros, P. et al. 1997. Contribution of vasogenic and cellular oedema to traumatic brain swelling. J Neurosurg, 87, 900–907. Black, P.M. 1978. Brain death. Parts 1 and 2. N Engl J Med, 299, 338–344 and 393–401. Bruce, D.A. 1984. Delayed deterioration of consciousness after trivial head injury in childhood. BMJ, 289, 715–716. Bruce, D.A., Alavi, A. and Bilaniuk, B. et al. 1981. Diffuse cerebral swelling following head injuries in children: the syndrome of ‘malignant brain edema’. J Neurosurg, 54, 170–178. Bullock, R., Maxwell, W.L., Graham, D.I., Teasdale, G.M. and Adams, J.H. 1994. Glial swelling following human cerebral contusion: an ultrastructural study. J Neurol Neurosurg Psychiatry, 54, 427–434. Byard, R.W., Bhatia, K.D. and Reilly, P.L. et al. 2009. How rapidly does cerebral swelling follow trauma? Observations using an animal model and possible implications in infancy. Legal Med, 11, S128–S131. Byard, R.W., Gabrielian, L. and Helps, S.C. et al. 2012. Further investigations into the speed of cerebral swelling following blunt cranial trauma. J Forensic Sci, 57, 973–975. Byard, R.W. and Fink, R. 2013. Speed of development of cerebral swelling following blunt cranial trauma. J Forensic Legal Med, 20, 598–600. Cernak, I., Vink, R. and Zappel, D.N. et al. 2004. The pathobiology of moderate diffuse traumatic brain injury as identified using a new experimental model of injury in rats. Neurobiol Dis, 17, 29–43.

137

138

Brain swelling, raised intracranial pressure and hypoxia-related brain injury Dada, M., Loftus, I. and Rutherfoord, G. 1993. Shotgun pellet embolism to the brain. Am J Forensic Med Pathol, 14, 58–60. Farin, A., Deutsch, R., Biegon, A. and Marshall, L.F. 2003. Sex-related differences in patients with severe head injury: greater susceptibility to brain swelling in female patients 50 years of age and younger. J Neurosurg, 98, 32–36. Fields, J.D., Lansberg, M.G. and Skirboll, S.L. et al. 2006. ‘Paradoxical’ transtentorial herniation due to CSF drainage in the presence of a hemicraniectomy. Neurology, 67, 1513. Graham, D.I., Ford, I. and Adams, J.H. et al. 1989. Fatal head injury in children. J Clin Pathol, 42, 18–22. Hassler, O. 1967. Arterial pattern of human brain stem. Normal appearance and deformation in expanding supratentorial conditions. Neurology, 17, 368–375. Inamasu, J., Miyatake, S., Suzuki, M., Nakatsukasa, M., Tomioka, H., Honda, M., Kase, K. and Kobayashi, K. 2010. Early CT signs in out of hospital cardiac arrest survivors: temporal profile and prognostic significance. Resuscitation, 81, 534–538. Itabashi, H.H., Andrews, J.M., Tomiyasu, U., Erlich, S.S. and Sathyavagiswaran, L. 2007. Responses of the central nervous system to acute hypoxicischaemic injury and related conditions. In Forensic Neuropathology – A  Practical Review of the Fundamentals. Burlington, MA: Elsevier Academic Press. Jortner, B.S. 2006. The return of the dark neuron. A histological artifact complicating contemporary neurotoxicologic evaluation. Neurotoxicology, 27, 628–34 Jung, H.J., Kim, D.M. and Kim, S.W. 2012. Paradoxical transtentorial herniation caused by lumbar puncture after decompressive craniotomy. J Korean Neurosurg Soc, 51, 102–104. Kalaria, R., Ferrer, I. and Love, S. 2015. Vascular disease, hypoxia and related conditions. In Love, S. Budka, H., Ironside, J.W. and Perry, A. (eds), Greenfields Neuropathology, 9th edn, Vol 1. CRC Press. Kamenar, E. and Burger, P.C. 1980. Cerebral fat embolism: a neuropathological study of a microembolic state. Stroke, 11, 477–484. Kitamura, O. 1994. Immunohistochemical investigation of hypoxic/­ ischemic brain damage in forensic autopsy cases. Int J Legal Med, 107, 69–76. Klatzo, I. 1967. Neuropathological aspects of brain edema. J Neuropathol Exp Neurol, 26, 1–14. Klintworth, G.K. 1966. Secondary brainstem haemorrhage. J Neurol Neurosurg Psychiatry, 29, 423–425. Kors, E.E., Terwindt, G.M. and Vermeulen, F.L. et al. 2001. Delayed cerebral edema and fatal coma after minor head trauma; role of the CACNA1A calcium channel subunit gene and relationship with familial hemiplegic migraine. Ann Neurol, 49, 753–760. Lang, D.A., Teasdale, G.M., Macpherson, P. and Lawrence, A. 1994. Diffuse brain swelling after head injury: more often malignant in adults than children? Journal of Neurosurgery. 80, 675–680. Leestma, J.E. 2014. Forensic aspects of adult general neuropathology. In Leestma, J.E., Forensic Neuropathology, 3rd edn. CRC Press.

Lynch, J.R., Pineda, J.A. and Morgan, D. et al. 2002. Apolipoprotein E affects the central nervous system response to injury and the development of cerebral edema. Ann Neurol, 51, 113–117. Márquez-Romero, J.M., Zermeño-Pohls, F. and Soto-Cabrera, E. 2010. Paradoxical herniation due to a continuous cerebrospinal fluid drain in a previously craniectomised patient. Neurologia, 25, 269–271. Miller-Fisher, C. 1995. Brain herniation: a revision of classical concepts. Can J Neurol Sci, 22, 83–91. Miyamoto, O. and Auer, R.N. 2000. Hypoxia, hyperoxia, ischemia, and brain necrosis. Neurology, 54, 362–371. Palmer, A.C., Calder, I.M. and Hughes, J.T. 1987. Spinal cord degeneration in divers. Lancet, ii, 1365–1366. Pulsinelli, W.A. 1997. Selective neuronal vulnerability and infarction in cerebrovascular disease. In Welch, K.M.A., Caplan, L.R., Reis, D.J., Siesjö, B.K. and Weir, B. (eds), Primer on Cerebrovascular Disease. Academic Press, pp 104–107. Sakas, D.E., Whittaker, K.W., Whitwell, H.L. and Singounas, E.G. 1997. Syndromes of post-traumatic neurological deterioration in children with no focal lesions revealed by cerebral imaging: evidence for a trigeminovascular pathophysiology. Neurosurgery, 41, 661–667. Scroop, S.R., Askridge, E.J. and Britz, G.W. 2002. Paradoxical cerebral arterial embolisation of cement during intraoperative vertebroplasty: case report. AJNR, 23, 868–870. Shapiro, H.A., Gordon, I. and Benson, S.D. 1988. Forensic Medicine – A Guide to Principles, 3rd edn. Churchill Livingstone. Siesjö, B.K. 1992. Pathophysiology and treatment of focal cerebral ischemia. Part II: mechanisms of damage and treatment. J Neurosurg, 77, 337–354. Snoek, J.W., Minderhoud, J.M. and Wilmink, J.T. 1984. Delayed deterioration following mild head injury in children. Brain, 107, 15–36. Stiver, S.I. 2009. Complications of decompressive craniectomy for traumatic brain injury. Neurosurg Focus, 26, E7. Tornheim, P.A. and McLaurin, R.L. 1981. Acute changes in regional brain water content following experimental closed head injury. J Neurosurg, 55, 407–413. Toro-Gonzalez, G., Navarro-Roman, L., Roman, G.C. et al. 1993. Acute ischemic stroke from fibrocartilaginous embolism to the middle cerebral artery. Stroke, 24, 738–740. Yoshino, E., Yamaki, T. and Higuchi, T. et al. 1985. Acute brain oedema in fatal head injury: analysis by dynamic CT scanning. J Neurosurg, 63, 830–839. Zamora, C.A., Nauen, D., Hynecek, R., Ilica, A.T., Izbudak, I., Sair, H.I., Gujar, S.K. and Pillai, J.J. 2015. Delayed posthypoxic leukoencephalopathy: a case series and review of the literature. Brain Behavior, 5, e000364. Źivković, V., Cvetković, D., Obradović, D. and Nikolić, S. 2020. Mechanism of brain swelling in cases of brain evisceration due to catastrophic craniocerebral injury – an autopsy study. Forensic Sci Med Pathol, 16, 107–112.

13

Sudden unexpected death in epilepsy Christopher Milroy and Daniel du Plessis

Definition of epilepsy and sudden unexpected death in epilepsy (SUDEP) Sudden unexpected death in patients with epilepsy (SUDEP) has been recognised since the first part of the 20th century (Jay and Leestma 1981). From the 1960s onwards, a number of papers on autopsy findings have been published (Krohn 1963; Hirsch and Martin 1971; Leestma et al. 1984). Epilepsy has been defined as a disorder of the brain characterised by an enduring predisposition to generate epileptic seizures and by neurobiological, cognitive, psychological and social consequences of this condition. The International League Against Epilepsy (IALE) produced a practical clinical definition of epilepsy in 2014 (Fisher et al. 2014). This states that epilepsy is a disease of the brain defined by the following conditions: 1. At least two unprovoked (or reflex) seizures occurring more than 24 hours apart. 2. One unprovoked seizure (or reflex seizure) and a probability of further seizures similar to the general recurrence risk (at least 60 per cent) after two unprovoked seizures, occurring over the next 10 years. 3. Diagnosis of an epileptic syndrome. By this definition of epilepsy, events such as a seizure after concussion (but not post-traumatic epilepsy), during alcohol withdrawal and seizure in fever do not lead to a diagnosis of epilepsy as they are provoked seizures. A number of definitions of what constitutes SUDEP have been made, but SUDEP can now be defined as: 1. The person had epilepsy. 2. Death was sudden and unexpected, whether witnessed or not. 3. The death was non-traumatic, not drowning and in benign circumstances. 4. Death occurred with or without seizure activity but excluding status epilepticus, i.e. seizure activity of more than 30 minutes. 5. No anatomical or pathological cause of death was identified after post-mortem examination. It should of course be noted that patients with epilepsy may die a sudden natural death unrelated to epilepsy.

Incidence The incidence of SUDEP is clearly going to depend upon accurate classification of deaths in patients with epilepsy and this will thus depend upon the thoroughness of death investigation, including how detailed the post-mortem examination is conducted. The rates in adults have varied. Leestma and colleagues recorded a mortality rate of 1.3–9.3 million per 1000 patient years (Leestma et al. 1997). Tomson and colleagues (2008) reported community-based rates of SUDEP of 0.35–2.3 per 1000

patient years with selected groups of having rates of 1.1–9.3 per 1000 patient years. In a study from the United Kingdom, it was estimated that death from SUDEP accounted for 0.1 per cent of all patients with epilepsy each year, thus resulting in 500 cases of SUDEP a year (Shankar et al. 2014). A comparative figure in the United States based upon the same rate would be 3000 cases per year.

Incidence of SUDEP in children The rate of SUDEP in children is lower than in adults (Milroy 2011; Morse and Kothare 2016). Donner and colleagues record a rate of 2/10,000 patient years (Donner et al. 2001). Rates of 1.1/10,000 patient years (Camfield and Camfield 2005), 4.3/10,000 years (Weber et al. 2005) and 3/10,000 (Geerts et al. 2010) have been reported. The highest rate was in those with remote symptomatic aetiology. In one series of 27 cases (Donner et al. 2001), 14 of her patients had symptomatic epilepsy, 5 cryptogenic epilepsy and 8 idiopathic epilepsy. Sudden death in epilepsy in childhood is rare in the absence of an underlying disorder sufficient to cause a functional neurological deficit (Camfield et al. 2002).

Mechanism of death in SUDEP While many deaths of patients with epilepsy remain unobserved, a few studies on observed cases have provided insight into what happens when these patients die, though there is still much that is not known. Studies of SUDEP have recorded cases with patients having witnessed seizure activity and seizure activity appears to be the most important factor in these deaths (Opeskin and Berkovic 2003). Poor seizure control increases the risk of sudden death and the prevention of death in patients with epilepsy includes advice to patients to be compliant with their medications to prevent seizures. Risk factors for sudden death in epilepsy include the frequency of seizures, with more seizures increasing the risk (So 2008). Similarly, patients on multiple anti-epileptic drugs have an increased risk, especially when subtherapeutic concentrations are present. Young adults and those of with a low intelligence quotient are at increased risk of death. Seizure activity affects both respiration and cardiac activity. Central-mediated apnoea has been observed in seizure activity. One of Weber’s cases was observed to gradually cease breathing without having convulsions (Weber et al. 2005). Many patients are found dead in bed face down (prone) and this has raised the possibility of an ‘asphyxial’ component. Pulmonary oedema is a very common finding and this appears to be neurogenic in origin. Postictal laryngospasm has also been proposed as a mechanism (So 2008). One hypothesis proposes that sudden death is mediated by postictal central respiratory depression, which could relate to underlying pathology in key respiratory nuclei and/or their neuromodulators. Alteration to neuronal populations in the medulla in SUDEP with evidence for greater reduction in neuromodulatory neuropeptidergic and monoaminergic

140

Sudden unexpected death in epilepsy systems, including for galanin and serotonin, could be a result of previous seizures and may represent a pathological risk factor for SUDEP through impaired respiratory homeostasis during a seizure (Patodia et al. 2018). Cardiac rhythm changes are increasingly recognised (Devinsky et al. 2016). The principal arrhythmia identified is bradycardia. While pre-ictal tachycardia has been observed, the evidence for malignant tachydysrhythmia as the terminal event is limited (Sevcencu and Struijk 2010). An analysis of patients with intractable epilepsy for a total of 220,000 patient hours with outpatient cardiac monitoring reported that 21 per cent had at least one bradycardic seizure (Rugg-Gunn et al. 2004). Most patients with ictal bradycardia or asystole had temporal lobe epilepsy but it may also be seen in frontal lobe epilepsy. In a 2013 study of monitored epilepsy patients with cardiorespiratory arrest, 11 cases of SUDEP were monitored, with 6 of these patients having an autopsy (Ryvlin et al. 2013). Eight patients had temporal lobe epilepsy. The longest documented seizure was 248 seconds. Out of 11 patients, 10 patients had what was termed early postictal neurodegenerative breakdown, which was defined as early, centrally mediated, severe alteration of both respiratory and cardiac function after generalised tonic-clonic seizures. In seven cases apnoea followed and then they went into asystole. These monitored patients showed four stages. The first was postictal rapid breathing between 18 and 50 breaths per minute (normal 12–18) with a heart rate of 55–145 beats a minute and a median rate of 90. Second, there was generalised postictal encephalography suppression. Third, early cardiorespiratory dysfunction developed in all patients during the first 3 minutes postictal. Lastly, terminal apnoeas always preceded asystole. Two patterns of evolution were noted. In some patients, the early cardiorespiratory arrest was terminal. In the other group, cardiorespiratory arrest spontaneously reversed after a median duration of 13 seconds of asystole. Restored respiration then progressively deteriorated until terminal apnoea and then terminal asystole.

(Pitkänen and Tamuna 2012). Some studies have not associated diffuse traumatic axonal injury (as an isolated diagnosis) with an increased risk of post-traumatic epilepsy (Scheid and von Cramon 2010).

Genetics of epilepsy There is an increasing realisation that genetic disorders may underlie epilepsy (Tu et al. 2011; Coll et al. 2016; Devinsky et al. 2016). Several genes linked to cardiac channelopathies are also associated with epilepsy. The two entities may overlap (Bagnall et al. 2016). Mutations in the SCN5A and KCNH2 genes have been associated with both long QT syndrome and epilepsy (Klassen et al. 2014). The KCNQ1 gene has been reported in a family with both epilepsy and long QT syndrome (Tiron et al. 2015).

The autopsy in SUDEP The autopsy, as it does in any investigation, begins with the history and scene information. The information that should be provided includes the medical history (specifically the epilepsy history), medications, including anti-epileptic medication, scene findings, including the position found in, whether witnesses saw the victim have a seizure or status epilepticus and whether in a bath with the head underwater or not. A full autopsy should be conducted. Often there will be no specific findings, but signs of seizure activity may be present, including bite marks on the tongue and a voided bladder. Their absence does not exclude seizure activity and these can only be considered ‘soft’ signs. Seizure activity may be a terminal event in non-epilepsy patients. Pulmonary oedema is a typical finding in SUDEP and there may be froth in the airway and at the mouth, with a mushroom plume present (champignon de mousse) (Figure 13.1). However, pulmonary oedema is such a common finding that it has little if any discriminatory value.

Post-traumatic epilepsy Sudden unexpected death following a previous head injury potentially introduces the possibility of death linked to the head injury if there was a history of post-traumatic seizures. This might have medico-legal implications, including the possibility of criminal proceedings should the original head injury have resulted from an assault. In such cases, it will be necessary to (1) confirm that epilepsy was due to the head injury, i.e. truly post-traumatic and linked to a specific injury event and (2) that a seizure-related death, including SUDEP, could be arrived at with confidence given the exclusion of other plausible competing causes of death. When attributing epilepsy to a previous head injury, it may be prudent to obtain the opinion of a neurologist/epileptologist aided by a thorough review of the clinical/historical background. Risk factors for the development of post-traumatic epilepsy should also be borne in mind. These include old age, penetrating brain injuries, a Glasgow Coma Score of less than 10, biparietal or multiple contusions, intracranial haemorrhage, frontal or temporal location of lesional pathology, greater than 5 mm midline shift, duration of coma in excess of 24 hours, loss of consciousness for more than 24 hours, prolonged posttraumatic amnesia, multiple intracranial neurosurgical procedures and the occurrence of early post-traumatic seizures

Figure 13.1 Mushroom plume present (champignon de mousse).

The neuropathology of SUDEP Other pathology, particularly cardiac pathology, may be present that accounts for apparent seizure activity but was in fact cardiac mediated and not true epilepsy. This may be particularly important in patients who have had genetic dysrhythmic disorders that have been diagnosed clinically as epilepsy, but which are in fact cardiac in origin, such as arrhythmogenic cardiomyopathy or channelopathies such as long QT syndrome. Apart from the usual macroscopic examination and sampling for microscopic examination, body fluids for toxicology, typically blood and urine if present, should be collected and submitted for analysis. Toxicology analysis in suspected SUDEP will depend upon the history and scene findings. Many death investigations systems conduct routine toxicology on all deaths, including recreational drugs such as stimulants and opioids; others do more selective testing. Analysis for anti-epileptic agents can be conducted, though not all laboratories will be able to test for all such drugs. In addition to the general problem of correlating post-mortem concentrations with concentrations at the time of death, there are problems determining what is the therapeutic concentration of an anti-epileptic medication, because of the wide therapeutic range seen clinically. However, the presence of apparent subtherapeutic concentration or no significantly detected amount may be important as non-compliance increases the risk of seizure. Because of the increasing recognition of genetic factors in both epilepsy and cardiac disorders, material for DNA extraction and genetic testing could be stored or sent. Genetic testing is now widely available and much less costly. DNA is best obtained from small samples of myocardium, spleen or kidney.

The neuropathology of SUDEP The reader is referred to the Royal College of Pathologists website for comprehensive guidelines on autopsy practice in deaths in patients with epilepsy, including sudden death (Thom and Allinson 2019). Coronal slices through the forebrain at the level of the amygdala and of the hippocampus at the level of the lateral geniculate body may provide sufficient multiple sample sites for assessment (frontal watershed cortex, insular cortex/ basal ganglia, amygdala, hippocampus, thalamus and temporal cortex) in addition to cerebellar and medullary samples. Tissue sampling may also be guided by clinical information of seizure localisation obtained in life. Samples from both left and right hemispheres should be taken where possible as epilepsy can be lateralised. Multiple samples of the hippocampi at multiple levels (anterior and posterior) may be beneficial. Fixation of the brain prior to examination is more likely to identify pathology relevant to the cause of death or epilepsy. Macroscopic and microscopic examination of the brain should address (1) the structural cause of epilepsy, (2) the effects of previous/recent seizures and (3) any unsuspected (unrelated) cause of death. Mild brain swelling without significant mass effect is not uncommonly identified in cases of SUDEP. Common lesions identified in SUDEP and epilepsy autopsy series include hippocampal sclerosis, cortical malformations (e.g. focal cortical dysplasia), vascular malformations such as cavernomas, brain tumours (primary and secondary) and old traumatic brain injuries. The spectrum of pathology identified in cases of SUDEP is comparable to that encountered in surgical epilepsy series with no over-representation of any single pathology type (Thom et al. 2018). Focal neuropathological abnormalities are found in approximately 50 per cent of SUDEP autopsies, ranging from 34 to 89 per cent in different series (Thom 2019).

Acute neuronal injury related to seizure activity (more likely in status epilepticus) or consequent to a period of survival following resuscitation may be limited to the hippocampus or also affect cerebral neocortex and other grey matter structures. Those with a long-term history of epilepsy may show evidence of cerebellar and/or thalamic atrophy with gliosis. Seizures may result in cortical atrophy and scarring. Weighing of the cerebellum may be beneficial (the hindbrain weight should represent 12–15 per cent of the total) (Thom 1997; Thom and Allinson 2019). MRI studies in SUDEP highlighted hippocampal volume asymmetries and atrophic changes in the brainstem. Focal neuronal loss and gliosis have been described within the amygdaloid subnuclei in SUDEP and it is thought that seizures spreading to the amygdala (which is functionally connected to the medulla) may cause cessation of spontaneous breathing. So far, no clear immunohistochemical signature for SUDEP in the human brain has been identified utilising a panel of antibodies to investigate inflammation, blood-brain barrier leakage, gliosis and acute neuronal injury in the hippocampus, amygdala and medulla (Michalak et al. 2017). There is no single neuropathological feature that can categorically confirm that seizures occurred during life, although some patterns of original neuronal damage and reorganisation such as mossy fibre sprouting are relatively specific for seizurerelated brain injury over time. Alterations in the granule cell composition/distribution in the dentate gyrus (such as dispersion or broadening of the granule cell layer) and reorganisation or sprouting of axons in the granule cells (mossy fibres, readily demonstrated by immunohistochemistry, including zinc transporter protein 3 [Figure 13.2]) can aid in the distinction of hippocampal damage due to epilepsy from other causes of the hippocampal neuronal loss such as due to hypoxic-ischaemic injury and neurodegenerative disease (Thom 2019). There is no evidence that hippocampal sclerosis is more common in SUDEP nor evidence of increased inflammation, including lymphocyte-mediated inflammation and microglial activation in SUDEP, compared to control groups. Acute neuronal injury seen as eosinophilic neuronal change is reported in approximately

Figure 13.2 Hippocampal sclerosis demonstrating mossy fibre sprouting as shown with zinc transporter 3 immunostaining. (Courtesy of Prof Maria Thom.)

141

142

Sudden unexpected death in epilepsy half of the cases of SUDEP subject to a post-mortem. The demonstration of neuronal HSP-70, HIF-1α and c-jun expression can be seen in such neurons in the CA1/subiculum region. Whilst such evidence of acute neuronal injury could reflect recent seizure-related neuronal hippocampal injury, similar changes could be explained by hypoxic-ischaemic events and/or excitotoxic cellular stresses and it can therefore not be regarded as reliable evidence of recent seizure activity (Thom 2019).

Sudden death, febrile seizures and hippocampal and temporal lobe maldevelopment in young children Sudden unexplained death in childhood (SUDC) is defined as the sudden and unexpected death in a child older than 1 year that remains unexplained after a complete autopsy, death scene investigation and ancillary testing (Kinney et al. 2012). Population-based studies subsequently reported a significant association amongst febrile seizures, hippocampal maldevelopment and sudden death. This led to the proposal that this SUDC subset represented a new entity, termed ‘hippocampal maldevelopment associated with sudden death’ (HMASD). HMASD was subsequently defined as focal granular cell bilamination (FGCB) in the dentate gyrus (Figure 13.3) with or without hippocampal asymmetry and/or malrotation (Kinney et al. 2009). Cases of a personal and/or family history of febrile seizures, but without HMASD, were classified as sudden unexplained death in childhood with febrile seizure phenotype (SUDC-FS) and the remaining cases, without HMASD, febrile seizures or an explained cause of death were classified as SUDC (Kinney et al. 2012). The explained group consisted of children in whom known/explained causes of sudden death were found such as infections or cardiac abnormalities (Hefti et al. 2016). The key morphological features of HMASD were further refined with the identification of significantly increased frequencies of other developmental abnormalities (Hefti et al. 2016). Reliable assessment of these abnormalities requires hippocampal sections orientated properly in the coronal plane with identifiable adjacent anatomical landmarks allowing for comparison with normative micro-atlases of the hippocampus to account for the well-recognised variation in cellular

Figure 13.4 Malrotation of the hippocampus.

architecture of the human hippocampus throughout its length. Sections for microscopy should be 6–10 µm thick. Features additional to FGCB, hippocampal malrotation (Figure 13.4) and hippocampal asymmetry found statistically more frequent in HMASD are (1) single ectopic cells in the molecular layer, (2) diffuse (widespread) dispersion of the dentate gyrus, (3) ectopic granule cells in the hilus, (4) irregular configuration of the dentate gyrus, (5) focal lack of granule cells in the dentate gyrus unaccompanied by gliosis and without an associated penetrating blood vessel, (6) clusters of ectopic granule cells in the molecular layer, (7) hyperconvolution of the dentate gyrus, (8) fluctuating thickness (caterpillarshaped) dentate gyrus, (9) granule cell heterotopias and (10) hilar gliosis. Neuronal loss and/or gliosis in CA1-3 is not significantly associated with HMASD. Only minor neuropathological abnormalities (red neuronal change, focal infiltrates of chronic inflammatory cells consistent with lymphocytes in the leptomeninges, focal subarachnoid haemorrhage, cerebral white matter gliosis, brainstem gliosis and periventricular leukomalacia) are observed outside of the temporal lobe in HMASD cases (Kinney et al. 2015; Hefti et al. 2016). Hefti and colleagues (2016) have found HMASD in almost 50 per cent of cases of SUDC in which hippocampal sections were available. The HMASD phenotype was characterised by the sudden, unexpected and sleep-related death of a young child with or without a personal and/or family history of febrile seizures and being discovered in a prone, face down sleep position. Approximately, half of the deaths were further associated with a fever around the time of death. In HMASD candidate, triggering events for a potential unwitnessed seizure during sleep are thought to be sleep itself, fever, an unwitnessed febrile seizure and/or prone sleep, the latter raising the possibility of hypoxaemia decreasing the seizure threshold. Common among HMASD, SUDC-FS and SUDC is a focal lack of granule cells in the dentate gyrus and ectopic single or clusters of granule cells in the dentate molecular layer and hilus suggesting a potential ‘core’ lesion among the three entities (Hefti et al. 2016).

Sudden unexpected death related to medullary brain lesions

Figure 13.3 Case of SUDC – bilamination of dentate gyrus with background prominence of pseudo-‘rod cell’ neuronal precursor cells.

In some cases of sudden unexpected death in all age groups, post-mortem examination will reveal no cause of death expect for the unexpected presence of a medullary brain lesion. Many such victims were considered healthy and neurologically normal at the time of sudden death, whilst others may recently have

References been clinically affected by a known lesion such as stroke, but their symptoms were mild and non-disabling. The spectrum of neuropathology is quite wide involving infections, infarcts, multiple sclerosis, neurosarcoidosis, congenital hypoplasia and neuronal immaturity of medullary nuclei and primary tumours. In most cases, the major motor and sensory pathways were grossly preserved, potentially explaining the absence of any related preceding symptoms. The medullary lesions underlying sudden unexpected death were often not visible on gross inspection at the time of brain sectioning and it has been suggested that in cases of sudden, otherwise unexplained death, at least three specimens of the brainstem be submitted for histological examination, a pontomesencephalic junction specimen, including the upper third of the pons, a second section extending from the upper third of the medulla to the adjacent portion of the pons and a third specimen taken at the level of the obex, including tissue above and below it (Jaster et al. 2008).

References Bagnall, R.D., Crompton, D.E., Petrovski, S., Lam, L., Cutmore, C., Garry, S.I., Sadleir, L.G., Dibbens, L.M., Cairns, A., Kivity, S. and Afawi, Z. 2016. Exome-based analysis of cardiac arrhythmia, respiratory control, and epilepsy genes in sudden unexpected death in epilepsy. Ann Neurol, 79, 522–534. Camfield, C.S., Camfield, P.R. and Veugelers, P.J. 2002. Death in children with epilepsy: a population-based study. Lancet, 359, 1891–1895. Camfield, P. and Camfield, C. 2005. Sudden unexpected death in people with epilepsy: a pediatric perspective. Semin Pediatr Neurol, 12, 10–14. Coll, M., Allegue, C., Partemi, S., Mates, J., Del Olmo, B., Campuzano, O., Pascali, V., Iglesias, A., Striano, P., Oliva, A. and Brugada, R. 2016. Genetic investigation of sudden unexpected death in epilepsy cohort by panel target resequencing. Int J Legal Med, 130, 331–339. Devinsky, O., Hesdorffer, D.C., Thurman, D.J., Lhatoo, S. and Richerson, G. 2016. Sudden unexpected death in epilepsy: epidemiology, mechanisms, and prevention. Lancet Neurol, 15, 1075–1088. Donner, E.J., Smith, C.R. and Snead, O.C. 2001. Sudden unexplained death in children with epilepsy. Neurology, 57, 430–434. Fisher, R.S., Acevedo, C., Arzimanoglou, A., Bogacz, A., Cross, J.H., Elger, C.E., Engel, J. Jr, Forsgren, L., French, J.A., Glynn, M. and Hesdorffer, D.C. 2014. ILAE official report: a practical clinical definition of epilepsy. Epilepsia, 55, 475–482. Geerts, A., Arts, W.F., Stroink, H., Peeters, E., Brouwer, O., Peters, B., Laan, L. and Van Donselaar, C. 2010. Course and outcome of childhood epilepsy: a 15-year follow-up of the Dutch Study of Epilepsy in Childhood. Epilepsia, 51, 1189–1197. Hefti, M.M., Kinney, H.C., Cryan, J.B., Haas, E.A., Chadwick, A.E. and Crandall, L.A. et al. 2016. Sudden unexpected death in early childhood: general observations in a series of 151 cases. Forensic Sci Med Pathol, 12, 14–25. Hirsch, C.S. and Martin, D.L. 1971. Unexpected death in young epileptics. Neurology, 21, 682–682. Jaster, J.H., Ottaviani, G. and Matturri, L. et al. 2008. Sudden unexpected death related to medullary brain lesions. Am J Forensic Med Pathol, 29, 371–374. Jay, G.W. and Leestma, J.E. 1981. Sudden death in epilepsy. A comprehensive review of all the literature and proposed mechanisms. Acta Neurol Scand Suppl, 81, 1–66. Kinney, H.C., Chadwick, A.E., Crandall, L.A., Grafe, M., Armstrong, D.L. and Kupsky, W.J. et al. 2009. Sudden death, febrile seizures, and ­hippocampal and temporal lobe maldevelopment in toddlers: a new entity. Pediatr Dev Pathol, 12, 455–463.

Kinney, H.C., Rognum, T.O., Nattie, E.E., Haddad, G.G., Hyma, B. and McEntire, B. et al. 2012. Sudden and unexpected death in early life: proceedings of a symposium in honor of Dr. Henry F. Krous. Forensic Sci Med Pathol, 8, 414–425. Kinney, H.C., Cryan, J.B., Haynes, R.L., Paterson, D.S., Haas, E.A. and Mena, O.J. et al. 2015. Dentate gyrus abnormalities in sudden unexplained death in infants: morphological marker of underlying brain vulnerability. Acta Neuropathol, 129, 65–80. Klassen, T.L., Bomben, V.C., Patel, A., Drabek, J., Chen, T.T., Gu, W., Zhang, F., Chapman, K., Lupski, J.R., Noebels, J.L. and Goldman, A.M. 2014. Highresolution molecular genomic autopsy reveals complex sudden unexpected death in epilepsy risk profile. Epilepsia, 55, e6–e12. Krohn, W. 1963. Causes of death among epileptics. Epilepsia, 4, 315–321. Leestma, J.E., Kalelkar, M.B., Teas, S.S., Jay, G.W. and Hughes, J.R. 1984. Sudden unexpected death associated with seizures: analysis of 66 cases. Epilepsia, 25, 84–88. Michalak, Z., Obari, D., Ellis, M., Thom, M. and Sisodiya, S.M. 2017. Neuropathology of SUDEP. The role of inflammation, blood-brain barrier impairment, and hypoxia. Neurology, 88, 551–561. Milroy, C.M. 2011. Sudden unexpected death in epilepsy in childhood. Forensic Sci Med Pathol, 7, 336–340. Morse, A.M. and Kothare, S.V. 2016. Pediatric sudden unexpected death in epilepsy. Pediatr Neurol, 57, 7–16. Opeskin, K. and Berkovic, S.F. 2003. Risk factors for sudden unexpected death in epilepsy: a controlled prospective study based on coroners cases. Seizure, 12, 456–464. Patodia, S., Somani, A., O’Hare, M., Venkateswaran, R., Liu, J., Michalak, Z., Ellis, M., Scheffer, I.E., Diehl, B., Sisodiya, S.M. and Thom, M. 2018. The ventrolateral medulla and medullary raphe in sudden unexpected death in epilepsy. Brain, 141, 1719–1733. Pitkänen, A. and Tamuna, B. 2012. Head trauma and epilepsy. In Noebels, J.L., Avoli, M. and Rogawski, M.A. et al. (eds). Jasper’s Basic Mechanisms of the Epilepsies, 4th edn. National Center for Biotechnology Information (US). Rugg-Gunn, F.J., Simister, R.J., Squirrell, M., Holdright, D.R. and Duncan, J.S. 2004. Cardiac arrhythmias in focal epilepsy: a prospective long-term study. Lancet, 364, 2212–2219. Ryvlin, P., Nashef, L., Lhatoo, S.D., Bateman, L.M., Bird, J., Bleasel, A., Boon, P., Crespel, A., Dworetzky, B.A., Høgenhaven, H. and Lerche, H. 2013. Incidence and mechanisms of cardiorespiratory arrests in epilepsy monitoring units (MORTEMUS): a retrospective study. Lancet Neurol, 12, 966–977. Scheid, R. and von Cramon, D.Y. 2010. Clinical findings in the chronic phase of traumatic brain injury: data from 12 years’ experience in the Cognitive Outpatient Clinic at the University of Leipzig. Dtsch Arztebl Int, 107, 199–205. Sevcencu, C. and Struijk, J.J. 2010. Autonomic alterations and cardiac changes in epilepsy. Epilepsia, 51, 725–737. Shankar, R., Jalihal, V., Walker, M., Laugharne, R., McLean, B., Carlyon, E., Hanna, J., Brown, S., Jory, C., Tripp, M. and Pace, A. 2014. A community study in Cornwall UK of sudden unexpected death in epilepsy (SUDEP) in a 9-year population sample. Seizure, 23, 382–385. So, E.L. 2008. What is known about the mechanisms underlying SUDEP? Epilepsia, 49, 93–98. Thom, M. 1997. Neuropathological findings in post-mortem studies of sudden death in epilepsy. Epilepsia, 38, S32–S34. Thom, M., Boldrini, M., Bundock, E., Sheppard, M.N. and Devinsky, O. 2018. Review: the past, present and future challenges in epilepsy-related and sudden deaths and biobanking. Neuropathol Appl Neurobiol, 44, 32–55. Thom, M. and Allinson, K. 2019. Guidelines on Autopsy Practise: Deaths in Patients with Epilepsy Including Sudden Death. The Royal College of Pathologists. Available on-line on College website. Thom, M. 2019. Neuropathology of epilepsy: epilepsy-related deaths and SUDEP. Diagn Histopathol, 25, 23–33.

143

144

Sudden unexpected death in epilepsy Tiron, C., Campuzano, O., Pérez-Serra, A., Mademont, I., Coll, M., Allegue, C., Iglesias, A., Partemi, S., Striano, P., Oliva, A. and Brugada, R. 2015. Further evidence of the association between LQT syndrome and epilepsy in a family with KCNQ1 pathogenic variant. Seizure, 25, 65–67. Tomson, T., Nashef, L. and Ryvlin, P. 2008. Sudden unexpected death in epilepsy: current knowledge and future directions. Lancet Neurol, 7, 1021–1031.

Tu, E., Bagnall, R.D., Duflou, J. and Semsarian, C. 2011. Post-mortem review and genetic analysis of sudden unexpected death in epilepsy (SUDEP) cases. Brain Pathol, 21, 201–208. Weber, P., Bubl, R., Blauenstein, U., Tillmann, B.U. and Lütschg, J. 2005. Sudden unexplained death in children with epilepsy: a cohort study with an eighteen-year follow-up. Acta Paediatr, 94, 564–567.

14

Contact sport and blast-related neuropathology Daniel du Plessis and Christopher Milroy

Sport injuries are leading causes of traumatic brain and spinal injuries, and concerns have been expressed about the development of chronic brain injury such as chronic traumatic encephalopathy (CTE) in sports participants, both professional and amateur. CTE is now also recognised as a complication in military personnel who have survived a blast injury. Second impact syndrome (SIS), a controversial entity, is suggested to occur following a second injury giving rise to cerebral swelling which may lead to death (Bey and Ostick 2009).

Death as a result of acute injury in boxers This is rare but well recognised (Ross and Ochsner 1999). The most common cause is acute intracranial haemorrhage, most commonly acute subdural haematoma. Traumatic subarachnoid haemorrhage is rare in boxing, probably as a result of the protective influence of well-developed neck muscles and lack of alcohol intoxication in the fighters. Traumatic subarachnoid haemorrhage has been suggested as the mechanism of death in a historic case (Plant and Butt 1993).

Other sports Head and neck injuries are seen in a variety of sporting activities. In general, unlike the issues relating to boxing, the injuries themselves are non-specific and present the picture of blunt injury. Golf and horse riding were identified as the most common cause of head injuries requiring hospital stay (Lindsay et al. 1980). Golf injuries may be from impact with the ball or club. Skiing injuries commonly result from a fall followed but also by collision with an object (Myles et al. 1992). Sailing accidents may cause severe head injury when the boom swings across. Blunt head injury in equestrian accidents may arise from falling or contact with the horse or other object such as a branch. Occasionally, horseshoe abrasions (which may be incomplete) may be identified (Lee 2000).

Chronic traumatic encephalopathy CTE was first described in boxing, a sport that has raised controversy over many years, in relation not only to acute injury, but also to the long-term chronic effects (Millspaugh 1937). In addition, there is now considerable literature concerning repetitive head injury in other sports such as Association football (soccer), American football (gridiron), rugby and ice hockey (Kirkendall et al. 2001; Rabadi and Jordan 2001; Marchie and Cusimano 2003; Guskiewicz et al. 2003). In the 1920s, dementia pugilistica commonly called the ‘punch drunk’ syndrome was described in both amateur and professional boxers (Martland 1928). It develops in boxers who have competed in many bouts over a long period of time. Intellectual deterioration develops in association with Parkinson-type symptoms, as well ataxia. In a large

neuropathological series of ex-boxers (Corsellis et al. 1973), characteristic macroscopic features were identified: • Enlargement of the cavum of the septum pellucidum and fenestration of its leaves and thinning of the fornices and corpus callosum (Figure 14.1). This was seen in 77 per cent of cases. • Cerebellar atrophy in relation to the folia adjacent to the foramen magnum with gliosis and Purkinje cell loss. This is thought to result from repetitive transient tonsillar herniation. • Depigmentation with neuronal loss of the substantia nigra and locus coeruleus with neurofibrillary tangles (NFT). Lewy bodies are not a feature. • Cortical pathology with NFT identified on silver staining in the absence of senile plaques was described by Corsellis and colleagues (Figure 14.2). This is most prominent in the medial temporal cortex but also seen elsewhere in the cerebral cortex and brainstem. Dementia pugilistica is now considered as part of the CTE spectrum. However, it has been pointed out that these cases represented prolonged exposure to traumatic brain injury (TBI) with extreme forces delivered in often mismatched boxers who underwent repeated fights and sparring without significant medical supervision. As such typical dementia pugilistica is not now encountered in the way it was in the past (Castellani and Perry 2017). The term CTE was first used for brain changes in the 1940s (Critchley 1949). When Roberts et al. (1990) re-examined the cases from Corsellis’s series with immunohistochemistry, they identified that nearly all the cases had extensive beta amyloid like that seen in Alzheimer’s disease. In 1990, Geddes described neuropathological changes consisting of neuropil threads sited predominantly around small intracortical vessels, often found at the depths of cortical sulci and tau positive NFT in five men including a soccer player and two boxers (Geddes et al. 1999). In 1996, Williams and Tannenberg reported related neuropathology in an alcoholic achondroplastic dwarf who had worked in a circus and ‘dwarf throwing’ competitions and had been knocked out on multiple occasions (Williams and Tannenberg 1996). In 2005 and 2006, Omalu and colleagues reported CTE in retired American Footballers (Omalu et al. 2005, 2006). In 2009, McKee et al. (2009) identified 48 cases in the literature. By 2014, Gardner and colleagues had identified over 150 cases, including cases related to physical abuse, head-banging, poorly controlled epilepsy and rugby. Montenigro and colleagues (2014) have proposed the name traumatic encephalopathy syndrome (TES) and found over 200 cases in the literature. These were mostly reported in boxers and American footballers, as well as ice hockey players and professional wrestlers. Montenigro and colleagues (2014) proposed four subtypes of TES (i) TES behavioural/mood variant, (ii) TES cognitive variant, (iii) TES mixed variant (behavioural/mood and cognitive features) and (iv) TES dementia.

146

Contact sport and blast-related neuropathology

Neuropathology of CTE Macroscopically there may be no obvious changes (McKee et al. 2015). In the early stages, there may be mild enlargement in the frontal and temporal horns of the lateral ventricles. Perivascular spaces may be prominent in the white matter, most significantly in the temporal lobes. As CTE advances, there is atrophy and brain weight is decreased. The most significant areas affected are grey and white matter in the frontal and anterior temporal lobes. There can be enlargement of the lateral and third ventricles and there may be a cavum septum pellucidum with septal fenestrations. The thalamus is atrophied along with the mammillary bodies and hypothalamus. The substantia nigra and locus coeruleus can become depigmented. The cerebellum is typically normal macroscopically. McKee and colleagues (2013, 2015) have identified four stages based upon macroscopy:

Figure 14.1 Cerebral atrophy, enlargement of the ventricular system and a wide cavum with torn septal walls. (Reproduced from Adams, J.H. 1992. Head injury. In Adams, J.H. and Duchern, L.W. (eds), Greenfield’s Neuropathology, 5th edn. Edward Arnold.)

Clinical features A number of clinical features have been reported in association with CTE (Iverson et al. 2018). These include memory impairment, attention and concentration difficulties, language impairment and visual-spatial difficulties. Behavioural features include difficulties with impulse control and aggression, disinhibition and socially inappropriate behaviour, and paranoia. There may be mood changes, depression, anxiety, apathy and suicidal ideation. Gait disorders, speech slowing and extrapyramidal signs may be present (Gardner et al. 2014). Suicide in patients with CTE has become an increasing concern, but a study of retired NFL American footballers did not indicate a higher suicide risk; indeed, they had a lower risk (Baron et al. 2012). Neuropathologists may be asked to evaluate a brain for CTE, especially in former sportsmen and military personnel, who have committed suicide.

Figure 14.2 Neurofibrillary tangles in a case of longstanding damage caused by boxing (Bielchowsky silver stain).

1. Macroscopically normal brains. Microscopically there are isolated perivascular focal epicentres of p-tau NFT. The tau pathology is most commonly localised to the depths of the cerebral sulci of the frontal, temporal, insular, septal and parietal cortices. Approximately 50 per cent of cases at this stage show rare TDP-43 neurites. Reactive microglia are found and perivascualar ­haemosiderin-laden macrophages are seen in the white matter. Aβ plaques are only seen in those over 50 years of age. 2. Subtle macroscopic changes may be present including mild enlargement of the frontal horns of the lateral ventricles and third ventricle, cavum septum pellucidum and pallor of the locus coeruleus and substantia nigra. Microscopically, multiple epicentres of perivascular foci of p-tau NFT and neurites are found in the depths of the frontal, temporal, parietal, insular and septal cortices as well as in the loci cerulei and substantitia nigra. Mild TDP-43 related pathology may be present. Reactive microglia are present. Aβ plaques are only seen in those over 50 years of age and are present in 19 per cent. 3. Brains are typically macroscopically abnormal, though more dramatic changes are seen microscopically. These changes include loss of brain weight, mild frontal and temporal atrophy with enlargement of the lateral and third ventricles. Septal abnormalities including cavum septum pellucidum, septal perforations or fenestrations are present in one-half of subjects. Characteristically there is pallor of the locus coeruleus and substantia nigra. Atrophy of the mammillary bodies, thalamus and hypothalamus and thinning of the corpus callosum are typically present. On microscopic examination NFTs are present in confluent patches. Neuropil threads and astrocytic tangles are found around blood vessels in the sulcal depths and as linear arrays in the superficial laminae of the cortex. NFT are widely present including in the superior, dorsolateral and inferior frontal, septal, insular, temporal pole, superior, middle and inferior temporal and inferior parietal cortices. NFT are also found in the olfactory bulbs, hippocampus, entorhinal cortex, amygdala, hypothalamus, mammillary bodies, nucleus basalis of Meynert, substantia nigra, dorsal and median raphe nuclei and locus coeruleus. NFTs may be present in the dentate nucleus of the cerebellum and the spinal cord grey matter. TDP-43-positive neurites and neuronal inclusions are found in the majority of cases. Thirteen

Blast injury per cent of stage III CTE cases show Aβ plaques. White matter abnormalities, loss of myelinated nerve fibres, axonal dystrophy and axonal loss can be conspicuous. 4. Brains show a decrease in weight and can be less than 1000 g. There is significant atrophy of the frontal and temporal lobes, medial temporal lobe and anterior thalamus. Diffuse atrophy of the white matter and thinning of the corpus callosum, particularly the isthmus is present. There is typically severe thinning of the hypothalamic floor and atrophy of the mammillary bodies. Most cases show septal abnormalities and pallor in the locus coeruleus and substantia nigra. Microscopically, there is widespread myelin loss, astrocytosis of the white matter and neuronal loss in the cerebral cortex, hippocampus and substantia nigra. Neuronal loss and astrocytosis may also be prominent in the frontal and temporal cortices and there may be microvacuolation of layer II. P-tau pathology is densely distributed throughout the cerebrum, thalamus, hypothalamus, mammillary bodies, basal ganglia, brainstem, cerebellar dentate nucleus and occasionally, spinal cord. NFT are found throughout the hippocampi. P-tau pathology is also seen extensively in the cerebellum. Widespread TDP-43 deposits are also found in neurites and as intraneuronal cytoplasmic inclusions. There is marked loss of myelinated nerve fibres, axonal dystrophy and loss in the white matter. Preliminary NINDS criteria for the pathological diagnosis for CTE have been published by McKee et al. (2016). In the view of this review group, the p-tau pathology of CTE is clearly distinct from other tauopathies. The pathognomonic lesion required for the diagnosis of CTE consists of p-tau aggregates in neurons, astrocytes and cell processes around small vessels in an irregular pattern at the depths of cortical sulci (Figure 14.3). It is recommended to follow NIA-AA sampling protocols for the evaluation of Alzheimer’s disease and Lewy body disease, supplemented by extra sections of frontal and temporal cortices (specifically superior frontal gyrus and temporal pole) and hypothalamus including the mammillary bodies (McKee et al. 2016). Bilateral representative sections from each region are recommended. Sites most valuable for detecting CTE neuropathology were

Figure 14.3 CTE typified by tau-immunopositive aggregates in neurons, astrocytes and cell processes around small vessels in an irregular pattern at the depths of cortical sulci.

identified as middle frontal gyrus, superior and middle temporal gyri and the inferior parietal lobule. The group also defined supportive but non-specific p-tau-immunoreactive features of CTE such as pretangles and NFTs affecting superficial layers (layers II–III) of cerebral cortex; pretangles, NFTs or extracellular tangles in CA2 and pretangles and proximal dendritic swellings in CA4 of the hippocampus; neuronal and astrocytic aggregates in subcortical nuclei; thorn-shaped astrocytes at the glial limitans of the subpial and periventricular regions; and large grain-like and dot-like structures. Supportive non-p-tau pathologies include TDP-43 immunoreactive neuronal cytoplasmic inclusions and dot-like structures in the hippocampus, anteromedial temporal cortex and amygdala. Macroscopic features such as disproportionate dilatation of the third ventricle, septal abnormalities, mammillary body atrophy and old traumatic injuries such as old contusions are also regarded as supportive evidence of CTE. Some caution has been proposed about the role of a tauopathy in CTE (Castellani 2015; Castellani et al. 2015) as clinicopathological correlation is poor and standard methodology is lacking (Castellani and Perry 2017). Historical and recent evidence suggests that CTE, as it is presented in the literature, might not be pathologically or clinically progressive in a substantial percentage of people. At present, it is not known whether the emergence, course, or severity of clinical symptoms, can be predicted by specific combinations of neuropathologies, thresholds for accumulation of pathology, or regional distributions of pathologies (Iverson et al. 2018).

CTE and other neurodegenerative disorders CTE is associated with other neurological disorders (McKee et al. 2010, 2013). Approximately 10 per cent of people with CTE develop features of motor neuron disease that is indistinguishable from amyotropic lateral sclerosis. CTE has also been seen in association with Alzheimer’s disease, Lewy body disease and frontotemporal lobar degeneration.

Blast injury Blast injury results in the rapid transmission of acoustic waves through the brain. It has been recognised for over 100 years that exposure to blast injury is associated with neurological disorders (Shively and Perl 2017; Kinch et al. 2019). In the First World War of 1914–1918, many servicemen were diagnosed with a neurological disorder called neurasthemia and ‘shell shock’, but correlative neuropathological studies were lacking. It is reported that 8.4 per cent of the US military serving in the Afghanistan and Iraq wars suffered TBI, most of these caused by blast injury (DePalma and Hoffman 2018). CTE has been described in military veterans (Goldstein et al. 2012; McKee and Robinson 2014). In contrast, a preliminary study of brain specimens of military personnel, including 43 with a history of TBI and three with a history of blast exposure did not identify changes associated with CTE (Tripathy et al. 2019). A history of non-blast-related head trauma in some cases further confounds establishing a causal relationship between blast exposure and CTE (Shively and Perl 2017). Single-blast exposure has resulted in injury in animal models with evidence of tau-related pathology. While sports-related trauma is relatively predictable, blast injury represents heterogenous circumstances, as injury may occur in closed or open environments such as buildings and

147

148

Contact sport and blast-related neuropathology vehicles or in open areas. The explosive charge will also vary as these are often improvised explosive devices. The blast may represent a single incident or repeated exposure to blast injury. Shively and Perl (2017) examined autopsy material from five cases with chronic blast TBI, three cases of acute blast injury, ten non-blast TBI cases and three cases with no history of TBI. The chronic blast cases showed a distinctive pattern of interface astroglial scarring at boundaries between brain parenchyma and fluids such as at junctions between grey and white matter, the subpial glial plate, penetrating superficial cortical blood vessels and structures lining ventricles. A case-control study of US veterans concluded that mild TBI without loss of consciousness increased the risk of dementia modestly and of Alzheimer’s disease specifically and that the risk was higher when there was loss of consciousness, and increased further with more severe TBI (contradicted by Weiner et al. 2017).

Second impact syndrome (SIS) The review by Cantu and Gean (2010) describes SIS as ‘typically occurring in athletes who had post-concussion symptoms after the first head injury which may include headache, visual, motor and sensory changes or mental difficulty (especially cognitive and memory problems) before the symptoms resolved, which might take days or weeks. The athletes who return to competition receive a second blow to the head. The second blow may be remarkably minor (perhaps involving only a blow to the chest and indirect injuries to the athlete’s head by causing accelerating forces to the brain). The affected athlete may appear stunned but usually does not experience loss of consciousness. The person usually remains on their feet for 15 seconds to 1 minute or so but appear dazed, like someone suffering from concussion. Frequently affected athletes remain on the playing field or walk off under their own power. After this, the athlete who is conscious yet stunned, collapses to the ground, semi-comatose with rapidly dilating pupils, loss of eye movement and respiratory failure’. The second injury, it is believed, results in catastrophic brain swelling and often fatal outcomes (Cantu 1998; McCrory et al. 2012; Saunders and Harbaugh 1984). The Cantu and Gean paper also described characteristic imaging features which are, 1. Thin subdural haemorrhage with a maximum thickness of about 0.5 cm; 2. Heterogeneous appearing subdural haematomas; 3. Complete effacement of the basal cisterns in cerebral sulci (indicating brain swelling and oedema); 4. A distorted brainstem due to uncal and diencephalic herniation; 5. No evidence of intra-axial haemorrhage such as contusions or white matter injury; 6. Preservation of grey-white matter differentiation within cerebral hemispheres on admission CT scan (indicating no early primary ischaemic damage); 7. Hemispheric asymmetry defined as a thickness of hemisphere measured at the level of mid lateral ventricles (1 cerebral hemisphere may be more swollen than the other one); 8. Multifocal post-traumatic ischaemic infarction on follow-up imaging studies (secondary ischaemia within the brain as a late complication). The concept of SIS and whether this constitutes a definable and reliably diagnosable entity is highly controversial and has not found general acceptance. The possibility that severe brain

swelling may result from a single blow questions whether a repeat head injury is essential to cause this entity (McCrory et al. 2012). The concept of SIS largely rests on the interpretation of anecdotal case reports, the low number of which suggests a vanishingly rare condition (if it exists). A substantial number of reported cases including the index case of this entity have actually not involved a ‘second’ impact. The only systematic review of this condition up to 2001 identified a total of 17 cases in the world literature (McCrory 2001). Of these, only five cases actually involved a repeated injury. Even in those cases, it was often not clear that the initial injury played a contributory role insofar as providing evidence that the athlete had no objective evidence of ongoing symptoms and/or injury prior to the putative second impact. The vast bulk of anecdotal reports are reported in the North American literature with near negligible numbers of case reports originating from the remainder of the world, including countries in which contact sports with a risk of concussive head injury (such as Australian football and rugby) are popular. It has been pointed out that a sport such as Australian football, which has both a high participation rate and a high concussive injury rate (approximately 15 times that of American football) curiously does not appear to be associated with reports of SIS. An Australian study examining all sports-related death over a 35-year period identified no case of SIS (McCrory et al. 2000). It is even more curious (were SIS to exist as an entity) that SIS appears to be extremely rare amongst young boxers, with those few cases involving death attributed to SIS open to challenge (McCrory et al. 2012). McCrory and Berkovic identified 12 reports in the SIS literature which clearly described sport-related catastrophic brain injury associated with unexplained cerebral swelling. In these cases though the players did not have a second impact; they either collapsed during sport participation or walked off and collapsed without any further injury occurring (McCrory and Berkovic 1998). A paper by Stovitz et al. (2017) pointed out that the definitions of SIS varied. The authors aptly introduced their paper with a quotation of Tryon Edwards, an American theologian (1809–1894) who stated ‘most controversies would soon be ended, if those engaged in them would first accurately define their terms and then rigidly adhere to their definitions’. This group’s systematic review on the definition of SIS found that the vast majority of articles defined SIS as a syndrome requiring catastrophic brain injury (e.g. cerebral oedema) in a person who suffers a head trauma whilst still recovering of the effects of recent concussion. The definition did not include diffuse cerebral oedema resulting from a single significant impact. Taking the most common definition of the outcome of cerebral oedema with or without death, the death rate is unknown and likely unknowable. To the authors’ knowledge, no one has ever tracked random cases of cerebral oedema after single or consecutive concussions. The authors stated that when ‘2 impacts occur in very close succession (during the same game or within a few days) it is impossible to determine whether it was the initial injury that may have given rise to the catastrophic consequences’. The paper of Hebert et al. (2016) investigated the diagnostic credibility of SIS. This also involved a systematic literature review. The purpose of this review was to examine current literature to determine whether or not enough evidence exists to support a World Health Organisation (WHO) recognised ICD-case definition for SIS. Concussion and post-concussion syndrome are recognised diagnoses, each having its own WHOderived ICD-10 code. To create a case definition for disease, the WHO requires significant information on signs, symptoms,

References diagnostic guidelines and at-risk population to create criteria for assessing a diagnosis. The authors stated that most resources that have discussed SIS up to the time of the submission of their paper are secondary sources or position papers and that to their knowledge, there were no contemporary systematic reviews of the literature that examined the evidence to support or refute the signs, symptoms, diagnostic guidelines and at-risk populations for this reported diagnosis. They refer to the 1998 review by McCrory and Berkovic (see aabove) who reviewed proposed cases of SIS and compared them to a set of four criteria to determine if any of the cases could be uniquely defined as SIS. Of the 17 cases that were examined, only 5/17 were considered to be probable cases of diagnosed SIS, and 0/17 were found to be definitive cases of SIS. Of these 17 cases, 13 reported sport-related catastrophic brain injury associated with unexplained cerebral swelling; yet this adverse event was typically not associated with a second impact. The conclusion of this systematic literature review was that ‘evidence to definitely support a unique diagnosis of SIS is lacking within current literature’. The study did identify trends associated with suspected cases of SIS, but they were able to identify definitive reasons why a unique diagnosis of SIS was so difficult to fully support. This included, a. Mixed results regarding the presence of cerebral swelling and/or subdural haemorrhage on imaging; b. Lack of a defined time component associated with sequela following a second impact; c. The presence of significant signs/symptom overlap with the following conditions: concussion, post-concussion syndrome, epidural haematoma and/or malignant brain oedema; and d. The absence of unique and required triage protocols to distinguish and diagnose SIS. Whilst it has been well documented that the population at risk for developing SIS appears to be adolescent and young adult males, further research is needed to be conducted to determine if males are at an increased risk of developing SIS due to differences in their genetic composition or if males tend to participate more in the sport/activities commonly associated with SIS. The authors further stated that ‘at this time, there does not appear to be adequate depth of high quality literature supporting definitive cases of SIS. As such, this author group concludes that, at this time, SIS is not a credible, literature-supported diagnosis’. There has been no additional data so far which challenges/ refutes this reasonable position. Some studies appear to suggest that the development of brain swelling after a mild head impact may be related to genetic risk rather than serial impacts. Multiple case reports have described disproportionate brain swelling to a single mild head injury in patients with a personal or family history of hemiplegic migraine related to a familial or de novo mutation in one of the CACNA1A calcium channel subunit genes. Based on current understanding of concussion pathophysiology, other types of ion channel dysfunction are thought plausible contributing mechanisms (Kamins and Giza 2016).

References Baron, S.L., Hein, M.J., Lehman, E. and Gersic, C.M. 2012. Body mass index, playing position, race, and the cardiovascular mortality of retired professional football players. Am J Cardiol, 109, 889–896. Bey, T. and Ostick, B. 2009. Second impact syndrome. West J Emerg Med, 10, 6–10.

Cantu, R.C. 1998. Second-impact syndrome. Clin Sports Med, 17, 37–44. Cantu, R. and Gean, A. 2010. Second-impact syndrome and a small subdural haematoma: an uncommon catastrophic result of repetitive head injury with a characteristic imaging appearance. J Neurotrauma, 27, 1557–1564. Castellani, R.J. 2015. Chronic traumatic encephalopathy: a paradigm in search of evidence? Lab Invest, 95, 576–584. Castellani, R.J., Perry, G. and Iverson, G.L. 2015. Chronic effects of mild neurotrauma: putting the cart before the horse? J Neuropathol Exp Neurol, 74, 493–499. Castellani, R.J. and Perry, G. 2017. Dementia pugilistica revisited. J Alzheimers Dis, 60, 1209–1221. Corsellis, J.A.N., Bruton, C.J. and Freeman Browne, D. 1973. The aftermath of boxing. Psychol Med, 3, 270–303. Critchley, M. 1949. Punch-drunk syndromes: the chronic traumatic encephalopathy of boxers. In Hommage a Clovis Vincent. Paris: Maloine. DePalma, R.G. and Hoffman, S.W. 2018. Combat blast related traumatic brain injury (TBI): decade of recognition; promise of progress. Behav Brain Res, 340, 102–105. Gardner, A., Iverson, G.L. and McCrory, P. 2014. Chronic traumatic encephalopathy in sport: a systematic review. Br J Sports Med, 48, 84–90. Geddes, J.F., Vowles, G.H., Nicoll, J.A.R. and Revesz, T. 1999. Neuronal cytoskeletal changes are an early consequence of repetitive head injury. Acta Neuropathol, 98, 171–178. Goldstein, L.E., Fisher, A.M., Tagge, C.A., Zhang, X.L., Velisek, L. and Sullivan, J.A. et al. 2012. Chronic traumatic encephalopathy in blast-exposed military veterans and a blast neurotrauma mouse model. Sci Transl Med, 4, 134ra160. Guskiewicz, K.M., McCrea, M. and Marshall, S.W. et al. 2003. Cumulative effects associated with recurrent concussion in collegiate football players. JAMA, 290, 2549–2555. Hebert, O., Schlueter, K., Hornsby, M., Van Gorder, S., Snodgrass, S. and Cook, C. 2016. The diagnostic credibility of second impact syndrome: a systematic literature review. J Sci Med Sports, 19, 789–794. Iverson, G.L., Keene, C.D., Perry, G. and Castellani, R.J. 2018. The need to separate chronic traumatic encephalopathy neuropathology from clinical features. J Alzheimers Dis, 61, 17–28. Kamins, J. and Giza, C.C. 2016. Concussion – mild traumatic brain injury: recoverable injury with potential for serious sequelae. Neurosurg Clin N Am, 27, 441–452. Kirkendall, D.T., Jordan, S.E. and Garrett, W.E. 2001. Heading and head injuries in soccer. Sports Med, 31, 369–386. Kinch, K., Fullerton, J.L. and Stewart, W. 2019. One hundred years (and counting) of blast-associated traumatic brain injury. J R Army Med Corps, 165, 180–182. Lee, K.A. 2000. Injuries caused by animals. In Mason, J.K. and Purdue, B.N. (eds), Pathology of Trauma. Arnold, pp 265–282. Lindsay, K.W., McLatchie, G. and Jennett, B. 1980. Serious head injury in sport. BMJ, 281, 789–791. Marchie, A. and Cusimano, M.D. 2003. Bodychecking and concussions in ice hockey: should our youth pay the price? CMAJ, 169, 124. Martland, H.S. 1928. Punch drunk. JAMA, 91, 1103–1107. McCrory, P.R. and Berkovic, S.F. 1998. Second impact syndrome. Neurology, 50, 677–683. McCrory, P.R., Berkovic, S.F. and Cordner, S.M. 2000. Death due to brain injury among footballers in Victoria, 1968–1999. Med J Aust, 172, 217–219. McCrory, P.R. 2001. Does second impact syndrome exist? Clin J Sport Med, 11, 144–149. McCrory, P., Davis, G. and Makdissi, M. 2012. Second impact syndrome or cerebral swelling after sporting head injury. Curr Sports Med Rep, 11, 21–23. McKee, A.C., Cantu, R.C., Nowinski, C.J., Hedley-Whyte, E.T., Gavett, B.E., Budson, A.E., Santini, V.E., Lee, H.S., Kubilus, C.A. and Stern, R.A. 2009. Chronic traumatic encephalopathy in athletes: progressive tauopathy after repetitive head injury. J Neuropathol Exp Neurol, 68, 709–735. McKee, A.C., Gavett, B.E., Stern, R.A., Nowinski, C.J., Cantu, R.C. and Kowall, N.W. et al. 2010. TDP-43 proteinopathy and motor neuron disease in chronic traumatic encephalopathy. J Neuropathol Exp Neurol, 69, 918–929.

149

150

Contact sport and blast-related neuropathology McKee, A.C., Stern, R.A., Nowinski, C.J., Stein, T.D., Alvarez, V.E. and Daneshvar, D.H., et al. 2013. The spectrum of disease in chronic traumatic encephalopathy. Brain, 13, 43–64. McKee, A.C. and Robinson, M.E. 2014. Military-related traumatic brain injury and neurodegeneration. Alzheimers Dement, 10, S242–S253. McKee, A.C., Stein, T.D., Kiernan, P.T. and Alvarez, V.E. 2015. The neuropathology of chronic traumatic encephalopathy. Brain Pathol, 25, 350–364. McKee, A.C., Cairns, N.J., Dickson, D.W., Folkerth, R.D., Keene, C.D. and Litvan, I. et al. 2016. The first NINDS/NIBIB consensus meeting to define neuropathological criteria for the diagnosis of chronic traumatic encephalopathy. Acta Neuropathol, 131, 75–86. Millspaugh, J. 1937. Dementia pugilistica. US Nav Med Bull, 35, 297–303. Montenigro, P.H., Baugh, C.M., Daneshvar, D.H., Mez, J., Budson, A.E., Au, R., Katz, D.I., Cantu, R.C. and Stern, R.A. 2014. Clinical subtypes of chronic traumatic encephalopathy: literature review and proposed research diagnostic criteria for traumatic encephalopathy syndrome. Alzheimers Res Ther, 6, 68–84. Myles, S.T., Mohtadi, N.G. and Schnittker, J. 1992. Injuries to the nervous system and spine in downhill skiing. Can J Surg, 35, 643–648. Omalu, B.I., DeKosky, S.T., Minster, R.L., Kamboh, M.I., Hamilton, R.L. and Wecht, C.H. 2005. Chronic traumatic encephalopathy in a national football league player. Neurosurgery, 57, 128–134. Omalu, B.I., DeKosky, S.T., Hamilton, R.L., Minster, R.L., Kamboh, M.I., Shakir, A.M. and Wecht, C.H. 2006. Chronic traumatic encephalopathy in a national football league player: Part II. Neurosurgery, 59, 1086–1093. Plant, J.R. and Butt, J.C. 1993. Laceration of the vertebral artery. An historic boxing death. Am J Forensic Med Pathol, 14, 61–64.

Rabadi, M.H. and Jordan, B.D. 2001. The cumulative effect of repetitive concussion in sports. Clin J Sport Med, 11, 194–198. Roberts, G.W., Allsop, D. and Bruton, C. 1990. The occult aftermath of boxing. J Neurol Neurosurg Psychiatry, 53, 373–378. Ross, R.T. and Ochsner, M.G. Jr. 1999. Acute intracranial boxing-related injuries in U.S. Marine Corps recruits: report of two cases. Mil Med, 164, 68–70. Saunders, R.L. and Harbaugh, R.E. 1984. The second impact in catastrophic contact-sports head trauma. JAMA, 252, 538–539. Shively, S.B. and Perl, D.P. 2017. Viewing the invisible wound: novel lesions identified in postmortem brains of US service members with military blast exposure. Mil Med, 182, 1461–1463. Stovitz, S.D., Weseman, J.D., Hooks, M.C., Schmidt, R.J., Koffel, J.B. and Patricios, J.S. 2017. What definition is used to describe second impact syndrome in sports? A systematic and critical review. Curr Sports Med Rep, 16, 50–55. Tripathy, A., Shade, A., Erskine, B., Bailey, K., Grande, A., deLong, J.J., Perry, G. and Castellani, R.J. 2019. No evidence of increased chronic traumatic encephalopathy pathology or neurodegenerative proteinopathy in former military service members: a preliminary study. J Alzheimers Dis, 67, 1277–1289. Weiner, M.W., Crane, P.K., Montine, T.J., Bennett, D.A. and Veitch, D.P. 2017. Traumatic brain injury may not increase the risk of Alzheimer disease. Neurology, 89, 1923–1925. Williams, D.J. and Tannenberg, A.E. 1996. Dementia pugilistica in an alcoholic achondroplastic dwarf. Pathology, 28, 102–104.

15

Head injury in the child Helen Whitwell and Christopher Milroy

Introduction This chapter covers the important topic of non-accidental head injury in the infant and child. This is an area, not only of expanding knowledge but also considerable debate.

Accidental injury The majority of incidences of accidental head injury in the paediatric age group arise as a result of motor vehicle accidents, including occupants as well as pedestrians and cyclists. Other causes injury include falls, particularly in the under 5 age group (Chaudhary et al. 2018) and injuries relating to recreational activities. The pattern of brain injury, particularly in the older child, resembles that seen in adults. There is some evidence that the maturing brain in the younger age group responds differently to that of the older child. There is a smaller risk of traumatic intracranial haematomas in children less than 5 years of age (Teasdale et al. 1990). Diffuse swelling is more common in the paediatric age group – this may be secondary to other pathologies such as ischaemia or contusional injury or may have no underling cause (Graham et al. 1989). This is covered further in Chapter 12.

Non-accidental head injury Various terms have been used for intentionally inflicted head injury, including non-accidental injury (NAI), shaken baby syndrome and more recently abusive head trauma (AHT) (Chaudhary et al. 2018; Narang et al. 2020). The use of AHT has been stated to be a medical diagnosis, but this has been challenged because of the use of abusive in the term (Lynøe and Eriksson 2019). Head injury is the major cause of death in cases of inflicted injury in young children (Hargrave and Earner 1992; Ellis 1997) and causes significant morbidity in the survivors (Caffey 1974; Ewing-Cobbs et al. 1998; Barlow and Minns 1999). Estimates of the true incidence are difficult and vary in the United Kingdom from between 11.2 per 100,000 children younger than 1 year to 24.6 per 100,000 (Barlow et al. 1998; Jayawant et al. 1998; Barlow and Minns 2000). In the United States, the incidence is estimated to be higher, 32–38 per 100,000 cases per year (Narang et al. 2020). There are significant differences in the neuropathology of injury in the child compared with the adult, as well as between children of different ages. The lack of reliable data has meant that opinions in this area, to a great extent, have been based on knowledge of adult head injury from both the biomechanical and the neuropathological aspects.

Post-mortem examination in a case of suspicious death Protocols have been developed for the post-mortem investigation of paediatric head injury. The investigation starts with as much history as possible. It should include details given by the carer(s) at the time of presentation as well as preceding events (Gill et al.

2014; Royal College of Pathologists 2016). Scene analysis may need to be correlated with the autopsy. Birth history, medical history and details of resuscitation should be obtained, including by carers and health professionals. Recognised resuscitation injuries include damage to the frenulum, bruising to the mouth area both internally and externally, as well as scalp bruising, particularly with neurosurgical procedures or, if in the posterior position, related to the position of the head (Leadbeatter 2001). With hospital survival, full clinical details, including radiological results and other investigative tests, must be reviewed. CT and magnetic resonance imaging (MRI) are now standard in the diagnosis of head injury in hospital and increasingly used post-mortem. If not performed clinically, a skeletal survey should be conducted. Post-mortem examination in these cases should follow a standard routine. Good-quality photography is important for both positive and pertinent negative findings. In addition to a detailed external and internal examination, extensive sampling should be conducted, including for histology, microbiology, metabolic studies, toxicology and samples for genetic testing and DNA analysis for identification. Assessment of extracranial injury is important. There are injuries considered hallmarks of child abuse but as in all cases, careful assessment is required (Milroy 2014; Kepron et al. 2016) Skeletal trauma is common and dealt with in standard textbooks (Bilo et al. 2010; Kleinman 2015). The main extracranial skeletal injuries seen in head injury cases are rib and metaphyseal fractures (Walker et al. 2016). It is for the forensic pathologist to assimilate these findings with the neuropathology.

External findings in head injury No external injury Atwal et al. (1998), in a series of 24 fatal cases, all younger than 5 years, identified 21 per cent with no visible external injuries. Careful inspection may reveal soft tissue swelling. It is recognised that bruising may develop over a period of time and deep bruising takes longer than superficial bruises to appear (Johnson 1990; Langlois and Gresham 1991). Bruising in darkcomplexioned children may be masked. Detailed subcutaneous dissection of the torso and limbs should be performed to identify or exclude ‘occult’ bruising.

Bruising This may comprise single or multiple areas and overlie any skull fracture present (Figure 15.1). Large diffuse areas are indicative of a broad surface impact, e.g. the floor or wall. It may occasionally be possible to identify patterns within the bruising relating to the surface such as a carpet weave or object that may produce a recognisable pattern. Small bruises may be the result of fingers, with larger more irregular areas being caused by slapping (Figure 15.2). Injuries to the ears may be as a result of direct blows or pinching. The classic picture of the battered child appears less often and presents less diagnostic difficulty than an isolated head injury

152

Head injury in the child with ocular pathology

Figure 15.3 Torn frenulum as a result of a direct blow to the mouth. non-prominent areas of the face. Dating of bruising, including histological assessment, is discussed in Chapter 6.

Abrasions Figure 15.1 Diffuse bruise overlying skull fracture.

(Whitwell 2001). In an attempt to clarify ‘normal’ accidental bruising in children, Carpenter (1999) surveyed 177 babies aged 6–12 months. There was an increase in bruises with mobility. Frequent locations were the bony prominences of the face and head. Mobility was again highlighted by Sugar et al. (1999). Atypical sites include the trunk and buttocks as well as the hands. Worrying sites on the head include the ear lobes and

These can occur in association with bruising. The features may be non-specific. Care should be taken in interpretation of some abrasions, because children are able to scratch themselves. Patterned abrasions may result when an object such as a ring or belt is involved.

Injury to the frenulum A direct blow to the mouth can cause frenular bruising or laceration (Figure 15.3). It may also occur when a feeding bottle is rammed into the mouth and has been reported due to resuscitation (Leadbeatter 2001, Kepron et al. 2016).

Traumatic alopecia These are irregular-shaped areas of hair loss, which occur in some cases. The force may be sufficient to cause an underlying subgaleal haematoma (Kempe 1975).

Internal examination

Figure 15.2 Fingertip bruising to the side of the face and ear.

Bruising to the subscalp tissues may be extensive or more discreet (Figure 15.4). If a hand has been used, small areas with a fingertip pattern may be identified. Bruising is usually seen where there is a recent fracture. Its absence should raise the possibility that the fracture may not be genuine, e.g. but a naturally occurring fissure. The latter is a well-recognised pitfall in the diagnosis of skull fractures in the infant. They are naturally occurring defects, which fuse as ageing occurs. Identification is usually radiological (Keats 1996; Swischuk 2003). If there is continuing doubt, it can be submitted for histology, which shows immature bone merging with mature lamellar bone.

Skull fractures

Figure 15.4 Diffuse subscalp bruise caused by impact with a wall. Care should be taken in interpretation of subscalp bruising in the very young infant, particularly with a history of instrumental delivery, which may show cephalohaematomas or subgaleal haemorrhage (O’Grady et al. 2000). In a small number of cases, there may be no evidence of either external or internal bruising. This is particularly so in the young infant (up to a few months in age). In one series, 8 of 53 cases showed no external or internal bruising and no skull fracture (Geddes et al. 2001a).

Skull fractures Skull fractures are frequent in inflicted head injury. They indicate impact injury – whether or not there has been additional shaking may be impossible to say. In one series of cases (Geddes et al. 2001a), 36 per cent had one or more skull fractures. This compares with Atwal et al. (1998), where the incidence was 42 per cent. Skull fractures may occur during natural and assisted birth (Figure 15.5) (Heise et al. 1996; King and Boothroyd 1998; Dupuis et al. 2005).

Figure 15.5 Neonatal fracture following difficult vacuum extraction.

Figure 15.6 Parietal fracture as a result of a blunt force impact.

A number of features relating to characteristics of fractures have been described in order to attempt to differentiate accidental injury from NAI (Hobbs 1984; Rao and Carty 1999). The features suggestive of NAI are as shown in the box (p. 161). Most authors, however, emphasise that there is no specific feature diagnostic of NAI and indeed most skull fractures in cases of NAI are linear with the more complex picture less commonly seen (Figure 15.6). The majority of fractures are located in the parietal or occipital regions. Diastatic fractures occur where the skull sutures are not completely fused and fractures may travel along with them. Separation of the sutures may also occur in severe cerebral oedema. Underlying abnormalities of nutritional, metabolic and genetic abnormalities such as rickets, Menkes disease and osteogenesis imperfecta should be excluded. (Brogdon et al. 2012; Kleinman 2015). Fractures of varying ages and of a number of bones may be seen and subdural haematomas have been reported (Tokoro et al. 1988; Steiner et al. 1996). Simple skull fractures have a low risk of intracranial complication (Harwood-Nash et al. 1971; Helfer et al. 1977). The literature on infants and children relating to falls has been extensively published (Chadwick et al. 1991; Williams 1991; Reiber 1993), and there has been a general consensus that lowlevel falls are rarely associated with skull fractures and are usually simple and linear, commonly parietal. However, Weber (1984, 1985), in a series of experiments in which he dropped dead infants on to a variety of surfaces, found skull fractures in all that were dropped from a height of 82 cm. This appears to be at odds with the clinical situation, although skull fractures may not necessarily indicate intracranial pathology (see also Chapter 4). Skull fractures occur in asymptomatic head-injured infants. Greenes and Schutzman (1998), reported a retrospective review of 101 infants aged under 2 years admitted to a paediatric hospital with head injury. In 20 per cent of cases, there were occult intracranial findings in asymptomatic children and all but one had a skull fracture.

153

Head injury in the child with ocular pathology

154

At post-mortem examination, histology should be taken of any fracture. Details of the histology of skull fractures are given in Chapter 7. However, the findings in infants and children are not well documented in the literature. Assessment of the age of any associated haemorrhage should be done which may aid interpretation.

Neuropathology of non-accidental injury The major neuropathological findings are as follows: • • • • • • • • •

Subdural haemorrhage Subarachnoid haemorrhage Cerebral swelling Intracerebral haemorrhage and contusional tears Hypoxic-ischaemic damage Traumatic axonal injury, both traumatic and ischaemic Chronic subdural haematomas and hygromas Spinal injury Brain damage in survivors

Subdural haemorrhages These are the most commonly identified intracranial lesion – 84 per cent of the infant age group with 81 per cent in the older child (Geddes et al. 2001a). In a population-based study by Jayawant et al. (1998), in 82 per cent of subdural haemorrhages (SDHs) in infants aged under 2 years, a cause was identified but not in the remaining 18 per cent. A lower incidence of 55 per cent was reported in a similar series from Australia (Tzioumi and Oates 1998). The haemorrhages are commonly bilateral and posteriorly sited (Duhaime et al. 1998). Particularly in the infant group, as opposed to the older child, they consist of thin films of blood rarely requiring neurosurgical intervention (Figure 15.7). At post-mortem examination, caution must be taken not to misinterpret blood that has drained from subdural sinuses, when these are incised, as ante-mortem subdural bleeding. The post-mortem appearance classically is of an extremely swollen brain with thin films of acute haemorrhage over the hemispheres, and not infrequently extension into the posterior interhemispheric region. In the older age group, space-occupying haematomas, which may be unilateral, are also seen, as

Figure 15.7 Thin films of haemorrhage in an infant with severe brain swelling.

is the more diffuse type of bleeding (Geddes et al. 2001a). The former do not necessarily lie on the same side as a fracture if present. The mechanism of the formation of SDHs is thought to be most commonly tearing of the bridging veins crossing the subdural space. Identification of ruptured bridging veins has been reported post-mortem (Maxeiner 2001). Recently, traumatic tearing of these veins as a mechanism for subdural bleeding has been questioned and indeed is of some controversy in the young. Intradural and SDH have been described in the very young paediatric population. The history of the mode of delivery is of extreme importance and ventouse extraction is recognised to have an incidence of SDHs, many of which resolve after birth (Liu et al. 1998, Towner et al. 1999, O’Grady et al. 2000). It has been suggested that hypoxic-ischaemic damage with raised intravascular pressure may cause the intradural haemorrhage with SDH (Geddes et al. 2003; Cohen and Scheimberg 2009). MRI in routine deliveries identifies asymptomatic SDH in a proportion. These resolve by 3 months (Whitby et al. 2003; Rooks et al. 2008). Subdural bleeding in utero, including frank haematomas, is also described (Barozzino et al. 1998). Underlying conditions, such as bleeding disorders, vascular malformations, sepsis and metabolic abnormalities, e.g. glutaric aciduria type I and hypernatraemia, should be sought (Ehrenforth et al. 1998; Ng et al. 1998; Renzulli et al. 1998; Handy et al. 1999; Rutty et al. 1999; Hartley et al. 2001; Menkes 2001; Pollanen 2011). Infants with enlarged extra-axial spaces, such as in shunted hydrocephalus, appear to be at increased risk of subdural bleeding with lesser degrees of trauma (Duhaime et al. 1998). Non-traumatic subdural haemorrhage in infants is rare. They have been associated with macrocrania, severe dehydration and arachnoid megaly (Lindberg et al. 2019) Not infrequently there is evidence of earlier subdural bleeding histologically (Geddes et al. 2001a) with haemosiderin staining of the dura. Evidence of such older haemorrhage should not be taken as certain evidence of previous NAI. If there is a definite and well-documented story of such, it is possible to say only that SDH could be as a result of such. However, the findings in themselves are non-specific and simply indicate the likelihood of previous trauma, including birth trauma (Figure 15.8). It is essential that several specimens of dura be taken to assess the presence of organising haemorrhage, membrane formation and haemosiderin deposition for the reasons above. This latter is said to take at least 48 hours to develop, although it should be said that this is only an estimate. Changes may only

Figure 15.8 Haemosiderin macrophages in the dura – away from acute haemorrhage.

Neuropathology of non-accidental injury be seen microscopically (Geddes et al. 2001a). Most of the work on ageing SDH relates to the adult population. This makes dating in the infant very difficult. It should be noted that extradural haemorrhage is extremely rare in NAI (Duhaime et al. 1998). Extradural haemorrhage may be a post-mortem artefact and caution must be exercised in its interpretation when present (Rutty et al. 2005; Pollanen et al. 2009).

Subarachnoid haemorrhage This is present in about half the cases of paediatric head injury (Geddes et al. 2001a). It is seen in association with either SDH and/or skull fracture. It is commonly seen in a patchy fashion over the hemisphere post-mortem, but after the brain has been fixed in formalin, it is often less extensive than originally thought. Subarachnoid haemorrhage is not clinically significant and, apart from where it is associated with a fracture, it is not an accurate indicator of location of injury. Very rare cases of traumatic subarachnoid haemorrhage have been reported with rupture of an intracranial vessel (Leestma 2014). This may be delayed with a traumatic aneurysm, which has been reported in NAI (Lam et al. 1996).

(a)

Cerebral swelling One of the major findings in inflicted head injury is cerebral swelling. This is most commonly seen in association with global neuronal hypoxic-ischaemic damage (see next) and is the most common mode of death. The most accurate way to assess this is by weight of the brain compared with the expected size for the age and other growth parameters of the child (Appendix 2, Chapter 5). It should be recognised that in the forensic setting congestive cerebral swelling, such as may occur in an ‘asphyxial’ type of death, may also show an increase in brain weight. The brain should be weighed fresh before suspension using accurate scales. The weight should be rechecked after fixation. One of the reasons that the infantile brain does not show the same features as the adult is that there is the possibility for the infant’s skull to expand. Thus, the various herniations are not usually present as they are in the adult. In infants with severe hypoxic damage, particularly when there has been a period of survival, often with assisted ventilation, the brain is extremely softened and ‘pours out’ on opening the skull. Artefactual disruption and distortion frequently happen even when extreme care is taken. Removal underwater may aid. After fixation, macroscopic examination may show only compression of the ventricles. Histological examination demonstrates predominantly white matter oedema with myelin swelling. With increasing survival time, additional features, including hypoxic-ischaemic damage, are commonly seen. Amyloid precursor protein (APP) staining often shows a vascular pattern (Figure 15.9a). This is usually widespread in the white matter. Focal geographic APP expression, commonly seen in the diencephalon and brainstem, is taken as outlining areas of incipient infarction caused by brain swelling (Geddes et  al. 2000). Distinguishing between vascular and traumatic damage to axons (Figure 15.9b) is not always easy and great care is needed, particularly in the brainstem (see also Chapter 11). The importance of this distinction is that, if axonal injury in this location (in the distribution of that seen in diffuse traumatic axonal injury rather than the localised corticospinal tract damage – see next) is claimed definitely to be traumatic, this conclusion implies that a significant degree of force has been

(b)

Figure 15.9 (a) ‘Vascular’ pattern of amyloid precursor protein (APP) expression and (b) traumatic axonal injury (APP – white matter).

applied and that there were immediate profound clinical effects at the time of injury. Brain swelling with compression of blood vessels leads to ischaemic white matter damage in similar areas to that seen in traumatic axonal injury (Shannon et al. 1998; Geddes et al. 2000) This is particularly so in cases of short survival which may be impossible to interpret. In some, albeit infrequent, instances where death occurs rapidly, swelling may not be a feature. If the cause of death is head injury, it is probable that the injury produced traumatic damage incompatible with life – presumably to the brainstem. In the author’s experience, such a situation occurs most commonly in association with significant impact. The mechanism of acute cerebral swelling after minor head injury is discussed in Chapter 12.

Intracerebral haemorrhage In association with contusional injury of adult type With increasing age of the child, the findings are essentially those seen in the adult with surface contusional/lacerating injury. In the younger infant, the ‘adult-type’ contusional injury is not a common finding, although if present it is most frequently seen in the olfactory bulbs or tracts and gyrus rectus (Rorke 1992). In severe trauma associated with fracture, true coup contusions relating to the fracture site may be seen. The various types of intracerebral haematoma seen are covered in Chapter 8.

155

156

Head injury in the child with ocular pathology

Contusional tears These relatively uncommon lesions appear to occur only in very young infants (Lindenberg and Freytag 1969; Calder et al. 1984). Contusional tears have been described at post-mortem examination by a number of authors (Lindenberg and Freytag 1969; Calder et al. 1984, Geddes et al. 2001a) but can also be identified radiologically, including by transcranial ultrasonography (Jaspan et al. 1992). They occur at the junction of the grey and white matter and are believed to result from shearing at that site. The classic locations are the frontal (beneath the superior frontal convolution and in the orbital gyri), temporal and occipital regions. Although many of these cases occur with impact injury, including severe impact, contusional tears may also be seen where there is minimal or no evidence of impact (Geddes et al. 2001a). It is the author’s view that these lesions are essentially focal in nature and represent the way that infants’ brains (poorly or not myelinated) react and are not comparable with contusional injury seen in the adult. In this context, it is clear that the degree of force required to produce these tissue tears is unknown, and saying that they necessarily require the same amount of force as needed to produce diffuse traumatic axonal injury cannot be justified. Larsen et al. (2019) have described a series of cases with subcortical clefts and hypothesised that subcortical clefts are not due to direct mechanical forces of trauma but are part of a secondary cascade caused by impaired venous drainage which may or may not follow trauma. A recent tear can be identified macroscopically as a slit, with variable haemorrhage. Caution is necessary, however, because cracks, slits or apparent tears in the white matter, which may look haemorrhagic, are relatively common findings in paediatric neuropathology and are not necessarily of any significance. Damage to tissue can occur if the brain is ‘roughly’ handled, and the author has seen one case where transportation immediately after removal and before adequate fixation resulted in artefactual slits in the tissue that resembled contusional tears. Even after some time in formalin, an infant’s brain falls apart as it is examined, particularly if it is swollen, poorly fixed or infarcted. The only way of proving beyond doubt that a white matter cavity is a contusional tear is by histology, showing haemorrhage or damaged axons around the edges, or a cellular reaction in or around the cavity. For that reason, it is essential, if possible, that the brain is well fixed and not cut fresh (Figure 15.10a). Older tears can be identified macroscopically as slits with yellow-brown discolouration indicative of old haemorrhage (Figure 15.10b). Histologically, a recent tear shows variable recent haemorrhage, with separation of the tissue (Figure 15.11). Depending on the timing after injury APP staining may demonstrate axonal damage, as well as a cellular reaction with neutrophil polymorphs (although these are most usually scanty). There may be focal shrinkage and ‘dark cell’ change in adjacent neurons, and after some hours’ survival genuine hypoxic-ischaemic changes. With the passage of time, macrophages accumulate along with reactive astrocytes and increase in microglial cells. Haemosiderin is usually identifiable after 48 hours. An old tear may be identifiable by haemosiderin macrophages and gliosis. Timing is obviously an issue, but caution should be applied and over-dogmatic opinions are ill-advised.

(a)

(b)

Figure 15.10 (a) Recent contusional tear – superior frontal region. (b) Bilateral older contusional tears – slit-like appearance. because similar lesions may be seen in so-called asphyxial deaths or unexplained deaths in infancy (see also Chapter 17).

Hypoxic-ischaemic injury This has become increasingly recognizsed as the major pathology in the infant. It is identifiable by neuropathological criteria of widespread neuronal eosinophilia and shrinkage

Petechial haemorrhages Not infrequently microscopic petechial haemorrhages, often perivascular, may be seen in the cortex, white matter or deepgrey matter. Their significance has to be interpreted with caution

Figure 15.11 Contusional tear – haemorrhage with surrounding macrophages.

Neuropathology of non-accidental injury

Figure 15.12 Hypoxic-ischaemic neuronal injury.

(Figure 15.12). It is usually widespread throughout the brain, see Chapter 12. However, it is possible in some cases to identify changes earlier. Differentiation from the so-called dark cell change may be difficult. The usual cause of death is brain swelling related to the hypoxic damage. Apnoea is a frequent presentation in head injury (Johnson et al. 1995) and it is thought that this, rather than the mechanism of injury, has the major role in prognosis (Prasad et al. 2002). In most infants (up to 9 months old), this is the most common finding, with 78 per cent of cases showing widespread neuronal hypoxia and a large proportion having a history of apnoea or respiratory arrest (Geddes et al. 2001b). Radiological studies using diffusion-weighted MRI in the investigation of non-accidental head injury have identified changes of cerebral ischaemia rather than traumatic lesions as the major abnormality (Suh et al. 2001; Biousse et al. 2002; Stoodley 2002). In some cases, older areas of hypoxic-ischaemic damage may be seen suggesting previous injury, although as in the cases of long-term survivors other causes, such as birth injury, should be sought. In long-term survivors with severe damage, gross cerebral atrophy may develop (Figure 15.13).

Axonal injury Immunohistochemistry revealed the spectrum of axonal injury (Snyder and Hansen 2016). It has been widely assumed that the brain damage was diffuse axonal injury (DAI) (now more correctly termed ‘diffuse traumatic axonal injury or dTAI). Earlier studies have claimed that DAI is common. Vowles et al. (1987) published a neuropathological series of 10 cases and reported DAI in 6. This study was done before APP and utilised silver stains. The cases were inadequately studied by today’s standards, and what the authors reported was not now understood to be dTAI (Geddes et al. 2001b, Chapter 11). Review of these cases using immunohistochemistry (unpublished data) showed that dTAI was present in only one case. Most showed lesser degrees of axonal injury, the aetiological nature of which was unclear because of the inadequate nature of the examination and paucity of clinical information. Shannon et al. (1998) demonstrated localised axonal injury in the spinal cervical roots in NAI cases. In addition, they found similarities between controls dying of hypoxic-ischaemic brain damage and those in NAI where hypoxic-ischaemic damage was also seen. The β-APP staining showed similar patterns in both, although these authors felt that additional axonal injury was also present as a

result of trauma in some cases. They pointed out that dorsolateral quadrant injury in the brainstem was often not identified – an important criterion for dTAI (Geddes et al. 2000). Gleckman et al. (1999) published a series of 10 cases of NAI in infants and diagnosed DAI in 5 cases of ‘shaken baby syndrome’. However, much of the APP staining appears to indicate the vascular type. In 2001, Geddes and colleagues (2001a) published the largest series to that date of 53 infants and children with inflicted head injury. This series also had defined diagnostic criteria for inclusion. In the 37 infants, only two of the cases showed DAI; both of these had evidence of significant impact with skull fractures. The authors concluded that DAI was uncommon and that the most common pathology related to widespread hypoxic-ischaemic damage and swelling. In the eight cases where there was no impact (no bruising or skull fracture), no DAI was present. In 13 of 37, there was evidence of either macroscopic or microscopic cervical cord damage, including epidural haemorrhage, APP positivity of cervical nerve roots, and corticospinal axonal damage in the lower pons and medulla (Figure 15.13). This last has been reported in adult hyperextension injury (Lindenberg and Freytag 1970) but not previously in NAI. There was no difference between cases with and those without impact (Geddes et al. 2001a). The experience of the author and her colleagues is that histological evidence of long tract or nerve root damage is more likely to be found if a full craniocervical examination is undertaken, and the whole brainstem and upper cervical cord are embedded for histology. The findings suggest that the significant mechanism of injury is primary brainstem damage leading to severe apnoea or respiratory abnormalities, which in turn result in raised intracranial pressure from widespread hypoxic-ischaemic damage, and that the diffuse brain damage that occurs in the infants is hypoxic, not traumatic. These findings that DAI is rare in infants has been confirmed by Reichard and his colleagues (2003), in a study of 28 cases of inflicted head injury which included 10 children aged 12 months or younger. Only one instance of DAI was seen in the subgroup of infants, in a child of 10 months. This is also supported by the work of Duhaime et al. (1987) in that the forces generated by shaking are not of sufficient severity to cause DAI without impact. What is unclear is how much force is required to damage the craniocervical region in the young infant. Within the diagnostic setting, it is important to sample the brain widely along with the spinal cord (see Chapter 5). APP staining may well, if nothing else, give some indication of survival time. Hemispheric injury is seen as either axons or bulbs in the white matter, corpus callosum and internal capsule. Axonal swellings are usually visible from around 12–18 hours after injury. Axonal injury varies from scattered foci in the white matter, which is more common than DAI where injury by definition is widespread and includes the rostral brainstem. CD 68 identifies microglial activity – becoming positive a few days after injury – and may be useful in identifying previous episodes of injury, which are not infrequently seen. As commented in the discussion relating to hypoxic-ischaemic damage, interpretation of the changes may be difficult.

Chronic subdural haematomas and subdural hygromas These most commonly result from trauma and are often difficult to distinguish from each other (Wittschieber et al. 2015). Radiologically, a mixed density pattern may be seen – it is not

157

158

Head injury in the child with ocular pathology

(a)

(d)

(e) (b)

(c)

Figure 15.13 (a) and (b) Corticospinal axonal swellings – haematoxylin and eosin (H&E). (c) and (d) Corticospinal axonal swellings – amyloid precursor protein (APP). (e) APP positivity in spinal nerve root.

clear as to whether such a picture equates to more than one episode of any identified trauma (Wittschieber et al. 2019). They are discussed in further in Chapter 8, including their potential relationships with each other. In addition, the issue as to whether hygromas represent the remains of an acute SDH or by traumatically induced arachnoid tears is unresolved. It may well be multiple mechanisms exist. Investigation into other manifestations of injury should take place.

Brain damage in survivors In cases where death occurs at a later stage, a variety of findings may be seen. The major brain pathology results from hypoxicischaemic damage and variable grey and white matter damage, either cystic or non-cystic, may be seen (Figure 15.14). Residua of gliding contusions, often in the frontal lobes, may show as well-demarcated cavities. Other findings include old SDH, old

Neuropathology of non-accidental injury

Figure 15.14 Long-term survivor after hypoxic-ischaemic injury.

Figure 15.15 Cervical spine decalcified and sectioned with arrows indicating dorsal nerve roots with haemorrhage.

surface contusional injury and secondary optic nerve degeneration.

for histology (Figure 15.15). Histochemical stains and immunohistochemistry can be employed on the blocks produced. They reported that where hyperextension was either suspected or confirmed all had evidence of unilateral or bilateral haemorrhages within the or surrounding the cervical spinal roots in C3–C5 (Figure 15.16) The technique was also employed by Ali and Fowler (2016) who stated that ‘epidural haemorrhage (diffuse or focal), focal intradural haemorrhage, focal subarachnoid haemorrhage, perivertebral artery/perivertebral space haemorrhage as well as minimal microscopic subdural haemorrhage, are not necessarily indicative of inflicted trauma and should be interpreted in context with other findings. The most compelling findings indicative of trauma (hyperflexion/hyperextension injuries) are dorsal root ganglia or nerve rootlet haemorrhage and paravertebral muscle haemorrhage’. A common artefact seen is epidural blood along the length of the cord, resulting from congestion in the epidural vascular fat network (Valdes-Dapena 1975). This should not be mistaken for true trauma. However, epidural bleeding localised to the cervical cord region, particularly in association with haemorrhage into the soft tissues of the neck, may be indicative of trauma to the craniocervical junction (Geddes et al. 2001a).

Spinal injury Various approaches have been proposed for examining the spinal column and spinal cord at autopsy (Judkins et al. 2004; Matshes et al. 2011) Injury to the spine is attracting increasing attention but remains understudied (Brennan et al. 2009). It is difficult to assess the true incidence of spinal fractures in child abuse. It is said to vary from 0 to 3 per cent (Akbarnia 1999). No cervical fractures were identified in the series of Brennan but soft tissue and muscular damage were identified. Many fractures are thought to go unidentified and they may be found without obvious clinical findings – this may have significance in timing (Cramer 1996; Akbarnia 1999). The majority of fractures involve the vertebral body with anterior compression fractures to the lower thoracic and upper lumbar areas. In addition, rupture of the spinal ligament and vertebral dislocation may be seen. The mechanism of injury varies but includes hyperextension-flexion, compression fractures as a result of compaction to the buttocks or head, and fractures of spinal processes as a result of direct impact. Spinal fracture sites should be examined histologically as should the spinal cord. It may be possible at least to indicate whether an injury is acute or shows ageing features. However, only general comments about timing may be achievable. Spinal subdural haematoma and extradural haemorrhage have been reported as a common finding in children with inflicted head injury when they had undergone imaging of the thoracolumbar spine (Choudhary et al. 2012; Choudhary 2020; Rabbitt et al. 2020). Cervical spine ligamentous injuries (predominantly the nuchal, atlanto-occipital and atlanto-axial ligaments) have radiologically (using STIR imaging) been shown to be common in inflicted head injury. Pathological correlation for such lesions is lacking (Choudhary et al. 2014). Matshes et al. (2011) published a detailed analysis of 35 infants and children where the neck had been excised and examined histologically. The technique involves the removal of the cervical spine en bloc, followed by decalcification, sectioning the spine transversely and the cut sections submitted

Figure 15.16 Haemorrhage in dorsal nerve root ganglia.

159

160

Head injury in the child with ocular pathology

Shaking or impact – the controversy Shaken baby syndrome remains a controversial area of forensic practice (Papetti 2018). Whether this is a real controversy has been challenged (Greeley 2014). Regardless those who are involved in this area need to be aware of the issues. A number of authors over the years have suggested that shaking alone was the explanation for cases of SDH where no scalp injury was identifiable (Guthkelch 1971; Caffey 1972). Caffey (1974) coined the phrase ‘whiplash shaken infant syndrome’. The term ‘shaken baby syndrome’ has now become virtually synonymous with the combination of SDH, retinal haemorrhages and brain injury (‘the triad’). The issue of how much trauma is necessary is debatable in this young age group and others have questioned whether unexplained subdural always equals either abuse or trauma (Fung et al. 2002). However, it is clear from pathological studies that there is commonly evidence of impact post-mortem, although this may not be identifiable externally. Shaking alone has, however, been thought to be inadequate to produce sufficient force to generate findings relating to shaken baby syndrome, including SDH. Biomechanical studies in this area are discussed in Chapter 4. Lynøe and Eriksson (2018) summating the literature with papers reporting confessions and convictions with paediatric head injury found that 33 per cent had no evidence of impact. Although in the past this has been attributed to impact with a soft surface, it is being re-questioned in the light of the recent pathological studies, which indicate that dTAI occurs only where there is significant impact and that the young infant more usually shows evidence of a focal stretch type injury to the craniocervical junction (Geddes et al. 2001a). Currently, there are no biomechanical data to indicate the degree of force required to produce this localised stretch injury, which, although similar to that seen in adult hyperextension injury (Lindenberg and Freytag 1970; Geddes et al. 2000), may be exceptionally variable in its severity. Is it necessary to postulate shaking where there is evidence of impact injury? The answer to this is most probably no. Impact has been shown to produce subdural bleeding and brain swelling, with various degrees of traumatic axonal injury. The series of Geddes et al. (2001a) showed similar findings in relation to the craniocervical injury in shaking alone as in those cases in which there was evidence of impact. This suggests that flexion and hyperextension of the neck may almost certainly occur as part of an impact injury. The controversy has been further debated in the literature and questioned by a systematic review from Sweden (SBU 2016; Lynøe et al. 2017; Elinder et al.2018 ). However, this review has been criticised by others (for example Debelle et al. 2018; Laurent-Vannier et al. 2018). The authors of the SBU have counter challenged this criticism stating that the evidence relating to shaking is scientifically poor and their review is valid (Lynøe and Eriksson 2020). The debate continues.

Re-bleeding in subdural haematomas It is well recognised in adults that a process of organisation and liquefaction of a subdural haematoma will occur within several days, with membrane formation. It has been said by some authors that re-bleeding in an SDH is not an explanation for symptoms or presentation in the young (David 1999; Lindberg 2019). This is also subject to debate. Evidence of SDHs of varying ages is seen in cases of infant head injury. Work referred to before (Greenes and Schutzman 1998) indicates that scanning reveals intracranial pathology, including SDH in a number of

asymptomatic children, and ‘disappearing subdurals’ have been described (Duhaime et al. 1996). It is well recognised that SDHs occur as the result of birth, including normal deliveries, however, as discussed previously, these resolve by 3 months. Data on re-bleeding remains limited.

Timing of injury Much of this is within the realms of clinical paediatrics. However, the forensic neuropathologist is often asked to comment on the effect of the injuries identified and the probable clinical symptomatology. Caution is advised in these circumstances, particularly so on the background of previous injury, including chronic SDH. In addition, the basis for traumatic unconsciousness was always thought to be that related to diffuse traumatic axonal injury. This has now been shown not to be so in most cases. It is likely that the injuries to the craniocervical junction found in a number of infants may lead to breathing abnormalities. In those infants, children who survive for some hours, pathological findings such as APP positivity, axonal retraction balls and other findings, e.g. CD68 positivity may assist in broad terms. It is recognised that the time interval between injury and onset of symptoms in a number of cases remains unclear (Nashelsky and Dix 1995; Gilliland 1998); however, in most cases with no complicating factors, it is generally accepted that an infant in the case of severe injury would not appear normal to a carer (David 1999). However, there are reported cases with delay in presentation – these appear to be related to delayed onset of complications such as cerebral swelling or presentation of an intracranial haematoma (Denton and Mileusnic 2003). This is one of the most difficult areas and the forensic neuropathologist would be wise to be limited in opinion.

How far to fall? This is another difficult area in which to draw conclusions in an individual case. There are many reported series of children with no fatalities from low-level falls-also, see earlier (Chadwick and Salerno 1993; Williams 1991; Warrington and Wright 2001; Lyons and Oates 1993; Chadwick et al. 2008). There have been individual case reports and series of documented genuine injuries, including fatalities from low-level falls (Aoki and Masuzawa 1984; Howard et al. 1993; Christian et al. 1999; Kim et al. 2000; Aly-Hamdy et al. 2001; Plunkett 2001; Reichelderfer et al. 1979; Hall et al. 1989), in infants/children with or without skull fractures as well as other intracranial pathology, including SDH. Stairway falls are also problematic – deaths are reported to be rare. Joffe and Ludwig (1988) found no correlation with injury and number of stairs, concluding these were less severe than a fall through the same vertical distance. Reconstruction in the individual situation is difficult, if not impossible. Detailed assessment of the known circumstances as well as assessment of other injuries is essential. Biomechanical engineering assessment is becoming more frequently used in these cases as well as the use of infant models to stimulate injury (see Chapter 4).

Conclusion Non-accidental head injury in infants and children is highly complex and probably the most controversial area of forensic neuropathology. The protection of children is paramount, however, in justice to the accused opinions must be given on the basis of solid scientific findings and careful assessment of the evidence.

References

Features of skull fractures suspicious of nonaccidental injury • • • • •

Non-parietal fracture Multiple or complex fractures Depressed fracture Diastatic fractures .5 mm Growing fractures

References Akbarnia, B.A. 1999. Pediatric spine fractures. Orthop Clin N Am, 30, 521– 536. Ali, Z. and Fowler, D.R. 2016. En bloc examination of the neck in pediatric homicide cases: a proper way for complete assessment of neck trauma. Acad Forensic Pathol, 6(4):622–637. Aly-Hamdy, N., Childs, A.M., Ferrie, C.D. and Livingstone, J.H. 2001. Subdural haemorrhage with bilateral retinal haemorrhage following accidental household trauma. Dev Med Child Neurol, 43(suppl 90), 24. Aoki, N. and Masuzawa, H. 1984. Infantile acute subdural haematoma: clinical analysis of 26 cases. J Neurosurg, 61, 273–280. Atwal, G.S., Rutty, G.N. and Carter, N. et al. 1998. Bruising in non-accidental head injured children: a retrospective study of the prevalence, distribution and pathological associations in 24 cases. Forensic Sci Int, 96, 215–230. Barlow, K.M. and Minns, R.A. 1999. The relation between intracranial pressure and out come in non-accidental head injury. Dev Med Child Neurol, 41, 220–225. Barlow, K.M. and Minns, R.A. 2000. Annual incidence of shaken impact syndrome in young children. Lancet, 356, 1571–1572. Barlow, K.M., Milne, S. and Minns, R.A. 1998. A retrospective epidemiological analysis of non-accidental head injury in children in Scotland over the last 15 years. Scott Med J, 43, 112–114. Barozzino, T., Sgro, M. and Toi, A. et al. 1998. Fetal bilateral subdural haemorrhages. Prenatal diagnosis and spontaneous resolution by time of delivery. Prenat Diagn, 18, 496–503. Bilo, R.A., Robben, S.G. and Van Rijn, R.R., 2010. Forensic Aspects of Pediatric Fractures: Differentiating Accidental Trauma from Child Abuse. Berlin: Springer Science & Business Media. Biousse, V., Suh, D.Y. and Newman, N.J. 2002. Diffusion weighted magnetic resonance imaging in shaken baby syndrome. Am J Ophthalmol, 133, 249–255. Brogdon, B.G., Shwayder, T. and Elifritz, J., 2012. Child Abuse and Its Mimics in Skin and Bone. Boca Raton: CRC Press. Brennan, L.K., Rubin, D. and Christian, C. et al. 2009. Neck injuries in young pediatric homicde victimes. J Neurosurg Pediatr, 3, 232–239. Caffey, J. 1972. On the theory and practice of shaking infants. Its potential residual effects of permanent brain damage and mental retardation. Am J Dis Child, 124, 161–163. Caffey, J. 1974. The whiplash shaken infant syndrome: manual shaking by the whiplash-induced intracranial and intraocular bleedings, linked with residual damage and mental retardation. Pediatrics, 54, 396–403. Calder, I.M., Hill, I. and Scholtz, C.L. 1984. Primary brain trauma in nonaccidental injury. J Clin Pathol, 37, 1095–1100. Carpenter, R.F. 1999. The prevalence and distribution of bruising in babies. Arch Dis Child, 80, 363–366. Chadwick, D.L. and Salerno, C. 1993. Likelihood of the death of an infant or young child in a short fall of less than 6 vertical feet. J Trauma, 35, 968. Chadwick, D.L., Chin, M. and Salerno, C. et al. 1991. Deaths from falls in children. How far is fatal? J Trauma, 31, 1353–1355. Chadwick, D.L., Bertocci, G. and Castillo, E. et al. 2008. Annual risk of death resulting from short falls among young children less than 1 in 1 ­million. Pediatrics, 121(6):1213–1224 Chaudhary, S., Figueroa, J., Shaikh, S., Mays, E.W., Bayakly, R., Javed, M., Smith, M.L., Moran, T.P., Rupp, J. and Nieb, S. 2018. Pediatric falls ages 0–4: understanding demographics, mechanisms, and injury severities. Injury Epidemiol, 5(1), 7.

Choudhary, A.K., Bradford, R.K., Dias, M.S., Moore, G.J. and Boal, D.K., 2012. Spinal subdural hemorrhage in abusive head trauma: a retrospective study. Radiology, 262(1), 216–223. Choudhary, A.K., Ishak, R., Zacharia, T.T. and Dias, M.S. 2014. Imaging of spinal injury in abusive head trauma: a retrospective study. Pediatr Radiol, 44, 1130–1140. Choudhary, A.K., 2020. Understanding the importance of spinal injury in abusive head trauma (AHT). Pediatr Radiol, 50(1), 15–16. Choudhary, A.K., Servaes, S., Slovis, T.L., Palusci, V.J., Hedlund, G.L., Narang, S.K., Moreno, J.A., Dias, M.S., Christian, C.W., Nelson, M.D. and Silvera, V.M. 2018. Consensus statement on abusive head trauma in infants and young children. Pediatr Radiol, 48(8), 1048–1065. Christian, C.W., Taylor, A.A., Hertle, R.W. and Duhaime, A.C., 1999. Retinal hemorrhages caused by accidental household trauma. J Pediatr, 135(1), 125–127. Cohen, M.C. and Scheimberg, I., 2009. Evidence of occurrence of intradural and subdural hemorrhage in the perinatal and neonatal period in the context of hypoxic ischemic encephalopathy: an observational study from two referral institutions in the United Kingdom. Pediatr Dev Pathol, 12(3), 169–176. Cramer, K.E. 1996. Orthopedic aspects of child abuse. Pediatr Clin N Am, 43, 1035–1051. David, T.J. 1999. Shaken baby (shaken impact) syndrome: non-accidental head injury in infancy. R Soc Med, 92, 556–561. Debelle, G.D., Maguire, S., Watts, P., Hernandez, R.N. and Kemp, A.M., 2018. Abusive head trauma and the triad: a critique on behalf of RCPCH of ‘Traumatic shaking: the role of the triad in medical investigations of suspected traumatic shaking’. Arch Dis Child, 103(6), 606–610. Denton, S. and Mileusnic, D. 2003. Delayed sudden death in an infant following an accidental fall. Am J For Med Pathol, 24, 371–376. Duhaime, A.C., Gennarelli, T.A. and Thibault, L.E. et al. 1987. The shaken baby syndrome. A clinical, pathological, and biochemical study. J Neurosurg, 66, 409–415. Duhaime, A.C., Christian, C. and Armonda, R. et al. 1996. Disappearing subdural haematomas in children. Paediatr Neurosurg, 25, 116–122. Duhaime, A.C., Christian, C.W., Rorke, L.-B. and Zimmerman, R.A. 1998. Nonaccidental head injury in infants-the ‘shaken-baby syndrome’. N Engl J Med, 338, 1822–1829. Dupuis, O., Silveira, R., Dupont, C., Mottolese, C., Kahn, P., Dittmar, A. and Rudigoz, R.C. 2005. Comparison of “instrument-associated” and “spontaneous” obstetric depressed skull fractures in a cohort of 68 neonates. Am J Obstet Gynecol, 192(1), 165–170. Ehrenforth, S., Klarmann, D. and Zabel, B. et al. 1998. Severe factor V deficiency presenting as subdural haematoma in the newborn. Eur J Pediatr, 157, 1032. Elinder, G., Eriksson, A., Hallberg, B., Lynøe, N., Sundgren, P.M., Rosén, M., Engström, I. and Erlandsson, B.E. 2018. Traumatic shaking: the role of the triad in medical investigations of suspected traumatic shaking. Acta Paediatr, 107, 3–23. Ellis, P.S.J. 1997. The pathology of fatal child abuse. Pathology, 29, 113–121. Ewing-Cobbs, L., Kramer, L. and Prasad, M. et al. 1998. New imaging, physical and developmental findings after inflicted and non-inflicted traumatic brain injury in young children. Paediatrics, 102, 300–307. Fung, E.L., Sung, R.Y. and Nelson, E.A. et al. 2002. Unexplained subdural hematoma in young children: is it always child abuse? Pediatr Int, 44, 37–42. Geddes, J.F., Hackshaw, A.K., Vowles, G.H., Nickols, C.D. and Whitwell, H.L. 2001a. Neuropathology of inflicted head injury in children. 1. Patterns of brain damage. Brain, 124, 1290–1298. Geddes, J.F., Tasker, R.C. and Hackshaw, C.D. et al. 2003. Dural haemorrhage in non-traumatic infant deaths: does it explain the bleeding in ‘shaken baby syndrome’? Neuropathol Appl Neurobiol, 29, 14–22. Geddes, J.F., Vowles, G.H. and Hackshaw, A.K. et al. 2001b. Neuropathology of inflicted head injury in children. 2. Microscopic brain injury in infants. Brain, 124, 1299–1306. Geddes, J.F., Whitwell, H.L. and Graham, D.I. 2000. Traumatic axonal injury: practical issues for diagnosis in medicolegal cases. Neuropathol Appl Neurobiol, 26, 105–116.

161

162

Head injury in the child with ocular pathology Gill, J.R., Andrew, T., Gilliland, M.G.F., Love, J., Matshes, E. and Reichard, R.R., 2014. National association of medical examiners position paper: recommendations for the postmortem assessment of suspected head trauma in infants and young children. Acad Forensic Pathol, 4(2), 206–213. Gilliland, M.G.F. 1998. Interval duration between injury and severe symptoms in nonaccidental head trauma in infants and young children. J Forensic Sci, 43, 723–725. Gleckman, A.M., Bell, M.D., Evans, R.J. and Smith, T.W. 1999. Diffuse axonal injury in infants with nonaccidental craniocerebral trauma: enhanced detection by beta-amyloid precursor protein immunohistochemical staining. Arch Pathol Lab Med, 123, 146–151. Graham, D.I., Ford, I. and Adams, J.H. et al. 1989. Fatal head injury in ­children. J Clin Pathol, 42, 18–22. Greeley, C.S., 2014. “Shaken baby syndrome” and forensic pathology. Forensic Sci Med Pathol, 10(2), 253–255. Greenes, D. and Schutzman, S. 1998. Occult intracranial injury in infants. Ann Emerg Med, 32, 680–686. Guthkelch, A.N. 1971. Infantile subdural haematoma and its relationship to whiplash injuries. BMJ, 11, 430–431. Hall, J.R., Reyes, H.M. and Horvat, M. et al. 1989 The mortality of childhood falls. J Trauma, 29, 1273–1275 Handy, T.C., Hanzlick, R. and Shields, L.B.E. et al. 1999. Hypernatremia and subdural hematoma in the pediatric age group: Is there a causal relationship? J Forensic Sci, 44, 1114–1118. Hargrave, D.R. and Earner, D.P. 1992. A study of child homicide over two decades. Med Sci Law, 32, 247–250. Hartley, L.M., Khwaja, O.S. and Verity, C.M. 2001. Glutaric aciduria type 1 and nonaccidental head injury. Pediatrics, 107, 174–176. Harwood-Nash, D.C., Hendrick, E.B. and Hudson, A.R. 1971. The significance of skull fractures in children. Radiology, 101, 151–156. Heise, R.H., Srivatsa, P.J. and Karsell, P.R., 1996. Spontaneous intrauterine linear skull fracture: a rare complication of spontaneous vaginal delivery. Obstet Gynecol, 87(5), 851–854. Helfer, R.E., Slovis, T.L. and Black, M. 1977. Injuries resulting when small children fall out of bed. Pediatrics, 60, 533–535. Hobbs, C.J. 1984. Skull fracture and the diagnosis of abuse. Arch Dis Child, 59, 246–252. Howard, M.A., Bell, B.A. and Uttley, D. 1993. The pathophysiology of infant subdural haematoma. Br J Neurosurg, 7, 355–365. Jaspan, T., Narborough, G., Punt, J.A. and Lowe, J. 1992. Cerebral contusional tears as a marker of child abuse-detection by cranial sonography. Pediatr Radiol, 22, 237–245. Jayawant, S., Rawlinson, A. and Gibbon, F. et al. 1998. Subdural haemorrhages in infants: population based study. BMJ, 317, 1558–1561. Joffe, M. and Ludwig, S. 1988. Stairway injuries in children. Paeditrics, 82, 457–461. Johnson, C.F. 1990. Inflicted injury versus accidental injury. Pediatr Clin N Am, 37, 791–814. Johnson, D.L., Boal, D. and Baule, R. 1995. Role of apnea in nonaccidental head injury. Pediatr Neurosurg, 23, 305–310. Judkins, A.R., Hood, I.G. and Mirchandani, H.G. et al. 2004. Technical communication: rationale and technique for examination of nervous system in suspected infant victims of abuse. Am J Forensic Med Patholol, 25, 29–32. Keats, T.E. 1996. Atlas of Normal Roentgen Variants that May Simulate Disease, 6th edn. Mosby. Kempe, C.H. 1975. Uncommon manifestations of the battered child syndrome. Am J Dis Child, 129, 1265. Kepron, C., Walker, A. and Milroy, C.M., 2016. Are there hallmarks of child abuse? II. Non-osseous injuries. Acad Forensic Pathol, 6(4), 591–607. King, S.J. and Boothroyd, A.E., 1998. Cranial trauma following birth in term infants. Br J Radiol, 71(842), 233–238. Kim, K.A., Wang, M.Y. and Griffith, P.M. et al. 2000. Analysis of pediatric head injury from falls. Neurosurg Focus, 8, 1–5. Kleinman, P.K. ed., 2015. Diagnostic Imaging of Child Abuse. Cambridge: Cambridge University Press. Lam, C.H., Montes, J., Farmer, J.-P. et al. 1996. Traumatic aneurysm from shaken baby syndrome: case report. Neurosurgery, 39, 1252–1255.

Langlois, N.E.I. and Gresham, G.A. 1991. The aging of bruises: a review and study of the color changes with time. Forensic Sci Int, 50, 227–238. Larsen, K.B., Barber, Z. and Squier, W. 2019. The pathology and aetiology of subcortical clefts in infants. Forensic Science International, 296, 115–122. Laurent-Vannier, A., Adamsbaum, C., Raul, J.S., Rey-Salmon, C. and Rambaud, C., 2018. Flawed Swedish study on traumatic shaking is already being used by defence lawyers and its findings must be ignored. Acta Paediatrica (Oslo, Norway: 1992), 107(12), 2048–2050. Leadbeatter, S. 2001. Resuscitation injury. In G.N. Rutty (ed), Essentials of Postmortem Examination Practice, 9th edn, Vol 1. Springer-Verlag. Leestma, J.E. 2014. Forensic Neuropathology. Boca Raton: CRC Press Lindberg, D.M., Dubowitz, H., Alexander, R.C. and Reece, R.M., 2019. The “New science” of abusive head trauma. Int J Child Maltreatment: Res Policy Practice, 2(1–2), 1–16. Lindenberg, R. and Freytag, E. 1969. Morphology of brain lesions from blunt trauma in early infancy. Arch Pathol, 87, 298–305. Lindenberg, R. and Freytag, E. 1970. Brainstem lesions characteristic of traumatic hyperextension of the head. Arch Pathol, 90, 509–515. Liu, H.Y., Huang, L.T. and Lui, C.C. 1998. Vacuum extraction delivery complicated with acute subdural haematoma and cerebral infarction: report of one case. Chung Hua Min Kuo Hsiao Erh Ko I Hsueh Hui Tsa Chih, 39, 119–122. Lynøe, N., Elinder, G., Hallberg, B., Rosén, M., Sundgren, P. and Eriksson, A., 2017. Insufficient evidence for ‘shaken baby syndrome’–a systematic review. Acta Paediatr, 106(7), 1021–1027. Lynøe, N. and Eriksson, A., 2018. Is focusing on the triad in suspected child abuse cases really irrelevant and of no practical use? Acta Paediatr, 107(10), 1675–1676. Lynøe, N. and Eriksson, A., 2019. Is the procedure of diagnosing abusive head trauma a purely medical matter? Pediatr Radiol, 49(3), 422–423. Lynøe, N. and Eriksson, A., 2020. No similarities between the Wakefield report on measles, mumps and rubella vaccine and the Swedish report on traumatic shaking. Acta Paediatr, 109(7), 1326–1329. Lyons, T.J. and Oates, R.K. 1993. Falling out of bed: a relatively benign occurrence. Pediatrics, 92, 125–127. Maxeiner, H., 2001. Demonstration and interpretation of bridging vein ruptures in cases of infantile subdural bleedings. J Forensic Sci, 46(1), 85–93. Matshes, E.W., Evans, R.M., Pinckard, J.K., Joseph, J.T. and Lew, E.O., 2011. Shaken infants die of neck trauma, not of brain trauma. Acad Forensic Pathol, 1(1), 82–91. Milroy, C.M. 2014. Blunt abdominal and chest injuries in children. In Byard, R.W. and Collins, K.A. (eds), Forensic Pathology of Infancy and Childhood. New York: Springer. Maxeiner, H. 2001. Demonstration and interpretation of bridging vein ruptures in cases of infantile subdural bleedings. J Forensic Sci, 46, 85–93. Menkes, J.H. 2001. Subdural haematoma, non-accidental injury or …? Eur J Paediatr Neurol, 5, 175–176. Narang, S.K., Fingarson, A., Lukefahr, J. and ABUSE, C.O.C., 2020. Abusive head trauma in infants and children. Pediatrics, 145(4). Nashelsky, M.B. and Dix, J.D. 1995. The time interval between lethal infant shaking and onset of symptoms. A review of the shaken baby syndrome literature. Am J Forensic Med Pathol, 16, 154–157. Ng, P.C., Fok, T.F. and Lee C.H. et al. 1998. Massive subdural haematoma: an unusual complication of septicaemia in preterm very low birthweight infants. J Pediatr Child Health, 34, 296–298. O’Grady, J.P., Pope, C.S. and Patel, S.S. 2000. Vacuum extraction in modern obstetric practice: a review and critique. Curr Opin Obstet Gynecol, 12, 475–480. Papetti, R. The unreliability of Shaken Baby Syndrome by R Papetti Academic Forensic Pathology International 2018 Plunkett, J. 2001. Fatal pediatric head injuries caused by short-distance falls. Am J Forensic Med Pathol, 22, 1–12. Pollanen, M.S., Perera, S.D.C. and Clutterbuck, D.J., 2009. Hemorrhagic lividity of the neck: controlled induction of postmortem hypostatic hemorrhages. Am J Forensic Med Pathol, 30(4), 322–326. Pollanen, M.S., 2011. Subdural hemorrhage in infancy: keep an open mind. Forensic Sci Med Pathol, 7(3), 298–300.

Retinal haemorrhages Prasad, M., Ewing-Cobbs, L., Swank, P.R. and Kramer, L. 2002. Predictors of outcome following traumatic brain injury in young children. Pediatr Neurosurg, 36, 64–74. Rabbitt, A.L., Kelly, T.G., Yan, K., Zhang, J., Bretl, D.A. and Quijano, C.V., 2020. Characteristics associated with spine injury on magnetic resonance imaging in children evaluated for abusive head trauma. Pediatr Radiol, 50(1), 83–97. Rao, P. and Carty, H. 1999. Non-accidental injury: review of the radiology. Clin Radiol, 54, 11–24. Reiber, G. 1993. Fatal falls in childhood; how far must children fall to sustain fatal head injury? Report of cases and review of the literature. Am J Forensic Med Pathol, 14, 201–207. Reichard, R.R., White, C.L., Hladik, C.L. and Dolinak, D. 2003. Beta-amyloid precursor protein staining of nonaccidental central nervous system injury in pediatric autopsies. J Neurotrauma, 20, 347–355. Reichelderfer, T.E., Overbach, A. and Greensher, J. 1979 Unsafe playgrounds Paediatrics, 64, 962–963 Renzulli, P., Tuchschmid, P. and Eich, G. et al. 1998. Early vitamin K deficiency bleeding after maternal phenobarbital intake: management of massive intracranial haemorrhage by minimal surgical intervention. Eur J Pediatr, 157, 663–665. Rooks, V.J., Eaton, J.P., Ruess, L., Petermann, G.W., Keck-Wherley, J. and Pedersen, R.C., 2008. Prevalence and evolution of intracranial hemorrhage in asymptomatic term infants. Am J Neuroradiol, 29(6), 1082–1089. Rorke, L.B. 1992. Neuropathology. In Ludwig, S. and Kornberg, A.E. (eds), Child Abuse: A Medical Reference, 2nd edn. Churchill Livingstone, pp 403–421. Royal College of Pathologists. 2016. Sudden unexpected death in infancy and childhood. https://www.rcpath.org/uploads/assets/874ae50ec754-4933-995a804e0ef728a4/Sudden-unexpected-death-ininfancy-and-childhood-2e.pdf Rutty, G.N., Smith, C.M. and Malia, R.G. 1999. Late-form hemorrhagic disease of the newborn. A fatal case report with illustration of investigations that may assist in avoiding mistaken diagnosis of child abuse. Am J Forensic Med Pathol, 20, 48–51. Rutty, G.N., Squier, W.M.V. and Padfield, C.J.H., 2005. Epidural haemorrhage of the cervical spinal cord: a post-mortem artefact?. Neuropathol Appl Neurobiol, 31(3), 247–257. SBU. 2016. Traumatic Shaking: The Role of the Triad in Medical Investigations of Suspected Traumatic Shaking. Stockholm: Swedish Agency for Health Technology Assessment and Assessment of Social Services (SBU). Shannon, P., Smith, C.R. and Deck, J. et al. 1998. Axonal injury and the neuropathology of shaken baby syndrome. Acta Neuropathol, 95, 625–631. Snyder, V.S. and Hansen, L.A., 2016. A conceptual overview of axonopathy in infants and children with allegedly inflicted head trauma. Acad Forensic Pathol, 6(4), 608–621. Steiner, R.D., Pepin, M. and Byers, P.H. 1996. Studies of collagen synthesis and structure in the differentiation of child abuse from osteogenesis imperfecta. J Pediatr, 128, 542–547. Stoodley, N. 2002. Non-accidental head injury in children: gathering the evidence. Lancet, 360, 271–272. Sugar, N.F., Taylor, J.A. and Feldman, K.W. 1999. Bruises in infants and toddlers. Arch Pediatr Adolesc Med, 153, 399–403. Suh, D.Y., Davis, P.C. and Hopkins, K.L. et al. 2001. Nonaccidental pediatric head injury: diffusion weighted imaging findings. Neurosurgery, 49, 309–320. Swischuk, L.E. (eds) 2003. Imaging of the Newborn, Infant and Young Child, 3rd edn. Lippincott Williams & Wilkins. Teasdale, G.M., Murray, G. and Anderson, E. et al. 1990. Risks of acute traumatic intracranial haematoma in children and adults: implications for managing head injuries, BMJ, 300, 363–367. Tokoro, K., Nakajima, F. and Yamataki, A. 1988. Infantile chronic subdural hematoma with local protrusion of the skull in a case of osteogenesis imperfecta. Neurosurgery, 22, 595–598. Towner, D., Castro, M.A. and Eby-Wilkins, E. et al. 1999. Effect of mode of delivery in nulliparous women on neonatal intracranial injury. New Engl J Med, 341, 1709–1714.

Tzioumi, D. and Oates, K.R. 1998. Subdural haematomas in children under 2 years accidental or inflicted? A 10 year experience. Child Abuse Neglect, 22, 1105–1112. Valdes-Dapena, M. 1975. Sudden death in infancy: a report for pathologists. Perspect Pediatr Pathol, 2, 1–14. Vowles, G.H., Scholtz, C.L. and Cameron, J.M. 1987. Diffuse axonal injury in early infancy. Clin Pathol, 40, 185–189. Walker, A., Kepron, C. and Milroy, C.M., 2016. Are there hallmarks of child abuse? I. Osseous injuries. Acad Forensic Pathol, 6(4), 568–590. Warrington, S.A. and Wright, C.M., ALSPAC Study Team. 2001. Accidents and resulting injuries in premobile infants: data from the ALSPAC study. Arch Dis Child, 85, 104–107. Weber, W. 1984. Experimental studies of skull fractures in infants. Z Rechtsmed, 92, 87–94. Weber, W. 1985. Biomechanical fragilty of the infant skull. Z Rechtsmed, 94, 93–101 Whitby, E.H., Griffiths, P.D. and Rutter, S. et al. 2003. Frequency and natural history of subdural haemorrhages in babies and relation to obstetric factors. Lancet, 362, 846–851. Whitwell, H.L. 2001. Non-accidental injury in children. In Lowe, D.G. and Underwood, J.C.E. (eds), Recent Advances in Histopathology, 9th edn, Vol 19. Churchill Livingstone. Williams, R.A. 1991. Injuries in infants and small children resulting from witnessed and corroborated free falls. J Trauma, 31, 1350–1353. Wittschieber, D., Karger, B., Niederstadt, T., Pfeiffer, H. and Hahnemann, M.L., 2015. Subdural hygromas in abusive head trauma: pathogenesis, diagnosis, and forensic implications. Am J Neuroradiol, 36(3), 432–439. Wittschieber, D., Karger, B., Pfeiffer, H. and Hahnemann, M.L. 2019. Understanding subdural collections in pediatric abusive head trauma. Am J Neuroradiol, 40(3), 388–395.

OCULAR PATHOLOGY AND HEAD INJURY Daniel du Plessis and Christopher Milroy

Retinal haemorrhages Retinal haemorrhages are strongly associated with non-accidental head injury in infants (NAHI) but not a prerequisite for such a diagnosis as it may be absent in 40–60 per cent of cases. Ocular findings must be viewed in the context of the whole case. These can be seen at autopsy using indirect ophthalmoscopy (Lantz and Adams 2005) and after the globe is dissected (Figure 15.17) and on histological examination (Figure 15.18), retinal haemorrhages have a very high specificity (94–97 per cent) and sensitivity (75 per cent) for inflicted head trauma (Bhardwaj et al. 2010). The specificity for inflicted head injury is further increased when there is intracranial pathology (Pierre-Kahn et al. 2003; Vinchon et al. 2010) and when there is bilateral involvement, peripheral involvement, intraretinal, preretinal, premacular haemorrhages and moderate-tosevere intraocular haemorrhages (Bechtel et al. 2004). In stark contrast, the incidence of retinal haemorrhages in accidental injury has been reported as less than 4 per cent (in most studies as low as zero, especially when head trauma is due to shortfalls (Binenbaum and Forbes 2014). In a true accidental setting, the mechanism involves a high-energy event such as severe and/or complex road traffic accidents, high-level falls or severe crush head injuries. In such situations, the retinal haemorrhages are confined to the posterior pole, few in number and rarely subretinal.

163

164

Head injury in the child with ocular pathology

Figure 15.19 Optic nerve sheath haemorrhage. Figure 15.17 Macroscopic retinal haemorrhages in the dissected globe.

infantile Terson’s syndrome due to a ruptured aneurysm (Mena et al. 2011).

Traumatic retinoschisis and perimacular folds Traumatic retinoschisis and perimacular folds are present in a minority of AHT (8 and 14 per cent, respectively), but rarely seen in other conditions (Bhardwaj et al. 2010).

Peripapillary scleral haemorrhage Bleeding at the optic nerve/scleral junctions (peripapillary scleral haemorrhages) is common in NAHI (47 per cent), but only rarely seen in accidental trauma (∼1–2 per cent) (RCPCH/ RCO 2013). Such bleeds might be a marker of injury severity; it does not invariably accompany ONSHs and it appears more commonly associated with the most severe retinal haemorrhages in cases of NAHI. Figure 15.18 Histology of extensive retinal haemorrhage.

Extraocular and intraorbital haemorrhages

Optic nerve sheath haemorrhages (ONSH)

Haemorrhage into the orbital fat has been reported in 28 per cent of cases with inf licted head injury by Wygnanski-Jaffe et al. (2006). Its presence, in particular posterior located bleeds, may aid discrimination of non-accidental from accidental injury. Orbital exenteration (en bloc removal of the entire orbital contents) rather than simple enucleation therefore appears best forensic practice in cases of suspected NAHI.

Optic nerve sheath haemorrhage (ONSH) may be seen grossly and on histology (Figure 15.19). In accidental head trauma cases, haemorrhage in the optic nerve sheath has been found in 5 per cent of cases and retinal haemorrhage in 6 per cent. ONSH restricted to microscopic foci only was found in only 1 of 18 cases of non-traumatic infant deaths (RCPCH/RCO 2013). Puanglumyai and Lekawanvijit (2017) found ONSH in all cases of inflicted head trauma investigated by them (1 unilateral, 10 bilateral), compared to none of their 5 cases of fatal closed accidental head trauma. Subarachnoid haemorrhage involving the optic nerve sheath was restricted to cases of non-accidental injury in an autopsy study (RCPCH/RCO 2013). Mena et al. (2011) reported that ‘mild-to-moderate’ subarachnoid haemorrhage was reported in the optic nerves of a case of

Other causes of retinal haemorrhage Although retinal haemorrhages are strongly associated with trauma, there is an extensive list of other causes of retinal haemorrhages (see Table 15.1). The most relevant are discussed next. Many of the entities listed are rare to extremely rare. Most may readily be diagnosed by their associated clinical and/or biochemical features, other ocular or systemic abnormalities, or relation to known trauma or surgery.

Terson syndrome

Table 15.1 Causes of retinal haemorrhages 1. Trauma a. Birth (normal and traumatic; especially with vacuum assisted deliveries) b. Accidental injury (direct and indirect injury) c. Non-accidental injury i. Direct eye trauma ii. Indirect injury (‘shaken baby syndrome’, ‘NAI’, ‘inflicted neurotrauma’ etc.) d. Surgical (during or following eye surgery) 2. Bleeding disorders (coagulopathies) a. Haemophilia b. von Willebrand’s disease c. Protein C deficiency d. Henoch-Schonlein purpura e. Severe anaemia f. Leukaemia g. Vitamin K deficiency (haemorrhagic disease of the newborn) h. Homocystinuria i. Thrombocytopaenia j. Haemophagocytic lymphohistiocytosis k. Factor XIII (fibrin stabilising factor) deficiency 3. Infection a. Meningitis (usually meningococcal, possibly streptococcal) b. Encephalitis c. Septicaemia d. Bacterial endocarditis e. Congenital infection (e.g. TORCH infections) f. Malaria (usually p. falciparum, occasionally p. vivax) g. Whooping cough (pertussis) – (anecdotal historic cases) 4. Other inflammatory diseases a. Sarcoidosis b. Behcet’s disease 5. Intrinsic diseases of the eye a. Retinopathy of prematurity (ROP) b. Hereditary vitreoretinopathy/retinal dysplasia, e.g. Familial exudative vitreoretinopathy; Incontinentia pigmenti; Norrie’s disease c. Coats’ disease d. Persistent foetal vasculature/persistent hyperplastic primary vitreous (PHPV) e. X-linked retinoschisis f. Retinal neoplasms (i.e. retinoblastoma) g. Retinal angiomatosis h. Optic disc drusen 6. Toxic/metabolic a. Glutaric aciduria b. Galactosaemia c. Methylmalonic aciduria d. Cobalamin deficiency e. Severe hypernatraemia f. Carbon monoxide poisoning g. Warfarin poisoning/overdosage h. Prolonged use of salicylates i. Vitamin C deficiency – (anecdotal historic cases) 7. Other conditions a. Intracranial vascular malformation b. Terson’s syndrome (retinal haemorrhage with subarachnoid haemorrhage)

c. Carotid-cavernous fistula d. Cardiopulmonary resuscitation (CPR) e. Valsalva retinopathy f. Purtscher retinopathy g. Severe hypertension h. Hyperviscosity syndrome i. Maternal use of cocaine (in the newborn) j. Extracorporeal membrane oxygenation (ECMO) k. Haemophagocytic histiocytosis

Birth-related head injury Birth-related retinal haemorrhages are commonly seen via indirect ophthalmoscopy (in 34 per cent of babies up to 30 hours of age). It may be seen in 75 per cent of vacuum-assisted deliveries, 33 per cent of vaginal deliveries and in 7 per cent of babies born by caesarean section. Such bleeds are almost invariably intraretinal, ranging from a single dot haemorrhage in one eye to bilateral widespread haemorrhages, occasionally with white centres. Multilayered haemorrhages have not been reported in the newborn. Almost all such haemorrhages have resolved by 2–4 weeks after birth (Emerson et al. 2001).

Resuscitation Retinal haemorrhage has rarely been reported in resuscitation (Odom et al. 1997). Prolonged and vigorous cardiopulmonary resuscitation has been associated with posterior pole and mid-periphery retinal haemorrhages in a 17-month-old child with normal blood clotting parameters (Kramer and Goldstein 1993). No evidence of subdural haemorrhage was found at post-mortem. The conditions of the optic nerve sheaths were not reported. When retinal haemorrhages do occur after resuscitation, there are usually other risk factors with a mild haemorrhagic retinopathy in the posterior pole (Pham et al. 2013).

Raised intracranial pressure Only a minority of older children (above 3 years) with measured significantly elevated intracranial pressure (ICP) due to a nontraumatic cause develop retinal haemorrhages. Most of these show papilloedema whereas only 9 per cent of cases of NAHI have papilloedema. Retinal haemorrhages in the setting of non-traumatically raised ICP tend to be superficial intraretinal haemorrhages located adjacent to (or on) the optic nerve head (Binenbaum et al. 2013).

Terson syndrome Intracranial subarachnoid haemorrhage has rarely been associated with intraocular bleeding and, when present, has predominantly been seen in older children (mean age: 10.3 years). Associated ocular haemorrhages tended to be mild and concentrated around the optic disc. McLellan et al. (1986) reported a case of a 6-week-old girl who survived rupture of an MCA aneurysm in which was complicated by a large intracerebral haematoma. Extensive, bilateral retinal haemorrhages and a large, right-sided subhyaloid haemorrhage were found on ophthalmoscopy. Two further cases have been reported where ruptured middle cerebral artery aneurysms in 7-month-old infants were associated with subarachnoid haemorrhage and extensive retinal haemorrhages (Bhardwaj et al. 2010; Mena et al. 2011).

165

166

Head injury in the child with ocular pathology

Figure 15.20 Extramedullary haemopoiesis in the eye.

Ocular findings in sudden unexpected death in infancy (SUDI) Arredondo et al. (2008) examined the eyes in 102 deaths under the age of 2. This series included 57 cases of SIDS (sudden infant death syndrome). Novel cytoid bodies were present in the retina of 72/102 cases. SIDS was the most associated with cytoid bodies (85 per cent). In 35 cases, there was extramedullary haematopoiesis (EMH), with 22 out of the 57 cases classified as SIDS having EMH (Figure 15.20)

References Arredondo, J.L., Fernandes, J.R. and Rao, C. 2008. Ocular findings in pediatric deaths under 2 years of age (1994–2004). J Forensic Sci, 53, 928–934. Bechtel, K., Stoessel, K., Leventhal, J.M., Ogle, E., Teague, B., Lavietes, S., Banyas, B., Allen, K., Dziura, J. and Duncan, C. 2004. Characteristics that distinguish accidental from abusive injury in hospitalized young children with head trauma. Pediatrics, 114, 165–168. Bhardwaj, G., Chowdhury, V., Jacobs, M.B., Moran, K.T., Martin, F.J. and Coroneo, M.T. 2010. A systematic review of the diagnostic accuracy of ocular signs in pediatric head trauma. Ophthalmol, 117, 983–992. Bhardwaj, G., Jacobs, M.B., Moran, K.T. and Tan, K. 2010. Terson syndrome with ipsilateral severe hemorrhagic retinopathy in a 7-month-old child. JAAPOS, 14, 441–443.

Binenbaum, G., Rogers, D.L., Forbes, B.J., Levin, A.V., Clark, S.A., Christian, C.W., Liu, G.T. and Avery, R. 2013. Patterns of retinal hemorrhage associated with increased intracranial pressure in children. Pediatrics 132, e430–e434. Binenbaum, G. and Forbes, B.J. 2014. The eye in child abuse: key points on retinal hemorrhages and abusive head trauma. Pediatr Radiol, 44, 571–577. Emerson, M.V., Pieramici, D.J., Stoessel, K.M., Berreen, J.P. and Gariano, R.F. 2001. Incidence and rate of disappearance of retinal hemorrhage in newborns. Ophthalmology, 108, 36–39. Kramer, K. and Goldstein, B. 1993. Retinal hemorrhages following cardiopulmonary resuscitation. Clin Pediatr (Phila), 32, 366–368. Lantz, P.E. and Adams, G.G.W. 2005. Postmortem monocular indirect ophthalmoscopy. J Forensic Sci, 50, 1450–1452. McLellan, N.J., Prasad, R. and Punt, J. 1986. Spontaneous subhyaloid and retinal haemorrhages in an infant. Arch Dis Child, 61, 1130–1132. Mena, O.J., Paul, I. and Reichard, R.R. 2011. Ocular findings in raised intracranial pressure: a case of Terson syndrome in a 7-month-old infant. Am J Forensic Med Pathol, 32, 55–57. Odom, A., Christ, E., Kerr, N., Byrd, K., Cochran, J., Barr, F., Bugnitz, M., Ring, J.C., Storgion, S., Walling, R. and Stidham, G. 1997. Prevalence of retinal hemorrhages in pediatric patients after in-hospital cardiopulmonary resuscitation: a prospective study. Pediatrics, 99, e3. Pham, H., Enzenauer, R.W., Elder, J.E. and Levin, A.V. 2013. Retinal hemorrhage after cardiopulmonary resuscitation with chest compressions. Am J Forensic Med Pathol, 34, 122–124. Pierre-Kahn, V., Roche, O., Dureau, P., Uteza, Y., Renier, D., Pierre-Kahn, A. and Dufier, J.L. 2003. Ophthalmologic findings in suspected child abuse victims with subdural hematomas. Ophthalmology, 110, 1718– 1723. Puanglumyai, S. and Lekawanvijit, S. 2017. The importance of optic nerve sheath hemorrhage as a postmortem finding in cases of fatal abusive head trauma: a 13-year study in a tertiary hospital. Forensic Sci Int, 276, 5–11. RCPCH/RCO – Royal College of Paediatrics and Child Health and The Royal College of Ophthalmologists. Abusive Head Trauma and the Eye in Infancy. June 2013. https://www.rcophth.ac.uk/wp-content/ uploads/2014/12/2013-SCI-292-ABUSIVE-HEAD-TRAUMA-AND-THEEYE-FINAL-at-June-2013.pdf. Vinchon, M., de Foort-Dhellemmes, S., Desurmont, M. and Delestret, I. 2010. Confessed abuse versus witnessed accidents in infants: comparison of clinical, radiological, and ophthalmological data in corroborated cases. Childs Nerv Syst, 26, 637–645. Wygnanski-Jaffe, T., Levin, A.V., Shafiq, A., Smith, C., Enzenauer, R.W., Elder, J.E., Morin, J.D., Stephens, D. and Atenafu, E. 2006. Postmortem orbital findings in shaken baby syndrome. Am J Ophthalmol, 142,233–240.

16

Spinal injuries Graham Flint and Helen Whitwell

Spinal injuries are uncommon, when compared with other conditions. Causes include road traffic incidents, falls, sports injuries and penetrating wounds. Estimates of annual incidence vary worldwide, with quoted figures of between 1.5 and 3.9 per 100,000 population, per annum, in economically advanced countries (Cripps et al. 2011). The majority of victims are young and their resultant burdens will be spread over many years. Prior to the pioneering work of Sir Ludwig Guttmann (Ross and Harris, 1980), the outlook for victims of spinal cord injury was grim. Many would succumb to pulmonary emboli and chest infections, as well as metabolic and renal complications. Subsequent advances and the development of guidelines for optimum management mean that life quality and expectancy is now very much improved for those with a spinal cord injury (World Health Organisation 2013; National Institute for Health and Care Excellence 2016).

Mechanisms and forces involved The forces to which the spinal column is subjected during injury can be (1) compression, (2) flexion, (3) extension, (4) distraction, (5) rotation and (6) shearing. The consequent damage will depend upon the level of the spine exposed to such energies and whether they act in isolation or combination. Penetrating injuries, from knife, bullet or impalement wounds, form a separate category, within which individual injuries may be difficult to categorise; neurological injury is a common consequence but skeletal stability is usually maintained.

Clinical assessment Many cases of spinal injury occur as part of polytrauma and the primary and secondary surveys are vital parts of their assessment, evaluating in particular any associated head injuries (American College of Surgeons 2012); see also Chapter 2. The principal concern with a spinal injury relates to any possible damage to the spinal cord and, to a lesser extent, the nerves issuing from it. The challenge, to all those managing an actual or potential spinal injury, is to ascertain whether or not the skeletal elements are stable, or at least to allow for this possibility until the actual state of the vertebral column is established. With most (but not all) stable injuries, the cord is not under threat. With unstable fractures, and/or dislocations, the cord is at risk. Even if it has already sustained some damage, this could be made worse if care is not taken in handling the skeletal components of the spine. The symptoms from which a spinal injury victim may complain are, for the most part, restricted to local pain, plus awareness of any loss of motor or sensory function. Signs of underlying skeletal damage may include local bruising, swelling, tenderness or deformity but the capacity to search for such signs may be limited, particularly in the victim of polytrauma. Neurological examination is needed as soon as practically possible. In addition to recording power (Table 16.1), together

with sensation and reflexes in the limbs, assessment should include the lower cranial nerves and function in the lower sacral dermatomes, including sacral reflexes.1 Findings should be documented, to provide a baseline from which to measure any subsequent deterioration or improvement. The absence of all motor, sensory and reflex function below the level of injury, soon after injury, may be due to spinal shock. This will resolve within 48 hours and, unless the conus has been damaged, sacral reflexes will then return. If, at this point, there is no voluntary motor or sensory function, the lesion is deemed to be complete. If, on the other hand, even a small amount of voluntary lower limb movement is seen, the individual will very likely recover sufficiently to walk again, given proper treatment. If there is sacral sparing evident at the outset, then the chances of functional recovery are also good. Trauma to the vertebral column can result in injury to individual nerve roots, the cauda equina or the spinal cord. The spinal cord can be damaged at any level and the resulting neurological deficit can be complete or incomplete. A number of patterns of neurological deficit may be identified (Table 16.2), although these are seldom in pure form and mixed patterns may be seen. Following the initial clinical assessment, it is important that the patient has regular, thorough neurological examinations. The neurological chart of the American Spinal Injuries Association (ASIA) provides for such documentation. The ASIA Impairment Scale (AIS), or the Frankel scale from which it was derived, can then be used for the classification of the patient’s neurology. These scales are very similar (Table 16.3) although the method of assignment to the various grades differs between the two. AIS limits the number of muscle groups tested, making the examination simpler. It also assigns a numerical score to the overall assessment. Frankel, instead, describes the outcome more in terms of the functional effect of any neurological change.

Radiological investigation This includes plain radiography, computerised tomography (CT) scanning and magnetic resonance (MR) imaging. Plain radiographs still provide the mainstay of such assessments in most trauma units but CT scanning is being used increasingly, to provide more anatomical information. CT imaging provides much more detail of bony anatomy and permits identification of various subtypes of the main categories of vertebral injury. MRI may be used, subsequently, to examine paraspinal soft tissues and the spinal cord. Key features to be noted on these various images are listed in Table 16.4. It is also important to remember that non-contiguous spinal injuries may be present, necessitating radiological assessment of the whole spine. The concept of spinal clearance is well established in trauma units. Cervical spinal injury is assumed to be present in any case of major trauma until it has been excluded. The need for 1

Anal wink and bulbocavernosus reflex.

Spinal injuries

168

Table 16.1 Medical Research Council scale for grading muscle power* Strength of movement

Table 16.3 Classification of spinal cord injuries

Grade

No muscle contraction evident

0

Flicker of movement noted

1

Movement of limb with gravity eliminated

2

Movement against gravity

3

Movement against gravity and resistance but not full power

4

Normal power

5

*

Medical Research Council (1976).

Table 16.2 Neurological injuries Single nerve root damage Brachial plexus avulsion injury

ASIA* definition

Grade

Frankel† definition

No motor or sensory function, including perianal area

A

Complete; no neurological function below injury level

Some sensory function retained, including perianal area

B

Preserved sensation only (even if only sacral)

More than half of muscles have power below grade 3‡

C

Motor activity (±sensation) but of no functional use

At least half of muscles have grade 3 power or more‡

D

Useful voluntary motor function

Normal motor and sensory function

E

Normal motor and sensory function

* † ‡

American Spinal Injuries Association (Kirshblum et al. 2011). Frankel et al. (1969). Medical Research Council scale (see Table 16.1).

Spinal cord syndromes Dorsal column: Uncommon; loss of proprioception and light touch; associated painful dysaesthesias. Central cord: Usually with cervical trauma; upper limb weakness more than legs; spinothalamic (pain/temperature) loss in arms and hands. Favourable prognosis. Anterior cord: Usually with thoracic injuries; dorsal column sensations preserved; loss of motor function and spinothalamic sensation; poor functional prognosis. Brown-Sequard: Penetrating injuries; ipsilateral weakness and dorsal column sensory loss plus contralateral spinothalamic sensory loss. Favourable prognosis. Cauda equina/conus syndromes Flaccid paralysis of legs, with loss of tendon reflexes; loss of bladder, bowel and sexual function. radiological investigation is determined by the clinical assessment. Two tools have been developed to aid such decisionmaking, these being the National Emergency X-Radiography Utilization Study (NEXUS) and the Canadian Cervical Spine Rule (Stiell et al. 2003). A number of classification systems have been devised, aiming at identifying stable and unstable injuries, and to provide a guide treatment and prognosis. The greater detail provided by CT and MR imaging permits more detailed systems to be devised, taking into account the morphology of bony injury, the state of the posterior ligamentous complex and the extent of any associated neurological damage. This last-mentioned consideration recognises that the presence of neurological injury means that stability is more likely than not to be compromised. The TLICS2 and SLIC3 systems generated numerical scores, to aid decisionmaking as regards the need for surgery (Vaccaro et al. 2005; 2007). With the development of increasingly sophisticated apparatus for internal fixation of the spine, additional anatomical details have been added to the classification criteria. The latest systems, albeit still subject to validation, generate a descriptive code, rather than a

2 3

Thoracolumbar Injury Classification and Severity Score. Subaxial Cervical Spine Injury Classification System.

Table 16.4 Plain radiographic/CT signs of spinal injury • Visible bony fractures • Loss of normal alignment, inspecting lines along: • Anterior bodies • Posterior bodies • Laminae • Tips of spinous processes. • Widening • Disc spaces • Facet joints • Spinous processes • Prevertebral soft tissue swelling numerical score (Tables 16.5 and 16.6). Whilst aiming to simplify and rationalise clinical practice, they still require the team managing spinal injuries to be well experienced in assessing patients and interpreting radiographs. Furthermore, established terms, such as ‘burst fracture’, ‘three-column injury’ and ‘Chance fracture’ (see below), are still very likely to be encountered in reports and correspondence. The spinal cord may sometimes be injured without any evidence of significant vertebral damage. This is referred to as spinal cord injury without obvious radiological abnormality (SCIWORA). It is seen most often affecting the cervical cord in children of primary school age or younger. This, presumably, is related to the increased laxity of juvenile ligaments, permitting a greater range of movement under acceleration/deceleration forces. Undiagnosed Chiari malformations and disc herniations are other, rare causes of SCIWORA.

Classification of vertebral injuries Injuries to the skeletal elements of the spine are usually considered in four main groups, by anatomical region: (1) suboccipital, (2) subaxial cervical, (3) thoracolumbar and (4) sacral. Within each group, individual types of injury may be stable or unstable, according to the extent of ligamentous damage. The following four sections of this chapter consider, in outline,

Classification of vertebral injuries

Table 16.5 AO spine thoracolumbar spinal injury classification system

Table 16.6 AO spine subaxial cervical spinal injury classification system

Morphology (categorisation is preceded by identification of the affected level)

Morphology (categorisation is preceded by the identification of affected level)

A Compression injuries A0 Minor: Not compromising integrity of column e.g. transverse or spinous process injury A1 Wedge compression: Single endplate involvement, posterior wall of body intact A2 Split: Coronal injury, both endplates involved but posterior wall intact A3 Incomplete burst: Posterior wall and single endplate involved A4 Complete burst: Posterior wall and both endplates involved

A Compression injuries A0 Minor: Not compromising integrity of column e.g. transverse or spinous process injury A1 Wedge compression: Single endplate involvement, posterior wall of body intact A2 Split: Coronal injury, both endplates involved but posterior wall intact A3 Incomplete burst: Posterior wall and single endplate involved A4 Complete burst: Posterior wall and both endplates involved

B Distraction (tension band) injuries B1 Classical Chance fracture (see main text), affecting bony but not ligamentous elements B2 Posterior ligamentous disruption + type A injury (above) B3 Hyperextension injury to anterior longitudinal ligament + disc or vertebral body C Translational (displacement) injuries (no specific subgrouping) Horizontal (anteroposterior or lateral) displacement, or vertical separation of adjacent bodies Neurology N0 Intact N1 Transient deficit, resolved by time of admission N2 Nerve root damage N3 Incomplete cord injury or any degree of cauda equine damage N4 Complete cord injury NX Unknown – sedated or unconscious Modifiers M1 Anterior bony elements intact but state of posterior tension band uncertain M2 Signifies the presence of (named) co-morbidity, relevant to any proposed surgery Source: Adapted from Vaccaro et al. 2013. commonly recognised types of vertebral column injuries. The reader should refer to more specialist works for further detail (Herkowitz et al. 2011), including the various subtypes that may occur.

Suboccipital injuries Skeletal anatomy at the craniovertebral junction provides for a wide range of movement of the head upon the neck but relies on the synovial joints and associated ligaments for stability. Because of the relatively wide diameter of the spinal canal at this level, the cord often escapes injury but when it is damaged deficits may be profound. Associated injuries to the vertebral artery may be overlooked as they may be clinically silent. Occipital condyle fractures usually result from high-energy impacts to the skull and are visualised on cranial CT bone windows, along with any associated head injuries. Three subtypes are described and the degree of stability varies. Neurological deficits range from minimal to quadriplegia. There may be lower cranial nerve deficits.

B Distraction (tension band) injuries B1 Fracture passing through posterior and anterior bony elements B2 Posterior ligamentous disruption + type A1-4 injury (above) B3 Hyperextension injury to anterior longitudinal ligament + disc or vertebral body C Translational (displacement) injuries (no specific subgrouping) Horizontal (anteroposterior or lateral) displacement, or vertical separation of adjacent bodies F Additional classification of facet injuries (‘BL’ designates a bilateral injury) F1 Stable, non-displaced fracture F2 Displaced fracture F3 Unstable, floating lateral mass F4 Subluxed, perched or dislocated facets Neurology N0 Intact N1 Transient deficit, resolved by time of admission N2 Nerve root injury N3 Incomplete cord injury or any degree of cauda equine damage N4 Complete cord injury NX Unknown – sedated or unconscious and therefore cannot be assessed Modifiers M1 Anterior bony elements intact but state of posterior tension band uncertain M2 Presence of a disc herniation having a bearing upon reduction of a facet dislocation M3 Coexisting ankylosing spondylitis, anterior or posterior longitudinal ligament ossification, or Forestier’s disease. M4 Presence of vertebral artery abnormality Source: Adapted from Vaccaro et al. 2016.

Atlanto-occipital dislocation was at one time a post-mortem diagnosis but improved recovery and resuscitation methods permit survival in some cases. It usually results from highenergy acceleration/deceleration forces, as occurr in road traffic collisions. It is more common in children. Many victims display lower cranial nerve or brainstem signs, as well as profound cord deficits, often requiring ventilator support. The injury is unstable, requiring external or internal fixation.

169

Spinal injuries

170

Figure 16.1 C1/C2 rotatory subluxation. Axial CT images through the atlanto-axial articulation. (a) The black line passes through the transverse processes, foramina transversarium and lateral masses of C1. Normal alignment is maintained with respect to the skull base, as indicated by the position of the mandible (arrows). (b) In contrast, the black line passing through the body and foramina transversarium of C2 is rotated with respect to the mandible (arrows) and skull. (c) The CT slice passing between (a) and (b) reveals the degree of rotatory displacement of C1 on C2. Atlanto-axial rotatory subluxation (Figure 16.1) can result from trauma or develop spontaneously in rheumatoid arthritis. It is sometimes seen in children, following an upper respiratory tract infection.4 It presents with local pain and torticollis although symptoms in some cases may be severe. Spontaneous reduction may occur; otherwise, intervention may be needed. Jefferson fractures (Figure 16.2) result from axial loads applied to the head, causing bilateral or multiple fractures of the arch of the atlas. Cord injury is uncommon as the spinal canal is wide at this level. Most cases are stable but transverse ligament damage makes the injury unstable and requiring immobilisation or fixation.

Odontoid fractures (Figure 16.3) have a bimodal age distribution, being seen following simple falls from standing in the elderly but following road traffic collisions or other high-energy impacts in children. Flexion or extension forces will cause

Figure 16.2 Jefferson fracture. Axial CT image through C1. (A) Odontoid peg; (B) foramina transversarium. The long arrows indicate the position of fractures in both the anterior and posterior arches.

Figure 16.3 Odontoid peg fracture (type 2). CT sagittal reconstruction of the cervical spine revealing a fracture through the base of the odontoid process (arrow). This has resulted in marked forward angulation of the peg in relation to the body of C2. The narrowed disc spaces and osteophytes seen on the vertebral bodies below indicate that this is the spondylotic neck of an older patient.

4

Grisel’s syndrome.

Classification of vertebral injuries

Figure 16.4 Anatomy of the hangman’s fracture. This model of the cervical spine shows the vertically aligned C2/3, C3/4, C4/5 and C5/6 facet joints, indicated by four pairs of parallel lines. Above these, the C1/2 articulation sits well anteriorly. The bar of bone (double-headed arrow) passing between this joint and the C2/3 articulation below constitutes the pars interarticularis of C2, which breaks bilaterally in the so-called hangman’s fracture. anterior or posterior displacement. The fracture may occur at the tip of the dens,5 at its base or through the body of C2. Hangman’s fracture usually results from axial loading in extension,6 such as occurs in road traffic incidents and diving accidents. Spondylolisthesis with dislocation of C2 on C3 results from bilateral pars interarticularis fractures (Figure 16.4).

Subaxial cervical injuries Burst fractures result from axial loading forces and commonly affect the lower cervical spine. There can be significant bony encroachment into the spinal canal. Axial loading with flexion may produce a less severe anterior compression fracture, which is usually stable. Teardrop fracture (Figure 16.5) is an innocent-sounding term, describing the small chip of bone seen, radiologically, broken off the anterior, inferior part of the body. The injury is, however, severe and unstable. This type of injury may be caused during a diving accident, with forceful flexion with posterior 5 6

Not to be confused with an os odontoideum. Judicial hanging, in contrast, causes hyperextension with vertical distraction, usually producing a complete spinal cord injury. One detailed study demonstrated a relative paucity of cervical fractures following hanging. Only 3 of 34 were of the typical hangman’s fracture and 3 were asymmetrical fractures involving C2 (James and Nasmyth-Jones 1992). A further anthropological review of six cases showed fractures of C1, C2, C3 and C5, as well as the C2 vertebral body (Spence et al. 1999). It appears that the effects of judicial hanging are variable although, even without bone or cord injury, death is usually rapid, sometimes due to carotid artery compression, reflex cardiac arrest, or venous or airway obstruction.

Figure 16.5 Teardrop fracture. CT sagittal reconstruction of the cervical spine, revealing near-normal overall alignment of the vertebral bodies. The prominent ‘teardrop’ fragment, broken off the anterior/inferior part of the C4 body reveals, however, that considerable flexion-compression force did, at the moment of impact, drive the C4 body backwards. As a result, there is also some residual posterior displacement of C4 body, in relation to C5 below. ligament disruption and axial loading. Neurological damage is common and stabilisation is required. Facet dislocations (Figure 16.6) result from flexion forces. If combined with rotation, the dislocation is likely to be unilateral but hyperflexion alone can produce bilateral dislocation, which is a more severe injury and very likely to cause cord injury. Distraction disc injury results from an impact to the head or face, such as following a fall. Hyperextension damages the anterior longitudinal ligament and disc. Middle-aged and elderly people with spondylotic necks are particularly vulnerable. The absence of bony damage may lead to the injury being missed and MR imaging may be required to visualise the damage. The clinical profile is that of a central cord syndrome, where the upper limbs are affected more than the legs (see Table 16.2). Spinous process avulsion fracture, commonly referred to as ‘clay-shoveler’s fracture’, results from forced paraspinal muscle contraction, although a similar injury can be caused by forced hyperextension or even hyperflexion injury to the neck.

171

Spinal injuries

172

Figure 16.6 Subaxial cervical subluxation. (a) Plain lateral radiograph of the cervical spine. The shadow of the shoulders partly obscures the C7 and T1 vertebrae but close inspection reveals anterior displacement of the C6 body with respect to C7 (arrows). (b) T2 sagittal MR images of the same injury, showing the displacement more clearly (arrow), with significant narrowing of the spinal canal posteriorly and ‘pinching’ of the spinal cord.

Thoracolumbar injuries These most often involve the thoracolumbar junction because the rib cage normally supports the mid and upper dorsal spine. Usually, considerable forces are involved, such as with falls and vehicular collisions. Polytrauma is common. Wedge compression fractures are relatively low-energy injuries, commonly seen in elderly people with osteoporotic bones. The anterior wall of the vertebral body, or its upper or lower endplates, gives way but the posterior parts of the body and the posterior elements of the spine are spared, so the injury is stable and neurological deficits are rare. Burst fractures (Figure 16.7) involve, additionally, the posterior parts of the vertebral body and there may be retropulsion of bony fragments into the spinal canal. The most severe forms include damage to the posterior spinal elements, causing instability and neurological deficits. Chance fractures7 typically result from the upper torso being thrown forwards rapidly whilst the pelvis is restrained, i.e. a seat belt injury.8 A ‘hinging’ mechanism results in tensile forces causing either a fracture through the spinous process (in the classical Chance fracture) or rupture of the interspinous and supraspinous ligaments. At the same time, the axis of rotation of the hinge, which lies just in front of the vertebral column, causes disruption of the intervertebral disc, or a horizontal fracture, slicing through the vertebral body. Between these two regions, the facet joints may dislocate. The injury is potentially but not inevitably unstable. Fracture-dislocations in the thoracic spine come about only as a result of considerable forces. There is damage to both anterior and posterior parts of the vertebral body and adjacent ligaments, plus disruption of the posterior ligament complex. There is horizontal displacement (translation) of the upper on 7

8

George Quentin Chance, British radiologist who first described this type of injury in 1948. More specifically, a lap seat belt injury, now seen less frequently with three-point fixation of in-car restraints.

Figure 16.7 Burst fractures. 3D CT reconstruction of the thoracic vertebral column. Distraction forces acting upon the posterior tension band, combined with flexion-compression forces applied to the vertebral bodies, has created these three-column injuries. There are burst fractures of two non-contiguous bodies (short arrows) and fractures through the posterior bony elements, seen most clearly at the lower level (long arrow). the lower vertebral body. These injuries are therefore highly unstable and neurological damage is common. Minor thoracolumbar fractures include avulsion injuries to the spinous or transverse processes, commonly the result of sudden, severe muscular effort, akin to the clay shoveler’s fracture (above). Laminae or articular processes may also sometimes be fractured in isolation.

Sacral fractures These injuries are usually seen in conjunction with damage to other parts of the pelvic ring, as a result of high-energy falls or collisions (Figure 16.8a). Occult bleeding, from damaged internal iliac vessel, may occur. Sacral fractures also form part of the spectrum of osteoporotic injuries, following falls in the elderly and may also be seen as stress fractures in athletes (Longhino et al. 2011). Fractures of the sacrum itself are classified according to whether they lie lateral to, along the line of, or medial to the neural foramina. They may lie transversely or vertically or may be multiplanar. They may be simple or comminuted. Laterally placed fractures are not associated with any neural damage. Fractures through one or more foramina may cause sacral nerve root deficits, affecting bladder, bowel and sexual function. A medially placed fracture will involve the sacral spinal canal and very likely cause damage to the lower sacral rootlets. Management is directed for the most part at the associated pelvic injuries, for which internal fixation is sometimes used. Fixation of sacral fractures is not carried out very often,

Neuropathology of spinal cord injury

Figure 16.8 Sacral fracture. (a) 3D CT reconstruction of the pelvis, revealing marked disruption of the pubic and ischial bones ­anteriorly (short arrows) and a fracture through the ala of the sacrum on the right side (long arrow). A fracture through the neck of the left femur is also evident. (b) Plain radiograph of the same injury following internal fixation. A considerable amount of apparatus was required to stabilise this injury against the forces that normally pass through the lumbosacral junction. unless there is major disruption of the lumbosacral junction, in which case complex internal fixation constructs are needed (Figure 16.8b).

Associated injuries Depending upon the mechanism of injury and energies involved, spinal trauma may be associated with injuries to the head and facial skeleton, the upper limbs and brachial plexus, the chest, abdomen, pelvis and the lower limbs. In addition, carotid or vertebral artery damage may result from sudden flexion/extension movements of the head. Vertebral artery injuries may also be seen with cervical dislocations and transverse process fractures; most are clinically silent.

Figure 16.10 Spinal cord showing acute injury with epidural haemorrhage. other than petechial intramedullary haemorrhages, although there may be associated subarachnoid haemorrhage. More severe damage occurs as a result of compression with vascular compromise or direct contusional injury (Smith et al. 2015) (Figure 16.10). Central haematomyelia, in the form of a large space-occupying haemorrhage with distension of the cord, may be seen, albeit rarely. In cases where there is significant trauma, for example the result of a penetrating injury, the cord may show gross physical disruption or laceration. Swelling and softening of the cord develops during the first 24 hours after initial injury. Necrosis of tissue is visible, beginning centrally with extension into white matter, with variable amounts of haemorrhage and loss of grey-white differentiation; there is usually subpial spraining. Axonal swellings are seen, most extensively at the site of damage but also extending distally and proximally (Figure 16.11). Acute inflammation begins, with leucocyte infiltration, followed by macrophitic removal of

Neuropathology of spinal cord injury In deaths that occur immediately or very soon after injury, indications of spinal cord damage may be evident only in adjacent structures e.g. fractures, dislocations or ligamentous tears (Figure 16.9). The cord itself may show minimal findings,

(a)

(b)

Figure 16.9 Macroscopic appearance of acute spinal cord injury with fracture-dislocation T2 following a fall from a height.

Figure 16.11 Axonal swellings in a high cervical fracture of a few days survival.

173

174

Spinal injuries debris, peaking at 3–7 days. Reactive astrocytes with gliosis then supervene over the next several days and weeks. Later, tracts damaged at the injury site show Wallerian degeneration in both rostral and caudal segments, with axonal degeneration and demyelination. The grey matter at the level of injury may become cavitated with hyaline thickening of blood vessels and collagenous fibrosis. In the absence of meningeal laceration, intradural scarring may be minimal but in cases where there has been breach of the dura and extensive cord disruption, connective scar tissue forms and is frequently extensive. Late deaths, occurring months or years after injury, may be the result of complications, such as chest infection or pulmonary emboli, or chronic renal tract infection with renal failure. It may then be necessary for the pathologist to demonstrate a direct link with the original injury, in any legal proceedings. Penetrating injuries of the spinal cord include stabbings and gunshot wounds. The general features of gunshot wounds are covered in Chapter 6, together with the ballistic characteristics. Injury to the spinal cord may be caused directly by the bullet, or may result from high-energy shock waves being transmitted to the cord. The level of cord injury may therefore not correspond with any external wound. Findings at post-mortem examination will depend on the time elapsed since the injury. Early examinations may show focalised haematoma formation, as well as cord disruption and haemorrhage. Additional damage may occur as a result of vascular involvement, with the formation of haematomas or infarction. Later deaths show the axonal degeneration and demyelination described above.

Table 16.7 Indications for surgery • Failure of traction to reduce facet dislocation • Neurological deterioration from baseline • Non-union after conservative management e.g. odontoid peg fracture • Unstable injury – optional • Persisting neural compression with an incomplete cord injury – controversial • Open injuries requiring debridement Table 16.8 Pathophysiological reactions to spinal cord injury • Structural disruption • Cells • Axons • Blood vessels • Vascular events • Loss or arterial autoregulation • Venous thrombosis • Biochemical reaction • Free radical formation • Release of neurotransmitters • Altered ionic membrane potentials • Release of autolytic enzymes • Inflammatory responses • Cellular destruction • Tissue repair • Axonal sprouting = abortive regeneration • Glial-collagen scar formation

Management of vertebral injuries Stable fractures allow the victim to be mobilised and pain control is the main requirement. In addition to appropriate levels of analgesia, temporary external support may be helpful be this some form of collar or a brace for the trunk. Unstable injuries need to be supported until natural healing occurs. This may be by external or internal stabilisation. The former includes bed rest with strictly controlled turning regimes, or with rigid external fixation apparatus. A wide range of well-designed and quality engineered apparatus now makes internal fixation increasingly used in more affluent countries. Conservative management remains, however, an effective alternative to fixation when such facilities are unavailable, or there are other reasons for wanting to avoid surgical intervention (El Masri and Kumar 2017). In the neurologically intact patient, who will not require lengthy in-patient rehabilitation, internal fixation does allow for early mobilisation and shorter stays in hospital. The risks of complications, commonly seen with prolonged bed rest when an expert dedicated multidisciplinary team is not available, are also lessened. These include thromboembolic events, chest infections and the development of pressure sores. If surgery is undertaken, the outcome may be better with early rather than delayed intervention (Fehlings et al. 2012). Surgery is also favoured for incomplete, as opposed to complete cord injuries, on the basis that there is function to preserve in the former group (Table 16.7).

Early management of spinal cord injuries A series of pathophysiological processes begins immediately after the spinal cord is damaged (Table 16.8), (El Masri 2006). The overriding need is to prevent the injured cord from being further damaged. In addition to the stability of the spinal column being secured, the cord must be protected from ischaemic

insults and secondary damage due to systemic hypoxia-hypotension. The latter is particularly relevant because the acute neural failure,9 which develops immediately after any significant spinal cord injury includes interruption of the intra-spinal sympathetic pathways. The resultant loss of vasomotor tone produces a state of low-resistance circulatory shock. Thus, the cord injury causes a state of shock to develop and the cord itself is then vulnerable to the effects of this acute drop in blood pressure. Many body systems are affected by a spinal cord injury and patients are vulnerable to developing a variety of complications (Table 16.9).

Long-term complications of spinal cord injury A number of problems may arise subsequently, including the development of neuropathic pains and painful, involuntary muscle spasms. Visceral disturbances, such as constipation or bladder infections, also occur. Patients with intercostal muscle paralysis remain at risk of developing chest infection. Osteoporosis is a consequence of paralysed limbs, affecting particularly the long bones adjacent to the knee joint (Battaglino et al. 2012). Heterotopic muscle calcification is not uncommon, often developing at an early stage (van Kuijk et al. 2002). Other long-term complications include traumatic neuromas in the regions of injured nerve roots. Autonomic dysreflexia is a potentially dangerous condition, caused by aberrant responses to visceral stimuli. It is seen 9

Commonly referred to as spinal shock.

Special categories of spinal injury

Table 16.9 Early needs of paraplegic patients Chest: Infections remain a significant cause of death in the early stages after injury. Ventilator dependency brings with it special problems. Blood pressure: Loss of sympathetic tone makes victims vulnerable to postural hypotension. Bladder: Drainage by condom or indwelling catheter. Later replaced by intermittent self-catheterisation, or a suprapubic catheter. Bowels: Attention to diet; use of laxatives and suppositories. Manual evacuation may be required. Skin: Regular turning to prevent pressure ulcerations, which take a long time to heal once established. Gastric protection: Peptic ulceration follows physiological stress, made more likely if steroids are used. Venous thromboembolic prophylaxis: Lower limb immobility, together with the catabolic response to trauma, put the patient at high risk of thrombosis and emboli.

Figure 16.12 Post-traumatic syringomyelia. T2 sagittal images of (a) the thoracolumbar and (b) the cervicothoracic spine. An old burst fracture is evident at T11 (a, white arrow). The spinal cord is abnormal throughout its length, being markedly distended throughout the whole of the thoracic canal, by a tense syringomyelia cavity (black arrows). The cavity extends into the cervical cord above, where it is less prominent.

Joints: Immobilised joints will stiffen rapidly if passive movements are not commenced early on. Equinus deformities of the ankle develop very readily and are difficult to reverse. Mental state: The psychological trauma of such a life-changing injury clearly requires sympathetic but positive management. Associated injuries: Head, pelvic, long bone and visceral injuries require specific management and will often delay the start of cord injury rehabilitation programmes. with lesions above the mid-thoracic level and results from the loss of the normal descending inhibition of spinal sympathetic reflexes. A bladder infection, development of a pressure sore or even an episode of constipation might trigger a sudden, severe bout of sweating and hypertension, with the potential of causing a cerebral haemorrhage. Syringomyelia is rare in the general population but is a common complication of spinal cord injury, affecting at least 1 in 20 individuals (Figure 16.12). Cavities form because cerebrospinal fluid (CSF) flow in the spinal subarachnoid channels becomes impeded by arachnoid scar tissue. This forms from the organisation of blood products, shed into the subarachnoid channels at the time of the original injury (Flint 2014). Discussion of the disturbances of spinal CSF dynamics underlying the filling of syringomyelia cavities is beyond the scope of this chapter (Flint 2019; Stoodley 2014). The importance of the condition is that it has the capacity to add to an already very significant neurological deficit. Surgery may be required for a propagating cavity that is causing additional motor deficits.

Special categories of spinal injury Ankylosing spondylitis – the rod-like and osteoporotic spine of the ankylosing spondylitis patient can break easily, after even moderate trauma (El Tecle et al. 2015). Injuries are typically caused by the result of a fall forwards (Figure 16.13). Fractures may occur at more than one level in the spine. Pathological fractures cause vertebral collapse due to bony infiltration by metastatic tumour. Spinal cord involvement is common.

Figure 16.13 Ankylosing spondylitis. Mid-sagittal, 3D CT reconstruction of the cervical and upper thoracic spine. The ‘bamboo spine’ appearances are evident in the thoracic region (short, solid arrows), together with marked ossification of the anterior longitudinal ligament in the neck (short, open arrows). A fracture is evident through the T1/2 vertebral bodies (long arrow).

175

176

Spinal injuries

Figure 16.14 Osteoporotic vertebral collapse. T2 sagittal MRI revealing osteoporotic collapse of three contiguous thoracic vertebrae (arrows).

Osteoporosis may cause vertebral collapse (Figure 16.14). Fortunately, in most such cases, the spinal cord is not compromised but the problem is usually that of pain. Whiplash injuries of the neck, resulting from rear-impact road traffic collisions, are common and are a frequent cause of compensation claims. Persisting pain and sensory symptoms are seldom accompanied by any hard neurological signs and standard radiological investigations are usually negative, beyond revealing any pre-existing spondylotic disease. There is likely to be joint capsule/ligamentous damage, beyond the resolution of routine MR imaging. They are effectively sprains but, clearly, more serious injuries need to be excluded before this conclusion is reached.

Paediatric aspects Children are most likely to sustain a spinal injury as vehicle passengers or pedestrians involved in road traffic collisions. They will often present with polytrauma and associated head injuries. Their large heads, underdeveloped axial musculature and relatively lax ligaments make them particularly vulnerable to acceleration/deceleration forces, causing wide excursions of neck movement with resultant cervical spinal injuries. Children are also more likely than adults to suffer spinal cord injury in the absence of obvious skeletal damage. Spinal injury caused by child abuse is covered in Chapter 15.

Figure 16.15 Josie Pearson MBE. Paralympic discus gold medal winner, 2012.

Conclusion Spinal cord injuries affect a relatively small number of people but frequently have life changing consequences. Modern care allows affected individuals to rehabilitate to an impressive degree (Figure 16.15) but the problem of the absence of regenerative capacity, in the mammalian central nervous system, still confronts modern medicine. Further neuropathological studies should undoubtedly help meet this challenge.

Acknowledgements The authors are indebted to Professor W S El Masri, former Director of the Midland Centre for Spinal Injuries, Oswestry, UK, and to Mr Antonino Russo, Consultant Spinal Neurosurgeon at the National Hospital for Neurology and Neurosurgery, London, UK, for their helpful comments and suggestions in reviewing the contents of this chapter.

References American College of Surgeons, Committee on Trauma. 2012. ATLS advanced trauma life support for doctors: student course manual, 9th edn. Battaglino, R.A., Lazzari, A.A., Garshick, E. and Morse, L.R. 2012. Spinal cord injury-induced osteoporosis: pathogenesis and emerging therapies. Curr Osteoporos Rep, 10, 278–285.

References Cripps, R.A., Lee, B.B., Wing, P., Weerts, E., Mackay, J. and Brown., D. 2011. A global map for traumatic spinal cord injury epidemiology: towards a living data repository for injury prevention. Spinal Cord, 49, 493–501. El Masri, W.S. 2006. Traumatic spinal cord injury: the relationship between pathology and clinical implications. Trauma, 8, 29–46. El Masri, W. and Kumar, N. 2017. Active physiological conservative management in traumatic spinal cord injuries – an evidence-based approach. Trauma, 19, 10–22. El Tecle, N.E., Abode-Iyamah, K.O., Hitchon, P.W. and Dahdaleh., N.S. 2015. Management of spinal fractures in patients with ankylosing spondylitis. Clin Neurol Neurosurg, 139, 177–182. Fehlings, M.G., Vaccaro, A., Wilson, J.R., Singh, A., Cadotte, D.W., Harrop, J.S. and Aarabi, B. et al. 2012. Early versus delayed decompression for traumatic cervical spinal cord injury: results of the surgical timing in acute spinal cord injury study (STASCIS). PLoS One, 7(2), e32037. Flint, G. 2014. Post-traumatic and post-inflammatory syringomyelia. In Flint, G. and Rusbridge, C. (eds), Syringomyelia, A Disorder of CSF Circulation. Springer, pp 167–184. Flint, G. 2019. Spinal cerebrospinal fluid dynamics. In Kirollos, R., Helmy, A., Thomson, S. and Hutchinson, P. (eds), Oxford Textbook of Neurological Surgery. UK: Oxford University Press. Frankel, H.L., Hancock, D.O., Hyslop, G., Melzak, J., Michaelis, L.S., Ungar, G.H., Vernon, J.D.S. and Walsh, J.J. 1969. The value of postural reduction in the initial management of closed injuries of the spine with paraplegia and tetraplegia. Spinal Cord, 7, 179–192. Herkowitz, H.N., Garfin, S., Eismont, F., Bell, G. and Balderston, R. 2011. Chapters 28, 75-78 & 80-84. In Rothman-Simeone. The Spine, 6th edn. Philadelphia, PA: Saunders-Elsevier. James, R. and Nasmyth-Jones, R. 1992. The occurrence of cervical fractures in victims of judicial hanging. Forensic Sci Int, 54, 81–91. Kirshblum, S.C., Burns, S.P., Biering-Sorensen, F., Donovan, W., Graves, D.E., Jha, A. and Johansen, M. et al. 2011. International standards for neurological classification of spinal cord injury. J Spinal Cord Med, 34, 535–546. Longhino, V., Bonora, C. and Valerio, S. 2011. The management of sacral stress fractures: current concepts. Clin Cases Miner Bone Metab, 8, 19–23. Medical Research Council. (eds) 1976. Aids to the Examination of the Peripheral Nerve System, 3rd edn. Her Majesty’s stationery office.

National Institute for Health and Care Excellence. 2016. Spinal injury: assessment and initial management. https://www.nice.org.uk/ guidance/ng41. Ross, J.C. and Harris, P. 1980. Tribute to Sir Ludwig Guttmann (3 July 1899 to 18 March 1980). Paraplegia, 18, 153–156. Smith, C., Margulies, S. and Duhaime, A.-C. 2015. Trauma. In Love, S., Budka, H., Ironside, J.W. and Perry, A. (eds), Greenfields Neuropathology, 9th edn, Vol 1. Boca Raton, FL: CRC Press. Spence, M.W., Shkrum, M.J., Ariss, A. and Regan, J. 1999. Craniocervical injuries in judicial hangings: an anthropologic analysis of six cases. Am J Forensic Med Pahol, 20, 309–322. Stiell, I.G., Clement, C.M., McKnight, R.D., Brison, R., Schull, M.J., Rowe, B.H. and Worthington, J.R. et al. 2003. The Canadian C-spine rule versus the NEXUS low-risk criteria in patients with trauma. N Engl J Med, 349, 2510–2518. Stoodley, M. 2014. The filling mechanism. In Flint, G. and Rusbridge, C. (eds), Syringomyelia, A Disorder of CSF Circulation. Berlin: Springer, pp 87–101. Vaccaro, A.R., Lehman, R.A. Jr, Hurlbert, R.J., Anderson, P.A., Harris, M., Hedlund, R. and Harrop, J. et al. 2005. A new classification of thoracolumbar injuries: the importance of injury morphology, the integrity of the posterior ligamentous complex, and neurologic status. Spine, 30, 2325–2333. Vaccaro, A.R., Hulbert, R.J., Patel, A.A., Fisher, C., Dvorak, M., Lehman, R.A. Jr and Anderson, P. et al. 2007. The subaxial cervical spine injury classification system: a novel approach to recognize the importance of morphology, neurology, and integrity of the disco-ligamentous complex. Spine, 32, 2365–2374. Vaccaro, A.R., Oner, C., Kepler, C.K., Dvorak, M., Schnake, K., Bellabarba, C. and Reinhold, M. et al. 2013. AOSpine thoracolumbar spine injury classification system: fracture description, neurological status, and key modifiers. Spine, 38, 2028–2037. Vaccaro, A.R., Koerner, J.D., Radcliff, K.E., Oner, F.C., Reinhold, M., Schnake, K.J. and Kandziora, F. et al. 2016. AOSpine subaxial cervical spine injury classification system. Eur Spine J, 25, 2173–2184. Van Kuijk, A.A., Geurts, A.C.H. and Van Kuppevelt, H.J.M. 2002. Neurogenic heterotopic ossification in spinal cord injury. Spinal Cord, 40, 313–326. World Health Organisation. 2013. Spinal cord injury fact sheet number 384. http://www.who.int/mediacentre/factsheets/fs384/en/.

177

17

Difficult areas in forensic neuropathology Homicide, suicide or accident Christopher Milroy and Helen Whitwell

In examining neuropathological aspects of a case, the pathologist will often be asked to give an opinion on the findings in the brain in the context of other pathological findings and the circumstances of a case. Such scenarios include falls, falls versus kicks, deaths in young children, sporting deaths and patterns of pathology in motor vehicle collisions. Inevitably certain areas of practice present more problems to the pathologist than others. The role of the pathologist is to assist the judicial system in determining what the manner of the death is. In some jurisdictions, notably the North American medical examiners’ systems, the determination of manner of death as well as cause of death lies with the pathologist. In other jurisdictions, such as the English coronial system or the Scottish system with the procurator fiscal, the pathologist does not decide manner of death, although clearly pathological evidence plays a pivotal role. Therefore, the identification of patterns of death that indicate accident, suicide or homicide is central to any death investigation. This chapter deals with areas that often cause problems for the pathologist. General features of injuries are covered in Chapter 6. The extent to which a pathologist will determine the cause of death and the manner of the death will depend on the information provided, and the full circumstances of a case may not be revealed until the legal proceedings have been completed. The detail of the information provided often depends on external agencies such as the police or health and safety executive, and most pathologists have a limited role in obtaining evidence not directly related to the autopsy. In such cases, caution should be exercised before coming to a firm conclusion. The pathologist is often asked to consider different scenarios, and one set of pathological facts may be consistent with more than one pattern of causation.

Homicide, suicide or accident Penetrating injuries Injuries may be self-inflicted with the intention of killing oneself, for psychiatric reasons or for other gain. The latter pattern of injuries is often seen where there is a claim that another person attacked the alleged victim. These injuries are typically caused by a sharp-bladed implement and are superficial incised wounds or abrasions. They often involve the face but not sensitive areas such as the eyes or lips. The injuries are frequently parallel and inflicted by the dominant hand in a direction that would be expected with a weapon held in that hand (Cordner 2003; Saukko and Knight 2016). Suicide by penetrating weapons other than firearms rarely involves the head, although, with some psychiatrically disordered people, bizarre injuries or methods of self-destruction may be used and case reports describing such events appear

from time to time. Accidental injury may also be seen with a pencil or other sharp implement. Most non-firearms deaths caused by penetrating injuries are going to be homicidal or accidental. The surrounding circumstances should help differentiate these cases, even if the pathology does not. The differentiation of self-inf licted gunshot wounds from homicide is a central role of the post-mortem examination. In the United States, firearms are the most common method of suicide for both men and women. In contrast, in the United Kingdom, suicide by firearm is a less common method of choice, and women very rarely kill themselves by firearms (Chapman and Milroy 1992). Any weapon may be used, but handguns are the most popular weapons in the United States. Typically the gunshot wound will be a contact wound (Figure 17.1), but intermediate wounds will occasionally be encountered. In considering whether or not a gunshot wound is self-inflicted, ancillary investigations can be very important. As well as examining for gunshot residue, blood spatter analysis may be helpful (Figure 17.2). In self-inflicted wounds, blood pattern analysis of the hands may provide important evidence (Yen et al. 2003). Multiple self-inflicted gunshot wounds to the head are well recorded and do not exclude suicide (Kury et al. 2000).

Road traffic deaths Deaths associated with transportation form an important part of forensic pathology practice, and they are the most significant cause of severe head injuries in modern society, accounting for over 50 per cent of severe head injuries in clinical and pathological practice in the United Kingdom. Many transportation deaths will present no significant problems in determining the cause of death and in reconstruction of the incident. However, in some cases, the pattern of injury is important in reconstructing the incident. This is most commonly encountered where the victim is a pedestrian and there are questions about what position the person was in when struck by a vehicle. A second problem case is where a body is found on the roadside, and the question raised is whether the person has been struck by a vehicle or died by some other means. A further problem is the run-over victim. Usually deaths of occupants of a vehicle pose no particular pathological questions, although one question that is occasionally raised is which of the occupants was driving when someone has been expelled from the car after the collision. Forensic post-mortem computed tomography (CT) combined with post-mortem examination has been shown to produce more detailed results in terms of collision reconstruction and cause of death (Chatzaraki et al. 2018).

Road traffic deaths

Figure 17.1 Contact shotgun wound to the temple.

Pedestrians Most pedestrians are struck by the front of the car (Clark and Milroy 2000). Most of the other impacts involve a pedestrian colliding with the side of a vehicle and a small percentage of incidents involve reversing. When the front of a car strikes an

Figure 17.2 Back spatter of blood in firearm suicide.

Figure 17.3 Bumper injuries to lower limbs. (Reproduced from Mason, J.K. and Purdue, B.N. 2000. Pathology of Trauma. Arnold.) upright pedestrian, there is damage to the impact site, which is usually on the legs, the bumper striking the lower limbs (Figure 17.3). These injuries are known as the primary impact site. The impact may cause bruising, abrasion or laceration of the skin. This is characteristically around the knee area but will obviously depend on the height of the victim and the car frontage. If the car is braking when it strikes the victim, the car bumper will be at a lower height than when stationary. If the impact is with sufficient force, the underlying bones of the leg may fracture. The injuries may be seen on the presenting leg, when the person has been struck side on. The other leg may have injury at a higher level as he or she walks or runs into the path of the vehicle. Where there are injuries to both legs at the same height, it supports the person being struck while facing forwards or being stuck from behind when the injuries are to the back of the legs. Deep dissection should be undertaken because bruising may be evident in the deeper tissues in the absence of surface injuries. When the tibia or femur is fractured, a wedge-shaped fracture site may occur, with the base indicating the site of impact and the front the direction of travel. Rotational movement may result in spiral fractures distant from the site of impact. It is important that the height from the heel of any injuries on the leg is measured, because this will allow comparison with any implicated vehicle. In a child, the primary impact site may be higher on the leg or at the level of the pelvis. Patterned abrasions or bruising may be encountered, but with modern car design, badges, mirrors and number plates are structured not to cause damage. Once struck, the body will be propelled on to the bonnet (hood) of the car, causing secondary injuries. These injuries may occur with impact on the bonnet, front pillars of the car windscreen (‘A’ pillar) or the windscreen itself, or by the body travelling on to the roof of the car. It is during this movement that the head and trunk are injured. The body of the victim will then typically fall to the ground, causing tertiary injuries, which may be more severe than the secondary injuries. External injuries to the head include abrasions, bruising and lacerations. The skull may be fractured, with the most common pattern of skull fracture in fatalities being a transverse hinge fracture caused when the head is impacted sideways. Linear vault fractures

179

180

Difficult areas in forensic neuropathology may occur and the facial skeleton may be fractured. Localised depressed fractures may be seen. Brain injury patterns will correspond to the degree of force applied to the head. Vehicular collisions are a major cause of diffuse traumatic axonal injury (dTAI). Extradural, subdural and subarachnoid haemorrhage may occur, as may coup and contre-coup contusions. In an analysis of 150 fatal pedestrian collisions, Ryan et al. (1994) identified 31 victims with evidence of a single head impact with the vehicle. By using sector-mapping techniques, they found that in lateral impacts (18 cases), there was a greater degree of damage than in occipital impacts (13 cases). In lateral impacts, greater cortical damage was seen in the inferior regions than in the superior or parasagittal regions of both left and right cerebral hemispheres. In the corpus callosum, more than 40 per cent of sectors were injured, as a consequence of shearing forces around the inferior edge of the falx cerebri and in the parasagittal regions. In occipital impacts, the distribution of cortical and white matter injury was symmetrical about the midline. There was less corpus callosum injury. Local effects caused by indentation of bone were not observed in this series of 31 victims. The frontal and temporal regions appear to be more sensitive to injury than other areas of the brain, explaining coup and contre-coup injuries in impacts in the sagittal plane (see also Chapter 10). The most common neck injury is fracture dislocation of the atlanto-occipital junction. This injury is inevitably fatal, and injury is most frequently encountered in motorcyclists but is seen in pedestrian fatalities. Other damage to the cervical vertebrae may be seen as well as fractures or dislocation to thoracic or lumbar spine (see also Chapter 16). Chest injuries are the most significant injuries in pedestrians after head injuries. Externally the same constellation of abrasions, bruising and lacerations may be seen. Abrasions to the trunk may be caused when the tertiary impact takes place and may consist of broad sheets of abrasion as the body moves along the road. These abrasions provide evidence that the body has travelled in a particular direction. Internally there may be more damage than is expected from the external injuries present. Rib fractures are common and may indicate the side of impact. They may puncture underlying lung resulting in contusional injury. The heart may be contused, with tears in the atria. The aorta may be ruptured, typically at the start of the descending thoracic aorta, although this injury is seen more often in vehicle occupants. Abdominal injuries are less common in pedestrian fatalities, the most common injury being lacerations of the liver. The spleen, kidney and mesentery may be damaged. Where a high-fronted vehicle, such as a van or lorry, strikes the pedestrian, he or she will be pushed forward. If remaining in the path of the vehicle, the person will be run over, presenting a further pattern of injuries. Pedestrians who have been struck by a car and suffered primary, secondary and tertiary injuries may also then be run over by another vehicle.

Run-over injuries Pedestrians may be run over because they have already been struck by another vehicle, or because they were lying in the road when initially struck. The pattern of injuries is different in run-over cases, although if the victim has already been struck or run over by more than one vehicle this may complicate the picture. With all road traffic deaths, the post-mortem findings should be correlated with any vehicle examination. It is often the case that an examination of the car that has allegedly struck

Figure 17.4 Sheet abrasions and grease to the lower back from being dragged underneath a vehicle. someone will show damage indicating the position of the victim. Furthermore, blood, tissue and other trace evidence may be recovered from the car to indicate whether the victim has been run over or struck in another position. Where a person has been lying in the road when struck, he or she is typically found to be intoxicated with alcohol or some other drug, but it is important to try to exclude natural disease or a previous attack by another person. If there is a possibility that the person was already dead when hit, a useful test is to perform a fat stain on the lung. The presence of fat embolism supports an active circulation at the time that the run-over injuries were inflicted. External examination typically reveals broad areas of abrasion indicating dragging of the body (Figure 17.4). There may be oil and grease from the underside of the car. Damage from the under-surface structures of the car may cause further abrasion, bruising and laceration. Tyre imprints may be present. Where no dragging has taken place, external injury may be slight. When the wheels of a vehicle run over the body, the head may not be significantly damaged, the principal areas of injury being to the chest and abdomen. If the head is run over, it will show crush injury patterns. Abrasions, bruising and lacerations may occur and the skull will be fractured. The brain may show direct damage but not the pattern of injury seen in secondary or tertiary impact injuries. Extensive thoracic and abdominal injury may indicate severe crushing, with extensive damage to the chest wall, lungs, heart and major vessels. The abdominal organs may be lacerated, with pulping of the liver and spleen seen in some cases. The vertebral column may show extensive damage, as may the pelvis. The bladder may be ruptured. The mesentery is often lacerated.

Vehicle occupants Post-mortem examinations on occupants of cars usually present less controversy than those on pedestrians, although, where occupants have been ejected from the car, questions over who was driving may be important. In these cases, seat-belt injuries (Figure 17.5) can be useful in determining the seating positions (Milroy and Clark 2000). Injuries to the driver may be caused by the steering wheel and foot injuries may be found where the

Falls pedestrian deaths, age range 15–86 years, Karger et al. (2000) found that brain damage did not depend on impact velocity. In their study, they found 64 spinal fractures in 38 victims. Nearly half of all fractures were in the cervical spine. Spinal fractures were seen with impact speeds above 27.5 km/h, with 36 of 38 victims having them at 45 km/h and in every case above 67.5 km/h. They concluded that, if there was no spinal fracture, the speed was below 70 km/h (43.5 mph) and probably below 50 km/h (31.25 mph). They also showed that aortic and inguinal ruptures and dismemberment correlated with higher impact speeds. However, in view of the variation in possible speeds with possible injuries, dogmatic statements on speed should not be made. Police traffic investigators are often very accurate in determining speed.

Falls Figure 17.5 Seat-belt marks to car passenger (right-hand drive). driver has been braking hard. The front-seat passenger may be injured when he or she strikes the front dashboard. Engine intrusion may occur in high-speed collisions, although modern car design structures the car so that crumple zones absorb the energy of an impact to decrease this event and increase occupant safety. Pathologists should be aware that injury to front-seat passengers may result from unrestrained rear-seat passengers moving forwards after a collision. As well as injury patterns, forensic examination of the vehicle is likely to provide important evidence as to who the driver was. The use of seat-belts and air bags has done much to reduce the death rate following vehicular collisions, but head injury remains an important cause of death in car occupants. Airbags themselves produce injuries to the head, including facial and neck abrasions as well as eye injuries. Children are particularly prone to head and neck injuries if placed in the rear position on the passenger side (Milroy and Clark 2000). In addition, fatalities are also reported where a child is restrained using an adult seat belt (Cooper et al. 1998) and where occupants are unrestrained or out of position (Boyd 2002).

Falls are a common problem in forensic practice. The pathologist may need to assess where there are suspicious findings in an individual who is found at the bottom of a flight of stairs or at the foot of a building; assessment of possible defence injuries, injuries caused by hanging on to a ledge, as well as those that could have been caused by an attack, should be done. A history of underlying disease such as ischaemic heart disease or epilepsy may also aid interpretation, as will evidence of use of drugs or alcohol. Infant head injury is covered in Chapter 15.

Falls from standing

Head and spinal injuries are a common cause of death in this group. Motorcyclists have a higher incidence of skull fractures than car occupants, including linear, hinge and rarely ring fractures. The pattern of injury in both types will be modified depending on the circumstances, including the presence or otherwise of bicycle helmets (Macpherson and Macarthur 2002; Wardlaw 2002) (see also Chapter 7). More recently studies from New York City have identified increased protectivity for head injury in urban cyclists (Sethi et al. 2015).

Falls leading to head injury are a common occurrence and differentiation between accidental and deliberate infliction is important. The presence of more than one laceration to the head should raise suspicion, although there may be an explanation, e.g. multiple falls in an intoxicated individual. A clinical analysis of 189 elderly patients (over 60 years of age) admitted to hospital (Nagurney et al. 1998) revealed that falls from a standing height (76 per cent) were more common than falls on stairs (19 per cent) or from a height (5 per cent). However, abnormalities on CT, identified in 16 per cent, were more common in stair falls (42 per cent) and from a height (40 per cent), with the most common patterns of pathology being cerebral contusions (38 per cent) and subdural haematoma (33 per cent). In fatal falls from a standing height, the typical pathology seen is an occipital impact site, contre-coup pattern of contusions and subdural haematoma formation. More recent studies confirm that ground level falls are the most frequent mechanism of head trauma in older people, with transient loss of consciousness a significant underlying cause (Timler et al. 2015). In a retrospective analysis of fatal falls of 291 post-mortem examinations (Thierauf et al. 2010), alcohol was a significant factor in the pattern and increased incidence of injury in ground level falls; whereas in falls from a height, alcohol did not influence the injury pattern.

Vehicular collision investigation

Falls from a height

The following is a question that may be asked of the pathologist: can a correlation be made between the injuries present and the speed of the vehicle? Although there is a general rule that the higher the speed, the greater the likelihood of significant injury, low-speed impacts may be associated with a fatal outcome. Where a limb is amputated or transection of the torso occurs, (Zivot and DiMaio (1993) found speeds were above 88 km/h (55 miles/h or mph). In a study of 47 standing or walking

High falls are a common method of suicide in some countries. In Singapore, where falls from heights are common, an analysis of 603 post-mortem examinations performed on falls from more than 3 m, mostly from 20 to 40 m, revealed the head and face as the primary impact site in 67 cases (11.1 per cent) (Lau et al. 1998). Over one-third of victims had cerebral lacerations, with a subset of about 20 per cent showing massive craniocerebral destruction (Lau et al. 2003). Subarachnoid and subdural

Motorcycles and pedal cycles

181

182

Difficult areas in forensic neuropathology haemorrhages were seen in a third of these cases. Extradural and intracerebral haemorrhages were uncommon, with cerebral contusions seen in 10 per cent of cases, presumably because of the rapidity of death in these cases. Beale and colleagues (2000) examined 341 people who had suffered a fall from more than 2 m in Scotland. Of these, 89 died, with mortality significantly higher in falls of greater than 10 m. Skull fractures were found in 59 victims with cranial pathology consisting of extradural haematoma, subdural haematoma, subarachnoid haemorrhage, cerebral contusions and intracerebral haematoma. The assessment of traumatic axonal injury, taken in isolation, does not show any significant difference between individuals falling from their standing height or those falling from a greater height (Abou-Hamden et al. 1997). Homicidal falls from a height are a rarity, as published in a series from Singapore (Lau 2004). He describes five cases of pure homicide and nine episodes of dyadic death from a total of 533 homicides and 3963 fatal falls from a height. The majority of victims were children.

Falls down stairs Falls down a flight of stairs are an important cause of head injury (Ragg et al. 2000). In an analysis of 51 deaths after falls down stairs by Wyatt and colleagues (1999), the peak age range was 71–80 years, with most victims being over 50 years of age. Males slightly outnumbered females; 28 victims had alcohol in the blood, with 20 victims having a blood alcohol in excess of 80 mg/100 mL; 35 had brain or brainstem damage, and 8 cervical injuries. The findings of head injury and an association with alcohol intoxication are also shown in a review by Preuss et al. (2004) in their series of 59 cases over 11 years. An unpublished analysis of 57 fatalities from the authors’ old department revealed a bimodal age distribution of victims. One group was 30–60 years of age, and there was a second peak of 70–90 years of age. Natural disease was common in this latter group. Of these 57 cases, 15 died of natural disease and 42 of injuries caused in the fall. Of this latter group, 30 died of brain injury and 8 of neck injuries. In the younger group, alcohol intoxication was a significant finding. Evidence of frontal impact was seen in 42 per cent, temporoparietal impact in 46.5 per cent and occipital in 30 per cent of cases. Lacerations were seen in half the cases. The most common site for lacerations was the occiput (Figure 17.6), with other sites being the frontal and parietal scalp and the face, including the supraorbital ridges. Most lacerations were single, although in four cases, there were two lacerations, and in one, there were three lacerations. More than one area of scalp damage may therefore be present, caused as the victim tumbles down the stairs. Of the victims, 27 had a skull fracture, 25 of these consisting of a linear non-displaced fracture. No depressed skull fractures were seen. The temporoparietal skull and occiput were the most common fracture sites. Frontal fractures were unusual. Subdural and subarachnoid were the most common type of intracranial haemorrhage. The principal pattern of brain injury was frontotemporal contusions with only one case of occipital contusions. The cervical spine was fractured in nine cases, the thoracic spine in five cases and the lumbar spine in one case. Rib fractures were identified in 12 cases. The position in which the body ends up at the bottom of the stairs may be very variable, and caution must be exercised in interpretation of the final position, in relation to the fall. The victim may still have some capacity to move after the fall, and

Figure 17.6 Occipito-parietal lacerations from a fall down stairs. the tumbling body can come to rest in unusual positions. An apparently unusual position should not lead to the automatic conclusion that the person could not have fallen down a flight of stairs. However, multiple lacerations should raise concerns of an attack with a blunt weapon. Elsewhere on the body, bruising may be found on the trunk and limbs, often with an abrading appearance – the skin overlying the bruise showing an abrasion in the direction of travel. Bruising with abrasion is commonly over body prominences, such as the iliac crest, and over the cervical prominence. However, these injuries to the trunk may not be as substantial as one might expect. Again, caution must be exercised in over-interpreting the relative absence of injuries.

Kicking versus falling In the medico-legal setting, the pathologist is not infrequently asked to distinguish between different types of assault by different assailants – a not uncommon scenario is one individual punching causing a fall to the ground and the other kicking him. dTAI is most likely to occur from a biomechanical aspect when the moving head is rapidly decelerated on the ground. Whether it can be produced by kicking alone without impact is unclear, although lesser degrees of traumatic axonal injury occur (see also Chapter 11). There are limited cases fully documented in the literature as the exact circumstances are frequently unclear and detailed neuropathological studies few. It may be difficult to be confident in any given situation as to the relative contribution of differing assaults to the neuropathological findings (Geddes et al. 2000). Correlation with external and scalp injuries may aid.

Sudden death in head injury Sudden death in the context of head injury is relatively uncommon contrary to the general perception of both non-medical and at times medical personnel. Clearly, where there is massive craniocerebral disruption such as after significant fall from a substantial height or a gunshot injury, death is likely to be

Contact sports

Figure 17.7 Microhaemorrhages in the corpus callosum in a case of sudden death in head injury.

instantaneous. With less severe injury, there is usually a period of survival with, for example, dTAI. Death occurring rapidly in this context is usually related to diffuse vascular injury (see Chapter 11). Sudden immediate death may also occur in the forensic setting in the context of basal traumatic subarachnoid haemorrhage – this is covered in detail in Chapter 9. Sudden death has been reported after a blow to the back of the neck (Davis and Glass 2001), where the likely mechanism of death was neurogenic shock caused by concussion of the particularly sensitive area at the junction of the medulla and spinal cord. This has also previously been reported by Freytag (1963) who reviewed the post-mortem findings in head injuries caused by blunt forces in 1367 deaths. In her series, 6 per cent of cases were concluded to result from ‘concussion’. In addition, a number of cases have now been reported of sudden death with head injury in association with alcohol intoxication (see also Chapter 19). The post-mortem findings vary in these in that some show evidence of significant trauma. However, in other cases, evidence of traumatic injury such as skull fractures or brain injury is minimal (Ramsay and Shkrum 1995; Milovanovic and DiMaio 1999, authors’ personal experience). A proposed mechanism is concussive brain injury which may produce prolonged post-injury apnoea in combination with alcohol (Zink and Feustel 1995). Pathological examination in these cases may reveal small haemorrhages, particularly in the corpus callosum and sometimes in the periventricular region or brainstem (Voigt 1981, author’s personal experience) (Figure 17.7). As a result of the short time interval between injury and death, amyloid precursor protein staining and other markers are negative.

SIDS and suffocation This is a difficult area in forensic pathology with serious implications in the medico-legal field. The term ‘sudden infant death syndrome’ (SIDS) became widely accepted in the 1970s and was defined as the sudden death of an infant or young child which is unexpected by history and in whom a thorough post-mortem examination fails to demonstrate an adequate cause of death (Saukko and Knight 2016). The incidence of true SIDS has fallen, partly as a result of the ‘Back to sleep’ campaign. There is now an increased risk of sudden death in association with co-sleeping, sleeping on a sofa, and where there is exposure to tobacco smoke (Blair et al. 1996, 1999).

The major difficulty for the pathologist is distinguishing a true SIDS from upper airway obstruction, as in suffocation. This topic is covered extensively in the literature (Byard and Collins 2014). Suspicion should be raised by facial petechiae, including conjunctival (although these can also be seen in SIDS and may not be evident in cases of suffocation) and facial or mouth bruising and/or abrasions. Attempts at resuscitation may complicate the issue (Hanzlick 2001; Leadbeatter 2001). The presence of extensive areas of intra-alveolar haemorrhage in the acute situation has been described (Yukawa et al. 1999). Bleeding from the nose is not infrequently seen. Substantial numbers of haemosiderin-containing macrophages have been described in repeated upper airway obstruction (Beecroft and Lockett 1997; Hanzlick and Delaney 2000; Milroy 1999). It appears that they are an uncommon finding in true SIDS and should lead to a search for their cause, which includes infections, pulmonary haemosiderosis and bleeding disorders. It should be recognised, however, that the above findings are non-diagnostic in themselves (Krous et al. 2006). It is essential to have a full evaluation of the history and scene, as well as detailed pathology. Increasingly metabolic or cardiac defects are being identified in such cases. As regards the neuropathology abnormalities of serotonin metabolism have been identified (Kinney at al. 2009). More recent reviews of the topic (Bright et al. 2018; Paine et al. 2014) outline work undertaken on other biomarkers; however, currently there are none which identify ‘at risk’ infants. Suffocation in the adult as a cause of death is not common. Most cases are in the elderly or infirm who not infrequently have other disease such as carcinoma. In addition, drugs such as morphine may be involved. Rare cases of homicide, such as in head injury, may be complicated by terminal suffocation. Neuropathological findings of suffocation in both the infant and adult are not well described. Survivors of attempted upper airway obstruction may show changes of hypoxic-ischaemic damage of the global type with widespread cortical involvement. Acute changes are non-specific. The weight of the brain may be increased in infants, presumably as a result of congestive swelling (Bamber et al. 2016). This may not be the case in adults however, as examination of organ weights in cases of asphyxiation as compared with trauma cases does not show any significant differences in brain weight across the various groups (Hadley and Fowler 2003). However, in a further series, brain weights of suicidal hangings were significantly higher than in those dying of overdose (Hamilton and McMahon 2002). Petechial haemorrhages may be seen histologically, often perivascular in the cortex; however, they cannot be taken as diagnostic of upper airway obstruction in the absence of other evidence as they may also be a feature of other causes of sudden death.

Contact sports Boxing has raised controversy over many years, in relation not only to acute injury, but also to the long-term chronic effects. Contact sports is covered in detail in Chapter 14.

Diving Investigation of diving deaths is complex. There has been an increase in deaths occurring during recreational diving with

183

184

Difficult areas in forensic neuropathology

Figure 17.8 Spinal cord in a case of acute decompression syndrome with dorsal column involvement. (Courtesy of Dr I M Calder.) a downturn in occupational deaths. Detailed discussion about the investigation of these is covered in detail by Ross in The Pathology of Trauma (Ross and Grieve 2000). Decompression illness is seen in divers who fail to control their rate of ascent. It comprises pulmonary barotrauma and cerebral gas embolism. Bubbles of inert gas come out of solution during decompression. Cerebral symptoms range from dizziness to unconsciousness, with spinal cord involvement manifesting as paraesthesias/ numbness. The lower spinal cord is particularly affected and damage may be permanent. Pathology shows acute infarcts typically in lateral and dorsal columns (Calder 1986) (Figure 17.8). Chronic damage may also be seen in the absence of overt illness (Palmer et al. 1987).

References Abou-Hamden, A., Blumbergs, P.C. and Scott, G. et al. 1997. Axonal injury in falls. J Neurotrauma, 14, 699–713. Bamber, A.R., Paine, M.L., Ridout, D.A., Pryce, J.W., Jaques, T.S. and Sebire, N.J. 2016. Brian weight in sudden unexpected death in infancy: experience from a large single centre cohort. Neuropathol Appl Neurobiol, 42(4), 344–351. Beale, J.P., Wyatt, J.P., Beard, D., Busuttil, A. and Graham, C.A. 2000. A five year study of high falls in Edinburgh. Injury, 31, 503–508. Beecroft, D.M. and Lockett, B.K. 1997. Intrapulmonary siderophages in sudden infant death: a marker for previous imposed suffocation. Pathology, 29, 60–63. Blair, P.S., Flemming, P.J. and Bensley, D. et al. 1996. Smoking and the sudden infant death syndrome: results from 1993–1995 case-control study for confidential inquiry into stillbirths and deaths in infancy. BMJ, 313, 195–198. Blair, P.S., Flemming, P.J. and Smith, I.J. et al. 1999. Babies sleeping with parents: case-control study of factors influencing the risk of sudden infant death syndrome. BMJ, 319, 1457–1462. Bright, F.M., Vink, R. and Byard, R.W. 2018.Neuropathologicial developments in sudden infant death syndrome. Pediatr Dev Pathol, 21, 515–521. Boyd, B.C. 2002. Automobile supplemental restraint system-induced injuries. Oral Surg Oral Med Oral Pathol, 94, 143–148. Byard, R.W. and Collins, K.A. (eds) 2014. Forensic Pathology of Infancy and Childhood. Springer. Calder, I.M. 1986. Dysbarism. A review. Forensic Sci Int, 30, 237–266. Chapman, J. and Milroy, C. 1992. Firearm deaths in Yorkshire and Humberside. Forensic Sci Int, 57, 181–191. Chatzaraki, V., Thali, M.J. and Ampanozi, G. et al. 2018. Am J Forensic Med Pathol, 39(2), 130–140. Clark, J.C. and Milroy, C.M. 2000. Injuries and deaths of pedestrians. In  Mason, J.K. and Purdue, B.N. (eds), Pathology of Trauma. Arnold, pp 17–29.

Cooper, J., Balding, L. and Jordan, F. 1998. Airbag mediated death of a two year-old child wearing a shoulder/lap belt. J Forensic Sci, 43, 1077–1081. Cordner, S. 2003. Suicide, accident or natural death. In Payne-James, J., Busuttil, A. and Smock, W. (eds), Forensic Medicine. Clinical and Pathological Aspects. Greenwich Medical Media, pp 135–147. Davis, G.G. and Glass, J.M. 2001. Case report of sudden death after a blow to the back of the neck. Am J Forensic Med Pathol, 22, 13–18. Freytag, E. 1963. Autopsy findings in head injuries from blunt forces: statistical evaluation of 1,367 cases. Arch Pathol, 75, 402–413. Geddes, J.F., Whitwell, H.L. and Graham, D.I. 2000. Traumatic axonal injury: practical issues for diagnosis in medicolegal cases. Neuropathol Appl Neurobiol, 26, 105–116. Hadley, J.A. and Fowler, D.R. 2003. Organ weight effects of drowning and asphyxiation on the lungs, liver, brain, heart, kidneys and spleen. Forensic Sci Int, 133, 190–196. Hamilton, S.J. and McMahon, R.F. 2002. Sudden death and suicide: a comparison of brain weight. Br J Psychiatry, 181, 72–75. Hanzlick, R. 2001. Pulmonary haemorrhages in deceased infants. Am J Forensic Med Pathol, 22, 188–192. Hanzlick, R. and Delaney, K. 2000. Pulmonary haemosiderin in deceased infants. Am J Forensic Med Pathol, 21, 319–322. Kinney, H.C., Richicherson, G.B., Dymecki, S.M., Danrnall, R.A. and Nattie, E.E. 2009. The brainstem and serotonin in the sudden infant death syndrome. Annu Rev Pathol, 4, 517–550. Karger, B., Teige, K., Bühren, W. and DuChesne, A. 2000. Relationship between impact velocity and injuries in fatal pedestrian-car collisions. Int J Legal Med, 113, 89–97. Krous, H.F., Wixom, C. and Chadwick, A.E. et al. 2006. Pulmonary intra-­ alveolar siderophages in SIDS and suffocation:a San Diego SUDS/ SUDC Research Project report. Pediatr Dev Pathol, 9(2), 103–114. Kury, G., Weiner, J. and Duval, J. 2000. Multiple self-inflicted gunshot wounds to the head: report of a case and review of the literature. Am J Forensic Med Pathol, 21, 32–35. Lau, G., Ooi, P.L. and Phoon, B. 1998. Fatal falls from a height: the use of mathematical models to estimate the height of fall from the injuries sustained. Forensic Sci Int, 93, 33–44. Lau, G., Teo, C.E.S. and Chao, T. 2003. The pathology of trauma and death associated with falls from heights. In Payne-James, J., Busuttil, A. and Smock, W. (eds), Forensic Medicine. Clinical and Pathological Aspects. Greenwich Medical Media. Lau, G. 2004. Homicidal and dyadic falls from a height: rarities in Singapore. Med Sci Law, 44, 93–106. Leadbeatter, S. 2001. Resuscitation injury. In Rutty, G.N. (ed), Essentials of Autopsy Practice, Vol 1. Springer-Verlag. Macpherson, A.K. and Macarthur, C. 2002. Bicycle helmet legislation: evidence for effectiveness. Pediatr Res, 52, 472. Milovanovic, A.V. and DiMaio, V.J.M. 1999. Death due to concussion and alcohol. Am J Forensic Med Pathol, 20, 6–9. Milroy, C.M. 1999. Munchausen syndrome by proxy and intra-alveolar haemosiderin. Int J Legal Med, 112, 309–312. Milroy, C.M. and Clark, J.C. 2000. Injuries and deaths in vehicle occupants. In Mason, J.K., Purdue, B.N. (eds), Pathology of Trauma. Arnold. Nagurney, J.T., Borczuk, P. and Thomas, S.H. 1998. Elderly patients with closed head trauma after a fall: mechanisms and outcome. J Emerg Med, 16, 709–713. Palmer, A.C., Calder, I.M. and Hughes, J.T. 1987. Spinal cord degeneration in divers. Lancet, ii, 1365–1366. Paine, S.M.L., Jaques, T.S. and Sebire, N.J. 2014. Review: neuropathological features of unexplained sudden unexpected death in infancy: current evidence and controversies. Neuropahol Appl Neurobiol, 40, 364–384. Preuss, J., Padosch, S.A. and Dettmeyer, R. et al. 2004. Injuries in fatal cases of falls downstairs. Forensic Sci Int, 141, 121–126. Ragg, M., Hwang, S. and Steinhart, B. 2000. Analysis of serious injuries caused by stairway falls. Emerg Med, 12, 45–49. Ramsay, D.A. and Shkrum, M.J. 1995. Homicidal blunt head trauma, diffuse axonal injury, alcoholic intoxication, and cardiorespiratory arrest: a case report of a forensic syndrome of acute brainstem dysfunction. Am J Forensic Med Pathol, 16, 107–114.

References Ross, A.S. and Grieve, J.H.K. 2000. Underwater diving. In Mason, J.K. and Purdue, B.N. (eds), Pathology of Trauma. Arnold, pp 341–362. Ryan, G.A., McClean, A.J. and Vilenius, A.T.S. et al. 1994. Brain injury patterns in fatally injured pedestrians. J Trauma, 36, 469–476. Saukko, P. and Knight, B. 2016. Knight’s Forensic Pathology, 4th edn. CRC Press. Sethi, M., Heidenberg, J. and Wall, S.P., et al. 2015. Bicycle helmets are highly protective against traumatic brain injury within a dense urban setting. Injury, 46, 2483–2490. Thierauf, A., Preuss, J., Lignitz, E., et al. 2010 Retrospective analysis of fatal falls. Forensic Sci Int, 198(1–3), 92–96. Timler, D., Dworzynski, M.J. and Szarpak, L., et al. 2015. Head trauma in elderly patient: mechanisms of injuries and CT findings. Adv Clin Exp Med, 24(6), 1045–1050. Voigt, G.E. 1981. Small haemorrhage in the brain stem: a sign of injury? Am J Forensic Med Pathol, 2, 115–119.

Wardlaw, M. 2002. Butting heads over bicycle helmets. JAMA, 167, 337–338. Wyatt, J.P., Beard, D. and Busuttil, A. 1999. Fatal falls down stairs. Injury, 30, 31–34. Yen, K., Thali, J.M., Kneubuehl, B.P., Peschel, O., Zollinger, U. and Dirnhofer, R. 2003. Blood-spatter patterns. Hands hold clues for the forensic reconstruction of the sequence of events. Am J Forensic Med Pathol, 24, 132–140. Yukawa, N., Carter, N. and Rutty, G. et al. 1999. Intra-alveolar haemorrhage in sudden infant death syndrome: a cause for concern? J Clin Pathol, 52, 581–587. Zink, B.J. and Feustel, P.J. 1995. Effects of ethanol on respiratory function in traumatic brain injury. J Neurosurg, 82, 822–828. Zivot, U. and DiMaio, V.J.M. 1993. Motor vehicle-pedestrian accidents in adults: relationship between impact speed, injuries and distance thrown. Eur Spine J, 14, 185–186.

185

18

Non-traumatic neurological conditions in medico-legal work Colin Smith

Pathology within the central nervous system (CNS) is a common finding in post-mortem examination. The intracranial pathology is, however, often an incidental finding and not directly related to the cause of death. This chapter deals with the non-traumatic neurological conditions that have a direct relationship to the cause of death not covered in other chapters and which the forensic pathologist may encounter (Stewart et al. 2004). Some can be associated with sudden death (Black and Graham 2002). This chapter will focus on adult pathology and, therefore, does not cover topics such as sudden infant death syndrome or neonatal/infant ischaemic pathology. Disorders of cerebral perfusion, including infarction and epilepsy and sudden death in epilepsy, are further covered in Chapters 12 and 13, respectively.

Non-traumatic haemorrhage Non-traumatic parenchymal haemorrhages can be caused by a number of conditions, but by far the two commonest conditions are chronic hypertension and cerebral amyloid angiopathy (CAA). Underlying neoplasms (discussed below) or vascular malformations, particularly arteriovenous malformations (AVM), can be a cause of fatal non-traumatic parenchymal haemorrhage.

Chronic hypertension Chronic hypertension is common and contributes to significant morbidity and mortality in society. It is not only an important contributor to cognitive decline in ageing but also the most significant cause of primary parenchymal brain haemorrhage. Primary hypertensive haemorrhages are typically seen in the central white matter, involving basal ganglia and thalamic nuclei, cerebellum and pons. Within the cerebral hemispheres, they are often extensive with localised tissue disruption, acting as a space-occupying lesion with mass effect, and they often extend into the ventricular system, which can cause ventricular dilatation due to sterile cerebrospinal fluid (CSF) flow obstruction. Histologically cerebral small vessel disease (SVD) is seen in the form of arteriolosclerosis and lipohyalinosis. Arteriolosclerosis is a concentric fibrotic thickening of the media in deep parenchymal vessels, whereas lipohyalinosis is a localised thickening of the media with foamy macrophages and often fibrinoid necrosis.

Cerebral amyloid angiopathy CAA refers to the abnormal deposition of β-amyloid protein in the walls of small leptomeningeal and cortical arterioles. CAA may be an incidental finding, particularly in cases with Alzheimer’s disease (AD), but it may also be seen in younger adults with no associated cognitive problems. CAA can cause parenchymal bleeding in a lobar distribution; that is, the haemorrhage is limited to one of the main cerebral lobes. Typically, there is an extension through the cortex into the subarachnoid

space. On routine H&E stains, the vessel walls have a hyaline appearance, and there may be splitting within the media. The abnormal protein deposition is highlighted by Aβ immunohistochemistry. Lobar haemorrhages can be a cause of death, and it has been suggested that bleeding can be initiated by relatively minor trauma (Pittella and da Silva Gusmao 2016).

Hypoglycaemia Hypoglycaemic brain damage will be produced in the adult when blood glucose levels fall below about 1.5 mmol/L. Hypoglycaemia produces a pattern of selective neuronal necrosis which, in its pure form, differs from that seen with ischaemia (Auer 2004). However, hypoglycaemic coma is often accompanied by seizures or cardiorespiratory depression such that ischaemic features may coexist with the pathology associated with hypoglycaemia. Insulin overdose, resulting in profound hypoglycaemia, may be accidental or intentional (suicide or homicide). The patient with diabetes may accidentally administer an incorrect dose of insulin or may have difficulty controlling the diabetes, resulting in episodes of hypoglycaemia or hypoglycaemic coma. Rare causes of hypoglycaemia include islet cell tumours of the pancreas (insulinoma). Macroscopically, the brain shows little abnormality with only mild swelling. The microscopic distribution of injury is detailed in Table18.1, where hypoglycaemic injury is compared and contrasted with hypoxic-ischaemic damage. The latter is covered in detail in Chapter 12.

Infection of the CNS Infections of the nervous system remain a significant cause of morbidity and mortality worldwide. With the increasing ease of world travel, particularly to destinations ‘off the beaten track’, and an increasing population of immunosuppressed individuals, there is a need for greater awareness of nervous system infections. Nervous system infection may be the cause of sudden unexpected deaths in adults, particularly in the elderly and immunocompromised individuals. This section covers some of the more common nervous system infections encountered; for a more detailed review of this subject, the interested reader is referred to the recent International Society of Neuropathology text (Chrétien et al. 2020). Identification of the underlying organisms requires appropriate tissue samples and swabs to be taken at the time of the post-mortem examination, often requiring consultation with microbiology and virology departments. Guidelines for practice at post-mortem examination in such cases are provided in Chapter 2. CSF can be removed from the brain in situ during the postmortem examination by passing a needle into the lateral ventricles after the dura has been removed and may be used for

Macroscopic appearance of infections

Table 18.1 The microscopic distribution of neuronal damage in ischaemia and hypoglycaemia Anatomical region

Ischaemia

Hypoglycaemia

Hippocampus

Maximal within sector CA1

Maximal within sector CA1 and within the dentate gyrus

Cerebral cortex

Accentuated within the depths of gyri, particularly at boundary zones, and involving the deeper layers of the cortex

Widespread and most pronounced within the superficial layers

Diffuse damage in the striatum, usually mediumsized neurons

Little damage seen in the striatum

Purkinje cell ischaemic damage or focal infarction within the boundary zones

The cerebellar cortex, and in particular Purkinje cells, shows little evidence of injury

Striatum

Cerebellum

Figure 18.1 Histology of cerebral mucormycosis showing fungal hyphae.

Protozoal infections

polymerase chain reaction identification of viruses. At the postmortem examination, the air sinuses, orbit and both the middle and inner ears should be examined for possible foci of infection. The examination of the heart and other viscera may indicate the site of infection. Infections of the nervous system can be considered based on their anatomical location or the underlying organism.

Bacterial infections Bacteria may enter the nervous system by the haematogenous route (most common) or secondary to osteomyelitis and sinusitis/mastoiditis or compound fracture. Bacterial infections are usually suppurative, although notable exceptions include mycobacteria.

Fungal infections These tend to be opportunistic infections, being seen in immunocompromised patients. CNS fungal infection is virtually always associated with systemic infection, although infection may be most obvious in the brain (Figure 18.1). Spread is usually haematogenous, although in some cases the infection is the result of direct spread from either the air sinuses or the orbit. In particular, mucormycosis may extend from paranasal sinuses to the orbit or brain. This is classically seen in diabetic ketoacidosis, although it may be associated with any immunosuppressed state.

Viral infections Viruses generally enter the body via the mucous membranes of the respiratory or gastrointestinal tract, although in some cases entry may be traumatic, e.g. after a dog bite that breaks the skin. Some viruses are neurotrophic and will move towards the nervous system, usually via peripheral nerves (e.g. rabies). Some viruses will show a latent period between primary infection and nervous system disease (e.g. human immunodeficiency virus [HIV]).

These are disseminated via haematogenous spread and can produce a range of pathological appearances depending on the infecting organism. Protozoal infections are common worldwide, although tend to be seen less frequently in Western populations.

Macroscopic appearance of infections Pachymeningitis This refers to a collection of pus in relation to the dura. Empyemas are collections of pus that are usually subdural and found in the supratentorial space, although in the spinal region extradural collections can be seen in association with osteomyelitis. Subdural empyemas may be associated with open skull fractures or neurosurgical procedures and are seen with untreated otitis or sinusitis (De Bonis et al. 2009). Macroscopically, the appearances are of a collection of pus in the subdural space, often encapsulated by granulation tissue.

Leptomeningitis Although anatomically correct to call this leptomeningitis, it is more commonly referred to as meningitis. A majority of cases at post-mortem examination are acute purulent (bacterial) meningitis (Figure 18.2), with viral meningitis rarely being fatal, and a purulent fungal meningitis being uncommon. In adults, bacterial meningitis is usually associated with bacteraemia, although skull fracture secondary to trauma or previous neurosurgical procedure may result in meningitis. The organisms commonly involved in adults are Streptococcus pneumoniae and Neisseria meningitidis. Similar organisms are seen in children with, in addition, Haemophilus influenzae. Macroscopically, the brain is usually swollen and a purulent membrane is seen overlying the brain; this is most pronounced over the convexities with S. pneumoniae, and usually most pronounced basally with other organisms (Love 2001). The infection may extend into the ventricles (ventriculitis), and there may be a degree of hydrocephalus secondary to CSF obstruction by pus. Microscopically a florid neutrophilic infiltrate is seen within the subarachnoid space, and there is often thrombosis of cortical vessels with associated superficial cortical infarction.

187

188

Non-traumatic neurological conditions in medico-legal work

Figure 18.2 Purulent exudate in the subarachnoid space in a case of acute bacterial meningitis. The reflected dura can be seen at the top of the image.

Figure 18.4 A large cerebral abscess acting as a space-occupying lesion with mass effect. There is midline shift and subfalcine herniation. Note the haemorrhages in the brainstem (Duret haemorrhages).

Figure 18.3 Tuberculous meningitis with a dense granulomatous exudate obscuring the base of the brain.

Granulomatous meningitis

degree by the route of entry. Haematogenous dissemination of the organism tends to produce multifocal cerebral abscesses, whereas direct invasion from local infection in the paranasal sinuses, middle ear or maxillary dental roots tends to produce a solitary lesion. For haematogenous dissemination, primary sites include cardiac valves and chronic lung infections (bronchiectasis). They also may be seen in intravenous drug abuse. Bacterial organisms are usually responsible for cerebral abscesses in immunocompetent individuals with Streptococcus viridans being the most common organism identified. In immunocompromised individuals, fungal organisms, such as Aspergillus and Candida species, and protozoa, such as Toxoplasma gondii, may be seen. Toxoplasmosis, in particular, is associated with HIV infection (Strittmatter et al. 1992) (Figure 18.5). Macroscopically, cerebral abscesses caused by haematogenous dissemination are classically said to develop at the cortical grey-white matter interface, although in some cases they are

Tuberculous meningitis macroscopically has a rather nodular exudate and is most pronounced basally and laterally, particularly overlying the Sylvian fissures (Figure 18.3). Tuberculosis may also present as a parenchymal mass (tuberculoma). The differential diagnosis of granulomatous inflammation of the meninges includes, amongst others, the non-infective condition neurosarcoidosis.

Infections of the brain parenchyma These may be suppurative or non-suppurative; the suppurative infections will produce an abscess, whereas non-suppurative infections tend to produce encephalitis, a non-suppurative diffuse lymphocytic inflammation of the brain parenchyma.

Cerebral abscesses Cerebral abscesses can act as space-occupying lesions resulting in raised intracranial pressure (Figure 18.4). Cerebral abscesses can be solitary or multifocal, determined to a

Figure 18.5 A toxoplasmosis cyst seen in the brain of a patient infected with HIV (H&E, ×40).

Macroscopic appearance of infections

Figure 18.6 Multiple cerebral abscesses: in this example the abscesses were caused by toxoplasmosis. widely distributed throughout the brain (Figure 18.6). Abscesses associated with paranasal air sinus and maxillary dental root infections are located in the adjacent frontal lobe, whereas those associated with middle-ear infections are usually found in the temporal lobe or, rarely, the cerebellum. Abscesses initially develop as an area of focal swelling and discolouration, often being haemorrhagic. By about 2 weeks, the abscess will have a purulent necrotic centre and a thick fibrotic capsule.

Encephalitis There are some general features of encephalitis that are not dependent on the infecting virus. Macroscopically, the brain is often swollen with congestion of parenchymal vessels. There is usually a perivascular lymphocytic infiltrate with both microglial and astrocytic (gliosis) activation. Cell necrosis is variable, ranging from individual neurons in poliomyelitis to extensive tissue infarction (necrotising encephalitis) in herpes simplex encephalitis. Inclusion bodies may be seen in neurons or glial cells, and this may help in identification of the infecting virus. Specific causes of encephalitis that are of importance are considered below.

HIV infection HIV infection is widespread throughout the world and remains a major clinical problem. In forensic practice, it is important to remember that a proportion of the intravenous drug abuser (IVDA) population are HIV positive. HIV infection results in the progressive depletion of CD4 T lymphocytes and subsequent immunosuppression causing acquired immune deficiency syndrome (AIDS). Neuropathology is seen in 70–90 per cent of AIDS cases, although much of this is caused by opportunistic infections (Morgello 2018). HIV itself can cause nervous system pathology, with neuropathology directly attributable to HIV being seen in 20–30 per cent of cases (Navia et al. 1986). HIV encephalopathy (HIVE) is a white matter disease characterised by multinucleated giant cells. HIVE has been shown to be more prevalent in IVDAs than in homosexuals. The HIV-related protein gp41 can be demonstrated immunohistochemically in relation to many multinucleated giant cells. Macroscopically, the HIVE brain often appears normal; microscopically, there are foci of perivascular lymphocytes and microglial nodules

Figure 18.7 Herpes simplex encephalitis: there is haemorrhagic necrosis predominantly involving the temporal lobes. often lying adjacent to multinucleated giant cells. Vacuolar myelopathy is seen in AIDS patients and resembles subacute combined degeneration of the cord. The degeneration develops in the posterior and lateral columns of the spinal cord, and there is axonal degeneration. In AIDS patients, the disorder is not associated with vitamin B12 deficiency.

Herpes simplex encephalitis This is the most common necrotising encephalitis and is usually caused by herpes simplex virus (HSV) type 1. The macroscopic appearances of established HSV encephalitis are characteristic, with asymmetrical necrosis of the anterior temporal lobes, cingulate gyri and insular cortices (Figure 18.7). In the early stages of the infection, however, the changes can be very mild with only temporal lobe congestion suggestive of infection. Microscopic examination of established cases reveals perivascular inflammation, necrosis with foamy macrophage infiltration and eosinophilic neuronal inclusions, which are predominantly nuclear (Esiri 1982). Immunohistochemistry can detect the virus up to 3 weeks from the start of an infection.

Malaria Malaria is a protozoal infection caused by Plasmodium species. Cerebral malaria may complicate Plasmodium falciparum (in up to 10 per cent of infections), the predominant type not only seen in West Africa but also encountered throughout Africa, Asia and South America. It should always be considered in cases of sudden death in travellers through these regions. Encephalopathy and death can develop rapidly. Macroscopically, petechial haemorrhages are usually widely distributed throughout the brain (Figure 18.8). Microscopically, the capillaries are engorged and malaria parasites or granules of pigment are seen in many of the red cells. Inflammation is usually sparse, although occasional aggregates of neutrophils, lymphocytes and macrophages (Dürck granuloma) may be seen.

COVID-19 infection COVID-19 infection may be associated with neurological complications. Around 2 per cent of patients admitted to hospital present with cerebrovascular disease which is associated with a dramatically poorer prognosis (Mao et al. 2020). Pathological and radiological features supportive of a thrombotic

189

190

Non-traumatic neurological conditions in medico-legal work microangiopathy caused by an endotheliopathy with a haemorrhagic predisposition has been described (HernandezFernandez et al. 2020). These changes are unaccompanied by vasculitis or necrotising encephalitis. COVID-19 associated cerebrovascular disease is further characterised by a high incidence of large vessel thrombotic occlusion, which shows a disproportionate vertebrobasilar location. Other potential associations between COVID-19 infection and cerebrovascular disease may include arterial dissections and a leukoencephalopathy consistent with posterior reversible encephalopathy syndrome. Apart from cerebrovascular disease, neuropathological findings so far reported appeared somewhat disparate with some of the mainly small series and case reports so far published describing no signs of encephalitis or vasculitic change in contrast to others reporting a panencephalitis with diffuse petechial haemorrhages (Glatzel 2020). The spectrum of encephalitic change may include an acute disseminated encephalomyelitis (ADEM)-like picture (Reichard et al. 2020).

Neoplasia involving the CNS A detailed description of tumours of the CNS is out with the scope of this book and the interested reader is referred to recent specialist accounts (Louis et al.). However, CNS tumours may be encountered at post-mortem examination, and the role of brain neoplasms as a cause of sudden death has been reviewed (Matschke and Tsokos 2005; Tsokos 2004). The tumours may be either primary or secondary; metastatic tumours to the CNS are common, whereas primary intrinsic tumours are rare. The epithelial tumours that metastasise most commonly to the nervous system are of the bronchus and breast, followed by melanoma (Delattre et al. 1988). In general, prostatic and female genital tract tumours rarely metastasise to the brain or meninges, although prostatic carcinomas do frequently cause neurological complications secondary to cord compression after vertebral body metastasis and collapse. Diffuse infiltration of the leptomeninges (malignant meningitis/meningeal carcinomatosis) may be seen with haematological malignancies (lymphoma and leukaemia) and epithelial malignancies such as breast and bronchial carcinomas. Of the intrinsic primary brain tumours, gliomas are the most common. There are a large number of histological subtypes that reside under the blanket term ‘glioma’. More specific

Figure 18.8 Section through the cerebellum and brainstem demonstrating multiple petechial haemorrhages secondary to malarial infection.

terms such as astrocytoma, oligodendroglioma, ependymoma, choroid plexus tumour or mixed glial tumour should be used, with an appropriate grade given on the basis of histological features (pleomorphism, mitotic activity, tumour necrosis, microvascular proliferation). Gliomas may occur at any age and any anatomical site within the nervous system. Tumours may present to the pathologist through a variety of mechanisms: • They can grow to a large size producing only rather nonspecific symptoms such as headache. When slow growing and large, the tumour acts as a mass lesion. Should the lesion then enlarge rapidly, e.g. after tumour haemorrhage, the brain is unable to compensate for the rapidly expanding mass lesion, resulting in brain herniation and death (Figure 18.9). Although intratumoural haemorrhage may occur in any tumour, primary or secondary, it is particularly associated with oligodendrogliomas, glioblastomas and metastatic melanomas. • As discussed below, seizures are often associated with malignancy, and death may follow a seizure. Macroscopically, gliomas result in diffuse expansion of both white and grey matter as they diffusely infiltrate brain tissue. They are often rather gelatinous and may have cystic areas. Oligodendrogliomas frequently show calcification, although this may be seen in other gliomas. Gliomas of the brainstem or spinal cord can be identified as an area of diffuse expansion with, on sectioning, loss of the normal anatomical structure. High-grade gliomas, such as glioblastoma, contain areas of necrosis that macroscopically appear yellow and granular. A specific benign tumour of which the pathologist should be aware is the colloid cyst of the third ventricle (Figure 18.10). This benign cystic lesion can be associated with sudden death, particularly in young to middle-aged individuals (Filkins et al. 1996). This lesion lies within the third ventricle where it can obstruct the foramen of Monro, resulting in acute hydrocephalus with dilatation of the lateral ventricles.

Figure 18.9 Acute haemorrhage into a glioblastoma, resulting in a rapid rise in intracranial pressure. There is compression of the ventricular system and an obvious subfalcine hernia.

Disorders associated with dementia

Figure 18.10 Colloid cyst of the third ventricle.

Demyelinating disorders Multiple sclerosis This is the most common form of demyelinating disease. It is a relapsing-remitting disorder in which plaques of demyelination can be widely distributed throughout the CNS. In general, the forensic pathologist will encounter multiple sclerosis (MS) as an incidental finding rather than a direct cause of death, although death may occur if there is acute demyelination in the brainstem. In chronic MS, the plaques have a predilection for subependymal (periventricular) and subpial sites, and for the cortical grey-white matter junction. Chronic plaques are sunken and grey and generally sharply defined. Acute plaques have a rather more granular appearance (Figure 18.11). The optic nerves should be examined if possible. In some rare situations, demyelinating disorders can present as an acute syndrome with rapid mortality (within days) (Bevan and Cree 2015). Some of these are discussed below. MS is associated with changes in thermoregulation and changes in cardiovascular function and is associated with sudden death, including in heatstroke and hypothermia (White et al. 1996; Davis et al. 2010). Sudden unexpected death is described in the setting of MS and is thought to be as a consequence of cardiovascular dysfunction (Kaplan et al. 2015).

Figure 18.12 Petechial haemorrhages within the cerebrum from a case of acute haemorrhagic leukoencephalitis. Ball- and ringtype haemorrhages were seen microscopically.

Acute haemorrhagic leukoencephalitis (Hurst’s disease) This rare disorder is considered to be a hyperacute allergic reaction, seen at all ages and usually fatal within 2–3 days of the onset of symptoms (Donnet et al. 1996). The illness usually follows an upper respiratory tract infection and is characterised by fever, impaired consciousness and focal neurological signs. Macroscopically, the brain is swollen and congested, and tentorial and tonsillar hernias are usually apparent. Petechial haemorrhages are widely distributed throughout the brain (Figure 18.12). Microscopically, these are ball- or ring-type haemorrhages with central necrotic small vessels. Foci of perivascular demyelination are seen if the individual had survived longer than about 4 days from the onset of symptoms. A similar condition is perivenous encephalomyelitis (ADEM), which may follow infection or vaccination (Anilkumar et al. 2020).

Central pontine myelinolysis This uncommon disorder is seen in individuals with a variety of underlying disorders (Lampl and Yazdi 2002), although it is most commonly seen in people with chronic alcohol problems. This is covered in more detail in Chapter 19.

Disorders associated with dementia

Figure 18.11 An acute plaque of demyelination with a granular appearance. Older plaques are seen in the periventricular area.

Like stroke, dementia is not a specific disorder but rather a generic term for cognitive and behavioural disturbances produced by a variety of pathological processes. The pathologist may encounter dementia as a cause of abnormal behaviour, which may underlie self-harm or a road traffic accident, or may be required to comment on neurodegenerative changes in assault victims who may have been displaying unusual aggressive behaviour. Allegations of maltreatment may also occur because the victims not infrequently become cachectic. Dementia is caused not only by neurodegenerative pathology, but also by neoplasia, infections and metabolic cerebral dysfunction induced by disorders such as chronic renal failure or endocrine disorders such as hypoparathyroidism. The neuropathology of disorders causing the clinical phenotype of dementia is complex, and the reader is recommended to

191

192

Non-traumatic neurological conditions in medico-legal work access more detailed texts for greater detail than can be offered in this chapter (Dickson; Kovacs 2015; Dickson 2011). This section will offer a very general overview and approach to cases with known clinical dementia, focusing on the more common disorders. Macroscopic examination should assess the presence or absence of cerebral atrophy and brain weight. The adult brain should weigh between 1200 and 1600 g, the weight usually being slightly greater in males. It must be remembered that, when interpreting brain weight, it is important to consider the stature of the individual; a person of small stature will have a small brain that may be out with these guide weights. Beyond the age of about 50 years, there is a gradual decrease in brain weight of about 2 per cent/year. If atrophy is present, the location and severity should be noted (frontotemporal distribution of AD and the knife-edge frontal atrophy of Pick’s disease and corticobasal degeneration being classic examples). In cases of suspected Creutzfeldt-Jakob disease, frozen tissue should be retained to assist with molecular diagnosis. The brain should be sectioned after adequate fixation. Histological sampling is a necessity to achieve a diagnosis and some immunohistochemical assessment is required to assess the extent of neurodegenerative changes. Guidelines for sampling the brain in cases of probable dementia have been published (Kovacs 2015), and recommended blocks and special stains are outlined in Chapter 5 (Appendix 7). In the absence of a good clinical history of cognitive and/or behavioural disturbances, the pathologist should not make a diagnosis of dementia. The pathological changes in such case should be described only with the comment is being made that such changes are associated with specific neurodegenerative conditions.

Alzheimer’s disease AD accounts for the majority of cases of dementia. It is most common after the age of 65 years, the incidence increasing with age, although early-onset forms are well recognised and are mostly genetic (familial AD). Macroscopically, the brain will show a frontotemporal pattern of atrophy, the changes being most severe in the medial aspects of the temporal lobe (Figure 18.13). Microscopically, the characteristic features are the abundance of neurofibrillary tangles and neuropil threads

(tau immunoreactive) and β-amyloid deposition (Aβ protein immunoreactive). These features can be associated with conditions other than AD, and it is the distribution and quantity that determine the diagnosis. The assessment is based on the published NIA-AA criteria (Montine et al. 2012), which will allow an assessment of the pathology indicative of not AD, low probability, intermediate probability or high probability of AD. CAA is strongly correlated with AD, and the affected vessels are usually seen in the subarachnoid space and superficial cortex. CAA is associated with lobar intracerebral haemorrhage (mentioned above).

Vascular dementia The degree to which vascular pathology contributes to dementia is controversial and poorly understood, but there have been recent attempts to standardise the neuropathological approach in assessing the vascular burden in cases of cognitive decline (Skrobot et al. 2016). Dementia may develop as the result of a single large infarct, as a consequence of multiple small cortical and lacunar infarcts (multi-infarct dementia), or in the setting of generalised white matter damage due to cerebral SVD. The location of the infarcts is considered to be important, with lesions in the left hemisphere being more likely to cause dementia. Unsurprisingly infarcts involving the limbic system will also result in cognitive impairment. At post-mortem examination of the brain, the small lacunar infarcts are seen predominantly in relation to the basal ganglia, thalamus and adjacent white matter. Histological examination of such cases should be carried out to exclude the possibility of coexistent pathology such as AD as the combination of vascular pathology with AD pathology is common.

Parkinson’s disease and dementia with Lewy bodies This disorder is often associated with Parkinson’s disease, and dementia with Lewy bodies (DLB) can also coexist with Alzheimer-type pathology (Walker et al. 2019). However, DLB appears to have a distinct clinical progression that differs from AD. The diagnosis is made by demonstrating Lewy bodies within the substantia nigra, limbic system and neocortical regions. DLB is of particular importance in that neuroleptic drugs, used to treat agitation and hallucinations, may induce rapid clinical deterioration that may be life-threatening. Cardiovascular abnormalities are common in Parkinson’s disease and associated with sudden death, referred to as sudden unexpected death in Parkinson’s disease (Scorza et al. 2018).

Iatrogenic neuropathology The pathologist may be asked to examine cases where an individual has died as a result of medical intervention in which the most significant pathology will be in the nervous system. This section considers neuropathology that can develop as a result of the invasive intervention and medical intervention.

Invasive intervention Neoplasm surgery Figure 18.13 Bilateral temporal lobe atrophy with associated enlargement of the ventricular system in a case of Alzheimer’s disease.

The surgical procedure may be either a burr-hole biopsy for diagnosis or a craniotomy for debulking surgery. Both are associated with the risk of raised intracranial pressure through either haemorrhage or oedema, the risk of mortality being

References greater with burr-hole biopsies where the surgeon is unable rapidly to control the bleeding and the intracranial pressure. In burr-hole biopsies, the risk is considered to be greater with freehand needle biopsy rather than using stereotactic techniques, particularly for deeply seated lesions, although freehand biopsy is now rarely done for such lesions (Lee et al. 1991).

Spinal cord injury Spinal cord infarction has been reported after surgical treatment of abdominal aortic aneurysms in which the artery of Adamkiewicz is damaged (Hamano et al. 2000).

Air emboli Air emboli may be introduced into the vascular system during spinal surgery, particularly when the sitting position is used for cervical spinal surgery (McCarthy et al. 1990). This is of the greatest concern if there is a right-to-left cardiac shunt (such as a patent foramen ovale), and, if untreated, this may result in a cardiorespiratory arrest (Pham Dang et al. 2002). Air emboli can also be associated with the insertion of central venous catheters.

Radiological coiling/embolisation Coiling of cerebral aneurysms is now frequently undertaken and is associated with a low rate of morbidity and mortality, but fatalities have been described after coiling. The coil may pass through the wall of the aneurysm resulting in subarachnoid haemorrhage or, alternatively, cause vasospasm or thrombosis of the stem vessel that results in massive infarction and subsequent death. In some cases, the complications can be delayed, sometimes for a period of more than 1 month (Rouchaud et al. 2016). Complications are less frequent than those associated with surgical clipping of aneurysms (Hohlrieder et al. 2002). At post-mortem examination, careful assessment of the coil and its relationship to the aneurysm is required. Embolisation is used in the treatment of AVM and can be used in some tumour types to reduce the size and vascularity of a tumour (particularly meningiomas) before surgery. Embolic material may not be confined to the target site, resulting in embolic infarcts and very rarely death (Gruber et al. 2000). At post-mortem examination, the embolic material is easily identified within vessels.

Medical intervention Anticoagulant, anti-platelet and thrombolytic drugs These drugs have a risk of intracerebral haemorrhage associated with them, as well as haemorrhage at other sites. The risk of intracerebral haemorrhage is dependent on the specific drug and the underlying medical disorders. Underlying CAA may predispose to intracerebral haemorrhage with anticoagulant therapy (McCarron et al. 1999). The intracerebral haemorrhages may be multiple small lesions throughout the brain and spinal cord or may be a single large lesion producing a significant mass effect.

Complications of previous irradiation In the acute situation, therapeutic cranial irradiation may be associated with oedema, although this is usually well treated with steroids. Delayed radiation injury may take the form of focal radionecrosis or diffuse radiation leukoencephalopathy.

Radionecrosis requires several months to develop and can be recognised histologically by the presence of hyaline vessels, some of which may show fibrinoid necrosis, and a geographic pattern of necrosis. Radiation leukoencephalopathy is rather poorly defined but has been reported in cases 12 months or more after therapeutic cranial irradiation (Partap et al. 2019). Clinically there is neurological and intellectual decline; histologically, the brain may show diffuse white matter vacuolation or focal white matter demyelination and necrosis.

Conclusion This chapter illustrates a range of non-traumatic neurological conditions that may be encountered in forensic practice. It will be clear that this chapter is not comprehensive and can act only as an introduction to the many potential causes of non-traumatic injury to the nervous system. When faced with a neurological condition as the cause of death, a successful post-mortem examination requires consideration of all the information available combined with a detailed examination of the nervous system.

References Anilkumar, A.C., Foris, L.A. and Tadi, P. 2020. Acute Disseminated Encephalomyelitis (ADEM). StatPearls. Auer, R.N. 2004. Hypoglycemic brain damage. Forensic Sci Int, 146(2–3), 105–110. Bevan, C.J. and Cree, B.A. 2015. Fulminant demyelinating diseases of the central nervous system. Semin Neurol, 35(6), 656–666. Black, M. and Graham, D.I. 2002. Sudden unexplained death in adults caused by intracranial pathology. J Clin Pathol, 55(1), 44–50. Chrétien, F., Wong, K.T., Sharer, L.R., Keohane, C. and Gray, F. (eds) 2020. Infections of the Central Nervous System: Pathology and Genetics, 3rd edn. Wiley-Blackwell. Davis, S.L., Wilson, T.E., White, A.T. and Frohman, E.M., 2010. Thermoregulation in multiple sclerosis. J Appl Physiol, 109(5), 1531–1537. De Bonis, P., Anile, C., Pompucci, A., Labonia, M., Lucantoni, C. and Mangiola, A. 2009. Cranial and spinal subdural empyema. Br J Neurosurg, 23(3), 335–340. Delattre, J.Y., Krol, G., Thaler, H.T. and Posner, J.B. 1988. Distribution of brain metastases. Arch Neurol, 45(7), 741–744. Dickson, D. 2011. Neurodegeneration: The Molecular Pathology of Dementia and Movement Disorders, 2nd edn. Oxford: Wiley-Blackwell. Dickson, D. 2011. Neurodegeneration: The Molecular Pathology of Dementia and Movement Disorders, 2nd edn. Wiley-Blackwell. Donnet, A., Dufour, H., Gambarelli, D., Bruder, N., Pellissier, J.F. and Grisoli, F. 1996. Acute Weston Hurst necrotizing hemorrhagic leukoencephalitis. Rev Neurol (Paris), 152(12), 748–751. Esiri, M.M. 1982. Herpes simplex encephalitis. An immunohistological study of the distribution of viral antigen within the brain. J Neurol Sci, 54(2), 209–226. Filkins, J.A., Cohle, S., Levy, B.K. and Graham, M. 1996. Unexpected deaths due to colloid cysts of the third ventricle. J Forensic Sci, 41(3), 521–523. Glatzel, M. 2020. Neuropathology of COVID-19: where are the neuropathologists? Brain Pathol, 30(4), 729. Gruber, A., Bavinzski, G., Killer, M. and Richling, B. 2000. Preoperative embolization of hypervascular skull base tumors. Minim Invasive Neurosurg, 43(2), 62–71. Hamano, T., Miyoshi, Y., Hirayama, M., Hiraki, S., Mutoh, T. and Kuriyama, M. 2000. Posterior thoracic spinal cord infarction: complication of thoracoabdominal aortic aneurysm. Eur Neurol, 44(1), 59–60. Hernandez-Fernandez, F., Valencia, H.S., Barbella-Aponte, R.A., ColladoJimenez, R., Ayo-Martin, O., Barrena, C., Molina-Nuevo, J.D., GarciaGarcia, J., Lozano-Setien, E., Alcahut-Rodriguez, C., Martinez-Martin, A., Sanchez-Lopez, A. and Segura, T. 2020. Cerebrovascular disease in patients with COVID-19: neuroimaging, histological and clinical description. Brain, 143, 3089–3103.

193

194

Non-traumatic neurological conditions in medico-legal work Hohlrieder, M., Spiegel, M., Hinterhoelzl, J., Engelhardt, K., Pfausler, B., Kampfl, A., Ulmer, H., Waldenberger, P., Mohsenipour, I. and Schmutzhard, E. 2002. Cerebral vasospasm and ischaemic infarction in clipped and coiled intracranial aneurysm patients. Eur J Neurol, 9(4), 389–399. Kaplan, T.B., Berkowitz, A.L. and Samuels, M.A. 2015. Cardiovascular dysfunction in multiple sclerosis. Neurologist, 20(6), 108–114. Kovacs, G.G. (eds) 2015. Neuropathology of Neurodegenerative Diseases: A Practical Guide, 3rd edn. Cambridge University Press. Lampl, C. and Yazdi, K. 2002. Central pontine myelinolysis. Eur Neurol, 47(1), 3–10. Lee, T., Kenny, B.G., Hitchock, E.R., Teddy, P.J., Palividas, H., Harkness, W. and Meyer, C.H. 1991. Supratentorial masses: stereotactic or freehand biopsy? Br J Neurosurg, 5(4), 331–338. Louis, D.N., Ohgaki, H., Wiestler, O.D. and Cavenee, W.K. 2016. WHO Classification of Tumours of the Central Nervous System, Revised 4th edn, France IARC Publications. Love, S. 2001. Autopsy approach to infections of the CNS. Curr Top Pathol, 95, 1–50. Mao, L., Jin, H., Wang, M., Hu, Y., Chen, S., He, Q., Chang, J., Hong, C., Zhou, Y., Wang, D., Miao, X., Li, Y. and Hu, B. 2020. Neurologic manifestations of hospitalized patients with coronavirus disease 2019 in Wuhan, China. JAMA Neurol, 77(6), 683–690. Matschke, J. and Tsokos, M. 2005. Sudden unexpected death due to undiagnosed glioblastoma: report of three cases and review of the literature. Int J Legal Med, 119(5), 280–284. McCarron, M.O., Nicoll, J.A., Ironside, J.W., Love, S., Alberts, M.J. and Bone, I. 1999. Cerebral amyloid angiopathy-related hemorrhage. Interaction of APOE epsilon2 with putative clinical risk factors. Stroke, 30(8), 1643–1646. McCarthy, R.E., Lonstein, J.E., Mertz, J.D. and Kuslich, S.D. 1990. Air embolism in spinal surgery. J Spinal Disord, 3(1), 1–5. Montine, T.J., Phelps, C.H., Beach, T.G., Bigio, E.H., Cairns, N.J., Dickson, D.W., Duyckaerts, C., Frosch, M.P., Masliah, E., Mirra, S.S., Nelson, P.T., Schneider, J.A., Thal, D.R., Trojanowski, J.Q., Vinters, H.V., Hyman, B.T. 2012. National Institute on Aging-Alzheimer’s Association guidelines for the neuropathologic assessment of Alzheimer’s disease: a practical approach. Acta Neuropathol, 123(1), 1–11. Morgello, S. 2018. HIV neuropathology. Handb Clin Neurol, 152, 3–19. Navia, B.A., Cho, E.S., Petito, C.K. and Price, R.W. 1986. The AIDS dementia complex: II. Neuropathology. Ann Neurol, 19(6), 525–535.

Partap, S., Russo, S., Esfahani, B., Yeom, K., Mazewski, C., Embry, L., Wheeler, G., Ullrich, N.J. and Bowers, D.C. 2019. A review of chronic leukoencephalopathy among survivors of childhood cancer. Pediatr Neurol, 101, 2–10. Pham Dang, C., Pereon, Y., Champin, P., Delecrin, J. and Passuti, N. 2002. Paradoxical air embolism from patent foramen ovale in scoliosis surgery. Spine (Phila Pa 1976), 27(11), E291–E295. Pittella, J.E. and da Silva Gusmao, S.N. 2016. Intracerebral hemorrhage due to cerebral amyloid angiopathy after head injury: report of a case and review of the literature. Neuropathology, 36(6), 566–572. Reichard, R.R., Kashani, K.B., Boire, N.A., Constantopoulos, E., Guo, Y. and Lucchinetti, C.F. 2020. Neuropathology of COVID-19: a spectrum of vascular and acute disseminated encephalomyelitis (ADEM)-like pathology. Acta Neuropathol, 140(1), 1–6. Rouchaud, A., Brinjikji, W., Lanzino, G., Cloft, H.J., Kadirvel, R. and Kallmes, D.F. 2016. Delayed hemorrhagic complications after flow diversion for intracranial aneurysms: a literature overview. Neuroradiology, 58(2), 171–177. Scorza, F.A., Fiorini, A.C., Scorza, C.A. and Finsterer, J. 2018. Sudden unexpected death in Parkinson’s disease (SUDPAR): a fatal event that James Parkinson did not address. Age Ageing, 47(4), 627. Skrobot, O.A., O’Brien, J., Black, S., Chen, C., DeCarli, C., Erkinjuntti, T., Ford, G.A., Kalaria, R.N., Pantoni, L., Pasquier, F., Roman, G.C., Wallin, A., Sachdev, P., Skoog, I., group, V., Ben-Shlomo, Y., Passmore, A.P., Love, S. and Kehoe, P.G. 2017. The vascular impairment of cognition classification consensus study. Alzheimers Dement, 13, 624–633. Stewart, W., Black, M., Kalimo, H. and Graham, D.I. 2004. Non-traumatic forensic neuropathology. Forensic Sci Int, 146(2–3), 125–147. Strittmatter, C., Lang, W., Wiestler, O.D. and Kleihues, P. 1992. The changing pattern of human immunodeficiency virus-associated cerebral toxoplasmosis: a study of 46 postmortem cases. Acta Neuropathol, 83(5), 475–481. Tsokos, M. (eds) 2004. Forensic Pathology Reviews, 3rd edn. Humana; Blackwell. Walker, L., Stefanis, L. and Attems, J. 2019. Clinical and neuropathological differences between Parkinson’s disease, Parkinson’s disease dementia and dementia with lewy bodies – current issues and future directions. J Neurochem, 150(5), 467–474. White, K.D., Scoones, D.J. and Newman, P.K. 1996. Hypothermia in multiple sclerosis. J Neurol Neurosurg Psychiatry, 61(4), 369–375.

19

Alcohol, drugs, toxins and post-mortem toxicology Colin Smith and Christopher Milroy

Introduction A total of 271 million people aged between 16 and 64 years, approximately 5.5 per cent of the world’s population, were estimated to use drugs at least once during 2017 according to the World Health Organization (WHO 2019). Alcohol (ethanol) and other recreational drugs are widely used in many societies and can result in significant neuropathological changes (Weis and Buttner 2017a,b). On average every person in the world aged 15 years or older drinks 6.2 L of pure alcohol per year. As less than half the population (38.3 per cent) actually drinks alcohol, this means that those who do drink consume on average 17 L of pure alcohol annually. Some 31 million persons have drug use disorders. Almost 11 million people inject drugs, of which 1.3 million are living with human immunodeficiency virus (HIV), 5.5 million with hepatitis C and 1 million with both HIV and hepatitis. The harmful use of alcohol results in 3.3 million deaths each year representing 5.9 per cent of global deaths (WHO 2018). In 2017, there were over 70,000 drug overdose deaths in the United States, with over 46,000 due to opioids (Scholl et al. 2018).

Alcohol (ethanol) Ethanol is an alcohol consumed in alcoholic beverages. It produces neurological dysfunction both directly and secondary to vitamin deficiencies and metabolic dysfunction (Sutherland et al. 2014a). Alcohol is typically consumed orally and rapidly absorbed by the gastrointestinal tract, mostly in the small bowel. It is not bound by plasma proteins, so is rapidly diffused in the blood stream. Because it is essentially not absorbed by fat or bone, its volume of distribution is related to the total body water. As such men have a lower blood ethanol compared with women when the same amount of ethanol is consumed. Alcohol is metabolised to acetaldehyde via three main enzyme pathways: alcohol dehydrogenase (ADH), microsomal ethanol oxidising system (MEOS), which requires the CYP2E1 enzyme system, and peroxisomal catalase, though the latter pathway is not significant in vivo. Non-oxidative pathways also occur. ADH is the main pathway at low concentrations but with chronic use MEOS becomes more important. The rate of metabolism of ethanol is typically 15–18 mg/100 mL per hour but may vary between 10 and 25 mg/100 mL per hour. Alcohol elimination is a zero order process at high concentrations, but this is not the case with very low concentrations. Most ethanol is metabolised in the liver. This results in the production of acetoacetate which is converted to acetate. The acute intoxicant effect of alcohol varies significantly from person to person depending on a number of factors. These include age, physique, sex, race and tolerance to the effects of alcohol, reflecting the individual’s drinking habits. The Mellanby effect, where the intoxicant effect of alcohol is greater when the concentration in blood is rising than when it is falling, also needs to be taken into account (Mellanby 1919). Table 19.1,

relating the effect of alcohol to the blood alcohol concentration, should be regarded as a very rough guide indeed (Dubowski 1980).

Death from acute alcohol intoxication Determining the cause of stupor or coma in someone who has the odour of an alcoholic beverage on their breath requires careful clinical assessment. Someone who is in an alcoholic stupor will normally be rousable, but all pathologists will come to deal with cases, sometimes including in police custody, where a person is assumed to have been drunk but actually sustained a head injury. As the concentration goes above 300 mg/100 mL of blood, there is an increasing risk of death, especially in the naïve drinker. Alcoholic patients in coma are at risk of inhaling gastric contents but may also die of the direct central nervous system depression. Where ethanol is consumed along with other drugs, the ethanol may act synergistically with other central nervous system depressants, such as benzodiazepines, zopiclone and opiates, to cause death when the concentrations of the individual drugs would not be expected to cause death. Heatley and Crane (1990) found the mean blood ethanol concentration in acute alcohol intoxication deaths was 326 mg/100 mL where there was evidence of inhalation of gastric contents, and 382 mg/100 mL without inhalation. Koski et al. (2002) found a mean ethanol concentration of 348 mg/100 mL in 615 deaths where ethanol was the only psychoactive substance present. Jones and Holmgren (2003) examined blood ethanol concentrations in 693 acute alcohol intoxication deaths and found the mean concentration of ethanol in acute intoxication deaths in 529 men was 356 and 373 mg/100 mL in 164 women. These figures must be contrasted with a number of anecdotal cases where substantially higher concentrations of ethanol have been survived (Johnson et al. 1982; Jones 1999).

Table 19.1 Acute intoxicant effects of alcohol Blood alcohol concentration (mg/100 mL)

Behavioural effects

 10–50

Subclinical

 30–120

Euphoria

 90–250

Excitement

180–300

Confusion

250–400

Stupor

350–500

Coma

450

Death

Source: Reproduced from Dubowski, K.M. 1989. Stages of Acute Alcoholic Influence/Intoxication. University of Oklahoma College of Medicine.

196

Alcohol, drugs, toxins and post-mortem toxicology The interaction of alcohol with the compromised brain is of considerable forensic interest. Although modest alcohol use may protect against ischaemic stroke, heavier drinking is associated with an increased incidence of both haemorrhagic and ischaemic strokes (Hillbom and Kaste 1990; Berger et al. 1999; Patra et al. 2010). Acute intoxication with alcohol is associated with a thrombophilic tendency (Numminen et al. 2000). This may contribute to the association between binge drinking and both thrombotic strokes and myocardial infarctions. Fluid shifts and the increase in blood pressure may contribute to the increased incidence of subarachnoid haemorrhage and binge drinking (Gray et al. 1999). Acute alcohol poisoning may be accompanied by haemorrhage in the ventral diencephalon, mesencephalon and basal ganglia (de la Monte and Kril 2014). Severe cerebral oedema in the white matter may be seen in the cerebral hemispheres and the pontine and medullary tegmentum.

Thiamine deficiency The neuropathological features associated with thiamine deficiency are most commonly seen in people with alcohol problems but may be seen in other rare causes of extreme malnutrition (Ihara et al. 1999). Thiamine deficiency underlies WernickeKorsakoff syndrome and cerebellar degeneration. Clinically, Wernicke’s encephalopathy typically presents with ophthalmoplegia, ataxia and alterations in mental status, which may progress to coma and death in acute cases. Chronic cases may progress to Korsakoff’s psychosis, a rare irreversible psychiatric condition seen in chronic alcoholism. Cerebellar degeneration typically results in an ataxic gait and clumsiness, particularly of the lower limbs. It must be remembered that chronic alcoholism is frequently associated with recurrent falls caused by intoxication and that the clinical symptoms associated with Wernicke’s syndrome and chronic cerebellar degeneration are likely to compound the problem. Evidence of both recent and old head injury is common in chronic alcoholism.

Wernicke’s encephalopathy

Figure 19.2 Chronic Wernicke’s encephalopathy: the mamillary bodies are shrunken and show brown discolouration (arrows). hypothalamic region surrounding the third ventricle. Similar lesions may be seen in the periaqueductal region of the brainstem. In long-standing cases, often with recurrent episodes, the typical pathology is that of shrunken mamillary bodies with brown discolouration (Figure 19.2).

Microscopic examination In the acute phase, the principal pathology is that of perivascular haemorrhages in the regions described above, often associated with microvascular proliferation. Unlike ischaemic necrosis, the neuronal population is relatively well preserved in the mamillary bodies. In subacute and chronic cases, gliosis is best demonstrated by immunohistochemistry for glial fibrillary acidic protein. The increased density of capillaries can be demonstrated by a reticulin stain and Perls’ stain is useful for highlighting haemosiderin indicative of previous haemorrhage. Necrotic lesions are frequently seen in the thalamus, although detailed examination of thalamic nuclei is out with the normal neuropathological examination in forensic cases (Figure 19.3a and b).

Macroscopic examination

Alcohol and atrophy

It is important to remember that microscopic examination is mandatory because the brain may appear macroscopically normal in up to 25 per cent of cases. Areas of haemorrhagic necrosis are seen in the mamillary bodies (Figure 19.1) and in the

Neuroimaging and neuropathological studies have shown that alcoholics have atrophic brains in comparison with controls (Fadda and Rossetti 1998; Harper et al. 2003). Atrophy in alcoholics can be accounted for predominantly by white matter loss. Alcohol is known to damage myelin and reduces a number of myelin-related genes causing atrophy (Sutherland et al. 2014b,c). Prefrontal white matter is most significantly reduced. In uncomplicated alcoholics, brain weights have been shown to be reduced to 94 per cent of controls and in alcoholics with Wernicke/Korsakoff syndrome 91 per cent of controls. An association between atrophy and the rate and amount of alcohol consumed over a lifetime has been demonstrated. There is some evidence that atrophic change may revert to the same rate as control populations following abstinence (Pfefferbaum et al. 1998). Neuropathologists may be asked to comment on the role of atrophy in alcoholics and the development of subdural haematoma formation.

Cerebellar degeneration Macroscopic Figure 19.1 Acute Wernicke’s encephalopathy: haemorrhages are seen within the mamillary bodies.

Cerebellar atrophy, predominantly in the superior vermis, is a frequent finding in chronic alcoholism (Figure 19.4).

Other metabolic disorders related to alcohol

   Figure 19.3 (a) H and E, (b) Perls. Histological appearances of Wernicke’s encephalopathy. As well as HE, hepatic myelopathy may occur. It is a subacute syndrome and clinically patients present with a spastic paraparesis. Neuropathologically, there is symmetrical demyelination of the lateral columns with axonal loss in the corticospinal tracts (de la Monte and Kril 2014).

Central pontine myelinolysis

Figure 19.4 Marked superior vermal atrophy secondary to chronic alcoholism. The vermal section on the left is from an age-matched control and shows only very minor superior vermal atrophy. The vermal section on the right, from someone who is alcoholic, by contrast shows marked atrophy. Clinically this presented as an ataxic gait.

Microscopic Microscopically, there is obvious loss of Purkinje cells with associated Bergmann gliosis in the superior vermal region. Regular consumption of alcohol over a 20–30 year period with 41–80 g/day results in a 15 per cent reduction in the Purkinje cell population and double these amounts in a 33.4 per cent loss of Purkinje cells. Purkinje cell degeneration and white cell loss in the vermis correlate with clinical ataxia (Baker et al. 1999).

This disorder is seen in individuals with a variety of underlying disorders (Lampl and Yazdi 2002), although it is most commonly seen in people with chronic alcoholism. The disorder is thought to result from electrolyte abnormalities, with profound hyponatraemia being seen in most cases. Macroscopically, a pale region is identified in the centre of the basis pontis (Figure 19.5), and this corresponds to a region of pallor when examined microscopically with the Luxol fast blue (LFB)/cresyl violet stain (Figure 19.6). As with any demyelinating disorder, β-amyloid precursor protein immunoreactivity may be seen in relation to the area of pontine demyelination (Medana and Esiri 2003). It is now recognised that extrapontine lesions are seen in approximately 10 per cent of cases (Singh et al. 2014), particularly in relation to the thalamus and striatum.

Marchiafava-Bignami disease This is a very rare condition particularly seen in people with alcoholism from Mediterranean regions. Clinically the acute form is associated with seizures and loss of consciousness and usually results in death. The chronic form is unlikely to come to the pathologist’s attention. The lesion is characterised by

Other metabolic disorders related to alcohol Hepatic encephalopathy Hepatic failure is commonly due to alcoholic liver disease. This may be accompanied by hepatic encephalopathy (HE). In acute HE, there is swelling and oedema, which has been shown to be due to swelling of perivascular astrocytes ultrastructurally. HE is associated with Alzheimer type II change in astrocytes, the cells having enlarged optically clear nuclei. Chronic HE can be associated with pseudo-laminar spongy degeneration in the deep layers of the cerebral cortex (Victor et al. 1965).

Figure 19.5 Sections through the pons from a case of central pontine myelinolysis. The area of myelin loss is central within the basis pontis and is seen as a shrunken grey area.

197

198

Alcohol, drugs, toxins and post-mortem toxicology

Figure 19.6 Histological section demonstrating the central area of myelin loss (area of pallor) (Luxol fast blue/cresyl violet) in central pontine myelinolysis. an area of demyelination in the corpus callosum, particularly anteriorly in the corpus callosum. Microscopically, the lesions are highlighted by a myelin stain such as LFB, and macrophages are abundant (Figure 19.7).

Alcohol and trauma to the brain Alcohol has effects upon the brain that may increase the risk of sudden death. Traumatic brain injury is common in intoxicated individuals from falls, vehicular collisions and interpersonal violence. Certain traumatic conditions are particularly associated with alcohol intoxication including traumatic subarachnoid haemorrhage with tears in the basilar and vertebral arteries and alcohol concussive syndrome, discussed later. This ‘J’-shaped effect, with low levels of alcohol having a protective effect and higher levels an adverse effect, is also seen in acute intoxication with alcohol associated with head injury, at least in animal models. Alcohol concentrations of around 100 mg/100 mL may have a neuroprotective effect after head

Figure 19.7 Histological section from a case of MarchiafavaBignami disease. The corpus callosum is reduced in size and there is a central area of myelin pallor (arrow) (Luxol fast blue/cresyl violet).

injury in animal models, possibly as a result of the inhibition of the activity of N-methyl-d-aspartate by alcohol (Kelly et al. 1997; Janis et al. 1998). The stimulation by alcohol of corticosterone production by the adrenals may also have a protective effect (Gottesfeld et al. 2002). Higher alcohol concentrations, of around 220 mg/100 mL, are associated with depressed respiratory drive following experimental head injury. This may be associated with increased endogenous opiate production (Zink et al. 2001). The practical effect of this in humans is moot; other variables such as the quality of immediate care and the maintenance of oxygenation and haemodynamic status may confound the putative protective effects of modest alcohol concentrations and the deleterious effects of higher amounts of alcohol. In humans, higher alcohol concentrations at the time of injury tend to be associated with a poorer early cognitive outcome. Whether this is an acute toxic effect of alcohol or a result of the interaction of the effects of an acute head injury and chronic brain damage associated with regular alcohol use, or a combination of both, is a matter for ongoing research (Bombardier and Thurber 1998). Therefore, whether alcohol intoxication results in more severe brain injury or is protective remains unclear as studies are contradictory. There are a number of recorded instances of sudden death following blows to the head with minimal neuropathological findings, typically in the setting of alcohol intoxication (Milovanovic and DiMaio 1999; Shkrum and Ramsay 2000; Molina and DiMaio 2015). The mechanism of injury and death appears to be impact brain apnoea. This is also recognised clinically as well as by forensic pathologists, though overall remains under appreciated. Wilson et al. (2016) described two clinical cases without alcohol use where there was immediate medical care. Both had apnoea but recovered spontaneous respiration within a few minutes following mouth to mouth resuscitation and ventilation. Therefore, if there is limited or absent resuscitation, these people can die with minimal findings, and autopsy does not disclose an obvious cause of death, which requires correlation with the pre-autopsy history. If the person is resuscitated, they may develop hypoxic brain damage, but the cause of the hypoxia may not be determinable without recourse to the history.

Foetal alcohol spectrum disorder When alcohol is consumed by pregnant women, there is a risk of damage to the growing fetus. Originally called foetal alcohol syndrome and first described in 1973, it is now referred to as foetal alcohol spectrum disorder (FASD) (Sokol et al. 2003). Its incidence is believed to be between 1 and 7/1000 live births and is the most preventable cause of neurodevelopmental defects. The severity of the disorder is related to the quantity of alcohol consumed and the duration of consumption during the pregnancy. Patients with FASD have a characteristic facial dysmorphology with midfacial hypoplasia, long smooth philtrum, thin upper lip, small eyes and inner epicanthal folds. There is growth restriction, including a degree of microcephaly and ophthalmic involvement. Neuropathological changes include microcephaly, hypogenesis of the corpus callosum, changes in the basal ganglia and hippocampus and cerebellar hypoplasia. Microscopic changes include leptomeningeal heterotopias, thinning, laminar disorganisation and evidence of neuronal and glial cell migration disorder in the cerebral cortex, agenesis and hypogenesis or

Other metabolic disorders related to alcohol thinning of the corpus callosum and anterior commissure, dysgenesis of the cerebellum and occasionally the brainstem, heterotopias in the cerebellum, hypoplasia and agenesis of the cerebellar vermis and hydrocephalus (de la Monte and Kril 2014). There can be a congenital absence of the olfactory bulbs and tracts, hippocampal hypoplasia, malformation of the basal ganglia and rarely cerebral dysgenesis with variable degrees of holoprosencephaly.

Methanol Methanol is found in a number of common products (solvents, toiletries) that can be drunk to produce intoxication and can be produced in the home brewing of spirits. Toxicity in the acute phase can lead to blindness and cardiorespiratory collapse. At post-mortem examination, the brain is swollen, and petechial haemorrhages may be seen. The haemorrhages are particularly pronounced in relation to the third and fourth ventricles and can also be subarachnoid. If the patient has survived for a period of time before death, bilateral necrosis of the putamen may be seen, and there is necrosis of the cerebellar cortex (McLean et al. 1980). In the eyes, there is loss of retinal ganglion cells and optic nerve axons (Naeser 1988).

Ethylene glycol Ethylene glycol is considered under alcohol, because it is usually consumed in antifreeze by people who are alcoholics. At postmortem examination there is non-specific cerebral oedema and petechial haemorrhages. In the central nervous system, oxalate crystals can be seen in the brain and meninges in vessels and in perivascular spaces (Figure 19.8).

Recreational drugs Opioids Opioids consist of naturally occurring opiates and synthetic opioids, which are commonly abused recreational drugs that can lead to death from drug toxicity. While opiates such as opium, morphine and heroin (diamorphine – a semisynthetic opiate) were traditionally the most widely misused, now synthetic opioids are widely available and powerful opioids like fentanyl are now responsible for a significant proportion of drug-related deaths. Opioid abuse can result in a wide range of lesions within the nervous system. Some of these lesions may be directly caused by opioids, whereas many others are the result

Figure 19.8 Oxalate crystals in a case of ethylene glycol poisoning.

of the impurities mixed with the drug. Drugs may be taken orally, insufflated, injected or smoked. Many impurities can be introduced into the bloodstream and can travel to the brain or spinal cord. Non-sterile injection techniques can introduce infection that, via haematogenous spread, can lead to cerebral abscesses or meningitis. HIV infection may result in central nervous system pathology (see Chapter 18). Cerebral infarcts may be seen as a consequence of emboli of injected materials (Karch and Drummer 2016). Most acute opioid deaths have no specific neuropathological findings. The most common pathology seen in relation to a fatal opioid overdose is, however, that of ischaemic damage secondary to hypotension, as described in Chapter 13. The brain is swollen, and lesions are seen at the parieto-occipital watershed region and often within the globus pallidus (Andersen and Skullerud 1999; Buttner et al. 2000). Hypoxic brain damage may occur in opioid users who are resuscitated following the administration of reversal agents such as naloxone, where cardiorespiratory arrest has been prolonged. Spongiform leukoencephalopathy was first reported in the Netherlands in 1982 amongst heroin smokers (Wolters et al. 1982), and subsequently cases have been reported from other countries. The brains were reported as oedematous, and on microscopy, there was damage to the white matter with vacuolation. There was attenuation of myelin around cavities formed by coalescence of the vacuoles. Oligodendrocytes were reduced in number. No inflammatory cells were present. MPTP (1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine) has been produced by clandestine laboratories attempting to produce MPPP (1-methyl-4-phenyl-4-propionoxypiperidine), which is a meperidine analogue. Meperidine is a synthetic opiate. MPTP is converted by astrocytes into a toxic compound. When taken in sufficient quantities, MPTP has been associated with Parkinsonism (Langston et al. 1983). Degenerative changes can be seen in the substantia nigra without Lewy body formation. Cases have been reported in intravenous drug taking populations using what they believed was synthetic heroin and in industrial chemists.

Cocaine Cocaine abuse is common and cocaine use is now recognised as a common cause of cerebrovascular pathology particularly in younger adults. Cerebral infarction is now well recognised as a complication of cocaine use. Many cases follow within 3 hours of cocaine use. Most individuals are in their mid-30s in comparison with non-drug related cerebral infarction where individuals are characteristically over 65 years of age (Karch and Drummer 2016). Although vasculitis has been reported as a cause of cerebrovascular disease in cocaine users, it appears to be rare and may be related to levamisole adulteration of cocaine (Brunt et al. 2017). Cocaine has been shown to cause vasoconstriction using MRI. Ischaemic infarcts may be the result of vasoconstriction, increased platelet aggregation and activation. Subarachnoid haemorrhage is seen in cocaine users (Winhusen et al. 2020). The majority are associated with berry aneurysms. Intracerebral haemorrhage occurs most commonly in the basal ganglia and thalamus, but may also occur in the cerebellum and subcortical white matter. With intracerebral haemorrhage, there is an association with hypertensive heart disease, heart weights in cocaine users dying of intracerebral haemorrhage being heavier than those with ruptured aneurysms. There is also a high incidence of underlying vascular

199

200

Alcohol, drugs, toxins and post-mortem toxicology anomalies and malformations (Buttner et al. 2003). A study by Fessler et al. (1997) revealed that of 33 patients with neurological symptoms following cocaine abuse nearly half (16 of 33) had subarachnoid haemorrhage, six intracerebral haemorrhage and seven cerebral infarction. With intracranial haemorrhage where there is a delayed between incident and death and alcohol or drugs use is suspected, the haematoma can be sent for toxicological examination (Hirsch and Adelson 1973; McIntyre et al. 2003). Cocaine abuse during pregnancy has been associated with foetal brain haemorrhage (Kapur et al. 1991). The actual incidence of vascular lesions varies between different published series, but there does appear to be an increased risk of foetal brain haemorrhages and cerebral infarcts. Cocaine use has been associated with movement disorders including choreoathetosis, akathisia and Parkinsonism and may cause grand mal seizures. Cocaine is associated with excited delirium, and changes in neurotransmitters have been reported in these deaths, including altered dopamine receptor status.

Amphetamines The amphetamines are a group of manufactured stimulants including amphetamine, methamphetamine and the amphetamines derivatives including 3,4-methylenedioxymethamphetamine (MDMA) and related drugs. Amphetamine and methamphetamine are easy to synthesise illicitly and are widely abused stimulants. A variety of amphetamine derivatives have been synthesised, and many have nonstimulant mind-altering properties that make them desirable as drugs of misuse. Like cocaine, amphetamines are particularly associated with effects on the cardiovascular and cerebrovascular systems. Haemorrhagic and ischaemic strokes are associated with amphetamine use. Intracerebral haemorrhage is commoner in the frontal lobes and may also be seen in the basal ganglia (Karch and Drummer 2016) (Figure 19.9). The neuropathological features related to MDMA use identified at post-mortem examination include severe cerebral oedema in water intoxication, and microscopic perivascular haemorrhages in cases of heatstroke (Milroy et al. 1996). In addition, intracerebral haemorrhages have been described on imaging and in one case there was a fatal outcome in an individual with an underlying arteriovenous malformation (Harries and De Silva 1992).

Novel psychoactive substances New drugs are always emerging into the recreational drug use scene and are known as novel psychoactive substances (NPS) (Schifano et al. 2015; Logan et al. 2017). The term was first used in respect of heroin-like derivatives, but when drugs such as MDMA emerged in the 1980s, they were also referred to as designer drugs. The term NPS been used for a number of different substances with different actions. Various groups of drugs that have emerged include synthetic cannabinoids, designer benzodiazepines, synthetic cathinones, novel hallucinogens and designer opioids being the main groups. These drugs have all been associated with adverse effects. Synthetic cannabinoids may cause central nervous symptom symptoms and signs, including delirium and seizure activity and have caused death. Synthetic cathinones are stimulants and can cause hyperpyrexia, hypertension and an excited delirium (ED) picture. Novel hallucinogens can also cause hyperpyrexia, hypertension and delirium. Designer benzodiazepines have similar actions to traditional benzodiazepines, and synthetic opioids have actions like other opioids and have been associated with many adverse events and deaths. Neuropathological changes similar to other stimulant and opioid drugs can be expected but neuropathological studies have not yet been reported.

Excited delirium Excited delirium (ED) is a clinico-pathological condition that presents to emergency medicine specialists and forensic pathologists (Gill 2014). It involves an agitated person with often violent, disordered and bizarre behaviour who by their behaviour often comes into contact with law enforcement agencies and healthcare professionals. Typically there is drug intoxication with a stimulant drug, most frequently cocaine and methamphetamine, but is also associated with psychiatric disorders. It is more common in warm weather and hyperpyrexia is a feature. It may be fatal and, when there has been engagement with law enforcement officers, controversial. Sudden death may occur during restraint, and the autopsy often fails to disclose a cause of death. The autopsy in ED is often non-specific or may have findings related to struggle and restraint. Toxicological analysis must be performed in fatal cases. The brain should be submitted for neuropathological examination in these cases to exclude neurotrauma or other underlying disorder. Examination for heat shock proteins (Hsp70) (Mash et al. 2009) and dopamine transporters has shown them to be elevated, though Johnson et al. (2012) questioned whether the elevated levels were a function of survival rather than ED. They pointed out that Hsp70 is rapidly induced by a number of cellular stresses, and they speculated that the rise in this biomarker was an (unsuccessful) adaptive response to limit cocaine neurotoxicity.

Neurotoxins Neurotoxins are discussed under the following headings: gases, heavy metals and organic solvents.

Gases Carbon monoxide

Figure 19.9 Intracerebral haemorrhage in a 33-year-old amphetamine user.

Carbon monoxide (CO) poisoning is still seen as a frequent form of suicide and in relation to poorly maintained gas heaters. Non-combustion sources of CO may be encountered and include the consumption of dichloromethane. At post-mortem examination, the brain is swollen and has a characteristic pinkred colour caused by the accumulation of carboxyhaemoglobin.

References Where there has been survival for several days, infarcts are seen within the globus pallidus and substantia nigra. Histologically, neuronal loss is common in the hippocampus and Purkinje cell layer of the cerebellar cortex (Betterman and Patel 2014). In long-term survivors, a myelinopathy may develop (Grinker’s myelinopathy), the damage being seen throughout the cerebrum, although being most severe posteriorly and deeply (Lapresle and Fardeau 1966).

Other gases Carbon disulphide (CS2) (Ku et al. 2003) and methyl bromide (de Souza et al. 2013) have been documented to cause fatal intoxications in industrial settings. CS2 has been used in textile and rubber industries, whereas methyl bromide was a component of fire extinguishers. Both can result in sudden loss of consciousness and coma. Post-mortem studies have demonstrated neuronal loss and gliosis. Methyl bromide intoxication has been associated with petechial haemorrhages and cerebral oedema in the acute stages, and in one case with 30-day survival, the lesions resembled Wernicke’s encephalopathy.

Heavy metals Arsenic Arsenic has been a classic poison in homicides, with such poisoners as Armstrong and the Seddons being notable British cases. Arsenic is available in a trivalent form and an organic pentavalent form. In acute deaths, the findings may be minimal. Where poisoning is more chronic, symptoms relating to the gastrointestinal tract may result in a mistaken diagnosis of gastroenteritis. A peripheral neuropathy may develop with Wallerian degeneration seen in the distal nerves. Whilst homicidal arsenic poisoning is rare, it still occurs and the last case seen by the authors involved the mistaken diagnosis of gastroenteritis. A peripheral neuropathy was ascribed to diabetes mellitus by the clinicians. Post-mortem toxicology confirmed arsenic poisoning with hair analysis confirming repeated administration.

Organic arsenic Organic arsenic has been used in the treatment of trypanosomiasis and can be associated with fatal intoxication. Acute haemorrhagic leukoencephalitis has been described with perivascular demyelination and ‘ball and ring’-type haemorrhages particularly in the pons and midbrain (Adams et al. 1986).

Mercury Mercury intoxication has been described in two situations: after industrial pollution of the environment resulting in the element being introduced into the human food chain and the use of mercury as a fungicide. Intoxication after industrial pollution was first described in Minamata in Japan in the mid-1950s (Takeuchi 1982), and intoxication with fungicide was the cause of many deaths in Iraq in 1971 (Bakir et al. 1973). The neuropathological features associated with mercury intoxication include severe atrophy of the calcarine cortex and cerebellar cortex, with less severe cortical atrophy elsewhere in the cerebrum. Within the cerebellum, the atrophy is most pronounced in the granule cell layer with preservation of the Purkinje cells. Axonal torpedoes can be seen in Purkinje cells and spiny dendritic processes have been described.

Thallium Thallium is used in the glass and semiconductor industry and, rarely in the developed world, as a pesticide and has been described in accidental, suicidal and homicidal poisoning.

The clinical presentation is of unexplained gastrointestinal disturbance followed by a peripheral neuropathy. Unexpected hair loss may occur. Motor weakness and cranial neuropathies can develop. Uniquely it has been identified in cremated remains following homicidal poisoning and may also be identified at autopsy by radiology, as it accumulates in the liver and is radiopaque. The neuropathological features are rather non-specific and include cerebral oedema with petechial haemorrhages. Chromatolysis of motor neurons is seen and sampling of the distal peripheral nervous system may show loss of myelinated axons (Cavanagh et al. 1974).

Other heavy metals Other heavy metals have been described to cause an encephalopathy in the past but are now rarely seen. These include aluminium (dialysis encephalopathy) and lead (used in paint and associated with childhood intoxication). Others are seen in only very specific situations (manganese intoxications have been described in manganese miners).

References Adams, J.H., Haller, L., Boa, F.Y., Doua, F., Dago, A. and Konian, K. 1986. Human African trypanosomiasis (T. b. gambiense): a study of 16 fatal cases of sleeping sickness with some observations on acute reactive arsenical encephalopathy. Neuropathol Appl Neurobiol, 12(1), 81–94. Andersen, S.N. and Skullerud, K. 1999. Hypoxic/ischaemic brain damage, especially pallidal lesions, in heroin addicts. Forensic Sci Int, 102(1), 51–59. Baker, K.G., Harding, A.J., Halliday, G.M., Kril, J.J. and Harper, C.G. 1999. Neuronal loss in functional zones of the cerebellum of chronic alcoholics with and without Wernicke’s encephalopathy. Neuroscience, 91(2), 429–438. Bakir, F., Damluji, S.F., Amin-Zaki, L., Murtadha, M., Khalidi, A. and al-Rawi, N.Y. et al. 1973. Methylmercury poisoning in Iraq. Science, 181(4096), 230–241. Berger, K., Ajani, U.A., Kase, C.S., Gaziano, J.M., Buring, J.E. and Glynn, R.J. et al. 1999. Light-to-moderate alcohol consumption and the risk of stroke among U.S. male physicians. N Engl J Med, 341(21), 1557–1564. Betterman, K. and Patel, S. 2014. Neurologic complications of carbon monoxide intoxication. Handb Clin Neurol, 120, 971–979. Bombardier, C.H. and Thurber, C.A. 1998. Blood alcohol level and early cognitive status after traumatic brain injury. Brain Inj, 12(9), 725–734. Brunt, T.M., van den Berg, J., Pennings, E. and Venhuis, B. 2017. Adverse effects of levamisole in cocaine users: a review and risk assessment. Arch Toxicol, 91(6), 2303–2313. Buttner, A., Mall, G., Penning, R., Sachs, H. and Weis, S. 2003. The neuropathology of cocaine abuse. Leg Med (Tokyo), 5(Suppl 1), S240–S242. Buttner, A., Mall, G., Penning, R. and Weis, S. 2000. The neuropathology of heroin abuse. Forensic Sci Int, 113(1–3), 435–442. Cavanagh, J.B., Fuller, N.H., Johnson, H.R. and Rudge, P. 1974. The effects of thallium salts, with particular reference to the nervous system changes. A report of three cases. Q J Med, 43(170), 293–319. de la Monte, S.M. and Kril, J.J. 2014. Human alcohol-related neuropathology. Acta Neuropathol, 127(1), 71–90. de Souza, A., Narvencar, K.P. and Sindhoora, K.V. 2013. The neurological effects of methyl bromide intoxication. J Neurol Sci, 335(1–2), 36–41. Dubowski, K.M. 1980. Alcohol determination in the clinical laboratory. Am J Clin Pathol, 74(5), 747–750. Fadda, F. and Rossetti, Z.L. 1998. Chronic ethanol consumption: from neuroadaptation to neurodegeneration. Prog Neurobiol, 56(4), 385–431. Fessler, R.D., Esshaki, C.M., Stankewitz, R.C., Johnson, R.R. and Diaz, F.G. 1997. The neurovascular complications of cocaine. Surg Neurol, 47(4), 339–345. Gill, J.R. 2014. The syndrome of excited delirium. Forensic Sci Med Pathol, 10(2), 223–228. Gottesfeld, Z., Moore, A.N. and Dash, P.K. 2002. Acute ethanol intake attenuates inflammatory cytokines after brain injury in rats: a possible role for corticosterone. J Neurotrauma, 19(3), 317–326.

201

202

Alcohol, drugs, toxins and post-mortem toxicology Gray, J.T., Puetz, S.M., Jackson, S.L. and Green, M.A. 1999. Traumatic subarachnoid haemorrhage: a 10-year case study and review. Forensic Sci Int, 105(1), 13–23. Harper, C., Dixon, G., Sheedy, D. and Garrick, T. 2003. Neuropathological alterations in alcoholic brains. Studies arising from the New South Wales Tissue Resource Centre. Prog Neuropsychopharmacol Biol Psychiatry, 27(6), 951–961. Harries, D.P. and De Silva, R. 1992. ‘Ecstasy’ and intracerebral haemorrhage. Scott Med J, 37(5), 150–152. Heatley, M.K. and Crane, J. 1990. The blood alcohol concentration at post-mortem in 175 fatal cases of alcohol intoxication. Med Sci Law, 30(2), 101–105. Hillbom, M. and Kaste, M. 1990. Alcohol abuse and brain infarction. Ann Med, 22(5), 347–352. Hirsch, C.S. and Adelson, L. 1973. Ethanol in sequestered hematomas. Am J Clin Pathol, 59(3), 429–433. Ihara, M., Ito, T., Yanagihara, C. and Nishimura, Y. 1999. Wernicke’s encephalopathy associated with hemodialysis: report of two cases and review of the literature. Clin Neurol Neurosurg, 101(2), 118–121. Janis, L.S., Hoane, M.R., Conde, D., Fulop, Z. and Stein, D.G. 1998. Acute ethanol administration reduces the cognitive deficits associated with traumatic brain injury in rats. J Neurotrauma, 15(2), 105–115. Johnson, M.M., David, J.A., Michelhaugh, S.K., Schmidt, C.J. and Bannon, M.J. 2012. Increased heat shock protein 70 gene expression in the brains of cocaine-related fatalities may be reflective of postdrug survival and intervention rather than excited delirium. J Forensic Sci, 57(6), 1519–1523. Johnson, R.A., Noll, E.C. and Rodney, W.M. 1982. Survival after a serum ethanol concentration of 1 1/2%. Lancet, 2(8312), 1394. Jones, A.W. 1999. The drunkest drinking driver in Sweden: blood alcohol concentration 0.545% w/v. J Stud Alcohol, 60(3), 400–406. Jones, A.W. and Holmgren, P. 2003. Comparison of blood-ethanol concentration in deaths attributed to acute alcohol poisoning and chronic alcoholism. J Forensic Sci, 48(4), 874–879. Kapur, R.P., Shaw, C.M. and Shepard, T.H. 1991. Brain hemorrhages in cocaine-exposed human fetuses. Teratology, 44(1), 11–18. Karch, S.B. and Drummer, O.H. 2016. Karch’s Pathology of Drug Abuse, 5th edn. CRC Press, Taylor & Francis Group. Kelly, D.F., Lee, S.M., Pinanong, P.A. and Hovda, D.A. 1997. Paradoxical effects of acute ethanolism in experimental brain injury. J Neurosurg, 86(5), 876–882. Koski, A., Ojanpera, I. and Vuori, E. 2002. Alcohol and benzodiazepines in fatal poisonings. Alcohol Clin Exp Res, 26(7), 956–959. Ku, M.C., Huang, C.C., Kuo, H.C., Yen, T.C., Chen, C.J. and Shih T.S. et al. 2003. Diffuse white matter lesions in carbon disulfide intoxication: microangiopathy or demyelination. Eur Neurol, 50(4), 220–224. Lampl, C. and Yazdi, K. 2002. Central pontine myelinolysis. Eur Neurol, 47(1), 3–10. Langston, J.W., Ballard, P., Tetrud, J.W. and Irwin, I. 1983. Chronic Parkinsonism in humans due to a product of meperidine-analog synthesis. Science, 219(4587), 979–980. Lapresle, J. and Fardeau, M. 1966. The leukoencephalopathies caused by carbon monoxide poisoning. Study of sixteen anatomo-clinical observations. Acta Neuropathol, 6(4), 327–348. Logan, B.K., Mohr, A.L.A., Friscia, M., Krotulski, A.J., Papsun, D.M. and Kacinko, S.L. et al. 2017. Reports of adverse events associated with use of novel psychoactive substances, 2013–2016: a review. J Anal Toxicol, 41(7), 573–610. Mash, D.C., Duque, L., Pablo, J., Qin, Y., Adi, N. and Hearn, W.L. et al. 2009. Brain biomarkers for identifying excited delirium as a cause of sudden death. Forensic Sci Int, 190(1–3), e13–e19. McIntyre, I.M., Hamm, C.E., Sherrard, J.L., Gary, R.D., Riley, A.C. and Lucas, J.R. 2003. The analysis of an intracerebral hematoma for drugs of abuse. J Forensic Sci, 48(3), 680–682. McLean, D.R., Jacobs, H. and Mielke, B.W. 1980. Methanol poisoning: a clinical and pathological study. Ann Neurol, 8(2), 161–167. Medana, I.M. and Esiri, M.M. 2003. Axonal damage: a key predictor of outcome in human CNS diseases. Brain, 126(Pt 3), 515–530.

Mellanby, E., 1919. Alcohol: Its absorption into and disappearance from the blood under different conditions (Medical Research Committee Special Report Series No. 31). Milovanovic, A.V. and DiMaio, V.J. 1999. Death due to concussion and alcohol. Am J Forensic Med Pathol, 20(1), 6–9. Milroy, C.M., Clark, J.C. and Forrest, A.R. 1996. Pathology of deaths associated with “ecstasy” and “eve” misuse. J Clin Pathol, 49(2), 149–153. Molina, D.K. and DiMaio, V.J. 2015. Head trauma and alcohol: a lethal combination. Am J Forensic Med Pathol, 36(4), 290–292. Naeser, P. 1988. Optic nerve involvement in a case of methanol poisoning. Br J Ophthalmol, 72(10), 778–781. Numminen, H., Syrjala, M., Benthin, G., Kaste, M. and Hillbom, M. 2000. The effect of acute ingestion of a large dose of alcohol on the hemostatic system and its circadian variation. Stroke, 31(6), 1269–1273. Patra, J., Taylor, B., Irving, H., Roerecke, M., Baliunas, D. and Mohapatra, S. et al. 2010. Alcohol consumption and the risk of morbidity and mortality for different stroke types – a systematic review and meta-analysis. BMC Public Health, 10, 258. Pfefferbaum, A., Sullivan, E.V., Rosenbloom, M.J., Mathalon, D.H. and Lim, K.O. 1998. A controlled study of cortical gray matter and ventricular changes in alcoholic men over a 5-year interval. Arch Gen Psychiatry, 55(10), 905–912. Schifano, F., Orsolini, L., Duccio Papanti, G. and Corkery, J.M. 2015. Novel psychoactive substances of interest for psychiatry. World Psychiatry, 14(1), 15–26. Scholl, L., Seth, P., Kariisa, M., Wilson, N. and Baldwin, G. 2018. Drug and opioid-involved overdose deaths – United States, 2013–2017. MMWR Morb Mortal Wkly Rep, 67(5152), 1419–1427. Shkrum, M.J. and Ramsay, D.A. 2000. Death due to concussion and alcohol. Am J Forensic Med Pathol, 21(1), 94. Singh, T.D., Fugate, J.E. and Rabinstein, A.A. 2014. Central pontine and extrapontine myelinolysis: a systematic review. Eur J Neurol, 21(12), 1443–1450. Sokol, R.J., Delaney-Black, V. and Nordstrom, B. 2003. Fetal alcohol spectrum disorder. JAMA, 290(22), 2996–2999. Sutherland, G.T., Sheedy, D. and Kril, J.J. 2014a. Neuropathology of alcoholism. Handb Clin Neurol, 125, 603–615. Sutherland, G.T., Sheedy, D. and Kril, J.J. 2014b. Using autopsy brain tissue to study alcohol-related brain damage in the genomic age. Alcohol Clin Exp Res, 38(1), 1–8. Sutherland, G.T., Sheedy, D., Sheahan, P.J., Kaplan, W. and Kril, J.J. 2014c. Comorbidities, confounders, and the white matter transcriptome in chronic alcoholism. Alcohol Clin Exp Res, 38(4), 994–1001. Takeuchi, T. 1982. Pathology of Minamata disease. With special reference to its pathogenesis. Acta Pathol Jpn, 32(Suppl 1), 73–99. Victor, M., Adams, R.D. and Cole, M. 1965. The acquired (non-Wilsonian) type of chronic hepatocerebral degeneration. Medicine (Baltimore), 44(5), 345–396. Weis, S. and Buttner, A. 2017a. Alcohol-related diseases. Handb Clin Neurol, 145, 175–180. Weis, S. and Buttner, A. 2017b. Neurotoxicology and drug-related disorders. Handb Clin Neurol, 145, 181–192. WHO. 2019. https://wdr.unodc.org/wdr2019/en/exsum.html Wilson, M.H., Hinds, J., Grier, G., Burns, B., Carley, S. and Davies, G. 2016. Impact brain apnoea – a forgotten cause of cardiovascular collapse in trauma. Resuscitation, 105, 52–58. Winhusen, T., Theobald, J., Kaelber, D.C. and Lewis, D. 2020. The association between regular cocaine use, with and without tobacco co-use, and adverse cardiovascular and respiratory outcomes. Drug Alcohol Depend, 214, 108136. Wolters, E.C., van Wijngaarden, G.K., Stam, F.C., Rengelink, H., Lousberg, R.J. and Schipper, M.E. et al. 1982. Leucoencephalopathy after inhaling “heroin” pyrolysate. Lancet, 2(8310), 1233–1237. Zink, B.J., Schultz, C.H., Stern, S.A., Mertz, M., Wang, X. and Johnston, P. et al. 2001. Effects of ethanol and naltrexone in a model of traumatic brain injury with hemorrhagic shock. Alcohol Clin Exp Res, 25(6), 916–923.

20

The role of the expert witness Paul Watson and Christopher Milroy

Introduction Anyone can be a witness of fact in a court case, but common law has developed rules about who can give an opinion as a witness, as opposed to the facts the opinion is based upon (Munday 2019). This chapter examines the role of the expert and discusses legal principles relating to the expert witness in a number of jurisdictions. The law on such matters as disclosure may vary between jurisdictions and in giving evidence the expert should familiarise themselves with the legal duties in the case they are involved in.

The admissibility of opinion evidence In general, evidence that does not relate to matters within the immediate experience of the witness (namely, what he has seen and heard) is inadmissible in common law. This is known as the hearsay rule. However, there are of necessity exceptions to this principle. Among the exceptions is the rule that, if a witness can be said to have a sufficient level of expertise in a particular field of knowledge that can be said to be outside the experience or knowledge of the trier of fact, witness may furnish the court with such information as is relevant to the issues of fact before the court and may assist the court by the giving of evidence of opinion as to matters that fall within the witness’s expertise. In a jury trial, the issue as to whether the matter in question calls for expertise, and whether the proposed expertise is sufficiently acknowledged, will be a matter for the judge, leaving it to the jury to decide on the assistance to be derived from it. The approach to the admissibility of novel science has been the subject of a number of significant legal judgements in multiple jurisdictions. A conservative approach was adopted in the important decision of Frye v United States (1923). In 1975, in the United States, the Federal Rules of Evidence were enacted (Federal Rules of Evidence 2019). These were modified following the Supreme Court decision in Daubert v Merrell Dow Pharmaceuticals, Inc. (1993) and was followed by the decision in Kumho Tire Co. v. Carmichael (1999), which clarified that all expert evidence, not just scientific evidence, is subject to the rules laid down in Daubert. Rule 702 governs admission of expert evidence and following Daubert currently states: A witness who is qualified as an expert by knowledge, skill, experience, training or education may testify in the form of an opinion or otherwise if: a. The expert’s scientific, technical or other specialised knowledge will help the trier of fact to understand the evidence or to determine a fact in issue. b. The testimony is based on sufficient facts or data. c. The testimony is the product of reliable principles and methods. d. The expert has reliably applied the principles and methods to the facts of the case.

Daubert remains the leading case on expert evidence in the United States and has had significant impact on common law jurisdictions outside the United States. In R v J-L J (2000), the Supreme Court of Canada endorsed Daubert. The key to the admissibility of such evidence is, first, whether the subject matter of the proposed evidence is such that a person without instruction or experience in the area of knowledge concerned would be able to form a sound judgement on the matter without the assistance of a witness possessing special knowledge in the area and, second, whether the subject matter of the opinion forms part of a body of knowledge or experience that is sufficiently recognised to be accepted as a reliable body of experience on which the opinion of the expert would be of assistance to the court. Of course, once a court has found that expert evidence is admissible, there follows the further, and quite separate, question as to whether the proposed expert is in fact possessed of the relevant degree of knowledge to render his or her evidence and opinion of sufficient value in resolving the issues before the court. Whether or not an expert is possessed of the relevant degree of expertise to be competent to give such evidence is a matter for the judge to determine. Once decided that such a witness is competent, the issue as to the weight to be given to the witness’s evidence and/or opinion is a matter for the tribunal of fact. In 2009, the England Court of Appeal in R v Reed, Reed and Garmson (2009) summarised admissibility of expert evidence as: It is important to distinguish the issue of the admissibility of expert evidence from the assessment of that evidence by the jury. In the present appeal, the issue related to admissibility. First, expert evidence of a scientific nature is not admissible where the scientific basis on which it is advanced is insufficiently reliable for it to be put before the jury. There is, however, no enhanced test of admissibility for such evidence. If the reliability of the scientific basis for the evidence is challenged, the court will consider whether there is a sufficiently reliable scientific basis for that evidence to be admitted, but, if satisfied that there is a sufficiently reliable scientific basis for the evidence to be admitted, then it will leave the opposing views to be tested in the trial.

Second, even if the scientific basis is sufficiently reliable, the evidence is not admissible unless it is within the scope of evidence an expert can properly give. Third, unless the admissibility is challenged, the judge will admit that evidence. That is the only pragmatic way in which it is possible to conduct trials, as sufficient safeguards are provided by Part 3 and Part 33 of the Criminal Procedure Rules. However, if objection to the admissibility is made, then it is for the party proffering the evidence to prove its admissibility. In Canada, the leading case on expert admissibility is the Supreme Court case R v Mohan (1994). This firmly states the gatekeeping role the judge plays in admissibility in Canada and

204

The role of the expert witness also states that expert evidence is admissible in Canada when four criteria are established: a. Relevance of the evidence b. The necessity of the evidence in assisting the trier of fact c. The absence of any exclusionary rule to the reception of evidence d. The proposed expert being properly qualified.

The role of the expert In a lecture to the Middle Temple, Lord Hodge stated that the expert witness has both privilege and power – privilege to give their opinion and power because of their knowledge and the difficulties a lawyer has in cross-examination of a well-prepared expert (Lord Hodge 2017). Because of their power and privilege, experts have responsibilities as well. The main responsibility is for the expert to assist the fact-finding tribunal as to the issues that it has to decide. The notion, if it ever existed, that the role of the expert witness was to assist the party whose cause he or she had espoused has gone. This view is now enshrined in England, in the context of civil litigation, in Part 35.3 of the Civil Procedure Rules, which provides as follows: Overriding duty to the court: 1. It is the duty of an expert to help the court on the matter within his expertise. 2. This duty overrides any obligation to the person from whom he has received instructions. In the context of family litigation (in cases involving children), it has been said that an expert witness bears a heavy responsibility to be as objective as possible, and to make clear the extent to which his or her opinion is based on hypothesis, especially where this view departs from the consensus opinion [see Re AB (A Minor) 1995]. These principles apply in criminal as well as civil litigation. Expert reports in criminal cases now routinely contain declarations about the duties enshrined in Rule 35.3 above. In practice, the duty of the expert is ‘to furnish the judge or jury with the necessary scientific criteria for testing the accuracy of their conclusions, so as to enable the judge or jury to form their own independent judgement as to the accuracy of their conclusions’. Davie v Magistrates of Edinburgh (1953) apply with equal force in whichever jurisdiction the evidence is given. This Scottish case also reconfirmed the principle that an expert may adopt statements made in scientific works (which would otherwise breach the hearsay rule) and have such works put to them in cross-examination. The weight that will be attached to the evidence of an expert witness will invariably be greater in those cases in which the expert has demonstrated their objectivity. The expert who has demonstrated partisanship, a personal agenda or an unblinking adherence to a particular hypothesis may too easily be discredited in the eyes of the fact-finding tribunal, so that, even if there is some significant merit in the view to which he or she holds, it may be significantly undermined or even rejected outright. Bias was examined in the Canadian case of White Burgess Langille Inman v Abbott and Halibuton Co. (2015). Bias on the part of the expert may be viewed as disqualifying them as an expert as well as to the subsequent question of whether the evidence in admissible, which is echoed in the UK Supreme Court case of Kennedy (Appellant) v Cordia (Services) LLP (Respondent) (Scotland) (2016).

The need for objectivity applies with perhaps greater force to evidence of medical expertise than to any other area of knowledge. Medical expert opinion is often readily accepted by the tribunal of fact, in particular by juries, precisely because of the respectable and respected body of opinion from which it derives. There tends to be less suspicion, on the part of either judges or jurors, as to the quality and credibility of medical expertise than there is in relation to other sources of opinion evidence, and it is for this very reason that the expert needs to be aware of the need for self-critical objectivity and an awareness of the weaknesses, as well as the strengths, of the view that he or she feels able to express. Moreover, if that point applies to expert evidence arising from general medical practice, it applies with greater force in the more specific and technical fields of pathology, in which failure to present all relevant information may lead to conclusions by the judge or jury that are not, in the context of all the other evidence in the case, justified. Proof of facts on which the expert opinion is based. Before a court can assess the value of an opinion it must know the facts upon which it is based. If the expert has been misinformed about the facts or has taken irrelevant facts into consideration or has omitted to consider relevant ones, the opinion is likely to be valueless. In our judgement, counsel calling an expert should in examination-in-chief ask his witness to state the facts upon which his opinion is based. It is wrong to leave the other side to elicit the facts by cross-examination. per Lawton LJ, R v Turner (1975) at p. 840.

Today, this statement of general principle can be expanded to refer to the desirability of experts, including in their reports the facts on which their opinion is based and, as importantly, the source of that information. In providing evidence an expert may give both give expert factual evidence either by itself or in combination with opinion evidence. This was reiterated in the UK Supreme Court case of Kennedy (Appellant) v. Cordia (Services) LLP (Respondent) (Scotland) (2016). Of course, much of the relevant material that will form the basis of expert opinions will derive from observations and findings made by the experts themselves. To a perhaps lesser extent, the expert may rely on what he or she has been told by others. Such evidence is admissible as being relevant to the process leading to the formulation of his opinion but, if no direct evidence as to the truth of those facts is called, or if the quality of the evidence of those facts is poor, the value of the opinion may suffer correspondingly. For this reason, if for this reason only, it is submitted that prudent experts will make some critical enquiry as to the information that they are given before placing undue reliance on it in formulating their view. It should go without saying that, where experts rely not on evidence of observation or information supplied by a third party, but on assumptions made for the purpose of the preparation of their report, they should say so in the body of that report and not await questioning as to the source of the information at a later stage. It is now clear that expert witnesses may make references to their own experience, research, the work of others, articles in respected publications, research papers and other such materials in forming their views. This principle was reiterated in the Privy Council case of Myers, Brangman and Cox (Appellants) v The Queen (Respondent) (Bermuda) (2015). In the case of R v Abadom (1983), evidence was permitted to be adduced of unpublished statistics of the Home Office Central Research Establishment, it being noted by the Court of Appeal that part of the value of

Pre-trial disclosure in criminal cases expert witnesses lies in their knowledge of such unpublished material; there can be no reason for not permitting experts to rely on such material, provided that they refer to it in their evidence so that the value of their conclusions can be tested by reference to it. Such reference does not infringe the rule against hearsay because it does no more than provide the basis on which the conclusions of the expert are drawn. Wherever reference is made to such material, experts should specify the source material in the body of their report – preferably by means of a short bibliography.

Duties of experts A number of cases, both civil and criminal, have laid down the duties of an expert. A helpful distillation of the case law relating to these duties is provided by the UK Forensic Science Regulator (2019). The duties of expert witnesses were laid out in the English civil case of the Ikarian Reefer (1993), a shipping case and repeated in the Court of Appeal in R v Harris and Others (2006) (appeals related to paediatric head injury) where the court stated: 1. Expert evidence presented to the court should be and seen to be the independent product of the expert uninfluenced as to form or content by the exigencies of litigation. 2. An expert witness should provide independent assistance to the court by way of objective unbiased opinion in relation to matters within his expertise. An expert witness in the High Court should never assume the role of advocate. 3. An expert witness should state the facts or assumptions on which his opinion is based. He should not omit to consider material facts which detract from his concluded opinions. 4. An expert should make it clear when a particular question or issue falls outside his expertise. 5. If an expert’s opinion is not properly researched because he considers that insufficient data is available, this must be stated with an indication that the opinion is no more than a provisional one. 6. If after exchange of reports, an expert witness changes his view on material matters, such change of view should be communicated to the other side without delay and when appropriate to the court. In the Court of Appeal case of R v Bowman (2006), the following points were added: 1. Details of the expert’s academic and professional qualifications, experience and accreditation relevant to the opinions expressed in the report and the range and extent of the expertise and any limitations upon the expertise. 2. A statement setting out the substance of all the instructions received (with written or oral), questions upon which an opinion is sought, the materials provided and considered and the documents, statements, evidence, information or assumptions which are material to the opinions expressed or upon which those opinions are based. 3. Information relating to who has carried out measurements, examinations, tests etc. and the methodology used, and whether or not such measurements etc. were carried out under the expert’s supervision.

4. Where there is a range of opinion in the matters dealt with in the report, a summary of the range of opinion and the reasons for the opinion given. In this connection, any material facts or matters which detract from the expert’s opinions and any points which should fairly be made against any opinions expressed should be set out. 5. Relevant extracts of literature or any other material which might assist the court. 6. A statement to the effect that the expert has complied with his/her duty to the court to provide independent assistance by way of objective unbiased opinion in relation to matters within his or her expertise and an acknowledgement that the expert will inform all parties and where appropriate the court in the event that his/her opinion changes on any material issues. 7. Where on an exchange of experts’ reports matters arise which require a further or supplemental report the above guidelines should, of course, be complied with. In Canada, a judgement of the Ontario Court of Appeal approved the guidelines for expert witnesses contained in the Ikarian Reefer in Moore v Getahun (2015) and these guidelines were also reiterated in the Privy Council case of Myers, Brangman and Cox (Appellants) v The Queen (Respondent) (Bermuda) (2015).

Pre-trial disclosure in criminal cases The expert giving evidence (or proposing to do so) for the prosecution is under an obligation, quite independently of the obligation on the police or the prosecution authorities, to bring to the attention of the lawyer instructing him or her, or the expert advising the other party, the records of all experiments and tests carried out by him or her, whether or not the results of those experiments or tests are of assistance to the prosecution (see R v Ward 1993). The obligation exists regardless of whether the expert witness is subsequently relied on by the prosecution to give expert evidence and is separate from his or her duty to assist the court. Experts for the prosecution should thus expect a request for all notes, emails and other materials to be disclosed to the defence. It is worth bearing in mind that there is no such thing as a private email in this context and a witness may come to be embarrassed by the content of an email written in the past and not necessarily written with litigation in mind. If you are not prepared to have an email displayed in the media, do not write it. In respect of disclosure and the duties of a pathologist, the words of Kay LJ in the case of R v Clark (2003) provide a salutary lesson to the pathologist, where at paragraph 168 he stated, … doctors reviewing the matter at a later stage are dependent upon the pathologist who conducted the original postmortem to draw to their attention not only any material which justifies the original pathologist’s conclusion but also any which reveals any abnormality that might need to be considered before being discounted. Where tests have been carried out and reported upon to the pathologist, his responsibility to make that material available for consideration by others is a clear one and his failure to do so may well mislead them into thinking that there has been negative findings when that is not the case.

For the defence in criminal cases expert witnesses are under no such disclosure obligation. In the event that defence pathologists do take issue with either the methodology or the conclusions of prosecution experts, it is incumbent on them to define the issues between the two with sufficient clarity to enable the

205

206

The role of the expert witness tribunal of fact to be able to resolve such issues as easily as possible. In this context, defence experts should always bear in mind that, if their evidence is to be relied on at trial, it may be subject to disclosure before trial. This is the case in England for criminal trials and civil matters, though in Canadian criminal case, the defence only has to disclose its evidence at the close of the prosecution case. In civil matters, there is a disclosure of both sides experts, where they are going to be relied upon. If an expert is appearing in a jurisdiction they are not familiar with, they should discuss with their instructing lawyers what rules apply about disclosure, including the disclosure of draft reports. In England and Wales, the Criminal Practice Directions require a party introducing evidence of expertise to give notice of anything of which that party is aware which might reasonably be thought to be capable of undermining the reliability of that evidence, or detracting from the credibility of the expert (Justice.gov.uk). This includes any fee arrangement under which the payment is in any way dependent on the outcome of the case, conflicts of interest, adverse peer or judicial comment, and any history of lax or inadequate scientific methods, and the declaration (required at the conclusion of every expert’s report) must include an acknowledgement that the expert has familiarised himself with the Practice Direction and the Rules.

Preparing for court Preparation for court may start with instructions from lawyers and writing a report or from the ordinary work of the expert being adopted by prosecuting authorities. Ordinarily, a report by a forensic pathologist or neuropathologist in the course of their work will be factually based with conclusions that are straightforward. In preparing any report that might be used in court, the potential expert must be cognisant of their duties. An expert should be careful about criticising another expert. This particularly applies where an expert is asked to write a report specifically for civil or criminal proceedings where other expert reports have been produced and the expert has been asked to comment on. The reports should be clearly produced detailing the material that the expert has seen. In R v Burridge (2010), a case involving paediatric head injury, the reports of an expert were described as ‘diffuse, poorly sourced where they are sourced, hard to interpret, infused by arrogance, and quite unnecessarily combative and dismissive of other experts, including those in fields which are not his own’. In R v Conaghan and Others (2017), the English Court of Appeal made clear that criticism of the work of another expert must be based on a proper consideration of the work of the expert. Where legitimate differences do occur, it was stated in the case of H-C (Finding of Fact: Rehearing) (2016) that the evidence must be presented in a professional manner. In using published research, it must not be partially quoted. In Squier v General Medical Council (2016), it was stated ‘… an expert must not cite the work of others as supporting her view when it does not. If it is capable of doing so, but only with significant qualification, she must say so’. ‘One of the overriding duties of an expert is not to mislead. Baldly stating, without qualification, that a research paper is a proper foundation for the proposition that the expert is seeking to advance is justified if that is the conclusion of the research paper; but if it is not, it should not be cited, without qualification, as supportive’.

It should be a matter of invariable routine that, in cases involving expert evidence, the expert should have had a conference in advance of the trial with the advocate who will actually be conducting the case. Frequently, in the course of such meetings (even where the evidence seems on its face to be relatively non-controversial), problems arise relating, for example, to terminology or a difference in approach by the pathologist and the lawyer, respectively, which would otherwise have to be resolved – somewhat unedifyingly – in the witness box. Best evidence requires careful thought and planning. Pathology material can be distressing for juries and photographs may not be the most appropriate method to demonstrate findings. Meetings conducted for the prosecution are disclosable to the defence. In preparing a report for a non-criminal case, it is advised to ask the instructing lawyer what communications and what drafts of reports may be disclosable, as this can vary. Before attending court, the expert should familiarise him- or herself with where the court is and when they are required to give evidence. In giving oral evidence, the expert should present a professional appearance and give clear evidence that can be heard by judge and jury. Technical terms should be explained and the expert should not stray beyond his or her field. Any temptation towards dogma, particularly if the state of scientific knowledge is limited or incomplete, is to be eschewed. Cross-examination techniques may appear annoying to the witness, but losing one’s temper, becoming sarcastic or arguing with counsel is ultimately detrimental to the evidence that needs to be given. The judge has the role to see that questioning is appropriate and fair, not the witness. A number of textbooks have been produced on how to conduct a cross-examination (see the example, Salhany et al. 2016). If a person is going to be a regular expert, their study can be valuable. There are limits to what an expert can give evidence, which may, on superficial examination, appear within their area of expertise. It is considered trite law that an expert should not stray from their area of expertise. Examples relating to pathology are whether a pathologist who is not a forensic pathologist may give opinion evidence on the cause of death. In Clarke and Anor v R (2013), an osteo-articular pathologist was not permitted to opine on such matters. Similarly, in R v Marquardt (2011), the Ontario Court of Appeal stated two paediatric neurologists had gone beyond their expertise in stating the cause of death in the case. Peer review and quality assurance are now standard procedures in many disciplines. A question arises as to how much information is permitted to be given about peer review in evidence. In the Canadian case of R v Ranger (2003), a crime scene expert stated that her opinion had been verified by 10 colleagues. This was ruled inadmissible on appeal. Evidence cannot be offered for the purpose of sharing an opinion about how believable testimony is. The process of stating that peer review has occurred but stating nothing further has been accepted.

Conclusion In conclusion, effective expert witnesses need to prepare well, maintain objectivity in the evidence, not favouring the side that instructed them and present their evidence in a professional manner. The stages of giving evidence have been summarised as dress up, turn up, stand up, speak up and shut up. Not a bad summary of the basic tenets for the giving of effective testimony!

Table of cases

References Department of Justice (UK). Rules and practice directions. https://www. justice.gov.uk/courts/procedure-rules/criminal/rulesmenu-2015. Federal Rules of Evidence. Updated 1st December 2019. https:// uscode.house.gov/view.xhtml?path=/prelim@title28/title28a/ node218&edition=prelim. Forensic Science Regulator. 2019. Legal obligations FSR-I-400 issue 7. https://assets.publishing.service.gov.uk/government/uploads/ system/uploads/at tachmnt t _data/file/795995/FSR _ Legal_ Obligations_-_Issue_7.pdf. Lord Hodge. 2017. Expert evidence: use, abuse and boundaries. https:// www.supremecourt.uk/docs/speech-171009.pdf. Munday, R. 2019. Cross and Tapper on Evidence, 13th edn. Butterworths. Salhany, R.E., Edelson, M.D. and Clifford, W.V. 2016. Cross-Examination. The Art of the Advocate, 4th edn. LexisNexis.

Table of cases Clarke and Anor v R [2013] EWCA Crim 162. Daubert v Merrell Dow Pharmaceuticals, Inc. [1993] 509 U.S. 579. Davie v Magistrates of Edinburgh [1953] SC 34. Frye v United States [1923] F. 1013 (D.C. Cir. 1923). H-C (Finding of Fact: Rehearing) [2016] EWFC 48.

Kennedy (Appellant) v Cordia (Services) LLP (Respondent) (Scotland) [2016] UKSC 6. Kumho Tire Co. v Carmichael [1999] 119 S.Ct. 1167. Moore v Getahun [2015] ONCA 55. Myers, Brangman and Cox (Appellants) v The Queen (Respondent) (Bermuda) [2015] UKPC 40; 3 WLR 1145. R v Abadom [1983] 1 WLR 126. R v Bowman [2006] EWCA Crim 417. R v Burridge [2010] EWCA Crim 2847. R v Clark [2003] EWCA Crim 1020. R v Conaghan and Others [2017] EWCA Crim 597. R v Harris and Others [2006] 1 Cr App. R.5. R v J L-J [2000] 2 S.C.R. 600. R v Marquardt [2011] ONCA 281. R v Mohan [1994] 2 S.C.R. 9. R v Ranger [2003] CanLII 32900 (ON CA). R v Reed, Reed and Garmson [2009] EWCA Crim 2698. R v Turner [1975] QB 834. R v Ward [1993] WLR 619. Re AB (A Minor) [1995] 1FLR 181. Squier v General Medical Council [2016] EWHC 2739 (Admin). White Burgess Langille Inman v Abbott and Halibuton Co. [2015] SCC 23.

207

Index Note: Locators in italics represent figures and bold indicate tables in the text.

A Abdominal injuries, 180 Abrading-type injuries, 66, 68 Abrasions, 66, 67, 152 Abusive head trauma (AHT), 151 Accidental injury, 151 Accident and emergency (A&E) department, 18 Acquired immune deficiency syndrome (AIDS), 189 Acute disseminated encephalomyelitis (ADEM), 190 Acute haemorrhagic leukoencephalitis, 191 Acute subdural haematoma (ASDH), 18, 48, 49, 51, 52 Air emboli, 133, 192–193 Air-powered weapons, 77 Alcohol (ethanol), 195 acute intoxicant effects of, 195 and atrophy, 196 intoxication deaths, 195–196 Alcohol, metabolic disorders related to alcohol and trauma to the brain, 198 central pontine myelinolysis, 197 ethylene glycol, 199 fetal alcohol spectrum disorder (FASD), 198–199 hepatic encephalopathy, 197 Marchiafava-Bignami disease, 197–198, 198 methanol, 199 recreational drugs amphetamines, 200 cocaine, 199–200 gases, 200–201 heavy metals, 201 novel psychoactive substances (NPS), 200 opioids, 199 Alcohol dehydrogenase (ADH), 195 Alzheimer’s disease (AD), 186, 192 Amphetamines, 200 Amyloid precursor protein (APP), 155, 158 Angular acceleration, 50 Angular kinematics, 50, 52 cavitation, effect of, 50–51 cerebral contusions, 50 concussion, 50 diffuse traumatic axonal injury (dTAI), 51

208

inertial loading, infant vulnerability to, 52 inertial loading, shaking and whiplash, 51–52 intracerebral haematomas (ICH), 51 shaking, characterisation of, 52 subdural haematoma, 51 Ankylosing spondylitis, 175, 175 Anterior compression fracture, 171 Anticoagulant, 193 Anti-platelet, 193 AO spine subaxial cervical spinal injury classification system, 169 AO spine thoracolumbar spinal injury classification system, 169 Arachnoid layer, 4 Arsenic, 201 Artefacts, 59 Arterial blood supply, 5–6, 6 Arterial dissection, 100–101 Arterial trauma, histology of, 98 Arteriovenous malformations, 114 Atlanto-axial rotatory subluxation, 170 Atlanto-occipital dislocation, 169 Autonomic dysreflexia, 174–175 Axonal damage; see also Traumatic axonal injury immunohistochemistry for, 121 localised to craniocervical junction, 125 pitfalls in the interpretation of, 124–125 Axonal injury, 157 Axonal swellings, 173, 173

B ‘Back to sleep’ campaign, 183 Bacterial infections, CNS, 187 Ball-pein hammer, crescent-shaped lacerations caused by blows with, 70, 70 Basilar artery, 6 Berry aneurysm, 100, 100 β-APP immunostainin, 98, 117, 121, 122 Biomechanical loading environment, characterisation of, 45 Birth-related head injury, 165 Bite mark, 66, 67 Blast injury, 147–148 Blocks, 62 in SUDI cases, 62

Blood arterial blood supply, 5–6, 6 face, blood supply to, 3 hyperdense, 32, 35 intraventricular, 36, 36 red blood cells, 79 spinal cord, blood supply of, 15 subarachnoid, 36, 36, 37, 38 Blood-brain barrier, 4, 128 Blood products, 27–28 Bodily injury, defined, 66 Brain adult brain weight, 61, 62 alcohol and trauma to, 198 appearance of, 77 blocking of, 58 blood-brain barrier, 4 congestive brain swelling, 128 contusional brain injury, see Contusional brain injury CT imaging of, 25 damage after cardiac arrest, 134, 134 after hypotension, 134, 134 in survivors, 158–159 distortion of, 59, 59 forebrain, 11 herniation, through craniotomy/burr holes, 58 hypoxic brain injury, 42 hypoxic-ischaemic brain injury, see Hypoxic-ischaemic brain injury infant and child brain weights, 61 malignant brain swelling, 132 medullary brain lesions, sudden unexpected death related to, 142–143 midbrain, 11 and nervous tissues, 4 non-perfused brain (respirator brain), 137 removal, in newborns and young infants, 57 respirator brain, 137 retention of, 57–58 traumatic brain injury (TBI), 46, 94, 145 venous drainage of, 6 Brain contusion injuries, 38–40, 40 Brain damage after cardiac arrest, 134, 134 after hypotension, 134, 134 Brain fixation, rapid microwave fixation method for, 62

Index Brain injury related to hypoxia and impaired perfusion delayed posthypoxic leukoencephalopathy (DPHL), 137 global ischaemia, 134 cardiac arrest, brain damage after, 134, 134 hypotension, brain damage after, 134, 134 hypoxic-ischaemic brain injury, 132–133 focal cerebral ischaemia and infarction, 133–134 non-perfused brain (respirator brain), 137 red neurons versus dark neurons, 135–137 timing of ischaemic changes, 134–135, 135 Brain parenchyma, infections of, 188 cerebral abscesses, 188–189, 188 encephalitis, 189 COVID-19 infection, 189–190 herpes simplex encephalitis, 189, 189 HIV infection, 189 malaria, 189 Brainstem, 11 Brainstem death, 22–23, 23 Brainstem haemorrhages, 112 Brain swelling, 128 cerebral oedema, 128 congestive, 128 with gyral flattening, 129 malignant, 132 Brain swelling and brain oedema, 128 brain weight, 129 herniations central transtentorial, 130 external cerebral, 130–131 paradoxical, 131 subfalcine, 129, 129 tonsillar, 130 transtentorial uncal, 129–130 upward tentorial, 130 macroscopy, 129 microscopy, 131, 131 rate of swelling, 132 Bruises, 66–69 Bruising, 85, 85, 151–152 Buccinator, 1 Buckshot skull showing external bevelling caused by, 75 skull showing internal bevelling caused by, 74 Bumper injuries to lower limbs, 179, 179 Burns, 71–72, 72 Burr holes, 58, 59 Burst fractures, 171, 172, 172 Burst lobe haematoma, 107, 109

C CACNA-1A gene, 132 Caisson’s disease, 133 Capillary telangiectasias, 114 Carbon disulphide, 201 Carbon monoxide, 200–201 Cardiac arrest, brain damage after, 134, 134 Carotid artery, 5 Carotid artery thrombosis, 101, 102 Carotid dissections, 101 Carotid siphon, 5 Cauda equina, 15 Cavernous angiomas, 114 Cavitation, effect of, 50–51 CD4 T lymphocytes, 189 CD68, 123 Central nervous system (CNS), infection of, 186 bacterial infections, 187 fungal infections, 187 neoplasia involving, 190 protozoal infections, 187 viral infections, 187 Central pontine myelinolysis, 191, 197 Central sulcus, 11 Central transtentorial herniation, 130 Cerebellar coning, 130 Cerebellar degeneration macroscopic, 196 microscopic, 197 Cerebellar haemorrhages, 111 Cerebellum, 14, 62 Cerebral abscesses, 188–189, 188 Cerebral amyloid angiopathy (CAA), 186 intracerebral haemorrhage (ICH) in association with, 113 histology, 114 macroscopic findings, 114, 114 problem areas, 114 Cerebral aqueduct, 11 Cerebral atrophy, 146 Cerebral concussion, 50 Cerebral contusions, 50 Cerebral damage caused by penetration with screwdriver, 71 Cerebral hemisphere, 8, 8–10 Cerebral oedema, 128 Cerebral swelling, 155; see also Brain swelling Cerebral venous thrombosis and infarction, 104 Cerebrospinal fluid (CSF), 4, 7, 11, 15, 175, 186 Cerebrovascular disease in cocaine users, 199 COVID-19 infection and, 189–190 Cerebrum, 6–11 Chance fractures, 172 Chest injuries, 71, 180 Children aneurysms in, 102

brain weights, 61 diffuse swelling, 132 extradural haematoma (EDH) in, 88 head injury in, see Head injury in children skull fracture, 87 sudden death in epilepsy (SUDEP) in, 139 sudden unexplained death in childhood (SUDC), 142 Chronic bilateral subdural haematoma, 90 Chronic hypertension, 186 Chronic subdural haematoma (CSDH), 90, 91, 157–158 histology of SDH, 91–92 macroscopic appearances, 90–91 problem areas in, 92 subdural hygromas, 92 Chronic traumatic encephalopathy (CTE), 145 neuropathology of, 146–147 and other neurodegenerative disorders, 147 Circle of Willis, 5, 6, 6, 99 Circulating cerebrospinal fluid (CSF), 4 CSF drainage, 20 CSF fistulas, 21 Clarke and Anor v R (2013), 206 Clay-shoveler’s fracture, see Spinous process avulsion fracture Clinicopathological entities infectious cases, 60 sudden death in epilepsy (SUDEP), 60 sudden unexpected death in infancy (SUDI), 59–60 suspected stroke, examination in cases of, 60 Clusterin (apoplipoprotein J), 123 Cocaine, 199–200 COL3A1 gene, 94 Commissural fibres, 8 Compound elevated skull fractures, 84 Compression forces, 48 Computerised tomography (CT) acute extradural haematoma on, 34 acute subarachnoid blood, 36 angiography, 19 diffuse traumatic axonal injury (dTAI) on, 41 global hypoxic-ischaemic brain injury, 42 head scan, 19 imaging of brain, 25 indications for admission and, 19 intraventricular haemorrhage on, 38 perfusion studies, 19 subdural haematoma (SDH), 30 Concussion, 50 Congestive brain swelling, 128 Contact gunshot wounds, 73, 73 Contact head loading, 45

209

Index

210

Contact sport, 145, 183 blast injury, 147–148 boxers, acute injury in, 145 chronic traumatic encephalopathy (CTE), 145 neuropathology of, 146–147 and other neurodegenerative disorders, 147 clinical features, 146 diving, 183–184 golf and horse riding, 145 second impact syndrome (SIS), 145, 148–149 Contact wounds, 75 Contre-coup fractures, 84, 84 Contre-coup orbital plate fractures, 81 Contusional brain injury, 108 macroscopic appearances, 107 microscopic appearances, 107 burst lobe, 107, 109 contre-coup contusions, 109–110 coup contusions, 107–109 delayed intra-cerebral haemorrhage, 107 fracture contusions, 110 Contusional tears, 156, 156 Contusions, 49, 50, 66 ‘Corkscrew’ neurons, Neuronal shrinkage with, 59, 60 Corpus callosum, 5, 8 ‘Country’ guns, 77 Court, preparing for, 206 COVID-19 infection, 189–190 Cranial nerve injuries, 21 Cranial vault, 4, 4 Craniocervical junction, axonal damage localised to, 125 Craniotomy/burr holes, herniation of brain through, 58 Crescent-shaped lacerations caused by blows with a ball-pein hammer, 70, 70 Creutzfeldt-Jakob disease, 55, 191 Criminal cases, pretrial disclosure in, 205–206 Crossbow injuries, 78 Crush loading, 45–47 Crush-related traumatic brain injury (TBI), 46

D Dark neurons, 137 red neurons versus, 135–137 Daubert v Merrell Dow Pharmaceuticals, Inc. (1993), 203 Davie v Magistrates of Edinburgh (1953), 204 Death from alcohol intoxication, 195–196 brainstem death, 22–23, 23 hippocampal maldevelopment associated with sudden death (HMASD), 142

late deaths, 174 medullary brain lesions, sudden unexpected death related to, 142–143 as a result of acute injury in boxers, 145 road traffic deaths, see Road traffic deaths sudden death in head injury, 182–183 sudden infant death syndrome (SIDS) and suffocation, 183 sudden unexpected death in epilepsy, see Sudden unexpected death in epilepsy (SUDEP) sudden unexpected death in infancy (SUDI), 59–60, 64 sudden unexplained death in childhood (SUDC), 142, 142 sudden unexplained death in childhood with febrile seizure phenotype (SUDC-FS), 142 suspicious death, post-mortem examination in a case of, 151 Decompressive craniectomy, 20 Defence, pathologists acting on behalf of, 58 Defence injuries, 78 Delayed posthypoxic leukoencephalopathy (DPHL), 137 Dementia assessment of, 62–64 disorders associated with, 191 Alzheimer’s disease, 192 dementia with Lewy bodies (DLB), 192 Parkinson’s disease, 192 vascular dementia, 192 Dementia with Lewy bodies (DLB), 192 Demyelinating disorders acute haemorrhagic leukoencephalitis, 191 central pontine myelinolysis, 191 multiple sclerosis (MS), 191 Depressed skull fractures, 83, 83 Diabetes insipidus, 21 Diatom analysis, procedure for, 64–65 Diatoms, 60, 61 Diencephalon, 11–14 Difficult areas in forensic neuropathology, 178 contact sports, 183 diving, 183–184 homicide, suicide or accident penetrating injuries, 178 road traffic deaths, 178 falls, 181 falls down stairs, 182 falls from a height, 181–182 falls from standing, 181 kicking versus falling, 182 motorcycles and pedal cycles, 181 pedestrians, 179–180

run-over injuries, 180 vehicle occupants, 180–181 vehicular collision investigation, 181 sudden death in head injury, 182–183 sudden infant death syndrome (SIDS) and suffocation, 183 Diffuse traumatic axonal injury (dTAI), 40–42, 41, 48, 51, 89, 111, 112, 116, 117, 118, 119, 124, 157, 160, 180 Diffuse traumatic vascular injury, 117, 118 Diffusion-weighted imaging (DWI), 25 Dislocation atlanto-occipital dislocation, 169 facet dislocations, 171 fracture-dislocations in thoracic spine, 172 Distant wounds, 74, 76 Distraction disc injury, 171 Diving, 183–184 Dural venous sinuses, 4 Dura mater, 4 Duret haemorrhage, 42, 119 Dynamic loading, 46

E Ecchymosis, 66 Ehlers-Danlos syndrome, 94 Electrical injury, 72 Encephalitis, 189 COVID-19 infection, 189–190 herpes simplex encephalitis, 189, 189 HIV infection, 189 malaria, 189 Endocrine disturbances, 21 Entrance wound, 75 Epidural haematoma, 49–50; see also Extradural haematoma Epithalamus, 11 Ethanol, 195 Ethmoidal sinuses, 2 Ethylene glycol, 199 Excited delirium (ED), 200 Exit wound, 75 slit-like, 75 Expert witness, role of, 203 criminal cases, pretrial disclosure in, 205–206 opinion evidence, admissibility of, 203–204 preparing for court, 206 role of the expert, 204 duties of experts, 205 External cerebral herniation, 130–131 External examination, 55, 66 External injury, 151 Extracellular methaemoglobin, 27 Extradural haematoma (EDH), 18, 34, 35, 55, 56, 87, 88 circumstances and clinical features, 87 external injury, 87 histology, 87–88

Index pathological appearances, 87 problem areas missed EDH, 88 timing, 88 rarities heat haematoma, 88–89, 88 slow evolution/resolving EDH, 88 spontaneous extradural haematoma, 88 Extradural haemorrhage, 32–35 Extradural space, 4 Extraocular and intraorbital haemorrhages, 164 Eyes, 1

F Face blood supply to, 3 external, 1 Face injury, 66 Facet dislocations, 171 Facial expression muscles, 1 Facial fractures, 81, 82 internal, 1 Falling, kicking versus, 182 Falls, 181 acceleration to a fall, 81 from a height, 181–182 low-level falls, 160 stairway falls, 160 from standing, 181 Falls down stairs, 182 Falx cerebelli, 5 Falx cerebri, 5, 7 Fat embolism, 133, 134 air emboli, 133 Fetal alcohol spectrum disorder (FASD), 198–199 Fibrocartilaginous emboli, 134 Fingertip bruising, 152 Firearm injuries, 72 Fixation, examination after, 58 Fixation artefacts, 59 FLAIR (FLuid Attenuation Inversion Recovery) imaging, 25, 36, 40 Focal cerebral ischaemia and infarction, 133 fat embolism, 133, 134 air emboli, 133 fibrocartilaginous emboli, 134 foreign body emboli, 134 Foramen of Luschka, 11 Forebrain, 11 Foreign body emboli, 134 Fracture anterior compression fracture, 171 burst, 171, 172, 172 chance, 172 compound elevated skull fractures, 84 contre-coup, 81, 84, 84 defined, 80

depressed skull fractures, 83, 83 facial, 81, 82 frontal, 82, 82 front-to-back hinge fracture, 82–83, 82, 84 haemosiderin staining over healing fracture, 85 Hangman’s fracture, 171, 171 hinge fractures, 84, 84 Jefferson fractures, 170, 170 linear frontal fractures, 81–82, 82 linear skull fractures, 80–83 linear vault fractures, 20 minor thoracolumbar fractures, 172 mosaic, 83 occipital, 81, 81 occipital condyle, 169 odontoid, 170–171, 170 pathological, 175 pond, 83 ring, 83–84 sacral, 172–173, 173 skull, see Skull fractures spinous process avulsion fracture, 171 squamous temporal bone fractures, 81, 81 stable fractures, mobilisation of, 174 teardrop, 171, 171 thoracic spine, fracture-dislocations in, 172 transmitted, 84, 84 wedge compression fractures, 172 Fracture lines, 75 Frenulum, injury to, 152, 152 Frontal fractures, 82, 82 Front-to-back hinge fracture, 82–83, 82, 84 Frye v United States (1923), 203 Fungal infections, CNS, 187

G Ganglionic haematomas, 111–112, 112 Genu, 14 Giant aneurysms, 99–100, 100 Giant capillaries with secondary bleeding, 91, 91 Glasgow Coma Scale (GCS) score, 19, 19, 112 Glasgow Outcome Scale, 22, 22 Glioblastoma, 190, 190 Glioma, 190 Global ischaemia, 134 cardiac arrest, brain damage after, 134, 134 hypotension, brain damage after, 134, 134 Granulomatous meningitis, 188 Gunshot wound, 178, 179 Gutter wounds, 74

H Haematoma, 27–28, 68 intracranial, see Intracranial haematomas within a skull fracture, 84, 85 Haemophilus influenzae, 187 Haemorrhage, 56 brainstem, 112 contusions with, 39 ganglionic haematomas, 111–112, 112 petechial white matter, 111 subarachnoid, 155 subdural, 154–155 tissue tear, 111, 112 traumatic intraventricular, 112 Haemosiderin, 156 Haemosiderin macrophages, 154, 154 Haemosiderin staining over healing fracture, 85 Hammer blows, 109 complex laceration caused by, 70, 70 coup contusion beneath area of, 109 Hangman’s fracture, 171, 171 Head crush loading of, 46 falls on to the back of, 66 Head and neck, anatomy of, 1 arterial blood supply, 5–6, 6 blood supply to the face, 3 brain and nervous tissues, 4 brainstem, 11 cerebellum, 14 cerebrum, 6–11 diencephalon, 11–14 face, external, 1 facial structures, internal, 1 meninges, 4–5 midbrain, 11, 12 nose, 1–2 oral cavity, 2–3 posterior structures of neck, 3 sagittal section through the head, 7–8, 8 scalp, 3–4 skull bones, 4, 4 spinal cord, 14, 16 internal structure of, 15–16 venous drainage, 6 vertebral column, 14, 15 Head injury, 17 acceleration measures of vulnerability and risk in, 47–48 acute surgical procedures, 20 decompressive craniectomy, 20 interventional radiological procedures, 20 skull fractures, 20 brainstem death, 22–23, 23 classification, 17 complications of cranial nerve injuries, 21 post-traumatic epilepsy, 20–21

211

212

Index diagnosis and management, 18 Glasgow Coma Scale (GCS), 19, 19 grading of, by severity, 17 impact-induced, 47, 48 acute subdural haematoma (ASDH), 49 contusions, 49 epidural haematoma, 49–50 skull fracture tolerance, 49 subdural haematoma (SDH), 49 translational kinematics, 48–49 incidence, 17 investigations, 19 long-term sequels, 22 persisting vegetative state, 22 non-surgical treatment critical care, 20 observation, 19–20 prevention, 23 primary injuries, 17 secondary damage, 17–18 sudden death in, 182–183 Head Injury Criterion (HIC), 48 Head injury in children, 151 accidental injury, 151 external findings abrasions, 152 bruising, 151–152 external injury, 151 frenulum, injury to, 152, 152 traumatic alopecia, 152 internal examination, 152–153 non-accidental injury, 151, 154 axonal injury, 157 brain damage in survivors, 158–159 cerebral swelling, 155 chronic subdural haematomas and subdural hygromas, 157–158 hypoxic-ischaemic injury, 156–157, 157, 159 impact injury, 160 intracerebral haemorrhage, 155–156 low-level falls, 160 shaken baby syndrome, 160 spinal injury, 159 stairway falls, 160 subarachnoid haemorrhage, 155 subdural haematomas, re-bleeding in, 160 subdural haemorrhages (SDHs), 154–155 timing of injury, 160 ocular pathology and head injury, 163 birth-related head injury, 165 extraocular and intraorbital haemorrhages, 164 optic nerve sheath haemorrhages (ONSH), 164 peripapillary scleral haemorrhage, 164 raised intracranial pressure, 165 resuscitation, 165

retinal haemorrhages, 163, 164, 165 sudden unexpected death in infancy (SUDI), ocular findings in, 166 Terson syndrome, 165 traumatic retinoschisis and perimacular folds, 164 post-mortem examination in a case of suspicious death, 151 skull fractures, 153–154, 153 Head trauma, imaging of, 25 blood products, 27–28 brain contusion injuries, 38–40, 40 CT imaging of brain, 25 diffuse traumatic axonal injury (dTAI), 40–42, 41 extradural haemorrhage, 32–35 MRI, 25–27, 27 primary versus secondary, 28–29 secondary effects herniation syndromes, 42 hypoxic-ischaemic brain injury, 42–43 paediatric non-accidental head injury, 43–44 skull fractures, 28 subarachnoid haemorrhage and intraventricular haemorrhage, 35–38, 36 subdural haemorrhage and effusions, 29–32 types of, 28–29 Hearsay rule, 203 Heat haematoma, 88–89, 88 Heat shock proteins (Hsp70), 200 Heavy metals arsenic, 201 mercury, 201 organic arsenic, 201 thallium, 201 Hepatic encephalopathy, 197 Herniation contusions, 111 Herniations central transtentorial, 130 external cerebral, 130–131 paradoxical, 131 subfalcine, 129 tonsillar, 130 transtentorial uncal, 129–130 upward tentorial, 130 Herniation syndromes, 42 Herpes simplex encephalitis, 189, 189 Herpes simplex virus (HSV) type 1, 189 High-energy transfer wounds, 77 Hinge fractures, 84, 84 Hippocampal maldevelopment associated with sudden death (HMASD), 142 Hippocampal sclerosis, 141, 141 Hippocampus, malrotation of, 142 Histological artefacts, 59 Histological stains, 58

HIV encephalopathy (HIVE), 189 HIV infection, 189 Homicide, suicide or accident penetrating injuries, 178 Hospital therapy, findings relating to, 58 Humane killers, 78 Hurst’s disease, 191 Hydrocephalus, 21, 21 Hydrostatic oedema, 128 Hyperaemia, 132 Hypertension, chronic, 186 Hypertensive intracerebral haemorrhage, 113 histology, 113 macroscopic findings, 113, 113 problem areas, 113 Hypoglycaemia, 186, 187 Hypotension, brain damage after, 134, 134 Hypothalamus, 11 Hypoxia, 133 classification of, 133 Hypoxic brain injury, 42 Hypoxic-ischaemic brain injury, 42–43, 132–133, 156–157, 157, 159 focal cerebral ischaemia and infarction, 133 fat embolism, 133, 134 fibrocartilaginous emboli, 134 foreign body emboli, 134

I Iatrogenic neuropathology, 192 invasive intervention air emboli, 192–193 neoplasm surgery, 192 radiological coiling/embolisation, 193 spinal cord injury, 192 medical intervention anticoagulant, anti-platelet and thrombolytic drugs, 193 previous irradiation, complications of, 193 Iba-1 immunostaining, 123 Ikarian Reefer in Moore v Getahun (2015), 205 Immediate neuropathological examination, 56–57 Impact-induced acceleration, 47 Impact-induced head injuries, 48 acute subdural haematoma (ASDH), 49 contusions, 49 epidural haematoma, 49–50 skull fracture tolerance, 49 subdural haematoma (SDH), 49 translational kinematics, 48–49 Impact injury, 160 Impact loading, 45 Imprint abrasions, 66 Incised wounds and stab wounds, 71

Index Inertial loading, 51–52 infant vulnerability to, 52 Infants brain removal in, 57 brain weights, 61 human-induced shaking of, 52 skull of, 49 subdural haemorrhages (SDHs) in, 154 sudden infant death syndrome’ (SIDS), 183 vulnerability to inertial loading, 52 Infarction, cerebral venous thrombosis and, 104 Infections, 60, 85 Injury abdominal, 180 abrading-type, 66, 68 abrasions, 66, 67 accidental, 151 axonal, 157 birth-related head injury, 165 blast, 147–148 bodily, 66 brain contusion, 38–40, 40 brain injury, see Brain injury related to hypoxia and impaired perfusion bruises, 66–69 bumper injuries to lower limbs, 179, 179 burns, 71–72, 72 caused by sharp-edged objects, 71 chest, 71, 180 chop injuries, 71 contusional brain injury, see Contusional brain injury cranial nerve injuries, 21 crossbow injuries, 78 crush-related traumatic brain injury, 46 defence injuries, 78 diffuse traumatic axonal injury (dTAI), 40–42, 41, 48, 51, 89, 111, 112, 116, 117, 118, 119, 124, 157, 160, 180 diffuse traumatic vascular injury, 117, 118 distraction disc injury, 171 electrical injury, 72 external injury, 151 face injury, 66 firearm injuries, 72 to frenulum, 152, 152 head injury, see Head injury hypoxic brain injury, 42 hypoxic-ischaemic brain injury, see Hypoxic-ischaemic brain injury impact-induced head injuries, see Impact-induced head injuries impact injury, 160 incised wounds and stab wounds, 71 lacerating injury, 110–111 lacerations, 67, 69–71, 70

non-accidental injury, see Non-accidental injury paediatric non-accidental head injury, 43–44 penetrating injuries, 174, 178 primary traumatic head injury, see Primary traumatic head injury, biomechanics of rifled weapon injuries, see Rifled weapon injuries run-over injuries, 180 scalp injuries caused by machete, 71, 72 seat-belt injuries, 180, 180 shotgun injuries, see Shotgun injuries spinal injuries, see Spinal injuries subaxial cervical injuries, 171, 172 suboccipital injuries, 169–171 sudden death in head injury, 182–183 surface injuries, 66 thoracolumbar injuries, 172 timing of, 160 translational kinematics, 48–49 traumatic axonal, see Traumatic axonal injury (TAI) traumatic brain injury (TBI), 46, 94, 145 traumatic vascular injuries, see Traumatic vascular injuries unstable injuries, 174 vertebral injuries, see Vertebral injuries whiplash injuries of neck, 176 Intermediate wound, 73–74, 75–76 Internal bevelling, 74 Internal examination, 152–153 skull, removal of, 55–56 subscalp bruising/haematomas, 55 Interstitial cell adhesion molecule (ICAM1), 79 Interstitial oedema, 128 Intracerebral haematoma (ICH), 27, 51 causes of arteriovenous malformations, 114 capillary telangiectasias, 114 cavernous angiomas, 114 vascular malformations, 114 venous angiomas, 114 Intracerebral haemorrhage (ICH), 107 in association with cerebral amyloid angiopathy (CAA), 113 histology, 114 macroscopic findings, 114, 114 problem areas, 114 in association with contusional injury of adult type, 155 contusional tears, 156, 156 petechial haemorrhages, 156 Intracranial haematomas, 87 acute subdural haematoma, 89 circumstances and clinical features, 89 histology, 90 macroscopic pathology, 89–90

mechanism of injury, 90 previous injury/disease, 90 chronic subdural haematoma (CSDH), 90, 91 histology of SDH, 91–92 macroscopic appearances, 90–91 problem areas in CSDH, 92 subdural hygromas, 92 extradural haematoma (EDH), 87, 88 circumstances and clinical features, 87 external injury, 87 histology, 87–88 pathological appearances, 87 problem areas, 88 rarities, 88–89 Intracranial hypotension, 33 Intracranial pressure (ICP) monitoring, 19–20 Intracranial surgery, 20 Intravenous drug abuser (IVDA), 189 Intraventricular haemorrhage, 35–38, 36, 38 Ipsilateral cerebral peduncular compression, 130 Ischaemia, 133, 187 Ischaemic changes, timing of, 134–135, 135

J Jefferson fractures, 170, 170

K Kennedy (Appellant) v Cordia (Services) LLP (Respondent) (Scotland) (2016), 204 Kernohan’s notch, 130 Kicking, 55 haemorrhage into temporalis muscle caused by, 56 laceration under the eyebrow caused by, 70, 70 versus falling, 182 Kiesselbach’s area, 2 Kinetic energy, 47 Kumho Tire Co. v. Carmichael (1999), 203

L Lacerating injury, 110 cerebellar haemorrhages, 111 herniation contusions, 111 Lacerations, 67, 69–71, 69, 70 caused by hammer blows, 70, 70 crescent-shaped, 70 under the eyebrow caused by kicking, 70, 70 Lands, 72 Late deaths, 174 Lateral ventricles, 7, 7 Leptomeningitis, 187

213

Index

214

Linear frontal fractures, 81–82, 82 Linear kinematics, 52 Linear skull fractures, 80 front-to-back hinge fracture, 82–83, 82 linear frontal fractures, 81–82, 82 linear squamous temporal bone fractures, 81, 81 occipital fracture, 81, 81 Linear squamous temporal bone fractures, 81, 81 Linear vault fractures, 20 Locked-in syndrome, 22 ‘Love’ bites, 69 Low-energy transfer wounds, 77 Lower limbs, bumper injuries to, 179, 179 Low-level falls, 160 ‘Lucid’ interval, 18 Lymphoid tissue, 3

M Machete, injuries to scalp caused by, 71, 72 Magnetic resonance imaging (MRI), 19, 25–27, 27, 36, 40, 141 Malaria, 189 Malignant brain swelling, 132 Marchiafava-Bignami disease, 197–198, 198 Marchi technique, 124 Maxillary sinuses, 2 Mechanical loading environment, 45 Medico-legal issues bruising, 85, 85 infections, 85 post-traumatic epilepsy, 85 Medico-legal work, non-traumatic neurological conditions in, 186 brain parenchyma, infections of, 188 cerebral abscesses, 188–189, 188 encephalitis, 189–190 central nervous system (CNS), 186 bacterial infections, 187 fungal infections, 187 neoplasia involving, 190 protozoal infections, 187 viral infections, 187 dementia, disorders associated with, 191 Alzheimer’s disease, 192 dementia with Lewy bodies (DLB), 192 Parkinson’s disease, 192 vascular dementia, 192 demyelinating disorders acute haemorrhagic leukoencephalitis (Hurst’s disease), 191 central pontine myelinolysis, 191 multiple sclerosis (MS), 191 granulomatous meningitis, 188 hypoglycaemia, 186 iatrogenic neuropathology, 192

invasive intervention, 192–193 medical intervention, 193 leptomeningitis, 187 non-traumatic haemorrhage, 186 cerebral amyloid angiopathy (CAA), 186 chronic hypertension, 186 pachymeningitis, 187 Medullary brain lesions, sudden unexpected death related to, 142–143 Meninges, 4–5 Meningitis, 187 Mercury intoxication, 201 Methanol, 199 3,4-Methylenedioxymethamphetamine (MDMA), 200 Microglial activation, 123 Microglial clusters, 123, 123, 124 Microsomal ethanol oxidizing system (MEOS), 195 Microwave fixation, 58 Midbrain, 11 Middle cerebral artery, 5 Middle cerebral artery berry aneurysm, 99 Minor thoracolumbar fractures, 172 Monro-Kellie principles, 131 Mosaic fracture, 83 Motorcycles and pedal cycles, 181 Mouth, 1 direct blow to, 152, 152 examination of the inside of, 75 punches to, 70 MPTP (1-methyl-4-phenyl-1,2,3,6tetrahydropyridine), 199 Multiple sclerosis (MS), 191 Multivessel trauma, 103 Mushroom plume, 140, 140 Myers, Brangman and Cox (Appellants) v The Queen (Respondent) (Bermuda) (2015), 204, 205

N Nasal cavities, 1–2, 2 Natural intracerebral haemorrhage, 112–113 Near contact and intermediate wounds, 73–74, 75–76 Neck, 3 external pressure to, 104 posterior structures of, 3 shotgun wound to, 76 whiplash injuries of, 176 Neisseria meningitidis, 187 Neoplasm surgery, 192 Neurofibrillary tangles, 146 Neuronal shrinkage with ‘corkscrew’ neurons, 59, 60 Neuropathological examination, 55 Neurotoxins, 200 Newborns, brain removal in, 57

Non-accidental injury, 151, 154 axonal injury, 157 brain damage in survivors, 158–159 cerebral swelling, 155 chronic subdural haematomas and subdural hygromas, 157–158 hypoxic-ischaemic injury, 156–157, 157, 159 impact injury, 160 intracerebral haemorrhage in association with contusional injury of adult type, 155 contusional tears, 156, 156 petechial haemorrhages, 156 low-level falls, 160 shaken baby syndrome, 160 spinal injury, 159 stairway falls, 160 subarachnoid haemorrhage, 155 subdural haematomas, re-bleeding in, 160 subdural haemorrhages (SDHs), 154–155 timing of injury, 160 Non-aneurysmal subarachnoid haemorrhage, 99 Non-contact head loading, 45 Non-perfused brain (respirator brain), 137 Non-traumatic haemorrhage, 186 cerebral amyloid angiopathy (CAA), 186 chronic hypertension, 186 Nose, 1–2 Novel psychoactive substances (NPS), 200 excited delirium, 200 neurotoxins, 200

O Occipital condyle fractures, 169 Occipital fracture, 81, 81 Occipital lobe, 11 Occipito-parietal lacerations from a fall down stairs, 182, 182 Ocular pathology and head injury, 163 birth-related head injury, 165 extraocular and intraorbital haemorrhages, 164 optic nerve sheath haemorrhages (ONSH), 164 peripapillary scleral haemorrhage, 164 raised intracranial pressure, 165 resuscitation, 165 retinal haemorrhages, 163, 164, 165 sudden unexpected death in infancy (SUDI), ocular findings in, 166 Terson syndrome, 165 traumatic retinoschisis and perimacular folds, 164 Odontoid fractures, 170–171, 170 Olive, medulla at the level of, 13

Index Ophthalmic artery, 5 Opinion evidence, admissibility of, 203–204 Opioids, 199 Optic nerve sheath haemorrhages (ONSH), 164 Oral cavity, 2–3 Orbicularis oculi, 1 Organic arsenic, 201 Osteoporosis, 176, 176 Oxyhaemoglobin, 27

P Pachymeningitis, 187 Paediatric non-accidental head injury, 43–44 Palpebral fibres, 1 Paradoxical herniation, 131 Parafalcine, 29 Parahippocampal herniation, 129–130 Paranasal sinuses, 2 Paraplegic patients, early needs of, 175 Parkinson’s disease, 192 Patchy subarachnoid haemorrhage, 94, 95 Pathological fractures, 175 Patterned bruise, 69 Peak g method, 48 Pedal cycles, 181 Pedestrians, 179–180 Penetrating injuries, 178 of spinal cord, 174 Peripapillary scleral haemorrhage, 164 Peripheral traumatic aneurysms, 102 Perivascular haemorrhages, 107, 108, 131, 132 Petechia, 66 Petechial haemorrhages, 66–67, 156, 189, 190, 191 Petechial perivascular haemorrhages, 131, 132 Petechial-type haemorrhages, 67, 68 Petechial white matter haemorrhages, 111 Pia mater layer, 4 Pick’s disease, 191 Pistol, self-loading, 72 Plasmodium falciparum, 189 Plastic bullet, 78 Pond fracture, 83 Pons, 11, 13 Post-concussion syndrome, 22, 22 Posterior inferior cerebellar artery (PICA), 130 Post-mortem angiography, 55 Post-mortem bruising, 69 Post-mortem examination before, 55 brain swelling and brain oedema, recognition of, 128–132 in COVID-19 deaths, 60 CT imaging, 95

general, 55 injuries’ appearance in, 66 of skull, 82 in suspicious death, 151 on vehicle occupants, 180–181 Post-traumatic epilepsy, 20–21, 85 Powder burns, gunshot wound with, 73, 74 Pressure, 45 Previous irradiation, complications of, 193 Primary sensory cortex, 11 Primary traumatic head injury, biomechanics of, 45 angular kinematics, 50 cerebral contusions, 50 concussion, 50 diffuse traumatic axonal injury (dTAI), 51 effect of cavitation, 50–51 inertial loading, shaking and whiplash, 51–52 infant vulnerability to inertial loading, 52 intracerebral haematomas (ICH), 51 shaking, characterisation of, 52 subdural haematoma (SDH), 51 biomechanical loading environment, characterising, 45 crush loading, 45–47 head impact response, characterisation of, 47–48 impact-induced head injuries, 48 acute subdural haematoma (ASDH), 49 contusions, 49 epidural haematoma, 49–50 skull fracture tolerance, 49 subdural haematoma (SDH), 49 translational kinematics, 48–49 mechanical loading environment, 45 Prognostic indicators, 22 Prosencephalon, 11 Protozoal infections, CNS, 187 Psychoactive substances, novel, 200 excited delirium (ED), 200 neurotoxins, 200 P-tau pathology, 147 ‘Punch drunk’ syndrome, 145 Punches lacerations on the face, 69 to the mouth, 70 Puppe’s rule, 83, 83 compound elevated skull fractures, 84 depressed skull fractures, 83, 83 hinge fractures, 84, 84 mosaic fracture, 83 pond fracture, 83 ring fractures, 83–84 transmitted fractures/contre-coup fractures, 84, 84 Purkinje cells, 201 Pyramids, medulla at the level of, 13

R Radiofrequency pulses, 25 Radiological coiling/embolisation, 193 Raised intracranial pressure, 165 Rapid microwave fixation method for brain fixation, 62 Reconstructive surgery, 21 Recording information, 55 Recreational drugs amphetamines, 200 cocaine, 199–200 gases carbon disulphide, 201 carbon monoxide, 200–201 heavy metals arsenic, 201 mercury, 201 organic arsenic, 201 thallium, 201 novel psychoactive substances (NPS), 200 excited delirium (ED), 200 neurotoxins, 200 opioids, 199 Red neurons versus dark neurons, 135–137 Respirator brain, 137 Resuscitation, 165 Retinal haemorrhage, 163, 164 causes of, 164, 165 Retrolentiform part, 14 Reversed tentorial herniation, 130 Revolvers, 72 Rifled weapon injuries, 72 contact wounds, 73, 73 distant wounds, 74 exit wounds, 75 near contact and intermediate wounds, 73–74 skull damage, 74–75 tangential wounds, 74 Ring fractures, 83–84 Road traffic deaths, 178 falls, 181 falls down stairs, 182 falls from a height, 181–182 falls from standing, 181 kicking versus falling, 182 motorcycles and pedal cycles, 181 pedestrians, 179–180 run-over injuries, 180 vehicle occupants, 180–181 vehicular collision investigation, 181 Rotational and angular velocities and accelerations, 50 Rubber bullets, 78 Run-over injuries, 180 R v Abadom (1983), 204 R v Bowman (2006), 205 R v Burridge (2010), 206 R v Clark (2003), 205 R v Conaghan and Others (2017), 206

215

Index

216

R v Harris and Others (2006), 205 R v J L-J (2000), 203 R v Marquardt (2011), 206 R v Mohan 1994, 203 R v Ranger (2003), 206 R v Reed, Reed and Garmson (2009), 203 R v Turner (1975), 204

S Sacral fractures, 172–173, 173 Sampling protocols, 119 Scalp, 1, 3–4 injuries, caused by machete, 71, 72 lacerations, 20 petechiae, 55, 56 Screwdriver cerebral damage caused by penetration with, 71 stab wounds caused by, 71 Seat-belt injuries, 180, 180 Secondary long tract degeneration, 123 Second impact syndrome (SIS), 145, 148–149 Self-loading pistol, 72 Semispinalis capitis, 3 Sensory decussation, medulla at the level of, 13 Shaken baby syndrome, 160 Shaking, 51–52 characterisation of angular kinematics, 52 linear kinematics, 52 Sharp-edged objects, injuries caused by, 71 Shotgun injuries, 75 brain, appearance of, 77 contact wounds, 75 distant wounds, 76 near contact and intermediate wounds, 75–76 to neck, 76 with spread shot, 76 wound ballistics, 76–77 Sinking skin flap syndrome, 131 Skin damage, caused by vomitus ­ post-mortem, 66, 68 electrical injury, 72 histology of wounds in, 78–79 laceration of, 69 of the neck, 3 sinking skin, 131 splitting of, 73 stippling of, 73 Skull, removal of, 55–56 Skull bones, 4, 4 Skull damage, 74–75 Skull fractures, 20, 28, 151, 152, 153–154, 153 complex, 86 tolerance, 49

Skull fractures, adult, 80 documentation, 80 histology, 84–85 linear skull fractures, 80 front-to-back hinge fracture, 82–83, 82 linear frontal fractures, 81–82, 82 linear squamous temporal bone fractures, 81, 81 occipital, 81, 81 medico-legal issues bruising, 85, 85 infections, 85 post-traumatic epilepsy, 85 Puppe’s rule, 83, 83 compound elevated skull fractures, 84 depressed skull fractures, 83, 83 hinge fractures, 84, 84 mosaic fracture, 83 pond fracture, 83 ring fractures, 83–84 transmitted fractures/contre-coup fractures, 84, 84 reconstruction of skull, 86 Skull showing external bevelling caused by buckshot, 75 Skull showing internal bevelling caused by buckshot, 74 Slit-like exit wound, 75 Small vessel disease (SVD), 186 Sphenoid air sinuses, 2 Spinal column, 14, 15 Spinal cord, 14, 16 artefact, 59 examination of, 58 internal structure of, 15–16 penetrating injuries of, 174 Spinal cord injury without obvious radiological abnormality (SCIWORA), 168 Spinal injuries, 159, 167, 192 classification of, 168 clinical assessment, 167, 168 early management of, 174 long-term complications of, 174–175 mechanisms and forces involved, 167 neuropathology of, 173–174 paediatric aspects, 176 pathophysiological reactions to, 174 plain radiographic/CT signs of, 168 radiological investigation, 167–168 special categories of, 175–176 vertebral injuries, classification of, 168 associated injuries, 173 sacral fractures, 172–173, 173 subaxial cervical injuries, 171, 172 suboccipital injuries, 169–171 thoracolumbar injuries, 172 vertebral injuries, management of, 174 Spinous process avulsion fracture, 171 Splenius capitis, 3

Spongiform leukoencephalopathy, 199 Spread shot, shotgun wound with, 76 Squamous temporal bone fractures, 81, 81 Squier v General Medical Council (2016), 206 Stable fractures, mobilisation of, 174 Stab wounds, 71 caused by a screwdriver, 71 Stains, histological, 58 Stairway falls, 160 Strain, 45, 46 Streptococcus pneumoniae, 187 Streptococcus viridans, 188 Stress, 113 ‘Stress-coat’ method, 80 Stroke suspected stroke, examination in cases of, 60 and trauma, 103 Subarachnoid blood, MRI of, 38 Subarachnoid haemorrhage, 55, 56, 155, 199 in association with traumatic brain injuries, 94 and intraventricular haemorrhage, 35–38, 36 patchy, 94, 95 spontaneous, 99 histological features, 100 medico-legal issues, 100 pathology, 99–100 traumatic basal, 94 circumstances and mechanism of injury, 94–95 histology, 96–99 macroscopic neuropathology, 96 medico-legal issues, 99 post-mortem examination, 95–96 Subaxial cervical injuries, 171, 172 Subdural effusions, 33 Subdural haematoma (SDH), 28–29, 30, 31, 32, 43, 49, 51, 89 acute subdural haematoma, 89 circumstances and clinical features, 89 histology, 90 macroscopic pathology, 89–90 mechanism of injury, 90 previous injury/disease, 90 ageing of, 91 chronic subdural haematoma (CSDH), 90, 91 histology of SDH, 91–92 macroscopic appearances, 90–91 problem areas in, 92 subdural hygromas, 92 re-bleeding in, 160 Subdural haemorrhages (SDHs), 154–155 and effusions, 29–32 Subdural hygromas, 32, 157–158 Subfalcine herniation, 129, 129

Index Suboccipital injuries, 169–171 Subscalp bruising/haematomas, 55, 56, 58 Sudden death in head injury, 182–183 Sudden infant death syndrome (SIDS) and suffocation, 183 Sudden unexpected death in epilepsy (SUDEP), 60, 139 autopsy in, 140–141 definition of epilepsy and, 139 genetics of epilepsy, 140 hippocampal maldevelopment associated with sudden death (HMASD), 142 incidence, 139 in children, 139 mechanism of death in, 139–140 medullary brain lesions, sudden unexpected death related to, 142–143 neuropathology of, 141–142 post-traumatic epilepsy, 140 Sudden unexplained death in childhood (SUDC), 59–60, 64, 142, 142, 166 Sudden unexplained death in childhood with febrile seizure phenotype (SUDC-FS), 142 Superior cerebellar artery (SCA), 130 Superior colliculus, 11, 12 Surface injuries, 66 Surgery intracranial, 20 neoplasm, 192 reconstructive, 21 Susceptibility-weighted imaging (SWI/ SWAN), 25 Suspected stroke, examination in cases of, 60 Suspicious death, post-mortem examination in a case of, 151 Swelling axonal, 173, 173 brain, see Brain swelling cerebral, 155 Swirl sign, 35, 36 ‘Swiss cheese’ artefact, 59 Sylvian fissure, 100 Sylvian fissure haematoma, 120 Syncope, 81 Syringomyelia, 175, 175

T T1-weighted imaging, 25 T2 gradient-echo sequence, 25 T2-weighted imaging, 25 Tangential wounds, 74 Teardrop fracture, 171, 171 Techniques, 55 adult brain weights, 61, 62 artefacts, 59 before post-mortem examination, 55

blocking of the brain, 58 blocks in SUDI cases, 62 brain removal in newborns and young infants, 57 clinicopathological entities, additional procedures for infectious cases, 60 miscellaneous, 60 sudden death in epilepsy (SUDEP), 60 sudden unexpected death in infancy (SUDI), 59–60 suspected stroke, examination in cases of, 60 defence, pathologists acting on behalf of, 58 dementia, assessment of, 62–64 diatom analysis, procedure for, 64–65 examination after fixation, 58 examination of the spinal cord, 58 external examination, 55 general post-mortem examination, 55 histological stains, 58 hospital therapy, findings relating to, 58 immediate neuropathological examination, 56–57 infant and child brain weights, 61 internal examination removal of the skull, 55–56 subscalp bruising/haematomas, 55 neuropathological examination, 55 rapid microwave fixation method for brain fixation, 62 recording information, 55 retention of brain, 57–58 Teeth, damage to, 69, 69 Tegmentum, 11 Tentorium cerebelli, 5, 5 Terson syndrome, 165 Thalamus, 11 Thallium, 201 Thiamine deficiency, 196 Thoracic spine, fracture-dislocations in, 172 Thoracolumbar injuries, 172 Thrombolytic drugs, 193 Tissue tear haemorrhages, 111 Tissue-tear lesions, 117–119 Tonsillar herniation, 130 ‘Toothpaste’ effect, 59 Tramline bruise, 69, 69 Translational kinematics, 48–49 Transmitted fractures, 84, 84 Transtentorial uncal herniation, 129–130 Trapezius muscle, 3 Trauma sampling protocol, 63 Traumatic alopecia, 152 Traumatic aneurysms, 101 aetiology and pathogenesis, 101–102 Traumatic arteriovenous fistula, 102

Traumatic axonal injury (TAI), 116, 124; see also Axonal damage clinical and forensic correlates, 116–117 diffuse traumatic axonal injury, 116 grading of, 116 evolution of, 121–124 recognising, 119–121 sampling protocols, 119 tissue-tear lesions, 117–119 Traumatic basal subarachnoid haemorrhage, 94 circumstances and mechanism of injury, 94–95 histology, 96–99 macroscopic neuropathology, 96 medico-legal issues alcohol and traumatic subarachnoid haemorrhage, 99 delayed presentations, 99 trauma severity, 99 post-mortem examination, 95–96 Traumatic brain injury (TBI), 46, 145 subarachnoid haemorrhage in association with, 94 Traumatic encephalopathy syndrome (TES), 145 Traumatic haemorrhage in septum pellucidum and ventricles, 39 Traumatic haemorrhagic contusion, 39 Traumatic intracerebral haematomas, 111 brainstem haemorrhages, 112 ganglionic haematomas, 111–112, 112 petechial white matter haemorrhages, 111 tissue tear haemorrhages, 111 traumatic intraventricular haemorrhage, 112, 112 Traumatic intraventricular haemorrhage, 36, 112, 112 Traumatic retinoschisis and perimacular folds, 164 Traumatic subarachnoid haemorrhage, 35, 37, 38 Traumatic vascular injuries, 100 arterial dissection, 100–101 traumatic aneurysms, 101 aetiology and pathogenesis, 101–102 traumatic arteriovenous fistula, 102 Tuberculous meningitis, 188, 188 Tumour necrosis factor alpha (TNF-α), 79

U Unstable injuries, 174 Unusual weapons air-powered weapons, 77 crossbow injuries, 78 humane killers, 78 rubber bullets, 78 Upward tentorial herniation, 130

217

Index

218 V

Vascular dementia, 192 Vascular malformations, 114 Vasodilatation, 132 Vasogenic oedema, 128 Vehicle occupants, 180–181 Vehicular collision investigation, 181 Venous angiomas, 114 Venous drainage of brain, 6 Ventricular catheters, 58 Ventriculo-peritoneal shunt (VP shunt), 131 Vertebral artery, 5 Vertebral column, 14, 15 Vertebral injuries classification of, 168 associated injuries, 173 sacral fractures, 172–173, 173 subaxial cervical injuries, 171, 172 suboccipital injuries, 169–171 thoracolumbar injuries, 172 management of, 174, 174

Vertebrobasilar arterial system, 96 Vestibule, 2 Viral infections, CNS, 187 Virchow-Robin spaces, 40

W Watershed infarction, 43 Wedge compression fractures, 172 Wernicke’s encephalopathy, 196, 197 macroscopic examination, 196 microscopic examination, 196 alcohol and atrophy, 196 cerebellar degeneration, 196–197 Whiplash, 51–52 Whiplash injuries of neck, 176 White Burgess Langille Inman v Abbott and Halibuton Co. (2015), 204 White cerebellum and pseudo-SAH signs, 43 Wounds contact gunshot wounds, 73, 73 contact wounds, 75

distant, 74, 76 entrance, 75 exit, 75, 75 gunshot, 73, 74, 178, 179 gutter, 74 high-energy transfer wounds, 77 incised wounds and stab wounds, 71 intermediate, 73–74, 75–76 low-energy transfer wounds, 77 shotgun wound to neck, 76 with spread shot, 76 skin wounds, histology of, 78–79 slit-like exit wound, 75 stab wounds, 71 caused by a screwdriver, 71 tangential, 74 Wound ballistics, 76–77 Wounding, defined, 66

Z Zip guns, 77