Forensic Pathology Reviews, Volume 6

Forensic Pathology Reviews, Volume 6

Forensic Pathology Reviews, Volume 6 Michael Tsokos, md, Series Editor Volume 1 (2004) • Hardcover: ISBN 1-58829-414-5 Full Text Download: E-ISBN 1-59259-786-6 Volume 2 (2005) • Hardcover: ISBN 1-58829-414-3 Full Text Download: E-ISBN 1-59259-872-2 Volume 3 (2005) • Hardcover: ISBN 1-58829-416-1 Full Text Download: E-ISBN 1-59259-910-9 Volume 4 (2006) • Hardcover: ISBN 1-58829-601-6 Full Text Download: E-ISBN 1-59259-921-4 Volume 5 (2008) • Hardcover: ISBN 978-1-58829-832-4 Full Text Download: E-ISBN 978-1-59745-110-9 Volume 6 (2011) • Hardcover: ISBN 978-1-61779-248-9 Full Text Download: E-ISBN 978-1-61779-249-6

Forensic Pathology Reviews Volume 6

by

Elisabeth E. Turk, MD Medizinische Abteilung, Asklepios Klinik Harburg, Hamburg, Germany

Editor Elisabeth E. Turk Medizinische Abteilung Asklepios Klinik Harburg Eißendorfer Pferdeweg 52 21075 Hamburg Germany [email protected]

ISSN 1556-5661 ISBN 978-1-61779-248-9 e-ISBN 978-1-61779-249-6 DOI 10.1007/978-1-61779-249-6 Springer New York Dordrecht Heidelberg London © Springer Science+Business Media, LLC 2011 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Humana Press, c/o Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights While the advice and information in this book are believed to be true and accurate at the date of going to press, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. Printed on acid-free paper Humana Press is part of Springer Science+Business Media (www.springer.com)

Preface

The field of forensic pathology is continuing to expand. Advances in research open up new fields of forensic expertise and exciting possibilities for national and international cooperation with a large variety of disciplines in medicine and other sciences. Autopsy and laboratory techniques are evolving rapidly, and existing techniques yield more and more accurate results. In Forensic Pathology Reviews, Volume 6, leading national and international authors provide cutting-edge reviews of key recent advances in the fields of traumatic death, sudden natural death and death time estimation. The new volume now features many color illustrations. The reviews are aimed at forensic experts across the world, serving as a guide to practical aspects as well as recent advances in forensic science and medicine. I wholeheartedly thank all authors who contributed to this volume. Their hard work and constant support has been a great inspiration. Hamburg, Germany

Elisabeth E. Turk

v

wwwwwwwwwwwwwwwww

Contents

  1 Sudden Natural Deaths in Infancy and Childhood.............................. Neil E.I. Langlois and Roger W. Byard   2 Post-mortem Investigation of Sudden Unexpected Death in Infancy: Role of Autopsy in Classification of Death........................ Martin A. Weber and Neil J. Sebire   3 Sudden Death from Pulmonary Causes................................................ Kris S. Cunningham and Michael S. Pollanen   4 Sectioning of the Heart, Searching for Pathology Under the Microscope, and the Cardiac Proteomics Approach in the Study of Sudden Cardiac Death Cases....................................... Vittorio Fineschi, M.S.B. Othman, and Emanuela Turillazzi   5 Endocrine Disorders with Potentially Fatal Outcome......................... Lars Hecht

1

27 47

75 93

  6 Sudden Death from Infectious Disease................................................. 121 James A. Morris, Linda M. Harrison, and Robert M. Lauder   7 Aviation Deaths....................................................................................... 145 S. Anthony Cullen   8 Fatalities in General Aviation: From Balloons to Helicopters............ 169 Alex de Voogt   9 The 9/11 Attacks: The Medicolegal Investigation of the World Trade Center Fatalities.................................................... 181 James R. Gill, Mark Desire, T. Dickerson, and Bradley J. Adams 10 Injuries and Fatalities in All-Terrain Vehicle Crashes........................ 197 Richard J. Mullins and J.H. Mullins

vii

viii

Contents

11 Advances in Entomological Methods for Death Time Estimation..................................................................... 213 Martin H. Villet and Jens Amendt 12 Tissue Fluorescence Spectroscopy in Death Time Estimation............ 239 Éverton S. Estracanholli, Cristina Kurachi, and Vanderlei S. Bagnato 13 Heat-Flow Finite-Element Models in Death Time Estimation............ 259 Holger Muggenthaler, Michael Hubig, and Gitta Mall 14 The Use of Protein Markers for the Estimation of the Postmortem Interval.................................................................... 277 Yekaterina Poloz and Danton H. O’Day 15 Alcohol and Drug Fatalities in Transportation: Forensic-Toxicological Implications...................................................... 295 F. Mußhoff Index................................................................................................................. 331

Contributors

Bradley J. Adams, PhD  Office of Chief Medical Examiner, 520 First Avenue, New York, NY 10016, USA [email protected] Jens Amendt, PhD  Institute of Forensic Medicine, University of Frankfurt, Kennedyallee 104, 60596 Frankfurt/Main, Germany [email protected] Vanderlei S. Bagnato, PhD  Instituto de Física de São Carlos, Universidade de São Paulo, Av. Trabalhador São-Carlense 400, São Carlos, SP 13566-590, Brazil [email protected] Roger W. Byard, MBBS, MD, FRCPC  Discipline of Pathology, The University of Adelaide, Level 3 Medical School North Building, Frome Road, Adelaide 5005, SA, Australia [email protected] S. Anthony Cullen, MD (R.I.P.)  Lately Consultant Pathologist, RAF Centre of Aviation Medicine, RAF Henlow, Bedfordshire, SG16 6DN, UK Kris S. Cunningham, MD, PhD  Provincial Forensic Pathology Unit, Ontario Forensic Pathology Service, Centre for Forensic Science and Medicine, University of Toronto, 26 Grenville Street, 2nd Floor, Toronto, ON, Canada M7A 2G9 [email protected] Mark Desire, MS, JD  Office of Chief Medical Examiner, 520 First Avenue, New York, NY 10016, USA [email protected] T. Dickerson, MS  Office of Chief Medical Examiner, 520 First Avenue, New York, NY 10016, USA [email protected] ix

x

Contributors

Éverton S. Estracanholli, MS  Instituto de Física de São Carlos, Universidade de São Paulo, Av. Trabalhador São-Carlense 400, São Carlos, SP 13566-590, Brazil [email protected] Vittorio Fineschi, MD, PhD  Department of Forensic Pathology, University of Foggia, Ospedale Colonnello D’Avanzo, Via degli Aviatori 1, 71100 Foggia, Italy [email protected] James R. Gill, MD  Office of Chief Medical Examiner, 520 First Avenue, New York, NY 10016, USA [email protected] Linda M. Harrison, MD  Department of Pathology, University Hospitals of Morecambe Bay NHS Trust, Royal Lancaster Infirmary, Lancaster, UK [email protected] Lars Hecht, MD  Institut für Pathologie, HELIOS Klinikum Bad Saarow, Pieskower Street 33, 15526 Bad Saarow, Germany [email protected] Michael Hubig, PhD  Institut für Rechtsmedizin, Universitätsklinikum Jena, 07740 Jena, Germany [email protected] Cristina Kurachi, MD  Instituto de Física de São Carlos, Universidade de São Paulo, Av. Trabalhador São-Carlense 400, São Carlos, SP 13566-590, Brazil [email protected] Neil E.I. Langlois, MD  Forensic Science SA and University of Adelaide, 21 Divett Place, Adelaide, SA 5000, Australia [email protected] Robert M. Lauder, PhD  Division of Biomedical and Life Sciences, School of Health and Medicine, Lancaster University, Lancaster, UK [email protected] Gitta Mall, MD  Institut für Rechtsmedizin, Universitätsklinikum Jena, 07740 Jena, Germany [email protected] James A. Morris, MD, FRCPath  Department of Pathology, University Hospitals of Morecambe Bay NHS Trust, Royal Lancaster Infirmary, Lancaster, UK [email protected] Holger Muggenthaler, PhD  Institut für Rechtsmedizin, Universitätsklinikum Jena, Jena 07743, Germany [email protected] J.H. Mullins, BA  Nursing School, University of Minnesota, Minneapolis, MN, USA [email protected]

Contributors

xi

Richard J. Mullins, MD  Department of Surgery, Trauma/Critical Care Section L611, Oregon Health and Science University, 3181 SW Sam Jackson Park Road, Portland, OR 97239, USA [email protected] F. Mußhoff, PhD  Department of Toxicology, Institute of Legal Medicine, Universitätsklinikum Bonn, Stiftsplatz 12, 53111 Bonn, Germany [email protected] Danton H. O’Day, BSc, MSc, PhD  Department of Cell and Systems Biology, University of Toronto, M5S 3G5 Toronto, ON, Canada Department of Biology, University of Toronto, L5L 1C6, Mississauga, ON, Canada [email protected] M.S.B. Othman,  Jabatan Perubatan Forensik Hospital Ipoh, Perak Darul Ridzuan, Malaysia, [email protected] Michael S. Pollanen, BSc, MD, PhD, FRCPC, DMJ (Path), MRCPath  Provincial Forensic Pathology Unit, Ontario Forensic Pathology Service, Centre for Forensic Science and Medicine, University of Toronto, 26 Grenville Street, 2nd Floor, Toronto, ON, Canada M7A 2G9 [email protected] Yekaterina Poloz, PhD  Department of Cell and Systems Biology, University of Toronto, M5S 3G5 Toronto, ON, Canada [email protected] Neil J. Sebire, MD, DRCOG, FRCPath  Department of Histopathology, Camelia Botnar Laboratories, Great Ormond Street Hospital, London WC1N 3JH, UK [email protected] Emanuela Turillazzi, MD, PhD  Department of Forensic Pathology, University of Foggia, Ospedale Colonnello D’Avanzo, Via degli Aviatori 1, 71100 Foggia, Italy [email protected] Martin H. Villet, MSc, PhD, PGDHE  Department of Zoology and Entomology, Rhodes University, Grahamstown 6140, South Africa [email protected] Alex de Voogt, MD  Division of Anthropology, American Museum of Natural History, 200 Central Park West, New York, NY 10024, USA [email protected] Martin A. Weber, DCH(SA), FRCPath  Department of Histopathology, Camelia Botnar Laboratories, Great Ormond Street Hospital, London WC1N 3JH, UK [email protected]

Chapter 1

Sudden Natural Deaths in Infancy and Childhood Neil E.I. Langlois and Roger W. Byard

Abstract  Nontraumatic sudden and unexpected death in the young is an ­uncommon event, but it is one that has tremendous impact on families and communities. Autopsy assessments may be difficult, as many entities are rare and have quite ­subtle manifestations. In the following chapter, a range of lethal (mostly) ­natural diseases that may be encountered in pediatric forensic practice are described ­involving central nervous system, respiratory, cardiovascular, gastrointestinal, hematopoetic, genitourinary, infectious, endocrine, metabolic, and miscellaneous conditions, the latter including connective tissue and chromosomal disorders. These are listed in Table 1.1. The issue of the “negative autopsy” is also discussed with the potential role of genetic screening. Given the broad nature of the subject matter, the chapter is not intended to be exhaustive, but covers a range of conditions that should be considered during the process of the postmortem examination process. Keywords  Sudden unexpected death • Childhood • Sudden infant death syndrome

Introduction The postmortem examination process in all cases of sudden death in the young begins with consideration of the history and circumstances surrounding the death [1]. A scene visit may be beneficial. Certain aspects may have to be deliberately sought. For example, symptoms of presyncope, syncope, and palpitations preceding

R.W. Byard (*) Discipline of Pathology, The University of Adelaide, Level 3 Medical School North Building, Frome Road, Adelaide 5005, SA, Australia e-mail: [email protected] E.E. Turk (ed.), Forensic Pathology Reviews, Volume 6, Forensic Pathology Reviews 6, DOI 10.1007/978-1-61779-249-6_1, © Springer Science+Business Media, LLC 2011

1

2

N.E.I. Langlois and R.W. Byard

death may not have been considered important and therefore not have been ­volunteered [2], or a history of circumstances that may have resulted in commotio cordis may not be forthcoming without direct questioning [3–5]. Certain medical conditions, such as myotonic or muscular dystrophies may not be immediately apparent on examination, but these conditions can be associated with sudden death due to cardiac arrhythmias [6–9] and obtaining a personal or family history may be revealing. Following on from the review of the history and circumstances of death, an external examination may reveal features suggestive of congenital abnormalities that may indicate an underlying syndrome and a possible cause for sudden death. Examples include William [10], Downs [11], and Marfan [12] syndromes. It is also possible to search databases for possible syndromes linked to a constellation of abnormal morphologic findings (http:ncbi.nlm.nih.gov/sites/entrez?db = omim). Photography is important for documentation of dysmorphic features, and extensive use of photography should be considered in pediatric autopsy cases for recording normal, as well as abnormal findings. Preautopsy radiography may also be informative in identifying abnormalities that may be linked to sudden death, such as the presence of a pneumothorax [13]. Consideration should be given to performing computerized tomographic (CT) scanning or magnetic resonance imaging (MRI) [14–17]. Progressing from the external examination to the internal examination may reveal an apparent cause of death. This will be reviewed using a systematic approach, while recognizing that the autopsy in practice is performed on a regional, cavity basis and that some causes of death are not isolated to a particular organ or system (summarized in Table 1.1).

Central Nervous System Examination of the head and cranial cavity should be performed in all cases to exclude trauma. Subarachnoid hemorrhage in children and young adults is more often due to arteriovenous malformation [18] than a berry aneurysm [19]. Subdural hemorrhage may result from blunt head trauma; however, microscopic hemorrhages in the dura may be due to an artifact caused in removing the dura [20]. There may be a history of epilepsy that is a recognized cause of sudden death at all ages. Sudden death in epilepsy (SUDEP) has been defined as “sudden, unexpected, witnessed or unwitnessed, nontraumatic, and nondrowning death in patients with epilepsy, with or without evidence for a seizure and excluding documented status epilepticus where necropsy examination does not reveal a toxicological or anatomical cause for death” [21]. Death tends to occur in poorly controlled patients, frequently occurring at night in bed [22–28]. It has been suggested that electrical discharges within the brain occurring during a seizure result in a lethal cardiac arrhythmia or fatal autonomic disturbance [29–34]; alternatively, apnea during the tonic phase of a fit may be a mechanism of death [21, 35]. Examination of the brain

1  Sudden Natural Deaths in Infancy and Childhood

3

Table 1.1  Causes of sudden death in childhood Central nervous system Infections Epilepsy Hemorrhage Tumors Metabolic disorders Structural abnormalities Miscellaneous

Respiratory Infections Upper airway obstruction Asthma Bronchopulmonary dysplasia Miscellaneous Cardiovascular Infections Congenital cardiac defects (before and after surgery) Cardiomyopathies Subaortic stenosis Valvular abnormalities Aortic abnormalities Coronary artery abnormalities

Venous abnormalities Vascular malformations Pulmonary hypertension Tumors Conduction defects Miscellaneous

Gastrointestinal Infections Intestinal obstruction Intestinal perforation Mesenteric defects Gastroesophageal reflux/aspiration Late-presentingcongenital diaphragmatic hernia Gastrointestinal hemorrhage Miscellaneous

Bleeding diatheses, vascular malformations

Moyamoya disease, fibromuscular dysplasia, Friedreich ataxia, tuberous sclerosis, von Recklinghausen disease, Guillain–Barre syndrome

Massive pulmonary hemorrhage, idiopathic pulmonary hemosiderosis, tension pneumothorax

Aortic stenosis, mitral valve prolapse syndrome, tricuspid valve prolapse Supravalvular stenosis, coarctation, William syndrome, DiGeorge syndrome Anomalous coronary arteries, aplasia/hypoplasia, idiopathic arterial calcinosis, coronary arteritis (Kawasaki disease) Total anomalous pulmonary venous drainage

Endocardial fibroelastosis, fibromuscular dysplasia, aortic cystic medial necrosis, Budd–Chiari syndrome, thromboembolism Gastroenteritis Intussusception, volvulus

Cystic fibrosis, pancreatitis, anorexia nervosa/malnutrition (continued)

N.E.I. Langlois and R.W. Byard

4 Table 1.1  (continued) Hematological Hemoglobinopathies Malignancies Bleeding diatheses Anemia Miscellaneous Genitourinary Primary renal disease Urinary tract obstruction Wilms tumor Complications of sexual maturity Miscellaneous Infectious Cardiovascular Respiratory Central nervous system Hematologic Gastrointestinal Genitourinary Generalized

Sickle cell disease Lymphoma, leukemia

Infections, polycythemia, splenic disorders Pyelonephritis, glomerulonephritis

Ruptured ectopic pregnancy, amniotic fluid embolism Hemolytic-uremic syndrome Myocarditis, endocarditis, rheumatic fever, aortitis Acute epiglottitis, acute bronchopneumonia Meningitis, encephalitis, poliomyelitis Malaria, infectious mononucleosis Gastroenteritis, botulism Pyelonephritis Septicemia, endotoxemia

Endocrine Congenital adrenal hyperplasia Diabetes mellitus Thyroid disease Metabolic Reye syndrome Fatty acid oxidation defects Carbohydrate disorders Amino acid disorders Urea cycle disorders Organic acid disorders

Acyl-CoA dehydrogenase deficiencies (MCAD, LCAD) Galactosemia, glycogen storage diseases Homocystinuria

Miscellaneous Connective tissue disorders Marfan syndrome, Ehlers–Danlos syndrome type IV Chromosomal disorders Trisomy 21 Skeletal disorders Achondroplasia Dermatological disorders Hypohydrotic ectodermal dysplasia Muscular conditions Malignant hyperthermia Immunological conditions Anaphylaxis (Adapted from Byard RW, Sudden Death in the Young, Cambridge University Press, 2010)

by a neuropathologist is useful in epilepsy-related deaths [36] as an underlying cerebral lesion responsible for seizures may be found in up to 60% of cases [37]. For this reason, retention of the brain for formal examination should be discussed with family members at the earliest available opportunity.

1  Sudden Natural Deaths in Infancy and Childhood

5

Occasionally, sudden death may be the first presentation of a central nervous system tumor, possibly associated with hemorrhage [38, 39]. Sudden death may also occur due to obstruction of cerebrospinal fluid flow in children with Chiari malformation or from centrally placed lesions such as colloid cysts [40]. Ventriculoatrial shunts have been associated with recurrent and lethal pulmonary thromboemboli [41] and ventriculoperitoneal shunts may result in lethal sepsis if the tip of a catheter perforates the intestine [42]. Function of the shunt can usually be tested during the postmortem examination by injection or manual operation of the reservoir. Although certain syndromic disorders may present with central nervous systemrelated manifestations, the final cause of death may not be due to cerebral disease. An example of this is tuberous sclerosis where there may be epilepsy associated with cortical tubers, with sudden death being due instead to the effects of cardiac rhabdomyomas. Lethal episodes involve outflow obstruction and arrhythmias [43]. Vascular malformations may be associated with lethal intracranial hemorrhage or seizures, the latter also occurring with a range of developmental abnormalities and metabolic conditions such as Lafora, Leigh, and Rett syndromes. The heritable nature of many of these conditions makes accurate diagnosis at autopsy imperative [44–48]. Causes of death in neurofibromatosis may involve vascular disease and the effects of tumors [49, 50].

Respiratory The oropharyngeal and laryngeal region should be carefully examined in all cases of unexpected death for causes of obstruction, including local infections such as acute epiglottitis (Fig. 1.1), congenital anomalies such as broncho-laryngomalacia [51, 52], laryngeal cysts [53], lingual thyroglossal cysts (Fig.  1.2) [54], hemangiomas [55], and aspirated foreign bodies [56, 57]. Infants with syndromes associated with mandibular hypoplasia are also at risk of acute airway obstruction [58]. Care should be taken to avoid over interpreting artifactual injuries caused by intubation [59]. However, medical procedures such as nasogastric tube insertion or inflicted injury may result in retropharyngeal abscess [60]. Individuals with tracheostomies may die unexpectedly if airway obstruction occurs from mucus plugging, or if an indwelling tube erodes into a major blood vessel [61]. Before opening the chest, the presence of a pneumothorax should be sought. This may be assessed by radiology [13], or at autopsy, by puncturing the chest wall beneath a level of water to look for escaping air [62]. Primary pneumothorax typically occurs in young, thin, tall males and is nonlethal; however, a tension pneumothorax may be rapidly fatal by causing midline shift of mediastinal structures [63]. Distended lungs with covering of the anterior surface of the heart with rib ­markings on the pleural surfaces raise the possibility of acute asthma. Other signs, such as plugging of bronchi, eosinophils, Charcot–Leyden crystals, bronchial wall thickening with edema, and basement membrane thickening [64–66] should be

6

N.E.I. Langlois and R.W. Byard

Fig. 1.1  Edema of the epiglottis with narrowing of the glottic inlet in a child with acute bacterial epiglottitis [1]

sought. However, these findings may not be present in up to 50% of cases [67], particularly if there has been attempted resuscitation [68], and a predominance of neutrophils rather than eosinophils has been described in acute deaths [69]. Primary diaphragmatic pathology may also result in a lethal outcome [70].

Cardiovascular Examination of the cardiovascular system begins with a general assessment of the anatomy, with particular regard to congenital abnormalities. Congenital abnorma­lities of the heart and great vessels may be unsuspected and can present with sudden death [1, 71–73]; death can occur following surgical correction, even late [71, 74–76]. Conditions include tetralogy of Fallot, transposition of the great vessels, left ventricular­ hypoplasia, and valvular stenoses (Fig. 1.3). Venous abnormalities are not usually associated with sudden death unless there is anomalous pulmonary venous drainage. In this condition, there is complete or partial drainage of pulmonary venous blood into the right side of the heart. In cases of compete anomalous drainage, there has to be a mechanism to enable shunting of

1  Sudden Natural Deaths in Infancy and Childhood Fig. 1.2  View of the posterior portion of the tongue showing a small lingual thyroglossal duct cyst at the foramen cecum. Enlargement would result in pressure on the epiglottis with posterior displacement and airway compromise

Fig. 1.3  Dysplasia of the aortic valve with significant stenosis

7

8

N.E.I. Langlois and R.W. Byard

Fig. 1.4  Thickening of the wall of a coronary artery due to vasculitis associated with Kawasaki disease (arrow)

oxygenated blood through a patent ductus arteriosus or a septal defect for survival to be possible [77–79]. Certain congenital anomalies of the coronary arteries, such as origin from the pulmonary trunk and left main origin from the right aortic sinus with the artery passing between the pulmonary artery and aorta, are associated with sudden death, particularly during exercise [80–83]. However, sudden death has been associated with all types of aberrantly sited arteries, and also in cases of arterial hypoplasia or aplasia [84–88]. Critical reduction of blood flow may occur with an acute angle of take off (>45°) or intimal ridges [89]. Rarely coronary artery thromboembolism may occur in childhood, usually in association with rheumatic fever or a congenital cardiac defect [90]. Atherosclerotic coronary artery disease is uncommon in childhood unless a hereditary hyperlipidemic syndrome is present. Coronary flow can also become compromised as a consequence of other conditions that cause arterial narrowing such as Kawasaki disease (Figs.  1.4 and 1.5) [91–93], fibromuscular ­dysplasia [94], and idiopathic arterial calcinosis (Fig. 1.6) [95]. The pulmonary arteries should be inspected, as pulmonary thromboembolism, although uncommon in the young does occur and has similar predisposing factors to older individuals, including conditions such as sepsis and malignancy [96, 97]. In  cases where there is no obvious initiating condition, the possibility of a ­thrombophilia such as mutations in the factor V Leiden and prothrombin genes (G20210A), or deficiencies in anti-thrombin III, plasminogen, and proteins C and S, or antiphospholipid syndrome should be considered [1]. Cerebral embolism may occur paradoxically if there is a septal defect present, or if the source of embolic material is cardiac, and may involve tumor, as in Wilms tumor [98], or infective material, as in cardiac ecchinococcosis [99].

1  Sudden Natural Deaths in Infancy and Childhood

9

Fig. 1.5  Obliteration of the lumen of a coronary artery following Kawasaki disease

Fig. 1.6  Concentric calcification of the media with fibrointimal proliferation in a case of ­idiopathic arterial calcinosis

Pulmonary hypertension may be idiopathic or caused by a variety of ­mechanisms including increased vascular flow from shunting, embolism, and chronic hypoxia. Sudden death may occur due to arrhythmias from right ventricular hypertrophy (Fig. 1.7) [1]. Aortic dissection may be associated with Marfan syndrome [100], in which case genetic counseling of family members may be required. However, the identification

10

N.E.I. Langlois and R.W. Byard

Fig. 1.7  Right ventricular and atrial hypertrophy and dilation in a case of sudden death from ­pulmonary hypertension

of over 500 mutations in the fibrillin gene, the lack of association between ­phenotype and genotype, and the fact that 25–30% of cases are sporadic, makes attempts at determining clinical prognosis difficult [12]. Cardiac hypertrophy is a risk factor for arrhythmia and for sudden death, and may be associated with ventricular septal defects [101, 102]. Coarctation of the aorta may also result in cardiac hypertrophy [103] and has been associated with aortitis [104]. Hypertrophic cardiomyopathy may be apparent if a markedly increased heart weight is found when compared to standard tables, and there may be asymmetrical thickening of the septum [105]. Endocardial thickening may be present over the septum beneath the aortic valve due to impact from the anterior mitral valve cusp [106]. Myocardial fiber disarray is considered a characteristic feature of this ­disorder, but this may be focal, necessitating adequate sampling [106]. Caution is also required, as disorganized myocytes can be found in the normal heart (particularly around the junction of the septum with the anterior and posterior walls [105, 107]), such that a significant amount of disarray must be observed to diagnose the ­condition (around 20% in two histological blocks has been recommended [106]). Nuclear changes, fibrosis, and abnormalities of small intramyocardial vessels may be also observed. Cardiac hypertrophy may also occur due to dilated cardiomyopathy [108], which has nonspecific histological features, with fibrosis being common [109]. It is also recognized that acute cardiac death may be a late complication of ­chemotherapy, for example, following anthracycline [110, 111]. Oncocytic cardiomyopathy refers to a condition of uncertain etiology where ­cardiomegaly is associated with subendocardial or epicardial tan-white nodules

1  Sudden Natural Deaths in Infancy and Childhood

11

with thickening of valves. The nodules are composed of aggregates of round to ­polygonal cells with granular cytoplasm. Sudden and unexpected death may occur in 20% of cases [112]. Myocardial noncompaction is a rare genetically heterogeneous cardiomyopathy where there is abnormal left ventricular development with failure of compaction of loose fetal myocardial fibers giving the left ventricle a spongy appearance with prominent trabeculation [113–117]. Involvement of the heart by tumor has been reported in association with sudden death [118], as has lipomatous hypertrophy of the interatrial septum, although this tends to occur in older individuals [119]. Floppy mitral valve is a recognized association with sudden death [120–123]. Although the mechanism is not understood, it is likely to be due to arrhythmia [124]. There is a reported association with Marfan’s syndrome, Ehlers–Danlos, osteogenesis imperfecta, and pseudoxanthoma elasticum [12, 121] and a familial element is present in some cases [120]. The valves should also be examined for the possibility of endocarditis [125]. Rheumatic heart disease may present acutely and a valvulitis of the mitral valve may be recognized [126]. Histological changes may be confined to the atria, or characteristic Ashoff bodies may be found throughout the heart [126, 127]. A method of sampling of the heart has been described by the TRAGADY group [128]. As part of this protocol, sampling of the right ventricle is advocated to ­identify or exclude right ventricular arrhythmogenic dysplasia (right ventricular cardiomyopathy), which may be associated with death during exertion [129, 130]. It has been proposed that there should be a separation between pure fatty replacement (that may also be related to Uhl anomaly when extensive), which may be regarded a ­congenital disorder, and fibrofatty replacement of the right ventricular myocardium (with or without inflammation), which is considered to be an acquired abnormality resulting from myocarditis with loss of myocytes [131–136]. It has been proposed that only the fibrofatty form should be regarded as true arrhythmogenic right ventricular ­dysplasia (cardiomyopathy) [137]. The changes are most marked in the right ventricular inflow, apex, and infundibular regions (the so-called triangle of dysplasia) [130, 134, 137]. However, involvement of the left ventricle has also been reported [138–140]. Sampling of the heart may also reveal evidence of myocarditis. Nonetheless, care must be taken not to over diagnose myocarditis as a cause of death as collections of lymphocytes without myocytolysis are a not uncommon incidental finding [141]. However, if changes meeting the Dallas criteria are present [142] in the absence of an alternative cause of death, then it is probably reasonable to attribute death to myocarditis [143]. The histological changes may be focal and sampling 8–10 histological blocks has been suggested [144, 145]. It has also been suggested that with advances in molecular biological techniques, it may be possible to make the diagnosis on the basis of PCR (polymerase chain reaction) demonstration of viral infection of the heart in the presence of a suggestive clinical history [146, 147]. Giant cell myocarditis may occur in children that may be an autoimmune disorder dependent on CD-4 positive T-lymphocytes [1].

12

N.E.I. Langlois and R.W. Byard

Gastrointestinal Gastrointestinal causes of sudden death in the young range from intestinal ­obstruction due to volvulus and intussusception, to gastric perforation with peritonitis, which should be apparent upon opening the abdomen [148, 149]. Malrotation of the gut may result in ischemia [150] and herniation of the intestine through congenital defects in the mesentery may also result in lethal obstruction with gangrene (Fig. 1.8) [151]. There may be underlying pathologies that should be sought, for example, gastric perforation is more common in children with severe developmental delay who may swallow air due to neuromuscular incoordination and who may have abnormally deep chest cavities predisposing to gastric torsion [152]. In addition, care must be taken as intussusception may also be a terminal event when death has resulted from other means [149, 153]. Rectal bleeding from an ischemic intestine in early childhood has been mistakenly attributed to trauma from sexual assault [154]. Late presenting congenital diaphragmatic hernia can be a cause of sudden death [155–157].

Fig. 1.8  Small intestinal infarction following herniation through a mesenteric defect (arrow)

1  Sudden Natural Deaths in Infancy and Childhood

13

Hematologic Inspection of the spleen may reveal enlargement, which, with or without the presence of lymphadenopathy, may suggest underlying hematological abnormalities. Abnormalities such as leukemia and sickle cell disease may be present [84, 158]. Sickle cell disease should be considered as a cause of sudden death in African and Mediterranean populations [159], which may follow exertion or minor infection and result in sickle cell sequestration crisis [160]. The diagnosis can be made from histological sections; however, sickling may be a postmortem phenomenon. Electrophoresis will display the presence of abnormal hemoglobin [161]. Leukemia may present with fatal intracerebral hemorrhage following a short period of nonspecific malaise.

Genitourinary Genitourinary causes of sudden death such as Wilms tumor hemorrhage or ­embolism occur rarely [98]. Pathologists may check for vesico-ureteric reflux during the course of the autopsy, due to its association with pyelonephritis. If pyelonephritis is present it may be recognized by the presence of punctate yellow collections of pus in the renal parenchyma, often with surrounding erythema. The diagnosis can be ­confirmed by microscopic examination. Hemolytic-uremic syndrome is caused by a systemic thrombotic microangiopathy that may follow an infection such as gastrointestinal infection with verotoxinproducing Escherichia coli or Shigella dysenteriae type 1. It has also been linked to drugs, tumors, other infectious agents, and possibly to immunization. An acute onset of microangiopathic hemolytic anemia, thrombocytopenia, and renal insufficiency may occur with sudden death from myocarditis or intracerebral hemorrhage [162]. Complications of pregnancy such as a ruptured ectopic pregnancy or amniotic fluid embolism should be suspected in any sexually mature female.

Infectious The examining pathologist must consider potential causes of death that are not ­system or body cavity specific, but which can be generalized in their manifestation and effects. Globally, fulminant infections remain major causes of rapid death in the young, with complications of acute gastroenteritis and malaria accounting for many deaths. Meningococcal sepsis and other types of bacterial meningitis (Fig. 1.9) are probably the best recognized infections causing rapid demise in previously well children in developed countries. Infection should always be considered a possibility at autopsy requiring the consideration of taking of additional samples for histology

14

N.E.I. Langlois and R.W. Byard

Fig. 1.9  Purulent exudate adherent to the base of the brain in a case of fatal bacterial meningitis

and other tests. Cerebrospinal fluid (CSF) can be taken by posterior spinal puncture before the body is opened, or by anterior puncture after evisceration of the organs. If CSF cannot be obtained by these routes, a swab of the basal region of the brain immediately after the skull is opened is a reasonable alternative. This can be supplemented by obtaining a sterile sample of brain tissue immediately following reflection of the dura. Once the chest is opened, the pericardial sac can be incised to allow a sterile puncture of the right atrium of the heart to obtain heart blood for culture. Samples from other organs, such as the heart, lungs, and spleen, should be taken as early as possible into the postmortem examination, before there has been significant handling of the organs. The surfaces of can be sterilized by searing or by washing by alcohol. Isopropyl alcohol should be used rather than ethanol to avoid contamination if samples are also being taken for toxicology. Testing of the heart for enteroviruses can be supplemented by taking of small bowel content. The use of the polymerase chain reaction (PCR) to search for molecular evidence of infecting organisms supplements viral culture, but has the advantage that it can be applied to formalin-fixed paraffin-embedded tissues [163]. Microscopic examination of tissues can prove a vital component of this ­assessment as entities such as pneumonia in the very young may result in sudden death following no specific, or only minimal, clinical signs [1]. While there may be a large difference in weight between the two lungs, with the involved lung being

1  Sudden Natural Deaths in Infancy and Childhood

15

firm and airless, in certain cases the changes may only been apparent on histological examination. Microbiological studies may again be contributory with polymerase chain reaction (PCR) analysis of samples supplementing bacterial and viral culture for identification of infectious agents such as influenza [164]. Local infection of the upper airway may produce critical reduction in lumenal diameter and thus compromise oxygenation. This used to occur with acute ­epipiglottitis due to Hemophilus influenzae type B infection, but this has declined dramatically since the introduction of vaccinations in the early 1990s. Viral infections such as infectious mononucleosis due to Ebstein–Barr virus may cause lethal acute upper airway obstruction due to tonsillar enlargement [165]. Lethal obstruction may also occur if there is infection of the lingual tonsils [166]. Tonsillitis may also cause lethal hemorrhage [167]. Bacterial endocarditis may complicate ­congenital cardiac malformations and cause death from a variety of mechanisms [168]. Lemierre syndrome is a necrotizing infection of the head due to Fusobacterium necrophorum that causes septic thrombophlebitis of the internal jugular vein, meningitis and descending necrotizing mediastinitis [169]. Sudden death in childhood may be associated with infection such as parvovirus B19 [170] and rare infections include those due to hydatid disease where sudden death may result from cyst rupture with anaphylaxis or embolization of the contents [99, 171]. In some cases, the etiological agent of infectious syndromes may not be ­identified, such as in cases of hemorrhagic shock and encephalopathy. This condition was first described in 1983; the pathogenesis is unclear, but the condition affects previously healthy infants and children. In approximately half of the cases there is a short ­prodroma of respiratory infection or gastrointestinal upset followed by a rapid onset of shock, encephalopathy, hemorrhage, diarrhea, and oliguria. The pathological findings­are nonspecific. Features of disseminated intravascular coagulation may be present. Softening, edema, and infarction have been described in the brain [172].

Endocrine Endocrine causes of sudden death should be considered in the absence of a cause of death in other systems. Adrenocortical insufficiency may arise spontaneously, as part of congenital adrenal hypoplasia, in the context of sepsis, or as a result of ­exogenous steroid therapy, and result in death [173, 174]. Abnormal pigmentation of the gingival membranes and skin creases may be apparent in Addison’s disease [173]. Serum cortisol can be measured in these cases [175]. The vitreous humor from the eye may be used to obtain an indication of the antemortem electrolyte levels; however, after death, changes in electrolyte levels occur due to necrosis of retinal cells [176–178]. Poorly controlled diabetes mellitus may be a cause of sudden death in ­childhood, but the mechanism may not be clear, particularly when death occurs when the ­victim is unattended in bed [179, 180]. Testing of urine using a clinical “dipstick”

16

N.E.I. Langlois and R.W. Byard

is not a perfect tool for detecting unsuspected diabetes [181] and vitreous ­biochemical ­analysis should be performed. In addition to testing the vitreous for glucose, further relevant information may be obtained by testing the vitreous for the ketone b-hydroxbutyrate [182] and by measuring blood HbA1c levels (with elevated levels ­indicating the presence of diabetes mellitus [183–185]); as well as analyzing for insulin and c-peptide [186, 187]. Basal vacuolization of renal tubular epithelial cells may be seen in the kidneys following ketoacidosis [188]. After death, glucose levels in the blood and ­vitreous tend to fall as cells anerobically metabolize any sugars present [177]; hence, hypoglycemia cannot be reliably diagnosed after death [189, 190]. However, measurement of ­lactate has been proposed as a means of identifying antemortem hypoglycemia [191].

Metabolic A wide variety of metabolic disorders may result in sudden and unexpected death in infancy and childhood, the features of which have been described in detail elsewhere [192]. One of the most common is medium-chain acyl-CoA dehydrogenase deficiency where abnormal b-oxidation of fatty acids causes episodic hypoglycemia, lethargy, vomiting, seizures, coma, respiratory depression/apnea, and sudden death. MCAD deficiency has an autosomal recessive inheritance and the gene for MCAD has been fully characterized on chromosome 1p31. The estimated ­frequency is between one in 9,000–22,000 with a higher incidence among ­northern Europeans [1]. Reye’s syndrome [193] has now largely disappeared as a cause of death following viral infection due to the avoidance of the use of aspirin, and the more precise identification of specific disorders that had been included under this diagnostic umbrella.

Miscellaneous A variety of other conditions may be associated with sudden death, including ­connective tissue disorders such as Marfan and Ehlers–Danlos syndromes. Sudden death in Marfan syndrome may result from aortic dissection, and vessel rupture is a feature of Type IV Ehlers–Danlos syndrome associated with absence type III ­collagen. These events may manifest early in life, including during infancy [194, 195]. Infants with any one of the three common trisomies, 21, 18, and 13, are predisposed to early death; for example, children with Down syndrome are at higher risk of death from congenital cardiovascular abnormalities and infections, in addition to a wide range of other less common conditions [11]. Anaphylaxis may result in rapid death, sometimes with minimal findings at autopsy. Sampling of postmortem serum for tryptase and specific immunoglobulin E levels will be required to establish the diagnosis [196].

1  Sudden Natural Deaths in Infancy and Childhood

17

The “Negative” Autopsy Despite meticulous examination considering potential anatomical (gross and ­microscopic), biochemical, microbiological, and metabolic causes of death, no apparent cause of death may be found in around 4–10% of postmortem examinations [145, 197–200]. There has been recent interest in molecular causes of sudden death as an explanation in such cases. One particular group of such disorders is associated with prolongation of the QT interval. The QT interval is the duration of the phase from ventricular depolarization (onset of ventricular contraction) to the end of ventricular repolarization (resetting of the electrical potential in the cardiac myocytes) [201, 202]. The QT interval varies with heart rate, but using Bazett’s formula [201, 203], a corrected QT interval (QTc) can be calculated that is independent of the rate. An individual with a QTc greater than 440 ms may have a 2.3 times increased risk of sudden death compared to controls [204]. Research into the long QT syndrome has revealed a number of genetic defects in components of the ion channels in cardiac myocytes [202, 205]; ion channels are essential for the control of muscular depolarization and depolarization. Mutations in the potassium channel genes KCNQ1 (LQT1) and KCHN2 (LQT2) and the sodium channel gene SCH5A (LQT3) are the most common causes of the long QT ­syndrome [206, 207]. The phenotypes associated with mutations vary, such that LQT1 and LQT2 are associated with death at times of emotional or physical stress and exercise (particularly swimming with LQT1); loud noise may cause arrhythmias in LQT2, but LQT3 is associated with death during sleep (and is one cause of Brugada ­syndrome [205, 208]) [202, 207]. Although the heritable nature of long QT has been known for many years, several mutations are known and the implications to ­families of finding polymorphisms may be uncertain [209–211]. It has been suggested that genetic screening should be performed in all cases of sudden death with negative postmortem examinations [212]. However, in the future it may also be shown that molecular abnormalities provide an explanation as to why sudden death occurs in some patients with chronic disease states such as epilepsy [213], asthma [214], and diabetes [215–217], but not others.

Conclusion Given the wide range of disorders and conditions that may result in sudden and/or unexpected deaths in the young [1], autopsies must be comprehensive and involve careful review of clinical and family histories, full dissections with specialized ­evaluations where necessary, and use of ancillary testing, such as microbiologic, metabolic, and genetic screening; in some cases, a scene visit may be beneficial. The complex and rare nature of many of these diseases also means that early involvement of additional specialists such as medical geneticists and pediatric ­cardiologists, radiologists, and microbiologists may provide invaluable information

18

N.E.I. Langlois and R.W. Byard

and guidance in the assessment of such cases. As always, determining what an ­individual has died from, and not with, may be the ultimate challenge.

References 1. Byard RW (2010) Sudden death in the young. Cambridge University Press, Cambridge 2. Wisten A, Messner T (2005) Symptoms preceding sudden cardiac death in the young are common but often misinterpreted. Scand Cardiovasc J 39:143–149 3. Frazer M, Mirchandai H (1984) Commotio cordis, revisited. Am J Forensic Med Pathol 5:249–251 4. Estes NAM (1995) Sudden death in young athletes. N Engl J Med 333:380–381 5. Marshall DT, Gilbert JD, Byard RW (2008) The spectrum of findings in cases of sudden death due to blunt cardiac trauma – “Commotio Cordis”. Am J Forensic Med Pathol 29:1–4 6. Phillips MF, Harper PS (1997) Cardiac disease in myotonic dystrophy. Cardiovasc Res 33:13–22 7. Fishbein MC, Siegel RJ, Thompson CE, Hopkins LC (1993) Sudden death of a carrier of x-linked Emery-Dreifuss muscular dystrophy. Ann Int Med 119:900–905 8. Finsterer J, Stöllberger C (2003) The heart in human dystrophies. Cardiology 99:1–19 9. Groth WJ, Groh MR, Saha C, Kincaid JC, Simmonds Z, Ciafaloni E et  al (2008) Electrocardiographic abnormalities and sudden death in myotonic dystrophy type 1. N Engl J Med 358:2688–2697 10. Petković SM, Savić S, Simić MA (2008) Sudden death of a 17-year-old boy due to suspected Williams syndrome. Am J Forensic Med Pathol 29:83–85 11. Byard RW (2007) Forensic issues in Down syndrome fatalities. J Forensic Legal Med 14:475–481 12. Hirani R, Koszyca B, Byard RW (2008) Marfan syndrome and sudden death within a family – aetiologic, molecular and diagnostic issues at autopsy. J Forensic Legal Med 15:205–209 13. Schmidt G, Kallieris D (1982) Use of radiographs in the forensic autopsy. Forensic Sci Int 19:263–270 14. Eustace SJ, Nelson E (2004) Whole body magnetic resonance imaging. A valuable adjunct to clinical examination. Brit Med J 328:1387–1388 15. Griffiths PD, Paley MNJ, Whitby EH (2005) Post-mortem MRI as an adjunct to fetal or neonatal autopsy. Lancet 365:1271–1273 16. Jeffery A, Morgan B, Rutty GN (2007) The role of CT in forensic investigations. B Roy Col Pathol July:20–23 17. Thali MJ, Yen K, Schweitzer W, Vock P, Boesch C, Oxdoba C et al (2003) Virtopsy, a new imaging horizon in forensic pathology: virtual autopsy by postmortem multislice computed tomography (MSCT) and magnetic resonance imaging (MRI) – a feasibility study. J For Sci 48:386–403 18. Racette S, Sauvageau A (2007) Brain arteriovenous malformation and its implication in forensic pathology. J For Sci 52:189–191 19. Weller RO (1995) Subarachnoid haemorrhage and myths about saccular aneurysms. J Clin Pathol 48:1078–1081 20. Croft PR, Reichard RR (2009) Microscopic examination of grossly unremarkable pediatric dura mater. Am J Forensic Med Pathol 30:10–13 21. Nashef L, Walker F, Allen P, Sander JWAS, Shorvon SD, Fish DR (1996) Apnoea and bradycardia during epileptic seizures: relation to sudden death in epilepsy. J Neurol Neurosur Ps 60:297–300 22. Schwender LA, Troncoso JC (1986) Evaluation of sudden death in epilepsy. Am J Forensic Med Pathol 7:283–287

1  Sudden Natural Deaths in Infancy and Childhood

19

23. Lip GYH, Brodie MJ (1992) Sudden death in epilepsy: an avoidable outcome? J Roy Soc Med 85:609–611 24. O’Donoghue MF, Sander J (1997) The mortality associated with epilepsy, with particular reference to sudden unexpected death. Eplepsia 38(Suppl 11):S15–S19 25. Brown SW (1992) Sudden death and epilepsy. Seizure 1:71–73 26. Timmings PL (1993) Sudden unexpected death in epilepsy: a local audit. Seizure 2:287–290 27. Lear-Kaul KC, Coughlin L, Dobersen MJ (2005) Sudden unexpected death in epilepsy. A retrospective study. Am J Forensic Med Pathol 26:11–17 28. Thom M (2007) The autopsy in sudden unexpected adult death: Epilepsy. Curr Diagn Pathol 13:389–400 29. Schraeder PL, Lathers CM (1989) Paroxysmal autonomic dysfunction, epileptogenic activity and sudden death. Epilepsy Red 3:55–62 30. Oppenheimer S (1990) Cardiac dysfunction during seizures and the sudden epileptic death syndrome. J Roy Soc Med 83:134–136 31. Mameli O, Melis F, Giraudi D, Cualbu M, Mameli S, De Riu PL et al (1993) The brainstem cardioarrhythmogenic triggers and their possible role in sudden epileptic death. Epilepsy Red 15:171–178 32. Schraeder PL, Lathers CM (1983) Cardiac neural discharge and epileptogenic activity in the cat: an animal model for unexplained death. Life Sci 32:1371–1382 33. Rugg-Gunn FJ, Simister RJ, Squirrell M, Holdright DR, Duncan JS (2004) Cardiac arrhythmias in focal epilepsy: a prospective long-term study. Lancet 364:2212–2219 34. P-Codrea Tigaran S, Dalager-Pedersen S, Baandrup U, Dam M, Vesterby-Charles A (2005) Sudden unexpected death in epilepsy. Is death by seizures a cardiac disease? Am J Forensic Med Pathol 26:99–105 35. Malatinský J, Procházka M, Sámel M (1975) Respiratory failure as a seizure phenomenon. Brit J Anaesth 47:1019–1021 36. Shields LBE, Hunsaker D, Hunsaker JC, Parker JC (2002) Sudden unexpected death in epilepsy. Neuropathologic findings. Am J Forensic Med Pathol 23:307–314 37. Leestma JE, Kalelkar MB, Teas SS, Jay GW, Hughes JR (1984) Sudden unexpected death associated with seizures: analysis of 66 cases. Epilepsia 25:84–88 38. Byard RW, Bourne AJ, Hanieh A (1991) Sudden and unexpected death due to hemorrhage from occult central nervous system lesions. Pediatr Neurosurg 17:88–94 39. Gleckman A, Smith T (1998) Sudden unexpected death from primary posterior fossa tumours. Am J Forensic Med Pathol 19:303–308 40. Byard RW, Moore L (1993) Sudden and unexpected death in childhood due to a colloid cyst of the third ventricle. J Forensic Sci 38:210–213 41. Byard RW (1996) Mechanisms of sudden death and autopsy findings in patients with ArnoldChiari malformation and ventriculoatrial catheters. Am J Forensic Med Pathol 17:260–263 42. Byard RW, Koszyca B, Qiao M (2001) Unexpected childhood death due to a rare complication of ventriculoperitoneal shunting. Am J Forensic Med Pathol 22:207–210 43. Byard RW, Blumbergs PC, James RA (2003) Mechanisms of unexpected death in tuberous sclerosis. J For Sci 48:172–6 44. Byard RW, Schliebs J, Koszyca BA (2001) Osler-Weber-Rendu syndrome – pathological manifestations and autopsy considerations. J Forensic Sci 46:698–701 45. Byard RW (2006) Forensic issues and possible mechanisms of sudden death in Rett ­syndrome. J Clin Forensic Med 13:96–99 46. Gilbert JD, Scott G, Byard RW (2001) Septo-optic dysplasia and unexpected adult death – an autopsy approach. J Forensic Sci 46:913–915 47. Wick R, Byard RW (2006) Mechanisms of unexpected and/or sudden death in Lafora disease. Forensic Sci Int 163:144–147 48. Wick R, Scott G, Byard RW (2007) Mechanisms of unexpected death and autopsy findings in Leigh syndrome (subacute necrotizing encephalomyelopathy). J Clin Forensic Med 14:42–45 49. Byard RW (2007) Forensic considerations in cases of neurofibromatosis – an overview. J Forensic Sci 52:1164–1170

20

N.E.I. Langlois and R.W. Byard

50. Koszyca B, Moore L, Byard RW (1993) Lethal manifestations of neurofibromatosis type 1 in childhood. Pediatr Pathol 13:573–581 51. Bye MR, Sharma G, Konop R, Callahan C, Cataletto ME, Hilman B (2003) Laryngomalacia. emedicine topic1280:1–7 52. Rohde M, Banner J, Byard RW (2005) Congenital lesions associated with airway narrowing, respiratory distress and unexpected infant and early childhood death. Forensic Sci Med Pathol 1:91–96 53. Dada MA (1995) Laryngeal cyst and sudden death. Med Sci Law 35:72–74 54. Byard RW, Bourne AJ, Silver MM (1990) The association of lingual thyroglossal duct remnants with sudden death in infancy. Int J Pediatr Otolaryngol 20:107–112 55. Byard RW, Burrows PE, Izakawa T, Silver MM (1991) Diffuse infantile haemangiomatosis: clinicopathological features and management problems in five fatal cases. Eur J Pediatr 150:224–227 56. Byard RW (1996) Mechanisms of unexpected death in infants and young children following foreign body ingestion. J Forensic Sci 41:438–441 57. Byard RW, Jensen L (2007) Fatal asphyxial episodes in the very young – classification and diagnostic issues. Forensic Sci Med Pathol 3:177–181 58. Byard RW, Kennedy JD (1996) Diagnostic difficulties in cases of sudden death in infants with mandibular hypoplasia. Am J Forensic Med Pathol 17:255–9 59. Raven K, Reay DT, Harruff RC (1999) Artifactual injuries of the larynx produced by resuscitative intubation. Am J Forensic Med Pathol 20:31–36 60. Byramji A, Gilbert J, Byard R. Fatal retropharyngeal abscess – a possible marker of inflicted injury in infancy and early childhood. Forensic Sci Med Pathol 5:302–306 61. Byard RW, Gilbert JD. Potentially lethal complications of tracheostomy – autopsy considerations. Am J Forensic Med Pathol (in press) 62. Burton J, Rutty G (2003) The hospital autopsy. Oxford University Press, Oxford 63. Currie GP, Alluri R, Christie GL, Legge JS (2007) Pneumothorax: an update. Postgrad Med J 83:461–465 64. Cardel BS, Pearson RSB (1959) Death in asthmatics. Thorax 14:341–352 65. Champ CS, Byard RW (1994) Sudden death in asthma in childhood. Forensic Sci Int 66:117–127 66. Sidebotham HJ, Roche WR (2003) Asthma deaths; persistent and preventable mortality. Histopathology 43:105–117 67. Copeland AR (1986) Asthmatic deaths in the medical examiner’s population. Forensic Sci Int 31:7–12 68. Preston HV, Bowen DAL (1987) Asthma deaths: A review. Med Sci Law 27:89–94 69. Sur S, Crotty TB, Kephart GM, Hyma BA, Colby TV, Reed CE et al (1993) Sudden-onset fatal asthma. A distinct entity with few eosinophils and relatively more neutrophils in the airway submucosa? Am Rev Respir Dis 148:713–719 70. Sundararajan S, Ostojic NS, Rushton DI, Cox PM (2005) Diaphragmatic pathology: a cause of clinically unexplained death in the perinatal/paediatric age group. Med Sci Law 45:110–114 71. Basso C, Frescura C, Corrado D, Angelini A, Daliento L, Thiene G (1995) Congenital heart disease and sudden death in the young. Hum Pathol 26:1065–1072 72. Pelech AN, Neisch SR (2004) Sudden death in congenital heart disease. In: Berger S (ed) Pediatric Clinics of North America. Saunders, Philadelphia, pp 1257–1271 73. Råsten-Almqvist P, Rajs J (2004) Cardiovascular malformations and sudden death in infancy. Am J Forensic Med Pathol 25:134–140 74. Bharati S, Lev M (1986) Conduction system in cases of sudden death in congential heart disease many years after surgical correction. Chest 90:861–868 75. Oechslin EN, Harrion DA, Connelly MS, Webb GD, Siu SC (2000) Mode of death in adults with congenital heart disease. Am J Cardiol 86:1111–1116 76. Silka MJ, Hardy BG, Menashe VD, Morris CD (1998) A population-based prospective evaluation of risk of sudden cardiac death after operation for common congenital heart defects. J Am Col Cardiol 32:245–251

1  Sudden Natural Deaths in Infancy and Childhood

21

77. Byard RW, Gilbert JD (2005) Total anomalous pulmonary venous connection. Autopsy considerations. Forensic Sci Med Pathol 1:215–220 78. Byard RW, Moore L (1991) Total anomalous pulmonary venous drainage and sudden death in infancy. Forensic Sci Int 51:197–202 79. James CL, Keeling JW, Smith NM, Byard RW (1994) Total anomalous pulmonary venous drainage associated with fatal outcome in infancy and early childhood: an autopsy study of 52 cases. Pediatr Pathol 14:665–678 80. Frescura C, Basso C, Thiene G, Corrado D, Pennelli T, Angelini A et al (1998) Anomalous origin of coronary arteries and risk of sudden death: a study based on an autopsy population of congenital heart disease. Hum Pathol 29:689–695 81. Roberts WC (1986) Major anomalies of coronary arterial origin seen in adulthood. Am Heart J 111:941–963 82. Lauridson JR (1988) Sudden death and anomalous origin of the coronary arteries from the aorta. A case report and review. Am J Forensic Med Pathol 9:236–240 83. Taylor AJ, Rogan KM, Virmani R (1991) Sudden cardiac death associated with isolated congenital coronary artery anomalities. J Am Col Cardiol 1992:640–647 84. Byard RW, Jimenez CL, Carpenter BF, Cutz E, Smith CR (1991) Four unusual cases of ­sudden and unexpected cardiovascular death in infancy and childhood. Med Sci Law 31: 157–161 85. Byard RW, Smith NM, Bourne AJ (1991) Association of right coronary artery hypoplasia with sudden death in an eleven-year-old child. J Forensic Sci 36:1234–1239 86. Lipsett J, Byard RW, Carpenter BF, Jimenez CL, Bourne AJ (1991) Anomalous coronary arteries arising from the aorta associated with sudden death in infancy and early childhood. Arch Pathol Lab Med 115:770–773 87. Lipsett J, Cohle SD, Berry PJ, Russell G, Byard RW (1994) Anomalous coronary arteries: a multicentre pediatric autopsy study. Pediatr Pathol 14:287–300 88. Moore L, Byard RW (1992) Sudden and unexpected death in infancy associated with a single coronary artery. Pediatr Pathol 12:231–236 89. Byard RW (1996) Vascular causes of sudden death in infancy, childhood, and adolescence. Views and Reviews in Vascular Medicine and Biology. Cardiovasc Pathol 5:243–257 90. Stahl J, Santos LD, Byard RW (1995) Coronary artery thromboembolism and unexpected death in childhood and adolescence. J Forensic Sci 40:599–601 91. Kato H, Ichinose E, Yoshioka F, Takechi T, Matsunaga S, Rikitake N (1982) Fate of coronary artery aneurysm in kawasaki disease: serial coronary angiography and long-term follow-up study. Am J Cardiol 49:1758–1766 92. Rozin L, Koehler S, Shakir A, Ladham S, Wecht CH (2003) Kawasaki disease. A review of pathologic features of stage IV disease and two cases of sudden death among asymptomatic young adults. Am J Forensic Med Pathol 24:45–50 93. Zuccollo JM, Byard RW (2001) Test and Teach. (Sudden infant death due to Kawasaki ­disease). Pathology 33:235–238 94. Ropponen KM, Alafuzoff I (1999) A case of sudden death caused by fibromuscular dysplasia. J Clin Pathol 52:541–542 95. Byard RW (1996) Idiopathic arterial calcification and unexpected infant death. Pediatr Pathol Lab Med 16:895–994 96. Byard RW, Cutz E (1990) Sudden and unexpected death in infancy and childhood due to pulmonary thromboembolism. Arch Pathol Lab Med 114:142–144 97. Champ C, Byard RW (1994) Pulmonary embolism and unexpected death in infancy. J Paediatr Child Health 30:550–551 98. Moore L, Byard RW (1992) Fatal paradoxical embolism to the left carotid artery during ­partial resection of Wilms’ tumor. Pediatr Pathol 12:551–556 99. Byard RW, Bourne AJ (1991) Cardiac echinococcosis with fatal intracerebral embolism. Arch Dis Child 66:155–156 100. Fothergill DF, Bowen DAL, Mason JK (1979) Dissecting and atherosclerotic aneurysms: a survey of post-mortem examinations, 1968–77. Med Sci Law 19:253–260

22

N.E.I. Langlois and R.W. Byard

101. Byard RW (1994) Ventricular septal defect and sudden death in early childhood. J Paediatr Child Health 30:439–440 102. Byard RW, Bourne AJ, Adams PS (1990) Subarterial ventricular septal defect in an infant with sudden unexpected death: cause or coincidence? Am J Cardiovasc Pathol 3:333–336 103. Cohle SD, Graham MA, Dowling G, Pounder DJ (1988) Sudden death and left ventricular outflow disease. Pathol Annu 2:97–124 104. Hennersdorf MG, Strauer BE (2001) Arterial hypertension and cardiac arrhythmias. J Hypertens 19:167–177 105. Davies MJ, McKenna WJ (1995) Hypertrophic cardiomyopathy – pathology and pathogenesis. Histopathology 26:493–500 106. Hughes SE (2004) The pathology of hypertrophic cardiomyopathy. Histopathology 44:412–427 107. Marron BJ, Bonow RO, Cannon RO, Leon MB, Epstein SE (1987) Hypertrophic cardiomyopathy. Interrelations of clinical manifestations, pathophysiology, and therapy. (First of two parts). N Engl J Med 316:780–789 108. Dec GW, Fuster V (1995) Idiopathic dilated cardiomyopathy. N Engl J Med 331:1564–1575 109. Roberts WC, Siegel RJ, McManus BM (1987) Idiopathic dilated cardiomyopathy: analysis of 152 necropsy patients. Am J Cardiol 60:1340–1355 110. Steinherz L, Steinherz P (1991) Delayed cardiac toxicity from anthracycline therapy. Pediatrician 18:49–52 111. Tsuda N, Oka R, Kajino H, Kajino M, Okuno A (2000) Sudden death of a patient in complete remission after anthracycline therapy for acute lymphoblastic leukaemia. Pediatr Int 42:319–321 112. Stahl J, Couper RTL, Byard RW (1997) Oncocytic cardiomyopathy: a rare cause of unexpected early childhood death associated with fitting. Med Sci Law 37:84–87 113. Chin TK, Perloff JK, Williams RG, Jue K, Mohrmann R (1990) Isolated noncompaction of left ventricular myocardium. A study of eight cases. Circulation 82:507–513 114. Ichida F, Hamamichi Y, Miyawaki T, Ono Y, Kamiya T, Akagi T et al (1999) Clinical features of isolated noncompaction of the ventricular myocardium. J Am Col Cardiol 34:233–240 115. Murphy RT, Thaman R, Blanes JG, Ward D, Sevedalis E, Papra E et al (2005) Natural history and familial characteristics of isolated left ventricular non-compaction. Eur Heart J 26:187–192 116. Varnava AM (2001) Isolated left ventricular non-compaction: a distinct cardiomyopathy? Heart 86:599–600 117. Weiford BC, Subbarao VD, Mulhern KM (2004) Noncompaction of the ventricular myocardium. Circulation 109:2965–2971 118. Burke AP, Afzal MN, Barnett DS, Virmani R (1999) Sudden death after a cold drink. Am J Forensic Med Pathol 20:37–39 119. Gay JD, Guileyardo JM, Townsend-Parchman JK, Ross K (1996) Clinical and morpho­ logical features of lipomatous hypertrophy (“massive fatty deposits”) of the interatrial ­septum. Am J Forensic Med Pathol 17:43–8 120. Anders S, Said S, Schulz F, Püschel K (2007) Mitral valve prolapse syndrome as a cause of sudden death in young adults. Forensic Sci Int 171:127–130 121. Hayek E, Gring CN, Griffin BP (2005) Mitral valve prolapse. Lancet 365:507–518 122. Pocock W, Bosman CK, Chesler E, Barlow JB, Edwards JE (1984) Sudden death in primary mitral valve prolapse. Am Heart J 107:378–382 123. Scheurman EH (1989) Myxoid heart disease: a review with special emphasis on sudden ­cardiac death. Forensic Sci Int 40:203–210 124. Kligfield P, Levy D, Devereux RB, Savage DD (1987) Arrhythmias and sudden death in mitral valve prolapse. Am Heart J 113:1298–1307 125. Cohle SD, Graham MA, Sperry K, Dowling G (1989) Unexpected death as a result of ­infective endocarditis. J For Sci 34:1374–86 126. Josselson A, Bagnall JW, Virmani R (1984) Acute rheumatic carditis causing sudden death. Am J Forensic Med Pathol 5:151–154 127. Vevaina SC, Dastur DK, Manghani DK, Shah NA (1985) Changes in atrial biopsies in chronic rheumatic heart disease II: muscle fibre reaction. Path Res Pract 179:600–609

1  Sudden Natural Deaths in Infancy and Childhood

23

128. Skinner JR, Duflou JA, Semsarian C (2008) Reducing sudden death in young people in Australia and New Zealand: the TRAGADY initiative. Med J Australia 10:539–540 129. Gallagher PJ (1994) The investigation of cardiac death. In: Anthony PP, MacSween RNM (eds) Recent advances in histopathology. Longman, Edinburgh, pp 123–146 130. McRae TA, Chung MK, Asher CR (2001) Arrhythmogenic right ventricular cardiomyopathy: a cause of sudden death in young people. Clev Clin J Med 68:459–467 131. Burke AP, Farb A, Tashko G, Virmani R (1998) Arrhythmogenic right ventricular cardiomyopathy and fatty replacement of the right ventricular myocardium. Are they different diseases? Circulation 97:1571–1580 132. Calabrese F, Basso C, Carturan E, Valente M, Thiene G (2006) Arrhythmogenic right ventricular cardiomyopathy/dysplasia: is there a role for viruses? Cardiovasc Pathol 15: 11–17 133. Corrado D, Basso C, Thiene G (2000) Arrhythmogenic right ventricular cardiomyopathy: diagnosis, prognosis and treatment. Heart 83:588–595 134. D’Amati G, Leone O, Di Gioia CRT, Magelli C, Arpesella G, Grillo P et  al (2001) Arrhythmogenic right ventricular cardiomyopathy: clinicopathologic correlation based on a revised definition of pathologic patterns. Hum Pathol 32:1078–1086 135. Sheppard MN (1998) Arrhythmogenic right ventricular dysplasia and arrhythmogenic right ventricular cardiomyopathy: do these entities exist and are they the same disease? Curr Diagn Pathol 5:150–156 136. Tabib A, Liore R, Chalabreysse L, Meyronnet D, Miras A, Malicier D et  al (2003) Circumstances of death and gross and microscopic observations in a series of 200 cases of sudden death associated with arrhythmogenic right ventricular cardiomyopathy and/or dysplasia. Circulation 108:3000–3005 137. Fornes P, Ratel S, Lecomte D (1998) Pathology of arrhythmogenic right ventricular cardiomyopathy/dysplasia – an autopsy study of 20 forensic cases. J For Sci 43:777–783 138. Gemayel C, Pelliccia A, Thompson PD (2001) Arrhythmogenic right ventricular cardiomyopathy. J Am Col Cardiol 38:1773–1781 139. Michalodimitrakis M, Papadomanolakis A, Stiakakis J, Kanaki K (2002) Left side right ventricular dysplasia. Med Sci Law 42:313–317 140. Shrapnel M, Gilbert JD, Byard RW (2001) “Arrhythmogenic left ventricular dysplasia” and sudden death. Med Sci Law 41:159–162 141. Davies MJ (1981) Pathological view of sudden cardiac death. Brit Heart J 45:88–96 142. Aretz HT (1987) Myocarditis: The Dallas criteria. Hum Pathol 18:619–624 143. Smith NM, Bourne AJ, Clapton WC, Byard RW (1992) The spectrum of presentation at autopsy of myocarditis in infancy and childhood. Pathology 24:129–131 144. Corby C (1960) Isolated myocarditis as a cause of sudden death. Med Sci Law 1:23–31 145. Cohle SD, Sampson BA (2001) The negative autopsy: sudden cardiac death or other? Cardiovasc Pathol 10:219–222 146. Dettmeyer R, Kandolf R, Schmidt P, Schlamann M, Madea B (2001) Lympho-monocytic enteroviral myocarditis: traditional, immunohistological and molecular pathological methods for diagnosis in a case of suspected sudden infant death syndrome (SIDS). Forensic Sci Int 119:141–144 147. Baughman KL (2006) Diagnosis of myocarditis. Death of Dallas criteria. Circulation 113:593–595 148. Byard RW (2000) Sudden infant death, large intestinal volvulus and a duplication cyst of the terminal ileum. Am J Forensic Med Pathol 21:62–64 149. Byard RW, Simpson A (2001) Sudden death and intussusception in infancy and childhood autopsy considerations. Med Sci Law 41:41–45 150. Levin TL, Liebling MS, Ruzal-Shapiro C, Berdon WE, Stolar CJ (1995) Midgut malfixation in patients with congenital diaphragmatic hernia: what is the risk of midgut volvulus? Pediatr Radiol 25:259–261 151. Byard RW, Wick R (2008) Congenital mesenteric anomalies and unexpected death. Ped Develop Pathol 15:205–209

24

N.E.I. Langlois and R.W. Byard

152. Byard RW, Couper RTL, Cohle S (2001) Gastric distension, cerebral palsy and unexpected death. J Clin Forensic Med 8:81–85 153. Cox DE (1997) Intussusception: agonal phenomenon or cause of death? Med Sci Law 37:355–358 154. Byard RW, Donald TG, Rutty G (2008) Non-traumatic causes of perianal hemorrhage and excoriation in the young. Forensic Sci Med Pathol 4:159–163 155. Byard RW, Bohn DJ, Wilson G, Smith CR, Ein SH (1990) Unsuspected diaphragmatic hernia: a potential cause of sudden and unexpected death in infancy and early childhood. J Pediatr Surg 25:1166–1168 156. Byard RW, Bourne AJ, Cockington RA (1991) Fatal gastric perforation in a 4-year-old child with a late-presenting congenital diaphragmatic hernia. Pediatr Surg Int 6:44–46 157. Vandy FC, Landrum JE, Gerig NR, Prahlow JA (2008) Death due to late-presenting congenital diaphragmatic hernia in a 2-year-old child. Am J Forensic Med Pathol 29:75–79 158. Whybourne A, Zillman MA, Miliauskas J, Byard RW (2001) Sudden and unexpected infant death due to occult lymphoblastic leukaemia. J Clin Forensic Med 8:160 159. Graham JK, Mosunjac M, Hanzlick RL, Mosunjac M (2007) Sickle cell lung disease and sudden death. A retrospective/prospective study of 21 autospy cases and literature review. Am J Forensic Med Pathol 28:168–172 160. Wirthwein DP, Spotswood SD, Barnard SD, Barnard JJ, Prahlow JA (2001) Death due to microvascular occlusion in sickle-cell trait following physical exertion. J For Sci 46:399–401 161. Thogmartin JR, Wilson CI, Palma NA, Ignacio SS, Pellan WA (2009) Histological diagnosis of sickle cell trait. A blinded analysis. Am J Forensic Med Pathol 30:36–39 162. Manton N, Smith NM, Byard RW (2000) Unexpected childhood deaths due to hemolytic uremic syndrome. Am J Forensic Med Pathol 21:90–92 163. Bonin S, Petrera F, Niccolini B, Stanta G (2003) PCR analysis in archival postmortem tissues. J Clin Pathol Mol Pathol 56:184–186 164. Landi KK, Coleman AT (2008) Sudden death in toddlers caused by influenza B infection: a report of two cases and a review of the literature. J For Sci 53:213–215 165. Byard RW (2002) Unexpected death due to infectious mononucleosis. J Forensic Sci 47:202–204 166. Byard RW, Silver MM (1993) Sudden infant death and acute posterior lingual inflammation. Int J Pediatr Otorhinolaryngol 28:77–81 167. Byard RW (2008) Tonsillitis and sudden childhood death. J Forensic Legal Med 15:516–518 168. Byramji A, Gilbert JD, Byard R. Sudden death as a complication of bacterial endocarditis. Am J Forensic Med Pathol 32:140–142 169. Gilbert JD, Warner M, Byard RW (2009) Lemierre syndrome and unexpected death in childhood. Am J Forensic Med Pathol 16:478–481 170. Zack F, Kilngel K, Kandolf R, Wegener R (2995) Sudden cardiac death in a 5-year-old girl associated with parvovirus B19 infection. Forensic Sci Int 155:13–17 171. Byard RW (2009) An analysis of possible mechanisms of unexpected death occurring in hydatid disease (echinococcosis). J For Sci 54:919–922 172. Little D, Wilkins B (1997) Haemorrhagic shock and encephalopathy syndrome. An unusual cause of sudden death in children. Am J Forensic Med Pathol 18:79–83 173. Burke MP, Opeskin K (1999) Adrenocortical insufficiency. Am J Forensic Med Pathol 20:60–65 174. Tough SC, Green FH, Paul JE, Wigle DT, Butt JC (1996) Sudden death from asthma in 108 children and young adults. J Asthma 33:179–188 175. Clapper A, Nashelsky MB, Dailey M (2008) Evaluation of serum cortisol in the postmortem diagnosis of acute adrenal insufficiency. Am J Forensic Med Pathol 29:181–184 176. Balasooriya BAW, St Hill CA, Williams AR (1984) The biochemistry of vitreous humour. A comparative study of the potassium, sodium and urate concentrations in the eyes at identical time intervals after death. Forensic Sci Int 26:85–91 177. Coe JI (1972) Use of chemical determinants on vitreous humor in forensic pathology. J For Sci 17:541–546

1  Sudden Natural Deaths in Infancy and Childhood

25

1 78. Coe I (1993) Postmortem Chemistry Update. Am J Forensic Med Pathol 14:91–117 179. Irwin J, Cohle SD (1988) Sudden death due to diabetic ketoacidosis. Am J Forensic Med Pathol 9:119–121 180. Tattersall RB, Gill GV (1990) Unexplained deaths of type I diabetic patients. Diabetic Med 8:49–58 181. Wei OY, Teece S (2006) Urine dipsticks in screening for diabetes mellitus. Emerg Med J 23:138 182. Osuna E, Vivero G, Conejero J, Abenza JM, Martínez P, Luna A et al (2005) Postmortem vitreous humor b-hydroxybutyrate: its utility for the postmortem interpretation of diabetes mellitus. Forensic Sci Int 153:189–195 183. Goullé J-P, Lacroix C, Bouige D (2002) Glycated hemoglobin: a useful post-mortem reference marker in determining diabetes. Forensic Sci Int 128:44–49 184. Khuu HM, Robinson AC, Brissie RM, Konrad RJ (1999) Postmortem diagnosis of unsuspected diabetes mellitus established by determination of decedent’s hemoglobin A1c level. J For Sci 44:643–646 185. Winecker RE, Hammett-Stabler C, Chapman JF, Ropero-Miller JD (2002) HbA1c as a postmortem tool to identify glycemic control. J For Sci 47:1373–1379 186. Batalis NI, Prahlow JA (2004) Accidental insulin overdose. J For Sci 49:1117–1120 187. Haibach H, Shah JH (1987) Homicide by insulin administration. J For Sci 32:208–216 188. Kock KF, Vestergaard V (1994) Armanni-Ebstein lesions of the kidney: diagnostic of death in diabetic coma? Forensic Sci Int 67:169–174 189. Gama R, Teale JD, Marks V (2003) Best Practice No 173. Clinical and laboratory investigation of adult spontaneous hypoglycaemia. J Clin Pathol 56:641–646 190. Patel F (1994) Diabetic death bed: post-mortem determination of hypoglycemia. Med Sci Law 34:84–87 191. Kernbach-Wighton G, Sprung R, Püschel K (2001) On the diagnosis of hypoglycemia in car drivers - including a review of the literature. Forensic Sci Int 115:89–94 192. Gilbert-Barness E, Barness L (2000) Metabolic diseases: foundations of clinical management, genetics and pathology. Eaton press. Natick, Mass 193. Young TW (1992) Reye’s syndrome. A diagnosis occasionally made first at medicolegal autopsy. Am J Forensic Med Pathol 13:21–27 194. Byard RW (2006) Sudden death in Marfan syndrome. In: Tsokos M (ed) Forensic pathology reviews. Humana Press, pp 93–106 195. Byard RW, Keeley FW, Smith CR (1990) Type IV Ehlers-Danlos syndrome presenting as sudden infant death. Am J Clin Pathol 93:579–582 196. Riches KJ, Byard RW (2004) The detection of fatal anaphylaxis at autopsy – an overview. Scand J Forensic Sci 10:61–63 197. Langlois NEI (2006) The use of histology in 638 coronial post-mortem examinations of adults: an audit. Med Sci Law 46:310–320 198. Langlois NEI (2009) Sudden adult death: a review. Forensic Sci Med Pathol 5:210–232 199. Lawler W (1990) The negative coroner’s necropsy: a personal approach and consideration of difficulties. J Clin Pathol 43:977–980 200. Thomas AC, Knapman PA, Krikler DM, Davies MJ (1988) Community study of the causes of “natural” sudden death. Br Med J 297:1453–1456 201. Meek S, Morris F (2002) ABC of clinical electrocardiography. Introduction. II-Basic terminology. Br Med J 324:470–473 202. Morita H, Wu J, Zipes DP (2008) The QT syndromes: long and short. Lancet 372:750–763 203. Gupta A, Lawrence A, Krishnan K, Kavinsky CJ, Trohman RG (2007) Current concepts in the mechanisms and management of drug-induced QT prolongation and torsade de pointes. Am Heart J 153:891–899 204. Algra A, Tijssen JGP, Roelandt JRTC, Pool J, Lubsen J (1991) QTc prolongation measured by standard 12-lead electrocardiology is an independent risk factor for sudden death due to ­cardiac arrest. Circulation 83:1888–1894 205. Towbin JA (2004) Molecular genetic basis of sudden cardiac death. In: Berger S (ed) Pediatric clinics of North America. Saunders, Philadelphia, pp 1229–1253

26

N.E.I. Langlois and R.W. Byard

206. Priori S, Schwartz PJ, Napolitano C, Bloise R, Ronchetti E, Grillo M et  al (2003) Risk ­stratification in the long-QT syndrome. N Engl J Med 348:1866–1874 207. Roden DM (2008) Long-QT syndrome. N Engl J Med 358:169–176 208. Brugada P, Brugada J (1992) Right bundle branch block, persistent ST segment elevation and sudden cardiac death: a distinct clinical and electrocardiographic syndrome. A multicentre report. J Am Col Cardiol 20:1391–1396 209. Doolan A, Langlois NEI, Chiu C, Ingles J, Lind J, Semsarian C (2008) Postmortem molecular analysis of KCNQ1 and SCN5A genes in sudden unexplained death in young Australians. Int J Cardiol 23:138–141 210. Oliva A, Pascali VL, Hong K, Brugada R (2005) Molecular autopsy of sudden cardiac death (SCD). The challenge of forensic pathologist to the complexity of genomics. Am J Forensic Med Pathol 26:369–370 211. Priori SG, Napolitano C (2006) Molecular underpinning of “Good Luck”. Circulation 114:360–362 212. Di Paolo M, Luchini D, Bloise R, Priori SG (2004) Postmortem molecular analysis in victims of sudden unexplained death. Am J Forensic Med Pathol 25:182–184 213. Johnson JN, Hofman N, Haglund CM, Cascino GD, Wilde AAM, Ackerman MJ (2009) Identification of a possible pathogenic link between congenital long QT syndrome and epilepsy. Neurology 72:224–231 214. Rosero SZ, Zareba W, Moss AJ, Robinson JL, Ali RH, Locati EH et al (1999) Asthma and the risk of cardiac events in the long QT syndrome. Am J Cardiol 84:1406–1411 215. Start RD, Barber C, Kaschula ROC, Robinson RTCE (2007) The “dead in bed syndrome” – cause of sudden death in Type 1 diabetes mellitus. Histopathology 51:843–845 216. Tu E, Twigg SM, Duflou J, Semsarian C (2008) Causes of death in young Australians with type 1 diabetes: a review of coronial postmortems. Med J Aust 188:699–702 217. Veglio M, Chinaglia A, Cavallo-Perin P (2004) QT interval, cardiovascular risk factors and risk of sudden death in diabetes. J Endocrinol Invest 27:175–181

Chapter 2

Post-mortem Investigation of Sudden Unexpected Death in Infancy: Role of Autopsy in Classification of Death Martin A. Weber and Neil J. Sebire

Abstract  Sudden unexpected deaths in infancy (SUDI) represent the commonest group of post-neonatal childhood deaths. Pathologists in the UK are currently recommended to follow the “Kennedy protocol” when performing such autopsies. This suggested protocol is primarily based on practice from expert opinion and the approach to the post-mortem examination has changed little over recent decades. The identification of specific medical causes of death at autopsy in SUDI has slightly improved in recent years, but around two-thirds of cases remain unexplained, being classified as SIDS or SUDI according to local protocols and circumstances. Current protocols include the autopsy with macroscopic examination of organs, but in the majority of cases in which a cause of death is identified, the diagnosis is based on a combination of ancillary investigations including histological examination and microbiological findings, which are mandatory studies in these infant deaths. However, with increasing evidence regarding the relative frequency with which the various components of the autopsy provide information regarding the cause of death, and recognition that immunological responses and/or bacterial products may be of increasing importance, alternative and/or additional diagnostic techniques are required which may result in modified evidence-based autopsy protocols. The aim of this article is to review the current evidence for protocols of post-mortem investigations of SUDI, with particular emphasis on features which may distinguish natural from unnatural deaths, and to evaluate the approach to investigations which maximise the likelihood of identifying natural causes of death. The article will not discuss issues related to non-accidental or inflicted injury, which remain complex and beyond the scope of this review. Keywords  Sudden infant death • SIDS • Autopsy • Microbiology • Rib fractures • Pulmonary haemosiderin-laden macrophages N.J. Sebire (*) Department of Histopathology, Camelia Botnar Laboratories, Great Ormond Street Hospital, London WC1N 3JH, UK e-mail: [email protected] E.E. Turk (ed.), Forensic Pathology Reviews, Volume 6, Forensic Pathology Reviews 6, DOI 10.1007/978-1-61779-249-6_2, © Springer Science+Business Media, LLC 2011

27

28

M.A. Weber and N.J. Sebire

Terminology and Classification of Sudden Infant Deaths Sudden unexpected death in infancy (SUDI) of an apparently healthy child has been reported for thousands of years, and became commonly known as “cot-/ crib-death.” This term is potentially misleading, as it implies that death invariably occurs in the cot [1] and the term was superseded by the definition of sudden infant death ­syndrome (SIDS) in 1969 [2]. However, “cot-death” is still widely used, particularly by the lay public, sometimes synonymously with SIDS, while others apply it to any sudden unexpected infant death, even when a cause of death is determined. SIDS was initially defined as, “the sudden death of an infant or young child, which is unexpected by history, and in which a thorough post-mortem examination fails to demonstrate an adequate cause of death” [2], and in 1971 SIDS became a registrable cause of death in England and Wales [3]. Decades later, in 1989, the National Institute of Child Health and Human Development (NICHD) proposed further ­modification of the definition, restricting the term to infants less than 1 year of age, and emphasising the importance of a thorough death scene investigation and review of the clinical history [4]. This was subsequently further modified into “typical” (Category I SIDS) and “atypical” deaths (Category II SIDS), as well as a third ­category (Category III SIDS) intended solely for epidemiological purposes to accommodate countries where autopsies were not routinely performed [5, 6], although this failed to gain wide acceptance. In 2004, an international panel [7] proposed a further classification, widely referred to as the “San Diego definition,” defined as, “the sudden unexpected death of an infant less than 1 year of age, with onset of the fatal episode apparently occurring during sleep, that remains unexplained after a thorough investigation, including performance of a complete autopsy and review of the circumstances of death and the c­ linical history” (Table 2.1). This current definition restricts the diagnosis of SIDS to cases where death is associated with sleep and makes it compulsory to investigate the circumstances surrounding death [7]. It is noteworthy that according to this classification, co-sleeping and prone sleeping-associated deaths are to be classified as SIDS (either as Category I or Category II deaths depending on other findings) if there is no convincing ­evidence of accidental asphyxia based on the review of the death scene and/or autopsy findings, a contentious issue for many pathologists, there being an ongoing debate as to whether sudden infant deaths in the presence of such clearly established risk factors constitute SIDS, accidents or neglect [3, 9]. Several variables intrinsic to the definition also remain poorly defined. For example, what criteria constitute “sudden” and what exactly is meant by “unexpected?” Equally, there is variation in pathologists’ interpretation of the potential significance of pathological changes and whether these may represent an adequate explanation of death or whether the death would better be classified as SIDS [1, 10, 11]. Furthermore, a “complete autopsy” according to this protocol requires toxicology and vitreous chemistry, neither of which is routinely performed in many centres in the UK (vide infra). Finally, the term “SIDS” remains controversial in view of its

(continued)

•

• As above

Category IB SIDS: (classic features of SIDS present but incompletely documented)

One or more not performed: –  Radiology –  Microbiology –  Vitreous chemistry –  Metabolic studies –  Toxicology

Death unexplained by autopsy findings, including –  Radiology –  Microbiology –  Vitreous chemistry –  Metabolic studies –  Toxicology No lethal pathological findings Minor respiratory system inflammatory infiltrates are acceptable • Intrathoracic petechiae supportive but not obligatory or diagnostic • No evidence of unexplained trauma, abuse, neglect, or unintentional injury • No significant thymic stress*

• Death unexplained >21 days and <9 months of age by investigation of Normal clinical history various scenes where Term pregnancy (³37 weeks) incidents leading to Normal growth and normal death might have development occurred • No similar deaths of siblings, close • Safe sleeping genetic relatives (uncles, aunts, environment or first-degree cousins), or other • No evidence of infants in the custody of the same accidental death caregiver

• Not performed

• • •

• • • •

Category IA SIDS: (classic features of SIDS present and completely documented)

Table 2.1  The San Diego classification of SIDS (adapted from Krous et al. [7] and Bajanowski et al. [8]) Clinical history Circumstances of death Autopsy • Death unexplained by autopsy findings • Death unexplained General definition of • Sudden unexpected death after review of SIDS • <1 year of age circumstances of • Onset of fatal episode during sleep death • Death unexplained by clinical history

2  Post-mortem Investigation of Sudden Unexpected Death in Infancy… 29

*Significant thymic stress is defined as thymic weight <15 g and/or moderate or severe cortical lymphocyte depletion; occasional “starry sky” macrophages or minor cortical depletion are acceptable

• Alternative diagnoses of natural or unnatural conditions are equivocal

• Alternative diagnoses of natural or unnatural conditions are equivocal, or • Autopsy not performed

Autopsy

Unclassified sudden infant • Criteria for Category I or II SIDS death not met

Circumstances of death

• Mechanical asphyxia • Abnormal growth and development not thought to have contributed to death or suffocation caused • Marked inflammatory changes or abnormalities not by overlaying not sufficient to be unequivocal causes of death determined with certainty

Clinical history

• Age 0–21 days or 9–12 months Category II SIDS: (includes infant deaths • Similar death of siblings, close relatives, or other infants in the that meet Category I custody of the same caregiver criteria except for ³1 (that are not considered suspicious of the following) of infanticide or of a recognised genetic disorder) • Neonatal/perinatal complications (e.g. prematurity) that have resolved by the time of death

Table 2.1  (continued)

30 M.A. Weber and N.J. Sebire

2  Post-mortem Investigation of Sudden Unexpected Death in Infancy…

31

inclusion of the term “syndrome,” with some arguing that SIDS merely represents a “dustbin diagnosis” for any sudden infant death which remains otherwise ­unexplained [12]; however, even ardent supporters for the continued use of the term “syndrome” agree that SIDS is unlikely to constitute a single cause of death, but is likely to ­represent a heterogeneous group of sudden infant deaths that currently remain unexplained but which share a common mechanism(s) of death [7]. The term sudden unexpected death in infancy (SUDI), in contrast, was adopted by the CESDI SUDI study [1] and is the preferred term in the UK for the presentation of all sudden and unexpected deaths in the first year of life, including non-­ natural sudden deaths due to accidents and inflicted injury, although the term is usually restricted to infants aged 7 days to 1 year of age [1]. The Kennedy report [13] also uses the acronym SUDI, since all cases are initially investigated in a ­similar manner. If no cause of death is identified following the autopsy, the Kennedy guidelines recommend that deaths are classified as SIDS, or simply as SUDI, depending on the results of the review of the clinical history and examination of the circumstances of death; however, we prefer the term “unexplained SUDI” as a ­pathological classification of death if the latter remains unexplained, as there is little agreement amongst pathologists, despite published guidelines [7, 8, 13], on classifying otherwise unexplained SUDI as SIDS, particularly in the presence of ­co-sleeping [3]. Furthermore, the term “SUDI” merely refers to the presentation of death, which can be divided into “explained SUDI” and “unexplained SUDI” ­following the autops and review of the clinical history and death scene; to use “SUDI” both as presentation of death and cause of death if the latter remains undetermined, is confusing and potentially ­misleading. In contrast, the term “unascertained” is generally used when the pathologist identifies features in the clinical history or at autopsy that raise the suspicion of possible inflicted injury but which are insufficient to account for the death [14]. It is not recommended that the term “unascertained” be used synonymously with SIDS or unexplained SUDI, as the ­latter terms imply a natural cause of death, whilst the term “unascertained” includes both natural and non-natural deaths, and as a consequence, affected families may be stigmatised by designating an otherwise non-suspicious but unexplained sudden death as “unascertained” [3, 13].

Risk Factors for Unexplained SUDI/SIDS Unexplained SUDI/SIDS affects around 0.5–1/1,000 infants, with recent decline in rates, predominantly attributed to modifications based on removal of risk factors, mainly placing babies supine for sleep, with little impact from other behaviour modification strategies such as reducing smoking during pregnancy or promoting breastfeeding (Fig. 2.1) [12]. Many risk factors for unexplained SUDI/SIDS have been identified by epidemiological studies, including associations with young maternal age, low social class, high parity, multiple pregnancy, intrauterine growth restriction, prematurity

32

M.A. Weber and N.J. Sebire 3

SIDS Rate (per 1000 Live Births)

England and Wales Australia

2.5

Netherlands Norway

2

Switzerland United States of America

1.5

1

0.5

0 1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

Fig. 2.1  SIDS Rates by Country, 1987–1997 (All data taken from Byard and Krous [12], red c­ ircles indicate the year the risk reduction campaign was launched in each country)

and maternal smoking [1, 15–20]. The most important – and readily modifiable – behavioural factor appears to be prone sleeping, and to a lesser extent, side sleeping [21–25], such that public health campaigns were introduced in many countries to promote supine sleeping, leading to a dramatic decline in the incidence of SIDS (Fig. 2.1). Currently, co-sleeping or bed-sharing remains a controversial risk factor for SUDI, but most studies demonstrate a positive association, especially when ­co-sleeping during the first 6 months of life, if the infant was born prematurely (<37 weeks gestation) or with low birth weight (<2.5 kg), or if either parent is a smoker, has been drinking alcohol, taken drugs or other medication, or is excessively tired, with co-sleeping with an infant on an armchair or sofa being especially high risk [1, 26–31]. Sharing a sofa has been shown to carry a very high risk of SIDS: in the CESDI study [1], co-sleeping on a sofa carried an almost 23-fold greater risk of SIDS than infants not co-sleeping with an adult. In a recent systematic review, Blair et al. [32] report a significantly increased risk of SIDS in infants whose heads were covered by bed clothes (adjusted OR 17). The mechanism of death in these and co-sleeping infants remains unclear; theories include hypoxia, accumulation of carbon dioxide, hyperthermia and partial asphyxiation, and there continues to be uncertainty about how such cases should be classified, most ­pathologists in the UK preferring not to use SIDS in these circumstances. Conversely, several studies have shown that the use of pacifiers (soothers or dummies) significantly reduces the risk of SIDS [20], a meta-analysis demonstrating a pooled ­multi-variate OR of around 0.4 for pacifier use during the last sleep [33]. Interestingly, many risk factors are shared between cases of explained and unexplained SUDI, including preterm birth, young maternal age, single mothers, low socioeconomic

2  Post-mortem Investigation of Sudden Unexpected Death in Infancy…

33

status, maternal smoking, and apparent mild illness in the preceding 24 h [34, 35]. However, explained SUDI deaths are generally younger or older than SIDS infants, with more explained SUDI deaths occurring in the first month of life and after 6 months of age. The current conceptual framework linking the interactions between established epidemiological risk factors and their biological, genetic and other associations is presented as the “triple-risk” hypothesis comprising an intrinsically vulnerable infant, at a critical developmental period, and exposure to exogenous stressors [12, 36].

Role of Infection in SUDI Some SUDI deaths clearly demonstrate evidence of definite acute bacterial infection, such as cases with pneumonia or meningitis. However, increasing evidence also suggests that other mechanisms may implicate bacteria and/or bacterial products, or an associated abnormal host response, in some cases of currently unexplained SUDI/ SIDS, although no proven mechanism has yet been established. The so-called common bacterial toxin hypothesis [37–39] postulates that some SIDS may be caused by bacterial toxins, most likely derived from upper respiratory tract organisms, present in young infants when maternal IgG concentrations are dwindling prior to maturity of the infants’ immune system, with some toxins acting as superantigens to cause massive release of cytokines with resulting toxic shock-like syndrome or septic shock, and death [38, 40]. Alternatively, it has been speculated that toxins may act directly on neural or myocyte membranes to induce sudden death [38]. It has been reported that asymptomatic nasopharyngeal carriage of Staphylococcus aureus in infants is more common within the first 3 months of life than in older infants [41], and significantly more common in SIDS infants £3 months old than in age-matched healthy controls [42]. Furthermore, it has been demonstrated that prone sleeping in the presence of an upper respiratory tract infection is associated with significantly increased bacterial carriage, including increased colonisation by staphylococci [41], and two recent retrospective studies have shown a significantly increased prevalence of S. aureus on post-mortem cultures of lung, blood and/or spleen in unexplained SUDI/SIDS compared to explained SUDI due to non-infective causes [43, 44]. The host response to such bacterial products may be further altered according to specific genetic polymorphisms resulting in abnormal cytokine responses [45]. Many other theories regarding the pathogenesis of unexplained SUDI/SIDS have been postulated, including exposure to immunological reactions such as hypersensitivity and/or anaphylaxis [46], impaired autonomic regulation with impaired arousal and ventilatory responses [20, 47–50], cardiac conduction system abnormalities [51, 52], and a range of neuropathological aberrations, mainly related to brainstem abnormalities, such as hypoplasia of brainstem nuclei [53, 54] and serotonergic aberrations [55–57], and even epilepsy [58], all of which are beyond the scope of this article, and at present are of little direct relevance to the pathologist, who is required to provide a probable cause of death for Her Majesty’s (HM) Coroner.

34

M.A. Weber and N.J. Sebire

The SUDI Autopsy Protocol and Determination of Cause of Death Current autopsy guidelines are primarily based on the “Kennedy Report” [13], which advocates a multi-agency approach to the investigation of all SUDI. The determination of a cause of death depends not only on the findings of the post-mortem examination, but includes a comprehensive review of the circumstances of death, including a home visit, and is dependent on the interacting roles of HM Coroner, the police, community paediatricians and other agencies such as social services. In this review, we shall focus specifically on the autopsy; according to current guidelines, the autopsy should be performed by a paediatric pathologist with training and expertise in the investigation of infant death. The autopsy in this setting not only ­comprises an external examination and systematic prosection, but includes a range of ancillary investigations including post-mortem radiology, histological examination, bacteriological and virological investigations, metabolic analyses for fatty acid oxidation defects, and other, more targeted and specific tests such as toxicology and biochemical assays in selected cases. Both the Kennedy autopsy protocol [13] and published international guidelines [8] recommend a variety of different investigations, although many are not evidence based, and the relative importance of these in determining cause of death in SUDI varies greatly. In a large study at one paediatric tertiary centre in which a total of 1,516 paediatric post-mortem examinations were performed over a 10-year period, there were 546 autopsies of infants aged 7–365 days presenting as SUDI [59]. Of these, in 202 (37%) an identifiable cause of death was present following post-mortem examination (“explained SUDI”), whilst the majority (344 cases; 63%) remained “unexplained SUDI.” Of the 202 explained deaths, the majority (around 60%) were due to an infective process, most commonly pneumonia (20%), whilst non-infective causes included congenital malformations (7%), and deaths due to accidental (6%) and non-accidental (10%) injuries (Fig. 2.2). Of the infective deaths, most were due to bacterial infections (80%), with the remainder presumed viral. Importantly, the component of the post-mortem examination which primarily determined the cause of death was the histological examination in 92 (46% of explained SUDI, 17% of all SUDI), the macroscopic examination at the time of autopsy in 61 (30% of explained SUDI, 11% of all 546 SUDI), microbiological examination in 38 (19% of explained SUDI, 7% of all 546 SUDI) and clinical history/death scene in 10 (5% of explained SUDI, 2% of all 546 SUDI; Fig.  2.3). Radiology showed fractures in around 6% of cases, but, whilst the findings contributed to the final diagnosis in many, the presence of fractures detected radiologically did not primarily determine the cause of death in any case. The majority of infection-related diagnoses were made primarily on histological examination, although microbiological analyses accounted for around 20% of the diagnoses which would have been missed had routine microbiological sampling not been performed. However, several investigations suggested in the current autopsy protocol appear less useful as a routine investigation. For example, tandem mass spectrometry of blood and/or bile did not reveal any deaths due to fatty acid ­oxidation

2  Post-mortem Investigation of Sudden Unexpected Death in Infancy…

35

Number of Cases 0

20

40

60

80

100

120

Infections

Congenital Malformations

Other Natural Deaths

Respiratory System

Accidents

Cardiovascular System Central Nervous System Lymphoreticular System Gastrointestinal System

Non-Accidental Injury

Urological System Endocrine System

Fig. 2.2  Causes of death in explained SUDI. Of explained deaths, the majority (60%) were due to an infective process, most commonly pneumonia (22%), whilst the commoner non-infective causes of death included congenital malformations (7%), and deaths due to accidental (6%) and nonaccidental (10%) injuries (adapted from Weber et al. [59]) Toxicology <1%

Clinical History 5%

Microbiology 19%

Macroscopic Examination 30%

Histology 46%

Fig. 2.3  Component of the post-mortem examination which primarily determined the final cause of death: histological examination in 46% of explained SUDI, 17% of all SUDI, macroscopic examination in 30% of explained SUDI, 11% of all SUDI, microbiological investigations in 19% of explained SUDI, 7% of all SUDI and the clinical history/circumstances of death in 5% of explained SUDI, 2% of all SUDI. Toxicological analyses were performed in only selected cases and revealed the cause of death in only one infant due to accidental heroin poisoning (adapted from Weber et al. [59])

disorders in this series of infants aged 1 week to 1 year, although previous studies have reported such metabolic diseases as causes of death in this patient group. Combining the data from previous studies, deaths due to possible metabolic disorders represent only 0.3% of all SUDI deaths [1, 59–62] and in such cases, it is

36

M.A. Weber and N.J. Sebire

expected that there would be abnormal fat accumulation in liver, renal tubules and muscle, and hence histological screening with retention of additional samples may be a more cost efficient protocol rather than routine tandem mass spectrometric analysis. In addition, at present, post-mortem microbiology interpretation may be problematic in individual cases, being based on empirical guidelines only [43] (vide infra). Further studies may allow identification of the abnormal host immune response in such cases rather than attempting to isolate the organism responsible, which will allow separation of cases with incidental colonisation from cases with death due to an inflammatory response to infection.

Sudden Unexpected Early Neonatal Death Neonatal deaths occur in around 1 in 250 births, most being expected complications of prenatally or clinically detected congenital abnormalities, or complications of prematurity [63]. However, sudden unexpected neonatal deaths of otherwise ­clinically healthy infants in the first week of life, analogous to SUDI in older infants, may also occur; such deaths may be referred to as sudden unexpected early neonatal deaths (SUEND), although the clinical usefulness of separating SUEND from SUDI remains undetermined. However, in contrast to SUDI, in which around two-thirds of deaths remain unexplained after a full investigation, in a series of 55 SUEND cases, representing about 30% of early neonatal deaths undergoing autopsy, and 6% of all infant autopsies in that series, almost two-thirds (60%) were medically explained following post-mortem examination, with only one-third remaining unexplained, similar to SIDS in older infants [64]. Interestingly, 70% of cases who died during sleep were co-sleeping with parents, including almost all (90%) of the unexplained deaths that occurred during sleep. Explained deaths were due to previously undiagnosed congenital abnormalities (40%), mainly (>90%) structural congenital heart disease. In addition, there were three deaths (5%) from unsuspected metabolic disease, including two infants with medium chain acyl-CoA dehydrogenase deficiencies and one case of carnitine acylcarnitine translocase deficiency; 15% of deaths were due to clinically unsuspected infections, with no accidental or non-accidental injury-associated deaths in this age group. Therefore, a significantly greater proportion of SUEND were explained following post-mortem examination compared to SUDI cases, and a significantly greater proportion of these were due to congenital abnormalities and metabolic diseases, with less infection-related or traumatic deaths (Fig. 2.4).

Interpretation of Rib Fractures in SUDI Autopsies Rib fractures are a well-recognised component of non-accidental injury (NAI) in infancy, usually being multiple, of varying ages and associated with other features of physical abuse [65]. The detection and confirmation of rib fractures is therefore

2  Post-mortem Investigation of Sudden Unexpected Death in Infancy…

37

60% SUEND (< 7 Days) SUDI (7 - 365 Days)

Explained Deaths

50%

40%

30%

20%

10%

0% ns

Infectio

s

rmalitie

al Abno

nit Conge

rders

lic Diso

Metabo

Fig. 2.4  Compared to SUDI, there were significantly less SUEND deaths due to infections, with significantly more SUEND deaths due to congenital abnormalities and metabolic disease (adapted from Weber et al. [64])

a mandatory component of the SUDI post-mortem examination, including by radiological examination, by macroscopic rib inspection and by histological sampling of apparent fractures for confirmation and dating. Recent data from an autopsy series in which ribs were routinely examined by stripping of the parietal pleura and macroscopic inspection of individual ribs, preceded by routine post-mortem radiology reported by specialist paediatric radiologists with expertise in NAI, rib fractures were detected in 4% of all SUDI autopsies [66]. These included healing fractures in 15 infants (3% of all SUDI; 63% of those with rib fractures) and fresh fractures with no histological evidence of a surrounding tissue reaction in 9 infants (2% of all SUDI; 37% of those with rib fractures; Fig. 2.5). Of the 15 infants with healing fractures, 10 (67% of healing fractures) were associated with additional features suggestive of NAI. In contrast, of the nine infants with exclusive fresh rib fractures, seven (<2% of all SUDI; 29% of those with rib fractures; 78% of those with fresh rib fractures) ­demonstrated no other injuries and no additional features suspicious of NAI or major trauma, suggesting that these fractures were caused by

38

M.A. Weber and N.J. Sebire

Fig. 2.5  Rib fractures in SUDI. (a) Macroscopic appearances of healing rib fractures in an infant with suspected NAI, with posterior healing rib fractures showing prominent callus formation (arrow). (b) (H&E) and (c) (Masson Trichrome). Photomicrographs of fresh rib fractures, demonstrating the fracture line (arrows) associated with minor fresh haemorrhage only. There is no surrounding inflammatory reaction or periosteal reaction

resuscitation-associated trauma. These apparent resuscitation-related fresh rib fractures were multiple, all involving the fourth and fifth ribs, and almost exclusively limited to the third, fourth, fifth, and/or sixth ribs, and all were located anterolaterally, compared to predominantly posterior fractures in cases with other features of NAI. No resuscitation-related fractures involved the costochondral junction, these being found only in cases with other injuries ­suggestive of NAI. Interestingly, healing rib fractures were identified on routine post-mortem radiological skeletal survey in >90% of cases, whereas only 20% of fresh fractures were identified radiologically. Pooled data from previous series suggest that about two-thirds of rib fractures are associated with NAI [67–71], with CPR-related fresh rib fractures occurring in about 1 in 300 episodes, using varying methods of ascertainment [72]. The frequency of detection of fresh fractures is greater in autopsy series in which specific stripping of the pleura and individual rib examination is carried out.

2  Post-mortem Investigation of Sudden Unexpected Death in Infancy…

39

Significance of Pulmonary Haemosiderin-Laden Macrophages in SUDI Haemosiderin-laden macrophages (HLMs) in histological sections of lung ­represent evidence of previous pulmonary haemorrhage [73]. Pulmonary haemorrhage (Fig. 2.6a) has been reported in association with accidental and inflicted suffocation, although the sensitivity, specificity and quantification of intra-alveolar haemorrhage in this setting remain controversial [73–78]. It has been suggested that the presence of alveolar HLMs in infant lungs at post-mortem examination may be an indicator of NAI [79–81]. In a large series of almost 600 sudden infant deaths (comprising both SUEND and SUDI cases), 5% demonstrated alveolar HLMs on Perls’ staining, often more prominent in the peripheral portions of the pulmonary lobule [82] (Fig.  2.6b). Around one-third with HLMs demonstrated additional features suggestive of NAI, about a third had a history of natural disease to explain their presence, and the remaining third were otherwise unexplained SUDI with no other significant post-mortem or clinical findings and no significant previous history or suspicious circumstances. Alveolar HLMs were almost ten times more frequent in those infant deaths associated with other features of NAI compared to those without other suspicious features, the effect remaining significant even when those with healing rib fractures were excluded. Thus, although the majority of cases with alveolar HLMs appear to be associated with previous natural disease, the presence of alveolar HLMs remains a potentially useful marker of NAI, although in isolation its interpretation is not diagnostic of previous asphyxia. Nevertheless, the presence of otherwise unexplained alveolar HLMs at autopsy should always prompt a careful exclusion of inflicted injury. In contrast, this association does not apply to interstitial HLMs (defined as HLMs limited to the interlobular septa, bronchovascular bundles and/or subpleural connective tissue), which are more prevalent in infants with increased birth weight and gestational age, and younger age at death, thought

Fig. 2.6  Intra-alveolar haemorrhage and alveolar haemosiderin-laden macrophages (HLMs) in the lungs in SUDI. (a) (H&E). Fresh intra-alveolar haemorrhage. (b) (Perls’ stain). Alveolar HLMs indicate previous haemorrhage; they can be detected within a few days after the haemorrhage occurred but may remain in the lungs for a long and variable period

40

M.A. Weber and N.J. Sebire

to represent the consequence of pulmonary haemorrhage caused during labour, ­particularly in larger infants of greater gestational age, the HLMs then being ­gradually cleared during early infancy [83].

Interpretation of Post-mortem Microbiology Whilst post-mortem microbiological sampling is clearly important and may provide the cause of death in some cases of SUDI [1, 59, 62, 84–86], in practice, interpretation of findings is difficult because of potential issues of contamination, post-­mortem translocation and overgrowth, as well as possible agonal spread [87] and distinguishing the clinical significance of positive findings from various sites. The role of infection contributing to the mechanism of death in some cases of otherwise ­unexplained SUDI is supported by the findings of a recent study in which routine ­post-mortem microbial isolates from cases of SUDI were classified as non-pathogens, group 1 pathogens (organisms usually associated with an identifiable focus of infection) and group 2 pathogens (organisms known to cause septicaemia without an obvious focus of infection: groups A and B beta-haemolytic streptococcus, pneumococcus, meningococcus, Escherichia coli and S. aureus) [43]. Of the 507 SUDI cases, including 379 unexplained SUDI, 72 non-infective explained SUDI and 56 explained SUDI deaths due to histologically confirmed bacterial infection (such as pneumonia or meningitis), including data on >2,000 samples with a median death to post-mortem interval of 3 days, around 30% were sterile, whilst the majority of cultures were positive for organisms. However, significantly more group 2 pathogens were isolated from the unexplained SUDI compared to the non-infective deaths; group 2 pathogens were detected in 50% of the unexplained SUDI group compared to only around 25% of the non-infective SUDI group. Specific organisms in this group most commonly detected included S. aureus and E. coli, predominantly from the lung. These data suggest that 10–35% of otherwise unexplained SUDI deaths in whom there is no histologically identifiable focus of infection at autopsy are related to infection with group 2 pathogens. However, interpretation in individual cases remains difficult, and discussion with a microbiologist is recommended in selected cases, particularly if there is no histological evidence of infection. Until further evidence becomes available, as a general rule, isolating the same pathogenic organism from multiple sample sites at post-mortem is likely to be more significant than isolating a pathogenic organism from a single culture site only; furthermore, it has been suggested that a pure growth or single isolate is more likely to be clinically significant than a mixed growth of organisms [87], although at present this is based on empirical grounds. The likelihood of clinical importance of a positive culture result of a recognised pathogen may be further modified by review of the clinical history, since symptoms and signs of a developing infection may have been present but unrecognised by carers prior to the death. Other markers of sepsis in life, such as white cell counts or CRP, which are not currently used in standard autopsy practice, may also prove to be helpful in this context, and may help to

2  Post-mortem Investigation of Sudden Unexpected Death in Infancy…

41

change the likelihood of a positive culture result from “possible contributor” to “probable contributor” to the cause of death [88].

Role of Neuropathology in SUDI Autopsies Formal examination of the central nervous system is an integral part of the SUDI autopsy investigation, since in a small proportion of cases a definite cause of death can be identified, such as meningitis or congenital malformation [1, 59]. Furthermore, and perhaps more importantly, such examination is required for the exclusion of traumatic brain injury, such as cerebral oedema, diffuse axonal injury, and subdural or retinal haemorrhage [89]. However, controversy exists surrounding even the significance of these findings in this clinical setting [90, 91]. Furthermore, a large volume of literature reports on neuropathological findings in SUDI, much of which relates to subtle abnormalities particularly of the brainstem, which may be relevant to the underlying pathogenesis of SUDI [53–57]. These specific issues regarding formal detailed neuropathological examination are therefore beyond the scope of this review.

Role of Toxicology in Routine SUDI Autopsies The Kennedy protocol [13] does not make toxicological analysis mandatory in every case, and most centres in the UK would not routinely perform such investigations in otherwise non-suspicious deaths in whom there is no history of parental drug use [1, 59]. There is no doubt that toxicological analysis may reveal a cause of death in a small proportion of cases, ranging from less than 1 to 4% of sudden infant deaths [59, 86, 92, 93]; in one study of 117 SUDI, the finding of possible methadone toxicity had not been expected from the clinical history available to the pathologist prior to the post-mortem examination in three cases (3%) [93]. It is therefore ­possible that a small number of deaths due to accidental or intentional poisoning are missed if toxicology is not routinely performed, but its role in unselected SUDI cases currently remains undetermined.

Towards an Evidence-Based Protocol The optimal autopsy protocol remains empirical at present, but increasing evidence is accumulating to guide the development of modifications to this approach. Full autopsy with histological sampling and multiple site bacteriology is required in all cases. Post-mortem radiology and examination of ribs are also mandatory in every case. Issues regarding the extent of histological sampling, and from which organs,

42

M.A. Weber and N.J. Sebire

remains uncertain and non-evidence based. The brain must be examined if no cause of death is identified, but the specific role of formal neuropathological examination as opposed to limited sampling for determining cause of death on behalf of HM Coroner remains unclear. Toxicological investigation clearly plays a role, but its usefulness in unselected SUDI cases remains undetermined, and, as outlined above, the interpretation of post-mortem microbiological cultures is also currently based on empirical principles. Future studies are required to address these issues in order to provide a true evidence base for the autopsy protocol in SUDI. Optimal pathological classification of these deaths, too, remains unclear, “unexplained SUDI” being most appropriate immediately post-autopsy if no cause is identified, with a case conference or HM Coroner decision on the final classification of death ­following a detailed and comprehensive review of the circumstances of death and to account for issues such as co-sleeping-associated deaths.

References 1. Fleming P, Blair P, Bacon C, Berry J (eds) (2000) Sudden unexpected deaths in infancy. The CESDI SUDI Studies 1993–1996. The Stationary Office, London 2. Beckwith JB (1970) Discussion of terminology and definition of the sudden infant death syndrome. In: Bergman AB, Beckwith JB, Ray CG (eds) Sudden infant death syndrome: proceedings of the second international conference on the causes of sudden death in infants. University of Washington Press, Seattle, pp 14–22 3. Limerick SR, Bacon CJ (2004) Terminology used by pathologists in reporting on sudden infant deaths. J Clin Pathol 57:309–311 4. Willinger M, Hoffman HJ, Hartford RB (1994) Infant sleep position and risk for sudden infant death syndrome: report of meeting held January 13 and 14, 1994, National Institutes of Health, Bethesda, MD. Pediatrics 93:814–819 5. Beckwith JB (1993) A proposed new definition of sudden infant death syndrome. In: Walker AM, McMillen C (eds) Second SIDS International Conference. Perinatology Press, Ithaca, pp 431–434 6. Beckwith JB (2003) Defining the sudden infant death syndrome. Arch Pediatr Adolesc Med 157:286–290 7. Krous HF, Beckwith JB, Byard RW, Rognum TO, Bajanowski T, Corey T, Cutz E, Hanzlick R, Keens TG, Mitchell EA (2004) Sudden infant death syndrome and unclassified sudden infant deaths: a definitional and diagnostic approach. Pediatrics 114:234–238 8. Bajanowski T, Vege A, Byard RW, Krous HF, Arnestad M, Bachs L et al (2007) Sudden infant death syndrome (SIDS) – standardised investigations and classification: recommendations. Forensic Sci Int 165:129–143 9. Bacon C (2008) A timely debate. Rapid Response to Gornall, 2008. Br Med J, http://www.bmj. com/cgi/eletters/336/7639/302#193852. 10. Bacon CJ (2000) Debate on cot death. Standard system for postmortem examination and certification needs to be agreed. BMJ 320:310–311 11. Gould SJ (2001) Sudden unexpected death in infancy. Curr Diagn Pathol 7:69–75 12. Byard RW, Krous HF (eds) (2001) Sudden infant death syndrome. Problems, progress & possibilities. Arnold, Hodder Headline Group, London 13. Kennedy H (2004) Sudden unexpected death in infancy. A multi-agency protocol for care and investigation. The report of a working group convened by The Royal College of Pathologists and The Royal College of Paediatrics and Child Health. London: The Royal College of Pathologists/The Royal College of Paediatrics and Child Health

2  Post-mortem Investigation of Sudden Unexpected Death in Infancy…

43

14. Green MA (1998) A practical approach to suspicious death in infancy – a personal view. J Clin Pathol 51:561–563 15. Fleming PJ, Blair PS, Ward PM, Tripp J, Smith IJ (2003) Sudden infant death syndrome and social deprivation: assessing epidemiological factors after post-matching for deprivation. Paediatr Perinat Epidemiol 17:272–280 16. Malloy MH (2004) Sudden infant death syndrome among extremely preterm infants: United States 1997–1999. J Perinatol 24:181–187 17. Anderson ME, Johnson DC, Batal HA (2005) Sudden Infant Death Syndrome and prenatal maternal smoking: rising attributed risk in the Back to Sleep era. BMC Med 3:4 18. Shah T, Sullivan K, Carter J (2006) Sudden infant death syndrome and reported maternal smoking during pregnancy. Am J Public Health 96:1757–1759 19. Blair PS, Sidebotham P, Berry PJ, Evans M, Fleming PJ (2006) Major epidemiological changes in sudden infant death syndrome: a 20-year population-based study in the UK. Lancet 367: 314–319 20. Moon RY, Horne RS, Hauck FR (2007) Sudden infant death syndrome. Lancet 370:1578–1587 21. Fleming PJ, Gilbert R, Azaz Y, Berry PJ, Rudd PT, Stewart A et al (1990) Interaction between bedding and sleeping position in the sudden infant death syndrome: a population based casecontrol study. BMJ 301:85–89 22. Mitchell EA, Scragg R, Stewart AW, Becroft DM, Taylor BJ, Ford RP et al (1991) Results from the first year of the New Zealand cot death study. N Z Med J 104:71–76 23. Dwyer T, Ponsonby AL, Newman NM, Gibbons LE (1991) Prospective cohort study of prone sleeping position and sudden infant death syndrome. Lancet 337:1244–1247 24. Scragg RK, Mitchell EA (1998) Side sleeping position and bed sharing in the sudden infant death syndrome. Ann Med 30:345–349 25. Gilbert R, Salanti G, Harden M, See S (2005) Infant sleeping position and the sudden infant death syndrome: systematic review of observational studies and historical review of recommendations from 1940 to 2002. Int J Epidemiol 34:874–887 26. Hauck FR, Herman SM, Donovan M et al (2003) Sleep environment and the risk of sudden infant death syndrome in an urban population: the Chicago Infant Mortality Study. Pediatrics 111:1207–1214 27. Vennemann MM, Findeisen M, Butterfass-Bahloul T, Jorch G, Brinkmann B, Kopcke W et  al (2005) Modifiable risk factors for SIDS in Germany: results of GeSID. Acta Paediatr 94:655–660 28. Tappin D, Ecob R, Brooke H (2005) Bedsharing, roomsharing, and sudden infant death syndrome in Scotland: a case-control study. J Pediatr 147:32–37 29. McGarvey C, McDonnell M, Hamilton K, O’Regan M, Matthews T (2006) An 8 year study of risk factors for SIDS: bed-sharing versus non-bed-sharing. Arch Dis Child 91:318–323 30. Horsley T, Clifford T, Barrowman N, Bennett S, Yazdi F, Sampson M, Moher D, Dingwall O, Schachter H, Côté A (2007) Benefits and harms associated with the practice of bed sharing: a systematic review. Arch Pediatr Adolesc Med 161:237–245 31. Ruys JH, de Jonge GA, Brand R, Engelberts AC, Semmekrot BA (2007) Bed-sharing in the first four months of life: a risk factor for sudden infant death. Acta Paediatr 96:1399–1403 32. Blair PS, Mitchell EA, Heckstall-Smith EM, Fleming PJ (2008) Head covering – a major modifiable risk factor for sudden infant death syndrome: a systematic review. Arch Dis Child 93:778–783 33. Hauck FR, Omojokun OO, Siadaty MS (2005) Do pacifiers reduce the risk of sudden infant death syndrome? A meta-analysis. Pediatrics 116:e716–e723 34. Leach CE, Blair PS, Fleming PJ, Smith IJ, Platt MW, Berry PJ et al (1999) Epidemiology of SIDS and explained sudden infant deaths. CESDI SUDI Research Group. Pediatrics 104:e43 35. Platt MW, Blair PS, Fleming PJ, Smith IJ, Cole TJ, Leach CE et al (2000) A clinical comparison of SIDS and explained sudden infant deaths: how healthy and how normal? CESDI SUDI Research Group. Confidential Inquiry into Stillbirths and Deaths in Infancy study. Arch Dis Child 82:98–106 36. Guntheroth WG, Spiers PS (2002) The triple risk hypotheses in sudden infant death syndrome. Pediatrics 110:e64

44

M.A. Weber and N.J. Sebire

37. Morris JA, Haran D, Smith A (1987) Hypothesis: common bacterial toxins are a possible cause of the sudden infant death syndrome. Med Hypotheses 22:211–222 38. Morris JA (1999) The common bacterial toxins hypothesis of sudden infant death syndrome. FEMS Immunol Med Microbiol 25:11–17 39. Morris JA, Harrison LM, Biswas J, Telford DR (2007) Transient bacteraemia: a possible cause of sudden life threatening events. Med Hypotheses 69:1032–1039 40. Blackwell CC, Weir DM (1999) The role of infection in sudden infant death syndrome. FEMS Immunol Med Microbiol 25:1–6 41. Harrison LM, Morris JA, Telford DR, Brown SM, Jones K (1999) The nasopharyngeal bacterial flora in infancy: effects of age, gender, season, viral upper respiratory tract infection and sleeping position. FEMS Immunol Med Microbiol 25:19–28 42. Blackwell CC, MacKenzie DA, James VS, Elton RA, Zorgani AA, Weir DM et  al (1999) Toxigenic bacteria and sudden infant death syndrome (SIDS): nasopharyngeal flora during the first year of life. FEMS Immunol Med Microbiol 25:51–58 43. Weber MA, Klein NJ, Hartley JC, Lock PE, Malone M, Sebire NJ (2008) Infection and sudden unexpected death in infancy: a systematic retrospective case review. Lancet 371:1848–1853 44. Goldwater PN (2009) Sterile site infection at autopsy in sudden unexpected deaths in infancy. Arch Dis Child 94:303–307 45. Blackwell CC, Moscovis SM, Gordon AE, Al Madani OM, Hall ST, Gleeson M et al (2005) Cytokine responses and sudden infant death syndrome: genetic, developmental, and environmental risk factors. J Leukoc Biol 78:1242–1254 46. Buckley MG, Variend S, Walls AF (2001) Elevated serum concentrations of beta-tryptase, but not alpha-tryptase, in Sudden Infant Death Syndrome (SIDS). An investigation of anaphylactic mechanisms. Clin Exp Allergy 31:1696–1704 47. Richardson HL, Walker AM, Horne RS (2009) Maternal smoking impairs arousal patterns in sleeping infants. Sleep 32:515–521 48. Rao H, Saiki T, Landolfo F, Smith AP, Hannam S, Rafferty GF, Milner AD, Greenough A (2009) Position and ventilatory response to added dead space in prematurely born infants. Pediatr Pulmonol 44:387–391 49. Richardson HL, Walker AM, Horne RS (2008) Sleep position alters arousal processes maximally at the high-risk age for sudden infant death syndrome. J Sleep Res 17:450–457 50. Gozal D (2004) New concepts in abnormalities of respiratory control in children. Curr Opin Pediatr 16:305–308 51. Baruteau AE, Baruteau J, Joomye R, Martins R, Treguer F, Baruteau R, Daubert JC, Mabo P, Roussey M (2009) Role of congenital long-QT syndrome in unexplained sudden infant death: proposal for an electrocardiographic screening in relatives. Eur J Pediatr 168:771–777 52. Millat G, Kugener B, Chevalier P, Chahine M, Huang H, Malicier D, Rodriguez-Lafrasse C, Rousson R (2009) Contribution of long-QT syndrome genetic variants in sudden infant death syndrome. Pediatr Cardiol 30:502–509 53. Ottaviani G, Matturri L, Mingrone R, Lavezzi AM (2006) Hypoplasia and neuronal immaturity of the hypoglossal nucleus in sudden infant death. J Clin Pathol 59:497–500 54. Matturri L, Biondo B, Mercurio P, Rossi L (2000) Severe hypoplasia of medullary arcuate nucleus: quantitative analysis in sudden infant death syndrome. Acta Neuropathol 99:371–375 55. Kinney HC, Richerson GB, Dymecki SM, Darnall RA, Nattie EE (2009) The brainstem and serotonin in the sudden infant death syndrome. Annu Rev Pathol 4:517–550 56. Kinney HC (2005) Abnormalities of the brainstem serotonergic system in the sudden infant death syndrome: a review. Pediatr Dev Pathol 8:507–524 57. Kinney HC, Randall LL, Sleeper LA, Willinger M, Belliveau RA, Zec N et  al (2003) Serotonergic brainstem abnormalities in Northern Plains Indians with the sudden infant death syndrome. J Neuropathol Exp Neurol 62:1178–1191 58. Rubens DD (2004) Are lethal audiogenic seizures a missing link to the sudden infant death syndrome? Med Hypotheses 63:87–91 59. Weber MA, Ashworth MT, Risdon RA, Hartley J, Malone M, Sebire NJ (2008) The role of post-mortem investigations in determining the cause of sudden unexpected death in infancy (SUDI). Arch Dis Child 93:1048–1053

2  Post-mortem Investigation of Sudden Unexpected Death in Infancy…

45

60. Vennemann M, Bajanowski T, Butterfass-Bahloul T, Sauerland C, Jorch G, Brinkmann B et al (2007) Do risk factors differ between explained sudden unexpected death in infancy (SUDI) and SIDS? Arch Dis Child 92:133–136 61. Landi K, Gutierrez C, Sampson B, Harruff R, Rubio I, Balbela B et al (2005) Investigation of the sudden death of infants: a multicenter analysis. Pediatr Dev Pathol 8:630–638 62. Mitchell E, Krous HF, Donald T, Byard RW (2000) An analysis of the usefulness of specific stages in the pathologic investigation of sudden infant death. Am J Forensic Med Pathol 21:395–400 63. http://www.statistics.gov.uk/STATBASE/xsdataset.asp?vlnk=4288&More=Y. 64. Weber MA, Ashworth MT, Risdon RA, Brooke I, Malone M, Sebire NJ (2009) Sudden unexpected neonatal death in the first week of life: autopsy findings from a specialist centre. J Matern Fetal Neonatal Med 22:398–404 65. Hobbs CJ (1989) ABC of child abuse. Fractures. BMJ 298:1015–1018 66. Weber MA, Risdon RA, Offiah A, Malone M, Sebire NJ (2009) Rib fractures identified at postmortem examination in sudden unexpected deaths in infancy (SUDI). Forensic Sci Int 189:75–81 67. Feldman KW, Brewer DK (1984) Child abuse, cardiopulmonary resuscitation, and rib fractures. Pediatrics 73:339–342 68. Schweich P, Fleisher G (1985) Rib fractures in children. Pediatr Emerg Care 1:187–189 69. Garcia VF, Gotschall CS, Eichelberger MR, Bowman LM (1990) Rib fractures in children: a marker of severe trauma. J Trauma 30:695–700 70. Bulloch B, Schubert CJ, Brophy PD, Johnson N, Reed MH, Shapiro RA (2000) Cause and clinical characteristics of rib fractures in infants. Pediatrics 105:E48 71. Cadzow SP, Armstrong KL (2000) Rib fractures in infants: red alert! The clinical features, investigations and child protection outcomes. J Paediatr Child Health 36:322–326 72. Maguire S, Mann M, John N, Ellaway B, Sibert JR, Kemp AM (2006) Welsh Child Protection Systematic Review Group. Does cardiopulmonary resuscitation cause rib fractures in children? A systematic review. Child Abuse Negl 30:739–751 73. Forbes A, Acland P (2004) What is the significance of haemosiderin in the lungs of deceased infants? Med Sci Law 44:348–352 74. Krous HF, Chadwick AE, Haas EA, Stanley C (2007) Pulmonary intra-alveolar hemorrhage in SIDS and suffocation. J Forensic Leg Med 14:461–470 75. Krous HF, Haas EA, Masoumi H, Chadwick AE, Stanley C (2007) A comparison of pulmonary intra-alveolar hemorrhage in cases of sudden infant death due to SIDS in a safe sleep environment or to suffocation. Forensic Sci Int 172:56–62 76. Hanzlick R (2001) Pulmonary hemorrhage in deceased infants: baseline data for further study of infant mortality. Am J Forensic Med Pathol 22:188–192 77. Berry PJ (1999) Intra-alveolar haemorrhage in sudden infant death syndrome: a cause for concern? J Clin Pathol 52:553–554 78. Yukawa N, Carter N, Rutty G, Green MA (1999) Intra-alveolar haemorrhage in sudden infant death syndrome: a cause for concern? J Clin Pathol 52:581–587 79. Becroft DM, Lockett BK (1997) Intra-alveolar pulmonary siderophages in sudden infant death: a marker for previous imposed suffocation. Pathology 29:60–63 80. Milroy CM (1999) Munchausen syndrome by proxy and intra-alveolar haemosiderin. Int J Legal Med 112:309–312 81. Dorandeu A, Perie G, Jouan H, Leroy B, Gray F, Durigon M (1999) Histological demonstration of haemosiderin deposits in lungs and liver from victims of chronic physical child abuse. Int J Legal Med 112:280–286 82. Weber MA, Ashworth MT, Risdon RA, Malone M, Sebire NJ (2009) The frequency and significance of alveolar haemosiderin-laden macrophages in sudden infants death. Forensic Sci Int 187:51–57 83. Becroft DM, Thompson JM, Mitchell EA (2005) Pulmonary interstitial haemosiderin in infancy: a common consequence of normal labor. Pediatr Dev Pathol 8:448–452 84. Byard RW, Carmichael E, Beal S (1994) How useful is postmortem examination in sudden infant death syndrome? Pediatr Pathol 14:817–822

46

M.A. Weber and N.J. Sebire

85. Sadler DW (1998) The value of a thorough protocol in the investigation of sudden infant deaths. J Clin Pathol 51(9):689–694 86. Arnestad M, Vege A, Rognum TO (2002) Evaluation of diagnostic tools applied in the examination of sudden unexpected deaths in infancy and early childhood. Forensic Sci Int 125:262–268 87. Morris JA, Harrison LM, Partridge SM (2006) Postmortem bacteriology: a re-evaluation. J Clin Pathol 59:1–9 88. Morris JA, Harrison LM, Partridge SM (2007) Practical and theoretical aspects of postmortem bacteriology. Curr Diagn Pathol 13:65–74 89. Oehmichen M, Schleiss D, Pedal I, Saternus KS, Gerling I, Meissner C (2008) Shaken baby syndrome: re-examination of diffuse axonal injury as cause of death. Acta Neuropathol 116:317–329 90. Cohen M, Cox P, Kiho L, Pollina E, Scheimberg I (2007) Letter to the editor. Lack of evidence for a causal relationship between hypoxic-ischemic encephalopathy and subdural hemorrhage in fetal life, infancy, and early childhood. Pediatr Dev Pathol 10:500–501 91. Squier W, Mack J (2009) The neuropathology of infant subdural haemorrhage. Forensic Sci Int 187:6–13 92. Bajanowski T, Vennemann M, Bohnert M, Rauch E, Brinkmann B, Mitchell EA (2005) Unnatural causes of sudden unexpected deaths initially thought to be sudden infant death syndrome. Int J Legal Med 119:213–216 93. Langlois NE, Ellis PS, Little D, Hulewicz B (2002) Toxicologic analysis in cases of possible sudden infant death syndrome: a worthwhile exercise? Am J Forensic Med Pathol 23:162–166

Chapter 3

Sudden Death from Pulmonary Causes Kris S. Cunningham and Michael S. Pollanen

Abstract  This chapter seeks to survey many of the common pathological entities identified in the lungs at autopsy and the potential role of pulmonary disease in formulating an opinion regarding the cause of death. Appreciation of pulmonary pathology in the medicolegal context is important as it frequently contributes to the immediate or underlying mechanisms of death. The primacy of the lungs in breathing and their coordinated function with the cardiovascular system means that ­pulmonary failure can rapidly compromise tissue oxygenation and body chemistry, leading to an alteration in blood pH, hypoxic damage to downstream tissues and ultimately multiorgan failure and death. Moreover, given that the lungs have direct contact with the environment through inhalation and receive approximately 50% of the cardiac output with each beat of the heart, they may be adversely affected by hazardous agents from the outside world or other pathologic processes not primarily located in the lungs. The range of topics discussed herein is limited by design to deaths due to disease and largely foregoes discussion of more forensically relevant issues relating to toxicology or trauma. Furthermore, the content and format of this chapter is not intended to be encyclopedic, but rather attempts to highlight selected issues regarding pulmonary disease of potential relevance to surgical or forensic pathologists who perform medicolegal postmortem examinations. Keywords  Sudden death • Pulmonary pathology • Autopsy • Forensic practice

K.S. Cunningham (*) Provincial Forensic Pathology Unit, Ontario Forensic Pathology Service, Centre for Forensic Science and Medicine, University of Toronto, 26 Grenville Street, 2nd Floor, Toronto, ON, Canada M7A 2G9 e-mail: [email protected] E.E. Turk (ed.), Forensic Pathology Reviews, Volume 6, Forensic Pathology Reviews 6, DOI 10.1007/978-1-61779-249-6_3, © Springer Science+Business Media, LLC 2011

47

48

K.S. Cunningham and M.S. Pollanen

Pediatric Pulmonary Pathology Pulmonary Hypoplasia Pulmonary hypoplasia in the neonate is an abnormal reduction in mass or volume of the lungs with a normal pattern of lobation. Depending on the extent of hypoplasia and the morphological stage of bronchiolar and alveolar maturation, severe ­hypoplasia is often incompatible with postnatal life. Multiple developmental and structural anomalies of the thoracic cavity and lungs are associated with pulmonary hypoplasia and neonatal death (Table 3.1) [1]. Co-existing systemic developmental abnormalities may occur with pulmonary hypoplasia. Examples include small or malformed thoracic cavities, hydrops fetalis, oligohydramnios sequence with ­associated genitourinary disorders, premature rupture of fetal membranes, large intrapulmonary or abdominal mass lesions, diaphragmatic malformations (Fig. 3.1), central nervous system disorders associated with impaired breathing and cardiovascular disorders associated with reduced pulmonary blood flow. In particular, the oligohydramnios or Potter’s Sequence highlights the central role of the genitourinary system and its associated production of urine in the staged development of the lung parenchyma. Examples of conditions associated with deficient amniotic fluid production or premature loss include renal agenesis, cystic renal dysplasia, ­polycystic kidney disease, cloacal agenesis, urethral atresia, posterior urethral valves as well as chronic leakage of amniotic fluid [1]. The neonate with severe pulmonary hypoplasia is hypoxic and eventually develops­pulmonary hypertension. This can be exacerbated by concurrent anatomical­ abnormalities of the heart and brain. Moreover, the degree of respiratory ­compromise

Table 3.1  Selected conditions associated with severe pulmonary hypoplasia Hydrops fetalis with large pleural effusions Oligohydramnios sequence Congenital diaphragmatic defects with herniation of abdominal contents Congenital cystic adenomatoid malformations Pulmonary sequestrations Congenital muscular dystrophy Spinal muscular atrophy Intrauterine hypoxic ischemic encephalopathy with secondary pulmonary hypoplasia Large foregut cysts Thanatophoric dysplasia Osteogenesis imperfecta II Jeune syndrome/asphyxiating thoracic dystrophy Achondroplasia Abdominal mass lesions Eventration of the diaphragm Central nervous system lesions associated with decreased breathing Congenital heart disease associated with decreased pulmonary arterial blood flow Cytogenetic abnormalities

3  Sudden Death from Pulmonary Causes

49

Fig. 3.1  Hypoplastic left lung in an infant with a congenital diaphragmatic hernia (Courtesy of Dr. C. Kepron)

is dependent not only on the reduction in size and volume of the lung, but also on the stage of development of the pulmonary parenchyma. Immature alveolar septae may have limited gas exchange as well as incomplete development of the pulmonary vasculature. Microscopically, the number of alveoli can be reduced with fewer bronchioles, less branching and more alveolar collapse as a consequence of surfactant deficiency. Secondary acute lung injury can develop as sequelae of reduced ventilation, infection or atelectasis. Examination of the lungs radiologically as part of the skeletal survey can reveal a pneumothorax prior to commencement of the internal examination. This may be particularly true following mechanical ventilation as the hypoplastic lung appears to be more susceptible to barotrauma. In addition, examination of the lungs in situ, prior to evisceration and disruption of the thoracic cavity is suggested to properly assess the underlying anatomy and associated pathology. Unfortunately, the degree of pulmonary hypoplasia necessary to cause clinically significant morbidity or mortality is not precisely defined by the available literature. Thus, correlation of the gross and histological findings with the clinical history is important for correct interpretation of the postmortem findings and evaluating their potential contribution to the ultimate cause of death.

Respiratory Distress Syndrome and Bronchopulmonary Dysplasia Acute lung injury may occur as a consequence of numerous insults such as ­infection, shock, drug effects, or assisted ventilation. In the setting of significant prematurity, surfactant deficiency may cause acute lung injury with the development of hyaline

50

K.S. Cunningham and M.S. Pollanen

membranes. Depending on the degree of prematurity and the duration of the insult, a spectrum of acute and chronic phases of diffuse alveolar damage (DAD) can be identified, which may ultimately lead to remodeling of the underlying lung architecture and reduced respiratory reserve. In the context of the appropriate clinical ­history, respiratory distress syndrome (RDS) or surfactant deficiency syndrome is the clinical correlate for hyaline membrane disease identified in the premature infant at autopsy. The histological changes identified in premature infants are mirrored in studies of surfactant-deficient mice (surfactant protein A: SP-A −/− mice) [2]. The  microscopic features can include collapsed distal airspaces with hyaline ­membranes, fibrinous intra-alveolar exudates, pneumocyte hyperplasia, leakage of erythrocytes, acute and chronic alveolar, and interstitial inflammation, as well as collections of siderophages and foamy macrophages. With time organizing pneumonia, granulation tissue plugs within respiratory bronchioles and architectural remodeling of the interstitium may be observed. Neonates born with surfactant ­deficiency during the late saccular and early alveolar stages (32–36 weeks) usually do not die as a consequence of their prematurity-associated lung disease. However, neonates with extreme prematurity (<28 weeks) in the canalicular stage of lung development (16–28 weeks) or those under 1,000 g are most at risk for RDS and bronchopulmonary dysplasia (BPD) [3]. BPD develops in those infants who have survived the acute phase of lung injury due to surfactant deficiency and develop a chronic fibrosing process of the lung parenchyma that is characterized by patchy, nonspecific, interstitial fibrosis, epithelial regeneration, and parenchymal collapse intermixed with regions of overdistended lung and squamous metaplasia of the distal conducting airways. In addition, medial hypertrophy of pulmonary arteries and arterioles may also be noted in regions of significant parenchymal remodeling (microscopic honeycomb change). Importantly, marked lung injury in infants born before 28 weeks gestation may cause arrest of alveolar development and subsequent deficiency of gas exchange [3]. At autopsy, children with severe BPD may possess firm, consolidated lungs (hepatization), as well as reduced lung volumes and decreased numbers of alveoli. Severe BPD has been attributed in part to adult-type positive pressure ventilatory techniques; however, current ventilation strategies adapted to neonates appear to limit the severity of BPD with more subtle morphological changes now identified in infants at autopsy. In addition to ventilatory changes, antenatal glucocorticoid administration, surfactant replacement therapy and alteration in ventilation strategies have reduced the incidence of RDS. Moreover, when examining premature lungs by microscopy, amorphous eosinophilic masses within the distal airways may be observed as a consequence of surfactant replacement therapy and should not be confused with alveolar proteinosis [3].

Acute Lung Injury and Ventilator-Associated Trauma In children with nonspecific acute lung injury, the mortality rate is high and increases in those who go on to develop acute respiratory distress syndrome (up to 50%) [1].

3  Sudden Death from Pulmonary Causes

51

Fig. 3.2  Chronic aspiration with exogenous lipoid pneumonia (orig. mag. 100×)

Fig. 3.3  Diffuse intravascular fat emboli highlighted with osmium stain (orig. mag. 25×)

Notwithstanding prematurity-associated surfactant deficiency, acute lung injury (ALI) in the pediatric population may mirror that observed in the adult population and be a consequence of direct or indirect causes. Direct causes of ALI can include infectious pneumonias, eosinophilic pneumonitis, aspiration (Fig.  3.2), fat emboli (Fig.  3.3),

52

K.S. Cunningham and M.S. Pollanen

reperfusion injury, cardiovascular disease, severe pulmonary hypertension­, ­pulmonary infarction and malignancy. Indirect causes may include sepsis, shock, transfusion reactions, cardiopulmonary bypass, drug reactions or acute pancreatitis [1]. Ventilation-associated injuries may contribute to respiratory distress syndrome in premature infants or acute lung injury in general. The histological changes induced by chronic positive pressure ventilation frequently overlap with those of the underlying­disease and include acute phase diffuse alveolar damage, interstitial edema, and inflammation. Mechanistically, ventilator-induced lung injury is thought to contribute to alveolar damage by multiple mechanisms. These include (1) barotrauma with elevated shear stress on the alveolar lining; (2) volutrauma with large tidal volumes that can disrupt the epithelial–capillary interface; (3) ­atelectrauma caused by mechanical ventilation-associated stress fracture of the epithelial–­ capillary interface following repetitive opening and closing of alveoli; and finally, (4) biotrauma, which follows the influx of neutrophils and inflammatory mediators in response to lung injury [1]. Ultimately, ventilatory strategies seek to prevent atelectasis and maintain the patency of airways without overdistention and ­pressure-related trauma. Air leak may be seen following ventilation-induced trauma to the parenchyma. Following pressure and volume-associated rupture of bronchiolo-alveolar ­junctions; intra-alveolar gas may pass into the peribronchovascular spaces and dissect along bronchovascular sheaths, leading to pulmonary interstitial emphysema. Pulmonary interstitial emphysema can potentially extend into adjacent hilar soft tissues, leading to the development of pneumomediastinum, pnueumopericardium, pneumothorax, or rarely air embolism, which has been associated with sudden death. These ventilation-associated injuries may be appreciated radiologically as meandering cystic and tubular lucencies that fail to conform to the predicted pattern of air bronchograms; thin-walled cysts (pneumatoceles) may also be identified [2]. When prominent and unilateral, pulmonary interstitial emphysema, pneumatoceles, and pneumothoracies can rarely cause shift of the mediastinal structures under tension and promote cardiovascular collapse and death.

Born Alive or Dead Determination of whether a fetus was born alive or was stillborn, not having had a separate living existence from its mother, is a frequent question posed to forensic pathologists. Under the best of circumstances, this determination is difficult. Classically, the flotation or hydrostatic test is utilized, which is based on the premise that en bloc lungs or an individual lung placed in water will float if aerated, ­suggesting that the neonate had at some point breathed or sink if nonaerated, ­signifying that the fetus was stillborn [4]. Interpretation of the results of this is fraught with difficulty in and it is not recommended to be used in isolation to make a determination of live birth. Potential causes for false-positive results could include air trapping due to attempted resuscitation or the production of postmortem gases by bacteria during putrefactive decomposition. Gas-producing bacteria may be

3  Sudden Death from Pulmonary Causes

53

introduced into the airways, for example, through postdelivery contamination when delivered into a toilet or as a consequence of choramnionitis. Furthermore, we have encountered the circumstance where lungs were examined separately and revealed that one lung sank and the other floated. Moreover, interpreting an increase in the lung’s mass as evidence of perimortem breathing, thought to be due to expansion and vascular perfusion of the parenchyma, is also a poor indicator of live birth [4]. Histological sections of the lung may show evidence of alveoli expanded by air following a live birth. However, the results must be tempered by the postmortem interval and the possibility of postmortem gas production by microorganisms. In addition, the presence of diffuse alveolar damage or pneumonia would strongly suggest that the infant was live born. Further, less stringent physical evidence of potential live birth may include a dirty diaper, food in the stomach, an inflammatory reaction at the umbilical cord stump and the absence of maceration. However, although the history provided to the pathologist at the time of postmortem examination must be considered in the ­context of the autopsy findings, one must guard against incorporation of circumstantial ­evidence when making a determination of live birth, as ultimately, in the absence of physical evidence this opinion may not be defendable. Teleologically, the best ­evidence of live birth is the identification of a clear cause for death.

Sudden Unexpected Death in Infancy and Sudden Infant Death Syndrome Sudden and unexpected death in infancy (SUDI) is often associated with potentially lethal upper and lower respiratory tract inflammation, most commonly as a ­consequence of bacterial and/or viral infections [5]. For children who are immunocompromised or from endemic regions, less common infectious agents that could include viruses (cytomegalovirus, herpes simplex virus, varicella zoster virus), fungi (aspergillus, candidiasis, histoplasmosis, blastomycosis, and mucormycosis) or myocobacteria (tuberculosis and mycobacterium avium intracellulare) may be identified within the upper and lower respiratory tract. Death as a consequence of acute viral pneumonitis and its complications in ­otherwise “healthy” children is observed. Correlation of the histomorphologic ­features from a virally infected lung with the results of microbiological culture is a mainstay of assessment. Sampling of the nasopharynx as well as the middle ears for virus may also be considered in addition to lung parenchyma, tracheal tissue, and blood. It has been suggested that in upward of 50–80% of SUDI cases that can be explained after autopsy, respiratory tract infections played a significant role in the ultimate cause of death [5]. Important risk factors for the development of pneumonia to extract from the clinical history could include cytogenetic abnormalities, ­congenital heart disease, prior aspiration of oropharyngeal or gastric contents, ­pulmonary masses or fistulas involving the respiratory tract, malignancy, diabetes mellitus, or other causes for a compromised immune system.

54

K.S. Cunningham and M.S. Pollanen

Fig. 3.4  Acute epiglottitis [Courtesy of Dr. J. Tanguay] (orig. mag. 16×)

Acute epiglottitis with swelling and enlargement of the epiglottis may abruptly occlude the airway causing respiratory insufficiency and death. Historically, Haemophilus influenzae infection has been implicated; however, immunization ­protocols have significantly reduced the incidence of acute epiglottitis in children [6]. However, other organisms such as Pneumococcus or parainfluenza viruses may also be isolated in both children as well as adults with acute epiglottitis (Fig. 3.4). Occasional cases of vaccinated children who have died from acute epiglottitis who subsequently test positive for Haemophilus colonization are identified. Thus, vaccine failure against Haemophilus influenzae can occur and must be considered in circumstances of positive postmortem cultures from affected children [7]. Other conditions affecting the upper airways that can lead to sudden death include retropharyngeal abscesses causing acute occlusion or massive hemorrhage within the larynx, tracheomalacea, laryngeal polyps (Fig. 3.5) [8] and bacterial tracheitis. In particular, acute bacterial tracheitis is often superimposed on a preceding viral infection and is associated with suppurative exudates and potentially pseudomembranes within the trachea at autopsy. Complications of bacterial tracheitis include bacterial pneumonitis, sepsis and acute respiratory distress syndrome. Microorganisms associated with acute tracheitis include S. aureus, H. influenzae, and parainfluenza virus [9]. Respiratory syncytial virus infection (RSV) is the most common cause for acute bronchitis and bronchiolitis (Fig. 3.6). In older children, parainfluenza, rhinovirus, influenza virus (Fig. 3.7), adenovirus and Mycoplasma pneumoniae are also commonly observed with clinical bronchitis. Histological findings with RSV include acute and chronic inflammatory infiltrates with airway obstruction by epithelial and inflammatory cell debris, mucus, fibrin, and hyperplastic lymphoid tissues. Occasional

3  Sudden Death from Pulmonary Causes

55

Fig. 3.5  Occlusive laryngeal polyp causing sudden death [Courtesy of Dr. J. Tanguay] (orig. mag. 16×)

Fig. 3.6  Immunohistochemistry for respiratory syncytial virus highlighting bronchial epithelium (orig. mag. 25×)

multinucleated syncytial cells with viral cytopathic changes may also be observed; however, this finding is uncommon and limited by sampling. Ancillary studies to detect RSV through viral culture, immunofluorescence, immunohistochemistry

56

K.S. Cunningham and M.S. Pollanen

Fig. 3.7  Acute bronchitis/bronchiolitis with influenza A infection [Courtesy of Dr. J. Tanguay] (orig. mag. 16×)

or  ELISA-based methods are suggested when possible in appropriate cases [10]. Finally, acute viral bronchitis may be complicated by the development of an acute bacterial pneumonia, particularly by superimposed S. aureus infection. In the circumstance of sudden infant death syndrome (SIDS) the spectrum of pulmonary changes identified are generally nonspecific in their appearance and are not obviously lethal. Typically, the lungs are congested with increased mass and pulmonary edema. It is important to not overinterpret pulmonary edema and congestion at autopsy as evidence of pneumonia. Parenchymal changes may include petechial hemorrhages, particularly in a pleural/subpleural distribution, focal acute congestive hemorrhage with or without occasional siderophages, as well as intraalveolar and interstitial edema. Atelectasis and focal interstitial emphysema may also be observed in children in whom resuscitation has been attempted [11]. The presence of a protein-rich fluid within centrilobular bronchioles, alveolar ducts, and centrilobular alveoli can be due to aspiration of gastric contents (e.g., milk), which may occur in extremis and should not necessarily be interpreted as acute aspiration causing death. Furthermore, depending on the age of the infant, occasional squamous cells (from amniotic fluid) and hyperplastic bronchus-­ associated lymphoid tissue can be normal for this age group. In cases of suspected SIDS, multiple sections from each lobe of lung are ­typically assessed with hematoxylin and eosin staining. In addition, the use of connective tissue stains (elastic trichrome, Movat pentachrome), histochemical stains for microorganisms and immunohistochemical stains to detect inflammatory cells, neuroendocrine hyperplasia [12], viruses, or other microorganisms may be considered for each case.

3  Sudden Death from Pulmonary Causes

57

An assessment of the radial-alveolar index as well as ­immunohistochemical stains for keratin and endothelial cell distribution may be considered if maturity of the lung and parenchymal architecture is to be assessed [13]. Interstitial and intra-alveolar hemosiderin and siderophages, the presence of which can be confirmed with iron stains, have in the past been used as evidence of asphyxia and in particular, suffocation. Although the nature of this manuscript precludes a detailed discussion of this controversial issue, multiple authors have demonstrated that the amount of hemosiderin does not necessarily indicate that an infant died an asphyxial death. Furthermore, iron in the lungs may be associated with multiple pathophysiological processes as part of natural, accidental or nonaccidental deaths. Thus, the presence of significant hemorrhage and hemosiderin should not be used as an independent predictor in determining the manner of death in an infant [14].

Pulmonary Pathology in the Adult Pulmonary Infections A comprehensive review of pulmonary infections is beyond the scope of this chapter. Nevertheless, acute bacterial pneumonia is a common immediate cause of death. However, pneumonia may also arise as a secondary complication of some other underlying disease process. Thus, while pneumonia may contribute to the immediate demise of an individual, the ultimate underlying pathophysiological abnormality or circumstance that initiated the causal chain leading to that pneumonia must be sought. Both community-acquired pneumonia and nosicomial pneumonia (those cases that develop after 72 h in hospital) are frequently found at postmortem examination (Table 3.2). The nature of the microorganisms identified following histomorphology and/or culture assessment may facilitate the reconstruction of events leading up to death and delineate those risk factors that promoted infection of the lung in the first place. In addition, the macroscopic pattern of inflammation within the lung may also assist with identification of the putative infectious agent and potentially direct the appropriate sampling of tissue for ancillary studies (Table 3.3). Ultimately, it is not necessarily the identification of an acute pneumonia at autopsy that is significant, but rather recognition of the underlying risk factors for that pneumonia that are potentially of medicolegal importance (Tables 3.4 and 3.5). Any factor that impairs normal respiratory function (e.g., obesity, thoracic cavity anomalies, prolonged immobility, alcoholism); diminishes the host immunological defense against microorganisms (e.g., chronic disease, medications); provides a permissive environment for opportunistic infections (e.g., intrapulmonary cavities, foreign bodies, tracheo-esophageal fistula); increases the risk of aspiration of oropharyngeal or gastric contents (e.g., edentulous state, neuromuscular disease, dementia) (Fig. 3.8) or seeding of the lung with microorganisms from an infected site (e.g., postoperative state, abscesses, endocarditis, intravenous catheters) may

58

K.S. Cunningham and M.S. Pollanen

Table 3.2  Classical community-acquired and nosicomial microorganisms Community acquired Nosicomial (³72 h in hospital) Streptococcus pneumoniae Streptococcus pneumoniae Haemophilus influenzae Haemophilus influenzae Legionella pneumophila Pseudomonas aeruginosa Moraxella catarrhalis Enterobacter spp. Mycoplasma pneumoniae Klebsiella pneumoniae Chlamydia psittaci Serratia marcescens Chlamydia pneumoniae Staphalococcus aureus Influenza A Escherichia coli Adenovirus Acinetobacter spp. Mycobacterium tuberculosis Anaerobes (Peptostreptococcus, Fusobacterium, Peptococcus, Bacteroides) Fungi (Aspergillus, Candida) Fungi (Aspergillus, Candida)

Table 3.3  Gross patterns of lung involvement with infections Macroscopic pattern Possible organism Patchy, centrilobular or lobar infiltrates Bacteria, mycobacteria Nodular with or without cavitation Fungi, pneumocystis, bacteria, septic emboli, nocardia Brochiectasis Mycobacterial, fungal infections Prominent cavitation Bacterial abscess, tuberculosis, aspergilloma Empyema Bacteria, mycobacteria Diffuse infiltrates Pneumocystis, CMV Enlarged mediastinal lymph nodes Mycobacteria, histoplasmosis, coccidiomycosis

Table 3.4  Risk factors for pneumonia of potential medicolegal relevance Mechanical ventilation/tracheostomy Neuromuscular disease Dementia Edentulous Severe chronic disease (e.g., diabetes mellitus, chronic renal failure, chronic heart failure) Drug effects/chemotherapy Acute alcohol consumption/chronic alcoholism Immunocompromised states Anatomical anomalies (e.g., Tracheal-esophageal fistulae, sequestrations) Thrombophlebitis/cellulitis/endocarditis/nonpulmonary abscesses/sepsis Pulmonary/nonpulmonary neoplasia Intrapulmonary foreign bodies Prolonged immobility/atelectasis Postoperative infection Thoracic cavity anatomical anomalies (e.g., scoliosis, contractures) Acute pulmonary infarcts Morbid obesity Underlying pulmonary disease (e.g., bronchiectasis, tuberculosis, emphysema) Intravenous drug abuse Intravenous catheters

3  Sudden Death from Pulmonary Causes

59

Table 3.5  Histomorphological appearance of selected organisms with hematoxylin and eosin stains Organism H&E appearance Cytomegalovirus Well-defined intranuclear inclusions (+/− halo) Adenovirus Poorly defined intranuclear inclusions (smudge cells) Herpes virus Well-defined intranuclear inclusions (­multinucleated, large, glassy) Respiratory syncytial virus Influenza virus Histoplasmaa Cryptococcus Blastomyces Coccidioides Candida Aspergillus Zygomces (Mucor)

Multinuclear syncytial cells (typically low numbers) Nonspecific cytopathic appearance 2–5 mm; narrow-based budding 5–20 mm; narrow-based budding, thick wall 15–30 mm; broad-based budding 20–200 mm; endospores Yeast and hyphal forms Acute angle branching; septate Right angle branching; broad-based ribbons; few septa

A mature erythrocyte is typically 7–9 mm Culture, histochemical stains, or immunostains may be used to confirm diagnosis a

Fig. 3.8  Aspiration of gastric contents. Note concentration of intra-alveolar edema in a centrilobular­ distribution, adjacent to anthracotic pigment deposition. Fragments of partially digested food, colonies of bacterial microorganisms and acute inflammation are also identified (orig. mag. 16×)

60

K.S. Cunningham and M.S. Pollanen

result in a potentially lethal pneumonitis. Moreover, the host reaction against an infectious agent may be associated with significant functional impairment of respiratory function. These changes can include the accumulation of acute and chronic inflammatory infiltrates, interstitial and intra-alveolar pulmonary edema as well as findings of diffuse alveolar damage. Complications of acute infections may include abscess formation, empyema, bronchopulmonary fistulae, significant hemoptysis, or sepsis. With time, acute infectious pneumonitis with its associated acute lung injury may resolve with complete functional reconstitution of the lung parenchyma. Pneumonitis can also develop into a chronic infection leading to a chronic inflammatory process causing architectural remodeling of the underlying lung parenchyma with impairment of gas exchange. Examples of chronic changes one may observe microscopically or at autopsy can include abscesses formation, nonspecific interstitial fibrosis, stromal and epithelial metaplasia with cyst formation, bronchiectasis, obstructive bronchiolitis, and honeycomb change. Pneumonia may be due to a compromised immune system. Relevant risk factors may include HIV infection, autoimmune disease, congenital immune deficiency, intravenous drug abuse, medication effects/chemotherapy, transplantation, advanced age, malignancy, chronic alcoholism, neglect, malnourishment, and diabetes mellitus. In some instances, consideration of one or more of these factors may be required when commenting on the significance of a potentially lethal pneumonia at autopsy. Furthermore, multiple conditions that promote acute lung injury such as ventilation may also increase the susceptibility for a secondary bacterial pneumonia. Other examples include antecedent influenza bronchiolitis, inhalational injuries following fire or chemical exposures or a chemical pneumonitis following aspiration of gastric acid. At the postmortem examination, appropriate sampling of tissues for ancillary testing must be considered. Depending on the clinical circumstances, sampling can include tissue for bacterial, fungal, or viral culture, tissue for snap freezing or electron microscopy, as well as blood for aerobic and anaerobic cultures. If indicated, communication with a public health laboratory or microbiologist before the autopsy may provide guidance to facilitate procurement of tissue specimens for ancillary testing. Respiratory Infection Outbreaks For infectious respiratory outbreaks caused by unknown pathogens, institutional protocols should be consulted for an approach to the autopsy that ensures the safety of autopsy staff, outlines the appropriate sampling of tissues for analysis and enables proper handling of tissues to minimize unintended transmission of virulent microorganisms after postmortem examination. It is important to consider undertaking a complete autopsy with sampling of body fluids and tissues examined when investigating an unknown infectious pathogen. This serves two purposes; first, it allows one to confirm that the cause of death was as a result of primary pulmonary disease caused by a putative pathogen, and second, permits procurement of varied tissues for analysis as it is not always clear prior to the autopsy that an unknown pathogen

3  Sudden Death from Pulmonary Causes

61

associated with respiratory failure is best isolated from lung tissue. Moreover, if pneumonia is identified, a complete autopsy may permit one to assess if that infection is primary or perhaps secondary to some other pathologic process, such as bronchial obstruction due to occult carcinoma. Anecdotally, during the SARS (severe acute respiratory syndrome) and Legionella pneumonia outbreaks in Toronto, a number of deaths that were clinically suspected to be due to the infectious agent in question were ultimately determined to be the result of other pathologies. The ultimate cause for death may have been missed had the autopsy been solely restricted to the lungs.

Interpretation of Suspected Aspiration A controversial finding in many autopsies is the presence of food fragments within the distal airways/alveolar ducts identified at microscopic examination. It is well recognized that gastric contents may contaminate the airways during the postmortem period and potentially migrate deep into the lung parenchyma. In the absence of evidence of a significant host response such as an acute inflammatory infiltrate, it is very difficult to substantiate a diagnosis of perimortem aspiration. Migration of gastric contents into the airways may also occur during attempted resuscitation of the decedent. In addition, even if early acute inflammatory infiltrates are identified in association with foodstuffs in the airway, aspiration of gastric contents can commonly occur in extremis. Consequently, these findings should not necessarily be interpreted to mean that airway occlusion by aspirated gastric contents contributed to the death of the individual being autopsied. The number of affected airways, the degree and nature of the host response and the clinical context should all be considered when confronted with this issue.

Cancer and Its Physiological Derangements Evidence of malignancy within the lungs is commonly observed in medicolegal autopsies and includes both primary pulmonary malignancies as well as metastatic tumor deposits. The traditional risk factors for primary lung carcinoma for which we may find evidence at autopsy includes chronic cigarette smoking, occupationbased pneumoconioses such as asbestosis or silicosis, idiopathic pulmonary fibrosis (or UIP, usual interstitial pneumonitis), coal miner’s lung as well as prior solid organ transplantation. Chronic infection with the human immunodeficiency virus (HIV) has also been associated with an increased risk of bronchogenic carcinoma as well as other pulmonary malignancies [15]. The lungs are a frequent site for metastatic disease. Classically, metastatic carcinoma, sarcoma, melanoma, or germ cell tumors present with multiple parenchymal nodules that may be unilateral or bilateral in their distribution, located either centrally or more often peripherally and frequently deposit within the pleura and subpleural tissues.

62

K.S. Cunningham and M.S. Pollanen

Fig. 3.9  Squamous cell carcinoma arising out of background of usual interstitial pneumonitis (orig. mag. 16×)

Carcinomas may invade the lung parenchyma directly from a contiguous site or enter through the vasculature and lymphatic channels. Lymphatic channels in the lungs are generally identified around the bronchovascular bundles, within the interlobular septae and within the pleura. With a more pronounced lymphangitic pattern of spread, known as lymphangitic carcinomatosis, the gross examination may reveal a number of small nodules bilaterally that are often associated with thickened parenchymal septae and visceral pleura. Microscopically, lymphangitic carcinomatosis may be obvious or very subtle. Sarcomas frequently metastasize to the lungs by hematogenous spread and can infiltrate or arbourize along the vasculature in a serpinginous manner. Occasionally, this pattern of tumor embolization and growth has been found to be fatal [16]. Primary bronchogenic carcinoma may also locally invade the lung parenchyma in atypical patterns. Examples can include restriction to the pleural/subpleural tissues that can mimic mesothelioma or arise from within regions of dense interstitial fibrosis such as with usual interstitial pneumonitis (UIP) (Fig. 3.9), which can be entirely missed macroscopically. Although metastatic bronchogenic carcinoma may metastasize to virtually any anatomical site, it has a predilection for the ipsilateral or contralateral lungs, mediastinal lymph nodes, brain, liver, bone, and adrenal glands. Given the altered appearance of decomposing tissues and the difficulty in ­interpreting immunohistochemistry, it is generally not necessary to specifically characterize the nature of the malignancy beyond a general class of neoplasm if possible; such examples would include carcinoma (non-small cell vs. small cell),

3  Sudden Death from Pulmonary Causes

63

sarcoma, lymphoma, melanoma, or germ cell tumor. In the end, however, the ­questions that may arise out of the case should ultimately dictate the degree of ­diagnostic detail required. Lung cancer may cause cachexia and death through numerous physiological derangements. A malignancy can compromise its surrounding anatomy, which may lead to a potentially lethal infection, infarction of surrounding normal tissues or massive hemorrhage. Examples of lethal complications attributable to lung cancer include erosion of tumor into large vascular structures (e.g., intrapulmonary, mediastinal, or cardiac) causing massive hemorrhage and shock; metastasis of tumor to sensitive regions of the heart or brain causing catastrophic organ failure; promotion of intravascular thrombosis with subsequent occlusion of in situ or downstream vasculature; empyema; development of bronchopulmonary fistulae and pneumothorax and finally, occlusion of bronchi with subsequent bacterial and lipoid pneumonias. In addition, recurrent pleural effusions as a result of the malignancy may lead to atelectasis and respiratory embarrassment, which can complicate respiratory function in an individual whom may already have compromised cardiorespiratory reserve. Metastatic carcinoma can extensively occlude a large percentage of the intrapulmonary microvasculature, which can lead to sudden death. Such pulmonary tumor microemboli have been identified in individuals with occult malignancies who present in extremis with apparent respiratory failure. Clinically, this condition can mimic pneumonia, tuberculosis, and interstitial lung disease [17]. Furthermore, the additional presence of numerous microscopic thromboemboli and fibrointimal proliferative lesions within the pulmonary microvasculature may also be observed, a condition referred to as pulmonary tumor thrombotic microangiopathy (Fig. 3.10). This associated condition may lead to potentially lethal pulmonary hypertension and is most commonly linked with metastatic adenocarcinomas from the upper gastrointestinal tract [18]. Bronchogenic carcinomas, especially small cell carcinoma, may be associated with numerous paraneoplastic syndromes that have physical manifestations that can be observed at autopsy and in certain circumstances may contribute to the mechanism of death (Table 3.6). Notable examples include central pontine myelinolysis in the setting of syndrome of inappropriate antidiuretic hormone secretion (SIADH), widespread paraneoplastic pemphigus or coagulopathic anomalies. Interpretation of postmortem physical and biochemical changes from suspected paraneoplastic syndromes requires careful correlation with the antemortem clinical history.

Pulmonary Vascular and Cardiovascular Disease Pulmonary Thromboembolism A wide range of pulmonary vascular anomalies may be associated with sudden death. One of the most common causes for sudden and unexpected death is acute,

64

K.S. Cunningham and M.S. Pollanen

Fig. 3.10  Tumor thrombotic microangiopathy adjacent to tumor-filled lymphatic space (orig. mag. 50×)

Table 3.6  Paraneoplastic syndromes with potential ­significance to the medicolegal autopsy Cushing syndrome Polymyositis-dermatomyositis Hypercalcemia Syndrome of inappropriate antidiuretic hormone secretion Paraneoplastic cerebellar degeneration Encephalomyelitis Sweet’s syndrome Pyoderma gangrenosum Nonbacterial thrombotic endocarditis Paraneoplastic pemphigus Disseminated intravascular coagulation Acquired thrombophelia Anemia Membranous glomerulonephritis

occlusive pulmonary arterial thromboembolism. The origin of intravascular thrombi is typically from deep femoral, popliteal, or crural veins. Sometimes thromboemboli originate from deep pelvic veins, the inferior caval vein, hepatic veins, deep veins of the upper extremities or mural thrombi from within the heart. The cause of deep venous thrombi may be a consequence of an identifiable thrombogenic risk factor(s) (Table 3.7).

3  Sudden Death from Pulmonary Causes

65

Table 3.7  Selected risk factors for intravascular thrombus formation of interest to the medicolegal autopsy Inherited coagulation factor mutations (factor V Leiden, mutations in prothrombin, etc.) Prolonged immobility Prolonged restraint Vascular trauma Vasculitis/vascular aneurysms/vascular malformations Pregnancy Oral contraceptive medications/estrogen replacement/estrogen supplementation Estrogen producing tumors (e.g., adult granulosa cell tumor) Postoperative states Lupus/antiphospholipid antibody syndrome Malignancy (especially adenocarcinoma) Nephrotic syndrome Dilated cardiomyopathy Acute or remote myocardial infarction/ventricular aneurysm/cardiac tumors Chronic arrhythmia (e.g., atrial fibrillation, atrial flutter) Intravascular foreign bodies (catheters, pacemaker leads, prosthetic valves, prosthetic pumps, etc.) Sepsis/diffuse intravascular coagulation Thrombotic thrombocytopenic purpura/hemolytic uremic syndrome/eclampsia

Acute thromboemboli may cause sudden death through occlusion of one or more large caliber pulmonary arteries. In addition, diffuse occlusion of smaller caliber ­vessels by thromboemboli, located in the more peripheral zones of the lung may also cause sudden death. The number and/or size of acute thrombi that cause sudden death likely depend on the underlying cardiopulmonary reserve of the affected individual. Any combination of proximal and/or peripherally located thromboemboli that cause an acute rise in the right ventricular systolic pressure by 40–50 mmHg can lead to acute heart failure [19]. Moreover, one may consider microscopically dating intravascular thrombi into, for example, recent, organizing, organized, or old thrombi in order to provide comment about the potential chronicity of the embolic process, which could have medicolegal significance. For example, if an individual dies 4 days following a surgical procedure of pulmonary thromboembolism, yet has evidence of organizing and organized thrombi in their pulmonary arteries and deep veins of their lower legs, one may at least suggest that the decedent had underlying thrombogenic risk factors that preceded the surgical procedure. If such facilities are available, one could consider testing for commonly inherited mutations within the genes of coagulation factors using postmortem blood and include the results within the autopsy report to enable clinical assessment of any first-degree family members.

Pulmonary Hypertension Pulmonary hypertension may also cause sudden and unexpected death and can be broadly divided into primary and secondary causes (Table 3.8) [20–23]. It is important to recognize that some forms of primary pulmonary hypertension can be inherited and

66

K.S. Cunningham and M.S. Pollanen

Table 3.8  Selected factors associated with pulmonary hypertension (PHTN) of interest to the medicolegal autopsy 1. Primary disease of the pulmonary vasculature:

Familial PHTN, idiopathic PHTN, pulmonary veno-occlusive disease, capillary hemangiomatosis, persistent pulmonary hypertension of the newborn 2. Pulmonary hypertension associated Collagen vascular disease, HIV infection, medication with systemic disease: effects, intravenous drug abuse with foreign body embolization, porto-pulmonary hypertension, sarcoidosis, sickle cell disease, chronic thromboembolic disease, tumor thrombotic microangiopathy, hematological disease, amyloidosis, congenital syndromes (e.g., Alagille syndrome), associated with chronically elevated altitudes, systemic to pulmonary shunts 3. Pulmonary hypertension associated Chronic fibrosing interstitial lung disease, chronic with underlying lung disease: obstructive lung disease/emphysema, marked architectural remodeling following organizing phase diffuse alveolar damage/bronchopulmonary dysplasia, pulmonary lymphangioleiomyomatosis 4. Pulmonary hypertension associated Left-sided ventricular failure, aortic or mitral valvular with underlying heart disease: disease, hypertrophic cardiomyopathy, infiltrative heart disease with restrictive physiology (e.g., amyloidosis, hemochromatosis), congenital heart disease with left to right shunts

that one should consider communicating this within the autopsy report; however, ­secondary causes for pulmonary hypertension are far more frequently detected at autopsy. The morphological changes associated with pulmonary hypertension may be subtle when examining the heart and lungs macroscopically at autopsy. Associated morphological lesions may include “cirrhosis” (cardiac sclerosis), right ventricular hypertrophy, thickening and myxomatous degeneration of the tricuspid and less commonly the pulmonary valves, ectasia of the pulmonary arterial vasculature with or without atherosclerosis, intravascular thrombi, pulmonary infarcts, and nonspecific interstitial fibrous tissue deposition. Microscopically, the lesions may be plexogenic or nonplexogenic in nature and depending on the nature of the underlying cause for hypertension, may involve both arterial and venous vessels. Complications of chronic pulmonary hypertension can include cor pulmonale, dissection of the pulmonary arteries [24] and massive intrapulmonary hemorrhage [25]. Pulmonary arteries in central and peripheral locations will often show intimal and medial hyperplasia, reduplication of elastic lamina and incorporation of organized thrombi. Microscopic assessment of peripheral lung tissue for evidence of small vessel disease, plexogenic lesions, remote intrapulmonary hemorrhage, interstitial fibrous tissue deposition and intravascular thrombi supports the diagnosis of chronic pulmonary hypertension (Fig. 3.11). Furthermore, special connective tissue stains such as elastic trichrome or Movat pentachrome will also facilitate the evaluation of underlying architectural changes within the lung. Rare diseases of the pulmonary vasculature that may also be associated with ­sudden death include pulmonary artery sarcoma; pulmonary vasculitities such as

3  Sudden Death from Pulmonary Causes

67

Fig. 3.11  Chronic vascular changes in patient with Eisenmenger Syndrome [movat pentachrome] (orig. mag. 100×)

Goodpasture syndrome, Wegener’s granulomatosis, microscopic polyangiitis, and Takayasu’s arteritis; as well as rupture of pulmonary artery aneurysms and pulmonary artery dissections (Fig. 3.12) [26, 27].

Emphysema and Asthma Acute exacerbations of obstructive lung diseases such as chronic obstructive ­pulmonary disease (COPD) as well as bronchial asthma frequently cause sudden death. With moderate to marked emphysema often associated with long-term ­smoking or chronic occupational exposures such as coal dust or with a1-antitrypsin deficiency, one may observe secondary pathological changes such as numerous pleural bullae or blebs, pneumothoracies with associated lobar collapse, loss of lung parenchyma in centriacinar, panacinar, or subpleural patterns, as well as diffuse mucus plugging. Histologically, the presence of centriacinar anthracotic pigment deposition, respiratory bronchiolitis and chronic bronchial/bronchiolar inflammation supports a diagnosis of smoking or organic dust-associated COPD. In addition, recent work has suggested that chronic, high quantities of alcohol consumption may also exacerbate the smoking-related risk of developing COPD [28]. Other smoking-related lesions that may occasionally be identified in the lungs of patients with COPD include bronchogenic carcinoma, pulmonary langerhans cell histiocytosis (Fig. 3.13) and desquamative interstitial pneumonitis (DIP). Although

68

K.S. Cunningham and M.S. Pollanen

Fig. 3.12  Acute pulmonary artery dissection in 16-day-old infant with complex congenital heart pathology [elastic trichrome] (orig. mag. 16×)

Fig. 3.13  Pulmonary Langerhans Cell Histiocytosis. Note cellular stellate-shaped nodule with Langerhans cells and eosinophils. Smoking-related lesion not to be confused with a malignancy. Immuno: S100+, CD1a + and Langerin + (orig. mag. 16×)

3  Sudden Death from Pulmonary Causes

69

Fig. 3.14  Acute exacerbation of bronchial asthma (orig. mag. 50×)

sudden death may be observed with seemingly stable severe emphysema with or without chronic bronchitis, an inciting pathology such as bronchopneumonia, myocardial ischemia/infarction or aspiration of gastric contents is often identified as an exacerbating factor that can lead to respiratory decompensation, hypoxemia, and sudden death. Sudden death due to bronchial asthma often correlates with severe, acute respiratory distress and increased usage of b-agonists [29], a delay in seeking medical attention and a clinical history of poorly controlled asthma (Fig. 3.14). It is within this context of multiple bronchial and bronchiolar mucus plugs, submucosal glandular hyperplasia, basement membrane thickening, bronchial/bronchiolar smooth muscle hypertrophy, and increased numbers of intramucosal eosinophils are virtually diagnostic of bronchial asthma. The lungs often appear hyperinflated at autopsy, frequently with overlapping borders of the right and left lungs across the anterior surface of the mediastinum. A frozen section at the time of autopsy may facilitate early diagnosis. Careful assessment of the history often suggests exposure to a potential trigger prior to respiratory distress. However, sudden and unexpected death is not uncommon in individuals with seemingly well-controlled asthma [30]. Finally, given that some individuals with bronchial asthma may also be at increased risk of systemic anaphylaxis following exposure to specific allergens, the differential diagnosis of systemic anaphylaxis causing acute respiratory distress and sudden death may also be considered when evaluating bronchial asthma as a potential cause for death. Furthermore, submitting serum for immunoglobulin E and tryptase levels in ­conjunction with other physical findings such as airway angio-edema may facilitate making this distinction in selected cases.

70

K.S. Cunningham and M.S. Pollanen

Fig. 3.15  Usual interstitial pneumonitis. Note paraseptal/subpleural distribution of fibrous tissue [spatial heterogeneity of fibrosis], regions of near normal parenchyma adjacent to dense fibrosis as well as fibroblastic foci [temporal heterogeneity of fibrosis] (orig. mag. 16×)

Interstitial Lung Disease Numerous acute or chronic interstitial lung diseases may be associated with sudden death. Acute conditions can include severe hypersensitivity pneumonitis/­eosinophilic pneumonia, Loeffler’s pneumonitis, acute allergic bronchopulmonary aspergillosis, and acute interstitial pneumonitis. Conditions that can cause a chronic fibrosing interstitial pneumonitis include usual interstitial pneumonitis (UIP) (Fig. 3.15) and nonspecific interstitial pneumonitis (NSIP). A giant cell interstitial pneumonitis may be observed following chronic heavy metal exposure. Additionally, exposure to asbestos fibers can lead to asbestosis, which typically presents in a UIP pattern of interstitial fibrosis. A nonspecific pattern of interstitial fibrous tissue deposition may also be identified with honeycomb lung, which represents the cystic and fibrosing architectural remodeling of the pulmonary parenchyma in response to various forms of acute lung injury and may occur at the end-stage of a multitude of interstitial and alveolar disease processes. Widespread interstitial and pleural nodules may occur as a consequence of ­infectious or noninfectious granulomatous disease, pneumoconioses, pulmonary langerhans cell histiocytosis (PLCH), amyloidosis, bronchocentric granulomatosis and chronic aspiration of oral or gastric contents. Finally, cystic lung changes may be observed in the setting of emphysema, lymphangioleiomyomatosis, PLCH, ­suppurative abscesses, malignancies, Wegener’s granulomatosis pneumatoceles, and bronchogenic cysts.

3  Sudden Death from Pulmonary Causes

71

Fig. 3.16  Hypersensitivity pneumonitis. Note poorly formed granulomas, occasional interstitial eosinophils and organizing pneumonia in a centrilobular distribution (orig. mag. 100×)

Diagnostic challenges with interstitial lung disease include identification of major patterns of fibrous tissue deposition, identification of any associated findings such as ferruginous bodies, fragments of aspirated food or inorganic dusts and recognition of secondary changes in the lung such as marked medial and intimal hyperplasia of pulmonary vessels with usual interstitial pneumonitis. It is important to sample less-affected lung tissue that has not gone onto end-stage honeycomb lung, as end-stage pulmonary fibrosis is generally not diagnostically specific and may not facilitate identification of the underlying interstitial pathology. All lung lobes should be sampled for microscopic assessment as many interstitial lung diseases have a predilection for different regions of the lung or possess a spectrum of histological changes that can be of diagnostic utility, such as with UIP, PLCH, or hypersensitivity pneumonitis (Fig. 3.16). Special connective tissue stains such as elastic trichrome or Movat pentachrome are often of great value in interpreting the underlying architecture of the lung which can aide in interpreting the aberrant pulmonary disease process. Granulomatous disease should prompt the use of special stains to help detect fungi, mycobacteria and bacterial microorganisms. Important information to garner from the history could include the rate of development of the clinical disease, the presence of any autoimmune or collagen vascular diseases, comparison to any prior thoracic radiology and identification of any prior infectious, environmental, occupational, or medicinal exposures. End-stage chronic interstitial pneumonitis is often associated with pulmonary hypertension and cor pulmonale. Furthermore, it is not uncommon to identify occult malignancy within regions of dense interstitial fibrous tissue such as with UIP.

72

K.S. Cunningham and M.S. Pollanen

Fig. 3.17  Wall of chronic pulmonary aspergilloma associated with intractable hemoptysis [movat pentachrome]. Note destruction of the underlying vessel wall (orig. mag. 2×)

Contribution of Pulmonary Disease to Systemic Disease Processes Multiple systemic disease processes involve the lung parenchyma and as such, may cause impairment of respiratory function and potentially contribute to the immediate cause for death. Examples include diffuse alveolar damage associated with systemic shock or “shock lung”; diffuse interstitial thickening as a consequence of collagen vascular disease, amyloidosis or sarcoidosis; hemoaspiration associated with upper digestive tract hemorrhage; aorto-bronchial fistulae; aspergillosis (Fig. 3.17) and bronchiectasis and finally, external restriction or compression of the lungs by chronic fibrous pleuritis, large hydrothoracies, empyema, ascities under tension and morbid obesity. The potential degree of respiratory dysfunction that each one of these conditions may play mechanistically in the immediate or underlying cause for death is highly dependent on the clinical history and the pathological context that these findings are identified.

Conclusions The identification of pulmonary pathology at autopsy may be central to determining the underlying cause for death. Alternatively, significant disease of the respiratory system may play a secondary role, contributing mechanistically to the more

3  Sudden Death from Pulmonary Causes

73

i­mmediate pathophysiological abnormalities occurring in the period just before death. Recognition of the various patterns of lung disease and appreciation of the differential diagnostic considerations for each of these pathologies will enable pathologists performing medicolegal autopsies to arrange for appropriate ancillary studies and ultimately opine about the immediate and underlying causes for death. Acknowledgments We would like to thank Dr. J. Tanguay, Dr. C. Kepron, Dr. D.M. Hwang, and Mr. D. Clutterbuck for their generous time and assistance.

References 1. Dahlem P, van Aalderen WM, Bos AP (2007) Pediatric acute lung injury. Paediatr Respir Rev 8:348–362 2. DeMello DE (2004) Pulmonary pathology. Semin Neonatol 9:311–329 3. Bendon RW, Coventry S (2004) Non-iatrogenic pathology of the preterm infant. Semin Neonatol 9:281–282 4. Byard RW (ed) (2004) Forensic pathology reviews. Humana Press, Totowa, USA 5. Bajanowski T, Vege A, Byard RW, Krous HF, Arnestad M, Bachs L et al (2007) Sudden infant death syndrome (SIDS) – standardised investigations and classification: recommendations. Forensic Sci Int 165:129–143 6. Gorelick MH, Baker MD (1994) Epiglottitis in children, 1979 through 1992. Effects of Haemophilus influenzae type b immunization. Arch Pediatr Adolesc Med 148:47–50 7. McEwan J, Giridharan W, Clarke RW, Shears P (2003) Paediatric acute epiglottitis: not a disappearing entity. Int J Pediatr Otorhinolaryngol 67:317–321 8. Tanguay J, Pollanen MS (2009) Sudden death by laryngeal polyp: a case report and review of the literature. Forensic Sci Med Pathol 5:17–21 9. Busuttil A, Keeling JW (2009) Paediatric forensic medicine and pathology. Hodder Arnold, London 10. Johnson JE, Gonzales RA, Olson SJ, Wright PF, Graham BS (2007) The histopathology of fatal untreated human respiratory syncytial virus infection. Mod Pathol 20:108–119 11. Valdes-Dapena M, McFeeley PA, Hoffman HJ, Damus KH, Franciosi RR, Allison DJ et  al (1993) Histopathological Atlas for the sudden infant death syndrome. Armed Forces Institute of Pathology, Washington, DC 12. Cutz E, Perrin DG, Pan J, Haas EA, Krous HF (2007) Pulmonary neuroendocrine cells and neuroepithelial bodies in sudden infant death syndrome: potential markers of airway chemoreceptor dysfunction. Pediatr Dev Pathol 10:106–116 13. Betz P, Nerlich A, Bussler J, Hausmann R, Eisenmenger W (1997) Radial alveolar count as a tool for the estimation of fetal age. Int J Legal Med 110:52–54 14. Krous HF, Chadwick AE, Haas EA, Stanley C (2007) Pulmonary intra-alveolar hemorrhage in SIDS and suffocation. J Forensic Leg Med 14:461–470 15. Kanmogne GD (2005) Noninfectious pulmonary complications of HIV/AIDS. Curr Opin Pulm Med 11:208–212 16. Ahmed AA, Heller DS (1999) Fatal pulmonary tumor embolism caused by chondroblastic osteosarcoma: report of a case and review of the literature. Arch Pathol Lab Med 123:437–440 17. Jakel J, Ramaswamy A, Kohler U, Barth PJ (2006) Massive pulmonary tumor microembolism from a hepatocellular carcinoma. Pathol Res Pract 202:395–399 18. Chinen K, Kazumoto T, Ohkura Y, Matsubara O, Tsuchiya E (2005) Pulmonary tumor thrombotic microangiopathy caused by a gastric carcinoma expressing vascular endothelial growth factor and tissue factor. Pathol Int 55:27–31

74

K.S. Cunningham and M.S. Pollanen

19. Rahimtoola A, Bergin JD (2005) Acute pulmonary embolism: an update on diagnosis and management. Curr Probl Cardiol 30:61–114 20. Montani D, Price LC, Dorfmuller P, Achouh L, Jais X, Yaici A et al (2009) Pulmonary venoocclusive disease. Eur Respir J 33:189–200 21. Srigley JA, Pollanen MS (2005) Sudden death with clinically undiagnosed pulmonary hypertension. J Clin Forensic Med 12:264–267 22. Diller GP, Gatzoulis MA (2007) Pulmonary vascular disease in adults with congenital heart disease. Circulation 115:1039–1050 23. Chan SY, Loscalzo J (2008) Pathogenic mechanisms of pulmonary arterial hypertension. J Mol Cell Cardiol 44:14–30 24. Arena V, De GF, Abbate A, Capelli A, De MD, Carbone A (2004) Fatal pulmonary arterial dissection and sudden death as initial manifestation of primary pulmonary hypertension: a case report. Cardiovasc Pathol 13:230–232 25. Leung WH, Lau CP, Wong CK, Cheng CH (1990) Fatal massive pulmonary hemorrhage complicating mitral stenosis. Clin Cardiol 13:136–138 26. Marchetti D, La MG, Ranalletta D (1996) Death due to a leiomyosarcoma of the pulmonary artery. Am J Forensic Med Pathol 17:315–318 27. Graham JK, Shehata B (2007) Sudden death due to dissecting pulmonary artery aneurysm: a case report and review of the literature. Am J Forensic Med Pathol 28:342–344 28. Sisson JH (2007) Alcohol and airways function in health and disease. Alcohol 41:293–307 29. Cazzola M, Matera MG, Donner CF (2005) Inhaled beta2-adrenoceptor agonists: cardiovascular safety in patients with obstructive lung disease. Drugs 65:1595–1610 30. Tough SC, Green FH, Paul JE, Wigle DT, Butt JC (1996) Sudden death from asthma in 108 children and young adults. J Asthma 33:179–188

Chapter 4

Sectioning of the Heart, Searching for Pathology Under the Microscope, and the Cardiac Proteomics Approach in the Study of Sudden Cardiac Death Cases Vittorio Fineschi, M.S.B. Othman, and Emanuela Turillazzi

…yet there is a method in’t. Hamlet, II, II, 205–206

Abstract  Cardiac causes account for a large majority of sudden natural deaths. The investigation of sudden cardiac death requires a careful correlation of circumstantial data with autopsy and laboratory findings. The correct postmortem dissection of the heart is crucial for the identification the exact cause of death. Standard samples should be used to perform histological, biochemical, toxicological, and genetic exams. Keywords  Sudden cardiac death • Autopsy • Dissection techniques

Introduction A good autopsy is one where the pathologist has considered the clinical situation in detail, and has performed the examination in a manner that attempts to answer the questions implicit in the case history [1]. This chapter is devoted to the analysis of “any death which is rapid (without prodrome), unexpected and/ or unforeseen, that occurs in apparently healthy people, or in ill patients during a benign phase of their disease.” Since any natural sudden death can be considered cardiac in origin after the exclusion of noncardiac causes, a full autopsy with a sequential approach, should always be performed to look for common and uncommon cardiac and extra-cardiac causes of sudden death [2]. Here we studied only the deaths which are cardiac in origin.

V. Fineschi (*) Department of Forensic Pathology, University of Foggia, Ospedale Colonnello D’Avanzo, Via degli Aviatori 1, 71100 Foggia, Italy e-mail: [email protected] E.E. Turk (ed.), Forensic Pathology Reviews, Volume 6, Forensic Pathology Reviews 6, DOI 10.1007/978-1-61779-249-6_4, © Springer Science+Business Media, LLC 2011

75

76

V. Fineschi et al.

The medico-legal autopsy must consist of more than just a final diagnosis. We are obliged to supply reasons for the diagnoses we provided, and we must be able to defend those reasons. We must be prepare to explain not only what we have done, but how we did it, and do so in sufficient detail that someone else looking at the same data set would reach the same conclusions. Simply put: personal experience is good, science is better [3]. Over the last three decades, landmark advances in the understanding and treatment of sudden cardiac death (SCD) have been made. Structural and electrical mechanisms of terminal arrhythmias have been elucidated. Over two dozen genetic mutations and polymorphisms have been identified, which, in turn, have increased our understanding of ion channel structure and function [4]. A complete autopsy, including detailed neuropathological and cardiovascular examination with toxicological studies, must be performed in the context of all available clinical information and on the circumstances of death, and discovering those that are cardiovascular in origin but not related to coronary atherosclerosis. At present, unique objective data are postmortem findings and, in a selected group, changes detectable by electrocardiography in monitored patients or clinical follow-ups from resuscitated patients. Almost all sudden cardiac death investigations require a careful correlation of circumstantial data with autopsy and laboratory findings. Relatively few natural death causes are self-evident by themselves at autopsy [5]. Approach to the autopsy in potentially sudden cardiac death cases is difficult and we then recommend a standard gross examination of the heart. In dissection of the heart, some fundamental requirements are essential: 1. Applying particular techniques to conditions and situations imposing to single out and localize reports or to delimit pathological tissues or samples (for instance, a different dissection or heart and big vessels opening when a valvular alteration is suspected). 2. Using fixed and standard samples to perform histological, biochemical, toxicological, genetic exams, and so forth. 3. Keeping parts of samples and reports in order to guarantee that the investigations will be repeatable. 4. Collect and file the necessary, preliminary evidence for forensic judgment. It presumes an absolutely clear objectivity (supported by photographic, microfilm documents) and a more trustful susceptibility to check.

Sectioning of the Heart Accurate information is required to make an accurate postmortem diagnosis and that, in turn, requires an accurate postmortem dissection. Reaching that goal requires an ongoing, continuous assessment of the methods used and the accuracy with which they have been applied. When you are examining a sudden death case, you always have to open the ­pulmonary artery (on the place) to look for the possible presence of embolism. After

4  Sectioning of the Heart, Searching for Pathology Under the Microscope…

77

you excised the inferior vena cava that comes out of the diaphragm and the aorta and pulmonary arteries (as high as possible), you extract the heart, and you will record the quantity and the appearance of the subepicardial fat and the epicardial’s features; then you weigh viscera after you empty the cavities. External examination of the heart also allows for examination of the shape and dimensions of the heart; these dimensions should be correlated with those made of nearby anatomic structures. Most of the old methods of dissection are really not practical for a routine diagnostic purpose, which is why only the inflow–outflow and the “short-axis” methods have survived. The last technique is useful in virtually every form of cardiac dysfunction. A number of recent methods have been developed that are very useful for teaching purposes, as they allow for easier comparisons, and easier demonstration of the normal heart structures. Regrettably, many forensic pathologists forgo the next step, but the heart should then be suspended in a large container, filled with neutral 10% formalin solution, for 24 h. Even though a period of only 24 h does not allow for complete fixation, it will harden the tissue, allowing myocardial dissection without any damage to the tissue itself. The mild fixation hardens the tissue and facilitates precise cutting of the organ. First, coronary arteries and their main extramural branches are cross-sectioned at 3 mm intervals along their whole course. Each 3 mm segment is removed together with the surrounding epimyocardial tissue to preserve any anatomical relationship. Before sectioning, when needed, the vessels are decalcified (Figs. 4.1 and 4.2).

The Standard Method For the opening of the heart, the most widely used technique is the one originally proposed by Virchow, and modified by Prausniz. The heart is opened in the direction of flow through the venae cavae, cutting the lateral margin of the right ventricle, its pulmonary cone, and its pulmonary artery. Then, from the left atrial veins and in sequence the left atrium, left ventricular margin, left outflow tract, and the aorta are opened. Then you keep cutting the origin of the big arteries in order to explore the intima and the valves. At this point, you finish cutting the atrium and ventricular valves, so you can explore the atrioventricular valves, their papilla muscles, and the state of the ventricular endocardium. The following step is to describe the myocardium in regard to its wall’s thickness (do not include the trabecular stratum in your measurement), the color, and their other features. Furthermore, by comparing clinical imaging, for example echocardiographic imaging, different plane dissections of the heart can be adopted (Fig. 4.3).

The Long-Axis Method The long-axis plane transects the heart from the aortic root to the left ventricular apex and includes the valves planes (Fig. 4.4). Using a long handled scalpel, make an

78

V. Fineschi et al.

Fig. 4.1  (Left) Plastic cast of the coronary arteries in a normal adult heart. By tridimensional plastic casts of the coronary arteries injected through the aorta – and therefore with reproduction of luminal ostia – we adopted the following classification (Fineschi, Baroldi, Silver 2006): Type I (77%; men 84%, women 68%): The right coronary artery supplies the posterior one-third of the interventricular septum and, to a variable extent, the posterior left ventricle according to these subtypes: Type Ia (5%): The RCA ends as a posterior descending branch without prominent branches to the left ventricle. This subtype is usually associated with a diagonal disposition of the LCX. Type Ib (55%): The RCA terminates between the posterior interventricular groove, and the left cardiac margin supplies about one-half of the posterior left ventricle. Type Ic (13%): The RCA ends at the left margin, supplying the whole posterior left ventricular wall. Type II (8%; men 6%, women 11%): The posterior descending branch arises from the LCX that supplies the whole posterior left ventricular wall, and the interventricular septum is entirely vascularized by the LCA through its two main branches. Type III (15%; men 10%, women 21%): Two posterior descending branches exist, one from the RCA and the other from the LCX. (Right) Comparison between the intimal thickening of the left anterior descending (LAD) artery (a) and the middle cerebral artery (b) of the same 18-year-old subject. In the latter artery, the intimal thickening is minimal in contrast to that of LAD, which is circumferential with a thickness greater than tunica media. (c) Difference in maximal thickness in microns found in main coronary arteries and branches in respect to the middle cerebral artery. (d) Absence of intimal ­thickening in the LAD of dogs, despite an identical morpho-function. This suggests a possible role of the neurogenic control of coronary arteries in humans. On the other hand, the absence of intimal thickening in the mural tract of coronary arterial vessels (e) emphasizes the role of systolic dynamic stresses on the arterial wall free to expand vs. those protected by encircling contracted myocardium

Fig. 4.2  Coronary arteries and their main extramural branches are cross-sectioned at 3 mm intervals along their whole course. Each 3  mm segment is removed together with the surrounding epimyocardial tissue to preserve any anatomical relationship

Fig. 4.3  Comparing clinical imaging, for example echocardiographic imaging, different plane dissections of the heart can be adopted. The long- and short-axis planes (modified from Standard Planes for Two-Dimensional Echocardiography, http://www.echoincontext.com/begin/skillB_07.asp)

80

V. Fineschi et al.

Fig. 4.4  The long-axis plane transects the heart from the aortic root to the left ventricular apex and includes the valves planes. RA right atrium, LA left atrium, TV tricuspid valve, MV mitral valve, LV left ventricle, RV right ventricle, S septum, PV pulmonary valve, AV aortic valve

incision at the apex of the heart, and extend the incision past the acute border of the right ventricle, the obtuse border of the left ventricle, and the interventricular septum. The incision must be extended through the mitral valve, the tricuspid, and the atria. When this type of incision is used, it is possible to split the heart in two, allowing for an easy comparison of all four chambers of the heart. The superior half will be opened along both ventricular outflow paths, following the “flow-down” flow method.

The Short-Axis Method The short axis approximates to the plane of the atrioventricular junction. Positioning the heart on the front surface, the first cut begins 2 cm up to the atrioventricular right sulco and continues in a circular way as a short-axis modification, passing the upper right atrioventricular valve, the pulmonary semilunar valves, and through the interventricular septum (Fig. 4.5). The cut is completed, finally, passing the upper left atrioventricular valve and the aortic semilunar valves. Now, the sufficiency of atrioventricular valves of the pulmonary and aortic semilunar valves could be investigated, putting some water into it, taking care not to let any coagulum of blood to form, because it could pass through the left ventricle up into the aorta, and it could prevent the semilunar valves from flattening under the water pressure: this could feign the insufficiency of valves of the aorta.

4  Sectioning of the Heart, Searching for Pathology Under the Microscope…

81

Fig. 4.5  The short axis approximates to the plane of the atrioventricular junction. Positioning the heart on the front surface, the first cut begins 2 cm up to the atrioventricular right sulco and continues in a circular way as a short-axis modification, passing upper the right atrioventricular valve, the pulmonary semilunar valves and through the interventricular septum. The cut is completed, finally, passing the upper left atrioventricular valve and the aortic semilunar valves. TV tricuspid valve, MV mitral valve AV aortic valve, PV pulmonary valve

Then, the origin of the big arteries has to be cut in order to explore the intima and the valves as well as the atrium and ventricular valves in order to explore the ­atrioventricular valves, their papilla muscles, and the state of the ventricular endocardium (Fig. 4.6).

Tissue Sampling A systematic sampling for histological and immunohistochemical examination is needed: 1. The first segment of coronary arteries and main extramural branches and any other segment with pathological changes are processed for histology, any heavily calcified coronary tract being previously decalcified; up to four slices of coronary arteries marked for orientation can be included in a histologic block. For a correct evaluation of adventitial and periadventitial structures, the coronary samples must contain epicardium and underlying myocardium. 2. A central heart slice is sampled at the anterior, lateral, and posterior walls of the left and right ventricle, and at the anterior and posterior interventricular

82

V. Fineschi et al.

Fig. 4.6  Comparing the long- and short-axis sections with the parasternal view of the middle and basal portions of the left ventricle (LV), the left atrium (LA), and the aortic root (Ao) by two-dimensional (2D) echocardiogram. An instant of the mid-systole of a cardiac cycle is “frozen” in this frame, as documented by the closed mitral valve as opposite to the open aortic valve. RV right ventricle, RA right atrium, AV aortic valve, LA left atrium, LV left ventricle, S septum, TV tricuspid valve, MV mitral valve

septum – as well as any pathological area seen grossly – by 2 cm wide sections of the whole wall thickness, including epicardium and endocardium. When a more complete examination is required, two heart slices, one superior and one inferior, are similarly processed. All samples are first placed in 10% buffered formalin to complete fixation. The uncut atriovalvular part and remaining tissue are preserved in a closed plastic bag containing a small amount of formalin, for any further quantitative study of the sinus node, conduction system, atria, valves, and so forth (Fig. 4.7).

The Microscopic Approach New methods of microscopic approach to heart investigation in SCD have been developed, and the accurate histochemical and immunohistochemical investigation of all cases of sudden cardiac death is now of particular importance. For an adequate assessment of sudden cardiac death, besides gross heart examination, histological

4  Sectioning of the Heart, Searching for Pathology Under the Microscope…

83

Fig. 4.7  A systematic sampling for histological and immunohistochemical examination is needed

sampling for further microscopic investigation is of paramount importance. Although traditional histological investigation (H&E, trichrome staining, etc.) under light microscopy is often rapidly available, it was found to be poorly informative in the assessment of sudden cardiac death causes. A detailed description of preparation of heart specimens for histological examination is beyond the scope of this paper and the reader is referred to the many detailed published tests on technical methods and interpretation of microscopic “traditional” findings in cases of SCD (i.e., acute inflammation in an endocarditis process; myocardial ischemic damage; substitution of myocardium with adipose or fibroadipose tissue in the arrhythmogenic right ventricular dysplasia; myocardial disarray in hypertrophic cardiomiopathy, etc.). However, most of cases of sudden cardiac death remain unexplained after gross examination of the heart and despite conventional histochemical staining procedures are performed. Immunohistochemistry has become a practical and widely used tool for diagnosis in human cardiopathology. In this respect, an extraordinary progress has been achieved with the production of a wide variety of antibodies to epitopes resistant to formalin fixation, and with the use of antigen retrieval methods and powerful amplifying systems. These techniques have proven efficiency in cardiopathology, permitting the utilization of a wide spectrum of markers whose usefulness in several cardiac diseases leading to sudden death is unquestioned. Some possible applications of immunohistochemistry are here discussed with specific regard to SCD investigations.

84

V. Fineschi et al.

Fig. 4.8  Although traditional histological investigation (H&E, trichrome staining, etc.) under light microscopy is often rapidly available, it was found to be poorly informative in the assessment of sudden cardiac death causes (a). Immunohistochemical detection of TNFa using antibodies antiTNFa (b, c). b1 receptor’s expression: multi- dimensional view produced by high-resolution confocal microscopy system allows a detailed view of ultrastructural changes in cardiomyocites (d)

The usefulness of immunohistochemical markers to the diagnosis of early ischemic myocardial damage has been suggested because most of them can be detectable as early as few minutes after the beginning of the myocardial injury. Many markers have been selected on the basis of their different diagnostic potential in early ischemic myocardial injury (C5b-9 complex, C9, fibronectin and fibrinogen, myoglobin, cardiac troponin C, and cardiac troponin T) [6–9]. The immunohistochemical demonstration of the selected markers on heart specimens can provide a good evidence of myocardial ischemia and/or necrosis, supporting the final diagnosis of SCD when no macroscopic or routine microscopic evidence is available and no other pathological changes can be observed. Evaluation of immunohistochemical expression of these markers appears to be a highly sensitive and specific marker of early myocardial infarction, very useful in forensic medicine. Due to new insights in myocardial tumor necrosis factor-alpha expression in sudden death cases [10–12] and to the demonstration that TNFa may be pathobiologically involved in the progression of dilated cardiomyopathy which is often associated to sudden death [12], immunohistochemical detection of TNFa using antibodies anti-TNFa is essential in many cases of SCD (Fig. 4.8).

4  Sectioning of the Heart, Searching for Pathology Under the Microscope…

85

Again, immunohistochemical studies play a very important role in the study of genetic causes of sudden death. This approach should open new avenues of cardiac investigation since it is well known that arrhythmic SCD may occur in the absence of gross and microscopic structural cardiac pathology due to ion channel disorders. Cardiac ion channelopathies are responsible for an ever-increasing number and diversity of familial cardiac arrhythmia syndromes. Multiple abnormalities in genes related to both potassium and cardiac sodium channels are thought to be involved in many cases of SCD (Long QT Syndromes, Brugada Syndrome). In these cases, where structural abnormality resides at molecular level, deoxyribonucleic acid (DNA) testing on peripheral blood and tissue has revolutionized the diagnosis of genetic causes of sudden death. In this setting, immunohistochemical techniques may be used on cardiac tissue samples to estimate the cellular localization and the distribution of particular ­subtypes of sodium and potassium channels as it is well known that alterations in these ion channel expressions and function have severe effects on excitability. Differential expression and localization of sodium and potassium channel su­bunits are likely to be important determinants of electric excitability of cardiac myocytes. The understanding of mechanisms underlying many cases of SCD may be implemented by the immunohistochemical study of expression and distribution of specific ion channel subunits in cardiac tissue samples. Using polyclonal antibodies anti-Na + CP type Va (C-20) we have demonstrated the specific localization of this subtype of sodium channel to the intercalated disk and a nearly 50% reduction in Na + channels expression in ventricular myocytes (compared with control cases) in an young man who suddenly died in a family with electrocardiographic pattern typical of Brugada Syndrome (Fig. 4.9) [13]. The same technique allowed us to confirm the genetic basis [14] in a simultaneous sudden infant death syndrome case. Immunohistochemistry combined with high-resolution imaging may be used to reveal the distribution of ryanodine receptors (RyRs) (the intracellular Ca(2+) release channels) whose mutations is involved in catecholaminergic polymorphic ventricular tachycardia which manifests as severe arrhythmias [15, 16]. Recent studies proposed a novel concept that the defects of non-ion channel proteins or channel-interacting proteins can affect ion channel gating kinetics, thereby causing secondary channel dysfunction leading to LQTS. Hence, this concept uncovers a cascade or domino effect that disturbs the “final common pathway” that causes arrhythmias, ion channels, and focuses attention on a novel class of proteins and candidate genes to explain the residual 25% of LQTS that remain genotype negative. Syntrophins are cytoplasmic submembranous proteins that are components of the dystrophin-associated protein complex whose isoforms have been identified in the heart. The PDZ domain of syntrophin-a1 (SNTA1), the most abundant isoform in the heart, has been reported to bind to the C-terminal domain of murine cardiac voltage-gated sodium channels. Immunohistochemistry should open new investigative scenarios allowing to demonstrate the localization and the distribution of SNTA1 [17].

86

V. Fineschi et al.

Fig. 4.9  Confocal laser scanning microscope: using polyclonal antibodies anti-Na + CP type Va (C-20) we have demonstrated the specific localization of this subtype of sodium channel to the intercalated disk and a nearly 50% reduction in Na + channels expression in ventricular myocytes (compared with control cases) in an young man who suddenly died in a family with electrocardiographic pattern typical of Brugada Syndrome

Due to the limited resolution of the light microscopical methods, the use of electron microscopy and confocal laser scanning microscopy analysis has to be implemented in cases of SCD when cardiac subcellular alterations have to be detected. Conventional transmission electron microscopy (TEM), scanning electron microscopy (SEM), and backscattered electron (BSE) emission can give a high-resolution image of the specimen’s intracellular structure showing pathological changes not found at light microscopy. SEM permits visualization of three-­dimensional features of cardiac myocytes with large visual fields. In addition to the three-dimensional advantages, the BSE imaging can obtain information from both cell surface and the intracellular myofibrils simultaneously [18, 19]. Myofibers’ structural changes which may represent the morphological substrate in many cases of SCD are very clearly visible and documented by ultrastructural electron microscopy view. We refer to disarrayed myocites with disorganized myofibrils and irregular Z lines in hypertrophic cardiomiopathy; short sarcomeres, highly thickened Z lines, myofibrillar rexis, and paradiscal lesions in contraction band necrosis; edematous myocell with few contracted myofibrils in colliquative myocitolysis.

4  Sectioning of the Heart, Searching for Pathology Under the Microscope…

87

Similarly, confocal laser scanning microscopy (CLSM) is a very valuable tool for obtaining high-resolution images in forensic pathology imaging. It allows greater resolution, optical sectioning of the sample and three-dimensional reconstruction of the same sample. The key feature of confocal microscopy is its ability to produce blurfree images of thick specimens at various depths. Images are taken point-by-point and reconstructed with a computer, rather than projected through an eyepiece. By combining additional technologies such as fluorescent lifetime imaging (FLIM)-based analysis, the capabilities of CLSM have been further extended so that now, in some cases, it is possible to speak of six-dimensional microscopy. The unquestionable value of confocal microscopy consists in the creation of optical ­sections: the laser beam ­sections the sample so that it is no longer necessary to prepare sections. Multidimensional view produced by high-resolution confocal microscopy system allows a detailed view of ultrastructural changes in cardiomyocites [20].

The Laboratory Technologies With the explosion of molecular techniques, deoxyribonucleic acid (DNA) testing on peripheral blood and tissues has revolutionized the diagnosis of genetic causes of sudden death. We describe the basic methods of tissue preparation and DNA ­analysis as a useful overview for the clinical pathologist and coroner. It is easy to amplify DNA that will be used for genetic testing when the DNA is taken from blood samples. Ideally, at the time of autopsy the coroner or pathologist collects 15 ml of blood in several tubes containing ethylenediaminetetraacetic acid (EDTA), which prevents coagulation and degradation of the DNA. The tubes are stored at 4°C until the DNA is extracted for analysis, which should be within 1 week, although sometimes we have extracted DNA 4 months after collection. If the blood samples are collected in tubes that do not contain an anticoagulant, the DNA should be extracted promptly (within days of the initial collection). Extraction of high-quality DNA from tissue that can be used for polymerase chain reaction (PCR) amplification is much more problematic than use of blood samples. It often is difficult to amplify long fragments of DNA from formalin-fixed and paraffin-embedded tissue because of the fixation time in the formalin, the often long storage time in the tissue blocks prior to analysis, and the formation of formic acid in the sample. Formic acid hydrolyzes the DNA and creates single-strand nicks in the DNA. In postmortem tissues fixed in nonbuffered formalin (usually in tissue preserved more than 20 years ago), DNA fragments longer than 90 base pairs cannot be amplified. A variety of published methods for extracting DNA from preserved tissue are available. Many involve a phenol-chloroform digestion and washing step. Commercial kits are available that may simplify the methodology. One article described a “pre-PCR restoration process” in which the singlestrand DNA nicks are repaired with Taq polymerase prior to PCR amplification, which greatly improved the length of DNA pieces that could be amplified. An alternative method for obtaining usable DNA from tissue is to snap-freeze fresh myocardial tissue collected at autopsy in liquid nitrogen and store at −80°C until

88

V. Fineschi et al.

DNA extraction is performed. Tissue processed in this manner can be used many months later. Clearly, collection and preservation of tissue or blood samples for future DNA ­analysis is cumbersome, time intensive, and costly. However, it is helpful for the pathologist or coroner to carefully preserve such biologic material for future DNA testing in cases where a genetic cause of sudden death is suspected, as outlined later. Affected persons could be identified and offered treatment while they are still alive [21]. Perhaps one of the most promising investigation tools in the field of cardiopathology is the study of proteomics. Since the term “proteome,” defined as the protein complement of a genome, was first coined by Wilkins working as part of a collaborative team at Macquarie and Sydney Universities (Australia) in 1995, proteomic studies are likely to provide new insights into cellular mechanisms involved in many heart disease and in sudden cardiac death [22]. Proteomic investigations of cardiovascular diseases, including dilated cardiomyopathy, atherosclerosis, and ischemia/ reperfusion injury, have identified candidate proteins altered with the pathologic states, so complementing past traditional biochemical observations [23, 24]. The causes of cardiac dysfunction in most heart diseases are still largely unknown, but are likely to result from underlying alterations in gene and protein expression. Proteomic studies are therefore likely to provide new insights into cellular mechanisms involved in cardiac dysfunction and may also provide new diagnostic. By enabling the separation of complex mixtures over numerous dimensions, exploiting the intrinsic properties of proteins, including charge state, molecular mass, and hydrophobicity, in addition to cellular location, the discrete alterations within the cell may be resolved. Proteomics-based studies are focused on the interactions of multiple proteins and their role as part of a biological system. Proteins are directly involved in virtually all cellular activities and as such influence cell phenotype and hence the tissue or organ. This phenotype varies under normal physiological conditions or in response to pathophysiological stresses. Proteomics is being applied in a number of other areas of cardiovascular medicine. Proteins have been identified that have the potential for use as novel markers of cardiac allograft rejection; proteomic analysis has been undertaken in cell culture models of hypertrophy and changes in protein expression profile have been examined in coronary arteries in relation to coronary artery disease progression, and formation of atherosclerotic plaques. Protein studies have revealed a number of proteins demonstrate altered expression during myocardial injury. In particular, troponin I and troponin T are currently used as markers of acute coronary syndromes based on the premise that the cellular ­contents of dead myocytes are released into the blood. Experimental studies have provided detailed molecular survey of global proteome changes linked to progressive dilated cardiomyopathy [25]. Proteomic analysis has been performed to identify possible HERG-interacting chaperones in Long QT syndrome [26]. Further application of proteomics investigation includes the study of cardiac hypertrophy, a well-known risk factor for QT-prolongation and cardiac sudden death [27]; the study of vulnerable plaques in coronary arteries which is thought to be associated to acute coronary syndromes, such as unstable angina, acute myocardial infarction, and sudden cardiac death [28]. Since there is a considerable body of

4  Sectioning of the Heart, Searching for Pathology Under the Microscope…

89

Fig. 4.10  Proteomic studies are likely to provide new insights into cellular mechanisms involved in many heart disease and in sudden cardiac death: the cycle study

evidence that common bacterial toxins, absorbed from the mucosal surface or delivered as part of a transient bacteremia, may have a pathogenic role in SIDS, the idea of toxin–gene interaction has been hypothesized and could be investigated using the new science of proteomics [29]. A detailed description of preparation of tissue specimens for proteomics technologies is beyond the scope of this paper. Briefly, analytical proteomics involves the incorporation of a vast array of new tools and technologies (Fig. 4.10). The basic overall approach to proteomic studies consists of sample preparation, protein separation, protein imaging, and identification. Mass spectrometry has become the technique of choice for protein identification as these methods are very sensitive, require small amounts of sample. In addition, expression of selected proteins may be confirmed by Western blot analysis and by immunohistochemistry (Fig. 4.11) [30].

Conclusions The substrate for SCD varies depending on the underlying structural heart disease, if any, and ranges from no obvious evidence of structural damage to advanced cardiomyopathic states [31]. SCD is the predominant cause of sudden death in young people, with structural cardiovascular abnormalities often evident at autopsy. However, not all SCD has an apparent attributable cause that can be determined at autopsy and the SCD is labeled as conventional autopsy-negative sudden unexplained death (SUD) [32]. The aims of forensic investigation must be direct to improve and maintain standards of autopsy investigation and to encourage ­subsequent analysis (toxicological, genetic, and microscopic) to formulate a

90

V. Fineschi et al.

Fig. 4.11  Cardiac expressions of IL1b (A-B) and IL6 (C-D) is confirmed by Western blot analysis in a sudden death case

d­ efinitive, objective diagnosis. Recent European guide-lines stated that the role of autopsy in sudden death must establish or consider: • Whether the death is attributable to a cardiac disease or to other causes of sudden death • The nature of the cardiac disease, and whether the mechanism was arrhythmic or mechanical • Whether the cardiac condition causing sudden death may be inherited, requiring screening and counseling of the next of kin • The possibility of toxic or illicit drug abuse and other unnatural deaths [33] Autopsy protocols actually represent no more than an official collection of information, certainly not capable ex nunc to ensure the improvement in competence of the single physician who uses them, since such protocols do not constitute alternatives to scientific knowledge, remaining procedures whose application can guarantee that this is exploited to the full extent of its potentialities. Actually, beyond the apparent rigidity of the standards and of the protocols, the only true limit to the use of the dissection experiment as effective tool for the “discovery of the truth” is represented by the lack of experience, preparation and imagination of the forensic pathologist, data that can be summarized in a single quality: skill.

4  Sectioning of the Heart, Searching for Pathology Under the Microscope…

91

All in all, it is necessary that the morphological observation, starting from the ­dissection of the corpse, becomes an increasingly more determined characterization and clearer documentation. Actually, the forensic pathologist may not be able to perform an autopsy and then he is an incomplete forensic pathologist. But when he performs an autopsy and presumes to be able to do it, he should do and should be able to do also a histological examination, which is and must stay the premise of every serious forensic study and evaluation in fatal sudden cardiac cases [3]. Finally, it is not possible to ignore particular techniques, based on imaging ­techniques which allow qualitative–quantitative evaluations of pathological entities that, until today, had been based, as Janssen cautions, only on “subjective” elements which much or little depend on the investigator’s interpretation. Moreover, we feel like to admonish how such investigations are often of fundamental importance for the whole postmortem examination, but with the understanding that such sophisticated methods should accompany and deepen but certainly not substitute, even partially, the postmortem examination as it is classically meant. The rarity with which, in autopsies, immunohistochemical techniques or sophisticated microscopic applications or proteomics techniques are used, which in other similar morphological– diagnostic fields are the rule, is absolutely to deprecate and constitutes, together with other obvious methodological distortions, the negative base why increasingly less autopsies are assigned to the forensic pathologist by the Courts.

References 1. Watson AJM, Benbow EW (2008) A surgeon’s guide to peri-operative death. Diagnostic Histopathol 15:27–32 2. Soilleux EJ, Burke MM (2008) Pathology and investigation of potentially hereditary sudden cardiac death syndromes in structurally normal hearts. Diagnostic Histopathol 15:1–26 3. Pomara C, Karch SB, Fineschi V (eds) (2010) Forensic autopsy: a handbook and Atlas. CRC Press, Boca Raton, London, New York 4. Turakia M, Tseng ZH (2007) Sudden cardiac death: epidemiology, mechanisms, and therapy. Curr Probl Cardiol 32:501–546 5. Fineschi V, Baroldi G, Silver MD (eds) (2006) Pathology of the heart and sudden death in forensic medicine. CRC Press, Boca Raton, London, New York 6. Ortmann C, Pfeiffer H, Brinkmann B (2000) A comparative study on the immunohistochemical detection of early myocardial damage. Int J Legal Med 113:215–220 7. Piercecchi-Marti MD, Lepidi H, Leonetti G et al (2001) Immunostaining by complement C9: a tool for early diagnosis of myocardial infarction and application in forensic medicine. J Forensic Sci 46:328–334 8. Martínez Díaz F, Rodríguez-Morlensín M, Pérez-Cárceles MD et  al (2005) Biochemical ­analysis and immunohistochemical determination of cardiac troponin for the postmortem diagnosis of myocardial damage. Histol Histopathol 20:475–481 9. Campobasso CP, Dell’Erba AS, Addante A et al (2008) Sudden cardiac death and myocardial ischemia indicators: a comparative study of four immunohistochemical markers. Am J Forensic Med Pathol 29:154–161 10. Xiao H, Chen Z, Liao Y et al (2008) Positive correlation of tumor necrosis factor-alpha early expression in myocardium and ventricular arrhythmias in rats with acute myocardial infarction. Arch Med Res 39:285–291

92

V. Fineschi et al.

11. Yao L, Huang K, Huang D et  al (2008) Acute myocardial infarction induced increases in plasma tumor necrosis factor-alpha and interleukin-10 are associated with the activation of poly(ADP-ribose) polymerase of circulating mononuclear cell. Int J Cardiol 123:366–368 12. Riezzo I, Pomara C, Neri M et al (2009) Myocardial tumor necrosis factor-alpha expression in a sudden death due to dilated cardiomyopathy with endomyoelastofibrosis. Int J Cardiol 134:e62–65 13. Turillazzi E, Pomara C, La Rocca G et al (2009) Immunohistochemical marker for Na + CP type Valpha (C-20) and heterozygous nonsense SCN5A mutation W822X in a sudden cardiac death induced by mild anaphylactic reaction. Appl Immunohistochem Mol Morphol 17:357–362 14. Turillazzi E, La Rocca G, Anzalone R et al (2008) Heterozygous nonsense SCN5A mutation W822X explains a simultaneous sudden infant death syndrome. Virchows Arch 453:209–216 15. Soeller C, Jayasinghe ID, Li P et  al (2009) Three-dimensional high-resolution imaging of cardiac proteins to construct models of intracellular Ca2+ signalling in rat ventricular myocytes. Exp Physiol 94:496–508 16. Marjamaa A, Laitinen-Forsblom P, Wronska A et  al (2011) Ryanodine receptor (RyR2) mutations in sudden cardiac death: Studies in extended pedigrees and phenotypic characterization in vitro. Int J Cardiol 147:246–252 17. Wu G, Ai T, Kim JJ et  al (2008) alpha-1-syntrophin mutation and the long-QT syndrome: a disease of sodium channel disruption. Circ Arrhythm Electrophysiol 1:193–201 18. Okabe M, Kanzaki Y, Shimomura H et  al (2000) Backscattered electron imaging: A new method for the study of cardiomyocyte architecture using scanning electron microscopy. Cardiovasc Pathol 9:103–109 19. Kanzaki Y, Terasaki F, Okabe M et  al (2009) Images in cardiovascular medicine. Threedimensional remodeling of cardiomyocytes in a patient with aortic stenosis: scanning electron microscopy. Circulation 119(2):e10 20. Turillazzi E, Karch SB, Neri M et al (2008) Confocal laser scanning microscopy. Using new technology to answer old questions in forensic investigations. Int J Legal Med 122:173–177 21. Turillazzi E, Glatter KA, Neri M (2006) Sudden cardiac death and channelopathies. In: Fineschi V, Baroldi G, Silver MD (eds) Pathology of the heart and sudden death in forensic medicine. CRC Press, Boca Raton, London, New York 22. McGregor E, Dunn MJ (2003) Proteomics of Heart Disease. Hum Mol Genet 12 Spec No 2:R135–144 23. McGregor E, Dunn MJ (2006) Proteomics of the heart: unraveling disease. Circ Res 98:309–321 24. White MY, Van Eyk JE (2007) Cardiovascular proteomics: past, present, and future. Mol Diagn Ther 11:83–95 25. Gramolini AO, Kislinger T, Alikhani-Koopaei R et al (2008) Comparative proteomics profiling of a phospholamban mutant mouse model of dilated cardiomyopathy reveals progressive intracellular stress responses. Mol Cell Proteomics 7:519–533 26. Walker VE, Atanasiu R, Lam H (2007) Co-chaperone FKBP38 promotes HERG trafficking. J Biol Chem 282:23509–23516 27. Kang YJ (2006) Cardiac hypertrophy: a risk factor for QT-prolongation and cardiac sudden death. Toxicol Pathol 34:58–66 28. Didangelos A, Simper D, Monaco C et al (2009) Proteomics of acute coronary syndromes. Curr Atheroscler Rep 11:188–195 29. Morris JA, Harrison L, Brodison A et al (2009) Sudden infant death syndrome and cardiac arrhythmias. Future Cardiol 5:201–207 30. De Souza AI, McGregor E, Dunn MJ et al (2004) Preparation of human heart for laser microdissection and proteomics. Proteomics 4:578–586 31. Maron BJ, Towbin JA, Thiene G et al (2006) Contemporary definitions and classification of the cardiomyopathies. Circulation 113:1807–1816 32. Carturan E, Tester DJ, Brost BC et  al (2008) Postmortem genetic testing for conventional autopsy-negative sudden unexplained death: an evaluation of different DNA extraction protocols and the feasibility of mutational analysis from archival paraffin-embedded heart tissue. Am J Clin Pathol 129:391–397 33. Basso C, Burke M, Fornet P et al (2008) Guidelines for autopsy investigation of sudden cardiac death. Virchows Archiv 452:11–18

Chapter 5

Endocrine Disorders with Potentially Fatal Outcome Lars Hecht

Abstract  Endocrine causes of sudden death are relatively rare compared to other internal causes. Thus, they may not be considered at autopsy and thus go undiagnosed. However, there is a large variety of endocrine disorders which can cause sudden, unexpected death and may thus gain forensic relevance. The post-mortem diagnosis of endocrine disorders is often difficult to establish. This chapter aims to provide an overview over those endocrine disorders which can cause sudden death and to illustrate their morphological and biochemical appearance. Keywords  Endocrinology • Metabolic crisis • Histopathology • Post-mortem biochemistry

Endocrine Disorders with Potentially Fatal Outcome Pituitary Coma Pituitary coma represents a severe course of pituitary insufficiency. Symptoms of hypopituitarism are dominated by a lack of ACTH and TSH with signs of secondary adrenal insufficiency and hypothyreosis (see below). Main symptoms are hypotension, pale skin, weakness, weight loss (ACTH deficiency), cold intolerance, sparse hair, constipation (TSH deficiency), reduced pubic hair (mostly in women), decreased libido, infertility, erectile dysfunction (gonadotropin deficiency). Pituitary coma results in severe hypotension, bradycardia, hypoventilation, hypothermia and

L. Hecht (*) Institut für Pathologie, HELIOS Klinikum Bad Saarow, Pieskower Street 33, 15526 Bad Saarow, Germany e-mail: [email protected] E.E. Turk (ed.), Forensic Pathology Reviews, Volume 6, Forensic Pathology Reviews 6, DOI 10.1007/978-1-61779-249-6_5, © Springer Science+Business Media, LLC 2011

93

94

L. Hecht

stupor. It may manifest after month or years as well, mostly in context of trivial infection or stress. General causes of hypopituitarism are pituitary tumours, craniocerebral injury, hypophysitis and cranial radiotherapy [1]. A particular cause of hypopituitarism is ischemic necrosis of adenohypophysis due to severe postpartum haemorrhage (Sheehan syndrome) which often becomes symptomatic with a failure of postpartum lactation [2].

Adrenal Crisis/Addison’s Disease Addison’s disease (primary adrenocortical insufficiency) is defined as an ­insufficient glucocorticoid and mineralocorticoid production as a consequence of destruction or atrophy of the adrenal cortex. Secondary adrenocortical insufficiency on the ­contrary is caused by a lack of glucocorticoids alone in the context of hypophyseal disturbances. Addison’s disease can remain subclinical for long time. Symptoms occur when more than 90% of the adrenal cortex is destroyed [3]. Addisonian crisis is typically precipitated by infection, stress, or in context of surgical interventions due to increased requirement of cortisol. It occurs acutely in adrenal apoplexy. The ­prevalence of Addison’s disease ranges from 30 to 60 cases per million with a maximum in the fourth and fifth decades [4]. Nevertheless, it may arise in childhood [5]. The main cause is autoimmune adrenalitis in industrialized countries and adrenal tuberculosis in developing countries [6, 7]. The disease is characterized by weight loss, vomiting, weakness, hypotension, loss of pubic hair (mostly in women), and increased skin pigmentation. The latter is absent in secondary adrenocortical insufficiency. A so-called white Addison’s disease is known mostly in individuals with skin type I (red or blond hair, pale skin) [8, 9]. Major symptoms of acute ­adrenal crisis are severe hypotension, fever, abdominal pain, and circulatory ­collapse [10]. The knowledge of circumstances of death is useful to confirm the post-mortem diagnosis. Secondary morphological changes should necessarily be included (see below).

Conn’s Syndrome Main symptom of Conn’s syndrome (primary hyperaldosteronism) is arterial hypertension which may be resistant to usual pharmacological therapy. Hypokaliemia is occasionally present and may induce muscle weakness, convulsion and polyuria [11]. Main cause of high aldosteron levels is bilateral hyperplasia of adrenal cortex (about 60%) followed by aldosteron-producing adenoma (about 35%) [12]. Mortality is affected by cardiovascular changes as a consequence of hypertension. Abdo et al. [13] described a case of primary hyperaldosteronism presenting as sudden death due to ventricular fibrillation.

5  Endocrine Disorders with Potentially Fatal Outcome

95

Thyrotoxicosis Thyrotoxicosis is a clinical syndrome resulting from excess of circulating thyroid hormones which develops in about 1% of pre-existing hyperthyroidism [14]. General features of increased thyroid function are among others hair loss, weight loss, warm and moist skin, tremor, palpitations, anxiety, osteoporosis and proximal myopathy. Depending on the stage of thyrotoxicosis the symptoms vary from tachycardia, hyperthermia, and tremor, diarrhoea to psychosis, convulsion, stupor and coma [15]. Congestive right heart failure with peripheral oedema is noted, as well as acute myocardial infarction in spite of unremarkable coronary arteries [14, 16–19]. In the latter, coronary vasospasms are assumed [20]. Mortality is up to 50% even if therapy is provided [21]. Precipitating factors in pre-existing hyperthyroidism are iodine contamination, surgical intervention, traumata, trivial infections or insufficient thyreostatic therapy [14]. An accidental or intended intake of inappropriate high dosages of thyroxin must be considered [22]. The antiarrhythmic agent amiodarone is known to cause thyroid dysfunction, hypothyroidism, as well as thyrotoxicosis [23–25].

Myxeodema Coma Myxoedema coma is a rare clinical syndrome developing from inappropriate low levels of thyroid hormones in pre-existing hypothyroidism. Because of different definition of the disease and difficulties in post-mortem diagnosis the exact actual prevalence is unknown. Typical features of hypothyroidism are slowing of activity, weight gain, and dry hair, loss of outer third of eyebrows, myxoedemic face, ptosis, dry and cold skin and constipation. Myxoedemic face is a result of accumulation of mucopolysaccharides in dermal connective tissue. Precipitating factors as infection, surgery, chronic thyroiditis or drugs (e.g. lithium, amiodarone) may lead to bradycardia, psychosis, hypothermia and shock [26–28]. Disturbances in heat balance may result in life-threatening situations. Myxoedema coma develops most often in older women (>60 years of age) during cold season [29, 30]. Mortality is up to 20% even if therapy is provided [31].

Parathyrotoxic Crisis The extreme rare parathyrotoxic crisis develops from chronic hyperparathyroidism (mainly primary and tertiary hyperparathyroidism) but also in vitamin D poisoning and malignant tumours (paraneoplastic hyperparathyroidism). In the latter a parathyroid-hormone-like peptide (PTHrP) is synthesized by several malignancies (mostly breast- and lung cancer). Chronic hyperparathyroidism may lead to osteomalacia,

96

L. Hecht

nephrocalcinosis, vessel- and soft tissue calcification [32]. Hypercalcaemia may exacerbate in life-threatening hypercalcaemic crisis in ­presence of dehydration, immobilisation and drugs (e.g. theophylline, thiazide) [33, 34]. Main symptoms are tachycardia, vomiting, constipation, muscle weakness, confusion and coma [35]. In 30–40% of patients a primary hyperparathyroidism and in 50% a malignancy is present. The prevalence is about 2 of 1,000 hospitalised patients [36]. Cases in childhood and adolescence have been reported [37].

Hypoparathyroidism Hypoparathyroidism may be congenital or in relation to autoimmune inflammatory lesions of parathyroid glands [32]. To our knowledge, cases of sudden unexpected death in hypoparathyroidism are not reported so far. However, Brown et  al. [38] described a case of severe hypocalcaemia with lethal outcome in EDTA-therapy for lead intoxication.

Morphological and Post-mortem Biochemical Findings in Endocrine Disorders with Potentially Fatal Course Pituitary Gland An autopsy study by Saeger and Hanke [39] has noted necroses of anterior pituitary in 21% of cases of clinical manifest circulatory shock. As referred above especially circulatory shock around the time of delivery may cause necroses of anterior pituitary. Necroses of neurohypophysis are more uncommon but also mostly related to shock [39]. Another, in the past often underestimated cause of necroses is craniocerebral injury. Histologically, early stages of necrosis are not easily recognisable and can be misinterpreted as autolytic changes. Pituicytes are dissociated from basallamina, becomes round and more eosinophilic with pale nuclei (Fig. 5.1). Later on the necrotic tissue becomes amorphous, granular and acellular apart from invading granulocytes. Finally, the tissue repairs itself with the formation of a scar (Fig. 5.2). Necroses of the neurohypophysis show oedema in early stages followed by a clearing reaction and a residual glial scar [40]. Inflammatory lesions of the pituitary gland are a heterogeneous group of diseases including lymphocytic hypophysitis, granulomatous and xanthomatous hypophysitis (primary hypophysitis) and hypophysitis resulting from systemic diseases like infection (secondary hypophysitis). Chronic lymphocytic hypophysitis is the most common among the primary chronic inflammations of the pituitary [41]. It presents most often in women during the peripartum period [42, 43]. However, there have been reported some cases of this pathological condition in men [44, 45]. Lymphocytic

5  Endocrine Disorders with Potentially Fatal Outcome

97

Fig. 5.1  Early stage of pituitary necrosis with rounded, dissociated pituicytes and pale nuclei (Hematoxylin and Eosin)

Fig. 5.2  Cicatrisation of pituitary gland with residual nests of pituicytes – Sheehan syndrome with sudden lethal course 20 years after delivery (hematoxylin and eosin)

hypophysitis has been attributed to an autoimmune process since it is frequently associated to chronic inflammation of other endocrine organs (Hashimoto thyroiditis, chronic adrenalitis) [42, 46]. Histological examination reveals diffuse and rarely follicular infiltration by lymphocytes, fibrosis and loss of pituitary cells. Also

98

L. Hecht

Fig. 5.3  High power micrograph of pituitary adenoma. Note anisonucleosis and prominent nucleoli (hematoxylin and eosin)

neurohypophysis may be affected by the chronic inflammation resulting in diabetes ­insipidus [47]. Primary hypophysitis usually causes symptoms due to ­deficits of the hormones produced by the adenohypophysis. Some cases with fatal course have been reported [48–50]. Secondary hypophysitis is associated with ­systemic disorders as generalised infection (e.g. tuberculosis [51]). Even various opportunistic infections like cryptococcosis, cytomegalovirus infection and toxoplasmosis may involve the pituitary of individuals suffering from acquired immune deficiency syndrome (AIDS) [52]. An extremely rare but potentially fatal pathologic condition of pituitary is Wegner’s granulomatosis. Histologically dominates a small-vessel vasculitis accompanied by diffuse granulomatous and necrotising inflammation [53]. Pituitary gland tumours can be detected in autopsies in up to 26% of cases and represent mostly benign lesions [54]. Pituitary adenomas arise in every age group but most often between the age of 40 and 50 [55]. A case of sudden death due to pituitary adenoma in a 3-month-old female infant was reported by Matsuura [56]. Functional disturbances results from tumour growth with compression of adjacent structures on the one hand and hormonal activity on the other (e.g. Cushing ­syndrome, acromegaly) [57]. In the majority of cases, immunohistochemistry is required to classify pituitary adenoma in respect of hormonal content (Fig. 5.3). A severe complication of pituitary adenoma is acute haemorrhage and infarction of the tumour (pituitary apoplexy). It is reported to occur in about 17% of the adenoma independent from subtype and hormonal activity [58]. Cases with fatal outcome have been reported [59, 60]. A specific cardiomyopathy is described in acromegaly (STH-adenoma) [61, 62]. Morphological changes include myocardial hypertrophy, increased content of ­collagen fibres, degenerative and inflammatory changes of sinus node and A-V node region [63, 64].

5  Endocrine Disorders with Potentially Fatal Outcome

99

Fig. 5.4  Abundant lymphocytic infiltration of the adrenal gland in autoimmune adrenalitis (hematoxylin and eosin). The infiltrate predominantly consists of T lymphocytes (inset; Immunohistochemistry: CD3)

In contrast to tumourous conditions, an extremely rare cause of pituitary h­ ypofunction is reported by Saeger et  al. [65] presenting as connatal ACTH cell aplasia in a 5-year-old boy with subsequent hypoglycaemia and cerebral seizures with fatal outcome.

Adrenal Glands Adrenal apoplexy predominantly arises in infants and newborns as a result of trauma during delivery or infection. Acute haemorrhagic necroses in the course of meningococcal septicaemia are known as Waterhouse–Fridrichsen Syndrome [66, 67]. In recent years, even cases of adrenal apoplexy in adulthood have been reported [68]. Heparin-induced thrombocytopenia (HIT) is known to cause adrenal haemorrhage rarely [69]. Confirming the diagnosis is quite easy. The adrenal cortices are necrotic while the medulla contains haemorrhagic substrate. As referred above, 90% of adrenal tissue must be destroyed in order that symptoms occur. Nowadays the most common cause of Addison’s disease is autoimmune adrenalitis (about 70% of cases). Induced by chronic lymphocytic infiltration the adrenals undergo severe atrophy with subsequent hormonal insufficiency (Fig.  5.4) [70]. Adrenocortical atrophy is an important finding since lymphocytic infiltrates are also seen occasionally in the adrenals of the elderly [71]. Autoimmune adrenalitis may be associated with other autoimmune diseases such as thyroiditis or hypophysitis and may lead to sudden death [5, 9, 72–74].

100

L. Hecht

Fig. 5.5  (a) Macroscopic appearance of adrenocortical adenoma. (b) Adrenocortical adenoma margined by a slight capsule of connective tissue and atrophic residual adrenocortical tissue (hematoxylin and eosin)

Among generalised infective diseases tuberculosis was a very common cause of Addison’s disease in the past. Nowadays, it is quite rare in industrialised countries [75]. Cytomegalovirus-associated adrenal damage in a patient suffering from HIV was reported by Uno [76]. Histologically, typical nucleal inclusions were present in the adrenal cortex. The incidence of adrenocortical adenoma is unknown. Most of them are found at autopsy or computed abdominal imaging as so-called incidentalomas. 85% of ­adenomas develop hormonal activity. The tumours usually show a yellow to brownish cut surface and smooth texture (Fig. 5.5a). The cells are enlarged with mostly clear cytoplasm (high content of fatty substrate) and are arranged in cords, nests or compact structures (Fig.  5.5b). Eosinophilic cytoplasmatic inclusions may be

5  Endocrine Disorders with Potentially Fatal Outcome

101

Fig. 5.6  Phaeochromocytoma of only 10  mm in maximal diameter. Note cellular arrangement (inset) (hematoxylin and eosin)

p­ resent in primary hyperaldosteronism during drug therapy (spironolactone ­bodies). Cortisol-producing adenomas may induce atrophy of zona fasciculata and zona reticularis of ipsi- and contralateral adrenal gland. In the presence of Conn-adenoma atrophy arises in zona gomerulosa. Structural or immunohistochemical findings are not reliable in differentiating functioning from non-functioning adenoma [11]. Distinguishing adrenocortical noduli (which are mostly reactive), cortical hyperplasia and adenoma is quite difficult. Congenital enzyme defects of the pathways of steroid synthesis may result in adrenocortical hyperplasia. The commonest example is 21-hydroxylase deficiency. Affected individuals suffer from a lack of mineralo- and corticosteroids and hypersecretion of adrenal sex steroids. Hormone replacement is essential to avoid lifethreatening “salt-wasting” crisis [77]. Pheochromocytomas are derived from chromaffin cells of the adrenal medulla. The incidence ranges from 0.4 to 9.5 per million [78]. Autopsy studies report a frequency of 0.05% [79, 80]. The tumour arises preferentially between the age of 40 and 50. Pheochromocytomas are mostly encapsulated and reach a maximum diameter of 3–6 cm [81]. The colour of the cut surface varies from reddish to yellow and is changing to brownish when the tissue is immersed in formalin. Haemorrhages and cysts may be present in bigger tumours. Pheochromocytomas histologically consists of large, polygonal cells with weakly eosinophilic to basophilic even ­granular cytoplasm and large, round nuclei with mostly distinct nucleoli (Fig. 5.6). The architecture ranges from alveolar to trabecular. A meshwork of sustentacular cells is developed. The tumour cells are immunohistochemically positive for ­chromogranin A, whereas the sustentacular cells show a positive reaction for S100 protein. Only  about two-third of the tumours show endocrine activity [81].

102

L. Hecht

Fig. 5.7  Chronic follicular lymphocytic infiltration of the thyroid gland and oxiphilic metaplasia (inset) in Hashimoto’s thyroiditis (hematoxylin and eosin)

Pheochromocytomas may lead to sudden death due to catecholamine excess with subsequent hypertensive crisis, cerebral haemorrhages and cardiogenic shock [82–86]. Occasionally noticed ischemic damage of cardiomyocytes has been explained by coronary vasospasms [87]. Symptoms include headache, chest pain, nausea and palpitations. Pheochromocytomas in pregnancy with lethal outcome have been described [88]. Extraadrenal sympathetic paragangliomas arise from paraaortic paraganglia (organs of Zuckerkandl) and show morphology and symptoms similar to pheochromocytomas. Paragangliomas more often produce norepinephrine and may rarely cause sudden unexpected death due to catecholamine excess [89–91].

Thyroid Gland Hashimoto’s thyroiditis (chronic lymphocytic thyroiditis) initially causes thyroid enlargement and hyperthyroidism. It shows a female predominance and mostly set in between the 40th and 60th year of age. Hashimoto’s thyroiditis is an autoimmune disease with autoantibodies reacting with thyroglobulin (Tg-Ab) and thyroperoxidase (TPO-Ab). Macroscopically, the thyroid gland appears uniformly pale and fleshy. In early stages of disease, a dense lymphocytic infiltrate with germinal centres is present histologically (Fig. 5.7). Thyroid follicles are partly reduced and typically show enlarged eosinophilic epithelia (oxiphilic metaplasia). In advanced stages the gland is captured by fibrosis and undergoes atrophy (Fig.  5.8) with concomitant hypothyroidism [92]. Two interesting cases of sudden death due to Hashimoto’s ­thyroiditis have been reported by Siegler [93] and Lorin de la Grandmaison [94].

5  Endocrine Disorders with Potentially Fatal Outcome

103

Fig. 5.8  Residual fibrosis of the thyroid gland in advanced stage of Hashimoto’s thyroiditis. Note oxiphilic metaplasia on the left lower margin (hematoxylin and eosin)

The most frequent cause of adult hypothyroidism is atrophic autoimmune t­hyroiditis (primary myxoedema) which is characterised be severe atrophy. Thyroid shows dense texture with pale cut surface. Only few follicles remain, oxyphilic metaplasia is lacking and a slight lymphocytic infiltrate is present [92]. Atrophic autoimmune thyroiditis must be distinguished from advanced stage of Hashimoto’s thyroiditis and the extremely rare Riedel’s thyroiditis. The latter even involves extrathyroidal structures. The granulomatous thyroiditis (de Quervain) is a selflimiting disease with minor impact on hormone production [92]. As referred above amiodarone may induce severe disturbances of thyroid function. In this context, an inflammatory reaction with epithelial necroses, lymphocytic, and foamy cell infiltration accompanied by fibrosis may occur [24]. The most common cause of hyperthyroidism is Grave’s disease (Morbus Basedow) which likewise develops against the background of autoimmune processes with stimulating autoantibodies reacting with TSH-receptor (long-acting thyroid stimulator – LATS). The disease is more common in women and tends to occur in patients between the 30th and 40th year of age. Goitre and exophthalmos are typical features. Histology reveals lymphocytic infiltration, micropapillary hyperplasia of follicular epithelia, reduction of stored colloid and retraction vacuoles (Fig. 5.9). The latter is an artefact produced due to tissue preservation. A coincidence of Grave’s disease and myxomatous mitral valve prolaps resulting from glucosamineglycane storage has been reported [95]. The stored substances can be visualised by alcian stain (Fig. 5.10). Functioning “toxic” adenoma is responsible for hyperthyroidism in one-third of cases. Histologically, the follicles are margined by tall and often papillary ­proliferated epithelia. Retraction vacuoles may be present. Regressive changes as haemorrhages,

104

L. Hecht

Fig. 5.9  Micropapillary epithelial proliferation, abundant retraction vacuoles and slight lymphocytic infiltration in Grave’s disease (Morbus Basedow) (hematoxylin and eosin)

Fig. 5.10  Storage of acid mucopolysaccharides in the mitral valve (Alcian blue staining)

fibrosis and calcification are frequent. The tumour is enclosed in a fibrous capsule. Pre-existing thyroid tissue is mostly atrophic because of increased hormone production of the adenoma [92, 96]. In nodular goitre one or more nodules with increased activity may develop. The nodules resemble “toxic” adenoma. Nevertheless, this condition represents an own

5  Endocrine Disorders with Potentially Fatal Outcome

105

Fig. 5.11  Autopsy appearance of the thyroid gland, left lower parathyroid gland (blue arrow) and lymph node for comparison (green arrow)

entity which is termed “toxic” nodular goitre (Plummer’s disease). This lesion may remain clinically inconspicuous for long time, but application of iodine-containing agents may give rise to thyrotoxic exacerbation [97]. “Black thyroid” should be mentioned as a special morphologic feature in this context. It is pathognomonic for chronic minocycline ingestion. An influence on thyroidal hormonal activity is arguable [98, 99].

Parathyroid Glands Most individuals have four parathyroid glands located on the posterior surface of the thyroid (upper pair) and near the lower pole of the thyroid (lower pair; Fig. 5.11). The weight of a single gland should not exceed 60 mg. Because of the varying content of fat cells the glands show a brownish colour in adults [32]. Three cell types can be distinguished – chief cells, water-clear cells and oxyphile cells (Fig. 5.12). Sudden death related to dysfunction of parathyroid glands is extremely uncommon. Therefore, only a few pathological conditions will be shortly described. Main cause of hyperparathyroidism is parathyroid adenoma (80% of cases). The cells of adenoma show slightly enlarged nuclei. Fat cells are lacking. The lesion is completely encapsulated [100, 101]. In 15% of cases hyperparathyroidism is caused by chief cell hyperplasia which occurs often in more than one single gland. Several fat cells may be present between the closely arranged chief cells. A pseudocapsule of connective tissue is inconstantly developed [101].

106

L. Hecht

Fig. 5.12  Normal histologic appearance of the parathyroid gland (hematoxylin and eosin)

The most uncommon cause of hyperparathyroidism is parathyroid carcinoma (3% of cases). Patients are mostly over 50 years of age but even infants may be affected [101]. Carcinoma induces usually higher calcium levels compared to ­adenoma and hyperplasia [102]. The lesion is characterised by a dense texture and a pale cut surface. Cellular atypia and infiltrative growth behaviour is seen histologically [103]. Metabolic changes are usually of higher impact on the clinical course than tumour progression [32].

The Endocrine Pancreas The endocrine pancreas is the central organ for the regulation of blood glucose level. The below described pathological processes of the pancreas may cause hypoglycaemia through inadequate high insulin secretion. Morphological changes of pancreas due to diabetes mellitus will be reviewed separately. The symptoms of hypoglycaemia can be grouped in two categories: neurological symptoms caused by neuroglycopenia and autonomic nervous system response by increased catecholamine release. Main symptoms are blurred vision, confusion and loss of consciousness, accompanied by sweating, hunger, tremor and nausea. Continuing severe hypoglycaemia causes unspecific histomorphological changes as cortical neuronal damage, perivascular and pericellular oedema, disseminated microhemorrhages and dissociation of myelin sheath [104, 105]. Insulinoma is the most common functioning endocrine tumour of the pancreas with an incidence of 2–4/million population/year. Insulinomas arise in every age

5  Endocrine Disorders with Potentially Fatal Outcome

107

Fig. 5.13  Cellular arrangement of insulinoma (Hematoxylin and Eosin)

group but mostly in adults of 40–60 years of age [106]. It is mainly found in ­pancreas but as well in intestine, gastric wall, hilus of the spleen and gastrosplenic ligament [106]. The tumour is well circumscribed with brownish cut surface and smooth texture. Its size is rarely more than 2 cm [107]. Bigger tumours may show regressive changes. A malignant transformation is possible. Insulinomas consists of middle-sized cells in trabecular and acinar growth pattern with round hyperchromatic nuclei (Fig. 5.13). Intracellular amyloid deposition is found in 3–5% of the tumours (Congo red staining). Its major component, islet amyloid polypeptide (IAPP), is visualizable immunohistochemically [108]. Insulin and Proinsulin can be detected in cytoplasm of the tumour cells by immunohistochemistry as well [109, 110]. The immunoreactivity of insulin does not correlate with circulating insulin levels. Nesidioblastosis is an uncommon cause of hyperinsulinemic hypoglycaemia. It arises mostly in newborns (10% of cases of hyperinsulinemic hypoglycaemia) [109]. However, in recent years increasing number of cases of nesidioblastosis in adulthood were reported [111–113]. “The pancreas is mostly of normal size, rarely of increased consistency” [109]. Pancreatic islets can be enlarged to more than 300 mm (normal: 100–200 mm). Distribution of different cell types within the islets remains unremarkable. The islet cells are partly swollen with enlarged nuclei showing distinct nucleoli [114, 115]. Ectopic islets arise occasionally in peripancreatic tissue [116]. Nesidioblastosis can be distinguished from acinar cell hyperplasia, which is a quite common and sometimes similar looking feature, by immunohistochemistry [117].

108

L. Hecht

Fig. 5.14  Islet amyloidosis of the pancreas in advanced stage of diabetes mellitus type 2 (Kongo red staining)

Diabetes Mellitus/Coma Diabeticum Diabetes mellitus-related changes of the pancreas are dependent on the type of ­diabetes and duration of the disease. In early stages of type 1 diabetes (insulindependent diabetes) an insulitis may occur. Some islets are affected by diffuse lymphocytic infiltrates. Others show an unremarkable arrangement or lack of beta cells. An autopsy study of under 20-years old diabetics showed an insulitis in 78% of cases with recent onset disease [118]. In advanced stages, an inflammation within the islets is missing. The count of beta cells decreases continuously. Löhr and Klöppel [119] found a rate of insulin-positive cells of near 40% with diabetes duration of more than 11 years. The exocrine pancreas undergoes atrophy. The average weight may be reduced to 40–50 g (normal: 80–90 g) [109]. Beta cells are decreased in type 2 diabetes (non-insulin-dependent diabetes) while an insulitis is lacking. Amyloidosis of the islets (Fig.  5.14) is present in 70–96% of cases [109, 120]. The deposits consist of islet amyloid polypeptide (IAPP; see also: insulinoma) which is secreted together with insulin under physiological conditions. Therefore, an amyloidosis of islets is lacking in type 1 diabetes. In diabetic nephropathy IAPP was also found in mesangium, Bowman’s capsule and Kimmelstiel–Wilson lesions [121]. As a consequence of the disease secondary histomorphological changes occur amongst others in liver and kidneys. In 1883, Ehrlich first described intra-nuclear depositions in liver tissue [122]. In 1906, Meixner could demonstrate glycogen as the substrate of these nuclei which are found in liver tissue from diabetics [123]. After conventional preparation of the tissue the nuclei seemed to be empty with

5  Endocrine Disorders with Potentially Fatal Outcome

109

Fig. 5.15  Empty nuclei of liver epithelia in diabetes mellitus type 2 (hematoxylin and eosin). Nuclear glycogen inclusions after fixation in absolute ethanol (inset) (PAS staining)

distinct nuclear membrane (Fig. 5.15). After fixation in highly concentrated ethanol the intra-nuclear substrate is strongly positive for periodic acid – Schiff staining (Fig. 5.15, inset). Intra-nuclear glycogen deposits in hepatocytes are not specific for diabetes mellitus but can be found in 47% of cases [124, 125]. They also appear in liver cirrhosis, non-alcoholic steatohepatitis and viral hepatitis [126, 127]. After Sellyei and Walton [128] the count of affected nuclei is higher in diabetics and increases with age (the latter apply for non-diabetics as well). There is no correlation between occurrence/intensity of intra-nuclear glycogen and a special therapy (oral anti-diabetics, diet and insulin). A further micromorphological phenomenon in diabetes mellitus is nodular glomerulosclerosis Kimmelstiel–Wilson (Fig. 5.16). It appears in both type 1 and type 2 diabetes. Dependent from duration of disease it is present in up to 70% of cases [129]. The nodules develop from a diffuse glomerulosclerosis adjacent to hyalinisation of glomerula and arterioles (Literature at Helmchen and Schubert [130]). Nodular glomerulosclerosis is also not specific for diabetes since an idiopathic glomerulosclerosis is described for non-diabetics [131, 132]. Nowadays, diabetics mostly die from secondary vascular changes. Coma diabeticum as cause of death is rather uncommon. A micromorphological sign of coma diabeticum are glycogen storing tubulus cells in the kidneys, so-called Armanni–Ebstein cells. They arise in the erected part of proximal tubules. The cytoplasmatic deposits are positive for periodic acid Schiff staining after fixation in highly concentrated ethanol. Glycogen may partly be found in the lumina of tubules. After fixation in formalin and haematoxylin–eosin staining the cells exhibit a water clear, vacuolated cytoplasm (resembling a plant cell) with unremarkable nucleus. Armanni–Ebstein cells are highly specific for coma diabeticum [133, 134].

110

L. Hecht

Fig. 5.16  Nodular glomerulosclerosis Kimmelstiel–Wilson. Also note the arteriolosclerosis (PAS staining)

Post-mortem Biochemistry Biochemical examinations are an essential supplement to autopsy and histology in cases of suspected endocrine death. About the field of post-mortem biochemistry Coe [135] gives a comprehensive overview. In the following, some important information to single substances will be offered. Levels of chemical substances such as hormones, carbohydrates and neurotransmitter change within certain time postmortem. Therefore, the exact time of death must be known. Generally, samples must be taken before the onset of decomposition (as early as possible). Blood should be obtained preferably from peripheral veins (e.g. femoral vein) before haemolysis set in. For analytical purposes, vitreous humour should be crystal-clear and obtained under avoidance of any contamination with tissue fragments.

Thyroid Hormones Levels of free thyroid hormones (fT4, fT3) reflect the thyroid function in living individuals and should be determined also in post-mortem diagnostic. Rachut et al. [136] found much higher levels of thyroid hormones in blood from inferior vena cava compared to femoral vein suggesting diffusion of hormones from the decomposing gland. Therefore, samples should be taken from peripheral veins. Vitreous humour is an inappropriate source because hormone levels do not correlate well with blood hormone levels [137].

5  Endocrine Disorders with Potentially Fatal Outcome

111

According to Coe [138] and Rachut [136] thyroxine levels tend to fall after death. This decline began at the earliest 2.75 h after death [136]. High thyroxine levels can be of use in diagnosing antemortem hyperthyroidism. Triiodthyronine levels showed an irregular course of increase in some individuals and decline in others. The increase of triiodthyronine can be explained by a conversion of thyroxine (T4->T3) after death. It is important to know that according to Bonnell [139] thyroxine levels may fall during antemortem period in individuals with prolonged agony. Also drug therapy may alter thyroid hormone levels. Lithium, phenylbutazone and sulfonylurea inhibit hormone synthesis, whereas salicylates cause a decreased hormone-­ protein binding [140]. TSH levels have been found to be stable in serum within 24 h after death [138] and correlate well with antemortem levels. TSH may penetrate into vitreous humour in some individuals [141]. However, hormone levels in vitreous humour do not correlate well with serum levels. In conclusion, post-mortem thyroid hormone levels should be interpreted with caution. Further investigations are strongly necessary to make post-mortem diagnosis of thyroid function more significant.

Cortisol In his survey of 20 adult individuals Finlayson [142] found a small and not significant decrease of serum cortisol levels (mean 3.7 mg/100 ml) within the first 18 h after death. Serum cortisol levels among a group of 15 infants (3 month to 3 years) were found to be comparable to those of the adult group. The concentration of cortisol was independent of the sampling site (cardiac blood, peripheral blood). Cortisol distribution is influenced by internal and external factors. Drucker and Shandling [143] showed that patients with acute medical illnesses have considerably higher levels. Furthermore, it must be taken into account that cortisol levels follow a circadian rhythm with high levels in the morning and lower levels during the night.

Catecholamines Post-mortem levels of catecholamines and their metabolites are very variable depending on the cause and circumstances of death. Catecholamine levels are ­supposed to be increased in hypothermia (urine levels) and suffocation (blood ­levels) [144, 145]. Tormey et al. [146] found elevated urine levels of adrenaline and noradrenaline in patients dying from myocardial infarction, alcohol poisoning, ­septicaemia and cerebral haemorrhage. Also l-dopa therapy may lead to high levels of catecholamines and their metabolites in urine [147]. Referring to phaeochromocytoma it is unreliable in assuming sudden death by catecholamine excess when circumstances of death are unknown.

112

L. Hecht

Adrenocorticotropic Hormone Ishikawa et al. [148] analysed post-mortem serum ACTH levels in different causes of death and found them similar to the clinical reference value for sharp instrument injury, fire fatality, drowning and hypothermia but lower in hyperthermia, asphyxia and drug poisoning. Cerebrospinal fluid levels of the same individuals were mostly higher compared to the serum levels. ACTH concentration was found to be independent of post-mortem time. Therefore, it may be a useful marker of pituitary function in cases of sudden death from endocrine disorders.

Blood Glucose Metabolism As a result of post-mortal glycolysis blood glucose levels decline rapidly in peripheral blood, whereas post-mortal hepatic glycogenolysis and diffusion leads to an increase of glucose levels in cardiac blood. A certain number of pathological conditions may cause high glucose levels also in peripheral vessels [149–151]. Therefore post-mortem diagnosis of diabetes mellitus and related metabolic crises should not be made from interpretation of blood glucose. Haemoglobin A1c is a helpful parameter to ensure the post-mortem diagnosis of diabetes mellitus. Normal values do not exceed 6.5%. According to Kernbach [152] values exceeding 12.1% are indicative for coma diabeticum. Ritz and Kaatsch [153] already suggest levels of more than 10%. Haemoglobin A1c has turned out to be quite resistant to autolysis for days [152]. To estimate antemortem blood glucose levels also today the use of the sum value of glucose and lactate according to Traub has been recommended as the most appropriate tool [154, 155]. This method is based on the fact that glucose in the fluids of the cadaver is converted into lactate and should be reliable for up to 105 h (for cerebrospinal fluid). Vitreous humour and cerebrospinal fluid (CSF) are applicable substrates because degradation of glucose takes a longer time compared to peripheral blood. Normal glucose levels of CSF range from 50 to 80 mg/dl (60–70% of normal blood glucose). The speed of post-mortem glucose degradation in CSF is approximately 10–15 mg/dl/h. That means if glucose is detectable in CSF 10–12 h post-mortem hyperglycaemia must be assumed. Lactate levels in CSF rise after death in the same manner as glucose levels decrease with a speed of 10–15 mg/dl/h. Normal levels range from 10  mg/dl in children and young adults to 20  mg/dl in older individuals but may be elevated in chronic inflammation of CNS, tumours and abrosia. Traub [155] determined a lower threshold level for CSF-glucose in lethal coma diabeticum of 362 mg/dl. Kernbach et al. [152] proposed a value of 415 mg/dl. Glucose levels of corpus vitreum range between 50 and 85% of blood glucose levels [156]. The sum value for non-diabetics should not exceed 277 mg/dl. Values exceeding 410 mg/dl may indicate diabetes mellitus with fatal outcome [157]. It is important to know that lactate levels of vitreous humour in the earliest post-mortem interval may range between 80 and 160 mg/dl and may increase up to 260 mg/dl

5  Endocrine Disorders with Potentially Fatal Outcome

113

after 20 h post-mortem [156]. The above-mentioned values must be interpreted with care in hospital patients because several diseases, drugs and infusions may produce abnormal high glucose levels [153]. Interpretation of glucose levels in urine requires special care because glucosuria not only occurs as a result of diabetes mellitus but also in other pathological conditions such as intoxication and leukaemia. On the other hand, lacking glucosuria does not exclude diabetes mellitus. Urine-glucose levels in lethal coma diabeticum mostly exceed 500 mg/dl [156]. Ketoacidotic coma typically develops in diabetes mellitus type 1 (unlike the hyperosmolar coma, which is typical for diabetes type 2) and is characterised by elevated levels of acetone and beta-hydroxybutyrate. Blood acetone levels range between 2.3 and 2.5 mg/l in non-diabetics and exceed 20 mg/l in diabetics. Levels of 5 mg/l and more in CSF are suspicious for diabetes mellitus. Ketosis may also occur as a result of alcoholism and abrosia [156]. It is a predisposing factor for lactate acidosis. Determination of lethal hypoglycaemia still remains problematic. Hypoglycaemia may be caused by over-dosage of insulin or sulfonylurea (external causes) or endogenous pathological conditions such as insulinoma and nesidioblastosis (as referred above). Sum values of less than 50 mg/dl in CSF or 100 mg/dl in vitreous humour may indicate antemortem hypoglycaemia [156, 158]. To exclude lethal application of insulin the hormone and c-peptide levels (which is low in exogenous insulin administration) should be measured in peripheral blood as soon as possible after death. In addition, the determination of insulin in CSF and vitreous humour by radio-immunoassay can be performed [159]. The estimation of insulin on the injection site (if identifiable) may be useful.

References 1. Herrmann BL, Mann K (2003) Hypophyseal coma. Internist 44(10):1253–1259 2. Sheehan HL (1937) Post-partum necrosis of the anterior pituitary. J Pathol Bact 45:189–214 3. Liddle GW (1981) The adrenals. In: Williams RH (ed) Textbook of endocrinology, 6th edn. WB Saunders, Philadelphia 4. Martín Martorell P, Roep BO, Smit JW (2002) Autoimmunity in Addison’s disease. Neth J Med 60(7):269–275 5. Al Sabri AM, Smith N, Busuttil A (1997) Sudden death due to auto-immune Addison’s disease in a 12-year-old girl. Int J Legal Med 110(5):278–280 6. Alevritis EM, Sarubbi FA, Jordan RM, Peiris AN (2003) Infectious causes of adrenal insufficiency. South Med J 96(9):888–890 7. Betterle C, Dal Pra C, Mantero F, Zanchetta R (2002) Autoimmune adrenal insufficiency and autoimmune polyendocrine syndromes: autoantibodies, autoantigens, and their applicability in diagnosis and disease prediction. Endocr Rev 23(3):327–364 8. Siebenlist D (1995) “White Addison’s disease” as an intensive care emergency. Dtsch Med Wochenschr 120(10):346–347 9. Schröder AS, Hecht L, Sperhake JP, Püschel K, Saeger W (2009) Sudden death due to ­endocinological crises. Rechtsmedizin 19:17–20

114

L. Hecht

10. Woenckhaus U, Vasold A, Bollheimer LC (2005) Adrenal crisis. Intensivmed 42:345–354 11. Sasano H, Young WF Jr, Chrousos GP, Koch CA, Giordano TJ, Kawashima A (2004) Adrenal cortical adenoma. In: DeLellis RA, Lloyd RV, Heitz PU, Eng C (eds) WHO classification of tumours, pathology and genetics, tumours of endocrine organs. IARC Press, Lyon 12. Young WF (2007) Primary aldosteronism: renaissance of a syndrome. Clin Endocrinol 66(5):607–618 13. Abdo A, Bebb RA, Wilkins GE (1999) Ventricular fibrillation: an extreme presentation of primary hyperaldosteronism. Can J Cardiol 15(3):347–348 14. Reschke K, Lehnert H (2003) Die thyreotoxische Krise. Internist 44(10):1221–1230 15. Hehrmann R (1996) Die thyreotoxische krise: fallstricke in der diagnostik-intensivtherapie. Fortschr Med 114:26–31 16. Amikam S, Riss E (1974) Acute myocardial infarction in a young patient with thyrotoxicosis. Harefuah 87(11):509–510 17. Cheah JS, Lee GS, Chew LS (1971) Myocardial infarction in thyrotoxicosis. Med J Aust 1(7):393–395 18. Kotler MN, Michaelides KM, Bouchard RJ, Warbasse JR (1973) Myocardial infarction associated with thyrotoxicosis. Arch Intern Med 132(5):723–728 19. Safe AF, Griffiths KD, Maxwell RT (1990) Thyrotoxic crisis presenting as status epilepticus. Postgrad Med J 66(772):150–152 20. Wei JY, Genecin A, Greene HL, Achuff SC (1979) Coronary spasm with ventricular fibrillation during thyrotoxicosis: response to attaining euthyroid state. Am J Cardiol 43(2):335–339 21. Burch HB, Wartofsky L (1993) Life-threatening thyrotoxicosis. Thyroid storm. Endocrinol Metab Clin North Am 22(2):263–277 22. Bhasin S, Wallace W, Lawrence JB, Lesch M (1981) Sudden death associated with thyroid hormone abuse. Am J Med 71(5):887–890 23. Newman CM, Price A, Davies DW, Gray TA, Weetman AP (1998) Amiodarone and the thyroid: a practical guide to the management of thyroid dysfunction induced by amiodarone therapy. Heart 79(2):121–127 24. Gough IR, Gough J (2002) Surgical management of amiodarone-associated thyrotoxicosis. Med J Aust 176(3):128–129 25. Mulligan DC, McHenry CR, Kinney W, Esselstyn CB Jr (1993) Amiodarone-induced thyrotoxicosis: clinical presentation and expanded indications for thyroidectomy. Surgery 114(6):1114–1119 26. Jordan RM (1995) Myxedema coma. Pathophysiology, therapy, and factors affecting prognosis. Med Clin North Am 79(1):185–194 27. Wall CR (2000) Myxedema coma: diagnosis and treatment. Am Fam Physician 62(11): 2485–2490 28. Ragaller M, Quintel M, Bender HJ, Albrecht DM (1993) Myxedema coma as a rare postoperative complication. Anaesthesist 42(3):179–183 29. Bailes BK (1999) Hypothyroidism in elderly patients. AORN J 69:1026–1030 30. Davis PJ, Davis FB (1984) Hypothyroidism in the elderly. Compr Ther 10:17–23 31. Arlot S, Debussche X, Lalau JD, Mesmacque A, Tolani M, Quichaud J, Fournier A (1991) Myxoedema coma: response of thyroid hormones with oral and intravenous high-dose L-thyroxine treatment. Intensive Care Med 17(1):16–18 32. Schmid KW, Böcker W (1997) Nebenschilddrüse. In: Remmele W (ed) Pathologie, 2. Aufl., Bd. 4. Springer, Berlin, Heidelberg, New York 33. Pfeilschifter J, Schatz H (2000) Differential diagnosis of hypercalcemia in adults. Med Klin 95(3):143–150 34. Raue F, Blind E, Schuler G, Otto G, Ziegler R (1992) Parathyrotoxic crisis with fatal outcome – what can be learned from this? Med Klin (Munich) 87(9):495–497 35. Lauschke H, Kaminski M, Stratmann H, Hirner A (2000) The parathyrotoxic crisis – a surgical emergency. Chirurg 71:58–62 36. Hehrmann R (1996) Hypercalcemic crisis exacerbation of an existing hypercalcemia ­syndrome. Fortschr Med 114(10):118–122

5  Endocrine Disorders with Potentially Fatal Outcome

115

37. Cronin CS, Reeve TS, Robinson B, Clifton-Bligh P, Guinea A, Delbridge L (1996) Primary hyperparathyroidism in childhood and adolescence. J Paediatr Child Health 32(5):397–399 38. Brown MJ, Willis T, Omalu B, Leiker R (2006) Deaths resulting from hypocalcemia after administration of edetate disodium: 2003–2005. Pediatrics 118(2):534–536 39. Saeger W, Hanke D (1978) Pituitary necroses: correlations to shock. Verh Dtsch Ges Pathol 62:301–306 40. Saeger W (1997) Hypophyse. In: Remmele W (ed) Pathologie, 2. Aufl., Bd. 4. Springer, Berlin, Heidelberg, New York 41. Caturegli P, Newschaffer C, Olivi A, Pomper MG, Burger PC, Rose NR (2005) Autoimmune hypophysitis. Endocr Rev 26(5):599–614 42. Cosman F, Post KD, Holub DA, Wardlaw SL (1989) Lymphocytic hypophysitis. Report of 3 new cases and review of the literature. Medicine 68(4):240–256 43. Fujimaki T, Hotta S, Mochizuki T, Ayabe T, Matsuno A, Takagi K, Nakagomi T, Tamura A (2005) Pituitary apoplexy as a consequence of lymphocytic adenohypophysitis in a pregnant woman: a case report. Neurol Res 27(4):399–402 44. Guay AT, Agnello V, Tronic BC, Gresham DG, Freidberg SR (1987) Lymphocytic hypophysitis in a man. J Clin Endocrinol Metab 64(3):631–634 45. Supler ML, Mickle JP (1992) Lymphocytic hypophysitis: report of a case in a man with cavernous sinus involvement. Surg Neurol 37(6):472–476 46. Gutenberg A, Hans V, Puchner MJA, Kreutzer J, Brück W, Caturegli P, Buchfelder M (2006) Primary hypophysitis: clinical-pathological correlations. Eur J Endocrinol 155(1):101–107 47. Nishioka H, Ito H, Sano T, Ito Y (1996) Two cases of lymphocytic hypophysitis presenting with diabetes insipidus: a variant of lymphocytic infundibulo-neurohypophysitis. Surg Neurol 46(3):285–290 48. Blisard KS, Pfalzgraf RR, Balko MG (1992) Sudden death due to lymphoplasmacytic hypophysitis. Am J Forensic Med Pathol 13(3):207–210 49. Gal R, Schwartz A, Gukovsky-Oren S, Peleg D, Goldman J, Kessler E (1986) Lymphoid hypophysitis associated with sudden maternal death: report of a case review of the literature. Obstet Gynecol Surv 41(10):619–621 50. Kaufmann P, Lax SF, Radner H, Eber B, Leuger A, Smolle KH (1995) Severe hypotension and coma secondary to unrecognized chronic anterior hypophysitis. Intensive Care Med 21(10):847–849 51. Brooks MH, Dumlao JS, Bronsky D, Waldstein SS (1973) Hypophysial tuberculoma with hypopituitarism. Am J Med 54(6):777–781 52. Klatt EC, Nichols L, Noguchi TT (1994) Evolving trends revealed by autopsies of patients with the acquired immunodeficiency syndrome. 565 autopsies in adults with the acquired immunodeficiency syndrome, Los Angeles, Calif, 1982–1993. Arch Pathol Lab Med 118(9): 884–890 53. Woywodt A, Knoblauch H, Kettritz R, Schneider W, Göbel U (2000) Sudden death and Wegner’s granulomatosis of the pituitary. Scand J Rheumathol 29(4):264–266 54. Tomita T, Gates E (1999) Pituitary adenomas and granular cell tumors. Incidence, cell type, and location of tumor in 100 pituitary glands at autopsy. Am J Clin Pathol 111(6):817–825 55. Gold EB (1981) Epidemiology of pituitary adenomas. Epidemiol Rev 3:163–183 56. Matsuura H, Kitazawa Y, Tanaka M, Morooka K (1994) Pituitary adenoma and unexpected sudden infant death: a case report. Med Pediatr Oncol 22(4):283–286 57. Püschel K (2004) Plötzlicher Tod im Erwachsenenalter. In: Brinkmann B, Madea B (Hrsg.) Handbuch Gerichtliche Medizin, Bd. 1. Springer, Berlin, Heidelberg, New York 58. Wakai S, Fukushima T, Teramoto A, Sano K (1981) Pituitary apoplexy: its incidence and clinical significance. J Neurosurg 55(2):187–193 59. Bauer M, Lang C, Patzelt D (2001) Sudden death due to pitiutary apoplexy. Leg Med (Tokyo) 3(3):183–186 60. Di Maio SM, Di Maio VJ, Kirkpatrick JB (1980) Sudden, unexpected deaths due to primary intracranial neoplasms. Am J Forensic Med Pathol 1(1):29–45 61. Ferramosca B, Bianchi D, Serra D, Savini R, Villecco AS, Bugiardini R (1987) Cardiological findings in acromegaly. Minerva Med 78(24):1813–1821

116

L. Hecht

62. Lopez-Velasco R, Escobar-Morreale HF, Vega B, Villa E, Sancho JM, Moya-Mur JL, ­Garcia-Robles R (1997) Cardiac involvement in acromegaly: specific myocardiopathy or consequence of systemic hypertension? J Clin Endocrinol Metab 82(4):1047–1053 63. Janssen OE, Mann K, Erbel R (2007) Endocrine disorders and the heart. Internist 48(3):246–254 64. Rossi L, Thiene G, Caregaro L, Giordano R, Lauro S (1977) Dysrhythmias and sudden death in acromegalic heart disease. A clinicopathologic study. Chest 72:495–498 65. Saeger W, Mitschke H, Lücking T (1972) On the cytology of the anterior pituitary in adrenocortical hypoplasia with hypoglycaemia. Dtsch Med Wochenschr 97:943–946 66. Fridrichsen C (1918) Nebennierenapoplexie bei kleinen Kindern. In: Bokay J, Czerny A, Feer E, Heuber O (eds) Jahrbuch für Kinderheilkunde und physische Erziehung, Bd 37. Karger, Berlin 67. Waterhouse R (1911) A case of suprarenal apoplexia. Lancet I:577–578 68. Grellner W, Madea B (1999) Sudden death due to Waterhouse-Fridrichsen syndrome in adulthood. Rechtsmedizin 9:102–106 69. Bakaeen FG, Walkes JC, Reardon MJ (2005) Heparin-induced thrombocytopenia associated with bilateral adrenal hemorrhage after coronary artery bypass surgery. Ann Thorac Surg 79(4):1388–1390 70. Saeger W (1997) Nebennierenrinde. In: Remmele W (Hrsg.) Pathologie 2. Aufl., Bd. 4. Springer, Berlin, Heidelberg, New York 71. Hayashi Y, Hiyoshi T, Takemura T, Kurashima C, Hirokawa K (1989) Focal lymphocytic infiltration in the adrenal cortex of the elderly: immunohistological analysis of infiltrating lymphocytes. Clin Exp Immunol 77(1):101–105 72. Burke MP, Opeskin K (1999) Adrenocortical insufficiency. Am J Forensic Med Pathol 20(1):60–65 73. Ringstad J, Rodge S, Loland W, Rode L (1991) Rapidly fatal Addison’s disease: three case reports. J Intern Med 230(5):465–467 74. Sobrinho-Simoes M, Brandao A, Paiva ME, Vilela B, Fernandes E, Carneiro-Chaves F (1985) Lymphoid hypophysitis in a patient with lymphoid thyroiditis, lymphoid adrenalitis, and idiopathic retroperitoneal fibrosis. Arch Pathol Lab Med 109(3):230–233 75. Ward S, Evans CC (1985) Sudden death due to isolated adrenal tuberculosis. Postgrad Med J 61(717):635–636 76. Uno K, Konishi M, Yoshimoto E, Kasahara K, Mori K, Maeda K, Ishida E, Konishi N, Murakawa K, Mikasa K (2007) Fatal cytomegalovirus-associated adrenal insufficiency in an AIDS patient receiving corticosteroid therapy. Intern Med 46(9):617–620 77. White PC, Speiser PW (2000) Congenital adrenal hyperplasia due to 21-hydroxylase deficiency. Endocr Rev 21(3):245–291 78. McNicol AM, Young WF Jr, Kawashima A, Komminoth P, Tischler AS (2004) Benign phaeochromocytoma. In: DeLellis RA, Lloyd RV, Heitz PU, Eng C (eds) WHO classification of tumours, pathology and genetics, tumours of endocrine organs. IARC Press, Lyon 79. Lo CY, Lam KY, Wat MS (2000) Adrenal pheochromocytoma remains a frequently overlooked diagnosis. Am J Surg 179(3):212–215 80. McNeil AR, Blok BH, Koelmeyer TD, Burke MP, Hilton JM (2000) Pheochromocytoma discovered during coronial autopsies in Sidney, Melbourne and Auckland. Aust N Z J Med 30(6):648–652 81. Klöppel G (2003) Tumors of the adrenal medulla and the paraganglia. Pathologe 24(5):280–286 82. Aikawa N, Suga S, Inamasu J, Takeda J, Kawase T, Tanahashi N, Murai M, Saruta T, Hata J (2000) A 35-year-old man with cerebral hemorrhage and pheochromocytoma: the second brain-dead organ donor in Japan. Keio J Med 49(3):117–130 83. Knisely AS, Sweeney K, Ambler MW (1988) Pheochromocytoma and sudden death as a result of cerebral infarction in Turner’s syndrome: report of a case. J Forensic Sci 33(6):1497–1502 84. Preuss J, Woenckhaus C, Schwesinger G, Madea B (2006) Non-diagnosed pheochromocytoma as a cause of sudden death in a 49-year-old man: a case report with medico-legal ­implications. Forensic Sci Int 156(2–3):223–228

5  Endocrine Disorders with Potentially Fatal Outcome

117

85. Scardigli K, Biller J, Brooks MH, Cespedes LE, Posniak HV (1985) Pontine hemorrhage in a patient with pheochromocytoma. Arch Intern Med 145(2):343–344 86. Türk EE, Sperhake JP, Saeger W, Tsokos M (2004) Pheochromocytoma as the cause of ­sudden death. Rechtsmedizin 14(5):409–411 87. Cardesi E, Cera G, Cassia A (1994) Pheochromocytoma and sudden death: a case of hyperacute myocardial ischemia. Pathologica 86(6):670–672 88. Ahlawat SK, Jain S, Kumari S, Varma S, Sharma BK (1999) Pheochromocytoma associated with pregnancy: case report and review of the literature. Obstet Gynecol Surv 54(11):728–737 89. Isaacson C, Rosenzweig D, Seftel HC (1960) Malignant pheochromocytoma of the organs of Zuckerkandl. Arch Pathol 70:725–729 90. Klein H (1969) Pheochromocytoma and pregnancy. Report on a pheochromocytoma of the Zuckerkandl’s organs in a bipara. Münch Med Wochenschr 111(17):977–981 91. Sperry K, Smialek JE (1986) Sudden death due to a paraganglioma of the organs of Zuckerkandl. Am J Forensic Med Pathol 7(1):23–29 92. Schmid KW, Böcker W (1997) Schilddrüse. In: Remmele W (Hrsg) Pathologie 2. Aufl., Bd. 4. Springer, Berlin, Heidelberg, New York 93. Siegler RW (1998) Fatal heatstroke in a young woman with previously undiagnosed Hashimoto’s thyroiditis. J Forensic Sci 43(6):1237–1240 94. Lorin De La Grandmaison G, Izembart M, Fornes P, Paraire F (2003) Myocarditis associated with Hashimoto’s disease: a case report. Int J Legal Med 117(6):361–364 95. Kahaly G, Mohr-Kahaly S, Beyer J, Meyer J (1995) Prevalence of myxomatous mitral valve prolapse in patients with lymphocytic thyroiditis. Am J Cardiol 76(17):1309–1310 96. Chan JKC, Hirokawa M, Evans H et al (2004) Follicular adenoma. In: DeLellis RA, Lloyd RV, Heitz PU, Eng C (eds) WHO classification of tumours, pathology and genetics, tumours of endocrine organs. IARC Press, Lyon 97. Sheu SY, Görges R, Schmid KW (2003) Hyperplasia of the thyroid gland. Pathologe 24(5):348–356 98. Bell CD, Kovacs K, Horvath E, Rotondo F (2001) Histologic, immunohistochemical and ultrastructural findings in a case of minocycline-associated “black thyroid”. Endocr Pathol 12(4):443–451 99. Tsokos M, Schröder S (2006) Black thyroid: report of an autopsy case. Int J Legal Med 120(3):157–159 100. Grimelius L, DeLellis RA, Bondeson L et al (2004) Parathyroid adenoma. In: DeLellis RA, Lloyd RV, Heitz PU, Eng C (eds) WHO classification of tumours, pathology and genetics, tumours of endocrine organs. IARC Press, Lyon 101. Sheu SY, Otterbach F, Frilling A, Schmid KW (2003) Hyperplasia and tumors of the parathyroid glands. Pathologe 24(5):373–381 102. Wenig BM, Heffes CS, Adair CF (1997) Neoplasms of the parathyroid glands. Atlas of endocrine pathology. Saunders, Philadelphia 103. Bondeson L, Grimelius L, DeLellis RA et al (2004) Parathyroid carcinoma. In: DeLellis RA, Lloyd RV, Heitz PU, Eng C (eds) WHO classification of tumours, pathology and genetics, tumours of endocrine organs. IARC Press, Lyon 104. Asmundo A, Aragona M, Gualniera P, Aragona F (1995) Sudden death from hypoglycemia. Pathologica 87(6):603–616 105. Ouyang YB, He QP, Li PA, Janelidze S, Wang GX, Siesjö BK (2000) Is neuronalinjury caused by hypoglycaemic coma of the necrotic or apoptotic type? Neurochem Res 25(5):661–667 106. Komminoth P, Perren A, Öberg K, Rindi G, Heitz PU, Klöppel G (2004) Insulinoma. In: DeLellis RA, Lloyd RV, Heitz PU, Eng C (eds) WHO classification of tumours, pathology and genetics, tumours of endocrine organs. IARC Press, Lyon 107. Donow C, Pipeleers-Marichal M, Stamm B, Heitz PU, Kloppel G (1990) The pathology of insulinoma and gastrinoma. The location, size, multicentricity, association with multiple endocrine type-I neoplasms and malignancy. Dtsch Med Wochenschr 115(37):1386–1391 108. Williams AJ, Coates PJ, Lowe DG, McLean C, Gale EA (1992) Immunochemical investigation of insulinomas for islet amyloid polypeptide and insulin: evidence for differential synthesis­and storage. Histopathology 21:215–223

118

L. Hecht

109. Klöppel G (1997) Endokrines pankreas. In: Remmele W (Hrsg) Pathologie 2. Aufl., Bd. 4. Springer, Berlin, Heidelberg, New York 110. Roth J, Klöppel G, Madsen OD, Storch MJ, Heitz PU (1992) Distribution patterns of proinsulin and insulin in human insulinomas: an immunohistochemical analysis in 76 tumors. Virchows Arch B Cell Pathol Incl Mol Pathol 63(1):51–61 111. Tsujino M, Sugiyama T, Nishida K, Takada Y, Takanishi K, Ishizawa M, Hirata Y (2005) Nonisulinoma pancreatogenous hypoglycemia syndrome: a rare case of adult-onset nesidioblastosis. Intern Med 44(8):843–847 112. van der Wal BC, de Krijger RR, de Herder WW, Kwekkeboom DJ, van der Ham F, Bonjer HJ, van Eijck CH (2000) Adult hyperinsulinemic hypoglycemia not caused by an insulinoma: a report of two cases. Virchows Arch 436(5):481–486 113. Wong JG, Tseng HS, Yang AH, Tang KT, Jap TS, Lee CH, Lin HD, Burcus N, Pittenger G, Vinik A (2006) Clinical features and morphological characterization of 10 patients with noninsulinoma pancreatogenous hypoglycemia syndroma (NIPHS). Clin Endocrinol 65(5):566–578 114. Anlauf M, Wieben D, Perren A, Sipos B, Komminoth P, Raffel A, Kruse ML, Fottner C, Knoefel WT, Mönig H, Heitz PU, Klöppel G (2005) Persistent hyperinsulinemic hypoglycemia in 15 adults with diffuse nesidioblastosis: diagnostic criteria, incidence, and characterization of beta-cell changes. Am J Surg Pathol 29(4):524–533 115. Raffel A, Krausch MM, Anlauf M, Wieben D, Braunstein S, Klöppel G, Röher HD, Knoefel WT (2007) Diffuse nesidioblastosis as a cause of hyperinsulinemic hypoglycemia in adults: a diagnostic and therapeutic challenge. Surgery 141(2):179–184 116. Sahloul R, Yaqub N, Driscoll HK, Leidy JW, Parkash J, Matthews KA, Chertow BS (2007) Noninsulinoma pancreatogenous hypoglycemia syndrome: quantitative and immunohistochemical analysis of islet cells for insulin, glucagon, somatostatin, and pancreatic and duodenal homeobox protein. Endocr Pract 13(2):187–193 117. Shinozuka H, Lee RE, Dunn JL, Longnecker DS (1980) Multiple atypical acinar cell nodules of the pancreas. Hum Pathol 11(4):389–391 118. Foulis AK, Liddle CN, Farquharson MA, Richmond JA, Weir RS (1986) The histopathology of the pancreas in type 1 (insulin-dependent) diabetes mellitus: a 25-year review of deaths in patients under 20 years of age in the United Kingdom. Diabetologia 29(5):267–274 119. Löhr M, Klöppel G (1987) Residual insulin positivity and pancreatic atrophy in relation to duration of chronic type 1 (insulin-dependent) diabetes mellitus and microangiopathy. Diabetologia 30(10):757–762 120. Clark A, de Koning EJ, Hattersley AT, Hansen BC, Yajnik CS, Poulton J (1995) Pancreatic ­pathology in non-insulin dependent diabetes (NIDDM). Diabetes Res Clin Pract 28(Suppl): S39–47 121. Gong W, Liu ZH, Zeng CH, Peng A, Chen HP, Zhou H, Li LS (2007) Amylin deposition in the kidney of patients with diabetic nephropathy. Kidney Int 72(2):213–218 122. Ehrlich P (1883) Über das Vorkommen von Glycogen im diabetischen und im normalen Organismus. Z Klin Med 6:33 123. Meixner K (1906) Mikroskopischer glycogennachweis. Münch Med Wschr 52:2175 124. Oldershausen HF (1966) Über Leberveränderungen bei diabetes mellitus. Acta HepatoSplenol 13:375 125. Schulz H (1969) Ultrastructure of intranuclear glycogen of human hepatocytes in cases of diabetes mellitus. Verh Dtsch Ges Pathol 53:304–309 126. Altmann HW (1949) Über die Abgabe von Kernstoffen in das Protoplasma der menschlichen Leberzelle. Z Natuforsch 4b:138 127. Dancygier H (2001) Nonalcoholic steatohepatitis. Dtsch Ärzteblatt 98:2511–2516 128. Sellyei M, Walton E (1969) On the occurrence and form of glycogen in liver cells. Zentralbl Allg Pathol 112(1):1–11 129. Ritz E (2006) Diabetic nephropathy. Saudi J Kidney Dis Transplant 17(4):481–490 130. Helmchen U, Schubert GE (1997) Niere. In: Remmele W (Hrsg) Pathologie Bd.5, 2.Aufl. Springer, Berlin, Heidelberg, New York 131. Herzenberg AM, Holden JK, Singh S, Magil AB (1999) Idiopathic nodular glomerulosclerosis. Am J Kidney Dis 34(3):560–564

5  Endocrine Disorders with Potentially Fatal Outcome

119

132. Müller-Höcker J, Weiss M, Thoenes GH, Grund A, Nerlich A (2002) A case of idiopathic nodular glomerulosclerosis mimicking diabetic glomerulosclerosis (Kimmelstiel-Wilson type). Pathol Res Pract 198(5):375–379 133. Kock KF, Vestergaard V (1994) Armanni-Ebstein lesions of the kidney: diagnostic of death in diabetic coma? Forensic Sci Int 67(3):169–174 134. Püschel K, Kernbach G, Brinkmann B (1987) Glycogennephrosis and postmortem biochemical equivalents of glucose metabolism. Verh Dtsch Ges Path 71:333–336 135. Coe JI (1993) Postmortem chemistry update. Emphasis on forensic application. Am J Forensic Med Pathol 14(2):91–117 136. Rachut E, Rynbrandt DJ, Doutt TW (1980) Postmortem behavior of serum thyroxine, triiodothyronine, and parathormone. J Forensic Sci 25(1):67–71 137. Edston E, Druid H, Holmgren P, Oström M (2001) Postmortem measurements of thyroid hormones in blood and vitreous humor combined with histology. Am J Forensic Med Pathol 22(1):78–83 138. Coe JI (1973) Postmortem values of thyroxine and thyroid stimulating hormone. J Forensic Sci 18:20–24 139. Bonnell HJ (1983) Antemortem chemical hypothyroxinemia. J Forensic Sci 28(1):242–248 140. Larsen PR (1992) The thyroid. In: Wyngaarden JB, Smith LH, Bennett JC (eds) Cecil textbook of medicine 19th edn. Saunders, Philadelphia, London, Toronto, Montreal, Sydney, Tokyo 141. Chong APY, Aw SE (1986) Postmortem endocrine levels in vitreous humor. Ann Acad Med Singapore 15:606–609 142. Finlayson NB (1965) Blood cortisol in infants and adults: a post mortem study. J Pediatr 67:284–289 143. Drucker D, Shandling M (1985) Variable adrenocortical fuction in acute medical illness. Crit Care Med 13:477–479 144. Lapinlampi TO, Hirvonen JI (1986) Catecholamines in the vitreous fluid and urine of guinea pigs dying of cold and the effect of postmortem freezing and autolysis. J Forensic Sci 31(4):1357–1365 145. Hirvonen J, Kortelainen ML, Huttunen P (1997) Pulmonary and serum surfactant phospholipids and serum catecholamines in strangulation. An experimental study on rats. Forensic Sci Int 90(1–2):17–24 146. Tormey WP, Carney M, FitzGerald RJ (1999) Catecholamines in urine after death. Forensic Sci Int 103(1):67–71 147. Davidson DF, Grosset K, Grosset D (2007) Parkinson’s disease: the effect of L-dopa therapy on urinary free catecholamines and metabolites. Ann Clin Biochem 44(Pt 4):364–368 148. Ishikawa T, Quan L, Li DR, Zhao D, Michiue T, Hamel M, Maeda H (2008) Postmortem biochemistry and immunohistochemistry of adrenocorticotropic hormone with special regard to fatal hypothermia. Forensic Sci Int 179(2–3):147–151 149. Coe JI (1976) Postmortem chemistry of blood, cerebrospinal fluid and vitreous humor. In: Wech CH (ed) Legal medicine annual 1976. Appleton, Century Croft, New York 150. Coe JI (1977) Postmortem chemistry of blood, cerebrospinal fluid and vitreous humor. In: Tedeschi CG, Eckert WC, Tedeschi LG (eds) Forensic medicine vol. 2. WB Saunders, Philadelphia 151. Coe JI (1975) Postmortem peripheral blood glucose and cardiopulmonary resuscitation. Forensic Sci Gaz 6(4):1–2 152. Kernbach G, Püschel K, Brinkmann B (1986) Die biochemischen Messgrößen des Glucosestoffwechsels in Abhängigkeit von der Todesart und postmortalen Einflüssen. Z Rechtsmed 96:199–213 153. Ritz S, Kaatsch HJ (1990) Postmortale Diagnostik von tödlichen diabetischen Stoffwechselentgleisungen: welchen Stellenwert haben Liquor- und Glaskörperflüssig­ keitssummenwerte sowie der HbA1-Wert? Pathologe 11:158–165 154. Zilg B, Alkass K, Berg S, Druid H (2009) Postmortem identification of hyperglycemia. Forensic Sci Int 185(1–3):89–95 155. Traub F (1969) Methode zur Erkennung von tödlichen Zuckerstoffwechselstörungen an der Leiche (Diabetes mellitus und Hypoglycämie). Zentralbl Allg Pathol 112(4):390–399

120

L. Hecht

156. Kernbach-Wighton G (2003) Postmortale biochemische Untersuchungen. In: Brinkmann B, Madea B (eds) Handbuch gerichtliche Medizin Band 1. Springer, Berlin, Heidelberg, New York 157. Sippel H, Möttönen M (1982) Combined glucose and lactate values in vitreous humour for postmortem diagnosis of diabetes mellitus. Forensic Sci Int 19:217–222 158. Karlovsek MZ (2004) Diagnostic values of combined glucose and lactate values in cerebrospinal fluid and vitreous humour – our experiences. Forensic Sci Int 146(Suppl):S19–23 159. Kernbach-Wighton G, Püschel K (1998) On the phenomenology of lethal application of insulin. Forensic Sci Int 93:61–73

Chapter 6

Sudden Death from Infectious Disease James A. Morris, Linda M. Harrison, and Robert M. Lauder

Abstract  This chapter is concerned with sudden death in infancy, childhood and adult life. Most of the evidence for a causal role for infection in sudden death comes from studies of sudden unexpected death in infancy (SUDI) and therefore the section on SUDI and closely related conditions forms the largest part of the material presented. There is a convincing body of evidence that infection has a role in at least some cases of sudden infant death. More specifically, there is evidence, based on sound theoretical principles and supported by laboratory experiments, that some cases of sudden death can be caused by common bacterial toxins absorbed from mucosal surfaces or delivered as part of a transient bacteraemia. This idea, termed the common bacterial toxin hypothesis, is not proven beyond doubt, it remains an hypothesis, but it does offer a plausible explanation for many of the features associated with sudden death at all ages. One reason for concentrating on this idea is that we are now in a position to prove or disprove the hypothesis using the techniques of genomics and proteomics. But we will only succeed if those who perform necropsy examinations in cases of sudden death are fully conversant with the theoretical background and are ready to obtain the appropriate specimens. Large-scale studies of the mucosal microbial flora with proteomic analysis of the bacterial secretome and parallel proteomic analysis of fluids obtained at autopsy are required. This is big science requiring experts in many disciplines. The potential rewards from these studies in terms of understanding disease are considerable and it will put autopsy pathology back in its rightful place at the centre of clinical academic medicine. Keywords  Sudden death • SUDI • Infection • Forensic pathology • Autopsy

J.A. Morris (*) Department of Pathology, University Hospitals of Morecambe Bay NHS Trust, Royal Lancaster Infirmary, Lancaster, UK e-mail: [email protected] E.E. Turk (ed.), Forensic Pathology Reviews, Volume 6, Forensic Pathology Reviews 6, DOI 10.1007/978-1-61779-249-6_6, © Springer Science+Business Media, LLC 2011

121

122

J.A. Morris et al.

Sudden Unexpected Death in Infancy Sudden unexpected death in infancy (SUDI) is simply defined as the death of an infant aged between 7 days and 12 months which is sudden and unexpected. A detailed epidemiological study of SUDI was carried out in England between 1993 and 1996 under the auspices of CESDI (confidential enquiry into sudden death in infancy) [1]. In the study population there were over 470,000 live births and 456 cases of SUDI occurred. Ninety three cases were fully explained following a detailed autopsy, but 363 cases remained unexplained and were classified as sudden infant death syndrome (SIDS). In this study, SIDS and unexplained SUDI were regarded as synonymous. In recent years, however, there has been a move to restrict the term SIDS to cases in which there has been a full death scene investigation, the autopsy protocol has met exacting standards and there are no suspicious circumstances. Thus, in current practice SIDS and unexplained SUDI are no longer synonymous, and it is a failure to appreciate this fact that has contributed to a number of miscarriages of justice. The inclusion criteria for the CESDI SUDI study were as follows: • Deaths that were unexpected, and unexplained at autopsy (i.e. those meeting the criteria for SIDS). • Deaths occurring in the course of an acute illness that was not recognized by ­carers and/or by health professionals as potentially life threatening. • Deaths occurring in the course of a sudden acute illness of less than 24 h duration in a previously healthy infant, or a death that occurred after this if intensive care had been instituted within 24 h of the onset of the illness. • Deaths arising from a pre-existing condition that had not been previously ­recognized by health professionals. • Deaths arising from any form of accident, trauma or poisoning. The 93 cases classified as explained SUDI are shown in Table 6.1.

Explained SUDI-Infection The commonest cause of explained SUDI is infection. Pneumonia, septicaemia and meningitis caused by bacteria or viruses can lead to sudden death in infancy. In the Table 6.1  CESDI SUDI study 1993–1996; cases classified as explained SUDI

Number of cases 35 15 21 10 5 4 3

Cause of death Infection Accidental death Non-accidental death Congenital abnormality Intestinal obstruction Metabolic disorder Other cause

6  Sudden Death from Infectious Disease

123

Table 6.2  Baby check scoring system; used in the CESDI SUDI study 1993–1996 Score Signs and symptoms 0–7 The infant is usually well 8–12 The baby is unwell but not seriously ill, carers should obtain advice and observe progress 13–19 The baby is ill and needs a doctor 20+ The baby is seriously ill and needs a doctor urgently

CESDI SUDI study the diagnosis of infection was based on histological criteria supplemented by bacteriological and virological studies. The bacteria that were isolated from lungs, blood or CSF and considered significant included, Streptococcus pneumoniae and Neisseria meningitidis but also more controversially, Staphylococcus aureus and Escherichia coli. In most cases of bacterial pneumonia, septicaemia and meningitis the causative organisms initially grow in the upper respiratory tract [2]. Spread then occurs to the lower respiratory tract to cause pneumonia and/or to the blood to cause septicaemia. Meningitis occurs secondary to blood borne spread. If the diagnosis of meningitis or pneumonia depends on the accumulation of neutrophils in sufficient numbers to be recognizable as inflammation then the entire process will take several hours at least. In the CESDI SUDI study the questionnaire to parents and controls included questions from the “Cambridge baby check” scoring system. This system uses easily recognized symptoms and signs, each of which is scored (Table 6.2). A striking finding was that 44.3% of explained SUDI cases had a score over 12 in the 24 h prior to death compared with 2.8% of controls (odds ratio 22; 95% confidence intervals 7–71). Thus, for many infants who die of histologically proven infection there will have been signs and symptoms of infection in the hours prior to death. The case control study also showed that for explained SUDI: • • • • •

Deaths were more common in the winter months There was a higher proportion of boys than girls (61%) Infants who died were of lower birth weight and shorter gestational age. The mothers of cases were younger and started their families earlier. The families of cases were more likely to live in rented and crowded accommodation, to be unemployed and to smoke.

Unexplained SUDI The majority of sudden unexpected deaths in infancy remain unexplained after a detailed autopsy. In the CESDI SUDI study 80% were classified as SIDS or unexplained SUDI. In a more recent retrospective review of 546 cases of SUDI examined at one institution over the preceding 10 years, 63% remained unexplained [3]. The key epidemiological features of the unexplained group are as follows: • The number of deaths rises to a peak at 2–3 months of age and then falls so that death is uncommon after 6 months and rare by 12 months.

124

J.A. Morris et al.

• Infants are least likely to die if they sleep on their back (supine) in a cot in the same room as their parents. The risk of death rises if they sleep prone, share a bed with their parents, are overwrapped or go to sleep on a sofa. • Infants of young mothers who start their family early are at increased risk. • The condition is commoner in male infants (60%) than female infants. • Low birth weight and reduced gestational age increase risk. • Parental smoking is a risk factor for SIDS. • Adverse social circumstances such as unemployment and overcrowded accommodation increase risk. • Later born infants are at increased risk compared with first born infants. • In the 1980s there was an increased incidence in the winter months but the seasonal effect became less marked in the 1990s following a reduction in the prevalence of prone sleeping. The epidemiological features of unexplained and explained SUDI are similar. They have a similar age incidence, there is a seasonal effect, a male excess, they are associated with poor social circumstances and parental smoking. These are all factors commonly associated with an increased risk of respiratory tract infection. We can readily understand why these factors are found in association with explained SUDI because infection originating in the respiratory tract is the commonest cause of death. For the same reason it has long been suspected that respiratory tract infection might play an important role in the pathogenesis of unexplained SUDI. The common bacterial toxin hypothesis is a specific idea linking respiratory tract infection to unexplained SUDI [4]. The hypothesis is that toxins produced by bacteria commonly found in the upper respiratory tract enter the blood stream and cause sudden collapse and sudden death by directly interfering with the neural control of respiratory and/or cardiac function. The toxins are either absorbed from the mucosal surface or delivered as part of a transient bacteraemia [5]. Evidence in support of this idea has gradually accumulated over the last 20 years [6]. • A mathematical model, based on the common bacterial toxin hypothesis, closely predicts the age incidence of SIDS which is the most consistent and characteristic feature of the syndrome [4]. Circulating immunoglobulin G (IgG) protects against infection by bacteria and bacterial toxins. Infants are born with high levels of IgG which has been pumped across the placenta from maternal blood. As the level of circulating IgG in the infant falls the risk of SIDS rises. When IgG reaches its nadir at 2–3 months of age the risk of SIDS is at its peak. Thereafter the level of circulating IgG rises, as infants meet and acquire immunity to common bacterial organisms, and the risk of SIDS falls. By 12 months of age infants are immune and therefore the causative organisms and toxins must be very common. • The nasopharyngeal and oropharyngeal bacterial flora in SIDS cases is disturbed with increased growth of staphylococci and Gram negative bacilli [7]. A study of 95 cases of SUDI was carried out in Bristol between 1987 and 1989 [8]. Throat swabs were obtained in the SUDI cases with a median time of 3.5 h from the discovery of the death. Throat swabs were also obtained in a comparison group of 190 healthy infants from the community matched for age, gender, season and locality.

6  Sudden Death from Infectious Disease

•

•

•

•

• •

125

Staphylococcus aureus was isolated in 24 of 95 cases (25%) compared with 12 of 190 controls (6%); odds ratio 5 (2.4–10). Coliforms were isolated in 15 of 95 cases (16%) compared with 2 of 190 controls (1%); odds ratio 29 (7.5–111). Streptococcus pneumoniae was isolated in 10 of 95 cases (11%) compared with 2 of 190 controls (1%); odds ratio 10 (2.9–34). Group B streptococcus was isolated in 6 of 95 cases (6%) compared with 2 of 190 controls (1%); odds ratio 11 (1.9–63). Isolates of S. aureus and E. coli from SIDS cases interacted synergistically to cause rapid septicaemic death in gnotobiotic weanling rats [9]. In this model, bacteria were isolated from the blood stream but there were minimal histological changes at autopsy. Toxin extracts of S. aureus and E. coli from SIDS cases were injected into the chorio-allantoic vein of chick embryos. The toxins interacted synergistically to cause death of the embryos [10]. Minute doses of nicotine, equivalent to the dose in man from smoking one-twentieth of a cigarette, interacted with bacterial toxins to produce a lethal effect in the chick embryos [11]. Nasopharyngeal secretions pool in the upper airways when adults suffering from an upper respiratory tract infection sleep prone [12]. This leads to increased bacterial carriage. A similar result is found in infants; nasopharyngeal secretions sampled in the early morning after an overnight sleep have increased bacterial counts if the infants slept prone rather than supine [13]. The increase was most marked in infants with a viral URTI. Not only is the total count raised but there is also increased carriage of S. aureus and coliforms. The staphylococcal pyrogenic toxins TSST and SEC were found in tissue, including brainstem, of 33 of 62 cases of SIDS (53%). The toxins were measured using ELISA and flow cytometry [14]. These results confirm those of earlier studies using immunohistochemistry [15]. IgG antibody to the conserved core of the endotoxin molecule (EndoCAb) is decreased in SIDS cases [16]. This indicates low levels of maternal antibody at birth or consumption of antibody due to endotoxaemia. Weber and colleagues [3] reviewed 546 cases of SUDI examined in one institution over a period of 10 years. In 470 cases samples of lung, spleen, blood and CSF had been obtained for bacterial culture. They found that S. aureus and E.coli were more commonly isolated in cases of unexplained SUDI than in cases where death had been explained by a non-infective process.

There is, therefore, an impressive body of evidence linking the respiratory tract bacterial flora, common bacterial toxins and unexplained SUDI.

Mode of Death in Unexplained SUDI Baby check scores for SIDS infants identified in the CESDI SUDI study are detailed in Table 6.3. 78.6% of the SIDS infants had a baby check score between 0 and 7 in the 24 h prior to death [1]. Thus, in terms of signs and symptoms the infants appeared to be generally well.

126

J.A. Morris et al.

Table 6.3  Baby check scores for SIDS infants identified in the CESDI SUDI study Score Signs and symptoms Percentage 0–7 The infant is usually well 78.6 8–12 The baby is unwell but not seriously ill, carers should 10.1 obtain advice and observe progress 13–19 The baby is ill and needs a doctor   6.0 20+ The baby is seriously ill and needs a doctor urgently   5.3

In the third year of the CESDI SUDI study information was collected on time from last seen to found dead. A total of 53 infants died in daytime hours and the median interval from last seen to found dead was 1 h 15 min (interquartile range 28 min to 1 h 35 min). For 5 infants the time gap was less than 10 min [17]. A few infants have died whilst on a monitor and their deaths classified, following autopsy, as unexplained SUDI. These cases, selected for monitoring, might not be typical of SIDS deaths in general but the physiological recordings are reasonably consistent. There is a general slowing of heart rate and respiratory rate with hypoxaemia followed by irregular grunting respiration and finally death. The entire process takes less than 20 min [18, 19]. Thus, it appears that in most cases of unexplained SUDI the infant is well, or at most only mildly unwell, when last seen and death occurs in a short period of time. This is not the mode of death that is usually associated with infection. The classical concept is that the organisms grow, which takes time, the body responds, which also takes time, the infant becomes progressively more unwell with obvious signs and symptoms of infection before death occurs. The idea that an infant can appear well and then die within 20 min due to infection is a serious challenge to conventional views. One infective process with a similar time frame to that of death in unexplained SUDI is transient bacteraemia [5]. Episodes of bacteraemia occur throughout life; the bacteria are quickly cleared from the blood stream by neutrophils and any toxins that are released will be neutralized by circulating IgG. But at 2–3 months of age the infant will be at risk from the toxins because of relative lack of IgG. There are a wide variety of bacterial toxins with a wide range of actions [20]. The staphylococcal pyrogenic toxins stimulate T lymphocytes to release cytokines including tumour necrosis factor (TNF). Widespread overproduction of cytokines can cause vascular dilatation with a catastrophic fall in blood pressure and generalized shock leading to death [21]. But the process of lymphocyte stimulation and cytokine production takes time and we would need to postulate more than one episode of transient bacteraemia, the first priming the system and a later episode leading to the cytokine cascade. Another difficulty with the idea of pyrogenic toxins causing cardiovascular collapse is – why is death the only end point and always sudden? Why do we not observe cardiovascular collapse in infants more often? This is in fact a general difficulty, which we will come back to later. Cytolysins and colicins are bacterial toxins which act by inserting ion channels into cell membranes [20]. Leakage of ions then disturbs the function of the cell and in the case of nerve cells or cardiac muscle fibres this could lead to cardiac arrhythmia­ or cessation of respiration. The alpha haemolysin toxin produced by S. aureus is a cytolysin. A number of bacteria, including E. coli, produce colicins which help in

6  Sudden Death from Infectious Disease

127

the competition for space on mucosal surfaces by lysing other bacteria. A similar mechanism is used by complement components to lyse cells and by natural killer cells to effect direct cell death. Rapid death by disturbing ion flow has a long evolutionary history; it is therefore of considerable interest to note that recent evidence suggests that up to 15% of previously unexplained SUDI cases are caused by single mutations in channelopathy genes [22] (see below).

“Near Miss” SUDI The majority of infants who die are well when last seen and succumb to a profound physiological disturbance in a short period of time caused by a natural disease process. If the disease process involves infection, as argued above, then one would anticipate that a few infants would be found, not dead, but in a state of cardiorespiratory collapse. The normal rules of biology and pathology would tend to exclude an infective disease process that is 100% fatal. The question posed is then: “in what manner does ‘near miss SUDI’ present?” Haemorrhagic Shock and Encephalopathy A candidate condition for “near miss SUDI” is haemorrhagic shock and encephalopathy (HSE), first described by Levin and colleagues [23]. They reported 10 infants with acute onset of encephalopathy, fever, shock, watery diarrhoea, severe disseminated intravascular coagulation, and renal and hepatic dysfunction. Seven of the infants died. No cause was found and blood cultures were negative. A similar syndrome had already been described by Bacon and colleagues [24]. They had reported five infants with sudden onset of fever, shock, convulsions, hepatic disturbance and a bleeding tendency. Blood cultures were negative. The authors suggested that the condition might be caused by heatstroke due to a combination of mild infection and overwrapping. Further cases have been described and the suggestion of a link between infection and overwrapping has received support [25, 26]. The age incidence of HSE rises to a peak at 3 months of age and then falls, supporting a role for infection [27]. In a case–control study of HSE it was found that cases were 30 times more likely than controls to be found with their head covered by bed clothes [28]. A report of a hybrid between staphylococcal toxic shock syndrome and HSE also supports an infective cause [29]. In two pairs of twins one twin developed HSE at the same time that the other died of SIDS supporting the idea that HSE is a candidate condition for “near miss SUDI” [26]. The clinical picture of HSE is essentially septicaemia but with negative blood cultures. It is tempting to suggest that bacterial toxaemia, secondary to a transient bacteraemia or due to direct absorption of toxins from mucosal surfaces, is a ­possible cause.

128

J.A. Morris et al.

Acute (Apparent) Life-Threatening Events In the CESDI SUDI study the parents of 317 cases of SIDS and 1,299 controls were asked whether their infant had ever had an episode in which he or she became lifeless. An affirmative response was obtained from 11.7% of SIDS parents and 3% of controls. The parents described cessation of breathing and change of colour as the main features. The parents were also asked whether their infant had ever had any form of convulsion, fit or seizure or other turn in which consciousness was lost or any part of the body made abnormal movements. A positive response was obtained from 3.5% of SIDS parents and 0.9% of controls. A positive response to these questions defines the condition ALTE. The results show that previous episodes of ALTE are associated with SIDS and that ALTE is not uncommon even in the control population. A systematic review of publications on ALTE identified 643 infants aged 0–13 months [30]. The calculated incidence figure was 0.6 per 1,000 live births. This is a much lower figure than found in the CESDI SUDI study and indicates that the majority of episodes of ALTE do not come to medical attention because they are short lived with no obvious harmful consequences. In the 643 cases in this review that came to medical attention there were 728 diagnoses (more than one per case). The most common were gastrointestinal reflux (n = 227), seizure (n = 83), lower respiratory tract infection (n = 58), and unknown (n = 169). In a prospective epidemiological study of 44,184 live-born infants in the Tyrol, carried out between 1993 and 2001, 164 cases of ALTE were identified (2.46 per 1,000 live births). This incidence figure is still tenfold lower than the CESDI SUDI study. In 73 of the 164 cases no cause was identified [31]. Although ALTE is likely to be a mixed bag of different conditions, physiological recordings during attacks show a consistent pattern which is similar to that seen in SIDS. There is hypoxaemia, bradycardia, gasping and then cessation of respiration followed after a variable interval by complete recovery. The entire event is short lived lasting less than 20 min. It is characterized by rapid onset and equally rapid offset [32]. In some cases of ALTE, subsequent clinical investigation reveals evidence of infection, most commonly of the lower respiratory tract. It is tempting to suggest that an episode of bacteraemia followed by clearance and toxin release is the explanation for the link. It would account for rapid onset and offset during the course of an otherwise prolonged clinical episode of infection. Negative blood cultures following the ALTE are consistent with a bacteraemia, which has been terminated. There are over 600,000 live births per annum in England and Wales. If the CESDI SUDI controls are a representative sample then there will be over 18,000 episodes of ALTE per annum. The vast majority are short lived and do not come to medical attention. A few of the more severe episodes lead to hospital admission and are investigated. But also some are followed by sudden death. Death appears to be more likely in those infants who have had previous episodes.

6  Sudden Death from Infectious Disease

129

Hypoxic Ischaemic Encephalopathy One difference between ALTE and SUDI is that most cases of the former are observed in infants who are awake whilst most cases of the latter occur during sleep. But in other respects the two conditions appear to be closely related and epidemiological, physiological and pathological evidence shows a continuum of change. The physiological hallmark of ALTE is an episode of hypoxaemia of varying length and severity with a rapid onset and rapid offset. In most cases, there is no hypoxic damage to body organs as the episodes are mild in severity and short in time. In a minority, however, the hypoxic episode is severe and terminates in death. But if there is a continuum of change we would anticipate that some cases would be sufficiently severe to cause hypoxic damage to the brain but terminate prior to death. These cases will go on to develop hypoxic ischaemic encephalopathy (HIE) which might prove fatal in the next 1–2 days or the infant might survive with residual brain damage. An infant who suffered ALTE followed by HIE and then died in the next 24 or 48 h after a period of intensive care would fulfil the criteria for SUDI. But in the CESDI SUDI study there were no cases of explained death classified as HIE. There were, however, 21 cases classified as non-accidental injury. The commonest form of non-accidental injury in infants is shaken baby syndrome leading to non-accidental head injury (NAHI). In these infants, the triad of retinal haemorrhage, bilateral thin film subdural haemorrhage and encephalopathy is found. In addition, there is often direct evidence of injury such as bruising at the site of impact, a fractured skull or grip marks on the arms and chest. The combination of the triad and marks of trauma is regarded as virtually diagnostic of violent shaking injury (with or without impact). The cause of death in NAHI is encephalopathy. A decade ago it was thought that violent shaking with or without impact directly damaged the brain leading to traumatic brain swelling and death. At the same time traumatic tearing of veins led to retinal haemorrhage and subdural haemorrhage. The triad was therefore regarded as virtually diagnostic of NAHI even in the absence of other evidence of trauma. Thus, in the CESDI SUDI study carried out in 1993–1996 infants with the triad but no other evidence of trauma would have been classified as NAHI. The conventional view has been challenged by Dr Geddes and colleagues [33, 34]. The authors undertook a detailed neuropathological study of the brain in 37 infants (20 days to 9 months) and 16 children (13 months to 8 years) in whom death had been classified as NAHI. They used conventional staining techniques together with immunohistochemistry for beta amyloid precursor protein. Severe hypoxic damage was present in 77% of cases but diffuse traumatic axonal injury was present in only 3 of 53 cases. Eleven cases, all infants, showed localized axonal injury to the craniocervical junction of the spinal cord. Clinical records revealed that in 75% of infants there was an episode of significant apnoea at presentation. Thus, the commonest cause of death in cases classified as NAHI is HIE. The damage to the brain is due to oxygen lack not trauma. This raises the possibility that

130

J.A. Morris et al.

some cases classified as NAHI are in fact due to natural disease that has precipitated ALTE and led to HIE as suggested above. This is an area of current controversy and cannot be dealt with fully in this chapter. But the main points for and against are set out below: Those who believe the triad alone is virtually diagnostic of NAHI argue: • There are documented cases in which shaking injury, with and without impact, has led to the triad. The proof is the presence of injury such as bruises at the site of impact, a fractured skull or grip marks on the arms and legs in some cases. In others there are witness accounts or a confession. • There are rare non-traumatic causes of the triad such as glutaric aciduria type 1 or copper deficiency. But there are no authenticated cases of non-traumatic HIE developing subdural haemorrhage [35]. The counter argument: • The claim that retinal and subdural haemorrhage is due to traumatic tearing of veins is unproven hypothesis. • Trauma to the head which tears veins is more likely to cause a large unilateral subdural haematoma rather than bilateral thin film haemorrhage. • The argument that non-traumatic HIE does not cause the triad is circular because once the triad is observed the diagnosis non-traumatic HIE is not accepted. • Haemorrhage is commonly seen in autopsy cases at all ages following hypoxic ischaemic damage to organs. Thus, how can anyone be so confident that it does not occur in HIE in infancy. • HIE in the perinatal period does lead to bilateral thin film subdural haemorrhage [36]. Our view is that the triad alone is not enough for a confident diagnosis of NAHI at the criminal level of proof. We suspect that some of these cases are due to episodes of ALTE caused by natural disease leading to HIE. We further suspect that one of the natural conditions that can precipitate this event is infection. However, at the time of writing this is still a controversial area and the majority of experts who give evidence in Court believe that the triad alone is a reliable indicator of NAHI.

Gene/Environmental Interactions in SUDI The nature nurture controversy that has bedeviled biology for over a century is gradually fading with the realization that we are the product of an interaction between our genes and our environment. Physiological and pathological events are not caused by genetic mutations alone or by environmental stresses alone but by an interaction of the two. In the same way, when considering the role of infection in SUDI we need to think in terms of infection and bacterial toxins acting in the context of genetic predisposition. The DNA of a human diploid cell has approximately 6.4 billion base pairs coding for 50,000 genes. Only 1% of DNA codes for proteins but up to 4% is conserved

6  Sudden Death from Infectious Disease

131

and is concerned with regulatory functions. When a cell divides in mitosis the DNA is copied but there is an error rate in the process and in spite of DNA repair approximately five in ten billion base pairs are miscopied per cell division. There are at least 30–40 cell divisions between the zygote and the oocyte, and over 60 between the zygote and spermatozoa. The result is that the DNA copy we transmit to our children is different from the DNA copy we receive from our parents [37–41]. The changes are mutations. The majority of base changes are neutral in an evolutionary sense. They have no effect on protein coding function or on regulatory function. The vast majority of these neutral changes disappear from the population after a few generations purely by chance. But a tiny minority gradually increases in the population, again by chance. These neutral changes are recognized as single nucleotide polymorphisms (SNPs) and the common ones have been in the population for in excess of 10,000 generations or 250,000 years. If a base change effects a protein coding gene and leads to a change in the amino acid sequence of the protein product then the change is often deleterious in that the protein does not function or functions less well than the original. However, sometimes the protein functions just as well as the original and this mutation is also neutral. It has the same dynamics as other neutral mutations and might disappear by a chance in a few generations or survive and expand in the population, again purely by chance. In the same way, changes in regulatory elements are usually deleterious but can be neutral. There are added complexities when we consider neutral mutations that change amino acid sequences in proteins or affect regulatory function. For instance, a change in the structure of a cell surface protein could make the cell and individual less susceptible to infection. Pathogenic organisms adhere specifically to cell surface proteins prior to invasion and a changed protein could be protective. This would in fact make the mutation advantageous in the short term but if it did expand in the population the pathogenic bacteria would evolve to recognize the change and it would become neutral once more. Different cell surface proteins in different individuals will lead to some individuals being at increased risk from certain bacteria and viruses but relatively protected from others; the changes remain neutral overall in evolutionary terms. A further complication is that heterozygous individuals will have different cell surface proteins reducing the degree of surface colonization by any one organism and therefore giving relative protection to the heterozygote compared with the homozygote. The diversity of blood group antigens and leukocyte surface antigens (HLA) reflect this evolutionary interplay between genetic change and susceptibility to different organisms. Changes in regulatory control of genes can also be neutral. An increased cytokine response to infection can be advantageous in certain situations and disadvantageous in others, once again neutral overall. Furthermore, heterozygous advantage will occur if the organism has a choice from two strategies rather than one. If SUDI is caused by a narrow range of organisms then it could be associated with specific neutral mutations and these can be detected by genome-wide association studies in which a large number of SNPs are assayed. But if a wide range of organisms is responsible for sudden death different neutral changes will be ­associated

132

J.A. Morris et al.

with different causal organisms and the associations will be much more difficult to detect. A number of smaller-scale studies of cytokine regulatory gene polymorphisms have been conducted to date with interesting results [42–48] and this does appear to be a promising area of investigation. These studies have provided ­evidence that genetic polymorphisms influencing the balance of pro-inflammatory and antiinflammatory cytokine responses to bacterial and viral infection are associated with risk of SIDS. But the results from larger-scale genome-wide association studies are still awaited at the time of writing.

Deleterious Mutations The mean number of deleterious mutations arising in spermatogenesis or oogenesis and passed on to the next generation is of the order of 0.5–1.5 [41]. The mean number of deleterious mutations in the zygote of individuals that survive and develop into healthy adults is of the order of 6–10. Deleterious mutations are distributed at random during meiosis and so there is a Poisson distribution in zygotes, a mean of 6–10 but a range of 0–12 or above. There is selection against the zygotes with most mutations and therefore the mean in those that survive is approximately 0.5–1.5 less than the mean number in the zygotes that are formed. The result is that the number of deleterious mutations in zygotes does not rise from generation to generation. Selection against deleterious mutations depends on synergistic interaction in genetic networks. Complex biological function is controlled by networks of hundreds or thousands of genes acting in concert. For a complex network to function it needs a degree of redundancy otherwise a single component failure will cause failure of the entire system. Thus one or two deleterious mutations in a complex genetic network have little effect but three or four lead to failure. System failure in a developing embryo causes selection against the deleterious mutations. The number of deleterious mutations in the germ line of live-born infants will vary from 0 through a mean of 8 to a maximum around 12. The lucky few at the lower end of the distribution will have robust systems capable of fighting disease and preserving health into old age. But those at the upper end of the distribution will have genetic networks with two or more deleterious mutations and they will be at increased risk of disease including infection. Bacterial toxins will be more likely to switch off cardio-respiratory function if the genetic networks controlling cardiorespiratory development contain deleterious mutations [49].

Channelopathy Mutations Sodium, potassium, calcium and chloride ions cross cell membranes through ­special ion channels. These channels open or close in response to chemical, electrical or mechanical stimuli. Cardiac rhythm and nerve impulses are controlled by the

6  Sudden Death from Infectious Disease

133

c­ o-ordinated phasic movement of ions through the ion channels. Deleterious ­mutations in genes specifying ion channels lead to defective function and the resulting disease is called a channelopathy [50]. Mutations in cardiac ion channels can cause delayed ventricular repolarization and a prolonged QT interval on the ECG. Individuals with prolonged QT are at increased risk of cardiac arrhythmia and sudden death. Mutations in seven cardiac ion channel genes (KCNQ1, KCNH2, SCN5A, KCNE1, KCNE2, KCNJ2, and CAV3) are a recognized cause of congenital LQTS [51]. The seven cardiac channelopathy genes were examined in 201 Norwegian SIDS cases and 182 controls from the same geographical area [22]. The authors found 8 mutations and 7 rare variants, which they considered to be potential causes of cardiac arrhythmia, in 19 of the 201 cases (9.5%; 95% CI: 5.8 – 14.4%). None of these occurred in the 182 controls. In this study, there were relatively more mutations in the sodium channel and fewer in the potassium channels than seen in congenital LQTS in the general population. Given 50,000 genes in the diploid set, 25,000 in the haploid set, one new deleterious mutation per generation and 7 cardiac channelopathy genes then we can calculate that approximately one in 3,500 live-born children will have a new channelopathy mutation. A mutation not present in the germ line of either parent. But only a minority will cause sudden infant death. Some children will have a prolonged QT and be at risk of cardiac arrhythmia in childhood, in the teenage years or in later life. Some may be phenotypically normal, lead a normal life and pass the mutant gene to their children. Others might manifest ion conduction disorders in the brain or elsewhere (?epilepsy). The prevalence of the channelopathy mutations in the general population is not known but if the ratio of 10 germ-line mutations to one new mutation applies the carriage rate could be of the order of 1 in 350. If these genes cause 10% of SIDS then only one in 50 carriers will die in this way. Obviously, we need to ask what precipitates death? It is not just the gene but a gene–environment interaction. Bacterial toxins that insert into cell membranes and leak ions are prime candidates for the environmental factor [52].

Investigation of SUDI There are standard protocols for the investigation of cases of SUDI and we do not intend to cover all aspects in this section. Instead, it is our intention to concentrate on areas directly relevant to bacterial infection and highlight potential areas for research. Cerebrospinal Fluid The investigation of meningitis in life involves obtaining a specimen of CSF using an aseptic technique to minimize the chance of contamination. The specimen is then examined urgently; a white cell count and differential count, a protein estimation,

134

J.A. Morris et al.

assay for glucose and lactate, and culture for bacteria. Part of the specimen will be sent to a specialized laboratory for viral culture and for nucleic acid amplification and immunoassay to identify specific organisms. The same approach should be used at autopsy. White Cell Count The most sensitive indicator of inflammation of the meninges is the white cell count in the CSF [53]. Normal CSF contains no more than 4 mononuclear cells per cubic mm and no polymorphs (neutrophils). Any increase in the number of white cells above these figures in life is an absolute indicator of meningitis if an acute CSF bleed can be excluded. An acute bleed will result in approximately 500–1,000 red cells per white cell; but subsequently an aseptic meningitis occurs due to breakdown of neutrophils in the CSF. Thus a CSF bleed initially presents with 500–1,000 red cells per white cell, but then over the next few days the number of red cells falls as the white cell count rises. Aseptic meningitis is caused by the breakdown of neutrophils and the release of toxic enzymes but the acute inflammatory reaction leads to an initial rise of neutrophils in the CSF. This potential positive feedback explosion is quickly switched off and the meningeal inflammation becomes predominantly mononuclear. A major problem of interpretation of the CSF white cell count is that the number of mononuclear cells increases after death even in the absence of evidence of meningitis [54, 55]. Platt and colleagues obtained CSF samples in 26 cases of SIDS between 2 and 28 h after death. The mean CSF white cell count was 647 per cubic mm (range 37–3,250 per cubic mm). In adult autopsies the increase was less marked with a mean count of 28 white cells per cubic mm after a mean post-mortem interval of 15 h. In SIDS cases the white cells were mononuclear; 60–70% were lymphocytes and 20–40% monocytes. The authors did not record the presence of neutrophils. This observation of an increased white cell count in the CSF in the absence of histological evidence of meningitis has led many to conclude that the CSF white cell count is of no value in the diagnosis of meningitis at autopsy. But in our opinion this needs to be reassessed [5, 56]. Counting cells in CSF is a far more sensitive indicator of meningitis than is histological examination of tissues. This can be shown using simple calculations. There are estimates of the number of neutrophils produced per day by the bone marrow and the time spent in the circulation is known. The neutrophils finally leave the blood and move rapidly to the body surface to undertake their major function of phagocytosing and destroying bacteria. Calculations indicate that uninflammed tissue has approximately one neutrophil per high power field and the neutrophil will be present in the blood and not in the extravascular tissue. Inflamed tissue has more than one neutrophil per high power field (usually considerably more) and many neutrophils will be marginated in blood vessels and present in extravascular­ tissue. If 800 neutrophils were dispersed in a cubic mm of tissue and the tissue sectioned and examined histologically then there would be one neutrophil per high power field; not enough for a diagnosis of inflammation. But 800 neutrophils per

6  Sudden Death from Infectious Disease

135

cubic mm of CSF would be clearly diagnostic of meningitis. In fact, even 8 ­neutrophils per cubic mm of CSF is diagnostic of meningitis – equivalent to one neutrophil per 100 high power fields of tissue. We cannot continue to regard histological examination of the meninges as the gold standard for the diagnosis of meningitis when it is clear that counting cells in the CSF is potentially much more sensitive. In our opinion, a sample of CSF should be obtained as soon after death as possible and a white cell count and differential count performed. The presence of neutrophils in the absence of red cells should be regarded as evidence of meningitis. A raised mononuclear count should also raise the suspicion of meningitis but more research is needed to define the range of counts that can occur post-mortem in the absence of inflammation. The diagnosis of meningitis by CSF cell count alone in the absence of histological changes does not give a cause of death. It is quite possible that non-fatal disseminated viral infections could be associated with a mild increase in the number of inflammatory cells in the meninges and lead to counts up to 3,000 cells per cubic mm of CSF as seen in SIDS. But we should attempt to diagnose the inflammation and correlate the cell counts with the results of CSF virology. Furthermore, if some cases of SIDS are due to bacterial toxaemia then we would expect to see generalized vascular dilatation with protein and cellular exudation into tissues producing a mild increase in cells in the CSF. Again we need to take the samples early and count the cells to detect these subtle indicators of inflammation. Protein The blood–brain barrier prevents the passive movement of molecules between the blood and the interstitial fluid of the brain and CSF. In adults, the protein concentration in the blood, as an example, is 200-fold higher than in the CSF. Meningeal inflammation leads to increased vascular permeability with exudation of cells and protein into the CSF. The CSF protein level is therefore a sensitive indicator of meningitis [53]. The blood–brain barrier was first demonstrated in post-mortem studies in animals [57]. Ehrlich and his students injected aniline dyes intravenously. The dyes became attached to albumin in the blood. At autopsy, the blood and extravascular tissues were stained blue but the brain and CSF were unstained. The dye had not penetrated the blood–brain barrier in life nor in the agonal phase of death nor in the post-mortem interval prior to the autopsy. The movement of molecules across the blood–brain is an active process and it ceases if the blood supply to the brain is impaired; therefore, there is no reason to assume that the blood–brain barrier will become leaky in the agonal phase. Mangin and colleagues [58] obtained CSF samples within 24  h of death and found a good correlation between protein concentration and the mode of death. If death was rapid, such as in cases of homicide, the CSF protein was within the normal range. But if death was prolonged and associated with inflammation and cytokine release as in patients on intensive care the CSF protein was raised. By ­contrast, Osuna et al. [59] found no relation between the CSF protein level and

136

J.A. Morris et al.

premorbid condition, but the mean post-mortem interval was 48  h. There are no published studies of CSF protein levels in SUDI but this is an obvious area for investigation. If a CSF sample is obtained soon after death the protein level could prove to be a useful marker of meningeal inflammation. CSF Glucose and Lactate Levels Neutrophils rely on anaerobic metabolism for energy. In bacterial meningitis they deplete glucose and produce lactic acid. These changes are useful in bacterial meningitis in life. But glucose levels fall and lactic acid rises after death and assay of these substances is likely to be of limited value. Polymerase Chain Reaction Amplification of DNA can be used to search for evidence of specific viral and bacterial organisms in CSF. The results, however, must be interpreted with care. The mere presence of a virus does not mean it is the cause of death although it could be a contributing factor, for instance by precipitating ALTE. If bacterial specific DNA is recognized it indicates previous bacteraemia but that is not uncommon in infancy. The results of PCR should be correlated with the CSF white cell count and protein level if only to validate the latter measures. Proteomics Examination of body fluids, including CSF, should allow us to prove or disprove the bacterial toxin hypothesis of SUDI. The CSF proteome is already being used as a source of biomarkers of normal and pathological processes [60, 61]. Proteins can be separated by two-dimensional electrophoresis and recognized by their position [62, 63]. A comparison of the secretome of individual bacteria with the protein composition of SUDI CSF samples can be made to identify bacterial products in the CSF samples. The presence of known proteins may be confirmed while this approach will also permit the identification of proteins not previously associated with toxic events. The individual proteins can then be examined by mass spectrometry (MS) and their amino acid sequence determined. The sensitive identification and quantification of bacterial toxins by MS approaches has been reported [63–66] confirming that this approach is feasible and has the capacity to detect ca. 3 pmol/ml of staphylococcal enterotoxin B (SEB) [67]. The next step would be to determine function and to develop specific immunoassays for candidate proteins. The presence of bacteria on the mucosal surfaces together with their toxic products in body fluids at sufficient concentration to derange normal physiology and in the presence of low levels of systemic anti-toxin IgG should be sufficient to convince the sceptic.

6  Sudden Death from Infectious Disease

137

Bacterial toxins released as part of a transient bacteraemia or absorbed directly into the blood stream from mucosal surfaces might not cross the blood–brain ­barrier. For this reason, it will be necessary to examine other body fluids, including blood and urine, using proteomic techniques. But it is probably best to start with CSF for the following reasons: • CSF contains fewer proteins than blood and is easier to analyse. • Post-mortem blood samples commonly contain bacteria of doubtful relevance to the cause of death. Bacterial growth in CSF samples by comparison is less common and more likely to be relevant to the cause of death. • S. aureus, a candidate organism, commonly produces pyrogenic toxins which cause shock and vascular collapse with capillary dilatation and protein exudation. Thus, there is a good chance that relevant toxins will be found in the CSF, and it is likely that the action of toxins will impact the presence/abundance of a range of other macromolecules. • It is relatively easy to obtain a clean specimen of CSF by cisternal puncture. Blood Culture A blood culture should be taken prior to autopsy and as soon after death as possible. A review of over 5,000 autopsies including 468 cases of SUDI with 4,992 blood cultures showed that if careful precautions are taken to reduce contamination then approximately two-thirds of blood cultures will be sterile, two in nine yield a single isolate and one in nine a mixed growth [68]. The bacteria that are grown could have entered the blood: • Prior to death. • During the agonal phase of death or during the period of resuscitation. • After death and prior to autopsy especially if the body is not stored at an appropriate temperature. • Contamination when the specimen is obtained. Careful technique will reduce the chance of contamination and the possibility of post-mortem growth and spread is greatly reduced if the body is refrigerated prior to autopsy. Thus, the majority of organisms found will have entered prior to death or in the agonal phase. Weber and colleagues [69] reviewed 546 cases of SUDI done at one institution over 10 years. Post-mortem isolates from blood, spleen, CSF, and lung were classified as follows: • Non-pathogens • Group 1 pathogens – organisms usually associated with an identifiable focus of infection • Group 2 pathogens – organisms known to cause septicaemia without an obvious focus of infection The authors found that group 2 pathogens were more commonly isolated from cases of unexplained SUDI than cases of SUDI that were explained by a non-infective

138

J.A. Morris et al.

cause. The group 2 pathogens included S. aureus (16% in unexplained SUDI vs. 9% in explained non-infective SUDI) and E. coli (6% in unexplained SUDI vs. 1% in explained non-infective SUDI). The conclusion to draw from this work is that if a significant pathogen, which includes S. aureus and E. coli, is isolated from blood, even in mixed culture, then it is a possible cause of death. If a significant pathogen is found in association with subtle or more marked evidence of an inflammatory reaction then it becomes the likely cause of death. However, the mere presence of bacteria in the blood is not an adequate explanation for death. For death to occur there must be either secretion of toxins or an abnormal and harmful immune response to the bacteria. Respiratory Tract Bacterial Flora A specimen of upper respiratory bacterial flora should be obtained by nasal, pernasal or throat swab. The upper respiratory tract is the likely site of origin for bacteria that cause lower respiratory tract infection, septicaemia, systemic toxaemia and transient bacteraemia. Isolation of S. aureus, E. coli, S. pneumoniae, N. meningitidis or group B streptococci would be noteworthy and could be the only evidence of the causative organism if blood, lung and CSF cultures are negative. Specimens of the lower respiratory tract should also be obtained and analysed as for blood cultures, significant pathogens give a possible cause of death. This becomes a likely cause if there is histological evidence of inflammation. Gastrointestinal Tract Flora Analysis of the gastrointestinal flora is much more difficult than the respiratory tract flora. There are between 40 and 80 bacterial species in faecal flora at any one time and many cannot be grown in conventional culture. A specimen can be examined for specific bacterial and viral pathogens but it is probably best to concentrate on the respiratory tract for the analysis of the contribution of common bacteria and common toxins to SUDI. However, bacteria growing in the upper respiratory tract are cleared into the oesophagus and will be carried transiently in the colon. Thus there could be value in analyzing faecal samples for organisms such as S. aureus to assess recent upper respiratory tract events. Urine The bladder is usually empty when cases of sudden infant death are examined at autopsy. But samples of urine can be obtained in infants presenting with ALTE, HSE and HIE. Furthermore, urine samples can be obtained from healthy infants acting as controls. One of the main problems in conducting research into SUDI is the lack of controls. One cannot obtain closely matched control samples of CSF or

6  Sudden Death from Infectious Disease

139

blood for SUDI cases. But it would be possible to get closely matched control ­samples of urine for ALTE, HSE, and HIE. Furthermore, any toxins detected in urine would reflect toxaemic events in the preceding few hours and could therefore monitor episodes of transient bacteraemia. We have preliminary results suggesting that an ELISA technique can detect staphylococcal toxins (TSST and SEC) in picogram quantities in urine and we have observed that whilst the majority of urine samples from healthy infants are negative up to 50% of infants with a viral URTI have detectable toxin excretion [70]. The latter is presumably a result of staphylococcal growth in the upper respiratory tract causing episodes of transient staphylococcal bacteraemia. In future, we need to use proteomic techniques to detect specific bacterial toxins in urine and then develop immunoassays for the toxins and compare urine samples in ALTE, HSE and HIE with closely matched normal healthy infants. In this way, we can prove or disprove the common bacterial toxin hypothesis.

Sudden Unexpected Death in Childhood Sudden unexpected death in childhood (SUDC) should be defined in a way that is closely analogous to SUDI. It is the death of a child over 1 year of age that is sudden and unexpected by history. If the cause is unexplained after a detailed autopsy, then the case is unexplained SUDC. If a cause is found, then the condition is explained SUDC. However, not all authors follow this convention and in some articles SUDC is defined as sudden unexplained death in childhood [71]. Krous and colleagues [71] reported 50 cases of SUDC in which 36 cases remained unexplained after a detailed autopsy and death scene investigation. The crude death rate for unknown and unspecified cause in children aged 1–4 in the USA is approximately 1.5 per 100,000 live births compared with 56 per 100,000 live births for SIDS (figures for 2001). SUDC is more common in boys than girls, death usually occurs during sleep, the child is often found face down, and there is commonly a family history of seizures. Detailed neuropathological studies have revealed dysmorphic changes in the hippocampus in a number of cases [72]. Children with LQTS are at increased risk of sudden death due to cardiac arrhythmia. In 3,015 LQTS children followed from 1 to 12 years, the cumulative probability of the combined end point aborted cardiac arrest and sudden death in males and females was 5% and 1%, respectively [73]. In 2,772 participants from the international long QT registry followed from aged 10 to 20 years; 81 had aborted cardiac arrests and 45 had sudden cardiac death [74]. The risk factors for cardiac arrest or death were corrected QT greater than 530 ms, recent syncope and male gender. It is important to realize that not all children with LQTS have recognized mutations in the seven channelopathy genes examined to date. Furthermore, not all children with significant mutations in the channelopathy genes have LQTS. This is an area of active research and we need to determine the prevalence of channelopathy mutations in SUDC. Since the mutations are rare this means we need to perform the examination in every case. Furthermore, since most children with mutations do not

140

J.A. Morris et al.

die or suffer cardiac arrest we need to search for environmental triggers and ­bacterial toxaemia is an obvious candidate. Pneumonia, septicaemia, pyelonephritis and meningitis can cause explained SUDC just as they can cause explained SUDI. The evidence depends on bacteriological and histological examination and the protocols for SUDC should be the same as for SUDI. Finding appropriate pathogens in association with inflammation in the lungs, kidneys or meninges allows a confident diagnosis of infection as the cause of death. But the situation is much less clear if a pathogen is found in blood, CSF or lung tissue in the absence of any evidence of inflammation. There is not a convincing body of evidence linking bacterial toxaemia to SUDC as there is for SUDI. The evidence does not exist because the studies have not been done and we recommend the following: 1. A comparison of the upper respiratory bacterial flora in SUDC and in age-, gender-, and season-matched healthy children from the same geographical area. 2. A comparison of bacterial isolates from lung, blood and CSF in explained and unexplained SUDC. 3. Proteomic analysis of CSF, blood and urine in all SUDC. Infants are at risk of sudden death from bacterial toxins because of a relative lack of protective IgG. This is most marked at 2–3 months of age when circulating IgG is at its lowest level. The risk then falls as IgG rises and most infants are fully protected by 12 months of age. The risk of toxin-induced sudden death in childhood is therefore much less. But in a biological system protection by IgG will never be absolute and we would anticipate some deaths after 12 months of age. Furthermore, we can predict that in later life as the protective mechanisms gradually wane the risk of toxin-induced sudden death will reappear.

Sudden Unexpected Death in Adults The majority of cases of SUDA in the UK are carried out by general histopathologists under the direction of HM Coroner. In most cases, this involves a detailed dissection with macroscopic examination of tissues but without any ancillary tests. If the cause is unexplained at the end of this procedure then ancillary investigations can be performed, but it is often too late to obtain uncontaminated lung, blood and CSF specimens. The quality of bacteriological examination, therefore, is often poor. The risk of sudden death due to LQTS continues throughout life and therefore analysis of channelopathy genes is appropriate in unexplained SUDA. The principles are the same as in SUDC and SUDI in that channelopathy mutations can cause sudden death even in those who do not have LQTS. Studies to establish the prevalence of channelopathy mutations in SUDA need to be undertaken. The role of bacterial toxins as a cause of SUDA also needs to be determined. This will require proteomic analysis of CSF, blood and urine in adult autopsies comparing the results in explained and unexplained deaths. Bacteria are commonly isolated

6  Sudden Death from Infectious Disease

141

if blood cultures are obtained prior to autopsy [68]. These findings are too often dismissed as contaminants, even though the specimens are obtained with great care; or as the result of agonal change, a mechanism which lacks experimental verification. Perhaps, in truth, transient bacteraemia is a common last event on the road to death at all ages.

Conclusion Most of the evidence for the role of infection in sudden death comes from studies of sudden unexpected death in infancy (SUDI). In this chapter, we have reviewed the evidence in relation to SUDI and then extrapolated to sudden unexpected death in childhood (SUDC) and sudden unexpected death in adults (SUDA). There is a considerable body of evidence, although it still falls short of absolute proof, that common bacterial toxins can cause sudden death in infancy. The fact that most infants are well when last seen and die quickly is consistent with direct toxic action on cardio-respiratory function; the toxins having been absorbed from the mucosal surface or delivered as part of a transient bacteraemia. Studies to date implicate Staphylococcus aureus and Escherichia coli as candidate organisms, but it is likely that other bacteria are also involved. The nature nurture controversy in biology has been resolved by simply accepting that nature and nurture act together in development, in determining function and in the pathogenesis of disease. Bacteria and viruses cause disease in those rendered susceptible by virtue of their genetic constitution. Neutral mutations in the germ line cause some individuals to have increased susceptibility to certain infectious agents but render them less susceptible to others; thus, neutral overall in evolutionary terms. The search for links between neutral mutations and common diseases can now be undertaken in genome-wide association studies by measuring single nucleotide polymorphisms (SNPs). Germ-line deleterious mutations interact synergistically to impair function in complex genetic networks and thereby increase susceptibility to disease. To date, deleterious mutations in channelopathy genes have been shown to cause sudden death in infants, children and adults; but others will be recognized in the near future as this is a very active area of research. The majority of live-born infants with germ-line channelopathy mutations do not die in infancy or childhood and so we must consider the role of gene–environment interactions. In this respect, the fact that both S. aureus and E. coli produce toxins (cytolysins and colicins) that create ion channels in cell membranes and disturb ion flow is worthy of note. If bacterial toxin–gene interactions can lead to death in infancy then it is likely that similar mechanisms act throughout life. This proposition can now be proven or disproven using the scientific techniques of genomics and proteomics. Large-scale autopsy studies are needed in which good quality specimens of CSF, blood and urine are obtained. Autopsy pathology has been displaced from its central role in academic medicine in recent years but there is now an opportunity for pathologists to take an intellectual lead in determining, at a molecular level, the mechanisms of sudden death following bacterial toxaemia.

142

J.A. Morris et al.

References 1. Fleming P, Blair P, Bacon C et al (2000) Sudden unexpected deaths in infancy: The CESDI SUDI studies 1993–1996. The Stationery Office, London 2. Brook I (2003) Unexplained fever in young children: how to manage severe bacterial infection. BMJ 327:1094–1097 3. Weber MA, Ashworth MT, Risdon RA et al (2008) The role of post-mortem investigations in determining the cause of sudden unexpected death in infancy (SUDI). Arch Dis Child 93:1048–1053 4. Morris JA, Haran D, Smith A (1987) Hypothesis: common bacterial toxins are a possible cause of the sudden infant death syndrome. Med Hypotheses 22:211–222 5. Morris JA, Harrison LM, Biswas J et al (2007) Transient bacteraemia: a possible cause of sudden life threatening events. Med Hypotheses 69:1032–1039 6. Morris JA (1999) The common bacterial toxin hypothesis of sudden infant death syndrome. FEMS Immunol Med Microbiol 25:11–17 7. Telford DR, Morris JA, Hughes P et al (1989) The nasopharyngeal flora in the sudden infant death syndrome. J Infect 18:125–130 8. Gilbert R, Rudd P, Berry PJ et al (1992) Combined effect of infection and heavy wrapping on the risk of sudden unexpected infant death. Arch Dis Child 67:171–177 9. Lee S, Barson AJ, Drucker DB et al (1987) Lethal challenge of gnotobiotic weanling rats with bacterial isolates from cases of sudden infant death syndrome (SIDS). J Clin Pathol 40:1393–1396 10. Sayers NM, Drucker DB, Morris JA (1995) Lethal synergy between toxins of staphylococci and enterobacteria: implications for sudden infant death syndrome. J Clin Pathol 48: 929–932 11. Sayers NM, Drucker DB, Telford DR et al (1995) Effects of nicotine on bacterial toxins associated with cot death. Arch Dis Child 73:549–551 12. Bell S, Crawley BA, Oppenheim BA et  al (1996) Sleeping position and the upper airways bacterial flora; possible relevance to cot death. J Clin Pathol 49:170–172 13. Harrison LM, Morris JA, Telford DR et al (1999) The nasopharyngeal bacterial flora in infancy: effects of age, gender, season, viral upper respiratory tract infection and sleeping position. FEMS Immunol Med Microbiol 25:19–28 14. Zorgani A, Essery SD, Al Madani OA et al (1999) Detection of pyrogenic toxins of Staphylococcus aureus in sudden infant death syndrome. FEMS Immunol Med Microbiol 25:101–108 15. Malam JE, Carrick GF, Telford DR et al (1992) Staphylococcal toxins and sudden infant death syndrome. J Clin Pathol 45:716–721 16. Oppenheim BA, Barclay GR, Morris JA et al (1994) Antibodies to endotoxin core in sudden infant death syndrome. Arch Dis Child 70:95–98 17. Blair PS, Platt MW, Smith IJ et al (2006) Sudden infant death syndrome and time of death: factors associated with night-time and day-time deaths. Int J Epidemiol 35:1563–1569 18. Meny RG, Carroll JL, Carbone MT et  al (1994) Cardiorepiratory recordings from infants dying suddenly and unexpectedly at home. Pediatrics 93:44–49 19. Poets CF, Meny RG, Chobanian MR et al (1999) Gasping and cardiorespiratory patterns during sudden infant deaths. Pediatrics 45:350–354 20. Alouf JE, Freer JH (1991) The comprehensive source book of bacterial protein toxins. Academic Press, London, UK 21. Raza MW, Balckwell CC et  al (1999) Sudden infant death syndrome, virus infections and cytokines. FEMS Immunol Med Microbiol 25:85–96 22. Arnestad M, Crotti L, Rognum TO et al (2007) Prevalence of Long QT syndrome gene variants in sudden infant death syndrome. Circulation 115:361–367 23. Levin M, Hjelm M, Kay LDS et al. (1983) Haemorrhagic shock and encephalopathy: a new syndrome with a high mortality in young children. Lancet ii:64–67 24. Bacon CJ, Scott D, Jones P (1979) Heatstroke in well wrapped infants. Lancet i:422–425 25. Bacon CJ, Bellman MH (1983) Heatstroke as a possible cause of encephalopathy in infants. BMJ 287:328

6  Sudden Death from Infectious Disease

143

26. Bacon CJ, Hall SM (1992) Haemorrhagic shock and encephalopathy syndrome in the British Isles. Arch Dis Child 67:985–993 27. Morris JA (1983) Haemorrhagic shock and encephalopathy. Lancet ii:686 28. Bacon CJ, Bell SA, Gaventa JM et al (1999) Case control study of thermal environment preceding haemorrhagic shock encephalopathy syndrome. Arch Dis Child 81:155–158 29. Van Lierde S, van Leeuwen WJ, Ceuppens J et al (1997) Toxic shock syndrome without rash in a young child: link with syndrome of haemorrhagic shock and encephalopathy. J Pediatrics 131:130–134 30. McGovern MC, Smith MBH (2004) Causes of apparent life threatening events in infancy; a systematic review. Arch Dis Child 89:1043–1048 31. Kiechl-Kohlendorfer U, Hof D, Peglow U et al (2005) Epidemiology of apparent life threatening events. Arch Dis Child 90:297–300 32. Poets CF, Samuels MP, Noyes JP et al (1993) Home event recordings of oxygenation, breathing movements, and heart rate and rhythm in infants with recurrent life threatening events. J Pediatr 123:693–701 33. Geddes JF, Hackshaw AK, Vowles JH et al (2001) Neuropathology of inflicted head injury in children: 1. patterns of brain damage. Brain 124:1290–1298 34. Geddes JF, Vowles JH, Hackshaw AK et al (2001) Neuropathology of inflicted head injury in children: 2 Microscopic brain injury in infants. Brain 124:1299–1306 35. Byard RW, Blumberg P, Rutty G et al (2007) Lack of evidence for a causal relationship between hypoxic-ischaemic encephalopathy and subdural haemorrhage in fetal life, infancy and childhood. Pediatric Develop Pathol 10:348–350 36. Cohen MC, Scheimberg I (2008) Evidence of occurrence of intradural and subdural haemorrhage in the perinatal and neonatal perid in the context of hypoxic ischaemic encephalopathy. An observational study from two referral centres. Pediatr Dev Pathol. doi:10.2350/08-08-0509.1 37. Nei M (1987) Molecular evolutionary genetics. Columbia University Press, New York 38. Maynard-Smith J (1997) Evolutionary genetics, 2nd edn. Oxford University Press, Oxford 39. Drake JW, Charlesworth B, Charlesworth D et  al (1998) Rates of spontaneous mutation. Genetics 148:1667–1686 40. Baltimore D (2001) Our genome unveiled. Nature 409:814–816 41. Morris JA (2005) Synergistic interaction of heterozygous deletions impairs performance and confers susceptibility to disease at all ages. Med Hypotheses 65:483–493 42. Summers AM, Summers CW, Drucker DB et al (2000) Association of IL-10 genotype with sudden infant death syndrome. Human Immunol 61:1270–1273 43. Gordon AE, MacKenzie DAC, El Ahmer OR et al (2002) Evidence for a genetic component in sudden infant death syndrome. Child Care Health Dev 28(Suppl 1):27–29 44. Korachi M, Pravica V, Barson AJ et al (2004) Interleukin-10 genotype as a risk factor for sudden infant death syndrome: determination of IL-10 genotype from wax embedded postmortem samples. FEMS Immunol Med Microbiol 42:125–129 45. Moscovis SM, Gordon AE, Al Madani OM et al (2004) Interleukin-10 and sudden infant death syndrome. FEMS Immunol Med Microbiol 42:130–138 46. Moscovis SM, Gordon AE, Hall ST et al (2004) Interleukin 1-beta responses to bacterial toxins and sudden infant death syndrome. FEMS Immunol Med Microbiol 42:139–145 47. Blackwell CC, Moscovis SM, Gordon AE et al (2005) Cytokine responses and sudden infant death syndrome: genetic, developmental, and environmental risk factors. J Leukocyte Biol 78:1242–1254 48. Dashash M, Pravica V, Hutchinson IV et al (2006) Association of sudden infant death syndrome with VEGF and IL-6 gene polymorphisms. Human Immunol 67:627–633 49. Morris JA (2004) Common bacterial toxins and physiological vulnerability to sudden infant death: the role of deleterious genetic mutations. FEMS Immunol Med Microbiol 42:42–47 50. Schwartz PJ, Crotti L et al (2008) Ion channel disease in children: manifestations and management. Curr Opin Cardiol 23:184–191 51. Marita H, Wu J, Zipes DP (2008) The QT syndromes: long and short. Lancet 372:750–763 52. Morris JA, Harrison LM, Brodison A et al (2009) Sudden infant death syndrome and cardiac arrhythmias. Future Cardiol 5:201–207

144

J.A. Morris et al.

53. Fishman RA (1992) Cerebrospinal fluid in diseases of the nervous system. W B Saunders, London 54. Platt MS, McClure S, Clarke R et al (1989) Postmortem cerebrospinal fluid pleocytosis. Am J Forens Med Pathol 10:209–212 55. Wyler D, Marty W, Bar W (1994) Correlation between the postmortem cell content of cerebrospinal fluid and time of death. Int J Legal Med 106:194–199 56. Morris JA, Harrison LM (2007) The microbiological investigation of sudden unexpected death in infancy. In: Kirkham N, Shepherd NA (eds) Progress in pathology. Cambridge University Press, Cambridge 57. Ehrlich P (1885) Das Sauerstoffbeduerfnis des Organismus. Eine farbenanalytische Studie, A Hirschfeld, Berlin 58. Mangin P, Lugnier AA, Chaumont AJ et al (1983) Forensic significance of postmortem estimation of the blood cerebrospinal fluid barrier permeability. Forens Sci Int 22:143–149 59. Osuna E, Perez-Carceles MD (1992) Luna A (1992) Efficacy of cerebrospinal fluid biochemistry in the diagnosis of brain insult. Forens Sci Int 52:193–198 60. Romeo MJ, Espina V, Lowenthal M et al (2005) CSF proteome: a protein repository for potential biomarker identification. Expert Rev Proteomics 2:57–70 61. Maurer MH (2008) Proteomics of brain extracellular fluid (ECF) and cerebrospinal fluid (CSF). Mass Spectrom Rev. doi:10.1002/mas.20213 62. Blackstock WP, Weir MP (1999) Proteomics: quantitative and physical mapping of cellular proteins. Trends Biotechnol 17:121–126 63. Pocsfalvi G, Cacace G, Cuccurullo M et al (2008) Proteomic analysis of exoproteins expressed by enterotoxigenic Staphylococcus aureus strains. Proteomics 8:2462–2476 64. Kawano Y, Ito Y, Yamakawa Y et al (2000) Rapid isolation and identification of staphylococcal exoproteins by reverse phase capilliary high performance liquid chromatography-electrospray ionization mass spectrometry. FEMS Microbiol Lett 189:103–108 65. Nedelkov D, Nelson RW (2003) Detection of staphylococcal enterotoxin B via biomolecular interaction analysis mass spectrometry. Appl Environ Microbiol 69:5212–5215 66. Callahan JH, Shefcheck KJ, Williams TL et al (2006) Detection, confirmation and quantification of staphylococcal enterotoxin B in food matrixes using liquid chromatography-mass spectrometry. Annal Chem 78:1789–1800 67. Kientz C, Hulst AG, Wils ER (1997) Determination of staphylococcal enterotoxin B by on-line (micro) liquid chromatography-electrospray mass spectrometry. J Cromatogr A 757:51–64 68. Morris JA, Harrison LM, Partridge SM (2006) Postmortem bacteriology: a re-evaluation. J Clin Pathol 59:1–9 69. Weber MA, Klein NJ, Hartley JC et  al (2008) Infection and sudden unexpected death in infancy: a systematic retrospective case review. Lancet 371:1848–1853 70. Harrison LM, Morris JA, Lauder RM et al (2009) Staphylococcal pyrogenic toxins in infant urine samples; a possible marker of transient bacteraemia. J Clin Path 62:735–738 71. Krous HF, Chadwick AE, Crandall L et al (2005) Sudden unexpected death in childhood: a report of 50 cases. Pediatr Dev Pathol 8:307–319 72. Kinney KC, Armstrong DL, Chadwick AE et al (2007) Sudden death in toddlers associated with developmental abnormalities of the hippocampus: a report of five cases. Paediatr Dev Pathol 10:208–223 73. Goldenberg I, Moss AJ, Peterson DR et al (2008) Risk factors for aborted cardiac arrest and sudden cardiac death in children with congenital long-QT syndrome. Circulation 117:2184–2191 74. Hobbs JB, Peterson DR, Moss AJ et al (2006) Risk of aborted cardiac arrest or sudden cardiac death during adolescence in the long-QT syndrome. JAMA 296:1249–1254

Chapter 7

Aviation Deaths S. Anthony Cullen

Abstract  Deaths in aircraft accidents are uncommon. Worldwide there were 11 fatal scheduled air service accidents in 2007 resulting in 587 fatalities (Insurance Information Institute, http://www.iii.org/media/research, 2008). General aviation accounts for a comparable number of deaths. In the USA alone in 2007 there were 284 general aviation accidents resulting in 491 fatalities (National Transportation Safety Board, http://www.ntsb.gov/aviation/Table10.htm, 2008). These figures are small compared with deaths in road traffic accidents; in Great Britain 2,946 people were killed in road traffic accidents in 2007 (Department for Transport, http://www. dft.gov.uk/pgr/statistics/datatablepublications/accidents/casuatiesmr/rcgbmainresults2007, 2008). However, major commercial aircraft accidents attract a great deal of publicity and there is pressure to complete a thorough investigation. Keywords  Aviation • Aircraft accidents • Trauma death • Forensic pathology • Reconstruction

International and National Laws and Treaties The Standards and Recommended Practices for Aircraft Accident Inquiries were first adopted in 1951 pursuant to Article 37 of the Convention on International Civil Aviation and were designated as Annex 13 to the Convention [1]. Article 26 imposes an obligation on the State in which an aircraft accident occurs to institute an inquiry in accordance with the procedure laid down by the International Civil Aviation Organisation, an agency of the United Nations representing 190 member states.

S.A. Cullen () Lately Consultant Pathologist, RAF Centre of Aviation Medicine, RAF Henlow, Bedfordshire SG16 6DN, UK E.E. Turk (ed.), Forensic Pathology Reviews, Volume 6, Forensic Pathology Reviews 6, DOI 10.1007/978-1-61779-249-6_7, © Springer Science+Business Media, LLC 2011

145

146

S.A. Cullen

Subsequent meetings of the Assembly directed efforts towards a procedure to make available promptly the reports of aircraft accident investigations, particularly when large transport aircraft were involved. The contracting States were urged to provide timely notification of all aircraft accidents to the State of manufacture of the aircraft and the State in which the aircraft is registered. Annex 13 defines an accident. This states, in part, that an accident is an occurrence associated with the operation of an aircraft which takes place between the time any person boards the aircraft with the intention of flight until such time as all such persons have disembarked, in which a person is fatally or seriously injured as a result of being in the aircraft, or being in direct contact of any part of the aircraft, including parts which have become detached from the aircraft, or by direct exposure to jet blast. Excluded from this definition are injuries from natural causes, selfinflicted injuries, injuries inflicted by other persons and injuries to stowaways hiding outside the areas normally available to the passengers and crew. It is clearly stated that the sole objective of the investigation shall be the prevention of accidents and incidents and not to apportion blame or liability. It places an obligation on the State in which the accident occurs to preserve the evidence and maintain safe custody of the aircraft and its contents. It requires the State of Occurrence to notify the States of Registry, Operation, Design and Manufacture of the aircraft and the International Civil Aviation Organisation when the accident involves an aircraft whose maximum mass is greater than 2,250 kg. These States have the right to appoint accredited representatives to the inquiry. They have the duty to provide information about the aircraft, its crew and any other relevant information pertaining to the inquiry. The State of Occurrence may delegate the whole or part of the conduct of the investigation to one of the other concerned States. Annex 13 to the Convention defines in detail the responsibilities of all concerned and the format of the reports and their distribution. The regulation of aircraft accident investigation in the Unite Kingdom is embodied in the Civil Aviation Acts of 1949, 1980, 1982 and 2006 that give the inspectors of accidents the authority to conduct their investigations. The detailed regulations are found in the Civil Aviation (Investigation of Air Accidents) Regulations 1996. These state amongst other things that “the fundamental purpose of investigating accidents under these regulations shall be to determine the circumstances and causes of the accident with a view to the preservation of life and the avoidance of accidents in the future; it is not the purpose to proportion blame or liability”. The Air Accident Investigation Branch (AAIB) of the Department for Transport employs the Inspectors of Accidents. The statutes that define the procedures to be followed in aircraft accident investigation and the powers of the Inspectors of Air Accidents are found in the Civil Aviation (Investigation of Air Accidents) Regulations 1996 and the Air Navigation (Investigations of Air Accidents Involving Civil and Military Aircraft or Installations) Regulations 1986. These regulations define accidents and incidents and these are summarised in the Air Accident Investigation Branch Memorandum on the Investigation of Civil Air Accidents [2]. The report prepared by the Inspector is submitted by the Chief Inspector to the Secretary of

7  Aviation Deaths

147

State who has no authority to alter its contents. It is the Chief Inspector of Accidents and not the Secretary of State who decides whether or not an investigation is to be carried out. Regulations allow for a review of the report if this is required. When there is a particular serious accident with many fatalities the Secretary of State may order a public inquiry. In this case, the Lord Chancellor in England will appoint the Commissioner who conducts the inquiry. Technical assessors will assist him. The AAIB investigation will continue and the inspectors from that organisation and others will be called as witnesses to the public inquiry. A public inquiry is seen to alleviate public anxiety and to remove all suspicion of a cover-up by any government department. Before a report is submitted to the Secretary of State the inspector conducting the investigation is obliged to notify everyone whose reputation is, in the inspector’s opinion, likely to be adversely affected by the report, inviting them to make representations. Such people may obtain legal advice and representation before responding. The inspector is required to consider all the points put to him and to amend his report, if necessary. If the participant is still unhappy he may ask for a review board, which the Secretary of State is not permitted to refuse. A Queen’s Counsel normally chairs a review board with two technical assessors. It reports to the Secretary of State who orders that both the inspector’s report and the review board’s report are published, usually within the same cover. Fatal accidents in England, Wales and Northern Ireland come under the jurisdiction of the coroner in whose area the accident occurs. He will appoint an appropriately qualified pathologist to perform the autopsies on his behalf (Coroner’s Rules 1984). Coroners have the right to impound evidence relating to the death being investigated. In an aircraft accident this can include the aircraft wreckage, flight data recorder and all the navigation and technical records of the flight. The inspectors of accidents also have the legal right to impound the very same evidence. Theoretically, there could be a conflict of interest. However, the roles of Coroners’ inquiries and AAIB inquiries are complementary and there is rarely any conflict between the two. In Scotland, the judicial office of Coroner does not exist and the Procurator Fiscal, whose office covers a small part of the Coroner’s duties, is a part of the Lord Advocate’s department. The Procurator Fiscal applies to the Sheriff’s Court for permission to order an autopsy in many cases where such an investigation is the normal practice in England. However, he may accept multiple injuries as a cause of death deduced solely from an external examination of the victim. He will look into the matter of fatal aviation accidents and will, if he considers it appropriate, raise the matter at a Fatal Accident Inquiry in the Sheriff’s Court. In the USA each State has its own laws regarding accidental deaths. As a rule the next of kin of the fatality must give permission for an autopsy. The exception to this requirement is when public interest outweighs the rights of the family members. In aviation accidents there is a federal interest because of the federal regulations, exercised through the Federal Aviation Administration, which controls aviation in America. This federal interest was formalised by the Federal Aviation Act of 1958

148

S.A. Cullen

that gave the federal government power to promulgate regulations governing reports of accidents involving civil aircraft. They were also empowered to make recommendations designed to prevent future accidents. The individual States do, however, have a legitimate interest in aircraft accident fatalities. The 1962 amendment to the Federal Aviation Act does not require that autopsies be obtained; it authorises the NTSB investigator to request them. The State Coroner or Medical Examiner has different reasons for requiring an autopsy. Some States authorise the coroner to obtain autopsies only in cases of persons who may have died by unlawful means. This conflict of authority may give rise to potential conflict and confusion, as some State officials may be unaware of the federal legislation. In continental Europe the law in most countries is based on the Napoleonic Code. The role of the investigating magistrate or chief of police is similar to that of the coroner in that he is mainly concerned to find out who has died and to ensure that there are no criminal aspects to the case. Some authorities require no more than a list of those on board the aircraft and, if the number of bodies agrees, they are willing to accept that those named on the manifest had all died. The investigating magistrate in these countries has the right to the wreckage, flight data recorder and other records. Many countries have a system by which the officials from the aircraft accident investigation organisation, such as the Bureau d’Enquete d’Accident in France, are formally recognised by the examining magistrate. This enables the professional investigators to have access to the wreckage of the aircraft and its records in their search for a cause of the accident. In most countries, there are several obligatory statutory investigations into deaths arising from aircraft accidents. In the UK, the Coroner is concerned with individual deaths while the AAIB or military Board of Inquiry are concerned with accident investigation. All have powers to impound the wreckage. If there is a possibility of crime the police may be involved and they will also represent the Coroner. The various investigations overlap and the possibility arises that different conclusions may be reached by the different investigating authorities. In the USA, the situation is even more complex. State or federal authorities may claim jurisdiction and the military may also be involved. Similarly, investigation by state police and the Federal Bureau of Investigation may be needed. The situation is particularly complex with military accidents when the authority comes from the Commander in Chief, the President. However, state officials may not recognise the claims of federal jurisdiction when it is based solely on military regulations. Civil authorities usually require compliance with state law before granting approval for autopsies. Comparable problems arise in those countries whose law is based upon the Napoleonic Code. In one accident to a British registered aircraft that occurred in a European country, the investigating magistrate seized the flight data recorder and refused to allow the UK accredited representative to take it to the UK for read out. It was not possible to obtain this in that country and after a considerable amount of time he took it to England himself. However, much time was wasted and the concern was that a similar aircraft might suffer the same fate while the cause of the accident was unknown.

7  Aviation Deaths

149

The Investigation of Aircraft Accidents The investigation covers three broad areas that explore the reasons for the ­breakdown in the interface between man, machine and environment. The human factors include the examination of the working environment with special emphasis on the medical, behavioural and survival aspects. Aircraft factors cover the design, performance and certification of the aircraft. The operational factors are concerned with the operational aspects including aviation practices, air traffic control, airports, weather, procedures, training and maintenance. These three broad interactive areas cover all facets of the investigation but experience has shown that, in all but the smallest accident, the number of “Groups” will need to be expanded. Pathologists will be involved with the “Human Factors Group” which is responsible for all aeromedical aspects of the crew’s performance including physical, physiological and psychological elements. They are concerned with the possibility of crew incapacitation. Evacuation and design factors that may have a bearing on survival are examined together with the crashworthiness of the aircraft. An examination of the injuries sustained by the fatalities may contribute to the resolution of the aircraft’s mode of impact. They work closely with the local authorities in the matter of body recovery, identification and subsequent post-mortem examination.

Action at the Scene The police and emergency services will be the first at the scene of a disaster. Their first concern will be the saving of life. If this is impossible, effective investigation aimed at saving lives in the future assumes major importance. Planning and training ensure that the emergency services are aware of their contribution to successful investigation. However, there is usually public pressure for a speedy disposal of the fatal casualties and restoration to their next of kin. Investigation can impede this process and co-operation between the police and the accident investigators is essential. A team of doctors who are not involved in caring for the injured will be needed in order to certify formally the fact of death. At the scene the main contributions of the police to the investigation are the maintenance of the security of the area and the preservation of evidence, particularly the distribution of the bodies and the aircraft wreckage. Security includes both security from bystanders and from the presence of well-intentioned officials. It is important to ascertain if the aircraft is carrying any hazardous freight before further action is taken. Photographs should be taken as soon as possible after the accident occurs and before the wreckage is moved or disturbed. Efforts should be made to preserve or record evidence of an ephemeral nature such as ice and soot deposits. If possible, photographs of the bodies should be taken before removal. However, there is considerable public pressure to remove the dead bodies as soon as possible and, in general, the lack of such photographs has not impeded the investigation provided the location of each body in the wreckage is known. Mapping the

150

S.A. Cullen

locations of the fatalities is more important. This is best achieved with a system of staking and labelling. This is normally the responsibility of the police. However, it is important that they are made aware of any potential hazards and matters of health and safety. They may need to obtain advice from environmental health officers and consultants in communicable disease control. All the personnel involved in the recovery procedures should be provided with appropriate protective clothing. Once death has been confirmed, a uniquely numbered label should be securely attached to the body. Most disaster kits contain pre-printed labels in triplicate for this purpose. The bodies should not be undressed or searched at the scene, nor should any property be removed from them or from their clothing. The recovery teams will then place the body or remains into a body bag and fix identically numbered labels to both the bag and the location from which it was removed. The location markers could be stakes or freestanding markers. Pin-on labels are used when fatalities are retained in their seats. It is sometimes useful for the pathologist to view the bodies in situ but this is often not possible. Parts of bodies should be treated in a similar manner to whole bodies. Once the coroner has given permission to remove the bodies they may be moved from the site but a strict procedure is needed to avoid cross contamination and to preserve the evidence. The bodies should then be taken to a holding area close to the accident site before being taken to the temporary mortuary. Ideally, this holding area should be under cover and out of public view. It should be accessible to vehicles and be secure. This holding area serves as a collection point and as a checking point to ensure that labelling of body bags is complete. Refrigerated lorries may also attend the holding area to allow appropriate storage and phased and orderly transfer of the bodies to the temporary mortuary. The body bags remain sealed throughout their time in the holding area. Accident investigators and police are now able to make detailed maps of the scene. This can be done when the bodies have been removed and the site is free and may be some days after the accident. The preparation of these maps can provide valuable information for the investigation. In one case with a suspected fire in the air, the body plot showed that the burnt bodies had been burnt on the ground in the post-crash fire after impact.

The Autopsy Before undertaking the autopsy examination, the body must be undressed and searched. The findings must be recorded accurately as they may contribute to the investigation. Tears may match cockpit projections and bloodstains may indicate the direction of forces. Controls may leave imprints on gloves and shoes. This may be helpful in the case of dual controls when there is doubt as to who was flying the aircraft at the moment of impact. If incapacitation of the pilot is suspected such evidence may help to eliminate this possibility. Photography before and during the autopsy is often very helpful. The post-mortem protocol is the standard one and merits no further description. As all aviators have regular medical examinations it is helpful to have obtained the

7  Aviation Deaths

151

information from their medical records before embarking on the autopsy; however, this is frequently impracticable. Such evidence may include a limitation for spectacles on the licence and this may prompt a search of the scene for evidence that they were worn. Details of distinguishing marks and scars helpful in identification may also be noted on the medical records. The following list details the minimal requirements for post-mortem examinations based on investigative, medico-legal and sociological needs: 1. Identification and complete examination of the operating crew on the flight deck or in the cockpit 2. A full external examination of all fatal casualties 3. Identification of the cabin attendants and comparison with the passengers 4. Minimal internal autopsy on all passengers and cabin crew to include: (a) Establishment of the cause of death (b) Discovery of major disease likely to influence life expectancy (c) Assessment of deceleration injury to: • Cardiovascular system, liver and diaphragm • Head, sternum, spine and pelvis 5 . Selection of blood specimens from all casualties for carboxyhaemoglobin studies 6. Collection of lung specimens from all casualties for estimation of the mode of death The pathologist must interpret his findings with caution. The head of the Human Factors Group and the Investigator-in-Charge must ensure that the pathological findings are taken as but part of the investigation as a whole and are fully correlated with evidence adduced within the Group and by other Groups. Experience has shown that this is facilitated and maximum advantage gained if the pathologist attends the periodic briefings by the Investigator-in-Charge. Multiple samples for toxicology need to be taken into appropriate containers. The quality of the analysis depends upon the quality of the specimens. If there are several victims it is very useful to have an assistant to undertake the labelling. Samples from all major organs should be taken for histology. Radiological examination may be very important. The discovery of embedded particles or foreign bodies from an explosion is clearly vital. However, availability and cost may limit the use of such facilities. Trauma accounts for most of the autopsy findings. The sometimes tedious task of describing all the injuries may prove valuable in the assessment of the forces suffered by the victims or the sequence of the injuries.

Disease in Aircrew The discovery of organic disease during the post-mortem examination of pilots presents the accident investigator with a number of problems. He wishes to know if the disease caused the accident, if it contributed to the cause or if its discovery was entirely coincidental. The difficulty is compounded as the normal pathological

152

S.A. Cullen

Table 7.1  Prevalence of coronary artery disease in pilot fatalities Percentage with coronary disease within age range Age range Number in age range No disease Mild Moderate Severe Under 31     411 81 12  4  3 31–40     361 62 21 12  5 41–50    226 44 27 16 13 Over 50    190 36 24 18 22 Total 1,188 61 20 11  8

sequence of events following an acute episode of disease may be cut short by the accident. The familiar morbid anatomical changes will not then be present. In order to answer this problem the pathologist will need to know the prevalence of the disease in the pilot population. Disease may cause accidents by incapacitating the pilot and the circumstances surrounding the crash may give valuable clues as to the likelihood of such an event. Pilots may become subtly impaired by virtue of their disease. The problem facing the accident investigator is the reverse of the one facing the clinician. The clinician elicits the symptoms and signs from the patient and tries to deduce the underlying pathology. The accident investigator has discovered the underlying pathology and tries to determine what symptoms the pilot may have had and how they may have affected his ability to fly his aircraft properly.

Coronary Artery Disease Examination of the coronary arteries of 1,188 pilots killed in aircraft accidents revealed some degree of coronary artery disease in just over 39%. The percentage with coronary artery diseased is shown in Table 7.1. Evidence of previous ischaemic damage was seen in 43 of the 1,000 pilots in whom histological examination of the heart was undertaken. This presents similar problems to the discovery of coronary artery atheroma, namely what role did the heart disease play in the causation of the accident. Coronary artery atheroma causing a reduction in the area of the lumen of greater that 50% was seen in just less than 20% of the pilots examined. The frequency of this finding makes the interpretation of its significance all the more difficult. In common with the investigation of any sudden death, it is the circumstances surrounding the death that determines the interpretation of the findings if no acute change is discovered. Not surprisingly, the ages of the pilots was relevant in those accidents in which coronary artery disease was thought to have played a part. The average age of those dying in their aircraft was 57 years while those whose disease was a probable cause of the accident averaged 53.7 years and those in whom the disease was thought to be a possible cause averaged 51.9 years. Table 7.2 summarises the relationship of the pilots’ coronary artery disease to the causes of the accidents. Coronary artery disease was thought to be the cause of the accident in 1.2% of cases and a possible factor in a further 0.9%.

7  Aviation Deaths

153

Table 7.2  The relationship of coronary artery disease to accident causation Cardiovascular disease Category Number Probable cause Possible cause

Total

Military Commercial General Other Mixed Total

1 (0.2%) 5 (3.2%) 9 (1.9%) 14 (6.4%) 0 29 (2.1%)

514 154 475 216 9 1,368

1 (0.2%) 3 (1.9%) 5 (1.0%) 7 (3.2%) 0 16 (1.2%)

0 2 (1.3) 4 (0.8%) 7 (3.2%) 0 13 (0.9%)

Other medical conditions rarely cause or contribute to the cause of an accident and they contribute to approximately five out of every 1,000 accidents.

Suicide Because of the implications of a suicide verdict it is rare for one to be recorded unless there is overwhelming evidence that this was the intention of the deceased. In practice this usually means that there should be a note or other clear indication that suicide was the purpose of the act leading to death. In England coroners are instructed by the Lord Chief Justice that “suicide must not be presumed merely because it seems on the face of it to be a likely verdict. Suicide must be proved by the evidence.” In many countries it seems likely that more suicides occur than are documented by coroner’s verdicts. In a review of possible suicides using aircraft in the USA, Gibbons and his colleagues [3] suggested that the same was true of aviation accidents. They believed that suicide accounted for less than 2% of fatal general aviation accidents. In an American review of nearly 6,000 general aviation accidents Ungs [4] recorded ten cases of suicide. In another 20 cases, suicide was possible but not proved, giving a rate of 0.17% for definite suicide and 0.51% if possible suicides are included. In Germany, Mäulen and Faust [5] stated that between 2 and 3% of aviation accidents could be attributed to suicide. The problem is compounded in aviation as the circumstances and findings in suicidal and accidental aircraft accidents are often nearly identical. Only careful investigation of the psychological problems encountered by the victim can help to elucidate this problem. This has been termed a “psychological autopsy” [6, 7]. A study of general aviation accidents in the UK revealed that suicide was the probable or possible cause of the crash in over 2.5% of cases [8]. Mastery of his machine and the environment, coupled with the sense of independence that flying gives, is, for many pilots, the source of the pleasure of flying. It has much in common with other dangerous sports such as mountaineering or motor racing. The pilot is at one with his machine. Suicide attempts using aircraft are unusual in that they are rarely unsuccessful and cause considerable press and public interest. The choice of his aircraft as the instrument of suicide reflects the particular relationship that the pilot has with his machine. In common with the Japanese kamikaze deaths aviation suicide has an element of heroism and may be seen as a “path to glory” [9].

154

S.A. Cullen Table 7.3  Common features of aviation suicide Suicide verdicts Total 7 Domestic problems 6 Psychiatric history 5 Left note 3 Aerobatics performed 1 Alcohol consumed 0

Other verdicts 7 5 2 2 3 2

Total 14 11  7  5  4  2

Many  pilots indulge in aerobatics [3] before the final impact, possibly as a final demonstration of mastery over their machine. Many authors [3, 10] have noted the precursors of suicide. Table 7.3 shows the frequency of some common factors in this study. A history of major domestic problems was the most common feature being seen in 80% of these cases. It has been suggested [6] that suicidal motives include a wish for release, notions of guilt, feelings of anger or revenge and a wish to “begin again.” These may all play a part in the genesis of these incidents. The significant number of accidents attributed to suicide serves to emphasise the need to include an examination of the victim’s domestic and psychological history in the traditional detailed examination of the factors leading to an aircraft accident. This examination should not be limited to obvious acts of suicide but should include those acts of “pilot error” such as flying while intoxicated, flying beyond one’s capabilities or taking chances with the weather. Many fatal general aviation accidents occur when pilots without instrument ratings venture into meteorological conditions that require flying by instruments [9]. It has been suggested [6] that such behaviour may fall into the category of subconscious self-destruction. No one would dispute that the Aviation Medical Examiner should ground a pilot who is clinically depressed. However, most of these cases do not fall into this category. Mental well-being is an important part of a pilot’s fitness for flight. Whether Medical Examiners should be required to explore the psychological aspects of private pilots’ lives is a difficult question. However, if he encounters a pilot under significant personal stress he should make detailed inquiry into all aspects of his psychological background. While some may feel that grounding the pilot or referring him for a psychiatric opinion will add more stress, this course of action should not be shirked. “Depression is a treatable condition; suicide is not” [10]. Medical conditions in the pilot are an infrequent cause or contributory factor in fatal accidents. In a study of 1,000 fatal aviation accidents organic disease was deemed to be a probable or possible cause or contributory factor in 3.5% of the cases studied [11]. However, it is a probable cause or contributory factor in only 1.3%. The rate of medical causes for fatal general aviation accidents is comparable to that reported from the USA [12]. It is difficult to see how this low rate may be reduced still further. Perhaps one might consider having an age bar above which pilots must fly as or with a co-pilot. It is unlikely that such a restriction would be placed upon general aviation and glider pilots, as it would be seen to interfere with an individual’s right to practice his sport. Most commercial pilots already fly in double-crewed aircraft and military pilots do not fly solo aircraft when they are older.

7  Aviation Deaths

155

Mechanisms of Injury in Aircraft Accidents The sudden deceleration that occurs when an aircraft hits the ground or water is the commonest cause of injury in aircraft accidents. Because the aircraft structures absorb some energy as they collapse or are crushed the forces acting upon the occupants may be less than those applied to the aircraft. Modern design can aid the collapse of the aircraft so that it is controlled and the forces applied to the occupants are reduced. However, lack of a safety harness may mean that the forces are magnified. The acceleration due to gravity is 9.81 ms−2 and is termed g. It is usual to refer to acceleration in terms of G, which is the acceleration applied to the individual divided by g. Therefore, 10G is 98.1 ms−2. It is to be noted that deceleration is the layman’s term for negative acceleration Fig. 7.1. Human tolerance to deceleration depends upon a number of factors including the duration, magnitude and direction of the inertial forces [13]. In most accidents, the duration of application is short – less than 0.5 s. The direction of forces is a major determinant of tolerance. Man can tolerate Gx deceleration better than Gz, and Gz better than Gy. Personal variables such as gender, age, build and level of fitness also influence the ability of man to tolerate deceleration. Long bones are most susceptible to bending injuries while short bones can withstand stress but are most affected by crushing. Impact injuries cause the greatest damage to flat bones. Horizontal forces frequently occur during aircraft crashes. These may be accompanied by collapse of the aircraft structure with injury to the occupant’s legs leading to incapacity and failure to escape. It is believed that a negative acceleration or deceleration (−Gx, “eyeballs out”) of 45G may be sustained for a short period and

Fig. 7.1  The effects of deceleration. The standard aeromedical terminology for describing the forces. The vector arrows indicate the direction of the resultant inertial forces

156

S.A. Cullen

25G for longer without incapacitating injury [14]. Much higher forces can be tolerated if rearwards facing seats with high backs are provided. Decelerations over 80G (+Gx, “eyeballs in”) have been tolerated with this configuration. This is the logic behind the drive to provide rearward facing seats in passenger-carrying aircraft. Perpendicular forces are also applied if the aircraft falls vertically. The occupants tolerate these forces less well. Minor injuries, including compression of the vertebrae, can occur with + Gz decelerations of 25G. Abrupt vertical deceleration frequently results in break up of the floor structure to which the seats are mounted and this failure of the seat mountings often leads to serious injury. The engines of helicopters are often mounted above the cockpit and collapse of the aircraft structure may cause these to encroach into the cockpit area and cause severe injury to the occupants. Injuries are caused by the interaction of the victim with the aircraft. In many crashes, the aircraft structure collapses and the individual is injured by impact with the airframe. These injuries can include amputations, major lacerations and crushing. When the structure collapses, the victims may become trapped within the wreckage and die of fire, drowning or traumatic asphyxia. Harness restraint systems are provided in aircraft and these may modify the injuries that are sustained. The unrestrained head will swing forward when the torso is effectively restrained and the body is exposed to eyeball-out or −Gx acceleration. This may put a strain on the atlanto-occipital articulation, which is increased if a heavy helmet with, for example, night vision goggles attached is worn. This joint, therefore, needs careful evaluation. Pivoting over a lap strap often produces tears in the lower part of the small bowel mesentery and other bowel injury. The restraints themselves may fail. This may occur in the harness, its mountings, or the seat or floor may fail. When this happens the unrestrained victim can be injured by secondary impact against fixed structures. Items of equipment within the cabin, which are not adequately secured, may break free in a crash and cause injury by secondary impact with the occupants. Overhead lockers are a particular source of loose items such as bottles that may cause significant injury. The heavier these items, the more likely are injuries. Flying debris from overhead lockers was a major cause of head injury in the Boeing 737 disaster at Kegworth in January 1989 [15].

Injury Analysis and Scoring Systems Injury scoring as a means of classifying the extent of trauma has been used for many years. The Abbreviated Injury Scale (AIS) defines the threat to life in anatomical terms and has been accepted at a method of assessing the severity of trauma in road traffic accidents. However, the majority of victims die from more than one fatal injury and injuries which on their own may not be life threatening may be significant when combined with other injuries. An Injury Severity Score (ISS) was devised [16] as a method of assessing victims with multiple injuries. Hill [17] used a modified injury scoring system that has been found to be useful in assessing the injuries in aircraft accidents (Table 7.4). The injuries sustained by the various anatomical regions are graded and the total for each victim is calculated.

7  Aviation Deaths Table 7.4  Modified injury scoring system Severity of trauma Score None 0 Mild 1 Moderate 2 Severe 3 Fatal 4

157

Force needed – Little Moderate Considerable Great

Outcome – No injury Possible injury Probable injury Fatal

In practice, this system has proved sufficient for the purpose of assessing injury patterns in fatal aircraft accidents.

Head Injury Head injury is very common in aviation accidents and is seen in two-thirds of cases. In most cases, the head injury caused or contributed to the cause of the death. A significant finding was that the base alone was fractured in 18.9% of the fatalities that were not disintegrated. The base alone was fractured in 15.7% of military aircraft accident victims, 17.1% of helicopter fatalities and 20.4% of light aircraft accident deaths. There are two mechanisms which cause this occult head injury. The first involves transmission of the impact forces through the mandible and temporo-mandibular joint to the base of the skull. This results in a transverse fracture that runs forward from the joint anterior and parallel to the petrous temporal bone. The two portions of this fracture join in or just posterior to the pituitary fossa. This is sometimes known as the “hinge fracture” (Fig. 7.2). There may be seemingly trivial external injury in these cases. These fractures may result in secondary shearing fractures of the vault. The second common fracture of the base is a result of the forces being transmitted through the vertebral column and is found particularly with + Gz deceleration. These severe vertical forces are seen in falls from a height and when aircraft descend vertically in situations such as a stall. The result of these forces is a “ring” fracture of the posterior fossa. This occurs around the foramen magnum and may be a complete ring or, more commonly, an incomplete one (Fig. 7.3). This fracture may communicate with the hinge fracture when the severe vertical forces also have a horizontal, −Gx, component. In severe cases, the forces may cause secondary “blowout” fractures of the vault of the skull.

Thoracic Injury Injuries to the bones of the thorax are the most common injuries seen and occur in 80% of all accident victims. These injuries in turn cause trauma to the cardiovascular system. 47.6% of all accident victims had a ruptured heart and in 35% there was also a ruptured aorta. Only 10.5% had ruptured their aorta without rupturing their heart.

158

S.A. Cullen

Fig. 7.2  (a) Shows the direction of the impact force; (b) shows the typical “hinge” fracture of the base of skull

Fig. 7.3  (a) Shows the direction of the impact force; (b) shows the “ring” fracture of the base of skull

Injury to the heart and aorta may arise in several ways. The most obvious is by direct penetration by the broken ends of ribs. However, the most frequent mechanism arises from compression of the heart between the sternum and spine. In forward-facing deceleration (−Gx) the chest is often compressed against fixed structures in the aircraft. Flexion injuries can also compress the chest as the chin

7  Aviation Deaths

159

falls forward and strikes the sternum – the so-called chin-sternum-heart syndrome, which was originally described in parachuting accidents [18]. Direct compression results usually in rupture of the atria and occasionally the ventricles. When the ventricles are lacerated this classically occurs in the right ventricle parallel to, and close to the left anterior descending coronary artery. When the rupture results from a sudden rise in intra-cardiac pressure, this may only cause endocardial laceration, typically, on the posterior wall of the atria. Ruptured aorta is caused by the downward displacement of the heart by compression of the base of the heart between the sternum and the spine. It also arises when the deceleration is in the vertical (+Gz) direction and the heart continues to move down while the aorta is anchored. Ruptures usually occur just above the aortic valve ring or at the end of the thoracic arch just distal to the attachment of the ligamentum arteriosum.

Abdominal Injury Damage to the gastro-intestinal tract is, with one exception, uncommon. The stomach seems resistant to rupture except when it is herniated through a ruptured diaphragm. The intestines are similarly rarely lacerated. The one exception is that they are often bruised. The distribution of the bruising suggests that this is caused by compression of the gut between a lap belt and the spine. This mechanism may also be responsible for the fenestration of the mesentery that often accompanies the bruising of the gut serosa. The association of these injuries with the use of harness restraint is helpful in accident reconstruction as it demonstrates the use of seat belts. Cabin crew spend little time seated and for most of the flight they are standing in the cabin going about their business. When such seat belt injuries are seen in cabin crew it indicates that they were seated. If the accident occurred at a time during the flight when one would not normally expect them to be seated one may infer that the emergency was anticipated or that there was another reason, such as turbulence, for them to be seated.

Patterns of Injury and Their Role in Accident Reconstruction The pattern of injuries sustained by the victims of aircraft accidents may give valuable clues that may aid the reconstruction of the sequence and circumstances of the accident. The “typical” passenger-carrying aircraft crash is likely to result in either a uniformity of injuries or a steady logical gradation of injuries. Study of the injury patterns may allow the investigators to compare different accidents. This is particularly important when the circumstances of an accident are unknown such as when an aircraft crashes into the sea when there is no wreckage trail from which the impact attitude may be deduced and when little or no aircraft wreckage may be available for engineering investigation.

160

S.A. Cullen

The Comet disasters of 1954 were the stimulus that prompted the formation of the RAF Department of Aviation Pathology. It was the study of the pattern of injury in the fatalities that pointed to the cause of these accidents [19]. Similar studies of the patterns of injury in subsequent accidents have often indicated the attitude of the aircraft at impact or the nature of the impact itself [20].

Who Was at the Controls at the Time of the Crash? Many aircraft fly with two pilots. In determining the cause of an accident it is important to know which pilot was in control of the aircraft at the time of the crash. The provision of cockpit voice recorders in commercial aircraft may help this task. When incapacitation is suspected in a single pilot aircraft it is crucial to know if he was controlling the aircraft at the time of the crash. Only one pilot is in charge of the controls at any one time while the other pilot concerns himself with observation of instruments and the airspace close to the aircraft. The second pilot may also be involved in cross-checking navigation. Either the pupil or instructor may be in control of training aircraft. Modern long-range aircraft are fitted with an automatic pilot. When this is engaged, the pilot may even leave his seat in the cockpit to perform other tasks. However, it is usual for one pilot to be in his seat at all times. The design of the controls, the pilot’s position and the manner of operation of the control system must all be known before one can determine which pilot was in control at the time of the accident. The design of control assemblies, rudder pedals and other control levers varies from aircraft to aircraft. Fixed wing aircraft frequently have a horn assembly as a control; this may be U-shaped or similar to a car steering wheel with spokes. Military aircraft frequently have a control stick contoured to fit the pilot’s right hand. This stick may incorporate switches for operation by the right thumb or fingers. Rudder pedals also differ in their construction. The abrupt deceleration when an aircraft crashes propels the pilot’s body in the direction of flight. Damage may occur in the hands and feet if they are on the controls at the moment of impact [21]. The injuries that are sustained may mirror the shape of the controls involved and depend on the direction and magnitude of the forces that are applied. The area between the thumb and index finger is particularly likely to be injured if the control column is being grasped at the moment of impact (Fig. 7.4). Patterns and abrasions may be seen which mirror the grips or switches on the control column. These injuries are seen on the palmar surface of the hand. In severe accidents the thumb may be severed. The injury caused by flailing of a hand that is manipulating the throttle is, in contrast, seen on the dorsal aspect between the wrist and the knuckles. The force directed between the thumb and index finger during control column injury may be transmitted to the wrist and forearm. This may cause fracture or ­dislocation of the wrist. The stress applied to the forearm may cause fractures of the arm. These are frequently found in the lower third and are usually in flexion; the distal fracture ends commonly penetrate the extensor surface of the arm. If the forces

7  Aviation Deaths

161

Fig. 7.4  Typical hand injury showing the palmar laceration caused by the control column

are applied to the elbow, posterior fracture dislocation may be seen. The control column frequently breaks and, in these circumstances, lacerations will occur on the palmar surfaces of the hands; fragments of the control column may occasionally be found in these injuries. When the pilot’s feet are resting on the rudder pedals at the moment of impact they are subjected to excessive force on the soles corresponding to the area of the pedals. The construction of the rudder pedals will determine the nature of the foot injuries; bar-shaped bruises and transverse fractures of the tarsal bones being often seen. Because of the angle of the feet on the pedals, the heel is subject to strain and comminuted fractures of the tarsus may occur. The injuries sustained in the feet that are caused by the rudder pedals are found on the plantar surface; those due to flailing are seen on the extensor surface of the feet and lower legs. Injuries due to controls will only be sustained when there is sufficient force. If such a force is present, the absence of these injuries may indicate that the pilot did not have his hands and feet on the controls at the moment of impact. It is important to note that persons other than the pilot may sustain similar injuries if their hands and feet adopt similar position to those of the pilot; feet resting on the bar of the seat in front will sustain injuries indistinguishable from those caused by pedals. Hence, it is important that these injuries are interpreted in the light of all the evidence that is available. The forces applied to the pilot may also cause injuries to the head and trunk. The head may strike parts of the instrument panel leaving imprints on the forehead or face. Patterns derived from the configuration of knobs and switches on the panel may be seen. Occasionally instruments may be embedded in the skull or face. Fragments of glass from the face of dials may be found in the wounds that arise from contact with the control panel. Blood or hair that is found on the control panel might mirror the wounds on the pilot’s head.

162

S.A. Cullen

Witness marks from lap strap and diagonal harness may indicate which seat the individual was occupying. If a shoulder harness in not worn, or if it fails, the upper part of the body will flex forward and frequently strikes the control column, causing characteristic wounds to the anterior chest. Injuries that are due to the manipulation of controls and pedals are found on the palmar surfaces of the hands and the plantar surfaces of the feet. Contact injuries caused by the limbs flailing and striking specific instruments or levers within the cockpit are almost always found on the extensor surfaces of the hand or lower limbs.

Pulmonary Fat and Bone Marrow Embolism Whether a victim was alive or dead at impact is frequently asked. This may arise in cases where pilot incapacitation is suspected. Fire is a significant problem and is seen in approximately one-third of aircraft accidents. When fatal burns occur or when the individual has died of drowning or other forms of asphyxia, the question arises whether the victim survived the impact. While the presence of carbon ­particles in the respiratory tract and carbon monoxide in the blood may help with burning other indicators are needed when there has been no burning. The gross appearance of injuries may give some indication of their age but in my experience the microscopic study of such injuries is rarely of value. The examination of the lungs for the presence of fat and bone marrow embolism may help to elucidate this problem. Their presence may indicate the need to explain injury before death from the crash. Does the injury suggest that the aircraft may have been subjected to turbulence before crashing? They may even indicate explosions on the aircraft with injury to those nearest to the seat of the explosion. Pulmonary fat and bone marrow embolism requires a fracture to a bone, a functioning cardiovascular system and a finite period of time after the injury. Mason [22] found that the severity of pulmonary fat embolism increased with the length of the agonal period. He also noted that bone marrow emboli tended to disappear within 5–24 h. It was noted that early bone marrow emboli retained their particulate structure for a while and then this was lost and the embolus was represented by a mass of marrow cells retained within the vessel until it was dissipated entirely (Figs. 7.5 and 7.6). Care is needed when searching for fat and bone marrow emboli as artefacts such as bronchial epithelium may mimic them.

Toxicology Toxicology is dealt with in detail elsewhere in this volume but aviation deaths present special problems that merit discussion here. The severe disruption and contamination that frequently are seen in aviation deaths means that robust methods of

7  Aviation Deaths

163

Fig. 7.5  The natural history of pulmonary fat and bone marrow embolism

analysis are needed. Headspace gas chromatography is used for ethanol analysis and screening for volatile compounds. Carbon monoxide may be measured by ­spectrophotometry using a CO-Oximeter (Instrumentation Laboratories Limited) but, with contaminated samples or decomposed blood, gas chromatography or derivative spectrophotometry is required [23]. Cyanide is separated from acidified blood by Conway diffusion and analysis undertaken by colourimetry or gas chromatography. Drug-screening methods should be capable of detecting therapeutic levels of prescription and over-the-counter drugs. As the victims of air accidents may not be recovered for some time after death the problem of the post-mortem production of alcohol is frequently encountered. This is best overcome by taking adequate samples in an appropriate preservative. Blood should be taken from at least two peripheral veins; urine and vitreous should also be taken. When urine is not available bile is a suitable alternative. The levels of alcohol found in the samples analysed should be concordant. If they are not postmortem production should be suspected. The performance of pilots has been shown to be impaired with blood alcohol levels as low as 11 mg/dl [24]. It has also been shown that the number and severity of the errors rises in proportion to the level of alcohol in the blood [25, 26]. In Europe it is illegal to fly an aircraft with a blood alcohol greater than 20 mg/dl. Alcohol may be a factor in 3–4% of general aviation accidents in the UK.

164

S.A. Cullen

Fig. 7.6  (a, b) Pulmonary fat and bone marrow embolism

Carbon Monoxide and Cyanide Carbon monoxide and cyanide are found in fire atmospheres and carbon monoxide is produced during incomplete combustion. Carbon monoxide binds to haemoglobin to form carboxyhaemoglobin. Carbon monoxide has an affinity for haemoglobin about 200 times that of oxygen and so a comparatively low level of carbon monoxide in the inhaled air will produce a comparatively high level of carboxyhaemoglobin. This may give rise to tissue hypoxia. The concentration of carbon monoxide in the blood is expressed as the percentage­ of haemoglobin that is saturated with carbon monoxide. A small amount, usually

7  Aviation Deaths

165

less than 1%, is produced during haem catabolism. This may rise to 5% in ­haemolytic anaemia. Increases in carboxyhaemoglobin are almost always the result of inhalation of smoke from fires, including cigarette smoke; vehicle exhaust fumes or poorly maintained heating appliances. It has a half-life of four to 5  h and is eliminated through the lungs. The half-life can be reduced to about 1 h by breathing pure oxygen and still further by breathing hyperbaric oxygen. The discovery of increased carbon monoxide or cyanide in the blood is indicative of either survival in a fire or contamination of the aircraft atmosphere with carbon monoxide possibly from defective heaters or exhaust systems.

Therapeutic and “Over-the-Counter” Drugs Therapeutic and “over-the-counter” drugs are being detected with increasing frequency. In the USA, Chaturvedi and his colleagues [27] report that over a five-year period they were detected in 35.2% of the 1,629 pilots tested. Therapeutic drugs were found in 19.3% and over-the-counter drugs in 15.9%. Drugs were detected about twice as often in private pilots when compared to commercial pilots. The commonest drugs found were analgesics, antihistamines, and benzodiazepines. The detection of any drug raises two important questions. Did the drug cause the accident, contribute to the cause or is its discovery entirely coincidental? The corollary is did the disease for which the drug is being taken cause the accident, contribute to the cause or is it entirely coincidental? The answers to these questions can be very difficult. A particular problem is faced by the discovery of first generation antihistamines. They may cause drowsiness and impair the pilot’s ability to fly his aircraft properly. They are frequently taken for colds or upper respiratory tract infections that may, because of their effect on the ear, give rise to an increased tendency to disorientation.

Crop-Spraying Accidents The aerial application of insecticides, weed control chemicals, defoliants and fertilisers is now common. Many chemicals used for these purposes are highly toxic and can cause illness and death. Aircraft accidents have occurred because of impairment caused by exposure to these substances. To understand the problems that may arise from crop spraying, it is essential to understand the toxic potential of the various chemicals used. Organophosphorous insecticides are toxic to both insects and humans by their inhibition of acetylcholinesterase enzymes within the nervous system. They have toxic effects on the central nervous system. Early symptoms of poisoning include giddiness, restlessness, blurred vision and respiratory depression. Death may ­follow. Plasma cholinesterase measurements are useful as indicators of exposure. Carbamate insecticides may cause similar signs and symptoms but the effect does not last as long. Red cell cholinesterase measurements may be used to assess the extent of the

166

S.A. Cullen

exposure. Chlorinated insecticides such as dieldrin or DDT are rarely used ­nowadays because of their toxicity on the central nervous system. Chlorinated herbicides have low toxicity for man. Their use by the USA of the herbicide known as Agent Orange during the Vietnam War has shown the low toxicity of these compounds. Nitrophenols produce thirst, excessive sweating, euphoria and fatigue. They have no antidote. The toxicity of paraquat is well known. Agricultural pilots in the former Soviet Union are required to maintain a logbook which records the specific chemicals used during crop spraying and the duration of the exposure. These records are valuable should the pilot develop symptoms that may be due to exposure to the chemicals used. Rarely is a reduced level of cholinesterase discovered in the pilot. Salvage workers involved in recovering the aircraft wreckage may become ill because of the toxic effects of the chemicals used for spraying. It is important that rescuers and salvage workers are made aware of the potential harmful effects of agricultural chemicals to which they may be exposed when they are dealing with the aftermath of an accident involving a crop-spraying aircraft.

The Accident Analysis More evidence is available to air accident investigators than in many other forms of transport accident. In addition to observations made at the crash scene and autopsy there may be evidence from flight data and cockpit voice recorders. The crew licensing and medical histories will be available and air traffic control may be able to provide radar plots of the final flight path. Meteorologists will provide weather information and the servicing history of the aircraft will be available. The correlation of all this information is the responsibility of the accident investigators but inevitably will be a team effort. The pathologist will be part of this team in the case of fatal accidents. Aircraft accidents rarely have a single cause but it is true that human error is the most frequent common factor. A psychologist may be a vital member of the team. The psychologist’s contribution is discussed in Chap. 8. The aim of accident investigation is to explain the cause of the accident so that similar accidents may be prevented in the future. Attributing blame has no part in the investigation process. Frequently, it is impossible to determine all the causes of an accident but the efforts of all investigators, including pathologists, have been rewarded by the improvement in air safety.

References 1. International Civil Aviation Association (1994) Annex 13 – aircraft accident investigation. Civil Aviation Authority, Cheltenham 2. Air Accidents Investigation Branch (1990) Memorandum on the investigation of civil air accidents. Air Accidents Investigation Branch, Farnborough 3. Gibbons HL, Plechus JL, Mohler SR (1967) Consideration of volitional acts in aircraft accident investigation. Aerospace Med 38:1057–1059

7  Aviation Deaths

167

4. Ungs TJ (1994) Suicide by use of aircraft in the United States, 1079–1989. Aviat Space Environ Med 65:953–956 5. Mäulen B, Faust V (1990) Suizid mit dem Flugzeug. Münch Med Wochenschr 132:572–574 6. Yanowitch RE, Mohler SR, Nichols EA (1972) Psychosocial reconstruction inventory: a postdictal instrument in aircraft accident reconstruction. Aerospace Med 43:551–554 7. Yanowitch RE, Bergin JM, Yanowitch EA (1973) Aircraft as an instrument of self-destruction. Aerospace Med 44:675–678 8. Cullen SA (1998) Aviation suicide: A review of general aviation accidents in the UK, 1970– 96. Aviat Space Environ Med 69:696–698 9. Mäulen B (1993) An aeronautical suicide attempt suicide and self-destructive behaviour in aviation. Crisis 14:68–70 10. Jones DR (1977) Suicide by aircraft: a case report. Aviat Space Environ Med 48:454–459 11. Cullen SA, Drysdale MRW (1997) Role of medical factors in 1000 fatal aviation accidents: case note study. BMJ 314:1592 12. Booze CF (1989) Sudden in-flight incapacitation in general aviation. Aviat Space Environ Med 60:332–335 13. Cugley J, Glaister DH (1999) Short duration acceleration. In: Ernsting J, Nicholson AJ, Rainford DJ (eds) Aviation medicine, 3rd edn. Butterworth-Heinemann, Oxford, pp 157–166 14. Anton DJ (1988) Crash dynamics and restraint systems. In: Ernsting J, King P (eds) Aviation medicine. Butterworths, London, pp 168–169 15. White BD, Firth JL, Rowles JM, NLDB Study Group (1993) The effects of structural failure on injuries sustained in the M1 Boeing 737/400 disaster, January 1989. Aviat Space Environ Med 64:95–102 16. Baker SP, O’Neill B, Haddon W et al (1974) The injury severity score; a method for describing patients with multiple injuries and evaluating trauma care. J Trauma 14:187–196 17. Hill IR (1987) The Air India jumbo jet disaster Kalishna – injury analysis. In: Caddy B (ed) Uses of forensic sciences. Scottish Academic Press, Edinburgh, pp 120–145 18. Simson LR (1971) “Chin-Sternum-Heart Syndrome”: Cardiac injury associated with parachuting mishaps. Aerospace Med 42:1214–1217 19. Armstrong JA, Fryer DI, Stewart WK, Whittingham HE (1955) Interpretation of injuries in the Comet aircraft disasters. Lancet i:1135–1144 20. Cullen SA, Turk EP (1980) The value of postmortem examination of passengers in fatal aviation accidents. Aviat Space Environ Med 51:1071–1073 21. Krefft S (1970) Who was at the aircraft’s controls when the fatal accident occurred? Aerospace Med 41:785–789 22. Mason JK (1968) Pulmonary fat and bone marrow embolism as an indication of ante-mortem violence. Med Sci Law 8:200–206 23. Mayes RW (1993) Measurement of carbon monoxide and cyanide in blood. J Clin Path 46:982–988 24. Ross LE, Teazel LM, Chau AU (1992) Pilot performance with blood alcohol below 0.04%. Aviat Space Environ Med 63:951–956 25. Billings CE, Wick RL, Gerke RJ, Chase RC (1973) Effects of ethyl alcohol on pilot performance. Aerospace Med 44:379–382 26. Billings CE, Demosthenes T, White TR, O’Hara DB (1991) Effects of alcohol on pilot performance in simulated flight. Aviat Space Environ Med 62:233–235 27. Chaturvedi AK, Craft KJ, Canfield DV, Whinnery JE (2005) Toxicological findings from 1587 civil aviation pilot fatalities, 1999–2003. Aviat Space Environ Med 76:1145–1150

wwwwwwwwwwwwwwwww

Chapter 8

Fatalities in General Aviation: From Balloons to Helicopters Alex de Voogt

Abstract  General Aviation (GA) has shown the highest proportion of accidents and fatalities in aviation. The countries with the highest number of GA flights (USA, UK, and Australia) all show high accident rates for this category. The developments in the approaches to aviation safety, in particular in relation to GA, show an increasing understanding of the determining factors of fatalities and the most promising strategies to prevent them. Occurrences in aviation that involve fatalities are universally classified as accidents. Accidents are in contrast with incidents, which only feature minor injury and minor damage to the aircraft. Accident statistics have dominated the study of aviation safety and have benefitted from new approaches from organizational psychology. The relevance of these approaches to GA accident investigations is limited due to the highly diversified operations and the limited organizational structure of this industry. Keywords  Aviation • Helicopters • Air balloons • Accident investigation

Accident Investigation James Reason [1, 2] developed a model of understanding aviation and other organizational accidents by showing layers within an organization that all contribute to the cause of the accident. Reason’s model became known as the Swiss Cheese Model and showed that “holes” in each layer create the conditions for an accident and that those at the end of the chain, for instance the pilots, are rarely solely responsible. Organizational factors, environmental conditions, and team actions all play a part.

A. de Voogt (*) Division of Anthropology, American Museum of Natural History, 200 Central Park West, New York, NY 10024, USA e-mail: [email protected] E.E. Turk (ed.), Forensic Pathology Reviews, Volume 6, Forensic Pathology Reviews 6, DOI 10.1007/978-1-61779-249-6_8, © Springer Science+Business Media, LLC 2011

169

170

A. de Voogt

The latent conditions within the organization turn into active failures when they are triggered by a certain event, such as deteriorating weather in aviation, that in combination may lead to an accident. Reason distinguishes slips and lapses that are unintended actions as well as mistakes and violations that are intended actions. Although this part of Reason’s work is based on concepts developed by Rasmussen [3, 4] and other parts were influenced by the information processing model by Norman [5], his approach resulted in a widespread improvement of understanding organizational accidents. Today’s safety investigation analysis methods for aviation are largely based on his ideas. The Human Factors Analysis and Classification System (HFACS) by Shappell and Wiegmann [6, 7] became a popular tool to investigate accidents, classify the underlying factors, and distinguish unintended and intended actions that may have caused the accident. It was to become particularly successful for aviation [6, 7]. A number of other models followed, for instance Threat and Error Management [8], that were more schematic but specifically designed for aviation. Criticism on such approaches by Dekker [9] and others found that these approaches were too limited and studied the process only after the accident had already happened. The approach of HFACS became dominant but was used to investigate accidents rather than practices that could possibly lead to an accident. Accidents lead to serious injury, fatal injury, or a substantially damaged or destroyed aircraft; reporting aircraft accidents is mandatory and aviation organizations, such as the National Transportation Safety Board in the USA, collect, analyze, and publish statistics on accidents. Incidents cause only minor damage or injury and, in case of the USA, are mostly collected by the Federal Aviation Administration. All such reports are historical and record the problems only after they have resulted in damage and injury. Efforts are made to collect reports on problems that only potentially lead to an incident or accident. The aviation organizations have endeavored to collect voluntary reports on such problems. In addition the International Civil Aviation Organization (ICAO) supports a program known as LOSA [8], a line operational safety audit, in which, for instance, a person joins the crew in the cockpit and monitors their actions. This approach records practices rather than accidents and is quick to identify areas in which procedures, training, or policies can be improved. Part of the criticism on accident investigation analysis was the focus on relatively rare events rather than current practices [overview in 10]. Although, this is often true for commercial aviation, general aviation shows a different picture. Military and Civil aviation have traditionally been studied separately although some of the flight characteristics resemble one another, particularly that of helicopters. Within Civil or Civilian aviation there is a distinction between commercial and non-commercial flights. The first concerns the major airlines and Reason’s work has transformed the investigation practices of airline accidents. Airlines and commercial flights, such as air taxis, transport goods or people to generate revenue. General aviation, on the other hand, commonly uses smaller aircraft and features a wide range of operations. They include fire-fighting, logging, crop-dusting, emergency medical services (EMS), and other operations often specific to helicopters as well as a range of flights with balloons, gliders, ultralights, gyroplanes as well as small

8  Fatalities in General Aviation: From Balloons to Helicopters

171

planes and helicopters that are used for personal or business use, for flight training or just for recreation. In 2007 [11], there were 228,000 active private pilots and 220,000 registered General Aviation aircraft in the USA. In the years prior, 91% of all aviation crashes and 94% of all aviation fatalities were in GA. The crash rate was reported as 1.31 fatal crashes per 100,000 flight hours, which is 82 times the rate for major airlines. This striking difference between general aviation and major airline operations has been in place for many years and the accident investigation models were not designed to address this difference. HFACS has been applied to general aviation and has classified the types of mistakes leading to accidents. The crucial element of the investigation models is the layers beyond the pilot’s mistake and in GA these layers are often not accessible or not investigated. But since accidents are not rare in general aviation, there is still much to gain from accident analysis without the danger of focusing on rare and unusual events only. Apart from relatively high numbers of accidents, the general aviation sector is characterized by highly diversified operations and aircraft. Next to overviews of general aviation accidents, each type of aircraft and each specific operation has required and has been given attention. The characteristics of these smaller groups of accidents create a better insight and more specific recommendations to improve the safety record of this industry.

Epidemiological Studies The main source of the literature for epidemiological studies in aviation is the journal of the Aerospace Medical Association: Aviation, Space, and Environmental Medicine, which is a continuation of its two predecessors Aerospace Medicine and the even earlier Journal of Aviation Medicine. Together they published almost 75 years of aviation-related epidemiological studies including countries other than the USA although rarely outside the UK, Australia, and New Zealand [12]. An important part of the early epidemiological studies concern the age and occupation of the pilots, topics that are less commonly studied today [13]. They did result in the 60-year rule that is still used as an upper limit for airline pilots but is highly debated [Overview in 14]. A review of this journal from 1994 [15] states that “in some cases, the deficiencies in study design and data analysis have resulted in controversial findings.” Li mentions the age-60 rule as a particular example and adds: “Many of the contributions were made by uncontrolled studies based on crash analysis in the form of case reports and case series studies.” Case–control and cohort studies are still rare, mostly because the case reports are easily available and still need much study while control studies require more effort. One case–control study of risk factors for fatal and nonfatal injury in New Zealand’s civil aviation [16] showed that environmental and operational factors were key determinants of the injury outcome rather than pilot or aircraft characteristics.

172

A. de Voogt

This confirms the shift in the literature from pilot characteristics to environmental and operational factors. The type of aircraft is still mentioned since O’Hare et al. [16] conclude that flying a twin-engine aircraft was a risk factor for fatal injury, while piloting a microlight aircraft was a risk factor for all serious injury. The most significant risk factors for fatal and serious injury were aerobatic flight, post-crash fire, not having a certificate of airworthiness and off-airport location. Publications of the last 10 years add much insight to the particularities of certain operations and aircraft. They identify the relevant environmental factors that are reported by the accident investigators. As such, the shift of research attention that was initiated by Reason has resulted in the inclusion of the environment and the operational factors but rarely the organizational factors when it comes to GA. GA is not part of one large organization but is governed by national laws, international aviation regulations, many different company policies as well as individual preferences, which in the case of, for instance, flight instruction [17] creates a complex picture for which improvements are possible but for which policies are not always related to the high number of accidents.

Types of Accidents Apart from aircraft types and, sometimes closely linked, aircraft operations, a number of studies have focused on a type of accident. For instance, studies on the geographical location of the aviation accident, the type of accident according to the gender of the pilot, accidents relating to disorientation including wrong airport landings and also mid-air collisions. Mid-air collisions are not limited to a type of aircraft nor to a particular operation but are considered the most dangerous type of accident. Apart from statistics, studies on mid-air collisions have focused on human factors, in particular the inability of pilots to detect converging aircraft in time to prevent a collision [18, 19]. Morris [19] analyzed mid-air collisions in GA to illustrate the limitations of the see-andavoid principle and a subsequent study [20] showed that radio communication also has little effect to prevent such mishaps. Extensive literature exists on spatial disorientation in both clinical and aviation studies [21]. Spatial disorientation is associated with aerobatic maneuvers, flying into clouds and vertigo in which pilots are no longer able to understand the position of the aircraft in relation to the ground. In aviation, way-finding problems have been termed geographical disorientation [22] and are often seen as part of a situation awareness problem. In such cases, the position of the aircraft in relation to a (mental) map is no longer understood. Landings at the wrong airport are a curious result of geographical disorientation [22, 23]. The development of improved displays and the introduction of GPS navigation systems in GA can be seen as the main solutions for the prevention of such accidents in the future. Aviation in terms of geography is partly made insightful by studies from different countries. In the USA Grabowski et al. [24] also identified localized problems

8  Fatalities in General Aviation: From Balloons to Helicopters

173

that point at climate, regulatory and flight activity differences within one country. In the USA, the state of Alaska has for that reason received increasing attention from researchers and aviation authorities because of its relatively high accident rate. Studies by Baker et al. [25] on gender show pilot characteristics that still play a role in GA accidents since female pilots make different types of errors compared to male pilots. Although pilot characteristics are no longer central in aviation accident studies, they are not to be ignored altogether.

Aircraft-Specific Studies Specific operations have received attention well before Reason and his followers. In GA it has been recognized that crop-spraying, fire-fighting, and logging operations by helicopters require specific attention and have little to gain from helicopter air taxi accident statistics even though air taxi accidents are more numerous. Although helicopters seem to have the most diverse types of operation, small airplanes are involved in crop-spraying and fire-fighting as well, allowing comparisons of the two machines. This is less common for the other types of aircraft that make up only a minor part of GA flights but which require separate recommendations (Table 8.1). Of these other aircraft, hot-air balloons have received most of the attention. Historically, the first aviation accident occurred with a hot-air balloon and the medical profession has benefitted from the studies on altitude sickness conducted almost exclusively in these aircraft up until the space age. Next to hot-air balloons there are less frequent gas balloon accidents that are sometimes included in these studies as well as zeppelins that are usually ignored. The number of zeppelin-related accidents at any given time has been too small even in countries such as Australia, the UK, and the USA to allow any statistical analysis. Fatal accidents in hot-air balloon crashes are few compared to those for small airplanes or helicopters but still warrant investigation [26, 27]. A study from the UK [28] highlighted victims with severe burns as one of the dangers in ballooning. De Voogt and van Doorn [29] showed that the severity of the injuries and the damage to the balloon were not significantly related. Crashes occur in which the balloon is destroyed but the passengers remain unharmed. The envelope of the balloon may

Table 8.1  Accident severity and aircraft specificity for 2,141 accidents occurring between 1982 and 2007 in the USA, extracted from the ntsb online database Glider Gyroplane Ultralight Balloon Blimp Fatal 152 112 113   37  2 Nonfatal 839 150 138 463 13 Destroyed 197 100 120   49  1 Nondestroyed 847 177 132 481 14 Homebuilt 91 241 184   6  0 Total 1,044 277 252 530 15

174

A. de Voogt

catch fire when it strikes an electrical wire in which case the passengers can often leave the basket unharmed. In contrast, hard landings cause most of the injuries in ballooning but the basket and the envelope commonly remain in one piece. When the basket drags across the ground injuries may become more severe or even fatal. Fatal glider accidents share characteristics with homebuilt airplanes. Van Doorn and de Voogt [30] found that fatal accidents were predicted by pilot error, flight phase, and homebuilt aircraft. Gliders and ultralights [31] are often homebuilt so that maintenance and construction errors are more common. Homebuilt aircraft make up 3% of all hours flown in GA but 10% of all crashes and also have a higher fatality rate [32]. Research on gyroplanes has shown that pilot experience is a significant predictor of accident fatality [33]. Pilots with less than 40 flight hours were four times as likely to be involved in a fatal accident as their more experienced colleagues. This result is similar to that found for ultralights in an earlier study [31]. Gyroplanes, also known as autogiros, were regarded as a relatively safe and stable type of GA aircraft. They were developed in the 1920s as a safe alternative to airplanes but were replaced by helicopters in the 1940s. Helicopters do not have a reputation as safe aircraft, partly because helicopter operations can be considered more dangerous.

Helicopters A survey of US helicopter accidents between 1994 and 2005 showed that accident rates were between 11.6 per 100,000 h in 1994 and 6.3 in 2005 [30]. A rare study of helicopter accidents in Russia [34] states that “the Western helicopter relative statistical data conform to the Russian helicopters ones” in particular the accident rate that is higher than for airplanes. For the period 1982 up to 2006, 15.2% of all US helicopter accidents were fatal [30]. Adverse weather and night conditions showed a significantly higher proportion of fatal accidents. At night the accidents were 3.0 times and in instrument-flying conditions 6.3 times more likely to be fatal. These results are similar to those found in specific helicopter operations. For instance, fatalities after helicopter EMS crashes are associated especially with post-crash fire and with crashes that occur in darkness or bad weather, according to Baker et al. [35]. Helicopter EMS operations from 1997 to 2001 showed a lower crash rate than General Aviation but fatal crashes showed a rate higher than all other categories of aviation: 1.7 per 100,000 h compared to 1.3 for general aviation. Taneja and Wiegmann [36] reviewed 84 autopsies of helicopter pilots from 1993 to 1999 in the USA. Blunt trauma was cited as the primary cause of death in 88.1% of these cases with ribs (73.8%), skull (51.2%), and facial bones (47.6%) as the most common bony injuries. There was no relation of the injury pattern with the age of the pilot or the phase of flight. They argue for crashworthy aircraft in order to reduce the number of fatalities. Hayden et al. [37] recommend crash-resistant fuel systems for Civil helicopter operations, which have shown close to 100% effectiveness in Army helicopters and which already prove to reduce post-crash fires in helicopter crashes. A study that

8  Fatalities in General Aviation: From Balloons to Helicopters

175

includes both GA helicopters and airplanes in the USA [38] also related fire or explosion to pilot death and found that pilots who failed to use both lap belt and shoulder harness were more likely to die as well as those who used only the lap belt compared to both restraints. Helicopters are uniquely equipped for specialized operations such as forestry and fire-fighting, often categorized as external load operations, as well as operations at low altitude in confined areas, such as crop-spraying also known as aerial application. The environment in which these operations take place allow little room for error and the machine’s performance is limited by the specialized loads. In low-altitude operations, helicopters have an increased risk of colliding with trees, transmission wires, power lines, or other high structures. One of few published studies on external-load helicopter operations, also known as sling-load operations, was presented by Manwaring et al. [39] who described the causes and circumstances of accidents between 1980 and 1995. They linked their outcomes to studies of Alaskan wood-logging helicopter operations in which progress on safety had been made. Research on aerial application flights has concentrated on the danger of the applied chemicals rather than the danger of the flight operation but a combined study of sling-load and aerial application accidents [40] shows that safety can be improved with additional crew and better fuel management. Additional crew may assuage the number of external load and aerial application accidents since the number of tree and wire strikes make up 18 and 24% of all such accidents, respectively.

Pathology The differences between aircraft is also reflected in pathology studies. Wolf and Harding [41] in their study of sport aircraft-related deaths included nine deaths in ultralight and experimental aircraft. The patterns of injuries included trauma predominantly to the chest, abdomen, or head in addition to blunt force injuries in chest and abdomen or head and torso. Only two extremity factures were found while injuries to the symphysis pubis appeared in six cases. They conclude that “these cases illustrate the varied pathologies associated with deaths due to crashes of sports aircraft and reveal the lack of uniformity associated with the investigations of such deaths.” A rare electrocution is reported for ballooning where collisions with powerlines are a frequent occurrence in accidents [42]. Deaths in ballooning are usually a result of blunt trauma from falls. Although burn victims are reported, burn injuries in balloon and small airplane crashes are usually survivable if the patient arrives at a burn center alive [43]. In gliders and light airplanes decelerative injury is a common pathological feature. A study by Byard and Tsokos [44] of an extreme case highlights the avulsion of the distal tibial shaft. The death of the two occupants of the light airplane was caused by severe injuries involving craniocerebral, skeletal, soft tissue, and organ trauma. The legs were shortened and fragments of distal tibial shaft had been forced through the soles of the occupants’ shoes, which indicated a fall from considerable height

176

A. de Voogt

and a direction along the axis of the legs. They conclude that the observation of extrusion of bone fragments downwards through the shoes and the fracturing of the lower limb bones can add to the understanding of the crash, particularly the position of the occupants prior to impact. Wiegmann and Taneja [45] report on the basis of 559 autopsy reports in the USA that in GA blunt trauma is the primary cause of death (86%) with fractured ribs, skull, facial bones, tibia, and pelvis as the most common. They also mention a fractured larynx in 14.7% of the cases; a common injury not reported before. Common organ injuries include laceration of the liver (48.1%), lung (37.6%), heart (35.6%), and lung (32.9%). They add that individuals who sustained brain hemorrhage were also more likely to have fractures of the facial bones more so than skull fractures. A study of fatal light aircraft accidents in Canada [46] showed that out of 68 victims that were studies about half of them suffered multiple trauma, 29% drowned, 16% died of head or neck injuries and 3% due to a coronary disease. Neck trauma occurred mostly with pilots and also occurred most often with drowned occupants. Passengers suffered relatively more craniofacial fractures and abdominal or retroperitoneal trauma. Ethanol intoxication was implicated in two cases but other drugs did not appear to be a definite factor in accident causation. Although drugs and alcohol are sometimes suspected, particularly with general aviation, Kuhlman et al. [47] already found that there is no consistent pattern of drug involvement in civil aviation fatalities. Not counting nicotine and ethanol they found 12.6% of 377 US cases positive for one or more drugs, with acetaminophen and salicylate as the most frequent ones. Ethanol at greater than 10  mg/dL was found in 14.8% of the cases but only 4.5% were concluded to be due to ingestion of ethanol.

Accidents and Fatality The least and the most dangerous operations in General Aviation, arguably flight instruction and aerobatics maneuvers, show the importance of detailing the study of GA accidents. Flight instruction is not limited to any type of aircraft and received special attention in the literature. Studies reporting flight accidents by student pilots [e.g., 48, 49] indicate that this is a substantial subcategory of accidents. Olson and Austin [50] found that the highest rate of errors in the landing phase was found during the flare and follow through phases of the landing. Benbassat and Abramson [51] state that pilots consider the flare to be more difficult than a number of other flight maneuvers. A follow-up study by Uitdewilligen and de Voogt [52] showed that solo flights and first-time solo flights in the USA show an even higher proportion of improper flares. As with most maneuvers taking place close to the ground, the number of fatalities is small and in the case of first-time solo flights zero. In contrast to flight instruction, aerobatic flight maneuvers are the most significant risk factor for fatal and serious injury, at least in the USA [16]. Nearly 50% of airport transport pilot-induced accidents in GA occur during aerobatics [53].

8  Fatalities in General Aviation: From Balloons to Helicopters

177

Instructional flights are the first flights of every airline pilot and aerobatics is mostly limited to the most experienced. Both show that it is difficult to separate General Aviation from Commercial Aviation since at least the pilots travel across these boundaries. The strategies to prevent fatalities in General Aviation differ widely and although shoulder restraints and crash-resistant fuel tanks are general recommendations, the operations and aircraft are too diverse to address them all at once. The relatively high accident and fatality rate for General Aviation flights emphasize the importance of both new approaches to safety and continuing epidemiological research in aviation.

References 1. Reason J (1990) Human error. Cambridge University Press, Cambridge 2. Reason J (1997) Managing the risks of organizational accidents. Ashgate, Aldershot, UK 3. Rasmussen J (1982) Human errors: a taxonomy for describing human malfunction in industrial installations. J Occ Acc 4(2–4):311–333 4. Rasmussen J (1988) Human error mechanisms in complex work environments. Reliability Eng System Safety 22:155–169 5. Norman DA (1981) Categorization of action slips. Psychol Rev 88:1–15 6. Shappell SA, Wiegmann DA (2001) Applying reason: the human factors analysis and classification system (HFACS). Hum Fact Aerospace Safety 1:59–86 7. Shappell SA, Wiegmann DA (1997) A human error approach to accident investigation: the taxonomy of unsafe operations. Int J Aviat Psych 7(4):269–291 8. Helmreich RL, Klinect JR, Wilhelm JA (1999) Models of threat, error and crm inflight operations. In: Jensen RS (ed) Proceedings of the Tenth International Symposium on Aviation Psychology. Ohio State University, Columbus, OH, pp 677–682 9. Dekker SWA (2003) Illusions of explanation: a critical essay on error classification. Int J Aviat Psych 13(2):95–106 10. Hayward B, Lowe A (2004) Safety investigation: systemic occurrence analysis methods. In: Goeters KM (ed) Aviation psychology: practice and research. Ashgate, Aldershot, UK, pp 363–380 11. Li G, Baker SP (2007) Crash risk in general aviation. JAMA 297:1596–1598 12. Rudge FW (1995) A review of articles published in aviation, space, and environmental medicine, 1975–94. Aviat Space Environ Med 66:1005–1009 13. Booze CF (1977) Epidemiological investigation of occupation, age, and exposure in general aviation accidents. Aviat Space Environ Med 48(1):1081–1091 14. Aerospace Medical Association, Aviation Safety Committee, Civil Aviation Safety Subcommittee (2004) The age 60 rule. Aviat Space Environ Med 75:708–715 1 5. Li G (1994) Pilot-related factors in aircraft crashes: a review of epidemiological studies. Aviat Space Environ Med 65:944–952 16. O’Hare D, Chalmers D, Schuffham P (2003) Case-control study of risk factors for fatal and non-fatal injury in crashes of civil aircraft. Aviat Space Environ Med 74(10):1061–1066 17. de Voogt AJ, van Doorn RRA (2007) The paradox of helicopter emergency training. Int J Aviat Psych 17(3):265–274 18. Wiener EL (1980) Midair collisions: the accidents, the systems, and the realpolitik. Human Factors 22(5):521–533 19. Morris CC (2005) Midair collisions: limitations of the see-and-avoid concept in civil aviation. Aviat Space Environ Med 76:357–65

178

A. de Voogt

20. de Voogt AJ, van Doorn RRA (2006) Midair collisions in U.S. civil aviation, 2000–2004: the roles of radio communication and altitude. Aviat Space Environ Med 77(12):1252–1255 21. Bellenkes A, Bason R, Yacavone DW (1992) Spatial disorientation in naval aviation mishaps: a review of class A incidents from 1980 through 1989. Aviat Space Environ Med 63:128–131 22. Antuñano MJ, Mohler SR, Gosbee JW (1989) Geographical disorientation: approaching and landing at the wrong airport. Aviat Space Environ Med 60:996–1004 23. de Voogt AJ, van Doorn RRA (2007) Approaches and landings at wrong airports: an analysis of 65 cases, 1981–2004. Aviat Space Environ Med 78(2):117–120 24. Grabowski JG, Curriero FC, Baker SP, Li G (2002) Exploratory spatial analysis of pilot fatality rates in general aviation crashes using geographic information systems. Am J Epidemiol 156(5):398–405 25. Baker SP, Lamb MPH, Grabowski JG et al (2001) Characteristics of general aviation crashes involving mature male and female pilots. Aviat Space Environ Med 72(5):447–452 26. Frankenfield DL, Baker SP (1994) Epidemiology of hot-air balloon crashes in the US, 1984– 1988. Aviat Space Environ Med 65:3–6 27. Cowl CT, Jones MP, Lynch CF et al (1998) Factors associated with fatalities and injuries from hot-air balloon crashes. J Am Med Assoc 279:1011–1014 28. Hasham S, Majumder S, Southern SJ et al (2004) Hot-air ballooning injuries in the United Kingdom (January 1976–January 2004). Burns 30:856–60 29. de Voogt AJ, van Doorn RRA (2006) Balloon crashes: accidents in the US 2000–2004. Aviat Space Environ Med 77(5):556–558 30. de Voogt AJ, van Doorn RRA (2007b) Helicopter accidents: data-mining the ntsb database. Proceedings of the 33rd Rotorcraft Forum, Kazan, Russia 31. Pagán BJ, de Voogt AJ, van Doorn RRA (2006) Ultralight aviation accident factors and latent failure: a 66-case study. Aviat Space Environ Med 77(9):950–952 32. Hasselquist A, Baker S (1999) Homebuilt aircraft crashes. Aviat Space Environ Med 70:543–7 33. Pagán BJ, de Voogt AJ (2008) Gyroplane accidents 1985–2005: epidemiological analysis and pilot factors in 223 events. Aviat Space Environ Med 79(10):983–985 34. Volodko AM (1996) Cause-factor analysis of helicopter accident rate. Proceedings of the 22nd European Rotorcraft Forum, pp. 31.1–31.13. Royal Aeronautical Society, Brighton, UK 35. Baker SP, Grabowski JG, Dodd RS, Shanahan DF, Lamb MW, Li G (2006) ems helicopter crashes: what influences fatal outcome? Ann Emerg Med 47(4):351–356 36. Taneja N, Wiegmann DA (2003) Analysis of injuries among pilots killed in fatal helicopter accidents. Aviat Space Environ Med 74:337–341 37. Hayden MS, Shanahan DF, Chen LH, Baker SP (2005) Crash-resistant fuel system effectiveness in Civil helicopter crashes. Aviat Space Environ Med 76:782–785 38. Rostykus PS, Cumming P, Mueller BA (1998) Risk factors for pilot fatalities in general aviation airplane crash landings. JAMA 280(11):997–999 39. Manwaring JC, Conway GA, Garrett LC (1998) Epidemiology and prevention of helicopter external load accidents. J Safety Res 29:107–121 40. de Voogt AJ, Uitdewilligen S, Eremenko N (2008) Safety in high-risk helicopter operations: the role of added crew. J Safety Sci. doi:10.1016/j.ssci.2008.09.009 41. Wolf BC, Harding BE (2008) Investigative and autopsy findings in sport aircraft-related deaths in southwest Florida. Am J Forensic Med Pathol 29(3):214–218 42. McConnell TS, Zumwalt RE, Wahe J, Haikal NA, McFeeley PJ (1992) Rare electrocution due to powerline contact in a hot-air balloon: comparison with fatalities from blunt trauma. J Forensic Sci 37(5):1393–1400 43. Moye SJ, Cruse CW, Watkings GM (1991) Burn injuries from small airplane crashes. Aviat Space Environ Med 62(11):1081–1083 44. Byard RW, Tsokos M (2006) Avulsion of the distal tibial shaft in aircraft crashes: a pathological feature of extreme decelerative injury. Am J Forensic Med Pathol 27(4):337–339 45. Wiegmann DA, Taneja N (2003) Analysis of injuries among pilots involved in fatal general aviation airplane accidents. Accid Anal Prev 35:571–577

8  Fatalities in General Aviation: From Balloons to Helicopters

179

46. Shkrum MJ, Hurlbut DJ, Young JG (1996) Fatal light aircraft accidents in Ontario: a five year study. J Forensic Sci 41(2):252–263 47. Kuhlman JJ Jr, Levine B, Smith ML, Hordinksy JR (1991) Toxicological findings in Federal Aviation Administration general aviation accidents. J Forensic Sci 36(4):1121–1128 48. Baker SP, Lamb MPH, Guohua LD et al (1996) Crashes of instructional flights. Aviat Space Environ Med 67(2):105–110 49. Li G, Baker SP (1999) Correlates of pilot fatality in general aviation crashes. Aviat Space Environ Med 70:305–309 50. Olson R, Austin J (2006) Performance-based evaluation of flight student landings: implications for risk management. Int J Aviat Psych 16(1):97–112 51. Benbassat D, Abramson CI (2002) Landing flare accident reports and pilot perception analysis. Int J Aviat Psych 12(2):137–152 52. Uitdewilligen S, de Voogt AJ (2009) Aircraft accidents with student pilots flying solo: analysis of 390 cases. Aviat Space Environ Med 80(9):803–6 53. Salvatore S, Stearns MD, Huntley MS, Mengert P (1986) Air transport pilot involvement in general aviation accidents. Ergonomics 29(11):1455–1467

wwwwwwwwwwwwwwwww

Chapter 9

The 9/11 Attacks: The Medicolegal Investigation of the World Trade Center Fatalities James R. Gill, Mark Desire, T. Dickerson, and Bradley J. Adams

Abstract  On September 11, 2001 two hijacked airplanes struck the Twin Towers at the World Trade Center (WTC) in New York City. All of the recovered human remains (21,741) to date have been examined by the New York City Office of Chief Medical Examiner (OCME). The major goals of the forensic medicine investigation of mass fatalities depends upon the circumstances and typically includes the determination of the cause and manner of death, accurate identification of the decedents, prompt issuance of death certificates, and collection of evidence from the remains. For the World Trade Center fatalities, the overriding concern was identification of the decedents. As of December 2008, there were 1,625 (59%) identifications of a total of 2,751 people reported missing. Of these, 996 were identified by a single means which included DNA analysis in 877 of the victims. DNA analysis markedly improves the ability to identify remains and has become the standard method for identification in these types of disasters. Expert anthropologic involvement also is a vital component. Keywords  Forensic pathology • Forensic biology • Terrorism • Fatalities • Mass disaster • DNA • Anthropology

Introduction The investigation of mass fatalities often involves transportation modalities (airplanes, trains, or ships) and currently terrorist involvement is a common concern [1]. The role of the forensic investigation of these deaths is twofold. One is the proper legal investigation of the scene and circumstances to determine if this is a criminal act and if so, attempt to answer the typical questions that follow (Who did J.R. Gill (*) Office of Chief Medical Examiner, 520 First Avenue, New York, NY 10016, USA e-mail: [email protected] E.E. Turk (ed.), Forensic Pathology Reviews, Volume 6, Forensic Pathology Reviews 6, DOI 10.1007/978-1-61779-249-6_9, © Springer Science+Business Media, LLC 2011

181

182

J.R. Gill et al.

it? Why? How?). If it is not criminal, the investigation is done by other agencies (FAA, OSHA, etc.). The second is the identification of the decedents. Regardless of criminality, the identification of the decedents is of prime concern. At the outset of any disaster, the forensic scientist must consider not only what needs to be done but also why it needs to be done. The forensic pathologist’s role in these deaths typically focuses on the autopsy in order to determine the cause of death and to collect any potential evidence. A forensic pathologist’s typical response to a homicide is to perform an autopsy. In the WTC attack, a new forensic paradigm was employed, and internal examinations were not done. The cause and manner of death were readily apparent, because the large majority (nearly 90%) of victims were fragmented. Furthermore, the time and place of death were known, the ­circumstances of the fatal episode were captured by television, and the identities of the perpetrator-terrorists were known. So when one asks why examine the ­decedents, the overwhelming answer is for identification of the decedents. In retrospect this may appear obvious; however, at the time of a disaster, it requires the rare ability to step back and think outside of the box. For identification purposes, the internal autopsy examination is of minimal value compared to the external examination and other techniques (dental comparison, DNA, radiology, and fingerprinting). DNA analysis is the most powerful technique in mass disaster identification of fragmented remains. Therefore, in some mass disasters, internal examinations do not need to be performed.

The Events of September 11, 2001 On September 11, 2001 two hijacked airplanes struck the Twin Towers at the World Trade Center (WTC) in New York City. The Office of Chief Medical Examiner (OCME) of New York City was responsible for the investigation of these deaths, the largest mass murder in US history [2, 3].

The Twin Towers The World Trade Center twin towers were 1,362 feet tall (110 stories) and 209 feet square. On a typical weekday, 55,000 people worked there. On September 11th, it was estimated that over 17,000 tenants and visitors were in the twin towers at the time of the attack. Underneath the towers was a transportation hub (four subway lines) and six basement levels with retail stores and a parking garage. There were 194 elevators in the towers and six stairwells (three in each). After the impacts, the elevators were inoperable and all the stairwells in the north tower were impassable. In the south tower, two stairwells were impassable and one was partially obstructed (18 people made it through the debris to safety). The towers’ main support was from the core and perimeter steel beams. At 1,100°F steel loses half of its strength. The subsequent fires reached 2,000°F.

9  The 9/11 Attacks: The Medicolegal Investigation of the World Trade Center Fatalities

183

The Airplanes Each airplane was a Boeing 767-200 that weighs 132 tons and carries 10,000 ­gallons of fuel. American Airlines flight #11 departed Boston for Los Angeles at 7:59 am with 87 passengers and crew and it struck the north tower (94–98th floors) at 8:46 am. It was estimated to be traveling at 429–494  mph. The north tower collapsed 102 min later (at 10:28 am). United Airlines flight #175 departed Boston for Los Angeles at 8:14 am with 60 passengers and crew. It struck the south tower (78–84th floors) at 9:03 am. It was estimated to be traveling at 537–586 mph. The south tower collapsed 55 min later (at 9:58 am). The velocity of the United flight was approximately 100 mph faster than the American flight which would have translated to an approximately 50% increase in kinetic energy. This increased energy and the lower impact height are two factors that accounted for the more rapid collapse of the south tower. The National Institute of Standards and Technology released estimates of where the tenants and visitors died. They reported that 1,466 people died in the north tower and 624 people in the south tower. In the north tower, at least 1,356 decedents were at or above the site of impact (94–98th floors) while 618 people who died in the south tower were at or above the impact site (78–84th floors). It has been estimated that fewer than 190 victims were below the impact site. Although the south tower was struck at a lower floor and was the first to collapse, these estimates corroborate the fact that people evacuated the south tower in the 17 min after the north tower and before the south tower were struck. It also reflects the people on the lower floors escaped before the collapse. In addition to the tenants and visitors, 421 rescue ­personnel, 147 airplane passengers and crew, and 18 people on the ground died [4].

Initial Investigation and Recovery of Remains [5] Initially 20,000 people were reported missing. Ultimately, at least 2,751 people from over 25 countries died in this disaster. The towers were reduced to a 70-ft tall pile of debris over a 16-acre area. Subterranean fires, which continued until December, also hampered the recovery. By May of 2002 the WTC excavation was complete after the removal of 1.7 million tons of debris and recovering 19,964 remains.

Initial Medical Examiner Investigations All of the remains were examined in a six-table arrangement in the Manhattan OCME (Fig. 9.1). A team approach involved medical examiners, forensic anthropologists, medicolegal investigators, forensic biologists, odontologists, and other support staff [6]. There were 30 medical examiners, more than 200 New York

184

J.R. Gill et al.

Fig. 9.1  The temporary examination area at the New York City Office of Chief Medical Examiner

Police Department (NYPD) general personnel, 5 NYPD print examiners, over 260 ­dentists, and more than 100 photographers and radiology and laboratory technicians. Other ancillary staff included the FBI, Federal Disaster Mortuary Operational Response Team (D-MORT), New York Fire Department (NYFD), NY Department of Corrections, police officers from adjacent Counties and States, medical students from New York University and Columbia, and the Salvation Army. Forensic anthropologists triaged the fragmented remains which then were ­examined by a medical examiner. During this triage, the anthropologists focus was on the recognition of commingling and determination of human vs. nonhuman. It was during this triage process that many instances of commingling were recognized and remains were segregated for subsequent testing [7]. Each identification team consisted of a medical examiner, a DNA technician, a missing person’s detective, a property clerk, and a scribe (medical students or physician volunteers). Remains were photographed and described; personal effects were logged and secured. Samples of tissue for DNA were collected from every potentially human remain. The environmental conditions and recovery time affected the ability to collect nondegraded DNA. During the prolonged recovery effort, the remains underwent the usual postmortem changes with the addition of heat and fire. Eventually, skeletal remains were the norm. Certain postmortem tissue samples were better suited for DNA analysis and yielded better results than others. The following are preferred: nonclotted blood, skeletal muscle (deep), or bone (compact bone such as humeral or femoral shaft). If possible, it is best to collect one muscle and one bone sample. For decomposed

9  The 9/11 Attacks: The Medicolegal Investigation of the World Trade Center Fatalities

185

remains, the preferred samples are deep muscle (if red), bone, and teeth. Samples that generally are not useful include decompositional fluid, clotted blood, hair, and gray muscle. One should attempt to obtain deep body samples to avoid possible surface contamination (e.g., by blood from another person) [8]. Depending upon the type and condition of the remains, they would undergo examination by radiography, fingerprint teams, and/or dentists. There were 288 intact bodies that were externally examined; no internal examinations were performed.

Family Assistance Center A Family Assistance Center (FAC) is a facility created in the aftermath of an incident with multiple fatalities. It has a number of objectives including providing information flow to family and friends of the deceased, the injured, and the missing. FAC personnel address the emotional needs of victim families and provide information in an appropriate setting and compassionate manner. Since the rapid and accurate identification of remains depends on the quality of information and items collected from families, it is important to begin this family outreach as soon as possible. To this end, the OCME collected ante mortem data from families including DNA ­samples and medical records. The OCME provided information and updates ­regarding the identification process. Collecting information from the WTC family members posed several problems. Many FAC intake specialists had not been trained in the use of DNA to identify a family member. OCME FAC specialists were not used at any of the WTC FACs. This led to some misunderstandings between the family and interviewer and a lack of standardization. Due to illegible handwriting and other individuals transcribing the information, data entry errors occurred [9]. Instructions were given to families to bring in personal effects of the victims (e.g., hairbrush, razor, undergarment, toothbrush) for DNA comparisons. Family members also gave “exemplars” for DNA to do kinship analysis. From the start of ante mortem collection a decision was made to collect DNA samples from family members of all victims, regardless of whether the decedents could be identified via photos, dental records, or fingerprinting. In many instances, it was the only opportunity to collect DNA from the family. Many of the victims’ families were not from the New York area and some family members did not want further contact with the OCME, even if the identification of their family member had been made or ­additional reference samples were needed from them. Having family DNA samples as a back up or to check existing identifications was important.

Identification The major goals of the Office of Chief Medical Examiner were accurate and prompt identifications of the decedents and issuance of death certificates [10]. The methods

186

J.R. Gill et al.

of identification included personal recognition, fingerprints, dental, radiographic, and unique marks (e.g., tattoos) or items (e.g., unique wedding band) [11]. A sevenpage comprehensive victim identification profile (V.I.P.) questionnaire was completed by the next of kin and entered into the database. Subsequent comparisons with recovered remains were performed by the OCME medicolegal investigation team. By April of 2002, just over 6 months after the disaster, 19,219 remains were examined and 968 decedents were identified. Of these, 40% of the identifications were by dental, 31% by DNA, 19% by fingerprints, and 10% by other. By October of 2002, 1,432 victims were identified and DNA was involved in 70% of the identifications. By December 2008, there were 1,625 (59%) identifications. Of these, 996 were identified by a single means which included DNA analysis in 877 (88%) followed by dental comparisons in 52 (5%). Without the use of DNA, almost onethird of all the decedents (over half of those ultimately identified) would not have been identified. DNA analysis markedly improves the ability to identify remains and is the standard method for identification in these types of disasters.

Death Certification Of those identified, the cause of death of the overwhelming majority was blunt injury (1,566). Asphyxia by debris accounted for 11 deaths and combinations of injury (blunt/thermal/smoke inhalation) caused 17. There were four deaths due to arteriosclerotic cardiovascular disease and three deaths certified outside of New York City. In addition, there were two delayed deaths from complications of WTC dust toxicity in which the exposures occurred on September 11th. Both involved complications of sarcoidosis. The unidentified remains were certified as “physical injures (body not found).” The manner of death for all the victims was homicide. A death certificate is needed by the next of kin for several reasons. Insurance companies, social security benefits for dependents, resolution of estates, etc., require a death certificate. Due to the vast amount of DNA analysis that would be needed for the majority of identifications, a plan was explored and implemented to help families quickly obtain death certificates in order to facilitate their insurance claims, etc. There is a law in New York State that allows the determination of death without physical remains (NYS section 2-1.7(b)). This law enabled our office to issue death certificates before the identification of remains after certain affidavits were produced. Use of legal statutes to assist in the timely issuance of death certificates benefit surviving family members. To have delayed issuing death certificates until all of the DNA analysis had been completed would have resulted in more than 1,000 families waiting for more than a year to receive a death certificate. Once positive identifications were made and the remains were ready for release, families were notified in person by local law enforcement. Since the DNA process was ongoing and multiple remains were found for many persons, families were given the option to claim the remains then or after all the DNA testing phases were complete. They also were given the option not to be notified if further remains were identified.

9  The 9/11 Attacks: The Medicolegal Investigation of the World Trade Center Fatalities

187

Fig. 9.2  Memorial Park (white tent with refrigerated tractor trailers), where the remains were kept, and other temporary supply and staff support buildings along East 30th street

It is important to engage family members and to be candid and forthcoming about the types of injuries and what to expect. Family group meetings were held regularly, information booklets (e.g., DNA) were created, and a DNA hotline ­number was established to facilitate communication. Remains were stored in 16 refrigerated tractor trailers adjacent to the OCME (Fig.  9.2). A family area was ­created nearby to allow the next of kin to visit. For potential future analysis, a tissuedrying facility preserved remains that could not be identified with current DNA technology. Tissue drying was chosen because it eliminates putrefaction, preserves DNA, simplifies environmental requirements for long-term storage of remains, and is acceptable to all religious groups.

Forensic Biology Challenges of WTC Identification The greatest challenge was the high number of decedents and human remains that would need to be identified. Decisions regarding the forensic biology identification process had to be made. Should every human remain be identified or should every

188

J.R. Gill et al.

victim be identified? If human remains are fragmented, what is the minimum size needed for DNA testing? The open population of the WTC disaster played a role in the decision that every human remain must be identified, rather than just identifying at least one remain for each victim (which might be done for a closed population such as a single airplane crash with a passenger manifest). DNA typing would be needed to identify all of the highly fragmented remains, as this technology does not require intact body parts such as fingers and mandibles that are necessary for other forms of identification. Any body fragment larger than the tip of a finger underwent DNA testing. As technology improved, even smaller fragments were analyzed. Factors that breakdown DNA or interfere with analysis were another major ­challenge. Techniques for DNA identification are not immune to external inhibiting agents. For the WTC disaster, various chemicals that are commonly found in building materials and in the ground were potential sources of such inhibitors. Environmental factors such as heat, fire, water, bacteria, and mold degrade DNA, a process that breaks the complete DNA strand into smaller pieces. These factors severely limit the ability to generate a usable DNA profile when applying traditional DNA technology. All of these factors were present during the months and years after the WTC collapse while remains still were being recovered. Another challenge was the commingling of remains. With the extensive body fragmentation and crowding of victims during the impacts and collapse, the likelihood of tissue integrating from two people was high. This integration was not always apparent during the initial gross examination; it would take DNA profiling to detect and correct this problem [8]. Commingling was discovered in two ways. One remain could yield a mixture of DNA from two victims, or two separate samples taken from an apparently single remain could give DNA profiles from two different victims. Remains were retested to confirm the multiple victims, and, when possible, the remains were separated so they could be properly identified. Even if the DNA results were consistent with a single individual, there was still the chance of a smaller fragment from one person combining with a larger body part of another. This made the teamwork of forensic biology, anthropology, and anthropology crucial [8]. Currently DNA analysis is performed in the 360,000 sq. ft, 15-floor, New York City OCME DNA building that has 62,000 sq. ft of laboratory space.

Reference Samples Three types of reference samples are utilized for mass fatality DNA identification. One is an antemortem sample generated while the victim was alive. This is a direct reference sample and includes such items as a toothbrush, razor, hairbrush, or clothing. Experience with WTC victim identification has shown that toothbrushes yield the highest quantity of DNA with the least chance of a mixture. One must be concerned

9  The 9/11 Attacks: The Medicolegal Investigation of the World Trade Center Fatalities

189

that a direct reference sample taken from a victim’s residence may have been used by another person. Several cases of DNA mixtures on a reference sample were detected. Razors are the second best source of DNA, although it is more difficult to recover the skin cells from between the blades of a disposable razor than simply pulling bristles from a toothbrush. Hairbrushes are good sources of DNA but have the highest risk of recovering more than one person’s DNA. Hair will remain in a hairbrush much longer than saliva cells on a toothbrush or skin cells on a razor, as these items are typically washed after each use. Other types of direct references include blood, tissue, and even semen samples collected during previous medical procedures. Teeth, umbilical cords, or hair saved as mementos from when the victim was younger also have proved useful. The second type of reference sample is derived from living family members. These are kinship samples used to compare with the DNA profile of recovered remains. The degree of relationship is calculated between living family members and the victim to be identified. Family samples typically are from buccal swabs. The closer the relationship between the family member and the victim, the more probative the sample will be. Parents, children, and siblings have the highest degree of relatedness. Other relatives can be used but are not as probative. A spouse is biologically unrelated to the victim; however, this sample may prove important when the victim’s child’s DNA profile is generated. Proving a relationship between the family member and the victim is imperative in this type of DNA comparison. Making sure the victim to family member relationship is correct was initially done at the FAC. Miscommunication at the FAC hindered the ability to make some identifications [9]. In some cases, the true relationship was not determined until after DNA profiling was performed on multiple family members of the same victim. One of the major hurdles was the vast number of profiles that needed to be collected and then compared. In a typical paternity test, a handful of profiles quickly can be compared by a single investigator. Similarly, with an unknown profile and a databank of known profiles, computer algorithms have been created to search for those direct matches as well [12]. But with the WTC cases there were 2,751 people with thousands of parental, children, and sibling profiles. There was no computer algorithm that could perform indirect comparisons to find matches. This software needed to be developed. Comparing family samples to remains when a direct reference sample was ­produced would seem like unneeded work; however, in some cases kinship analysis along with direct-reference-sample testing had shown mislabeled samples. This also helped support the fact that a reference sample was used for the correct individual. Using a combination of family and direct reference samples can give the identification a higher degree of confidence. The third type of reference sample comes from the remains themselves. If a victim is recovered and identified through visual, dental, or some other non-DNA method these remains would still have DNA work performed. This would generate the known DNA profile from that particular victim. Other fragments from that individual could then be identified through DNA.

190

J.R. Gill et al.

Types of Testing The DNA extractions were performed at the OCME (for tissue), Bode Technology Group (bone), and the New York State Police Laboratory (personal effects, exemplars). DNA extracts were profiled at Myriad Genetics. The DNA profiles were stored in computerized databases and all DNA identifications were done at OCME. The WTC DNA identification process used three separate types of DNA tests: short tandem repeats (STR), mitochondrial (MtDNA), and single nucleotide polymorphisms (SNPs) [13]. The majority of the work was performed with STRs. mtDNA and SNPs were initially thought to be more useful due to the amount of degradation that was expected, but the results of these additional tests did not ­significantly increase the number of victims who were identified. mtDNA was ­useful in confirming identifications, because a maternal relative would have the same mtDNA profile as the victim. Both mtDNA and SNP were used as additional tools and not for large-scale database searches. Therefore, it was not necessary to test large volumes of samples using this technology. The OCME made the decision to preserve samples for future work when DNA profiling advanced enough to complete work previously deemed impossible. Choosing bone over muscle samples led to fewer comingling problems and less chance of inhibition through degradation.

WTC DNA Database The database that holds information necessary to make a DNA identification is called the Mass Fatality Identification System™ (M-FISys), developed by Gene Codes Forensics, Inc. (Ann Arbor, MI). It contains a list of victims’ names with all of the reference samples tested in association with each decedent. The database also includes all of the profiles generated from remains recovered from all phases of work. A comparison is made of the more than 100,000 DNA profiles in the database and an automatic sorting is performed to group together samples that have matching DNA profiles. All of the short tandem repeat, mitochondrial, and single nucleotide polymorphism profiles are stored together in this fashion. With this variety of techniques and partial profiles, virtual (phased DNA) profiles were capable of making statistical thresholds. Samples that have discrepancies can be marked as well for further comparison. Quality control reports are built into the system as well. Contamination can be tracked and compared to all members of the DNA collection and analysis teams. The program available to the OCME performs straightforward allele frequency statistics on the made comparisons. The chance of another individual randomly ­having the same profile must be calculated. Based on the estimated number of victims, statistics for the threshold of direct matching was finalized at 1 in 200,000,000 for females and 1 in 2,000,000,000 for males. This number would be smaller for a population of victims in a closed manifest while being greater for a completely open manifest. Gender discrepancies are allowed since there were more male victims.

9  The 9/11 Attacks: The Medicolegal Investigation of the World Trade Center Fatalities

191

Making a New Identification When a postmortem sample is linked to a reference sample an administrative review is performed to confirm the match. Furthermore, an anthropological review is ­completed to confirm that newly identified remains are not incompatible with previously identified remains for the victim [7]. Both the reference sample and postmortem sample documentation are audited for potential errors. This work includes review of the chain of custody for all samples and FAC paperwork. Several agencies were involved in collecting and handling samples resulting in many different ways of recording information, assigning labels, and sending reports to the OCME [9]. The OCME was the only agency with the authority to issue identifications and death certificates. It was because of the abundance of WTC information, groups involved, and potential room for error that the OCME developed a Unified Victim Identification System (UVIS) in the aftermath of WTC work. This system tracks all antemortem data collection while maintaining the chain of custody. Having this system online for WTC work from the beginning would have made identifications much more efficient.

DNA Phases of Recovery From September 2001 to the end of 2002, Phase I of the recovery from Ground Zero resulted in 20,078 human remains; each was tested for DNA. Of these, 6,318 did not yield a usable profile due to insufficient sample size or degradation and so DNA work was discontinued. Phase II recovery began in 2006 due to the discovery of human remains in ­several unexpected locations around the WTC site (see below). These remains also underwent DNA analysis. At this time a modified bone extraction procedure, optimized by Bode Technology Group, is used (the OCME had been working with Bode and other private laboratories since 2001 to perform DNA extraction and profiling). The success in generating profiles from Phase II recoveries was surprisingly high considering these bone fragments were either buried or exposed to the elements for 5 years. As a result, this modified protocol also was used to reinvestigate the 6,318 bone fragments that were originally negative for DNA (from Phase I). This ­combined project has obtained results in 60% of the samples tested.

Anthropologic Search, Recovery, and Identification Efforts: Phase II Since 2006, there have been several unexpected discoveries of human remains around the World Trade Center site including in a manhole, on a rooftop, and buried beneath existing roads. Most of these remains were small bone fragments, usually

192

J.R. Gill et al.

ranging from a few millimeters to several centimeters. These discoveries triggered a renewed search and recovery effort of the area. For these renewed efforts (Phase II), OCME was assigned as the lead agency responsible for the search and identification work. As of December 2008, the Phase II project has included large-scale excavations within a two-block perimeter of the WTC site, as well as the search of several building rooftops and hundreds of subterranean structures. This has resulted in the removal of nearly 16,000 cubic yards of material, all of which was meticulously searched for human remains and associated personal effects from the 9/11 attacks on the World Trade Center. Forensic anthropology has played a critical role in almost every aspect of the Phase II work. For example, forensic anthropologists employ archeological ­techniques and direct the excavations at and around the WTC site. All construction areas associated with the rebuilding efforts around the World Trade Center site must be monitored by OCME anthropology staff for the presence of World Trade Center debris. If World Trade Center debris is located, this is taken as indication that human remains also could potentially be present. In areas where debris is discovered, it is the forensic anthropologists that direct the excavation progress and implement the appropriate techniques to ensure that material is completely removed, appropriately documented, and thoroughly sifted. Understanding and interpreting soil stratigraphy is crucial for undertaking this large-scale recovery endeavor. Forensic anthropologists have expertise in human osteology and are able to ­identify even very small bone fragments and distinguish them from nonhuman remains. With eight full-time members of the Forensic Anthropology Unit, OCME has the largest number of forensic anthropologists employed in a medical examiner’s office. It is this group that carries the responsibility of maintaining the forensic integrity of the Phase II fieldwork and the subsequent hand-sifting of the soil. During the height of the Phase II project (most of 2007), a crew of over 30 permanent and temporary anthropologists participated in various aspects of the daily operations. These anthropologists worked with other OCME staff that included medicolegal investigators, medical examiners, DNA analysts, evidence technicians, and safetyhazmat specialists. Although there have been numerous project areas associated with the Phase II recovery efforts, the Deutsche Bank has played a prominent role. The Deutsche Bank is a 40-story building at the southern edge of the World Trade Center site that was damaged during the 9/11 attacks. As a result, the building was slated for demolition and due to contamination (e.g., heavy metals and asbestos) much of the deconstruction had to be conducted under hazmat conditions. During some work on the roof to remove the ballast (gravel used to support the roof membrane), laborers found several small bone fragments intermixed with the rocks. The OCME was notified and the bones were assessed by a forensic anthropologist. Although the bone fragments were very small (<3 cm), they were confirmed to be consistent with human remains. Due to an increased frequency of similar discoveries on the roof, OCME initiated an on-site presence at the Deutsche Bank starting in April 2006. The work on the Deutsche Bank roof consisted of an on-site visual inspection of all the ballast for the presence of human remains and associated WTC debris.

9  The 9/11 Attacks: The Medicolegal Investigation of the World Trade Center Fatalities

193

All material was water-screened through fine wire mesh in order to recover the bone fragments and associated WTC debris. Due to the contamination, OCME personnel had to perform these operations in the appropriate personal protective equipment (e.g., respirators and hazmat suits) which added a layer of complexity to the ­process. The rooftop search for human remains at Deutsche Bank concluded in August 2006. In total, 766 bone fragments were recovered (most were smaller than 1 cm). All of these fragments were submitted for DNA testing and results indicated that the remains were largely associated with passengers from American Airlines Flight 11, which struck the north tower. In October 2006, human remains and personal effects again were unexpectedly discovered at the World Trade Center site in an abandoned utility manhole. The manhole was located on an access road along the western edge of the site. This find initiated a large-scale evaluation of other subterranean structures (e.g., sewers and manholes), additional building rooftops, and roadways. The results of a multi-agency evaluation identified numerous locations that were deemed to be potential locations for buried World Trade Center material. The primary Phase II project areas included complete excavation of two large areas bordering the World Trade center site (a construction access road measuring approximately 1,000 by 60 ft, as well as the former site of the St. Nicholas Greek Orthodox Church which is nearly 1 acre), over 600 subterranean structures including utility manholes and sewers, and five building rooftops. In order to maintain provenience information (spatial context), the Phase II recovery sites were divided into discrete excavation units. Every excavation area was tested for asbestos and heavy metals by taking soil samples before excavation. In order to sift the huge quantity of recovered material, a unique, two-part system was designed. The goal of this system was to make the handling and inspection of the material as efficient as possible. The first step in the process was to utilize a trommel when possible. A trommel is a large machine fitted with metal screens and conveyor belts for the purpose of mechanically separating fine soil from the larger material. The fine soil is not forensically significant and is mechanically removed, while the larger material is collected for subsequent hand-sifting. Use of the trommel was crucial for the Phase II project as it provided a means to mechanically reduce the quantity of soil that had to be hand-sifted by approximately 45% without impacting the forensic integrity of the material. During this automated sifting process, the original provenience information was maintained. The second step of the sifting process was the meticulous evaluation of the material by trained personnel. In order to complete the hand-sifting of the WTC material, a custom facility was constructed for this Phase II project (Fig. 9.3). Twenty-two hand-screening stations were constructed and approximately 35 cubic yards of soil were processed each day through the facility. All material was water-screened to ensure the most thorough recovery. To maximize efficiency, this hand-screening facility was designed to automate the transport of material through the use of conveyor belts and specially designed equipment. The only time material was manually handled was during placement into the screens and during the actual hand-sifting by trained personnel. Items of interest that were found during the sifting process were evaluated by the facility’s lead forensic anthropologist. As thousands of yards of

194

J.R. Gill et al.

Fig. 9.3  Custom hand-sifting operation used to search for human remains and personal effects as part of the Phase II recovery efforts

soil were removed from the excavation sites, the hand-sifting operation was the most time-consuming process. It also was one of the most crucial since the vast majority of the human remains and personal effects were discovered at this point. When the hand-sifting facility was deconstructed in December 2007, all of the equipment was retained for conversion into two mobile sifting platforms. These unique mobile platforms can be pulled by a truck and can be rapidly set up to ­facilitate the sifting of large quantities of material. As part of the renewed WTC recovery work, hundreds of additional human remains have been recovered and numerous remains have been associated to victims that were previously unidentified, while other remains have been linked to previously identified victims. It is the policy of the OCME to continue testing remains and to never give up in the still ongoing identification of victims from the World Trade Center attacks.

Conclusion The forensic medicine response to a mass disaster is a team approach [14]. Forensic pathologists, biologists, police detectives, anthropologists, dentists, and assorted technicians all have important parts in the evaluation of these fatalities. DNA analysis markedly improves the ability to identify remains and is the standard method for identification in these types of disasters with very large numbers of fatalities and

9  The 9/11 Attacks: The Medicolegal Investigation of the World Trade Center Fatalities

195

when fragmentation and/or decomposition preclude other methods of identification. Since each disaster is unique, the forensic pathologist must evaluate his/her role in the investigation to keep perspective on the specific objectives required by the individual situation. The main role of the forensic medicine team in the World Trade Center disaster was identification of the decedents. Expert anthropologic involvement is a vital component. The use of legal statutes that facilitate the prompt issuance of death certificates benefits surviving family members. Acknowledgments We thank Dr. Charles Hirsch for his review of the manuscript and insights to the WTC disaster. We also acknowledge Christian Crowder and Frank DePaolo for their contribution to the Phase II anthropologic investigation and Ben Figura for his assistance with the WTC statistics.

References 1. Natta D, Sciolino E (2005) Timers used in blasts, police say; parallels to Madrid are found. New York Times, New York. July 8, 2005 2. Hirsch CS, Shaler R (2002) 9/11 through the eyes of a medical examiner. J Investig Med 50:1–3 3. Gill JR (2006) 9/11 and the New York City Medical Examiner’s Office. Forensic Sci Med Pathol 2:29–32 4. Kirkpatrick SW, Bocchieri RT, Sadek F, MacNeill RA, Holmes S, Peterson BD et al. (2005) Federal Building and Fire Safety Investigation of the World Trade Center Disaster: Analysis of Aircraft Impacts into the World Trade Center Towers. U.S. Department of Commerce, Washington, D.C. (NIST NCSTAR 1-2B) 5. Marchi E, Chastain T (2002) The sequence of structural events that challenged the forensic effort of the World Trade Center disaster. Am Lab 34:13–17 6. Zelson-Mundorff A, Steadman D (2003) Anthropological perspectives on the forensic response at the World Trade Center disaster. General Anthropol 10:1–5 7. Mundorff AZ (2008) Anthropologist-directed triage: three distinct mass fatality events involving fragmentation of human remains. In: Adams BJ, Byrd JE (eds) Recovery, analysis, and identification of commingled human remains. Humana Press, Totowa, NJ, pp 123–144 8. Mundorff AZ, Shaler R, Bieschke E, Mar-Cash E (2008) Marrying anthropology and DNA: essential for solving complex commingling problems in cases of extreme fragmentation. In: Adams BJ, Byrd JE (eds) Recovery, analysis, and identification of commingled human remains. Humana Press, Totowa, NJ, pp 285–300 9. Hennessey M (2008) Data management and commingled remains at mass fatality incidents (MFIs). In: Adams BJ, Byrd JE (eds) Recovery, analysis, and identification of commingled human remains. Humana Press, Totowa, NJ, pp 337–356 10. Weedn VW (1998) Postmortem identifications of remains. Clin Lab Med 18:115–137 11. Simpson EK, Byard RW (2008) Unique characteristics at autopsy that may be useful in identifying human remains. In: Tsokos M (ed) Forensic pathology reviews. Humana Press, Totowa, NJ, pp 175–195 12. Cash HD, Hoyle JW, Sutton AJ (2003) Development under extreme conditions: forensic bioinformatics in the wake of the World Trade Center disaster. In: Altman RB, Dunker AK, Hunter L, Jung T, Klein T (eds) Pacific symposium on biocomputing. Singapore, World Scientific Printers, pp 638–653 13. Pinckard J (2008) Forensic DNA analysis for the medical examiner. Am J Forensic Med Pathol 29:375–381 14. Jordan FB (1999) The role of the medical examiner in mass casualty situations with special reference to the Alfred P. Murrah Building bombing. J Okla State Med Assoc 92:159–163

wwwwwwwwwwwwwwwww

Chapter 10

Injuries and Fatalities in All-Terrain Vehicle Crashes Richard J. Mullins and J.H. Mullins

Abstract  All-terrain vehicles (ATV) are gasoline-powered vehicles with a bicycle-like seat for the rider and handle bars for steering. They have large soft tires designed to provide optimal traction in the irregular terrain encountered in off-road conditions. Although compared to road traffic accidents involving cars, bikes, and pedestrians, ATV crash victims constitute a relatively small group of trauma victims, the forensic expert should be aware of the specific problems which can be encountered in ATV crashes. This chapter gives an overview over the characteristics of ATV and typical injury patterns resulting from ATV crashes. Keywords  ATV • Off-road vehicles • Injury pattern • Forensic pathology

Introduction All-terrain vehicles (ATVs) were first manufactured in the USA in 1971. ATVs enabled outdoor workers, i.e., farmers, dairymen, loggers, oilmen, to rapidly ­traverse rough terrain. ATVs proved to be reliable, rugged, and increased the efficiency of workers. By the 1980, ATVs have become popular with citizens participating in recreational activities such as camping, fishing, hunting, and racing over sand dunes. In the late 1980s, as sales in the USA of ATVs soared, public health officials and scholars reported that ATV riders, especially children, were at risk for serious injury and death. It is estimated that in 2006 in the USA there were 8.6 million ATV vehicles in use and a substantial proportion of riders are using the vehicles for recreation.

R.J. Mullins (*) Department of Surgery, Trauma/Critical Care Section L611, Oregon Health and Science University, 3181 SW Sam Jackson Park Road, Portland, OR 97239, USA e-mail: [email protected] E.E. Turk (ed.), Forensic Pathology Reviews, Volume 6, Forensic Pathology Reviews 6, DOI 10.1007/978-1-61779-249-6_10, © Springer Science+Business Media, LLC 2011

197

198

R.J. Mullins and J.H. Mullins

Definition of ATV The defining characteristics of an ATV are a gasoline-powered vehicle with bicycle-like seat for the rider who uses handle bars for steering. ATVs have large soft tires designed to provide optimal traction in soft soils, sand and irregular terrain encountered in off-road conditions. These soft tires have inferior traction on hardsurfaced roads. Initially, ATVs were designed with either a three wheel or four wheel chassis. Following a report in 1986 by Smith and Middaugh that three-wheeled chassis have high risk for rollover, evidence accumulated that three-wheeled ATVs were exceptionally dangerous [1]. Manufactures in the USA discontinued production of three-wheeled ATVs in the late 1980s and by 2005 less than 5% of the ATVs in use in the USA were of the tricycle configuration. For three decades ATVs were intended to carry a single rider, the driver. Nonetheless, many investigators of ATV injuries reported a substantial proportion of injured riders were passengers. Starting in 2005, manufactures have produced four-wheeled ATVs designed to accommodate a passenger riding in tandem behind the driver on a second straddle seat.

The US Consumer Product Safety Commission Authorities at the Consumer Product Safety Commission (CPSC) are US government employees vested with the responsibility of “protecting the public from unreasonable risks of serious injury or death from consumer products.” Since 1982, the CPSC investigators have compiled and published an annual report of ATV-related deaths in the USA. The report is entitled Annual Report: All-Terrain Vehicle Related Deaths and Injuries, and this report is available online at http://www.cpsc.gov/ library/atv2005.pdf [2]. CPSC investigators require strict case inclusion criteria in the Annual Report. Their intent is to only include decedents killed in an ATV crash. Other off-road vehicles include carts, dune buggies, golf carts, utility vehicles, which share a characteristic that the driver uses a wheel to turn the vehicle. Two-wheeled motorcycles, sometimes referred to as dirt bikes, are specifically designed for off-road use. CPSC investigators compiling a list of citizens killed each year on ATVs have a problem when the used nonspecific screening codes to select cases from large databases such as the accumulated annual death certificates in the vital statistics records of an individual state. Another problem is that some ATV riders are killed riding an ATV on a public road and are designated traffic deaths. To assure the accuracy of the Annual Report the CPSC investigators verify with secondary sources that decedents listed as killed in an ATV crash in fact died while on a vehicle that meets the definition of an ATV. In addition, the CPSC investigators use additional sources of information including police reports, obituaries, news agency reports, and medical records to identify all the individuals killed in the USA while riding an ATV. Two additional specific categories of decedents killed in ATV-related crashes are pedestrians or riders of other vehicles struck and killed by an ATV.

10  Injuries and Fatalities in All-Terrain Vehicle Crashes

199

The CPSC Annual Reports document that from 1982 through 1986 each year the number of ATV-related deaths in the USA increased. Then for 8 years annual deaths declined. CPSC investigators attributed this favorable reversal in death rates in part to a fall in the number of three-wheeled ATVs. In the late 1990s, the annual number of US citizens killed while riding an ATV increased. Since 1982 through the end of 2007, the CPSC Annual Reports indicates that at least 8,995 individuals have died from ATV-related crashes. For two decades children (under the age of 16) have been 25% of decedents. In recent years, CPSC officials have added to the death statistics in their Annual Reports estimates of ATV-related injuries in the USA. CPSC analysts have calculated an estimate of the annual rate of ATV injury using the National Electronic Injury Surveillance System (NEISS). NEISS is a database compiled annually by federal government statisticians that is probability-based sample of medical record information recorded on patients treated in the emergency departments of selected hospitals. The mechanism of injury code in the medical record is used to identify in NEISS the number of patients treated for injuries following an ATV crash. The CPSC Annual Report estimated that in 2000 92,000 patients sought treatment for injuries in an emergency department related to an ATV, and this number had increased 60% by 2007 to 151,000. Twenty seven percent were children. CPSC statisticians calculated an annual rate of ATV injury in the USA by dividing the number of injured by an estimate of ATVs operational based upon the number sold. The rate of 2 injuries per hundred ATVs in 2000 declined to 1.5 injuries per hundred ATVs in 2007. The CPSC statisticians would prefer to calculate annual rates based upon number of ATVs registered for use, but all state do not require registration of ATVs, and the number of functional ATVs is not known.

International Classification of Disease Codes and the Definition of ATV Public Health officials who analyze large databases to identify individuals killed or injured on an ATV have depended upon the International Classification of Disease codes. The CPSC used the International Classification of Disease, 9th revision, Clinical Modification (ICD-9-CM) through 1998 and then transition to the 10th revision. ICD-9-CM system listed the External Cause of Injury, E Code, as a numerical code. E Code 821.X is assigned to “non-traffic accident involving other off-road motor vehicle.” E Code 821 excludes an ATV rider injured while riding illegally on a public highway. Investigators who used the 821.X E code depend on the fourth-digit subdivisions to provide additional information. The codes 821.0 indicate the driver of the ATV, 821.1 indicate the ATV rider was a passenger, 821.8 indicates other occupant, and 821.9 indicates unspecified person, which may be applied in cases where individuals killed in an ATV crash are discovered adjacent to the vehicle, with no specific information regarding the circumstances of the crash.

200

R.J. Mullins and J.H. Mullins

Table 10.1  International classification of disease 10th revision, clinical modification (ICD10-CM) as listed on the World Health Organization Web site, version 2007 V86 is one category in the group V80-V89 “Other land transport accidents” V86: Occupant of special all-terrain or other motor vehicle designed primarily for off-road use, injured in transport accident Excludes: vehicle in stationary use or maintenance (W31) Subcategory definitions V86.0 Driver of all-terrain or other off-road motor vehicle injured in traffic accident V86.1 Passenger of all-terrain or other off-road motor vehicle injured in traffic accident V86.2 Person on outside of all-terrain or other off-road motor vehicle injured in traffic accident V86.3 Unspecified occupant of all-terrain or other off-road motor vehicle injured in traffic accident V86.4 Person injured while boarding or alighting from all-terrain or other off-road motor vehicle V86.5 Driver of all-terrain or other off-road vehicle injured in nontraffic accident V86.6 Passenger of all-terrain or other off-road motor vehicle injured in nontraffic accident V86.7 Person on outside of all-terrain or other off-road motor vehicle injured in nontraffic accident V86.9 Unspecified occupant of all-terrain or other off-road motor vehicle injured in nontraffic accident

Since 1999, the US National Center for Health Statistics, the government agency compiles the death certificate data recorded in the 50 states, has required that only ICD-10-CM coding be used. CPSC analysts welcomed the use of ICD-10-CM coding because the new code used for indicating ATV-related death, V86.X was more specific regarding ATV-related injuries (Table 10.1). However, the ICD-10-CM coding system is still a problem for authorities trying to compile information on exclusively ATV-related crashes because V86.X includes “all-terrain or other off-road motor vehicles” which means included within this group are individuals who were riding dune buggies, and the increasingly popular and dangerous two-wheeled dirt bikes [3]. A consequence of the change from ICD-9-CM to ICD-10-CM coding has been a distortion in the estimate over the past two decades of trends in rates of ATV injury death.

Public Health Authorities Response to Risks to the Public from ATVs Scholars and Public Health officials in the USA have been calling for 30 years for injury prevention interventions to be directed toward the risk of ATVs. From the mid-1970s to the late 1980s as the number of people riding ATVs escalated, reports of injuries in the medical literature appeared. Margolis reported in 1985 that ATVs caused 243 serious injuries and six deaths among “vocational and recreational users” in Maine [4]. Traumatic brain injury was the predominant lethal injury. Margolis observed that 90% of injured patients were male, and 60% were under the age of 20. Margolis highlighted that summer was the ATV accident season, and the

10  Injuries and Fatalities in All-Terrain Vehicle Crashes

201

poor weather conditions was rarely a contributing factor in an ATV crash. Conditions Margolis did report were associated with ATV crashes were alcohol intoxication and “poor judgment or carelessness.” Margolis’s paper is one of the first to report on ATV risks, and of historical interest is that 90% of ATVs involved in a crash in this study had low horsepower (range 16–30). In later years, the size of ATV motors will increase several fold. In an influential paper published in 1986, Smith and Middaugh identified that three-wheeled ATVs were substantially more likely to roll over than four-wheeled ATVs [1]. These authors in their analysis of ATV injuries and deaths in Alaska during 1984 and 1985 identified four rider-related risk factors: immature drivers, alcohol intoxication, failure to wear helmets and rider “inattention and excessive speed.” The authors called for regulations that prohibited children from driving heavy powerful ATVs that exceeded their capacity to physically control. Golladay and colleagues confirmed in 1985 that three-wheeled ATVs were a menace for young riders [5]. Sneed and colleagues highlighted in their report that children injured in ATV crashes are at risk to sustain permanent disability from spinal cord injury [6]. In 1987 and 1988 CPSC authorities and executives of the leading companies that manufacture ATVs negotiated a consent agreement in US District Court that was signed in March 1988 [7]. Seven legally binding rules were stipulated by the consent decree. ATV manufactures would cease production of three-wheeled ATVs. ATV manufactures would warn against passengers being permitted to ride on ATVs. The educational materials produced by ATV manufactures, including videos and curriculum for training courses would emphasize safety when riding an ATV. ATV manufactures endorse that the size of ATVs sold to children be compatible with the driver’s body size and strength. ATV manufactures would encourage those riders under the age of 16 to ride with adult supervision. ATV manufactures would endorse and promote the use of ATV only in off-road environments. ATV manufacturers would endorse and encourage riders to wear safety devices including helmets. The jurisdiction of this consent decree was for 10 years. After it expired in 1998 the consent decree was not renewed in court, but ATV manufactures decided to perpetuate voluntary standards “regarding ATV equipment, design, and configuration.” Signing the consent agreement in 1988 was followed by a decline in the annual number of deaths in the USA attributed to ATVs. That favorable trend reversed in the late 1990s. Helmkamp and colleagues used the National Inpatient Sample, a population-based dataset that provides an annual estimate of the number of patients hospitalized in the USA, to evaluate the burden of ATV-related crashes [8]. They used data from the 5-year period beginning January 2000, and estimated that 58,254 patients were hospitalized. These investigators used ICD-9-CM E codes to identify the individuals included in the cohort. The number of individuals injured per year doubled over the time period, and additionally the hospital death rate increased from 0.4 to 1.4%. Many of these patients were younger, 30% under the age of 18, but also 9% were over the age of 60 indicating that seniors are at risk, and may need to be the target of injury prevention interventions. An indication of the prevalence of recreational use of ATVs was that 50% of hospitalized patients were admitted between Friday evening and Monday morning. Helmkamp determined that 46% of

202

R.J. Mullins and J.H. Mullins

patients were treated in urban teaching hospitals, and 23% were admitted to rural hospitals. This distribution of patients may not represent the proportion of patients injured in rural locations because in many regions of the US regional trauma systems have been established that foster the expeditious transfer of seriously injured patients first treated in rural emergency departments to tertiary care trauma centers [3]. Helmkamp and colleagues used information in the National Inpatient Sample on patient charges for hospitalization to make an estimate of the financial burden of ATV-related injury in the first 5 years of the new millennium. Patients’ costs averaged $20,000 per hospitalization, and the US annual cost in 2005 exceeded one billion dollars. Ten percent of patients were sufficiently disabled by their injuries that they were discharged to a rehabilitation hospital or a nursing facility highlighting an additional burden to patient from being injured in an ATV crash.

ATV Pattern of Injury Authors of large series of patients injured in ATV-related crashes have reported patterns of injury that provide insights regarding what injury prevention measures might be used to reduce the risk to riders of ATVs. Acosta and Rodriquez reported that while 48% of patients treated at their tertiary trauma center had brain injury, fewer than 10% were wearing a helmet [9]. Sibley and Tallon reported that at their tertiary care trauma center major injury was a problem among ATV adult riders in the Canadian Province of Nova Scotia. A substantial majority of patients were males between the age of 16 and 34 years of age and ethanol intoxication was a risk factor in 25% of patients [10]. Seventy-six percent of patients were injured in the spring and summer months, and it took patient an average of 8 h to arrive at the tertiary center. An indication of the limitations that regulatory efforts is the fact that Sibley and Tallon reported in their series from a Canadian Province with a law mandating helmet use by ATV riders, only 25% of the injured patients had a helmet on when injury. The patterns of injury among children injured in ATVs showed the same range of diversity observed in adults. Cvijanovich and colleagues studied ATV injuries in children in Utah, by evaluating three databases; records of children treated in an emergency department and released, hospitalized injured children and dead children identified in death certificates [11]. The authors identified 268 injured children, and four of the eight fatally injured children were younger than 8 all were driving an ATV despite a law in Utah prohibiting these young children from driving ATVs. Three children were killed at the scene. The authors calculated that the majority of deaths and many of the injuries would have been prevented if children’s access to ATVs had been restricted as required by law. This study indicates the challenge faced by public health officials who expect that government regulations may be an effective method for reducing the risks related to unsafe ATV use. Murphy and Yanchar examined patterns of injury in children injured in ATVs crashes in Nova Scotia over a 14-year period ending in 2003 [12]. During that time period helmet use increase from 53 to 76%, and children who were injured in an ATV crash and who were wearing a helmet were better protected. An additional observation by these

10  Injuries and Fatalities in All-Terrain Vehicle Crashes

203

authors was the trend for more children with attention deficit hyperactivity to be injured on an ATV, suggesting the critical role that parents, guardians and older siblings should play in preventing injury. Keenan and Bratten attempted to determine the value of public laws and regulations intended to improve the safety of ATVs [13]. The authors compared children injured in Pennsylvania with children injured in North Carolina. Pennsylvania has laws that prohibit children under the age of 10 from driving ATVs on public lands (e.g., state parks), and require children wear helmets and have passed an ATV driving safety course. North Carolina did not have these laws and regulations. Children included in this study were seriously injured and entered into the statewide trauma systems operational in both states. The authors concluded that Pennsylvania’s laws were beneficial for the youngest children because the children killed in ATV crashes in Pennsylvania were older, and helmet use was more prevalent. Nonetheless, death rates for children in both states indicated there was still a problem of substantial needless risk. Kirkpatrick and colleagues analyzed 73 children whose average age was 10, and had injuries sustained in an ATV crash that required their hospitalization [14]. These authors noted a wide range of injuries but concluded especially alarming was the 5% death rate among children killed by a traumatic brain injury who were not wearing a helmet when the ATV they were on collided. These orthopedic surgeons considered fractures of the olecranon and elbow to be a particularly worrisome problem because of long-term disability. Upperman and colleagues analyzed data in the CPSC database for children killed in the USA over the years 1982–1998 [15]. In their analyses they demonstrated a wide range of annual death rates among states. The mortality rate for the most dangerous states was 0.57–0.10 per 100,000 children. Three sparsely populated states West Virginia, Alaska, and North Dakota had extraordinary high rates of death. The states with the lowest mortality rates had 0.09–0.01 deaths per 100,000 children. The authors speculate that in states with higher ATV death rates, there may be a greater emphasis on encouraging children to consider ATV riding as a risk-free entertainment. The authors of this analysis could not find that states with a higher profile of injury prevention regulations had lower death rates, although it cannot be known without more detailed data the extent to which unenforced regulations undermine the effectiveness of regulations at improving ATV safety. They observed that a substantial proportion of children killed on ATVs were passengers, and children as young as 3 years old were killed driving an ATV. Humphries and colleagues examined the pattern of injury among children treated in rural Kentucky following an ATV crash [16]. Brain injury was the leading cause of death and long-term disability, but fewer than 4% were wearing a helmet when they were injured while either driving (average age: 13) or riding as a passenger (average age: 9). Alarmed by the growing problem of more children injured annually as ATVs become more popular, the American Academy of Pediatrics has prepared recommendations for interventions and regulations that they expect could improve the safety of ATVs for children [17]. One idea advocated by many concerned about the marketing of ATVs as exceptionally exciting and risk-free pleasure is to emphasize to children and adolescents that ATVs are not toys, but rather potentially dangerous vehicles capable of disfiguring and maiming those who ride them in a careless manner. Wang and co investigators concluded from their review of the literature on children injured while riding an

204

R.J. Mullins and J.H. Mullins

Fig. 10.1  Child Injured riding an adult size ATV. An 11-year-old boy accustomed to driving his own small ATV decided to try to drive his uncle’s new adult size ATV. He did not wear a helmet because he planned to just ride it around the yard. He lost control and collided with a shed. He complained of face pain, headache, and leg pain. In the emergency department he was noted to have a 15 cm deep gash of his left thigh that was not actively bleeding. On CT of his facial bones (a) there is a fracture with fluid in the right maxillary sinus. There was a 7 mm punctate hemorrhage identified by CT scan in his frontal lobes and (b) a fracture of floor of the right orbit with orbital fat herniated into the maxillary sinus. Vision in the right eye was intact. The thigh laceration was explored and a 3-cm piece of plastic was retrieved from deep in the wound

ATV that commonly the injured child is a novice adolescent rider inadequately trained in the procedures for safe vehicle operation, and whose initial rides were with without proper adult supervision [18].

Mechanism of Injury Published manuscripts reporting ATV-related injuries have documented details regarding the mechanisms of injury sustained by riders of ATVs. These mechanisms of injury are valuable information in planning how to train riders to safely operate and to manufactures who seek improved designs that increase ATV safety. Collisions of ATV with immovable objects like tree stumps, boulders, or rocky terrain abruptly stop the vehicle, and the rider is ejected. High-speed ejection increases the rider’s risk of brain injury and spine fracture [19]. Risk factors for collisions are rapidly traversing unfamiliar terrain, driving in conditions of impaired vision such as between sunset and sunrise, and in rain, snow, or fog. Collisions occur when large powerful ATVs are driven by inexperienced or physically impaired drivers (Fig. 10.1). Collisions between two or more ATVs can lead to multiple casualties. Multi-vehicle crashes occur when multiple vehicles in a pack are riding at a high

10  Injuries and Fatalities in All-Terrain Vehicle Crashes

205

Fig. 10.2  Crushed right upper face. This man was driving an ATV up a ramp onto a flatbed truck. The ATV abruptly accelerated from under him. The ATV flipped backwards and crushed the right side of the driver’s face between the ramp and the ATV’s handlebar. CT scans demonstrate widely displaced fractures of the zygoma, inferior orbital rim and mandible. Mid-face sinus walls are fractured, and sinuses are opacified with fluid. The right globe was not injured, and vision was intact. Multiple procedures were required to reconstruct his face, but residual diplopia was a problem due to limited motion of the right globe

rate of speed over the undulating terrain of sand dunes. ATVs are big powerful machines, and loading them on to trailers and other transport vehicles can be risky (Fig. 10.2). A reason for prohibiting the driving of ATVs vehicles on traffic road is that these vehicles are not readily visible to larger vehicles, and the large soft tires that are effective in soft soils slip on hard pavement (Fig. 10.3). ATVs have struck and killed pedestrians and bicyclists.

206

R.J. Mullins and J.H. Mullins

Fig. 10.3  ATV vs. truck on public road. This 38-year-old man was riding an ATV on a remote highway near his ranch. He was struck on his left side by a pickup truck when it drove through an intersection, the truck driver never seeing the ATV until he collided with it. In the emergency department of a local hospital the ATV rider complained of left arm pain and dyspnea. A chest radiograph showed a ruptured left diaphragm. Image A. Computerized tomography of the chest and abdomen confirmed the diaphragm tear, and demonstrated the stomach had herniated into the chest posterior to the heart. Image B The patient’s left distal humerus was fractured. At emergency surgery, the patient had a laparotomy after the herniated stomach was returned to the peritoneal cavity the surgeon repaired the diaphragm. Three days post-ATV crash he had an open reduction and internal fixation of his distal humerus fracture. He made a satisfactory recovery

ATV rollover leading to injury and death of drivers and passengers is a common mechanism of injury. ATV rollovers can occur in three directions; sideways, forwards, or backwards. Each has a unique set of circumstances that lead to the rollover. ATVs can be top heavy, and when the rider is heavy and there is a load of cargo, the ATV can abruptly pitch sideways while traversing in parallel to the crest of an incline. The propensity for an ATV to roll sideways can be set off by the rider shifting his body’s weight although this may require in a large ATV an exceptionally strong agile adult. ATV rollover risk is increased by impaired riders who are more prone to error, failure to avoid dangerous terrain, carrying on the ATV the added weight of a passenger and imprudent loading of cargo (Fig. 10.4). ATVs roll in a rear-over-front forward rollover when a rider steers down too steep a grade and leans forward. The rider can be pitched over the handlebars on a downward incline if the rider abruptly shifts his or her center of gravity forward. The risk of forward rollover is greatest in steep terrains with an irregular surface where the forward wheels can become wedged. Riding an ATV at an excessive speed down a steep incline can increase the risk of forward rollover. A particularly dangerous ATV rollover occurs when an individual on an ATV that is driving up an incline leans backwards. A front-over-rear roll over can occur because the rider shifts the center of gravity to the rear, because a surge of power leads to the rear wheels accelerating out from under the rider, or because the rider falls or is bounced off the back of the vehicle. Another mechanism for front-over-rear rollover is for a rider to experience a decline in power riding an ATV up a hill and the ATV stalls. Many ATVs do not have a reverse gear. The rider must have strength and skill to turn a large ATV around that is stalled in an upward-facing direction. Riders who attempt

10  Injuries and Fatalities in All-Terrain Vehicle Crashes

207

Fig. 10.4  ATV Rollover on to right hip. This 38-year-old man was riding an ATV on the side of a hill. The vehicle rolled over sideways and his right hip was trapped beneath the vehicle. His colleagues lifted the vehicle off him and transported him to a local hospital. Computer tomography scans of his abdomen and pelvis revealed his right hip was dislocated (a, b). The patient was flown by helicopter to a tertiary trauma center. Six hours after injury he arrived, was immediately sedated, and his dislocated hip reduced (c). The patient subsequently had open reduction and fixation of the posterior wall of the acetabulum (d, e)

to accelerate upwards have had the vehicle pull out from under them and roll backwards. An added risk to the rider in the circumstance of a front-over-back rollover is that as the rider rolls down to the bottom of the incline, the vehicle bounces backwards as well, and crushes them (Fig. 10.5).

208

R.J. Mullins and J.H. Mullins

Fig. 10.5  Backward rollover of ATV. 55-year-old was riding up a steep hill “when the ATV reared up and rolled on top of me.” Found at the bottom of the hill, he complained of back pain and reported he could not move his legs. As he fell backward off the vehicle and rolled down the hill, the ATV followed, also rolling down the hill, and landed on top of him. On examination in the emergency department he had limited movement of his lower extremities. His lumbar spine X-rays demonstrated an L1 fracture. The patient has a posterior fusion of T11 to L2. On follow-up 8 weeks postinjury, his muscle strength in his legs was 4-out-of-5 flexion of his hips by his quadriceps femoris muscles and 3-out-of-5 extension of his toe by his tibialis anterior

Several mechanisms of injury that are not common, but associated with riding ATVs have been noted. While riding a high rate of speed, the rider can strike an object with their face or neck which can produce devastating brain injury, disfigurement, and blindness, especially if a helmet with face guard is not being worn [20, 21]. An exceptionally devastating injury occurs when the neck strikes a strung length of wire or rope (Fig. 10.6) [22]. Thompson and colleagues reported a group

Fig. 10.6  (continued) An hour was required for his riding companions to transport him to the closest hospital, a rural Level IV trauma center. During the ride he developed respiratory distress and on presentation to the emergency department had severe stridor. His anterior neck was swollen and there was subcutaneous air palpable from his face to his anterior chest wall. His larynx was tender and he could not speak. The emergency medicine physician managed his impending airway obstruction by performing an emergency tracheostomy. A CT scan of his neck revealed his tracheal cartilage was disrupted with edema and subcutaneous air. (a) There was rotary subluxation of the first cervical vertebra on the occipital condyle with the posterior margin of the left superior articular facet of C1 perched on a small compression fracture of the left occipital condyle. (b, c) An angiogram demonstrated a circumferential disruption of his right common carotid artery. (d, e) The right common carotid artery was explored, and a short segment with intimal disruption was excised, and a primary anastomosis performed

10  Injuries and Fatalities in All-Terrain Vehicle Crashes

209

Fig. 10.6  ATV rider garroted by a wire slung across a field. This young man was riding an ATV at a high rate of speed as dusk. He did not see a wire slung low across a field, and his neck collided with the wire. Jerked from the ATV he got up and was initially able to talk, although hoarse.

210

R.J. Mullins and J.H. Mullins

of children whose feet and ankles were lacerated and mangled in ATV power chain mechanisms [23]. Brandenburg and colleagues studied mechanism of injury in patients admitted to their trauma center [24]. Sixty-three percent of riders were injured in a roll over; 31% to the right, 29% back, 20% forward, and 19% to the left. Rollovers were associated with a collision one-third of circumstances. The rider was struck by the vehicle in over 30% of rollovers. Humphries and colleagues determine in a group of seriously injured children under the age of 18, that rider rolled the ATV 43% of the time, the rider collided with another vehicle or object 30% or the time, and rider and vehicle were “separated” from the ATV at the moment of crash 24% of the time [16]. A very similar distribution of mechanisms of injury was reported by Demas and colleagues, who also noted that 6% of injuries occurred when the ATV was driven over of a steep embankment or cliff. All of these mechanisms of injury have been reported to occur when children, immature in body strength and impulsive reckless judgment, drive ATVs [20]. The growing number of elderly citizens injured on ATVs suggests that as the age of the population in the USA gets older the elderly should be expected to be group proportion of the patients needing treatment following a crash of their ATV [25]. Sudden unexplained syncope of ATV riders has been reported, with the suspicion that a cardiac arrhythymia was the immediate cause of the patient’s death [26]. The ATV Safety Institute is a not-for-profit division of the Specialty Vehicle Institute of America. The ATV Safety Institute is sponsored by leading ATV manufacturers. The Institute has an informational Web site that instructs ATV riders regarding safe operation of their off-road vehicle. Videos are available from manufactures that demonstrate safe technique in riding an ATV, as well as proper use of safety equipment. The evidence published in the medical literature regarding patterns of injury suggest that injuries and deaths can be reduced and prevented if there is nationwide initiative to emphasize to ATV riders the benefits of learning how to safely operate these machines. An additional characteristic of off-road vehicle injuries that adds to this group of patients’ risk of poor outcome is the tendency for these to occur in remote rural locations [3]. Timely treatment of seriously injured patients reduces their risk of poor recovery and death particularly if they have sustained a traumatic brain injury. As part of a comprehensive response to the growing threat of ATV injury, manufactures should consider incorporating in vehicles methods for transmission that a crash has occurred and the location of the crash. This would accelerate the response of prehospital care providers in responding to the injured ATV rider and enable a prompt evacuation to a hospital. In addition, because ATV-injured patients often first present to small rural hospital that are typically closest to the remote rural scenes where ATV riders prefer to ride, a means for improving the ability of these hospitals to care for injured patients would be beneficial. Thus, statewide trauma systems are an asset for ATV enthusiasts because these assure injured patients with acute life, limb, brain, and vision-threatening injuries are promptly transferred to an appropriate tertiary care trauma center where they have access to expert trauma care.

10  Injuries and Fatalities in All-Terrain Vehicle Crashes

211

Conclusion ATVs are enormously popular in the USA primarily for recreational reasons, and the popularity of these vehicles worldwide is anticipated to grow several fold in the decade to come. As evidence accumulates regarding the specific mechanism of injury, causes of crash and risk factors for riders, public health-care officials must call for a comprehensive injury prevention response to the epidemic of death and disability related to ATV crashes. Rodgers examined the factors associated with death among ATV riders in the USA, and concluded that injury prevention initiatives need to target children, adolescents, and young adults [27]. Children, and perhaps infirm or obese geriatric riders, must be convinced that ATVs are not toys and easily handled, but instead are potentially dangerous. The risk from ATVs may substantially increase as larger more powerful machines are sold unless safety is incorporated in new designs. Safety equipment that is effective and also acceptable to ATV advocates who enjoy riding the vehicles needs to be developed. Manufactures should investigate modifications that improve safety and not just speed and power performance. Research into the details that cause ATVs to roll over is needed, and manufactures need to develop and test the effectiveness of roll bars, restraint devices and innovative methods of controlling acceleration and breaking. To reduce the dangers of ATVs, riders must be fully trained. Health-Care Policy advocates need to consider ways of convincing government authorities that what is needed are increased regulation over who drives ATVs and where these vehicles should be operated. Furthermore, methods for enforcement of regulations are needed. Perhaps an expansion and extension of the previously signed and enacted consent agreement between the CPSC and manufacturers is needed. In the process of developing the consent agreement, CPSC authorities, injury prevention advocates, ATV enthusiasts, and ATV manufactures must form a coalition that effectively reduces the death and disability brought to the citizens with the use of these all-terrain vehicles.

References 1. Smith SM, Middaugh JP (1986) Injuries associated with three-wheeled, all-terrain vehicles, Alaska, 1983 and 1984. JAMA 255:2454–2458 2. U.S. Consumer Product Safety Commission (2005) Annual report: all-terrain vehicle related deaths and injuries, Directorate for Epidemiology, Washington DC Available online at: http:// www.cpsc.gov/library/atv2005.pdf. 3. Mullins RJ, Brand D, Lenfesty B, Newgard CD, Hedges JR, Ham B (2007) Statewide assessment of injury and death rates among riders of off-road vehicles treated at trauma centers. JACS 204:216–224 4. Margolis JL (1988) All-terrain vehicle accidents in Maine. J Trauma 28:395–399 5. Golladay ES, Slezak JW, Mollit DI et al (1985) The three wheeler – a menace to the preadolescent child. J Trauma 255:2454–2458 6. Sneed RC, Stover SI, Fine PR (1986) Spinal cord injury associated with all terrain vehicle accidents. Pediatrics 77:271–274

212

R.J. Mullins and J.H. Mullins

7. U.S. Distric Court for the District of Columbia (1988) United States of America versus American Honda Motor Co., et al: Final consent decree – civil action number 87–3525; March 14 8. Helmkamp JC, Furbee PM, Coben JH, Tadros A (2003) All-terrain vehicle-related hospitalizations in the United States, 2000–2004. Am J Prev Med 34:39–45 9. Acosta JA, Rodriguez P (2003) Morbidity associated with four-wheel all-terrain vehicles and comparison with that of motorcycles. J Trauma 55:282–284 10. Sibley AK, Tallon JM (2002) Major injury associated with all-terrain vehicle use in Nova Scotia: a 5 year review. CJEM 4:263–267 11. Cvijanovich NZ, Cook LJ, Mann NC, Dean JM (2001) A population-based assessment of pediatric all-terrain vehicle injuries. Pediatrics 108:631–635 12. Murphy N, Yanchar NL (2004) Yet more pediatric injuries associated with all-terrain vehicles: should kids be using them? J Trauma 56:1186–1190 13. Keenan HT, Bratton SL (2004) All-terrain vehicle legislation for children: a comparison of a state without and without a helmet law. Pediatrics 113:330–334 14. Kirkpatrick R, Puffinbarger W, Sullivan JA (2007) All-terrain vehicle injuries in children. J Ped Orth 27:725–728 15. Upperman JS, Shultz B, Gaines BA, Hackam D, Cassidy LD, Ford HR, Helmkemp J (2003) All-terrain vehicle rules and regulations: impact on pediatric mortality. J Pediatric Surg 38:1284–1286 16. Humphries RL, Stone CK, Stapczynski JS, Florea S (2006) An assessment of pediatric allterrain vehicle injuries. Ped Emerg Care 22:491–494 17. Anonymous (2000) All-terrain vehicle injury prevention: two-, three-, and four-wheeled unlicensed motor vehicles. Pediatrics 105:1352–1354 18. Wang BS, Smith SL, Periera KD (2007) Pediatric head and neck trauma from all-terrain vehicle accidents. Otolaryngol Head Neck Surg 137:201–205 19. Sanfilippo JA, Winegar CD, Harrop JS, Albert TJ, Vaccaro AR (2008) All-terrain vehicles and associated spinal injuries. Spine 33:1982–1985 20. Demas PN, Braun TW (1992) Pediatric facial injuries associated with all-terrain vehicles. J Oral Maxillofac Surg 33:329–332 21. Carr AM, Bailes JE, Helmkamp JC, Rosen CL, Miele VJ (2004) Neurological injury and death in all-terrain vehicle crashes in West Virginia: a 10-year retrospective review. Neurosurgery 54:861–866 22. Graham J, Dick R, Parnell D, Aitken ME (2006) Clothesline injury mechanism associated with all-terrain vehicle use by children. Ped Emerg Care 22:45–47 23. Thompson TM, Latch R, Parnell D, Dick R, Aitken ME, Graham J (2008) Foot injuries associated with all-terrain vehicle use in children and adolescents. Ped Emerg Care 24:466–467 24. Brandenburg MA, Brown SJ, Archer P, Brandt EN Jr (2007) All-terrain vehicle crash factors and associated injuries in patients presenting to a regional trauma center. J Trauma 63:994–999 25. Deladisma AM, Parker W, Medeiros R, Hawkins ML (2008) All-terrain vehicle trauma in the elderly: an analysis of a national database. Am Surg 74:767–769 26. Motozawa Y, Hitosugi M, Kido M, Kurosu A, Nagai T, Tokudome S (2008) Sudden death while driving a four-wheeled vehicle: an autopsy analysis. Med Sci Law 48:64–68 27. Rodgers GB (2008) Factors associated with the all-terrain vehicle mortality rate in the United States: an analysis of state-level data. Accident Anal Prev 40:725–732

Chapter 11

Advances in Entomological Methods for Death Time Estimation Martin H. Villet and Jens Amendt

Abstract  The development of entomological methods for estimating the time of death has been rapid in the last decade, and new methods are on the horizon. These developments are reviewed with specific reference to experimental design, established and new techniques, and mathematical modelling for forensic retrodiction. The techniques include the use of electron microscopy, magnetic resonance imaging, computed tomography, cuticular hydrocarbon profiles, and real-time PCR to estimate the age of immature insects found on corpses, based on their stage of development. Near-infrared spectrograph and pteridine fluorescence techniques can be applied to this task on adult insects. The use of ecological succession in the carrion insect community is also introduced briefly. Finally, the creation and uses of standard techniques in forensic entomology is discussed. We recommend that two steps in this standardisation process are that physiological and ecological studies should be reported in physiological time wherever this is appropriate, and that the type of post-mortem interval being estimated should be stipulated more explicitly than is currently common. Keywords  Death time estimation • Post-mortem interval • Entomology • Insects

Introduction Forensic entomology is best known as a means of providing evidence of the time of death of a body based on the insects associated with it, although the discipline covers other topics too. It uses a variety of ecological and developmental processes

J. Amendt (*) Institute of Forensic Medicine, University of Frankfurt, Kennedyallee 104, 60596 Frankfurt/Main, Germany e-mail: [email protected] E.E. Turk (ed.), Forensic Pathology Reviews, Volume 6, Forensic Pathology Reviews 6, DOI 10.1007/978-1-61779-249-6_11, © Springer Science+Business Media, LLC 2011

213

214

M.H. Villet and J. Amendt

to estimate the passage of time, each with their own strengths and caveats. The first concern of this review is to explain what “time of death” means in terms of these methods. The first comprehensive textbook of entomological methods for estimating time of death was provided by Smith [1], and subsequently updated in books edited by Catts and Haskell [2], Byrd and Castner [3] and Amendt et al. [4]. The rate at which new methods for obtaining evidence are becoming available has escalated dramatically since 2000, and forensic entomology is, in a sense, coming of age. The second purpose of this review is to summarise the advances made since 2000, and to point to future research directions. This section of the review is structured in terms of the types of questions that forensic entomologists need to answer when making their estimates: the identity and age of specimens, and their biological meaning. In other fields of forensic science, progress has been marked by the development, acceptance, and publication of quality assurance standards [5, 6], and forensic entomology is still at the start of that process [7]. Thus, the third aim of this article is to describe a framework within which reliable methods can be brought into the mainstream of applied forensic entomology.

Time of Death in Forensic Entomology Post-mortem Events and Intervals There is some contention about what the definition of “time of death” means in terms of entomological evidence because it depends on the form of evidence being considered (Fig. 11.1). In most literature [1–4], reference is made simply to a postmortem interval or PMI, while more recent literature [8, 9] refers to period of insect activity (PIA) or time of colonisation (TOC, i.e., duration of colonisation, because it is an interval and not an event). In the case of submerged corpses, the concept of post-mortem submergence interval (PMSI) has been developed [9]. The reason for these fine-tuned terms is that they apply to different means of estimating the duration of time since death (Fig. 11.1). In the case of estimates made from the development of insects, there is a lag between the onset of decomposition at death, the arrival of adult insects, and the deposition of the first immature insects. The developmental clock is started only once eggs are laid (Fig.  11.1) and the resulting estimate of PMI is, strictly speaking, a minimum post-mortem interval (PMImin) because it represents the most recent date or time at which death may have occurred [10, 11]. Seen in that light, PMImin refers to an event and not a duration or interval, and it would be less ambiguous to speak of the duration of insect development. Exceptionally, the duration of development is not a PMI when the immature stage of flies infest living tissue pre-mortem (termed myiasis), for instance in cases of neglect [e.g. 12].The lag between death and the oviposition of insects may be of the magnitude of less than an hour to several days [13], but constitutes an increasingly negligible proportion (technically termed relative error) of the estimate as the PMI increases [10].

11  Advances in Entomological Methods for Death Time Estimation

215

Fig. 11.1  Relationships of duration of insect development (PMImin), time of colonisation (TOC), period of insect activity (PIA), ecological succession, maximum post-mortem interval (PMImax) and events surrounding a death. The discovery of the corpse may occur after the insects complete development. The lags between death, the first arrival of insects and oviposition by insects may represent minutes or days depending on the contingencies surrounding the death. The shaded boxes represent the windows of prediction associated with each estimator of the time since death

On the other hand, the clock of ecological succession on corpses starts when the body dies and decomposition processes like autolysis and bacterial putrefaction start to produce the chemical signals that subsequently attract insects [e.g. 14, 15], and therefore includes the lag period before insects oviposit (Fig. 11.1). The resulting estimates of PMI are therefore not minima [cf. 11]. The time when a person was last seen alive provides an estimate of the earliest date or time when they may have died, and could therefore be termed the maximum post-mortem interval (PMImax: Fig. 11.1) or the last time seen alive. There is a strong argument to be made for disambiguating the terminology about time of death by referring directly and explicitly to the duration or event that is being estimated, e.g. initiation of insect development, duration of insect development, initiation of ecological succession, or duration of ecological succession. The possibility of artificially disrupting ecological succession and oviposition [e.g. 16] may

216

M.H. Villet and J. Amendt

require specific terminology for such unusual situations. However, the purpose of this discussion is primarily to raise awareness of the meaning of “time of death,” and not to generate new terminology or three-letter acronyms. Estimates of time of death are given as windows of prediction around the event (Fig. 11.1) because of the uncertainties inherent in estimation [e.g. 10]. The boundaries of such windows are an expression of the accuracy of the estimate, and reflect the precision and any bias of the underlying methods and techniques.

Physiological Time A second important point about entomological estimates of time of death is that insects are heterotherms (their body temperatures fluctuate greatly), which means that environmental temperature has an important influence on their metabolic rates. Within lethal limits, when conditions are warmer, insect metabolism and activity is faster, so that both developmental [10] and ecological processes [17] unfold faster when measured in standard chronological units, e.g. hours or days. The clock-like properties of insects’ biology are therefore calibrated in what is termed physiological time, which is generally transformed to chronological time by taking into account the effects of concurrent temperatures [18]. Comparable accommodations are necessary in pathologists’ estimates of time of death based on livor mortis or rigor mortis. Physiological time is usually measured in thermal accumulation units such as hour-degrees (h°C) or day-degrees (d°C) in forensic entomology [10]. Measuring physiological time also requires that one establishes the threshold temperature below which the physiological activity effectively ceases and therefore stops accumulating [cf. 19–21]. An approximation of this threshold temperature is available in some models [e.g. 17, 18], and requires that the physiological process is measured over a range of realistic temperatures [22]. Unfortunately, this is not always possible, and many contemporary studies have been compelled to report their results in chronological time. It may be possible to estimate a physiological time from such studies if a suitable threshold temperature is available for the particular species from another study. Any attempt to do this should take into account the effects of developmental plasticity and geographical variation on these thresholds [23–25]. We recommend that future forensic entomological studies report their results in physiological time (and the associated threshold temperature), rather than chronological time, where it is appropriate.

Identifying Specimens A crucial step in forensic entomological investigations is to identify the insects involved. Each species of insect has its own specific biology and suite of adaptations. This means that even closely related species [26–28] or even populations [23]

11  Advances in Entomological Methods for Death Time Estimation

217

Fig. 11.2  Typical life cycles of (a) beetles and (b) flies, with the associated terminology for the stages and the events that punctuate them. The larval stage is divided into several instars by the shedding of the exoskeleton (ecdysis) for growth. In flies, pupation takes place within the skeleton of the larva, so that it cannot be detected without dissection or an anatomical imaging technique, and the transition between these stages is therefore indicated vaguely

are distinct sources of forensic information and, conversely, that the most detailed and precise interpretation of entomological evidence begins with an accurate identification of the insects involved. A variety of techniques are relevant.

Morphology The insects that are of greatest forensic significance in estimating time of death are the flies and the beetles, which both have life cycles that involve four sequential stages: egg, larva, pupa, and adult (Fig 11.2). The earliest stage of the insect life cycle that occurs on a corpse and which is useful in estimating PMImin is the egg [29–31]. A great deal of work has been published on the identification of fly eggs based on their morphology [32–35]. This work has generally involved sophisticated electron microscopy, but a simple potassium permanganate staining technique for light microscopy has been published [32]. Eggs hatch into the larval stage, which is subdivided into several instars by the shedding of the exoskeleton (an event termed ecdysis: Fig. 11.2) to allow growth. Szpila et al. [36] and Szpila [37] have also used electron microscopy to study the first instar larvae of blowflies, which are notoriously difficult to identify by traditional means. They have described a suite of new diagnostic characters that can help to solve this problem, allowing the differentiation of first instar larvae and the application of species-specific developmental models.

218

M.H. Villet and J. Amendt

The third stage of the life cycle is the pupa (Fig. 11.2). Because fly larvae pupate inside their final larval skins in a process termed pupariation, the external morphology of mature maggots can be used to identify puparia too. The pupa gives rise to an adult at an event termed eclosion (Fig. 11.2). Identifications of insects can often be made quickly and cheaply based on their external structure. However, the diagnostic characteristics are specific to particular stages of the life cycle, and if these are physically damaged or under-researched, the following molecular techniques should be tried instead.

Genetic Techniques Molecular techniques for identification have developed enormously, and include genetic methods, which can be applied to all stages of insect life cycles, and cuticular hydrocarbons, which are discussed in the next section. The genetic methods include direct DNA sequencing [e.g. 38–45], randomly amplified polymorphic DNA (RAPD) analysis [e.g. 39], restriction fragment length polymorphism (RFLP) analysis [46], short tandem repeat (STR) analysis [e.g. 47], and amplified fragment length polymorphism (AFLP) analysis [e.g. 48]. These well-known methods are mentioned primarily to show the data sources that have been tested; details of the methods and results should be sought in the original publications. These methods have well-known caveats, many of which are discussed by Wells and Stevens [49].

Cuticular Hydrocarbons Cuticular hydrocarbons show some promise for identification of specific insect species [50]. They are chain-like hydrophobic lipids that accumulate on insects’ cuticles until they are shed, waterproofing the cuticle, repelling pathogens and mediating copulatory behaviour in carrion flies [51, 52]. Apparently, most cuticular hydrocarbons are synthesised by decarboxylation and chain elongation by the insects themselves, with only a small proportion derived directly from their diets [52, 53], making them quite genome-specific and therefore useful in identification of species. Their chemistry and some means of analysing their abundance are reviewed by Drijfhout [52]. They show characteristic variation that is generally more quantitative than qualitative, even between species [54], and which is therefore best analysed using their ratios or relative frequencies. Cuticular hydrocarbons derived from freshly shed puparial cuticles can be used to discriminate between the carrion flies Aldrichina grahami, Chrysomya megacephala, C. rufifacies, Lucilia sericata, Sarcophaga peregrina and S. crassipalpis [54]. In most cases, the distinctions were qualitative, but C. megacephala and L. sericata had to be separated quantitatively. Cuticular hydrocarbons may even be able to provide taxonomic resolution at the level of conspecific populations. Quantitative differences were found between adults

11  Advances in Entomological Methods for Death Time Estimation

219

of three different geographical populations of P. regina using discriminant function analysis [55], and fifteen populations of Chrysomya bezziana could be distinguished by canonical variate analysis [56]. Although Roux et  al. [57] suggest that cuticular hydrocarbons can serve as markers for evolutionary relationships (i.e. phylogenetic patterns) for blow flies, it is not possible to test this rigorously with less than four terminal taxa, and they used only three species. Phylogenetic patterns in the qualitative differences in the hydrocarbon pools of puparia exuviae of five species of blow flies and flesh flies were not immediately obvious [54], and cuticular hydrocarbons have proved unreliable as phylogenetic indicators in insects in general [50]. This means that, unlike phylogenetic analysis of DNA data, such analyses of cuticular hydrocarbons may not have power to predict identifications of taxa that have not been investigated already [cf. 38, 40, 41].

Estimating the Duration of Insect Development The development of immature insects has clock-like predictability. It can also be adjusted relatively easily to take into account the effects on temperature on insect metabolic rates. Blowflies and fleshflies have been used to provide estimates of time of death for decades, and are useful in a time frame of as much as 5 weeks after death, depending on season. Midgley and Villet [58] showed that the silphid beetle, Thanatophilus micans, arrives at carrion very soon after its exposure and has a longer development than flies, thus extending the period over which PMIs can be estimated by this method. Similar advantages exist with hymenopteran parasitoids, which emerge from their hosts after unparasitised hosts have eclosed [59, 60]. Development data are available for five cosmopolitan parasitoids [59–62], and several other species await detailed investigation [63, 64].

Measurement Accuracy, Precision and Bias One of the first concerns of any investigation, experimental or criminal, is the quality of the evidence [6]. Villet et al. [10] have discussed the factors affecting accuracy in studies of insect development. Accuracy comprises both bias and precision, and each of these can be compromised by the way that the sample is selected (sample precision and sample bias), the way that the sample is measured (measurement precision and measurement bias), and the way that the data are analysed (estimate precision and estimate bias). Richards and Villet [22] have discussed the relative impact of various aspects of sampling protocols in the experimental quantification of insect development, and empirically shown that an increase in temporal resolution is more crucial than an increase in the numbers of immature insects sampled at each time. This resulted in recommendations that will help to standardise the quantification of developmental data [22].

220

M.H. Villet and J. Amendt

Table 11.1  Common factors affecting the duration of insect development, and their largest likely effect Effect Magnitude of effect References Precocious egg development Hours [142], [143] Circadian rhythms Hours [144], [145], [146] Light-induced variability Hours [147] Preservation and measurement Hours to days [148], [149], [134] Tissue type of diet Days [139], [140] Drugs Days [150], [151] Maggot-generated heat Days [152] Intra- and inter-specific competition Days [140], [153] Wandering Days [154] Developmental plasticity Days [115], [23] Chilling Days to weeks [155], [156], [157] Diapause Months [158]

The rate of development of insects is affected by a suite of influences (Table 11.1), the most significant of which is the temperature of the immature insects [18]. Villet et al. [10] reviewed the biological sources of variation that affect estimates of PMImin, and tabulated the relative precision and potential bias of each one. Comparisons of the development of the sister species of blowflies Chrysomya chloropyga and C. putoria from the same geographical place (thus controlling for many ecological variables) showed that it cannot be taken for granted that the biology of one species can be used to estimate that of another [26], highlighting the importance of correct identification. These studies offer insight into the limitations of estimates based on insect development (see below).

Gene Expression and RNA Analysis The most accurate results for age estimation can be expected for the larval stage, which is illustrated by the fact that many studies give the times from oviposition or larval hatching to pupariation on a quite precise scale [7, 65–67]. On the other hand, estimating the age of pupae is currently not easy without rearing them to the adult stage. Apart from some changes in colour and length that occur during pupariation (Fig. 11.2), the first few hours of a developmental stage that lasts for days (Fig. 11.2), pupae show very few outward changes that might be quantified. Hence rearing to the developmental landmark of eclosion has traditionally been required, which is unfortunately time-consuming and not always successful. However, the process of metamorphosis that occurs inside the puparium represents about 50% of the immature development time and therefore its progress may serve as an important tool in entomological PMI estimation. Metamorphosis is characterized by developmental modifications such as cellular proliferation, tissue remodelling, cell migration and apoptosis.

11  Advances in Entomological Methods for Death Time Estimation

221

Not surprisingly, many genes are involved in these actions [68] and display agedependent expression, which might provide a tool for estimating the age of blowfly pupae. Gene expression studies have demonstrated consistent age-dependent changes in global gene transcription profiles in Drosophila [69, 70] and mosquitoes of the genera Aedes [71] and Anopheles [72]. Ali et al. [73] recently applied this approach on a blowfly species, and Tarone et al. [74] finally introduced the technique into the field of forensic entomology. Zehner et  al. [75] performed a differential-display reverse transcription PCR (ddRT-PCR) and quantified the transcript abundance in a real-time PCR system. They identified nine differentially expressed genes in the forensically important blowfly Calliphora vicina and were able to discriminate between pupae at the beginning and at the end of metamorphosis and with an age of at least 120 h (at 25°C).

Hormone Production Gene transcription produces RNA, some of which is translated into proteins that catalyse the production of specific metabolites such as hormones. During the pupal stage, the secretion of a hormone, 20-hydroxyecdysone, fluctuates in a highly predictable way, reaching a peak 36–96 h after pupariation at about 20°C in the blowfly Protophormia terraenovae [76]. Levels of this hormone can be measured in individual pupae using enzyme immunoassay (EIA), and offer an indirect measure of gene expression that can be used to estimate the age of pupae. The pattern was not detectable in pupae that had been preserved by freezing, and its limitations are discussed by Gaudry et al. [76].

Cuticular Hydrocarbons Like patterns of gene expression, cuticular hydrocarbons show promise for estimating the age of individual insects in various stages of development [10]. Larvae of C. rufifacies had age-dependent signatures in the ratios of their cuticular hydrocarbons [77]. The coefficients of determination of the relationships between age and various hydrocarbon ratios were all above 0.8, which corresponds to correlations above 0.89, although the variances were not stationary. The relationships were approximately exponential, so that they provide narrower inverse prediction intervals for older larvae, which is a desirable characteristic. Using discriminant function analysis of the entire cuticular hydrocarbon pool, Roux et al. [78] were able to distinguish larvae of different ages (in days) for C. vomitoria, C. vicina and P. terraenovae, but the pattern of variation was irregular. There are promising indications that similar patterns may exist in Lucilia cuprina [79], and probably in flies generally.

222

M.H. Villet and J. Amendt

Puparia of Lucilia cuprina,C. vomitoria, C. vicina and P. terraenovae yielded no hydrocarbons [78, 79]. However, by opening the head end of the puparium, Roux et  al. [57, 78] extracted hydrocarbons from the “nymph’s membrane.” Even-length linear alkanes accumulated progressively on puparia of C. vicina [78], but showed stable levels in C. vomitoria and P. terraenovae after a couple of days, so that the age of the puparia could not be estimated from them with further precision. Fortunately, discriminant function analysis of the suite of cuticular hydrocarbons was usually successful in distinguishing specimens of different ages, and the pattern of variation was less irregular than in larvae [78]. Puparia of Sarcophaga bullata showed a strong age-related pattern in their total hydrocarbon secretions [80]. Exuviae and puparia may remain at decomposition sites for months or, depending on environmental parameters [81], even years after other traces have disappeared. The hydrocarbon profiles of puparia change as the puparia weather after eclosion, gradually converting longer chains to shorter ones [82]. This involves physical, chemical and biological processes [52], and may provide a very useful means to estimate post-mortem intervals in certain situations. Adult flies also show variation in their cuticular hydrocarbons that may have forensic significance in estimating PMIs, although the results to date are ambi­ guous. Stoffolano et al. [51] reported that there was no variation related to age, diet or sex in adults of Phormia regina that were caught wild or kept in culture since 1971, but quantitative differences were found between adults of three different geographical populations of P. regina using discriminant function analysis [55]. Quantitative variation in hydrocarbon profiles attributable to age and sex was found in Chrysomya bezziana using canonical variate analysis [56] and in Calliphora vomitoria [83]. As expected in heterotherms, adults of Musca domestica produce more hydrocarbons at higher temperatures, and the relative proportions of some of them also change with temperature [84], probably due to differences in the rate kinetics of the relevant enzymes. The production of cuticular hydrocarbons is also affected by relative humidity, especially in females [84], as might be expected from their role in reducing water loss. Sex-specific differences occur in C. bezziana, Cochliomyia hominivorax and M. domestica [56, 84–86]. The effects of ageing therefore need to be distinguished from taxonomic, geographical, sexual and environmental causes. Similar patterns of variation related to species, sex and age have been found in sexton beetles (Silphidae) of the genus Nicrophorus [87] and the findings related to carrion flies probably apply to all of the carrion beetles too. In summary, cuticular hydrocarbons can provide information about the age of carrion insects during stages of development that are difficult to age by other means, and about the age of the remains of pupae and puparia that have eclosed. The matter is complicated by species identity, geographical variation, sex and temperature, but methods for taking these confounding variables into account already exist [56]. What remains is to develop adequate databases to implement the approach.

11  Advances in Entomological Methods for Death Time Estimation

223

Developmental Anatomy of Pupae Gene expression and cuticular hydrocarbon analysis are techniques that can be used to estimate the age of most stages of the insect life cycle. Some additional techniques are specifically applicable to pupae. The pupal stage of an insect’s life cycle (Fig. 11.2) is when metamorphosis occurs. Although few changes are visible externally, the insect undergoes dramatic internal reorganisation of tissues and organs, and this process offers a large suite of developmental landmarks. Dissection is generally not a practical approach to determining the progress of metamorphosis because some stages are composed largely of body fluid, and because one may want to keep the specimen alive. The medical methods of multi-slice computed tomography (MSCT), magnetic resonance imaging (MRI) and X-ray imaging were developed for non-invasive, high-resolution imaging of internal anatomy of live humans. This led to the development of virtopsy [88], non-invasive virtual autopsy, as complementary or alternative means of post-mortem examination. MSCT scans are usually performed with 1.2 mm thick slices and a reconstruction increment of 0.7 mm, but small features such as teeth and delicate fractures are scanned in slices 0.63 mm thick and reconstruction increments of 0.5 mm [88]. This level of spatial resolution may be adequate to identify gross anatomical structures inside fly puparia, such as the head, flight muscles and legs. Flight muscles are soft tissues that are better visualised by MRI, which resolves soft materials excellently, while the legs contain a large proportion of exoskeletal chitin that is better visualised by MSCT. MRI shows hard materials as artefact-producing gaps in the image, and MSCT has poorer contrast resolution in soft tissues [also see 89]. X-ray micrographs have been made of characteristic internal structures and their changes in various small insects like ants and aphids, and used to illustrate metamorphosis inside a puparia of fruit flies [90]. X-ray microtomography (microcomputed tomography or microCT), identical in its basic principles to medical CT scanning, has been increasingly used in non-clinical research in the last decade [91, 92] and microCT imaging is already being applied to quantitative studies of variation and development in insects [93]. The spatial resolution of this method is astonishing: the maximum resolution is presently ca. 1 mm [91]. One current drawback is that insect tissue must be fixed and stained (with, e.g. iodine, osmium or silver) to enhance the contrast between different tissues [91, 92], which kills the insect and makes rearing to adulthood impossible, undermining the value of its non-invasiveness. Socha et al. [94] used synchrotron X-ray phase contrast imaging to study the physiology and internal biomechanics in small animals like ants, beetles and fruitflies. According to these authors, this is the only generally applicable technique that has the necessary spatial and temporal resolutions, penetrating power, and sensitivity to soft tissue that is required to visualise the internal physiology of living animals on the scale from millimetres to microns. Because all these techniques need a complicated and often expensive setting, histology might be a practical alternative that has a long tradition in insect system-

224

M.H. Villet and J. Amendt

atics. Weismann [95] was the first to make detailed analyses of metamorphosis in higher Diptera and Pérez [96] described the histological changes during metamorphosis in Calliphora erythrocephala (now C. vicina) with particular emphasis on the role of phagocytes in the destruction of larval tissues. Numerous studies [e.g. 97, 98] have also improved our knowledge about morphological and histological aspects of dipteran metamorphosis and they are now used in forensic entomology, too [99, 100]. Microtome sectioning was greatly improved by the use of new embedding materials and modern diamond knives. Even strongly sclerotised insects can be sectioned at a thickness of 1.0–1.5  mm and artefacts are greatly reduced compared to paraffin wax sections. Linking this technique with computerbased three-dimensional reconstructions provides a qualitative improvement in the study of insect anatomy. An impressive example is a study [101] of the thoracic morphology of the first instar larva of a species of Strepsiptera, Mengenilla chobauti, which, with an average total length of 230 mm, is one of the smallest of insects, even smaller than many protozoans.

Estimating the Age of Adults In cases where a body has decomposed indoors, insects may complete their life cycle before the body is discovered (Fig.  11.1), and the resulting adults may be recovered from the scene. Their ages can be estimated using the general techniques of cuticular hydrocarbon analysis [56, 83], and perhaps gene expression analysis. Additional techniques specific to adults include the measurement of pteridine accumulation, near-infrared spectroscopy, wing wear and ovarian development (in females). Pteridine is a fluorescing pigment that accumulates in an adult fly’s head as it ages. As a metabolic by-product of purine degradation, the accumulation of pteridine occurs in physiological time and is therefore affected by temperature [102–105]. Its concentration can be measured as relative fluorescence using a spectrofluorimeter with an emission monochromator set at 450 nm and an excitation filter of 340 nm [106–108]. The value obtained has to be standardised to account for body size, because larger flies have larger heads, and therefore contain more pteridine at the same age, and the result is also influenced by sex, being more predictable in males (Table 11.2). The data do not have to be corrected for diet. The average precision of this method is about 2 or 3 days (Table 11.2), but better prediction might be obtained using one of the new generation of General Additive Models [cf. 109]. Estimating age using near-infrared spectroscopy (NIRS) was tested on houseflies (Musca domestica), face flies (Musca autumnalis) and stable flies (Stomoxys calcitrans) [104]. The technique measures the near-infrared reflectance of the adult’s cuticle, which changes with age. Compared to the pteridine accumulation technique, NIRS predicted ages well (r² = 0.78–0.84 in fresh or preserved houseflies), independent of size, sex or temperature, and with smaller confidence limits. It could be done on whole flies or their heads alone, and on fresh, dried or

11  Advances in Entomological Methods for Death Time Estimation Table 11.2  Relationship between pteridine accumulation and age of adult flies Coefficient of Average precision ­determination (r²) (days) Species Females Males Females Males L. sericata 0.910 0.945 C. megacephala 0.999 0.956 ±2.55 C. rufifacies 0.87 ±3.6 ±2.3

225

Sources [107] [141] [105]

ethanol-preserved specimens. The equipment is conveniently portable, and the specimens do not need the lengthy extraction procedure that pteridine requires [104]. The lack of an influence of temperature on NIRS makes it a very appealing method when insects are found at windows indoors, because estimating the temperature of these microenvironments is complicated if they receive direct sunlight. Wing wear can be calibrated against a known marker of age, e.g. pteridine accumulation, but this leaves the estimators inter-correlated and non-independent, which means that they cannot be used to cross-validate one another. It is also a very coarse index that is not independent of levels of activity (and hence temperature) or predation. We do not recommend it for forensic work. The development of the ovary, and specifically the length of the most advanced egg follicle, has been used to give an estimate of age [107, 108, 110]. Egg follicle length followed a sigmoidal relationship with age (r² = 0.815) in Lucilia sericata [107]. It was a useful predictor of age for the first 3 days of adulthood, was very variable on the 4th day, and settled at a mature length on the 5th and 6th days. There is a confounding effect from females’ body size (r² = 0.784), and the method is obviously not useable on males. In summary, NIRS seems ideal for estimating the age of adult flies, and validation and calibration studies should be done. Pteridine accumulation measurements are less convenient, but provide useful results, while ovarian development and wing wear are much more approximate sources of evidence.

Data Analysis Traditionally, forensic entomology has drawn on mathematical models of insect development that were formulated for agricultural entomology and modified to suit the needs of criminal investigations. This has covered a spectrum from relatively simple numerical models that can be implemented on paper [18] to sophisticated, cohortbased models that require a fair amount of computing power [111]. Graphical methods of estimating post-mortem intervals from larval growth (isomegalen diagrams) and developmental events like pupation or eclosion (isomorphen diagrams) were proposed by Reiter [112] and Grassberger and Reiter [62, 65–67]. These diagrams are easy to use when air temperatures are relatively constant, as may happen in indoor situations,

226

M.H. Villet and J. Amendt

and recently a refined method has been proposed for using them under conditions of fluctuating environmental temperatures [10]. VanLaerhoven [113] has shown that a simple linear regression model can be effective in providing credible estimates of PMImin, a crucial validation of the approach. At least part of the reason for this lies in the inherent variation characteristic of insect development [10], which limits the sample precision that is possible even under ideal conditions. The technique for calculating one of the simpler and most popular linear regression models has been revised to improve its robustness to underlying assumptions and to provide measures of confidence in the estimates [114]. However, insect development is not actually rectilinear, and violates several assumptions of traditional linear regression models [109, 115]. With the advent of increased readily available computing power and more sophisticated statistical models and modelling environments, it has become possible to construct a new generation of comprehensive, non-linear models of insect development. Ieno et al. [109] illustrated the implementation of a General Additive Model for the development of Calliphora vicina under various combinations of three drugs. Apart from showing the power of this approach, this exercise also illustrated the inherent limitations of using length as a proxy for age: the temporal resolution of estimates of time since death is excellent during the initial exponential phase of growth, but deteriorates sharply after the larvae wander from the carcass to pupate [109]. This ability to supply information about the relative reliability of the prediction is another strength of the approach, one of the four criteria of the Daubert standard for expert evidence [[6], (Daubert vs Merrell Dow Pharmaceuticals Inc (1992) 509 US 579)].

Estimating the Duration of Ecological Succession Changes in the structure of the insect community present on a corpse, referred to as ecological succession, also provide a clock for estimating longer PMIs [116–119]. This clock appears to be quantitatively predictable [17, 120] despite influences from season and habitat on community structure and duration of the succession process [120]. It is less precise than the clock provided by development because it involves more processes and occurs over longer time spans than development, and therefore has more time to accumulate complicating influences. Although the succession process is commonly described in terms of a variable number of stages, there is no statistical support for this [121].

Experimental Design The relevance of experimental studies of ecological succession have been challenged on several grounds, which has led to experimental verification of at least

11  Advances in Entomological Methods for Death Time Estimation

227

some of the assumptions of the approach. In particular, most experimental studies have used non-human remains as a substitute for humans because of legislative restrictions. Pigs are generally the model animal of choice, and Schoenly et al. [122] provided some experimental evidence that they were a statistically reasonable substitute. This experiment involved limited numbers of pigs and humans, but the outcome was encouraging. The study was conducted at the Anthropology Research Facility, Knoxville, Tennessee, a site where such research has been conducted intensively. Because of this, it could be argued that insect populations and communities there had become enriched and therefore unrepresentative, but this was shown empirically to be of little concern [123, 124]. Numerous other studies have offered baseline data on community composition on pig carcasses in various parts of the world, but these studies are still largely qualitative. Well-replicated studies have been published [e.g. 116, 117, 125–127], but in most studies few replicates have been used and variation within and between seasons in the succession pattern has not received sufficient attention. Ease of access to a body had an effect on the rate of colonisation and community composition of pigs in vehicles [16], and there are significant influences from season and habitat on community structure and duration of the succession process [120, 128, 129]. Studies of carrion of a variety of other vertebrates have been published, but their relevance to the succession pattern of human corpses has not been tested. Species, size, and mode of death of the carcass possibly influencing the blow fly composition and diversity. There is still much experimental work to be done on ecological patterns in decomposition, and Goff [130] has proposed a standard protocol for such studies.

Data Analysis Progress is being made on understanding the mechanisms underlying ecological succession [118, 119], which will increase our ability to model and predict the process. For instance, larder beetles have been found to accelerate decomposition of mummified corpses [131]. Warmer environmental temperatures hasten succession [17, 120], not only through accelerating metabolic processes in ectothermic necrophages and decomposers, but also by stimulating more species and specimens to be active, leading to more comprehensive colonisation of carrion [cf. 118]. Quantitative methods for retrodicting these changes have been proposed by Lamotte and Wells [132] and Michaud and Moreau [17], and are implicit in the work of Voss et  al. [120]. Schoenly et  al. [133] introduced an “occurrence matrix” method that notes the presence or absence of each species at each point in time after death [128, 129]. Lamotte and Wells [132] proposed two means (a likelihood ratio statistic and an unconditioned Fisher’s exact test) of calculating the probability of species co-occurring at a given stage in succession. Michaud and Moreau [17] measured time using a modified thermal accumulation method, and used it (fairly successfully: in three out of five cases, r² = 0.55–0.77) to model

228

M.H. Villet and J. Amendt

the probability of a species being present using logistic regression. They then used the successful models to derive odds-based estimates of the place in the succession from which a hypothetical sample might be collected. Voss et al. [120] showed that the community composition at different stages of the succession process and in different seasons can be ordinated, and that the first axis of the ordination summarised the changes during succession, while the second axis summarised the seasonal effect (apparently primarily due to temperature), and that there was a small amount of interaction between these sources of variation. This means that a new sample can be projected into this ordination space to estimate the stage of succession that it represents. By calibrating the ordination space using the known ages of the original samples used to construct it, one can estimate the PMI represented by the forensic sample. Estimates derived from ecological succession are recognised as being moderately coarse and in need of refinement to improve their precision [17, 118, 119]. One relevant modification of the approach is to move beyond simply scoring the presence (or absence) of species, and to score the presence of particular stages of the life cycles of each species because this improves the temporal resolution of the data. Generating more temporal precision would require changing from collecting categorical data (e.g. presence and absence), which is typical of traditional approaches to ecological succession in carrion, to collecting continuous data (e.g. age and abundance), which is typical of the following methods.

Framework for Standardising Methods Modern courts have become increasingly concerned with the admissibility and validity of expert evidence, especially in the USA, where the Frye (Frye vs. United States 293F. 1013 (D.C. Cir. 1923)), Daubert (Daubert vs Merrell Dow Pharmaceuticals Inc (1992) 509 US 579) and Khumo Tire (Kumho Tire Co. v. Carmichael, 119S.Ct. 1167, 1176 (1999)) judgements set criteria for expert evidence that emphasise the peer review, reliability and general acceptance of methods [6]. The Daubert judgement set four tests for scientific theories and techniques: whether they could be (and had been) tested, whether they have been subjected to peer review and publication, whether their known or potential rates of error have been quantified, and whether they had gained widespread acceptance within the scientific community. If the plethora of methods emerging in forensic entomology is to be useful, there is a need for standardisation to raise the validity of forensic entomology as a source of evidence [7]. The creation of such standards is a process that involves peer review in the scientific literature, consensus amongst professional practitioners and eventual formalisation through an agency. Explicit proposals of standards for forensic entomology have started to appear in the scientific literature part of the process of peer review that leads to consensus about good practice. An outline of standard methods for the collection and handling of most traditional sources of entomological evidence has been published by Amendt

11  Advances in Entomological Methods for Death Time Estimation

229

et al. [7, also see 3], and a revised standard protocol for the preservation of beetle larvae has been described by Midgley and Villet [134]. Explicit guidelines for the design of experiments to gather evidence of developmental biology have been discussed by Richards and Villet [22, 135], and Goff [130] has described a standard experimental protocol for ecological succession studies. The issues of consensus and how it is to be recognised lie in the near future of the discipline. Professional bodies such as the American Board of Forensic Entomology and the European Association for Forensic Entomology are obvious candidates to provide an arena for facilitating this step of the standardisation process, but several other relevant bodies exist, particularly in the USA [5, 6]. Finally, the formalisation of standards provides a means to help courts to decide the admissibility of exert evidence. This issue will be partially contingent on the particulars of national legal systems and the existence of relevant and competent authorities. Forensic entomology can also take advantage of existing formal standards developed for other forensic disciplines, such as standard operating procedures for forensic laboratories and equipment [e.g. 5, 136]. The International Organization for Standardization oversees the ISO/IEC 17025 standard, which is the main standard used by testing and calibration laboratories, and deals not only with technical requirements such as methods, calibration and testing of equipment, and competence of staff, but also management matters, including the operation and effectiveness of a laboratory quality management system. Accreditation under this standard is carried out through an accrediting body, and these tend to exist at a regional level (although the USA has several). The American Society of Crime Laboratory Directors/Laboratory Accreditation Board is an example relevant to forensic entomology [5]. The International Laboratory Accreditation Cooperation (ILAC) has formalised methods of evaluating accreditation bodies against another ISO standard (ISO/IEC 17011), providing a means for international benchmarking and sharing of standards. Apart from their deployment in juridical matters, standards also serve other functions, such as the certification of training, practitioners and facilities. There is a need for certification procedures such as the certification program of the American Board of Forensic Entomology, and for standardised criteria for evaluating the associated examinations. Similarly, accreditation of training courses may take into account quality standards relating to the methods covered in the course [6, 137].

Conclusion A variety of new methods for estimating time of death have been added to the repertoire of forensic entomologists in the last decade, and methods are also being developed in botany [138]. Some of the methods have been internally validated [113, 122], and further work is urgently needed in this direction to meet the demands that courts place on scientific evidence. In addition, the spectrum of methods provides a means of cross-validating estimates of time since death. There is a pressing need to set stan-

230

M.H. Villet and J. Amendt

dards for the validation and deployment of methods and for certification procedures. International bodies already exist to facilitate these steps in the maturation of forensic entomology into a robustly established branch for forensic science. The task facing forensic entomologists is therefore to publish rigorous studies that include quantified error rates, and to reach consensus about which methods will stand up in court. We recommend that two steps in the standardisation process for forensic entomology are that physiological and ecological studies should be reported in physiological time wherever it is appropriate, and that the post-mortem interval being estimated should be stipulated more explicitly than is currently common. Acknowledgements We thank E. Türk (University of Homburg/Saarland) for inviting us to write this review.

References 1. Smith KGV (1986) A manual of forensic entomology. London, British Museum (Natural History) 2. Catts EP, Haskell NH (1990) Entomology and death: a procedural guide. Clemson, South Carolina 3. Byrd JH, Castner JL (eds) (2010) Forensic entomology: the utility of arthropods in legal investigations. (2nd ed.) CRC Press, Boca Raton 4. Amendt J, Goff ML, Campobasso CP, Grassberger M (eds) (2010) Current trends in forensic entomology. Springer, Dordrecht 5. Lentini JJ (2009) Forensic science standards: where they come from and how they are used. Forens Sci Policy Mgmt 1:10–16 6. National Research Council Committee on Identifying the Needs of the Forensic Sciences Community [NRCCINFSC] 2009. Strengthening forensic science in the United States: a path forward. pp 1–328. The National Academies Press, Washington, DC 7. Amendt J, Campobasso CP, Gaudry E et al (2007) Best practice in forensic entomology – standards and guidelines. Int J Legal Med 121:90–104 8. Byrd JH, Lord WD, Wallace JR et al (2010). Collection of entomological evidence during legal investigations. In: Byrd JH, Castner JL (eds) Forensic entomology: the utility of arthropods in legal investigations (2nd ed.). CRC Press, Boca Raton 9. Merritt RW, Wallace JR (2010) The role of aquatic insects in forensic investigations. In: Byrd JH, Castner JL (eds) Forensic entomology: the utility of arthropods in legal investigations (2nd ed.). CRC Press, Boca Raton 10. Villet MH, Richards CS, Midgley JM (2010) Contemporary precision, bias and accuracy of minimum post-mortem intervals estimated using development of carrion-feeding insects. In: Amendt J, Campobasso CP, Goff ML et al (eds) Current concepts in forensic entomology. Springer, Dordrecht 11. Wells JD, Lamotte LR (2010) Estimating the postmortem interval. In: Byrd JH, Castner JL (eds) Forensic entomology: the utility of arthropods in legal investigations. CRC Press, Boca Raton 12. Benecke M (2010) Cases of neglect involving entomological evidence. In: Byrd JH, Castener JL (eds) Forensic entomology: the utility of arthropods in legal investigations. CRC Press, Boca Raton 13. Mahat NA, Zafarina Z, Jayaprakash PT (2009) Influence of rain and malathion on the oviposition and development of blowflies (Diptera: Calliphoridae) infesting rabbit carcasses in Kelantan, Malaysia. Forens Sci Int 192:19–28

11  Advances in Entomological Methods for Death Time Estimation

231

14. Voss SC, Spafford H, Dadour IR (2009) Host location and behavioural response patterns of the parasitoid, Tachinaephagus zealandicus Ashmead (Hymenoptera: Encyrtidae), to host and host-habitat odours. Ecol Ent 34:204–213 15. LeBlanc HN, Jogan JG (2010) Exploiting insect olfaction in forensic entomology. In: Amendt J, Campobasso CP, Goff ML et al (eds). Current concepts in forensic entomology. Springer, Dordrecht 16. Voss SC, Forbes SL, Dadour IR (2008) Decomposition and insect succession on cadavers inside a vehicle environment. Forens Med Sci Path 4:22–32 17. Michaud JP, Moreau G (2009) Predicting the visitation of carcasses by carrion-related insects under different rates of degree-day accumulation. Forens Sci Int 185:78–83 18. Higley LG, Haskell NH (2000) Insect development and forensic entomology. In: Byrd JH, Castner JL (eds) Forensic entomology: the utility of arthropods in legal investigations. CRC Press, Boca Raton 19. Sharpe PJH, Demichele DW (1977) Reaction kinetics of poikilotherm development. J Theor Biol 64:649–670 20. Wagner TL, Wu H, Sharpe PJH et al (1984) Modeling insect development rates: a literature review and application of a biophysical model. Ann Entomol Soc Am 77:208–225 21. Wagner TL, Wu H, Sharpe PJH et al (1984) Modeling distributions of insect development time: a literature review and application of the Weibull function. Ann Entomol Soc Am 77:475–483 22. Richards CS, Villet MH (2008) Factors affecting accuracy and precision of thermal summation models of insect development used to estimate postmortem intervals. Int J Legal Med 122:401–408 23. Richards CS, Paterson ID, Villet MH (2008) Estimating the age of immature Chrysomya albiceps (Diptera: Calliphoridae), correcting for temperature and geographical latitude. Int J Legal Med 122:271–279 24. Hwang CC, Turner BD (2009) Small-scaled geographical variation in life-history traits of the blowfly Calliphora vicina between rural and urban populations. Entomol Exp Appl 132: 218–224 25. Gallagher MB, Sandhu S, Kimsey R (2010) Variation in developmental time for geographically distinct populations of the common Green Bottle Fly, Lucilia sericata. J Forens Sci 55:438–442 26. Richards CS, Crous KL, Villet MH (2009) Models of development for the blow fly sister species Chrysomya chloropyga and C. putoria. Med Vet Entomol 23:56–61 27. Richards CS, Price BW, Villet MH (2009) Thermal ecophysiology of seven carrion-feeding blowflies in Southern Africa. Ent Exp App 131:11–19 28. Richards CS, Williams KA, Villet MH (2009) Predicting geographic distribution of seven blowfly species (Diptera: Calliphoridae) in South Africa. Afr Entomol 17:170–182 29. VanLaerhoven SL, Anderson GS (2001) Implications of using development rates of blow fly (Diptera:Calliphoridae) eggs to determine postmortem interval. J Entomol Soc B C98: 189–194 30. Bourel B, Callet B, Hédouin V et al (2003) Flies eggs: a new method for the estimation of short-term post-mortem interval? Forens Sci Int 135:27–34 31. Anderson GS (2004) Determining time of death using blow fly eggs in the early postmortem interval. Int J Legal Med 118:240–241 32. Sukontason K, Sukontason KL, Piangjai S et al (2004) Identification of forensically important fly eggs using a potasium permanganate staining technique. Micron 35:391–395 33. Sukontason K, Sukontason KL, Boonchu N et  al (2004) Ultrastructure of eggshell of Chrysomya nigripes Aubertin (Diptera: Calliphoridae). Parasitol Res 93:151–154 34. Sukontason K, Sukontason KL, Piangjai S et al (2004) Fine structure of the eggs of blowflies Aldrichina grahami and Chrysomya pacifica (Diptera: Calliphoridae). Biol Res 37: 483–487 35. Sukontason KL, Sukontason K, Vogtsberger RC et al (2005) Ultramorphology of eggshell of flesh fly Liosarcophaga dux (Diptera: Sarcophagidae). J Med Entomol 42:86–88

232

M.H. Villet and J. Amendt

36. Szpila K, Pape T, Rusinek A (2008) Morphology of the first instar of Calliphora vicina, Phormia regina and Lucilia illustris (Diptera, Calliphoridae). Med Vet Entomol 22:16–25 37. Szpila K (2010) Key for the identification of third instars of European blowflies (Diptera: Calliphoridae) of forensic importance. In: Amendt J, Campobasso CP, Goff ML et al (eds) Current concepts in forensic entomology. Springer, Dordrecht 38. Sperling FAH, Anderson GS, Hickley DA (1994) A DNA-based approach to the identification of insect species used for postmortem interval estimation. J Forens Sci 39:418–427 39. Stevens JR, Wall R (1997) Genetic variation in populations of the blowflies Lucilia cuprina and Lucilia sericata (Diptera: Calliphoridae). Random amplified polymorphic DNA analysis and mitochondrial DNA sequences. Biochem Syst Ecol 25:81–97 40. Wells JD, Sperling FAH (2001) DNA-based identification of forensically important Chrysomyinae (Diptera: Calliphoridae). Forens Sci Int 120:110–115 41. Wells JD, Pape T, Sperling FAH (2001) DNA-based identification and molecular systematics of forensically important Sarcophagidae (Diptera). Forens Sci Int 46:87–91 42. Harvey ML, Mansell MW, Villet MH, Dadour IR (2003) Molecular identification of some forensically important blowflies of southern Africa and Australia. Med Vet Entomol 17:363–369 43. Harvey ML, Gaudieri S, Villet MH, Dadour IR (2008) A global study of forensically significant calliphorids: implications for identification. Forens Sci Int 177:66–76 44. Zehner R, Amendt J, Schütt S, Sauer J, Krettek R, Povolný D (2004) Genetic identification of forensically important flesh flies (Diptera: Sarcophagidae). Int J Legal Med 118:245–247 45. Boehme P, Amendt J, Disney RHL et al (2010) Molecular identification of carrion-breeding scuttle flies (Diptera: Phoridae) using COI barcodes. Int J Legal Med. doi:10.1007/s00414010-0429-5 46. Ratcliffe S, Webb D, Weinzierl R, Robertson H (2003) PCR-RFLP identification of Diptera (Calliphoridae, Muscidae and Sarcophagidae) – a generally applicable method. J Forens Sci 48:1–3 47. Torres TT, Azerdo-Espin AML (2005) Development of new polymorphic microsatellite markers for the New World screw-worm Cochliomyia hominivorax (Diptera: Calliphoridae). Mol Ecol Notes 5:815–817 48. Picard CJ, Wells JD (2009) Survey of the genetic diversity of Phormia regina (Diptera: Calliphoridae) using amplified fragment length polymorphisms. J Med Entomol 46:664–670 49. Wells JD, Stevens JR (2008) Application of DNA-based methods in forensic entomology. Ann Rev Entomol 53:103–120 50. Bagnères AG, Wicker-Thomas C (2010) Chemical taxonomy with hydrocarbons. In: Blomquist GJ, Bagnères AG (eds) Insect hydrocarbons: biology, biochemistry and chemical ecology. Cambridge University Press, Cambridge 51. Stoffolano JG, Schauber E, Yin CM et al (1997) Cuticular hydrocarbons and their role in copulatory behavior in Phormia regina (Meigen). J Insect Physiol 43:1065–1076 52. Drijfhout FP (2010) Cuticular hydrocarbons: a new tool in forensic entomology? In: Amendt J, Campobasso CP, Goff ML et al (eds). Current concepts in forensic entomology. Springer, Dordrecht 53. Blomquist GJ, Jackson LL (1973) Hydroxylation of n-alkanes to secondary alcohols and their esterification in the grasshopper Melanoplus sanguinipes. Biochem Biophys Res Commun 53:703–708 54. Ye G, Li K, Zhu et al (2007) Cuticular hydrocarbon composition in pupal exuviae for taxonomic differentiation of six necrophagous flies. J Med Entomol 44:450–456 55. Byrne AL, Camann MA, Cyr TL et al (1995) Forensic implications of biochemical differences among geographic populations of the black blow fly, Phormia regina (Meigen). J Forens Sci 40:372–377 56. Brown WV, Morton R, Lacey MJ et al (1998) Identification of the geographical source of adults of the old world screw-worm fly, Chrysomya bezziana Villeneuve (Diptera:Calliphoridae), by multivariate analysis of cuticular hydrocarbons. Comp Biochem Physiol B119:391–399

11  Advances in Entomological Methods for Death Time Estimation

233

57. Roux O, Gers C, Legal L (2006) When, during ontogeny, waxes in the blowfly (Calliphoridae) cuticle can act as phylogenetic markers. Biochem Syst Ecol 34:406–416 58. Midgley JM, Villet MH (2009) Development of Thanatophilus micans (Fabricius 1794) (Coleoptera:Silphidae) at constant temperatures. Int J Legal Med 23:285–292 59. Voss SC, Spafford H, Dadour IR (2010) Temperature-dependant development of Nasonia vitripennis on five forensically important carrion fly species. Entomol Exp Appl 135:37–47 60. Voss SC, Spafford H, Dadour IR (2010) Temperature-dependent development of the parasitoid Tachinaephagus zealandicus on five forensically important carrion fly species. Med Vet Entomol 24:189–198 61. Geden CJ (1997) Development models for the filth fly parasitoids Spalangia gemina, S. cameroni, and Muscidifurax raptor (Hymenoptera:Pteromalidae) under constant and variable temperatures. Biol Control 9:185–192 62. Grassberger M, Frank C (2003) Temperature-related development of the parasitoid wasp Nasoniavitripennis as forensic indicator. Med Vet Entomol 17:257–262 63. De Carvalho AR, Pinto de Mello R, D’Almeida JM (2003) Microhimenópteros parasitóides de Chrysomya megacephala. Rev Saude Publica 37:810–812 64. Disney RHL, Munk T (2004) Potential use of Braconidae (Hymenoptera) in forensic cases. Med Vet Entomol 18:442–444 65. Grassberger M, Reiter C (2001) Effect of temperature on Lucilia sericata (Diptera: Calliphoridae) development with special reference to the isomegalen- and isomorphendiagram. Forens Sci Int 120:32–36 66. Grassberger M, Reiter C (2002) Effect of temperature on development of the forensically important Holarctic blow fly Protophormia terraenovae (Robineau-Desvoidy) (Diptera: Calliphoridae). Forens Sci Int 128:177–182 67. Grassberger M, Reiter C (2002) Effects of temperature on development of Liopygia (=Sarcophaga) argyrostoma (Robineau-Desvoidy) (Diptera: Sarcophagidae) and its forensic implications. J Forens Sci 47:1332–1336 68. White KP, Rifkin SA, Hurban P et al (1999) Microarray analysis of Drosophila development during metamorphosis. Science 286:2179–2184 69. Arbeitman MN, Furlong EEM, Imam F et al (2002) Gene expression during the life cycle of Drosophila melanogaster. Science 297:2270–2275 70. Sullivan AA, Thummel CS (2003) Temporal profiles of nuclear receptor gene expression reveal coordinate transcriptional responses during Drosophila development. Mol Endocrinol 17:2125–2137 71. Cook PE, Hugo LE, Iturbe-Ormaetxe I (2006) The use of transcriptional profiles to predict adult mosquito age under field conditions. PNAS 103:18060–18065 72. Marinotti O, Calvo E, Nguyen QK et al (2006) Genome-wide analysis of gene expression in adult Anopheles gambiae. Insect Mol Biol 15:1–12 73. Ali RA, Mellenthin K, Fahmy K et al (2005) Structural conservation of the salivary glandspecific slalom gene in the blowfly Lucilia sericata. Dev Genes Evol 215:537–543 74. Tarone AM, Jennings KC, Foran DR (2007) Aging blow fly eggs using gene expression: a feasibility study. J Forens Sci 52:1350–1354 75. Zehner R, Boehme P, Amendt J (2009) Gene expression analysis as a tool for age estimation of blowfly pupae. Forens Sci Int: Genet Suppl Series 2(2009):292–293 76. Gaudry E, Blais C, Maria AA et al (2006) Study of steroidogenesis in pupae of the forensically important blow fly Protophormia terraenovae (Robineau-Desvoidy) (Diptera: Calliphoridae). Forens Sci Int 160:27–34 77. Zhu GH, Ye GY, Hu C et  al (2006) Development changes of cuticular hydrocarbons in Chrysomya rufifacies larvae:potential for determining larval age. Med Vet Entomol 20:438–444 78. Roux O, Gers C, Legal L (2008) Ontogenetic study of three Calliphoridae of forensic importance through cuticular hydrocarbon analysis. Med Vet Entomol 22:309–317 79. Goodrich BS (1970) Cuticular lipids of adults and puparia of the Australian sheep blowfly Lucilia cuprina (Wied.). J Lipid Res 11:1–6

234

M.H. Villet and J. Amendt

80. Arnold MT, Regnier FE (1975) A developmental study of the cuticular hydrocarbons of Sarcophaga bullata. J Insect Physiol 21:1827–1833 81. Whitaker AP (2010) How long do empty puparia survive in woodland soil? In: Abstracts of the 8th meeting of the European Association for Forensic Entomology, University of Murcia/ Spain, 8–11 September 2010 82. Zhu GH, Xu XH, Yu XJ et al (2007a) Puparial case hydrocarbons of Chrysomya megacephala as an indicator of the postmortem interval. Forens Sci Int 169:1–5 83. Trabalon M, Campan M, Clement JL et  al (1992) Cuticular hydrocarbons of Calliphora vomitoria (Diptera): relation to age and sex. Gen Comp Endocrin 85:208–216 84. Noorman N, Den Otter CJ (2002) Effects of relative humidity, temperature, and population density on production of cuticular hydrocarbons in housefly Musca domestica L. J Chem Ecol 28:1819–1829 85. Pomonis JG (1989) Cuticular hydrocarbons of the screwworm, Cochliomyia hominivorax (Diptera:Calliphoridae). Isolation, identification, and quantification as a function of age, sex, and irradiation. J Chem Ecol 15:2301–2317 86. Mpuru S, Blomquist GJ, Schal C et al (2001) Effect of age and sex on the production of internal and external hydrocarbons and pheromones in the housefly, Musca domestica. Insect Biochem Mol Biol 31:139–155 87. Whitlow S (2003) Recognition in burying beetles (Nicrophorusspp., Silphidae, Coleoptera). Dissertation, Albert-Ludwigs-Universität, Freiburg 88. Bolliger SA, Thalli MJ, Ross S et al (2008) Virtual autopsy using imaging: bridging radiologic and forensic sciences. A review of the Virtopsy and similar projects. Europ Radiol 18:273–282 89. Hart AG, Bowtell RW, Kockenberger W, Wenseleers T, Ratnieks FLW (2003) Magnetic resonance imaging in entomology: a critical review. J Insect Sci 3:5 90. Yoshimura H, Shoutsu D, Horikoshi T et  al (2000) Application of SEM-modified X-ray microscope to entomology and histology, and effects of X-ray coherence in imaging. J Electron Microsc 49:621–628 91. Metscher BD (2009) MicroCT for comparative morphology: simple staining methods allow high-contrast 3D imaging of diverse non-mineralized animal Tissues. BMC Physiol 9:11. doi:10.1186/1472-6793-9-11 92. Metscher BD (2009) MicroCT for developmental biology: a versatile tool for high-contrast 3d imagingat histological resolutions. Dev Dyn 238:632–640 93. Hörnschemeyer T, Beutel RG, Pasop F (2002) Head structures of Priacma serrata LeConte (Coleoptera, Archostemata) inferred from X-ray tomography. J Morphol 252:298–314 94. Socha JJ, Westneat MW, Harrison JF et al (2007) Real-time phase-contrast X-ray imaging: a new technique for the study of animal form and function. BMC Biol 5:6. doi:10.1186/1741-7007-5-6 95. Weismann A (1864) Die nachembryonale Entwicklung der Musciden nach Bebobachungen an Musca vomitoria und Sarcophaga carnaria. Z Wiss Zool 14:187–336 96. Pérez C (1910) Recherches histologique sur la metamorphose des muscides (Calliphora erythrocephala Mg.). Arch Zool Exp Gen 4:1–274 97. Brainridge SP, Bownes M (1981) Staging the metamorphosis of Drosophila melanogaster. J Embryol Exp Morphol 66:57–80 98. Sivasubramaniam P, Biagi M (1983) Morphology of the pupal stages of the fleshfly, Sarcophaga bullata (Parker) (Diptera: Sarcophagidae). Int J Insect Morph Embryol 12:355–359 99. Davies K, Harvey M (2010) Preservation of blowfly pupae for age and post mortem interval estimation. In: Abstracts of the 8th meeting of the European Association for Forensic Entomology, University of Murcia/Spain, 8–11 September 2010 100. Defilippo F, Bonilauri P, Massirio I et al (2010) Development of the pupal stages of the blowfly Calliphora vicina. In: Abstracts of the 8th meeting of the European Association for Forensic Entomology, University of Murcia/Spain, 8–11 September 2010 101. Osswald J, Pohl H, Beutel RG (2010) Extremely miniaturised and highly complex: The thoracic morphology of the first instar larva of Mengenilla chobauti (Insecta, Strepsiptera). Arthropod Struct Dev 39:287–304

11  Advances in Entomological Methods for Death Time Estimation

235

102. Lehane MJ, Chadwick J, Howe MA et al (1986) Improvements in the pteridine method for determining age in adult male and female Stomoxys calcitrans (Diptera: Muscidae). J Econ Entomol 79:1714–1719 103. Thomas DB, Chen AC (1989) Age determination in the adult screwworm (Diptera: Calliphoridae) by pteridine levels. J Econ Entomol 82:1140–1144 104. Perez-Mendoza J, Dowell FE, Broce AB et al (2002) Chronological age-grading of house flies by using near-infrared spectroscopy. J Med Entomol 39:499–508 105. Zhu GH, Ye GY, Hu C et al (2007) Determining the age of adult Achoetandrus rufifacies by pteridine fluorescence analysis. J Shantou Univ Med College 2007-02 106. Wall R, Langley PA, Stevens J et al (1990) Age determination in the old-world screw-worm fly Chrysomya bezziana by pteridine fluorescence. J Insect Physiol 36:213–218 107. Wall R, Langley PA, Morgan KL (1991) Ovarian development and pteridine accumulation for age determination in the blowfly Lucilia sericata. J Ins Physiol 37:863–868 108. Hayes EJ, Wall R, Smith KE (1998) Measurement of age and population age structure in the blowfly, Lucilia sericata (Meigen) (Diptera: Calliphoridae). J Insect Physiol 44:895–901 109. Ieno EN, Amendt J, Fremdt H et al (2010) Analysing forensic entomology data using additive mixed effects modelling. In: Amendt J, Campobasso CP, Goff ML et  al (eds) Current concepts in forensic entomology. Springer, Dordrecht 110. Linhares AX, Avancini RP (1989) Ovarian development in the blowflies Chrysomya putoria and C. megacephala on natural diets. Med Vet Entomol 3:293–295 111. Byrd JH, Allen JC (2001) Computer modeling of insect growth and its application to forensic entomology. In: Byrd JH, Castner JL (eds) Forensic entomology: the utility of arthropods in legal investigations. CRC Press, Boca Raton 112. Reiter C (1984) Zum Wachstumsverhalten der Maden der blauen Schmeißfliege Calliphoravicina. Z Rechtsmed 91:295–308 113. VanLaerhoven SL (2008) Blind validation of postmortem interval estimates using developmental rates of blow flies. Forens Sci Int 180:76–80 114. Ikemoto T, Takai K (2000) A new linearised formula for the law of total effective temperature and the evaluation of line-fitting methods with both variables subject to error. Environ Entomol 29:671–682 115. Tarone AM, Foran DR (2006) Components of developmental plasticity in a Michigan population of Lucilia sericata (Diptera Calliphoridae). J Med Entomol 43:1023–1033 116. Hobischak NR, Vanlaerhoven SL, Anderson GS (2006) Successional patterns of diversity in insect fauna on carrion in sun and shade in the Boreal Forest Region of Canada, near Edmonton, Alberta. Can Entomol 138:376–383 117. Matuszewski S, Bajerlein D, Konwerski S et al (2008) An initial study of insect succession and carrion decomposition in various forest habitats of Central Europe. Forens Sci Int 180:61–69 118. VanLaerhoven SL (2010) Ecological theory and its application in forensic entomology. In: Byrd JH, Castner JL (eds) Forensic entomology: the utility of arthropods in legal investigations. CRC Press, Boca Raton 119. Villet MH (in press) African carrion ecosystems and their insect communities in relation to forensic entomology. Func Ecosyst Commun (special issue) 120. Voss SC, Spafford H, Dadour IR (2009) Annual and seasonal patterns of insect succession on decomposing remains at two locations in Western Australia. Forens Sci Int 193:26–36 121. Schoenly K, Reid W (1987) Dynamics of heterotrophic succession in carrion arthropod assemblages: discrete series or a continuum of change? Oecologia 73:192–202 122. Schoenly KG, Haskell NH, Hall RD (2007) Comparative performance and complementarity of four sampling methods and arthropod preference tests from human and porcine remains at the Forensic Anthropology Center in Knoxville, Tennessee. J Med Entomol 44:881–894 123. Shahid SA, Schoenly KG, Haskell NH et al (2003) Carcass enrichment does not alter decay rates or arthropod community structure:a test of the arthropod saturation hypothesis at the Anthropology Research Facility in Knoxville, Tennessee. J Med Entomol 40:559–569

236

M.H. Villet and J. Amendt

124. Schoenly KG, Shahid SA, Haskell NH et al (2005) Does carcass enrichment alter community structure of predaceous and parasitic arthropods? A second test of the arthropod saturation hypothesis at the Anthropology Research Facility in Knoxville, Tennessee. J Forens Sci 50:134–141 125. VanLaerhoven SL, Anderson GS (1999) Insect succession on buried carrion in two biogeoclimatic zones of British Columbia. J Forens Sci 44:31–41 126. Matuszewski S, Bajerlein D, Konwerski S et al (2010) Insect succession and carrion decomposition in selected forests of Central Europe. Part 1: Pattern and rate of decomposition. Forens Sci Int 194:85–93 127. Matuszewski S, Bajerlein D, Konwerski S et al (2010) Insect succession and carrion decomposition in selected forests of Central Europe. Part 2: Composition and residency patterns of carrion fauna. Forens Sci Int 195:42–51 128. Tabor KL, Brewester CC, Fell RD (2004) Analysis of the successional patterns of insects on carrion in southwest Virginia. J Med Entomol 41:785–795 129. Sharanowski BJ, Walker EG, Anderson GS (2008) Insect succession and decomposition patterns on shaded and sunlit carrion in Saskatchewan in three different seasons. Forens Sci Int 179:219–240 130. Goff ML (2010) Early post-mortem changes and stages of decomposition. In: Amendt J, Campobasso CP, Goff ML et  al (eds). Current concepts in forensic entomology. Springer, Dordrecht 131. Schröder H, Klotzbach H, Oesterhelweg L et  al (2002) Larder beetles (Coleoptera, Dermestidae) as an accelerating factor for decomposition of a human corpse. Forens Sci Int 127:231–236 132. Lamotte LR, Wells JD (2000) p-Values for postmortem intervals from arthropod succession data. J Agric Biol Environ Stat 5:58–68 133. Schoenly K (1992) A statistical analysis of succesional patterns in carrion-arthropod assemblages: implications for forensic entomology and determination of the post-mortem interval. J Forens Sci 37:1489–1513 134. Midgley JM, Villet MH (2009b) Effect of the killing method on post-mortem change in length of larvae of Thanatophilus micans (Fabricius, 1794) (Coleoptera:Silphidae) stored in 70% ethanol. Int J Legal Med 123:103–108 135. Richards CS, Villet MH (2009) Data quality in thermal summation models of development of forensically important blowflies (Diptera:Calliphoridae):a case study. Med Vet Entomol 23:269–276 136. Penders J, Verstraete A (2006) Laboratory guidelines and standards in clinical and forensic toxicology. Accred Qual Assur 11:284–290 137. Jackson GP (2009) The status of forensic science degree programs in the United States. Forensic Sci Policy Manag 1:2–9 138. Cardoso HFV, Santos A, Dias R et al (2009) Establishing a minimum postmortem interval of human remains in an advanced state of skeletonization using the growth rate of bryophytes and plant roots. Int J Legal Med. doi:10.1007/s00414-009-0372-5 139. Kaneshrajah G, Turner BD (2004) Calliphora vicina larvae grow at different rates on different body tissues. Int J Legal Med 118:242–244 140. Ireland S, Turner BD (2006) The effects of larval crowding and food type on the size and development of the blowfly, Calliphora vomitoria. Forens Sci Int 159:175–181 141. Zhu GH, Ye G, Hu C (2003) Determining the adult age of the oriental latrine fly, Chrysomya megacephala (Fabricius) (Diptera: Calliphoridae) by pteridine fluorescence analysis. Insect Sci 10:245–255 142. Erzinçlioglu YZ (1990) On the interpretation of maggot evidence in forensic cases. Med Sci Law 30:65–66 143. Wells JD, King J (2001) Incidence of precocious egg development in flies of forensic importance (Calliphoridae). Pan-Pac Entomol 77:235–239 144. Saunders DS (1972) Circadian control of larval growth rate in Sarcophaga argyrostoma. Proc Natl Acad Sci USA 69:2738–2740

11  Advances in Entomological Methods for Death Time Estimation

237

145. Joplin KL, Moore D (1999) Effects of environmental factors on circadian activity in the flesh fly, Sarcophaga crassipalpis. Physiol Entomol 24:64–71 146. Kocárek P (2001) Diurnal patterns of postfeeding larval dispersal in carrion blowflies (Diptera: Calliphoridae). European J Entomol 98:117–119 147. Nabity PD, Higley LG, Heng-Moss TM (2007) Light-induced variability in the development of the forensically important blow fly, Phormia regina (Meigen) (Diptera: Calliphoridae). J Med Entomol 44:351–358 148. Tantawi TI, Greenberg B (1993) The effect of killing and preservative solutions on estimates of maggot age in forensic cases. J Forens Sci 38:702–707 149. Adams ZJO, Hall MJR (2003) Methods used for the killing and preservation of blowfly larvae, and their effect on post-mortem larval length. Forens Sci Int 138:50–61 150. Introna F, Campobasso CP, Goff ML (2001) Entomotoxicology. Forens Sci Int 120:42–47 151. Lopes de Carvalho LM (2010) Toxicology and forensic entomology.In: Amendt J, Campobasso CP, Goff ML, Grassberger M, eds. Current concepts in forensic entomology. Heidelberg: Springer, 163–178 152. Slone DH, Gruner SV (2007) Thermoregulation in larval aggregations of carrion-feeding blow flies (Diptera: Calliphoridae). J Med Entomol 44:516–523 153. Shiao SF, Yeh TC (2008) Larval competition of Chrysomya megacephala and Chrysomya rufifacies (Diptera: Calliphoridae): behavior and ecological studies of two blow fly species of forensic significance. J Med Entomol 45:785–799 154. Arnott S, Turner BD (2008) Post-feeding larval behaviour in the blowfly, Calliphora vicina: effects on post-mortem interval estimates. Forens Sci Int 177:162–167 155. Myskowiak JB, Doums C (2002) Effects of refrigeration on the biometry and development of Protophormia terraenovae (Robineau-Desvoidy) (Diptera: Calliphoridae) and its consequences in estimating post-mortem interval in forensic investigations. Forens Sci Int 125:254–261 156. Ames C, Turner BD (2003) Low temperature episodes in development of blowflies: implications for postmortem interval estimation. Med Vet Ent 17:178–186 157. Huntington TE, Higley LG, Baxendale FP (2007) Maggot development during morgue storage and its effect on estimating the Post-Mortem Interval. J Forens Sci 52:453–458 158. Saunders DS, Cymborowski B (2003) Selection for high diapause incidence in blow flies (Calliphora vicina) maintained under long days increases the maternal critical daylength: some consequences for the photoperiodic clock. J Ins Physiol 49:777–784

wwwwwwwwwwwwwwwww

Chapter 12

Tissue Fluorescence Spectroscopy in Death Time Estimation Éverton S. Estracanholli, Cristina Kurachi, and Vanderlei S. Bagnato

Abstract  One of the limitations of conventional methods for PMI determination is the fact that the measurements cannot be performed in situ nor in real time. Several factors, environmental and body, influence tissue decomposition and time evolution, resulting in a actual poor resolution. Considering this limitation for PMI determination, our group proposes a more objective method based on a tissue characterization of the degradation phases through optical information using fluorescence spectroscopy. The results obtained so far show the potential of the fluorescence spectroscopy for the PMI determination, with at least similar resolution of the actual methods. A portable interrogation system allows its use in outdoor areas with no need of laboratory supplies. Keywords  Postmortem interval • Death time estimation • Fluorescence spectroscopy

Introduction The great majority of the chronognosis methods for postmortem estimation are based on the detection of the phenomena taken place at the cadaver. The main limitation of these methods is that postmortem phenomena do not always take place at the same time interval due to several environmental and individual cadaver factors that influence their time course. There are changes, destructive and even conservative events that may accelerate, slow down, or suppress some stages of tissue modification. These events may take place or not depending on a large set of variables

É.S. Estracanholli (*) Instituto de Física de São Carlos, Universidade de São Paulo, Av. Trabalhador São-Carlense 400, São Carlos, SP 13566-590, Brazil e-mail: [email protected] E.E. Turk (ed.), Forensic Pathology Reviews, Volume 6, Forensic Pathology Reviews 6, DOI 10.1007/978-1-61779-249-6_12, © Springer Science+Business Media, LLC 2011

239

240

É.S. Estracanholli et al.

composed by cadaver characteristics and the environment conditions to which the body was subjected. In this way, most of the used methods for postmortem interval (PMI) determination can only provide an estimated value, since there is great variability in time course of the cadaveric phenomena, especially for the methods based on visual characterization. Considering this limitation for PMI determination, our group proposes a more objective method based on a tissue characterization of the degradation phases through optical information using fluorescence spectroscopy [1]. If proven sensitive enough, this method has as main advantages when comparing to conventional ­methods: a less interobserver variance and quantitative tissue information. These characteristics are relevant because they are less influenced by individual skills. Another attractive feature of optical technologies is the fact that in situ information is achieved through a noninvasive and nondestructive interrogation with fast response. Conventional laboratorial techniques for PMI determination available are time demanding and also require cadaver removal from the location it was found to a forensic lab facility. This operation is already suitable to introduce extra variance on the determination. Several equipments as nuclear magnetic resonance (NMR), electrophoresis, chromatography techniques and X-ray diffraction analyses are of high cost, resulting in few specialized centers [2–14]. Lab exams may take hours for its completion and especially for crime investigations, time is a relevant issue to be considered. Fluorescence spectroscopy has been presented as a sensitive technique to biochemical and structural changes of tissues. Examples of application for diagnostics purpose are detection of cancer [15–17], dental caries [18–20], and canker disease at orange tree leaves [21]. Tissue changes begin to take place at the cadaver as soon as there is cessation of life, optical characteristics changes, and they may be detected by fluorescence spectroscopy. There is a potential correlation of the tissue fluorescence changes and the PMI, even though the same limitations concerning the time course variability of the cadaveric phenomena are also present, we believe the ­optical spectra information can provide a more objective estimation. With the cessation of the metabolic reactions, tissue modifications are induced by several distinct factors as lack of oxygen and adenosine triphosphate, and intestinal microorganism proliferation. In our study, we first aimed to establish a proof of concept that fluorescence spectral variations for distinct PMIs are higher than the variance observed within each PMI. If this is positive, a spectral time behavior could be determined, i.e. fluorescence spectral changes identifying each PMI. We propose a method that has the potential to determine an unknown PMI based on a comparative analysis of a pattern spectral database. The organ that we chose for the initial tests was the skin, basically because of its direct access, but factors influencing its spectral response include tissue dryness, contact with environment contaminants, and diffusion of degradation products of internal organs, resulting in a fluorescence spectrum that, it is from a skin interrogation site, but also contains nondermal optical information. Other variations within skin investigation is the anatomical site, hypostatis due to cadaver positioning

12  Tissue Fluorescence Spectroscopy in Death Time Estimation

241

p­ resents distinct optical characteristics and evolution course compared to other skin sites. We believe that skin can be the site of choice when considering a feasible application for onsite PMI interrogation. There are several factors that still must to be investigated and correlated to PMI, but we here present some of our data to show the potential of fluorescence spectroscopy for PMI determination. Taking into account the resolution limitation for PMI determination in biological tissues, where the degradation process is nonhomogenous and influenced by environment and cadaver intrinsic factors, we believe the optical techniques may show better PMI prediction when compared to the actual techniques. In this way, we oversee a portable equipment with a dedicated software that can be taken to the site where the cadaver was found and after a fast interrogation and data analysis, a predicted PMI is given.

Fluorescence Spectroscopy Fluorescence spectroscopy is a good option if an objective and fast response is desired. A brief introduction to the basic principles of this technique is presented. According to quantum mechanics, an electron orbits around an atom or a molecule nucleus and can present only some specific energy amounts. In this sense, the orbit radius are not present freely in all distance from the nucleus. Energy amounts are well defined, representing the electronic levels. The allowed energy sets are determined by the potential which the electron is subjected to, i.e., the nucleus characteristics around which the electron is orbitating and its interaction with the other electrons. Each chemical element is characterized by its nucleus composition (­proton number), allowing a characterization of the possible energy levels that the electron can occupy. When light is shine on the surface and light waves propagate through the material distinct events can take place, one of great interest for light/tissue interactions is absorption. Absorption occurs when the incident energy is of the same amount of the difference between the lower electronic level and one higher, in this case the atom or the molecule absorbs this energy and one electron goes to a higher electronic level. The higher electronic levels are instable and the electron usually returns to the lower more stable level. When the electron decays to a lower state, the emission of a photon with the same excitation energy can occur. These energy levels may exist in two states, singlet (S = 0) or triplet (S = 1) state, depending on the spin direction of a pair of electrons. In the singlet state, the spins of the electrons are of opposite orientations. On the other hand, a triplet state exists when the two electrons show the same orientation [22]. The electrons that are at excited states, i.e., higher electronic levels, can only release part of the energy through nonradiative processes, as transferring energy to vibrational modes of the molecule, and so this electron goes to a lower energy level but still higher than the lowest ground state. When electronic transitions result in emission of light, this process is called fluorescence, that occurs only with electrons of same spin

242

É.S. Estracanholli et al.

­ ultiplicity. This transition is allowed by quantum mechanics and present a typical m rate of 108 s−1, resulting in a lifetime (time the electron stays in the excited state before a photon is emitted by the molecule) of approximately 10−8 s. The transitions between states with distinct spin multiplicities, usually from an excited triplet state to the singlet ground state, results in a process called phosphorescence. Phosphorescence shows much less probability to occur a longer lifetime, varying from milliseconds to seconds [22]. When the electron decays from an excited state to the ground state, part of the energy was previously transferred to nonradiative processes, and consequently, the emitted photon has a lower energy than the excitation photon, and a longer wavelength. The fluorescence or the phosphorescence light is shifted to longer wavelengths when considering the excitation light used for the electronic transition of the molecule. The internal transitions within vibrational levels are much more faster than the fluorescence, and so the fluorescence emission occurs in the transitions between the lowest vibrational level of an excited state to the ground state. In other words, the fluorescence spectrum represents the lowest levels of the vibrational modes, which is characteristic of each specific molecule. In this way, the use of the obtained information using emitted fluorescence can be used in discrimination of molecules. The investigation of biological tissues is much more complex than of individual molecules. Photons interact with biomolecules in several ways, and depending on the type of interaction, they can be classified in three groups [23, 24]. The absorbers are the biomolecules that absorb photon energy. The fluorophores are biomolecules that absorb and emit fluorescent light. The scatterers are biomolecules that does not absorb the photons but changes its direction. Several endogenous cromophores contribute and modify the final tissue spectrum. Distinct fluorophores emit light but the collected spectrum will be modified depending on the presence of absorbers and scatterers in the microenvironment on the path of the emitted photons and the probe interrogator. Each biomolecule presents distinct light interaction as a function of the excitation wavelength, in this way, for a specific wavelength the biomolecule acts as an absorber and for another wavelength it acts as scatterer. The tissue composition plays an important role for the fluorescence spectrum. The type and quantities of the biomolecules influence the resulted tissue fluorescence. If the interrogated tissue presents a high amount of absorbers, less photons will interact with the fluorophores, the emitted fluorescence may be also absorbed by these biomolecules, and less scattering is observed. As a result, the fluorescence signal is low, and its collection for data analysis is difficult. A high concentration of biomolecule and structure ­scatterers may also decrease the collected fluorescence, since the photons may travel away from the collection probe [23, 24]. Taking into account all these light interactions that occur within the biological tissues, it is important to keep in mind that a tissue fluorescence spectrum is a result of the combination of all these processes occurring in the pathway between excitation and collection: excitation absorption and scattering, fluorescence emission, and fluorescence absorption and scattering.

12  Tissue Fluorescence Spectroscopy in Death Time Estimation

243

In this new approach, our group wants to present a proof of concept for the use of fluorescence spectroscopy for PMI determination. Considering that the biochemical composition of the tissues is continuously modified after the cessation of metabolic reactions, we believe that the induced variation of concentration and types of fluorophores, absorbers and scatterers as a function of time may be correlated to fluorescence spectra changes. There is an implicit high variance in tissue degradation at same interrogation time, which includes both the variance between measurement sites at the same interrogation area (that means, fluorescence spectrum taken at each measurement point) and the one existent among individuals. The proposed technique will only be valuable if the differences between the fluorescence spectra of distinct PMIs are higher than the ones observed within the same PMI. In this sense, the data processing must identify spectral changes induced by tissue degradation and determine a pattern behavior as a function of time. The best light source for this proposed methodology is a laser system, due to its monochromaticity and intensity. The monochromaticity allows a higher selectivity in the biomolecules excitation. It is possible to choose the excitation wavelength based on a target cromophore, then the specific electronic transition is more efficient, and finally the fluorescence signal is higher and detection is improved [25–27]. The excitation wavelength is within the ultraviolet-green spectral range because of its higher photon energy and high absorption by biomolecules. A large number of cromophores is related to tissue degradation, but the most relevant ones can be pointed out. The deoxydated form of nicotinamide adenine dinucleotide (NADH) and the oxidized form of flavine adenine dinucleotide (FAD) are fluorophores related to the energetic cell metabolism. Tryptophan is an aromatic amino acid that most contributes in the fluorescence emitted by proteins. Collagen and elastin are structural proteins that shows high fluorescence, mainly due to the cross-linkage between the amino acid chains. The most important biological absorber is the hemoglobin. Proteins are also important absorbers, and together with the hemoglobin, are present in high concentrations and show absorption in UV spectrum, which is the main spectral range used for fluorescence investigation of biological systems. Besides the influence provided by the intrinsic biochemical factors, there are other factors related to the probe/tissue interface, like surface tissue conditions, that modifies the coupling of excitation light to the tissue and later of the emitted light to the collection probe. The light coupling will vary depending on surface roughness and dryness of the tissue, and also on the probe positioning. All these factors may be taken into account during the experiment and data analysis. Based on the fact that there is a variation of quantity and diversity of endogenous components, mainly fluorophores, that alter tissue fluorescence and are related to PMI, we want to demonstrate the viability to monitor these tissue changes through spectral information. Using fluorescence spectral analysis we aim to detect tissue modification at distinct PMIs without the need of dedicated laboratorial chemical tests for quantification of biological components.

244

É.S. Estracanholli et al.

Spectra Interrogation A fluorescence spectroscopy system is basically composed of: a light source, a Y-type probe, a spectrometer, and a computer. The more commonly used light sources are lamps, as mercury lamps, and lasers. The Y-type probe has one end connected to the light source and the other one to the spectrometer. The interrogation tip is the part of the probe that is placed in contact with the target tissue surface. A fiber delivers the excitation light and one or a set of fibers collects the re-emitted light from the tissue surface. In the front of the spectrometer, a longpass filter is placed to reject most part of the backscattered light, allowing the detection of the longer wavelengths, i.e., the fluorescence. The spectrometer detects the light signal and converts it into digital data, which is saved and graphically presented by a software. For our experiments, a portable system was assembled. Two excitation lasers were used, a 408 nm diode laser and a 532 nm doubled frequency Nd:YAG laser. A SF2000 spectrometer (SF2000 – Ocean Optics, USA), a Y-type probe with two 400 mm fibers (Ocean Optics, USA), and a laptop were the other main optical components of our fluorescence spectroscopy system. Two longpass filters at 420 and 545  nm were used for 408 and 532  nm excitation, respectively. The system was assembled inside a case, it is a robust and carry-on device (Fig. 12.1). The fluorescence spectrum measurements are taken with a gentle positioning of the probe at the tissue surface. The probe is perpendicularly placed at the surface and the measurement is acquired in about 2 s. The fluorescence spectrum can be viewed in real time at the laptop monitor and several measurement points can be interrogated. To switch the excitation laser, the probe end connected to the light source is interchanged. An example of the measurement is presented in Fig. 12.2.

Fig. 12.1  Experimental portable system for fluorescence spectroscopy

12  Tissue Fluorescence Spectroscopy in Death Time Estimation

245

Fig. 12.2  Example of fluorescence spectra collected at rats with 96 h of PMI

Our first approach was the interrogation of rat skin. The skin was chosen as the target organ for the initial experiments due to its direct access and also because due to degradation, the internal organs, as liver and kidney, suffer much more influence of the intraperitoneal liquids and after 48–72 h, all the tissue components and degradation liquids are mixed. Since a high variance is observed in biological systems, in our case, the fluorescence spectra taken from different site points at an 2.5 × 2.5 cm interrogation area show distinct behavior even at the same investigation time, a sampling of 40 measurements for each PMI and animal was set in order to get enough number for statistic analysis. Thirty-five male Wistar rats were used in this study (n = 7 for each PMI group). The animals were killed in a CO2 chamber, the hair was removed from a 2.5 × 2.5 cm abdominal area, and the fluorescence measurements were taken at 0, 24, 48, 72, and 96 h of PMI.

Spectral Analysis The first spectral analysis performed was a qualitative evaluation of the fluorescence emission for both excitation wavelengths. An example of the obtained fluorescence spectra at 3 animals and a 24 h PMI is presented in Fig. 12.3a for 408 nm excitation and Fig. 12.3b for 532 nm excitation. Based on a qualitative analysis of the spectra presented in Fig. 12.3, it is possible to note that there is a fluorescence pattern that is common to the investigated ­animals

246

É.S. Estracanholli et al.

Fig. 12.3  Fluorescence spectra of three animals with PMI equal to 24 h, obtained under 408 nm (a) and 532 nm (b) excitation wavelengths

Fig. 12.4  Fluorescence spectra of several PMIs, obtained with excitation wavelengths of 532 nm (a) and 408 nm (b)

at 24 h of PMI. This spectral feature can be used as a potential identifier for tissue degradation occurring in this period of time. The correlation of the fluorescence emission and the biochemical compound must be verified, but hypothesis of this presented feature include, the presence of this fluorophore in higher concentration and/or decrease in the amount of an absorber that absorbs photons at this wavelength. Another possibility is also structural changes in the tissue that result in a higher optical coupling of photons at this wavelength. Even though, several factors can be pointed out and must be further investigated, at this stage, we first have to check if the induced changes in fluorescence spectrum can be correlate with the PMI, no matter which tissue biomolecules are contributing for that. As it can be seen at Fig. 12.3, there is an intrinsic spectral variance that results from differences between the animals, and also between measurements points at the same animal. The interrogated tissue volume represents only a small part of the total investigated area, and the tissue degradation is not uniform considering the time course and degradation products. In Fig. 12.4, the fluorescence behavior as a function of PMI is presented. Besides the intrinsic variance observed at the 24 h group, differences can also be noted on the fluorescence spectra of the other PMIs.

12  Tissue Fluorescence Spectroscopy in Death Time Estimation

247

Through this simple qualitative analysis, it seems that fluorescence spectroscopy is a potential technique, sensitive enough to detect tissue changes during ­postmortem degradation. The variations in fluorescence spectra may be high enough to overcome the intrinsic variations related to biological systems. To test this sensitivity, we performed spectral processing with statistical analysis.

Spectral Processing One of the methods used for the monitoring of spectral variations as a function of PMI is based on the observation that the skin fluorescence under 532 nm excitation, shows a shape change in the spectrum. It was noted that the relative intensities at 600 and 630 nm vary as a function of time. In this way, the first spectral processing was the monitoring of the sample distribution of intensity ratio at 600 and 630 nm for 0, 24, 48, 72, and 96 h of PMI. The other statistical method used was the Principal Component Analysis (PCA), which is widely used for spectrum processing. PCA is a multivariate data analysis that cluster similar samples, discriminating them from distinct ones. It is a classification method and it is useful for multivariate data, since it reduces the number of variables, selecting and using the most relevant ones for clustering the samples in another dimension based on principle components. In our case, the samples are the fluorescence spectra and the variables are the emission wavelengths. The statistical analysis recognizes the emission wavelengths that are most important for sample grouping in the distinct PMIs (classification groups).

Analysis Using the Intensity Ratio It is known that the collected fluorescence intensity is influenced by several factors as excitation laser intensity, fiberoptic positioning and handling, and surface tissue conditions. Based on this fact, the single analysis of the emission intensity can drive to uncorrected conclusions. Although this is a relevant issue, the procedure influence on the fluorescence collection is expected to similarly affect the intensity emission at all wavelengths of the same spectrum. Spectral shape modifications are not expected. The emission spectrum can be described as follows:

F (λ ) = Af (λ )

(12.1)

Where F(l) is the intensity at the emission wavelength l measured by the equipment; f(l) is the intensity at the emission wavelength l an ideal condition (without influence of factors as fiber inclination); A is a coefficient dependent of all procedure-influencing factors with no l-dependence.

248

É.S. Estracanholli et al.

The intensity ratio of two emission wavelengths of the same spectrum is defined by:



α=

F (λi ) F (λ j )

(12.2)

Where a is the intensity ratio, F(li) is the intensity at emission wavelength i, F(lj) is the intensity at emission wavelength j. With this type of analysis, the influencing procedure factors should be eliminated. A total of 280 fluorescence spectra were taken for each investigated PMI. The intensity ratio a was determined for each spectrum and the distribution as a function of PMI is presented in Fig. 12.5. In Fig. 12.5a it is possible to note that there is a high sample distribution in each PMI, but an increase of the average a is observed as the interval increases. This trend can be better observed in Fig. 12.5b, where the average a is presented as a function of time. These results were validated in fluorescence data taken from another group of animals (validation group) with 10 animals for each PMI. Figure 12.6 shows the predicted PMIs based on a analysis vs. the real PMIs for the validation data. If there was a perfect prediction, i.e., a perfect match between the predicted and the real values of PMI, the plot would be the dashed line. Even though these values were not coincident, a high correlation was obtained (solid line). The angular coefficient of the perfect match (dashed line) is equal to 1. Our proposed model (solid line) shows an angular coefficient equals do 0.75. The model presents a higher performance when this value approaches to 1. Although our result is not high, the acquired resolution is at least as good as the actual methods, with the advantage of a fast in situ response.

Analysis Using Principal Component Analysis The principal component analysis (PCA) is one of the most used methods of data processing in Chemistry, Geology and Physics, among other areas that need ­analysis of correlation in a multivariate data [28–32]. This method was first proposed by Pearson in 1901 and further developed by Hottelling 30 years later, but it only begin to be widely used by chemicists in the 1970s for multivariate analysis [33, 34]. The collected fluorescence spectra is usually presented in a plot as the one showed at Fig. 12.7, and this data, intensity for each emission wavelength is saved in a matrix. The columns present all the investigated emission wavelengths and each line represents one collected spectrum. Each wavelength is a variable that shows distinct intensity values depending on the interrogation time and tissue characteristics. This matrix has 900 variables and 1,400 spectrum for 408 nm excitation, and 600 × 1,400 matrix for 532 nm excitation.

12  Tissue Fluorescence Spectroscopy in Death Time Estimation

249

Fig. 12.5  (a) Distribution of the relationship between two bands of fluorescence emission. (b) Temporal evolution of the relationship between emission bands

The lower number of variables for the green excitation is due to the fixed ­detection range of the spectrometer. The samples (lines) were grouped in the PMI groups of 0, 24, 48, 72, and 96 h. The final matrix representation, called Q matrix, is ­presented in Fig. 12.8. At this moment, it is relevant to point out that the pre-normalization procedure of the spectra can induce changes on the correlations between the variables, ­influencing the final results. Anyway, spectrum normalization is still needed due to the already

250

É.S. Estracanholli et al.

Fig. 12.6  Correlation plots of real and estimated PMI determined by the intensity ratio I630/I600

Fig. 12.7  Fluorescence spectra of in vivo animal’s skin represented in graphical form (above) and matrix form (below)

12  Tissue Fluorescence Spectroscopy in Death Time Estimation

251

Fig. 12.8  Representation of data in matricial form

mentioned measurement influences on fluorescence intensity. We decided to normalize our spectra, setting the highest emission at each excitation wavelength equals to 1. For 408 nm excitation the highest emission at 0 h PMI was at 488 nm and for 532 nm excitation at 597 nm. After the spectra normalization, the matrix is transformed in order to delocate the data around an average variable value, following this equation:



 qij − q j  xij =  ,  σj 

(12.3)

where xij is the element of the new data matrix; qij is the element of the data matrix of the measurement ith of the variable j; qj is the average value of variable j; sj is the standard deviation of variable j. This matrix transformation is performed to have each variable presented relatively to the inherent variables of the system. This modification of the data presentation keeps the statistic information of the system and transforms the variables so they have zero average value and standard deviation of 1 [33–35]. The used transformation is only one type of processing that can be performed. Considering X this new data matrix composed by the elements xij from (12.3), the correlation matrix R composed by the correlation coefficients of the same data set is represented by:

R = XT X

(12.4)

252

É.S. Estracanholli et al.

And the elements of matrix R are defined as [31–36] follows:



n

n

i =1

i =1

rjj ′ = ∑ xij xij ′ = ∑

(qij − q j )(qij ′ − q—j ′ ) σ jσ j ′

.

(12.5)

The elements rjj are a standard covariance in the range of −1 and 1. It is ­interesting to note that the matrix R is Hermetian (symmetrical in the case of real variables, as it is our case). The elements at the main diagonal of matrix R (elements where j = j¢) correspond to the variable variance qj [37]. Also for Hermitian matrices, the autovalores are real positive numbers, and its autovetores are ortogonals. When the correlation matrix R is diagonalized, the autovalores and the autovetores can be determined and represented by matrices K and V, respectively.



 λ1 0 K=  ...  0

0 λ2 ... 0

... 0  ↑ ... 0   V =  v1 ... 0    ↓ ... λ n 

↑ v2 ↓

↑  v3 ... ... vn  ↓ ↓  ↑

(12.6)

The diagonalized matrix is a correlation matrix, with elements proportional to the standard deviation. Therefore, the resulted new base is composed by the eigenvectors of R, where each eigenvector represents a part of total variance of the system. This representation is unique and exclusive for each eigenvector, and so they are orthogonal. Each li of matrix K corresponds to the data variance defined by each autovetor ni, and finally when using this eigenvalues it is possible to determine which principal components (PCs), or which component of the new base formed by the eigenvectors, explain most part of the data. Once this base that maximizes the variances of the data set is determined, a projection of the data on this new base is performed:

S = VQ,

(12.7)

where S is the data set projected on the new base and it is called score matrix. This transformation of database from matrix X in matrix S is known as Karhunen– Loève Transposition [37]. The score matrix presents the data in this new database and is represented as it is shown in Fig. 12.9. Initially the data was presented in intensity values as a function of emission wavelength, now the analysis is performed in the variance space of this values. This change on the analysis base allows a significant reduction on the number of dimensions (or variables), just a few PCs have almost all the information of the dataset. In the present study, 600–900 variables could be reduced to 2 PCs with 90% of the data information. The PC1 × PC2 plot presented in Fig. 12.10 shows the PCA results for our dataset. Each dot represents one fluorescence spectrum. The dataset has a total of 1,400 spectra taken at 5 PMIs, 40 measurements/animal, in 7 animals.

12  Tissue Fluorescence Spectroscopy in Death Time Estimation

253

Fig. 12.9  Representation of data in the new base

Fig. 12.10  PC1 vs. PC2 in the interval from 0 to 96 h of PMI, with excitation at 532 nm (a) and 408 nm (b)

The samples from the same PMI group are clustered, but there is dispersion and also overlap with other groups. The clustering occurs because the samples show similar characteristics. The dispersion and overlap is observed because there is a large sample distribution due to the nonhomogenous evolution of tissue degradation that was already discussed. One classification algorithm for the PMI prediction was determined for each excitation wavelength, since the PCA results showed distinct patterns. For the induced fluorescence at 532 nm excitation, box areas were determined around the PMI groups (Fig. 12.11). The probability of a fluorescence measurement of an unknown PMI of being classified in each of the PMI groups can then be determined.

254

É.S. Estracanholli et al.

Fig. 12.11  Representative regions obtained from samples of each PMI

Figure  12.12 presents the data from 4 animals, one for each PMI, that were i­nvestigated for validation of the proposed classification algorithm. The fluorescence spectra were taken using the same procedure previously presented, and then were normalized. This database is projected on the eigenvectors matrix determined in the training group, and then the data is already represented in the PC1 × PC2 plot (Fig. 12.12). Based on the sample distribution within this PC1 × PC2 space, the probability of the animal being part of each PMI group is determined (Table 12.1). As it can be noted, the developed algorithm classifies the animals at the real PMI. We believe that there are several implementations for the fluorescence measurements that may improve the final resolution. Some of these implementations are as follows: multispectral excitation wavelengths, fluorescence lifetime measurements, combined use of exogenous dyes, and imaging techniques. Considering the PCA results for the induced fluorescence at 408 nm excitation, the sample distribution was distinct. A trend can be observed within each PMI group, where the samples are distributed following a linear behavior, instead of the surface distribution of the fluorescence data under 532 nm excitation. In this way, for the 408 nm excitation data, we decided to set a prediction algorithm based on the least-square analysis for the determination of the best fitting for the samples of each PMI group. The correlation between the postmortem hours and the angular coefficient of each PMI group is presented in Fig. 12.13. The plot shows a linear behavior, so the two factors, average angular coefficient and the postmortem hours, are highly correlated. In other words, our proposed analysis can well explain the fluorescence data and correlate with the PMI.

12  Tissue Fluorescence Spectroscopy in Death Time Estimation

255

Fig. 12.12  Validation of PMI determination for four distinct animals, 0 h-animal 1 (a), 24 h-­animal 2 (b), 48 h-animal 3 (c), and 72 h-animal 4 (d)

Table 12.1  PMI (h) 0 24 48 72 96

Probability of each animal from validation group being classified in each of the PMI Prob. Anim1 (%) Prob. Anim2 (%) Prob. Anim3 (%) Prob. Anim4 (%) 97.5 0 1.6 0 2.5 53.4 39.2 8.1 0 45.7 43.5 37 0 0.9 15.7 39 0 0 0 15.9

This algorithm was validated in another group of 10 animals. The same measurement procedure was repeated and the fluorescence data analyzed. Figure  12.14 shows the correlation between the predicted PMI and the real PMI for the validation animals. The dashed line represents a perfect correlation between the predicted and real PMI values, i.e., if the model could perfectly predict all the PMIs. The solid line shows the correlation between the predicted PMI values by our proposed algorithm and the real PMI values. The perfect correlation has an angular coefficient of 1, our model shows an angular coefficient of 0.92. This result shows the high correlation obtained by the proposed model.

256

É.S. Estracanholli et al.

Fig. 12.13  Temporal evolution of angular coefficient obtained from principal components PC1 vs. PC2 with excitation at 408 nm

Fig. 12.14  Correlation plots of real and estimated PMI determined by PCA with excitation at 408 nm

12  Tissue Fluorescence Spectroscopy in Death Time Estimation

257

Conclusion and Perspectives The technique here presented is a new method proposed for PMI determination. This technique is not completely established and the improvements and studies are under development at University of São Paulo, Brazil. The results obtained so far show the potential of the fluorescence spectroscopy for the PMI determination, with at least similar resolution of the actual methods. The main advantage, that we believe this optical technique presents, is a fast response at in situ interrogation. A portable interrogation system allows its use in outdoor areas with no need of laboratory supplies. There is still a lot of room for improvements to make this optical technique more powerful, with a higher resolution on PMI determination. In this study, we used only two excitation wavelengths and three types of spectral analysis. The obtained results may provide complementary information for the PMI prediction, and this can be further investigated. Here we only performed individual algorithms for each excitation wavelength, but it may be worth to try a combined algorithm taking into account both fluorescence data. It is possible that with the implementation of the analysis of an excitation–emission matrix (EEM), or other mathematical processing of the spectra, the obtained resolution here presented may be further improved, increasing the efficacy in the PMI determination.

References 1. Estracanholli ES, Bagnato VS, Kurachi C, Menezes PFC, Vicente JR (2009) Determination of post-mortem interval using in situ tissue optical fluorescence. Optics Express 17:8185–8192 2. Vanezis P, Trujillo O (1996) Evaluation of hypostasis using a colorimeter measuring system and its application to assessment to the post-mortem interval (time of death). Forensic Sci Int 78:19–28 3. Scheurer E, Ith M, Dietrich D, Kreis R, Hüsler J, Dirnhofer R, Boesch C (2005) Statistical evaluation of time-dependent metabolite concentrations: estimation of post-mortem intervals based on in situ H-1-MRS the brain. J Forensic Sci 18:163–172 4. Sabucedo AJ, Furton KG (2003) Estimation of postmortem interval using the protein marker cardiac troponin I. Forensic Sci Int 134:11–16 5. Johnson LA, Ferris JAJ (2002) Analysis of postmortem DNA degradation by single-cell gel electrophoresis. Forensic Sci Int 126:43–47 6. Stan AD, Ghose S, Gao XM et al (2006) Human postmortem tissue: What quality markers matter? Brain Res 1123:1–11 7. Zhou B, Zhang L, Zhang G, Zhang X, Jiang X (2007) The determination of potassium concentration in vitreous humor by low pressure ion chromatography and its application in the estimation of postmortem interval. J Chromatography B Analyt Technol Biomed Life Sci 852:278–281 8. Madea B, Rodig A (2006) Time of death dependent criteria in vitreous humor – accuracy of estimating the time since death. Forensic Sci Int 164:87–92 9. Querido D (1998) A preliminary investigation into postmortem changes in skinfold impedance during the early postmortem period in rats. Forensic Sci Int 96:107–114 10. Querido D, Philipps MRB (2001) Estimation of postmortem interval temperature-correction of extracellular abdominal impedance during the first 21 days of death. Forensic Sci Int 116:133–138

258

É.S. Estracanholli et al.

11. Prieto-Castelló MJ, Hernández del Rincón JP, Pérez-Sirvent C et  al (2007) Application of biochemical and X-ray diffraction analyses to establish the postmortem interval. Forensic Sci Int 172:112–118 12. Vanrell JP (2008) Mecanismo da morte. Disponível em: http://www.pericias-forenses.com.br/ mecamorte.htm 13. Huntington TE, Higley LG, Baxendale FP (2007) Maggot development during morgue storage and its effect on estimating the post-mortem interval. J Forensic Sci 52:453–458 14. Amendt J, Campobasso CP, Gaudry E et al (2007) Best practice in forensic entomology – standards and guidelines. Int J Legal Med 121:90–104 15. Bagizadegan K et  al (2004) Spectroscopic diagnosis and imaging of invisible pre-cancer. Faraday Discuss 126:265–279 16. Wagnieres GA, Star WM, Wilson BC (1998) In vivo fluorescence spectroscopy and imaging for oncological applications. Photochem Photobiol 5:603–632 17. Loschenov VB, Konov VI, Prokhorov AM (2000) Photodynamic therapy and fluorescence diagnostics. Laser Phys 10:1188–1207 18. Stookey GK (2003) The evolution of caries detection. Dimensions of Dental Hygiene Oct:12–15 19. Meller C, Heyduck C, Tranaeus S, Splieth C (2006) A new in vivo method for measuring caries activity using quantitative light-induced fluorescence. Caries Res 40:90–96 20. Figueiredo ACR, Kurachi C, Bagnato VS (2005) Comparison of fluorescence detection of carious dentin for different excitation wavelengths. Caries Res 39:393–396 21. Marcassa LG, Gasporato MCG, Belasque J et al (2006) Fluorescence spectroscopy applied to orange trees. Laser Phys 16:884–888 22. Cohen-Tannoudji C, Diu B, Laloë F (1977) Quantum mechanics. Wiley, New York, p 1528 23. Ramanujam N (2000) Fluorescence spectroscopy in vivo. Encyclopedia of analytical chemistry. Wiley, Chichester, pp 20–56 24. Lakowicz JR (1983) Principle of fluorescence spectroscopy. Plenum Press, New York, p 496 25. Georgakoudi I (2002) NAD(P)H and collagen as in vivo quantitative fluorescent biomarkers of epithelial precancerous changes. Cancer Res 62:682–687 26. Cheong W, Prahl AS, Welch AJ (1990) A review of the optical properties of biological tissues. IEEE Quantum Electron 26:2166–2185 27. Kurachi C (2005) Espectroscopia de fluorescência na detecção de lesões quimicamene induzida por agentes carcinogênicos na borda lateral da língua – estudo in vivo. Universidade de São Paulo, São Carlos, Tese de doutorado – Instituto de Fisica de São Carlos 28. Davies AMC, Fearn T (2005) Back to basics: the principles of principal component analysis. Spectroscopy Europe, pp 20–23 29. Landim PMB (2008) Análise estatística de dados geológicos multivariados. Departamento de Geologia Aplicada, Universidade do estado de São Paulo. Disponível em http://www.rc.unesp. br/igce/aplicada/DIDATICOS/LANDIM/multivariados.pdf 30. Sena MM, Poppi RJ, Fighetto RTS, Valarini PJ (2000) Avaliação do uso de métodos quimiométricos em análise de solos. Quimica Nova 23:547–556 31. Valderrama P, Braga JWB, Poppi RJ (2007) Validation of multivariate calibration models in the determination of sugar cane quality parameters by near infrared spectroscopy. J Brazil Chem Soc 18:259–266 32. Lopez JA (2007) Principal components analysis. Lisboa: Instituto Superior de Engenharia de Lisboa. Disponível em: http://www.deetc.isel.ipl.pt/comunicacoesep/disciplinas/pes/pes_pt.htm. 33. Barros Neto B, Scarminio IS, Bruns R (2006) 25 anos de quimiometria no Brasil. Quimica Nova 29:1401–1406 34. Ferreira MMC, Antunes AM, Melo MS, Volpe PLO (1999) Quimiometria I: calibração multivariada, um tutorial. Química Nova 22:724–773 35. Moita Neto JM, Moita GC (1998) Uma introdução à análise exploratória de dados multivariados. Química Nova 21:467–469 36. Gonçalves de Souza AJ (2008) Análise em componentes principais. Technical University of Lisbon. Disponivel em: http://biomonitor.ist.utl.pt/~ajsousa/. 37. Costa LF (2001) Shape analysis and classification: theory and practice. CRC Press, Boca Raton, p 456

Chapter 13

Heat-Flow Finite-Element Models in Death Time Estimation Holger Muggenthaler, Michael Hubig, and Gitta Mall

Abstract  Heat-flow finite-element (FE) models are based on the principles of thermodynamics. The numerical finite-element method (FEM) allows solving the partial differential equation of heat flow for complex geometrical, initial and boundary conditions. This article provides an overview of heat-flow mechanisms, their FE-modelling and their application to post-mortem cooling. A FE-model of the human body is presented which can be used to simulate standard and non-standard post-mortem cooling scenarios. Keywords  Thermodynamics • Death time estimation • Heat flow mechanisms • Body cooling

Introduction Death time estimation is of major importance for reconstructing homicide cases and for verifying claims by alleged perpetrators. In forensic casework, different methods can be applied in the early post-mortem phase. The evaluation of the signs of death (post-mortem lividity and post-mortem rigidity), of supravital reactions (mechanical stimulation of brachial muscles, electrical stimulation of facial muscles, pharmacological stimulation of pupillar muscles) and of post-mortem body cooling is common in forensic routine. Each of these evaluations is subject to inevitable inaccuracies. While evaluating post-mortem lividity varies with the examiner, ATPconsumptive processes prior to death influence post-mortem rigidity and supravital reactions. The post-mortem cooling rate of a corpse strongly depends on the environmental conditions like ambient temperature, ground, clothing, covering, etc.

H. Muggenthaler (*) Institut für Rechtsmedizin, Universitätsklinikum Jena, Jena 07743, Germany e-mail: [email protected] E.E. Turk (ed.), Forensic Pathology Reviews, Volume 6, Forensic Pathology Reviews 6, DOI 10.1007/978-1-61779-249-6_13, © Springer Science+Business Media, LLC 2011

259

260

H. Muggenthaler et al.

Nevertheless, due to the fact that human temperature regulation maintains a core temperature of 37°C under physiological conditions and fails at the time of death (cardio-respiratory arrest), temperature-based methods still provide the most reliable results. The measurement of the deep rectal temperature at the crime scene has been established in practical case work, because the measurement location is non-invasively accessible and provides a core temperature. The basis for death time back calculation using the temperature measured at the crime scene is a temperature–time model curve valid for the specific case. Two different approaches can be applied to obtain model curves. Empirical modelling consists in describing experimental temperature–time curves by ad hoc mathematical formulae. They are easily applicable but valid only for the experimental standard cooling conditions. Most forensic methods belong to this class of models [1–10]. Heat-flow models are developed on the basis of the physical laws of heat transfer. They are in principle valid for non-standard cooling conditions but time-consuming and difficult to apply. Only few forensic methods belong to this class of models [11–14]. To cope with the complex geometrical, initial and boundary conditions of the cooling scenarios met in forensic routine, our working group first applied the FEM to simulate post-mortem cooling [15–17]. We developed a FE-model of the whole human body with 14 different body compartments to which different thermal properties are assigned. The model can be scaled according to real body height and weight. Case-specific clothing, covering and ground can be modelled. Meanwhile, a case of death time estimation using a FE-model of a human torso with three different tissue layers was reported by den Hartog and Lotens [18]. This article gives a brief overview of the principles of thermodynamics and finite-element simulation followed by a detailed description of the model developed by our working group including model validation and model application.

Thermodynamics Heat Transfer Mechanisms Three general heat transfer mechanisms can be distinguished: Conduction, convection and radiation. Heat conduction is defined as a transfer of heat within a medium by particle interaction without particle flow. Heat is transferred from more energetic particles to adjacent less energetic particles without the interacting particles changing their location. Heat conduction is expressed by Fourier’s law (which is cited here for the 1-dimensional case):



dT Q = −kA , dx

13  Heat-Flow Finite-Element Models in Death Time Estimation

261

where Q is the amount of heat transferred per unit of time whereas dT/dx represents the temperature gradient in x-direction. A is the area orthogonal to the heat-flow direction. The ability of a material to transfer heat via conduction is quantified by the thermal conductivity k. Fatty tissue with a low thermal conductivity of 0.2 W/(m°K) possesses better isolating material properties than bone with a higher thermal conductivity of 0.75 W/(m°K). Heat transfer by particle flow is referred to as heat convection. In case of heat flow through a solid–fluid boundary it can be described by Newton’s law of cooling: Q = hA(TS − T∞ ),



where h is the convection heat transfer coefficient and the term in parenthesis represents the temperature difference between the surface and a location in the fluid (or gas) far from the surface. Unlike thermal conductivity, the convection coefficient does not depend on the material properties of the solid body but amongst others on the fluid (or gas) properties and on the surface geometry. If the fluid (or gas) is in the state of an externally caused motion (e.g., current or wind) convection is called forced convection; otherwise, it is called natural or free convection. A third heat transfer mechanism is radiation. According to the law of Stefan and Boltzmann heat loss by electromagnetic radiation can be expressed by:

(

)

Q rad = εσ A Ts4 − Te4 ,

where e is the emissivity of a surface, s the Stefan–Boltzmann constant, A the surface area, TS the surface temperature and Te the environmental temperature. The emissivity e corresponds to the ability of a material to emit radiation. Human skin or water (>0.9) possess high emissivities, polished metallic surfaces low emissivities (0.03–0.07). For convenience, the transfer mechanisms were formulated for one spatial dimension only.

Equation of Heat Transfer The general form of the homogeneous heat transfer equation can be expressed as a partial differential equation for the temperature field T as a function of the space and time coordinates including the thermal conductivity k, the density r and the specific heat capacity c:

c· ρ ·T − k·∆T = 0.

The equation relates the time-dependent change of temperature T to the spatial Laplace operator of the temperature within a control volume. It is only valid in

262

H. Muggenthaler et al.

cases without external or internal heat sources or heat sinks. The higher the specific heat capacity of a material, the higher the energy input to increase the temperature of the material. By introducing a term for the change of thermal energy with time by sources and sinks the equation can be written in the more general form:

c· ρ ·T − k·∆T = Q .

Radiation and convection can be interpreted as heat energy sources and sinks denoted by Q .

Finite-Element Method Theory The incorporation of the boundary conditions radiation and convection in the heat transfer equation leads to a partial differential equation, which cannot be solved analytically. In engineering sciences, the FEM represents an established numerical approach for solving partial differential equations dealing with structural, electrostatic or thermodynamic problems. Automotive engineering is one field of application where numerical simulation has been widely used for crash and occupant safety development. Current research projects deal with the development and validation of FE-based human models [19]. In the following, a short introduction to the theory of the FEM is presented. The FEM is based on the idea to split up a complex geometrical structure into basic geometrical elements. This step is called discretisation and results in a set of elements and nodes. Within every element an interpolation function (also called shape function) is defined for each node of the element. The value of the interpolation function is 1 for the related node and 0 for all other nodes. Then the unknown course of the temperature distribution T within the element can be expressed as a linear combination of the shape functions. By using the values of the shape functions on the nodes as weights element matrices can be developed. The compilation of the entire set of element matrices and the boundary conditions as well as the discretisation with respect to time provide a set of linear equations which can be solved using standard numerical algorithms. The FE-interpolation approach can be demonstrated for the 1-dimensional case in a linear bar, modelled by two finite elements E1 and E2, with unknown temperature distribution (Fig. 13.1). The temperatures at the locations x0, x1 and x2 (or nodes N0, N1 and N2) are known. Two simple linear shape functions h1 and h2 are defined such that in element E1 the function h1 assumes the value 1 at node N0 and 0 at node N1 and the function h2 assumes the value 0 at node N0 and 1 at node N1. The estimate T #(x) of the real temperature T(x) value at location x can now be obtained by a linear

13  Heat-Flow Finite-Element Models in Death Time Estimation

263

Fig. 13.1  Explanatory scheme for finite-element method

combination of the known temperature values T(x0) and T(x1) at the nodes N0 and N1 and the values of the functions h1 and h2 at the location x:

T # ( x ) = T ( x0 )h1 ( x ) + T ( x1 )h2 ( x ).

The application of the FEM for solving the homogeneous heat transfer equation is presented. The partial differential equation of order 2: cρ T − λ∆T = 0



contains a time derivative T of the temperature field T and the Laplacian of the temperature field T which combines its second order spatial derivatives: ∆T =

∂2 ∂2 ∂2 T + T + T. ∂x 2 ∂y 2 ∂z 2

The remaining letters denote thermal material properties: c is the specific heat capacity, r the density and l the thermal conductivity. Transformations of the heat transfer equation using test functions and the Gaussian integral theorem are performed, by which one spatial derivative disappears and the other spatial derivative is transposed to the shape functions. A system

264

H. Muggenthaler et al.

of ordinary differential equations of order 1 remains containing only the time derivative T # of the approximated temperature field T #. The heat transfer equation then becomes the finite-element equation:

MT # − KT # = 0,

where M and K represent n × n-matrices where n is the total number of shape functions hi of the model. M is the heat capacity matrix:

M = cρ ∫ H t H dV V

with

H = (h1 ,...hn )

K is the conductivity matrix:

K = λ ∫ B t B dV V

with



 ∂h1   ∂x  ∂h B= 1  ∂y  ∂h1   ∂z

∂hn ∂x ∂hn ... ∂y ∂hn ... ∂z ...

    .    

Matrix B contains the spatial derivatives of the shape functions hi. As described above, the spatial derivatives of the temperature field were transposed to the shape functions by approximating the temperature field by a linear combination of the shape functions. Since the shape functions are simple arithmetic functions, their spatial derivatives can easily be computed. This equation system can be solved numerically by the finite-difference method replacing the derivatives by difference quotients. For the sake of simplicity, the effects of the element deformation transformations on the formulae were omitted.

Finite-Element Software In general, FE-software packages consist of three parts. The pre-processor provides features for generating FE-models including geometrical entities as well as for defining the initial and boundary conditions. The calculation is performed by the finite-element solver and the simulation results can be viewed and analysed in the post-processor.

13  Heat-Flow Finite-Element Models in Death Time Estimation

265

Explicit finite-element solvers can be differentiated from implicit finite-element solvers. For highly dynamic mechanical problems like crash simulation the explicit solution scheme is commonly applied. In case of mechanical computations accelerations and displacements are calculated by finite-difference equations for each time step. The time step has to fulfil certain conditions to ensure a numerical stable solution. An implicit finite-element solution scheme follows the approach described above resulting in a set of linear equations for each time step that can be solved with numerical standard algorithms. For thermo-dynamical problems only implicit solvers are applied.

Finite-Element Model An FE-model of the human body was developed by Mall [15] using the pre-/ post-processor MSC Mentat and the finite-element solver MSC Marc. The original model consists of an unclothed body and a steel support representing the steel trolley used in most of the published experimental setups.

Geometry For saving modelling and computation time symmetry of the human body in the sagittal plane was assumed and only one-half of the body was modelled. The model (Fig. 13.2) with body and support consists of 9,804 solid cuboid 3D elements and 12,503 nodes, the body alone is made up of 8,328 elements and 10,154 nodes. The model contains 14 different tissue (material) compartments. The meshing of the 3D geometry was performed manually, height and weight fit a 50-percentile man. In order to use the standard model for different body constitutions, the model can be individually scaled to the actual body height and weight. The scaling factor in z-direction for the body length is

l=

L′ , L

where L denotes the original length of the body model and L¢ the actual body length. The model can be scaled in x- and y-direction using the factor s:



s=

M ′L , ML ′

where M is the mass of the original model and M¢ represents the actual mass.

266

H. Muggenthaler et al.

Fig. 13.2  FE-model with grey-scaled materials including the steel trolley support

Material Properties The model was divided into 14 approximate anatomical compartments in order to get an approximate representation of the human tissue distribution. The thermal properties of human tissues were extensively measured by other studies (Table 13.1) in post-mortem tissues and already applied to physiological models. Table  13.1 summarizes the thermal properties of different tissues taken from Werner and Buse [20]. The values for skin, subcutaneous fat, muscle, bone, brain and lungs were directly attributed to the elements in the model. For the remaining compartments containing different organs and tissues the thermal properties were interpolated according to the approximate proportions of the organs and tissues in the compartments. The thermal properties of different clothing and building materials can be found in (textile) engineering literature [21].

13  Heat-Flow Finite-Element Models in Death Time Estimation

267

Table 13.1  Thermal tissue properties: conductivity k, specific heat capacity c, density r and emissivity e [20]

Tissue Skin Subcutaneous fat Muscle Bone Brain Facial tissues Cervical organs Mediastinal organs Lungs Upper abdominal organs Gastrointestinal organs Kidneys Retroperitoneal tissues Pelvis organs

 W  k   m·°C 

 J  c   kg·°C 

 kg  ρ  3 m 

0.47 0.21 0.51 0.75 0.49 0.51 0.48 0.47 0.28 0.48 0.46 0.39 0.51 0.49

3.680 2.300 3.800 1.700 3.850 3.245 3.363 3.375 3.520 3.730 3.346 3.158 3.800 3.350

1.085 920 1.085 1.357 1.080 1.056 1.006 1.033 560 1.080 933 1.026 1.085 1.008

ε 0.95

Initial Conditions The finite-element solver requires initial temperature values for each node. This initial 3D temperature field can be chosen arbitrarily. In the standard model the initial temperature values were attributed according to the temperature gradients between body core and shell in a living unclothed body at room temperature [22].

Boundary Conditions Besides convection, conduction and radiation the so-called internal power was introduced as a further boundary condition to account for a heat production during intermediary life between clinical and biological death. Conduction Heat conduction is described by the homogeneous part of the heat transfer equation:

c· ρ ·T − k·∆T = Q

with specific heat capacity c, density r and thermal conductivity k.

268

H. Muggenthaler et al.

Convection Heat convection is described by an inhomogeneous part of the heat transfer equation: Q = hA(TS − T∞ ),



where h represents the heat transfer coefficient on the surface A. This coefficient is not material specific but depends on parameters like fluid viscosity, fluid velocity or surface characteristics. h can be expressed as a function of a dimensionless number called the Nusselt number Nu: h=



k Nu δ

with the conductivity k of the fluid and the characteristic length d of the geometry. Convection coefficients in natural convection differ from those in forced convection. In case of combined convection the Nusselt number can be expressed by the Nusselt numbers determined for natural and forced convection. For technical applications the Nusselt numbers have been determined empirically for different geometries and for different flow conditions (e.g. laminar or turbulent). Convection coefficients for the geometry of a human body have been measured [23]. For natural convection a value of h = 3.3 W/(m² °C) is assumed. Radiation Heat loss and heat gain by thermal radiation is described by the law of Stefan and Boltzmann

(

)

Q rad = εσ A Ts4 − Te4 ,

where e is the emissivity of a surface, s the Stefan–Boltzmann constant, A the surface area, TS the surface temperature and Te the environmental temperature. The heat loss and heat gain of emitting and radiating surfaces depend on the geometry and orientation of the surfaces. The view factor of two surfaces A1 and A2 is defined as the ratio of the radiation leaving surface A1 and striking surface A2. A view factor of 1 expresses, that all radiation emitted by the one surface reaches the other surface. If no radiation from surface A1 strikes surface A2 the view factor is 0. The view factor for two surfaces is calculated by: Vf ( A1 , A2 ) =

cos β1 cos β 2 1 dA1dA2 , F ( A1 ) ∫A1 ∫A2 π r2

where A1 represents the emitting surface, A2 the absorbing surface and b1 and b2 denote the angles between the emitted/absorbed ray and the normal of the emitting/ absorbing surface (Fig. 13.3). The distance between the surfaces is r.

13  Heat-Flow Finite-Element Models in Death Time Estimation

269

Fig. 13.3  Explanatory scheme for radiation view factors

By incorporating the view factors within the Stefan and Boltzmann law the thermal power transmitted from an emitting surface A1 to an absorbing surface A2 can be calculated as

(

)

Q1−> 2 = F ( A1 )ε1ε 2Vf ( A1 , A2 ) σ T14 − T24 ,

where F(A1) is the area of the surface A1. View factors for simple 3D and 2D geometries can be looked up in tables [24]. In case of multiple objects and complex arrangements of emitting and absorbing surfaces a simple mathematical description of view factors is not possible. FE-environments provide tools for Monte Carlo simulation where a random generator produces many (up to several thousands) rays of different directions which are virtually emitted from each defined element surface. The view factors for each pair of surfaces can be calculated from the ratio of the number of rays emitted from the one surface and the number of rays hitting the other surface. According to the majority of authors, energy loss due to radiation is thought to be negligible. However, theoretical investigations by Mall et al. [25, 26] showed that radiation can considerably contribute to post-mortem energy loss, primarily within the first few hours post-mortem.

Internal Power The internal energy represents the production of thermal energy in the phase of intermediary life. Assuming that the internal heat production via supravital reactions decreases exponentially, the internal heat production can be expressed as a function of time

P = P(0)·e −α t ,

where P(0) is the initial value of internal power instantaneously after death and a is the rate of decrease in energy production. P(0) can be approximated by the basal metabolic rate of a living individual related to body mass. Different body compartments

270

H. Muggenthaler et al.

exhibit different basal metabolic rates. According to Klinke and Silbernagel [22] the overall metabolic rate can be split as follows: • Body core (mediastinum, lungs, upper abdominal organs, gastrointestinal organs, kidneys and pelvic organs): 60% • Muscle tissue: 18% • Brain: 17% • Bone tissue, facial and cervical connective tissues: 5% Lundquist [27] estimated the amount of energy produced from the breakdown of glycogen for an individual of 70  kg (approximately 587  kJ) during intermediary life. Due to hydrolysis and phosphorous compounds he assumed that half of the estimated energy production from glycogen breakdown has to be added. The calibration of the exponential formula was done on the basis of the empirical model by Marshall and Hoare [4] founded on numerous cooling experiments (n = 176). FE-analyses were performed with varying values for the decrease rate a. Least-square measures representing the distances between the Marshall and Hoare curves and the curves received from the simulations were calculated. The best fit between simulation results and literature data was obtained for a = 0.0000770164 s−1, which corresponds to a half-life rate tH = ln2/a of tH » 2.5 h.

Validation Model Validation The model with the calibrated value of the decrease rate of the internal energy production was indirectly validated on the basis of the empirical model of Marshall and Hoare. The validation was performed for two different body constitutions (height: 1.60 m, weight: 61 kg and height: 1.90 m, weight: 103 kg) as well as for two different environmental temperature levels (8°C and 18°C) with good results (Figs. 13.4–13.7). In addition to this indirect validation a direct validation of the model by own body cooling experiments under strictly controlled conditions is being performed. On the whole 84 temperature–time curves of recently deceased with known time of death were recorded inside a climate simulation chamber by our working group between 2003 and 2007. In order to minimize possible influence factors two experiments with unclothed corpses, normal body mass indices and similar body constitutions were selected from the data base. Model validity was assessed for a node that corresponds to the experimental measurement location in the rectum. Prior to simulation, the model was scaled to the height and weight given in Table 13.2 for case 1. After a slight modification of the thickness of the subcutaneous fat of the torso, the simulation results were in good correspondence with the experimental curve in case 1. To verify the validity of the modified model a

13  Heat-Flow Finite-Element Models in Death Time Estimation

271

Fig. 13.4  Indirect validation (1.60 m, 61 kg, 13°C), MH Marshall and Hoare, FEM finite-element simulation

Fig. 13.5  Indirect validation (1.90 m, 103 kg, 13°C), MH Marshall and Hoare, FEM finiteelement simulation

simulation was performed using the boundary conditions and the body height and weight for case 2. Figures 13.8 and 13.9 show the simulation results for case 1 and case 2 as well as the temperature curves predicted by the Henssge model. As can be seen from both diagrams, the entire rectal cooling curves predicted by the Henssge

272

H. Muggenthaler et al.

Fig. 13.6  Indirect validation (1.75 m, 80 kg, 8°C), MH Marshall and Hoare, FEM finite-element simulation

Fig. 13.7  Indirect validation (1.75 m, 80 kg, 18°C), MH Marshall and Hoare, FEM finite-element simulation

model (thick grey lines) lie above the temperature values measured in the experimental setup (black solid line). In the first case, the temperatures calculated in the simulation (black dashed line) fit the measured values well, although the calculated values lie slightly below the measured curve in the first 20 h and slightly above the

13  Heat-Flow Finite-Element Models in Death Time Estimation Table 13.2  Measurement cases for validation (t0: time between death and start of measurement, Ta: ambient temperature) Case h [m] m [kg] Ta [°C] t0 Cause of death 1 1.64 64 18 2 h, 44 min Suicide, shoot 2 1.68 65.3 18.7 1 h, 40 min Fall, staircase

Fig. 13.8  Validation results (case 1)

Fig. 13.9  Validation results (case 2)

273

274

H. Muggenthaler et al.

measured temperatures in the second part of the measurement. For the second case the temperatures calculated in the simulation show similar characteristics. In case of “standard conditions” (normal body mass index, no clothing, steel trolley support) the finite-element model is able to reproduce realistic body core temperature curves. By changing the boundary conditions or by adding elements representing different clothing layers the model can be adjusted to address the boundary conditions of the real case.

Conclusions Heat-flow models of post-mortem cooling are based on the physical laws of heat transfer. The differential equation of heat transfer can be solved analytically for simplified geometries only, like the cylinder of infinite length [11]. The finiteelement method (FEM) represents an established numerical computational procedure in engineering sciences for solving partial differential equations like the heat transfer equation under complex geometrical, initial and boundary conditions. By applying the FEM to post-mortem cooling our working group was able to develop a model of the human body with 14 different tissue and organ compartments, to which the different thermal material properties are assigned. Depending on the routine case clothing, covering and ground can be modelled specifically. The initial temperature field is modelled inhomogeneously with a temperature gradient from body core to shell and can be varied to account for hypothermia or hyperthermia at death. Boundary conditions like convection (natural as well as forced convection) and radiation (including irradiation by sun or radiant heater) can be simulated according to their physical laws. The environmental conditions can be varied with time to account for variable environmental temperatures. Thus the FE-model of post-mortem cooling provides an additional approach to determine the time since death from body temperature decrease. Unlike by other commonly applied methods the cooling conditions met in forensic routine can be simulated specifically.

References 1. Henssge C, Althaus L, Bolt J, Freislederer A, Haffner HT, Henssge CA, Hoppe B, Schneider V (2000) Experiences with a compound method for estimating the time since death. I. Rectal temperature nomogram for time since death. Int J Legal Med 113(6):303–319 2. De Saram GSW, Webster G, Kathirgamatamby N (1955) Post-mortem temperature and the time of death. J Crim Law Pol Sci 46:562–577 3. Fiddes FS, Patten TD (1958) A percentage method for representing the fall in body temperature after death. J Forensic Med 5(1):2–15 4. Marshall TK, Hoare FE (1962) Estimating the time of death. J Forensic Sci 7:56–81; 189–210; 211–221 5. Henssge C (1979) Precision of estimating the time of death by mathematical expression of rectal body cooling. Z Rechtsmed 83:49–67

13  Heat-Flow Finite-Element Models in Death Time Estimation

275

6. Henssge C (1981) Todeszeitschätzungen durch die mathematische Beschreibung der rektalen Leichenabkühlung unter verschiedenen Abkühlbedingungen. Z Rechtsmed 87:147–178 7. Green MA, Wright JC (1985) Postmortem interval estimation from body temperature data only. Forensic Sci Int 28:35–46 8. Green MA, Wright JC (1985) The theoretical aspects of the time dependent Z equation as a means of postmortem interval estimation using body temperature data only. Forensic Sci Int 28:53–62 9. Morgan C, Nokes LDM, Williams JH, Knight BH (1988) Estimation of the post mortem period by multiple-site temperature measurements and the use of a new algorithm. Forensic Sci Int 39:89–95 10. Al-Alousi LM, Anderson RA, Worster DM, Land DV (2001) Multiple-probe thermography for estimating the postmortem interval: I. Continuous monitoring and data analysis of brain, liver, rectal and environmental temperatures in 117 forensic cases. J Forensic Sci 46(2):317–322 11. Sellier K (1958) Determination of the time of death by extrapolation of the temperature decrease curve. Acta Medic Soc Leg 11:279–302 12. Joseph AEA, Schickele E (1970) A general method for assessing factors controlling postmortem cooling. J Forensic Sci 15(3):364–391 13. Hiraiwa K, Kudo T, Kuroda F, Ohno Y, Sebetan IM, Oshida S (1981) Estimation of postmortem interval from rectal tremperature by use of computer – relationship between the rectal and skin cooling curves. Med Sci Law 21(1):4–9 14. Hiraiwa K, Ohno Y, Kuroda F, Sebetan IM. Oshida S (1980) Estimation of post-mortem interval from rectal tremperature by use of computer. Med Sci Law 20(2):115–125 15. Mall G (2000) Temperaturgestützte Bestimmung der Todeszeit mit Hilfe der Methode der Finiten Elemente. Habilitation, München 16. Mall G, Eisenmenger W (2005) Estimation of time since death by heat-flow Finite-Element model. Part I: method, model, calibration and validation. Leg Med 7(1):1–14 17. Mall G, Eisenmenger W (2005) Estimation of time since death by heat-flow Finite-Element model. Part II: application of non-standard cooling conditions and preliminary results in practical casework. Leg Med 7(1):69–80 18. den Hartog EA, Lotens WA (2004) Postmortem time estimation using body temperature and a finite-element computer model. Eur J Appl Physiol 92:734–737 19. Muggenthaler H, von Merten K, Peldschus S, Holley S, Adamec J, Praxl N, Graw M (2008) Experimental tests for the validation of active numerical human models. Forensic Sci Int 177(2–3):184–191 20. Werner J, Buse M (1988) Temperature profiles with respect to inhomogeneity and geometry of the human body. J Appl Physiol 65(3):1110–1118 21. Schneider K-J, Goris A (2006) Bautabellen für Ingenieure, 17th edn. Werner-Verlag, Neuwied 22. Klinke R, Silbernagel AS (1994) Lehrbuch der Physiologie. Thieme, Stuttgart 23. de Dear RJ, Arens E, Hui Z, Oguro M (1997) Convective and radiative heat transfer coefficients for individual human body segments. J Biometeorol 40(3):141–156 24. Çengel AC (1998) Heat transfer: a practical approach. McGraw-Hill, Boston a.o 25. Mall G, Hubig M, Beier G, Eisenmenger W (1998) Energy loss due to radiation in postmortem cooling. Part A: quantitative estimation of radiation using the Stefan-Boltzmann law. Int J Legal Med 111(6):299–304 26. Mall G, Hubig M, Beier G, Büttner A, Eisenmenger W (1999) Energy loss due to radiation in postmortem cooling. Part B: energy balance with respect to radiation. Int J Legal Med 112(4):233–240 27. Lundquist F (1956) Physical and chemical methods for the estimation of the time of death. Acta Med Leg Soc (Liège) 9:205–213

wwwwwwwwwwwwwwwww

Chapter 14

The Use of Protein Markers for the Estimation of the Postmortem Interval Yekaterina Poloz and Danton H. O’Day

Abstract  The postmortem interval (PMI) is a critical variable in any death ­investigation. The goal is to elucidate a new and precise biochemical method for PMI determination in humans. Specific protein tissue markers show predictable and precise postmortem degradation patterns that could be utilized to develop a new method for PMI determination. We provide a compilation of available research in this field and review the use of calmodulin-binding proteins as PMI markers. Finally, we propose the development of a portable, hand-held device that will allow the use of protein markers for PMI determination at the scene of a crime. Keywords  Postmortem interval • Protein • Degradation • Calmodulin • Calpain

Introduction The postmortem interval (PMI) is the time between the death of an individual and the postmortem analysis of the body. This variable is an essential component of any death investigation. According to Vass et al. [72] early PMI is up to 72 h postmortem, before the onset of putrefaction, while late PMI is any time after 3 days [72]. Unfortunately, PMI estimation is not an exact science. The various methods employed in estimating PMI yield wide postmortem windows and sometimes contradict one another. The field of forensic pathology will benefit immensely from the development of more accurate and precise methods for the estimation of PMI. As Sabucedo and Furton [55] mentioned, “time since death markers have lagged behind the advance in technology of the past 150 years” [55]. In this chapter, we will D.H. O’Day (*) Department of Cell and Systems Biology, University of Toronto, M5S 3G5 Toronto, ON, Canada Department of Biology, University of Toronto, L5L 1C6, Mississauga, ON, Canada e-mail: [email protected] E.E. Turk (ed.), Forensic Pathology Reviews, Volume 6, Forensic Pathology Reviews 6, DOI 10.1007/978-1-61779-249-6_14, © Springer Science+Business Media, LLC 2011

277

278

Y. Poloz and D.H. O’Day

focus on postmortem protein degradation patterns as markers for PMI determination. We will first cover the available research on postmortem protein degradation in the English literature and then introduce our own research on calmodulin-binding proteins. The goal is to develop a field-based assay kit for PMI determination utilizing protein degradation patterns as markers of specific PMIs. Death results in global anoxia, the global lack of oxygen after blood circulation ceases. When cells are deprived of oxygen they cannot undergo respiration and thereby cannot produce energy in the form of ATP. If energy production ceases cells die. As a result, one of the hallmarks of cell death is the change in cell membrane permeability. The cell membrane acts as a selective barrier, controlling entry and exit of molecules, thereby maintaining homeostasis inside the cell. Once membrane permeability is compromised, a cell cannot control the passage of molecules. This leads to the breakdown of cellular homeostasis, as ions and other molecules flow in and out uncontrollably. Signal transduction pathways become deregulated that lead to the degradation of cellular components including ribonucleic acid (RNA), deoxyribonucleic acid (DNA) and proteins. Biochemical methods for PMI estimation have been explored for many years. The most accurate biochemical method that is being used is the estimation of potassium content of the vitreous humor [64]. Due to new discoveries and developments in the biochemical field a new era is slowly emerging. RNA, DNA and various proteins are the new targets. These molecules are precisely regulated during life and often predictably degraded after death. After death, protein levels have been shown to be more stable than RNA or DNA. There is an extensive amount of research available on the degradation patterns of specific proteins in an array of tissues after death, though many researchers conduct such research for purposes other than finding potential markers for PMI determination. Data on protein degradation patterns exists that could be useful for the estimation of early (up to 72 h) and late (after 72 h) PMI. Immunohistochemistry, western blotting, and 2D gel electrophoresis coupled with mass spectrometry are the most common methods that have been used to assess protein degradation patterns after death. Western blotting and immunohistochemistry rely on the immunological reaction of a specific antibody to its target antigen, usually a specific protein. The antibody recognizes and binds to a sequence in the protein called the epitope. Degradation of the protein after death involves cleavage of the protein at specific sites. The cleavage pattern depends on the types of degradative enzymes that are involved. Once the portion of the protein containing the epitope is cleaved the antibody can no longer bind to and detect the protein. Therefore, both techniques detect the intact protein and any degradation products that still contain the epitope. In western blotting one can determine whether the antibody detects the intact protein or its degradation product(s) by monitoring the changes in the size of the protein. This cannot be done using immunohistochemistry. Furthermore, the results of an immunohistochemistry are usually scored in terms of a positive vs. a negative reaction rather than giving a quantitative value. 2D gel electrophoresis with mass spectrometry works differently. One can also follow the degradation pattern of a protein by monitoring the size of the cleavage products. This more complex technique does not rely on antibodies or

14  The Use of Protein Markers for the Estimation of the Postmortem Interval

279

the presence of a specific epitope so all of the degradation products can be detected. But it is somewhat cumbersome because there is a two-stage process to generate a gel in which hundreds of spots are revealed. Subsequently each protein spot needs to be isolated and then identified using mass spectrometry. Identification is not always correct and the procedure is very time consuming and requires expensive equipment. Clearly such an approach could not be used to develop a field-based assay. However, the cleavage pattern provides another piece of valuable information that could be used along with western blotting, for example, to understand the way specific proteins are broken down postmortem. The ability to quantify the amount of protein present in a sample is crucial to the application of the aforementioned techniques to PMI estimation. While the quantification of protein levels is possible with all of these techniques we believe that approaches that use western blotting will provide the way to developing a comparatively simple, effective, and accurate method of PMI determination. For these and other reasons, western blotting is the technique chosen most often for this line of research. Here we will summarize this approach and the results obtained by others before overviewing our research in this area. Table  14.1 lists various proteins that are potential postmortem protein markers grouped by tissue. The data were compiled mostly from English literature and only include mammalian species. Protein changes in body fluids are not included due to the reported variability in previously analyzed biochemical markers [4].

Postmortem Calmodulin-Binding Protein Degradation Calcium is one of the ions whose passage across cell membrane is tightly controlled during life. After death, the loss of cell membrane permeability leads to an influx of calcium into the cell and release of calcium from intracellular storage compartments (reviewed in [8, 74]). Inside the cells calcium works as one of the most widely spread secondary messengers and has roles in processes-like cell division, differentiation, membrane fusion, contraction and movement. Calmodulin is the primary calcium sensor and signal transducer [5]. Calcium binds to and induces a conformational change in calmodulin thereby activating it. Calmodulin then binds to and activates a variety of calmodulin-binding proteins (CaMBPs) that control a multitude of cellular processes. Calcium also activates calpain, the neutral, ubiquitous cysteine protease. Ante-mortem, calpain functions in many signaling pathways regulating cell cycle, adhesion, motility, apoptosis and differentiation (reviewed in [16]). Postmortem, calpain has been shown to be the primary proteolytic agent functioning in the first 72 h of PMI in mouse skeletal muscle, leading to degradation of cellular proteins [14]. Calpain can degrade a variety of proteins, including CaMBPs. In postmortem studies, calpain has been shown to degrade skeletal muscle cell ­proteins like metavinculin, vinculin, dystrophin, titin, troponin, nebulin, and fodrin, a CaMBP [14, 36, 63]. In addition, calmodulin binding can protect certain proteins from calpain’s proteolytic activity while rendering some more vulnerable

rat rat human

human human human human human human rat rat rat

rat rat

human

rat human

liver liver liver

liver liver liver pancreas pancreas pancreas spleen lung lung

lung lung

lung

kidney kidney

rat rat

rat

brain brain

brain

thyroid gland human thyroid gland human

Species

Tissue

tubulin b tubulin vascular endothelial growth factor bifunctional protein 3-oxoacyl-CoA thiolase acyl-CoA oxidase somatostatin insulin glucagon tubulin tubulin a Ca+2/calmodulin-dependent protein kinase II calcineurin myristoylated alanine-rich C-kinase substrate vascular endothelial growth factor tubulin vascular endothelial growth factor cacitonin thyroglobulin

tau

actin fodrin

cytoskeletal cytoskeletal signalling

cytoskeletal

petide hormone hormone precursor cytoskeletal cytoskeletal

cytoskeletal signalling

signalling

signalling signalling

enzyme enzyme enzyme peptide hormone peptide hormone peptide hormone cytoskeletal cytoskeletal signalling

Protein type

Protein name

✔

✔ ✔

✔

✔

✔ ✔

✔ ✔ ✔

✔

✔

Early PMI

✔ ✔

✔ ✔

✔ ✔

✔

✔ ✔

✔ ✔ ✔ ✔ ✔ ✔ ✔ ✔ ✔

✔ ✔

Late PMI

RT

RT RT

4°C 4°C

RT unknown

unknown

RT RT

RT RT RT 4°C 4°C 4°C RT RT RT

21°C RT unkown

Wehner et al. 2001b Wehner et al. 2000; Liu et al. 2005b

Liu et al. 2008b Thaik-Oo et al. 2002

Thaik-Oo et al. 2002

Kang et al. 2003 Kang et al. 2003

Espeel et al. 1990 Espeel et al. 1990 Espeel et al. 1990 Wehner et al. 2006 Wehner et al. 1999 Wehner et al. 2001a Liu et al. 2008b Kuai et al. 2008 Kang et al. 2003

Xiao and Chen 2005 Liu et al. 2008b Thaik-Oo et al. 2002

Publication

WB Liu et al. 2008a WB, 2D Sorimachi et al. 1996; Fountoulakis et al. 2001 WB, IH Schwab et al. 1994; Irving et al. 1996

IH IH

WB WB

WB

WB WB

IH IH IH IH IH IH WB WB WB

WB WB WB

Temperature Method

Table 14.1  A compilation of the available data on the postmortem protein degradation in several tissues of several mammalian species

rat rat rat rat

rat rat rat rat rat rat rat rat mouse mouse mouse mouse human

rat mouse

rat human

rat mouse human mouse human human human human

brain brain brain brain

brain brain brain brain brain brain brain brain brain brain brain brain brain

brain

brain

brain brain

brain brain brain

brain

Species

Tissue

stathmin ezrin internexin alpha

dihydropyrimidinase-like2

glial fibrillary acidic protein glial fibrillary acidic protein

protein phosphatase 2A

neurofilament

microtubule-associated proteins tubulin 14-3-3 protein Rab GDP dissociation inhibitor alpha synaptotagmin turned on after division G protein a subunits vascular endothelial growth factor vasopressin albumin transferrin fetoprotein alpha villin 2 vessicle-fusing ATPase calpastatin glycogen synthase kinase 3 neurofilament

Protein name

cytoskeletal cytoskeletal cytoskeletal

enzyme

cytoskeletal cytoskeletal

signalling

cytoskeletal

signalling signalling signalling signalling peptide hormone carrier carrier carrier cytoskeletal enzyme signalling signalling cytoskeletal

cytoskeletal cytoskeletal signalling signalling

Protein type

✔ ✔ ✔

✔

✔

✔

✔ ✔ ✔ ✔ ✔ ✔

✔ ✔ ✔ ✔

✔

Early PMI

✔ ✔

✔

✔

✔ ✔ ✔ ✔ ✔ ✔ ✔

✔ ✔ ✔

Late PMI

RT RT RT

RT, 25°C

4°C RT, 25°C

4°C, RT

RT, 25°C

RT RT RT unknown RT RT RT RT 25°C 25°C RT RT 4°C

RT RT RT RT

Schwab et al. 1994 Fountoulakis et al. 2001 Fountoulakis et al. 2001 Fountoulakis et al. 2001

Publication

Fountoulakis et al. 2001 Fountoulakis et al. 2001 Li et al. 1996 Thaik-Oo et al. 2002 Van Zwieten et al. 1991 Fountoulakis et al. 2001 Fountoulakis et al. 2001 Fountoulakis et al. 2001 Hunsucker et al. 2008 Hunsucker et al. 2008 Goni-Oliver et al. 2009 Goni-Oliver et al. 2009 Grange-Midroit et al. 2002; Crecelius et al. 2008 2D Fountoulakis et al. 2001; Hunsucker et al. 2008 WB, 2D Fountoulakis et al. 2001; Grange-Midroit et al. 2002 IH Wehner et al. 2006 2D Crecelius et al., 2008; Hunsucker et al. 2008 2D Franzén et al. 2003; Crecelius et al. 2003; Hunsucker et al. 2008 2D Crecelius et al. 2008 2D Crecelius et al. 2008 2D Crecelius et al. 2008 (continued)

2D 2D WB WB WB 2D 2D 2D 2D 2D WB WB WB, 2D

WB, IH 2D 2D 2D

Temperature Method

Species

human

human human human

human human human human human rat cow human rat

rat mouse pig pig pig pig pig pig

pig pig pig pig pig

Tissue

brain

brain brain brain

brain brain brain brain brain cr.muscle cr.muscle sk.muscle

sk.muscle sk.muscle sk.muscle sk.muscle sk.muscle sk.muscle sk.muscle sk.muscle

sk.muscle sk.muscle sk.muscle sk.muscle sk.muscle

Table 14.1  (continued)

peroxiredoxin-1 G protein b subunit G protein coupled receptor kinases arrestin 2 b apolipoprotein A-1 precursor septin myelin basic protein-isoform 5 complexin 2 tubulin a troponin I cardiac Ca+2/calmodulin-dependent protein kinase II calcineurin calpain 2 cofilin 2 talin myozenin 1 myosin regulatory light chain myokinase eukaryotic translation initiation factor 5a phosphoglycerate kinase albumin fatty acid-binding protein hemoglobin enolase 1

synuclein beta

Protein name

signalling carrier carrier carrier enzyme

signalling signalling cytoskeletal cytoskeletal cytoskeletal cytoskeletal signalling signalling

signalling carrier scaffolding other other cytoskeletal cytoskeletal signalling

cytoskeletal, chaperone signalling signalling signalling

Protein type

✔

✔

✔ ✔ ✔ ✔ ✔ ✔ ✔

✔

✔

Early PMI

✔ ✔ ✔ ✔ ✔

✔ ✔ ✔ ✔

✔ ✔

✔ ✔ ✔

✔

✔ ✔

Late PMI

4°C 4°C 4°C 1°C 4°C

RT 4°C 1°C unknown 4°C 4°C 4°C 4°C

4°C 4°C, RT RT RT RT 3–12°C, 20°C RT RT

RT 4°C 4°C

RT

2D 2D 2D 2D 2D

WB WB 2D WB 2D 2D 2D 2D

WB 2D 2D 2D 2D WB, 2D WB WB

2D WB WB

2D

Temperature Method

Lametsch et al. 2003 Park et al. 2007 Park et al. 2007 Hwang et al. 2005 Lametsch et al. 2003

Kang et al. 2003 Geesink et al. 2005 Hwang et al. 2005 Bee et al. 2007 Park et al. 2007 Park et al. 2007 Lametsch et al. 2003 Lametsch et al. 2003

Grange-Midroit et al. 2002 Crecelius et al. 2008 Crecelius et al. 2008 Crecelius et al. 2008 Crecelius et al. 2008 Jia et al. 2006b; Kuai et al. 2008 Sabucedo and Furton 2003 Kang et al. 2003

Crecelius et al. 2008 Grange-Midroit et al. 2002 Grange-Midroit et al. 2002

Crecelius et al. 2008

Publication

pig lamb cow pig cow

lamb cow lamb cow

nebulin mouse lamb cow sheep

mouse lamb troponin T cow

sk.muscle

sk.muscle sk.muscle

sk.muscle

sk.muscle

sk.muscle

lamb cow cow cow cow cow pig pig cow pig cow

sk.muscle sk.muscle sk.muscle sk.muscle sk.muscle sk.muscle sk.muscle sk.muscle sk.muscle sk.muscle

cytoskeletal cytoskeletal

actinin a titin

cytoskeletal

cytoskeletal

cytoskeletal

cytoskeletal

enzyme enzyme, signalling signalling cytoskeletal cytoskeletal cytoskeletal signalling enzyme cytoskeletal cytoskeletal cytoskeletal cytoskeletal

Protein type

myosin heavy chain

myosin heavy chain

ATP synthase glyceraldehyde 3-phosphate dehydrogenase calpastatin filamin tubulin a tropomyosin 14-3-3 protein ADP ribosylhydrolase actin actin actin cofilin 1

pig pig

sk.muscle sk.muscle

Protein name

Species

Tissue

✔

✔

✔

✔

✔ ✔ ✔

✔ ✔ ✔ ✔

✔

Early PMI

✔

✔

✔ ✔

✔

✔

✔ ✔

✔

✔ ✔

Late PMI

WB WB 2D 2D 2D 2D 2D 2D 2D 2D

Publication Hwang et al. 2005 Park et al. 2007

Geesink and Koohmaraie 1999 Huff-Lonergan et al. 1996a Jia et al. 2007 Jia et al. 2006a,b Jia et al. 2006a,b Jia et al. 2007 Hwang et al. 2005 Lametsch et al. 2003 Jia et al. 2007 Lametsch et al. 2003; Jia et al. 2006b; Jia et al. 2007 4°C WB, 2D Bandman and Zdanis 1998; Geesink and Koohmaraie 1999; Lametsch et al. 2008 4°C, 37°C WB, 2D Bandman and Zdanis 1998; Lametsch et al. 2003 4°C WB Ho et al. 1996; Geesink and Koohmaraie 1999 4°C WB, 2D Bandman and Zdanis 1998; Fritz and Greaser 1991; Taylor et al. 1995; Huff-Lonergan et al. 1995, 1996b; Ho et al. 1997; Geesink and Koohmaraie 1999; Lametsch et al. 2003 4°C WB Fritz and Greaser 1991; Huff-Lonergan et al. 1995; Taylor et al. 1995; Ho et al. 1997; Geesink and Koohmaraie 1999; Ilian et al. 2004; Geesink et al. 2006, 2005 2, 4, 16, 37°C WB Ho et al. 1994; Huff-Lonergan et al. 1996a; Geesink and Koohmaraie 1999; Rhee et al. 2000; Rowe et al. 2004; Geesink et al. 2005, 2006 (continued)

4°C 4°C 4–10°C 3–12°C, NA 3–12°C, NA 4–10°C 1°C 4°C 4–10°C 3–12°C

2D 2D

Temperature Method 1°C 4°C

pig lamb cow

mouse pig lamb cow

cow sheep

mouse cow

mouse lamb cow mouse lamb cow sheep mouse sheep human human human cow human

sk.muscle sk.muscle

sk.muscle sk.muscle

sk.muscle

sk.muscle sk.muscle

sk.muscle

vinculin

desmin desmin

desmin

dystophin desmin

troponin T dystophin

Protein name

cytoskeletal

cytoskeletal cytoskeletal

cytoskeletal

cytoskeletal cytoskeletal

cytoskeletal cytoskeletal

Protein type

✔

✔

✔

✔

Early PMI

✔

✔

✔

✔

✔ ✔

Late PMI

4°C 2, 16 and 37°C 4°C

1, 4–5°C

4°C 1, 4–5°C

Publication

WB

Hwang et al. 2005; Park et al. 2007 Taylor et al. 1995; Huff-Lonergan et al. 1996a; Geesink and Koohmaraie 1999; Geesink et al. 2006 WB Geesink et al. 2005, 2006 WB, 2D Taylor et al. 1995; Ho et al. 1996; Geesink and Koohmaraie 1999; Melody et al. 2003; Hwang et al., 2005; Bee et al. 2007; Park et al. 2007 WB Huff-Lonergan et al. 1996a; Ilian et al. 2004; Rowe et al. 2004; Veiseth et al. 2004 WB Geesink et al. 2005, 2006 WB Rhee et al. 2000

2D WB

Temperature Method 1, 4°C 4°C

Taylor et al. 1995; Geesink and Koohmaraie 1999; Geesink et al. 2006 calpain 1 signalling ✔ ✔ 1°C, 4°C WB Geesink and Koohmaraie 1999; Ilian sk.muscle et al. 2004; Rowe et al. 2004; Vieseth et al. 2004; Geesink et al. 2005 sk.muscle calpain 3 signalling ✔ 4°C WB Geesink et al. 2005 sk.muscle calpain 4 signalling ✔ ✔ 4°C WB Ilian et al. 2004 sk.muscle troponin I cytoskeletal ✔ ✔ 4°C WB Zheng et al. 2006 sk.muscle creatine kinase signalling ✔ ✔ 4°C, 25°C 2D Tavichakorntrakool et al. 2008 sk.muscle myoglobin carrier ✔ 3–12°C, 25°C 2D Tavichakorntrakool et al. 2008 sk.muscle heat shock 27kDa protein chaperone, ✔ 3–12°C, 25°C 2D Jia et al. 2006a,2006b; Tavichakorntrakool signalling et al. 2008 sk.muscle pig cow heat shock 27kDa protein chaperone, ✔ 1, 4°C 2D Hwang et al. 2005; Park et al. 2007; signalling Morzel et al. 2008 Proteins that have the potential to be used for the estimation of early vs. late PMI are indicated. Proteins are classified into functional groups. Temperature of the PMI is indicated (RT = room temperature) along with the method that was used to analyze the protein degradation patterns (2D = 2-dimensional gel electrophoresis; IH = immunohistochemistry; WB = western blotting). Some of the contributors to this research are listed in the last column

Species

Tissue

Table 14.1  (continued)

14  The Use of Protein Markers for the Estimation of the Postmortem Interval

285

[33, 61, 76]. Thus, there is an interaction between calcium, calmodulin, calpain and CaMBPs in cells postmortem. Our laboratory has focused on analyzing the postmortem degradation of CaMBPs in mice and rats in hopes of finding protein markers of specific tissues whose postmortem degradation pattern could be used to estimate PMI in forensics [35, 52]. Probing of postmortem tissues of rats revealed a diversity of calcium-dependent CaMBPs [35]. Tissues that were analyzed include the heart, lung, skeletal muscle, brain, kidney, liver and spleen. Each tissue revealed a unique complement of CaMBPs and differential expression profiles after death. From this study, skeletal muscle and lung were chosen for further analysis. Skeletal muscle is a dense tissue that is isolated from the sources of contamination and is easily accessible for examination. On the other hand, lung is a soft tissue that is more easily accessible to forces of putrefaction and environmental influences. All protein analysis in rats was performed at 21°C. To analyze the effect of temperature on the postmortem protein degradation mice were subjected to three different temperatures, namely 4, 10 and 21°C. The CaMBPs that were chosen for in-depth postmortem analysis include myristoylated alanine-rich C-kinase substrate (MARCKS), calcium/calmodulin-dependent kinase II (CaMKII), inducible nitric oxide synthase (iNOS), and calcineurin (CnA). The postmortem degradation patterns of these proteins had not been previously analyzed. MARCKS is a signaling protein that translocates between the plasma membrane and the cytosol depending on its phosphorylation status [70]. It is widely spread and functions in diverse cellular processes including cell division, cell shape, motility, secretion and transmembrane transport (reviewed in [1]). CnA was shown to dephosphorylate MARCKS in the rat brain [58]. Calpain cleaves MARCKS in vitro and in differentiating myoblasts [9]. Since the postmortem degradation pattern of MARCKS had not previously been analyzed we examined it in rats and mice. MARCKS could not be detected in rat and mouse skeletal muscle so its postmortem degradation pattern was analyzed only in lung. At 21°C, in the rat and mouse lung, MARCKS degraded to approximately 10% of 0 h levels by 4 days postmortem. The degradation pattern was linear in rats but showed some variability at 24 and 48 h in mice [35, 52]. Temperature had an effect on the postmortem MARCKS degradation pattern as about 70% of the protein was still present by 4 days postmortem at 4°C and 50% at 10°C [52]. CaMKII is a serine/threonine protein kinase that regulates a variety of target proteins through phosphorylation (reviewed in [62]). CaMKII has been shown to phosphorylate and thereby deactivate CnA [46]. Ischemia-like conditions lead to changes in CaMKII localization and CaMKII clustering in the brain [7, 65, 66]. CaMKII has also been shown to play a major role in apoptosis and necrosis of cells following ischemic injury [75]. Calpain can cleave CaMKII in excitotoxic brain conditions [19]. In spite of this, the postmortem degradation pattern of CaMKII remained to be studied. In rats, CaMKII levels increased slightly over 4 days postmortem in lung but were stable in skeletal muscle [35]. In mice, CaMKII showed a very different degradation pattern. CaMKII levels became undetectable by 48  h postmortem in both tissues at 21°C. Lower temperature did not slow down CaMKII

286

Y. Poloz and D.H. O’Day

degradation as the protein was also undetectable by 48 h postmortem at 4 and 10°C [52]. Due to interspecies variability further investigation of CaMKII as a potential marker for PMI is not recommended. Inducible nitric oxide synthase (iNOS or NOS2) catalyzes the production of nitric oxide from l-arginine. Nitric oxide then functions as a signaling molecule, having diverse roles in vasodilation, the immune response, neurotransmission, and cytotoxicity. iNOS expression can be induced by ischemia and other stress conditions (reviewed in [85]). Furthermore, calcineurin signaling has been shown to enhance iNOS expression in cardiac muscle and other cells [50]. CaMKII has also been shown to regulate iNOS activity [34]. Calpain has been shown to cleave iNOS in a macrophage cell line [76]. However, the postmortem degradation pattern of iNOS was yet to be studied. iNOS showed a highly irregular and unpredictable degradation pattern over 4 days postmortem in the rat lung and was therefore not analyzed in skeletal muscle or in any mouse tissues [35]. This protein does not appear to be a good potential marker for PMI estimation in humans. CnA is a protein serine/threonine phosphatase formerly known as protein phosphatase 2B. It is composed of two subunits, a catalytic, calmodulin-binding 60 kDa subunit A (CnA) and a regulatory 19 kDa subunit B (CnB). It works in cellular signal transduction pathways by dephosphorylating target proteins. CnA is activated by calcium-bound calmodulin, CnB and proteolytic cleavage (reviewed in [48]). Calpain has been shown to cleave CnA in vitro, in ischemic tissue and in Alzheimer’s disease brains. Calmodulin also plays a critical role in calpain-mediated CnA cleavage. The three potential cleavage products of CnA localize at 57, 48, and 45 kDa in western blots. The 48 and 45 kDa cleavage products lose their calmodulin-dependence and become constitutively active while the 57 kDa product retains its dependence on calmodulin [42, 61, 82]. Overactivation of CnA, such as that occurs after cleavage, has been linked to cell death [60]. In keeping with the results of previous research we examined the postmortem CnA cleavage pattern in mice and rats. We also studied the role of proteolytic activity in regulating the observed changes. In the rat lung and skeletal muscle, CnA exhibited a change in mobility in the first 24 h postmortem. The 65 kDa band, corresponding to the intact protein, disappeared by 48 h while the faster migrating cleavage product (57 kDa) increased in intensity by 48 h and was still present at 4 days postmortem [35]. CnA showed a very similar degradation pattern in the mouse skeletal muscle (Fig. 14.1a–c). At 21°C, 65 kDa intact protein was cleaved to produce 57 kDa product within the first 24 h postmortem. Levels of the cleavage product declined to almost undetectable by 4 days postmortem. Lower temperatures slowed down degradation of the cleavage product since it was still detected by 4 days postmortem [52]. We next set out to examine the involvement of calcium, calmodulin, calpain and the proteasome in postmortem cleavage of CnA in mouse skeletal muscle (Fig. 14.1d). We used the complete protease inhibitor tablet from SantaCruz Biotechnology to inhibit all protease activity in the tissue. MDL-28170 from Biomol was used as a potent and selective calpain inhibitor while MG-132 from Biomol was used as a proteasome inhibitor. Calmidazolium from Sigma was used to antagonize calmodulin and EGTA (ethylene glycol tetra acetic acid) from Sigma was used to chelate extracellular calcium

14  The Use of Protein Markers for the Estimation of the Postmortem Interval

287

Fig. 14.1  Postmortem CnA degradation in mouse skeletal muscle. (a) Western blot quantification of the 65 kDa intact CnA over 4 days postmortem. (b) Western blot quantification of the 57 kDa cleavage product of CnA over 4 days postmortem. (c) Western blot of postmortem CnA degradation pattern. (d) The involvement of calcium, calmodulin, calpain, and proteasome in the postmortem degradation of CnA. 1 = complete protease inhibitor tablet, 2 = 1/10 complete protease inhibitor tablet, 3 = 1 mM MDL-28170, 4 = 10  mM MDL-28170, 5 = 2  mM MG-132, 6 = 20  mM MG-132, 7 = 1 mM calmidazolium, 8 = 10 mM calmidazolium, 9 = 1 mM EGTA, 10 = 10 mM EGTA, 11 = PBS only (control)

ions. Tissue incubated in PBS (phosphate-buffered saline) was used as the control. Protease inhibitor tablet blocked CnA cleavage when it was used at the recommended concentration suggesting that proteolytic enzymes are responsible for postmortem CnA cleavage. MDL-28170 showed a dose-dependent effect on CnA cleavage. At 10 mm MDL-28170 almost completely inhibited CnA cleavage suggesting that calpain is involved in postmortem CnA cleavage. MG-132 also had a dose-dependent effect on CnA cleavage. At 20  mm MG-132 completely blocked CnA cleavage, suggesting that not only calpain but also the proteasome plays an important role in the postmortem CnA cleavage. This could occur if the proteasome inhibition led to the accumulation of calpastatin, an endogenous calpain inhibitor, thereby inhibiting calpain and preventing CnA cleavage. Antagonism of calmodulin by calmidazolium and chelation of extracellular calcium by EGTA also inhibited CnA cleavage dose-dependently. This suggests that calmodulin and calcium also

288

Y. Poloz and D.H. O’Day

play an important role in the postmortem CnA cleavage [52]. The appearance of 57 kDa cleavage product of CnA could be attributed to the action of calpain 1. It appears to be the same cleavage product as was observed in tissues after ischemic damage and in Alzheimer’s disease brains [42, 82]. The common link between the above conditions is the breakdown of calcium homeostasis, which is known to occur postmortem. From the above results it is clear that calcium, calpain and calmodulin all play an important role in the postmortem CnA cleavage. It is likely that calcium activates calpain which then cleaves and activates CnA. Calmodulin is somehow necessary for calpain-mediated cleavage of CnA. It could be that calmodulin binding leads to a conformational change in CnA thereby opening up a site for calpain cleavage. This type of interaction has been demonstrated in mouse brains where the constitutively active cleavage products of CnA (45 and 48 kDa) could only be produced in the presence of calmodulin [61]. The 57 kDa cleavage product was demonstrated to have increased phosphatase activity in Alzheimer’s disease brains [42]. During the PMI the enhanced phosphatase activity could lead to dephosphorylation of critical proteins further contributing to cell death [40]. This work was the first to show the calpain-mediated CnA cleavage in postmortem tissue. It is also possible that the proteasome is involved in the ­postmortem CnA cleavage. The CSFantrogin-1 E3 ubiquitin ligase complex is known to interact with CnA in mouse cardiac and skeletal muscle in vitro (reviewed in [86]). The complex attaches ubiquitin to a lysine residue on the proteins it interacts with thereby targeting them for degradation by the proteasome. Protein phosphatase 2A (PP2A) is another protein serine/threonine phosphatase from the phosphoprotein phosphatase (PP) family that CnA belongs to (reviewed in [59]). This protein is not a CaMBP but its postmortem degradation pattern was analyzed in mice for comparison to CnA. We also studied the role of proteolytic activity in regulating the observed changes in PP2A. In lung, PP2A levels decreased to about 30% of 0 h levels by 24 h and then remained relatively stable for up to 4 days. The same pattern was observed at all temperatures. In skeletal muscle at 21°C PP2A levels increased up to 120% by 24 h and then declined to 10% by 4 days (Fig. 14.2a, b). Temperature had an effect on postmortem PP2A changes in skeletal muscle. At 10°C, PP2A levels increased up to 130% by 48 h and declined to 0 h levels by 4 days. At 4°C, PP2A was stable for 4 days postmortem [52]. The increase in PP2A levels seen at 24–48 h postmortem could possibly be attributed to the production of a more accessible epitope or to postmortem translation of PP2A mRNA [56]. Sanoudou et al. [56] have demonstrated that cells of human skeletal muscle show active gene transcription and possibly protein translation in the first 46 h postmortem. Since PP2A levels in skeletal muscle are relatively stable for the first 4 days postmortem at 4 and 10°C while declining after 48 h postmortem at 21°C, PP2A should be further examined as a potential marker for later PMI. We next set out to examine the involvement of calcium, calmodulin, calpain, and the proteasome in the postmortem cleavage of PP2A in mouse skeletal muscle (Fig. 14.2c). After treatment with the complete protease inhibitor tablet, MDL-28170, MG-132 or calmidazolium, there was an increase in PP2A levels. This suggests that calpain, proteasome, and calmodulin could be involved in PP2A degradation or in the degradation of

14  The Use of Protein Markers for the Estimation of the Postmortem Interval

289

Fig. 14.2  Postmortem PP2A changes in mouse skeletal muscle. (a) Western blot quantification of PP2A over 4 days postmortem. (b) Western blot of postmortem PP2A degradation pattern. (c) The involvement of calcium, calmodulin, calpain and proteasome in the postmortem degradation of PP2A. 1 = complete protease inhibitor tablet, 2 = 1/10 complete protease inhibitor tablet, 3 = 1 mMMDL-28170, 4 = 10 mM MDL-28170, 5 = 2 mM MG-132, 6 = 20 mM MG-132, 7 = 1 mM calmidazolium, 8 = 10  mM calmidazolium, 9 = 1  mM EGTA, 10 = 10  mM EGTA, 11 = PBS only (control)

proteins that are responsible for PP2A stabilization or translation. EGTA had no effect on PP2A levels suggesting that calcium is involved in the postmortem increase in PP2A levels [52]. Enhanced PP2A levels and activity have been demonstrated in hypoxic lung tissue and postmortem in mice and humans [39].

Recommendations for Future Research Differences in protein sequence, function, stability, interacting partners and cleavage sites lead to very unique postmortem degradation patterns. This could be exploited to develop a new, more precise biochemical method for the estimation of PMI. Research in model organisms like mice and rats can identify proteins that show significant and reproducible degradation patterns in specific tissues. Such work makes it possible to carry out large-scale studies allowing for the accumulation of statistically significant data. Manipulations of environmental conditions such as temperature, humidity, body covering, etc., are easily carried out giving insight into their effects on the degradation of specific proteins. Based on the results from

290

Y. Poloz and D.H. O’Day

such work, correction factors could be developed for the environmental influences on the degradation of these proteins. Consequently, this type of research will ultimately reveal a group of potential PMI markers that could then be further analyzed in human postmortem tissues. The proteins that should be chosen for the human studies should be highly conserved and ubiquitously expressed. CnA, for example, is present in most human tissues and CnA in mice shares 99% sequence identity with CnA in humans. Our research has identified CnA as a potential marker for the estimation of early PMI and PP2A for the estimation of late PMI. We propose skeletal muscle as the tissue of choice for PMI estimation for several reasons. It is highly isolated and therefore relatively removed from the forces of putrefaction. It presents low risk of contamination. Furthermore, it is easy to harvest and there are several sites available for harvest depending on the postmortem condition of the body. It is also evident that there is already a wealth of data available on the skeletal muscle protein degradation from journals like Meat Science where the research focuses on finding ways to ­create more tender meat and has not been previously linked to PMI estimation.

The Long-Term Goal: Development of a Device to Measure the PMI via Protein Markers One of the main reasons that new biochemical methods for PMI estimation have not been developed is due to the inability of the methods to produce results quickly and at the scene of the crime. We have previously proposed the development of a handheld device, much like the pregnancy test unit. The device would have specific antibodies embedded in specific locations on a test strip. A ground up tissue sample would be applied to the input slot of the strip. The proteins from the tissue would be carried over the embedded antibody slots via buffer containing specific protease inhibitors. The antibodies would bind to their specific target proteins and the reaction would be visualized via the formation of color or another reporter method. The presence vs. absence of specific bands on the strip would then be compared to reference data to generate a precise PMI. Coupling this with other traditional field measurements (e.g., core body temperature) could increase the significance of the time since death determination. We have made the first step to reach this goal by showing that protein degradation patterns are not altered when the centrifugation step is removed during sample preparation. Centrifugation is a time-consuming step that requires comparatively expensive and heavy equipment. When this step was removed from our protocol the degradation pattern of CnA did not change. Without the need for centrifugation the postmortem protein degradation could be analyzed at the scene of the crime. The antibodies are relatively inexpensive and commercially available for all of the proteins detailed in this chapter. In its simplest form, the PMI test kit could contain a mini grinder (to grind up the tissue sample), the strip, buffer and an eye dropper (for tissue application). It is also possible that a mixture of

14  The Use of Protein Markers for the Estimation of the Postmortem Interval

291

digestive enzymes for tissue breakdown could be employed rather than a tissue grinder but this research remains to be done.

References 1. Arbuzova A, Schmitz AAP, Vergères G (2002) Cross-talk unfolded: MARCKS proteins. Biochem J 362(1):1–12 2. Bandman E, Zdanis D (1988) An immunological method to assess protein degradation in postmortem muscle. Meat Sci 22(1):1–19 3. Bee G, Anderson AL, Lonergan SM et al (2007) Rate and extent of pH decline affect proteolysis of cytoskeletal proteins and water-holding capacity in pork. Meat Sci 76(2):359–365 4. Coe JI (1993) Postmortem chemistry update: emphasis on forensic application. Am J Forensic Med Pathol 14:91–117 5. Cohen P, Klee CB (1988) Calmodulin. Elsevier, Amsterdam 6. Crecelius A, Götz A, Arzberger T et al (2008) Assessing quantitative post-mortem changes in the gray matter of the human frontal cortex proteome by 2-D DIGE. Proteomics 8(6):1276–1291 7. Dosemeci A, Reese TS, Petersen J et al (2000) A novel particulate form of Ca2+/CaMKIIdependent protein kinase II in neurons. J Neurosci 20(9):3076–3084 8. Dong Z, Saikumar P, Weinberg JM et al (2006) Calcium in cell injury and death. Annu Rev Pathol 1:405–434 9. Dulong S, Goudenege S, Vuillier-Devillers K et al (2004) Myristoylated alanine-rich C kinase substrate (MARCKS) is involved in myoblast fusion through its regulation by protein kinase Ca and calpain proteolytic cleavage. Biochem J 382(3):1015–1023 10. Fountoulakis M, Hardmeier R, Höger H et al (2001) Postmortem changes in the level of brain proteins. Exp Neurol 167(1):86–94 11. Franzén B, Yang Y, Sunnemark D et  al (2003) Dihydropyrimidinase related protein-2 as a biomarker for temperature and time dependent post mortem changes in the mouse brain proteome. Proteomics 3(10):1920–1929 12. Fritz JD, Greaser ML (1991) Changes in titin and nebulin in postmortem bovine muscle as revealed by gel electrophoresis, western blotting and immunofluorescence microscopy. J Food Sci 56:607–615 13. Geesink GH, Koohmaraie M (1999) Effect of calpastatin on degradation of myofibrillar proteins by m-calpain under postmortem conditions. J Anim Sci 77(10):2685–2692 14. Geesink GH, Kuchay S, Chishti AH (2006) m-Calpain is essential for postmortem proteolysis of muscle proteins. J Anim Sci 84(10):2834–2840 15. Geesink GH, Taylor RG, Koohmaraie M (2005) Calpain 3/p94 is not involved in postmortem proteolysis. J Anim Sci 83(7):1646–1652 16. Goll DE, Thompson VF, Li H et al (2003) The calpain system. Physiol Rev 83(3):731–801 17. Goni-Oliver P, Avila J, Hernandez F (2009) Calpain-mediated cleavage of GSK-3 in postmortem brain samples. J Neurosci Res 87:1156–1161 18. Grange-Midroit M, García-Sevilla JA, Ferrer-Alcón M et al (2002) G protein-coupled receptor kinases, b-arrestin-2 and associated regulatory proteins in the human brain: Postmortem changes, effect of age and subcellular distribution. Mol Brain Res 101(1–2):39–51 19. Hajimohammadreza I, Raser KJ, Nath R et al (1997) Neuronal nitric oxide synthase and calmodulin-dependent protein kinase IIa undergo neurotoxin-induced proteolysis. J Neurochem 69(3):1006–1013 20. Ho CY, Stromer MH, Robson RM (1994) Identification of the 30  kDa polypeptide in post mortem skeletal muscle as a degradation product of troponin-T. Biochimie 76(5):369–375

292

Y. Poloz and D.H. O’Day

21. Ho C, Stromer MH, Robson RM (1996) Effect of electrical stimulation on postmortem titin, nebulin, desmin, and troponin-T degradation and ultrastructural changes in bovine longissimus muscle. J Anim Sci 74(7):1563–1575 22. Ho C, Stromer MH, Rouse G et al (1997) Effects of electrical stimulation and postmortem storage on changes in titin, nebulin, desmin, troponin-T, and muscle ultrastructure in bos indicus crossbred cattle. J Anim Sci 75(2):366–376 23. Huff-Lonergan E, Mitsuhashi T, Beekman DD et  al (1996a) Proteolysis of specific muscle structural proteins by m-calpain at low pH and temperature is similar to degradation in postmortem bovine muscle. J Anim Sci 74(5):993–1008 24. Huff-Lonergan E, Mitsuhashi T, Parrish FC Jr et al (1996b) Sodium dodecyl sulfate-polyacrylamide gel electrophoresis and western blotting comparisons of purified myofibrils and whole muscle preparations for evaluating titin and nebulin in postmortem bovine muscle. J Anim Sci 74(4):779–785 25. Huff-Lonergan E, Parrish FC Jr, Robson RM (1995) Effects of postmortem aging time, animal age, and sex on degradation of titin and nebulin in bovine longissimus muscle. J Anim Sci 73(4):1064–1073 26. Hunsucker SW, Solomon B, Gawryluk J et  al (2008) Assessment of post-mortem-induced changes to the mouse brain proteome. J Neurochem 105(3):725–737 27. Hwang IH, Park BY, Kim JH et al (2005) Assessment of postmortem proteolysis by gel-based proteome analysis and its relationship to meat quality traits in pig longissimus. Meat Sci 69(1):79–91 28. Ilian MA, Bekhit AE, Bickerstaffe R (2004) The relationship between meat tenderization, myofibril fragmentation and autolysis of calpain 3 during post-mortem aging. Meat Sci 66(2):387–397 29. Irving EA, Nicoll J, Graham DI et al (1996) Increased tau immunoreactivity in oligodendrocytes following human stroke and head injury. Neurosci Lett 213(3):189–192 30. Jia X, Ekman M, Grove H, Færgestad EM et al (2007) Proteome changes in bovine longissimus thoracis muscle during the early postmortem storage period. J Proteome Res 6(7):2720–2731 31. Jia X, Hildrum KI, Westad F et al (2006a) Changes in enzymes associated with energy metabolism during the early post mortem period in longissimus thoracis bovine muscle analyzed by proteomics. J Proteome Res 5(7):1763–1769 32. Jia X, Hollung K, Therkildsen M et al (2006b) Proteome analysis of early post-mortem changes in two bovine muscle types: M. longissimus dorsi and M. semitendinosis. Proteomics 6(3):936–944 33. Johnson GVW, Greenwood JA, Costello AC et al (1991) The regulatory role of calmodulin in the proteolysis of individual neurofilament proteins by calpain. Neurochem Res 16(8):869–873 34. Jones RJ, Jourd’heuil D, Salerno JC et  al (2007) iNOS regulation by calcium/calmodulindependent protein kinase II in vascular smooth muscle. Am J Physiol Heart Circ Physiol 292(6):2634–2642 35. Kang S, Kassam N, Gauthier ML et al (2003) Post-mortem changes in calmodulin binding proteins in muscle and lung. Forensic Sci Int 131(2–3):140–147 36. Koohmaraie M, Geesink GH (2006) Contribution of postmortem muscle biochemistry to the delivery of consistent meat quality with particular focus on the calpain system. Meat Sci 74(1):34–43 37. Kuai J, Liu Y, Zhang Y (2008) A study on the relationship between the degradation of tubulin in cardiac muscle and lung of rat and the postmortem interval. Chin J Forensic Med 23(2):96–98 38. Lametsch R, Karlsson A, Rosenvold K et al (2003) Postmortem proteome changes of porcine muscle related to tenderness. J Agric Food Chem 51(24):6992–6997 39. Larsen K, Lygren B, Sjaastad I et al (2008) Diastolic dysfunction in alveolar hypoxia: A role for interleukin-18- mediated increase in protein phosphatase 2A. Cardiovasc Res 80(1):47–54

14  The Use of Protein Markers for the Estimation of the Postmortem Interval

293

40. Li J, Gould TD, Yuan P et al (2003) Post-mortem interval effects on the phosphorylation of signaling proteins. Neuropsychopharmacology 28(6):1017–1025 41. Li X, Greenwood AF, Powers R et al (1996) Effects of postmortem interval, age, and alzheimer’s disease on G-proteins in human brain. Neurobiol Aging 17(1):115–122 42. Liu F, Iqbal-Grundke I, Iqbal K et al (2005a) Truncation and activation of calcineurin A by calpain I in alzheimer disease brain. J Biol Chem 280(45):37755–37762 43. Liu L, Cheng J, Wang F (2005b) An experimental study on relationship between degradation of thyroglobulin and postmortem interval by immunohistochemistry. Chin J Forensic Med 20(5):265–267 44. Liu Y, Kuai J-X, Zhang Y-W et al (2008a) The relationship between the degradation of actin and the postmortem interval in rats. J Forensic Med 24(3):165–167 45. Liu Y, Kuai J-X, Zhang Y-W (2008b) A study on the content changes of tubulin in liver, spleen and kidney of rats killed by bleeding. Chin J Forensic Med 23(5):315–317 46. MacDonnell SM, Weisser-Thomas J, Kubo H et al (2009) CaMKII negatively regulates calcineurin-NFAT signaling in cardiac myocytes. Circ Res 105(4):316–325 47. Melody JL, Lonergan SM, Rowe LJ et al (2003) Early postmortem biochemical factors influence tenderness and water-holding capacity of three porcine muscles. J Anim Sci 82(4):1195–1205 48. Morioka M, Hamada J, Ushio Y et al (1999) Potential role of calcineurin for brain ischemia and traumatic injury. Prog Neurobiol 58(1):1–30 49. Morzel M, Terlouw C, Chambon C et al (2008) Muscle proteome and meat eating qualities of longissimus thoracis of “blonde d’aquitaine” young bulls: A central role of HSP27 isoforms. Meat Sci 78(3):297–304 50. Obasanjo-Blackshire K, Mesquita R, Jabr RI et  al (2006) Calcineurin regulates NFATdependent iNOS expression and protection of cardiomyocytes: Co-operation with src tyrosine kinase. Cardiovasc Res 71(4):672–683 51. Park BY, Kim NK, Lee CS et al (2007) Effect of fiber type on postmortem proteolysis in longissimus muscle of landrace and korean native black pigs. Meat Sci 77(4):482–491 52. Poloz YO, O’Day DH (2009) Determining time of death: temperature-dependent postmortem changes in calcineurin A, MARCKS, CaMKII, and protein phosphatase 2A in mouse. Int J Leg Med 123(4):305–314 53. Rhee MS, Ryu YC, Imm JY et al (2000) Combination of low voltage electrical stimulation and early postmortem temperature conditioning on degradation of myofibrillar proteins in korean native cattle (hanwoo). Meat Sci 55(4):391–396 54. Rowe LJ, Maddock KR, Lonergan SM et al (2004) Oxidative environments decrease tenderization of beef steaks through inactivation of m-calpain. J Anim Sci 82(11):3254–3266 55. Sabucedo AJ, Furton KG (2003) Estimation of postmortem interval using the protein marker cardiac troponin I. Forensic Sci Int 134(1):11–16 56. Sanoudou D, Kang PB, Haslett JN et al (2004) Transcriptional profile of postmortem skeletal muscle. Physiol Genomics 16:222–228 57. Schwab C, Bondada V, Sparks DL et al (1994) Postmortem changes in the levels and localization of microtubule-associated proteins (tau, MAP2 and MAP1B) in the rat and human hippocampus. Hippocampus 4(2):210–225 58. Seki K, Chen H, Huang K (1995) Dephosphorylation of protein kinase C substrates, neurogranin, neuromodulin, and MARCKS, by calcineurin and protein phosphatases 1 and 2A. Arch Biochem Biophys 316(2):673–679 59. Shi Y (2009) Serine/threonine phosphatases: mechanism through structure. Cell 139(3):468–484 60. Shibasaki F, McKeon F (1995) Calcineurin functions in Ca2+−activated cell death in mammalian cells. J Cell Biol 131(3):735–473 61. Shioda N, Moriguchi S, Shirasaki Y et al (2006) Generation of constitutively active calcineurin by calpain contributes to delayed neuronal death following mouse brain ischemia. J Neurochem 98(1):310–320 62. Skelding KA, Rostas JAP (2009) Regulation of CaMKII in vivo: the importance of targeting and the intracellular microenvironment. Neurochem Res 34(10):1792–1804

294

Y. Poloz and D.H. O’Day

63. Sorimachi Y, Harada K, Yoshida K (1996) Involvement of calpain in postmortem proteolysis in the rat brain. Forensic Sci Int 81(2–3):165–174 64. Sturner WQ (1963) The vitreous humour: postmortem potassium changes. Lancet 1:807–808 65. Tao-Cheng J, Gallant PE, Brightman MW et al (2007) Structural changes at synapses after delayed perfusion fixation in different regions of the mouse brain. J Comp Neurol 501(5):731–740 66. Tao-Cheng J, Vinade L, Smith C et al (2001) Sustained elevation of calcium induces Ca2+/ calmodulin-dependent protein kinase II clusters in hippocampal neurons. Neuroscience 106(1):69–78 67. Tavichakorntrakool R, Prasongwattana V, Sriboonlue P et al (2008) Serial analyses of postmortem changes in human skeletal muscle: a case study of alterations in proteome profile, histology, electrolyte contents, water composition, and enzyme activity. Proteomics Clin Appl 2(9):1255–1264 68. Taylor RG, Geesink GH, Thompson VF et  al (1995) Is Z-disk degradation responsible for postmortem tenderization? J Anim Sci 73(5):1351–1367 69. Thaik-Oo M, Tanaka E, Tsuchiya T et  al (2002) Estimation of postmortem interval from hypoxic inducible levels of vascular endothelial growth factor. J Forensic Sci 47(1):186–189 70. Thelen M, Rosen A, Nairn AC et al (1991) Regulation by phosphorylation of reversible association of a myristoylated protein kinase C substrate with the plasma membrane. Nature 351(6324):320–322 71. Van Zwieten EJ, Ravid R, Van Der Sluis PJ et al (1991) Increased vasopressin immunoreactivity in the rat brain after a postmortem interval of 6 hours. Brain Res 550(2):263–267 72. Vass AA, Barshick S, Sega G et al (2002) Decomposition chemistry of human remains: a new methodology for determining the postmortem interval. J Forensic Sci 47(3):542–553 73. Veiseth E, Shackelford SD, Wheeler TL et al (2004) Indicators of tenderization are detectable by 12 h postmortem in ovine longissimus. J Anim Sci 82(5):1428–1436 74. Verkhratsky A (2007) Calcium and cell death. Subcell Biochem 45:465–480 75. Vila-Petroff M, Salas MA, Said M et al (2007) CaMKII inhibition protects against necrosis and apoptosis in irreversible ischemia-reperfusion injury. Cardiovasc Res 73(4):689–698 76. Walker G, Pfeilschifter J, Otten U et al (2001) Proteolytic cleavage of inducible nitric oxide synthase (iNOS) by calpain I. Biochim Biophys Acta Gen Subj 1568(3):216–224 77. Wehner F, Steinriede A, Martin D (2006) Two-tailed delimitation of the time of death by immunohistochemical detection of somatostatin and GFAP. Forensic Sci Med Pathol 2(4):241–248 78. Wehner F, Wehner H, Schieffer MC et al (2000) Delimitation of the time of death by immunohistochemical detection of thyroglobulin. Forensic Sci Int 110(3):199–206 79. Wehner F, Wehner H, Schieffer MC et al (1999) Delimitation of the time of death by immunohistochemical detection of insulin in pancreatic b-cells. Forensic Sci Int 105(3):161–169 80. Wehner F, Wehner H, Subke J (2001a) Delimitation of the time of death by immunohistochemical detection of glucagon in pancreatic a-cells. Forensic Sci Int 124(2–3):192–199 81. Wehner F, Wehner H, Subke J (2001b) Delimitation of the time of death by immunohistochemical detection of calcitonin. Forensic Sci Int 122(2–3):89–94 82. Wu H, Tomizawa K, Oda Y et al (2004) Critical role of calpain-mediated cleavage of calcineurin in excitotoxic neurodegeneration. J Biol Chem 279(6):4929–4940 83. Xiao JH, Chen YC (2005) A study on the relationship between the degradation of protein and the postmortem interval. Fa Yi Xue Za Zhi 21(2):110–112 84. Zheng X, Zhang Y, Zhi X (2006) Estimation of postmortem interval by determination of troponin I using western blot technique in human pectoralis major. Chin J Forensic Med 21(3):146–148 85. Zhou L, Zhu D (2009) Neuronal nitric oxide synthase: Structure, subcellular localization, regulation, and clinical implications. Nitric Oxide Biol Chem 20(4):223–230 86. Zolk O, Schenke C, Sarikas A (2006) The ubiquitin-proteasome system: focus on the heart. Cardiovasc Res 70(3):410–421 87. Espeel M, Hashimoto T, De Craemer D, Roels F (1990) Immunocytochemical detection of peroxisomal beta-oxidation enzymes in cryostat and paraffin sections of human post mortem liver. Histochem J 22(1):57–62

Chapter 15

Alcohol and Drug Fatalities in Transportation: Forensic-Toxicological Implications F. Mußhoff

Abstract  Alcohol, drugs and medicines are recognized as leading factors for traffic accidents even with fatal outcome. Epidemiological data as well as basic information about alcohol, drugs and medicines are given, especially concerning driving effects. Finally, forensic-toxicological aspects concerning drug fatalities in transportation are discussed. Keywords  Forensic toxicology • Drugs • Alcohol • Toxicological analysis

Introduction In general, alcohol is recognized as leading factor contributing to motor vehicle crashes. Several studies clearly demonstrated an over-involvement of alcohol in crashes. Compared to alcohol less evidence is available about the accident risk caused by other drugs. In recent years, it has been clearly demonstrated that driving under the influence (DUI) of drugs and medicines also plays an important role, especially in (fatal) traffic accidents. Therefore, a comprehensive chemicaltoxicological analysis should be performed in every case of traffic accident. Forensic-toxicological analysis and interpretation of results can be complicated in post-mortem cases. Special features and recommendations are given.

F. Mußhoff (*) Department of Toxicology, Institute of Legal Medicine, Universitätsklinikum Bonn, Stiftsplatz 12, 53111 Bonn, Germany e-mail: [email protected] E.E. Turk (ed.), Forensic Pathology Reviews, Volume 6, Forensic Pathology Reviews 6, DOI 10.1007/978-1-61779-249-6_15, © Springer Science+Business Media, LLC 2011

295

296

F. Mußhoff

Epidemiological Data Krüger et al. [1] summarized roadside test results about alcohol and drugs in traffic of 69 epidemiological studies (Table  15.1). Compared to a random sampling of drivers the mean percentage of positives for alcohol, illicit drugs and medicines in injured or deceased drivers was 7, 15 and 2.5 times higher, respectively. Also De Gier (cited according to [2]) could show that the rate of drivers impaired by substances is considerably higher for the injured and fatalities than the rate revealed in roadside tests (Table 15.2). Cannabis and opiates are the most frequently observed substances in the general driver population. In some countries (e.g. Norway) the number of drivers impaired by amphetamines is increasing. Within the group of licit drugs benzodiazepines are most frequently found [2]. Another way to determine the prevalence of drug use in the general population is roadside surveys where drivers are stopped on the road and tested for drugs and alcohol. Maes et al. [2] critically pointed out that there are different ways of data collection and analysis methods, especially the individual disposition regarding consumption behaviour. A further way to get data is to test drivers that are suspected to drive under influence (DUI) (for summaries [2, 3]). Analyzing this group, a higher prevalence of licit and illicit drugs can be found but detection of this group depends on the perception of police officers. Cannabis and opiates are found in 57 and 42%, amphetamines in 21%. There are remarkable differences between countries, possibly because there exist different national road traffic acts and different attention is paid to the problem. Epidemiological studies demonstrate the frequency of specific substances in traffic but do not explore the real impact of drugs on accidents. Studies about responsibility analysis might give answers. In ROSITA, Maes et al. [2] reported a study by Drummer [4] who investigated data of 1,045 killed drivers (Table 15.3). Responsibility was defined by eight factors (responsibility index without knowing the result of the drug analysis) and revealed in three groups of drivers: culpable drivers, contributory drivers and inculpable drivers. The culpability ratio was calculated and drivers that Table 15.1  Median exposure rate from 69 epidemiological studies [1] Roadside (% positives) Injured (% positives) Fatalities (% positives) Drugs 1 17 19 Medicines 4 13 10 Alcohol 6 35 52

Table 15.2  Prevalence of drugs in different driver populations (De Gier according to [2]) General driver population Collision-involved drivers Illicit drugs 1–5% 10–25% Legal drugs 5–15% 6–21%

15  Alcohol and Drug Fatalities in Transportation: Forensic-Toxicological Implications

297

Table 15.3  Responsibility analysis results in fatal accidents [4] Relative risk Relative risk Relative risk Drug group Prevalence (all cases) (drug alone) (drug + alcohol) Drug free 51% 1.0 Alcohol 27% 6.0 Alcoho + drugs 9% 9.0 Drugs 13% 1.4 Cannabis 11% 1.6 0.6 5.6 Stimulants 3.7% 2.7 1.6 8.7 Opiates 2.7% 5.0 2.3 2.9 Benzodiazepines 3.1% 5.8 1.9 9.5 Misc. drugs 5.6% 4.0 8.7 Bold = statistically significant Table 15.4  Responsibility analysis results for fatally injured drivers [5] Drug groups Responsible (substance only) Responsible (drug + alcohol) Drug free 67.7% (=reference group) Alcohol <1 g/l ³1 g/l

75.8% 93.9%

Cannabis THC + THCCOOH THCCOOH

57.9% 83.3%

94.6% 93.1%

Cocaine Benzodiazepines Amphetamines

57.1% 66.7% 83.3%

87.8% 100% 91.7%

Bold = statistically significant

were positive for substances showed a higher culpability ratio. Drivers with high drug concentrations were either culpable or contributory to the accident. Persons driving under the influence of more than one substance were found to be culpable of the accident in every case. Terhune et  al. [5] examined data of 1,882 fatally injured drivers in the USA (Table 15.4). They found that the culpability ratio increased significantly when only alcohol or alcohol–drug combinations were consumed. The responsibility rate for drivers with delta-9-tetrahydrocannabinol (THC, an active cannabis agent) in blood decreased compared to a drug-free control group. In contrast, the responsibility rate for amphetamine-positive drivers was higher than the drug-free group. Crash responsibility rates increased significantly the way the number of non-alcohol drugs in a driver’s blood increased. Drummer [4] found higher culpability ratios in drivers that were positive for benzodiazepines, whereas Terhune [5] found no difference. However, both studies showed statistically significantly higher culpability when benzodiazepines and alcohol were combined. Longo et al. [6] found that drivers tested positive for alcohol only, for benzodiazepines only, for the combination of alcohol and THC and for alcohol combined with

298

F. Mußhoff

benzodiazepines were significantly more likely responsible for a car crash than those of a drug-free group. For THC only (with increasing concentration) the authors reported no increasing risk in contrast to alcohol where highly increasing risk was reconfirmed. This study again demonstrated that combination of substances increases the accident risk. In general, responsibility analysis suggested little relation between drug use and crash risk but sample sizes were small. Obviously, there is some potential for increased crash risk when specific drugs were combined with alcohol. As for medicines also pharmaco-epidemiological studies give very useful information and showed an increasing risk for accident involvement in cases of DUI. Especially, elderly drivers have a relative risk of injuring crash involvement if they are under medication (benzodiazepines, antidepressants and others) which is dosedependent markedly enhanced [2]. Recently, Raes et al. [3] summarized actual epidemiological data and classified it in as follows: • Roadside surveys • Subsets of drivers –– –– –– –– ––

Drivers injured in traffic accidents Drivers killed in traffic accidents Drivers involved in traffic accidents Drivers suspected of driving under the influence of drugs Drivers suspected of driving under the influence of alcohol

• Surveys on driving under the influence of drugs Data were also presented in tabular form. Data of drivers killed in traffic accidents are demonstrated in Table 15.5. However, epidemiological research is limited because there may be risk factors associated with drug use that are not indicated in studies. Further, it is not possible to distinguish between “real” risk factors and other factors that might be highly correlated with the risk factor [19]. Epidemiological studies are also difficult to compare because they show several kinds of differences as the following: • Different sample populations (age, gender, etc.). • Time at which the study is performed differs (year and day of the week; studies conducted on weekend nights with higher percentages of drug-positive drivers). • Biological samples are analyzed for different types of psychoactive substances (prevalence results can depend on number and type of analyzed substances). • Different types of biological samples are used with varying detection times (e.g. cannabis can be detected in urine for a relatively long period and the presence in urine does not necessarily proof actual influence). • Different analytical techniques are used to analyze samples with different limits of detection and quantification. • Different cut-off levels are used in analyses.

3.1%

3.0% 28.9%

11.7%

0.4%

4.7% 13.1% 9.2%

Drugs and alcohol 9.7%

Amphetamine 4.1%a Methamphetamine a MDA a MDMA a Cocaine Cannabis 13.5% Benzodiazepines 4.1% 2.0% 1.0%

1.5%

1.5%

2.5% >0.5‰: 2.0%

4.7% 4.7% 3.9%

0.8% 0.0% 8.4% 8.4% 5.9%

0.0%

16.0%

Table 15.5  Prevalence of alcohol and drugs in drivers killed in traffic accidents according to [3] Australia [7] Canada [8] France [9] Hong Kong [10] Italy [11] Italy [12] Year 1990–1999 1999–2002 2003–2004 1996–2000 1986–1996 1997–1999 Sample size 3,398 855 2,003 197 129 119 Blood Sample Blood Blood and Blood Blood alcohol, Blood alcohol, urine blood/urine blood/ for drugs urine for drugs Remarks <30 years 1 vehicle Dead within crash 4h 47.9% >0.5‰: 24.9% 55.8% Alcohol >0.5‰: 33.5% >0.1‰: 29.1% >0.8‰: 48.1% 28.5% Drugs 26.7% 24.7% 6.1% 21.9% Spain [14] 1996–1998 33 Blood and urine

0.6% 5.2% 2.2% 3.4%

1.2%

58.9% 43.8% >0.8‰: 48.1% Illicit: 8.8% 10.0% Licit: 4.7% 5.6% 7.0%

Spain [13] 1991–2000 5,745 Blood

36.36%

Glasgow

UK [16] 1998–2002 22 Blood

0.9% 0.5% 3.9% 7.6%

5.2% 0.8%

4.9%

17.0%

35%

>0.1‰: 44.0%

USA [17, 18] 2001–2002 370 Blood and serum

3.5% 12./% Diazepam 4.1% 4.6% (continued)

b

b

4.6%b

Illicit: 8.1% 13.6% Licit: 19.4% >0.2‰: 4.9%

>0.2‰: 22.2%

Sweden [15] 2000–2002 855 Blood and urine

1.0% 0.5% 0.5%

MOR morphine, HC hydrocodon a Includes Methamphetamine, MDMA, cocaine, (pseudo)ephedrin, phentermine b Including methamphetamine, MDMA

3.1%

2.3%

Barbiturates Phencyclidine Ephedrine Ketamine Propyphenazone Antidepr Phenytoin Pain killers Others

0.4% 0.5%

3.9%

Methadone

1.9%

0.8% 1.7%

3.4%

1.3%

Opiates

4.9%

Italy [12]

Table 15.5  (continued) Australia [7] Canada [8] France [9] Hong Kong [10] Italy [11]

0.5%

0.6%

0.3%

3.2%

Spain [13]

0.6%

0.1%

USA [17, 18]

1.9%

4.6% 1.6% (MOR) (MOR) 1.9% (HC)

Spain [14] Sweden [15] UK [16]

15  Alcohol and Drug Fatalities in Transportation: Forensic-Toxicological Implications

301

These factors can influence the outcome of epidemiological studies and make it nearly impossible to compare results. Thus, there is need for methodological guidelines. Recently, a consensus report was published defining guidelines, standards, core data variables and other controls that could be basis for future research [20].

Alcohol, Drugs or Medicines and Driving In the following, basic information is given about alcohol, drugs and medicines, especially concerning effects in driving based on various sources [2, 3, 21–24].

Effects of Alcohol on Driving It is well known that alcohol has a profound effect on driving skills. Because of its depressant effects, drivers can misjudge their capabilities. The effects of alcohol are determined by body weight and time. Some of the effects of alcohol that affect driving include the following: • Reaction time – slow reflexes can decrease the ability to react swiftly to situations. • Vision – eye muscles function more slowly; eye movement and perception are altered, possibly resulting in blurred vision; night vision and colour perception are also impaired. • Tracking – the ability to judge the car’s position on the road, the location of other vehicles, centre line, road signs, etc. can be affected adversely. • Concentration – attention to driving may decrease and/or drowsiness may occur. • Comprehension – the depressant effect of alcohol hinders the ability to make rational decisions. • Coordination – the mechanics of driving can be affected by reduced eye/hand/ foot coordination. If alcohol is used in combination with other drugs (legal or illegal) the effects of both substances can increase. In case of prescribed drugs health professionals must educate patients about the danger of combining alcohol with medication: • • • • • •

Impaired vision. Reduced reaction times. Reduced concentration and vigilance. Feeling more relaxed and drowsy, the driver may fall asleep at the wheel. Difficulties in interpreting complex sensory information. Difficulties with multiple tasks such as a person’s ability to keep a vehicle within the lane limits and keep the correct direction while paying attention to other important things during driving.

302

F. Mußhoff

• Drivers might ignore traffic rules. • Drivers may feel more confident which may lead to increased risk tolerance. Hangover effects of alcohol the day after can also affect a person’s ability to concentrate and to drive safely and might let him fall asleep while driving. Someone who drinks alcohol may think that he is able to alter his driving to counteract any impairment of his driving ability; however, it is a well-known effect of alcohol that drivers have an altered view and experience of reality. This means that actions and responses may be quite different to what is actually required and drivers may be unaware of how much their driving skills are affected by alcohol. In general, blood alcohol concentrations (BACs) of 0.5  g/l or lower lead to driving impairments. Reduced capacity, for example demonstrated in tunnel vision (impairment of peripheral vision caused by reduced performance of the retinal neurones and of the corresponding areas of the cerebral cortex) already occurs at a BAC of 0.3 g/l [25]. As of 0.3 g/l, driving up close to another car, misperceptions of speed and incorrect passing may be caused by impaired depth of visual acuity [26]. Examining eye movements and reactions of drunken drivers with a BAC of 0.7 g/l revealed that the average length of visually fixing an object was significantly longer for intoxicated drivers. With regard to reaction behaviour it was ascertained that inebriated drivers obtained significantly poorer results than sober drivers. For example, inebriated drivers had more problems to react appropriately to sudden appearing objects. Moreover, sober drivers remembered more observation exercises (cyclists, pedestrians, traffic signs) along a test road than intoxicated drivers. Responsiveness, endurance and visual structuring capacity decrease under the influence of alcohol. Furthermore, a significant increase of the readiness for taking risks can be observed. According to Bartl et  al. [27], an intoxication of 0.5  g/l up to 0.84 g/l lead to a decrease of relevant driving capacities (ability to observe, to react and to concentrate as well as mental capacity). In particular, it is interesting that the most reduced capacities are seen in test situations with highest correlation to the number of traffic accidents. In general, effects become more apparent as BAC increases. BACs of 1.0–1.5  g/l will lead to a marked loss of coordination and perception. 1.5–2.0 g/l will be evident as drunkenness and at 2.0 g/l the “passing out” stage occurs. 3.0 g/l will greatly heighten the risk of poisoning. Concentrations higher than 4.5/5.0 g/l are probably fatal due to respiratory paralysis. The causal effects of alcohol on impaired driving are well established, to the extent where it has been possible to enact legislation for the use of alcohol by drivers based on a valid classification system (BAC 0.5–0.8‰ = risk curve begins to ascend; BAC > 0.8‰ = considerable risk increase for most drivers; BAC 1.0‰ = definite increase of crash risk for all drivers [28]). The so-called Grand-Rapids Study by Borkenstein et  al. [29] proved the correlation between alcohol intoxication and accident risk by comparing more than 12,000 accidents of inebriated and sober drivers (Fig. 15.1). The probabilities of causing an accident could be attributed to the different BACs indicated in Table 15.6.

15  Alcohol and Drug Fatalities in Transportation: Forensic-Toxicological Implications

303

Fig. 15.1  Correlation between blood alcohol concentration (BAC) and accident risk according to Borkenstein [29]

Table 15.6  Probabilities of causing an accident for different BACs (according to [29])

g/l BAC 0.0 0.6 1.0 1.2

Probability   0×   2×   6× 25×

Effects of Cannabis on Driving Cannabis terms the preparation of the Cannabis sativa plant. Used products are marihuana (dried parts of the plant), hashish (resin of female flowering tops) and hashish oil (extract from resin). Further data are summarized in Table  15.7. Depressant THC effects slow down brain activity and other areas of the central

304

F. Mußhoff

Table 15.7  Data sheet concerning cannabinoids Cannabinoids Active agent

Metabolites

Intoxication

Effects on driving

Tetrahydrocannabinol (THC) Effective dose by inhalation at approx. 15 mg of THC (cannabis 0.1–0.2 g hashish) with maximum plasma concentration after 15–20 min Half-life approx. 45 min in absorption phase, 3.5 h in distribution phase and up to 24 h in the terminal elimination phase Window of detection in plasma 4–6 h (after singular consumption) 11-OH-THC (psychotropic substance) Half-life 12–18 h in the terminal elimination phase Window of detection in plasma 4–6 h (after singular consumption) THC-COOH (inactive) Half-life 25–37 h, in the terminal elimination phase up to 6 days Window of detection 2–3 days in plasma after singular consumption, approx. 3 weeks after regular consumption and in urine samples 2–3 d after singular consumption and up to 3 m after regular consumption Acute phase (1–2 h): central sedating effects with dysfunction in motor activity/speech; reddened glassy eyes; mydriasis; in general slowing down and maybe slow-witted Subacute phase (approx. 4–6 h after consumption): drowsiness gone, more exuberant, unconcerned basic mood with euphoria, serenity and internal calmness, under elimination of negative environmental factors; lowered criticism ability; over-confidence of capacity Postacute phase (approx. 12–24 h): decreased impulse, passivity, dizziness Sedation; severe fatigue, motor disturbances, concentration and attention weaknesses; extensions of reaction time; sensitivity to light; accumulation of false, inadequate reactions; disorders of entrenched automatisms (changing speeds, deviations or drift from the lane, violation of right of way signs and red lights; inadequate reactions to perceptions on the edge of the visual field)

nervous system (CNS) while the minor hallucinogenic effects can distort a person’s perception of the world. The effects of cannabis can be different for each person and are influenced by factors such as: • Dose (ingested and also dependent on the THC content of the preparation). • Ingestions route: when cannabis is smoked the effects are experienced very quickly and may last up to 5 or 8 h; after oral consumption the onset of effects can be delayed by about 60–90 min and can last up to 24 h. • Consumer’s psychological and physical attributes; general factors such as mental or emotional state and physical health can influence the effects of a drug. Therefore, it is difficult to prognose exactly in which way and how long cannabis will affect a person’s ability to drive safely. As a general guideline some of the effects of cannabis that can affect a person’s driving ability include the following: • Reduced coordination and slower reaction times • Slower information processing ability, confusion and impaired thinking • Changes in visual, auditory, time and space perception

15  Alcohol and Drug Fatalities in Transportation: Forensic-Toxicological Implications

305

The “hangover” effects of cannabis which can last for several hours can also impair the driving ability. The following studies concerning cannabis and driving constitute only an excerpt of studies recently evaluated by Raes et al. [3]. Ramaekers et al. [30] defined performance impairment (in terms of motor control, motor impulsivity and executive function) as a function of THC concentration in serum. It was concluded that 2 and 5  ng/ml are the lower and upper ranges of a serum THC limit for impairment. Binomial tests showed an initial and significant shift towards impairment in the critical tracking task for serum THC concentrations between 2 and 5  ng/ml. At concentrations between 5 and 10 ng/ml approximately 75–90% of the observations were indicative of significant impairment in every performance test. At THC concentrations above 30  ng/ml 100% of observations in every performance test were indicative of significant impairment. Some deleterious effects of cannabis appear to be additive or even synergistic with those of alcohol; combination of both substances results in prolongation as well as enhancement of their effects [31]. Driving studies showed that drivers under the influence of alcohol and cannabis are less attentive to traffic approaching from side streets while the use of either cannabis or alcohol had no effect [32]. Furthermore, the combination of cannabis and alcohol generates an additional decrement in lateral control on top of the decrement caused by either cannabis or alcohol [31]. Few epidemiological studies investigated the risk of being involved in traffic accidents while driving under the influence of cannabis. A Canadian case–control study of driving under the influence of cannabis alone was associated with an odds ratio (OR) of 2.2 (95% CI: 1.5–3.4), while taking all cannabis cases into account, an OR of 4.6 (95% CI: 3.4–6.2) was found [33]. Driving under the influence of alcohol (BAC > 0.8‰) and cannabis was associated with an increased accident risk of 80.5 (OR, 95% CI: 28.2–230.2). Another French study with drivers below the age of 27 years who were driving under the influence of cannabis alone was associated with an increased accident risk of 2.5 (OR, 95% CI: 1.5–4.2); alcohol (BAC > 0.5‰) and cannabis led to an increased risk of 4.6 (OR, 95% CI: 2.0–10.7) [34]. A Dutch and Norwegian study showed an increased (statistically insignificant) accident risk for driving under the influence of cannabis alone [16]. Another study realized in Australia found an OR of 2.7 (95% CI: 1.02–7.0) for being responsible for an accident while driving under the influence of cannabis alone [7]. For drivers with blood THC concentrations ³5  ng/ml, the OR was greater and statistically more significant (OR 6.6, 95% CI: 1.5–28.0). A significantly stronger positive association with accident responsibility was also seen in drivers who were positive for cannabis and had a BAC ³ 0.5‰ compared with drivers having a BAC ³ 0.5‰ with no cannabis consumption (OR 2.9, 95% CI: 1.1–7.7). Drivers in France involved in fatal crashes with positive cannabis detection were associated with decreased sense of responsibility (OR 3.3, 95% CI: 2.6–4.2) [35]. Moreover, a significant dose effect was identified with OR increasing from 1.6 (95% CI: 0.8–3.0) for THC concentrations in blood of 0–1 ng/ml to 2.1 (95% CI: 1.3–3.4) for THC concentrations above 5 ng/ml. For driving under the influence of alcohol combined with cannabis an OR of 14 (95% CI: 8.0–24.7) was calculated.

306

F. Mußhoff

In a recent review, Ramaekers et al. [36] concluded that a person who consumed cannabis may think that he can alter his driving to counteract any impairment to his driving ability; however, the effects of cannabis may mean lead to an altered view and experience of reality. This means that actions and responses may be quite different to what is actually required and drivers may be unaware of how much their driving skills are affected by cannabis. THC has been shown to impair cognition, psychomotor function and actual driving performance in a doserelated manner. Detrimental effects appear more prominent in highly automated driving behaviour, compared to more complex driving tasks that require conscious control.

Effects of Heroin and Other Opioids on Driving Opium is the dried milky exudation from the unripe capsules of Papaver somniferum, containing morphine (4–21%) as well as noscapine (2–8%), codeine (0.7–3%), papaverine (0.5–1.3%) and the baine (0.2–1%) which are combined under the term “opiate.” After further extraction heroin (=diacetylmorphine) is synthesized from morphine by double acetylation. Heroin is more lipophilic and reaches the brain more rapidly than morphine and euphoric effects are more intense. Heroin can be inhaled (by smoking or inhaling the vapors of heated powder), snorted or injected intravenously (usual dose 10–15 mg). Further information is summarized in Table 15.8. Morphine is a potent analgesic, therapeutically used for the relief of moderate to severe pain. Abuse is rare in Europe. Codeine is a licit opiate with antitussive and analgesic properties. Other so-called opioids include synthetic drugs such as pethidine, oxy- or hydrocodone, tramadol, tilidine, methadone or buprenorphine which are used for the relief of moderate to severe pain (e.g. palliative care), the latter in substitution programs. Heroin and other opioids are depressant drugs. They slow down the activity of the brain and other parts of the CNS. Sedation induces sleepiness, apathy and indifference to external stimuli and a decrease of reaction time. In combination with alcohol sedation is enhanced. Miosis has a negative influence on the accommodation of the eyes to darkness (entering a tunnel, driving at night). The effects of opioids are influenced by a range of factors such as: • The type of opioid and its strength. • The dose because larger amounts can produce different effects (e.g. heroin can vary in strength and purity, thus it can be difficult to predict the extent to which a person’s driving skills will be impaired after using heroin). • The ingestion route (injected > swallowed or smoked). • The consumer’s psychological and physical attributes.

15  Alcohol and Drug Fatalities in Transportation: Forensic-Toxicological Implications

307

Table 15.8  Data sheet concerning opioids Opioids Application form Heroin Effective dose parenteral approx. 50–250 mg “street heroine” Half-life 2–9 min Intermediate 6-Monoacetylmorphine with a half-life of approx. 38 min Active agent Morphine (Analgetic effective single dose 5–20 mg) Half-life 1.1–3.1 h Window of detection in plasma several hours (dose-dependent), in urine approx. 2–3 days Codeine Oral single dose 10–60 mg Half-life 1.9–3.9 h Window of detection in plasma several hours up to a few days (dosedependent), in urine approx. 2–3 days Dihydrocodeine Oral single dose 10–30 (60) mg Half-life 3.3–4.5 h Window of detection in plasma several hours up to a few days (dosedependent), in urine approx. 2–3 days Intoxication Acute phase: in the primal phase incapacity for minutes after i.v. application; in the second phase mild euphoria; indifference; euphoric as well as dysphoric moods; fluent transition to deprivation syndrome with merely physical insufficiencies as well as a decrease in attention and perception; in general an uncontrolled additional consumption of CNS-depressant substances (benzodiazepine, cannabinoids, methadone, codeine, dihydrocodeine, alcohol) but also centrally stimulating substances (amphetamine, ecstasy, cocaine), so synergetic as well as antagonistic effects are possible Effects on driving CNS depression and sedation, apathy, drowsiness, and dizziness with concentration weakness; motoric reduction; extended time of reaction; miosis (hell-dark-adaptation); short after application/during detoxification: slow, “shaky” way of driving including drift from the lane/ collisions; as the strong hypnotic effect wears off, perhaps aggressive, unrestrained way of driving (aggressive driving, inadequate, risky over-taking; violation of right of way signs, etc.)

So it is difficult to say exactly in which way and how long opioids will affect a person’s ability to drive safely. As a general guideline, some of the effects of heroin and the misuse of some opioids that may affect a person’s driving ability include the following: • Slow reaction time • Taking longer to respond to events or situations and possibly choosing an inappropriate response • Reduced coordination

308

• • • • •

F. Mußhoff

Reduced ability to think clearly Changes to visual acuity such as blurred vision Drowsiness or starting to “nod off” Miosis Nausea and vomiting

When the effects of heroin or opioid diminish some people may experience withdrawal symptoms like cravings or “hanging out,” cramps and flu-like symptoms which may also affect a person’s driving ability. As summarized by Raes et al. [3], a French case–control study demonstrated that morphine consumption is associated with increased accident risk (OR 8.2, 95% CI: 2.5–27.3) [35]. A Dutch study revealed that use of codeine is not associated with increased accident risk (RR 3.0, 95% CI: 0.7–14.2) while heroin and morphine (not combined) are associated with increased accident risk of 32.4 (OR, 95% CI: 1.8–592.0) [16]. Furthermore, a study of Norway showed that driving under the influence of opiates alone (morphine, heroin or codeine) is associated with an increased accident risk of 13.8 (OR, 95% CI: 1.2–154.2). A meta-analysis of data from the Dutch and Norwegian studies indicates that drivers under the influence of opiates only have an increased risk of being involved in an accident, as indicated by RR of 3.2 (95% CI: 1.4–6.9) and an OR of 3.7 (95% CI: 1.4–10.0). The question is whether patients under opioid therapy, for example palliative care patients, are able to drive. A structured evidence-based review by Fishbain et al. [37] evaluated 48 studies to determine what kind of evidences exists for and against opioid-related impairment of driving skills in opioid-dependent/tolerant patients. Fishbain concluded that the majority of studies indicated that opioids apparently do not impair driving-related skills in opioid-dependent patients and that – under certain conditions – patients stabilized on long-term opioid therapy are able to drive. Thus, physicians should not necessarily take the position that being on opioids precludes driving. It was recommended that patients on long-term opioid treatment should be advised of the current status of this research. It should be the patient’s decision whether he does or does not drive but the following rules should be respected: 1. After starting the opioid treatment or after increase in dosage patients should not drive for at least 4–5 days. 2. They should not drive if they feel sedated. 3. They should report sedation/unsteadiness/cognitive decline immediately to the physician so that reduction in dosage can be initiated. 4. Under no circumstances they should consume alcohol or other illicit drugs such as cannabinoids and drive. 5. They should avoid taking further medication, especially if it causes sleepiness, like for example antihistamines or fever medicines prescribed by a doctor or bought over-the-counter. 6. They should change their way of taking the medication without visiting the physician.

15  Alcohol and Drug Fatalities in Transportation: Forensic-Toxicological Implications

309

In summary – as pointed out by Raes et al. [3] – opioids cause cognitive and psychomotor impairment highly dependent on the type of opioid and the administered dose. Morphine tends to slow users’ responses although accuracy is not diminished. Fentanyl produces cognitive impairment in doses used in out-patient surgical procedures but shows little impairment effect when used for long-term pain management. Heroin users show clear impairment of psychomotor and cognitive skills, some of which can last for more than a year after the last drug use. Some of the impairments are related to severity of dependence and duration of use. Acute effects of methadone can be avoided by dividing the daily dose. Methadone maintenance treatment does cause impairment, including additional impairment higher than impairment associated with heroin dependence although the latter can be better explained by other associated risk factors in some cases. Buprenorphine users do not generally show impairment but at high doses. Patients on long-term opioid therapy develop some impairment of psychomotor and cognitive performance. However, the effect of the opioid drug itself concerning the impairment of patients under opioid maintenance therapy is unclear. Other factors like diseases and pain seem to be of greater importance than opioid effects.

Effects of Cocaine on Driving Cocaine is extracted from leaves of the coca plant Erythroxylon coca. The leaves can be chewed. Cocaine base (“free base” or “crack”) is smoked while the hydrochloride is snorted (25–100  mg) or injected. Further information is given in Table 15.9. Cocaine is a stimulant drug and speeds up the activity of the brain and other parts of the CNS. It can vary in purity and strength which makes it difficult to predict the extent to which a person’s driving ability will be impaired after using the drug. As a general guideline, some of the effects of cocaine that can affect a person’s driving are as follows: • Impaired ability to react appropriately • Poor concentration and judgement • Over-confidence in driving skills, not necessarily supported by an actual improvement in driving ability • Feelings of increased confidence which may increase the risk that the person will take unnecessary risks • Feelings of aggression which may lead to dangerous driving • Drowsiness as the cocaine wears off which may increase the risk that the driver suddenly falls asleep It has to be taken into consideration that the “come down” effects (e.g. exhaustion, mood swings and depression) after cocaine consumption may also impair a person’s driving ability.

310

F. Mußhoff

Table 15.9  Data sheet concerning cocaine Cocaine Active agent:

Metabolites:

Intoxication:

Effects on driving:

Cocaine Single dose of approx. 10 mg pure cocaine-HCl i.v., 20–50 mg i.n., max. 100 mg Half-life 42–90 min Window of detection in plasma 4–6 h (instable!), in urine 6–8 h Benzoylecgonine Half-life 4.5–7 h Window of detection in plasma a few days (dose-dependent), in urine 3–6 days Ecgoninmethylester Half-life 3.1–5 h Window of detection in plasma a few days (dose-dependent), in urine 3–6 days Euphoric phase: feeling high with euphoric moods, strong positive sensations, bravery, increasing risk-taking, increase of impetus without getting exhausted or tired, decrease of inhibitions, over-confidence, lowered criticism ability Inebriation phase: often negative and anxious misperception of environment (possibly. hallucinatory state with paranoia) Depressive phase: loss of impetus; tiredness and exhaustion; irritability and depression Unrestrained and risky way of driving with inadequate high speed and over-confidence in driving skills; restlessness; poor concentration; nervousness; sensitivity to light (dilation of the pupils); irritability and aggressiveness; decreasing concentration and attention but also: due to physical exhaustion, tiredness and depressive alienation (possibly disorientation, confusion, paranoia) with slow or changing speed, drift from the lane, etc.; frequently, due to paranoia, hit and run incl. car chases

Because of a low number of positive cases in most studies it was impossible to calculate the risks concerning traffic accidents after cocaine abuse [3]. A study in Canada revealed that driving under the influence of cocaine is associated with an increased accident risk of 12.2 (OR, 95% CI: 7.2–20.6) [33, 38]. Driving under the influence of either cocaine or a combination of cocaine and cannabis, of cocaine and alcohol (BAC > 0.8‰) or cocaine, cannabis and alcohol (BAC > 0.8‰) was associated with an increased accident risk of 4.9 (OR, 95% CI: 1.4–17.4), 8.0 (OR, 95% CI: 3.1–20.7), 170.5 (OR, 95% CI: 21.2–1,371.2) and 85.3 (OR, 95% CI: 9.5–767.0), respectively.

Effects of Amphetamine, Methamphetamine and “Ecstasy” on Driving As well as cocaine, amphetamines are central stimulants. They belong to the group of drugs that include prescription medicines and illegally produced amphetamines

15  Alcohol and Drug Fatalities in Transportation: Forensic-Toxicological Implications

311

Table 15.10  Data sheet concerning amphetamines Amphetamines Active agents:

Intoxication:

Effects on driving:

Amphetamine, methamphetamine, MDMA, MDA, and MDEA Effective doses approx. 10–50 mg amphetamine/methamphetamine and approx. up to 100 mg MDMA/MDEA/MDA Half-life 4–12 (34) h for amphetamine, approx. 9 h for methamphetamine and approx. 7–25 h for MDMA/MDA/MDEA Window of detection in plasma for each 6–24 h, in urine 1–3 days With short-term application: stimulating, sleep- and tiredness-restraining, mood-lightening effect with feeling of increased concentration ability, talkativeness, etc. With long-term application: disturbance, anxiety and depression as well as restlessness, agitation and confusion or even violent behaviour Detectable from the outside: nervousness; motor disturbances, inability to concentrate; dilated, fixed pupils; trembling; insomnia respectively tiredness and later loss of impetus, fatigue and exhaustion,; irritability; depression Acute phase: unrestrained and risky driving with inadequate high speed increased sense of capacity (Over-confidence, misjudgement, disturbance, fidgetiness, nervousness, sensitivity to light (mydriasis), irritability and aggressiveness), sometimes dramatic capacity decrease when subsiding with physical exhaustion, tiredness and depressive alienation (reduced concentration and attention, disorientation, confusion, loss of reality, etc.) with typical driving features due to fatigue (slow/changing speed, drift from the lane, etc.)

and methamphetamine (sometimes known as “speed,” “base” or “paste”; crystal methamphetamine is sometimes called “ice” or “crystal meth”). Ecstasy is the term for a range of drugs that are structurally similar to 3,4-methylenedioxymethamphetamine (MDMA) and shows effects like amphetamines (stimulants) and hallucinogens. Amphetamines suppress feelings like tiredness and hunger and increase mental alertness and physical energy. In addition, they enhance the mood and increase self-confidence. Therapeutic applications of amphetamines are the treatment of narcolepsy, obesity and hyperactive behaviour in children. Amphetamines are used by truckers and students to stay awake for long periods. MDMA and analogues are stimulants (“dance-pills”) and entactogens (leading to emotional disinhibition and increased social communication abilities). As the substances loose their effects it comes to negative feelings like fatigue, anxiety, emptiness and depression. Later, “hangovers” associated with headaches, muscle aches, exhaustion, apathy, sweating, nausea and further effects are experienced. Further data are given in Table 15.10. It is considered that the effects of amphetamines/ecstasy are influenced by a range of factors, thus each person can show different reactions. Things to consider include the following: • Quality and dose (there is no quality control on illegally manufactured drugs so that manufacturers may substitute a wide range of substances) • Consumer’s psychological and physical attributes

312

F. Mußhoff

As a general guideline, some of the effects of amphetamines that can affect a person’s driving ability include the following: • • • • •

Attention difficulties and a tendency to fidget Feeling of disorientation Lack of coordination Impaired ability to react appropriately and safely control a vehicle Aggressive and dangerous driving and increased chance of taking unnecessary risks • Over-confidence in driving skills which is not necessarily supported by actual improvement of driving ability • Drowsiness as the effects of amphetamine wears off and the driver possibly falls asleep (rebound fatigue) Similar to cocaine, the “come down” effects after amphetamines consumption (exhaustion, concentration difficulties, irritability and depression) may impair a person’s driving ability, too. Raes et  al. [3] summarized several studies concerning amphetamines and driving. Driving simulation showed that the intake of dexamphetamine (0.42 mg/kg) causes a decrease of all driving performances that were simulated by inducing problems like incorrect signalling, failing to stop at red lights and slow response time [39]. Brookhuis et al. [40] performed driving simulator tests on a group of young people who had admitted to take MDMA regularly. They were tested shortly after MDMA consumption as well as before they were intend to go to a party and again during a control night at a corresponding time when they were sober. Under the influence of MDMA subjects drove faster only in residential areas with speed limit of 50 km/h. Speed variance increased in the city as well as on the motorway. Lateral control and gap acceptance was not affected. Crashes occurred during two of the 20 control rides and four times while under the influence of MDMA which is a 100% increase. In Norway, Gustavsen et al. [41] analyzed the concentration– effect relationship between blood amphetamine concentrations and impairment in selected cases with amphetamine or methamphetamine as the only drug present in blood samples from impaired drivers. According to the police physician, 27% were judged to be not impaired while 73% were judged to be impaired. A positive relationship was found between blood amphetamine concentration and impairment but it showed a limit at concentrations of 270–530 ng/ml. In the Canadian study, it was demonstrated that driving under the influence of amphetamines is associated with increased accident risk of 12.8 (OR, 95% CI: 3.0–54.0) [33]. In the responsibility analysis of Drummer et  al. [7] the amphetamine-associated risk was not calculated but was calculated for a group of substances acting as stimulants, namely amphetamine, methamphetamine, MDMA, ephedrine, pseudoephedrine, phentermine and cocaine. There was no significant association between stimulant use and crash responsibility. However, when truckers were considered as “discrete driver type” the OR increased to 8.8 and was of borderline statistical significance (95% CI: 1.0–77.8).

15  Alcohol and Drug Fatalities in Transportation: Forensic-Toxicological Implications

313

Effects of Hallucinogens on Driving Hallucinogens like ketamine, LSD (lysergic acid diethylamide), magic mushrooms (psilocybin), mescaline (peyote cactus), and PCP (phencyclidine) distort a person’s perception of reality. The effects of hallucinogens vary greatly and combination of hallucinogenic drugs with other hallucinogens might have unpredictable effects on a person’s ability to drive safely. Some of the effects of hallucinogenic drugs that can affect a person’s driving include the following: • • • •

Distorted perception of reality Impaired thinking Blurred vision Reduced coordination

Effects of Medicines on Driving In general, the highest risk of driving impairment and having an accident due to pharmaceutical drug usage is thought to be during the first 2 weeks after a person began taking medication. In Table  15.11, relevant medicines with potential relevance to influence driving ability are summarized. The effects of medicines can be different for each person. Type of medicine, dosage and the consumer’s psychological and physical attributes are just some of many other factors that can influence the drug-induced effects on a person. Thus, it is difficult to prognose exactly in which way and how long specific medicines will affect a person’s ability to drive safely. In general, the health professional has to inform a patient about potential side-effects of new prescribed medication. The patient himself should read the package insert or the Consumer Medication Information sheet (CMI) before taking any medicine. It is highly important to check whether a new prescribed medicine or combination of medicines may have an impact on driving ability. It is possible that herbal remedies may also impair a person’s ability to drive safely. In contrast, a patient who is concerned about a prescription should not stop taking the medication but should not drive until he expressed his concerns to his health professional. The medical condition itself may impair a person’s ability to drive safely; in this case, the medication may rather assist than hinder safe driving. Extra care should be taken with medicines used to treat: • • • • • • •

Sleeping difficulties Anxiety, depression and stress Pain such as strong painkillers containing codeine and other opioids Allergies and hay fever Colds and flu Arthritis Blood pressure

Table 15.11  Short summary of medicaments relevant in traffic accidents according to Schubert [42] Pharmaceutical classes Examples for substances/agents Risk Analgetica Strong analgesic, sedation, perhaps withdrawal, change in Opioids Morphine derivatives mood and impulse, change of cognitive and sensory Morphine, codeine dihydrocodeine, substitutes capacity (methadone, buprenorphine), oxycodone, hydromorphone, tilidine, tramadol Uncritical mono-compounds (maybe. headache, nausea, Non-opioid analgetica Salicylates vertigo); danger due to mixing compounds, e. g. with Acetaminophene caffeine Propyphenazone Phenacetine Antidiabetica Insulin Hyper- and hypoglycemia, esp. in phases of re-/adjustment Sulfonylurea derivatives Impairment of CNS functions and sedating Antiepileptica Clonazepam Phenobarbital Phenytoine Primidone Due to substance class more or less sedating effect Antihistaminica Diphenhydramine Promethazine Ketotifene Partly sedating or impairing the cardiovascular system Antihypertensiva Clonidine (vertigo, tiredness, headache) Reserpine Guanethidine Prazosine, enalapril captopril, lisinopril, betablockers Narcotics Mixed compounds for ambulant short or local Partly differing half-times and due to this reduction of anaesthesia/e.g., lidocaine, nitrous oxide, etc. psychomotility Ophtalmica Anticholinergica, atropin, belladonna e.g. dysfunction of accomodation

++ + up to ++++

++ ++ up to ++

+ up to ++++

+ up to + +

+ + up to +++

++ up to +++

− up to +++

++ up to ++++

Level

314 F. Mußhoff

+ up to ++++ + up to ++++

Dullness, impulse reduction, dysfunction of coordination and psychomotility Increase or reduction of impulse, CNS accompanying symptoms, lowering of spasm threshold Soothing, sleep-inducing effect with reduction of performance and reaction Long half-times and cumulation; due to this effects on psychomotility Long-term reduced concentration and restlessness with cancelling tiredness for a short time

+ up to +++

++ + up to ++++

++ ++ up to ++

Level

Risk

− no effect, + little effect, ++ slight effect, +++ obvious effect, ++++ severe effect

Stimulants

Sedativa, Hypnotika

Examples for substances/agents

Neuroleptica e.g. chlorpromazine haloperidol thioridazin Antidepressiva e.g. amitriptyline, trimipramine Tranquilizer e.g. diazepam, flunitrazepam, oxazepam Barbiturates, benzodiazepines, bromureides, chloralhydrate, piperidin derivatives Caffeine Ephedrine norpseudoephedrine

Pharmaceutical classes

Psychopharmaceuticals

15  Alcohol and Drug Fatalities in Transportation: Forensic-Toxicological Implications 315

316

• • • • • • •

F. Mußhoff

Epilepsy Heart conditions Fluid retention Nausea Stomach problems Diabetes Several types of infections

As a general guideline some of the effects of medicines that can affect a person’s driving ability include the following • • • • • •

Drowsiness or tiredness Dizziness or feeling light-headed Difficulty to concentrate Feeling edgy, angry or aggressive Feeling nauseous or otherwise unwell Reduced coordination or feeling shaky and unstable

Using benzodiazepines and driving, for example may cause an increase in the risk of a person to have an accident [3]. People on higher doses or those who just began to take higher doses are the most at risk of impaired driving and to have a resulting accident which is due to some of the effects of benzodiazepines that may affect driving ability such as drowsiness and fatigue, blurred vision, lack of muscular coordination, slower reaction time, slower information processing, reduced concentration and impaired judgement.

Effects of Mixing Drugs on Driving Taking more than one drug at the same time or taking one drug followed by another can have unpredictable results [21]. This includes mixing illegal drugs and legal drugs such as alcohol and medicines (prescribed and over-the-counter medication). If a person takes multiple drugs (including alcohol, medicines and illegal drugs) each drug could alter the effects of the other, often in an unpredictable way. The effects of mixing drugs are influenced by a range of factors and can be different for each person. Things to consider include the following: • • • •

Type of drugs Dose of each drug Intake of drugs (taken at the same time, at different times and in which order) Consumer’s psychological and physical attributes (person’s mental or emotional state and physical health)

The risk of a crash while someone is under the influence of two or more drugs (illicit, licit or pharmaceutical) may be even higher than under the influence of one drug.

15  Alcohol and Drug Fatalities in Transportation: Forensic-Toxicological Implications

317

Combining drugs with similar effects such as alcohol and cannabis, alcohol and benzodiazepines, or amphetamines and ecstasy can increase the effects of each drug. Depressant drugs like alcohol, cannabis, heroin or other opioids, and benzodiazepines slow down brain activity and other parts of the central nervous system. Combining depressants can multiply the depressant effects (slowing down) in an additive as well as in an exponential way. Stimulant drugs like amphetamines, cocaine and ecstasy speed up brain activity and other parts of the CNS and combination of stimulants can multiply the stimulant effects. This lead to greater stress on the body, particularly on the heart and other vital organs, and can impair a person’s driving ability more than in the case that only one drug was consumed. Combining drugs with different effects like alcohol and ecstasy or cocaine and benzodiazepines seem to “cancel each other out.” This makes it difficult for someone to estimate how much his driving ability has been impaired. If a person has been drinking alcohol and using amphetamines, for example, he may not feel the depressant effects of the alcohol as they have been masked by the stimulant effects of amphetamines. The person may feel capable of driving when he might be drunk in fact. These combinations also put a lot of stress on the body since it tries to balance the different effects of the drugs which simultaneously exert their effects.

Forensic-Toxicological Analysis As described above, in the epidemiological aspect alcohol, drugs and medicines play an important role especially in (fatal) traffic accidents. Therefore, a chemicaltoxicological analysis should be performed in every case of traffic accident. Not only drivers responsible for the crash but also pedestrians or others persons who were recognized as victims can be affected. During the course of clarification drug-related mistakes of victims could be interpreted as jointly responsible for an event. After an accident without injuries or fatalities drug analysis is performed, as routinely done in every case that is suspicious of DUI using routine laboratory procedures. Analysis is complicated if injured or fatal persons have to be tested due to some specifics which have to be considered like potential influences of emergency medical aid (medication, infusions, loss of blood), a longer time period between event and sampling of material for analyses (also between the time of death and sampling) or no availability of definite sample material even in fatalities (e.g. blood).

Matrices In principle, various matrices are useful for chemical-toxicological analysis to investigate cases of drugged driving. Generally, urine is the sample of choice for screening and identification of unknown drugs because concentrations of substances

318

F. Mußhoff

are relatively high compared with other matrices such as blood, saliva or sweat. Saliva and sweat are potentially useful for on-site drug-testing procedures in living persons but not in fatalities. In urine, the window of detection for a preceding consumption of drugs is markedly enhanced compared to blood. Additionally, the sometimes active or inactive metabolites of drugs in urine samples can be identified in addition, or even exclusively. Otherwise, a positive result in a urine sample does not necessarily prove an actual impairment. The dose of a drug and as a general rule impairments due to drug consumption are most closely correlated with its concentration in blood. For this purpose, the relevant matrices that should be analyzed for quantification are whole blood or rather serum or plasma. In these matrices, the unchanged drug is detectable in most cases and the sample material is relatively homogenous because physiologic parameters vary only within narrow limits. In addition, blood or plasma samples are mandatory in cases of DUI in a relevant number of countries all over the world. Difficulties arise when only aged or haemolysed or post-mortem blood is available, even in cases of fatal traffic accidents.

Post-mortem Material for Toxicological Analysis in Traffic Fatalities In general, specimens available in post-mortem toxicology investigations can be numerous and variable and may be selected based on case history, requests, legal aspects and availability in a given case. Further information about post-mortem toxicology is given in some excellent reviews [43, 44].

Body Fluids Urine As described above, the accumulation of drugs and metabolites in urine usually results in high concentrations facilitating detection of drug use. Therefore, urine has a great potential to provide information on ante-mortem drug exposure and is frequently used as a screening specimen. However, there is no correlation between urine drug concentration and pharmacological effects and urine may not always be indicative for acute poisoning and impairment.

Bile Even in post-mortem cases in which urine is not available bile may be substituted. Bile represents a collection and storage depot for many drugs and corresponding metabolites that have a biliary excretion and are subject to enterohepatic circulation.

15  Alcohol and Drug Fatalities in Transportation: Forensic-Toxicological Implications

319

Many drugs have been shown to accumulate in bile and the detection of drugs and/ or major metabolites in bile may indicate previous or chronic exposure to a drug. Cerebrospinal Fluid and Vitreous Humour Cerebrospinal fluid and vitreous humour are thought to be closer to the site of action of drugs than blood and are useful to screen for a variety of drugs. They are less subject to contamination and bacterial invasion by virtue of their protected environment inside the brain, the spinal column or the eye. Both cerebrospinal fluid and vitreous humour also contain very little enzymes and proteins. Therefore, drugs which are highly protein bound or those which are lipophilic tend to be found in lower concentrations in these fluids than in blood. Vitreous humour is also useful for alcohol analysis and has been used to distinguish ante-mortem alcohol ingestion from post-mortem alcohol formation. Gastric and Intestinal Contents Oral ingestion remains the most popular means of exposure, especially to medicines. Therefore, the gastrointestinal contents are essential for screening. Undigested pills and tablets are often present. Blood As described above, blood is the specimen of choice for quantifying and interpreting concentrations of drugs and corresponding metabolites. For further interpretation quantitation is usually performed on specimens from peripheral sites, for example from the femoral vein, because post-mortem drug concentration can vary from site to site. Caution has to be made concerning samples labelled as “heart blood.” Samples may not have been collected from the heart itself but drawn blind through the chest wall and may include pleural or chest fluid, pericardial fluid and even gastric content if the death was traumatic. Following severe injury or trauma, samples from defined sources are often not available and blood may be collected only from the thoracic or abdominal cavity. The composition of these specimens markedly differs from whole blood. Therefore, these “blood” samples only provide a qualitative documentation of the presence of a drug. Tables  15.12 and 15.13 summarize factors that should be taken into consideration when interpreting blood analysis results. Tissue Samples Tissues usually collected for post-mortem toxicological investigations include liver, kidney, lung, brain and skeletal muscle specimens, as well as adipose tissue. Drug

320

F. Mußhoff

Table 15.12  Special considerations in the interpretation of blood analysis results in post-mortem toxicology 1. Cardiac vs. peripheral blood: site dependence and distribution/redistribution phenomena •  Cardiac blood is usually more abundant than peripheral blood •  In contrast to concentrations derived from peripheral sources, reference data of cardiac blood drug concentrations (from living persons) are not available •  Important factors affecting the concentration ratio cardiac/peripheral blood are the type of drug, its volume of distribution, concentration range, protein binding, pKa-value and the post-mortem interval between death and autopsy (sampling) •  Site-dependent differences can also arise from an incomplete distribution of the drug at the time of death, and/or from post-mortem redistribution at the cellular level by passive diffusion or via the vascular pathway from the major organs; the vascular pathway may depend on the blood remaining fluid after death •  The ante-mortem interval may be of importance, as one of the main factors behind the redistribution phenomena: during the distribution phase, the arterial blood concentration can be appreciably higher than the venous blood concentration   To consider: •  Drug levels in heart blood are mainly higher than in femoral venous blood and often there is a wide range of ratios of drug concentrations in cardiac vs. peripheral blood (basic drugs with a large volume of distribution showed the greatest range) Heart blood is only useful for qualitative screening procedures! 2. Blood to serum/plasma ratios •  Blood is a complex mixture containing solubilized proteins, dissolved fats, solids, and suspended cells but drug concentrations provided in literature are usually determined in serum •  The water content and pH of a post-mortem blood sample may also differ significantly from physiological ranges; samples are often haemolysed, putrefied, and may be quite inhomogeneous mixtures   To consider: •  Literature data of serum/plasma concentrations cannot be absolutely used to classify the concentrations determined from post-mortem blood •  Blood to plasma concentration ratios for drugs of forensic interest has to be taken into consideration Be careful in interpretation of quantitative data with respect to the characteristics of the drug of interest! 3. Post-mortem instability of substances and metabolic production (Table 15.14) •  Bioconversion in situ after death as well as in collection vessels can occur •  Decomposition and bacterial production occur dependent on time intervals and temperature   To consider: •  Target analytes in living and deceased can vary •  Possible influences of storage conditions prior to sampling as well as between sampling and analysis has to be taken into consideration Special target analytes and artefacts due to storage have to be considered in post-mortem toxicology!

detection in tissue specimens may be considered whenever drugs are involved that are highly lipophilic in nature and are preferably bound to tissue. Tissue samples may also be useful in cases with extended post-mortem time period (time interval between accident and time of death) and whenever body fluids are not available or difficult to obtain.

15  Alcohol and Drug Fatalities in Transportation: Forensic-Toxicological Implications

321

Table 15.13  Special mechanisms in post-mortem samples between time of death and sampling and during further storage, modified according to [42, 59] Mechanisms Example(s) Chemical instability • Hydrolysis • Heroin, cocaine, acetyl salicylic acid • Oxidation • Oxidation of sulphur containing drugs, morphine Metabolic instability • Hydrolysis of ester type drugs or prodrugs • Esterases (endogenous) • Enzyme activities derived from • Hydrolysis of glucuronides, e.g. morphine glucuronides • Reduction, e.g. nitrobenzodiazepines, THCCOOH bacteria • Oxidation, e.g. thioridazine Metabolic production Ethanol, GHB, carbon monoxide, cyanide

Kidney Even in cases without urine samples a kidney specimen can be useful for screening purposes since most drugs and metabolites are excreted into urine and will pass through the kidneys. Liver Liver is favoured as a specimen when blood is not available due to exsanguination, fire or decomposition. Since most drugs are metabolized in the liver, both the parent compound and its metabolites may be present in high concentrations. Analysis of a liver tissue specimen may also help to differentiate acute overdose from therapeutic use of drugs with a narrow dosing window.

Lungs Often high drug concentrations can be found in lung tissue, especially in cases of inhalation or intravenous poisoning.

Brain Brain is a useful specimen for the measurement of drugs because it is the principal site of action for many drugs. Additionally, lipophilic substances like, for example antidepressants, narcotics and halogenated hydrocarbons accumulate in central nervous tissue. According to Mura et al. [45], THC for example can be detected in brain regions that are influenced by its effects even if the substance is no longer detectable in blood. This could be of special interest in traffic fatalities.

322

F. Mußhoff

Skeletal Muscle Skeletal muscle as a post-mortem sample for analysis is used when blood is not available, for example due to exsanguination. It is present in large quantities and less affected by decomposition than the internal organs. However, the muscle to blood ratio is influenced by the time lapse between drug exposure and death as well as by the volume of distribution of the drug and drug analysis on skeletal muscle is rather qualitative than quantitative in nature. Muscle specimens had also been considered as an alternative sample for alcohol analysis but the muscle to blood ethanol concentration ratio was found to depend on the time course of ethanol absorption, distribution and elimination. Hair Samples Hair samples may provide retrospective information about drug (ab)use [46, 47]. The amounts deposited in hair are functions of both ingestion/exposure and of the metabolic regimen. There are various factors that influence drug concentrations in hair. However, hair analysis revealed information about previous drug consumption and tolerance.

Laboratory Approaches to Drug Testing in DUI Cases In cases with suspicion of driving while impaired by drugs and especially in traffic fatalities samples should be subjected to a broad screen for common drugs not only relevant to the patterns of recreational drug use in jurisdiction but also for common CNS-acting prescription and over-the-counter drugs [48]. Analysis starts with an immunoassay screening procedure in most cases. These assays are often class-specific rather than drug-specific and their value is to rule out the presence of certain drug classes in form of presumptive tests without further forensic relevance, especially at court. Positive immunoassay results should not be considered as proof of identification of a compound without complementary confirmatory analysis. Furthermore, it has to be taken into consideration that immunoassays will not detect all types of drugs that are present; it may produce different intensities of response to members of the same drug class and may fail to identify important members of drug classes completely. Due to cross-reactivity and considering interferences false-negative and false-positive immunoassay results can be revealed, especially in biological samples of road casualties (after medical treatment with various drugs and infusions or if sample material is available after a crash that is poorly defined). Especially in cases of fatal traffic injuries immunoassay tests should be supplemented by chromatographic tests to include as many of the relevant drugs as possible. Without this procedure there is significant possibility that the drug or the metabolite that has caused impairment is not identified.

15  Alcohol and Drug Fatalities in Transportation: Forensic-Toxicological Implications

323

“Gold standard” for confirmatory identification of drugs in biological samples are gas chromatographic or liquid chromatographic procedures coupled with mass spectrometry (GC/MS or LC/MS(−MS)). Generally, a clean-up procedure is applied to the sample to separate drug(s) and metabolites from the sample matrix and to concentrate the analytes since then they can be detected and quantified more readily. Gas chromatographic – mass spectrometric (GC/MS) as well as liquid chromatographic – mass spectrometric (LC/MS) procedures for drugs of abuse monitoring in cases of DUI were recently reviewed and are not matter of the present article [49–52]. Each laboratory is free to perform chemical-toxicological analysis using home-made procedures. However, analytical methods should be validated and a linear range and limits of detection and quantification have to be established according to accepted guidelines [53]. Recently, the German speaking Society of Toxicological and Forensic Chemistry (GTFCh) published new guidelines and recommendations for forensic-toxicological analyses containing some more details and especially concerning the method validation in forensic-toxicological analytical techniques including pre-testing with immunoassays [54]. Recommendations for appropriate cut-offs for screening and confirmation in both blood and urine are provided for some of the most important analytes in DUI investigations (Table  15.14). As described by Farrell et  al. [48], too, thresholds were established to reflect the performance of both commercially screening technology and confirmatory techniques, both routinely used in forensic toxicology laboratories. These recommended cutoffs were based on analytical methodology rather than pharmacology or the probability of impairment. It has to be realized that in many cases laboratories use an immunoassay cut-off concentration that is lower than the manufacturer’s recommended cutoff, particularly if the assay is marketed towards clinical testing (intoxication with high drug concentrations). This is common practice as long as laboratories properly validate their procedures and establish in-house cut-off concentrations using appropriate matrix. In contrast to other groups, the German speaking Society of Toxicological and Forensic Chemistry (GTFCh) only recommended values for confirmatory chromatographic procedures. If an immunoassay is used as preliminary test the laboratory has to demonstrate the required sensitivity for the target substance. The list of drugs in Table  15.11, proposed according to statistical data of the USA, does not claim completeness and does not represent an exhaustive or comprehensive list of analytes or impairing substances. There is a substantial regional variability in patterns of illicit drug use and some drugs that are specific or unspecific to the laboratory’s demographic area can be in- or excluded. Because some drugs such as, for example, GHB are only analyzed in few cases at the request of local authorities or with special strong suspicion, probably the real number of cases DUI is underestimated. Recent studies demonstrated that the combination of an amphetaminebased drug with GHB could be very popular [55, 56]. Table 15.14 does not contain recommendations for hallucinogens and inhalants. However, hallucinogens such as LSD, peyote and psilocybin, as well as commonly abused inhalants such as butane, ether or other anaesthetics, freon, nitrous oxide, toluene and xylene will significantly impair the user’s ability to operate a motor vehicle safely. Currently, there

10 50 10

–a 50 –a

–d –d –d –d –d 50 –d 50 –d –e –e –e –e

CNS depressants Alprazolam Chlordiazepoxide Clonazepam 7-Aminoclonazepam Diazepam Nordiazepam Lorazepam Oxazepam Temazepam Trazodone Amitriptyline Nortriptyline Diphenhydramine

10 50 10 10 20 20 10 50 50 25 25 25 25

20 20 20 20

20 20 20 20

CNS stimulants Amphetamine Methamphetamine MDMA MDA MDEA Cocaine Benzoylecgonine Cocaethylene 25 25 25 25 25 10 30

–d –d –d –d –d 100 –d 100 –d –e –e –e –e

–a 300 –a

200 200 200 200

50 totalc 50 totalc 50 totalc 50 totalc 50 totalc 50 totalc 50 totalc 50 totalc 50 totalc 50 50 50 50

20 50 20

50 50 50 50

Table 15.14  Recommended scope and analytical cutoffs of toxicological analysis in DUI investigations Blood (ng/ml) Urine (ng/ml) Target analyte Screen Confirmation (Farrell et al. [47]) GTFCh [53] (Serum) Screen Confirmation (Farrell et al. [47]) Cannabis 2 1 –a 2 THC –a Carboxy-THC 10 5 10 20b 5 11-OH-THC –a 2 (1) –a 2

30

200 200 200 200 200

10 totalc

GTFCh [53]

324 F. Mußhoff

–f –f –f 50 20 freeg –f 50 –e

–e 10

Narcotic analgesics Codeine Hydrocodone Hydromorphone Methadone Morphine Oxycodone Propoxyphene Tramadol

Dissociative drugs Dextromethorphan Phencyclidine

20 10

10 10 10 10 10 10 50 20

500 500 20 100 100 100 500 500 1,000 5,000

50 10

10

GTFCh [53] (Serum)

–e 25

–f –f –f 300 200 –f 300 –e

–e –e –e –a –a 200 –e –e –e –e

50 10

50 50 50 50 50 totalc 50 50 20

500 500 20 100 100 100 5,000 5,000 1,000 10,000

Urine (ng/ml) Screen Confirmation (Farrell et al. [47])

200 25 totalc

25 totalc

GTFCh [53]

THC = Delta-9-tetrahydrocannabinol; Carboxy-THC = 11-nor-9-carboxy-delta-9-tetrahydrocannabinol; 11-OH-THC = 11-hydroxy-delta-9-tetrahydrocannabinol; MDMA = 3,4-methylenedioxymethamphetamine a  Immunoassay screening not targeted to this analyte b  Combination of free and conjugated analyte c  Immunoassay screening targeted to nordiazepam, oxazepam or both; not an effective tool for screening all drugs in this class d  Not routinely screened for by immunoassay e  Immunoassay screening targeted to morphine; not an effective tool for screening all drugs in this class f  Free drug, not conjugated

–e –e –e –a –a 100 –e –e –e –e

Blood (ng/ml) Screen Confirmation (Farrell et al. [47])

Carisoprodol Meprobamate Zolpidem Butalbital Phenobarbital Secobarbital Phenytoin Carbamazepin Topiramate Gamma-hydroxybutyrate

Target analyte

15  Alcohol and Drug Fatalities in Transportation: Forensic-Toxicological Implications 325

326

F. Mußhoff

exist limited techniques for routine screening of blood and urine for these compounds and usually target analysis is performed if case information suggest possible involvement of such substances.

Interpretation of Post-mortem Toxicology Results in Traffic Fatalities Because in cases of fatal traffic fatalities the sampling of biological specimens for toxicological analysis takes place immediately post-mortem formation of alcohol is negligible for further interpretation. In contrast, it has been demonstrated that alcohol in stomach content may diffuse through the gastric wall and the diaphragm and eventually enters into the heart and central blood vessels [57, 58]. Severe trauma with rupture of the stomach and the diaphragm might lead to a passing of gastric content into the chest cavity which reveals in artefactual high chest blood alcohol concentrations. In addition, agonal or post-mortem movement of gastric contents into the trachea and the lungs can lead to elevated alcohol concentrations in the major central pulmonary and in cardiac vessels and may cause erroneous interpretation [59]. Interpretation of the concentration of other drugs in post-mortem specimens is complicated because many drugs are unstable in vivo and in vitro [60] (Table 15.14). For example, cocaine is hydrolyzed readily before and after death. It was demonstrated that serum cholinesterase is responsible for the hydrolysis to ecgonine methyl ester while the formation of benzoylecgonine may arise from spontaneous nonenzymatic hydrolysis [61]. Therefore, interpretation must not only be based on the measured concentration of cocaine but also on ecgonine methyl ester and benzoylecgonine. Some benzodiazepines (e.g. flunitrazepam, nitrazepam or clonazepam) are also known to be unstable in vitro and exact calculation of peri-mortem concentration is not practical [62, 63]. Many other drugs have poor stability in post-mortem blood, too (e.g. chlordiazepoxide, olanzapine, zopiclone). However, one of the most important factors to affect the interpretation of postmortem drug concentrations is post-mortem redistribution as described above. The use of reference tables of therapeutic and toxic drug concentrations should be treated with caution. Tables are useful in clinical toxicology but they are of limited value if it comes to interpretation of post-mortem concentrations. Inappropriate use of those tables can result in over- or underestimation of potential effects of the drug since they depend on the degree of tolerance, pre-existing diseases and if other substances are present. Tolerance as well as inter-individual variations in pharmacological response (e.g. pharmacogenetics), drug interactions and the presence of natural diseases always has to be taken into consideration [64]. Potential tolerance can be proven by hair analysis which gives information about recent substance use [46, 47]. Medical prescriptions of the last weeks or months should also be checked for further interpretation. In cases of injured or dead suspects in traffic accidents it is decisive that law enforcement officers perform broad investigation including crash reconstruction at the scene – perhaps with the help of technical experts – and recording of

15  Alcohol and Drug Fatalities in Transportation: Forensic-Toxicological Implications

327

witness statements to document impairment. At court, influence of technical defects has to be excluded and responsibility or culpability of a suspect or contributory impairing effects due to alcohol or other drugs have to be demonstrated by forensic experts in individual cases. DUI investigations must be consistently, scientifically and objectively. Toxicologists help to establish the connection between driving behaviour and drug use [48].

References 1. Krüger HP, Schulz E, Magerl H (1995) The German roadside survey 1992–1994. Saliva analyses from an unselected driver population: licit and illicit drugs. In: Kloeden CN, McLean AJ (eds) Alcohol, drugs and safety – T’95. University of Adelaide, NHMRC Road Accident Research Unit, pp 55–62 2. Maes V, Charlier C, Grenez O, Verstraete A (1999) Drugs and medicines that are suspected to have a detrimental impact on road user performance. Rosita EU research project, Deliverable 1 3. Raes E, van den Neste T, Verstraete A, Lopez D, Hughes B, Griffiths P (2008) Drug use, impaired driving and traffic accidents. European Monitoring Centre for Drugs and Drug Addiction (EMCDDA). EMCDDA Insights, Luxembourg 4. Drummer OH (1994) Drugs in drivers killed in Australian road traffic accidents. The use of responsibility analysis to investrigate the contribution of drugs to fatal accidents. Report No. 0594. Dep. of Forensic Medicine, Monash University, Melborne 5. Terhune KW, Ippolito CA, Hendricks DL, Micjalovic JG, Bogema SC, Santinga P, Blomberg R, Preusser D F (1992) The incidence and role of drugs to fatal accidents. DOT HS 808 065. National Highway Safety Administration 6. Longo MC, Hunter CE, Lokan RJ, White JM, White MA (2000) The prevalence of alcohol, cannabinoids, benzodiazepines and stimulants amongst injured drivers and their role in driver culpability: part II: the relationship between drug prevalence and drug concentration, and driver culpability. Accid Anal Prev 32:623–632 7. Drummer OH, Gerostamoulos J, Batziris H, Chu M, Caplehorn J, Robertson MD, Swann P (2004) The involvement of drugs in drivers of motor vehicles killed in Australian road traffic crashes. Accid Anal Prev 36:239–248 8. Brault M, Dussault C, Bouchard J, Lemire AM (2004) The contribution of alcohol and other drugs among fatally injured drivers in Quebec: final results. In: Oliver J, Williams P, Clayton A (eds.) Proceedings of the 17th International Conference of Alcohol, Drugs and Traffic Safety (Glasgow) (CD); International Council on Alcohol, Drugs and Traffic Safety (ICADTS), Oslo 9. Mura P, Chatelain C, Dumestre V, Gaulier JM, Ghysel MH, Lacroix C, Kergueris MF, Lhermitte M, Moulsma M, Pepin G, Vincent F, Kintz P (2006) Use of drugs of abuse in less than 30-yearold drivers killed in a road crash in France: a spectacular increase for cannabis, cocaine and amphetamines. Forensic Sci Int 160:168–172 10. Cheng JY, Chan DT, Mok V (2005) An epidemiological study on alcohol/drugs related fatal traffic crash cases of deceased drivers in Hong Kong between 1996 and 2000. Forensic Sci Int 153:196–201 11. Sironi L, Molendini LO, Bernabei C, Marozzi F (1999) Incidence of xenobiotics among drivers killed in single-vehicle crashes. Forensic Sci Int 104:37–46 12. Vignali C, Groppi A, Polettini A, Valli A, Sali A, Cancelli B, Montagna M (2001) Drugs and driving. Toxicological findings in 119 fatally injured drivers. In: Novakova E, Habrdova V (eds) Proceedings of the 39th International Meeting of the International Association of Forensic Toxicologists (TIAFT), pp 368–373

328

F. Mußhoff

13. del Rio CM, Gomez J, Sancho M, Alvarez FJ (2002) Alcohol, illicit drugs and medicinal drugs in fatally injured drivers in Spain between 1991 and 2000. Forensic Sci Int 127:63–70 14. Lopez-Rivadulla M, Cruz A (2000) Drugs and driving in Spain. Blutalkohol 37(Suppl): 28–33 15. Holmgren P, Holmgren A, Ahlner J (2005) Alcohol and drugs in drivers fatally injured in traffic accidents in Sweden during the years 2000–2002. Forensic Sci Int 151:11–17 16. Assum T, Mathijssen MPM, Houwing S (2005) The prevalence of drug driving and relative risk estimations. A study conducted in the Netherlands, Norway and United Kingdom; Immortal Deliverable D-R4.2. Austrian Road Safety Board, Vienna 17. Logan BK, Schwilke EW (2004) Changing patterns of alcohol and drug use in fatally injured drivers in Washington State 1992–2002. In: Oliver J, Williams P, Clayton A (eds) Proceedings of the 17th International Conference on Alcohol, Drugs and Traffic Safety (Glasgow) (CD-ROM). International Council on Alcohol, Drugs and Traffic Safety (ICADTS), Oslo 18. Logan BK (2005) Adverse effects of stimulants and cannabis on driving. Presentation at the 17th Meeting of the International Association of Forensic Sciences (IAFS), Hong Kong 19. Berghaus G (2007) Meta-analyses in research in forensic medicine: alcohol, drugs, diseases and traffic safety. Forensic Sci Int 165:108–110 20. Walsh JM, Verstraete AG, Huestis MA, Morland J (2008) Guidelines for research on drugged driving. Addiction 103:1258–1268 21. Australian Drug Foundation (2009) Drugs and driving. http://www.drugsdriving.adf.org.au. 22. Austrian Road Safety Board (2003) Preventative measures to prevent driving while under the influence of alcohol/drugs. Kuratorium für Verkehrssicherheit, Vienna 23. Madea B, Dettmeyer R with collaboration of Musshoff F (2007) Basiswissen Rechtsmedizin. Springer, Heidelberg 24. Madea B, Musshoff F, Berghaus G (2007) Verkehrsmedizin. Deutscher Ärzte-Verlag, Cologne 25. Krüger HP, Kazenwadel J, Vollrath M (1995) Das Unfallrisiko unter Alkohol mit besonderer Berücksichtigung risikoerhöhender Faktoren. Krüger HP (ed) Das Unfallrisiko unter Alkohol. Analyse, Konsequenzen, Maßnahmen. Fischer Verlag, Stuttgart, pp 1–114 26. Grohmann P (1996) Deutsches Promillerecht – Überblick und Leitfaden für die verkehrsgerichtliche Praxis. Blutalkohol 33:177–200 27. Bartl G, Lager F, Domesle L (1998) Testleistung bei Minderalkoholisierung. Lebensraum Verkehr – Kleine Fachbuchreihe des KfV, 97–141. Kuratorium für Verkehrssicherheit. Vienna. pp 97–141 28. Klebelsberg D (1988) Drugs and traffic safety – some basic reflections. IATSS Res 12:24–32 29. Borkenstein RF, Crowther RF, Shumate RP, Ziel WB, Zylman R (1964) The role of drinking driver in traffic accidents (The Grand Rapids Study). Department of Police Administration, Indiana University. Bloomington, Indiana, US 30. Ramaekers JG, Moeller MR, van Ruitenbeek P, Theunissen EL, Schneider E, Kauert G (2006) Cognition and motor control as a function of Delta9-THC concentration in serum and oral fluid: limits of impairment. Drug Alcohol Depend 85:114–122 31. Sexton B, Tunbridge RJ, Board A, Jackson PG, Wright K, Stark MM, Engelhart K (2002) The influence of cannabis and alcohol on driving (TRL 543). Transport Research Laboratory, Wokingham, UK 32. Lamers CT, Ramaekers JG (2001) Visual search and urban driving under the influence of marijuana and alcohol. Hum Psychopharmacol 16:393–401 33. Dussault C, Brault M, Bouchard J, Lemire AM (2002) The contribution of alcohol and other drugs among fatally injured drivers in Quebec: some preliminary results. In: Mayhew D, Dussault C (eds) Proceedings of the 16th International Conference on Alcohol, Drugs, and Traffic Safety (CD-ROM). La Societe deI’Assurance Automobile du Quebec. Montreal, Canada 34. Mura P, Kintz P, Ludes B, Gaulier JM, Marquet P, Martin-Dupont S, Vincent F, Kaddour A, Goulle JP, Nouveau J, Moulsma M, Tilhet-Coartet S, Pourrat O (2003) Comparison of the

15  Alcohol and Drug Fatalities in Transportation: Forensic-Toxicological Implications

329

prevalence of alcohol, cannabis and other drugs between 900 injured drivers and 900 control subjects: results of a French collaborative study. Forensic Sci Int 133:79–85 35. Laumon B, Gadegbeku B, Martin JL, Biecheler MB (2005) Cannabis intoxication and fatal road crashes in France: population based case-control study. BMJ 331:1371 36. Ramaekers JG, Berghaus G, van Laar M, Drummer OH (2004) Dose related risk of motor vehicle crashes after cannabis use. Drug Alcohol Depend 73:109–119 37. Fishbain DA, Cutler RB, Rosomoff HL, Rosomoff RS (2003) Are opioid-dependent/tolerant patients impaired in driving-related skills? A structured evidence-based review. J Pain Symptom Manag 25:559–577 38. Dussault C, Brault M, Lemire AM, Bouchard J (2001) The role of cocaine in fatal crashes: first results of the Quebec drug study. Annu Proc Assoc Adv Automot Med 45:125–137 39. Silber BY, Papafotiou K, Croft RJ, Ogden E, Swann P, Stough (2005) The effects of dexamphetamine on simulated driving performance. Psychopharmacology (Berl). 179:536–543 40. Brookhuis KA, de Waard D, Samyn N (2004) Effects of MDMA (ecstasy), and multiple drugs use on (simulated) driving performance and traffic safety. Psychopharmacology (Berl) 173:440–445 41. Gustavsen I, Morland J, Bramness JG (2006) Impairment related to blood amphetamine and/ or methamphetamine concentrations in suspected drugged drivers. Accid Anal Prev 38:490–495 42. Schubert W, Schneider W, Eisenmenger W, Stephan E (2005) Begutachtungsleitlinien zur Kraftfahrereignung, Kommentar. 2. Aufl., Kirschbaum Verlag, Bonn 43. Skopp G (2004) Preanalytic aspects in postmortem toxicology. Forensic Sci Int 142:75–100 44. Drummer OH (2004) Postmortem toxicology of drugs of abuse. Forensic Sci Int 142: 101–113 45. Mura P, Kintz P, Dumestre V, Raul S, Hauet T (2005) THC can be detected in brain while absent in blood. J Anal Toxicol 29:842–843 46. Musshoff F, Madea B (2007) New trends in hair analysis and scientific demands on validation and technical notes. Forensic Sci Int 165:204–215 47. Musshoff F, Madea B (2007) Analytical pitfalls in hair testing. Anal Bioanal Chem 388:1475–1494 48. Farrell LJ, Kerrigan S, Logan BK (2007) Recommendations for toxicological investigation of drug impaired driving. J Forensic Sci 52:1214–1218 49. Moeller MR, Kraemer T (2002) Drugs of abuse monitoring in blood for control of driving under the influence of drugs. Ther Drug Monit 24:210–221 50. Maurer HH (2009) Mass spectrometric approaches in impaired driving toxicology. Anal Bioanal Chem 393:97–107 51. Maurer HH (2005) Multi-analyte procedures for screening for and quantification of drugs in blood, plasma, or serum by liquid chromatography-single stage or tandem mass spectrometry (LC-MS or LC-MS/MS) relevant to clinical and forensic toxicology. Clin Biochem 38:310–318 52. Kraemer T, Paul LD (2007) Bioanalytical procedures for determination of drugs of abuse in blood. Anal Bioanal Chem 388:1415–1435 53. Peters FT, Drummer OH, Musshoff F (2007) Validation of new methods. Forensic Sci Int 165:216–224 5 4. Society of Toxicological and Forensic Chemistry (GTFCh) (2009) Richtlinie der GTFCh zur Qualitätssicherung bei forensisch-toxikologischen Untersuchungen und Anforderungen an die Validierung von Analysenmethoden. http://www.gtfch.org/cms/ index.php/guidelines?Itemid=126 55. Dresen S, Kempf J, Weinmann W (2007) Prevalence of gamma-hydroxybutyrate (GHB) in serum samples of amphetamine, metamphetamine and ecstasy impaired drivers. Forensic Sci Int 173:112–116 56. Verschraagen M, Maes A, Ruiter B, Bosman IJ, Smink BE, Lusthof KJ (2007) Post-mortem cases involving amphetamine-based drugs in The Netherlands. Comparison with driving under the influence cases. Forensic Sci Int 170:163–170

330

F. Mußhoff

57. Pounder DJ, Smith DR (1995) Postmortem diffusion of alcohol from the stomach. Am J Forensic Med Pathol 16:89–96 58. Iwasaki Y, Yashiki M, Namera A, Miyazaki T, Kojima T (1998) On the influence of postmortem alcohol diffusion from the stomach contents to the heart blood. Forensic Sci Int 94:111–118 59. Pounder DJ, Yonemitsu K (1991) Postmortem absorption of drugs and ethanol from aspirated vomitus – an experimental model. Forensic Sci Int 51:189–195 60. Drummer OH (1997). Stability of drugs postmortem: a review. Proceedings of the 35th Annual Meeting of TIAFT, Padua 61. Isenschmid DS, Levine BS, Caplan YH (1989) A comprehensive study of the stability of cocaine and its metabolites. J Anal Toxicol 13:250–256 62. Robertson MD, Drummer OH (1995) Postmortem drug metabolism by bacteria. J Forensic Sci 40:382–386 63. Robertson MD, Drummer OH (1998) Stability of nitrobenzodiazepines in postmortem blood. J Forensic Sci 43:5–8 64. Jones G (2004) Postmortem toxicology. In: Moffat AC, Osselton MD, Widdop B (eds) Clarke’s analysis of drugs and poisons, 3rd edn. Pharmaceutical Press, London, UK

Index

A Abdominal injury, 161 Accident investigation characteristics, 173 crash rate, 173 HFACS, 172 LOSA, 172 Swiss Cheese Model, 171 Acute (apparent) life threatening events (ALTEs), 130 Acute lung injury (ALI) surfactant deficiency, 49 and ventilator associated trauma, 50–52 Adult pulmonary pathology cancer, physiological derangements bronchogenic carcinomas, 63 metastatic carcinoma, 63 paraneoplastic syndromes, 63, 64 primary bronchogenic carcinoma, 62 primary malignancies and metastatic tumor deposits, 61 squamous cell carcinoma, 62 emphysema and asthma acute exacerbation, 69 COPD, 67 Langerhans cell histiocytosis, 68 long term smoking/chronic occupational exposures, 67 severe and acute respiratory distress, 69 systemic anaphylaxis, 69 infections acute, complications, 60 ancillary testing, tissues sampling, 60 aspiration, gastric contents, 59 community acquired and nosicomial pneumonia, 57, 58 histomorphological appearance, 59

host immunological defense, 57 lung involvement, 57, 58 pneumonia, 60 respiratory outbreaks, 60–61 risk factors, pneumonia, 58 interstitial lung disease acute conditions, 70 connective tissue stains, 71 diagnostic challenges, 71 end-stage chronic interstitial pneumonitis, 71 hypersensitivity pneumonitis, 71 pleural and interstitial nodules, 70 pneumonitis, 70 suspected aspiration, 61 systemic disease processes, 72 vascular and cardiovascular disease pulmonary hypertension, 65–67 pulmonary thromboembolism, 63–65 Air Accident Investigation Branch (AAIB), 148–150 Air balloons aviation accident, 175 crashes, 175–176 Aircraft accidents analysis, 168 in continental Europe, 150 disease, aircrew coronary artery, 154–155 post-mortem examination, 153–154 suicide, 155–156 injury mechanisms abdominal injury, 161 control, crash time, 162–164 deceleration effects, 157 head injury, 159 horizontal forces, 157–158

E.E. Turk (ed.), Forensic Pathology Reviews, Volume 6, Forensic Pathology Reviews 6, DOI 10.1007/978-1-61779-249-6, © Springer Science+Business Media, LLC 2011

331

332 Aircraft accidents (cont.) human tolerance to deceleration, 157 lack, safety harness, 157 overhead lockers, 158 pattern, 161–162 perpendicular forces, 158 pulmonary fat and bone marrow embolism, 164 scoring systems, 158–159 thoracic injury, 159–161 investigation action, scene, 151–152 autopsy, 152–153 human, aircraft and operational factors, 151 “Human Factors Group”, 151 obligatory statutory investigations, 150 toxicology blood alcohol levels, 165 carbon monoxide and cyanide, 165–166 crop spraying accidents, 167–168 disruption and contamination, 164–165 therapeutic and “over-the-counter” drugs, 167 in UK Coroner’s and AAIB inquiries, 149 definition, Annex 13, 148 investigation, 148–149 Standards and Recommended Practices for Aircraft Accident Inquiries, 147 State of Occurrence, 148 in USA, 149–150 Alcohol and drug fatality, transportation amphetamine, methamphetamine and “ecstasy” effects data sheet, 315 described, 314–315 guideline, person’s driving ability, 316 MDMA consumption, 316 responsibility analysis, 316 therapeutic applications, 315 cannabis effects altered view and reality experiences, 310 data sheet, 308 deleterious, 309 epidemiological studies, 309 guideline, person’s driving ability, 308 performance impairment, 309 used products and factors, 307–308 cocaine effects altered view and reality experiences, 310 calculation, accident risk, 314 data sheet, 314 described, 313

Index guideline, person’s driving ability, 313 effects altered view and reality experiences, 306 BACs, 306–307 correlation, BAC and accident risk, 307 medication, 305–306 probabilities, accident cause, 307 problems, 305 epidemiological data classification, 302 drivers rate, 300 DUI and responsibility analysis, 300 fatal accidents, responsibility analysis, 300 fatally injured drivers, 301 median exposure rate, 300 “real” risk factors vs. other factors, 302 roadside surveys, drivers, 300 THC, responsibility rate, 301 traffic accidents, prevalence, 302–304 forensic-toxicological analysis body fluids, post-mortem material, 322–326 drug, 321 drug testing, 326–330 hair samples, post-mortem material, 326 matrices, 321–322 tissue samples, post-mortem material, 323–326 hallucinogens effects, 317 heroin and opioids effects codeine use, 312 data sheet, 307, 308 factors, 310 guideline, person’s driving ability, 311–312 morphine, methadone and buprenorphine, 313 opium and morphine, 310 rules, 312 structured evidence-based review, 312 symptoms, 312 medicines effects benzodiazepines use, 320 CMI and conditions, 317 guideline, person’s driving ability, 320 medications, 317, 318 mixing drugs effects alcohol and medicines, 320 depressant and stimulant, 321 factors, 320 motor vehicle crashes, 299 post-mortem toxicology alcohol formation, 330 cocaine, 330

Index tolerance and reference tables use, 330–331 ALI. See Acute lung injury All-terrain vehicles (ATV) crashes CPSC, 200–201 definition, 200 ICD, 201–202 injury pattern, 204–212 manufacture, 199 public health authorities response, 202–204 Anthropology Deutsche bank, 194–195 human remains discoveries, 193–194 and personal effects, 195 osteology, 194 phase II recovery, 194 role, 194 sifting process, 195–196 WTC recovery work, 196 9/11 Attacks anthropology, 193–196 autopsy, 184 events airplanes, 185 description, 184 twin towers, 184 fatalities investigation, 183–184 forensic biology, 189–193 investigation and remains recovery, 185–189 pathologist, 184 Autopsy blood culture, 139–140, 142–143 fundamental requirements, 76 good, medico-legal and complete, 75–76 haemorrhage, 132 pathology, 143 pneumonia, 60 primary lung carcinoma, 61 protocols, 90 pulmonary edema and congestion, 56 report, pulmonary hypertension, 65–67 respiratory distress syndrome (RDS), 50 role (see SUDI post-mortem investigation)severe BPD, 50 staff safety, 60 sudden death, role, 90 Aviation accidents and fatality, 178–179 investigation, 171–173 aircraft-specific studies

333 helicopter air taxi accident, 175 homebuilt airplanes and gyroplanes, 176 hot-air balloons, 175–176 epidemiological studies Aerospace Medical Association, 173 environment and operational factors, 174 fatal and non-fatal injury, 173 helicopters, 176–177 pathology distal tibial shaft, 177–178 drugs and alcohol, 178 neck trauma, 178 powerline collisions, 177 types, accident mid-air collisions, 174 pilot characteristics, 175 spatial disorientation, 174 Aviation deaths. See Aircraft accidents B Backscattered electron (BSE), 86 Blood alcohol concentrations (BACs) description, 306 probabilities, accident cause, 307 Body cooling, 261, 272 Bronchopulmonary dysplasia (BPD), 50 Brugada syndrome, 85 BSE. See Backscattered electron C Calmodulin calcium sensor and signal transducer, 281 calpain-mediated CnA cleavage, 290 PP2A degradation, 292–293 Calmodulin-binding proteins (CaMBPs) degradation, postmorterm calpain, 281, 289 CaMKII, 289–290 cellular process control, 281 CnA, 290–292 iNOS activity regulation, 290 MARCKS, 289 PP2A stabilization/translation, 292–293 tissue probing, 289 Calpain cleavage CaMKII, 289 CnA, 290–292 iNOS, 290 MARCKS, 289 PP2A, 292–293 proteolytic agent, 281

334 Cerebrospinal fluid (CSF) glucose and lactate levels, 139 PCR, 138 protein, 137–138 proteomics causes, 139 proteins, 138 white cell count meningitis, 136 neutrophils, 136 SIDS, 136, 137 Childhood, sudden death. See also Sudden natural deaths, infancy and childhoodcoronary artery thromboembolism, 6 diabetes mellitus, 14 metabolic disorders, 14–16 parvovirus B19 and hydatid disease, 13 rectal bleeding, 10 CLSM. See Confocal laser scanning microscopy Confocal laser scanning microscopy (CLSM) description, 87 FLIM-based analysis, 87 structure, 86 Consumer Product Safety Commission (CPSC) annual report, ATV deaths, 200, 201 estimation, 201 NEISS, 201 authorities and executives, 203 data base, 205 ICD-9-CM, 201 Coronary artery disease, 154–155 CPSC. See Consumer Product Safety Commission D Death time estimation cadaver, 241 ecological succession data analysis, 229–230 experimental design, 228–229 insect community structure changes, 228 entomology, 216–218 FEM software, 266–267 theory, 264–266 FE-model boundary conditions, 269–272 geometry, 267–268 initial conditions, 269 material properties, 268–269

Index fluorescence spectroscopy, 243–245 forensic entomology, 215 insects, 221–228 methods, 261 model curves approaches, 262 PMI, 242 post mortem cooling, 261, 262, 276 specimens identification cuticular hydrocarbons, 220–221 genetic techniques, 220 insects, 218 morphology, 219–220 spectra interrogation light source, Y-type probe and spectrometer, 246 rat skin, 247 spectral analysis fluorescence, PMI, 247–249 tissue degradation, 248 spectral processing intensity ratio analysis, 249–250 PCA, 250–258 standardising methods Daubert judgement, 230 forensic entomology, 230–231 ISO/IEC, 231 thermodynamics, 262–264 tissue changes, 242 validation body constitutions and environmental temperature levels, 272–274 measurement cases, 275 simulation results, 273, 275 thermal tissue properties, 268, 269 Degradation, CaMBPs. See Calmodulin-binding proteins degradation, postmorterm Deoxyribonucleic acid (DNA) analysis, 184 database, 192 human diploid cell, 132–133 identification family member, 187 reference samples, 190–191 techniques, 190 profiles, 191 recovery phases, 193 sources, 191 test types, 192 Dissection techniques inflow-outflow and “short-axis” method, 77 plane, 77, 79 requirements, 76 Driving under influence (DUI) drivers test, 300

Index drug testing, laboratory approaches (see Forensic-toxicological analysis)investigations, 331 DUI. See Driving under influence E EEM. See Excitation Emission Matrix Ehlers–Danlos syndrome, 17 Endocrinology adrenal crisis/Addison’s disease, 96 Conn’s syndrome, 96 hypoparathyroidism, 98 morphological and post-mortem biochemical findings adrenal gland, 101–104 diabetes mellitus/coma diabeticum, 110–112 endocrine pancreas, 108–110 parathyroid gland, 107–108 pituitary gland, 98–101 thyroid gland, 104–107 myxeodema coma, 97 parathyrotoxic crisis chronic hyperparathyroidism, 97 hypercalcaemia, 98 pituitary coma postpartum haemorrhage, 96 symptoms, hypopituitarism, 95 post-mortem biochemistry adrenocorticotropic hormone, 114 blood glucose metabolism, 114–115 catecholamines, 113 cortisol, 113 thyroid hormones, 112–113 thyrotoxicosis, 97 Entomology physiological time, 218 post mortem events and intervals ecological succession and oviposition, 217–218 PMImin and PMImax, 216–217 Excitation Emission Matrix (EEM), 259 F FAC. See Family Assistance Center Family Assistance Center (FAC) family member relationship, 191 objectives, 187 Fatalities alcohol and drug (see Alcohol and drug fatality, transportation)ATV (see All-terrain vehicles (ATV)

335 crashes)coronary artery disease prevalence, pilot, 154 GA (see Aviation)WTC anthropology, 193–196 events, 184–185 forensic biology, 189–193 investigation, 185–189 FEM. See Finite-element-method Finite-element-method (FEM) software, 266–267 theory discretisation and results, 264 explanatory scheme, 264, 265 functions values, locations, 264–265 heat transfer equation, 265, 266 partial differential equation and second order spatial derivatives, 265 shape functions, 266 Finite element (FE)-model boundary conditions conduction, 269 convection, 270 internal power, 271–272 radiation factors, 270–271 geometry 3D, 267 grey-scaled materials, 267, 268 initial conditions, 269 material properties, 268–269 Flavine adenine dinucleotide (FAD), 245 Forensic biology DNA phases, recovery, 193 identification, 193 reference samples direct, 190–191 family members, 191 test types, 192 WTC DNA database, 192 identification, 189–190 Forensic pathologist response, homicide, 184 role, 184, 197 Forensic-toxicological analysis body fluids, post-mortem material bile, 322–323 blood, 323 blood analysis results, interpretation, 324 brain, 325 cerebrospinal fluid and vitreous humor, 323 gastric and intestinal contents, 323 skeletal muscle, 326 urine, 321–322

336 Forensic-toxicological analysis (cont.) drug analysis, 321 drug testing, laboratory approaches cutoffs recommendations, 327–329 GC/MS and LC/MS procedures, 327 GHB, 327 “gold standard”, 327 GTFCh, 327 immunoassay screening procedure, 326 hair samples, post-mortem material, 326 matrices, 321–322 tissue samples, post-mortem material brain, 325 drug detection, 323–324 kidney and liver, 325 lungs, 325 skeletal muscle, 326 G General aviation (GA) accident and fatality, 178–179 investigation, 171–173 types, 174–175 aircraft-specific studies, 175–176 epidemiological studies, 173–174 helicopters, 176–177 pathology, 177–178 H Haemorrhagic shock and encephalopathy (HSE), 129 Haemosiderin-laden macrophages (HLMs) birth weight and gestational age, 39–40 comparison, NAI, 39 structure, and intra-alveolar haemorrhage, 39 Head injury, 159 Heart sectioning, SCD examination, 76–77 long axis method described, steps, 77, 80 structure, 77, 80 long vs. short-axis sections, 79, 81, 82 requirement, information, 76 “short-axis” methods described, steps, 80–81 inflow-outflow and, 77 structure, 80, 81 standard method described, steps, 77 plane dissections, 77, 79 structure, 77–79

Index tissue sampling histological and immunohistochemical examination, 81–82 structure, 82, 83 Heat flow mechanisms, 262–263 Helicopters air taxi accident, 175 crashworthy aircraft, 176 EMS crashes, 176 lap belt and shoulder harness, 176–177 sling-load operations, 177 Hemolytic-uremic syndrome, 12 Heparin-induced thrombocytopenia (HIT), 101 HIT. See Heparin-induced thrombocytopenia HLMs. See Haemosiderin-laden macrophages Human factors analysis and classification system (HFACS), 172 Hypoxic ischaemic encephalopathy (HIE) and ALTE, 131 brain damage, 131 counter argument, 132 NAHI, 132 non-accidental injury, 131 I ILAC. See International Laboratory Accreditation Cooperation Immunoglobulin G (IgG), 126, 142 Immunohistochemistry techniques Brugada syndrome, 85 description, 83 detection, TNFa, 84 DNA testing, 85 “final common pathway”, 85 high-resolution imaging, 85 LQTS and SNTA1, 85 markers, 84 sodium and potassium channel subunits, 85 Infection explained SUDI, 124–125 HSE age incidence, 129 IgG, 126 respiratory tract, 126, 130 role, 143 Injury mechanisms, aircraft accidents abdominal injury, 161 control, crash time, 162–164 deceleration effects, 157 head injury, 169 horizontal forces, 157–158 human tolerance to deceleration, 157

Index lack, safety harness, 157 overhead lockers, 158 pattern of injuries, 161–162 perpendicular forces, 158 pulmonary fat and bone marrow embolism, 164 scoring systems, 158–159 thoracic injury, 159–161 Injury pattern, ATV brain, 205 children, 204–205 CPSC database, 205 death rate, children, 205 driving safety course, 205 mechanism backward rollover, 209, 210 crushed right upper face, 207 drivers and passengers, 208 driving prohibition, 207 front-over-rear rollover, 208, 209 neck, wire slung, 210, 211 off-road vehicle injuries, 212 prehospital care providers, 212 riding rate, 210 risk factors, 206 rollovers, right hip, 208, 209 Safety Institute, 212 vs. truck on public road, 208 related crashes, 204 riding, 205–206 Insects adults age estimation NIRS, 226–227 ovarian development and wing wear, 227 pteridine accumulation and flies relationship, 226, 227 techniques, 226 cuticular hydrocarbons adult flies, 224 discriminant function analysis, 223 larvae, 223 puparia, 224 data analysis insect development, 227–228 linear regression model, 228 gene expression and RNA analysis, 222–223 hormone production, 223 measurement accuracy, precision and bias, 221–222 pupae developmental anatomy beetles and flies life cycles, 219 metamorphosis, 226

337 MSCT and MRI, 225 spatial and temoporal resolution, 225 International Classification of Disease, 9th revision, Clinical Modification (ICD-9-CM), 201–202 International Laboratory Accreditation Cooperation (ILAC), 231 Investigation and remains recovery, WTC fatality death certification asphyxia, 188 memorial park, 189 positive identifications, 188 FAC, 187 identification, 187–188 medical examiner examination area, 185–186 postmortem tissue samples, 186–187 triage, 186 K Kawasaki disease, 6–7 L Lemierre syndrome, 13 M Magnetic resonance imaging (MRI), 225 Marfan syndrome, 7–8, 16 Mass disaster forensic medicine response, 196 identification, 184 Metabolic crisis, 114 Microbiology, post-mortem group 1 and 2 pathogens, 40 positive culture result, 40–41 Staphylococcus aureus and Escherichia coli, 40 MRI. See Magnetic resonance imaging Multi-slice computed tomography (MSCT), 225 N National Electronic Injury Surveillance System (NEISS), 201 “Negative” autopsy, 17 Non-accidental head injury (NAHI), 131 Nuclear magnetic resonance (NMR), 242

338 O Off-road vehicles dirt bikes, 200 E Code 821.X, 201 injuries, 212 P PCA. See Principal component analysis Pediatric pulmonary pathology ALI and ventilator associated trauma chronic aspiration, 51 diffuse intravascular fat emboli, 51 mechanisms, alveolar damage, 52 mortality rate and causes, 50–51 ventilation associated injuries, 52 born alive/dead flotation/hydrostatic test, 52 gas producing bacteria, 52–53 stringent physical evidence, 53 hypoplasia co-existing systemic developmental abnormalities, 48 description, 48 lungs radiological examination, 49 severe, 48–49 respiratory distress syndrome and bronchopulmonary dysplasia BPD, 50 surfactant deficiency, 49–50 SUDI and SIDS acute epiglottitis, 54 acute viral pneumonitis, 53 bacterial tracheitis, 54 bacterial/viral infections, 53 Haemophilus influenzae and pneumococcus/parainfluenza viruses, 54 hematoxylin and eosin staining, 56 immunohistochemistry, respiratory syncytial virus, 55 interstitial and intra-alveolar hemosiderin and siderophages, 57 nasopharynx sampling, 53 occlusive laryngeal polyp, 55 protein rich fluid use, 56 RSV, 54–56 Period of insect activity (PIA), 216 PMI. See Post-mortem interval PMSI. See Post mortem submergence interval Polymerase chain reaction (PCR), 138 Post-mortem biochemistry adrenal gland adrenocortical atrophy, 101

Index chronic follicular lymphocytic infiltration, 104 congenital enzyme defect, 103 eosinophilic cytoplasmatic inclusions, 102–103 incidentalomas, 102 lymphocytic infiltration, 101 macroscopic appearance, adrenocortical adenoma, 102 phaeochromocytoma, 103 Waterhouse-Fridrichsen syndrome, 101 adrenocorticotropic hormone, 114 blood glucose metabolism haemoglobin A1c, 114 hormone and c-peptide levels, 115 ketoacidotic coma, 115 vitreous humour and CSF, 114 catecholamines, 113 cortisol, 113 diabetes mellitus/coma diabeticum lymphocytic infiltration, 110 nodular glomerulosclerosis, 111–112 nuclei, liver epithelia, 111 periodic acid Schiff staining, 111 endocrine pancreas blood glucose level regulation, 108 insulinoma, cellular arrangement, 108–109 islet amyloid polypeptide (IAPP), 109, 110 nesidioblastosis, 109 symptoms, hypoglycaemia, 108 parathyroid gland cellular atypia and infiltrative growth behaviour, 108 histologic appearance, 107, 108 hyperparathyroidism, 107 pituitary gland cicatrisation, 99 connatal ACTH cell aplasia, 101 high power micrograph, adenoma, 100 hypophysitis, 100 inflammatory lesions, 98 necroses, neurohypophysis, 98 necrosis, early stage, 99 thyroid gland acid mucopolysaccharides, mitral valve, 106 atrophic autoimmune thyroiditis, 105 autopsy appearance, 107 Grave’s disease, 105, 106 Hashimoto’s thyroiditis, 104 regressive change, 105–106 residual fibrosis, 105

Index thyroid hormones, 112–113 Post-mortem interval (PMI) adult flies, 224 CaMBPs degradation (see Calmodulinbinding proteins (CaMBPs) degradation, postmorterm) determination, 242, 243, 245 device development, protein markers, 294 environmental influences, 293–294 fluorescence spectroscopy, 243, 245, 259 forensic pathology, 279 markers, protein, 281, 282–288 membrane permeability, 280 metamorphosis process, 222–223 prediction, 243, 255, 257, 259 protein degradation assessment, 280–281 Post mortem submergence interval (PMSI), 216 Principal component analysis (PCA) angular coefficient temporal evolution, 258 animal probability, 257 autovetores determination, 254 correlation plots, real and estimated PMI, 258 data projection, 254 representation, matricial form, 253 EEM, 259 in vivo animal’s skin fluorescence spectra, 252 matrix transformation, 253 new base data representation, 255 optical technique, 259 PC1 vs. PC2, 258 PMI, representative regions, 255–256 spectrum processing, 249 validation, PMI determination, 257 Protein CaMBPs degradation, 281–293 markers, 281–288 PMI measurement, markers, 294 Public health authorities response, ATVs crashes, 203 financial burden, 204 hospitalized patients, 203–204 injuries, 202 manufactures, 203 National Inpatient Sample, 203 rider-related risk factor, 203 Pulmonary pathology adult (see Adult pulmonary pathology) identification, autopsy, 72–73 pediatric (see Pediatric pulmonary pathology)

339 R Reconstruction, injury patterns, 161–162 Rib fractures interpretationSee SUDI post-mortem investigation S Scanning electron microscopy (SEM), 86 SCDSee Sudden cardiac death Scoring systems, 158–159 SEMSee Scanning electron microscopy SIDSSee Sudden infant death syndrome Single nucleotide polymorphisms (SNPs), 143 Spectral processing, intensity ratio analysis distribution, PMI function, 250 emission spectrum, 249 real and estimated PMI correlation plots, 252, 256 Sudden and unexpected death in infancy (SUDI), 53–57 Sudden cardiac death (SCD) cases autopsy good, medico-legal and complete, 75–76 role and protocols, 90 described, 76, 89 heart sectioning examination, 76–77 inflow-outflow and “short-axis” methods, 77 long axis method, 77, 80 long vs. short-axis sections, 81, 82 requirement, information, 76 short axis method, 80–81 standard method, 77, 79 structure, 77–79 tissue sampling, 81–82 Janssen cautions, 91 laboratory technologies analytical proteomics, 89 cardiac hypertrophy, 88–89 defined, proteome, 88 DNA testing and extraction, 87 high-quality DNA extraction, 87 mass spectrometry, 89 “pre-PCR restoration process”, 87–88 proteomic investigations, 88 Western blot analysis, 89, 90 microscopic approach Brugada syndrome, 85 CLSM, 87–88 high-resolution imaging, 85 histological examination, 82, 83

340 Sudden cardiac death (SCD) cases (cont.) immunohistochemistry, 83–84 LQTS and SNTA1, 85 markers use, immunohistochemical, 84 myofibers structural changes, 86 significance, 82, 85 sodium and potassium channel, 85 TEM, SEM and BSE, 86 TNFa, immunohistochemical detection, 84 requirements, 76 Sudden death autopsy pathology, 143 infection role, 143 pulmonary causes (see Sudden death, pulmonary pathology)SNPs, 143 SUDA, 142–143 SUDC, 141–142 SUDI, 124–141 Sudden death in epilepsy (SUDEP), 2 Sudden death, pulmonary pathology adult cancer and physiological derangements, 61–63 emphysema and asthma, 67–69 infections, 57–61 interstitial lung disease, 70–71 suspected aspiration, 61 systemic disease processes, 72 vascular and cardiovascular disease, 63–67 pediatric ALI and ventilator associated trauma, 50–52 born alive/dead, 52–53 hypoplasia, 48–49 respiratory distress syndrome and bronchopulmonary dysplasia, 49–50 SUDI and SIDS, 53–57 Sudden infant death syndrome (see Sudden infant death syndrome)terminology and classification, 28–31 Sudden infant death syndrome (SIDS). See also Sudden natural deaths, infancy and childhood definition and categories, 28 diagnosis, 28 “dustbin diagnosis”, 31 explained infection, 124–125 mandibular hypoplasia, 3 metabolic disorders, 14–16 SUDI, 31, 53–57

Index trisomies, 17 unexplained, risk factors co-sleeping/bed-sharing, 32 pacifiers use, 32 rates, 31, 32 “triple-risk” hypothesis, 33 Sudden natural deaths, infancy and childhood anaphylaxis, 17 cardiovascular system cardiac hypertrophy, 8 congenital abnormalities, 4 coronary arteries, 6, 7 dysplasia, aortic valve, 5 floppy mitral valve, 9 hypertrophic cardiomyopathy, 8 Marfan syndrome, 7–8 myocardial noncompaction, 9 myocarditis, 9 oncocytic cardiomyopathy, 8–9 pulmonary arteries, 6 pulmonary hypertension, 7, 8 venous abnormalities, 4–6 central nervous system subarachnoid hemorrhage, 2 SUDEP, 2–3 tuberous sclerosis, 3 ventriculoatrial shunts, 3 Ehlers–Danlos syndrome, 17 endocrine, 14 gastrointestinal diaphragmatic defect, 11 gut malrotation, 10 small intestinal infarction, 10 genitourinary hemolytic-uremic syndrome, 12 pyelonephritis, 11–12 hematologic, 11 infectious causes Hemophilus influenzae type B infection, 13 Lemierre syndrome, 13 meningococcal sepsis and bacterial meningitis, 11, 12 pneumonia, 11–12 Marfan syndrome, 16 metabolic disorders, 14–16 “negative” autopsy, 17 preautopsy radiography, 2 respiratory acute asthma, 3–4 epiglottis edema, 4 lingual thyroglossal duct cyst, tongue, 5 oropharyngeal and laryngeal region, 3 pneumothorax, 3

Index Sudden unexpected death in childhood (SUDC) IgG, 142 LQTS, 141 SUDI, 141 Sudden unexpected death in infancy (SUDI) ALTEs, 130 blood culture bacteria, 139 post mortem isolation, 139 S. aureus and E. coli, 140 cerebrospinal fluid, 135–139 channelopathy mutations ion channels, 135 prolonged QT, 135 classification, 124 CSF, 135–139 death mode baby check scores, SIDS infants, 127–128 bacterial toxins, 128 classical concept, 128 transient bacteraemia, 128 deleterious mutations genetic networks, 134 zygotes, 134 explained SUDI–infection baby check scoring system, 125 case control study, 125 pneumonia, septicaemia and meningitis, 125 gastrointestinal tract flora, 140 gene/environmental Interactions base changes, 133 changes, genes regulatory control, 133 DNA, 132 neutral mutations, 133 HIE, 131–132 inclusion criteria, CESDI, 124 “Near Miss”, 129 respiratory tract bacterial flora, 140 SIDS, 124 unexplained epidemiological features, 128 impressive body, evidence, 127 respiratory tract infection, 126 urine, 140–141 Sudden unexpected early neonatal death (SUEND), 36 SUDI. See Sudden unexpected death in infancy SUDI post-mortem investigation autopsy protocol and cause of death ancillary, 34 component, 34, 35

341 macroscopic examination, 34 microbiological analyses, 34 multi-agency approach, 34 paediatric tertiary centre, 34 tandem mass spectrometry, 34–36 evidence-based protocol, 41–42 HLMs, pulmonary birth weight and gestational age, 39–40 comparison, NAI, 39 structure and intra-alveolar haemorrhage, 39 infection role bacterial toxin hypothesis, 33 pathogenesis, 33 Staphylococcus aureus, 33 neuropathology role, 41 post-mortem microbiology interpretation group 1 and 2 pathogens, 40 positive culture result, 40–41 Staphylococcus aureus and Escherichia coli, 40 rib fractures interpretation data, autopsy series, 37 healing, macroscopic appearances, 37, 38 NAI component, 36–37 pooled data, 38 resuscitation-related fractures, 38 sudden infant deaths classification, San Diego, 28–30 complete autopsy, 28 cot/crib death and SIDS, 28 described, SUDI, 31 diagnosis, 28 “dustbin diagnosis”, 31 San Diego definition, 28 SUEND description, 36 vs. SUDI, 36 toxicology role, 41 unexplained SUDI/SIDS, risk factors behaviour modification strategies, 31 co-sleeping/bed-sharing, 32 pacifiers use, 32 rates, 31, 32 “triple-risk” hypothesis, 33 SUEND. See Sudden unexpected early neonatal death Suicide, 155–156 T TEM. See Transmission electron microscopy Terrorism. See 9/11 Attacks

342 Thermodynamics, heat transfer equation, 263–264 mechanisms, 262–263 Thoracic injury, 159–161 Time of colonisation (TOC), 216 Tissue fluorescence spectroscopy absorption, 243 biomolecule, 244 energy levels and electrons states, 243 laser system, 245 phosphorescence, 244 photons interaction and groups, 244 PMI determination, 245 sensitive technique, 242

Index TOC. See Time of colonisation Toxicology, aircraft accidents blood alcohol levels, 165 carbon monoxide and cyanide, 165–166 crop spraying accidents, 167–168 disruption and contamination, 164–165 therapeutic and “over-the-counter” drugs, 167 Transmission electron microscopy (TEM), 86 Trauma, 153, 158, 159 W World Trade Center (WTC) See 9/11 Attacks