morphology and pathoanatomy of the cervical spine facet ...€¦ · lars uhrenholt, edith nielsen,...

Morphology and pathoanatomy of the cervical spine facet joints in road traffic crash fatalities with

emphasis on whiplash – a pathoanatomical and diagnostic imaging study

PhD thesis

Lars Uhrenholt

Faculty of Health Sciences University of Aarhus

Institute of Forensic Medicine, University of Aarhus

Nordic Institute of Chiropractic and Clinical Biomechanics, Part of Clinical Locomotion

Science, University of Southern Denmark, Odense

Department of Neuroradiology, Aarhus University Hospital, Aarhus Sygehus (NBG)

Institute of Pathology, Aarhus University Hospital, Aarhus Sygehus (THG)

Morphology and pathoanatomy of the cervical spine facet joints in road traffic crash fatalities with emphasis on whiplash – a pathoanatomical and diagnostic imaging study Cover illustration Parasaggital histological section of a lower cervical spine facet joint illustrating a synovial fold. The thesis has been submitted to Faculty of Health Sciences, University of Aarhus, Aarhus, Denmark, in partial fulfilment of the requirements for the PhD degree. ©2007 Lars Uhrenholt Faculty of Health Sciences University of Aarhus Denmark Printed by SUN-TRYK, Aarhus 2007

ii

Preface

The present PhD thesis is based on the original work carried out in the period November 2002 to august

2007 at the Institute of Forensic Medicine, University of Aarhus in collaboration with the Nordic Institute

of Chiropractic and Clinical Biomechanics, Part of Clinical Locomotion Science, University of Southern

Denmark, Odense, the Department of Neuroradiology, Aarhus University Hospital, Aarhus Sygehus

(NBG) and the Institute of Pathology, Aarhus University Hospital, Aarhus Sygehus (THG).

The thesis has been submitted to the Faculty of Health Sciences at the University of Aarhus in order to

fulfil the requirements for attaining the PhD degree. The study was approved by the Scientific Ethics

Committee, Central Denmark Region, and The Danish Data Protection Agency.

All the material included in the work has been obtained at the Institute of Forensic Medicine, University of

Aarhus. Neuroradiological diagnostic imaging procedures have been conducted at the Department of

Neuroradiology, Aarhus University Hospital, Aarhus Sygehus (NBG). Macroscopical processing has

been performed at the Institute of Forensic Medicine, University of Aarhus whereas microtomal

sectioning has been performed at the Research Unit of Rheumatology and Bone Biology located at the

Department of Connective Tissue, Institute of Anatomy, University of Aarhus (formerly at the Institute of

Pathology, Aarhus University Hospital, Aarhus Sygehus (THG)).

The author sincerely appreciates the affiliations with the Nordic Institute of Chiropractic and Clinical

Biomechanics (NIKKB), Odense and the current part-time research position at the Institute of Forensic

Medicine, University of Aarhus. Financial support to the study has been generously granted from the

Foundation for the Advancement of Chiropractic Research, Denmark, the European Chiropractors Union

Research Fund (Grant # A.03-5), Switzerland, the Aarhus University Research Foundation, Denmark,

the Faculty of Health Sciences, University of Aarhus, Denmark, the A.P. Møller Foundation for the

Advancement of Medical Sciences, Denmark, the “Helga og Peter Kornings Fond”, Denmark and a

scholarship from the Spinal Research Institute of San Diego, USA.

iii

Acknowledgements

This thesis is the result of several years of fruitful collaboration between several institutes at the

University of Aarhus, Aarhus University Hospital and the Nordic Institute of Chiropractic and Clinical

Biomechanics (NIKKB), Odense and without the involvement of many people throughout its course the

completion of the thesis would not have been possible.

I would like to express my sincere gratitude to my chief-supervisor Markil Gregersen for giving me the

unique opportunity to explore a branch of forensic medicine, his genuine interest in the project and the

contributions, advices and support throughout the course of the project. Special thanks to my

supervisors, Annie Vesterby Charles, Edith Nielsen and Ellen Hauge who have all contributed immensely

in large periods of the project. I am grateful for your undeniable enthusiasm and positive approach to

solving many of the difficulties this project has encountered along the way. I would like to express my

deepest respects to my supervisor professor Flemming Melsen, DMSc. who sadly passed away

unexpectedly in the winter of 2004. Flemming Melsen contributed significantly to the original design and

concept of the project with his extensive experience in biology and pathology of cartilage and bone, and

his fantastic personality remains greatly missed.

The staff at the department of Neuroradiology, Aarhus University Hospital, Aarhus Sygehus (NBG) are

thanked for their enthusiastic commitment in the project. Without the logistic control of Niels Movim and

his administrative staff booking for diagnostic imaging would have been impossible as I always requested

timeslots within a few hours notice. This was furthermore supported by the staff on the ward that assisted

in a smooth and effective completion of each imaging session, including the nurses, porters,

radiographers and radiologists. A special thank goes to Vibeke Fink-Jensen who contributed with

unbiased second opinion evaluation of the diagnostic images.

My colleagues at the Institute of Forensic Medicine deserve a special acknowledgement. I appreciate the

skilful and professional cooperation of the forensic technicians who were engaged in the practical part of

this project – Karsten Mildahl Nielsen, Arne Nissen Ernst, Allan Laursen, Iuri Podgorbuschih and Jan

Schou. From the other side of the “barbed wire” the secretaries supported my endeavours emphatically

during the ever so necessary coffee break which coincidentally was also the most likely location to

v

receive the invaluable computer support from Ole Lundskov. Particularly, Tove Hansen must be thanked

for the many hours she spent entering data into Epidata©. Furthermore, my small office enjoyed the

companionship of many different doctors during the last few years, with some of whom moments of

despair as well as exhilaration have been shared. Also the medical staff, who were so fortunate not to

share my office, deserve my compliments as they reinforced an excellent environment for scientific

research. A tremendous effort has been put into this project by Rita Ullerup at the Research Unit of

Rheumatology and Bone Biology. Her unique insight and experience in the fields of processing hard

tissue samples has been vital for the results and I appreciate the many occasions where we have shared

thoughts and sought for answers. Similarly, the discussions with Eva Olesen and Jytte Jakobsen

concerning histological problems and last minute decisions have always been of great use. Also thanks

to the Orthopaedic Research Laboratory, Aarhus University Hospital, for their assistance with laboratory

procedures when the time schedule was pressed to the limits.

The Research Unit of Rheumatology and Bone Biology, Aarhus University Hospital, has been a

formidable setting for discussion and presentation of preliminary findings which I have enjoyed. The

thoughtful and educative statistical counselling from Niels Trolle and Morten Frydenberg is greatly

appreciated as this allowed detailed evaluation of the large datasets. The support from NIKKB has been

important with Ulla Dinesen being the window of opportunity as she strictly managed the funding

obtained and finally the staff at Kiropraktisk Klinik Nortvig & Uhrenholt deserves my compliments for their

ever positive attitude despite my continuing changes in schedules.

To those who mean the most to me; my loving wife Hedvig and our dearest children Sofie and Sebastian.

To the fond memory of my mother.

Lars Uhrenholt

August 2007

vi

List of appended papers

The PhD thesis is based on the following papers:

I. Imaging occult lesions in the cervical spine facet joints

Lars Uhrenholt, Edith Nielsen, Annie Vesterby Charles, Ellen Hauge and Markil Gregersen

(Accepted for publication in American Journal of Forensic Medicine and Pathology)

II. Pathoanatomy of the lower cervical spine facet joints in motor vehicle crash fatalities

Lars Uhrenholt, Annie Vesterby Charles, Ellen Hauge and Markil Gregersen

Submitted

III. Histomorphology of the lower cervical spine facet joints

Lars Uhrenholt, Ellen Hauge, Annie Vesterby Charles and Markil Gregersen

(Submitted to Scandinavian Journal of Rheumatology, under revision)

vii

Abstract

A major issue in neck pain syndromes following road traffic crashes, including whiplash trauma, is the lack of

objectifiable organic injury in the majority of cases. Hence, detailed description of potential somatic injuries in the

cervical spine is necessary in order to improve differential diagnostics and therapeutic management strategies. The

present PhD study examined the lower cervical spine facet joints from 19 persons killed in road traffic crashes and

21 persons dead due to non-traumatic causes. Conventional X-rays, computed tomography, magnetic resonance

imaging, stereomicroscopy, and light microscopy were used. The aim was to examine whether discrete osseous and

soft tissue injuries could be detected in the lower cervical spine facet joints of road traffic crash fatalities, in

comparison to people who had died due to non-traumatic causes, using these techniques. The study showed that

discrete osseous lesions in the lower cervical spine facet joints could be detected by means of advanced diagnostic

imaging procedures, in particular using computed tomography, although approximately half of the unique discrete

facet fractures could not be detected. Soft tissue injuries including bleeding in the cervical facet joints and synovial

folds could not be determined reliably on any diagnostic imaging modality, including MRI, despite histological

evidence hereof. Furthermore, discrete non-fatal injuries were commonly identified in the road traffic crash fatalities,

including facet fractures, haemarthrosis, disruption and bleeding in the synovial folds. Pathology was best

appreciated on light microscopy in comparison to stereomicroscopic evaluation, and conventional autopsy

procedures did not reveal any of the injuries. Histomorphological examination showed that structural changes occur

early in life in both cartilage and subchondral bone. The most common findings were superficial flaking, horizontal

split and vertical fissures of the cartilage. The number of facets with flaking, horizontal split and osteophytes, and the

subchondral bone thickness increased with age. Males were generally more severely and frequently affected by

these changes than females. The presence of two synovial folds in each facet joint was documented irrespective of

age and gender. The calcified cartilage was found to be constantly equal to about 10 % of the total articular cartilage

thickness in all ages. Similarly, the articular facets were almost completely covered by hyaline cartilage with non-

covered bony edges being a rare finding.

In conclusion, this study has provided a detailed description of the histomorphology of the lower cervical spine facet

joints of a young population with regard to the soft tissues, cartilage and subchondral bone. It confirmed that there is

an under-representation of soft tissue injuries and discrete fractures in the cervical spine facet joints in people killed

in road traffic crashes examined by standardised autopsy and diagnostic imaging. It is possible that such injuries

may be present in casualties after non-fatal road traffic crashes, however the potential clinical implications of the

lesions identified in this study are not known.

viii

Dansk resumé

Et væsentligt problem med nakkesmerter efter trafikulykker, herunder whiplash traumer, er manglen på

objektiviserbare vævsskader i størstedelen af tilfældene. Af denne grund er en detaljeret beskrivelse af potentielle

vævsskader i halshvirvelsøjlen nødvendig for at forbedre diagnostik og behandlingsmæssige tiltag. Det nærværende

ph.d.-projekt undersøgte de nedre facetled i halshvirvelsøjlen fra 19 trafikdræbte og 21 personer døde af non-

traumatiske årsager. Der benyttedes konventionel røntgen, computer tomografi skanning, magnetisk resonans

skanning, stereomikroskopi og lysmikroskopi. Formålet var at undersøge om diskrete osseøse skader og

bløddelsskader kunne identificeres i halshvirvelsøjlens nedre facetled hos trafikdræbte til sammenligning med døde

af ikke-traumatiske årsager ved brug af disse teknikker. Studiet viste at diskrete osseøse skader i facetleddene

kunne identificeres ved hjælp af avancerede billeddiagnostiske teknikker, især computer tomografi, til trods for at

omkring halvdelen af alle unikke facetfrakturer ikke blevet opdaget. Bløddelsskader, inklusive blødning i ledspalten

og leddets folder kunne ikke sikkert bestemmes på nogen af de billeddiagnostiske teknikker, herunder magnetisk

resonans skanning, til trods for histologisk evidens for tilstedeværelse. Desuden fandtes diskrete ikke-dødelige

skader hyppigt hos de trafikdræbte, herunder frakturer af leddets facet, blødning i ledspalten, ødelæggelse af og

blødning i leddets folder. Skaderne kunne bedst identificeres ved lysmikroskopi til sammenligning med

stereomikroskopi, og konventionel autopsi kunne ikke identificere nogen af skaderne. Histomorfologisk undersøgelse

viste at strukturelle forandringer sker tidligt i livet i ledbrusken såvel som i den subkondrale knogle. De hyppigste

fund var overfladisk fibrillation, horisontale og vertikale revner/fissurer i ledbrusken. Antallet af facetter med

overfladisk fibrillation, horisontale revner og osteofytter, og den subkondrale knogletykkelse steg med alderen.

Mænd var påvirket hyppigere heraf og i sværere grad end kvinder. To synoviale folder blev identificeret i hvert

facetled uafhængigt af køn og alder. Den forkalkede ledbrusk var konsekvent omkring 10 % af den totale tykkelse af

ledbrusken i alle aldre. Ligeledes var ledfacetterne næsten fuldstændigt dækkede med hyalin ledbrusk og udækkede

knogleender var sjældne.

Sammenfattende har dette studium bidraget med en detaljeret beskrivelse af halshvirvelsøjlens nederste facetleds

histomorfologi i en ung population, med hensyn til bløddelene, brusken og den subkondrale knogle. Det blev

bekræftet, at der er en under-rapportering af bløddelsskader og små frakturer i halshvirvelsøjlens facetled hos

trafikdræbte undersøgte ved konventionel autopsi og billeddiagnostisk undersøgelse. Det er muligt at sådanne

skader også kan være til stede hos tilskadekomne efter ikke-dødelige trafikulykker, men de mulige kliniske

betydninger af læsionerne identificeret i dette arbejde er ikke kendte.

ix

Content

1. Introduction 15

1.1. Aim 17

2. Types of cervical spine trauma during road traffic crashes 18

2.1. High energy crashes 18

2.2. Low energy crashes – whiplash trauma 18

2.2.1. Pathomechanics of the cervical spine during whiplash trauma 19

2.2.2. Incidence rates of acute whiplash injury 19

2.2.3. Risk of acute whiplash injury 21

2.2.4. Risk of chronic whiplash symptoms 21

3. Anatomy and degenerative changes of the cervical spine 22

3.1. The cervical spine 22

3.2. Anatomy of the lower cervical spine facet joints 24

3.2.1. Histology and histomorphometry of the cervical spine facet joints 26

3.2.2. Degenerative and age-related changes of the cervical spine facet joints 28

3.2.3. Grading and staging of osteoarthritis of the cervical spine facet joints 28

4. Pathology of the cervical spine after road traffic crashes 30

4.1. Fatal injuries in fatal road traffic crashes 30

4.2. Non-fatal injuries in fatal road traffic crashes 32

4.3. Injuries following survivable road traffic crashes 32

4.3.1. Whiplash injury 32

5. Diagnostic imaging of the cervical spine after road traffic crashes 36

5.1. Imaging modalities 36

5.2. Fatal road traffic crashes 37

5.3. Survivable road traffic crashes 38

6. Materials and methods 39

6.1. Materials 39

6.2. Retrieval of specimens 39

6.3. Diagnostic imaging procedures 40

6.3.1. The first examination of the radiological images 40

6.3.2. The second examination of the radiological images 41

6.3.3. The third examination of the radiological images 41

x

6.4. Histolaboratory procedures 41

6.4.1. Production of the 3-mm thick anatomical slices 43

6.4.2. Examination of the 3-mm thick anatomical slices 43

6.4.3. Production of the 10 μm thick histological sections 45

6.4.4. Examination of the 10 μm thick histological sections 45

6.4.5. Handling of data from the 10 μm thick histological sections 47

6.4.6. Evaluation of the inter- and intraobserver agreement of the morphological findings 51

6.5. Police records and information from the medico-legal autopsy 51

6.6. Statistical analysis 51

7. Summary of results 53

7.1. Diagnostic imaging findings (Paper I) 53

7.2. Pathological findings (Paper II) 58

7.3. Anatomical and age-related findings (Paper III) 62

7.4. Other findings 68

7.4.1. Inter- and intraobserver agreement 68

7.4.2. Police records 68

8. Discussion 70

8.1. Diagnostic imaging of the cervical spine facet joints after trauma 70

8.2. Pathology of the cervical spine facet joints after trauma 72

8.3. Anatomy of the lower cervical spine facet joints 74

8.4. Methodological considerations 76

8.5. External validity and clinical implications 80

9. Conclusions and perspectives 83

10. References 85

Appendices 105

Papers

xi

List of abbreviations

BAC Blood alcohol concentration

CT Computed tomography

IAP Inferior articular process

LOSRIC Low speed rear-impact collision

MMA Methylmethacrylate

MRI Magnetic resonance imaging

OA Osteoarthritis

RTC Road traffic crash

SAP Superior articular process

WAD Whiplash-Associated Disorders

V Velocity change (Delta-V)

xii

List of figures

Figure 2.1 Segmental pathokinematics with segmental hyperextension “S-shape”

Figure 3.1 The cervical spine

Figure 3.2 Overview of the lower cervical spine facet joint structures

Figure 3.3 The arcadial and zonal organisation of articular cartilage of a cervical spine facet

Figure 4.1 Lesions reported in humans subjected to cervical acceleration-deceleration trauma

Figure 6.1 Illustration of a cervical spine specimen

Figure 6.2 Illustration of the sawing of 3-mm thick slices from a cervical spine specimen

Figure 6.3 Setup for histomorphometric measurements

Figure 6.4 Histomorphometric quantitative measurements

Figure 6.5 Random location of histomorphometric measurements

Figure 7.1 Discrete injuries to the cervical spine facet joint soft tissue structures

Figure 7.2 Cervical spine facet fracture in a motor vehicle crash fatality

Figure 7.3 Injuries to the cervical spine facet joint in a motor vehicle crash fatality

Figure 7.4 Haemarthrosis in a cervical spine facet joint in a fatal motor vehicle crash victim

Figure 7.5 The morphological variables

Figure 7.6 Complete covering of the articular surface by the synovial folds

xiii

List of tables

Table 4.1 The Quebec Task Force grading system of Whiplash-Associated Disorders (WAD)

Table 6.1 Description of the morphological and histomorphometric variables

Table 7.1 Subjects with unique facet fractures on either histology or CT consensus

Table 7.2 Histopathological findings in the road traffic crash fatalities

Table 7.3 Morphological findings of the lower cervical spine articular facet

Table 7.4 Histomorphometric findings of the lower cervical spine articular facet

Table 7.5 Inter- and intraobserver agreement regarding microscopy of morphological variables

xiv

15

1. Introduction

Each year several millions of people are injured in road traffic crashes (RTC) world-wide with an

estimated 1.7 million injuries and more than 40.000 fatalities in Europe alone, although the figures are

declining 1-3. In Denmark, the official total number of injured people recorded by the police was 6.588 with

331 people killed in 2005. In comparison to the “White Paper” published in 2001 by the European

Communities on road traffic safety 2, the Danish figures are approximately equivalent to a reduction of 34

% and 28 % in fatalities and casualties respectively over the first five-year period (2000 – 2005) and thus

fulfils the part-requirements of the White Paper 4.

Despite the overall positive development, the number of relatively slight or minor injuries following RTC’s

have been reported to be increasing, particularly involving whiplash injury which is the most common

“minor” injury following RTC’s 5-10. In Denmark, a conservatively estimated 6.000 new cases of whiplash

injuries occur each year 11. Whiplash injuries, and in particular related symptoms in and from the neck,

are responsible for substantial morbidity with adverse effects on the quality of life for the affected

individuals as well as large health care expenditures 1;2;11-17. The exact costs related to whiplash injuries

are unknown, however the annual costs in the United States has been estimated to be in the range of

billions of US dollars annually 18;19. In Western Europe the annual costs have been estimated to be 8

billion Euros each year 20, and in Sweden the expected future costs related to one years casualties has

been estimated to four billion Swedish Kroner 12.

The most common symptom after whiplash injury is neck pain and the cervical spine facet joints have

been found to be closely related to persistent symptoms based on several clinical studies of patients

suffering from chronic symptoms after whiplash injury 21-25. Post-mortem studies have identified

pathoanatomical lesions in the cervical spine in both people suffering from chronic neck pain after

survivable road traffic crashes 26-29, and fatal road traffic crashes with high incidence rates of lower

cervical spine facet joint injuries 30-32.

In many cases the exact cause of the patient’s symptoms after whiplash injury cannot be established and

currently no pathognomonic whiplash injury has been identified. Despite the fact that a possible

16

explanation of the most common acute symptoms after whiplash injury can be found in organic pathology

of lower cervical spine facet joints in whiplash patients, the knowledge of tissue injuries following

whiplash trauma is limited and remains inconclusive for the majority of whiplash related conditions 14;24;33-

35. Similarly, only few controlled post-mortem studies have examined the incidence rates of specific

lesion types in these joints 31;32;36. If injuries are identified in these joints in people killed in road traffic

crashes this may support the theories of an underlying organic cause of symptomatology in subgroups of

whiplash patients 30;33;34;37;38.

Conventional X-ray studies of the lower cervical spine facet joints of whiplash patients rarely report

pathological findings, and fractures of the articular columns have usually been identified in severely

injured patients 39-41. In a number of imaging studies of acutely injured whiplash patients using magnetic

resonance imaging (MRI) injuries have been identified primarily in the intervertebral discs and correlating

the findings with the symptoms and the preceding trauma have been found difficult in many cases 42-45.

However, none of these studies have evaluated the cervical spine facet joints in any great detail. In

several studies of people killed in road traffic crashes, examined with conventional X-rays, computed

tomography (CT) and MRI, the diagnostic imaging procedures have only identified a small part of

pathological lesions that were observed microscopically in the lower cervical spine facet joints 30-32;36;46.

Despite these inconclusive results no studies have evaluated the cervical spine facet joints in detail,

including correlating findings from microscopy with diagnostic imaging. Furthermore, the interobserver

agreement of the diagnostic imaging examinations of cervical spine facet has not been tested in a

blinded and controlled fashion. Hence, the incidence rates of injuries to the cervical spine facets detected

on diagnostic imaging, the correlation of these findings to microscopy and the reliability of the radiological

evaluations have not been vigorously tested.

Numerous studies have investigated the cervical spine morphology and histomorphometry including the

properties of the articular facets and their soft tissues 47;48;48-56. However, quantitative data regarding the

properties of the hyaline cartilage, calcified cartilage and subchondral bone of the cervical spine facets is

very limited 48.

17

1.1. Aim

It was hypothesised that there is an under-representation of pathoanatomical lesions in the cervical spine

facet joints in people killed in RTC’s examined by standardised autopsy and conventional radiological

evaluation.

The primary aim of the present PhD thesis was to investigate whether discrete lesions in the lower

cervical spine facet joints could be detected in people killed in a passenger car crash by advanced

diagnostic imaging procedures, including conventional X-rays, CT and MRI, and specialised

microscopical evaluation as compared to non-traumatized decedents. The morphology of the lower

cervical spine facet joints in a group of young individuals (age 20-49 years) should be evaluated

qualitatively and by histomorphometry. Furthermore, attempts were made to compare the pathological

findings in the people killed in the passenger car crashes with data from the police records with regard to

mechanism of trauma.

The ultimate goal was to improve our knowledge of discrete spinal injuries in road traffic crash victims

which may help improving diagnostic and therapeutic possibilities in survivors of RTC’s sustaining non-

fatal injuries to the cervical spine facet joints.

18

2. Types of cervical spine trauma during road traffic crashes

There are an infinite number of possible types of RTC’s. With the purpose of broad classification these

are often described according to the direction of first impact (which relates to the primary direction of

force) and severity as it relates to the relative energy transfer (high energy versus low energy crashes),

which is often referred to as high speed and low speed crashes respectively.

2.1. High energy crashes

Far the majority of fatal car crashes involve high energy transfer to the decedent as a consequence of a

high speed crash involving a single vehicle in a rollover or a frontal collision between two vehicles 57-59.

However, other vectors are also implicated in fatal crashes, for example the American National Highway

Traffic Safety Administration found that 6.5 % of people killed in passenger car crashes in the United

States were killed in a rear-impact collision in 2005 58. The mechanics of injury in high energy crashes

are often complex with multiple different types of injurious loading of the occupants 60;61. The exposure of

high energy load over a short time-span carries a high risk of serious bodily injury which is characteristic

of high speed crashes, with the risk of serious injury and death being closely related to the velocity

change ( V) as a function of time, i.e. acceleration, of the involved vehicles and occupants 57;59.

2.2. Low energy crashes - whiplash trauma

In contrast to high speed RTC’s, low speed crashes (low energy crashes) are far more common. The

majority of all injurious crashes take place with velocity changes below 30 km/hour 14;59;62-65. The classical

whiplash trauma is a low speed rear-impact collision (LOSRIC), where the collision between two vehicles

exposes the driver and/or passengers of the “target” vehicle to a biphasic acceleration-deceleration force

loading with principal inertial loading of the occupants, i.e. without head strike 14;59. However, other types

of trauma, e.g. motor vehicle collisions involving head strike, skiing accidents and assaults, are regularly

classified as whiplash trauma due to similar symptomatology which makes the group of whiplash patients

heterogeneous affecting the efficiency of clinical classification systems, e.g. the WAD-grading system 66.

19

2.2.1. Pathomechanics of the cervical spine during whiplash trauma

The early whiplash models hypothesised that the most injurious mechanism involved hyperflexion and in

particular hyperextension of the cervical spine during rear-impact 67. Later more complex kinematics

were identified and the observations were described as a “serpentine configuration” of the cervical spine

during a rear-impact collision 68. However, it was not until the mid-1990’es that experiments using human

volunteers and post-mortem human subjects identified segmental biphasic pathomechanics during

experimental indirect rear-impact acceleration loading 69;70. Biomechanically these rear-impact collisions

usually take place within a timeframe of approximately 150 milliseconds, during which the occupants are

exposed to well-described patterns of kinematics. This involves forward acceleration of the torso with

ramping causing compression of the cervico-thoracic spine and potentially pathophysiological segmental

movements of the cervical spine due to changes in the instantaneous axis of rotation. Consequently,

shear, compression and distraction loading of particularly the facet joints takes place with the

simultaneous “s-shape” configuration of the cervical spine (Figure 2.1) 69-74.

The abnormal kinematics are potentially injurious to the regional anatomical structures as the normal

anatomical boundaries of mobility are exceeded within the first 75-100 milliseconds of the collision

14;19;37;70;75;76. This has recently been supported in biomechanical experiments where the facet joint

capsular ligaments and adjacent soft tissues are at risk of injury at acceleration forces typical for low

speed rear-impact collisions 73;77-81. Since the kinematic changes take place very early on during

physiological loading there is no or only limited protective reflex contraction of the neck muscles 37;82;83,

and as the pathophysiological movements take place within the normal range of motion there is often no

global hyperextension or hyperflexion of the cervical spine 14.

2.2.2. Incidence rates of acute whiplash injury

Despite the recent decrease in the number of people killed and severely injured in RTC’s in most

European countries there appears to have been an increase in the number of people sustaining non-fatal

injuries 7;10. The incidence rates of whiplash injuries have not been confidently established, probably due

to differences in sampling systems and geographical and cultural differences. However, incidence rates

have been estimated in several countries ranging from 10 per 100.000 of the general population of New

Zeeland 84, 70 per 100.000 inhabitants annually in the province of Quebec (Canada) 66, 417 per 100.000

inh

20

Figure 2.1 Segmental pathokinematics with segmental hyperextension ”S-shape”

The normal correlation in the cervical spine facet joint (A). During typical rear-impact loading (i.e. loading from the

posterior) the cervical spine undergoes a segmental pathophysiological hyperextension “S-shape” associated with

shear forces, compression forces posteriorly and distraction forces anteriorly acting on the joint structures (B).

21

inhabitants in the United Kingdom 7 to as high as 1.172 per 100.000 inhabitants annually in USA 14. In

Denmark, an estimated 120 per 100.000 inhabitants sustain a whiplash injury each year 11;85. However,

since most of the official figures are based primarily on police records and hospital admission records

they must be regarded with caution. Under-reporting of injuries sustained in crashes with limited or no

property damage is likely as these crashes are rarely attended by the police. Furthermore, hospital

admission records do not include injured people contacting the general practitioner or other

physicians/therapist’s directly, as well those people not contacting any medical personnel 86;87.

2.2.3. Risk of acute whiplash injury

Although the most common form of whiplash injury carry a remote life-threatening potential, a substantial

risk of acute injury (here defined as relevant symptoms after the trauma) is present and relates to multiple

crash- and personal related factors 14;59. The immediate risk of sustaining an acute whiplash injury after a

whiplash trauma is approximately 30-60 % 14;18;88-90, with a wide range of severity of symptoms,

developing most often within the first 48 hours of the crash, although delayed onset of symptoms has

been reported 14;91-94. As a combination of crash severity parameters and individual susceptibility to injury,

which if affected by biological variability, uniquely relates to the outcome of any given crash, no

generalised injury threshold has been established for low speed crashes 88;95-97.

2.2.4. Risk of chronic whiplash symptoms

Approximately 33-43 % of the originally injured subjects in a whiplash trauma develop longstanding

symptoms of varying severity 18;91. Notwithstanding these figures, official figures from several countries,

including the National Board of Health in Denmark, estimate that the risk of developing chronic symptoms

in Denmark is below 5 % 11, and Swedish risk figures range between 5-10 % 98. The inconsistencies

between international publications and the official national figures are furthermore troublesome taking into

consideration recent results from prospective studies of Danish whiplash injuries in which persistent

symptomatology at one year after injury were reported in approximately 50 % of the patients 13;99. The risk

of sustaining chronic disabling symptoms that affects the working ability has recently been found to be as

high as 12 % 17;99 and 25 % 13 in two independent prospective studies. Although the term disability is not

uniquely defined in all studies, similar reports of lower percentages of long-term disability have previously

been described 100-105.

22

3. Anatomy and degenerative changes of the cervical spine

The cervical spine consists of seven vertebrae, with the two upper vertebrae; atlas (C1) and axis (C2)

being atypical vertebrae and the remaining five vertebrae (C3-C7) being classified as typical vertebrae.

The vertebral bodies of atlas and axis have unique anatomical appearances and consequently unique

biomechanical properties. In the lower cervical spine (C3-C7) osseous elements extends posteriorly from

the vertebral bodies, consisting of pedicles, articular columns, transverse processes, laminae and

spinous processes (Figure 3.1). The primary functions of the cervical spine are to supply support and

mobility to the head and provide protection to the spinal cord.

3.1. The cervical spine

With the exception of occiput-C1 and C1-C2, each cervical vertebral body is connected to each other by

an intervertebral disc which is bordered postero-laterally by the uncinate processes forming the

uncovertebral joints of Luschka 49;106;107. Posteriorly the facet joints of the articular columns supplements

to the mobility and stability of the cervical spine and guide weight bearing. The cervical spine mobility is

highly specialised depending on the anatomical location 108-110. In the upper cervical spine the primary

movement between occiput-C1 is flexion/extension and between C1-C2 it is rotation 108. In the lower

cervical spine all vertebral levels contribute to flexion, extension, rotation and lateral flexion although

some segments are more active than others 108;109. Ligaments are widely dispersed in the cervical spine

and their appearance reflects their supportive function. In the upper cervical spine the principal ligaments

are the transverse ligament, the apical ligament, the alar ligament, the capsular ligaments and the

anterior and posterior atlanto-occipital membrane. These ligaments check movement and are

responsible for limiting motion thereby adding stability. In the lower cervical spine the primary ligament

are the anterior and posterior longitudinal ligaments, the ligamentum flavum, the interspinous ligaments

and the capsular ligaments. Surrounding the skeletal cervical spine, uniquely specialised musculatures

add function to the highly mobile region. In the upper cervical spine the suboccipital muscles control

mobility in combination with the sternocleidomastoideus, trapezius and some of the deeper neck

23

Figure 3.1 The cervical spine

The cervical spine is illustrated in overview from the side with a typical vertebra shown from superior and the

lateral aspect.

24

muscles. In the lower cervical spine muscles controlling mobility include primarily the scaleneus group,

the levator scapulae, the trapezius and the anterior flexor muscles. There is an intricate nerve supply to

the cervical spine. Nerves from the cervical and brachial plexus divide in branches supplying the cervical

spine bones, ligaments, intervertebral discs, muscles and other soft tissues in a segmental organisation.

Autonomic fibers, relaying in the cervical sympathetic ganglion and coursing with the vasculature as well

as arising from the spinal nerves, also supply some of the cervical spine structures. The blood supply to

the cervical spine arises primarily from the brachio-cephalic trunk, the common carotid arteries and the

vertebral arteries. Spinal and muscular branches supply the vertebral bodies, the posterior elements and

the adjacent musculature of the posterior part of the cervical spine.

3.2. Anatomy of the lower cervical spine facet joints

The lower cervical spine (C3-C7) facet joints are arranged in pairs bilaterally and each joint consist of

opposing articular facets, i.e. the inferior articular process (IAP) of the vertebra above and the superior

articular process (SAP) of the vertebra below (Figure 3.1). The facets are oriented slightly oblique to the

saggital plane, with the superior articular facets facing laterally and upwards approximately 45o to the

horizontal plane, with gradual changes in orientation in the cervico-thoracic transitional zone. The

articular facets are covered completely or in part by hyaline cartilage as the presence of cartilage free

areas, i.e. “cartilage gaps”, have recently been identified (Figure 3.2) 48. The facet joint capsules are

loose and thin with the strongest part in the antero-lateral aspect. Medially the capsular fibers are

contiguous to those of the ligamentum flavum and the weakest fibers are in the dorsal region 107;111. The

posterior joint capsule is supported by the attachment of the multifidus muscles at all levels 107. The

presence of synovial folds (fat pads, meniscoids or inclusions) of the anterior and posterior margins of

the lower cervical spine facet joints have been described inconsistently in several anatomical studies

(Figure 3.2) 49;111-115. A recent review noted that these structures are present as small vascular synovial-

lined fat pads, although no mentioning to the exact prevalence was noted 107, however one study have

identified meniscoids in all facet joints of 20 spines from decedents aged 20 to 80 years 116, and another

study identified anterior and posterior meniscoids in all six facet joints from one subject examined with

microscopy and high-field magnetic resonance imaging 117. The exact function and importance of the

synovial folds is not clearly understood. Immediately anterior to each facet joint the cervical spine nerve

root exits the spinal canal through the intervertebral foramen formed by the pedicles, vertebral bodies,

25

Figure 3.2 Overview of the lower cervical spine facet joint structures

Three facet joints from the lower cervical spine illustrated with two different staining techniques (Masson-Goldner

trichrome and Safranin-O fast green) showing hyaline cartilage (a), an anterior synovial fold (b), a posterior joint

capsule (c), and a nerveroot (d) on selected joints, original magnification x1.

26

intervertebral disc and the articular column 106;107;118. At each spinal segment the recurrent (spinal)

meningeal nerve of Luschka re-enters the spinal canal via the foramen in order to supply the outer part of

the intervertebral disc with sensory and sympathetic autonomic fibers. The medial branches and

accessory nerves of the primary dorsal rami supply the articular structures, including the bone,

musculature, capsule and synovial folds, but not the articular cartilage, with nociceptive fibers and

mechanoreceptive nerve endings at a segmental organisation 119;120. Each medial branch curves the

ipsilateral articular column and sends ascending and descending branches to the joints above and below

121. The spinal and accessory branches of the vertebral arteries anastomise longitudinally and course the

intervertebral foramen where they continue as anterior and posterior radicular arteries with vascular

supply to the facet joints, spinal cord and posterior elements of the spinal column and are surrounded by

the accessory vertebral vein 106.

3.2.1. Histology and histomorphometry of the cervical spine facet joints

The general histological appearance of human synovial joints, including the cervical spine facet joints,

has been described in detail 47-50;122;123. The hyaline cartilage covering of the cervical spine facets exhibit

the typical arcadial appearance (Bennington’s arcades) of cartilaginous matrix with zonal distribution

(Figure 3.3), although this is not always readily identified, e.g. at the periphery of the joint. The function of

the cartilage is to increase the area of load distribution and provide a smooth surface for mobility 122;124.

Articular cartilage is composed of more than 70% water and only 5% of the total volume is accounted for

by the chondrocytes 122-124. The major constituents of the extracellular matrix are the collagen type II and

proteoglycan aggrecan 122-124. Histomorphometrical evaluations of the cervical spine facets have been

performed and data are available concerning the anatomical properties of the posterior articular columns

and the orientation of the cervical facets 51-53;56. In macroscopical studies the facet joint width, as

measured in the coronal plane, has been measured to be about 11-12 mm 48;55 and the depth in the

saggital plane to be about 10 mm 55. The mean area of the inferior articular surfaces have been found to

increase from approximately 1.25 cm2 at C3 to 1.6 cm2 at C7 125. The mean overall thickness of the

cervical spine facet cartilage has been shown to be somewhat smaller in females (0.4 mm) compared to

males (0.5 mm) based on a cryomicrotomal study 48. In joints, other than the cervical spine facet joints,

the thickness of the calcified cartilage has been found fairly constant occupying approximately 5 % (3-

8%) of the total cartilage thickness 126-129. Apparently, no detailed histomorphometric information exists of

27

Figure 3.3 The arcadial and zonal organisation of articular cartilage of a cervical spine facet

Overview of a facet, original magnification x4 (A). Zonal organisation of the cartilage with the hyaline cartilage

divided into a superficial, middle and deep zone, original magnification x10 (B). Stained with Masson Goldner-

Trichrome.

28

the quantitative properties of the cartilage and subchondral bone of the lower cervical spine facets.

3.2.2. Degenerative and age-related changes of the cervical spine facet joints

The facet joints and the articular cartilage are likely to degenerate as a consequence of overuse and

repeated injury as well as ageing 49;107;122;130;131. However, it is important to note that osteoarthritis of the

facet joints is not necessarily a consequence of ageing, and that these joints are frequently unaffected in

spondylosis 49;106.

Generally, degeneration of the cartilage takes place over time, where the initial changes consisting of

only discrete superficial flaking (fibrillation) of the cartilage are gradually replaced by more aggressive

changes in the shape of vertical fissures, erosions, denudation and deformation 49;107;122;123;131. As a

consequence of the degenerative process the changes are most vividly visualised in the unmineralised

hyaline cartilage, but changes also take place in the calcified cartilage as well as the subchondral bone

which undergoes remodelling 122;127;129;132. Vascular invasion through the basophilic tidemark (the border

between the unmineralised hyaline cartilage and calcified cartilage) with increasing duplications of the

tidemark, has been related to ageing 122;133. New bone formation in the shape of osteophytes may

develop at the borders of the joints and cysts may form in the subchondral region 49;106;122;127;133. Hence,

age-related changes in the facet joints are generalised affecting the osseous, cartilaginous and the soft

tissue structures.

3.2.3. Grading and staging of osteoarthritis of the cervical spine facet joints

General histological classification and grading systems are available for the description of human joints

49;130;131;134-136. However, these are primarily based upon the examination of large joints, such as the hip

and knee joints and have not been validated for the use in the lower cervical spine facet joints, with the

exception of recent classification systems, some of which have been developed for the lumbar spine

facet joints 130;134;135;137. Grading of osteoarthritis of the cervical spine facet joints is made difficult by the

lack of a validated system for histological grading of these joints, although several general classification

systems are available 49;123;130;131;134-137.

A grading system has been developed for the lumbar spine facet joints, and has recently been utilized in

the evaluation of the cervical spine 130;135;137. However, it is questionable whether this grading system

supplies a valid interpretation of the true extent of degenerative changes as differentiation between

29

grades of severity is difficult. Recently, an important paper assessed current knowledge based on large

joints and proposed a system that takes into consideration both the grade (extent of widespread

changes) and the stage (severity) of the OA changes 131. Although this system is based on the evaluation

of large joints, and it has not been tested on smaller synovial joints such as the cervical spine facet joints,

it seems suitable for the examination of facet joints.

30

4. Pathology of the cervical spineafter road traffic crashes

By the Quebec Task Force on Whiplash-Associated Disorders (WAD) classification, whiplash is

classified according to the clinical presentation and the clinical findings (Table 4.1) 66. In case of severe

injuries such as brain injuries, cervical spine fracture or spinal cord injury, the grading is not applicable.

Hence, studies evaluating the clinical consequences of whiplash injuries most often refer to subjects

graded as WAD II or III, and the implications of WAD IV has not been investigated in great detail.

Therefore, somatic injuries have rarely been reported in studies of people sustaining whiplash injury

probably due to that fact that evidence of tissue damage often excludes participation in these studies

whereby the included subjects most often are graded WAD II. Hence, in subgroups of whiplash injured

subjects (e.g. WAD III and WAD IV) the structural integrity of relevant anatomical structures has not been

evaluated in detail which may have implications, not only for the management of these patients, but also

influence management strategies and utilization of diagnostic procedures in apparently less severely

injured people (e.g. WAD II).

4.1. Fatal injuries in fatal road traffic crashes

Far the most common non-survivable types of injuries are those related to the head- and cervical spine

complex. Injuries to the cervical spinal cord, brainstem and brain are often not survivable, and in many

cases part of complex injury patterns. Other fatal injuries include for example cardiac tamponade,

disruption of the aorta and severe contusion injuries of the pulmonary tree 60;138. Generally, the

decedents have in common, that they suffer several types of injuries to different body parts, and that

often several competing causes of death are present. The primary role of the medico-legal autopsy is to

identify and describe all injuries sustained by the decedent and conclude upon the cause/causes of

death 60;138. Hence, registration of fatal and non-fatal injuries is an important role of the medico-legal

autopsy.

31

Grades of severity Clinical presentation

0 No neck complaints; no physical sign(s)

I Neck pain, stiffness, or tenderness only; no physical signs

II Neck complaints AND musculoskeletal sign(s)*

III Neck complaints AND neurological sign(s)**

IV Neck complaints AND fracture or dislocation

*: Musculoskeletal signs include decreased range of motion and point tenderness

**: Neurological signs include decreased or absent deep tendon reflexes, weakness, and sensory deficits

Table 4.1 The Quebec Task Force grading system of Whiplash-Associated Disorders (WAD)

32

4.2. Non-fatal injuries in fatal road traffic crashes

Minor injuries after fatal road traffic crashes are not always described in same detail as severe injuries,

and discrete injuries may not be readily visualised during standardised autopsy 30;32;139;140. In the cervical

spine this is particularly true for micro-fractures, haemarthrosis and injuries to the facet joints,

intervertebral discs, ligaments and muscles 27;28;30-32;36;46;139;141.

4.3. Injuries following survivable road traffic crashes

A wide range of lesions have been reported in humans subjected to survivable road traffic trauma,

including cervical acceleration-deceleration trauma, i.e. also whiplash trauma (Figure 4.1) 14.

Furthermore, post-mortem studies of people previously suffering from chronic neck pain after road traffic

crashes have identified pathoanatomical lesions in the cervical spine that may have been caused by the

preceding trauma and have had clinical relevance for the decedents prior to death 26-29. It has been

shown that these lesions by in large are identical to those lesions identified in several post-mortem

studies of people killed in fatal road traffic crashes 30, and that injuries to the facet joints are particularly

common in the lower cervical spine 31;32. Although interesting, it is not known what relevance these post-

mortem findings have in comparison to injuries sustained in survivable road traffic crashes, and uncritical

extrapolation of autopsy findings to clinical settings is not admissible 19;30;33. Nonetheless, it seems likely

that in some cases injuries are sustained in survivable road traffic crashes that are similar to those

identified at post-mortem 30;33;34;38, and it has previously been suggested that convergent validity is

constituted by the fact that neck pain is common, particularly following whiplash injury to the cervical

spine and that discrete injuries to the cervical spine is common in post-mortem subjects killed in road

traffic crashes 34.

4.3.1. Whiplash injury

The exact cause of the patient’s symptoms after whiplash injury has not been established and currently

no pathognomonic whiplash injury has been identified. Hence, the definition of whiplash injury does not

include the description of an organic lesion in the cervical spine. Rather, the term whiplash injury refers

to the mechanism of trauma. Hence, a prerequisite for a whiplash injury is the exposure to a preceding

whiplash trauma, which can be strictly defined as an acceleration-deceleration trauma during a motor

vehicle collision, although other mechanisms of trauma have been known to cause similar sympt

33

Figure 4.1 Lesions reported in humans subjected to cervical acceleration-deceleration trauma

(Illustration modified from Foreman et al. 14)

34

vehicle collision, although other mechanisms of trauma have been known to cause similar

symptomatology 14;59;66. More recently the alternative term; Whiplash-Associated Disorders (WAD), has

been proposed together with a whiplash injury grading system (Table 4.1) 66. However, although the

WAD grading system allows some considerations regarding the severity of injury, the grading is primarily

related to the cervical spine and does not consider other signs or symptoms, e.g. lumbar spine pain.

Similarly, casualties eligible for the WAD grading IV, e.g. cervical spine fractures with spinal cord

compromise, are rarely scored according to the WAD grading system as their injuries are often life

threatening with consequently different therapeutic considerations 142. Nonetheless, the WAD-system

has some justification as the most common symptom following a whiplash injury is neck pain 14;21;59;66;142-

145. Since different directions of crash impulses, e.g. frontal versus rear-impact, causes different

interactions throughout the human body, and thereby carries different tissue damage and injury potential,

these parameters should be integrated in the diagnostic possibilities 14, and the duration of symptoms

and stage of recovery should also be more specifically incorporated 66;146. This was actually proposed in

an earlier cervical acceleration-deceleration classification system, published prior to the WAD grading

system 146;147.

In theory, acute neck pain can be explained by somatic injury, such as sprain/strain of cervical spine soft

tissues, contusion of articular structures, disruption of the cervical discs, damage to nerve roots and

microfractures of osseous structures (Figure 4.1) 24;33-35, however, it remains unclear to what extent these

injuries may apply to whiplash patients. Similarly, the exact mechanism of sustained chronic pain in the

cervical spine after whiplash is unclear, although it has been shown that alterations in the central

processing of sensory stimuli (central hypersensitivity) may be present 148-150. Such conditions are

thought to affect, not only the brains interpretation of ascending nociceptive information, but also

influence the segmental circuits on a spinal cord level whereby the patients’ experience of pain is altered

148-151. Alternatively, other non-organic underlying causes have been proposed with the bio-psychosocial

model and post-traumatic stress disorder defining the symptoms’ origin and maintenance as primarily

psychogenic 152-155.

The most common symptoms following an acceleration-deceleration injury are neck pain and headache

14;21;59;66;142-145 and the cervical spine facet joints have been found to be closely related to chronic neck

pain based on several clinical studies of patients suffering from chronic symptoms after whiplash injury 21-

25. Furthermore, several studies have found that abolishing the nerve supply to these joints, by nerve

35

blocks, radiofrequency neurotomy or intra-articular prolotherapy, neck pain may be alleviated for periods

of months 156-160. Hence, the cervical spine facet joints have high priority in the diagnostic evaluation of

whiplash patients. However, despite extensive research into the mechanisms of trauma and clinical

parameters, the knowledge of tissue injuries in the cervical spine facet joints following whiplash trauma is

limited and remains inconclusive for the majority of whiplash related conditions.

36

5. Diagnostic imaging of the cervical spine after road traffic crashes

5.1. Imaging modalities

In the diagnostic evaluation of victims from road traffic crashes the most commonly utilised diagnostic

imaging modalities include conventional X-rays, CT and MRI. For the purpose of screening for skeletal

injury after trauma, the American College of Radiology have proposed in their appropriateness criteria 161

that conventional X-rays are recommended as the primary choice of screening 162;163. However, the

appropriateness of any specific radiological examination is made by the referring physician and

radiologist in light of the particular circumstances for any given case 161. Hence, in increasingly severe

injuries CT is more likely chosen as the primary tool of screening if there is clinical indication of cervical

spine fractures 161;164-169.

Conventional X-rays are the most commonly utilised diagnostic imaging procedures for the screening of

injuries following trauma worldwide 162;170. The advantage of conventional X-rays is that it is

geographically easy accessible around the world, it is cost-efficient and it does identify orthopaedic

damage, e.g. extremity fractures, tumors and foreign objects. However, soft tissue injuries and minor

fractures are often not detectable on conventional X-rays. Furthermore, radiation safety must be

considered every time a patient is exposed to an X-ray exam 171;172.

Computed tomography is particularly sensitive towards detection of skeletal injury because of its

unparalleled resolution and illustration of bony detail and is therefore used extensively in the diagnostic

evaluation of severely injured casualties 14;164;167. Furthermore, CT allows good appreciation of

abdominal, thoracic, and pelvic organs as well as the brain. The accessibility to CT-scanners is good in

developed countries, and as the complete examination takes only minutes it is central in vital diagnostics.

The disadvantages of CT are the high costs and the small increased risk of adverse health effects from

cancer related to the radiation doses given to the patients 172.

Magnetic resonance imaging is based upon principles of magnetic fields and does not expose the

patients to radiation. It is superior with regard to diagnostics of the soft tissues and organs. Magnetic

resonance imaging is however not displaying fractures as well as CT, and albeit rare, injuries induced

37

during MRI procedures have been reported as the highly magnetic field may cause adverse effects on

implanted devices 173. The accessibility to MRI is fairly good in most developed countries, however the

major disadvantage of MRI is the accessibility, the time needed per procedure and the high costs.

5.2. Fatal road traffic crashes

There is an increasing demand for advanced diagnostic imaging in forensic settings, and CT and MRI

are regarded as useful supplements to the forensic autopsy 168;174-177.

Conventional X-rays inadequately identify discrete osseous lesions in the cervical spine in people killed

after significant trauma, including road traffic crashes 27;30-33;36;46;178-181. Furthermore, in a number of post-

mortem studies, detailed histological examination has shown injuries in the cervical spine that could not

be detected on conventional X-rays or CT 30;46;115;167;175;178;179;182;182-187, and MRI 46;162;167;168;178;187;188.

Despite these findings, a few studies have identified lesions in the cervical spine facet joints using

conventional X-rays 180;189. The soft tissue structures of the facet joints are poorly displayed on MRI

112;115;178. However, a recent publication have identified the cervical spine facet joint folds on a high

energy-field 3.0 Tesla MRI research unit 117 suggesting that with future advancements visualisation of the

facet joint soft tissues may become possible in clinical settings. Overall, there is a scarcity of publications

regarding the lower cervical spine facet joints and the incidence rates of facet fractures, haemarthrosis

and synovial fold injury in these joints based on different diagnostic imaging procedures of people killed

in RTC’s are unknown.

Currently, no large scaled evidence based research has evaluated the validity and reproducibility of

forensic diagnostic imaging procedures and the exact value of diagnostic imaging findings at post-

mortem remains unclear. Although the sensitivity and specificity of advanced diagnostic imaging

modalities (CT and MRI) gradually increase little is known with regard to the inter- and intraobserver

agreement in the field of post mortem imaging. Furthermore, as many such studies have not included

detailed microscopical evaluation of the anatomy visualised on diagnostic imaging, a gold standard (e.g.

microscopical findings) has not been used as reference. Rather, MRI and CT have not been found

adequate as gold standard when it comes to the identification of morphology and pathoanatomical

conditions 42;162;190.

38

5.3. Survivable road traffic crashes

Conventional X-ray studies of the lower cervical spine facet joints of whiplash patients rarely report

somatic lesions, although fractures of the articular columns have been identified in several studies of

traumatised patients, including road traffic crash victims 39-41;191;192. In contrast, conventional X-rays have

been shown to miss large numbers of discrete lesions in the cervical spine of survivors after road traffic

trauma 26;40;164-166;186;192-198. Many of these lesions are likely to be diagnosed on CT 162;164;167;186;198-200

whereas MRI is not expected to have the same sensitivity 42;162;190. Soft tissue lesions are not readily

detected on conventional X-rays as identification is based upon observed distortions of fat lines and

shadow lines. Prevertebral swelling may be present following road traffic crash trauma causing distortion

of the prevertebral shadows which can be identified on conventional X-rays 201-205, although this is much

better visualised on MRI. Another indication of soft tissue injury, and/or osseous injury, is the fanning of

the spinous processes which may be seen on conventional X-rays. This may be caused by injury to the

facet joints, e.g. joint capsule, synovial folds or articular surface which may be identified in some cases

on either conventional X-rays, CT or MRI 39-41;162;164;167;186;191;192;198-200. Changes in the cervical spine

lordosis is a frequent finding in whiplash patients and is thought to be related to factors such as post-

traumatic muscular hypertonicity and segmental articular dysfunction (biomechanical dysfunction)

14;202;206-208, although its exact correlation to trauma has been questioned 209;210. In follow-up studies of

patients suffering from chronic pain after whiplash injury it has been shown that high percentages of

patients develop degenerative changes in the cervical spine in comparison to non-traumatised subjects

211-213, however other studies have not found similar correlations 214.

In several MRI studies of people injured in road traffic crashes, including whiplash trauma, injuries to the

cervical spine have been identified, primarily in the intervertebral discs and ligaments 42-45;190;206-208;215-218.

However, the majority of findings in these studies were degenerative changes rather than acute injuries.

Attempts to correlate MRI findings with the symptoms and the preceding trauma have most often been

inconclusive 45;207;208;215;217;219. In recent years the primary focus for diagnostic imaging has been on the

upper cervical spine where different types of ligamentous injuries have been identified in whiplash

patients 220-222. However, there has been some debate with regard to the validity of these findings

although the body of evidence is mounting 223;224. Nonetheless, this anatomical region is beyond the

purpose of this study.

39

6. Materials and methods

6.1. Materials

Twenty motor vehicle crash fatalities (cases) and 22 non-traumatic decedents (controls) were examined

at medico legal autopsy within 1-5 days (median 3 days) after death and included in this study. The

cases were decedents from motor vehicle crashes in which the decedent had been an occupant of an

involved vehicle and the control subjects had died due to non-traumatic causes. All subjects had to be in

the age range of 20-49 years of either gender. Subjects were excluded if the case record showed

evidence of history of drug and/or alcohol abuse, if there had been any previous injury to the cervical

spine, if there was significant external evidence of cervical spine injury, and if diagnostic imaging

procedures could not be performed within 24 hours of the autopsy. After completion of the inclusion

period, two of the 42 subjects (one case and one control) were excluded due to extensive decomposition

causing generalized tissue damage. The final study group consisted of 40 subjects divided into 19 cases

(15 males (median age 33, range 20-47) and 4 females (median age 27, range 22-45)) and 21 controls

(13 males (median age 35, range 22-49) and 8 females (median age 41, range 27-49)). For the purpose

of this study only the lower cervical spine facet joints from C4-C5, C5-C6, C6-C7 and C7-Th1 were

examined.

The study was approved by the Scientific Ethics Committee, Central Denmark Region, and The Danish

Data Protection Agency.

6.2. Retrieval of specimens

During autopsy the lower cervical spine motions segments from C4 to C7 were retrieved en bloc from

each subject through a conventional “Y”-incision from the front of the neck according to previously

described procedures 225, and stability was restored with a specially designed brace, enabling completion

of the autopsy in such manner that no additional external evidence of the intervention was visible other

than was is normal following post-mortem autopsy. The osteo-cartilaginous specimens consisted of the

lower four cervical vertebral segments (C4-C7) including the vertebral body, pedicles, articular columns,

40

laminae, spinous processes and surrounding muscles without the covering skin. Hence, the facet joints

from C4-C5, C5-C6, C6-C7 and C7-Th1 bilaterally were included. Each specimen was assigned a

reference number (LU-reference), which uniquely identified the subject.

6.3. Diagnostic imaging procedures (see Article I for more details)

After retrieval each specimen was enveloped in a clear plastic bag and placed in a clear plastic box in

which a firm foam rubber structure gave support. The box with the specimen was placed in a clear

airtight envelope. The specimens were examined with the presence of the lead investigator and

consisted of conventional X-rays, computed tomography (CT) and magnetic resonance imaging (MRI).

All the procedures were standard clinical procedures equivalent to what were used on a daily basis in the

diagnostic evaluation of cervical spine trauma patients. The conventional X-ray examination consisted of

anterior-posterior, lateral, bilateral articular pillar and oblique views using an Arcosphere, Arcoma©

equipment. Spiral scanning of the total volume of the specimen was performed on a Phillips© four-slice

MX8000 CT scanner with a slice thickness of 1,3 mm, increments of 0.6 and a matrix of 512x512. Axial,

coronal, saggital and angled reconstructions were performed with bone and soft tissue algorithms.

Magnetic resonance imaging was performed on a General Electrics© 1.5 T Signa scanner using a surface

“temporomandibular” coil. Saggital T1, T2, STIR-sequences, axial T1, T2 and T2*gradient-sequences

were produced of the facet joint structures with the majority of sequences being made with a thickness of

3 mm, spacing 0 and FOV 16x16, and the matrix was 512x512 on the saggital sequences and 256x256

on the remaining.

Results from the first, second and third neuroradiological examinations (see below) were entered directly

onto response sheets, produced for the purpose of this study, independently by either observer. From

there direct entry into Excel, Microsoft© spreadsheets were possible.

At all times were the staff , including the participating radiologists, blinded with regard to the descriptive

characteristics of the subject examined and the lead investigator did not influence the interpretations or

conclusions made by the radiologists.

6.3.1. The first examination of the radiological images

In the first evaluation of radiological images all images from conventional X-rays, CT and MRI were

scored independently by two experienced neuroradiologists, with regard to general facet fractures and

41

haemarthrosis/excess fluid in the facet joints. No training or coaching was given prior to the first

evaluation and the scoring was based on routine clinical interpretation.

6.3.2. The second examination of the radiological images

A second evaluation was performed by both observers together during which consensus agreement was

obtained with regard to fractures on all imaging modalities, and haemarthrosis/excess fluid in a facet joint

on conventional X-rays and CT-scanning only.

6.3.3. The third examination of the radiological images

A third evaluation of the MR-images was performed by both observers independently and blinded with

regard to haemarthrosis/excess fluid in the facet joints only. Prior to the third evaluation the observers

defined consensus regarding scoring this variable. Hence, the variable was defined as a joint showing

focally increased signal, with signal differences between adjacent joints on any sequence or relatively

increased signals in all joints.

6.4. Histolaboratory procedures (see Article II & III for more details)

Immediately after the neuroradiological examinations the specimens were embedded in five litres 70 %

ethanol. After two weeks the specimens were divided by hemisection in the median plane and re-

embedded in 70 % ethanol for an additional three weeks, followed by one week in 96 % ethanol under

vacuum conditions fixation and a final week in 99 % ethanol. The hemisection was performed using an

industrial bandsaw and improved the efficiency of the alcohol fixation (Figure 6.1). After fixation each

specimen was embedded in liquid methylmethacrylate (MMA) and dibutylphtalate (a softener) in a

refrigerator for six weeks under continuous supervision, after which percadox (a catalyst) was added to

complete polymerisation and hardening into a plastic bloc 226-228. The hardened plastic bloc contained

one side of the un-decalcified specimen. At no time was the specimens frozen or decalcified.

An independent registrar who was blinded with regard to the descriptive characteristics of each subject

assigned each pair of plastic blocs with a two-digit reference number (k-reference subject number) as

well as a side label identifying the left and right side allowing identification of each unique subject. The

registrar was in charge of a codebook consisting of the reference number (LU-reference), originally

assigned by the lead investigator, prior to neuroradiological examination, and the corresponding k-

42

Figure 6.1 Illustration of a cervical spine specimen

The lower four cervical spine segments (corpora C4 to C7) viewed from the posterior (A). Illustration of

hemisectioning of the cervical spine sample approximately through the midline (median) with crude estimated

location of the individual spinal segments (B).

43

reference number. The codebook was disclosed to the participating physicians only at the time when all

evaluations had been completed. Hence, adequate blinding was obtained for all consequent evaluations

of the material by the lead investigator and the senior pathologists.

In addition to the k-reference and the LU-number a case number uniquely identified each specific facet.

The case number consisted of a 6-digit number made according to the order of appearance; two digits

from the k-reference number (legal values: 12-54), one digit from the side (legal values: 1 left, 2 right),

one digit from the slice (legal values: 1-9), one digit from the segment (legal values: 1 (Th1), 4-7 C4-C7))

and one digit from the facet orientation (legal values: 1 superior, 2 inferior). For example, case 422562

was equivalent to the subject with k-reference 42, right side, slice 5, segment C6 and the inferior facet.

6.4.1. Production of the 3-mm thick anatomical slices

From a random medial starting point, each plastic embedded hemisected cervical spine was sawed by

serial sectioning, into continuous approximately 3-mm thick parasaggital slices using a precision guided

bandsaw (Femi©, Bologna, Italy) (Figure 6.2). The guiding equipment ensured comparable thicknesses of

the slices with an additional tissue loss of approximately 1 mm per slice due to sawing. A total of 280 3-

mm thick slices were produced.

Each slice was given consecutive numbers with the most medial slice being numbered one and the most

lateral slice receiving the highest number, and a label containing the slice number, side and case

identification was fitted on the lateral aspect of the slice. The slices were cleaned and placed in a

photographic documentation bay where digital photographs were made in overview using a “μ [mju:] 400

digital”, Olympus© camera. These photographs were printed and used as reference material for the

subsequent stereomicroscopical examination of the 3-mm thick slices.

Detailed photographs of each facet joint were made using a Camedia C5050, Olympus© through a

stereomicroscope (SZX9, Olympus© with a 0.5X optical) with a direct capture function storing the images

directly in a database (DP-Soft, Olympus©). However, these photographs were not used in the present

study.

6.4.2. Examination of the 3-mm thick anatomical slices

Each 3-mm thick slice was examined with stereomicroscopy by one of two experienced pathologists and

scored with reference to injury of the facet cartilage and/or bone, bleeding in a joint, the presence of

44

Figure 6.2 Illustration of the sawing of 3-mm thick slices from a cervical spine specimen

The hemisected cervical spine viewed from the posterior with a black frame illustrating the line of parasaggital

sectioning for a 3-mm thick slice (A). The parasaggital 3-mm thick slice from the medial part of the facet joint is

viewed from the side visualising the articular column with the facet joints from C4-5, C5-6, C6-7 and C7-Th1 (B).

45

synovial folds and injury to the synovial folds. The findings were recorded on a reply sheet produced for

the purpose and entered into an Excel sheet, Microsoft©.

Furthermore, all 3-mm thick slices were documented with direct radiography using a faxitron medical

imaging device, Hewlett-Packard©. However, these images have not been evaluated in the present

study.

6.4.3. Production of the 10 μm thick histological sections

All the 3-mm thick slices containing facet joints were re-embedded in MMA. After polymerisation a

cleaning layer was removed from the surface using a heavy-duty microtome (SM 2500, Leica©) after

which two consecutive histological sections of 10 μm thickness were produced from each bloc. All the

histological sections were stained with Masson-Goldner trichrome, and mounted un-deplastified on glass

slides. The Masson-Goldner trichrome was chosen as it is particularly useful when staining alcohol

fixated plastic embedded undecalcified osteo-cartilaginous specimens 226-228. In selected cases additional

sections were produced and stained with Safranin O fast-green, haematoxylin-eosin and/or touluidine

blue.

6.4.4. Examination of the 10 μm thick histological sections

A total of 636 unique facets were available for examination as 39 subjects each had 16 unique facets

and the remaining one subject had only 12 due to a bloc vertebra with partial fusion of the facet joints

bilaterally of one motion segment. A total of 1830 unique observations (~46 observations per subject)

were made each containing information regarding morphological and histomorphometric variables (Table

6.1).

The morphological variables were examined with light microscopy using a BX51, Olympus© microscope

and relevant images were photographically documented through a light microscope with a digital camera

(Camedia C5050, Olympus© or D70s, Nikon©). The morphological variables included; cartilage flaking,

cartilage fissures (mild, moderate and severe), cartilage split, vascular invasion, osteophytes, presence

and integrity of the anterior and posterior fold, facet fractures, osteochondral fissures with or without

bleeding, bleeding in the joint space (haemarthrosis), bleeding in the folds (anterior, posterior or both

simultaneously) and bleeding in the underlying bone (Table 6.1).

The histomorphometric variables (Table 6.1) were obtained by measurements of projected images from

46

Variables Description Measurement/scoring

Morphological

flaking Flaking is defined as superficial fibrillation (discontinuity or cracks of the superficial matrix) anywhere on the surface

of the hyaline articular cartilage

no, yes, don’t know and missing

split Split is defined as one or more horizontal (0-60o to the articular cartilage surface) matrix separations of any size of the

hyaline cartilage anywhere within the calcified cartilage


fissure A fissure is defined as one or more vertical fissures (60-90o to the articular cartilage surface) anywhere within the

hyaline cartilage, and the most extensive fissure as defined as the highest percentage of the total non-calcified

hyaline cartilage thickness at the line of measurement is scored

no, more than 0% and less than 50%, 50% to less

than 100%, 100%, don’t know and missing

fracture A fracture is defined as a separation/discontinuity of osseous components (cortex, spongious and subchondral bone)

with the presence of erythrocytes in and around the lesion site


osteochondral fissure An osteochondral fissure is defined as a continuous line of separation of tissue components affecting both the

subchondral bone and the adjacent cartilage. This is furthermore evaluated for the presence of blood in the line of

separation

no, present without blood, present with blood, don’t

know and missing

vascular invasion Vascular invasion is defined as a blood vessel, or structure resembling a vascular structure in which erythrocytes are

present, that indents or penetrates the tidemark at the base of hyaline cartilage


osteophytes An osteophyte is defined as a prominent outgrowth of bony structures at the periphery of the joint margins in close

proximity to the anterior or posterior cartilage borders


blood in the joint space Blood in the joint space is defined as the presence of erythrocytes within the confines of the joint spaces. Is there

blood in the joint space not including the synovial fold area? Defined as the presence of significant erythrocytes in the

joint space (intra-articular)


blood in the synovial folds Blood in the synovial fold is defined as presence of extra vascular erythrocytes within any of the anterior or posterior

synovial folds

no, yes but anterior fold only, yes but posterior fold

only, yes and both folds, don’t know and missing

blood in the bone Blood in the bone is identified as significant localised extra vascular bleeding in either a fracture site or excessive

blood in the marrow space of the bone


presence of the anterior &

posterior fold The presence of tissue that exhibits the structure and location of an anterior or posterior synovial fold respectively no, yes and intact, yes and disrupted, don’t know and

missing

Histomorphometric

cartilage length The linear distance of the cartilage covering of each facet, measured on all 10μm thick sections. The distance is

measured in millimetres from the most anterior to most posterior aspect of the cartilage. The measure points of the

cartilage borders are determined by the observer as the estimated point where hyaline cartilage no longer covers the

articular surface

The length is measured on the projected image using

1x magnification, 1 mm equals 19.25 mm projected

on the table. Not measurable and missing values are

recorded

the anterior & posterior

fold overlap The anterior and posterior fold overlaps of the hyaline cartilage are measured. A line perpendicular to and through

the cartilage length line at the anterior and posterior endpoints form the basis. From this basis line the maximal length

of the respective folds extending towards the joint centre is recorded in millimetres. If the length of the fold is more

than the total measured length of the cartilage, then it is scored “100%” (if the overlap can confidently be ascribed to

the respective fold alone) or “complete” (if the overlap is a combined overlap of the anterior and posterior fold)

The overlap is measured on the projected image by

4x magnification, 1 mm equals 77.00 mm projected

on the table. Complete and 100% overlap, not

measurable and missing values are recorded

joint-tidemark The joint-tidemark is the distance measured from the joint space surface of the articular hyaline cartilage to the

tidemark separating the hyaline from the calcified cartilage. The tidemark is identified as the line dividing the green

and red staining on the Masson Goldner-Trichrome staining in the part of the deep cartilage. The measurement is

performed along a gridline that is perpendicular to the cartilage length line. The variable is recorded at maximally 5

intersections through each facet

The recorded value is measured in millimetres on the

projected image, which is at 4x magnification, 1 mm

equals 77.00 mm projected on the table. Not


tidemark-osteochondral The tidemark-osteochondral is the distance measured from the tidemark to the osteochondral junction. The

osteochondral junction is defined as the border between the light green and darker green stained areas on the

Goldner Trichrome staining. The measurement is performed along a gridline that is perpendicular to the cartilage

length line. The variable is recorded at maximally 5 intersections through each facet





osteochondral-bone The osteochondral-bone is the distance measured from the osteochondral junction to the bony edge. The bony edge

is defined as the first coming area devoid of bone with a minimum size of five millimetres in any length between bone

borders. The measurement is performed along a gridline that is perpendicular to the cartilage length line. The

variable is recorded at maximally 5 intersections through each facet





Table 6.1 Description of the morphological and histomorphometric variables

47

a BH2 Olympus© microscope with objectives x1 or x4, through a projection arm to a measuring table with

a final magnification factor on the table of 19.25 or 77.00 respectively (Figure 6.3).

These variables included information regarding; the hyaline cartilage length, the degree of overlap of the

anterior and posterior fold in relation to the underlying hyaline cartilage covering of the facet (Figure 6.4),

and the thicknesses of the hyaline cartilage, the calcified cartilage and the subchondral bone (Figure

6.5). The measurements were performed with a metallic ruler of the projected images and the measured

values were recorded. During statistical analysis these values were transformed by the known

magnification factor in order to correct for the effect of the magnification of the projected image (Table

6.1).

In order to randomise the recording sites for the thickness of the hyaline cartilage, calcified cartilage and

the subchondral bone the starting point for each observation was always anteriorly. The location was

determined by a random value produced in Excel, Microsoft© whereby the consequent recordings could

be made from the selected point moving from anterior to posterior with an absolute interval of 3.5 mm (on

the original object) between the lines of each measurement.

6.4.5. Handling of data from the 10 μm thick histological sections

The results from the examination of the 10μm thick histological sections were entered directly onto a

specially designed A-3 sheet containing all variables. After completed examination of all the histological

sections the data were entered twice into Epidata© vs. 3.02, a freeware software program for data entry

and documentation 229, by two independent persons in two separate files. Each person entered a total of

114.028 unique entries using the Epidata© data entry software. By having two separate files comparison

of the entries could be performed using an integrated validation function in Epidata©. Furthermore,

double entry of the case identifying variables (k-reference and case number) ensured entry of data

relevant to each unique facet. After the complete entry, comparison was performed on selected key fields

and missing observations were identified and corrected producing a 100 % response rate for all variables

in the dataset. The Epidata© function “validation of duplicate files” allowed identification and correction of

discrepancies between the two data files producing the final data set. This identified 34 field errors

(inconsistencies) in 20 observations of the morphological data (equivalent to error in 0.07% of the 51116

field registrations), and 153 field errors (inconsistencies) in 112 observations of the histomorphometric

data (equivalent to error in 0.24% of the 62912 field registrations). All field errors were corrected in both

48

Figure 6.3 Setup for histomorphometric measurements

The projection arm mounted on the BH2, Olympus© microscope through which the images are projected onto the

table. The magnification factor for the projected image on the table was measured for each objective used on the

revolver, i.e. objective x1 resulted in a magnification factor of 19.25 whereas objective 4x resulted in a

magnification factor of 77.00.

49

Figure 6.4 Histomorphometric quantitative measurements

Histomorphometric measurements of the cartilage length (a), and the anterior fold (b) and the posterior fold (c)

overlap of the underlying facet.

50

Figure 6.5 Random location of histomorphometric measurements

A “birdseye” view of the facet cartilage of a superior articular facet with the parallel lines resembling the

parasaggital 3-mm thick slices (A). A parasaggital histological section, equivalent to one of the 3-mm thick slices,

through the articular facet with the localisation of the measurement lines (1-5) through the cartilage and bone

determined at random prior to the first reading (B).

51

independent data files and following repeated comparison confirmation of complete agreement between

the files was obtained.

6.4.6. Evaluation of the inter- and intraobserver agreement of the morphological findings

For the purpose of interobserver agreement testing 10 μm thick sections from four randomly chosen

facets from each subject were used. Hence, a total of 160 facets were eligible for this evaluation,

however only 159 unique facets were evaluated as the remaining one facet did not contain articular

structures. All facets were examined independently by two examiners (the lead investigator and a senior

pathologist) who were blinded with regard to the descriptive characteristics of the subjects. Only the

morphological variables were tested in this manner. The results from each observer were entered into an

Excel, Microsoft© spreadsheet. Approximately six months after the first examination of the 160 randomly

chosen facets, the lead investigator conducted a repeated examination of the same facets with the

purpose of intraobserver agreement testing. Prior to this, two independent laboratory assistant had

assigned the facets with new running numbers whereby blinding and randomisation of the facets was

obtained. The results from the repeated examination were entered into an Excel, Microsoft© spreadsheet

and compared with the results from the first examination.

6.5. Police records and information from the medico-legal autopsy

Data from each medico-legal autopsy were obtained by a forensic pathologist who recorded pathological

conditions and concluded upon the diagnosis. Included in the information available to the forensic

pathologist were data from the law enforcement (police records). These observations were kept in the

classified archives of the Institute of Forensic Medicine. Access to the archive was available on-site only

and the discretion of personal data was maintained when obtaining relevant information, which was

copied and entered into an Excel, Microsoft© spreadsheet.

6.6. Statistical analysis

The association of groups was tested with Fisher’s exact test and Chi-squared test. Measurements of

agreement between the same and different observers, as well as between different diagnostic methods

were tested with kappa ( ) statistics 230. The data were analysed with a linear regression model

assuming linear association, normal distribution and independence (homoskedasticity) of residuals. The

52

regression model used for the morphological data examined for correlation between gender, age and

trauma versus no trauma and a reference person was defined as a 35 years old male who had been

killed in a road traffic crash. The regression model used for the histomorphometric data examined for

correlation between gender, age, trauma versus no trauma, side and facet per segment and a reference

person (level) was defined as the left C4 inferior facet of a 35 years old male who had been killed in a

road traffic crash. The potential effect of interaction was not tested due to the limited number of subjects

(n=40). Results including 95 % confidence intervals were presented in brackets, e.g. 75 μm [69-80]. In all

statistical tests the significance level was p < 0.05. All statistical analyses were performed using Stata© 9

(StataCorp LP, College Station, USA).

53

7. Summary of results

7.1. Diagnostic imaging findings (Paper I)

Initial examination of all neuroradiological images obtained of the lower cervical spine facet joints,

including conventional X-rays, CT and MRI, was performed by two senior radiologists in a blinded and

independent fashion with regard to articular facet fracture and haemarthrosis/excess fluid in a joint. This

revealed that CT was the most sensitive for the detection of facet fractures, followed by MRI and

conventional X-rays. Overall, agreement was “substantial” between the observers with kappa scores

ranging from 0.66–0.79 [0.37-1.09] (range [95 % confidence intervals]), p < 0.001. Based on consensus

agreement, facet fractures were present on neuroradiology in four trauma cases (4/19) only. Computed

tomography was the only modality capable of identifying all cases with facet fractures, where MRI

identified three and conventional X-rays one case with facet fractures. The agreement between the

observers initial (first) evaluation and the consensus (second) evaluation was substantial to almost

perfect for both observers with regard to facet fractures with kappa scores ranging from 0.66-1.00 [0.37-

1.31], p < 0.001 (unpublished data).

On neither conventional X-rays nor CT was haemarthrosis/excess fluid in a joint observed. There was

poor agreement between observers in a third evaluation on MRI only of haemarthrosis/excess fluid in a

joint in which 38 subjects (38/40) were scored positive by one or both observers ( = 0.31 [0.00-0.62], p <

0.05) with no statistically significant correlation to trauma.

Comparing the initial MR findings (first evaluation) to the third evaluation findings (MRI only) with regard

to haemarthrosis/excess fluid in a joint both observers obtained poor agreement, with one scoring ( =

0.04 [-0.05-0.13], p = 0.183) and the other scoring ( = 0.17 [0.00-0.34], p < 0.05) (unpublished data).

The findings on microscopy of haemarthrosis and bleeding in the synovial folds could not be confirmed

with any diagnostic imaging procedure (Figure 7.1).

In comparison to microscopical evaluation of the 10 μm thick sections only three subjects with facet

fractures were identified by both microscopy and CT (Figure 7.2) (Table 7.1) (Appendix 1). The fourth

facet fracture identified on CT and MRI was not identified on microscopy. A discrete fifth facet fracture at

54

Figure 7.1 Discrete injuries to the cervical spine facet joint soft tissue structures

Overview of an injured joint (A). The disrupted synovial fold with bleeding (a), and haemarthrosis in the joint

space (b) was only identified on microscopical examination of the 10 μm thick section (B), original magnification

x1.25 and x4. Stained with Masson-Goldner trichrome.

55

Figure 7.2 Cervical spine facet fracture in a motor vehicle crash fatality

The fracture through the superior facet of C6 was identified on both microscopical examination of the 10 μm thick

section, original magnification x1, Masson-Goldner trichrome (left) and computed tomography 3-D reconstructed

image (right).

56

Subject # 10 μm histology CT consensus 22 (case) C5 right IAP fracture -

25 (case)

C5 left IAP fracture C6 left SAP fracture C6 left IAP fracture C7 left SAP fracture C6 right SAP fracture

-C6 left SAP fracture ---

39 (case) Th1 left SAP fracture -C5 right IAP fracture

Th1 left SAP fracture C7 left IAP fracture -

53 (case) C7 left IAP fracture C6 right IAP fracture C7 right SAP fracture

-C6 right IAP fracture C7 right SAP fracture

24 (case) - Th1 left SAP fracture

The total number of unique facets is 636 among the 40 subjects SAP: superior articular process IAP: inferior articular process

Table 7.1 Subjects with unique facet fractures on either histology or CT consensus

57

the osteochondral junction was identified on microscopy only. There was “substantial” agreement

between the CT consensus findings and the microscopical findings with regard to identifying subjects

with facet fractures, and “moderate” agreement when identifying unique facet fractures. Computed

tomography had a sensitivity of 0.67 (95 % confidence interval: 0.29-1.04) and a specificity of 0.99 (95 %

confidence interval: 0.98-1.00) with regard to detecting facet fractures identified on the microscopy.

Key points Paper I CT was the most sensitive imaging modality with regard to detection of facet fractures

less than ½ of all unique facet fractures were identified on diagnostic imaging procedures

observers generally agreed on the presence of facet fractures on all imaging modalities

haemarthrosis could not be reliably identified on any diagnostic imaging procedure

58

7.2. Pathological findings (Paper II)

Stereomicroscopy of 3-mm thick slices and microscopical examination of 10 μm thick histological

sections from the lower cervical spine facet joints of motor vehicle crash victims and controls was

performed. Using stereomicroscopy damage to the facet cartilage and/or bone and haemarthrosis was

identified in a several subjects, however there was no significant correlation between exposure to trauma

and the findings, and the folds were poorly identifiable and no injuries were detected in the folds. In

contrast to the stereomicroscopy, light microscopy identified discrete injuries to the lower cervical spine

facet joints in large numbers among the victims of fatal motor vehicles crashes (Table 7.2). Light

microscopy identified four trauma cases with facet fractures (Figure 7.3) and three had osteochondral

fissures with bleeding that corresponded significantly with trauma, p < 0.05 and p < 0.01, respectively.

Soft tissue lesions were particularly common, with bleeding in any fold being the most frequent finding,

and simultaneous bleeding in both folds at any one segment was significantly correlated to trauma (p <

0.05) and to the presence of a fracture (p < 0.01). Haemarthrosis was identified in nine trauma cases

(Figure 7.4) and correlated significantly with trauma (p < 0.01), as well as with the presence of a fracture

(p < 0.05) and bleeding in both synovial folds at the same segment (p < 0.01). Disruption of the synovial

folds was common in the whole study population, but when the number of disrupted folds per subject

were calculated this showed significant correlation with trauma (p < 0.01). There was no statistically

significant correlation between the stereomicroscopy and microscopical findings concerning damages to

the facet cartilage and/or bone or the synovial folds.

Key points Paper II facet fractures, haemarthrosis and bleeding in folds were common after fatal traffic crashes

there was significant correlation between pathology and trauma

light microscopy was superior to stereomicroscopy in the evaluation of pathology

conventional autopsy procedures did not reveal any of the injuries to the facet joints

Subj

ect

Age

Gen

der

Frac

ture

of a

fa

cet

OC

F w

ith b

lood

(0

-16

face

t)O

CF

with

out

bloo

d

(0-1

6 fa

cet)

Blo

od in

any

jo

int s

pace

Blo

od in

any

fo

ld*

Blo

od in

bot

h fo

lds±

Ant

erio

r fol

d di

srup

ted

(0-8

join

ts)

Post

erio

r fol

d di

srup

ted

(0-8

join

ts)

Blo

od in

the

unde

rlyin

g bo

ne

2621

MN

o0

10N

oN

oN

o3

7Ye

s

2422

FN

o0

10N

oYe

sN

o0

1Ye

s

3533

MN

o0

16N

oN

oN

o1

4N

o

1420

MN

o0

5Ye

sYe

sN

o3

6Ye

s

2029

MN

o0

9N

oN

oN

o0

0N

o

2536

MYe

s4

7Ye

sYe

sYe

s8

3Ye

s

4229

MN

o0

6N

oN

oN

o0

1Ye

s

4723

FN

o0

10Ye

sYe

sYe

s3

5Ye

s

2131

FN

o0

6N

oYe

sno

76

Yes

1334

MN

o0

6N

oYe

sN

o2

8Ye

s

1745

FN

o0

8N

oN

oN

o0

2Ye

s

1935

MN

o0

9Ye

sN

oN

o1

4Ye

s

2220

MYe

s1

13Ye

sYe

sYe

s8

7Ye

s

3629

MN

o0

10Ye

sYe

sN

o5

7Ye

s

4637

MN

o0

10Ye

sYe

sN

o3

2N

o

5331

MYe

s0

13N

oN

oN

o3

7N

o

3941

MYe

s1

15Ye

sYe

sYe

s6

5Ye

s

3447

MN

o0

10Ye

sYe

sYe

s1

3Ye

s

2838

MN

o0

8N

oN

oN

o2

5Ye

s

19 c

ases

46

181

(59.

2%)

911

556

(37.

3%‡)

83 (5

5.3%

‡)15

(p<0

.05)

(p<0

.01)

(n.s

.)(p

<0.0

1)(n

.s.)

(p<0

.05)

(p<0

.01)

(p<0

.01)

(n.s

.)

‡ p

erce

ntag

e of

all

fold

s in

the

grou

ps (t

wo

fold

s ar

e no

n-ex

istin

g du

e to

a b

loc

verte

brae

in o

ne tr

aum

a ca

se)

OC

F: o

steo

chon

dral

fiss

ure

n.s.

: not

sta

tistic

ally

sig

nific

ant

* bl

eedi

ng in

the

ante

rior a

nd/o

r pos

terio

r fol

ds a

t any

seg

men

t per

sub

ject

± s

imul

tane

ous

blee

ding

in b

oth

fold

s at

any

seg

men

t per

sub

ject

Gen

der:

M (m

ale)

, F (f

emal

e)

Tabl

e 7.

2 H

isto

path

olog

ical

find

ings

in th

e ro

ad tr

affic

cra

sh fa

talit

ies

59

Figure 7.3 Injuries to the cervical spine facet joint in a motor vehicle crash fatality

Large overview of the anterior part of a facet joint, original magnification x1.25 (A). Close-ups of an osteo-

chondral fracture at the superior articular facet (B) and widespread bleeding in the anterior synovial fold and in

the joint space, original magnification x4 (C). Stained with Masson Goldner-Trichrome.

60

61

Figure 7.4 Haemarthrosis in a cervical spine facet joint in a fatal motor vehicle crash victim

Large overview of the joint with the inferior and superior facets, original magnification x1.25 (A). The close-up

illustrates bleeding in the joint space with an intact synovial fold, original magnification x4 (B). Stained with

Masson Goldner-Trichrome.

62

7.3. Anatomical and age-related findings (Paper III)

Microscopy of all histological sections containing articular facets (n=636) was performed with regard to

morphological and histomorphometric variables providing 1830 unique observations in total. Inter- and

intraobserver agreement was tested on four randomly selected facets from each subject with regard to

selected morphological variables. Significant age-, gender and trauma related changes in the bone,

cartilage, and soft tissues were observed. Flaking of the cartilage surface was the commonest finding in

more than half of the facets, followed by split and fissures (Figure 7.5). Osteophytes were only present in

approximately 4 % of the facets at the age of 35 years. Females were less affected by flaking, split and

fissures than males. The length of the cartilage in males was approximately 11.4 mm, whereas in

females it was significantly shorter (by approximately 1.3 mm) and the articular facets appeared to be

completely covered with cartilage without free bony edges (“cartilage gaps”) in far the majority of cases.

Ageing was significantly related to an increase in the number of facets with flaking, split and osteophytes.

The subchondral bone thickness increased significantly with increasing age equivalent to an increase of

approximately 10 % per decade. The presence of fissures did not correlate significantly with age. In the

trauma group there were significantly more disrupted anterior and posterior folds, and more articular

facets with cartilage split. However, fissures and flaking did not correlate to trauma. In the reference

person (a 35 years old male) the anterior and posterior fold overlap of the underlying cartilage was

approximately 16 % and did not differ between genders or with ageing. Furthermore, synovial folds were

consistently identified anteriorly and posteriorly in all but one facet joint, and in some histological section

the folds completely filled the joint space (Figure 7.6). Significant systematic differences were observed

between individual cervical spine segments from C4 to Th1, although none of these segmental variations

appeared to be linear. The thicknesses of the hyaline cartilage, the calcified cartilage, the total articular

cartilage and the calcified cartilage percentage of the total articular cartilage were not different between

genders and showed no correlation with ageing. There was overall moderate interobserver agreement

and good intraobserver agreement for the description of the morphological variables. The most important

morphological and histomorphometric findings are presented in Table 7.3 and Table 7.4 respectively.

63

Figure 7.5 The morphological variables

Vertical fissures are present in the deep and middle part of the cartilage, original magnification x4 (A).

Widespread superficial flaking/fibrillation of the cartilage, original magnification x10 (B). Large horizontal split

through the middle part of the hyaline cartilage, original magnification x4 (C). Example of vascular invasion of the

tidemark, original magnification x10 (D). All stained with Masson-Goldner trichrome.

64

Figure 7.6 Complete covering of the articular surface by the synovial folds

In this parasaggital section through the lateral part of a cervical spine facet joint there is complete covering of the

articular surfaces with a synovial fold, with the posterior fold (A) and anterior fold (B) illustrated in close-up,

original magnification x1.25 and x4, Masson-Goldner trichrome.

Estim

ate

95%

CI

Estim

ate

95%

CI

p-va

lue

Estim

ate

95%

CI

p-va

lue

Estim

ate

95%

CI

p-va

lue

Flak

ing

(%)

75[6

7-83

]-1

8[-2

9-(-7

)]<0

.01

13[7

-20]

<0.0

01-4

[-15-

(7)]

0.46

0S

plit

(%)

72[6

5-78

]-1

8[-2

7-(-9

)]<0

.001

8[3

-13]

<0.0

1-1

0[-1

8-(-1

)]<0

.05

Any

fiss

ure

(%)

58[4

9-67

]-1

3[-2

6-(-1

)]<0

.05

0[-7

-7]

0.98

5-2

[-14-

10]

0.73

1A

nter

ior f

old

disr

uptio

n (%

)16

[8-2

3]2

[-8-1

3]0.

655

-1[-7

-6]

0.87

9-1

1[-2

1-(-1

)]<0

.05

Pos

terio

r fol

d di

srup

tion

(%)

33[2

3-43

]-8

[-22-

5]0.

210

-10

[-18-

(-3)]

<0.0

5-1

5[-2

7-(-2

)]<0

.05

Ref

eren

ce p

erso

n: 3

5 ye

ars

old

mal

e ki

lled

in a

road

traf

fic c

rash

Varia

ble

Exam

ple:

a 3

5 ye

ars

old

fem

ale

is p

redi

cted

to h

ave

flaki

ng in

app

roxi

mat

ely

57%

, i.e

. 18%

few

er fa

cets

than

mal

es (P

<0.0

1). T

here

is a

n ag

e-re

late

d in

crea

se in

the

num

ber o

f fac

et w

ith fl

akin

g eq

uiva

lent

to

13%

per

dec

ade

(p<0

.001

) and

no

sign

ifica

nt c

orre

latio

n w

ith tr

aum

a (p

=0.4

60)

Incr

ease

per

10

year

s

AGE

Con

trol v

s. C

ase

CAS

E/C

ON

TRO

LG

END

ER

Fem

ales

vs.

Mal

esR

efer

ence

per

son

Tabl

e 7.

3 M

orph

olog

ical

find

ings

of t

he lo

wer

cer

vica

l spi

ne a

rtic

ular

face

t

65

66

Estim

ate

95%

CI

Estim

ate

95%

CI

p-va

lue

Estim

ate

95%

CI

p-va

lue

Estim

ate

95%

CI

p-va

lue

Hya

line

carti

lage

thic

knes

s (m

m)

0.72

[0.6

7-0.

77]

-0.0

5[-0

.11-

0.02

]0.

152

0.00

[-0.0

4-0.

04]

0.96

60.

04[-0

.02-

0.10

]0.

180

Cal

cifie

d ca

rtila

ge th

ickn

ess

(μm

)79

[68-

90]

5[-8

-17]

0.47

17

[-1-1

4]0.

080

0[-1

2-12

]0.

972

Sub

chon

dral

bon

e th

ickn

ess

(mm

)0.

35[0

.31-

0.39

]0.

03[-0

.02-

0.07

]0.

229

0.04

[0.0

1-0.

06]

<0.0

10.

00[-0

.04-

0.04

]0.

852

Thic

knes

s of

the

tota

l arti

cula

r car

tilag

e (m

m)

0.80

[0.7

5-0.

85]

-0.0

4[-0

.10-

0.02

]0.

207

0.01

[-0.0

3-0.

04]

0.70

80.

04[-0

.02-

0.10

]0.

184

Max

imum

car

tilag

e le

ngth

(mm

)11

.35

[10.

66-1

2.04

]-1

.33

[-2.1

9-(-0

.46)

]<0

.01

-0.2

0[-0

.70-

0.30

]0.

435

0.54

[-0.2

9-1.

37]

0.20

4C

alci

fied

carti

lage

of t

otal

arti

cula

r car

tilag

e th

ickn

ess

(%)

10[9

-12]

1[-1

-3]

0.17

71

[0-2

]0.

123

-1[-2

-1]

0.48

9A

nter

ior f

old

over

lap

(%)

16[1

1-21

]0

[-6-6

]0.

974

1[-2

-4]

0.62

25

[0-1

1]0.

055

Pos

terio

r fol

d ov

erla

p (%

)17

[13-

21]

2[-2

-7]

0.31

41

[-2-4

]0.

460

3[-2

-7]

0.24

5

Ref

eren

ce le

vel:

the

C4

infe

rior f

acet

of a

35

year

s ol

d m

ale

kille

d in

a ro

ad tr

affic

cra

sh

Varia

ble

Exam

ple:

a 3

5 ye

ars

old

fem

ale

has

a hy

alin

e ca

rtila

ge th

ickn

ess

equi

vale

nt to

that

of m

ales

(0.0

5 m

m th

inne

r, p=

0.15

2). T

here

are

no

age-

rela

ted

diffe

renc

es w

ith re

gard

to th

e th

ickn

ess

(p=0

.966

) and

ther

e is

no

sign

ifica

nt c

orre

latio

n w

ith e

xpos

ure

to

traum

a (p

=0.1

80).

Incr

ease

per

10

year

s

AGE

Con

trol v

s. C

ase

CAS

E/C

ON

TRO

LG

END

ER

Fem

ales

vs.

Mal

esR

efer

ence

leve

l

Tabl

e 7.

4 H

isto

mor

phom

etric

find

ings

of t

he lo

wer

cer

vica

l spi

ne a

rtic

ular

face

t

Key points Paper III males are more affected by degenerative changes in the cervical spine facets than females

degenerative changes in the articular cartilage are common from a young age

two synovial folds are consistently present in the cervical spine facet joints

pathological lesions are produced in facet cartilage and soft tissues after road traffic crashes

67

68

7.4. Other findings

7.4.1. Inter- and intraobserver agreement

All morphological variables were evaluated for the purpose of inter- and intraobserver agreement testing

using the 10 μm thick sections from four randomly chosen facets per subject. Based on the random

number generator in Excel, Microscoft© random slices, segments and orientations of two facets from

each side from each subject were selected. Some of the findings have been presented in Paper III,

however, for the purpose of clarity all observer scores are tabulated here (Table 7.5).

The agreement between the two observers was “moderate” in four variables (fissures, flaking, posterior

fold disruption and haemarthrosis) with kappa scores between 0.41 and 0.60 (p < 0.001). In four

variables (bleeding in folds, split, vascular invasion, and facet fracture) the agreement was “fair” with

kappa scores between 0.21 and 0.40 (p < 0.001). The agreement was “slight” with kappa scores

between 0 and 0.20 for the variables blood in the subchondral bone, osteochondral fissure (p < 0.05),

and anterior fold disruption (p = 0.156). There was “poor” agreement with a kappa score below 0 (p =

0.688) for the presence of osteophytes 230. As expected, the intraobserver agreement was better than the

interobserver agreement. Where the interobserver agreement generally was fair to moderate, the

intraobserver agreement was generally substantial.

7.4.2. Police records

For all subjects the police records were reviewed in detail with regard to information pertaining to the

circumstances of death and for all trauma cases data were retrieved concerning the mechanism of

trauma. In eighteen of the trauma cases (18/19 cases) the decedent was the driver of a passenger car.

The crash occurred mostly on dry roads, in darkness, with the involved vehicle driving straight ahead on

a main road without any road lightning. The collision was most often a frontal collision involving more

than one party. For practical purposes only the most important data are presented in Appendix 2.

Toxicological evaluation revealed a blood alcohol concentration (BAC) > 0.8 mg/ml in eight subjects

(5/19 cases and 3/20 controls tested), with four drivers (4/18 drivers) killed in a motor vehicle crash being

under the influence of alcohol (BAC > 0.8 mg/ml in three cases and alcohol concentration in the vitreous

humor of 2.4 mg/ml in one case). Medication was found in four subjects (3/9 cases tested and 1/10

controls tested) and narcotics were found in three subjects (1/9 cases tested and 2/10 controls tested).

The toxicological data were presented in part in Paper II and are presented in Appendix 3.

69

Simple agreement Kappa p-value

Simple agreement Kappa p-value

Variablefissure* 89.1% 0.65 0.53 ; 0.76 < 0.001 95.4% 0.84 0.72 ; 0.95 < 0.001

flaking 79.7% 0.54 0.37 ; 0.71 < 0.001 91.8% 0.83 0.67 ; 0.98 < 0.001

posterior fold disruption 83.2% 0.51 0.32 ; 0.70 < 0.001 92.1% 0.74 0.57 ; 0.90 < 0.001

haemarthrosis 87.3% 0.48 0.32 ; 0.64 < 0.001 98.7% 0.88 0.72 ; 1.04 < 0.001

bleeding in folds 81.6% 0.36 0.23 ; 0.49 < 0.001 97.3% 0.84 0.72 ; 0.96 < 0.001

split 72.9% 0.31 0.15 ; 0.47 < 0.001 89.9% 0.79 0.63 ; 0.94 < 0.001

vascular invasion 82.1% 0.29 0.13 ; 0.45 < 0.001 93.0% 0.70 0.55 ; 0.86 < 0.001

facet fracture 87.4% 0.23 0.12 ; 0.34 < 0.001 99.4% 0.85 0.70 ; 1.01 < 0.001

blood in subchondral bone 81.3% 0.15 0.00 ; 0.30 < 0.05 88.4% 0.59 0.44 ; 0.74 < 0.001

osteochondral fissure 56.0% 0.13 -0.01 ; 0.27 < 0.05 82.8% 0.61 0.46 ; 0.75 < 0.001

anterior fold disruption 84.6% 0.09 -0.08 ; 0.26 0.156 94.9% 0.75 0.59 ; 0.91 < 0.001

osteophytes 91.9% -0.04 -0.21 ; 0.13 0.688 96.8% 0.65 0.50 ; 0.80 < 0.001

*: weighted kappa

95% CI 95% CI

Interobserver agreement Intraobserver agreement

Table 7.5 Inter- and intraobserver agreement regarding microscopy of morphological variables

70

8. Discussion

8.1. Diagnostic imaging of the cervical spine facet joints after trauma

Using conventional X-rays, CT and MRI this study identified several lesions in the lower cervical spine

facet joints of road traffic crash fatalities in comparison to non-traumatised decedents, thereby

supplementing the results from previous advanced imaging studies 30;46;168;169;178;231. Fractures of the

articular facets were identified in four subjects (4/19 trauma cases) using diagnostic imaging techniques

which is a somewhat higher prevalence rate than reported in previous studies 30-32;36;46;169;178;231. As only

one of the facet fractures were identified on conventional X-rays the inadequacy of this imaging modality

towards visualising this type of lesion was underlined which is in agreement with previous findings

26;40;164-166;186;192-198. This finding strongly suggests that previous studies based solely on conventional X-

ray examination may have been significantly influenced by under-reporting of facet fractures. However,

although CT was the most sensitive imaging modality towards identifying facet fractures, similar to other

studies 164-168, only four of the 11 unique facet fractures identified on microscopy could be visualized

suggesting that, when present, facet fractures are more often affecting several facets rather than solitary

levels and that CT may miss discrete fractures 168;184. Hence, the overall sensitivity of advanced

diagnostic imaging procedures is inferior to microscopical examination with regard to the identification of

discrete facet fractures in the lower cervical spine.

This study revealed that post-mortem diagnostic imaging procedures, including MRI, do not reliably

identify injuries to the synovial folds or cases with haemarthrosis in comparison to the microscopical

findings, which has been reported previously 30-32;36;46;112;115;178;231. This was despite optimal imaging

conditions with no motion artefacts, long acquisition times and several different protocols including

sensitive STIR sequences. In contrast, a recent study identified the cervical spine facet joint folds on

MRI, although this required a high energy-field 3.0 Tesla MRI unit, with long acquisition time, limited field

of view and a small specimen size 117. In the current study a 1.5 Tesla MRI unit was used and the

protocols used may not have allowed adequate resolution introducing technical limitations that may have

71

affected the observations 117;168. Possibly, these findings suggest that increased signals in the facet joints

spaces in post-mortem subjects are unspecific, potentially irrelevant findings in relation to pre-mortem

status of these joints. However, as a large number of the trauma cases (almost half) had incurred

haemarthrosis and/or bleeding in the synovial folds (evident on microscopy) lesions were present in

many cases although they could not be verified reliably. In contrast to these findings, previous studies

have identified bleeding in the cervical spine facet joints on MRI in people killed in road traffic crashes,

using both independent observations and blinding to the pathological findings although the findings were

not statistically tested 178. In none of the conventional X-ray or CT investigations did the observers

identify soft tissue injuries in the facet joints.

The agreement between the observers with regard to detection of facet fractures was good for all

imaging modalities. The substantial agreements between the initial findings (the first evaluation) and the

consensus findings (the second evaluation) obtained by each observer supports the opinion that the

radiological findings of facet fractures are credible. In contrast, the observers obtained a poor agreement

with regard to the presence of bleeding/excess fluid in the joint on MRI. As this was thought to be a

problem of definition the observers agreed upon improved criteria prior to the third evaluation of

bleeding/excess fluid in the joint on MRI only. However, this did not improve the interobserver agreement

and comparing the results between the two evaluations poor intraobserver agreement was obtained,

which was probably explained by the newly introduced criteria. Actually, using the consensus criteria with

regard to scoring the MRI images for haemarthrosis/excess fluid in a joint, a significant increase in the

number of joints positive for this variable was observed with no correlation to trauma, suggesting either

improper definition of the variable or clinically irrelevant structural changes at post-mortem. Furthermore,

the integrity of the synovial folds could not be evaluated as these were not visualised specifically on any

of the imaging modalities. Overall, MRI of the soft tissues of the facet joints did not offer reliable

information and the role of this modality in the post-mortem evaluation of facet joint soft tissue injury is

questionable. This finding must be regarded in the context of the high frequency of soft tissue lesions in

the traumatised subjects (approximately half had soft tissue injuries). However, the role of MRI in the

evaluation of soft tissue injuries to the facet joints may have different relevance in clinical settings.

Survivors of road traffic crashes may sustain injury to the soft tissues of the facet joints that is likely to

cause hyperaemia, oedema, swelling, inflammation and pain developing over hours to days.

Theoretically, this should improve the likelihood of diagnosing injury to these joints as these processes

72

introduce physiological and potentially structural changes. Nonetheless, the findings from this study

suggest that soft tissue lesions in the facet joints, including haemarthrosis and bleeding in the synovial

folds, are difficult to diagnose on MRI.

No previous studies have published inter- and intraobserver agreement scores of the radiological

examination of post-mortem cervical spines nor the exact effect of training of observers with regard to

scoring of forensic radiological images, although recent comments in this regard have been proposed 168.

However, a recent clinical study evaluated the reliability of in vivo CT-scanning, in which a clinical

classification system for lower cervical spine injuries based on conventional radiology and CT-scanning

was proposed based on excellent intra- and interobserver agreement 232. Hence, although a comparison

with the findings from the current study is impossible future trials should evaluate the reliability of post-

mortem diagnostic imaging findings in order to ensure quality control.

The current generation of multislice CT-scanners (i.e. 64 slice scanners) typically produce image slice

thicknesses of 0.64 mm, which is a considerable improvement in comparison to the four-slice scanner

used in this study, and with the enhanced resolution the diagnostic sensitivity can be expected to

increase significantly. However, despite these advancements in CT technique, the specificity and

sensitivity of negative clinical and post-mortem CT examinations remains questionable as particularly

discrete injuries in the soft tissues of the facet joints were very common in the trauma cases without

being identified on the CT.

8.2. Pathology of the cervical spine facet joints after trauma

This study identified a total of 11 unique facet fractures in four of 19 subjects exposed to a fatal road

traffic crash, which is somewhat higher than previously reported 30-32;36;46;169;178;231. For example, in one

study examining road traffic crash fatalities, one lower cervical spine facet fracture was detected in one

of 22 crash victims 32. In another study, four unique lower cervical spine fractures were identified in one

subject out of 15 examined 36. The differences between the reported figures are likely due to sample

size, population characteristic (e.g. age and gender) anatomical region of interest and the mechanisms of

trauma, where the exposure to a road traffic crash causing life-threatening injuries and death indicates

the exchange of forces of a substantial magnitude. Hence, in this study all the trauma subjects were

young and had been occupants of a passenger car, which had crashed subjecting the decedent to fatal

bodily injuries.

73

The presence of haemarthrosis of the cervical spine facet joints have previously been described in post-

mortem studies of road traffic crash victims using histological and/or microtomal methods 30-32;36;46;178;231,

although the incidence rates have not previously been estimated. However, as haemarthrosis has not

generally been confirmed by histological evaluation in studies using cryomicrotomy these findings may

not be comparable. In this study, haemarthrosis was confirmed by the presence of erythrocytes in the

joint cavity which in some cases was difficult. Therefore, comparison with blood vessel content allowed

identification of erythrocytes as pale cell ghosts in a honeycomb pattern and by this definition

haemarthrosis was identified in almost half the trauma cases with statistical significance. This may

however have caused underestimation as tissue necrosis may have caused the erythrocytes to be

interpreted as decomposed synoviocytes and debris in the joint rather than blood. Nonetheless, the high

incidence of haemarthrosis in the trauma group was statistically significant in this study indicating that

these findings are common following fatal road traffic crashes with or without the presence of facet

fracture. The presence of bleeding in a joint has been shown to have negative effects on the cartilage, as

even brief exposures to blood may have detrimental long-lasting effects on the cartilage, by inhibiting the

proteoglycan synthesis, thereby predisposing to premature degeneration 233-236. Hence, acute synovitis

after haemarthrosis and/or cartilage damage may occur after trauma 233;236, which may also have

relevance for development of neck pain following adequate trauma.

The finding of bleeding in several folds is similar to previous reports 27;30-32;36;46;178. However, in some of

the control subjects bleeding was also encountered indicating that artefacts are being produced during

handling of the decedent or that this may be a natural physiological post-mortem finding. Despite this

finding, bleeding and disruption of the synovial folds correlated significantly to the exposure trauma. This

suggests that the folds are at risk of injury during a fatal road traffic crash. Although this may seem

irrelevant in a forensic perspective, this may have clinical implications as the folds are nociceptive

structures which may explain neck pain at least in the acute phase in casualties after survivable road

traffic crashes 36;37;107;113;114;119.

The presence of horizontal split of the articular cartilage was also closely related to trauma. As the

cartilage does not receive sensory nerve supply these lesions do not directly cause pain. However, it can

be conjectured that such lesions predispose to degeneration of the articular facets as the internal

architecture is disorganised with consequently altered biomechanics of the cartilage 122;124.

74

Theoretically, the advanced degenerative changes observed on diagnostic imaging of chronic neck pain

patients following survivable road traffic crashes 211-213, may be explained by the potential consequences

of any of the findings including; cartilage split, facet fractures, haemarthrosis as well as disruption of

synovial folds. These conditions may alter the biomechanics of the articular cartilage and the joint

whereby degradation of cartilage matrix may predispose to OA years after the injury 33;114;212;213;237.

Stereomicroscopic evaluation of plastic embedded 3-mm thick slices were found unreliable with regard to

the identification of osseous and/or soft tissue injury to the lower cervical spine facet joints, and findings

from previous reports could not be confirmed. Hence, light microscopy was found superior to

stereomicroscopy in the evaluation of pathology of the lower cervical spine facet joints.

The conventional medicolegal autopsy, which utilised an anterior approach 225, did not identify any of the

injuries in the cervical spine facet joints despite histological and neuroradiological evidence of injuries.

However, the poor sensitivity towards identifying injuries in the facet joints is not surprising as

conventional medicolegal autopsy does not allow detailed description of the posterior elements of the

cervical spine. This is in accordance with other studies reaching the same conclusions regarding

standardised autopsies 30;31;36;46;178;238. Hence, post-mortem investigations of cervical spine injury

necessitate detailed autopsy procedures similar to those utilised in this study.

8.3. Anatomy of the lower cervical spine facet joints

In this study, morphological and histomorphometric evaluation of the lower cervical spine facet joints

revealed a number of age and gender related differences in anatomical properties.

The mean articular cartilage thickness of the cervical spine facet joint at C4 inferior process was

approximately 0.8 mm, independent of gender, but depending on the spinal level. In a related study the

mean overall thickness of the cervical spine facet cartilage was found to be less (females 0.4 mm and

males 0.5 mm) 48. However, these findings were based on a cryomicrotomal study of six elderly

embalmed cadavers which makes precise determination of the cartilage thickness difficult as the

measurements are performed on photographic illustrations. The differences observed in the current

study between the spinal segments probably are consequences of developmental adaptations to

different functional requirements depending on the spinal segment. It was interesting to observe that the

calcified cartilage occupied approximately 10 % of the complete cartilage thickness which is somewhat

higher than previous studies of other joints reporting values of approximately 5 % (3-8 %) 127-129. The

75

finding of anterior and posterior synovial folds in all but one joint, irrespective of age and gender,

supports a number of previous publications 47;107;116;117;135. In contrast to these findings other studies have

not been able to identify synovial folds consistently in the facet joints 112;185. Fletcher et al 185, examined

20 cadavers (two aged 10 and 19 years, and the remaining 18 subjects between the age of 37-86 years)

with photographic documentation of cryomicrotomal sections and haematoxylin-eosin staining of

decalcified sections and concluded that the menisci [folds] were nonexistent at the age of 37 years and

older. In the study by Yu et al 112, 10 decedents (mean age 48 years, range 10-69 years) were examined

with photographic evaluation of cryomicrotomal sections and four types of menisci [folds] were proposed

despite the low number of subjects and the authors concluded that in the lower cervical spine of adults,

no menisci [folds] were present within the joint and the articular surfaces. Hence, both these studies

relied on photographic documentation, which probably in part explains their results. Similarly, all the folds

had a mean overlap of approximately 16 % of the underlying cartilage. However, in several parasaggital

sections of the facets the synovial fold covered the articular surface completely which indicates that the

folds have a heterogenous covering of the articular surface. The consistent presence of the synovial

folds suggest that they may have a unique physiological function, similar to the protective and lubricative

function of menisci, although this has not been elucidated 33;107;239.

Hyaline cartilage was found to cover the joint surfaces in the majority of the joints. Occasionally, a

transitional zone was observed at the borders of the cartilage covering of the facets in which a gradual

replacement of the hyaline cartilage by fibrous cartilage took place. Hence, only rarely were free bony

edges encountered with periosteal covering, previously described as “cartilage gaps” 48. The

discrepancies between these findings may be explained by the different populations examined, where

the previous study only included six elderly subjects in contrast to this study’s 40 subjects in the age

range 20-49 years, and the orientation of sectioning (although both methods used parasaggital

orientation). It is believed, that detailed morphology is best visualised by microscopy rather than by

evaluation of photographic reprints of anatomical structures. It can be speculated that the previously

reported findings are due to technical limitations, i.e. poor resolution, which does not allow identification

of all articular cartilage and fibrous cartilage. Hence, the current study does not offer support for the

presence of cartilage gaps and consequently no support of the pathomechanical theory of bony impact

during trauma as a cause of persistent neck pain.

76

A number of morphological variables were significantly correlated with age, including; cartilage flaking,

cartilage split, and osteophytes, and are all related to OA 49;122;131. The commonest finding was flaking of

the cartilage, followed by cartilage split and fissures, all present in more than half the articular facets,

indicating that these changes occur early in life. Age was also significantly related to an increase in the

subchondral bone thickness, although no age-related differences were observed concerning the

thickness of the hyaline and calcified cartilage. The subchondral bone thickness increased approximately

10 % per decade which is in agreement with previous observations 122;127;129. Other indicators of OA,

such as osteophytes, duplication of the tidemark and vascular invasion of the tidemark, were only

present in limited numbers which is explained by the median age (35 years) of the entire study

population. However, despite the young population examined, cartilage flaking, split and fissures were

very common in all ages in agreement with other studies 49;130. Although, fissures were common they did

not reach significant correlation with age despite published correlations with OA and increasing age

122;131.

The overall finding of significant correlation between gender and histological findings seen in OA (e.g.

cartilage flaking and splitting) is in agreement with other reports of males having a preponderance for

pathological facet joints, i.e. males were more severely affected by these changes than females 49;122.

Similarly, the cartilage length was significantly shorter in females which has also been reported

previously 54;56. Nonetheless, several variables were not affected by gender; e.g. overlap of the synovial

folds, and the thickness of the articular cartilage and the subchondral bone.

8.4. Methodological considerations

The effective study population consisted of 40 subjects, providing a total of 636 unique lower cervical

spine articular facets, which considering the fact these were post-mortem human subjects is a quite high

number. In order to improve the likelihood of obtaining significance of the statistical analyses we used

strict inclusion criteria, which produced a homogenous group of subjects. Although this increased the

statistical strength this also had a detrimental effect as a number of biological age-related variables (e.g.

osteophytes) were much less likely to be observed in this population of a median age of 35 years (range

20-49). Furthermore, the population size did not allow evaluation of the potential effects of interaction in

relation to age, as the statistical significance was not present. The difference from time of death to time

of autopsy, the influence of outside temperature and surroundings was not taken into consideration in

77

this study. Although advanced decomposition was an exclusion criterion, two subjects had to be

excluded after inclusion due to advanced tissue decomposition, which was identified on microscopical

examination and combined with the findings during autopsy. It can be difficult to evaluate the exact

extent of putrefaction in a body, since the early changes in particular may vary according to anatomical

region. Nonetheless, it is a prerequisite for comparability that the subjects are in a comparable state

exhibiting only limited post-mortem changes.

In this study the most common cause of death in the trauma group was multiple injuries with a high

incidence rate of skull fractures and organic injuries. Forensic toxicology revealed that four of the 18

drivers killed were under the influence of alcohol during the time of the crash which is similar to

previously reported figures of 20-30 % of decedents from fatal road traffic crashes having BAC values

above 0.5 mg/ml 240-242. Similarly, we found one trauma case who tested positive for drugs which also

supports previous reports of drugs in road traffic crash fatalities 241. Hence, the subjects included in this

study were comparable with road traffic crash fatalities examined in other studies 60;138.

This study was performed on tissue samples that had been removed en bloc from the decedent, rather

than examined in situ. Removal of tissue samples from decedents is known to enhance the rate of

putrefaction in the tissues. Although this may have influenced the quality of the diagnostic images and

the following histological samples the detrimental effect hereof is thought to be minimal since the

samples were examined within 24 hours, most often within two to four hours after removal. Furthermore,

cooling of the samples was used until diagnostic imaging after which fixation was initiated immediately.

When preparing bone samples for microscopical examination of un-decalcified stained sections, paraffin

embedding cannot be utilised since this does not supply adequate support of the osseous tissue during

microtomy. Therefore, alcohol (ethanol) fixation followed by embedding in liquid methylmethacrylate was

chosen as the most suitable method. Nonetheless, we found that fixation was extremely time-consuming

and that the quality of the consequent histological sections probably was affected by the quality of

fixation of these large specimens.

Embedding large tissue samples in MMA demands skill and experience as the process takes place over

several weeks. In some cases the polymerisation (hardening) of the MMA had not been completed

evenly throughout the bloc, which made consequent sawing difficult with the risk of inducing artefact in

the tissues. Alternative methods have been used extensively in the last couple of decades which involves

cryosectioning (cryomicrotomy) of frozen samples rather than fixation 28;32;48;184;185;243-245. However, for the

78

purpose of the present study this method would require a large cryomicrotome, which was only present

in Denmark in one centre (privately owned) at the start of the project. It is however interesting whether

future studies would be able to utilize such apparatus, as that would significantly reduce the time factor.

Furthermore, cryosectioning may perform automated continued sectioning throughout a specimen of a

set thickness, e.g. 50 μm, with the production of a few microns thick section per interval, which may be

stained prior to microscopy.

We found that staining the plastic mounted histological sections with Masson-Goldner trichrome or

haematoxylin-eosin did not visualise the erythrocytes, as they would normally appear on a conventional

section, e.g. paraffin embedded pulmonary tissue sample. It was speculated that the remaining plastic in

the histological sections caused incomplete staining of the erythrocytes, however staining of sections

without plastic did not influence the staining characteristics. Possibly, the prolonged ethanol fixation and

embedding period may have resulted in osmotic gradients and dissolution of the erythrocyte cells

whereby consequent staining of un-deplastified histological sections may have caused the poor

visualisation of erythrocytes in the lesions sites thereby underestimating bleeding.

From other histological studies it is known that tissue may shrink during handling 246. As shrinkage most

often takes place non-homogenously within tissue samples this is likely to have affected the results in

this study. However, no attempts to quantify the effects of shrinkage were made.

We evaluated the histomorphometrical properties of the articular cartilage, the calcified cartilage and the

underlying subchondral bone in typically 9-12 locations of each cervical spine facet between C4 and Th1.

However, although the histological sections were produces from random lines through the facets the

orientation was not completely randomised, i.a. all lines were parasaggital and parallel. Other studies

have used the same direction of sectioning (i.e. parasaggital) 27;31;32;48;135;178;185, and only few studies

have used supplemental planes of sectioning 36. Although we randomised the locations of measurements

the effect of orientation is unclear. Consequently, systematic false readings could be obtained due to the

line of sectioning chosen as this may skew the true anatomical sizes and appearances. Preferably, a

stereological unbiased evaluation using random sampling and orientation should be used in order to

examine a larger population with a wider age-range for the extent of age-related changes.

Notwithstanding these facts, we believe that our results are acceptable estimates of the

histomorphometry of the lower cervical spine facet joints, and they provide a systematic presentation of

the cartilage, subchondral bone and soft tissue properties.

79

The potential consequences of a three-dimensional structure (i.e. the facet joint) being sectioned in one

plane for the purpose of two-dimensional evaluation likewise affect the morphological interpretation.

The morphological variables were evaluated dichotomously, with the exception of few variables where

more possibilities were present. For example, it is conceivable that the synovial folds identified in the

anterior and posterior parts of the joints are continuous circular structures surrounding the entire surface

of the cartilage, rather than only located anteriorly and posteriorly. This would not be detectable using the

current methods. Furthermore, as a consequence of the thickness of sawing (i.e. 3-mm thick slices) facet

fractures parallel to the line of sectioning would not necessarily appear on the histological sections as

discrete fractures could be present inside the 3-mm slices but outside the area of consequent microtomal

sectioning.

The purpose of this study was not to design a scoring system for the histological grading of OA in the

facet joints. However, using a pragmatic approach several morphological variables known to have

relevance to OA were selected and described. The method used in this study did not allow evaluation of

the true severity of the morphological variables as only the worst case finding was recorded for each

observation. A recently proposed classification system, by the OARSI Working Group, integrates the

grade (extent) and stage of degenerative changes in general cartilage with limited considerations to the

subchondral bone 131. Grading and staging of age-related and OA changes in the cervical spine facet

joints may be performed with the method of Pritzker et al 131. However, this system is based upon

examination of decalcified paraffin embedded sections rather than un-decalcified plastic embedded

sections which makes the method inappropriate for the evaluation of large un-decalcified osteo-

cartilaginous specimens. Ideally, however, severity measures should be integrated in the method by

considering the extent of tissue affected 131. Hence, an adjusted classification system based on the

principles of the OARSI Working Group methods targeting the cervical spine facet joints specifically

seems to be the most ideal solution 131.

An alternative scoring system for cervical spine facet joint OA has been published 135. However, the

weighted contributions from each variable were accumulated whereby the likelihood of observing

differences in pathological severity between individuals was reduced. This probably explains why

insignificant differences were identified between age groups and severity of the changes.

In the evaluation of inter-observer agreement between a senior pathologist and the lead investigator

generally moderate agreement was achieved for the majority of the morphological variables. The

80

interobserver agreement was probably influenced by improperly defined variables as well as different

levels of experience between the observers in combination with the low prevalence rate. For example,

the osteophytes identified were discrete changes rather than the typical appearance of large bony spurs,

hence differentiation of positive versus negative scoring was difficult, even within the same observer.

Similarly, osteochondral fissures were believed to have little, if any, biological relevance as they most

often were unstained empty spaces indicating artefacts rather expression of true anatomical morphology.

Hence, the observers may have scored the osteochondral fissures differently, depending on whether

they were regarded as true fissures rather than artefacts. Excessive bleeding in the subchondral bone

was extremely difficult to confirm, as the presence of blood was a frequent normal finding. We can offer

no sensible explanation why the integrity of the anterior folds could not be established reliably between

the observers as the evaluation of the posterior counterpart resulted in a higher and significant

agreement score. The overall intraobserver agreement was good, and for all variables better than the

interobserver agreement, which could be expected. Despite the overall acceptable inter- and

intraobserver agreement scores, future studies should attempt to improve the definitions of the variables

in consensus. Furthermore, the analysis underlines the importance of training and consensus of

definitions of the anatomical structures under investigation.

Initially attempts were made to compare the results from the investigation of the cervical spine facet

joints and pathological evaluations with data from the police records with regard to the mechanism of

trauma. However, in far the majority of the road traffic crash cases the mechanisms of trauma were

complex and the extent of bodily injury was significant. Furthermore, only limited information concerning

the details of the fatal crashes were available which did not allow optimal correlation of the findings from

our study to the police records. Hence, future studies investigating the specific causative mechanisms of

trauma in relation to unique pathological findings need to address these difficulties early in the planning

phase, as detailed onsite investigations of the crash is necessary for adequate information pertaining to

the crash scene and circumstances.

8.5. External validity and clinical implications

It is clear from this study that the cervical spine facet joints have a unique morphology which is probably

closely related to the highly specialised physiological (biomechanical) demands. With consideration to

the methodological limitations in this study, the findings are thought to be in good agreement with

81

previously published studies and therefore reliable measures of the anatomical properties of the facet

joints.

Regarding the pathological findings uncritical extrapolation from post-mortem findings to survivors

sustaining similar injuries cannot be justified. However, it is interesting to consider whether the

pathological findings in this study are likely to be present in survivors of RTC’s, and whether such

discrete injuries have clinical implications.

A number of studies have found similar lesions in survivors from road traffic crashes who later died due

to un-related causes 26-29;139;169, and clinical studies using diagnostic imaging procedures have identified

osseous injuries in survivors from road traffic crashes 41. Biomechanical studies have pointed to

pathological force loading on facet joint structures as a consequence of the exposure to relatively minor

acceleration forces 69-71;78;79. Furthermore, in clinical practice, posterior neck pain is an extremely

common symptom following RTC’s for which specific organic causes seldom can be documented

21;25;156;157. Hence, these findings converge on the theory that injuries are likely to be incurred based on

violent acceleration-deceleration forces to the cervical spine complex and that these injuries may well

arise in traumas that expose the occupants to forces short of lethal forces 21;30;95. However, the frequency

of facet joint fractures and soft tissue lesions in humans subjected to cervical spine acceleration-

deceleration forces in survivable RTC’s is unknown 21;30;34.

Although this study has not studied the classical whiplash trauma (LOSRIC) per se, it has investigated

injuries that may be related to motor vehicle crashes of a wide range of severities. It is noteworthy that

not all decedents due to a road traffic crash suffered significant external evidence of neck, head and/or

brain injury. As a matter of fact, in several cases the primary cause of death was not related to lesions in

the head or neck. Hence, it is conceivable that the discrete non-fatal injuries of the lower cervical spine

facet joints detected in this study of high-energy crash fatalities would also have been present had the

decedents been exposed to sub-lethal forces to other vital organs in which the primary cause of death

was identified.

Causal injury thresholds for facet fractures and soft tissue injuries of the lower cervical spine facet joints

are not available as these are dependent upon many factors 14;95;96;247. The majority of whiplash trauma’s

involve the exchange of relatively low energy (force) 14;59;63-65, and presumably the average resultant

forces acting on the occupants subjected to a LOSRIC are much less than what is expected to have

been inflicted on the fatalities in this study. This is probably also the case for the loading of the facet

82

joints in the lower cervical spine. However, as the forces in common rear-impact collisions involve

pathophysiological compressive, distractive and shear forces to the facet joints 37;77-80;82 the differences

in magnitude of loading between the different types of crashes cannot be evaluated based on this study.

Hence, as the individual thresholds of injury for any particular subject under any given crash

circumstances cannot be reliably estimated 95, the findings in this study may well have relevance and

clinical implications for subgroups of people injured in whiplash trauma.

83

9. Conclusions and perspectives

This study provided a detailed description of the histomorphology of the lower cervical spine facet joints

of a young population with regard to the soft tissues, cartilage and subchondral bone. Degenerative

changes in the facet joints were found to occur early in life in both the cartilage and the subchondral

bone. Two synovial folds were identified in all but one facet joints which indicate a yet unproven

biological function. The calcified cartilage was constantly equal to about 10 % of the total articular

cartilage thickness and the articular facets were generally covered in their entirety by primarily hyaline

cartilage.

The original hypothesis that there is an under-representation of pathoanatomical lesions in the cervical

spine facet joints in people killed in road traffic crashes examined by standardised autopsy and

conventional radiological evaluation was confirmed. Non-fatal injuries to the lower cervical spine facet

joints were common in road traffic crash fatalities, including facet fractures, haemarthrosis, disruption and

bleeding in the synovial folds. The lesions were best appreciated on light microscopy in comparison to

stereomicroscopic evaluation, and conventional autopsy procedures did not reveal any of the injuries.

Discrete osseous lesions in the lower cervical spine facet joints were detected in people killed in

passenger car crashes by means of advanced diagnostic imaging procedures, including conventional X-

rays, magnetic resonance imaging, and in particular computed tomography. However, approximately half

of the unique facet fractures were not detected on any imaging modality despite optimal technical

circumstances, confirming that not all discrete facet fractures are identified on diagnostic imaging

procedures. Furthermore, bleeding in the cervical facet joints and synovial folds could not be determined

reliably on any diagnostic imaging modality, including MRI, despite histological evidence hereof. This

study emphasizes the need for scientific evidence of validity and reliability of advanced diagnostic

imaging procedures in forensic settings, in particular with regard to non-fatal soft tissue lesions.

The data concerning the circumstances of the traffic crashes were insufficient in order to conduct

detailed comparison with the pathological findings, as the majority of decedents killed in a road traffic

crash suffered multiple injuries due to complex kinematics and force loading.

84

The utilisation of increasingly advanced imaging modalities as well as specialised autopsy techniques

should be encouraged in forensic settings in order to increase the sensitivity towards identifying discrete

injuries. The potential clinical implications of the lesions identified in this study are unknown. However,

although uncritical extrapolation of the results to survivors from road traffic crashes cannot be justified, it

is plausible that discrete injuries similar to those identified in this study will be present in subgroups of

casualties from survivable road traffic crashes.

85

References

1. European Communities. Keep Europe Moving - sustainable mobility for our continent: Mid-term

review of the European Commision's 2001 Transport White Paper. Accessed at

http://ec.europa.eu/transport on June 16, 2007. Luxembourg: Office for Official Publications of

the European Communities, 2006.

2. European Communities. White Paper - European transport policy for 2010: time to decide.

Accessed online at http://ec.europa.eu/transport/white_paper on June 11, 2007. Luxembourg:

Office for Official Publications of the European Communities, 2001.

3. European Road Safety Observatory (2007). Traffic safety basic facts 2006. Accessed at

http://www.erso.eu/safetynet on May 23. 2007.

4. Danmarks Statistik. Road traffic accidents (Færdselsuheld) 2005. Danmarks Statistiks trykkeri,

Copenhagen, 2006.

5. Quinlan KP, Annest JL, Myers B et al. Neck strains and sprains among motor vehicle occupants-

United States, 2000. Accid.Anal.Prev. 2004;36:21-7.

6. Lovell ME, Galasko CS. Whiplash disorders--a review. Injury 2002;33:97-101.

7. Galasko CS, Murray PM, Pitcher M et al. Neck sprains after road traffic accidents: a modern

epidemic. Injury 1993;24:155-7.

8. Richter M, Otte D, Pohlemann T et al. Whiplash-type neck distortion in restrained car drivers:

frequency, causes and long-term results. Eur Spine J 2000;9(2):109-17.

9. Versteegen GJ, Kingma J, Meijler WJ et al. Neck sprain in patients injured in car accidents: a

retrospective study covering the period 1970-1994. Eur.Spine J 1998;7:195-200.

10. Versteegen GJ, Kingma J, Meijler WJ et al. Neck sprain after motor vehicle accidents in drivers

and passengers. Eur.Spine J 2000;9:547-52.

11. Sundhedsstyrelsen. Redegørelse om whiplash. Sundhedsstyrelsen, København, 2000.

86

12. Whiplashkommisionen. Whiplashkommisionens slutrapport. Accessed at

http://www.whiplashkommissionen.se on October 5, 2006. Sandviken tryckeri, Sandviken,

Whiplashkommisionen, Stockholm, Sweden, 2005.

13. Kongsted A, Qerama E, Kasch H et al. Neck collar, "act-as-usual" or active mobilization for

whiplash injury? A randomized parallel-group trial. Spine 2007;32:618-26.

14. Foreman SM, Croft AC. Whiplash injuries - the cervical acceleration/deceleration syndrome, 3rd.

ed.Lippincott Williams & Wilkins; Baltimore, MD, USA, 2002.

15. Rebbeck T, Sindhusake D, Cameron ID et al. A prospective cohort study of health outcomes

following whiplash associated disorders in an Australian population. Inj.Prev. 2006;12:93-8.

16. Heikkila H, Heikkila E, Eisemann M. Predictive factors for the outcome of a multidisciplinary pain

rehabilitation programme on sick-leave and life satisfaction in patients with whiplash trauma and

other myofascial pain: a follow-up study. Clin.Rehabil. 1998;12:487-96.

17. Kasch H, Bach FW, Jensen TS. Handicap after acute whiplash injury: a 1-year prospective study

of risk factors. Neurology 2001;56:1637-43.

18. Freeman MD, Croft AC, Rossignol AM et al. A review and methodologic critique of the literature

refuting whiplash syndrome. Spine 1999;24:86-96.

19. Yoganandan N, Pintar FA, Kleinberger M. Whiplash injury. Biomechanical experimentation.

Spine 1999;24:83-5.

20. European Commision. Whiplash prevention: a major car safety issue. Accessed online at

http://ec.europa.eu/research/growth/gcc/projects/in-action-whiplash.html#03], June 6, 2000.

21. Barnsley L, Lord S, Bogduk N. Whiplash injury. Pain 1994;58:283-307.

22. Barnsley L. An evidence-based approach to the treatment of acute whiplash injury. Pain

Res.Manag. 2003;8:33-6.

23. Lord SM. Treatment strategies for chronic cases. Pain Res.Manag. 2003;8:37-9.

24. Bogduk N, Teasell R. Whiplash: the evidence for an organic etiology. Arch.Neurol. 2000;57:590-

1.

87

25. Barnsley L, Lord SM, Wallis BJ et al. The prevalence of chronic cervical zygapophysial joint pain

after whiplash. Spine 1995;20:20-5.

26. Abel MS. Occult traumatic lesions of the cervical vertebrae. CRC Crit Rev.Clin.Radiol.Nucl.Med.

1975;6:469-553.

27. Schonstrom N, Twomey L, Taylor J. The lateral atlanto-axial joints and their synovial folds: an in

vitro study of soft tissue injuries and fractures. J.Trauma 1993;35:886-92.

28. Bring G, Jonsson H, Jr., Hamberg J et al. Pathoanatomical Changes of the Brachial Plexus and

of C5-C6 Following Whiplash-Type Injury: A Case Report. Journal of Whiplash & Related

Disorders 2005;1:53-64.

29. Taylor JR, Kakulas BA. Neck injuries. Lancet 1991;338:1343.

30. Uhrenholt L, Grunnet-Nilsson N, Hartvigsen J. Cervical spine lesions after road traffic accidents:

a systematic review. Spine. 2002;27:1934-41.

31. Taylor JR, Taylor MM. Cervical spinal injuries: an autopsy study of 109 blunt injuries. J

Musculoskeletal Pain 1996;4:61-79.

32. Jonsson H, Jr., Bring G, Rauschning W et al. Hidden cervical spine injuries in traffic accident

victims with skull fractures. J.Spinal Disord. 1991;4:251-63.

33. Taylor J, Twomey L. Whiplash injury and neck sprain: a review of their prevalence, mechanisms,

risk factors, and pathology. Crit Rev.Phys.Rehab Med. 2005;17:285-99.

34. Bogduk N. Point of view. Spine 2002;27:1940-1.

35. Bogduk N. The anatomy and pathophysiology of whiplash. Clin Biomech 1986;1:92-101.

36. Taylor JR, Twomey LT. Acute injuries to cervical joints. An autopsy study of neck sprain. Spine

1993;18:1115-22.

37. Stemper BD, Yoganandan N, Pintar FA. Gender- and region-dependent local facet joint

kinematics in rear impact: implications in whiplash injury. Spine 2004;29:1764-71.

38. Gargan MF. What is the evidence for an organic lesion in whiplash injury? J Psychosom Res

1995;Aug;39(6):777-81.

88

39. Smith GR, Beckly DE, Abel MS. Articular mass fracture: a neglected cause of post-traumatic

neck pain? Clin.Radiol. 1976;27:335-40.

40. Woodring JH, Goldstein SJ. Fractures of the articular processes of the cervical spine. AJR Am.J

Roentgenol. 1982;139:341-4.

41. Kotani Y, Abumi K, Ito M et al. Cervical spine injuries associated with lateral mass and facet joint

fractures: new classification and surgical treatment with pedicle screw fixation. Eur.Spine J.

2005;14:69-77.

42. Davis SJ, Teresi LM, Bradley WG, Jr. et al. Cervical spine hyperextension injuries: MR findings.

Radiology 1991;180:245-51.

43. Jonsson H, Jr., Cesarini K, Sahlstedt B et al. Findings and outcome in whiplash-type neck

distortions. Spine 1994;19:2733-43.

44. Karlsborg M, Smed A, Jespersen H et al. A prospective study of 39 patients with whiplash injury.

Acta Neurol.Scand. 1997;95:65-72.

45. Pettersson K, Hildingsson C, Toolanen G et al. MRI and neurology in acute whiplash trauma. No

correlation in prospective examination of 39 cases. Acta Orthop.Scand. 1994;65:525-8.

46. Uhrenholt L, Nielsen E, Vesterby A et al. Detailed examination of the lower cervical spine facet

joints in a road traffic crash fatality - a case study. IRCOBI (annual proceedings) 2005;paper

701;411-414.

47. Bland JH, Boushey DR. Anatomy and physiology of the cervical spine. Semin.Arthritis Rheum.

1990;20:1-20.

48. Yoganandan N, Knowles SA, Maiman DJ et al. Anatomic study of the morphology of human

cervical facet joint. Spine 2003;28:2317-23.

49. Sager P. Spondylosis cervicalis - a pathological and osteoarchaeological study. Thesis,

University of Copenhagen, Denmark, 1969.

50. Hirsch C, Schajowicz R, Galante J. Structural changes in the cervical spine. A study on autopsy

specimens in different age groups. Acta Orthop.Scand. Suppl 109, 7-77. 1967.

51. Ebraheim NA, Xu R, Challgren E et al. Quantitative anatomy of the cervical facet and the

posterior projection of its inferior facet. J.Spinal Disord. 1997;10:308-16.

89

52. Ebraheim NA, Patil V, Liu J et al. Morphometric analyses of the cervical superior facets and

implications for facet dislocation. Int.Orthop. 2006.

53. Panjabi MM, Duranceau J, Goel V et al. Cervical human vertebrae. Quantitative three-

dimensional anatomy of the middle and lower regions. Spine 1991;16:861-9.

54. Francis CC. Variations in the articular facets of the cervical vertebrae. Anat.Rec. 1955;122:589-

602.

55. Milne N. The role of zygapophysial joint orientation and uncinate processes in controlling motion

in the cervical spine. J.Anat. 1991;178:189-201.

56. Panjabi MM, Oxland T, Takata K et al. Articular facets of the human spine. Quantitative three-

dimensional anatomy. Spine 1993;18:1298-310.

57. Evans L. Traffic Safety. Bloomfiels Hills, Michigan, USA: Science Serving Society, 2004.

58. National Highway Traffic Safety Administration. Traffic safety facts 2005: A compilation of motor

vehicle crash data from the fatality analysis reporting system and the general estimates system.

Accessed at http://www-nrd.nhtsa.dot.gov, on June 4, 2007. 2007. DOT HS 810 936.

Washington, DC: Department of Transportation.

59. Nordhoff LS, Jr. Motor Vehicle Collision Injuries - Biomechanics, Diagnosis, and Management.

2nd. ed. London: Jones and Bartlett Publishers, Inc., 2005.

60. Saukko P, Knight B. Knight's Forensic Pathology. 3rd. ed. London, UK: Arnold, a member of the

Hodder Headline Group, 2004.

61. Dix J, Graham M, Hanzlick R. Investigation of road traffic fatalities - an atlas. Boca Raton,

FLorida, USA: CRC Press LLC, 2000.

62. Eichberger A, Geigl BC, Moser A. Comparison of different car seats regarding head-neck

kinematics of volunteers during rear end impact. 153-164. 1996. Dublin, Ireland; IRCOBI. In:

International IRCOBI Conference on the Biomechanics of Impact.

63. Hell W, Langwieder K, Walz F. Reported soft-tissue neck injuries after rear-end collisions. 261-

274. 1998. Göteborg, Sweden; IRCOBI. In: International IRCOBI Conference on the

Biomechanics of Impact.

90

64. Otte D, Pohlemann T, Blauth M. Significance of soft tissue neck injuries AIS1 in the accident

scene and deformation characteristics of cars with delta-V up to 10 km/h. 265-283. 1997.

Hannover, Germany; IRCOBI. In: International IRCOBI Conference on the Biomechanics of

Impact.

65. Krafft M, Kullgren A, Malm S et al. Influence of crash severity on various whiplash injury

symptoms: A study based on real-life rear-end crashes with recorded crash pulses. Paper

number: 05-0363. 2005. 19th. International Technical Conference on the Enhanced Safety of

Vehicles Conference (ESV) 2005.

66. Spitzer WO, Skovron ML, Salmi LR et al. Scientific monograph of the Quebec Task Force on

Whiplash-Associated Disorders: redefining "whiplash" and its management. Spine 1995;20:1S-

73S.

67. Macnab I. Acceleration injuries of the cervical spine. J Bone Joint Surg.Am. 1964;46:1797-9.

68. Huelke DF, Mendelsohn RA, States JD et al. Cervical fractures and fracture-dislocations

sustained without head impact. J Trauma 1978;18:533-8.

69. Grauer JN, Panjabi MM, Cholewicki J et al. Whiplash produces an S-shaped curvature of the

neck with hyperextension at lower levels. Spine 1997;22:2489-94.

70. Kaneoka K, Ono K, Inami S et al. Motion analysis of cervical vertebrae during whiplash loading.

Spine 1999;24:763-9.

71. Ono K, Kaneoka K, Wittek A et al. Cervical injury mechanism based on the analysis of human

cervical vertebral motion and head-neck-torso kinematics during low-speed rear impacts. SAE

paper 973340, 339-356. 1997. Society of Automotive Engineers, Inc., Warrendale,

Pennsylvania, USA. 41st Stapp Car Crash Conference.

72. Siegmund GP, Brault JR, Wheeler JB. The relationship between clinical and kinematic

responses from human subject testing in rear-end automobile collisions. Accid.Anal.Prev.

2000;32:207-17.

73. Siegmund GP, Myers BS, Davis MB et al. Mechanical evidence of cervical facet capsule injury

during whiplash: a cadaveric study using combined shear, compression, and extension loading.

Spine 2001;26:2095-101.

74. Stemper BD, Yoganandan N, Pintar FA. Kinetics of the head-neck complex in low-speed rear

impact. Biomed.Sci.Instrum. 2003;39:245-50.

91

75. Stemper BD, Kumaresan S, Yoganandan N et al. Head-neck finite element model for motor

vehicle inertial impact: material sensitivity analysis. Biomed.Sci.Instrum. 2000;36:331-5.

76. Cusick JF, Pintar FA, Yoganandan N. Whiplash syndrome: kinematic factors influencing pain

patterns. Spine 2001;26:1252-8.

77. Panjabi MM, Maak TG, Ivancic PC et al. Dynamic intervertebral foramen narrowing during

simulated rear impact. Spine 2006;31:E128-E134.

78. Ito S, Ivancic PC, Panjabi MM et al. Soft tissue injury threshold during simulated whiplash: a

biomechanical investigation. Spine 2004;29:979-87.

79. Pearson AM, Ivancic PC, Ito S et al. Facet joint kinematics and injury mechanisms during

simulated whiplash. Spine 2004;29:390-7.

80. Ivancic PC, Panjabi MM, Ito S. Cervical spine loads and intervertebral motions during whiplash.

Traffic.Inj.Prev. 2006;7:389-99.

81. Winkelstein BA, Nightingale RW, Richardson WJ et al. The cervical facet capsule and its role in

whiplash injury: a biomechanical investigation. Spine 2000;25:1238-46.

82. Stemper BD, Yoganandan N, Rao RD et al. Reflex muscle contraction in the unaware occupant

in whiplash injury. Spine 2005;30:2794-8.

83. Szabo TJ, Welcher JB. Human subject kinematics and electromyographic activity during low

speed rear impacts. SAE paper 962432, 295-315. 1996. Society of Automotive Engineers, Inc.,

Warrendale, Pennsylvania, USA. 40th Stapp Car Crash Conference.

84. Mills H, Horne G. Whiplash--manmade disease? N.Z.Med.J 1986;99:373-4.

85. Larsen LB, Holm R. [Prolonged neck pain following automobile accidents. Gender and age

related risk calculated on basis of data from an emergency department]. Ugeskr.Laeger

2000;162:178-81.

86. Dolinis. Risk factors for 'whiplash' in drivers: a cohort study of rear-end traffic crashes. Injury

1997;Apr;28(3):173-9.

87. Chapline JF, Ferguson SA, Lillis RP et al. Neck pain and head restraint position relative to the

driver's head in rear-end collisions. Accid.Anal.Prev. 2000;32:287-97.

92

88. Krafft M. Non-fatal injuries to car occupants - injury assessment and analysis of impacts causing

short- and long-term consequences with special reference to neck injuries. 1998. Karolinska

Institute, Stockholm, Sweden.

89. Krafft M, Kullgren A, Tingvall C et al. How crash severity in rear impacts influences short- and

long-term consequences to the neck. Accid.Anal.Prev. 2000;32:187-95.

90. Ono K, Kaneoka K. Influences of the physical parameters on the risk to neck injuries in low

impact speed rear-end collisions. 201-212. 1993. International IRCOBI Conference on the

Biomechanics of Impact, Eindhoven, Netherlands.

91. Murphy DR, Freeman M. Management of neck pain and related disorders . In: Haldeman S, ed.

Principles and practice of chiropractic. 3rd. ed. USA: McGraw-Hill Companies, Inc., 2005:969-98.

92. Gotten N. Survey of one hundred cases of whiplash injury after settlement of litigation. J

Am.Med.Assoc. 1956;162:865-7.

93. Hildingsson C, Toolanen G. Outcome after soft-tissue injury of the cervical spine. A prospective

study of 93 car-accident victims. Acta Orthop.Scand. 1990;61:357-9.

94. Deans GT, Magalliard JN, Kerr M et al. Neck sprain--a major cause of disability following car

accidents. Injury 1987;18:10-2.

95. Uhrenholt L, Gregersen M. Trafikulykker ved lav hastighed (del 1) – grænseværdier for whiplash

associated disorders (WAD). (In Press) 2007.

96. Croft AC, Freeman MD. Correlating crash severity with injury risk, injury severity, and long-term

symptoms in low velocity motor vehicle collisions. Med.Sci.Monit. 2005;11:RA316-RA321.

97. Moss RT, Bardas AM, Hughes MC et al. Injury symptom risk curves for occupants involved in

rear end low speed motor vehicle collisions. SAE Technical Paper Series 2005;01-0296.

98. Whiplashkommisionen och Svenska Läkaresëllskapet. "Diagnostik och tidigt omhändertagande

av whiplashskador". Accessed at http://www.whiplashkommissionen.se on October 5, 2006.

Sandvikens Tryckeri, Sandviken, Whiplashkommisionen, Stockholm, Sweden, 2005.

99. Kasch H, Bach FW, Stengaard-Pedersen K et al. Development in pain and neurologic

complaints after whiplash: A 1-year prospective study. Neurology 2003;60:743-9.

93

100. Squires B, Gargan MF, Bannister GC. Soft-tissue injuries of the cervical spine. 15-year follow-up.

J Bone Joint Surg Br 1996;Nov;78(6):955-7.

101. Sterner Y, Toolanen G, Gerdle B et al. The incidence of whiplash trauma and the effects of

different factors on recovery. J.Spinal Disord.Tech. 2003;16:195-9.

102. Bylund PO, Bjornstig U. Sick leave and disability pension among passenger car occupants

injured in urban traffic. Spine 1998;23:1023-8.

103. Ettlin TM, Kischka U, Reichmann S et al. Cerebral symptoms after whiplash injury of the neck: a

prospective clinical and neuropsychological study of whiplash injury. J

Neurol.Neurosurg.Psychiatry 1992;55:943-8.

104. Pettersson K, Brandstrom S, Toolanen G et al. Temperament and character: prognostic factors

in whiplash patients? Eur.Spine J 2004;13:408-14.

105. Radanov BP, Di SG, Schnidrig A et al. Psychosocial stress, cognitive performance and disability

after common whiplash. J Psychosom.Res. 1993;37:1-10.

106. Payne EE, Spillane JD. The cervical spine; an anatomico-pathological study of 70 specimens

(using a special technique) with particular reference to the problem of cervical spondylosis. Brain

1957;80:571-96.

107. Taylor J, Twomey L, Levander B. Contrasts between cervical and lumbar motion segments. Crit

Rev.Phys.Rehab Med. 2000;12:345-71.

108. White AA, III, Panjabi MM. The basic kinematics of the human spine. A review of past and

current knowledge. Spine 1978;3:12-20.

109. Panjabi MM, White AA. Physical properties and functional biomechanics of the spine. In: White

AA, Panjabi MM, eds. Clincial Biomechanics of the Spine. Philadelphia: JB Lippincott, 1978.

110. Schafer RC. Clinical Biomechanics - Musculoskeletal Actions and Reactions. 2nd. ed. Baltimore:

Williams & Wilkins, 1987.

111. Tonetti J, Peoc'h M, Merloz P et al. Elastic reinforcement and thickness of the joint capsules of

the lower cervical spine. Surg.Radiol.Anat. 1999;21:35-9.

112. Yu SW, Sether L, Haughton VM. Facet joint menisci of the cervical spine: correlative MR imaging

and cryomicrotomy study. Radiology 1987;164:79-82.

94

113. Inami S, Kaneoka K, Hayashi K et al. Types of synovial fold in the cervical facet joint.

J.Orthop.Sci. 2000;5:475-80.

114. Mercer S, Bogduk N. Intra-articular inclusions of the cervical synovial joints. Br.J.Rheumatol.

1993;32:705-10.

115. Fletcher G, Haughton VM, Ho KC et al. Age-related changes in the cervical facet joints: studies

with cryomicrotomy, MR, and CT. AJR Am.J.Roentgenol. 1990;154:817-20.

116. Kos J, Hert J, Sevcik P. [Meniscoids of the intervertebral joints]. Acta Chir

Orthop.Traumatol.Cech. 2002;69:149-57.

117. Friedrich KM, Trattnig S, Millington SA et al. High-field magnetic resonance imaging of

meniscoids in the zygapophyseal joints of the human cervical spine. Spine 2007;32:244-8.

118. Ebraheim NA, Xu R, Stanescu S et al. Anatomic relationship of the cervical nerves to the lateral

masses. Am.J.Orthop. 1999;28:39-42.

119. Inami S, Shiga T, Tsujino A et al. Immunohistochemical demonstration of nerve fibers in the

synovial fold of the human cervical facet joint. J.Orthop.Res. 2001;19:593-6.

120. McLain RF. Mechanoreceptor endings in human cervical facet joints. Spine 1994;19:495-501.

121. Bogduk N. Cervical zygapophysial joint pain and percutaneous neurotomy - an update to the

Quebec Task Force on Whiplash-Associateed Disorders. In: Gunzburg R, Szpalski M, eds.

Whiplash Injuries: Current Concepts in Prevention, Diagnosis, and Treatment of the Cervical

Whiplash Syndrome. Philadelphia: Lippincott-Raven Publishers, 1998:211-9.

122. Brandt KD, Doherty M, Lohmander LS et al. Osteoarthritis. 2nd. ed. Oxford, New York: Oxford

University Press Inc., 2003.

123. Ostergaard K, Salter DM. Immunohistochemistry in the study of normal and osteoarthritic

articular cartilage. Prog.Histochem.Cytochem. 1998;33:93-165.

124. Mow VC, Hung CT. Biomechanics of Articular Cartilage. In: Nordin M, Frankel VH, eds. Basic

biomechanics of the musculoskeletal system. 3rd. ed. Philadelphia: Lippincott Williams &

Wilkins, 2001:60-101.

125. Pal GP, Routal RV. The role of the vertebral laminae in the stability of the cervical spine. J.Anat.

1996;188 ( Pt 2):485-9.

95

126. Müller-Gerbl M, Schulte E, Putz R. The thickness of the calcified layer of articular cartilage: a

function of the load supported? J.Anat. 1987;154:103-11.

127. Oegema TR, Jr., Carpenter RJ, Hofmeister F et al. The interaction of the zone of calcified

cartilage and subchondral bone in osteoarthritis. Microsc.Res.Tech. 1997;37:324-32.

128. Oettmeier R, Abendroth K, Oettmeier S. Analyses of the tidemark on human femoral heads. II.

Tidemark changes in osteoarthrosis--a histological and histomorphometric study in non-

decalcified preparations. Acta Morphol.Hung. 1989;37:169-80.

129. Müller-Gerbl M. The Subchondral Bone Plate.Springer-Verlag, Berlin, Germany, 1998.

130. Gries NC, Berlemann U, Moore RJ et al. Early histologic changes in lower lumbar discs and

facet joints and their correlation. Eur.Spine J 2000;9:23-9.

131. Pritzker KP, Gay S, Jimenez SA et al. Osteoarthritis cartilage histopathology: grading and

staging. Osteoarthritis.Cartilage. 2006;14:13-29.

132. Lane LB, Bullough PG. Age-related changes in the thickness of the calcified zone and the

number of tidemarks in adult human articular cartilage. J.Bone Joint Surg.Br. 1980;62:372-5.

133. Oegema TR, Jr., Bradford DS. The inter-relationship of facet joint osteoarthritis and degenerative

disc disease. Br.J Rheumatol. 1991;30 Suppl 1:16-20.

134. Kettler A, Werner K, Wilke HJ. Morphological changes of cervical facet joints in elderly

individuals. Eur.Spine J 2007;16:987-92.

135. Berlemann U, Laubli R, Moore RJ. Degeneration of the atlanto-axial joints: a histological study of

9 cases. Acta Orthop.Scand. 2002;73:130-3.

136. Collins DH. Osteoarthritis. The pathology of articular and spinal diseases. 1949:74-115.

137. Laeubli RT, Moore RJ, Berlemann U. Histologic degeneration of the aging cervical spine.

(International Cartilage Repair Society annual conference), International Cartilage Repair Society

annual conference 1999, poster # 1. 1999.

138. Löwenhielm P, Bayer Kristensen I. Trafik. In: Lange Thomsen J. eds. Retsmedicin - Nordisk

Lærebog. 1st. ed. 2004.

96

139. Kakulas BA, Taylor JR. Pathology of injuries of the vertebral column and spinal. In: Vinken PJ,

Bruyn GW, Klawans HL et al., eds. Handbook of Clinical Neurology: Spinal Cord Trauma.

Amsterdam: Elsevier Science Publishers, 1992:21-51.

140. Cain CM, Ryan GA, Fraser R et al. Cervical spine injuries in road traffic crashes in South

Australia, 1981-86. Aust.N.Z.J Surg. 1989;59:15-9.

141. Twomey LT, Taylor JR, Taylor MM. Unsuspected damage to lumbar zygapophyseal (facet) joints

after motor-vehicle accidents. Med.J.Aust. 1989;151:210-7.

142. Kongsted A. Whiplash injuries - the effect or early interventions and the value of smooth pursuit

eye movement testing. Thesis, University of Southern Denmark, 2005.

143. Sturzenegger M, DiStefano G, Radanov BP et al. Presenting symptoms and signs after whiplash

injury: the influence of accident mechanisms. Neurology 1994;44:688-93.

144. Bunketorp L, Nordholm L, Carlsson J. A descriptive analysis of disorders in patients 17 years

following motor vehicle accidents. Eur.Spine J. 2002;11:227-34.

145. Bunketorp L, Stener-Victorin E, Carlsson J. Neck pain and disability following motor vehicle

accidents--a cohort study. Eur.Spine J 2005;14:84-9.

146. Croft AC. Treatment paradigm for cervical acceleration/deceleration injuries (whiplash). ACA J

Chiro 1993;30:41-5.

147. Steigerwald DP, Croft AC eds. Whiplash and temporomaniducbalr joint dysfuntion: An

interdisciplinary approach to case management. Encinitas, CA: Keiser Publishing Co., 1992.

148. Banic B, Petersen-Felix S, Andersen OK et al. Evidence for spinal cord hypersensitivity in

chronic pain after whiplash injury and in fibromyalgia. Pain 2004;107:7-15.

149. Curatolo M, rendt-Nielsen L, Petersen-Felix S. Evidence, mechanisms, and clinical implications

of central hypersensitivity in chronic pain after whiplash injury. Clin.J Pain 2004;20:469-76.

150. Curatolo M, rendt-Nielsen L, Petersen-Felix S. Central hypersensitivity in chronic pain:

mechanisms and clinical implications. Phys.Med.Rehabil.Clin.N.Am. 2006;17:287-302.

151. Koelbaek Johansen M, Graven-Nielsen T, Schou Olesen A et al. Generalised muscular

hyperalgesia in chronic whiplash syndrome. Pain 1999;Nov 1;83(2):229-34.

97

152. Nederhand MJ, Hermens HJ, Ijzerman MJ et al. The effect of fear of movement on muscle

activation in posttraumatic neck pain disability. Clin.J Pain 2006;22:519-25.

153. McLean SA, Clauw DJ, Abelson JL et al. The development of persistent pain and psychological

morbidity after motor vehicle collision: integrating the potential role of stress response systems

into a biopsychosocial model. Psychosom.Med. 2005;67:783-90.

154. Adams H, Ellis T, Stanish WD et al. Psychosocial factors related to return to work following

rehabilitation of whiplash injuries. J Occup.Rehabil. 2007;17:305-15.

155. Williamson E, Williams M, Gates S et al. A systematic literature review of psychological factors

and the development of late whiplash syndrome. Pain 2007.

156. Sapir DA, Gorup JM. Radiofrequency medial branch neurotomy in litigant and nonlitigant patients

with cervical whiplash: a prospective study. Spine 2001;26:E268-E273.

157. Barnsley L. Percutaneous radiofrequency neurotomy for chronic neck pain: outcomes in a series

of consecutive patients. Pain Med. 2005;6:282-6.

158. McDonald GJ, Lord SM, Bogduk N. Long-term follow-up of patients treated with cervical

radiofrequency neurotomy for chronic neck pain. Neurosurgery 1999;Jul;45(1):61-7; discussion

67-8.

159. Hooper RA, Frizzell JB, Faris P. Case series on chronic whiplash related neck pain treated with

intraarticular zygapophysial joint regeneration injection therapy. Pain Physician 2007;10:313-8.

160. Mikeladze G, Espinal R, Finnegan R et al. Pulsed radiofrequency application in treatment of

chronic zygapophyseal joint pain. Spine J 2003;3:360-2.

161. Keats TE, Dalinka MK, Alazraki N et al. Cervical spine trauma. American College of Radiology.

ACR Appropriateness Criteria. Radiology 2000;215 Suppl:243-6.

162. el-Khoury GY, Kathol MH, Daniel WW. Imaging of acute injuries of the cervical spine: value of

plain radiography, CT, and MR imaging. AJR Am.J Roentgenol. 1995;164:43-50.

163. Kaiser JA, Holland BA. Imaging of the cervical spine. Spine 1998;23:2701-12.

164. Holmes JF, Akkinepalli R. Computed tomography versus plain radiography to screen for cervical

spine injury: a meta-analysis. J.Trauma 2005;58:902-5.

98

165. Griffen MM, Frykberg ER, Kerwin AJ et al. Radiographic clearance of blunt cervical spine injury:

plain radiograph or computed tomography scan? J.Trauma 2003;55:222-6.

166. Gale SC, Gracias VH, Reilly PM et al. The inefficiency of plain radiography to evaluate the

cervical spine after blunt trauma. J.Trauma 2005;59:1121-5.

167. Holmes JF, Mirvis SE, Panacek EA et al. Variability in computed tomography and magnetic

resonance imaging in patients with cervical spine injuries. J.Trauma 2002;53:524-9.

168. Yen K, Lovblad KO, Scheurer E et al. Post-mortem forensic neuroimaging: Correlation of MSCT

and MRI findings with autopsy results. Forensic Sci.Int. 2007;Epub ahead of print.

169. Rauschning W, McAfee PC, Jonsson H, Jr. Pathoanatomical and surgical findings in cervical

spinal injuries. J.Spinal Disord. 1989;2:213-22.

170. Daffner RH, Daffner SD. Vertebral injuries: detection and implications. Eur.J.Radiol.

2002;42:100-16.

171. U.S.Food and Drug Administration. Medical X-rays. Accessed at http://www.fda.gov, on August

25, 2007.

172. U.S.Food and Drug Administration. What are the radiation risks from CT? Accessed at

http://www.fda.gov, on August 25, 2007.

173. U.S.Food and Drug Administration. FDA Public Health Notification: MRI-caused injuries in

patients with implanted neurological stimulators. Accessed at http://www.fda.gov, on August 25,

2007.

174. Buck U, Naether S, Braun M et al. Application of 3D documentation and geometric

reconstruction methods in traffic accident analysis: With high resolution surface scanning,

radiological MSCT/MRI scanning and real data based animation. Forensic Sci.Int. 2006;Epub

ahead of print.

175. Poulsen K, Simonsen J. Computed tomography as routine in connection with medico-legal

autopsies. Forensic Sci.Int. 2006;Epub ahead of print.

176. Dirnhofer R, Jackowski C, Vock P et al. Virtopsy: minimally invasive, imaging-guided virtual

autopsy. Radiographics 2006;26:1305-33.

177. Leth P. The use of CT scanning in forensic Autopsy. Forensic Sci.Med.Pathol. 2007;3:65-9.

99

178. Stäbler A, Eck J, Penning R et al. Cervical spine: postmortem assessment of accident injuries--

comparison of radiographic, MR imaging, anatomic, and pathologic findings. Radiology

2001;221:340-6.

179. Sweeney JF, Rosemurgy AS, Gill S et al. Is the cervical spine clear? Undetected cervical

fractures diagnosed only at autopsy. Ann.Emerg.Med. 1992;21:1288-90.

180. Abel MS, Teague JH. Unilateral lateral mass compression fractures of the axis. Skeletal Radiol.

1979;4:92-8.

181. Leditschke J, Anderson RM, Hare WS. The cervical spine in fatal motor vehicle accidents. Clin

Exp Neurol 1992;29:263-71.

182. Pech P, Kilgore DP, Pojunas KW et al. Cervical spinal fractures: CT detection. Radiology

1985;157:117-20.

183. Rauschning W, Bergstrom K, Pech P. Correlative craniospinal anatomy studies by computed

tomography and cryomicrotomy. J.Comput.Assist.Tomogr. 1983;7:9-13.

184. Yoganandan N, Cusick JF, Pintar FA et al. Whiplash injury determination with conventional spine

imaging and cryomicrotomy. Spine 2001;26:2443-8.

185. Fletcher G, Haughton VM, Ho KC et al. Age-related changes in the cervical facet joints: studies

with cryomicrotomy, MR, and CT. AJNR Am.J.Neuroradiol. 1990;11:27-30.

186. Woodring JH, Lee C. The role and limitations of computed tomographic scanning in the

evaluation of cervical trauma. J Trauma 1992;33:698-708.

187. Klein GR, Vaccaro AR, Albert TJ et al. Efficacy of magnetic resonance imaging in the evaluation

of posterior cervical spine fractures. Spine 1999;24:771-4.

188. Yenerich DO, Haughton VM. Oblique plane MR imaging of the cervical spine. J

Comput.Assist.Tomogr. 1986;10:823-6.

189. Abel MS. Moderately severe shiplash injuries of the cervical spine and their roentgenologic

diagnosis. Clin Orthop. 1958;12:189-208.

190. Benzel EC, Hart BL, Ball PA et al. Magnetic resonance imaging for the evaluation of patients

with occult cervical spine injury. J Neurosurg. 1996;85:824-9.

100

191. Bohlman HH. Acute fractures and dislocations of the cervical spine. An analysis of three hundred

hospitalized patients and review of the literature. J Bone Joint Surg.Am. 1979;61:1119-42.

192. Clark CR, Igram CM, el-Khoury GY et al. Radiographic evaluation of cervical spine injuries.

Spine 1988;13:742-7.

193. Abel MS. Roentgenographic aspects of post-traumatic arthritis, instability and disk deterioration

following occult fractures of cervical spine. Clin.Orthop. 1962;24:49-60.

194. Abel MS. The radiology of chronic neck pain: sequelae of occult traumatic lesions. Crit

Rev.Diagn.Imaging 1983;20:27-78.

195. Abel MS. Clinical and roentgenological aspects of occult fractures of the smaller elements of the

cervical vertebrae. Am.J Surg. 1959;97:530-8.

196. Daffner RH, Brown RR, Goldberg AL. A new classification for cervical vertebral injuries: influence

of CT. Skeletal Radiol. 2000;29:125-32.

197. Davis JW, Phreaner DL, Hoyt DB et al. The etiology of missed cervical spine injuries. J.Trauma

1993;34:342-6.

198. Woodring JH, Lee C. Limitations of cervical radiography in the evaluation of acute cervical

trauma. J.Trauma 1993;34:32-9.

199. Crone-Munzebrock W, Jend HH, Heller M et al. Spinal fractures: results and experience with

computer tomography. Arch.Orthop.Trauma Surg. 1984;103:36-41.

200. Lin JT, Lee JL, Lee ST. Evaluation of occult cervical spine fractures on radiographs and CT.

Emerg.Radiol. 2003;10:128-34.

201. Dai LY. Significance of prevertebral soft tissue measurement in cervical spine injuries. Eur.J

Radiol. 2004;51:73-6.

202. Miles KA, Maimaris C, Finlay D et al. The incidence and prognostic significance of radiological

abnormalities in soft tissue injuries to the cervical spine. Skeletal Radiol. 1988;17:493-6.

203. Anagnostara A, Athanassopoulou A, Kailidou E et al. Traumatic retropharyngeal hematoma and

prevertebral edema induced by whiplash injury. Emerg.Radiol. 2005;11:145-9.

101

204. Birsner JW, ask WH. Retropharyngeal soft tissue swelling due to whiplash injury.

AMA.Arch.Surg. 1954;68:369-73.

205. Paakkala T. Prevertebral soft tissue changes in cervical spine injury. Crit Rev.Diagn.Imaging

1985;24:201-36.

206. Ronnen HR, de Korte PJ, Brink PR et al. Acute whiplash injury: is there a role for MR imaging?--

a prospective study of 100 patients. Radiology 1996;201:93-6.

207. Borchgrevink G, Smevik O, Haave I et al. MRI of cerebrum and cervical columna within two days

after whiplash neck sprain injury. Injury 1997;Jun-Jul;28(5-6):331-5.

208. Borchgrevink G, Smevik O, Nordby A et al. MR imaging and radiography of patients with cervical

hyperextension-flexion injuries after car accidents. Acta Radiol 1995;Jul;36(4):425-8.

209. Matsumoto M, Suzuki N, Toyama Y et al. Cervical curvature in acute whiplash injuries:

prospective comparative study with asymptomatic subjects. Injury 1998;Dec;29(10):775-8.

210. Gore DR, Sepic SB, Gardner GM. Roentgenographic findings of the cervical spine in

asymptomatic people. Spine 1986;11:521-4.

211. Carroll C, Mcafee P, Riley P. Objective findings for diagnosis of whiplash. J Musculoskel Med

1986;3:57-76.

212. Hohl M. Soft-tissue injuries of the neck in automobile accidents. Factors influencing prognosis. J

Bone Joint Surg.Am. 1974;56:1675-82.

213. Watkinson A, Gargan MF, Bannister GC. Prognostic factors in soft tissue injuries of the cervical

spine. Injury 1991;22:307-9.

214. Robinson DD, Cassar-Pullicino VN. Acute neck sprain after road traffic accident: a long-term

clinical and radiological review. Injury 1993;24:79-82.

215. Pettersson K, Hildingsson C, Toolanen G et al. Disc pathology after whiplash injury. A

prospective magnetic resonance imaging and clinical investigation. Spine 1997;22:283-7.

216. Goldberg AL, Rothfus WE, Deeb ZL et al. Hyperextension injuries of the cervical spine. Magnetic

resonance findings. Skeletal Radiol. 1989;18:283-8.

102

217. Bonuccelli U, Pavese N, Lucetti C et al. Late whiplash syndrome: a clinical and magnetic

resonance imaging study. Funct.Neurol. 1999;14:219-25.

218. Harris JH, Yeakley JW. Hyperextension-dislocation of the cervical spine. Ligament injuries

demonstrated by magnetic resonance imaging. J Bone Joint Surg.Br. 1992;74:567-70.

219. Voyvodic F, Dolinis J, Moore VM et al. MRI of car occupants with whiplash injury. Neuroradiology

1997;39:35-40.

220. Krakenes J, Kaale BR, Moen G et al. MRI of the tectorial and posterior atlanto-occipital

membranes in the late stage of whiplash injury. Neuroradiology 2003;45:585-91.

221. Krakenes J, Kaale BR, Nordli H et al. MR analysis of the transverse ligament in the late stage of

whiplash injury. Acta Radiol. 2003;44:637-44.

222. Krakenes J, Kaale BR. Magnetic resonance imaging assessment of craniovertebral ligaments

and membranes after whiplash trauma. Spine 2006;31:2820-6.

223. Kwan O. MRI assessment of the alar ligaments in the late stage of whiplash injury.

Neuroradiology 2003;45:195-6.

224. Kwan O, Friel J. MRI of the posterior tectorial and atlanto-occipital membranes in the late stage

of whiplash injury. Neuroradiology 2004;46:165-6.

225. Vanezis P. Pathology of neck injury. Butterworth & Co. (Publishers) Ltd., 1989.

226. Uhrenholt L, Ullerup R, Vesterby A et al. Examination of cervical spine histological sections - a

technical note. J Whiplash & Rel Disorders 2006;5:55-60.

227. Erben RG. Embedding of bone samples in methylmethacrylate: an improved method suitable for

bone histomorphometry, histochemistry, and immunohistochemistry. J.Histochem.Cytochem.

1997;45:307-13.

228. Hahn M, Vogel M, Delling G. Undecalcified preparation of bone tissue: report of technical

experience and development of new methods. Virchows Arch.A Pathol.Anat.Histopathol.

1991;418:1-7.

229. Epidata - a comprehensive tool for validated entry and documentation of data. [computer

program]. Version Version 3.02, available at http://www.epidata.dk/.The EpiData Association,

Odense, Denmark, 2004.

103

230. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics

1977;33:159-74.

231. Bergstrom K, Nyberg G, Pech P et al. Multiplanar spinal anatomy: comparison of CT and

cryomicrotomy in postmortem specimens. AJNR Am.J.Neuroradiol. 1983;4:590-2.

232. Moore TA, Vaccaro AR, Anderson PA. Classification of lower cervical spine injuries. Spine

2006;31:S37-S43.

233. Roosendaal G, Vianen ME, Marx JJ et al. Blood-induced joint damage: a human in vitro study.

Arthritis Rheum. 1999;42:1025-32.

234. Hooiveld M, Roosendaal G, Vianen M et al. Blood-induced joint damage: longterm effects in vitro

and in vivo. J Rheumatol. 2003;30:339-44.

235. Hooiveld M, Roosendaal G, Wenting M et al. Short-term exposure of cartilage to blood results in

chondrocyte apoptosis. Am.J Pathol. 2003;162:943-51.

236. Jansen NW, Roosendaal G, Bijlsma JW et al. Exposure of human cartilage tissue to low

concentrations of blood for a short period of time leads to prolonged cartilage damage: an in vitro

study. Arthritis Rheum. 2007;56:199-207.

237. Gore DR, Sepic SB, Gardner GM et al. Neck pain: a long-term follow-up of 205 patients. Spine

1987;12:1-5.

238. Uhrenholt L, Nielsen E, Vesterby Charles A et al. Post-mortem identification of cervical spine

injuries in a road traffic crash fatality - a case report. (Submitted) 2007.

239. Thambyah A, Nather A, Goh J. Mechanical properties of articular cartilage covered by the

meniscus. Osteoarthritis.Cartilage. 2006;14:580-8.

240. Hansen AC, Kristensen IB, Dragsholt C et al. Alcohol and drugs (medical and illicit) in fatal road

accidents in a city of 300,000 inhabitants. Forensic Sci.Int. 1996;79:49-52.

241. Drummer OH, Gerostamoulos J, Batziris H et al. The incidence of drugs in drivers killed in

Australian road traffic crashes. Forensic Sci.Int. 2003;134:154-62.

242. European Road Safety Observatory (2006). Alcohol. Accessed at http://www.erso.eu/safetynet

on May 27, 2007.

104

243. Jonsson H, Jr., Rauschning W. Postoperative cervical spine specimens studied with the

cryoplaning technique. J.Orthop.Trauma 1992;6:1-9.

244. An HS, Wise JJ, Xu R. Anatomy of the cervicothoracic junction: a study of cadaveric dissection,

cryomicrotomy, and magnetic resonance imaging. J.Spinal Disord. 1999;12:519-25.

245. Tanno I, Murakami G, Oguma H et al. Morphometry of the lumbar zygapophyseal facet capsule

and cartilage with special reference to degenerative osteoarthritic changes: an anatomical study

using fresh cadavers of elderly Japanese and Korean subjects. J Orthop.Sci. 2004;9:468-77.

246. Dorph-Petersen KA, Nyengaard JR, Gundersen HJ. Tissue shrinkage and unbiased

stereological estimation of particle number and size. J Microsc. 2001;204:232-46.

247. Freeman MD, Croft AC, Nicodemus CN et al. Significant spinal injury resulting from low-level

accelerations: a case series of roller coaster injuries. Arch.Phys.Med.Rehabil. 2005;86:2126-30.

105

Appendix 1

Facet fractures on radiology and histology

RADIOLOGY (CONSENSUS) HISTOLOGY

Subject Age Gender Facet facture

on X-ray Facet facture

on CT Facet facture

on MRI Facet fracture on

10 μm 26 21 M No No No No 24 22 F No Yes Yes No 35 33 M No No No No 14 20 M No No No No 20 29 M No No No No 25 36 M Yes Yes Yes Yes 42 29 M No No No No 47 23 F No No No No 21 31 F No No No No 13 34 M No No No No 17 45 F No No No No 19 35 M No No No No 22 20 M No No No Yes 36 29 M No No No No 46 37 M No No No No 53 31 M No Yes Yes Yes 39 41 M No Yes No Yes 34 47 M No No No No 28 38 M No No No No 19 cases 1 4 3 4 16 39 F No No No No 48 46 F No No No No 12 41 M No No No No 29 34 F No No No No 43 33 M No No No No 30 43 F No No No No 38 49 M No No No No 18 37 M No No No No 45 35 M No No No No 51 23 M No No No No 15 32 M No No No No 23 37 M No No No No 31 31 M No No No No 44 40 M No No No No 52 48 M No No No No 50 41 F No No No No 49 41 F No No No No 40 27 F No No No No 54 22 M No No No No 37 49 F No No No No 33 35 M No No No No 21 controls 0 0 0 0

Gender: male (M), female (F)

App

endi

x 2

Info

rmat

ion

from

the

polic

e re

cord

s co

ncer

ning

the

road

traf

fic c

rash

fata

litie

s

106

Subj

ect

Mon

thAc

tion

# in

volv

ed

part

ies

# ki

lled

# ca

sual

ties

Roa

d co

nditi

ons

Visi

bilit

yLi

ght

Wea

ther

Allo

wed

sp

eed

km/h

Shap

e ro

adC

ircum

stan

ces

Type

of r

oad

Prim

ary

colli

sion

po

int

Estim

ated

spe

ed

km/h

26no

vem

ber

pass

enge

r2

13

dry

good

dark

ness

no ra

in70

stra

ight

obst

acle

mai

n ro

adrig

htun

know

n

24no

vem

ber

driv

er2

17

wet

good

dayl

ight

no ra

in80

stra

ight

front

al k

ollis

ion

mai

n ro

adfro

ntun

know

n

35ja

nuar

ydr

iver

11

0w

etgo

odda

ylig

htno

rain

110

stra

ight

obst

acle

mai

n ro

adfro

ntun

know

n

14ja

nuar

ydr

iver

11

0ic

ygo

odda

rkne

ssno

rain

80st

raig

htso

lom

ain

road

com

bine

dun

know

n

20ja

nuar

ydr

iver

21

4dr

y-ic

ygo

odda

rkne

ssno

rain

50st

raig

htfro

ntal

kol

lisio

nci

ty ro

adfro

nt40

25m

aydr

iver

21

0dr

ygo

oddu

skno

rain

80st

raig

htfro

ntal

kol

lisio

nm

ain

road

front

80

42se

ptem

ber

driv

er2

12

dry

unkn

own

dusk

unkn

own

80st

raig

htfro

ntal

kol

lisio

nm

ain

road

front

unkn

own

47de

cem

ber

driv

er1

11

wet

good

dark

ness

no ra

in80

stra

ight

obst

acle

mai

n ro

adco

mbi

ned

unkn

own

21ja

nuar

ydr

iver

21

0ic

yre

duce

ddu

skno

rain

80cu

rve

front

al k

ollis

ion

mai

n ro

adco

mbi

ned

70

13ja

nuar

ydr

iver

21

1w

etgo

odda

ylig

htno

rain

80st

raig

htfro

ntal

kol

lisio

nm

ain

road

front

unkn

own

17fe

brua

rydr

iver

21

0w

etgo

odda

rkne

ssno

rain

80st

raig

htfro

ntal

kol

lisio

nco

untry

road

front

unkn

own

19m

arch

driv

er2

11

wet

good

dark

ness

rain

80st

raig

htfro

ntal

kol

lisio

nm

ain

road

front

unkn

own

22ap

rildr

iver

11

0dr

ygo

odda

rkne

ssno

rain

80st

raig

htso

lom

ain

road

com

bine

dun

know

n

36m

aydr

iver

21

1dr

ygo

odda

ylig

htno

rain

80st

raig

htfro

ntal

kol

lisio

nm

ain

road

front

unkn

own

46no

vem

ber

driv

er2

12

wet

good

dark

ness

no ra

in80

inte

rsec

tion

side

impa

ctm

ain

road

left

unkn

own

53de

cem

ber

driv

er1

10

wet

good

dark

ness

no ra

in80

curv

eob

stac

lem

ain

road

left

unkn

own

39m

arch

driv

er1

10

dry

good

dark

ness

no ra

in80

curv

eob

stac

lem

ain

road

com

bine

dun

know

n

34ap

rildr

iver

11

0dr

ygo

odda

ylig

htno

rain

80cu

rve

solo

mai

n ro

adco

mbi

ned

unkn

own

28m

aydr

iver

31

2dr

ygo

oddu

skno

rain

130

stra

ight

com

bine

dhi

ghw

ayco

mbi

ned

unkn

own

Appendix 3

Toxicological findings

Trauma cases

Subject Age Gender Blood alcohol

concentration (‰) Drugs/ medicine

in blood and/or urine

26 21 M 2,08 No

24 22 F 0,00 No

35 33 M 0 No

14 20 M 0,88 No

20 29 M 0,10 Not tested

25 36 M 0,90 Metoprolol

42 29 M 0 Not tested

47 23 F 0 Not tested

21 31 F 0 No

13 34 M 0 Ketamine¥



22 20 M 0 Valproate




39 41 M (2,40)± Not tested

34 47 M 0 Citalopram

28 38 M 0 Not tested 19 cases 5 (BAC>0.5) 4

Controls


48 46 F 1,82 No


29 34 F 1,80 Cannabis¥


30 43 F 0 Clomipramine & paracetamol†


18 37 M 0 No

45 35 M 0 No


15 32 M 0 No






49 41 F 0 No

40 27 F 0 Cocaine¥

54 22 M 0 No

37 49 F Not tested Not tested

33 35 M 0 No 21 controls 3 (BAC>0.5) 3

BAC: Blood alcohol concentration Gender: M (male), F (female) ± conc. in eyefluid (vitrous humor) † conc. of intoxication (overdose) ¥ narcotics/drugs

107

Imaging occult lesions in the cervical spine facet joints

Lars Uhrenholt, DC1,2,3, Edith Nielsen, MD4, Annie Vesterby Charles, Professor, DMSc.1,3,

Ellen Hauge, PhD, MD3,5,, Markil Gregersen, Professor, DMSc.1

1 Institute of Forensic Medicine, University of Aarhus, Denmark

2 Nordic Institute of Chiropractic and Clinical Biomechanics, Part of Clinical Locomotion Science, Odense,

Denmark

3 Research Unit of Rheumatology and Bone Biology, Aarhus University Hospital, Aarhus Sygehus (NBG),

Denmark

4 Department of Neuroradiology, Aarhus University Hospital, Aarhus Sygehus (NBG), Denmark

5 Department of Rheumatology, Aarhus University Hospital, Aarhus Sygehus (NBG), Denmark

Corresponding author:

Lars Uhrenholt

Institute of Forensic Medicine

University of Aarhus

Brendstrupgaardsvej 100, 8200 Aarhus N, Denmark

E-mail: [email protected]

Facsimile: 45-86125995, Phone: 45-89429800

PAPER I

Abstract

Discrete injuries in the lower cervical spine facet joints have been reported in studies of motor vehicle crash

victims. We conducted a detailed investigation of these joints from 20 motor vehicle crash fatalities and 22

decedents due to non-traumatic causes using conventional radiology, computed tomography and magnetic

resonance imaging in order to examine whether the diagnostic imaging procedures could identify injuries in

the facet joints. The diagnostic imaging procedures identified facet joint fractures in 4 of the 19 trauma cases

with computed tomography having the highest sensitivity and obtaining good correlation with findings from

the microscopical evaluation. No diagnostic imaging procedure could reliably evaluate the integrity of the

synovial folds or the joint spaces for bleeding despite microscopical evidence of such findings in these

structures in a large proportion of the motor vehicle crash fatalities. This study emphasizes the need for

scientific evidence of validity and reliability of advanced diagnostic imaging procedures in forensic settings, in

particular with regard to occult soft tissue lesions, and cautions uncritical use of negative results from these

procedures until such evidence has been produced.

Introduction

One of the primary functions of radiological evaluation in forensic medical settings is to confirm the presence

or absence of fractures in victims from assaults, and investigate road traffic crashes, child abuse cases and

other criminal or suspect cases1. With Computed Tomography (CT) and Magnetic Resonance Imaging (MRI)

procedures more detailed analysis can be performed, compared to previous methods such a conventional x-

rays, and these procedures are being utilised increasingly in forensic settings around the world2-5.

Furthermore, diagnostic imaging findings, such as typical skull fractures, spinal and extremity fractures can

often be verified by localised incision of the lesion site2;4. However, recent studies have suggested that

discrete injuries in the spinal column may remain undetected on advanced diagnostic imaging only to be

identified on microscopical examination6-8. In a systematic review of studies examining the cervical spine at

post-mortem of road traffic crash fatalities discrete injuries in the cervical spine, including intervertebral discs

lesions, facet joint fractures and injuries to the joint capsules, synovial folds and ligamentous structures,

were common and many of the discrete injuries remained un-diagnosed on conventional radiological

examination7, which has been supported in subsequent studies9-11. Hence, several studies have found that

minor injuries are often not detected on conventional x-rays, but even CT an MRI have been found to miss

discrete injuries4;6;8;12-18. The highest frequency of injury to the cervical spine facet joints in post-mortem

studies has been reported to be in the lower cervical spine between C4 and C719 20. The purpose of this

study was to investigate whether advanced diagnostic imaging procedures could detect facet joint fractures

and injuries to the synovial folds in the lower cervical spine in a controlled group of post-mortem individuals

using microscopical data as the gold standard.

Materials and methods

Twenty motor vehicle crash fatalities (cases) and 22 non-traumatic decedents (controls) were included,

consisting of 12 females (median age 40, range 22–49 years) and 30 males (mean age 34.5, range 20–

49 years). From the subjects the lower four cervical spine segments were removed and the facet joints were

examined. Due to post-mortem decomposition two subjects were withdrawn from statistical analysis. The

material has been described in detail elsewhere21. The study was approved by the Scientific Ethics

Committee.

All subjects were examined within 24 hours after autopsy using conventional x-rays, computed tomography

(CT) and magnetic resonance imaging (MRI). Conventional x-ray examination was performed on an

Arcosphere, Arcoma© and consisted of the following projections; AP, lateral and bilateral articular pillar and

oblique views. The film to object distance was one meter, the consol setting of 50 Kv and 32-50 mAs, and

the use of high focus extremity cassettes ensured optimal quality of the images. Computed tomography was

performed on a Phillips© four-slice MX8000 CT scanner. The examination was performed as spiral scanning

of the total volume of the specimen, with a slice thickness of 1,3 mm, increments of 0.6, pitch 0,875, 120 Kv

and 250 mAs per slice using a matrix of 512x512. Axial, coronal, saggital and angled reconstructions were

performed with bone and soft tissue algorithms. Magnetic resonance imaging was performed on a 1.5 T

Signa scanner from General Electrics© using a surface coil. Saggital T1, T2, STIR-sequences, axial T1, T2

and T2*gradient-sequences were performed of the facet joint structures. The majority of sequences were

made with a thickness of 3 mm, spacing 0 and FOV 16x16. The matrix was 512x512 on the saggital

sequences and 256x256 on the remaining.

In the first evaluation of radiological images, two experienced neuroradiologists (obs1 and obs2) scored all

the retrieved images dichotomously, independently and blinded, with regard to general facet fractures and

haemarthrosis/excess fluid in the facet joints. Prior to the first evaluation the observers received no training

or coaching, hence the scoring of observations was based on routine clinical interpretation. During a second

evaluation the observers obtained consensus agreement with regard to the presence or absence of fractures

on all imaging modalities, and haemarthrosis/excess fluid in a facet joint on conventional x-ray and CT-

scanning only. A third evaluation of the MR-images was performed by both observers independently and

blinded with regard to the presence or absence of haemarthrosis/excess fluid in the facet joints only.

Consensus was obtained by the observers, with the variable defined as a joint that on MRI showed focally

increased signal, with signal differences between adjacent joints on any sequence or relatively increased

signals in all joints.

Each specimen was prepared for microscopical evaluation according to previously published methods21-23,

and the microscopical findings of the 10 μm thick histological sections have been described in detail

elsewhere21;24.

The association of groups was tested with Fisher’s exact test (significance level was p<0.05) and

measurement of agreement between observers was tested with kappa statistics25. All statistical analyses

were performed using Stata© 9 (StataCorp LP, College Station, USA).

Results

The first evaluation was performed on images from the conventional x-ray examination, CT-scanning and

MRI (Table 1). On the conventional x-ray images, one subject with facet joint fracture was identified by both

observers (1/19 cases) and one subject with facet joint fracture by one observer only (1/19 cases), with a

simple agreement of 97.5% ( = 0.66, 95% confidence interval: 0.37-0.95). On CT-scanning three subjects

with facet joint fractures (3/19 trauma cases) were identified by both observers and two subjects with facet

joint fractures (1/19 cases and 1/21 controls) by one observer only, with an interobserver agreement of

95.0% ( = 0.72, 95% confidence interval: 0.42-1.02). On MRI two subjects with facet joint fractures were

identified by both observers (2/19 cases) and one subject with facet joint fracture by one observer only (1/19

cases), with an interobserver agreement of 97.5% ( = 0.79, 95% confidence interval: 0.49-1.09). No blood

or excess fluid in a joint could be determined on conventional x-ray. One subject (1/19 cases) was scored

positive for bleeding or excess fluid in a facet joint on CT. On MRI three subjects (2/19 cases and 1/21

controls) were scored positive for haemarthrosis/excess fluid in a joint by both observers, and 15 were

scored positive by one observer (7/19 cases and 8/21 controls). The interobserver agreement on the

presence of haemarthrosis/excess fluid in a facet joint on MRI was 62.5% ( = 0.12, 95% confidence interval:

-0.12 to 0.36).

Following the consensus agreement obtained during the second evaluation (Table 1) the presence of

fractures were agreed upon in four trauma cases (4/19) and no controls. Computed tomography was the only

modality capable of identifying all four cases with facet fractures (Figure 1), where MRI identified three and

conventional x-ray one case with facet fractures. Agreement between MRI and CT with regard to subjects

with fractures was 97.5% ( = 0.84, 95% confidence interval: 0.53-1.15), whereas the agreement between

conventional x-ray and CT was 92.5% ( = 0.38, 95% confidence interval: 0.13-0.62). Only on CT there was

statistically significant correlation between subjects with fractures and trauma (p = 0.042, two-sided Fisher’s

exact test). On neither conventional x-ray nor CT was there any positive scoring for the presence of

haemarthrosis/excess fluid in a joint.

The third evaluation (Table 1) revealed haemarthrosis/excess fluid in a joint on MRI in 32 subjects (14/19

cases and 18/21 controls) by both observers, and six were scored positive by one observer (3/19 cases and

3/21 controls). The interobserver agreement of the presence of haemarthrosis/excess fluid in a facet joint on

MRI after the third evaluation was 85.0% ( = 0.31, 95% confidence interval: 0.00-0.62). There was no

statistically significant correlation between being a trauma case or control subject and haemarthrosis/excess

fluid in a joint on the MRI. Consensus agreement with regard to bleeding on MRI was not obtained.

Microscopical evaluation of the 10 μm thick sections revealed facet fractures in four subjects (4/19 cases)26,

three of whom were identified on neuroradiological evaluation and the remaining case with facet fracture only

identified on histology. Among the four subjects with microscopical fractures a total of 11 unique facet

fractures were identified (Table 2). Furthermore, a fifth subject with a facet fracture identified on CT could not

be confirmed on microscopy. The subject based analysis between the CT consensus findings and the

microscopical findings regarding subjects with fractures resulted in 95% agreement ( = 0.72, 95%

confidence interval: 0.41-1.03) (Table 3) and the correlation between the microscopical findings and CT

consensus findings with regard to unique facet fractures was statistically significant (p < 0.001)(Table 2).

Bleeding in any joint space was present in nine (9/19 cases) and one (1/21 controls) subjects, and

simultaneous bleeding in both synovial folds at any one segment was present in five cases (5/19) and no

controls.

Discussion

In this post-mortem diagnostic imaging study of the lower cervical spine facet joints, facet fractures were

identified on diagnostic imaging in four of the trauma cases, with CT-scanning being the only imaging

modality capable of identifying all these fractures and a statistically significant correlation between trauma

and fractures. In comparison microscopical evaluation confirmed three of the four facet fractures identified at

CT and revealed another facet fracture not seen on diagnostic imaging. There was substantial agreement

between the CT consensus findings and the microscopical findings with regard to facet fractures. Bleeding in

the facet joints and the facet joint synovial folds could not be confirmed reliably on any diagnostic imaging

procedure, despite microscopical evidence of bleeding in the facet joints.

Fractures of the lower cervical spine facet joints have been described in previous investigations of road

traffic crash fatalities using diagnostic imaging and autopsy data6-8;19;20;27-29. In a post-mortem study of 22

road traffic crash victims, only one fracture of a facet was identified on x-rays and confirmed

microscopically19. Although, x-ray evaluation initially identified four facet fractures these were all read false-

positive due to the lack of microscopical confirmation, which was based on photographic documentation of

the specimens at interval of approximately 1-mm thickness. In a study of 32 post-mortem subjects, two

subjects (2/15 motor vehicle crash fatalities) had suffered fractures to the lower cervical spine facet joints

that could not be identified on conventional x-rays27. More recently, a study of 10 accident victims revealed

two fractures identified on microscopical evaluation in the vicinity of the cervical facet joint (lamina fractures)

that could not be detected on MRI or conventional x-rays, however no distinct fractures to the facets were

identified6. Our study comprised 40 well described subjects (19 road traffic crash fatalities and 21 controls)

that had been investigated in detail using techniques very similar to those utilised in these previous studies.

However, we found an incidence rate of 4/19 subject with facet fractures on the histological sections in the

trauma group which is markedly higher than previously reported, with good agreement between the

observations from CT and microscopy of the 10 μm sections. Furthermore, we found a total of 11 unique

facet fractures on microscopy, divided among these four subjects, of which seven were not reported on the

CT evaluation, and a fifth subject with potential fractures was identified on CT that could not be verified on

microscopy (Table 2). This may suggest that the previous studies6;19;27 of mixed trauma mechanisms (falls,

sporting injuries and different types of road traffic crashes, e.g. motorcycle crashes and pedestrian injuries)

have different outcomes with regard to the incidence rate of facet joint injuries due to differences in the

mechanism of trauma, which may be very different from passenger car crashes. It also suggests that, when

present, facet fractures are more often affecting several facets rather than solitary facets. None of these

findings could be confidently confirmed on the stereomicroscopic evaluation of the 3-mm thick slices, since

the results from this evaluation showed no correlation with trauma or the microscopical findings21.

Haemarthrosis of the cervical spine facet joints have been described in post-mortem studies of road traffic

crash victims using histological and/or microtomal methods6-8;19;20;27;28. In a study of 10 accident victims

(unknown proportion of road traffic crash victims), haemarthrosis was reported in 10 joints on microscopical

evaluation, with only six of the 10 joints with haemarthrosis identified on MRI, which was conducted

retrospectively6. Besides this study, no other forensic diagnostic imaging studies have described this

condition in detail. Our results support these finding6, in that the presence of haemarthrosis or bleeding in the

synovial folds cannot reliably be established based on diagnostic imaging procedures, despite microscopical

evidence of bleeding in one or more joints in almost half the trauma cases and simultaneous bleeding in the

segmental synovial folds in a quarter of the trauma cases21. Numerous studies have been published

investigating the contributions from MRI and CT in forensic settings, and it can be anticipated that the

increasing use of these modalities gradually will replace the role of conventional x-rays in the screening of

the deceased during autopsy2-4;8. However, in the study of the cervical spine of road traffic crash fatalities

only few studies have utilised advanced diagnostic imaging procedures2;6;8;28;29, whereas a substantial

number of studies have included conventional x-ray evaluation7. In a large post-mortem study of 109 trauma

fatalities, a substantial number of injuries were missed on conventional x-ray evaluation although the exact

detection rate of facet joint fractures was not reported20. Except for one recent study6, the majority of

previous post-mortem studies examining the cervical spine have not attempted to confirm microscopical or x-

ray findings with CT or MRI19;20;27. We found that only one of the four fractures identified on CT could be

visualised on conventional x-ray evaluation, which suggests that the previous studies utilizing conventional x-

rays only may have under-reported the true incidence of fractures that potentially could have been detected

on advanced diagnostic imaging procedures. Similarly, our results supports other studies where CT has

been shown to be the most reliable diagnostic imaging procedure with regard to detecting fractures of the

cervical spine2;9-11;18. Notwithstanding this fact, a recent post-mortem forensic neuroimaging study, reported

that CT and MRI missed two (2/2) atlas C1 fractures, two fractures at axis C2 (2/4) on CT and three fractures

at axis C2 (3/4) on MRI, with the injuries being confirmed on autopsy2. With regard to soft tissue lesions in

the facet joints, such as bleeding in the joint space and the synovial folds, we found, in accordance with

other studies, that the presence and structural integrity of synovial folds of the lower cervical spine could not

be determined reliably on neither conventional x-ray, CT or MRI6;8;12;30. Recently, however, the cervical spine

facet joint folds have clearly been visualised on MRI, although this required a high energy-field 3.0 Tesla MRI

unit, with long acquisition time, limited field of view and a small specimen size31. In our study we used a 1.5

Tesla MRI unit and the protocols used did not allow adequate resolution, and the slight interobserver

agreement is therefore probably more a consequence of technical limitations rather that academic2;31. With

increasing availability of high energy-field MR-scanners reliable evaluation of the structural integrity of the

synovial folds and identification of bleeding in the joint spaces and the synovial folds becomes increasingly

possible. We found good indication for utilising CT and to a certain extent MRI for the purpose of identifying

fractures of post-mortem cervical spine facet joints. However, we did not find support in our study for use of

these modalities for the purpose of identifying bleeding in facet joints or synovial folds.

There is a scarcity of publications addressing the reliability and validity of forensic diagnostic imaging

evaluation. No previous publication have described the degree of inter- and intraobserver agreement nor the

exact effect of training of observers with regard to scoring of forensic radiological images, although recent

comments in this regard have been proposed2. In comparison, a recent clinical study evaluated the reliability

of in vivo CT-scanning, in which a clinical classification system for lower cervical spine injuries based on

conventional radiology and CT-scanning was proposed based on excellent intra- and interobserver

agreement32. We found substantial interobserver agreement with regard to facet fractures, however the

interobserver agreement of the MRI evaluation regarding the presence of haemarthrosis/excess fluid in a

facet joint was slight despite training of the observers with regard to scoring this variable. Actually, the

training towards scoring the facet joints on MRI with regard to haemarthrosis/excess fluid in a joint caused a

significant increase in the number of joints positive for this variable which had no correlation to cases or

controls, suggesting either improper definition of the variable or clinically irrelevant structural changes in

these joints at post-mortem.

Histological procedures have the advantages of allowing microscopical evaluation of relevant sections as

well as increasing sensitivity by adequate staining procedures. However, similar to previous studies6;20;27, the

sensitivity towards identifying very discrete fractures could be limited by the section thickness, which was 3-

mm in our study. Linear lesions, parallel with the parasaggital line of sectioning, and lesions smaller than 3

millimetres could potentially remain undetected on the microscopical examination21. Furthermore, the long

fixation time in alcohol and the embedding of the un-decalcified specimens in MMA caused the erythrocytes

to appear less vividly on the consequent un-deplastified staining. This may explain why some fractures were

difficult to identify on microscopical evaluation. By reducing the section thickness more injuries would be

detectable on microscopical examination. Likewise, increasing the technical specifications of the

radiographic equipment and improving the radiologists knowledge of post-mortem anatomical changes

should improve the sensitivity of these procedures2.

Advanced diagnostic imaging procedures, in particular CT, did detect facet joint fractures in a substantial

proportion of the lower cervical spine facet joints of post-mortem individuals killed in a motor vehicle crash

and showed good correlation with findings from microscopical evaluation. The findings confirm that discrete

osseous injuries are common in the facet joints following fatal road traffic crashes, and that not all discrete

facet fractures are identified on diagnostic imaging procedures. Furthermore, bleeding in the cervical facet

joints and synovial folds could not be determined reliably on any diagnostic imaging modality although

microscopical evaluation revealed bleeding in a large proportion of the facet joints and synovial folds of the

motor vehicle crash fatalities. This study emphasizes the need for scientific evidence of validity and reliability

of advanced diagnostic imaging procedures in forensic settings, in particular with regard to non-fatal soft

tissue lesions, and until such evidence has been produced these procedures must be considered adjunct

techniques to the post-mortem autopsy.

Acknowledgements

This study has complied with the national laws on scientific research projects and has been approved by the

Scientific Ethics Committee, Aarhus County. The study was generously supported by the European

Chiropractors Union Research Fund Grant no. A.03-5, Switzerland, the Foundation for the Advancement of

Chiropractic Research, Denmark and the University of Aarhus Research Foundation, Denmark. The authors

would like to acknowledge the original contributions from Professor DMSc. Flemming Melsen†, the timely and

professional cooperation of the staff at the Institute of Forensic Medicine, University of Aarhus and the

Department of Neuroradiology, Aarhus Sygehus. Furthermore, we acknowledge the efforts of Rita Ullerup for

the careful preparation of the histological sections and the neuroradiological observations made by senior

consultant Vibeke Fink-Jensen.

References



2. Yen K, Lovblad KO, Scheurer E et al. Post-mortem forensic neuroimaging: Correlation of MSCT and

MRI findings with autopsy results. Forensic Sci.Int. 2007;Epub ahead of print.

3. Buck U, Naether S, Braun M et al. Application of 3D documentation and geometric reconstruction

methods in traffic accident analysis: With high resolution surface scanning, radiological MSCT/MRI

scanning and real data based animation. Forensic Sci.Int. 2006;Epub ahead of print.

4. Poulsen K, Simonsen J. Computed tomography as routine in connection with medico-legal autopsies.

Forensic Sci.Int. 2006;Epub ahead of print.

5. Aghayev E, Jackowski C, Sonnenschein M et al. Virtopsy hemorrhage of the posterior cricoarytenoid

muscle by blunt force to the neck in postmortem multislice computed tomography and magnetic

resonance imaging. Am.J Forensic Med.Pathol. 2006;27:25-9.


comparison of radiographic, MR imaging, anatomic, and pathologic findings. Radiology 2001;221:340-

6.

7. Uhrenholt L, Grunnet-Nilsson N, Hartvigsen J. Cervical spine lesions after road traffic accidents: a

systematic review. Spine. 2002;27:1934-41.

8. Uhrenholt L, Nielsen E, Vesterby A et al. Detailed examination of the lower cervical spine facet joints in

a road traffic crash fatality - a case study. IRCOBI (annual proceedings) 2005;paper 701;411-414.

9. Griffen MM, Frykberg ER, Kerwin AJ et al. Radiographic clearance of blunt cervical spine injury: plain

radiograph or computed tomography scan? J.Trauma 2003;55:222-6.

10. Gale SC, Gracias VH, Reilly PM et al. The inefficiency of plain radiography to evaluate the cervical

spine after blunt trauma. J.Trauma 2005;59:1121-5.

11. Holmes JF, Akkinepalli R. Computed tomography versus plain radiography to screen for cervical spine

injury: a meta-analysis. J.Trauma 2005;58:902-5.

12. Fletcher G, Haughton VM, Ho KC et al. Age-related changes in the cervical facet joints: studies with

cryomicrotomy, MR, and CT. AJR Am.J.Roentgenol. 1990;154:817-20.

13. Pech P, Kilgore DP, Pojunas KW et al. Cervical spinal fractures: CT detection. Radiology

1985;157:117-20.

14. Rauschning W, Bergstrom K, Pech P. Correlative craniospinal anatomy studies by computed

tomography and cryomicrotomy. J.Comput.Assist.Tomogr. 1983;7:9-13.

15. Sweeney JF, Rosemurgy AS, Gill S et al. Is the cervical spine clear? Undetected cervical fractures

diagnosed only at autopsy. Ann.Emerg.Med. 1992;21:1288-90.

16. Woodring JH, Lee C. The role and limitations of computed tomographic scanning in the evaluation of

cervical trauma. J Trauma 1992;33:698-708.

17. Klein GR, Vaccaro AR, Albert TJ et al. Efficacy of magnetic resonance imaging in the evaluation of

posterior cervical spine fractures. Spine 1999;24:771-4.

18. Holmes JF, Mirvis SE, Panacek EA et al. Variability in computed tomography and magnetic resonance

imaging in patients with cervical spine injuries. J.Trauma 2002;53:524-9.

19. Jonsson H, Jr., Bring G, Rauschning W et al. Hidden cervical spine injuries in traffic accident victims

with skull fractures. J.Spinal Disord. 1991;4:251-63.



21. Uhrenholt L, Vesterby A, Hauge E et al. Pathoanatomy of the lower cervical spine facet joints in motor

vehicle crash fatalities. (Submitted) 2007.

22. Uhrenholt L, Ullerup R, Vesterby A et al. Examination of cervical spine histological sections - a

technical note. J Whiplash & Rel Disorders 2006;5:55-60.

23. Hahn M, Vogel M, Delling G. Undecalcified preparation of bone tissue: report of technical experience

and development of new methods. Virchows Arch.A Pathol.Anat.Histopathol. 1991;418:1-7.

24. Uhrenholt L, Vesterby Charles A, Hauge E et al. Semi-quantitative morphology of the lower cervical

spine facet joints. (Submitted) 2007.

25. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics

1977;33:159-74.

26. Uhrenholt L, Nielsen E, Vesterby Charles A et al. Imaging occult lesions in the cervical spine facet

joints. (Submitted) 2007.


1993;18:1115-22.



29. Rauschning W, McAfee PC, Jonsson H, Jr. Pathoanatomical and surgical findings in cervical spinal

injuries. J.Spinal Disord. 1989;2:213-22.

30. Yu SW, Sether L, Haughton VM. Facet joint menisci of the cervical spine: correlative MR imaging and

cryomicrotomy study. Radiology 1987;164:79-82.

31. Friedrich KM, Trattnig S, Millington SA et al. High-field magnetic resonance imaging of meniscoids in

the zygapophyseal joints of the human cervical spine. Spine 2007;32:244-8.

32. Moore TA, Vaccaro AR, Anderson PA. Classification of lower cervical spine injuries. Spine

2006;31:S37-S43.

Figure 1 3-D reconstruction of CT-images of a facet joint fracture in a motor vehicle crash fatality

A fracture (arrow) is localised in the superior articular process of C6 on the left.

Figure 2 Histological microimage of a fracture of a lower cervical spine facet joint following a fatal motor

vehicle crash

Overview of a parasaggital histological section through the lower cervical spine facet joints shows a

fracture (F) of the superior articular process (SAP) of C6 on the left, and the shaded area is visualised in

a close-up illustrating the fractures (F) and haemarthrosis in the joint space (JS), original magnification

1.25x, Masson Goldner-Trichrome.

Tabl

e 1

Eva

luat

ions

of t

he n

euro

radi

olog

ical

imag

es w

ith re

gard

to fr

actu

res

and

haem

arth

rosi

s/ex

cess

flui

d in

a fa

cet j

oint

X-

ray

CT

MR

I

Cas

es

obs1

-obs

2 fra

ctur

e ha

em.

fract

ure

haem

. fra

ctur

e ha

em.

1. e

valu

atio

n (n

=19)

0-

0 17

19

15

18

16

10

0-

1 1

0 0

1 0

5

1-

0 0

0 1

0 1

2

1-

1 1

0 3

0 2

2

C

ontr

ols

0-0

21

21

20

21

21

12

(n

=21)

0-

1 0

0 0

0 0

8

1-

0 0

0 1

0 0

0

1-

1 0

0 0

0 0

1

2.

eva

luat

ion

Cas

es

0-0

18

19

15

19

16

(n

=19)

0-

1 0

0 0

0 0

1-0

0 0

0 0

0

1-

1 1

0 4

0 3

-

Con

trol

s 0-

0 21

21

21

21

21

(n=2

1)

0-1

0 0

0 0

0

1-

0 0

0 0

0 0

1-1

0 0

0 0

0

-

3. e

valu

atio

n C

ases

0-

0 2

(n

=19)

0-

1 2

1-0

1

1-

1

- -

-

14

Con

trol

s 0-

0 0

(n

=21)

0-

1 1

1-0

2

1-

1

- -

-

18

fract

ure:

frac

ture

of a

cer

vica

l spi

ne fa

cet j

oint

ha

em.:

haem

arth

rosi

s/ex

cess

flui

d in

a c

ervi

cal s

pine

face

t joi

nt

obs1

: obs

erve

r one

ob

s2: o

bser

ver t

wo

Table 2 Subjects scoring positive for facet fracture of either histology or CT consensus

Subject # 10 μm histology CT consensus

22 (case) C5 right IAP fracture -

25 (case)

C5 left IAP fracture

C6 left SAP fracture C6 left IAP fracture

C7 left SAP fracture

C6 right SAP fracture

-

C6 left SAP fracture -

-

-

39 (case)

Th1 left SAP fracture -

C5 right IAP fracture

Th1 left SAP fracture C7 left IAP fracture

-

53 (case)

C7 left IAP fracture

C6 right IAP fracture C7 right SAP fracture

-

C6 right IAP fracture C7 right SAP fracture

24 (case) - Th1 left SAP fracture

The total number of unique facets is 16 per subject, except for one subjects with a bloc vertebra with 12 facets (636 facets in total among the 40 subjects)

Correlation per unique facet fracture (Fishers exact test): risk ratio 60 (95% confidence interval: 24 - 152, p < 0.001)

SAP: superior articular process

IAP: inferior articular process

Table 3 Agreement between CT consensus and histological findings regarding facet fracture

Subjects (n = 40) CT – 10 μm Fracture

0 - 0 35

0 - 1 1

1 - 0 1

1 - 1 3 ( = 0.72, 95% confidence interval: 0.41-1.03)

CT: computed tomography scanning

10 μm: microscopy of 10 μm thick histological sections

Fracture: cervical spine facet joint fractures

Table 4 Subject based description of facet fractures on radiology and histology

RADIOLOGY (CONSENSUS) HISTOLOGY

Subject # Age Gender Facet facture

on X-ray Facet facture

on CT Facet facture

on MRI Facet fracture on

10 μm 26 21 M No No No No 24 22 F No Yes Yes No 35 33 M No No No No 14 20 M No No No No 20 29 M No No No No 25 36 M Yes Yes Yes Yes 42 29 M No No No No 47 23 F No No No No 21 31 F No No No No 13 34 M No No No No 17 45 F No No No No 19 35 M No No No No 22 20 M No No No Yes 36 29 M No No No No 46 37 M No No No No 53 31 M No Yes Yes Yes 39 41 M No Yes No Yes 34 47 M No No No No 28 38 M No No No No 19 cases 1 4 3 4 16 39 F No No No No 48 46 F No No No No 12 41 M No No No No 29 34 F No No No No 43 33 M No No No No 30 43 F No No No No 38 49 M No No No No 18 37 M No No No No 45 35 M No No No No 51 23 M No No No No 15 32 M No No No No 23 37 M No No No No 31 31 M No No No No 44 40 M No No No No 52 48 M No No No No 50 41 F No No No No 49 41 F No No No No 40 27 F No No No No 54 22 M No No No No 37 49 F No No No No 33 35 M No No No No 21 controls 0 0 0 0

Gender: male (M), female (F)

Pathoanatomy of the lower cervical spine facet joints in motor

vehicle crash fatalities

Lars Uhrenholt1,2,3, Annie Vesterby Charles1,3, Ellen Hauge3,4, Markil Gregersen1



Denmark


Denmark



Lars Uhrenholt

Institute of Forensic Medicine

University of Aarhus


E-mail: [email protected]

Facsimile: 45-86125995

Phone: 45-89429800

PAPER II

Acknowledgements

The authors would like to acknowledge the original contributions from Professor DMSc. Flemming Melsen†,

the professional cooperation of the staff at the Institute of Forensic Medicine, University of Aarhus. The

efforts of Rita Ullerup for the careful preparation of the histological sections are greatly appreciated. The

study received support from the European Chiropractors Union Research Fund Grant no. A.03-5,

Switzerland, the Foundation for the Advancement of Chiropractic Research, Denmark and the University of

Aarhus Research Foundation, Denmark.

Abstract

In decedents from motor vehicle crashes, lesions to the cervical spine facet joints have previously been

described that were non-lethal. The aim of this study was to conduct a detailed examination of the lower

cervical spine facet joints in a population of motor vehicle crash victims and controls using comparable data

from medicolegal autopsy, stereomicroscopy and histological evaluations. Injuries to the cervical spine facet

joints were common and included facet fractures, haemarthrosis and bleeding in the synovial folds. The

majority of injuries could not be verified by stereomicroscopic evaluation, and conventional autopsy

procedures did not reveal any of the injuries to the facet joints. Despite the presence of these

pathoanatomical lesions in road traffic crash fatalities their potential clinical implications in survivors from

motor vehicle crashes are unknown.

Introduction

Road traffic crashes remain responsible for the death of more than 40.000 people in Europe annually, and in

addition to the fatalities there is an estimated 3.5 million casualties per year in Europe with estimated costs

of 250 billion Euro1;2. Non-fatal injuries to the cervical spine facet joints have been related to the clinical

picture of chronic neck pain following road traffic crashes which is responsible for substantial morbidity3-6.

However, the aetiology of chronic neck pain with regard to which somatic structures may be injured,

damaged or diseased remains unclear. The limited anatomical knowledge of non-fatal injuries in the cervical

spine facet joints is related to often inconclusive clinical diagnostic imaging studies despite subjective

complaints from the patients. Despite advancements in the field of diagnostic imaging procedures, including

improved resolution of computed tomography and magnetic resonance imaging, discrete injuries may remain

undetected on these investigations7-9. Since histological procedures are not viable for in vivo human studies,

post-mortem studies have been conducted in order to examine the potential anatomical basis for sustained

neck pain following trauma. In several post-mortem studies, non-fatal lesions in the cervical spine of road

traffic crash fatalities have been identified, including injuries to the facet joint capsules, folds and articular

surfaces9-14. Only few studies, however, have evaluated homogenous study groups including details of the

mechanism of trauma, cause of death and controls subjects. Therefore, the extent of non-fatal injuries to the

cervical spine facet joints with regard to types of lesions and anatomical locations is unknown in both clinical

settings and post-mortem material.

The aim of this study was to examine the lower cervical spine facet joints in a population of motor vehicle

crash fatalities compared with an age-matched control group for the presence of cervical spine facet joint

injuries using comparable data from medicolegal autopsy, stereomicroscopy and histological evaluations

using the histological findings as gold standard.


Materials

Forty-two subjects were examined at medicolegal autopsy within 1-5 days (median 3 days) after death and

included in this study. During autopsy the lower four cervical vertebral segments (C4-C7) including

paravertebral muscles were removed en bloc, from twelve females (median age 40 years, range 22–49) and

30 males (median age 34.5 years, range 20–49 years). Twenty died in a passenger car crash (cases) and 22

died due to non-traumatic causes (controls). Subjects were excluded if there was evidence of previous

history of drug and/or alcohol abuse, previous cervical spine injury, and extensive post mortem

decomposition. Two subjects (one case and one control) were excluded from statistical analysis due to

extensive decomposition causing generalized tissue damage. The study was approved by the Scientific

Ethics Committee.

Methods

The specimens were fixated in 70 % ethanol for a minimum of five weeks with hemisection in the median

plane after two weeks, followed by one week in 96 % ethanol under vacuum conditions fixation and a final

week in 99 % ethanol. Then each specimen was embedded in liquid methylmethacrylate (MMA) and

dibutylphtalate (a softener) in a refrigerator for six weeks, after which percadox (a catalyst) was added to

complete polymerisation and hardening into a plastic bloc. From a random central starting point, each

plastic bloc was sawed by serial sectioning, into approximately seven 3-mm thick parasaggital slices using a

precision guided bandsaw (Femi©, Bologna, Italy), which ensured exact thickness of the slices with a tissue

loss of approximately 1 mm per slice. Each 3-mm thick slice was examined using a stereomicroscope and

scored with reference to damage of the facet cartilage and/or bone, bleeding in a joint, the presence of

synovial folds and injury to the folds.

The 3-mm thick slices containing facet joints were re-embedded in MMA. Using a heavy-duty microtome (SM

2500, Leica©) two consecutive histological sections of 10 μm thickness were produced from each bloc,

stained with Masson Goldner-Trichrome, and mounted un-deplastified on glass slides. All facet joints on

histological sections were examined by light microscopy using a BX51, Olympus© microscope. An average of

45.8 unique observations per subject (2.9 observations per unique facet) were made including the scoring of

predetermined variables; articular facet fractures, osteochondral fissures with or without bleeding, bleeding in

the joint space, bleeding in the anterior, posterior or both synovial folds, bleeding in the underlying bone and

the integrity of the anterior and posterior folds.

Fisher’s exact test was used to examine the significance of association between categorical variables and

Chi-squared test was used to examine the correlation of the number of disrupted folds per subject

(case/control) versus exposure to trauma. Agreement between diagnostic methods was examined with

kappa statistics. The significance level was p<0.05. All statistical analyses were performed using Stata© 9

(StataCorp LP, College Station, USA).

Results

Autopsy findings

The medicolegal autopsy identified no injuries to the lower cervical spine facet joints (Table 1). In the trauma

group seven unique fractures in cervical spine vertebral bodies were identified in four subjects (4/19 cases),

with one classified as a disco-vertebral avulsion fracture at C6-C7, and the remaining fractures situated at or

above C3. Skull fractures were identified in eight cases (8/19). There were significant injuries to cranial

organs in 11/19 cases, thoracic organs in 18/19 cases, and abdominal organs in 17/19 cases. Injuries to the

spleen and liver affected 15/19 cases and injuries to the lungs affected 15/19 cases. The primary cause of

death in the trauma group was multiple injuries in 10/19 cases and bleeding in 6/19 cases. In eighteen of the

cases (18/19 cases) the decedent was the driver of a passenger car and the remaining one case (1/19

cases) was a passenger in a motor vehicle. In the control group subjects (n=21) there were neither injuries to

the musculoskeletal system nor traumatically induced organic injuries. The most common causes of death

among the control group subjects was cardiovascular disease in 12/21 subjects. Toxicological evaluation

revealed a blood alcohol concentration (BAC) > 0.8 mg/ml in eight subjects (5/19 cases and 3/20 controls

tested), with four drivers (4/18 drivers) killed in a motor vehicle crash being under the influence of alcohol

(BAC > 0.8 mg/ml in three cases and alcohol concentration in the vitreous humor of 2.4 mg/ml in one case)

(Table 2). Medication was found in four subjects (3/9 cases tested and 1/10 controls tested) and narcotics

were found in three subjects (1/9 cases tested and 2/10 controls tested).

Stereomicroscopy of the 3-mm thick slices

On the stereomicroscopy of the 3-mm slices damage to the facet cartilage and/or bone was present in 10

subjects (3/19 cases and 7/21 controls) and the presence of damage could not be evaluated in 29 subjects

(15/19 cases and 14/21 controls). There was no significant correlation between exposure of trauma and the

presence of damage to the facet cartilage and/or bone (p = 0.28). Bleeding in a joint was present in 3

subjects (3/19 cases), which did not correlate with exposure to trauma (p = 0.10), and the presence of

bleeding in a joint could not be evaluated in 26 subjects (12/19 cases and 14/21 controls). Synovial folds

were found in 21 subjects (21/40) and could not be evaluated in 21 subjects (10/19 cases and 11/21

controls) and no injuries to the folds were detected.

Light microscopy of the 10 μm thick sections

Light microscopy of the 10 μm sections initially identified facet fractures (Figure 1) in three subjects (3/19

cases) however, after review of journals with “possible fractures” a fourth case with facet fractures was

identified. The histological findings of a facet fracture correlated with the exposure to trauma (p < 0.05)

(Table 3). Osteochondral fissures without blood were detected in all subjects (n = 40) affecting on average

59 % of the facets among the cases, and 57 % of the facets among the controls, whereas osteochondral

fissures with bleeding were identified in three cases (3/19) and no controls (p < 0.01) (Table 3). Blood in the

underlying bone was present in 32 subjects (15/19 cases and 17/21 controls). Bleeding in any joint space

(haemarthrosis) was present in nine cases (9/19) and one control subject (1/21) (Figure 2). There was

significant correlation between bleeding in any joint and exposure to trauma (p < 0.01), between bleeding in

any joint and the presence of a fracture on histological evaluation (p < 0.05) and between bleeding in any

joint and bleeding in both synovial folds at the same segment (p < 0.001).

Bleeding in any fold was present in 11 cases (11/19) and seven controls (7/21) and was not statistically

significant (p = 0.20), however simultaneous bleeding in both folds at any one segment was present in five

cases (5/19) and no controls (p < 0.05) (Figure 3). There was significant correlation between simultaneous

bleeding in both folds at one spinal segment and the presence of a fracture (p < 0.01). A total of 27 subjects

(15/19 cases and 12/21 controls) had a disrupted anterior fold and thirty-three subjects (18/19 cases and

15/21 controls) had a disrupted posterior fold with no significant correlation to trauma. However, disruption of

the posterior fold combined with bleeding in the same fold at the same level revealed five trauma cases

(5/19) and no controls (p < 0.02). Analysis of the incidence rates of disrupted folds per subject (using Chi-

squared test) revealed a risk ratio of 1.8 (95 % confidence interval: 1.5 – 2.2, p < 0.001) and 1.8 (95 %

confidence interval: 1.4 – 2.2, p < 0.001), for disruption of the anterior and posterior fold respectively with

regard to exposure to trauma (Table 3). Simultaneous disruption of both folds at any one segment was found

in 15 subjects (10/19 cases and 5/21 controls, p = 0.10), and if combined with bleeding in any of the same

two folds or both, seven cases (3/19) and no controls scored positive (p < 0.01).

There was no correlation between the stereomicroscopy and histological findings concerning damages to the

facet cartilage and/or bone or the synovial folds. It is worth noticing that none of the injuries to the facet joints

were observed during the autopsy.

Discussion

The detection of non-fatal injuries to the cervical spine facet joints was not reported equally among the

diagnostic methods used in this study. Based on conventional autopsy detailed description of fatal injuries

could be performed, however no injuries to the cervical spine facet joints were detected. Results from the

stereomicroscopic evaluation of the 3-mm slices did not reach statistically significant correlation between any

of the parameters evaluated. In contrast, microscopical evaluation of the 10 μm thick sections revealed

statistically significant correlations between exposure to trauma, and the presence of non-fatal articular facet

fractures, disruption of the folds, bleeding in the facet joints and simultaneous bleeding in the both folds.

In this study the most common cause of death in the trauma group was multiple injuries, and the high

incidence rates of skull fractures and organic injuries were all in concurrence with previously reported injuries

in road traffic crash fatalities15. Forensic toxicology revealed that four of the 18 drivers killed were under the

influence of alcohol during the time of the crash which is similar to previously reported figures of 20-30 % of

decedents from fatal road traffic crashes having BAC values above 0.5 mg/ml16-18. Similarly we found one

trauma case who tested positive for drugs which also supports previous reports of drugs in road traffic crash

fatalities17. The medicolegal autopsy did not identify injuries in the cervical spine facet joints despite

histological and neuroradiological evidence of injuries. However, this is likely due to the choice of autopsy

technique utilized (i.e. anterior approach) which does not allow detailed description of the posterior elements

of the cervical spine. Hence, the poor sensitivity towards identifying injuries in the facet joints is not

surprising and concurs with other studies reaching the same conclusions regarding standardised

autopsies7;9-11;14;19.

Several post-mortem studies of trauma fatalities have reported subjects with fractures of the lower cervical

spine facet joints similar to those identified in our study9-12;14;19-21. However, focusing on studies examining

road traffic victims, the 1/22 and 1/15 traffic accident victims sustaining lower cervical spine facet joint

fractures, examined in the studies by Jónsson et al12 and Taylor and Twomey10 respectively, is significantly

lower than our findings of facet fractures in 4/19 trauma cases. The differences between the reported figures

may be due to the primary focus on facet joint injuries in our study in contrast to the previous studies

examining the whole cervical spine. Furthermore, our material exclusively involved subjects killed in motor

vehicle crashes which indicate that all the decedents have been exposed to forces of significant magnitudes.

The low number of subjects examined in all the studies as well as technical issues such as the histological

methods utilized could influence the results. Osteochondral fissures without bleeding were non-specific

findings and in many cases likely due to artefacts caused by the microtome sectioning. In contrast, bleeding

was only identified in three subjects (3/19 cases) and consequently recorded as a fractures.

Haemarthrosis of the cervical spine facet joints have been described in post-mortem studies of road traffic

crash victims using histological and/or microtomal methods9-12;14;19;20. However, due to the nature of the

different methods used in these studies the presence of erythrocytes in the joint cavities have not been

clearly demonstrated in all studies, in particular not in studies utilising photographic documentation of the

remaining tissue bloc after removal of micrometer thick sections. In our study the erythrocytes were, after

and in comparison with blood vessel content, identified as pale cell ghosts in a honeycomb pattern. By this

definition, haemarthrosis was identified in almost half the trauma cases with statistical significance. Similarly,

a recent study investigated 10 consecutive trauma fatalities, with no reference to the proportion of traffic

crash victims in the material and the specific segments examined, and found haemarthrosis in the facet

joints in 6 of the 10 cases based on microscopy of 3-mm thick slices and in some cases histopathologic

examination19. In another post-mortem study haemarthrosis was present in six of 15 trauma cases, with only

three affecting the lower cervical spine facet joints below C410. However, the exact manner of identification of

bleeding within the joints was not described in detail in any of these studies and combined with our finding

that stereomicroscopic evaluation of 3-mm thick slices do not justify reliable conclusions regarding the

presence of bleeding in the facet joints we find the previously reported results questionable.

Although there were a higher number of subjects with disrupted folds in the trauma group than the control

group, statistical significance was only present when analysis of the individual number of disrupted folds per

subject was performed. This revealed that the posterior fold was at higher risk of injury during a fatal road

traffic crash compared to the anterior counterpart. Similarly, simultaneous bleeding in the anterior and

posterior fold was present in several trauma cases only which indicates significant trauma. No previous

publications have reported specific subject based incidence rates for the occurrences or locations of

bleeding in the folds, although the lesions have been reported9-14;19. The presence of erythrocytes in and the

disruption of synovial folds in a number of the control group subjects suggest that artefacts may be

introduced at post-mortem.

Despite good homogeneity (all cases were passenger car occupants, the age ranges were restricted to 20-

49 years and the controls were non-traumatised) and detailed multifaceted evaluation the study population,

the gold standard chosen did not have the possibility of identifying all lesions. Facet fractures parallel to the

line of sectioning would not necessarily appear on the histological sections and discrete fractures could

potentially be present inside the 3-mm slices but outside the area of microtomal sectioning. Furthermore,

prolonged ethanol fixation and staining of un-deplastified histological sections may have caused the poor

visualisation of erythrocytes in lesions sites and thereby underestimation of bleeding.

The findings of non-fatal disruption with or without bleeding in a significant number of folds suggest that such

injuries may be present in injured survivors from severe road traffic crashes possibly acting as a source of

pain mediated through pain pathways. We furthermore observed haemarthrosis which may have clinical

significance in the light of recent studies finding that even brief exposures to blood may have detrimental

long-lasting effect on the cartilage, by inhibiting the proteoglycan synthesis, thereby predisposing to

premature degeneration22-25. Acute synovitis after haemarthrosis and/or cartilage damage after trauma has

also been described in the literature22;25, which also may have clinical implications for development of neck

pain following adequate trauma. The possibility of degradation of cartilage matrix and subsequent

development of degenerative changes in articular structures may correlate with diagnostic imaging findings

of increasing osteoarthrosis in road traffic trauma patients years after the injury26-28, although opinions are

divergent27;29. The true incidence rates of bleeding in cervical spine facet joints and folds after survivable

road traffic crashes have not been reported and the potential clinical implications are inconclusive.

Conclusions

In this controlled study of motor vehicle crash fatalities, non-fatal injuries to the lower cervical spine facet

joints were a common finding, including facet fractures, haemarthrosis, disruption and bleeding in the

synovial folds. The majority of lesions were not present on stereomicroscopic evaluation, and conventional

autopsy procedures did not reveal any of the injuries. The utilisation of advanced imaging modalities as well

as specialized autopsy techniques should be encouraged in forensic settings in order to increase the

sensitivity towards identifying such discrete injuries. The clinical implications of the lesions identified in these

subjects are unknown, however, it is possible that discrete injuries similar to those identified in this study will

be present in a number of survivors from severe motor vehicle crashes.

References

1. European Road Safety Observatory (2007). Traffic safety basic facts 2006. Accessed at

http://www.erso.eu/safetynet on May 23, 2007.

2. European Road Safety Observatory (2006). Deliverable D4.2 Database Transparency. Accessed at

http://www.erso.eu/safetynet on May 27, 2007.

3. Cavanaugh JM, Lu Y, Chen C et al. Pain generation in lumbar and cervical facet joints. J Bone Joint

Surg.Am. 2006;88 Suppl 2:63-7.

4. Barnsley L. Percutaneous radiofrequency neurotomy for chronic neck pain: outcomes in a series of

consecutive patients. Pain Med. 2005;6:282-6.

5. Banic B, Petersen-Felix S, Andersen OK et al. Evidence for spinal cord hypersensitivity in chronic pain

after whiplash injury and in fibromyalgia. Pain 2004;107:7-15.

6. Curatolo M, rendt-Nielsen L, Petersen-Felix S. Central hypersensitivity in chronic pain: mechanisms

and clinical implications. Phys.Med.Rehabil.Clin.N.Am. 2006;17:287-302.

7. Uhrenholt L, Nielsen E, Vesterby Charles A et al. Imaging occult lesions in the cervical spine facet

joints. (Submitted) 2007.

8. Holmes JF, Mirvis SE, Panacek EA et al. Variability in computed tomography and magnetic resonance

imaging in patients with cervical spine injuries. J.Trauma 2002;53:524-9.

9. Uhrenholt L, Grunnet-Nilsson N, Hartvigsen J. Cervical spine lesions after road traffic accidents: a

systematic review. Spine. 2002;27:1934-41.


1993;18:1115-22.



12. Jonsson H, Jr., Bring G, Rauschning W et al. Hidden cervical spine injuries in traffic accident victims

with skull fractures. J.Spinal Disord. 1991;4:251-63.

13. Schonstrom N, Twomey L, Taylor J. The lateral atlanto-axial joints and their synovial folds: an in vitro

study of soft tissue injuries and fractures. J.Trauma 1993;35:886-92.

14. Uhrenholt L, Nielsen E, Vesterby A et al. Detailed examination of the lower cervical spine facet joints in

a road traffic crash fatality - a case study. IRCOBI (annual proceedings) 2005;paper 701;411-414.



16. Hansen AC, Kristensen IB, Dragsholt C et al. Alcohol and drugs (medical and illicit) in fatal road

accidents in a city of 300,000 inhabitants. Forensic Sci.Int. 1996;79:49-52.

17. Drummer OH, Gerostamoulos J, Batziris H et al. The incidence of drugs in drivers killed in Australian

road traffic crashes. Forensic Sci.Int. 2003;134:154-62.

18. European Road Safety Observatory (2006). Alcohol. Accessed at http://www.erso.eu/safetynet on

May 27, 2007.


comparison of radiographic, MR imaging, anatomic, and pathologic findings. Radiology 2001;221:340-

6.



21. Rauschning W, McAfee PC, Jonsson H, Jr. Pathoanatomical and surgical findings in cervical spinal

injuries. J.Spinal Disord. 1989;2:213-22.

22. Roosendaal G, Vianen ME, Marx JJ et al. Blood-induced joint damage: a human in vitro study. Arthritis

Rheum. 1999;42:1025-32.

23. Hooiveld M, Roosendaal G, Vianen M et al. Blood-induced joint damage: longterm effects in vitro and

in vivo. J Rheumatol. 2003;30:339-44.

24. Hooiveld M, Roosendaal G, Wenting M et al. Short-term exposure of cartilage to blood results in

chondrocyte apoptosis. Am.J Pathol. 2003;162:943-51.

25. Jansen NW, Roosendaal G, Bijlsma JW et al. Exposure of human cartilage tissue to low

concentrations of blood for a short period of time leads to prolonged cartilage damage: an in vitro

study. Arthritis Rheum. 2007;56:199-207.

26. Hohl M. Soft-tissue injuries of the neck in automobile accidents. Factors influencing prognosis. J Bone

Joint Surg.Am. 1974;56:1675-82.

27. Gore DR, Sepic SB, Gardner GM et al. Neck pain: a long-term follow-up of 205 patients. Spine

1987;12:1-5.

28. Watkinson A, Gargan MF, Bannister GC. Prognostic factors in soft tissue injuries of the cervical spine.

Injury 1991;22:307-9.

29. Robinson DD, Cassar-Pullicino VN. Acute neck sprain after road traffic accident: a long-term clinical

and radiological review. Injury 1993;24:79-82.

Figure 1 Fractures of the opposing cervical spine facets in a motor vehicle crash fatality

Overview of the joint, inferior articular process (IAP), superior articular process (SAP), joint space (JS)

and facet fracture (F), original magnification x1.25, Masson Goldner-Trichrome

Figure 2 Haemarthrosis in a cervical spine facet joint in a motor vehicle crash fatality

Large overview of the joint, inferior articular process (IAP), superior articular process (SAP), synovial fold

(SF), joint space (JS), original magnification x1.25, Masson Goldner-Trichrome, and a close-up of

shaded area illustrating bleeding in the joint space (B), original magnification x4, Masson Goldner-

Trichrome

Figure 3 Injuries to the cervical spine facet joint in a motor vehicle crash fatality

General overview of the joint, inferior articular process (IAP), superior articular process (SAP), synovial

fold (SF), joint space (JS), original magnification x1.25, Masson Goldner-Trichrome, and close-up of

shaded areas illustrating bleeding in the fold (BF) and joint (B), and an osteochondral fracture (F),

original magnification x4, Masson Goldner-Trichrome

Tabl

e 1

Det

aile

d de

scrip

tion

of th

e au

tops

y fin

ding

s

Subj

ect

Age

Gen

der

Circ

umst

ance

s Pr

imar

y ca

use

of d

eath

Sk

ull

frac

ture

C

ervi

cal s

pine

ve

rteb

ral f

ract

ure

Thor

acic

spi

ne,

lum

bar s

pine

and

/or

pelv

ic fr

actu

res

Extr

emity

frac

ture

s In

trac

rani

alor

gan

in

jurie

s Th

orac

ic o

rgan

in

jurie

s Ab

dom

inal

org

an

inju

ries

26

21

M

Pass

enge

r, R

TC

Mul

tiple

trau

mat

ic in

jurie

s Ye

s N

o Pe

lvis

H

umer

us, f

emur

Ye

s Ye

s Ye

s

24

22

F D

river

, RTC

M

ultip

le tr

aum

atic

inju

ries

Yes

No

No

ulna

, fem

ur a

nd ti

bia

Yes

Yes

Yes

35

33

M

Driv

er, R

TC

Blee

ding

, hae

mot

hora

x, ru

ptur

e th

orac

ic a

orta

N

o N

o N

o tib

ia, p

edis

N

o Ye

s Ye

s

14

20

M

Driv

er, R

TC

Mul

tiple

trau

mat

ic in

jurie

s N

o C

1, C

2 Th

3, T

h4

No

Yes

Yes

No

20

29

M

Driv

er, R

TC

Mul

tiple

trau

mat

ic in

jurie

s Ye

s N

o Pe

lvis

fe

mur

N

o Ye

s Ye

s

25

36

M

Driv

er, R

TC

Mul

tiple

trau

mat

ic in

jurie

s N

o C

1, C

6-7

No

fem

ur

Yes

Yes

Yes

42

29

M

Driv

er, R

TC

Blee

ding

, inj

urie

s to

the

hear

t and

aor

ta

No

No

No

fem

ur, t

ibia

, fib

ula

No

Yes

Yes

47

23

F D

river

, RTC

Bl

eedi

ng, m

ultip

le tr

aum

atic

inju

ries

No

No

No

hum

erus

N

o Ye

s Ye

s 21

31

F

Driv

er, R

TC

Cer

ebra

l con

tusi

on/m

edul

la o

blon

gata

lesi

ons

Yes

C0-

1,C

2-3

No

tibia

, fib

ula

Yes

Yes

Yes

13

34

M

Driv

er, R

TC

Blee

ding

, mul

tiple

trau

mat

ic in

jurie

s N

o N

o Pe

lvis

hu

mer

us, f

emur

, tib

ia, f

ibul

a Ye

s Ye

s Ye

s

17

45

F D

river

, RTC

H

aem

orrh

agic

med

iast

inum

/rupt

ure

vena

cav

a N

o N

o N

o tib

ia, p

edis

N

o Ye

s Ye

s

19

35

M

Driv

er, R

TC

Suba

rach

noid

hae

mor

rhag

e/ce

rebr

al c

ontu

sion

N

o N

o N

o ul

na, r

adiu

s, fe

mur

, tib

ia, f

ibul

a Ye

s N

o Ye

s

22

20

M

Driv

er, R

TC

Pulm

onar

y co

ntus

ion

and

lace

ratio

n N

o N

o Th

9 N

o N

o Ye

s Ye

s

36

29

M

Driv

er, R

TC

Mul

tiple

trau

mat

ic in

jurie

s Ye

s N

o Th

5-6

hum

erus

, uln

a, ti

bia,

ped

is

Yes

Yes

Yes

46

37

M

Driv

er, R

TC

Mul

tiple

trau

mat

ic in

jurie

s Ye

s C

1-2

Pelv

is

No

Yes

Yes

Yes

53

31

M

Driv

er, R

TC

Mul

tiple

trau

mat

ic in

jurie

s Ye

s N

o Th

12, p

elvi

s N

o N

o Ye

s Ye

s

39

41

M

Driv

er, R

TC

Hae

mot

hora

x, a

ortic

rupt

ure

N

o N

o Pe

lvis

tib

ia

Yes

Yes

Yes

34

47

M

Driv

er, R

TC

Blee

ding

and

aor

tic la

cera

tions

N

o N

o Th

11, T

h12

No

No

Yes

No

28

38

M

Driv

er, R

TC

Blee

ding

and

mul

tiple

trau

mat

ic in

jurie

s Ye

s N

o N

o fe

mur

, cal

cane

us

Yes

Yes

Yes

19 c

ases

8

4 10

14

11

18

17

16

39

F Vi

tal a

ctiv

ity

Myo

card

ial i

nfar

ct

No

No

No

No

No

No

No

48

46

F H

ypot

herm

ia

Hyp

othe

rmia

and

into

xica

tion

(alc

ohol

) N

o N

o N

o N

o N

o N

o N

o

12

41

M

Vita

l act

ivity

M

yoca

rdia

l inf

arct

N

o N

o N

o N

o N

o N

o N

o

29

34

F Vi

tal a

ctiv

ity

Unk

now

n, in

toxi

catio

n (a

lcoh

ol)

No

No

No

No

No

No

No

43

33

M

Oth

er a

ctiv

ity

Blee

ding

due

to c

uttin

g in

jury

N

o N

o N

o N

o N

o N

o N

o

30

43

F Su

rger

y C

ardi

ac d

ysfu

nctio

n se

cond

ary

to s

urge

ry

No

No

No

No

No

No

No

38

49

M

Vita

l act

ivity

Ar

terio

scle

rotic

car

diac

dis

ease

N

o N

o N

o N

o N

o N

o N

o 18

37

M

Vi

tal a

ctiv

ity

Epile

ptic

sei

zure

/sub

arac

hnoi

d in

flam

mat

ion

No

No

No

No

No

No

No

45

35

M

Unk

now

n

Dia

betic

com

a N

o N

o N

o N

o N

o N

o N

o

51

23

M

Vita

l act

ivity

D

iabe

tic c

oma

No

No

No

No

No

No

No

15

32

M

Leis

ure

activ

ity

Unk

own

No

No

No

No

No

No

No

23

37

M

Oth

er a

ctiv

ity

Car

diac

dis

ease

N

o N

o N

o N

o N

o N

o N

o

31

31

M

Vita

l act

ivity

D

ehyd

ratio

n, p

erito

nitis

, par

alyt

ic s

mal

l bow

el

No

No

No

No

No

No

No

44

40

M

Oth

er a

ctiv

ity

Car

diac

dis

ease

N

o N

o N

o N

o N

o N

o N

o

52

48

M

Wor

king

Ar

terio

scle

rotic

car

diac

dis

ease

N

o N

o N

o N

o N

o N

o N

o

50

41

F Vi

tal a

ctiv

ity

Suba

rach

noid

hae

mor

rhag

e/ce

rebr

al a

neur

ysm

N

o N

o N

o N

o N

o N

o N

o

49

41

F Vi

tal a

ctiv

ity

Car

diac

dis

ease

N

o N

o N

o N

o N

o N

o N

o

40

27

F U

nkno

wn

Car

diac

dis

ease

, dia

bete

s m

ellit

us, i

ntox

icat

ion

No

No

No

No

No

No

No

54

22

M

Vita

l act

ivity

Ao

rtic

aneu

rysm

rupt

ure

and

card

iac

dise

ase

No

No

No

No

No

No

No

37

49

F Vi

tal a

ctiv

ity

Myo

card

ial i

nfar

ct

No

No

No

No

No

No

No

33

35

M

Vita

l act

ivity

Ar

terio

scle

rotic

car

diac

dis

ease

N

o N

o N

o N

o N

o N

o N

o 21

con

trols

0 0

0 0

0 0

0

Gen

der:

M (

mal

e), F

(fem

ale)

R

CT:

road

traf

fic c

rash

Vi

tal a

ctiv

ity: s

leep

, res

t

page 19

Table 2 Detailed description of the toxicological findings

Subject Age Gender Blood alcohol

concentration (‰) Drugs/ medicine

in blood/urine 26 21 M 2,08 No 24 22 F 0,00 No 35 33 M 0 No 14 20 M 0,88 No 20 29 M 0,10 Not tested 25 36 M 0,90 Metoprolol 42 29 M 0 Not tested 47 23 F 0 Not tested 21 31 F 0 No 13 34 M 0 Ketamine¥ 17 45 F 0 Not tested 19 35 M 1,89 Not tested 22 20 M 0 Valproate 36 29 M 0 Not tested 46 37 M 0 Not tested 53 31 M 0,22 Not tested 39 41 M (2,40)± Not tested 34 47 M 0 Citalopram 28 38 M 0 Not tested 19 cases 5 (BAC>0.5) 4

16 39 F 0 Not tested 48 46 F 1,82 No 12 41 M 0,78 Not tested 29 34 F 1,80 Cannabis¥ 43 33 M 0 Not tested

30 43 F 0 Clomipramine &

paracetamol† 38 49 M 0 Not tested 18 37 M 0 No 45 35 M 0 No 51 23 M 0 Not tested 15 32 M 0 No 23 37 M 0 Not tested 31 31 M 0 Not tested 44 40 M 0,28 Not tested 52 48 M 0 Not tested 50 41 F 0 Not tested 49 41 F 0 No 40 27 F 0 Cocaine¥ 54 22 M 0 No 37 49 F Not tested Not tested 33 35 M 0 No 21 controls 3 (BAC>0.5) 3

BAC: Blood alcohol concentration Gender: M (male), F (female) ± conc. in eyefluid (vitrous humor) † conc. of intoxication (overdose) ¥ narcotics/drugs

page 20

Tabl

e 3

Det

aile

d de

scrip

tion

of th

e m

icro

scop

ical

find

ings

on

the

10 μ

m s

ectio

ns

Subj

ect

Age

Gen

der

Frac

ture

of

a fa

cet

OC

F w

ith

bloo

d (0

-16

face

t)

OC

F w

ithou

t bl

ood

(0-1

6 fa

cet)

Blo

od in

any

jo

int s

pace

B

lood

in

any

fold

* B

lood

in

both

fold

s±

Ant

erio

r fol

d di

srup

ted

(0

-8 jo

ints

)

Post

erio

r fol

d di

srup

ted

(0

-8 jo

ints

)

Blo

od in

the

unde

rlyin

g bo

ne

26

21

M

No

0 10

N

o N

o N

o 3

7 Ye

s

24

22

F N

o 0

10

No

Yes

N

o 0

1 Ye

s

35

33

M

No

0 16

N

o N

o N

o 1

4 N

o

14

20

M

No

0 5

Yes

Yes

No

3 6

Yes

20

29

M

No

0 9

No

No

No

0 0

No

25

36

M

Yes

4 7

Yes

Yes

Yes

8 3

Yes

42

29

M

No

0 6

No

No

No

0 1

Yes

47

23

F N

o 0

10

Yes

Yes

Yes

3 5

Yes

21

31

F N

o 0

6 N

o Ye

s no

7

6 Ye

s

13

34

M

No

0 6

No

Yes

No

2 8

Yes

17

45

F N

o 0

8 N

o N

o N

o 0

2 Ye

s

19

35

M

No

0 9

Yes

No

No

1 4

Yes

22

20

M

Yes

1 13

Ye

s Ye

s Ye

s 8

7 Ye

s

36

29

M

No

0 10

Y

es

Yes

N

o 5

7 Ye

s

46

37

M

No

0 10

Y

es

Yes

N

o 3

2 N

o

53

31

M

Yes

0 13

N

o N

o N

o 3

7 N

o

39

41

M

Yes

1 15

Ye

s Ye

s Ye

s 6

5 Ye

s

34

47

M

No

0 10

Ye

s Ye

s Ye

s 1

3 Ye

s

28

38

M

No

0 8

No

No

No

2 5

Yes

19 c

ases

4

(p<0

.05)

6

(p<0

.01)

18

1 (5

9.2%

) (n

.s.)

9 (p

<0.0

1)

11

(n.s

.) 5

(p<0

.05)

56

(37.

3%‡ )

(p<0

.01)

83

(55.

3%‡ )

(p<0

.01)

15

(n

.s.)

16

39

F N

o 0

8 N

o N

o N

o 2

1 Ye

s

48

46

F N

o 0

9 N

o Ye

s N

o 2

0 Ye

s

12

41

M

No

0 13

N

o N

o N

o 0

3 N

o

29

34

F N

o 0

3 N

o N

o N

o 0

0 Ye

s

43

33

M

No

0 11

N

o N

o N

o 0

2 Ye

s

30

43

F N

o 0

8 N

o N

o N

o 0

3 N

o

38

49

M

No

0 13

N

o N

o N

o 0

1 Ye

s

18

37

M

No

0 10

N

o N

o N

o 0

0 Ye

s

45

35

M

No

0 13

N

o N

o N

o 1

6 Ye

s

51

23

M

No

0 13

N

o N

o N

o 2

6 Ye

s

15

32

M

No

0 10

N

o Ye

s N

o 0

2 Ye

s

23

37

M

No

0 9

No

No

No

1 7

Yes

31

31

M

No

0 11

N

o N

o N

o 1

0 Ye

s

44

40

M

No

0 7

No

Yes

No

2 2

No

52

48

M

No

0 7

No

No

No

0 1

Yes

50

41

F N

o 0

8 N

o Ye

s N

o 8

0 Ye

s

49

41

F N

o 0

7 N

o Ye

s N

o 1

6 Ye

s

40

27

F N

o 0

5 N

o Ye

s N

o 1

0 Ye

s

54

22

M

No

0 6

No

No

No

1 6

Yes

37

49

F N

o 0

7 Ye

s Ye

s N

o 0

1 Ye

s

33

35

M

No

0 13

N

o N

o N

o 1

1 N

o

21 c

ontro

ls

0

0 19

1 (5

6.8%

) of

all f

acet

s 1

7 0

23 (1

3.7%

‡ ) of

all m

enis

ci

48 (2

8.6%

‡ ) of

all m

enis

ci

17

Gen

der:

M (m

ale)

, F (f

emal

e)

OC

F: o

steo

chon

dral

fiss

ure

n.s.

: not

sta

tistic

ally

sig

nific

ant

* bl

eedi

ng in

the

ante

rior a

nd/o

r pos

terio

r fol

ds a

t any

seg

men

t per

sub

ject

±

sim

ulta

neou

s bl

eedi

ng in

bot

h fo

lds

at a

ny s

egm

ent p

er s

ubje

ct

‡ p

erce

ntag

e of

all

fold

s in

the

grou

ps (t

wo

fold

s ar

e no

n-ex

istin

g du

e to

a b

loc

verte

brae

in o

ne tr

aum

a ca

se)

page 21

Histomorphology of the lower cervical spine facet joints

Lars Uhrenholt1,2,3, Ellen Hauge3,4, Annie Vesterby Charles1,3, Markil Gregersen1



Denmark


Denmark



Lars Uhrenholt, Institute of Forensic Medicine, University of Aarhus


E-mail: [email protected], facsimile: 45-86125995, phone: 45-89429800

PAPER III

Abstract

Objectives

The purpose of this study was to examine anatomical variables of the lower cervical spine facet joints with

regard to age, gender, and exposure to trauma.

Methods

The lower four cervical spine motion segments (C4-C7 included) were obtained from forty subjects during

autopsy, twelve females (median age 40 years, range 22–49) and 28 males (median age 34.5 years, range

20–49 years). Ten μm thick histological sections were produced from 3-mm thick parasaggital slices going

through the facet joints bilaterally and evaluated with microscopy at random locations. Inter- and

intraobserver agreement was tested on four randomly selected facets from each subject.

Results

Significant age-, gender and trauma related changes in the cartilage and soft tissues were observed.

Females were less affected by changes in the cartilage than males. Anterior ad posterior synovial folds were

present in all but one joint. Moderate interobserver and good intraobserver agreement were achieved.

Conclusions

This study provides knowledge of the anatomy of the cervical spine facet joints. The findings support the

existing knowledge that males are more commonly affected by degenerative changes than females and that

these are common from young age. Histomorphometry confirms the presence of synovial folds throughout

the facet joints. Following spinal trauma pathological lesions may be produced in the facet joints and/or

accentuate already existing pathology. This study provides knowledge of the anatomy of the cervical spine

facet joints that may have relevance for patients with neck pain.

Keywords

cervical spine, facet joint, zygapophysial joint, histology, histomorphometry, autopsy, osteoarthritis, age,

gender

Introduction

The clinical relevance of age-related changes and osteoarthritis (OA) in the cervical spine facet joints is not

completely understood and the exact relation to neck pain is unknown, beyond the general fact that

increasing severity of OA is related to stiffness, decreased mobility and pain (1). The prevalence of early

degenerative changes such as superficial flaking (fibrillation) and fissuring of the articular hyaline cartilage,

vascular invasion of the tidemark and sclerosis of the subchondral bone plate is not known for the cervical

spine facet joints (1-3). Although numerous studies have investigated the cervical spine morphology

including the properties of the articular facets and related soft tissues (4-9) only limited knowledge is

available concerning studies of age-related cartilage and subchondral bony changes of human subjects (6).

The synovial folds (fat pads, meniscoids and inclusions) of the anterior and posterior margins of the lower

cervical spine facet joints have been described inconsistently in anatomical studies (4;10-16) and the joint

capsules of the cervical spine facet joints have been described in some detail (15). Histomorphometrical

studies have described the cervical spine anatomy, however in comparison to larger joints only limited

quantitative data currently exist regarding the properties of the cartilage, calcified cartilage and subchondral

bone of the cervical spine articular facets. (17-24). The general microscopical appearance of human synovial

joints, including the cervical spine facet joints, has been described in some detail (4-7) and general

histological classification systems are available for the description of these joints (2;6;16;25-28). However,

no validated system for microscopical grading of age-related changes in the cervical spine facet joints is

available, although a recent system originally developed for the lumbar spine facet joints has been utilised in

a preliminary evaluation of the upper cervical spine facets (16;27). Neck pain is an extremely common

symptom with high prevalence rates in Scandinavian countries compared to the rest of Europe and Asia

(29). Chronic neck pain following relatively minor trauma, e.g. low speed rear-impact collisions, is particularly

common and the lower cervical spine facet joints have been identified as a potential source of pain (30-32),

and nociceptive fibers have been identified in these structures (33;34). These findings have been supported

by nerve bloc and radio frequency neurotomy studies providing relief of symptoms after ablation of the

segmental nerve supply to the facet joints (35-37).

The purpose of this study was to describe the histomorphology of the lower cervical spine facet joints using

quantitative histomorphometric methods and to compare finding to age, gender and trauma.


Materials

The lower cervical spines were obtained from forty subjects, twelve females (median age 40 years, range

22–49) and 28 males (median age 34.5 years, range 20–49 years). Nineteen died in a passenger car crash

and 21 died due to non-traumatic causes. Each specimen consisted of the cervical spine segments C4 to C7

including the facet joints bilaterally from C4-C5 to C7-Th1. In 39 subjects there were eight facet joints

available from each subject and in the remaining one subject there were only six facet joints available for

evaluation due to a bloc vertebrae at the C6-C7 level. Hence, the total number of facet joints available for

evaluation was 318, with 636 unique facets. The study was approved by the Scientific Ethics Committee,

Central Denmark Region.

Laboratory methods

The methods have previously been described (38). Each specimen was fixated in 70% ethanol followed by

hemisection along the midline saggital plane and embedded in methylmethacrylate (MMA) from where 3-mm

thick parasaggital slices were produced. All slices containing facet joint structures were re-embedded in

MMA and heavy-duty microtomal sections of 10 μm thickness were produced from each final bloc. These

histological sections were mounted on glass and stained un-deplastified with Masson Goldner-Trichrome

and examined with light microscopy using a BX51 Olympus© microscope.

Microscopical evaluation

A total of 1830 unique observations (approximately 46 observations per subject) were made, each with

reference to morphological variables including cartilage flaking (fibrillation), cartilage fissures, cartilage split,

vascular invasion, osteophytes, presence of the anterior and posterior folds and the integrity of these folds

(Table 1). Furthermore, included were observations obtained by measurements of projected images from a

BH2 Olympus© microscope with objectives x1 or x4, through a projection arm to a measuring table with a

final magnification factor on the table of 19.25 or 77.00 respectively (Table 1). This included the cartilage

length, the degree of overlap of the anterior and posterior folds in relation to the hyaline cartilage covering of

the facet, the thickness of the hyaline cartilage, the calcified cartilage and the subchondral bone (Figure 1).

The thicknesses were recorded of the facets from each histological section at one to five randomly chosen

points depending on the anterior to posterior length of the facet surface. The total articular cartilage

thickness was calculated as the sum of the calcified and hyaline cartilage thickness. All microscopical

evaluations were performed blinded with regard to the characteristics of the subjects (i.e. age, gender,

exposure to trauma, side, segment and level of the facet). For the purpose of inter- and intraobserver

agreement testing the morphological variables of four randomly chosen facets from each subject were

evaluated independently by two examiners who were blinded with regard to the characteristics of the

subjects. Hence, a total of 160 facets were eligible for this evaluation, however only 159 unique facets were

evaluated as the remaining one facet did not contain articular structures.

Statistical methods

The data were analysed with a linear regression model assuming linear association, normal distribution and

independence (homoskedasticity) of residuals. The regression model used for the morphological data

examined for correlation between gender, age and trauma versus no trauma and a reference person was

defined as a 35 years old male who had been killed in a road traffic crash (Table 2). The regression model

used for the histomorphometric data examined for correlation between gender, age, trauma versus no

trauma, side and facet per segment and a reference person (level) was defined as the left C4 inferior facet of

a 35 years old male who had been killed in a road traffic crash (Table 3). The potential effect of interaction

was not tested due to the limited number of subjects (n=40). The morphological variables were analysed for

inter- and intraobserver agreement using kappa statistics (39). The significance level was p < 0.05. Results

including 95 % confidence intervals were presented in square brackets, e.g. 75 μm [69-80]. All statistical

analyses were performed using Stata© 9 (StataCorp LP, College Station, USA).

Results

Descriptive morphology

Significant correlations to gender were predicted with females having 18 % [7–29] fewer facets with flaking

(p<0.01), 18 % [9–27] fewer facets with split (p<0.001), 13 % [1-26] fewer fissures overall (p<0.05) and 8 %

[1-15] fewer moderate fissures (p<0.05) than the male reference person (Figure 2) (Table 2). Significant

correlations to age were predicted with each decade contributing with an increase of 13 % [7–20] of facets

with flaking (p<0.001), 8 % [3-13] of facets with split (p<0.01) and 2 % [1-4] of facets with osteophytes

(p<0.01), and a decrease in the number of facets with a disrupted posterior fold by 10 % [3-18] (p<0.05). The

presence of split was significantly correlated to trauma with 10 % [1-18] more facets with split in the

traumatised group (p<0.05). Similarly, the presence of disrupted anterior and/or posterior folds was

associated with trauma, with 11 % [1-21] and 15 % [2-27] more folds being disrupted anteriorly and

posteriorly in the subject killed in a road traffic crash (p<0.05). No significant correlations were found for

vascular invasion, mild fissures and severe fissures with regard to any of the independent variables. The

presence of fissures did not correlate significantly with age.

Presence of the anterior and posterior folds

Intra-articular synovial folds were identified in all subjects (n=40), with anterior folds present in 100%

(318/318 anterior folds) and posterior folds present in 99.7% (317/318 posterior folds) of the respective folds

(Figure 3). In the remaining one joint (1/318) without identification of a posterior fold no evaluation could be

performed due to poor quality of the histological section.

Histomorphometry

Only the maximum cartilage length showed significant correlation to gender with a predicted 1.33 mm [0.46-

2.19] shorter cartilage in the female subjects (p<0.01) (Figure 4) (Table 3). There was a significant

correlation between age and an increase in the subchondral bone thickness of 0.04 mm [0.01-0.06] per

decade (p<0.01) approximately equivalent to an increase of 10% in comparison to the original thickness.

Significant systematic variations were generally observed between the cervical spine segments (i.e. C4-Th1)

(Figure 4 & 5). No statistically significant correlations were observed between; the thickness of the hyaline

cartilage, the calcified cartilage, the calcified cartilage percentage of the total cartilage, the anterior and

posterior fold overlap versus gender or age (Table 3). In none of these parameters were a significant

differences present between subjects exposed to trauma versus no trauma. The side was generally not

significant, except for minor differences in the subchondral bone thickness, maximum cartilage length and

anterior fold overlap.

Inter- and intraobserver agreement

The results from the inter- and intraobserver agreement evaluation are tabulated in detail in Table 4.

Regarding the interobserver agreement several variables had statistically significant kappa scores (p<0.001);

fissures (k=0.65), flaking (k=0.54), posterior fold disruption (k=0.51), split (k=0.31), and vascular invasion

(k=0.29). There were no statistically significant kappa scores for anterior fold disruption (k=0.09) and

osteophytes (k=-0.04), with (p=0.156) and (p=0.688) respectively. With regard to the intraobserver

agreement all variables had statistically significant kappa scores (p<0.001). The kappa scores were; fissures

(k=0.84), flaking (k=0.83), posterior fold disruption (k=0.74), split (k=0.79), vascular invasion (k=0.70),

anterior fold disruption (k=0.75), and osteophytes (k=0.65).

Discussion

This study of young individuals revealed significant relationship between age and flaking, split and

osteophytes. Age was significantly related to an increase in the subchondral bone thickness, however no

age-related differences were observed concerning the thickness of the hyaline and calcified cartilage.

Females had significantly fewer facets with flaking, splitting and fissures and the cartilage length was shorter

in comparison to males. The exposure to trauma was significantly correlated to cartilage split and disruption

of the folds. Flaking was the most common finding followed by split of the cartilage and the presence of

fissures of any severity. The mean overlap of the folds in relation to the underlying cartilage was almost

equal anteriorly and posteriorly with no significant differences with regard to gender or age.

Age-related changes

It is noteworthy that the age-range in this study was limited to 20-49 years which reduced the expected

alterations in age-dependent changes. Despite this we found, in accordance with previous studies,

significant correlation between age and several morphological variables (flaking, split and osteophytes) that

are all common in elderly subjects and related to OA (1;6;28). Similarly, the subchondral bone thickness was

correlating significantly with age, increasing approximately 10 % per decade which is in agreement with

previous observations (1;3;40). Other indicators of OA, such as osteophytes, duplication of the tidemark and

vascular invasion of the tidemark, were only present in limited numbers which is explained by the median

age (35 years) of the entire study population. Nonetheless, cartilage flaking, split and fissures were very

common in all ages in agreement with other studies (6;27). Fissures were common but did not reach

significant correlation with age although they are commonly encountered in OA and elderly subjects (1;28). It

was surprising to observe a significant decrease in the number of disrupted posterior folds with increasing

age and it can be speculated whether these observations occurred by chance or were consequences of

technical preparation of the histological sections.

Gender

The overall finding of significant correlation between gender and histological findings seen in OA (e.g.

cartilage flaking and splitting) is in agreement with other reports of males having a preponderance for

pathological facet joints (1;6). Similarly, the cartilage length was significantly shorter in females which has

also been reported previously (20;24).

Trauma

In a related study (unpublished observations) of the same material discrete fractures of the facets, bleeding

in the joints spaces and bleeding in the synovial folds were significantly related to the exposure to fatal road

traffic crash trauma. Disruption of both the anterior and posterior folds correlated significantly to trauma as

did cartilage split contributing to previous findings. The exact clinical relevance of these pathological findings

is uncertain. However, due to widespread individual biological variation in injury thresholds for these

variables it can be conjectures that some survivors from serious road traffic crashes may suffer similar

pathology and contribute to symptomatology (unpublished observations)(41). Furthermore, the lower cervical

spine facet joints have been found vulnerable to relatively minor acceleration forces which may cause

potentially injurious pathophysiological kinematics of these joints (42-44). As several post-mortem studies of

trauma victims have identified subtle lesions in the cervical spine facet joint capsules and synovial folds,

including disruption and bleeding, these structures are likely to have clinical importance in selected cases

(41;45).

Histology

The mean articular cartilage thickness of the cervical spine facet joint (C4 inferior process) in a reference

person was approximately 0.8 mm. The thickness varied depending on the spinal level indicating anatomical

differences in geometry reflective of developmental adaptations to different functional requirements. It was

interesting to observe that the calcified cartilage occupied approximately 10% percent of the complete

cartilage thickness which is somewhat higher than previous studies of other joints reporting values of

approximately 5% (3-8%) (3;40;46). The finding of synovial folds in all joints supports a number of previous

publications (4;14-16;47), although other studies have not been able to identify folds consistently (10;48).

Fletcher et al (48), examined 20 cadavers (two aged 10 and 19 years, and the remaining 18 subjects

between the age of 37-86 years) with photographic documentation of cryomicrotomal sections and

haematoxylin-eosin staining of decalcified sections and concluded that the menisci [folds] were nonexistent

at the age of 37 years and older. In the study by Yu et al (10), 10 decedents (mean age 48.3 years, range

10-69 years) were examined with photographic evaluation of cryomicrotomal sections and four types of

menisci [folds] were proposed despite the low number of subjects and the authors concluded that in the

lower cervical spine of adults, no menisci [folds] were present within the joint and the articular surfaces. In

contrast we found anterior and posterior synovial folds in all joints (except one), irrespective of age and

gender. The mean overlap of the anterior folds (16.1%) and posterior folds (16.9%) expressed as a

percentage of the underlying cartilage illustrates the overall overlap. However, in several parasaggital

sections of the facets we observed that the synovial fold covered the articular surface completely which

indicates that the fold have a heterogenous covering of the articular surface that is continuous over some

larger areas (Figure 6). It can be speculated that synovial folds in the cervical spine facet joints have a

protective and lubricative function as that of menisci (15;49).

Hyaline cartilage was found to cover the opposing joint surfaces in the majority of the joints. Occasionally, a

transitional zone was observed at the borders of the cartilage covering of the facets in which a gradual

replacement of the hyaline cartilage by fibrous cartilage took place. Hence, only rarely were free bony edges

encountered with periosteal covering, previously described as “cartilage gaps”(5). The discrepancies

between these findings may be explained by the different populations examined, where the previous study

only included six elderly subjects in contrast to 40 subjects in the age 20-29 years, and the orientation of

sectioning (although both methods used parasaggital orientation). Furthermore, we believe that detailed

morphology is best visualised by microscopy. Hence, in our study we find no support for the reported

findings of cartilage gaps and consequently no support of the pathomechanical theory of bony impact during

trauma as a cause of persistent neck pain.

Methodology

The overall interobserver agreement was moderate and was probably influenced by improperly defined

variables as well as different levels of experience between the observers. The overall intraobserver

agreement was good, and for all variables better than the interobserver agreement. This study used a similar

direction of sectioning (i.e. parasaggital) as related studies increasing the comparability (5;16;48;50-54).

However, despite randomisation of the locations of measurements the effect of orientation is unclear.

Preferably, a stereological unbiased evaluation using random sampling and orientation should be used in

order to examine a larger population with a wider age-range for the extent of age-related changes. The

potential effect of tissue shrinkage was not evaluated in this study. The method used in this study did not

allow evaluation of the overall severity of the morphological variables as only the worst case finding was

recorded for each observation. Ideally, the overall severity should be evaluated by considering the extent of

tissue affected (28).

Grading and staging of age-related and OA changes in the cervical spine facet joints may be performed with

the method of Pritzker et al (28). An alternative scoring system for cervical spine facet joint OA has been

published (16). However, as the weighted contributions from the variables were accumulated potential

differences in pathological severity between individuals were reduced. Hence, an adjusted classification

system based on the Pritzker et al (28) model targeting the cervical spine facet joints specifically, similar to

the methods used in this study, seems to be the most ideal solution.

Conclusion

Degenerative changes occur early in life in the lower cervical spine facet joints in both the cartilage and

subchondral bone, and have significant correlation to age. Males are more frequently and severely affected

by degenerative changes in the facet joints than females. Synovial folds are consistently found anteriorly and

posteriorly in the facet joints suggesting a yet unproven biological function. Spinal trauma produces

microscopical injuries to the soft tissues and cartilage in the facet joints, and may accentuate already existing

degenerative changes. It can be speculated that similar injuries may cause pain in survivors after road traffic

crashes.

Acknowledgements

The authors would like to acknowledge the original contributions from Professor DMSc. Flemming Melsen†,

the professional cooperation of the staff at the Institute of Forensic Medicine, University of Aarhus and the

Knoglelaboratoriet, Aarhus University Hospital. The strenuous and time-consuming efforts of Rita Ullerup in

careful preparation of the histological sections and the statistical counselling from Morten Frydenberg are

greatly appreciated. The study received support from the European Chiropractors Union Research Fund

Grant no. A.03-5, Switzerland, the Foundation for the Advancement of Chiropractic Research, Denmark and

the University of Aarhus Research Foundation, Denmark.

References

(1) Brandt KD, Doherty M, Lohmander LS, eds. Osteoarthritis. 2nd. ed. Oxford, New York: Oxford

University Press Inc., 2003.

(2) Collins DH. Osteoarthritis. In: The pathology of articular and spinal diseases. 1949:74-115.

(3) Oegema TR, Jr., Carpenter RJ, Hofmeister F, Thompson RC, Jr. The interaction of the zone of

calcified cartilage and subchondral bone in osteoarthritis. Microsc Res Tech 1997;37:324-32.

(4) Bland JH, Boushey DR. Anatomy and physiology of the cervical spine. Semin Arthritis Rheum

1990;20:1-20.

(5) Yoganandan N, Knowles SA, Maiman DJ, Pintar FA. Anatomic study of the morphology of human

cervical facet joint. Spine 2003;28:2317-23.

(6) Sager P. Spondylosis cervicalis - a pathological and osteoarchaeological study. Thesis, University of

Copenhagen, Denmark, 1969.

(7) Hirsch C, Schajowicz R, Galante J. Structural changes in the cervical spine. A study on autopsy

specimens in different age groups. Acta Orthop.Scand 1967 Suppl 109, 7-77.

(8) Boyle JJ, Singer KP, Milne N. Morphological survey of the cervicothoracic junctional region. Spine

1996;21:544-8.

(9) Payne EE, Spillane JD. The cervical spine; an anatomico-pathological study of 70 specimens (using

a special technique) with particular reference to the problem of cervical spondylosis. Brain

1957;80:571-96.

(10) Yu SW, Sether L, Haughton VM. Facet joint menisci of the cervical spine: correlative MR imaging

and cryomicrotomy study. Radiology 1987;164:79-82.

(11) Inami S, Kaneoka K, Hayashi K, Ochiai N. Types of synovial fold in the cervical facet joint. J Orthop

Sci 2000;5:475-80.

(12) Mercer S, Bogduk N. Intra-articular inclusions of the cervical synovial joints. Br J Rheumatol

1993;32:705-10.

(13) Fletcher G, Haughton VM, Ho KC, Yu SW. Age-related changes in the cervical facet joints: studies

with cryomicrotomy, MR, and CT. AJR Am J Roentgenol 1990;154:817-20.

(14) Kos J, Hert J, Sevcik P. [Meniscoids of the intervertebral joints]. Acta Chir Orthop Traumatol Cech

2002;69:149-57.

(15) Taylor J, Twomey L, Levander B. Contrasts between cervical and lumbar motion segments. Crit Rev

Phys Rehab Med 2000;12:345-71.

(16) Berlemann U, Laubli R, Moore RJ. Degeneration of the atlanto-axial joints: a histological study of 9

cases. Acta Orthop Scand 2002;73:130-3.

(17) Ebraheim NA, Xu R, Challgren E, Yeasting RA. Quantitative anatomy of the cervical facet and the

posterior projection of its inferior facet. J Spinal Disord 1997;10:308-16.

(18) Ebraheim NA, Patil V, Liu J, Haman SP, Yeasting RA. Morphometric analyses of the cervical

superior facets and implications for facet dislocation. Int Orthop 2006 [Epub ahead of print].

(19) Panjabi MM, Duranceau J, Goel V, Oxland T, Takata K. Cervical human vertebrae. Quantitative

three-dimensional anatomy of the middle and lower regions. Spine 1991;16:861-9.

(20) Panjabi MM, Oxland T, Takata K, Goel V, Duranceau J, Krag M. Articular facets of the human spine.

Quantitative three-dimensional anatomy. Spine 1993;18:1298-310.

(21) An HS, Wise JJ, Xu R. Anatomy of the cervicothoracic junction: a study of cadaveric dissection,

cryomicrotomy, and magnetic resonance imaging. J Spinal Disord 1999;12:519-25.

(22) Pal GP, Routal RV, Saggu SK. The orientation of the articular facets of the zygapophyseal joints at

the cervical and upper thoracic region. J Anat 2001;198:431-41.

(23) Panjabi MM, Shin EK, Chen NC, Wang JL. Internal morphology of human cervical pedicles. Spine

2000;25:1197-205.

(24) Francis CC. Variations in the articular facets of the cervical vertebrae. Anat Rec 1955;122:589-602.

(25) Laeubli RT, Moore RJ, Berlemann U. Histologic degeneration of the aging cervical spine.

International Cartilage Repair Society annual conference 1999, poster # 1.

(26) Kettler A, Werner K, Wilke HJ. Morphological changes of cervical facet joints in elderly individuals.

Eur Spine J 2007;16:987-92.

(27) Gries NC, Berlemann U, Moore RJ, Vernon-Roberts B. Early histologic changes in lower lumbar

discs and facet joints and their correlation. Eur Spine J 2000;9:23-9.

(28) Pritzker KP, Gay S, Jimenez SA, Ostergaard K, Pelletier JP, Revell PA, et al. Osteoarthritis cartilage

histopathology: grading and staging. Osteoarthritis Cartilage 2006;14:13-29.

(29) Fejer R, Kyvik KO, Hartvigsen J. The prevalence of neck pain in the world population: a systematic

critical review of the literature. Eur Spine J 2006;15:834-48.

(30) Barnsley L, Lord SM, Wallis BJ, Bogduk N. The prevalence of chronic cervical zygapophysial joint

pain after whiplash. Spine 1995;20:20-5.

(31) Manchikanti L, Boswell MV, Singh V, Pampati V, Damron KS, Beyer CD. Prevalence of facet joint

pain in chronic spinal pain of cervical, thoracic, and lumbar regions. BMC Musculoskelet Disord

2004;5:15.

(32) Cavanaugh JM, Lu Y, Chen C, Kallakuri S. Pain generation in lumbar and cervical facet joints. J

Bone Joint Surg Am 2006;88 Suppl 2:63-7.

(33) Inami S, Shiga T, Tsujino A, Yabuki T, Okado N, Ochiai N. Immunohistochemical demonstration of

nerve fibers in the synovial fold of the human cervical facet joint. J Orthop Res 2001;19:593-6.

(34) McLain RF. Mechanoreceptor endings in human cervical facet joints. Spine 1994;19:495-501.

(35) Sapir DA, Gorup JM. Radiofrequency medial branch neurotomy in litigant and nonlitigant patients

with cervical whiplash: a prospective study. Spine 2001;26:E268-E273.

(36) Barnsley L. Percutaneous radiofrequency neurotomy for chronic neck pain: outcomes in a series of

consecutive patients. Pain Med 2005;6:282-6.

(37) Manchikanti L, Damron K, Cash K, Manchukonda R, Pampati V. Therapeutic cervical medial branch

blocks in managing chronic neck pain: a preliminary report of a randomized, double-blind, controlled

trial: clinical trial NCT0033272. Pain Physician 2006;9:333-46.

(38) Uhrenholt L, Ullerup R, Vesterby A, Gregersen M. Examination of cervical spine histological sections

- a technical note. J Whiplash & Rel Disorders 2006;5:55-60.

(39) Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics

1977;33:159-74.

(40) Müller-Gerbl M. The Subchondral Bone Plate. Springer-Verlag, Berlin, Germany, 1998.

(41) Uhrenholt L, Grunnet-Nilsson N, Hartvigsen J. Cervical spine lesions after road traffic accidents: a

systematic review. Spine 2002;27:1934-41.

(42) Ito S, Ivancic PC, Panjabi MM, Cunningham BW. Soft tissue injury threshold during simulated

whiplash: a biomechanical investigation. Spine 2004;29:979-87.

(43) Ivancic PC, Panjabi MM, Ito S. Cervical spine loads and intervertebral motions during whiplash.

Traffic Inj Prev 2006;7:389-99.

(44) Yoganandan N, Cusick JF, Pintar FA, Rao RD. Whiplash injury determination with conventional

spine imaging and cryomicrotomy. Spine 2001;26:2443-8.

(45) Uhrenholt L, Nielsen E, Vesterby A, Gregersen M, Melsen F. Detailed examination of the lower

cervical spine facet joints in a road traffic crash fatality - a case study. International Research

Counsil on the Biomeckanics of Impact (IRCOBI) annual proceedings, (Prague, Czech Republic)

2005;paper 701:411-414.

(46) Oettmeier R, Abendroth K, Oettmeier S. Analyses of the tidemark on human femoral heads. II.

Tidemark changes in osteoarthrosis--a histological and histomorphometric study in non-decalcified

preparations. Acta Morphol Hung 1989;37:169-80.

(47) Friedrich KM, Trattnig S, Millington SA, Friedrich M, Groschmidt K, Pretterklieber ML. High-field

magnetic resonance imaging of meniscoids in the zygapophyseal joints of the human cervical spine.

Spine 2007;32:244-8.

(48) Fletcher G, Haughton VM, Ho KC, Yu SW. Age-related changes in the cervical facet joints: studies

with cryomicrotomy, MR, and CT. AJNR Am J Neuroradiol 1990;11:27-30.

(49) Thambyah A, Nather A, Goh J. Mechanical properties of articular cartilage covered by the meniscus.

Osteoarthritis Cartilage 2006;14:580-8.

(50) Stäbler A, Eck J, Penning R, Milz SP, Bartl R, Resnick D, et al. Cervical spine: postmortem

assessment of accident injuries--comparison of radiographic, MR imaging, anatomic, and pathologic

findings. Radiology 2001;221:340-6.

(51) Taylor JR, Taylor MM. Cervical spinal injuries: an autopsy study of 109 blunt injuries. J


(52) Jonsson H, Jr., Bring G, Rauschning W, Sahlstedt B. Hidden cervical spine injuries in traffic accident

victims with skull fractures. J Spinal Disord 1991;4:251-63.

(53) Schonstrom N, Twomey L, Taylor J. The lateral atlanto-axial joints and their synovial folds: an in vitro

study of soft tissue injuries and fractures. J Trauma 1993;35:886-92.

(54) Taylor JR, Twomey LT. Acute injuries to cervical joints. An autopsy study of neck sprain. Spine

1993;18:1115-22.

Figure 1 Schematic overview of the histomorphometric measurements

The cartilage length (d), and the anterior and posterior folds overlap of the underlying cartilage is

measured (d & e)(A). Each facet is measured at one to five randomly selected vertical lines through the

cartilage and underlying bone with regard to the thickness of the hyaline cartilage, the calcified cartilage

and the subchondral bone (B). The parallel lines resemble the 3-mm thick anatomical slices from where

each facet is measured and the numbers one to five correspond to the vertical lines in the figure above

(C).

Figure 2 The morphological variables

Vertical fissures are present in the deep and middle part of the cartilage, original magnification x4 (A).

Widespread superficial flaking/fibrillation of the cartilage, original magnification x10 (B). Large horizontal

split through the middle part of the hyaline cartilage, original magnification x4 (C). Example of vascular

invasion of the tidemark, original magnification x10 (D). All stained with Masson-Goldner trichrome.

Figure 3 Synovial folds of the cervical spine facet joints

A close-up view of the anterior synovial fold (F) which consist of fatty adipose tissue and strands of

fibrous tissue extending into the joint cavity, original magnification x4, Masson-Goldner trichrome

staining.

Figure 4 Mean maximum cartilage length

The mean values of the maximum cartilage length with 95 % confidence intervals as a function of

cervical spine segment and gender. Females have significantly shorter cartilage lengths than males and

these differ significantly between the cervical spine segments.

Figure 5 Mean articular cartilage thickness

The mean values of the mean thickness of the articular cartilage with 95 % confidence intervals as a

function of cervical spine segment and gender. There is no significant difference in thickness in relation

to gender or age, however the articular cartilage is significantly different between the spinal segments

being thinner in the lower cervical spine.

Figure 6 Heterogenous covering of the articular surface by the synovial folds

In this parasaggital section through the lateral part of a cervical spine facet joint there is complete

covering of the articular surfaces with a synovial fold, with the posterior fold (A) and anterior fold (B)

illustrated in close-up, original magnification x1 and x4, Masson-Goldner trichrome.

Table 1

Description of the anatomical variables

Definition Comments Morphological variables Flaking

Is there any tangential flaking or fibrillation of the surface of the hyaline articular surface?

Fibrillation/flaking anywhere of the cartilage surface

Split

Are there one or more horizontal splits of the hyaline cartilage?

Eyeballed percentage (a split is a matrix separation horizontal (0-60o to the articular surface)

Fissure

Are there one or more vertical fissures of the hyaline cartilage and what percentage does the worst represent of the total non-calcified hyaline cartilage thickness?

Eyeballed percentage (a fissure is a matrix separation vertical (60-90o to the articular surface). The fissures were grouped depending on whether the fissure affected none, < 50 % (mild), 50 – 99 % (moderate) or 100% (severe) of the cartilage

Vascular invasion Is there vascular invasion extending into the hyaline cartilage through the tidemark?

A blood vessel with or without cells resembling erythrocytes must be present extending into the hyaline cartilage

Osteophytes Are there any osteophytes developing at the joint margins?

Defined as new bone formation in the periphery of the joint margin – “an added outgrowth”

Anterior fold Is an anterior fold present and intact? Is tissue present that exhibits the structure and location of an anterior synovial fold and is it intact?

Posterior fold Is a posterior fold present and intact? Is tissue present that exhibits the structure and location of a posterior synovial fold and is it intact?

Histomorphometrical variables Cartilage length Length of the articular cartilage

Defined as the direct line between the anterior and posterior eyeballed borders of cartilage covered facets. Measured on the projected image and converted by a factor 1:19.25

Anterior fold overlap The degree of overlap of the facet in question by the anterior synovial fold

Defined as the part of the anterior fold extending into the joint centre from the line perpendicular to the cartilage length line at the anterior border. Measured on the projected image and converted by a factor 1:77

Posterior fold overlap The degree of overlap of the facet in question by the posterior synovial fold

Defined as the part of the posterior fold extending into the joint centre from the line perpendicular to the cartilage length line at the posterior border. Measured on the projected image and converted by a factor 1:77

Joint to tidemark (hyaline cartilage)*

Distance from joint space surface to the tidemark

Measured as the length along the perpendicular line to the articular cartilage surface on the projected image and converted by a factor 1:77

Tidemark to osteochondral junction (calcified cartilage)*

Distance from the tidemark to the osteochondral junction


Osteochondral junction to subchondral bone edge*

Distance from the osteochondral junction to the first free space with bone marrow


* each variable is measured at 1-5 sites on each 10 μm section

Estim

ate

95%

CI

Estim

ate

95%

CI

p-va

lue

Estim

ate

95%

CI

p-va

lue

Estim

ate

95%

CI

p-va

lue

Flak

ing

(%)

75[6

7-83

]-1

8[-2

9-(-7

)]<0

.01

13[7

-20]

<0.0

01-4

[-15-

(7)]

0.46

0S

plit

(%)

72[6

5-78

]-1

8[-2

7-(-9

)]<0

.001

8[3

-13]

<0.0

1-1

0[-1

8-(-1

)]<0

.05

Any

fiss

ure

(%)

58[4

9-67

]-1

3[-2

6-(-1

)]<0

.05

0[-7

-7]

0.98

5-2

[-14-

10]

0.73

10%

< fi

ssur

e <

50%

(%)

37[3

1-43

]-4

[-13-

4]0.

304

-5[-1

0-0]

0.05

93

[-6-1

1]0.

527

50%

= fi

ssur

e <

100%

(%)

18[1

3-23

]-8

[-15-

(-1)]

<0.0

54

[0-8

]0.

051

-4[-1

1-3]

0.23

2C

ompl

ete

fissu

re =

100

% (%

)3

[1-4

]-1

[-3-1

]0.

354

1[-1

-2]

0.32

9-1

[-2-1

]0.

562

Vas

cula

r inv

asio

n (%

)15

[9-2

0]-5

[-13-

3]0.

181

-4[-8

-1]

0.09

51

[-6-9

]0.

701

Ost

eoph

ytes

(%)

4[2

-6]

-1[-4

-1]

0.30

72

[1-4

]<0

.01

-1[-3

-1]

0.43

3A

nter

ior f

old

disr

uptio

n (%

)16

[8-2

3]2

[-8-1

3]0.

655

-1[-7

-6]

0.87

9-1

1[-2

1-(-1

)]<0

.05

Pos

terio

r fol

d di

srup

tion

(%)

33[2

3-43

]-8

[-22-

5]0.

210

-10

[-18-

(-3)]

<0.0

5-1

5[-2

7-(-2

)]<0

.05

Ref

eren

ce p

erso

n: 3

5 ye

ars

old

mal

e ki

lled

in a

road

traf

fic c

rash

Ref

eren

ce p

erso

n

Varia

ble

Exam

ple:

a 3

5 ye

ars

old

fem

ale

is p

redi

cted

to h

ave

flaki

ng in

app

roxi

mat

ely

57%

, i.e

. 18%

few

er fa

cets

than

mal

es (P

<0.0

1). T

here

is a

n ag

e-re

late

d in

crea

se in

the

num

ber o

f fac

et w

ith fl

akin

g eq

uiva

lent

to

13%

per

dec

ade

(p<0

.001

) and

no

sign

ifica

nt c

orre

latio

n w

ith tr

aum

a (p

=0.4

60)

Des

crip

tion

of th

e m

orph

olog

ical

find

ings

Tabl

e 2

Incr

ease

per

10

year

s

AGE

Con

trol v

s. C

ase

CAS

E/C

ON

TRO

LG

END

ER

Fem

ales

vs.

Mal

es

Estim

ate

95%

CI

Estim

ate

95%

CI

p-va

lue

Estim

ate

95%

CI

p-va

lue

Estim

ate

95%

CI

p-va

lue

Hya

line

carti

lage

thic

knes

s (m

m)

0.72

[0.6

7-0.

77]

-0.0

5[-0

.11-

0.02

]0.

152

0.00

[-0.0

4-0.

04]

0.96

60.

04[-0

.02-

0.10

]0.

180

Cal

cifie

d ca

rtila

ge th

ickn

ess

(μm

)79

[68-

90]

5[-8

-17]

0.47

17

[-1-1

4]0.

080

0[-1

2-12

]0.

972

Sub

chon

dral

bon

e th

ickn

ess

(mm

)0.

35[0

.31-

0.39

]0.

03[-0

.02-

0.07

]0.

229

0.04

[0.0

1-0.

06]

<0.0

10.

00[-0

.04-

0.04

]0.

852

Thic

knes

s of

the

tota

l arti

cula

r car

tilag

e (m

m)

0.80

[0.7

5-0.

85]

-0.0

4[-0

.10-

0.02

]0.

207

0.01

[-0.0

3-0.

04]

0.70

80.

04[-0

.02-

0.10

]0.

184

Max

imum

car

tilag

e le

ngth

(mm

)11

.35

[10.

66-1

2.04

]-1

.33

[-2.1

9-(-0

.46)

]<0

.01

-0.2

0[-0

.70-

0.30

]0.

435

0.54

[-0.2

9-1.

37]

0.20

4C

alci

fied

carti

lage

of t

otal

arti

cula

r car

tilag

e th

ickn

ess

(%)

10[9

-12]

1[-1

-3]

0.17

71

[0-2

]0.

123

-1[-2

-1]

0.48

9A

nter

ior f

old

over

lap

(%)

16[1

1-21

]0

[-6-6

]0.

974

1[-2

-4]

0.62

25

[0-1

1]0.

055

Pos

terio

r fol

d ov

erla

p (%

)17

[13-

21]

2[-2

-7]

0.31

41

[-2-4

]0.

460

3[-2

-7]

0.24

5

Ref

eren

ce le

vel:

the

C4

infe

rior f

acet

of a

35

year

s ol

d m

ale

kille

d in

a ro

ad tr

affic

cra

sh

Des

crip

tion

of th

e hi

stom

orph

omet

ric fi

ndin

gs

Tabl

e 3

Incr

ease

per

10

year

s

AGE

Con

trol v

s. C

ase

CAS

E/C

ON

TRO

LG

END

ER

Fem

ales

vs.

Mal

esR

efer

ence

leve

l

Varia

ble

Exam

ple:

a 3

5 ye

ars

old

fem

ale

has

a hy

alin

e ca

rtila

ge th

ickn

ess

equi

vale

nt to

that

of m

ales

(0.0

5 m

m th

inne

r, p=

0.15

2). T

here

are

no

age-

rela

ted

diffe

renc

es w

ith re

gard

to th

e th

ickn

ess

(p=0

.966

) and

ther

e is

no

sign

ifica

nt c

orre

latio

n w

ith e

xpos

ure

to

traum

a (p

=0.1

80).

page 26

Table 4

Inter- and intraobserver agreement of morphological variables

Simple agreement Kappa 95% CI p-value

Simple agreement Kappa 95% CI p-value

Variable

fissure* 89.1% 0.65 [0.53-0.76] < 0.001 95.4% 0.84 [0.72-0.95] < 0.001

flaking 79.7% 0.54 [0.37-0.71] < 0.001 91.8% 0.83 [0.67-0.98] < 0.001

posterior fold disruption 83.2% 0.51 [0.32-0.70] < 0.001 92.1% 0.74 [0.57-0.90] < 0.001

split 72.9% 0.31 [0.15-0.47] < 0.001 89.9% 0.79 [0.63-0.94] < 0.001

vascular invasion 82.1% 0.29 [0.13-0.45] < 0.001 93.0% 0.70 [0.55-0.86] < 0.001

anterior fold disruption 84.6% 0.09 [-0.08-0.26] 0.156 94.9% 0.75 [0.59-0.91] < 0.001

osteophytes 91.9% -0.04 [-0.21-0.13] 0.688 96.8% 0.65 [0.50-0.80] < 0.001

*: weighted kappa95% CI: 95% confidence interval

Interobserver agreement Intraobserver agreement

page 27

morphology and pathoanatomy of the cervical spine facet ...€¦ · lars uhrenholt, edith nielsen,...

Documents